Encyclopedia of Islands

ENCYCLOPEDIA OF ISLANDS

ENCYCLOPEDIA OF ISLANDS E D I T E D BY

ROSEMARY G. GILLESPIE University of California, Berkeley

DAVID A. CLAGUE Monterey Bay Aquarium Research Institute

UNIVERSITY OF CALIFORNIA PRESS Berkeley

Los Angeles

London

University of California Press, one of the most distinguished university presses in the United States, enriches lives around the world by advancing scholarship in the humanities, social sciences, and natural sciences. Its activities are supported by the UC Press Foundation and by philanthropic contributions from individuals and institutions. For more information, visit www.ucpress.edu. Encyclopedias of the Natural World, No. 2 University of California Press Berkeley and Los Angeles, California University of California Press, Ltd. London, England © 2009 by the Regents of the University of California Library of Congress Cataloging-in-Publication Data Encyclopedia of islands / edited by Rosemary G. Gillespie and David A. Clague. p. cm. — (Encyclopedias of the natural world ; 2) Includes bibliographical references and index. ISBN 978-0-520-25649-1 (cloth : alk. paper) 1. Islands—Encyclopedias. I. Gillespie, Rosemary G., 1957II. Clague, D. A. GB471.E53 2009 551.4203—dc22 2008037221 Printed in China 16 15 14 13 12 11 10 09 10 9 8 7 6 5 4 3 2 1 The paper used in this publication meets the minimum requirements of ANSI/NISO Z39.48-1992 (R 1997) (Permanence of Paper).{infcir} Cover photograph: Mercherchar Island, Palau, courtesy Patrick L. Colin. Insets, from left: Lava from Piton de la Fournaise, Rèunion, © KM KRAFFT/CRI-Nancy-Lorraine, used with permission; Greene’s dudleya (Dudleya greenei) on Santa Cruz Island, © Kathy deWet-Oleson; male orange dove (Chrysoenas victor), endemic to Fiji, courtesy Paddy Ryan; aerial view of Bermuda, courtesy Bermuda Zoological Society (see also frontispiece).

CO NTEN TS

Contents by Subject Area / xi Contributors / xv Guide to the Encyclopedia / xxix Preface / xxxi

Adaptive Radiation / 1 Rosemary G. Gillespie

Atlantic Region / 63

Britain and Ireland / 116

Andreas Klügel

Rosemary G. Gillespie; Mark Williamson

Atolls / 67 Edward L. Winterer

Azores / 70

Canary Islands, Biology / 127

Paulo A. V. Borges; Isabel R. Amorim;

Javier Francisco-Ortega; Arnoldo Santos-

Rosalina Gabriel; Regina Cunha; António

Guerra; Juan José Bacallado

Frias Martins; Luís Silva; Ana Costa;

Canary Islands, Geology / 133

Virgílio Vieira

Kaj Hoernle; Juan-Carlos Carracedo

Cape Verde Islands / 143

Anagenesis / 8

Maria Cristina Duarte; Maria Manuel

Tod F. Stuessy

Antarctic Islands, Biology / 10

Baffin / 76 Lynda Dredge

Steven L. Chown; Jennifer E. Lee

Antarctic Islands, Geology / 17 John L. Smellie

Antilles, Biology / 20 Charles A. Woods; Florence E. Sergile

Antilles, Geology / 29 Richard E. A. Robertson

Ants / 35 Brian L. Fisher

Archaeology / 41 Marshall I. Weisler

Arctic Islands, Biology / 47 Inger Greve Alsos; Lynn Gillespie;

Baja California: Offshore Islands / 78 Martin L. Cody

Barrier Islands / 82 Miles O. Hayes

Barro Colorado / 88

Romeiras

Caroline Islands / 148 James R. Hein

Caves, as Islands / 150 David C. Culver; Tanja Pipan

Channel Islands (British Isles) / 154 Edward P. F. Rose

Egbert Giles Leigh, Jr.

Channel Islands (California), Biology / 155

Beaches / 91

Aaron Moody

Bruce Richmond

Bermuda / 95

Channel Islands (California), Geology / 161

Anne F. Glasspool; Wolfgang Sterrer

Janet Hammond Gordon

Biological Control / 99

Cichlid Fish / 165

Mark Gillespie; Steve Wratten

Ole Seehausen

Yuri M. Marusik

Arctic Islands, Geology / 55

Bird Disease / 103

Climate Change / 169

David Cameron Duffy

David A. Burney

Bird Radiations / 105

Climate on Islands / 171

Jeffrey Podos; David C. Lahti

Thomas A. Schroeder

Borneo / 111

Cold Seeps / 174

Swee-Peck Quek

Charles K. Paull

Michael J. Hambrey

Arctic Region / 59 Stephen D. Gurney

Ascension / 61 David M. Wilkinson

v

Comoros / 177

Eruptions: Laki and Tambora / 263

Galápagos Islands, Geology / 367

D. James Harris; Sara Rocha

T. Thordarson

Dennis Geist; Karen Harpp

Continental Islands / 180

Ethnobotany / 271

Gigantism / 372

David M. Watson

W. Arthur Whistler

Pasquale Raia

Convergence / 188

Exploration and Discovery / 276

Global Warming / 376

Tadashi Fukami

Scott M. Fitzpatrick

David R. Lindberg

Cook Islands / 191

Extinction / 281

Granitic Islands / 380

Tegan Hoffmann

Kevin J. Gaston

Millard F. Coffin

Coral / 197

Great Barrier Reef Islands, Biology / 382

Daphne G. Fautin; Robert W. Buddemeier

Cozumel / 203

Farallon Islands / 287

Faroe Islands / 291

Great Barrier Reef Islands, Geology / 386

Gina E. Hannon; Simun V. Arge; Anna-

Scott G. Smithers

Alfredo D. Cuarón

Crickets / 206 Daniel Otte; Greg Cowper

Harold Heatwole

Phil Capitolo

Maria Fosaa; Ditlev L. Mahler; Bergur

Cyprus / 212

Olsen; Richard H. W. Bradshaw

Ioannis Panayides

Fernando de Noronha / 297

Greek Islands, Biology / 388 Kostas A. Triantis; M. Mylonas

R. V. Fodor

Greek Islands, Geology / 392 Michael D. Higgins

Darwin and Geologic History / 217 James H. Natland

Deforestation / 221 Barry V. Rolett

Dispersal / 224 Isabelle Olivieri

Dodo / 228 J. P. Hume

Drosophila / 232 Patrick M. O’Grady; Karl N. Magnacca; Richard T. LaPoint

Dwarfism / 235 Shai Meiri; Pasquale Raia

Earthquakes / 240

Fiji, Biology / 298 Paddy Ryan

Fiji, Geology / 305 Howard Colley

Fish Stocks/Overfishing / 310 Jon Brodziak

Flightlessness / 311 Curtis Ewing

Fossil Birds / 318

Grant McCall

Ecological Release / 251

Honeycreepers, Hawaiian / 410 Robert C. Fleischer

Founder Effects / 326

Patrick V. Kirch

Michael C. Whitlock

Hurricanes and Typhoons / 418

Fragmentation / 328

Thomas A. Schroeder

Luis Cayuela

Hydrology / 420

Fraser Island / 330

Christian Depraetere; Marc Morell

Brad Balukjian

Hydrothermal Vents / 424

French Polynesia, Biology / 332

Robert C. Vrijenhoek

Jean-Yves Meyer; Bernard Salvat

French Polynesia, Geology / 338 Alain Bonneville

Freshwater Habitats / 343 Alan P. Covich

Quentin C. Cronk; Diana M. Percy

Frogs / 347 Ephemeral Islands, Biology / 258

David R. Sherrod

Human Impacts, Pre-European / 414

Rosemary G. Gillespie

Endemism / 253

Hawaiian Islands, Geology / 404

Trevor H. Worthy

Paul Okubo; David A. Clague

Easter Island / 244

Hawaiian Islands, Biology / 397 Jonathan Price

Rafe M. Brown

Sofie Vandendriessche; Magda Vincx;

Iceland / 428 Sigurdur Steinthorsson

Inbreeding / 436 Leonard Nunney

Indian Region / 437 Frederick A. Frey

Steven Degraer

Indonesia, Biology / 446 Ephemeral Islands, Geology / 259

Galápagos Finches / 352

Kazuhiko Kano

Heather Farrington; Kenneth Petren

Erosion, Coastal / 261

Galápagos Islands, Biology / 357

Wayne Stephenson

Terrence M. Gosliner

Tigga Kingston

Indonesia, Geology / 454

vi

CONTENTS

Robert Hall

Insect Radiations / 460

Landslides / 535

Marshall Islands / 610

Diana M. Percy

Simon J. Day

Nancy Vander Velde

Inselbergs / 466

Land Snails / 537

Mascarene Islands, Biology / 612

Stefan Porembski

Brenden S. Holland

Christophe Thébaud; Ben H. Warren;

Introduced Species / 469

Lava and Ash / 542

Daniel Simberloff

Katharine V. Cashman

Invasion Biology / 475

Lava Tubes / 544

George Roderick; Philippe Vernon

Jim Kauahikaua; Frank Howarth; Ken Hon

Island Arcs / 481

Lemurs and Tarsiers / 549

Richard J. Arculus

Robert D. Martin

Island Biogeography, Theory of / 486

Line Islands / 553

José María Fernández-Palacios

Island Formation / 490 Patrick D. Nunn

Island Rule / 492 Shai Meiri

Christopher Charles; Stuart Sandin

Lizard Radiations / 558 Miguel Vences

Lophelia Oases / 564 Sandra Brooke

Lord Howe Island / 568 Carole S. Hickman

Dominique Strasberg; Anthony Cheke

Mascarene Islands, Geology / 620 Robert A. Duncan

Mediterranean Region / 622 John Wainwright

Metapopulations / 629 Dag Øystein Hjermann

Midway / 631 Elizabeth Flint

Missionaries, Effects of / 633 Alan I. Kaplan; Vincent H. Resh

Moa / 638 Allan J. Baker

Motu / 641 Francis J. Murphy

Japan’s Islands, Biology / 497 Lázaro M. Echenique-Diaz; Masakado Kawata; Jun Yokoyama

Japan’s Islands, Geology / 500 S. Maruyama; S. Yanai; Y. Isozaki; D. Hirata

Juan Fernandez Islands / 507 Simon Haberle

Macquarie, Biology / 573 Jenny Scott

Macquarie, Geology / 575 Arjan Dijkstra

Madagascar / 577 Steven M. Goodman

Madeira Archipelago / 582 Kick ’em Jenny / 510

New Caledonia, Biology / 643 Jérôme Murienne

New Caledonia, Geology / 645 Timothy J. Rawling

Newfoundland / 649 Harold Williams

Dora Aguin-Pombo; Miguel A. A. Pinheiro

New Guinea, Biology / 652

de Carvalho

Allen Allison

Jan Lindsay

Kı–puka / 512

Makatea Islands / 585

New Guinea, Geology / 659

Lucien F. Montaggioni

Hugh L. Davies

Maldives / 586

New Zealand, Biology / 665

Paul Kench

Steven A. Trewick; Mary Morgan-Richards

Amy G. Vandergast

Komodo Dragons / 513 Tim Jessop

Kon-Tiki / 515

Mammal Radiations / 588

New Zealand, Geology / 673

Lawrence R. Heaney; Steven M. Goodman

Hamish Campbell; Charles A. Landis

Mangrove Islands / 591

Nuclear Bomb Testing / 680

Peter Saenger

Edward L. Winterer

Robert C. Suggs

Krakatau / 517 Robert J. Whittaker

Marianas, Biology / 593 Kurile Islands / 520

Gordon H. Rodda

Alexander Belousov; Marina Belousova; Thomas P. Miller

Lakes, as Islands / 526 Shelley Arnott

Land Crabs on Christmas Island / 532 Peter Green

Marianas, Geology / 598

Oases / 686

Frank A. Trusdell

Slaheddine Selmi; Thierry Boulinier

Marine Lakes / 603

Oceanic Islands / 689

Michael N Dawson; Laura E. Martin;

Patrick D. Nunn

Lori J. Bell; Sharon Patris

Orchids / 696

Marine Protected Areas / 607

David L. Roberts; Richard M. Bateman

Alan M. Friedlander

Organic Falls on the Ocean Floor / 700 Craig R. Smith

CONTENTS

vii

Refugia / 785

Species–Area Relationship / 857

Angus Davison

David A. Spiller; Thomas W. Schoener

Anthony A. P. Koppers

Relaxation / 787

Spiders / 861

Palau / 715

Kenneth J. Feeley

Miquel A. Arnedo

Alan R. Olsen

Research Stations / 788

Spitsbergen / 865

Pantepui / 717

Neil Davies

Maria Włodarska-Kowalczuk

Valentí Rull

Rodents / 792

Sri Lanka / 866

Peopling the Pacific / 720

David Towns

Colin Groves; Kelum Manamendra-

Patrick V. Kirch

Rottnest Island / 796

Philippines, Biology / 723

Anne Brearley

Pacific Region / 702

St. Helena / 870

Rafe M. Brown; Arvin C. Diesmos

Philip Ashmole; Myrtle Ashmole

Philippines, Geology / 732

Sticklebacks / 873

Graciano Yumul, Jr.; Carla Dimalanta;

Samoa, Biology / 799

Michael A. Bell

Karlo Queaño; Edanjarlo Marquez

A. C. Medeiros

Succession / 877

Phosphate Islands / 738

Samoa, Geology / 802

Beatrijs Bossuyt

James R. Hein

James H. Natland

Surf in the Tropics / 879

São Tomé, Príncipe, and Annobon / 808

Graham Symonds; Thomas C. Lippmann

D. James Harris

Sturla Fridriksson

Seabirds / 811

Sustainability / 888

Mark J. Rauzon; Sheila Conant

R. R. Thaman

Pigs and Goats / 741 Elizabeth Matisoo-Smith

Pitcairn / 744 Naomi Kingston; Noeleen Smyth

Plant Disease / 747

Surtsey / 883

Ulla Carlsson-Granér; Lars Ericson;

Sea-Level Change / 815

Barbara E. Giles

W. H. Berger

Plate Tectonics / 752

Seamounts, Biology / 818

Taiwan, Biology / 897

Roger C. Searle

Malcolm Clark

Man-Miao Yang; Kuang-Ying Huang

Pocket Basins and Deep-Sea Speciation / 755

Seamounts, Geology / 821

Taiwan, Geology / 902

Paul Wessel

Yue-Gau Chen

Sexual Selection / 825

Tasmania / 904

Kenneth Y. Kaneshiro; Richard T. LaPoint

Alastair M. M. Richardson

Bruce H. Robison; William M. Hamner

Polynesian Voyaging / 758 Atholl Anderson

Popular Culture, Islands in / 761 Vincent H. Resh; Jonathan P. Resh

Population Genetics, Island Models in / 766 Jeffrey D. Lozier

Prisons and Penal Settlements / 767 Ephraim Cohen

Seychelles / 829

Tatoosh / 909

Justin Gerlach

Egbert Giles Leigh, Jr.; Robert T. Paine

Shipwrecks / 833

Taxon Cycle / 912

James Hayward

Mandy L. Heddle

Silverswords / 835

Tides / 914

Bruce G. Baldwin

Marlene Noble

Sky Islands / 839

Tierra del Fuego / 917

John E. McCormack; Huateng Huang;

Matthew J. James; John M. Woram

L. Lacey Knowles

Snakes / 843 Gordon H. Rodda

Radiation Zone / 772 Kostas A. Triantis; Robert J. Whittaker

Socotra Archipelago / 846 Kay Van Damme

Rafting / 775 Christophe Abegg

Solomon Islands, Biology / 851 Orlo C. Steele

Tonga / 918 Donald R. Drake

Tortoises / 921 Charles R. Crumly

Trinidad and Tobago / 926 Christopher K. Starr

Reef Ecology and Conservation / 779

Solomon Islands, Geology / 854

Tristan da Cunha and Gough Island / 929

Robert H. Richmond; Willy Kostka;

Hugh L. Davies

Peter G. Ryan

Noah Idechong

viii

Arachchi

CONTENTS

Tsunamis / 933

Volcanic Islands / 950

Whale Falls / 973

Emile A. Okal

John M. Sinton

Amy Baco

Voyage of the Beagle / 954

Whales and Whaling / 975

Jere H. Lipps

Joe Roman

Vancouver / 937

Wizard Island / 979

Martin L. Cody

David W. Ramsey

Vanuatu / 939

Wallace, Alfred Russel / 962

Jérôme Munzinger

Elin Claridge

Vegetation / 941

Wallace’s Line and Other Biogeographic Boundaries / 967

Dieter Mueller-Dombois

Vicariance / 947 Michael Heads

Zanzibar / 982 N. D. Burgess; R. A. D. Burgess

Jeremy D. Holloway

Warming Island / 971 Kurt M. Cuffey

Glossary / 987 Index / 1019

CONTENTS

ix

This page intentionally left blank

CO NTENTS BY S UBJE CT AREA

GEOGRAPHY

Organic Falls on the Ocean Floor

Arctic Region

Pantepui

Atlantic Region

Phosphate Islands

Indian Region

Pocket Basins and Deep-Sea Speciation

Mediterranean Region

Seamounts, Biology

Pacific Region

Seamounts, Geology Sky Islands

ISLAND TYPES

Atolls

Volcanic Islands Whale Falls

Barrier Islands Caves, as Islands

IMPORTANT ISLANDS

Cold Seeps

Ascension

Continental Islands

Azores

Ephemeral Islands, Biology

Baffin

Ephemeral Islands, Geology

Baja California: Offshore Islands

Granitic Islands

Bermuda

Hydrothermal Vents

Borneo

Inselbergs

Britain and Ireland

Island Arcs

Cape Verde Islands

KXpuka

Caroline Islands

Lakes, as Islands

Channel Islands (British Isles)

Lava Tubes

Comoros

Lophelia Oases

Cook Islands

Makatea Islands

Cozumel

Mangrove Islands

Cyprus

Marine Lakes

Easter Island

Motu

Farallon Islands

Oases

Faroe Islands

Oceanic Islands

Fernando de Noronha xi

Fraser Island

Great Barrier Reef Islands, Geology

Iceland

Greek Islands, Geology

Juan Fernandez Islands

Hawaiian Islands, Geology

Krakatau

Indonesia, Geology

Kurile Islands

Japan’s Islands, Geology

Line Islands

Kick ‘em Jenny

Lord Howe Island

Macquarie, Geology

Madagascar

Marianas, Geology

Madeira

Mascarene Islands, Geology

Maldives

New Caledonia, Geology

Marshall Islands

New Guinea, Geology

Newfoundland

New Zealand, Geology

Palau

Philippines, Geology

Rottnest Island

Samoa, Geology

São Tomé, Príncipe, and Annobon

Solomon Islands, Geology

Seychelles

Taiwan, Geology

Socotra Archipelago Spitsbergen Sri Lanka

Earthquakes

St. Helena

Erosion, Coastal

Surtsey

Eruptions: Laki and Tambora

Tasmania

Island Formation

Tonga

Landslides

Trinidad and Tobago

Lava and Ash

Tristan da Cunha and Gough Island

Plate Tectonics

Vancouver Vanuatu Warming Island Wizard Island Zanzibar GEOLOGY

xii

GEOLOGIC PROCESSES

BIOGEOGRAPHY

Antarctic Islands, Biology Antilles, Biology Arctic Islands, Biology Canary Islands, Biology Channel Islands (California), Biology

Antarctic Islands, Geology

Fiji, Biology

Antilles, Geology

French Polynesia, Biology

Arctic Islands, Geology

Galápagos Islands, Biology

Beaches

Great Barrier Reef Islands, Biology

Canary Islands, Geology

Greek Islands, Biology

Channel Islands (California), Geology

Hawaiian Islands, Biology

Fiji, Geology

Indonesia, Biology

French Polynesia, Geology

Japan’s Islands, Biology

Galápagos Islands, Geology

Macquarie, Biology

CONTENTS BY SUBJECT AREA

Marianas, Biology

OCEANOGRAPHY AND CLIMATOLOGY

Mascarene Islands, Biology

Climate Change

New Caledonia, Biology

Climate on Islands

New Guinea, Biology

Hurricanes and Typhoons

New Zealand, Biology

Hydrology

Philippines, Biology

Sea-Level Change

Samoa, Biology

Surf in the Tropics

Solomon Islands, Biology

Tides

Taiwan, Biology

Tsunamis

ECOLOGY AND EVOLUTION

Adaptive Radiation Anagenesis Convergence Dispersal Dwarfism Ecological Release Endemism Flightlessness Fossil Birds Founder Effects Fragmentation Freshwater Habitats Gigantism Inbreeding Invasion Biology Island Biogeography, Theory of Island Rule Metapopulations Population Genetics, Island Models in Radiation Zone Rafting Refugia Relaxation Sexual Selection

PLANTS AND ANIMALS

Bird Radiations Cichlid Fish Coral Crickets Dodo Drosophila Frogs Galápagos Finches Honeycreepers, Hawaiian Insect Radiations Komodo Dragons Land Crabs on Christmas Island Land Snails Lemurs and Tarsiers Lizard Radiations Mammal Radiations Moa Orchids Reef Ecology and Conservation Seabirds Silverswords Spiders Sticklebacks Tortoises

Species-Area Relationship Succession

HUMAN IMPACT

Taxon Cycle

Ants

Vegetation

Biological Control

Vicariance

Bird Disease

Wallace’s Line and Other Biogeographic Boundaries

Deforestation

CONTENTS BY SUBJECT AREA

xiii

Ethnobotany

Barro Colorado Island

Extinction

Darwin and Geologic History

Fish Stocks/Overfishing

Exploration and Discovery

Global Warming

Human Impacts, Pre-European

Introduced Species

Kon-Tiki

Marine Protected Areas

Midway

Pigs and Goats

Missionaries, Effects of

Plant Disease

Nuclear Bomb Testing

Popular Culture, Islands in

Peopling the Pacific

Research Stations

Pitcairn

Rodents

Polynesian Voyaging

Snakes

Prisons and Penal Settlements

Sustainability

Shipwrecks

Whales and Whaling

Tatoosh

HISTORY AND PRE-HISTORY

Archaeology

xiv

CONTENTS BY SUBJECT AREA

Tierra del Fuego Voyage of the Beagle Wallace, Alfred Russel

CO NTRIBUTOR S

CHRISTOPHE ABEGG

SHELLEY ARNOTT

German Primate Centre Padang, Sumatra Barat, Indonesia Rafting

Queen’s University Kingston, Ontario, Canada Lakes, as Islands

DORA AGUIN-POMBO

MYRTLE ASHMOLE

University of Madeira, Portugal Madeira Archipelago

Kidston Mill Peebles, Scotland, United Kingdom St. Helena

ALLEN ALLISON

Bishop Museum Honolulu, Hawaii New Guinea, Biology INGER GREVE ALSOS

University Centre of Svalbard Longyearbyen, Norway Arctic Islands, Biology ISABEL R. AMORIM

University of the Azores Terceira, Portugal Azores ATHOLL ANDERSON

Australian National University, Canberra Polynesian Voyaging RICHARD J. ARCULUS

Australian National University, Canberra Island Arcs SIMUN V. ARGE

National Museum of the Faroe Islands, Tórshavn Faroe Islands MIQUEL A. ARNEDO

University of Barcelona, Spain Spiders

PHILIP ASHMOLE

Kidston Mill Peebles, Scotland, United Kingdom St. Helena JUAN JOSÉ BACALLADO

Museum of Natural Sciences Tenerife, Spain Canary Islands, Biology AMY BACO

Associated Scientists at Woods Hole Woods Hole, Massachusetts Whale Falls ALLAN J. BAKER

Royal Ontario Museum Toronto, Ontario, Canada Moa BRUCE G. BALDWIN

University of California, Berkeley Silverswords BRAD BALUKJIAN

University of California, Berkeley Fraser Island

xv

RICHARD M. BATEMAN

JON BRODZIAK

Royal Botanic Gardens, Kew Richmond, Surrey, United Kingdom Orchids

Pacific Islands Fisheries Science Center, NOAA Fisheries Service Honolulu, Hawaii Fish Stocks/Overfishing

LORI J. BELL

Coral Reef Research Foundation Koror, Palau Marine Lakes MICHAEL A. BELL

Stony Brook University Stony Brook, New York Sticklebacks ALEXANDER BELOUSOV

Institute of Volcanology and Seismology Petropavlovsk, Russia Kurile Islands MARINA BELOUSOVA

Institute of Volcanology and Seismology Petropavlovsk, Russia Kurile Islands W. H. BERGER

University of California, San Diego Sea-Level Change ALAIN BONNEVILLE

Institut de Physique du Globe de Paris, France French Polynesia, Geology PAULO A. V. BORGES

University of the Azores Terceira, Portugal Azores

SANDRA BROOKE

Marine Conservation Biology Institute Bellevue, WA Lophelia Oases RAFE M. BROWN

University of Kansas, Lawrence Frogs Philippines, Biology ROBERT W. BUDDEMEIER

Kansas Geological Survey Lawrence, Kansas Coral N. D. BURGESS

University of Cambridge Cambridge, United Kingdom Zanzibar R. A. D. BURGESS

Cambridge Regional College Cambridge, United Kingdom Zanzibar DAVID A. BURNEY

National Tropical Botanical Garden Kalaheo, Hawaii Climate Change HAMISH CAMPBELL

University of Ghent, Belgium Succession

GNS Science Lower Hutt, New Zealand New Zealand, Geology

THIERRY BOULINIER

PHIL CAPITOLO

Centre d’Écologie Fonctionnelle et Évolutive Montpellier, France Oases

Berkeley, California Farallon Islands

BEATRIJS BOSSUYT

JUAN-CARLOS CARRACEDO

Liverpool University, United Kingdom Faroe Islands

Estación Volcanológica de Canarias Tenerife, Spain Canary Islands, Geology

ANNE BREARLEY

ULLA CARLSSON-GRANÉR

University of Western Australia Crawley, Western Australia, Australia Rottnest Island

Umeå University, Sweden Plant Disease

RICHARD H. W. BRADSHAW

KATHARINE V. CASHMAN

University of Oregon, Eugene Lava and Ash

xvi

CONTRIBUTORS

LUIS CAYUELA

SHEILA CONANT

University of Alcalá Madrid, Spain Fragmentation

University of Hawaii, Honolulu Seabirds

CHRISTOPHER CHARLES

University of the Azores Vairão, Portugal Azores

Scripps Institution of Oceanography La Jolla, California Line Islands ANTHONY CHEKE

Oxford, United Kingdom Mascarene Islands, Biology YUE-GAU CHEN

National Taiwan University Taipei, Taiwan Taiwan, Geology STEVEN L. CHOWN

Stellenbosch University Matieland, South Africa Antarctic Islands, Biology DAVID A. CLAGUE

Monterey Bay Aquarium Research Institute Moss Landing, California Earthquakes ELIN CLARIDGE

Richard B. Gump Moorea Field Station University of California, Berkeley Moorea, French Polynesia Wallace, Alfred Russel MALCOLM CLARK

National Institute of Water and Atmospheric Research Kilbirnie, Wellington, New Zealand Seamounts, Biology MARTIN L. CODY

University of California, Los Angeles Baja California: Offshore Islands Vancouver MILLARD F. COFFIN

University of Southampton, United Kingdom Granitic Islands EPHRAIM COHEN

Hebrew University of Jerusalem Rehovot, Israel Prisons and Penal Settlements HOWARD COLLEY

Higher Education Academy Heslington, York, United Kingdom Fiji, Geology

ANA COSTA

ALAN P. COVICH

University of Georgia, Athens Freshwater Habitats GREG COWPER

Academy of Natural Sciences Philadelphia, Pennsylvania Crickets QUENTIN C. CRONK

University of British Columbia Vancouver, British Columbia, Canada Endemism CHARLES R. CRUMLY

University of California, Berkeley Tortoises ALFREDO D. CUARÓN

Servicios Ambientales, Conservación Biológica y Educación Morelia, Michoacan, Mexico Cozumel KURT M. CUFFEY

University of California, Berkeley Warming Island DAVID C. CULVER

American University Washington, DC Caves, as Islands REGINA CUNHA

University of the Azores Varão, Portugal Azores HUGH L. DAVIES

University of Papua New Guinea New Guinea, Geology Solomon Islands, Geology NEIL DAVIES

Richard B. Gump Moorea Field Station University of California, Berkeley Moorea, French Polynesia Research Stations

CONTRIBUTORS

xvii

ANGUS DAVISON

LARS ERICSON

University of Nottingham, United Kingdom Refugia

Umeå University, Sweden Plant Disease

MICHAEL N DAWSON

CURTIS EWING

University of California, Merced Marine Lakes

University of California, Berkeley Flightlessness

SIMON J. DAY

HEATHER FARRINGTON

University College London, United Kingdom Landslides

University of Cincinnati, Ohio Galápagos Finches

STEVEN DEGRAER

DAPHNE G. FAUTIN

Ghent University, Belgium Ephemeral Islands, Biology

University of Kansas, Lawrence Coral

CHRISTIAN DEPRAETERE

KENNETH J. FEELEY

Global Islands Network Grenoble, France Hydrology

Wake Forest University Winston-Salem, North Carolina Relaxation

ARVIN C. DIESMOS

JOSÉ MARÍA FERNÁNDEZ-PALACIOS

National Museum of the Philippines Manila, Philippines Philippines, Biology

La Laguna University Tenerife, Spain Island Biogeography, Theory of

ARJAN DIJKSTRA

BRIAN L. FISHER

University of Neuchâtel, Switzerland Macquarie, Geology

California Academy of Sciences San Francisco, California Ants

CARLA DIMALANTA

University of the Philippines Diliman, Quezon City, Philippines Philippines, Geology

SCOTT M. FITZPATRICK

DONALD R. DRAKE

ROBERT C. FLEISCHER

University of Hawaii, Manoa Tonga

Smithsonian Institution Washington, DC Honeycreepers, Hawaiian

LYNDA DREDGE

Geological Survey of Canada Ottawa, Ontario, Canada Baffin MARIA CRISTINA DUARTE

xviii

North Carolina State University, Raleigh Exploration and Discovery

ELIZABETH FLINT

U.S. Fish and Wildlife Service Honolulu, Hawaii Midway

Instituto de Investigação Científica Tropical Lisbon, Portugal Cape Verde Islands

R. V. FODOR

DAVID CAMERON DUFFY

ANNA-MARIA FOSAA

University of Hawaii, Manoa Bird Disease

Faroese Museum of Natural History, Tórshavn Faroe Islands

ROBERT A. DUNCAN

JAVIER FRANCISCO-ORTEGA

Oregon State University, Corvallis Mascarene Islands, Geology

Florida International University, Miami Canary Islands, Biology

LÁZARO M. ECHENIQUE-DIAZ

FREDERICK A. FREY

Tohoku University Aoba-ku, Sendai, Japan Japan’s Islands, Biology

Massachusetts Institute of Technology, Cambridge Indian Region

CONTRIBUTORS

North Carolina State University, Raleigh Fernando de Noronha

STURLA FRIDRIKSSON

JANET HAMMOND GORDON

Reykjavik, Iceland Surtsey

Pasadena City College, California Channel Islands (California), Geology

ALAN M. FRIEDLANDER

TERRENCE M. GOSLINER

University of Hawaii, Honolulu Marine Protected Areas

California Academy of Sciences San Francisco, California Galápagos Islands, Biology

TADASHI FUKAMI

Stanford University Stanford, California Convergence ROSALINA GABRIEL

PETER GREEN

La Trobe University Bundoora, Victoria, Australia Land Crabs on Christmas Island

University of the Azores Terceira, Portugal Azores

COLIN GROVES

KEVIN J. GASTON

STEPHEN D. GURNEY

University of Sheffield, United Kingdom Extinction

University of Reading, United Kingdom Arctic Region

DENNIS GEIST

SIMON HABERLE

University of Idaho, Moscow Galápagos Islands, Geology

Australian National University, Canberra Juan Fernandez Islands

JUSTIN GERLACH

ROBERT HALL

Nature Protection Trust of Seychelles Cambridge, United Kingdom Seychelles

Royal Holloway University of London Egham, Surrey, United Kingdom Indonesia, Geology

BARBARA GILES

MICHAEL J. HAMBREY

Umeå University, Sweden Plant Disease

Aberystwyth University Ceredigion, Wales, United Kingdom Arctic Islands, Geology

LYNN GILLESPIE

Canadian Museum of Nature, Ottawa Arctic Islands, Biology MARK GILLESPIE

Lincoln University Christchurch, New Zealand Biological Control ROSEMARY G. GILLESPIE

Australian National University, Canberra Sri Lanka

WILLIAM M. HAMNER

University of California, Los Angeles Pocket Basins and Deep-Sea Speciation GINA E. HANNON

Southern Swedish Forest Research Centre Alnarp, Sweden Faroe Islands

University of California, Berkeley Adaptive Radiation Britain and Ireland Ecological Release

KAREN HARPP

ANNE F. GLASSPOOL

D. JAMES HARRIS

Bermuda Zoological Society Flatts, Bermuda Bermuda

University of Porto Vila do Conde, Portugal Comoros São Tomé, Príncipe, and Annobon

STEVEN M. GOODMAN

The Field Museum Chicago, Illinois Madagascar Mammal Radiations

Colgate University Hamilton, New York Galápagos Islands, Geology

MILES O. HAYES

Research Planning, Inc. Columbia, South Carolina Barrier Islands CONTRIBUTORS

xix

JAMES HAYWARD

JEREMY D. HOLLOWAY

University of California, Berkeley Shipwrecks

The Natural History Museum London, United Kingdom Wallace’s Line and Other Biogeographic Boundaries

MICHAEL HEADS

Ngaio, Wellington, New Zealand Vicariance LAWRENCE R. HEANEY

The Field Museum Chicago, Illinois Mammal Radiations HAROLD HEATWOLE

North Carolina State University, Raleigh Great Barrier Reef Islands, Biology MANDY L. HEDDLE

Center for Environmental Education Ahmedabad, India Taxon Cycle JAMES R. HEIN

University of Hawaii, Hilo Lava Tubes FRANK HOWARTH

Hawaii Biological Survey, Bishop Museum Honolulu, Hawaii Lava Tubes HUATENG HUANG

University of Michigan, Ann Arbor Sky Islands KUANG-YING HUANG

Yangmingshan National Park Taipei, Taiwan Taiwan, Biology

U.S. Geological Survey Menlo Park, California Caroline Islands Phosphate Islands

J. P. HUME

CAROLE S. HICKMAN

NOAH IDECHONG

University of California, Berkeley Lord Howe Island

Palau National Congress, Koror Reef Ecology and Conservation

MICHAEL D. HIGGINS

Y. ISOZAKI

University of Québec Chicoutimi, Canada Greek Islands, Geology

University of Tokyo, Japan Japan’s Islands, Geology

D. HIRATA

Sonoma State University Rohnert Park, California Tierra del Fuego

Kanagawa Prefectural Museum of Natural History Odawara, Japan Japan’s Islands, Geology

The Natural History Museum London, United Kingdom Dodo

MATTHEW J. JAMES

TIM JESSOP

University of Oslo, Norway Metapopulations

Zoos Victoria Parkville, Victoria, Australia Komodo Dragons

KAJ HOERNLE

KENNETH Y. KANESHIRO

Leibniz Institute of Marine Sciences (IFM–GEOMAR) Kiel, Germany Canary Islands, Geology

University of Hawaii, Manoa Sexual Selection

TEGAN HOFFMANN

Geological Survey of Japan Tsukuba, Ibaraki, Japan Ephemeral Islands, Geology

DAG ØYSTEIN HJERMANN

T. C. Hoffmann and Associates, LLC Oakland, California Cook Islands BRENDEN S. HOLLAND

University of Hawaii, Manoa Land Snails

xx

KEN HON

CONTRIBUTORS

KAZUHIKO KANO

ALAN I. KAPLAN

El Cerrito, California Missionaries, Effects of

JIM KAUAHIKAUA

JENNIFER E. LEE

Hawaiian Volcano Observatory U.S. Geological Survey, Hawaii National Park Lava Tubes

Stellenbosch University Matieland, South Africa Antarctic Islands, Biology

MASAKADO KAWATA

EGBERT GILES LEIGH, JR.

Tohoku University Aoba-ku, Sendai, Japan Japan’s Islands, Biology

Smithsonian Tropical Research Institute Balboa, Panama Barro Colorado Tatoosh

PAUL KENCH

University of Auckland, New Zealand Maldives NAOMI KINGSTON

DAVID R. LINDBERG

University of California, Berkeley Global Warming

National Parks and Wildlife Service Dublin, Ireland Pitcairn

JAN LINDSAY

TIGGA KINGSTON

THOMAS C. LIPPMANN

Texas Tech University Lubbock, Texas Indonesia, Biology

Ohio State University, Columbus Surf in the Tropics

PATRICK V. KIRCH

University of California, Berkeley Voyage of the Beagle

University of California, Berkeley Human Impacts, Pre-European Peopling the Pacific ANDREAS KLÜGEL

University of Bremen, Germany Atlantic Region L. LACEY KNOWLES

University of Michigan, Ann Arbor Sky Islands ANTHONY A. P. KOPPERS

Oregon State University, Corvallis Pacific Region WILLY KOSTKA

Micronesia Conservation Trust Kolonia, Pohnpei Reef Ecology and Conservation DAVID C. LAHTI

University of Massachusetts, Amherst Bird Radiations CHARLES A. LANDIS

University of Otago Dunedin, New Zealand New Zealand, Geology RICHARD T. LAPOINT

University of California, Berkeley Drosophila Sexual Selection

University of Auckland, New Zealand Kick ’em Jenny

JERE H. LIPPS

JEFFREY D. LOZIER

University of Illinois, Urbana-Champaign Population Genetics, Island Models in KARL N. MAGNACCA

University of California, Berkeley Drosophila DITLEV L. MAHLER

National Museum Copenhagen, Denmark Faroe Islands KELUM MANAMENDRA-ARACHCHI

Wildlife Heritage Trust Colombo, Sri Lanka Sri Lanka EDANJARLO MARQUEZ

University of the Philippines, Manila Philippines, Geology LAURA E. MARTIN

University of California, Merced Marine Lakes ROBERT D. MARTIN

The Field Museum Chicago, Illinois Lemurs and Tarsiers

CONTRIBUTORS

xxi

ANTÓNIO FRIAS MARTINS

AARON MOODY

University of the Azores Vairão, Portugal Azores

University of North Carolina, Chapel Hill Channel Islands (California), Biology

YURI M. MARUSIK

Institut de Recherche pour le Développement Fort de France, Martinique Hydrology

Institute for Biological Problems of the North Magadan, Russia Arctic Islands, Biology S. MARUYAMA

Tokyo Institute of Technology Meguro, Tokyo, Japan Japan’s Islands, Geology ELIZABETH MATISOO-SMITH

University of Auckland, New Zealand Pigs and Goats GRANT MCCALL

University of New South Wales Sydney, Australia Easter Island

MARC MORELL

MARY MORGAN-RICHARDS

Massey University Palmerston North, New Zealand New Zealand, Biology DIETER MUELLER-DOMBOIS

University of Hawaii, Manoa Vegetation JÉRÔME MUNZINGER

Institut de Recherche pour le Développement Nouméa, New Caledonia Vanuatu JÉRÔME MURIENNE

University of Michigan, Ann Arbor Sky Islands

Harvard University Cambridge, Massachusetts New Caledonia, Biology

A. C. MEDEIROS

FRANCIS J. MURPHY

Pacific Island Ecosystems Research Center, U.S. Geological Survey Makawao, Hawaii Samoa, Biology

Richard B. Gump Moorea Field Station University of California, Berkeley Moorea, French Polynesia Motu

SHAI MEIRI

M. MYLONAS

Imperial College London Ascot, United Kingdom Dwarfism Island Rule

University of Crete Irakleio, Greece Greek Islands, Biology

JEAN-YVES MEYER

University of Miami, Florida Darwin and Geologic History Samoa, Geology

JOHN E. MCCORMACK

Department of Research of the Government of French Polynesia Papeete, Tahiti, French Polynesia French Polynesia, Biology THOMAS P. MILLER

Alaska Volcano Observatory, U.S. Geological Survey Anchorage, Alaska Kurile Islands LUCIEN F. MONTAGGIONI

University of Provence Marseille, France Makatea Islands

JAMES H. NATLAND

MARLENE NOBLE

U.S. Geological Survey Menlo Park, California Tides PATRICK D. NUNN

University of the South Pacific Suva, Fiji Island Formation Oceanic Islands LEONARD NUNNEY

University of California, Riverside Inbreeding

xxii

CONTRIBUTORS

PATRICK M. O’GRADY

KENNETH PETREN

University of California, Berkeley Drosophila

University of Cincinnati, Ohio Galápagos Finches

EMILE A. OKAL

MIGUEL A. A. PINHEIRO DE CARVALHO

Northwestern University Evanston, Illinois Tsunamis

University of Madeira, Portugal Madeira Archipelago

PAUL OKUBO

Karst Research Institute Postojna, Slovenia Caves, as Islands

Hawaiian Volcano Observatory U.S. Geological Survey, Hawaii National Park Earthquakes ISABELLE OLIVIERI

University of Montpellier, France Dispersal ALAN R. OLSEN

Belau National Museum Koror, Palau Palau BERGUR OLSEN

Faroese Fisheries Laboratory, Tórshavn Faroe Islands DANIEL OTTE

Academy of Natural Sciences Philadelphia, Pennsylvania Crickets ROBERT T. PAINE

University of Washington, Seattle Tatoosh IOANNIS PANAYIDES

Cyprus Geological Survey Lefkosai, Nicosia, Cyprus Cyprus SHARON PATRIS

Coral Reef Research Foundation Koror, Palau Marine Lakes CHARLES K. PAULL

Monterey Bay Aquarium Research Institute Moss Landing, California Cold Seeps DIANA M. PERCY

University of British Columbia Vancouver, British Columbia, Canada Endemism Insect Radiations

TANJA PIPAN

JEFFREY PODOS

University of Massachusetts, Amherst Bird Radiations STEFAN POREMBSKI

University of Rostock, Germany Inselbergs JONATHAN PRICE

University of Hawaii, Hilo Hawaiian Islands, Biology KARLO QUEAÑO

Department of Environment and Natural Resources Diliman, Quezon City, Philippines Philippines, Geology SWEE-PECK QUEK

Harvard University Cambridge, Massachusetts Borneo PASQUALE RAIA

University of Naples, Italy Dwarfism Gigantism DAVID W. RAMSEY

U.S. Geological Survey Vancouver, Washington Wizard Island MARK J. RAUZON

Marine Endeavours Oakland, California Seabirds TIMOTHY J. RAWLING

University of Melbourne Melbourne, Victoria, Australia New Caledonia, Geology JONATHAN P. RESH

Undaunted Design Co. Chicago, Illinois Popular Culture, Islands in

CONTRIBUTORS

xxiii

VINCENT H. RESH

MARIA MANUEL ROMEIRAS

University of California, Berkeley Missionaries, Effects of Popular Culture, Islands in

Instituto de Investigação Científica Tropical Lisbon, Portugal Cape Verde Islands

ALASTAIR M. M. RICHARDSON

EDWARD P. F. ROSE

University of Tasmania Hobart, Tasmania, Australia Tasmania

University of London Bucks, United Kingdom Channel Islands (British Isles)

BRUCE RICHMOND

VALENTÍ RULL

U.S. Geological Survey Santa Cruz, California Beaches

Botanic Institute of Barcelona, Spain Pantepui

ROBERT H. RICHMOND

University of Hawaii, Honolulu Reef Ecology and Conservation

Johnson and Wales University Denver, Colorado Fiji, Biology

DAVID L. ROBERTS

PETER G. RYAN

Royal Botanic Gardens, Kew Richmond, Surrey, United Kingdom Orchids

University of Cape Town Rondebosch, South Africa Tristan da Cunha and Gough Island

RICHARD E. A. ROBERTSON

PETER SAENGER

University of the West Indies St. Augustine, Trinidad and Tobago, West Indies Antilles, Geology

Southern Cross University Lismore, New South Wales, Australia Mangrove Islands

BRUCE H. ROBISON

BERNARD SALVAT

Monterey Bay Aquarium Research Institute Moss Landing, California Pocket Basins and Deep-Sea Speciation

University of Perpignan, France French Polynesia, Biology

SARA ROCHA

Scripps Institution of Oceanography La Jolla, California Line Islands

University of Porto Vila do Conde, Portugal Comoros GORDON H. RODDA

U.S. Geological Survey Fort Collins, Colorado Marianas, Biology Snakes GEORGE RODERICK

University of California, Berkeley Invasion Biology BARRY V. ROLETT

University of Hawaii, Honolulu Deforestation JOE ROMAN

University of Vermont, Burlington Whales and Whaling

xxiv

CONTRIBUTORS

PADDY RYAN

STUART SANDIN

ARNOLDO SANTOS-GUERRA

Jardín de Aclimatación de La Orotava Tenerife, Spain Canary Islands, Biology THOMAS W. SCHOENER

University of California, Davis Species–Area Relationship THOMAS A. SCHROEDER

University of Hawaii, Manoa Climate on Islands Hurricanes and Typhoons JENNY SCOTT

University of Tasmania Hobart, Tasmania, Australia Macquarie, Biology

ROGER C. SEARLE

CHRISTOPHER K. STARR

Durham University, United Kingdom Plate Tectonics

University of the West Indies St. Augustine, Trinidad and Tobago, West Indies Trinidad and Tobago

OLE SEEHAUSEN

EAWAG Center for Ecology, Evolution, and Biogeochemistry Kastanienbaum, Switzerland Cichlid Fish

ORLO C. STEELE

SLAHEDDINE SELMI

University of Iceland, Reykjavik Iceland

Faculté des Sciences de Gabès, Tunisia Oases

University of Hawaii, Hilo Solomon Islands, Biology SIGURDUR STEINTHORSSON

WAYNE STEPHENSON

University of Florida, Gainesville Antilles, Biology

University of Melbourne Melbourne, Victoria, Australia Erosion, Coastal

DAVID R. SHERROD

WOLFGANG STERRER

U.S. Geological Survey Vancouver, Washington Hawaiian Islands, Geology

Bermuda Zoological Society Flatts, Bermuda Bermuda

LUÍS SILVA

DOMINIQUE STRASBERG

University of the Azores Vairão, Portugal Azores

University of La Réunion Saint-Denis, Réunion Mascarene Islands, Biology

DANIEL SIMBERLOFF

TOD F. STUESSY

University of Tennessee, Knoxville Introduced Species

University of Vienna, Austria Anagenesis

JOHN M. SINTON

ROBERT C. SUGGS

University of Hawaii, Honolulu Volcanic Islands

Boise, Idaho Kon-Tiki

JOHN L. SMELLIE

GRAHAM SYMONDS

British Antarctic Survey Cambridge, United Kingdom Antarctic Islands, Geology

CSIRO Marine and Atmospheric Research Wembley, Western Australia, Australia Surf in the Tropics

CRAIG R. SMITH

R. R. THAMAN

University of Hawaii, Manoa Organic Falls on the Ocean Floor

University of the South Pacific Suva, Fiji Sustainability

FLORENCE E. SERGILE

SCOTT G. SMITHERS

James Cook University Townsville, Queensland, Australia Great Barrier Reef Islands, Geology NOELEEN SMYTH

CHRISTOPHE THÉBAUD

Paul Sabatier University Toulouse, France Mascarene Islands, Biology

National Botanic Gardens Dublin, Ireland Pitcairn

T. THORDARSON

DAVID A. SPILLER

DAVID TOWNS

University of California, Davis Species–Area Relationship

New Zealand Department of Conservation Newton, Auckland, New Zealand Rodents

University of Edinburgh, United Kingdom Eruptions: Laki and Tambora

CONTRIBUTORS

xxv

STEVEN A. TREWICK

JOHN WAINWRIGHT

Massey University Palmerston North, New Zealand New Zealand, Biology

University of Sheffield, United Kingdom Mediterranean Region

KOSTAS A. TRIANTIS

University of La Réunion Saint-Denis, Réunion Mascarene Islands, Biology

Oxford University, United Kingdom Greek Islands, Biology Radiation Zone FRANK A. TRUSDELL

Hawaiian Volcano Observatory U.S. Geological Survey, Hawaii National Park Marianas, Geology

DAVID M. WATSON

Charles Sturt University Albury, New South Wales, Australia Continental Islands MARSHALL I. WEISLER

Ghent University, Belgium Socotra Archipelago

University of Queensland St. Lucia, Queensland, Australia Archaeology

SOFIE VANDENDRIESSCHE

PAUL WESSEL

Instituut voor Landbouw- en Visserijonderzoek Oostende, Belgium Ephemeral Islands, Biology

University of Hawaii, Manoa Seamounts, Geology

AMY G. VANDERGAST

University of Hawaii, Manoa Ethnobotany

KAY VAN DAMME

U.S. Geological Survey San Diego, California Kı¯puka NANCY VANDER VELDE

Majuro, Marshall Islands Marshall Islands

W. ARTHUR WHISTLER

MICHAEL C. WHITLOCK

University of British Columbia Vancouver, British Columbia, Canada Founder Effects ROBERT J. WHITTAKER

Technical University of Braunschweig, Germany Lizard Radiations

Oxford University, United Kingdom Krakatau Radiation Zone

PHILIPPE VERNON

DAVID M. WILKINSON

University of Rennes 1 Paimpont, France Invasion Biology

Liverpool John Moores University, United Kingdom Ascension

VIRGILIO VIEIRA

Memorial University St. John’s, Newfoundland, Canada Newfoundland

MIGUEL VENCES

University of the Azores Ponta Delgada, Portugal Azores MAGDA VINCX

Ghent University, Belgium Ephemeral Islands, Biology ROBERT C. VRIJENHOEK

Monterey Bay Aquarium Research Institute Moss Landing, California Hydrothermal Vents

xxvi

BEN H. WARREN

CONTRIBUTORS

HAROLD WILLIAMS

MARK WILLIAMSON

University of York, United Kingdom Britain and Ireland EDWARD L. WINTERER

Scripps Institution of Oceanography La Jolla, California Atolls Nuclear Bomb Testing

MARIA WŁODARSKA-KOWALCZUK

S. YANAI

Institute of Oceanology, Polish Academy of Sciences Sopot, Poland Spitsbergen

Japan Communications, Co. Ltd. Tokyo, Japan Japan’s Islands, Geology

CHARLES A. WOODS

MAN-MIAO YANG

University of Vermont, Island Pond Antilles, Biology

National Chung Hsing University Taichung, Taiwan Taiwan, Biology

JOHN M. WORAM

Rockville Centre, New York Tierra del Fuego TREVOR H. WORTHY

University of Adelaide Unley, Adelaide, Australia Fossil Birds STEVE WRATTEN

JUN YOKOYAMA

Yamagata University, Japan Japan’s Islands, Biology GRACIANO YUMUL, JR.

University of the Philippines Diliman, Quezon City, Philippines Philippines, Geology

Lincoln University Christchurch, New Zealand Biological Control

CONTRIBUTORS

xxvii

This page intentionally left blank

GUID E TO TH E ENCYC LO P EDI A

The Encyclopedia of Islands is a comprehensive, complete, and authoritative reference dealing with all of the physical and biological aspects of islands and island habitats. Articles are written by researchers and scientific experts and provide a broad overview of the current state of knowledge on these fascinating places. Biologists, ecologists, geologists, climatologists, oceanographers, geographers, and zoologists have contributed reviews intended for students as well as the interested general public. In order for the reader to easily use this reference, the following summary describes the features, reviews the organization and format of the articles, and is a guide to the many ways to maximize the utility of this Encyclopedia.

Article titles have been selected to make it easy to locate information about a particular topic. Each title begins with a key word or phrase, sometimes followed by a descriptive term. For example, “Seamounts, Biology” is the title assigned rather than “Biology of Seamounts,” because seamounts is the key term and, therefore, more likely to be sought by readers. Articles that might reasonably appear in different places in the Encyclopedia are listed under alternative titles—one title appears as the full entry; the alternative title directs the reader to the full entry. For example, the alternative title “Darwin’s Finches” refers readers to the entry entitled Galápagos Finches. ARTICLE FORMAT

SUBJECT AREAS

The Encyclopedia of Islands includes 236 topics that review the various ways scholars have studied islands. The Encyclopedia comprises the following subject areas: • Geography • Island Types • Important Islands • Geology • Geologic Processes • Biogeography • Ecology and Evolution • Oceanography and Climatology • Plants and Animals • Human Impact • History and Prehistory ORGANIZATION

Articles are arranged alphabetically by title. An alphabetical table of contents begins on page v, and another table of contents with articles arranged by subject area begins on page xi.

The articles in the Encyclopedia are all intended for the interested general public. Therefore, each article begins with an introduction that gives the reader a short definition of the topic and its significance. Here is an example of an introduction from the article “Biological Control”: Biological control (or biocontrol) is the use of natural enemies to suppress pest species populations to less damaging densities. When certain native and introduced invertebrates, plants, pathogens and vertebrates increase in abundance and become pests through human influence or through other causes, economic crop damage and threats to natural resources are likely. Islands are particularly vulnerable to pest outbreaks. With high endemism, low species diversity, small land areas, and a history less affected by forces that develop adaptability compared to continents, islands are more susceptible to the effects of habitat changes and species introductions. The enhancement of the efficacy of the natural enemies of pest organisms is a potentially more environmentally sound and sustainable control option than chemical or mechanical control strategies.

xxix

Within most articles, especially the longer ones, major headings help the reader identify important subtopics within each article. The article “Cook Islands” includes the following headings: “Geology and Geomorphology,” “Climate and Weather,” “Biological Resources,” “Settlement, History, and Culture,” and “Environmental Issues.”

especially new and newsworthy. Thus, the reader interested in delving more deeply into any particular topic may elect to consult these secondary sources. The Encyclopedia functions as ingress into a body of research only summarized herein. GLOSSARY

CROSS-REFERENCES

Many of the articles in this Encyclopedia concern topics for which articles on related topics are also included. In order to alert readers to these articles of potential interest, cross-references are provided at the conclusion of each article. At the end of “Drosophila,” the following text directs readers to other articles that may be of special interest:

Almost every topic in the Encyclopedia deals with a subject that has specialized scientific vocabulary. An effort was made to avoid the use of scientific jargon, but introducing a topic can be very difficult without using some unfamiliar terminology. Therefore, the contributors were asked to define a selection of terms used commonly in discussion of their topics. All these terms have been collated into a glossary at the back of the volume after the last article. The glossary in this work includes over 900 terms.

SEE ALSO THE FOLLOWING ARTICLES

Flightlessness / Founder Effect / Hawaiian Islands, Biology /

INDEX

Insect Radiations / Sexual Selection

The last section of the Encyclopedia of Islands is a subject index consisting of more than 3,200 entries. This index includes subjects dealt with in each article, scientific names, topics mentioned within individual articles, and subjects that might not have warranted a separate standalone article.

Readers will find additional information relating to Drosophila in the articles listed. BIBLIOGRAPHY

Every article ends with a short list of suggestions for “Further Reading.” The sources offer reliable in-depth information and are recommended by the author as the best available publications for more lengthy, detailed, or comprehensive coverage of a topic than can be feasibly presented within the Encyclopedia. The citations do not represent all of the sources employed by the contributor in preparing the article. Most of the listed citations are to review articles, recent books, or specialized textbooks, except in rare cases of a classic ground-breaking scientific article or an article dealing with subject matter that is

xxx

G U I D E TO T H E E N C YC LO P E D I A

ENCYCLOPEDIA WEBSITE

To access a website for the Encyclopedia of Islands, please visit: http://www.ucpress.edu/books/pages/10384.php

This site provides a list of the articles, the contributors, several sample articles, published reviews, and links to a secure website for ordering copies of the Encyclopedia. The content of this site will evolve with the addition of new information.

P REFACE

Islands are peaceful and tranquil, and also mysterious and intriguing—paradigm, paradox, paradise, and prison—and as such have inspired scientists, writers, and painters for centuries. Metaphorically, writers have used the idea of “island” over and over—islands in the sky, an island of discontent, an island of sanity, and a tropical island of mystery—Robinson Crusoe, Lord of the Flies, Joe vs. the Volcano. Everybody knows what an island is. More prosaically, scholars generally define an island as an isolated piece of habitat that is surrounded by dramatically different habitat, such as water. A high moist mountain surrounded by a dry desert, a patch of fertile soil in a “sea” of jumbled rocks, or the corpse of a whale at the bottom of the ocean—all of these are also islands. Islands have been studied precisely because isolation, and many biological ideas about isolation, have been so essential in uncovering biological patterns and processes. Darwin’s theory of evolution might not have emerged in the way it did had he not experienced the reality and significance of islands. Scale and context matter. A lake in the middle of a continent may be isolated for a fish, but not so much for a horse; an oceanic island may be isolated for a frog or an insect, but not for a whale; and a rock in the middle of a forest may be isolated for an ant, though not for a pig. The Encyclopedia of Islands is meant to pull together all the traditional ideas of islands, as well as the biological and geological concepts that have been erected around islands as isolated locations. Isolation can be achieved in many ways, but most often through some geological process. New volcanoes grow from the sea floor to create islands; global plate tectonics fragment continents, recombine them in different configurations, and fragment them again. This book examines the geologic and other

processes that create isolation, and then looks at the biologic consequences of isolation. The extent of biological isolation depends on the ability of organisms to move and disperse as well as their ability to withstand terrain that is inhospitable. In visualizing biological islands, it becomes clear that almost anything can serve as an island—a water-filled tree hole for many invertebrates; a human body for the parasites it contains; a crack in the sidewalk for a weed. Given the astonishingly broad importance of the idea of “island,” our purpose is to cover a range of topics in sufficient detail to give the reader a general understanding of guiding principles, and lead them to the vast literature on island science. Circumstantially, this reference might be thought of as a collection of islands, each article being an island itself. Our hope is that readers of this reference will “travel” from island to island, from topic to topic, and learn the myriad ways that the idea of “island” has contributed to a remarkable scientific voyage whose destination will always lie just over the horizon. We wish to thank the many busy scientists who took the time and went to the effort to create the individual articles that constitute this volume. Many of our authors helped define the book by suggesting additional topics and proposing other potential authors. We also thank the staff at the University of California Press for their assistance in preparing this volume. In particular, Chuck Crumly originated the idea of an encyclopedia of islands, convinced us we were the right scientists to assemble it, and oversaw the developmental stages. Gail Rice expertly managed the manuscripts and revisions and persisted in extracting articles from even the most recalcitrant of contributors. Without her efforts there would be no Encyclopedia of Islands. We finally thank our families—George,

xxxi

William, and Melrose; Andrea and Gillian—for their patience, understanding, and encouragement as our time and thoughts were diverted to islands around the world and of all types. Rosemary G. Gillespie University of California, Berkeley David A. Clague Monterey Bay Aquarium Research Institute Moss Landing, California

xxxii

P R E FA C E

A ADAPTIVE RADIATION

contribute to the emergence of novel evolutionary changes. Key Innovations

ROSEMARY G. GILLESPIE University of California, Berkeley

Adaptive radiation is one of the most important outcomes of the process of evolution, and islands are places where it is best observed. The term itself was first used by H. F. Osborn in 1902 in describing parallel adaptations and convergence of species groups on different land masses. Since then, adaptive radiation has been widely recognized and defined in multiple ways to emphasize the contribution of key features thought to underlie the phenomenon, including adaptive change, speciation within a monophyletic lineage, and time. Dolph Schluter, a prominent researcher in the field, defines it as “the evolution of ecological diversity within a rapidly multiplying lineage.” There are radiations that are not adaptive. These are mostly caused by changes in topography that, instead of opening up new habitats, have served simply to isolate a previously more widespread species. Island radiations are likely to include adaptive as well as non-adaptive evolutionary elements. In this article, the focus is on adaptive evolutionary radiations.

Key innovations are, for example, adjustments in morphology/physiology that are essential to the origin of new major groups, or features that are necessary, but not sufficient, for a subsequent radiation. It appears that the new trait can enhance the efficiency with which a resource is

FIGURE 1 Adaptive radiation of weevils, genus Cratopus (Coleoptera,

FACTORS UNDERLYING ADAPTIVE RADIATION

Curculionidae), on the small island of La Réunion. The southwest-

The most widely recognized “trigger” for adaptive radiation is the opening up of ecological space. This may occur following evolution of a key innovation or when climatological or geological changes lead to the appearance of novel environments. Islands of all sorts provide examples of such dynamic environments, which

ern Indian Ocean is a hotspot of terrestrial diversity and endemism. The genus Cratopus has undergone intense diversification within the islands of the Mascarenes, including a total of 91 species, of which 43 are present on La Réunion (21 endemic), 36 on Mauritius (28 endemic), and five on Rodrigues (all endemic). Research is currently under way to elucidate the nature of diversification in this group. Photographs by Ben Warren, Christophe Thébaud, Dominique Strasberg, and Antoine Franck, with permission.

1

utilized and hence can allow species to enter a “new” adaptive zone in which diversification occurs. For example, with the origin of powered flight in birds, new feeding strategies, life history patterns, and habitats became available. Likewise, the origin of jaws in vertebrates allowed rapid diversification of predatory lineages. However, key innovations only set the stage for changes in diversity; they do not, by themselves, cause the change. Key innovations can occur over and over, such as heterospory in plants. They have been extensively implicated in the adaptive radiation of interacting species. For example, symbioses can be an “evolutionary innovation,” allowing the abrupt appearance of evolutionary novelty, providing a possible avenue through which taxonomic partners can enter into a new set of habitats unavailable to one or both of the symbiotic partners alone. The development of toxicity in plants (which can allow them to “escape” the predatory pressure of insects) can lead to a subsequent development of tolerance to the toxin in insects, allowing the insects to radiate onto the plants. Paul Ehrlich and Peter Raven envisioned this as a steplike process in which the major radiations of herbivorous insects and plants have arisen as a consequence of repeated openings of novel adaptive zones that each has presented to the other over evolutionary history. This idea, termed the “escalation/diversification” hypothesis, has been supported by studies of phytophagous beetles. A recent study in butterflies has shown that a key innovation (evolution of the NSP glucosinolate detoxification gene) has allowed butterflies (Pierinae) to radiate onto plants (Brassicales) that had temporarily escaped herbivory from the caterpillars.

FIGURE 2 Adaptive radiation of Hawaiian lobeliad plants. The very

large and spectacular radiation of plants (125 species in six genera) appears to have occurred together with that of its pollinators, the Hawaiian honeycreepers. Species shown are (A) Trematolobelia macrostachys, Poamoho Trail, Ko‘olau Mountains, O‘ahu; (B) Clermontia samuelii, Upper Hana rain forest, East Maui; (C) Cyanea koolauensis, Ko‘olau Mountains, O‘ahu; (D) Trematolobelia macrostachys, Wai’anae Mountains, Ka‘ala, O‘ahu; (E) Clermontia kakeana, O‘ahu; (F) Cyanea

2

Novel Environments

leptostegia, Koke‘e, Kaua‘i; (G) Brighamia insignis, Kaua‘i (photograph

When a new habitat forms close to a source, it is generally colonized by propagules from the adjacent mainland. However, if the new habitat is isolated, as in the case of the formation of islands in the ocean, then colonization may be very slow, to the extent that evolution may contribute to the formation of new species more rapidly than does colonization. It is under such conditions that some of the most remarkable adaptive radiations have occurred (e.g., Fig. 1). Geological history is punctuated with episodes of extinction, presumably induced by catastrophic environmental changes, with each extinction episode setting the stage for the subsequent adaptive radiation of new (related or unrelated) groups. Environmental change has, for example, been implicated in Phanerozoic revolutions and the repeated radiations of ammonoids throughout the geological record, in which each radiation appears to have originated from a few taxa

by H. St. John, courtesy of Gerald D. Carr). All photographs by Gerald

A D A P T I V E R A D I AT I O N

D. Carr, with permission, except where noted.

that went on to give rise to morphologically diverse lineages. Some of the best examples of recent, or ongoing, adaptive radiations come from isolated islands, including oceanic archipelagoes, continental lakes, and mountaintops. Galápagos island finches, Hawaiian honeycreepers, Madagascar lemurs, and African cichlids are frequently cited as some of the best examples of adaptive radiation. The finches are well known because of their historical importance, being a focus of discussion by Darwin in the Origin of Species. Thirteen species of Galápagos finch (Geospizinae, Emberizidae) have diversified on the Galápagos Islands, with each filling a different niche on a given island. The clear similarity among species in this radiation, coupled with the dietary-associated modifications in beak morphologies, played a key role in

the development of Darwin’s theory of evolution through natural selection. Research on the finches has continued to illuminate evolutionary principles through the work of Rosemary and Peter Grant. Hawaiian honeycreepers (Drepaniinae, Emberizidae) also originated from a finchlike ancestor and diversified into approximately 51 species in three tribes, with extraordinary morphological and ecological diversity. However, a large proportion of these species (about 35) are now extinct.

Isolation is a relative phenomenon. Accordingly, islands that are isolated for mammals and lizards are less so for many insects, with mammal radiations known in less isolated islands such as the Philippines (radiation of Old World mice/rats, Murinae), and lizards in the Caribbean (Anolis diversification in the Greater Antilles); prior to human transportation, few vertebrates, except for birds and an occasional bat, succeeded in colonizing more remote islands. However, the most remote islands of the

A D A P T I V E R A D I AT I O N

3

central Pacific are well known for some of the most spectacular radiations of arthropods and plants. Examples of Insular Radiation

Some adaptive radiations are perhaps best known because of the number and rate of species that have been produced. One of the most spectacular in this regard is the Hawaiian Drosophilidae, with an estimated 1000 species divided into two major lineages (Hawaiian Drosophila and Scaptomyza). However, the cricket genus Laupala has been documented as having the highest rate of speciation so far recorded in arthropods, with 4.17 species per million years, supporting the argument that divergence in courtship or sexual behavior can drive rapid speciation in animals. Very rapid rates of speciation have also been reported for the African cichlid fish, with an estimated 2.02–2.09 species being formed per million years in Lake Malawi and Lake Victoria. Here again, sexual selection is implicated in the diversification of the group. A recent study, which awaits explanation, is that of Hawaiian bees (Hylaeus, Colletidae), which indicates that the 60 species known to occur on the islands originated and radiated on the island of Hawaii less than 700,000 years ago. Most island adaptive radiations occur within an archipelago setting, and the argument is often made that different islands are necessary to provide sufficient isolation for adaptive radiation to take place. Interesting in this regard is a radiation of small flightless weevils in the genus Miocalles (Coleoptera: Curculionidae: Cryptorhynchinae) on the single small island of Rapa in the southern Australs of French Polynesia. Rapa is home to almost half of the 140 species that occur across the western Pacific and Australia. Here, the beetles collectively feed on 24 genera of native plants and show varying degrees of host specificity, thus utilizing almost all genera of native plants found on Rapa. Some radiations have been studied because they are readily accessible to scrutiny of the process of adaptive radiation, in part because they have occurred very recently. In this context, research of Dolph Schluter has focused on stickleback fish of the deglaciated lakes of coastal British Columbia, Canada. These lakes harbor a number of sibling species of fish, and repeated co-occurrence of pairs of species has been attributed to novel ecological opportunities provided by deglaciation and recolonization. In particular, the threespine stickleback, Gasterosteus aculeatus, is a species complex that has diversified in each lake such that no more than two species occur in any one lake. Interestingly, it appears that pairs of species in different lakes have evolved independently

4

A D A P T I V E R A D I AT I O N

of other pairs, and species have diverged as a result of parallel episodes of selection for alternate feeding environments. Research in this system has highlighted the role of divergent natural selection as a mechanism underlying adaptive radiation. Among plants, the largest known radiation is that of the Hawaiian lobeliads (Brighamia, Cyanea, Clermontia, Delissia, Lobelia, and Trematolobelia—Campanulaceae), with the more than 100 species now thought to have arisen from a single colonization event (Fig. 2). The radiation exhibits extraordinary diversity in vegetative and flower morphology, with species inhabiting a huge array of habitats. The diversity is considered to have arisen in concert with that of the Hawaiian honeycreepers, the lobeliads displaying a suite of morphological characteristics associated with bird pollination, including deep, tubular, longlived inflorescences, with an abundance of nectar and no odor. Another spectacular example of adaptive radiation in plants is the Hawaiian silversword alliance, Argyroxiphium, Dubautia, Wilkesia (Asteraceae–Madiinae), in which 28 species are known that display a huge diversity in life form, from trees to shrubs, mats, rosettes, cushions, and vines, occurring across habitats from rain forests and bogs to desert-like settings. An analogous radiation of 23 species in the genus Argyranthemum (Asteraceae– Anthemideae) has occurred in the Macaronesian islands, although the largest radiation of plants in the Canaries is that of the succulent, rosette-forming species of the genus Aeonium (Crassulaceae). Predisposition to Adaptive Radiation

Are certain taxa predisposed to adaptive radiation? Some have suggested that plant-associated insects are constrained by their narrow host range, which prevents adaptive diversification. However, this argument is not well supported, as multiple insects with specialized host affinities have succeeded in colonizing remote islands and have also undergone some of the most spectacular adaptive radiations. Based on the available information, there is no a priori means of predicting whether or not a species will undergo adaptive radiation upon being provided an ecological opportunity that it can exploit. At the same time, some lineages do show multiple independent episodes of adaptive radiation; for example, Hawaiian silverswords have a parallel sister radiation of California tarweeds, and one sister group, the shrubby tarweeds (Deinandra), has undergone adaptive radiation on the California Channel Islands (Fig. 3). Hawaiian long-jawed spiders (genus Tetragnatha) have undergone independent radiations on different archi-

pelagoes of the Pacific, presumably because they are adept at overwater dispersal and readily adapt to insular environments. In general, although there may be a substantial random element to colonization, successful colonization of very isolated locations requires high dispersal abilities, so representation of taxa within biotas in isolated areas will be skewed toward those with high dispersal abilities. However, subsequent establishment of the initial colonists on remote islands is frequently associated with a dramatic loss of dispersal ability and/or attainment of a more specialized habitat. Indeed, loss of dispersal ability is implicated as a key factor in allowing diversification to proceed. PROCESS OF ADAPTIVE RADIATION Ecological Release

Ecological release has been inferred to occur at the outset of adaptive radiation. This sort of release is the expansion of range, habitat, and/or resource usage by an organism when it reaches a community from which competitors, predators, and/or parasites may be lacking. Indeed, regular cycles of ecological and distributional expansion following colonization of islands are well known and have been documented in a number of groups, with some showing subsequent shifts toward specialization, as part of the phenomenon of “taxon cycles.” Adaptive Plasticity

Behavioral and ecological plasticity have recently been implicated as playing a key role in adaptive radiation. Although initially thought to impede adaptive diversification because it allows a single taxon to exploit a broad environmental range without requiring evolutionary shifts, recent work by Mary Jane West-Eberhard indicates that adaptive plasticity (including behavior) can promote evolutionary shifts, in particular when the environment is variable. Accordingly, plasticity in Caribbean Anolis lizards may allow species to occupy new habitats in which they otherwise might not survive. Once in these habitats, selection may act to accentuate attributes that allowed them to live in this habitat. Recent studies of threespine stickleback, in which ancestral oceanic species have changed little since colonization and diversification of freshwater species, have lent support to the importance of behavioral plasticity in allowing adaptive radiation to proceed. An interesting example of how natural selection can promote plasticity, potentially leading to species radiation, is found in pitohui birds in New Guinea (Fig. 4).

FIGURE 3 Radiation

of the shrubby tarweed genus Deinandra

on Guadalupe Island, Mexico. Guadalupe, the highest and most remote of the California Islands, is home to three endemic taxa of shrubby tarweeds (Deinandra). This small radiation in the California Islands parallels diversification of the sister lineage, the Hawaiian silversword alliance (Compositae-Madiinae), indicating the propensity of this group of plants to diversify in isolated settings. From B. G. Baldwin, 2007.

Speciation in Adaptive Radiation

A key feature of adaptive radiation is rapid speciation coupled with phenotypic diversification. This introduces an interesting paradox in that adaptive radiations are simultaneously characterized by minimal genetic diversity (very small numbers of individuals involved in the initial colonization) and very high morphological/ecological/behavioral diversity. Given the circumstances, a number of processes have been implicated as operating together to allow adaptive radiation to occur. Founder Events

A founder event occurs when a new population is composed of only a few colonists, inevitably carrying only a small sample of the genetic diversity of the parent population. This small population size means that the colony may have reduced genetic variation and a non-random sample of the genes relative to the original population. Many studies have suggested that founder events play a role in adaptive radiation, as taxa within a radiation are generally characterized by small population sizes with ample opportunity for isolation. This could potentially lead to a cascade of genetic changes leading to evolutionary differentiation, an idea first formulated in the “genetic revolution” model,

A D A P T I V E R A D I AT I O N

5

are formed through adaptive shifts is generally considered to require competition between similar taxa. The ecological theory of adaptive radiation suggests that speciation and the evolution of morphological and ecological differences are caused by divergent natural selection resulting from interspecific competition coupled with environmental differences. Recent studies with walking stick insects and bacteria have suggested that predation may operate with (or instead of ) competition to allow divergent natural selection. Hybridization and Gene Flow FIGURE 4 Example of morphological plasticity in a single clade show-

ing how natural selection might lead to rapid diversification under strong natural selection. Bird species in the genus Pitohui, endemic to the island of Papua New Guinea, are chemically defended by a potent neurotoxic alkaloid in their skin and feathers. It appears that they cannot produce the toxin themselves but rather rely on eating a melyrid beetle and sequestering the beetle toxin. The two most toxic species are the hooded pitohui (P. dichrous) and the variable pitohui (P. kirhocephalus). Pitohui kirhocephalus is considered to be a single species based on the similarity of its members’ songs and habits, because of clinal variation between certain races, and because no two races have been found in sympatry; Pitohui dichrous, in contrast, shows little geographic variation throughout its range. However, the

A traditional argument has been that gene flow among diverging populations, or hybridization between incipient species, acts to slow the process of diversification. However, recent research suggests that divergence between lineages can be increased through moderate levels of gene flow, a phenomenon termed “collective evolution.” Likewise, interspecific hybridization may be a possible source of additional genetic variation within species, increasing the size of the gene pool on which selection may act.

two species are virtually identical in color pattern in many areas of co-occurrence. It has been suggested that Müllerian mimicry is driv-

Sexual Selection

ing the similar color patterns between the “mimetic” P. kirhocephalus phenotype and P. dichrous. The map shows P. kirhocephalus subspecific ranges and phenotypes; letters indicate the subspecies, in which certain phenotypes (ranges shown in orange and green in b, c, f, h, i, p, q) are thought to be potential mimics of P. dichrous. From Dumbacher and Fleischer (2001).

which posits that the reduced levels of heterozygosity following founder events affect the nature of co-adapted gene complexes. However, the precise role of founder events remains unclear: During the bottleneck, (1) a large proportion of alleles is lost, and few new mutations can occur with the population at small size; (2) genetic change will occur through drift, but the effect becomes weaker as the population starts to grow. As a result of these opposing forces, the number of beneficial mutations fixed per generation will change little because of the bottleneck. However, subsequent to a genetic bottleneck, selection can preserve alleles that are initially rare and that would otherwise tend to be lost through stochastic events. Founder events have been implicated in the adaptive radiation of such large groups as Hawaiian Drosophila and other insects, but there is little empirical evidence to support their role in species formation among vertebrates. Divergent Natural Selection

Although adaptive radiation is generally associated with reduced competition, the process through which species

6

A D A P T I V E R A D I AT I O N

Sexual selection has been linked to the diversification of species within some of the most rapid adaptive radiations. In particular, although ecological diversification still plays a role, it appears that sexual selection may drive species proliferation in African haplochromine cichlids, Hawaiian Drosophila flies, Laupala crickets, and Australasian birds of paradise. In each case, female choice is implicated in driving speciation: Males (rather than females) exhibit striking colors (in the case of the cichlids), demonstrate complex mating systems that involve modifications of the mouthparts, wings, and/ or forelegs with associated elaborate behaviors and leks which are visited by females (in the flies), and have distinct courtship songs (in the crickets). Likewise, diversification in jumping spiders in the sky islands of the western United States appears to be the product of female preference for greater signal complexity or novelty. COMMUNITY ASSEMBLY

Communities on an island are generally assembled by the interplay between colonization of species from the same niche in another region (e.g., on a mainland source or another island) or, in cases where isolation means that the number of available colonists are insufficient to fill a community, by adaptive shifts from one niche to

allow a taxon to exploit a new niche. Accordingly, during the course of adaptive radiation, speciation appears to play a role similar to that of immigration—although over an extended time period—in adding species to a community. One striking aspect of adaptive radiation is the role of convergent evolution in similar habitats, and the associated parallel evolution of similar ecological forms, resulting in the production of strikingly predictable communities during the course of diversification. Some of the best examples here are the Anolis lizards of the Caribbean, where studies by Jonathan Losos and colleagues have shown that on the islands of the Greater Antilles (Cuba, Hispaniola, Jamaica, and Puerto Rico) multiple species co-occur. Each species can be recognized as an “ecomorph,” occupying a characteristic microhabitat (e.g., tree twigs, grass, tree trunks) with corresponding morphological and behavioral attributes named for the part of the habitat they occupy. Most remarkably, similar sets of ecomorphs are found on each island and have generally arisen through convergent evolution on each island, showing that similar communities on different islands evolved independently. Studies have now demonstrated similar patterns of multiple convergences and independent evolution of similar sets of ecomorphs among multiple island settings—for example, spiders in the Hawaiian Islands, cichlid fishes of the African Rift lakes, and Madagascan and Asian ranid frogs. The phenomenon of repeated evolution of similar forms among species undergoing adaptive radiation has led to research on the molecular basis of such apparently complex changes. Increasingly, these studies are indicating that rather small developmental shifts may lead to large shifts in morphology (Fig. 5). Accordingly, very minor developmental shifts may allow some very striking morphological and ecological shifts, which can readily be lost or gained. SEE ALSO THE FOLLOWING ARTICLES

Ecological Release / Founder Effects / Radiation Zone / Sexual Selection / Taxon Cycle FURTHER READING

Baldwin, B. G. 2007. Adaptive radiation of shrubby tarweeds (Deinandra) in the California Islands parallels diversification of the Hawaiian silversword alliance (Compositae-Madiinae). American Journal of Botany 94: 237–248. Carlquist, S., B. G. Baldwin, and G. D. Carr, eds. 2003. Tarweeds and silverswords: evolution of the Madiinae (Asteraceae). St. Louis: Missouri Botanical Garden Press. Dumbacher, J. P., and R. C. Fleischer. 2001. Phylogenetic evidence for colour pattern convergence in toxic pitohuis: Müllerian mimicry in birds? Proceedings of the Royal Society Biological Sciences 268: 1971–1976.

FIGURE 5 Diversity of scarab beetle “horns,” used as weapons in

males for access to mates. Dung beetles (Scarabaeinae) and rhinoceros beetles (Dynastinae) are both common inhabitants of islands, and the dung pats that the beetles use can themselves be considered ephemeral islands of nutrients. Sample taxa illustrated are (A) Dynastes hercules (Dynastinae) from the New World tropics, including different subspecies on the islands of the Lesser Antilles; (B) representative of the genus Golofa (G. porteri) (Dynastinae), a neotropical lineage known for its “Mesoamerican mountaintop” pattern of biogogeography; (C) Allomyrina dichotoma (Dynastinae) from the Old World tropics, including Japan, Taiwan, and associated islands; (D) Proagoderus tersidorsis (Scarabaeinae) from South Africa; (E) Onthophagus nigriventris (Scarabaeidae: Scarabaeinae) from the East Africa highlands, and now in Australia and Hawaii. The huge diversity of horns appears to result from subtle changes in the relative activities of different developmental pathways. Photograph montage by Douglas J. Emlen, with permission.

Ehrlich, P. R., and P. H. Raven. 1964 Butterflies and plants: a study in coevolution. Evolution 18: 586–608. Emlen, D. J., Q. Szafran, L. S. Corley, and I. Dworkin. 2006. Insulin signaling and limb-patterning: candidate pathways for the origin and evolutionary diversification of beetle “horns.” Heredity 97: 179–191. Givnish, T. J., and K. J. Sytsma. 1997. Molecular evolution and adaptive radiation. Cambridge: Cambridge University Press. Grant, P. R., and B. R. Grant. 2007. How and why species multiply: the radiation of Darwin’s finches. Princeton, NJ: Princeton University Press. Losos, J. B. 2009. Lizards in the evolutionary tree: the ecology of adaptive radiation in Anoles. Berkeley: University of California Press. Ricklefs, R. E., and E. Bermingham. 2007. The causes of evolutionary radiations in archipelagoes: passerine birds in the Lesser Antilles. American Naturalist 169: 285–297. Schluter, D. 2000. The ecology of adaptive radiation. Oxford: Oxford University Press. Seehausen, O. 2006. African cichlid fish: a model system in adaptive radiation research. Proceedings of the Royal Society B—Biological Sciences 273: 1987–1998. West-Eberhard, M. J. 2003. Developmental plasticity and evolution. New York: Oxford University Press.

ALEUTIAN ISLANDS SEE PACIFIC REGION

A D A P T I V E R A D I AT I O N

7

AMSTERDAM SEE INDIAN REGION

ANAGENESIS TOD F. STUESSY University of Vienna, Austria

Anagenesis is a process of gradual speciation whereby only single endemic species within respective genera diverge within oceanic islands (Fig. 1). In this case, the founding immigrant population does not dramatically and rapidly change and split (cladogenesis) into two or more different species after arrival, which is what happens during the better-known pattern of adaptive radiation. THE PROCESS OF ANAGENETIC SPECIATION

The process of speciation through anagenesis is very different from that via cladogenesis (resulting in adaptive radiation). During anagenesis an immigrant population establishes in a suitable new oceanic island habitat. In the absence of ecological opportunity, such as may occur on an island with low elevation, the pioneer population proliferates but maintains genetic cohesiveness because of the absence of geographical and ecological isolating barriers. Genetic variation begins to accumulate throughout the growing population as a result of mutation and recombination. Through time and genetic drift, enough variation accumulates such that the island population (or

Anagenesis

Cladogenesis

B

anagenetic change

B C D

A

immigrant population

A

adaptive radiation

immigrant population

A

A

ancestral continental population

ancestral continental population

FIGURE 1 Diagrammatic contrast between anagenetic and cladoge-

netic speciation in oceanic islands.

8

ANAGENESIS

metapopulation) appears morphologically and genetically distinct at the specific level from continental relatives. The level of morphological divergence is not great, as selection has not been directional and intense. Despite initial genetic reduction due to founder effect in the immigrant population, over time the level of genetic variation within the island metapopulation may approximate that of the continental progenitor. This is in marked contrast to the genetic consequences of cladogenesis and adaptive radiation, whereby the amount of genetic divergence between species is minimal, accompanied, however, by dramatic morphological differences. EVIDENCE FOR ANAGENESIS IN ISLAND ARCHIPELAGOES

Evidence from endemic vascular plants of Ullung Island, Korea, provides a good example for the process of anagenesis. Ullung is a single low island 130 km east of South Korea in the Eastern (Japan) Sea, just under 1000 m elevation and with only moderate vegetational zonation. Of the 23 endemic vascular plant taxa found on Ullung Island, most are alone in their respective genera. This suggests that these endemic species have originated by anagenetic speciation. They clearly do not belong to large, adaptively radiated species complexes. Recent molecular studies in Ullung Island of the endemic and anagenetically derived Dystaenia takesimana (Apiaceae) from its Japanese progenitor, D. ibukiensis, provide further insight on the genetic components of anagenetic speciation. The advantage of this pair of species is that (1) they are the only two species in the genus, which excludes complex origins (e.g., via hybridization) with other potential relatives; and (2) Ullung island is only 1.8 million years old, much younger than the island of Japan, and therefore it is more likely that D. takesimana was derived from D. ibukiensis rather than the reverse. A deletion in one of the DNA sequences (trnL-F ) in the chloroplast genome of the island endemic species, which affects the secondary structure of the molecule, also supports this interpretation. Surprisingly, the levels of genetic variation are similar in both species. This suggests that the colonizing populations may have passed through a genetic bottleneck as a result of the original founder effect, but that subsequently the level of genetic variation built back up to that of the progenitor. FREQUENT OCCURRENCE OF ANAGENESIS IN ISLAND SYSTEMS

Ullung Island is, no doubt, an exceptional case with very high levels of endemic species having been derived

TABLE 1

Features of Island Systems, Numbers of Endemic Species, and Estimated Levels of Anagenetic vs. Cladogenetic Speciation in Selected Continental and Oceanic Islands/Archipelagoes Distance Number

from

No.

Anagenetic

Island

of

Size

Mainland

Age (million

Elevation

Vegetation

Endemic

Speciation

Cladogenetic Speciation

System

Islands

(km2)

(km)

years)

(m)

Heterogeneitya

Species

(%)

(%)

8 7 4 3 12 16 3 12 1 1

16885 7601 208 100 4033 7847 792 99 123 73

3660 100 2580 600 570 930 630 800 1850 130

5 21 18 4 23 5 14 Tertiary 15 2

4250 3710 2060 1319 2829 1707 1862 916 826 984

6 6 3 5 2 4 3 3 2 2

828 429 27 97 68 133 96 118 36 33

7 16 33 36 37 43 48 53 53 88

93 84 67 64 63 57 52 47 47 6

1 1 2

35800 963 8500

130 300 410

5 80 Tertiary

3950 294 705

3 1 2

724 37 14

29 62 71

71 38 29

Oceanic

Hawaii Canary Tristan da Cunha Juan Fernandez Cape Verde Galapagos Madeira Ogasawara St. Helena Ullung Continental

Taiwan Chatham Falkland

note: From Stuessy et al. 2006. a The higher the value, the greater the vegetation heterogeneity (vegetation zones). For calculations, refer to Stuessy et al. (2006).

by anagenesis (at least 88% of endemic species). It is of interest, therefore, to examine other islands/archipelagoes of the world, of both oceanic and continental origins, to determine relative levels of anagenesis and cladogenesis (Table 1). Results show very different levels of both processes, the lowest level of anagenesis being in the Hawaiian archipelago (7%) and the highest in Ullung Island (88%). The levels of anagenesis or cladogenesis in different islands/archipelagoes clearly relate to two factors: (1) elevation of the island and (2) habitat heterogeneity. In islands that are high in elevation and with strong and diverse vegetational zones, the highest levels of cladogenesis occur. In contrast, low islands with few vegetation zones show the highest level of anagenetically derived species. Likewise, in continental islands also with high elevation and habitat diversity, higher levels of cladogenesis are also seen (Table 1). These levels do not relate directly to size of island, age, or distance from mainland source areas. THE EVOLUTIONARY IMPORTANCE OF ANAGENESIS

The importance of anagenesis is that it represents another major model for speciation in oceanic islands. Cladogenetic examples involving adaptive radiation will continue to capture our imagination by virtue of their dramatic

natures, but this process does not explain all endemic species in all islands. In fact, for some islands, anagenesis is fundamental for explaining patterns of diversity. We have estimated conservatively that 25% of all island species have originated by anagenesis. An important related point is that the genetic and specific diversity of species originating from anagenesis or cladogenesis will not only be different but will also vary over time (Fig. 2). Oceanic islands have short existences, often enduring little more than 6 million years before they erode and subside under the sea. Depending upon the stage of island ontogeny, the existing genetic and specific diversity will vary, starting with little variation, peaking in mid-life of the island, and declining as the size and ecological diversity of the island diminishes through time. In making comparisons of genetic variation in populations between islands, therefore, as well as between those of continental regions and islands, we need to keep the island ontogeny firmly in mind. It makes little sense to compare implications of the founder effect, for example, between populations in continental regions and those on two different islands of very different ages. Likewise, it is important to know whether the genetic variation being analyzed occurs in a species having evolved cladogenetically or anagenetically, as this can also help suggest reasons for the observed patterns.

ANAGENESIS

9

high

an ag en es is

ANTARCTIC ISLANDS, BIOLOGY STEVEN L. CHOWN AND JENNIFER E. LEE Stellenbosch University, South Africa

genes clado

is

low

Sp ec i f i c ( — ) an d g en et i c ( ---- ) d i v er s i t y

A

Arrival and establishment (0 - 10,000 y)

Early development (10,000 - 3 my)

Maturation (3 - 5 my)

Senescence and extinction (5 - 6 my)

high

B

s ne ge o ad cl

is

es gen an a

is

ANTARCTIC ISLANDS

low

Sp ec i f i c ( — ) an d g en et i c ( ---- ) d i v er s i t y

Ontogenetic phase

Arrival and establishment (0 - 10,000 y)

Early development (10,000 - 3 my)

Maturation

(3 - 5 my)

Senescence and extinction (5 - 6 my)

Ontogenetic phase FIGURE 2 Change in genetic and specific diversity during ontogeny of

floras of low (A) and high (B) elevation oceanic islands. From Stuessy (2007).

SEE ALSO THE FOLLOWING ARTICLES

Endemism / Founder Effects / Metapopulations / Population Genetics, Island Models in FURTHER READING

Crawford, D. J., and T. F. Stuessy. 1997. Plant speciation in oceanic islands, in Evolution and Diversification of Land Plants. K. Iwatsuki and P. H. Raven, eds. Tokyo: Springer-Verlag, 249–267. Pfosser, M., G. Jakubowsky, P. M. Schlüter, T. Fer, H. Kato, T. F. Stuessy, and B.-Y. Sun. 2005. Evolution of Dystaenia takesimana (Apiaceae) endemic to Ullung Island, Korea. Plant Systematics and Evolution 256: 159–170. Stuessy, T. F. 2007. Evolution of specific and genetic diversity during ontogeny of island floras: the importance of understanding process for interpreting island biogeographic patterns, in Biogeography in a Changing World. M. C. Ebach and R. S. Tangney, eds. Boca Raton, FL: CRC Press, 111–123. Stuessy, T. F., D. J. Crawford, and C. Marticorena. 1990. Patterns of phylogeny in the endemic vascular flora of the Juan Fernandez Islands, Chile. Systematic Botany 15: 338–346. Stuessy, T. F., G. Jakubowsky, R. Salguero Gómez, M. Pfosser, P. M. Schlüter, T. Fer, B.-Y. Sun, and H. Kato. 2006. Anagenetic evolution in island plants. Journal of Biogeography 33: 1259–1265. Stuessy, T. F., and M. Ono, eds. 1998. Evolution and Speciation of Island Plants. Cambridge, UK: Cambridge University Press. Whittaker, R. J., and J. M. Fernández-Palacios. 2007. Island Biogeography: Ecology, Evolution, and Conservation, 2nd ed. Oxford: Oxford University Press. 10

Antarctic islands vary substantially, from small exposed mountain peaks and large dry valleys surrounded by ice to the highly variable Southern Ocean islands, which have a considerable range of sizes and geological histories. Reflecting the diverse locations and origins of Antarctic islands, their biota varies substantially. Some ice-free areas of the continent are devoid of anything except microbial life, while others support bryophytes, lichens, nematodes, arthropods, and, occasionally, breeding seabirds. The Southern Ocean islands range from those virtually covered by glaciers to others that have lush, vegetated landscapes at their lower elevations, riddled by the burrows of breeding seabirds and home to large colonies of penguins, albatrosses, and seals.

A N TA R C T I C I S L A N D S , B I O L O G Y

Despite its considerable area, only 0.32% of the Antarctic continent is exposed. Thus, land-surfaces are effectively islands surrounded by a sea of ice. These “islands” vary from small nunataks (Fig. 1), no larger in some cases than tens of square meters, to much larger areas such as the McMurdo Dry Valleys (4000 km2), Bunger Hills (950 km2), and Vestfold Hills (420 km2). Along the Antarctic Peninsula, more conventional islands abound. Among the better known of these are those of the Scotia Arc: the South Shetland, South Orkney, and South Sandwich Islands and South Georgia. The latter is typically classified as a sub-Antarctic island, together with several archipelagoes

FIGURE 1 Cairn Peak, a nunatak in the Robertskollen nunatak group,

Western Dronning Maud Land (Queen Maud Land), Antarctica.

s. an dI Fa lkl Antarctic Polar Frontal Zone

So u

th Am e

ric

a

Tristan da Cuhna Group Gough Is.

ground and the open fellfield on the higher, exposed areas.

to the east, including the Prince Edward Islands (Fig. 2), Crozet Islands, Kerguelen Islands, Heard and McDonald Islands, and Macquarie Island. These islands all lie in the Antarctic Polar Frontal Zone, on either side of the Antarctic Polar Front. More remote and poorly known Antarctic islands include the extremely isolated Bouvetøya (∼54° S, 3° E), Balleny Islands (∼67° S, 163° E), and Peter I Øy (∼69° S, 90° W). On biogeographic grounds, the Antarctic is typically divided into three major regions: the continental Antarctic (most of the continent, the eastern and southern regions of the Peninsula, and the Balleny Islands), maritime Antarctic (western coastal regions of the Peninsula south to Alexander Island, as well as the South Shetlands, South Orkneys, Bouvetøya, and Peter I Øy), and sub-Antarctic. Biogeographically, the sub-Antarctic islands form an almost indistinguishable continuum with the other islands of the Southern Ocean. Indeed, in terms of the relationships among their biotas, the New Zealand subAntarctic islands (Auckland, Campbell, Snares, Antipodes, Bounty) and the islands of the Tristan da Cunha/Gough and St. Paul/New Amsterdam archipelagoes are not entirely distinguishable from those further to the south and are often treated as part of the Antarctic region. This is true also of the Falkland Islands, though these various treatments are not without controversy. Here, we include these island groups along with those in the sub-Antarctic as Southern Ocean islands (Fig. 3). Given their occurrence right around the Southern Ocean, the islands differ considerably in their geological histories, past and current glacial extents, current climates, and vegetation. Although some of the islands, such as Prince Edward Island, are entirely volcanic, young (< 500,000 years), and show no signs of glaciation at the height of the last glacial maximum (LGM), others

th A

frica

gia Bouvetøya eor th G ich Is. Sou South Sandw Is. ney . k r th O nd Is Sou hetla th S Sou

Prince Edward Is. Marion Is. Crozet Is.

Kerguelen Is. McDonald Is.

Amsterdam Is.

Heard Is.

St Paul Is.

Antarctica

Ma

rie

Is.

ua

cq

FIGURE 2 An aerial view of the Tafelberg area on sub-Antarctic Marion

Island. Take note of the mire and fernbrake vegetation in the fore-

Sou

Campbell Is. Auckland Is. Antipodes Is. Snares Is. Bounty Is. lia

New Zealand

a str Au

FIGURE 3 Schematic map of the position of the Southern Ocean

Islands, which straddle the Antarctic Polar Frontal Zone.

have a more complex geology and history. Macquarie Island constitutes a raised section of seafloor. The Kerguelen Islands (a large archipelago) are still partly glaciated and have a complex history associated with the 100-million-year geological evolution of the large igneous province of the Kerguelen Plateau, of which some parts were subaerial (i.e., above sea level) at least as far back as 93 million years ago and consistently for at least 40 million years. The geological history of the Crozet archipelago remains something of a conundrum, whereas the Falkland Islands have a complex geological history spanning some 2500 million years. In terms of climates, similar variation can be found, from the temperate, warmer islands, such as Gough Island, the Falklands, and the Auckland Islands, to the north of the Polar Frontal Zone, to the much colder islands south of the zone, such as South Georgia and Heard Island. The islands also differ in the extent to which they are influenced by frontal weather, and in some cases the considerable height of the islands (e.g., highest peaks of 2950 m for South Georgia; 2745 m for Heard Island) means that the climates on the weather and lee sides of the islands are wholly different. BIOLOGICAL DIVERSITY

The Antarctic continent is depauperate by comparison with other terrestrial regions. Microbes predominate, but, unlike most areas elsewhere on the planet, terrestrial areas of the Antarctic are frequently either devoid of or very poor in other life forms. How the identity and diversity A N TA R C T I C I S L A N D S , B I O L O G Y

11

of Antarctic microbial communities compares with other areas globally is not yet certain, largely because similar sampling regimes have not been as comprehensively implemented in many other areas. Algae are the most widespread and diverse primary photosynthetic organisms in the terrestrial Antarctic, with 700–1000 species likely present on the continent. Lichens are represented by approximately 420 species, with the majority known from the maritime Antarctic. The mosses are also more diverse in the maritime than in the continental Antarctic. In total, 113 species are known from the region, of which only 18 have been recorded in continental Antarctica. Liverworts are much less diverse, with approximately 27 species known from the maritime Antarctic and only one from the continental Antarctic. Only two species of vascular plants, the grass Deschampsia antarctica and the forb Colobanthus quitensis, are known from Antarctica, and specifically the Peninsula area. Multicellular animal life is absent from several ice-free areas. Moreover, some sites are unusual in that they lack groups of animals that are characteristic of virtually all systems globally (such as nematode worms, springtails, and mites). The continental and maritime Antarctic house approximately 29 and 54 mite species, respectively, and some 30 springtail species. Approximately 70 tardigrade (water bear) species are known from the region, and rotifer richness is in the region of 150. Nematode diversity is likely high but is poorly known because of the lack of comprehensive sampling. It is only on the Antarctic Peninsula that true insects (two species of chironomid midge: Belgica antarctica and Parochlus steinenii) can be found. As might be expected, species richness of most groups is substantially higher on the Southern Ocean islands than on the Antarctic Continent. For example, the sub-Antarctic Prince Edward Islands (∼340 km2) are characterized by an indigenous biota comprising 90 moss species, 40 liverworts, 118 lichens, 22 vascular plants, ∼63 mites, 11 springtails, and 19 insect species. The warmer Gough Island is home to 57 indigenous vascular plant species and 28 insect species. Islands further to the north, such as the Falklands and the Auckland Islands, are considerably more species rich. For example, East Falkland is home to ∼149 indigenous vascular plant and 132 indigenous insect species, and in total 188 indigenous vascular plant and 237 indigenous insect species have been recorded from the Auckland islands. For many islands, higher taxa such as the bryophytes, springtails, and mites remain poorly known, owing to lack of systematic surveys and modern systematic treatments. In consequence, biogeographic conclusions con-

12

A N TA R C T I C I S L A N D S , B I O L O G Y

cerning these groups as a whole across the region are likely to be speculative at best. By contrast, the pelagic, though land-breeding, seabirds and seals are well known. Islands such as Possession Island in the Crozet group and Prince Edward Island in the Prince Edward Islands group are home to 33 and 28 species of seabirds respectively. On Prince Edward Island these include albatrosses (wandering, grayheaded, yellow-nosed, dark-mantled sooty and lightmantled sooty), penguins (king, rockhopper, macaroni, and gentoo), petrels (including the white-chinned, gray, great-winged, Kerguelen, soft-plumaged, Georgian diving, and gray-backed storm), and several other species, including the unusual, nonpelagic lesser sheathbill. The diversity of seals is lower, but on some islands includes breeding southern elephant seal, Antarctic fur seal, and sub-Antarctic fur seal, and vagrant species such as leopard seal. BIOGEOGRAPHY

Although the Antarctic is typically divided into three biogeographic regions (continental Antarctic, maritime Antarctic, and sub-Antarctic, as just described), more recently an extremely clear distinction between faunas of the Antarctic Peninsula and the remainder of continental Antarctica has been recognized. The zone of division between them has been named the Gressitt Line (roughly south of Alexander Island and the English and Bryan Coasts, north of inland Ellsworth nunatak ranges, and south of the Wakefield Mountains on the east coast of the Peninsula). This distinction is not shown by the bryophytes, perhaps reflecting their greater dispersal ability. Variation in dispersal ability also accounts for differences in the biogeography of groups found on the Southern Ocean islands. For those taxa with limited dispersal ability, biogeographic affinities of the islands tend to be to the nearest continental landmass, whilst in the case of more mobile taxa, distributions tend to be more circumAntarctic, with mean annual temperature, rather than identity of the nearest landmass, being more significant a correlate of the composition of the assemblages. Nonetheless, some relationships remain enigmatic. For example, although recent phylogeographic work has revealed that islands such as the Prince Edwards were probably colonized by ectemnorhine weevils shortly after they became subaerial (despite the present flightlessness of all members of the group), and it is known that the weevils probably arose on older islands (e.g., Kerguelen, which has been subaerial since the Cretaceous), the closest relatives of the group remain a matter of conjecture. Indeed, the origins

of many groups in the Kerguelen Biogeographic Province (or South Indian Ocean Islands) remain controversial. In other areas of the Antarctic, phylogeographic work has resolved similar questions, demonstrating, for example, that the midge species endemic to the Antarctic Peninsula achieved their distributions as a consequence of vicariance following opening of the Drake Passage ∼20–40 million years ago. Similarly, springtails within and those closely related to the genus Cryptopygus likely diverged following glaciation of the continent 10–23 million years ago and colonized the sub-Antarctic islands much later on several different occasions (< 2 million years ago). By contrast, mites in the genera Halozetes and Alaskozetes likely colonized the continent and the sub-Antarctic islands over the last 2–3 million years. Within particular regions, the smaller-scale phylogeographic signal may vary substantially. For example, a relatively straightforward isolation by distance signal is characteristic of the springtail Gomphiocephalus hodgsoni in Victoria Land. By contrast, dozens of haplotypes are found in each of several mite and springtail species at Marion Island (e.g., Cryptopygus antarcticus travei; Eupodes minutus; Halozetes fulvus), showing no isolation by distance, but rather a complex pattern of relationships relating to glacial and volcanic dissection of the landscape. Reflecting their origins from a small number of founding individuals, virtually all alien species that have been examined to date show minimal phylogeographic structure, and often very large populations are characterized by a single haplotype. As is the case on other isolated islands, some groups have shown adaptive radiation. For example, the ectemnorhine weevils are restricted to the islands of the South Indian Ocean province and comprise some 36 species. Endemicity in the insects is also high for other archipelagoes such as the Aucklands. By contrast, the vascular plants show much lower levels of endemicity, and this is true also of more mobile taxa such as bryophytes and the pelagic birds and seals. From the perspective of ecological biogeography, and for better known groups, such as vascular plants, insects, seabirds and, where these are present, landbirds, variation in indigenous and alien species richness across the Southern Ocean islands has been thoroughly investigated. Vascular plant species richness covaries strongly with available energy and with island area, insect richness covaries with available energy, seabird richness with energy at sea (chlorophyll concentration) and energy, and landbird richness with indigenous insect and plant richness. These patterns are in keeping with the kinds of species–energy relationships found elsewhere on both continents and

islands. However, their mechanistic basis remains as elusive here as it does elsewhere. That the alien vascular plant and insect species richness (see below) also show strong relationships with energy suggests that extinction, rather than speciation, is a major determinant of richness variation, at least over the short term. Of course, for islands, colonization rates must also be significant, and indeed for the alien species, human visitor frequency to the islands is also a significant covariate of richness variation. SMALL-SCALE PATTERNS IN DIVERSITY

Over small spatial scales, variation in the presence, richness and abundance of Antarctic plants and animals is determined mainly by water availability, temperature (which also influences water availability), protection from wind, the availability of nutrients, the extent of lateral water movement, and the extent of soil movement and ice formation. Of these, water availability (and the elevated temperatures that drive it) is thought to be most significant on the Antarctic continent and Peninsula, while nutrient availability, soil water movement and temperature are most significant in the sub-Antarctic. At least on the continent, extreme abiotic conditions preclude life in many ice-free areas and, unlike the situation across most of the planet, abiotic rather than biotic stressors exert a controlling influence on life histories. Indeed, a few abiotic and spatial factors together may account for more than 80% of the variation in the abundance of arthropods at a given site. In addition to free water, the availability of nitrogen has a substantial influence on assemblage patterns at some continental Antarctic sites (e.g. nunataks supporting vs. those devoid of seabird colonies have very different assemblages). Nonetheless, the limiting resource in many continental Antarctic systems is carbon. Indeed, at some sites, such as the McMurdo Dry Valleys, it is clear that palaeo deposition of carbon supports current, slow biogeochemical activity. Carbon limitation is highly unusual by comparison with other regions globally. Among the Southern Ocean islands, plant assemblages vary substantially mostly owing to climate. For example, Heard Island (53° S) has closed vegetation communities only in coastal areas and in some deglaciated valleys. Above approximately 50 m in elevation, vegetation is open, and above about 200 m comprises cryptogams only. From about 300 m the slopes are almost entirely ice-covered. By contrast, Gough Island (40° S) supports trees (Phylica arborea, Sophora macnabiana) and tree ferns (Blechnum palmiforme) at lower elevations, but above ∼300 m it comprises mostly wet heath and moorland vegetation. Similarly, whereas at South Georgia closed vegetation is

A N TA R C T I C I S L A N D S , B I O L O G Y

13

mostly restricted to the lowlands, a structurally complex flora can be found at the Auckland Islands. On particular islands, plant assemblage variation is a function of the influence of salt spray and manuring by pelagic species that come ashore to breed; water availability and the extent of lateral water movement; and temperature. For example, on Marion Island at the whole-island level the most pronounced variation is with altitude. High-elevation areas are free of vascular plants and are classified as polar desert. Intermediate elevations are dominated by fellfield vegetation (with the cushionforming Azorella selago (Apiaceae) most abundant). Coastal plains are characterized by mires and fernbrakes (Fig. 2), and where seabird and seal colonies are present, tussock grasslands and herbfield predominate, characterized by nitrophilous or salt-tolerant species. In low-land areas, finer spatial-scale variation is considerable and is most clearly reflected by changes in the species composition of bryophytes, in response especially to the extent of waterlogging and likelihood of lateral movement of water. Similar patterns characterize other islands, though with the absolute distance over which assemblage differences develop and the nature of the high elevation assemblages depending on the latitudinal position and altitudinal extent of the island in question. Invertebrate abundances and distributions tend to reflect those of the major plant assemblages, mostly because the invertebrates respond to similar environmental conditions. Nonetheless, pronounced differences do occur, such as the much higher richness and habitat specificity of species in the rocky, epilithic biotope dominated by cryptogams, compared with the lower richness and habitat specificity of those of the vegetated biotope that is dominated by vascular plants. Because temperature and water availability are major determinants of the abundance and distribution of life (and therefore of survival and reproduction at the population level), much attention has been focussed on the survival strategies and life histories of Antarctic and subAntarctic organisms. A variety of strategies are adopted by Antarctic organisms to survive low temperatures, and, at least on the Peninsula and in the sub-Antarctic, unpredictable changes in weather. These strategies range from anhydrobiosis in tardigrades and nematodes, to very rapid alterations (within hours) of the freezing point of springtails, to substantial protection against ultraviolet radiation in a variety of bryophytes. In sub-Antarctic insects it appears that moderate freeze tolerance is used as a strategy to overcome the marked and unpredictable short-term variation in temperature so characteristic of

14

A N TA R C T I C I S L A N D S , B I O L O G Y

these cold islands. Indeed, sub-Antarctic insects are very different from their sub-Arctic counterparts, which tend to be characterized mostly by freezing intolerance and a strategy of lowering their freezing points in anticipation of winter conditions. This asymmetry in strategies seems typical of many groups and may indeed represent one of a variety of ways, at several levels of the biological hierarchy, in which the cold temperate zones of the northern and southern hemispheres differ. HUMAN IMPACTS

In the terrestrial ecosystems of the Antarctic, human impacts are smaller than they have been elsewhere. Humans first landed on the continent in the Peninsula area around 1821, and on East Antarctica in 1895. Many of the sub-Antarctic islands have equally short human histories (sealing commenced mostly in the early to late nineteenth century). Intriguingly, the number of humans likely to be present on an island annually is strongly related to the area of the island and to its mean annual temperature – people prefer large, warm islands. However, the situation might now be changing because numbers of tourists to the Antarctic, and particularly to the Peninsula region and a few sub-Antarctic islands, are growing almost exponentially. Early human impacts were mostly restricted to marine systems as a consequence of sealing and whaling, with changes to the terrestrial environment being localized in their extent and nature. Now the situation is quite different, and both the direct local and indirect influences of humans are increasing across the region. For example, invasive alien species have profoundly altered species assemblages and ecosystem functioning on most subAntarctic islands, and their direct effects are starting to be felt on the continent itself, often in ways that are not immediately obvious. At only two sites on the Antarctic continent have reproducing populations of alien multicellular species established themselves outside research stations (the same is not true of microbes): the grasses Poa pratensis (Cierva Point, northern Antarctic Peninsula) and Poa annua (King George Island). A single individual of an alien grass (Poa trivialis) is known from the vicinity of Syowa station (East Antarctica). Two populations of Lycoriella midge have established on the continent: one survives in the sewage system at Casey Station (East Antarctica) and one, now eradicated, in the alcohol bond store at Rothera station (Antarctic Peninsula). The origins of these species remain poorly known. On Signy Island (South Orkney Islands), the chironomid midge Eretmoptera murphyi

and the enchytraeid worm Christensenidrilus blocki have become established after accidental introduction during reciprocal transplant experiments investigating plant performance. Alien species are typical of the large majority of Southern Ocean islands. Many of these species are thought to have been introduced either by sealing and whaling activities in the nineteenth and twentieth centuries or by scientific and logistic operations that have taken place since the early to mid-twentieth century. On some islands, the alien component is substantially larger than the indigenous species richness of the same groups, such as for the vascular plants on Possession Island (Crozet Group) and the insects on Gough Island. Alien species include aggressive transformer species such as feral cats, rabbits, mice, reindeer, and weedy grasses that have had or continue to have substantial impacts on local ecosystems and on several of their constituent species. The overwhelming influence of abiotic factors on the distribution and abundance of organisms in the Antarctic and on the Southern Ocean islands suggests that assemblages will be highly sensitive to climate change. At least in some parts of the continent, change in temperature and precipitation has been considerable. Thus, over the past 50 years, mean annual temperature at Faraday/Vernadsky station (Western Peninsula) has increased by 0.56 °C per decade, with much of the warming taking place in the winter months. Liquid precipitation is also on the increase. By contrast, temperatures in some Eastern Antarctic locations are declining, with the Amundsen-Scott station (South Pole) experiencing a decline at a rate of –0.17 °C per decade. Elsewhere, little change on the ground has taken place, although generally tropospheric warming is characteristic of the region. Regional variation in local responses to global climate change is also characteristic of the Southern Ocean islands, with some islands, such as Gough Island, showing little more than the average change for the planet in terms of temperature and little alteration in rainfall, while others, such as Kerguelen, Macquarie, and Marion Islands, have shown substantial increases in mean annual temperature and declines in precipitation (∼1.2 °C increase in mean annual temperature and a decline of 600 mm in total annual precipitation on Marion Island). Less obvious changes include an increase in the frequency of strong winds, a change in wind direction, and an increase in the number of clear sky evenings. In response to warming on the Antarctic Peninsula, the two vascular plant species have shown a considerable increase in local abundance and in overall distributional range. Not all species have shown such clear-cut

responses. Indeed, the effects of interactions between changes in UV-B radiation, temperature and water availability, and the life histories and physiological responses of different taxa make responses to environmental change complex. Although experimental work has demonstrated increases in abundance of some groups with warming, in others the effects may be in the opposite direction, especially where water stress increases. Moreover, even within higher taxa, responses vary considerably between sites and between studies. Nonetheless, the exposure of new ground as glaciers retreat and the increasing availability of water in some areas suggest that the overall area occupied by multicellular life is likely to increase, at least in the Peninsula region. In the sub-Antarctic, increases in temperature and declines in water availability are having considerable impacts on arthropod assemblages and on several plant species. Glacial retreat on some islands is exposing new ground, which is often colonized by invasive alien species. For example, pioneer species on glacier forelands on Kerguelen Island include Poa annua and Cerastium fontanum, species that are invasive throughout the region. On other islands the upper altitudinal distribution of vascular plants has increased substantially over the past half century. The impacts of climate change can also be more subtle. For example, increases in freeze-thaw cycles associated with more frequent clear skies are predicted to have a substantial impact on a keystone caterpillar species on Marion Island (the moth Pringleophaga marioni) because of its inability to sustain growth after multiple low-temperature exposures. These impacts will be compounded by the predation of caterpillars by introduced feral house mice, which are showing a population increase due to their positive response to warming, drying conditions. Indeed, mice are substantially altering nutrient cycling and community level changes in plant abundances on Marion and many other Southern Ocean islands. Interactions between climate change and invasive species show further subtleties. For example, it appears that alien springtail species on Marion Island have greater desiccation resistance following acclimation to high temperatures (15 °C), whereas the converse is true of the indigenous species. These differences in the form of phenotypic plasticity account for the positive response to warming and drying in the alien species and the negative response in the indigenous ones in large-scale manipulative field trials. Field surveys have also demonstrated the predominance of alien springtails in lowaltitude assemblages on the island and their absence from higher elevations. Laboratory work has in turn demonstrated steeper development rate–temperature relationships

A N TA R C T I C I S L A N D S , B I O L O G Y

15

in the alien than in the indigenous species, and a greater tolerance of low temperatures and less of a tolerance of high temperatures in the indigenous compared with the alien species. Thus, abiotic controls are likely playing a major role in influencing the way in which climate change affects the interplay between indigenous and alien species on the island. Interactions between alien and indigenous species are predicted to grow in scope and complexity with climate change because more alien species are likely to establish in warmer climates. Indeed, an increasing level of establishment of alien species is predicted both for the Antarctic Peninsula and for the Southern Ocean islands because of ameliorating climates and increases in human activity (scientific, logistic, and tourist). Although ongoing monitoring has not taken place at many sites, it is clear that at several of the sub-Antarctic islands, alien species continue to establish despite strict management controls. In many instances these introductions have substantially increased food web complexity by adding trophic groups (e.g., parasitoids) that were either scarce or absent from terrestrial systems. CONSERVATION

Conservation south of 60 °S is implemented through the Antarctic Treaty System (the Antarctic Treaty itself was signed on December 1, 1959, and entered into force on June 23, 1961), specifically via the provisions of the Protocol on Environmental Protection to the Antarctic Treaty, or the Madrid Protocol. The Committee for Environmental Protection is the body responsible for conservation decision making in the region, by consensus among the Treaty Parties, which are then individually responsible for implementing decisions via their national legislation. Although formal conservation planning at a regional scale has been absent from the Antarctic region until recently, new efforts are underway to remedy the situation. Because the Southern Ocean Islands have sovereignty claims, each island is administered by the claimant nation, and the level of management varies considerably. Nonetheless, the exceptional biological value of these islands is broadly recognized. The islands are governed by five different nations: the United Kingdom, South Africa, Australia, New Zealand, and France. Although the Antarctic Treaty does not apply to the islands, international agreements to which the above states are party, such as the Convention on Biodiversity and the Agreement for the Conservation of Albatrosses and Petrels, apply to the islands. Moreover, most of the islands enjoy a high conservation status. The five New Zealand sub-Antarctic island groups

16

A N TA R C T I C I S L A N D S , B I O L O G Y

(Snares, Bounty, Antipodes, Auckland, and Campbell Islands), Heard and McDonald Islands, Gough Island, and Macquarie Island are all World Heritage Areas (at the highest IUCN Reserve Status of Category Ia). Macquarie Island is listed as a UNESCO Biosphere Reserve. At a national level, the New Zealand sub-Antarctic islands are all National Nature Reserves. Macquarie Island and Heard and McDonald Islands (Australia) have the highest reservation status, Nature Reserve and Commonwealth Reserve, respectively, under their governing legislations (state and federal). Marion and Prince Edward Island (South Africa) are classified as a Special Nature Reserve under South African legislation, South Georgia has National status (United Kingdom) as a Protected Area, and the Kerguelen and Crozet Islands (France) and Gough Island (United Kingdom) are National Nature Reserves, either as a whole or, for larger islands, in part. In most cases, the indigenous biotas of the broader Antarctic region are strictly protected, and considerable care is given to their management and the assessment of population trends, especially for seabirds and seals. At least for many of the Southern Ocean islands, strict quarantine procedures are typically in place to prevent the introduction of new alien species, although for some of the islands more needs to be done in terms of their implementation. For the region, more also needs to be done in terms of the documentation of indigenous biotas and their spatial variability so that appropriate conservation management procedures can be developed and implemented. Nonetheless, most Antarctic and Southern Ocean islands enjoy considerable protection, probably more so than many other parts of the world. Whether this situation will change when it becomes financially viable to mine the Antarctic continent or the seabed for fossil fuels or minerals remains to be seen. SEE ALSO THE FOLLOWING ARTICLES

Antarctic Islands, Geology / Climate Change / Flightlessness / Invasion Biology / Whales and Whaling FURTHER READING

Bergstrom, D., and S. L. Chown. 1999. Life at the front: history, ecology and change on Southern Ocean islands. Trends in Ecology and Evolution 14: 472–477. Bergstrom, D., P. Convey, and A. H. L. Huiskes, eds. 2006. Trends in Antarctic terrestrial and limnetic ecosystems. Berlin: Springer-Verlag. Chown, S. L., and P. Convey. 2007. Spatial and temporal variability across life’s hierarchies in the terrestrial Antarctic. Philosophical Transactions of the Royal Society of London Series B Biological Sciences 362: 2307–2331. de Villiers, M. S., J. Cooper, N. Carmichael, J. P. Glass, G. M. Liddle, E. McIvor, T. Micol, and A. Roberts. 2006. Conservation management at Southern Ocean Islands: towards the development of best-practice guidelines. Polarforschung 75: 113–131.

Frenot, Y., S. L. Chown, J. Whinam, P. M. Selkirk, P. Convey, M. Skotnicki, and D. M. Bergstrom. 2005. Biological invasions in the Antarctic: extent, impacts and implications. Biological Reviews 80: 45–72. Riffenburgh, B., ed. 2007. Encyclopedia of the Antarctic. New York: Routledge. Turner, J., J. E Overland, and J. E. Walsh. 2007. An Arctic and Antarctic perspective on recent climate change. International Journal of Climatology 27: 277–293.

ANTARCTIC ISLANDS, GEOLOGY JOHN L. SMELLIE British Antarctic Survey, Cambridge, United Kingdom

Islands in the Antarctic region (south of 60° S) have an importance beyond a simple curiosity related to their current geographical and climatic isolation. They contain a repository of geology that is representative of much of the geology of Antarctica; they are typically more accessible and often better exposed than elsewhere in Antarctica as a result of recent marine-related stripping of superimposed snow and ice; and some islands may also have acted as persistent ice-free refugia for fugitive plant and animal communities at the end of each interglacial, when the continent itself was swathed in extensive ice sheets. The survival of those refugia may have been critical in subsequently determining evolutionary trends. DISTRIBUTION AND CLASSIFICATION OF ISLAND TYPES

Islands included in this article form two geographically, geomorphologically, and tectonically distinct groups, comprising: (1) South Orkney Islands, South Shetland Islands, Joinville Island group, James Ross Island group, Seal Nunataks, Anvers and Brabant islands, Adelaide Island, Alexander Island, and Thurston Island; and (2) Peter I Island, Siple Island, Ross Island, Black and White islands, Franklin and Beaufort islands, Coulman Island, Scott Island, and Balleny Islands. The first group is mainly situated close to the Antarctic Peninsula and owes its origins geologically and tectonically to subduction of Pacific Ocean crust (Fig. 1), whereas the second group of islands is geographically more scattered and is composed of several isolated large alkaline volcanoes situated in an intraplate setting (both continental and oceanic) and probably related to impingement of one or more deep thermal anomalies (also called mantle

FIGURE 1 Map showing the location of subduction-related islands in

Antarctica.

plumes) (Fig. 2). Although some of the islands are surrounded by open sea, in summer, at least, others are joined to mainland Antarctica by permanent thick shelf ice, such as the Seal Nunataks (originally named Seal Islands), Alexander Island, Thurston Island, Siple Island, Ross Island, and Black and White Islands. Others, such as Berkner Island and Roosevelt Island, are wholly enshrouded by the large ice shelves facing the Weddell and Ross seas; although covered by topographically prominent ice domes, the rock surfaces in these are probably well below sea level. SUBDUCTION-RELATED ISLANDS

By Antarctic standards, the subduction-related islands are comparatively well exposed (e.g., large snow-free peninsulas on King George and Livingston islands in the South Shetland Islands; Deception Island; northern James Ross Island; and Seymour Island). Others are almost wholly encased in snow and ice (e.g., Joinville Island group, Biscoe Islands). They range from mountainous islands with sharp peaks rising to summits over 2000 m above sea level (e.g., South Orkney Islands; Elephant and Clarence Islands group; Smith Island; Brabant Island; and eastern Anvers, Adelaide, and Alexander islands) to subdued lowlying islands typically < ∼1000 m in elevation, dominated

A N TA R C T I C I S L A N D S , G E O L O G Y

17

FIGURE 2 Map showing the location of intra-plate islands in Antarctica.

by thick snow and ice domes (e.g., many of the South Shetland Islands; western Anvers, Adelaide, and Alexander islands). Unusually for the region, the South Shetland Islands are rich in raised beaches and raised marine platforms. Similar features have also been described below sea level. The geology of these islands can be described within a plate tectonic framework of eastward subduction of oceanic crust beneath the Antarctic Peninsula, a process that may have started as early as the mid-Paleozoic but that is mainly represented by rocks < 200 million years old. Thus, the region is dominated by remnants of a near-complete Mesozoic arc–trench system. An alternative view, involving the sequential accretion of suspect terranes, has also been proposed. However, the terrane accretion also took place within a plate tectonic framework that does not greatly detract from the basic description of the principal tectonic elements given here. Subduction ceased northward as a series of sec-

18

A N TA R C T I C I S L A N D S , G E O L O G Y

tions of a mid-ocean ridge arrived at the Antarctic Peninsula trench, beginning about 50 million years ago off southern Alexander Island and continuing until about 4 million years ago off Anvers Island. The South Shetland trench is the sole surviving remnant. Although this trench is largely quiet seismically, subduction is believed to be occurring there still, but at a very slow rate. Active tectonic processes may have been responsible for the relatively recent rapid uplift of sections of crust that were formerly buried very deeply and are now quite mountainous (e.g., Smith Island, and Elephant and Clarence Islands Group). Subduction complexes are exposed in islands in the Joinville Island group, South Shetland Islands, South Orkney Islands, and Alexander Island. The oldest, which may be part of the local basement on which the Mesozoic magmatic arc was founded, comprises outcrops of deformed metasedimentary sequences, mainly quartzose sandstones and mudstones (Trinity Peninsula Group). The strata are sparsely exposed on Joinville Island, and they form Laurie Island and part of Powell Island (South Orkney Islands). Although commonly attributed to a subduction complex, the tectonic setting is not well understood. Deposition may have been Permo-Triassic (250–230 million years ago), with accretion probably completed by Early Jurassic time (210–190 million years ago). Further south, quartzofeldspathic metasedimentary and rarer volcanic rocks occupy an outcrop over 300 km in width in Alexander Island (LeMay Group). They are deformed and metamorphosed up to blueschist grade. Sedimentation was probably mainly Early Jurassic–Early Cretaceous (200–100 million years ago), but it may extend back to Permo-Carboniferous times (350–260 million years ago). Other distinctive subduction complex outcrops form Smith Island and the Elephant and Clarence Islands Group (South Shetland Islands) and parts of the South Orkney Islands. The metamorphic outcrops include distinctive blueschists; ages around 200–240 million years have been obtained in the South Orkney Islands and 90–110 million years in the South Shetland Islands. The South Orkney Islands owe their present isolated position, some 600 km northeast of the Antarctic Peninsula, to late Cenozoic faulting associated with Scotia Sea formation during the separation of Antarctica from southern South America. Undeformed, highly fossiliferous fore-arc basin sedimentary sequences are preserved in Alexander Island, where they unconformably overlie the LeMay Group.

The sediments are known as the Fossil Bluff Group, and they comprise a weakly deformed and essentially unmetamorphosed marine to fluvial sequence up to 7 km in thickness that is Early Jurassic to mid-Cretaceous in age (200–100 million years). Magmatic rocks, both volcanic and plutonic, are widely exposed all the way from Thurston Island to the South Shetland Islands. They were formed between about 170 and 20 million years ago and comprise mainly basaltic to andesitic lavas and fragmental deposits, and more rarely dacites to rhyolites. The volcanic rocks are hydrothermally altered by a range of compositionally similar coeval plutonic intrusions. Most of the magmatic rocks are related to partial melting caused by subduction of old Pacific Ocean crust, but those on Alexander Island include distinctive magnesium-rich andesites linked to subduction of a segmented ancient mid-ocean ridge after about 80 million years ago, causing anomalously shallow heating and melting in the fore-arc region and the northward migration of the volcanic centers on the island. Volcanic centers in the South Shetland Islands also migrated in a northeasterly direction, but the cause is unknown. The youngest magmatic arc rocks (< 30 million years ago) on King George Island are interbedded with three important glacial sedimentary units. The two older sequences contain a rich diversity of exotic stones that indicate that glaciation was also under way in East Antarctica. Late Cretaceous to Eocene/Oligocene (90–35 million years ago) back-arc basin sedimentary sequences are widely exposed on Seymour, Snow Hill, and James Ross islands. The basin fill is possibly 5 km thick, and it contains a highresolution record of both terrestrial and marine climate change. The excellent exposure of highly fossiliferous sediments contains a wide range of invertebrate and vertebrate fossils, which are often very well preserved, but the very soft nature of the strata means the outcrop areas form very lowlying ground. Extension and rifting of the northwestern margin of the Antarctic Peninsula within the past 20 million years culminated in the formation of a wide marginal basin in Bransfield Strait and eruption of basalt lavas. Deception Island, an active volcano with a large flooded caldera, is related to this process (Fig. 3), as are Bridgeman Island, Penguin Island, and numerous small volcanic centers on Livingston and Greenwich Islands. Three large overlapping shield volcanoes less than 200,000 year in age were also constructed on Anvers and Brabant islands, at the likely southern limit of marginal basin tectonic effects. Topographic expression

FIGURE 3 Aerial view of Deception Island, a largely ice-free active

volcano with extensive areas of snow-free ground. The volcano has a flooded interior caused by a major collapse following a very large eruption, probably less than 10,000 years ago. Photograph by John Smellie.

of Cenozoic extension in the Antarctic Peninsula region is widespread and is shown by deep north–south channels separating the islands from the mainland. Outlet glaciers formerly occupied the channels, and buttressing of the large shield volcanoes on Brabant and Anvers islands may have enhanced the glacial erosion. Back-arc volcanism is well represented in the James Ross Island group, which contains an extensive basaltic volcanic field dominated by a single large volcano (Mt. Haddington). Many of the smaller islands in the group are satellite volcanic centers, and the volcanic field may still be active. There are numerous interbedded glacial sedimentary rocks, some highly fossiliferous. The volcanic rocks also show unequivocal evidence for repeated eruptions beneath an ice sheet, and, together with the sedimentary rocks, they are a major source of paleoenvironmental information. Possibly as a consequence of mid-ocean ridge collisions with the Antarctic Peninsula trench, gaps were created in the subducting slab and resulted in eruption of post-subduction volcanic rocks, mainly in Alexander Island and Seal Nunataks, from 7 million years ago on. The volcanism is basaltic, related to upwelling of mantle through the gaps in the slab. Surprisingly, the timing of post-subduction eruptions shows no obvious progression mirroring the ridge collisions. INTRAPLATE ISLANDS

More than 90% of the surfaces on the islands in this group are covered by permanent snow and ice. Almost all are well-formed volcanic cones that contrast with the much more severely eroded subduction-related islands. This contrast is probably due to the very different ages and glacier thermal regimes in the two island groups: almost all of the subduction-related islands are older

A N TA R C T I C I S L A N D S , G E O L O G Y

19

and are deeply eroded by wet-based ice, whereas the intraplate islands (excepting the oceanic Peter I Island, Scott Island, and Balleny Islands) are less than 7 or 8 million years old and have been affected by a largely nonerosive cold polar regime. The intraplate islands are all volcanoes, typically rising steeply to elevations between 1000 and 3000 m above sea level. A few have snow- and ice-filled calderas, but most simply have small summit craters; in one case (Mt. Erebus) the crater contains a persistently active lava lake that frequently erupts pumiceous bombs with distinctive large (to 10 cm) feldspar crystals. At least one other volcano in the group may be active (Mt. Siple, Siple Island). A few are still undated (Scott Island, Balleny Islands, Beaufort Island). The volcanoes range in composition from basalt to mugearite, trachyte and phonolite, characteristic of an intraplate extensional environment. However, there are differences in tectonic setting. Mt. Siple is probably part of the large continental alkaline volcanic province in Marie Byrd Land, which is causally related to a mantle plume, regional updoming, and progressive fracturing along orthogonal continental crustal fractures. Volcanoes in the western Ross Sea region may be associated with reactivated deep faults related to large-scale Cenozoic Australia–Antarctica plate separation tectonics. Finally, Peter I Island, Scott Island, and Balleny Islands have been linked to eruptions along “leaky” oceanic fracture zones.

ANTILLES, BIOLOGY CHARLES A. WOODS University of Vermont, Island Pond

FLORENCE E. SERGILE University of Florida, Gainesville

The “Antilles” is an archipelago of over 7000 large and small islands, cays, reefs, and exposed offshore banks with long and diverse geological and biological histories. The total human population of the Antilles is 34.5 million. The heterogeneous assemblage of islands stretches in an arc over 3200 km long. The islands originated in a variety of ways (volcanism, uplifted island arcs, exposed and uplifted coral banks, movements of major plates, changes in sea levels) and have had a history of numerous vicariance (separation) events, as pieces of islands as well as whole islands became attached and unattached, submerged and reemerged. This dynamic geological history has resulted in extremely high levels of biodiversity and insular endemism. Because of the unique combination of a well-preserved fossil record of vertebrates in sinkholes and caves (mostly found in areas of karst geology) and the extraordinarily rich record of microfossils found in Dominican amber, the past and present biodiversity of these islands is especially well documented.

SEE ALSO THE FOLLOWING ARTICLES

Antarctic Islands, Biology / Plate Tectonics / Refugia / Volcanic Islands FURTHER READING

Barker, P. F. 1982. The Cenozoic subduction history of the Pacific Margin of the Antarctic Peninsula: ridge crest–trench interactions. Journal of the Geological Society of London 139: 787–801. LeMasurier, W. E., and J. W. Thomson, eds. 1990. Volcanoes of the Antarctic Plate and Southern Oceans. American Geophysical Union, Antarctic Research Series 48: 1–487. Smellie, J. L., W. C. McIntosh, and R. Esser. 2006. Eruptive environment of volcanism on Brabant Island: evidence for thin wet-based ice in northern Antarctic Peninsula during the late Quaternary. Palaeogeography, Palaeoclimatology, Palaeoecology 231: 233–252. Storey, B. C., and S. W. Garrett. 1985. Crustal growth of the Antarctic Peninsula by accretion, magmatism and extension. Geological Magazine 122: 5–14. Storey, B. C., R. J. Pankhurst, I. L. Millar, I. W. D. Dalziel, and A. M. Grunow. 1991. A new look at the geology of Thurston Island, in Geological Evolution of Antarctica. M. R. A. Thomson, J. A. Crame, and J. W. Thomson, eds. Cambridge, UK: Cambridge University Press, 399–403. Vaughan, A. P. M., and B. C. Storey. 2000. Terrane accretion and collision: a new tectonic model for the Mesozoic development of the Antarctic Peninsula magmatic arc. Journal of the Geological Society of London 157: 1243–1256.

20

ANTILLES, BIOLOGY

BIOGEOGRAPHY OF THE GREATER ANTILLES

The Greater Antilles includes Cuba, Hispaniola (Haiti and the Dominican Republic), Jamaica, and Puerto Rico (including the Virgin Islands) (Fig. 1). Parts of many of these islands have changed position in the past. For example, the southern part of Hispaniola (“south island”) was separate from the rest of Hispaniola for much of its early history and was located well west of that island. The mountains of eastern Cuba were likely part of northern Hispaniola. It is also likely that parts of eastern Hispaniola were united with Puerto Rico. The Greater Antilles has extensive areas of limestone, and some have been uplifted into high mountain plateaus. Rainfall falling on these karst areas creates crevices, hollowing out sinkholes and even deep caverns. These features provide additional important habitats that further increase biodiversity. These sinkholes and caves are also places where fossil and semifossil remains of animals are found. Their presence is one reason that there is such an excellent record of the kinds and ages of vertebrates that lived on the islands.

70

80

90

GULF OF MEXICO

Grand Bahama

FLORIDA

BAHAMAS BANK

Great New Abaco Providence Bimini Eleuthera

Tropic of Cancer

Great Exuma

CU AN AT C U Y

BA

Limit of region

Cat Island San Salvador Rum Cay

Andros

20

60

Tropic of Cancer

Long Island

Mayaguana Crooked Island Caicos Islands

Isle of Youth

Great Inagua

HAITI

Grand Cayman Gonave

Santo Domingo

HI SPA NIOL A

BELIZE

Leeward Islands

PUERTO RICO

DOMINICAN REPUBLIC

JAMAICA

St.

ix Cro

illa gu An rtin St. Ma ST. Barthelemy

Barbuda

ts

St. Kit

t

rra nse Mo

Islas de la Bahia

GUATEMALA

CA

HONDURAS

ARUBA

PACIFIC OCEAN

Dominica Martinique St. Lucia Barbados

Netherland Antilles

NICARAGUA

Guadeloupe

Martinique Passage

R I B B E AN S E A

EL SALVADOR

20

LESSER ANTILLES

GREATER ANTILLES

Windward Islands

CURACAO BONAIRE

Isla Margarita

St. Vincent Grenada Tobago Trinidad

COSTA RICA

10

VENEZUELA

200 km PANAMA

COLOMBIA

200 Mi. 90

10

80

70

60

FIGURE 1 Map of the Antilles showing its biological regions.

In the Greater Antilles, many plants and animals are derived from South American ancestors. One hypothesis as to how South American terrestrial vertebrates dispersed to the Greater Antilles is the GAARlandia theory proposed by Ross MacPhee and Manuel Iturralde-Vinent. They believe that about 35 million years ago proto-Cuba, Hispaniola, and Puerto Rico were connected to northwestern mainland South America via the now submerged Aves Ridge, and they give the name GAARlandia (from Greater Antilles + Aves Ridge) to the overall structure. This bridge of emergent islands would have been above sea level for about 2 million years (between 35 and 33 million years ago). Terrestrial mammals such as sloths, monkeys, and rodents would have been able to make their way to GAARlandia at that time, later dispersing to other islands over water or via vicariance events. Rodents, sloths, monkeys, and even the Antillean piculet (Nesoctites micromegas), a primitive woodpecker, may have dispersed to the central Greater Antilles via this route. Other plants and animals, such as soricomorphs (including the large solenodons and the much smaller island shrews of the genus Nesophontes), most likely originated in North America and were able to disperse to the Greater Antilles via vicariance events when pieces of the proto-Antilles

passed close to the southern tip of the North American continent. There is recent evidence for the antiquity and uniqueness of solenodons. Analysis of a sequence of 13,885 base pairs of both nuclear and mitochondrial genes of Solenodon paradoxus indicates that solenodons separated from other placental mammals 76 million years ago. The last connection between the proto-Antilles and mainland North America was severed 70–80 million years ago, and it is likely that solenodons became part of the Antillean fauna by vicariance at that time. Cuba

Cuba is the largest island of the Antilles (110,860 km2). It is a long and relatively narrow island organized into areas of high mountains surrounded by lowlands. In the past, these lowland areas were flooded by rising sea levels, and Cuba was fragmented into as many as three separate islands. The main island is fringed along the north and south coasts by small islands and cays that are protected from invasive species and are rich in biodiversity. Off the south coast is the Isle of Youth, itself a complex island formed of two parts, which is especially rich in biodiversity. Overall, the Cuban Bank, with its diversity of habitats and smaller islands, contains the most complex and

ANTILLES, BIOLOGY

21

varied biota in the Antilles. Cuba has the largest number of educational and conservation programs, and the status of the islands biota is therefore in better overall shape than any other island in Antilles. Hispaniola

Hispaniola (76,193 km2) is located on the tectonically active Caribbean plate and is the most geologically complex of the Antilles. Pico Duarte in the Central Mountains of the Dominican Republic (3175 m) is the highest mountain peak east of the Mississippi River and the Andes. There are a number of high ranges on the island with cool, well-forested peaks (that is, sky islands), and these high mountain ranges often create rain shadows downwind with dry, and even desert, habitats. Along the fault zone where the old north and south islands join sits Lake Enriquillo, which is 45 m below sea level. Hispaniola lacks the numerous offshore cays that in Cuba have protected endemic species from competition from invasive species, but there are several large offshore islands such as Gonave Island near Port-au-Prince and Turtle Island in the north. Jamaica

Jamaica (11,424 km2) is much smaller than Cuba and Hispaniola, but the island has high mountain areas (the Blue Mountains rise to 2256 m) and significant habitat diversity, including semi-desert areas. After the Middle Eocene, Jamaica was completely submerged. It was once again fully above sea level in the Middle Miocene, about 16 million years ago, thus making the island biologically the youngest of the Greater Antilles based on current (extant) biota. The submerged platform forming the base of Jamaica is narrow, and as a result, unlike Cuba or Hispaniola, the island has few offshore cays and islands that form disjunct ecosystems. As a consequence, Jamaica has much less biodiversity than either Cuba or Hispaniola. The Seven Rivers vertebrate site in west central Jamaica (15 km south of Montego Bay) is of late Early or early Middle Eocene age. This site has produced the remains of many aquatic vertebrates, including the earliest known fully quadrupedal sea cow (sirenian), Pezosiren portelli. The site also produced the remains (right dentary) of a large rhinoceros-like perissodactyl (Hyrachys sp.). The specimen is important because it is unlikely that such a terrestrial mammal as a rhinoceros could have dispersed to Jamaica across a broad open-water area, making it likely that Jamaica had a land bridge connection with North America at the time the rhino dispersed to the island. 22

ANTILLES, BIOLOGY

Puerto Rico

Puerto Rico was part of northern Hispaniola before it separated and moved eastward about 15 million years ago. After that, and before the high mountains of the center of the island arose, the island largely submerged again before emerging for good about 12 million years ago. The submerged platform on which the island rests is very broad, and Puerto Rico has numerous surrounding cays and offshore islands such as Culebra Island, Vieques Island, and the northern Virgin Islands. THE BAHAMA BANKS

The Bahama Banks has had a dynamic history as a result of changing sea levels during the Pleistocene ice age cycles. This region of over 2000 small islets (“cays”) and 700 larger islands has a separate geological origin and is biogeographically distinct from the Greater Antilles. The geological origin of the Bahamas is controversial. The archipelago may have arisen via activity associated with plate tectonics 200 million years ago, or these low-lying islands and cays may be the remnants of a much larger platform that formed in tropical waters. The islands are all less than 61 m above sea level. The flora and fauna are mostly derived from Cuba and Hispaniola, having dispersed there when sea levels were much lower. At the height of the last ice age, when the sea level was over 90 m lower, many islands of the Bahamas became contiguous and formed two mega-islands corresponding to the “Great Bahama Bank” and the “Little Bahama Bank.” There is little topographical or ecological diversity in the Bahamas, and as a result there is relatively little diversity of plants and animals. Biogeographically, the Bahamas also include the Turks and Caicos Islands, as well as the Navidad Bank and Silver Bank off the north coast of the Dominican Republic, which are important wintering areas for whales, especially the humpback whale. LESSER ANTILLES

The Lesser Antilles begins east of the Virgin Islands (which here are considered part of the Puerto Rican Bank). The 25 larger and hundreds of smaller-to-tiny islands lie along the leading edge of the Caribbean plate, which is colliding with the North American plate (subduction). Many of the islands are of volcanic origin, and most of the remaining small cays and islets are of coral origin. Overall, this region of small volcanic and coral islands contains only 6% of the overall landmass of the Antilles. The islands in the northwest are collectively called the Leeward Islands. Within this subregion is the St. Martin Bank, which includes the islands of St. Martin, Anguilla,

and St. Barthelemy. At times of low sea levels, they would have formed a single island approximately 5949 km2. South of this, but still within the Leeward Islands, is the St. Kitts Bank, including the islands of St. Eustatius, St. Christopher (i.e., St. Kitts), and Nevis. At low sea levels, this bank would have been a single island of approximately 1546 km2. The islands of Barbuda and Antigua are located on a bank east of the above, which at times of low sea levels would have been over 4274 km2 in area. The Leeward Islands are relatively low-lying and similar to the Virgin Islands. South of these are the much more mountainous islands of Montserrat, Guadeloupe, and Dominica, still considered part of the Leeward Islands by most biogeographers. South of the Martinique Passage (located between the islands of Dominica and Martinique), a series of islands are isolated from one another by deep-water passages. These islands are collectively known as the Windward Islands and range from Martinique in the north to Grenada in the south. There is biological evidence for excluding Trinidad, Tobago, Margarita, and the Netherlands Antilles from the true Antilles.

The Poinars examined over 3000 pieces of amber, and the most frequently occurring organisms were worker ants (497), winged adult ants (286), gall midges (197), bark lice (173), and stingless bees (156). They also documented the remains of small vertebrates such anoles and geckos, soft ticks and hairs from rodents, the feather of an Antillean piculet (a small primitive woodpecker-like bird), and a few vertebrae and ribs of a tiny island shrew (Nesophontes). What is remarkable is how many Antillean plants and animals documented in Dominican amber, including the algarrobo tree itself, are now extinct or extirpated from the Antilles. For example, of the seven genera and subgenera of bees documented by the Poinars in amber, all are now extinct in the Antilles (whereas relatives survive in tropical Central and South America). The reasons and timing of these extinction events in the Antilles may relate to climate change, because ice ages during the Pleistocene in North America led to drying and cooling in the Antilles, and lowering sea levels changed the size and shape of Antillean islands. MAMMALIAN FAUNA

MICROFAUNA KNOWN FROM DOMINICAN AMBER

The nature of the biodiversity of small organisms such as ants, bees, and forest insects is poorly understood on many islands and archipelagoes but is fabulously preserved in amber deposits of the Dominican Republic. This record, preserved in fossilized resin of a now extinct algarrobo tree (Hymenaea protera), is a record of the broad biodiversity of plants and animals living together in a complex moist tropical forest between 15 and 45 million years ago in the western part of what is now Hispaniola. This tall forest habitat, referred to by George and Roberta Poinar (1999) as the “Amber Forest,” was an ancient forest ecosystem of 40-m-tall canopy trees such as the caoba (Swietenia), the algarrobo (Hymenaea), and the nazareno (Peltogyne), as well as a diverse understory layer of shrubs, ferns, and flowers. The algarrobo trees produced copious quantities of sticky resin in which small organisms were trapped and preserved. The trees formed a complex multistory habitat rich in animal biodiversity attracted by the leaves, flowers, pollen, and fruits of the trees. Ants, bees, wasps, moths, butterflies, birds, bats, rodents, and even small insectivores became trapped in the sticky resin and are now preserved in chunks of amber. The result is one of the best-documented histories of forest biodiversity known from an ancient tropical ecosystem.

There are few surviving examples of the rich mammalian fauna that characterized the Antilles 18,000 years ago. At that time, land areas of the Greater Antilles were much larger than today because of lower sea levels. The mammals that inhabited these islands included a variety of rodents, sloths, primates, and primitive insectivores. The rich diversity of endemic terrestrial mammals was especially true on the large, topographically diverse islands of Hispaniola and Cuba. Antillean Rodents

The best-known taxa are members of the rodent family Capromyidae. They are known as “jutias” in Cuba and the Dominican Republic, “zagoutis” in Haiti, and “conies” in Jamaica and the Bahamas. We will call them hutias. They reached high levels of diversity in the Cuban archipelago (five genera, three endemic) and on Hispaniola (four endemic genera). They are not known to have occurred in the Lesser Antilles, but one genus (Geocapromys) with three subspecies occurred in the Bahamas and dispersed to the Cayman Islands and to very small Little Swan Island. There were at least 55 species, of which 42 have become extinct (76%). The reasons for the extinction of these rodents, many of which were the size of squirrels and large house cats, was likely overhunting by Amerindians, habitat destruction, and predation by introduced dogs, cats, and the mongoose. The largest numbers of surviving hutias

ANTILLES, BIOLOGY

23

occur on Cuba, where 11 species are found in habitats ranging from small isolated offshore islands to mainland Cuban swamps and forested areas. One species (Capromys pilorides) is common throughout Cuba, where it is well known and frequently eaten as “bush meat.” On Hispaniola, the single surviving species of hutia (Plagiodontia aedium) has two extant subspecies, one of which is mainly restricted to southern Hispaniola (P. a. aedium) and the other to northern Hispaniola (P. a. hylaeum). A closely related hutia (the now extinct Rhizoplagiodontia lemkei) was restricted to the far west of the southern peninsula of Haiti in the Massif de la Hotte. It was very abundant and was part of the cluster of endemic plants and animals that characterized the Massif de la Hotte hotspot. On Cuba, a fossil rodent of early Miocene age (Zazamys veronicae) has been discovered, which is ancestral to Hispaniolan hutias of the genus Isolobodon. This species, like most other capromyid rodents, was a delicacy, and its remains are common in Amerindian kitchen middens. “Conies” of the genus Geocapromys still survive in Jamaica and the Bahamas. Thus, it is still possible to observe the last remnants of the once great Antillean radiation of capromyid rodents. In addition to capromyid rodents, large, heavy-bodied, wide-toothed, hutia-like rodents were present on Jamaica, Hispaniola, and Puerto Rico as well as on two small islands (Anguilla and St. Martin [St. Martin Bank]) in the northernmost Lesser Antilles. Some of these forms, such as the gigantic Amblyrhiza inundata from Anguilla and St. Martin, were as large as 200 kg. These giant rodents (family Heptaxodontidae) were known as “twisted-toothed giant hutias.” They apparently did not radiate into many species, but they were of exceptional mass. Hispaniola, Cuba, and Puerto Rico shared a radiation of small, spiny, rat-like rodents. They were smaller than hutias in mass and had cheek teeth similar in morphology to South American spiny rats. Two species are common in cave deposits on Hispaniola, and another two species occurred in Cuba. Three species are known from Puerto Rico. The remains of most of these species are fresh in appearance, indicating that they became extinct in historical times, especially on the high plateaus of southwestern Haiti. Remains are common in sinkholes, in cave deposits, and in Amerindian kitchen middens. They were likely driven to extinction by competition from introduced rats and predation by introduced dogs, cats, and possibly even the mongoose. The radiation of rodents in the Lesser Antilles was much more limited and reflects the very different origin of this long chain of volcanic islands. There are no capro-

24

ANTILLES, BIOLOGY

myids, spiny rats, or (with the exception of Anguilla and St. Martin) giant hutias. Instead, the rodents of these small islands were sigmodontine rodents of the genera Oryzomys and Megalomys that made their way onto this chain of oceanic islands by overwater dispersal from South America. There are described and undescribed taxa of sigmodontines from Anguilla, Antigua, Barbados, Barbuda, Guadeloupe, Marie Galante, Montserrat, St. Eustatius, St. Kitts, St. Lucia, and St. Vincent. One species of Oryzomys is known from Jamaica, which likely dispersed there over water from Central America. All of these forms are now extinct. Antillean Sloths

The second largest adaptive radiation of land mammals in the Antilles occurred on Hispaniola (four genera and six species) and Cuba (five genera and six species), where sloths filled a number of niches. They ranged in mass from the size of a very large dog to smaller than a house cat (the smallest known sloth). The radiations of sloths on Cuba and Hispaniola were remarkably similar to each other. Sloths are also known to have occurred on Puerto Rico (one named form and one unnamed form). This distribution confirms the close biogeographic affinity of these three islands. Sloths are not known to have occurred on Jamaica. Fragmentary dental remains of what is clearly sloth material are known from Grenada in the Lesser Antilles. No sloths have survived in the Antilles into historical times. Two large, ground-dwelling (i.e., megafaunal) sloths were present on Hispaniola, and they were likely hunted to extinction by early Amerindians. Both are closely related to the giant megalonychid ground sloth from Cuba (Megalocnus rodens), estimated to have weighed 270 kg. Another part of the sloth radiation on Hispaniola and Cuba (subfamily Choloepodinae, tribe Acratocnini) are forms similar to the two-toed sloth (Choloepis) from Central and South America. These much smaller sloths, such as “Yesterday’s Acratocnus” (Acratocnus ye) are known from well-forested and savanna upland habitats in the southwest. It is likely that this form hung from branches as two-toed sloths do. Hispaniola also had three species of a smaller sloth group (tribe Cubanocnini). The larger form was the “Hispaniolan Neocnus” (Neocnus comes), which was widespread on Hispaniola. Also present on Hispaniola was the “Slow Neocnus” (Neocnus dousman) and the “Least Neocnus” (Neocnus toupiti). All three forms were excellent climbers and moved easily in the treetops but did not hang under branches (according to sloth expert Jennifer White). Why and when the extensive radiation of sloths became extinct, in spite of their wide range of size, food

habits, and locomotor habits, is not yet resolved. The well-preserved remains of sloths are common in sinkholes along the karst-covered plateaus of the southern peninsula of Haiti. These plateaus were well forested and probably represented the last haunts of sloths on the island. It is likely that they became extinct within the last 2000 years as a result of human activity. Antillean Primates

Monkeys were first documented in the Antilles as an endemic platyrrhine (Antillothrix bernensis) in Hispaniola. Fossil endemic monkeys are also known from Cuba (Paraloutta varonai) and Jamaica (Xenothrix mcgregori). Fossil evidence of monkeys in the Greater Antilles extends back as far as the Early Miocene. The Hispaniolan monkey is similar to a living squirrel monkey (Saimiri). It was originally described as Saimiri bernensis but is now considered to be an endemic genus closely related to the Cuban monkey. The extinction of monkeys in the Greater Antilles was again probably the result of overhunting and deforestation by Amerindians. Living Fossils and “Island Shrews”

The surviving solenodons and very recently extinct (if they are extinct) island shrews are like “living fossils” from an earlier and long-extinct North American radiation. They are distributed on Hispaniola, Cuba (and the nearby Cayman Islands), and Puerto Rico (including adjacent Vieques Island). Presumably, solenodontids dispersed from North America to proto-Cuba between 70 and 80 million years ago in the Late Cretaceous, when the proto-Antilles and North America were last connected. They could have dispersed to what is now northern Hispaniola when eastern Cuba and northern Hispaniola were connected during the Oligocene. There is DNA evidence that Solenodon paradoxus and S. cubanus diverged 25 million years ago. This is about the time that eastern Cuba and northern Hispaniola began separating. The dispersal of Solenodon and Nesophontes to southern Haiti and the adjacent southern Dominican Republic could not have occurred until the Late Miocene or Early Pliocene, when the isolated “south island” crunched into true Hispaniola.

other placental mammals 76 million years ago. DNA analyses suggest that the two solenodons are so genetically divergent that the Cuban form could be considered as a separate genus (Atopogale). Solenodons never radiated into many species (an additional extinct species is known from each island). Solenodon paradoxus has a widespread distribution in the mountains and in appropriate lowland karst zones of southern Haiti and much of the Dominican Republic. In Cuba, Solenodon cubanus is now restricted to the high karst mountains of eastern Cuba. These living fossils have poor vision and are slow moving. They are easily preyed upon by dogs and the ever more abundant introduced mongoose, so their future survival, even under the best of conservation efforts, is questionable. ISLAND SHREWS

Endemic “island shrews” of the genus Nesophontes (Fig. 2) are part of the other soricomorph radiation in the Greater Antilles. The best current hypothesis is that solenodons and island shrews are closely related to each other. Cuba and Hispaniola each have three species of island shrew. In Hispaniola the remains of all three species of Nesophontes are abundant in some cave and sinkhole deposits under barn owl roosting sites. Some bones still have bits of dried tissue on them, and they are often mixed with the fresh-looking remains of introduced rats. The extinction of island shrews in both Cuba and Hispaniola was likely quite recent and may have been caused by competition and predation from black rats and mongooses. It is believed that the final extinction of Nesophontes in both Cuba and Hispaniola may have been as recent as the last 60 years. In Puerto Rico, there is just one, much larger species (Nesophontes edithae), which may have been an ecomorph of Solenodon. Remains are abundant on the island but are not nearly as recent in appearance as remains from Cuba and Hispaniola, and it is likely that this species became extinct before the arrival of Europeans in the Antilles, and before rats and dogs were introduced.

SOLENODONS

Solenodons are found on Cuba (Solenodon cubanus) and Hispaniola (Solenodon paradoxus). Both are now very rare and are threatened with extinction. Saving these forms from extinction has been given top priority by the Zoological Society of London’s EDGE scheme (Evolutionarily Distinct and Globally Endangered). They are truly evolutionarily distinct, having become isolated from all

FIGURE 2 Reconstruction of the largest Hispaniolan island shrew

(Nesophontes paramicrus).

ANTILLES, BIOLOGY

25

Other Antillean Mammals

Bats are the best known of the other major groups of Antillean mammals. There are 56 known extant species of bats in the Antilles, 28 (50%) of which are endemic. Cuba has the largest and most diverse bat fauna, with 26 living species (plus four extinct and three extirpated species). The next largest assemblage of bats is surprisingly (because it is so much smaller and less diverse than Hispaniola) found on Jamaica, with 21 extant species and three extirpated forms. The much larger island of Hispaniola has 18 extant species and three extinct. In the Lesser Antilles, most major islands have between 10 and 13 extant bat species, with small islands such as Saba (three species) and the Grenadines (four species) having many fewer. Fossil species known from the Lesser Antilles include one extinct form and three locally extinct species (but which still occur in the Greater Antilles). The five most common and widespread species of Antillean bats are Monophyllus redmani, Brachyphylla cavernarum, Artibeus jamaicensis, Noctilio leporinus, and Molossus molossus. Carnivores appear to be lacking as native species in the Antilles, although they are ecologically present in the form of feral dogs and cats, and the small Indian mongoose (Herpestes javanicus) that was introduced to Trinidad from India in 1870, and has now spread to 29 Antillean islands. The extinct “wild dogs” described from fossils in Cuban cave deposits (Cubacyon and Paracyon) are most likely remains of deformed domestic dogs. There are also reports of native raccoons in the Antilles. Recent molecular studies of the DNA of raccoons have confirmed that what were previously considered to be endemic island species (Procyon maynardi from the Bahamas, P. minor from Guadeloupe, and P. gloveralleni from Barbados) are conspecific with the North American raccoon Procyon lotor. BIRDS

There are 604 bird species in the Antilles, and 163 of these are endemic (27% endemism), some with very restricted distributions. In the Antilles, there are 36 endemic genera and two endemic families (palmchats [Dulidae] and todies [Todidae]). Within the boundaries of the Antilles, 48 bird species are currently threatened with extinction. Fourteen bird species and subspecies have already become extinct. Because the Antilles lacked predatory mammals, an unusually large number of birds became flightless during prehistoric times, and some birds filled predatory niches. There was at least one large ground-dwelling predatory owl (Ornimegalonyx oteroi), several other large 26

ANTILLES, BIOLOGY

barn owls, and huge birds of prey. There were 15 or 16 parrot-like species endemic to the Antilles, all of which are now extinct. Past climate and sea-level changes during the Pleistocene, as well as the impact of humans, introduced predators (dogs, cats, mongoose), and invasive species have combined to destroy most ground-nesting and flightless species. The Antilles, especially the Greater Antilles, is an important area for migratory birds. Many North American migrants spend the winter months on these islands. Many individuals return to specific locations each winter. This creates numerous biological challenges, such as competition for limited resources with native species. Some migrants such as black-throated blue warblers have developed winter strategies whereby males and females winter at different elevations and in different habitats. A small percentage of the population of some migrants remains behind each year (e.g., in the case of redstart warblers in Hispaniola and Cuba). There are subspecies of North American migrants that have become endemic breeding forms (e.g., pine warblers, yellow warblers, and whitewinged crossbills). One of the more interesting of sitespecific North American migrants is the Bicknell’s thrush, which nests on specific mountaintops in northern New England and winters in Hispaniola in specific locations, usually primary montane forests above 1000 m elevation. These examples reveal the complex mixture of resident, endemic, migratory, and modified migratory species that makes up the avifauna of the Antilles. The distribution of Hispaniolan birds reflects the way the island was formed. There are two species of palm tanagers, with the black-crowned palm tanager (Phaenicophilus palmarum) being restricted to northern Hispaniola (the old north island). The gray-crowned palm tanager (Phaenicophilus poliocephalus) likely arose from a dispersal of a small group of juvenile P. palmarum to a western piece of the south island in the Pleistocene. The gray-crowned palm tanager is still restricted to southwestern Haiti west of Jacmel (an old separate part of the south island). There are two species of endemic tody on Hispaniola. Cuba, Jamaica, and Puerto Rico each have only one species. There is also a species pair of high-mountain chat-tanagers (Calyptophilus) on Hispaniola. The western chat-tanager (C. tertius) is restricted to high mountain areas of the old south island. The eastern chat-tanager (C. frugivorus) is restricted to areas of the old north island in Haiti (including Gonave Island) and to the northern part of the Dominican Republic. The morphology and distribution of the western chat-tanager further reflects the past geographical history of the island. The western species can be

divided into a far western subspecies (C. t. tertius) in the Massif de la Hotte of Haiti and an eastern subspecies (C. t. selleanus) in the Massif de la Selle of Haiti and the Sierra de Bahoruco of the Dominican Republic, reflecting the division of the old south island into two separate sub-islands. The above examples of bird species, subspecies, and distributions that mirror the past geological and biological history of Hispaniola is biogeography at its living best. REPTILES AND AMPHIBIANS

There are 499 native species of reptiles in the Antilles, 468 (94%) of which are endemic. This high level of endemism is one important reason that the Antilles is considered as a world hotspot by many conservation organizations. The principal radiations of reptiles in the Antilles include several species swarms such as in the genus Anolis with 154 species, 150 (97%) of which are endemic. Other species swarms include the beautiful dwarf geckos of the genus Sphaerodactylus with 86 species, 82 (95%) of which are endemic (Fig. 3), and curly-tailed lizards (Leiocephalus) with 23 species, all of which are endemic to the Antilles. Other important adaptive radiations of reptiles in the Antilles include 11 species of large rock iguanas (Cyclura) that thrive in dryer areas of the islands, nine species of large boa constrictors of the genus Epicrates, many of which are large enough to kill and eat large capromyid rodents. Other Antillean snakes include racers (Alsophis) with 13 endemic species and boldly colored snakes of the genus Tropidophis with 26 endemic species. There are two species of crocodiles in the Antilles: Crocodylus acutus (American crocodile), found in Cuba, the Cayman Islands, Jamaica, Hispaniola, and possibly Martinique, and Crocodylus rhombifer (the Cuban crocodile), which is now restricted to the Zapata Swamp and to a large

FIGURE 3 A dwarf gecko (Sphaerodactylus) from the Massif de la

Hotte hotspot in southwestern Haiti. Courtesy of Charles Woods.

swamp in the central area of the nearby Isle of Youth. The Cuban crocodile has the smallest range of any species of crocodile in the world. The only poisonous snakes in the Antilles are pit vipers (fer-de-lance) of the genus Bothrops in Martinique and Guadeloupe (B. lanceolatus). Amphibians are not as diverse as reptiles in the Antilles. There are only 165 species, and all of them are frogs and toads. All but one of these frogs is endemic to the Antilles. Almost all of these frogs are from the single genus Eleutherodactylus (139 species). Frogs of this genus lay their eggs on land and hatch into adults without passing through a tadpole stage. Some species can be tiny, such as the Cuban species Eleutherodactylus iberia, one of the smallest tetrapods in the world (only 10 mm in length). The largest concentration of Eleutherodactylus in the Antilles is on mid-elevation slopes of the Massif de la Hotte of western Haiti, where 19 species occur sympatrically. The Alliance for Zero Extinction (AZE) has designated the Macaya Biosphere Reserve in this area of Haiti as the site in the world with the largest number of critically endangered species, 13 of which are Eleutherodactylus frogs. The most famous frog in the Antilles is the coqui of Puerto Rico (Eleutherodactylus coqui), whose loud vocalizations in the treetops are recognized as the typical sound of the night by almost all Puerto Ricans. In addition to frogs, the Antilles also has 12 species of endemic toads (Bufonidae), with eight species on Cuba, three on Hispaniola, and one on Puerto Rico. Most islands of the Greater and Lesser Antilles also have breeding populations of the introduced Bufo marinus. The greatest biodiversity of amphibians and reptiles in the Antilles is on Hispaniola (a pattern that is true for almost all Antillean flora and fauna). There are 217 species, 209 of which are endemic (96% endemism). On Hispaniola, there are 64 amphibian species, 62 of which are endemic, and 153 reptiles, 147 of which are endemic. The small frogs of the genus Eleutherodactylus, which have very limited abilities to disperse, have 28 species in Haiti’s southwestern mountain range (the Massif de la Hotte) and 15 species in the Central Mountains of the Dominican Republic. The two areas share only six species. The Massif de la Hotte has the highest density of frog species anywhere in the Antilles (Fig. 4). There, tall pine trees (Pinus occidentalis) form the forest canopy along with mid- and understory layers of other trees and shrubs rich in biodiversity. This relictual pine forest is in an area of tremendous rainfall and is the closest remaining forest in the Antilles structured like the long-lost moist tropical algarrobo forest of the Dominican Republic, the record of which is preserved in Dominican amber.

ANTILLES, BIOLOGY

27

BUTTERFLIES AND OTHER INSECTS

FIGURE 4 Photograph of Pic Macaya (right) and Pic Formon (left) in

the core area of the Massif de la Hotte of southwestern Haiti. Courtesy of Charles Woods.

FRESHWATER FISHES

The Antilles have over 160 species of freshwater fishes, but many of these are introduced species or species of marine origin that occupy freshwater and brackish habitats. The true “freshwater” fish fauna of the Antilles includes 71 species in nine families. Most (65) freshwater species are endemic. It is not a surprise that Cuba and Hispaniola, with their complex biomes, have the largest number of species (89% of the total freshwater fish fauna). Jamaica (six species), the Bahamas (five species), and the Cayman Islands (four species) have much smaller freshwater fish faunas than do Cuba (28 species) and Hispaniola (32 species), in spite of these islands being geographically close to one another. Puerto Rico, in spite of being close to Hispaniola and having a broad platform with wide coastal features, totally lacks a native freshwater fish fauna. A measure of the complexity of any analysis of freshwater fishes, which are often kept (and sold) as pets in the aquarium trade and are sometimes raised for food (i.e., Tilapia in fish farms), is the abundance of 24 well-established introduced freshwater fishes on Puerto Rico. The Lesser Antilles have many fewer freshwater fishes, with one species (Rivulus cryptocallus) endemic to Martinique, and a second (Rivulus ocellatus) widespread in the Greater Antilles and many islands in the Lesser Antilles. The most successful family of Antillean freshwater fishes is the live-bearing killifishes of the family Poeciliidae (five genera and 46 species). The complex freshwater fish fauna on the Antilles includes a total of 71 species, including 27 exotic species and five species that are of marine origin. The component of freshwater fishes derived from North America (seven genera, 27 species) is almost equal to the number derived from South America (five genera, 39 species). Almost all of these fishes are endemic to a single island, or at least to a single island group. 28

ANTILLES, BIOLOGY

There are about 350 butterflies known in the Antilles. In a pan-Antillean sense, the biogeography of butterflies is similar to the pattern observed for other groups. The greatest diversity of butterflies is on Hispaniola. Hispaniola also has the greatest number of species (201), 75 of which are endemic. The multipartite history of Hispaniola increases species numbers and diversity of butterflies just as has been observed in West Indian mammals, birds, reptiles, amphibians, and freshwater fishes. The north island–south island species dichotomy observed in other groups shows up clearly in butterflies too. For example, in the genus Callisto, 11 species originated on the old south island, and 22 species are associated with areas that were on the north island. There are several important questions that are still unresolved in studies on the biology of Antillean butterflies. One is how butterflies dispersed to the islands: specifically whether strong fliers dispersed (flew) over open water whereas generally poor fliers mainly dispersed to the Antilles by vicariance. The large number of endemic species of weak flying butterflies such as satyrids (genus Calisto) is in marked contrast with the distribution of strong fliers such as sphingids (hawkmoths) with 47 species, only seven of which are endemic. It is likely that weak-flying groups of butterflies are descendents of ancestors that arrived in the Antilles by vicariance events long ago, whereas strong fliers dispersed on numerous occasions by overwater dispersal. CONSERVATION BIOLOGY

The Antilles is recognized as one of the world’s most diverse and richest biological regions, with an unusually large number of island endemics. For example, of the 13,000 plant species, 6550 are endemic to single islands in the Antilles. This long (3200 km) curving arch of islands has an overall land area of 236,000 km2. The problem of protecting the biodiversity of the Antilles, however, is reflected in the equilibrium theory of island biogeography as well as in the size and shape of the islands. The slope of species-area curves for most species jumps sharply above a threshold of island size of about 3000 km2. On islands above this threshold, speciation exceeds immigration, and species proliferation increases. This is one reason that biodiversity is especially rich on Cuba, which has over 6505 plant species, 3224 of which are endemic. The percentage of endemic plant species on Cuba alone is 54% of all of the endemic plants of the Antilles. Thus, as a consequence of the size (105,806 km2) and shape (numerous offshore archipelagoes) of Cuba,

it is geographically and biologically advantaged over smaller islands of the Antilles. Indeed, about 50% of the entire land area of the Antilles is found on Cuba alone. Ninety percent of the area of the Antilles occurs on the four major islands of the archipelago (Cuba, Hispaniola [73,929 km2], Jamaica [11,190 km2], and Puerto Rico [9100 km2]). But island size alone does not determine the chances of retaining an islands biodiversity. Other important influences are cultural history, population size and distribution, and the percentage of an islands area that is “protected.” Once again, Cuba (with a population of 11.5 million and a population density of 103 people per km2) sets the standard for the Antilles, with 15% of its area being “protected” (about the same percentage of the island that remains forested). Overall, in the Antilles, 12.9% of the area is “protected” (although only 7.1% is designated conserved in IUCN categories I–IV). Keeping in mind that sustainable biodiversity requires habitat (remember the lesson of species-area curves), what are the prospects for biodiversity in other parts of the Antilles? Island size alone is not the critical indicator, but rather population density and the presence of suitable habitat. For example, in the Dominican Republic, with a population of over 9 million (186 people per km2), 20% of the land area of the country is “protected” (19 national parks, six scientific reserves, 15 natural preserves, two marine sanctuaries). In adjacent Haiti, with a population of 8.7 million (population density 316 people per km2), only 1.7% of the land area is protected in three barely functional national parks. The standard for protected land area in the Antilles is the small Lesser Antillean island of Dominica (754 km2) with a population of 71,540 (a population density of 101 people per km2 and declining), which protects 21.4% of the island in three national parks. Barbados and Aruba have preserved less that 1% of their land areas, and tiny Anguilla (only 102 km2) is developing its first national park (Fountain Cavern). SEE ALSO THE FOLLOWING ARTICLES

Antilles, Geology / Freshwater Habitats / Lizard Radiations / Mammal Radiations / Vicariance

FURTHER READING

Crother, B. I. 1999. Caribbean amphibians and reptiles. San Diego, CA: Academic Press. Fernández, E. 2007. Hispaniola: a photographic journey through island biodiversity/Bioversidad a través de un recorrido fotografico [bilingual ed.]. Cambridge, MA: Harvard University Press. Poinar, G., and R. Poinar. 1999. The Amber Forest. Princeton, NJ: Princeton University Press.

Raffaele, H., J. Wiley, O. Garrido, A. Keith, and J. Raffaele. 1998. Field guide to the birds of the West Indies. Princeton, NJ: Princeton University Press. Ricklefs, R., and E. Bermingham. 2007. The West Indies as a laboratory of biogeography and evolution. Philosophical Transactions of the Royal Society of London B—Biological Sciences. doi: 10.1098/rstb.2007.2068. Schwartz, A., and R. W. Henderson. 1991. Amphibians and reptiles of the West Indies: descriptions, distributions, and natural history. Gainesville: University of Florida Press. Sergile, F. E. 2005. A la découverte des oiseaux d’Haïti. Port-au-Prince: Société Audubon Haïti. Smith, D. S., L. D. Miller, and J. Y. Miller. 1994. The butterflies of the West Indies and South Florida. Oxford: Oxford University Press. Woods, C. A. 1989. Biogeography of the West Indies: past, present and future. Gainesville, FL: Sandhill Crane Press. Woods, C. A., and F. E. Sergile. 2001. Biogeography of the West Indies: patterns and perspectives, 2nd ed. Boca Raton, FL: CRC Press.

ANTILLES, GEOLOGY RICHARD E. A. ROBERTSON University of the West Indies, St. Augustine, Trinidad and Tobago

Located at the eastern edge of the Caribbean plate where it borders with the North and South America plates, the Lesser Antilles is a region of high seismicity, tectonism, and active volcanism that typifies oceanic-island arc volcanism. The island chain forms an 850-km arc that is convex toward the Atlantic and extends from Sombrero in the north to the southern island of Grenada. The arc splits north of St. Lucia into two: an Eocene-to-Miocene eastern arc and a Pliocene-to-Recent western arc. The northeastern islands extending from Marie Galante to Sombrero are characterized by Cenozoic limestones and are called the Limestone Caribbees. The inner arc extending from Grenada to Saba comprises the present day volcanic front and is called the Volcanic Caribbees. These have been mainly active from Eocene to midOligocene (Fig. 1). The entire region is one of active volcanism, which exhibits both explosive and effusive activity and which currently poses a threat to vulnerable island communities throughout the Eastern Caribbean. STRUCTURE AND TECTONICS

The main structural components of the arc consist of the Atlantic oceanic crust; a fore-arc upon which is built an accretionary wedge (with the island of Barbados as the exposed top); the island arc itself; and the Grenada basin behind the arc. The dominant tectonic process is active

ANTILLES, GEOLOGY

29

o 62 W

Anguilla St. Martin

o 18 N

St. Bartholomew Barbuda

Saba St. Eustatius St. Kitts Nevis

Antigua

Redonda Montserrat Desirade o Marie 16 N Galante

Guadeloupe

Pleistocene

Dominica

Pliocene Miocene Limestone

Guadeloupe

Eo-Oligocene

St. Lucia

o 14 N

St. Vincent Mustique Canouan Carriacou Kick 'em Jenny Grenada

o 12 N

FIGURE 1 Map of the Lesser Antilles island arc, showing the ages of

the exposed rocks and the positions of the volcanic front during the Eocene-Oligocene (red line), Pliocene (blue line), and Pleistocene (black line). The dashed line is the 200-m isobath (adapted from Fig. 9.3 of Wadge 1994).

crustal subduction with underthrusting of the Caribbean plate, by the Atlantic Ocean crust of Cretaceous to Jurassic age along the axis of the arc (Fig. 2). Subduction rates are estimated to be 2.2–3.8 cm/year. Plate subduction with volcanism commenced along the arc after cessation of similar activity along the Aves ridge to the west (at about the time of the Oligocene). The island arc is bounded to the north and northwest by the Greater Antilles, which consists of deformed and metamorphosed sediments and volcanics of Jurassic to Eocene age located to the northwest of the arc. The Puerto Rico trench separates the Greater Antilles from the island arc. To the west, the island arc is bounded by the Venezuela basin and Grenada trough, between which lies the north–south striking Aves ridge, the site of active subduction during the Upper Cretaceous. Investigations of regional tectonics have indicated that the Caribbean plate is moving eastward relative to the American plate, although the velocity and sense of

30

ANTILLES, GEOLOGY

movement is debated. The suggested direction of relative movement has included east–west, left-lateral strike slip to northeast–southwest oblique convergence. Various estimates exist for the rate of convergence within the region, depending on the period considered and the specific part of the arc examined. All estimates suggest that the rate of convergence in the Lesser Antilles (from as low as 1.3 cm per year to as high as 4 cm per year) is lower than in most arc systems. The Benioff zone has an average dip of 45° and is at 100- to 120-km depth beneath the active volcanic arc. The configuration of the Benioff zone beneath the Lesser Antilles has been established using the hypocentral locations of earthquakes recorded from 1978 to 1984. The arc is segmented: To the north of Martinique, the Benioff zone dips at 60–50° and trends about 330°. In contrast, the southern segment trends at 20°, dips at 50–45° in the north, and is vertical in the south at Grenada. Gravity anomalies occur as an arcuate pair associated with the fore-arc (negative anomalies) and the arc massif (positive anomalies). The anomalies are due to either the presence of dense igneous rocks (positive anomalies) or the presence of thick, low-density sedimentary rocks (negative anomalies). Crustal thickness beneath the arc is about 30 km and has typical island arc and oceanic crustal seismic velocities. Seismic reflection data indicate that the arc crust consists of three layers: an upper layer of sedimentary and volcanic rocks, a middle layer of intermediate plutonic rocks, and a lower layer of basic rocks. Gravity anomalies supported by seismic refraction data indicate significant differences between the northern and southern segments of the arc. EVOLUTION AND STRATIGRAPHY

The Caribbean plate was generated over the Galápagos hotspot 100–75 million years ago and was inserted between the North American and South American plates sometime between the Late Campanian and Late Eocene. A subduction zone and associated volcanic arc developed at its leading edge, and volcanic activity, now represented by the Aves ridge, extended from the Upper Turonian to the Lower Palaeocene. The Lesser Antilles volcanic arc developed in the area where the accretionary prism associated with the Aves ridge developed. The Aves ridge accretionary prism was separated from the active volcanic arc by the development of a back-arc basin, now represented by the Grenada trough, a 3-km-deep basin, which contains a sedimentary pile estimated to vary from 4.2–7 km to 7–12 km thick.

ST. VINCENT (Active Island Arc)

AVES RIDGE EXTINCT ISLAND ARC

FAULTS

SEA LEVEL

0

EAST BARBADOS

SEA FLOOR

DETACHMENT FAULT

0

SEDIMENT VOLCANIC ROCKS

BASEMENT

20

20 BASEMENT

ACCRETIONARY SEDIMENT PILE

CARIBBEAN PLATE

40 BASE OF CRUST

80 0

RISING MAGMA

EL TI NG

60 S

H UT O

IC ER AM

E PLAT AN

40 60

VERTICAL EXAGGERATION 2:1

M

DEPTH IN KILOMETERS

WEST

100

200

300 400 DISTANCE IN KILOMETERS

500

600

80

700

FIGURE 2 Cross-section drawn through the Lesser Antilles at the latitude of St. Vincent, showing the subduction of the South American plate

underneath the Caribbean plate (after Westerbrook et al. 1984).

Basement rocks in the Lesser Antilles consist of Cretaceous/Palaeocene island arc rocks that underlie much of the arc from Guadeloupe northward, but much of the evidence for pre-Cenozoic arc rocks is fragmentary and difficult to correlate. Barbados has a unique position in the arc as the top of the accretionary wedge of sediments scraped off the ocean floor during the subduction process. The island is composed mostly of Pleistocene reef limestones that have been rapidly uplifted, but there are two groups of older accretionary rocks exposed on the northeast coast. The Limestone Caribbees, which extend from Anguilla to Marie Galante, are composed of Cenozoic limestones of varying ages (Eocene in St. Barthelemy; Oligocene on Antigua; Miocene on Anguilla and St. Martin; and Pliocene-Quaternary on Barbuda, Guadeloupe, and Marie Galante). They are underlain by volcanic rocks that represent the older, eastern branch of the arc where volcanism ceased millions of years before. Generally, the rocks exposed on these islands increase in age as one moves northward along this part of the island chain. The Volcanic Caribbees, which extend from Grenada to Saba, represent the areas where magma produced by the subduction process reaches the surface. The oldest exposed rocks date back to the Eocene and occur in the islands from Martinique southward. The age of the rocks in the islands north of Martinique extend only to the Miocene, and in some cases to the Early Pliocene. In the southernmost part of the arc, Grenada and the Grenadines, sedimentary rocks of Middle Eocene to Middle Miocene age, are abundant. Evidence for the transition from the eastern Limestone Caribbees front to the current Volcanic Caribbees is not well preserved on the islands. Only in Martinique, where a record of almost continuous migration of volcanic activity

from the tholeiitic products in the east (∼16 million years ago) to calc-alkaline rocks in the west (∼6 million years ago) exists, is the evidence well preserved. In fact, determination of stratigraphic relationships in the arc is fraught with problems given the abundance of volcanic deposits and the essentially point-source evidence for arc evolution provided by the islands. GEOCHEMISTRY

The Lesser Antilles arc consists of three geochemically and structurally distinct zones. The northern segment, from Saba to Montserrat, contains andesites and minor dacites and belongs to the island-arc tholeiitic magma suite. The islands of this group have low volumes of basalts and rare rhyolites (St. Kitts and St. Eustatius). The central group (Guadeloupe to St. Lucia) contains the most prolific volcanoes in the Quaternary and has total erupted volumes among the largest in the Lesser Antilles. The predominant rock type is again andesites, with some basalts and dacites and rare rhyolites, but the magmas belong to the calc-alkaline magma suite. The southern group extends from St. Vincent to Grenada and consists of predominantly basalts and basaltic andesites with rare andesites. This group includes an alkalic suite of magmas associated with highly undersaturated lavas enriched in incompatible and transition elements. In addition to major rock types outlined above, three types of plutonic nodules have been found among the strata exposed in the Lesser Antilles. Cognate inclusions are phenocryst clusters and fine-grained-to-porphyritic crystal clots of differing textures to the host magma. Metamorphic xenoliths are cordierite-bearing hornfels and metasediments with relict bedding and cross-stratification. Rare samples of these xenoliths have been found at the Soufrière volcano. Finally, there are plutonic cumulate inclusions and nodules.

ANTILLES, GEOLOGY

31

Cumulate-textured blocks are a common occurrence in most islands of the Lesser Antilles. The blocks vary in size from 1 cm to several tens of centimeters and are particularly abundant in some areas (e.g., the Soufrière of St. Vincent). The rocks exhibit a wide variety of textures and mineralogies with plagioclase and amphibole being dominant in most islands. SEISMICITY

The Eastern Caribbean is significantly seismically active (Fig. 3). Tectonic earthquakes associated with the subduction process and volcanic earthquakes associated with the rise of magma are the two types of earthquakes experienced in the region. As noted previously, the hypocenters of the tectonic earthquakes define the shape of the subducting plate or the Wadati-Benioff zone. Energy release from major historical earthquakes indicates a slip rate of 1–5 mm/year, significantly less than the 20 mm/year predicted from global plate tectonic models. The relatively slow plate convergence rate of 2 cm per year contributes to long intervals between the largest earthquakes generated by the system. These earthquakes occur in the rigid crustal material on either side of the colli-

sion boundary and within the descending slab. In the subduction zone environment, earthquakes within the crust are described as shallow, and those occurring in the descending slab are described as occurring at intermediate depth or as being deep. Annually, there are over 1000 earthquakes recorded by the seismograph networks in the Eastern Caribbean with epicenters located in the Lesser Antilles island arc. The general pattern observed during the more than 50 years that continuous monitoring has been taking place is a broad zone of shallow seismicity with better defined, overlapping bands of intermediate depth and deeper seismicity. In general, the deepest events occur to the west of the arc. Earthquake activity in the Eastern Caribbean is not distributed uniformly throughout the region, and some areas exhibit more intense activity than others do. The zones near Antigua, north of the Paria Peninsula and Gulf of Paria, are the areas where higher levels of seismicity are manifested. The lowest level of seismicity is seen in the area from Grenada to St. Lucia. That pattern has been attributed to a smoothly descending slab or to the accumulation of strain energy, which is yet to reach its limit. The area around Barbados also displays a relatively low level of seismic output, which is considered consistent with its location away from active subduction. VOLCANISM

FIGURE 3 Eastern Caribbean earthquakes for the period July 1, 2004,

to July 31, 2006. (From Fig. 3 of Seismic Research Centre, 2007).

32

ANTILLES, GEOLOGY

Volcanism in the Lesser Antilles dates back as far as the Eocene, and its general nature appears to have remained unchanged throughout. Volcanic centers exhibit a wide range of isotopic and chemical compositions, which reflect the variety and nature of sources and evolutionary processes that led to their genesis. Westward translation of the volcanic arc occurred in the northern islands during the Miocene, but in the south, new volcanic centers formed adjacent to the older ones. Although the nature of volcanism has remained unchanged with time, there has been migration of the center of activity within islands. There are at least three examples of progressive intra-island migration of volcanism during the Plio-Pliestocene period in the Lesser Antilles: St. Kitts (northward), Guadeloupe (southward), and St. Vincent (northward). In each case, migration (at rates of 4 to 10 km/million years), may represent the movement of a single magma source or plume trace along the active front, creating lines of volcanoes with linearly decreasing age. The largest volcanoes occur in the central part of the arc extending from St. Vincent to Guadeloupe. These have created large islands from overlapping volcanic deposits produced by repeated eruptions of volcanic cen-

ters. Large, mature, and complex stratovolcanic centers have been created on these islands usually with a central core of lava domes surrounded by primary and reworked pyroclastic deposits. Determination of stratigraphic relationships is often difficult and has relied in the past heavily on the incidence of dateable charcoal created by the carbonization of wood during pyroclastic flows. The islands from Montserrat to Saba in the northern segment of the arc are smaller and are comprised of loweraltitude volcanoes that appear incapable of producing the size of eruptions preserved in the rock record in the central islands. One effect of the smaller size of volcanic mountains is the greater interaction with the sea. As such, on St. Kitts and St. Eustatius there are raised limestone platforms created by the uplifting of shallow, submarine shelves during the emplacement of cryptodomes. VOLCANIC HAZARDS

Pleistocene-to-Recent volcanoes (occurring less than 2 million years ago) occurs in narrow zones (less than 10 km wide), which appear to define three segments: Saba to Montserrat, Guadeloupe to Martinique, and St. Lucia to Grenada. Active volcanism has been characterized both by effusive eruptions producing lava flows and domes and by explosive eruptions producing various types of pyroclastic deposits. The total volumetric volcanic production over the past 0.1 million years is symmetrical about Dominica, which has produced ∼40 km3 of volcanic deposits. This compares with 8 km3 for Guadeloupe and Martinique and 0–5 km3 for the islands located to the north and south of these central islands. The mean spacing between active volcanoes during the past 0.1 million years is in the range of 15 to 125 km. The Lesser Antilles contain 21 live volcanoes distributed among 11 volcanically active islands (Fig. 4). There have been at least 34 eruptions of Lesser Antilles volcanoes (Table 1) during the “historic period” (the past 400 years), and currently about 1 million people live within the areas that can be affected by the direct impacts of eruptions in the future. Volcanic eruptions have killed over 30,000 people in the past, and volcanic hazards are some of the main geologic hazards that threaten the Eastern Caribbean region. All of the islands of the Volcanic Caribbees have at least one volcano that may erupt in the future. Twenty-one of the historic eruptions have occurred since 1900: nine on land from volcanoes on Guadeloupe, Martinique, St. Vincent, Montserrat, and Dominica, and 12 from the submarine volcano Kick ’em Jenny, ∼9 km north of Grenada. These eruptions have all shown a wide variety of eruptive style, magnitude, and impact on the local population. Several eruptions have been phreatic in nature, involv-

FIGURE 4 The Lesser Antilles region showing the location of the 21

volcanic centers considered active or potentially active. The islands that make up the Volcanic Caribbees are shown in brown and those of the Limestone Caribbees in yellow.

ing the interaction of groundwater with rising magma. One of these was a minor phreatic eruption in Dominica in 1997 that went largely unnoticed; two were much more serious phreatic eruptions in Guadeloupe in 1956 and 1976–1977. The 1902–1907 magmatic eruption of Montagne Pelée on Martinique was characterized by both effusive dome formation and explosive dome collapse, and led to the total destruction of the town of St. Pierre and the deaths of approximately 30,000 people. A similar eruption occurred from Montagne Pelée several years later, between 1929 and 1932, this time with no reported casualties. Explosive magmatic eruptions have included the 1902 eruption from the Soufrière in St. Vincent that resulted in the deaths of at least 1500 people. In contrast, the 1971– 1972 eruption of this volcano was an effusive magmatic

ANTILLES, GEOLOGY

33

TABLE 1

Historical Volcanic Eruptions of the Lesser Antilles Country

Volcano

Date

Mt. Scenery

∼1670 (280 ± 80 years BP)

Montserrat

Soufrière Hills Soufrière Hills

1667 ± 40 (285 years BP; n =10) 1995–present

Dome-forming Dome-forming

Guadeloupe

La Soufrière La Soufrière La Soufrière La Soufrière La Soufrière La Soufrière

1690 1797–1798 1812 1836–1837 1956 1976–1977

Phreatic Phreatic Phreatic Phreatic Phreatic Phreatic

Dominica

Valley of Desolation Valley of Desolation

1880 1997

Phreatic Phreatic

Martinique

Montagne Pelée Montagne Pelée Montagne Pelée Montagne Pelée

1792 1851–1852 1902–1905 1929–1932

Phreatic Phreatic Explosive and dome-forming Explosive and dome-forming

St. Lucia

Soufrière Volcanic Center

1766

Series of phreatic eruptions

St. Vincent

The Soufrière The Soufrière The Soufrière The Soufrière The Soufrière The Soufrière

1718 1780 1812–1814 1902–1903 1971–1972 1979

Explosive magmatic Dome-forming Explosive magmatic Explosive magmatic Dome-forming Phreatomagmatic and dome-forming

Grenada

Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny Kick ’em Jenny

July 24, 1939 October 5–6, 1943 October 30, 1953 October 24, 1965 May 5–7, 1966 August 3–6, 1966 July 5, 1972 September 6, 1974 January 14, 1977 December 29–30, 1988

Kick ’em Jenny Kick ’em Jenny

March 26–April 5, 1990 December 4–6, 2001

Phreatomagmatic Submarine Submarine Submarine Submarine Submarine Submarine Phreatomagmatic Submarine, dome-forming Phreatomagmatic, dome-collapsing Submarine Submarine

Saba

Type of Eruption a

Explosive b

note: From Lindsay et al. 2005. a There are no written accounts of this eruption, but radiocarbon dates place it well within the historical period for the region. b There are no written accounts of this eruption, but radiocarbon dates place it well within the historical period for the region.

eruption that resulted in the formation of a lava dome confined within the summit crater. The 1979 eruption of the Soufrière was again explosive but was followed by dome growth, and although there was some property damage, no lives were lost. The 12 submarine eruptions from Kick ’em Jenny are believed to have been dominantly explosive, although in at least one case a lava dome was extruded. The Soufrière Hills volcano on Montserrat has been in active eruption since 1995 and has had a major impact on the island’s population. The eruption is characterized by periods of dome growth interspersed with dome collapse and minor 34

ANTILLES, GEOLOGY

explosions. The Soufrière Hills volcano is the only volcano currently erupting in the Eastern Caribbean. VOLCANO MONITORING IN THE LESSER ANTILLES

The responsibility for monitoring volcanic and seismic activity in the Lesser Antilles is divided between three main organizations. The Seismic Research Centre, which is part of the University of the West Indies (UWI), is based in St. Augustine, Trinidad, and is responsible for monitoring activity in the independent Commonwealth countries of the Lesser Antilles, namely St. Kitts and

Nevis, Dominica, St. Lucia, St. Vincent and the Grenadines, and Grenada. The Institut de Physique du Globe de Paris (IPGP) monitors volcanic activity in Martinique and Guadeloupe, and the Montserrat Volcano Observatory (MVO) operates a monitoring network in Montserrat. In all of the islands of the Lesser Antilles, these agencies work closely with the civil authorities (typically known locally as national disaster preparedness organizations), which represent the respective local governments. The mainstay of all volcanic monitoring in the Lesser Antilles is the seismograph network. The Seismic Research Centre of UWI maintains 40 seismic stations in the volcanic islands for which they are responsible; these are located near the 18 live volcanoes spread across these countries. The IPGP maintains eight stations in Guadeloupe and eight in Martinique to monitor La Soufrière and Montagne Pelée, respectively. The seismograph network on Montserrat comprises 11 stations, eight maintained by the MVO and three by the Seismic Research Centre. All these stations form part of the regional seismograph network, which includes a further 16 UWI stations on the surrounding non-volcanic islands (Trinidad, Tobago, Barbados, Antigua, Barbuda, St. Martin); nine stations in northeast Venezuela maintained by the Universidad de Oriente, Cumana, Venezuela; and the Fundacion Venezolana de Investigaciones Sismologicas and several French stations in eastern Guadeloupe and southern Martinique. In addition to seismic monitoring, programs of volcanic gas surveillance and ground deformation monitoring are also maintained in the volcanic islands of the Lesser Antilles. SEE ALSO THE FOLLOWING ARTICLES

Antilles, Biology / Earthquakes / Island Arcs / Kick ’em Jenny / Lava and Ash FURTHER READING

Biju-Duval, B., and J. C. Moore. 1984. Initial report of the Deep Sea Drilling Project. 78A. Washington, DC: Government Printing Office. Briden, J. C., D. C. Rex, A. M. Faller, and J. F. Tomblin. 1979. K-Ar geochronology and palaeomagnetism of volcanic rocks in the Lesser Antilles island arc. Philosophical Transactions of the Royal Society of London A291: 485–528. Brown, G. M., J. G. Holland, H. Sigurdsson, and J. F. Tomblin. 1977. Geochemistry of the Lesser Antilles volcanic island arc. Geochimica et Cosmochimica Acta 41: 785–801. Fox, P. J., and B. C. Heezen. 1975. Geology of the Caribbean crust. The Ocean Basins and Margins series, volume 3. A. E. M. Nair and F. G. Stehli, eds. New York: Plenum. Lindsay, J. M., R. E. A. Robertson, J. B. Shepherd, and S. Ali, eds. 2005. Volcanic hazard atlas of the Lesser Antilles. Trinidad and Tobago: Seismic Research Unit,University of the West Indies. Macdonald, R., C. J. Hawkesworth, and E. Heath. 2000. The Lesser Antilles volcanic chain: a study in arc magmatism. Earth-Science Reviews 49: 1–76.

Pindell, J. L., and S. F. Barrett. 1990. Geological evolution of the Caribbean region: a plate tectonic perspective, in The geology of North America, Vol. H, The Caribbean region. G. Dengo and J. E. Case, eds. Boulder, CO: Geological Society of America, 405–432. Speed, R. C., and G. K. Westbrook. 1984. Lesser Antilles and adjacent terranes, Atlas 10, Ocean Margin Drilling Program regional atlas series 28. Woods Hole, MA: Marine Science International. Wadge, G. 1994. The Lesser Antilles, in Caribbean geology: an introduction. S. K. Donovan and T. A. Jackson, eds. Kingston, Jamaica: UWI Publishers Association. Westerbrook, G. K., A. Mauffret, A. Munschy, R. Jackson, B. BijuDival, A. Mascle, and J. W. Ladd. 1984. Thickness of sediments above acoustic basement, in Lesser Antilles arc and adjacent terranes, Atlas 10, Ocean Margin Drilling Program regional atlas series. R. C. Speed and G. K. Westbrook, eds. Woods Hole, MA: Marine Science International.

ANTS BRIAN L. FISHER California Academy of Sciences, San Francisco

The rise of ants to ecological dominance has been called one of the great epics in evolution. The same features associated with their ecological success also make them destructive invaders. Islands provide an exceptional model for studying ant dispersal, extinction, and radiation. Ants often reach oceanic islands via accidental “sweepstake routes,” leading to a unique cluster of ant species on different islands. The chance dispersal to islands results in high species turnover between islands and within islands over time. The composition of the ant fauna on any particular island can therefore reflect the age, size, and relative isolation of the island. At the same time, the limited land area and biodiversity of islands also increase their vulnerability to incursions by invasive ant species. Increased habitat fragmentation and the accelerated pace of ant species introductions put endemic island ecosystems at increased risk for invasional meltdown. ISLAND ANTS

Ants are the glue that holds ecosystems together. These social insects dominate almost every terrestrial habitat throughout the world, in terms of both sheer numbers and ecological interactions. This dominance is particularly remarkable because ants constitute only about 1% of all described insect species. Understanding the processes driving the phenomenal success of ants is an active area of research. Current techniques involve the careful analysis

ANTS

35

of species distributions as well as the historical and geographical factors affecting dispersal and radiation. Overall, the study of ant ecology is limited by the fact that up to half of the estimated 20,000 species of ants in the world have yet to be described. However, because islands are much smaller in area and harbor less diverse faunas than do continents, an exhaustive inventory of their ant species is feasible. Analysis of this more limited assemblage can then shed light on processes that affect ant composition and dispersal. In general, the bigger an island, the more diverse its assemblage of ants. This is certainly true for the world’s three largest tropical islands: New Guinea, Borneo, and Madagascar. It is no coincidence that these three land masses also feature more endemic ant genera and species than any other islands on Earth. Chance Dispersal

Ants on oceanic islands often arrive via a so-called “sweepstakes route,” a term that aptly describes the rarity of a successful island landing but also the huge potential payoff. Ant species that do manage to establish themselves on a large island can often then radiate to fill many empty ecological niches. Island ants typically arrive via one of four common dispersal routes. In many ant species, the reproductive form is a winged queen. Newly inseminated queens taking wing to establish a new colony can be blown across the open ocean to distant shores. Alternatively, an entire ant colony can raft to an island inside a rotten log or tree washed out to sea during a storm. Some islands have received new fauna from the mainland or other islands via land bridges exposed during periods of low sea level. Finally, humans have been transporting ants inadvertently wherever they travel, including on island voyages. An island’s geography can be the determining factor in whether or not dispersing ants can land and establish a foothold. Islands in proximity to other sources of ants are easily reached by prevailing winds or ocean currents and will be colonized more often. The older an island, the more time ants will have had to arrive and establish themselves. By the same token, larger islands offer a bigger target to dispersers. The extreme isolation and relative youth of the Polynesian islands east of Samoa, including Hawaii, have placed these islands among the few places on Earth that lack native ant species. On these islands, ants have had little opportunity to arrive on their own. In fact, Hawaii’s native fauna includes no social insects of any kind. Yet today, 50 ant species are established in Hawaii, having

36

ANTS

FIGURE 1 Endemic

species

on

islands

such

as

Mauritius

and

Madagascar are characterized by species with wingless reproductive queens. These colonies reproduce by fission, when the newly inseminated queen walks away from the parent nest with a few workers to start her new colony. (A) Odontomachus coquereli; (B) Aphaenogaster sp.; (C) Pristomyrmex bispinosus.

been brought to the islands by humans over the past century. Among them are some of the world’s most widespread and damaging invasive species (see below). These recent arrivals have been devastating the highly endemic arthropod fauna, which never evolved defenses against ants and lacks specialized ant predators. In contrast, as a very old island long isolated in the southwestern Indian Ocean, Madagascar today harbors an unusually diverse ant fauna. Of its more than 1000 ant species, over 95% are endemic to the island. Madagascar originally formed part of the ancient supercontinent of Gondwana but

broke away from Africa approximately 120 million years ago. Its assemblage of native ant species likely evolved only after this breakup. Biologists now believe the extant ant lineages on Madagascar arrived via oceanic dispersal, primarily from Africa, but also from Asia. Although Madagascar is much closer to Africa, ocean and wind currents from the east may explain the connection to Southeast Asia. On the other side of the world, the Antilles, which arc across the Caribbean in a chain of more than 7000 islands, provide another good example of sweepstakes colonization. Here, islands at increasing distance from the mainland show a corresponding drop in the number of ant genera. Moreover, the larger the size of an island, the greater the number of endemic species it contains; few to no endemics live on Caribbean islands under 1000 km2. The exception is Trinidad, a large continental island just 11 km from the mainland of Venezuela. The ant fauna there is an extension of species found in South America and includes 17 genera widespread on the continent but absent from the rest of the archipelago. The existing ant community, vegetation, and habitat of the island also help shape how hospitable the island will be once an ant makes landfall. Despite the African and Asian origins of Madagascar ants, the ant fauna is quite unique when compared to the faunas of neighboring continents. Several of the ant species that dominate ecosystems in Africa and Asia are absent from Madagascar. Among these are army ants (Aenictus, Dorylus) of the forest floor and weaver ants (Oecophylla) of the forest canopy. Weaver and army ants, especially of the genus Dorylus (driver ants), are major predators of other ants. Their presence influences the structure of ant populations as well as the diversity of ant communities. The absence of such keystone species is a common feature of many other islands (e.g., Cuba, Hispaniola, Fiji) where ants have radiated. These island systems thus constitute a natural experiment for evaluating how these dominant ants affect biological communities and how their absence allows the diversification of some unusual groups. The characteristics of an individual species may also affect how likely it is to become a pioneer. For example, in most continental species, ants reproduce and disperse through a winged queen. After mating, she flies to a new location to establish a new colony. Winged queens are clearly an advantage for dispersing to an island. Interestingly, winged queens may become a drawback once a species has reached an island, because winged queens are more likely to be blown offshore and into the ocean. This may explain why many endemic ant species of the southwestern Indian Ocean islands have evolved wingless queens (Figs. 1, 2).

FIGURE 2 Endemic species with wingless reproductive queens. (A)

Terataner sp.; (B) Mystrium mysticum; (C) Cerapachys sp.

Radiation

Once an ant species establishes on an island, it may undergo adaptive radiation to fill vacant niches. The number of endemic species tends to be greater on older and larger islands, where ants have had more time to evolve and there has been a greater complexity of local habitats to occupy. In the Caribbean, for example, endemic ants make up a disproportionate share of the ant faunas of both Cuba and Hispaniola. Endemic radiations of a single genus, Temnothorax, now constitute more than 25% of the ant fauna in Cuba alone. The group includes species that have become specialists at nesting in habitats such as soil, limestone crevices, or epiphytic plants. The stunning morphological diversity of these species is comparable to the range usually seen in several genera. The absence of

ANTS

37

army or driver ants on these islands may have encouraged this evolutionary profusion. Likewise, on the island of Fiji, the diversification of the genus Leptogenys might have been possible because of the absence of army ants and the relatively low number of other endemic species from similar genera. On Madagascar, many groups of ants have undergone an equally spectacular radiation. The five most speciesrich ant genera on the island, Camponotus, Hypoponera, Pheidole, Strumigenys, and Tetramorium, all contain over 100 species. Each group exhibits remarkable morphological and niche diversity. Local diversity is also amazing. For example, on the Masoala Peninsula alone, 98 species from these five genera co-occur. As on Hispaniola, the absence of army and weaver ants likely allowed certain lineages to persist and others to radiate and flourish on the island. An example is the diversification of the tribe Cerapachyinae (Cerapachys and Simopone), which includes an unprecedented, morphologically diverse assemblage of more than 50 species on Madagascar. In fact, certain Cerapachys species show morphological similarity to the army ant genus Aenictus found in Africa. Whether the absence of army ants led to the diversification of Cerapachys in Madagascar, or simply permitted their persistence, remains unclear. Another remarkable example of ant radiation in Madagascar is illustrated by the two closely related genera of the ant tribe Dacetini. Dacetine ants rely on the traplike action of their mandibles to capture and subdue live food. Most of the differences between species reflect various methods of seizing prey. With 89 described species in Madagascar, the dacetines are the island’s dominant predatory leaf-litter insect. Local diversity, too, is off the charts. For example, 25 species of dacetines have been recorded in an area roughly 1 km2 on the Masoala Peninsula. The ecological context in which ant colonists find themselves may also influence whether or not a lineage has the opportunity to radiate. For example, the relative diversity of Strumigenys and Pyramica in Africa is very different from their diversity on Madagascar. These specialized trap-jaw predators have oddly shaped mandibles and pear-shaped heads, and are often covered in bizarre hairs and strange outgrowths of whitish sponge-like tissue around their waist segments. Strumigenys are quite diverse in Madagascar, with 74 species on the island compared to 50 in Africa. Pyramica, in contrast, are much more diverse on the continent, with 81 species in Africa and only 15 in Madagascar. Why Strumigenys has undergone this diverse island radiation whereas Pyramica has not remains somewhat of a mystery. Maybe the first Pyramica arrived after Strumigenys had already radiated and filled potential niches.

38

ANTS

Taxon Cycle

While studying ants in Melanesia, E. O. Wilson observed that species pass through sequential phases of expansion and contraction in distribution. He coined the term “taxon cycle” to describe the phenomenon. On islands, expanding taxa tend to be recent arrivals from the mainland that occupy lowland habitats along the coastlines of islands. By contrast, contracting taxa have reduced and fragmented ranges and tend to occupy interior and montane habitats. These differences suggest that existing species in the contraction phase of the cycle are pushed upslope and into new habitats by competition from more recent arrivals along the coast. This pattern certainly holds in Madagascar, where endemic and possibly older ant groups are restricted to mountaintops. An example is the genus Anochetus, also known as little trap-jaw ants. Within this group, two species thought to have resided on the island for a long period of time are related to taxa in Asia and found only at interior higher elevation habitats. Two other species, related to taxa in Africa, are widespread across the entire lowlands. Together, these data support the idea that Anochetus species arriving on Madagascar first settled in marginal coastal habitats before shifting to anterior lowland forest and finally up to montane forest. Turnover

The composition of ant species can vary considerably across an island’s history. The primary forces that impact island biogeography—size, isolation, and habitats—also exert great influence on species turnover through time. On Hispaniola, distance to source populations has had a dramatic effect on faunal assembly. Studies of Dominican amber indicate that 20 million years ago, during the Miocene, the island’s ant fauna was closely related to the continental fauna of Mexico. Of the 38 genera and subgenera found in amber, only 22 persist today on Hispaniola, whereas 15 native genera have colonized the island since. Interestingly, the amber fauna includes army ants no longer present on the islands. This dramatic species turnover reflects the fact that during the Miocene, the Greater Antilles (including Cuba, Jamaica, Hispaniola, and Puerto Rico) were located nearer to the mainland. Lying further from the mainland today, Hispaniola has lost some of its continental taxa. Highly specialized species or those less able to establish themselves on new ground were the most likely to disappear. Similar turnovers in ant species have probably swept across many islands as shifts in climate, volcanic eruptions, or other geological factors changed their habitats.

Further Research

Islands are natural laboratories for understanding the processes of faunal distribution and diversification. New methods combining species inventories, taxonomic research, and phylogenetic findings are enabling scientists to investigate these processes through the study of island ants. A comprehensive research effort is now in progress in the southwestern Indian Ocean (SWIO) islands. The coralline, volcanic, and Gondwanaland fragments of this region vary widely in age, size, degree of isolation, and habitat type, making them an ideal place to explore how each of these factors affects species diversity. Some of the questions researchers seek to answer include the geographic origins of the ant fauna and whether the estimated ages of endemic groups correlate with the ages of the islands themselves.

Wasmannia auropunctata, the big-headed ant Phiedole megacephala, the tropical fire ant Solenopsis geminata, and the Argentine ant Linepithema humile (Figs. 3, 4). These ants are especially dangerous when they join forces with another insect like mealybugs, a type of sap-sucking plant parasite that is often invasive on islands. In exchange for protection, the mealybugs provide drops of sugary honeydew to the ants. Fueled by these bonus sources of sugar, invasive ant populations can easily multiply out of control on an island system and trigger an invasional meltdown. This scenario has played out on Christmas Island, where such interactions and the concomitant vast numbers of ants (Anoplolepis gracilipes) have led to decimation of the native land crab–dominated ecosystem.

INVASIVE ANTS

Although colonization and species turnover are natural island processes, modern-day incursions of invasive ants are a major threat to natural ecosystems. Small size and limited biodiversity make islands inherently vulnerable to new species introductions. Today, habitat fragmentation caused by development, together with species introductions accelerated by global trade, has further increased this vulnerability. Combating invasive species is of particular importance on smaller islands. In Mauritius, where only a few patches of original forest still remain, invasive ants may have driven the entire lowland ant fauna to extinction. The bigheaded ant, Pheidole megacephala, has been implicated in the blanket decimation of Hawaii’s lowland arthropods. Entomologists in the early twentieth century described in detail how the native beetle fauna was defenseless again the onslaught of the invading big-headed ant. On the smaller, granitic islands of the Seychelles, Christmas Island, and Zanzibar, invasive ants such as P. megacephala have already extirpated native ants and are now threatening nesting bird populations. Larger islands such as Madagascar, where habitats are severely fragmented, can be just as vulnerable to invasion as their smaller counterparts. And although parks and reserves bolster the chances of survival of native species by protecting habitat, they cannot prevent aggressive exotic ants from driving native species locally extinct. An island’s ant fauna may include dozens of invasive species. However, a few bad actors can cause enough damage to destroy an island system. The usual culprits include the yellow crazy ant Anoplolepis gracilipes, the white-footed ant Technomyrmex albipes, the little fire ant

FIGURE 3 The most notorious invasive ants on islands. (A) yellow

crazy ant, Anoplolepis gracilipes; (B) little fire ant, Wasmannia auropunctata; (C) tropical fire ant, Solenopsis geminata.

ANTS

39

Invasional meltdowns of island ecosystems may be caused in part by the formation of ant supercolonies. The entire population of a newly arrived species may derive from the landing of a single queen or colony. Because all ants of this species are so closely related, they may lose aggression toward others of their own kind, a feature that normally limits colony densities. The resulting supercolonies can attain extremely high densities, can decimate local arthropod communities in the region, and can lead to an oversimplified invertebrate community that fails to provide essential ecosystem services such as nutrient cycling, plant seed dispersal, and a prey base for higher trophic levels. On Christmas Island, for example, researchers have documented the devastating impact of a supercolony on a local ecosystem. After an accidental introduction, the invasive crazy ant (Anoplolepis gracilipes) formed massive supercolonies that tended scale insects. Through direct predation, the supercolonies practically eliminated the red ground crab in the infested area. Without the crab, the principal litter consumer and seed disperser, the habitat changed dramatically. The scale insects killed off many trees and impacted ground-nesting birds. History

FIGURE 4 The invasive (A) white-footed ant, Technomyrmex albipes;

(B) big-headed ant, Phiedole megacephala; and (C) Argentine ant, Linepitherna humile.

Once non-native, invasive ants become established in natural settings, they are difficult, if not impossible, to eradicate. Thus, when preserving an island’s native ant species, an ounce of prevention is truly worth a pound of cure. Impact of Invasives

Worldwide, invading ants have caused impacts that reverberate throughout local ecosystems. In some cases, invasive ants have reduced the abundance and diversity of native ants by more than 90%. Nor are the consequences of ant invasions limited to other ants. The intruders also cause decreases in the diversity of insect herbivores, mammals, lizards, birds, and even plants.

40

ANTS

Ants have probably hitched rides with humans since the dawn of history. A recent dig at a Roman bath in Britain uncovered the bodies of 2000-year-old invasive ants. Over the past 500 years, there have been many accounts of ant plagues on different islands. For example, between the sixteenth and eighteenth centuries, several tropical West Indian islands were stricken with a series of ant plagues. Historical documents prove that environmental problems caused by invasive ants ensued just decades after Europeans came to the New World. At least two different ant species were the culprits of the Caribbean ant plagues. Solenopsis geminata, the tropical fire ant, was brought to the islands in the early 1500s; Pheidole megacephala, an African ant, came in the late 1700s. The plagues caused widespread crop destruction and may have been accelerated by the arrival of sapsucking insects. The same scenario has been replayed again and again on other islands. One interesting feature of ant plagues is that they are relatively short lived. The invasive ants are soon either repressed or driven extinct by later invading ants. For example, Pheidole megacephala was the dominant ant species on the Atlantic islands of Bermuda for much of the twentieth century. In 1940, however, the Argentine ant (Linepithema humile) arrived in the area and quickly

outcompeted the earlier champion. Pheidole megacephala, however, persists, and ever-shifting battlefronts now crisscross most of the islands. An important note is that islands have a long history of cycling through taxa. For example, historical records for the Indian Ocean island of Réunion indicate that the invasive ant Anoplolepis gracilipes was already abundant on the island in 1895. With the capacity to attain extremely high densities, this species can decimate resident vertebrates and invertebrate populations. In the ensuing 100 years, however, A. gracilipes has become rare. Research suggests that competition with dominant species such as Pheidole megacephala and Solenopsis geminata may have reduced its foraging efficiency and, therefore, its abundance. Although A. gracilipes is less competitive than other invasives, it thrives on a wide range of islands and must possess superior colonizing ability. The fortunes of this species have followed a similar trajectory in the Seychelles. The modern twist to this phenomenon is the speed with which species turnover now occurs. As planes and ships have multiplied, and transport times have shrunk, pressures on native species have increased apace.

Ricklefs, R. E., and E. Bermingham. 2002. The concept of the taxon cycle in biogeography. Global Ecology and Biogeography 11: 353–361. Suarez, A. V., D. A. Holway, and P. S. Ward. 2005. The role of opportunity in the unintentional introduction of invasive ants. Proceedings of the National Academy of Sciences USA 102: 17,032–17,035. Underwood, E. C., and B. L. Fisher. 2006. The role of ants in conservation monitoring: if, when, and how. Biological Conservation 132: 166–182. Wilson, E. O. 1988. The biogeography of the West Indian Ants (Hymenoptera: Formicidae), in Zoogeography of Caribbean insects. J. K. Liebherr, ed. Ithaca, NY: Cornell University Press, 214–230.

ARCHAEOLOGY MARSHALL I. WEISLER University of Queensland, St. Lucia, Australia

Archaeology is the systematic study of material remains. The diversity of island environments and biota presents fascinating opportunities for studying the evolution of Oceanic societies that developed from ∼40,000 to ∼800 years ago.

Further Research

The first step in understanding the level of threat posed by invasive ants is to inventory and map the extent of ant invasions. However, an assessment of invasives has yet to be initiated for many island systems. Once the invasives have been catalogued and mapped, a number of critical questions can be addressed. These include evaluating the risk of spread to other regions and habitats, predicting the effects on native fauna and flora, and assessing the impact of climate change on species interactions. The answers can help provide guidelines for conservation policies and control or eradication initiatives such that native island ants can persist. SEE ALSO THE FOLLOWING ARTICLES

Dispersal / Insect Radiations / Land Crabs / Madagascar / Species– Area Relationship / Taxon Cycle FURTHER READING

Fisher, B. L. 2005. A model for a global inventory of ants: a case study in Madagascar. Proceedings of the California Academy of Sciences 56: 86–97. Holway, D. A., L. Lach, A. V. Suarez, N. D. Tsutsui, and T. J. Case. 2002. The causes and consequences of ant invasions. Annual Review of Ecological Systems 33: 181–233. Molet, M., C. Peeters, and B. L. Fisher. 2007. Permanent loss of wings in queens of the ant Odontomachus coquereli from Madagascar. Insectes Sociaux 54: 174–182. O’Dowd, D. J., P. T. Green, and P. S. Lake. 2003. Invasional ‘meltdown’ on an oceanic island. Ecology Letters 6: 812–817.

THE FASCINATION OF ISLANDS

The vast expanse of the Pacific Islands hosted the greatest maritime migration in human history. Venturing across uncharted waters, Pacific colonists over millennia developed remarkable voyaging skills; built massive doublehulled sail-rigged canoes up to 10 m long; and carried vital plants, animals, and indeed the mental templates of their community layouts or “transported landscapes,” to every speck of land in this watery world known as Oceania. This journey began ∼40,000 years ago in the western Pacific and culminated a millennium ago in the settlement of the isolated outposts of Hawai‘i, New Zealand, and Easter Island—the final islands colonized before encountering the continental barrier of South America. Thousands of islands, scattered over more than onethird of the Earth’s surface, display a wide range of size and elevation fostering a diversity of geology, landforms, soils, vegetation, climate, marine biota, and degrees of isolation, providing an endless array of “natural experiments” for human colonists. Archaeologists are challenged to understand, for example, how and why founding human groups settling in the Hawai‘ian Islands about AD 900 developed into one of the most highly stratified Oceanic societies, yet with more than 30,000 years of occupation, communities in the southwestern Pacific never attained this level of social complexity. How did humans adapt to

ARCHAEOLOGY

41

the remarkable variability of environments found on continental, high volcanic, and raised limestone (makatea) islands and on low coral atolls? Why were some societies sustainable, whereas others fell into an endless spiral of environmental degradation, warfare, and ultimate collapse? Oceanic islands, in all their endless variety, provide remarkable opportunities for investigating the historical development of human societies. NEAR VS. REMOTE OCEANIA

Having traversed the great breadth of Oceania nearly two centuries ago, the great French explorer and naval commander Dumont d’Urville was the first to classify Pacific peoples into three seemingly distinct groups. Melanesians (“dark islanders”) occupied islands in the southwest Pacific including New Guinea, the Bismarck Archipelago, the Solomons, New Caledonia, Fiji, and a few others. These are generally dark-skinned people who have inhabited Oceania the longest and speak the greatest diversity of languages. Consisting mainly of low coral atolls scattered along the northern tropical latitudes, Micronesians (“small islanders”) are not a homogeneous group linguistically, in terms of the range of human phenotypes, nor do they share a common point of origin. Decades before d’Urville, the famous British commander Capt. James Cook, after visiting Tahiti, New Zealand, and Hawai‘i,

remarked on the similarities of language, physical appearance, and customs of the “Indians of the South Seas”— the region d’Urville labeled Polynesia or “many islands.” Although not without its problems, this tripartite division of Melanesia, Micronesia, and Polynesia has remained in use to this day. Roger Green, in 1991, developed far more meaningful terms to differentiate among the islands east of Wallace’s Line and reflect the biogeographic differences of Oceania, importantly reflecting the constraints and opportunities for human colonists and the subsequent evolution of island societies. Green’s Near Oceania, consisting mostly of New Guinea, the Bismarck Archipelago, and the Solomon Islands, has the greatest biodiversity in Oceania and the only Pleistocene human occupation (Fig. 1). Some resident Papuan speakers eventually mixed with the Austronesians who would settle the region much later. The islands are generally larger and more closely spaced than those of Remote Oceania, a fact that had important implications for human colonization and subsequent exploratory voyaging. Geoffrey Irwin refers to this region, sheltered from cyclones, as a “voyaging nursery” where the knowledge of open-water voyaging was learned—an essential skill for traversing the vast distances separating the far-flung islands of Remote Oceania. Here, the biodiversity declines progressively in an easterly direction.

FIGURE 1 Map of the Pacific showing the division between Near and Remote Oceania and trade (interaction) spheres defined by the identification

of exotic artifacts, including the movement of adze rock from Hawaii to the Tuamotu Archipelago (adapted from Weisler 1998).

42

ARCHAEOLOGY

Settled by Austronesian speakers no earlier than about 1500 BC, the islands of Remote Oceania were among the last places settled on Earth. It is little wonder, given that the islands are generally smaller and spaced much farther apart than those in Near Oceania. Human colonists had to develop the voyaging skills required to routinely sail out of sight of land in the quest for new islands. Given the larger distances between landfalls, it was previously thought that once colonized, island societies developed in near isolation. Recent studies of ancient Polynesian trade have shown otherwise and are discussed below after a brief overview of the foundation culture of Remote Oceania, known as Lapita. LAPITA

Nearly four millennia ago, pottery-using agricultural peoples inhabiting Taiwan, the Philippines, and the immediate islands arcing from Southeast Asia initiated a second wave of Oceanic exploration. Venturing into New Guinea, the Bismarck Archipelago, and the Solomon Islands, these Austronesian speakers encountered small human populations in scattered settlements, unlike the larger agricultural communities from whence they came. The small, kin-based social groups of Near Oceania made some use of marine resources, developing a shell technology for making adzes, fishhooks, arm rings, and beads, but they had a more inland focus—a gathering-hunting adaptation for exploiting brackish to freshwater environments and the forest for nut- and fruit-bearing trees, tubers, rats, reptiles, and introduced marsupials such as the cuscus. Inter-island movement is indicated by the transport of obsidian. This way of life, the dominant pattern for ∼30,000 years, was supplanted by “late” arrivals from the west—the Lapita culture. Jack Golson was the first to realize that highly decorated pottery finds from archaeological sites spanning the Near and Remote Oceania divide represented a “community of culture.” This ceramic horizon developed in Near Oceania and was identified by the unique, fine dentate-stamped designs on low-fired, earthenware pottery (Fig. 2). The geographic spread of Lapita covers Near Oceania and the islands east to Tonga and Samoa and is the foundation culture for all of Polynesia. In the west the earliest sites are ∼3500 years old and consist of large villages on coastal beach terraces or clusters of stilt-houses built over the shallow lagoons or calm shorelines. The average settlement size was ∼5000 m2, consisting of 15 to 30 dwellings, whereas the largest communities may have had up to 100 houses. These were sophisticated agriculturalists who managed root and tree crops and introduced

FIGURE 2 A Lapita pottery sherd exhibiting characteristic dentate-

stamping and lime-filled design (photograph by M. I. Weisler; used with permission of P. Kirch).

pigs, dogs, and chickens from their homeland. There was nearby inter-island and long-distance trade spanning hundreds of kilometers for exchanging pottery, obsidian, chert, volcanic oven stones, and undoubtedly marriage partners as well as perishable foodstuffs, which rarely survive in archaeological contexts. In fact, one of the longest documented movements of any commodity during world prehistory is that of obsidian, traded 4500 km across the breadth of Near Oceania into Remote Oceania, undoubtedly as a series of linked moves. The Lapita diaspora is also remarkable for its speed, taking perhaps two to three centuries to traverse the 5000 km from New Guinea and the Bismarck Archipelago to Tonga and Samoa in West Polynesia, establishing viable communities en route. The reasons for such rapid migration are clearly not based on population pressure, because density-dependent models require more time between stages of migration. Part of the reason lies in the search for pristine islands with their abundant stocks of seabirds (which could number into the tens of thousands) and unexploited marine shores teaming with fish, molluscs, crabs, and lobsters—or perhaps even new sources of stone for making the ubiquitous adze and other cutting tools. Imagine laying claim to a new, high-grade stone source—a veritable gold mine by today’s standards. Economics aside, there were important social reasons for risking all to found new colonies. In traditional Oceanic societies the first-born is often bestowed with all the property and accumulated family wealth. This, alone, would be a real incentive for junior siblings to split off from the parent community to establish new settlements where they could assume seniority. Archaeologists

ARCHAEOLOGY

43

are now charged with filling in the details of the Lapita horizon and all its variability across this large swath of the Pacific. Using Roger Green’s terms of intrusion, innovation, and integration of the Lapita phenomenon, how did these processes play out over the entire spatial and temporal span that this horizon occupied? What was the sequence of island colonization? With rare exception, excavation samples for any one Lapita site are still small, thus limiting our understanding of intra-site variation or archaeology at the “household” level. What was the role of exchange, not only for establishing new colonies, but also throughout the duration of settlement? These are just a few of the questions that will occupy archaeologists in the coming decades. DATING HUMAN COLONIZATION OF PACIFIC ISLANDS

There is scarcely any Pacific island where archaeologists agree on the timing of human colonization. This is no small problem, as the arrival of humans on previously uninhabited islands “starts the clock” for examining the speed and tempo of adaptations, the rate of human-caused animal extinctions, and transformations in the economic and social fabric of society. What kinds of data are useful for clearly documenting the first presence of people on islands? All archaeologists agree that artefacts, charred and patterned fragments of food remains, subterranean ovens and hearths, and post molds are obvious evidence for human habitation. In addition to radiocarbon chronologies for anchoring island sequences, pollen, sediment, and charcoal depositional events can signal human colonization of pristine islands and subsequent human-induced landscape modification, which often coincides with bird extinctions or reductions, changes in land snail assemblages, and marked alterations in vegetation communities. Paleoenvironmental estimates for human arrival are often inferred from pollen sequences, developed from sediment cores taken most often in lakes. Colonists cleared land for agricultural production, which resulted in a marked decline in forest taxa, an increase of disturbance indicators (such as open ground ferns), and an influx of charcoal particles. However, these characteristics can also develop in response to climatic change such as drought, or short-term perturbations, including hurricanes, which can influence vegetation dynamics. For example, in Palau, western Micronesia, sediment cores show a decline in forest pollen and increases in ferns and microscopic charcoal particles ∼4500 years ago. This is more than 1000 years earlier than the documentation of cultural materials (artifacts, food remains, etc.). In Fiji, charcoal particles from pollen

44

ARCHAEOLOGY

cores were identified at ∼4300 years ago, yet there is no reliable indication of human settlement until 2900 years ago. Mangaia, in the southern Cook Islands, shows evidence of habitat disturbance at 2500 years ago, but no in situ cultural remains until ∼1000 years later. Pollen sequences are most useful when they include evidence of humanly transported plants such as aroids (e.g., taro, giant swamp taro) and ornamental and medicinal tree and shrub species. Whereas pollen cores are an important source of environmental data—whether they document human disturbance or not—multiple, well-dated sediment cores displaying similar disturbance-influenced environmental sequences and pollen indicative of human-introduced plants are robust data for inferring the presence of people on islands. More holistic approaches to determining human colonization of islands can be achieved by using a broader array of data. Are the earliest habitation sites located in favorable environmental settings? Using a settlement pattern approach, we would expect to find the earliest sites situated near formerly abundant terrestrial and marine resources, with easy access to the sea. In other words, is the “earliest” archaeological site located in a logical environmental position? Exploitation of abundant and pristine resources should be found at the earliest cultural layers. The lowest, and oldest, cultural layers in early archaeological sites usually have the densest concentrations of food bones with evidence of extinct species, often those of flightless birds. In fact, extensive research has shown that most Pacific islands had one or more species of flightless bird that became extinct after human arrival. This contextual information, of settlement patterns, subsistence, and extinction events, provides a better platform for clearly documenting early sites. THE MYTH OF ISOLATION: DOCUMENTING ANCIENT TRADE AND EXCHANGE

Once settled, small founding groups on previously uninhabited islands were thought to have diversified and evolved in relative isolation. However, early trade links between parent and daughter communities—documented through the identification of exotic artifacts—were seen as integral to the process of colonization where “lifelines” were maintained until founding colonists were well established. (The term “trade” is used loosely here to mean interaction at some level, but not necessarily the two-way movement of commodities.) The classic example is that of the Pitcairn Group in southeast Polynesia. The four islands of the Pitcairn Group consist of the Pleistocene volcano of Pitcairn (only 4.5 km2), the raised limestone (or makatea) island of Henderson (37 km2), and the two

diminutive atolls of Oeno and Ducie. These islands are characterized by their isolation, small size, lack of reliable freshwater, nutrient-poor soils, and limited reefs (especially in the case of Pitcairn and Henderson). It is little wonder that Peter Bellwood coined the term “mystery islands” for these and nearly two dozen other ecologically marginal islands found throughout Polynesia, for it is these “mystery” islands that have ample evidence of prehistoric habitation, but were found abandoned when rediscovered by Europeans. Why were these islands settled in the first place, and why were they abandoned? The Pitcairn Group provides an illuminating example. The parent population for the Pitcairn communities was the diverse and well-watered volcanic islands of Mangareva, situated ∼400 km west. Extensive archaeological survey and excavations throughout Mangareva and the Pitcairn Group by Weisler in the 1990s detailed a history of trade between Pitcairn and Henderson islands, and on to Mangareva. Traded commodities included fine-grained basalt adzes (Fig. 3) and volcanic glass (a silica-rich rock made into sharp flakes for cutting) from Pitcairn, which were exported to Henderson because it had no sources of volcanic rocks for vital tool making. Mangareva received adzes from Pitcairn and exchanged black-lipped pearlshell (primarily for making fishhooks), which is ubiquitous in its 20-km-wide lagoon. Mangareva also supplied Henderson with volcanic oven stones, planting stock for coconuts, giant swamp taro, and bananas,

FIGURE 3 A typical late prehistoric, fine-grained basalt adze blade

from Pitcairn Island. Adze material from Pitcairn Island was traded to Henderson Island (100 km north) and Mangareva (400 km west).

as well as marriage partners and pigs. Trade was essential for the long-term viability of the Henderson community because the island had no sources of these necessary resources. The commodities transferred within this interaction sphere are shown in Fig. 4. The detailed radiocarbon chronology for Henderson settlements documented a period of sustained interaction between Pitcairn Island and Mangareva for nearly six centuries beginning about AD 900. After AD 1450 imported artefacts are no longer found in the Henderson archaeological sites, signaling an

FIGURE 4 The southeast Polynesia interaction sphere between the ecologically marginal Pitcairn Group and the high volcanic islands of Mangareva.

This system operated for six centuries beginning about AD 900 (Green and Weisler 2002).

ARCHAEOLOGY

45

sources of information for documenting changes in the social organization of island societies. East Polynesia is known for its household architecture in which dry-laid stones were stacked to delimit various functional spaces and to provide support for pole-and-thatch superstructures. Fig. 5 illustrates a substantial L-shaped wall (house foundation), an attached stone terrace (shrine), and a C-shaped shelter (cookhouse) forming a late prehistoric residential complex on Moloka‘i, Hawai‘ian Islands. A comparative analysis of the size and content of similar residential complexes across the landscape can provide information about the status of the occupants. This was clearly shown in the late prehistoric settlement pattern of Kawela, located along the central south shore of Moloka‘i, where extensive excavations of ten residential complexes and a contrastive analysis of the artefacts, food remains, and architectural complexity revealed the status of the occupants. Religious architecture has a wide range of forms and sizes: small family shrines (Fig. 5); small to large stone platforms, some larger than 2000 m2; and the well-known statues (moai) on Easter Island, an extreme example of a familiar East Polynesian pattern of temple layout (Fig. 6). Monumental architecture serves a multitude of functions within society. Ritualized ideology is codified by a series of temples that were strategically placed to reinforce social

end to external trade. Perhaps one or two human generations later, after the lifeline to Mangareva was severed, the small communities on Henderson died out or relocated to their parent homeland. During the past two decades archaeologists, often working in concert with geochemists, have identified exotic volcanic adzes, pottery, and obsidian in distant archipelagoes. When exotic artefacts are found in dated layers of archaeological sites, the spatial and temporal dynamics of long-distance interaction can be defined. Fig. 1 shows the limits of Polynesian interaction spheres based on the “sourcing” of exotic adzes. Recent research has also identified a rock from Hawai‘i that was fashioned into an adze in the Tuamotu Archipelago, a distance of more than 4000 km, making this one of the longest uninterrupted maritime movements in world prehistory. This interdisciplinary research makes it quite clear that Oceanic communities were far from isolated and, in many cases, had a long history of interaction with distant islands. It is these interactions that archaeologists must define to better understand the external forces that shaped and transformed island societies. SOCIOPOLITICAL CHANGE

Throughout much of Remote Oceania, surface residential, religious, and fortification architecture are important

(30) (40)

(25)

(75)

(70) (45)

4

(110)

Compound Feature B

hearths

1

2 (110)

1

lithic scatter

2 60m to

Feature

3 (75)

(70)

C

Anahaki 50-60-02-16 Features A and B Habitation Structures

E

MN

r

o

d

e

d

Feature A 1 (50)

(55)

FIGURE 5 An early seventeenth-century late pre-

historic residential complex from Moloka‘i, Hawaiian Islands. The long wall anchored a pole and thatch house with an internal slab-lined hearth. The small stone terrace, immediately east, was a family

0

1

2

3

4

5m

shrine. Feature A was a cookhouse with a stone tool-working area just north.

46

ARCHAEOLOGY

FURTHER READING

FIGURE 6 Monumental architecture on Easter Island (Rapa Nui). This

temple site (marae) consists of statues (moai), a platform (ahu), and a paved court. Photograph courtesy of M. I. Weisler.

order. This can be seen in the distribution of Easter Island temples, whose placement relates to traditional land unit boundaries. The size, number, and position of temples coincide with sociopolitical complexity where human activities were regulated. Prominent examples of prehistoric defensive architecture are found along the ridges of Babeldoab on the Palau Islands in western Micronesia, on the hilltops of Rapa on the Austral Islands in East Polynesia, and throughout much of New Zealand. Many fortifications consisted of a series of ditches, embankments, and palisades that ensured protection from warring tribes. These sites were built in response to social conflicts that developed, in part, from increasing populations, low soil productivity for crop production, and intertribal competition between late prehistoric social groups. In essence, fortifications symbolize periods of great social upheaval. CONCLUSIONS

The archaeology of the Pacific Islands provides ample opportunities to investigate many of the important problems in world prehistory, not all of which could be discussed in this short essay. Oceania has some of the earliest examples of crop domestication, the longest transport of commodities, some of the most complex chiefdoms in the world, and certainly the most linguistically diverse regions on Earth. Modern archaeological research is only a few decades old in the Pacific, and many exciting and innovative studies await future generations. SEE ALSO THE FOLLOWING ARTICLES

Easter Island / Exploration and Discovery / Human Impacts, Pre-European / Peopling the Pacific / Polynesian Voyaging / Wallace’s Line

Collerson, K. D., and M. I. Weisler. 2007. Stone adze compositions and the extent of ancient Polynesian voyaging and trade. Science 317: 1907–1911. Green, R. C., and M. I. Weisler. 2002. The Mangarevan sequence and dating of the geographic expansion into Southeast Polynesia. Asian Perspectives 41: 213–241. Irwin, G. 1992. The prehistoric exploration and colonisation of the Pacific. Cambridge, UK: Cambridge University Press. Kirch, P. V. 2000. On the road of the winds: an archaeological history of the Pacific islands before European contact. Berkeley: University of California Press. Rainbird, P. 2004. The archaeology of Micronesia. New York: Cambridge University Press. Weisler, M. I., ed. 1997. Prehistoric long-distance interaction in Oceania: an interdisciplinary approach. New Zealand Archaeological Association Monograph 21. Auckland: New Zealand Archaeological Association. Weisler, M. I. 1998. Hard evidence for prehistoric interaction in Polynesia. Current Anthropology 39: 521–532.

ARCTIC ISLANDS, BIOLOGY INGER GREVE ALSOS University Centre of Svalbard, Longyearbyen, Norway

LYNN GILLESPIE Canadian Museum of Nature, Ottawa

YURI M. MARUSIK Institute for Biological Problems of the North, Magadan, Russia

Arctic islands constitute a major part of the arctic land masses. Low temperatures and short summers are strong environmental filters that exclude most organisms from living there. Thus, the diversity of most species groups is lower on arctic islands than on the arctic mainland and more southern latitudes. Arctic species exhibit many different adaptations to cope with these harsh environmental conditions. EFFECT OF PAST AND PRESENT CLIMATE ON PATTERNS OF BIODIVERSITY AND ENDEMISM

Repeated periods of glaciation during the Pleistocene have strongly influenced the biota of arctic islands. During the Last Glacial Maximum (LGM; about 20,000 years ago), major ice caps wiped out most species in the Canadian Arctic Archipelago (CAA), Greenland, Novaya Zemlya, Severnaya Zemlya, Franz

ARCTIC ISLANDS, BIOLOGY

47

St. Lawrence

Wrangel

Novosibirskiye

Canadian Arctic Archipelago

Zevernaya Zemlya

Franz Josef Land

Novaya Zemlya Greenland

Svalbard

Jan Mayen

Bioclimatic zones: A Arctic polar desert (1-3 oC) o B Northern arctic tundra (4-5 C) o C Middle arctic tundra (6-7 C) o D Southern arctic tundra (8-9 C) o E Arctic shrub tundra (10-12 C)

Current glaciers Last Glacial Maximum (LGM) Bering Land Bridge during LGM Ice-free uplands/nunataks during LGM

FIGURE 1 Bioclimatic zones (http://www.arcticatlas.org/maps/catalog/

index.shtml) and glaciations of Arctic islands.

St. Lawrence

Ayon Chetyrekhstolbovoy

Wrangel

Bol’shaya Lyakhovskiy Genriyetta

Kotel’nyy

Bennetta Banks Victoria

Bolshoi Begichev

Parry Prince of Wales

Andreya

King William Somerset

Ellef Ringnes Cornwallis Axel Heiberg Devon Ellesmere Southampton

Oktyabr’skov Revolyutsii Russkiy Pioner

Coats Baffin

Troynoy Dickson Uyedineniya Brekhovskiye Sverdrup Sibiryakova Belyy Franz Josef Novaya Zemlya N Vize

N Greenland

Novaya Zemlya S

Total Greenland Svalbard W Greenland E Greenland

S Greenland

Jan Mayen

hypothesis), or became locally extinct and later recolonized from areas outside the main ice caps (tabula rasa hypothesis). Although a few molecular studies have found support for the glacial survival hypothesis, no paleorecords support continuous in situ existence of life within the glaciated islands. In contrast, the islands and areas around the Bering Strait (Beringian islands such as Novosibirskiye Ostrova, Wrangel, St. Lawrence, and Diomede) remained unglaciated throughout the Pleistocene. The lowered sea levels during glacial periods resulted in a large shelf area connecting present-day islands with the Russian and Alaskan mainland (the Bering land bridge). These altering connections to the mainland, and the Beringian islands remaining unglaciated, have strongly influenced both speciation processes and distribution of species on these islands. The number of endemic species is larger on Beringian islands than on other arctic islands and the diversity of vascular plants and springtails on, for example, Wrangel Island is extremely high (Fig. 2). The current summer temperatures of the warmest month range from 10–12 °C in the arctic shrub tundra zone to 1–3 °C in the arctic polar desert zone. The polar desert and northern Arctic tundra zones are almost exclusively found on arctic islands. Within a geographical region, summer temperature is the environmental variable that best predicts the diversity of species. For example, the number of vascular plants decreases towards the north in the Canadian Arctic Archipelago and from Novaya Zemlya to Franz Josef Land in Russia. However, some exceptions exist. In bryophytes, species diversity depends more on substrate than on temperature, and thus the difference in species numbers between north and south Greenland is low. Although the total number of species decreases towards the north or with lower temperatures, the proportion of widespread species increases. Of 115 taxa of vascular plants found in the arctic polar desert zone, 91.3% occur in both North America and Eurasia, and only one species is endemic to a region. Similarly, in the small arthropods known as springtails, the proportion of widespread species is highest in previously glaciated, high arctic islands.

FIGURE 2 Vascular plant (red bars) and springtail (Collembola, blue

bars) diversity on arctic islands. Data compiled from various sources. The bar for total species of vascular plants in Greenland represents 515 species. The bars for springtails have been doubled to visualize them.

ISLAND GROUPS Svalbard and Jan Mayen

Josef Land, and Svalbard (Fig. 1). Some ice-free nunataks and uplands existed, however, and it is debated whether some plants and animals survived the periods of glaciation (glacial persistence or glacial survival 48

ARCTIC ISLANDS, BIOLOGY

The Svalbard archipelago is situated from 74° to 81° N and from 10° to 30° E. The land area is 61,000 km2, but about 60% of this is covered by glaciers. The influence of the warm North Atlantic Current gives a more oceanic

TABLE 1

Dwarf Birch (Betula nana)

A

Estimated Numbers of Different Species Groups

3

Canadian Arctic

Size (km2) Ice-free area Vascular plants Mosses Liverworts Fungi Lichens Terrestrial mammals Marine mammals Birds (regularly nesting) Freshwater fishes

Wrangel

Archipelago

Greenland

Svalbard

Island

1,420,000 1,260,000 349 346 — — >750 11

2,170,000 410,000 515 (32) 477 135 1600 1094 8

61,000 24,000 165 (4) 288 85 624 764 (12) 2 (1)

7,600 7,608 417 (23) 239 87 — 350 3 (2)

7 61

22 58

8 38

12 47 (1–3)

10

3

1

0

41 9

Mountain Avens (Dryas octopetala)

B

11

note: Data are for Canada (various sources), Greenland (Jensen and Christensen 2003), Svalbard (updated from Elvebakk and Prestrud 1996; Prestrud et al. 2004), and Wrangel Island (Stishov 2004). Estimated numbers of endemic species are given in parentheses. Note that differences in degree of exploration and as well as taxonomical view make the numbers inaccurate.

and warmer climate compared to other islands at this latitude. This is also reflected in the species diversity, which is comparatively high in Svalbard (Table 1). Svalbard was almost entirely glaciated during the last glacial maximum, and paleorecords show a sparse arctic vegetation subsequent to 10,000 years ago. Although this is one of the most remote arctic archipelagoes, molecular analyses of plant species show that it was repeatedly colonized during the Holocene from several source areas (Fig. 3). The main source areas were in northern Russia/ Siberia and northeastern Greenland, areas connected to Svalbard by winter sea ice. Thus, sea ice, probably in combination with wind, might be an important dispersal mechanism to arctic islands. Exceptionally warm winds may also carry insects directly from areas such as Russia to Svalbard, as was observed for the nonresident migratory diamondback moth Plutella xylostella. The few endemic species or subspecies in Svalbard (i.e., the Svalbard reindeer, the Svalbard aphid, and four plant species) have probably evolved recently from species that immigrated after the LGM. In contrast to most arctic islands, there are no rodents on the archipelago (except the locally introduced sibling vole). The main herbivores are geese and reindeer. The arctic fox feeds mainly on eggs and chicks of sea birds and

1

67 2

White Arctic Bell-heather (Cassiope tetragona)

C

4

1

6 45

D

Bog Bilberry (Vaccinium uliginosum)

12

76

FIGURE 3 Source regions for past colonization of (A) dwarf birch

(Betula nana), (B) mountain avens (Dryas octopetala), (C) white arctic bell-heather (Cassiope tetragona), and (D) bog bilberry (Vaccinium uliginosum) to Svalbard. Source regions are inferred from genetic data (amplified fragment length polymorphism). Colors represent main genetic groups, and symbols represent sub-groups. Numbers on the arrows are percentages assumed to have arrived from the source region. The geographic distribution of the species is shaded. Reproduced with permission from Science.

ARCTIC ISLANDS, BIOLOGY

49

geese, as well as carcasses of seals and reindeer. There are many seabirds breeding in the archipelago, and these contribute a significant input of nutrients for plant growth. The only resident bird is the Svalbard ptarmigan (Lagopus mutus hyperboreus), which is endemic to Svalbard and Franz Josef Land. Jan Mayen is a small (373 km2) volcanic island east of Greenland. It has an extremely oceanic climate with mild winters and relatively cold and wet summers. About twothirds of the 66 vascular plant species found there are circumpolar, whereas the other third is amphi-Atlantic. The only endemic species found are apomictic dandelions (Taraxacum). The arctic fox is the only terrestrial mammal on the island. Large seabird colonies are found during the summer, but only the fulmar (Fulmarus glacialis) stays during winter. Greenland

Greenland is the world’s largest island. Including the numerous smaller islands along the shore, its total size is 2.17 million km2. The majority of species are confined to the ice-free margins, which cover only approximately 410,000 km2. Greenland stretches from 59°45′ N to almost 84° N and spans a vegetation gradient from birch forest in the south to polar desert in the north (Fig. 4). Considering the large size of Greenland, species diversity is relatively low, and it decreases from south to north as, for example, in vascular plants. Also, there are only a limited number of species that are endemic to Greenland. Of the total 515 vascular plant species, 32 taxa are endemic. However, 15 of the endemics belong to the apomictic hawkweed genus (Hieracium), which rapidly evolves new

FIGURE 4 Low-stature vegetation with prostrate or cushion-formed

herbs such as moss champion (Silene acaulis) dominate the middle arctic tundra zone. Ammassalik district in southeastern Greenland. Photograph by Inger Greve Alsos.

50

ARCTIC ISLANDS, BIOLOGY

species. Endemic species of algae and three spider species have also been recorded, and a few bird subspecies breed only in Greenland, but they overwinter elsewhere. The relatively low diversity and endemism found in this large island are probably due to its glacial history. Ice-free areas existed in Greenland throughout the glacial period, but according to climate data derived from ice cores, it was so cold during the LGM that only the most cold-adapted species could have survived there. Thus, it is assumed that the majority of species colonized Greenland during the last 11,500 years. This view is supported by molecular studies of several plant species. A large proportion of Greenland’s plants and animals are also found in northwestern Europe, indicating that they arrived from there. Although this distance is long, the Faroe Islands and Iceland form steppingstones along the route. Further, the majority of Greenlandic birds migrate from Europe and could have transported seeds, spores, and even some invertebrates. The majority of spiders and some groups of insects are Nearctic, indicating a high proportion of immigration also from northeastern Canada. Canadian Arctic Archipelago

The Canadian Arctic Archipelago (CAA) covers an immense area, ∼1.42 million km2, and comprises numerous large and many more smaller islands. It extends about 3000 km from below the Arctic Circle to the northern tip of Ellesmere, and 3000 km east-west from Baffin to Banks Islands. Ice caps occur in the mountainous northern and eastern parts on Axel Heiberg, Ellesmere, Devon, and Baffin Islands (maximum elevation 2615 m), but overall, glaciers cover only about 11% of the archipelago. Thus, the ice-free area of CAA is three times as large as the icefree area of Greenland and more than 50 times as large as the ice-free area in Svalbard. Recolonization after the LGM occurred primarily from mainland areas to the south, a distance as short as 1–20 km at several locations. Unglaciated Beringia was also an important source area for many groups, contributing to east-west differences in species composition (e.g., higher diversity of legumes on Banks and Victoria Islands). Glacial refugia on Banks Island provided additional source areas, while postulated refugia on Ellesmere and other islands have yet to be confirmed. Considering its large land area, species diversity is low on the Canadian Arctic Archipelago. The strong southto-north decrease in diversity is correlated with summer temperature and distance from the mainland. However, topography and oceanic influences modify this gradient,

creating a more complex pattern. Rain shadow effects are responsible for warmer drier summers and the relatively high diversity of the “polar oases” of the Forsheim Peninsula and Lake Hazen area on Ellesmere Island. Cool summers with extensive cloud and fog are responsible for the low vascular plant and arthropod diversity on Ellef Ringnes and nearby islands. Located north of the “shrub line,” this barren region lacks woody plants, which are so characteristic of tundra vegetation (e.g., willows, mountain avens, Ericaceae). No endemic vascular plants, bryophytes, lichens, mammals, or arthropods are known from the CAA, but several species are confined to the Archipelago and Greenland, such as Peary caribou, the alkali grass Puccinellia bruggemannii, the moth Gynaephora groenlandica, and the wolf spider Alopecosa exasperans. Also, at least one undescribed species of spider has been found on Banks Island. Russian and Beringian Islands

The Russian arctic islands can be divided into five main groups: (1) Novaya Zemlya (“New Land”) with adjacent Vaigach, Kolguyev, and some smaller islands; (2) Franz Josef Land; (3) Severnaya Zemlya (“North Land”); (4) Novosibroskiye Ostrova (“New Siberian Islands”); and (5) Wrangel and Gerald Islands. Besides these main groups there many small islands at the Ob’, Yenisei, Kheta, Lena, and Kolyma deltas and near Taimyr Peninsula. In addition, there are several arctic islands belonging to Russia and the United States in the Bering Strait and Bering Sea (e.g., St. Lawrence, Yttygran, Arakamchechen, Diomede and King Island). The most biologically diverse and well studied island in the Russian Arctic is Wrangel Island. It is a remote, relatively small island of about 7,600 km2, with the highest elevation above 1000 m. A unique feature of this island is the very limited extent of Pleistocene glaciations combined with the lowered sea level during LGM, making Wrangel a part of the Bering land bridge. This has enabled enrichment of the fauna and flora by very different elements originating in boreal, forest-tundra, tundra, arctic polar desert, and even steppe zones from Asia as well as North America, which has resulted in a species composition on Wrangel Island different from those on all other islands. For vascular plants, 417 species are known, more than for the whole CAA (349 species) and the northwest sector of the Siberian arctic (387 species), and approaching that of Taimyr Peninsula (494) and Greenland (515). Similarly, the diversity of spiders, beetles, and birds is high on Wrangel Island compared to Svalbard and Greenland.

The diversity of many insect families and orders is higher than on any other arctic island, including Greenland. The recurrent connections and disconnections of Wrangel Island also led to speciation in mammals, vascular plants, and some groups of arthropods. The number of endemic species on the island is extraordinarily high for the Arctic in general and for arctic islands particularly. There are 23 endemic vascular plant species, four spider species, 20% of weevils, both of the rodents Dicrostonyx groenlandicus vinogradovi and Lemmus sibiricus portenkoi (Fig. 5), and at least one bird subspecies Cepphus grylli tajani. If the recently (3500 years old) extinct dwarf mammoth is counted, the level of endemism of mammals would be higher. In the late Pleistocene several other ungulates such as Przewalski’s horse, woolly rhinoceros, primeval bison (Bison priscus), musk ox, woolly mammoth, and reindeer occurred on the island. The Novosibroskiye Ostrova (New Siberian Islands) consists of two larger groups of islands, Lyakhovsky and Anzhu, and one small group called De-Longa. With its area of about 36,000 km2, this region is about five times larger than Wrangel Island, and like Wrangel, it was also unglaciated and connected by the Bering land bridge. However, the archipelago is rather flat, which limits habitat diversity, and the diversity of plants, springtails, birds, and beetles is less than on Wrangel. There are some species on Novosibroskiye Ostrova that do not occur on Wrangel, including a willow grouse species and two goose species. The mammal fauna consist of wolf, wild reindeer, one lemming species, and arctic fox. The fossil mammoth fauna is even richer than on Wrangel Island with additional species such as saiga antelope, cave lion, and voles. Also, the fossil beetle fauna is much richer

FIGURE 5 Portenkoi’s lemming, Lemmus sibiricus portenkoi, endemic

to Wrangel Island, has an important role in the ecosystem. Photograph by Igor Dorogoi.

ARCTIC ISLANDS, BIOLOGY

51

(about 100 species during the past 200,000 years, or 58 species during last 115,000 years) than the present beetle fauna (about 10 species). In the nineteenth and beginning of the twentieth centuries, digging up and selling ivory from fossil mammoths was a profitable business. The Severnaya Zemlya Ostrova consist of four large (October Revolution, Bolshevik, Komsomolets, Pioneer) and about 70 smaller islands, covering a total area of about 37,000 km2. Although the island group remained partly unglaciated throughout the Pleistocene, it was not connected by the Bering land bridge, and it is also situated rather far north. Thus, species diversity is lower than on most other Russian archipelagoes with, for example, about 78 species of vascular plants. Only 17 of the 32 bird species that have been observed on the islands breed there. Six terrestrial mammals are known there: lemming, arctic fox, wolf, ermine, arctic hare, and reindeer. Four species of beetles have been found on the archipelago, but only on the southernmost island. The Novaya Zemlya Archipelago consists of two large and several smaller islands, in total about 81,000 km2. The number of vascular plants and springtails is similar to that on many other islands that have been previously glaciated. The flora represents a transition between the arctic Europe and Asia but with a separate mountain range element connected to the Urals. Some endemic vascular plants in Novaya Zemlya have been proposed, but they are dubious. There are two lemming species, a local reindeer subspecies, and arctic fox on the archipelago. Franz Josef Land (16,100 km2) is the northernmost archipelago and consists of almost 200 islands. Glaciers cover 85% of the archipelago. It is the most species-poor arctic archipelago, with only about 50 vascular plant species, about 150 bryophytes, over 300 lichens, one terrestrial mammal (arctic fox), at least 14 springtails, two spiders, and no beetles. Only 14 bird species breed on the island, but almost 30 other bird species have been observed visiting. Aleutian Islands

The Aleutian and Commander Islands are a 2,100-km long archipelago to the south of the Arctic, separating the Bering Sea from the North Pacific Ocean. Although the U.S. government includes this archipelago in its definition of the Arctic because of the treeless landscape that prevails here, this is caused largely by strong winds rather than low temperatures and short growing season, as is the case in the Arctic. These islands serve as a natural bridge between Old World and New World flora and fauna, although

52

ARCTIC ISLANDS, BIOLOGY

FIGURE 6 The sedge Carex bigelowii ssp. arctisibirica in Svalbard.

Most sedges have clonal growth and can become very old; up to 3000-year-old clones have been found in the Arctic. Photograph by Inger Greve Alsos.

physical evidence suggests that this archipelago was under ice during the LGM, so terrestrial species on these islands should be recent colonists (i.e., since the last glaciation, or less than 10,000 years ago). However, the relatively high levels of endemism (for high-latitude organisms) that characterize the Aleutian and Commander Islands suggests that many of these taxa were isolated for longer periods of time, probably in “cryptic” glacial refugia: ice-free areas that harbored multiple taxa through the Quaternary glacial cycles, though so far only evident through the biological record. Moreover, natural selection has resulted in local adaptations to the harsh conditions of the islands, with evidence of traits such as increased body size in some bird species. In the same way, the Alexander archipelago, a chain of over 1000 islands off the southeastern coast of Alaska that is also recently glaciated, though currently covered by evergreen forest and even temperate rain forest, is characterized by a number of monophyletic lineages, which may be attributed to multiple Holocene invasions or the persistence of taxa in refugia during Pleistocene glacial advances. CHARACTERISTICS OF SPECIES Plants

The arctic flora ranges from shrub tundra in the south to almost barren polar desert, where no woody plants live and only scattered herbs, bryophytes, and lichens are found. The majority of species are long-lived perennials with

relatively low resource allocation to sexual reproduction and high reliance on asexual reproduction for population maintenance, but a high variety of life strategies exists. There are fewer pollinators and lower pollinator activity in the Arctic than in other regions. Plant species adapted for wind pollination are dominant, and self-pollination is common. The growth forms are often prostrate mats, tussocks, rosettes, or cushions, which reduce desiccation and mechanical damage from the strong wind and maximize heat absorption (Fig. 6). In addition to the low temperatures and short growing season, drought places a very significant stress on plant life. The majority of bryophytes and lichen species are well adapted to periods of drought and increase both in abundance and ecosystem importance northwards. Invertebrates

Invertebrate groups occurring on arctic islands have evolved from species in the boreal biome, where winter temperatures are often lower than in the tundra zone, and they have similar adaptations to cold resistance as boreal species. The main limiting factor for invertebrates is heat deficit in the short (2–3 month), cool growing season, and therefore the main adaptations are directed towards shortening of the life cycle (vivipary, reduction of size) or extension of the life cycle to several years. They survive the winter by producing cryoprotectors, being able to dehydrate, or overwintering as cold-resistant eggs. In addition, behavioral strategies may assist in avoiding low-temperature extremes, for example seeking protected places to avoid winter cold, such as under thick snow cover or close to non freezing water currents. In arthropods, adaptations to the arctic climate lead to dominance by small-sized groups such as mites, spiders, and springtails, which have relatively high species diversity on arctic islands. Large sized insects, such as large ground beetles and bumblebees, are lacking on most arctic islands, whereas beetles and some other megadiverse groups, such as moths and true bugs, are represented on arctic islands by fewer species than small-sized groups. There is a decrease in species numbers of herbivorous insects (especially among beetles, butterflies, moths, and true bugs) in comparison to predaceous ones.

FIGURE 7 In the Arctic, many animals are white in winter and brown

during summer, which gives them a good camouflage. In the High Arctic some animals, such as this arctic hare on Ellesmere Island, remain white year-round, an adaptation to the very short snow-free summer period. Photograph by Lynn Gillespie.

high as 90 °C. Larger arctic animals have developed thicker and denser fur or plumage or a thick fat/blubber layer to keep warm without spending much energy. Smaller animals such as lemmings and voles live mainly under the snow, which acts as a thick insulating layer. Ptarmigans stay in “dock” (under snow) during bad weather conditions to reduce heat loss. The blood circulation system is also adapted to minimize heat loss by countercurrent heat exchange and by slowing down the circulation to extremities. Many arctic mammals have enlarged nasal cavities, and circulatory adjustments in the nose reduce water and heat loss. Some arctic animals, such as the Svalbard reindeer and ptarmigan, store large amounts of fat during the summer and autumn season, which is used to survive the winter (Fig. 8). Most arctic birds species migrate south before the winter period. When they arrive

Mammals and Birds

To survive the harsh winter of the Arctic, mammals and birds have developed morphological, physiological, and behavioral adaptations (Fig. 7). The difference between ambient temperature and body temperature may be as

FIGURE 8 Many arctic mammals, such as the Svalbard reindeer, put on

large fat reserves during the autumn. Photograph by Inger Greve Alsos.

ARCTIC ISLANDS, BIOLOGY

53

on the breeding ground in spring, their breeding success is closely linked to the timing of breeding relative to snow melt and peak food production. HUMANS ON ARCTIC ISLANDS

Humans have long been a part of the arctic environment, intimately connected to the local resources on land and sea. The indigenous peoples harvest natural resources both from the terrestrial (arctic fox, ptarmigan, reindeer, caribou, musk ox) and the marine environment (fish, whale, seal, polar bear; Fig. 9). In Greenland, fishing is the all-dominating trade and accounts for 95% of total exports, but in the hunting districts of the outlying areas the seal and whale catch is of great importance and forms a stable existence for one-fifth of the Greenlandic population. Reindeer herding is of local importance only on few arctic islands, such as in northernmost Norway. The 15 communities in the CAA are mainly inhabited by Inuit. Most Russian arctic islands are not inhabited except by the staff of small military camps, nature reserve stations and weather stations, but indigenous people live on some islands; for example, some Nenet families live on Novaya Zemlya. Fifty-seven thousand people, predominantly Inuit, live in towns and small settlements on Greenland. In Svalbard there is one Norwegian settlement with around 2000 inhabitants and one Russian settlement with about 500 inhabitants. Industrial activity is found only on a few arctic islands, such as mining activities in Svalbard, on Kolguyev, and, until recently, on Little Cornwallis and Baffin Island. There are huge reserves of oil and gas on arctic islands and the surrounding sea floor, such as the Sverdrup Basin in the Canadian high arctic, and exploration drilling is done in several locations.

Tourism and research activities are increasing on some arctic islands such as Svalbard and Baffin. CONSERVATION OF ARCTIC FLORA AND FAUNA

The arctic flora and vegetation are vulnerable to physical disturbance, and vehicle tracks often last for decades. Humans have overexploited many species, such as whales, polar bear, and arctic fox. Although some species and populations have recovered, others are still threatened. Long-range pollution from the industrial part of the world, such as heavy metals, persistent organic pollutants (POPs), and radionuclides, has reached arctic islands, and such pollutants are accumulating in some organisms. Climate change is predicted to be of higher magnitude in the Arctic than in other places in the word. Because the arctic islands represent the “end of land,” high arctic species have no further place to migrate if they are outcompeted by more southern species, and they may thus become extinct. Global warming will also open up the northern sea routes both in Canada and Russia and make the arctic oil and gas reserves more accessible, which would potentially lead to increased pollution and disturbance. Knowledge necessary for conservation is lacking for many islands, species, and ecological processes in the Arctic. For example, the identification and classification of arctic invertebrates, fungi, bryophytes, and microorganisms is limited. Although some monitoring programs exist, information on the status and trends of arctic populations is fragmentary. For proper management in a changing climate, more knowledge is needed about the species found on arctic islands, the ways they interact, and how they respond to the changing physical environment, especially climate. SEE ALSO THE FOLLOWING ARTICLES

Arctic Islands, Geology / Global Warming / Mammal Radiations / Refugia / Whales and Whaling FURTHER READING

FIGURE 9 Inuit hunters skinning a seal in Grise Fiord, Ellesmere Island,

Canada. Photograph by Olivier Gilg.

54

ARCTIC ISLANDS, BIOLOGY

Aiken, S. G., M. J. Dallwitz, L. L. Consaul, C. L McJannet, L. J. Gillespie, R. L. Boles, G. W. Argus, J. M. Gillett, P. J. Scott, R. Elven, and M. C. LeBlanc. 2007. Flora of the Canadian Arctic Archipelago. Ottawa: NRC Press (CD-ROM). Born, E. W., and J. Böcher. 2001. The ecology of Greenland. Nuuk, Greenland: Ministry of Environment and Natural Resources. Conservation of Arctic Flora and Fauna. 2001. Arctic flora and fauna. Status and conservation. Helsinki: Edita. Chapin, F. S. III, and C. Körner. 1995. Arctic and alpine biodiversity: pattern, causes and ecosystem consequences. Berlin: Springer-Verlag. Danks, H. V. 1981. Arctic arthopods: a review of the systematics and ecology with particular reference to the North American fauna. Ottawa: Entomological Society of Canada.

ES

ALASKA

RI

DG

Banks Is.

Severnaya Zemlya

AL PH A& M M O EN RI N DE D OS G O LE E V V

Victoria Is.

CHUKCHI PLATEAU AMERICA

Siberia

Franz Josef Land

LO

Elvebakk, A., and P. Prestrud. 1996. A catalogue of Svalbard plant, fungi, algae and cyanobacteria. Norsk Polarinstitutt Skrifter 198. Jensen, D. B., and K. D. Christensen. 2003. The biodiversity of Greenland—a country study. Technical report. Pinngortitaleriffik: Grønlands Naturinstitut (Greenland Institute of Natural Resources). Kristinsson, H., E. S. Hansen, and M. Zhurbenko. 2008. Panarctic lichen checklist. Conservation of Arctic Flora and Fauna. http://arcticportal .org/en/caff/. Prestrud, P., H. Strøm, and H. V. Goldman. 2004. A catalogue of the terrestrial and marine animals of Svalbard. Norsk Polarinstitutt Skrifter 201. Stishov, M. S. 2004. [Wrangel Island—master pattern of nature and nature anomaly.] Yoshkar-Ola: Mariyski Printing Factory Press (In Russian).

Baffin Is.

Ellesmere Is.

Novaya Zemlya Svalbard

EU R A S I A

GREENLAND Scandinavia

ARCTIC ISLANDS, GEOLOGY

FIGURE 1 Plate tectonic reconstruction of the Arctic as it is believed

to have appeared approximately 70 million years ago prior to opening of the Arctic, North Atlantic, and North Pacific oceans. (Reprinted

MICHAEL J. HAMBREY Aberystwyth University, United Kingdom

The geological history of the Arctic spans nearly four billion years and includes some of the oldest rocks on Earth. A vast range of sedimentary, igneous, and metamorphic rocks are present, but few were formed in their current position. The geological record for many Arctic islands reflects the drift of fragments of continental crust from a position south of the equator to their current polar position. As a consequence, the rocks record a range of climates from tropical to glacial, as well as a fascinating glimpse of biological evolution from the algae of the Precambrian to the high-order animals and plants of today. TECTONIC EVOLUTION

By outlining the tectonic components and history of the whole Arctic region, a context is provided for the main phases of geological evolution of the region. The geological attributes of the Arctic islands reflect the disparate nature of individual continental fragments and their movement by plate tectonic processes through time. These processes involved continental breakup, continental collision, and sea floor spreading. Indeed, many parts of the Arctic have rocks that once were formed south of the equator; plate movements have resulted in their slow progression to a northern polar position today. All the continental fragments are believed to have been united as one supercontinent about 70 million years ago (Fig. 1). Since then, the Arctic Ocean basin has opened, along with the North Pacific Ocean, Baffin Bay, and the Norwegian-Greenland Sea. The oldest rocks are Archean to Proterozoic meta-

with permission of Cambridge University Press from Dowdeswell and Hambrey 2002: 32, adapted from Worsley and Aga 1986.)

morphic rocks that represent stable crystalline shields. In many areas they are overlain by sedimentary rocks. Periodically, the sedimentary strata were intruded by igneous rocks, subjected to metamorphism deep in the crust, and deformed during mountain-building events or “orogenies,” when continents collided. The sedimentary strata reflect the climatic regimes and topographic/bathymetric settings under which they formed, including under ice age, temperate, and tropical climates and deep-sea, continental shelf, fluvial, glacial, estuarine, and mountain environments. This pattern of northward drift has been determined with a reasonable degree of certainty over the past 600 million years, especially for Svalbard and East Greenland. Ongoing tectonic processes are focussed on the continuing opening of the North Atlantic. New oceanic crust continues to form along the Mid-Atlantic Ridge, upon which the volcanic islands of Iceland and Jan Mayen are located, surrounded by deep ocean. New basaltic rocks formed on the ridge continue to push Europe away from Greenland and North America, a process that began about 60 million years ago. In contrast, continental margins are “passive” and thus relatively stable today, lacking significant earthquake activity. The other main plate tectonic process, continental collision, is not a feature of the Arctic at the presentday. However, the older geological record shows dramatic evidence of this process on several Arctic islands. The “Caledonian Orogeny” of the early Paleozoic Era

ARCTIC ISLANDS, GEOLOGY

55

resulted in a chain of mountains of Himalayan proportions, with evidence of volcanism, metamorphism, folding, strike-slip and thrust faulting, and igneous intrusion on a vast scale. It will be evident from the above that the islands do not coincide neatly with geological boundaries. Some, such as Svalbard, consist of a number of slivers of crust called terranes, each of which has a distinct geological history. In contrast, some now widely separated regions share a common history. Strike-slip faulting on a scale of hundreds or thousands of kilometers was the process whereby terranes joined or separated. Although the geology of the Arctic islands is complex, a number of key stages of evolution (summarized in Fig. 2) can be identified, as outlined below. EARLY CRUSTAL EVOLUTION

The geological foundations of many continental areas are represented by ancient crystalline shields, the Arctic region being no exception. The islands contain fragments of two main shields that were joined together prior to 70 million years ago (Fig.1). The Laurentian shield includes much of North America, including Greenland, and Baltica is the foundation of much of Europe, including parts of Svalbard. Even where these

Million Years

Period

Some key events in Arctic

0 Quaternary

Glacial/interglacial cycles

Neogene

Arctic acquires current configuration

Palaeogene

Cool temperate climates Extensive forests coal North Atlantic & Arctic Oceans begin opening

1. 8 23. 8 65 Cretaceous 142 Jurassi Jurassic

Dinosaurs

206 Triassic

248 Permian 290

Younger sedimentary basins

rocks are not exposed, they underlie younger sedimentary cover rocks. The rocks are strongly deformed and metamorphosed, the dominant rock types being gneiss, schist, and igneous rocks, some of which are among the oldest rocks on the planet; rocks in southwest Greenland date back to 3800 million years. Typical shield landscapes are characterized by low rocky hills with intervening boggy areas and lakes that are commonly aligned parallel to the dominant structures such as the metamorphic layering (foliation) or faults. EARLY SEDIMENTARY BASINS

The crystalline shield areas are overlain or bordered by zones of sedimentary rocks that range in age from approximately 1000 to 400 million years (late Proterozoic to early Paleozoic). The rocks are variably metamorphosed, some of which were partially melted deep in the crust (Fig. 3). However, those least affected give unique insights to early climates and life. These sedimentary rocks were laid down in rift basins where the crust was being stretched and flooded by the sea, and supplied by detritus from the rift margins. The shallow marine sediments are dominated by limestone, dolomite, mudstone, and sandstone that collectively attain a thickness of 10 to 20 kilometers. Particularly striking are stromatolites, layers of algae that trapped carbonate sediment, forming mounds or columns several meters high. Impressive examples occur in eastern Greenland and northeastern Svalbard. The best known modern examples are found in Shark Bay, Western Australia, but they never attain the spectacular scale of those of Proterozoic age in the Arctic. In contrast to these probably warm-water features, there is abundant evidence of glaciation, in the form of tillites (Fig. 4). These rocks are best known from eastern and northern Greenland and from Svalbard, and represent the Arctic manifestation of a global ice age, when according to some geologists the Earth

Tropical climates

Carboniferous 354

Old Red Sandstones deserts and rivers

Devonian 417 Silurian 443 Ordovician 495 Cambrian

Continental collision between Eurasia and mountain building: the Caledonian Orogeny

2500 4600

56

PreCambrian

545 Proterozoic Eon Archaean Eon

Major ice age Early life forms (algae) preserved Oldest rocks, 3800m.y., in Greenland

FIGURE 2 Summary of the geological time scale and key events

FIGURE 3 Ice-smoothed wall of inner Nordvestfjord, East Greenland,

affecting the Arctic islands. (Reprinted with permission of Cambridge

displaying deformed and highly metamorphosed rocks of Proterozoic

University Press from Dowdeswell and Hambrey 2002: 34.)

age called migmatites. Photograph by M. J. Hambrey.

ARCTIC ISLANDS, GEOLOGY

FIGURE 4 Neoproterozoic tillite, indicative of a global ice age, Ella Ø,

East Greenland. Photograph by M. J. Hambrey.

was locked in a deep freeze of global proportions. This controversial hypothesis is known as “Snowball Earth,” but even if this extreme view is incorrect, the evidence for continental-scale glaciation is unequivocal. It is possible to match strata in these early sedimentary basins across different areas, sometimes even on a bed-by-bed basis, as between eastern Greenland and northeastern Svalbard. Even though the rocks are today separated by not only an ocean but other tectonic terranes, the evidence points to their being formed in the same sedimentary basin. Subsequent plate movements, notably strike-slip faulting, have since separated these two areas. THE CALEDONIAN OROGENY

An ancient Celtic tribe, the Caledones, in Scotland gives its name to the most important mountain-building phase that affected the Arctic islands. The Caledonian Orogeny was the result of closure of an ancient ocean, now called Iapetus, that separated North America (with Scotland) from Europe (with England and Wales). Closure of the ocean and subduction led to the growth of a mountain range 100–200 km wide that extended from the Appalachians, through the northwestern British Isles, western Scandinavia, eastern Greenland, and eastern Svalbard. Closure of the ocean was accompanied by metamorphism, folding, and thrusting of the predominately shallow marine strata of the early sedimentary basins. Metamorphism ranged from low-grade (low temperature, high pressure) to high-grade (high temperature, high pressure). The greatest pressures and highest temperatures occurred at depths of 30–40 km, permitting the growth of new minerals such as garnet, hornblende, and mica. New layering, or foliation, defined by new minerals, overprinted the original sedimentary structures. Simultaneously, the rocks behaved plastically and were

subject to folding on scales ranging from kilometers to millimeters. In the upper few kilometers of the crust, the rocks were only slightly metamorphosed and were more brittle and prone to fracturing. Thrusts allowed translocations of large bodies of rocks (known as nappes) by tens of kilometers. Spectacular examples are present along the walls of eastern Greenland’s fjords. Intense deformation took place sporadically over the period from Ordovician to Silurian times (480–420 million years ago). The latter stages of the Caledonian Orogeny were accompanied by extensive intrusion of granite as a result of melting of the upper crust; again these rocks are well exposed in Svalbard and eastern Greenland. The Orogeny ended with reordering of continental fragments along strike-slip faults in a manner analogous to movements along the San Andreas Fault in California today. The end result of the Caledonian Orogeny was the unification of North America and Europe, a state of affairs that lasted until the modern North Atlantic began to open some 360 million years later. Similar, but smaller scale orogenies affected western Svalbard, northern Greenland, and Ellesmere Island and the Urals, Severnaya Zemlya, and Novaya Zemlya. The vast, high mountain chain was subject to erosion, and as the crust relaxed, intermontane basins developed. The erosion products accumulated in these dry continental basins as red beds: sandstone and conglomerate delivered by flash floods. Ephemeral lakes developed, but some lasted sufficiently long for early fish to flourish. These strata are referred to as the Old Red Sandstones and, broadly speaking, were deposited during the Devonian Period. YOUNGER SEDIMENTARY BASINS

Devonian events heralded a change throughout the Caledonian fold belt from continental collision through strike-slip motions to crustal extension. Sedimentary basins bounded by faults developed next, not only in the fold belt but in other parts of the High-Arctic. The Canadian Arctic Archipelago, western and eastern Greenland, Svalbard (Fig. 5), and Franz Josef Land have well-developed sedimentary sequences that span all geological periods from Carboniferous to Cretaceous (380–65 million years ago). Active phases of faulting led to pulses of sedimentation in both terrestrial and marine settings. Sediments include mud, sand, limestone, dolomite, and occasional coal and evaporite deposits such as salt. Marine environments were commonly rich in shelly faunas, whereas on land, plants and animals (including dinosaurs later on Svalbard) were abundant at times. Organic matter, especially in muddy sediments, decayed

ARCTIC ISLANDS, GEOLOGY

57

FIGURE 5 Carboniferous limestone and dolomite exposed in the walls

of Billefjorden, Svalbard. Photograph by M. J. Hambrey.

to produce hydrocarbons, with oil and gas migrating into adjacent porous sandstones. Extensive hydrocarbon exploration has taken place in Svalbard, Greenland, and the Canadian Arctic, with limited success, but large reserves of gas and oil are most likely to be found on the surrounding continental shelves. Oil and gas reserves have already been heavily exploited in equivalent rocks on the mainland areas, such as the North Slope of Alaska and Siberia. In Svalbard, localized deposits of Carboniferous coal were extracted until a decade ago. These deposits all carry a strong climatic signal and, in some regions, reflect movements from equatorial through tropical and temperate climatic zones. Intrusive activity occurred sporadically within these sedimentary basins, reflecting enhanced phases of crustal stretching. Prominent sills of Mesozoic basalt (a basic igneous rock) intrude sedimentary strata in Svalbard and the Canadian Arctic Archipelago, forming resistant cliffs in otherwise relatively soft sediment. VOLCANISM AND OPENING OF THE NORTH ATLANTIC OCEAN

The supercontinent that embraced Europe and North America was beginning to split apart toward the end of the Cretaceous Period. Separation began in the south with rifting, followed by sea floor spreading and the initiation of the North Atlantic Ocean. Opening of the ocean propagated northward, branching on either side of Greenland to form Baffin Bay and the NorwegianGreenland Sea. The opening of the ocean was heralded by extensive igneous activity, notably large-scale intrusion of granite derived from melting of the crust, and gabbros and basalts derived from the mantle. In addition, huge volumes of basalt were extruded, forming

58

ARCTIC ISLANDS, GEOLOGY

columnar cliffs and flat-topped plateaus. The most spectacular examples of these volcanic processes occur in eastern Greenland, where glaciers have carved out cross-sections through the intrusions and lavas, producing the most rugged scenery in the Arctic. These igneous events took place in the early Paleogene Period, being dated mainly to ∼52–55 million years ago in eastern Greenland and 65–55 million years in western Greenland and on Baffin Island. As new ocean floor was created at the northward-extending Mid-Atlantic Ridge, the continental areas became increasingly separated. Igneous activity ceased in the continental areas and became focused mainly on the submarine ridges. The most obvious manifestation of continental separation is evidenced where the Mid-Atlantic Ridge rises above the ocean, in Iceland and Jan Mayen. Sea floor spreading and active volcanism continues to this day, pushing North America away from Europe at a rate of several millimeters a year. Centered on the North Pole, the Arctic Ocean formed simultaneously, as a result of spreading from several midoceanic ridges, such as the Nansen-Gakkel Ridge. Today the Arctic Ocean consists two main basins, the Amerasia and the Eurasia, underlain by sea floor basalt. EMERGENCE OF PRESENT-DAY GEOGRAPHY UNDER A TEMPERATE CLIMATE

In many parts of the Arctic, igneous activity was accompanied by uplift, and the Arctic islands configuration began to emerge at this time. Crustal deformation and even localized mountain building (in western Svalbard) took place in the early Paleogene Period, as tectonic plates shuffled into their current positions. On the land areas, temperate climatic conditions prevailed, promoting the growth of extensive forests, comprising both coniferous and broadleaved trees. Extensive tree fossils are preserved on Axel Heiberg Island and Spitsbergen, and on the latter island vegetation accumulated sufficiently to generate exploitable coal measures, notably at Barentsburg, Longyearbyen, and Svea. These fossil forests clearly indicate that the climate was warmer and wetter than the present day, even though the islands were already in a high latitude. THE ICE AGE

The succeeding Neogene Period, approximately 10 million years ago, saw most of the principal elements of the Arctic land masses achieving approximately their current configuration. Along with many other parts of the world, the Arctic experienced sharp climatic cooling at this time. Glaciers began to form over Greenland, possibly as early as 11

million years ago, as evidenced by ice-rafted sediments on the ocean floor, delivered to the coast by tidewater glaciers and transported by icebergs offshore. The main growth of the northern hemisphere ice sheets over the Canadian Arctic, Greenland, and northern Eurasia is believed to have taken place in late Neogene/early Quaternary times, approximately 2.5 million years ago. The Arctic islands, especially Greenland, were affected by many glacial periods, but evidence remains only of the last one, named the Wisconsinan in North America, and the Weichselian in northwestern Europe, which spanned the interval 80,000 to 10,000 years ago. Glaciers remain extensively developed on many of the Arctic islands, and reached a recent historical peak around AD 1900, leaving behind extensive glacial and glaciofluvial deposits (Fig. 6). Beyond the ice limits, most of the Arctic islands are influenced by frozen ground (permafrost), in some places to depths of several hundred meters. Modern coastal and fluvial processes continue to rework sediments, especially those of glacial origin. Braided river plains are a characteristic feature of many parts of the Arctic.

and shrinkage of the area covered by sea ice, to name but a few. Belated recognition of global warming, induced by humans, and concerted action to reduce emissions, is too late to halt these changes on a centennial time scale; the best that can be anticipated is a reduction in the rates of change. SEE ALSO THE FOLLOWING ARTICLES

Arctic Islands, Biology / Climate Change / Continental Islands / Plate Tectonics FURTHER READING

Dowdeswell, J. A., and M. J. Hambrey. 2002. Islands of the Arctic. Cambridge, UK: Cambridge University Press. Haller, J. 1971. Geology of the East Greenland Caledonides. New York: Wiley & Sons. Harland, W. B. 1997. The geology of Svalbard. Memoir no. 17. London and Bath: Geological Society. Henriksen, N. 2008. Geological history of Greenland. Copenhagen: Geological Survey of Denmark and Greenland (GEUS). Hjelle, A. 1993. Geology of Svalbard. Polarhåndbok No. 7. Oslo: Norsk Polarinstitutt. Escher, A., and S. W. Watt, eds. 1976. The geology of Greenland. Copenhagen: The Geological Survey of Greenland. Worsley, D., and O. J. Aga. 1986. The geological history of Svalbard. Stavanger, Norway: Den Norske Stats Oljeselskap A.S.

THE FUTURE

Arctic island geology continues to evolve. On the multimillion-year time scale, plate movements will continue to change the geographical configuration of the Arctic, and islands will come and go. Today, in a sense, several Arctic islands are still under “ice age” conditions, with extensive glaciers and permafrost, and the surrounding seas are covered by sea ice in winter. However, under the influence of human-generated atmospheric pollutants, the Arctic region is warming at a rate faster than almost anywhere else on the planet, and major changes are already occurring, including recession of glaciers, melting of permafrost, and thinning

ARCTIC REGION STEPHEN D. GURNEY University of Reading, United Kingdom

The Arctic is that region of the northern hemisphere where the sun, for some time in summer, does not set and, for some time in winter, does not rise. It is a land of contrasts, of the polar night and the polar day, and its southerly limit is the Arctic Circle at a latitude of 66°33′ N. Islands within the Arctic region (see Fig. 1) include the Canadian Arctic islands, Greenland (considered to be the largest island on the planet) and its surrounding islands, Jan Mayen, Bjørnøya, the Svalbard archipelago, the Lofoten Islands, and the islands of the Russian Arctic including Novaya Zemlya, Zemlya Frantsa Iosifa, Severnaya Zemlya, Novosibirskiye Ostrova, and Wrangel Island. The North Pole is not on land; rather, it can be visited only by venturing onto the frozen surface of the Arctic Ocean. CLIMATIC SETTING

FIGURE 6 Thompson Glacier with push-moraine and braid-plain on

Axel Heiberg Island, Canadian Arctic Archipelago. Photograph by M. J. Hambrey.

The climate of the islands of the Arctic region varies greatly, although it is typified by a negative heat balance

ARCTIC REGION

59

glaciers of the Greenland Ice Sheet, and these drift out into the North Atlantic via fjord systems.

180° Arc

ARCTIC REGION

tic C

ircl e

66 °3 3'

N

PERMAFROST

WRANGEL ISLAND

CANADIAN ARCTIC ISLANDS

80°N

NOVOSIBIRSKIYE OSTROVA

Arctic Ocean

SEVERNAYA ZEMLYA

North Pole

90°W

ZEMLYA FRANTSA IOSIFA

90°E

NOVAYA ZEMLYA

SVALBARD GREENLAND

In areas without glaciers, the cold climate results in the formation and maintenance of permafrost. Permafrost is a thermal condition of the ground whereby it is frozen throughout the year to depths of up to several hundred meters. In the brief Arctic summer, however, the positive air temperatures thaw a thin layer of the surface materials, known as the “active layer.” Under certain conditions the summer warmth can lead to extensive thawing of ice lenses at the base of the active layer, which can cause it to become detached, resulting in large slope failures.

Barents Sea BJØRNØYA JAN MAYEN

800 km 500 miles

NORWEGIAN ARCTIC ISLANDS 0°

FIGURE 1 The Arctic Region and the islands within it.

or budget, meaning that mean annual air temperatures can be lower than -15 °C and winter temperatures may fall to lower than -50 °C. Winters are generally cold and stormy, and snow cover can last six months or longer. Summers are short and often cloudy, but they may be mild. The proximity of sea ice or the existence of a warm ocean current can result in colder or warmer conditions than would otherwise be expected. All areas experience the “Arctic Night” in winter when the sun does not rise above the horizon for a period of up to several months (depending on latitude). In summer, of course, the “Midnight Sun” creates 24-hour daylight. GLACIERS

The largest body of ice found on the Arctic islands is the Greenland Ice Sheet; however, it does not lie totally within the Arctic (the southern tip of Greenland is found at 60° N). Although smaller than the East Antarctic and West Antarctic Ice Sheets, it is still vast at over 1.8 million km2. The nature of other ice masses on the Arctic islands is generally dictated by the terrain and precipitation regime. For example, Baffin Island supports lowland ice caps, whereas Novaya Zemlya supports mountain ice caps and valley glaciers of the outlet type. In some areas the glaciers flow down to meet the sea and form floating glacier tongues which calve (create) icebergs into the sea. The largest icebergs originate from the outlet

60

ARCTIC REGION

GEOMORPHOLOGY

The geomorphology of the Arctic islands is varied but is dominated by glacial processes and landforms where glaciers are present and by permafrost and periglacial processes where they are not. Permafrost landscapes are often dominated by ground-ice landforms, the most important of which are ice wedges. These features are created over many years because the ground cracks open in winter as a result of thermal contraction. The cracks become filled, first with veins and then later with wedges of ice, and these grow over successive winters until they may be as much as 4 m wide at their tops and 6 m deep. At the surface they can be seen through the polygons that they produce, which may be low- or high-centered and can have a surface relief of up to 2 m. The ice wedges themselves are hidden from view by the active layer, which closes the uppermost portion of the ground cracks in summer as it thaws. The slopes in permafrost terrain may be composed of solifluction sheets or lobes, which slowly creep downhill through the growth and subsequent thaw of ice within the active layer on a seasonal basis; such processes can operate on slopes with angles of only a few degrees. Pingos are small ice-cored hills and are another feature unique to permafrost. Some of the best examples are seen in the Arctic islands, for example on Traill, in east Greenland, and in the Svalbard archipelago, where they can attain heights of up to 40 m and diameters of over 500 m. They grow when water under pressure makes its way toward the surface but freezes just beneath it to form a conical ice core. As the ice core grows it forces up the ground above it into a hill, which is the pingo. In lowland tundra areas the source of water under pressure results from permafrost growth into the saturated sediments exposed by lake drainage. In mountainous regions

the source is upwelling groundwater that originates on the hillsides high above the valley floor. Rock glaciers also form in the mountainous permafrost environments of the Arctic islands. These take the form of lobes or tongues of rock debris where the rock fragments are cemented together by ice and flow slowly downhill through the slow deformation of the ice. COASTAL PROCESSES

The coastal processes and geomorphology of the islands of the Arctic region are similar to those found in other areas, except where there is a role played by permafrost or sea ice. Where ground ice contents are high, the seawater has the ability to erode thermally—that is, its relative warmth thaws the ice within the sediments that it contacts—and this can lead to a loss of strength and to enhanced erosion. In a microtidal environment, such as is found in much of the Arctic, this can lead to the undercutting of soft sediment cliffs and to block collapse. In areas with sea ice, coastal morphology can be shaped by the effects of the ice being blown or driven onto the island fringes by winds, tides, or currents, especially where the islands are low-lying. FURTHER READING

Burn, C. R. 1995. Where does the polar night begin? The Canadian Geographer 39: 68–74 Dowdeswell, J. A., and M. J. Hambrey. 2002. Islands of the Arctic. Cambridge, UK: Cambridge University Press. Gurney, S. D. 1998. Aspects of the genesis and geomorphology of pingos: perennial cryogenic mounds. Progress in Physical Geography 22: 307–324. Humlum, O. 2000. The geomorphic significance of rock glaciers: estimates of rock glacier debris volumes and headwall recession rates in west Greenland. Geomorphology 35: 41–67. Lewkowicz, A. G. 2007. Dynamics of active-layer detachment failures, Fosheim peninsula, Ellesmere Island, Nunavut, Canada. Permafrost and Periglacial Processes 18: 89–103. Mackay, J. R. 2000. Thermally induced movements in ice wedge polygons, western Arctic coast: a long term study. Géographie Physique et Quaternaire 54: 41–68.

ited by some of the leading nineteenth-century traveling scientists, including Charles Darwin and Joseph Dalton Hooker, and it has a range of endemic species of both scientific and conservation importance. PHYSICAL GEOGRAPHY AND LANDSCAPE

Ascension is a volcanic island located in the tropical south Atlantic (7°57′ S, 14°22′ W) with an area of about 97 km2. Much of its scientific interest comes from its remote location. The nearest point in Africa is 1504 km away, and the island is 2232 km from South America; the closest other island is St. Helena, 1300 km to the southeast. The oldest recorded rocks on the island are approximately 1 million years old. Therefore, the ancestors of all the organisms living on the island either dispersed to it naturally during the last million years or have been brought there by humans in the last few centuries. Most of the island is very arid, providing a dramatic volcanic landscape dotted with scoria cones—large cones of volcanic material formed by explosive eruptions. When the marine biologist and amateur watercolorist Alister Hardy visited in 1925, he described these cones thus: “although barren they present a great variety of colour: raw sienna, reds, browns, dark and light grays and yellows, while some are almost crimson—all changing tone with the light and shade from passing clouds” (Fig. 1). The highest point on the island is Green Mountain (845 m; declared a national park in 2005). Here there is lush tropical vegetation mainly composed of introduced plant species. It is probably best described as cloud forest—the water supply for this vegetation coming from clouds blown onto the summit from the ocean by the trade wind. The very top of the “mountain” is dominated by bamboo

ASCENSION DAVID M. WILKINSON Liverpool John Moores University, United Kingdom

Remote oceanic islands such as Ascension have long played an important role in the study of ecology and evolutionary biology. Although historically less important for these subjects than the Galápagos, Ascension was vis-

FIGURE 1 Much of Ascension is an arid volcanic landscape. The pho-

tograph shows the volcanic cone of “Sisters Peak.” The white marks on the rocks in the foreground are the remains of guano from a former seabird colony (often referred to as “ghost colonies”); such easily accessible colonies failed to survive predation by introduced cats.

ASCENSION

61

FIGURE 2 Green Mountain photographed from the top of Breakneck

Valley. The peak is enveloped in cloud, which provides the water to support the cloud forest vegetation on the summit, largely composed of introduced plant species. Much of the northeastern side of the “mountain” is exposed to the full force of the trade winds and is covered in a grass-dominated vegetation—again mainly comprising introduced species.

and has an artificially dug pond, making it one of the very few freshwater habitats on the island (Fig. 2). At sea level the temperature is relatively constant, usually varying between 27 and 31 °C, whereas at 660 m on Green Mountain, temperatures in the low 20s are more often recorded. Rainfall at sea level is approximately 140 mm per year, although the island was presumably wetter at some point in the past; the crater called the “Devil’s Riding School” contains old lake deposits—first recognized as such by Charles Darwin when he visited in 1836 on the return leg of the Beagle voyage. BIOLOGY

As with many remote islands, one of the reasons Ascension is of interest to conservationists and evolutionary biologists is its endemic species—which must have evolved there during the last 1 million years. The flora, prior to human introduction, was very limited, presumably due to a combination of the island’s remote location and geologically recent origin. When humans first arrived on the island, in the early sixteenth century, it is thought that there were around 25 native plant species, ten of which were endemic (four of these endemics are now extinct). Today an additional 280-plus introduced plant species grow on the island, threatening the survival of some of the remaining endemics. Many of these species were deliberately introduced during the nineteenth century in an attempt to improve the environment for humans. The only tree in the native flora was Oldenlandia adscensionis (this may always have been rare as Darwin failed to notice it when he visited); it was last recorded in 1889. 62

ASCENSION

There is currently one species of endemic bird on the island, the Ascension Frigatebird Fregata aquila; however, sub-fossil remains of two extinct endemic birds have been found in cave deposits. There was a flightless rail Mundia elpenor (formerly placed in the genus Atlantisia), which was still present when humans arrived in the sixteenth century, and a small night heron Nycticorax olsoni, whose date of extinction is not known. The island is an important breeding ground, not only for the frigatebird but also for many species of tropical seabird, and it is also famous for the green turtles, Chelonia mydas, which nest on its beaches and have formed the subject of extensive scientific research. There is an equally interesting invertebrate fauna, containing at least 25 endemic terrestrial species; this includes five species of endemic pseudoscorpions. One of these, Garypus titanius, confined to Boatswain Bird Island, is the largest known pseudoscorpion in the world. There is also a land crab, Johngarthia (Gecarcinus) lagostoma, which is restricted to Ascension and a few islands off the coast of Brazil. On Ascension it mainly lives above 200 m, where there is more vegetation than in the arid lowlands; however, it has to migrate across these lowlands to the sea to breed. Little is known about the microorganisms on Ascension; a recent study of the soil protozoa recorded 52 species, all of which have wide global distributions. CURRENT STATUS AND CONSERVATION

The discovery of the island in the early sixteenth century led to the introduction of many non-native species. These included a long list of plant species, which have competed with the limited native flora, and animals such as rats and cats which proved very destructive to seabird populations. During the twentieth century, with the exception of sooty terns, Sterna fuscata, all these seabirds have been confined to breeding either on the steep cliffs of the main island or on Boatswain Bird Island off the east coast—to escape predation by introduced cats and rats. Guano deposits show that seabird colonies were much more extensive on the main island in the past (Fig. 1). The successful eradication of feral cats (completed in 2004) has lead to some seabirds starting to breed on the main island again. Ascension is also famous for the green turtles, Chelonia mydas, which nest on its beaches. Recent analyses of historical records suggests that there was a decline in the numbers breeding between 1822 and 1935, during which time there was a commercial harvest of turtles, this harvest stopped in the 1940s. More recently numbers have increased greatly—by an estimated 285% since the 1970s.

Plant conservation has been more limited than seabird conservation, but it has recently included attempts to propagate some of the endemic species—such as the fern Pteris adscensionis—under nursery conditions. SEE ALSO THE FOLLOWING ARTICLES

St. Helena / Tristan da Cunha and Gough Island / Voyage of the Beagle FURTHER READING

Ashmole, P., and M. J. Ashmole. 2000. St Helena and Ascension Island: a natural history. Oswestry, UK: Anthony Nelson (current distributor: www.kidstonmill.org.uk.). Gray, A., T. Pelembe, and S. Stroud. 2005. The conservation of the endemic vascular flora of Ascension Island and threats from alien species. Oryx 39: 449–453. Hartnoll, R. G., T. MacKintosh, and T. J. Pelembe. 2006. Johngarthia lagostoma (H. Milne Edwards, 1837) on Ascension Island: a very isolated land crab population. Crustaceana 79: 197–215. Wilkinson, D. M. 2004. The parable of Green Mountain: Ascension Island, ecosystem construction and ecological fitting. Journal of Biogeography 31: 1–4. Wilkinson, D. M., and H. G. Smith. 2006. An initial account of the terrestrial protozoa of Ascension Island. Acta Protozoologica 45: 407–413.

ATLANTIC REGION FIGURE 1 Map of the Atlantic with locations of islands covered in this

ANDREAS KLÜGEL University of Bremen, Germany

article (GLOBE Task Team and others. 1999. The Global Land Onekilometer Base Elevation (GLOBE) Digital Elevation Model, Version 1.0. Boulder, CO: National Oceanic and Atmospheric Administration, National Geophysical Data Center, URL: http://www.ngdc.noaa.gov/

The Atlantic is the world’s second largest ocean, forming an elongated basin between the Arctic Ocean in the north and Antarctica in the south. It owes its existence to the break-up of the supercontinent Pangaea, which began around 180 million years ago in the North Atlantic and 130 million years ago in the South Atlantic. Today the Atlantic continues to widen at rates of approximately 1.8–3.5 cm per year by seafloor spreading along the MidAtlantic Ridge, a submarine mountain range separating the ocean into an eastern and a western basin. The Atlantic harbors a number of islands of mostly volcanic origin that experience a wide range of maritime climates, from polar in the high latitudes to tropical around the equator (Fig. 1). Many of these islands and associated seamounts, for example the Canary Islands, form volcanic chains, which reflect movement of the underlying tectonic plates over a hotspot. Overall, these volcanic chains are less conspicuous and less frequent in the Atlantic than in the Pacific. This article gives an overview of the Atlantic’s islands, their ages, and how they were formed; it is orga-

mgg/topo/globe.html).

nized by geographic-geologic groups roughly from north to south. ICELAND, SURTSEY, FAROE ISLANDS

Iceland is a highly active volcanic island, located just where a hotspot meets the Mid-Atlantic Ridge. It is the only place where the west-east spreading of the Eurasian and North American tectonic plates can be witnessed on land. The oldest rocks of Iceland (formed 13 to 16 million years ago) are therefore exposed at its western and eastern ends, and the youngest rocks are found closer to the center. The cause for the hotspot and for the uplift of the Iceland plateau is a deep-seated mantle plume, the track of which is represented by the Greenland-Faroe ridge, which formed during the opening of the northeast Atlantic. The Faroe Islands represent an earlier stage of the Iceland hotspot; they are of volcanic origin and consist predominantly of lava flows between 54 and 58 million years old.

AT L A N T I C R E G I O N

63

Iceland’s volcanically active areas are the Western, Eastern, and Northern Volcanic Zones, which are rift zones crossing the island in north-south to northeast-southwest directions. The current locus of the Iceland plume axis is at the Vatnajökull glacier, where subglacial volcanic eruptions periodically cause large meltwater floods known as jökulhlaups. Between 1963 and 1967, volcanism off southwest Iceland gave birth to Surtsey Island, which is one of the youngest islands in the world and a classic natural reserve for the study of biocolonization. The volcanic activity of Iceland is also reflected by abundant hot springs and geysers. Iceland is composed primarily of basaltic lavas and pyroclastics erupted from fissures and shield volcanoes. Central volcanoes located within fissure swarms have also produced rhyolitic ash-flow deposits in calderaforming eruptions. About 11% of Iceland’s land area is covered by glaciers, and about half is of Quaternary volcanic origin forming much of the inhospitable and uninhabited central highland. Despite its location immediately south of the Arctic Circle, Iceland’s climate is temperate because of the Gulf Stream’s moderating influence. ISLANDS OFF NORTHWESTERN AFRICA (MACARONESIA)

Macaronesia (Greek for “fortunate islands”) covers the archipelagoes of the Azores (Portugal), Madeira with Selvagens (Portugal), the Canary Islands (Spain), and Cape Verde, which have pleasant subtropical to tropical climates. All these islands are entirely of volcanic origin and are the expression of several hotspots likely underlain by mantle plumes. Apart from the Azores, the Macaronesia islands and associated seamounts form part of a remarkable volcanic belt off northwestern Africa extending from the Azores-Gibraltar Fracture Zone to the Sierra Leone Rise. The origin of this belt, however, is not yet fully understood. The Azores archipelago consists of nine major islands rising from a broad submarine plateau; hence, there is a topographic Azores high in addition to the meteorologic one. The archipelago likely owes its existence to a mantle plume and to its special tectonic setting near a triple junction. The westernmost islands Flores and Corvo are located on the North American plate, and the other islands are on the Azores microplate, which is bounded by the North American plate along the MidAtlantic Ridge in the west, by the Eurasian plate along the Terceira Rift (a spreading center) in the northeast, and by the African plate along the East Azores Fracture Zone in

64

AT L A N T I C R E G I O N

the south. The oldest island of the archipelago is Santa Maria (more than 8 million years old) and the youngest is Pico, where the archipelago’s highest peak (2351 m) is located. Historic volcanic eruptions occurred on São Miguel, Pico, São Jorge, Faial, and Terceira, with São Miguel being the most active island. The last volcanoes to erupt were Capelinhos in the western part of Faial in 1957 and the shallow João de Castro seamount between São Miguel and Terceira in 1997. The volcanic rocks of the Azores span a wide range in composition from alkalic basalts to alkali-rich rhyolites. The Madeira Archipelago comprises Porto Santo (14.3–11.1 million years old), Madeira (5.3 million to ∼6,000 years old), and the three inhabited Desertas Islands (5.1–1.9 million years old). Madeira island is the present locus of a discontinuous hotspot track that can be traced back 67 million years along Seine, Ampère, and Ormonde seamounts to the northeast. The archipelago’s highest peak (1862 m) is located within the deeply eroded central highland of Madeira. The islands consist dominantly of basaltic lava flows and pyroclastics, and the elongated and spectacular Desertas Islands show the deeply eroded interior of a volcanic rift zone. At Porto Santo the shallow submarine stage of an emerging seamount is well exposed. Most rocks of the archipelago have compositions ranging from basanite and alkalic basalt to basaltic trachyandesite to trachyte. The Selvagen Islands are located between the Madeira and Canary Archipelagoes and form the summits of two extinct shield volcanoes. Geochemical studies reveal that they are part of the Canary hotspot track (thus, though the Selvagens belong to Portugal politically, they belong to Spain geologically). The small islands consist predominantly of alkalic basaltic, basanitic, and phonolithic volcanic rocks ranging in age from 29 to 3 million years old. The presence of marine carbonate sediments on Selvagem Grande show that the top of the volcano was eroded and submerged beneath sea level during a volcanic hiatus between 24 and 12 million years ago. The Canary Islands form a chain of seven major volcanic islands that are all underlain by Jurassic oceanic crust. The ages of the islands’ oldest subaerially erupted rocks decrease roughly from 21 million years at Fuerteventura in the east to 1.7 million at La Palma and 1.1 million at El Hierro in the west, consistent with a hotspot origin. The hotspot track can be traced back 68 million years along a chain of seamounts to the northeast. Despite their age progression, all of the Canary Islands except for La Gomera are still volcanically active. Historic eruptions occurred on La Palma (the most active

Canary Island in historic times), Tenerife, and Lanzarote. La Palma features the world’s largest erosional crater, in which an uplifted seamount is exposed, and Pico de Teide on Tenerife is the Atlantic’s highest peak (3717 m). The Canary Islands comprise almost the entire spectrum of volcanic eruption products including lava flows, fallout, nonwelded and welded ash flows, and surge deposits. The compositional spectrum of the rocks is also extremely large, ranging from tholeiitic basalts to highly silica-undersaturated alkalic basalts to phonolites and peralkaline rhyolites. The Cape Verde Archipelago is horseshoe shaped and comprises ten main islands along a northern (Barlavento) and southern (Sotavento) group. The islands are situated on the Cape Verde Rise, which represents a hotspot swell. Because of large local uplift in the lithosphere, Jurassic oceanic crust is now exposed on Maio and São Tiago islands. Subaerial volcanic activity on the archipelago took place from around 16 million years ago (Sal island) until the present, and although there is no simple age progression, a crude decrease in age from east to west is recognized, with Santo Antão and Fogo islands showing the greatest amount of recent volcanism. Historic volcanic activity occurred on Fogo only, which also features the highest peak of the archipelago (2829 m) and a huge recent collapse scarp. The islands show a variety of rock types ranging from nephelinites and alkalic basalts to phonolites. Carbonatites (carbonate-rich igneous rocks), which are exceptional in ocean settings, occur on Brava, Fogo, São Tiago, Maio, and São Vicente. BERMUDA ISLANDS

The Bermuda islands belong to the 1500-km-long Bermuda Rise, an elongated platform built on 100–105-million-yearold oceanic crust that was caused by some type of hotspot. The islands are the surface expression of an extinct shield volcano that had been uplifted and eroded. Submarine intrusive and extrusive activity of this volcano occurred between 45 and 33 million years ago. The present Bermuda islands rise up to 76 m above sea level, forming a 15– 100 m-thin carbonate cap on the 4500-m-tall truncated stump of the old volcano. They are predominantly composed of Quaternary carbonate sandstones originally eroded from biogenic, primarily coral reef limestones during the low sea levels of several glaciations. The islands show a central lagoon, but the relation of the present morphology to the volcanic basement is unknown. There is also no evidence for a volcanic caldera often said to be outlined by the islands. The Bermuda platform is the most northerly coral reef habitat in the Atlantic Ocean

and shows similarities to Pacific atolls. The climate of the islands is warm-temperate or oceanic with high humidity. GULF OF GUINEA ISLANDS

The islands of Bioko (or Fernando Po), Príncipe, São Tomé, and Pagalu (or Annobón) are of volcanic origin and are extensions of the Cameroon line, a 1600-km-long chain of intra-plate volcanoes extending from inland Cameroon to Pagalu island. The ages of the oldest exposed rocks decrease from 31 million years on Príncipe to 13 million years on São Tomé to 5 million years on Pagalu, consistent with movement of the African plate over a hotspot. Bioko is an exception from this age progression, being a young volcanic island located close to the continental Cameroon line volcanoes. Although none of the islands have had historic eruptions, São Tomé and Pagalu show signs of recent volcanic activity, whereas Príncipe is more deeply eroded. The volcanic rocks range in composition from alkalic basalts to phonolites and rare trachytes, some of which form morphological domes and plugs. Being situated close to the equator, the islands experience a hot and humid tropical climate. ISLANDS OFF EASTERN BRAZIL

The Fernando de Noronha archipelago consists of the main island and 20 small islets, which are all of volcanic origin and rise more than 4000 m above the seafloor. The volcanic rocks of Fernando de Noronha are between 12 and 1.7 million years old, and the youngest volcanic activity is represented by a single lava flow on the islet of São José. The islands are part of a 500-km-long hotspot track that can be traced back towards Fortaleza, where 30 million-year-old volcanic rocks are exposed. A seamount to the east of Fernando de Noronha may be volcanically active and represent the current hotspot location. The archipelago is composed mainly of highly alkaline, silica-undersaturated rocks including phonolitic and trachytic plugs and domes and various basalts. The climate is tropical oceanic with well-defined rainy and dry seasons. Trindade and Martín Vaz are small rugged volcanic islands rising 5500 m above the seafloor. They are the easternmost and youngest expression of a 1200-km-long hotspot track that is well defined by a seamount chain to the west. Trindade is about 10 km2 in size, up to 600 m high and ∼5 million years old, with the last eruption having produced Vulcão de Paredão 10,000 to 5000 years ago. It is uninhabited except for a permanent Brazilian navy base. The tiny Martín Vaz islets (0.3 km2), located 47 km east of Trindade and the easternmost Brazilian territory, are hard to access and are about 1 million years old. Trindade and

AT L A N T I C R E G I O N

65

Martín Vaz, together with Fernando de Noronha, represent the most alkaline province among oceanic volcanic islands of the world. The extremely sodium-rich and silica-undersaturated rocks comprise abundant phonolitic domes and plugs and different types of basalts. The Saint Peter and Paul Rocks are located 870 km northeast of Fernando de Noronha close to the MidAtlantic Ridge. They consist of five rugged islets and four rocks reaching less than 20 m above sea level within an area of about 200 by 350 m, and they are the surface expression of a 3800-m-high sigmoidal massif. Only the largest rock, on which a lighthouse was built, shows some low vegetation. The rocks are not of volcanic origin but are composed of peridotite, an olivine-rich rock typical of the Earth’s mantle, and are thus a unique surface outcrop of mantle rocks within an ocean. The cause for the uplift of the St. Peter and Paul Rocks massif is still-occurring tectonic activity along the St. Paul Fracture Zone. ISLANDS IN THE SOUTH CENTRAL ATLANTIC

These volcanic sister islands are British overseas territories and are among the world’s most remote islands. Ascension (98 km2) rises near the Mid-Atlantic Ridge from a depth of 3000 m to a height of 859 m above sea level. It is a hotspottype island but is not associated with a hotspot track. Subaerial volcanism occurred from approximately 1 million years ago to the present, and much of the island is morphologically young, with many cinder cones; however, no eruptions have been reported since its discovery in 1501. The volcanic rocks range from mildly alkaline basalts to trachytes and some rhyolites comprising abundant pyroclastics and lava flows. St. Helena (120 km2) rises from a 4400-m depth to 823 m above sea level and is near the end of a broad and diffuse hotspot track heading northeast toward Cameroon (but not belonging to the Cameroon line hotspot track). Subaerial volcanic activity on St. Helena occurred from 14 to 7 million years ago, and since then erosion has produced tall cliffs. The island is dominated by basaltic to trachytic lava flows, some trachyte intrusions, and subordinate pyroclastics. Tristan da Cunha and the adjacent uninhabited Nightingale and Inaccessible Islands form the end of a prominent but broad hotspot track that includes Walvis Ridge to the northeast and is the expression of the Tristan/ Gough plume. Tristan (98 km2) is a huge, almost circular volcano rising from a 3700-m depth to 2062 m above sea level; the subaerial rocks are less than 1 million years old. Because of prevailing steep slopes and tall cliffs, a small plateau where the settlement of Edinburgh is located is the only habitable part of the island. It is exactly there that a volcanic eruption occurred in 1961. The Tristan

66

AT L A N T I C R E G I O N

consists predominantly of pyroclastics and lava flows of alkali basaltic through trachytic composition. Gough Island (or Diego Alvarez island; 65 km2) is located about 350 km southeast of Tristan da Cunha and is part of the same hotspot chain. It rises to 907 m above sea level and is less than 1 million years old. The compositions of the lavas, pyroclastics, and dikes on the island range from mildly alkaline basalt through trachyte. Except for a weather station, the remote island is uninhabited. Gough is a protected wildlife reserve and harbors one of the Atlantic’s least disrupted ecosystems. The climate of Gough and Tristan da Cunha is marine subtropical with only small temperature variations. FALKLAND ISLANDS

The Falkland Archipelago (Spanish: Islas Malvinas) consists of two main islands, East Falkland and West Falkland, and several hundred smaller ones. It has a total land area of about 12,000 km2 and rises to 705 m. The islands are located 500 km off southern Argentina on the Patagonian shelf and have a cold, windy, and humid maritime climate. They represent fragments of continental crust resulting from the break-up of Gondwana and the opening of the South Atlantic that began about 130 million years ago. Because of this breakup, the Falkland Islands display similar rock sequences as southern Africa and Queen Maud Land, Antarctica. The crystalline basement consists of 1100–1000-million-year-old gneisses and granitoids, which are overlain by a Devonian through Triassic siliciclastic succession (about 400 to 210 million years old) dominated by quartzite, sandstone, and mudstone. Beginning about 280 million years ago, the rocks came to be strongly deformed and folded, and subsequently intruded by Jurassic mafic dikes (190 million years old), the youngest rocks. The islands’ surface was finally modified during the Pleistocene glaciations. SOUTH GEORGIA, SOUTH SANDWICH ISLANDS, SOUTH ORKNEY ISLANDS

These islands are the subaerial part of the Scotia arc that encircles the Scotia Sea (a back-arc basin) and extends from Tierra del Fuego along the submarine Scotia Ridge, South Georgia, South Sandwich, and South Orkney Islands to the Antarctic Peninsula. They owe their existence to complex tectonic processes related to the breakup of Gondwana and the opening of the South Atlantic. The climate of the islands is harsh, cold, wet, and windy. The South Georgia Islands lie 1400 km east of the Falkland Islands and are the emergent part of an isolated microcontinental block with similar geology as Tierra del Fuego. The main island (3528 km2) rises to 2934 m and is largely covered

by glaciers, ice caps, and snow fields. The islands largely consist of thick late Jurassic to early Cretaceous (140–110-millionyear-old) turbidite sequences of volcaniclastic and siliciclastic sandstones and shales that were strongly deformed and folded around 91–82 million years ago. The oldest rocks are gneisses and schists intruded by middle Jurassic (164-million-year-old) granites and gabbros. In southeastern South Georgia some uplifted 150-million-year-old basaltic ocean crust (an ophiolite) is exposed. The youngest rocks, about 80–100 million years old, are remnants of a former volcanic arc. The South Sandwich Islands comprise 11 uninhabited volcanic islands with a total area of 310 km2 and the highest peak, a stratovolcano, reaching 1370 m in height. They represent a volcanic arc west of the Scotia Trench where the South American plate is being subducted from the east. The islands are volcanically active with rocks created between about 4 million years ago and the present. The compositions of the volcanic rocks is calc-alkaline ranging from basalt to rhyolite. The South Orkney Islands are part of a continental fragment on the southern Scotia arc. They comprise five main islands about 600 km northeast of the Antarctic Peninsula with a total area of 620 km2 and a height of up to 1266 m. The islands consist largely of metamorphosed and folded sedimentary and volcanic rocks of early to middle Jurassic age (205–176 million years old) as well as Triassic to early Cretaceous sediments. The subantarctic islands are barren, uninhabited, and largely covered by ice and snow.

FURTHER READING

Faure, G. 2001. Origin of igneous rocks: the isotopic evidence. Berlin: Springer. Mitchell-Thomé, R. C. 1970. Geology of the South Atlantic islands. Berlin: Borntraeger. Sigurdsson, H., ed. 2000. Encyclopedia of volcanoes. San Diego, CA: Academic Press. Vogt, P. R., and B. E. Tucholke. 1986. The geology of North America, vol. M: the western North Atlantic region. Boulder, CO: Geological Society of America. The Great Plume Debate Website. http://www.mantleplumes.org/.

ATOLLS EDWARD L. WINTERER Scripps Institution of Oceanography, La Jolla, California

Atolls are a special type of coral reef complex formed in tropical seas by a ring of reef coral enclosing a lagoon. The characteristic features of an atoll include a reef rim, from 100 to 500 m across, which is mainly awash at high tide, and flattish islands (motu), which remain a few meters above sea level and on which people may live. The word atoll itself comes from the language of the Maldive Islands, a chain of large atolls in the Indian Ocean (Fig. 1). Atolls range in size from a few to as many

BOUVET ISLAND

Bouvet is a small volcanic island (about 50 km2) rising from near the southernmost end of the Mid-Atlantic Ridge to 780 m above sea level. It is the world’s most remote island lying about 1600 km from the nearest land, Queen Maud Land, Antarctica. Bouvet owes its existence to a melting anomaly of the underlying mantle or a small mantle plume. The island forms the top of a large stratovolcano, the maximum age of which is constrained by the age of the underlying oceanic crust of 6 million years. The volcano is considered active, as shown by the formation of a lava shelf on the west coast in the 1950s. The main eruptive center is in the northwest of the island, where fumaroles occur and a caldera may have collapsed. The lava compositions range from basalts to trachyandesites to alkalic rhyolites and obsidian. Most of Bouvet is covered by glaciers, which together with high cliffs and harsh weather conditions make landing extremely difficult. The harsh polar climate at Bouvet restricts the vegetation to litchens and mosses, and the fauna comprises seals, penguins, and seabirds. The island is uninhabited and was designated a nature reserve in 1971.

FIGURE 1 Air view of southern part of the Maldive chain of atolls in the

(cloudy) Indian Ocean. Taken looking north from 2.4° N, 73.3° E. Kolumadulu Atoll, in center (at 2°25' N, 73°10' E), is about 50 km across. NASA photograph STS037-97-46. Image courtesy of the Image Science and Analysis Laboratory, NASA Johnson Space Center.

AT O L L S

67

as 40,000 km2, and lagoon depths range from almost nothing to as much as 100 m. DISTRIBUTION

The minimum water temperature for the approximately 600 atolls around the globe is about 20 °C, a temperature that prevails from about 24° N to 29° S. Reef-building corals are partly dependent on their embedded photosynthesizing algal symbionts for adequate oxygen, and thus grow healthily only in lighted waters less than a few tens of meters deep. Slow-growing, deep-water coral reefs, without algal symbionts, are known to exist at depths down to about 6000 m. Many atoll lagoons are connected directly to the sea by inlet channels, but few channels are wide or deep enough to be navigable by anything but canoes. Atolls with wide, deep channels are favored for tourism and for administrative headquarters. People from other atolls in the region tend to migrate to these centers, and overcrowding is a problem on some. The main source of income on most atolls is copra from coconuts, and tourism is important on a few. THE REEF RIM

Between the ocean and the lagoon, the reef comprises an outer, wave-resistant algal ridge a few tens to a hundred meters wide at sea level, against which the ocean swell breaks. Seaward, the ridge drops steeply off into deeper, less turbulent water, and corals thrive along these slopes. The outer part of the algal ridge is commonly notched to a depth of 10–20 m by a reticulate network of narrow (1–3 m) grooves and surge channels that funnel water across the reef rim toward the lagoon. Behind the algal ridge is a reef flat, close to low tide level, where a pavement of living and dead coral and coral debris from the size of sand grains to the size of boulders extends lagoonward. Reef debris thrown onto the flat during typhoons lies in patches and windrows 1–2 m high in some places. Large and small islands, termed motu, surmount the reef flat in places and rise to as much as 4 m above sea level (Fig. 2). They are commonly edged by a steep scarp toward the sea, created by waves beating against the island. It is only on these islands that the people of the atolls can live, where they are safely above the surf. The motu are weathered remnants of reef flats dating from the time of a regionally (in low latitudes) higher stand of sea level, about 2000–4000 years ago. The highstand resulted from the isostatic response to a redistribution of ocean waters after deglaciation of the northern continental ice sheets, beginning about 18,000 years ago. Between motu are shallow channels (hoa) that direct washover from ocean

68

AT O L L S

FIGURE 2 Leeward side of Tikehau atoll in the Tuamotu archipelago (15°

S, 148° W), showing closely spaced surge channels next to the shore, the reef flat surmounted by wooded motu separated by channels (hoa) leading to the lagoon. The opposite shore of the lagoon is visible in the background as a dark line. Photograph by Pierre Labout, with permission of the Institut de Recherche pour le Développement, Paris.

waves toward the lagoon. Shallow lakelike depressions, termed faros, pock reef flats on some atolls. These are the now-drowned remnants of karstic depressions dissolved by meteoric waters during the last lowstand of sea level. In addition to carefully collecting and storing rainwater, atoll islanders depend on a precarious groundwater supply beneath the motu, which accumulates as a freshwater lens floating on the denser saltwater beneath. Seepages of this water are used to water vegetable gardens, and wells tap the lens for potable water. LAGOONS

Atoll lagoons are mainly of fairly normal ocean salinity, except for completely landlocked lagoons surrounded by relatively high islands. These lagoons may include shallow salt pans and briny lakes. Detailed charts of atoll lagoons commonly show patch reefs, some rising to sea level but others submerged so as to constitute hazards to navigation. Sediment builds up in the lagoon almost entirely during sea-level highstand times, when ocean waves can wash fine sediment across the reef. Shells of organisms living in the lagoon add to this thickness. At lowstand times, rain falling into the perched lagoonal depression may drain to the surrounding sea through erosional channels cut through the reef rim. ORIGIN OF THE ATOLL FORM

The origin of the ring-around-a-lagoon form has attracted the attention of geologists for two centuries, beginning with Charles Darwin (1842), who never actually visited an atoll during his Beagle voyage but made a painstaking study of charts, supplemented by his own observations on

other coral-surrounded islands. He showed that tropical islands displayed a series of reef forms, from fringing reefs that cling to the shore, to barrier reefs separated from the island shore by a lagoon, to atolls, with a lagoon but no bedrock island. From his experiences in South America, where his hikes took him to shelly marine deposits high on the Andes slopes, connoting tectonic uplift, he reasoned that there must be a compensating subsidence in the Pacific basin and that the array of reef forms—fringing, barrier, atoll—must be due to regional subsidence of the islands. According to Darwin, the original fringing reef would grow upward, keeping pace with the relative rise of sea level during island subsidence, but as subsidence continued, the original fringing reef would continue to grow upward, and coral growth would be inhibited in the less turbulent and less fertile waters inland from the reef front. The perimeter would keep up with the rising sea level, whereas the more interior parts of the reef would fall further and further behind, resulting in a lagoon, first behind a barrier reef and then within an atoll. (Fig. 3) Since Darwin’s theory was published, in 1842, the reality of island subsidence has been confirmed by the drilling of atolls in several places, starting with a testing expedition by the Royal Society to the Pacific atoll of Funafuti, in the Ellice Islands, in 1896–1898, The drill reached a depth of 340 m, at a level near the Pleistocene–Pliocene boundary (∼2 million years old according to Ohde et al. 2002) in dolomitized reefal deposits, a finding that demonstrated rapid subsidence of the island foundations. Later drilling at Bikini and Enewetak (Eniwetok) atolls in

FIGURE 3 The stages in the evolution of an atoll, according to Darwin

(1842). (A) A subsiding island is first girt by a fringing reef; then, as subsidence continues, the island is surrounded by a barrier reef, with a lagoon formed because growth rates of “interior” corals do not keep pace with better nourished “exterior” corals. (B) Continued subsidence, with the same growth-rate difference, leads to disappearance of the island and to an atoll form.

the Marshall Islands, as part of the nuclear bomb testing program, reaffirmed the subsidence theory. The history of reef construction at Enewetak goes back some 50 million years, to Eocene times. Difficulties with Darwin’s hypothesis for the origin of the lagoon provoked others (e.g., Daly 1915) to advocate other schemes, mainly in hypothesizing planation (including dissolution) of carbonate banks during lowered sealevel periods followed by construction of a perimeter reef during the following sea-level rise, thus leaving a central lagoon. These hypotheses were disproved by drilling at Bikini, Enewetak, and Mururoa, as part of nuclear bomb testing, which showed no such erosive platform. THE DISSOLUTION, OR KARST, THEORY

The Darwinian hypothesis for the origin of the lagoon has been replaced gradually by the dissolution, or karst, theory, in which the lagoonal depression is caused by dissolution of the reef carbonate by rainwater during periodic lowstands of sea level in the Pleistocene, when the sea level dropped repeatedly by about 100 m, on a 100,000year time scale. The freshwater dissolved a hollow in the emergent carbonate bank, a hollow that was later filled with seawater when the sea level rose again. The two most important controls on lagoon depth are rainfall catchment area and average rainfall. The bigger the atoll and the rainier the climate, the deeper the lagoon. Present-day global rainfall patterns (Fig. 4) probably reasonably reflect rain patterns (wet north and south of the equator, with an equatorial dry belt in the eastern Pacific) during drier glacial times. Erosion of the atoll rim, generally by about 8 m, occurs during lowstands, but the rim is rebuilt during sea-level rise, and then still more rim carbonate is added during slow tectonic subsidence at highstand times. At Enewetak atoll, drilling shows that since Eocene times, a flattish carbonate bank built up to a thickness of about 1400 m, keeping pace with tectonic subsidence, but interrupted from time to time by periods of emergence, when soils developed. The drilling and seismic work there also showed that the atoll form itself likely did not develop until Pleistocene times, when sea-level fluctuations were large and glacial intervals long, whereas earlier sea-level fluctuations were of lesser magnitude and duration. Atolls are thus mainly a phenomenon of the Pleistocene. THE FUTURE OF ATOLLS

Atolls and their peoples are threatened today both by wave attacks on the fragile motu and by rising sea levels associated with global warming. These combine to reduce atoll relief to ephemeral patches of storm debris on reef flats awash at

AT O L L S

69

FIGURE 4 Average annual rainfall in millimeters per year for the 17-year period from 1979 to 1995. Interpolated and modified from Xie and Arkin (1997).

high tide. This is to say nothing of the threats to healthy coral growth posed by high water temperatures and possible increasing acidification of the ocean. Very high carbon dioxide levels probably occurred during the Cretaceous, a time of global warmth but also of very healthy populations of calcareous animals and plants. Whatever trends toward major acidification in the Cretaceous world existed were partly offset by dissolution of calcareous sediments on the deep seafloor. Ultimately, the same balancing may happen in the modern ocean, but this will take time. Over the shorter term, Waterworld awaits. SEE ALSO THE FOLLOWING ARTICLES

Darwin and Geologic History / Makatea Islands / Marshall Islands / Motu / Reef Ecology and Conservation

ATOMIC TESTING SEE NUCLEAR BOMB TESTING

AZORES PAULO A. V. BORGES, ISABEL R. AMORIM, AND ROSALINA GABRIEL University of the Azores, Terceira, Portugal

REGINA CUNHA, ANTÓNIO FRIAS MARTINS, LUÍS SILVA, AND ANA COSTA University of the Azores, Vairão, Portugal

VIRGÍLIO VIEIRA

70

FURTHER READING

University of the Azores, Ponta Delgada, Portugal

Daly, R. A. 1915. The glacial control-theory of coral reefs. Proceedings of the American Academy of Arts and Sciences 51: 155–251. Darwin, C. 1962. The structure and distribution of coral reefs. H. W. Menard, ed. Berkeley: University of California Press. Dickinson, W. R. 2003. Impact of mid-Holocene hydro-isostatic highstand in regional sea level on habitability of islands in Pacific Oceania. Journal of Coastal Research 19: 489–502. Purdy, E. G., and E. L. Winterer., 2001. Origin of atoll lagoons. Geological Society of America Bulletin 113: 837–854.

The Azores are a remote and geologically recent archipelago consisting of nine volcanic islands located in the North Atlantic Ocean (Fig. 1). Of the 4467 species and subspecies of terrestrial plants and animals known to inhabit this archipelago, 420 are endemics. These islands were discovered in the fifteenth century, and more than 500 years of

AZORES

human settlement have taken their toll on the local fauna and flora. Approximately 70% of the vascular plants and 58% of the arthropods found in the Azores are exotic, many of them invasive, and only 20% of the archipelago’s terrestrial realm is protected, which raises serious long-term conservation concerns for the Azorean endemic biota. GEOLOGICAL SETTING AND ENVIRONMENT

The Azorean archipelago is located in the North Atlantic Ocean, at the junction of the Eurasian, African, and North American plates (Fig. 1). The archipelago consists of nine volcanic islands, aligned on a west-northwest–east-southeast trend, which are divided into three groups: the western group of Corvo and Flores; the central group of Faial, Pico, Graciosa, São Jorge, and Terceira; and the eastern group of São Miguel and Santa Maria (Fig. 1). The largest island is São Miguel (745 km2), and the smallest is Corvo (17 km2). Santa Maria is the southern- and easternmost island (37° N, 25° W), Flores is the westernmost (31° W), and Corvo (39°42′ N) is the northernmost island. Pico has the highest elevation point (2351 m above sea level), and Graciosa the lowest (402 m above sea level). Five other islands have elevations near 1000 m above sea level. The three island groups are separated by 1000–2000-m-deep sea channels, except for Faial and Pico islands, between which the channel is, in many parts, only 20 to 50 m deep. The Azores are separated from the most western point of mainland Europe (i.e., Cabo da Roca, Portugal)

by 1390 km. Located in the Atlantic Ocean at a mean latitude of 38°30′ N, the Azores enjoy a distinctly oceanic climate. The insignificant variation in the seasonal temperature and the high humidity and precipitation that characterize the archipelago’s climate are mostly due to the influence of the Gulf Stream, which transports warm waters and humid air masses and is responsible for the high-pressure systems over the Azores. Geologically, the Azores comprise a 20–36-million-yearold volcanic plateau; the oldest rocks (composing Santa Maria Island) emerged 8.120 million years ago, whereas the youngest (forming Pico Island) are about 250,000 years old. The geostructural environment of the Azores Plateau, defined by the 2000-m bathymetric contour line, is dominated by the confluence of the American, Eurasian, and African lithospheric plates. Thus the Azores are characterized by high volcanic activity typical of a ridge-hotspot interaction (i.e., a hotspot on a slow-moving plate). As opposed to the Hawaiian islands, which are chronologically arranged, the Azorean islands do not show any correlation between their distances to the hotspot and their individual ages of emergence. The eastern parts of all Azorean islands are geologically the oldest, which is the result of the particular seismovolcanic mechanisms of this archipelago. This tectonic feature is responsible for many volcanic eruptions (e.g., Capelinhos, Faial Island, 1957–1958) and tectonic earthquakes (e.g., Terceira and São Jorge islands, 1980; Faial and Pico islands, 1998).

FIGURE 1 The location of the Azorean archipelago in the North Atlantic, and the nine islands of the Azores with estimated geological age. Shaded

areas correspond to protected island areas based on recent IUCN classification (almost 20% of the total territory of the archipelago).

AZORES

71

As a result of several recent historical lava flows, there is a great concentration of lava tube caves and pits in the Azores. A total of 250 underground cavities, including lava tubes, volcanic pits, pit-caves, and sea-erosion caves, are known to exist on the Azores, creating many kilometers of cave passages, extraordinary geological formations, and unique fauna adapted to caves. Regarding native plant communities, laurisilva,—a humid evergreen broadleaf laurel forest—was considered in the past to be the predominant vegetation form in the Azores. However, more recent studies have shown the existence of a wide variety of plant communities, including coastal vegetation, wetland vegetation (lakeshore and seashore communities and a variety of bogs), several types of meadows, and different types of native scrub and forest. Moreover, the Azorean laurisilva differs from that found on Madeira and on the Canary Islands, as it includes a single species of Lauraceae, several species of sclerophyllous and microphyllous trees and shrubs, and luxuriant bryophyte communities, covering all available substrata. In contrast to other Macaronesian archipelagoes, the Azores only has one endemic genus of vascular plants (Azorina). After human settlement, other types of vegetation cover have become progressively dominant. Presently, they include pastureland, production forest (mostly with Cryptomeria japonica), mixed woodland (dominated by nonindigenous taxa), field crops and orchards, vineyards, hedgerows, and gardens. OVERALL BIODIVERSITY: SPECIES INHABITING THE AZORES

The terrestrial flora and fauna of the archipelago were recently listed (summary in Table 1). It is believed that the Azores, especially the younger islands, are not saturated with species. The islands are probably in a nonequilibrium condition as a consequence of (1) the dispersal difficulties imposed by the isolation of the archipelago, which are much greater than the dispersal abilities of a wide range of taxa; (2) the vicissitudes of the Pleistocene environment; (3) the destructive influence of volcanic activity; and, more recently, (4) the impact of human activities. BIOGEOGRAPHY

Even factoring out the area of the islands, the native fauna and flora of the Azores is impoverished when compared to the other Macaronesian archipelagoes (Madeira and the Canary Islands). For example, the number of Azorean endemic species is about three times less than

72

AZORES

TABLE 1

Number of Currently Known Terrestrial Species and Subspecies in the Fauna and Flora of the Azores

Algicolous fungi Lichenicolous fungi Lichens Bryophyta Plantae Nematoda Annelida Mollusca Arthropoda Chordata Total

Total

Endemic

1 22 551 438 947 80 21 111 2227 69 4467

0 0 12 9 68 2 0 49 267 13 420

note: Table based on the catalog of Borges et al. (2005).

the number of endemics of the Madeira archipelago (three times smaller but older and nearer to the mainland) and ten times less than the number of endemics of the Canary Islands (three times larger). Given the isolation of the Azores, the ancestors of all the terrestrial endemic species found in the archipelago had to travel over a significant water distance (more than 1200 km) from neighboring Europe and about 800 km from Madeira Island. Additionally, colonization of the Azores has occurred over a short geological period, since the oldest island (Santa Maria) emerged 8.12 million years ago. Accordingly, it is of no surprise that the only indigenous terrestrial vertebrates are bats (two species) and birds (16 species). Nevertheless, the presence of many endemic flightless beetles in the Azores, whose ancestors are believed to have also been flightless, suggests that long-distance ocean dispersal, by air and on the water surface (rafting), must have been an important colonization mechanism. Most storms and prevailing winds come from the West, but the Azorean biota is mainly of Palearctic and Macaronesian origin. A similar situation occurs with the Azorean terrestrial molluscs, which are clearly of Palearctic origin and, at the same time, exhibit Macaronesian relationships in some taxa: Leptaxis lives also in Madeira, and Napaeus in the Canary Islands. The preferred explanation for this “anomaly” is the large distance to the American continent and the possibility that the paleo-winds may have blown in a different direction from the current prevailing winds. However, a simpler explanation for the biota composition of the Azores is the arrival of colonizers through “sandstorm” dispersal originally coming from the Sahara. Clear exceptions to the Palearctic/Macaronesian origin can only be found

in organisms with great dispersal abilities, such as bryophytes, some species of which, found in the Azores, are unequivocally of American origin. The most general pattern in ecology is the species-area relationship (SAR). Considering only the area above an altitude of 300 m (because native habitats can only be found above that elevation in almost all Azorean islands), a significant relationship is observed in the Azores for indigenous bryophytes, vascular plants, and arthropods, but not for land molluscs (Fig. 2). Arthropods show the steepest slope of the SAR curves, which implies a higher beta-diversity for this group and, consequently, a more heterogeneous species composition among the islands. However, the time factor should be accounted for in the case of endemics; this could explain the absence of SAR for land snails, as small, older islands (e.g., Santa Maria) harbor more endemic species. Some of the most diverse Azorean genera with endemic species are also diverse in Madeira and the Canary Islands (e.g., the beetles Trechus and Tarphius, the hemipteran Cixius, and the land-snails Napaeus and Plutonia), thus reinforcing the hypothesis of a Macaronesian interarchipelago dispersal. EVOLUTION

There are several aspects of the evolutionary history of the Azorean biota that are still not clear. Traditionally, many

biologists considered the Azorean and the Macaronesian endemic flora in general to be ancient, consisting of many paleoendemic species, relicts of the vegetation that originally covered most of Western Europe during the Tertiary period. However, there is increasing evidence from molecular data that many of the endemic Macaronesian plant species are the result of in situ evolution after a relatively recent colonization (neoendemics). This theory may also apply to invertebrates (e.g., arthropods and terrestrial molluscs), in which colonization followed by isolation has led to the evolution of a highly original neoendemic fauna. The most diverse genera in the Azores belong to the animal realm (classes Gastropoda and Insecta). Molecular data on insects show that many endemic species belonging to speciose genera are monophyletic in the Azores (e.g., Tarphius, Trechus, Hipparchia), thus implying that all species within a particular genus originated by speciation events occurring after the arrival of a single ancestor to the archipelago. In spite of some evidence that evolution has proceeded to a subgeneric level in some land molluscs (e.g., Macaronapaeus, Atlantoxychilus, Drouetia), natural arrival to the Azores is an uncommon event, and most of the Azorean endemics are neoendemics. Thus, dispersal limitation may be viewed as one of the main driving forces that has shaped the Azorean native biota. Available data suggest that the “progression rule”

A

B

C

D

FIGURE 2 Species-area curves for (A) indigenous bryophytes, (B) vascular plants, (C) terrestrial molluscs, and (D) arthropods (see text for further

explanations).

AZORES

73

(i.e., a nonstochastic pattern of colonization from older to younger islands) applies to the Azores: Santa Maria has generally been the first to be colonized, accompanied by subsequent lineage splitting as individuals disperse to the younger western islands. Future phylogenetic studies, aimed at understanding patterns of dispersal, colonization, and speciation in the Azores, should clarify the presence of more complex patterns within and among island speciation as well as the possibility of back colonization events from younger to older islands. The high volcanic activity in the Azores is responsible for the formation of new habitats, such as lava tubes and volcanic pits, from which 20 neoendemic troglobitic arthropod species have been described to date. These cave-limited species exhibit different levels of adaptation to the underground environment and therefore constitute an excellent opportunity to investigate ongoing evolutionary processes. Many of the cave species known to occur in the Azores belong to genera that have representatives in the troglobitic fauna of other Macaronesian islands (e.g., the Canary Islands), thus serving as a model for further studies of inter-archipelago speciation. An interesting example of an island syndrome in the Azores is given by the damselfly Ischnura hastata (Insecta, Odonata): There is no evidence of parthenogenetic populations in the New World, so the Azorean populations probably developed parthenogenesis after colonization. CONSERVATION REMARKS

The relatively high level of endemism in the Azores gives to the archipelago’s biota great conservation relevance. Its preservation has also been recognized by the local government through the establishment of protected areas for conservation purposes since the early 1980s. The Azorean Protected Areas Network is currently being reformulated according to IUCN criteria and includes 23 Sites of Community Importance and 15 Special Protected Areas, which are part of the NATURA 2000 network of nature protection areas. Most of this protected area includes the richest sites in endemic arthropods and also in rare European bryophyte species, but the area does not protect all native fauna and flora. Although expanding, unregulated tourism has not yet raised conservation concerns in the archipelago. Fragmentation and degradation of habitats together with the spread of nonindigenous species are the greatest threats to the terrestrial biodiversity in the Azores. Intentional introduction of many plant species for agriculture, forestry, and aesthetic purposes has had an enormous impact on the current flora of these islands. Many of the imported species “escaped 74

AZORES

into the wild,” and a considerable proportion have become naturalized, causing problems in agriculture and forestry. The impact of these species—in particular, invasive vascular plants, which are disrupting native plant communities with unknown consequences for overall native biodiversity—is of great concern. A negative impact on the indigenous community of phytophagous insects is expected, as well as changes in vegetation structure, difficulties in the regeneration of endemic species, and competition for dispersal agents, leading to a reduction in the frequency and abundance of indigenous plant taxa. Humans are clearly implicated in the establishment of exotic species: 70% of the vascular plants and 58% of the arthropod species and subspecies have been introduced on purpose or as stowaways. Moreover, the density of human population is correlated with the diversity of exotic taxa (vascular plants: r = 0.86; p = 0.003; arthropods: r = 0.93; p = 0.0002), and there is a remarkable correlation between the richness of exotic plant species and that of exotic arthropod species (r = 0.96; p < 0.0001). Protected areas are strategically important in order to guarantee a successful management of biodiversity conservation policy in the Azores. Progress in the conservation of Azorean biodiversity depends predominantly on longterm studies on the distribution and abundance of focal species and the control of invasive species. This research requires serious commitment from scientists, politicians, and the general public. The definition of genetic units for conservation purposes in the Azores is also extremely important, particularly for widespread endemic species. For some of those endemics that are geographically structured, part of their genetic variability is locally endangered due to threats to specific populations or to the refuge-type distribution. The conservation of the Azorean natural heritage will largely depend on the definition of a global and integrated global strategy focusing on the management of both indigenous and nonindigenous species, and paramount attention needs to be paid to the implementation of a sustainable use of the archipelago’s natural resources, including its biodiversity, in a trade-off with human activities and increasing inhabitants’ commitment to environmental values. SEE ALSO THE FOLLOWING ARTICLES

Canary Islands, Biology / Fragmentation / Lava Tubes / Madeira / Species–Area Relationship FURTHER READING

Borges, P. A. V., and V. K. Brown. 1999. Effect of island geological age on the arthropod species richness of Azorean pastures. Biological Journal of the Linnean Society 66: 373–410.

Borges, P. A. V., R. Cunha, R. Gabriel, A. F. Martins, L. Silva, and V. Vieira., eds. 2005. A list of the terrestrial fauna (Mollusca and Arthropoda) and flora (Bryophyta, Pteridophyta and Spermatophyta) from the Azores. Direcção Regional do Ambiente and Universidade dos Açores, Horta, Angra do Heroísmo and Ponta Delgada. Cameron, R. A. D., R. M. T. da Cunha, and A. M. Frias Martins. 2007. Chance and necessity: land snail faunas of São Miguel, Açores,

compared with those of Madeira. Journal of Molluscan Studies, 73: 11–21. Gabriel, R., and J. W. Bates. 2005. Bryophyte community composition and habitat specificity in the natural forests of Terceira, Azores. Plant Ecology 177: 125–144. Silva, L., and C. W. Smith. 2004. A characterization of the non-indigenous flora of the Azores Archipelago. Biological Invasions 6: 193–204.

AZORES

75

B was known previously to the Norse as Helluland, the land of flat stones.

BAFFIN

HISTORY LYNDA DREDGE

Early settlement of the island dates back about 4000 years to the Paleoeskimo Pre-Dorset people, and their successors, the Dorset people. These cultural groups are named after key sites at Cape Dorset, along the south coast of Baffin Island. Both cultures were based on a maritime economy dominated by the hunting of sea mammals, especially walrus, narwhal, and beluga. The Dorset culture is particularly renowned for superb miniature ivory carvings of sea animals, polar bears, humans, and magical beings. About 1000 years ago, as the climate warmed, the Thule people, the ancestors of the present Inuit, migrated eastward into the region, and for the first time dog teams were used to pull sleds. Although seals and bowhead

Geological Survey of Canada, Ottawa

Baffin Island, with an area of more than 500,000 km2, is one of the principal islands of the Canadian Arctic archipelago and the world’s fifth-largest island (Fig. 1). It lies within the territory of Nunavut, at the eastern portal of the Northwest Passage, the grail of early explorers and a potential shipping route in years to come. Half the island’s 11,000 inhabitants live in Iqaluit, the administrative capital for Nunavut; the remainder live in seven other coastal communities. The island is named after William Baffin, a seventeenth-century British explorer, although it 80° Northwest Passage

Cenozoic

Baffin Island

Tertiary basalts

Paleozoic Carbonate platforms

Paleoproterozoic

Baffin Bay

Cover sequences

Trans-Hudson Orogen Continental magmatic arcs Island arcs / oceanic crust

Penny Ice Cap

Barnes Ice Cap

Continental shelf / foredeep prism

68°

Archean um be r

Cratons

C

Nettilling Lake

lan dS

0

d.

km

300

Iqaluit C. Dorset Hu d son S trait 80°

76

Fro bish

er Bay

FIGURE 1 Map showing location (inset)

and geology of Baffin Island.

whales were the main food sources of the Thule people, all early peoples of Baffin Island had a diversified diet that included sea mammals, caribou, fish, and birds’ eggs. The first definitive European exploration began in 1576 when Martin Frobisher, an English privateer-turnedexplorer, sailed into what is now known as Frobisher Bay. There, he found “black ore,” which he promoted as gold-bearing. He brought back more than 1200 tons to England in three mining expeditions backed by investors, but all the ore was hornblende and pyrite—fool’s gold. Frobisher thus instigated the first mining scam in the New World. Today, the island’s main inhabitants are the Inuit, and their economy is based on government administration, hunting and fishing, mineral resources, and tourism, as well as the carving and printmaking first established at Cape Dorset. GEOLOGY AND LANDFORMS

Baffin Island is part of the Canadian Shield. The oldest rocks are Archean-age (2.8 billion years ago) granite, gneiss, and metamorphosed volcanic and sedimentary rocks that form part of the Rae Craton, a proto-continent. A large fold belt crossing the central and southern parts of the island consists of a continental margin succession of quartzite, marble, shale, and turbidite wackes of Proterozoic age (2.16–1.90 billion years ago), deposited on the flank of the craton. Associated volcanic rocks resulted from eruptions about 1.93 billion years ago. The Precambrian rocks were deformed by plate convergence and continental accretion during the Trans-Hudson Orogeny 1.8 billion years ago, a major event in the welding together of the Canadian Shield. Later, in the Paleozoic era (450 million years ago), calcareous marine sediments were laid down over the craton. The flat-lying strata of dolostone and limestone that form surface rocks on western parts of the island are remnants of this event. Gold, iron ore, and sapphires from the Precambrian rocks, leadzinc deposits in the Paleozoic carbonates, and diamonds in younger kimberlite intrusions are the main mineral resources on the island. In the Cenozoic era, passive-margin plate tectonics associated with the opening of the Atlantic Ocean and the drifting of Greenland caused rifting and uplift along the northern and eastern edge of Baffin Island and shaped the island into what we know today—a plateau tilted downward toward the southwest. The eastern side of the island rises 2100 m out of the waters of Baffin Bay, creating an icecapped mountainous edge that is deeply dissected by fjords and sounds, some of which penetrate more that 100 km

FIGURE 2 The northern coast. Rock-walled Royal Society Fjord, with

snowfields on the highlands and small outlet glaciers descending to tidewater. Photograph GSC 2007-193 by D. A. Hodgson.

inland (Fig. 2). The land slopes gently across a central plateau to the relatively shallow waters of Foxe Basin. In the last 2 million years, Baffin’s landforms have been modified by the great ice sheets that covered substantial parts of the northern hemisphere. During the last glaciation, culminating 20,000 years ago, Laurentide ice from the Canadian mainland crossed to the outermost edges of Baffin Island, scraping and polishing rocks, deepening fjords, and depositing a layer of till (a mixture of glacially eroded debris). Ice began to melt away from the northeast highlands about 8600 years ago and from Foxe Basin about 7800 years ago, shifting one of the glacial dispersal centers onto central Baffin Island. The 700-m thick Barnes Ice Cap (Fig. 3) on the central plateau and the Penny Ice Cap are the last remaining remnants of the once-vast continental ice sheet. The numerous ice fields and small outlet glaciers in the mountains covering the northern island rim are more recent and have formed in

FIGURE 3 The edge of the Barnes Ice Cap, ice-marginal moraine

ridges, proglacial meltwater streams, and nearby tundra barrens. Photograph GSC 2007-195 by L. A. Dredge.

BAFFIN

77

the postglacial period. As the ice sheet melted, marine waters inundated the lowlands along Foxe Basin, depositing sand and mud. With glacioisostatic rebound, these deposits now lie between present sea level and an elevation of about 100 m and form the grassy coastal lowlands that are major wildlife habitat.

open waters, and walruses, polar bears, and ringed and bearded seals live on the pack ice that surrounds the island for much of the year.

FLORA AND FAUNA

FURTHER READING

Most of Baffin Island lies above the Arctic Circle and is thus subjected to midnight sun in summer and polar night during the winter. It experiences a maritime arctic to continental-arctic climate. Mean annual temperatures average about –15 °C; precipitation ranges from 400 mm on the eastern lowlands to less than 200 mm on the upland plateau, which forms a polar desert. Below a shallow layer of soil that thaws seasonally, the ground remains frozen to a depth of about 400 m. Most of the land is rocky and supports tundra-barrens vegetation, mainly a cover of lichen-heath (Fig. 4). Foxes, wolves, hares, and lemmings are found in this habitat. However, extensive areas of the plateau above an elevation of 550 m were covered with persistent snowfields during the Little Ice Age, between AD 1600 and 1850. These areas remain almost bare of any vegetation, including lichens, and are devoid of wildlife. In contrast, the lowlands around Foxe Basin consist of sedge and grass meadows containing dwarf birch and willow shrubs, interspersed with shallow tundra ponds. These areas are prime habitats for barren-ground caribou and a myriad of migratory birds, including various species of geese, sandpipers, murres, plovers, and gulls. The lowland around Nettilling Lake, known as the Great Plain of the Koudjuak, lies along the Eastern Flyway and is a major nesting area for Canada and brant geese. Whales, particularly the bowhead, beluga, and narwhal, are prevalent in

Andrews, J. T. 1989. Quaternary geology of the northeastern Canadian Shield. Geological Survey of Canada, Geology of Canada 1: 276–318. Bird, J. B. 1967. The physiography of Arctic Canada. Baltimore, MD: Johns Hopkins Press. Dredge, L. A. 2002. Surficial materials, central Baffin Island. Geological Survey of Canada, Current Research 2002-C20. McGhee, R. 1996. Ancient people of the Arctic. Vancouver: University of British Columbia Press. Porsild, A. E. 1964. Illustrated flora of the Canadian Arctic Archepelago. National Museum of Canada Bulletin 146. Scott, D. J., M. St-Onge, and D. Corrigan. 2003. Geology of the Archaean Rae Craton and Mary River Group and the Paleoproterozoic Piling Group, central Baffin Island, Nunavut. Geological Survey of Canada, Current Research 2003-C26.

FIGURE 4 Lichen-covered, glaciated, granitic rocks and tundra bar-

rens of central Baffin Island. Stacked boulders are left where rocks transported within glaciers have been gently deposited as the ice melts. Photograph GSC 2007-196 by L. A. Dredge.

78

BAJA CALIFORNIA: OFFSHORE ISLANDS

SEE ALSO THE FOLLOWING ARTICLES

Arctic Region / Climate Change / Sea-Level Change

BAJA CALIFORNIA: OFFSHORE ISLANDS MARTIN L. CODY University of California, Los Angeles

The Baja (lower) California Peninsula, in extreme northwest Mexico, stretches more than 1000 km south from the southern (Alta) California border to the Tropic of Cancer, spanning latitudes of 32°30′ to 23° N. Originating about 500 km south of its present position and rifting north some 4 cm per year since the Miocene, the peninsula was sheared from the North American plate by contact with the northwesterly moving Pacific plate. The East Pacific Rise runs up the center of the Sea of Cortés (or Gulf of California), which separates the peninsula from mainland Mexico; it continues north as the classic strike-slip San Andreas Fault. Resulting largely from complex plateboundary dynamics involving, besides those mentioned, the fragmented remnants of the largely subducted Farallon plate, islands have been generated around the peninsula, west and south in the Pacific Ocean and east in the Sea of Cortés off the peninsula’s trailing edge. Many islands had Pleistocene connections to the mainland; others are deepwater islands, some being blocks faulted from the trailing

edge of the peninsula, variously tilted and uplifted, and others being the products of seafloor spreading at submarine fracture zones. The larger and more interesting of these islands are discussed here. ISLANDS TO THE WEST OF THE PENINSULA Isla Guadalupe

Isla Guadalupe is an oceanic island at latitude 29° N, with an area of 244 km2 and an elevation of 1295 m, and distant some 259 km from the nearest point on the Baja California Peninsula. It is the emergent tip of submarine shield volcanoes of plate-boundary origin and Miocene age. Guadalupe had evolved an extremely interesting flora and vertebrate fauna (birds only; no mammals reached the island, and the only herpetological record is an old one of treefrog tadpoles with no further substantiation). At least 20% of the native biota is endemic at genus (e.g., Baeropsis, Asteraceae; Hesperelaea, Oleaceae), species (e.g., Caracara Polyborus lutosa, Falconidae), or subspecies level; snails and insects may be similarly distinct. The endemic shrubby tarweeds Deinandra (first studied by Sherwin Carlquist) have radiated from ancestral annuals in a similar fashion to the classical tarweed of the Hawaiian islands. Mere remnants of the biota remained into the twentieth century, and fewer still survive to the present. Around two centuries of residence by goats almost completely denuded the island, such that many endemics (and dependent fauna) became extinct (26 of 154 native species) and much of the remainder nearly so. To add to the debacle, goatherds deliberately hounded to extinction the endemic caracara (perceived as a goat predator) by around 1901. Just three of the original nine endemic bird taxa are still extant: the rock wren Salpinctes obsoletus guadalupensis, the house finch Carpodacus mexicanus amplus, and the junco Junco insularis. A recently instigated (2003) goat removal program was successful, and it promises a limited rebound among the plants (e.g., with recent rediscoveries of taxa thought extinct, and renewed regeneration in the endemic trees Pinus radiata binata, Cupressus guadalupensis). The island is a now a biosphere reserve receiving international attention and a pinniped sanctuary (with hopeful great white sharks Carcharodon megalodon in attendance—another tourist attraction!); extant marine mammals include seals, sea lions, elephant seals, and the one remaining (relictual) breeding colony of Guadalupe fur seal Arctocephalus townsendi. Isla de Cedros

At latitude 28° N, Isla de Cedros is the largest island west of Baja California (360 km2, elevation 1200 m), 19 km

northwest of the tip of the Vizcaino Peninsula, to which it was attached during late Pleistocene times when sea levels were lower. It supports a diverse flora of conspicuous (relictual) northern affinities, including pine forest (Pinus radiata) at higher elevations, California juniper (Juniperus californicus), and various chaparral shrub taxa (e.g., Arctostaphylos) on lower slopes. Around a dozen plant species are endemic at the species level; there are also endemic species of reptiles (horned-lizard Phrynosoma, alligatorlizard Elgaria, rattlesnake Crotalus) and mammals (pocket mouse Chaetodipus, packrat Neotoma) as well as a plethora of endemic subspecies (especially in other taxa indicative of landbridge status, such as deer and cottontails). The island’s birds, unsurprisingly, are not distinct. There is a substantial human presence on the island, with many of the 10,000 souls occupied with inshore fishing, shellfish exploitation, and canneries; recent archaeological excavation shows basically similar occupations in sizable human populations that thrived there 11,000 years ago. Among the usual wide variety of introduced mammals, feral cats and dogs pose the most serious conservation threats. ISLANDS TO THE SOUTH OF THE PENINSULA Las Tres Marías

The Tres Marías (Maria Madre, Maria Magdalena, and Maria Cleofas, with the much smaller San Juanito lying to the north) are an island trio at 21°30′ N, with a total area of 245 km2 and a maximum elevation of 615 m. They are situated 100 km off Nayarit, western Mexico, and around 350 km southeast of Cabo San Lucas at the tip of the Baja California Peninsula; before rising sea levels partitioned them, they were likely parts of a single large (80 km2) island. They are underlain mostly by uplifted Miocene volcanics of plate-boundary origin, although shallow mainland-island channel depth assures landbridge status. There is extensive and varied habitat on the islands, from tall tropical dry forest and shorter deciduous woodland to low thornscrub, comparable to vegetation on the adjacent mainland (e.g., south of Puerto Vallarta, Jalisco, Mexico). Some 40 breeding bird species have been recorded, 16 represented by endemic subspecies. All populations are thought to be quite healthy except that of the bluerumped parrotlet Forpus cyanopygius insularis, which may be close to extinction. There is an endemic raccoon (Procyon insularis), cottontail Sylvilagus graysoni, bat Myotis findleyi, and mouse Peromyscus madrensis, but the endemic rice rat Oryzomys nelsoni is now thought extinct, replaced apparently by ubiquitous Rattus rattus. Around a thousand people are resident, most on Maria Madre and involved with a penal colony. Inmates cultivate agave to

BAJA CALIFORNIA: OFFSHORE ISLANDS

79

supply their henniquen mill, an operation that is an obvious threat to the persistence of the native biota. Las Islas Revillagigedo

Some 480 km just west of south from Cabo San Lucas is Isla Socorro (132 km2; elevation 1130 m), largest of the oceanic Islas Revillagigedo archipelago and around 600 km from the closest coast in Colima. It was discovered in 1533 by the Spanish explorer Hernando de Grijalva and named after the 53rd viceroy of New Spain. Of the four islands between latitudes 18° and 19°20′ N, the second largest, Clarión (20 km2; elevation 335 m), lies a further 370 km to the west. Their volcanic origins along the Clarión Fracture Zone were highlighted in August 1952 with the birth of Volcan Bárcena on the third-largest and more northerly Isla San Benedicto (6 km2); in the ensuing six months, 300 million m3 of fresh tephra and lava added both height and area to this islet. Clarión’s native vegetation is composed mostly of grasses and scrubby cacti, but on Socorro endemic trees of Bumelia, Ilex, Guettarda, and Psidium form low and dense woodlands at higher elevations. The native reptiles (other than marine sea turtles) are an endemic snake Masticophis anthonyi on the older Clarión and two endemic brush lizards, Urosaurus clarionensis on Clarión and U. auriculatus on Socorro. These brush lizards are not closely related, apparently representing independent invasions from the mainland. Among the ten species of nesting seabirds, one (Townsend’s shearwater, Puffinus auricularis) is endemic, persisting on Socorro but thought to be near extinction on Clarión, where feral pigs excavate its burrows. Fifteen species of landbirds are all endemic at generic, species, or subspecies levels, and none are shared among the three larger islands. San Benedicto lost its endemic rock wren Salpinctes obsoletus exsul after the volcanic eruption there. Two nonendemic landbirds, both on Socorro, appear to be playing out a recurrent island biogeographic theme: Repeated invasion of an island by the same mainland taxon causes the demise of the resident endemic relative and drives a “taxon cycle.” Northern mockingbird Mimus polyglottos reached Soccoro sometime between the 1950s and 1978; it has become common since, during which time the endemic Mimoides graysoni, the product of an earlier mockingbird invasion, has become rare and restricted. During the same pre-1978 period, mourning dove Zenaida macroura reached Socorro, where earlier colonization had produced an endemic species Zenaida graysoni. Mourning doves have become abundant on Socorro, whereas the endemic species has declined precipitously and may now be extinct. These might be considered natural

80

BAJA CALIFORNIA: OFFSHORE ISLANDS

extinctions—inevitable faunal turnover—but casual introductions of herbivores and predators via the small naval garrisons on Socorro and Clarión (house mice, cats, and sheep, with pigs and rabbits on Clarión as well) likely accelerate the process. ISLANDS TO THE EAST OF THE PENINSULA

Islands in the Sea of Cortés have received much attention over the last several decades from biogeographers and ecologists, as well as conservationists, ecotourists, and politicians. Travel to the region is now well served by land, sea, and air, and the spectacular natural beauty and biological wonders of the area are readily available for study, admiration, or exploitation. The two largest islands, both in the northern gulf between 28°30′ N and 29°30′ N, are Ángel de la Guarda (west) and Tiburón (east). These and the several islands nearby (Partida Norte, Rasa, Salsipuedes, San Lorenzo, San Esteban) constitute the “Midriff,” and biotic affinities change sharply from east to west, being mainland Sonoran on Tiburón, intermediate on San Esteban, and peninsular on Ángel and the remaining islands. North of this area the gulf is shallow, filled with sediments from the Colorado River and bordered by broad alluvial fans. From Ángel (area 936 km2) south to San Lorenzo, the islands are faulted blocks separated from the peninsula by channels of modest width (12–20 km), but with troughs and basins reaching depths of up to 1400 m, some of which are active spreading centers. Tiny Rasa (0.68 km2) is volcanic and formed of basalts derived from seafloor spreading; centrally located San Esteban is comprised of Miocene volcanics with a similar genesis. Across by the Sonoran coast, Tiburón is both very large (1224 km2) and a landbridge island with prior mainland connections; it has essentially a mainland flora (298 species) and fauna (reptiles: 29 species; mammals: 14 species; and landbirds: 34 species; totals far richer than those of other Sea of Cortés islands). From latitude 27° N south to 24° N and closely adjacent to the peninsula is a series of landbridge islands: San Marcos, Coronados, Carmen, Danzante, Santa Cruz, San Diego, San José, and San Francisco, with Monserrat of questionable classification. The troughs and basins, with depths to 2400 m, here lie to the east in the center of the gulf. Northeast of the city and bay of La Paz are Espíritu Santo and Partida Sur, nearly united and likewise landbridge islands across a channel 6 km across and 12 m deep from the peninsula. Carmen, San José, and Espíritu Santo/Partida Sur are the largest of these islands, at 143, 187, and 107 km2, respectively; almost all are faulted and variously uplifted blocks.

Several remaining islands are deepwater islands and are more interesting because of their increased isolation over time and space. South of the Midriff, Isla San Pedro Mártir (2.9 km2) in the central gulf is a towering (320 m) basement rock, a biosphere reserve, and a celebrated breeding site for seabirds and marine mammals. Further south Isla Tortuga (latitude 27°26′ N, 11.4 km2) sits in water 1200 m deep and, like Rasa, is a product of recent (Holocene) volcanic activity. At latitude 25°39′ N, Santa Catalina is another block-faulted island, 41 km2, which in Pleistocene times lay just a few kilometers east from the now-submerged peninsular shelf. Lastly, the southernmost island of Cerralvo is old (Pliocene), large (140 km2), and high (767 m), with no record of past peninsular or mainland connections. It is thus an island on which, a priori, interesting biology might be expected, with expectations fulfilled by numerous endemic plant, lizard, snake, and mammal species, and even a locally endemic lizard genus (Sator; Iguanidae). The prevalent vegetation throughout the Sea of Cortés region is Sonoran Desert scrub, open and xeric with a preponderance of stem-succulent cacti, droughtdeciduous shrubs, and short, phreatophytic trees (Fig. 1). Some 80–150 mm of annual precipitation sustain life here; most is winter rainfall except in the south (e.g., near Cerralvo), where summer precipitation increases, and a taller, denser thorn-scrub prevails. With many recent (late Pleistocene) land bridge connections, similar conditions island to mainland, and modest isolation distances, there is very little endemism in the more vagile taxa, namely plants and birds. About 650 plant species are found on the islands; 17 of them endemic, 8 of these cacti (Echinocereus,

FIGURE 1 Sonoran Desert scrub vegetation is typical of most of the

islands in the Sea of Cortés. Note the prevalence of cacti, such as the giant tree-like cardon cacti (Pachycereus pringlei), the sprawling Stenocereus gummosus in the left foreground, and the bushy cholla (Opuntia cholla) behind it.

Ferocactus, Mammillaria), and are arguably no more than single-island growth form variants. Of more interest are plants whose present island occurrences reflect historical legacies. When Mojave Desert vegetation moved south onto the Baja California Peninsula in cooler and wetter Pleistocene climates, several typical Mojave Desert taxa colonized Ángel de la Guarda: for example, the phreatophyte Acacia greggii, the shrub Gutierrezia microcephala, the composite forb Trichoptilum incisum, and the annual Plagiobothrys jonesii. When peninsular relatives retreated north with the onset of warmer, drier times, insular populations remained marooned on that island. There are, unremarkably, only a few weakly defined subspecies in the island avifauna, composed of typical Sonoran Desert birds that are island-area dependent in terms of species richness and even species identity. The thrasher genus Toxostoma has allopatric ecological counterparts across the gulf, the curve-billed thrasher T. curvirostre to the east and the gray thrasher T. cinereum to the west. Each occupies several gulf islands but no islands are co-occupied. Several seabird species are near-endemic to the gulf in their breeding ranges; one such is Heermann’s gull Larus heermanni, of which 200,000 congregate on Rasa yearly and defend nesting territories a meter or so across. Following El Niño winters, food availability declines with the higher sea surface temperatures, females lose body weight, and chick survival drops to near zero; similar trends are recorded in other breeding seabirds such as elegant terns, Sterna elegans, and brown pelicans, Pelecanus occidentalis. Birds are regarded as good colonists because they fly, but their high metabolic rates and short longevities translate to poor persistence. Mammals also are poor persisters for the same reason, but in addition, they are poor colonists (excepting bats, they do not fly). On the gulf islands mammals are scarce overall, and many species are found only on land bridge islands. This is illustrated by taxa such as shrews (Notiosorex on Tiburón), jackrabbits (Lepus on Carmen, endemic species on Espíritu Santo), cottontails (endemic Sylvilagus species on San José), coyote Canis latrans, ringtail Bassariscus astutus, and mule deer Odontocoileus hemionus on San José as well as Tiburón, and many others. The packrat Neotoma lepida reached one non–land bridge island (Ángel), but the only taxon with a respectable colonization record throughout the gulf islands is the mouse Peromyscus, with ten endemic species and numerous subspecies described. However, most taxa are singleisland records, and all islands except Tiburón have but a single taxon. Thus, there has been a substantial evolution of island forms, but no adaptive radiation.

BAJA CALIFORNIA: OFFSHORE ISLANDS

81

From an island biogeography viewpoint, the most interesting animals on Sea of Cortés islands are reptiles. They colonize infrequently but persist well, even on small, unproductive islands; thus, they readily form endemics (one-fourth to one-third are endemic at the species level). The endemic genus Sator occurs on three islands: S. grandaevus on Cerralvo and S. angustus on adjacent San Diego and Santa Cruz. The gecko Phyllodactylus, the sideblotched lizard Uta (with six described endemic species), and several snake genera have demonstrated an ability to colonize islands over water. The Sonoran (mainland) western diamondback rattlesnake has reached San Pedro Mártir, where it is an island dwarf, and also Tortuga, where it constitutes an endemic species Crotalus tortugensis. The endemic San Esteban rattlesnake likewise is a dwarf and also is derived from a Sonoran species, the black-tailed C. molossus. All the preceding are one-snake islands, but on Ángel there are two rattlesnake species. Peninsular red rattlesnake C. ruber is present but as a dwarf, much smaller than peninsular specimens. On the other hand, the speckled rattlesnake derivative there, from peninsular C. mitchelli, is a veritable giant. The two have reversed positions on the island, relative to the peninsula, in a size segregation sequence. Dramatic body size shifts are evidenced in island chuckawallas Sauromalus. Whereas peninsular and mainland chuckawallas S. obesus average 300 g in mass, the endemic island derivative S. hispidus on Ángel de la Guarda and neighbors is a 1400-g giant, and another endemic, S. varius on San Esteban, is truly huge (1800 g). These herbivorous lizards reproduce successfully only in El Niño years, when higher rainfall produces more edible plant material (i.e., a pattern opposite that of seabirds). The island forms show rapid growth rates and occur in high densities; they are thought to evolve gigantism in response to the absence of large predators on islands, and possibly also to a general lack of herbivorous competitors. One historic predator was the Seri Indians, who visited at least the Midriff islands on foraging trips. The giant chuckawallas must have made attractive and easy prey, so much so that the Indians are known to have transplanted the lizards to chuckawalla-free islands to make living food caches, just as the old oceangoing mariners did with goats. SEE ALSO THE FOLLOWING ARTICLES

Gigantism / Pigs and Goats / Snakes / Taxon Cycle / Vegetation FURTHER READING

Axelrod, D. 1979. Age and origin of Sonoran Desert vegetation. Occasional Papers of the California Academy of Sciences 132: 1–74.

82

BARRIER ISLANDS

Baldwin, B. A. 2007. Adaptive radiation of shrubby tarweeds (Deinandra) in the California Islands parallels diversification of the Hawaiian silversword alliance (Compositae-Madiinae). American Journal of Botany 94: 237–248. Brattstrom, B. H. 1990. Biogeography of the Islas Revillagigedo, Mexico. Journal of Biogeography 17: 177–183. Case, T. J., M. L. Cody, and E. Ezcurra. 2002. A new island biogeography of the Sea of Cortés. Oxford: Oxford University Press. De la Luz, J. L. L., J. P. Rebman, and T. Oberbauer. 2003. On the urgency of conservation on Guadalupe Island, Mexico: is it a lost paradise? Biodiversity and Conservation 12: 1073–1082. Des Lauriers, M. R. 2006. Terminal Pleistocene and early Holocene occupations of Isla de Cedros, Baja California, Mexico. Journal of Island and Coastal Archaeology, 1: 255–270. Jehl, J. R., and K. C. Parkes. 1982. The status of the avifauna of the Revillagigedo Islands, Mexico. Wilson Bulletin 94: 1–19. Mellinck, E. 1993. Biological conservation of Isla de Cedros, Baja California, México: assessing multiple threats. Biodiversity & Conservation 2: 62–69. Shreve, F., and I. L. Wiggins. 1964. Vegetation and flora of the Sonoran Desert. Stanford, CA: Stanford University Press. Stager, K. 1957. The avifauna of the Tres Marias Islands, Mexico. Auk 74: 413–432.

BALEARIC ISLANDS SEE MEDITERRANEAN REGION

BARRIER ISLANDS MILES O. HAYES Research Planning, Inc., Columbia, South Carolina

Barrier islands are elongate, shore-parallel accumulations of unconsolidated sediment, some parts of which are situated above the high-tide line (supratidal) most of the time, except during major storms (see example in Fig. 1). They are separated from the mainland by bays, lagoons, estuaries, or wetland complexes and are typically intersected by deep tidal channels called tidal inlets. A large percentage of the major barrier islands of the world occur along the coastlines of the trailing edges of continental plates and of epicontinental seas and lakes (e.g., Caspian and Black Seas). Because they are composed of unconsolidated sediments (primarily sand, with gravel being present in some Arctic regions), they most commonly occur on coastal plain and deltaic shorelines (depositional coasts), where the sediment that makes up the islands was ultimately brought to the shore by rivers and streams. Some barrier islands do occur, primarily as spit forms, on leading edges of continental plates and on some glaciated coasts, but they are a minority coastline type in those areas.

FIGURE 1 Oblique view with infrared film looking southwest at Kiawah

Island, South Carolina, which clearly illustrates the drumstick-like configuration of the 13.7-km-long, prograding barrier island. Note the presence of linear ridges of sand vegetated by maritime-forest, which indicate the positions of the backbeach foredunes at earlier stages in the growth of the island. Photograph taken by Dennis K. Hubbard on March 18, 1976. Since that time, this island has been developed for residences and golf courses.

OCCURRENCE AND IMPORTANCE

The largest single chain of barrier islands in the world occurs along the East and Gulf Coasts of North America. Many of the largest of these North American barrier islands are highly developed with human habitation, some with moderate-sized cities, such as Galveston, Texas, and Atlantic City, New Jersey. Because of the dynamic nature of the coastal zone, beach erosion is a serious problem for some of the more developed barrier islands, invoking major engineering efforts to stem the erosion. One commonly applied technique is the addition of massive volumes of sand derived from elsewhere along the eroding shoreline, a process called beach nourishment. The largest of these beach nourishment projects commonly cost many millions of dollars. Some of the islands have been preserved as national and state parks, to which vacationers flock in season. Consequently, barrier islands have a high socioeconomic profile in North America, especially for the large percentage of the population that lives in coastal areas. Barrier islands also occur on the shorelines of northwestern Europe, the Mediterranean and Caspian Seas, West Africa, and elsewhere.

or T. R. = 0–2 m), mesotidal (T.R. = 2–4 m), and macrotidal (T.R. = > 4 m). As a generalization, depositional features on microtidal coasts are highly influenced by waves (wave-dominated coasts), whereas those on macrotidal coats are highly influenced by tides (tide-dominated coasts) and those on mesotidal coasts respond to the effects of both waves and tides (mixed-energy coasts). For example, barrier islands do not occur on open-ocean, coastal-plain shorelines with tidal ranges greater than about 4 meters (macrotidal coasts). This is because their primary mechanism of formation, wave action, is not focused long enough at a single level during the tidal cycle to form the island. Furthermore, the strong tidal currents associated with such large tides transport the available sediments to the offshore regions. Barrier islands exposed to open-ocean waves and tides that are in a progradational mode (i.e., consistently building in an offshore direction) show major differences depending on whether the tides are microtidal or mesotidal. Prograding barrier islands along microtidal shorelines are long and linear, commonly over 25–50 km in length, with a predominance of storm washover features. Those on mesotidal shorelines are stunted, usually less than 16 km in length, with an abundance of large tidal inlets. More tidal inlets are required on mesotidal coasts, because of the large amount of water that moves into and out of the backbarrier regions during a single tidal cycle. During major storms with significant storm surges, microtidal barrier islands are usually washed over and permanent washover fans are formed (e.g., those on the Texas coast), whereas on mesotidal barrier islands, permanent washover fans are not so common, because the system is already adjusted to major influx and outflow of ocean waters during normal spring tides (e.g., those on the Georgia and South Carolina coasts). The long-term patterns of morphology and sedimentation on most coastal plain and deltaic shorelines have been significantly impacted by the major changes in sea level that occurred during the glacial episodes of the Pleistocene Epoch. During each major glaciation, sea level was lowered significantly, over 100 m during the last glaciation (Wisconsin). When the sea level was lower, major valleys, called lowstand valleys, were carved across the coastal plains and continental shelves. As sea level rose, the valleys were flooded to become major estuaries.

ROLE OF TIDES, WAVES, AND SEA LEVEL

Depositional coasts have characteristic morphology and sediment distribution patterns controlled by the interaction of waves and tides, with the magnitude of the tides being of particular importance. Accordingly, depositional coasts are commonly classified as microtidal (tidal range,

MORPHOLOGY AND STRATIGRAPHY

A major consideration with regard to the morphology and stratigraphy of barrier islands is whether they consistently migrate landward (transgressive) or build in an offshore direction (prograding). The general patterns of

BARRIER ISLANDS

83

FIGURE 2 View of the southwest end of Cape Romain, South Carolina

a few days after the passage of Hurricane Hugo (1989). The white band of sediment, a mixture of sand and shell, is a washover terrace that advanced landward some 10 m during the storm. The dark layer seaward of the washover terrace is exposed, muddy backbarrier sediments. As a result of the landward migration of this transgressive barrier island, a new tidal inlet was created where it intersected the large tidal channel in the foreground.

prograding barrier islands with respect to the effect of tidal range were discussed in the previous section of this article. However, both types of barrier islands, transgressive and prograding, may be present on either microtidal or mesotidal coastlines, depending upon the rate of sealevel change relative to sediment (usually sand) supply at that location (not the tidal range). Diminished, or low, sand supply and rapid sea-level rise both promote the development of landward-migrating islands, and vice versa for those that build in an offshore direction. Both prograding and transgressive barrier islands clearly tend to change to some extent over time, with the rates of landward migration and offshore growth being different from place to place. In some parts of the South Carolina coast, for example, the landward-migrating barrier islands may move 3 m or more per year (example in Fig. 2). However, those that build seaward usually grow more slowly, at rates of < 3 m per decade (example in Fig. 1).

FIGURE 3 Morphology and subsurface three-dimensional configuration (stratigraphy) of prograding and transgressive barrier islands.

84

BARRIER ISLANDS

As illustrated in Figs. 2 and 3, transgressive (landwardmigrating) barrier islands are composed of coalescing washover fans, or a washover terrace, that is overtopped at high tides, usually several times a year. In the process of migration, the entire washover terrace complex moves landward, leaving an eroded nearshore zone in its wake. As a result of this type of migration, in three dimensions the entire complex consists of a relatively thin (< 1–3 m) wedge of sand and shell of the washover terrace, which overlies muddy sediment originally deposited in the lagoons or wetlands landward of the islands (see cross section in Fig. 3). Because of their continual landward migration, these types of islands are, needless to say, impractical sites for human development. The transgressive (landward-migrating) barrier islands in South Carolina are relatively short, 2–8 km on the average, because new inlets are created where the migrating islands intersect tidal channels (see Fig. 2). Prograding (seaward-building) barrier islands (Figs. 1 and 3) are typically composed of multiple, relatively parallel linear ridges of sand topped by vegetated sand dunes that originally formed as front-line dunes on the backbeach (called foredunes). The most notable changes on these types of islands occur where adjacent tidal inlets migrate into them or when the inlets expand dramatically during hurricane storm surges. As a result of their offshore growth, in three dimensions these types of barrier islands typically consist of a wedge of sand 7–9 m thick that has built over offshore muds (see cross section in Fig. 3). Most of the major developed barrier islands along the east coast of the United States, which typically are greater than 16 km long, are of this type (e.g., Kiawah Island and Hilton Head Island, South Carolina). When human development occurs on these types of islands, buildings are usually secure from all but the most extreme hurricanes if they have been set back an adequate distance from the front-line dunes and tidal inlets. That security will vanish, however, if a major rise of sea level occurs in the near future as a result of global warming. The morphology of the prograding barrier islands longer than about 11 km takes on a characteristic drumstick appearance, as shown in Fig. 4. This pattern is most common on mesotidal shorelines. Two factors that enhance the development of the drumstick shape are: 1. The occurrence of significant masses of sand in the form of large, wave-built intertidal sand bars (swash bars) that develop along the outer margin of a large lobe of sand deposited on the seaward side of the tidal

FIGURE 4 Barrier island drumstick model, primarily the result of weld-

ing of masses of sand derived from the ebb-tidal delta in the form of large swash bars. A sediment transport reversal resulting from the refraction of the dominant waves around the ebb-tidal delta is also a factor. These types of barrier islands are most common on prograding, mesotidal (tidal range = 2–4 m) shorelines.

inlet by ebb-tidal currents (called the ebb-tidal delta). These huge swash bars eventually move toward shore and weld to the beach (Fig. 4). This welding process builds out the end of the island that faces the direction from which the sediments come, accentuating its drumstick shape (see photographs in Figs. 1 and 5). 2. Refraction of the dominant waves around the ebbtidal delta, a process that enhances deposition on the same end of the island where the huge swash bars come ashore. The refracted waves create currents that transport sediment in the opposite direction from that on the open coast. This is a relatively minor reversal from the normal longshore transport direction (see model in Fig. 4), but it allows sand to remain in the inlet area and aids in its accumulation on that end of the island. ROLE OF CLIMATE

Climate plays a significant role in the nature of barrier islands, not so much regarding their morphology and stratigraphy, as was demonstrated for tides and waves, but in the production of sediment types, occurrence of storms, and the types of vegetation present. This is especially true in extreme climates, such as polar regions. For example, the barrier islands on the North Slope of Alaska are eroding

BARRIER ISLANDS

85

away at alarming rates. This erosion continues despite the fact the Arctic Ocean is frozen for many months of the year, producing limited fetch for the waves, hence relatively small waves, and a short season for waves to occur. Even in August, blocks of ice sometimes occur near shore. Composed mostly of gravel, these islands are short (average length < 3 km) and low, with numerous wide inlets. Warming of the Arctic Ocean may be playing a role in the islands’ demise, with melting of the permafrost possibly being a major factor.

probably the hottest and driest barrier island chain in the world, these islands have many of the characteristics of mesotidal barrier islands (the tidal range in Abu Dhabi is 2.5 m). Features such as stunted islands, large tidal inlets with huge ebb-tidal deltas, and complex backbarrier regions stand out. However, typical inlet migration and beach erosion patterns are inhibited by the beachrock, producing some jagged shorelines cemented in place. Also, the ebb-tidal delta sediments are composed of carbonate oolite sand, another signature of arid tropic regions. ORIGIN OF BARRIER ISLANDS

The origin of barrier islands has been a matter of conjecture in the geological literature for well over 100 years. The processes of beach accretion between storms and the formation of a foredune landward of the beach, primary components of barrier islands, are well understood, and their common occurrence on depositional shorelines is not surprising. The aspect of barrier islands that is most difficult to understand is the fact that they occur offshore of the mainland separated by a topographically low area, a lagoon or estuarine complex in most cases. Any theory for the origin of these islands must account for their mysterious offshore location. Numerous hypotheses for the origin of barrier islands have been proposed, including the following three examples.

FIGURE 5 View looking east of a prograding, drumstick-shaped bar-

rier island, Egg Island, Alaska (on the Copper River Delta). Note the recurved spits on the western end of the island, clear evidence of the strong east-to-west longshore transport of sediment along the bar-

1. Elongation of sand spits away from some kind of headland, with segmentation of the spits as they grow as a result of the formation of permanent tidal inlets through them during storms, creating individual islands separated from the mainland by an openwater lagoon (as illustrated in Fig. 7). 2. Elevation of an offshore bar or flooding of a line of foredunes along the shore. 3. The transgressive–regressive interfluve hypothesis, which has been well documented on the coasts of North Carolina and South Carolina.

rier islands in this area. Also note where several large swash bars have welded to the eastern end of the island (compare with model in Fig. 4). Photograph of this mesotidal barrier island taken at low tide in the spring of 1975.

In the other extreme, barrier islands in hot, arid regions are more prone to have barren sand dunes and extensive sand transport across them. In shallow seas with warm water and high salinities, the sediments of the islands beaches may be cemented with beachrock. As the satellite image of the shoreline of Abu Dhabi in Fig. 6 shows, although

86

BARRIER ISLANDS

One of the major proponents of the spit elongation hypothesis, John Fisher, cited the northern part of the Outer Banks of North Carolina as one of his examples of barrier islands that have been formed by that mechanism. This mode of origin has been suggested for other areas in the world, for example those islands on the central Texas coast. This clearly is one way barrier islands can form, but not the only one. There is no doubt that many barrier islands have formed without the aid of spit elongation. Two major ideas have

been proposed to account for the elevation and permanence of barrier islands independent of spit elongation. Some observations along the Gulf Coast of the United States illustrate that during the elevated tides and unusually high water levels that accompany hurricanes, a major offshore bar may be formed by the large waves. When the water level recedes, the bar emerges and, under the right circumstances, may survive to become a barrier island. A second idea is that a line of typical foredunes forms along the beach during a stillstand or slowly rising sea level. A relatively abrupt rise in sea level floods the line of foredunes. During this abrupt rise, the foredunes are not completely eroded away by the waves. The remnant of the original dune line becomes an island. Both of these processes seem reasonable, but their documentation is still somewhat in question. The mode of formation for most of the larger prograding barrier islands on the South Carolina coast, such as Kiawah Island, is clear and well documented in studies by Tom Moslow and D. J. Colquhoun. Four major steps take place in this mode of formation:

FIGURE 6 Satellite image of the barrier island chain in Abu Dhabi

acquired in 2000, courtesy of Earth Science Data Interface (ESDI) at

1. A narrow, landward-migrating barrier island moved rapidly across what is now the inner continental shelf, leaving behind a thin lag of coarse material on top of an erosion surface across the continental shelf, called the transgressive surface of erosion. 2. The topography over which the shoreline advanced was irregular, and estuarine waters flooded the numerous river valleys formed when the shoreline was further offshore. Isolated, primary transgressive barrier islands, consisting of washover terraces composed of coarse-grained sand and shell, continued to migrate landward on the exposed interfluves between the drowned lowstand valleys. 3. When sea level stopped rising and a relative stillstand occurred about 4500 years ago, shoals developed at the entrances of the estuaries created by the drowning of the valleys, and a longshore sediment transport system was initiated along the face of the stranded barrier islands. Over time, beach ridges began to develop, eventually impinging upon the adjacent estuary entrances. As a well defined inlet throat evolved, a shoal off the entrance (ebb-tidal delta) formed, around which sediment was bypassed, augmenting beachridge growth on the adjacent barrier island. 4. As the barrier island matured, and minor fluctuations of sea level occurred, parts of some of the originally prograding beach ridges were eroded as a result of tidal-creek and tidal-inlet migration.

the Global Land Cover Facility. These barrier islands show many of the characteristics of mesotidal barrier islands—short length, large ebb-tidal deltas, and absent flood-tidal deltas (tidal range = 2.4 m this area). The huge ebb-tidal deltas are composed of carbonate sand (with abundant oolites).

The end result of all this was a prograding, drumstickshaped barrier island, such as the ones illustrated in Figs. 1, 4, and 5. However, this hypothesis does not account for how the transgressive element, the original barrier island, formed.

FIGURE 7 The spit elongation hypothesis for the origin of barrier

islands, based on ideas presented by Fisher (1967) and others.

BARRIER ISLANDS

87

SEE ALSO THE FOLLOWING ARTICLES

Beaches / Hurricanes and Typhoons / Sea-Level Change / Tides FURTHER READING

Curray, J. R. 1964. Transgressions and regressions, in Papers in marine geology: Shepard commemorative volume. R. L. Miller, ed. New York: MacMillian and Co, 175–203. Davis, R. A. Jr., ed. 1994. Geology of the Holocene barrier island systems. Berlin: Springer-Verlag. Fischer, J. J. 1967. Origin of barrier island chain shoreline, Middle Atlantic States. Geological Society of America Special Paper 115: 66–67. Hayes, M. O. 1979. Barrier island morphology as a function of tidal and wave regime, in Barrier islands, from the Gulf of St. Lawrence to the Gulf of Mexico. S. Leatherman, ed. New York: Academic Press, 1–27. Hoyt, J. H. 1967. Barrier island formation. Geological Society of America Bulletin 78: 1125–1136. Pilkey, O. H., and M. E. Fraser. 2003. A celebration of the world’s barrier islands. New York: Columbia University Press.

BARRO COLORADO EGBERT GILES LEIGH, JR. Smithsonian Tropical Research Institute, Balboa, Panama

Barro Colorado is a 1500-ha island situated at 9˚9′ N, 79˚51′ W in central Panama, first isolated from the surrounding mainland in 1914 by the rising waters of Gatun Lake, after the Chagres River was dammed to form part of the Panama Canal. The island is covered by seasonal tropical forest, half of it old growth, which offers beauty and fascination enough to fill any biologist’s lifetime. Barro Colorado’s primary claim to the reader’s attention is its contribution to our understanding of tropical biology. HISTORY

The governor of the Canal Zone declared this island a reserve in 1923 in response to two groups. Biologists from Harvard and the American Museum of Natural History wished to preserve this island for their research, whereas Canal Zone entomologists, who were spraying mainland areas to eliminate malaria-carrying mosquitoes, wished to keep the island unsprayed, a standard for measuring the effectiveness of their spraying. Their standard was high. Most mosquitoes on this island breed in water-filled tree holes, which also shelter resident nymphs of 100-mm-long giant damselflies. These nymphs eat most of the wrigglers, keeping Barro Colorado relatively free of mosquitoes. Barro Colorado’s first buildings were erected in 1924. James Zetek, an entomologist, directed the laboratory

88

BARRO COLORADO

during its first, impoverished, 30 years. The Smithsonian was given charge of Barro Colorado in 1946. In 1957, Martin Moynihan, a student of animal behavior, was appointed resident naturalist. He set about assembling a staff of resident scientists. In 1966, the Smithsonian Tropical Research Institute, including Barro Colorado and marine laboratories on both coasts, was declared a bureau of the Smithsonian Institution. Moynihan was extraordinarily perceptive and was interested in nearly everything. He distrusted team research, believing that the best ideas come from people pursuing projects they devised themselves. He established a sound intellectual basis for further research at his Institute. Ira Rubinoff succeeded him in 1973, established agreements with the Republic of Panama for the Institute’s continued activity, broadened the range of scientific specialties at the Institute, and built the laboratories needed to support this research. THE CENTRAL PROBLEM OF ANIMAL BEHAVIOR AND ITS ANALOGUES

An early theme of Barro Colorado’s research was animal behavior. In the 1930s, C. R. Carpenter was studying howler monkeys, which live in groups of 15 or more, and T. C. Schnierla was studying army ants, whose workers cooperate in organized swarms of up to a half million ants to overcome prey. A central problem of animal behavior is why animals live in groups. An animal’s fellow group members are usually its closest competitors for food, shelter, and mates: What keeps competition among a group’s members from becoming a cheating contest? This problem has many analogues. One is far older than Plato: How can the advantage of individuals be aligned with the good of their society? Second, a person’s body consists of a multitude of cells that normally cooperate for that person’s good, yet rogue cell lines—cancers—can spread and kill that person. Why are cancers not much more common? Third, partnerships between members of different species confront similar tensions. For example, each of our cells has mitochondria, organelles that turn sugar into energy. These mitochondria descend from parasitic bacteria that invaded our one-celled ancestors over a billion years ago. Conflicts of interest between cells and their mitochondria sometimes erupt, but why so rarely? One partnership between species, a classical example of coevolution, profoundly influences this island’s ecosystem. In 1980, Allen Herre began working on fig trees, each species of which maintains one or more species of minute wasp as dedicated pollinators. A fig “fruit” is a flowerhead turned outside in, so the flowers line the inside of a perforated ball. One or more pollen-bearing wasps enter each

mycorrhizae, root fungi, to help extract mineral nutrients from the soil. Many need animals to disperse their seeds, and perhaps even bury them, beyond the reach of the parents’ pests. Barro Colorado’s early emphasis on social behavior encouraged study of the many different ways in which the plants and animals of tropical forest depend on each other, and the ways participants defend cooperative relationships against subversion by cheaters. DIVERSITY OF ANIMALS AND PLANTS

FIGURE 1 A view of mature forest on Barro Colorado Island. From the

left, the drought-sensitive species Poulsenia armata; the long-lived species Anacardium excelsum, which colonizes large tree-fall gaps; and Quararibea asterolepis, a shade-tolerant canopy species. At far right, the midstory tree Gustavia superba, with rosettes of very long leaves. Drawing by Daniel Glanz.

“fruit,” pollinate its flowers, and lay eggs in about half of them. Each wasp larva grows within a single fig seed and metamorphoses into an adult. In each fruit, hundreds of wasps emerge and mate with each other. Fertilized females fly off in search of trees in “fruit” in which to lay their eggs; they pick up pollen and carry it from tree to tree.In the 1990s, John Nason found that although these wasps live less than 72 hours, they often pollinate trees over 10 km away, maintaining great genetic diversity in their fig species even if it is very rare. To maintain its pollinators, each species of fig must always have some trees in fruit. On Barro Colorado, this year-round fruit supply supports ten species of fig-eating bat. The island’s 17 fig species bear different-sized fruit. Larger fruit attracts bigger bats, which carry seeds further from parents and their pests. Seedlings further from adults of their species are more likely to grow to a safe size before pests specialized on their species find them, so these plants need invest less in defenses against their pests, freeing resources for faster growth. Wasps die, however, if their fruit overheats: Large fruits must evaporate much water to keep cool, so large-fruited figs must have reliable access to water. This mutualism, however, has its tensions. In 2005, Charlotte Jander began to study how fig trees make their wasps pollinate their fruit. Another question, as yet unanswered, is how fig trees keep their wasps from laying eggs in more than half their flowers. Barro Colorado’s denizens live by such partnerships. Most species of plants need animals that will carry pollen from one plant to another of its species, even when this species is kept rare by its pests. Likewise, most plants need

A second research theme is the great diversity of Barro Colorado’s plants and animals. The first step in this study is identifying them. The Smithsonian’s Paul Standley first catalogued the island’s plants in 1927. In 1935, a Swarthmore professor, Robert Enders, summarized what was known of the island’s mammals. In 1952, the American Museum’s Eugene Eisenmann published a list of its birds. Such work continues. How can forests like Barro Colorado’s harbor so many kinds of plants and animals? One light in this darkness is the “principle of competitive exclusion”: If two species make their living in the same way, in the same place, one will be better and will replace the other. Diversity reflects trade-offs, circumstances where enhancing the aptitude for one task diminishes the aptitude for others. Where trade-offs occur, the threat of competitive exclusion drives adaptive divergence in competing populations. Work on Barro Colorado helped to reveal factors driving the tradeoff plants face between growing fast in bright light and surviving in shade. This trade-off allows fast-growing species of “weed tree” that survive only in tree-fall gaps to coexist with shade-tolerant tree species of mature forest (Fig. 1). Insects and disease-causing microbes face tradeoffs in ability to handle defenses of different kinds of plant. The most damaging pests specialize on particular species or genera of plants. Pests play a major role in the lives of this island’s trees. From observations that he began when he first came to the island in 1967, Robin Foster found that trees of the canopy species Tachigali versicolor flower and fruit once, then die. Only every four years do some Tachigali have fruit. Yet a weevil’s specialist larvae destroy 20% of their fruit. Did trees that fruited a second time lose their whole second crop to these weevils? Work on this island helped to show how, in some tree species, a parent’s pests destroy its nearby young, making room for trees of other species to grow up between a parent and those of its young far enough away to survive. To document and explain the island’s tree diversity, Robin Foster and Stephen Hubbell set up a 50-ha Forest

BARRO COLORADO

89

Dynamics Plot in 1980, mapping, marking, measuring, and identifying every free-standing woody stem over 1 cm diameter. In 1982, this plot had 305 species among 235,000 stems over 1 cm in diameter, and 238 species among its 20,881 trees over 10 cm trunk diameter. The plot was recensused every five years from 1985 on. To document tree diversity, plots like it were established all through the tropics. A 52-ha plot in Lambir, Sarawak, had 1008 species among 32,661 trees over 10 cm trunk diameter. Comparing these plots shed light on the causes of tree diversity. In these plots, trees usually reproduce less well where they are more common, so no one kind of tree can crowd out the others, allowing many species to coexist. The next question: Are these forests so diverse because each species is kept rare by its specialist pests? HERBIVORES AND THE FOREST

The third research theme is, What keeps the animals from stripping the forest bare? Exploring this theme illustrates the importance of “unity of place,” showing how many different kinds of study, done in one small area, can cohere into a unified picture. Starting in August 1969, Robin Foster measured fruit fall on Barro Colorado every week or two, for two years. In both years, little fruit fell from November through February. In striking contrast to 1969, however, little fruit fell from August through October in 1970. Many mammals starved, the vultures could not keep up with the corpses, and the forest stank. Were populations of vertebrate fruit-eaters limited by seasonal shortage of suitable food? This question dominated the next 15 years of research on Barro Colorado. The staff scientist Stanley Rand established an environmental monitoring program that maintained records on weather, stream flow, soil moisture, fruit fall, the times of flowering, fruiting and leaf flush in many tree species, and the responses to these events of many animal populations. Year-long studies of the behavior of agoutis, pacas, sloths, coatis, howler monkeys, many kinds of bats, and selected birds suggested that their populations were limited by seasonal shortage of fruit or new leaves. The island’s mammals were censused. Knowing their numbers and sizes, one can calculate their food requirement. In December and January, the measured fruit fall was too little to feed the ground-dwelling fruit-eaters. True, these animals do not starve then. Indeed, each species has a different fallback specialty, which is why no species is entirely replaced by its competitors. These specialties, however, barely tide them over to the next fruiting season: The season of fruit shortage is obviously a time of dearth.

90

BARRO COLORADO

Especially where many different scientific specialties are represented, questions beget questions. How do different kinds of plants know when to flower, fruit, or flush new leaves? In particular, what keeps most plants from fruiting during the season of shortage? Observations suggest that some plant species flower only after sensing a clear transition from dry to rainy season: If the dry season is too wet, they will not flower. Others flower in response to dry season rains, if the soil is dry enough before the rain. To test these ideas, the staff scientist S. Joseph Wright watered two 2.25-ha plots through five successive dry seasons. The behavior of nearly all the larger trees was unchanged by watering. Everyone was astonished by this lack of response. Plant physiologists found that most of the island’s big trees had reliable access to water all year long. They must time their activities in response to atmospheric conditions, such as humidity or solar radiation. Some trees even flower in response to seasonal changes in day length. There is still much to learn about how different trees choose the times for their activities. Insects also eat plant leaves. Mature leaves are too tough for most insects to eat, but young leaves are tender and nutritious. In 1977, Foster’s student Phyllis Coley began work that eventually showed that young leaves on Barro Colorado are eaten 20 times more rapidly than their temperate-zone counterparts, despite being more poisonous than the latter. To protect their young leaves, some plants shoot them out and toughen them as soon as possible. Others stock slower-growing leaves with a cocktail of poisons, different for each species. Coley found some of these poisons to be of medical interest. Specialist herbivores deal best with the poisons in their host’s leaves. These specialists do the most damage. Do they keep their host species rare enough that many of its plants escape the pest’s attentions until big enough to survive them? Barro Colorado’s forest loses at least 7% of its leaf production— over a half ton dry weight of leaves per hectare per year— to leaf-eating insects. Judging from the numbers, weights, and diets of the island’s birds, a third of this eaten foliage feeds insects birds eat. The forest can defend itself from vertebrates fond of vegetable matter without help from jaguars, but it needs help from birds, wasps, and spiders to deal with its insect pests. THE CONSEQUENCES OF BECOMING AN ISLAND

The fourth research theme is, What are the consequences of becoming an island? Since Barro Colorado became an island, its species diversity has been declining toward a new, lower equilibrium as predicted by the theory of

island biogeography. Barro Colorado is now too small to support white-lipped peccaries, which live in herds of a hundred or more. It is too small to guarantee the presence of very young forest. Accordingly, a pygmy squirrel that lived in such forest died out after its forest matured. In 1982, Barro Colorado’s 50-ha plot had 238 species among its trees over 10 cm trunk diameter, but it had only 226 in 2000: Did this happen because the plot is on an island? Many of the island’s bird species, 35 edge and 30 forest species, died out after their home became an island. Most of the 30 forest species are understory insectivores. Experiments suggest that in most of the extinct understory species, birds cannot fly 100 m over open water, either because bright sunlight dazzles them or because they lack the stamina to fly so far. When these species die out, they cannot return. On the other hand, Barro Colorado, with 74 species of bat, lacks few of the bat species found on the nearby mainland. Indeed, Barro Colorado’s ecosystem is not sufficient unto itself. Some of its insect populations are replaced from the nearby mainland when they die out. Some North American songbirds spend the winter there. Some bats leave the island during certain seasons: No one knows where they go. The island lies athwart the route of many species of butterfly that migrate seasonally within the tropics. Many species of bat and bird, even hummingbirds, regularly fly between the island and the nearby mainland. Many animals, including pumas and ocelots, swim to the mainland and back. This traffic has delayed the onset of inbreeding in many populations, prevented extinctions, and played an essential role in maintaining the integrity of Barro Colorado’s ecosystem. Ironically, by becoming a (reasonably large) island, Barro Colorado became a good place to learn how mainland forest ecosystems are organized. Concentrating research there provided “unity of place”: Each project there provided both the empirical background data and the intellectual foundation for further work, and essential context for other, very different projects. These many projects accordingly cohere into a common story of this forest’s function. Work there has also helped to answer many other questions: How mutualism evolves, why there are so many kinds of tropical trees, where to look for chemical compounds most likely to be useful for future medicines, and how a large group of mindless army ants following simple rules can form a well-organized swarm of raiders. SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Deforestation / Island Biogeography, Theory of / Vegetation

FURTHER READING

Leigh, E. G., Jr. 1999. Tropical forest ecology: a view from Barro Colorado Island. New York: Oxford University Press. Leigh, E. G., Jr., A. S. Rand, and D. M. Windsor, eds. 1996. The ecology of a tropical forest: seasonal rhythms and long-term changes. Washington, DC: Smithsonian Institution Press. Losos, E. C., and E. G. Leigh, Jr., eds. 2004. Tropical forest diversity and dynamism. Chicago: University of Chicago Press. Robinson, W. D. 1999. Long-term changes in the avifauna of Barro Colorado Island, Panama, a tropical forest isolate. Conservation Biology 13: 85–99. Ziegler, C., and E. G. Leigh, Jr. 2002. A magic web. New York: Oxford University Press.

BEACHES BRUCE RICHMOND U.S. Geological Survey, Santa Cruz, California

Beaches are shoreline accumulations of loose sand, gravel, or a mixture of the two, that are formed primarily by the action of waves. Beach sediment can be derived from a variety of sources including insular shelves, the adjacent land and upland sources, or other beach locations through alongshore movement of material. Beaches provide critical coastal habitat, such as nesting sites for sea turtles; they act as a buffer protecting adjacent land from storm wave attack; and they are an important cultural and recreational resource. Island beaches are the same as those on the continents, but island beach characteristics typically change over very short distances on account of rapid changes in coastline orientation, exposure to waves, and sediment source. BEACH MORPHOLOGY

Beaches straddle the boundary between land and water, extending from the low-water line to the supratidal backshore. A typical beach consists of a beach face or foreshore, a berm or crest, and backbeach or backshore (Fig. 1). The foreshore occurs between the low- and high-tide water levels and typically has a concave-upward topography. The base of the beach commonly ends with a steep step or “beach toe.” In areas of ample sediment supply and suitable nearshore conditions, wave-formed bars occur in the surf zone. The crest, or berm, of the beach is the boundary between the relatively steep beach face and the flatter backbeach area. The landward limit of the backshore is typically marked by permanent vegetation, dunes, or

BEACHES

91

valley. Drowned stream valleys may have a structural origin that is further emphasized by stream development, whereas embayments of a strictly structural origin have no pronounced stream development. Structural embayments can be created by a variety of processes including volcano growth characteristics (including coalescing volcanoes), volcano decay (e.g., landslides and crater collapse), faulting, and vertical tectonic displacement. Pocket Beach

FIGURE 1 Photograph of a beach from the west end of St. Croix, U. S.

Virgin Islands, showing the beach toe, foreshore, berm crest, backshore, and an older berm marked by an erosional scarp. The vegetation

Pocket beaches are common features of rocky coasts worldwide and are essentially smaller (< 1 km long) versions of the drowned stream valley or structural embayments (Fig. 2). Pocket beach sediment can be relatively isolated from adjacent systems, and there may be a marked difference in composition from one beach to the next.

marks the more stable section of beach. This is a popular sea turtle nesting site. Photograph by the author.

coastal cliffs and indicates the highest reach of the waves during storms. Beach planform is dependent primarily upon the local physiographic setting. For example, crenulated coastlines typically produce confined pocket beaches, whereas straighter coastlines tend to be bordered by long, gently curving beaches. Beach types include stream valley with bayhead beach, structural embayment with beach, pocket beach, coastal plain/cuspate foreland beach, delta beach, and, perched beach. In addition to these enclosed beach systems, some islands such as barrier islands and sand cays are for the most part entirely bordered by beaches along their shorelines. The following are brief descriptions of island beach physiographic settings and their important characteristics (adapted from Richmond 2002).

FIGURE 2 Oblique aerial photograph of the beach at Waimea Bay on

the north shore of O‘ahu, Hawai‘i. This pocket beach, composed mostly of carbonate sand, is formed at the mouth of a stream within a structural embayment. Photograph by the author.

Coastal Plain/Cuspate Foreland Beach Stream Valley with Bayhead Beach

Stream valleys, which are typically cut during sea-level lowstands and subsequently drowned during subsequent sea-level rise, are often the sites of bayhead or barrier beach deposition. Where there is an open inlet to the sea, the beach may form a spit, which seasonally opens or closes depending upon stream outflow and wave energy. Sediment composition is typically a mixture of terrigenous and marine-derived material. Structural Embayment with Beach

Embayments of either fluvial or structural origin can be distinguished by the presence or absence of a major stream

92

BEACHES

Broad coastal plains (hundreds of meters wide, 1+ km long), composed mostly of aggraded beach deposits, are indicative of long-term stability and deposition. Many of the larger coastal plains on islands tend to occur along leeward shores. The convex-seaward plan form of the cuspate foreland (Fig. 3) distinguishes them from beaches within embayments. Cuspate forelands often migrate laterally by erosion on one side accompanied by deposition on the other. Changes in weather patterns can lead to reversals in transport directions, resulting in a very dynamic coastline prone to large changes in shoreline position on both seasonal and decadal scales. For example, the seasonal shifts in monsoon winds in the Maldives results in a significant

FIGURE 3 Oblique aerial photograph of the cuspate foreland shoreline

FIGURE 4 Perched beach at Pu‘uhonua o Honaunau National Historic

and coastal plain on the southwest coast of Isla de Mona, Puerto Rico.

Park (PUHO) on the west (Kona) coast of the island of Hawai‘i. The

The arrows point to areas of exposed beachrock that denote former

beach, which is composed mostly of reef-derived carbonate sand and

shoreline positions. Photograph by the author.

gravel and minor amounts of volcanic debris, is perched on top of a basalt platform that lies just above mean sea level. View to the north.

shift in shoreline position between seasons, and El Niño– Southern Oscillation (ENSO) events commonly increase erosion of leeward shores of Pacific islands through an increase of westerly winds and waves. Delta Beach

On coastlines where stream deposition is greater than the rate of longshore dispersion of sediment, a seaward-protruding delta beach will form. Delta beaches are usually limited to older islands with well-developed stream systems. Sediment composition is typically mostly terrigenous. Ephemeral deltas may form as a result of a large storm/rainfall event and slowly disappear as sediment is reworked alongshore.

Photograph by the author.

Barrier Island

Barrier islands are depositional features offshore of a mainland coast with shorelines that are typically composed almost entirely of beach deposits (Fig. 5). Barrier islands are more common in temperate and higher latitude settings where there is a sufficient supply of sediment and a gently sloping continental shelf.

Perched Beach

Along rocky coasts where low terraces are developed, storms can form beaches above the normal wash of the tides (Fig. 4). These perched beaches are active only during storms and other large-wave events. Perched beaches occur on both volcanic islands where basalt benches have formed near sea level and on limestone islands where elevated limestone platforms border the coast. Sand Cay

Sand cays are small, roughly circular to oval-shaped islands, which occupy the tops of intertidal to shallow subtidal platforms (commonly a reef platform in tropical settings). The entire shoreline of a sand cay is beach deposits. The location of the sand cay on the underlying platform represents some long-term average sediment depocenter. The shape of the cay can change in response to seasonal weather variations and possibly during individual storms.

FIGURE 5 Oblique aerial photograph of Cross Island, a barrier island in

the Beaufort Sea near Prudhoe Bay off the north coast of Alaska. The entire island is bounded by beaches that are modified by waves during ice-free periods, typically a few months during the summer, and by sea ice the rest of the year. Photograph by the author.

BEACHES

93

TABLE 1

Approximate Coastline Length, Percent Beach Shoreline, and Island Type for Selected Pacific Islands Island

Hawai‘i Maui Moloka‘i Lana‘i O‘ahu Kaua‘i Kosrae, FSM Pohnpei, FSM Chuuk, FSM Majuro, RMI Yap, FSM Guam Saipan, CNMI

Approx. Coastline Length (km)

Approx. % Beach

492 255 170 84 320 182 90 180 110 140 124 60 55

8 21 22 34 28 43 31 1 30 100 9 33 31

Island and Reef Type

Active volcanic Young volcanic Volcanic (partial fringing reef ) Volcanic (partial fringing reef ) Volcanic (partial fringing reef ) Volcanic (partial fringing reef ) Volcanic (continuous fringing reef ) Volcanic (barrier reef ) Volcanic (almost-atoll/barrier) Atoll Mixed (uplifted; fringing reef ) Mixed (uplifted; fringing and barrier reef ) Mixed (uplifted; fringing and barrier reef )

note: Approximate coastline length and percent of beach shoreline for the main Hawaiian Islands are modified from Moberly and Chamberlain (1964), and the values for the islands in the Federated States of Micronesia (FSM) are interpolated from U.S. Geological Survey topographic maps. CNMI = Commonwealth of the Northern Mariana Islands; RMI = Republic of the Marshall Islands. Modified from Richmond (2002).

BEACH SEDIMENT

Beaches comprising sand-size material are the most common and occur throughout the world. Gravel beaches are much less common and tend to occur where there is a local gravel source, such as near gravel-rich rivers, gravel-bearing cliffs, or coral reef environments. Although mixed sand and gravel beaches exist, in general beach sediment tends to be moderately well-sorted, with mean grain size decreasing in a downdrift direction away from the source area. The composition of the beach sediment is directly related to the local source material. For example, in tropical environments away from the effects of rivers, reefderived biogenic carbonate sediment dominates the beach composition. Carbonate components may include fragments of coral, mollusc, foraminifera, coralline algae, and calcareous green alga (Halimeda). The amount of carbonate sediment decreases with distance from the reef and/or increasing terrigenous input, mostly from streams. Temperate and high-latitude beaches are typically composed of terrigenous sediment delivered primarily from rivers but with contributions from erosion of the adjacent landscape. In islands formed by continental-type rocks, quartz and feldspar sediments are dominant with smaller amounts of heavy minerals. Young volcanic island beaches often have high heavy mineral content and volcanic clasts. The percentage of beach that is comprised in island shorelines is related to the type of island and island age (Table 1). For example, in the main Hawaiian Islands, the percent of beach volume increases with island age, with a low percentage on Hawai‘i (youngest) and a much higher percentage on Kaua‘i (oldest). The increase in beach volume on older islands is due to a combination of well-developed stream systems, which can deliver large amounts of sediment to the 94

BEACHES

coast, and more pronounced reef development, which provides a consistent supply of reef-derived carbonate sediment. The development of large beach systems requires a relatively stable landmass, adequate coastal accommodation space for beach deposition, and a steady and abundant supply of sediment. Although modern beaches are essentially Holocene features, well-developed beach systems require a much longer time frame to evolve. Longer time for development results in increased reef formation and greater land denudation—both of which lead to increased sediment production. In summary, island beaches are best developed and most stable where there is an abundant source of sediment and sufficient time and space for the sediment to accumulate, the rate of vertical tectonic movement is low (i.e., a stable island), and anthropogenic disturbances are minimal. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Barrier Islands / Erosion, Coastal / Motu / Surf in the Tropics / Tides FURTHER READING

Bascom, W. 1980. Waves and beaches, revised and updated. New York: Anchor Doubleday. Komar, P. D. 1998. Beach processes and Sedimentation, 2nd Ed. Upper Saddle River, NJ: Prentice Hall. Moberly, R., and Chamberlain, T. 1964. Hawaiian beach systems. Final Report prepared for Department of Planning and Economic Development, State of Hawai‘i, Hawai‘i Institute of Geophysics HIG-41. Richmond, B. M. 2002. Overview of Pacific Island carbonate beach systems, in Carbonate beaches 2000. Proceedings of the First International Symposium on Carbonate Sand Beaches. L. L. Robbins, O. T. Magoon, and L. Ewing, eds. Reston, VA: ASCE, 218–228. Robbins, L. L., O. T. Magoon, and L. Ewing. 2002. Carbonate beaches 2000. Proceedings of the First International Symposium on Carbonate Sand Beaches. Reston, VA: ASCE. Short, A. D., ed. 1999. Handbook of Beach and Shoreface Morphodynamics. Chichester, UK: John Wiley and Sons.

BERMUDA ANNE F. GLASSPOOL AND WOLFGANG STERRER Bermuda Zoological Society, Flatts

Located at 32°18′ N, 64°46′ W in the Sargasso Sea (northwestern Atlantic Ocean), Bermuda is a small, low-lying oceanic archipelago of solidified wind-blown dunes (aeolianite) lying atop an eroded volcanic platform. Because of the Gulf Stream’s influence Bermuda boasts a subtropical climate that supports the world’s northernmost coral reef and mangrove systems. GEOLOGY AND GEOGRAPHY

The Bermuda Seamount originated at least 33 million years ago, probably on top of a much earlier eruptive episode (110 million years ago), as a towering mid-ocean volcano with two side peaks (now Challenger Bank and Argus Bank). Drifting westward on the North American Plate, the extinct volcano eventually eroded down to sea level and, over the Pleistocene (1.8 million to 10,000 years ago), acquired a surface topography of solidified calcareous sand dunes derived from surrounding coral reefs (Fig. 1). Aeolian (wind-blown) dunes accumulated during warmer, interglacial periods of rising sea levels (e.g., 400,000 ago, to 22 m above today’s). During cooler, glacial periods of falling sea levels (e.g., 18,000 years ago, to 120 m below today’s), dunes stabilized and solidified when rain first dissolved, then recrystallized carbonate sand grains. The youngest limestone formations (called

Southampton), therefore, are barely consolidated calcareous sand, whereas the oldest (Walsingham) are largely recrystallized, containing extensive karst features such as solutional caves and cave collapses, most of which are now below sea level. Limestone formations are separated by geosols (fossil soils) made of dissolved limestone, plant debris, and atmospheric dust from as far as the Sahara Desert. Present-day Bermuda is a low-lying (maximum elevation 79 m), fishhook-shaped chain of four larger islands surrounded by hundreds of islets totaling 53.7 km2. Together these enclose significant inshore basins and line the southeastern margin of an extensive (750 km2) but shallow (average depth 10 m) oval lagoon, which is, in fact, formed by the truncated top of the volcanic pinnacle rising from the deep sea. During the Pleistocene, eustatic sea level fluctuations alternately exposed and flooded the platform about every 100,000 years, and the Island’s land mass oscillated between two extremes—a contiguous area of nearly 1000 km2 surrounded by a narrow reef fringe at low sea levels, and a string of a few islets totaling less than today’s land area surrounded by an extensive shallow reef lagoon at high sea levels. CLIMATE

Bermuda’s climate is influenced by the Mid-Atlantic (or Bermuda-Azores) High. In the summer (May–October), winds are relatively weak and southeasterly, whereas during the winter (November–March) westerlies predominate, including gales. The tropical storm season is between May and November, with one hurricane approaching Bermuda every year on average and a severe hurricane expected every 4–5 years. The annual total rainfall of 150 cm is evenly distributed through the year, and humidity is uniformly high (70–82%). Mean monthly air temperatures range between 18.5 °C (February) and 29.6 °C (August), and mean sea temperatures between 17.3 °C (February) and 28.0 °C (August). Tides are semidiurnal, with a mean annual range of 0.75 m. PRECOLONIAL ECOLOGY

FIGURE 1 Bermuda, an aerial view taken in 1997, showing the land and

surrounding reef complex, the most northerly in the world. Photograph courtesy of Bermuda Zoological Society.

Precolonial Bermuda was a low, hilly landscape with freshwater marshes (but no streams), densely wooded with some 15 species of endemic evergreen plants such as the Bermuda cedar (Juniperus bermudiana), Bermuda palmetto (Sabal bermudana), olivewood bark (Cassine laneana), and some 150 native plants. Forests and marshes were populated by a species-poor invertebrate fauna consisting of about 200 native and some 92 endemic species (including the only endemic genus, the land snail

B E R M U DA

95

Poecilozonites). Bermuda was a breeding ground for at least four species of Atlantic sea turtles, a dozen species of landbirds (including an endemic subspecies of the white-eyed vireo, Vireo griseus bermudianus), and at least six species of seabirds including the endemic Bermuda petrel or cahow (Pterodroma cahow, Fig. 2), which may have had a nesting colony of a million birds. There were no amphibians or terrestrial mammals, and a small lizard, the endemic Bermuda skink (Eumeces longirostris) was the only four-legged land animal. Thanks to the warm Gulf Stream, which passes halfway between the island and Cape Hatteras, Bermuda has the complete range of tropical marine habitats, from drowned karst caves and brackish ponds to inshore lagoons, sandy and rocky shores, mangroves, sea grass beds, and coral reefs. Bermuda’s coral reefs extend over the entire lagoon.

FIGURE 2 The endemic Bermuda petrel or cahow (Pterodroma cahow),

“rediscovered” after it was believed to have been extinct for nearly 300 years, is now the subject of an intensive translocation initiative. Photograph by Andrew Dobson.

the major vector for drifting and rafting colonizers. Of about 60 species of shallow-water reef corals (Scleractinia) known from the Caribbean, for example, 24 species (40%) also occur in Bermuda. The island’s fossil record does not predate the Pleistocene, with salient terrestrial records of snails and birds showing significant extinctions and recolonizations, including an endemic tortoise (Hesperotestudo bermudae) known only as a single 300,000-year-old fossil. Despite its isolation and great age, Bermuda has very few endemic species in both marine and terrestrial habitats, as highlighted in a comparison with Hawaii (Sterrer, 1998), where marine endemism is five times greater (10.6% vs. 2.1% in Bermuda), and terrestrial/freshwater endemism is 13 times greater (48.3% vs. 3.7% in Bermuda). Most of Bermuda’s endemics are found in drowned (anchialine) caves (Fig. 3), where of a total of 86 species recorded (mostly Crustacea), 80 (93%) are endemic (including three endemic genera). The Island’s low rate of endemism is best explained by its relative paucity of terrestrial habitats, its down-current position with regard to the Gulf Stream (which maintains genetic continuity with Caribbean biota), and by habitat discontinuity during the Pleistocene, when low sea levels would have favored extensive forests and marshes at the expense of shallow reef environments, with high sea levels reversing this ratio. Karst caves, whose depth exceeded sea level fluctuations, may have been among the few temporally continuous habitats, and thus favored the origin and persistence of endemics. HUMAN COLONIZATION AND ITS IMPACT

Discovery by the Spanish in 1503 marked a significant turning point in the natural history of Bermuda. Initially,

The entire southern coast is paralleled by a string of circular “algal-vermetid” reefs (also called mini-atolls, or “boilers”) which, arising from an ancient shoreline, are constructed mainly by sedentary worm snails (Vermetidae) and encrusting red algae. The northern lagoon is littered with “coralgal” reefs (i.e., constructed mainly by corals and coralline algae), which have irregular round or elliptic shapes in the lagoon, but are rather linear at the rim of the platform where they form a near-impenetrable bulwark of breakers. BIOGEOGRAPHY

Bermuda’s native flora and fauna are a subset of southeastern North America and the tropical Caribbean, as might be expected if one accepts the Gulf Stream as

96

B E R M U DA

FIGURE 3 Divers explore one of the many submerged passages in

Bermuda’s anchialine cave system. Photograph by Christian Lascu.

fears of the treacherous reefs, coupled with haunting nocturnal cries of the cahow, deterred sailors from actually landing on the “Isle of Devils.” However, in 1609 the shipwrecking of the Sea Venture off Bermuda during a hurricane, followed by the survival of all on board as they sought refuge on the island, inspired a new perspective, and permanent settlement followed. As word of the Island’s bountiful resources spread (notably fish, turtles, and whales), and new settlers experimented with a range of introduced crop species and domestic animals, Bermuda for a time came to be viewed as a mid-Atlantic provision store. Rapid depletion of the natural resources inevitably followed, and this, coupled with a suite of less desirable introductions—including soon-to-be terrestrial invasives (such as the common rat [Rattus rattus], American cockroach [Periplaneta americana], weevils, and wireweed [Sida carpinifolia])—also marked the beginning of widespread and largely irreversible changes to the native biodiversity. By 1684, when it was apparent that many of the early economic speculations (ambergris, pearls, and silk) and cash crops (tobacco and sugar cane) were not viable, Bermudians had to reinvent their survival strategy and embarked upon a 150-year era of shipbuilding, whaling, maritime commerce, and privateering. Fueling this activity was the endemic cedar, which provided food and lumber, inevitably leading to heavy deforestation, which in turn provided the opportunity for the continual influx of exotic species to firmly establish themselves. Agriculture reemerged as an export business in the mid-1800s until the 1930s, and for 200 years after 1783 Bermuda was also a strategic naval and military outpost, most significantly during World War II. Inspired by this rich history and the more than 300 documented shipwrecks, Bermudians have been pioneers in the development of modern archeological methods. The twentieth century saw Bermuda become transformed into an almost exclusively service-based economy, first through tourism and more recently through international business. Today, Bermuda supports a resident population of 68,000, with 400,000 visitors a year, and the island has been transformed into a largely suburban landscape (Fig. 4). More than 50% of the land mass is considered developed, with 14% covered by artificial surfaces. Gardens, golf courses, and arable land occupy a further 20%. No undisturbed upland valleys remain, 75% of upland coastal habitat has been developed, and remaining upland forest occupies just 39%

FIGURE 4 Most of Bermuda’s landscape is now considered suburban,

as evidenced in this aerial photograph. Photograph courtesy of Bermuda Government Information Services.

of its former area. Only the coastal habitats remain relatively unchanged (Table 1). Bermuda’s close proximity to North America and easy access to deep water has encouraged a rich history of scientific exploration, including the HMS Challenger expedition of the 1870s, and William Beebe’s deep-sea bathysphere triumph in the 1930s. Furthermore, inspired by the rediscovery of the cahow in 1951, which was believed to have been extinct for nearly 300 years, the creation of the Nonsuch Island Living Museum is acknowledged as one of the first global examples of an island restoration project. It also TABLE 1

Change in the Area of Bermuda’s Habitats since Colonization Habitat, in Hectares

Beach and dune Mangrove swamp Salt marsh Marine pond Rocky coastal Upland coastal Upland hillside Brackish/fresh pond Peat marsh Upland valley Limestone sink Arable field/pasture Garden Golf course Developed, including hedgerows, walls, wayside, road verges Total

AD 1600

%

AD 2000

%

76 24 4 17 162 1382 2303 8 119 921 125 0 0 0 0

1.5 0.5 0.1 0.3 3.2 26.9 44.8 0.1 2.3 17.9 2.4 0.0 0.0 0.0 0.0

76 18 1 17 90 348 903 10 45 0 67 178 669 260 2689

1.4 0.3 0.0 0.3 1.7 6.5 16.8 0.2 0.8 0.0 1.2 3.3 12.5 4.8 50.1

5141

100

5371

100

note: Bermuda’s total land area has increased since AD 1600 as a result of land reclamation, primarily for the airport.

B E R M U DA

97

TABLE 2

Change in Bermuda’s Major Terrestrial Taxa since Colonization

Terrestrial Species

Flowering plantsa Ferns, mosses Molluscs Insectsb Spiders Amphibians Reptiles Birds Mammals Total

Endemic

10 6 18 44 2 0 1 4 0 85

(of Which

Non-Endemic

Introduced

Total

Extinct)

Native

Naturalised

Species

% Aliens

371 17 33 703 34 3 4 9 4 1178

531 38 57 919 41 3 5 20 4 1618

70 45 58 76 83 100 80 45 100 73

0 0 6 16 0 0 0 3 0 25

150 15 6 172 5 0 0 7 0 355

a

The total for introduced species includes only naturalized, self-propagating species. The total for introduced species excludes interceptions and isolated records.

b

marked the real start of the conservation movement in Bermuda, triggered by the devastating effects of the cedar blight, a disease of Bermuda’s cedars caused by two accidentally introduced scale insects, the juniper scale (Carulaspis minima) and the oyster-shell scale (Insulaspis pallida), in the 1940s. This prompted perhaps one of the biggest attempts at biological control ever tried, with the introduction of over 100 species of insect predators, mainly beetles and wasps. With the loss nevertheless of 94% of the island’s cedar trees, a huge reforestation effort was initiated with a suite of alien species, including casuarina (Casuarina equisetifolia), Brazil pepper (Schinus terebinthifolius), and Indian laurel (Ficus retusa), resulting in wholesale change of the landscape. Recent islandwide vegetation surveys have revealed that 22 invasive plant species are now a dominant feature of the 33% of Bermuda’s land area that remains undeveloped. And of more than 1600 resident terrestrial plant and animal species, only 27% are native (Table 2). Twentyfive endemic species have become extinct, 200 native species have declined significantly, and at least 1200 exotic species have become naturalized (mainly flowering plants, insects, spiders, snails, birds, reptiles, and amphibians). Bermuda’s shallow water marine platform has been less impacted by humans, with only ten native species and one endemic known to have been extirpated. Today, shipping and shoreline development pose the main threats to the inshore marine environment; however, the reefs remain extremely healthy particularly in comparison with the neighboring Caribbean, with coral coverage as high as 75% on some main terrace reefs. With the

98

B E R M U DA

establishment of the first Marine Protected Areas in 1966, legislative enactment of the whole island as a coral reef preserve in 1972, and the banning of fish traps in 1990, Bermuda has generally been ahead of the game in marine resource management. SEE ALSO THE FOLLOWING ARTICLES

Caves as Islands / Coral / Hurricanes and Typhoons / Mangrove Islands / Marine Protected Areas / Seamounts, Geology FURTHER READING

Anderson, C., H. De Silva, J. Furbert, A. Glasspool, L. Rodrigues, W. Sterrer, and J. Ward. 2001. Bermuda biodiversity country study. Bermuda Zoological Society. Curran, H. A., and B. White, eds. 1995. Terrestrial and shallow marine geology of the Bahamas and Bermuda. The Geological Society of America Special Paper 300. Boulder, CO: The Geological Society of America. Iliffe, T. M. 1993. A review of submarine caves and cave biology of Bermuda. Boletino Sociedad Venezolana de Espeleologia 27: 39–45. http://www.tamug.edu/cavebiology/Bermuda/BermudaIntro.html Morris, B., J. Barnes, F. Brown, and J. Markham. 1977. The Bermuda marine environment. Bermuda Biological Station Special Publication #15. Sterrer, W. 1998. How many species are there in Bermuda? Bulletin of Marine Science 62: 809–840. Sterrer, W., A. Glasspool, H. De Silva, and J. Furbert. 2004. Bermuda—an island biodiversity transported, in The effects of human transport on ecosystems: cars and planes, boats and trains. J. Davenport and J. Davenport, eds. Dublin: Royal Irish Academy, 118–170. Thomas, M. 2004. The natural history of Bermuda. Bermuda Zoological Society. Vesey, T. 2002. When disaster struck. The Bermudian Sept.: 10–21. Vogt, P. R., and W.-Y. Jung. 2007. Origin of the Bermuda volcanoes and Bermuda rise: history, observations, models and puzzles, In Plates, plumes and planetary processes. G. R. Fowler and D. M. Jurdy, eds. Geological Society of America Special Paper. 430: http://www.mantleplumes .org/P%5E4/P%5E4Chapters/VogtBermudaP4AcceptedMS.pdf Wingate, D. B. 1997. The pre-colonial and current status of Bermuda’s seabird population. El Pitirre 10: 36–37.

BIOLOGICAL CONTROL MARK GILLESPIE AND STEVE WRATTEN Lincoln University, Christchurch, New Zealand

Biological control (or biocontrol) is the use of natural enemies to suppress pest species populations to less damaging densities. When certain native and introduced invertebrates, plants, pathogens, and vertebrates increase in abundance and become pests through human influence or through other causes, economic crop damage and threats to natural resources are likely. Islands are particularly vulnerable to pest outbreaks. With high endemism, low species diversity, small land areas, and a history less affected by forces that develop adaptability compared to continents, islands are more susceptible to the effects of habitat changes and species introductions. The enhancement of the efficacy of the natural enemies of pest organisms is a potentially more environmentally sound and sustainable control option than chemical or mechanical control strategies. THE IMPORTANCE OF PESTS

The presence of pest animal and plant species can heavily impact agricultural, forest, and urban ecosystems, costing national economies billions of dollars in management and lost revenue, compromising biosecurity internationally, and threatening natural ecosystems and endangered species. Weeds alone are estimated to cost the New Zealand economy NZ$100 million annually, for example. However, the dairy industry in that country has an annual economic value of NZ$7 billion, so in the short term at least, weeds could be considered to be a “minor” problem. There are two main causes of pest outbreaks: the structure of agricultural systems and deliberate or accidental introductions of non-native species. In many agricultural systems, the selection of crops for rapid growth at the expense of defense mechanisms such as herbivore resistance and the planting of large uniform monocultures bereft of biodiverse non-crop habitats can disproportionately benefit herbivores, unwanted plants, and pathogens over their natural enemies, such that they reach economically damaging densities. It is estimated that about 37% of global crops are lost to native and introduced pests. Non-native species, introduced to new areas either deliberately or accidentally, are also one of the biggest threats to indigenous flora and fauna. These adventive

species often arrive without their coevolved natural enemies, and many are then able to thrive in an enemy-free environment. They may even change fundamental ecological properties such as water availability or soil chemistry, resulting in a further loss of competitiveness among natives. Before the advent of global transport and commerce, the arrival of introduced species is likely to have been once every 35,000 years in Hawaii, for example. Between 1962 and 1985, there were on average 19–20 immigrant species a year entering that island group, 3.5 of which became pest species. The Issues for Islands

This article focuses on the issues facing the more remote islands such as those in the Pacific. Although pest origins and control principles are similar in all parts of the world, the characteristics of the more remote islands make them particularly vulnerable to both the damaging impacts of exotic pests and potential nontarget effects of the introduced biocontrol species (the agent). Islands that have been geographically and evolutionarily isolated for millions of years tend to accommodate large proportions of endemic species, low species diversity, restricted genetic diversity, low natural immigration rates, and narrow home ranges of species, compared with larger land masses. During their geological development, they have not faced the effects of mammalian herbivore browsing and virulent diseases that are routinely encountered on continents. Because of these factors, natural enemies of invasive species are less likely to be present on such islands, and other affected species are less likely to adapt to and compete with new arrivals. Intense human activities on the smaller land areas of islands can also marginalize native species by reducing the availability of refugia. With low adaptability, restricted ranges, and sudden occurrence of hitherto absent competitive or predatory species, islands are at greater risk of losing globally important endemic species than are continents or less fragile islands. An example of this vulnerability is the arrival of mammals in New Zealand. Before Polynesian and European settlement, the New Zealand archipelago held unique species assemblages that had evolved in isolation for over 80 million years. When humans arrived, they deliberately or inadvertently brought rats, livestock, cats, domestic dogs, mustelids, and plants and irrevocably altered unique habitats. Such rapid changes were unlike any the native flora and fauna had been faced with before, and species such as ground-dwelling, flightless birds with

BIOLOGICAL CONTROL

99

ineffective mammalian predator defense strategies were easy prey. As a result of this and widespread hunting, many of New Zealand’s archaic and mammal-niche species were driven to extinction, and the new, enemy-free species thrived. With increased global travel and current tourism trends toward “island paradises,” the numbers of introduced species are reaching high levels, a situation that further increases the need for control. However, the above factors are also applicable to the introduction of natural enemies to control the pests, and examples given later of agents impacting upon nontarget species demonstrate the fragility of islands and the difficulty of achieving sustainable pest control. The Limitations of Nonbiological Control

The negative effects of pesticides have been documented comprehensively, from adversely affecting nontarget natural enemies and other beneficial organisms to accumulating in soil, water, and even humans. Of particular concern is the capability of many pests to become resistant to pesticides. Many of them can also “create” pests by killing the natural enemies of previously harmless species, allowing them to thrive as “secondary pests.” In many cases, chemical and mechanical strategies such as manual removal are not desirable because of the fragility of island flora and fauna, or are not possible because of the cost and labor requirements of such methods. Concerns about these control methods have led many countries to seek ways to reduce chemical use, and biocontrol is billed as a technique that can help to limit the use of more damaging or inappropriate control methods. THE PRINCIPLES OF BIOLOGICAL CONTROL

Although biological control occurs naturally everywhere, it is the human manipulation of the processes that forms the techniques we call biocontrol. These were first utilized in China and Yemen thousands of years ago, but they have gained popularity only since the turn of the nineteenth century, alongside advances in scientific knowledge of the ecological processes involved. Since then, many programs have been attempted with the aim of developing a sustainable and cost-effective technology. The most common targets of biocontrol are insects and weeds, although vertebrates, other invertebrates such as mites and snails, and pathogens are just as problematic and are currently receiving much attention in the scientific literature. The agents, natural enemy organisms that can be used in biological control programs, also vary

100

BIOLOGICAL CONTROL

widely and include invertebrates, vertebrates, and pathogens, with parasitic wasps that attack insect pests being the most common type of program. Broadly, biocontrol considers that a contribution to solving pest problems rests with conserving natural enemies and/or reconnecting disrupted food webs. Three general applications of this theory have evolved over the last 120 years. Classical biocontrol, the introduction of one or more appropriate indigenous and/or exotic natural enemies to pest-infested areas, is the oldest and most common method. This method also carries the most risk, because, although trials are made to ensure the agent has no nontarget effects, some effects are unpredictable. Augmentation biocontrol, largely a commercial industrial technology and application, aims to increase the abundance of natural enemies already present in low densities or arriving late relative to the pests, through timely releases either inundatively or by innoculation of smaller numbers of natural enemies into the cropping system. Finally, conservation biocontrol, a more recently recognized method, attempts to conserve natural enemies by rectifying negative influences on these beneficial organisms. These influences may include pesticide use and the removal of natural-enemy overwintering sites. This method is often employed as part of an integrated pest management program or following an initial classical introduction of natural enemies in the case of exotic pests. In successful programs, the agents will continue to provide control after establishment or conservation without the need for continued management and will also persist when pest densities are low, greatly reducing both the effort and cost of the control method. Similarly, both the pest and the introduced specialized enemy ideally decline in abundance over time, reaching densities similar to those in the pest’s home range; preferably, a community structure is achieved that is similar to that existing before the pest invasion. BIOLOGICAL CONTROL ON ISLANDS The Benefits of Successful Biological Control

When biocontrol works, it is self sustaining, nonpolluting, and cost-effective. In Hawaii alone, programs have contributed to the control of over 200 agricultural pest species, saving millions of dollars and tons of pesticides annually. Worldwide, successful introduction programs number over 700 (with more than 200 providing complete control and over 500 giving substantial or partial control). Economic analyses of these programs suggest that benefit-cost ratios for successful programs can exceed 145:1. A well known example is the control of the

floating fern, Salvinia molesta (Fig. 1), which was introduced to Papua New Guinea from Brazil as an aquarium plant and botanical curiosity. By the 1970s, the fern had established sufficiently to spread over the surface of the Sepik River in a mat up to 1 m thick, preventing navigation, killing submerged vegetation, and isolating villages. In 1978, the introduction of a weevil from the fern’s home range led to effective control and rid the river of the fern permanently, delivering control at a moderate cost when all other methods were considered unfeasible.

FIGURE 2 The glassy-winged sharpshooter, Homalodisca coagulata,

alongside native species of cicadellid. The Tharra sp. (middle) is the largest species native to Tahiti, whereas the Nesophyla sp. (left), is an average sized native cicadellid in Tahiti. (Courtesy of J. Grandgirard and J. Petit, Gump Station, University of California, Berkeley).

The Limitations of Biological Control FIGURE 1 A farm pond covered in the aquatic weed Salvinia molesta.

Such infestations can kill submerged flora and fauna and threaten livelihoods. (Courtesy of Ted D. Center, USDA Agricultural Research Service.)

Success has also recently been seen in French Polynesia, where a bug, the glassy-winged sharpshooter, Homalodisca coagulata (Fig. 2), is a major pest of agricultural, ornamental, and native plants, reaching densities up to 1000 times greater than in its home range in the southeastern United States and Mexico. However, biocontrol scientists identified a suitable egg parasitoid from the home range of H. coagulata, and in September 2005, 14,000 parasitoids were introduced to 27 sites on Tahiti. By December 2006, a 99% decrease in H. coagulata nymph densities was recorded. To date, both pest and natural enemy remain at low densities on every island to which they were introduced. In both these cases, a high degree of scientific expertise was employed, strict biocontrol standards were adhered to, and measures were taken to ensure that the pest control agent did not impact native species. The selection of the most appropriate natural enemy to introduce is not as straightforward as it may seem however, and past errors have detracted from successes.

Biological control has thus far not been the panacea that the early pioneers had hoped. There have been some attempts that have had disastrous consequences over the years, which some critics claim have led to nontarget effects as serious as species extinctions. The rosy wolf snail, Euglandina rosea, for example, was introduced to Hawaii in the 1950s after the government responded under pressure from residents to control the invasive giant African land snail, Achatina fulica, despite incomplete biocontrol trials and the fact that the snails do not share the same home range. Although A. fulica did decrease in abundance following the introduction of E. rosea, reasons other than the introduction of biocontrol agents have been posited. In contrast, there is ample evidence that E. rosea negatively impacted many native snail species, to the extent that they caused extinctions of several endemic species. Despite this evidence, E. rosea was subsequently introduced to the Society Islands for the same reason and with the same devastating impact on native Partula snails. In the 1920s, a moth species, Levuana iridescens, was heavily damaging coconut production on Fiji. Entomologists searching outside Fiji failed to locate L. iridescens, and therefore its natural enemies, and instead selected and imported a parasitic fly from Malaysia that attacked a similar host species. Levuana iridescens was brought under control

BIOLOGICAL CONTROL

101

rapidly and has been cited as a leading example of biocontrol. However, recent work argues that L. iridescens was endemic to the Fijian archipelago and that the agent introduced for control may have driven it to extinction. The matter is as yet unresolved, but it illustrates the fragility of island species and the risks involved with biocontrol programs. When programs have drifted from the fundamental principles of biocontrol (e.g., natural enemies should be highly host specific), nontarget effects have often occurred that could have been avoided. Badly conceived projects fail largely because inappropriate natural enemies are selected. Generalist predators are inappropriate, for example, because of their adverse effects on native organisms and lack of “fidelity” to the target pest. In some cases, groups other than skilled technicians have carried out the control with no scientific grounding or government oversight, often in order to sidestep rigorous procedures to achieve “quick fixes.” Alternatively, projects were carried out before the implementation of government regulation of control agent importation. Many nontarget effects are actually also open to controversy because direct evidence linking the exotic natural enemy to the decline in indigenous organisms is often not available. For example, the parasitoid wasp Pteromalus puparum, which was introduced to New Zealand in 1933 to control the cabbage white butterfly, Pieris rapae, also attacks the endemic Red Admiral butterfly, Bassaris gonerilla. However, a self-introduced parasitoid, Echthromorpha indicatoria, has been shown to have a greater impact on B. gonerilla abundance than does the introduced biocontrol agent. Although poor planning and technical knowledge may be blamed for many inappropriate biocontrol programs, even the strictest program can suffer from risk and unpredictability. Importantly, though, problematic control attempts have taught lessons and led to the guidelines that most countries now follow when embarking upon a biological control program. Criticism of the adverse affects of classical control has also led to guidelines set down by the Food and Agriculture Organization of the United Nations (FAO). New Zealand and Hawaii, for example, currently have the some of the most rigorous national legal regulations for the importation of potential biological control agents, and this is designed to limit such risks, although they cannot be eliminated. FUTURE PROSPECTS FOR BIOLOGICAL CONTROL TECHNOLOGY

The risks of nontarget impacts need to be considered alongside the risks of inaction and those of pesticide use on insecticide resistant pests. The risks of biocontrol are

102

BIOLOGICAL CONTROL

usually taken seriously in today’s environmentally aware social climate. In the years since strict quarantine screening and risk-benefit analysis were implemented, the safety record of biocontrol science has been particularly strong. For example, although a study in Hawaii found that 83% of parasitoids reared from native moths were biological control agents with parasitism reaching 28% in some species, all of those agents were species released before 1945. This suggests that recent guidelines have been effective. The future of biological control is, however, hampered by a lack of three things: research funding, continuity of effort, and a long-term emphasis. Pacific islands for example, make up only 2% of the Pacific Ocean; have great differences in flora between them; and have low population densities, a low proportion of commercial agriculture, and a certain reliance on foreign aid. As a result, funding agencies are reluctant to contribute to such apparently low-benefit causes, overlooking the conservation value of these islands and the fact that collaboration with a developed country is important if control and quarantine of new species is to be successful. Indeed, biological control is often the only economically viable technology in such regions. Nonetheless, although young, the science of biocontrol is growing: Novel biocontrol techniques are being constantly explored, and new targets are being frequently studied. New approaches to thus-far difficult targets, such as feral cats, including sexual transmission of host-specific diseases and immunocontraception as means to reduce the fecundity of noxious vertebrates, are being actively pursued in New Zealand and Australia. This continued development and further international and interdisciplinary collaboration are vital if biocontrol is to fulfill the potential it promises. Furthermore, in today’s scientific and political environment, risks are likely to be minimized, and biological control will develop as a viable, safe, and sustainable solution to pest invasions of islands. SEE ALSO THE FOLLOWING ARTICLES

Extinction / Fiji, Biology / Hawaiian Islands, Biology / Introduced Species / Invasion Biology / New Zealand, Biology FURTHER READING

DeBach, P., and D. Rosen. 1991. Biological control by natural enemies. Cambridge: Cambridge University Press. Gurr, G., and S. Wratten. 2000. Biological control: measures of success. Dordrecht: Kluwer Academic Publishers. Hoddle, M. S. 2002. Restoring balance: using exotic species to control invasive exotic species. Conservation Biology 18: 38–49. Julien, M. H., J. K. Scott, W. Orapa, and Q. Paynter. 2007. History, opportunities and challenges for biological control in Australia, New Zealand and the Pacific Islands. Crop Protection 26: 255–265.

Messing, R. H., and M. G. Wright. 2006. Biological control of invasive species: solution or pollution? Frontiers in Ecology and Environment 4: 132–140. Van Driesche, R. G., and T. S. Bellows. 1996. Biological control. New York: Chapman and Hall. Wilson, K.-J. 2004. Flight of the huia: ecology and conservation of New Zealand’s frogs, reptiles, birds and mammals. Christchurch: Canterbury University Press.

BIRD DISEASE DAVID CAMERON DUFFY University of Hawai‘i, Manoa

Bird diseases have had a profound effect on some island avifauna, but, as with human diseases, our full understanding of their extent and impact has been limited by isolation, the paucity of qualified observers, and the possibility that many of the strongest effects have already occurred. Despite these limitations, research suggests that bird diseases on islands fall into two main, contrasting groups, both the result of human intervention: alien diseases attacking naïve, susceptible populations, with often devastating effects on an island’s species and ecosystems, and (more rarely) “emerging diseases,” resulting from human contact with endemic avian pathogens on previously isolated islands. DISEASE SUSCEPTIBILITY IN ISLAND BIRDS

An island’s size and isolation affect its avian disease ecology much as they determine its avifauna. Only a few birds reach more isolated islands, representing a subset of the source population’s genes, and they may also carry a reduced set of pathogens compared to the source pool of pathogens. With a small host population, birds and pathogens would undergo rapid selection for resistance and reduced mortality. At the community level, small avifaunas would theoretically also reduce the total pool of pathogens and the subsequent risk of inter-specific transfer of diseases between bird species. Unfortunately, the role of disease in organizing natural island bird communities in the absence of humans can probably no longer be explored because avifaunas have been changed so greatly by human activity. The lack of pathogen challenge may have led to reduced immunologic defenses in island birds, making them more vulnerable to new diseases, but this theory remains controversial. In theory, there are costs to immune defenses,

and, in the absence of a challenging disease environment, birds should redirect their resources to other activities such as reproduction. On the other hand, island birds remain in an evolutionary race with those diseases that are present, they may be able to adapt different components of their immune systems to the pathogen community, and changes in their immune systems may be limited by mutation and genetic drift. It does appear that island birds exhibit reduced immune functions following inbreeding during bottlenecks, when populations are reduced to small numbers, whether caused by anthropogenic damage, reintroduction, or perhaps even during natural colonization. The resulting combination of small populations and weakened immune systems would make populations vulnerable to extinction from disease. INTRODUCED DISEASES Acute Catastrophic Effects

The first human contact with any island has frequently been coupled with the introduction of alien pathogens, often with disastrous consequences to native ecosystems. Sailing vessels carried a menagerie of rats, cats, chickens, and mosquitoes, and their long voyages meant they had to stop at isolated islands to take on water and food, leaving ample opportunity for various hosts and vectors of pathogens to disembark. Such introductions and the resulting diseases often had devastating effects for island birds, much as the diseases of sailors themselves had for island peoples. The best examples of epidemics or epizootics are mosquito-borne avian pox virus and avian malaria (Plasmodium relictum) in the Hawaiian Islands. Although a mosquito vector was present from the early 1800s, it was probably not until the late nineteenth or early twentieth centuries that the two diseases devastated the native avifauna. One observer reported “scores” of forest birds “dead or dying” and others transformed with hideous pox lesions that in some cases left them unable to move. Although a few species remained unaffected, most retreated up the slopes of the Hawaiian volcanoes, above the range of the mosquito vector. In subsequent experiments, native birds developed severe pox and malaria, and many died. Although avian malaria and avian pox had catastrophic effects on Hawaii’s bird populations, there is some evidence that an “equilibrium” is forming between pathogen and host. Within a century of the introduction of avian malaria, two species in Hawaii show signs of developing resistance to the disease. Avian pox, initially associated with mortality in red-tailed tropicbirds (Phaethon rubricauda)

BIRD DISEASE

103

on Midway Island in 1963, continues to affect nestling Laysan albatross (Pheobastria [Diomedea] immutabilis) populations in Hawai‘i, but with little mortality, suggesting either a decrease in virulence by the virus or an increase in resistance by the birds. As alien diseases continue to be introduced to island ecosystems, other significant mortality events for island birds have included avian malaria in New Zealand on yellow-eyed penguins (Megadyptes antipodes) and in South Africa on African penguins (Spheniscus demersus); avian cholera and Newcastle disease on the guano islands of Peru (various seabird species); avian cholera, coccidiosis, and influenza A off South Africa (various seabird species); Newcastle disease on double-crested cormorants (Phalacrocorax auritus) in the Great Lakes and California; a strain of Salmonella on an endangered New Zealand bird, the hihi (Notiomystis cincta), and avian pox in Galapagos (various species). Avian pox has also infected two endemic land birds in the Canary Islands, Berthelot’s pipits (Anthus berthelotti) and short-toed larks (Calandrella rufescens), exhibiting 28% and 50% frequencies of lesions referable to pox, thus suggesting that pox is a serious threat to the two species. The route of transmission of disease to a new population is not always direct or obvious. An ornithosis outbreak in the 1930s in the Faeroe Islands and Iceland involving northern fulmars (Fulmarus glacialis) may have been triggered by avian scavenging on diseased parrots tossed overboard while being transported to Europe to be sold. The disease spread to islanders through consumption of young birds. The death rate among infected people was 20% (up to 80% in pregnant women), and the situation forced a ban on the traditional harvest of nestlings. In many cases, the causes of mortality events remain unknown or are suspected to be the result of a combination of factors, with disease sometimes playing only a secondary role. Chronic Effects or Their Absence

Not all new pathogens produce epidemics. Some pathogens may not have the best mode of transmission, or the social structures of bird species may not be conducive to spreading the disease, or bird species may not be phylogenetically vulnerable to a particular pathogen. On the other hand, it is possible that observers arrived only after a pathogen had lost its initial virulence or that the disease had sublethal effects that would not be obvious without study. Some diseases can be hosted in a carrier population of birds that remain unaffected even as they transmit the disease to other species. Campylobacter jejuni, a cause of

104

BIRD DISEASE

human bacterial enteritis, has been detected in Macaroni penguins (Eudyptes chrysolophus) on an island off South Georgia, occurring without apparent negative effects on the birds. Similarly, infectious bursal disease virus, which impairs the immune system in chickens, is present in Adelie penguins (Pygoscelis adeliae) in the Antarctic with no apparent clinical effects. Borrelia burgdorferi, the infectious agent for Lyme disease, has been found in a wide range of seabirds in both hemispheres, and antibodies to it have been detected in Faeroe Islanders who harvest seabirds, without any evidence of clinical symptoms in either birds or humans. Some pathogens cause devastation in some geographic areas while occurring without detriment in others. Avian malaria, despite its devastation elsewhere, appears to have no effect on the avifaunas of American Samoa, Bermuda, Moorea, and some South Pacific islands, suggesting the pathogen has been present for long enough for the avifaunas to evolve resistance to it or for any vulnerable species to have already gone extinct. ENDEMIC DISEASES

A long list of obscure avian pathogens, especially viruses, have been discovered on islands throughout the world, following what were usually only superficial surveys. After so much ship traffic, we cannot now always be sure which diseases are indigenous, just as we have no idea what diseases might have already vanished after exterminating their hosts. Much research has focused on pathogens that have the potential to affect humans, especially viruses associated with seabird-parasitizing ticks (Acari), the soft-bodied genus Ornithodoros in the tropics and the hard-bodied genus Ixodes in the temperate and polar regions. Whereas the ticks are known to cause anemia, exanguination, nest desertion, and resulting mass mortality of abandoned young, little to nothing is known of the effect of viruses on seabirds themselves. Apparent human illness associated with tick bites, but perhaps caused by viruses, has been reported from Arabia, the Seychelles, Peru, Morocco, and France, primarily in guano workers, farmers, and researchers. Finally, there is the enigmatic Laysan fever from the northwest Hawaiian Islands that attacked field biologists for a period and then apparently died out. Mosquito-borne viruses that use birds as hosts have the potential to cause widespread infection among humans, but endemic forms appear rare on islands. Whataroa virus, affecting humans in New Zealand with an influenzalike disease, now appears to be sustained by the introduced song thrush (Turdus philomelos), but it is spread

by endemic mosquitoes in a human-modified landscape with relatively few native birds. Its original host remains unknown, but unlike many islands, New Zealand had endemic mosquitoes, which could have coevolved with viruses. Another significant disease because of its pandemic potential is avian influenza. The normal hosts are waterfowl, shorebirds, and seabirds. Sampling for influenza antibodies in Australian seabirds shows variation between different species (wedge-tailed shearwaters Puffinus pacificus and black noddies Anous minutus) and local breeding islands, suggesting they may host local lineages of influenza.

Van Riper, C., S. G. Van Riper, M. L. Goff, and M. Laird. 1986. The epizoootiology and ecological significance of malaria in Hawaiian land birds. Ecological Monographs 56: 327–344. Wikelski, M., J. Foufopoulos, H. Vargas, and H. Snell. 2004. Galápagos birds and diseases: invasive pathogens as threats for island species. Ecology and Society 9:5. http://www.ecologyandsociety.org/vol9/iss1/art5/.

BIRD RADIATIONS JEFFREY PODOS AND DAVID C. LAHTI University of Massachusetts, Amherst

THE FUTURE

Our understanding of the effects of introduced bird diseases on island avifauna has begun to lead to protective measures such as the removal of alien-host rock doves (Columba livia) in Galapagos and vector mosquitoes (Culex quiquefasciatus) on Midway Island in the Pacific, the implementation of quarantine systems of varying degrees of competence, the sterilization of boots in the Antarctic, and the institution of bans on importation of high-risk hosts. Despite these efforts, diseases such as Newcastle disease, West Nile, and avian influenza remain ever ready to make an appearance, facilitated by today’s rapid forms of transportation and international trade. The potential for emerging diseases from islands also continues and may even increase as ecotourism brings travelers onto ever more remote islands and then rapidly back to sophisticated medical facilities where their ailments can be diagnosed as something beyond “fever.” The study of bird diseases on islands remains one of the last frontiers for exploratory biology. It is likely to continue to provide unexpected insights into immunology, public health, community ecology, and conservation biology, as additional islands, their avifaunas, and their diseases are examined. SEE ALSO THE FOLLOWING ARTICLES

Extinction / Honeycreepers, Hawaiian / Inbreeding / Seabirds FURTHER READING

Clifford, C. M. 1979. Tick-borne viruses of seabirds, in Arctic and tropical arboviruses. E. Kurstak, ed. New York: Academic Press, 83–99. Daszak, P., A. A. Cunningham, and A. D. Hyatt. 2001. Anthropogenic environmental change and the emergence of infectious diseases in wildlife. Acta Tropica 78: 103–116. Matson, K. D. 2006. Are there differences in immune function between continental and insular birds? Proceedings of the Royal Society B 273: 2267–2274.

Bird radiations provide informative illustrations of ecological and evolutionary processes, including those that help to generate biodiversity as ancestral species radiate into multiple descendent species. Adaptive radiation involves initial phases of divergence among populations or incipient species, accompanied by or followed by the evolution of reproductive isolation. In this article, aspects of these two key processes—divergence and the evolution of reproductive isolation—are outlined and examined with specific reference to island bird radiations. BIRD RADIATIONS ON ISLANDS: ECOLOGICAL AND EVOLUTIONARY INSIGHTS

Island habitats provide biologists valuable if not unique opportunities for the study of ecology, evolution, and animal behavior. Relative to continental habitats, islands tend to express low biological diversity, high abundance of constituent species, streamlined food webs, and environmental fluctuations that can be both marked and unpredictable. The simplified and dynamic profile of island biology has facilitated the study of a wealth of biological patterns and processes, with results gleaned from island studies often extrapolated to more complex and empirically less accessible habitats such as tropical rain forests. Birds have taken center stage in many field studies of island ecology and evolution, following a tradition set by David Lack in the mid-twentieth century. Bird radiations have been numerous and varied, featuring outcomes such as flightlessness and robust variations in plumage, size, and beak form and function (Table 1). This entry provides a brief overview of some ecological and evolutionary processes that drive island bird radiations and then discusses recent findings from the authors’ own studies of two island bird systems: introduced island populations of African weaverbirds, and Darwin’s finches of the Galápagos Islands.

B I R D R A D I AT I O N S

105

TABLE 1

Representative Examples of Bird Radiations on Islands Bird Assemblage

Islands

Distinctive Features of the Radiation

Vangas (15–22 species) and Malagasy songbirds (9+ species)

Madagascar

Moas (14 species)

New Zealand

Darwin’s finches (14–15 species)

Galápagos and Cocos islands

Myiarchus flycatchers (five species)

West Indies

Honeycreepers (∼52 species)

Hawaiian Islands

Golden whistler Pachycephala pectoralis (66 forms, perhaps can be grouped into 5 species) Chaffinches Fringilla coelebs (four forms, designated as subspecies)

Northern Melanesia and New Guinea

The vangas’ common ancestor presumably arrived on Madagascar about 25 million years ago and subsequently radiated into a group unusually diverse in both plumage and bill morphology. Another diverse group, Malagasy songbirds (previously placed into three different families of bulbuls, babblers, and warblers), represent a radiation from a single colonizing species 9–17 million years ago. At 587,000 km2, Madagascar is the smallest single island in the world with a prominent bird radiation. Beginning about 20 million years ago, these herbivorous and flightless birds evolved in the absence of predators, radiating mostly within the last 10 million years into a variety of body forms and sizes. A grassquit or warbler-like ancestral species colonized the Galápagos about 2–3 million years ago, followed by a pattern of island-hopping and speciation; descendant species eventually redistributed themselves throughout the islands without interbreeding. Bill morphology diverged by adaptation to available seed types, song subsequently diverged as a by-product; female preference for both traits is thought to have promoted reproductive isolation upon secondary contact. A single immigration to Jamaica from Central America about 4 million years ago, led to subsequent island-hopping and divergence. Several island-specific forms are designated as subspecies; thus this is apparently a radiation in process. Beginning about 4–5 million years ago, the descendants of a single colonizing finch species radiated explosively; highly diverse feeding methods and bill types were faciliated by a broad range of vacant feeding niches. Called the world’s “greatest speciator” by Mayr and Diamond (2001), this bird has diverged greatly in color patterns, but species boundaries are difficult to determine. Some forms arose by divergence but others through hybridization (mixing) of existing forms. DNA evidence has indicated the history of chaffinch expansion from an ancestral stock that ranges across Europe and Africa, to several nearby island groups: a Portuguese chaffinch population colonized the Azores, birds from the Azores colonized Madeira, and birds from Madeira then colonized the Canary Islands twice.

North Atlantic islands

Island Ecology COMPETITION AND CHARACTER DISPLACEMENT

Any given habitat supports a diversity of species. Species that share habitats are thought to partition limited sets of resources such as space and nutrients. Shorebirds, for instance, have evolved to forage in distinct shoreline zones. Moreover, resource partitioning appears to be adaptive; sandpipers and plovers, to illustrate, are adept in extracting distinct food sources at particular depths within the sand or mud, by means of distinct bill lengths and shapes. Adaptive partitioning of resources among present-day species implies two historical processes: competition for limited resources, followed by selection favoring the occupation of divergent niches (“character displacement”). These processes are thought to be highly generalized, explaining, for example, how tropical rainforests can support a rich biota. However, these processes have proven notoriously difficult to study in field environments.

106

B I R D R A D I AT I O N S

A clear demonstration of competition leading to character displacement has been reported by Peter and Rosemary Grant, who have been conducting field studies on Galápagos finches for over three decades. During an initial decade of study, beginning in 1973, the small central island of Daphne Major was found to support a large breeding population of the medium ground finch Geospiza fortis. This population featured birds of a wide range of beak sizes feeding on a diverse array of seed types, including the relatively hard seeds of caltrop Tribulus cistoides. Observations of feeding finches revealed that only the largest-beaked G. fortis were able to crack Tribulus seeds, a capability that served these large-beaked birds well during a severe drought in 1977. With exclusive access to this food resource, large-beaked G. fortis were able to survive and proliferate. In 1982, however, a breeding population of the large ground finch (Geospiza magnirostris) began to take root on Daphne, building slowly and peaking at about

350 individuals in 2003. G. magnirostris also feeds on the seeds of Tribulus, and thus introduced a competitive threat to large-beaked G. fortis. The effects of interspecies competition were realized during an intense drought in 2003, during which time the population of G. fortis crashed as a result of widespread starvation. Subsequent analysis revealed that large-beaked G. fortis had suffered disproportionately, presumably because of the depletion of the favored Tribulus seed resource by G. magnirostris. Smaller-beaked G. fortis survived and bred in disproportionate numbers, thus leading to an evolutionary reduction in G. fortis beak size in the next generation. The character of beak size had thus been displaced through competition from another species, through selection favoring those individuals that could avoid direct interspecies competition. ISLAND BIOGEOGRAPHY

Given evidence of resource partitioning through competition and displacement, it follows that habitats with more ecological niches might support relatively greater levels of species biodiversity. This idea has received particular attention in the theory of island biogeography, which focuses on “rules” that might govern the diversity and evolution of species on terrestrial islands. A first rule is that large islands are expected to support more species than small islands, because of a greater abundance of ecological niches, resulting not only from their larger areas but also from broader elevational gradients. A tenfold increase in the size of an island is expected to lead to an approximate doubling of species number. A second rule of island biogeography is that more remote islands should be relatively depauperate, because chance immigrations of new species are less likely when mainland source populations are more distant. Larger and more isolated islands are also expected to support greater numbers of endemic species, because of increased opportunities for genetic isolation. Island species diversity can of course be influenced by other factors including the presence or absence of predators, the geological age of islands, the extent and diversity of appropriate habitats, and factors that can affect animal movement such as storm paths and migration routes. In modern times, humans have also imposed profound impacts on island species diversity through activities such as hunting, habitat conversion, and species introductions. Birds on the islands of the Indian Ocean well illustrate some of the principles of island biogeography, particularly those related to endemism. Of fourteen islands, endemic species occur in much greater frequency on the larger islands. The two smallest islands (< 100 km2 area)

support no endemics, whereas the three largest islands (> 1000 km2) support over a dozen each. A parallel pattern is observed across the four inhabited Comoro Islands. The largest of these islands, Grande Comore, supports 14 endemic species, whereas the other three islands support only two or three endemics each. The Comoro Islands also highlight the importance of elevation to bird diversity; all of the Grande Comore endemics can be found on its Mount Karthala, the only mountain of the archipelago. The Indian Ocean Islands also illustrate a relationship, albeit more subtle, between endemic species and degrees of island isolation. Very small and distant islands have no endemics at all, as predicted given that they are only seldom visited and that any occasional founding population is likely to go extinct. Of islands with at least one endemic species, the more remote ones tend to have higher proportions of endemics. For instance, Rodrigues, Mauritius, and Réunion make up the Mascarenes, the most remote archipelago in the Indian Ocean. Not surprisingly, none of the birds native to these islands are found anywhere else. The small, isolated island of Rodrigues illustrates both predictions of low total species and a high proportion of endemism. Before it was inhabited by humans, 13 landbird species lived on the island, 12 of which were endemic. Today there are only two native landbird species on the island, both of which are endemic (the Rodrigues fody and the Rodrigues warbler). The sharp decline in species on Rodrigues highlights the often overwhelming influence of human impacts, which usually affect species diversity to at least as great a degree as island biogeographic processes and thus interfere with our ability to detect those processes. The Indian Ocean islands are unusual in that anthropogenic extinctions and introductions apparently have not yet obliterated the effects of island size and distance. Evolution on Islands POPULATION DIVERGENCE

Populations that colonize island habitats can undergo rapid evolutionary change, through any number of evolutionary mechanisms. Colonizing populations are small (normally comprising a small fraction of individuals from source populations) and thus normally experience severe population bottlenecks, with subsequent trajectories of evolution altered by founder effects. Founder effects may be sustained in “island hopping” scenarios, because of multiple sequential bottlenecks. Small population sizes also amplify the effects of genetic drift, which can theoretically result in the elimination or the fixation of genetic traits. Evolutionary change may also be facili-

B I R D R A D I AT I O N S

107

tated by the fact that island habitats often differ substantially from mainland habitats in resource structure and competition regimes. The genetic profiles of different populations can diverge even as they adapt to similar environmental challenges, because evolutionary changes in phenotypes can be achieved through distinct, parallel genetic modifications. A well-known, comprehensive illustration of adaptive divergence is found again in the research of the Grants and colleagues. Comparative and phylogenetic analysis of Galápagos finches and sister taxa suggests that the common ancestor of Darwin’s finches was relatively smallbodied and small-beaked, akin to present-day warbler finches. The radiation of Darwin’s finches toward mostly larger-beaked forms reflects a general trend in island radiations toward larger beak sizes, which may be powered both by “release” from competition in low-diversity environments and by a relative abundance of hard foods in ocean island habitats. The microevolutionary process of adaptation through natural selection has been documented several times for the Daphne finch populations. During the severe drought of 1977 (mentioned previously in the discussion of competition and character displacement), large-beaked members of the Geospiza fortis population held a relative survival advantage in being able to crack Tribulus seeds, which were available in quantity at that time. The softer, smaller seeds on which smaller-beaked G. fortis relied had become depleted, both because of a lack of rain and new seed growth and because of high demand for seeds given the relatively large G. fortis population resident on Daphne at the time. As the G. fortis population crashed, large-beaked G. fortis thus survived to breed in greater numbers, and the subsequent generation exhibited larger beaks than their parental generation. REPRODUCTIVE ISOLATION

Divergence is a prelude to radiation. In the opinion of many biologists, the currency of radiation—the generation of new species—is achieved only when divergence causes populations to become reproductively isolated from each other. Consider an ancestral population that colonizes multiple island habitats. If offshoot populations remain geographically isolated, they will accumulate, via divergent trajectories of drift or selection, differences in genetic and phenotypic traits. On secondary contact, insufficient reproductive isolation among offshoot populations would result in cross-breeding and reunification of gene pools, thus obstructing speciation. Alternatively, limited cross-breeding success would prevent mixing of

108

B I R D R A D I AT I O N S

gene pools and thus foster continued interpopulation divergence. In many groups of birds, reproductive isolation is driven in large measure by the divergence of mating signals and mate recognition systems. As mating signals diverge among populations, individuals are less likely to mate erroneously with members of other populations. Hybrid offspring, if viable, may suffer disadvantages in survival and reproduction, thus depressing the fitness of the hybrid’s parents. Selection is thus thought to favor individuals that breed within-population, a pressure that in turn favors the divergence of mating signals. In a recent review of this literature, Trevor Price finds that island bird populations show unusually wide divergence from both ancestral and sister species in signals used in mate and species recognition, including color patterns, feather ornaments, and vocal structure. Reproductive isolation of diverging island birds has been documented by Peter Ryan and colleagues for Neospiza buntings of the Tristan da Cunha archipelago. Two species of bunting, Neospiza acunhae and N. wilkinsi, are recognized on two islands, Nightingale and Inaccessible. On Inaccessible Island, the two species interbreed regularly, as indicated by an abundance of hybrids within a specific habitat (coastal tussock grass) favored by both species. On Nightingale Island, by contrast, reproductive isolation appears to be complete, with the two species forms showing highly distinctive body and beak forms and habitat preferences. Genetic analyses suggest that the diversity of subspecies in this group of birds has emerged through parallel trajectories of within-island interpopulation divergence, with reproductive isolation building up over time within each island, but not yet having reached completion on Inaccessible Island. Divergence among these species appears to have been facilitated not only by habitat segregation but also by the evolutionary divergence of mating signals. FOCUS EXAMPLE: EVOLUTION OF VILLAGE WEAVERBIRDS ON ISLANDS

The village weaverbird Ploceus cucullatus, a species that is common across subsaharan Africa, has been introduced to a number of different islands worldwide (Fig. 1). In the 1700s village weaverbirds were introduced from West Africa to the Caribbean island of Hispaniola (presentday Dominican Republic and Haiti). In the 1880s weaverbirds from South Africa were introduced to Mauritius and Réunion, two islands in the Indian Ocean. In the 1980s the village weaverbird also became established on the Caribbean island of Martinique. The Cape Verde and

H

5 Ma)

Estación Volcanológica de Canarias, Tenerife, Spain

The Canary Islands, located between 100 and 500 km from the coast of northwestern Africa (Morocco), consist of seven major volcanic islands forming a rough west-southwest to east-northeast trending archipelago. Together with the Selvagen Islands and a group of seven major seamount complexes (some of which were former

Lars (68 Ma)

Porto Santo (14 Ma)

20

00

Anika (55 Ma)

Canary Volcanic Province

Dacia (47 Ma) Selvagens (30 Ma)

Hierro (1 Ma)

Nico Conception Bank 10

Tenerife (12 Ma)

La Palma (4 Ma)

JUAN-CARLOS CARRACEDO

Ampère & Coral Patch (31 Ma)

Unicorn (27 Ma)

Lanzarote Fuerteventura (24 Ma)

Gomera (9 Ma)

30˚N

00

Africa

Gran Canaria (15 Ma)

FIGURE 1 Bathymetric map showing the Canary (red) and Madeira

(blue) volcanic provinces, including islands and associated seamounts, in the eastern central North Atlantic. Thick dashed lines mark centers of possible hotspot tracks. For clarity, only depth contours above 3500 m are shown. Bathymetric data from Smith and Sandwell (1997); ages and location of the Azores–Gibraltar fracture zone from Geldmacher et al. (2005) and Guillou et al. (1996).

CANARY ISLANDS, GEOLOGY

133

80

Lr

70 Shield Stage

60

A

Age (Ma)

Late Stage

50

C

Plate velocity = 12 mm/yr

40

Lz S

30

F

D

?

?

GC

20

G

10

H

T

LP

0 0

100

200

300

400

500

600

700

800

Distance from Hierro Along Hotspot Track (km) FIGURE 2 Radiometric ages of shield stages and late (rejuvenated

or posterosional) stages of magmatic activity on islands and seamounts in the Canary volcanic province versus distance from Hierro. The center of each island is projected onto the proposed curved hotspot track in Fig. 1 along a line perpendicular to the hotspot track. Distances were measured from Hierro along the proposed hotspot track. The regression line calculated for oldest available ages of shield-stage volcanism gives an average rate of plate motion of 12 mm/year. On older islands and seamounts, volcanic units that had low

206

Pb/204Pb

(≤ 19.5) were assigned to the late stage of volcanism, with the exception of the single sample from Lars seamount, which had a slightly lower

206

Pb/204Pb (19.44). Abbreviations for islands and seamounts:

H = Hierro, LP = La Palma, G = Gomera, T = Tenerife, GC = Gran Canaria, S = Selvagen Islands, F = Fuerteventura, Lz = Lanzarote, C = Conception seamount, D = Dacia seamount, A = Anika seamount, and Lr = Lars seamount. References for age data from Guillou et al. (1996) and Geldmacher et al. (2005).

destruction through mass wasting (e.g., landsliding) and erosion. The constructive phase occurs primarily during the shield-building (or shield) cycle of activity, during which eruptive rates are high, and most of the volcanic edifice is formed. Even though mass wasting is an important process during the shield stage, the volcano continues to increase in size, despite short-term setbacks. The constructive phase of island/volcano evolution can extend into the first late (also commonly referred to as post-erosional or rejuvenated) cycle of volcanism, during which volcanic eruptive rates are drastically lower, but magmas can be more evolved (silicious and thus more viscuous), contributing to an increase in volcano height. Late cycles of volcanism are generally separated from the shield stage of volcanism by extended periods of volcanic inactivity or drastically reduced activity. During the destructive phase of evolution, mass wasting and erosion outpace volcanic growth, and the volcanoes (islands) decrease in size until they are eroded to sea level. As the plate moves away from the magma source, it cools and subsides, and the now flat-topped volcanic edifices sink beneath sea level, forming guyots. Despite the differences in age of the volcanoes,

134

CANARY ISLANDS, GEOLOGY

all of the islands have had Holocene acitivity except La Gomera and Fuerteventura. The islands of La Palma, Tenerife, and Lanzarote have had historical volcanic activity, and thus youthful volcanic structures can be found across the entire archipelago, even on the oldest islands. The two youngest and westernmost islands of Hierro (1500 m above sea level, with oldest dated subaerially erupted rocks at 1.2 million years) and La Palma (2426 m; 1.8 million years old) have been the most active within the Holocene. Both volcanic islands are characterized by mafic alkaline volcanism, high eruptive rates, and volcanism along magmatic rift zones, commonly associated with the shield cycle of volcanism on ocean island volcanoes. Both volcanoes (and associated islands) are expected to continue to grow in size in the future. The three central islands, intermediate in age, were in their shield stage in the Middle to Late Miocene and have had low levels of late or rejuvenated volcanism in the Pliocene and/or Quaternary. Tenerife in the central western part of the archipelago forms the largest island and is also the third largest and highest (more than 7000 m in elevation above the sea floor and 3718 m in elevation above sea level; 11.9 million years old) volcanic structure on Earth after the Hawaiian volcanoes of Mauna Loa and Mauna Kea. The highest peak of Tenerife is formed by the highly differentiated (phonolitic) Teide volcano, nested in a lateral collapse caldera (formed through mass wasting), indicating that this volcano is at the transition from its constructive to destructive phase. Considering the significantly lower eruption rates of the Canaries as compared to the Hawaiian volcanoes, the similarity in size reflects Tenerife’s older age (longer life-span), most likely related to the almost-order-of-magnitude slower motion of the plate beneath the Canary Islands (∼12 mm/year) as compared to the motion of the plate beneath the Hawaiian Islands. Therefore, it has taken much longer for Tenerife to move away from its magmatic source than has been the case with the Hawaiian volcanoes. Although crudely round in outline, the central volcanoes of La Gomera (1487 m; 9.4 million years old) and Gran Canaria (1950 m; 14.5 million years old) no longer have conical shapes and are characterized by deeply incised canyons, indicating that these two volcanoes are well into their destructive phase of evolution, with erosion and mass wasting outpacing growth through magmatic activity. Erosion has exposed intrusive complexes and dike swarms on both islands. The compositions of volcanic rocks are highly variable, ranging from mafic (transitional tholeiite to melilite nephelinite) to highly evolved (peralkaline rhyolite to

trachyte to phonolite), reflecting the mature nature of these volcanoes. On the two easternmost, oldest, and lowest islands of Fuerteventura (807 m; 20.2 million years old) and Lanzarote (670 m; 15.6 million years old), erosion is clearly the main process shaping the morphology of the islands, even though both islands have had rejuvenated volcanism within the last 150,000 years, and Lanzarote even within historical times. Both islands are realtively flat, with little of the original shield volcano morphology being preserved. Instead, they are characterized by isolated older volcanic sequences or erosional remnants and broad valleys. Surprisingly, the largest historical eruption in the Canary Islands and the second largest basaltic fissure eruption ever recorded (after the 1783 Laki eruption on Iceland) was the 1730–1736 Timanfaya eruption on Lanzarote, which produced ∼1 km3 of primarily tholeiitic basalts. AGE OF VOLCANISM

The age of volcanism in the Canary Islands is well known from radiometric age dating. More than 600 K/Ar, Ar/ Ar, and 14C ages have been obtained by different groups on igneous samples from the islands. There have been a number of problems, however, primarily in dating older volcanic rocks and the uplifted portions of the seamount formations, some of which are clearly affected by alteration. Samples with excess Ar have produced ages that are too old, artifically increasing the duration of the basaltic shield stage on the islands. Ages obtained by newer techniques, such as Ar/Ar step heating, single crystal laser age dating, and the K/Ar unspiked method, and the employment of stringent age-dating requirements (e.g., sampling from well-controlled stratigraphic sections, and performing replicated analyses and systematic comparison of the palaeomagnetic polarities of the samples with the currently accepted geomagnetic reversal timescales) demonstrate that there is a progression of increasing age of the shield stage of volcanism from west to east in the Canary archipelago and from southwest to northeast in the Canary volcanic province, even though most of the Canary volcanoes have had a very long, complicated history, often including multiple cycles of late-stage volcanism (Figs. 1 and 2). VOLCANIC HAZARDS

Geological hazards are moderate in the Canary Islands compared, for instance, with the Hawaiian Islands, which have a similar area and population but much more frequent and intense volcanic activity and seismicity. Although high magnitude eruptions (plinian) occurred

in the geological past of the Canarian archipelago, only moderately explosive activity (strombolian to subplinian) took place in the last 200,000 years. Holocene eruptions, predominantly basaltic fissure eruptions, occurred on all the islands except La Gomera and Fuerteventura. Most of these, however, have been located on La Palma, El Hierro, and Tenerife, with only 10–12 events on Gran Canaria during this period and two on Lanzarote (1730–1736 and 1824). The most recent eruption in the Canary Islands was in 1971, from the Teneguia volcano at the southern tip of La Palma. During the Holocene, phonolitic strombolian to subplinian eruptions were associated with lava dome growth in the Teide volcanic complex on Tenerife, and to a lesser extent on the Cumbre Vieja rift on La Palma. No casualties have been reported in the 16 eruptions recorded after the colonization of the archipelago at the end of the sixteenth century. Reliable prediction of when future eruptions will occur is not feasible because of the low frequency of events and great variability of inter-eruptive periods, from a few years to several hundred years (e.g., on Cumbre Vieja, the most active volcano in the historic epoch, repose periods varied from 26 to 237 years). GEOCHEMICAL OVERVIEW OF ISLAND EVOLUTION

The geochemistry of the volcanic rocks from the Canary Islands is well understood in the context of volcano evolution. During each growth cycle (shield and late cycles), highly silica undersaturated rocks (nephelinites and basanites) dominate the early stage, more silicasaturated basaltic melts (alkali basalts and transitional thoeleiites) are produced during the peak of the cycle, and both mafic and evolved alkalic volcanic rocks (alkali basalts–basanites–nephelinites to trachytes–phonolites) are erupted during the waning stages. The almost complete lack of tholeiitic rocks and the low eruption rates during shield-cycle volcanism on the Canary volcanoes is likely to reflect a combination of low sublithospheric upwelling (mass flux) rates of the Canary plume and deep depths of melting beneath the thick Jurassic oceanic lithosphere. Although there are no systematic differences in major and trace element composition, systematic variations in isotopic composition do exist, with shield-stage volcanism being generally characterized by higher Pb and Sr and lower Nd and Hf isotopic ratios as compared to late-stage volcanism. These differences are likely to reflect greater amounts of melting of depleted upper or plume type mantle during the late stages of

CANARY ISLANDS, GEOLOGY

135

volcanism, as compared to enriched plume material during the shield stage of volcanism. GEOLOGY OF THE INDIVIDUAL CANARY ISLANDS El Hierro

The youngest and smallest of the Canary Islands is formed by three overlapping Quaternary stages of primarily mafic alkaline volcanism from oldest to youngest: Tiñor, El Golfo, and Rift (Fig. 3). A prominent threebranched rift system and related arcuate lateral collapse embayments (landslide scars) between the rifts define the characteristic shape of the island. Four lateral collapses have been recognized on El Hierro. The Tiñor collapse (∼0.8-million-year-old) embayment in the northwestern part of the island was rapidly filled by subsequent volcanism, developing a 2000-m-high, 20-kmwide volcano that collapsed (between 130,000 and 20,000 years ago) to form the present El Golfo embayment. The El Julan lateral collapse (occurring more than 158,000 years ago) removed the southwestern flank of the island, whereas the incomplete failure of the southeastern flank (between ∼261,000 and 176,000 years ago) generated the San Andrés fault system and the Las Playas slump. The latest eruptive activity of El Hierro, occuring over the last ∼145,000 years, forms a conspicuous three-branched rift system, capping most of the island with eruptive vents (at the rift crests) and lava flows (at the flanks), partially filling the respective collapse embayments. The last eruption on the island, from a small vent on the northeastern rift, was dated by 14C at 2500 years ago. An intense seismic crisis, believed to be related to an impending eruption, almost caused the total evacuation of the island in 1793. La Palma

La Palma, the most active island of the Canaries in the Holocene, is formed by two coalesced volcanoes: the northern circular Taburiente Volcano and the younger north–south elongated Cumbre Vieja volcano to the south (Fig. 3). Two lateral collapses removed much of the southwestern flank of the Taburiente volcano. The uplifted (by about 1000 m) and tilted Pliocene seamount formations are exposed on the floor and walls of the Caldera de Taburiente, formed by the younger gravitational landslide (∼0.5 million years ago). The seamount volcanism is composed of basaltic to trachytic pillow lavas and hyaloclastites. Fossils in submarine sequences suggest that the uplifted portion of the seamount stage may be 3–4 million

136

CANARY ISLANDS, GEOLOGY

years old, but there is continuing controversy about the validity of these fossil ages. The oldest subaerial volcanism is separated from the underlying submarine volcanics by an angular unconformity and a 400–600-m-thick sedimentary unit made up of breccias, agglomerates, and sediments. During the Taburiente stage of volcanism, continuous mafic alkaline eruptive activity formed a 3000-m-high shield volcano. Subsequent volcanism migrated southward along the southern Cumbre Nueva rift zone. Continuing southward migration of volcanism ultimately resulted in the extinction of the northern shield ∼0.4 million years ago and in the formation in the last 130,000 years of the Cumbre Vieja rift zone, a 20-km-long, 1949-m-high ridge, composed predominantly of mafic alkaline lavas. Half of the historical (i.e., occurring in the last 500 years) eruptions of the Canary Islands, including the most recent event in 1971, occurred along the Cumbre Vieja rift system (Fig. 3). These eruptions have been characterized by Strombolian activity forming cinder cones and lava flows. The 1971 eruption added several square kilometers of new land to the island, clearly demonstrating that the island is still growing. Phonolites have been extruded in several of the historical eruptions and can form from basanitic parental magmas within 1000–2000 years. The 1585 eruption is famous for the emplacement of giant phonolitic spines (tens of meters high), which, according to an eyewitness account of a monk, rose from the ground like “the devil’s horns” at the beginning of the almost exclusively mafic eruption. La Gomera

La Gomera, presently undergoing a volcanic hiatus, is a heavily eroded, circular (22 km in diameter) volcano (Fig. 4). During the Late Miocene, the mafic alkaline shield volcano (∼9–8 million years old) experienced a northward lateral collapse. Continued volcanic activity filled the collapse embayment and spread over the entire island, forming a central volcano with differentiated rocks at its terminus. The remains of a central caldera in this volcano crop out north of Vallehermoso, comprising trachytic and phonolitic lavas and intrusives with the latter forming radial dike swarms and cone sheets. A local unconformity separates the late-cycle Pliocene (occurring primarily 5–4 million years ago) mafic alkaline eruptions from the Miocene shield. Numerous phonolitic domes, some of the most conspicuous and spectacular volcanic features of the island, intrude both the Miocene and Pliocene mafic sequences. Volcanic activity was sparse during the last 4 million years and ended completely on the island

EL HIERRO

RIFT VOLCANISM Last Glacial Maximum Upper 35% 11% 9% 8.5% 3%

59

3%

does not reach genus level, a pattern consistent with other less isolated and/or relatively younger archipelagoes. The differentiation of various organisms observed today in the terrestrial ecosystems of the Greek Islands can be categorized largely as either geographical within species or differentiation of closely related taxa with vicariant distributional pattern and doubtful genetic isolation. This is why hybridization is intensive in the contact zones of these taxa. In the case of the Cretan flora, despite the long isolation of the island (more than 5 million years), its complex geological history, and its great topographical heterogeneity, a significant number of its endemic species seem to be paleoendemics that differentiated before the splitting of the Aegean land mass, and have since been conserved in restricted habitats (e.g., the monotypic endemic plant genus Petromarula in Crete). In general, it can be argued that all the highly diversified taxa of the Aegean belong to genera that have been present in the region for some 5 to 12 million years (e.g., Nigella [plants], Mastus and Albinaria [land snails], Dendarus [beetles], Schizidium [terrestrial isopods], Podarcis [reptiles], Rana [amphibians]). These genera exhibit large numbers of species and high percentages of endemism. Another striking characteristic of Greek Islands, especially the Aegean ones, is the presence of a vast number of small islets. More than 90% of the Aegean islands are smaller than 10 km2. Most have been formed quite recently on a geological scale and still “behave” as parts of a continuous land mass, even for taxa with reduced dispersal abilities: species numbers are high, extinctions are marginal, net effects of island size are limited, and there is a significant effect of environmental heterogeneity on species richness. These islets in many cases serve as refugia for endangered species, as in the case of the land

snail Helix godetiana. In terms of plants, several species are considered “islet specialists.” On the basis of biotic processes such as invasion, differentiation, and extinction, we can distinguish three areas in the Aegean archipelago. The first includes the surrounding mainland and the offshore islands. This zone has faced all the waves of biotic invasions and extinctions. A mixture of relict species, new invaders, and microdifferentiated forms composes the richer and more balanced ecosystems of the Aegean. The second area includes the islands of the central Kyklades, western and central Crete, Ikaria, and Karpathos. Very few of the Pleistocene invaders succeeded in establishing on these archipelagoes. On the contrary, many Pliocene elements have survived to the present day. The ecosystems of the aforementioned islands are more robust and more resistant to human alterations. The third zone includes the most isolated areas of the Aegean (southeastern Kyklades, the more remote islets in the southeastern Aegean, and eastern Crete). Their ecosystems appear to be very poor, with low biodiversity and disharmonic faunas. However, the percentage of Aegean endemics is very high, and their microdifferentiation is distinctive. Relict species are very rare. HUMAN PRESENCE, HABITAT MODIFICATION, AND ADAPTATION

The Greek islands have undergone intensive human influence for more than 8000 years. The area was the birthplace of many ancient civilizations (e.g., the Minoan of Crete, the Cycladic of the Kyklades, and the Mycenean of the Peloponnesos). The continuous and intense presence of humans in the area led to major alterations of the environment predating the last 2000 years. The first human societies transformed the natural environment through hunting, cultivation, introduction of domestic animals, manipulation of the frequency of fires, and cutting of trees. Thus, the landscape we see now has been shaped almost completely by human intervention. Nevertheless, this intervention was relatively modest until the middle of the last century, and many practices, especially agricultural ones, enhance certain components of biodiversity. Most recent taxa of the Greek Islands exhibit effective adaptations to the environmental changes and to human impacts. Although human settlements can be disastrous for fragile ecosystems, such as those of oceanic islands, Mediterranean and especially Greek islands are different. Plants for example, have had ample evolutionary experience of repeated fires and browsing and have evolved associated adaptations to avoid or recover from such events.

These adaptations have resulted from the establishment of Mediterranean-type ecosystems (MTE) on almost all the islands during the last 2 million years. By the time of the arrival of humans, whose impacts bear analogies with natural temporal changes in MTEs, many species were already “adapted.” These are the species that dominate today on most islands. CURRENT STATUS AND CONSERVATION

Despite these biological features, the scale and intensity of modern human intervention pose serious threats to many components of biodiversity. With the ongoing depletion of water resources, mainly as a result of development driven by tourism, among the most vulnerable ecosystems are those related to inland waters. Hundreds of streams and small wetlands are scattered all over the islands of the Aegean Sea, with small estuaries of seasonal streams being the most common. The insular inland waters act as refuges for most hydrophilic species and host high levels of diversity within very restricted areas, but they are vulnerable to human intrusion and encroachment: They are drained, deprived of their crucial freshwater inputs, overpumped, overgrazed, dumped, split by roads, polluted by sewage, filled with rubble, cultivated, or turned into airports. The reduction of wetlands on the Aegean Islands will drive to extinction many species dependent on them but will also result in a significant reduction of the stopover sites of millions of migrant birds. A number of conservation plans have been initiated with the recognition of the need to protect and conserve the biodiversity of the Greek Islands. Thus, 96 sites, 82 in the Aegean and 14 in the Ionian Islands, have been included in the NATURA 2000 network of protected areas. Besides the NATURA sites, several other types of sites, such as the National Park of Samaria Gorge (Crete), the National Marine Park of Alonnissos (northern Sporades) and the Protected Natural Monument of the Petrified Forest of Lesvos (Lesvos Island), as well as other types of specially protected areas (aesthetic forests and “important bird areas”), can also be found throughout the Greek Islands. As far as wetlands are concerned, approximately half are under some type of protected status. Nevertheless, progress in the conservation of the Greek Islands’ biodiversity depends predominantly on long-term studies in the area and a more organized and targeted research agenda. CONCLUDING REMARKS

The Greek Islands collectively constitute one of the most outstanding laboratories of nature. The intensiveness and

GREEK ISLANDS, BIOLOGY

391

frequency of the environmental dynamics (geological and climatic), both in the past and at present, drastically diminishes the time available for immigration, extinction, and differentiation to establish conditions of equilibrium. In addition, they provide an excellent case study for the dynamics of the interplay between human activities and biodiversity. At the same time, as a result of being a favorite destination for millions of tourists each year during the last several decades, their biota is under serious threat. There is an urgent need for intensification of conservation efforts so that their value, both scientific and cultural, can be preserved. SEE ALSO THE FOLLOWING ARTICLES

Endemism / Greek Islands, Geology / Refugia / Relaxation / Vicariance FURTHER READING

Blondel, J., and J. Aronson. 1999. Biology and wildlife of the Mediterranean region. Oxford: Oxford University Press. Poulakakis, N., A. Parmakelis, P. Lymberakis, M. Mylonas, E. Zouros, D. S. Reese, S. Glaberman, and A. Caccone. 2006. Ancient DNA forces reconsideration of evolutionary history of Mediterranean pygmy elephantids. Biology Letters 2: 451–454. Sfenthourakis, S., S. Giokas, and E. Tzanatos. 2004. From sampling stations to archipelagos: investigating aspects of the assemblage of insular biota. Global Ecology and Biogeography 13: 23–35. Stamou, G. P. 1998. Arthropods of Mediterranean-type ecosystems. Berlin: Springer-Verlag. Thompson, J. D. 2005. Plant evolution in the Mediterranean. Oxford: University of Oxford Press.

INTRODUCTION TO THE GEOLOGY OF GREECE

Two hundred million years ago the Tethys Ocean lay between Eurasia and Africa, opening out to the east. Since that time, continental fragments have spalled off Africa and been propelled toward Europe by the creation of oceanic tectonic plate material to the south and its consumption in a subduction zone to the north. When these mini-continents collided with Europe, they made mountain ranges—for example, Italy’s collision created the Alps. During these collisions some rocks were forced deep into the Earth, where the action of temperature and pressure metamorphosed the rock, changing its mineralogy and appearance. Greece has been the locus of many such collisions, which have contributed to its complex geology. At the present time the Mediterranean tectonic plate is being subducted beneath the Aegean Sea. Melting of the plate produces molten rock, which rises to the surface as the South Aegean volcanic arc (Fig. 1). The region north of the arc is expanding, opening up tectonic valleys, such as the Gulf of Corinth, and forcing Crete southward. In addition, the Anatolian plate is pushing eastward, separated from the Eurasian plate by the North Anatolian fault and its extension, the North Aegean fault zone. Movements along these plate boundaries occur during earthquakes, and

Eurasian plate

GREEK ISLANDS, GEOLOGY

Anatolian plate

MICHAEL D. HIGGINS University of Québec, Chicoutimi, Canada

The geological diversity of the Greek islands reflects long and complex interactions between the Eurasian, Mediterranean (African), and Anatolian tectonic plates. The Mediterranean climate and common paucity of soil have augmented the influence of geology on the cultural development of these islands for the last 5000 years: The nature of the bedrock and water supply has controlled agriculture; exploitation of marble and metals have been important economic activities; volcanic eruptions and earthquakes have directly influenced the lives of the inhabitants. In turn, study of the geology of the islands has contributed much to our knowledge of geological processes elsewhere.

392

GREEK ISLANDS, GEOLOGY

Folding Normal Fault Strike-slip Fault Thrust Fault Volcano

Mediterranean plate FIGURE 1 Plate tectonics of the Aegean region overlaid on a MODIS sat-

ellite image. The Anatolian plate is moving west into the Aegean region. The Mediterranean (African) plate is subducted along a major thrust fault and melts at a depth of 100 km to feed the active volcanoes.

100 km

Greece is the most seismically active part of Europe. Fault movements also produce vertical changes in the height of the land, commonly observed by local changes in sea level. Finally, faults provide channels for surface water to descend deep in the Earth and rise as hot springs. The most common rock in the region is limestone or its metamorphosed equivalent, marble (Fig. 2). Erosion of these rocks produces a special landscape with closed basins, springs, and caves. Early agriculture was enabled by the perennial springs, and caves were important as shelter and for religious purposes. In some areas subsurface water evaporates before it reaches the surface, cementing beach sand to make “beach rock.” It should be remembered that we live in geologically unusual times: Just 20,000 years ago, much of the Northern Hemisphere was covered by ice, and the sea was 120 m below its current level. Most of the Greek Islands were connected to the mainland, and there were vast expanses of shallow sea, with abundant molluscs. The shells of the molluscs were worn down to sand that formed dunes, which were rapidly cemented and transformed into a porous limestone. This useful building material is locally called Poros or Panchina and has been much used in the region for rough construction. ISLANDS OF THE SARONIC GULF

The Saronic and Corinthian Gulfs are broad, partly flooded valleys produced by almost north–south extension of the crust. The oldest rocks are hard, gray limestones (250–65 million years old) that were deposited in shallow seas to the south. These rocks are well exposed on Salamis, the island closest to Athens. They are also seen on Aegina, the largest island of the group, but only in a small area. Volcanic eruptions started 4 million years ago and covered the southern and eastern parts of the island with lavas and tuffs. After volcanism ceased, the northern part of island was submerged, and marls were deposited. Volcanic activity has continued recently on Methana, a peninsula 10 km to the south, and on the island of Poros, close to the Peloponnese.

Samothrace Thasos Imroz

Lemnos Lesvos

Kerkira

Chios Evvoia

Levkas

Samos Salamis Aegina Methana

Kefallinia Zakinthos

0

Ikaria Syros Naxos Kos

Paros Nisyros

100 km Milos Young sedimentary rocks

Thera Rhodes

Limestone Shales and sandstones Marble and schist

Crete

Serpentinite Granite Volcanic rocks

FIGURE 2 Simplified geology of the Aegean region and the Greek

Islands.

appearance of an onion (Fig. 3). Variegated colored marble (Fior di Pesca) was also exploited in antiquity near Eretria. Closure of small ocean basins thrust parts of the ocean floor over the metamorphic rocks. The whole package was then uplifted and weathered under tropical conditions to produce iron- and aluminum-rich “soils” called laterites and bauxites. The former were exploited in antiquity as a source of iron and more recently as a nickel ore. Finally, parts of the island sank down to form swampy basins. Low-grade coal, lignite, formed here and has been exploited for power generation. IONIAN ISLANDS

Kerkira (Corfu) lies close to the mainland and was indeed connected 8000 years ago when sea level was lower.

EVVOIA

Evvoia (Euboea) is a long island that runs parallel to the Greek mainland, separated from it by a strait that narrows to only 80 m at Khalkis. Here, tidal movements in the North and South Evvoikos gulfs interact to produce chaotic currents that reverse 6 to 14 times a day. The oldest rocks on the island are schists and marbles. The marble in the south of the island was exploited extensively by the Romans, especially for columns—it is now called Cipollino (Italian for onion) because layers rich in muscovite and chlorite give the

FIGURE 3 Partly finished Roman columns 5 m long, from the Cipollino

marble quarries on southern Evvoia.

GREEK ISLANDS, GEOLOGY

393

However, the western coast of the island follows a fault, and the sea-floor drops rapidly to over 1000 m. The oldest rocks are hard, gray limestones (250–145 million years old), which crop out in the north and make the highest hill. Further south, the rocks are younger and softer and have developed thick, red soils. Paleolithic implements have been found in this soil, testifying to the long occupation of this fertile island. The southern Ionian islands (Levkas, Ithaca, Kefallinia, and Zakinthos) lie close to the western edge of a tectonic plate, which is why earthquakes are so common (Fig. 1). Over the long term, such activity has produced strong relief, which is expressed as hills, islands, and lakes. All the islands are dominated by limestone (210–36 million years old), which has been cut by many faults during compression of the region. On Kefallinia this combination has led to an extensive system of sinkholes, caves, and springs. Near Argostoli, on the west coast, there is a very unusual phenomenon: The sea drains into a sinkhole (katavothre) and reappears, mixed with freshwater, on the other side of the island. The process is driven by density differences between seawater and freshwater. The oldest rocks on Zakinthos (Zante) resemble those of the islands further north but have been overlain by younger rocks. These include gypsum that was formed when the Mediterranean almost completely dried up 6 million years ago. There are natural pools of bitumen (pitch) in the southern part of the island, which formed by seepage of petroleum and evaporation of the more volatile components. Bitumen was used extensively in antiquity for waterproofing ships and jars, as well as for medical purposes. However, there are no significant oil deposits in this region. CYCLADES

The Cyclades are part of a band of complex metamorphic rocks that stretches north to Attica and Evvoia. Marble and schist dominate, but there are traces of less common minerals and rocks: The blue/mauve mineral glaucophane is widespread and was used as a pigment, jadeite from Syros is a form of jade that may have been used in Neolithic times to make axe heads, and corundum (emery) from Naxos was used to shape and polish marble. White marble was exploited in antiquity from Naxos and Paros; the translucent nature of that from the latter was particularly prized. Granite was intruded into the metamorphic rocks and is abundant on Naxos, Mykonos, and the sacred island of Delos. Milos and its surrounding islands are dominated by volcanic rocks but have a foundation similar to their

394

GREEK ISLANDS, GEOLOGY

FIGURE 4 The cliffs of the Thera caldera at Oia. Gray lavas at the base

of the cliff are covered by red agglomerates. The pale tuff at the top of the cliff is from the Minoan eruption in 1640 BC.

neighbors. Volcanism started 4 million years ago with the eruption of tuffs and lavas and has been expressed most recently by swarms of phreatic explosions, the latest about 2000 years ago. In Paleolithic to Neolithic times, the natural volcanic glass obsidian was exploited for the production of blades. Two domes of obsidian were used, both from the area north of the Bay of Milos. More recently, volcanic rocks have been exploited to make perlite. Melos is also a major producer of the clay bentonite, which is formed by hydrothermal alteration of volcanic rocks. The most famous and spectacular volcano in the Aegean is on the island of Thera (Santorini, Fig. 4). The volcano was built on a foundation of marble and schist, now exposed on the hills around Ancient Thera. Volcanism started 1.5 million years ago in the southern part of the island, but the main phase only dates from 200,000 years ago. There have been many major eruptions, which are exemplified by the Minoan eruption about 3600 years ago. This started with the rapid eruption of 35 km3 of volcanic ash, which buried a Bronze Age town in the southern part of the island near Akrotiri. The volcanic summit then collapsed, leading to the formation of a caldera that now makes up the northern part of the Bay of Thera. Construction of a new volcano started shortly afterward with the eruption of lavas in the center of the bay. Volcanic activity continues on the Kameni Islands, which last erupted in 1950. The Colombo Bank underwater volcano, located only 10 km northeast of Thera, erupted in 1650. It will probably make a new volcanic island in a few thousand years. CRETE

Crete is part of the Hellenic Arc, a series of islands and shallow water that extends from the Peloponnese to Turkey. It formed in response to the subduction of the African

plate beneath the Aegean. The plate boundary is immediately to the south, which accounts for the frequency of earthquakes. The lowest rocks exposed on the island are limestones (250–210 million years old), which have been partly recrystallized. During crustal compression almost horizontal faulting has emplaced limestones and other rocks of similar age on top. About 12 million years ago, subduction started to the south and, in response, the Aegean sea to the north expanded. Crete was faulted into many blocks, which moved independently. Some blocks became the mountains, whereas others dropped down, leaving troughs that became filled with sedimentary rocks. These large, and commonly rapid, movements continue to this day: The extreme relief of the Samaria Gorge in western Crete was produced by erosion in response to rapid uplift during the last few thousand years. More recently the harbor at Phalasarna was uplifted by 7 m, possibly during a single earthquake in the fifth century. ISLANDS OF THE NORTHERN AEGEAN

Thasos is almost completely made up of schist, gneiss, and marble and is an extension of the Rhodope metamorphic massif on the mainland 8 km to the north. The western part of the island has many small metallic mineral deposits. The oldest mines were for red ochre, hydrated iron oxide, which was exploited in Paleolithic times for cult purposes. Indeed, these underground mines were some of the largest in Europe at that time. From the ninth century BC, the same ore was used to make iron metal. There were also significant silver and gold mines, some of which were reopened in the nineteenth century for antimony and zinc. Thasos was also well known in antiquity for white marble. Samothrace lies to the north of the North Aegean trough, an important plate boundary fault. The island itself is a horst, a block of rock uplifted along faults to the north and south. The oldest rocks are parts of the ocean floor, formed about 150 million years ago. Volcanism 45 million years ago was followed 20 million years later by more volcanism and the emplacement of granite that now makes up some of the highest parts of the island. Lemnos and Imroz (a Turkish island) lie on the south side of the North Aegean trough. The sea around here is shallow, and indeed both islands were connected to the mainland 20,000 years ago. The oldest rocks on Lemnos are sandstones and marls that were shed from a rising mountain range about 45 million years ago. Similar rocks occur in the Meteora region of central Greece. Much of Lemnos and Imroz are covered by volcanic rocks that were erupted about 20 million years ago. Similar rocks also occur

on Lesvos and the Turkish mainland. Although Lemnos is associated with the blacksmith god Hephaestus, there is no evidence of recent volcanic activity. One of the chief products of Lemnos from antiquity onward was Lemnian earth, a medicament. The nature of the earth is not entirely clear: It may have been ochre deposited from springs or a mixture of clay and alum. EASTERN SPORADES

Lesvos (Mytilene) is a large island close to the Turkish coast. The eastern part of the island is composed of metamorphic rocks—schist and marble. Further west, there is a wide band of serpentinite, part of a section of ocean floor that was thrust up during continental collision. Such rocks do not produce good soils but have been exploited for magnesite. The western part of the island is covered by volcanic rocks, lavas, and tuffs, which are mostly 16–18 million years old. They are part of a much larger volcanic province that extends about 150 km to the east. Fossil pine and sequoia trees have been preserved in volcanic ashes in the western part of the island. Chios also lies close to the mainland, but its geology is quite different from that of Lesvos except for minor volcanic activity. The oldest rocks are sandstones and shale shed from a mountain range earlier than 250 million years ago. Later, these rocks were overlaid by the limestones that dominate the center of the island. In antiquity the island produced a marble called Marmo Chium or Portasanta, which is salmon pink with red and white inclusions. The rock is a metamorphosed limestone breccia. Samos is dominated by marble and schist, which are an extension of a metamorphic massif to the east. There are two basins where the younger rocks have been deposited. The eastern basin is well known for fossils, deposited 7 to 9 million years ago around small lakes when Samos was connected to the mainland. The remains include lions, mastodons, rhinos, gazelles, and Samotherium, an ancestral giraffe unique to Samos. In antiquity the island was known for its engineering works, including a 1000-m-long tunnel cut through a hill to transport spring water to the city. DODECANESE ISLANDS

Patmos is a small island built on marble but now largely covered by volcanic rocks 6–7 million years old. These are well exposed in a sacred rock shelter where St. John wrote the Book of Revelation. Marble and schist continue south to Kos but have been largely covered by limestone on the islands of Lipsos and Kalymnos. The highlands of Kos are made of marble and schist but also contain the earliest volcanic rocks erupted about

GREEK ISLANDS, GEOLOGY

395

quakes. One of the most notorious occurred in 226 BC, when it toppled the Colossus of Rhodes, a 30-m high statue of the sun god Helios, one of the seven wonders of the ancient world. The early geological history of Rhodes resembles that of Crete and much of the Peloponnese: Cherty limestones were deposited 210–65 million years ago in shallow water to the south. Later on, overall compression of the crust raised mountains that were eroded. Finally, basins developed and were filled with marls, which make fertile soils. The oldest limestones are resistant to erosion and form the highest point on the island. FIGURE 5 Recent phreatic explosion craters on Nisyros. These have

partly destroyed young volcanic domes visible to the left.

10 million years ago. Volcanism restarted 3 million years ago with the eruption of two volcanic domes (short, thick flows) in the west. Major eruptions 555,000 and 145,000 years ago produced tuffs that covered most of the island and also created calderas, which underlie the sea between Kos and Nisyros. Cold springs at the Asklepieion deposited terraces of travertine, which probably initially attracted attention to the site. Later on, the travertine was quarried to construct the temple and ancient “health center.” Nisyros is the easternmost volcano of the active South Aegean volcanic arc. Volcanic activity started about 200,000 years ago and has continued until recent times. The island is now a single simple volcano with a large crater partly occupied by young volcanic domes. The last volcanic activity was a series of phreatic explosions in 1871–1873 (Fig. 5). Deep wells have drilled for exploitation of geothermal power, but this resource is yet to be exploited. Rhodes lies just northwest of a major tectonic plate boundary, which accounts for the frequency of earth-

396

GREEK ISLANDS, GEOLOGY

SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Eruptions / Greek Islands, Biology / Mediterranean Region FURTHER READING

Fassoulas, C. G. 2000. Field guide to the geology of Crete. Natural History Museum of Crete. Friedrich, W. L. 2000. Fire in the sea: the Santorini volcano: natural history and the legend of Atlantis. Cambridge: Cambridge University Press. Higgins, M. D., and R. Higgins. 1996. A Geological companion to Greece and the Aegean. Ithaca, NY: Cornell University Press. Institute of Geology and Mineral Exploration (IGME), Athens, Greece. www.igme.gr Jacobshagen, V. 1986. Geologie von Griechenland. Beitraäge zur regionalen Geologie der Erde, Bd. 19. Berlin: Gebruder Borntraeger. Pe-Piper, G., and D. J. W. Piper. 2002. The igneous rocks of Greece: the anatomy of an orogen. Berlin: Gebruder Borntraeger. Perissoratis, C., and N. Conispoliatis. 2003. The impacts of sea-level changes during latest Pleistocene and Holocene times on the morphology of the Ionian and Aegean Seas (SE Alpine Europe). Marine Geology 196: 145–156.

GREENLAND SEE ARCTIC REGION

H HAWAIIAN ISLANDS, BIOLOGY JONATHAN PRICE University of Hawaii, Hilo

The biology of the Hawaiian Islands consists of organisms evolving and interacting in an archipelago characterized by extreme isolation, a distinctive geologic history, and a diverse physical environment. The Hawaiian biota exhibits classic examples of the evolution of endemism, the emergence of ecologic traits typical of islands, and the problems associated with human occupancy and invasive species. PHYSICAL ENVIRONMENT AND HISTORY

Lying just inside the tropics in the central Pacific, the Hawaiian archipelago is among the most isolated archipelagoes on Earth. The closest point of continental land is the west coast of North America, nearly 4000 km away. The nearest islands of comparable size and with similar environments, the Marquesas, are equally distant. Even the closest tiny atolls are over 1000 km away. These oceanic islands have never been connected to or in proximity to continental land masses. The geologic history of the Hawaiian Islands has ensured their continuous isolation from the time they originally formed to the present. The Hawaiian Islands consist of a series of volcanoes that emerged from the sea in sequence, with the youngest island, Hawaii, in the southeast and progressively older volcanoes lying to the northwest, extending out to Midway and Kure atolls. After islands form as large

land masses, subsidence, or the sinking down of volcanic masses under their tremendous weight, causes them to shrink quickly. Erosion is another process shaping the environment available to organisms. When volcanism ceases, erosive processes gradually reduce islands from large volcanic shields, to mature islands with deeply eroded valleys, to small pinnacles of rock, and ultimately to atolls with no volcanic rock above the sea surface. The duration of these life stages varies among islands, with clear implications for the long-term evolution of organisms. For example, by the time Kauai emerged about five million years ago, the islands that preceded it had largely diminished to small, distantly spaced islands. The Hawaiian Islands exhibit an enormous range of climates for so small a land area. Within a short distance on a single island, one may encounter arid lowlands, extremely rainy mountain slopes, and cold, dry alpine climates. Although the Hawaiian Islands are situated in the tropics, the presence of tall mountains there results in tremendous variation in average annual temperatures. At the lowest elevations, warm temperatures may persist year-round, while at the highest elevations on Maui, which rises to over 3000 m, and Hawaii, which rises to over 4000 m, freezing temperatures may occur throughout the year. Variation in annual precipitation is produced by the very frequent influence of trade winds. These winds bring moisture-laden clouds from the northeast, which rise against mountain slopes and produce large amounts of rain. Windward slopes and mountaintops receive over 2500 mm of rain per year, with Mount Waialeale on Kauai receiving an average of over 10,000 mm of rain per year. An inversion layer keeps most clouds and rainfall below an elevation of about 2000 m most of the time; consequently a drier climate is present on the highest mountains on the

397

islands of Maui and Hawaii, which rise above the inversion layer. Additionally, a strong rain shadow produces very dry climates at lower elevations on the leeward (southwestern) sides of all of the major islands. The net result is that most islands contain a wide range of moisture regimes. The history and climate of the Hawaiian Islands generate numerous landforms and variable soils. As erosion ensues, rugged lava fields give way to complex landscapes that contain cliffs, ridge tops, and valleys with running streams. Porous volcanic rock weathers into viable soil, gradually releasing nutrients. The youngest soils are deficient in important nutrients such as nitrogen and phosphorus, with somewhat older soils (between 20,000 and 500,000 years old) having much greater concentrations. Older soils, however, are subjected to leaching in wet climate zones and lose nutrients over time. This produces a clear pattern in terrestrial vegetation in which growth rate, forest height, and total biomass all track concentrations of soil nutrients. These processes combine to produce a diverse set of terrestrial environments. COLONIZATION AND EVOLUTION

Being isolated throughout their history, the Hawaiian Islands have been entirely dependent on long-distance dispersal to populate its ecosystems with organisms. The difficulties of dispersing such distances and becoming established have greatly restricted the kinds of organisms that colonized the archipelago. Despite their tropical setting, the Hawaiian Islands lack many groups of plants and animals that are otherwise important components of tropical communities. For example, ants, termites, and members of the philodendron, bromeliad, and ginger plant families all failed to colonize the Hawaiian Islands. Except for a species of bat and a species of monk seal, no mammals have colonized the Hawaiian Islands. Terrestrial reptiles and amphibians entirely failed to colonize. Instead, organisms with well-developed dispersal mechanisms were much more likely to colonize. Plants that were predisposed to arrive had saltwater-tolerant seeds that could float, tiny seeds that could blow long distances, or small seeds that could be transported by birds either attached to the outside or carried internally after ingestion of fruits. Arthropods that could fly or were small enough to be suspended in wind currents were similarly predisposed to colonize. Nearshore marine organisms with adults or larvae that can survive in the open ocean long enough to reach the Hawaiian Islands also had an advantage in colonization. Colonists arrived in Hawaii from throughout the Pacific region. The closest relatives of different groups of Hawaiian plants may be found in the Arctic, the desert

398

H AWA I I A N I S L A N D S , B I O L O G Y

southwest of North America, Australia, New Zealand, and Mainland Asia. Birds and arthropods share these diverse origins. The progenitors of the Hawaiian biota were therefore limited in number yet diverse in geographic origin. This disharmonic biota is markedly different from those of any other biogeographic region. The timing of colonization is constrained by the ages of the islands and the dynamic nature of their size and configuration. High islands have been continuously available since about 30 million years ago. Over that period, however, the number and size of islands have fluctuated greatly. Genetic analyses indicate that much of the biota arrived into the archipelago around the time that Kauai formed; few of the organisms that had arrived on older islands were able to disperse to Kauai, both because those islands were largely deteriorated and because they were distantly spaced relative to today’s islands. Those organisms whose origins in Hawaii do predate Kauai appear to have arrived during earlier times when larger islands were available. Upon arrival colonists encountered a diverse physical environment and novel communities of organisms. Facilitated by isolation, which prevented gene flow with ancestral populations, a majority of colonists evolved endemic species, producing very high rates of endemicity for some taxonomic groups (Table 1). These endemics exhibit adaptations to the ecological conditions of the islands, permitting them to occupy habitats quite unlike those of their ancestors. For example, some Hawaiian violet species (Violaceae) live in warm, moderately dry lowland areas, despite being descended from Arctic ancestors. True bugs in the genus Nysius (Lygaeidae) have adapted to frigid environments at the summit of Hawaii’s highest mountain, Mauna Kea, despite being related to species of tropical lowland origin. Other adaptations have helped species become more competitive or fill ecological niches that the original colonists could not fill. For example, numerous plants, including violets (Violaceae), beggar’s ticks in the genus Bidens (Asteraceae), and members of the carnation family (Caryophyllaceae) have evolved woody anatomy from weedy herbaceous ancestors. Birds, especially Hawaiian honeycreepers, adapted feeding behaviors quite different from those of their ancestors, including specialization in floral nectar, eating seeds of particular species of plants, and foraging for insects in very specialized ways with uniquely shaped bills. In one noteworthy insect group, the Eupithecia moths, caterpillars have shifted from being herbivores into being ambush predators. Other organisms experienced a release from predation and other pressures experienced by ancestors in their homelands. Birds, including ducks, geese, rails, and ibises, evolved flightless-

TABLE 1

Numbers of Species for Selected Taxa in the Hawaiian Islands Taxon

Flowering plants Other plantsa Insects Other arthropodsb Molluscs Crustaceans Fishes Amphibians Reptilesc Birdsd Mammals

Total Natives

Endemic

Endemic (%)

Naturalized

Naturalized (%)

Total Species

1006 719 5818 456 1243 1106 1143 0 4 241 25

905 241 5462 366 962 68 149 0 0 63 2

90 34 94 80 77 6 13 0 0 26 8

1101 47 2609 735 96 73 73 8 23 53 19

52 6 31 62 7 6 6 100 85 18 43

2107 766 8427 1191 1339 1179 1216 8 27 294 44

note: Adapted from Eldredge and Evenhuis (2002). a Includes ferns, fern allies, and bryophytes. b Includes arachnids, isopods, centipedes, and millipedes. c Includes marine reptiles (sea turtles). d Includes seabirds and migratory birds.

ness in the absence of mammalian predators. Numerous independent groups of insects also lost the ability to fly. In many instances, a single ancestral species diversified into numerous endemic descendant species. Speciation occurred in some groups as species moved from older to younger islands, with isolation creating species unique to each given island (a process called the progression rule). In other groups, species differentiated within a given island,

resulting in dramatically different morphology and ecology from one another, being adapted to diverse habitats and occupying different ecological niches. In these cases of adaptive radiation, a handful of colonists have produced large portions of the biota representing the full range of adaptations (Table 2). Prominent examples among the plants include the Hawaiian lobelioids (with a total of 125 species in the genera Brighamia, Clermontia, Cyanea, Delissea, Lobelia,

TABLE 2

Notable Adaptive Radiations in the Hawaiian Islands Number of Group

Taxonomic Group

Species

Notable Characteristics

Lobelioidae

Plants (Campanulaceae)

125+

Silversword alliance Schiedea

Plants (Asteraceae)

30

Plants (Caryophyllaceae) Insects (Drosophilidae)

32

Largest radiation of plant species, includes 6 genera; considerable variation in floral morphology and pollination syndrome Includes 3 endemic genera; extreme variation in physiognomy and habitat preference Considerable variation in floral morphology, breeding system, and physiognomy Largest radiation in Hawaii with over 400 described species and many more undescribed; larvae feed on fruit, bark, fungi, and other tissue Species differentiated by distinctive songs; genetic studies suggest extremely rapid speciation rate Largest radiation of arachnids; species with a variety cryptic colorations adapted for different habitats Largest radiation of terrestrial snails; species exhibit a variety of shell patterns and shapes Largest radiation of bird species; extreme variation in bill morphology, feeding behavior, and plumage Unique radiation of nectar-feeding birds; previously considered to be true honeyeaters (Meliphagidae), but now family is uncertain; all species now extinct Originally derived from dabbling ducks, evolved into several large flightless species; all species now extinct

Hawaiian drosophilids Hawaiian gryllid crickets Tetragnatha spiders Achatinelline snails Hawaiian honeycreepers Hawaiian honeyeaters Moa nalos

1000?

Insects (Gryllidae)

169+

Arachnids (Tetragnathidae) Molluscs (Achatinellidae)

29

Birds (Fringillidae)

50+

Birds (family?)

7

Birds (Anatidae)

4

H AWA I I A N I S L A N D S , B I O L O G Y

399

and Trematolobelia, in the Campanulaceae), the Hawaiian endemic mints (with a total of 57 species in the genera Haplostachys, Phyllostegia, and Stenogyne in the Lamiaceae), and the silversword alliance (with a total of 30 species in the genera Argyroxiphium, Dubautia, and Wilkesia in the Asteraceae). Along with other groups, these havediversified into multiple endemic genera with endemic species specializing in remarkably disparate habitats. Among the birds, the Hawaiian honeycreepers (numerous genera in the Fringillidae) represent the premier adaptive radiation, with over 50 highly specialized and varied species derived from a single finchlike ancestor. Among arthropods, the Hawaiian drosophilid fruit flies have generated the largest number of species from a single ancestor, perhaps as many as 1000. Other noteworthy adaptive radiations of arthropods include crickets (Gryllidae), Tetragnatha spiders (Tetragnathidae), and numerous groups of true bugs and moths. In addition, a diverse array of land snails (in the families Achatinellidae and Succinidae) has evolved from a small number of colonists.

FIGURE 2 An example of a coastal community, including cliff, beach,

and pond, on the north coast of Molokai near Olokui.

NATIVE ECOSYSTEMS

In the Hawaiian Islands, the diverse physical environment and the evolution of specialized endemic organisms have promoted the development of highly unique native ecosystems. Following are general descriptions of the major types of native ecosystems (examples can be seen in Figs. 1–8). Nearshore Marine Communities

Unlike the open ocean or deep-sea habitats, shallow habitats near the shore such as coral reefs are isolated from similar environments elsewhere and therefore make up distinct communities in Hawaii. This isolation has restricted dispersal such that Hawaii harbors only a fraction of the

FIGURE 1 A shallow-water coral reef off the leeward coast of Oahu, an

example of a Hawaiian native ecosystem.

400

H AWA I I A N I S L A N D S , B I O L O G Y

number of coral reef organisms found in the Indonesian region, a major source area for colonists. Nevertheless the isolation is sufficient that about 20% of coral reef organisms are endemic to the Hawaiian Islands, a comparatively high rate of endemism. The endemic Hawaiian monk seal can be found throughout the Hawaiian archipelago. Sea turtles can also be found in these waters, including the endangered hawksbill turtle. Coastal Communities

The coast of the Hawaiian Islands is highly varied, including arid to very wet climatic conditions and containing a variety of landforms. Sea cliffs occur particularly on windward aspects, and often fall within somewhat moist climates. Molokai and Kauai have especially tall and extensive sea cliffs, supporting a rich vegetation of ferns, sedges, succulents, and shrubs. Beaches and sand dunes occur primarily on leeward sides of islands, but also in protected locations on windward sides. These may be calcareous or (particularly on the Island of Hawaii) black volcanic sand and support a vegetation of sprawling plants such as beach morning glory (Ipomoeia pes-caprae) and ohai (Sesbania tomentosa). Anchialine ponds occur in primarily leeward areas of the island of Hawaii, where fresh groundwater mixes with seawater in ponds that are connected to the ocean only underground through porous lava. Here various species of tiny native shrimp, including opaeula (Halocaridina rubra), thrive in the absence of predatory fish.

Dry Forest and Shrubland

The leeward slopes of the Hawaiian Islands typically receive less than about 1250 mm of rain annually. Here a summer dry season presents a challenge to plants. In the very driest areas, which receive less than 500 mm of rain per year, as well as on steep slopes and areas with poor soil development, a sparse vegetation of shrubs and grasses existed, although little is known about vegetation in these areas because they have been strongly modified by human activity. More moderately dry areas receiving more than 500 mm support vegetation consisting of drought-adapted trees. Trees such as lama (Diospyros sandwicensis) have small thick leaves that persist through the dry season. A small number of trees, including wiliwili (Erythrina sandwicensis), have larger leaves that are shed during drought months, when they flower profusely.

FIGURE 4 A diverse mesic forest at Waimea canyon on Kauai.

Wet Forests and Bogs

FIGURE 3 An example of a lowland dry forest of Wiliwili (Erythrina

sandwicensis), taken at Ahihi-Kinau on Maui.

Plant communities in the wettest areas (receiving over 2500 mm of rain per year) typically consist of wet forest. These forests are usually dominated by the ohia tree (Metrosideros polymorpha) and often contain a rich assortment of ferns including hapuu tree ferns (Cibotium spp.) and the mat-forming uluhe (Dicranopteris linearis). A diverse assortment of epiphytic plants often forms a dense cover on the branches of trees. In poorly drained areas, and particularly on the wettest mountain summits on Kauai and West Maui, bogs occur (although technically these are fens) with continually saturated ground supporting an interesting community of tussock sedges, dwarf shrubs, and showy species such as greenswords (closely related to silverswords) and lobelias.

Mesic Forests

Areas receiving moderate amounts of rainfall spread throughout the year are typically referred to as mesic. Here conditions are favorable for numerous tree species, and consequently mesic forests are the most diverse, particularly on the older islands of Kauai and Oahu. While ohia trees (Metrosideros polymorpha) often dominate in these and other plant communities, in some cases there is no clear dominant tree species. Higher-elevation mesic forests, especially on Maui and Hawaii, may be dominated by very large koa trees (Acacia koa). Such forests provide habitat for the endangered akiapolaau (Hemignathus munroi), a honeycreeper with a peculiar bill that specializes in digging arthropods out of koa branches.

FIGURE 5 An example of wet forest on the island of Hawaii.

H AWA I I A N I S L A N D S , B I O L O G Y

401

Subalpine and Alpine Communities

Freshwater Habitats

Above 2000 m elevation, the trade wind inversion permits little cloud formation or rainfall. In addition, temperatures often drop to below freezing, particularly at night. Consequently, vegetation is adapted to cold, dry conditions. Where warmer and moister habitats directly abut this zone, a treeline often demarcates the upper limit of forest vegetation. Immediately above the treeline, vegetation is stunted and takes the form of a dense shrubland, although trees may be found in gulches, which retain moisture and are protected from colder temperatures. Shrubs adapted to cold, dry conditions, including pukiawe (Leptecophylla tameiameiae), hinahina (Geranium cuneatum), and a‘ali‘i (Dodonaea viscosa), form dense thickets. On the slopes of Mauna Kea, forests of mamane (Sophora chrysophylla) support populations of palila (Loxioides bailleui), a specialist honeycreeper that feeds on mamane seed pods. At higher elevations, vegetation becomes sparser as conditions become more extreme, and species such as the silversword (Argyroxiphium sandwicense) thrive in an otherwise barren landscape. At the highest elevation in the islands, the summit of Mauna Kea is essentially devoid of vegetation, yet it supports populations of the very unique wekiu bug (Nysius wekiuicola), which tolerates subfreezing temperatures and feeds on insects that are blown up from lower elevations and stunned by the cold.

Many perennial streams run particularly along windward slopes, although large watersheds such as that of Waimea Canyon on Kauai begin in wet summit regions, then drain primarily through drier leeward areas. Hawaiian streams are home to numerous types of endemic arthropods, especially conspicuous dragonflies (Anax spp.) and damselflies (Megalagrion spp.). Oopu (five native species of fish, all but one of which are gobies) spend much of their lives in fresh water and can climb up rocks and even waterfalls against fast-moving stream currents. Larger streams drain into estuaries and small bays, which provide protection and serve as nurseries for many fish species.

FIGURE 7 A lava pioneer community with ohia tree, at Kilauea Iki on

Hawaii.

Lava Pioneer Communities

FIGURE 6 An alpine community with silversword (Argyroxiphium

sandwicense), at Haleakala on Hawaii.

402

H AWA I I A N I S L A N D S , B I O L O G Y

On the island of Hawaii, volcanic eruptions are frequent enough that primary succession occurs across large areas recently covered by lava. Young lava flows tend to support little vegetation, even in areas with very wet climate, because the porous nature of the substrate retains little water and has few available nutrients. The first pioneer species to colonize lava include lichens, ferns (especially ae, Polypodium pellucidum), and the nearly ubiquitous ohia. Over time, ohia trees may grow larger, and an increasing number of species may add to the community until a fully formed forest develops (in wet regions) within about 500 years. Young lava flows also may flow around an area of older substrate, leaving an island-like patch of forest surrounded by barren vegetation. These patches, known as kı¯puka, behave like miniature islands where certain species (especially Drosophila fruit flies) may become isolated and even begin to differentiate from nearby relatives as a result of this natural fragmentation.

recently, introduced arthropod pests, game animals such as deer and mouflon sheep, and escaped domestic animals such as cats and large European pigs have further contributed to ecosystem disruption. Large mammals damage native vegetation, which is not adapted to such disturbance, facilitating invasions of invasive plant species that are better adapted. Introduced cats and mongooses prey on native birds, especially their young. Feral European pigs create habitat for introduced mosquitoes, which spread devastating diseases to native forest birds, which have no resistance. Rats continue to degrade vegetation by stripping bark and eating the seeds of endemic plants that are not adapted to such predation. Many more bird species have gone extinct since Western contact, as well as nearly 10% of endemic plant species, and unknown numbers of arthropod and tree snail species. FIGURE 8 A freshwater stream and waterfall at Kopiliula on Maui.

HUMAN IMPACTS

The arrival of humans initiated prodigious changes to Hawaiian native ecosystems. The first people to arrive, over 1000 years ago, Polynesians brought a sophisticated island culture, numerous crop and utility plants, and animals, including Polynesian pigs, chickens, dogs, and (unintentionally) the Polynesian rat. These first Hawaiians cleared land for crops, using fire extensively in lowland and dry areas, and likely hunted larger flightless bird species. Rats fed widely on seeds, bird eggs, and nestlings, and probably on arthropods and tree snails. This combination of stresses led to the extinction of approximately half of all endemic bird species, as evidenced by an extensive sub-fossil record. Some plant, arthropod, and snail species also may have become extinct or at least greatly reduced during this period. Western contact, beginning with Captain Cook in 1778, heralded an acceleration of losses to the Hawaiian biota. Goats, cattle, and two new species of rats were all introduced within decades of European contact. Extensive cutting of sandalwood for export to Asian markets, clearing of more land for plantation agriculture, and the introduction of numerous plant species from around the world (many of which became invasive) greatly increased the rate of habitat loss and degradation. More

CONSERVATION

As a result of human activities, a large proportion of the remaining Hawaiian biota is endangered. Consequently, a number of efforts have been employed to stem the loss of species. First, large areas have been set aside for conservation in the Hawaiian Islands, including national parks and federal, state and private nature reserves. These include many terrestrial habitats as well as the marine and coastal habitats of the Northwest Hawaiian Islands National Marine Sanctuary, the largest marine protected area in the world. Strong measures have been taken to reduce and reverse the damage done by invasive species. By fencing extensive areas and removing large mammals, native vegetation has recovered to some degree, although full recovery may take time and require additional measures. Invasive plant species have been controlled where possible, although new species are becoming established every year. In some cases biocontrol has been used, although with mixed results. Control of rats, cats, mongooses, and ants has been effective only at very local scales. Some birds have been bred in captivity, although only the puaiohi, or small Kauai thrush (Myadestes palmeri), and the nene (Branta sandwicensis) have been released with any degree of success. Similarly, numerous plant species have been propagated in nurseries, with many being outplanted in safe locations. Even some land snails have been raised successfully in captivity. However, it is not clear to what degree any of these will establish stable, self-sustaining populations in the long term.

H AWA I I A N I S L A N D S , B I O L O G Y

403

Protection of important nesting sites has become important to the reproduction of sea turtles, and likewise marine protected areas may bolster reproduction of various fish populations. Despite continued and even very recent extinctions, the combined approach of establishing natural areas and engaging in active management holds promise for longterm conservation. Cooperation among government agencies, nonprofit conservation organizations, and private landowners makes possible coordinated, landscapescale conservation efforts needed to save what remains of the unique Hawaiian biota.

of atolls, extends 2600 km to Kure island (Fig. 1). The ridge is the southeastern part of the Hawaiian–Emperor volcanic chain, the balance of which comprises submarine seamounts that reach to Kamchatka. The chain is convincing evidence for a hotspot melting anomaly deep in the Earth’s mantle, even as the stability and depth of some hotspots are being reexamined by new scientific investigations.

SEE ALSO THE FOLLOWING ARTICLES

Crickets / Drosophila / Hawaiian Islands, Geology / Honeycreepers, Hawaiian / Invasion Biology / Kı¯puka / Silverswords FURTHER READING

Cuddihy, L. W. and C. P. Stone. 1990. Alteration of native Hawaiian vegetation: effects of humans, their activities and introductions. Honolulu: University of Hawaii Press. Culliney, J. L. 2006. Islands in a far sea: nature and man in Hawaii. Honolulu: University of Hawaii Press. Eldredge, L. G., and N. L. Evenhuis. 2002. Numbers of Hawaiian species for 2000. Bishop Museum Occasional Papers 68: 71–78. Stone, C. P., C. W. Smith, and J. T. Tunison, eds.. 1992. Alien plant invasions in native ecosystems of Hawaii. Honolulu: Cooperative National Park Resources Studies Unit, University of Hawaii. Price, J. P., and D. A. Clague. 2002. How old is the Hawaiian biota? Geology and phylogeny suggest recent divergence. Proceedings of the Royal Society of London Series B 269: 2429–2435. Price, J. P., and W. L. Wagner. 2004. Speciation in Hawaiian angiosperm lineages: cause, consequence, and mode. Evolution 58: 2185–2200. Vitousek, P. M. 1995. The Hawaiian Islands as a model system for ecosystem studies. Pacific Science 49: 2–16. Wagner, W. L., and V. A. Funk, eds. 1995. Hawaiiian biogeography: evolution on a hot spot archipelago. Washington, DC: Smithsonian Institution Press. Ziegler, A. C. 2002. Hawaiian natural history, ecology, and evolution. Honolulu: University of Hawaii Press.

HAWAIIAN ISLANDS, GEOLOGY DAVID R. SHERROD U.S. Geological Survey, Vancouver, Washington

The Hawaiian Islands described here are the eight principal islands of the Hawaiian Ridge, which, including a series

404

H AWA I I A N I S L A N D S , G E O L O G Y

FIGURE 1 Hawaiian Islands and Hawaiian–Emperor volcanic chain,

most of which consists of submarine seamounts. Vectors indicate Pacific plate motion, in millimeters per year, relative to presumed fixed mantle hotspot.

HAWAIIAN VOLCANOES BUILD THROUGH A SEQUENCE OF STAGES

The main Hawaiian Islands are built of 15 emergent volcanoes. Two other volcanoes, although now submerged, were once emergent, and a third, newly born about 250,000 years ago, is still building toward sea level. An idealized Hawaiian volcano passes through four eruptive stages: preshield, shield, postshield, and rejuvenated. These stages likely reflect variations in the amount and rate of heat supplied to the lithosphere as the Pacific plate overrides the Hawaiian hotspot. Volcanic extinction ensues as a volcano moves away from the hotspot. At inception a Hawaiian volcano is in its preshield stage. Eruptive products, chemically all basalt, are slightly richer in the alkali elements (for example, sodium and potassium) than the typical shield-stage basaltic lava flows. The alkalic character of preshield lava is a consequence of a nascent magma-transport system and less extensive melting at the periphery of the mantle plume

fed by the hotspot. Such lava has been dredged from the archetypal preshield-stage volcano, Lo¯‘ihi, youngest of the Hawaiian volcanoes. Lo¯‘ihi’s summit lies submerged about 980 m beneath sea level, 30 km south of the Island of Hawai‘i (Fig. 2). Similar volcanic rocks have been recovered from Kı¯lauea volcano’s south flank by remotely operated and manned submersibles. Recently obtained samples from Huala¯lai volcano’s northwest rift zone, also on the Island of Hawai‘i, may indicate that the preshield stage is still exposed there. Elsewhere along the chain, these early strata are buried by lava flows of shield-stage volcanism. The shield stage is the most productive volcanically, and each Hawaiian volcano erupts an estimated 80–95 percent of its ultimate volume during this stage. Shieldstage volcanism marks the time when a volcano is near or above the hotspot and its magma supply system is robust. The degree of mantle melting increases by factor of 2–2.5 compared to the melting that drives preshield-stage magmatism. Magma ascending from at least 60–70 km depth (deepest earthquakes) is stored in reservoirs at the base of the oceanic crust (approximately 20 km depth) and then, at shallow level, in a nexus of sheetlike intrusions and more equant reservoirs 3–7 km beneath the volcano’s summit. Some magma erupts at a volcano’s summit, but much is shunted into rift zone dikes to feed eruptions and intrusions downslope (Fig. 3). The pa¯hoehoe and ‘a‘a¯ lava flows of the ongoing eruption that began in 1983 along Kı¯lauea’s east rift zone, 20 km from the volcano’s summit, are characteristic of shield-stage volcanism in both style and composition. Rift zones are prominent topographic features of many Hawaiian shields. Two or three rift zones are typical at an individual volcano. Some of the volcanoes, such as Kaua‘i, are nearly equant in plain view because their three rift zones are equally active (Fig. 2). The presence of an adjacent volcano, however, changes the regional stress regime to favor extension along only two of the rift zones, leading to a substantially elongate volcano. The growth of East Maui has been almost entirely along Haleakala¯’s east rift zone and its offshore continuation, the Ha¯na Ridge (Fig. 2); in contrast, the volcano’s southwest and north rift zones are stunted. Similarly, Kı¯lauea’s east rift zone and offshore Puna Ridge have been the eruptive loci for most of that volcano’s lava flows. Seismic and gravity data indicate that the rift zone dikes of shield-stage volcanism have roots as deep as the base of the volcano and possibly penetrate into the underlying oceanic crust (Fig. 3).

FIGURE 2 Eight main Hawaiian Islands. Shown named are submarine

landslides discussed in text. Red lines are contours showing depth, in meters below sea level, to top of oceanic crust (Watts and ten Brink 1989), which is depressed because of the load of young volcanoes and high heat flow in area of Hawaiian hotspot. Some volcanoes indicated by initials: W, Wai‘anae; K, Ko‘olau (on O‘ahu); H, Haleakala ¯ (on Maui); KO, Kohala; HU, Huala ¯lai; MK, Mauna Kea; ML, Mauna Loa; KI, Kı¯lauea (on Hawai‘i).

FIGURE 3 Zone of magma throughput into summit area of volcano

and downrift transport into rift zone dike system during shield-stage volcanism (generalized from Tilling et al. 1987).

Duration of Shield Stage

Hawaiian volcanoes require at least 0.6 million years to grow from the ocean floor to nearly full size by late in their shield-building stage. They breach the ocean surface about midway through this period. But the entire shield stage persists for 1 million years or more, judging

H AWA I I A N I S L A N D S , G E O L O G Y

405

by the 0.6–0.8-million-year span of late-shield ages from subaerially exposed lava on Wai‘anae volcano (O‘ahu) and West Maui. Stratigraphic accumulation rates on the order of 7.8–8.6 (± 3.1) m per 1000 years are indicated by ages from a deep drill hole into Mauna Kea’s shield-stage lava near Hilo. A similar rate, about 6 m per 1000 years, was derived from drill core data on Kı¯lauea’s east rift zone. Volumetrically the magma supply rate at the currently active Kı¯lauea is in the range 0.09 to 0.11 km3 per year (calculated as vesicle-free, denserock-equivalent magma), on the basis of geodetic and lava effusion-rate data from eruptions in the twentieth century. Compositional Variation

Of 1000 whole-rock analyses from Mauna Loa and Kı¯lauea, 99 percent contain between 47% and 54% SiO2—the tholeiitic basalt that builds the islands (Fig. 4). The few analytical outliers are typically only slightly more or less silicic. Early and late-shield strata extend the silica

range as alkali basalt and even hawaiite lava flows are sparsely interlayered with tholeiite at some volcanoes. The notable but unusual example of more highly fractionated shield-stage lava comes from Wai‘anae volcano, O‘ahu. There, the chemical trend includes the Mount Kuwale rhyodacite flow, the most silicic lava in the islands (68 percent SiO2). Intervolcano compositional differences result mainly from variations in the part of the mantle plume sampled by magmatism and the zoning of sources within it. These distinctions are tracked most successfully by the trace element variation between volcanoes, notably in the radiogenic isotopes of Pb, Sr, and Nd. These data add geochemical significance to the spatial concept of “Loa” and “Kea” trends, in which the volcanoes from O‘ahu southeast to the Island of Hawai‘i fall into one of two geographic alignments and geochemical groupings. The trend names come from Mauna Loa and Mauna Kea, prominent volcanoes that fix the southeast position of the alignments. VOLCANO BIRTH PROGRESSES SOUTHEAST

The age of Hawaiian shield-stage volcanism is successively younger from northwest to southeast, an observation made first by the early Polynesian settlers 1200 or more years ago and preserved in their oral history. In the plate tectonic paradigm, this age progression results from the northwestward movement of the Pacific plate over the Hawaiian hotspot at the rate of about 10 cm per year. Radiometric ages from the Ni‘ihau shield are as old as about 6.3–5.5 million years. Similar ages have been obtained from Kaua‘i (Table 1). Lava from O‘ahu dates back to about 3.9 million years from Wai‘anae volcano and 3.2 million years from the younger Ko‘olau volcano. Ages from Moloka‘i and Maui are as old as about 2 million years. All exposed lava flows on the Island of Hawai‘i are younger than 0.6 million years. VOLCANISM COMMONLY PERSISTS INTO THE POSTSHIELD STAGE

FIGURE 4 Silica vs. total alkalies variation diagram for representative

chemical analyses of Hawaiian lava. Rock classification grid from LeBas et al. (1986) and dashed boundary separating tholeiitic and alkalic basalt from Macdonald and Katsura (1964). Stratigraphic formation names shown parenthetically in explanation box. Hawaiian volcanoes evolve compositionally from shield to transitional to post-shield as their magmatic systems become less robust. Rejuvenated-stage lava, where present, follows after a repose. Hawaiian lava data available in electronic format (Sherrod et al. 2007).

406

H AWA I I A N I S L A N D S , G E O L O G Y

As Pacific plate motion rafts Hawaiian volcanoes away from the hotspot, volcanism wanes gradually, passing from the shield stage into the postshield stage. The shallow magma reservoirs (3–7 km depth) of the shield stage volcanoes cannot be sustained as magmatic supply lessens, but smaller reservoirs at 20–30 km depth persist. The rate of extrusion, as measured by upward stratigraphic accumulation, diminishes by a factor of ten late in the shield stage. The composition of erupted lava also changes,

TABLE 1

Summary of Volume, Eruptive Age, and Volcanic Stages of Volcanoes on the Main Hawaiian Islands Volume, km3

Stages Presenta

Island

Volcano

Ni‘ihau Kaua‘i O‘ahu

Ni‘ihau Kaua‘i Wai‘anae Ko‘olau West Moloka‘i East Moloka‘i La¯na‘i Kaho‘olawe West Maui Haleakala¯ Kohala

21.7 57.6 53.5 34.1 30.3 23.9 21.1 26.3 9 69.8 36.4

S, (PS), R S, (PS), R S, PS S, R S, PS S, PS, R S S, (PS) S, PS, R S, PS S, PS

Mauna Kea

41.9

S, PS

Huala¯lai Mauna Loa Kı¯lauea

14.2 74.0 31.6

S, PS S S

Moloka‘i La¯na‘i Kaho‘olawe Maui Hawai‘i

Oldest Radiometric Ageb

6.3; 5.54 5.77 3.93 3.19 1.99 1.75 1.5 1.25 2.15 1.1 0.6; 0.46 260 ka in outcrop; 330 ka in drill core 114–92 ka 100–200 ka 70 ka in outcrop; 220 ka from submarine exposures)

Youngest Volcanism

0.35 0.375 2.89; 2.77 0.1 1.73 0.35 1.24 0.9 0.4 ca. AD 1600 120 ka, perhaps as young as 80 ka 4.6 ka, calibrated age AD 1801 AD 1984 Ongoing continuously since AD 1983

note: Volumetric data from Robinson and Eakins (2006). Stage distribution and age data compiled from Sherrod et al. (2007). a Stages: S, shield; PS, postshield (shown parenthetically where volumetrically minor or not mapped separately); R, rejuvenated. b Radiometric ages in millions of years unless specified; ka = thousands of years.

becoming more alkalic as the degree of melting diminishes (Fig. 4). Of the volcanoes old enough to have seen the transition, only two, Ko‘olau and La¯na‘i, lack rocks of postshield composition. Eight have postshield strata sufficiently distinct and widespread to map separately. The transition to postshield volcanism is brief (less than 0.1–0.2 million years) and commonly too short to measure confidently by radiometric dating. Postshieldstage volcanism generally lasts for another 0.1–0.2 million years, although its endurance is variable. At Mauna Kea it has been ongoing for 0.3 million years. Exceptional is the 1-million-year duration at Haleakala¯, where eruptive products of the Kula and Ha¯na Volcanics have coated the volcano with a thickness as great as 1 km. The occurrence of postshield volcanism at Hawaiian volcanoes may be explained by the thermal structure imposed on the overriding lithospheric plate, as determined from numeric modeling of lithospheric and mantle heat transport, viscosity, and kinematics (Fig. 5), but other, still poorly defined factors must play into the final explanation to account for the widely varying durations mentioned.

to 2.0 million years in duration. The prominent volcanic landforms on O‘ahu, such as Punchbowl, Diamond Head, and Hanauma crater, are products of the rejuvenatedstage Honolulu Volcanics, which began about 0.8 million years ago and have persisted sporadically until as recently

FIGURE 5 Thermal structure of lithosphere beneath Hawaiian Islands,

REJUVENATED-STAGE VOLCANISM FOLLOWS QUIESCENCE

Five Hawaiian volcanoes have seen rejuvenated-stage volcanism following quiescent periods that ranged from 0.5

derived from numeric modeling of heat transport, viscosity, and kinematics needed to account for ocean-floor rebound, lateral extent of Hawaiian Arch, and melt production of Hawaiian Ridge. Motion of Pacific plate is from right to left, which leads to asymmetric isotherms leeward of the hotspot. Modified from Ribe and Christenson (1999).

H AWA I I A N I S L A N D S , G E O L O G Y

407

as about 0.1 million years ago. The rejuvenated stage can be brief—only one or two eruptive episodes—or notably durable. That on Ni‘ihau lasted from 2.2 to 0.4 million years ago; on Kaua‘i, it has been ongoing since 3 million years ago. The causal mechanisms of rejuvenated-stage volcanism remain difficult to explain. Rejuvenation is likely to be in some way related to decompression melting, either as the lithosphere rebounds from the zone of depression beneath the largest young volcanoes or when hot mantle, dragged initially downward in response to plate motion, rises naturally by its lower density until pressure and temperature are suitable for melt production (Fig. 5). Some models postulate that enhanced crustal fracturing, as might result from the lithospheric rebound, enables magma to percolate upward more easily. The combination of shield, postshield, and rejuvenatedstage volcanism accounts for the wide range in the age of volcanism along the island chain, even though the age of Hawaiian shields is progressively younger to the southeast. For example, almost every island from Ni‘ihau to Hawai‘i had an eruption in the time between 0.3 and 0.4 million years ago, even though only the Island of Hawai‘i was then hosting volcanoes in their shield stage.

Nui—that encompassed about 14,600 km2, larger by 50% than today’s Big Island of Hawai‘i. Included in this landmass was Penguin Bank, now a broad shoal west of Moloka‘i but originally a volcano that completed its shieldbuilding stage about 2.2 million years ago, when it was briefly connected to O‘ahu. The islands of Maui, La¯na‘i, and Moloka‘i were connected as recently as 18,000 years ago—a connection in this case owing to glacioeustatic sea-level lowering but nonetheless pointing out the shallow ocean depth across the submerged land bridges. In its heyday, Maui Nui, or “Greater Maui,” encompassed extensive lowlands dotted by upland areas. Plants and animals, including flightless or slowly dispersing species, could disperse readily across this terrane; consequently the biologic communities among the Maui Nui islands are more similar than would be found on islands always isolated from each other. Maui Nui’s large area increased the chance that airborne or waterborne species would make landfall. Its diverse environments provided a rich storehouse of plants and animals for the subsequent colonization of Hawai‘i, the youngest island in the chain.

GROWTH OF THE ISLANDS DEPRESSES THE CRUST

Large landslides ring the submarine flanks and seafloor adjacent to Hawaiian volcanoes, a significant discovery of the past 30 years. Recognized first at the emergent volcanoes, these landslides can occur at any stage of volcanic growth or quiescence. For example, submarine Lo¯‘ihi is already gutted by slope failures that have sculpted it into a narrow ridge. The larger landslides may reach onshore, their headwalls defined by normal faults. These faults may have prominent topographic expression, as seen on the west side of Mauna Loa, for example (Fig. 2). Some faults may result from perturbations in the local stress field after large landslides have removed buttressing slopes. Examples include faults along the crest of La¯na‘i and Kohala, which resulted from, respectively, the Clark debris avalanche about 0.8 million years ago and the Pololu¯ Slump in the past 0.1 million years.

The massive outpouring of lava flows from Hawaiian volcanoes weighs upon the oceanic crust, depressing it along an axial Hawaiian Moat (Fig. 2). The periphery of subsidence is marked by the surrounding Hawaiian Arch. The ocean floor, at about 4.5–5.0 km depth adjacent to the Hawaiian Ridge, has subsided to depths as great as 9.5 km along the moat (Fig. 2). Subsidence is ongoing throughout the shield stage and probably into postshield time. The shoreline of Hawai‘i Island today, loaded down by its five exposed shields, is subsiding at 2–3 mm per year as measured by tide gauges, after correcting for global sealevel rise. The rate in the center of the island is probably twice as great. One consequence of subsidence is the drowning of coral reefs that drape the submarine flanks of the actively subsiding volcanoes. At least six reefs northwest of Hawai‘i Island form a stairstep configuration, the oldest being deepest. Drowned reefs on Maui are tilted toward nearby Hawai‘i Island, where downwarping is greatest. Large islands built by several volcanoes become a sequence of small islands once volcanic upbuilding ceases. The four islands of Moloka‘i, La¯na‘i, Kaho‘olawe, and Maui were once a single subaerial landmass—Maui

408

H AWA I I A N I S L A N D S , G E O L O G Y

DISSECTION BY LARGE LANDSLIDES MAY OCCUR ANY TIME

TSUNAMIGENIC DEPOSITS

Disagreement still surrounds the origin of poorly sorted, coralline-bearing sedimentary breccia found at widely ranging altitudes as high as 170 m on the leeward sides of Kohala, West Maui, La¯na‘i, and East Moloka‘i volcanoes. The most widely accepted hypothesis explains these deposits as the consequence of catastrophic, giant

waves (megatsunami) generated by prehistoric submarine landslides. The interpretation stems partly from the landward fining of the La¯na‘i deposits and landward fining in the carbonate-clast component of the Moloka‘i deposits. The La¯na‘i deposits were specifically attributed to the ‘A¯lika 2 Slide, a slope failure occurring about 125,000 years ago from the west side of Hawai‘i Island (Fig. 2). The breccia deposits were originally interpreted as ancient shorelines and attributed to glacioeustatic marine high sea level stands, an explanation that requires substantial uplift of La¯na‘i and Moloka‘i to account for their vertical positioning. Recent estimates for uplift of O‘ahu suggest rates of 0.020–0.024 m per 1000 years for the past 400,000 years. The result has been to expose calcareous reef rock and marine sediment, the heart of the cement industry in the Hawaiian Islands. The absence of these emerged reefs and lagoonal limestone beds elsewhere in the Hawaiian Islands suggests that neither La¯na‘i nor Moloka‘i have seen much uplift. Although rates are imprecisely defined for La¯na‘i, during the past 30,000 years that island has been relatively stable, with uplift or subsidence bracketed between +0.1 and −0.4 m per 1,000 years, on the basis of the depositional character of carbonate deposits on submerged terraces adjacent to the island. Compelling evidence in favor of the giant-wave hypothesis comes from deposits on Kohala volcano, Island of Hawai‘i, where the question of uplift is made moot by the ongoing subsidence that has characterized Hawai‘i Island since its emergence. The calcareous breccia of Kohala, found today at altitudes ranging from sea level to 100 m, must have been deposited originally at altitude 350 to 390 m higher, after correcting for modern rates of subsidence and the age of the deposits. MAUNA KEA—THE ONLY GLACIATED VOLCANO IN THE CHAIN

Mauna Kea, on the Island of Hawai‘i, is the only volcano known to have undergone glaciation in the past 200,000 years. Adjacent Mauna Loa had sufficient area above the ice equilibrium-line altitude to maintain an ice cap during the last glaciation 25,000–15,000 years ago, but till and outwash deposits, if they formed, must be buried now by younger lava flows throughout the summit region and down to at least 2,000 m altitude. That altitude is too low to expect depositional traces, if the mapped till and outwash from Mauna Kea are useful benchmarks. Haleakala¯ (East Maui) lacked sufficient high-altitude terrain to accumulate glaciers in the past 200,000 years, and no deposits have ever been found there.

Mauna Kea’s three known glaciations have scoured lava flows at higher altitudes, leaving striated bedrock surfaces that lead down to the moraines and outwash left by the glaciers. Some lava flows in the summit area have ice-contact features such as steep margins, pillow basalt, glassy faces, and palagonitized zones. The estimated age and duration of Mauna Kea’s three glaciations are drawn from the ages of bracketing, dated lava flows, and worldwide correlations using oxygen isotope data to match glacial maxima. The oldest recognized glaciation, the Po¯hakuloa (using the Mauna Kea terminology), corresponds to marine oxygen isotope stage 6 and likely occurred sometime between about 180,000 and 130,000 years ago. The Waihu¯ glaciation is thought to have occurred during oxygen isotope stage 4, or roughly 80,000–60,000 years ago. The youngest, the Ma¯kanaka, was under way by about 40,000 years ago. It had ended by 13,000 years ago, the time when a small summit depression became an ice-free lake capable of accumulating sediment. PERMEABLE VOLCANIC GEOLOGY PUTS GROUNDWATER MOSTLY AT SEA LEVEL

Rain and snow percolate into the highly permeable lava flows of Hawaiian volcanoes. The resulting groundwater may become perched on impermeable ashy beds and zones of secondary mineralization or hydrothermal alteration, but most penetrates to sea level within the island edifice. There the freshwater floats upon the saltwater owing to its lower specific gravity, depressing the interface between them and, in compensation, mounding slightly. The resulting lensshaped body of freshwater thickens slightly inland, albeit to no more than a few meters above sea level in the center of each island. The slope of the water table is nearly flat, sloping downward roughly 0.3–0.4 m per km toward the coast. Rift zones, with an internal structure rich in vertical dikes, possess somewhat different hydrologic character than a volcano’s flanks because the dikes can act as dams to retard the lateral migration of groundwater. Groundwater in the Hawaiian Islands is derived chiefly from the freshwater lens beneath each island. The resource is fragile, because overpumping allows underlying brackish water to intrude, destroying the quality of the freshwater. Upslope water development typically is limited to elevations where drilling and recovery are economically feasible, currently about 400–600 m above sea level. Potable water at higher altitude comes chiefly from rain catchment systems or by diverting streams into canal systems built in the late nineteenth and early twentieth centuries to bring water from the windward, wetter sides of the higher islands for irrigation purposes.

H AWA I I A N I S L A N D S , G E O L O G Y

409

SEE ALSO THE FOLLOWING ARTICLES

Hawaiian Islands, Biology / Landslides / Lava Tubes / Oceanic Islands / Plate Tectonics / Tsunamis / Volcanic Islands

HONEYCREEPERS, HAWAIIAN

FURTHER READING

Clague, D. A., and G. B. Dalrymple. 1987. The Hawaiian–Emperor volcanic chain. Part I, Geologic evolution. U. S. Geological Survey Professional Paper 1350, vol. 1, 5–54. Eakins, B. W., J. E. Robinson, T. Kanamatsu, J. Naka, J. R. Smith, E. Takahashi, and D. A. Clague. 2003. Hawaii’s volcanoes revealed. U. S. Geological Survey Geologic Investigations Series Map I–2809, scale about 1:850,000. [available online at http://geopubs.wr.usgs. gov/i-map/i2809]. Gingerich, S. B., and D. S. Oki. 2000. Ground water in Hawaii. U. S. Geological Survey Fact Sheet 126-00. http://hi.water.usgs. gov/publications/pubs/fs/fs126-00.pdf. Moore, J. G., and D. A. Clague. 1992. Volcano growth and evolution of the island of Hawaii. Geological Society of America Bulletin 104(11): 1471–1484. Moore, J. G., D. A. Clague, R. T. Holcomb, P. W. Lipman, W. R. Normark, and M. E. Torresan. 1989. Prodigious submarine landslides on the Hawaiian Ridge. Journal of Geophysical Research 94(B12): 17465–17484. Owen, S., P. Segall, M. Lisowski, A. Miklius, R. Denlinger, and M. Sako. 2000. Rapid deformation of Kilauea Volcano: Global Positioning System measurements between 1990 and 1996. Journal of Geophysical Research 105(B8): 18983–18998. Price, J. P., and D. Elliott-Fisk. 2004. Topographic history of the Maui Nui complex, Hawaii, and its implications for biogeography. Pacific Science 58(1): 27–45. Sherrod, D. R., J. M. Sinton, S. E. Watkins, and K. M. Brunt. 2007. Geologic map of the State of Hawai‘i. U. S. Geological Survey Open-File Report 2007-1089. http://pubs.usgs.gov/of/2007/1089. Tilling, R. I., C. Heliker, and T. L. Wright. 1987. Eruptions of Hawaiian volcanoes: past, present, and future. U. S. Geological Survey General Interest Publication.

ROBERT C. FLEISCHER Smithsonian Institution, Washington, DC

The Hawaiian honeycreepers, presently classified in the subfamily Drepanidinae (also called drepanidines or Hawaiian finches), are a morphologically and ecologically diverse group of more than 56 species of cardueline finches endemic to the Hawaiian Islands. Molecular data and rate calibrations based on geological age of the islands suggest that they radiated from a single colonizing ancestral cardueline species beginning as little as 3–4 million years ago. The Hawaiian honeycreepers are a highly endangered avian group, with about 70% of the species recently extinct (i.e., recent Holocene), and at least ten of the remaining 17 species currently considered endangered. TAXONOMY AND PALEONTOLOGY

Specimens of Hawaiian honeycreepers were originally described by early ornithologists as members of many

REFERENCES

Le Bas, M. J., R. W. Le Maitre, A. Streckeisen, and B. Zanettin. 1986. A chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of Petrology 27: 745–750. Macdonald, G. A., and T. Katsura. 1964. Chemical composition of Hawaiian lavas. Journal of Petrology 5: 82–113. Ribe, N. M., and U. R. Christenson. 1999. The dynamical origin of Hawaiian volcanism. Earth and Planetary Science Letters 171: 517–531. Robinson, J. E., and B. W. Eakins. 2006. Calculated volumes of individual shield volcanoes at the young end of the Hawaiian Ridge. Journal of Volcanology and Geothermal Research 151: 309–317. Watts, A. B., and U. S. ten Brink. 1989. Crustal structure, flexure, and subsidence of the Hawaiian Islands. Journal of Geophysical Research 94: 10473–10500.

FIGURE 1 Photograph of museum specimens showing the diversity of

extant and recently extinct Hawaiian honeycreepers. Note the diversity in bill shapes, body size, and plumage pattern and color. This represents only a sample of the more than 55 species and the total diversity of the group. Clockwise, from top: Hawaii akepa (Loxops coccineus), akiapolaau (Hemignatus wilsoni), akohekohe (Palmeria dolei), greater

HAZARDS

koa finch (Rhodocanthis palmeri), Nihoa finch (Telespiza ultima), apapane (Himatione sanguinea), Hawaii akialoa (Hemignathus obscunis),

410

SEE EARTHQUAKES; ERUPTIONS; HURRICANES AND

palila (Loxoides balleuxi), and iiwi (Vestiaria coccinea). Photograph by

TYPHOONS; LANDSLIDE; TSUNAMIS

John Steiner.

H O N E YC R E E P E R S , H AWA I I A N

FIGURE 2 Photographs of several species of living Hawaiian honey-

creepers. (A) Nihoa finch (Telespiza ultima); (B) akiapolaau (Hemignathus wilsoni); (C) palila (Loxioides balleuxi); (D) Kauai amakihi (Loxops stejnegeri); (E) iiwi (Vestiaria coccinea); (F) Hawaii creeper (Loxops mana); (G) Hawaii akepa (Loxops coccineus); (H) Maui creeper (Paroreomyza montana). Photographs by Jack Jeffreys.

different songbird families because of their great diversity in morphology and plumage (Figs. 1 and 2). However, ornithologists studying Hawaiian birds around 1900, such as Hans Gadow, Walter Rothschild, and R. C. L. Perkins, proposed that the diverse species were actually all closely related and belonged to a single family, the Drepanididae. They identified a number

of plumage and anatomical characters that united the species, and also noted that most have a distinctive “canvas” odor. More recently, exhaustive osteological and molecular analyses (see below) have largely confirmed the monophyly (descent from a single ancestral species) of the honeycreepers and concluded that the radiation arose from within the cardueline finches (a cosmopolitan group that includes birds such as rosefinches, goldfinches, hawfinches, canaries, and crossbills). Thus, the Hawaiian honeycreepers represent one of the premier examples of insular adaptive radiation within birds. There is a very limited ancient fossil record for birds in the Hawaiian Islands, but a very rich one from the Holocene (about 10,000 years to present). In fact, the discovery and study of these Holocene fossil birds by Smithsonian scientists Storrs Olson and Helen James revealed that massive extinctions of birds had occurred in the Hawaiian Islands subsequent to their colonization by humans some 1000–1500 years ago. These extinctions included at least 23 species of Hawaiian honeycreepers known only from subfossils. Some of the species had unique morphologies not represented among the remaining extant species. The subfossils are found in old lava tubes, limestone caves, and sand dunes, and many have yielded DNA suitable for sequencing. Taxonomists have traditionally divided the honeycreeper species into two or three taxonomic subfamilies or tribes, corresponding to whether they are (a) mostly greenish or black and red, or (b) have finchlike bills (Psittirostrini) or warbler-like or long bills (Hemignathini), or are mostly nectar-feeding (Drepanidini). Based on recent osteological studies by Helen James, the 56–60 known species can be placed in 22 genera, but genericlevel taxonomies by other authors differ. Of these 56–60 species, 33–37 species were known historically (extant or known from museum specimens collected since the eighteenth century). Several species have only recently become extinct. For example, the poouli (Melamprosops phaeosoma) was the most recently discovered Hawaiian honeycreeper. It was found in small numbers on Maui in 1973, and two juvenile specimens were collected by a group of university students. Paleontological studies indicate that it was once one of the most common honeycreepers on Maui at low elevations. Its numbers declined precipitously over the three decades since its discovery, and it was considered extinct in the wild by 2005. The rapid decline and loss of this species serves as a cautionary tale that drastic measures and quick action may often be necessary to keep species from extinction.

H O N E YC R E E P E R S , H AWA I I A N

411

EVOLUTIONARY RELATIONSHIPS BASED ON MORPHOLOGY AND MOLECULES

The phylogeny or evolutionary tree of the Hawaiian honeycreepers is difficult to reconstruct, as in many other adaptive radiations, probably because of a relatively low number of defining morphological characters, considerable evidence for convergent or parallel evolution (homoplasy), and an apparently rapid radiation that likely obscures resolution of many branches of the tree. The Hawaiian honeycreepers represent perhaps the most extreme radiation within a single subfamily or family of birds, with forms that have bills of almost any type known from songbirds (thick finch-like, thin warbler-like, parrot-like, long and woodhewer-like, long and decurved to probe flowers for nectar, etc.), and variable plumages that can contain drab browns, blacks, grays, greens, yellows, and bright reds in many patterns (Fig. 1), but never blue hues or iridescence. One species, the akiapolaau (Hemignathus wilsoni) has a “dualpurpose” woodpecker bill (Figs. 1 and 2), with a short, straight, robust mandible for hammering holes in bark and a longer, decurved maxilla for probing insects from the hole it makes. Most of the honeycreeper species that take nectar have evolved tongues folded into tubes to facilitate nectar flow, convergent to those found in many other nectar-feeding birds in different families or orders. The rapidly expanding field of molecular genetics offers powerful methods of DNA amplification via the polymerase chain reaction (PCR) and DNA sequencing to help unravel the evolutionary relationships of the Hawaiian honeycreepers and to offer a means of estimating the timing of their origin and radiation. Evolutionary trees have been constructed using DNA sequences from most of the extant species, and the power of PCR, applied carefully and with proper controls, has allowed the amplification and sequencing of “ancient” DNA from many extinct or endangered honeycreeper species only available as museum specimens or even subfossil bones (Fig. 3). However, even trees based on a large amount of DNA sequence data do not provide strong statistical support to resolve the branching order, and the pattern of the trees suggests a very rapid initial divergence into different morphological types of honeycreepers. The main Hawaiian Islands follow a progression of ages, with the oldest in the northwest (Kauai at about 5 million years) and the youngest in the southeast (Hawaii at less than 1 million years). These island ages put an age limit on each species, and, with certain assumptions,

412

H O N E YC R E E P E R S , H AWA I I A N

5 changes

Anianiau (K) Poouli (M)† 91 Kauai Nukupuu† 66 Maui Nukupuu† Oahu Nukupuu† Akiapolaau (H) Maui Parrotbill 85 Akohekohe (M) 85 Iiwi (A) Apapane (A) Kauai Amakihi 55 Oahu Amakihi Hawai Amakihi 100 Maui Amakihi Hawaii Akepa 70 Kauai Akepa 59 Hawaii Creeper Kauai Akialoa† 93 Laysan Finch Nihoa Finch Palila (H) Greater Koa Finch (H)† Molokai Creeper† 56 Maui Creeper Oahu Creeper† Kauai Creeper House Finch (non-honeycreeper outgroup)

FIGURE 3 A phylogenetic or evolutionary tree based on mitochondrial

DNA sequences showing the relationships of a sampling of extant and extinct Hawaiian honeycreepers. The numbers along a branch are the support values from a statistical bootstrap analysis; values over 70 indicate good support for the node. Letters in parentheses indicate islands on which the species exists (when not noted in the name of the species; A = all islands, K = Kauai, M = Maui, H = Hawaii). A † indicates a presumed extinct species.

allow the estimation of rates of DNA sequence evolution. These rates, in turn, can be used to estimate the timing of species formation and even the date when the original finch species colonized the islands. Such calculations indicate that the Hawaiian honeycreepers probably began their radiation about 3 to 4 million years ago and that the original colonization must have occurred when Kauai was the youngest (and largest) island. The timing of the radiation (when most of the morphological types evolved) correlates with the formation of Oahu, the next island that formed after Kauai. This suggests that the formation of the new and uninhabited Oahu may have been a major factor initiating or promoting the adaptive radiation. ECOLOGY AND BEHAVIOR

The Hawaiian honeycreepers show tremendous diversity in life history characteristics, feeding mode, and other behavioral and ecological traits. They radiated,

probably from an ancestral finch-billed granivore, to fit into almost every possible ecological niche of songbirds. There were as many as 19 species of ground-feeding granivores (finches or grosbeaks), at least one specialist frugivore, and at least five flower-probing nectarivores (with a few other species more facultative in their nectar feeding). Among the many insectivorous species, some glean prey from foliage, while others probe bark. Several species, including the aforementioned akiapolaau and the Maui parrotbill (Pseudonestor xanthophrys), puncture or tear off bark to find insects within. At least one species, the aforementioned recently extinct poouli, appeared to be a snail specialist but also took other invertebrates, and it had a spatulate tongue well shaped for removing snails from shells or insect prey from crevices. The akepas in the genus Loxops have slightly crossed bill tips, which they use like forceps to separate bracts on the ohia tree (Metrosideros polymorpha) in search of insect prey. The “creepers” include two related genera (Oreomystis and Paroreomyza) and a third species on Hawaii Island (Loxops mana, formerly Oreomystis mana) that is apparently convergent in its “creeping” behavior to the other two genera and is more closely related to the akepas based on osteology and DNA sequences. Two species in the genus Vangulifer, known only from subfossil remains, had a shovel-like bill with a rounded tip, unlike any other known songbird. Upon what and how these extinct species fed can only be speculated, but they may have been aerial flycatchers. Whereas some species are highly specialized in feeding mode or prey type, others feed more generally. Some researchers have suggested that these more generalized species are usually less likely to be endangered with extinction than the more specialized forms. Hawaiian honeycreepers show mostly typical songbird nesting behavior, with most species constructing cup-shaped nests but with a few species known also to nest in cavities. Clutch sizes range from 1 to 4 eggs, and there is considerable variation in other breeding characteristics such as incubation times, growth rates, and egg and nestling survival. Honeycreepers appear to be more highly impacted by introduced mammalian nest predators than are introduced birds, probably because they evolved for millions of years in isolation from such predation pressure. Hawaiian honeycreepers are like most other songbirds and have a socially monogamous mating system. In the case of the palila (Loxioides balleuxi), an endangered, finch-billed species that lives on the island of Hawaii, this

social monogamy is matched by genetic monogamy, as a DNA fingerprinting analysis of parentage revealed no cases of extra-pair fertilization. A similar DNA analysis of parentage conducted in the Pearl and Hermes Reef population of the endangered Laysan finch (Telespiza cantans) revealed that about 12% of the offspring were the result of extra-pair mating. Both of these values are on the low end of the extra-pair mating range for songbirds, but within the range found for other cardueline finches such as American goldfinches (Carduelis tristis), house finches (Carpodacus mexicanus), and serins (Serinus serinus), and for tropical, sedentary songbirds. Why carduelines and drepanidines, or tropical songbirds, may have lower extra-pair mating levels than other songbirds is not known. The Hawaiian honeycreepers also vary considerably in vocalizations. The finch-billed forms sing more melodious songs reminiscent of the songs of canaries and other cardueline finches. Most of the insectivorous honeycreepers sing relatively simple songs, such as the flat-toned trill or rattle of the amakihi or the descending trill of the Kauai creeper (Oreomystis bairdi), and that are in some ways more convergent to songs of warblers or sparrows. And the nectarivorous honeycreepers (such as the iiwi, Vestiaria coccinea, and the apapane, Himatione sanguinea) have amazingly versatile and varied songs, with complex series of pure tones, buzzes, clicks, and trills. The apapane has a particularly varied song, with repertoires containing literally dozens of distinct notes, while the iiwi has one of the most divergent and unusual songs, with highly discordant and “creaky” tones. These variable songs are used for a variety of social purposes: for maintaining contact while in feeding flocks, and to effect agonistic and mating interactions. CONSERVATION AND DISEASE

The Hawaiian honeycreepers have been subjected to a massive Holocene extinction, with nearly 40 of the original 56–60 species disappearing over the past few hundred years. Many of these species disappeared without being discovered by Western naturalists and are known only from subfossil bones. Studies by paleontologists Helen James and Storrs Olson showed that most of these subfossil species disappeared from the fossil record during the period of Polynesian residence in the Hawaiian Islands. This suggests that direct (e.g., hunting) or indirect (e.g., habitat modification, or introduction of predators such as the Polynesian rat, or perhaps disease

H O N E YC R E E P E R S , H AWA I I A N

413

from chickens) impacts from the Polynesians caused these extinctions. Of the 17 Hawaiian honeycreeper species that are believed extant, ten species are currently or will soon be listed as endangered under the U. S. Endangered Species Act (note that an additional six species are on the list, but the consensus of most biologists working in Hawaii is that these are actually extinct). Because of this high level of endangerment of the honeycreepers, and 16 other types of Hawaiian birds, Hawaii has been given the moniker of the “capital of endangered species” in the United States. The primary threats to these extant species include habitat loss and degradation (often from introduced pigs and ungulates), introduced mammalian predators (rats, cats, mongooses), invasive diseases vectored by introduced mosquitoes, and possibly competition from introduced birds and insects. Hawaiian honeycreepers appear to be particularly susceptible to introduced mosquito-vectored diseases such as avian poxvirus and avian malaria (Plasmodium relictum). Infection prevalences and parasitemias are particularly high, and mortality is, unfortunately, a common result of infection. Several species of Hawaiian honeycreepers were infected with introduced malaria in a series of controlled aviary experiments by Richard Warner, Charles van Riper, Carter Atkinson, and their colleagues. One species, the amakihi (Loxops virens), consistently had lower mortality rates, but for the remaining species tested, such as the iiwi (Vestiaria coccinea) and Laysan finch, nearly all individuals succumbed. Introduced birds generally show no mortality and few symptoms when infected with this same malarial strain. As might be expected, amakihi are the only Hawaiian honeycreeper species that regularly breed at lower elevations where Culex mosquitoes and malaria are common, whereas iiwi and most other honeycreeper species are almost nonexistent at these elevations. Genetic structure data suggest that malaria-resistant amakihi survived in small pockets at low elevations and subsequently spread out to fill available habitat, rather than spreading from high to low elevations. Molecular evidence also indicates that the type of avian malaria introduced to Hawaii is widespread across most of the world, particularly on oceanic islands, and likely originated in Africa. Both poxvirus and malaria impact Hawaiian honeycreeper survival, and, along with a variety of additional stressors, continue to challenge this unique and varied taxon with threat of extinction.

414

H U M A N I M PAC T S , P R E - E U R O P E A N

SEE ALSO THE FOLLOWING ARTICLES

Bird Disease / Bird Radiations / Fossil Birds / Galápagos Finches / Hawaiian Islands, Biology FURTHER READING

Beadell, J. S., F. Ishtiaq, R. Covas, M. Melo, B. H. Warren, C. T. Atkinson, T. Bensch, G. R. Graves, Y. V. Jhala, M. A. Peirce, A. R. Rahmani, D. M. Fonseca, and R. C. Fleischer. 2006. Global phylogeographic limits of Hawaii’s avian malaria. Proceedings of the Royal Society, B 273: 2935–2944. Fleischer, R. C., C. E. McIntosh, and C. L. Tarr. 1998. Evolution on a volcanic conveyor belt: using phylogeographic reconstructions and K-Ar based ages of the Hawaiian Islands to estimate molecular evolutionary rates. Molecular Ecology 7: 533–545. Fleischer, R. C., C. L. Tarr, H. F. James, B. Slikas, and C. E. McIntosh. 2001. Phylogenetic placement of the po‘o-uli Melamprosops phaeosoma based on mitochondrial DNA sequence and osteological characters. Studies in Avian Biology 22: 98–103. Foster, J. T., B. L. Woodworth, L. E. Eggert, P. J. Hart, D. Palmer, D. C. Duffy, and R. C. Fleischer. 2007. Genetic structure and evolved malaria resistance in Hawaiian honeycreepers. Molecular Ecology 16: 4738–4746. James, H. F. 2004. The osteology and phylogeny of the Hawaiian finch radiation (Fringillidae: Drepanidini), including extinct taxa. Zoological Journal of the Linnean Society 141: 207–255. James, H. F., and S. L. Olson. 1991. Descriptions of thirty-two new species of birds from the Hawaiian Islands: Part 2. Passeriformes. Ornithological Monographs 46: 1–88. Jarvi, S. I., C. T. Atkinson, and R. C. Fleischer. 2001. Immunogenetics and resistance to avian malaria (Plasmodium relictum) in Hawaiian honeycreepers. Studies in Avian Biology 22: 254–263. Pratt, H. D. 2005. The Hawaiian honeycreepers. Oxford: Oxford University Press. Scott, J. M., S. Mountainspring, F. L. Ramsey, and C. B. Kepler. 1986. Forest bird communities of the Hawaiian Islands: their dynamics, ecology and conservation. Studies in Avian Biology 9. Lawrence, KS: Allen Press. van Riper, C. III, S. G. van Riper, M. L. Goff, and M. Laird. 1986. The epizootiology and ecological significance of malaria in Hawaiian land birds. Ecological Monographs 56: 327–344. Warner, R. E. 1968. The role of introduced diseases in the extinction of the endemic Hawaiian avifauna. Condor 70: 101–120.

HUMAN IMPACTS, PRE-EUROPEAN PATRICK V. KIRCH University of California, Berkeley

Islands, and especially truly oceanic islands (such as those situated on the Pacific Plate), offer numerous historical “experiments” of interactions between previously isolated and often vulnerable ecosystems and

colonizing human populations. Because of isolation, biotic disharmony, and lack of competition, island biotas are characterized by high species-level endemism and vulnerability to invasive taxa. Pioneering human populations, such as those of the Lapita and later Polynesian groups, introduced a portmanteau biota along with cultural concepts of land use and ecosystem management. Over time spans ranging from greater than 40,000 to less than 1,000 years, these invading human populations and their “transported landscapes” irreversibly altered these fragile and previously isolated island ecosystems. TAXONOMY OF HUMAN IMPACTS

Indigenous human populations on islands throughout the Pacific, Indian, and Caribbean Oceans altered their environments in ways that can be characterized as both direct and indirect, with various subcategories, as indicated in Fig. 1. Direct impacts are those that result from a variety of consciously directed human actions in the ecosystem, including hunting and gathering of a variety of plant and animal resources, forest clearance and vegetation modification (through both clearing and the use of fire), introduction and planting of agricultural crops and other useful plants, and a wide array of often permanent physical modifications of the landscape, such as

FIGURE 1 A taxonomy of human impacts to island environments.

terracing slopes, building fish ponds, and digging canals and drains. Such physical manipulation of landscapes often occurred some time after initial island colonization, when human population density had reached relatively high levels, and efforts were made to intensify food production. Indirect impacts derive in the first instance from the array of portmanteau biota that typically accompanies the movement of a human population into virgin territory. Human bodies themselves may carry disease pathogens and ectoparasites, and along with them go domestic animals, crop and other useful plants, weeds, vermin, and other synanthropic species. This array of synanthropic plants and animals often included species that were highly competitive with the vulnerable native biota of remote oceanic ecosystems. A prime example is the Pacific rat (Rattus exulans), which was spread by humans to virtually every island group in the Pacific, and which has been implicated in ecological changes on islands ranging from Rapa Nui (Easter Island) to Hawai‘i. TRANSPORTED LANDSCAPES

Botanist Edgar Anderson originally introduced the concept of “man’s transported landscapes” to refer to the fact that when people move into a new area, they carry with them not just physical artifacts, but typically a host of plants and animals, including crops, weeds, and vermin. In addition, they bring cognitive models of how a landscape should be managed, how it should look, and how it should be manipulated. Historian Andrew Crosby extended the notion of transported landscapes with his analysis of the imperial expansion of the West (“ecological imperialism”), and coined the term “portmanteau biota” to refer to the assemblage of plants and animals that accompany expanding human populations. These concepts apply equally well to prehistoric peoples who migrated from continents and larger island masses out into truly oceanic islands. In the case of the early Polynesians, for example, the portmanteau biota carried on double-hulled voyaging canoes and introduced to islands included pigs, dog, and chickens as purposeful introductions; a small rat (Rattus exulans) and several species of gecko and skink (Lepidodactylus lugubris, Gehyra mutilata, Cryptoblepharus poecilopleurus, and other species) as probable inadvertent “stowaways”; at least 27 species of crop, narcotic, medicinal, or otherwise useful plants; an unknown number of weed species; and several

H U M A N I M PAC T S , P R E - E U R O P E A N

415

kinds of terrestrial gastropods (including Alopeas gracile, Lamellidea pusilla, and Gastrocopta pediculus) the last of these probably adhering to crop plants or existing in associated soil. Some of these newly introduced species quickly escaped beyond the confines of human settlements and began to compete with or have impacts on the native biota. For example, rapid population increases in the Pacific rat R. exulans have been invoked as a likely cause for rapid dryland forest reduction in the Hawaiian Islands, and possibly also on Easter Island (Rapa Nui), through the rats’ consumption of seeds and young shoots. The early Polynesians also brought with them cognitive models of land use and management. These included the use of swidden or shifting cultivation (“slash-andburn”) agriculture, in which natural forest vegetation is cut, dried, and fired prior to planting, as well as the planting of taro and other aroids in naturally swampy terrain, or in terraces constructed along stream banks. These cognitive models also included concepts of land ownership and division, with permanent house sites and garden lands forming hereditary estates held by extended family groups. Applying these land-use concepts, the Polynesians transformed island after island into highly managed, anthropogenic landscapes. Recognizing that society and nature become inextricably interconnected in such complex landscapes, following C. Barton we may call them socioecosystems. HUMAN TRANSFORMATION OF ISLAND ECOSYSTEMS

Pre-European transformation of island ecosystems by indigenous human populations has now been well documented through archaeological and associated paleoecological research on many islands in the Pacific, the Caribbean, and the Mediterranean, and on the large island of Madagascar. This article draws mainly on Pacific island examples, but the same fundamental processes can be found on islands throughout the world. Deforestation was a frequent consequence of preEuropean land-use practices, including shifting cultivation, and the use of fire. On many islands in the humid tropics lacking active volcanoes, natural fires appear to have been largely absent prior to human arrival. This is indicated by the absence of microscopic charcoal particles in sediment cores for layers predating human occupation, and by sharp influxes of such charcoal particles following the arrival of humans. Pollen grains extracted from the same sediment cores (on such islands as New

416

H U M A N I M PAC T S , P R E - E U R O P E A N

Caledonia, Fiji, Atiu, Mangaia, O‘ahu, and Rapa Nui) typically display radical changes in their floristic composition, likewise correlating with the period following initial human settlement. Native arboreal taxa show significant declines whereas pyrophytic or fire-adapted plants (such as Dicranopteris ferns and Pandanus) increase. The extent and degree of deforestation appears to have varied depending on a number of factors, including island age and consequent nutrient depletion of soils, climate and frequency of drought, human landuse practices, and the size and density of human populations. Some younger volcanic islands such as Tahiti or Rarotonga seem to have been fairly resilient to intensive land use, whereas others (particularly geologically older islands) such as Mangaia, Mangareva, and Rapa Nui were extensively deforested. On the large, nearcontinental islands of New Zealand, pollen analysis has demonstrated that vast areas were deforested following Polynesian arrival around AD 1200. In part, this may have been due to burning to drive and hunt the large flightless moa birds, and in later times to encourage the growth of indigenous bracken fern, an important food resource. In part because of deforestation and habitat destruction, and in part because of the direct pressure of hunting for food and feathers (and in some cases also because of predation by rats on ground-nesting eggs and chicks), island bird populations frequently underwent substantial declines, leading to the extirpation and even extinction of many species. Some island birds evolved flightlessness (including flightless ducks and ibises in Hawai‘i, flightless megapodes in New Caledonia, and the famous moa of New Zealand), and these taxa were especially vulnerable to predation and extinction. Archaeological and paleontological evidence for such major impact on avifauna is now well documented for many Pacific islands. One of the most dramatic cases is New Zealand, where at least 13 species of moa (in the genus Dinornis and other taxa), the largest standing up to 3 m tall with neck extended, were driven to extinction in probably less than 200 years after Polynesian colonization. But many other islands, such as Mangareva, Rapa Nui, and Tahuata, show dramatic decreases in bird populations such those of nesting seabirds (petrels, shearwaters, terns) that were evidently abundant in large numbers prior to human arrival but became rare or extirpated thereafter. Human impacts on native biota extended well beyond birds. On some islands, particularly where the

areas of exploitable reef or lagoon were limited, archaeologists have detected evidence for resource depression in marine fauna. On Mangaia Island, for example, this is demonstrated both in significantly diminished size distributions in shellfish (Turbo sp.) and in the fish. In Hawai‘i, similar impacts have been detected for late prehistoric harvesting pressure on the prized limpet species Cellana exarata. On other islands, however, such as Mangareva, marine resources appear to have been more resilient and able to withstand continued pressure of human exploitation without measurable impacts. The degree to which pre-European human populations irreversibly transformed island ecosystems depended on a number of factors. As noted above, island age and nutrient status was clearly one factor influencing the extent of deforestation. Another important factor was the overall size and density of the human population. In late prehistory, many islands in Remote Oceania saw their human populations attain density levels of 200 or more persons per square kilometer. Under such high population levels, intensification of production systems and conversion of island landscapes to highly managed socioecosystems was inevitable. In some cases, trophic competition between humans and domestic animals (especially pigs) became such that the latter were eliminated (this is now documented for Tikopia, Mangaia, and Mangareva). IMPLICATIONS FOR BIOGEOGRAPHY

As recently as the 1960s, many naturalists and anthropologists alike assumed that the pre-European populations of the world’s islands had had relatively little impact on these insular ecosystems. This viewpoint was a holdover of the eighteenth-century Rousseauian myth of the “noble savage” and can no longer be sustained. Every island that was settled for any length of time in prehistory, and which has now been subjected to archaeological and paleoecological study, has been shown to have had some degree of irreversible impact. In many cases the effect on native biota, through deforestation, species loss, and resource depression, was substantial. For biogeographical studies, this means that the historically recorded data on biodiversity, taxonomic richness, and so on cannot be taken as representative of the truly natural or pristine conditions on islands. Rather, it is essential that paleoecological and archaeological data be used to establish the actual biotic conditions on particular islands as a baseline against which both

pre-European and post-European human impacts can be assessed. INDIGENOUS MANAGEMENT OF LANDSCAPES

The fact that pre-European populations extensively modified island ecosystems through their mix of direct and indirect impacts has now been empirically documented for many islands. The historical record of human impacts should not, however, be misconstrued to mean that indigenous populations were some sort of “eco-vandals” intent on destroying their environments, even though in some cases the consequences of cumulative human actions were indeed calamitous (as on Rapa Nui or Mangaia). To the contrary, island peoples struggled to bring their necessary exploitation of island resources into balance with human population numbers and developed various cultural approaches to land and resource management. On many Pacific islands, pre-European populations developed intensified methods of agricultural production, especially terracing and irrigation, which allowed some landscapes to be put into near-continuous cropping and food production. In the Hawaiian Islands, for example, alluvial valley soils and colluvial slopes were sculpted into systems of pondfields, with stone-faced embankments, which were flooded with water diverted from streams through stone-lined canals. These pondfield complexes were used to cultivate taro (Colocasia esculenta), which responds to such irrigation with high yields. In New Caledonia taro irrigation was also practiced, with earth-banked terraces extending up even very steep hillsides. In addition, the New Caledonians developed intensive yam (Dioscorea spp.) culture, planting these xerophytic crops in extensive linear mounds. On other islands, such as Tikopia and Kosrae, and in the Marquesas, land management practices emphasized arboriculture or tree cropping, especially of breadfruit (Artocarpus altilis). The Tikopia system of orchard gardening is especially remarkable for the way that it artificially mimics the multi-storey structure of a tropical rainforest. Tikopia appears to represent a true case of long-term sustainability of an indigenous economic production system. Resource management also extended to reef and lagoon resources. Many island cultures imposed regular prohibitions (tapu, rahui) on the taking of certain fish or shellfish resources, in order to allow stocks to replenish. In the Hawaiian Islands, true pond aquaculture was developed for the raising of mullet (Mugil cephalis) and

H U M A N I M PAC T S , P R E - E U R O P E A N

417

milkfish (Chanos chanos) in large ponds whose rock walls extended out onto shallow reef flats. More than 400 of these ponds have been archaeologically identified, and they probably greatly augmented pre-European subsistence production. SEE ALSO THE FOLLOWING ARTICLES

Deforestation / Easter Island / Extinction / Flightlessness / Introduced Species / Peopling the Pacific FURTHER READING

Dodson, J., ed. 1992. The naive lands: prehistory and environmental change in Australia and the Southwest Pacific. Melbourne: Longman Chesire. Fosberg, R., 1963. Man’s place in the island ecosystem: a symposium. Honolulu: Bishop Museum Press. Kirch, P. V. 1997. Microcosmic histories: island perspectives on ‘global change.’ American Anthropologist 99: 30–42. Kirch, P. V. 2007. Three islands and an archipelago: reciprocal interactions between humans and island ecosystems in Polynesia. Earth and Environmental Science Transactions of the Royal Society of Edinburgh 98: 85–99. Kirch, P. V., and T. L. Hunt, eds. 1997. Historical ecology in the Pacific Islands: prehistoric environmental and landscape change. New Haven, CT: Yale University Press. Rolett, B., and J. Diamond. 2004. Environmental predictors of preEuropean deforestation on Pacific Islands. Nature 431: 443–446. Steadman, D. W. 2006. Extinction and biogeography of tropical Pacific birds. Chicago: University of Chicago Press.

HURRICANES AND TYPHOONS THOMAS A. SCHROEDER University of Hawaii, Manoa

“Hurricane” and “typhoon” are regional terms for intense tropical cyclones. Tropical cyclones are warmcore vortices that develop over tropical oceans. Tropical cyclones are more intense and compact than the extratropical cyclones of middle latitudes (Fig. 1). They impact islands through damaging winds, heavy rains, and coastal flooding. INTENSITY CLASSIFICATION

Tropical cyclone intensity is determined from either sealevel pressure in the storm center or maximum sustained winds anywhere in the vortex. Intensity class terminol-

418

HURRICANES AND TYPHOONS

FIGURE 1 Surface weather map for January 8, 1980. Superimposed on

the 1980 map is Hurricane Iniki (September 11, 1992). The area of Iniki within the 1004 mb (hPa) isobar is stippled. The equivalent isobar for the 1980 winter storm is in bold. Note the differences in scale of the two systems. Wind symbols (feathered) represent winds reported on January 8, 1980. Meteorological convention is to plot isobars as two digits, where 96 = 996 mb and 04 = 1004 mb.

ogy is basin- and/or forecast-agency dependent. The United States Navy/Air Force Joint Typhoon Warning Center, which issues forecasts for U. S. interests in the western North Pacific Ocean, classifies systems as depressions (winds less than 17 m/s), tropical storms (winds between 17 and 32 m/s) and typhoons (winds greater than 32 m/s) . The Japanese Meteorological Agency “typhoon” includes U.S.-defined “tropical storm” and “typhoon.” Sustained winds may be one-minute average (U.S. practice) or ten-minute average. This article uses U. S. Atlantic basin (“hurricane” instead of “typhoon”) classifications. STRUCTURE

A hurricane consists of three regions (Fig. 2): 1. A relatively calm and clear central “eye.” The winds are relatively light and speeds increase linearly with increasing radius. Strong sinking motion limits cloud formation. There may be some low clouds and thin, high overcast. Development of an eye is considered to indicate intensification from “tropical storm” to

“hurricane” intensity. Typical eye diameters range from 20 km to 60 km. 2. A “core” or “eyewall.” Winds attain maximum intensity, and a solid band of heavy rains normally completely encircles the eye. Smaller intense vortices may exist within the eyewall. Typical eyewall thickness is 20 km to 60 km. 3. An outer region. Winds gradually diminish, and rain squalls (“rain bands”) are common. The outer region may extend 500 km from the storm center. N 7 6 5

Storm genesis is poorly observed, and the mechanism is debated. Scientists agree upon the fundamental conditions for hurricane formation. These include a minimum sea-surface temperature, a minimum depth of the oceanic mixed layer (a measure of heat content), critical thermodynamic structures in the atmosphere, critical vertical shear of the horizontal winds, and a preexisting seedling disturbance. The necessity that each condition be met explains the rarity of hurricane development. Annual global storm totals have been constant, but location and frequency of genesis within and among basins varies. One controlling factor is the El Niño– Southern Oscillation (ENSO), which in its warm phase shifts Northwest Pacific typhoon formation zones and inhibits Atlantic hurricanes. Theory suggests that current global warming should lead to a detectable increase in hurricane intensity. IMPACTS UPON ISLANDS

1 4 2

3 56

FIGURE 2 Schematic of a hurricane approaching a mountainous

island. Featured are (1) eye wall (stippling indicates radar signature),

Hurricanes bring damaging winds, coastal flooding, and torrential rains. The degree of impact varies with the size and topography of the island, with offshore and near shore bathymetry, and with the path of the storm relative to the island. Hurricane winds destroy vegetation and infrastructure (Fig. 3). These winds extend upward for 3 km. Thus, even on large, high islands, extreme winds penetrate well inland. Storms have easily transited Hispaniola, Jamaica, and New Caledonia with minimal weakening. In steep topography,

(2) rain band (also stippled), (3) location of maximum sustained winds, (4) waves propagating ahead of the storm, (5) point of expected peak coastal inundation, (6) enhanced rainfall, and (7) likely downslope winds.

The hurricane just described is circularly symmetric. Actual storms are asymmetric. The strongest asymmetry arises from storm motion, which augments the winds in the right-hand sector (relative to storm motion). FREQUENCIES, FORMATION (GENESIS), AND VARIABILITY

Tropical cyclones are rare events. Annually about 85 storms of all intensity classes occur. One to two tropical cyclones are present daily over an area comprising 50% of the Earth’s surface. Storm frequency varies among ocean basins. The most active, in order, are the western North Pacific, the eastern North Pacific, the Southwest Pacific and Australia, the Southwest Indian, and the North Atlantic.

FIGURE 3 Hurricane winds battering palm trees on Key West, Florida.

Photograph taken at the Weather Forecast Office in Key West, Florida, during the height of Hurricane Charley on Friday, August 13, 2004. Photograph courtesy of the National Weather Service, Key West, Florida.

HURRICANES AND TYPHOONS

419

enhanced downslope winds may devastate supposedly sheltered lee sides (e.g., Hurricane Iniki, 1992). Embedded in the eyewall and outlying rain bands are intense tornadic vortices. Eyewall vortices were prominent in both Hurricanes Andrew and Iniki in 1992. Rainband tornadoes were observed in the Florida Keys in 1972 during the passage of Hurricane Agnes well to the west. Coastal flooding is caused by a dome of water (the “storm surge”) which grows at the center and right flank of a moving storm. The low pressure in the eye causes an “inverted barometer” which raises the sea level. Additionally, the winds in the right front sector pile water ahead of the storm. Water may rise even higher, depending upon the nearshore bathymetry. The U. S. Gulf Coast, with its broad shelf, favors large storm surges. In 1900 the Galveston Hurricane killed over 6000, all by storm surge. The Hawaiian Islands feature fringing reefs and no significant shelf, but they have nevertheless experienced significant flooding. Hurricane Iniki caused an extreme high-water level of 9.1 m through a combination of surge, high tide during landfall, and run-up. On high islands coastal inundation is limited to narrow coastal strips. However, low-lying atolls and barrier islands may be completely inundated. Wave action damages reefs and may drastically alter local bathymetry. In one notable instance, Hurricane Bebe (1972) created a completely new debris rampart (the Bebe Rampart) 35 km long, 35 m thick, and up to 3.5 m high along the east side of Funafuti Atoll, Tuvalu. Hurricanes are prolific rain producers. Precipitation accumulations vary with island topography and speed of storm passage. Most world record rains for periods from one to 15 days have fallen at Réunion Island in the Southwest Indian Ocean. Réunion is a high island featuring a 2.5-km shield volcano. Records range from 1869 mm for 24 hours to 6433 mm for 15 days. The latter occurred as a tropical cyclone (Hyacinthe, 1980) executed two complete loops around the island. The combination of surge and torrential rains may produce extreme flooding of coastal river deltas and barrier islands. Stream waters running down to the sea encounter elevated sea level. This combination has been especially deadly in the northern Bay of Bengal. SEE ALSO THE FOLLOWING ARTICLES

Barrier Islands / Climate Change / Climate on Islands / Global Warming / Surf in the Tropics

420

HYDROLOGY

FURTHER READING

Anthes, R. A., R. W. Corell, G. Holland, J. W. Hurrell, M. C. McCracken, and K. E. Trenberth. 2006. Hurricanes and global warming: potential linkages and consequences. Bulletin of the American Meteorological Society 87: 623–628. Elsberry, R. L., ed. 1995. Global perspectives on tropical cyclones. Report No. TCP-38. World Meteorological Organization. Emmanuel, K. E. 2005. Divine wind: the history and science of hurricanes. New York: Oxford University Press. Fletcher, C. H., B. M. Richmond, G. M. Barnes, and T. A. Schroeder. 1995. Marine flooding on the coast of Kaua‘i during Hurricane Iniki: hindcasting inundation components and delineating washover. Journal of Coastal Research 11: 188–204. Larson, E. 1999. Isaac’s storm. New York: Crown Publishers. Maragos, J. E., G. B. K. Baines, and P. J. Beveridge. 1973. Tropical Cyclone Bebe creates a new land formation on Funafuti Atoll. Science 181: 1161–1164. Pielke, R. A., Jr., C. Landsea, M. Mayfield, J. Laver, and R. Pasch. 2005. Hurricanes and global warming. Bulletin of the American Meteorological Society 86: 1571–1575. Simpson, R. H., ed. 2003. Hurricane! Coping with disaster. Washington, DC: American Geophysical Union. Simpson, R. H., and H. Riehl. 1981. The hurricane and its impact. Baton Rouge, LA: LSU Press.

HYDROLOGY CHRISTIAN DEPRAETERE Global Islands Network, Grenoble, France

MARC MORELL Institut de Recherche pour le Développement, Fort de France, Martinique

Freshwater is often a critical resource on islands. Island hydrology takes into account the specific effects of the surrounding ocean on the physical processes and water resources budget of islands, and the specific approaches and considerations required to understand and manage freshwater on islands. UNIQUE CHARACTERISTICS OF ISLAND HYDROLOGY

Islands are special compared to continents when it comes to water resources. This is a consequence of the defining characteristics of islands: their limited size and remoteness from large sources of freshwater. In general, groundwater is the main freshwater resource, although its predominance depends on the area and relief of the island: the smaller and the lower the island, the greater

the relative importance of groundwater compared to other sources. Island hydrological contexts are also as diverse as their continental counterparts. A global survey of islands shows that they can be found in all climatic zones and correspond to a wide range of geological and ecological settings. Nevertheless, it is a geographic fact that they are more directly exposed to marine atmospheric advection, with humid air constantly passing over the island, and undergo less seasonal variation in temperature than larger land masses. Although islands also have great geological diversity, a majority of them are composed of volcanic rock, coralline limestone, or alluvial deposits as direct consequence of specific oceanic, coastal, and tectonic processes. WATER RESOURCES ON LOW-LYING AND MOUNTAINOUS TROPICAL ISLANDS

The tropical zone includes the contrasted cases of flat low-lying coral atolls and motu (the Tuamotus, the Bahamas, the Maldives) and steep mountainous volcanic islands (Hawaii, the Cape Verde archipelago, the Comoros). On coral islands, all the precipitation immediately percolates and gets stored inside the porous carbonate rocks. Pumping is the only way to get water to the surface. The soils and geological basement of volcanic islands also have high infiltration rates, but during large storms and cyclones the heavy rainfall produces rapid surface runoff on steep slopes, which generates devastating flash floods in the foothills and on coastal plains. In both island types, the use of surface water is limited if not impossible. Both coral and volcanic tropical islands are therefore mostly dependent on their groundwater reserves. Exploiting this groundwater resource is particularly complicated in the island context. The first difficulty comes from rainfall representing the input variable into the island hydrologic system. The direct advection of humid tropical air masses over mountains produces very complex rain patterns and strong orographic effects that make precise monitoring and estimation of rainfall difficult at the island scale. Another major drawback for coastal groundwater lies in the fragile equilibrium of the freshwater lens with the surrounding seawater. This body of water is also called a “Ghyben–Herzberg lens,” after the two scientists who described how rainwater that percolates into the porous rock and sand floats on the saltwater beneath, depressing it into a profile in the shape of a lens. The form of the

lens depends on the hydrostatic equilibrium between the density of saltwater (es) and freshwater (ef ): Hb = Ha ef /(es − ef ) where Hb and Ha are respectively the depths of freshwater below and above sea level. The difference of density between saltwater (es = 1.025) and freshwater (ef = 1.0) is directly responsible for the thickness of the water table at a specific point. For example, if the top stands at 1 m above sea level, then the bottom would be 40 m below sea level. The main challenge is to ensure that pumping from this floating lens will not contaminate it with underlying brackish water. When excessive pumping creates a saltwater wedge, recovery is a slow process with high impact on the island’s economy and people. Island groundwater is a fragile resource that is difficult to estimate and shows limited resilience. AN EXAMPLE OF CONTRASTING ISLANDS WITHIN A TROPICAL ARCHIPELAGO

The archipelago of Guadeloupe in the French Caribbean may be taken as a case study of contrasting hydrological regimes in tropical islands (Fig. 1). On the eastern, windward side are several flat, low-lying islands made of limestone. Consequently, the annual rainfall is below 1500 mm/yr due to the lack of orographic uplift, and the drainage network is poor with no perennial rivers. These properties are typical of dry islands with their water resources mainly stored in underground lenses. Conversely, the western, downwind island is volcanic and features the large mountain of Basse-Terre, which is the major hydrological feature of the whole archipelago. The windward eastern slopes of Basse-Terre experience a drastic rainfall gradient from 2500 mm at sea level up to 11,000 mm at 1400 meters elevation, while the coast on the leeward side receives less than 1400 mm. Although a tropical island represents a small surface area, it usually presents extreme hydroclimatic contrasts. OVERVIEW OF THE HYDROLOGY OF ISLANDS ACROSS ALL CLIMATIC ZONES

Outside the humid tropical zone, islands face the whole spectrum of hydrological situations depending on the climate. A simple hydroclimatic classification of all the world’s islands larger than 0.1 km2 is given in Table 1. In polar regions, all precipitation occurs as snow, permafrost makes the soil impermeable, and no runoff occurs. At subpolar latitudes, melting snow and perma-

HYDROLOGY

421

West

East Florida

Bahamas

Cuba Puerto Rice

Max elevation (meters)

Guadeloupe

Basse-Terre

Venezuela

Isohyet (mm/year) elevation >400 meters

Trinidad

Watershed

La Désirade

Grande-Terre

Marie-Galante

Les Saintes

Upwind

Downwind

Wet and montainous island with volcanic basement Leeward

Windward

Elevation (meters)

>11,000mm/y.

Dry and low-lying island with coral basement Hydrological fluxes and wter bodies Average rainfall 40

00m m/y .

>250

20 °C

Dry P < 400 mm

Wrangel (RU) 6979

Spitsbergen (NO) 2366

Isla Porcia (CL) 108

Lanzarote (ES) 319

Socotra (YE) 2675

12447

Moderate 400 mm < P < 1000 mm

Peter the Great (AQ) 520

Disco (GL) 7459

Jan Mayen (NO) 10974

Sicily (IT) 2730

Bahamas (BS) 2773

24456

Sub-humid 1000 mm < P < 1500 mm

Balleny Islands (AQ) 78

South Sandwich (UK) 174

Iceland (IS) 7927

Tasmania (AU) 3241

Puerto-Rico (PR) 5287

16707

Humid 1500 mm < P < 2000 mm

White Island (AQ) 26

0

Kodiak (US) 4318

Honshu (JP) 1307

Hainan (CN) 5877

11528

Chiloe (CL) 1399

Java (ID) 13975

21170

8996

30587

86308

Very humid P > 2000 mm

0

0

Vancouver (CA) 5796

Total

7603

9999

29123

Total

note: The number of islands is given for each class of temperature (T ) and precipitation (P ). For each class, the name of an island is given as an example along with its ISO country code: AQ = Antarctica, AU = Australia, BS = Bahamas, CA = Canada, CL = Chile, CN = China, ES = Spain, GL = Greenland, ID = Indonesia, IS = Iceland, IT = Italy, JP = Japan, NO = Norway, RU = Russia, PR = Puerto-Rico, UK = United-Kingdom, US = United-States, YE = Yemen.

high-cost alternatives such as desalinization of seawater, piping or barging freshwater from the mainland. THE LIMITATIONS PLACED ON ISLANDS BY WATER RESOURCES

The UNESCO report on Hydrology and Water Resources of Small Islands (1991) is a major milestone in the formal recognition of the specificity of islands in relation to water. The report concludes that the problem of water resources on islands is “usually very serious.” The carrying capacity of an island is often determined by available water and is highly vulnerable to the forces of nature. Scientists have highlighted in the International Panel on Climate Change (IPCC) reports that expected climate change and sea level rise will have far-reaching consequences on coastal aquifers. For instance, the atoll groundwater of Kiribati is so reduced by El Niño–Southern Oscillation (ENSO)-related droughts that domestic water wells become too salty to drink. The case of the highly populated, low-lying islands of the Maldives (max elevation 2.3 m!) illustrates the complex problem of sanitation and the thin water lens. With 66,000 inhabitants living on 1.77 km2, the islet of Male is exclusively urban, and, despite a great deal of effort by the authorities, preserving the shallow lens from pollutants and sewage is extremely difficult. There is no doubt that islands more than elsewhere are limited in their development by hydrological issues, both in terms of freshwater quantity and quality. This is true both for insular microstates and for any other island ruled

by or dependent on continental countries. The challenge of sustainable development in the insular context is therefore largely dependent on the capacity of local people to manage this crucial and fragile resource for the dual benefit of ecology and economy. ISLANDS AS BELLWETHERS OF WATER ISSUES

A political leader has recently declared that “Islands are the bellwethers of international environmental policy” at the Global Islands Partnership meeting (GLISPA meeting, 2007). This statement is particularly true for water issues. It is emblematically exemplified in the film The Naked Island by Kaneto Shindo (1961, Japan). The action takes place on a small and steep island on which the everyday life of a poor peasant family focuses on collecting water from the neighboring mainland to water their garden on their dry patch of land. Far from the cliché of the Garden of Eden, with water portrayed as the film’s heroine, it stresses the hardship of island conditions when hydrologic resources are limited. SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Freshwater Habitats / Maldives / Socotra / Sustainability FURTHER READING

Baldacchino, G. 2007. A world of islands: An island studies reader. Malta: Agenda, 2007. Falkland, A. C. 1991. Hydrology and Water Resources of Small Islands: A Practical Guide. IHP Studies and Reports in Hydrology No. 49. Paris: UNESCO.

HYDROLOGY

423

Michel, J. A. 2007, Keynote address on the opening of the First Global Island Partnership Strategy Meeting. http://www.cbd.int/island/glispa.shtml, Shindo, K. 1960. Hadaka no shima. (released in English [1962] as The Naked Island). BAFTA Awards from Moscow International Film Festival (1961) and BAFTA (1963).

HYDROTHERMAL VENTS ROBERT C. VRIJENHOEK Monterey Bay Aquarium Research Institute, Moss Landing, California

Deep-sea hydrothermal vents are submarine hot springs located along the global mid-ocean ridge system, in backarc basins, and on volcanic seamounts. The vents support lush animal communities fueled by reduced sulfur compounds and methane, rather than sunlight. Chemosynthetic microbes are the primary producers at vents, and they, in turn, are grazed and filtered by a variety of animals or hosted as symbionts by others. Dependence on geochemical energy restricts vent-endemic animals to small island-like habitats scattered throughout the world’s oceans. Studies of larval development and ocean circulation coupled with population genetic analyses have revealed a range of physical and biological factors that either facilitate or impede the dispersal of vent animals among these islands. Although some ancient taxa occur at vents, phy-

logenetic analyses of the dominant vent organisms suggest that these chemosynthetic deep-sea habitats are not stable refugia for living fossils. Instead, species turnover may not differ much from that found in other marine environments. VENT ISLANDS

Since Darwin’s epic voyage on the HMS Beagle, the Galápagos Islands have played a seminal role in the growth of our ideas about island biogeography and organic evolution. It is fitting, therefore, that deep-sea hydrothermal vents were discovered along the Galápagos Spreading Center. In 1977, geologists J. Corliss and R. Ballard reported dense aggregations of meter-long tubeworms, giant clams, mussels, crabs, and unusual eel-like fish thriving around submarine hot springs laden with toxic volcanic gases. Hydrothermal vents have since been found along most explored portions of the global mid-ocean ridge system, a 55,000-km-long mountain chain that circles the globe (Fig. 1). Vent communities also occur in back-arc spreading centers and on hydrothermally active seamounts. The discovery of chemosynthetic communities flourishing in darkness, extreme temperatures, and immense pressures of the deep sea changed our views about the limits to life on Earth and opened our minds to the potential for life in extraterrestrial environments. Most vent animals depend entirely on chemosynthetic primary productivity, constraining them to live around submarine hot springs. Consequently, these inhabitants are sus-

FIGURE 1 Global distribution of well known hydrothermal vent communities. Colors represent biogeographic provinces: dark blue, East Pacific

Rise; green, northeast Pacific; pink, western Pacific; red, Mid-Atlantic Ridge; yellow, Azores Plateau; orange, Central Indian Ridge. Spreading centers are shown with double lines, and areas of subduction are marked with arrowheads that point in the direction of subduction. The apparent absence of vents at higher northern and southern latitudes reflects limited exploration in these remote regions.

424

HYDROTHERMAL VENTS

V

V

B

V

Ex p Rid lore ge r

V

V V

nco

V

Sova

Fractu re Zo ne Ende avou r Seg ment Cobb Segm ent Vance Segm ent Cleft Segm ent Blan co T rans form Fault V

Endeavour Segment

V

V

V V V

V

V

V

V

Portland

Mendocino Fracture Zone

V V

V V

1 km

Gor da R

idg

e

Jua nd

10 k

m

e Fu

ca R

idg

e

V

V

V

Depending on their unique life histories and modes of dispersal, various vent species are differentially affected by disruptions in the ridge system. Most of the annelids and mollusks are sessile or sedentary as adults and disperse as larvae. The mussel Bathymodiolus thermophilus produces numerous small eggs (∼50 µm) that hatch into swimming planktotrophic larvae, but the altitude they achieve in the water column is not known. In contrast, the palm worm Alvinella pompejana produces large eggs (∼200 µm) that hatch into lecithotrophic larvae, which are capable of arresting development in cold abyssal waters and continuing development upon encountering warm waters. The giant tubeworm Riftia pachyptila produces intermediate sized eggs (80–100 µm) with sufficient yolk to live for about one month in cold

V

DISPERSAL BARRIERS

A

V

ceptible to sporadic volcanic eruptions and tectonic events that extinguish old vents and create new ones. The spacing between active vents and the tempo of habitat turnover are linked to the spreading rates of ridge segments. Large transform faults with extended fracture zones occur at frequent intervals along slow-spreading ridge systems like the MidAtlantic Ridge. Without active hydrothermal venting, these large offsets are expected to disrupt animal dispersal along a ridge system, particularly if propagules are transported along rift valleys formed between mountainous walls of the ridge axis (Fig. 2). On the other hand, buoyant larvae produced by some vent animals may disperse in hydrothermal plumes that rise above the axial walls. Occasional megaplumes associated with volcanic events are hypothesized to aid longdistance dispersal of these organisms, and the frequency of such events is greater along fast-spreading axes. Discrete vent fields are typically composed of multiple chimneys and mineralized structures that emit waters as hot as 400 °C. These focused hot vents are often flanked with diffuse flows at much lower temperatures. A vent field can stretch for a few hundred meters along the ridge axis and persist for a few decades before the subterranean plumbing is clogged or the field is obliterated by a lava flow. Along a medium-rate spreading center like the Endeavour Segment of the Juan de Fuca Ridge (Fig. 2A), gaps between vent fields may be a few kilometers long, and gaps between adjacent segments may extend from tens to hundreds of kilometers (Fig. 2B). Lateral offsets like the Blanco Transform Fault displace contiguous ridge segments, disrupting along-axis water currents, and consequently impeding animal dispersal. Intervening seamounts, oceanic microplates, and inflated bathymetry can also displace ridge segments and disrupt currents that contribute to along-axis dispersal.

FIGURE 2 Hierarchical structure of vent fields in the northeastern

Pacific. (A) Portion of the Endeavour segment (modified from Thomson et al., 2003) has a series of vent fields portrayed as chimneys. (B) Contiguous segments along the Juan de Fuca Ridge are separated by transform faults from the Explorer Ridge and Gorda Ridge. These northeastern Pacific ridge systems are separated from the East Pacific Rise (Fig. 1) by subduction below the North American Plate.

abyssal waters and disperse about 100 km. Vent shrimp and crabs are mobile as adults, but actively swimming larvae and juveniles may feed and grow in the photic zone where they can potentially disperse great distances. Biological and geographical limits to dispersal are often reflected in patterns of gene flow and population subdivision of a species. For example, large ridge offsets such as the 450-km-long Blanco Transform Fault (Fig. 2) can restrict dispersal between the flanking ridge axes. This fault effectively isolates sister species of limpets Lepetodrilus fucensis and L. gordensis, which live respectively on the Juan de Fuca and Gorda ridge systems. Northern and southern populations of the tubeworm Ridgeia piscesae are not similarly isolated across this region, but gene flow is about six times greater in a southerly direction than in the reverse. Apparently, deep-ocean currents that flow in a southeasterly direction along Blanco Transform Fault force this unidirectional pattern of dispersal. Genetic studies of vent-endemic animals living along the East Pacific Rise and Gálapagos Rift also reveal physical oceanographic barriers to dispersal in some species, complete isolation of others, and no discernable effects on yet others. Various species with differing larval life histories (e.g., planktotrophy vs. lecithotrophy) and buoyancy will spend more or less time and rise to differing heights in the water column. These factors will expose larvae to different current patterns that might facilitate the retention of some species within the axial valleys and subject other species to crossaxis currents that sweep them away from the ridge.

HYDROTHERMAL VENTS

425

BIOGEOGRAPHIC REGIONS

Hydrothermal vent fauna can be subdivided into six biogeographic provinces associated with various ocean basins and discontinuities in the mid-ocean ridge system (Fig. 1). The northeastern and southeastern Pacific ridge systems (the Explorer, Juan de Fuca, and Gorda ridges versus the East Pacific Rise) were once connected, but subduction under the North American plate eliminated the interconnecting Farallon Ridge about 25 million years ago. Today, these northeastern and southeastern Pacific ridge systems support sister-species pairs of annelids and gastropods that diverged following this vicariant event. Closure of the Isthmus of Panama severed connections between eastern Pacific and Atlantic vent fauna (∼10 million years ago), and closure of the Tethys Sea severed connections between Atlantic and Indian Ocean vent fauna (∼60 million years ago). Elevation of the Azores Plateau (∼20 million years ago) may have separated northern and southern fauna along the Mid-Atlantic Ridge. The relevant timing of these events will likely vary among taxa, as various species differ in the altitudes they achieve when dispersing. Thus, some species, such as vent shrimp, may transcend these boundaries among these provinces, whereas others will not. Chemosynthetic fauna are also found at hydrocarbon seeps and at sites of organic deposition (e.g., fjords, submarine canyons, wood and whale falls). Some researchers hypothesize that these habitats may serve as stepping-stones for the modern-day dispersal of some animals between widely separated vent systems (C. Smith, this volume). EVOLUTION OF VENT FAUNA

Deep-sea hydrothermal vents were once thought to be refuges for ancient relics, unaffected by catastrophic events that led to global mass extinctions in shallow marine environments at the close of the Paleozoic and Mesozoic eras. Nearly 500 new species of vent-endemic animals have been described since the discovery of vents in 1977. Certain stalked barnacles appear to have occupied vent habitats since the early to middle Mesozoic and are considered living fossils. The great diversity of new genera, families, and putatively new phyla was hypothesized to be a consequence of the stability and longevity of these chemosynthetic environments. Nonetheless, molecular phylogenetic studies reveal that some of the common families of vent invertebrates, such as bresiliid shrimp and bathymodiolin mussels, are products of recent evolutionary radiations following the K-T mass extinction (less than 65 million years ago). Molecular and fossil evidence suggests that chemosynthetic clams and vestimentiferan tubeworms

426

HYDROTHERMAL VENTS

may be somewhat older, having radiated during the late Cretaceous (less than 90 million years ago). Some primitive gastropod families may be older yet (∼100 million years), but they are not exclusive to chemosynthetic environments. Ancient deposits from Silurian (∼400-millionyear-old) hydrothermal vents contain fossils of molluscan groups and brachiopods that are not found in modern vent communities. Thus, deep-sea vents, too, suffered from mass extinction events that revised animal diversity in the photic zone at the close of the Paleozoic. Although hydrothermal vents have existed since the early eras of our planet, some modern vent taxa may have first diversified in sulfidic cold seeps or in organic deposits, and then subsequently radiated at vents. Molecular evidence suggests that the oldest evolutionary lineages of vestimentiferan tubeworms diversified first in seeps. Vesicomyid clams also radiated first in seeps and then generated a few species capable of exploiting hydrothermal vents. In contrast, bresiliid shrimp and alvinellid polychaetes appear to have radiated recently in vent environments. It remains uncertain whether the most ancient lineages of bathymodiolin mussels first occupied cold seeps, wood falls, or hydrothermal seamounts. All these scenarios are equally probable based on molecular phylogenetic analyses, but the mussels have invaded progressively deeper habitats and have diversified in their use of sulphur- and methane-oxidizing endosymbionts. THE FUTURE FOR VENTS

Deep-sea hydrothermal vents are not eternally stable refuges, immune to forces affecting Earth’s surface. All the animals found at vents are anaerobes, making their livings by eating microbes or hosting them as symbionts. In either case, these primary producers and their hosts must live in a narrow redox zone that provides reduced volcanic gases on one side and oxygen on the other. Oxygen is a byproduct of photosynthesis at the planet’s surface, and it is delivered to ocean depths by circulation patterns driven by the heating of surface waters near the equator and cooling near the poles. Periods of intense global warming during the late Cretaceous (∼95 million years ago) and associated releases of greenhouse gases during the early Tertiary (∼55 million years ago) altered deep-ocean circulation and created anoxic basins that led to extinctions. Vent animals, as well, should be vulnerable to events that increase the boundaries of narrow redox zones that support them. Present-day accumulations of greenhouse gases and global warming, although small compared to catastrophic events during the late Cretaceous, may nevertheless threaten deep-ocean circulation regionally and thereby affect vents similarly.

Mining poses another threat to vent communities. The Papua New Guinea government has granted commercial leases to mine high-grade gold and copper deposits from extensive sulfide mounds at Manus Basin vents. The animals living on these mounds will surely be disturbed, but it not known whether such local disturbances will have greater impacts than the submarine volcanic events that regularly disrupt this region. Finally, scientific visits to vents with manned and robotic submersibles may carry hitchhiking animals, microbes, and potentially even diseases between vents. Thus, the same anthropogenic factors that affect surface islands worldwide (exploitation, habitat destruction, invasive species, and diseases) will also affect deep-sea hydrothermal events. Consequently, two nations have created deep-sea marine protected areas (MPAs). The Lucky Strike and Menez Gwen vent fields were designated the first deep-sea MPAs by the Azores government in June 2002. Canada designated the Endeavour vent field as an MPA in March 2003. Other vent fields that lie in national waters may eventually gain similar status, but the protection of vents that lie outside

of exclusive economic zones will require international cooperation. SEE ALSO THE FOLLOWING ARTICLES

Cold Seeps / Dispersal / Refugia / Vicariance / Whale Falls

FURTHER READING

Corliss, J. B., and R. D. Ballard. 1977. Oasis of life in the cold abyss. National Geographic Magazine 152: 441–453. Little, C. T. S., and R. C. Vrijenhoek. 2003. Are hydrothermal vent animals living fossils? Trends in Ecology and Evolution 18: 582–588. Thomson, R. E., S. F. Mihály, A. B. Rabinovich, R. E. McDuff, S. R. Veirs, and F. R. Stahr. 2003. Constrained circulation at Endeavour Ridge facilitates colonization by vent larvae. Nature 24: 545–549. Tunnicliffe, V., and M. R. Fowler. 1996. Influence of sea-floor spreading on the global hydrothermal vent fauna. Nature 379: 531–533. Tyler, P. A., and C. M. Young. 1999. Reproduction and dispersal at vents and cold seeps. Journal of the Marine Biological Association of the United Kingdom 79: 193–208. Van Dover, C. L. 2000. The ecology of deep-sea hydrothermal vents. Princeton, NJ: Princeton University Press. Van Dover, C. L., C. R. German, K. G. Speer, L. M. Parson, and R. C. Vrijenhoek. 2002. Evolution and biogeography of deep-sea vent and seep invertebrates. Science 295: 1253–1257.

HYDROTHERMAL VENTS

427

I ICELAND SIGURDUR STEINTHORSSON University of Iceland, Reykjavik

Iceland is the largest volcanic island (103,000 km2) in the Atlantic Ocean. It is one of the most volcanically active places on Earth, with more than 20 eruptions per century, and owing to its northerly location in the middle of the sea, the forces of erosion are very active as well. For these reasons, Iceland is a “natural laboratory” in which a continuous tug-of-war exists between constructive and destructive processes that can be studied in real time as well as in the geological record. TECTONICS Tectonic Setting

In terms of global tectonics, Iceland is a hotspot located near a constructive plate boundary. It is the largest mass of land found on the Mid-Atlantic Ridge, and the transverse ridge crossing Iceland from Greenland to Scotland is among the most substantial of aseismic ridges in the oceans (Fig. 1). A progression in age exists from the active central rift of Iceland to the Tertiary basalts of eastern Greenland (65 million years old) and Britain (45–50 million years old). The high-rising Iceland Plateau reflects the buoyancy of a more-than-400km-deep plume of ascending hot upper mantle presently centered below southeastern Iceland, and the transverse ridge, with over 20-km thickness of basaltic crust, is the passive trail of the hotspot, which has been active since before the opening of the North Atlantic. The V-shaped ridges southwest of Iceland find their explanation in magma pulses travelling ~20 cm/year southward along the ridge.

428

FIGURE 1 Satellite gravity map of the North Atlantic region reflects

the submarine topography (Smith and Sandwell, 1995). The black line is the active plate boundary, and the red dot is the assumed center of the Iceland mantle plume. Vectors show the plate movements relative to the spreading center, 1 cm/a in each direction, striking 104°. (Modified by I. Þ. Bjarnason.)

Tectonic Evolution

The Mid-Atlantic Ridge crosses Iceland from southwest to northeast, at which latitude the rate and direction of spreading is 1.95 cm/year and N 104° E, respectively. Relative to the adjacent oceanic Reykjanes and Kolbeinsey Ridges, the plate boundary in Iceland is shifted some 100 km to the east by a set of transform faults: the Tjörnes Fracture Zone (TFZ) in the north and the South Iceland Seismic Zone (SISZ) in the south (Fig. 2). During the geological history of Iceland, the present configuration of volcanic zones has come about in a series of ridge jumps and rift propagations, caused by the gradual west-northwest drift

volcanic outlier in the North American plate, representing a dying remnant of an earlier configuration. Prior to the relocation of the rift system 6 million years ago, it corresponded to the Mid-Iceland volcanic zone. Conversely, in southern Iceland, the rift zone is propagating southwestward into a 10-million-year-old plate; the Surtsey eruption of 1963–1967 represents the southernmost site of activity yet. Finally, the Öræfajökull zone may represent the opening up of old crust above the mantle plume.

16 Ma

6 Ma

Kinematics and Crustal Structure

3 Ma

TFZ

SNA

SISZ Z

Alk Trans Thol

ORAE

SVZ PRESENT FIGURE 2 Summary of tectonic evolution with reference to the Ice-

land mantle plume (red circle). About 16 million years ago, the spreading center jumped from west of present Iceland toward the east. The rift zone as shown in the top diagram was active until about 6 million years ago when it again jumped east (center). The Southern Volcanic Zone (SVZ) has been propagating southwestward for the past 3 million years. Present configuration (bottom): SNA: Snæfellsnes; ORA: Öræfajökull; SISZ: South Iceland Seismic Zone; TFZ: Tjörnes Fracture Zone. The different colors denote petrological affinity.

of the North Atlantic plate system relative to the Iceland mantle plume. The three flank zones termed, respectively, the Snæfellsnes, Southern, and Öræfajökull volcanic zones, are all distinguished by volcanics of alkalic tendency resting unconformably upon much older basement, as opposed to the tholeiitic rocks of the rift zones. Each of the three has its own distinct tectonic origin. The Snæfellsnes Zone is a

Present-day crustal movements in Iceland are monitored by a range of measurements, including seismics, satellite radar interferometry (SAR), measurements of ground tilt, and measurements of strain in boreholes. A network of seismometers has delineated the plate boundary, relative movement on active faults, and the fact that spreading is sporadic (episodic) along the plate boundary. Tilt measurements (and, more recently, SAR) show that the Icelandic crust is extremely labile, responding almost instantaneously to changes in glacial loading. Recently, the monitoring effort has shown some success in shortterm prediction of earthquakes and volcanic eruptions. Seismics indicate that the basaltic crust in eastern Iceland is 27–33 km thick, and in western Iceland 20–26 km. The upper crust, down to 6–9 km, is characterized by rapid linear increase in P-wave velocity with depth, whereas in the lower crust the velocity changes little with depth. The reason for the difference in thickness may be the west-northwest drift of the plate system, such that the eastern plate is all but stationary above the mantle plume whereas the western plate moves almost 2 cm/year relative to it. Clearly, and in view of the structure of ophiolites, the uppermost crust is composed mostly of subaerial lava flows; dike density increases with depth, and at a certain level gabbros dominate. Perhaps the most exceptional feature of Iceland’s geology, as compared with “classical geology,” is embodied in a kinematic model of crustal accretion (Fig. 3). In contrast with the customary idea of horizontal layers accumulating one upon the other—in the fashion of a layer cake— the accumulation of Iceland’s basalt succession takes place more in the fashion of a conveyor belt: The source is almost exclusively in the rift zone, from where the formations drift away as new rocks are formed. Both the tilt of the succession toward the rift zone and the seismic layering of the crust are formed in the rift zone as an integral part of the accreting process. Regional synclines in Iceland, therefore, dip toward the rift zone from which they originated, whereas regional anticlines result from rift jumps.

ICELAND

429

FIGURE 3 Kinematics of crustal accretion. (A) If the 43,000 km3 of

A

volcanics that erupt in Iceland in 1 million years simply piled up along the rift zone, a 300-km-long ridge would result, 20 km wide at the base and 13.6 km high—an unrealistic scenario because the surface remains at constant elevation while the ground sags below new formations. (B) Schematic block diagram of northeastern Iceland showing volcanics formed in the last 2 million years (dark gray) and in the 2 million years prior to that (lighter gray). Blue curves are isochrones showing the surface 2 million years ago, 4 million years ago, and so forth. Black

B

lines are dikes. (C) Kinematic model of the upper crust, quantifying the section in (B): Isochrones (surface at 2 million years ago, four million years ago, etc.), blue; calculated isotherms, red; material trajectories (describing movement of rock erupted in the rift zone), black. Following one such trajectory (arrow): after sinking and heating up for 2 million years, the rock is 400 °C. A maximum of 600 °C is reached in 3 million years, at which time the rock starts cooling as it drifts out of the rift zone (400 °C at 6 million years). (D) As the rocks heat up, largely irreversible recrystallization (progressive metamorphism) takes place. Upon cooling, the mineral assemblage of the maximum temperature is retained, resulting in horizontal metamorphic layering. Rocks forming the upper crust (seismic layers 1, 2, and 3) are volcanic, and the gab-

C

°

° ° °

°

°

D

bros of the lower crust plutonic.

boundary, is frequently the site of a volcanic center, characterized by evolved rock compositions in addition to basalts and by a high-temperature geothermal system. The volcanic center together with the fissure swarm transecting it form a comagmatic entity termed a volcanic system. The relationship between the plate boundary and volcanic systems has, in southwestern Iceland, been shown seismically: Earthquakes defining the plate boundary below about a 3-km depth cluster along a vertical plane striking N 80° E, whereas shallow earthquakes originate along the strike of the fissure swarms, at about N 40° E. A similar relationship exists for the northeastern branch of the rift. STRATIGRAPHY AND GEOLOGICAL HISTORY Overview

The model also explains the decrease in dip and lava thickness with stratigraphic height, the quasi-horizontal zoning of amygdules, and the seismic layering in the upper crust. Volcanic Systems

Crustal spreading and volcanism in the axial rift zone takes place on discrete fissure swarms straddling the plate boundary in an en-echelon array (Fig. 4). The swarms may be 10–15 km wide and up to 100 km long. They are characterized by extensional tectonic features such as open fissures, graben structures, and crater rows at the surface, and dikes and normal faults at deeper levels. The central, most active part of each system, overlying the plate

430

ICELAND

The predominantly volcanic rocks of Iceland, which range back about 16 million years in age, are conventionally divided into four stratigraphic formations or series. This division is based on climatic evidence from interlava sediment or volcanic breccias and on paleomagnetic reversal patterns supported by radiometric age datings. The categories are as follows: Postglacial (Holocene) lavas and detrital rocks of the last 9000–13,000 years. This formation (25,000 km2) occupies, in addition to the regolith, the presently active volcanic areas of Iceland including the median rift zone—the subaerial Mid-Atlantic ridge— traversing the country from southwest to northeast. Upper Pleistocene “palagonite formation,” dating back 0.7 million years, corresponding to the present

normal geomagnetic epoch, Brunhes. This formation coincides spatially more or less with the postglacial formations. Plio-Pleistocene “gray basalt formation,” dating back 0.7–3.1 million years ago and including the Matuyama epoch and the Gauss epoch down to the Mammoth event. These rocks occupy a broad belt (25,000 km2) on either side of the median rift zone. Tertiary “plateau basalts,” older than 3.1 million years. This series covers about half of Iceland (50,000 km2) in the east, west, and north.

°

°

°

°

°

Tertiary formation

The Tertiary plateau basalts, well exposed due to glacial erosion, are made up of regular sequences of flat-lying subaerial lava flows, 5–15 m thick and separated by minor clastic interbeds of volcanic origin. The sequences dip toward the respective rift zone in which they formed; where rift jumps have occurred, the resulting unconformity lies at the crest of an anticline where the beds dip toward the old and the new rift, respectively. The lava pile is composed of elongated lenses, each representing the products of a volcanic system. Along each lens stretches a dike swarm, in the core of which a volcanic center may have developed, characterized by evolved rocks, intrusions of gabbro and granophyre, and hydrothermal alteration. Mapping the apparently monotonous Tertiary flood basalts calls for special techniques for correlation and age determinations. In the last half century, several “mapping campaigns” employing these new methods have been launched in all Tertiary regions in Iceland. The longest continuous succession studied so far is 8.5 km thick, spanning some 10 million years and encompassing 700 lava flows. The build-up rate varied between 360 and 2600 m per million years, whereas, on average, the eruption rate was one flow per 14,000 years—the common range in the Icelandic Tertiary is 6000–10,000 years between adjacent flows. This means that only exceptionally large lava flows, having spread far out of the volcanic zone, are represented in the exposed sections. In the flood basalts, three lithologic types can be distinguished in the field for stratigraphic mapping: olivine tholeiite, tholeiite, and feldspar-porphyritic basalt. Apart from the phenocryst content, different appearance of weathered surfaces and diagnostic mineral assemblages in amygdules enable the distinction between rock types. Regional-scale paleomagnetic mapping of lava groups is standard in stratigraphic correlations of basalt successions. The first radiometric datings (1968) of Tertiary samples

FIGURE 4 Simplified geological map of Iceland showing bedrock and

glaciers. (1) ice sheet; (2) Upper Pleistocene and Holocene; (3) PlioPleistocene; (4) Tertiary; (5) volcanic system (yellow) with a volcanic center and a caldera.

from Iceland showed the oldest exposed rocks to be about 16 million years old, in sharp contrast to the ~60-millionyear-old age assumed on the basis of Iceland being part of an old east-west igneous province extending from eastern Greenland to Britain. Since then, several hundred such age determinations exist of Tertiary basalt successions as well as of silicic volcanic centers within the lava pile. Sedimentary horizons of terrestrial origin, lacustrine or fluvial, occur interspersed with lava flows in the Tertiary basalt succession. These include beds of lignite, which yield a more or less continuous paleobotanical record ranging back 16 million years. By comparison with modern analogues, the mean annual temperature in the North Atlantic region was about 5 °C warmer than now (4–6 °C) some 15 million years ago. A maximum was reached around 12 million years ago (13.5 ± 1 °C, similar to the present southeastern United States) with gradual cooling to 7.4 ± 2 °C at 6 million years ago. The first signs of glaciation appear at about 3.1 million years ago. Plio-Pleistocene Formation

The boundary between the Tertiary and Plio-Pleistocene is somewhat arbitrarily fixed at the base of the Mammoth geomagnetic event, 3.1 million years ago. A continuously cooling climate in the upper Tertiary led to the appearance of the first tillite horizons interstratified with basalt at about this time. The sedimentary interbeds become thicker and coarser than before; subglacial pillow lavas and hyaloclastites interchanged with thick subaerial lava flows. At least ten glacial/interglacial cycles are known

ICELAND

431

from the Upper Pliocene, as well as at least that many from the Pleistocene. The most complete single section from this period is found on the Tjörnes Peninsula in northern Iceland, where the Plio-Pleistocene sequence is 600 m thick, 400 m of which are shellbearing sediment of mainly marine and estuarine facies, but in the upper half, also of glacial origin, at least six tillite horizons have been identified. With the onset of glaciation, the topography of Iceland started undergoing radical change from the flat-lying Tertiary lava plain: Elongated ridges and isolated mounds of hyaloclastite piled up in subglacial eruptions, whereas the glaciers sculpted out the fjord landscape now typifying the Tertiary. Upper Pleistocene Formation

This series encompasses the Brunhes geomagnetic epoch—the last 700,000 years, up to the Holocene. In space it coincides essentially with the neovolcanic zones where its volcanics form two distinct facies: subglacial and subaerial. Within volcanic zones, accumulation proceeded at a much higher rate than did denudation, but in some places groups of lava flows can be seen alternating with tillite beds and hyaloclastite rocks giving information about several interglacial/glacial cycles. Outside the volcanic zones, glacial erosion became very effective. In western Iceland, 800–1000-m-deep valleys were carved out in less than 1.8 million years, and on the Snæfellsnes Peninsula this took place in less than 1 million years. SUBGLACIAL VOLCANICS

The topography most typical for the Upper Pleistocene are ridges and mounds of hyaloclastite, often with a core of pillow basalts. These are formed in subglacial eruptions, the ridges on long volcanic fissures, the mounds on short fissures or circular craters. Two eruptions of this kind have been followed closely by scientists in Iceland: the submarine Surtsey eruption of 1963–1967 and the subglacial 1996 Gjalp eruption. With decreasing water pressure as the crater builds up below water or ice, pillow lava forms first, followed by tuff, and finally by lava flows as the crater becomes subaerial. By no means will all volcanoes show all three facies: pillow lava at the base and/or subaerial lava on top are often missing, signifying, respectively, that the overlying glacier was too thin or that the eruption failed to melt the overlying ice. At Surtsey it emerged that palagonitiztion—the lithification process transforming the pile of hyaloclastite tuff into rock—takes

432

ICELAND

place between 80 and 150 °C and is completed shortly after the eruption. Many examples are known of intermediate or silicic volcanics having erupted beneath the Pleistocene ice sheets. The silicic ones are characteristically steep-sided domes consisting of distinctive rock varieties. Among these are lobes of glassy rhyolite encrusted with a thick layer of pitchstone resembling huge pillows. The lobes are embedded in a granulated glass matrix evidently derived from the crust of the lobes themselves. INTERGLACIAL VOLCANICS

During the interglacials, lava flows and tephra issued from the volcanoes. The lavas tend to have a different aspect from those of the Tertiary, being gray as opposed to black in color and often doleritic. The slaggy surfaces have been scraped off by glaciers, exposing the coarsergrained interiors. Holocene

Evidence of the Pleistocene-Holocene transition is seen in raised beaches (up to 100 m, but generally 40–60 m), erosional platforms at –30 m, moraines signifying oscillatory retreat of the glaciers, sediments in lakes yielding a continuous record back to 13,000 years ago, and profuse and singular volcanism. The final glacial advance (Younger Dryas) at 10,000 years ago followed an interstadial (Alleröd) at 11,000. The rapid eustatic rise of sea level caused the lowlands to be inundated, but because of the low viscosity of the underlying mantle (1018 poise), the land rose isostatically in the span of 500–1000 years, with the formation of erosional platforms at about 30 m below present sea level. The isostatic recovery caused rapid release of pressure in the upper mantle, resulting in very intense volcanism, probably more than 30 times that of present day. Two types of volcanoes appear to be confined to this time (~13,000 to 6000 years ago): small picritic lava shields and large (up to 50 km3) shields of olivine tholeiite. The table mountains (tuyas) are probably subglacial equivalents of the latter, formed near the end of the glaciation. Postglacial formations comprise lava flows and pyroclastics, loess, unconsolidated marine clays, fluvioglacial and fluvial outwash and soil formed after the deglaciation of the land area. VOLCANICS

Postglacial volcanism has been confined to the same areas as that of the Upper Pleistocene, with 30 active volcanic systems in the rift zones and the three off-rift zones. Total

production in the Holocene is estimated at 400–500 km3, with postglacial lavas covering some 12,000 km2. Basaltic lavas predominate (~90%), but the high proportion of intermediate and silicic rocks is unusual for the oceanic rift system—the evolved rocks are confined to volcanic centers. Almost all types of volcanoes known on Earth are found in Iceland (see Volcanoes, this volume). REGOLITH

The regolith in Iceland is mainly the till of the Weichselian glacial, in addition to fluvioglacial outwash. The latter covers about 5000 km2, and much of the southern coast is sandur plains, up to 200 m thick, formed largely by glacial floods (Icelandic: jökulhlaup): sediment-charged meltwater from subglacial eruptions. The “organic soil” is mostly peat, rich in mineral matter (often 40–60%) because of frequent tephra falls and deposition of eolian material. Tephrochronology Numerous eruptions have left tephra layers in the soil, some of which cover most of Iceland (and are found on the sea floor, in northwestern Europe, and in the Greenland ice sheet). These layers, particularly the silicic (light-colored) ones, make useful marker horizons in the soil or sediment, as well as being records of the eruption history of the various volcanoes. Dr. S. Thorarinsson, founder of this branch of geology, worked out the eruption history of Hekla and some other volcanoes, and with improved analytical techniques, most basaltic (black) layers can now be relegated to their respective volcanic system of origin. Climatic Variations Climatic variations in the Holocene can be traced from biota (including pollen and diatoms) in bogs and lake deposits, and from types of sediment. This work is greatly facilitated by tephrochronological marker horizons. In the early Holocene, 10,000–9000 years ago, Betula (birch) appeared in northern Iceland, from which it spread rapidly to the south 9000 years ago, possibly indicating that birch survived the glaciation in some nunataks. This first Betula maximum in pollen diagrams corresponds to the Boreal and Lower Atlantic in continental Europe, with annual temperatures some 2 °C higher than present and with lower precipitation. A Betula minimum, corresponding to the wet (but still warm) Atlantic, started at about 6500 years ago with bogs replacing the birch forests. During the second Betula maximum at 5000–2500 years ago (sub-Boreal), brushes covered half the country, with temperatures 2–3 °C higher than now, low precipitation, and mild winters. At about 2500 years ago, the climate deteriorated, and the birch gave way to bogs. A third Betula maximum, correspond-

ing to the Neo-Atlantic, beginning about 1650 years ago, was interupted by the settlement of Iceland around AD 870, followed by the “Little Ice Age” between AD 1400 and 1900. PETROLOGY AND CHEMISTRY

Igneous rocks are classified on basis of chemistry (basalticintermediate-silicic) and crystallinity (rate of cooling: e.g., glassy hyaloclastite–microcrystalline basaltic lava–coarsegrained gabbro). Hyaloclastites are particularly common in Iceland as a result of subglacial volcanism. Plutonic rocks, gabbro, and granophyre are found in eroded central volcanoes and as xenoliths in volcanic rocks. Increasing alteration and metamorphism with depth is seen in the amygdule-zoning in eastern Iceland and elsewhere (Figs. 3–4), and similar sequences obtain in drill holes into geothermal systems and around intrusions in central volcanoes. The proportion of evolved rocks in the Tertiary, predominantly rhyolites and dacites (9%), is surprisingly high in Iceland considering its tectonic setting at a spreading center. In this respect Iceland is entirely different from, for instance, the Hawaiian islands, where such rocks hardly occur at all. Intermediate rocks are less common (3%; basalt 80–85%) than silicic, and a Daly gap is frequently present in individual volcanoes. In a typical Tertiary section, volcanogenic sediments amount to some 5–10%; in the Quaternary areas, however, they form a much greater part of the succession owing to the influence of glaciation on both volcanism and denudation. The silicic rocks are confined to volcanic centers, of which about 65 are known in the Tertiary formation and around 30 in the Quaternary and Holocene areas. Investigations have shown that (1) the rift zones give rise to tholeiitic basalts and their derivatives; (2) the off-rift Snæfellsnes, Southern, and Öræfajökull volcanic zones produce alkaline to transitional-alkaline rocks; (3) there is a systematic variation along the rift zone and the adjacent oceanic ridges in various petrochemical properties, including the range of basalt compositions; (4) the volcanic productivity varies systematically along the rift zone, with a maximum in central Iceland (Fig. 5); and (5) each volcanic system shows a range in compositions, from “primitive oceanic tholeiite” to quartz tholeiite, and even to silicic and minor intermediate rocks. Isotope geochemistry indicates that anatexis of hydrated basalt plays an important role in magmatic evolution in rift-zone volcanoes, whereas in the off-rift areas, crystal fractionation is the dominant process. The compositional variation along the rift zones has by different authors been ascribed to heterogeneities in

ICELAND

433

°

A

°

B

C

10

Na2O + K2O wt.%

fajo

Tor

8 y

rg

ae

6

eim

H

e etb

l

kul

muli

Thing

S

la Hek

4 fla Kra

2 0 45

50

55

60 65 SiO2 wt.%

70

75

FIGURE 5 (A) Distribution of the estimated 420 km3 of volcanic rocks

erupted in Iceland in the Holocene. Dark blue, tholeiites; light blue, alkalic and transitional basalts; yellow, evolved rocks. (B) K2O-concentration in tholeiitic basalts along the rift zone of the Iceland region (red trace in inset). Note the increase in compositional range toward central Iceland. Yellow arrows are volcanic centers with more evolved rocks. (C) TAS diagram (total alkalis vs. silica) for six volcanic centers in Iceland. Blue, tholeiitic; green, transitional; red, alkalic. Dashed black is the “Hawaii Division Line” separating alkalic and tholeiitic compositions. (Redrawn based on S. P. Jakobsson, in Saemundsson 1979).

the upper-mantle source, to crystal fractionation, or to anatexis and magma mixing in the crust. It now appears that all processes are at work. VOLCANOES

Three main types of volcanoes have been active in the Holocene: lava shields (shield volcanoes), crater rows, and central volcanoes. Lava shields are typical for the rift-zone volcanism in the early Holocene (before 6000 years ago). Their age, and their similarity to the subglacial table mountains both with regard to petrology and size, suggests that both are related to the isostatic rebound at the close of the Weichselian glaciation about 10,000 years ago. Similar structures exist from earlier glacials and interglacials. The Holocene shields, about 30 in number, are large and small; the apparent volume of the largest ones, Skjaldbreidur and Trölladyngja, is 15–20 km3, but

434

ICELAND

gravity measurements indicate hidden lava flows at depth, increasing their volumes up to 40 km3. A volcanic center, together with the fissure swarm transecting it, constitutes a volcanic system. Some 24 such volcanic systems have been active in the Holocene, 18 since Iceland’s settlement around AD 870. During that time, between 30 and 40 sites have erupted, and during the last 300 years an eruption has started on the average every fifth year. Based on geological mapping of extinct volcanic centers, it is estimated that their lifespan may be 0.3 to more than 1 million years. The various stages of their evolution can be seen in Iceland, from the “primitive” fissure swarms on the westernmost Reykjanes Peninsula to “mature” centers like Krafla or Askja. The relationship, shown in Fig. 4, indicates that magma enters the upper crust at the intersections between the fissure swarms and the plate boundary. At the depth where isostatic equilibrium is reached, magma pools to gradually form a magma chamber. During spreading episodes, magma empties from the magma chamber into the adjoining fissure swarm, as seen in the Krafla Fires of 1974–1985. As the rift-zone volcanic centers develop, silicic magma starts forming through the remelting of hydrated basalt. The magma chamber, as it evolves chemically, rises in the crust, and finally the roof caves in, forming a caldera. When the crust is relatively thin, as in the rift zones, it is unable to support lofty edifices, and consequently the volcanic centers never take the form of “classical” central volcanoes, unlike, for example, Snæfellsjökull (1446 m) and Öræfajökull (2111 m), which stand on thick off-rift crust. The largest eruptions in historical time were Eldgja in AD 934 (30-km-long fissure) and Laki in 1783 (25-kmlong crater row), each producing 15–20 km3 of basalt. The Laki Fires released huge amounts of sulfur into the atmosphere, affecting climate in the Northern Hemisphere for two years and causing a famine in Iceland that killed over a fifth of the population. Both fissures are fed by a volcanic center, Eldgja by the subglacial Katla volcano and Laki by the subglacial Grimsvötn volcano, the most active in Iceland. Subglacial volcanoes that have given rise to jökulhlaups (debris-laden meltwater floods) in historical time are Öræfajökull in 1362 and 1727, Grimsvötn every decade or so, Katla about 20 times (most recently in 1918), and Eyjafjallajökull in 1821. GLACIERS

Glacier ice (Icelandic: jökull) covers about 11% of Iceland. Almost all forms of glaciers are represented, from cirque glaciers to extensive plateau ice caps. By far the largest is

the Vatnajökull ice sheet (8300 km2), the world’s largest ice mass after Antarctica and Greenland. Other ice sheets larger than 500 km2 are Myrdalsjökull (596 km2), Langjökull (953 km2), and Hofsjökull (925 km2), but many high mountains are capped with small glaciers. In the rugged northwest, and especially at the Tröllaskagi Peninsula in north-central Iceland where many mountains reach 1300–1500 m in elevation, there are a great number of “alpine” cirque and valley glaciers. All glaciers in Iceland are of the temperate type (ice in thermal equilibrium with water) and are therefore highly responsive to climatic fluctuations. In the first centuries of settlement in Iceland (870 to the thirteenth century), during the last part of the Medieval Climatic Optimum, the glaciers were much smaller than now. After that, with a cooling climate that culminated in the Little Ice Age, the glaciers grew to their maximum size in postglacial time around 1890. Since then, they have been receding at an ever-increasing rate, and some of the smaller ones have already disappeared. Glaciers in Iceland are very dynamically active. The most active glacier outlets flow southward from the high plateaus of Myrdalsjökull and Vatnajökull. Only one quarter of the accumulation (annual precipitation exceeds 4000 mm) is melted within the accumulation areas, so enormous amounts of ice are transported by the outlets down to the ablation areas. Surface velocities of these outlet glaciers commonly exceed 1 m/day. Most of the broadlobed outlets of Vatnajökull are liable to periodic surges (catastrophic advances) with return periods of several decades. The surge in Brúarjökull of 1963–1964 involved 40% of the area of Vatnajökull, with the ice front advancing 8 km at velocities up to 4–5 m/hour. Surges are also typical for outlets of Hofsjökull and Langjökull. GEOTHERMICS

At present, over 50% of Iceland’s energy consumption is geothermal, mostly in the form of space heating (homes and greenhouses), but also for industry and the production of electricity (26%, the rest hydroelectric). The regional heat flow in Iceland falls within the heat flow anomaly of the Mid-Atlantic Ridge crest and varies between about 80 and 300 mW/m2. Low- and hightemperature thermal areas are distinguished empirically by the geothermal gradient in the uppermost 1 km of the crust. The low-temperature areas are characterized by a gradient of less than 150 °C/km, by a relatively low degree of thermal metamorphism, and by hot springs and geysers. The water is exclusively percolating groundwater, usually mildly alkaline. Hot springs are found in more than 300 localities spread all over the country, although

there are very few in the east and southeast. The largest spring, Deildartunguhver in western Iceland, has a discharge of 180 L/s of 100 °C water. The 22 known high-temperature areas are part of active volcanic systems in the rift zones. Of these, five have so far been harnessed for energy production. The geothermal gradient in the uppermost 1 km exceeds 200 °C/km. Steam vents in the high-temperature areas discharge carbon dioxide, hydrogen sulfide, and hydrogen. Their fluid is derived almost exclusively from percolating groundwater, whereas the heat and the volcanic volatiles are probably derived from cooling intrusions at relatively shallow depths. The mineral content of the high-temperature water depends systematically on the temperature. Large deposits of silica sinter have formed in two high-temperature areas, including that of the Great Geysir in southern Iceland. On the Reykjanes Peninsula, some of the geothermal systems are saline due to infiltration of seawater. The well-known Blue Lagoon is formed by effluent water from the Svartsengi geothermal power station. EARTHQUAKES

Earthquakes in Iceland are primarily caused by (1) tectonic movements at the plate boundary and (2) volcanic activity. Most large earthquakes (M > 6) occur within two transform zones, the SISZ in the south and the TFZ in the north. The SISZ is marked by a 10–15-km-wide, easterly trending epicentral belt. The large earthquakes occur by faulting on north-south striking right-lateral faults. The left-lateral transform motion along the zone thus appears to be taken up by slip on numerous parallel faults and counter-clockwise rotation of the blocks between them (bookshelf tectonics). Within the Tjörnes Fracture Zone, at least three parallel, northwest-trending seismic belts have been identified. The seismicity of the volcanic zones is characterized by spatial clustering of epicenters. Most clusters coincide with central volcanoes. Rifting structures such as fissure swarms and normal faults are mostly aseismic except during episodes of rifting and magmatism. Several volcanoes exhibit persistent, low-magnitude seismicity. In the Hengill volcano in southwestern Iceland, seismicity is interpreted as the result of extensional failure and heat extraction from a cooling magma chamber. RATES OF GEOLOGICAL PROCESSES

Iceland’s volume above sea level is about 51,500 km3 (mean altitude 500 m, area 103,000 km2). The volcanic productivity in the Holocene has been estimated at 420 km3/10,000 years, which equals 42,000 km3/million years. At that rate, the whole volume of Iceland, 16 million years old, would

ICELAND

435

be produced in 1.2 million years. This means that powerful “sinks” are at work counterbalancing the high productivity: (1) thermal contraction in the cooling crust as it drifts away from the rift zone, bringing rocks older than ~16 million years old below sea level; (2) isostatic sagging in the rift zones beneath younger volcanics (Fig. 3); and (3) erosion by glaciers, water, and wind. Measured sagging of the surface in the rift zones, such as the Thingvellir graben, is 0.4–1 cm/ year (4–10 km/million years), whereas the westward tilt of the Tertiary succession in eastern Iceland indicates an average rate of sinking in the center of the rift zone amounting to 2.7 km/million years. The Pleistocene glaciers carved the fjord topography in the Tertiary lava pile, the weight loss being partly counterbalanced by isostatic uplift. The glacial rivers carry about 0.025 km3/year (25,000 km3/million years) of sediment out to sea, where it is temporarily deposited on the insular shelf before continuing down to the abyssal plain. SEE ALSO THE FOLLOWING ARTICLES

Atlantic Region / Earthquakes / Eruptions: Laki and Tambora / Faroe Islands / Surtsey / Volcanic Islands FURTHER READING

Jacoby, W. R., and M. T. Gudmundsson, eds. 2007. Hotspot Iceland. Journal of Geodynamics, Special Issue, 43. Saemundsson, K., ed. 1979. Geology of Iceland. Jökull, Special Issue, 29. Sigmundsson, F., L. A. Simonarson, O. Sigmarsson, and O. Ingolfsson, eds. 2008. The dynamic geology of Iceland. Jö Kull, Special Issue, 58. Steinthorsson, S., and S. Thorarinsson. 1997. Iceland, in Encyclopedia of European and Asian regional geology. E. M. Moores and R. W. Fairbridge, eds. New York: Chapman and Hall. Thordarson, Th., and A. Hoskuldsson. 2002. Iceland. Classic geology in Europe 3. Hertfordshire, UK. Terra Publishing.

INBREEDING LEONARD NUNNEY University of California, Riverside

Inbreeding is a process central to understanding the genetics of populations. Although usually thought of as the mating of close relatives, more generally inbreeding is the local build-up of genetic similarity due to common ancestry. It can cause genetic differentiation among island populations of the same species and the loss of genetic variability within populations. It can also reduce population survival over the short term by inbreeding depression

436

INBREEDING

and over the longer term by compromising the ability of the population to adapt to environmental change. WRIGHT’S THEORY

The theory of inbreeding was developed by Sewell Wright between about 1920 and 1950. Inbreeding is a simple process, but the genetic consequences are not, and Wright’s complex theory can be difficult to understand. A good starting point for understanding inbreeding is “identity by descent.” Two gene copies are identical by descent (IBD) if they both originated from the same ancestral gene copy. Using this idea, the inbreeding coefficient (F ) of a diploid individual is the probability that its two copies (one maternal and one paternal) of any gene are IBD. For example, an offspring of full siblings has a 25% chance of carrying two copies of a gene that are IBD because they were originally present as a single copy in one of its two grandparents. However, if we go back far enough in time (a period called the coalescence time), all gene copies currently segregating in any isolated population (or species) are IBD. This apparent paradox is resolved by recognizing that inbreeding is a relative measure: inbreeding at one level must be interpreted relative to a higher level. The relativity is made explicit in the notation of Wright’s hierarchical inbreeding coefficients (e.g., FST refers to the IBD of gene copies in subpopulations (S ) relative to the total (T ) population). To illustrate this point, consider a simple spatial hierarchy of an archipelago (T ) of many islands (S ), each supporting a population of many individuals (I ) of a plant species, and that inter-island movement (via seed and/or pollen) is infrequent. To make the example more interesting, it is assumed that the plants occasionally self-pollinate. Thus, the two gene copies from a randomly chosen individual are, on average, inbred relative to randomly chosen gene copies within the island (FIS > 0), a result driven by the presence of the inbred selfed individuals. Similarly, gene copies randomly chosen from any one island are inbred relative to gene copies chosen randomly from the whole archipelago (FST > 0). This island-level inbreeding arises because, on average, gene copies from the same island share more recent common ancestry (i.e., a shorter coalescence time) than those from the larger spatial unit, the archipelago. GENETIC CHANGES UNDER INBREEDING

A common misconception is that inbred populations generally have a deficit of heterozygotes relative to Hardy-Weinberg (H-W) expectations. In fact, an island population, inbred because of its isolation relative to other populations of the species, is expected to exhibit H-W ratios.

Any relative decrease in heterozygotes compared to H-W ratios would be due to processes internal to the population such as selfing or sibling mating. However, inbreeding does decrease the absolute level of heterozygosity in a population, because genetic variability decreases as a result of the random loss of alleles by genetic drift. This loss also causes initially identical populations to become genetically different at a rate proportional to 1/Ne, where Ne is the effective size of each population. As a rough guide, significant genetic divergence takes about Ne generations. For neutral alleles, this random loss affects large and small populations equally (but at different rates); however, as Ne decreases, more genetic variation becomes effectively neutral. The result is that in small populations, deleterious alleles can increase in frequency through random sampling, lowering the average fitness of the population and thereby inducing population-wide inbreeding depression. Thus, in the extreme case of the establishment of a population by few individuals, the founder effect typically includes a dramatic increase in the frequency of some previously rare deleterious alleles (e.g., the high frequency of porphyria variegata in the Afrikaner population of South Africa). There can also be a positive effect on fitness of such a founding event. Although a few highly deleterious alleles can, by chance, increase in frequency, many more disadvantageous alleles are lost through the same sampling process. Thus, provided that the population can survive the initial founding and grow, selection can reduce the frequency of the remaining deleterious alleles. The net result is that, for a significant period of time, the population can have less of a fitness loss due to disadvantageous alleles than would the parent population (although eventually, new mutations would restore the typical balance). This is one form of a process called purging. Another form of purging is restricted to managed populations and involves mating close relatives within a population to create inbred individuals (i.e., creating a population with more homozygotes that expected under Hardy-Weinberg ratios), allowing selection to more effectively reduce the frequency of deleterious recessives. INBREEDING DEPRESSION AND EXTINCTION RISK

Inbred populations typically have elevated extinction rates, over the short term because of the negative effects of inbreeding depression and over the long term because the loss of genetic variability resulting from inbreeding reduces adaptive potential, thus increasing susceptibility to environmental change. Island populations are particularly vulnerable to inbreeding and its negative fitness consequences for two

reasons. First, they are very often established by a few immigrants, resulting in a strong founder effect. Second, even after initial growth, island populations tend to be small and isolated, such that additional inbreeding can result in a further loss of genetic variation. These potentially negative effects on fitness may be somewhat offset by the purging of deleterious alleles noted above. Island populations are especially vulnerable to biotic change such as the arrival of a novel competitor, predator, or pathogen. Examples of the extinction of island endemics following the invasion of such species are well documented; however, when native populations persist, genetic management to reduce inbreeding and promote adaptation to the changing environment can be beneficial. For example, where island populations have become fragmented through habitat loss, inbreeding within the fragments can be minimized by the exchange of migrants among the fragments. On a larger scale, if island populations are not too different, inter-island exchange may be a viable strategy for increasing genetic variability. SEE ALSO THE FOLLOWING ARTICLES

Extinction / Founder Effects / Invasion Biology / Population Genetics, Island Models in FURTHER READING

Frankham, R. 1998. Inbreeding and extinction: island populations. Conservation Biology 12: 665–675. Frankham, R. 2005. Genetics and extinction. Biological Conservation 126: 131–140. Hartl, D. L., and A. G. Clark. 2007. Inbreeding, population subdivision and migration, in Principles of population genetics. Sunderland, MA: Sinauer Associates, Chapter 6. Jamieson, I. G. 2007. Has the debate over genetics and extinction of island endemics truly been resolved? Animal Conservation 10: 139–144. Nunney, L. 2001. Population structure, in Evolutionary ecology: concepts and case studies. C.R. Fox, D. Roff, and D. Fairbairn, eds. Oxford: Oxford University Press, 70–83. Willi, Y., J. Van Buskirk, and A. A. Hoffman. 2006. Limits to the adaptive potential of small populations. Annual Review of Ecology, Evolution, and Systematics 37: 433–458.

INDIAN REGION FREDERICK A. FREY Massachusetts Institute of Technology, Cambridge

Like the submarine igneous crust of the Indian Ocean, most Indian Ocean islands are predominantly formed of volcanic rock, usually basalt; however, others, such as Sri

INDIAN REGION

437

Lanka, Madagascar, Zanzibar, and the Seychelles archipelago, are predominantly formed of continental crustal rocks, such as granite, sediment, and their metamorphosed equivalents. This article summarizes major geologic and geographic features of many but not all islands within the Indian Ocean basin and discusses the plate tectonic implications of these islands. GEOLOGIC SETTINGS FOR TERRESTRIAL VOLCANISM

There are three types of geologic settings that lead to basaltic volcanism on Earth. Regions of Plate Divergence

Volcanoes form when plates move apart because the melting temperature of deep Earth material, the mantle, decreases with decreasing pressure; therefore, where plates move apart, the mantle rocks ascend and partially melt, creating basaltic magmas that erupt because they have lower density than their surroundings. Such magmas form the submarine volcanic crust of the Indian Ocean along the major spreading ridges within the Indian Ocean (i.e., the Southeast, Southwest, and Central Indian ridges) (Fig. 1).

Regions of Plate Convergence

Basaltic volcanoes also form when two plates collide and the denser plate, typically an oceanic plate, subducts beneath the less dense plate, typically a continental plate. Prior to subduction, the subducting oceanic plate interacts with overlying seawater and hydrated minerals form within the oceanic plate; as this plate is subducted to depths of 100 to 200 km, these hydrated phases become unstable and release water into the overlying mantle. Because water, like decreasing pressure, lowers the melting temperature of mantle rocks, basaltic magmas are created in subduction zones and ascend to form the arc volcanoes that are characteristic of convergent zones (e.g., the well-known “Ring of Fire” that surrounds the Pacific Ocean). Arc volcanoes form the eastern boundary of the Indian Ocean (e.g., the islands forming Indonesia). Intraplate Volcanism

Most of the volcanic islands in the Indian Ocean are distant from plate boundaries; they reflect intraplate volcanism. Often they are referred to as hotspot volcanoes because they apparently form above buoyant mantle, referred to as mantle plumes, which ascend and partially melt because the hotspot (plume) material is hotter and less dense than

FIGURE 1 Map of the Indian Ocean showing islands and submarine ridges discussed in text. Satellite altimetry data: Geoware GMT Companion CD-R

Vol. 1, Version 1.9, June 2006. Image prepared by Jenny Paduan and David Clague, MBARI.

438

INDIAN REGION

the surrounding mantle. In some regions these hotspots create volcanism for tens of millions of years, and they are relatively fixed in location; consequently, a linear chain of volcanoes forms as an oceanic plate migrates over the hotspot. The Indian Ocean contains two well-defined hotspot tracks, the Ninetyeast ridge in the eastern Indian Ocean and the Laccadive-Maldive-Chagos ridge in the central Indian Ocean.

ring in the 1990s. The majority of Andaman and Nicobar islands, however, are composed of metamorphosed sedimentary rocks (i.e., sandstones, shales, and limestones), which have been related to the Arakan Yoma mountain range that dominates western Myanmar. THE EASTERN INDIAN OCEAN ISLANDS: INTRAPLATE VOLCANOES Christmas and Cocos-Keeling Islands

THE EASTERN INDIAN OCEAN MARGIN: PLATE CONVERGENCE VOLCANISM Indonesian Volcanic Arc

The Indonesian volcanic arc includes many islands extending over 6000 km from northern Sumatra to the Banda Sea. From northwest to southeast, the arc can be divided into Sumatra; the Sunda arc, extending from Java to Flores; and the Banda arc. Along the Indonesia volcanic arc, there are more than 500 volcanoes, and more than 100 have had Holocene eruptions. These volcanoes are a consequence of eastward subduction of the Indian Ocean plate beneath continental rocks, such as those exposed in Sumatra. Volcanic rocks as old as Paleocene (~55 to 65 million years ago) are present along the Indonesian arc, but volcanism in the Indonesian region is most famous because two of the Earth’s largest and most explosive historic volcanic eruptions occurred at Tambora (1815) and Krakatau (1883); more recently, a major submarine earthquake west of Sumatra caused the devastating tsunami on December 26, 2004. Andaman and Nicobar Islands

The Andaman and Nicobar Islands are in the northeastern Indian Ocean, several hundred kilometers north of Sumatra and south of Myanmar; they are a Union Territory of India. There are ~258 Andaman islands (6408 km2) and 61 Nicobar islands (1841 km2); of these, only 37 are inhabited. More than 6000 people were killed by the 10–15-m tsunami arising from the 2004 submarine earthquake off the west coast of Sumatra. In the vicinity of these islands, the convergence of the Indian Ocean plate with southeast Asia is highly oblique, in contrast to the perpendicular convergence at Java. As is typical of convergent zones, there is a deep submarine trench west of these islands and seismic evidence that the Indian Ocean plate has subducted to at least 250 km below the islands; however, the oblique subduction zone geometry has not led to an active volcanic arc. Two outlying Andaman islands, 100 km to the east of the main group, are described as stratovolcanoes; Narcondam Island is an extinct volcano, but Barren Island is an active volcano with eruptions occur-

Christmas and Cocos-Keeling Islands, territories of Australia, are several hundred kilometers west of Java. They originally formed as large volcanoes rising several kilometers above the sea floor. Christmas Island, with a population of 1500 and 135 km2 in area, is composed of interbedded Tertiary volcanic rocks and carbonate sediments with overlying economic deposits of phosphate derived from guano. There are two volcanic sequences of alkalic basalt; the youngest and most voluminous erupted in the Lower Miocene. Subsequent to volcanism, the island was capped by a coral atoll and repeatedly uplifted, as documented by a terraced coastline. The Cocos-Keeling Islands comprise 27 small coral islands (14 km2), presumed to be constructed on old submerged volcanoes, which form two distinct coral atolls 24 km apart. Two islands are inhabited by a population of ~620. Kerguelen Plateau, Broken Ridge, Ninety East Ridge

The Kerguelen Plateau, the Broken Ridge, and the Ninety East Ridge are submarine volcanic structures on the eastern Indian Ocean sea floor. They are broadly consistent with a simple end-member model for intraplate volcanism arising from a hotspot represented by a mantle plume (Fig. 2). In this idealized model, a large igneous province (LIP) is created over a brief time interval, perhaps less than 1 million years, as the large plume head partially melts upon decompression; subsequently, an age-progressive and linear hotspot track (i.e., a line of volcanoes) forms as the oceanic plate migrates for many million years over the less voluminous hotspot tail. The geologic evolution of volcanoes forming a hotspot track as a migrating oceanic plate, such as the Indian plate, overrides a hotspot is as follows: (1) submarine eruptions construct a volcano that eventually emerges above sea level to form an island; (2) subsequently, volcanism wanes as the plate migrates away from the source of volcanism (i.e., the hotspot); (3) erosional processes (e.g., landsliding) become more important than volcanic construction processes; (4) the volcanic island subsides below sea level. INDIAN REGION

439

emerged above sea level and subsequently subsided as the Indian plate migrated northward over what is known as the Kerguelen hotspot. Where is the current volcanism created by the Kerguelen hotspot? Three possibilities are recently active volcanic islands constructed on the Kerguelen Plateau, the Kerguelen Archipelago, Heard Island, and the McDonald Islands. Kerguelen Archipelago

FIGURE 2 Simple plume head/tail model for hotspot volcanism. Pho-

tograph of a laboratory model of a starting thermal plume. The red fluid is buoyant hot material derived from a thermal boundary layer; the yellow fluid is cooler material entrained during ascent (modified, with permission, from Campbell 2005). Within the Earth, the buoyancy of mantle plumes are controlled by their composition as well as by their higher temperature; in this case, the plume head and tail structure is likely to be much more complex.

In this hypothesis, the Kerguelen Plateau/Broken Ridge LIP represents volcanism derived from the plume head, and the Ninety East Ridge, a long (more than 5000-km), age-progressive, and linear chain of volcanoes, represents volcanism derived from the plume tail. The Kerguelen Plateau and Broken Ridge were contiguous prior to their separation at ~40 million years ago by the Southeast Indian ridge. Both structures are now submarine, at water depths of 2 to 3 km, with plateaus rising several kilometers above the surrounding deep ocean floor. However, basalt samples recovered from Kerguelen Plateau and Broken Ridge by the Deep Sea Drilling Project and the Ocean Drilling Program, both of which utilized the JOIDES Resolution drilling ship, were erupted above sea level (i.e., they formed islands of unknown size). Subsequently, like the Hawaiian volcanoes that form the submarine Hawaiian Ridge, these islands subsided below sea level as volcanism ceased. Depending upon subsidence rates, the emergent parts of the Kerguelen Plateau may have had a role in paleobiogeography by serving as a land bridge between Antarctica and India/Madagascar. Also, basaltic drill cores from the age-progressive Ninety East Ridge (i.e., ~42 million-year-old volcanism in the south to ~77 million-year-old volcanism in the north) show that the Ninety East Ridge formed as a series of islands that

440

INDIAN REGION

The Kerguelen Archipelago is part of Terres Australes et Antarctique Françaises (TAAF) and is occupied at Port aux Français by a rotating group of ~100 scientists who study the geology, flora, and fauna of the islands. The archipelago consists of a major island (Grande Terre, 6500 km2) and many nearby, much smaller islands. There is a glacier (Cook) in the western highlands of Grande Terre, and the land surface was extensively glaciated during the Pleistocene; consequently, the coastal morphology is dominated by fjords. These fjords expose steep, ~1 km sections of nearly flat-lying, tholeiitic to transitional flood basalt flows erupted from ~30 to 24 million years ago. These flood basalts, erupted from unexposed fissures, cover 85% of the land surface. In some areas, especially in the west, plutonic rocks, gabbro, syenite, and granite are exposed; such occurrences of slowly cooled (i.e., intrusive) rocks are uncommon in oceanic islands. More recent volcanism, occurring 1 to 0.1 million years ago, in the archipelago erupted from localized vents and created steep-sided volcanoes formed by alkalic lavas, ranging from basanite to trachyte; an example is Mount Ross (1850 m), the highest point in the archipelago. The most recent volcanism (26,000 years ago) is a trachytic ignimbrite. Overall, there has been a longterm trend for archipelago volcanoes to change from tholeiitic and transitional basalt, erupted from 30 to 25 million years ago, to alkalic basalt and more differentiated lavas, such as phonolite and trachyte, erupted from 24 million years ago to the present. Some of these alkalic lavas contain abundant xenoliths of lower crust (granulites) and mantle (peridotite) rocks. Heard Island

Heard Island, a territory of Australia with no permanent residents, lies 450 km southeast of the Kerguelen Archipelago. It is a world heritage site. The island, largely (more than 70%) covered by glaciers, has two distinct parts: (1) a near circular (20 by 25–km) main island dominated by Big Ben, a steep-sided stratovolcano formed of alkalic basalt and trachyte (Fig. 3). The historically active (2007 eruption) Mawson Peak of Big Ben, with a summit elevation of

FIGURE 3 Big Ben volcano, the largest volcano on Heard Island in the

Central Kerguelen plateau; Compton glacier is in the foreground. Photograph courtesy of Dr. W. Powell (GEMOC ARC National Key Centre, Macquarie University, Sydney, Australia).

2745 m, is constructed within a 5–6-m wide caldera that is breached on the southwest flank of the volcano. Like the alkalic basalts in the Kerguelen Archipelago, xenoliths of mantle rock (peridotite) are found in alkalic basalt flows and boulders at Heard Island. The Laurens Peninsula is connected to the main island by a narrow, low isthmus; Mt. Dixon is the principal feature, and the oldest rocks outcrop on this peninsula. They are limestone deposited 45–50 million years ago that are intruded by thin (less than 2-m) sills of basalt. McDonald Islands

The McDonald Islands, also a territory of Australia and a world heritage site, are a group of islets that lie 40 km west of the Laurens Peninsula. The largest, McDonald Island (1 km2), is volcanically very active, erupting phonolite lava (2005 eruption); this island doubled in size from 2000 to 2001, destroying all vegetation. In summary, the current location of the Kerguelen hotspot is uncertain; however, the region occupied by Heard and McDonald Islands is the location of the most recent volcanism. Amsterdam and St. Paul Islands

Amsterdam and St. Paul Islands are young (less than 1 million years old) tholeiitic basalt volcanoes constructed on the submarine Amsterdam St. Paul (ASP) plateau. They are close—60 and 100 km, respectively—to the actively spreading Southeast Indian ridge (SEIR). Both islands are French (TAAF) territories, and a small scientific base is maintained on Amsterdam Island. This island, with an area of 55 km2 and a peak elevation of 881 m, is a subcircular (10 by 7–km) volcanic cone that has steep erosional cliffs on the west flank; they perhaps reflect a major

landslide collapse. Amsterdam lavas were first inferred to be 0.4 to 0.2 million years in age, which is considerably older than St. Paul Island lavas, but younger, 0.01 to 0.2–million-year-old ages have been recently reported. For Amsterdam Island lavas, no historical eruptions are known, but in 2000 recent volcanism was discovered at Boomerang sea mount, only 18 km northeast of Amsterdam Island. St. Paul Island, 100 km south of Amsterdam Island, is a small cone, 8.4 km2 in area and 264 m in elevation, whose northwest sector has collapsed enabling seawater access to the 1.6-m diameter crater. St. Paul Island has been inferred to be younger (by 0.4 to 0 million years) than Amsterdam Island. Although basalts from these two islands on the ASP plateau are geochemically distinct, they share some geochemical characteristics with basalt erupted along the Ninety East Ridge. This observation, coupled with a series of sea mounts extending to the northeast from the ASP plateau, suggests that the hotspot forming the ASP plateau and its islands may have also played a role in forming the NER. In summary, volcanism in these islands and a sea mount on the ASP plateau represent recent hotspotderived volcanism located near a spreading ridge axis, much like the Galápagos Islands and unlike volcanism on the Kerguelen archipelago, Heard Island, and the McDonald Islands, which are many hundreds of kilometers distant from the SEIR. THE DECCAN PLATEAU, THE LACCADIVE–MALDIVE–CHAGOS RIDGES, AND THE MASCARENE ISLANDS IN THE CENTRAL INDIAN OCEAN: PLUME HEAD AND TAIL VOLCANISM

The Deccan plateau, a continental flood basalt covering much of northwestern India, is a large igneous province derived from a plume head, and the Laccadive–Maldive– Chagos ridges and Mascarene Islands form a hotspot track, which ends at the currently active volcano, Piton de la Fournaise, on Réunion Island. Like the Ninety East Ridge, the oldest part of the Réunion to Deccan hotspot track strikes south–north (i.e., the Laccadive–Maldive– Chagos Ridge), reflecting northward migration of the Indian Plate over the Réunion hotspot. However, the younger part of the track strikes southwest–northeast because it was separated from the south–north segment of the hotspot track by post-Eocene spreading on the relatively young Central Indian Ridge. Age determinations of volcanic rocks recovered from ten geographically

INDIAN REGION

441

and temporally distinct locations along the Réunion to Deccan hotspot track are in excellent agreement with the volcanism originating from a near stationary Réunion hotspot, as the Indian plate migrated northward with subsequent southwest–northeast spreading on the Central Indian ridge. Deccan Plateau

The Deccan Plateau (often referred to as Deccan Traps) is a well-known continental flood basalt, a type of large igneous province because of its very large volume, 1–10 million km3, of tholeiitic basalt, which erupted over a relatively narrow time interval of less than 500,000 years 65.5 million years ago, an age that is indistinguishable from the massive extinction event that occurred at the Cretaceous–Tertiary boundary. Extending southward from the Deccan Plateau, the north–south-trending Laccadive–Maldive–Chagos submarine ridges, which mark the western boundary of the Central Indian basin, extend for more than 2500 km between 14° N and 9° S. Although their upper surfaces are carbonate bank and reef deposits up to 2 km thick, seismic data show that these ridges form a continuous volcanic structure. Sea floor drilling and coring by the JOIDES Resolution on the northern margin of the Maldive Ridge and the northern edge of Chagos Ridge recovered tholeiitic basalt that erupted subaerially or at shallow marine depths; these basalts are from ancient, submerged islands underlying carbonate sediments and reef deposits. Laccadive Islands

The Laccadive Islands, now known as Lakshadweep, are the subaerial portion of the Laccadive Ridge; they are a Union Territory of India in the Arabian Sea, 220–440 km from the southwest (Malabar) coast of India. The archipelago consists of 36 islands, largely atolls and reefs with a land area of 32 km2. The western rim of the archipelago, exposed to the southwest monsoon, is mostly submerged coral reef, whereas the relatively protected eastern rim includes ten inhabited islands with a combined population of 60,000. These low-lying coral atolls overlie extinct, submerged volcanoes that form the northern end of the Réunion to Deccan hotspot track; these volcanoes are inferred to have formed subsequent to the Deccan Plateau (i.e., less than 65.5 million years ago); however, they have not been sampled by drilling. Maldive Islands

The Maldive Islands, 595 km south of India, consist of 26 atolls comprising ~1200 islands (298 km2) extending from

442

INDIAN REGION

7°10´ N to 0°45´ S. A population of 300,000 inhabitants on ~200 islands form the Republic of the Maldives. These islands are exclusively formed of coral reef deposits with a maximum land elevation of 2.3 m, average of 1.2 m; there is much concern about the future effects of a rising sea level, which has already risen ~20 cm over the last century. Like the Laccadive Islands to the north, the coral atolls of the Maldives were constructed on ancient submerged volcanoes. These volcanoes were sampled by the Ocean Drilling Program at 5°05´ N, 73°50´ E on the eastern margin of the submarine Maldive ridge. Based on a hotspot track model, an age of 55 million years was predicted for the basaltic core recovered from beneath 211 m of shallow water limestone (i.e., reef deposits). Subsequent dating of the recovered tholeiitic basalt using the K-Ar radiogenic decay system determined an eruption age of 57 million years ago, an age that is in excellent agreement with that predicted by the hotspot model. Chagos Archipelago

The Chagos Archipelago, a British Indian Ocean territory, 456 km south of the southernmost Maldive Islands, is the youngest expression of the submarine Laccadive– Maldives–Chagos volcanic ridge. This archipelago, extending from 04°54´ S to 07°39´ S, consists of seven coral atolls and many islets with a land area of 63 km2 and a population of 3500. The largest island, Diego Garcia (27 km2), is a naval and air military base jointly operated by the United States and United Kingdom. Construction of this base required forced evacuation of the local inhabitants to the Seychelles and Mauritius. The consequent human-rights controversy coupled with the strategic military location (i.e., close to volatile Middle Eastern countries) has led to Diego Garcia becoming the most well-known coral atoll in the Indian Ocean. Drilling of the sea floor by the Ocean Drilling Program on the northern margin of Chagos Bank at 4°12´ S recovered tholeiitic basalt below 107 m of sediment. K-Ar dating of this basalt yielded an age of 49 million years, again in excellent agreement with the age range 45–50 million years, predicted by a simple hotspot track model. Mascarene Islands

The Mascarene Islands in the western Indian Ocean include the volcanic islands of Réunion, Mauritius, and Rodrigues. Reunion and Mauritius define the less-than10-million-year southwest–northeast trend of the Réunion to Deccan hotspot track that began with formation of the Deccan Plateau at 65.5 million years.

Réunion Island

Réunion Island, 750 km east of Madagascar with an area of 2512 km2 and a population of 793,000, is an overseas department of France, and as one of the 26 regions of France, its inhabitants have the same status as citizens of European France. The island, rising 7 km above the surrounding sea floor, consists of two volcanoes, Piton des Neiges and Piton de la Fournaise; the latter is the current volcanic expression of the hotspot that created the Deccan plateau 65.5 million years ago. These two volcanoes at Réunion were simultaneously active, but Piton des Neiges in the central part of the island, with a peak elevation of 3070 m, is older; the earliest volcanism, basalt to hawaiite, began ~2 million years ago and continued until at least 30,000 years ago. Piton des Neiges now appears to be extinct. The surface of the central part of Piton des Neiges is spectacularly rugged because several deeply dissected gorges broaden upstream into large erosional cirques. These cirques provide windows into the early igneous history of the volcano by exposing a basement complex that includes breccias and intrusive rocks, such as layered gabbros. Piton de la Fournaise, 2631 m at its peak elevation, occupies the eastern part of the island and is crowned by a concentric series of caldera collapse structures. The oldest lavas, dominantly alkalic basalt, erupted ~0.5 million years ago, and volcanism continues today with more than 150 eruptions in the last 300 years; the most recent occurred in April 2007 (Fig. 4). Although the oldest lavas were alkalic basalt, tholeiitic basalt is the most abundant lava type. Piton de la Fournaise and the much smaller McDonald Island on the Kerguelen Plateau are currently the most active volcanoes in the Indian Ocean. The eruptions and lavas at Piton de la Fournaise have been intensively studied, and a volcano observatory, 15 km from the summit, was constructed in 1979 and is operated by the Institut de Physique du Globe in Paris. Mauritius

Mauritius is an island nation, the Republic of Mauritius, located 1000 km east of Madagascar and 220 km northeast of Réunion Island; it has an area of 2040 km2 and a population of 1.25 million. The volcanism that created Mauritius is divided into three stages; the older series, erupted from 7.8 to 5.4 million years ago, is the erosional remnant of a large subaerial shield volcano with a current maximum elevation of 828 m. The oldest lavas of the older series are alkalic olivine basalt; some of these lavas contain xenoliths of cumulate rocks such as dunite, anorthosite, and syenite. Such samples of cumulate rocks provide insight into the

FIGURE 4 Lava from Piton de la Fournaise entering the sea (©KM

KRAFFT/CRI-Nancy-Lorraine, with permission).

partial crystallization and resulting mineral segregation processes that control magma evolution during magma ascent. In fact, coarse-grained xenoliths are abundant at each of the Mascarene islands, principally ultramafic and gabbro xenoliths in Réunion lavas and gabbros in Rodrigues lavas. The younger lavas of the older series, aged between 6.8 and 5.5 million years, include more differentiated lavas such as hawaiite and trachyte. After a 2-million-year hiatus in volcanism, the intermediate series lavas, dominantly alkalic basalt, erupted from 3.5 to 1.9 million years ago. Finally, from 0.7 to 0.3 million years ago, the younger series lavas erupted and covered most of the island with alkalic olivine basalt. In summary, the lava compositions erupted at Réunion and Mauritius Islands are quite different, with predominantly alkalic basalt lavas at Mauritius and predominantly tholeiitic basalt at Réunion. Nevertheless, the radiogenic isotopic ratios of Sr, Nd, Hf, and Pb are similar in the older series (i.e., shield-building lavas on Mauritius and the shield lavas on Réunion Island). This similarity in isotopic ratios and, even more importantly, the inferred 7–8 million year age for construction of the Mauritius shield INDIAN REGION

443

are consistent with the interpretation that Mauritius belongs to the Deccan to Réunion hotspot track (i.e., the Mauritius shield is 7–8-million-year-old volcanism resulting from the hotspot now underlying the active Piton de la Fournaise volcano on Réunion Island). Rodrigues Island

Rodrigues Island, the easternmost Mascarene Island, 560 km east of Mauritius, has a population of 40,000; it is part of the Republic of Mauritius. Located on the eastern end of submarine Rodrigues Ridge, a 600-km-long east– west structure, the volcanic island surrounded by a coral reef is 109 km2 with a peak elevation of 396 m. The submarine Rodrigues Ridge has been sampled by dredging and consists of basalt and trachyte erupted 7–11 million years ago. It has been suggested that the Rodrigues Ridge represents channelized flow of hotspot-related mantle toward the Central Indian Ridge. Rodrigues Island consists predominantly of alkalic basalt erupted ~1.5 million years ago; consequently, unlike lavas erupted at Réunion and Mauritius, Rodrigues Island with its relatively young age is not on the Deccan to Réunion hotspot track. THE WESTERN INDIAN OCEAN ISLANDS: INTRAPLATE VOLCANOES Crozet Archipelago

The Crozet Archipelago consists of five main islands located south of the Southwest Indian Ridge (SWIR). They are a French (TAAF) territory and a National Park. There are no permanent residents, but as at Kerguelen and Amsterdam Islands, a rotating group of scientists occupy a permanent base, Port Alfred, on Ile de la Possession. These islands, on the large (4500-km2) west-northwest– east-southeast–trending submarine Crozet plateau, are subdivided into a western group of three relatively small islands—Ile aux Cochons, a 67 km2 stratovolcano, and two smaller reef islands, Ile des Pingouins (3 km2) and Ilots des Apotres (2 km2)—and an eastern group, 100 km to the east of two larger islands: Ile de la Possession (150 km2) and Ile de l’Est (130 km2). These eastern islands, with maximum elevation of ~1 km, are eroded stratovolcanoes, predominantly composed of alkali basalt and associated cumulate (e.g., ankaramite) and differentiated lavas (e.g., phonolite). Ile de l’Est is deeply dissected by 0.5- to 1-km, glacially carved U-shaped valleys, whereas Ile de la Possession has a more subdued surface. The main volcano on Ile de la Possession erupted lavas from ~8 to 0.5 million years ago, but very young appearing, more-than-100-mhigh scoria cones, probably Holocene in age, are common on the surface; hot springs are also present. 444

INDIAN REGION

Marion and Prince Edward Islands

Marion and Prince Edward Islands are South African territories, on the Antarctic plate, 160 to 200 km south of the actively spreading Southwest Indian ridge. The pronounced, north–south trending submarine Madagascar plateau, extending from the southern tip of Madagascar Island to the SWIR, is inferred to be a hotspot track formed as the Indian plate migrated northward relative to the Marion hotspot. The presumed volcanic rocks forming this submarine plateau have not been sampled. Marion Island (290 km2) hosts a research station. The island has a low dome-like profile, has a 1230-m summit elevation, and is formed of two basaltic shield volcanoes overlain by ~150 cinder cones. The interior highlands, including a glacier, rise from a coastal plain via a 200–400 m escarpment. The lavas, predominantly alkalic basalt and hawaiite, are divided into two age groups: (1) an older, glaciated gray lava series erupted less than 500,000 years ago (i.e., Pleistocene-age lavas) and (2) a younger black lava series that postdates glaciation (i.e., Holoceneage lavas). Unvegetated lava flows indicate recent volcanism; the first historical eruption occurred in 1980, and a recent eruption occurred in 2004. The smaller (45-km2 and 72-m-maximum-elevation) and uninhabited Prince Edward Island, 22 km northeast of Marion Island, also features central highlands bounded by a prominent escarpment. It is a remnant of a shield volcano; Holocene volcanism is reflected by numerous scoria and tuff cones. Apparently, Marion and Prince Edward Islands experienced contemporary volcanism. Comoros Archipelago

The Comoros Archipelago consists of four islands: Grand Comoro (Ngazidja), Moheli (Mwali), Anjouan (Nzawani), and Mayotte (Maore). The first three islands form the Union of the Comoros, with a population of 676,000; the island of Mayotte is administered by France but claimed by the Union of Comoros. The four islands, midway between Madagascar and northern Mozambique, are the summits of a northwest to southeast trending volcanic ridge that crosses the Mozambique Channel northwest of Madagascar. Grand Comoro, at the northwest end of the archipelago, is the largest (1013 km2) of the four islands; it consists of two undissected, coalescing shield volcanoes made up largely of alkalic basalt. These shields are La Grille (1087 m), active in the Holocene, and Karthala (2361 m), which has a 3 by 4–km summit caldera; it last erupted in 2006. To the southwest, Moheli and Anjouan Islands are deeply eroded alkalic basalt shields, modified by extensive faulting. At the southwest

extremity of the archipelago, Mayotte Island is a subdued (660 m maximum elevation), intensively eroded volcano with an embayed shoreline reflecting subsidence of the shield; the island has a continuous barrier reef reaching 12 km in diameter. The southeast to northwest age progression, obvious from geomorphology and confirmed by K-Ar dating, appears to reflect a hotspot track formed by southeast migration of the Somali plate over a Comoros hotspot. However, unlike the Hawaiian and Réunion shields, which are predominantly composed of tholeiitic basalt, the Comoros volcanoes are composed of alkalic basalt. A distinctive feature of these alkalic lavas is the abundance of xenoliths (i.e., exotic blocks of coarsegrained rocks, predominantly gabbros, dunites, and peridotite). These xenoliths provide insights into the geologic processes occurring at depth within the volcanoes. INDIAN OCEAN ISLANDS PREDOMINANTLY COMPOSED OF CONTINENTAL CRUST Sri Lanka

Sri Lanka is a heavily populated (~20 million), large island (65,610 km2), less than 130 km south of India. At some time there may have been a direct land-bridge connection between Sri Lanka and southern India known as Ramu’s Bridge in Hindu mythology. More than 90% of the Sri Lankan land surface is metamorphic Precambrian rock (as old as 2 × 109 years); these rocks are ancient granites and sediments that were strongly recrystallized (i.e., metamorphosed at high pressures and temperatures to form granulites). The Precambrian rocks of Sri Lanka were once part of the ancient supercontinent of Gondwana, originally composed of Antarctica, Australia, India, and Africa, before it began to break apart 200 million years ago. Ongoing research is investigating similarities between the Precambrian rocks of Madagascar, Antarctica, and Southern India. Madagascar

Madagascar, which has a population of 17 million people, is one of the largest islands on Earth and is the largest island (587,000 km2) in the Indian Ocean. The Mozambique Channel separates Madagascar from Africa by 250–400 km. As with Sri Lanka, more than two-thirds of Madagascar’s land surface is Precambrian rock, as old as 3.2 × 109 years. Also like Sri Lanka, now far to the northwest, Madagascar was part of Gondwana and was probably contiguous with southern India prior to breakup of Gondwana. Madagascar was created as an island by two major rifting events during Gondwana’s break-up: The first was the separation of Madagascar and India

from Africa ~180 million years ago; the second resulted in the separation of India and Madagascar 88 million years ago—an event that was accompanied by extensive basaltic (i.e., magma derived from mantle rocks) volcanism. Although much less voluminous than the old continental crustal rocks, this volcanism 88 million years ago is important in the context of intraplate volcanism because the volcanic rocks occurring on the east and west coasts of Madagascar are an extension of the Madagascar submarine plateau that extends from the southern tip of the island to the Southwest Indian ridge. The Madagascar basaltic lavas are inferred to be early volcanism related to the Marion mantle plume, which is currently creating Marion Island. Zanzibar Archipelago

The Zanzibar Archipelago, only 40 km east of mainland Africa, consists of two main islands, Pemba and Unguja, and several surrounding islets; Pemba is larger and older than Unguja. In contrast to Sri Lanka and Madagascar, these islands are relatively young, Miocene to recent. They formed as a river delta, their maximum elevations are 120 m, and they consist of sedimentary rocks, such as limestone, sand, silt, and clay. Seychelles Archipelago

The Seychelles Archipelago (455 km2) consists of ~115 islands, population of ~80,000, covering a large area (1.4 × 106 km2) of the western Indian Ocean north of Madagascar. Surprisingly for islands distant from continental margins—the Seychelles are 480 to 1600 km from the east coast of Africa—~40 Seychelles islands are predominantly composed of Precambrian continental crustal rocks. For example, undeformed and unmetamorphosed granitic rocks formed 750 million years ago are abundant on the major island of Mahé. Further to the south, ~75 smaller Seychelles islands are composed of sedimentary carbonates and coral. However, northwest of Mahé, two islands, Silhouette and Isle du Nord, consist of syenite and related rocks that are ~63 million years old, an age close to that of the Cretaceous–Tertiary boundary. This age is also close to the eruption age (65.5 million years ago) of the Deccan Plateau, a LIP formed when tremendous volumes of basaltic magma (~8 × 106 km3) erupted in northwestern India over less than 1 million years. Moreover, basaltic dikes on Praslin Island, northeast of Mahé, have geochemical characteristics very similar to a distinctive geochemical group of Deccan Plateau basalts. Consequently, it is inferred that the Seychelles granitic islands were previously located at the northwest margin

INDIAN REGION

445

of India, along with Madagascar, and that they were subsequently transported to their present location, far south of India, by sea floor spreading. In summary, Sri Lanka, Madagascar, and the Seychelles contain ancient (i.e., Precambrian) continental crust that formed well before formation of the Indian Ocean basin, which was created when Gondwana, a large supercontinent composed of Africa, India, Australia, and Antarctica, broke apart between ~200 and 132 million years ago. Therefore, these islands are continental fragments that were stranded within the Indian Ocean after the break-up of the Gondwana supercontinent. FURTHER READING

Campbell, I. H. 2005. Large igneous provinces and the mantle plume hypothesis. Elements 1: 271–275. Global Volcanism Programs. http://www.volcano.si.edu/ The Great Plume Debate. http://www.mantleplumes.org/ Nairn, A. E. M., and F. G. Stehli, eds. 1982. The ocean basis and margins, volume 6, the Indian Ocean. New York: Plenum. Yoshida, M., B. F. Windley, and S. Dasgupta, eds. 2003. Proterozoic East Gondwana: supercontinent assembly and breakup. Geological Society, London, Special Publications 206.

INDONESIA, BIOLOGY TIGGA KINGSTON Texas Tech University, Lubbock

Indonesia is one of the most biologically rich countries in the world, yet dramatic land-use changes in recent decades place much of this biodiversity in peril. The high levels of species richness and endemism are in large part attributable to a complex geological history that has both generated a profusion of island speciation centers and brought together two very different biological realms. BIOGEOGRAPHY OF INDONESIA Geographic Setting

Straddling the equator from 6° N to 11° S, the Indonesian archipelago comprises over 18,000 islands that extend from Sumatra in the Paleotropical realm to New Guinea and the Aru Islands in the Notogean (Australian) realm. Between these two extremes lies a transitional region, Wallacea, with biological affinities to both realms and a high incidence of endemic species. Fluctuating sea levels and tectonic movements throughout the Cenozoic facilitated connections

446

INDONESIA, BIOLOGY

among many of the island groups, but others, particularly those in the deep seas of the Wallacea region, have been isolated for millions of years, limiting overall diversity but resulting in a profusion of endemic species. As mountain systems arose, many were periodically isolated by rising sea levels, creating additional opportunities for speciation. Biogeographic History

In 1854, Alfred Russel Wallace set sail from Singapore to explore the biology of the Malay Archipelago (now Malaysia and Indonesia) (Fig. 1). He spent eight years crisscrossing between the islands and collected over 125,000 mammal, bird, and butterfly specimens. His interpretations of the distributions of the species he collected were published in his book The Malay Archipelago in 1869, laying the foundation for much of modern biogeography. One of his most intriguing observations was the apparent lack of relationship between the distances among some islands and their faunal affinities. On Borneo he found a diversity of Old World monkeys (Cercopithecidae), wild cats (Felidae), deer (Cervidae), and civets (Viverridae) that were largely absent from Sulawesi (Celebes), some 200 km away across the Makassar Strait. Moreover, Sulawesi lacked the overall diversity of Borneo, but had numerous endemic species. Some endemics had clear Asian affinities: the tusked babirusa (Babyrousa babyrussa), a member of the pig family; the dwarf buffaloes or anoas (Bubalus depressicornis and B. quarlesi); and the endemic macaques (Macaca) and tarsiers (Tarsius). Others, such as the marsupial cuscus species (Strigocuscus pelengensis and S. celebensis) and the maleo (Macrocephalon maleo), a large turkey-like bird (Megapodiidae) that uses geothermal heat to incubate its eggs, had clear Australasian affinities. Conversely, despite far greater intervening distances, the Borneo fauna was very like that of Sumatra, Java, and the Malay Peninsula, notable for the shared presence of gibbons (Hylobatidae), rhinoceros (Rhinocerotidae), and elephants (Elephantidae). A similar disjuncture occurred between Lombok and Bali in the Lesser Sundas. The separation provided by the Lombok Strait was as little as 6 km in some places, yet many families of birds that were common on Bali and its western neighbors were greatly reduced or absent east of Lombok (e.g., the barbets [Megalaimidae], bulbuls [Pycnonotidae], and woodpeckers [Picidae]), whereas Lombok seemed to mark the western limit of Australian birds such as the honeyeaters (Meliphagidae), cockatoos (Cacatuidae), and brush-turkeys (Megapodiidae).

The Wallace Line

To delineate the biogeographical divisions he observed, Wallace described a line following the deep water of the Lombok Strait up through the Makassar Strait between Borneo and Sulawesi. Today, the Wallace Line is best viewed as the eastern edge of a region dominated by Asian species and the western limit of Australian species, and Huxley’s modification of this line marked the boundary of the Oriental region of the Paleotropical Realm. Lydekker’s Line delineates the boundary of the Australian region, part of the Notogean (or Australian) Realm, and represents the eastern limit of Asian species and the western edge of dominance by Australian species (Fig. 2). The area between these lines is now generally accepted as a biogeographic region in its own right, Wallacea. Named for Wallace, the region includes Sulawesi, the Lesser Sunda Islands (except Bali), and the Moluccas (excluding the Aru Islands). Although some view this as a just transition zone between the Oriental and Australian regions, others argue that the high degree of endemicity in most taxonomic groups confers region status. Biogeographic divisions based on flora differ slightly in that the Australian kingdom does not include New Guinea, and as a consequence all the flora of Indonesia falls into the Malesian subkingdom of the Paleotropical kingdom. The Malesian subkingdom is further split into subdivisions that, again, differ slightly from the faunal regions. The western subdivision includes Sumatra and Borneo (with the Malay Peninsula and the Philippines); Java and the Lesser Sundas are in the southern division; the eastern division encompasses Sulawesi, the Moluccas, and New Guinea. Tectonic Origins and Pleistocene Sea Levels

Wallace correctly inferred that Sumatra, Borneo, Java, and the Malay Peninsula, separated by wide expanses of uniformly shallow seas (Fig. 2), were at one point part of a single land mass. His boundary traces the edge of the Sunda Shelf, the continental shelf extension of the Southeast Asian mainland. Similarly, Lydekker’s Line traces the Sahul Shelf underlying New Guinea, Australia, and the Aru Islands (shallow, smooth water in Fig. 2). Wallace speculated that the similarities among species on islands within these areas might be explained by periods of low sea level, connecting the islands and promoting biotic exchange. However, it was not until the field of plate tectonics came to the fore in the 1970s that a mechanistic explanation for the complex mix of Australian and Oriental faunas could be proffered. Although there are limits to the extent to which geology alone can explain

Mainland South East Asia

Philippines

Pacific Ocean

South China Sea Malay Peninsula

Mentawi Islands

Halmahera Kalimantan (Borneo)

Sum

atra

West Papua & Papua (New Guinea) Sulawesi

Seram Buru Moluccas

Bali Flores Indian Ocean

Java

500

1000

Papua New Guinea

Banda Islands Lombok Sumba

0

Aru

2000

Timor

Lesser Sundas Australia

Kilometers

FIGURE 1 Geography of Indonesia detailing the main islands and those

discussed in the text. Areas shaded dark green are part of Indonesia; light green areas are other Southeast Asian countries.

biogeography, it does provide a general explanation for the broader distribution patterns seen in Indonesia. Prior to the Cenozoic, the various land masses that now constitute Sundaland were last in contact with the northern Australia–New Guinea continental margin about 200 million years ago, when they formed part of the great southern continent Gondwana. Continental slivers successively rifted from Gondwana during the Mesozoic, drifting north to eventually collide with Asia and join to form the Sundaland core. Ocean basins created in the process kept the Oriental biota far removed from that of Australia, and it was not until approximately 50 million years ago that tectonic movements split Australia and New Guinea from Gondwana, and the Australia plate began to rapidly drift northward. It brought with it very different flora and fauna from that seen in Asia, with Gondwanan groups that are considered diagnostic for the Notogean Realm including four of the six orders of mar-

FIGURE 2 Shaded relief map of Indonesia and surrounding Southeast

Asian countries to illustrate topography and bathymetry, key faunal boundary lines, and the Sundaland and Wallacea biodiversity hotspots (dashed red lines).

INDONESIA, BIOLOGY

447

FIGURE 3 Terrestrial Ecoregions of Indonesia. Constructed from the Terrestrial Ecoregion GIS Database of Olson et al. (2001).

supials, monotremes (egg-laying mammals such as the platypus [Ornithorhynchus anatinus]), ratites (flightless birds such as the cassowary [Casuarius]), lungfish, snakenecked turtles (Chelydae), and tree frogs (Hylidae). In the last 30 million years there has been progressive closure of the marine gap between the Oriental and Australian faunas as the Australian plate collided with the Eurasian plate. The collision also led to the emergence of many of the Wallacea islands, as well as to the uplift of the mountains of New Guinea. However, connections that might have permitted direct migrations between the two faunas appear to have been limited; tectonic movements also opened up deep ocean basins in the Wallacea region, and sea-level fluctuations were such that many of the islands that, as we recognize today, would not have been exposed (and in some cases would not have originated) until approximately 5 million years ago. In contrast, exposure of the Sunda Shelf would have linked Borneo, Java, Sumatra, the Malay Peninsula, and intervening small islands to create a single geographical unit, Sundaland, which was largely emergent or intermittently transected

448

INDONESIA, BIOLOGY

by a very shallow sea throughout the Cenozoic, providing a biogeographic connection from Borneo through to Asia. Cenozoic connections between Australia and New Guinea were limited, as the land masses were not consistently emergent before the last 5 million years. The most recent opportunity for land connections among the Indonesian islands was during the last glacial maximum, approximately 18,000 years ago. Sea levels were about 120 m lower than today, exposing much of the Sunda Shelf. Although there is some debate about the nature of the exposed vegetation (whether forest and mangrove swamp or savannah), Borneo, Java, Sumatra, and the Malay Peninsula would have been connected. Similar exposure of the Sahul Shelf would have connected New Guinea with Australia and the Aru Islands. The point at which connections between islands were broken coming out of the last ice age varied with the depth of the surrounding sea floor, but by 7000 years ago, sea levels would have risen enough to present the Indonesian archipelago configuration much as we recognize it today.

TABLE 1

Biodiversity of Indonesia Number of species

Vascular plants Mammals Birds Reptiles Amphibians

Projected species loss by 2100

Recorded

Endemic

Indonesian extinctions

Global extinctions

(% world total)

(% Indonesian total)

(min.–max.)

(min.–max.)

5797–28,057 161–537 231–624 32–407 12–251

3454–16,715 97–323 149–402 13–167 10–219

29,375 (11) 667 (14) 1604 (16) 749 (9) 347 (6)

17,500 (60) 216 (32) 443 (28) 227 (32) 159 (12)

note: Table shows number of species recorded, endemic species, and projected species losses by 2100 for five major components of biodiversity.

Terrestrial Ecoregions

The biotic composition of an individual island also reflects the size of the island and the diversity of habitats that occur on it. Moreover, the extent to which particular groups of species are able to overcome dispersal barriers varies significantly with the ecology of the group or even of the species. Nonetheless, despite the complex interplay of geology, sea levels, and ecology, geographically and biotically distinct assemblages of natural communities have evolved and can be grouped as ecoregions. Thirty-seven ecoregions are described for Indonesia (Fig. 3), and their distribution reflects the synergy between biogeographic processes. Larger islands with complex topography and geological history typically support several ecoregions. New Guinea is bisected by a very high mountain cordillera (peaks reaching 4000 m) separating the island into distinct north and south regions and supporting several montane habitats. Thus, there are 11 ecoregions in the western half of the island alone. At the same time, the distribution of ecoregions on different islands, particularly those of Wallacea, frequently reflects land connections or low sea levels during the Pleistocene. Despite the proximity of the islands of the Lesser Sundas, they constitute three ecoregions. The Lesser Sundas deciduous forests include Lombok, Sumbawa, and Moya (which were merged as Greater Sumbawa) and Komodo, Flores, and Lembata (Greater Flores). In the outer Banda arc, Sumba was separated from Greater Flores by some 50 km of deeper water and supports its own ecoregion, as do the islands that made up Greater Timor (Roti, Semau, and Timor) and the island of Wetar. In the Moluccas, the Halmahera Islands were most likely connected during the Pleistocene, and they now compose a single ecoregion (Halmahera rainforests). Similarly, Ambon and Seram both fall within the Seram rainforests, but neighboring Buru, which has always been isolated by deep seas, has its own ecoregion.

BIODIVERSITY OF INDONESIA: TRENDS AND KEY FEATURES

Home to nearly 12% of the world’s land vertebrates and 10% of the world’s vascular plants, Indonesia is one of the most biologically rich countries in the world. Critically, much of the country’s flora and fauna is found nowhere else; endemic species account for 45% of the amphibians, 32% of the mammals, 28% of the birds, and a staggering 60% of the vascular plants. Tragically, only about 50% of the original natural forest area remains, and based on current deforestation rates, as little as 11% may survive to the end of this century. As a consequence, even the minimum projected species losses for some groups are of major concern, with the global extinction of 97 of the 667 mammal species and 149 of the 1604 bird species anticipated to occur by 2100 (Table 1). Indonesia has consequently been considered a “megadiversity country” since the introduction of the concept by Russell Mittermeier in the late 1980s. It encompasses the Sundaland and Wallacea biodiversity hotspots, apolitical areas delineated by high levels of vascular plant endemism (more than 1500 species) and a high degree of habitat threat (70% loss of original vegetation), and prioritized further by considerations of endemism in birds, mammals, reptiles, and amphibians. In the east, about half the New Guinea Tropical Wilderness Area falls within Indonesia (Fig. 2, Table 2). Sundaland Biodiversity Hotspot

Biotic exchange during periods of exposure of the Sunda Shelf has resulted in major floral and faunal similarities across the hotspot, particularly in the lowland forests, which are dominated by the family Dipterocarpaceae. Nonetheless both species richness and endemism are high in all groups (Table 2), with particularly high levels of endemism in the montane ecoregions.

INDONESIA, BIOLOGY

449

TABLE 2

Geography and Diversity of the Sundaland and Wallacea Biodiversity Hotspots and New Guinea Wilderness Area

Geography Area (km2) Vegetation remaining (%) Protected area (%)a Population density (people/km2) Terrestrial ecoregions

Sundaland

Wallacea

New Guinea

biodiversity hotspot

biodiversity hotspot

wilderness area

1,501,063 6.7

338,494 15.0

828,818 70.0

5.2 153

5.8 81

11.0 5.1

16

9

17b

Species richness

Recorded

Endemic (%)

Recorded

Endemic (%)

Recorded

Endemic (%)

Vascular plants Mammals Birds Reptiles Amphibians

25,000 380 769 452 244

15,000 (60) 172 (45) 142 (18) 243 (54) 196 (80)

10,000 222 647 222 48

1500 (15) 127 (57) 262 (40) 99 (45) 33 (69)

17,000 233 650 275 237

10,200 (60) 146 (63) 334 (58) 159 (58) 215 (91)

note: Figures include contributions from Malaysia and Brunei to the Sundaland Hotspot and from Papua New Guinea to the New Guinea Wilderness Area. a Protected areas in IUCN categories I–IV are those managed primarily for science, conservation, or ecosystem function and afford the highest levels of protection. b 12 ecoregions in Indonesian New Guinea.

The hotspot is perhaps best known for its many charismatic but threatened mammals, including two species of orangutan (Pongo pygmaeus and P. abelii ); two critically endangered rhinoceros (Rhinoceros sondaicus and Dicerorhinus sumatrensis); the Sumatran tiger (Panthera tigris sumatrae); two subspecies of Asian elephants (Elephas maximus sumatrensis and E. m. borneensis); the vulnerable Malayan tapir (Tapirus indicus), and the endangered proboscis monkey (Nasalis larvatus). Sundaland is also noteworthy for several large endemic threatened reptiles such as the endangered false gharial (Tomistoma schlegelii), a freshwater crocodilian species found mostly in Sumatra and Borneo, and two species of critically endangered large river terrapins: the mangrove terrapin (Batagur baska) and the painted terrapin (Callagur borneoensis). SUMATRA

Species richness on Sumatra is comparable with the larger islands of Borneo and New Guinea, with approximately 10,000 plant species, 201 mammal species, and 465 breeding bird species. The diverse lowland fauna has much in common with other parts of Sundaland, but the proximity and past connections to peninsular Malaysia mean that a number of Asian species, including 22 mammals, are found that do not occur on other Indonesia islands (e.g., the siamang [Symphalangus syndactylus], the great hornbill [Buceros bicornis], the monotypic fire-tufted barbet [Psilopogon pyrolophus]). In contrast, the more isolated montane forests support eight endemic bird species, such as the threatened Sumatran cochoa (Cochoa beccarii) and 450

INDONESIA, BIOLOGY

Sumatran ground-cuckoo (Carpococcyx viridis), and seven endemic mammals, including Thomas’s leaf-monkey (Presbytis thomasi) and the critically endangered Sumatran rabbit (Nesolagus netscheri). The drier montane forests are dominated by the Sumatran pine (Pinus merkusii), which forms the only coniferous ecoregion in Indonesia. The forests of both Sumatra and Borneo are also noteworthy for 16 species of Rafflesia, a genus of parasitic plants that derive energy and nutrients from the tissues of Tretrastigma vines. Among their number is the famous R. arnoldii, which produces the largest single flower (1 m in diameter) in the world. Most of the ecoregions have been severely reduced in extent, and total loss of the lowland forests is predicted if current exploitation rates continue. Sumatra is the last refuge for the critically endangered Sumatran orangutan, with less than 3500 individuals in the wild, and it also still contains small populations of the Sumatran rhino (D. sumatrensis), the Sumatran tiger, the elephant E. m. sumatrensis, and the false gharial. JAVA AND BALI

Java comprises an island arc of volcanoes that have coalesced to form a large single island, which has probably been consistently emergent only since the Early Pliocene (5 million years ago). It is smaller, more isolated, and less species-rich than are Borneo and Sumatra, with the moister evergreen forests of western Java supporting greater species richness than do the drier forests of eastern Java and Bali. These drier climates may have

facilitated the persistence of about 30 seasonal forest birds from mainland Asia (e.g., the green peafowl [Pavo muticus], the lineated barbet [Megalaima lineate]), which are presumed to have gone extinct on Sumatra and Borneo with the reestablishment of rainforest after the last glacial maximum. Java is one of the most densely populated islands in the world, and little of its original habitat remains. Each of the four ecoregions harbors critically endangered animals: the western Java rainforests, reduced to less than 5% of their original extent, have the last viable population (less than 40 to 50 individuals) of Javan Rhinoceros (R. sondaicus); the western Java montane rainforests (less than 20% of which remain) are home to the most endangered primates in Indonesia, the Javan leaf monkey (Presbytis comata) and the Javan gibbon (Hylobates moloch); and the eastern Java-Bali rainforest on the island of Bawean is home to an endangered deer (Axis kuhlii). Also of concern are the endangered Javan warty pig (Sus verrucosus) and several Javan subspecies of carnivore (e.g., the yellow-throated marten [Martes flavigula robinsoni] and the leopard Panthera pardus melas). The Javan and Balinese subspecies of tigers (Panthera tigris sondaica and P. t. balica) were last recorded on the islands in 1976 and in the late 1930s, respectively, and both are believed extinct. Several native bird species are equally imperiled, most famously the Bali starling (Leucopsar rothschildi), a critically endangered Bali endemic reduced to just six wild individuals by 2001 by trapping for the illegal cagebird trade. There have been no records of the critically endangered Javanese lapwing (Vanellus macropterus) since 1940, and the endangered Javan hawk eagle (Spizaetus bartelsi) is confined to the remaining patches of forest. KRAKATAU

The Krakatau Islands lie in the Sunda Strait between Java and Sumatra and are famous for the cataclysmic eruption of the Krakatau volcano on the main island (Krakatau or Rakata) in 1883. No life was left on the island, and patterns of recolonization of flora and fauna have been central tests of MacArthur and Wilson’s Equilibrium Theory of Island Biogeography. Colonization rates of resident land birds to 1933 seemed to suggest that the number of species on the island reflected a dynamic equilibrium between immigration and extinction processes, as MacArthur and Wilson proposed. However, more recent studies, based on longer time series, suggest that persistence probabilities for bird and butterfly spe-

cies are strongly influenced by successional changes in plant communities. MENTAWAI ISLANDS

The four Mentawai islands have been separated by an oceanic trench from the west coast of Sumatra for more than 0.5 million years. They support a distinct ecoregion characterized by no less than 17 endemic mammals, including four threatened primates: Mentawai gibbon (Hylobates klossii); Mentawai macaque (Macaca pagensis); Mentawai leaf-monkey (Presbytis potenziani); and the sole representative of an endemic genus, the snub-nosed monkey (Simias concolor). INDONESIAN BORNEO—KALIMANTAN

Kalimantan is the Indonesian region of Borneo and occupies the central and southern two-thirds of the island. Borneo is the third largest island in the world and supports the highest floral diversity in Malesia (more than 15,000 plant species, with 59 endemic genera). It is at the center of dipterocarp diversity with over 265 species, of which at least 155 are endemic, and it further boasts over 3000 tree species and 2000 orchids. Despite a large Sundaland element to the lowland fauna, endemism on Borneo is high for most groups. There are 39 (out of 350) endemic bird species and at least 44 (out of 222) mammals, including the endangered proboscis monkey, the Bornean orangutan (Pongo pygmaeus), and the Borneo pygmy elephant (Elephas maximus borneensis). There are in the region of 100 endemic amphibians, 47 lizards and 41 snakes, and the island is the only home for the monotypic Lanthanotidae, represented by the very rare Bornean earless monitor lizard (Lanthanotus borneensis). Since 1994, over 400 new species have been discovered on Borneo, and Borneo and Sumatran populations of the clouded leopard have been shown to have diverged from mainland Asian populations about 1.4 million years ago, elevating them to specific status (Neofelis diardii). The clouded leopard is the largest predator on Borneo, but in the absence of tigers and leopards, there are at least 26 species of small- to medium-sized carnivores. The highlands of Borneo are the most extensive, and are among the oldest (with a 20-milion-year history), in Indonesia, and a mix of Asian and Australian plant families has produced the most species-rich montane forests in the Indo-Pacific region, supporting some 150 mammals and 250 birds. In 2007, representatives of the three Bornean governments of Malaysia, Brunei Darussalam,

INDONESIA, BIOLOGY

451

and Indonesia signed the Declaration on the Heart of Borneo Initiative. This historic initiative is intended to conserve and sustainably manage approximately 220,000 km2 of rainforest, primarily encompassing the transboundary highlands of Indonesia and Malaysia.

neck turtle (Chelodina mccordi) represents the westernmost extent of the Chelidae, the dominant turtle family in Australia and New Guinea. It is critically endangered and limited to three populations on the tiny island of Roti where just 70 km2 of suitable habitat remain.

Wallacea Biodiversity Hotspot

SULAWESI

Wallacea comprises the large island of Sulawesi and thousands of smaller islands grouped into two archipelagoes, the Lesser Sundas (Nusa Tenggara) and the Moluccas (Maluku). Although the total species richness is generally less than in Sundaland, this is in part due to the much smaller total land area of the region. For most of the major taxonomic groups, richness is actually 1.5–3 times greater than would be expected for an equivalent area on Sundaland. The exception is the amphibians, for which the deep seas appear to have represented a major barrier to dispersal, and only 48 species have been described to date. Levels of endemism are high for all groups, ranging from 15% of vascular plants to 69% of amphibians (Table 2). Although the invertebrate fauna of Wallacea is generally poorly known, endemism is known to be high in the more conspicuous groups such as the huge bird-wing butterflies (Ornithoptera) (40 endemic species of 80 recorded) and the tiger beetles (Cicindelidae) (79 of 109). Endemism at the level of the island is also high, and it seems that the many opportunities for allopatric speciation provided by the thousands of islands have led to rapid radiations in some groups. The avifauna provides a particularly good illustration, with about 25% of the world’s accipiters (Accipitridae), 19% of pigeons and doves (Columbidae), 28% of kingfishers (Alcedinidae), and 37% of Megapodiidae. Other groups, such as the Phasianidae and Anseriformes, are greatly underrepresented.

Geologically, Sulawesi is a complex result of convergence of the Australian, Pacific, and Eurasian plates, bringing together five separate paleo-islands. Critical to the biogeography of the region are the timing and positioning of emergent fragments relative to the depth and extent of separating straits and oceans. Parts of southeastern Sulawesi were likely emergent about 20 million years ago, but with few other islands above water, connections to either Sundaland or Australia would have been limited. It was not until 10–5 million years ago that substantial land is known for Sulawesi, but at this point the Makassar Strait would have been fairly wide, limiting colonization from Borneo. This would have likely been the best chance (about 10 million years ago) for land animals and plants to cross Wallacea, as seas were relatively narrow prior to the opening of several deep ocean basins. These deep seas further prevented land connections during the low sea levels of the Pleistocene (with the exceptions of links to the Sula Islands and Tukang Besi Islands), although the Makassar Strait would have been narrower. Thus, opportunities for “easy” colonization of Sulawesi have been limited, and it provides a good example of an unbalanced endemic fauna, depauperate in species such as large herbivores and large carnivores, which have poor dispersal abilities. Although 62% of the mammals are endemic, there is only a single native carnivore, the endemic Sulawesi palm civet (Macrogalidia musschenbroekii), and the only large herbivores are the two anoa species and a deer (Cervus timorensis). Similarly, although primate diversity appears relatively high with 12 species, seven are endemic macaques with allopatric distributions, and the rest are tarsiers. Thus, at a single locality, there will typically be one species of macaque and one (or two) tarsier(s). The macaque distributions coincide with the distributions of genetically differentiated populations of toads (Bufo celebensis), grasshoppers (Chitaura), fanged frogs (Limnonectes), and flying lizards (Draco lineatus), suggesting that periodic historic barriers to dispersal across the island (e.g., oceanic inundation, large rivers, soil or vegetation shifts) have promoted speciation and created centers of genetic endemism.

LESSER SUNDA ISLANDS

The Lesser Sundas are an inner volcanic island arc and are unusual in Indonesia because a much drier and more seasonal climate has resulted in deciduous or “monsoon” forests, dominated by the legume Pterocarpus indicus. Forest subtypes range from dry thorn to moist deciduous, but human activity has degraded many areas to savanna. The driest islands of Komodo, Padar, Rinca, and Flores are home to the largest lizard in the world, the vulnerable Komodo dragon (Varanus komodoensis). Males reach 2.8 m in length and can weigh 50 kg. The only turtle found in Wallacea is also in the Lesser Sundas. McCord’s side-

452

INDONESIA, BIOLOGY

MOLUCCAS

Collectively the Moluccas are distinguished by their more even mix of Asian and Australian faunas and by their extraordinary degree of bird endemism, which is greater than anywhere else in Asia. There are 21 species endemic to the Banda Sea Islands moist deciduous forests (225 species), 16 in the Seram deciduous forests (213 species), ten in the Buru rainforests (173 species); the Halmahera rainforests include four endemic monotypic genera (Habroptila, Melitorgrais, Lycocorax, and Semioptera) among the 26 endemics (223 species). Many islands are small and uninhabitable but are critical breeding sites for seabirds such as frigatebirds (Fregatidae), tropicbirds (Phaethontidae), boobies (Sula), terns (Sterna), and smaller species. Typically, the mammal fauna is unbalanced and depauperate, but with both Asian and Australasian affinities, and each ecoregion has several endemic species. New Guinea Tropical Wilderness Area

New Guinea is the second largest island in the world and is one of the Earth’s three remaining major tropical wilderness areas—regions characterized by extent (greater than 10,000 km2), non-urban population density (fewer than 5 people per km2), and amount of intact habitat (greater than 70%) (Table 2). The eastern half of the island is an independent nation, Papua New Guinea, but two Indonesian provinces, West Papua and Papua (formerly Irian Jaya), cover the western half. Although much of the forest remains intact, it was also little explored until recent expeditions by Conservation International and the Indonesian Institute of Science uncovered dozens of new species, including 20 new frog species and a giant Mallomys rat weighing 1.4 kg. It is likely that all measures of diversity for the area are greatly underestimated. Levels of diversity and endemism are comparable to Sundaland and Wallacea (Table 1), but, bounded by Lydekker’s Line, the flora and fauna of New Guinea are predominantly Australian. Three of the four species of extant monotremes (all echidnas in the family Tachyglossidae) are present, and the genus Zaglossus is endemic to New Guinea. The mammalian fauna is otherwise dominated by marsupials, with rats and bats being the only native placental mammals. There are no living large carnivores, although there is fossil evidence of the thylacine, but 16 species of the small- to mid-sized carnivorous Dasyuridae (quolls, marsupial shrews, dasyures, dunnarts, and a planigale) are present. Of the 22 Macropodidae (kangaroos and wallabies), it is the ten endemic tree kangaroos (Dendrogalus) and six forest wallabies (Dorcopsis and Dorcopsulus) that distinguish New Guinea and the surrounding

islands, although there are also five species of the smallest macropods, the pademelons (Thylogale), mainly in eastern New Guinea (Papua New Guinea). Cuscus (Phalangeridae) evolved on New Guinea and abound with at least 13 species, and there are many small herbivorous species in the families Acrobatidae, Burramyidae, Petauridae, and Pseudocheiridae. Although the avifauna is largely Australian, there are a number of Asian elements, some of which do not extend into Australia, namely the tree swifts (Hemiprocnidae), shrikes (Laniidae), and sandpipers (Scolopacidae). Australian families exhibiting high diversity in the ancient forests of New Guinea include the famous birds of paradise (Paradisaeidae), bower birds (Ptilonorhynchidae), and honeyeaters (Meliphagidae). New Guinea is also home to all three species of large flightless cassowaries (Casuarius). Gondwanan floral elements include the conifers Podocarpus and the rainforest emergents Araucaria and Agathis, as well as tree ferns and several species of Eucalyptus. SEE ALSO THE FOLLOWING ARTICLES

Borneo / Indonesia, Geology / Island Biogeography, Theory of / Krakatau / Wallace’s Line FURTHER READING

Conservation International’s Biodiversity Hotspots website http://www .biodiversityhotspots.org/Pages/default.aspx MacKinnon, K., G. Hatta, H. Halim, and A. Mangalik. 1997. The ecology of Kalimantan, Indonesian Borneo. The Ecology of Indonesia Series, volume III. Hong Kong: Periplus Editions. Marshall, A. J., and B. M. Beehler. 2007. The ecology of Indonesian Papua: part one. The Ecology of Indonesia Series, volume VI. Hong Kong: Periplus Editions. Metcalf, I., J. M. B. Smith, M. Morwood, and L. D. Davidson, eds. 2001. Faunal and floral migrations and evolution in S E Asia–Australasia. Lisse, Netherlands: A. A. Balkema (Swets & Zeitlinger Publishers). Monk, K. A., Y. de Fretes, and G. Reksodiharjo-Lilley. 1997. The ecology of Nusa Tenggara and Maluku. The Ecology of Indonesia Series, volume V. Hong Kong: Periplus Editions. Olson, D. M., E. Dinerstein, E. D. Wikramanayake, N. D. Burgess, G. V. N. Powell, E. C. Underwood, J. A. D’amico, I. Itoua, H. E. Strand, J. C. Morrison, C. J. Loucks, T. F. Allnutt, T. H. Ricketts, Y. Kura, J. F. Lamoreux, W. W. Wettengel, P. Hedao, and K. R. Kassem. 2001. Terrestrial ecoregions of the world: a new map of life on Earth. BioScience 51: 933–938. http:// www.worldwildlife.org/science/ecoregions/terrestrial.cfm Sodhi, N., and B. W. Brook. 2006. Southeast Asian biodiversity in crisis. Cambridge: Cambridge University Press. Whitten, T. J., S. J. Damanik, J. Anwar, and N. Hisyan. 2000. The ecology of Sumatra. The Ecology of Indonesia Series, volume I. Hong Kong: Periplus Editions. Whitten, T. J., G. S. Henderson, and M. Mustafa. 2002. The ecology of Sulawesi. The Ecology of Indonesia Series, volume IV. Hong Kong: Periplus Editions. Whitten, T. J., R. E. Soeriaatmadja, and S. A. Affif. 1997. The ecology of Java and Bali. The Ecology of Indonesia Series, volume II. Hong Kong: Periplus Editions.

INDONESIA, BIOLOGY

453

INDONESIA, GEOLOGY ROBERT HALL Royal Holloway University of London, United Kingdom

Indonesia is a geologically complex region situated at the southeastern edge of the Eurasian continent. It is bordered by tectonically active zones characterized by intense seismicity and volcanism resulting from subduction. Western Indonesia is largely underlain by continental crust, but in eastern Indonesia there is more arc and ophiolitic crust, and several young ocean basins. The Indonesian archipelago formed over the past 300 million years by reassembly of fragments rifted from the Gondwana supercontinent that arrived at the Eurasian subduction margin. The present-day geology of Indonesia is broadly the result of Cenozoic subduction and collision at this margin. PRESENT-DAY TECTONIC SETTING

Indonesia is an immense archipelago of more than 18,000 islands extending over 5000 km from east to west between 95° and 141° E, and crossing the equator from 6° N to 11° S (Figs. 1 and 2). It is situated at the boundaries of three major plates: Eurasia, India-Australia, and Pacific-Philippine Sea (Fig. 1). In western Indonesia, the boundary between the

Eurasian and Indian plates is the Sunda Trench. Parallel to this in Sumatra is the right-lateral strike-slip Sumatran Fault, which results from the partitioning of oblique plate convergence into normal convergence at the trench and trench-parallel movement further north. Most active deformation in Sumatra occurs between the trench and the Sumatran fault. In contrast, east of Java, active deformation occurs within a complex suture zone up to 2000 km wide, including several small plates and multiple subduction zones; plate boundaries (Fig. 1) are trenches and another major strike-slip zone, the left-lateral Sorong Fault, which runs from New Guinea into Sulawesi. Global Positioning System (GPS) measurements indicate very high rates of relative motions, typically more than several centimeters per year, between tectonic blocks in Indonesia. Volcanism and Seismicity

The subduction zones are mainly well defined by seismicity extending to depths of about 600 km (Fig. 3) and by volcanoes (Fig. 1). There are at least 95 volcanoes in Indonesia that have erupted since 1500, and most are situated between 100 and 120 km above descending lithospheric slabs. Thirty-two have records of very large eruptions with a volcanic explosivity index (VEI) of greater than 4; 19 have erupted in the last 200 years, including Tambora in 1815 (VEI = 7) and Krakatau in 1883 (VEI = 6). Tambora, on the island of Sumbawa, is known for its impact

FIGURE 1 Geography of Indonesia and surrounding regions showing present-day tectonic boundaries and volcanic activity. Indonesia is shaded

green, and neighboring countries are shaded in pale gray. Bathymetric contours are at 200 m, 1000 m, 3000 m, 5000 m, and 6000 m. The location of the three most famous explosive eruptions known from Indonesia are shown in red text. Red arrows show plate convergence vectors for the Indian plate (IND-EUR) and the Philippine Sea plate (PSP-EUR) relative to Eurasia, and for the Australian plate relative to the Pacific plate (AUSPAC). There is little thrusting at the Timor trough. The Seram trough and Flores-Wetar thrusts are the sites of active thrusting.

454

INDONESIA, GEOLOGY

FIGURE 2 Digital elevation model showing topography and bathymetry of the Indonesian region. Compare to Fig. 1 for tectonic and geographic features.

on global climate, and its 1815 eruption resulted in the Northern Hemisphere’s 1816 “year without a summer,” when crops failed, causing famine and population movements. The eruption of Toba on Sumatra 74,000 years ago was even bigger (estimated VEI of 8) and is the largest eruption known on Earth in the last 2 million years. Sundaland

The interior of Indonesia (Fig. 2), particularly the Java Sea, Sunda Shelf, and surrounding emergent, but topo-

graphically low, areas of Sumatra and Kalimantan (Indonesian Borneo), is largely free of seismicity and volcanism (Figs. 1 and 3). This tectonically quiet region forms part of the continental core of the region known as Sundaland (Fig. 4). Sundaland extends north to the Thai-Malay Peninsula and Indochina, and formed an exposed landmass during the Pleistocene. Most of the Sunda Shelf is shallow, with water depths considerably less than 200 m (Fig. 2), and its lack of relief has led to the misconception that it is a

FIGURE 3 Seismicity in the Indonesian region between 1964 and 2000. Bathymetric contours are shown at 200 m and 6000 m.

INDONESIA, GEOLOGY

455

FIGURE 4 Growth of the Indonesian region. Collision between the Sibumasu and East Malaya-Indochina blocks occurred in the Triassic. Additional

crust has been added to this Sundaland core, largely by later collisions of continental blocks. The present-day zone of active deformation is shaded yellow. Gray areas within this complex plate-boundary zone are areas underlain by Cenozoic ocean crust.

stable area. Sundaland is often described as a shield or craton, but geological observations, heat flow, and seismic tomography show that this is not the case. There has been significant deformation during the Cenozoic with the formation of deep sedimentary basins and the localized but widespread elevation of mountains. Unlike well-known shields or cratons (e.g., Baltic or Canadian), Sundaland is not underlain by a thick, cold lithosphere formed in the Precambrian. Its interior has high surface heat flow values, typically greater than 80 mW/m2. At the Indonesian margins, high heat flows reflect subduction-related magmatism, but the hot interior of Sundaland appears to be the consequence of high upper-crustal heat flow from radiogenic granites and their erosional products, the insulation effects of sediments, and a high mantle heat flow. P- and S-wave seismic tomography show that Sundaland is an area of low velocities in the lithosphere and underlying asthenosphere, in contrast to the colder and thicker Indian and Australian continental lithosphere to the northwest and southeast. Such low mantle velocities are commonly interpreted in terms of elevated temperature, and this is consistent with regional high heat flow, but they may also partly reflect a different composition or elevated volatile contents. Consequences of Long Subduction History

The upper mantle velocities and heat flow observations suggest the region is underlain by a thin and weak lithosphere. In contrast, in the lower mantle beneath Indo-

456

INDONESIA, GEOLOGY

nesia, there is a high velocity anomaly, suggesting the accumulation of subducted lithosphere. These features are the consequence of long-term subduction at the Indonesian margins. The active margins are the site of magmatism, heating, and weakening, but the region of weak lithosphere extends many hundreds of kilometers from the volcanic margins. The character of the lithosphere has been a major influence on the development of Indonesia, combined with repeated collisions at the subduction margins that have led to continental growth. PRE-CENOZOIC HISTORY OF INDONESIA

Western Indonesia, notably the islands of Sumatra and Borneo, contains most of the oldest rocks in Indonesia. Sumatra: Basement

Sumatra represents the geological continuation of the Malay Peninsula and contains the most extensive outcrops of Paleozoic and Mesozoic rocks. The oldest rocks at the surface are Carboniferous sediments, but possible Devonian rocks have been reported from petroleum boreholes in the Malacca Straits, and granites from boreholes in central Sumatra have been dated as Silurian. Xenoliths in dykes, granite clasts in sediments, and various highgrade metamorphic rocks from different parts of Sumatra suggest a pre-Carboniferous crystalline basement similar to that beneath the Malay Peninsula, which is Proterozoic at depth.

Sumatra: Cathaysian and Gondwana Blocks

In western Sumatra there are Paleozoic sediments that range in age from Carboniferous to Triassic, and Permian volcanic rocks with Cathaysian affinities. These are interpreted to form part of an Indochina–East Malaya block (Fig. 4) that separated from Gondwana in the Devonian and by the Carboniferous was in warm tropical low latitudes where a distinctive flora developed. In contrast, in eastern Sumatra, Carboniferous sediments include pebbly mudstones interpreted as diamictites that formed in a glacio-marine setting. These indicate cool Gondwana affinities. The Carboniferous rocks and associated Permian and Triassic sediments belong to the Sinoburmalaya or Sibumasu block (Fig. 5), which was at high southern latitudes during the Carboniferous, separated from Gondwana in the Permian, and collided with the Indochina–East Malaya block, already amalgamated with the South and North China blocks, in the Triassic. Sumatra: Triassic Collision

The collision of the Sibumasu and Indochina–East Malaya blocks was the first stage in the geological development of Indonesia. The widespread Permian and Triassic granites of the Thai-Malay tin belt extend into western Indonesia and are the products of subduction and post-collisional magmatism associated with this event.

plutons, associated with volcanic rocks, intrude the metamorphic rocks in the Schwaner Mountains of southwestern Borneo. To the north, the northwestern Kalimantan domain, or Kuching zone, includes fossiliferous Carboniferous limestones, Permo-Triassic granites, Triassic marine shales, ammonite-bearing Jurassic sediments, and Cretaceous melanges. In Sarawak, Triassic floras suggest Cathaysian affinities and correlations with Indochina. The Kuching zone may mark a subduction margin continuing south from East Asia, at which ophiolitic, island arc, and microcontinental crustal fragments collided and were deformed during the Mesozoic. Sumatra, Java, Kalimantan, Sulawesi: Cretaceous Active Margin

A Cretaceous active margin is interpreted to have run the length of Sumatra into western Java and then continued northeast through southeastern Borneo and into western Sulawesi, as suggested by the distribution of Cretaceous high pressure–low temperature subduction-related metamorphic rocks in central Java, the Meratus Mountains of southeastern Kalimantan and western Sulawesi. Western Sulawesi and eastern Java (Fig. 4) are underlain in part by Archean continental crust, and geochemistry and zircon dating indicates derivation of this crust from the west Australian margin. Subduction ceased in the Late Cretaceous following collision of this block with Sundaland.

Sumatra: Mesozoic

The Mesozoic sedimentary record is very limited but suggests that much of Sundaland, including most of its Indonesian margin, was emergent. During the Mesozoic, there is interpreted to have been reorganization of Sumatran crustal blocks, possibly by strike-slip faulting at an active margin. Isotopic dating in Sumatra indicates that there were several episodes of granite magmatism, interpreted to have occurred at an Andean-type margin, during the Jurassic and Cretaceous. Marine sedimentary rocks were deposited in an intra-oceanic arc that collided with the Sumatran margin in the Middle Cretaceous. The collision added arc and ophiolitic rocks to the southern margin of Sumatra. Borneo: Mesozoic Collisions

Southwestern Borneo (Fig. 4) may be the eastern part of Triassic Sundaland, or it could be a continental block added in the Early Cretaceous, at a suture that runs south from the Natuna Islands. The Paleozoic is represented mainly by Carboniferous to Permian metamorphic rocks, although Devonian limestones have been found as river boulders in eastern Kalimantan. Cretaceous granitoid

Sundaland: Cretaceous Granites

Cretaceous granites are widespread in the Schwaner Mountains, in western Sarawak, on the Sunda Shelf, and on the Thai-Malay Peninsula. They have been interpreted as the product of Andean-type magmatism at active margins but are spread over a large area and are far from any likely subduction zones. They probably represent postcollisional magmatism following Cretaceous addition of continental fragments in Borneo, eastern Java, and western Sulawesi. CENOZOIC HISTORY OF INDONESIA

Cenozoic rocks cover most of Indonesia. They were deposited in sedimentary basins in and around Sundaland. There are products of volcanic activity at subduction margins, and there are ophiolites, arc rocks, and Australian continental crust added during collision. Little is known of the Late Cretaceous and Paleocene because of the paucity of rocks of this age. From Sumatra to Sulawesi, the southern part of Sundaland was probably mostly emergent during the Late Cretaceous and Early Cenozoic, and there was widespread erosion; the

INDONESIA, GEOLOGY

457

oldest Cenozoic rocks typically rest unconformably on Cretaceous or older rocks. There is little evidence of subduction, although there was minor volcanic activity in southern Sumatra and Sulawesi. At the beginning of the Cenozoic, there were probably passive margins around most of Indonesia. Eocene Subduction Initiation

India moved north during the Cretaceous to collide with Asia in the Early Cenozoic but passed to the west of Sumatra. Australia began to move rapidly northward from about 45 million years ago, in the Eocene. At this time, northward subduction resumed beneath Indonesia, producing widespread volcanic rocks at the active margin. The Sunda arc stretched from Sumatra, through Java and the north arm of Sulawesi, and then continued into the western Pacific. From the Eocene to the Early Miocene, the Halmahera arc was active in the western Pacific, far north of Australia, above a north-dipping subduction zone. Also in the Eocene, southward subduction of the proto–South China Sea began on the northern side of Sundaland. Sediment carried north from southwestern Borneo was deposited in deep marine fans at this active margin. Early Cenozoic volcanic activity in Kalimantan is not well dated or characterized but appears to be related to this subduction. Eocene Rifting

In the interior of Sundaland, widespread rifting began at a similar time as subduction and led to the formation of numerous sedimentary basins. These basins, some more than 10-km deep, are filled with Cenozoic sediments and are rich in hydrocarbons. The largest of these are in Sumatra, offshore Java, and eastern Kalimantan. In southeastern Sundaland, Eocene rifting led to the separation of Borneo and western Sulawesi, forming the Makassar Straits (Fig. 2), and by the Oligocene, much of eastern Kalimantan and the straits was an extensive area of deep water. In western Sulawesi, shallow water deposition continued, and there are extensive platform limestones in the south arm. The southern part of the straits is relatively shallow (about 1 km) and underlain by continental crust. The northern straits are connected to the oceanic Celebes Sea, but it is not known whether they are underlain by oceanic or stretched continental crust because there is up to 14 km of sediment below the 2.5-km-deep sea floor. The Makassar Straits are today a major passageway for water and heat from the Pacific to the Indian Ocean and have been an important influence on biogeography. The Wallace Line,

458

INDONESIA, GEOLOGY

marking an important boundary between Asian and Australasian faunas, follows the Makassar Straits south to pass between the islands of Bali and Lombok. Miocene: Continental Collisions

At the beginning of the Miocene, ophiolites were emplaced by collision between the Australian and the Sundaland continents in Sulawesi, and between the Australian continent and the Halmahera arc much further to the east in the Pacific. The ophiolites of Sulawesi are remnants of fore-arc and oceanic crust between Sundaland and the Australian plate, whereas those of the North Moluccas are fragments of Philippine Sea plate arcs. Later in the Early Miocene, there was collision in north Borneo with the extended passive continental margin of South China. These collisions led to mountain building in eastern Sulawesi and in Borneo. The first Australian continental crust was added in eastern Indonesia (Fig. 5), but as northward movement of Australia continued, there was a change in eastern Indonesia to extension, complicated by minor collisions as microcontinental blocks moved along strike-slip faults. In the Sunda arc, volcanic activity declined in Java for a few million years before a late Miocene increase from Sumatra to the Banda arc. Neogene: Java-Sumatra

In the Java–Sulawesi sector of the Sunda arc, volcanism greatly diminished during the Early and Middle Miocene, although northward subduction continued. The decline in magmatism resulted from Australian collision in eastern Indonesia, causing rotation of Borneo and Java. Northward movement of the subduction zone prevented replenishment of the upper mantle source until rotation ceased in the late Middle Miocene. Then, about 10 million years ago, volcanic activity resumed in abundance along the Sunda arc from Java eastward. Since the Late Miocene, there has been thrusting and contractional deformation in Sumatra and Java related to arrival of buoyant features at the trench, or increased coupling between the overriding and downgoing plates. Both islands have been elevated above sea level in the last few million years. Neogene: Borneo

The rise of mountains on Borneo increased the output of sediment to circum-Borneo sedimentary basins. In eastern Kalimantan, thick Miocene to recent sediments filled accommodation space created during Eocene rifting. Most was derived from erosion of the Borneo highlands and inversion of older parts of the basin margins to the north and west, which began in the Early Miocene.

Sedimentation continues today in the Mahakam delta and in the offshore deepwater Makassar Straits. In parts of central Kalimantan, there was some Miocene magmatism and associated gold mineralization, but volcanic activity largely ceased in Kalimantan after collision. Minor Plio-Pleistocene basaltic magmatism in Borneo may reflect a deep cause such as lithospheric delamination after Miocene collisional thickening. Neogene: Sulawesi

Sulawesi is inadequately understood and has a complex history still to be unraveled. In eastern Sulawesi, collision initially resulted in thrusting of ophiolitic and Australian continental rocks. However, contractional deformation was followed in the Middle Miocene by new extension. There was Miocene core complex metamorphism in north Sulawesi, extensional magmatism in south Sulawesi, and formation of the deep Gorontalo Bay and Bone Gulf basins between the arms of Sulawesi. Compressional deformation began in the Pliocene, partly as result of the collision of the Banggai-Sula microcontinent in east Sulawesi, which caused contraction and uplift. Geological mapping, paleomagnetic investigations, and GPS observations indicate complex Neogene deformation in Sulawesi, including extension, block rotations, and strike-slip faulting. There are rapidly exhumed upper mantle and lower crustal rocks, and young granites, near to the prominent Palu-Koro strike-slip fault (Fig. 1). During the Pliocene, coarse clastic sedimentation predominated across most of Sulawesi as mountains rose. The western Sulawesi fold-thrust belt has now propagated west into the Makassar Straits. At present, there is southward subduction of the Celebes Sea beneath the north arm of Sulawesi and subduction on the east side of the north arm of the Molucca Sea toward the west (Fig. 1). Neogene: Banda

The Banda arc is the horseshoe-shaped arc that extends from Flores to Buru, including Timor and Seram, with islands forming an outer non-volcanic arc and an inner volcanic arc. It is an unusual region of young extension that developed within the Australian-Sundaland collision zone and formed by subduction of an oceanic embayment within the northward-moving Australian plate. In the Middle Miocene, Jurassic ocean lithosphere of the Banda embayment began to subduct at the Java Trench. The subduction hinge rolled back rapidly to the south and east, inducing massive extension in the upper plate. Extension began in Sulawesi in the Middle Miocene. As the hinge rolled back into the Banda embayment,

it led to formation of the Neogene Banda volcanic arc and the opening of the North Banda Sea, the Flores Sea, and later the South Banda Sea. About 3–4 million years ago, the volcanic arc collided with the Australian margin in Timor, causing thrusting. Remnants of the Asian margin and Paleogene Sunda arc are found in the uppermost nappes of Timor and other Banda islands. After collision, convergence and volcanic activity ceased in the Timor sector, although volcanic activity continued to the west and east. New plate boundaries developed north of the arc between Flores and Wetar (Fig. 1), and to the north of the South Banda Sea, associated with subduction polarity reversal. The Banda region is now contracting. During the last 3 million years, there have been significant shortening and probable intra-continental subduction at the southern margins of the Bird’s Head microcontinent south of the Seram trough (Fig. 1). Neogene: North Moluccas

In eastern Indonesia, the Halmahera and Sangihe arcs (Fig. 1) are the only arcs on Earth currently colliding. Both of the currently active volcanic arcs formed during the Neogene. The Sangihe arc is constructed on Eocene oceanic crust and initially formed at the Sundaland margin in the Early Cenozoic. The modern Halmahera arc is built upon older arcs, of which the oldest is a Mesozoic intra-oceanic arc that formed in the Pacific. Early Miocene arc–Australian continent collision terminated northward subduction, and the north Australian plate boundary became a major left-lateral strike-slip zone in New Guinea. Volcanism ceased, and there was widespread deposition of shallow marine limestones. Arc terranes were moved westward within the Sorong fault zone. At the western end of the fault system, there was subduction beneath the Sangihe arc and collision in Sulawesi of continental fragments sliced from New Guinea. Initiation of east-directed Halmahera subduction probably resulted from locking of strands of the left-lateral Sorong fault zone at its western end in Sulawesi. The present-day Molucca Sea double subduction system was initiated in the Middle Miocene, and volcanism began in the Halmahera arc about 11 million years ago. The Molucca Sea has since been eliminated by subduction at both its eastern and western sides. The two arcs first came into contact about 3 million years ago, and this contact was followed by repeated thrusting of the Halmahera fore-arc and back-arc toward the active volcanic arc. Collision has formed a central Molucca Sea (Fig. 1) melange wedge, including ophiolite slices from the basement of the Sangihe arc. There are small fragments of continental crust between splays of the Sorong fault.

INDONESIA, GEOLOGY

459

Neogene: New Guinea

In New Guinea, there was rifting from the Late Triassic onward to form a Mesozoic northern passive margin of the Australian continent, on which there was widespread carbonate deposition during the Cenozoic. To the north of the passive margin were a number of small oceanic basins and arcs developed above subduction zones; the region was probably as complex as the western Pacific today. At the beginning of the Miocene, the arc–Australian continent collision began emplacement of arc and ophiolite terranes, which then moved west in a complex strike-slip zone. The Halmahera arc was one of these. Today, northern New Guinea is underlain by these arc and ophiolitic rocks, fragmented by faulting. In the Pliocene, subduction probably began at the New Guinea Trench (Fig. 1), as there is now a poorly defined slab dipping south that has reached depths of about 150 km. There was isolated, but important, magmatism associated with world-class copper and gold mineralization including the Grasberg and Ertsberg complexes. The rise of the New Guinea main ranges accelerated, and mountains reached their present elevations with peaks more than 5-km high (Fig. 2) capped by glaciers. SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Eruptions: Laki and Tambora / Indonesia, Biology / Island Arcs / New Guinea, Geology / Philippines, Geology FURTHER READING

Barber, A. J., M. J. Crow, and J. S. Milsom, eds. 2005. Sumatra: geology, resources and tectonic evolution. Geological Society London Memoir 31. Bijwaard, H., W. Spakman, and E. R. Engdahl. 1998. Closing the gap between regional and global travel time tomography. Journal of Geophysical Research 103: 30,055–30,078. Bock, Y., L. Prawirodirdjo, J. F. Genrich, C. W. Stevens, R. McCaffrey, C. Subarya, S. S. O. Puntodewo, and E. Calais. 2003. Crustal motion in Indonesia from global positioning system measurements. Journal of Geophysical Research 108: doi:10.1029/2001JB000324. Hall, R. 2002. Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions, model and animations. Journal of Asian Earth Sciences 20: 353–434. Hall, R., and D. J. Blundell, eds. 1996. Tectonic evolution of SE Asia. Geological Society of London Special Publication 106. Hall, R., and C. K. Morley. 2004. Sundaland basins, in Continent-ocean interactions within the East Asian marginal seas. P. Clift, P. Wang, W. Kuhnt, and D. E. Hayes, eds. American Geophysical Union, Geophysical Monograph 149, 55–85. Hamilton, W. 1979. Tectonics of the Indonesian region. US Geological Survey Professional Paper 1078. Hutchison, C. S. 1989. Geological evolution of South-East Asia. Oxford Monographs on Geology and Geophysics, Oxford, UK: Clarendon Press. Metcalfe, I. 1996. Pre-Cretaceous evolution of SE Asian terranes, in Tectonic evolution of SE Asia. R. Hall, R. Blundell, and D. J. Blundell, eds. Geological Society of London Special Publication 106, 97–122. van Bemmelen, R.W. 1949. The geology of Indonesia. The Hague, Netherlands: Government Printing Office, Nijhoff.

460

I N S E C T R A D I AT I O N S

INSECT RADIATIONS DIANA M. PERCY University of British Columbia, Vancouver, Canada

Insect radiations on islands are the evolutionary product of diversification within an insect lineage on an island or a series of islands forming an archipelago. Radiations, by definition, represent in situ diversification and are often characterized by evolutionarily novel adaptations. Island radiations are one of the most important natural phenomena for evolutionary biologists, and, like other animal and plant groups, insects on islands have undergone radiations that range from a modest diversification of species to explosive radiations over a short period of time. Insect radiations vary not only in the numbers of species but in the rates of speciation and in the diversity of adaptive traits that evolve. COLONIZATION, ESTABLISHMENT, AND DIVERSIFICATION

Island radiations are by definition the product of in situ diversification and the evolution of multiple species from a founding ancestor, producing a lineage of closely related island endemics. All islands represent habitable patches surrounded by uninhabitable environments. These can be terrestrial habitat “islands” such as shifting sand dunes in Namibia, where a lineage of Scarabaeus dung beetles has radiated into 12 endemic species, or “sky islands” in the Rocky Mountains, where Melanoplus grasshoppers have radiated into 37 species after isolation in multiple glacial refugia. More commonly, islands represent terrestrial habitats surrounded by water. Insect radiations on islands necessarily begin with a stepwise process involving dispersal from a source population, establishment of a viable population upon colonization, and diversification into multiple species within an island or between islands in an archipelago. In many cases, the distances between source populations on continental land masses and remote oceanic islands (islands that are formed de novo and are not terrestrial fragments separated from a once-larger land mass) are great enough that active dispersal by insects (e.g., by flying) is ruled out and passive dispersal methods (e.g., by wind currents, attachment to migrating birds, or rafting with tidal debris) are considered more plausible means for insects to have traveled the distance. Four endemic flightless insect genera, including large flightless beetles, cave crickets,

and cockroaches, have dispersed from New Zealand to the Chatham Islands (~800 km) relatively recently (2–6 million years ago), precluding isolation by vicariance. The considerable impediments to such long-distance dispersal events are one of the primary filters that limit which and how many of the taxa in any particular source area will be found on an island. In addition to such dispersal challenges, for many insects the establishment of a viable population in the novel island environment is an equally formidable filter, particularly for habitat specialists (e.g., monophagous herbivores). These barriers to successful island colonization are the reason why islands are typically lineage-poor, with limited overall representation of the continental biota. Conversely, in situ species radiations can result in unusual species richness within select lineages that manage to establish successfully. Founding species may be at an advantage where colonization barriers have reduced the number of competitors, predators, and parasites; this advantage may promote speciation and radiation in successful colonists. Radiations that are thought to have been initiated under such conditions fall under the “escape and radiate” hypothesis. Conversely, competition between newly formed taxa within a lineage may promote speciation by leading to character displacement that reinforces reproductive isolation in sympatric island taxa (e.g., Aphanarthrum bark beetles, Drosophila, and Laupala crickets). The more remote and isolated islands are from a source biota, the more asymmetrical or disharmonic the biotas are likely to be. For instance, the Kurile archipelago is an extensive chain of volcanic continental rim islands that are only a few hundred kilometers from continental Asia. These islands have a broad representation of the continental biota and little endemism (90% in many groups), but the Hawaiian biota has a highly asymmetric representation of its source biotas. On islands where species lineages have radiated and there are numerous multiplications of species from a few ancestral colonists, levels of endemism dramatically increase. Theoretically, any number of species greater than two is eligible to be considered a radiation, but in practice “radiation” usually refers to higher numbers of derived species. Radiations may be relatively modest in species numbers but notable for their adaptive shifts. The Galápagos finches, which attracted the attention of Charles Darwin, have only 14 species but are considered a classic example of adaptive radiation on the basis of ecologically driven morphological changes between the different spe-

cies. Insect radiations take many forms, from impressive explosions in the number of species (a primary example being the Hawaiian drosophilid flies, with approximately 1000 extant species derived from a single colonization event) to radiations that are more modest in species number but exhibit dramatic adaptive shifts. A relatively modest radiation of Caconemobius ground crickets in the Hawaiian Islands has only nine endemic species, but there have been substantial adaptive shifts in this lineage from inhabiting rocky coastal areas to inhabiting subterranean lava tubes. The derived subterranean species exhibit adaptive morphological changes associated with the shift in habitat, including eye reduction and loss of pigmentation. Many island radiations combine both the evolution of different adaptive traits and diversification into large numbers of species. There are also nonadaptive radiations, which refer to diversification occurring within a narrow or similar ecological range, such as speciation by specialist herbivores on the same host plant but on different islands (i.e., geographic rather than ecological separation) or nonspecialist, dietary generalists speciating within an island (e.g., Rhyncogonus weevils on the island of Rapa). Diversification in the absence of adaptive shifts may be precipitated by other conditions such as escape from a restraint that served to limit diversification on continents (e.g., predators and competitors), resulting in the “escape and radiate” phenomenon already mentioned. Interpretations of radiation processes are necessarily influenced by the number of species locally extant at a given time, which is determined by a number of factors, including the age of the lineage and rates of extinction. On islands, lineages may have undergone relatively minimal extinction compared to continents, because islands often have stable, temperate climates compared to continents, and in addition, in the case of recent radiations on young islands, there has been less time to the present for extinctions to have occurred. Where much of the diversity in recent radiations survives, evolutionary biologists are presented with some of the clearest examples of speciation processes. In contrast, the process by which an insect lineage dispersed and first established on an island will necessarily be more speculative, because these events, and the ancestral forms and habits of those initial colonists, are frequently no longer apparent from the derived species’ morphology and preferred island habitats observed today. In some cases, such as the Hawaiian tree and swordtail crickets, the extant species have diverged from their original founder lineage to such a degree that island groups such as these are often taxonomically misplaced by specialists.

I N S E C T R A D I AT I O N S

461

CONSTRAINTS AND PROMOTERS: ISLANDS AND ARCHIPELAGOES

Insect radiations on islands are fairly common, but they are not ubiquitous. Not every insect lineage that successfully colonizes an island or archipelago undergoes a radiation. Finding explanations for what drives or constrains radiations is one of the most challenging aspects of island biology. Potential explanations include morphological or genetic constraints on diversification, and available ecological niches favoring one group over another. An example of a radiation occurring in one insect group but not in a related group can be found in the Canary Islands, which have been colonized by several related lineages of specialist herbivores (psyllids). Diversification in the different psyllid lineages appears to have been limited or promoted by the presence and abundance of familiar host plant species. In contrast, in several Hawaiian herbivore groups, host range has dramatically expanded after colonization. In the case of Nesosydne planthoppers, a Hawaiian lineage with more than 80 species derived from continental ancestors typically specialized on monocotyledonous plants, colonization of the Hawaiian Islands has resulted in diversification on many novel (i.e., previously unused) host plant groups, mostly in dicotyledonous plant families. Similarly, the Hawaiian endemic plant bug genus Sarona, with at least 40 species, also has a greatly expanded host plant range (even though individual species remain specialists) as compared to continental sister groups, including several novel host plant families. Because the Canary Islands are less isolated from a continental source than the Hawaiian Islands, the presence of multiple colonizing lineages in the Canary Islands may serve to increase competition and maintain ancestral host preferences. In these herbivorous insect groups, radiation is in large part driven by ecological adaptations to feeding on particular host plants. Macromorphological changes are typically not required in shifting to different hosts, and there is little morphological variation between these species. In contrast, Hawaiian drosophilid flies, in addition to marked ecological habitat specialization, exhibit considerable morphological and behavioral diversity, reflecting the development of divergent secondary sexual characteristics correlated with male mating behavior. In this case, separating out the roles of different processes (e.g., ecological, behavioral) becomes more complex, in particular determining which processes act as a primary promoter rather than secondarily reinforcing speciation. An important factor in determining the scale of insect radiations is the type of island colonized. Islands come in many forms, from large single- or two-island groups such 462

I N S E C T R A D I AT I O N S

as Madagascar, New Caledonia, and New Zealand, which share some geological history with other land masses, and are markedly old compared to islands formed de novo from volcanic activity; to small isolated singleton islands of oceanic origin such as the Atlantic island of St. Helena; through to chains of islands sharing a common volcanic hotspot origin such as the Hawaiian Islands. Many of the most impressive examples of species radiations are to be found in these hotspot archipelagoes, such as the Pacific Hawaiian and Galápagos Islands, and the Atlantic Canary Islands. However, a number of insect lineages have also radiated on isolated singleton islands that have no close neighboring islands. Single-island radiations have been less well studied. For instance, there are more than 70 endemic curculionid weevils on the island of St. Helena (an island that is more than 1900 km from the nearest continent and 1100 km from Ascension Island), but there has been no comprehensive systematic study undertaken to assess the number of independent colonizations and therefore the number of individual lineages represented by this diversity. Similarly, on the isolated island of Rapa (part of the Austral Islands in southern French Polynesia, but the most remote of these islands), the weevil genus Miocalles is represented by more than 60 species on this small island that is only 40 km2. As with the St. Helena weevils, the number of independent colonizations has not been established. In New Zealand, at least two radiations of jumping plant lice (comprising more than 80 species) are associated with two highly diverse host plant genera in the Asteraceae. A radiation of Helictopleurini dung beetles in Madagascar is thought to have been promoted by a parallel radiation in lemurs, whose dung the beetles specialize on. These and other insect lineages that diversify within an island may do so by ecological shifts in sympatry or micro-allopatric shifts. For instance, at least 50% of the speciation events in the Hawaiian plant bug genus Sarona are estimated to have occurred in sympatry within islands via host plant shifts. In contrast, the monophyletic lineage of Rhyncogonus weevils on Rapa are not habitat specialists and therefore may have diversified via small-scale geographic isolation (micro-allopatry) (Fig. 1). Other Hawaiian herbivorous groups, such as the endemic Hawaiian seed bug genus Neseis, have speciated primarily by colonizing new islands, often without involving a host plant change. In the case of the Neseis seed bugs, an archipelago with multiple islands as geographic isolates has been a necessary promoter for speciation and radiation. Arguably, it is chains of islands that have produced the most dramatic radiations of insects. For instance, both drosophilid flies and crickets in the Hawaiian Islands have

FIGURE 1 Rhyncogonus gracilis, endemic to the island of Rapa, has

radiated in the Austral Islands; this species is not a habitat specialist. Photograph by Ronald Englund.

lineages that have radiated into more than 100 species, and the Hawaiian drosophilid radiation is unchallenged for species diversity, with approximately 1000 extant species arising from a single founder event, including around 25% of the world’s species of Drosophila. In terms of rapidity of radiation, the Hawaiian Laupala cricket radiation is proposed to have the fastest speciation rate in an insect radiation (4.17 species per million years). Among animals this rate is exceeded only by speciation rates in African cichlid fishes. The Hawaiian Laupala crickets are ecologically and morphologically cryptic; they are all forest dwellers with generalist diets, and the primary differences between species are their acoustic mating signals, suggesting that this radiation is driven by sexual selection rather than ecologically adaptive shifts. One of the signature patterns of archipelago radiations emphasizes the role of multiple island chains. In both Hawaiian crickets and flies, as with many archipelago radiations, there is a pervasive pattern of single-island endemics. However, as with the herbivorous bug groups referred to earlier, there are contrasting modes of speciation. Among the flies, diversification is primarily via inter-island colonization, such that sister taxa typically occur on different islands, whereas among the crickets, diversity is primarily a product of secondary radiations within islands, such that sister taxa are likely to be found on the same island. Island chains may be particularly productive in terms of species radiations for several reasons. The first is that the additional land area of multiple islands increases both the quantity and usually also the diversity of habitats, particularly if volcanic activity on some islands dramatically increases the elevation available for colonization. Pacific blackfly diversity is greatly increased on younger islands such as Tahiti in French Polynesia, where there are 29 species, because the newly uplifted

topography creates the habitats necessary (steep slopes) for a radiation of waterfall-inhabiting species. Secondly, and perhaps more importantly, each island and its habitats are discretely circumscribed and isolated from other islands by the intervening marine environment. Barriers to gene flow between islands are therefore both persistent (barring dramatic changes in sea level) and repeated with the colonization of each island. This important feature of island chains clearly serves to promote radiations, because each individual island potentially becomes a relay point for a diversifying lineage, and each new island colonized can serve to continue and expand a radiation as well as serving as a refugium if a lineage suffers extinction elsewhere in the archipelago. This is most clearly seen in hotspot archipelagoes, where the successively formed islands are frequently in relatively close proximity to the preceding and following islands and can act as steppingstones for a radiating lineage. The isolation and uniqueness of individual island lineages is promoted by the frequent loss of dispersal capabilities in island taxa (e.g., marked reduction in wing size and flightlessness). Loss of dispersal capability can dramatically limit inter-island movement, virtually ensuring that populations on different islands become rapidly genetically isolated. The partial or complete loss of forewings in Hawaiian crickets is a classic example. The resulting genealogical patterns, through a combination of intra-island diversification and single-island endemism, are monophyletic island lineages. Each individual island lineage may be characterized by novel adaptations determined by the environmental character of that island. Islands in an archipelago that have similar environmental variables (e.g., altitude, climate, geomorphology, soil, and vegetation zones) may produce independent lineages where species evolve convergent adaptive traits on each island. The species in these lineages may look morphologically and/or ecologically similar but have evolved independently on each island. Oliarus planthoppers, with more than 70 species in the Hawaiian Islands, have independently colonized subterranean lava tubes on different islands, but these cave-dwelling species are remarkably morphologically convergent as a result of parallel adaptations to the cavernous habitat. Diversity in a radiation is therefore promoted by both evolution within islands and dispersal between islands. Novel species arising in isolation on one island can “seed” this diversity “forward” onto other, more recently formed islands or “backward” to older islands. In these situations, the presence or absence of genetic or reproductive incompatibility between species that evolved in isolation on different islands will

I N S E C T R A D I AT I O N S

463

determine to what extent, if any, introgression between different islands will occur after species formation. Furthermore, the characteristics and extent of any island radiation is influenced by the geological history of the island(s). In an archipelago for instance, whether ancient radiations are able to maintain a continuum by populating new islands as they emerge will depend on whether the replacement rate of old submerged islands by new emergent islands is sufficiently regular in space and time. Molecular dating methods have revealed several insect lineages to be considerably older than the age of the islands inhabited by extant taxa (e.g., Hawaiian drosophilid flies, Galápagos flightless weevils). The likely explanation in these cases is that diversification of the lineage began earlier and outside the current geographic range, and in most cases it is proposed that this earlier diversification took place on older islands now submerged under the sea. Studying insect radiations in archipelagoes with a progressive age range of islands (e.g., the Hawaiian Islands with several islands ranging in age from ~0.5–~5 million years) can reveal how an insect lineage responds to increasing island age and the formation of successively newer islands.

FIGURE 2 This hypothetical phylogeny of island colonization and

radiation illustrates how phylogenetic analyses can help interpret patterns of radiation. Illustrated is the colonization and radiation of a hotspot archipelago by a lineage dispersing from a continent. The data depicted here are clocklike, and therefore, shorter branches on the youngest islands indicate more recent colonization and speciation events. The overall branching pattern of individual island lineages within the phylogeny follows the “rule of progression” with successional colonization from older to younger islands. Also illustrated are secondary within-island radiations leading to monophyletic island lineages and species that are single-island endemics. Larger islands with

464

INTERPRETING ISLAND RADIATIONS: THE RULE OF PROGRESSION

more habitat types result in more intra-island speciation events, but

Much of our current understanding of island radiations derives from the power of modern molecular phylogenetics. A common pattern found in several archipelagoes is that ancestral species tend to occur on older islands and more recently derived branches of an insect radiation are found on younger islands (Fig. 2). This pattern occurs frequently enough to have become known as the “rule of progression” because the diversification of an insect lineage progresses in step with the age of the islands. There are exceptions to the rule of progression, and these are found even in lineages that show a general progression pattern. A radiation that in large part adheres to the rule of progression may also exhibit retrogressive patterns, such as back-colonizations from younger to older islands, and branches of a radiation that undergo recent secondary radiations within older islands. In this respect, the geological history of individual islands can dramatically influence patterns of island radiations. For instance, new environments on older islands can be created by geological disturbances such as volcanic activity and landslides. Islands of volcanic origin can remain highly volatile, with eruptions and landslides altering the landscape and ecology of large areas at a time. Such events may cause extinctions, but conversely they can also promote new diversification. Volcanically altered landscapes can result

island to an older island is illustrated.

I N S E C T R A D I AT I O N S

over time, those lineages on older islands may incur more extinction events. Finally, a retrogressive back-colonization from the youngest

in geographic range expansions in dominant plant species (Pinus in the Canary Islands; Metrosideros in the Hawaiian Islands) with similar range expansions in the associated insect fauna. In addition, insect lineages that have already radiated above ground may diversify further in the unique habitats provided by subterranean lava tubes (e.g., crickets and planthoppers). Insect diversity on isolated singleton islands may be much more severely impacted by the loss of habitat from catastrophic natural phenomena and introduced species, because there are no nearby islands to replenish the lost diversity. On the island of St. Helena in the South Atlantic a number of species (including ground beetles, weevils, earwigs, and grasshoppers) have become endangered or extinct, attributed to ecological changes and habitat destruction. With older islands such as St. Helena (14.5 million years old), it is possible that some species that have become extinct over time may have provided evidence of historic radiations. Unlike the relatively good fossil record for extinct island birds, there is little fossil evidence with which to assess extinct insect faunas on islands, and it is possible that many species are lost before they are ever recorded as having been present.

The recolonization of continental land masses from lineages that have evolved on islands is a less common biogeographic pattern, and most of the known cases are from less remote islands/archipelagoes back to continents, such as Aphanarthrum bark beetles from the Canary Islands to North Africa, a distance of 100–500 km. Examples of continental recolonization from more remote oceanic islands are much rarer, but one possible example is the drosophilid genus Scaptomyza. This genus appears to have originated in the Hawaiian Islands and subsequently spread to the rest of the world. The loss of dispersal capability (e.g., wing reduction) common in many island lineages likely contributes to the asymmetrical rates of colonization between islands and continents. A number of other ways in which phylogenetic analyses can reveal patterns and processes of radiation include identifying outgroups and thus determining ancestral source biotas; the number and direction of colonization events; and using sister species pair comparisons to elucidate speciation mechanisms. Clarification of these processes can greatly facilitate our interpretation of evolutionary processes. Molecular data is also critical in revealing radiations among cryptic species such as the Laupala crickets. Increasingly, the functional genes involved in radiations are being studied. Radiations that are characterized by multiple adaptive shifts suggests complex evolutionary processes, but adaptively driven radiations can happen rapidly and may involve changes in only one or few genetic loci. Conversely, relatively simple shifts in mate recognition systems that have promoted radiation in Hawaiian crickets may be controlled by multiple genes of small effect. THE ROLE OF REPRODUCTIVE BEHAVIOR

As a potentially important process in insect radiations, the role of reproductive behavior is less well known and characterized than either geographic or ecological processes. However, in at least three island lineages (crickets, planthoppers, and drosophilid flies in the Hawaiian Islands), reproductive behavior has been shown to be important to interspecific isolation and intraspecific recognition and even a primary factor driving species radiations. In the Hawaiian swordtail cricket genus Laupala, the rapid explosion of speciation in this lineage is thought to be driven primarily by divergence in mate recognition systems. Multiple signals are involved in mate recognition in this group, including acoustic signals and, at close range, cuticular hydrocarbons. In contrast to ecologically driven radiations, the Laupala cricket species remain ecologically and morphologically similar. Among Hawaiian Drosophila several different modes of mate recognition are thought to contribute to species isolation,

including acoustic, visual, chemical, and tactile factors. In New Zealand, a radiation of Kikihia cicadas (~30 species) includes several morphologically cryptic species that can be differentiated by their acoustic signals (Fig. 3).

FIGURE 3 Two morphologically cryptic Kikihia cicada species that are

part of a radiation in New Zealand (predominant background color of both species varies from green to yellow-green, as does overall degree of darkness). These species are difficult to distinguish morphologically, but they have well-differentiated songs (scale bar of each oscillogram indicates one second). Photograph by David Marshall.

Whether these behaviors contributed to species isolation and radiation on islands, or developed as important species reinforcement mechanisms post isolation, is not well known. All the Hawaiian crickets exhibit classic island flightlessness, and there is complete loss of forewings in some species that no longer produce sound, but many species have retained the forewings for the purposes of sound production. In Hawaiian drosophilids, morphological characteristics, courtship behavior, and acoustic songs combine to form species-specific mate recognition systems, but in this system there have also been numerous adaptive shifts to different habitats and different larval feeding substrates. Thus, in this group both sexual selection and ecological adaptations are likely to be important in promoting speciation and radiation. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Crickets / Drosophila / Endemism / Extinction / Flightlessness / Sexual Selection / Species–Area Relationship FURTHER READING

Emerson, B. C. 2002. Evolution on oceanic islands: molecular phylogenetic approaches to understanding pattern and process. Molecular Ecology 11: 951–966.

I N S E C T R A D I AT I O N S

465

Gillespie, R. G., and G. K. Roderick. 2002. Arthropods on islands: colonization, speciation, and conservation. Annual Review of Entomology 47: 595–632. Juan, C., B. C. Emerson, P. Oromi, and G. M. Hewitt. 2000. Colonization and diversification: towards a phylogeographic synthesis for the Canary Islands. Trends in Ecology and Evolution 15: 104–109. Roderick, G. K., and R. G. Gillespie. 1998. Speciation and phylogeography of Hawaiian terrestrial arthropods. Molecular Ecology 7: 519–531. Roderick, G. K., and R. G. Gillespie. 2003. Island biogeography and evolution, in Encyclopedia of Insects. V. Resh and R. Cardé, eds. San Diego: Academic Press, 602–604. Wagner, W. L., and V. A. Funk. 1995. Hawaiian biogeography: evolution on a hot spot archipelago. Washington, DC: Smithsonian Institution Press.

INSELBERGS STEFAN POREMBSKI University of Rostock, Germany

Inselbergs are isolated rock outcrops that frequently consist of granites and gneisses and form old landscape elements that are widespread on crystalline continental shields. The environmental conditions on inselbergs are extreme both edaphically (because they lack soil) and microclimatically (because they are exposed to intense irradiation and high temperatures), and their vegetation is thus demarcated against the surroundings. Widespread are desiccation-tolerant lichens, mosses, ferns, and angiosperms.

erosional processes, they possess a considerable age, frequently surpassing millions of years. The environmental conditions on inselbergs are extreme both edaphically (i.e., in terms of soil and moisture conditions, because they lack soil and nutrients are scarce) and microclimatically (i.e., in terms of atmospheric conditions, because they undergo intense irradiation and temperatures regularly exceeding 60 °C). Even when situated in rainforests they form “micro-environmental deserts.” Despite the general lack of moisture, several locally restricted habitat types occur that are seasonally wet. Most prominent are seasonally water-filled depressions (“rock pools”), which may carry water for several consecutive weeks or months. Wet conditions are also present where water seeps continuously over longer periods at the feet of steep rocky slopes. However, unpredictable periods of drought may cause the drying up of wet sites even during the rainy season, thus triggering local extinction events. With regard to absolute height and surface area, inselbergs cover a vast spectrum. Absolute height ranges from a few meters (“shield inselbergs”) to several hundred meters, and surface area reaches from small outcrops cov-

ENVIRONMENTAL CONDITIONS

Patchily distributed habitats that are dominated by a hard, stony surface are known as rock outcrops when they protrude above the surroundings. This definition encompasses a broad range of landforms such as the tepuis of the Guayana shield, the tower karst mountains along the Guilin Li River in China, and inselbergs such as the Pao de Açúcar of Rio de Janeiro (Fig. 1). Inselbergs embody a spectacular example of island-like rock outcrops that consist of freely exposed slopes. Typically consisting of Precambrian granites and gneisses, they form prominent landscape features in all vegetational and climatic zones. The term “inselberg” (from German Insel, island, and Berg, mountain) was coined by the German geologist Wilhelm Bornhardt, who noticed their distinctive character as forming island-like habitats. In his honor, a particular type of inselberg that is characterized by a dome-shaped appearance with steeply precipitous flanks is named a “Bornhardt.” Inselbergs are particularly frequent on the old crystalline continental shields. Because of their resistance to

466

INSELBERGS

FIGURE 1 Typical examples of steep-flanked inselbergs in a tropical

landscape (near Pancas, Espirito Santo, Brazil).

ering several square meters to large domes extending over square kilometers. The degree of geographic isolation of inselbergs varies considerably. They can occur as hills isolated over hundreds of kilometers, or they can form dense clusters with individual outcrops occurring at distances of only a few kilometers. Throughout the world, cave or rock paintings dating back to prehistoric times are impressive testimonies of the close links of humans to inselbergs. They thus possess an immense cultural importance. However, throughout the tropical and temperate zones, negative human impacts on inselbergs have increased dramatically over the past decades. Of particular importance have been fire, quarry-

ing, and tourism, all of which have led to the degradation of numerous inselbergs, in particular those located near human settlements. PLANT AND ANIMAL LIFE Habitat Types

The extent of the expression of the island-like character of inselbergs (i.e., ecological isolation) depends on the surrounding vegetation types, with forests emphasizing particularly profoundly the floristic differentiation between rock outcrops and their surrounding matrix. Based on plant species composition and physiognomic criteria, a limited set of typical plant communities and habitat types can be distinguished. The most important ones are described concisely as follows: CRYPTOGAMIC CRUSTS

Exposed rocks are covered by cyanobacterial lichens (frequently Peltula spp.) and cyanobacteria (e.g., Stigonema spp. and Scytonema spp.) which are responsible for the often dark coloration of inselbergs. Individual microhabitats such as exposed rocky slopes, boulders, and drainage channels are differentiated floristically. Lichens and cyanobacteria may form dense epilithic surface layers, but they also occur endolithically. Cryptogamic communities on inselbergs have very close relationships to those on other rock surfaces. Particularly well developed affinities exist to the Tintenstrich communities that are ubiquitous on both anthropogenic and natural rocks in temperate and tropical regions.

two families contain tree-like, woody-stemmed species that are desiccation tolerant. On South American inselbergs, both tank-forming (e.g., Alcantarea spp., Vriesea spp.) and xerophytic (e.g., Dyckia spp., Encholirium spp.) Bromeliaceae occur from sea level up to high altitudes, as is the case in southeastern Brazil. In addition, the bromeliad genus Tillandsia occurs with mat-forming species throughout many parts of that country. In tropical Africa and Madagascar, desiccation-tolerant Cyperaceae and Velloziaceae are widespread. The Cyperaceae Afrotrilepis pilosa and Microdracoides squamosus occur as mat-formers in West Africa (Fig. 2). In southern and eastern Africa and on Madagascar, they are replaced by the Cyperaceae genus Coleochloa and by numerous species of the genus Xerophyta (Velloziaceae). Remarkably, the woody fibrous stems of mat-forming Cyperaceae and Velloziaceae are occasionally colonized by specific epiphytic orchids. Among the orchids, the genera Polystachya (tropical Africa), Constantia (Brazil), and Pseudolaelia (Brazil) comprise species that are restricted to monocotyledonous mats.

EPILITHIC VASCULAR PLANTS

This group comprises higher plant species that grow directly on open rock. Frequently, these epilithic species are succulents or xerophytes. Prominent examples are provided by numerous orchids (e.g., in the neotropical genera Cyrtopodium and Laelia and the aroid genus Anthurium, which possess water-storing stems or leaves). Bromeliaceae, too, are very rich in epilithic species, with Brazil being their center of diversity. MONOCOTYLEDONOUS MATS

On both level and inclined open rocky slopes, dense stands of long-lived (i.e., hundreds of years) monocotyledons occur. These epilithic species are of matlike appearance and are firmly attached to the rock by dense wiry roots. Frequently, monocotyledonous mats occur as isolated patches surrounded by open rock, but large, continuous expanses of mats can also be found. Most typical are Bromeliaceae, Cyperaceae, and Velloziaceae. The last

FIGURE 2 Specialized mat-forming monocotyledonous plants (here

Microdracoides squamosus, Cyperaceae, Cameroon) colonize freely exposed rocky slopes. In being desiccation-tolerant, they are well adapted to survive periods of prolonged droughts.

ROCK POOLS

Usually on level parts, seasonally water-filled rock pools covering a wide range of sizes, forms, and depths occur (Fig. 3). They are products of natural solution processes, have a considerable age, and are typically irregularly shaped depressions of variable depth, covering up to several square meters. They form an unreliable habitat for higher plants because they may dry out even in the rainy season during rainless periods. Typically, epilithic and endolithic cyanobacteria and lichens form a dense cover on open rock walls. Cover by vascular plants is generally very sparse. Widespread are plants that are otherwise

INSELBERGS

467

SUCCULENTS

FIGURE 3 Seasonally water-filled rock pools (in Australia called gnam-

mas) provide establishment and growth sites for aquatic plants (e.g., Dopatrium longidens, Scrophulariaceae, Ivory Coast). In a sense, they form “islands on islands.”

colonizers of marshy ground and ponds. The number of species restricted to rock pools is relatively low. Prominent examples occur within Scrophulariaceae (e.g., in the genera Amphianthus, Dopatrium, Glossostigma, and Lindernia) and in the fern genus Isoetes (with highly specialized species in the southeastern United States, tropical Africa, and southwestern Australia). EPHEMERAL FLUSH VEGETATION

Located at the foot of rocky slopes or along the downslope fringes of monocotyledonous mats, this plant community depends on seepage water that is only available during the rainy season. The basic matrix is formed by Poaceae and Cyperaceae with numerous, mostly diminutive annuals imbedded within. Typically the substrate is very shallow with the lowest values occurring toward the transition to the open rock. Nutrient availability is restricted, a situation reflected in the floristic composition of the ephemeral flush vegetation with Lentibulariaceae, Eriocaulaceae, and Xyridaceae being usually well represented. In general, carnivorous plant species (e.g. Drosera spp., Genlisea spp., Utricularia spp.) are widespread and have a center of diversity here. Plant Adaptive Traits

Inselbergs form centers of diversity for certain plant functional types that are well adapted for survival under environmentally extreme conditions. In particular, water scarcity and low nutrient availability have had a deep impact on the floristic composition of inselbergs. The microclimatical and edaphic dryness of inselbergs is reflected in the presence of numerous drought-adapted plants. Among them, succulents and desiccation-tolerant vascular plants are particularly prominent. 468

INSELBERGS

On exposed rocky slopes of tropical inselbergs, succulents occur as perennial lithophytes. In the paleotropics, inselbergs in East Africa and Madagascar are particularly rich in succulents. Here, Aloaceae (Aloe spp.), Apocynaceae (e.g., Pachypodium), Crassulaceae (e.g., Kalanchoe), and Euphorbiaceae (e.g., Euphorbia) comprise numerous endemics. On neotropical inselbergs, Bromeliaceae (e.g., Encholirium), Cactaceae (e.g., Coleocephalocereus), and Orchidaceae (e.g., Cyrtopodium) occur with succulent species. Pachycaulous and caudiciformous species are widespread on tropical inselbergs. These plants possess fat water-storing trunks or a subterranean caudex. Comparatively rare are annual leaf succulents that occur on inselbergs in both tropical and temperate regions. Examples are Cyanotis lanata (tropical Africa), Sedum smallii (southeastern United States), and the genera Crassula and Calandrinia, which occur on Australian inselbergs. DESICCATION-TOLERANT VASCULAR PLANTS

Only ~300 species of vascular plants can be classified as being absolutely desiccation tolerant and are known as the so-called resurrection plants. Desiccation-tolerant vascular plants are well adapted to withstand long periods of drought by resting in a state of dormancy. Most resurrection plants lose their chlorophyll and other photosynthetic pigments (i.e., they are poikilochlorophyllous, in contrast to relatively few desiccation-tolerant plants that keep their photosynthetic pigments during the process of desiccation and which are thus homoiochlorophyllous). Monocots outnumber dicots among desiccation-tolerant vascular plants, with Velloziaceae, Cyperaceae, and Poaceae being particularly important. Desiccation-tolerant arborescent monocots are unique with regard to the possession of certain morphological and anatomical features. Their fibrous stems consist mainly of adventitious roots and old persistent leaf bases and may attain a height of several meters. Surprisingly, the adventitious roots possess a velamen radicum that might be of functional importance for the rapid capillary uptake of rain water. On inselbergs desiccationtolerant vascular plants are mainly found as mat-formers, but they also occur in shallow depressions and even in seasonally water-filled rock pools. A striking and highly specialized desiccation-tolerant species in rock pools is the Scrophulariaceae Lindernia intrepidus (syn. Chamaegigas intrepidus), an endemic to Namibia. Geographic Patterns of Species Richness

Floristically, inselbergs in different geographical regions are clearly distinct. Based on comparative floristic data,

three hotspots of inselberg plant diversity can be identified, which are rich in both species and endemics: (1) southeastern Brazil, (2) Madagascar, and (3) southwestern Australia. It has to be emphasized, however, that for several tropical regions (e.g., Angola, India), our knowledge of inselbergs is still sparse.

resented by the monocotyledonous genus Borya. Apart from a few tiny, short-lived leaf succulents (Calandrinia spp., Crassula spp.) succulents are absent. Very rich in species are terrestrial Utricularia species, which are typical components of seasonally wet vegetation types. Faunistic Aspects

SOUTHEASTERN BRAZIL

That the forest vegetation of the Mata Atlântica is rich in species and endemics is well known. However, it is frequently overlooked that in this region (i.e., in particular parts of the Brazilian federal states of Rio de Janeiro, Minas Gerais, and Bahia) rock outcrops not only form dominant landscape elements but also support large numbers of endemics. Remarkably high is the beta diversity (i.e., the degree of floristic differentiation over small distances) in southeastern Brazil, with considerable species turnover between individual outcrops. Their vegetation is extremely rich in drought-resistant perennial species, whereas annuals are relatively rare. Prominent examples are xerophytic and succulent bromeliads (e.g., Encholirium, Orthophytum, Pitcairnia, Vriesea, Tillandsia), cacti (e.g., Coleocephalocereus, Melocactus), and orchids (e.g., Cyrtopodium, Laelia). Moreover, resurrection plants occur abundantly and belong to genera such as Vellozia and Trilepis and to the fern genera Anemia, Doryopteris, and Selaginella. MADAGASCAR

Madagascan inselbergs are particularly frequent on the central plateau, where they are famous for their richness in succulent plants. Moreover, desiccation-tolerant vascular plants (e.g., Coleochloa, Myrothamnus, Selaginella, Xerophyta) occur profusely. The preliminary data available for this region show that the rock outcrop flora contains an extraordinarily high percentage of endemics. A considerable number of genera (e.g., Aloe, Cynanchum, Euphorbia, Kalanchoe, Pachypodium, Senecio) have obviously radiated on Madagascan inselbergs. Remarkably, many species show a high degree of morphological differentiation over short distances, making the limitation of taxa difficult (e.g., within Euphorbia and Xerophyta).

Inselbergs provide structural niche components that can be used as nest sites, for shade, or for water supply. Special attention among the inselberg specialists is merited by rock lizards, hyraxes, klipspringers, or rock wallabies. Moreover, inselbergs are visited by a large number of animals that use rock outcrops as part of their range. SEE ALSO THE FOLLOWING ARTICLES

Madagascar / Orchids / Pantepui / Vegetation FURTHER READING

Barthlott, W., S. Porembski, R. Seine, and I. Theisen. 2007. The curious world of carnivorous plants: a comprehensive guide to their biology and cultivation. Portland, OR: Timber Press. Hopper, S. D., A. P. Brown, and N. G. Marchant. 1997. Plants of western Australian granite outcrops. Journal of the Royal Society of Western Australia 80: 141–158. Hunter, J. T. 2003. Persistence on inselbergs: the role of obligate seeders and respouters. Journal of Biogeography 30: 497–510. Kruckeberg, A. R. 2002. Geology and plant life: the effects of landforms and rock types on plants. Seattle: University of Washington Press. Parmentier, I., T. Stévart, and O. J. Hardy. 2005. The inselberg flora of Atlantic Central Africa: I. determinants of species assemblages. Journal of Biogeography 32: 685–696. Porembski, S., and W. Barthlott, eds. 2000. Inselbergs—biotic diversity of isolated rock outcrops in tropical and temperate regions. Ecological Studies, volume 146. Berlin: Springer–Verlag. Porembski, S., and W. Barthlott. 2000. Granitic and gneissic outcrops (inselbergs) as centers of diversity for desiccation-tolerant vascular plants. Plant Ecology 151: 19–28. Sarthou, C., S. Samadi, and M.-C. Boisselier-Dubayle. 2001. Genetic structure of the saxicole Pitcairnia geykesii (Bromeliaceae) on inselbergs in French Guiana. American Journal of Botany 88: 861–868. Twidale, C. R., and J. R. Vidal Romani. 2005. Landforms and geology of granite terrains. Leiden, Netherlands: Balkema.

INTRODUCED SPECIES

SOUTHWESTERN AUSTRALIA

DANIEL SIMBERLOFF

In this region, inselbergs occur along steep climatic gradients from winter rainfall climates to inland deserts. Species richness and endemism decline with increasing aridity. Particularly striking is the richness in annuals, with Asteraceae, Stylidiaceae, Poaceae, and Amaranthaceae being particularly speciose. Resurrection plants are mainly rep-

University of Tennessee, Knoxville

Introduced species are those brought purposefully or inadvertently by humans to new regions, whereas a biological invasion is the establishment and spread of an introduced species into one or more habitats in its new home. Bio-

INTRODUCED SPECIES

469

logical invasions are particularly pronounced on islands; worldwide, ~1.6 times as many mammal species and three times as many bird species have been introduced successfully to islands than to continents. Introduced carnivores eliminated many island bird species and subspecies, as well as reptiles and amphibians. Introduced grazers such as sheep, rabbits, and reindeer have eliminated many endemic island plants, and rooting by introduced pigs has massively eroded mountainous islands. Introduced plants have transformed island forests into grassland. GLOBAL PATTERNS

For birds, 129 species are believed to have gone extinct worldwide since AD 1500; of these, 119 were island endemics. Of these 129 extinctions, scientists implicate a cause for 93 species; of these 93, 78 (84%) were eliminated wholly or partly by introduced species (primarily carnivores, but also herbivores that devastated habitats and birds that vectored introduced diseases). Even these figures do not convey the full impact of introduced species on island birds. In New Zealand, for example, 24 endemic birds were eliminated after the arrival of ancestors of the Maori (probably AD ~1250–1300) but before AD 1400, and some of these may have fallen prey to Pacific rats (Rattus exulans), which arrived with the earliest settlers. Island birds are not unique in their susceptibility to invaders. Even a group such as the freshwater fishes, relatively poorly represented on islands because of dispersal difficulties, has been devastated. Of 69 extinctions since AD 1500 (not counting Lake Victoria cichlids, or one extinction on Madagascar), 21 have been on islands; of these, causes are suspected for 20, and in each instance but one, introduced species are implicated. Insufficient data exist for similar analyses on invertebrate extinctions, but introduced species have had catastrophic impacts on a few well-studied groups on islands. For instance, all 144 known extinctions of land snails since AD 1500 were on islands. Causes are not known for many, but ~50 were caused at least partly by the rosy wolf snail (Euglandina

FIGURE 1 Rosy wolf snail, Euglandina rosea. Photograph courtesy of

Jack Jeffrey Photography.

470

INTRODUCED SPECIES

rosea; Fig. 1), a predatory snail introduced to islands worldwide, but especially in the Pacific, in a failed attempt to control the introduced giant African snail Achatina fulica. Predation by introduced rats has also contributed to many island snail extinctions. Native island plants are at risk from introduced species, just as native animals are. For example, on Phillip Island in the Norfolk group (Australia), 13 indigenous plant species (including two endemics) were eliminated within 140 years of the establishment of pigs, goats, and rabbits, primarily because of grazing by the last of these. Similarly, on Laysan Island (Hawaii), rabbits introduced in 1903 eliminated 26 plant species in just two decades. On St. Helena, beginning with Portuguese discovery in 1502, goats, pigs, cattle, rabbits, horses, donkeys, rats, and mice arrived, with kilometer-long goat herds numbering in the thousands wreaking particular havoc on vegetation. Seven of the 46 endemic plant species known to have been present in 1502 have since been extinguished (plus an unknown number that vanished without records or collections), primarily by these introduced animals. The exotic animals were abetted by introduced plants, far more adapted to the ravages of goats than were the native plants. Today, introduced plant species (at least 77) outnumber native species (53) and dominate most of the island. Even these statistics understate the effect of the introduced species: Many of the native plant species consist of just a few individuals, and two species have now each been reduced to one plant. For some islands, many impacts of introduced species have been documented. This is especially true of the Hawaiian Islands, New Zealand, the Mascarenes, Macaronesia, the Antilles, and St. Helena. However, all islands have probably been modified by introduced species, whereas only a small fraction of these possible impacts have been studied. TYPES OF IMPACTS

Impacts of introduced species that lead to endangerment or extinction are varied. Some of the most dramatic and obvious ones, such as an introduced mongoose (Fig. 2) eating a native ground-nesting bird, may not be as consequential as subtler impacts that indirectly affect an entire community, as when introduced plants change nitrogen availability or fire frequency. A subtle but potentially drastic impact is the fertilization of parts of the island of Hawaii by the shrub Morella (Myrica) faya, native to Macaronesia and introduced a century ago. Hawaii is a young, volcanic island, so soil is nitrogen-poor, and the native plants have adapted to this soil regime, which

most introduced plants cannot tolerate. A nitrogen-fixer, M. faya quadruples the input of fixed nitrogen, which in turn stimulates invasions, and ultimately dominance, by a number of other introduced plants. Changing the plant community, in turn, affects many associated animal species. On many Pacific islands, introduced grasses (e.g., in Hawaii, Melinus minutiflora from Africa and Schizachyrium condensatum from North America) foster more frequent and intense fires that kill most native trees and shrubs, changing diverse, nativedominated woodlands into low-diversity grasslands dominated by exotics. Introduced herbivores, by removing the dominant plant species, can also affect entire island communities. The goats of St. Helena, noted above, are an example. Similarly, on the Antarctic island of South Georgia, reindeer (Rangifer tarandus), introduced from Norway by whalers in the early twentieth century, destroyed great areas of the native tussock grass Poa flabellata and also grazed heavily on large lichens. The reindeer have spurred a massive invasion by introduced Poa annua, negatively affecting seabird colonies. In addition, the introduced grass has led to a decline in body size of a native beetle that cannot digest it. Impacts on other species are likely but unstudied. North American beavers (Castor canadensis), introduced in 1946 to Isla Grande (Tierra del Fuego), have spread to several other islands, and they have changed many closed southern beech forests to grass- and sedgedominated meadows. By virtue of gnawing on seeds, Pacific rats now appear to have been at least partly responsible for massive deforestation of endemic palm forests on Oahu and Easter Island, an event that had been attributed wholly to humans. In addition to this impact by herbivory, rats can affect entire island ecosystems by changing soil fertility. On small offshore islands of New Zealand invaded by black rats (Rattus rattus) or Norway rats (R. norvegicus), predation of seabirds led the latter to abandon these islands, thus disrupting transport of nutrients to the islands, thereby lowering soil fertility. This change, in turn, triggered a cascade of effects on belowground organisms, which in turn affected many ecosystem processes and features, both above and below ground. Many other impacts of introduced species, including some of the most noteworthy island cases, are seen primarily as particular species affecting particular other species, rather than entire ecosystems. Determining which natives an introduced species affects is not always simple, as many impacts on one species can be propagated to others (for example, when reindeer browsing on a native grass on South Georgia ultimately affect a native beetle).

Predation is foremost among these direct species-level impacts. The hecatomb attributed to the rosy wolf snail has already been depicted. On Guam, the brown tree snake (Boiga irregularis), introduced in cargo from the Admiralty Islands just after World War II, has eliminated all but one of the native forest bird species and subspecies. Feral populations of introduced housecats have devastated seabird colonies on Ascension Island, Kerguelen Island, Marion Island, and others. On Kerguelen alone, feral cats are estimated to kill ~1.3 million birds each year. The small Indian mongoose, Herpestes auropunctatus (Fig. 2), was introduced in the late nineteenth and early twentieth centuries to the West Indies, Hawaiian Islands, Fiji, Mauritius, Okinawa, and elsewhere, primarily to control rats in agriculture, but also, on the Adriatic islands, to control snakes. It is almost certainly wholly responsible for the global extinction of a rail in Fiji and a petrel on Jamaica, and it contributed to several other avian extinctions on islands, interacting with introduced rats, cats, dogs, and pigs, as well as anthropogenic habitat destruction. H. auropunctatus has eliminated several birds that persist on nearby mongoose-free islands. The mongoose is also believed responsible for extinction of four endemic Hispaniolan mammals and possibly several West Indian snake species; it is likely to have caused extirpation of other West Indian snakes on those islands to which it was introduced. Mongooses were almost surely the key cause of extirpation of lizards on particular islands in the West Indies and Fiji, as evidenced by the persistence of these species only on mongoose-free islands. The mongoose similarly extirpated a frog species from three Caribbean islands. In many instances, the exact role of each introduced predator is uncertain, but a battery of them have

FIGURE 2 Small Indian mongoose, Herpestes auropunctatus. Photo-

graph courtesy of Jack Jeffrey Photography.

INTRODUCED SPECIES

471

driven a species to the brink of extinction or beyond. For example, the New Zealand kakapo (Strigops habrotilus), a large, flightless parrot, now exists only on a few small, predator-free islands to which it was translocated, having been eliminated on all main islands by introduced Pacific rats (R. exulans), Norway and black rats, cats, dogs, pigs, stoats (Mustela erminea), weasels (Mustela nivalis), and ferrets (Mustela putorius furo). Another endangered New Zealand bird, the kokako (Callaeas cinerea wilsoni), exemplifies competition for food with introduced species, in this case brush-tailed possums (Trichosurus vulpecula), red deer (Cervus elaphus), and goats (Capra hircus). Two palearctic introduced yellowjacket wasps, Vespula vulgaris and V. germanica, also compete with threatened native New Zealand birds, especially another endemic parrot, the kaka (Nestor meridionalis), in southern beech forests. The competitive effect of V. vulgaris is greatly exacerbated by a native scale insect that exudes a dark, sweet “honeydew” on beech trunks. Feeding on this honeydew, V. vulgaris reaches densities of 360 wasps/m2, turning the trunks yellow and harvesting 8.1 kg of honeydew/ha, at least equal to consumption by all birds together. Competition with an invader for food is strongly implicated in the decline of an island native in other cases. Further examples include the ongoing replacement of the European red squirrel (Sciurus vulgaris) by the North American gray squirrel (S. carolinensis) in Britain and the impact of the introduced house gecko, Hemidactylus frenatus, on resident native gecko species on several Pacific islands. Competition can also be for resources other than food: The endangered Puerto Rican parrot (Amazona vittata) competes for nest sites with an introduced bird, the pearly-eyed thrasher (Margarops fuscatus), as well as introduced honeybees (Apis mellifera). Herbivory by introduced species can affect entire ecosystems, as has been noted above for goats on St. Helena and reindeer on South Georgia. However, many introduced herbivores devastate particular plant species that are not key players in structuring an entire ecosystem. For example, 17 plant species and subspecies are threatened in the California Channel Islands by herbivory by introduced herbivores (primarily goats, pigs, and sheep); most are small plants unlikely to have been community dominants. Introduced parasites and diseases have frequently devastated island species, contributing to the extinction of many. In the Hawaiian Islands, avian pox and malaria arrived with the extensive introduction of exotic songbirds in the late nineteenth and early twentieth centuries. These diseases have contributed to the decline and

472

INTRODUCED SPECIES

extinction of many Hawaiian bird species. Their spread was fostered by introduced mosquitoes, particularly the tropical species Culex quinquefasciatus, which gradually adapted to cooler temperatures and thus advanced higher up mountains, restricting surviving native birds to eversmaller, higher refugia. A particularly subtle impact of some introduced species is hybridization with native species, which can lead to a sort of genetic extinction even though no lineage is terminated. For instance, both the New Zealand gray duck (Anas superciliosa superciliosa) and the Hawaiian duck (A. wyvilliana) are endangered partly because they hybridize with the introduced North American mallard (A. platyrhynchos). Likewise, the Seychelles turtledove (Streptopelia picturata rostrata) has hybridized so extensively with S. p. picturata, introduced from Madagascar, that the Seychelles population now consists of a hybrid swarm more similar on average to the invader than to the native. Sometimes two or more introduced species exacerbate one another’s impact on natives, a phenomenon known as invasional meltdown. On Christmas Island, the introduced yellow crazy ant (Anoplolepis gracilipes) had long been present but quite innocuous, until the late 1980s. Then, introduction of a scale insect, plus outbreaks of a native scale insect, led to massive production of honeydew. The ants protect the scales and harvest the honeydew, and populations of both scales and ants exploded. The ants then devastated populations of the native red land crab (Gecarcoidea natalis), which in turn caused greatly increased growth of ground cover plants, seeds and seedlings of which had previously been harvested by crabs. The invasion of nitrogen-fixing Morella faya in Hawaii, described previously, is part of a meltdown in which the seeds of the plant are primarily dispersed by the introduced Japanese white-eye (Zosterops japonicus) as well as introduced rats and feral pigs, whereas greatly increased densities of introduced earthworms under Morella trees enhance the rate at which nitrogen-rich litter is buried, thereby aiding the invasion of previously nitrogen-limited introduced plants. Often, introduced birds exacerbate invasions of exotic plants by dispersing their seeds. The red-whiskered bulbul (Pycnonotus jocosus), introduced to the Mascarenes, has exacerbated invasions by several alien plants, even into previously undisturbed areas. On Isla Grande, beaver-altered sites have higher levels of soil nitrogen, which favors invasion by a number of introduced plants. Sometimes the impacts of an introduced species initiate a chain reaction that ends up indirectly affecting native species in surprising ways. In Great Britain, caterpillars of

the native large blue butterfly (Maculinea arion) matured underground in nests of the ant Myrmica sabuleti. This ant cannot nest in overgrown habitats, and during the course of centuries, reduced livestock grazing and changing land-use patterns left rabbits (Oryctolagus cuniculus), introduced from continental Europe in AD ~1150, as the main species maintaining suitable habitat for the ant. The South American myxoma virus, introduced to France in 1952 to control the rabbit, quickly spread to Great Britain and devastated rabbit populations. Habitats became overgrown, ant populations plummeted, and the butterfly disappeared. ARE ISLANDS MORE VULNERABLE THAN MAINLAND AREAS TO INTRODUCED SPECIES?

Conventional wisdom is that introduced species are more likely to survive on islands than on mainland areas and to have greater impacts there, because island ecosystems are particularly fragile, and island species weak. Island species have been termed evolutionary “backwaters and dead ends,” whereas island ecosystems are seen as presenting less “biotic resistance” to invaders than do continental ecosystems. Probably the most characteristic doomed island species, in the mind of the public, is the hapless, extinct Mauritius dodo. However, this view of generic weakness of island species and communities is suspect for both a theoretical reason and an empirical one. The theoretical reason is that island species, at least endemic ones, have evolved on their respective islands, and one would expect natural selection to have adapted them to these islands better than a species arriving from elsewhere. The empirical reason is that, for every sort of impact described in the previous section, it is possible to cite similar impacts of species introduced to mainland communities. There are even island species that have wrought havoc with continental communities, such as the New Zealand mud snail Potamopyrgus antipodarum, introduced to the Greater Yellowstone region of North America. The percentage of introduced species is often higher on islands than on the mainland, a fact adduced as evidence of island susceptibility to invasion. For instance, ~35% of Hawaii’s insect species are introduced, whereas only ~3% of the insect species of the continental United States are introduced. However, in fact there are as many introduced insect species in the continental United States as in Hawaii: ~3000. The percentage is greater in Hawaii than in the continental United States simply because the latter has ~90,000 native species, and the former only ~6000. The other difficulty in interpreting higher per-

centages of introduced species in island biotas is that no account is taken of failed introductions and amount of effort devoted to attempting to establish exotic species. Many islands (e.g., Hawaii, New Zealand, La Réunion, Mauritius) had active acclimatization societies that tried to redress the paucity of native bird life by introducing species from elsewhere. Thus, for example, ~70 species of perching birds and doves have been introduced to Hawaii (compared to only 40 in all of the continental United States), of which about half survive, a number that rivals the surviving native species in these groups (most of which are now endangered). Only 13 introduced species of perching birds and doves established populations in the continental United States. However, several acclimatization societies strove mightily to establish these birds in Hawaii, with large propagules and multiple introductions for many. It has been established that propagule pressure—the number of individuals introduced and number of attempts—is a key determinant of the likelihood that an introduced species will establish a population. Though island species are not generically maladapted weaklings, there does appear to be an aspect of many island biotas that predisposes them to be particularly susceptible to certain kinds of introductions. This is the fact that many islands, especially oceanic islands, lack large vertebrate predators and social insects. Native species on these islands therefore did not evolve adaptations to such species, and an absence of such adaptations often led them to be devastated upon introduction of predators and ants. Many island birds, for example, nest on the ground, making them easy prey for the small Indian mongoose, rats, cats, pigs, and mustelids. Approximately 90% of bird extinctions since AD 1500 that can be ascribed to predation have occurred on islands. Similarly, Guam has no native snakes or predators that could have countered the introduced brown tree snake, so native birds (and other species) were easy prey for a new type of predator. Furthermore, continental birds introduced to islands had evolved in the presence of vertebrate predators, so they were preadapted to replace vulnerable native birds. Isolated islands also lack large mammalian grazers and browsers, so it is no wonder that the dominant native plants were devastated by introduced goats, reindeer, sheep, and cattle. By contrast, continental plants introduced to islands had evolved with large grazers and browsers, so it is unsurprising that they replaced many native island plants. A second factor predisposing island communities to be vulnerable to certain kinds of invaders is that islands are smaller than continents, usually much smaller. Thus, it is far less likely on an island that there will be a refuge

INTRODUCED SPECIES

473

available to a native species that an invader cannot reach. Even hurricanes have far greater biodiversity consequences for this reason; a number of bird species and subspecies have been eliminated by hurricanes on islands (e.g., five on Kauai from Hurricane Iniki in 1992), but none on continents, and this is not because island birds are somehow weaker than continental ones; rather, their entire range can be exposed to a hurricane. ERADICATING INTRODUCED SPECIES ON ISLANDS

A growing number of introduced species have been eradicated on islands around the world, and the size of such projects has increased substantially as eradication technology advances. Terrestrial island invaders are particularly tempting targets for eradication because islands are generally far less likely to be reinvaded than are continental regions, simply by virtue of the water barrier separating islands from sources. The first island eradication of an introduced insect was the elimination of the tsetse fly (Glossina spp.) from Principe in the Gulf of Guinea. The flies had arrived in cargo from Africa in 1825, and sleeping sickness devastated the human population beginning in 1859. However, the fly and the disease were eradicated between 1911 and 1914. In 1956, a new tsetse invasion was noted on Principe, and the fly was again eradicated in a massive effort using traps, insecticides, extensive brush clearing, and hunting to reduce populations of pigs and wild dogs. Perhaps the most famous island eradication of an introduced insect entailed a demonstration in 1954–1955 on Curaçao. There, release of enormous numbers of sterile males eliminated an entire population (the screw-worm fly, Cochliomyia hominivorax) by keeping females from mating with fertile males. This technique has since been employed in several insect eradications on islands. Notable recent eradications of introduced insects from islands include the Oriental fruit fly (Dacus dorsalis) from Rota and Guam and the melon fly (Bactrocera cucurbitae) from the entire Ryukyu Archipelago, including Okinawa. Many invasive introduced mammals have been eradicated from islands. The most widely publicized of such projects have eliminated rats. Black, Norway, and Pacific rats have been eradicated from at least 100 islands worldwide, usually with a goal of saving threatened bird species. The largest island cleared of rats to date is 113-km2 Campbell Island, cleared in 2001 by the New Zealand government, which is in the intial stages of planning an attempt to eliminate rats from Great Barrier Island (274 km2). Probably the best-known eradication of an inva-

474

INTRODUCED SPECIES

sive mammal was the successful campaign completed in 1986 to eradicate nutria (Myocaster coypus) from Great Britain. Among other introduced mammals eradicated from islands have been feral cats and dogs, house mice, rabbits, muskrats (Ondatra zibethicus), and burros. Very recently, large, longstanding populations of goats and pigs have been eradicated from Santiago (58,000 ha) and Isabella (400,000 ha) in the Galápagos. The New Zealand government is currently planning a campaign to eradicate Norway, black, and Pacific rats as well as rabbits, stoats, hedgehogs (Erinaceus europaeus), and feral cats simultaneously from Rangitoto and Motutapu Islands. There have been far fewer attempted eradications of plants on islands and very few successes. However, the United States government recently succeeded in its 16-year campaign to rid Laysan Island of an introduced sandbur (Cenchrus echinatus), which had spread to replace native plants on 30% of the vegetated part of the island, threatening the endangered endemic Laysan finch (Telespiza cantans) and Laysan duck (Anas laysanensis). Eradicating invaders from an island does not automatically restore the desired state, and unexpected results are common. Eradication of rats has, on several islands, been followed by explosions in density of house mice (Mus musculus), which may prove equally detrimental to the species targeted for restoration. Elimination of a predator may also increase herbivore populations, to the detriment of native plants, whereas eradicating an introduced herbivore can profit introduced weeds at the expense of native plants. Eradication of rabbits from Motunau Island (New Zealand) led to increased populations of introduced boxthorn (Lycium ferocissimum), whereas removal of grazing livestock from Santa Cruz Island (California) caused explosive increases in fennel (Foeniculum vulgare) and other introduced plants. Changes in plant community composition and structure after herbivore eradication can, in turn, affect animal populations. On Mana Island (New Zealand), removal of cattle decreased native lizard populations by modifying vegetation. SEE ALSO THE FOLLOWING ARTICLES

Biological Control / Invasion Biology / Land Snails / Pigs and Goats / Rodents FURTHER READING

D’Antonio, C. M., and T. L. Dudley. 1995. Biological invasions as agents of change on islands versus mainlands, in Islands: biological diversity and ecosystem function. P. M. Vitousek, L. L. Loope, and H. Adsersen, eds. Berlin: Springer-Verlag, 103–121.

Elton, C. S. 1958. The ecology of invasions by animals and plants. London: Methuen. Lockwood, J. L., M. F. Hoopes, and M. P. Marchetti. 2007. Invasion ecology. Malden, MA: Blackwell. Simberloff, D. 1995. Why do introduced species appear to devastate islands more than mainland areas? Pacific Science 49: 87–97. Towns, D. R., I. A. E. Atkinson, and C. H. Daugherty. 2006. Have the harmful effects of introduced rats on islands been exaggerated? Biological Invasions 8: 863–891. Veitch, C. R., and M. N. Clout, eds. 2002. Turning the tide: the eradication of invasive species. Gland, Switzerland: IUCN Species Survival Commission. Williamson, M. 1996. Biological invasions. London: Chapman & Hall.

INVASION BIOLOGY GEORGE RODERICK University of California, Berkeley

PHILIPPE VERNON University of Rennes 1, Paimpont, France

A biological invasion is the establishment and spread of a locally nonindigenous species. Invasive species have had significant ecological and economic impacts on islands, including displacement or extirpation of native species, changes in physical geography such as erosion and silting of streams and offshore habitats, and rising costs associated with management of urban and agricultural pests. Invasive species can colonize new areas on their own, inadvertently through actions of humans, or as a result of purposeful introductions with unexpected consequences. With the expansion of global trade and connectivity, the importance of invasive species is intensifying, and biological invasion is appropriately considered an agent of global change. Invasive species are a particular environmental concern for islands, especially those that have been isolated ecologically. STATUS AND IMPACTS

The structural isolation of islands limits the number of species that colonize, resulting in communities comprising an unrepresentative collection of founding species. Adaptive radiation and other processes, which often scale with isolation, further accentuate the unrepresentative nature of species within island communities. It is for these reasons that new invasive species on islands can have such large impacts. The importance of invasive species has been recognized since the time of Darwin. However, with the recent increase in global trade and the accompanying homogenization of the world’s flora and fauna, biological communities on islands are becoming

less isolated and are experiencing increasing ecological and socioeconomic impacts associated with the arrival of many invasive species. An invasive species is a locally nonindigenous species that colonizes a new geographical area and thus expands its range. Invasive species have a “pestlike” connotation, in that they typically directly or indirectly cause ecological or economic damage. Government entities often use the term invasive alien species (IAS), which, like the term exotic species, emphasizes that such species are nonindigenous. Knowledge of the ecological origins of a species (i.e., indigenous or nonindigenous) is crucial in understanding the biology of invasive species as well as in their management. Determining which species are endemic (i.e., native and found nowhere else) is usually not difficult. However, resolving which species are indigenous (i.e., native, but also found elsewhere) can be problematic—accordingly, many types of data are used to make this determination, including biogeography, comparative systematics, molecular genetic variation, and anthropological associations. Using these approaches, biodiversity surveys show that nonindigenous species can make up a large proportion of island biota. For example, in the Hawaiian archipelago, of the 2142 flowering plant species and 8151 insect species, 53% and 34%, respectively, are nonindigenous (Table 1). In the paucispecific cold oceanic islands, a dramatic increase in the number of insect species has been recently observed—on the Kerguelen Islands, for example, of 39 known insect species (Fig. 1), 16 (41%) are alien. Severe ecological and environmental impacts have been attributed to invasive species on islands, such as displacement of native species, changes in demographic rates, gene flow modifications, and drastic perturbations in pristine ecosystems. For example, in the Hawaiian archipelago the introduction of the plant Myrica faya has displaced endemic plants, and because it also fixes nitrogen through a symbiotic relationship with an actinomycete bacteria, its presence alters the soil, facilitating colonization by a diversity of other alien species. Several features of islands and their biota may accentuate these effects: large edge effects (relative to area), large topological diversity within small geographical areas, demographic and genetic consequences related to small population sizes that characterize indigenous species on islands, and the nonrepresentative species composition and subsequent evolution in the absence of many mainland taxa. Although the role of these features requires more study, it is broadly understood that invasive species have a disproportionate impact in insular environments in general, whether oceanic islands, mountaintops, ponds and lakes, or even old trees in a field.

I N VA S I O N B I O L O G Y

475

TABLE 1

Numbers of Hawaiian Species Tabulated by the Bishop Museum’s Hawaii Biological Survey, Honolulu Taxon

Total Species

Endemic Species

Indigenous Species

Nonindigenous Species

Algae Other protists Angiosperms Other plants Fungi and lichens Cnidaria Annelida Mollusca Crustacea Insectaa Araneae (spiders) Acari (mites) Other Arthropoda Other invertebrates Echinodermata Lower Chordata Fish Amphibians Reptiles Aves (birds)b Mammalia

1118 1229 2142 639+ 3185+ 108 352 2073+ 1407 8151 248 656 97 2115 309 77 1245 7 29 183 44

104 ? 896 226 972 32 81 1096 ? 5245 144 168 42 ? 150 ? 157 0 0 63 2

? ? 107 ? ? 31 ? 848+ ? 124 ? 17 14 ? 159 ? 1033 0 3 65 23

? ? 1139 37+ ? 45 ? 129 ? 2782 104 471 41 ? 0 ? 55 7 26 55 19

Totals

25,615

9378

2424

4910

source: Eldredge and Evenhuis (2003). a Current estimates suggest there may be as many as 10,000 endemic insects and spiders in Hawaii, with many still undescribed. b Many birds are known extinct, and there are thought to have been previously 200–300 species.

PROPAGULES

FIGURE 1 In the Kerguelen Islands, the subantarctic wingless fly Anata-

lanta aptera (shown here) has suffered great impacts from the predatory beetle Oopterus soledadinus, introduced in 1912 from the Falkland Islands. This fly also interacts with the blowfly Calliphora vicina, accidentally introduced in 1978. Photograph by M. Laparie.

One of the primary determinants that appears to dictate the impact of invasive species is isolation. Isolation is a complex concept, incorporating not only distance, but also island size and topological diversity, the nature of the matrix, and the dispersal capabilities of the taxa of interest. The timing of colonization of invasive species is also associated with isolation. For example, extremely isolated habitats have only relatively recently experienced the impact of global homogenization, accentuating the changes currently observed. Here, we consider invasive species of islands generally, and focus on the processes of invasion and establishment.

476

I N VA S I O N B I O L O G Y

Propagules are individuals, or sets of individuals of a given species, that can give rise to new populations. Propagule pressure, or the rate at which propagules arrive or attempt to colonize, is an important predictor of the likelihood that a species will become established. Propagule pressure typically scales inversely with isolation, so that remote islands experience fewer propagules with only the most dispersive propagules being capable of reaching the most remote islands under natural conditions. For example, to account for the 8000 described indigenous terrestrial arthropod species in the Hawaiian Archipelago, it has been estimated that approximately 400 colonists must have arrived to initiate the various lineages, many of which diversified subsequently through adaptive radiation. If the age of the oldest high island, Kauai, of 4–6 million years is used as a reasonable older limit for most lineages, this number of colonists translates to a rate of successful establishment of approximately one every 10,000–15,000 years. Of course, many propagules that arrive are not successful, and thus it is difficult to estimate the magnitude of natural (historical) propagule pressure experienced by a given island. Even using the roughest estimates of propagule pressure, it is clear that recent growth of traffic and transportation routes have resulted in a huge increase in the number

of propagules reaching remote locations. For example, ballast water in ships is thought to account for the movements of 7,000–10,000 species simultaneously. In a study of ships arriving at 243 ports worldwide, expected invasion rates have been estimated at up to 2.94 × 10–4 species per km2 per year, with large increases predicted in so-called hotspots. A study of 44,000 air transportation routes estimates that numbers of passengers transported will increase by 8% per year. International air travel also figures prominently in the movement of invasive species— for example, 73% of pest interceptions in the United States Port Information Network database were at international airports. Modeling efforts including both passenger and freight traffic suggest the greatest risks of invasion are between airports that are connected by numerous highcapacity routes and that have similar climates. The relationship between establishment and propagule pressure can be important also for deciding how to manage invasive species. For example, greater propagule pressure may lead to a monotonic increase in probability of establishment, in which case any attempt to limit the number of propagules can limit further establishment. However, if the probability of establishment does not continue to increase with propagule pressure but levels off at some point, then additional effort to limit propagules after this point will have little effect. For ballast water, it has been suggested that moderate reductions in the per-ship-visit chance of introducing an invasive species can reduce the probability of invasion, and, by contrast, eliminating key port-to-port shipping connections will have negligible effects. This analysis suggests that reducing or eliminating organisms in ballast water will be effective, as for example in onshore treatment at particular ports or by onboard ballast water treatment. Propagules are only a small sample of their source populations. Accordingly, colonists of islands typically show a reduction in genetic diversity, a genetic bottleneck, relative to their source. However, it is not obvious that genetic bottlenecks limit the spread of invasive species. One reason for this is that if a population can increase in size rapidly following a bottleneck, much genetic variation can be maintained among and within individuals, even though the rare alleles have been lost. Some loss of alleles may actually contribute to invasive success. For example, different species of invasive ants, such as the Argentine ant Linepithema humile, demonstrate a tendency toward larger multi-queen unicolonial structures that facilitate cooperation among otherwise competitive colonies. It has been hypothesized that bottlenecks have resulted in a loss of alleles that genetically code for diverse cuticular hydrocarbons, which are responsible for colonies recognizing each other. Hence colonies

can grow largely unchecked in their new environment. In the red imported fire ant, Solenopsis invicta, the transition from monogyne form to a more damaging polygyne form appears to be associated with ecological constraints favoring cooperative breeding, but it is also controlled by genetic factors. Multiple colonization events can mitigate, or even increase the genetic diversity of founding populations if the propagules stem from sources that are genetically different. For example, the brown anole, Anolis sagrei, has invaded Florida multiple times from separate, genetically differentiated locations in Cuba, resulting in invasive populations that are more genetically diverse than source populations. When invasive populations of the same species are subdivided, both genetic drift and independent selection in the subpopulations may preserve genetic diversity. Other species may show greater genetic diversity in invasive populations than in source populations. For example, in the common myna bird, Acridotheres tristis, populations introduced to New Zealand, South Africa, Hawaii, Australia, and Fiji show higher genetic variation than in native populations from India. In the house finch, Carpodacus mexicanus, populations introduced into Hawaii and eastern North America have greater variation in amplified fragment length polymorphism (AFLP) markers than populations in their native western North America range, differences that are consistent with morphological variation. Invasive and colonizing often can adapt quickly to new environments, despite the effects of genetic bottlenecks. For example, in invasive Australian cane toads, Bufo marinus, populations at the edge of the expanding range have evolved longer legs and greater speed, suggesting that there is an advantage to founding a site first. One explanation for adaptive response to selection despite a recent genetic bottleneck is that novel interactions among genes in lowdensity and growing populations may enable populations to retain and even increase genetic diversity, upon which selection can then act. It has also been hypothesized that many invasive species have been the objects of selection over many generations for the ability to cope with low genetic diversity, and those that could not withstand low genetic diversity have been selected against. This hypothesis may help to explain why one of the best predictors for whether a plant species will be invasive is that it is invasive elsewhere. One way that an invasive species may overcome a potential loss in the ability to adapt genetically is though phenotypic plasticity. If a species can naturally expand its ecological range upon colonization, adaptation through natural selection may not be necessary. Plasticity associated with colonization has been better studied in plants than

I N VA S I O N B I O L O G Y

477

animals, and there is some evidence to support two observations, both of which deserve more study in a greater variety of systems: (1) invasive species may be more plastic than noninvasive or native species, and (2) populations in the introduced range of an invasive species may evolve greater plasticity than populations in the native range. While the role of plasticity in invasions is likely more important than is currently appreciated, there are limits to the novelty of an environment that an organism is capable of exploiting. For example, for insects, climate-matching models do fairly well in predicting the geographic limits of species, suggesting that at least some organisms are already at their physiological limits, particularly with respect to temperature. The recent interest in predicting spread of invasive species associated with global temperature change will no doubt shed more light on this topic. Multiple colonization events may also overcome demographic stochasticity—the loss of a population through chance events of survival and/or reproduction—in small invasive populations. In studies of conservation biology, it is thought that demographic stochasticity, coupled with Allee effects (see the following paragraph), are likely more important in determining the persistence of populations on an ecological time frame, and the same is likely true for invasive species in the early stages before populations have grown to sizable numbers. A difficulty in understanding the relationship between establishment and propagule pressure, and in understanding the early phases of a biological invasion in general, is that individuals can be more difficult to detect at low population sizes. Thus, invasive species can be resident and established in an area for many years while escaping detection. One reason that invasive populations may remain at a low level in the initial stages and then suddenly appear may simply be a consequence of exponential growth. Additionally, small populations may display so called Allee effects, in which their growth rate may be reduced at low population sizes, as for example, when there are so few individuals that it is difficult for an individual to find mates. Both these phenomena make it difficult to use population numbers to estimate how long an invasive species has been established, making predictive models of invasions more difficult. Islands may be the best places to study demographic processes associated with invasions. For example, because of the unrepresentative collection of species on islands and well-defined boundaries, invasions may be discovered at lower population sizes than might be possible in more complex ecological communities. For example, data on abundance and spread of the glassy-winged sharpshooter,

478

I N VA S I O N B I O L O G Y

FIGURE 2 The glassy-winged sharpshooter, Homalodisca vitripennis

(shown here on papaya), has been accidentally introduced into Hawaii, French Polynesia, and other Pacific Islands, where it is an urban nuisance because of the honeydew it produces, but is also an agricultural threat because of physical damage and its ability to vector a bacterial plant pathogen Xylella fastidiosa, which causes a variety of lethal scorchlike diseases in susceptible hosts. The Pacific invasions of the glassy-winged sharpshooter may have come from California; the insect is native to the southeastern United States and northern Mexico. Photograph by J. Grandgirard.

Homalodisca vitripennis (Fig 2), in Tahiti suggests that this leafhopper was discovered very quickly following invasion, prompting a successful program in biological control. It has also been possible to document the growth and spread of island populations of the cane toad, Bufo marinus, leading to the development of novel molecular genetic and statistical methods to investigate demographic history of invasive species. VECTORS AND TRANSPORTATION

Many native species made use of vectors, or biological transportation agents, to reach islands naturally. For example, crab spiders, snails, and multiple plant lineages likely arrived on remote islands with birds. Likewise, the use of vectors and analogous transportation vehicles (noted in the preceding section) is characteristic of many invasive species, allowing them to spread to habitats and regions they would otherwise not be able to reach. Because the movement of the vector or other agent of transportation determines in large part, or completely, the movement of associated invasive species, management often focuses on the biology of the vector or mode of transportation. Propagules and/or their vectors often follow well-defined pathways of introduction, though the pathways can take many forms. For example, invasive weeds are often found along routes used by human vectors, such as hiking trails. In the marine realm, ballast water is a well-known pathway for marine invasions worldwide (see the preceding section), as are ship hulls for fouling organisms. Similarly, the transport of fruit and vegetables is associated with movement of agricultural pests, as for example the introduction of tephritid flies, including the Mediterra-

nean fruit fly, Ceratitis capitata, associated with citrus and other crops. Likewise, nursery or horticultural stock trade is a common mode of introduction of pests, including the glassy-winged sharpshooter, which mostly likely came to Tahiti from California as inconspicuous eggs on horticultural plants. Avian malaria, Plasmodium relictum, arrived in Hawaii with alien birds and vector mosquitoes around 1825, which has resulted in eliminating native birds at lower elevations, where mosquitoes and parasites persist. Pathways associated with global trade, particularly regular shipping or airplane routes, can also be used to predict the arrival of potential invasive species. For example, invasive species managers in Hawaii are constantly vigilant for the brown tree snake, Boiga irregularis, which has been established in Guam for more than 50 years and has decimated the island’s indigenous bird fauna. Risk modeling approaches are also used to predict routes of transportation. In a model that included volume of commodities transported, infestation rates of pests, efficacy of inspection, and the probability of establishment, transport of mosquitoes on airplanes has been identified as the most likely route for West Nile virus (an encephalitis) to reach the Galápagos, which would have devastating effects on endemic avifauna but also on reptiles. FEATURES OF THE COMMUNITY

The presence or absence of other organisms is thought to play a significant role in the success of invasive species in novel habitats. Numerous studies have attributed the success of invasive species in new environments to having escaped many of their predators and parasites present in their indigenous range, although in some communities the invasive species may also pick up new predators and parasites. In a broad study of 26 host species of molluscs, crustacea, fish, birds, mammals, amphibia, and reptiles, the number of parasite species in native populations was double that in newly established populations, which also had lower rates of parasitism. A contributing factor may be that introduced populations are less likely to be infected upon arrival—propagules typically originate from small subsets of native populations and perhaps also from uninfected life-history stages. Invasive species have likely escaped pressures of direct competitors, though this effect is not as well studied. Over time, invasive species are likely to accumulate more predators, parasites, and competitors as a result of functional, numerical, or evolutionary responses. An increase in biological interactions over time may explain the downward slope of the often observed “hump” in population dynamics of invasive species, when, following an initial increase in popu-

lation size, numbers eventually decline. One example is the glassy-winged sharpshooter in Hawaii, whose numbers declined without intervention, likely as a result of parasitoids introduced for the biological control of other sap-feeding species. Although the foregoing examples indicate the importance of lack of elements of the biotic community in contributing to invasion success, the reverse can also be true. Accordingly, in some cases the establishment of one invasive species may facilitate the establishment and growth of a second (and third or more) invasive species, in which case the impact of the invasive species can be extreme and lead to a phenomenon termed invasional meltdown. For example, invasive ants have been shown to facilitate the invasion of sap-feeding insects, and invasive plants can facilitate invasive insect herbivores. In an example from Christmas Island, an invasion of the yellow crazy ant, Anoplolepis gracilipes, has decimated populations of a land crab, a keystone species in the native forest dynamics. The ants have also facilitated the increase of scale insects, whose honeydew causes sooty mold, reducing the health of the forest vegetation, eventually causing light gaps. The light gaps and a decrease in crab numbers have resulted in a decrease in other native species and invasions by weeds. A factor that is often associated with invaded habitats is ecological disturbance and/or open space. For example, following Hurricane Iniki on the island of Kauai in 1992 there was a large increase in the abundance of invasive plants, which colonized open space. One might ask why the endemic species do not themselves fill open space caused by disturbance? One part of the answer may be that island endemics frequently have reduced dispersal abilities compared to invasive species that might occupy similar niches. This difference in dispersal between indigenous and invasive species, coupled with the unrepresentative assembly of island species, may also help to explain why invasive species are thought to have such a large ecological impact on islands. Low species diversity may also encourage invasibility as a result of availability of open niches to a new spectrum on invasive species. Theory and empirical studies both suggest that historical accumulation of species in a community takes time, with time being dictated by isolation. Accordingly, particularly remote islands may take a long time to reach an equilibrium number of species (i.e., a number that stays consistent over time), and some islands may never reach such an equilibrium. Likewise, somewhat less isolated habitats that are relatively “young,” such as San Francisco Bay in California, have been heavily

I N VA S I O N B I O L O G Y

479

impacted by invasive species, and this may also be because the natural community had not reached equilibrium in terms of species diversity. The unrepresentative nature of taxa on islands also seems to be important in explaining the ability of invasive species to establish. For example, for plants, the current diversity in Hawaii is greater than what it was historically, and in New Zealand recent invasions have doubled the numbers of species from approximately 2000 to 4000. For freshwater fish in Hawaii the relative increase in numbers of species is even greater, with 40 alien species and only five indigenous species. This does not seem to be true for all taxa. For example, the total diversity of birds currently on the Hawaiian archipelago, including nonindigenous species, is approximately the same as it has been historically, despite the loss of bird diversity since human colonization. Phylogenetic relationships between invaders and indigenous taxa may be used to predict not only which species can colonize but also the relative impact of invasive species. For grasses in California, for example, the most invasive species are not closely related to indigenous species in comparison to the less invasive exotics. A similar analysis of the association between invasiveness of species and their phylogenetic similarity to native species has not been conducted for remote islands, though the unrepresentative nature of species on remote islands may make such a comparison difficult. THE FUTURE

As noted in other chapters, invasive and introduced species have large economic, environmental, and societal impacts on island communities. As a result, and because of the well-defined and compartmental nature of islands, invasive species on islands continue to be the object of basic and applied research and can provide much needed cross-fertilization among disciplines. Island systems can offer much to basic research in invasion biology. Several features are particularly worth noting. First, islands offer replicate systems in which to observe invasions and colonizations. Though invasions are natural experiments that are unplanned, observations of invasions can span a diversity of taxa under a range of conditions and time scales and thus complement more controlled manipulations. Second, island invasions are frequently noted very soon after arrival, so the initial stages can be studied. Thus, the early period of an invasion, which usually goes undetected, may be shorter on islands. Samples from the early stages are critical as benchmarks for studies of ecological and genetic change. Third, because of the clear bound-

480

I N VA S I O N B I O L O G Y

aries and typically restricted size of islands it should be possible to obtain accurate, and replicated, climate data that can then be used to improve the parameterization of models used to predict the spread of species in the context of global climate change. Such models will both contribute to an understanding of responses to global change and also be used to manage threatened island ecosystems. In sum, while the threats of invasive species on islands are real and daunting, islands can offer a promise of better understanding of processes over a range of temporal and spatial scales. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Ants / Biological Control / Dispersal / Introduced Species FURTHER READING

Carlton, J. T. 1999. The scale and ecological consequences of biological invasions in the world’s oceans, in Invasive species and biodiversity management. O. Sandlund, P. Schei, and Å. Viken, eds. Dordrecht, The Netherlands: Kluwer, 195–212. D’Antonio, C. M., and T. L. Dudley. 1995. Biological invasions as agents of change on islands versus mainlands, in Islands: biological diversity and ecosystem function. Ecological Studies 115. P. M. Vitousek, L. L. Loupe, H. Anderson, eds. Berlin: Springer-Verlag, 103–121. Eldredge, L. G., and N. L. Evenhuis. 2003. Hawaii’s biodiversity: A detailed assessment of the numbers of species in the Hawaiian Islands. Bishop Museum Occasional Papers 76: 1–30. Gillespie, R. G., E. M. Claridge, and G. K. Roderick. 2008. Biodiversity dynamics in isolated island communities: interaction between natural and human-mediated processes. Molecular Ecology 17: 45–57. Lockwood, J. A., P. Cassey, and T. Blackburn. 2005. The role of propagule pressure in explaining species invasions. Trends in Ecology and Evolution 20: 223–228. Pyšek, P., D. M. Richardson, J. Pergl, V. Jarošík, Z. Sixtová, and E. Weber. 2008. Geographical and taxonomic biases in invasion ecology. Trends in Ecology and Evolution 23: 237–244. Reaser, J. K., L. A. Meyerson, Q. Cronk, M. De Poorter, L. G. Eldrege, E. Green, M. Kairo, P. Latasi, R. N. Mack, J. Mauremootoo, D. O’Dowd, W. Orapa, S. Sastroutomo, A. Saunders, C. Shine, S. Thrainsson, and L. Vaiutu. 2007. Ecological and socioeconomic impacts of invasive alien species in island ecosystems. Environmental Conservation 34: 98–111. Richards, C. L., O. Bossdorf, N. Z. Muth, J. Gurevitch, and M. Pigliucci. 2006. Jack of all trades, master of some? On the role of phenotypic plasticity in plant invasions. Ecology Letters 9: 981–993. Sax, D. F., J. J. Stachowicz, J. H. Brown, J. F. Bruno, M. N. Dawson, S. D. Gaines, R. K. Grosberg, A. Hastings, R. D. Holt, M. M. Mayfield, M. I. O’Connor, and W. R. Rice. 2007. Ecological and evolutionary insights from species invasions. Trends in Ecology and Evolution 22: 465–471. Simberloff, D. 1995. Why do introduced species appear to devastate islands more than mainland areas? Pacific Science 49: 87–97. Vernon, P., G. Vannier, and P. Trehen. 1998. A comparative approach to the entomological diversity of polar regions. Acta Oecologica 19: 303–308.

IRELAND SEE BRITAIN AND IRELAND

ISLAND ARCS RICHARD J. ARCULUS Australian National University, Canberra

Island arcs are chains of concurrently or potentially active volcanic islands, consistently associated but displaced spatially more than 100 km from a deep-sea trench. Much of the eruptive activity is strongly explosive. Although some chains are strongly arcuate, others are linear. Adjacent to many island arcs in the western Pacific are back-arc basins floored by crustal spreading centers. Some arcs have associated nonvolcanic individual islands or island chains between the volcanic arc and trench, comprising uplifted, trench-accreted sediment. CONTEXT

Oceanic lithosphere created at zones of plate divergence (mid-ocean ridges) is returned to the Earth’s interior at sites of plate convergence, specifically at deep-sea trenches, which mark the surface trace of gently to steeply dipping subduction zones. During its interactive exposure at Earth’s surface, oceanic lithosphere acquires an imprint of chemical exchange with the oceans and a superstrate of sediment derived predominantly from biological and hydrothermal activity in the oceans plus detritus transported from continents. This diverse rock and sediment package is variably recycled into the Earth’s mantle, leading to a complex series of processes including energetic earthquake activity and fluid release accompanying metamorphic changes to the package; the fluids rise buoyantly into the wedge of mantle overlying the subducted lithosphere (or “slab”), triggering magma generation through partial melting. The magma rises towards the Earth’s surface, eventually forming the chains of volcanoes known as island arcs. The world’s greatest ocean deeps, biggest gravity anomalies, largest earthquakes, most explosive volcanic eruptions, and highest mountains all occur at zones of plate convergence and subduction. In the simplest terms, island arcs are the surface expression of a complex spectrum of magmatic and tectonic processes triggered by plate subduction at zones of convergence. In addition, the geochemical characteristics of island arc magmas most closely match those of the continental crust, unlike mid-ocean ridge and hotspot types; accordingly, a persistent stimulus for island arc research has been the potential link with genesis and evolution of the continents.

Island arcs differ in crucial ways from other chains and clusters of islands produced by volcanic activity, such as those of Hawaii or Galápagos. The magmas constructing the latter are the partial melt products of isolated, individual plumes of mantle material rising from thermal boundary layers within the Earth (e.g., the core–mantle or lower– upper mantle transition zones). Plumes (or “hotspots”) are generally independent of zones of plate divergence or convergence, and consequently their magmatic products may be passively transported away from the active plume locus by ambient motion of the plate they happen to be erupted onto. In contrast, island arcs develop parallel to sites of plate subduction, and numerous loci of magmatic activity are concurrently formed over distances that can extend in individual arc segments for thousands of kilometers. Although portions of a previously active arc may be abandoned as a “remnant arc” by zones of crustal spreading, rifting an arc apart during “back-arc basin” formation, regular progressions from fringing reef to atoll are generally not well developed on the subsiding remnant arc edifices. Historically, the development of arcuate chains of islands in some western Pacific arcs attracted attention: In 1903 W. J. Sollas demonstrated that such an arc could trace an outcropping planar fault on the Earth’s spherical surface, and in 1931 P. Lake suggested these fault surfaces were defined by continentward-dipping zones of earthquake foci, documented the same year by K. Wadati for Japan. Note, however, that many so-called arcs, such as the Izu–Bonin, Solomons, New Hebrides, and Tonga– Kermadec, are linear geographically, presumably related to the local morphology of the subducted slab. GEOGRAPHIC DISTRIBUTION

There is a gradation globally between island arcs apparently formed in intra-oceanic settings (Izu–Bonin–Mariana, New Hebrides, Tonga–Kermadec), those developed on fragments of continental crust that have migrated away from an adjacent continental land mass (Japan, Philippines, New Zealand), and those subduction-related volcanoes active at the margins of continents (Cascades in North America, Central America, and the Andes in South America) (Fig. 1). Specific gradational examples include the Aleutians, which extend onto the Alaskan Peninsula and Mainland; the Kurile chain, which extends between Kamchatka and Japan, a portion of continental crust that migrated eastward from the East Asian continental plate about 20 million years ago; and the Kermadec Arc, which strikes southward onto the continental crust of New Zealand and which migrated away from the margin of Antarctica some 80 to 55 million years ago.

ISLAND ARCS

481

Ka

Al

Ku C Ho IB

R

Mx LA

M

P

CA Sa

Ha

Su J

Ba

Bi-NB

NAn

So NH

CAn

To

Ke NZ

SAn Ao

Ae

Tu

I

SS

FIGURE 1 Global distribution of island arcs based on Global Topography base (http://topex.ucsd.edu/marine_topo/mar_topo.html); main panel

centered on the Pacific with inset of Mediterranean and Middle East. Abbreviations: Ae, Aegean; Al, Aleutians; Ao, Aolian; Ba, Banda; Bi-NB, Bismarck–New Britain; C, Cascades; CA, Central America; CAn, Central Andes; Ha, Halmahera; Ho, Honshu; I, Iran; IB, Izu–Bonin; J, Java; Ka, Kamchatka; Ke, Kermadec; Ku, Kurile; LA, Lesser Antilles; M, Mariana; Mx, Mexico; NAn, Northern Andes; NH, New Hebrides; P, Philippines; R, Ryukyu; Sa, Sangihe; Su, Sumatra; So, Solomons; SAn, Southern Andes; SS, South Sandwich; To, Tonga; Tu, Turkey.

The Andean continental arcs are connected by transform faults to two island arcs in the western Atlantic: (1) in the north of the Andes to the Lesser Antilles Island arc and the subduction of the Atlantic portion of the North American plate beneath the Caribbean plate and (2) in the south of the Andes to the South Sandwich Island Arc, where the Atlantic portion of the South American plate is being subducted beneath the Scotia plate. Although the Sumatra–Java–Banda arc system appears superficially to comprise particularly large subaerial and emergent chains of volcanoes forming extensive islands, in fact these eruptive centers are developed on the margin of a mostly submerged continental crust extended southward from the Asian plate (Fig. 1). Other arcs include the Aeolian and Aegean in the Mediterranean, involving subduction of the African beneath the Eurasian plate. Eastward, the further collision between the African and Eurasian plates is marked by a limited chain of continental arc-type volcanoes striking from Turkey southeastward to Iran. But eruptive activity in arcs generally ceases where prolonged collision between extensive continental portions of plates has occurred. For example, much of the subduction zone active between India and Asia lacks normal arc volcanism. Elsewhere, volcanic arc activity has also ceased where subduction of oceanic lithosphere is taking place at a relatively shallow

482

ISLAND ARCS

dip, as is the case currently beneath much of Peru and parts of central Chile. REPRESENTATIVE CROSS-SECTION

Although the structure of plate convergence zones along strikes (i.e., parallel to the deep-sea trench) is fundamentally important, as a first approach, a two-dimensional cross-section is useful for graphically illustrating the physical and chemical processes involved in arc genesis. A representative composite cross-section (Fig. 2) shows the following important features: SUBDUCTING SLAB

The subducting slab (feature 1 in Fig. 2) at the trench comprises a superficial sediment layer (Layer 1; depending on specific oceanic setting, these sediments can include carbonate ooze, siliceous ooze, hydrothermal deposits, and continent-derived muds and sands) up to several hundred meters thick; Layer 2, formed by basaltic pillow lavas underlain by feeder dikes that are tapped from Layer 3, the gabbroic and olivine–clinopyroxene-dominated, intra-crustal magma chamber cumulative fractions. Depending on the magma production rate at a mid-ocean ridge, the igneous crustal thickness can range between 0 (at amagmatic transform fault–ridge intersections) and about 10 km.

Remnant Arc

6 Backarc Basin

Volcanic Arc

5

Forearc

4

Serpentine Diapir Accretionary Prism

3

Upper Crust

an

Lower Crust d

e at

dr

Hy

Fluid Fluxed Melting

8

2

tle

M

7

1

e

er

ph

s ho

t

g

Li

in ct

Advecting Wedge

u

bd

Su

FIGURE 2 Schematic cross-section of an island arc from trench to back-arc basin. Numbers refer to the features described in the section titled

“Representative Cross-Section.” Stars indicate major thrust-mechanism earthquakes in the shallower (seismogenic) portion of the Wadati–Benioff zone of earthquakes.

Below the crustal layers is the residual upper mantle from which the basaltic portion of the ocean crust is derived. Alteration of the crust and underlying mantle by circulation of seawater produces hydrated minerals including clays, epidote and amphibole groups, and serpentine in the case of olivine-pyroxene-dominated lithologies. ACCRETIONARY PRISM

An accretionary prism (feature 2 in Fig. 2) has developed in the fore-arc of only some systems; it consists of sediments and sporadic igneous crustal fragments, including major portions of oceanic seamounts. Packages of sediment are scraped off and sequentially underplated by thrust faulting; uplift of the prism as more and more sediment is underplated can develop to the point of emergence, forming islands such as Barbados between the southern Lesser Antilles Island arc and trench; Simelue, Nias, and Siberut between the Sumatra volcanic arc and adjacent trench; and Kodiak Island in Alaska. FORE-ARC

In those arcs lacking an accretionary prism, the fore-arc (feature 3 in Fig. 2) is dominated by igneous rocks and volcaniclastic-rich sediments shed from the arc itself. The igneous rocks can include the crust upon which the active arc has been built and arc crust constructed during early stages of arc development. Emergent examples of the latter include the islands of Shikotan (Kurile), Chichi- and Hahajima (Bonin), Tinian–Saipan–Guam (Marianas), and Vava‘u–Tongatapu–Eua (Tonga). Some nonaccretionary fore-arcs, such as those of the Izu–Bonin– Mariana system, are penetrated in diapiric manner by major seamounts dominated by serpentinized rocks and mud,

derived by serpentinization of the shallow mantle wedge through fluid release from the subducted Pacific slab. VOLCANIC FRONT

The most trenchward locus of eruptive volcanic activity (sometimes known as the “volcanic front,” feature 4 in Fig. 2) comprises the volcanic arc sensu stricto; volcanic islands are the emergent tips of large, steeply sided submarine structures, typically of several thousand meters relative elevation between summits and local crustal basement. It is not clear why focused spacing of magma supply, leading to island formation rather than a continuous eruptive sheet is developed. Globally, there are major differences between the total number and proportions of emergent and submerged arc volcanoes in any given arc. In the past decade, detailed multi-beam sonar swath mapping accompanied by hydrothermal plume surveys and rock dredging has documented the true incidence of volcanic centers in a number of island arcs. For example, in the case of the Aleutians, the majority of individual volcanoes are emergent with an along-strike spacing of ~100 km. In contrast, in the Tonga–Kermadec system, there are only about 12 subaerial and more than 80 submerged volcanic centers with an average spacing of ~40 km. Some ephemeral islands in this region are instructive in terms of understanding the stabilization required of emergent submarine structures, particularly by blanketing lava flows. It is known that the crustal loads exerted by larger volcanic structures tend to capture more of the rising magma flux from the mantle— larger edifices grow at the expense of smaller vents. Thus, subaerial emergence is controlled by several major factors: the localized magmatic flux; the bathymetric depth of the plate surface overriding the subduction zone reflecting

ISLAND ARCS

483

overall compressional, tensional, and gravitational force balance; the extent of collapse and degradation of the arc volcanoes; and the degree of tectonic subsidence or uplift of the arc accompanying formation of back-arc basins. Along the strike of the volcanic front in the Izu–Bonin Arc are also coupled variations in degree of emergence, crustal thickness, and dominant magma composition. For example, the islands in this arc tend to be dominated by basaltic eruptions overlying relatively thick (~20-km) crust, whereas the submerged structures include many rhyolitedominated calderas overlying thin (~15-km) crust. REAR-ARC CHAINS

Cross- and parallel rear-arc chains of volcanoes (feature 5 in Fig. 2) extending orthogonal to and along strike with the volcanic front. In a number of arcs such as the Izu– Bonin–Mariana system, there are subaerial (e.g., Niijima and Kozu-shima) to submarine chains of seamounts extending at a high angle away from the arc front. In the case of Izu–Bonin, the volcanoes are young toward the arc front and may relate to arc migration following back-arc basin formation. In other intraoceanic island arcs, there are volumetrically minor but volcanically active edifices, some of which form islands (e.g., Alaid, Kuriles) and others of which are submerged. Cross-chains can be quite localized in development, restricted to a small region of the arc, as is the case with the volcanically active Willaumez Peninsula and Witu Islands in the New Britain Arc of Papua New Guinea. In the case of major islands such as Honshu in Japan and Java in Indonesia, there are prominent volcanoes distributed irregularly further from the trench but parallel to the strike of the volcanic front, some of which have experienced major catastrophic eruptions in historical time, such as Tambora (AD 1815 eruption) on the island of Sumbawa. In general, the magma types forming cross- and arc-parallel rear-arc chains tend to be more alkali (Na + K)–rich at any given bulk SiO2 content than those forming the volcanic front.

western Pacific have magmatically active zones of ocean floor spreading and new crustal creation, similar in many morphological respects to the major mid-ocean ridges. Actively spreading basins include the Mariana Trough, Manus Basin, Andaman Sea, Coriolis Troughs–North Fiji Basin (with several spreading loci), Lau Basin (with several spreading loci)–Havre Trough, and Scotia Sea. The only island in any of these basins is Barren Island in the Andaman Sea. The duration of spreading in back-arc basins appears to be restricted from a few to ~15 million years, and several other inactive basins in the western Pacific are known. These include the Kurile back-arc basin, Sea of Japan, West Philippine Basin, Shikoku–Parece Vela Basins, and South Fiji Basin. In a number of cases, rifting occurs within the volcanic arc, leading to development of a spreading system splitting a remnant arc from the sustainedly active volcanic front. Cessation of magmatic supply to the remnant arc generally leads to subsidence below sea level as with the cases of the remnant Kyushu– Palau Ridge–active Izu-Bonin-Mariana Arc and remnant Lau Ridge–Tofua Arc pairs. In the latter case, however, the Lau Group of islands at the northern end of the Lau Ridge are still emergent. In other cases, rifting takes place totally behind the arc, leading to migration of the entire arc structure and no remnant arc formation (e.g., New Hebrides). During episodes of back-arc basin initiation, volcanic front activity may temporarily cease; it appears that the back-arc basin rift captures the total mantle wedge–derived magmatic flux. This is currently the situation in the case of the volcanic front in the northern Tofua Arc, adjacent to the nascent spreading occurring in the Fonualei Rifts, where volcanic structures north of the volcanic island of Fonualei are planated ~30 m below current sea level and are carbonate capped. In the case of the Mariana Arc, the volcanic front became reestablished on newly created back-arc basin crust, requiring a longer period of time to emerge above sea level for a given arc magma flux than would have been the case if they were constructed on the bathymetrically shallower fore-arc.

BACK-ARC BASINS

Although island arcs are located in zones of overall plate convergence, the most significant driving force behind plate motions is slab pull in subduction zones, and in many of these systems, no horizontal compressive stress is transmitted by the downgoing to the overriding plate. Slab roll-back is common in the western Pacific, resulting in advance of the trench toward the incoming plate, fore-arc collapse, and extension in arc and back-arc, leading in some instances to spreading and formation of a basin (feature 6 in Fig. 2). Many of the island arcs of the

484

ISLAND ARCS

DOWN DIP IN THE SUBDUCTED SLAB

Increases in pressure and temperature control a sequence of metamorphic dehydration reactions, accompanied at depths of less than ~30 km along the slab–overriding plate interface by large thrust earthquakes (feature 7 in Fig. 2). Some of these may propagate to the sea floor, triggering large tsunamis, as was the case with the December 2004 earthquake off the northwestern coast of Sumatra. The term Wadati–Benioff Zone is given to the continuum of a dipping, shallow-to-deep plane of earthquakes from the

deep-sea trench, beneath the island arc and on to great depths (~600 to 700 km, transition zone in the mantle). The descending slab imposes an inverted temperature gradient in and forces advective motion of the wedge. Further down the slab dip, a complex and continuous series of dehydration reactions, involving breakdown of phases such as zoisite, epidote, lawsonite, amphibole, and mica, releases fluid into the overlying mantle wedge. Where sufficiently detailed studies can be made, double zones of earthquakes are found, one close to the slab–wedge interface and the other ~30 km deeper within the slab. FOCUSED MAGMA CORNER

Motion of the advecting wedge results in focusing of the magma by porous flow to a “corner” (feature 8 in Fig. 2). It is an empirical fact that with some rare exceptions, the volcanic front is located on average about 100 km above the subjacent slab–wedge interface, and possibly above the corner of focused magma flow. Magmas do erupt above greater depths to the slab. In general, and unlike the situation at mid-ocean ridges and hotspots, where adiabatic decompression is the prime partial melting trigger, it is believed that fluid-fluxed partial melting of the mantle wedge is particularly important in arcs. A very wide variety of magma compositions is erupted globally, ranging at the basaltic end from low-K2O olivine tholeiite through to high-K2O feldspathoid-bearing compositions. Where cross-arc magmatism exists, a zonation from low- to high-K2O types with increasing depth to slab is generally observed. Other restricted magma types include high-MgO, low-SiO2 picrites. Rare high-MgO intermediate (~55–60 wt% SiO2) types called boninite are particularly found accompanying arc inception and more rarely in active arcs developed above strongly basalticmelt depleted mantle wedges, such as beneath northern Tonga. Arc basalts generally contain significantly more H2O (~2 to 6 wt%) than is the case with mid-ocean ridge (less than 0.2 wt%) and hotspot (less than 1 wt%) basalts. With reduction in ambient pressure during ascent of the magma to the Earth’s surface, exsolution of an H2Orich volatile phase drives the characteristically explosive volcanism of arcs. Return of H2O to the Earth’s atmosphere is part of a large-scale H2O recycling process in the Earth from hydrothermal alteration at a mid-ocean ridge through fixation in and subduction of oceanic lithosphere, and fluid-fluxed melting of the mantle wedge to explosive submarine or atmospheric emissions. Other volatile compounds are exsolved in this gas phase including CO2 and sulfur- and halogen-bearing molecules. Recognizing the importance of H2O in the genesis of arc magmas and

its plausible link with genesis of the continental crust, in 1983 I. H. Campbell and S. R. Taylor coined the aphorism “no water, no granites—no oceans, no continents.” Numerous factors control the composition of the magma erupted in arcs: composition of the mantle wedge, particularly with respect to previous melting history; depth of melting of the wedge; composition of the fluid addition from the slab, which is a function of subducted sediment characteristics; type of alteration of igneous lithologies; extent of lithologic mixing in the surface layers of the slab (the fluids may range from H2O-dominant through to H2O-bearing silica-rich melts and possibly supercritical); ascent paths and degree of interaction between magma and host lithologies during ascent; extent and depth of fractional crystallization of the magma; and extent of magma mixing. The most important process generating the spectrum of basalt–andesite–dacite–rhyolite volcanic rock compositions is fractional crystallization of olivine–pyroxene–plagioclase–spinel (fom Cr- to Ferich) assemblages. These cumulative assemblages can be found as relic plutons in deeply exposed arc crustal sections. Although the low-MgO, intermediate-SiO2 rock type called andesite is characteristic of many arcs, it is not volumetrically always the most abundant, particularly in the submarine realm of intraoceanic arcs where basalt is prominent. INCEPTION, DEVELOPMENT, AND DEMISE

Several decades of research combining land-based studies with marine research voyages, deep-sea drilling, and submersible investigation have established the history of a number of western Pacific arcs in particular, from inception through to current stages of development. For example, it is known that the Izu–Bonin–Mariana and former Solomons–New Hebrides–Tonga–Kermadec (“Vitiaz”) systems were initiated ~50 million years ago. Reconstruction of this episode requires closure of subsequently developed back-arc basins (e.g., Mariana Trough, Parece Vela, and Shikoku) juxtaposing the remanant arc of the Kyushu–Palau Ridge with the current amagmatic forearc. Paleomagnetic investigations have shown a clockwise rotation of ~90° of a formerly east-west trending system at equatorial latitudes for Izu–Bonin–Mariana since initiation. Inception through cannibalization of transform faults accompanying a change in plate motion has been a popular hypothesis, constituting an example of forced initiation. It is also recognized that gravitational instability of oceanic lithosphere more than ~10 million years old with respect to the underlying asthenosphere might plausibly lead to spontaneous subduction initiation.

ISLAND ARCS

485

Although the present location of systems such as Izu– Bonin–Mariana and Tonga–Kermadec appear to be remote (hence the term intraoceanic) from the nearest continents, it is important to bear in mind the evolution of these systems. For example, a series of continental fragments of eastern Australia (Lord Howe Rise, Norfolk Ridge) have been rifted off during the late Cretaceous through Tertiary periods, accompanied by a series of arc–back-arc developments along their eastern and northern margins. Similarly, the Kyushu–Palau Ridge is developed across a series of Cretaceous arc–back-arc systems that may have been proximal to Asia. The important point is that nominally intraoceanic arcs may have been initiated much closer to continental masses. We have few examples of subduction initiation in truly remote intraoceanic settings: the New Britain–New Ireland–Bougainville–Solomons–New Hebrides may be one such. Other island arcs are clearly developed on continental fragments such as Japan; during the Miocene epoch (20 million years ago), fragments of modern Japan migrated away from eastern Asia with the development of a back-arc basin (Sea of Japan). Some of this archipelago’s most frequent earthquakes are on the Sea of Japan side of Honshu, marking the inception of a new subduction zone and closure of the Sea. Japan is destined to collide with its formerly rifted parent land mass. More generally, despite the inherent gravitational instability of oceanic lithosphere, it is plausible that subduction initiation is propagated from elsewhere as a type of “tectonic infection,” emphasizing the importance of the third dimension in our cross-sectional depiction of arc systems. We know that initial development of the Izu–Bonin– Mariana system was submarine in an extensional setting involving extensive pillow lava–dike (including boninite) emplacement. Ocean drilling has shown that subaerial, explosively eruptive conditions were temporarily reached a few million years after inception. Submarine growth stages are likely characteristic of many island arcs, as a growing body of survey data in the western Pacific has revealed. But we have sparse current examples of subduction inception: The Hjort Trench–Macquarie Ridge complex south of New Zealand is a possibility, but detailed swath mapping has failed to reveal volcanic activity, with the exception of a single active subaerial volcano (Solander). The demise of arc systems includes the cessation of eruptive activity on remnant arcs and their general submergence. Changes in general plate tectonic frameworks can also lead to subduction and arc demise, as in the case of the Greater Antilles (Cuba–Hispaniola–Puerto Rico), where relatively large emergent islands remain. Another example is the South Shetlands off the northern coast

486

I S L A N D B I O G E O G R A P H Y, T H E O R Y O F

of the Antarctic Peninsula, where sparse volcanic activity continues (e.g., Deception Island). Collision with a neighboring continent also leads to arc demise: Examples include multiple collisions of arc systems with western North America and the southern margin of central Asia during the Mesozoic. An ongoing example is the collision of the West Bismarck Arc with mainland Papua New Guinea. Volcanic activity is terminating progressively in this collision zone from northwest to southeast. In conclusion, island arc systems are dynamic constructs with geologically ephemeral subaerial island development at varying distances from large land masses. Peripheral addition to continents and reworking of the constituent lithologies through metamorphism and igneous processes are important in continued continental evolution. The mobility of arc systems from initiation through migration and potential collision with continents is clearly significant from biogeographic perspectives, and there is much to learn from the combination of such studies with geological understanding. SEE ALSO THE FOLLOWING ARTICLES

Antilles, Geology / Hydrothermal Vents / Island Formation / Japan’s Islands, Geology / Kurile Islands / Oceanic Islands / Plate Tectonics FURTHER READING

Abers, G. A., and P. E. van Keken. 2006. The thermal structure of subduction zones constrained by seismic imaging: Implications for slab dehydration and wedge flow. Earth and Planetary Science Letters 241: 387–397. Arculus, R. J. 2004. Evolution of arc magmas and their volatiles. Geophysical Monograph 150: 95–108. Grove, T. L., and R. J. Kinzler. 1986. Petrogenesis of andesites. Annual Review of Earth and Planetary Sciences 14: 417–454. Stern, R. J. 2002. Subduction zones. Reviews of Geophysics 40: doi:10.1029/2001RG000108. Tatsumi, Y., and S. M. Eggins. 1995. Subduction zone magmatism. Oxford, UK: Blackwell.

ISLAND BIOGEOGRAPHY, THEORY OF JOSÉ MARÍA FERNÁNDEZ-PALACIOS La Laguna University, Tenerife, Spain

The theory of island biogeography, developed by Robert H. MacArthur and Edward O. Wilson successively in 1963 and 1967, argues for the existence of a dynamic balance in species richness on islands, as a function of the addition

of species through the immigration of propagules to the island, plus any speciation within it (dictated by the degree of isolation from the mainland), and of species extinction from the island (dictated by island area). The result of these opposing forces, given enough time, is a dynamic equilibrium in which the species number remains approximately constant through time but species composition is continually changing. As a result of its quantitative approach and predictive power, the theory has transformed island biogeography into a mature scientific discipline. It rapidly achieved paradigmatic status, with numerous studies setting out either to confirm or reject its predictions, focused both on a large variety of taxonomic groups (plants, vertebrates, invertebrates, protozoa) and island types (real and habitat islands). It has thus inspired a substantial advance in our knowledge of insular biotas and processes. Finally, it has provided the foundation of a substantial (if controversial) body of theoretical work on the design/implications of protected area networks. FUNDAMENTAL PRINCIPLES OF THE THEORY

The theory of island biogeography was developed in large measure to account for apparently systematic variation in species–area relationships. Basically, the theory considers that there are two major processes shaping the species richness that a given island can carry. How they vary is described in a straightforward graphical model (Fig. 1), considered by some a key ingredient in the adoption of the theory. The first process is the immigration (arrival) E small

Rate

I near

E large Tns

I far

Tfs Sfs

Sfl Sns

Snl

P

Number of species present

FIGURE 1 A version of MacArthur and Wilson’s (1963, 1967) equilib-

rium model of island biogeography, showing how immigration rates are postulated to vary as a function of distance, and extinction rates as a function of island area. The model predicts different values for S (species number), which can be read off the ordinate and for turnover rate (T ) (i.e., I or E, as they are identical at equilibrium). Each combination of island area and isolation should produce a unique combination of S and T. To prevent clutter, only two values for T are shown. Source: Whittaker and Fernández-Palacios (2007).

of a propagule of a species that is new to the island, and its rate depends on the island’s isolation (the distance from the mainland source). Nearer islands are easier to reach by mainland species than farther ones, therefore the immigration rate curve (I ) tends both to flatten with increasing islands isolation, and to decline exponentially through time as initially empty islands fill up with species. Hypothetically, the immigration rate would decline to zero if the islands were near enough to the mainland to enable the whole mainland species pool (P ) to disperse to them. However, because islands are smaller than mainland sources this point is never reached because of the counterbalancing second process, extinction. The theory of island biogeography posits that the extinction, or total disappearance, of a species from an island is dependent on the island area. So, larger islands can support larger populations than smaller ones, and as extinction risk is inversely related to population size, the extinction rate curve (E ) tends both to flatten with increasing island area and to increase exponentially as initially empty islands (where E = 0) fill up with species. The increase of E with species richness is explained by the fact that the higher the number of species on a given island, the smaller the population sizes that can be maintained for a given species, hence the higher the extinction risks. When the trajectories of the immigration and extinction rates (as the ordinates) for a given island with a specific area and isolation are plotted on a graph against species richness on the abscissa (Fig. 1), the projection to the abscissa from the point at which the two curves intersect defines the species richness that this island can carry, whereas the projection from this point to the ordinate defines the turnover rate (T), or number of species extinguished and replaced per unit time. Thus, species richness, as well as immigration, extinction, and turnover rates, are island-specific parameters that vary with the island’s area and isolation in a dynamic equilibrium, where species richness tends to remain constant through time although species composition continues to turn over. Each combination of island area and isolation should produce a specific combination of species richness and turnover rate. Thus, the theory predicts that large, near islands tend to have higher species richness and lower species turnover than small, far islands, whereas small, near islands as well as far, large islands should have intermediate values (Fig. 1). Finally, colonization, defined as the relatively lengthy persistence (e.g., through at least one life cycle) of an immigrant species on an island, can be plotted through time to obtain the colonization curve, or temporal change of numbers of species found together on an island (Fig. 2). The colonization curve starts from zero, when the island

I S L A N D B I O G E O G R A P H Y, T H E O R Y O F

487

Number of species present

Rate

I

E

Colonization curve

Time

Time

FIGURE 2 Integration of immigration and extinction curves (left)

should theoretically produce the colonization curve as shown (right). Source: Whittaker and Fernández-Palacios (2007).

has not yet been colonized by any species, then rises rapidly as the island accumulates new species (notwithstanding the gradual rise in extinction events) until the curve flattens approaching asymptotically the island’s species carrying capacity, as determined by its area and isolation. OBJECTIONS AND EMBELLISHMENTS TO THE THEORY OF ISLAND BIOGEOGRAPHY

The theory of island biogeography assumes that all species are equal in their probabilities of immigrating onto the island or of going extinct once there. With the development of Hubbell’s Unified Neutral Theory of Biodiversity and Biogeography, there has been a resurgence of theory regarding the importance of such neutral processes in dictating species composition in a given community. It appears that species do differ in their chances of colonizing and persisting on an island. Although MacArthur and Wilson’s monograph (1967) includes a chapter called “Evolutionary Changes Following Colonization,” the mechanism through which speciation can substitute for immigration is not well developed, and there is general agreement that the theory is more readily applied to islands driven by “ecological” processes, where

A

In

B

E

the frequency of immigration events precludes speciation, than for very remote islands, where species diversity is dictated mainly by “evolutionary” processes and in situ speciation is more frequent than immigration as the source of new species. Several approaches have been developed to include more information about speciation, the third main biogeographical process, together with immigration and extinction, within the framework of the theory of island biogeography (M. V. Lomolino, L. R. Heaney, etc.), in order to increase the generality of the theory. Another limitation in the model is that isolation is considered to dictate rates of immigration only. However, island isolation has been shown also to play an important role in influencing extinction rates through a phenomenon known as the rescue effect (Fig. 3): the supplementary immigration from the mainland of propagules of species present on the island in small population sizes and that would otherwise go extinct (such supplementation does not count as immigration if the focal species is rescued prior to actually going extinct). This possibility was noted by MacArthur and Wilson but was not included in the model, yet the effect can be important on small islands near to the mainland, thus modifying the extinction rate, and equilibrium point. In addition, area is considered to dictate only extinction in the model, although MacArthur and Wilson noted the potential importance of island area in influencing the immigration rate as well. Through the latter phenomenon, known as the target effect (Fig. 4), larger islands provide easier targets for passively dispersing propagules, such as windborne and waterborne plants, thus increasing their immigration rates. These two effects undermine the predictive power of the simplified model in Fig. 1. MacArthur and Wilson also recognized that biologists can rarely, if ever, be certain of recording all immigration and extinction events in real-world systems, especially if large islands are considered, and thus species turnover calculations for large islands can contain important

In

Ef

FIGURE 3 The rescue effect is

the reduction in the extinction

Rate

Rate

rate of near islands versus

Tn If Tf

distant ones. Whereas the MacArthur and Wilson model

If

predicts higher turnover on near islands (A), the rescue

Tf Tn

En

effect may increase turnover on more distant islands (B). Tn is the turnover rate on the

Sf

Sn

Number of species on island (S)

488

I S L A N D B I O G E O G R A P H Y, T H E O R Y O F

P

Sf

Sn

Number of species on island (S)

P

near island; Tf is the turnover rate of the far island. Source: Gotelli (2001).

A

Es

I

B

FIGURE

Ii

Es

4 The

target

effect is the increase in the immigration rate on

Ts

Ei

Ti

ones. Whereas the Mac-

Rate

Rate

large islands versus small Arthur and Wilson model predicts higher turnover

Ei Ti Is

Si

P

Number of species on island (S)

small

islands

(A),

the target effect may increase the turnover on large islands (B). Ts is the

Ts Ss

on

Ss

Si

P

Number of species on island (S)

turnover rate on the small island; Tl is the turnover rate on the large island. Source: Gotelli (2001).

biases. Two main problems arise when trying to calculate species turnover rates: cryptoturnover and pseudoturnover. Cryptoturnover is the real turnover of species not detected because of too great a gap between census intervals, as for instance, when a species undergoes extinction and later re-colonization, or vice versa, in the time interval occurring between two consecutive inventories. The effect leads to underestimation of the real turnover rate. Pseudoturnover, conversely, is the apparent disappearance and reimmigration of species in consecutive surveys when they were actually present throughout, or alternatively, were only ever present as vagrant individuals. Such incomplete surveying leads to overestimation of the real turnover rates. Attempts have been made to quantify both of these sources of error, but they are inherently hard to estimate precisely, especially for larger islands. It has been argued that habitat diversity and not area per se is the true determinant of island species richness, because more habitats will offer more opportunities for the colonization of species differing in their ecological requirements. Although this is undoubtedly true, it is also the case that habitat diversity is generally correlated with island area. Indeed, it was partly for the lack of adequate data on habitat diversity (still a limitation today) that MacArthur and Wilson focused only on area in their monograph. Their specific mention of this point makes clear that they used area in the theory as a measure of the combined effects of area per se and habitat diversity. Finally, it has been argued that some insular systems do not achieve equilibrium, even after extended periods of time. Three alternative states have been recognized. First, “static” nonequilibrium systems include those where species losses (such as those due to postglacial isolation of mountaintop systems preventing further immigration across arid lowlands) are occurring, but so slowly as to be effectively unmeasurable on ecological time scales. Second, “dynamic” nonequilibrium systems are those that are frequently impacted by

extreme events, such as volcanic eruptions or hurricanes, which resets the community development iteratively, preventing the attainment of equilibrium. Third, “static” equilibrium systems are those archipelagoes that, over ecological time scales, have a clear species–area relationship (consistent with the theory of island biogeography) but with no turnover (not consistent with the theory), as for instance appears to apply to some oceanic island avifaunas. These different scenarios have allowed fine-tuning of the theory of island biogeography, with an expanded framework. THE THEORY OF ISLAND BIOGEOGRAPHY AND CONSERVATION

One of the main applications of the theory of island biogeography outside its academic context has been within the field of conservation biology. The increasing worldwide anthropogenic fragmentation experienced by continental ecosystems has transformed once continuous habitats into complex landscapes containing many patches of relict habitats, differing in size, shape, isolation, or degree of disturbance, surrounded by a more or less penetrable matrix of arable land, pasture, urbanized areas, and/or built infrastructure (e.g., roads). As the new geographical framework for the species of the relictual habitats increasingly resembles an archipelago more than a continent, many authors have attempted to apply the principles of the theory to generate guidelines for these new anthropogenic landscapes. Although there have been many theoretical developments (e.g., metapopulation scenarios, source–sink relationships, edge effects), at the core of this work is the use of the theory of island biogeography to predict the eventual species losses following fragmentation, as immigration rates decline as a result of increased isolation, and extinction rates rise as a result of reduction in contiguous area of habitat. The process of species losses is termed relaxation. At the point of habitat disruption, residual fragments may become supersaturated (i.e., have

I S L A N D B I O G E O G R A P H Y, T H E O R Y O F

489

temporarily high species numbers), and may take time to reach their new equilibrium. The terms “lag effect” (duration of delay) and “extinction debt” (magnitude of the losses) are each used to describe the delay in species losses following fragmentation of habitat. Rather more controversially, the theory of island biogeography has also been invoked in the design of protected areas networks. It has been proposed that, in general, the theory favors deploying resources to protect fewer large reserves rather than many smaller ones (the “SLOSS Debate”), short rather than long inter-reserve distances, circular rather than elongated reserves (to minimize edge effects), and the use of corridors, when ever possible, to connect reserves and facilitate dispersal between them. SEE ALSO THE FOLLOWING ARTICLES

Extinction / Fragmentation / Relaxation / Species–Area Relationship FURTHER READING

Cody, M. 2006. Plants on islands. Diversity and dynamics of a continental archipelago. Berkeley: University of California Press. Gotelli, N. J. 2001. A primer of ecology, 3rd ed. Sunderland, MA: Sinauer. Lomolino, M. V. 2000. A call for a new paradigm of island biogeography. Global Ecology and Biogeography 9: 1–6. Losos, J. B., and D. Schluter. 2000. Analysis of an evolutionary speciesarea relationship. Nature 408: 847–850. MacArthur, R. 1972. Geographical ecology. Patterns in the distribution of species. Princeton, NJ: Princeton University Press. MacArthur, R., and E. O. Wilson. 1963. An equilibrium theory of insular zoogeography. Evolution 17: 373–387. MacArthur, R., and E. O. Wilson. 1967. The theory of island biogeography. Princeton, NJ: Princeton University Press. Rosenzweig, M. L. 1995. Species diversity in space and time. Cambridge, UK: Cambridge University Press. Simberloff, D. 1974. Equilibrium theory of island biogeography and ecology. Annual Review of Ecology and Systematics 5: 161–182. Whittaker, R. J., and J. M. Fernández-Palacios. 2007. Island biogeography. Ecology, evolution and conservation, 2nd ed. Oxford: Oxford University Press.

ISLAND FORMATION PATRICK D. NUNN University of the South Pacific, Suva, Fiji

To understand how islands form, continental islands must be distinguished from oceanic islands, the former being pieces of continents with the connection submerged, the latter being younger islands that originated exclusively within the ocean basins. However they appear today— low or high, limestone or volcanic—all oceanic islands began life as ocean-floor volcanoes. Those that have not

490

I S L A N D F O R M AT I O N

yet reached the ocean surface (and many never do so) are referred to as seamounts, whereas those that were once emergent but have since been submerged are often distinctively flat-topped and are called guyots. OCEAN-FLOOR VOLCANOES: ORIGINS AND GROWTH

It comes as no surprise to learn that we are not very knowledgeable about ocean-floor volcanism because of the difficulties in actually observing it. Most ocean-floor volcanism occurs in the dark beneath 4 km of ocean water. It is not that the technological difficulties are insurmountable, just that it is difficult to be sure that researchers are getting an accurate picture of what is going on. In this regard, places where the ocean floor actually rises above the ocean surface are extremely valuable as observation sites. Second best are places where seamounts have been thrust up above sea level and have their insides exposed for scientists to see how they were built up. The finest example of the first situation—where the ocean floor actually rises above the ocean surface—is the island of Iceland in the northern Atlantic Ocean. Iceland is part of the Mid-Atlantic Ridge (a divergent plate boundary) that lies at a plate triple junction and where eruptive activity has been unusually voluminous over the past few million years. The mid-ocean ridge—a common site of ocean-floor volcanism—actually passes through the center of Iceland. From studies of this, we learn that the earliest type of ocean-floor volcanism is commonly along fissures. As fissure eruptions continue, some parts of the fissure become blocked, and eruptions begin to occur at points. Point volcanism results in the build-up of the earliest types of seamounts. Studies of emerged seamounts—which rise from ocean floor that has been thrust upward by tectonic forces— have also given us a lot of information about the undersea development of oceanic islands. In particular, it is clear that intrusion of igneous rocks is at least as important as extrusion is in building seamounts in many parts of the ocean basins. Another important issue is the depth of overlying ocean water in places where seamount eruption occurs. In most places below about 600 m the weight of overlying water is so great that, however powerful the volcanic eruption, it will not be explosive, and the material produced will generally be pillow lava. At depths shallower than 600 m (the hydroexplosive zone), on the other hand, the weight of overlying water is not always sufficient to subdue explosive eruptions, and there is a reaction between the cold ocean water and the hot magma (liquid rock)

that causes the latter to solidify rapidly as numerous small fragments. Some of this fragmental (clastic) material may reach the ocean surface, where it floats as pumice, but most of it sinks and becomes draped over the sides of the more solid seamount as a sediment apron. Given the fragmented nature of the eruptive products produced in the hydroexplosive zone, a growing seamount often takes far longer to build itself up through this part of the ocean than through deeper areas. HOW ISLANDS RISE ABOVE SEA LEVEL

The three main ways in which a seamount can grow above the ocean surface are extrusion, intrusion, and uplift; each is discussed separately below. These categories should not be regarded as exclusive, for it is often a combination of these processes that actually leads to emergence. Island Emergence by Extrusion

Underwater volcanoes that erupt within the hydroexplosive zone sometimes cause islands to form. These are often referred to as “jack-in-the-box” islands because they alternately appear (during eruptions) and then disappear (between eruptions because of wave erosion). Examples are comparatively common in southwestern Pacific island groups such as Tonga and Vanuatu, located close to convergent plate boundaries. The reason that these islands do not endure long above the ocean surface is because they are composed entirely of unconsolidated and uncemented rock fragments and are promptly destroyed by wave erosion after the islandforming eruption comes to an end. For such an island to endure above the ocean surface, clearly it needs to be made of more resistant material. The way this happens has been recorded only once, with the exceptionally voluminous and lengthy eruption of an underwater volcano off the south coast of Iceland from 1964 to 1967, which produced the island Surtsey. In the case of this eruption, the amount of volcaniclastic (fragmented volcanic) material erupted created such a large island that at one point it isolated the volcano’s vent from ocean water. Once this happened, there was no longer any explosive reaction between ocean water and magma, and as a result, volcaniclastic rocks were replaced by lava. The lava flowed out over the surface of the volcaniclastic island, armoring it against erosion and leading to the establishment of a permanently emerged island. Island Emergence by Intrusion

One of the big unknowns in the emergence of oceanic islands is the precise role of intrusion—the emplacement

and solidification of igneous rocks within (not outside) an existing island edifice. Studies of former oceanic islands, now uplifted far above sea level and partly denuded to allow glimpses of their anatomy, show that intrusions can comprise as much as 70–80% of the total mass of such islands. Whether or not this is typical is uncertain, but it does underline the importance of intrusion. Intrusion begins to affect island growth from almost its beginnings, but it appears to be generally less important when an island is significantly emergent. Intrusion in early stages of island growth may be mostly through sill formation, although later, when the island edifice is sufficiently large to accommodate them, large intrusive bodies (batholiths, stocks) may come to dominate. Island Emergence by Uplift

Uplift refers to the upward forcing of an island, irrespective of its extrusive or intrusive activity. Island uplift is most common at convergent plate boundaries, where arcs of non-volcanic islands are often found, produced by the movement of the overriding (upper) plate across the top of the downgoing (lower) plate. Examples include the limestone islands of Tonga (South Pacific) and those of the Mentawai group (Indonesia). At convergent plate boundaries, the downgoing plate is commonly flexed (bent) upward before it goes down into the Earth’s interior, producing island emergence; examples include the Loyalty Islands (Southwest Pacific). Island uplift can also occur in the middle of plates, where islands are carried on moving plates across hotspots or other ocean-floor irregularities such as intraplate swells. Some of the Tuamotu Islands (South Pacific) have emerged as a result of such a process. More complex cases of island emergence result from a variety of causes. For example, there are various islands that are largely composed of pieces of ocean floors (ophiolites) that have been peeled off and pushed upward across a crustal irregularity. Examples include Crete (Mediterranean Sea) and La Grande Terre in New Caledonia (Southwest Pacific). ROLE OF SEA-LEVEL CHANGE IN ISLAND FORMATION

It may seem paradoxical, but both sea-level rise and sealevel fall can cause islands to form. Sea-level rise floods dry land and, in doing so, can transform a large island, for example, into a series of smaller islands. This is what happened as sea level rose after the last glacial maximum in the Channel Islands off the coast of California, where people about 12,000 years ago occu-

I S L A N D F O R M AT I O N

491

today. A good example is provided by the remote Pitcairn island group in the southeastern Pacific Ocean (Fig. 1). SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / Earthquakes / Motu / Oceanic Islands / Sea-Level Change / Seamounts, Geology / Surtsey FURTHER READING

FIGURE 1 Islands of the east central and southeast parts of the Pacific

Ocean. The biota of high Pitcairn Island shows affinities with the Gambier and eastern Tuamotu island groups. Henderson Island was uplifted during the late Quaternary (about 200,000 years ago) but was periodically submerged in earlier times. Ducie and Oeno atolls have been regularly submerged during high sea-level stages of the Quaternary; their terrestrial biota is derived from Pitcairn yet is consequently much less diverse. The existence of shallow-water (less than 130 m) submerged islands (marked by crosses on the map) that emerged during glacial low sea-level stages (most recently between 22,000 and 16,000 years

Kayanne, H., H. Yamano, and R. H. Randall. 2002. Holocene sea-level changes and barrier reef formation on an oceanic island, Palau Islands, Western Pacific. Sedimentary Geology 150: 47–60. Menard, H. W. 1986. Islands. New York: Scientific American Books. Nunn, P. D. 1994. Oceanic islands. Oxford: Blackwell. Nunn, P. D. 2007. Holocene sea-level change and human response in Pacific Islands. Transactions of the Royal Society of Edinburgh: Earth and Environmental Sciences 98: 117–125. Nunn, P. D. 2009. Vanished islands and hidden continents of the Pacific. Honolulu: University of Hawaii Press. Sinton, J., E. Bergmanis, K. Rubin, R. Batiza, T. K. P. Gregg, K. Gronvold, K. C. Macdonald, and S. M. White. 2002. Volcanic eruptions on mid-ocean ridges: new evidence from the superfast spreading East Pacific Rise, 17°–19° S. Journal of Geophysical Research 107: doi:10.1029/2000JB000090.

ago) has undoubtedly aided the successive recolonization of Ducie and Oeno. The lack of any submerged islands between the Pitcairn group and Easter Island–Sala y Gómez accounts for the marked lack of biotic similarity between these island groups.

pied a large island (named Santarosae), which was subsequently broken up into several smaller ones, something that had discernible effects on societal evolution there. Yet in the same period of sea-level rise, some ten of the 26 islands that existed during the last glacial maximum in the Channel Islands were completely submerged. A similar situation occurred in East Asia. During Quaternary glaciations, when sea level was low, the main islands of Japan often formed a single land mass (named Hondo), sometimes connected to the Asian mainland. Yet this connection was severed when sea level rose at the end of these glaciations, isolating Japan (and its occupants) from the continent and transforming one large island into a number of smaller ones. When Japan became a land of islands, people’s diets changed focus from terrestrial- to marine-dominated. For almost all of the past 2–3 million years, the ocean surface has lain tens of meters below its present level. Thus, for most of this time, the geography of islands has been quite different from the way it appears today. Many more islands were emergent at times of lower sea level, having emerged solely because of exposure as sea level fell to below their surface level. Many biogeographers have speculated that at such times the dispersal of plants and animals throughout the ocean basins was much easier than it would have been in a drowned world such as we occupy

492

ISLAND RULE

ISLAND RULE SHAI MEIRI Imperial College London, United Kingdom

The island rule is a name given for the supposed tendency of small-bodied animals to evolve larger sizes on islands whereas large animals evolve toward relatively smaller body sizes. The evolutionary forces that drive the observed patterns and the circumstances under which the phenomenon is manifest are widely debated but are thought to include changes in resource abundance, lower interspecific competition, elevated levels of intraspecific competition, and reduced predation on islands. SIZE EXTREMES ON ISLANDS

Anecdotal observation can lead to the conclusion that in many clades, island-dwelling species are characterized by extreme body sizes relative to their mainland counterparts, especially on large oceanic islands. Several animal groups have their largest or smallest representatives on islands. For example, the St. Helena earwig Labidura herculeana and the New Zealand wetas (Deinacrida spp.) may be the largest representatives of their clades. The world’s largest bat is the Philippine-endemic golden-crowned flying fox (Acerodon jubatus ). Brown bears on Kodiak (Ursus arctos

COMPARATIVE STUDIES OF SIZE EVOLUTION ON ISLANDS

In the 1960s John Bristol Foster compared closely related mammals (usually conspecifics) on islands and mainlands. He found that the body size of most of the island rodent populations he studied (mainly of the deer mouse, Peromyscus maniculatus) was larger than on the adjacent mainland. Most of the carnivore, even-toed ungulate, and lagomorph (rabbits and hares) populations he examined, however, were characterized by smaller body size on islands. Island marsupial and shrew populations showed no general tendency toward either gigantism or dwarfism. Lee Van Valen interpreted Foster’s findings as a tendency of large animals to grow small on islands and of small animals to grow larger on islands. He named this phenomenon “the island rule.” Such a pattern was later shown statistically in mammals in general by

Mark Lomolino, who quantified the body size of insular mammals as a function of the body size of the mainland population. Statistically the island rule is shown when the slope of this relationship is significantly lower than 1 (Fig. 1). Alternatively such a relationship can be quantified when island/mainland body size ratios are regressed against the body size of the mainland population. The island rule will be manifest in cases where the slope of this regression is negative (Fig. 2). Although the latter method is easier to visualize, it is fraught with statistical difficulties, because a regression of a ratio against its denominator is likely to produce more negative than positive results even if island and mainland body sizes are chosen at random. An interesting property of the aforementioned regression equations is that it is possible to compute a value at which body size will be the same on both islands and mainlands. Under the island rule, insular mammals that are smaller than this value tend to be larger than their mainland relatives, and those that are larger than this value tend to be smaller than their mainland relatives. The threshold itself is sometimes viewed as an evolutionary attractor toward which

Body size on island

middendorffi) are considered the largest land carnivores. At the other end of the scale, at roughly 100 kg, Elephas falconeri from the Middle Pleistocene of Sicily was by far the world’s smallest elephant. The largest bird known is the Madagascar elephant bird, Aepyornis maximus (although Dromornis stirtoni from the late Miocene of Australia may have been even larger), and the smallest snake is probably the Lesser Antillean threadsnake, Leptotyphlops bilineata. Both the largest and smallest lizards (Varanus komodoensis and Sphaerodactylus ariasae, respectively) are insular endemics. However impressive this array of examples may seem, they offer no more than anecdotal evidence that extremesized animals arise on islands. Even under an appropriate null model, the largest and/or smallest members of some clades are expected to be insular endemics by chance. The smallest member in three of 23 orders of terrestrial mammals, for example, is an insular endemic. This may not be surprising given that about 18% of the world’s mammal species are insular endemics. Thus, although many insular dwarfs and giants are known from several taxa, it is not clear whether the numbers are really different from those expected by drawing species randomly from a global size distribution. This issue can be easily examined using randomization tests, where a same number of species equivalent to the known number of insular endemics are drawn from a global species pool and their masses compared to those actually observed. As yet, such tests have not been conducted, and the issue of whether islands really harbor an unusual number of giants and dwarfs is therefore unresolved.

Body size on mainland FIGURE 1 A regression of island body size on mainland body size.

Each dot represents a pair of conspecific populations: one from an island, the other from a nearby mainland. Under the island rule, at small body sizes, dots will tend to fall above the line of equality (dashed line, where body size is the same on the island and the mainland) because small insular animals are larger than their mainland conspecifics. For large animals, dots will tend to fall below the line of equality (dotted line), because large animals get smaller on islands. The regression line (filled line) will thus have a slope of less than 1 (shallower than that of the line of equality). At body sizes where the regression line intersects the line of equality, little evolution is predicted: mainland and island animals will be roughly the same size (circle). This is the predicted “optimal” body size.

ISLAND RULE

493

Island / mainland body size ratio

100%

Body size on mainland FIGURE 2 A regression of size ratio on mainland body size. Each dot

represents a ratio: island body size/mainland body size (on the y axis) versus the body size of the mainland population. Under the island rule, at small body sizes, dots will tend to fall above the line of equality (dotted line, size ratio = 1): size ratios will be greater than 1 (island sizes larger). At large sizes, dots will tend to below the line of equality (size ratios of less than 1, large animals dwarf). The regression line (filled line) will thus have a negative slope. At body sizes where the regression line intersects the line of equality, little evolution is predicted: Mainland and island animals will be roughly the same size (circle). This is the predicted “optimal” body size.

size always tends to evolve (sometimes interpreted as an “optimal body size”; see below). Similar patterns to those described for mammals were shown for birds, turtles, snakes, and primates, but not for members of the mammalian order Carnivora or for lizards. MECHANISMS OF SIZE EVOLUTION ON ISLANDS

The island rule is thought to emanate from a combination of evolutionary forces, most of which are related to the overall low species richness on islands. Because islands support fewer animal species than the adjacent mainland, insular animals face fewer potential competitor species on islands, fewer predator species and, for carnivores, fewer prey species as well. This in turn is taken to imply that interspecific competition and predation risk are also lower on islands. Thus, if predation and competition influence size evolution on the mainland, these selective forces are thought to be relaxed on islands. Insular animals are therefore thought to be evolving “back” toward the mean (or modal) body sizes of their group (the most common size classes within taxa are usually perceived to be closer to the evolutionary attractors discussed above). Alternatively, if species occupying a certain niche in the ecological space on the mainland are missing from an island altogether, members of some other group may

494

ISLAND RULE

evolve to fill the niche. The existence of very large birds on oceanic islands (e.g., moas on New Zealand), for example, is often thought to result from the absence of large mammals. Thus, birds that perhaps cannot compete with large mammals on the mainland can evolve larger size only where mammals are absent. A feature of islands that is thought to promote dwarfism is their small area, supposedly via an overall shortage of resources. However, it is not clear why this should be so, because the extent of dwarfism needed to alleviate a food scarcity problem is probably far beyond that actually seen in nature, even taking into account the fast evolution of island animals: Body size usually does not diminish drastically enough for overall resource consumption to be meaningfully lower, and population densities on islands are often higher than on the mainland. Furthermore, it is not clear why size reduction, even if it benefits the population as a whole, is advantageous to an individual: If large individuals have the upper hand in intraspecific confrontation, then they will be able to control more food resources and have better reproductive success even when the population as a whole is stressed for food. Sicily, home to the smallest of the insular elephants, the 100 kg Elephas falconeri, was shown to be large enough to maintain a healthy population of 15-ton elephants, yet E. falconeri dwarfed there to a mere 1% of the size of its mainland ancestor. SELECTION PRESSURES AND IMPLICATIONS

The existence of an evolutionary attractor at medium body sizes assumes that both large and small sizes have their advantages. Large size is often associated with better ability to control territories, resources, and mates; larger prey size (for carnivores); better ability to fend off predators; higher number of offspring (in reptiles); better cold endurance; and a capacity to survive longer without food. Small size is correlated with earlier maturation and higher number of offspring (in mammals), shorter inter-birth intervals, better heat dissipation ability, lower overall resource requirements, and a higher chance of avoiding predation via the use of micro-habitat characteristics (e.g., burrows). Different species or groups of species were shown to respond differently to these selection pressures. Being at the very end of a group’s size spectrum is often problematic because of physiological and mechanical constraints. For example, very small homeotherm vertebrates have high surface-to-weight ratios. If resource acquisition scales isometrically with body size, then they must spend more time foraging than do larger homeotherms, and

thus heat loss may preclude maintaining homeothermy below a size of one gram or so. A medium size is perceived as a compromise between the selective pressures favoring size extremes. The forces that are thought to maintain a wide spectrum of body sizes on mainlands are, therefore, predation and interspecific competition. If predation exerts a diverging force on body size (small animals benefit from being smaller still and large animals from being yet larger), and if the maintenance of these sizes is costly, then under relaxed predation pressure, as on islands, size will evolve toward medium. With competition, maintenance of a specific size need not even be costly, as long as by evolving a more similar size to a competitor, a species is able to exploit some resources used by that competitor (assuming resources are limiting). A small species is likely to have fewer large competitors and a large species fewer small competitors on islands by chance alone, and so reduced number of competitors can lead to medium size. It has even been claimed repeatedly (but for different reasons) that large groups of animals (e.g., mammals) each have a single optimal size (the optima suggested for mammals differ by an order of magnitude: 100 g to 1 kg), and that the sole force maintaining the entire mammalian size distribution is interspecific competition. The island rule, with its regression toward intermediate sizes, was used as empirical evidence in support of all these theories. However, these models face some serious theoretical and empirical difficulties, not least from some patterns of insular size evolutions themselves: It has been shown empirically that different “optima” can be calculated for different mammalian subclades using island–mainland comparisons and that these perceived optima are probably artifacts of the specific data sets used rather than having a real biological meaning. FUTURE PROSPECTS

Although the existence of insular giants and dwarfs among many clades is undeniable, much work needs to be done to understand better both the patterns and processes of insular size evolution. Merely recording extreme body sizes is not enough: the existence of 200 kg rodents and 450 kg birds on islands may be consistent with one formulation of the rule (Foster’s) but not with others (Van Valen’s and Lomolino’s). Furthermore, it needs to be shown statistically that the numbers of dwarfs and giants on islands really deviate from a random draw of a similar number of species actually found on islands from an appropriate species pool. Otherwise, insularity need not be invoked to explain the existence of extreme body sizes on islands.

The terminology gets even more messy with attempts to invoke the island rule for phenomena that are unrelated in either time or space, such as susceptibility to extinction (e.g., Australian “time dwarfs,” where only smaller relatives of large Australian mammals survived the wave of extinctions at the end of the Pleistocene), size evolution in mainland areas during the nineteenth and twentieth centuries, or size patterns in the deep sea. Paradoxically, we may understand the processes that shape size evolution on islands, because selection pressures affecting size can be demonstrated, but the resulting patterns are obscured by our inability to predict which of the prevailing evolutionary forces will be most influential under a given set of ecological circumstances. More importantly, we have a limited theoretical framework that can be used to predict size evolution on islands. Mainland body size, which is assumed to be ancestral, is usually the sole factor used. Island area and island isolation feature less prominently and are often shown to have low explanatory power, or no effects on size evolution at all. This may be because patterns in nature are highly complex and may often be clade-specific, islandspecific, and contingent on the colonization history of the island. The same animals may show strikingly different patterns of size evolution even within archipelagos in relation to the prevailing ecological conditions on different islands. Rather than viewing Homo floresiensis, the hotly debated “hobbit” of Flores, as an obvious case of insular dwarfism in large mammals, it may be more rewarding to ask why humans should dwarf on highly productive islands such as Flores, with its large predators (Komodo dragons) and potential prey (Stegodon). Features such as species-specific body masses and island areas are easy to quantify but may not be good predictors of the direction and magnitude of size evolution on islands. Quantifying aspects of the biotic and abiotic environments on different islands that will be relevant for different species is much more challenging, but it may offer our best chance of obtaining a better understanding of size evolution. The identity of community members is likely to prove more important than their numbers. To make good predictions we are likely to need data on the feeding habits, territoriality, reproductive strategy, guild membership, and numbers of smaller and larger guild members for the species in question. We will also need estimates of the island-mainland differences in the identity of predators and (for carnivorous animals) prey, and perhaps primary productivity and population density as well.

ISLAND RULE

495

Understanding which animals will show which pattern of size evolution on different islands can offer a unique opportunity to decipher the mechanisms of size evolution in general, and better and more varied data may promise the best chance of achieving this. SEE ALSO THE FOLLOWING ARTICLES

Dwarfism / Gigantism / Komodo Dragons / Moa / Rodents FURTHER READING

Brown, J. H., P. A. Marquet, and M. L. Taper. 1993. Evolution of body size: consequences of an energetic definition of fitness. American Naturalist 142: 573–584.

496

ISLAND RULE

Case, T. J. 1978. A general explanation for insular body size trends in terrestrial vertebrates. Ecology 59: 1–18. Clegg, S. M., and I. P. F. 2002. The ‘island rule’ in birds: medium body size and its ecological explanation. Proceedings of the Royal Society B 269: 1359–1365. Lomolino, M. V. 1985. Body size of mammals on islands: the island rule reexamined. American Naturalist 125: 310–316. Meiri, S. 2007. Size evolution in island lizards. Global Ecology and Biogeography 16: 702–708. Raia, P., and S. Meiri. 2006. The island rule in large mammals: paleontology meets ecology. Evolution 60: 1731–1742.

J JAN MAYEN SEE ARCTIC REGION

Continental East Asia

JAPAN’S ISLANDS, BIOLOGY

Hokkaido

LÁZARO M. ECHENIQUE-DIAZ AND MASAKADO KAWATA Tohoku University, Aoba-ku, Japan

Honshu

Holarctic (Flora) Palearctic (Fauna)

Shikoku Kyushu

JUN YOKOYAMA Palaeotropic (Flora)

Yamagata University, Japan

Indo-Malay (Fauna)

The Japanese Archipelago

The Japanese Archipelago consists of 4 major islands (Hokkaido, Honshu, Kyushu, and Shikoku) and 6848 smaller islands, arranged across 3500 km parallel to the eastern coast of the Asian continent, and separated from it by the Sea of Japan. Overall, the Japanese Archipelago extends over a latitudinal range of 25° and a longitudinal range of 31°. Japan’s rich biota is reflective of the complex geological history of the archipelago and of the great diversity of climates it encompasses. It lies within two major biogeographic regions: the Palearctic and the Indo-Malay in faunal categories and the Holarctic and the Paleotropic in floristic categories. The biogeographic boundaries for flora and fauna do not coincide, the latter being slightly shifted northward (Fig. 1).

500 km

Biogeographic boundary for Fauna Biogeographic boundary for Flora FIGURE 1 Geographic setting of the Japanese Archipelago, and the

major biogeographic regions where it lies. The red dotted line marks the boundary between the Palearctic and the Indo-Malay biogeographic regions (defined for global patterns of animal distributions), and the black-dotted line marks the boundary for the Holarctic and the Paleotropic biogeographic regions (defined for global patterns of plants distribution). Smaller islands appear in red, and the four major islands appear differently colored.

BIOTA OF JAPAN: FLORA

1900 of them being endemic to the Archipelago. In addition, about 1000 species of bryophytes and more than 4500 algal species are recognized. This diversity is in part due to the different climate regimes that characterize the following distributional zones.

The flora of Japan is characterized by its rich diversity. About 7100 vascular plant taxa are recognized, more than

1. The subtropical zone, including the Ryukyu and Ogasawara island groups, and the lowland areas of

497

Kyushu and Shikoku islands, is characterized by the presence of evergreen broad-leaved trees (e.g., species of Camellia, Machilus, Distylium, Lithocarpus, and Castanopsis). In oceanic islands such as the Ogasawara group and a part of the Ryukyu group (Daito Islands), members of Fagaceae are absent as a result of isolation from mainland Japan. Several conifers are also present in the subtropical zone, such as Pinus luchuensis and Podocarpus macrophyllus. 2. The temperate zone ranges between the sea level to an altitude of about 1500 m in the mountainous areas of central and northeastern Honshu. In Hokkaido, it expands from the sea level to an altitude of about 600 m. In southern Honshu, Kyushu, and Shikoku, this zone is sparsely represented given the low occurrence of high mountains. It is characterized by the predominance of a large number of deciduous tree species (Quercus, Acer, Kalopanax, Fagus, Aesculus, Juglans, and other genera). Conifers such as Abies firma and Tsuga sieboldi are also typical. This zone is usually subdivided into three: the warm-temperature zone of broad-leaved evergreen forests dominated by evergreen Quercus species, covering most of southern Honshu, Shikoku, and Kyushu; the cooltemperature zone of broad-leaved deciduous forests dominated by Fagus and deciduous Quercus species, covering central and northern Honshu; and the southeastern part of Hokkaido (Oshima peninsula), and temperate broadleaf and mixed forests established in the transitional zone between the temperate and boreal zone. The third type of forests is dominated by Betula, deciduous Quercus species, and coniferous genera Abies and Picea. 3. The boreal zone covers areas above 1500 m on Honshu, above 600 m in central Hokkaido, and from the sea level in restricted areas in the easternmost part of Hokkaido. This zone is dominated by coniferous species of the genera Abies, Tsuga, Picea, and Larix. Deciduous broad-leaved trees such as Betula ermani also represent important members of this zone. Most of the components of Japanese flora came from continental East Asia, and some families endemic to East Asia are also distributed in Japan (e.g., Cercidiphyllaceae, Eupteleaceae, Stachyuraceae, Trochodendraceae). Two monotypic families, Sciadopityaceae and Glaucidiaceae, and more than 20 genera (e.g., Anemonopsis, Peltoboykinia, Pteridophyllum, Ranzania) are endemic to Japan. About 90 genera of plants in Japan show a disjunct distribution in East Asia and North America. Distinctive genera include

498

J A PA N ’ S I S L A N D S , B I O LO G Y

Buckleya, Caulophyllum, Diphylleia, Stewartia, Schisandra, Illicium, Wisteria, Lespedeza, Penthorum, Itea, Astilbe, Menispermum, and Shortia. The Ogasawara Islands and most of the Ryukyu Islands belong to the Paleotropic floristic kingdom, having common floral components with Polynesia (Ogasawara) and Southeast Asia (Ryukyus). For example, some of the endemic species of the Ogasawara Islands (e.g., Clinostigma savoryana, Loberia boninensis, Metrosideros boninensis, and two species of Boninia) are considered as having a Polynesian origin. Many plant genera such as Archidendron, Arenga, Argostemma, Bischofia, Cananga, Dipteris, Erycibe, Epipremnum, Flagellaria, Grewia, Lepidagathis, Macaranga, Melicope, Murraya, Rhynchotechum, Schima, or Wendlandia, are considered as South-East Asian elements of the Ryukyu Islands. BIOTA OF JAPAN: TERRESTRIAL FAUNA

The Japanese fauna is relatively rich in comparison with those of other countries of similar size and given the fact that most of Japan belongs to the temperate climatic zone. This richness has a geographical foundation. Most islands of the Japanese Archipelago are continental islands and have a large basic stock of fauna that migrated from the Eurasian continent by different routes. Isolation and vicariant events due to glacial retreats that cut off connections with the mainland are the foundations of many cases of relict distributions. The Tokara Straits (Watase line) and the Tsugaru Strait (Blakiston line) (Fig. 2) represent connectedness gaps in the geological history of the archipelago and divide it into

FIGURE 2 Major island groups of the Japanese Archipelago, and zoo-

geographical regions as determined by distributional patterns and historical connectedness gaps. Islands with the same color have greater faunistic affinities than islands with different colors.

three zoogeographic regions: Hokkaido; other main islands of the Archipelago (Honshu, Shikoku, and Kyushu) with their surrounding islands; and the Ryukyu and Ogasawara island groups. Each part has a specific fauna (more noticeable in birds and mammals), resulting in a very diverse and rich fauna for the whole archipelago (Fig. 2). Japanese Mammals

Japanese Mammals are characterized by their endemism (40%), species abundance, and the scarcity of large species. Endemic and relict mammals as the result of isolation are found all across the Japanese Archipelago. On the Ryukyu Islands, separated from the continent in the late Tertiary, there are 14 endemic terrestrial species (48.3%), including two endemic genera (Pentalagus and Tokudaia). Honshu, Shikoku, Kyushu, and the surrounding islands (known collectively as Hondo) appeared in the Quaternary (around a million years ago), and together host 27 endemic species (46.5%). Most of the species in the genera Apodemis, Eothenomys, and Aschizomys, very common in Honshu, are endemic. On the other hand, Hokkaido shows a more monotonous fauna, with no endemics at the species level and many species in common with the Eurasian continent, such as Ursus arctos, Sciurus vulgaris, Tamias sibiricus, and Lepus timidus. Japanese Birds

The prime characteristic of Japanese birds is regionality. Their distribution is largely affected by the Watase and Blakiston lines (Fig. 2), so that in the Ryukyu Islands, birds belonging to the Indo-Malay biogeographic region are more common, while Palearctic avifauna mainly from the Siberian region is distributed in Hokkaido. Southern Palearctic species from East Asia are more frequently found in Honshu, Kyushu, and Shikoku. In Hokkaido and the Ryukyu Islands, migratory birds represent more than 80% of the population, while in Honshu they account for 60%, and in Shikoku and Kyushu 40%. Many endemic birds are found across the Japanese Archipelago (e.g., Phasianus soemmerringii, Phasianus versicolor, Synthliboramphus wumizusume, Picus awokera), and some taxa are found only in remote oceanic islands. These include Turdus celaenops (Izu Islands), Apalopteron familiare hahasima (Bonin Islands), Diomedea albatrus (Torishima Islands), and Otus scops interpositus (Daito Islands) (Fig. 2).

snakes all across the Japanese Archipelago are characterized by their high endemism (over 80%). The same occurs for amphibian species, showing over 75% endemism. Another characteristic of the Japanese herpetofauna is the abundance of salamander species. Particularly interesting examples are hynobiid salamanders, which had undergone a significant geographic separation in restricted areas of Japan, resulting in more species of this group in the Japanese Archipelago than on the continental mainland. The world's largest living amphibian, the Japanese giant salamander Andrias japonicus, a primitive species, is also an example of the evolutionary significance of the Japanese herpetofauna, where many relict species are found. Japanese Invertebrates (Insects, Spiders, and Land Snails)

Most insects in Japan are common East Asian species, mainly derived from northeast Asia and from the South China–Himalayan region. They are characterized by a large number of species (e.g., over 8000 species of beetles and weevils, 310 of butterflies, 5000 of moths, 140 of cicadas and aquatic hemipterans, and 175 of dragonflies). Around 100,000 species are estimated for the whole archipelago, of which 30,000 are known). Some groups show high endemism (e.g., 70% in Carabinae ground beetles), and some species are representative of relict insect groups. Among these is an archaic dragonfly, Tanypteryx pryeri, a member of a group of which only ten species survive in the Pacific Rim, and subterranean species of the family Grylloblattidae (a group of wingless insects that live on top of mountains, sometimes called ice crawlers or icebugs). Spiders in Japan comprise over 1400 species in 53 families. Endemic species can be found across the Archipelago and some oceanic islands. Conspicuous examples are Doosia japonica, Desis japonica, Verpulus boninensis, and Alopecosa hokkaidensis. The characteristic for Japanese land snail fauna is a north-to-south cline in species richness, with more species found in southern parts of the archipelago than in northern areas. Endemism is also another feature of this group, with some oceanic islands showing the highest values (e.g., over 100 species in the Ogasawara Islands, 90% of which are endemic, including seven endemic genera such as Ogasawarana, Mandarina, Hirasea, Hirasiella, and Boninena). Japanese Freshwater Fishes

Japanese Reptiles and Amphibians

About 87 taxa (species and subspecies) of reptiles and 60 of amphibians are known in Japan, symbolizing the evolutionary importance of the archipelago. Lizards and

The indigenous freshwater ichthyofauna of Japan is closely related to the fauna of continental East Asia and consists of about 214 native species, of which 24.3% are endemics. Strictly freshwater fish fauna is more abundant

J A PA N ’ S I S L A N D S , B I O LO G Y

499

in central and western Honshu, where Lake Biwa is the main species reservoir. Of the peripheral freshwater fish fauna, northern species such as salmonids, sticklebacks, and sculpins are well represented in northeast Honshu and Hokkaido, and southern elements such as gobies occur mostly in western and southern areas.

places (26% of the national land). The ratio of natural vegetation is regionally unbalanced, with Hokkaido and Okinawa showing the highest values (48.7% and 47.9%, respectively), while other regions had less than 20%. A large proportion of the natural vegetation lies within National Parks and other nature reserves, a network of 37,500 km2 of legally enforced conservation areas.

JAPANESE MARINE BIOTA

Three major ecosystems with a diverse fauna can be found along the Japanese shores. Algal beds on rocky shores (the subtidal ecosystems) are dominated by Sargassum species, kelp beds are typically represented by Laminaria, Eisenia, and Ecklonia, while the sandy mud bottom is dominated by Zostera marina sea grass. In the Ryukyu Islands, the latter species is replaced by Thalassia hemprichii. The coral reef ecosystem in Japan covers an area of approximately 900 km2, extending from the Ryukyu Islands to the southern areas of Kyushu, Shikoku, and the Kii Peninsula, with several gaps in between. It is characterized by a large number of genera and species, favored by the warm Kuroshio current. This region is considered a marine biodiversity hotspot, with 1315 fish species of which 26 are endemic (1.97%), ranking fourth among the world’s top ten areas for endemic reef fishes. Conspicuous species include Stegastes altus, Chromis albomaculata, and Chaetodon daedalma. It also harbors a large number (400) of coral species, including the oldest and bigger community of blue coral (Heliopora coerulea) in the Northern Hemisphere. The tideland ecosystem is found in estuarine environments, supporting a rich fauna of marine invertebrates.

SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Endemism / Extinction / Freshwater Habitats / Japan’s Islands, Geology FURTHER READING

Abe, H. 1999. Mammals of Japan, their diversity and conservation, in Recent advances in the biology of Japanese Insectivora: proceedings of the symposium on the biology of insectivores in Japan and on wildlife conservation. Y. Yokohata, ed. Hiwa, Japan: Hiwa Museum for Natural History. Goris, R., and N. Maeda. 2004. Guide to the amphibians and reptiles of Japan. Melbourne, FL: Krieger Publishing. Japan Integrated Biodiversity Information System (J-IBIS). http://www .biodic.go.jp/english/J-IBIS.html. Masuda, H., K. Amaoka, C. Araga, T. Uyeno, and T. Yoshino. 1984. The fishes of the Japanese Archipelago. Tokyo: Tokai University Press. Ministry of the Environment, Government of Japan. Nature and Parks. http://www.env.go.jp/en/nature/. Murata, J. 2000. Flora of Japan, in The botany of biodiversity 1. Floristic Research. K. Iwatsuki and M. Kato, eds. Tokyo: University of Tokyo Press, 24–47. Otsuka, H., H. Ota, and M. Hotta, eds. 2000. International symposium: The Ryukyu Islands—arena for adaptive radiation and extinction of island fauna. Tropics 10.1: 1–241. Wen, J. 1999. Evolution of eastern Asian and eastern North American disjunct pattern in flowering plants. Annual Review of Ecology and Systematics 30: 421–455.

EXTINCTIONS IN THE JAPANESE BIOTA

Recent extinctions of plants and animals in Japan are well documented. Among the relatively recent extinctions are at least four species of mammals (two endemic bat species, the Japanese sea lion, and the Japanese wolf), while a fourth endemic species, the Japanese otter (Lutra nippon) is seemingly extinct. Similarly, 13 species of birds, 3 fish species, 2 insect species, and 41 species of plants (including eight extinct in the wild) are reportedly extinct. These extinctions are attributed to human influences, although natural causes cannot be excluded (e.g., impact of volcanic activity). Drastic changes in the natural lands of Japan began with the process of modernization and were especially severe after World War II, resulting in the decline and disappearance of wildlife habitats throughout the archipelago.

JAPAN’S ISLANDS, GEOLOGY S. MARUYAMA Tokyo Institute of Technology, Japan

S. YANAI Japan Geocommunications, Co., Ltd., Tokyo

Y. ISOZAKI University of Tokyo, Japan

D. HIRATA Kanagawa Prefectural Museum of Natural History, Odawara, Japan

NATURE CONSERVATION IN THE JAPANESE ARCHIPELAGO

Although about 67.0% of the Japanese archipelago is forested, original primary vegetation only remains in few

500

J A PA N ’ S I S L A N D S , G E O LO G Y

The Japanese islands are a bow-shaped chain of islands extending over 3000 km along the eastern margin of Asia. Four arcs—from north to south, the Kurile, Japan, Izu–

Mariana, and Ryukyu arcs—are combined to form Japan. The Izu–Bonin arc extends south from the Japan arc (Fig. 1). The climate of Japanese islands is variable depending on latitude, ranging from snowy (from Hokkaido to northeastern Japan) to moderate (southwestern Japan) to subtropical (on the Ryukyu and Ogasawara arcs). One hundred and thirty million people live on the Japanese islands. Over 200 active volcanoes and their accompanying hot springs are present. Moreover, earthquakes along subduction zones and inland inner arcs pose additional geologic hazards. EARLY GEOLOGIC STUDIES

The geology of the Japanese islands was first studied about 120 years ago by German geologist Edmund Nauman (1854–1927), who initially, at the age of 20, thought Japan was composed entirely of volcanic rocks, because they are located in the Pacific Ocean. Performing extensive geologic study during his 10-year stay in Japan, he recognized that high-grade regional metamorphic rocks in Japan were similar to those in continental areas, and therefore he realized that Japan had been a part of Asia before the opening of the Sea of Japan in the Tertiary and had rifted away and become isolated from Asia. Since 1950, modern geologic mapping has been carried out, and

the evolution of the Japanese islands is now well defined, as outlined here. The history of the Japanese islands is now confirmed back to 1.9 billion years ago, as a part of the South China continent. GEOLOGIC OUTLINE Plate Boundaries around Japan

Japan faces four major plates, with the Pacific plate subducting at 10 cm/yr under Northeast Japan, where seismologically detected slabs reach to a depth of 660 km under Beijing, China. The Japan Trench is a topographically depressed region ~5–6 km deep, corresponding to the western end of the Pacific plate. The Philippine Sea plate subducts from southeast to northwest at 4 cm/yr under Southwest Japan. The boundary is called the Nankai Trough or Trench and can be traced on-land through Honshu to its eastern end, south of Tokyo. Northeast Japan (northern Honshu) and Hokkaido are part of the North American plate, which is underthrust from the west by the Eurasian plate. Two trench–trench–trench (TTT) triple junctions are present near Tokyo, one near Mt. Fuji and another to the south of Tokyo. Northeast Japan has been colliding against Southwest Japan to form the Japanese Alps by uplift of the leading edge of the Eurasian plate since 0.5 million years ago. To

FIGURE 1 Geographic position of the Japanese islands. The island arc chains around Japan consist of the Kurile, Japan, Ryukyu, and Izu-Mariana

arcs. The Fossa Magna separates the Japan arc into two segments; Northeast Japan on the North American plate and Southwest Japan on the Eurasian plate. The Pacific and Philippine Sea plates are juxtaposed with the Northeast Japan–Izu-Mariana arcs and the Japan–Ryukyu arc, respectively. The Sea of Japan behind the Japan arc is a rifted basin. The Okinawa trough behind the Ryukyu arc is currently active. Tectonic frameworks of the Japan and Ryukyu arcs are linked with the Cathaysia block, the southern part of the South China block.

J A PA N ’ S I S L A N D S , G E O LO G Y

501

First out-ofsequence

Rigid forearc rocks 0

F

BSR

thrust zone (OST)

Cover sequence undeformed

E

km

2 4

Imbricate thrust zone (ITZ)

D

C

Nankai Trough Axis

Large thrust slice zone (LTSZ)

Proto-thrust zone

Landward dipping reflectors zone (LDRZ)

B

A

OST Cover 1st OST sediment

OST

LDR Old fault traces

8 10 km

Trough

off

Southwest

Japan. The principal fabrics around the converging plate boundary are marked by development of decollement and associated imbrication pileups moto et al. (2000).

6 ? enic zone

ary complex along Nankai

(duplexes). Source: Kura-

LDR

Seismog

FIGURE 2 Seismic reflec-

tion profile of accretion-

Imbricate packages

Décollement

Imbricate packages

Hemipelagic Unit

Décollement Turbidites

PHILIPPINE SEA PLATE

V.E.~3x

the north of the Ryukyu arc, the Okinawa Trough marks where back-arc spreading has been on-going since 5 million years ago. The spreading began from the southern end and propagated northeastward to northern Kyushu. Accretionary Front

One of the characteristic geologic phenomena is the formation of an accretionary complex on the leading edge of the Eurasian plate. The seismic reflection profile off Shikoku island (Fig. 2) shows thick turbidite units, right above the subducting Philippine Sea plate, that are highly folded and have thrust-bounded oceanward vergency. A series of highangle reverse faults flatten at depth. Thrust-converged parts form a decollement zone between the underlying Philippine Sea plate and the accretionary wedge above, successively forming Zones A, B, C, D, E, and F. Zone A, the youngest one, is not yet deformed and originally derived from trench turbidites formed along the submarine canyons. The material and physical boundaries are somewhat different. Channelized fluid pipes pass along the decollement zone and branch to several active reverse faults and to the surface, where biological communities form colonies. Oceanic materials are often accreted together with trench turbidites. Japan has grown by adding accretionary complexes through time since 500 million years ago, but also has eroded tectonically at trenches to lose already-formed accretionary complexes. The eroded materials either underplate the hanging wall at great depth or are removed down to the deep mantle. The near-trench zone in Northeast Japan is an example where only small amounts of post-Miocene accretionary complex are present. Arc Volcanism and Underlying Mantle

Japan’s arc volcanoes form as a result of secondary mantle convection triggered by cold slab subduction, and the arc runs parallel to the active trench. Volcanoes occur 100–120

502

J A PA N ’ S I S L A N D S , G E O LO G Y

km above the subducted slab, regardless of the angle of the subducting slab. If the angle is large, as in the case of the Mariana arc, the distance between volcano and trench becomes short. An example from Northeast Japan is shown as a cross section (Fig. 3). High resolution P-wave tomographic images clearly document the high-velocity subducting cold Pacific slab (blue). The hot, low-velocity perturbation (orange-yellow) in the hanging wall corresponds to secondary mantle convection rising up to intersect the arc crust where the majority of volcanoes are formed (called the volcanic front). Minor volcanoes form behind the front, called the second chain of volcanoes on the Sea of Japan side. On the cross sections, earthquakes are observed in the Pacific slab and the Japan arc. Earthquakes are absent along the top boundary of the slab. Double seismic zones occur within the Pacific slab, namely at the topmost slab mantle and in the center of the slab. These earthquakes

FIGURE 3 P-wave seismo-tomographic image of the Northeast Japan

arc. The Pacific slab, blue-colored (high-P wave velocity), can be well discerned. The orange-yellow-colored low-velocity zone is developed in the hanging wall of the mantle wedge, probably driven as a counterflow of mantle convection triggered by the Pacific slab subduction. A rising counterflow of mantle intersects with arc crust to release arc magma by melting mantle as a result of high water content derived from the Pacific slab and decompression. Source: Hasegawa et al. (2008).

are triggered by dehydration embrittlement of the hydrous magnesium silicates. Water-rich fluids liberated from dehydration reactions lower the stress in the plate under extremely high-pressure conditions, triggering earthquakes. Deep earthquakes in the mid-ocean ridge basalt (MORB) crust are abundant to 50–60 km depth, and limited to 100 km depth, by dehydration reactions in the hydrated MORB crust. Earthquakes are also abundant in the upper half of the arc crust as a result of the brittle-ductile transformation of granitic materials. The boundary between absence and presence of earthquakes, corresponds to the 350 °C isotherm, above which granitic materials behave in a brittle fashion, hence forming earthquakes. Geotectonic Divisions

Rocks of the Japanese islands are classified into four kinds: (1) accretionary complexes dominated by trench turbidites with enclosed oceanic materials, (2) igneous rocks derived from arc magma, (3) regionally metamorphosed rocks from 10–60 km depth above the subducting slab, and (4) normal sedimentary sequences. These four kinds can be grouped into one set of units, formed at the same time in different

positions. For example, during the Cretaceous, ~120–80 million years ago, extensive volcano-plutonism occurred to form a huge batholith belt accompanied by acidic effusive rocks. Simultaneously, the deep-seated subduction zone complex was metamorphosed along the Benioff plane at 10–60 km depth and returned to the surface near the forearc region, a new accretionary complex formed oceanward, and active sedimentation took place in the fore-arc region. This group of geologic units is termed the Cretaceous orogenic unit. Japan is composed of six of these orogenic units formed 450, 320, 250, 120–80, and 60 million years ago and post-Tertiary. These are well-documented in Northeast and Southwest Japan, and presumably in the Ryukyu islands, but not in Izu-Ogasawara and eastern Hokkaido. These two regions originally formed above the ~100-million-year oceanic lithosphere by subduction of slabs to become arcs, and subsequently they collided to become parts of the Japanese islands. The major parts of Japan are composed predominantly of accretionary complexes, including even regionally metamorphosed ones. The material incorporated into the accretionary complexes is a key to estimate the age of subducted slabs with ocean plate stratigraphy (Fig. 4). Thus, the

FIGURE 4 Ocean plate stratigraphy through time in Japan. Simplified ridge-subduction system and the concept of ocean plate stratigraphy indi-

cate the age gap between two distinct horizons; the horizon between pillowed MORB and pelagic chert marking the oceanic plate; the horizon between hemipelagic mudstone and terrigenous clastics marking the arrival at the trench, which represents the total travel time of the subducting oceanic plate from mid-oceanic ridge to trench or, in other words, the age of the subducting oceanic plate at trench.

J A PA N ’ S I S L A N D S , G E O LO G Y

503

formation of accretionary complexes has been estimated to have occurred 500–450, 400–320, 300–250, 200–120, and 80–70 million years old, and the post-Tertiary. The Japanese islands were initiated by continental rifting of the Cathaysia continent with 1.9–billion-yearold rocks. These old basement rocks are now exposed on small islands in the Sea of Japan, Oki-Dogo island. Further to the south, the structural top unit is the regionally metamorphosed unit formed 240 million years ago by the collision of South China against North China. The oldest basement is underneath this unit. The oldest ophiolite sequence (~600 million years old) is the oceanic lithosphere made by the opening of the paleo-Pacific ocean when the Rodinia supercontinent was formed and is equivalent to the oldest ophiolites along the western margin of North America, such as the Klamath ophiolite. After the conversion of the plate boundary from passive to active margin 450 million years ago, an accretionary complex formed and periodically exhumed the deep-seated accretionary complex to the surface, while the volume of arc crust increased rapidly by calc-alkaline volcano-plutonism. Later tectonic modification, specifically the opening of the Sea of Japan and the collision of the Izu-Ogasawara arc and eastern Hokkaido, strongly modified pre-existing geologic units of Japan along the Itoigawa–Shizuoka Tectonic Line (originally called the Fossa Magna by E. Nauman), the Median Tectonic Line, the Tanakura Tectonic Line, and the Ishikari Tectonic Line and Kannawa thrust fault. Oceanic plate stratigraphy and regional ultrahigh- and high-pressure metamorphic suites of accreting units of southwestern Japan are assigned dates from late Paleozoic to Neogene. The oceanward and tectonically downwardyounging polarity is noted, as shown in the structural profile compiled schematically in Fig. 1.

FORMATION OF GONDWANA AND ITS BREAKUP AND DISPERSION (540–300 MILLION YEARS AGO)

In the following sections, the geotectonic evolution of Japan is illustrated by a series of paleogeographic maps.

The smaller continents were amalgamated to form the semi-supercontinent Gondwana at 540 million years ago. North America subsequently docked to Baltica to form Caledonides and collided later against Africa to form Hercynides. On the other hand, Gondwana began fragmenting in Early Paleozoic, and the pieces were later amalgamated to form Laurasia during the late Paleozoic. The new supercontinent Pangaea formed near the end of Paleozoic to early Mesozoic. Subduction of the Pacific seafloor occurred at almost all of the continental margins around the Pacific in the Early Ordovician, ~480 million years ago. Proto-Japan, near South China, had remained near the same corner of the Australian continent since the Cambrian, but it was now enrolled for the first time into a subduction regime. According to the distribution of the continental blocks in the Early Devonian, ~400 million years ago, South China and Japan were close to Gondwana. Following the collision and closure of the seaway between Laurentia and Baltica that formed Laurasia (the Caledonian orogenic belt, colored in red), the remnant segments of the Iapetus Ocean were narrowed by successive subduction. South China moved to the north of Australia and became isolated in the Pacific. North China was also isolated from the other continental blocks and belonged to an independent faunal province.

RODINIA BREAKUP, ~750–600 MILLION YEARS AGO

FORMATION OF PALEOASIA (LAURASIA),

During the late Proterozoic, 1.1–0.9 billion years ago, all fragmented continents were united through extensive collisions to form a supercontinent, Rodinia, cemented by several Grenvillian orogens. The Rodinia supercontinent was present in the Southern Hemisphere, where most continents (except for West Africa, Amazonia, or both) formed another supercontinent, were rifted, and drifted apart under the force a huge upwelling mantle plume, the Pacific superplume at the triple junction. South China

~200 MILLION YEARS AGO

Geologic History

504

(with proto-Japan) was juxtaposed with Laurentia (present-day western North America) and situated close to the divergent triple junction. One of the important Grenville orogens is called the Shibao, which was a failed rift in South China separating the southern Cathaysia and northern Yangtze cratons. The Cathaysia was a part of North America, whereas Yangtze was a part of East Antarctica or Australia. Rodinia had broken up by ~600 million years ago, and more than 10 smaller continents had separated and migrated.

J A PA N ’ S I S L A N D S , G E O LO G Y

In the Late Carboniferous ~300 million years ago, several continental blocks, including South China, North China, Tarim, and Indochina, moved northward from the Southern Hemisphere (Fig. 5) because a large-scale cold superplume (colored in purple in the figure) developed and dragged the blocks. In addition, the other continental blocks in modern Asia, such as Siberia and Kazakhstan, were also dragged in the same domain. On

FIGURE 5 Birth of the Asian cold superplume ~300 million years ago.

Am = South America; Au = Australia; Ba = Baltica; Co = Congo; EA = Eastern Antarctica; I = Indochina; In = India; K = Kazakhstan; Ka = Kalahari; La = Laurentia; Si = Siberia; SK = North China; T = Tarim; WA =

FIGURE 6 Collision of South China (Yangtze) against North China

West Africa; Yg = South China.

(Sino–Korea) during the Triassic period. The figure shows the modern coastlines of east Asia as dotted lines for reference, land as dark yel-

the other hand, the closure of the Iapetus ocean was completed along the Hercynian–Appalachian orogenic belts, and the merger of Gondwana and Laurasia consequently formed the supercontinent Pangaea. As shown in Fig. 5, North China and South China continents moved northward to merge with Siberia, closing seaways among these blocks ~280 million years ago. Their mutual isolation resulted in the development of three distinct floristic provinces, namely the northern Cathaysia, southern Cathaysia, and Angara flora, respectively. Proto-Japan was on the southeastern continental margin of South China and has grown with the pasted materials of the Pacific Ocean, or Panthalassa. The Carboniferous to Permian Akiyoshi–Sawadani seamount chains, capped by a reef limestone, were approaching the proto-Japan margin of the South China continent, driven by subduction of the Farallon plate.

low, sea as blue, green seamount and oceanic plateau as green, granite batholith belt as red, and mid-oceanic ridges as light yellow.

from the subducted Akiyoshi–Sawadani seamount chain and the Permian Maizuru oceanic plateau. After or around 210 million years ago, the Mongolian seaway mostly closed, and a delta formed in the eastern end of the suture. The Qinling–Dabie suture exhumed the ultrahigh-pressure metamorphic unit from the depth where diamonds form. Abundant terrigenous clastics were transported along the suture and finally brought to a deep-sea fan formed at the trench along the Pacific

COLLISION OF SOUTH CHINA AGAINST NORTH CHINA 240 MILLION YEARS AGO

The Yangtze, or northern South China, started to collide against North China from Dabie Promontory during the Triassic period, ~240 million years ago, closing the paleo-Tethyan seaway. The collision suture developed along the Qinling–Dabie mountains in central China. The Mongolian seaway between the North China and Siberia continents also narrowed through double-vergent subduction on both sides. Along the southern margin of South China, the active subduction of the Farallon plate formed the Late Permian Akiyoshi accretionary complex (colored in light pink in Fig. 6), incorporating fragments

FIGURE 7 Separation of Japan from the Asian continental margin

around 20 million years ago.

J A PA N ’ S I S L A N D S , G E O LO G Y

505

Philippine Sea plate generated slab melting along the subduction zone. COLLISION OF IZU–BONIN ARC AGAINST HONSHU ARC SINCE 5 MILLION YEARS AGO

FIGURE 8 Collision of Izu–Bonin arc 5 million years ago.

ocean. Owing to the abundant supply of clasts from the suture, accretion along the Yangtze margin revived and constructed the Late Triassic–Early Cretaceous accretionary complex of the Mino–Tanba–Chichibu belt, Southwest Japan. As the triple junction of ridge–trench–trench moved northeastward, the young and buoyant Farallon and Izanagi Plates subducted, tectonically exhuming the high-pressure/low-temperature Sangun metamorphics and generating a granite batholith belt beneath the coeval volcanic arc. OPENING OF THE SEA OF JAPAN, ~20–15 MILLION YEARS AGO

The Japanese islands became an island arc. The activity of the sub-Asia plume peaked, accelerating the rifting in major basins in East Asia, such as the Sea of Japan, the Baikal basin, and the Bohai basin in northern China. Bimodal (both basaltic and rhyolitic) volcanism characterized the initial phase of rift-related basin formation (Fig. 7). The back-arc basins reached their full extent, generating several across-arc strike-slip faults. In contrast to the back-arc extension, the fore-arc of the Japanese islands suffered from a local contraction that tectonically juxtaposed the Ryoke granites above the Sanbagawa high-pressure/low-temperature metamorphics by the subhorizontal paleo–Median Tectonic Line in Southwest Japan. Another back-arc basin opened in the Philippine Sea plate, in which the proto–Izu-Bonin arc split the Kyushu–Palau ridge (arc) and created the Shikoku– Parece Vela basin. The subduction of the young and hot

506

J A PA N ’ S I S L A N D S , G E O LO G Y

The back-arc basins that were opened during the Miocene, namely, the Sea of Japan and the South China Sea, had already been turned into a contraction regime by 5 million years ago. The subduction of the Philippine Sea Plate beneath Asia accompanied the collision/subduction of the Izu–Bonin arc against the Northeast Japan arc in central Japan (Fig. 8). A fore-arc sliver at the front of the Kurile arc was formed by the oblique subduction of the Pacific Plate. The Japanese islands are currently located on four distinctive plates: the Eurasian, Okhotsk (North American), Philippine Sea, and Pacific Plates. Three active arc-trench systems developed in Southwest Japan–Ryukyu, Northwest Japan, and Izu–Bonin. The oblique subduction of the Philippine Sea plate generated fore-arc slivers along the Southwest Japan–Ryukyu arc. The linear neo–Median Tectonic Line cutting the low-angle paleo–Median Tectonic Line corresponds to the landward margin of the fore-arc sliver in Southwest Japan. Another back-arc basin with hydrothermal activity, the Okinawa trough, is opening in the Ryukyu. SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Island Arcs / Japan’s Islands, Biology / Kurile Islands / Pacific Region FURTHER READING

Maruyama, S., Y. Isozaki, and G. Kimura. 1997. Paleogeographic maps of the Japanese islands: plate tectonic synthesis from 750 Ma to the present. Island Arc 6: 121–142. Maruyama, S. 1997. Pacific-type orogeny revisited; Miyashiro-type orogeny proposed. Island Arc 6: 91–120. Maruyama, S., M. Santosh, and D. Zhao. 2007. Superplume, supercontinent, and post-perovskite: mantle dynamics and anti-plate tectonics on the core-mantle boundary. Gondwana Research 11: 7–37. Maruyama, S., D. A. Yuen, and S. Karato. 2007. Plumes and superplumes through Earth’s history, in Superplumes: beyond plate tectonics. D. A. Yuen, S. Maruyama, S. Karato, and B. F. Windly, eds. Berlin: SpringerVerlag, 441–502. REFERENCES

Hasegawa, A., J. Nakajima, S. Kita, Y. Tsuji, K. Nii, T. Okada, T. Matsuzawa, and D. Zhao. 2008. Transportation of H2O in the NE Japan Subduction Zone as inferred from seismic observations: supply of H2O from the slab to the arc crust. Journal of Geography (Tokyo Geographic Society) 117.1: 59–75. Kuramoto, S., A. Taira, N. L. Bangs, T. H. Shipley, G. F. Moore, and EW99-07, 08 Scientific Parties. 2000. Seismogenic zone in the Nankai Accretionary Wedge: General summary of Japan–U.S. Collaborative 3-D Seismic Investigation. Journal of Geography (Tokyo Geographic Society) 109.4: 531–539.

JUAN FERNANDEZ ISLANDS

to be almost perpetually cloud covered, which along with occasional frosts and snow falls, contribute to the existence of a climatic forest limit at an elevation of around 700–750 m.

SIMON HABERLE

BIOGEOGRAPHY

Australian National University, Canberra

The Juan Fernandez Islands have a very limited fauna, with no native mammals, reptiles, or amphibians. Around 30 landbird and seabird species breed on the islands. The islands have three endemic bird species, the most striking being the picaflor or Juan Fernandez firecrown, Sephanoides fernandensis (Fig. 1), the only endemic hummingbird known on oceanic islands. The present vegetation of the islands is well known with a total of 383 species of flowering plants, which includes 104 endemic, 52 native, and 227 introduced species. There are 51 species of ferns. Table

The Juan Fernandez Islands are located in a warm temperate region of the far southeastern Pacific Ocean and consist of three large volcanic islands that harbor a flora of remarkably high endemism (about 67%). Historic human-induced changes to the island environments and their isolation from a continental landmass have contributed to degradation of the island biota, which is rapidly becoming one of the most threatened in the world. GEOLOGY

The Juan Fernandez archipelago is made up of three large volcanic islands—Isla Robinson Crusoe (or Masatierra; 33°37´ S, 78°51´ W; area 47.9 km2; elevation 915 m), Isla Alejandro Selkirk (or Masafuara; 33°45’ S,´80°45´W; area 49.7 km2; elevation 1380m), and Isla Santa Clara (33°41´ S, 79°00´ W; area 2.2 km2, elevation 374 m)—and several small rocky islets that lie between 570 and 720 km west of the Chilean coast. Currently the islands are administered by the Chilean government and are inhabited by around 600 residents based in the present village of San Juan Bautista on Isla Robinson Crusoe and whose livelihoods are derived primarily from fishing (lobsters), tourism, and cattle. The islands are volcanic in origin, forming over a hotspot that underlies the eastward drifting Nazca plate. These isolated volcanic islands range in age from around 4 to 1 million years old, with the oldest, Isla Robinson Crusoe, having a very rugged and deeply dissected topography compared to the younger Isla Alejandro Selkirk. The islands lie within the northern margins of the subantarctic region of the South Pacific Ocean and are under the influence of major ocean currents: the subantarctic Chile-Peru Current (Humboldt Current) and the subtropical Peru Oceanic Countercurrent. The climate data available for Isla Robinson Crusoe Island records a warm temperate climate with a wet winter and dry summer, a mean annual rainfall of around 1000 mm, and a mean annual temperature of 15.2 °C at sea level. The higher Isla Alejandro Selkirk, 187 km to the west, generates abundant orographic rain, primarily from southwesterly and southeasterly winds, in addition to precipitation delivered from the western storm track. The high altitudes are considered

FIGURE 1 (A) The Juan Fernandez firecrown, Sephanoides fernandensis,

is an endemic hummingbird on Isla Robinson Crusoe, seen here feeding on the nectar of Dendroseris littoralis. (B) Robinsonia gayana, an endemic Asteraceae on exposed southern cliff face, Robinson Crusoe Island.

JUAN FERNANDEZ ISLANDS

507

TABLE 1

Diversity of Species on the Juan Fernandez Islands Endemic

Flowering plants Dicotyledons Monocotyledons Ferns (pteridophytes) Mosses and liverworts Terrestrial birds Insects

Native

Totals

Total

% before 1574

Total

% before 1574

Total

% after 1574

before 1574

after 1574

104 90 14 17 38 3 440

67 58 9 33 24 42 72

52 29 23 34 119 4 170

33 19 14 66 76 58 28

227 195 32 0 0 7 77

59 50 8 0 0 50 11

156 119 37 51 157 7 687

383 314 69 51 157 14 687

1 shows the composition and origin of the flora with the pre-1574 flora consisting of around 67% endemic species. The origin of these plants is dominantly from the southern South American mainland (55%), with smaller proportions being widespread and derived from the Neotropics, Pacific, Australia, and New Zealand. The distinctive rosette-like trees, shrubs and herbs within the Asteraceae (five endemic genera including 26 endemic species derived predominantly from the Americas) have diversified most dramatically on the islands and occupy a wide range of habitats from lowland forest to alpine areas (Fig. 1B). Other genera that have numerous endemic species include Gunnera, Peperomia, Wahlenbergia, Chenopodium, and Eryngium. The remaining forest cover on the islands falls into a subtropical montane rainforest classification that is dominated by trees of Myrceugenia, Drimys, and Fagara. The endemic palm species Juania australis forms an occasional emergent. At higher elevation, dwarf trees, shrubs (Ugni, Pernettya, and Escallonia), and tree ferns (Blechnum and Dicksonia) become dominant. Above the climatic tree line on Isla Alejandro Selkirk, the “alpine tundra” is distinguished by cushion plants (Abrotanella). In all these vegetation zones, invasive species have taken hold, mainly in the lowlands, resulting in more than 70% of endemic species being classified as threatened under IUCN threatened-species criteria. Insects are also very diverse and have evolved extensive adaptations to island ecosystems. This is reflected in the high percentage of endemics (~70%), which is comparable to endemicity within the flowering plants of the Juan Fernandez Islands. The majority of these species are derived from southern Chile, with some Pacific and IndoMalaysian elements present. DISCOVERY AND EXPLORATION

The Juan Fernandez Islands were discovered in 1574, and a small colony of Spanish and South American Indians

508

Introduced

JUAN FERNANDEZ ISLANDS

was established in 1591–1596. They introduced goats and pigs, cut firewood, grew vegetables, and caught and dried fish for the Spanish colonies in Chile. Short-lived settlement of the islands characterized much of the seventeenth and eighteenth centuries, which included groups of castaways or deserters inhabiting the islands for brief periods as British and Spanish interests competed for control. The most famous of these was marooned in 1704, when Alexander Selkirk took voluntary leave of Dampier’s squadron and remained ashore for four and a half years—with his experiences later becoming the basis of Daniel Defoe’s allegorical romance Robinson Crusoe. In 1749 the Spanish Viceroy ordered the formal colonization and protection of the Juan Fernandez Islands, with the construction of a substantial fort in Cumberland Bay and the arrival of 62 soldiers, 171 colonists, and 22 convicts, together with cows, sheep, mules, pigs, and poultry. By the 1790s, the settlement, essentially a penal colony, consisted of about 300 people, although this was abandoned by 1817. Permanent settlement began in 1877, and the islands were declared a Chilean national park in 1935; UNESCO declared the park a world biosphere reserve in 1977. As of 1976, the Chilean national park service, CONAF, has delivered administrative an environmental protection services to the islands. CONSERVATION AND FUTURE CHALLENGES

Despite a relatively short history of permanent human settlement dating back only 130 years, the island flora and fauna are threatened by changes brought about by human activity including over-exploitation of forest and animal resources (e.g., indigenous sandalwood, Santalum fernándezianum, probably extinct since the beginning of the twentieth century, and earlier devastation of fur seal and elephant seal colonies—with 3 million skins being taken from Isla Alejandro Selkirk alone during 1797–1804), soil erosion, fire, introduced vascular plants (around 227

at this time, suggesting that the vegetation became much more degraded after the mid-eighteenth century under the influence of increased human activity. Extensive botanical surveys and photographic comparisons spanning the twentieth century (Fig. 2) show that this process is ongoing and that the rapid invasion of exotic species will need to be halted if threatened and rare species are to survive to the end of the twenty-first century. Since the islands were discovered in AD 1574 by the explorer Juan Fernandez, the conversion of natural vegetation into pastures, the occurrence of extensive fires, and the introduction of alien plant and animal species have had a profound impact on the composition and extent of natural biotic communities. Only one extinction has so far been observed (Santalum fernandezianum), but population sizes of many endemics have become small, with some having less than 25 known individuals left. Continued conservation efforts by CONAF, local residents, and international interests have the potential to ensure the ongoing preservation of this remarkable island environment. SEE ALSO THE FOLLOWING ARTICLES

Deforestation / Exploration and Discovery / Insect Radiations / Invasion Biology / Pigs and Goats FURTHER READING FIGURE 2 Comparison of landscape images in Cumberland Bay, Isla

Robinson Crusoe by (A) George Anson, 1740; (B) Carl Skottsberg, 1918; (C) Simon Haberle, 2000. Notice the replacement of natural forest vegetation by degraded grassland and introduced tree species, including Pinus sp., Eucalyptus sp., Aristotelia chilensis, and Ugni molinae.

invasive species known in 1998), and introduced animals including cattle, goats, and the European rabbit. Descriptions of the islands by explorers such as George Anson dating to the mid-eighteenth century suggest that the lowland valleys were forested. Scientific exploration of the islands has been extensive with the work of the Swedish naturalist Carl Skottsberg in the early twentieth century providing a remarkable account of island environments

Bernardello, G., G. J. Anderson, T. F. Stuessy, and D. J. Crawford. 2006. The angiosperm flora of the Archipelago Juan Fernandez (Chile): origin and dispersal. Canadian Journal of Botany 84: 1266–1281. Castilla, J. C., ed. 1987. Islas océanicas Chilenas: conociemento cientifico y necesidades de investigationes. Santiago: Universidad Catolica de Chile. Dirnböck, T., J. Greimler, P. Lopez, and T. F. Stuessy. 2003. Predicting future threats to the native vegetation of Robinson Crusoe Island, Juan Fernandez Archipelago, Chile. Conservation Biology 17: 1650–1659. Haberle, S. G. 2003. Late Quaternary vegetation dynamics and human impact on Alexander Selkirk Island, Chile. Journal of Biogeography 30: 239–255. Mueller-Dombois, D., and F. R. Fosberg. 1998. Vegetation of the tropical Pacific Islands. New York: Springer. Skottsberg, C. 1953. The vegetation of the Juan Fernandez Islands, in The natural history of Juan Fernandez and Easter Islands, vol. 2 (botany). C. Skottsberg, ed. Uppsala: Almquist & Wiksells, 793–960. Woodward, R. L. 1969. Robinson Crusoe’s island: a history of the Juan Fernandez Islands. Chapel Hill: University of North Carolina Press.

JUAN FERNANDEZ ISLANDS

509

K KARST ISLANDS SEE MAKATEA ISLANDS

KERGUELEN SEE INDIAN REGION

KERMADEC ISLANDS SEE PACIFIC REGION

KICK ’EM JENNY JAN LINDSAY University of Auckland, New Zealand FIGURE 1 Location of Kick ’em Jenny in the Grenadines. Inset shows

Kick ’em Jenny is a submarine basaltic volcano located approximately 8 km north of Grenada in the eastern Caribbean (Fig. 1). It is the most frequently active volcano in the Lesser Antilles island arc and the only known submarine volcano in the region. VOLCANO MORPHOLOGY AND GEOLOGY

Kick ’em Jenny is a conical-shaped volcano that rises 1300 m from the sea floor. It is asymmetric, as it abuts the Grenadines shelf to the east. It has a summit crater about 320 m in diameter, and the highest point on the crater rim is about 180 m below sea level. There is an actively degassing inner crater in the northwest of the essentially flat crater floor, which, at about 265 m below sea level, is the deepest point within the crater. Growth of a dome within the

510

the islands of the Lesser Antilles.

crater between 1976 and 1978 brought the highest point on the volcano to within 160 m of the sea surface, but subsequent collapse during an eruption in 1988 destroyed the dome and breached the crater wall to the northeast. The volcano lies within a large (5 km wide), horseshoeshaped collapse scarp produced during a major sector collapse at some stage in Kick ’em Jenny’s past. Debris avalanche deposits extending up to 30 km from the volcano into the Grenada Basin to the west represent the material carved off the volcano during this event (Fig. 2). The large volume of collapsed material (> 10 km3) indicates that the summit of the ancestral Kick ’em Jenny volcano might have protruded above sea level before the collapse.

HISTORICAL ERUPTIVE ACTIVITY AND HAZARDS

FIGURE 2 View of Kick ’em Jenny and debris avalanche deposits, with

axes shown in km (approx. 2× vertical exaggeration). Inset shows side view at 5× vertical exaggeration.

Kick ’em Jenny eruptions are typically either nonexplosive (producing pillow lava) or explosive (producing tephra), depending on the rate and extent of magma degassing and magma–seawater interaction. The rocks are typically olivine basalt and basaltic andesite and contain unusually high abundances of large amphibole crystals that have been assimilated by the ascending magma from the surrounding crust. The Kick ’em Jenny volcano lies on the western edge of a field of other shallow submarine volcanic features (Fig. 3), and nearby islands represent points where this bathymetric high rises above the sea surface. A sustained eruption from Kick ’em Jenny might cause the volcano to breach the sea surface and develop a small island, as was the case at Surtsey, Iceland, in 1963. With continuing eruptive activity over a long time period (thousands to millions of years), it may grow big enough to join with nearby islands to form one large island, thus illustrating the typical early stages of island growth in the Lesser Antilles arc.

Throughout historical time Kick ’em Jenny has been the most frequently active volcano in the Lesser Antilles, erupting at least 12 times since 1939. The presence of the volcano was first revealed by a phreatomagmatic eruption in July 1939. The eruptions ejected ash-laden columns to heights of up to 300 m above the sea surface and generated water turbulence and numerous small earthquakes. Material was also ejected into the air during eruptions in 1974 and 1988; all other eruptions, including the most recent in December 2001, were entirely submarine. Among the world’s many submarine volcanoes, Kick ’em Jenny is one of the few that is in shallow water close to significant population centers. It lies directly beneath a major trade route frequented by local interisland traffic, large numbers of yachts, and cruise liners. During eruptions, water disturbances and ejection of hot rocks pose a threat to boats in the vicinity of the crater. Continuous hydrothermal venting is occurring in the inner crater as well as on the southwestern flanks of the volcano, and periods of elevated magma degassing (both during and between eruptions) may reduce the water density (and thus the buoyancy of boats) above the volcano. Although there has been significant and variable activity in the crater area of Kick ’em Jenny over the past 70 years, including at least one dome collapse event, to date there have been no confirmed accounts of any tsunami generated by this activity. At the current depth to the crater floor of about 265 m, hydrostatic pressures inhibit the occurrence of large explosive eruptions that could generate tsunamis, but this could change if the volcano grows closer to the surface. Collapse of the ancestral Kick ’em Jenny probably generated a large tsunami. Many similar large collapse scarps and deposits have been identified in other volcanic island settings around the world, including the Caribbean, and this appears to be a normal part of the evolution of these islands. SEE ALSO THE FOLLOWING ARTICLES

Antilles, Geology / Hydrothermal Vents / Island Arcs / Tsunamis / Volcanic Islands FURTHER READING

FIGURE 3 Seabeam image of Kick ’em Jenny and nearby submarine

features (view from the southwest). Colors reflect depth and grade from red (shallow) to dark blue (deep).

Devine, J. D., and H. Sigurdsson. 1995. Petrology and eruption styles at Kick’em Jenny submarine volcano, Lesser Antilles island arc. Journal of Volcanology and Geothermal Research 69: 35–58. Koschinsky, A., R. Seifert, A. Knappe, K. Schmidt, and P. Halbach. 2007. Hydrothermal fluid emanations from the submarine Kick ’em Jenny volcano, Lesser Antilles island arc. Marine Geology 244: 129–141.

KICK ‘EM JENNY

511

Lindsay, J. M., J. B. Shepherd, and D. Wilson. 2005. Volcanic and scientific activity at Kick ’em Jenny submarine volcano 2001–2002: implications for volcanic hazard in the southern Grenadines, Lesser Antilles. Natural Hazards 34: 1–24. Lindsay, J. M., and J. B. Shepherd. 2005. Kick ’em Jenny and Ile de Caille, in Volcanic Hazard Atlas of the Lesser Antilles. J. M. Lindsay, R. Robertson, J. Shepherd, and S. Ali, eds. St. Augustine, Trinidad and Tobago: University of the West Indies, Seismic Research Unit, 108–126. . Smith, M., and J. Shepherd. 1993. Preliminary investigations of the tsunami hazard of Kick ’em Jenny submarine volcano. Natural Hazards 7: 257–277. Smith, M. and J. Shepherd. 1995. Potential Cauchy-Poisson waves generated by submarine eruptions of Kick ’em Jenny volcano. Natural Hazards 11: 75–94. Watlington, R. A., W. D. Wilson, W. E. Johns, and C. Nelson. 2002. Updated bathymetric survey of Kick ’em Jenny submarine volcano. Marine Geophysical Research 23: 271–276.

KI¯PUKA AMY G. VANDERGAST U.S. Geological Survey, San Diego

“Kı¯puka” is one of several Hawaiian terms adopted by geologists to describe volcanic features, and is defined as a fragment of land surrounded by one or more younger lava flows (Fig. 1). Kı¯puka are essentially habitat islands, and when present on islands themselves, kı¯puka can be thought of as islands within islands. Kı¯puka may play a unique role in shaping the ecology and evolutionary trajectories of species. Kı¯puka undoubtedly act as refugia during flow events and provide source populations for colonists throughout ecosystem succession on new lava flows. As lava flows age, they are gradually recolonized by plant and animal communities until the distinction between kı¯puka and surrounding lava flows becomes negligible. Thus, populations of

FIGURE 1 K¯ıpuka formed during the Pu‘u ‘O‘o-Kupaianaha eruption

on the east rift zone of Kilauea Volcano, Hawai‘i. Photographed by J. D. Griggs on January 13, 1987, U.S. Geological Survey.

512

K ¯I P U K A

plants and animals can become cyclically isolated and connected in this landscape, promoting population divergence and, potentially, speciation. Kı¯puka systems also provide an ideal natural laboratory for examining invasions, fragmentation effects, metapopulation dynamics, and the state factors that regulate ecosystem development. KI¯PUKA AS REFUGIA

Kı¯puka can remain ecologically distinct from surrounding lava flows for many decades (and even centuries) after isolation. For example, mature, closed-canopy forest kı¯puka on the island of Hawai‘i are more stable in temperature and humidity (generally cooler and wetter), supporting markedly different understory vegetation and invertebrate communities than surrounding 150-year-old flows. However, the main canopy tree, Metrosideros polymorpha, is one of the first colonizers of bare lava. As these trees mature and fill in, the closed-canopy forest regenerates, colonized from source populations in adjacent kı¯puka. Kı¯puka also provide refuges for native organisms from invasive species, as they seem to be less susceptible to invasions than surrounding lava flows. For example, invasive plant and spider species are largely restricted to younger surrounding lava flows in kı¯puka systems on Hawai‘i. This difference is presumably due to differences in niche availability, with younger and more open environments providing less competition for colonizing species. KI¯PUKA AS ACCELERANTS OF EVOLUTIONARY CHANGE

On volcanic archipelagoes, geologic processes play a central role in shaping patterns of biodiversity. On Mauna Loa and Kilauea volcanoes on Hawai‘i, lava flows cover surfaces at rates of about 40–90% per 1000 years. This ongoing volcanic activity has created a shifting mosaic of habitats as large areas are destroyed or fragmented into kı¯puka, and the new lava flows are subsequently recolonized. Populations of plant and animal species on the slopes of these volcanoes have been subject to repetitive extinctions, fragmentation, founder events, and population growth on time scales of hundreds to thousands of years. Based on studies of Hawaiian Drosophila flies, it has been hypothesized that a combination of metapopulation structure and founder effects can promote genetic differences among populations, sometimes leading to the evolution of new character states and species. Concordantly, recent work has shown that genetic differentiation increases among kı¯puka in native Tetragnatha spiders. The potential importance of isolation of small populations due to lava flows has been invoked in explanations of diversification in several other Hawaiian lineages, as well as those from other systems

such as the Galápagos and Canary Islands and the Albertine Rift of central Africa. KI¯PUKA AS NATURAL LABORATORIES

Although kı¯puka have been studied globally, ranging from sagebrush patches in the northwestern United States to tropical forest fragments in Africa, forested kı¯puka systems in the Hawaiian Islands are understood particularly well. Kı¯puka constitute a dominant landscape feature in these volcanic islands, with hundreds of these habitat fragments found on the Island of Hawai‘i. The substrate ages of many of these kı¯puka and their surrounding flows have been dated using historical records and geologic methods. The juxtaposition of older and younger lava substrates, ranging across gradients of elevation, temperature, soil type, and nutrient content, creates a unique system in which to study the state factors that regulate ecosystem development and function. Primary succession and ecosystem development can be studied across lava flows in a chronological sequence while other potentially confounding factors (such as altitude, substrate type, nutrient load, slope, and aspect) are held constant. Studies of vegetation communities in forest kı¯puka have revealed that mature closed canopy forest can develop on new lava flows in as few as 300 or as many as 3000 years. These temporal differences depend on interactions among both biotic and abiotic factors (slope, aspect, altitude, lava type). Kı¯puka systems can also be useful in investigations of the effects of habitat fragmentation over time and in comparisons of natural versus human-mediated fragmentation. Because habitat loss and fragmentation are major threats to global biodiversity, understanding the long term consequences of natural fragmentation processes may lead to a greater ability to protect biodiversity.

of Hawai‘i. These animals physically alter the understory through feeding and rooting. Impacts to the forest understory in kı¯puka may be severe, particularly in very small fragments, where large proportions of the understory can be affected. Perhaps even more importantly, introduced mammals transport seeds, pollen, and invertebrates as they move easily from patch to patch. These activities can homogenize once-distinct fragments, disrupting ecological and evolutionary processes associated with isolation. In some cases, active management, particularly control of non-native species, may be necessary to preserve the functionality of kı¯puka systems. SEE ALSO THE FOLLOWING ARTICLES

Fragmentation / Invasion Biology / Lava and Ash / Metapopulations / Refugia / Succession / Volcanic Islands FURTHER READING

Carson, H. L., J. P. Lockwood, and E. M. Craddock. 1990. Extinction and recolonization of local populations on a growing shield volcano. Proceedings of the National Academy of Sciences of the United States of America 87: 7055–7057. Vandergast, A. G., and R. G. Gillespie. 2004. Effects of natural forest fragmentation on a Hawaiian spider community. Environmental Entomology 33: 1296–1305. Vandergast, A. G., R. G. Gillespie, and G. K. Roderick. 2004. Influence of volcanic activity on the population genetic structure of Hawaiian Tetragnatha spiders: fragmentation, rapid population growth and the potential for accelerated evolution. Molecular Ecology 13: 1729–1743. Vitousek, P. M., G. H. Aplet, J. W. Raich, and J. P. Lockwood. 1995. Biological perspectives on Mauna Loa Volcano: a model system for ecological research, in Mauna Loa revealed: structure, composition, history and hazards. J. M. Rhodes and J. P. Lockwood, eds. Washington, DC: American Geophysical Union, 117–125.

KOMODO DRAGONS

THREATS TO KI¯PUKA SYSTEMS

Kı¯puka on active volcanoes may not undergo the impact of human use and urbanization to the same degree as other, geologically more stable, environments. However, threats to these systems still exist. Threats from invasive species are prominent in nearly all island systems, even though kı¯puka themselves may be more resistant to invasion than their surrounding younger lava flows. For example, invasions on Hawaiian lava flows by the nitrogen-fixing tree Myrica faya and the fountain grass Pennisetum setaceum drastically alter nutrient availability and fire regimes, decreasing the ability of native species to colonize these flows and disrupting the natural successional progression. Introduced ungulates are well established even in remote forest kı¯puka on the island

TIM JESSOP Zoos Victoria, Parkville, Australia

The Komodo dragon (Varanus komodoensis) is infamous for being the world’s largest lizard, reaching a maximum length of 3 m and a maximum body mass of up to 87 kg (Fig. 1). As adults, this monitor lizard is capable of killing and consuming mammals, including water buffalo (Bubalus bubalis), Timor deer (Cervus timorensis), and wild pigs (Sus scrofa), which coexist on five rugged islands in eastern Indonesia. Monitor lizards are monophyletic and comprise three clades; the Komodo dragon is assigned to the Indo-Australian lineage. Its closest sister species, as inferred from a mitochondrial gene tree, is the Eastern Australian lace monitor (Varanus

KO M O D O D R AG O N S

513

ISLAND POPULATION DEMOGRAPHY

FIGURE 1 Two adult male Komodo dragons engaged in combat during

the winter mating period. Photograph by Achmad Ariefiandy.

varius). The Komodo dragon is estimated to have differentiated from Australasian ancestors ~4 million years ago, and historically, its range extended further east across to the island of Timor. DIET AND DISTRIBUTION

Ungulate prey constitutes the majority of the diet of large Komodo dragons (i.e., > 15 kg in body mass), whereas smaller dragons consume small reptiles, birds, and rodents. Within Komodo National Park, where this species is best studied, Komodo dragons persist on four islands that vary in area from 10 km2 up to 350 km2. Across these islands, the availability and distribution of ungulate prey varies considerably, with deer being found on all islands but buffalo and pigs only on the two larger, and human-habited, islands. Differences in large-prey density and availability appear to be an important factor underpinning ecological and evolutionary processes influencing Komodo dragons. PHENOTYPIC DIVERGENCE AMONG ISLANDS

Major differences in the morphology of the Komodo dragon appear to be associated with conspicuous island differences in large-prey availability. For instance, maximal body size among the four islands varies nearly fourfold, indicating that the Komodo dragon exhibits both dwarf and giant populations. Maximal body size is positively correlated with insular prey density, independent of genetic relatedness among lizard populations. Hence, even genetically similar island populations can exhibit large differences in body size associated with local prey density, suggesting considerable phenotypic plasticity in body size. Other ecological hypotheses that are implicated in morphological divergence among island populations, including character displacement due to competition or predation, are intuitively unlikely given that this lizard is the sole large predator persisting on these islands.

514

KO M O D O D R AG O N S

Differences in island prey density appear to be associated with conspicuous differences in the population demography of Komodo dragons in Komodo National Park. For example, low densities of dwarf Komodo dragons persist on the two small islands, in contrast to high densities of large-bodied lizards occurring on the two big islands. Despite individuals exhibiting a much smaller body size, and in turn decreased energetic requirements, the two small-island dragon populations appear to be uncompensated by increased population density. Thus an inverse relationship between body size and population density, as predicted by the energetic equivalence rule, is not observed for Komodo dragons. SPATIAL HABITS AND DISPERSAL

On emergence from their subterranean nests, hatchling Komodo dragons (~100 g) directly climb trees and exhibit an arboreal life-history stage. Limited natal dispersal is apparent, with juveniles moving slowly in a mostly linear direction, away from their nests. Arboreal living by juvenile Komodo dragons presumably reduces predation from larger terrestrial dragons, while enabling access to smaller arboreal prey. Once reaching several kilograms, juvenile lizards transition to predominantly terrestrial activity and develop home ranges of ~0.25 km2. Spatial requirements increase considerably with body size, with the largest adult lizards occupying home ranges of ~7 km2. Long-distance dispersal of Komodo dragons within and among islands appears to be an ecologically rare event. Even within the large islands, the exchange of individuals (based on mark-recapture estimates and telemetry) between closely adjacent valleys is infrequent. Komodo dragons thus exhibit strong philopatry to their resident valleys (and to islands), a feature also generally supported by molecular data indicating significant genetic structure within, and among, island populations. POPULATION GENETICS AND CONSERVATION

Neutral genetic estimates suggest varying population differentiation among extant populations. The populations of Rinca, Nusa Kode, and Western Flores are the most closely related, reflecting their proximity to one another and their higher rates of gene flow. In contrast, the more isolated populations are more differentiated. The Komodo island population exhibits the highest level of genetic divergence and allelic distinctiveness. In contrast, the small island of Gili Motang exhibits a low level of hetrozygosity, consistent with its small population size, suggesting that this island is most at risk from stochastic processes.

Demographic evidence from Komodo National Park suggests the two small island populations are declining in abundance, while the two large island populations appear relatively stable. The factors underpinning the declines on the two small islands are not clearly identified. Nevertheless, given the inherent vulnerability of island populations to extinction, and that Komodo dragons appear to exhibit limited dispersal among islands, ongoing and robust population monitoring is advocated to ensure that extant populations of Komodo dragons, both within and outside Komodo National Park, are conserved. SEE ALSO THE FOLLOWING ARTICLES

Dispersal / Dwarfism / Gigantism / Indonesia, Biology / Island Rule / Lizard Radiations FURTHER READING

Auffenberg, W. 1981. The behavioral ecology of the Komodo monitor. Gainesville: University Presses of Florida. Ciofi, C., J. Puswati, D. Winana, M. E. De Boer, G. Chelazzi, and P. Sastrawan. 2007. Preliminary analysis of home range structure in the Komodo monitor, Varanus komodoensis. Copeia 2007: 462–470. Imansyah, M. J., T. S. Jessop, C. Ciofi, and Z. Akbar. 2007. Ontogenetic differences in the spatial ecology of immature Komodo dragons. Journal of Zoology doi: 10.1111/j.1469-7998.2007.00368.x. Jessop, T. S., T. Madsen, J. Sumner, H. Rudiharto, J. A. Phillips, and C. Ciofi. 2006. Maximum body size among insular Komodo dragon populations covaries with large prey density. Oikos 112: 422–429. Jessop, T. S., T. Madsen, J. Sumner, H. Rudiharto, J. A. Phillips, and C. Ciofi. 2007. Differences in population size structure and body condition: conservation implications for Komodo dragons. Biological Conservation 135: 247–255. Murphy, J. B., C. Ciofi, T. Walsh, and C. de la Panouse, 2002. Komodo dragons: biology and conservation. Washington DC: Smithsonian Institution Press.

as a bearded, white-skinned individual with reddish or blond hair and blue eyes who came to South America from across the Atlantic. Heyerdahl stated that this god remained in Peru for a time, before being driven out into the Pacific, where he was subsequently known to Polynesians as “Tiki” and worshipped as a sun god. The name “Kon-Tiki” is a distorted rendering of the Quechua title for one of the many aspects of the sun god in Peru: ápu qon téXsi wiraqúcha, where the letter X represents a sound similar to that of k in milk. The word teXsi is therefore not a cognate of the Marquesan word tiki. The Polynesian Tiki was not a god but a Marquesan mythological character who created the first human by copulating with a heap of sand. No sun gods were present in the Polynesian pantheon; all Polynesian gods were deified ancestors. The legends of Viracocha do not describe a migration from Peru into the Pacific by a white-skinned god and his followers. They describe the solar deity, with shining hair and beard, moving across the sky on his annual cyclical course between solstices. At the winter solstice, the sun, as seen from Peru, drops lower on the northwestern horizon, as though sinking into the Pacific. The supposed Caucasoid features of Viracocha appealed to Heyerdahl, who at one time was a correspondent of Hans Guenther, the author of Nazi racist anthropological texts that emphasized the superiority of the Nordic “race.” The “red” beard and blue eyes of Viracocha are nowhere mentioned in Peruvian legends; these attributes were added by Heyerdahl to strengthen the Nordic connection. THE KON-TIKI RAFT: CONSTRUCTION AND LOGISTICS

KON-TIKI ROBERT C. SUGGS Boise, Idaho

In 1947, the Norwegian explorer Thor Heyerdahl made a drift voyage of 101 days from Peru to Polynesia on the balsa-log raft Kon-Tiki. Heyerdahl’s stated purpose was to prove his theory that Polynesia was settled from South America by light-skinned followers of a foreign god. WHO WAS “KON-TIKI?”

Heyerdahl named the Kon-Tiki after the god he identified as “Kon-Tiki Viracocha.” This god was described

The Kon-Tiki raft was constructed from balsa logs. It generally resembled balsa log rafts seen by early European explorers of South America, such as Bartolomé Ruiz in 1526. There was never any serious question about the ability of balsa logs to survive long immersion in sea water: balsa rafts were a standard part of pre-European South American maritime technology. Kon-Tiki had a square-rigged sail, a large steering oar, and keel boards, guaras, to aid in tacking. A deckhouse sheltered Heyerdahl and his five companions, including a cameraman and a radioman. The main food source was a large quantity of canned U.S. Army rations. Coconuts, bananas, and sweet potatoes were also taken aboard for symbolic value, but they spoiled within approximately three weeks. Primus stoves were used for cooking. The canned food was supplemented by 275

K O N -T I K I

515

liters of water in cans. This quantity of water would have provided each man with approximately a half-liter per day, an amount insufficient for life under exposed conditions in subequatorial seas. Additional liquid was obtained from raw fish, from rain, or by mixing sea water with canned water, although solar stills supplemented the water supply. The essential modern life support supplies, including canned food and water, stove, lanterns, radio, electric generator, and solar water stills, invalidate the voyage as a test of pre-European Peruvian voyaging capability. AT SEA

The Kon-Tiki was unable to cross the strong Humboldt Current, which flows northward along the Peruvian coast. The craft was therefore towed out to sea by the Peruvian Coast Guard and was released more than 50 miles from the coast. Once the raft was released from the tow, steering proved impossible. The raft drifted at the whim of winds and currents. Upon reaching the Tuamotu Archipelago, the Kon-Tiki passed near several atolls, finally washing ashore on the atoll of Raroia after 101 days adrift. The crew escaped without injury.

FIGURE 1 A photograph used in advertising the documentary film

depicts the Kon-Tiki approaching Raroia, Tuamotu Archipelago. Image courtesy of Janson Media.

SIGNIFICANCE

Heyerdahl claimed that the voyage proved his theory according to which Caucasoid Peruvians first settled Polynesia and were later followed by an invasion of Native Americans from the Northwest Coast. This single drift voyage provided no new facts or insights to challenge the data already available from the disciplines of archeology, ethnology, physical anthropology, and linguistics, all of which pointed to a Central-Western Pacific origin for the Polynesians. Additional archeological evidence against Heyerdahl’s theory appeared in the decade following the Kon-Tiki voyage, and today, after 60 years of further archeological investigations in Polynesia, no evidence of South American contact has ever been discovered anywhere in Polynesia. Although Peruvians never reached Polynesia, the Polynesians, descendants of an Asian maritime culture that originated on the coast of China and Taiwan approximately 4500 years BCE, are now known to have reached the New World in pre-Columbian times. The Kon-Tiki raft voyage was a great financial and public relations success. Heyerdahl’s book sold millions of copies, in many languages, and the film of the voyage won an Academy Award (Fig. 1). It also successfully proved that a group of physically fit personnel, on a crude balsa raft, with modern survival supplies and equipment, could survive an uncontrolled drift voyage from South

516

K O N -T I K I

America to Polynesia—once they were towed out beyond the Humboldt Current, which otherwise would have carried them up the South American coast to Panama. SEE ALSO THE FOLLOWING ARTICLES

Easter Island / Peopling the Pacific / Polynesian Voyaging / Popular Culture, Islands in FURTHER READING

Bierbach, A., and H. Cain. 1996. Religion and Language of Easter Island. Baessler-Archiv, Neue Folge. Beiheft 9. Berlin: Dietrich Reimer Verlag. Clark, R. 1990. Austronesian languages, in The world’s major languages. B. Combe, ed. New York: Oxford University Press, 899–912. Demarest, A. A. 1981.Viracocha: the nature and antiquity of the Andean high god. Peabody Museum Monographs, No. 6. Cambridge, MA: Harvard University. Finney, B. 1994. Voyage of rediscovery: a cultural odyssey through Polynesia. Berkeley: University of California Press. Heyerdahl, T. 1950. Kon-Tiki: across the Pacific by raft. New York: Rand McNally and Company. Kirch, P. V. 2000. On the road of the winds: an archeological history of the Pacific islands before European contact. Berkeley: University of California Press. Liping Jiang and Li Liu. 2005. The discovery of an 8000-year-old dugout canoe at Kuahuqiao in the Lower Yangzi River, China. Antiquity 79 (305). Suggs, R. C.1960. Island civilizations of Polynesia. New York: New American Library. Storey, A. A., J. M. Ramirez, D. Quiroz, D. V. Burley, D. J. Addison, R. Walter, A. J. Anderson, T. L. Hunt, J. S. Athens, L. Huynen, and E. A. Matisoo-Smith. 2007. Radiocarbon and DNA evidence for a preColumbian introduction of Polynesian chickens to Chile. Proceedings of the National Academy of Sciences of the United States of America 104: 10335–10339.

KRAKATAU ROBERT J. WHITTAKER Oxford University, United Kingdom

The complete, or near-complete, sterilization of the Krakatau Islands in a devastating sequence of volcanic eruptions in 1883 provided a remarkable opportunity for natural scientists to monitor the processes and patterns of island recolonization and primary succession. Subsequent survey data, although intermittent in nature, extend to the present day and collectively (1) form a well-specified descriptive account of ecosystem development and (2) provide valuable opportunities for testing theories concerning the turnover and dynamics of insular systems. Most notably, Krakatau was the first case study system of colonization and turnover used by Robert H. MacArthur and Edward O. Wilson in evaluating their dynamic equilibrium model of island biogeography. Their findings, based on species data for the period 1883–1934, were equivocal, showing the apparent establishment of dynamic equilibrium for birds, but not for plants. Subsequent research suggested a slight upward drift in numbers of bird species and a continuing increase in plant species number and a variety of patterns for other taxa, consistent with strong successional structure in the patterns of recolonization. An overall dynamic equilibrium across different taxonomic and ecological groups had not been established by the start of the twenty-first century. THE ENVIRONMENTAL CONTEXT AND HISTORY OF ERUPTIVE ACTIVITY

The Krakatau Islands (6° S, 105° E) are located in the Sunda Strait and are roughly equidistant between Java (approximately 40 km away) and Sumatra (30 km). The islands experience a tropical seasonal climate with a few dry months, classified as “Afa” in the Koeppen system. Because of their proximity to a point of lateral stress crossing a destructive plate margin, the Krakatau Islands have undergone repeated phases of volcanic activity. They experienced at least one caldera collapse event prior to 1883, when the group consisted of three islands, in order of diminishing area, Rakata (730 m maximum altitude, 17 km2 area; formerly “Krakatau” or Pulau Rakata Besar), Sertung (180 m, 12 km2; Verlaten Island), and Panjang (140 m, 3 km2; Rakata Kecil, Lang Island). Little is known about the islands prior to the events of 1883, although it is established that they were forestcovered and had been largely dormant from 1680 until

May of 1883, when one of three volcanic cones (Perboewatan, Danan, and Rakata) on the largest island began a series of eruptions that ended on August 27, 1883, in exceptionally destructive eruptions. Two-thirds of that island, including the volcanoes Perboewatan and Danan and part of Rakata, were violently displaced, with vast quantities of ejecta thrown into the atmosphere, and pyroclastic surges crossing the sea to slam into mainland Sumatra. An estimated 36,000 people lost their lives in the coastal fringes of the Sunda Strait, mostly as a result of a series of tsunamis generated in the collapse. The advent of the telegraph shortly before the eruptions meant that news of the event traveled rapidly around the world, initiating an enduring scientific interest in the causes and consequences of the eruption, with the latter including pressure waves that encircled the world several times and a slight global cooling over the following 1–2 years. ISLAND STERILIZATION

Although the main island (now known as Rakata) lost the majority of its land area, all three islands also gained extensive areas of new land resulting from the emplacement of pyroclastic deposits on to the preexisting foundations. These strata were some 60 to 80 m in thickness in the lowlands, and to this day the vast majority of the three islands remain mantled in these unconsolidated ashes, with little solid geology exposed at the surface. The islands can be taken to have been as near to completely sterilized in August 1883 as to make no practical difference, although this conclusion was bitterly contested by C. A. Backer in the early twentieth century. No evidence for any surviving plant or animal life was found by the scientific team led by the geologist R. D. M. Verbeek later in 1883, and in May 1884 the only life spotted by visiting scientists was a spider. The first signs of plant life, a “few blades of grass,” were detected in September 1884. ANAK KRAKATAU

The islands remained dormant until June 1927, when activity commenced once again from the sea bed in the center of the group. By August 12, 1931, a new island, Anak Krakatau (Child of Krakatau), had established a permanent presence, reaching nearly 50 m in height in little over a year. This island has remained highly active and now exceeds 300 m in elevation and 3.5 km2 in area. As a result of its active volcanism, relatively little of this island has become vegetated, and Anak Krakatau has also caused repeated episodes of widespread accelerated mortality of trees within forests on Panjang and Sertung Islands but not, to date, on Rakata.

K R A K ATA U

517

FAUNAL AND FLORISTIC SURVEYS

The fauna and flora of Krakatau have thus colonized since 1883 from an array of potential source areas, many of which were also impacted by the eruption; the closest of these, the island of Sebesi, is 12 km distant. The first botanical survey data are from 1886, with biological survey work being carried out intermittently since then (e.g., around 1897, 1905, 1919–1932, 1951, 1979, and 1983 onward). The 1886 survey recorded a few beach plants, and in the interiors it recorded mosses, blue-green algae, ferns, and a few higher plants (grasses, composites). Other life forms arrived quickly thereafter, and by 1897 Rakata supported young trees interspersed within tall, dense grasslands and an abundance of ferns. Since then, cumulative data from

botanical surveys indicate a marked and rapid increase in the colonization of vascular plants on Krakatau, although few additional solely sea-dispersed plant species have colonized since 1930 (Fig. 1). By 1934, Rakata, Panjang, and Sertung islands collectively held nearly 300 plant species, and by 1983, between 423 and 456 species had been documented. The cumulative total of vascular plants now stands at approximately 540 species. Colonization by animals was less well documented than that of the flora, but nonetheless, data that have been gathered provide valuable information for the analysis of trends in colonization, extinction, and turnover. The islands have been colonized by a wide variety of invertebrate species and at least 89 vertebrate species, including

Phase 1

Phase 3

Sea Rapid colonization of strand-line Simple succession

Animal Delayed colonization of interior

1.64

Sea

Animal

Colonization rate declines Diplochorous species spread to interior by birds and bats

More forest trees arrive Local populations of bats and birds critical to mozaic development

0.29

0.14

1.0 Ferns

0.83 Other

0.83 Ferns

1.0 Other

Wind

Wind

Forest epiphytes increase as habitat availability increases

Rapid colonization of interior by pioneers (ferns, grasses, composites)

Phase 2 Sea

1.26

Restricted groups: colonization improbable Animal

Colonization continues at a reduced pace

Colonization rate increases as attraction to frugivores increases 1.32

1.18

0.86 Other

0.18 Ferns

Wind Pioneers lose ground

Sea Habitat restrictions only

Animal Large seeded bat-spread or terrestrial mammal-spread species

Wind Species with large, heavy, winged seeds, such as many mature forest canopy trees

FIGURE 1 Plant recolonization of Rakata Island (Krakatau group) since sterilization in 1883. The three phases correspond with survey periods and

represent convenient subdivisions of the successional process. Phase 1, 1883–1897; phase 2, 1898–1919; phase 3, 1920–1989 (subsequent survey data are not included in this analysis). Arrow widths are proportional to the increase in cumulative species number in species year1 (these values are also given by each arrow). The flora is subdivided into the primary dispersal categories (i.e., the means by which each species is considered most likely to have colonized: animal-dispersed [zoochorous], wind-dispersed [anemochorous], and sea-dispersed [thalassochorous]). The model distinguishes between strandline (outer circle) and interior habitats, and in the fourth figure identifies constraints on further colonization. (From Whittaker and Jones, 1994b.)

518

K R A K ATA U

54 birds, 11 microchiropterans (micro-bats), 8 pteropodid fruit bats, 11 reptiles, two snakes, two rats, and a pig (of uncertain origin and identity). COLONIZATION, SUCCESSION, AND TURNOVER

Trends in colonization and loss of species have been analyzed by several authors within the framework of MacArthur and Wilson’s equilibrium theory of island biogeography, largely to determine whether the data support a smooth (i.e., monotonic) progression of declining immigration and increasing extinction rates, approaching asymptotically the dynamic equilibrium condition predicted in their model. Based on data for the first 50 years, MacArthur and Wilson found support for this process in birds but not plants, which continued to increase steadily in number and showed no sign of approaching an asymptote. Subsequent analyses reported a further slight rise in bird numbers, a continued increase in plant species numbers, and a variety of patterns for other taxa (e.g., only two species of lizards have gone extinct, in both cases due to habitat loss, with no evidence of a dynamic equilibrium condition featuring continued turnover). Opinions have differed as to whether these various results and analyses can be incorporated within the framework of MacArthur and Wilson’s theory: The findings clearly require at least some modification of their basic colonization model to recognize the successional structure inherent in the system (Fig. 1). Summarizing a great deal of empirical detail, succession along the coastline was driven by the colonization of sea-dispersed plants, was rapid, and involved little compositional change and little turnover. The set of coastal plant species establishing early and on each island have undergone next to no species extinction, although some species (e.g., mangrove species) repeatedly reach the islands but fail to establish because of lack of habitat. In the interiors, the first colonists were exclusively winddispersed species: Presumably the islands held no interest to passing frugivorous birds or bats at this stage. However, within the first decade of the recolonization commencing, a limited local source of fruit would have become available in the strand lines, as some of the early colonizing sea-dispersed species were diplochores, producing fruit that, in addition to their primary dispersal mode, provided a fruit resource for vertebrate frugivores. This inference is consistent with reports of the first restrictively animal-dispersed (zoochorous) plants being found in close association with the strandline vegetation. Some of these early colonizing strictly zoochorous plant species were fig (Ficus) species, subsequently shown to be bat dispersed. Indeed, fruit bats are now understood to have had

a pivotal role in the early stages of forest development, introducing several ecologically important species and extending their distributions across the island interiors. However, frugivorous birds have introduced a much larger number of plant species, with fruit pigeons playing a particularly important role in the process. The initially open fern-, herb-, and grass-dominated vegetation types (most species of which were wind-dispered) gradually gave way to forest, in which most trees and shrubs were bat- or bird-spread. As the forest closed over during the 1920s, the rapid reduction in open habitat drove crashes in populations of open-habitat animal and plant species, leading to a measurable pulse of islandwide species extinctions. Coincidentally, the wide availability of new forest habitat was matched by the establishment of forest-specialist species, including many wind-dispersed epiphytic plants (e.g., ferns and orchids). The turnover pulse during forest closure is evidenced in the island colonization data for several taxa and is responsible, for example, for MacArthur and Wilson’s premature claim for a dynamic equilibrium in bird species numbers. ONGOING ACTIVITY

The emergence of Anak Krakatau around 1927–1931 has added a further dimension to the “natural experiment” of the Krakatau islands, both providing a further island for a rerun of recolonization and succession and impacting heavily on two of the other islands, Sertung and Panjang, which intermittently suffer significant disturbance from volcanic ejecta produced by Anak Krakatau. This is evidenced in deposits of between 1 and 2 m in depth of mostly fine-grained volcanic products across the great majority of each of these islands. To date, Rakata Island has not been directly affected in this way. Partly as a consequence, there are some differences in the succession of forest types evident across the different islands. The Krakatau Islands remain of considerable interest to natural scientists. Geologists continue to debate the details of the 1883 events and to monitor Anak Krakatau’s activity against the prospect of future significant hazard to the inhabitants of the Sunda Strait region. Recent biological work has included monitoring of forest dynamics, studies of bat–plant and fig wasp–fig mutualisms, and analyses of the genetic relationships of colonist plant species to populations from potential source areas. In recognition of their special scientific interest, the islands enjoy protected status and have been largely uninhabited since 1883, although they are regularly visited by scientists, fishermen, tourists, illegal pumice gathering teams, and others.

K R A K ATA U

519

SEE ALSO THE FOLLOWING ARTICLES

Eruptions: Laki and Tambora / Extinction / Island Biogeography, Theory of / Succession / Tsunamis FURTHER READING

Compton, S. G., S. J. Ross, and I. W. B. Thornton. 1994. Pollinator limitation of fig tree reproduction on the island of Anak Krakatau (Indonesia). Biotropica 26: 180–186. Shilton, L. A., J. D. Altringham, S. G. Compton, and R. J. Whittaker. 1999. Old World fruit bats can be long-distance seed dispersers through extended retention of viable seeds in the gut. Proceedings of the Royal Society, London B 266: 219–223. Simkin, T., and R. S. Fiske, eds. 1983. Krakatau 1883—the volcanic eruption and its effects. Washington, DC: Smithsonian Institution Press. Thornton, I. W. B. 1996. Krakatau: the destruction and reassembly of an island ecosystem. Cambridge, MA: Harvard University Press. Whittaker, R. J., M. B. Bush, and K. Richards. 1989. Plant recolonization and vegetation succession on the Krakatau Islands, Indonesia. Ecological Monographs 59: 59–123. Whittaker, R. J., R. Field, & T. Partomihardjo. 2000. How to go extinct: lessons from the lost plants of Krakatau. Journal of Biogeography 27: 1049–1064. Whittaker, R. J., and S. H. Jones. 1994a. The role of frugivorous bats and birds in the rebuilding of a tropical forest ecosystem, Krakatau, Indonesia. Journal of Biogeography 21: 245–258. Whittaker, R. J., and S. H. Jones. 1994b. Structure in re-building insular ecosystems: an empirically derived model. Oikos 69: 524–530. Whittaker, R. J., S. H. Jones, and T. Partomihardjo. 1997. The re-building of an isolated rain forest assemblage: how disharmonic is the flora of Krakatau? Biodiversity and Conservation 6: 1671–1696.

KURILE ISLANDS ALEXANDER BELOUSOV AND MARINA BELOUSOVA Institute of Volcanology and Seismology, Petropavlovsk, Russia

THOMAS P. MILLER U.S. Geological Survey, Anchorage, Alaska

The Kurile (or Kuril) Islands are one of the last blank spots on the world map, and their very remoteness results in a uniquely pristine environment. The biodiversity of the islands is remarkable, ranging from broad-leaved subtropical forests with magnolia, ligneous lianas, and Kurile bamboo in the south to subarctic moss tundra, alder shrubs, and stunted birches in the north. The landscapes are impressive, combining rocky capes, heavy fogs, surrealistic volcanic cones, boiling crater lakes, and almost impenetrable giant grasses. The Kurile Islands have often been compared to the nearby Aleutian Islands, and with good reason in terms of geology, remoteness, and noto-

520

KURILE ISLANDS

riously bad weather. But the critical, though usually ignored, difference in orientation between the 1200-kmlong, northeast-trending Kuriles and the 1800-km-long, east-west-trending Aleutians results in major differences in climate and accordingly in flora and fauna. GEOGRAPHY General Description

The Kurile Islands are located in the northwestern part of the Pacific Ocean, forming a 1200-km long island arc stretched over 8° of latitude from the Kamchatka Peninsula (Russia) southward to the island of Hokkaido (Japan) (Fig.1). This island arc separates the Sea of Okhotsk from the Pacific Ocean and represents an important geographical and geological boundary. The arc consists of 22 main islands and 30 smaller islets with a total area of 15,600 km2. The largest islands—Iturup (3200 km2), Paramushir (2000 km2), Kunashir (1500 km2), and Urup (1450 km2)—are very narrow across the arc and extended along the arc, in contrast to the smaller islands, which tend to be oval or irregularly shaped. The Kuriles are subdivided longitudinally into two approximately parallel island chains: the Greater Kuriles and the much shorter Lesser Kuriles, located in the southern part of the arc. The chains are separated by the 50-km-wide and 130-m-deep Southern Kurile Strait. The Lesser Kuriles consist of Shikotan Island and six small islands (the Habomai group). The Greater Kuriles include all of the remaining Kurile Islands, from Shumshu southward to Kunashir. Both island chains represent emerged summits of approximately parallel undersea ridges: the Greater Kurile Ridge, which connects to the Shiretoko Peninsula of Hokkaido, and the Lesser Kurile Ridge or Vityaz Ridge, which connects to the Nemuro Peninsula of Hokkaido. The oceanic slope of the Vityaz Ridge descends into the deep (10,542 m) KurileKamchatka Trench, which lies along the entire length of the archipelago and represents the surface expression of subduction of the Pacific Plate under the Okhotsk Plate (formerly considered part of the North American Plate). The island arc is subdivided transversely into three groups of islands separated by deep and wide straits. The Northern Kurile Islands (Shumshu to Shiashkotan) are separated from the Central Kurile Islands (Matua to Simushir) by the Kruzenstern Strait (1900 m deep and 80 km wide). The Central Kurile Islands are, in turn, separated from the Southern Kurile Islands (Chirpoy to Kunashir) by the Boussole Strait (2300 m deep and 67 km wide). Most of the Kurile Islands have mountainous relief punctuated by tall volcanoes, many of which are active. The highest volcanoes are Alaid (2339 m, Atlasova Island), Tyatya (1819 m, Kunashir Island), and

Chikurachki (1816 m, Paramushir Island). Tyatya is considered one of the most beautiful volcanic cones in the world (Fig. 2). Shumshu Island and islands of the Lesser Kuriles have no volcanoes and a low flat relief. Although rivers and lakes are common on the larger Kurile Islands, several small islands have no sources of drinking water. Island rivers are commonly short and rapid (whitewater rivers) with many waterfalls. The 140-m high Ilia Murometz waterfall on Iturup is one of the highest in Russia. Many of the lakes are located in volcanic craters and calderas. The deepest (>264 m) and most beautiful is Kol’tsevoye (Circular) Lake (Fig. 3), which is located inside the Tao-Rusyr caldera at Onekotan Island. Some lakes and rivers located in hydrothermal areas have acid thermal waters, where special thermophilic microorganisms and algae flourish (e.g., Kipyasheye Lake on Kunashir, with a pH of 2.8, and Yur’eva river on Paramushir, with a pH of 1.6). Climate

The Kuriles, in general, have a maritime monsoon climate influenced by sea currents (both cold and warm, Fig. 1) of the Pacific Ocean and the Sea of Okhotsk, as well as by air masses coming from eastern Asia or the Bering Sea region. The southern part of the Sea of Okhotsk is under the influence of the warm Soya sea current, whereas the cold Kamchatka current travels south along the Pacific coast. As a result, the climate of the western slopes of the largest southernmost islands, warmed by the Soya current and protected by high ridges from the cold Pacific, is close to subtropical. The climate of the eastern slopes is notably colder, resulting in strikingly different vegetation. The northernmost islands and the small islands of the Central Kuriles are surrounded by the cold sea and have a subarctic climate. Precipitation is high throughout the year, from 700 to 1000 mm on the northern islands and 1000 to 1100 mm on the southern islands. Thus, the climate is rather humid, and the islands are almost continuously shrouded by cloud and fog; extended rain (drizzle) is common. In winter the precipitation occurs in the form of heavy snowfalls; snowstorms are frequent. By the end of winter, the Sea of Okhotsk is extensively choked by ice fields that can block the western coasts of the islands. Population

Similar artifacts (pottery, stone tools, etc.) dated 14,000– 11,000 years ago have been found in Japan, southern Kamchatka, and southern Alaska, indicating that in the past the Kurile Islands formed a migration route between Japan and Kamchatka that could have been involved in maritime migrations in and out of the North American continent.

FIGURE 1 Schematic map showing the location of the Kurile Islands

with inset showing their location in the Northwest Pacific. In the inset, the surface currents are shown with directions of circulation. SWC: Soya warm current; KC: Kamchatka Current.

Ancient settlements have been discovered on almost all the Kurile Islands. During the last 7000 years, the Kurile Islands were inhabited by several ethnically different groups of people that replaced one another. First known settlers were people of the Jomon (7000–2000 years ago), Epi-Jomon (2000–1300 years ago), and Okhotsk I and II (1300–700 years ago) cultures. Ainu people inhabited the islands after 700 years ago but were gradually displaced by Japanese and Russians in the eighteenth century. Since the end of World War II in 1945, all the islands politically belong to the Sakhalinskaya oblast’ (Sakhalin District) of Russia, although Japan claims the southernmost islands (Iturup, Kunashir, and the Lesser Kuriles).

KURILE ISLANDS

521

FIGURE 2 Tyatya volcano, Kunashir Island. View from the southwest

from the Pacific coast. Photograph by A. Belousov.

Only the largest islands (Paramushir, Iturup, Kunashir, and Shikotan) are presently inhabited, and the population of about 20,000, mostly Russian fishermen and coast guard personnel, is concentrated in several small towns.

of 5000. About 50 small to moderate-scale tsunamis have been recorded since 1952, and studies of paleotsunami deposits revealed multiple strong tsunamis throughout the Holocene in the Kuriles. Great earthquakes on November 16, 2006, and January 13, 2007, with magnitudes of 8.3 and 8.1, generated tsunamis more than 20 m high that affected unpopulated shores of the Central Kuriles. At the present time, the Kurile Islands experience slow ground deformation between major local earthquakes and more rapid deformation (commonly in a reverse direction) during earthquakes. Long-term tide gauge data show that the west coast of Shikotan Island was uplifting at a rate of 12.6 mm/yr until the October 5, 1994, earthquake, when it experienced a 50-cm drop. Recent GPS measurements have shown horizontal motion of the south of Urup Island, with a rate of 18 mm/year in a direc-

GEOLOGY Tectonic Setting

The Kurile Islands were formed by geological processes associated with subduction of the Pacific Plate under the Okhotsk Plate. The rate of subduction of the Pacific Plate is estimated at 95 mm/yr in the north (where the plate motion is normal to the Kurile Trench) and 100 mm/yr south of Boussole Strait (where the arc makes a sharp 22–23º turn to the west and plate subduction becomes oblique). Boussole Strait is considered to be a graben formed by northeast-southwest tension, caused by westward motion of the southern part of the arc due to oblique subduction of the Pacific plate since the Late Miocene (6–7 million years ago). The dip angle of the subduction plane is 48–55º in the northern part and 38–46º in the southern part of the arc. Crustal thickness is 25–36 km below the Northern Kuriles, 26–32 km below the Central Kuriles, and 25–44 km below the Southern Kuriles. Overall, the Kurile Arc is seismically and volcanically much more active than the Izu-Bonin–Mariana and Ryukyu Arcs to the south, but less active than Kamchatka to the north. Six earthquakes with magnitude >8 were recorded in the twentieth century. Seismicity in the subducting slab occurs to the depth up to 650 km. The most intense seismicity is recorded in the southern sector of the Kurile Arc. Some earthquakes and volcanic eruptions have generated tsunamis. The most deadly historic tsunami, up to 20 m high, occurred in 1952 in the northern part of the arc, when the town Severo-Kurilsk (Paramushir Island) along with multiple fishing settlements of the Pacific coast of the islands were demolished, with an estimated death toll

522

KURILE ISLANDS

FIGURE 3 Kol’tsevoye lake located in Tao-Rusyr caldera. Since the cal-

dera formed 7500 years ago, Krenitzin Peak stratovolcano has formed in the central part of the caldera. The 1952 crater and the lava dome are visible on the slope and at the foot of the volcano. Photograph by A. Belousov.

tion coinciding with the direction of subduction. During the period of the winter 2006–2007 earthquakes, Ketoy Island experienced horizontal motion exceeding 60 cm in a direction opposite to the subduction direction. Stratigraphy

The Kuriles are built of predominantly volcanic rocks (both volcaniclastic and effusive), and chemical and biochemical sedimentary rocks are rare. The geology of the Lesser and Greater Kuriles is notably different. The Lesser Kuriles are built of Late Cretaceous–Paleogene mafic volcaniclastic rocks intercalated with basalt and basaltic andesite lava flows. The lower part of the sequence (K/Ar ages 105–62.5 million years ago) was deposited in submarine conditions, while the upper (K/Ar ages 61–59 million

years ago) formed subaerial shield volcanoes. No Neogene or younger rocks occur in the Lesser Kuriles. The Greater Kuriles are built of a much wider spectrum of volcanic rocks of Late Miocene age and younger ( 15 m (on coasts 60–80 km distant) > 10 m

1792

Unzen, Japan

1883

Augustine, Alaska

1888

Ritter Island, ~4 to 5 km3 a Papua New Guinea

> 15 m (on coasts up to 50 km distant)

1928

Paluweh, Indonesia

3 waves, from 5 to 10 m

1933 1966 1979 2002

Volume poorly constrained due to subsequent eruption Harimkotan, Kuriles ~1 km3 (collapse scar volume) Tinakula, < 0.01 km3 b Solomon Islands Ili Werung, 0.05 km3 Indonesia Stromboli, Italy 0.02 km3

a

> 19 m?

20 m

Ocean entry by landslide from subaerial volcano lateral collapse 3–4 m on Korean coast, Partly submarine volcano lateral collapse 1200 km distant none: collapse into Ocean entry by landslide from subaerial enclosed bay volcano lateral collapse 6–8 m, ~100 km distant Ocean entry by landslide from subaerial volcano lateral collapse 8 m at Hatzfeldhafen, Partly submarine volcano lateral collapse 370 km distant; 4.5 m at Rabaul, 540 km distant Ocean entry by subaerial landslide at start of large explosive eruption Significant damage on adjacent islands

Ocean entry by landslide from subaerial volcano lateral collapse Ocean entry by landslide from small failure near summit Ocean entry by subaerial landslide

2 m (140 km distant)

Two thin-slope parallel landslides, one subaerial and one submarine

Small local waves only 9m 10 m

Landslide type

Volume estimate from collapse scar, not deposit volume. Tinakula is sometimes cited as an example of a volcano collapse without a significant tsunami. However, the main collapse scar was described some years before 1966, and examination of aerial photographs taken soon after the 1966 event show most of the collapse scar to be vegetated, with bare rock confined to a narrow chute from a small rockfall scar near the apex of the collapse scar: the latter is a prehistoric feature. b

536

LANDSLIDES

Siebert, L. 1996. Hazards of large volcanic debris avalanches and associated eruptive phenomena, in Monitoring and mitigation of volcano hazards. R. Scarpa and R. I. Tilling, eds. Berlin: Springer-Verlag, 541–572.

LAND SNAILS BRENDEN S. HOLLAND University of Hawaii, Manoa FIGURE 2 Cumbre Nueva giant lateral collapse scar, La Palma, Canary

Islands. This collapse scar is at least 20 km wide. Although it has been partly filled by the younger Bejenado volcano (center of view) and the active Cumbre Vieja volcano (foreground), the collapse scar headwall cliff (right) is still about 1 km high.

giant ocean island volcano collapses are rare, with a worldwide frequency of known events of around 1 every 20,000 years. Lateral collapses at island arc volcanoes are smaller, with volumes usually in the range of 1–10 km3, but they occur much more frequently. The most recent island arc volcano lateral collapses occurred at Ritter Island in 1888 and at Oshima-Oshima in 1741 (Table 1). LANDSLIDE HAZARDS AND ECOSYSTEM EFFECTS DUE TO LANDSLIDES AT ISLANDS

In addition to destruction of areas covered by landslides themselves, large landslides at islands, especially volcano lateral collapse landslides, produce destructive tsunamis when they enter the ocean. Well-documented tsunamis produced by events such as the 1741 Oshima-Oshima and 1888 Ritter Island collapses provide key evidence to support computer models that predict catastrophic ocean-wide tsunamis as a result of giant ocean island volcano collapses. More locally, volcano collapses such as that of Ritter Island in 1888 can eliminate the precollapse island ecosystems, leading to instances of island recolonization. SEE ALSO THE FOLLOWING ARTICLES

Canary Islands, Geology / Island Arcs / New Guinea, Geology / Taiwan, Geology / Tsunamis FURTHER READING

Blong, R. J., and G. O. Eyles. 1989. Landslides: extent and economic significance in Australia, New Zealand and Papua New Guinea, in Landslides: extent and economic significance. E. E. Brabb and B. L. Harrod, eds. Rotterdam: Balkema, 343–355. Moore, J. G., W. R. Normark, and R. T. Holcomb. 1994. Giant Hawaiian landslides. Annual Reviews of Earth Planetary Science 22: 119–144.

Land snails are surprisingly adept at dispersing across vast stretches of open ocean, a fact supported by their presence on virtually all tropical and subtropical islands globally. Island snail radiations make fascinating subjects for the study of biogeography and diversification, as many archipelagoes have well-developed and diverse endemic snail faunas. WHAT MAKES A SNAIL A SNAIL

Land snails are familiar molluscs with several characteristics that make them easily identifiable. They usually have paired eyes located at the tips of tentacles, a second pair of sensory tentacles, and a single, coiled shell into which the animals can generally withdraw their soft bodies for protection from predators and desiccation. Beneath the head is a mouth equipped with a radula, a highly specialized, elongated, rasping tongue-like organ used to scrape plant or fungal material, or in some cases to bore holes in the shells of other molluscs. Hard, calcified snail shells are secreted by a specialized layer of tissue called the mantle. Most snails have a flattened, muscular, tapering foot on which the animals glide. Land snails typically dwell on the ground or in or under bushes, shrubs, or trees; feed on decaying organic matter; and deposit eggs on or in damp soil. Taxonomically, snails are members of the Gastropoda, a globally distributed class that contains more species than all of the other classes in the phylum combined. The often drab coloration of familiar snails from temperate continental Asia, Europe, and North America, with the notable exception of a few taxa, such as the Cepaea species of Europe, contrasts dramatically with the often beautiful, brightly colored, and intricately patterned and banded shells of many island snails, such as tropical treedwelling groups like the Achatinellinae of the Hawaiian Islands, Amphidromus spp. of Indonesia, Liguus spp. and Polymita spp. of Cuba, Papuina spp. of Melanesia, and

LAND SNAILS

537

FIGURE 1 Images of a number of conspicuous endemic island land snail shells from several Pacific diversity hotspot regions including Melanesia

(New Guinea with over 1000 species) and Polynesia (Hawaiian Islands with over 750 species). A number of species represented are extinct, some are endangered, and the remainder are surviving but threatened. These images are shown at approximately accurate relative sizes and provide just a hint of the tremendous diversity and dazzling beauty of endemic island land snail faunas. Species identifications and geographic origins are as follows: (A) Placostylus albersi, New Caledonia; (B) Papuina micans, Solomons; (C) Placostylus hargravesi, Solomons; (D) Achatinella decipiens, Oahu; (E) Placostylus strangei, Solomons; (F) Achatinella fulgens, Oahu; (G) Laminella sanguinea, Oahu; (H) Coxia sp., New Guinea; (I) Parahytida dictyodes, New Caledonia; (J) Papuina pulcherrima, New Guinea; (K) Trocomorpha sp., Fiji; (L) Carelia sp., Kauai; (M) Carelia sp., Kauai; (N) Amastra spirizona, Oahu; (O) Succinea kuhnsi, Hawaii; (P) Papuina mendana, New Guinea; (Q) Partulina proxima, Molokai; (R) Papustyla hindei, New Guinea. All photographs by B. S. Holland. See also Fig. 2.

Partulidae of French Polynesia and the islands of the South Pacific (see Figs. 1 and 2 for selected examples). DIVERSITY OF ISLAND SNAILS

Island snail diversity and biogeography have fascinated biologists since the time of Darwin, who wrote in a letter to A. R. Wallace in 1857, “One of the subjects on which I have been experimentising and which cost me much trouble, is the means of distribution of all organic beings found on oceanic islands and any facts on this subject would be most gratefully received: Land-Molluscs are a great perplexity to me.” In spite of weak active dispersal, a relatively sedentary lifestyle, and minimal seawater tolerance, a number of land snail families have

538

LAND SNAILS

distributions that span ocean basins. In fact, land snails constitute some of the major terrestrial species radiations on oceanic islands, offering excellent opportunities for the study of historical biogeography, microevolution, and the diversification of insular lineages. Examples of island snail species diversity estimates include the Canaries (350), the Caribbean (1200), the Galápagos (100), the Hawaiian Islands (750), the Marquesas (300), Micronesia (400), New Guinea (1000), New Caledonia (400), the Ogasawaras (100), Pitcairn (30), Rapa (100), Rota (40), the Samoan Islands (100), and the Society Islands (200). The broad distribution of snails on oceanic Pacific islands contributed to the development of early biogeographic theories including the “mid-Pacific continent”

FIGURE 2 More conspicuous endemic island land snail shells from Pacific hotspot regions. (A) Placostylus koroensis, Fiji; (B) Amastra cylindrica,

Oahu; (C) Achatinella byronii, Oahu; (D) Samoana stevensoniana, Samoa; (E) Samoana sp., Marquesas; (F) Partula affinis, Tahiti; (G) Crystallopsis tricolor, Solomons; (H) Placostylus fulguratus, Fiji; (I) Samoana ganymedes, Marquesas; (J) Achatinella livida, Oahu; (K) Achatinella pulcherrima, Oahu; (L) Partulina sp., Hawaiian Islands; (M) Corilla sp., Samoa; (N) Trochonanina rectangular, Marquesas; (O) Achatinella fulgens, Oahu; (P) Opiella pfeifferi, Fiji. All photographs by B. S. Holland.

hypothesis, which proposed that a massive continent stretching across the South Pacific had once existed and had subsequently sunk beneath the ocean. Origins of most island snail faunas are poorly understood. Multiple dispersal events from multiple geographic source regions, in which the nearer the potential geographic source, the higher the probability of colonization, is thought to be the predominant mode of origin for oceanic island snail faunas. Rafting attached to drifting tree trunks and other floating items, aerial dispersal including by tropical storms and hurricanes, and transport attached to birds have all been suggested as mechanisms of dispersal to and among oceanic islands. For long-distance dispersal, most malacologists agree that rafting is not likely to have been an important passive vector because of a general inability of land snails to tolerate saltwater. Small stones and pebbles of the mass of certain snails have been sampled in aerial plankton studies. Transport of land

snails attached to the feet and feathers of migratory birds has been documented. To complicate the understanding of dispersal pathways and history further, founding propagules may not have originated on islands that are currently emergent but may have come from those that, because of erosion and or subsidence, no longer exist. The dynamic geological and climatic processes that build up, tear down, and profoundly transform volcanic islands have no doubt influenced the diversity of land snail faunas that have persisted through long periods of island evolution. But vicariant processes such as island separation and coalescence, sea-level fluctuation, and lava flows do not hold the potential to generate novel lineages. Such processes fragment populations and have therefore played a role in enhancing allopatric speciation on a local scale, resulting in complexes of closely related sister taxa. Thus, although there is evidence that vicariance has impacted island snail species diversity, the results

LAND SNAILS

539

of island vicariance are phylogenetically relatively shallow (affecting only more recent tip clades), compared with the effects of long-distance dispersal and multiple colonizations by divergent, independent lineages, from various geographic sources. Deeper phylogenetic divisions, and therefore more ancient levels of diversity, are driven and shaped by passive long-distance dispersal. Therefore, both dispersal and vicariance have played important roles in shaping island snail faunas, but to differing degrees and at different phylogenetic levels. Although no single compilation of the overall numbers of non-marine island snail species exists, species lists are available for various geographic regions, including most island groups, some recent, others more than 100 years old. Such lists are notorious for taxonomic inaccuracy, including the fact that many species remain undescribed, whereas others have been described multiple times as different species. Nevertheless, using these compilations, it is possible to arrive at rough estimates of diversity. One such recent estimate suggests that there are perhaps 24,000 described species of terrestrial snails, FIGURE 4 More live oceanic island snails in their natural habitats. (A)

Unidentified helicarionids, Ua Pou, Marquesas; (B) Achatinella sowerbyana, Oahu, Hawaiian Islands; (C) Unidentified helicarionid, Nuku Hiva, Marquesas; (D) Partulina tappaniana, Maui, Hawaiian Islands; (E) Partulina redfieldi, Molokai, Hawaiian Islands; (F) Perdicella helena, Molokai, Hawaiian Islands. All photographs by B. S. Holland.

FIGURE 3 Live oceanic island snails in their natural habitats. (A) Lam-

procystis sp., Bora Bora, Society Islands; (B) Partulina crocea, Maui, Hawaiian Islands; (C) Samoana sp., Tahiti, Society Islands; (D) Catinella sp., Oahu, Hawaiian Islands; (E) Succinea lumbalis, Kauai, Hawaiian Islands; (F) Succinea sp., Tahiti, Society Islands. All photographs by B. S. Holland. See also Figs. 4 and 5.

540

LAND SNAILS

although estimates as high as 80,000 species have been published. Likewise, a rough estimate of the total number of snails that have evolved on islands (excluding New Zealand, Papua New Guinea, and Madagascar) is about 6000, or 25% of the lowest total global estimate of snail species. In light of the minute fraction of global land area constituted by islands, this estimate shows the disproportionately important role played by islands in the generation of land snail biodiversity. For example, a comparison of the number of species in all of North America with the species diversity of the main Hawaiian Islands reveals that, in spite of the fact that the land area of North America exceeds that of the Hawaiian Islands by about a thousandfold, snail species diversity is roughly equivalent between the two regions. The presence of diverse endemic assemblages of land snails on islands, including oceanic island groups that are thousands of kilometers from the nearest neighboring island or continent, is a testament not only to the extraordinary ability of these molluscs to passively disperse over long distances and establish new colonies, but also to their tendency to radiate into large numbers of species from relatively rare initial propagules, often in

FIGURE 5 More live oceanic island snails in their natural habitats. (A) Unidentified helicarionid, Hiva Oa, Marquesas; (B) Samoana sp., Tahiti, Soci-

ety Islands; (C) Achatinella sowerbyana, Oahu, Hawaiian Islands. All photographs by B. S. Holland.

adaptive species complexes (see Fig. 3). Because of these characteristics, island land snail radiations are increasingly valued as informative systems in illuminating and advancing the general understanding of the processes and patterns of evolution and adaptive radiation on islands. Factors considered most important in the generation of high species diversity include island area, latitude, age, elevation, rainfall, and plant community diversity. These factors are frequently interrelated, and probably influence species diversity on islands in concert, and in complex ways. CONSERVATION CONCERNS

Non-marine molluscs comprise one of the most threatened groups of animals on Earth and include 99% of all molluscan extinctions. Habitat loss, degradation, and fragmentation and environmental stresses such as climate change, pollution, and the introduction of invasive species such as black rats (Rattus rattus), the wolf snail (Euglandina rosea), and the flatworm (Platydemus manokwari) all play important and often cascading synergistic roles in island land snail extinction. Although terrestrial vertebrate extinctions are well documented, invertebrate extinctions usually go unnoticed by the general public and undocumented by biologists and conservation agencies. Only a tiny fraction (less than 2%) of known molluscan species have had their conservation status scientifically assessed. Among the high-diversity island faunas, the Hawaiian land snails are relatively well studied, and species losses are daunting, estimated at 65– 90%. Although most archipelagoes have not received the level of scientific scrutiny as have the Hawaiian Islands, similarly dire conditions probably exist on islands across all of the ocean basins. Ecologists have begun to recognize important inputs to ecosystem function for a variety of invertebrate taxa,

whether they be in the form of aeration of soils by earthworms, pollination of flowers by insects, decomposition of dead timber by termites, or other processes. Clearly, removal of such species has the potential to induce negative impacts on ecosystem balance, energy flow, health, and function. Although the precise roles of abundant, conspicuous terrestrial island snails in ecosystem function are not well understood, snails may play important roles in the maintenance of healthy forest ecological equilibrium (they may be involved in the calcium cycle, the carbon budget, and the breakdown of leaf litter and other organic material, and they may have a role as predator/ prey in food webs and nutrient cycles). Island snail assemblages are highly sensitive to habitat alteration and to the presence of invasive predatory species. On the majority of oceanic high islands, especially at elevations below about 300 m, native flora has been replaced by introduced weeds, ornamentals, and agricultural species. Most island land snails require native plants. We know from the notes of early twentieth-century expedition scientists that island floras were altered long ago; for example, Adamson in 1932 wrote of the Marquesas that “The native flora below 1000 ft has been replaced in large measure by immigrants, and to a considerable extent up to 2,500 feet.” Thus, in general, the native snail fauna is presently restricted to upper elevations of high islands, because these areas harbor the only remaining native forest in many island environments. The degree of island land snail imperilment is poorly documented and almost certainly underestimated. This view is supported by the continual discovery of undescribed snails, especially on tropical island archipelagoes throughout the world, many of which have been largely deforested and on which numerous harmful invasive species have become established. For example, a multi-

LAND SNAILS

541

year field survey of the terrestrial snails of the Papuan Peninsula and nearby islands by Florida Museum of Natural History biologists has recently uncovered dozens of undescribed species, suggesting that although the area has long been considered a biodiversity hotspot, the land snail fauna is far more diverse than was previously known. Surveys conducted in the southeastern United States focusing on small snails restricted to specialized habitats, including seeps and springs, are also uncovering previously undescribed species. Such surveys are also important in uncovering previously undocumented introduced species. Early detection of introduced species is crucial in the facilitation of efforts to control and prevent establishment, because after establishment, eradication is impossible. CONCLUSION

This brief overview is intended to demonstrate that in terms of biodiversity hotspots, tropical and subtropical island land snails are important yet frequently overlooked components in terms of their contributions to unique island biodiversity. Ongoing efforts to preserve island land snails have had some success in recent years and have included captive rearing programs for rare Hawaiian achatinelline tree snails (University of Hawaii, Manoa) and French Polynesian partulid tree snails (mainly the London Zoological Society and the Jersey Zoo, but also the Detroit Zoological Park, the John G. Shedd Aquarium, and the San Diego Zoo); efforts to fence out, trap, or poison predators; habitat restoration; population translocation; and conservation genetics studies. However, the need persists for intensification of such efforts and for creative novel conservation strategies and solutions in the face of accelerating environmental change.

LAVA AND ASH KATHARINE V. CASHMAN University of Oregon, Eugene

Lava and ash are two different products of volcanic eruptions that cover the surfaces of volcanic islands. These two substrates have different volcanic origins and different physical properties, particularly with respect to their interaction with water (water transmission, storage, and susceptibility to erosion). The prevalence of one component or the other depends on the types of volcanic eruptions responsible for island formation. Eruption style, in turn, is determined primarily by the island’s location with respect to tectonic plates. LAVA FLOWS

Basaltic lava flows cover much of the surface area of volcanic islands that form over hotspots, such as Hawai‘i. In these settings, the surfaces of active volcanoes are almost entirely composed of vast lava flow fields. J. D. Dana, an early geologist to visit Hawai‘i on Charles Wilkes’s U.S. Exploring Expedition, commented that “Areas, hundreds of square miles in extent, are covered with the refrigerated lava floods, over which the twistings and contortions of the sluggish stream as it flowed onward are everywhere apparent; other parts are desolate areas of ragged scoria.” Dana also noted that degradation of lava-mantled surfaces was dependent

SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Dispersal / Hawaiian Islands, Biology / Invasion Biology / Oceanic Islands / Vicariance FURTHER READING

FIGURE 1 Two NASA images of the island of Hawai‘i, USA, available

on the Earth from Space website (http://earth.jsc.nasa.gov/sseop/

Barker, G. M. 2001. The biology of terrestrial molluscs. CAB International. Cowie, R. H. 1996. Pacific Island land snails: relationships, origins and determinants of diversity, in The origin and evolution of Pacific Island biotas, New Guinea to Eastern Polynesia: patterns and processes. A. Keast and S. E. Miller, eds. Amsterdam: SPB Academic Publishing, 347–372. Cowie, R. H., and B. S. Holland. 2006. Dispersal is fundamental to biogeography and the evolution of biodiversity on oceanic islands. Journal of Biogeography 33: 193–198. Lydeard, C., S. A. Clark, K. E. Perez, R. H. Cowie, W. F. Ponder, A. E. Bogan, P. Bouchet, O. Gargominy, K. S. Cummings, T. J. Frest, D. G. Herbert, R. Hershler, B. Roth, M. Seddon, E. E. Strong, and F. G. Thompson. 2004. The global decline of nonmarine mollusks. BioScience 54: 321–330.

542

L AV A A N D A S H

efs/). (A) Image STS61A-050-0057 (November 1985). A low-oblique, north-looking photograph that shows the five volcanoes that form the island. From youngest to oldest these are Kilauea (KL), Mauna Loa (ML), Hualalai (HL), Mauna Kea (MK), and Kohala (KH). Also evident in this image is the prominent rain shadow on the western and southern flanks of Mauna Kea and Kohala (seen as brown desert in contrast to the lush greenery of the northeastern slopes of the volcanoes). (B) Image STS051-102-083 (September 1993). Kohala volcano, the oldest on the island, shows the effect of topography-induced variations in rainfall on patterns of vegetation (green vs. brown in the image) and rainfall-induced erosion (seen as deep erosional gullies that have formed on the windward side of the volcano).

on both time and rainfall; in Hawai‘i this is evidenced by the pronounced difference in vegetative cover and erosional dissection of slopes on the windward (high rainfall) and leeward (near-desert) sides of the island of Hawai‘i (Fig. 1). Basaltic lava flows are traditionally classified into two types on the basis of their surface morphology: pa¯hoehoe and ‘a‘a¯. Pa¯hoehoe flows are characterized by smooth glassy surfaces that are often folded into ropy textures as a consequence of surface compression while they are still flowing (Fig. 2A). Pa¯hoehoe flow fields are fed through lava tubes, which develop within large lava sheets by protracted flow of lava at low to moderate rates. Crevices between surface folds may trap moisture and allow early colonization by plants with windblown spores, such as ferns (Fig. 2B). The surface crust of pa¯hoehoe flows is easily removed, and resulting shallow holes were used by native Hawaiians to plant upland taro and sweet potato. The thin glassy surface of pa¯hoehoe flows was also easily chipped away to create petroglyphs (Fig. 2C). ‘A‘a¯ flow surfaces are covered in rough clinkers that are typically 20–50 cm in diameter and form by shearing and tearing of partially crystalline lava along the flow margins and upper surface (Fig. 2D). ‘A‘a¯ flows form when eruption rates are high or in distal parts of flow fields (when the lava has had time to cool and crystallize). Because the rubbly character of ‘a‘a¯ flow surfaces makes them difficult to traverse, native Hawaiians constructed paths across these flows by breaking large clinkers into small pieces. The clinkery flow surfaces also make the boundaries between flows highly porous and permeable to water, despite less porous flow interiors. For this reason, active volcanic regions that are dominated by ‘a‘a¯ (or blocky) flows rarely have rivers or other manifestations of surface water, but instead are characterized by large aquifers and low-elevation springs. ASH

Volcanic ash, in a strict sense, is a term used for solid fragments of volcanic material that are less than 2 mm in size. These fragments can include both frothy glass (quenched melt) and fragments of crystals (Fig. 3A). Ash is formed by explosive eruptions, where rapidly expanding gases within the ascending and cooling magma literally blow the magma apart, or by disintegration of growing lava domes during collapse. Ash-forming eruptions are most common along volcanic arcs that border subduction zones. Magma generated in this tectonic setting is rich in dissolved water, which powers highly explosive eruptions when magma ascends toward the Earth’s surface. Like lava, ash is relatively impermeable

FIGURE 2 Photographs of Hawaiian basaltic lava flows. (A) An active

pa ¯hoehoe flow from Kı¯lauea Volcano, Hawai‘i, slowly advances; surface wrinkles form by compression of the newly formed surface crust during flow. (B) Ferns growing on a young (1992) pa ¯hoehoe flow, Kı¯lauea Volcano. (C) Petroglyph carved in an older pa ¯hoehoe flow, Kı¯lauea Volcano. (D) Active ‘a‘a ¯ flow from Kı¯lauea Volcano; flow surface is covered with clinkers that form as the flow surface is pulled apart during flow advance. These clinkers form a rubbly, and permeable, upper and lower surface on solidified flows.

to surface water, but unlike lava it is also highly mobile, easily eroded, and prone to airborne redistribution. Explosive volcanic eruptions occur when water (or other magmatic gases such as carbon dioxide and sulfur) that is originally dissolved in the melt comes out of solution (vesiculates) as magma rises toward the surface (depressurizes), in a process similar to the release of carbon dioxide in a bottle of soda when pressure is released as the cap is removed. As bubbles form and expand within the melt, they both drive the bubble–melt mixture upward and blow the magma apart, ultimately creating violent eruptions that produce large volcanic plumes (Fig. 3B). Such plumes may carry fine ash fragments into the stratosphere, where they can then be transported, literally, around the globe. Most ash particles, however, fall out of the plume within tens to hundreds of kilometers from the volcano, along the trajectory of the wind. More common are less violent eruptions that are associated with the growth of lava domes. Such eruptions may last for years, with ash posing persistent hazards to aircraft and covering fields and houses with thick deposits that are easily mobilized by heavy rainfall. A recent example of this type of activity can be found on the island of Montserrat, part of the Lesser Antilles island arc in the Caribbean. Here the Soufriere Hills volcano became active in 1995; by the end of 1997 the capital city of Plymouth had

L AV A A N D A S H

543

FURTHER READING

Cashman, K. V. 2004. Terrestrial volcanism, in Volcanic worlds. R. Lopes and T. Gregg, eds. Worthing, UK: Praxis Publishing Ltd., 5–42. Druitt, T. H., and B. P. Kokelaar, eds. 2002. The eruption of Soufrière Hills volcano, Montserrat, from 1995 to 1999. Geological Society of London Memoir 21. Macdonald, G. A. 1953. Pahoehoe, ‘a‘a¯, and block lava. American Journal of Science 251: 169–191. Sigurdsson, H., B. F. Houghton, S. McNutt, H. Rhymer, and J. Stix, eds. 2000. Encyclopedia of volcanology. San Diego, CA: Academic Press. Soule, S. A., and K. V. Cashman. 2005. The shear rate dependence of the pahoehoe-to-‘a‘a¯ transition: analog experiments. Geology 33: 361–364.

FIGURE 3 (A) Scanning electron microscope image of ash particles

from Mount St. Helens volcano, Washington. Seen here are ragged

LAVA TUBES

bubbly fragments of quenched melt and smooth-surfaced crystal fragments. Image is about 300 microns in width. (B) Eruption column from

JIM KAUAHIKAUA

Mount Cleveland, Alaska, as photographed by J. N. Williams (NASA)

U.S. Geological Survey, Hawaii National Park

from the International Space Station on May 23, 2006. Image available at http://antwrp.gsfc.nasa.gov/apod/image/0606/volcanoplume_iss_ big.jpg. (C) Aerial view of ash-damaged buildings on Montserrat; both roof collapse and partial burial are the result of ash fall. (D) Summit

FRANK HOWARTH Hawaii Biological Survey, Honolulu

region of Stromboli volcano, Italy. Summit was covered with large frothy lava “bombs” as well as ash during the eruption of April 5, 2003.

KEN HON University of Hawaii, Hilo

been largely covered by ash and pyroclastic debris. Activity persists today, and nearly 60% of this small island is currently uninhabitable (Fig. 3C). Ash is also a component of basaltic Strombolian eruptions, named for the island of Stromboli (Italy), which has been erupting almost continuously for at least the past 1000 years, thus earning it the name “Lighthouse of the Mediterranean.” Here several small eruptions occur every hour; less frequent, larger eruptions, such as one that occurred on April 5, 2003, cover the volcano’s summit with frothy volcanic bombs as well as basaltic ash (Fig. 3D). More energetic basaltic eruptions produce more ash. Although not located on an island, an example of such an eruption is that of Parícutin volcano in Mexico, which was born in a cornfield on February 20, 1943. For the next nine years, strong explosions built up over 12 m of ash and scoria deposits in areas near the volcano, forcing local residents to move to more distant locations. Now, more than 50 years after the end of the eruption, much of the ash remains, but much has also been eroded and redistributed by annual monsoon rains. SEE ALSO THE FOLLOWING ARTICLES

Eruptions: Laki and Tambora / Hawaiian Islands, Geology / Island Formation / Volcanic Islands

544

L AV A T U B E S

Lava tubes, originally called “pyroducts,” form within lava flows as thermally insulated conduits through which lava is carried away from a volcanic vent and supplied to an active lava flow. After draining and cooling, these same lava tubes, along with the right mixture of water, in the form of trapped humid air, and food, in the form of organic debris and root systems, can provide a very specialized ecological niche that requires organism adaptation. The opportunity to colonize these forbidding environments is responsible for some of the fastest evolutionary adaptation currently known. VOLCANIC ISLANDS AND LAVA TUBES

Volcanic islands, known to have lava tube caves, grow by repeated eruptions of basaltic lava. These islands form in two distinctly different settings—over hotspots or in island arcs. Most basaltic ocean islands are related to high rates of magma production either in the middle of tectonic plates or near spreading ridges. These are the “hotspot” volcanoes and include Hawai‘i, Iceland, the Azores, Réunion, and many other island chains scattered throughout the world’s oceans. The processes that lead to the formation of these islands are not well understood but seem to be related to varying contributions of deep mantle plumes. As these islands grow older or move away

from their source, the erupted lavas often become richer in silica and alkalis (sodium and potassium). Basaltic islands may also form along “primitive” sections of island arcs and are related to subduction of one ocean plate beneath another. Eruptions of tube-forming basalt are most common as arc island volcanoes emerge from the sea in places like the Izu–Bonin, Tonga–Kermadec, Banda Api, Aeolian, and South Sandwich subduction zones. As the arc ages and the islands grow, the magmas become much higher in SiO2, and eruptions take on the explosive nature that characterizes most “Ring of Fire” subduction zone volcanoes. This change in lava chemistry diminishes the possibility that tubes will form. Most basaltic volcanic islands begin as submarine seamounts that grow by eruptions of pillow lavas. As they near the surface and emerge from the sea, interaction with seawater produces pyroclastic eruptions, like those of Surtsey in Iceland, that mantle the seamount in glassy sand and ash. Ensuing subaerial eruptions of lava flows build shield and cinder cone complexes that give the islands their distinctive appearance. FORMATION OF LAVA FLOWS

Effusive eruptions are the result of gentle gas expansion that drives magma from the ground in spectacular fissure eruptions and lava fountains. The erupted lava is very hot (typically 1100–1200 °C) and fluid, allowing the gases to escape easily and non-explosively. The resulting lava flows pour down the slopes of the volcanoes and form either pa¯hoehoe or ‘a‘a¯, depending on a combination of composition, temperature, water content, ground slope, and effusion rate. Effusion rates in excess of about 5 m3/s favor formation of ‘a‘a¯ flows, whereas lower effusion rates favor pa¯hoehoe flows. In regions of low ground slope, pa¯hoehoe lava flows are favored, even during high effusion rate eruptions or at great distances from the vent. Effusion rates are generally highest at the beginning of an eruption and decrease as the eruption continues. Therefore, long-lived eruptions and flow fields typically become dominated by pa¯hoehoe flows even if they were initially ‘a‘a¯.

molten surface solidifies rapidly upon contact with air, forming an insulating crust that may inflate in response to continued lava supply. As the flow advances downslope, the lava eventually develops preferred pathways that become lava tubes. Open lava tube caves are less common in inflated flows because they occur most often on flat slopes that do not drain at the end of an eruption. As both ‘a‘a¯ and pa¯hoehoe flows mature, lava tubes develop. For broader flows, the tube network can develop a complicated, braided, or anastomosing network. With time, the network becomes consolidated, sometimes into a single tube. The development of tubes favors efficient transport of lava because solid basalt is a very good thermal insulator. Lava flowing through tubes on Kı¯lauea rarely loses more than a single degree C per kilometer of travel. Lava tubes allow lava flows to extend great distances away from their vents. The Undara flows of Australia exceed 100 km in length, but on ocean islands, the length of the flows and lava tube systems are limited by the size of the island. Examples of long lava tube caves are the 42-km-long Kazumura lava tube cave on Hawai‘i Island and a single sealed lava tube 13 km in length on Jeju Island, South Korea. The concentration of all shallow tubes into a single tube begins the process of downcutting through a combination of melting and abrasion. Within a month, tubes can become deeper, creating a keyhole shape in cross section. A cross section through an entire mature flow will reveal the original anastomosing shallow caves and a “master tube” that is greatly enlarged and downcut by flowing lava. Any of these conduits can form caves when drained. Localized collapse of the roof of active tubes creates skylights (Fig. 1) that frequently form over or near lava falls. Tiered lava tube systems may form around skylights

FORMATION OF LAVA TUBES

Lava tubes may form in both p¯a hoehoe and ‘a‘a¯ flows. ‘A‘a¯ channels may begin to crust over, allowing the transport of hotter, more fluid lava that favors pa¯ hoehoe-morphology flows. Within a short period of time, the ‘a‘a¯ flow is covered with pa¯hoehoe channel overflows, and the channel becomes roofed over, forming a lava tube. Pa¯hoehoe channels may also crust over to form lava tubes. Pa¯hoehoe flows characteristically “self-seal” as the

FIGURE 1 View through a skylight of lava flowing through one of the

many lava tubes formed during the current eruption of Kı¯lauea volcano. The lava stream is about 1150 °C, about 2 m wide, and flowing away from the viewer at approximately 3 m/s. Note the incipient shelf formation over the stream surface.

L AV A T U B E S

545

by several processes. Cooling through a skylight can cause the lava channel to crust over, forming an internally roofed, two-tiered tube section. Overflows from skylights and ruptures in the tube may be caused by blockages of the tube due to collapse or by increased lava supply in the tube. The resulting overflows from the tube can form a new set of lava tubes and create a stacked lava tube system. LAVA TUBE STRUCTURES

Internal ornamental structures that form while lava tubes are active include shark’s tooth lava stalactites, hollow straw lava stalactites, anhydrite encrustations, driblet spire lava stalagmites, extruded lava buds, secondary lava flows, elephant skin textures, wall slumps, channel gutters, fluted walls, and floor volcanoes. Shark’s tooth lava stalactites are drip structures that form during the rise and fall of lava against the tube roof. Growth rings inside these stalactites record individual coats of lava added during these events. Channel gutters and ledges on the walls of lava tubes record decreasing lava levels within the tubes due to either downcutting or declining lava supply. Many shelves are incomplete roofs downslope of skylights. Slumped and crumpled walls (elephant skin textures) result when partially molten tube walls begin to deform. Water-rich partial melts within the tube walls, roof, and

floor may be expulsed to form hollow straw stalactites and cogenetic driblet spires (Fig. 2), secondary lava flows, and floor volcanoes. All of these features have distinctive chemical compositions that show that they were produced by remelting previously solidified tube roof, wall, and floor material. Prolonged reaction of the remelted wall rock with the high-oxygen atmosphere within lava tubes creates a unique suite of high-temperature minerals. Most of the interior surfaces of lava tubes are coated with magnesioferrite, which gives many cave interiors a metallic gray appearance. Parts of the tube interior may also have an appearance of brownish ceramic-like glaze, where surfaces have not been exposed for sufficient time to develop crystalline magnesioferrite and related minerals. Hightemperature anhydrite sublimates have also been found as drusy crystals coating the walls and roof of active lava tubes. Many secondary minerals form during cooling of lava tubes after lava ceases to flow within the system. This process can take many months, and during this time the original high-temperature oxides and sulfates react and form lower-temperature iron oxides and hydroxides as well as a host of low-temperature sulfate minerals. Sections of lava tubes that cool near open skylights commonly develop bright red hematite glazes. Deep within the cave systems, the original magnesioferrite coatings are transformed into dull metallic to brownish iron oxides. Many of the sulfate minerals and speleothems form rapidly when the temperature of the cave drops below 100 °C, allowing water to infiltrate the tube. In wet climates, these deposits are washed away within months, whereas in arid climates they may last for years. Several lava tube caves on Jeju Island, South Korea, have well-developed calcite speleothems and flowstones due to leaching of water through calcareous sand dunes that overlie the tube systems. LAVA TUBE CAVES

FIGURE 2 Soda straw lava stalactites and stalagmites in a lava tube

cave that last carried lava about two years before this photograph was taken.

546

L AV A T U B E S

After a lava tube system has drained and cooled enough to allow human entry, each lava tube cave is defined by its traversibility. Therefore, a single lava tube system may result in several different lava tube caves, with each cave being defined as any length of tube passable by humans that includes at least one entrance. Although the lava tube cave systems may not have apparent physical connections, the extensive crack systems within lava flows allow small organisms ample mobility. Besides lava tube caves (Fig. 3), basaltic lava flows produce abundant additional voids of varying sizes that may not be passable by humans. These are made by degassing,

FIGURE 3 Schematic of a lava tube cave.

expanding flow lobes, gaps between new flows and the older substrate, buried ‘a‘a¯ rubble, and cracking of the cooling lava. These voids are interconnected by a vast system of cracks that radiate through the flow. The cracking is so pervasive that, after cooling, few lava tube caves can carry water except during times of high rainfall. The smaller, capillary-sized spaces created by gas vesicles are less important biologically because their small size limits the amount of food resources they can hold and transport. Voids larger than about 5 cm can transport large volumes of food into the underworld.

the flow. This aeolian, or wind-supported, ecosystem disappears shortly after plants arrive and the surface cracks become blocked. Within a year, cave-adapted animals begin colonizing the subterranean habitat, arriving from the voids in neighboring older flows. Predators and omnivores arrive first and feed on lost windblown animals and organic matter that washes or falls into the deeper moist voids. The arrival of pioneering plants on the surface paves the way for colonization by the root- and litter-associated fauna. The plants can begin within a decade in wet climates but may take more than a century under desert conditions. Because of the lack of soil and surface water (due to the lava flow’s high porosity and extensive cracking), colonizing plants must send their roots deep underground to obtain water and nutrients. In Hawai‘i, a 1-m-tall pioneering Metrosideros tree on the surface requires the support of a branched root system that may extend 15 m or more underground (Fig. 4). In addition to plant roots, food resources in lava tube caves include chemoautotrophic bacteria and organic material brought in by roosting or lost animals washed in by rain or blown in by wind. The young flow remains inhospitable for most soil and litter animals. The abundant underground food resources have allowed a few surface animals living on each volcanic island to adapt to live permanently underground on the same island.

COLONIZATION OF THE LAVA FLOW SURFACE

The surfaces of basaltic lava flows are marked by numerous cracks, crevices, and broken lava tubes. Many of these surface openings connect with the system of subterranean voids. Any organic material falling on the surface is likely to fall or be washed into these voids and to subsequently sink underground. Thus, the surface of these flows can appear barren for decades to centuries, depending on the time it takes for the surface voids to fill and soil to form. The barren environment is an extreme one for life: The dark lava absorbs solar radiation and heats the surface to intolerable temperatures. Rain disappears underground almost as it falls, making the surface extremely dry (xeric). The changing heat and open environment generate strong surface winds. However, on each island, a group of specialized organisms soon colonize the surface. The first to arrive are crickets, spiders, and other invertebrates, which colonize the flow within a month of the surface solidifying. Many of these are specifically adapted to live only on these barren lava flows. They hide deep in moist cracks during the day and emerge at night to feed on windborne organic matter concentrated in cracks on

FIGURE 4 Metrosideros polymorpha (‘o ¯hi‘a) roots in a lava tube cave

within a lava flow erupted from Mauna Loa, Hawai‘i, in 1881.

L AV A T U B E S

547

LAVA TUBE CAVE HABITATS

The cave habitat is dominated by water vapor and carbon dioxide. Water vapor is lighter than air and tends to rise out of caves via openings. However, long passages with convoluted shapes and smaller voids tend to trap water vapor, and the air remains saturated. Saturated air is above the equilibrium humidity of blood, and cold-blooded animals literally drown without special adaptations to cope with the excess water. Carbon dioxide, which is produced by respiration, is heavier than air and tends to concentrate in deeper and more isolated passages, especially within the intermediate-sized spaces in the lava. Uniquely high CO2 concentrations may be found in lava tube caves located in active volcanic regions that are emitting the gas. High concentrations of CO2 force animals to breathe more, intensifying the respiratory effect of excess water. Cave-adapted animals prefer the stagnant, watersaturated atmosphere found in the innermost deep zones. Lava tube caves are three-dimensional, complex, mazelike passages with lethal or near-lethal gas concentrations; scattered, hard-to-find food resources; unforgiving rocky substrates; and wet, slippery, vertical surfaces. This perpetually dark and humid environment is, at least, stressful or outright inhospitable for most organisms. Few animal groups can exploit this habitat. ADAPTATION BY CAVE ORGANISMS

In Hawai‘i, both the lava flow crickets and big-eyed lava flow spiders (Fig. 5) have founded cave populations. Other colonizers have come from rainforest and marine littoral habitats. More than 75 species of cave arthropods have been discovered to date in the Hawaiian Islands, with 26 species known from Hawai‘i Island, including beetles, crickets, planthoppers, spiders, moths, flies, earwigs, and true bugs.

FIGURE 5 Adelocosa anops (no-eyed big-eyed hunting spider) in a

Kaua‘i, Hawai‘i, lava tube cave.

548

L AV A T U B E S

The ancestors of cave-adapted animals lived in damp, dark environments and could already cope with some of the stresses in the lava tube environment. The planthoppers are a good example. Nocturnal animals living in damp, wet-rock habitats, such as rocky sea coasts, stream margins, and lava flows, already possessed many of the necessary adaptations for subterranean life. The accumulating food resources provided the impetus to give up surface life and move underground permanently. Cave adaptation has occurred independently on each island from surface ancestors living on the island. At least 12 separate groups have adapted independently to lava tubes at least twice in Hawai‘i, indicating that cave adaptation is a natural process wherever the environment is suitable and animals are present. Because the cave animals move underground from older lava tubes to newer tubes, the cave species are nearly always older than the tubes in which they occur. Thus, the maximum time necessary for cave adaptation could be as old as the island itself. The hallmark of cave-adapted species is the loss of conspicuous structures such as eyes, body color, protective armor, and for some insects, wings. These structures are useless underground but expensive for the body to make and maintain. Therefore, they can be lost quickly when selection is relaxed. A clue to how such losses might happen quickly is demonstrated by the cave-adapted planthoppers. The nymphs of surface relatives also feed on plant roots and have reduced eyes and body color, whereas the adults have big eyes, cryptic colors, and strong wings. The cave-adapted adult descendents maintain the nymphal eyes, color, and other characters into adulthood, a phenomenon known as neoteny. Organisms adapting to live permanently underground also need to make changes in their physiology, form, and behavior to cope with the stressful environment, especially the high relative humidity and occasional episodes of high carbon dioxide concentrations. Because many of the normal cues used by surface animals behave abnormally or do not occur underground (e.g., wind, light-dark cycles, sound), the organisms also need to adapt to take advantage of new cues to find mates and food. Consider trying to follow a scent plume, whether from a potential mate or food resource, in a three-dimensional, dark maze; an animal would need to take an indirect path to the source. In addition, predators might take advantage of the same plume and be waiting for the unwary. Jumping or falling might land a hapless animal in a pitfall trap. Most lava tube arthropods have specialized elongated claws to hold on to the glassy wet rock surfaces. Many have elongated

legs to step across cracks rather than having to descend and climb the other side. Small surface insects are too heavy or are unable to climb the meniscus at the edge of small pools and eventually drown. However, many caveadapted insects have unique knobs near the base of each elongated claw that allow them to climb the meniscus and escape. Some of the latter species are predators or scavengers, who wait on pools for victims. SEE ALSO THE FOLLOWING ARTICLES

Caves, as Islands / Crickets / Insect Radiations / Lava and Ash / Seamounts, Geology / Spiders FURTHER READING

Calvari, S., and H. Pinkerton. 1998. Formation of lava tubes and extensive flow field during the 1991–1993 eruption of Mount Etna. Journal of Geophysical Research. 103: 27,291–27,301. Cashman, K. 2004. Volcanoes on Earth: our basis for understanding volcanism, in Volcanic worlds: exploring the solar system’s volcanoes. R. M. Lopes and T. P. Gregg, eds. Chichester, UK: Springer and Praxis Publishing, 5–42. Culver, D. C., and W. B. White, ed. 2004. The encyclopedia of caves. Burlington, MA: Elsevier Academic Press. Helz, R. T., C. Heliker, K. Hon, and M. Mangan. 2003. Thermal effi¯ ‘o¯-Kupaianaha eruption, in The ciency of lava tubes in the Pu‘u ‘O Pu‘u ‘O¯‘o¯-Kupaianaha eruption of Kı¯lauea volcano, Hawai‘i: the first 20 years, USGS Professional Paper 1676. C. Heliker, D. Swanson, and T. J. Takahashi, eds. Reston, VA: U.S. Geological Survey, 105–120. Howarth, F. G. 1983. Ecology of cave arthropods. Annual Review of Entomology 28: 365–389. Kauahikaua, J., K. Cashman, T. N. Mattox, C. C. Heliker, K. Hon, M. Mangan, and C. R. Thornber. 1998. Observations on basaltic lava streams in tubes from Kilauea volcano, island of Hawai‘i. Journal of Geophysical Research 103: 27,303–27,323. Sigurdsson, H., B. F. Haughton, S. R. McNutt, H. Rymer, and J. Stix, eds. 2000. Encyclopedia of volcanoes. San Diego, CA: Academic Press. Wilkins, H., D. C. Culver, and W. F. Humphreys, eds. 2000. Subterranean ecosystems, ecosystems of the world, 30. Amsterdam: Elsevier Press.

LEMURS AND TARSIERS ROBERT D. MARTIN The Field Museum, Chicago, Illinois

Lemurs and tarsiers—two of the five main groups of living primates—are island inhabitants. Lemurs have undergone a major diversification on Madagascar, resulting in over 100 modern species (including 16 recently extinct representatives). Diversification of tarsiers on islands of Southeast Asia has been more modest, generating 17 modern species. However, 15 of those tarsier species are found on the relatively small island complex of Sulawesi, which is a “hotspot” for evolutionary divergence.

PRIMATES ON ISLANDS

The mammalian order Primates, to which we ourselves belong, contains five major groups: (1) lemurs; (2) lorisiforms (bushbabies and lorises); (3) tarsiers; (4) New World monkeys; and (5) Old World simians (monkeys, apes, and humans). Nonhuman primates are generally restricted to tropical and subtropical regions, and most are forest inhabitants. They are found on all mainland areas of the southern continents, except Australia. New World monkeys occur in South and Central America, whereas both lorisiforms and Old World simians are widely distributed throughout Africa and Asia. However, two of the five major groups of primates are restricted to island areas in the south: lemurs on Madagascar, and tarsiers on the island archipelago of Southeast Asia. Perhaps because of their isolation on islands and the resulting lower levels of competition with other animals, both the lemurs and the tarsiers have remained relatively primitive in many features. Although tarsiers are actually quite advanced in certain respects and more closely related to higher primates (monkeys, apes, and humans), they have also retained many primitive features. At least in terms of the morphology of their cheek teeth, tarsiers have remained relatively unchanged over the past 40 million years or more. This has led some authors to describe them loosely as “living fossils.” LEMURS

Lemurs are found only on Madagascar, and they are the only primates to occur there. Molecular evidence has now confirmed previous indications from shared morphological features that lemurs are all derived from a single common ancestor. Within Madagascar, they have undergone a remarkable adaptive radiation. For example, dentitions are more diverse in lemurs than in any other living primate group, and the range of dietary adaptations is correspondingly extensive. The modern lemur fauna contains almost 100 species, ranging in size from mouse lemurs, weighing just 40 g, up to the indri, with a body mass of about 6.3 kg. In addition, there are 16 species that died out just a few thousand years ago, probably because of climatic change combined with human colonization of Madagascar. These subfossil lemurs, which are really part of the modern fauna, are generally very big and extend the body size range up to 100 kg or more. There is no true fossil record for lemurs. Relationships among lemurs must therefore be established from a combination of morphological and molecular evidence. The remarkable diversification of Malagasy lemurs has occurred within a geographical area far smaller than that L E M U R S A N D TA R S I E R S

549

occupied by any other major group of primates. Lemurs account for around one-fifth of the genera and species of modern primates, and the ranges occupied by individual species are correspondingly small. The diversity of lemurs is due at least in part to limited competition from other mammals. Only three other land mammal groups are long-established inhabitants of Madagascar: tenrecs, rodents, and carnivores. Lemurs show a number of general trends with increasing body size: from nocturnal to diurnal habits, from insectivory through frugivory to folivory, and from solitary foraging to gregarious social life, although most show strict seasonal breeding. Five distinct subgroups of lemurs are recognized: (1) mouse and dwarf lemurs (Allocebus, Cheirogaleus, Microcebus, Mirza, Phaner); (2) sportive lemurs (Lepilemur); (3) true lemurs and bamboo lemurs (Eulemur, Hapalemur, Lemur, Prolemur, Varecia); (4) the indri and its relatives (Avahi, Indri, Propithecus); and (5) the aye-aye (Daubentonia). Molecular and chromosomal evidence indicates that the aye-aye diverged first, but subsequent separations between the other four subgroups occurred in quite rapid succession. A key question regarding the adaptive radiation of Madagascar lemurs concerns the process by which so many species could have arisen within the island. As a rule, a barrier of some kind between populations is required for speciation to occur. The combination of a cooler, elevated central plateau and several major rivers running down to the sea apparently accounts for effective isolation of a ring of forest regions around the coasts of Madagascar. Marked climatic zonation in Madagascar endows each region with a particular range of environmental conditions. Regarding rainfall, trade winds deposit most of their moisture along the east coast, leaving the western part of the island relatively dry. There is a latitudinal gradient for mean temperatures in the coldest month, which are higher in the north than in the south. The vegetation of any region isolated between two major rivers reflects the local climatic conditions, with a spectrum ranging from dense tropical rain forest in the northeast down to semi-arid, shrubby vegetation in the southwest. It thus seems highly likely that the animals in any coastal or lowland region are typically isolated and adapted to local conditions, including special features of both climate and plant life. Occasional migration of closely related species between lowland regions would lead to competition between sibling species, often resulting in divergent specialization. This proposed model of lemur speciation in Madagascar has recently been subjected to a number of

550

L E M U R S A N D TA R S I E R S

tests. Molecular evidence has shown that major rivers do, indeed, represent boundaries between extant sister species in many cases. Moreover, chromosomal and/or molecular evidence has allowed identification of many new species, in particular small-bodied nocturnal species (“cryptic species”), which correspond quite well to the major lowland forest regions recognized in the model. The most spectacular example for new cryptic species is the sportive lemur, Lepilemur: For many years, classifications recognized only a single species, Lepilemur mustelinus, with an extensive distribution throughout the forests of Madagascar. Eventually, this species was divided into two, with the red-tailed sportive lemur (L. ruficaudatus) inhabiting the humid forests in the east and the weasel mouse lemur (Lepilemur mustelinus) typically inhabiting the drier forests in the west. However, chromosomal and molecular evidence has progressively revealed the existence of at least 24 Lepilemur species. An equally striking case is that of the lesser mouse lemur, Microcebus, with the long-recognized island-wide species M. murinus initially divided into two: the brown mouse lemur (Microcebus rufus) in the east and the gray mouse lemur (Microcebus murinus; Fig. 1) in the west.

FIGURE 1 Gray lesser mouse lemur (Microcebus murinus), eating an

insect. This small-bodied nocturnal species has an omnivorous diet of arthropods and various plant items. At one time, all lesser mouse lemurs were allocated to this single species, but molecular evidence has now revealed the existence of at least 16 species. Photograph by David Haring.

FIGURE 2 Topography of Madagascar. The elevated central plateau is the source of major river drainages that serve as routes of retreat into refuges

and subsequent dispersion (white zones with letters). Centers of endemism (colored zones with numbers) are assemblages of smaller watersheds with sources at lower elevations isolated between retreat-dispersion watersheds. Recent distributions of lemurs at the level of species or subspecies are broadly concordant with the proposed centers of endemism. Illustration courtesy of Steven M. Goodman. Modified from Wilmé et al. (2006), Science 312: 1063–1065. Reprinted with permission from AAAS.

Subsequently, molecular evidence has gradually revealed the existence of at least 16 species of Microcebus, and it is likely that more remain to be discovered. A more elaborate model has been developed to explain the general confinement of individual animal species to restricted geographical ranges (microendemism) in Madagascar using an analysis of watersheds in the context of Quaternary climatic shifts (Figs. 2–3). River

catchments with sources at relatively low elevations were identified as zones of isolation that led to speciation of locally endemic taxa. By contrast, river catchments with sources at higher elevations were identified as zones of retreat and dispersion in which microendemism is correspondingly less pronounced. The resulting model provides a valuable framework for interpreting the process of explosive speciation on the island.

L E M U R S A N D TA R S I E R S

551

FIGURE 3 A female ringtail lemur (Lemur catta) with her ventrally

carried infant. Ringtail lemurs, which are restricted to relatively dry regions of southwestern Madagascar (areas 6, d6, and e6 in Fig. 2), are day-active and live in troops containing around two dozen individuals. Photograph by the author.

FIGURE 4 Like lesser mouse lemurs, the Philippine tarsier (Tarsius

syrichta), shown here with a lizard in its mouth, is a small-bodied nocturnal species, but it shows a number of special features. The very long hindlimbs reflect the specialized locomotor pattern of vertical-clinging and leaping between thin, vertical trunks. Relative to body size, the

TARSIERS

In contrast to lemurs, the tarsiers are members of a modest adaptive radiation, now confined to islands in the Southeast Asian archipelago. (Early fossil relatives have been documented from China and Thailand.) All species are allocated to the single genus Tarsius. One striking feature of tarsiers is the long hindlimb, notably including an elongated ankle region (tarsus). This

enormous eyes are bigger than those of any other mammal. Tarsiers are also the only primates that feed exclusively on animal prey. Photograph by David Haring.

feature, linked to their specialized locomotor pattern of vertical-clinging and leaping between thin, vertical trunks of saplings, has given rise to their common and scientific names. In connection with their nocturnal habits, they also have enormous eyes, which are by far

FIGURE 5 Distribution of tarsier spe-

cies on islands of the Southeast Asian archipelago. Myron Shekelle has suggested that the species complex on Sulawesi may contain 15 tarsier species. Updated from Niemitz, 1994.

552

L E M U R S A N D TA R S I E R S

the largest (relative to body size) of any mammal. All tarsiers are quite small, with body masses ranging from 60 to 135 g. Uniquely among primates, tarsiers feed exclusively on animal prey: mainly arthropods, but also small vertebrates—including highly venomous snakes. They have never been seen to ingest plant food of any kind. Until 1987 only three tarsier species were recognized: the western tarsier (Tarsius bancanus) on Borneo and Sumatra, the Philippine tarsier (Tarsius syrichta; Fig. 4), and the eastern tarsier (Tarsius spectrum) on Sulawesi. However, following a 1987 report on a pygmy species (Tarsius pumilus) discovered in museum collections, a fifth species (Tarsius dianae), also found on Sulawesi, was announced in 1991. By 2003, eight Sulawesi species had been recognized, and a study of vocalizations then revealed 15 distinct populations, all of which may be separate species (Fig. 5). As with the lemurs of Madagascar, it must be asked how so many tarsier species could have evolved on Sulawesi. There are numerous other examples of animal groups, such as macaques and frogs, that have shown equally spectacular speciation on Sulawesi, the eleventh largest island in the world. In contrast with the model developed for Madagascar, geographical barriers such as major rivers are not prominent features of Sulawesi, there is little climatic zonation, and the original vegetation covering the island was relatively uniform tropical rainforest. On the other hand, Sulawesi has a peculiar geological history, and this seems to account for much of the zonation connected with subdivision between individual species. Reconstructions have shown that four major land fragments drifted together from different directions to form the island. SEE ALSO THE FOLLOWING ARTICLES

Madagascar / Mammal Radiations

LINE ISLANDS CHRISTOPHER CHARLES AND STUART SANDIN Scripps Institution of Oceanography, La Jolla, California

Oceanographers are often given to building transects, where contrasting properties can be observed within a relatively narrow geographic area. Such is the opportunity afforded by the Line Islands in the central tropical Pacific, an island chain that stretches from Johnston atoll in the north to the Tuamotu Islands in the south. The very name of the chain invites this transect approach, and, although the basic connotation of “The Line Islands” does not do justice to the complexity of its geological origin, the appellation does quite aptly describe the gradients in geographic properties manifested on the various islands. A “NATURAL LABORATORY”

The atolls of the Line Island chain span several oceanographic and climatic zones. As a result, the terrestrial characteristics of the islands vary from lushly vegetated to nearly desert-like terrains, while the submerged coral reef fauna is subject to varying influences of the El Niño phenomenon. Furthermore, although the Line Islands archipelago is the most isolated collection of islands on the planet—no continent is closer than 5000 km from any of the Line Islands—the chain is now characterized by concentrated pockets of human disturbance; this concentration of human influence has important implications for the study of island resource management. All of these aspects of the Line Islands make the archipelago an essential natural laboratory for understanding the development, evolution, and governing processes of the tropical atoll environment.

FURTHER READING

Alterman, L., G. A. Doyle, and M. K. Izard, eds. 1995. Creatures of the dark: the nocturnal prosimians. New York: Plenum Press. Martin, R. D. 2000. Origins, diversity and relationships of lemurs. International Journal of Primatology 21: 1021–1049. Pastorini, J., U. Thalmann, and R. D. Martin. 2003. A molecular approach to comparative phylogeography of extant Malagasy lemurs. Proceedings of the National Academy of Sciences USA 100: 5879–5884. Tattersall, I. 1982. The primates of Madagascar. New York: Columbia University Press. Wilmé, L., S. M. Goodman, and J. U. Ganzhorn. 2006. Biogeographic evolution of Madagascar’s microendemic biota. Science 312: 1063–1065. Wright, P. C., E. L. Simons, and S. Gursky. 2003. Tarsiers: past, present, and future. New Brunswick: Rutgers University Press.

ORIGIN AND GEOLOGICAL FEATURES

Many linear island chains are often assumed to have arisen from the passage of a crustal plate over a fixed “hotspot” or mantle plume. In the case of the Line Islands, it is clear that this hotspot model cannot explain the age, distribution, and origin of the chain. First, the Line Islands are not a simple chain; instead, they must be understood as part of a complex array of islands and seamounts. Given this geometry, reconstruction of plate movement, or plate “backtracking,” does not lead to a convergence of all the islands and seamounts back to a

LINE ISLANDS

553

FIGURE 1 General bathymetric chart illustrating the geological con-

text of the Line Islands. The numbers in parentheses are the ages (in millions of years) of sea-floor rock samples collected at the indicated locations.

single point source. Furthermore, there is no northward progression of age of the islands or seamounts, in contrast to what one might expect if a plate passed over a single hotspot or mantle plume. In fact, the crustal age of the northernmost islands is similar to that of their southernmost counterparts, and the ages of the ridges and platforms upon which the Line Islands rest fall into two main clusters with the formation of the northern Line Islands in the Late Cretaceous (70–80 million years ago) and the subsequent formation of some of the islands and neighboring seamounts in the Paleogene (35–50 million years ago) (Fig. 1). Finally, although regular age-depth relationships often characterize the sea floor away from spreading centers (because of the progressive cooling and sinking of the marine crust), the sea floor associated with the Line Islands deviates markedly from the expected trend with age—the northern Line Islands are more shallow than one would predict on the basis of the age of the surrounding marine crust. Taken together, the observations argue that the formation of the Line Islands was most likely an extension of the same general volcanic construction processes that created the Marquesas and the neighboring islands of the South Pacific—a phenomenon of intraplate volcanism termed the “South Pacific Superswell.” Detailed study of the history of this volcanism and the age-depth relationships in the Line Island area may ultimately lead to a more complete picture of convection processes in the mantle. The relative antiquity of the Line Islands implies that the present morphology of the islands must be largely a product of the complex interplay between (1) reef con554

LINE ISLANDS

struction and sedimentation during sea level transgressions and (2) erosion and karstification of the emergent surface during sea-level regressions. The cycle of construction and erosion must have occurred repeatedly for all the Line Islands, given the history of sea-level fluctuations over the past 40 million years. As of yet, there have been no deep drilling campaigns that capture the entire sequence from island surface to basaltic crust, but one might nevertheless surmise the overriding importance of preexisting (antecedent) topography in determining the architecture of modern reef and land surfaces. This is especially true for the Line Islands, because there are no active tectonics in the region that otherwise might create significant structure. This antecedent topography must surely account for much of the variability in the morphology of the islands. Only five of the islands have typical lagoons that are connected directly to the surrounding ocean water (Kingman, Palmyra, Tabuaeran [Fanning], Kiritimati [Christmas], and Karoraina [Millennium]). Flint and Vostok are small islands occupying most of the reef platform upon which they rest, whereas Karoraina is a crescent-shaped atoll. The margins of Starbuck and Malden consist of continuous land, whereas the interiors of these islands have salty inland lakes. Teraina (Washington), on the other hand, features an inland freshwater lake. Tabuaeran has discontinuous islands around its rim and a complex network of reefs within its lagoon. Kiritimati Island is by far the largest of the Line Islands, although at least 25% of its interior surface consists of an interconnected network of saline lakes of variable depths. In general, the Line Islands are characterized by uniformly low-lying topography, with elevations rarely exceeding 10 m above present mean sea level. Terrestrial substrates are typically reef sands and rubble, but one notable feature of many of the islands is that the coral rubble can reach boulder size (these coral boulders were presumably deposited by either past tsunamis or large storms). CLIMATE

The characteristics of the Line Islands are shaped not only by their geological history but also by their location with respect to major oceanographic and climatic zones. The southern Line Islands (e.g., Flint, Vostok, Jarvis, Kiritimati) fall within the westward-flowing south equatorial current; the northern islands (Teraina, Palmyra, and to a lesser extent, Tabuaeran) lie in the realm of influence of the north equatorial countercurrent; and the northernmost islands (Kingman Reef, Johnston) are bathed by the westward-flowing north equatorial current. Forced by the winds, the surface ocean

currents deliver differing heat and nutrient concentrations to the coral reef communities of the Line Islands. For example, Jarvis and Kiritimati lie on western edge of the equatorial “cold tongue,” the region most directly influenced by the upwelling of colder, nutrient-rich subsurface water. Across the Line Islands, average annual rainfall varies over tenfold, ranging from approximately 500 mm yr-1 in the most southern islands (Flint and Vostok) to over 5000 mm yr-1 in the north (Palmyra). These average rainfall patterns are related directly to the mean annual position of the intertropical convergence zone, which, while undergoing some seasonal north–south migration, nevertheless remains north of the equator throughout the year in the region of the Line Islands. Annual averages of climate statistics do not provide a complete picture for the Line Islands because the interannual variability of temperature and especially rainfall in this part of the Pacific Ocean is highly significant. In fact, of all island chains across the tropics, the Line Islands are closest to the center of action for the El Niño–Southern Oscillation phenomenon (although the nature of the climatic influence does vary from island to island; for example, the temperature effect is maximized for the more equatorial islands) (Fig. 2). During strong El Niño years, the surface temperature in the region encompassing the Line Islands increases by as much as 3 °C during the December–March period, and the increase in rainfall may exceed historical averages by 200 to 300%. Kiritimati, for example, is subject to interannual sea-level variations of the order of 30–40 cm associated with El Niño warm events, and the greatly increased rainfall floods much of the atoll’s interior, transforming the normally dry hypersaline flats into shallow brackish lakes. Thus, the fauna and flora must be capable of withstanding significant interannual extremes in climate. On a serendipitous note, the coral boulders exposed on the Line Island beaches offer one of the best available means to access the history of El Niño over the last few millennia (the chemistry of coral skeletons records a snapshot of ambient conditions at the time of coral growth). BIOGEOGRAPHY

The extreme degree of isolation influences much of the islands’ biology. Relative to less isolated sites, the islands have few species, either terrestrial or marine. The difficulty of dispersal across wide expanses of deep ocean limits the number of species that arrive to the Line Islands, and the small sizes of the islands provide only a limited number of habitats for arriving species to exploit. The species that are successful in surviving in the archipelago have created biological communities that efficiently exploit the available

FIGURE 2 The Line Islands (white circles) are close to the center of the

climatic disturbance associated with the El Niño–Southern Oscillation phenomenon. Shown here are the anomalies in temperature (colors, in °C) and rainfall (contours, in mm/day, negative is dashed) during the 1982–1983 El Niño event.

resources. A subset of species that use the Line Islands are temporary visitors. As islands in the middle of the deep central Pacific, the Line Islands provide rare shallow water and land resources to wide-ranging species such as whales, dolphins, pelagic sharks, and other fishes, sea turtles, and seabirds. The isolation of these islands also has limited the history and extent of human activities in the region. Terrestrial Biota

Each of the Line Islands (except for Kingman) supports either forest or shrubland. The calcareous sand, low topography, and consequent lack of significant stored groundwater all imply that plants on the Line Islands cannot depend upon the soil to retain water between rainstorms. These conditions then create an intimate association between the floral assemblage and the amount of rainfall reaching the island. The most thickly vegetated islands are, not surprisingly, those that receive the most annual rainfall. There is limited species diversity within the plant assemblage across the archipelago. Most species have large, buoyant seeds or small, light seeds capable of dispersing very long distances across the sea (for example, by flotation, by wind, or by attaching to feathers of seabirds). Because of the propensity for long-distance dispersal, there are few endemic plants and instead the Line Islands are populated by a host of wide-ranging, pan-Pacific species. One of the most notable species is Pisonia grandis, a tree species that historically dominated many Pacific coral atolls. Pisonia forests are renowned for their great stature and dense foliage, but they are becoming rare across the Line Islands because of a host of stressors. An introduced pest (a green scale insect, Pulvinaria urbicola) is believed to be causing mass mortality within the Pisonia population on Palmyra, whereas direct felling and replacement with coconut palms (Cocos nucifera) has hastened decline on other islands. The flora of the islands is home to a limited, yet distinctive, collection of land animals. Most numerous on many of the Line Islands are the hermit crabs (principally

LINE ISLANDS

555

Coenobita brevimanus). Like their marine cousins, these hermit crabs use large snail shells (with openings up to 8 cm in diameter) to protect their bodies as they roam the forest floor for food. Also abundant is the larger land crab (Cardisoma carnifex), with a carapace width typically from 6 to 12 cm, which live in burrows and typically scavenge at night. Most impressive, however, is the coconut crab (Birgus latro), named for its ability to crack open coconuts with its claws. These crabs can grow to be 70 cm in thoracic length, have a leg span of 1 m, weigh a few kilograms (with some estimates over 4 kg), and live for decades. Because of their large size and their fabled taste, the coconut crab has been harvested extensively from the inhabited Line Islands. However, remnant populations of crabs thrive on the uninhabited islands (Palmyra, Millennium, Flint), with population estimates of hundreds of thousands to millions of crabs on the southern islands. The land and vegetation of the Line Islands serve a critically important role as nesting areas for at least 19 species of seabirds. Because of the paucity of land in the central Pacific, the Line Islands attract millions of birds annually to complete their life cycles. The birds nest in the native vegetation, including the Pisonia forests, the shrubby beach naupaka (Scaevola sericea), and the tree heliotrope (Tournefortia argentea). Some of the more spectacular seabirds to visit these islands are the redFIGURE 4 Representative 0.5-m2 photos of the bottom at (A) King-

man and (B) Kiritimati, again showing reef degradation.

footed boobies, the great frigatebirds, and both the redand white-tailed tropicbirds; enormous aggregations of sooty terns can fill the sky in pre-nesting displays. Some harvesting of seabirds and their eggs has reduced colony sizes on Kiritimati, Tabuaeran, and Teraina, but the uninhabited Line Islands continue to support a significant proportion of the seabird nesting for the central Pacific populations. Marine Biota

FIGURE 3 General aspect of fore reef habitats at (A) Kingman and

(B) Kiritimati, showing the degradation from a reef dominated by top predators and corals to a reef dominated by small planktivorous fishes and algae.

556

LINE ISLANDS

The Line Islands are each surrounded by expansive fringing reefs. On the windward sides of the islands, the fringing reefs are exposed to frequent large Pacific swells and have thus been structured into sharp, steep slopes beginning near to shore. Leeward reefs typically include an extended shallow terrace (100–300 m wide and 10–20 m deep), dropping off to a steeper seaward reef slope. In contrast to much of French Polynesia, the majority of coral reef development in the Line Islands is fore-reef (habitats seaward of the reef crest).

The Line Islands are part of the central tropical Pacific bioregion, with marine species diversity showing much overlap with that from neighboring archipelagoes. Each of these island groups is fairly isolated from the rest of the shallow Pacific, and thus the total number of species in the bioregion is relatively low. Nevertheless, the Line Islands are still home to over 400 species of fishes, 200 species of algae, and at least 50 genera of reef-building corals. However, the taxonomic catalogs for the Line Islands are far from complete and are likely to be augmented during each new research survey. Although not spectacular in their species diversity, the Line Islands are home to among the most healthy, ecologically intact coral reefs remaining on the planet. The reefs are characterized by very high fish biomass, dominated by predatory species such as snappers, groupers, and sharks, and a complex reef substrate, dominated by reef-building corals and coralline algae. Because of their isolation, the reefs of the Line Islands have remained greatly protected from the major stress caused by local human activities, predominately the stress of fishing. Many insights are being gained through the study of coral reef ecology in the Line Islands, in large part because of the opportunity to compare the structure and functioning of the community with and without local human influences (Figs. 3, 4). The lagoons of the Line Islands, as in many other places, tend to serve as nursery habitats. For example, the lagoons on Kingman and Palmyra support abundant populations of juvenile reef sharks that find safety from predators that frequent the fore-reef. Much of the benthic fauna of these lagoons is distinctive from forereef habitats. In particular, many species of clams live in the reef matrix of these lagoons. Most impressive among these are the giant clams (mainly Tridacna maxima), a species that can grow to upward of 40 cm in length and has been reported at densities of over 20 adults per square meter at Kingman. The flora and fauna of the freshwater and hypersaline pools contain a number of unique and endemic species about which very little is known to date. On the other side of the fore-reef habitat is the nearshore, pelagic environment. The Line Islands attract a large number of pelagic visitors, just as they attract seabirds above the sea’s surface. Predatory fishes, such as yellowfin tuna, wahoo, and mahi mahi, are found frequently on the leeward sides of the islands. These species feed on smaller bait fishes that aggregate in areas of nearshore upwelling and on larval reef fishes produced and advected from the nearby reefs. Three species of dolphins are found commonly across the archipelago: bottlenose dolphins (Tursiops truncates), spinner dolphins (Stenella longirostris), and melon-headed whales

(Peponocephala electra). Genetic analyses have revealed that these animals have significant site fidelity, with each island having its own resident population that mixes little, if at all, with populations from nearby islands. A number of other cetacean species are found in the area, including various beaked whales that have been found nowhere besides the Line Islands. Humans and the Line Islands

At the time of European discovery in the late eighteenth century, there were no human inhabitants in the Line Islands. Archaeological remains found on a number of the islands suggest that the islands were visited sporadically by seafaring Pacific islanders. However, the lack of sufficient rainfall on many of the islands and adequate soil for agriculture would have limited the people’s ability to develop long-term communities in the archipelago. Since the beginning of the nineteenth century, the islands have been used for a number of purposes. In the late nineteenth century, a number of the Line Islands were colonized by small guano-mining operations. Additionally, a number of copra plantations were created on a number of islands by clearing native forests and planting coconut palms. Most importantly, however, three of the islands (Kiritimati, Tabuaeran, and Teraina) were populated by the I-Kiribati people. During the first part of the twentieth century, these populations grew slowly, but during the later part of the century, an active population relocation program began moving people to the Line Islands from the overpopulated Kiribati island of Tarawa. As of 2005, the I-Kiribati population in the northern Line Islands was estimated at 8500 people (5100 on Kiritimati, 2500 on Tabuaeran, and 900 on Teraina). Although fishing and limited agriculture supports a fraction of the food needs, cargo ships from Tarawa and Hawaii heavily subsidize the food and fuel demands of the population. In recognition of the unique aspects of the Line Islands, a number of national and international efforts are in place to ensure the protection of the island chain. The three United States protectorates in the Line Islands (Kingman, Palmyra, and Jarvis) are designated as national wildlife refuges, managed by the U.S. Fish and Wildlife Service. A number of the Kiribati protectorates in the southern Line Islands also are protected as Kiribati wildlife sanctuaries. Finally, international efforts are under way to include the Line Islands as part of the nominated Central Pacific World Heritage Project, a program of the United Nations. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Climate Change / Coral / Pacific Region / Seamounts, Geology

LINE ISLANDS

557

FURTHER READING

Agardy, D. T. 2001. Scientific research opportunities at Palmyra atoll. A report submitted to The Nature Conservancy. Cobb, K. M., C. D. Charles, H. Cheng, and R. L. Edwards. 2003. El Niño–Southern Oscillation and tropical Pacific climate during the last millennium. Nature 424: 271–276. Davis, A. S., L. B. Gray, D. A. Clague, and J. R. Hein. 2002. The Line Islands revisited: New 40Ar/39Ar geochronologic evidence for episodes of volcanism due to lithospheric extension. Geochemistry, Geophysics, Geosystems: doi: 10.1029/2001GC000190. Dawson, E. Y. 1959. Changes in Palmyra atoll and its vegetation through the activities of man 1913–1958. Pacific Naturalist 1: 51. Dinsdale, E. A., O. Pantos, S. Smriga, R. A. Edwards, F. Angly, L. Wegley, M. Hatay, D. Hall, E. Brown, M. Haynes, L. Krause, E. Sala, S. A. Sandin, R. Vega Thurber, B. L. Willis, F. Azam, N. Knowlton, and F. Rohwer. 2008. Microbial ecology of four coral atolls in the northern Line Islands. PLoS ONE 3: e1584. Handler, A. T., D. S. Gruner, W. P. Haines, M. W. Lange, and K. Y. Kaneshiro. 2007. Arthropod surveys on Palmyra atoll, Line Islands, and insights into the decline of the native tree Pisonia grandis (Nyctaginaceae). Pacific Science 61: 485–502. Keating, B. H. 1992. Insular geology of the Line Islands, in Geology and offshore mineral resources of the central Pacific basin, Circum-Pacific Council for Energy and Mineral Resources Earth Science Series, vol. 14. B. H. Keating and B. R. Bolton, eds. New York: Springer-Verlag, 77–99. Sandin, S. A., J. E. Smith, E. E. DeMartini, E. A. Dinsdale, S. D. Donner, A. M. Friedlander, T. Konotchick, M. Malay, J. E. Maragos, D. Obura, O. Pantos, G. Paulay, M. Richie, F. Rohwer, R. E. Schroeder, S. Walsh, J. B. C. Jackson, N. Knowlton, and E. Sala. 2008. Baselines and degradation of coral reefs in the northern Line Islands. PLoS ONE 3: e1548. UNESCO. 2003. Central Pacific World Heritage Project, International Workshop Report. Woodrofe, C. D., and R. F. McLean. 1998. Pleistocene morphology and Holocene emergence of Christmas (Kiritimati) Island, Pacific Ocean. Coral Reefs 17: 235–248.

LIZARD RADIATIONS MIGUEL VENCES Technical University of Braunschweig, Germany

Lizards belong to the clade Squamata, together with snakes, and among nonflying terrestrial vertebrates, they are the ones most commonly observed on islands. Lizards are characterized by a great facility in colonizing islands and adapting to novel ecological circumstances by changes in their morphology, physiology, and reproductive biology. They have consequently become an important model group for the inferential and experimental study of adaptive radiations. LIZARDS ON ISLANDS

On major land-bridge islands with favorable climates (i.e., in the tropical, dry, and temperate zones) both liz-

558

L I Z A R D R A D I AT I O N S

FIGURE 1 Emblematic island lizards. (A) Gallotia stehlini, Gran Canaria.

(B) Gallotia galloti, Tenerife. (C) Chalcides sexlineatus, Gran Canaria. (D) Tarentola delalandii, Tenerife. These species and their relatives have originated on the Canary Islands. Photographs by Miguel Vences.

ards and snakes are commonly encountered, with snake species richness often being similar to lizard species richness. On 14 major islands and island groups of the Mediterranean Sea, there are 152 occurrences of 30 species of lizards and 28 species of snakes. However, native extant snakes are missing on many smaller islands and on oceanic archipelagoes such as the Macaronesian Islands (Canary and Cape Verde Islands, Savage Islands, Madeira, and the Azores), where native species and even endemic radiations of lizards are present (Fig. 1). The most remote oceanic islands (e.g., Hawaii) are devoid of both native lizards and snakes. Although within-island diversification is rare in snakes and is limited to very large islands such as Madagascar, lizards have diversified on medium-sized islands such as the Greater Antilles as well (see below). Of the currently known ~5000 species and 26 families of lizards, representatives of the Gekkonidae, Iguanidae, Lacertidae, and Scincidae are most commonly encountered on islands. Continental islands, especially, may frequently act as an evolutionary reservoir by enabling the survival of remnants of lineages that became extinct or very rare on the mainland. Such is the case of the tuataras, two species of lizard-like reptiles which are the last extant representatives of the Sphenodontia (the sister group of squamates). At present, tuataras are confined to various small islands off New Zealand, although fossil remains demonstrate their past presence on the New Zealand mainland, and that of their relatives on other continents. On the Balearic Islands in the Mediterranean Sea, the lizard Podarcis lilfordi is present only on tiny offshore islands surrounding the larger islands of Mallorca and Menorca, where they are extinct. On Madagascar, the radiation of snakes in the subfamily Pseudoxyrhophiinae is very diverse, but

FIGURE 2 The largest lizard worldwide, the Komodo dragon, Varanus

komodoensis. Photograph by Thomas Ziegler.

FIGURE 3 Both of the smallest lizards worldwide occur on islands. (A)

The gecko Sphaerodactylus ariasae occurs on Isla Beata and adjacent areas of Hispaniola. Photograph by S. Blair Hedges. (B) Adult male Malagasy leaf chameleon of an undescribed species in the genus Brookesia from the extreme north of Madagascar. Photograph by Frank Glaw.

this lineage has only a few representatives in Africa, where it probably has been replaced by other snakes. Both the largest and smallest extant lizards occur on islands: The largest is the Komodo dragon (Varanus komodoensis) with a maximum snout–vent length of over 1500 mm (Fig. 2); the smallest (Fig. 3) are two species of Sphaerodactylus geckos (S. ariasae and S. parthenopion) from the Caribbean, with adult snout–vent lengths of about 16 mm, and several species of Malagasy leaf chameleons (Brookesia) with adult snout–vent lengths of 14–19 mm. Lizards appear to show a trend of island gigantism and dwarfism opposite to what is generally considered as a rule: In lineages of small lizards, the island populations and species become even smaller, and in lineages of large forms, the island representatives become even larger, especially in carnivorous taxa. Snakes also show size changes in island populations and species, and snakes that evolved to become small on islands did so to a relatively greater degree than those that became large. The observed pattern suggests that snake body size is principally influenced by prey size, with large snakes mainly feeding on nesting seabirds and small snakes mainly feeding on lizards. Many island lizards have adapted to resources that differ from those available on the nearby mainland. The most famous is the marine iguana from the Galápagos (Amblyrhynchus cristatus), the only lizard that feeds on algae while diving in the ocean. Many lizards of the family Lacertidae were originally insectivorous but became herbivorous on islands. In fact, herbivory in mainland lineages may be an important “preadaptation” that allows for successful colonization of island habitats. A further intriguing difference between island and mainland populations of lizards is population density, which is generally one order of magnitude higher on islands. This phenomenon is likely driven by distinctly lower numbers of predators and competitors. These same factors may also have allowed island lizards to expand

L I Z A R D R A D I AT I O N S

559

their diet to include nectar, pollen, and fruit. Indeed, in several island ecosystems, lizards also occupy an important role as pollinators and seed dispersers. Few studies have addressed changes in reproductive strategy in island populations of lizards, but in species of the family Lacertidae a trend of reduced clutch size and larger egg size on islands has been noted. COLONIZATION OF ISLANDS BY LIZARDS

Recent years have seen a paradigm shift in our understanding of the occurrence of many taxa on islands. This has involved a shift from the dominance of vicariance explanations to hypotheses in which dispersal plays at least an equally important role. In general, the mode of reproduction of lizards and snakes, with internal fertilization, favors overseas dispersal because the arrival of a single gravid female to an island can be sufficient to give rise to a new population. For lizards, there is no doubt that their dispersal capacities are high and that they have on many occasions colonized islands over water from the mainland or from other islands. For green iguanas, direct evidence exists that after a hurricane in 1995, at least 15 individuals arrived on a mat of logs and uprooted trees on the eastern beaches of Anguilla and other islands in the Caribbean, and some specimens survived there for at least three years. Molecular genetic analyses have provided evidence for various events of long-distance dispersal between Africa and South America (e.g., in geckos of the genera Tarentola and Hemidactylus, and in skinks of the genus Trachylepis). For example, Trachylepis atlantica from the Fernando de Noronha Archipelago in the Atlantic, 350 km east of the Brazilian coast, belongs to this mainly African and Malagasy genus rather than to the related Neotropical genus Mabuya. Its ancestors presumably colonized by overseas dispersal from Africa rather than from nearby South America. Native populations of lizards (and often endemic species) are found on many oceanic islands: on major archipelagos such as Macaronesia, the Galápagos, the Gulf of Guinea islands, the Comoros, and the Mascarenes, but also on many small and isolated islands. The Australian region, including small islands such as those of the Solomon and Bismarck archipelagoes, harbors a massive radiation of the scincid genus Sphenomorphus, and other skinks (genus Emoia) have radiated on most islands in the southwestern Pacific, including, among many others, the Fiji, New Caledonia, Solomon, and Bismarck archipelagoes. This further demonstrates the capacity for overseas dispersal of lizards. 560

L I Z A R D R A D I AT I O N S

FIGURE 4 An endemic species of chameleon from the Comoro island

of Mayotte, Furcifer polleni. Photograph by Frank Glaw.

Inverse routes of colonization, from islands back to the mainland, have occurred as well. This appears to be the case for a Central and South American clade within the genus Anolis, which probably originated from a West Indian ancestor, and it is possibly also true for chameleons, which may have dispersed multiple times from Madagascar to mainland Africa, and which certainly have dispersed from Madagascar to Mayotte (Fig. 4). On the Gulf of Guinea islands (São Tomé, Principe, and Annobon), a relatively high proportion of endemic burrowing species of lizards and snakes occur, indicating that the capacity of overseas dispersal also extends to species living in humid soil and leaf litter. A combination of ocean currents, floating islands, and reduced surface salinity caused by freshwater discharges from large rivers may be favorable to overseas dispersal events in general and may also enable such soil-dwelling species to colonize islands. Eggs of some lizards are known to be resistant to immersion in seawater. In the case of Anolis sagrei, this may explain the survival of populations of this lizard on small islands vulnerable to hurricanes, but it also may allow the overseas rafting of lizard eggs in tree holes or mats of vegetation. In some cases, commensal species of lizards have been translocated by humans. Several species of geckos of the genus Hemidactylus have a transcontinental distribution that in some cases is due to natural colonization but often may reflect deliberate or, more probably, accidental introductions. Lipinia noctua, a scincid lizard that lives alongside humans on islands of the central and eastern Pacific, displays a phylogeographic pattern concordant with the “express train” hypothesis: Specimens may have been transported as stowaways on early Polynesian canoes during the rapid human colonization of Polynesian islands.

PATTERNS OF INSULAR LIZARD RADIATIONS

The process of speciation can be either (1) adaptive (i.e., the process of an ancestral population diverging and giving rise to two daughter lineages adapted to different niches) or (2) nonadaptive (e.g., the separation of the daughter species by geographic barriers or by differentiation of features that serve for species recognition). Most lizard radiations on smaller islands probably belong to the category of nonadaptive and allopatric speciation on different islands. This same mode of speciation has also taken place within some islands of sufficient size. A few possible examples also exist for sympatric adaptive speciation within an island. As an instance of nonadaptive speciation on different islands, the western Canary Islands are populated by small radiations of skinks and geckos (Chalcides and Tarentola), but on each island or group of islands, only one species of each genus occurs. The situation is slightly more complex in the Canarian lacertid lizards, genus Gallotia: Here an initial split is observed between large-sized and small-sized species, and sympatry occurs only between (ecologically strongly differentiated) representatives of either group (on Hierro, Gomera, Tenerife, and probably La Palma, if extant species and natural occurrences are considered). Day geckos of the genus Phelsuma have radiated on the Seychelles and Mascarenes, and on each of these two archipelagoes there is a monophyletic lineage of various species and subspecies. At least on the Seychelles, the available evidence favors allopatric speciation of the three endemic taxa on different islands, with secondary sympatry in some cases. Crucial to test hypotheses of radiation on islands are robust phylogenies. However, critical data on the interplay of dispersal and vicariance can be provided by the geological age of an island or of its last connection to the mainland, and hence the age of evolutionary splits in the lineage under study. For example, the two Galápagos iguanas (the terrestrial genus Conolophus and the marine iguana Amblyrhynchus; Fig. 5) occur on the same islands and do form a monophyletic group. This could be interpreted as an example of speciation by ecological specialization under sympatric conditions. However, the age of the evolutionary divergence between these species predates the geological origin of the current Galápagos Islands. This indicates that either (1) they must have diverged on a previous, now submerged land mass, or (2) both species originated on the mainland, they colonized the Galápagos independently, and their mainland relatives subsequently went extinct. In general, the possibility of extinction must always be taken into account to under-

FIGURE 5 Galápagos iguanas. (A) The marine iguana, Amblyrhynchus

cristatus. Photograph by Ylenia Chiari. (B) A terrestrial iguana, Conolophus subcristatus. Photograph by Scott Glaberman.

stand the biogeographic history of lizard populations on islands. The best-studied case of an insular lizard radiation is that of the Caribbean genus Anolis (Iguanidae), the anoles, which are among the most common terrestrial vertebrates in the Caribbean and are found on almost every island in this region. There are over 400 species of anoles, of which nearly 150 are Caribbean. Their origin has been estimated at around 40 million years ago, and fossil specimens preserved in amber are known from the Oligocene to the Miocene of the Dominican Republic. The patterns of anole radiation have been intensively studied by Jonathan B. Losos and colleagues. Summaries are found in Losos (1998) and Losos and Thorpe (2004), from where much of the following information has been extracted. Anoles are very good dispersers, evidenced by cases of related taxa occurring on islands of great geographic distance. However, by far the highest proportion of Caribbean anoles are endemic to single island banks (more than 85%). A few cases of natural hybridization are known, but L I Z A R D R A D I AT I O N S

561

in general, mismating among species of these lizards is prevented by the throat fans (“dewlaps”) of males, which show specific colors and patterns used in species recognition. In fact, sympatric species of anoles always differ in the size, color, or patterning of their dewlaps. Up to 11 species of anoles can coexist at a single site, and such sympatric species almost always differ in terms of habitat use and morphology or physiology. The number of anole species coexisting on a certain island is significantly correlated with island size. Considering only small islands (i.e., islands of a surface of 1500 km2 or less), the species–area relationship is stronger for islands that were in the past connected by land bridges to other land masses than for isolated islands, highlighting the importance of historical effects: Land-bridge islands probably had a higher number of species at the time of isolation, and through subsequent extinctions, species numbers adjusted to the island-specific ecological carrying capacity. In contrast, isolated islands depend fully on over-water colonization as the source for species. Isolated islands mostly are populated by a single species of anole only, with a maximum of two species per island (which then differ in their ecology). Apparently, colonization of small isolated islands by anoles can be successful only if (1) the island does not yet harbor any anole population or (2) the island is populated by an anole species that differs in ecological requirements from the new colonizers. Evolutionary diversification of anoles appears to occur on a single island when its size is above a certain threshold. In the Caribbean, within-island diversification has occurred on the Greater Antilles (Jamaica, Puerto Rico, Hispaniola, and Cuba). Each of these large islands harbors endemic divergent lineages, which contain various species and, hence, very probably originated on the island. Within-island speciation can be invoked for at least 70% of the Greater Antillean anoles. A few examples from smaller islands or island groups exist of cooccurrence of endemic taxa that could have arisen on the same island, but these cases are not compelling. Hence, a certain island area is necessary for within-island speciation, a conclusion that highlights the importance of geography for this process. The Anolis radiations on the four Greater Antillean islands (although phylogenetically independent) show recurrent patterns. As was first pointed out by Ernest Williams, different types of habitat specialists (ecomorphs) occur on all or most of the Greater Antilles. These are usually represented by several species on each island (Figs. 6–8). Initially six ecomorphs were proposed, but others have since been distinguished. Interestingly, molecular

562

L I Z A R D R A D I AT I O N S

FIGURE 6 Ecomorphs of Caribbean Anolis. All species shown are from

Hispaniola. Names roughly denote the preferred habitat of each ecomorph. (A) Crown giant: Anolis baleatus. (B) Trunk crown: A. coelestinus. Note that the photographs are not to scale; Crown Giants are much larger than all other ecomorphs. Photographs by S. Blair Hedges.

FIGURE 7 Ecomorphs of Caribbean Anolis, continued. (A) Trunk: A.

christophei. (B) Trunk ground: A. cybotes. Photographs are not to scale. Photographs by S. Blair Hedges.

Cuba was fragmented during the Miocene. The Anolis alutaceus group, also on Cuba, contains 12 species with narrow distributions, mostly centered on different mountain ranges, a pattern that is also seen in other groups. Which prevalent pattern of species formation gave rise to the current diversity of anoles? Adaptive speciation in sympatry or parapatry may occur in Caribbean anoles, but it is probably not the main driving force explaining their diversity. In many cases, populations became isolated on small land-bridge islands or reached isolated small islands by overseas dispersal. Geographically and thus genetically separated from other anole populations, they evolved different morphologies and dewlaps, probably largely because of adaptation to new ecological conditions. On the larger islands, species belonging to the main ecomorphs underwent allopatric speciation (e.g., on different mountain ranges or on parts of their island that were separated by water barriers in periods of rising sea levels). As summarized in the following section, many examples indicate that adaptation can occur in the absence of speciation in Caribbean anoles. But it is still uncertain how the initial differentiation of ecomorphs on each of the Greater Antillean islands took place. PHYLOGEOGRAPHY AND EXPERIMENTAL TESTS OF SELECTION

FIGURE 8 Ecomorphs of Caribbean Anolis, continued. (A) Stream: A.

eugenegrahami. (B) Grass: A. semilineatus. (C) Twig: A. placidus. Photographs are not to scale. Photographs by S. Blair Hedges.

data show that, with two exceptions, the ecomorphs arose independently on the different islands: Different ancestors diversified independently and gave rise to the same ecological and morphological adaptations. Species belonging to different ecomorphs usually occur sympatrically, but species belonging to the same ecomorph generally are geographically separated within an island (and have different dewlap colors or patterning). In addition to the six main ecomorphs, many islands harbor further habitat specialists, but these usually occur on a single island only. In several cases, the different species of one ecomorph occur in geographically separated populations scattered across an island. In the Anolis carolinensis group, three evolutionary lineages can be distinguished and have ranges corresponding to three paleo-archipelagoes into which

Deciphering radiations is possible by looking at general patterns across a whole group or by examining in more detail the microevolutionary processes. Comparison of DNA sequences allows phylogeographical analyses where chiefly the geographical distribution of differentiated alleles (haplotypes) is mapped, and the phylogenetic relationships among these haplotypes is determined. The assumption is that haplotypes evolve through mutation, and different haplotypes get fixed in genetically isolated populations. In various studies on anoles and Canarian lizards, Roger S. Thorpe and colleagues have found evidence for discordance between historical and adaptive patterns. For example, in Gallotia lizards on the Canarian island of Tenerife, a historical boundary of mitochondrial haplotype lineages exists between western and northeastern areas, whereas within both groups, morphological differences were found between northern and southern populations, reflecting strong ecological differences between the humid north and arid south of the island. On Dominica, Anolis oculatus shows a complex phylogeographical structure that is not fully concordant with the phenotypic variability encountered. These examples demonstrate that morphological adaptations to local conditions, especially in terms of col-

L I Z A R D R A D I AT I O N S

563

oration, can evolve very fast in island lizards. This is also witnessed by the large variability of lacertid lizard species inhabiting Mediterranean islands (e.g., Adriatic islands, satellite islands of the Balearics, or Tyrrhenic islands in Greece). From many of these archipelagoes, a plethora of subspecies have been described based on color patterns and partly on variation in scale numbers, but molecular studies have rarely found any significant differentiation between these populations, indicating that the external differences evolved extremely rapidly, on a geological timescale. Other work has yielded evidence that in Anolis sagrei, the number of body scales increases with increasing precipitation and with decreasing temperature in open arid habitats, and the variation in scale numbers is probably heritable. In further experiments, the effects of a potential predator (the ground-dwelling lizard Leiocephalus carinatus) on the behavior of Anolis sagrei was tested by introducing the potential predator on six small islands on the Bahamas and using six other predator-free islands as control sites. As a result, anoles altered their behavior by using the ground less often, but in addition, a strong selection took place: Surviving specimens on the experimental islands had larger body sizes and longer hindlimbs than those on control sites, probably reflecting their better capacities to escape. Evidence for strong selection pressures acting on island lizards also comes from further experimental studies. The Dominican Anolis oculatus displays various ecomorphological variants related to different conditions between the east and west coasts and the montane regions of the island. In experiments, lizards were translocated to large lizard-proof enclosures in regions occupied by other habitat types than those in their source population. Morphology (coloration, scale counts, body proportions) of the translocated lizards were scored, and each lizard individually marked. Several months later, survivors were collected and identified. Morphological differences were found between survivors and non-survivors (e.g., of specimens of the montane population in enclosures of the relatively xeric west coast), and the intensity of selection was dependent on the magnitude of ecological change experienced by the specimens in the enclosures. How these intraspecific processes of fast morphological variation relate to the actual process of species formation and adaptive radiation is not clear. Evidence of parapatric forms with restricted gene flow among them comes from the islands of Dominica and Martinique; on Martinique this may constitute evidence for adaptive (ecological) species formation because the forms are distinguished by current

564

LO P H E L I A OA S E S

habitat and not by historical allopatry. It seems clear that these lizards have a strong potential to adapt to new ecological conditions by changes in morphology and coloration, and this may have favored adaptive speciation (mostly under allopatric conditions). This may also be a factor explaining the recurrent evolution of similar ecomorphs. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Convergence / Dispersal / Komodo Dragons / Snakes FURTHER READING

Losos, J. B. 1994. Integrative approaches to evolutionary ecology: Anolis lizards as model systems. Annual Reviews of Ecology and Systematics 25: 467–493. Losos, J. B. 1998. Ecological and evolutionary determinants of the speciesarea relationship in Caribbean anoline lizards, in Evolution on islands. P. R. Grant, ed. Oxford: Oxford University Press, 210–224. Losos, J. B., and R. S. Thorpe. 2004. Evolutionary diversification of Caribbean Anolis lizards, in Adaptive speciation. U. Dieckmann, M. Doebeli, J. A. J. Metz, and D. Tautz, eds. Cambridge: Cambridge University Press, 322–344. Olesen, J. M., and A. Valido. 2003. Lizards as pollinators and seed dispersers: an island phenomenon. Trends in Ecology and Evolution 18: 177–181. Williams, E. E. 1983. Ecomorphs, faunas, island size, and diverse end points in island radiations of Anolis, in Lizard ecology. R. B. Huey, E. R. Pianka, and T. W. Schoener, eds. Cambridge, MA: Harvard University Press, 326–370.

LOPHELIA OASES SANDRA BROOKE Marine Conservation Biology Institute, Bellevue, WA

The deep-water stony coral Lophelia pertusa (Linnaeus 1758) creates extensive and complex structures on hardbottomed areas in the deep sea, including continental shelf bedrock, lithified sediment mounds, volcanic basalt, and (microbially mediated) authigenic carbonate. Large colonies of L. pertusa have abundant tangled branches that provide habitats for diverse and abundant associated communities. These long-lived and slow-growing coral ecosystems are currently under threat globally from negative human impact, and although some areas have been placed under protective legislation, continued international effort is needed to ensure the future of these valuable resources. CORAL BIOLOGY

There are several species of “framework-building” deep-water corals (Lophelia pertusa, Oculina varicosa,

Enallopsammia profunda, Solenosmilia variabilis, and Goniocorella dumosa), all of which have similar characteristics. Individual colonies are complex branching structures that can be several meters high. These corals are broadcast-spawning species, releasing eggs and sperm into the water column for larval development. In the North Atlantic, Lophelia spawns during late February and early March, and each new coral colony is formed from the settlement of a single larva onto hard substrate; the resulting coral polyp divides asexually, and as the colony grows, the outer branches block the flow to the inner (older) parts of the colony, which eventually die. Subsequent invasion by boring and encrusting organisms weakens the inner core of the colony, and it falls apart, exposing the dead center to overgrowth by living coral. The standing dead coral provides hard substrate for a variety of other fauna and provides a micro-habitat for many small organisms. The living coral branches may come into contact with each other as the colony grows, and the braches commonly fuse with each other, which provides additional stability. Over time, this sequential growth, death, and overgrowth can create massive mounds composed of unconsolidated dead coral and sediment, with an outer layer of live coral (Fig. 1). These are referred to as bioherms and may be hundreds of meters deep. Deep-water corals do not have the algal symbionts common in shallow-water species, and they survive primarily on zooplankton. Corals require a moderate (and fairly continuous) current; water flow delivers food and oxygen, removes metabolic waste products, and reduces accumulation of sediment, which

FIGURE 1 A well-developed thicket of Lophelia from deep-water

bioherms along the southeast coast of Florida. The bright white branches are live Lophelia, with open polyps giving the colony a fuzzy appearance. The darker branches are dead skeleton, which is often colonized by other animals. Photograph courtesy of Brooke et al., 2005, NOAA Office of Ocean Exploration (http://oceanexplorer.noaa .gov/explorations/05deepcorals).

can suffocate the polyps. Growth rates for Lophelia have been estimated at 5–25 mm yr-1. At this rate, it would take thousands of years to form the extensive coral ecosystems that have been discovered in recent years. Experimental and field observations have shown that the upper thermal limit for Lophelia survival is approximately 12 °C, and temperature is undoubtedly a controlling factor in the depth distribution of this coral. Deep coral communities may also provide information about climate change in the deep ocean. Because of their worldwide distribution and longevity, cold-water corals are an excellent proxy for reconstructing past changes in global climate and ocean conditions. GLOBAL DISTRIBUTION OF LOPHELIA PERTUSA ECOSYSTEMS

Deep-water corals have been recorded from many topographic features throughout most of the world’s oceans. The highest known density of deep-water coral ecosystems occurs in the North Atlantic, but this may be an artifact of the high level of research effort in that area. These communities have provided the basis for much of the current knowledge of deep-water corals. Norway has the largest known Lophelia systems (approximately 2000 km2) particularly along the midNorwegian shelf at a 200–400 m depth. The shallowest known Lophelia systems (40 m in depth) occur in the fjords of Norway, where deep oceanic water intrudes into narrow channels and recreates deep-water conditions at shallow depths. At the other extreme, Lophelia has also been found on the Mid-Atlantic Ridge at more than 3000 m, but is most commonly found between 200 m and 1000 m. The most northerly Lophelia structure in the western Atlantic covers a small, 1-km-long area at the mouth of the Laurentian Channel in Canada, but the continental margin between the Blake plateau off North Carolina and the Miami terrace in South Florida supports the most well developed and extensive Lophelia complexes off the North American coast. In contrast to the Atlantic coast, there are few well-developed coral complexes in the Gulf of Mexico. The limestone bedrock of the western Florida shelf supports a series of Lophelia mounds at 500 m depth, and elsewhere in the Gulf the soft sediment sea floor is interspersed with boulders of authigenic carbonate, which supports the development of hard-bottom communities. The most extensive Lophelia communities in the Gulf of Mexico have been found at depths of 420–530 m on Viosca Knoll, a large mound approximately 100 km south of Mobile Bay.

LO P H E L I A OA S E S

565

Colonies of Lophelia have been observed in the eastern Pacific from California to Alaska. In 2006, Lophelia was discovered in the Olympic Coast National Marine Sanctuary off Washington State. Although there are currently no records of Lophelia elsewhere in the Pacific Ocean, it is probable that there are communities still to be discovered. Information on deep-water coral distribution is particularly lacking from tropical and subtropical regions where countries do not have the funds or technology to conduct deep-water exploration. Cold-water corals have also been documented on artificial substrates such as oil installations and wrecks. Hundreds of Lophelia colonies were observed on oil platforms in the North Sea. Large Lophelia colonies were also observed growing on a World War II wreck at 554 m in the northern Gulf of Mexico. With fossil fuel operations moving into deeper waters, oil and gas platforms may provide substrate for deep-water coral development in the Gulf of Mexico as they do in the North Sea. CORAL-ASSOCIATED COMMUNITIES

The complex structure produced by the coral branches provides substrate for sessile benthic organisms, food and refuge for many small invertebrates and fish, and food for larger predators. Like shallow tropical reefs, deep-water corals also provide structures for spawning aggregations and nursery habitats for various fish species. Coral complexes form biodiversity hotspots in deep water, and although there have been few quantitative studies of coral-associated fauna, a great deal of census information has been collected over the past decade. A complex reef structure can be divided into three general areas: the live coral, the dead coral rubble underneath, and the rubble/soft sediment at the base of the coral mound.

FIGURE 2 View of Lophelia thicket showing the bright red squat

lobster Eumunida sp., with claws outstretched, presumably waiting for prey. Small anemones and a pencil urchin are also visible among the branches. Photograph courtesy of Brooke et al., 2005, NOAA

Office

of

Ocean

Exploration

.gov/explorations/05deepcorals).

566

LO P H E L I A OA S E S

(http://oceanexplorer.noaa

FIGURE 3 Close-up view of a spectacular yellow glass sponge (fam-

ily Hexactinellidae) nestled between dead Lophelia branches. A new species of amphipod was found living inside this species of sponge from the South Atlantic. Photograph courtesy of Brooke et al., 2005, NOAA

Office

of

Ocean

Exploration

(http://oceanexplorer.noaa

.gov/explorations/05deepcorals).

Relatively few organisms live on the live coral (polyps contain stinging cells that may repel larval settlement); however, the carnivorous polychaete (Eunice sp.) is frequently found within the live branches. The worm excretes a soft tube, which the coral then overlays with calcareous skeleton, thus strengthening the tube for greater protection. In return, these calcified tubes make the colony more robust. Mobile organisms are also observed among the live branches; one of the most common is a red squat lobster (Eumunida sp.), which sits in the coral with claws outstretched, apparently waiting for food to pass by (Fig. 2). The dead coral framework supports a much greater abundance and diversity of organisms than does the live coral. Hundreds of different species from many taxonomic groups live in this habitat, with gorgonians, black corals, anemones, and a diverse array of glass sponges (Fig. 3) being among the most common of the larger fauna. Some species bore into the skeleton, weakening it and causing collapse over time; some encrust the branches; and others simply use the dead structure as a refuge or hunting ground. In the northeast Atlantic, a census identified over 1300 species associated with Lophelia colonies. Deep coral communities have also been identified as habitats for hundreds of species of fish, many of which are commercially valuable. ANTHROPOGENIC IMPACTS ON LOPHELIA ECOSYSTEMS

There are several potential threats to deep-water coral ecosystems, some more prevalent and potentially damaging than others. These include physical impacts such as commercial bottom fishing, hydrocarbon extraction,

deployment of cables and pipelines, bio-prospecting, and coral harvesting. The greatest of these is bottom trawling, which drags weighted nets held open by heavy “doors” over the fragile corals. Large trawl nets can cover many square kilometers in a single fishing trip, and reccovery (if it occurs) may take hundreds of years. Another potential threat to deep water coral communities comes from impending changes in the ocean as a byproduct of burning fossil fuels. As carbon dioxide increases in the atmosphere, it diffuses into the ocean and upsets the balance of ocean chemistry, making the ocean more acidic. Coral skeleton is made from a calcium carbonate, which dissolves under acidic conditions, therefore as the pH drops, the corals will find it harder to make their supporting skeletons. Fossil fuel exploration and extraction activity is a potential threat to deep-water corals if they are found in close proximity to large oil and gas deposits, particularly in the North Sea and the Gulf of Mexico. Live Lophelia has been observed close to drilling operations; however, more research is needed to determine the effects of fossil fuel extraction on deep-water coral communities. Laying gas pipelines and communication cables may also damage deep-water communities because large anchors are often used to stabilize the surface vessel during deployment. Other threats to deep-water corals exist, but their impact is relatively minor at present.

the coast of Scotland, a series of Lophelia bioherms called the Darwin Mounds was placed under special protection in 2004; bottom fishing was permanently prohibited over a 1300 km2 area. Other countries such as Ireland, the Azores, the Canary Islands, and Madeira also implemented similar policies in 2004. In the United States, several large areas have been placed under protection from bottom-damaging activites in the past five years, and more protected areas are currently being proposed. FUTURE CHALLENGES

As advances in technology allow humans to further exploit the deep ocean, the need for protective legislation is essential. Deep-water coral systems are usually far offshore and encompass large areas, so, even with protection in place within national EEZs, enforcement of regulations is extremely difficult and expensive, and protection from trawling on the high seas is yet a greater challenge. In the past decade, our understanding of deep-water corals has greatly increased due to funding from national and international sources and to the collaborative efforts of scientists and governmental institutions. However, these oases of biodiversity and abundance in the vastness of the deep sea are in urgent need of research, conservation, and protection from human exploitation. SEE ALSO THE FOLLOWING ARTICLES

CONSERVATION AND MANAGEMENT OF DEEP-WATER CORALS

Many deep-water coral systems lie outside national exclusive economic zones (EEZ) and are therefore not covered by any legal jurisdiction; however, these vulnerable ecosystems are currently the focus of international efforts to create legal protection under the United Nations Convention on the Law of the Sea (UNCLOS). Over the past five years, several countries have enacted or are in the process of establishing regional measures to protect and manage their deep-water coral ecosystems. These measures vary greatly and range from requirements for environmental impact assessments prior to conducting activities around deep-water corals to designation of marine protected areas with specific regulations in place. Norway was the first country to implement protection of Lophelia reefs after it was estimated that 30–50% of the Lophelia had been damaged by bottom trawling. The Norwegian fisheries authorities established a regulation that prohibited intentional destruction of coral habitat and provided special protection to five selected areas by banning the use of bottom gear altogether. Norway is also in the process of implementing an additional series of marine protected areas. Off

Coral / Fish Stocks/Overfishing / Marine Protected Areas FURTHER READING

Continental Shelf Associates. 2007. Characterization of northern Gulf of Mexico deepwater hard bottom communities with emphasis on Lophelia coral. U.S. Department of the Interior, Minerals Management Service, Gulf of Mexico OCS Region, New Orleans, LA. OCS Study MMS 2007-044. Freiwald, A., J. H. Fossa, A. Grehan,T. Koslow, and J. M. Roberts, 2004. Coldwater coral reefs: out of sight no longer out of mind. Biodiversity Series No. 22, Cambridge: UNEP-WCMC. Available online at http://www .unep-wcmc.org/resources/publications/UNEP_WCMC_bio_series/22htm Freiwald, A., and J. M. Roberts, eds. 2005. Cold-water corals and ecosystems. Erlangen Earth Conference Series. Heidelberg, Germany: Springer Publishing House. Gianni, M. 2004. High seas bottom trawl fisheries and their impacts on biodiversity of vulnerable deep sea ecosystems: options for international action. Gland, Switzerland: IUCN. Lumsden, S. E., T. F. Hourigan, and A. W. Bruckner, eds. 2007. The state of deep coral ecosystems of the United States. NOAA Technical Memorandum NOS-CRCP-3. Silver Spring, MD. Messing, C. G., J. K. Reed, S. D. Brooke, and S. W. Ross. 2008. Deepwater coral reefs of the United States, in Coral Reefs of the USA. B. Riegel and R. Dodge, eds. Heidelberg, Germany: Sprnger Publishing House, 763–787. Rogers, A. D. 1999. The biology of Lophelia pertusa (Linnaeus 1758) and other deepwater reef-forming corals and impacts from human activities. International Review of Hydrobiology 84: 315–406.

LO P H E L I A OA S E S

567

LORD HOWE ISLAND CAROLE S. HICKMAN University of California, Berkeley

Lord Howe Island is the spectacular remnant of a large shield volcano that was active 7 million years ago in an isolated part of the southwestern Pacific Ocean between Australia and New Zealand. The high proportion of rare and endemic plants and animals, the rugged natural beauty of the landscape, and the occurrence of the southernmost coral reef in the world contribute to its status as a UNESCO World Heritage Site. The Permanent Park Preserve encompasses 75% of the island. Scientific studies dating from the 1850s have provided an unusually detailed documentation of natural history phenomena, making Lord Howe one of the most intensively studied sites in Australia. Because it is one of the last island biodiversity hotspots on the planet to be discovered and colonized by humans, the original composition of the biota is relatively well known. Successful management efforts to eliminate introduced species and site-based initiatives for conserving native habitats and threatened species place Lord Howe at the forefront of island conservation practice. GEOGRAPHIC SETTING

Lord Howe Island is the largest emergent feature on the western flank of the Lord Howe Rise, at the southern end of a north–south chain of seamounts and guyots extending from southwest of New Caledonia to the Challenger Plateau, west of New Zealand. The island is 702 km northeast of Sydney, Australia, at a latitude of 31° S. The Lord Howe Rise is bounded by the Tasman Sea on the west and the New Caledonia Basin on the east. The rise is no more than 2000 m below sea level, whereas depths in the Tasman Sea between the rise and Australia exceed 4000 m. The rise is underlain by continental crust that separated from eastern Australia during the Cretaceous, moving eastward to its current position during the 30-millionyear-old opening of the Tasman Basin. The Lord Howe World Heritage Site includes offshore islands, islets, and rocks: the Admiralty group to the northeast, Mutton Bird and Sail Rock to the east, Rabbit Island within the lagoon of the Lord Howe reef, Gower Island off the southern end, and Ball’s Pyramid 25 km south of the island. The main island is crescent shaped, measuring 10 km from north to south and approximately 2 km in width.

568

LORD HOWE ISLAND

FIGURE 1 Prominent volcanic features of the southern peaks of Lord

Howe Island. (A) Mt. Lidgbird (left) and Mt. Gower, rising spectacularly out of the ocean to nearly 100 meters and capped by cloud. (B) Layers of horizontal basalt on Mt. Lidgbird, remnants of the final sequence of lavas that filled the ancient caldera after collapse of the Lord Howe volcano. (C) Oblique basalt dikes representing a final swarm of intrusions of magma into the horizontal layers of Mt. Gower.

PHYSICAL FEATURES

Topographically, Lord Howe is dominated by the southern basalt peaks of Mount Lidgbird (777 m) and Mount Gower (875 m), which rise nearly vertically out of the sea forming steep cliffs and slopes covered with dense subtropical rain forest (Fig. 1A). A narrow lowland isthmus separates the southern peaks from an older northern high-

land that forms the shear cliffs of the northern coastline. Only the lowland isthmus is habitable. The two major rock types exposed on Lord Howe are basalt and calcarenite. The majority of the exposed rock is layered basalt, with individual flows ranging in thickness from a few centimeters to more than 30 m (Fig. 1B). The nearly horizontal basalt layers of Mount Lidgbird are part of a sequence that formed near the close of volcanic activity, filling an immense caldera that resulted from collapse of the summit of the original volcano. In addition to basalt, the older sequences of volcanic rock at the north end of the island and in the Admiralty Islands include tuff and volcanic breccia, consolidated fragmental deposits from a more explosive earlier phase of volcanism. The basalt layers on Lord Howe are extensively crosscut by dikes. These features formed when younger basalt was intruded into fissures cutting obliquely across the older basalt layers (Fig. 1C). The low-lying coastal strip of Lord Howe is dominated by calcarenite, a sedimentary rock that normally does not occur on high volcanic islands. Calcarenite consists of calcium carbonate sand, formed by mechanical breakdown of the skeletons of coral, calcareous algae, and shells. The calcarenites of Lord Howe represent major episodes of erosion of the coral reef during ice-age fluctuations of sea level and subsequent formation of beach sands and windblown dunes around the flanks of the old volcano. Cementatation of the deposits and their subsequent erosion has formed many interesting sedimentary features (Fig. 2A). The calcarenite is also notable for its fossil content, which includes both marine and terrestrial biota. Fragments of the bones of seabirds are common (Fig. 2B) along with shells of endemic land snails and large quantities of bone of the extinct giant horned turtle Meiolania platyceps. The British Museum of Natural History and the Australian Museum both contain hundreds of specimens resulting from collections as early as the 1880s. The skull of a new species of extinct endemic bat, an extinct species of penguin, and bird eggs are included in the interesting fossil discoveries. Some of the best-preserved fossil invertebrates are in beach rock exposed at sea level when storms wash away the overlying beach sand (Fig. 2C). GEOLOGIC HISTORY

When the emergent portion of Lord Howe Island formed, 7 million years ago during the Miocene epoch, its subaerial extent was 40 times greater than it is today. The subaerial phase of eruption and building of the shield volcano culminated in a collapse of a caldera estimated to be 900 m deep and 5 km by 2 km. Rapid filling of the caldera

FIGURE 2 Sedimentary rocks and fossils. (A) Outcrop of the Neds

Beach Calcarenite, a sequence of cross-bedded, windblown, calcium carbonate skeletal grains that formed during low stands of Pleistocene sea level and erosion of exposed reef. (B) A fragment of fossil bird bone in place in the Neds Beach Calcarenite. (C) Fossil coral imbedded in beachrock on Neds Beach.

with horizontal basalt layers ended the volcanic phase of island history 6.4 million years ago. The modern landscape is predominantly a consequence of erosion of the original edifice. Very little is known of the larger submarine portion of the island, which is more than 25 km in diameter at its base. The Admiralty Islands and small rocks and islands adjacent to Lord Howe are part of the original shield volcano. Ball’s Pyramid, the dramatic

LORD HOWE ISLAND

569

551-m pinnacle located 25 km southeast of Lord Howe Island, is the remnant of a different shield volcano. It is separated from Lord Howe by depths of approximately 500 m, and the two were never connected by land. Beach sand units and windblown dune units include fossil soil layers. The history and ages of the Lord Howe calcarenites have been studied in detail using five different dating techniques. The cyclical pattern of formation can be linked to periods of sea-level change during successive ice ages. CLIMATE

Lord Howe Island has a humid subtropical climate, with mean summer temperatures of 23 °C and mean winter temperatures of 16 °C. Mean annual rainfall is approximately 165 cm, with higher rainfall during the winter. The peaks of Mt. Lidgbird and Mt. Gower must receive significantly higher precipitation throughout the year. The marine climate is notable for its mix of temperate and tropical water. The tropical East Australia Current flows south along the Great Barrier Reef and into the northern Tasman Sea, but in some years there are strong incursions of cold subantarctic currents from the south. BIOTA Fauna

There are no large terrestrial vertebrates in the native fauna, which is restricted to a skink, a gecko, and a small bat. Of the 15 native land bird species at the time of discovery of Lord Howe, nine are now extinct. Two were eaten to extinction by sailors (white gallinule, whitethroated pigeon) and a third was eliminated by early settlers because it was a crop pest (red-fronted parakeet). Five additional species were eliminated when the black rat reached the island in 1918. Two of the seven remaining native landbirds are endemic (Lord Howe woodhen, Lord Howe island silvereye). The most conspicuous members of the avifauna are seabirds (Fig. 3A). Fourteen species nest on Lord Howe and adjacent islets, including huge colonies of tens of thousands of individuals of flesh-footed shearwaters, sooty terns, and providence petrels. The terrestrial invertebrate fauna is rich in unusual, rare, and endemic species. There are at least 85 endemic species of land snails and a remarkable evolutionary radiation of freshwater hydrobiid snails. More than 100 species of spiders have been identified, and 50% of these are believed to be endemic to the island (Fig. 3B). There is an endemic freshwater shrimp, an endemic freshwater crab, an endemic leech, ten endemic earthworm species, an endemic amphipod, 12 endemic species and one endemic genus of terrestrial isopod, and an endemic cicada. Genus570

LORD HOWE ISLAND

FIGURE 3 Native flora and fauna of Lord Howe Island. (A) A nesting

colony of masked boobies at Mutton Bird Point. (B) A colorful orbweaving spider, part of the large and poorly known fauna of native spiders. (C) Native cloud forest on the slope of Mt. Lidgbird.

level endemism in the insects includes the hemipterin bug Howeria and the cricket Howeta. One of the most unusual insects is the Lord Howe Island phasmid, Dryococelus australis, a giant stick insect reaching lengths of 15 cm. It was common until the arrival of the black rat in 1918 and extinct by 1920. Rediscovery of a small population of the “extinct” phasmid on Ball’s Pyramid in 2001 is an example of the “Lazarus phenomenon.” It has triggered a vigorous debate about conservation options for reappearing species.

The marine fauna of Lord Howe is remarkable for its mix of tropical and temperate species. The reef has attracted attention not only as the most southerly coral reef in the Pacific, but also because it has unusually high coral cover and high algal biomass. The 83 reported species of coral form some unique associations of tropical species. More than 400 species of fish have been reported. As with the rich marine invertebrate fauna, the fish are a unique mix of tropical and temperate species.

of the immense continental crustal block containing the modern emergent islands of New Zealand, Lord Howe, New Caledonia, and Norfolk from Antarctica and Australia. The two main ridges on the block, the western Lord Howe Rise and the eastern Norfolk Ridge, separated 65 million years ago with the opening of the New Caledonia Basin. The probability of preexisting islands, connections, and extinct island biotas is strong, but it lacks preserved geologic evidence.

Flora

HUMAN HISTORY AND CULTURAL HERITAGE

There are 241 native vascular plant species in the Lord Howe Island group, and 105 (44%) are endemic. The richness of the flora can be attributed in part to the variety of habitats. Twenty-five vegetation associations have been recognized. It is unusual for an island as small as Lord Howe to have altitudes supporting a true cloud forest (Fig. 3C). The moss forests at the summits of Mt. Gower and Mt. Lidgbird are rich in orchids as well as mosses. The endemics are not all rare, high-elevation species. There are four endemic species of palm in three endemic genera. Howea forsteriana is the most notable, forming dense lowland forests. Exotic species pose one of the greatest threats to the native flora. There are 230 introduced species, including 17 that are considered noxious weeds. Most of these are restricted to the settlement area, and most have not invaded the indigenous plant communities.

There is no evidence of prehistoric habitation of Lord Howe Island. The earliest recorded sighting of the island was by Henry Lidgbird Ball, in command of HMS Supply bound from Sydney to Norfolk Island in 1788. He landed on his return trip to Sydney, claiming the island for Great Britain and naming several prominent features (Mt. Lidgbird, Ball’s Pyramid) for himself. The island itself he named for the first lord of the British Admiralty, Lord Howe. Although ships in search of food and freshwater visited the island, it was not settled until 45 years after its discovery. Sparse early settlement in 1833 and 1844 was restricted to the lowland and supported by subsistence farming and supplying passing ships. The only “industry” in the latter part of the nineteenth century was the marketing of seeds of the Howea palm, an adaptable and hearty indoor plant that achieved great popularity during the Victorian era. Tourism has been the only other “industry,” initially by steamship service and small rustic guesthouses. Today the number of tourist beds on the island is controlled, and the natural history, beauty, and tranquility are the major attractions. The island is under jurisdiction of the New South Wales Government and is administered locally by the Lord Howe Island Board. There is no private land ownership. Leaseholders must reside on the island, under a management plan for the settlement. The board also manages the Permanent Park Preserve, which encompasses 75% of the land on the island and has its own management plan. Listing of the island group as a world heritage property included 1455 hectares of the main island, offshore islets, and Ball’s Pyramid. In 1988, the New South Wales government created the marine park that expanded the area to 145,000 hectares.

Biogeographic Affinities

The Lord Howe Island biota is a composite of organisms with different geographic affinities. Many endemic elements in the flora have their closest affinities with New Zealand, but there is a mix of tropical and temperate components. Insects show many different patterns. There are beetles whose closest relatives are on New Caledonia and Norfolk Island. The endemic species and genera of Lord Howe crickets also have their closest affinities with crickets on New Caledonia and Norfolk. Four species of caddis flies endemic to Lord Howe are in a genus endemic to Australia, and the endemic muscid flies of both Lord Howe and Norfolk have an Australian origin. The Lord Howe stick insect is closely related to a genus in New Guinea. The land snails also show several different biogeographic patterns. In one family there is a close affinity with Norfolk Island, whereas the Lord Howe Placostylus is closest to a species in New Zealand. Reconstructing the deep history of island biotas of the western Pacific requires understanding 100 million years of plate tectonics events that both created and destroyed islands. These events were set in motion with separation

CONSERVATION

Although Lord Howe Island has not been drastically modified relative to most Pacific islands, the most obvious steps in conservation have focused on eradicating introduced

LORD HOWE ISLAND

571

feral animals. Feral pigs were eliminated in 1995. Feral cats have been eliminated, and goats may be totally eliminated. Rodent eradication assessments have been made, and eliminating rats appears feasible if external funding can be obtained to meet the relatively high costs. Intensive efforts have been made on behalf of a number of endemic species listed as threatened or endangered. The Lord Howe Island woodhen, Tricholimnas sylvistris, had been reduced to three adult pairs by 1980. They were transferred to a captive breeding facility on the island. By the end of three breeding seasons, 57 individuals had been released on the island. By 1992, the population was estimated at 250–300 birds. Eradication of pigs and cats has been critical to the success of the captive breeding program. There is a recovery plan for the Lord Howe Placostylus, a large, critically endangered land snail. It occurs in the fossil calcarenites of the island and is most often encountered today as empty shells (Fig. 4). The recovery plan emphasizes community involvement in conserving suitable habitat for the species. Although Lord Howe Island has not been drastically modified relative to most Pacific islands, the “people pressure” is a constant threat. An Australian voluntary conservation movement views Lord Howe as “a paradise in peril,” and has generated a “management strategy” to afford better protection to its world heritage values.

SEE ALSO THE FOLLOWING ARTICLES

Endemism / Extinction / Fossil Birds / Island Formation / Land Snails / Spiders

AUSTRALIAN ISLAND TERRITORIES

FURTHER READING

The Australian Commonwealth includes several island territories in addition to the islands such as Lord Howe

Brooke, B. P., C. D. Woodroffe, C. V. Murray-Wallace, H. Heijnis, and B. G. Jones. 2003. Quaternary calcarenite stratigraphy on Lord Howe Island, southwestern Pacific Ocean and the record of coastal carbonate deposition. Quaternary Science Reviews 22: 859–880. Francis, M. P. 1993. Checklist of the coastal fishes of Lord Howe, Norfolk, and Kermadec Islands, southwest Pacific Ocean. Pacific Science 47: 136–170. Harriott, V. J., P. L. Harrison, and S. A. Banks. 1995. The coral communities of Lord Howe Island. Marine and Freshwater Research 46: 457–465. Hutton, I. 1986. Lord Howe Island. Australian Capital Territory: Conservation Press. Hutton, I., J. P. Parkes, and A. R. E. Sinclair. 2007. Reassembling island ecosystems: the case of Lord Howe Island. Animal Conservation 10: 22–29. McDougall, I., B. J. J. Embleton, and D. B. Stone. 1981. Origin and evolution of Lord Howe Island, southwest Pacific Ocean. Journal of the Geological Society of Australia 28: 155–176. Miller, B., and K. J. Mullette. 1985. Rehabilitation of an endangered Australian bird: the Lord Howe Island woodhen Tricholimnas sylvestris (schlater). Biological Conservation 34: 55–95. Pickard, J. 1983. Vegetation of Lord Howe Island. Cunninghamia 1: 133–266. Priddel, D., N. Carlile, M. Humphrey, S. Fellenberg, and D. Hiscox. 2003. Rediscovery of the ‘extinct’ Lord Howe Island stick-insect (Dryococelus australis (montrouzier)) (Phasmatoidea) and recommendations for its conservation. Biodiversity and Conservation 12: 1391–1403. Standard, J. C. 1963. Geology of Lord Howe Island. Proceedings of the Royal Society of New South Wales 96: 107–121.

FIGURE 4 Empty shell of the endangered Lord Howe Placostylus

(right) and an unidentified land snail (left).

572

that are under the jurisdiction of mainland territories. Off the east coast there are two island territories. The Coral Sea Islands Territory is a vast complex of uninhabited reefs and atolls northeast of Queensland and the Great Barrier Reef. Norfolk Island is a small, subtropical, volcanic island territory on the Norfolk Ridge, midway between New Zealand and New Caledonia, twice as far from Sydney as Lord Howe. The territory comprises three islands: Norfolk and the small adjacent Phillip and Nepean Islands. Norfolk Island is the only Australian territory with self-governance. Like Lord Howe, Norfolk is the remnant of a submarine volcano, but it differs from Lord Howe in its low topographic relief and considerably younger age (2–3 million years). In contrast to Lord Howe, there is archaeological evidence of late prehistoric Polynesian occupation, although there was no indigenous population at the time of its discovery in 1774 by Captain James Cook. It was more heavily colonized and disturbed following European discovery, and only 5% of the native forest remains intact. The origins of its endemic plants and animals have been of considerable interest to biogeographers.

LORD HOWE ISLAND

M Hasselb orou gh Ba y

MACARONESIA SEE ATLANTIC REGION

rt h No

■ ANARE

Station

Halfmoon Bay 0

2

4

d Hea

6 km

Nuggets Pt

+ Mt Elder

Bauer Bay

MACQUARIE, BIOLOGY

Sandy Bay

Cormorant Pt

Brothers Pt

JENNY SCOTT University of Tasmania, Hobart, Australia

Macquarie Island (54°30′ S, 158°56′ E) is a remote subantarctic island 1500 km south-southeast of Tasmania in the southern Pacific Ocean (Fig. 1). It has luxuriant herbaceous vegetation but no trees or shrubs, and it supports huge concentrations of seabirds and seals. As with all subantarctic islands, the terrestrial ecosystem of Macquarie evolved without mammals. Their introduction by humans has resulted in significant ecological impacts.

Macquarie N

Aurora Pt Green Gorge

Mt Waite + Davis Bay

Island Saddle Pt

Sandell Bay

Mt Hamilton +

ern uth So ce a n O

Cape Star

Carrick Bay

Macquarie Island

South West Pt

r Hu d

BIOLOGICAL SETTING

Macquarie Island’s position just north of the Antarctic Polar Frontal Zone gives it a uniformly cool, wet, and windy climate (3.3–7.0 °C, 920 mm precipitation) with no permanent ice or snow. Its remote location and wholly oceanic origin means that all its native terrestrial flora and fauna arrived by long-distance dispersal, either by wind or sea. The marine environment is dominated by the Macquarie Ridge and associated trenches, with the westward-flowing Antarctic Circumpolar Current passing through gaps in the ridge north and south of the island. The island is 34 km long, up to 5.5 km wide, and

Pa ci Oc fic ea n

ALIA STR AU

Mt Fletcher +

Refuges Walking tracks

Pt

NE ZEA W LA N

D

●

ANTARCTICA

INFO COURTESY: AUSTRALIAN ANTARCTIC DIVISION AUSTRALIAN GEOGRAPHIC CARTOGRAPHIC DIVISION

FIGURE 1 Map of Macquarie Island. Map reproduced with permission

of Australian Geographic Magazine.

almost 13,000 hectares in area. Several small groups of islets lie to the north and south. The main island forms an elongated plateau, with a maximum elevation of 433 m, that is surrounded by steep coastal slopes 100– 250 m in height (Fig. 2). The coastal environment is

573

species have a greater abundance and vigor on Macquarie than anywhere else in their restricted global distribution. Bryophytes (mosses and liverworts) and lichens are abundant and diverse, with over 130 bryophytes and over 150 lichens recorded. More than 127 species of freshwater and terrestrial algae, and over 200 species of macro-fungi, have been reported. There are no known endemic bryophyte species, and endemic levels of the lichen, algal, and macro-fungal taxa are not yet known. Fauna

FIGURE 2 Rugged tussock-covered slopes on the west coast, with

rocky coastline and small sandy beaches. Photograph by J. J. Scott.

dominated by rocky shores with numerous small sandy beaches. MARINE AND TERRESTRIAL LIFE Vegetation

The rocky intertidal zone has a diverse and extensive benthic algal flora (seaweeds), dominated by spectacular growths of the giant Antarctic kelp Durvillea antarctica. At least 103 species of benthic algae have been recorded, over twice the number of terrestrial vascular (flowering) plants recorded (47 species, including four endemics and five alien species, two of which have been eradicated). The major plant communities are fellfield, herbfields, short grassland, mires, and tall tussock grassland. The most visually striking plants are the large tussock grass Poa foliosa (Fig. 3) and the megaherbs Stilbocarpa polaris (Macquarie Island cabbage) and Pleurophyllum hookeri. The latter two

The marine benthic and pelagic invertebrate and fish fauna around Macquarie Island is not well known, but it appears to be typical of other subantarctic islands. The coastal environment of Macquarie Island supports enormous concentrations of marine vertebrates during the breeding season (Figs. 3–4), with an estimated 80,000 seals, mainly southern elephant seals (Mirounga leonina) and small numbers of three fur seal species. At least 27 bird species breed on the island. Around 3.5–4 million seabirds, mainly four species of penguins, congregate along the coasts during summer. The endemic royal penguin (Eudyptes schlegeli) is the most abundant, with an estimated 850,000 breeding pairs (Fig. 4). Four species of albatross and at least 19 species of petrels, prions, and other birds also breed on the island. These include three alien bird species (a fourth species was eradicated). Less than 300 species of terrestrial invertebrates have been reported, mainly of unknown or cosmopolitan distribution. Approximately 10% are believed to be endemic, and there are at least nine alien species. Currently there are three species of alien mammals (European rabbit Oryctolagus cuniculus, ship rat Rattus rattus, and house mouse Mus musculus), and a fourth (cat, Felis catus) was recently eradicated.

FIGURE 3 The south coast during summer breeding season with

black-browed albatross (Thalassarche melanophris) in the foreground,

574

and rockhopper penguins (Eudyptes chrysochome), royal penguins

FIGURE 4 Royal penguins (Eudyptes schlegeli) and southern elephant

(E. schlegeli), and the large tussock grass Poa foliosa in the back-

seals (Mirounga leonina) crowd the beaches during summer. Photo-

ground. Photograph by J. J. Scott.

graph by J. J. Scott.

MACQUARIE, BIOLOGY

HUMAN IMPACTS AND CONSERVATION

Human presence on Macquarie Island initially involved exploitation of seals and penguins, mostly during the eighteenth and nineteenth centuries, followed thereafter by scientific studies, conservation protection, and a limited amount of commercial fishing and tourism. In recognition of its substantial natural values, the island is a UNESCO World Heritage Area and Biosphere Reserve, and a Tasmanian Nature Reserve. In recognition of the island’s marine values and the interconnectedness of its marine and terrestrial environments, the World Heritage Area extends out to 22.2 km from low water mark and the Nature Reserve to 5.6 km, whereas the majority of the 370-km Exclusive Economic Zone around Macquarie Island has been declared a Commonwealth Marine Park covering around 16.2 million hectares. Since the 1960s, natural resource conservation on the island itself has largely involved the management or eradication of mammals (cats, rabbits, and rodents) that were introduced during the nineteenth century and adapted to the subantarctic conditions, resulting in substantial impacts. In the first decade of the twenty-first century, a massive increase in rabbit numbers has caused widespread damage to vegetation through grazing and digging, with associated impacts to erosion and hydrological regimes and flow-through effects to habitats of threatened seabird species. This has prompted the preparation of a detailed eradication plan for all three remaining mammal pest species (rabbits, rats, and mice). Implementation is planned for 2010, after which substantial recovery of the terrestrial ecosystem is expected.

Selkirk, P. M., R. D. Seppelt, and D. R. Selkirk. 1990. Subantarctic Macquarie Island: environment and biology. Cambridge: Cambridge University Press. Terauds, A., and F. Stewart. 2008. Subantarctic wilderness: Macquarie Island. Crows Nest, Australia: Allen and Unwin.

MACQUARIE, GEOLOGY ARJAN DIJKSTRA University of Neuchâtel, Switzerland

Macquarie Island is a unique island of great geological importance, because it is the only locality in the world where a complete section of young ocean crust formed at a spreading center is exposed above sea level. It has thus become a type-locality for ocean crust and has played a major role in the development of theories of seafloor spreading, one of the key processes of plate tectonics. The island was put on the UNESCO World Heritage List in 1997 for this reason. The island also provides a key geological record of a spreading ridge system that became a convergent plate boundary and that will probably develop into a subduction zone over time. GEOLOGICAL SETTING

Macquarie Island is the subaerially exposed summit of a 1500km-long submarine mountain chain, the Macquarie Ridge Complex (Fig. 1). The island measures 34 by 5.5 km, with a

SEE ALSO THE FOLLOWING ARTICLES

Biological Control / Introduced Species / Macquarie, Geology / Marine Protected Areas / Seabirds FURTHER READING

Banks, M. R., and S. J. Smith, eds. 1988. Macquarie Island Symposium, Hobart, May 1987. Symposium proceedings. Papers and Proceedings of the Royal Society of Tasmania 122: 1–318. Environment Australia. 2001. Macquarie Island Marine Park management plan 2001–2008. Commonwealth of Australia. http://www.environment .gov.au/coasts/mpa/publications/pubs/macquarie-plan.pdf Parks and Wildlife Service. 2006. Macquarie Island Nature Reserve and World Heritage Area management plan. Parks and Wildlife Service, Department of Tourism, Arts and the Environment, Hobart, Tasmania, Australia. http://www.parks.tas.gov.au/publications/tech/macquarie/macquarie.pdf Parks and Wildlife Service and Biodiversity Conservation Branch. 2007. Plan for the eradication of rabbits and rodents on subantarctic Macquarie Island. Parks and Wildlife Service, Department of Tourism, Arts and the Environment, Tasmania, and Biodiversity Conservation Branch, Department of Primary Industries and Water, Tasmania, Australia, 1–30, http://www.environment.gov.au/heritage/publications/draft-macquarierabbit-eradication-plan.html

FIGURE 1 Location of Macquarie Island on the Australian-Pacific plate

boundary. Continents in red, continental margins in tan, oceanic crust in green. Blue indicates ocean floor that was created at plate boundary prior to 5 million years ago, when it was a zone of seafloor spreading. Abbreviations: MI, Macquarie Island; MRC, Macquarie ridge complex; NZ, New Zealand. Diagram modified from Sutherland (1995).

MACQUARIE, GEOLOGY

575

maximum elevation of 433 m above sea level. Two groups of small islets also lie on this topographic ridge, namely the Judge and Clerk Islets 14 km to the north and the Bishop and Clerk Islets 34 km to the south of Macquarie Island. The Macquarie Ridge Complex marks the boundary between the Australian and Pacific tectonic plates. The adjacent sea floor lies at 4000 m below sea level, but trenches immediately next to the ridge system are locally as deep as 6500 m. At present, the Macquarie Ridge is a zone of oblique, rightlateral convergence. This plate boundary can be traced to the northeast into the Alpine fault zone of New Zealand and the Tonga–Kermadec subduction zone. The plate boundary at and around Macquarie Island is seismically very active, with several recorded earthquakes of a magnitude greater than 8. The absence of deep (greater than 35-km) earthquakes along the plate boundary southwest of New Zealand shows that subduction has not started yet in the region, although subduction may just be initiating in the Hjort Trench segment. Before 5 million years ago, the plate boundary was divergent (i.e., it was a zone of seafloor spreading). It was during this phase of sea-floor spreading that the rocks on Macquarie Island were formed; biostratigraphic and radiometric dating give ages for the formation of the rocks on the island between 10 and 6 million years ago. ROCKS AND STRUCTURES RELATED TO SEAFLOOR SPREADING

Because of recent tilting and erosion of the rock units, a unique section of the oceanic crust and into the upper part of the mantle is exposed on the northern part of Macquarie Island. Here, a top-to-bottom sequence of basaltic pillow lavas, a sheeted-dike complex, various types of gabbros (Fig. 2), and other associated plutonic rocks is exposed

FIGURE 3 Schematic section through the rock sequence exposed on

the northern part of Macquarie Island, restored to its assumed original orientation in which the sheeted dikes are vertical, and the layering in the lowermost gabbros is horizontal. Simplified from Dijkstra and Cawood (2004).

(Fig. 3). Finally, peridotites, rocks from the upper part of the mantle, are found at the base of the sequence. Such a sequence is characteristic for ophiolite complexes, and it is generally assumed that typical ocean floor consists of a similar rock sequence, as near-identical rocks are also found exposed at modern mid-ocean ridges. However, some volcanic rocks on the island are nearer in composition to volcanics typically found in seamounts formed at some distance from mid-ocean ridges. In contrast to the northern part of the island, the middle and southern parts of the island have only lavas and some sheeted-dike outcrops. Sedimentary rocks on the island include conglomerates, breccias, and sandstones, interlayered with volcanic rocks. These rocks were deposited at the base of early, sea-floor spreading–related fault scarps. UPLIFT AND EROSION

FIGURE 2 Outcrop of layered gabbros representing the lower levels

of the oceanic crust exposed on the northern part of Macquarie Island (Elizabeth and Mary Point).

576

MACQUARIE, GEOLOGY

Oblique convergence of the Pacific and Australian plates, starting 5 million years ago, has created a 50km-wide zone of uplifted oceanic crust, the Macquarie Ridge Complex. Interestingly, the numerous recent fault scarps found on the island itself seem to be produced by normal rather than the reverse fault movements that are typical for convergent plate boundaries, with minor strike slip. These recent faults control the shapes and locations of most landforms, including the lakes on the island. Raised Pleistocene beach deposits at various

levels clearly attest to recent uplift and marine erosion. Dating of paleobeaches has yielded an average uplift rate for the island of 0.8 mm per year and has shown that the island probably emerged above sea level 600,000– 700,000 years ago. Since then, the island has been shaped by periglacial processes during the Pleistocene glacial cycles. SEE ALSO THE FOLLOWING ARTICLES

Lava and Ash / Macquarie, Biology / Oceanic Islands / Plate Tectonics / Seamounts, Geology FURTHER READING

Adamson, D. A., P. M. Selkirk, D. M. Price, N. Ward, and J. M. Selkirk. 1996. Pleistocene uplift and palaeoenvironments of Macquarie Island: evidence from palaeobeaches and sedimentary deposits. Papers and Proceedings of the Royal Society of Tasmania 130: 25–32. Daczko, N. R., S. Mosher, M. F. Coffin, and T. A. Meckel. 2005. Tectonic implications of fault-scarp-derived volcaniclastic deposits on Macquarie Island: sedimentation at a fossil ridge-transform intersection. Geological Society of America Bulletin 117: 18–31. Dijkstra, A. H., and P. A. Cawood. 2004. Base-up growth of ocean crust by multiple phases of magmatism: field evidence from Macquarie Island. Journal of the Geological Society London 161: 739–742. Goscombe, B. D., and J. L. Everard. 2001. Tectonic evolution of Macquarie Island: extensional structures and block rotations in oceanic crust. Journal of Structural Geology 23: 639–673. Varne, R., A. V. Brown, and T. Falloon. 2000. Macquarie Island: its geology, structural history, and the timing and tectonic setting of its N-MORB to E-MORB transition. Geological Society of America Special Paper 349: 301–320. Wertz, K. L., S. Mosher, N. R. Daczko, and M. F. Coffin. 2003. Macquarie Island’s Finch-Langdon fault: A ridge-transform insider-corner structure. Geology 31: 661–664.

nation is the number of new taxa being described each year. As it is further explored, measures of species richness, already extraordinary, increase at a nearly exponential rate. GEOGRAPHY OF THE TERRITORY

Madagascar is located in the western portion of the Indian Ocean and is separated from the African continent by 400 km of the Mozambique Channel (Fig. 1). Madagascar is the fourth largest island in the world, measuring 581,500 km2, which makes it slightly larger than France or California. It is one of the oldest existing islands, with rocks dating back 3200 million years, and it was one of the last large island masses in the world to be colonized by humans—an event currently estimated by archaeological evidence to have taken place about 2300 years ago. This mini-continent is approximately 1600 km long and 600 km at its widest point, spanning the latitudinal range from 12° to 25.5° S; hence, the southern portion of the island falls outside of the Tropic of Capricorn. To the northwest are the Comoros Islands (Grande Comoro, Anjouan, Mohéli, and Mayotte), which are of volcanic origin and are no more than 7 million years old, and to the north is the western edge of the Seychelles archipelago (Aldabra group), which scatters across another 1000 km of ocean to the east. Approximately 850 km to the east of coastal eastern Madagascar are the isolated Mascarene Islands (La Réunion and Mauritius), in the Indian Ocean.

MADAGASCAR STEVEN M. GOODMAN The Field Museum, Chicago, Illinois

As a result of its plant and animal endemism, nearly unparalleled in other portions of the world, and of the notable levels of threat associated with human activities, Madagascar has been designated as one of the priority biodiversity hotspots. This outstanding biological richness is a result of Madagascar’s long isolation, having been separated from Gondwana some 160 million years ago and having had no subsequent land connection. Furthermore, the island has notable topographic and geological complexity, providing different mechanisms for speciation. One remarkable aspect of this island

FIGURE 1 Map of Madagascar and surrounding islands. Aldabra and

Cosmoledo form islands in the western portion of the Seychelles, and the balance of this archipelago continues off the map in an eastnortheast direction. Map by Lucienne Wilmé.

M A DAG AS C A R

577

Given its size, latitudinal breadth, and different oceanic currents, Madagascar shows remarkably different regional climatic regimes. Portions of the extreme northeast receive over 6.5 m of rainfall per year, whereas the extreme southwest, south of Toliara, generally receives less than 0.5 m per year. Furthermore, given the considerable elevation gradient of the island (Fig. 2A)—particularly the north–south oriented eastern mountain chain, which has numerous peaks surpassing 2000 m—overlaid on the latitudinal gradient, temperatures are highly variable. This includes nightly low temperatures well below freezing in the summital zone of Andringitra in the central south to daily highs reaching 45 °C in lowland areas of the west and southwest. As a generalization, for the humid eastern portions of the island, the rainy and warm season is from December to April, and the dry and colder season is from May to October or November. In contrast, the drier southwestern portions of the island have a short and generally unpredictable rainy season from December to February, and the balance of the year is dry. These different climatological patterns are closely associated with different natural vegetation cover (Fig. 2B).

Since the detachment of Madagascar from Gondwana 170–155 million years ago, there has been considerable geological activity on the island, including volcanic eruptions, mountain uplifting, and substantial formation of sedimentary rock. The island can be roughly divided into an eastern two-thirds composed of Precambrian rocks and a western one-third composed of Phanerozoic unmetamorphosed sedimentary rocks. BIOLOGICAL CHARACTERISTICS

There is no other land mass in the world of equivalent size to Madagascar that surpasses its level of species richness and endemism at different taxonomic levels. Certain groups of organisms are considered Gondwana relicts, and others have colonized the island since its separation and undergone extensive speciation, forming distinctive adaptive radiations. These patterns can be found across different groups of plants and animals, and in some cases they rival the diversity of body forms classically cited for adaptive radiations, such as the Galápagos finches (Fig. 3). The humid evergreen forest is located on the eastern side of Madagascar with an extension into the extreme

FIGURE 2 (A) Altitude and (B) vegetation maps of Madagascar. Little of the original natural vegetation of this biodiversity hotspot island

nation remains. Maps by Lucienne Wilmé.

578

M A DAG AS C A R

FIGURE 3 Line drawing of an extraordinary example of an adaptive

radiation on Madagascar, here as illustrated by the vangas (family Vangidae), which are endemic to Madagascar and the Comoro Islands. Note the differing body sizes and bill shapes that function in a similar fashion to different tools (forceps, thin pliers, large pliers, etc.) and allow the different species of this monophyletic group to exploit different food resources. Drawing by John W. Fitzpatrick.

northwest. This zone has the least pronounced dry season on the island and, as a result, supports considerable floristic and associated faunistic diversity. Classically, the humid evergreen forest formation was divided into several vegetational types, from lowland (sea level to around 800 m) to mid-elevation (800 to 1800 m) to mountain (1800 to 2000 m). However, these different formations show a continuous gradient, and the precise divisions associated with this classification are oversimplified. Furthermore, variation in slope and aspect can have a dramatic influence on vegetational structure. In considerable portions of the zone formerly covered with humid evergreen forest, the natural vegetation has been removed by human actions, forming a secondary savanna or pseudosteppe, where regenerating vegetation is dominated by introduced plants and animals (ants, earthworms, rodents, etc.). On the higher mountains in the upper portion of the humid evergreen forest, overlapping with the mountain zone or above, there is a particular vegetational formation known as sclerophyllous forest. This vegetation type is dominated by ericaceous plants of the family Ericaceae, particularly the genera Erica (= ex Philippia), Vaccinium,

and Aguaria, and is generally elfin (2–3-m tall). The sclerophyllous forest zones experience remarkable variation in weather patterns. For example, in mid-August (end of austral winter) in the summital zone of the Andringitra Massif (2600 m), daily temperatures vary from –11 to 32 °C. Given this temperature differential of over 40 °C and the fact that snow has been recorded on this massif, the local biota has numerous adaptations to withstand such extreme temperature vicissitudes. Much of the central portion of Madagascar, generally referred to as the Central Highlands, representing about 40% of the island and defined as the zone above 800 m, has little remaining natural forest cover. The eastern side of the Central Highlands is well defined by an escarpment, whereas the western side is less so, and it generally descends gradually to the sea. The Central Highlands are currently covered by vast areas of open habitat, including a grassland formation dominated by introduced plants. Based on pollen cores from paleontological sites dating from the Late Quaternary, it has been shown that portions of the Central Highlands were not covered by dense forest formations before the recent large-scale habitat transformation of this zone by humans. Hence, the previous hypothesis that Madagascar was solid forest from coast to coast just before human colonization needs to be reconsidered. Rather, the Central Highlands were probably a mosaic of different forest types, including marshlands, dense humid evergreen forests, and open wooded savannas. The vast majority of the former marshlands of this zone were converted into agricultural lands, particularly rice paddies. The vegetation formation along the western slope in the northern half of the island, below the Central Highlands, is dry deciduous forest. In the lowland portions of the extreme southwest and south, the natural vegetation is composed largely of spiny bush (also referred to as Didieraceae–Euphorbia thicket or xerophytic bush). The dry deciduous forest zone can experience annual dry seasons of up to 6 to 8 months, during which a considerable portion of the vegetation drops its leaves, although some plants remain green using an assortment of adaptations to retain moisture. In contrast, the spiny bush region can experience annual dry seasons of up to 10 months, and in some cases, several years can pass without any substantial precipitation. Here, the majority of the plants show adaptations to survive these long periods of drought, and they exhibit a remarkable level of micro-endemism. Other vegetation communities on the island include mangrove (particularly on the western coast), aquatic

M A DAG AS C A R

579

freshwater zones along rivers and inland lakes, and a particular Central Highland vegetation known as tapia forest, dominated by the genus Uapaca. Botanists have been actively working on the flora of Madagascar for several hundred years, and even with this historical foundation, the estimate of the number of vascular plants on the island has increased from about 7400 in 1936 to 12,000 in 1971 to 14,000 in 2006. Madagascar possesses numerous endemic plant families. Although the figures differ between groups from about 40% to 100%, the fauna of the island is largely endemic, including families not found elsewhere in the world. As with plants, a remarkable number of new animal taxa are described each year, running the gamut from invertebrates to mammals. To provide a few examples, the ant fauna of Madagascar contained about 300 species in the 1990s, and it is now estimated that 1000 species occur on the island, approaching 96% endemism. The amphibian fauna was estimated in 1984 to comprise 150 species; by 2008 about 230 have been described, and it is projected that 500 species occur on the island, with a level of endemism approaching 99%. Of the native mammal terrestrial fauna, 100% are endemic and composed of four lineages (carnivorans of the family Eupleridae, tenrecs of the family Tenrecidae, rodents of the family Nesomyidae, and lemurs of the five different families within the order Primata); a considerable number of the over 112 species have been named since 1984. Slightly over 50% of the island’s bird fauna is endemic, including several families and subfamilies restricted to the Madagascar region (including the Comoro Islands). ORIGINS OF BIODIVERSITY

As would be expected given Madagascar’s size, geological history, and isolation from other land masses, the sequence of events that led to the evolution of its remarkably rich and unique biota is complicated, and a portion of these events have their origin deep in time. The island is a fragment of the former supercontinent Gondwana. During the earlier portions of the Mesozoic (the geological period spanning from 245 to 65 million years ago), this large land mass comprised areas now referred to as Africa, South America, Australia, Madagascar, India, Sri Lanka, Antarctica, and portions of the Seychelles. Gondwana started to break apart about 160 million years ago, and the eastern section of this massive land mass (composed of Madagascar, India, Australia, Antarctica, and portions of the Seychelles) started to drift toward the east. Approximately 140 million years ago, the landmass referred to as Indo-Madagascar, composed of Madagascar and the

580

M A DAG AS C A R

Indian subcontinent attached to its eastern flank, was completely separated from the African land mass, hence severing any direct means of plant or animal dispersal from Africa not passing over oceanic waters. Several groups of plants and animals occurring on Madagascar presumably have their origins during the period before the breakup of Gondwana. Examples include the plant Podocarpus, found across portions of the Southern Hemisphere; different groups of invertebrates, such as trapdoor spiders (family Migidae), known from South America, Africa, Madagascar, Australia, New Zealand, and New Caledonia; and several different vertebrates, including the iguanid lizards (family Iguanidae) from the southern portion of the Americas and Madagascar. Hence, when Indo-Madagascar broke off from other portions of this former continent, these groups were already present on Madagascar. The severance of the former distributional range of these organisms, known as a vicariant event, associated with plate tectonics, led to distinct patterns of speciation on Madagascar. Subsequently, about 80 million years ago, India broke away from Madagascar and moved north toward Asia, carrying with it a number of Gondwana relicts. After the Gondwana breakup, different groups, predominantly from Africa and to a lesser extent from Asia, colonized Madagascar by a variety of means. For plants, these means include wind-dispersed pollen, seeds dispersed in the guts of frugivores, buoyant seeds dispersed by sea currents, and sticky or spiny seeds dispersed in the feathers of birds. For numerous species of ferns, for example, spores can be transported in the stratosphere and dispersed over remarkable distances. In the case of animals, the mechanism of dispersal to Madagascar is often cited as rafting on vegetation, aestivating or outright hibernating in floating clumps of vegetation or holes in tree trunks, swimming, or flying directly (in the case of certain insects, bats, and birds). For larger land vertebrates, the process of colonization was logistically more complicated and rare than it was, for example, for certain groups of plants. Recent research on the native land mammals of the island indicate that all four living groups (see above) can be explained by merely four independent colonization events, underlining how exceptional successful colonization was and how such events led to subsequent diversification into different adaptive forms. THREATS TO BIODIVERSITY

Other than New Zealand, Madagascar was the last large island mass in the world to be colonized by humans,

who arrived from southeastern Asia. Based on current archeological information, people colonized Madagascar about 2300 years ago. There is no evidence of Neolithic or Paleolithic cultures on the island. Over the past few thousand years, Madagascar has experienced a considerable number of extinctions of large-bodied land vertebrates, including lemurs, tortoises, elephant birds, and a variety of other remarkable animals. For example, Bibymalagasia is an extinct order of mammals endemic to the island known only from subfossil bones dating from the Holocene. The period some of these animals disappeared roughly coincides with the first colonization of humans on the island. Explanations for the disappearance of these animals range from natural cycles of climatic change and associated shifts in resources, resulting in extinction events, to humans, through hunting or massive habitat modification that pushed these animals beyond the brink. Although this debate is partially unresolved, the best explanation is probably a combination of these two explanations. Radiocarbon dates from the bones of the extinct megafauna indicate that numerous species continued to exist until a few hundred years ago and existed contemporaneously with humans for over two millennia. This would indicate, at least for these animals, that the idea of massive extinction associated with direct human pressure or the introduction of a virulent hyperdisease associated with domestic animals are not tenable hypotheses to explain the disappearance of these animals. However, over the past 1000 years, humans have colonized virtually all areas of Madagascar, and in their wake, there has been considerable habitat modification and destruction. This has reached a level such that less than 8% of the natural habitats existing before human colonization of the island remain today. Numerous forested areas have become isolated and fragmented, others show clear signs of extensive human usage and degradation, and hundreds of thousands of hectares have completely disappeared. Several decades of recent socioeconomic turmoil has considerably exacerbated this situation, giving rise to the “biological crisis” of the island. Major steps have been taken in past decades to conserve the remaining biotopes of the island and the organisms that they contain. Madagascar was the first country in the African region to develop a national environmental action plan, dating from the early 1990s. It is one of the few regional examples of a country where biological data were applied to advance conservation programs. Madagascar has been at the forefront in conceptualizing and implementing numerous new policies into its national conservation programs to safeguard the unique biota of the island and,

at the same time, allow human communities to retain their identities and advance in socioeconomic development. However, these often very innovative programs, even when properly implemented, have still produced few results at the needed level for changing the economic situation of people living near the forest edge or patterns of habitat disturbance. The Malagasy Government has taken a series of more recent steps to rectify and ameliorate the protection of the island’s unique natural patrimony. In September 2003, the President of Madagascar, M. Marc Ravolomanana, declared at the World Parks Congress held in Durban, South Africa, that over the subsequent five years the total protected areas system of Madagascar would be increased threefold, reaching 10% of the island’s surface. Excellent progress is being made with regard to this commitment, and a series of new parks and reserves have and are about to be named; this process will help to protect aspects of the unique biota of Madagascar. Furthermore, several generations of Malagasy conservation biologists have been trained, and they have exceptional capacity to lead their nation. This, combined with programs to ameliorate the island’s economic problems and improve its educational system, hold considerable promise for the future of the unique organisms and ecosystems found on Madagascar. Some of the critical aspects that need to be properly reinforced in the system of protected area management are the training of staff, the means of enforcing laws, and the revamping of the associated judicial system. A variety of introduced organisms pose serious problems as invasive colonizers, particularly in naturally (e.g., cyclone damaged) or human disturbed habitats. These include plants such as Lantana camera and Psidium cattleianum, which form the principal vegetation in certain zones, and animals such as the ant Technomyrmex, the fish Channa, and the rodent Rattus that apparently outcompete or prey upon their endemic counterparts. However, the problems associated with invasive species on Madagascar, although a serious matter, are notably less pervasive than on neighboring islands such as the Comoros and the Mascarenes. With the opening in recent years of Madagascar to foreign investors and companies, a new problem has developed with respect to the exploitation of the island’s considerable mineral resources. There are current plans at various stages to commercially exploit a variety of different sites for ilmenite, nickel, cobalt, and different precious and semi-precious gems. Furthermore, there is currently considerable exploration of Madagascar’s offshore and deep-sea oil reserves, with the intent of exploitation.

M A DAG AS C A R

581

Unfortunately, from a conservation perspective, many of these sites are associated with important tracts of forest or with freshwater habitats, some within, some on the perimeter, and some lying considerable distances from protected areas. The Malagasy Government is currently formulating policies on how these proposed exploitation projects will unfold, and it is hoped that the conservation concerns will not be completely overridden by the means of important economic gain for the country. SEE ALSO THE FOLLOWING ARTICLES

Archaeology / Comoros / Endemism / Lemurs and Tarsiers / Mammal Radiations / Mascarene Islands, Geology / Rafting FURTHER READING

Battistini, R., and G. Richard-Vindard, eds. 1972. Biogeography and ecology of Madagascar. The Hague, Netherlands: W. Junk. Burney, D. A., L. P. Burney, L. R. Godfrey, W. L. Jungers, S. M. Goodman, H. T. Wright, and A. J. T. Jull. 2004. A chronology for late Prehistoric Madagascar. Journal of Human Evolution 47: 25–63. Dewar, R. E., and H. T. Wright. 1993. The culture history of Madagascar. Journal of World Prehistory 7: 417–466. de Wit, M. J. 2003. Madagascar: heads it’s a continent, tails it’s an island. Annual Review of Earth Planetary Science 31: 213–248. Donque, G. 1975. Contribution géographique à l’étude du climat de Madagascar. Tananarive: Nouvelle Imprimerie des Arts Graphiques. Goodman, S. M., ed. 2008. Paysages naturels et biodiversité de Madagascar. Paris: Muséum national d’Histoire naturelle. Goodman, S. M., and J. P. Benstead, eds. 2003. The natural history of Madagascar. Chicago: The University of Chicago Press. Lowry, P. P., G. E. Schatz, and P. B. Phillipson. 1997. The classification of natural and anthropogenic vegetation in Madagascar, in Natural change and human impact in Madagascar. S. M. Goodman and B. D. Patterson, eds. Washington, D.C.: Smithsonian Institution Press, 93–132. Paulian, R. 1961. La zoogéographie de Madagascar et des îles voisines. Volume 13 de Faune de Madagascar. Tananarive: l’Institut de Recherche Scientifique.

LOCATION, ORIGIN, AND CLIMATE

The archipelago of Madeira is located in the Atlantic Ocean between 32 and 33° N and between 16 and 17° W, lying closer to Africa (∼635 km) than to Europe (∼794 km). Despite its size being less than 800 km2, it comprises two inhabited islands, Madeira and Porto Santo; three islets of only 15 km2 known as the Desertas Islands (Ilheu do Chão, Deserta Grande, and Deserta Pequena); and about ten offshore rocks. It is believed that this archipelago originated from a plume of the Earth’s mantle in the Mid-Atlantic Ridge about 70 million years ago. Of all the islands, Porto Santo is the oldest (∼14 million years), whereas Madeira and Desertas are more recent (from more than 4.6 to ∼0.7 million years old). The composition of these islands is mainly basaltic, and there has not been any recent volcanic event in the last 6000 years. The island of Madeira represents almost 93% of the archipelago extension and is rugged and steep, with about 90% of its surface above 500 m. It has a mountain ridge running east–west, reaching 1862 m at it highest point and much lower in its eastern part (below 200 m). The climate of the archipelago is Mediterranean, but because of differences in sun exposure, humidity, and annual mean temperature, on the island of Madeira there is a clear north–south differentiation. Here, the temperatures are mild year-round, between 15 and 22 °C at lower altitudes and between 5 and 15 °C at the highest altitudes. Humidity varies according to altitude, being greater at high and medium altitudes in the forest areas, where there is fog and a persistent cloud cover from 600–800 m up to 1600 m. In contrast, Desertas and Porto Santo are smaller and lower in altitude (less than 520 m) with temperatures similar to those in the warmer parts of Madeira, but they are much drier (less than 400 mm per year). FLORA AND VEGETATION

MADEIRA ARCHIPELAGO DORA AGUIN-POMBO AND MIGUEL A. A. PINHEIRO DE CARVALHO University of Madeira, Portugal

Madeira is a small archipelago of volcanic origin with a highly diversified flora and fauna. A substantial amount of this diversity is harbored in the evergreen laurel forest, a formerly widespread type of vegetation that covered southern Europe and North Africa during the Tertiary period.

582

MADEIRA ARCHIPELAGO

On the basis of flora, the archipelago is biogeographically included in the Macaronesian region and shows affinities to the Mediterranean. In contrast to Desertas and Porto Santo, which are covered mainly with herbaceous vegetation, the flora of Madeira Island, as a result of its dimension, altitude, and orography, is more diverse, showing a marked altitudinal stratification, which is related mainly to temperature. There are four main types of vegetation: coastal vegetation, evergreen dry and wet forest, and upland vegetation. Coastal vegetation is below 300 m and includes a community of herbs and shrubs that have as dominant species Euphorbia piscatoria, Echium nervosum, and Globularia salicina, all endemic to Macaronesian archipelagoes (Fig. 1).

FIGURE 1 Vegetation of Madeira. Eastern part of Madeira (Ponta de

São Lourenço) showing herbaceous vegetation and Matthiola maderensis, an endemic plant species occurring at low altitudes mainly in coastal rocks and cliffs. Photograph by T. Dellinger.

The dry evergreen forest has been much reduced and is typical of lower altitudes with high mean temperatures and low annual precipitation. Apollonias barbujana, Laurus novocanariensis, Myrica faya, and Ilex canariensis are the dominant canopy tree species. The evergreen wet laurel forest, which occupies 20% of the island, is the main type of vegetation. This luxuriant forest grows from 300–800 m to 1400 m in humid areas with mild temperatures, high precipitation, and frequent coastal fogs (Fig. 2) and contains many rare endemic species. Here are found hygrophilous tree species of Lauraceae exclusive to Macaronesia, such as Laurus novocanariensis, Ocotea foetens, and Clethra arborea. At higher altitudes, this forest is replaced mainly by herbaceous plants and shrubs, with Erica arborea being the dominant shrub species. As occurs on other islands, the flora and fauna of this archipelago are species-poor but very interesting in terms of endemic species and taxonomically isolated groups. The vascular flora comprises about 1200 species including native and naturalized plants, of which 10% are endemic (Figs. 3, 4). Most of the endemics are found among trees and shrubs, and fewer among annuals. There are no endemic taxa above genus level in vascular plants, but there are five endemic genera, with three being monospecific, and about 18 genera out of 44 being exclusive to Macaronesia. Ferns and bryophytes are very diverse probably because of the high humidity; they are represented by approximately 75 and 512 species, respectively. Within the flowering plants, some genera such as Argyranthemum (four spp.), Helichrysum (four spp.), and Sinapidendron (five spp.) have diversified prolifically. One of the most remarkable features of vascular plants is the high number of woody endemic species with herbaceous relatives on the mainland, such as the genera Euphorbia

FIGURE 2 Vegetation of Madeira. Wet laurel forest with Euphorbia

mellifera, an endemic woody species characteristic of moist and shady places on Madeira. Photograph by T. Dellinger.

(Fig. 2) and Echium (Fig. 4). Some genera show biogeographic disjunctions with taxa from far territories, such as the tree species Apollonias barbujana, with its closest relatives outside the islands being A. arnottii from southern India, or the genus Picconia, closely related to the genus Notelaea from Australia and Tasmania. In contrast to plants, lichens and fungi have been less studied.

FIGURE 3 Andryala glandulosa, a common endemic species on

Madeira. Photograph by T. Dellinger.

MADEIRA ARCHIPELAGO

583

FIGURE 4 Echium candicans, an endemic rare woody species found

at high altitudes on cliffs in the laurel forest on Madeira. Photograph by T. Dellinger.

However, a checklist of the lichens and lichenicolous fungi of Madeira indicates more than 700 species already. FAUNA

Among all animal species, about 25% are endemic, and all show great affinities to those of the Mediterranean and Europe, with about 4200 species having been reported thus far. Like other oceanic islands, Madeira has a characteristic disharmonic fauna, and this is especially visible in vertebrates. Native vertebrates include around 50 species, including one fish (an eel), five species of mammals (bats), one reptile (a lizard), and about 38 species of known resident birds, of which seabirds are the most diverse. However, there are only seven known endemic species of vertebrates: a lizard (Teira dugessi) (Fig. 5); three bird species including the Zino’s petrel (Pterodroma madeira), the Trocaz pigeon (Columba trocaz), and the Madeira firecrest (Regulus madeirensis); and one bat and two bird species endemic to Madeira and the Canary Islands.

Although invertebrates are still insufficiently known, they are the most diverse, with land snails showing the most remarkable speciation rates and one of the highest rates of endemic speciation per square kilometer of all oceanic islands. About 70% of the approximately 250 recorded species are exclusive, and 18 genera are endemic, probably as the result of no more than 40 colonization events. High levels of single-island endemism and speciation have been predominantly a within-island phenomenon: In some extreme cases such as Discula, 15 endemic taxa occur on an island (Porto Santo and offshore rock) of only 42 km2. Other genera, such as Leiostyla (30 spp.), Actinella (19 spp.), and Amphorella (12 spp.), have also undergone considerable radiation on Madeira. Arthropods, especially millipedes, woodlice, and insects, represent about 87% of all known fauna. Of these, beetles stand out for the largest number of species, with approximately 900. There are 13 endemic genera of arthropods, most of which are monospecific. In contrast to this, some non-endemic genera—especially of beetles, millipedes, and moths—have diversified greatly. Particularly remarkable are the millipedes of the genus Cylindroiulus, which gave rise to the largest number of endemic species (30 spp.) within a single genus, and beetles of the genus Laparocerus (33 spp.). In contrast to terrestrial fauna, marine biodiversity has been less studied and, although diverse, is not so remarkably rich in endemic species. However, the isolation of Madeira from the closest mainland and nearby islands by great depths (greater than 2000 m) and the warm Gulf Stream current that reaches this archipelago are both responsible for its unique fauna. Because of this, the marine fauna has its greatest affinities to the Atlantic Mediterranean areas and to the tropical and subtropical species of the eastern Atlantic, but there are also pantropical and amphiatlantic tropical species. These characteristics also explain the presence of Macaronesian and endemic species, among which are especially remarkable marine molluscs. CURRENT STATUS AND CONSERVATION

FIGURE 5 Teira dugessi, a very common endemic lizard species on

Madeira. Photograph by T. Dellinger.

584

MADEIRA ARCHIPELAGO

After its discovery, the strategic position of the archipelago resulted in its extensive use by early sailors as a point of recharge for food on their way to America. Human population increased considerably after colonization in 1425, and today the number of inhabitants per square kilometer is about 330, the highest rate in Portugal and one of the largest in Europe. In addition to this large resident human population, Madeira receives about 1 million tourists per year. Thus, demographic pressure and tourism, together

with exotic species, agriculture, forest degradation, and erosion, represent major challenges for conservation. During and after colonization several non-native species of vertebrates such as rabbits, rats, mice, ferrets, goats, frogs, fishes, pigs, cats, and geckos were introduced, either purposely or accidentally, by humans, and many have become naturalized. Of these, mammals have been responsible for causing significant modifications to the native flora. Moreover, some, such as the house mouse (Mus musculus domesticus), have undergone considerable evolution during the short period of approximately 500 years since introduction. Indeed, this species represents an extreme case of rapid speciation, in which a single species has given rise to six different chromosomal forms probably because of geographical isolation and fragmentation. The outstanding biodiversity of Madeira has been well known since the time of Wollaston and Darwin, yet much work remains in order to understand the extent and nature of this diversity. However, protected areas have been established for about 40% of the archipelago’s surface in order to preserve the biodiversity of the island. The laurel forest, although it has been considerably reduced, still represents the largest surviving area of this kind in Macaronesia and, because of its size and conservation status, was recognized in 1999 by UNESCO as a World Natural Heritage Site. In addition, two marine reserves have been established on Madeira Island, whereas the entire Desertas Islands, both marine and terrestrial habitats, are considered natural reserves. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Azores / Canary Islands, Biology / Cape Verde Islands / Endemism / Insect Radiations FURTHER READING

Britton-Davidian, J., J. Catalan, M. G. Ramalhinho, G. Ganem, J.-C. Auffray, R. Capela, M. Biscoito, J. B. Searle, and M. L. Mathias. 2000. Environmental genetics: rapid chromosomal evolution in island mice. Nature 403: 158. Carine, M. A., S. J. Russell, A. Santos-Guerra, and J. Francisco-Ortega. 2004. Relationships of the Macaronesian and Mediterranean floras: molecular evidence for multiple colonizations into Macaronesia and back-colonization of the continent in Convolvulus (Convolvulaceae). American Journal of Botany 91: 1070–1085. Cook. L. M. 1996. Habitat, isolation and the evolution of Madeiran landsnails. Biological Journal of the Linnean Society 59: 457–470. Gittenberger, E., D. S. J. Groenenberg, B. Kokshoorn, and R. C. Preece. 2006. Biogeography: molecular trails from hitch-hiking snails. Nature 439: 409–409. Vanderpoorten, A., F. J. Rumsey, and M. A. Carine. 2007. Does Macaronesia exist? Conflicting signal in the bryophyte and pteridophyte floras. American Journal of Botany 94: 625–639.

Wetterer, J. K., X. Espadaler, A. L. Wetterer, D. Aguin-Pombo, and A. M. Franquinho-Aguiar. 2006. Long-term impact of exotic ants on the native ants of Madeira. Ecological Entomology 31: 358–368.

MAKATEA ISLANDS LUCIEN F. MONTAGGIONI University of Provence, Marseille, France

The term makatea, derived from Polynesian words (maka: slingstone; tea: white), relates to tropical Pacific islands possessing emergent (uplifted) limestones, mainly of coral reef origin and dissected by karst. ORIGIN AND TECTONIC EVOLUTION

Makatea islands possess a volcanic basement built by a number of different mechanisms affecting the Earth’s tectonic plates (hotspots, volcanism at or near divergent plate boundaries, arc volcanism at or near convergent plate boundaries). As the islands drowned due to crustal cooling, coral reefs developed around the volcanic cores. Locally, the volcanic pedestals were totally overtopped by reefal deposits. The time of deposition varies from site to site but was usually between the early Miocene and mid Quaternary. The time of uplift also differs from island to island, but it was generally not before the early Quaternary. Mechanisms vary and reflect regional tectonics. On islands close to recent hotspot volcanoes (Makatea island in the Tuamotus), uplift was driven by the overload of the volcano (e.g., Tahiti), forming a bulge on the surrounding crust. By contrast, on Nauru (southwestern Pacific), emergence is attributed to the uplift of sea floor carried up onto a mantle bump by plate motion. Guam (in the Marianas) was uplifted at the crest of the frontal arc formed at the convergence of Pacific and Philippine plates. GEOMORPHIC EVOLUTION

The major topographic features of the islands were acquired when the limestones were subaerially exposed as a result of uplift and/or global sea-level falls. At the time of emergence, two main island types existed: those with a volcanic core surrounded by an annular limestone platform and those high in elevation, composed of limestone capped by extensive plateaus (Fig. 1). Landscapes resulted from differential karst erosion during periods of humid climate. This process promoted the

M A K AT E A I S L A N D S

585

sesses and their mode of evolution. The reef islands are low-lying and small in size, which has generated widespread concern as to their vulnerability to future sea-level and climatic change. LOCATION AND GEOLOGIC STRUCTURE

FIGURE 1 Idealized cross-sections of the borders of makatea island

Located in the northern Indian Ocean, the Maldives archipelago is an 868-km-long network of coral reefs that extends from the northern atoll of Ihavandhippolhu (6°57′ N) to Addu Atoll (0°34′ S), just south of the equator (Fig. 1). The Maldives constitute the central section

types. These islands are commonly fringed by fossil (mid- to late Quaternary) and modern coral reefs. Heights and distances (in meters) are given as indicative measures. (A) Island with a volcanic core surrounded by barrier of limestones. (B) High, atoll-like island, deeply dissected by karst. The volcanic pedestal lies at an unkown depth (presumably several hundreds of meters) below the carbonate pile.

formation of rim ramparts on the margins of the plateaus. Depressions developed coevally within plateau interiors. The final result was upland basin-and-rim structures and coastal cliffs. SEE ALSO THE FOLLOWING ARTICLES

Atolls / French Polynesia, Geology / Pacific Region / Phosphate Islands / Sea-Level Change FURTHER READING

Dickinson, W. R. 2004. Impacts of eustasy and hydro-isostasy on the evolution and landforms of Pacific atolls. Palaeogeography, Palaeoclimatology, Palaeoecology 213: 251–269. Mylroie, J. E., J. W. Jenson, D. Taborosi, J. M. U. Jocson, D. T. Vann, and C. Wexel. 2001. Karst features of Guam in terms of a general model of carbonate island karst. Journal of Cave and Karst Studies 63: 9–22. Nunn, P. D. 1994. Oceanic islands. Oxford: Blackwell Publishers. Stoddart, D. R., C. D Woodroffe, and T. Spencer. 1990. Mauke, Mitiaro and Atiu: geomorphology of makatea islands in the Southern Cooks. Atoll Research Bulletin 341: 1–65. Vacher, H. L., and T. M. Quinn. 1997. Geology and hydrogeology of carbonate islands. Amsterdam: Elsevier Science B.V.

MALDIVES PAUL KENCH The University of Auckland, New Zealand

The Republic of Maldives consists of 21 atolls and four reef platforms that straddle the equator in the northern Indian Ocean. Comprising 2041 reefs and 1190 reef islands, the archipelago is globally unique in the reef structures it pos-

586

MALDIVES

FIGURE 1 Configuration of the Maldives archipelago in the northern

Indian Ocean showing latitudinal gradients in climate, oceanography, and physical atoll characteristics. Direction of arrow indicates an increase in each characteristic.

of a submarine ridge that stretches from the islands of the Lakshadweep (Laccadives) group in the north to the Chagos archipelago in the south. The archipelago consists of a double chain of atolls on either side of an inner sea, tapering to single atolls to the north and south (Fig. 1). The basement rocks of the atoll chain are Eocene volcanics (55 million years old), which are capped with over 2000 m of Tertiary limestones. The atoll system has not formed according to the model of subsidence proposed by Charles Darwin. Rather, carbonate accumulation through the Tertiary has been dominated by lateral progradation from the outer twin atoll chain toward the inner sea. During the Quaternary, vertical reef growth has dominated over lateral progradation. This vertical growth is controlled by oscillations in sea level, with solutional lowering of reef platforms and lagoons during low sealevel stages and vertical accretion at times of higher sea level. CORAL REEFS AND ATOLLS

The Maldives consists of 2041 individual reefs, with a total reef area of 4513 km2. This complex network of reefs is organized into a number of distinct atoll and reef types. Open atolls are the dominant atoll type (16 in total) and are characterized by heavily dissected atoll rims, which in plan form appear as a sequence of individual reef platforms enclosing a central lagoon. Open atolls are large structures, ranging from 290 to 3790 km2 in area, that contain numerous lagoonal reefs and collectively account for 99.5% of all Maldivian reefs. A striking feature of Maldivian open atolls is the presence of faros, ring-shaped coral reefs located within atoll lagoons. At a global scale, faros are scarce, yet they are abundant in the Maldives, where their formation remains a puzzle. There are five closed atolls in the archipelago, which are smaller in area than open atolls and have near-continuous reef platforms enclosing lagoons. Four oceanic reef platforms also occur. These reefs have no lagoon, and islands cover a large proportion of the reef surface. REEF ISLANDS

The archipelago contains 1190 reef islands perched on top of reef surfaces, 200 of which are inhabited. The islands provide the only living space for the Maldivian population of approximately 330,000. Mid-Holocene in age, the reef islands formed during a major phase of deposition 5500–4000 years ago. Composed entirely of carbonate sands and gravels derived from the surrounding reefs, the

islands are typically small and have a mean elevation of less than 1 m above sea level. They are dynamic landforms that exhibit rapid morphological adjustments in response to changing climatic (particularly wave-energy) and sealevel conditions. LATITUDINAL GRADIENTS IN ATOLL CHARACTERISTICS AND PROCESSES

The physical characteristics of open atolls in the Maldives show marked spatial variations along the north–south gradient (Fig. 1). Northern atolls are characterized by a heavily dissected atoll reef rim, numerous lagoonal patch reefs and faros, and moderate lagoon depths (40–50 m). Reef islands are located on the peripheral and lagoonal patch reefs. Toward the south, atolls are characterized by more continuous atoll reef rims, a higher proportion of peripheral reef rim containing islands, deeper lagoons (70–80 m), and fewer lagoonal patch reefs. Latitudinal variations in atoll morphology have been attributed to broad north–south gradients in climate and oceanographic conditions. Annual rainfall reduces from south to north along the archipelago (Fig. 1). Over the longer term, this rainfall gradient has influenced solutional lowering of lagoons during Quaternary glacial periods. The archipelago is subject to monsoon conditions that switch from west to northeast in a predictable fashion and influence wave and current patterns. The intensity of oscillating monsoon conditions increases to the north. In contrast, incident wave energy reduces in magnitude in the northerly direction. This energy gradient controls contemporary coral reef growth and island building processes. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Climate Change / Indian Region / Reef Ecology and Conservation / Sea-Level Change FURTHER READING

Gardiner, J. S. 1903. The fauna and geography of the Maldives and Laccadive archipelagoes. Cambridge: Cambridge University Press. Kench, P. S., and R. W. Brander. 2006. Response of reef island shorelines to seasonal climate oscillations: South Maalhosmadulu atoll, Maldives. Journal of Geophysical Research 111: F01001:1-12. Kench, P. S., R. F. McLean, and S. L. Nichol. 2005. New model of reefisland evolution: Maldives, Indian Ocean. Geology 33: 145–148. Nasser, A., and B. G. Hatcher. 2004. Inventory of the Maldives’ coral reefs using morphometrics generated from Landsat ETM+ imagery. Coral Reefs 23: 161–168. Purdy, E. G., and G. T. Bertram. 1993. Atoll and carbonate platform development in the Maldives, Indian Ocean. AAPG Studies in Geology No. 34. Tulsa, OK: American Association of Petroleum Geologists.

MALDIVES

587

MAMMAL RADIATIONS LAWRENCE R. HEANEY AND STEVEN M. GOODMAN The Field Museum, Chicago, Illinois

Islands that have been isolated since their formation or for very long periods of time often have land mammal faunas that are made up largely or entirely of endemic species, and often these species are members of speciesrich endemic clades. Though members of these endemic clades are each other’s closest relatives, they typically show highly diverse body size, morphology, behavior, and ecology. These constitute classic cases of adaptive radiation, in which local speciation has produced spectacular diversity. LOCATION AND EXTENT OF ISLAND MAMMAL RADIATIONS

On very isolated and small islands, such as those in the central Pacific and southern Atlantic Oceans, native mammals are either absent or very low in diversity. On progressively less isolated and larger islands, especially in warm seas, bats are present in increasing diversity, with moderately rich faunas often present before non-volant mammals make an appearance. Even in some archipelagoes, such as the Galápagos and the Canary Islands, which are relatively near to continents and have a dozen or more islands, non-flying mammals have low total diversity and show evidence of only limited diversification within the archipelago. However, on single large islands that once were parts of continents but have been isolated for tens of millions of years, such as Madagascar; in groups of large, geologically old islands with mixed continental and oceanic origins, such as the Greater Antilles and the islands west of New Guinea; and in large, complex oceanic archipelagoes, such as the Philippines, remarkably diverse endemic faunas are present, the great majority produced by local speciation. Perhaps the best known of the island mammalian radiations occur on Madagascar, with a surface area of 587,000 km2. Here the modern non-flying and native mammal fauna of well over 110 species, all of which are endemic, includes 26 species of rodents (family Nesomyidae), eight carnivorans (family Eupleridae), 31 tenrecs (family Tenrecidae), and over 50 lemurs (five different families). Within each of these different radiations, one can find a range of body forms and sizes, with striking convergence to other mammalian groups found 588

M A M M A L R A D I AT I O N S

elsewhere in the world. This is well exemplified by the tenrecs, which span three orders of magnitude in body mass from about 2.5 to over 2000 g (Fig. 1). Within one of the tenrec genera, Microgale, the most speciose on the island with 22 recognized species, are included, for example, small shrew- or mole-like animals, some of which are semi-fossorial; at least partially arboreal, moderately sized species with partially prehensile tails three times their head and body length; and primarily terrestrial, relatively large carnivorous or omnivorous species with robust dentitions. Furthermore, what is extraordinary about these four large Malagasy mammalian radiations is that this notable endemic diversity can be explained by only four successful and independent colonizations of the island by terrestrial mammals, which in turn indicates how rare successful colonization events were through geological time. Finally, recent biological exploration of Madagascar has revealed that a considerable proportion of its diversity was unknown to science, with, for example, over 30 land mammals being described as new since 1986. (These figures do not include descriptions of new lemur taxa based exclusively on genetic data.)

FIGURE 1 Malagasy tenrecs have diversified into a spectacular radia-

tion. Here several forms are represented (clockwise from upper righthand corner): Setifer setosus, Geogale aurita, Echinops telfairi, and Hemicentetes semispinosus. Reprinted from Goodman and Benstead (2003), with permission from Link Olson.

Less widely known but similarly diverse mammalian radiations have taken place in the Philippines, a group of 7000 islands totaling a land surface area of about 300,000 km2. Most remarkable among these are the rodents in the family Muridae. Although members of this family are generally known as “rats” and “mice,” most of the Philippine species look and behave little like the pest species familiar to most people. One endemic clade includes 15–20 species of cloud rats and their relatives. These nearly

all live in trees; have long, furry, or hairy tails; and feed in the rainforest canopy on tender young leaves, fruits, and/ or seeds. Another endemic clade includes at least 35 species of smaller, short-tailed animals that live mostly in cloud forest and feed on earthworms and other soft-bodied soil invertebrates (Fig. 2). As with Madagascar, recent field studies have documented many previously unknown species; 20 species have been described since 1986, and many others are currently under study. THE ORIGINS OF DIVERSITY

In the cases of Madagascar, the oceanic Philippines, and many other islands, the non-flying mammals colonized from a nearby continent across ocean waters, probably floating on a mass of vegetation or in hollow tree trunks. In some of these cases (e.g., cloud rats and “vermivore mice” from the Philippines), species that are most closely related to one another occur in different parts of the archipelago, and the evidence suggests that much speciation takes places in populations that share a recent common ancestor but have become geographically isolated because of dispersal over water barriers. However, we can also typically see not just one or two, but many species from the same clade living sympatrically on the large island of Luzon. For example, in the high, wet Central Cordillera of Luzon Island, up to five species of cloud rats may live together on one forested hillside; their weights are 15 g, 125 g, 200 g, 1.4 kg, and 2.6 kg, with the smaller members feeding on seeds and small fruit and the larger ones feeding on leaves and buds. Up to seven species of “vermivore mice” may occur on the same hillside, with two species that burrow through the thick layer of humus, two that forage in the leaf litter, one that patrols long trails over the moss-covered ground, and three that forage opportunistically on the ground surface and in the lower portions of trees and shrubs. Similarly, on Madagascar, up to 14 species of lemurs and 17 species of tenrecs may occur in close proximity in the same forest block, and in many cases, the sympatric species are each others’ closest living relatives. In such cases, the speciation most likely took place within the island, with isolation caused by land features such as rivers, mountain ranges (for lowland species), or forest types that are not suitable habitat. These examples illustrate several primary features of mammalian radiations. First, certain species within a given island or archipelago occurring in geographic isolation differentiate genetically, morphologically, and ecologically from one another. In some cases, with the passage of time, changing circumstances reduce or

FIGURE 2 Four genera containing at least 35 species make up the

endemic clade of vermivorous Philippine rodents, with a phylogeny showing the estimated times of divergence since their common ancestor arrived in the Philippines 12–15 million years ago from the Asian mainland. Phylogeny adapted from Jansa et al. (2006). Drawings by V. Simeonovski, copyright Field Museum of Natural History.

remove the geographic isolating barrier, and they are able to occur sympatrically. This process is repeated many times, usually over the passage of several hundred thousand or million years, so that species richness within communities increases over time. The members of a clade that live sympatrically—all descended from a single common ancestral species, and all derived by speciation within the island or archipelago—nearly always differ from each other in conspicuous ways, such as body size, pelage coloration, vocalizations, preferred food or foraging zones, time of day they are active, and so forth. They may continue to use the same kinds of resources as their shared ancestor, but they evolve in response to selection to exploit different portions of the resource base efficiently. The net result is a diverse community of species, each using resources in a unique way (i.e., each with its own niche). Often these communities parallel the extent of mammalian diversity that we might see in a continental area, and the species on the islands superficially resemble those on continental landmasses, but they originated separately. For example, all tenrecs are related to one another,

M A M M A L R A D I AT I O N S

589

but certain species are shrewlike (small Microgale), hedgehog-like (several genera of the subfamily Tenrecinae), molelike (Oryzorictes), otter-like (Limnogale), and mouselike (Geogale). The carnivorans of Madagascar form an endemic radiation, superficially resembling other groups of African and Asian carnivores, such as genets and mongooses. The vermivorous shrew-mice of Luzon (Archboldomys spp.) are very shrewlike, as their English name implies, and the cloud rats of the Philippines are, in some respects, ecological equivalents of the leaf-eating monkeys of continental Asia and Africa, with some similar aspects to their anatomy (e.g., very large cecums for digesting tough leaves) and body size. Such comparisons have led to the concept of convergence, in which species in different places come to resemble one another because of similar selective pressures associated with use of similar resources. Although such comparisons are general and can rarely be made precisely, the concept does call attention to the fact that island mammal radiations produce a great many distinctive species from a single ancestor, and those species fill a wide range of niches, with overall patterns of resource use being similar to those in very distantly related groups of mammals. CONSERVATION OF ISLAND MAMMAL RADIATIONS

A disproportionately large number of mammal species that have become extinct in the last 400 years were island endemics, leading to fears that island mammals may, in general, be vulnerable to extinction. Several lines of evidence indicate that the picture is complex, and no single pattern can be identified. One of the clearest and most extreme cases for examining this issue involves the mammals of the West Indies, which have an exceptionally rich fossil and subfossil history. Of 76 species of Antillean non-flying mammals that existed 20,000 years ago (including insectivores, sloths, primates, and rodents) about 90% have become extinct, including 14% of the 59 species of bats (and 24% of those that still survive have undergone local but not complete extinctions). Thirty-six of the 67 extinct non-flying species (54%) disappeared prior to the arrival of humans on these islands, due to loss of open, grassy habitats (replaced by wetter and more densely forested habitats), reduction in island area (due to the 120 m rise in sea level at the close of the last glacial maximum), and flooding of coastal caves (especially in the case of bats). Hunting and massive habitat disturbance that came with the subsequent arrival of Amerindians about 4500 years ago, followed by Europeans with their attendant house rats and feral dogs,

590

M A M M A L R A D I AT I O N S

cats, goats, and pigs about 500 years ago, and mongooses more recently, were associated with the extinction of another 37 species of non-flying mammals, leaving only nine as survivors today. The picture that emerges from studies in Madagascar is similar in some respects, but different in others. Rich deposits of subfossil lemur bones dating from the Late Quaternary include 17 species that have gone extinct, all larger in body mass than any living taxon. Three different species of dwarf hippos were recovered from these deposits, as well as the enigmatic genus Plesiorycteropus, which is placed in its own order of mammals (Bibymalagasia), and an assortment of other mammals, all of which have vanished. Some of these extinctions are associated with natural climatic change, including shifts in the west and south associated with aridification, and in other cases humans, who colonized the island some 2500 years ago, presumably delivered the coup de grace. The non-native rat Rattus rattus has colonized virtually all of the remaining humid forest zones of the island and has introduced bubonic plague to the native nesomyine rodents, which has had epidemic consequences for the latter. Overall, unlike the case of the Caribbean islands, a great diversity of Malagasy mammals is still extant. Yet a different picture emerges from the Philippines. Several species of large mammals (an elephant, a rhinoceros, and a dwarf water buffalo) are known only as fossils or subfossils; their age and cause of extinction is unknown. The extant fauna is highly diverse, with about 200 land-living species. Recent studies have shown that alien pest species (primarily Rattus) are abundant in agricultural and residential areas but are entirely absent from areas with natural vegetation, where native small mammals are abundant. In disturbed natural vegetation, the alien Rattus are present but uncommon and are replaced by the native species as the natural forest regenerates. The paucity of fossils or subfossils makes the earlier history of extinction very uncertain, but there is no evidence that the extant native species are especially vulnerable to extinction; rather, they seem to persist well in the face of anything short of overwhelming habitat destruction and/ or severe overhunting. Unfortunately, those conditions do occur in parts of the Philippines, and some extinction of mammals, though not yet recorded, seems inevitable. Taken together, these examples indicate that there is a wide range in the susceptibility to extinction of endemic mammals on islands, just as there is on continents. Continued biological inventories of different islands around the world have documented far greater species diversity than previously thought and provide a wealth of new ecological data. This information provides new insights into

understanding the evolutionary history of mammalian radiations, as well as new information on how we may conserve these remarkable fauna. SEE ALSO THE FOLLOWING ARTICLES

Convergence / Endemism / Lemurs and Tarsiers / Madagascar / Philippines, Biology / Radiation Zone / Rodents FURTHER READING

Burney, D. A., L. P. Burney, L. R. Godfrey, W. L. Jungers, S. M. Goodman, H. T. Wright, and A. J. T. Jull. 2004. A chronology for late Prehistoric Madagascar. Journal of Human Evolution 47: 25–63. Goodman, S. M., ed. 2008. Paysages naturels et biodiversité de Madagascar. Paris: Muséum national d’Histoire Naturelle. Goodman, S. M., and J. P. Benstead, eds. 2003. The natural history of Madagascar. Chicago: The University of Chicago Press. Heaney, L. R. 2000. Dynamic disequilibrium: a long-term, large-scale perspective on the equilibrium model of island biogeography. Global Ecology and Biogeography 9: 59–74. Jansa, S., K. Barker, and L. R. Heaney. 2006. Molecular phylogenetics and divergence time estimates for the endemic rodents of the Philippine Islands: evidence from mitochondrial and nuclear gene sequences. Systematic Biology 55: 73–88. Steppan, S., C. Zawadski, and L. R. Heaney. 2003. Molecular phylogeny of the endemic Philippine rodent Apomys and the dynamics of diversification in an oceanic archipelago. Biological Journal of the Linnean Society 80: 699–715. Woods, C. A., and Florence E. Sergile, eds. 2001. Biogeography of the West Indies: patterns and perspectives, 2nd ed. Boca Raton, FL: CRC Press.

MANGROVE ISLANDS

mangroves form a part of the island vegetation. Those mangrove islands consisting only of mangroves commonly occur where sediments have built up sufficiently to form an island at low tide, which has then been colonized by mangroves (Fig. 1). These are known as overwash mangroves as the islands are completely submerged at high tide, and only the mangroves are visible. Such overwash forests are commonly associated with deltas of rivers carrying high terrigenous silt loads (e.g., the Meghna–Brahmaputra delta of Bangladesh or the Fly River delta of Papua New Guinea) or in shallow waters where abundant sediments of biogenic carbonates accumulate (e.g., the overwash mangrove islands of the Florida Keys). Clearly, however, the mangrove vegetation will differ structurally and functionally in the deltaic, terrigenous setting from that of the carbonate setting.

FIGURE 1 Overwash mangroves on carbonate shoals in the Ras

Mohammed National Park in the northern Red Sea, Egypt. These man-

PETER SAENGER Southern Cross University, Lismore, Australia

Mangrove islands are islands composed entirely or partially of mangrove vegetation and comprise microcosms of generally high productivity, high biodiversity, and structurally complex habitats in the nearshore environment. Such mangrove islands are largely confined to the tropics, with their latitudinal limits occurring in those regions where water temperatures never exceed 24 °C throughout the year. In comparison to shoreline mangroves, the vegetation of mangrove islands is subjected to much greater fluctuations in hydrological and meteorological conditions, resulting in its high dynamism and leading to its enhanced susceptibility to drastic change. MANGROVE ISLAND TYPES

Mangrove islands can be subdivided into two broad types: islands consisting only of mangroves and islands where

groves consist of only one species, Avicennia marina, whose pneumatophores form a 20–30-cm-high dense fringe to the stand.

The second type of mangrove islands, where mangroves form a significant but non-exclusive part of the island vegetation, are usually associated with coral reef islands. There mangroves may grow directly over a reef flat or a fossil reefal platform, or in the lee of shingle ramparts, which provide protection from trade winds. Mangrove islands on reef flats are common off the coast of Central America (e.g., Belize) and on many islands of the Pacific (e.g., Palau and Ponape), where low energy conditions and relatively low tidal ranges occur. Although such mangroves are generally more extensive on “high islands”—which are not included here as mangrove islands—they are also common on “low islands” (e.g., the atoll islands of Nui, Vaitupu, and Nanumanga of Tuvalu). On the Great Barrier Reef of Australia, persistent southeasterly trade winds occur during April to November, resulting in shingle ramparts on the windward, southeastern shores of these “low islands” and leeward sand cays with

MANGROVE ISLANDS

591

592

intervening mangroves in the lee of the shingle. These are referred to as “low wooded islands” and occur commonly in northeastern Queensland, north of 15° S. Low wooded islands display a complex range of features, which occupy around 30–50% of the reef flat. The windward shingle banks are often partially colonized by mangrove scrub (e.g., Avicennia marina, Aegialitis annulata, Pemphis acidula) and other halophytes (e.g., Sarcocornia, Sesuvium, Suaeda), and moats form to the leeward of the shingle ramparts, which retain water at low tide and allow micro-atoll growth (commonly of Porites andrewsi) to occur. The micro-atolls grow upward to mean low water level, and provide a convenient platform for the anchoring of drifting propagules and the subsequent growth of mangroves, which ultimately cover the entire moat with a closed canopy mangrove forest. On the Great Barrier Reef, the primary colonizing species is Rhizophora stylosa, although other species, such as Avicennia marina and Lumnitzera racemosa, quickly follow. In the northern Great Barrier Reef, closed canopy mangroves on low wooded islands consist of around 15 species. Because of the small catchment areas of these islands, freshwater availability is restricted, and mangrove communities on low wooded islands are restricted to regions where the annual rainfall exceeds 1200 mm. Low Isles, off Cooktown in North Queensland, is the most studied low wooded island on the Great Barrier Reef, being originally surveyed by the 1928–1929 Great Barrier Reef Expedition, which spent a year surveying the biota of this island. The island and its mangroves have been resurveyed in 1945, 1954, 1965, and 1973, and over that period, the Rhizophora woodland has increased markedly. Some, but not all, other low wooded islands also showed mangrove expansion, and it seems that reef-top topography and the extent of micro-atoll formation appear to be the major factors in the extent of mangrove development. Cyclones are an annual occurrence in this area, and destructive winds can arrest or even reverse mangrove expansion on low wooded islands for considerable periods.

flocculate and are deposited when mixed with seawater. As a result, nutrient-rich sediments are deposited around mangrove pneumatophores and stiltroots on mangrove islands, and luxuriant and diverse mangrove communities rapidly develop. Organic-matter production is generally washed out of the system (strongly outwelling) and replenished by the river-borne nutrient supply. In turn, the outwelling of organic matter trophically supports nearshore food webs, particularly for molluscs, penaeid shrimp, and juvenile fish. By way of contrast, reefal settings bear the brunt of the tides, which are generally full-strength seawater. Only biogenic sediments are available, brought in by daily tides (bidirectional flux), and nutrients are generally deficient. As a consequence, mangroves are slowed in terms of their organic-matter production, nutrient coupling is usually tight, and organic matter tends to accumulate amongst the vegetation. Reefal mangrove islands are susceptible to erosion, particularly around their periphery, and tend to have less luxuriant mangrove vegetation, often wind-pruned and limited in species richness to those species capable of growing in full-strength seawater. The tendency toward erosion, however, may be offset in this setting, as root material does not break down as rapidly as in deltaic systems and may accumulate as mangrove peat, which facilitates vertical accretion and habitat stability. In Central America, in particular, wherever coring of reefal mangrove islands has been undertaken, peat layers of up to 10 m in depth have been found, and carbon dating suggests that the mangrove communities were initiated through rising sea levels around 8000 years ago and have accumulated peat at a rate that allowed them to keep pace with the rising sea levels of the late Holocene. On the Great Barrier Reef, the mangrove communities of low wooded islands, less reliant on peat accumulation than on micro-atoll growth, were initiated around 6000 years ago.

ENVIRONMENTAL SETTINGS OF MANGROVE ISLANDS

ECOLOGICAL SERVICES PROVIDED BY MANGROVE ISLANDS

As outlined above, mangrove islands occur in two contrasting settings: in river mouth deltas and on reefal platforms. Clearly, the deltaic setting is river-dominated, and the reefal setting is tide-dominated. These dominating factors result in significant differences in the structure and functioning of mangrove islands. In the deltaic setting, mangrove islands are flooded by river water as well as by tides, so salinity remains moderate. Silt and nutrients are carried by the river water, but they

Mangrove islands play an important role in stabilizing and protecting shoaling sand and mud flats. Their roots bind the sediments, and their stiltroots and pneumatophores reduce the water velocity around them, leading to the further deposition of sediments. Mangrove islands, particularly those associated with deltas, tend to be strongly outwelling. Organic matter is broken down by various biotic and physical process into small particles, and this detritus supports a range of dependent

MANGROVE ISLANDS

Stoddart, D. R. 1980. Mangroves as successional stages, inner reefs of the northern Great Barrier Reef. Journal of Biogeography 7: 269–284. Woodroffe, C. D. 1987. Pacific Island mangroves: distribution and environmental settings. Pacific Science 41: 166–185.

MARIANAS, BIOLOGY GORDON H. RODDA U. S. Geological Survey, Fort Collins, Colorado

FIGURE 2 Brown pelicans and darters roost in the canopy of Avicennia

germinans on reefal mangrove islands in Quintana Roo, Mexico. Although such roosting aggregations may provide significant nutrient enrichment, it does not offset the physical damage they do to the upper canopy.

nearshore species, including penaeid shrimp (prawns) and detritivorous fish, such as mullet (family Mugilidae) and bream (family Sparidae). Many commercial and subsistence fisheries are focused in and around such mangrove islands. Finally, mangrove islands constitute structurally complex habitats that provide roosting and nesting sites, particularly for bats and seabirds. The surrounding waters, the mangrove vegetation, and the mangrove-associated invertebrates provide a diverse source of food for such nesting and roosting aggregations. On the Great Barrier Reef, around 30 species of seabirds nest on low wooded islands; one species, the Torres Strait pigeon (Myristicivora spilorrhoa), relies on such mangrove islands in its annual migrations between Australia and Papua New Guinea. Although such nesting aggregations in the reefal setting bring much needed nutrients, this comes at a physical cost in that dense aggregations of seabirds damage the mangroves (Fig. 2). SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Coral / Great Barrier Reef Islands, Biology / Hurricanes and Typhoons / Hydrology / Tides FURTHER READING

Cintrón, G., A. E. Lugo, D. J. Pool, and G. Morris. 1978. Mangroves of arid environments in Puerto Rico and adjacent islands. Biotropica 10: 110–121. Hopley, D. 1982. The geomorphology of the Great Barrier Reef: Quaternary development of coral reefs. New York: John Wiley & Sons Inc. Macintyre, I. G., M. A. Toscano, R. G. Lighty, and G. B. Bond. 2004. Holocene history of the Mangrove Islands of Twin Cays, Belize, Central America. Atoll Research Bulletin 510: 1–16. Saenger, P. 2002. Mangrove ecology, silviculture and conservation. Dordrecht, Netherlands: Kluwer Academic Publishers.

The Mariana archipelago is a line of small oceanic islands in the tropical northern Pacific Ocean about 1800 km east of the Philippine Islands. Because of its remoteness from Melanesian and Asian source areas, its small land area, its limited elevational range, and the sparse opportunities for faunal exchange among islands, the native biota of the Mariana Islands is relatively depauperate (species-poor), and native vertebrates are limited to species that can fly (birds, bats) or raft on floating vegetation and withstand contact with seawater (small lizards). Many of these vertebrates have been extirpated, and all but one of the Mariana endemic vertebrates have experienced range reductions consequent to the arrival of humans and human-introduced species. BIOGEOGRAPHY

The Mariana island chain consists of 15 primary islands stretching 920 km from north to south and ranging in size from 541-km2 Guam (13.5° N) to 2-km2 Farallon de Pajaros (20.5° N). Guam is more than four times the size of the next larger Mariana island (Saipan, at 123 km2) and by itself makes up over half the land area of the Marianas. Guam is also the largest island in Micronesia, the cluster of mostly tiny islands (hence the “micro” in Micronesia) stretching more than 5000 km east–west across the vast area of tropical northern Pacific Ocean west of the International Date Line. The Mariana Islands arose from the depths of the ocean, in response to melting of the Pacific plate as it pushed under the Philippine plate. Accordingly, none of the islands has ever been connected to a continental land mass or to its neighboring islands. The southern arc of islands (Guam to Farallon de Medinilla) arose at about the same time, during a land-building period that occurred from 42 to 8 million years ago. Although seismically active, the southern arc has experienced no volcanism in the modern era.

MARIANAS, BIOLOGY

593

The more northerly islands arose about 5 million years ago, with continued volcanism. The northern islands are relatively small (the largest is Pagan, 18.2° N, at 48 km2). Periodic eruptions throughout historic time have eliminated much of the vegetation on Pagan. In addition to being younger and smaller, the northern islands experience prevailing westerly winds, lower rainfall (Pagan receives about 1900 mm/y compared to Guam’s 2400 mm/y), and greater seasonal variation in temperature and rainfall. In general, seasonal variation in rainfall and temperature increases from south to north, with Guam being the most equable (mean temperature: 26.3 °C; annual variation in monthly mean temperature: 1.2 °C; driest month: March [78 mm]; wettest month: September [411 mm]). Unlike the low atolls of eastern Micronesia, which are mostly limited to seashore or strand vegetation, the Mariana Islands are considered “high” islands by virtue of their uplifted volcanic and limestone peaks and plateaus, most (11 of 15) of which have maximum elevations greater than 300 m (965 m on Agrihan, the highest point in the archipelago). Consequently they have extensive interior forests that benefit from high-elevation cloud cover and coolness. Thus, whereas the coastal plants and animals of the Mariana Islands are similar to those found throughout Micronesia, the interior species of the Mariana Islands are generally unique to high islands. However, the islands rarely attain enough elevation for independent life zones to have developed, as they have in Hawaii and many other large Pacific islands, and very few plant or animal species are limited to high elevation. The Mariana Islands are relatively remote; no larger land masses are found within 1500 km. The nearest larger island to Guam is the slightly larger island of Manus, just south of the equator and north of New Guinea, 1740 km south of the Mariana Islands. Furthermore, the tropical source area for species climatically suitable for the Mariana Islands is to the south and west, whereas the prevailing winds and currents are from the northeast. Drift dispersal northward from the south is further inhibited by the powerful cross-currents supplied by the easterly equatorial current and the westerly equatorial countercurrent. Thus, the natural immigration rate of new species to the Marianas is very low, and biotic exchange among the islands is fairly limited. The southern islands are comparatively species-rich, as befits their greater age and larger area. The Mariana island chain well illustrates the ecological factors affecting species colonization and diversification; few species have colonized, and little diversification has occurred. Clusters of remote islands, such as Hawaii, have

594

MARIANAS, BIOLOGY

proven to be incubators of evolutionary diversification if the islands are old enough, are close enough together, and have enough native habitat differences to promote, sustain, and exchange evolutionary novelties. However, most of the Marianas are too young, too far apart, and too limited in habitat diversity to support the dramatic evolutionary radiations seen in the Galápagos or Hawaii. Nonetheless, many Mariana Island populations have been isolated for such a long period that they have diverged into unique species; thus, the Mariana Islands are a hotspot of endemism. ORIGINAL BIOTA OF THE MARIANA ISLANDS

With the exception of extant vertebrates, the biota of the Mariana Islands is not well documented. The earliest vertebrate inventory data come primarily from the subfossil excavations of David Steadman and colleagues on the islands of Rota, Tinian, and Aguijan. Steadman estimates that the prehuman vertebrate fauna of Rota consisted of 30–35 land birds, about ten lizards, one subterranean snake, and 2–3 bats. Originally all of the Mariana Islands probably supported breeding colonies of pelagic seabirds. Presumably, the primordial species richness would have been substantially greater on the much larger island of Guam, for which the prehuman fauna has yet to be excavated. Guam now has about 325 native species of plants, of which about 25% are endemic to the Mariana Islands. About half of the 60 species of land molluscs (snails) are endemic, and an estimated 45% of the native insects are believed endemic. Endemism is lower in lizards (perhaps because of genetic swamping by repeated colonization) and higher in birds (as a result of a relatively high diversification rate). Near-shore marine fish species in the Marianas number around 900, but relatively few appear to be endemic to the Marianas, presumably because of a steady influx of planktonic larvae (which swamp out local adaptation). All of these taxa reached the Marianas primarily from the Caroline Islands to the south, but most Caroline Island lineages arose on the Asian mainland (fish) or in Australasia (other vertebrates and plants) and reached the Caroline Islands by island hopping via Indonesia (especially birds), the Philippine Islands (especially fish), or New Guinea and its environs (especially lizards). HUMAN IMPACT

The Mariana Islands were first settled between 3500 and 4000 years ago, by people who originated in southeastern Asia, island hopped to Indonesia, and sailed eastward to the western Caroline and Mariana Islands. Their seclu-

sion in the isolation of western Micronesia was abruptly terminated by the arrival in 1521 of Ferdinand Magellan and his starving crew. Over the ensuing several centuries (until 1896) the Mariana Islands were considered a possession of Spain, which forcibly converted the islanders to Catholicism through a series of small wars that—along with exotic diseases—largely depopulated the islands. In 1896 the Spanish administration of Guam was terminated by U. S. warships, which neglected to visit the more northerly Marianas, as the U. S. government erroneously believed them to be unpopulated. The U. S. took Guam as a spoil of the Spanish-American war, and Spain sold the Northern Marianas to Germany, which lost the islands to the Japanese during World War I. During World War II, the islands north of Guam were taken from the Japanese and later folded into the U. S. Strategic Trust Territory of the Pacific, a trusteeship sanctioned by the United Nations. The Strategic Trust Territory was dissolved in the 1980s following various plebiscites on self-government, which in the case of the Northern Marianas took the form of a request to be made part of the United States. The Mariana islands north of Guam are today recognized as the “Commonwealth of the Northern Mariana Islands” or CNMI. The CNMI and Guam are each territories of the United States. Mariana Island residents are U. S. citizens, but they cannot vote in U. S. elections, have no direct say in U. S. governance, and do not pay federal taxes. The four large southern Mariana Islands have dense human populations, especially Guam (168,000) and Saipan (62,400). Since the cessation of Japanese colonization of the islands north of Saipan after World War II, the far northern islands have been generally uninhabited. Six of these (Guguan, Asuncion, the three islands of Maug, and Farallon de Pajaros) are permanently preserved as nature reserves under the terms of the CNMI constitution. Most of the bird and mammal species of the Mariana Islands have been progressively eliminated by humans or by species introduced by humans. For example, Rota today has only ten land birds, eight native lizards, one native snake, and one bat. Guam is a more extreme example, having lost all but two of its native land birds. The circumstances of these extirpations warrant a closer look. Prehistoric Extinctions and Introductions

About half of the bird species loss occurred during prehistoric times. Presumably, large, edible, flightless birds were the most vulnerable and disappeared first. For example, Steadman lists six species of undescribed rails eliminated from the southern Marianas. The proximate

cause was probably direct human consumption, as their bones were found burned. On many oceanic islands, rats were an important early pressure on native birds, but this appears to have been less important in the Mariana Islands. The only rat present in the Marianas prior to the modern era was the small Pacific rat (Rattus exulans), which appeared relatively late in the prehistoric period (about 1000 years before present, or more than 2000 years after human colonization), and it was the only mammal introduced prior to Magellan’s arrival in 1521. Historic Introductions

Chickens, cattle, dogs, pigs, cats, deer, and goats were all intentionally introduced soon after Magellan’s discovery. Of these, cats are likely to have taken the largest toll on the native fauna, especially flightless birds; ungulates (deer, pigs, cattle, goats) undoubtedly had the greatest impact on vegetation. Potentially even more destructive, however, were species introduced by accident: rats, ants, mosquitoes, lizards, snakes, and diseases. Early Western observers wrote with great awe about the irruption of rats in the Mariana Islands and the loss of food crops to the rodent plague. Mosquitoes appear to have arrived with Magellan; invertebrate introductions have continued to the present time, and the rate of new species’ arrival has increased each decade. Plant introductions have probably followed the same temporal pattern, with nearly 600 species now recorded as having been introduced to Guam. Vertebrate introductions have been more carefully documented, with six frog species established as of 2007, as well as two or more turtle species, six lizards, one snake, eight birds, and 11 mammals. Impacts of Historical Introductions

Every introduced species has the potential to transform the ecosystem that it invades, but generally only a few species have such a large impact. Introduced plants and herbivores often alter the structure and composition of the vegetation. Predatory species tend to depress the abundance of, or even eliminate, vulnerable prey. Prey species tend to subsidize the abundance of predators, which then become abundant and depress or eliminate native prey species. The Mariana Islands have notable examples of all three types of impacts. ECOSYSTEM TRANSFORMERS

The key vegetation transformers in the Mariana Islands were an introduced legume tree (Leucaena leucocephala,

MARIANAS, BIOLOGY

595

known locally as tangantangan), introduced forestcovering vines (of diverse species), pigs (Sus scrofa), deer (Cervus mariannus), and goats (Capra hircus). The tangantangan tree readily colonizes disturbed sites, often forming dense monotypic stands that persist for decades or perhaps centuries, thereby preempting natural succession to native forest species. Introduced vines also play a major role in inhibiting natural succession, primarily by covering the forest canopy and preventing light penetration to native forest stands attempting to regrow after typhoons. Regeneration is also inhibited by browsing of introduced ungulates. Curiously, the deer species found in the Marianas, although native to the Philippine Islands, was first described scientifically from a specimen taken in the Mariana Islands after the deer was introduced to Guam in 1771—hence the misleading species name mariannus. Regenerating plants are especially nutritive and defenseless, and are thus especially vulnerable to the unnatural browsing from introduced ungulates. Goats have been particularly damaging to native vegetation, especially on uninhabited islands, where the ungulates can sometimes reach densities enabling them to remove virtually all leaves within reach; plans are progressing for goat removal from selected northern islands. INTRODUCED PREDATORS

Many predators have been introduced to the Mariana Islands, of which the most important were the rat (Rattus rattus), shrew (Suncus murinus), and brown tree snake (Boiga irregularis). Although R. rattus is primarily a herbivore, this species climbs trees very well and preys opportunistically on eggs and small animals. Worldwide, R. rattus has a strong negative influence on nesting birds, but data specific to the Mariana Islands are generally lacking. The large, primarily insectivorous shrew (20–45 g) arrived in the 1960s and can depress abundances of small vertebrates such as lizards. The brown tree snake has had a dramatic impact on birds, mammals, and lizards, although fortunately it has colonized only Guam to date. Losses include three pelagic seabirds no longer nesting on Guam (a fourth disappeared for other reasons), and extirpations of ten of 12 species of native forest birds, one or more of the three native bats, and between one and five of the nine to 12 native lizards. Although introduced predators often depress the abundances of their prey, it is unusual for this depression to progress to outright extirpation. It is even more unusual for an introduced predator to extirpate species from more than one vertebrate class.

596

MARIANAS, BIOLOGY

PREY AS SUBSIDIES FOR INTRODUCED PREDATORS

One reason for the snake’s exceptionally severe impact is that native prey species in the Mariana Islands evolved without defenses against a nocturnal arboreal predator similar to the brown tree snake. An additional consideration is that several introduced prey species were present to subsidize the introduced predator when the predator’s abundance would otherwise have declined in response to declining native prey populations. The key introduced prey species were the rat, shrew, and lizard Carlia ailanpalai. All three species have reached spectacular densities in the Mariana Islands, and these abundances have sustained the snake when it would otherwise have declined in response to disappearing native prey species. For example, in spite of the brown tree snake’s suppression of lizard populations on Guam, more than 180 million Carlia ailanpalai lizards are estimated to live on the islands of Guam, Saipan, and Tinian (mean density: 2364/ha). Thus, the primary ecological impact of Carlia’s introduction to the Mariana Islands is that this prey item subsidized populations of the introduced snake and made possible the extinction of several species of alternate prey. CURRENT STATUS OF BIOTA

From the perspective of conserving global biodiversity, the greatest extinction risk is to species with very limited ranges: endemic species. Although the Mariana Islands are relatively depauperate, they are a hotspot of endemism. There are at least 15 endemic vertebrates, primarily birds, that were originally found nowhere beyond the Mariana Islands (Table 1). Endemic island species tend to be especially vulnerable to introduced predators and ecosystem transformers. Both factors have been detrimental to the Mariana Island endemics, most of which are now endangered or extinct. One unlisted endemic bird (the bridled white-eye) still occupies all of its original range. The other unlisted endemic species occupy from 17 to 45% of their original ranges; listed endemic species occupy from 0 to 17%. The same introduced species have affected the nonendemic resident species, although perhaps to a lesser degree. The seabird colonies have been lost from most inhabited islands, with the exception of Rota. Thus, the Mariana Islands exemplify the challenges facing conservation of island biotas: loss of endemic and resident species, introduced predators, ecosystem transformation, and introduced prey as predator subsidies. These problems affect all archipelagoes to a degree, but have affected the Marianas more than most. Although there are various small nature

TABLE 1

Endemic Land Vertebrates of the Mariana Islands Percent Common Name

Species Name

Guam

Rota

Agui

Tini

Emoia slevini

X

X

X

X

Pteropus tokudae

X

Saip

Anat

Range

ESA Status

Sari

Gugu

Alam

Paga

Agri

Asun

Occupied

R

R

R

R

R

R

17

Reptiles

Mariana skink

R

Mammals

Little Mariana fruit bat

0

Extinct

Birds

Bridled white-eye Golden white-eye Guam bridled white-eye Guam flycatcher Guam kingfisher Guam rail Mariana crow Mariana fruit dove Mariana mallard Mariana swiftlet Nightingale reedwarbler Rota whiteeye Tinian monarch

Zosterops saypani Cleptornis marchei Zosterops conspicillatus Myiagra freycineti Todiramphus cinnamominus Gallirallus owstoni Corvus kubaryi Ptilinopus roseicapilla Anas oustaleti Aerodramus bartschi Acrocephalus luscinia Zosterops rotensis Monarcha takatsukasae

X

R

R

R

100

R

X

R

41

X X

X

X X

X

X X

D R

X

X

Extinct

0

Extinct

0

Endangereda

0

Endangereda Endangered

R

R

13 37

X

X

0

R

X

R

16

Endangered

?

X

R

17

Endangered

10

Endangered

R

X D

0

D R

X

R

Extinct

45

note: Distribution, percent of original range still occupied, and status under the Endangered Species Act (ESA). Original range with reference to Steadman (2006); fossil-only taxa excluded. Major islands are represented by the initial four letters of their names and are ordered from south to north (see Fig. 1). R = resident islandwide in suitable habitat; X = extirpated from that island; D = diminished range on that island; ? = uncertain. a Species persists in zoos or experimental populations.

preserves in the Mariana Islands, these have been powerless to protect the fauna from introduced species. SEE ALSO THE FOLLOWING ARTICLES

Introduced Species / Invasion Biology / Land Snails / Marianas, Geology / Rodents / Snakes FURTHER READING

Fritts, T. H., and G. H. Rodda. 1998. The role of introduced species in the degradation of island ecosystems: a case history of Guam. Annual Review of Ecology and Systematics 29: 113–140.

Furey, J., ed. 2006. Island ecology and resource management: Commonwealth of the Northern Mariana Islands. Saipan: Northern Marianas College Press. Jaffe, M. 1994. And no birds sing. New York: Simon and Schuster. Quammen, D. 1996. The song of the dodo. New York: Scribner’s. Rodda, G. H., T. H. Fritts, and D. Chiszar. 1997. The disappearance of Guam’s wildlife: new insights for herpetology, evolutionary ecology, and conservation. BioScience 47: 565–574. Rodda, G. H., Y. Sawai, D. Chiszar, and H. Tanaka, eds. 1999. Problem snake management: the habu and the brown treesnake. Ithaca, NY: Cornell University Press. Steadman, D. W. 2006. Extinction and biogeography of tropical Pacific birds. Chicago: University of Chicago Press.

MARIANAS, BIOLOGY

597

MARIANAS, GEOLOGY FRANK A. TRUSDELL U.S. Geological Survey, Hawaii National Park

The Mariana Islands are the summits of a large volcanic mountain range that stretches 650 km from Guam to Uracas (Farallon de Pajaros; Fig. 1). The subaerial islands are but a small fraction of the mass, estimated at 0.5 to 1.5% of the volcanoes that form the Mariana Arc. GEOLOGIC SETTING AND STRUCTURE

The Northern Mariana volcanic islands form the upper 2–3 km of the East Mariana Ridge, which rises about 2–4 km

above the ocean floor. To the east of the Mariana Ridge is the Mariana Trench, which is, at a depth of nearly 10 km, the deepest in the world. To the west of the Northern Mariana Islands is the Mariana Trough, which is partly filled with young lava flows and volcaniclastic sediment (Fig. 1). The Mariana Trench and the Northern Mariana Islands (East Mariana Ridge) overlie an active subduction zone, where the Pacific plate, moving northwest at about 11 cm/year, passes beneath the Philippine plate, moving westnorthwest at 8.6 cm/year. Beneath the Northern Mariana Islands, earthquake hypocenters at depths of 50–250 km mark the location of the west-dipping subducting slab. Farther west, the slab becomes nearly vertical and extends to a 700-km depth. During the past century, more than 40 earthquakes of magnitude 6.5–8.1 have occurred along this subduction zone. The Mariana Islands form two subparallel, concentric, concave-west arcs (Fig. 1). The southern islands make up the outer arc and extend north from Guam to Farallon de Medinilla. They consist of Eocene to Miocene volcanic rocks and uplifted Tertiary and Quaternary limestone. Converging plates cause uplift of the outer arc volcanoes. The nine northern volcanic islands extend from Anatahan to Uracas and form part of the active inner arc. This inner arc extends south from Anatahan, where some of the volcanoes form seamounts west of the older, outer arc (Fig. 1). Other volcanic seamounts of the active arc formed on top of the East Mariana Ridge. Six volcanoes (Uracas, Asuncion, Agrigan, Mount Pagan, Guguan, and Anatahan) and several seamounts (Ruby, Northwest Rota-1, Ahyi, Supply Reef, and Esmeralda) of the volcanic arc have erupted during the past century. Rock Composition

The southern Mariana Islands comprise a basement of dacitic and andesitic rocks and a cap of marine limestone. The Northern Mariana Islands are made up of volcanic rocks ranging in composition from basalt to dacite. The suites of rocks of the Mariana Islands are transitional between tholeiites and cal-alkaline types. Previous Investigations FIGURE 1 Regional map of the Mariana Arc volcanoes and of the adja-

cent Mariana Trench. The Commonwealth of the Northern Mariana Arc volcanoes extends from Guam in the south to Uracas in the north. Figure courtesy of Susan Merle, Oregon State University NOAA Vents Program.

598

MARIANAS, GEOLOGY

Despite centuries of settlement, few detailed studies have been made of the geology and historic eruptive activity of most Mariana volcanoes. A few eruptions were reported as early as the 1600s by seafaring explorers. A large eruption was recorded from Pagan in 1872–1873, and relatively

minor activity was noted during the early part of the twentieth century (Fig. 2). After World War II, the U. S. military commissioned geologic studies of Pagan, Saipan, Tinian, and Guam. More recently, the geology of Alamagan, North Pagan, and Anatahan were compiled. The large Plinian eruption of Mount Pagan on May 15, 1981, resulted in the evacuation of Pagan residents. Intermittent ejection of mainly phreatic ash and other products continued for another 15 years. The first historical eruption on Anatahan Island began on May 10, 2003, from the east crater of the volcano. Ash emissions continued until September 2005, and gas emissions and low levels of seismicity continue to the present. The eruption was not an immediate threat to human life, but volcanic ash posed a great hazard to aircraft.

Uracus

? ?

Maug ?

Asuncion

?

?

?

?

?

?

?

?

?

?

?

?

?

Agrigan

? ?

Pagan Alamagan

?

Guguan

? ?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

Sarigan Anatahan

? ?

Farallon de Medinilla

?

?

Saipan

?

?

?

Tinian

?

?

Aguijan

?

?

?

Rota

?

?

?

?

?

Guam

0 0,

,0

Soils usually form from chemical and physical breakdown of parent material. Soils on the southern Mariana Islands are a direct result of those processes. Most of the soils are alkaline, due to the calcareous parental material. In the Northern Marianas, most soils are the products of volcanic eruptions and are therefore composed of volcanic ash and tephra, along with organic matter. STRUCTURE AND MORPHOLOGY

Sector Collapse

Geomorphic Shape and Age

Landslides and other large-scale mass movements are common on high volcanic islands, leading to the degradation of the volcanic edifice. In the Northern Mariana Islands, landslides are small and represent localized slope failure of unconsolidated volcanic pyroclasts. On most islands, they do not play as important a role in the deterioration of the Mariana volcanoes as in other ocean island environments, such as Hawai‘i, Tenerife, and the Aleutians.

The young volcanoes have a nearly pristine conical shape and form, but older volcanoes are deeply incised with arroyos and stream valleys. The youngest volcanoes include Uracas, Asuncion, Mount Pagan, South Pagan (Butkan Paliat), North Guguan, and the east crater of Anatahan. Over time, some of the southern Mariana volcanic islands subsided beneath the sea, and coral began

00 0, 10

,0

00

0, 10

00

FIGURE 2 Generalized ages of Mariana Islands, modified after Simkin

and Siebert (1994).

to cover the volcanic rocks with limestone. Currently, all the southern Mariana Islands are being uplifted as a result of the convergence of the Pacific and Philippine plates. Tinian has been uplifted by more than 100 m since the Pleistocene and by an additional 1.8 m during the Holocene. Rota has been uplifted ∼500 m. In all cases, the southern Mariana Islands are now raised volcanic islands with caps of limestone. The limestone grew either as coral growth kept pace with the subsidence rate or during higher stands of the sea. The Northern Marianas range in age from Holocene to Quaternary, whereas the southern islands range in age from 15 to 56 million years. Islands sizes are shown in Fig. 1.

SOILS

00

00

0 00

0

Years B.P.

1,

0, 10

10

,0

00

00

0 00 1,

0 10

The Northern Mariana Islands are dominated by easterly trade winds, with seasonal variations, throughout the year. Winds blow from the east from July to November and from the northeast from December to March. In the summer, the winds are light and variable. The rainy season, from July through November, brings frequent heavy showers. Typhoons and tropical storms are common from July to December. Average annual rainfall is about 180–310 cm. In general, the climate of the Mariana Islands ranges from subtropical in the north to tropical in the south.

10

1

CLIMATE

MARIANAS, GEOLOGY

599

ISLANDS Uracas

Uracas, or Farallon de Pajaros, is a single stratocone, mostly devoid of vegetation. The volcano stands 319 m above sea level, is ∼2 by 1.5 km in size (north–south by east–west), with a surface area of 3.1 km2. The volcano’s subaerial volume is 0.25 km3. The northeast flank has an embayment, semicircular in shape, suggesting flank failure. The lack of vegetation indicates recurring volcanic activity on a 1–3-year frequency. The island’s remote location makes an absolute determination of eruptive frequency difficult. Pyroclastic flow deposits underlie the northern two-thirds of the island. The southern one-third is covered by ‘a‘¯a lava flows. On the southeast and southern coast are remnants of two older scarps of a former volcanic edifice, now destroyed. Resurgent activity (the current Uracas) has replaced the older edifice. The summit crater contains solfataras that fume continuously. Slopes greater than 30° are found on all flanks of the volcano, the northeast embayment, and the west coast. Maug

Maug comprises three islets that were once part of a stratovolcano. Explosive activity destroyed the original edifice. The volcano stands 227 m above sea level and is ∼3 by 3 km (north–south by east–west) in size, with a surface area of 4.2 km2. The volcano’s subaerial volume is 0.1 km3. Bathymetric surveying by the National Oceanic and Atmospheric Administration (NOAA) in 2004 shows a resurgent cone within the lagoon between the islets. Maug is made up of stacked lava flows intercalated with minor pyroclastic deposits. Dikes are common in the interior walls. Slopes in excess of 30° are found on all flanks of the volcano except the southwest, where slopes of less than 5° are present. The coastline is mostly rocky, with one or two pocket beaches. Asuncion

Asuncion is a conical volcano with slopes at the angle of repose. The volcano stands 857 m above sea level and is ∼3.5 by 3 km (north–south by east–west) in size, with a surface area of 11.7 km2. The volcano’s subaerial volume is ∼2.2 km3. The summit of the volcano contains a shallow crater with a spatter cone from the 1906 eruption. Steaming, probably related to residual heat from the 1906 eruption, can be observed on clear days. The coastline is rocky, with no beaches. Slopes in excess of 30° are found on all flanks of the volcano; slopes of less than 5° are uncom-

600

MARIANAS, GEOLOGY

mon and are found only on the southwest flank of the volcano. Volcanics erupted on the island are andesite in composition. Agrihan

Agrihan, the highest-standing stratovolcano and largest (by subaerial volume) in the Northern Mariana Islands, stands 882 m above sea level. The island is ∼10 by 6.5 km (north–south by east–west) in size, with a surface area of 52.7 km2. The volcano’s subaerial volume is ∼15.9 km3. The summit contains a large depression, roughly 1.5 by 1.2 km in diameter, and 380 m deep. A spatter cone and flows from the 1917 eruption cover ∼50% of the crater floor. This large crater implies a local edifice with shallow magma storage within the volcano. The flanks of the volcano are steep (more than 30°), with deep furrows extending radially away from the crater. To the north is a large canyon into which a recent, large ‘a‘¯a flow advanced to form a delta on the coast. Pyroclastic flow deposits mantle most of the interior of the island. Rocks erupted on the island range from basalt to andesite. The southwest coast has several beaches composed of mineral sands; otherwise, the coast is rocky. Pagan

Pagan is an island made up of a string of volcanoes originating from three volcanic centers trending northeast to southwest, distinguishing the island from the rest of the Mariana Islands. A large caldera with a resurgent volcano at its center, Mount Pagan (579 m above sea level) is at the northern end of the island. The central part of the island is an isthmus composed of another volcanic center (the highest point is Togari Mountain at 579 m above sea level) containing a chain of deeply incised volcanoes representing the oldest rocks on the island. Another caldera occurs in the south and includes three volcanic cones. The largest, Bulikan Paliat, is 548 m above sea level and rises 248 m above the caldera floor, which is covered by recent ‘a‘¯a flows. Pagan Island is ∼14 km long (north–south) and varies in width from 1.5 to 6 km (east–west), with a surface area of 64.1 km2. The island’s subaerial volume is ∼7.1 km3. Mount Pagan is the second most active volcano in the Northern Mariana Islands, after Uracas (Fig. 2). The large Plinian eruption in 1981 caused the evacuation of residents, and intermittent ejection of mainly phreatic ash and other products continued for another 15 years. A few eruptions were reported as early as the 1600s; a large eruption was reported in 1872–1873, and relatively minor activity was noted during the 1920s. Mount Pagan most recently

erupted on December 5, 2006, emitting fine ash for four days; elevated activity continued until December 19, 2006. Rocks erupted on the island range from basalt to andesite. Eruptive products include scoria, ‘a‘¯a and p¯ahoehoe flows, tuffs, and other pyroclastic deposits. The roughly circular caldera of north Pagan is ∼5.5 km in diameter and contains a prominent southern wall, with more subdued, partly buried northern and eastern scarps. Preliminary calculations of the volume of the volcanic edifice that occupied the region of the former caldera are between ∼4.7 and 5.8 km3 implying that the caldera-forming eruption was as large as volcano explosivity index (VEI) 5, (the same as the May 18, 1980, eruption of Mount St. Helens). On the western flank of Mount Pagan is a maar, a ∼2 by 2–km depression with a lake on its floor. Satellitic vents define a northwest to southeast trend on the flanks of Mount Pagan. To the northwest is a pair of cones, and on the south-southeast, four more cones are located on the southeast flank of Mount Pagan, with the largest standing 244 m high. The shoreline of north Pagan is rocky on all sides, except for a narrow beach 500 m long, flanked by a fringing reef, on the north coast. The Bandeera Peninsula, on the west coast, is a remnant caldera rim that forms two small bays with mineral sands. The central volcanic region has fringing reefs and organic (coral) sand beaches on the east side. The west-side beaches are a mixture of mineral and coral sands. The southernmost coastline is rocky, as a result of recent eruptions and wave erosion. The central volcanic deposits are deeply eroded and exhibit evidence of local mass wasting and/or landslide events. The oldest rocks on the island, located in this region, are probably Late Quaternary in age. North Pagan has two lakes on its western flank. The inner, Laguna Sanhalom, at 10.8 ha, is 8 m deep and filled with brackish water. The southwest corner of the lake was the former site of a hot spring. Since the 1981 eruption, Laguna Sanhalom decreased in area by 40% because of infilling by lahars and debris flows. The outer lake, Laguna Sanhiyon, covers 12.8 ha, is 12 m deep, and also contains brackish water. South Pagan has a caldera, containing three cones of relatively young age. The caldera measures 3 km (northeast– southwest) by 2 km (northwest–southeast), and ‘a‘¯a flows cover ∼60% of the floor. To the northwest of Bulikan Paliat is another cone, Bulikan Bulifli. This cone is the oldest of the group and contains mud pots and a solfatara field. The southern tip of Pagan Island consists of tuffaceous deposits and lava flows. Slopes of the northern half of Pagan Island are low to moderate (less than 20°), thus making this part of the

island the flattest and most inhabitable of the Northern Marianas. Slopes in excess of 30° are common in the central and southern volcanic regions. The southern caldera is perched 200–300 m above sea level and has low slopes, excluding the volcanic cones. The outermost caldera flanks are steep (greater than 30°). Alamagan

Alamagan stands 744 m above sea level, is ∼4 by 3.5 km (north–south by east–west) in size, and has a surface area of 13.4 km2. Its subaerial volume is ∼2.5 km3. Alamagan has had no eruptions during historical time. The island has a large crater just south of the topographic high, and its steep flanks are deeply furrowed. Near the summit are several steaming areas. The two most recent eruptions produced extensive pyroclastic flow deposits, dated by radiocarbon at 1077 ± 87 and 1410 ± 80 years ago. Pyroclastic flow deposits underlie about two-thirds of Alamagan. The northern coast is composed entirely of a single massive lava flow. Rocks erupted on the island range from basalt to andesite. The shoreline is rocky, but two small beaches can be found—on the northwest and southwest coasts. Both are rocky, composed of boulders, cobbles, and broken rocks. Guguan

Guguan last erupted in 1883 and produced a tephra cone and ‘a‘¯a lava flows on the northern half of the island. The southern half of the island is made of an older edifice, eroded and faulted, composed of interbedded lava flows and pyroclastic deposits. The highest peak, found in the south, stands 287 m above sea level. Guguan is ∼3.0 km by 2.5 km (north–south by east–west) in size, with a surface area of 16.8 km2. The volcano’s subaerial volume is ∼0.6 km3. Rocks erupted on the island range from basalt to andesite. The shoreline is rocky, composed of cobbles and boulders. Sarigan

Sarigan, an old volcano with steep slopes (greater than 30°) as a result of mass wasting and/or erosion on the southeast and southwest flanks, has had no historic eruptions. The volcano stands 538 m above sea level and is ∼3 km by 2 km (north–south by east–west) in size, with a surface area of 6.2 km2. The volcano’s subaerial volume is ∼0.8 km3. The summit of the volcano has a high plateau composed of dense spatter and blocky ‘a‘¯a lava flows. North of the summit is another flat mesa that was the site of a second eruptive center. These two regions are the flattest areas on the island. Sarigan consists of ∼30%

MARIANAS, GEOLOGY

601

lava flows and 70% pyroclastic material. The shoreline is rocky, composed of broken-down flows, cobbles, and boulders.

along the west coast. Slopes in excess of 30° are rare and are found in mountainous regions; most slopes are moderate to low (less than 7°).

Anatahan

Tinian

Anatahan measures 9 km (east–west) by 3.7 km (north– south), with a surface area of 46.2 km2. The highest peak is 787 m above sea level, and the island has a subaerial volume of 9 km3. The summit of the island is marked by an elongated caldera, 5 km by 2 km. Located in the eastern part of the caldera is a pit crater, 1.4 by 1.2 km and ∼200 m deep. Pyroclastic flow deposits mantle about 80% of Anatahan. Prior to the eruption in 2003, Anatahan had active solfataras, springs, and mud pots on the floor of the east crater. Temperatures in the springs and mud pots ranged from a maximum of 98.8 °C to a minimum of 67.4 °C, and the pH ranged from 1.7 to 4.3. The first historical eruption on Anatahan Island began on May 10, 2003, from the pit crater of the volcano. The eruption was preceded by several hours of seismicity. Two and a half hours before the outbreak, the number of earthquakes surged to more than 100 events per hour. Plume heights were 4500 to 13,000 m for the initial phases of the eruption. Prior to 2003, an earthquake swarm in April 1990 resulted in the evacuation of the island. From submarine bathymetry, it appears that Anatahan is a single edifice (Chadwick et al., 2005), as opposed to two coalesced volcanoes, as was previously reported. The flanks of the volcano are steep (greater than 30°), with deep furrows extending away from the highlands. Within the elongated caldera, the slopes are low (less than 10°). The shoreline of Anatahan is rocky, composed of sea cliffs, broken-down flows, blocks, and boulders. A small cobble beach is located on the west coast.

Tinian measures 19 km (north–south) by 3–8 km (east– west) and has a surface area of 110 km2. The highest peak is 187 m above sea level, and the island has a subaerial volume of 6.4 km3. The island is composed of Eocene tuffs and volcanic breccias at its core, mantled by Miocene limestone and Plio-Pleistocene coralline algae and coral limestone, similar to Saipan. Holocene beach and raised reef deposits are also found on Tinian. Limestone covers 99% of the island, which is mostly flat, except for remnant fault scarps. The shoreline is rocky, composed of broken-down flows, limestone cobbles, and boulders.

Saipan

Saipan measures 23 km (north–south) by 2.5–10 km (east–west), with a surface area of 141 km2. The highest peak is 473 m above sea level, and the island has a subaerial volume of 10.3 km3. The island is composed of mid-Eocene tuffs, lava flows, and volcanic breccias at its core, mantled by Miocene limestone and Plio-Pleistocene coralline algae and coral limestone. Ninety-five percent of the island is covered by limestone. The oldest rocks on the island have been dated to 41 million years ago. Fringing reefs are present on Saipan’s east coast, and a barrier reef with a shallow lagoon and coral sand beaches is found 602

MARIANAS, GEOLOGY

Rota

Rota is 16 km (north–south) by 6 km (east–west) in size, with a surface area of 110 km2. The highest peak is 187 m above sea level, and the island has a subaerial volume of 6.4 km3. The geology of Rota has not been systematically studied but appears similar to that of adjacent islands. The island is composed of Eocene tuffs and volcanic breccias at its core, mantled by Miocene limestone and PlioPleistocene coralline algae and coral limestone. Limestone covers more than 90% of the island, which is mostly flat, except for remnant fault scarps. The island is surrounded by a fringing reef. Guam

Guam is 48 km long (north–south) by 13 km wide (east– west), with a surface area of 546.2 km2. The highest peak is 406 m above sea level, and the island has a subaerial volume of 53 km3. Approximately 35% of the island is composed of exposed volcanic rocks, located mostly in the south. The northern portions of the island are exposed limestone of Neogene age. SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Island Arcs / Lava and Ash / Marianas, Biology FURTHER READING

Banks, N. G., R. Y. Koyanagi, J. M. Sinton, and K. T. Honma. 1984. The eruption of Mount Pagan volcano, Mariana Islands, 15 May 1981. Journal of Volcanology and Geothermal Research 22: 225–269. Chadwick, W. W., R. W. Embley Jr., P. D. Johnson, S. G. Merle, S. Ristau, and A. Bobbitt. 2005. The submarine flanks of Anatahan volcano, Commonwealth of the Northern Mariana Island. Journal of Volcanology and Geothermal Research 146: 8–25.

Cloud, P. E., R. G. Schmidt Jr., and H. W. Burke. 1956. Geology of Saipan, Mariana Islands, part 1. General geology. U. S. Geological Survey Professional Paper 280-A. Dickinson, W. R. 2000. Hydro-isostatic and tectonic influences on emergent Holocene paleoshorelines in the Mariana Islands, western Pacific Ocean. Journal of Coastal Research 16: 735–746. Karig, D. E. 1971. Structural history of the Mariana Island arc system. Geological Society of America Bulletin 82: 323–344. Kato, T., Y. Kotake, S. Nakao, J. Beavan, K. Hirahara, M. Okada, M. Hoshiba, O. Kamigaichi, R. B. Feir, P. H. Park, M. D. Gerasimenko, and M. Kasahara. 1998. Initial results from WING, the continuous GPS network in the western Pacific area. Geophysical Research Letters 25: 369–372. Simkin, T., and L. Siebert. 1994. Volcanoes of the world. Tucson, AZ: Geoscience Press, Inc. Simkin, T., R. I. Tilling, P. R. Vogt, S. H. Kirbey, P. Kimberly, and D. B. Stewart. 2006. This dynamic planet: world map of volcanoes, earthquakes, impact craters, and plate tectonics. U. S. Geological Survey Map I-2800, scale 1:12, 500. Tanakadate, H. 1940. Volcanoes in the Mariana Islands in the Japanese mandated South Seas. Bulletin Volcanologique 18: 199–225. Trusdell, F. A., R. B. Moore, and M. K. Sako. 2006. Preliminary geologic map of Mount Pagan volcano, Pagan Island, Commonwealth of the Northern Mariana Islands. U. S. Geological Survey Open-File Report 2006-1386.

MARINE LABORATORIES SEE RESEARCH STATIONS

MARINE LAKES MICHAEL N DAWSON AND LAURA E. MARTIN University of California, Merced

LORI J. BELL AND SHARON PATRIS Coral Reef Research Foundation, Koror, Palau

Marine lakes are bodies of seawater entirely surrounded by land. They come in a great variety of shapes, sizes, and distances from the “mainland” sea and can be described further in terms of their water-column characteristics and biotic complements, which may exhibit differences due, in part, to dissimilar physical connections with the sea. Marine lakes are “habitat islands” that exhibit the biogeographic, ecological, and evolutionary characteristics of “true islands” (with varying degrees of isolation), mainland fragments, or otherwise patchily distributed habitat. DISTRIBUTION AND FORMATION

The distribution of marine lakes is poorly documented. They are mentioned, often incidentally, in literature for

tourists or on wetlands, and they can be found most easily by searching high-resolution topographic maps, aerial photographs, and satellite images. Such sources, with some ground-truthing, clearly indicate over 200 marine lakes, concentrated in four regions worldwide (Fig. 1). These regions are characterized by coastal karst semisubmerged in the sea (Fig. 2). This association has two important implications. First, considering the distribution of karst in maritime areas globally, the true number of marine lakes, and regions with marine lakes, likely far exceeds our modest estimates. Marine lakes thus provide abundant natural experiments in ecology and evolution, which may be used to investigate patterns of biodiversity and biogeography globally. Second, all known marine lakes likely formed as depressions in porous, fissured, karst landscapes flooded by rising sea level after the last glacial maximum (LGM) around 18,000 years ago. Measurements of water depths of modern lakes in Berau, Palau, and Papua demonstrate that none is deeper than 60 m, and most are much shallower. Measurement of downward flux of sediment in a subset of lakes in Palau, using sediment traps and radioisotope analyses of cores, indicate sedimentation rates of 1–7 mm per year. Assuming even the slow compaction rates typical of highly flocculent rich organic sediments, all but the highest sedimentation rates applied to the deepest lakes indicate that the underlying karst is less than 100 m below current sea level and therefore must have been covered in dry valleys, not lakes, during the LGM. Indeed, the age of one lake in Palau has been corroborated at ∼10,000 years by coring and radiocarbon analyses of sediments immediately overlaying the karst bedrock. The modern physical diversity of karstic marine lakes therefore formed during the last two decamillenia. PHYSICAL CHARACTERISTICS

Modern marine lakes range from around 1.5 m to 60 m deep and from about 50 m to 2 km in their maximum horizontal dimension. They are a few to many hundreds of meters from the sea, surrounded by forest-covered ridges and peaks from several (although usually several tens) and occasionally several hundreds of meters high. All have a measurable tidal cycle, the amplitude and timing of which can differ greatly from those occurring in the surrounding ocean, indicating variously restricted physical connections. The lakes closest to the sea may have tunnels that are short enough and wide enough for a person to swim through; the lakes farthest from the sea have no such obvious, direct connections. Partly as a consequence of tidal mixing, lakes range from those

MARINE LAKES

603

FIGURE 1 Global distribution of marine lakes. The distribution and frequency of marine lakes are inferred from peer-reviewed scientific literature,

natural history and dive magazines, satellite images, personal experience, and conservation management publications (including Hamner and Hamner 1998; Donachie et al. 1999; Porter et al. 2001; hsea.unep-wcmc.org/sites/wetlands/; oceancolor.gsfc.nasa.gov/cgi/landsat.pl; GoogleEarth). , tens of marine lakes occur in clusters among drowned karst in at least four regions (Bahamas, Vietnam, Palau, Papua). , marine lakes also occur in smaller numbers in additional locations.

, other enclosed and semi-enclosed ecosystems with marine origins are widespread, numerous, and

may share some of the ecological and evolutionary characteristics of marine lakes. A global total of approximately 200 marine lakes is a conservative minimum estimate due to extremely poor knowledge and underreporting of this class of marine ecosystem.

FIGURE 2 Oblique aerial photograph of Mecherchar Island, Palau, showing elevated karst ridges covered in tropical forest among which are nestled

a handful of marine lakes (an additional four are obscured by the ridges). The largest, in the foreground, is approximately 2 km long. Ongeim‘l Tketau, or “Jellyfish Lake,” is in the background (top right). Photograph courtesy of Patrick L. Colin.

that are physically very similar with the surrounding ocean—“holomictic,” isothermal and well-oxygenated to depth, and having a salinity of 34 practical salinity units (psu). Others, because of damped tidal flux, shelter from wind, high rainfall, high insolation, and the lack of distinct seasons, are physically and chemically very different—“meromictic,” vertically stratified by temperature and salinity, anoxic at depth, and brackish. Thus, marine lakes are variously connected to the ocean, forming a continuum from lagoon-like to highly isolated, and they constitute a wide variety of habitats and corresponding assemblages. Such environmental heterogeneity yields a range of novel selective regimes with the potential to structure communities and modify species over time. BIOLOGICAL CHARACTERISTICS

Marine lakes in Palau are famous for the Mastigias jellyfish that populate them. In one such jellyfish lake, Ongeim’l Tketau, millions of medusae, from the size of a pea to a small soccer ball, migrate a quarter-mile eastward in the morning, turn around at noon, and migrate the quartermile westward back to where they began. The results are spectacular aggregations of many hundreds of medusae per square meter during the late morning and late afternoon (Fig. 3). This migration, like the morphology of this jellyfish (Fig. 4), which is just one of five subspecies each restricted to a single lake in Palau, must have evolved in situ (in ≤ 10,000–15,000 years) after colonization from an ancestral population in the surrounding lagoon. Remarkably, that ancestral population, the most recent common ancestor to all marine lake populations, still exists in the lagoon in Palau, providing a situation rare in its clarity for studying the genotypic and phenotypic evolution of Mastigias. What is true for Mastigias logically should also be true for many of the other tens to hundreds of species that inhabit marine lakes. However, with only preliminary biotic inventories completed in a subset of lakes in Palau, Berau, and West Papua, much remains unknown about these other taxa, including other cnidarians, a few fishes, bivalves and gastropods, echinoderms, green algae, bryozoans, copepods, ascidians, and many sponges. Some species are found in all regions, whereas others are regionalized; some are shared among lakes, and some are not; some are common in the surrounding sea, whereas others are known only from a single individual in a single lake. As many as 10–20% of species and subspecies in marine lakes may be new to science, either because they are endemic to marine lakes or

FIGURE 3 Mastigias swarm in Ongeim’l Tketau, Palau, viewed from

above and below water. The perspectives of these two photographs are perpendicular to each other, with the forest to the right in the lower photograph.

because they are simply very rare elsewhere. Determining their taxonomic status is a necessary precursor to further biogeographic research. Geographic variation in the physical structure and community composition of marine lakes is also accompanied by variation in population dynamics. Because marine lakes are much smaller, and isolated, bodies of seawater, they are expected to amplify the effects of changing weather patterns and climate, relative to larger, and open, marine environments that have greater thermal inertia and mixing. Marine lakes in Palau, at the far western edge of the Western Pacific Warm Pool, the region in which El Niño–Southern Oscillation (ENSO) events initiate, are indeed particularly sensitive to ENSO conditions in the eastern Pacific, extreme events of which may cause the Mastigias population in Ongeim’l Tketau to vary in size by seven orders of magnitude. Interestingly, given the high sedimentation rates in the marine lakes, such physical variation, linked population dynamics—as well as colonization and extirpation dynamics affecting community constitution—should be recorded in great detail.

MARINE LAKES

605

FIGURE 4 The evolution of Mastigias medusae in marine lakes occurs

in parallel within species and archipelagoes, but also remarkably between species separated by many millions of years of evolution and thousands of kilometers of ocean (plus a few hundred meters of land). (A) Palau—the morphotype that occurs in coves within the lagoon (Ngerchaol Island) colonized marine lakes giving rise to, after about 10,000 years, the morphotype now found in Ongeim‘l Tketau. (B) West Papua—the ocean morphotype from Gam Island and a derived marine lake morphotype from southeastern Misool (December 2007); in this image, the top of the “bell” is reflected by the water’s surface. Photograph in right half of (B) courtesy of Precious Planet: Eric Battistoni.

RELEVANCE, THEORY

The analogies between marine lakes and other kinds of islands are remarkable because for much of the last three decades, the majority of marine science literature discounted the existence of marine islands. The example of marine lakes, which provide a particularly clear venue for studying patterns and processes in the ecology and evolution of marine taxa, demonstrate beyond doubt that marine species and marine communities can be influenced by the processes described by ecological and evolutionary theory regularly applied to freshwater and terrestrial systems. Concomitantly, over the last decade, publications reporting geographic isolation have blossomed, and papers on rapid evolution, species-area relationships, and “island rule”

606

MARINE LAKES

evolution have appeared. Such theory may be particularly important in understanding the ecological and evolutionary dynamics of marine resources on island archipelagoes, networks of marine protected areas, and species invasions. Beyond basic tests of island theory such as species– area relationships, colonization–diversity curves, and “the island rule,” the range of studies to which marine lakes are likely to contribute in the future is immense. Deep lakes can be cored to compare trajectories of community assembly, population dynamics (related to colors of environmental variation), colonization, and extirpation dynamics. By analogy, modern lakes of different depths may provide analogues of various stages in the flooding and formation of what are now deep lakes, illuminating biotic changes associated with transitions from holomictic to meromictic lakes. Because lakes are small and easy to sample comprehensively, and because they yield natural subsamples of regional species pools, variation between modern lakes’ physical structures can be clearly related to α-, β-, and γ-diversity tied into comparisons of latitudinal and longitudinal trends in diversity. Thus, correlates (and causal factors) affecting physical connectivity, gene flow, and community similarity—and also those influencing physical heterogeneity, heterozygosity, and species diversity—can be studied, including in a community phylogenetics framework. Such analyses are likely to result in greater synthesis in studies of marine and terrestrial systems. An important question, given this long list of similarities, is whether studies of marine lakes will provide anything novel? The combination of a very high-resolution sediment record of subfossils and associated climate is likely unparalleled in almost every freshwater and terrestrial system. In addition, the on-average larger range size of and strong stabilizing selection on marine taxa in ocean environments means that many marine lakes are colonized independently from the same ancestral population (or equivalent conspecific phenotypes), providing perhaps more possibilities to examine the combined stochastic and deterministic effects of genetic drift, genomic canalization or genetic lines of least resistance, and altered adaptive landscapes than any other system. Moreover, the pattern has been repeated in region after region (Fig. 4). In the case of marine lakes, the tape of life has been run again, and again, and again. SEE ALSO THE FOLLOWING ARTICLES

Climate Change / Freshwater Habitats / Island Rule / Lakes as Islands / Palau / Sea-Level Change / Species–Area Relationship / Tides

Dawson, M. N, and W. M. Hamner. 2005. Rapid evolutionary radiation of marine zooplankton in peripheral environments. Proceedings of the National Academy of Sciences of the USA 102: 9235–9240. Dawson, M. N, and W. M. Hamner. 2008. A biophysical perspective on dispersal and the geography of evolution in marine and terrestrial systems. Journal of the Royal Society Interface 5: 135–150. Donachie, S. et al. 2004. The Hawaiian archipelago: a microbial diversity hotspot. Microbial Ecology 48: 509–520. Hamner, W. M. 1982. Strange world of Palau’s salt lakes. National Geographic 161: 264–282. Hamner, W. M., and P. P. Hamner. 1998. Stratified marine lakes of Palau (Western Caroline Islands). Physical Geography 19: 175–220. Martin, L. E., M. N Dawson, L. J. Bell, and P. L. Colin. 2005. Marine lake ecosystem dynamics illustrate ENSO variation in the tropical western Pacific. Biology Letters 2: 144–147. Porter, J. S., P. E. J. Dyrynda, J. S. Ryland, and G. R. Carvalho. 2001. Morphological and genetic adaptation to a lagoon environment: a case study in the bryozoan genus Alcyonidium. Marine Biology 139: 575–585. Tomascik, T., A. J. Mah, A. Nontji, and M. K. Moosa. 1997. The ecology of Indonesian seas. Part I & II. The Ecology of Indonesia Series, Vol. VII. Singapore: Periplus Editions.

MARINE PROTECTED AREAS ALAN M. FRIEDLANDER University of Hawaii, Honolulu

Marine protected areas (MPAs) are any intertidal or subtidal areas, together with their associated flora, fauna, and historical and cultural features, that have been set aside by law or other effective means to protect part or all of the designated environments. Marine reserves are a more restrictive subset of MPAs and are defined as areas permanently and completely protected from extractive harvest and other major human uses. MPA PRINCIPLES AND THEORY

As a result of overfishing and overall degradation of marine ecosystems, marine protected areas (MPAs) have increasingly been proposed as an ecosystem-based management tool to conserve biodiversity and manage fisheries. Closing certain areas to harvest for periods of time has been practiced for centuries by Pacific Islanders to help sustain healthy populations of marine resources; area closure has more recently come into increased use because of the failure of more “modern” management methods. By protecting populations, habitats, and ecosystems within their borders, MPAs provide a spatial refuge for the entire ecological system they contain and provide a powerful buffer against human uncertainty

and natural variability. In addition to resource management, MPAs also contribute to the long-term livelihoods of island people though the strong cultural and economic connections between islanders and the sea, as well as their interdependence on a healthy marine environment for survival and prosperity. Theory and experience show that populations of exploited species, when protected within MPAs, respond by producing larger and more abundant individuals (Fig. 1). Larger individuals produce exponentially more, and healthier, offspring, and higher population densities improve the likelihood of reproductive success. By increasing reproductive output, MPAs can serve as a source area for larvae that can restock the protected area itself, as well as export larvae to adjacent areas open to fishing. These changes in population structure help to conserve fish stocks within MPA boundaries and provide fisheries benefits outside these protected areas through enhanced reproductive output and adult spillover (Fig. 2). Spillover occurs when high population densities within MPAs result in net movement of individuals into nearby areas (Fig. 3). This density-dependent emigration can enhance adjacent populations but may also diminish the reproductive potential of the MPA itself if the protected area is too small and the home ranges of the organisms extend out into fished areas where they can be caught. MPAs can protect entire marine ecosystems by conserving multiple species and essential habitats such as spawning areas and nursery grounds. Island ecosystems often contain limited numbers of these critical habitats, which are interconnected by the movement of organisms at a variety of spatial and temporal scales. By protecting 1000 Percent increase within MPA

FURTHER READING

800 600 400 200 0 Biomass Numbers

Size

Diversity

FIGURE 1 Species protected inside MPAs have been shown to increase

in total biomass, numbers, size, and diversity over time. Average increases are most pronounced for biomass (413%) and number of individuals (200%). Box plot showing 25th, 50th, and 75th percentiles, with 10th and 90th percentiles as error bars and red bars as means. Adapted from Halpern (2003).

MARINE PROTECTED AREAS

607

ecosystem processes necessary for productive populations; (6) providing a reference point to guide future management decisions; (7) increasing overall catches despite reducing the fishing area; and (8) ensuring future catches against management mistakes. Non-Fisheries Benefits

FIGURE 2 Larger-bodied fishes in higher densities within MPAs pro-

duce exponentially more and healthier offspring that can replenish stocks both inside and outside the protected area. Adults and juveniles may spillover from MPAs into adjacent areas as a result of densitydependent effects.

Boundary

Fish abundance

Partial Reduction

MPA

Fished Area

FIGURE 3 Spillover from MPAs to adjacent areas may result from

emigration of fishes from a reserve due to increased competition for resources or other density-dependent mechanisms (black arrows). Shading represents partial reduction in abundance of fishes from inside the MPA that occurs when home ranges extend beyond the MPA boundaries and fish are caught. Adapted from Abesamis et al (2006).

the diversity of habitats and interactions necessary for proper ecosystem function, MPAs represent a holistic ecosystem-based approach to management. BENEFITS OF MPAS Fisheries Benefits

From a fisheries standpoint, MPAs may be used for a number of purposes. Goals include (1) protecting genetic diversity, size distributions, sex ratios, or other stock characteristics; (2) reducing bycatch impacts on vulnerable species in multispecies assemblages; (3) rebuilding overfished stocks; (4) maintaining habitat characteristics necessary for productive populations; (5) maintaining

608

MARINE PROTECTED AREAS

MPAs also have many non-fisheries benefits, such as protecting biodiversity and ecosystem structure, serving as biological reference areas, providing nonconsumptive recreational activities, and maintaining other ecosystem services such as shoreline protection, nutrient cycling, climate control, and so forth. Nonconsumptive access to protected areas includes enhanced economic opportunities, diversified social activities, and increased public awareness. Because of the importance of nearshore ecosystems to islanders, MPAs can also provide a mechanism for cultural maintenance and revival through increased food and economic security. With species loss in the sea accelerating, the irreplaceability of these species makes MPAs a powerful tool for marine conservation by protecting species and their associated habitats. The long-term decline in marine ecosystem health has led to the “shifting baseline syndrome,” where there are no truly natural places left to compare against current conditions. The establishment of MPAs provides an unparalleled opportunity to study marine ecosystems and to better understand ecosystem function in the absence of fishing pressure and other major human impacts. MPA DESIGN CRITERIA

Results suggest that in order to help sustain fisheries, MPAs should cover from 10 to 20% of an area, whereas 30 to 50% protection may be necessary to ensure high long-term fisheries catch levels. The size and shape of individual MPAs can have important effects on ecological and socioeconomic performance. Individual areas need to be large enough to contain the short-distancedispersing larvae (∼1 kilometer) and spaced far enough apart so that long-distance-dispersing larvae (tens to hundreds of kilometers) released from one MPA can settle in adjacent ones. Whether the goals are to enhance fishing or to conserve natural ecosystems, it is desirable to design MPAs so that most adults remain inside whereas some of their reproduction flows out. There is no ideal shape for protecting species and their associated habitats, although swaths stretching from shore into deep water are more likely to contain a diversity of habitats than reserves without as much depth range, and

120°30'W

120°20'W

120°10'W

120°W

119°50'W

119°40'W

119°30'W

119°20'W

119°10'W

34°30'N

34°30'N

120°40'W

Pt. Conception

34°20'N

34°20'N

Santa Barbara

Ventura

34°N

34°N

34°10'N

34°10'N

Santa Barbara Channel

Santa Cruz

33°50'N

33°50'N

Santa Rosa

E

W

Marine Reserve

Channel Islands National Park

1 Marine Conservation Area 2 Marine Conservation Area

State Waters (Mean High Water to 3 nm) Channel Islands National Marine Sanctuary

120°30'W

120°20'W

120°10'W

120°W

119°50'W

119°40'W

5

10

119°30'W

20 Nautical Miles

119°20'W

33°30'N

S

0

120°40'W

33°40'N

N

Marine Protected Areas Network

33°30'N

33°40'N

Santa Barbara (inset)

119°10'W

FIGURE 4 Example of a network of MPAs varying in size, use, and level of protection. Channel Islands National Marine Sanctuary boundaries

showing the location of existing state and proposed federal marine reserves (no-take) and marine conservation areas that allow limited harvest of lobster and pelagic finfish (1, recreational harvest only; 2, recreational harvest of finfish and both recreational and commercial harvest of lobster). The total MPA network covers about 22% of the sanctuary. Image courtesy of NOAA/Channel Islands National Marine Sanctuary.

they may also encompass common natural migration pathways from shallower nearshore to deeper habitats. Individual MPAs need to be networked in order to provide large-scale ecosystem benefits (Fig. 4). An MPA network consists of a series of protected areas that are connected by larval dispersal or juvenile and adult movement. MPA networks have the greatest chance of protecting all species, life stages, and ecological linkages if they encompass representative portions of all ecologically relevant habitat types in a replicated manner. For fisheries purposes, many small reserves in a network may be preferred because of the higher rates of juvenile and adult spillover and more regional benefits through greater larval export than from fewer, larger areas. However, a smaller number of larger MPAs that limit fishing and preserve a greater amount of habitat will provide more benefits for biodiversity conservation. Enforcement and compliance will be greatly aided if reserve borders are straight lines or utilize other obvious navigational reference points. SOCIOECONOMIC FACTORS

The traditional ecological knowledge held by many island peoples is critical to the development and design of MPAs. Traditional customary management systems

have included various forms of area protection, and incorporating elements of these established and recognized practices into a contemporary framework can increase the legitimacy of decisions regarding MPAs, as well as aid in compliance with regulations. Locally managed marine areas that incorporate traditional concepts of customary marine tenure have been effective on many Pacific islands, and participatory community approaches in other parts of the world have also proven to be effective means by which to involve stakeholders in the process, and therefore achieve the intended benefits established for MPAs. SEE ALSO THE FOLLOWING ARTICLES

Fish Stocks/Overfishing / Refugia / Sustainability FURTHER READING

Abesamis, R. A., G. R. Russ, and A. C. Alcala. 2006. Gradients of abundance of fish across no-take marine reserve boundaries: evidence from Philippine coral reefs. Aquatic Conservation: Marine and Freshwater Ecosystems 16: 349–371. Halpern, B. S. 2003. The impact of marine reserves: do reserves work and does reserve size matter? Ecological Applications 13: S117–S137. Palumbi, S. R. 2004. Marine reserves and ocean neighborhoods: the spatial scale of marine populations and their management. Annual Review of Environment and Resources 29: 31–68.

MARINE PROTECTED AREAS

609

Roberts, C. M. 2005. Marine protected areas and biodiversity conservation, in Marine conservation biology: the science of maintaining the sea’s biodiversity. E. Norse and L. Crowder, eds. Washington, DC: Island Press, 265–279. Russ, G. R. 2002. Yet another review of marine reserves as reef fisheries management tools, in Coral reef fishes: dynamics and diversity in a complex ecosystem. P. F. Sale, ed. San Diego, CA: Academic Press, 421–443. Sladek Nowles, J., and A. M. Friedlander. 2005. Marine reserve design and function for fisheries management, in Marine conservation biology: the science of maintaining the sea’s biodiversity. E. Norse and L. Crowder, eds. Washington, DC: Island Press, 280–301.

FIGURE 2 Sandy beach along the ocean shore of Eneu, Bikini atoll.

MARQUESAS ISLANDS

shrublands, with a relatively limited diversity of plant and animal species. All the atolls and islands are low in elevation. Some of the northern atolls were used for nuclear tests, the impact of which is still being studied.

SEE PACIFIC REGION

MARSHALL ISLANDS

GEOLOGY NANCY VANDER VELDE Majuro, Marshall Islands

The loosely strung double archipelagoes of Ratak and R¯alik, with their 29 atolls and five solitary coral islands, make up what is now known as the Marshall Islands. Marine life associated with these north central Pacific islands is rich and varied. Terrestrial environments range from lush forests and inland mangrove ponds to dry 160˚

165˚

170˚ 15˚

Bokak

Bikini

Enewetak

Rongelap 10˚

t Ra

Ujelang

in ha

C ak

Kwajalein

Ra

120˚

130˚

140˚

150˚

160˚

170˚

Majuro

lik

110˚

180˚

Ch ain

20˚

The atolls and islands of the Marshall Islands were formed as marine animals and plants continually built upon the foundation of submerging volcanoes. Sea level was 1 to 1.5 m higher 4000–6000 years ago, during which time the present Marshall Islands were probably just coral reef. After sea level dropped slightly, an estimated 2000 years ago, the atolls became inhabitable. Deep drilling in the Marshall Islands has shown evidence of volcanoes beneath generations of accumulated marine organisms. The Marshalls’ atolls, with their thousand-plus individual islets, are part of the geographical region called “Micronesia” (“small islands”). Total land is now about 180 km2 (roughly the area of Santa Catalina Island, California) spread over 2 million km2 of ocean, located from 14°37′to 04°37′ N and 172°9′ to 160°55′ E. (Fig. 1). Although the land is relatively flat, rarely reaching even 4 or 5 m in elevation, marine topography is varied, with seamounts, guyots, and pinnacles scattered across oceanic plains. The narrow, irregularly shaped rings of islets encircling lagoons of the Marshalls’ atolls have many features of classic, idyllic paradises (Figs. 2 and 3). Nevertheless, there are harsh environmental conditions—desiccating winds, salt spray, and an annual dry season. Typhoons are uncommon but are devastating when they occur.

10˚

0˚

5˚

-10˚

Ebon

-20˚

-6000

-5250

-4500

-3750

-3000

-2250 Depth (m)

-1500

-750

0

750

1500

FIGURE 1 Map of the Marshall Islands, with locator inset. Satellite

altimetry data: Geoware GMT Companion CD-R Vol. 1, Version 1.9, June 2006. Image prepared by Jenny Paduan and David Clague, MBARI.

610

MARSHALL ISLANDS

BIOLOGY

The equatorial countercurrent, equatorial currents, seabirds, and wind continually bring animals and plants. Both above and below water, the biota displays a close relationship with the far western Pacific. Over 1000 fish species and 100 coral species have been reported; however, there are no native terrestrial mammals, less than three

there are probably less than 75 native species of land plants, but almost ten times that number of introduced plants can be found in urban areas. Concern has been expressed over the possible impacts of sea-level rise and climate change. HUMAN HABITATION

FIGURE 3 Mangrove wetlands, Jaluit atoll.

dozen reptile species, and only about 100 native bird species (although isolated nesting islets can be home to spectacular numbers of individual birds) (Fig. 4). No one atoll is home to more than a few dozen native plants. Endemics are limited and are primarily marine, with the most visible being the three-banded anemonefish (Amphiprion tricinctus). On land, there are a few endemics, including unique pseudoscorpions (i.e., Garupus ornatus of Bikini), insects (i.e., several species of longhorn beetles and sandflies), a landsnail (Assiminea nitida marshallensis), and lizards (i.e., the Arno skink, Emoia arnoensis). The only extant land bird, the Ratak Micronesian pigeon, is an endemic subspecies from the Marshalls’ eastern chain. There are distinct horticultural varieties of pandanus (Fig. 5). The array of the atolls’ overall biodiversity is far from being homogenous, largely because of species’ serendipitous arrival and survival. The atolls’ relative isolation affords habitat to globally endangered migratory species, such as the hawksbill turtle and bristle-thighed curlew. Although less impacted by invasives than other parts of the world, foreign introductions are of concern; for example,

The first inhabitants likely came from the south in outrigger canoes capable of heading into the wind, and the people continued to interact with distant places even after settlement. They imported valuable plants and animals, including giant swamp taro (Cyrtosperma merkussi), breadfruit (Artocarpus spp.), and red jungle fowl (Gallus gallus). Modern Marshallese people are probably largely descendants of the original inhabitants and still largely speak their own language, also called Marshallese, a Micronesian language within the Austronesian family. English, although rapidly becoming widely known, remains the secondary language. INFLUENCES BEYOND THE REGION

Spain first “discovered” the Marshall Islands in the sixteenth century but distant countries had little impact until the mid- to late nineteenth century, when Germany established the copra industry and Christendom’s missionaries from the United States caused major changes. Forests were cleared for coconut plantations, and exotic species were introduced. Change continued under foreign administration (Germany, Japan, the United States Navy, Trust Territory of the Pacific). In 1979, the Republic of the Marshall Islands came into being, and modernization continues

FIGURE 5 A horticultural variety of pandanus from Likiep atoll. The

fruits of most varieties are used extensively for food. Other varieties FIGURE 4 Baby white tern from southwestern Bikini atoll.

serve other purposes.

MARSHALL ISLANDS

611

apace. By the beginning of the twenty-first century, the atolls were home to over 51,000 people. NUCLEAR TESTING

Some of the northwestern atolls were used for the first postwar nuclear testing ever conducted. From 1946 to 1958, 67 nuclear devices were detonated on Bikini and Enewetak atolls, with radioactive material that spread over the rest of the country with long-term ramifications that are still being investigated and debated. Such is only part of what remains to be discovered about the Marshall Islands. The atolls’ overall charming appearance and outwardly simple environment belies their true diversity, complexity, and importance. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Introduced Species / Nuclear Bomb Testing / Pacific Region FURTHER READING

Amerson, A. B., Jr. 1969. Ornithology of the Marshall and Gilbert Islands. Atoll Research Bulletin 127: 1–216. Crisostomo, Y. A. 2000. Initial communication under the United Nations framework convention on climate change. Majuro: Republic of the Marshall Islands Environmental Protection Authority. Erdland, A. 1914. Die Marshall Insulanur. Leben und sitte, sinn und religion eines sudseevolkes. Antropos Bibliothek 2: 1–376. National Biodiversity Team of the Republic of the Marshall Islands. 2000. The Marshall Islands: living atolls amidst the living sea. The National Biodiversity Report of the Marshall Islands. Majuro: RMI Biodiversity Project. Neidenthal, J. 2001. For the good of mankind: a history of the people of Bikini and their islands, 2nd ed. Majuro, Marshall Islands: Bravo Publishers. Republic of the Marshall Islands Biodiversity Clearing House Mechanism. http://www.biormi.org/.

MASCARENE ISLANDS, BIOLOGY CHRISTOPHE THÉBAUD Paul Sabatier University, Toulouse, France

BEN H. WARREN AND DOMINIQUE STRASBERG University of La Réunion, Saint-Denis, Réunion.

ANTHONY CHEKE Oxford, United Kingdom

The Mascarenes are an island group lying near the Tropic of Capricorn in the southwestern Indian Ocean ∼700 km east of Madagascar. This archipelago comprises three high volcanic islands (Réunion, Mauritius, Rodrigues), scat-

612

MASCARENE ISLANDS, BIOLOGY

tered along a ∼600 km west-east axis, and a group of small coralline islands (Cargados Carajos Shoals) ∼400 km to the north of Mauritius, which sit upon a submarine bank of volcanic origin that extends a further 700 km or more to the northeast. The Mascarene Islands have an extraordinary status among islands: Mauritius was the former home of the dodo, the universal symbol of human-caused species extinction on islands. Although their recent history, since the first permanent human settlements in the seventeenth century, has been an endless series of ecological disasters and species extinctions, these islands still harbor up to 25% of their original forest cover and are extremely rich in species and habitats, with high degrees of endemism. Consequently, they are listed among the world’s top biodiversity hotspots. THE GEOGRAPHICAL CONTEXT FOR THE EVOLUTION OF BIODIVERSITY IN THE MASCARENE ISLANDS

Although less well-studied than the Hawaiian Islands, the Mascarene Islands, Rodrigues excluded, are generally believed to result from the same process of plate movement over a stationary hotspot. Today the Réunion hotspot is the source of frequent volcanism on the island of Réunion. The Réunion hotspot’s activity, however, can be traced northeast along the Mascarene Plateau to India, where massive Deccan volcanism coincided with the Cretaceous–Tertiary (K/T) mass extinction event. Rodrigues, which sits next to the Central Indian ridge, is thought to have arisen in relation to the tectonic evolution of the Rodrigues triple junction, located 950 km to the southeast of the island. Typical of such archipelagoes, the Mascarene islands of today have have never been connected to larger land masses. Thus the biogeography and endemic biodiversity of these islands are the product of oceanic dispersal alone. The three main islands of today, Réunion, Mauritius, and Rodrigues, are very different in their size and current topography but are united by their relative geographic proximity and volcanic origin. Réunion, the largest (2512 km2) and most southerly (21° S, 55.5° E), is nearest to Madagascar (665 km), whereas Mauritius, next in size (1865 km2) is 164 km east-northeast of Réunion. Rodrigues, the smallest (104 km2) and currently the most isolated, is located 574 km east of Mauritius. The islands are separated from each other by fracture zones, and each island has developed independently. The most ancient dated lavas from Réunion, Mauritius, and Rodrigues are dated at 2.1, 7.8, and 1.5 million years ago, respectively. However, many exposed lavas in the Mascarenes are the result of recent reactivation, and new

data suggests that Rodrigues, instead of being the youngest, is at least as old as Mauritius. Thus Mauritius and Rodrigues have been available for colonization by diverse biota for about 8–15 million years, while Réunion became habitable much later, about 2–3 million years ago. The extent to which Mauritius and Rodrigues have been isolated from larger land masses over the course of their history has been influenced by Pliocene and Quaternary sea level changes. Information from past sea level curves and current ocean floor bathymetry supports the existence of several large islands, as recently as 18,000–10,000 years ago, along the 115,000-km2 Mascarene Plateau (currently under water with depths ranging from 8 to 150 m) between the granitic Seychelles and the Mascarenes. Drilling projects establish a volcanic origin for (or volcanic contribution to) these islands, with erosion and subsidence thereafter. It is likely that these islands, and also the Chagos and Maldives when fully above water, have played a role as a source of colonists for the present islands. In addition, chains of smaller islands would have reduced the distance for oceanic dispersal and could have served as steppingstones for dispersal between India and the Mascarenes. As a consequence of erosion and subsidence on older islands and volcanic activity on younger islands, the greatest elevations above sea level are currently found on Réunion, with two main summits: Piton des Neiges (3070 m), which is the highest peak in the Indian Ocean, and Piton de La Fournaise (2631 m), one of the most active volcanoes in the world. The highest points of Mauritius (Black River Peak, 828 m) and Rodrigues (Mt. Limon, 398 m) are low in comparison. Réunion, like other young volcanic islands, has a very dramatic topography, being highly dissected into huge caldera-like valleys (cirques) caused by erosion under very high rainfall, with very narrow outlets to the sea through deep gorges. Mauritius, in spite of being an old island, has undergone dramatic geological transformation until recently. Volcanic eruptions have reshaped the island into a series of small, eroded, “geological” islands (age 7.5–5.1 million years) embedded in a matrix of recent lava flows (0.7–0.025 million years old). Thus, both Mauritius and Réunion show considerable spatial heterogeneity in their topography. While the significance of such heterogeneity for the evolution of colonist lineages is evident in the case of Réunion, the biological implications of the “islands within the island” structure of Mauritius, though obvious, have been overlooked by most biologists until very recently. As on the other islands, the main relief of Rodrigues is composed of basaltic lava, but Rodrigues also has an area of limestone plateau made of consolidated coral sands and punctuated with caves.

Owing to their geological history, geographic isolation, and current climate, the Mascarene Islands show more similarities to the Hawaiian Islands than to any other archipelago, even though these two island systems differ greatly in the numbers of islands currently present and the degree of isolation from the nearest other masses. Colonization of the Mascarene Islands by immigrating lineages has occurred relatively recently, but in spite of the simplicity of the present geographic setting, evolutionary diversification in the archipelago has been strongly influenced by a rather complex volcanic evolution combined with a regional geographic configuration that has greatly changed since the first island was formed. THE ECOLOGICAL THEATER

The Mascarene Islands have a tropical climate; that is, temperatures are warm and show little seasonal variation. The climate is strongly influenced by the humid prevailing winds blowing from the southeast, with annual rainfall varying from 500 mm in the driest leeward areas to about 12 m in the wettest areas on the windward slopes of Réunion. Such climate generally promotes the development of forests. From early reports and ecological inference from what is left of the original vegetation, all three main islands were completely forested when discovered. Exceptions are the high-elevation environments above 1900 m on Réunion, where the forests give way to a subalpine scrub. The Mascarene Islands share with other oceanic islands the habitat destruction and transformation associated with human activity. Low-altitude areas have been subject to a much higher impact than high-altitude areas. Although native vegetation remains, all the original forest covering Rodrigues has been destroyed, and a mere 2% of the original cover has been left in Mauritius. In contrast, about 25% of the estimated original extent of Réunion’s habitats are still in a good state. As a result of deforestation and rugged topography, Réunion’s forest remnants are severely fragmented, with large tracts found only above 500 m elevation, and with no more than 1% of lowland forest remaining. Lowland forest remnants are mostly located on the slopes of the active volcano, where they often take the form of forest islands embedded in a matrix of lava flows of various ages. Unfortunately, many of these forest islands were wiped out by a massive volcanic eruption in 2007. Descriptions of the vegetation zones, using historical accounts and subfossil record when necessary, have emphasized five natural plant formations arranged in broad moisture and altitudinal zones. Dry lowland forests dominated by palms (Latania spp., Dictyosperma album),

MASCARENE ISLANDS, BIOLOGY

613

screw-pines (Pandanus spp.), and trees such as Terminalia bentzoe (Combretaceae) were present from sea level to 200 m elevation in areas with less than 1000 mm average annual rainfall. This ecosystem was probably the habitat of some of the most spectacular endemic animals, in particular the now extinct giant tortoises (Cylindraspis spp., Testudinidae), but it no longer exists on the main islands. Some relicts may be found on a small islet (Round Island) off the northern tip of Mauritius and in a few places on Réunion. Semi-dry sclerophyllous forests occurred between coastal areas and 360 m on all sides of Mauritius and Rodrigues, but were restricted to 200–750 m elevation on the western slopes of Réunion, where they still exist in small forest remnants. This ecosystem has an average annual rainfall of 1000–1500 mm and is characterized by ebonies (Diospyros spp., Ebenaceae) and other trees such as Pleurostylia spp. (Celastraceae), Foetidia spp. (Lecythidaceae), Olea europea subsp. africana (Oleaceae), Cossinia pinnata (Sapindaceae), Dombeya spp. (Sterculiaceae), and a variety of Sapotaceae species (Sideroxylon boutonianum, Mimusops spp.). The ecosystem is also home to several spectacular endemic species of Hibiscus (Malvaceae). Many species of this zone, such as Zanthoxylum spp. (Rutaceae), Obetia ficifolia (Urticaceae), and Scolopia heterophylla (Flacourtiaceae), exhibit developmental heterophylly, with juvenile leaves being more divided than those of adults. Such convergence may have evolved to deter herbivory by extinct giant tortoises. Lowland rainforests occur above 360 m (on Mauritius) and all over the eastern lowlands from the coast to 800– 900 m and, on the western side, from 750 to 1100 m (on Réunion) (average annual rainfall 1500–6000 mm). These forests have a canopy of tall trees up to 30 m high and represent the richest plant communities of the Mascarene Islands. Characteristic plants include trees in the plant family Sapotaceae (e.g. Mimusops spp., Labourdonnaisia spp., Sideroxylon spp.), Hernandiaceae (Hernandia mascarenensis), Clusiaceae (Calophyllum spp.), and Myrtaceae (Syzygium spp., Eugenia spp., Monimiastrum spp.); shrubs in the plant family Rubiaceae (Gaertnera spp., Chassalia spp., Bertiera spp., Coffea spp.); and numerous species of orchids (e.g., Angraecum spp., Bulbophyllum spp.) and ferns (e.g., Asplenium spp., Hymenophyllum spp., Trichomanes spp., Elaphoglossum spp.). Dense cloud forests occur on Réunion between 800 and 1900 m on eastern slopes (average annual rainfall 2000–10,000 mm) and between 1100 to 2000 m on western slopes (average annual rainfall 2000–3000 mm) and are also restricted to a small area of Mauritius around

614

MASCARENE ISLANDS, BIOLOGY

Montagne Cocotte above 750 m on Mauritius (average annual rainfall 4500–5500 mm). On both islands these low forests, with a canopy of 6 to 10 m high, are rich in epiphytes (orchids, ferns, mosses, lichens), emergent tree ferns (Cyathea spp.), and, originally, palms (Acanthophoenix rubra), but these now survive only in areas of Réunion where poaching has not wiped them out. Untransformed cloud forests still cover large areas on Réunion (44,000 ha in 2005). These forests are characterized by trees such as Dombeya spp. (on Réunion only) and species in the plant family Monimiaceae (Monimia spp., Tambourissa spp.) as canopy species, with small trees and shrubs such as Psiadia spp. (Asteraceae) and Melicope spp. (Rutaceae) in the understory. They also include large areas of three monodominant plant communities, forests with Acacia heterophylla (Fabaceae) as canopy species that are very similar to Acacia koa forests in Hawaii, thickets dominated by Erica reunionensis (Ericaceae), or hyperhumid screw-pine forest (Pandanus montanus). Finally, above the tree line, at elevations where frosts occur regularly in winter (1800–2000 m), is a unique subalpine scrub dominated by shrubs in the plant families of Ericaceae (Erica spp.), Asteraceae (Hubertia spp., Psiadia spp., Stoebe passerinoides), and Rhamnaceae (Phylica nitida), with some notable endemic species such as Heterochaenia rivalsii (Campanulaceae), Eriotrix commersonii (Asteraceae), and Cynoglossum borbonicum (Boraginaceae) (average annual rainfall 2000–6000 mm). The summits of the volcanoes are covered by large mineral areas with sparse grasslands rich in endemic grasses (Poaceae, e.g., Festuca borbonica, Agrostis salaziensis, Pennisetum caffrum) and orchids (Orchidaceae, e.g., Disa borbonica), ericoid thickets, or thickets of the small tree Sophora denudata (Fabaceae), depending on substrate texture and age. The Mascarene Islands are surrounded by approximately 750 km2 of coral reef. Rodrigues has nearly continuous fringing reefs bounding an extensive lagoon with deep channels, whereas Mauritius is surrounded by a discontinuous fringing reef and a small barrier reef. In contrast, Réunion has very short stretches of narrow fringing reefs along the western and southwestern coasts only. The islets of the Cargados Carajos Shoals, which have a very depauperate terrestrial biota owing to being so low-lying and swamped during cyclones, are bound to the east by an extensive arc of fringing reef, which accounts for ∼30% of the reefs of the Mascarene Islands. Lagoon reefs and reef flats are dominated by scleractinian corals such as branching and tabular Acropora, Porites massives, foliaceous Montipora and Pavona, and sand consolidated

with beds of seagrass such as Halophila spp. (Hydrocharitaceae). Among coral reef fishes, wrasses (Labridae), damselfish (Pomacentridae), carnivorous groupers (Serranidae), and surgeonfishes (Acanthuridae) are particularly well represented. BIODIVERSITY AND ENDEMISM

Identifying species that are unique (endemic) to a group of islands or individual islands relies upon extensive biological inventories, including in-depth systematic investigations, and the ability to recognize cryptic species in lineages that lack substantial morphological differentiation across their range. Although there is currently much effort to fill the gaps, there are still many taxonomic groups in the Mascarene Islands that have not been thoroughly investigated and for which rigorous figures for endemism are not yet available. The Mascarenes therefore present an exciting relatively unchartered study system for island biologists. However, the Mascarene biota, like most other oceanic island biotas, has suffered many recent extinctions that are a source of information bias, particularly in groups of organisms that leave no subfossil materials or that were not recorded and described by the early travelers. The Mascarene biota exhibits high levels of endemism in many groups of related taxa: about three-quarters of the approximately 960 native flowering plant species, ∼65% of the Coleoptera (∼1550 species), and 90% of the nonmarine molluscs (∼200 species) are endemic (Table 1). These degrees of endemism are very close to those observed in similar groups in the Hawaiian Islands or New Caledonia. The Mascarene Islands had the richest oceanic island reptile fauna before the arrival of people. For nonmarine reptiles, the percentage of endemic species

TABLE 1

A Summary of Mascarene Biodiversity and Endemism for Different Taxonomic Groups with Adequate Systematic Knowledge Number of Taxonomic Group

Flowering plants Ferns and allies Mosses and allies Nonmarine mammals Reef fishes Landbirds Seabirds Nonmarine reptiles Nonmarine molluscs Coleoptera

Percent of

Number of

Endemic

Endemic

Species

Species

Species

959 265 ∼800 7 923 60 21 32 200 1538

691 58 40–80 4 42 51 3 30 180 979

72 22 5–10 57 5 85 14 94 90 64

has been estimated to be 94%, but more than half of the 30 endemic species known to have occurred in the Mascarene Islands have gone extinct in the last four centuries, including five species of Indian Ocean giant tortoises. Of three endemic snakes, only one boa still exists (Casarea dussumieri). Apart from bats there are no terrestrial mammals, but all three species of fruit bats (Pteropus spp.; 1 extinct) and at least two (two newly recognized species of Mormopterus) of the four species of microbats are endemic. The Mascarene Islands once had a very rich avifauna, with an estimated 81 native species, 54 of which (67%) were endemic to the archipelago. Apart from three seabirds (Pterodroma baraui, Pseudobulweria aterrima, and an undescribed Pterodroma known only from subfossil bones), most endemic species were landbirds. A large fraction of these endemic birds, especially the larger ones, have now gone extinct, among which were found the legendary dodo (Raphus cucullatus, Columbidae, formerly Rhaphidae), its flightless relative the Rodrigues solitaire (Pezophaps solitarius), and the Réunion solitaire (Threskiornis solitarius, which was not related to the dodo but was an ibis, Threskiornithidae). For the marine biota, there is a dearth of comprehensive systematic investigation at the scale of the archipelago. The average percentage of marine species in the Mascarene Islands that are endemic is apparently lower (2–15%) than in the terrestrial biota. However, the degree of endemism may vary considerably among taxonomic groups, and it is likely that cryptic species, having virtually indistinguishable morphologies, are more widespread in the sea than previously thought. Future broad-scale examination of groups with many wide-ranging species, such as marine molluscs (with at least 3000 species of gastropods occurring in the western Indian Ocean region), crustaceans (with a minimum total of 780 species for the western Indian Ocean), or bryozoans (with at least 500 species in the western Indian Ocean), using molecular taxonomy and new morphometric approaches, may reveal similar levels of endemism to those observed in some terrestrial groups. There are many endemic genera of plants and animals in the terrestrial biota. For example, 32 genera of flowering plants (11% of the total number of genera) and 89 genera of Coleoptera (14% of the total number of genera) are restricted to the Mascarene Islands. Some of these genera provide spectacular examples of diversification within the archipelago, such as weevils (Cratopus, 86 species), leaf beetles (Trichostola, >25 species) and several shrubs (e.g., Badula [14 species], Heterochaenia [3 species], Trochetia [6 species]) (Fig. 1). However, the highest numbers of

MASCARENE ISLANDS, BIOLOGY

615

FIGURE 2 Forgesia racemosa, or bois de Laurent-Martin. This enigFIGURE 1 Badula borbonica, or bois de savon (Myrsinaceae). Badula is

matic and beautiful species, common in cloud forests, is in the Escal-

a species-rich genus, endemic to the Mascarene Islands. Here is shown

loniaceae family. Its closest known relatives live in the Andes, South

a large-leaved species that forms a medium-sized unbranched shrub

America. Photograph by Christophe Thébaud.

in the dense cloud forests of Réunion. Photograph by Christophe Thébaud.

endemics are often found in nonendemic genera (numbers of species in parenthesis), for example, Gonospira landsnails (28), Phelsuma geckos (9), Diospyros trees (14), Dombeya trees (13), Gaertnera shrubs (14), Pandanus screw pines (22), or daisy trees Psiadia (26). Such a pattern suggests that a number of very recent species radiations have been a significant factor in the buildup of endemic biodiversity in the Mascarene Islands. PHYLOGEOGRAPHY, PROCESS OF SPECIES FORMATION, AND ADAPTIVE RADIATION

An important question for understanding endemic biodiversity is the geographical origin of colonizing lineages. As expected from current geography, Mascarene biota has close affinities with Madagascar and Africa. However, unexpectedly, many elements are related to more remote regions, notably Asia and the Indo-Pacific region. In flowering plants, about two-thirds of the genera are shared between the Mascarene Islands and Madagascar and Africa (e.g., Angraecum, Diospyros, Dombeya, Psiadia) whereas at least 20% are shared with Asia and the IndoPacific region (e.g., Astelia, Ochrosia, Terminalia) (Fig. 2). This pattern has been suggested for many other groups, including birds, insects, and even reptiles, including the endemic Mauritian boa family Bolyeridae. Recent phylogenetic analyses using DNA markers have confirmed either the western (e.g., the fruit fly Drosophila mauritiana, Falco kestrels, Cylindrapsis tortoises, Phelsuma geckos, Angraecum orchids, Gaertnera shrubs, Phylica shrubs, Polyscias trees, Psiadia daisy trees) or the eastern (e.g., Mormopterus free-tailed bats, Leiolopisma skinks, Nactus geckos, Hypsipetes bulbuls, Aerodramus swiftlets, 616

MASCARENE ISLANDS, BIOLOGY

Psittacula parakeets, Alectroenas pigeons, the climbing shrub Roussea simplex) origins of some groups (Fig. 3). They have also revealed the Asian origins in groups with poorly understood evolutionary history. The dodo and the Rodrigues solitaire, whose geographic origins have long been mysterious, appear to have dispersed from southeast Asia to the Mascarene Islands at some point in the past. The phylogenetic relationships of Indian Ocean white-eyes (Zosterops) point to an Asian origin for Mascarene species, contrary to intuition (Fig. 4). That a significant portion of colonist lineages comes from the east emphasizes the likely role played by now-submerged land masses between the Mascarene Islands and India and also sea currents and winds in drawing high numbers of colonists from Asia and the Indo-Pacific region into the southwestern Indian Ocean region.

FIGURE 3 The day gecko Phelsuma cepediana, one of the seven sur-

viving Mascarene species, is currently the sole pollinator and seed disperser of Roussea simplex, a climbing shrub endemic to the mountains of Mauritius that was named after Jean-Jacques Rousseau, the Swiss philosopher of the Enlightenment. Photograph by Dennis Hansen.

FIGURE 4 Zosterops mauritianus belongs to Mascarene gray white-

eyes, an anomalous group of warbler-like white-eyes with no “whiteeye” with Asian affinities that appears to have undergone a cryptic adaptive radiation in the Mascarenes. Photograph by Charlie Moores.

How biodiversity builds up in an archipelago like the Mascarene Islands after the first island has appeared above sea level depends on patterns of dispersal and subsequent diversification of founding lineages within the nascent archipelago, including within-island speciation. Some taxa were never successful in colonizing the archipelago. For example, amphibians were absent from the original fauna. Among lineages that colonized the archipelago, some may have repeatedly colonized the archipelago or diversified more than others. Data on the diversification of the Mascarene biota are still too scanty to draw any generalization, but recent molecular studies provide good illustrations of the processes that have led to species diversity in this region. An example in which species diversity within the archipelago reflects multiple successful colonizations from source areas comes from fig trees (Ficus), fruit bats (Pteropus), and orchids (Angraecum). Recent phylogenetic hypotheses imply that the five species of figs found in the Mascarenes, three of them being endemic, have arisen from five separate colonization events. The three endemic fruit bat species apparently originated from three distinct colonizations. Concerning Angraecum, a genus represented by approximately 30 species in the Mascarene Islands, 21 of which are endemic, phylogenetic data implies at least 20 independent colonization events. In contrast, biodiversity in other groups appears to result from single or a few colonization events followed by the evolution of species radiations. The daisy trees (Psiadia), the second most species-rich genus of flowering plant in the Mascarene Islands (Angraecum is the first), display phylogenetic relationships that are consistent with a double archipelago colonization, followed by extensive species radiations within both Mauritius and Réunion. The phylogenetic hypothesis for day geckos (Phelsuma)

is concordant with a single colonization of the archipelago, something that is also true for other endemic reptile taxa in the Mascarene Islands (e.g., Cylindrapsis tortoises, Nactus geckos). The nine Mascarene species of day geckos are best explained by a combination of inter-island dispersal and intra-island speciation events. Understanding how speciation proceeds within small islands in groups as diverse as flowering plants, reptiles, insects, and even birds is not well understood yet in the Mascarenes. Considerable morphological variation is found in diverse organisms on both Mauritius and Réunion. Many species of streptaxid land snails and several species of day geckos display high among-population morphological and/or genetic variation within the islands. The Mascarene gray white-eye (Zosterops borbonicus), a bird endemic to Réunion, shows spectacular variation in plumage traits that coincide with differences in the habitats occupied within this topographically and climatologically diverse island. Thus, it seems likely that natural selection is the key to diversification both among and within islands, although the detailed processes involved remain to be studied. CONSERVATION ISSUES

Before the arrival of the Europeans in the sixteenth century, the Mascarene Islands had evaded discovery by seafarers. The early visitors released ungulates and, on Mauritius, rats and monkeys, but the islands were settled only in the the mid-seventeenth century, when commercial rivalry induced European trading nations to annex and settle the islands. The Mascarene Islands had not experienced major perturbations of the biota when early visitors started to describe what they found. Hence, nowhere in the world has the tragic loss of species and the alteration of pristine tropical island ecosystems been documented as thoroughly as in the Mascarenes. Historical records demonstrate unambiguously that forest clearance, human hunting, and the introduction of nonindigenous predators have been the primary causes of species extinctions in these islands. In total, the Mascarenes have lost about 40% of their native vertebrate species, and most of these were already gone by the middle of the nineteenth century. Some iconic species such as the dodo, the Rodrigues solitaire, and the Réunion solitaire even disappeared from Mascarene landscapes by the late seventeenth or early to mid-eighteenth century. The last giant tortoises were seen in Rodrigues around 1800, and when Charles Darwin visited Mauritius aboard the Beagle in 1836, there had been no record of any tortoise on this island for more than a century. These extinctions and several others mostly predate the first spells of extensive

MASCARENE ISLANDS, BIOLOGY

617

human-caused habitat destruction. This fact implicates human hunting and predation by rats, cats, and pigs, rather than forest clearance, as the primary cause of vertebrate extinctions in the Mascarenes. Patterns of vertebrate species loss also show clearly that the impacts of human settlement and introductions of predators occurred very rapidly. They may have been exacerbated by the fact that many species possessed characteristics that increased their susceptibility to human hunting (large body size) or the impact of nonindigenous predators (lack of mammalian predator escape response, including, e.g., flightlessness, tameness). As a consequence of growing human populations and colonial policies, the rate of forest destruction peaked during the nineteenth century. Forests were cleared for agriculture development to produce sugar cane in both Mauritius and Réunion and maize, coffee, and geranium oil in Réunion, and for slash-and-burn agriculture combined with free-range livestock production in Rodrigues. Such forest destruction likely caused many extinctions by wiping out entire habitats (e.g., semi-dry sclerophyllous forests). Another, almost inevitable, consequence of large-scale forest destruction was increased fragmentation of habitats coupled with invasion by a wide range of introduced plants, additional animals (e.g., Herpestes mongooses, Calotes agamid lizard, and Lycodon wolf snakes), and pathogens. Notable extinctions that occurred during this period include the hoopoe starling (Fregilupus varius, a bird species that was still common in forested areas of Réunion into the early 1850s but had vanished before 1860), the pigeon hollandais (Alectroenas nitidissima), the Mauritius lizard-owl (Mascarenotus sauzieri, last seen in the 1820s or 1830s), the slit-eared skinks (Gongolymorphus spp., which disappeared with the arrival of the wolf snake and survive only on islets offshore), and the anomalous hole-roosting flying-fox (Pteropus subniger, which vanished on Réunion around 1840). During the twentieth century, destruction of forests continued (often for short-lived agricultural initiatives or even make-work programs), fragmentation of forest remnants increased, nonindigenous species continued to arrive in the Mascarenes, and the number of species on the verge of extinction rose steadily. Today, Mauritius and Rodrigues are so extensively altered that most of their native biotas are already extinct or severely threatened. In addition, Mauritius forest remnants are permeated with hordes of invasive mammal species such as monkeys, deer, pigs, and mongoose. By contrast, Réunion is relatively free of these animals. Hence, the survival of

618

MASCARENE ISLANDS, BIOLOGY

relatively intact Mascarene ecosystems largely depends on adequate conservation on the island of Réunion, although even here the best surviving lowland forest was mostly lost to ill-conceived forestry in the 1970s. From a network of reserves finally set up in the 1980s and onwards, approximately 1000 km2 of Réunion (40% of the island area) were designated as a national park in 2007, with a further 35 km2 of marine nature reserves. In Mauritius, nominally protected areas are much longer established (dating from 1951), and a national park covering 66 km2 and much of the best remaining habitat was established in 1994. Total reserves cover a mere 75 km2, but considerable effort has been devoted to conservation management work. Pioneering habitat restoration programs have been underway since the late 1960s, but until recently the programs have been focused on a handful of endangered bird species. Through captive-breeding and release programs, including eradication of nonindigenous predators, the Mauritius kestrel (Falco punctatus), pink pigeon (Nesoenas mayeri), and echo parakeet (Psittacula eques) were rescued from imminent extinction, while the future of Mauritius fody (Foudia rubra) and Mauritius olive-white-eye (Zosterops chloronothus) now also looks more secure. An extraordinary result is that the population of Mauritius kestrels recovered from a single wild breeding pair in 1974, when its prospects were considered to be hopeless, to over 900 individuals in the wild today. The populations of echo parakeets and pink pigeons have also bounced back from 10 (early 1980s) to 300 and 10 (1991) to 360 individuals in the wild today, respectively. In the Mascarene Islands, eradication of nonindigenous predators has become a high conservation priority to prevent further extinctions and is often a prerequisite to ecosystem restoration work. On islets around Mauritius and Rodrigues such as Round Island, Ile aux Aigrettes, or Gunner’s Quoin, eradication programs have succeeded in clearing these islands from feral goats, rabbits, rats, cats, and mice, although house shrews, agamid lizards, and wolf snakes have proved harder to remove. Removal of these predators has led to an increase in numbers of native plant and animal species living on the islets, and the now stabilized conditions on Round Island have allowed the palm forest to recover and the unique reptiles there to thrive, and one species, Telfair’s skink (Leiolopisma telfairii), has been reintroduced onto islands it formerly inhabited, now again rat-free. However, complete eradication of all predators is not always possible, and eradication programs easily become a difficult conundrum for conservation. In Réunion, where

cats and rats threaten endemic petrels that breed on mountain tops, conservation managers have to take into account possible mesopredator effects. While intuition is that cats should be eliminated first, this may not be the best strategy if a population explosion of rats might follow and hit the petrel populations even harder. However, cats mostly eat adult petrels, while rats only attack eggs and chicks. Thus, even if the eradication of cats leads to an increase in rat density, this might not necessarily mean a decline in the petrel populations. Removal of cats and rats in nesting areas is under way, with long-term monitoring projects to ensure that conservation action leads to increased population sizes, not to unwanted decreases due to mesopredator effects. Nonindigenous plant invasions are widespread in the remnant native ecosystems of the Mascarenes. Most invaders colonize human-disturbed sites most successfully, with sizable forest remnants being still dominated by native species. In Mauritius and Rodrigues, and to a lesser extent in the lowlands of Réunion, forest remnants are small, disturbed, and heavily invaded. To improve the prospects of long-term survival of these fragments and the species that inhabit them, pioneering restoration programs began in the late 1960s in Mauritius, and have been much extended, with Rodrigues added, from the early 1980s. Conservation management areas were established in remnants of the major original habitat types and have demonstrated that habitat restoration can effectively reduce the rate of species loss if based on sound ecological knowledge. These areas were fenced to reduce access by introduced herbivores and regularly weeded to control populations of nonindigenous plants. They usually showed an improvement in natural regeneration of native plants in less than ten years, including the legendary tambalacoque tree (Sideroxylon grandiflorum), popularly supposed to be dependent on the extinct dodo. However, although some species that were not regenerating before management were now thriving, nearly half of the species were still not regenerating. This lack is likely related to alterations of plant-pollinator and plant-disperser interactions as a result of extreme habitat fragmentation and animal species extinctions. The loss of native mutualists can limit natural regeneration of native plants that were once dependent on them, through the shortage of pollinators or seed dispersers (Fig. 5). Pioneering work is currently under way to reconstruct missing elements of these critically endangered ecosystems by filling the gaps left by the lost species using analogues (e.g., related species surviving in other parts of the Indian Ocean). However, suggestions to reintroduce from Mauritius species lost

FIGURE 5 Giant Aldabra tortoises (Aldabrachelys gigantea), intro-

duced into Iles aux Aigrettes (Mauritius), can be used as ecological analogue seed dispersers of Syzygium mamillatum, a rare endemic tree. Photograph by Dennis Hansen.

in Réunion (several birds and a fruit bat), and vice versa (Réunion harrier, Circus maillardi), have so far not been acted on. SEE ALSO THE FOLLOWING ARTICLES

Biological Control / Coral / Deforestation / Dodo / Mascarene Islands, Geology FURTHER READING

Atkinson, R., J. C. Sevathian, C. N. Kaiser, and D. M. Hansen. 2005. A guide to the plants in Mauritius. Vacoas, Mauritius: Mauritius Wildlife Foundation. Austin, J. J., E. N. Arnold, and C. G. Jones. 2004. Reconstructing an island radiation using ancient and recent DNA: the extinct and living day geckos (Phelsuma) of the Mascarene islands. Molecular Phylogenetics and Evolution 31: 109–122. Blanchard, F. 2000. Guide des milieux naturels: La Réunion–Maurice– Rodrigues. Paris: Editions Eugen Ulmer. Bosser J., T. Cadet, J. Guého, and W. Marais. 1976–2005. Flore des Mascareignes. Paris: Editions de l’Institut de Recherche pour le Développement (IRD). Cheke, A., and J. Hume. 2008. Lost land of the Dodo: an ecological history of the Mascarene Islands. London: T & AD Poyser. Griffiths, O. L., and V. F. B. Florens. 2006. A field guide to the non-marine molluscs of the Mascarene Islands (Mauritius, Rodrigues and Réunion) and the northern Dependencies of Mauritius. Mauritius: Bioculture Press. Motala, S. M., F.-T. Krell, Y. Mungroo, and S. E. Donovan. 2007. The terrestrial arthropods of Mauritius: a neglected conservation target. Biodiversity and Conservation 16: 2867–2881. Probst, J. M. 1997. Animaux de La Réunion: guide d’identification des oiseaux, mammifères, reptiles, et amphibiens. Saint-Denis, Réunion: Editions Azalées. Turner, J., and R. Klaus. 2005. Coral reefs of the Mascarenes, Western Indian Ocean. Philosophical Transactions of the Royal Society A 363: 229–250. Warren, B. H., E. Bermingham, R. P. Prys-Jones, and C. Thébaud. 2006. Immigration, species radiation, and extinction in a highly diverse songbird lineage: white-eyes on Indian Ocean islands. Molecular Ecology 15: 3769–3786.

MASCARENE ISLANDS, BIOLOGY

619

MASCARENE ISLANDS, GEOLOGY

evolution and (2) a northward increase in age of volcanic activity consistent with rapid northward motion of the Indian plate, followed by slower northeastward motion of the African plate over the last 65 million years. TEMPORAL DISTRIBUTION OF VOLCANIC ACTIVITY

ROBERT A. DUNCAN Oregon State University, Corvallis

The west-central Indian Ocean volcanic islands of Reunion, Mauritius, and Rodrigues are collectively known as the Mascarene Islands. They are volcanoes related to an age-progressive trend of north-to-south volcanic activity that includes the coral-capped Mascarene plateau and the Chagos-Maldive-Laccadive ridge, extending northward to the Deccan flood basalts of western India (Fig. 1). The origin and distribution of these elevated features, rising from ocean floor depths of 4–5 km, are thought to be the result of plate motions over the Reunion hotspot, a persistent upper mantle melting anomaly maintained by focused mantle upwelling. The primary evidence for this idea is (1) common elements of volcano composition and 80°E

60°E

Deccan Traps (65-66)

The volcanic islands of Reunion, Mauritius, and Rodrigues lie between 19° and 22° S and between 55° and 64° E (Fig. 1). Radiometric age determinations, by K–Ar total fusion and 40Ar–39Ar incremental heating methods, provide the time frame for volcanism on the three islands. Reunion Island (21°12′ S, 55°32′ E) rises from an ocean floor depth of ∼5 km to a maximum elevation of 3069 m above sea level (Fig. 2). Two volcanoes make up the island: Piton des Neiges is inactive and forms the northwest twothirds of the island, whereas Piton de la Fournaise forms the southeastern part of the island and is one of the most productive volcanoes in the world. The centers of the two volcanoes are 30 km apart. The oldest lava flows found on the island are about 2 million years old. Piton des Neiges has not been active for 70,000 years, and the volcano is eroding into steep valleys and cirques. Active for 360,000 years, Piton de la Fournaise is being built on the flank of the older volcano in much the same way as Kilauea is forming on the flank of Mauna Loa, at the island of Hawaii. The island’s west coast hosts an intermittent fringing coral reef.

20°N 55°20’E

55°40’E

N Laccadives 21°00’S Ca

rls

be

rg

Ri

dg

e

Piton des Neiges

60

Maldives

21°10’S Central Indian Ridge

0°

Seychelles

50 Chagos

0

Piton de la Fournaise

15km

40

Mascarene Plateau 3000m

30

Ridge Ninety Ea st

ge id R

SW

an di In

In dia n

Ri dg e

0 Ma Reunion

SE

20°S

Piton des Neiges

Rodrigues

Mauritius

1

2

Piton de la Fournaise

3

FIGURE 2 Geological map of Reunion Island, composed of two volca-

620

FIGURE 1 Regional plate tectonic map of the western Indian Ocean,

noes: the older Piton des Neiges and the younger Piton de la Fournaise.

with the volcanic trail of the Reunion hotspot. Black line shows pre-

The oldest shield lava flows of Piton des Neiges are exposed in steep-

dicted trail of the hotspot, in 10-million-year ticks of activity, modeled

sided river canyons (unit 1), overlain by surface lava flows of more

from an assumed stationary hotspot and plate motions over the last

evolved composition (unit 2). The young Piton de la Fournaise volcano

65 million years.

(unit 3) is growing on the southeast flank of the older volcano.

MASCARENE ISLANDS, GEOLOGY

o

57 o 40 'E

o

57 20 'E

o

57 30 'E

N

20 00 'S

20 o 10 'S

0

OLDER SERIES CRATER CALDERA LIMIT

1

5km

Rodrigues

Piton des Piton de la Neiges Fournaise

YOUNGER SERIES INTERMEDIATE SERIES

0 2

Mauritius

Reunion

o

20 00 'S

2

o 20 10 'S

3

20 o 20 'S

20 o 20 'S

Age (Ma)

4 5 6 7 8 20 o 30 'S

o

20 30 'S

A

9 57 o 30 'E

o

57 20 'E

57 o 40 'E

10 0

0 10

teau e Pla ren ca as M

10˚S

m

ChagosLacadive Ridge

11 12

a ntr Ce

Rodriguez

FIGURE 4 Time frame of volcanic construction at the Mascarene

l In

Reunion

30

00

Islands, from radiometric age determinations.

m

idg nR

Mauritius

di a

20˚S

e

B 60˚E

70˚E

FIGURE 3 Geological map of Mauritius Island, with distribution of

three phases of volcanic activity.

Mauritius Island (20°20′ S, 57°30′ E) is inactive and rises from a coastal plain to a central plateau of between 275 and 730 m elevation. The island was built in at least three eruptive episodes (Fig. 3). The Older Series lavas comprise the erosional remnants of a single large volcano that was built up above sea level from ocean depths of 4.5 km, between 7.8 and 6.8 million years ago (Fig. 4). Waning activity persisted intermittently until about 5.5 million years ago. After a 2-million-year hiatus and extensive erosion of the Older Series rocks, smaller volumes of lavas were erupted between 3.5 and 1.9 million years ago and form the Intermediate Series. The Younger Series lavas are confined to localized vents along a system of NNW–SSE fissures that traverse the island. These were erupted between 1.0 and 0.1 million years ago. The island is surrounded by a fringing coral reef at a distance of several hundred meters to 5 km offshore. Rodrigues Island (19°42′ S, 63°25′ E) lies about halfway between the Central Indian sea floor spreading ridge and

Mauritius (570 km distant to the south-southwest), at the eastern end of the Rodrigues ridge. It rises only a few hundred meters above sea level as the erosional remnant of a volcano that ceased activity about 1.5 million years ago. The entire island sits within an extensive, reef-fringed lagoon of around 200 km2. Rocks dredged from the Rodrigues ridge are 7.5 to 11.0 million years old, indicating that the island is the most recent, emergent part of an older volcanic feature. COMPOSITIONAL VARIATIONS

As products of hotspot volcanic activity, the Mascarene Islands show many similarities in composition and evolution to the classic Hawaiian volcanoes. In both Reunion and Mauritius, construction of the main volcanic edifice (“shield”) occurred with eruptions of rather homogeneous magmas derived from moderately high degrees (5–10%) of melting of the upper mantle, followed by low-pressure (i.e., crustal level) fractional crystallization (gravitational removal of crystals of olivine and pyroxene). Many of these early lavas contain significant accumulation of olivine crystals. At the termination of the shield-building phase at both islands, highly differentiated lavas were erupted, consistent with smaller degrees of partial melting of the mantle and more extended cooling in crustal magma chambers prior to

MASCARENE ISLANDS, GEOLOGY

621

eruption. Late-stage, shield-capping lavas in Hawaii are distinctly more alkaline than earlier ones, whereas this is not the case on Piton des Neiges. The Intermediate Series at Mauritius, consisting of highly silica-undersaturated lavas erupted in small volumes following a long hiatus and period of erosion, is analogous to the post-erosional series of Hawaii. The Younger Series lavas at Mauritius, which are less alkaline and were erupted in larger volumes than the Intermediate Series lavas, have no obvious compositional analog in Hawaii. Geochemical data (major and trace element concentrations, and Sr, Nd, Pb isotopic compositions) from lava flows from Reunion, Mauritius, and Rodrigues islands support the proposed relationship to a common mantle plume. These data suggest that the plume composition is heterogeneous, however, with a more “fertile” (lower temperature and higher degree melt fraction) component dominating the early shield-building phase, and a less “fertile” (higher temperature and lower degree melt fraction) playing a more significant role in post-erosional phases (Intermediate and Younger Series at Mauritius). The protracted time scale of volcanic activity at Mauritius (7.8 to 0.1 million years ago) presents a challenge for the hotspot model, because even at relatively slow rates of African plate motion, eruptions have taken place over 200 km away from the hotspot (now beneath Reunion). Explanations involve entrainment of the plume “downstream” (in the direction of plate motion) from the hotspot along the underside of the African plate, together with flexing of the oceanic lithosphere caused by the load created by the growing island of Reunion (2 million years ago to present). RELATIONSHIP TO REGIONAL GEOLOGICAL DEVELOPMENT

The location and timing of volcanic activity at Mauritius and the Rodrigues ridge imply that they are products of the focused mantle upwelling now beneath Reunion (the Reunion mantle “plume”), which apparently produced extraordinary melting at the time of the Deccan flood basalts in western India, 65 million years ago. Northward motion of the Indian plate away from the plume left a trail of volcanic islands and ridges now seen as the Laccadive-MaldiveChagos island chains between 65 and 36 million years ago. The Central Indian spreading ridge then migrated across the plume, from southwest to northeast, causing the subsequent Mascarene plateau to Mascarene islands trend of volcanic activity, 36 million years ago to present, to be left on the African plate (Fig. 1). SEE ALSO THE FOLLOWING ARTICLES

Indian Region / Mascarene Islands, Biology / Volcanic Islands

622

MEDITERRANANEAN REGION

FURTHER READING

Duncan, R. A. 1981. Hotspots in the southern oceans—an absolute frame of reference for motion of the Gondwana continents. Tectonophysics 74: 29–42. Duncan, R. A., and M. A. Richards. 1991. Hotspots, mantle plumes, flood basalts, and true polar wander. Reviews of Geophysics 29: 31–50. Gillot, P.-Y., and P. Nativel. 1989. Eruptive history of the Piton de la Fournaise volcano, Reunion Island, Indian Ocean. Journal of Volcanology and Geothermal Research 36: 53–65. Mahoney, J. J., R. A. Duncan, W. Khan, E. Gnos, and G. R. McCormick. 2002. Cretaceous volcanic rocks of the South Tethyan suture zone, Pakistan: implications for the Reunion hotspot and Deccan Traps. Earth and Planetary Science Letters 203: 295–310. McDougall, I. 1971. The geochronology and evolution of the young oceanic island of Reunion, Indian Ocean. Geochimica et Cosmochimica Acta 35: 261–270. McDougall, I., and F. H. Chamalaun. 1969. Isotopic dating and geomagnetic polarity studies on volcanic rocks from Mauritius, Indian Ocean. Geological Society of America Bulletin 80: 1419–1442. Morgan, W. J. 1981. Hotspot tracks and the opening of the Atlantic and Indian Oceans, in The Sea, Vol. 7. C. Emiliani, ed. New York: Wiley, 443–487. Paul, D., W. M. White, and J. Blichert-Toft. 2005. Geochemistry of Mauritius and the origin of rejuvenescent volcanism on oceanic island volcanoes. Geochemistry, Geophysics, and Geosystems 6, doi: 10.1029/2004GC000883. White, W. M., M. M. Cheatham, and R. A. Duncan. 1990. Isotope geochemistry of Leg 115 basalts and inferences on the history of the Reunion mantle plume. Proceedings of the Ocean Drilling Program Scientific Results 115: 53–61.

MAURITIUS SEE INDIAN REGION

MEDITERRANEAN REGION JOHN WAINWRIGHT University of Sheffield, United Kingdom

Islands make up a significant component of the Mediterranean coastline, with about 19,000 km of length compared to the total of some 46,000 km. However, there are only five large islands—Sicily, Sardinia, Cyprus, Corsica, and Crete, in decreasing order—but numerous groups of smaller islands (Fig. 1). This article describes the groups and individual islands following a clockwise direction from the northwest. GEOGRAPHIC SETTING

The Mediterranean Sea has an area of ∼2.54 million km2, with an average water depth of about 1500 m and a total volume of about 3.7 million km3. It stretches from about 45°42′ N in the northern Adriatic to 30°15′ N off the coast

ic O cea

n

France Slovenia

Croatia Languedoc

Atl

ant

Golfe de Lion

Spain gh

u ro aT

i nc

le Va

Mallorca

Balearics Balearic Basin

Strait of Gibraltar

Alboran Sea

Bay of Naples

Corsica

Adriatic Islands

Italy Tuscan Archipelago

SardinoBalearic Abyssal Plain

Montenegro

Adriatic Sea

Tyrrhenian Abyssal Plain

Albania

Tyrrhenian Sea

Sardinia Galite Archipelago

St Si rait ci of ly

Pelagie Islands

Tunisia

Greece

Lipari Islands

Sicily

Pantelleria

Morocco

Black Sea

Bosnia

Ligurian Sea

Gozo Comino

Malta

Northern Norther Aegean Sporades

Ionian Sea

Ionian Islands

Saronic Sar onic Islands

De

do

Cyclades

ca

Ionian Abyssal Plain

ne

se

Rhodes Crete

Gulf of Gab s Djerba

Turkey

Aegean

Eu bo Sea ea

Mediterranean Sea

Cyprus

Levantine Sea Herodotus Abyssal Plain

0

500

Egypt

Libya

Nil

e

Algeria

km

FIGURE 1 The Mediterranean Sea, showing the location of the main islands and groups of islands. Drawing by Paul Coles.

of Libya and from 35°57′ E off northern Lebanon. At its westernmost point, approximately 35°56′ N, 5°36′ W, it joins the Atlantic Ocean via the Strait of Gibraltar. Just over 14 km wide at its narrowest, the Strait is formed of a shallow sill no more than 400 m deep, which restricts inflow from the Atlantic to about 0.65–0.85 Sverdrup (i.e., 650,000–850,000 m3 per second). Freshwater inputs to the basin are dominated by a small number of major rivers fed from the mountains surrounding the sea and from the Nile and the Black Sea. The size of the Mediterranean and the dominance of the encircling mountains lead the region to have a characteristic “Mediterranean” climate, which is typified by hot, dry summers and relatively warm, wet winters with rainfall dominated by convective events. However, the broad definition of a Mediterranean climate conceals a great deal of variability, with both temperature and rainfall gradients from south to north and east to west. Within these broad gradients, west-facing land masses will also typically have greater rainfall because of the dominance of westerly air flows. As the Mediterranean is an enclosed basin with generally high temperatures and low inflows from both rainfall and runoff, evaporation rates are high. Circulation from west to east produces increasingly saline waters that eventually sink and return at depth in the currents of the Levantine

Intermediate Water. Outflow occurs below the surface across the sill at the Strait of Gibraltar and amounts to about 0.6–0.8 Sverdrup (600,000–800,000 m3/s), reflecting the net evaporative loss across the basin. Physiologically, the Mediterranean is divided into two large basins by the constriction of the Strait of Sicily. The position of islands and the irregular coastline further divides the area into more or less discrete units, with the Alboran Sea, Balearic Basin, the Ligurian Sea, and Tyrrhenian Sea in the west and the Adriatic, Ionian, Aegean, and Levantine Seas in the east. The western basin is the smaller, with a surface area of 0.87 million km2, and descends to depths of 2806 m on the Sardino-Balearic Abyssal Plain and 3427 m on the Tyrrhenian Abyssal Plain. The continental shelf tends to fall off rapidly except for the area of the Alboran Sea, the Valencia Trough, Languedoc, and areas north and east of Corsica and south of Sardinia. The eastern basin has an area of 1.67 million m2, and its deepest points are on the Ionian Abyssal Plain and the Herodotus Abyssal Plain, at 4140 m and 3219 m, respectively. There are large areas of much shallower water between Sicily and Tunisia and to the east of Tunisia, in the Adriatic, and parts of the Aegean. Elsewhere, the continental shelf again tends to fall off rapidly.

MEDITERRANEAN REGION

623

The formation of the Mediterranean is geologically complex and began around 260 million years ago with the breakup of the Pangaea supercontinent. This breakup produced at least 11 major crustal blocks as well as the Iberian subplate (which included the crust underlying Corsica and Sardinia) and the main African and Eurasian plates. Sea floor spreading produced a narrow ocean that originally linked the proto-Atlantic to the protoIndian Oceans. This ocean has been called Tethys (after the woman in Greek mythology who was sister and wife of the god of the sea, Okeanus). Rotation of the African plate by about 65 million years ago produced a series of subduction zones that produced a contraction of the width of Tethys and collisional zones that produced most of the surrounding mountain ranges. The modern Mediterranean came into being around 9 million years ago with the final closure of the seaways to the east through what is now southeastern Turkey. A subsequent, relatively short-lived series of closures of the Strait of Gibraltar during the period from about 5.75 to 5.32 million years ago produced the so-called Messinian salinity crisis, in which large parts of the Mediterranean, and possibly the entire sea, dried up. Many Mediterranean rivers have overdeepened gorges as a result of this event, and the resulting evaporite deposits are now exposed in a number of areas, not least at the eponymous site on Sicily. The region is tectonically and volcanically active. BALEARICS

The main island of the group is Mallorca, with an area of 3640 km2 and a highest point of 1445 m. Menorca lies to the northeast while Ibiza to the southwest. There are a number of smaller islands in the group, the biggest of which are Formentera and Cabrera. Part of the Iberian subplate, which became sutured onto the rest of the Euroasian plate with the formation of the Pyrenees by 35 million years ago, the Balearics rotated clockwise away from the mainland by limited sea floor spreading in the Valencia Trough between around 23 and 10 million years ago. The Serra de Tramuntana mountain range to the north and the Serres de Llevant in the east of Mallorca are dominated by limestones of Triassic to Cretaceous age. The central area is characterized by some Tertiary limestones as well as alluvial fan and other unconsolidated deposits. Menorca has a similar division between older limestones of the Tramuntana to the north and Tertiary sediments to the south. Ibiza similarly has a series of southwest-northeast-trending, uplifted limestone massifs with Tertiary basins between them.

624

MEDITERRANEAN REGION

CORSICA AND SARDINIA

Corsica and Sardinia were also originally part of the Iberian subplate. By about 35 million years ago, lateral movement along the Pyrenean fault zone had emplaced the area that is now Corsica to be adjacent to the Provençal coast of southeastern France and the present west coast of Sardinia to run across the Golfe du Lion. The spreading of the Valencia Trough from around 23 million years ago caused movement away from the French coast, and subsequent displacement of the spreading center to the east of the Balearics rotated Corsica and Sardinia away from these islands, forming the extensive, deep SardinoBalearic Abyssal Plain by 10 million years ago. Further back-arc spreading over a similar time period created the Tyrrhenian Sea and Abyssal Plain, thus rotating Corsica and Sardinia away from what is now mainland Italy. Corsica has an area of 8680 km2 and is dominated by steeply sided, north-south-trending chains of mountains. The highest points are Monte Cinto (2706 m), Monte Rotondo (2622 m), and Monte Padro (2393 m) in the north-central part of the island. This part of the island is made up predominantly of granites as well as some gabbros and related volcanic rocks that date to around 280– 250 million years ago. To the east are a series of schists that were formed during the early Tertiary, together with limited limestone outcrops. The narrow coastal plain on the eastern side of the island is composed of breccias and marls that date to around 18 million years ago. At 23,813 km2, Sardinia is the second largest of the Mediterranean islands. The eastern half of the island is composed of the same batholith structure that dominates western and central Corsica. The 11-km-wide Strait of Bonifacio separates the two islands. The highest points in Sardinia are in this eastern zone, reaching 1359 m at Punta Batestrieri in the north and 1834 m at Punta La Marmora in the center of the island. Granites and related igneous rocks are again intruded into an early Palaeozoic metamorphic basement. The southwest of the island is also made up of similar granites and metamorphic rocks, reaching a high point of 1017 m at Punta Maxia. Uplift and thrusting in this zone were the result of the Pyrenean faulting of the early Tertiary. The remainder of the eastern part of the island is made up of Jurassic-Cretaceous limestones overlain by Tertiary molassic sediments. Rifting in the southwest produced the Campidano Graben during the Late Oligocene to Early Miocene, with the opening of the Sardino-Balearic Basin. The graben contains carbonate deposits as well as subsequent conglomerates, and there are related Tertiary volcanics in the western part of the island.

ISLANDS OF THE TYRRHENIAN SEA

ISLANDS OF THE STRAIT OF SICILY

Islands in the Tyrrhenian can be divided into three groups. The Tuscan Archipelago is located between the northeast of Corsica and the Italian coast. The main island is Elba (224 km2), which has a complex geology with multiple nappes relating to the Apennine development and intrusive and volcanic igneous activity dating to the later Tertiary. The island was known from prehistory for its copper and iron sources. The other islands of the archipelago—Giglia, Capraia, Pianosa, Gorgona, Montecristo, and Giannutri—have late Tertiary volcanic origins and are all less than 24 km2 in area. The Bay of Naples has a number of small islands, again of volcanic origin. Ischia is the largest, with an area of 46 km2, and reaching a maximum height of 788 m at Mount Epomeo. The islands of Procida and Capri also belong to this group. This volcanic activity is ongoing on the nearby mainland at Vesuvius, which also relates to the subduction of the Ionian Abyssal Plain beneath southern Italy. The Lipari or Aeolian Islands are also related to this subduction zone. It contains the eponymous island of Vulcano (from Vulcan, the Roman god of fire and metalwork), with its typical moderately explosive activity. Further to the north, Stromboli exhibits the typical, almost continuous, rhythmic eruptions of this type of volcano. Other islands in the group are Lipari (the largest at 38 km2), Salina, Panarea, Alicudi, Filicudi, and Ustica. Volcanic activity is recent and ongoing.

Malta (245 km2), Gozo (65 km2), and Comino (4 km2) are located about 80 km to the south of Sicily, from which they are separated by the Malta Channel. Both are composed of more or less horizontally bedded coralline limestone overlain by Globigerina limestone and blue clays and topped by a further coralline limestone. Perched water tables in the upper coralline limestone have almost totally been exhausted by overpumping, and groundwater tables in the lower coralline limestone are becoming increasingly vulnerable to seawater intrusion. Pantelleria, located midway between southwest Sicily and Tunisia, is a volcanic island of 83 km2 composed of basalts, pantellerites (alkaline rhyolites), trachytes, and tuff; there have been a complex series of eruptions since ∼320,000 years ago. Most volcanic rocks exposed date to eruptions over the last 50,000 years. Obsidian is found on the island and was widely distributed in the western Mediterranean in the Neolithic period. The Pelagie Islands, made up of Lampedusa and the smaller islands of Linosa and Lampione, are also volcanic but sit on the edge of the African continental shelf. They are important sites for the endangered loggerhead turtle.

SICILY

Sicily’s 25,700 km2 make it the largest of the Mediterranean islands. It rises to a maximum in the east of the island of 3323 m at the summit of Mount Etna, which has built up over the last 700,000 years in a series of eruptions producing subalkaline lavas. Major eruptions have continued to occur through the historical period, some of which have been highly explosive and caused widespread damage. The volcano is also an important global source of CO2, producing on average 26 million metric tons per year, and of SO2, with an average production of 0.6 million metric tons per year. The north of the island is dominated by the Peloritani, Nebrodi, and Madonie mountains. The mountains are principally of limestone with some metamorphic rocks, and relate to Tertiary thrusting events as part of the development of the Apennine chains through Italy. The south-central and northeastern tip (around the type location of Messina) of the island contain extensive evaporitic deposits relating to the Messinian salinity crisis in the Mediterranean, described earlier in this article. These deposits are locally overlain by extensive Pliocene and Pleistocene sediments.

ADRIATIC ISLANDS

There are approximately 1250 islands in the Adriatic, mostly located along the eastern edge. They are predominantly located off the coast of Croatia, with smaller numbers off Bosnia-Herzegovina and Montenegro. Krk and Cres are both located towards the north of the chain and are approximately the same size at 405 km2. Of the other main islands, Pag, Braˇc, Hvar, and Korˇcula are the biggest. Most of the islands have steep slopes and are deeply incised. They are composed of Jurassic and Cretaceous limestones and dolomites and exhibit extreme karstic features. IONIAN ISLANDS

The Ionian Islands are mainly located off the west coast of the Greek mainland. Kerkira (Corfu) is the northernmost, 180 km2, with Paxos (80 km2), Levkas (360 km2), Ithaka (120 km2), Cephalonia (910 km2), and Zakinthos (or Zante, 405 km2) progressively further south. Many of the islands were extensively affected by a major earthquake occurring in 1953. Kithira (280 km2) is separate from the main group of Ionian Islands; it is located off the southern Peleponnese and joined via a submarine ridge to Crete. The Ionian geotectonic zone also continues onto the Peleponnese mainland and consists of a series of Triassic evaporites, Jurassic-Cretaceous limestones in extensional terranes, and Tertiary flysch deposits, reflecting the compression of the Alpine phase.

MEDITERRANEAN REGION

625

CRETE

Crete is bounded to the south by the subduction zone of the Hellenic Trench, which forms the deepest part of the Mediterranean (at 4661 m) to the west of the island. The island covers an area of 8340 km2 and is principally composed of east-west-trending high mountains. There are three principal massifs of crystalline limestone: the White Mountains (Leuka Ori) in the west, peaking at 2452 m; Mount Idhi or Psiloritis in the center (2456 m); and the Dikti Mountains in the east (2148 m). These Triassic, Jurassic, and Eocene limestone areas are made up of high plateaux, numerous caves and other karstic features, and deep gorges. Locally, there are upthrust phyllites and quartzites, as well as flysch deposits. Lower areas were infilled with clays, marls, and conglomerates in the Tertiary, and there are localized Quaternary fan and gravel deposits. Igneous rocks are sparsely distributed through the island, and the southeast and central parts of the island are made up of Jurassic ophiolites. There is a small area of gneiss near the southernmost tip of the island. AEGEAN ISLANDS

Although there are approximately 1200 islands in the Aegean, most of them are very small (only 50 are larger than 40 km2), like the Adriatic islands, and are sparsely inhabited or uninhabited. The region as a whole is undergoing extensional tectonics as a result of indentation from the Turkish subplate, and there are extensive areas of normal faulting as a result. The islands can be considered in six main groups. The Cyclades are located in an arc approximately 110–250 km to the north of Crete. A number of these islands are volcanically active and are related to the subduction of oceanic crust at the Hellenic Trench to the south of Crete. The volcanic islands include Milos (another prehistoric source of obsidian), Antiparos, and Santorini. The latter surrounds an 85 km2 caldera that is the result of the prehistoric megaeruption of Thera ca. 1629 BCE, which buried the Bronze Age city of Akrotiri on the island to a depth of 7 m; ejected an estimated 13–40 km3 of material, some of which reached as far as the Black Sea, eastern Turkey, and the Nile Delta; and possibly caused a tsunami that destroyed coastal settlement on Crete. Islands closer to the subduction zone such as Santorini and Milos tend to produce calc-alkaline volcanics, while Antiparos is made up of alkaline lavas. Some of the larger islands such as Naxos (430 km2) and Andros (380 km2) are made up of high-grade metamorphic rocks, such as blueschists and greenschists, that formed through the Cretaceous and Tertiary and have been uplifted as a result of back-arc extension. Naxos and Paros also have ophiolites exposed. Other parts of the group are made of limestone with karst morphology well developed.

626

MEDITERRANEAN REGION

The Saronic Islands are located in the Saronic Gulf between the Attic peninsula and the Peleponnese. There are two principal islands of the group. Aegina (85 km2) is an extension of the calc-alkaline volcanics in the Cycladic arc, while Salamis (96 km2) is essentially a drowned part of the mainland, located only 2 km from the port of Piraeus. Evvoia (Euboea) is aligned on a similar southeastnorthwest axis as the Cycladic islands of Andros, Tinos, and Mikonos from which it extends, and runs subparallel to the mainland Greek coastline of Attica, Boeotia, and Phthiotis. Evvoia has an area of 3685 km2, and its highest point is Mount Dirphys at 1745 m. At the Euripus Strait, Evvoia is only 40 m away from the mainland, and the island would have been connected to the mainland during lower sea levels in the Pleistocene. The island has a complex geology, with thrusting juxtaposing shallowwater limestones of Jurassic age, with Triassic basalts and metamorphic rocks including ophiolites, and sediments. The Sporades are composed of four main islands— Alonnisos (65 km2), Skiathos (50 km2), Skopelos (95 km2), and Skiros (210 km2)—and 20 smaller islands, to the north and east of Evvoia. They are dominated by calc-alkaline igneous rocks relating to Hellenic Trench subduction. The Dodecanese occur to the east of the Cyclades and close to the Turkish mainland. Again, some of the northern part of the archipelago relates to volcanic activity from subduction at the Hellenic Trench. Active alkaline volcanics are found on Kos (290 km2), while potassic lavas dominate further south at Yali (20 km2) and Nisyros (42 km2). Rodhos (Rhodes) is the largest island at 1400 km2, reaching a peak of 1216 m on Mount Attavyros. The uplands are dominated by Mesozoic limestones with karst morphology and deeply incised gorges, and there are Tertiary flysch and molasse deposits as well as ophiolites to the west of the island. These ophiolites as well as those on the neighboring island of Karpathos are late Cretaceous and thus more closely linked to the ophiolites of Turkey than to those of Crete, the Cyclades, and the Balkans. Lower-lying areas to the north of the island have been infilled with Plio-Pleistocene clastic sediments and shallow-water limestones. Karpathos also exhibits karst morphology in its limestone areas. Northern Aegean islands include Samos (480 km2), with recent alkaline volcanics relating to Hellenic Trench subduction. Further north, Khios (Chios, 840 km2) preserves the remains of an earlier collisional belt, with turbidites and phyllites dating to the Early Carboniferous. Overlying these are Mesozoic limestones with developed karst; there are also middle Miocene andesites, basalts, and

alkaline and calc-alkaline rhyolites in the southeast of the island. Lesvos (1630 km2) can be broadly divided into two parts. To the northwest are early Miocene rhyolites and pyroclastic deposits; to the southeast, ophiolites separate the younger volcanics from metamorphic rocks including marbles, schists, quartzites, and phyllites. Limnos (480 km2) also contains Miocene volcanics above localized Eocene marls and sandstones. There are basaltic, andesitic, and trachyandesitic lava flows and tuffs and small outcrops of quartz monzonite. Gökçeada (280 km2, also known as Imbros or Imroz) is composed of an Eocene sedimentary sequence passing from turbidites to nearshore and fluvial sediments overlain by limestone. Locally there are Miocene potassic volcanics. Samothraki (180 km2) also has potassic volcanics, Miocene granites, and ophiolites. The furthest north of the Aegean islands, Thasos (380 km2), is 7 km from the mainland coast of Macedonia. It is dominated by metamorphic rocks including gneiss, marble, and migmatites. The metamorphism is Oligocene to Miocene in age and relates to the development of the Rhodope metamorphic core complex on the adjacent mainland. In antiquity, the island was well known for its gold mines. CYPRUS

Cyprus is the third largest of the Mediterranean islands, with a surface area of 9250 km2. Except for the small islands of Arwad in Syria and Tyre in Lebanon (joined to the mainland since Alexander the Great built a causeway after conquering the city), it is also the furthest east of the Mediterranean islands. There are two main mountain ranges: the Troödos to the center west and the Kyrenia range running along the northern edge of the island. Between these are the lower-lying plains of the Mesaoria. The summit of Mount Troödos is at 1952 m and is composed of an Upper Cretaceous ophiolite suite surrounded by pillow-lava basalts. The area is extensively mineralized and has some of the world’s major sources of copper (the name for which comes from the Roman name for the island) as well as iron, nickel, cobalt, and chrome. To the west of Troödos, the Mamonia terrane contains Palaeozoic and Cretaceous metamorphic rocks and limestones. The lower parts of the south of the island are made up of Upper Cretaceous to mid-Tertiary limestones and marls. The Kyrenia or Pentadaktylos Range of the north of the island is composed of Permian to mid-Cretaceous limestones. These rocks are steeply dipping following the compression that closed this part of the Tethys Ocean in the late Cretaceous and early Tertiary, thrusting up material from the former sea bed. Some scattered late Miocene evaporites are found in the west and south. The central

part of the island has unconsolidated marls and clastic sedimentary deposits from the Pliocene and Quaternary. ISLANDS OF THE NORTH AFRICAN COAST

The North African coast has relatively few islands compared to the rest of the basin, and none are of any real size. The Gulf of Gabès in eastern Tunisia is the location of Jerba and the Kerkennah Islands. Jerba has an area of 515 km2 and is the only African Mediterranean island with any significant human population. It is just over 2 km from the mainland and has a maximum elevation of 53 m. The island is made up of Miocene and later sediments and is adjacent to the major offshore natural gas field of Ezzaouia. The Kerkenneh Islands are a small archipelago about 20 km from the mainland at Sfax. They are dominated by Quaternary sands and typically only a few meters above sea level. Along the north Tunisian coast there are the small island of Zembra in the entrance to the Bay of Tunis, and further west the Galite Archipelago, which is about 40 km from Cape Serrat. The six islands are made of granite and related volcanics. Other small island groups are the Habibas off Oran (Algeria) and the Chararinas Islands off Morocco. Finally, Alborán, in the center of the sea of the same name, is located about 50 km off the Moroccan coast and 80 km from Spain. GEOMORPHOLOGY

Surface features on the islands are dominated in nonlimestone terrains by intense surface erosion. The dominance of convective storm events, coupled with the relatively sparse vegetation cover, produces rapid surface runoff. Soil erosion accelerates rapidly where rills and gullies form. Coupled with the active tectonic regime over most of the region, these processes produce typically steep slopes, and the lack of significant shelf areas around most islands means that there is often little in the way of fan or alluvial plain buildup in the coastal zone. Wildfire is also an important process that acts to remove vegetation cover and thus increase the potential for erosion. The west-east trend of rainfall suggests that there should be an asymmetry in erosion rates, with the western side of most islands exhibiting more erosion, but this pattern is complicated not only by high geological and soil variability but also by the fact that enhanced water supply will tend to increase vegetation cover and thus reduce this effect. Many of the steeper slopes have been terrraced so that agriculture can be carried out without too high soil loss rates. Landslides are also a common feature, especially in unconsolidated Neogene sediments, and rockfalls are also common in limestone areas, especially in areas of oversteepened slopes. Local

MEDITERRANEAN REGION

627

badlands also form (e.g., on Sardinia) as a result of intense erosion, usually in the areas of Neogene sediments. Rivers are typically steep and short, with gravel beds. In many areas the irregular rainfall and high temperatures mean that rivers are ephemeral and the only permanent water supplies are from groundwater. In limestone areas, karst topography often dominates. Rapid infiltration produces little in the way of surface runoff, especially once soil cover is removed. Residual soils are often found in deep solutional hollows. On many of the islands of the eastern Mediterranean, but also on Sardinia and Corsica, dust deposited on the surface carried from the Sahara is often a significant soil-forming material. It has often been used to explain the distinctive bright red color of many terra rossa soils formed on limestone in the Mediterranean. Crete and some of the larger Adriatic islands have extensive upland plateaux with limestone pavement and poljes. Cave systems and deep-cut river gorges are typical of these landscapes, and rivers on the larger islands may be perennial here because they are fed from groundwater from deep springs, as evaporative loss is lower. The more upland areas exhibit relict periglacial features such as rock debris slopes, from the cold periods of the Pleistocene, but none of the islands are sufficiently high to have been glaciated, unlike parts of the adjacent mainland. A more important consequence of Quaternary climate change is the sea-level change and its effect on joining present-day islands to the mainland, with important consequences for their biogeography as well as for oceanic circulation patterns. For example, Kerkira and Levkas in the Ionian Islands and Jerba would have been joined to the mainland in this way. Of the larger islands, only Sicily would have become joined to the mainland along a narrow isthmus, as the shallowest part of the Strait of Messina is approximately 70 m deep. Many of the Cyclades would have been joined into a single larger island under such conditions, although there was never a complete land bridge across the southern Aegean during the Pleistocene either by this route or via Crete. The Black Sea was cut off from the Mediterreanen during lowstands. Many islands exhibit wave-cut notches from higher sea levels both in the past interglacials and the mid-Holocene. Raised (and drowned) beaches and cliff lines are frequent because of neotectonic movements. Coastal erosion is reduced in many areas because of the very low amplitude of the tidal range in the Mediterranean. ECOLOGY

The Mediterranean has a wide range of species that are considered to be characteristic, although close observation of their extents shows that most are more restricted by other

628

MEDITERRANEAN REGION

conditions rather than a strict adaptation to drought and other features of the Mediterranean climate. The highly accentuated relief on most islands leads to a strong vertical distribution of plants. Of the arboreal species, Aleppo pine (Pinus halepensis or other pines such as P. pinaster or P. pinea) and holm oak (Quercus ilex) tend to dominate below 500–600 m above sea level. From about 500 m to 1500 m, they may be joined with white (deciduous) oak (Q. pubescens) and chestnut (Castanea sativa), then beech (Fagus sylvatica) above 1400 m, and finally Scots pine (Pinus sylvestris) and fir (Abies spp.). Some islands have species that are endemic, for example Corsican pine (P. laricio), or only found in Europe on certain islands; for example, Cretan pine (P. brutia) and Zelkova are restricted to Crete. Nonarboreal species are also commonly endemic—about 10% of the Cretan flora is endemic. Plants may also be highly specialized to local conditions even within the same group of islands. Within the Balearics, 40 endemic taxa are only found on a single island in the archipelago, for example. Even before extensive human disturbance, forest cover may not have been extensive in the Mediterranean because of frequent disturbance from drought and fire. At lower levels, a plagioclimax community tends to develop, with tree species occurring in shrubby forms or even with just shrub forms (for example, the kermes oak Q. coccifera). These two stages are usually characterized as maquis (or macchia in Corsican dialect) and garrigue after the French terms. In Spanish, the use is to call both matorral, while phrygana in Greek is close to garrigue. The extent to which garrigue replaces maquis is most likely related to the frequency of disturbance, including humaninduced. Islands where the geology and water availability has limited soil production may be reduced to desert-like conditions with bare or steppe grassland surfaces. Because of the isolation of many of the Mediterranean islands, animal species also exhibited a high degree of endemism, although many of these species have subsequently become extinct following human settlement. Cyprus was home to species of pygmy hippopotamus (Hippopotamus minutis) and elephant (Elephas cypriotes), which seem to have been hunted to extinction in the late Pleistocene or early Holocene. Dwarf forms of these animals (H. melitensis and E. falconeri) have also been found in Pleistocene sediments on Malta, Crete (H. creutzbergi), and Sicily (H. pentlandi and E. falconeri). Myotragus balearicus is a now-extinct dwarf antelope that inhabited the Balearic Islands. On Sardinia and Corsica, an endemic rabbit (Prolagus sardus) survived into the Holocene, and there are numerous smaller endemic species throughout the Mediterranean islands. Not all species that have been

considered endemic have proved to be so. The mouflon of Corsica and Sardinia is now thought to descend from feral sheep following human settlement in the Neolithic. HUMAN SETTLEMENT OF THE MEDITERRANEAN ISLANDS

Of the major islands, only Sicily seems to have been settled through the Pleistocene, although during the colder periods, there would have been a connection to the mainland. Seafaring is attested for the last few millennia at least of the Pleistocene, by the occurrence of obsidian from Milos on the Argolid mainland of Greece. Cyprus was first occupied around 12,300–11,200 BCE, and the island may have undergone several phases of settlement, as at present there are hiatuses in the early Neolithic archaeological record of the island. Early settlers from the Levantine mainland introduced sheep, cattle, pigs, and deer. Crete does not seem to have been settled until 7000–6200 BCE, and, as on many of the Mediterranean islands, the settlement seems related to the development of simple agriculture in the Neolithic. Sicily, Kerkira, and the Dalmatian islands seem to have been settled around the same time, and these settlements may reflect underlying seafaring activities of hunter-gatherer populations in the Mediterranean. Mesolithic populations are known from slightly earlier than this date on Corsica and Sardinia, and shortly afterward agriculture seems to have been practiced on these islands. Malta was settled by the start of the sixth millennium BCE, and Pantelleria was known as a source of obsidian in the middle Neolithic, indicating that even the smallest of islands were visited. Although there were also previous explorations of the Balearics, agriculture arrived fairly late on the island in the second half of the fourth millennium BCE. From the later Neolithic onward, settlement frequently intensified on the islands, and as elsewhere in the Mediterranean this led to increased disturbance of the vegetation and often the acceleration of erosion rates. FURTHER READING

Allen, H. D. 2000. Mediterranean ecogeography. London: Prentice-Hall. Blondel, J., and J. Aronson. 1999. Biology and wildlife of the Mediterranean region. Oxford: Oxford University Press. Grove, A. T., and O. Rackham. 2001. The nature of Mediterranean Europe: an ecological history. New Haven, CT: Yale University Press. Ricou, L. E. 1994. Tethys reconstructed: plates, continental fragments and their boundaries since 260 Ma from Central America to south-eastern Asia. Geodinamica Acta 7: 169–218. Thompson, J. D. 2005. Plant evolution in the Mediterranean. Oxford: Oxford University Press. Wainwright, J., and Thornes, J. B. 2003. Environmental issues in the Mediterranean: processes and perspectives from the past and present. London: Routledge. Woodward, J. C., ed. 2009. The physical geography of the Mediterranean. Oxford: Oxford University Press.

METAPOPULATIONS DAG ØYSTEIN HJERMANN University of Oslo, Norway

Metapopulations are “populations of populations”— collections of island populations bound loosely together by occasional migration between the islands. Metapopulation theory has been applied to terrestrial environments (where the “islands” are patches of habitat in an “ocean” of unsuitable habitat) as well as to real island archipelagoes. The theory is especially useful in conservation biology. THE CONCEPT

A basic concept of MacArthur and Wilson’s theory of island biogeography was that the collection of species found on an island is dynamic. The metapopulation theory, first used by Richard Levins in 1970, shares this dynamic view of animal populations. In contrast to MacArthur and Wilson’s theory, few people paid much attention to metapopulation theory before about 1990. It differs from the theory of island biogeography in two ways. First, it focuses typically on just one species (animal or plant). This species inhabits islands—which may be real islands in the ocean, or terrestrial or oceanic habitat islands (in an “ocean” of unsuitable habitat). Secondly, there is no mainland where the species is constantly present. Each island holds only a small population of the species, so the population of a specific island will sooner or later die out. However, from time to time, animals are able to migrate from an inhabited island to an uninhabited one and establish (or reestablish) a new population (Fig. 1). As in MacArthur and Wilson’s theory, over time, the system will reach equilibrium. In the metapopulation case, the fraction of islands that are inhabited by the species is roughly constant over time; however, specifically which islands are inhabited by the species will change over time. The scale of a metapopulation can be anywhere from millimeters to thousands of kilometers, depending on the migration capability of the species. MATHEMATICAL MODEL

Although it is not essential for understanding, Levins’s original mathematical model of a metapopulation may shed further light on metapopulation theory. Let us say that there are 100 identical islands with suitable habitat for a given species. At a given moment, p islands are occupied by different populations of the species (whereas

M E TA P O P U L AT I O N S

629

realistic, for instance by allowing for differences in island size and isolation.

Year 1

†

LOCAL EXTINCTION

For most metapopulation theories, it is assumed that the habitat is suitable at all times; that is, extinction events are stochastic (the model can be modified to include deterministic extinction, for instance if the vegetation over time changes to be unsuitable). Four types of processes can lead to stochastic extinction events: (1) demographic stochasticity (caused by inherent variation in, for example, litter size and sex ratio), (2) environmental stochasticity (e.g., variation in survival and recruitment as a result of weather conditions), (3) genetic stochasticity (the loss of genetic variance by genetic drift and inbreeding), and (4) catastrophes (e.g., catastrophic weather events). The risk of population extinction as a result of the first three extinction types increases strongly as the population becomes smaller, whereas the fourth is quite independent of population size. Also, immigration from other islands may rescue that population from extinction (“the rescue effect”). In practice, habitat quality may differ between islands, so some islands (sinks) have a tendency toward extinctions, which is balanced by inflow from islands with high habitat quality (sources).

Year 2

PRACTICAL USE OF THE THEORY FIGURE 1 A schematic map of a metapopulation showing islands with

(•) and without (°) a population of a certain species. In year 1, one population becomes locally extinct (marked with a ), but this is bal-

†

anced by a colonization of an empty island from a neighboring populated island (shown with an arrow).

100 − p are “empty”). For a single occupied island, the probability that the population becomes extinct during a time step (say, one year) is e. For the entire metapopulation, p × e extinctions occur every year (the number of islands multiplied by each island’s extinction probability). The probability that animals from a specific occupied island will colonize a specific “empty” island is c. The number of colonization events in the entire metapopulation during one year is p × (100 − p) × c, depending on the number of “donor” and “recipient” islands as well as the colonization frequency. At equilibrium, the number of extinction events necessarily equals the number of colonization events: p × e = p × (100 − p) × c. Solving this equation for p, we find that at equilibrium, the number of islands that are occupied is p = 100 − e /c. Thus, the entire metapopulation can be maintained only if the colonization rate c is high enough (in this case, c > 0.01e). Metapopulation models have, since 1980, been developed to be more

630

M E TA P O P U L AT I O N S

Metapopulation theory has a number of important implications, especially in the field of conservation biology. Threatened species are often found as small, scattered populations on the brink of extinction, a population structure that typically is the result of fragmentation of the species habitat by human land use. Thus, the metapopulation concept is well suited to such species. One implication of metapopulation theory is that every habitat island counts if we want to save a threatened species—even habitat islands where the species currently does not exist. Because the species survive by recolonizing new habitat islands as fast as local populations go extinct, the entire collection of habitat islands is vital to avoid regional extinction in the long term. A second message of metapopulation theory is that destruction of habitat has a non-linear effect on species abundance and long-term survival: As habitat area is reduced, the species will become regionally extinct long before total habitat area reaches zero. A third message is that migration is important: Without migration, regional extinction is inevitable. It is important to realize, however, that a species that lives in patchy environments does not necessarily function as a metapopulation. Species with good migration capabili-

ties, such as many birds, easily travel between the islands of an oceanic archipelago. For such species, the entire archipelago functions as a single habitat for a single population. SEE ALSO THE FOLLOWING ARTICLES

Dispersal / Extinction / Fragmentation / Island Biogeography, Theory of / Population Genetics, Island Models in FURTHER READING

Fahrig, L. 2002. Effect of habitat fragmentation on the extinction threshold: A synthesis. Ecological Applications 12: 346–353. Hanski, I. 1999. Metapopulation ecology. Oxford: Oxford University Press. Hanski, I. A., and M. E. Gilpin, eds. 1997. Metapopulation biology: ecology, genetics, and evolution. San Diego, CA: Academic Press. Harrison, S. 1991. Local extinction in a metapopulation context: an empirical evaluation. Biological Journal of the Linnean Society 42: 73–88. Hastings, A., and S. Harrison. 1994. Metapopulation dynamics and genetics. Annual Review of Ecology and Systematics 25: 167–188. McCullough, D. R., ed. 1996. Metapopulations and wildlife conservation. Washington, DC: Island Press. FIGURE 1 Midway atoll. Image courtesy of DigitalGlobe.

MICRONESIA SEE PACIFIC REGION

MIDWAY ELIZABETH FLINT U.S. Fish and Wildlife Service, Honolulu, Hawaii

Midway atoll (28°15′ N, 177°20′ W) consists of three sandy islets (Sand Island: 4.56 km2, Eastern Island: 1.36 km2, and Spit Island: 0.05 km2), for a total of 5.98 km2 in terrestrial area, lying within a large, elliptical barrier reef measuring approximately 8 km in diameter (Fig. 1). Although geographically part of the Hawaiian archipelago, Midway is not part of the State of Hawaii and is an unincorporated territory of the United States. CLIMATE

The climate of Midway is influenced by the marine tropical or marine Pacific air masses, depending upon the season. During the summer, the Pacific high pressure system becomes dominant with the ridge line extending across the Pacific north of Kure and Midway. This places the region under the influence of easterly winds, with marine tropical and trade winds prevailing. During the winter, especially from November through January, the Aleutian low moves southward over the North Pacific, displacing the Pacific high before it. The Kure-Midway region is

then affected by either marine Pacific or marine tropical air, depending upon the intensity of the Aleutian low or the Pacific high pressure system. GEOLOGY

Nowhere else on the planet is the tropical island evolution process, with examples of every stage of development, illustrated so beautifully and linearly as in the northwestern Hawaiian Islands. The 1200-mile-long string of islands represents the longest, clearest, and oldest example of island formation and atoll evolution in the world. The ten islands and atolls extending northward from Kauai represent a classic geomorphological sequence, consisting of highly eroded high islands, nearatolls with volcanic pinnacles jutting from surrounding lagoons, true ring-shaped atolls with roughly circular rims and central lagoons, and secondarily raised atolls, one of which bears an interior hypersaline lake. These islands are also surrounded by over 30 submerged banks and seamounts. This geological progression along the Hawaiian ridge continues northwestward beyond the last emergent island northwest of Midway, Kure atoll, as a chain of submerged platforms. Numerous patch reefs formed by reef-building coralline algae and 16 species of corals provide habitat for a wide variety of coral reef species. Midway is at the northern end of the Hawaiian archipelago. The atoll, which is 28.7 million years old, is surrounded by more than 356 km2 of coral reefs.

M I D W AY

631

FLORA AND FAUNA

The biota of Midway is a combination of native and introduced species that are a product of its land-use history. The islands boast enormous nesting colonies of Laysan albatrosses (Phoebastria immutabilis; ∼450,000 breeding pairs) (Fig. 2) and black-footed albatrosses (Phoebastria nigripes; ∼24,000 breeding pairs), making the atoll the largest single colony of albatrosses in the world. Another 14 species of tropical seabirds also breed in abundance at Midway. Introduced canaries breed among historic buildings that mark the beginning of cable communication across the Pacific in the early twentieth century.

atoll to breed. There are 163 species of reef fish reported from Midway, and the shallow-reef fish community is remarkable in the abundance and size of fish in the highest trophic levels. Recent research efforts in the Hawaiian Islands and the Line archipelago have demonstrated that coral reef communities with low levels of human exploitation and disturbance are characterized by fish communities dominated by apex predators. HUMAN HISTORY

There are no records yet discovered of Polynesian visits to Midway, but Captain N. C. Brooks of the Gambia first claimed it under the Guano Act of 1856 after discovering it on July 5, 1859. Midway atoll’s central location in the Pacific made it a critical link in communications and transportation history in the Pacific in the early twentieth century. One of the most important battles of World War II in the Pacific, the Battle of Midway, was fought both at Midway atoll and in the waters beyond it. Midway continued to have a significant military role after World War II; it was an active Navy installation during the Cold War and served as an aircraft and ship refueling station during the Vietnam War. CONSERVATION

FIGURE 2 Laysan albatrosses (Phoebastria immutabilis) incubating

on Sand Island, Midway atoll. Photograph by U. S. Fish and Wildlife Service.

Currently, the land cover on all of the islands at Midway is approximately 30% paved or structures, 23% grass and forbs, 18% woodland, 7% sand and bare ground, 22% shrublands, and less than 0.23% wetland. Of the at least 354 species of plants that have ever been observed at Midway, only 14 are considered indigenous, and only three of those are endemic to the Hawaiian Islands. There are no plant species considered endemic to Midway atoll alone. A total of 508 species of terrestrial arthropods are listed from Midway. Only 41 of these species (8%) are endemic to Midway, and 50 additional species (10%) are indigenous to the tropical Pacific. The large number of alien species that have been introduced and the failure to re-collect three of the four native seed bugs and all three native moths in recent years suggests that the high rate of alien species introductions has reduced the numbers of native insects. Hawaiian monk seals and green turtles forage in the waters offshore but come to the sandy beaches of the

632

M I D W AY

Noteworthy is the role that Midway played in the early history of wildlife conservation in the United States. In 1903 President Theodore Roosevelt sent in the U. S. Marines to stop the slaughter of seabirds at Midway atoll by feather hunters. In 1988 the atoll was designated an overlay national wildlife refuge, and in 1993 the Navy closed the naval air facility and embarked on a major environmental cleanup in which many buildings and underground fuel tanks were removed. In 1996 the Navy formally transferred Midway atoll and the ocean waters out to 19 km to the Department of Interior as Midway Atoll National Wildlife Refuge. In 1999, the U. S. Navy identified the Battle of Midway as one of the two most important events in its history, and in 2000, the U. S. government designated Midway as the Battle of Midway National Memorial. Papah¯anaumoku¯akea Marine National Monument was established by Presidential Proclamation 8031 on June 15, 2006. This resulted in bringing waters out to 80 km around Midway atoll, along with the rest of the northwestern Hawaiian Islands waters, into the largest fully protected marine conservation area in the world. Midway’s terrestrial native vegetation and insect communities have been greatly altered by more than a century

of human occupation. The U. S. Navy, the U. S. Fish and Wildlife Service, and the U. S. Department of Agriculture successfully eradicated black rats (Rattus rattus), introduced in 1943, from all of Midway, and invasive ironwood trees (Casuarina equisetifolia) have been entirely removed from Eastern Island. An active program of invasive weed eradication and native plant propagation is ongoing to restore the plant community present prior to human occupation. This has improved living conditions for the 16 species of tropical seabirds that breed at the atoll as well as for migrant shorebirds such as bristle-thighed curlews (Numenius tahitiensis) and Pacific golden plovers (Pluvialis fulva) that winter there. A translocated population of Laysan ducks (Anas laysanensis, an endangered species whose sole remaining population was at Laysan Island until the translocation) is thriving as it forages on the introduced insect community at Midway. Tragically, the Laysan rail population that was moved to Midway prior to its extirpation at Laysan Island in the 1920s went extinct in 1943 when Rattus rattus arrived at Midway and before individuals could be returned to repopulate Laysan Island. Today, Midway serves as a window to the rest of the Marine National Monument as the only atoll in the chain open for public visitation. It is the site of active ecological restoration efforts and research for conservation science. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Hawaiian Islands, Geology / Invasion Biology / Reef Ecology and Conservation / Seabirds FURTHER READING

Amerson, A. B., Jr., R. B. Clapp, and W. O. Wirtz II. 1974. The natural history of Pearl and Hermes Reef, northwestern Hawaiian Islands. Atoll Research Bulletin 174. Washington, DC: Smithsonian Institution. Clague, D. A. 1996. Growth and subsidence of the Hawaiian-Emperor volcanic chain, in The origin and evolution of Pacific Island biotas, New Guinea to eastern Polynesia: patterns and processes. A. Keast and S. E. Miller, eds. Amsterdam: SPB Academic Publishers, 35–50. Healy, M. 1993. Midway 1942, turning point in the Pacific. Oxford, UK: Osprey Publishing. Hinz, E. 1995. Pacific Island battlegrounds of World War II: then and now. Honolulu, HI: The Bess Press. Nishida, G. 1998. Midway terrestrial arthropod survey. Final report prepared for the U. S. Fish and Wildlife Service. Honolulu, HI: Hawaii Biological Survey, Bishop Museum. Rauzon, M. J. 2001. Isles of refuge: wildlife and history of the northwestern Hawaiian Islands. Honolulu: University of Hawaii Press.

MIGRATION SEE DISPERSAL

MISSIONARIES, EFFECTS OF ALAN I. KAPLAN El Cerrito, California

VINCENT H. RESH University of California, Berkeley

Missionaries have gone to islands to convert local populations away from indigenous religions for millennia. They have been successful because of geographical isolation and cultural aspects of island life, as well as internal and external political influences. The activities of missionaries on islands have resulted in improvements in health, agricultural development, and education but also have brought about drastic cultural changes. Missionaries often had positive influences on island peoples’ lives but, having the range of human frailties, sometimes also did irreparable harm. ISLANDS AND MISSIONS

Islands have been the recipients of a great deal of missionary activity relative to the size of the area they comprise worldwide, although missionary efforts have been far greater in China, Africa, India, and the Americas than on oceanic islands. Missionaries have been going to islands in their search for converts for millennia, proselytizing for Buddhism (the Indian Emperor Asoka sent Buddhist missionaries to Sri Lanka in the third century BC) and Islam (the ruler of Brunei invited Islamic missionaries there after his conversion from Hinduism around 1425) as well as for Christianity, as is most familiar today. As an early example of the latter, the legendary St. Patrick went to Ireland in the mid-fifth century AD to convert the population, ordain priests, and establish churches. In response, Ireland began to send missionaries into the rest of Europe shortly after the death of Patrick, and these represented one of the most vibrant Christian missionary movements in history. Ireland continues this influence today as the country that sends the highest number of missionaries per capita in the world. There is evidence of some groups concentrating their early evangelizing efforts on islands. For example, Catholic missionaries went to Guam in 1668 to establish the first Pacific outpost of European civilization and religion. The Moravians sent the first Protestant missionaries of the modern era to the West Indies in 1732. Mormon

MISSIONARIES, EFFECTS OF

633

missionaries who were originally sent to the British Isles in 1837 sent British converts in turn to Illinois. Mormons also came early to Polynesia in 1844, when Addison Pratt and others sailed there from Boston. Geopolitical and historical contingencies also have played a role in bringing missionaries to islands. For example, Roman Catholic missionaries went to the Philippines with Spanish traders, Lutherans to New Guinea after it became a German protectorate in 1884, Roman Catholics to New Caledonia and Tahiti with French military intervention, Russian Orthodox to the Alaska islands with fur traders from Tsarist Russia, and American varieties of Protestantism to Haiti during the U. S. military occupation there from 1915 to 1934. EVANGELIZATION AND CONVERSIONS ON ISLANDS

Conversions by missionaries on islands can be seen to occur at multiple levels: the option taken by an individual; a collective decision (e.g., an entire group makes this decision); or through the authority of a leader. For example, on New Guinea, missionaries withheld individual baptisms (i.e., the outward signs of personal conversions to Christianity) until there were large numbers of conversions to take place. Wesleyan Methodists converted Tongan royalty in 1830s, which greatly encouraged the conversion of others. Today, the Tongan Free Wesleyan Church (Methodist Church of the United States) is headed by the Tongan monarch, and about one-third of the population belongs to it. On Fiji, little progress was made by either the London Missionary Society or Methodist missionaries until the principal chief Thakombau was baptized in 1854. The conversion of the King and Queen of Huahine (a tributary of the Tahitian ruler King Pomare) resulted in almost the entire population of the island converting to Christianity. There is a clear model of how missionary programs can be successful: the charismatic authority or leader who succeeds in making converts because of his or her personal qualities. The problem is that when that leader leaves or dies, the mission often fails. Even with a charismatic leader, the isolation of missions and missionaries was often a problem. When a missionary died, a long time often passed before a replacement arrived. Isolation sometimes also caused missionaries to become “tropo” or “to go native.” Related to this is the question of whether missionaries should dress like locals, or induce the locals to dress like them? The background and training of missionaries varied greatly. For example, the London Missionary Society sent

634

MISSIONARIES, EFFECTS OF

competent lay tradesmen (blacksmiths, carpenters) with religious zeal but little or no formal theological training. In contrast, other groups (e.g., Jesuits) emphasized religious training. Sometimes, societies with long-established religious ties to one sect are very resistant to conversion to other sects. On Guam and Saipan, Underwood (2005) noted that the Catholic Church is “religion, family, culture, custom, and country,” and the Chamorro culture there fiercely ostracizes converts who leave the Catholic religion. These converts to other religions often find themselves rejected by their immediate family as well as the greater society. As a consequence, converts in the Marianas are usually from the non-Chamorro population on those islands. On predominantly Catholic Guam, Mormon missionaries were chased with rocks, machetes, and even shotguns as recently as 1975 as they went about evangelizing. Missionaries on islands often deal with small populations, which results in increased effort in translating the Bible and other holy texts into local languages. For example, speech-group size might be as small as a few hundred, and in Melanesia even a large speech group might only have a few thousand speakers. This contrasts with some Polynesian islands with many thousands, or continents with hundreds of thousands to millions, speaking the same language. Missionaries often had to provide written languages for populations with largely oral traditions. The introduction of the alphabet and written languages (done by such organizations as the Wycliffe Society to aid in the translation of the Bible, but also done by Buddhists on some islands of Southeast Asia) provided a chance to record oral tradition more permanently. Moreover, translation of sacred texts into local languages raised the status of a language, because their written language now included the “Word of God.” Arguably, it also often resulted in the loss of the oral tradition as biblical stories were used in place of local legends. By providing a written language, missionaries could have great and lasting influence on islands. Wesleyan missionaries developed a written form of Fijian in 1850 by choosing one of the 300 available island dialects as the language for translation of the Bible; this dialect then became Standard Fijian. Fijians were quick to convert to Methodism because of the use of this Fijian Bible and the use of the Fijian language in their services. Today, 90% of Fijian Christians are Methodists. The use of local legends by missionaries was widespread on islands. For example, the traditional creation story of the island of Moorea in the Society Islands featured an

octopus. For this reason an octagonal-shaped building was constructed as the first Protestant church in French Polynesia (Fig. 1). It is widely accepted today that missions have been most successful when they incorporate elements of local beliefs into the religion they are promoting.

FIGURE 1 Octagonal-shaped church on the island of Moorea, French

Polynesia, built over a marae (a ceremonial site) honoring an octopus deity. Photograph by Cheryl Resh.

THE IMPACT OF MISSIONARIES ON ISLANDS

A vivid, popular culture image of the influence that missionaries have had on island communities is the “fire and brimstone”-preaching Rev. Abner Hale, who is the protagonist in the novel and movie Hawaii by James Michener. Even the most ardent believer in evangelization and missionary activities cringes at the havoc Hale creates. In contrast, the presentation of the young midtwentieth-century Mormon missionary John Groberg on Tonga in the autobiographical book In the Eye of the Storm and movie The Other Side of Heaven is extremely positive and even inspiring. These contrasting views reflect current perceptions about both the activities and the role that missionaries have played on islands. Sociologists of religion are clearly divided or at least ambivalent in terms of evaluating the effects of missionary activities on native peoples. In part, this is because societies change with evangelization, and change can be viewed as either positive or negative depending on the perspective of the observer. It can be argued, however, that people choose to be missionaries because they believe that converting people to their own religious beliefs truly will benefit the converted (e.g., through achieving salvation) and thus make their lives better. An individual missionary on an island, perhaps because of the small population usually present, can have

great influence in the process of evangelization and on the people. On the Scottish coastal island of St. Kilda, two sides of this influence can be seen. The Church of Scotland’s Reverend Neil MacKenzie arrived in 1830 and improved agriculture and education, introducing formal training in reading, writing, and arithmetic. However, a later-arriving missionary instituted day-long, mandatory church services on Sunday, forbade children from playing on that day, and required them to carry a Bible wherever they went. For 24 years his strictures seriously interfered with the practical necessities of maintaining island life. Slavery was fought by missionaries in the South Pacific, who tried to protect native populations from “blackbirding,” which included the capture of people for labor on sugar cane fields in Queensland or mines in Peru. But some missionaries actually engaged in slavery themselves. Jesuit priest Honoré Laval terrorized and enslaved inhabitants of the Polynesian island of Mangareva for 37 years. He destroyed statues of their gods and forced them to build a 1200-seat coral and mother-of-pearl cathedral on nearby Rikitea. He was later tried for murder in Tahiti, and not surprisingly found insane. In contrast to Fr. Laval, William Knibb, sent to Jamaica by the Baptist Missionary Society, worked to end slavery. Although firmly instructed by his sending society not to interfere in civil or political affairs, Knibb was outspoken in his support of abolition, resulting in the burning of his chapel and school in Jamaica, his arrest there, and frequent libelous accusations that appeared about him in the press. However, through his efforts, slavery ended in the British colonies in 1838. There have been many other examples of missionaries intervening on behalf of local people. For example, Russian Orthodox priests protected Kodiak Island Alaskans from the warlike behavior of fur traders. Modern missionaries on islands also build schools, orphanages, and hospitals (as well as churches), teach hygiene and public health, provide sanitation and clean water facilities, and educate. Father Damien de Veuster, and the story of his leper colony on Molokai in Hawaii, is still presented to children as an example of self-sacrifice for others. Missionaries, in fact, have been presented as the primary means of modernization in education, medicine, and technology on islands. For example, the introduction of metal to Pacific Islands for creation of longer-lasting tools has been attributed to missionaries. Undoubtedly, some early missionaries were the inspiration for out-of-country aid activities (for example, the Peace Corps program in the United States) that involve volunteers “doing good to help others.” In many cases, the activities of modern missionaries are indistinguishable

MISSIONARIES, EFFECTS OF

635

from those of secular NGO (nongovernmental organization) participants. More recently, there has been an increase in faith-based foreign aid programs, which have been reported to increase prosperity of local economies, increase standard of living, and even contribute to greater global security through poverty alleviation. Even critics of missionary influence would agree that some missionary activities have improved life on the islands. However, in doing so they have changed cultures dramatically. On islands, perhaps because of their isolation, the impact of missionaries is magnified compared to the effects on continents. A common criticism applied to some missionaries on oceanic islands is that “They came to do good and did well,” implying personal economic and political gain resulting from their activities. Throughout the islands of the world, missionaries stopped what Western culture would view as horrific practices such as infanticide (which was often done as a means of reducing overpopulation) and human sacrifices. On the Marquesas in the Society Islands, missionaries put a stop to death rituals performed by female relatives of deceased people that included harming themselves by cutting their hands and faces with sharks’ teeth and other sharp objects. On Samoa, although instructed by the London Missionary Society to restrict their activities to the religious sphere, missionaries ended warfare, polygamy, abortion, and other activities, fundamentally restructuring Samoan society within a few years in the nineteenth century. Some cultural effects of missionary activities are truly indicative of, at least, cultural insensitivity and would today be viewed as foolishness. For example, throughout the South Pacific, social mores such as tattoos and betel-nut chewing were outlawed and upper-body nudity replaced by full body-covering garments. Religious shrines were destroyed and, more unfortunately, oral histories were lost. There has more recently been a response by some island people to preserve their past. Although in the Cook Islands and French Polynesia, tribal religions are now virtually without followers, elements of tribal religion clearly influence the expression of island Christianity. Likewise, Chamorros (Guamanian natives) today are nearly entirely Roman Catholic, but their belief practices reflect Filipino animism, Chamorro ancestor veneration, and religious icon worship. Similar practices can also be seen elsewhere in Micronesia. Tattooing, which was banned by Catholic missionaries on the Marquesas in 1867, was allowed again in 1985 because the Catholic bishop felt that the Marquesans needed these marks as a means of identifying themselves to the rest of the world.

636

MISSIONARIES, EFFECTS OF

Islands today are filled with contradictions because of the missionary influence. On Tahiti, the celebration of the Bible arriving there is a national holiday. However, it is now celebrated with festivals in which the islanders compete in dances that the missionaries tried to forbid! Ironically, missionaries came to islands, had the local people dress conservatively, and required them to abstain from activities on Sunday except churchgoing. Today, tourists from these same missionary-sending countries attend Sunday church for the island-cultural experience, but they often dress skimpily and then follow church services with the activities previously banned on Sunday! Arguably, the advances in agriculture and animal husbandry brought by the missionaries were significant contributions to island economies. However, the subsistence levels of agriculture on Pacific islands probably were sufficient and did not necessarily require greater food production. These increased activities often were instigated because of the missionaries’ view that hard work kept people from “idleness.” Furthermore, the high incidence of diabetes among Pacific Islanders may have resulted in part from changes in diet. Punishment for failing to obey religious laws could be severe. For example, banishment (which on a small island typically meant death) was sometimes a punishment for repeatedly disobeying church laws such as drinking alcohol and Sabbath nonobservance. On Rarotonga in the Cook Islands, missionaries created a “virtual religious police state” (one in six people worked for the police), and in the period 1835–1880, when missionaries were at the height of their powers there, the police rigidly enforced a variety of morality laws. For example, any man who walked with his arm around a woman after dark had to carry a torch in his other hand! Although difficult to separate from modernization in general, higher divorce rates, less stable family lives, and loss of many traditional taboos have resulted after conversions because of the abandonment of the more rigid social structure. For example, in the Cook Islands, the nuclear family replaced communal structure, and living quarters were limited to the smaller units, which reduced society cohesiveness. It should be noted that missionaries with good intentions were often duped by others with less noble motives. In 1863, four evangelical missionaries on Penryn in the Cook Islands, hoping to raise funds for a new church, were tricked by Peruvian slave traders into recruiting their congregation to work in Peru, with the missionaries serving as overseers. With most of the chiefs and men gone (and never to return), the line of leadership succession

was eliminated and the social structure of the island was destroyed. COMPETITION AND CONFLICTS AMONG MISSIONARIES

Evangelizing efforts on islands often take the form of a series of consecutive arrivals of competing sects. After the U. S. purchase of Alaska from Russia in 1867, Baptist and other Protestant missions came into the Alaska islands. They told the Alaskan natives that they had accepted the “wrong” form of Christianity (Russian Orthodoxy) and needed to convert over again. This competition among Christian sects caused entire villages to revert to shamanism. Violence and warfare sometimes resulted from evangelization. The original missionaries’ success on Tonga brought more missionaries to the island, and consequent religious wars between Christian and non-Christian Tongans were fought in 1826, 1837, 1840, and 1852. Twentieth-century missionary conflicts have been less bloody but no less disruptive. In recent years, a great deal of competition among new religious groups on the outer atolls of the Marshall Islands has hurt these communities. Furthermore, competition among sects for reconversions can be seen in American Samoa, where Congregationalists have declined over a 40-year period because of losses to other churches, chiefly to Latter-day Saints (Mormon), Catholic, and Seventh Day Adventist churches. Ironically, increased evangelization does not always result in increased conversions. On many predominantly Buddhist islands in Southeast Asia, a small portion of the population is Christian. Christian missionary activities typically fail to make new converts among any of the Buddhists but are successful in drawing previous converts from different Christian sects. Even on nearby islands, differences in religion and conflicts may be pronounced. Tokelau (north of Samoa) consists of three clusters of islets that have a long history of fighting among each other. The ∼1500 residents occupying the 11 km2 of land comprised by the islands ascribe to different branches of Christianity. Two of the three clusters are predominantly Congregationalist, and the third is entirely Roman Catholic. NATIVE MISSIONARIES

From the very earliest days of evangelization, there has been a zeal on the part of the recently converted to spread the message of their new faith, and religious organizations have used these native missionaries in their conversion activities. Both the Anglican Melanesian Mission and the London Missionary Society used native converts as mis-

sionaries in the Pacific Islands from the beginning of their efforts in the early and middle nineteenth century. From the Cook Islands, the Congregationalists sent 70 of their own native missionaries to Papua between 1872 and 1896, and over 200 have been sent since the practice began in 1830. Maori Mormon missionaries today are in the Cook Islands, working to convert (native) Congregationalists. Fijians have sent out missionaries to other countries for more than 150 years, and in 2001 nearly 100 native Fijians served overseas. Tahitians were sent as missionaries throughout the Pacific in the nineteenth century, but this practice has declined in the last 40 years and likely will not be revived. Besides sending native missionaries to other islands, some groups send them to continental countries that have large emigrant populations from those islands. For example, many of Jamaica’s 60 independent churches have missionary programs to Great Britain because of the large number of Jamaicans who have migrated there. However, Jamaica still receives far more missionaries than it sends. Because it is a predominantly Christian nation today, these missionaries are competing among sects for reconversion to other ones. The most recent trend in missionary evangelization has been the increased number of Protestant missionaries from countries that themselves were fields of missions. For example, in 2002 there were 44,000 missionaries from India, with 60% going on external missions. Today, missionary recruitment is greater from the Southern Hemisphere than from the West, and soon the majority of missionaries will be from Latin America, Africa, and Asia. Clearly, the pattern of South-to-South mission activity (rather than North-to-South) will be a major trend in the future on both islands and continents. CONCLUSION

Missionaries have clearly had major effects on island culture and practice. This is not surprising: as agents of change, missionaries have been highly successful. The authors believe that much of this influence also resulted from the perceptions of the missionaries among themselves and their view of potential converts. The missionaries often viewed themselves in terms of their social positions, and the relationships among them were important. In contrast, their relationship toward the indigenous people was sometimes remote, superior, estranged, and often they viewed the local people as dangerous. The latter view clearly would evoke a need for changing the population’s beliefs. Sometimes, however, native values helped missionaries. For example,

MISSIONARIES, EFFECTS OF

637

the generosity and kindness of Tongans set an example for the behavior of the Mormon missionaries operating there. In The Voyage of the Beagle, Charles Darwin wrote that a marooned sailor would be grateful if missionaries had previously arrived on an island where the unfortunate seaman might find himself. He said that this was because the missionaries would have tamed the “murderous habits” of the natives toward sailors. Unfortunately, the behavior of some missionaries toward natives on islands has often been less than honorable.

creepers, both of which have undergone adaptive radiations primarily involving incredible differences in bill morphology. Another equally spectacular but less wellknown adaptive radiation occurred in the now extinct moa on the two larger islands of New Zealand, the North and South Islands. Moa were giants of the bird world; species varied in body mass from 20 to 250 kg, with the largest genus (Dinornis) reaching up to about 3.5 m when the neck was extended, although recent studies suggest moa carried their heads much lower than this because of the articulation of the vertebral column at the back of the skull (Fig. 1).

SEE ALSO THE FOLLOWING ARTICLES

Cook Islands / Easter Island / Popular Culture, Islands in / Tonga / Voyage of the Beagle FURTHER READING

Barrett, D. B., G. T. Kurian, and T. M. Johnson. 2001. World Christian encyclopedia, 2nd ed. Oxford: Oxford University Press. Hiney, T. 2000. On the missionary trail: a journey through Polynesia, Asia and Africa with the London Missionary Society. New York: Atlantic Monthly Press. Johnston, A. 2003. Missionary writing and empire, 1800–1860. Cambridge, UK: Cambridge University Press. Montgomery, R. L. 1999. Introduction to the sociology of missions. Westport, CT: Praeger. Neill, S. C. 2005. Missions: Christian missions, in Encyclopedia of religions, 2nd ed. L. Jones, ed. Farmington Hills, MI: Thomson Gale. Stackhouse, M. L. 2005. Missions: missionary activity, in Encyclopedia of religions, 2nd ed. L. Jones, ed. Farmington Hills, MI: Thomson Gale. Underwood, G., ed. 2005. Pioneers in the Pacific: memory, history, and cultural identity among the Latter-day Saints. Provo, UT: Religious Studies Center, Brigham Young University. Walters, J. S. 2005. Missions: Buddhist missions, in Encyclopedia of religions, 2nd ed. L. Jones, ed. Farmington Hills, MI: Thomson Gale. Whiteman, D. L. 1993. Oceania, in Toward the 21st century in Christian mission. J. M. Phillips and R. T. Coote, eds. Grand Rapids, MI: Wm. B. Eerdmans Publishing Co. Wright, L. B., and M. I. Fry. 1936. Puritans in the South Seas. New York: Henry Holt and Co.

FIGURE 1 Cast of a skeleton of the South Island giant moa (Dinornis

robustus), standing 2.5 m high in this mount. Image by Claiton Martins Ferreira and Daniel Baker.

MOA ALLAN J. BAKER Royal Ontario Museum, Toronto, Canada

Isolated island archipelagoes with ecologically diverse habitats can be veritable laboratories of evolution for many organisms. Some of the most spectacular examples involve colonizing species of birds, such as Darwin’s finches in the Galápagos Islands and Hawaiian honey-

638

MOA

DISCOVERY AND ORIGIN OF THE MOA

Moa were first discovered scientifically in the nineteenth century when European colonists shipped bones of the giant birds back to England. The renowned comparative anatomist Sir Richard Owen originally dismissed a partial bone fragment as coming from an ox but realized later on closer inspection that it represented an “unknown struthious bird,” a discovery published in 1840 amid much skepticism. As more collections were received and studied, he described a number of new

species from isolated bones. Part of the doubt about Owen’s original deduction of a new type of flightless bird was that it seemed beyond belief that such a large bird, thought to be about the size of an ostrich, could have evolved solely on the relatively small islands of New Zealand. This issue has puzzled biogeographers ever since, leading to the formulation of two competing hypotheses. Many scientists argued that the large flightless birds called ratites—such as kiwis and moa of New Zealand, emus and cassowaries of Australia, rheas of South America, and ostriches and elephant birds of Africa and Madagascar—could not have dispersed to these widespread regions of the world. Thus, they could not have evolved from a common ancestor and do not form a monophyletic group. Others have argued that they do share a common ancestor and were rafted on drifting land masses following the fragmentation of the supercontinent Gondwana to their present geographic locations in the southern hemisphere. A third hypothesis based on the discovery of flighted fossil lithornithiforms in the northern hemisphere suggests that the ancestor of ratites may have flown to these locations and become secondarily flightless. So how did moa originate and become confined to an isolated island archipelago like New Zealand? Studies of shared morphological characters produced a phylogenetic tree that placed ratites in a monophyletic group. Additionally, moa were most closely related to kiwis on a basal branch of the tree, inferring that this was an ancient divergence event consistent with the separation of New Zealand from Gondwana about 82 million years ago. With the discovery that DNA could be extracted from well-preserved moa bones retrieved from swamps or from mud in cave floors, it became possible to test the Gondwana hypothesis by amplifying and sequencing fragments of mitochondrial DNA (mtDNA) because of the high copy number of mitochondria in cells. Furthermore, these sequences could be used to build a tree of relationships from multiple genes, and two laboratories were even able to reconstruct the complete mtDNA genomes of three genera of moa. The approximately even rate of molecular evolution of the genes provided a molecular clock to date the divergence of moa about the time New Zealand separated from Gondwana. As in other DNA studies, kiwis were shown not to be closely related to moa at all, but instead to be close relatives of emus and cassowaries of Australia. The ancestor of moa must have originated separately in Gondwana, and there is no need to posit flying ancestors or island hopping as a

way for moa to reach and proliferate in the isolation of New Zealand. Sadly, moa became extinct around 1300, about 100 years after humans arrived in New Zealand, as they apparently were tame and were an easy source of meat. TIMING AND MODE OF ADAPTIVE RADIATION OF MOA

The diversity of moa species that evolved in New Zealand has been gradually whittled down from an earlier estimate of 64 species to a low of ten species. Once again, ancient DNA extracted from bones provided completely new insights about species diversity and showed why previous estimates based on bone morphology had been so difficult to interpret. Innovative studies that were able to amplify fragments of a nuclear gene that identified the sex of moa species came up with a really surprising result—several genera contained a mixture of smaller and larger individuals that had been classified into two species in each genus. The sexing test instead showed that the smaller individuals were males, the larger ones were females, and intermediate-sized birds could be of either sex. Consequently, this confirmed the earlier removal of a number of (but not all) putative species from the list, consistent with an earlier statistical analysis of morphometric variation suggesting that bimodal distributions represented sexual dimorphism within species. Genetic typing of bones from 125 specimens detected 14 major lineages of moa and indicated misidentifications of bones in collections. By combining the control region sequences with sequences from nine other mtDNA genes, a strongly supported phylogeny of these lineages was derived. If all these lineages represent species, then moa were about as diverse as Darwin’s finches (13 species) and Hawaiian honeycreepers (14 surviving and eight extinct species). To date the divergence times of these lineages, and thus to estimate the tempo of moa evolution, a method of molecular dating was employed that allowed the rate of evolutionary change to vary across the phylogeny. The beginning of the diversification of moa lineages was dated to 15–23 million years ago, coinciding with the Oligocene “drowning” of New Zealand, in which much of the land surface was thought to have been submerged under water. This probably erased the earlier history of moa, especially if the surviving ancestral stock was small. However, by dating the control region tree of 125 specimens, a cycle of lineage-splitting was estimated to have occurred about 4– 6 million years ago, when the landmass was fragmented by tectonic plate movements and mountain-building

MOA

639

FIGURE 2 Bayesian tree constructed by using 658-bp control region sequences from 125 moa specimens. Numbers at the branch tips identify the

14 major lineages. Specimens are color-coded according to geographic locations plotted together with place names. Major mountain ranges are represented by peaks. Unfilled bars at the branch tips indicate specimens without locality data. Asterisks at the nodes indicate posterior probabilities (above the nodes) or maximum likelihood bootstrap values (below the nodes) of 1.0 and 100%, respectively. (Lower Insets) Extent of the New Zealand landmass and movement of tectonic plates from 25 million years ago to present. Faults, subduction zones, and sea-floor spreading centers are shown in red. Modified from Baker et al., 2005.

events, with ocean-level rises severing the link between the North and South Islands, and with general cooling of the climate occurring, including Pleistocene glacial cycles (Fig. 2). This resulted in the geographic isolation of lineages and ecological specialization in different habitats in the North and South Islands. Evidence that moa exploited different ecological niches comes from subfossil remains and observations that no more than four species in different genera existed in the same biogeographic regions of New Zealand and ate different foods. Diets ranged primarily from coarse twigs (Dinor-

640

MOA

nis), soft leaves and berries (Euryapteryx and Emeus), and tough leaves (Pachyornis and Anomalopteryx) to plants in forest edges and adjacent high altitude grasslands (Megalapteryx). Because moa did not have foregut fermentation, it has been speculated that they evolved long intestines to ferment their plant diets, and therefore have correlated large body sizes. These findings jointly provide strong evidence that the mode of evolution was by allopatric speciation, and the flowering of lineages in the last 6 million years was indeed a striking example of an adaptive radiation in the isolation of the islands of New Zealand.

SEE ALSO THE FOLLOWING ARTICLES

TERMINOLOGY

Bird Radiations / Flightlessness / Fossil Birds / Gigantism

Motu is a Polynesian-language term that means “small island” (the plural is also motu), and it is most commonly used for all reef islands associated with high islands, almost-atolls, and atolls. In the scientific classification of reef islands, a distinction is made between motu and cays. In this context, a “motu” is a coral reef island that is found in high-energy environments, has a seaward shingle ridge and a leeward sand deposit, and is relatively permanent. A “cay” is normally made of homogeneous sediment (either sand or shingle) and is ephemeral. Often cays do not support vegetation.

FURTHER READING

Anderson, A. 1989. Prodigious birds: moas and moa hunting in New Zealand. Cambridge: Cambridge University Press. Baker, A. J., L. Huynen, O. Haddrath, C. D. Millar, and D. M. Lambert. 2005. Reconstructing the tempo and mode of evolution in an extinct clade of birds with ancient DNA: the giant moa of New Zealand. Proceedings of the National Academy of Sciences of the USA 102: 8257–8262. Haddrath, O., and A. J. Baker. 2001. Complete mitochondrial DNA genome sequences of extinct birds: ratite phylogenetics and the vicariance biogeography hypothesis. Proceedings of the Royal Society of London, Series B 268: 1–7. Holdaway, R. N., and T. H. Worthy. 1991. Lost in time. New Zealand Geographic 12: 51–68. Huynen, L., C. D. Millar, R. P. Scofield, and D. M. Lambert. 2003. Nuclear DNA sequences detect species limits in ancient moa. Nature 425: 175–178. Paton, T., O. Haddrath, and A. J. Baker. 2002. Complete mtDNA genome sequences show that modern birds are not descended from transitional shorebirds. Proceedings of the Royal Society of London, Series B 269: 839–846. Worthy, T. H., and R. N. Holdaway. 2002. The lost world of the moa: prehistoric life in New Zealand. Bloomington: Indiana University Press.

MOTU FRANCIS J. MURPHY University of California, Berkeley

A motu is a small island, made entirely of coral reef sediment, which forms on top of a barrier reef (Fig. 1). These islands are variably capable of supporting vegetation, seabird colonies, and human settlements, but because of their low elevation and small size, they are also highly susceptible to the effects of major storms and changes in sea level.

FIGURE 1 This small motu on the southeast corner of Raiatea has

formed just a short distance from the reef crest. Photograph © DANEEHAZAMA.com.

MOTU GEOMORPHOLOGY

Because motu are derived from coral reef material, they can be found only where coral reefs form (i.e., in tropical seas where the sea-surface temperatures range between 18 and 34 °C). Furthermore, motu require large storms for their development, so their distribution is only common in tropical areas that support coral reefs and experience cyclonic storms. The Pacific has the largest number of motu, which make up the bulk of the land on the atolls of the Federated States of Micronesia, the Marshall Islands, Tuvalu, the Cook Islands, and French Polynesia. Atolls, such as those of Kiribati, that lie close to the equator and are therefore out of the dominant storm tracks do have some motu but are mostly covered with sand cays. Toward the end of the last glacial period, around 14,000 years ago, sea level was at a lowstand, at approximately 120 m below its present level. As sea level rose during the Holocene, old exposed coral reefs were once again inundated, and existing barrier reefs developed. Dates vary from island to island but generally in the Pacific, sea level approached present levels from 8000 to 6000 years ago, and coral reefs, lagging somewhat, reached present levels about 4000 years ago. Once sea level and reef crests reached a state where sedimentation on the reef flat was possible, then the stage was set for motu to form. There is evidence that in some places where motu occur there was a late-Holocene highstand, with relative sea levels up to 1 m above present that persisted until as recently as 2000 years ago. In this situation, motu could not have formed until relative sea levels dropped to close to present levels. Motu have a pattern of development and maintenance whereby the two different sediment classes experience erosion and sedimentation at different times and rates: The large-scale sediment deposits quickly during occasional high-energy storms and erodes slowly during normal periods, and the small-scale sediment deposits slowly during normal times and erodes quickly during high-energy events. Large storms often create or add to steep berms on the

MOTU

641

seaward shore made of sediment that ranges from shingle to boulders. These then create a protected area directly leeward, where during calm periods finer sediment—sand and gravel—accumulates. These same large storms can, however, create waves that completely submerge the island. In these cases, small sediment is usually swept into the lagoon. Motu are found both on reefs that show no sign of emergence (drop in sea level) and on reefs that are slightly emerged, such as those on the south shore of Rangiroa in the Tuamotu Archipelago. Where emergence has raised a shoreline to the point that wave-driven sedimentation can no longer occur, there are no motu—for instance, on Niau in the Tuamotu. Motu are stabilized somewhat by the cementation of sediment into features such as conglomerate platforms and beachrock. Reef sediment that is wetted sufficiently (i.e., in or just above the intertidal zone) will be bonded together by calcareous cement that precipitates between sediment particles. On beaches, this creates beachrock, which has the slope and sediment characteristics of the original beach. Rubble banks created by large storms cement from the bottom up to create conglomerate platforms that have a characteristically level surface and a wide range of clast sizes. Both these features become exposed over time when the loose sediment covering them is removed by subsequent storm action. MOTU LIFE

Today, motu and sand cays are often thought of as the ideal paradise islands, with iconic palm-shaded white-sand beaches adjacent to a turquoise lagoon. They are, in fact, great places to visit, but long-term settlement is difficult because they are harsh environments for both plants and animals. The native vegetation of motu is characterized by hardy plants that can withstand salt spray and survive with little freshwater and a very narrow spectrum of organic nutrients. These include shrubs such as Pemphis acidula and Suriana maritima, and trees such as Hernandia sonora, Pandanus tectorius, Pisonia grandis, and Tournefortia argentea. As plants become established, they stabilize and add nutrients to the motu sediment, and they create roosting areas for birds, which in turn also contribute nutrients to the motu. Freshwater that falls on motu seeps into the porous coralline base of the island and sits on top of the heavier saltwater. The potential size of this reservoir is relative to the area and elevation of the motu, such that there can be 40 times the amount of water below sea level as there is above it. The actual volume is regulated by rainfall and utilization of the aquifer by plants and people. It is clear from this relationship that small changes in sea level can effect large changes in freshwater capacity.

642

MOTU

The habitability and carrying capacity of motu for humans varies with the size of the motu and the amount of rainfall that it receives. People living on motu have to depend heavily on marine life for food because it is difficult to grow food crops in this harsh environment. Crops that can be cultivated on motu include coconuts, taro, bananas, and arrowroot. Because of these generally harsh conditions, high-density populations on motu are unusual, and the normal high end is around 500 people per square kilometer, such as in the villages of Avatoru and Tiputa on Rangiroa in the Tuamotu Archipelago. In extreme cases such as Ebeye Island on Kwajalein Atoll in the Marshall Islands there are more than 33,000 people per square kilometer. More common densities of inhabited motu, however, are below 50 people per square kilometer. Radiocarbon dating of conglomerate platforms across the Pacific (which ranges from 4000 to 1500 years ago), and archaeological evidence for the same region, shows that the creation of motu only slightly predates the period over which most of the islands in the central and eastern Pacific were being settled, with atolls in general being settled later than the high islands. Even as human populations were moving out through Oceania, searching for new landfall, the combined effects of sea level, storms, reef growth, and plant dispersal and settlement, were producing new motu for them to live on. Rising sea levels, an increase in frequency of major storms, and the lowering of pH in ocean surface water are all likely consequences of global climate change. For motu this may have grave consequences. The predicted sea-level rise of 30–80 cm over the next 100 years will erode motu shores and reduce the volume of the freshwater lens. Major storms will do the same, only more drastically, as well as directly removing flora and fauna. A lowering of pH will slow calcium carbonate production and eventually increase chemical erosion of reef sediments. Because of humaninduced climate change, these landforms, and all of the biota and human populations associated with them, will increasingly become endangered and may in fact disappear completely within a few generations. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Coral / Global Warming / Oceanic Islands / Sea-Level Change FURTHER READING

Alkire, W. H. 1978. Coral islanders. Arlington Heights, IL: AHM Publishing. Guilcher, A. 1988. Coral reef geomorphology. New York: John Wiley and Sons. Nunn, P. D. 1994. Oceanic islands. Oxford: Blackwell. Stoddart, D. R. 1975. Almost-atoll of Aitutake: geomorphology or reefs and islands. Atoll Research Bulletin 190: 31–57. Stoddart, D. R., and J. A. Steers. 1977. The nature and origin of coral reef islands, in Biology and geology of coral reefs. O. A. Jones and R. Endean, eds. New York: Academic Press, 59–105.

N NEW CALEDONIA, BIOLOGY JÉRÔME MURIENNE Harvard University, Cambridge, Massachusetts

Because of its extremely diverse biology, New Caledonia is one of few islands to be designated a biodiversity hotspot. The long isolation of the territory (since the breakup of Gondwana 80 million years ago), in conjunction with its climatic stability, was often proposed as an explanation for its outstanding biodiversity. Recent evolutionary studies are more in accordance with submersion of the territory from 65 to 45 million years ago, establishing a new paradigm for the origin of biodiversity in this island. GEOGRAPHY OF THE TERRITORY

New Caledonia is a Melanesian archipelago comprising two sets of islands (Fig. 1). The first includes the main island (Grande Terre, approximately 500 km long and 50 km wide), the Belep Islands to the north, and the Isle of Pines to the south. The second includes more recent volcanic islands (Ouvéa, Lifou, Tiga, and Maré), called the Loyalty Islands. The inclusion of other small dependencies (Chesterfield, Matthew, and Hunter) brings the total land area to 18,972 km2. The climate is subtropical, and the mean annual temperature varies between 22 and 24 °C. There are two major seasons: a hot season from mid-November to mid-April and a cool season from mid-May to mid-September. The Grande Terre harbors an asymmetric mountain chain. The eastern part is abrupt, whereas the western part ends

in long plains. The culminating points are Mont Panié (1629 m) in the north and Mont Humboldt (1618 m) in the south. The mean annual rainfall is 1700 mm, but precipitation is also uneven, mainly because of the mountains. Eastern parts of the island can receive five times more rain than their western counterparts. Different soils are found in New Caledonia. Among them, ultramafic soils cover one-third of the territory (mainly in the south). Their richness in metal and poverty in nutritive elements makes that area a very particular environment for the flora and its associated fauna. BIOLOGICAL CHARACTERISTICS

Compared to other islands, New Caledonia, despite its small size, exhibits an extraordinary species richness and rate of regional endemism (Fig. 2). In addition, several groups are considered relictual. Local endemism to restricted areas is a general characteristic of groups such as terrestrial squamates or some insects and can also be found in some plants. New Caledonia has long been seen as a distinctive floristic and faunal entity and, as mentioned, is now recognized as a biodiversity hotspot. The dense evergreen rain forest is the richest vegetation type. It is mainly located on the east coast of the island because of the high precipitation there. This forest is present on every kind of substrate, from 300 m to the highest altitudes. The sclerophyll forest is the most endangered vegetation type. Including small-sized trees (10 to 15 m), this dry forest is located on the west coast between the altitudes of 0 and 400 m. Twenty-four percent of its plant species are locally endemic. The scrubland, or maquis, because it is mostly associated with ultramafic soils, is often referred to as maquis minier (“mining maquis”). High-altitude maquis covers 100 km2, and low- to mid-altitude maquis covers 4400 km2.

643

FIGURE 1 Principal types of vegetation found in New Caledonia Grande Terre (current coverage). Plant rates of endemism are noted in the legend.

Diagrams represent the evolution from the original to the current coverage.

The savanna covers 6000 km2, and its major tree component is the niaouli, Melaleuca quinquenervia. Other types of land include mangrove forests, swampland, and secondary shrubland. New Caledonia presents a very distinctive flora and fauna. Thirteen endemic species of coniferous trees in the genus Araucaria are found in this territory, out of only 19 worldwide species. Illustrating the unique nature of the island, the endemic Amborella trichopoda is the sole member of the family Amborellaceae, which is a sister group to all the remaining flowering plants. Likewise, the kagu (Rhynochetos jubatus), an emblematic, almost flightless forest bird, is the sole representative of the endemic family Rhynochetidae, an additional lowland species endemic to the island having gone extinct in the Holocene. ORIGIN OF THE BIODIVERSITY

New Caledonia has often been described as a “museum” in which its tremendous biodiversity has been explained by a long accumulation of species in ancient groups since Gondwanan times. The geological history of the region, however, contradicts this classical view. The total submersion of the territory between 65 to 45 million years ago

644

NEW CALEDONIA, BIOLOGY

(subsequent to separation from Gondwana) was either unrecognized or disregarded because of the presence of the “relictual” groups. The recent use of molecular phylogenetic methods has enabled the investigation of the origin of the biodiversity in New Caledonia in an evolutionary framework. Most studies on plants, vertebrates, and invertebrates show that for supposedly Gondwanan groups, their origin in New Caledonia is never older than 40 million years ago and can involve multiple dispersal events from the neighboring regions. Additionally, their diversification can be extremely recent (a few million years), especially in insect groups. A new paradigm has now emerged that presents a more balanced view, with either single or multiple dispersal events and possibly different tempos of evolution inside New Caledonia. The potential existence of refugium islands during submersion times could explain the presence of “relictual” groups in New Caledonia. On a local scale (inside New Caledonia) the high heterogeneity of the territory, rather than its long isolation, is certainly an important factor of diversification. The presence of ultramafic soils (toxic and poor in nutrients) could have also been involved in some process of adaptive radiation.

THREATS TO THE BIODIVERSITY

Fire is an important factor in the degradation and the transformation of the vegetation. Because of the fire resistance of the niaouli tree, the expansion of the savanna (to the detriment of the primary vegetation) has been significant. Even though fires are attested as a natural process before human arrival, recent fire is mainly due to human activities. Invasive species constitute a major threat to the biodiversity. Among plants, around 1300 species are nonnative, and 67 taxa are considered invasive. The rusa deer, Cervus timorensis, was introduced less than 150 years ago and retards the regeneration of the forest. With its fierce sting, the little fire ant, Wasmannia auropunctata, was introduced in 1960. In addition to harming humans, cattle, and cultures, it induces a serious ecological problem by competing with other arthropods (it is known to use its venom against other ant species) and potentially affecting the entire food web. Even though open-pit mining is better managed than in the past, it is still a concern. The high international demand for nickel has led to the development of new projects that may threaten some relictual forests. The necessary economical development of the island (mining representing the major source of income) needs to better integrate the question of the preservation of New Caledonia’s exceptional biodiversity. Numerous protected areas exist, but their impact on the conservation of plant diversity has proven to be relatively low. Perhaps the most urgent and promising initiative has been the creation in 2001 of the “dry forest conservation program.” This program groups ten different partners (including national and local agencies and international NGOs) that join together to put a stop to the erosion of the most endangered New Caledonian ecosystem.

FIGURE 2 Rates of endemism in different groups. Common names

have been used even though they do not always represent monophyletic groups. Numbers for plants include only native species. Endemism in invertebrates varies between 38% in the butterflies to 100% in the less mobile groups.

Documents Scientifiques et Techniques II 4. Nouméa, New Caledonia: Institute de Recherche pour la Développement. Marquet, G., P. Keith, and E. Vigneux. 2003. Atlas des poissons et des crustacés d’eau douce de Nouvelle-Calédonie. Collections Patrimoines Naturels 58. Paris: Muséum National d’histoire Naturelle. Morat, P. 1993. The terrestrial biota of New Caledonia. Biodiversity Letters 1: 69–71. [This special edition includes numerous review papers on the subject.]

NEW CALEDONIA, GEOLOGY

SEE ALSO THE FOLLOWING ARTICLES

TIMOTHY J. RAWLING

Ants / Endemism / Invasion Biology / New Caledonia, Geology / Vegetation

University of Melbourne, Australia

FURTHER READINGS

Bauer, A. M., and R. A. Sadlier. 2000. The herpetofauna of New Caledonia. Ithaca, NY: Society for the Study of Amphibians and Reptiles. Beauvais, M.-L., A. Coléno, and H. Jourdan. 2006. Invasive species in the New Caledonian archipelago. Paris: Institut de Recherche pour le Développement. Dry Forest Conservation Program. http://www.foretseche.nc [in French and English]. Jaffré, T., P. Bouchet, and J. M. Veillon. 1998. Threatened plants of New Caledonia: is the system of protected areas adequate? Biodiversity and Conservation 7: 109–135. Jaffré, T., P. Morat, J.-M. Veillon, F. Rigault, and G. Dagnostini. 2004. Composition and characterization of the native flora of New Caledonia.

The New Caledonia archipelago is a group of islands that form the emergent northern part of the Norfolk Ridge in the Southwest Pacific Ocean. The archipelago consists of six major islands as well as numerous smaller islands. The closest neighboring island group is Vanuatu, 500 km to the northeast, while New Zealand and Australia are situated 1500 km to the south and west, respectively. The New Caledonian territorial boundaries are approximately between latitude 18° and 23° S and longitude 158° and 172° E, and the land area of the archipelago is in the order of 18,600 km2.

NEW CALEDONIA, GEOLOGY

645

FIGURE 1 Regional geology map of New Caledonia including major faults and fold hinges.

GEOGRAPHIC SETTING

The largest island in the New Caledonia group, commonly referred to as “Grande Terre,” is elongate (about 400 km long and 60 km wide; 16,350 km2) and strikes northwest–southeast. The geomorphology of Grande Terre is highly variable and includes a mountain chain that extends from the northeast coast through the central chain to the southern plateau, numerous evolved erosional river valleys, and a gently undulating west coast disrupted by several steep-sided, mesa-like plateaus. The highest peaks are Mt. Panié (1628 m), Mt. Colnett (1505 m), and Mt. Ignambé (1311 m) on the northeast coast and Mt. Humboldt (1618 m) in the south. The climate on Grande Terre is highly variable. New Caledonia is situated immediately north of the Tropic of Capricorn and thus is on the very edge of the tropics. The large range of mountains that extend much of the way down the east coast act as a barrier to rain-bearing clouds approaching from the east or northeast. As a result, the much flatter west coast lies within a rain shadow and receives considerably less rain. Rainfall on the northeast coast is typically in excess of 4000 mm per year, whereas the west coast may receive as little as 1000 mm per year. The other major islands of the New Caledonia archipelago include the Loyalty Islands, Maré, Lifou, and Ouvéa, as well as the smaller Île des Pines and Belep Island. The Loyalty Islands are situated on the Loyalty Ridge, which runs parallel to the northern Norfolk Ridge, about 100 km to the east. These islands are volcanic in

646

NEW CALEDONIA, GEOLOGY

origin and are covered by low-lying coral reef. Recent uplift has resulted in the emergence of the reefs and exposure at present levels. These islands have little relief and no developed drainage systems (Fig. 1). GEOLOGY

The main island of New Caledonia is composed of a complex mixture of metamorphic, igneous, and sedimentary rocks with both continental and oceanic affinities. These rocks record a geological evolution that spans 300 million years and is punctuated by three major phases of tectonic activity (or deformation and heating): (1) late Paleozoic to early Mesozoic convergent-margin tectonism along the eastern margin of Gondwana, (2) mid- to late-Cretaceous extensional tectonism associated with the dismemberment of the eastern passive margin of Gondwana, (3) collisional orogenesis, peridotite nappe emplacement, and high-pressure metamorphism during the Tertiary. The geology of New Caledonia can be divided into six geologically distinct tectonostratigraphic terranes: (1) the ultramafic nappe, (2) the central mountain chain, (3) the western coastal belt, (4) the eastern coastal belt, (5) the northern region, and (6) the Loyalty Islands. The Ultramafic Nappe

The ultramafic nappe is an extensive ophiolite complex, or sheet of ocean crust, that was originally thrust over much of the island during the late Eocene during a major

collisional tectonic event but has subsequently been disrupted and locally removed by faulting. The ultramafic rocks, primarily harzburgite and dunite, comprise over 40% of the rocks exposed in New Caledonia. The age of ophiolite emplacement has been established in the western part of the island, because the peridotite at the base of the sheet is in thrust contact with underlying Upper Cretaceous and Lower Tertiary sediments. In addition, there is a Late Eocene flysch (or marine sediment formed at the time of collision) that unconformably overlies part of the ophiolite sheet. This unit contains basalt pebbles derived from the basement, as well as serpentinized peridotite pebbles derived from the overthrust ophiolite sheet. During emplacement the ophiolite can be seen to have overpushed material to the southwest, indicating northeast to southwest–directed collision and thrusting. The ultramafic rocks are thought to be relics of a once continuous nappe of Cretaceous oceanic crust and mantle, derived from the southern flank of a WNW-ESE oceanic ridge in the Loyalty Basin. Geophysical data confirms that the ultramafic rocks exposed in New Caledonia are continuous with crust-mantle structures in the Loyalty Basin. Tropical weathering of the ophiolite material has formed widespread and deep laterites that are strongly enriched in nickel. These deposits account for about 25% of the world’s nickel reserves, and the associated mining industry forms a critical part of the local economy. The Central Chain

The oldest rocks in New Caledonia are the PermianJurassic quartzo-feldspathic metasediments and metavolcanics of the central chain. These rocks are interpreted to have been deposited along an active continental margin and are thought to have been sourced from emergent land to the southwest at that time. They were accreted to the eastern margin of Gondwana by the early Cretaceous. The igneous rocks of the central chain include mainly tholeiitic basalts, dolerites, and gabbros, indicative of an active continental margin setting. Extrusive material from this suite is overlain by Senonian-age sedimentary sequences. The rocks of the central chain have undergone varying amounts of metamorphism, or burial and heating, and metamorphic grade ranges from low-pressure/lowtemperature prehnite-pumpellyite facies in the southwest to low-pressure/moderate-temperature greenschist facies with in the northeast. In this region some localized highpressure/low-temperature metamorphic assemblages containing the minerals glaucophane and lawsonite have also been identified, which may be indicative of collision- or

subduction-related tectonic activity. Metamorphism occurred between 152 and 132 million years ago and has been interpreted to be associated with either (1) southwest dipping subduction beneath the Gondwana margin or (2) uplift and emergence of the New Caledonian core region, which at this time is thought to have been attached to the eastern margin of Gondwana. The Western Coastal Belt

The western coastal belt contains a diverse mixture of rock types. Deformed rocks of Jurassic and older age are overlain by Cretaceous sediments associated with deepening sea levels (or transgression), which were probably deposited contemporaneously with the opening of the Tasman Sea. The stratigraphy and palaeontology of these sequences closely resemble rocks of the similarly aged Murihiku terrane in New Zealand, and both are interpreted to be part of a marginal Gondwana sedimentary package. These in turn are overlain by Eocene conglomerates and shallow-water sequences, as well as Upper Eocene flysch sequences. Basaltic igneous material overlies rhyolitic tuffs and is intercalated with marine sediments varying in age from Senonian to late Eocene. Dating of basalts from the western coastal belt indicates that they were erupted between 42 and 59 million years ago; however, the lower age limit is absolutely constrained by emplacement of the ophiolite in the late Eocene. Basalts of the Poya Formation are typically mid-ocean ridge type. The temporal relationship between these basalts to the rhyolitic tuffs is interpreted to indicate basalt eruption occurred shortly after the rift formed during the late Cretaceous and that rifting continued at least until the early Tertiary. The Eastern Coastal Belt

The eastern coastal belt includes a thin strip of sediments and volcanics that separate rocks of the Central Chain from the east coast of Grande Terre. These rocks are Cretaceous to Eocene in age and are broadly similar to those of the western coastal belt. Based on age and stratigraphic affinities they are also generally considered to have formed in a similar tectonic environment. Also included in the eastern coastal belt is the root zone of the ophiolitic nappe complex described above. The Northern Region

The northern region of Grande Terre is dominated by variably metamorphosed metasedimentary and metavolcanic sequences associated with an Eocene-aged highpressure/low-temperature metamorphic belt. This region

NEW CALEDONIA, GEOLOGY

647

contains one of the world’s largest, most spectacular and continuous exposures of high-pressure (blueschist- and eclogite-facies) metamorphic rocks in the world. The western part of the belt is dominated by fine-grained siliceous siltstones (phtanites) and limestones that are variably metamorphosed from prehnite-pumpellyite facies to lawsonite facies, increasing in grade to the northeast. The eastern part of the belt, or Pouébo terrane, contains a mixture of mafic material (derived from oceanic precursors) and more siliceous material (with continental origins) that was metamorphosed to blueschist or eclogite facies during the Eocene. The western and eastern sectors of the northern region are geologically distinct; however, they do represent different levels of the same structural and metamorphic zone. The northern region is strongly dissected by numerous generations of brittle faults and shear zones that formed at various stages during the exhumation history of this once deeply buried package of rocks. High-pressure/low-temperature metamorphic rocks are important because they typically form in subduction zones where cold oceanic crust is forced down to great depths (and pressures) as it is overridden by another crustal block. The eclogite facies rocks in northern New Caledonia are believed to have been metamorphosed at pressures of 20 kbar, which equates to burial to depths of 60 km or more. These rocks were then transported via some (still relatively poorly understood) tectonic process back to the Earth’s surface, where they are now exposed in mountainous ridges high above sea level. This all happened in a geologically short period of time (on the order of 20–40 million years).

SEE ALSO THE FOLLOWING ARTICLES

New Caledonia, Biology / Pacific Region / Plate Tectonics / Vanuatu

Loyalty Islands

ADDITIONAL READING

The Loyalty Islands are a series of four emergent islands (Ouvéa, Lifou, Maré, and Walpole) that lie on the Loyalty Ridge, 100 km to the east of Grande Terre. The Loyalty Ridge probably represents an extinct volcanic arc associated with the Eocene subduction described above. The islands are probably volcanic, but, apart from a small outcrop of basalt on Maré, the basement rocks have not been identified because they are overlain by the thick coral and limestone sequences that have developed on top of the islands.

Aitchison, J. C., G. L. Clarke, D. Cluzel, and S. Meffre. 1995. Eocene arc– continent collision in New Caledonia and implications for regional SW Pacific tectonic evolution. Geology 23: 161–164. Black, P. M., and R. N. Brothers. 1977. Blueschist ophiolites in the melange zone, Northern New Caledonia. Contributions to Mineralogy and Petrology 65: 69–78. Clarke, G. L., J. C. Aitchison, and D. Cluzel. 1997. Eclogites and blueschists of the Pam Peninsula, NE New Caledonia: a reappraisal. Journal of Petrology 38: 843–876. Cluzel, D., J. C. Aitchison, G. L. Clarke, S. Meffre, and C. Picard. 1994. Point de vue sur l’évolution tectonique et géodynamique de la Nouvelle-Calédonie. Comptes Rendus de l’Academie des Sciences, Serie II 319: 683–690. Rawling, T. J., and G. S. Lister. 1999. Oscillating modes of orogeny in the Southwest Pacific and the tectonic evolution of New Caledonia, in Exhumation processes: normal faulting, ductile flow, and erosion. U. Ring, M. T. Brandon, G. S. Lister, S. D. Willett, eds. Special Publication 154. London: Geological Society of London, 101– 102. Schellart, W. P., G. S. Lister, and V. G. Toy. 2006. A Late Cretaceous and Cenozoic reconstruction of the Southwest Pacific region: tectonics controlled by subduction and slab rollback processes. Earth-Science Reviews 76: 191–233.

Evolution of the High-Pressure Schist Belt

Since plate tectonic theory was first applied to the Southwest Pacific, it has been recognized that a detailed understanding of the evolution of New Caledonia’s highpressure schist belt was critical to any tectonic interpretation or reconstruction of the region.

648

The most recent tectonic models for the evolution of the region suggest that a collision took place between two segments of oceanic crust, resulting in east-to-northeastdirected subduction beneath the Loyalty Basin. This subduction was interrupted by the arrival of a sliver of continental crust from the west. This material was partially subducted, resulting in stacking of continental thrust slices at depth and crustal thickening. Buoyancy of this material resulted in failure of the subduction zone and emplacement of the ophiolite from the northeast via obduction. Exhumation was accomplished by a combination of buoyancy-driven diapiric uplift, erosion, and tectonic unroofing. Subsequent to the collision, late-stage extensional tectonism resulted in the formation of a series of low-angle ductile shear zones that caused the unroofing and rapid exhumation of the region and, ultimately, the exposure of the blueschists and eclogite rocks. New Caledonia’s steep topography, its high elevation, and a number of geomorphologic clues (such as stranded rivers, elevated coral reefs, and weathering surfaces) indicate that tectonic processes are actively maintaining the relief of the island. This uplift does not appear to be related to any active faulting, because there is very little record of recent seismicity beneath the island. It is likely, however, that this relief is being maintained at least in part by the lithospheric bulge associated with the subduction of the Australian plate in the New Hebrides subduction zone several hundred kilometres to the east.

NEW CALEDONIA, GEOLOGY

NEWFOUNDLAND HAROLD WILLIAMS Memorial University, St. John’s, Canada

The island of Newfoundland is a textbook example of a collisional geologic mountain belt. It formed through the opening and closing of an ancient Iapetus Ocean, which preceded the modern North Atlantic. The cycle of opening and closing lasted for about 300 million years. The major geologic divisions of Newfoundland represent the margins and vestiges of Iapetus. GEOGRAPHIC SETTING

The island of Newfoundland forms the northeast extremity of the Appalachian Mountain Belt, or Appalachian Orogen, in eastern North America (Fig. 1). The northeast coastline of the island displays a complete cross section of the Appalachians in superb wave-washed cliff

exposures. Since the wide acceptance of continental drift and plate tectonics, many of the concepts of geological mountain building, especially those of continental collisions, originated in Newfoundland. The island exhibits many of the best examples of plate tectonic processes in the remote geological past. Some of its rocks, such as those of Gros Morne National Park, are designated by UNESCO as World Heritage; others, such as those of Fortune Head and Green Point, are the world’s typeexamples of certain geological boundaries—in this case, the Precambrian–Cambrian and Cambrian–Ordovician, respectively. Many other rocks and fossil localities are of provincial heritage status. MAJOR GEOLOGIC DIVISIONS

The major divisions of rocks of the Newfoundland Appalachians, from west to east, are the Humber, Dunnage, Gander, and Avalon Zones (Fig. 2). These geological divisions in Newfoundland represent the continental margins and vestiges of an ancient Iapetus Ocean that opened and closed between 600 and 300 million years ago. The

FIGURE 1 Extent of the Appalachian Mountain Belt in eastern North America and interpretation of its major geological divisions.

NEWFOUNDLAND

649

FIGURE 2 Generalized interpretive map of the Newfoundland Appalachians. Modified after a compilation by J. P. Hayes, 1987, and digitized by

T. Paltanavage, 1994.

Newfoundland zones have been extended the full length of the Appalachian Mountain Belt from Newfoundland 3000 km southward to Alabama. The closure of Iapetus resulted in the assembly of a supercontinent, Pangea. The break-up of Pangea, beginning 250 million years ago, gave rise to the modern Atlantic Ocean. Opening of the Atlantic Ocean dispersed segments of the Appalachians and correlatives to the North Atlantic borderlands of northwestern Africa, Europe, Scandinavia, and eastern Greenland (Fig. 3).

650

NEWFOUNDLAND

HUMBER ZONE

The Humber Zone represents the ancient continental margin of eastern North America or the western margin of Iapetus. The Dunnage Zone represents vestiges of Iapetus. The Gander Zone represents the eastern margin of Iapetus, and the Avalon Zone originated somewhere east of Iapetus and is of African affinity. Rocks and structures of the Humber Zone fit the model of an evolving continental margin and spreading Iapetus Ocean. This began with rifting of existing con-

tinental crust at about 600 to 550 million years ago. The rifting is evidenced by liquid injections that filled cracks in the older crust and fed volcanic eruptions. It also led to the deposition of coarse fragmental sedimentary rocks. This was followed by the development of a passive continental shelf with mainly limestone deposition, like that of the present Bahamas, with contemporary continental slope/rise deposits. After about 100 million years, this ended with the deposition of coarse sandstones that are a harbinger of forthcoming catastrophic events. Destruction of the Humber margin is marked by the transport of rocks from the compressed uplifted continental slope and rise landward above the former continental shelf. These transported slope/rise rocks are, in turn, structurally overlain by slabs of Iapetan oceanic crust and mantle, such as the Tablelands of Gros Morne National Park. The western boundary of the Humber Zone is drawn where deformed rocks of the Appalachians pass into flat-lying rocks of the continental interior. The eastern boundary of the Humber Zone is a steep belt marked by discontinuous occurrences of Dunnage Zone ocean crust and underlying mantle. DUNNAGE, GANDER, AND AVALON ZONES

The Dunnage Zone is recognized by its abundant volcanic rocks and oceanic crust and mantle rocks. It also contains chaotic mixtures of discrete resistant blocks surrounded by shales. Sedimentary rocks are all of deep marine deposition. The Dunnage Zone is the widest and best preserved in northeastern Newfoundland because of matching morphological embayments in the opposing margins of Iapetus. It is narrow or absent in southwestern Newfoundland, where matching promontories took the brunt of the collision. The boundary between the Dunnage and Gander Zones is also marked by the occurrence of Iapetan oceanic crust and mantle rocks. The Gander Zone has a thick, monotonous sequence of sandstones, siltstones, and shales that grade eastward into deformed and altered rocks. Its analysis is far less sophisticated than that of the Humber Zone. Almost half of its rocks are granitic intrusions, and half of the remainder are deformed and altered beyond recognition. The Avalon Zone is defined by its well-preserved sedimentary and volcanic rocks, dated at about 650 to 550 million years. Overlying shales contain a trilobite fauna completely different from trilobites in equivalent rocks of the Humber Zone. The Avalon Zone extends 600 km offshore, making it the broadest zone of the entire

FIGURE 3 Restored North Atlantic region showing the spreading axis

of the imminent North Atlantic Ocean and dispersed correlative rocks (blue and yellow colors) of the Appalachian mountain belt.

NEWFOUNDLAND

651

Appalachians, more than twice the combined width of all other zones. The western boundary of the Avalon Zone is a major fault. LATER HISTORY

Rocks younger than about 400 million years overlie those of the fundamental zones. They are sedimentary and volcanic rocks that show an upward change from marine to terrestrial, with all rocks deformed together and cut by granites. These changes mark the final closing phases of Iapetus. After Iapetus closure, the youngest Appalachian rocks are everywhere the same, mainly subaerial red and gray sandstones, coal measures, shallow marine limestones, and salt deposits. OTHER FEATURES

Rocks of the fundamental zones and overlying deposits record the full history of the Iapetus cycle. Apart from kinds of rocks and structures, zones are expressed also by geophysics, paleontology, metallogeny, plutonism, metamorphism, isotopic signatures, and other features. Younger rocks show less contrast because they were deposited during the dying phases of the Iapetus cycle or after its closure.

652

CONCLUDING REMARKS

Newfoundland has attracted the attention of geologists worldwide, as well as other prominent dignitaries. Visitors are encouraged to visit Gros Morne World Heritage Site as well as the World Heritage Viking Site at nearby L’Anse aux Meadows. The Johnson GeoCentre in the capital of St. John’s is also a place of more than passing geological interest. SEE ALSO THE FOLLOWING ARTICLES

Atlantic Region / Britain and Ireland FURTHER READING

Williams, H., ed. 1995. Geology of the Appalachian-Caledonian Orogen in Canada and Greenland. Geological Survey of Canada, Geology of Canada, no. 6 (also Geological Society of America, The Geology of North America, v. F-1). Williams, H., S. A. Dehler, A. C. Grant, and G. N. Oakey. 1999. Tectonics of Atlantic Canada. Geoscience Canada 26: 51–70.

NEW GUINEA, BIOLOGY ALLEN ALLISON

RELATIONSHIPS BETWEEN THE NORTH ATLANTIC AND IAPETUS

Bishop Museum, Honolulu, Hawaii

The North Atlantic Ocean and its margins provide an actualistic model for the Iapetus Ocean, the destruction of which led to the Appalachian Mountain Belt. Just as Atlantic rifting involved a broad area of several hundred kilometers, so too did Iapetan rifting extend well inland toward the continental interior. The transition from rifting to continental drifting at the Atlantic margin, defined by seismic reflection, deep drilling, and the age of adjacent oceanic crust, has an Iapetan counterpart in the Appalachian Humber Zone. The widths of the North Atlantic shelf/slope/rise and the thickness of modern sediments are comparable to restored widths of the Humber Zone and thicknesses of its sedimentary rocks. The offshore form of the North Atlantic margin at the Grand Banks mimics an Iapetan promontory in the Gulf of St. Lawrence and provides an explanation for the sinuosity of the Humber Zone along the Appalachian Orogen. The crust and mantle beneath the North Atlantic are analogous to Iapetan volcanic rocks and its oceanic crust and mantle rocks of the Dunnage Zone, and Atlantic micro-continents and oceanic volcanic islands and sea mounts are typical of some Iapetan unexplained terranes.

New Guinea, the world’s largest and highest tropical island, was one of the last parts of the globe to be explored, earning it the nickname “the last unknown.” Although many of its species have yet to be scientifically named, we do know that is inhabited by an extraordinarily rich assemblage of plants and animals, derived from both Southeast Asia and Australia, with diversity exceeding that of the much larger Australian continent and rivaling that of the Amazon Basin. Overall it has ∼8% of the world’s biota. At least 70% of species are endemic.

NEW GUINEA, BIOLOGY

GEOGRAPHICAL COVERAGE

In this treatment, New Guinea includes the main island and associated satellite islands, such as the Raja Ampat group, the islands in Cenderawasih Bay, and the Aru Islands in the west, and in the east the d’Entrecasteaux and Louisiade groups together with Woodlark Island and the Trobriand Islands. The western half of mainland New Guinea and satellite archipelagoes are politically part of Indonesia. The eastern half, together with archipelagoes to the north and east (which are not included in this treatment), are comprised in the sovereign state of Papua New Guinea.

GEOLOGICAL SETTING

Elements of New Guinea began forming in the late Cretaceous at the leading edge of the Australian tectonic plate. As this plate separated from Antarctica at the beginning of the Cenozoic, it began moving northward, colliding with a complex subduction system that included island arcs, oceanic plateaux, seamounts, and plate fragments. Today, New Guinea is a geological composite consisting of at least 32 separate terranes. The evolutionary history of the biota is linked to the accretion of these terranes to the Australian craton, and to the uplift, volcanism, and rifting that accompanied these tectonic events. The island can be naturally divided into five main biogeographic provinces, based on geological origin of these regions (Fig. 1): (1) the Australian craton, (2) fold belt (leading edge of the Australian craton), (3) accreted terranes, (4) the Vogelkop composite terrane, and (5) the East Papuan composite terrane. The Aru Islands, located off the south coast of New Guinea, have important biotic differences from the New Guinea mainland, particularly for fishes, and are sometimes recognized as a separate biogeographic province, although part of the Australian craton. Similarly, some of the the Raja Ampat islands at the western tip of New Guinea and a sliver of the northern Vogelkop have a separate geological origin from the adjacent Vogelkop Peninsula and are sometimes treated as a separate province. Although many genera are widespread, a conspicuous number of genera, or species groups within genera, are restricted to single biogeographic provinces. In addition, the fold belt, which comprises the mountainous spine of the island, has tended to promote speciation through the creation of extensive montane habitat and by dividing the north and south coast lowlands.

The composition of the New Guinea flora is generally similar to that of Southeast Asia, and the two areas are often floristically grouped together into a region called Malesia. However, the New Guinea flora tends to have fewer plant families and a much higher proportion of endemic genera and species than do other parts of Malesia. New Guinea is also noteworthy in having a high proportion of orchids and ferns; collectively these groups make up about a quarter of the named plant species. The orchids, which are represented by more than 3000 species, comprise the largest family of flowering plants in New Guinea. The madder family, Rubiaceae, is next, with about 1000 species, followed by the grasses (Poaceae), spurges (Euphorbiaceae [sensu lato]) and palms (Arecaceae). Animals

The vertebrates are far better understood than invertebrates or plants, but many groups remain incompletely known. Although the New Guinea vertebrates form a rich assemblage, many lineages found in Southeast Asia are missing from New Guinea. For example, there are only four major groups, comprising six orders, of terrestrial mammals (monotremes, marsupials [three orders]), rodents, and bats) in New Guinea, in contrast to 10–11 orders in most of Southeast Asia. Only one of the three orders of amphibians (frogs) occurs in New Guinea. A number of bird families that are widespread in Asia (e.g., woodpeckers) do not occur in New Guinea. The largest vertebrate group in New Guinea is the marine fishes, which make up ∼62% of the New Guinea

COMPOSITION AND RICHNESS OF THE BIOTA

New Guinea was, and arguably still is, one of the world’s great biological unknowns. Although early collections were made from coastal areas beginning in the late 1700s, it was not until the mid-1870s that biologists first penetrated the mountainous interior. Large parts of the island still remain unexplored, particularly in the west. It is likely that upwards of half the biota remains unknown to science. Plants

There are no comprehensive treatments of the flora of New Guinea, and the number of vascular plant species is the subject of considerable controversy, with estimates ranging from 11,000 to more than 30,000.

FIGURE 1 Main geological components of New Guinea (based on vari-

ous sources, particularly Pigram and Davies 1987; Hill and Hall 2003).

NEW GUINEA, BIOLOGY

653

vertebrate fauna. Freshwater and brackish fish species comprise nearly 8%. Birds make up nearly 15% of New Guinea vertebrates, followed by amphibians and reptiles at nearly 10%. Mammals account for only 5% of the total. If we focus on land and freshwater vertebrates (i.e., exclude marine and brackish-water fishes as well as other exclusively marine vertebrates such as sea turtles and sea snakes), there are 1759 land and freshwater vertebrates known from New Guinea; 1247 (71%) of these are endemic. The level of endemism ranges from a low of 50% for crocodiles, which tend to have large geographic ranges, to a high of 93% for frogs, which have limited dispersal capacity and are represented by a high percentage of restricted-range species. Most native species that are not endemic to the island of New Guinea are shared with Australia and to a lesser extent with Maluku, Southeast Asia, or the Bismarck Archipelago and Solomon Islands. A few indigenous species also have extensive ranges in the Pacific Basin. New Guinea vertebrates comprise 8% of currently recognized world vertebrate species and range from about 4% of the world total for snakes and lizards to a high of nearly 11% for fishes (Table 1). If we restrict the comparison to land and freshwater vertebrates, including fishes, the New Guinea proportion drops to about 4% of the world total; it is 6.5% if fishes are excluded. The freshwater fish fauna is relatively small, with about 400 species known, representing less than 1% of world fishes. A few of these are primary-division fishes (species belonging to families that have always been intolerant of saltwater), but most are thought to have evolved from TABLE 1

The Vertebrate Fauna of New Guinea in Comparison with That of the World New Guinea Taxon

New Guinea

World

as % of World

Fishes Frogs Turtles Crocodiles Lizards Snakes Birds Mammals

3,200 314 18 2 208 113 553 341

30,000 5,500 308 23 5,002 3,113 9,704 5,415

10.7 5.7 5.8 8.7 4.2 3.6 5.7 6.3

Total

4,749

59,065

8.0

note: Figures for fishes are from Allen (1991, 2007) and Allen and Swainston (1992). Amphibian and reptile data for New Guinea are from Bishop Museum (2007), and for the world are from Frost (2007). Data for world reptiles are from J. Craig Venter Institute (2007). Data for birds are from Mack and Dumbacher (2007). Mammal data are from Helgen (2007). Where necessary, figures have been updated to include recently described species.

654

NEW GUINEA, BIOLOGY

FIGURE 2 Choerophryne sp. This small treefrog is about 25 mm long

and inhabits mid-montane rain forest on the northeast coast of New Guinea.

FIGURE 3 Varanus salvadorii. This large lizard, which is widespread in

the lowlands of western and central New Guinea, can exceed 3 m in total length.

marine ancestors. About three-quarters of the freshwater fishes lack a marine larval stage, and of these more than 80% are endemic to New Guinea. The marine fishes present quite a different pattern and include an estimated 2600 to 3000 species, which represent ∼11% of the world total. About a third of the world’s reef fishes occur in New Guinea, although endemism is low and many species occur widely in the Indo-Pacific region. The frogs are probably the most poorly known group of New Guinea vertebrates. There are currently 314 species known from New Guinea, but this total will almost certainly double when all species have been scientifically named. Each major herpetofaunal survey during the past five years has turned up five to ten, and sometimes more, new species in a single drainage basin. All but 23 of New Guinea’s frogs are endemic. The indigenous species are generally shared with Australia. The lizards are somewhat better known than frogs, with 208 described New Guinea species. However, dozens of new species are known to reside in collections, and ongoing studies of several large genera are likely to result in the

recognition and description of many additional species. The overall total is likely to reach around 300 species. Lizards tend to have a higher capacity for dispersal than do frogs, and only about two-thirds of them are endemic. The indigenous species include widespread Indo-Pacific species as well as species that are shared only with Australia or Maluku. The New Guinea snake fauna is relatively depauperate, with 113 species. Only 87 of these are terrestrial; the rest are marine. About half the terrestrial species are endemic, while only two of the marine species are endemic. As a result of ongoing systematic studies and field surveys, the total number of New Guinea snake species is likely to reach about 140. The New Guinea herpetofauna includes a number of charismatic species, such as the largest tree frog in the world (Litoria infrafrenata), some of the world’s smallest frogs (genus Oreophryne), frogs with peculiar snouts (genus Choerophryne; Fig. 2), the world’s largest crocodile (Crocodylus porosus), and the world’s longest lizard (Varanus salvadorii; Fig. 3). Birds are generally the most conspicuous elements of the vertebrate fauna and, because of their vagility, tend to have larger geographic ranges than do the more sedentary amphibians and reptiles. They also have a much larger popular following and for these reasons are far better studied than other vertebrate groups. New species are still occasionally discovered, but the cumulative number of species for New Guinea has reached a plateau, and the current total of 579 resident species, of which 325 (56%) are endemic, is not expected to increase by much with further surveys. Perhaps the most famous group of birds in New Guinea are the birds of paradise (family Paradisaeidae; Fig. 4), which comprise 42 species, most of which are endemic. They have long fascinated naturalists because of the males have extremely colorful plumage and elaborate courtship displays. In addition, 11 of the world’s 18 species of bowerbirds (family Ptilonorhynchidae) occur in New Guinea. Bowerbirds, which also occur in Australia, are renowned for constructing elaborate structures called “bowers” that are used for courtship displays. New Guinea is also home to three species of cassowaries—large flightless birds distantly related to ostriches—as well as the world’s largest pigeon (Goura victoria) and, at the other extreme, the world’s smallest parrots (Micropsitta spp.). Another group of interesting birds in New Guinea are the pitohuis, six passerine bird species that contain neurotoxic alkaloids. These compounds, which are chemically related to batrachotoxin (found in Central and South American poison arrow frogs of the family Dendrobati-

FIGURE 4 Paradisaea raggiana. The raggiana bird of paradise is found

throughout much of eastern New Guinea and is a national symbol of Papua New Guinea.

dae), are thought to ward off ectoparasites and predators. Several species are brightly colored, suggesting that this is aposematic or warning coloration. Geographic races of other pitohui species sometimes mimic these brightly colored species, a phenomenon known as Müllerian mimicry, in which the mimic and model are both toxic, but gain a mutual advantage by sharing, and therefore reinforcing, the same anti-predator warning coloration and pattern. Mammals are much more poorly known than birds. More than half the ∼340 mammal species known from New Guinea have been described since 1900. At least two-thirds of these species are endemic. Recent taxonomic work has revealed the existence of many cryptic species, and field surveys have turned up dozens of undescribed taxa. It is likely that more than 100 species remain to be named. The monotremes, an ancient group of egg-laying mammals now restricted to Australia and New Guinea, are perhaps the most noteworthy elements of the vertebrate fauna (Fig. 5), followed by the marsupials, which include tree kangaroos and small rat-sized carnivores. There are at least two, and probably at least four, species of monotremes in New Guinea (one shared with Australia). There are also large radiations of bats and rodents. The major groups of insects and other invertebrate groups inhabiting New Guinea are similar to those found

NEW GUINEA, BIOLOGY

655

in which the males joust for access to females for mating. There are also long-lived, high-elevation beetles (genus Gymnopholus) that have pits on their dorsum that support a growth of algae and other primitive plants, providing excellent camouflage for the host, which would otherwise be highly visible in its mossy forest habitat. Rotifers, protozoans, and other microorganisms live within the plants, a phenomenon termed episymbiosis. DISTRIBUTION PATTERNS Biogeographical Patterns FIGURE 5 Zaglossus bruijni. One of three monotremes or egg-laying

mammals in the world, this species is endemic to New Guinea and is found in montane regions throughout much of the island.

in Southeast Asia. Although there are no comprehensive checklists for most groups of invertebrates, an estimated 80,000 species are known. This probably represents less than a quarter of the actual total, and it is likely that the overall number exceeds 300,000 species. However, a few groups are relatively well known, and some trends can be identified. In the best-known group, butterflies, the descriptions of new species appears to be approaching an asymptote, suggesting that New Guinea has around 960 species or 5.5% of the world total. The Odonata, which includes dragonflies and damselflies, are also relatively well known at a world level. New Guinea has around 580 species, or 9.4% of the world total. Aquatic Heteroptera, which include nearly 5000 species worldwide, have been well surveyed in New Guinea and include around 350 species, or about 7% of the world total. Approximately 70–80% of New Guinea insects are endemic. Recent ecological studies have demonstrated that trees in New Guinea and those in temperate regions support similar numbers of insect species, and that levels of host-specificity are similar in both areas. This suggests that the high diversity of insects found in New Guinea, and the tropics generally, may be related to much higher levels of plant diversity found in the tropics compared to temperate regions (a sevenfold difference). Interestingly, these recent studies have also demonstrated that herbivorous insects tend to feed on clades of closely-related species (e.g., genera) and are not, as had been commonly assumed, mostly restricted to individual tree species. Some of the particularly noteworthy elements of the New Guinea insect fauna include the world’s largest butterflies and moths (the Queen Alexandra’s Birdwing, Ornithoptera alexanderae, and the atlas moth, Attacus atlas, respectively); stick insects reaching 20 cm or more in total length), and the bizarre stalk-eyed and moose-antlered flies

656

NEW GUINEA, BIOLOGY

The biogeographic affinities of New Guinea vertebrates are complex. Repeated land connections during the late Cenozoic facilitated faunal exchange between New Guinea and Australia, resulting in many similarities of the vertebrate faunas of these two areas, particularly at the level of genus and higher. This is particularly true for mammals, for which marsupials and monotremes are such important and distinctive components of the fauna, but is also true for many groups of birds and other vertebrates. For example, of the 119 genera of amphibians and reptiles that occur in New Guinea, 56 (47%) also occur in Australia. New Guinea also has a strong affinity with Southeast Asia, but oceanic barriers have always separated it from that region, as demarcated by the famous “Wallace’s Line.” These oceanic barriers, which prevented a wide array of vertebrate lineages from reaching New Guinea, have been less of a barrier to plants and invertebrates, which are largely of Southeast Asian origin. The geological history of New Guinea has strongly influenced the distribution of plants and animals. Its many isolated mountain ranges and lowland basins have produced pockets of endemism, particularly beyond the northern margins of the Australian craton, where, for example, lizard species richness is greatest. In contrast, the central mountains have the richest assemblages of frogs. The separation of northern and southern lowlands by the central mountains also strongly influences biotic distributions, even in relatively recently evolved groups such as the seven birds of paradise in the genus Paradisaea, which are probably are no more than 500,000 years old (Fig. 6). Ecological Patterns

Although mean temperatures do not show much seasonality, temperatures are strongly influenced by elevation, ranging from a mean of around 26 °C in the lowlands down to 5–7 °C on the high peaks. These differences in

temperature, together with considerable geographic variation in the total amounts and seasonal patterns of rainfall, have combined with the geological history to produce a rich array of habitats and ecological diversity. Focusing on vegetation, which is closely linked to climate and substrate, much of the coast, particularly in the south, is fringed by mangroves, which make up about 4% of forested areas. Other broadleaf forests—mostly rain forests—cover about two-thirds of the island. These forests, which are floristically diverse and occur to around 3000 m, generally decrease in canopy height and tend to have fewer lianas and a lower incidence of tree buttressing with increasing elevation. A band from about 3000 to 3200 m is dominated by shrubs and small trees. There is a shrub-grassland mosaic from 3200 to 3400 m. Areas above 3400 m and extending to about 4200 m are dominated by grasslands, sometimes with pockets of shrubbery, particularly on slopes. Areas above 4200 m, which include a total of area of 5000 km2, support a tundra-like vegetation or are covered in bare rock. The highest mountain in New Guinea, Puncak Jaya (4884 m), supports a permanent glacier. Large parts of southern New Guinea, south of the Fly and Digul Rivers and covering about 15% of the island area, have strongly seasonal rainfall patterns and are covered in woodland savanna or seasonally deciduous monsoon forest. Similar forests are found along the southeast coast of Papua New Guinea, and patches extend nearly to the eastern tip of the island and comprise another 5% of the island’s land area. More than 31% of the land area of New Guinea is above 1000 m, and nearly 2% is above 3000 m elevation. Many of the mountains are the result of rapid uplift during the Tertiary period. This has tended to promote speciation along elevational gradients in which closely related species occupy contiguous and often mutually exclusive elevational ranges. This is seen, for example, in the lizard genus Papuascincus. One species, P. morokanus, occurs from about 900 m to 1800 m throughout much of montane New Guinea. A second species, currently unnamed, occurs from 1800 m to nearly 2600 m. A third species, P. stanleyanus, occurs from about 2000 m to nearly 2800 m. Similarly, three species of acanthizid mouse-warblers of the genus Crateroscelis occur throughout much of New Guinea. Crateroscelis murina occurs from sea level to about 1700 m. A montane species, C. robusta, generally occurs from 1750 to 3600 m. A third species, C. nigrorufa, which has an extremely patchy distribution, occupies an intermediate altitudinal band from 1300 to 2000 m.

FIGURE 6 Distribution of the seven species of birds of paradise of the

genus Paradisaea in New Guinea (data from Frith et al. 1998). The red bird of paradise, P. rubra, is endemic to islands of Waigeo and Batanta in the Raja Ampat group at the western trip of New Guinea. Adjacent areas of the New Guinea mainland, such as the Vogelkop Peninsula, are occupied by the lesser bird of paradise, P. minor, which also occurs throughout much of western New Guinea. The greater bird of paradise, P. apoda, occupies southwestern New Guinea including the Aru Islands. The Raggiana bird of paradise, P. raggiana, occupies much of eastern New Guinea. The blue bird of paradise, P. rudolphi, occupies montane regions of central New Guinea. The final two species have restricted ranges: the emperor bird of paradise, P. guilielmi, is endemic to the Huon Peninsula of northern Papua New Guinea; and Goldie’s bird of paradise, P. decora, is endemic to Fergusson and Normanby islands in the Entrecasteaux Archipelago near the eastern tip of New Guinea. Both these areas were created by relatively recent geological uplift.

COMPARISON OF NEW GUINEA AND BORNEO

In order to put the biota of New Guinea into a regional perspective, it is useful to compare it with the island of Borneo. The second largest tropical island in the world, Borneo is moderately smaller than New Guinea (743,330 km2 vs. a land area of 790,000 km2 for New Guinea). Both islands have high mountains (>4000 m) and formed through complex processes of accretion of island arcs, oceanic crustal material, and other tectonic fragments onto Paleozoic continental cores. Borneo began forming during the Mesozoic and was connected by land to mainland Southeast Asia until at least the Eocene; it has subsequently been reconnected to that region for varying amounts of time coinciding with periods of lowered sea level during the Tertiary and Quaternary. It is today topographically far less complex than New Guinea, with wide expanses of lowlands surrounding the central mountains. The occurrence of past land connections between Borneo and continental Asia, the uniformity of the climate throughout the region during the Tertiary, and the absence of much topographic relief has resulted in the

NEW GUINEA, BIOLOGY

657

fauna of Borneo being rather similar to that of other parts of SE Asia. New Guinea, by contrast, with its more complex geological history, its relative isolation from continental areas with similar climates, its greater topographic diversity, and the impact of these factors on the evolutionary diversification of species, has developed a richer and more endemic biota than has Borneo. However, Borneo is generally richer in higher taxa or lineages (e.g., families and orders). The two islands are thought to have about the same number of plant species, approximately 14,000. Forests in Borneo have been shown to be exceedingly rich in numbers of species; by some accounts they are the richest forests in the world. New Guinea’s forests are less rich but have a much higher proportion of endemic plant species than do Bornean forests. New Guinea is much more poorly known than Borneo and it is likely that with further collecting the total number of plants occurring there will ultimately exceed the total for Borneo. There are 11 orders and at least 219 species of mammals from Borneo; 46 (21%) species are endemic. New Guinea has six orders of mammals (monotremes, three orders of marsupials [Dasyuromorphia, Peramelemorphia, and Diprotodontia], rodents, and bats) with 341 species, of which 252 (74%) are endemic. A total of 434 species of breeding birds are known from Borneo, but only 39 of these are endemic. New Guinea. by contrast. has 579 species of breeding birds, of which 325 species are endemic. Borneo has a higher number of families and genera of amphibians and reptiles (30) than does New Guinea (23) and, in addition, has caecilians, an amphibian order that is absent from New Guinea. At the species level Borneo also has a higher number of snakes (157) than does New Guinea (113), reflecting the importance of this group in Southeast Asia. However, it has only 150 species of frogs and 117 species of lizards. In comparison New Guinea has 314 frogs and 208 lizards. Overall Borneo has only about two-thirds the number of species that are found in New Guinea. Borneo and New Guinea both have a similar number of freshwater fishes, about 400 species, but around a quarter of the freshwater ichthyofaunas of both areas are composed of fishes with a marine larval stage. Borneo is dominated by primary-division fishes; about 40% of these species are endemic. In contrast, New Guinea essentially lacks primary-division fishes, and most species of its freshwater fishes are thought to have evolved from marine ancestors; approximately 180 (84%) of currently recognized New Guinea freshwater species lacking a marine larval stage are endemic. In general, species richness in Borneo tends to be higher within a drainage basin than in New Guinea. 658

NEW GUINEA, BIOLOGY

For example, the large Kapuas River system in western Kalimantan has 320 species. By contrast, the Fly River drainage in New Guinea has fewer than 110 species. The New Guinea region is estimated to harbor around 2600 to 3000 species of marine fishes, including ∼30% of the world’s reef fishes; Borneo likely has a comparable number of marine species. Many of the marine species are widespread in the Indo-Pacific region and are found in both Borneo and New Guinea. Species richness of the marine biota of the Indo-Australian region is among the highest is the world. These differences tend to highlight the fact that while New Guinea has been colonized by fewer evolutionary lineages than has Borneo, the complex geological history of New Guinea and its topographic diversity has resulted in enormous adaptive radiations of species that have produced much higher levels of species richness in New Guinea than have been attained in Borneo. HUMAN IMPACTS AND FUTURE CONSIDERATIONS

New Guinea has been inhabited by humans for 50,000 years. Agriculture first developed there some 6000 years ago, about the same time as food plants were first cultivated in the Fertile Crescent of Asia. Initial human population density in New Guinea was low and is still only around 10–15 people per square kilometer, less than half that of the United States. About a third of New Guinea’s human population of about nine million people is concentrated along the coastal fringe, where slash and burn is the prevailing type of agriculture, and the other two-thirds are in highland areas, where large valleys are covered in grassland, likely the result of frequent burning to clear fields for agriculture. However, human impacts tend to be localized, and overall less than 1% of the island is covered in urban or built-up areas or agricultural systems; vast areas of the island are virtually uninhabited and seemingly pristine. This has led to a persistent belief among biologists, conservation professionals, and others that New Guinea is in no immediate conservation peril. However, huge tracts of old-growth forests are being lost to timber harvesting and mining. Approximately half of forests suitable for logging (on slopes < 30%) have been logged, and logging concessions have been granted for much of the rest. Land conversion to cash-crop agriculture, particularly oil palm plantations, and to urban development are taking an increasing toll. The introduction of alien species, illegal collecting driven by the pet trade, and other factors are also endangering elements of New Guinea’s biodiversity.

The 1992 Papua New Guinea Conservation Needs Assessment and the 1997 Irian Jaya Conservation PrioritySetting Workshop helped to focus attention on areas of endemism in New Guinea. Subsequent work has largely tended to confirm the importance of areas identified in these exercises and highlighted the need for additional field surveys. There is a compelling, indeed crucial, need for a comprehensive biological survey of the island to better document the biota and to guide and inform the designation of protected areas to help preserve some of the richest assemblages of biodiversity on the planet.

NEW GUINEA, GEOLOGY HUGH L. DAVIES University of Papua New Guinea

Borneo / New Guinea, Geology / Wallace’s Line

The island of New Guinea is made up of elements of Australian and Pacific geology. It is in a dynamic part of the world and has been the subject of benchmark studies into plate tectonics, ophiolite, rifting of continents, Quaternary sea levels, and other fields. Because the geological history of the island is relatively short, it is more readily deciphered than for older regions of Earth’s crust.

FURTHER READING

PHYSIOGRAPHY

Beehler, B. M., T. K. Pratt, and D. A. Zimmerman. 1986. Birds of New Guinea. Princeton, NJ: Princeton University Press. d’Abrera, B. 1977. Butterflies of the Australian region. Melbourne: Lansdowne Press. Flannery, T. F. 1995. Mammals of New Guinea. Ithaca, NY: Comstock/ Cornell. Gressitt, J. L. 1982. Biogeography and ecology of New Guinea. Monographiae Biologicae. The Hague: W. Junk. Marshall, A. J., and B. M. Beehler, eds. 2007. The ecology of Papua. Singapore: Periplus Editions. Novotny, V., P. Drozd, S. E. Miller, M. Kulfan, M. Janda, Y. Basset, and G. D. Weiblen. 2006. Why are there so many species of herbivorous insects in tropical rainforests? Science 313: 1115–1118.

New Guinea lies across the northern margin of Australia. It is the second largest island in the world, 2200 km long and up to 750 km wide, and one of the most mountainous, with peaks to almost 4900 m above sea level. A central mountain range runs the length of the island and is bounded to the north by lesser mountain ranges and plains and to the south by a broad plain. The Mamberamo and Sepik rivers drain the north side of the central range, and the Digul and Fly rivers drain the south (Fig. 1). Beyond the southern plains a broad, shallow shelf extends to the Australian coast. Other shorelines are steeper, and some are bounded by deep-sea trenches (Fig. 2). Small ocean basins lie to the northeast and southeast and a great submarine plateau (Ontong Java Plateau) lies to the extreme northeast, beyond the islands of the Bismarck Archipelago. Smaller submarine plateaus lie south of the eastern part of New Guinea. Politically, the island is divided between the independent state of Papua New Guinea (PNG) in the east and

SEE ALSO THE FOLLOWING ARTICLES:

REFERENCES

Allen, G. R. 1991. Field guide to the freshwater fishes of New Guinea. Madang, Papua New Guinea: Christensen Research Institute. Allen, G. R. 2007. Fishes of Papua, in The ecology of Papua: Part 1. B. M. Beehler and A. J. Marshall, eds. Singapore: Periplus Editions, 637–653. Allen, G. R., and R. Swainston. 1992. Reef fishes of New Guinea: a field guide for divers, anglers and naturalists. Madang, Papua New Guinea: Christensen Research Institute. Bishop Museum. 2007. Papuan herpetofauna: a Bishop Museum project. http://www.bishopmuseum.org/research/pbs/papuanherps/. Frith, C. B., B. M. Beehler, and W. T. Cooper. 1998. The birds of paradise: Paradisaeidae. Oxford: Oxford University Press. Frost, D. R. 2007. Amphibian species of the world: an online reference. Version 5.1 (October 10, 2007). http://research.amnh.org/herpetology/ amphibia/index.php. Helgen, K. M. 2007. A taxonomic and geographic overview of the mammals of Papua, in The ecology of Papua: Part 1. B. M. Beehler and A. J. Marshall, eds. Singapore: Periplus Editions, 689–749. Hill, K. C., and R. Hall. 2003. Mesozoic–Cenozoic evolution of Australia’s New Guinea margin in a West Pacific context, in Evolution and dynamics of the Australian plate. R. R. Hillis and R. D. Müller, eds. Special Paper 372. Boulder, CO: Geological Society of America, 265–290. J. Craig Venter Institute. 2007. The reptile database. http://www.tigr.org/ reptiles/search.php. Mack, A., and C. Dumbacher. 2007. Birds of Papua, in The ecology of Papua: Part 1. B. M. Beehler and A. J. Marshall, eds. Singapore: Periplus Editions, 654–688. Pigram, C. J., and H. L. Davies. 1987. Terranes and the accretion history of the New Guinea orogen. BMR Journal of Australian Geology 10(3): 193–211.

FIGURE 1 Physiographic map of New Guinea (seafloor topography

from Smith and Sandwell 1997; Satellite Geodesy, Scripps Institute of Oceanography 2008). OJP, Ontong Java Plateau; EP, Eastern Plateau; PP, Papuan Platform. NEW GUINEA, GEOLOGY

659

FIGURE 2 Geological map of New Guinea. AB, Aru Basin; AFB, Aure fold belt; B, Bougainville; BB, Bintuni Basin; BK, Biak; BT, Bismarck Sea Trans-

form; C, Cyclops Mountains; CB, Cenderawasih Bay; FR, Finisterre Range; G, Gauttier Range; GR, Grasberg Mine; KT, Kilinailau Trench; L, Lihir Island (mine); LFB, Lengguru Fold Belt; M, Manus; MB, Manus Basin; MI, Misool; MT, Manus Trench; MU, Mussau; NB, New Britain; OT, Ok Tedi; P, Porgera; PT, Pocklington Trough; R, Rabaul; SB, Salawati Basin; SF, Sorong Fault; ST, Seram Trench; T, Timor Trough; TT, Trobriand Trough; W, Wau; WA, Waipona Basin; WB, Woodlark Basin; WM, Wamena; WN, Wandamen Peninsula; WO, Waigeo; WT, Weyland Thrust; Y, Yapen.

Indonesia in the west, with a boundary that coincides, for the most part, with the 141° E meridian. The western half was known as Irian Jaya and is now divided into two provinces known as Western Irian Jaya (the westernmost point of the island, commonly known as the Bird’s Head and Neck peninsula) and Papua (the rest of the Indonesian part of New Guinea). The population of PNG is 6 million and is predominantly Melanesian. The population of the Indonesian provinces is 2.1 million and comprises 60–70% indigenous Melanesian and 30–40% migrants from Java, Sulawesi, and Ambon. GEOLOGICAL SETTING

New Guinea is at the interface between the northwardmoving Australian plate and the west-northwest-moving Pacific plate. The resultant motion is convergence at a rate of 110 mm/yr on an azimuth close to 070°.

660

NEW GUINEA, GEOLOGY

Convergence has led to a succession of collisions of the Australian craton with the microcontinents and volcanic islands of the Pacific and with fragments of the craton that had been separated from the craton and then docked again. While the southern half of the island was always part of the Australian continent, the northern part has been added by successive collisions. In geological terms, the southern part is autochthonous and the northern part allochthonous, being made up of accreted terranes. The boundary between the Pacific and Australian plates is marked by a number of microplates. Those offshore are bounded by spreading ridges, deep-sea trenches, and transform faults, and those onshore by thrust, extensional, and strike-slip faults and folds. Earthquakes are located on the microplate boundaries (Fig. 3). Volcanic activity is associated with the deep-sea trenches and spreading ridges.

THE PAPUAN BASIN

The Papuan Basin occupies all of autochthonous New Guinea. Sediments of the Papuan Basin underlie the southern plains and are exposed in the fold belt, where the strata have been folded and faulted. The basin is underlain by Australian craton of Precambrian age in the west and of Paleozoic age in the east. The sedimentary section in the west is 16 km thick and includes Late Proterozoic and Paleozoic strata. The sedimentary section in the east is 4 km thick and is entirely Mesozoic and Cenozoic. The Mesozoic sediments in the east are similar to those in the west (Kembelangen Group) but tend to be less mature and less well sorted. Quaternary volcanoes are present only in the east. In both the east and the west, the sediments at the northern margin of the Papuan Basin fold belt have been metamorphosed to phyllitic and micaceous graphitic schists. In the east the transition from unmetamorphosed to metamorphosed sediments is gradational, and the metamorphic rocks are included in the area shown as fold belt on the map. In the west the contact is faulted and the metamorphosed sediments, along with other metamorphic rocks, have been picked out as a separate rock unit. The evolution of the Papuan Basin after the Permian period took place in five stages (Figs. 3–4): 1. Triassic and Jurassic rifting of the northern margin of the Australian continent, accompanied by the development of rift-related volcanics and syn-rift pockets of sediment 2. Jurassic and Cretaceous postrift copious siliciclastic sedimentation on the subsiding rifted margin, with some volcanic activity in Early Cretaceous 3. End-Cretaceous uplift and erosion of Cretaceous sediments in the east, due to thermal uplift associated with the opening of the Coral Sea Basin 4. Paleocene and Eocene to mid-Miocene: Slow subsidence and deposition of limestone followed by calcareous shale, except in the extreme northeast, where there was rapid clastic sedimentation from an emerging part-volcanic mountain mass 5. Late Miocene (12–8 million years ago) to present: Development of fold belt, uplift of mountains, rapid erosion and deposition of coarse clastic sediments, accompanied by volcanic activity Much of the fold belt comprises a broad, asymmetric, south-facing, thrust-bounded anticline upon which are superimposed lesser structures. East of 142° E the single

130˚ 0˚

132˚

134˚

136˚

138˚

140˚

142˚

144˚

146˚

148˚

150˚

152˚

154˚ 0˚

-2˚

-2˚

-4˚

-4˚

-6˚

-6˚

-8˚

-8˚

-10˚

-10˚

-12˚ 130˚

132˚

134˚

136˚

138˚

140˚

142˚

144˚

146˚

148˚

150˚

152˚

-12˚ 154˚

FIGURE 3 Earthquakes in the period 1963–2004, stronger than mag-

nitude 6. Focal depths as follows: red < 50 km, yellow < 100 km, green < 200 km, blue < 300 km, purple < 400 km, brown < 500 km, gray > 500 km. Shallow earthquakes (< 50 km) mark active plate boundaries. Map by Emile Okal.

anticline gives way to parallel thrust slices and valley-andridge topography. The strike of the fold belt changes at the international border. This may indicate that the border coincides with the contact between Paleozoic basement in the east and Precambrian in the west. However, the presence of Permo-Triassic metamorphic rocks further west, in the upper reaches of the Eilanden River at 140.2° E, suggests that the contact may be more complex. JIMI–KUBOR TERRANE

The Mesozoic and Cenozoic sediments of the Kubor Range and the Jimi River drainage rest on Paleozoic basement. The sequence is broadly similar to the Papuan Basin sequence and is regarded by some as an in situ part of the Papuan Basin. However, there are sufficient differences to suggest that Jimi–Kubor is a terrane, probably one that separated from the Australian margin in the Triassic and redocked with the margin in the Paleocene or Early Eocene. TERRANES OF THE CENTRAL RANGE, 136–141° E

In western New Guinea, along the north side of the central range, there is a belt of metamorphic rocks. This is bounded northward by ophiolite and, beyond the ophiolite, Cenozoic volcanic arc rocks. The metamorphic rocks, for the most part, are graphitic schists formed from the black shales and siltstones of the Kembelangen Formation and are thus part of the Papuan Basin sequence. However, there are blueschist facies metabasites, eclogites, and high-temperature amphibolites toward the northern margin. Such rocks typically are associated with arc-continent collision and the emplacement of ophiolite.

NEW GUINEA, GEOLOGY

661

Ultramafic and gabbroic rocks of the ophiolite, with minor basalt, are exposed for a length of 450 km on the north side of the central range. The ophiolite is thought to represent Late Cretaceous oceanic crust and mantle and to have been emplaced in the Late Cretaceous. Soils overlying ultramafic rocks are depleted in the lighter elements and enriched in iron and support vegetation poorly, with the result that the boundary of the ultramafic rocks can be mapped on vegetation pattern alone. Arc-type volcanic rocks and associated dioritic intrusive rocks that adjoin the ophiolite in the northwest are Late Miocene, whereas those further east are Late Eocene to Oligocene. ISOLATED TERRANES NORTH OF THE CENTRAL RANGE, 136–141° E

North of the central range is the Mamberamo basin, a vast area underlain by Late Miocene to Quaternary folded and faulted clastic sediments and some limestone. Within the basin are isolated blocks of basement rocks. The Gauttier mountain block comprises ultramafic rocks with some Cenozoic volcanic arc rocks. The Efar and Sidoas mountain blocks, adjacent to the north, are of ultramafic rocks. The Cyclops Mountains, near Jayapura, comprise ultramafic and gabbroic rocks and moderate to highgrade metamorphic rocks: amphibolite, gneiss, and schist. Immediately adjacent are Miocene sediments and Cenozoic volcanic arc rocks. Directly south of the Cyclops Mountains are the Border Mountains, made up of Paleozoic metasedimentary rocks intruded by Permo-Triassic granodiorite, diorite, and gabbro with some ultramafic rocks and andesitic porphyry. The metamorphic rocks include amphibolite, schist, quartzite, and garnet gneiss. TERRANES OF EASTERN NEW GUINEA 141–146° E

The Sepik complex comprises faulted blocks of metamorphic, ophiolitic, Eocene volcanic arc, and sedimentary rocks, with Oligocene to Miocene dioritic intrusive rocks FIGURE 4 History of accretion of terranes. (A) Late Cretaceous: Arc-

continent collision in Late Cretaceous in west emplaced Irian ophiolite. (B) Paleocene: Rifting of continent margin to produce Jimi–Kubor terrane, opening of Coral Sea; arc-continent collisions formed East Papua Composite Terrane (EPCT) and captured Marum ophiolite. (C) Eocene: Jimi–Kubor sutured to craton; Salumei volcanic arc develops. (D) Oligocene: Arc-continent collision formed the Sepik Complex; Finisterre volcanic arc developed; EPCT sutured to craton toward end of Oligocene.

662

NEW GUINEA, GEOLOGY

and Miocene-Quaternary sedimentary cover. The metamorphic rocks include blueschist facies metabasites and eclogite. The complex extends from the northern margin of the fold belt north to about the center line of the northern ranges and underlies the Miocene-Quaternary sediments of the Sepik basin. The complex includes Cretaceous high-grade metamorphic rocks associated with ultramafic rocks in the northeast. Basement rocks of the northern slopes of the Bewani– Torricelli ranges are Oligocene to earliest Miocene arc volcanic rocks and associated sediments and may be related to the arc volcanic rocks of the Finisterre Range. Near the international border, Miocene limestone partly conceals ultramafic rocks that may be an extension of the ultramafic rocks of the Cyclops Mountains. The Adelbert, Finisterre, and Saruwaged Ranges comprise Oligocene to Early Miocene arc volcanic rocks overlain by Miocene and younger limestone. The structure of the mountains is a south-facing thrust-based antiform marked by limestone dip-slopes on the north side and by rapidly eroding volcanic rocks in the south. WESTERNMOST NEW GUINEA: THE BIRD’S HEAD AND NECK, 130–136° E

The area north of the Sorong Fault comprises terranes that are mostly but not entirely of oceanic affinity. South of the Sorong Fault are terranes of continental affinity. Rocks north of the Sorong Fault include PermoTriassic granite, Cenozoic volcanic arc rocks, and younger sediments; ophiolite is present in adjacent islands. South of the Sorong Fault a basement of Paleozoic low-grade metamorphics is exposed in the northern mountains and on Misool Island. This is overlain by a thick sequence of platform sediments of Permian to Middle Miocene age that comprise the Salawati and Bintuni sedimentary basins. In the Bird’s Neck area, south of Cendrawasih Bay, the Weyland Thrust has transported Paleozoic sediments, metamorphic rocks, and ophiolite slices and Miocene diorites southward over a terrane of continental affinity. FIGURE 5 History of accretion of terranes. (A) Miocene: Finisterre

volcanic arc sutured to continent, development of Ramu-Markham and Sepik basins, Maramuni volcanic arc igneous activity triggered by uplift rather than subduction? (B) Pliocene: Converging of continent and Bismarck volcanic arc caused thrust faults to develop in Finisterre terrane and in Papuan Basin fold belt. (C) Present day. (D) Bird’s Head terrane history: 1. Australian Precambrian craton; 2. terranes that had docked by 25 million years ago; 3. terranes that had docked by 10 million years ago, those in Bird’s Head have Paleozoic continental basement; 4. terranes that had docked by 2 million years ago.

NEW GUINEA, GEOLOGY

663

SOUTHEASTERN NEW GUINEA, 146–154.4° E

The mountainous peninsula that forms southeastern New Guinea is sometimes referred to as the Papuan peninsula because it is part of what was formerly the Australian Territory of Papua. At the heart of the peninsula and islands is the East Papua Composite Terrane (EPCT). This comprises ophiolite on the northeast side and metamorphic rocks on the southwest. The two are separated by a major fault, which allows crustal extension and uplift of the metamorphic rocks. Overlying the ophiolite are Middle Eocene arc-type volcanics, Middle Miocene and younger volcanics, rapidly deposited clastic sedimentary rocks and some limestone, and Pliocene–Quaternary volcanic rocks, including those of the intermittently active major volcanoes Lamington and Victory. The association of metamorphic rocks and ophiolite extends northeast to the D’Entrecasteaux Islands and east-southeast to Misima, Sudest, and Rossel islands. The Trobriand Islands are raised coral and may be underpinned by Pliocene–Quaternary volcanic rocks. Woodlark (Muyua) Island comprises Quaternary limestone cover on Early Cenozoic volcanic arc rocks. The Aure Fold Belt is made up of a thick sequence of rapidly deposited Late Oligocene to Miocene and Pliocene mostly-clastic sediments, folded and faulted in response to westward movement of the EPCT. The Aure Fold Belt extends offshore as far east as 146.8° E. East of 146.8° E the fold belt gives way to thrustbounded strike ridges of Paleocene and Eocene fine siliceous sediments and minor Oligocene coarser sediments intruded by Oligocene gabbro. The sequence is interpreted to have formed as an accretionary prism above an Eocene–Oligocene northeast-dipping subduction system. Late Cretaceous and Middle Eocene tholeiitic basalts with rare interbeds of pelagic limestone form much of the peninsula east of 148° E. The basalts are 3000–4000 m thick and represent former ocean crust. Scattered stocks of syenite and related alkali-rich rocks of Middle Miocene age intrude the basalts. ISLANDS OF THE BISMARCK ARCHIPELAGO

The large islands of the Bismarck Archipelago have a common origin, as is indicated by the similarities in their geology. Each has a basement of volcanic arc rocks of Eocene or Oligocene age unconformably overlain by Miocene limestone, which is in turn unconformably overlain by Pliocene and Quaternary clastic sediments and volcanics. The sequence is similar to that seen in the Finisterre Range on the mainland. 664

NEW GUINEA, GEOLOGY

The volcanoes of the Bismarck volcanic arc include the caldera collapse volcanoes Long Island, Dakataua, Witori, and Rabaul, each of which has been the source of devastating eruptions in the past. Ash emission from an eruption of Long Island in about 1665 AD was sufficiently voluminous to block out the light of the sun for several days, an event recalled in legend as a time of darkness. ONTONG JAVA PLATEAU

The Ontong Java Plateau is a continent-scale mass that comprises a sequence of basaltic lava flows overlain by 1 km of pelagic sediments. The lavas were emplaced in one remarkable major magmatic event over a period of less than 7 million years around 122 million years ago (Early Cretaceous, Aptian). The plateau is entirely submerged except for isolated atolls. SMALL OCEAN BASINS

The Coral Sea Basin opened in the Paleocene. Adjacent submarine plateaus are rifted fragments of Australian craton. The Solomon Sea basin, north of the Trobriand Trough, opened in the Late Eocene–Early Oligocene and has been subducted at both the New Britain Trench and the Trobriand Trough. The Manus and Woodlark basins both opened in the last 6 million years. TECTONIC ACTIVITY

The rapid oblique convergence of the Australian and Pacific plates results in ongoing tectonic activity. In northwestern New Guinea a series of east-westtrending faults links the Bismarck Transform in the east with the Sorong Fault in the west. Movement on the faults is complemented by subduction at the New Guinea Trench and transpressional folding and faulting in the fold belt and in the Mamberamo Basin. Repeated occupation of survey stations shows that the western side of Cendrawasih Bay is moving west-southwest at a rate of 93 mm/yr. This motion likely caused the opening of Cendrawasih Bay, the development of Waipona Basin, and the development of the Lengguru Fold Belt. The motion is almost the same motion as that of the Pacific Plate (110 mm/yr) and suggests that the western part of the New Guinea Trench is locked. Further east, the New Guinea Trench is active. Seismic tomography suggests that a subducted slab dips at a gentle angle southward from the trench beneath the central ranges. The subducted slab may be be the trigger for igneous activity in the fold belt, such as the intrusive rocks at Grasberg and Ok Tedi, and it may explain the transfer of convergent stress from the New Guinea Trench to the southern front of the fold belt, a distance of 300 km.

In northeastern New Guinea, the ongoing collision of the Finisterre and Sarawaged ranges with the Bismarck volcanic arc causes uplift of the north coast of the Huon Peninsula at averaged rates of 1–3 mm/yr. Study of raised coral terraces on the peninsula has yielded a highquality record of fluctuations in sea level during the Late Quaternary. At the same time the Finisterre mountain mass rides southward and causes the down-warping of the northern end of the Papuan peninsula, which is subsiding at rates of up to 5 mm/yr. In eastern New Guinea, active sea floor spreading in the Woodlark Basin is advancing westward and causing north-south extension of the mainland and adjacent islands. One result is the emergence in the Pliocene of domes and half-domes of metamorphic rocks by lowangle extensional faulting. Another is the opening of small rift basins offshore. Spreading within the last 1.2 million years has caused the separation of Misima Island from a position adjacent to Woodlark (Muyua) Island (152.8° E).

Pigram, C. J., and H. L. Davies. 1987. Terrranes and the accretion history of the New Guinea orogen. BMR Journal of Australian Geology and Geophysics 10:193–211. Visser, W. A., and J. J. Hermes. 1962. Geological results of the exploration for oil in Netherlands New Guinea. Verhandelingen, Geologische Serie 20. The Hague: Koninklijk Nederlands Geologisch Mijnbouwkundig Genootschap. REFERENCES

Satellite Geodesy, Scripps Institute of Oceanography. 2008. Global topography. http://topex.ucsd.edu/marine_topo/mar_topo.html. Smith, W. H. F., and D. T. Sandwell. 1997. Global seafloor topography from satellite altimetry and ship depth soundings. Science 277: 1957–1962.

NEW ZEALAND, BIOLOGY STEVEN A. TREWICK AND MARY MORGAN-RICHARDS Massey University, Palmerston North, New Zealand

ECONOMIC ASPECTS

Oil is produced from Miocene reefs in the Salawati Basin. A 17-trillion cubic foot reserve of gas beneath Bintuni Bay is being developed for export as liquefied natural gas (LNG). In Papua New Guinea oil and gas are produced from structures in the fold belt. Copper and gold are produced from major mines at Grasberg and Ok Tedi, and gold from Porgera and Lihir Island. In the eastern Bismarck Sea (Manus Basin), gold and base metal sulfide mineralization on the sea floor is associated with active spreading ridges. SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / New Guinea, Biology / Plate Tectonics / Pocket Basins and Deep-Sea Speciation FURTHER READING

Cloos, M., B. Sapiie, A. Quarles van Ufford, R. J. Weiland, P. Q. Warren, and T. P. McMahon. 2005. Collision delamination in New Guinea: the geotectonics of subducting slab breakoff. Special Paper 400. Boulder, CO: Geological Society of America. Dow, D. B., G. P. Robinson, U. Hartono, and N. Ratman. 1988. Geology of Irian Jaya. Indonesia: Geological Research and Development Centre. Hill, K. C., and R. Hall. 2003. Mesozoic–Cenozoic evolution of Australia’s New Guinea margin in a west Pacific context. , in Evolution and dynamics of the Australian plate. R. R. Hillis and R. D. Müller, eds. Special Paper 372. Boulder, CO: Geological Society of America, 265–290. Hope, G. S., and K. P. Aplin. 2007. Paleontology of Papua, in The ecology of Papua: Part 1. B. M. Beehler and A. J. Marshall, eds. Singapore: Periplus Editions, 247–254. Parris, K. 1996. Central Range Irian Jaya Geology Compilation 1:500,000 scale geological map. Jakarta: P. T. Freeport Indonesia. Pieters, P. E., C. J. Pigram, D. S. Trail, D. B. Dow, N. Ratman, and R. Sukamto. 1983. The stratigraphy of western Irian Jaya. Bulletin of the Geological Research and Development Centre, Bandung, Indonesia 8:14–48.

New Zealand, spanning more than 1400 km of latitude on the southwest edge of the Pacific Ocean, supports a distinct assemblage of plant and animal groups. Species-level endemism in the wet temperate forests and alpine habitats is high; however, compared to many other oceanic islands, species diversity is not. New Zealand is a small part of a large continent, Zealandia, that sank beneath the surface of the sea after separation from Gondwanaland and is thus often considered a continental island. Whether any of the New Zealand biota originated in Zealandia is uncertain, but a number of animals that lack close living relatives elsewhere in the world may have arrived in that way. As with true oceanic islands, New Zealand biodiversity is dominated by speciation in relatively recent geological time, mostly from overseas colonists. GEOLOGICAL HISTORY AND GEOGRAPHIC SETTING

New Zealand is composed of continental crust; a property it shares with just a few other islands, including New Caledonia and Madagascar. In fact, the geological histories of New Zealand and New Caledonia are closely linked, as both are small, emergent parts of an otherwise submerged continental fragment, Zealandia. The continent of Zealandia was somewhat larger than India when it rifted from Gondwana starting about 85 million years ago. During the following 60 million years the continental crust of Zealandia stretched, thinned, and sank, so that today 93% of its area is beneath the sea. Subsequently, NEW ZEALAND, BIOLOGY

665

parts of Zealandia have re-emerged as a result of local tectonic activity to form islands, and these include New Caledonia, the Chatham Islands, and most if not all of New Zealand. The New Zealand archipelago lies in the southwest Pacific Ocean and has a total area of about 270,000 km2. The nearest continent, Australia, is about 1,500 km to the west. New Zealand consists of three main islands: North Island and South Island separated by the Cook Strait, and the rather smaller Stewart Island separated by Foveaux Strait (Fig. 1). There are a number of other smaller inhabited islands: the Chatham Islands (963 km2) about 800 km east and the islands of Waiheke and Great Barrier in the Hauraki Gulf of North Island, plus numerous uninhabited islands near the mainland coast that share a recent biological history, because they were connected when sea level was lower in the Pleistocene (Fig. 2). In contrast, the more distant islands to the north (Poor Knights, Three Kings), south (sub-Antarctic islands), and east (Chatham Islands) have been isolated longer or were never linked to mainland New Zealand. Diversity and endemicity are not homogeneous even across the main islands; distinct zones of higher endemicity are particularly conspicuous among the flora (Fig. 2). The climate is predominantly temperate but ranges from cool temperate in the south (latitude 47º) to subtropical in the north (latitude 34º). A large proportion of New Zealand can be classified as mountain land (60% of South Island and 20% of North Island), and this contributes to habitat diversity. The majority of alpine habitat is on the Southern Alps, which run the length of South Island, but there are smaller ranges and a number of volcanic mountains in North Island (Fig. 1). The Southern Alps reach to ∼3000 m above sea level (the highest, Mt. Aoraki/Cook, is 3753 m), but the treeline in New Zealand is relatively low (averaging 1300 m above sea level), so the alpine zone is a relatively large and important ecotone. For students of island biogeography, New Zealand is an enigma. The biology has features reminiscent of both continental lands and oceanic islands. For example, the absence of native terrestrial mammals is typical of an island fauna, but the presence of an endemic order of reptiles (tuatara; Sphenodontia), two endemic orders of birds (moa and kiwi), and an endemic family of amphibians (frogs; Leiopelmatidae) is seen as more continental in nature. Recognition that New Zealand is a continental island, and that fragmentation of Gondwanaland played an important role in its geological formation, has strongly influenced interpretation of its biogeography. In contrast to dispersal origination of biota on true oceanic islands,

666

NEW ZEALAND, BIOLOGY

Three Kings Is. Northland

North Island

Cook Strait Tasman Sea

South Island Bank's Peninsula Chatham Is.

Stewart Island

Fouveax Strait

FIGURE 1 The islands, straits, and and other locations in modern New

Zealand, with the distribution of montane land (1000 m) indicated (black), formed largely since the Pliocene (5 million years ago).

the biota of New Zealand have been widely treated as elements of an ancient, continental “ark.” There has been an overemphasis on inferring biogeographic origins from distribution patterns rather than recognizing the role of speciation in New Zealand biology. PATTERNS OF SPECIES DIVERSITY

Prior to human arrival the natural vegetation over about 85% of New Zealand was mixed temperate rain forest, with southern beech (Nothofagus), tree ferns (Cyathea, Dicksonia), and species of the Southern Hemisphere gymnosperm family Podocarpacea being prominent features of the forests. The biggest New Zealand tree is the endemic kauri (Agathis australis), the only representative of this genus in New Zealand. Kauri belongs to a group of ancient conifers Araucariaceae, which has most of its diversity in subtropical and tropical Oceania. New Zealand plants tend not to have prominent, showy flowers; instead, a large proportion of angiosperms have small, often pale or white flowers pollinated by generalists, including flies, rather than specialist butterflies and bees. Among the few flamboyant flowering trees are kowhai (Sophora), with large yellow flowers, and pohutakawa or Christmas tree (Meterosidros), which produces abundant scarlet inflorescences in December; both have close relatives elsewhere through the southern hemisphere and Pacific. Like the flora, New Zealand’s birds are not showy; native forest birds tend to have cryptic plumage, many are nocturnal (21%), and there is also little sexual dimorphism in plumage. For example, New Zealand species of Petroica robins are monomorphic and monochrome, whereas male Australian Petroica robins have bright pink or scarlet chests. However, several New Zealand birds show traits indicative of resource

partitioning, including pronounced beak dimorphism in the huia (Heteralocha acutirostris, extinct) and probably also, but to a lesser extent, the Chatham Island rail (Gallirallus modestus, extinct) and kaka parrot (Nestor meridionalis), and size dimorphism in some moa that was so extreme that bones from males and females were initially identified as belonging to different species. The alpine habitat is rather youthful in New Zealand, probably largely developed during the last 5 million years as the mountain ranges formed and the climate cooled in the late Pliocene. Nevertheless there are many alpine specialists, including snow tussock grass (Chionochloa), cushion-forming herbs such as “vegetable sheep” (Haastia and Raoulia), buttercups (Ranunculus), alpine daisies (Celmisia), skinks, geckos, birds including the rock wren (Xenicus gilviventris), and the world’s only alpine parrot, the kea (Nestor notablis). Specialist alpine invertebrates are numerous and include cicadas (Maoricicada), a black ringlet butterfly (Percnodaimon), short-horned grasshoppers (e.g., Sigaus), cockroaches (Celatoblatta), and weta. Some insects, including the

A

alpine species of weta (Hemideina and Deinacrida), grasshoppers (Sigaus Brachaspis), and cockroaches (Celatoblatta), are tolerant of freezing and over-winter under snow and ice as adults and juveniles (Figs. 3–6). These alpine-adapted species all belong to recent radiations that include alpine and lowland representatives. ASSEMBLY OF THE BIOTA

The New Zealand biota has been described as ill-balanced because of the variance in diversity and endemicity exhibited among different plant and animal groups, and this has frequently been attributed to New Zealand’s supposed long isolation in the Pacific. However, the composition is broadly consistent with other island assemblages and indicative of long-standing (albeit intermittent) interactions with other biotas. Its geographic position, hundreds of kilometers from other land for some millions of years, has inevitably restricted successful migration of plants and animals, but not precluded it. As a result, a large proportion of species in many groups are endemic to New Zealand, but endemicity above this level is much lower. For example, New Zealand

C

B

Continuous forest

125 Volcanic activity starts ~1.75 mya

Scrub, grassland and forest patches 36

189 Orogenic activity starts ~1mya

Deposition of glacial aluvial outwash

Glacial ice

C.30

Bare ground and grassland 90

km 0

200

400

FIGURE 2 The changing shape of New Zealand and its biota. Changes in the distribution, extent and topography of land during the last 5

million years are thought to have influenced the distribution of endemism within New Zealand. (A) During the early Pleistocene (1.8 million years ago) there were two major islands separated more than today. A number of smaller islands were subsequently joined to the mainland (for example, Bank’s Peninsular and the Northland archipelago, which still have endemic species). Estimated outline of land during the early Pleistocene (green) is superimposed on the present shoreline. (B) During glaciation sea level fell, and at the last glacial maximum (0.02 million years ago) a single major island would have existed (green). During this time glaciers extended across the Southern Alps (black) (McGlone 1985). (C) Five broad zones of plant endemicity have been identified among plants (dashed lines, values indicate numbers of endemic species) (Wardle 1963). Not surprisingly, the land areas with fewest endemics are those that are youngest or most disturbed during the late Pleistocene. The extent of forests was much reduced during glacials.

NEW ZEALAND, BIOLOGY

667

H. thoracica

D. connectens

0.01 substitutions/site

FIGURE 4 Biodiversity patterns in New Zealand, exemplified by anos-

tostomatid weta. Genetic structure within two widespread species of weta in New Zealand. The Auckland tree weta (Hemideina thoracica) in North Island has greatest genetic diversity in the far North, where islands exisited during the Pliocene and where forests grew even during Pleistocene glacial cycles and after the volcanic eruptions of the central North Island. The giant scree weta (Deinacrida connectens) in South Island is fragmented on mountain peaks, but during glacial cycles its distribution would have been more continuous at a lower altitude. Genetic diversity within this species dates to uplift of the southern mountains.

FIGURE 3 Biodiversity patterns in New Zealand, exemplified by anos-

tostomatid weta. The Anostostomatidae are a relatively diverse group of orthopterans in New Zealand, expressing a range of biological features characteristic of the biota generally. The tusked weta (A) have their closest relatives in New Caledonia. The (B) giant (Deinacrida) and (C) tree (Hemideina) weta of New Zealand are unique among Anostostomatidae for eating leaves, fruit, and flowers. Secondary sexual characteristics such as enlarged heads and jaws in males have evolved in Hemideina tree weta, and a similar evolutionary path has resulted in the tusks of male tusked weta (A).

668

NEW ZEALAND, BIOLOGY

has more than 20,000 endemic invertebrate species (95% endemism at species-level) but just five endemic invertebrate families, and 2000 endemic species of vascular plants but no endemic families. New Zealand is subjected to a prevailing westerly wind and circumpolar oceanic current, and some animals are regular visitors. For instance, many seabirds regularly traverse the oceans but nest in New Zealand (e.g., sooty shearwater, Puffinus griseus); cuckoos (Chrysococcyx lucidus, Eudynamys taitensis) travel to and from islands in the Pacific Ocean; and godwits (Limosa lapponica) migrate to breed in Alaska. For other taxa, migration is less frequent, but among the landbirds there are many post-human colonists (see the section “Human Contact”), and Australian species that arrived before recording began (e.g., the pukeko Porphyrio porphyrio, the harrier Circus approximans, the fantail Rhipidura fuliginosa, and the owl Ninox novaezealandiae), in addition to endemic species that have a close affinity (genus level) to

Anisoura nicobarica Motuweta isolata Motuweta riparia Deinacrida fallai

H. thoracica

D. heteracantha D. mahoenui

D. rugosa

H. trewicki

D. tibiospina D. talpa H. broughi D. pluvialis

H. crassidens D. parva

H. femorata D. elegans

H. ricta D. connectens

H. maori

D. carinata FIGURE 5 Biodiversity patterns in New Zealand, exemplified by anos-

FIGURE 6 Biodiversity patterns in New Zealand, exemplified by anos-

tostomatid weta. The distribution of tusked and giant weta in New

tostomatid weta. The distribution of tree weta reveals broad regional

Zealand. The greatest diversity of Deinacrida is in the habitat-diverse

allopatry and evidence for range expansion following climate change.

South Island mostly associated with the mountains. The three tusked

H. thoracica extend its range south probably after the last glacial maxi-

weta species are restricted to northern North Island.

mum, excluding the cold-adapted H. crassidens from lowland forests of central North Island but leaving isolated populations of H. crassidens marooned in the subalpine zone of mountains in the region.

taxa elsewhere (e.g., Petroica robins in Australia). In many instances, New Zealand species have extraordinary forms compared to their nearest (frequently Australian) counterparts. For example the large, endemic flightless takahe (Porphyrio hochstetteri) shares a common ancestor with the smaller, flying purple swamphen or pukeko (P. porphyio), and the giant extinct eagle (Harpagornis moorei) shares a recent (Plio-Pleistonce) ancestor with the Australian little eagle (Aquila morphnoides). However, New Zealand also has taxa that have been classified as distinct at higher levels (e.g., bird families including wrens Acanthisittidae, kiwis Apterygidae). In the freshwater realm, the fish are either themselves diadromous or are related to diadromous taxa. Eels are common (Anguilla) and have been an important food resource of people. Other native fish in New Zealand lakes and streams include species of Galaxias, Neochanna, and Gobiomorphus, most of which are endemic, but the genera also occur elsewhere, including Australia. Endemic freshwater invertebrates include two species of crayfish (Paranephrops) (related taxa exist among the more diverse Parastacidae fauna of Australia) and a number of insects with unusual life histories, such as caddis flies and

dragonflies with semi-terrestrial larvae. Freshwater invertebrates tend to be highly distinctive (e.g., 20 of 21 genera of stoneflies, Plecoptera, are endemic), although there is a species of freshwater crab that also occurs in Australia. An analogous pattern is evident in the terrestrial flora too, with endemism high at species level (80%) but less so at higher taxonomic levels. The flora, and other elements of the biota, have been subjected to substantial change over time. There is evidence for bouts of diversification associated with geophysical events in New Zealand’s prehistory, including significant changes in area, habitat diversification, and climate fluctuation. During and since the Miocene, changing diversity has resulted from extinction (e.g., Eucalyptus gum trees disappeared from New Zealand), colonization (e.g.; Fuscospora beech arrived), and speciation (e.g., Coprosma). Many elements of the biota are described as “Gondwanan.” Such taxa are simply those that have a distribution largely restricted to modern land areas that originated from the breakup of Gondwanaland: Africa (usually just southern Africa), Madagascar, South America, India, Australia, and New Caledonia. It is often assumed that a Gondwana distribution is evidence of

NEW ZEALAND, BIOLOGY

669

a vicariant Gondwanan origin, and this concept has been a central tenet in New Zealand biogeography since the acceptance of continental drift in the early 1970s. A broad range of New Zealand taxa are attributed to this origin, including moa and kiwi (ratite birds), southern beech trees (Nothofagus), weta (anostostomatid crickets), land snails (Punctidae, Charopidae), and peripatus (Onychophora or velvet worms). Following a form of reciprocal illumination, some biogeographers have confounded the apparent evidence from Gondwana biological distributions with geological evidence for the Gondwanan origin of the continental crust from which New Zealand is formed and concluded that the biology of New Zealand is first and foremost Gondwanan. That is, that the biota of New Zealand has evolved in isolation since separation by continental drift from Gondwana some 62–80 million years ago. A stream of recent evidence from molecular studies in particular reveal that this is, in fact, far from true. A prime example comes from Nothofagus beech, an iconic Gondwanan taxon, long assumed to be incapable of transoceanic dispersal. Yet, recent research has revealed that extant New Zealand beech arrived after separation of Zealandia from Gondwana. In a similar vein, although the presence of peripatus (Onychophora) in New Zealand is consistent with a vicariant Gondwanan origin, this is not the only explanation. There are also peripatus in Jamaica (an emergent Miocene island more than 600 km from a continent). Similarly, weta (Anostostomatidae), which have a largely Gondwanan distribution (i.e., Southern Hemisphere), also occur in Japan, and speciose land snail families (Charopidae, Punctidae) are well represented on oceanic Pacific islands. It may remain useful to describe some of these as Gondwanan in terms of distribution (extant and fossil), but not necessarily in terms of the process that led to their current distribution. Gondwanan taxa are essentially southern hemisphere taxa and these are the most likely to arrive in southern lands. For most plants and for almost all terrestrial animals, the fossil record in New Zealand is, as yet, too patchy to be highly informative about timing of origin and persistence in the biota. Birds and some insects and plants are represented by fossils in caves, middens, swamps, and sand dunes; such sites often have abundant material, but because they are of Holocene age, they are informative about the last few thousand of years of New Zealand’s prehistory and not the preceding millions. Recent discoveries of rich Miocene fossil sites in southern South Island will no doubt open important windows into the past biology. Already a small mammal, a crocodile, and a community of now-extinct aquatic birds are known to 670

NEW ZEALAND, BIOLOGY

have existed 18 million years ago and reveal much about the turnover of the New Zealand biota. FEATURES OF THE BIOTA Eclectic Mix

There are no living snakes, land turtles, or crocodiles in New Zealand, and lizards belong to just two groups (geckos and skinks), but the tuatara is the only living representative of the sphenodontid reptiles anywhere in the world. Among invertebrates, many orders are missing (e.g., Scorpionida, Embioptera, Zoraptera), and others are poorly represented (e.g., Mecoptera, one species; Formicidae, 12 species, compared to ∼1300 in Australia). Similarly, within orders, representation is patchy even in comparison to neighboring land areas. For example, Orthoptera have high endemicity and diversity among the Rhaphidophoridae (18 genera, ∼50 species), Anostostomatidae (five genera, ∼70 species), and to a lesser extent Acrididae (four endemic and two native genera, ∼16 species, mostly associated with alpine habitat), fewer still in Tettigonidae (two Australian katydids), Gryllidae (two endemic and one Australian species), and Gryllotalpidae (one endemic), while others are absent (e.g., Gryllacrididae). Curious Endemics

Among the birds, the large parrot kakapo (Strigops habaproptilus) is exceptional in its combination of life history characteristics and is unique among parrots for any one of these characteristics: lek breeding, nocturnal, and flightless. The sporadic breeding cycle of the kakapo is linked to the masting of forest trees. The New Zealand avifauna included some 16 ratites (all endemic genera), including five of the smallest living members of the group, the kiwi (Apteryx). Kiwi invest all their reproductive effort into a single, very large (up to 450 g) egg per nesting. They are predominantly nocturnal and are the only birds with nostrils at the tip of the bill. Olfactory cues are important to kiwi, which forage on the ground at night, but it was recently discovered that kiwi have vibration sensors in the bill tip like those of wading birds (Scolopacidae), which are used when probing for prey. New Zealand amphibian diversity consists of just four extant frogs, belonging to an endemic family (Leiopelmatidae). New Zealand frogs lack ears, pass through the tadpole stage in their eggs, and have paternal care of froglets. The only recent native mammals are three species of bat. The long-tailed bat (Chalinolobus tuberculatus) probably arrived from Australia in the Pleistocene. The two short-tailed bats (one now extinct) belong to an endemic family (Mystacinidae), and Mystacina tuberculate is unusual

in having evolved the ability to move efficiently on the ground, where they forage on insects, fruit, pollen, and nectar. The largest known eagle in the world (Harpagornis moorei) hunted moa (as revealed by talon marks in Holocene fossil moa hip bones), had a wingspan of 2.5 m, and had evidently evolved to exploit this large prey. Two genera of New Zealand weta (Deinacrida and Hemideina) are unusual in having adopted a largely herbivorous diet (Fig. 3). They feed primarily on green leaves or fruit or flowers, and rarely on animals, which is the normal diet of other genera in this family of crickets (Anostostomatide) in New Zealand and elsewhere. The only significantly poisonous animal in New Zealand is the katipo spider (Latrodectus katipo), which, judging by its close relation to the infamous Australian red-back (Latrodectus hasselti), must have arrived on New Zealand beaches in recent geological time.

All modern ratites are flightless, and New Zealand had diverse fauna of moa (11) and kiwi (5). Similarly, six species of penguin (Spheniscidae) breed in New Zealand. Other flightless species were members of volant groups. Rails have the greatest propensity to evolve flightless forms, as observed on many other islands, and 11 species (70%) are known from the New Zealand archipelago. Porphyrio, Gallirallus, Gallinula, and Fullica are each represented by one or more flightless species on New Zealand main or offshore islands; each is a product of a separate colonization by a flying ancestor. Other flightless taxa include a parrot (Strigops), the strange rail-like predator (Aptornis), ducks (Anas), geese (Cnemiornis), and wrens (Xenicus). Much, and in some cases everything, that is known of these birds has been gleaned from their Holocene bones preserved in sand dunes, swamps, and caves.

Gigantism

Gigantism has in the past been identified as a distinctive feature of the biota, but most of the animal groups usually identified as being represented by giant forms in New Zealand (e.g., earthworms, centipedes, land snails, flatworms, millipedes, slugs, stick insects, weta, longhorn beetle, a weevil, a moa, and an eagle) also have species as large or, in most cases, larger in other parts of the world. A weta (cricket) is credited in New Zealand as being the heaviest insect recorded anywhere; female Deinacrida heteracantha in the wild average about 32 g (a much higher, anomalous weight of 71 g is famously recorded from a captive egg-engorged female). On the whole, animal groups with large species in New Zealand also contain many (usually the majority) smaller taxa. This is true even for the famous moa; although Dinornis giganteas was very big (240 kg), a bigger ratite (Aepyornis maximus) is known to have existed in Madagascar, and other moa were considerably smaller (Megalapteryx didinus ∼25 kg). There are a number of “megaherbs” with large, glossy leaves and large, colorful floral displays associated with the moist environments of some offshore islands, including Chatham Island forget-menot (Myosotidium hortensia) and Campbell Island daisy (Pleurophyllum speciosum). The extinct New Zealand eagle (Harpagornis moorei) and the giraffe weevil (Lasiorhynchus barbicornis, which owes half of its length to a long, thin snout) are the largest of their respective kinds, but there is little evidence for a dominant evolutionary pattern across the biota. Flightlessness

About a third of New Zealand land birds at the point of human contact were flightless; many are now extinct.

Floral Peculiarities

The New Zealand flora consists of some 2300 native species, of which 85% are endemic. A relatively large proportion of the angiosperms have separate sexes or some degree of sexual dimorphism (23% of genera). There is a predominance of white flowers and unspecialized pollination systems among the largely evergreen trees and shrubs of New Zealand. Many (e.g., Nothofagus, Dacrydium, Chionocloa) display a high variance in fruiting from one year to the next (masting), and this has a significant impact on breeding of endemic birds (in particular kakapo, Strigops) and introduced mammals (mice, Mus). There is a high frequency of small-leaved, tangle-branched shrubs (divaricating habit), a form that has evolved independently in 20 plant families (e.g., Coprosma, Myrsine, Melicytus, Pseudopanax, Pittosporum, Olearia). There are about 60 species with tiny leaves and interlacing branchlets, and about 14% of these represent juvenile stages of plants that later grow into adult leafy trees of normal habit and foliage (e.g., matai, Prumnopitys taxifolia; putaputaweta, Corpodetus serratus). Two competing hypotheses explain the unusual abundance of this plant growth form: moa browsing and climate. From subfossil remains we know that moa did indeed eat these divaricating plants. However, the probable recent origin and distribution of the divaricating species, added to the fact that they have not been replaced in the >200 years since moa went extinct, suggests that climate is the more likely selective force. Skinks and geckos appear to be important seed dispersers for many of these small/divaricating shrubs. Honeydew is a sugar-rich secretion produced by sapsucking insects. In New Zealand, beech trees (Nothofagus) are frequently infested with scale insects (Ultracoelostoma) NEW ZEALAND, BIOLOGY

671

that secrete honeydew through a long, hairlike, waxy tube that extends from the bark of host trees and is an important resource for native birds (e.g., kaka, tui, bellbird). Exotic bees and wasps (especially Vespula) compete for honeydew, and in some forests they take almost all of it, depriving local birds. Group Diversity and Radiations

Many invertebrate groups have high diversity and endemism that are the product of species radiations, including some carabid beetles (e.g., Mecodema), alpine cicadas (Maoricicada, Kikihia), weta (Fig. 2), and land snails large (predatory taxa of Rhytidae including Powelliphanta) and small (including Punctidae and Charopidae). The latter show high levels of sympatric species diversity in some parts of New Zealand. Species diversity within bird groups is less prominent; there are relatively few endemic species of birds in New Zealand (176 endemics of 245 species at human contact) but high representation of taxonomic diversity (20 of 27 orders). Among the more speciose are kiwi (5) and moa (11), and the Cyanoramphus parakeets represent a young radiation of some 10 species in the New Zealand archipelago. Seabirds are well represented, and even today albatross (Diomedea epomophora), gannet (Sula bassana), and two penguins (Megadyptes antipodes, Eudyptula minor) nest on the mainland, although many others are now restricted to offshore islands. Notably speciose plant groups include Coprosoma, Hebe, Ranunculus (buttercup), Celmisia (daisy), and Asplenium (fern); the products of relatively young radiations. HUMANS IN AOTEAROA Human Contact

As with most oceanic islands, the arrival of humans in New Zealand had a major impact on the composition and structure of the flora and fauna. The first people (Maori) to colonize the New Zealand archipelago (Aotearoa), came from central Polynesia and made first contact with these islands 1000 to 600 years ago. They introduced, during a succession of exchanges, a number of commensal animals, including the Pacific rat (kiore, Rattus exulans) and dogs (the extinct kuri) and some plants for cultivation (e.g., kumara, Ipomoea batatas). Early Maori hunted birds that provided a ready and abundant resource of food and materials (e.g., feathers woven into cloaks known as korowai). Moa were probably hunted to near extinction within about 100 years of colonization, and it is likely that low fecundity and slow growth resulted in their subsequent extinction. Other large ground birds (the flightless goose Cnemiornis, the adzebill Aptornis)

672

NEW ZEALAND, BIOLOGY

were also soon extinct, and other forest birds (the weka rail Gallirallus and the keruru pigeon Hemiphaga) were harvested using specialized techniques including snares and traps. Sea mammals including seals and sea lions were taken in great numbers at their coastal rookeries, and seabirds (titi petrels, Puffinus) were gathered at their nesting grounds that were often far inland. Maori also used fire to clear land as tribal structure developed. The first confirmed contact with New Zealand by a European was made by Abel Tasman in 1642. James Cook and Jean François Marie de Surville reached the islands more than 100 years later in 1769, and subsequently whalers, traders, and missionaries began arriving. Colonization by European people did not start in earnest until 1840 but resulted in accelerated modification of the landscape and biology of New Zealand. In particular, land was cleared of trees as timber was harvested, and pasture farming developed. Today, about 22% of the prehuman vegetation/ habitat remains in a relatively pristine condition. Introduced wild or feral animals included mice, cats, pigs, rats (ship and Norway), deer, goats, hares, and rabbits, and early European settlers continued to add familiar species from “home” after settlement. Many “garden” birds (e.g., song thrush, sparrow, blackbird, several finches, pheasant) and hedgehogs were also introduced under the auspices of regional Acclimatization Societies. A total of 33 bird species and 34 terrestrial mammal species have established in New Zealand. Trout, salmon, and 18 other freshwater fish have been introduced, to the detriment of native freshwater fish and invertebrates. In an early (1882) misguided attempt at biocontrol, mustelids (stoats, weasels) were introduced to limit the burgeoning rabbit population. It soon became apparent that, as predicted by a few scientists of the time, the predators did not restrict their attentions to rabbits, and today mustelids remain major predators of native birds, lizards, and invertebrates. Despite early attempts in the late nineteenth century to protect notable endemic species, the impacts of introduced predators and natural history collecting continued to be felt. For instance, a species of wren (Xenicus insularis) was extinguished from Stephen’s Island in the Cook Strait in 1895 through the attention of cats brought to the island by the lighthouse keepers. In the 1890s a pair of huia (Heteralocha acutirostris), captured for translocation to a reserve island, were in fact sold illegally to a collector in the expectation that more could be found and conserved, but this never happened and the species was lost. The brush-tailed possum (Trichosurus vulpecula) was introduced from Australia in 1837 with the aim of supplying the fur trade; this possum is now a major pest in New

Zealand. Eating leaves, flowers, and fruits of native trees, it competes with native birds, and it also directly preys on eggs and nestlings. Some 76 native bird species have thus become extinct since the arrival of people in New Zealand, including 41% of the 176 endemic species. A large number of plant species have also been introduced (about 1630 alien plant species have established), and many of these are now invasive weeds. For vascular plants, land mammals, land birds, and freshwater fish, over 40% of the species now found in New Zealand are exotics. Habitat modification also appears to have opened the way for a number of self-introductions; first records for birds include the silvereye Zosterops in 1832, the welcome swallow Hirundo tahitica in the 1950s, the spur-wing plover Vanellus miles in 1932, and the cattle egret Bubulcus ibis in 1963. Exotic insects include the monarch butterfly Danaus plexippus, which arrived in the 1880s. Although perhaps facilitated by human activities, this process very much continues a prior persistent feature of sporadic colonization (see the section on “Assembly of the Biota”). Conservation

Despite the ravages of habitat modification and exotic taxa, a number of distinctive bird species have been kept from extinction by the efforts of New Zealand conservationists, who have pioneered management techniques now applied worldwide. Prominent successes include the black robin (Petroica traversi), resurrected from just five birds (two females, three males) in 1980 using a combination of translocations, cross-fostering, and supplementary feeding. The intensely managed night-parrot, kakapo (Strigops habroptilus), which, despite an extreme male bias and intermittent breeding, has experienced a gradual improvement in its meager population, which today stands at 90. Among the most valuable resources available to New Zealand conservation are the offshore islands, from which introduced predators are removed to provide vital reserves for protected species. Several endangered taxa owe their survival thus far to remnant populations on offshore islands (e.g., Hamilton’s frog, Leiopelma hamiltoni, on Stephens Island; hihi, Notiomystis cincta, Little Barrier Island; saddleback or tieke, Philesturnus carunculatus, on Hen Island and Big South Cape Island). New Zealanders also pioneered mammal eradication techniques to clear islands of introduced pest species; the largest island to have been successfully cleared of rats is Campbell Island (114 km2). A dozen species of exotic mammal have been eradicated from other islands in the region including mice, possums, cattle, pigs, goats, rabbits, and cats. The use of sophisticated fencing techniques and predator control programs have also

allowed the development of “mainland islands” that protect dwindling biodiversity and provide a valuable point of contact between people and the natural environment. Conservation efforts are now supplemented by stringent biosecurity measures that strictly limit the importation of further exotic species to New Zealand. CONCLUSION

New Zealand as a land mass has a long and complex biological history. Proximity to a large continent (Australia) and composition of continental crust have complicated inferences about the origins of the biota. The biota has assembled over many millions of years, but relatively few (conceivably none) of the lineages that must have been present when Zealandia broke from Gondwana survive today. Episodes of extinction and colonization have acted upon the biological assemblage, but, as with most islands, the key influence on New Zealand’s biological character has been speciation. SEE ALSO THE FOLLOWING ARTICLES

Bird Radiations / Flightlessness / Gigantism / Madagascar / New Caledonia, Biology / New Zealand, Geology / Vicariance FURTHER READING

Campbell, H., and G. Hutching. 2007. In search of ancient New Zealand. Hong Kong: Penguin Books. Gibbs, G. 2006. Ghosts of Gondwana: the history of life in New Zealand. Nelson, New Zealand: Craig Potton Publishing. Goldberg, J., S. A. Trewick, and A. M. Paterson. 2008. Evolution of New Zealand’s terrestrial fauna: a review of molecular evidence. Philosophical Transactions of the Royal Society B 363: 3319–3334. McGlone, M. 1985. Plant biogeography and the late Cenozoic history of New Zealand. New Zealand Journal of Botany 23: 723–749. Pole, M. S. 2001. Can long-distance dispersal be inferred from the New Zealand plant fossil record? Australian Journal of Botany 49: 357–366. Trewick, S. A., A. M. Paterson, and H. J. Campbell. 2007. Hello New Zealand. Journal of Biogeography 34: 1–6. Wardle, P. 1963. Evolution and distribution of the New Zealand flora, as affected by Quaternary climates. New Zealand Journal of Botany 1: 3–17.

NEW ZEALAND, GEOLOGY HAMISH CAMPBELL GNS Science, Lower Hutt, New Zealand

CHARLES A. LANDIS University of Otago, Dunedin, New Zealand

In geological terms, New Zealand may be regarded as an emergent portion of a sunken continent. This is unusual globally; the Kerguelen Plateau may be the only other

NEW ZEALAND, GEOLOGY

673

modern example of a large tract of submerged continental crust, or sunken continent. The New Zealand land area represents just 7% of a much larger area of submerged continental crust referred to as Zealandia. New Zealand owes its existence as land to tectonic collision along a northeastsouthwest-oriented segment of the plate boundary separating the Australian Plate (to the west) from the Pacific Plate (to the east), a process that has been especially active since the onset of Miocene time 24 million years ago. THE ISLANDS OF NEW ZEALAND

The New Zealand “mainland” is oriented more or less northeast-southwest, stretching some 1500 km between latitudes 34° and 47° S (Figs. 1, 2). It is largely the product of plate collision tectonism and consists of two large islands, North Island (150,437 km2) and South Island (113,729 km2); the smaller Stewart Island (1,680 km2); and about 700 much smaller islands, including islets that are in close proximity to the coast. The largest of these include Great Barrier, D’Urville, Resolution, Waiheke, Secretary, Arapawa, Ruapuke, Codfish, Big South Cape, and Kapiti. Beyond the “mainland” but considered part of New Zealand are the “offshore islands.” These are White Island (Fig. 3), the Three Kings Islands, and Kermadec Islands to the north; the Chatham Islands (including Chatham, Pitt, Mangere, Southeast, Little Mangere, Forty Fours, Sisters,

FIGURE 2 The major islands of New Zealand presented on a map of

the “New Zealand Continent.” This is a bathymetric map color-coded to show water depth. The 2,500-m isobath serves as a proxy for the boundary between oceanic crust and continental crust. New Zealand represents 7% of the area of Zealandia. In this context New Zealand can be regarded as an emergent part of a large sunken continent. Location of the active plate boundary between the Australian and Pacific plates is shown. The “teeth” indicate direction of down-going subducting slabs of oceanic crust. Diagram by GNS Science.

FIGURE 1 “Mainland” New Zealand is the product of active plate colli-

sion between the Pacific Plate (to the east) and the Australian Plate (to the west). This image, taken from the Space Shuttle (NASA), is looking north from above southernmost South Island. From space, the South Island appears as a giant welt or bruise within the Earth’s crust. This is not surprising as it is the locus of continent-continent collision, giving rise to the Southern Alps. The plate boundary is the Alpine Fault, which appears as a remarkably straight line that runs northeast–southwest down the western side of the snow-capped high peaks of the Southern Alps. In a sense, the New Zealand land surface is being pushed up and held up against its will. If and when the tectonic forces responsible diminish or cease, the land will slowly subside. Photograph by NASA.

674

NEW ZEALAND, GEOLOGY

and Pyramid), Antipodes Islands, and Bounty Islands to the east; and Solander Island, the Snares Islands, and the Subantarctic Islands to the south, namely the Auckland Islands and Campbell Island. Whereas “mainland” New Zealand consists mainly of diverse continental rocks (igneous, metamorphic, and sedimentary) of Paleozoic, Mesozoic, and Cenozoic age, a number of the smaller New Zealand islands are volcanic. Some owe their existence to subduction-related late Cenozoic (Miocene-Recent) arc volcanism including Great Barrier, Little Barrier, Great Mercury, Solander, White, and the Kermadec Islands, but of these only White Island and the Kermadec Islands (Raoul, Macauley, Curtis, Cheeseman, and L’Esperance Rock) are classed as active. Conversely, two islands in the Bay of Plenty, Mayor and Motiti, relate to extensional tectonism associated with active rifting of the Taupo Volcanic Zone (TVZ).

FIGURE 3 White Island, New Zealand’s most active volcano, located

offshore some 50 km north of the Bay of Plenty coast of the northeastern North Island. This is an example of an active subduction-related volcano. Photograph by GNS Science/Lloyd Homer.

The TVZ is often referred to as a “back arc basin,” but it is nevertheless a rift within continental crust and hence is the locus of new continental crust. “Fresh” granite magmatism at depth manifests itself at the surface as rhyolite volcanism accompanied by voluminous production of pumice, ash, and ignimbrite. This continental rift is responsible for the Y shape of the North Island, involving relative rotation of Northland (trending to the northwest) and East Cape (trending northeast) with respect to Mount Ruapehu (approximately). The TVZ is the main locus of rifting with a history of active normal faulting and east-west extension that has been measured at rates of about 10 mm per year. Islands proximal to (Mercury Islands) and to the north of the Coromandel Peninsula (Great Barrier, Little Barrier) in the eastern part of the Hauraki Gulf may also be regarded as rift related but are of older Miocene– Pliocene age. A number of other more substantial islands relate to extinct Cenozoic (Miocene) intraplate basalt volcanism, including the Auckland Islands and Campbell Island. The Chatham Islands owe their existence to Neogene uplift, but their origins can be traced back to older Cretaceous intraplate basalt volcanism with subsequent lesser Cenozoic (Eocene and Pliocene) activity. New research suggests that some tectonic process (other than volcanism) must also be at work. Such a process is necessary to explain uplift within the past 3 million years during Late Pliocene to Pleistocene time. Most probably the uplift relates to localized perturbation (thermal inflation or upwelling) within the mantle.

Several islands in the western part of the Hauraki Gulf near Auckland, including Rangitoto and Ponui, relate to active intraplate basalt volcanism. Most nonvolcanic islands around New Zealand relate to erosion resulting from Pleistocene glaciation and sealevel rise, especially in Fiordland (Resolution, Secretary, Anchor, Cooper, Chalky, Long, and Coal) and Stewart Island (Ruapuke, Codfish, and Big South Cape), but also the Bounty Islands and Antipodes Islands. Note that, of course, volcanic islands have also been subject to these processes, including the Auckland Islands and Campbell Island. Some islands relate to downdrop (crustal sag) associated with subduction of the Pacific Plate below the Australian Plate, such as in Cook Strait between the North and South islands, and in the Marlborough Sounds (Kapiti, Mana, D’Urville, Arapawa, Blumine, Forsyth, Chetwode, Stephens). In this context, the Marlborough Sounds are indeed a drowned landscape, but not because sea level has risen; rather, the crust has sunk (Fig. 4). It is thought that continental crust on the Australian Plate has been drawn down by the descending “slab” of oceanic crust on the adjacent Pacific Plate. In summary, the islands of New Zealand are of diverse origin. Most, including the largest islands (North, South, Stewart), are the product of tectonic uplift, localized tectonic downdrop or subduction-related volcanism as a result of collision between the Pacific Plate and the Australian Plate. Other islands relate to processes such as localized continental rifting, intraplate volcanism, and possibly localized mantle inflation. Yet others are the

FIGURE 4 The Marlborough Sounds, northeastern South Island, an

example of island formation in a drowned landscape caused by tectonic downwarp. This view is looking north across Queen Charlotte Sound with part of D’Urville Island (top left) and Arapawa Island (right). Photograph by GNS Science/Lloyd Homer.

NEW ZEALAND, GEOLOGY

675

product of erosion relating to a combination of Pleistocene glaciation and/or sea level rise. In order to explain this diversity, it is necessary to consider the geology of New Zealand. Let us start by considering aspects of New Zealand today. YOUNG NEW ZEALAND

New Zealand is widely regarded as “young” geologically. Perhaps the main basis for this claim is the presence of active and very conspicuous volcanoes in the central North Island. These include subduction-related arc volcanoes that represent a southwestern extension of the Pacific “Ring of Fire”: the Taranaki, Ruapehu, Tongariro (Ngauruhoe), Tarawera, and White Island volcanoes. Of these, only Taranaki is more than 100,000 years old. They each have long eruption histories that vary in frequency. Other less conspicuous volcanoes, also classed as active but considered much more dangerous and productive, are the eight rift-related rhyolite calderas within the TVZ including Taupo and Okataina (Rotorua). The eruption products (tephras, ignimbrites) of these “supervolcanoes” are widespread and conspicuous throughout central North Island and are present in Auckland. These volcanoes are less than two million years old, and Taupo is less than 500,000 years old. Other reasons for regarding New Zealand as geologically young relate to earthquake activity, the presence of active faults, and spectacular fault scarps. More than 16,000 earthquakes are recorded annually in New Zealand. Auckland, New Zealand’s largest city, is built on an active volcanic field of about 50 volcanic centers (cones, maars, craters). The age of eruption of each cone is imprecise, but the entire field is less than 250,000 years old, and the youngest eruption (Rangitoto Island) was about 600 years ago. By contrast, the Canterbury Plains are composed of an extensive sheet of youthful alluvial gravels related primarily to active braided river development and redistribution of glacial outwash during Pleistocene time (the past 1.8 million years), mainly since the last ice age just 20,000 years ago. Correlative gravel accumulations, widespread throughout the South Island on both sides of the Southern Alps, relate to vigorous uplift and erosion of the Southern Alps, largely within the last five million years. In summary, there is ample evidence of active geological processes in New Zealand. The effects of active volcanism, active deformation of the landscape, and mountain building are conspicuous. However, it cannot be claimed

676

NEW ZEALAND, GEOLOGY

that New Zealand is necessarily younger than many other parts of the world, especially those that are also affected by active plate boundary collision, such as Japan, Indonesia, Alpine Europe, or Himalayan Asia. To explain and understand these active geological processes better, it is first necessary to consider the presentday tectonic setting of New Zealand. TECTONIC SETTING OF NEW ZEALAND

Modern New Zealand straddles an active plate boundary zone. The actual boundary can be drawn as a line that extends from the southern end of the Tonga–Kermadec Trench and runs down (heading from northeast to southwest) but entirely offshore of the eastern coast of the North Island, cuts across the upper half of the South Island along the Hope Fault, then connects with the Alpine Fault on the western side of the Southern Alps, and heads offshore at the entrance of Milford Sound (western Fiordland) and into the Puysegur Trench. In broad terms, the Pacific Plate is moving westwards and the Australian Plate is moving northwards. The relative rate of plate motion convergence (collision) has been measured using satellite GPS surveillance at about 4–5 cm per year. In terms of the crust, the collision varies greatly along the New Zealand segment of the plate boundary, and it is this variation that dictates the diversity of topography and geological processes manifest in the New Zealand landscape. The entire North Island consists of continental crust of the Australian Plate and is subject to collision with oceanic crust on the subducting Pacific Plate. This is an example of conventional continent–ocean collision, whereby dense oceanic crust slides down or is drawn down below less dense continental crust (subduction). Hence, the active subduction-related volcanoes are restricted to the North Island. The South Island is utterly different. It represents collision between continental crust of the Australian Plate and continental crust of the Pacific Plate (continent–continent collision). Hence, the Southern Alps are comparable to the European Alps and the Himalayas, and there is an absence of subduction-related volcanism. At the southern end of the South Island the plate boundary reverts back to continent–oceanic collision, but here it is subduction of oceanic crust on the Australian Plate beneath continental crust of the Pacific Plate. This is the reverse of what is happening in the North Island. Solander Island is the only subductionrelated arc volcano above sea level on this segment of the plate boundary and may rightly be regarded as part of the Pacific Ring of Fire.

Much of the interest and geological intrigue of New Zealand relates to the diversity of geology afforded by the present tectonic setting, all within one small part of the globe. The nature and age of the actual rocks that form the substrate or basement of New Zealand are considered next in terms of the three major episodes in the geological history of the New Zealand land mass: Gondwanaland (545– 83 million years ago), Zealandia (83–23 million years ago), and New Zealand (23 million years ago to the present). GEOLOGICAL HISTORY OF NEW ZEALAND Gondwanaland

The oldest rocks known from mainland New Zealand are of Early Cambrian age, about 505 million years old (Tasman Formation). These rocks are fossil-bearing limestone and subduction-related volcanic rocks exposed in northwest Nelson, South Island. However, older rocks are known from Campbell Island and are of latest Precambrian age, about 545 million years old (Complex Point Schist). In geological terms, these occurrences fall within the Takaka and Buller terranes (respectively) of the Western Province. Modern geological interpretation (mapping) of the older basement rock of New Zealand recognizes tectonostratigraphic units referred to as terranes (Fig. 5). These are elongate belts of rock measurable in terms of crustal thickness (between 5 and 25 km thick), tens of kilometers wide, and hundreds of kilometres long. Each terrane is composed of variably metamorphosed sedimentary or volcanic rocks that are fault-bounded and share a common history. Series of related volcano–sedimentary terranes are grouped into “provinces.” In New Zealand there are two provinces, Western and Eastern, separated by the predominantly igneous Paleozoic–Mesozoic Median Batholith (effectively a long-lived magmatic province) that is dominated by granite and gneiss. Two Paleozoic terranes are recognised in the Western Province and seven Paleozoic–Mesozoic terranes in the Eastern Province (from west to east): Brook Street, Murihiku, Dun Mountain–Maitai, Caples, Waipapa, Rakaia, and Pahau. The dominant sedimentary rock type that is common to all nine terranes is greywacke and its metamorphic equivalent, schist. More accurately described perhaps as muddy sandstone or silty sandstone derived from erosion of either granite or volcanic rocks, greywacke, more than any other rock type, is characteristic of New Zealand. More importantly, it is characteristic of sediment accu-

FIGURE 5 Distribution of basement rocks in New Zealand in terms of

tectonostratigraphic units: nine terranes grouped into provinces. The orientation of these elongate belts of rock reflects systematic continental growth of eastern Gondwanaland by tectonic accretion during 459 million years of Paleozoic–Mesozoic time. The oldest continentward terranes are to the west, the youngest oceanward terranes are to the east. Diagram by GNS Science.

mulation in marine continental accretionary margin settings. Greywacke dominates and pervades the New Zealand landscape. Systematic research on the provenance or source of the original sediments that make up the sedimentary (greywacke) rocks of all nine New Zealand terranes has established that they are all derived from unique yet closely related source areas. Broadly speaking, these terranes represent a series of more or less independent sedimentary basins. Each terrane represents sediment accumulation from distinctive source areas within eastern Gondwanaland, and mainly the Queensland sector of what is now Australia. Furthermore, the terranes are all allochthonous to varying degrees. Those of the Eastern Province are more so than those of the Western Province and indeed have been referred to as “exotic.” This means that they

NEW ZEALAND, GEOLOGY

677

have been tectonically removed (excised along major faults) and are now detached from both their original source rocks and their original geographic setting. This prolonged period of accumulation and subsequent tectonic displacement spans 459 million years, from Cambrian to Cretaceous time—from about 542 to 83 million years ago—and represents the Gondwanan or Gondwanaland history of New Zealand. Both provinces, along with the Median Batholith, were assembled into their present configuration with respect to each other within Early Cretaceous time. They were assembled during oblique plate boundary convergence (collision tectonism), similar to what is happening today in the North Island. It involved prolonged subduction of oceanic crust of the Panthalassa Ocean floor on the paleo–Pacific Plate beneath continental crust of eastern Gondwanaland. In this regard, the basement rock of New Zealand may be regarded as a remnant of a Cretaceous land mass composed of continental crust that developed from sustained growth by continental accretion over a period of about 460 million years along a segment of the eastern margin of Gondwanaland. As would be expected from any orderly accretionary process, the terranes of New Zealand reflect systematic eastward growth from the continent outwards. In summary, in Cretaceous time this part of the world, best described as proto–New Zealand, established itself in two significant respects: as continental crust and as land. The basement rocks of New Zealand are largely composed of sediments that accumulated in the Panthalassa Ocean, deposited by rivers draining eastern Gondwanaland, and that were subsequently bulldozed (accreted) onto the eastern Gondwanaland continental margin by subduction. The “bulldozer” was the oceanic crust of the paleo–Pacific Plate, and it was at work for about 460 million years. Contrast this with the present “bulldozing” of the Pacific Plate, which has been at work for only about 25 million years. Cretaceous time was seminal in the geological history of New Zealand, with comparative stability reigning for about 50 million years, until the rifting of Zealandia away from Gondwanaland. Zealandia

Zealandia is the name of the large fragment of eastern Gondwanaland that broke away about 83 million years ago (Fig. 6). Almost half the size of Australia, this new continent drifted in a northeasterly direction in response to continental rifting and the formation of fresh oceanic crust in

678

NEW ZEALAND, GEOLOGY

FIGURE 6 A reconstruction showing the configuration and extent of

Zealandia about 90 million years ago, prior to rifting from eastern Gondwanaland about 83 million years ago and opening of the Tasman Sea. Note the location of “mainland” New Zealand (dark green shading) and also New Caledonia. Diagram by GNS Science.

what is now the Tasman Sea floor. The age and duration of this sea floor spreading process is based on paleomagnetic interpretation coupled with radiometric dating. The oldest sea floor is dated at 83 million years, and the youngest is about 60 million years old, so the Tasman Sea floor formed in less than 25 million years and has been more or less inactive since earliest Cenozoic (Paleocene) time. As it drifted away, the continental crust of Zealandia was stretched, thinned, and cooled, and as a consequence the continent slowly sank. It did so over a sustained period of 60 million years until about 23 million years ago. The crust of Zealandia effectively lost buoyancy and sank up to 2500 meters with respect to sea level. Modern bathymetric mapping reveals the extent of continental crust of Zealandia as shallower than the 2500-meter isobath. The Late Cretaceous to mid-Cenozoic (earliest Miocene) geology of New Zealand provides an excellent record of this process: the slow, steady submergence and drowning of Zealandia as a function of marine transgression—inundation by the sea. A transgressive sequence becomes increasingly more oceanic as the sea encroaches farther onto the land. A New Zealand–wide marine trangressive sequence accumulated during the time period from 83 to 23 million years ago. The geology suggests that sea level was rising, but what was really happening was tectonic sinking of the crust. Superimposed on this primary long-term process, involving sea level changes measurable in hundreds to thousands of meters, was second-order, much shorter-term sea level fluctuation involving tens of meters of rise and fall. Recent analysis of the geological evidence for land in the New Zealand region of Zealandia in earliest Miocene time (24–22 million years ago) has raised the possibility

of total submergence. Current geological evidence cannot conclusively confirm nor deny the existence of continuous land at that time. Furthermore, a strong geological argument can be made in favor of total submergence. Needless to say, this idea is controversial, because it has major implications for the origins and antiquity of the native terrestrial biota of New Zealand. This does not mean to say that all of Zealandia was submerged 23 million years ago. New Caledonia has been emergent for at least 30 million years, although it was totally submerged prior to the onset of Oligocene time (34 million years ago). So, whereas the New Zealand region of Zealandia may have been completely submerged about 23 million years ago, New Caledonia was land at that time. This is an important consideration, because New Caledonia may have been a significant source of the ancestors of the native terrestrial biota of modern New Zealand, even though it lies more that 1200 km to the north of New Zealand. Maximum submergence of the New Zealand region of Zealandia culminated in latest Oligocene and Early Miocene time, about 23 million years ago, with an abrupt change in regional tectonism. This heralded a new active configuration along a segment of the boundary between the Pacific and Australian plates that cut its way clean through continental crust of old Zealandia. Zealandia no longer represented a single tract of largely submerged continental crust. It was now split asunder, shared between two plates. New Zealand

The present plate boundary configuration can be traced back to Eocene time, 45 million years ago. However, collision tectonism became vigorous with the onset of Miocene time. This process manifests itself in the geological record as a New Zealand–wide regressive sequence. In other words, the geology reflects retreat of the sea from the New Zealand region, commencing in Early Miocene time and continuing to the present day. The cause, however, is not sea-level change per se, but sustained tectonic uplift as a function of plate collision. It is this process that gave rise to the New Zealand land mass as we know it and it continues today. In this context, the geological history of New Zealand, strictly speaking, only relates to the last 23 million years. DISCUSSION

In light of the preceding account of the geology of New Zealand, both “mainland” New Zealand and “off-shore islands,” it is clear that New Zealand is very much part

of a much bigger entity, namely the Zealandia continent, which rifted away from Gondwanaland about 83 million years ago and slowly submerged over a period of 60 million years. New Zealand has gradually become emergent in response to tectonism associated with a major change in plate configuration about 23 million years ago. New Zealand owes its location to the modern plate boundary. Without it, New Zealand would still be substantially underwater. New Zealand can be regarded as part of a largely sunken continent. The Kerguelen Plateau in the southern Indian Ocean may be considered as another possible example of a sunken continent or large tract of submerged continental crust. This modern understanding of New Zealand has been realized primarily from fresh insight into our understanding of the nature of the Earth’s crust, how it works, and at what rates. This enables us to better explain the origin of the New Zealand islands, and yet some uncertainties and mysteries remain. For instance, an explanation for the Pliocene–Recent uplift of the Chatham Islands is required. The Chatham Islands are too far removed to be affected by any uplift associated with the active plate boundary through “mainland” New Zealand, and late Cenozoic volcanism seems insufficient (volumetrically) to account for tectonic uplift that appears to involve the entire Chatham Islands area of a least 10,000 km2. One possible explanation, a largerscale mantle inflation effect, is being considered. Like “mainland” New Zealand, the Chatham Islands and several other “offshore” islands are continental: the Snares Islands, Auckland Islands, Campbell Island, the Antipodes Islands, and Bounty Islands. They all have basement rocks of continental affinity: granite, greywacke, or schist. In addition, most of these are also the centers of basaltic intraplate volcanism. The presence of basement on these islands either at or above sea level is a function of chance preservation beneath a resistant cap of much younger basaltic rock. In a sense, the volcanoes act as storage vessels, harboring for scientific posterity a record of the rock they buried when they erupted. All of these islands appear to be Miocene or younger. They are variously subduction-related, associated with continental rifting in the TVZ, or intraplate basalt volcanoes. The latter are of particular interest because a satisfactory explanation of their location remains as yet uncertain. These include the Auckland Islands, Campbell Island, the Chatham Islands and the islands of the Auckland Volcanic Field. The idea that the New Zealand ‘mainland’ may not have existed in earliest Miocene time, about 23 million years

NEW ZEALAND, GEOLOGY

679

ago, is of considerable interest. Some of the key evidence is geomorphological,and in particular the origin or formation of regional planar surfaces in the New Zealand landscape. These features have been reinterpreted as remnants, not of a terrestrial “peneplain” but a regional wave-cut surface referred to as the Waipounamu Erosion Surface (WES). This surface formed as a result of marine planation between Cretaceous and Miocene time, commensurate with the slow sinking and submergence of Zealandia. Inevitably, the WES is commonly superimposed upon an older Cretaceous land surface. However, the Cretaceous land surface is easily recognized because it does not form planar geomorphic features in the landscape but exhibits demonstrable relief and is invariably overlain by fluvial sediments, whereas the younger WES is invariably overlain by marine sediments. The geological origins of New Zealand’s islands continue to fascinate the research world, and yet it is somewhat surprising that so much new insight has been gained in recent years. It is not easy work. Most difficult is determining the age of surfaces in the landscape. Conversely, it is relatively easy to determine the age of the rocks into which the surfaces are cut. Geomorphology, stratigraphy, radiometric dating, and a keen understanding of crustal processes remain the principal keys to further geological understanding of islands and their histories. SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / New Caledonia, Geology / New Zealand, Biology / Volcanic Islands FURTHER READING

Campbell, H. J., and G. J. Hutching. 2007. In search of ancient New Zealand. Auckland, New Zealand: Penguin Books and GNS Science. GNS Science Website. http://www.gns.cri.nz. Graham, I. J. (ed.) 2008. Continent on the move. Geological Society of New Zealand Miscellaneous Publication 124. Wellington, New Zealand: Geological Society of New Zealand Inc. Hicks, G., and H. J. Campbell. 1998. Awesome forces: the natural hazards that threaten New Zealand. Wellington, New Zealand: Te Papa Press. Landis, C. A., H. J. Campbell, J. G. Begg, D. C. Mildenhall, A. M. Paterson, and S. A. Trewick. 2008. The Waipounamu erosion surface: questioning the antiquity of the New Zealand land surface and the terrestrial fauna and flora. Geological Magazine 145: 173–197. LINZ Website (for inventory of New Zealand islands): http://www.linz .govt.nz. Mortimer, N. 2004. New Zealand’s geological foundations. Gondwana Research 7: 261–272. Te Ara online encyclopedia of New Zealand Website. http://www.teara .govt.nz.

NORFOLK ISLAND SEE PACIFIC REGION

680

NUCLEAR BOMB TESTING

NUCLEAR BOMB TESTING EDWARD L. WINTERER Scripps Institution of Oceanography, La Jolla, California

Nuclear bomb testing by the United States began just two years after the end of World War II and continued intermittently until late 1962, running up a total of 105 tests of fission (A-bomb) and fusion (H-bomb) weapons, all of them in the atmosphere or in shallow lagoon waters. The tests were conducted not only to evaluate successive bomb designs but also to observe the effects of nuclear explosions on ships and crews. Islanders were evacuated from Bikini atoll (now Pikini) successively to a series of other atolls, where they fared poorly because of poor environments for fishing and growing crops. Their home island is still not safe. On Eniwetok atoll (now Enewetak), A-bomb tests began in 1948, followed by H-bomb tests in 1952, which later alternated between there and Bikini. A further 24 U.S. tests were made as air drops close to Christmas atoll, a British island in the Line Islands chain. Two very highaltitude rocket tests over Johnston atoll in 1962 closed out the American atoll testing. Meanwhile, the British, at Christmas and Malden, exploded nine devices, and the French at Mururoa conducted some 175 tests, including 134 underground tests in lagoon boreholes, finally ending all testing there in 1996. EARLY BIKINI TESTS

Only a few months after the detonation of the atomic bombs over Hiroshima and Nagasaki in August 1945, U.S. military and political authorities began to plan for improvements in the U.S. atomic arsenal and to plan for testing of these new bombs at Bikini (now Pikini) atoll in the Marshall Islands in the western Pacific (Fig. 1). The Marshall Islands had been under Japanese rule under a League of Nations Mandate but were occupied by U.S. military forces during 1944 and 1945. From the war’s end until the establishment of a UN trust territory in 1947, the islands were under U.S. military control. In February 1946, the U.S. military commander of the Marshall Islands obtained the permission of the chief of the group of 167 Bikini people to move all the people to another atoll (Rongerik, about 300 km east of Bikini) during the tests. By early March 1947, the Bikinians were on Rongerik, and Americans began arriving at Bikini

165

Northern Mariana Islands (U.S.)

Guam (U.S.)

170

Wake Island (U.S.)

NORTH PACIFIC OCEAN

National Capital 0

MARSHALL ISLANDS

100

200 killometers

0

100

200 Miles

FED. STATES OF MICRONESIA Equator KIRIBATI

NAURU

15

PAPUA NEW GUINEA

Coral Sea

15

SOLOMON ISLANDS

Taongi TUVALU SOUTH PACIFIC OCEAN

VANUATU

FIJI

AUSTRALIA

Bikar

Bikini

Enewetok

Utirik

Rongerik Ailinginae

Taka

Rongelap

Ailuk Jemo

Wotho

10

Mejit

10

Likiep

Ujelang

Ujae

Lae

NORTH

Watje

Roi

Ebadon

reef

Kwajalein

Erikub

Kwajalein

Maloelap

Lib

Aur Namu

PACIFIC

Jabwot Ailinglapalap

OCEAN

Arno

Majuro

Majuro

Mili

Jaluit

FEDERATED STATES OF MICRONESIA

Namorik

Knox

Kili

Kosrae

5

5 Ebon

KIRIBATI

Makin

Butaritari

165

170

FIGURE 1 Marshall Islands. Image courtesy of the U.S. Central Intelligence Agency. (As a work of the United States Government, all images by the

CIA are in the public domain, with the exception of classified information.)

atoll to prepare for the tests, termed Operation Crossroads, which were conducted in July. Besides the military and the bomb scientists and technicians, the arriving group included teams of scientists—oceanographers, biologists, and geologists—to map the atoll and its waters prior to any explosions. One of the main purposes of the tests was to evaluate the effects of atomic bombs on ships, planes, and equipment, so a fleet of some 90 target vessels, along with some

animals was assembled in the lagoon and on Bikini atoll. The first test, named Able, an air drop about 500 feet over the lagoon, caused little visible damage to the islands, whereas the second, Baker, submerged at a depth of about 30 m, threw up a huge cloud of water and irradiated debris from the floor of the lagoon (Fig. 2). The two test bombs were each 23 kilotons. Able damaged ships, but radioactivity faded enough for crews to board the vessels in two days. Baker sank eight ships, and the radiation persisted

NUCLEAR BOMB TESTING

681

49 kilotons, and the tests (successful) were to evaluate a new, more efficient design than the ones for the bombs used over Japan and at Bikini. Most of the people of Eniwetok had already, in May of 1946, been evacuated to Kwajalein atoll, in preparation for the Bikini tests. In December 1947, the balance of the Eniwetok population was taken to Ujelang atoll. After three more years, in April 1951, a new series of fission bomb tests, termed Operation Greenhouse, was conducted at Eniwetok. The series comprised at least four shots, with yields from 14 to 81 kilotons, and included an odd design (test George) that was intended to test the mechanism of radiation implosion, an element in the design of the H-bomb (Fig. 3). FIGURE 2 Cloud of water vapor and dust from Bikini lagoonal bottom

sediments resulting from Baker explosion, July 25, 1946. The dome evolved to a mushroom cloud as it rose. Image courtesy of the U.S. Government Defense Threat Reduction Agency. (As a Work of the United States Government, all images by the Defense Threat Reduction Agency are in the public domain.)

for weeks, preventing all but brief visits to remaining ships. The ships were towed to Kwajalein atoll for decontamination and offloading ammunition, work that continued into 1947. Eight ships were towed to the United States for inspection, and 12 were sailed there by their crews. All the rest were sunk. In 1947, scientists returned to Bikini atoll for a resurvey. Meanwhile, the Bikini people were in worsening physical condition on Rongerik and pleaded to be relocated. Plans were made to move them to Ujelang atoll, about 60 km southwest of Eniwetok atoll, but instead, Eniwetok was evacuated and its people moved to Ujelang atoll. The Bikinians were moved first to one of the islands on Kwajelein atoll and finally, in November 1948, to Kili atoll, a tiny (0.75 km2) island about 800 km southeast of Bikini. There they stayed until October 1972, when a few families moved back to Bikini, which was declared safe by U.S. authorities. In 1975, a restudy discovered that the water and food available on Bikini were too radioactive for human consumption, but the population was not moved; rather, food was brought in by ship. In 1978, the remaining people on Bikini were moved back to Kili. ENIWETOK TESTS

Test operations after Crossroads were moved to Eniwetok atoll (now Enewetak), where a series of three fission bombs were exploded from 60-m towers on three different islands in April and May 1948, during a 10,000man operation named Sandstone. Yields were from 18 to

682

NUCLEAR BOMB TESTING

FIGURE 3 Fireball from test device George, 20 milliseconds after deto-

nation. Eniwetok, 1951. (This work is in the public domain in the United States as a work of the U.S. Federal Government under the terms of Title 17, Chapter 1, Section 105 of the U.S. Code.)

In November 1952, there began the first tests of the thermonuclear hydrogen fusion bomb, first declared a goal by President Truman in 1950, following shortly after the first successful Soviet fission bomb test. In test Mike, the first H-bomb was detonated, with a yield of 10.4 megatons (versus the 15 kilotons of the Hiroshima fissiontype bomb). The cylindrical device, called Sausage by its builders, was sheltered in a “cab” and weighed about 80 tons (Fig. 4). It comprised two stages: a fission bomb as the primary stage and a flask of liquid deuterium (a heavy isotope of hydrogen) with a central rod of plutonium and an enveloping 5-ton layer of uranium as the secondary. The device was encased in a cylinder of 30-cmthick steel about 2 m across and 6 m long, and left a crater more than 1.5 km wide and 50 m deep. The fireball was over 5 km wide, and the mushroom cloud rose to 17 km in less than 90 seconds. Little more than a minute later it had reached 36 km in altitude, with the top eventually

(Shrimp) of an appearance similar to the huge Sausage, but weighing only about 24 tons. The novel feature was that about 40% of the fuel was of enriched lithium-6 deuteride. The case was of aluminum rather than heavy steel. The unexpectedly high yield of 15 megatons was a surprise and resulted from the supposedly inert lithium-7, which, when struck by high-energy neutrons, could fragment into a helium and a tritium atom. The tritium caused very energetic fusion that increased the expected 4–8-megaton yield. Bravo created a crater on the reef rim and adjacent part of the lagoon about 2 km wide and 75 m deep, and threw all the excavated debris into the sky. The cloud rose in only six minutes to a height of 40 km. After 8 minutes, even the cloud bottom had risen to about 15 km. A few weeks after this test, the new bomb design was weaponized, and a bomb was put into production (275 bombs, 4.5-megaton yield) during 1955 and 1956. The fallout moved mainly east and northeast and blanketed atolls and ships as distant as 250 km with ash, exposing the (completely unwarned) people to dangerous levels of radiation (Fig. 5). A Japanese fishing vessel, the Fifth Lucky Dragon, working about 80 km northeast of Bikini, was blanketed by ash, and its crew received about 300 rad of radiation. One member of the crew died of this exposure. It took about two days for the military to send evacuation ships to the affected atolls of Rongelap, Rongerik, Utirik, and Ailinginae, where people received estimated doses of from 14 to 175 rad. Bikini atoll was declared too contaminated for people, and further tests, which used non-enriched lithium deuteride, were controlled by radio from a ship. The first device was detonated

FIGURE 4 “Sausage” device Mike (vertical cylinder) at Eniwetok.

Seated man with mandolin gives scale. 1952. (This work is in the public domain in the United States as a work of the U.S. Federal Government under the terms of Title 17, Chapter 1, Section 105 of the U.S. Code.)

spreading out to a diameter of 160 km. The blast created a crater about 1800 m in diameter and almost 50 m deep where the isle of Elugelab had once stood; the blast generated 6-m water waves. Irradiated coral debris fell on ships stationed 50 km from the blast. Two weeks later, fission bomb King was airdropped, testing a 500-kiloton design, at about the limit for a fission bomb at that time and finishing the test series. LATER TESTS AT BIKINI

Further testing of H-bombs, with the first test on March 1, 1954, took place on an isle at the northwestern corner of Bikini atoll. This test, named Bravo, was of a device

EFFECTIVE ARRIVAL TIME (HOURS) 0

6

7

8

9

10

11

12

13

14

15

16

17

18 19 20

BIKAR ATOLL

100

GROUND ZERO

61

300 1000

1500

3000

3400

RONGELAP ATOLL

2100 80 100

200

30

2700 2000 1900 1300 3300 1500 880

900

60 120

BIKINI ATOLL 134 104

420

830

170

220

120 ALINGINAE ATOLL

0 10 20

40

60

80

100

120

430

UTIRIK ATOLL

550 240

970

RONGERIK ATOLL

140

160

TAKA ATOLL

180

200

220

240

260

280

14 18

300

320

340

DISTANCE FROM GROUND ZERO (MILES) FIGURE 5 Map showing levels of radiation (in millirads) downwind from event Bravo, March 1, 1954. (This work is in the public domain in the United

States as a work of the U.S. Federal Government under the terms of Title 17, Chapter 1, Section 105 of the U.S. Code.)

NUCLEAR BOMB TESTING

683

on a barge in the lagoon, rather than onshore like Bravo, and yielded about 11 megatons. There then followed four more tests, from 110 kilotons to 11 megatons, all from barges, ending on May 14, 1954. Most of these produced significant fallout at the atolls east of Bikini. BACK TO ENIWETOK

Testing was resumed, this time back on Eniwetok, in May 1955, in which a small (50-cm diameter) warhead and very small (12-, 20-, and 28-cm diameter), lightweight systems were detonated. It was a test series for proving thermonuclear designs of actual weapons. Through July 21, 1955, 17 shots were detonated, of which seven were at Bikini and ten at Eniwetok, some from barges, some from towers, and one from the ground. Yields ranged from 1 kiloton to 5 megatons. Further operations began at Eniwetok and Bikini in May 1958 and comprised 35 detonations. One shot was from a balloon at 25-km altitude, about135 km northeast of Eniwetok, and another from about 150 m underwater in waters about 975 m deep. Yields ranged from to 1.7 kilotons to 9.3 megatons. The last test (OAK) of a thermonuclear device, with a yield of 8.9 megatons, left a large crater and debris field on the edge of the reef flat (Fig. 6). Two tests over Johnston atoll, at about 16°45’ N, 169°31’ W, were of a warhead designed for antiballistic missiles, which was fired from a rocket to altitudes of

about 80 km. On August 18, 1958, the last explosion was detonated, completing all nuclear testing in the Marshall Islands. Total cost is estimated at about $2.5 billion. BRITISH TESTS AT CHRISTMAS ATOLL AND MALDEN ISLAND

Christmas atoll (1°50’ N, 157°20’ W), now the island of Kiritimati in the Republic of Kiribati (formerly the Gilbert Islands), and Malden Island (4°1’ S, 154°56’ W), about 300 km south of Christmas, were used for a series of nuclear bomb tests beginning in 1956, with the arrival of British military personnel and civilian scientific people. Next, extensive construction was begun on Christmas atoll for wharves, airfields, and living quarters. Between May 15 and 19, 1957, three megaton-size bombs were dropped by airplanes and detonated at about 2500 m over the sea about 50 km south of Malden. Next, tests were conducted over Christmas atoll itself. Women and children had been evacuated to Fanning Island, about 400 km to the north, and the remaining plantation staff was put on a ship, below decks, during the early morning tests. The thermonuclear devices were dropped from airplanes off the island, one (1.8 megatons) on November 8, 1957, and three more (2–3 megatons) from April 28 to September 11, 1958. A bomb in the megaton range was detonated from a balloon on August 22, 1959, and another balloon-dropped device with a yield in the kiloton range on September 23. AMERICAN TESTS AT CHRISTMAS ATOLL

Further tests at Christmas Atoll resumed in April 1962, this time by the United States. In all, 24 devices, ranging in yield from 3 kilotons to 7.6 megatons, were dropped from airplanes and detonated at altitudes of 750 to 4500 m in he vicinity of the atoll. The tests were concluded on July 11, 1962. TESTS AT JOHNSTON ATOLL

FIGURE 6 Side-scan sonar image of crater and debris field created

during 8.9-megaton test OAK on the lagoon side of the west rim of Eniwetok atoll, June 28, 1958. (From Foulger et al., 1986; this work is in the public domain in the United States as a work of the U.S. Federal Government under the terms of Title 17, Chapter 1, Section 105 of the U.S. Code.)

684

NUCLEAR BOMB TESTING

To evaluate the effects of nuclear bombs to be used in the U.S. missile defense system, the United States, on July 9, 1962, at 0900 GMT, conducted a high-altitude nuclear test over Johnston atoll, using a rocket that carried the thermonuclear device to an altitude of about 400 km, where it detonated with a yield of 1.4 megatons. The result was a spectacular light show, visible from Kwajalein, some 2500 km to the southwest, as a white flash that quickly turned into a green ball projecting long white streaks arcing toward the poles, followed by a set of concentric rings that rose high in the sky. Then, a red glow grew to a broad red arc in the northeast, stretching high up toward

the zenith. The Kwajalein show lasted for about 10 minutes. The show at Johnston was even more flamboyant: a red disk at the zenith and down to about 45 degrees from there. A yellowish-white streak grew from the zenith to the north along magnetic north–south. The auroral band widened quickly, then receded from the north and extended to the south. It was over in less than 10 minutes. In Hawaii, about 1600 km away, street lights went dark, TVs and alarms malfunctioned, and microwave phone links were shut down. Radiation belts generated by the explosion crippled some satellites and destroyed others. The experiment had produced a global belt of highenergy electrons, some persisting for five years. Testing of eight more nuclear bombs at Johnston atoll continued to November 4, 1962. These were partly of bombs dropped from airplanes at altitudes of 3000 to 3600 m and missile airbursts at altitudes of 13 to 90 miles. FRENCH TESTS AT MURUROA AND FANGATAUFA ATOLLS

Following a series of 17 nuclear tests in the Sahara from 1960 to 1966, France elected to move its testing to a remote site in the South Pacific, at Mururoa atoll (21°52’ S, 138°55’ W) and Fangataufa (22°14’ S, 138°45’ W atoll, 65 km southeast of Mururoa, where testing began in July 1966. After 41 atmospheric tests of yields from 20 to 1000 kilotons ended in September 1974, there followed a series of 134 underground tests in boreholes, ranging from 5 to 150 kilotons in yield. That series ended on January 27, 1996. Shortly afterward, in March, France and Britain signed the Raratonga Treaty, which created a nuclear-free zone in the South Pacific, thereby foreclosing further tests in Polynesia. The United States signed the treaty, but the Senate never ratified it.

LESSONS LEARNED FROM TESTING AT PACIFIC ATOLLS

Without question, the military preparedness objectives of three nations—the United States, Great Britain, and France—were advanced. Many new designs of nuclear weapons were tested, and the results translated into the construction of both strategic and tactical weapons. The human costs were large: People died or were sickened by exposure to fallout, and people were deprived of their ancestral homelands and exiled—for some people, forever—to unsuitable substitute islands. The biota of the atolls was, at least locally, seriously impacted by some of the test explosions, but the long-term effects are probably very minor compared to more far-reaching environmental changes associated with, for example, global warming. Scientifically, the surveys and test borings of the testrange atolls, especially Bikini, Eniwetok, and Mururoa, have deepened our understanding of the origin of atolls and their physical, biological, and geological makeup. It is highly unlikely that this knowledge could ever have been achieved without the science being done in conjunction with the testing. It was a devil’s bargain. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Marshall Islands FURTHER READING

Foulger, D. 1954. Geology of Bikini and nearby atolls. U.S. Geological Survey Professional Paper 260-A. Johnston, W. R. 2005. Nuclear tests—databases and other material. http:// www.johnstonsarchive.net/nuclear/tests/index.html. Micronesia Support Committee. 1981. Marshall Islands: a chronology: 1944–1981. Honolulu, HI: Sublette, C. 1994–2007. The nuclear weapon archive. http://nuclearweaponarchive.org/.

NUCLEAR BOMB TESTING

685

O OASES SLAHEDDINE SELMI Faculté des Sciences de Gabès, Tunisia

THIERRY BOULINIER Centre d’Ecologie Fonctionnelle et Evolutive, Montpellier, France

The term “oasis” is often taken in its metaphorical and very broad sense: a spot of life within an inhospitable environment. In that way, it has repeatedly been used to designate patches of vegetation in less-vegetated and dry landscapes, isolated ice-free areas in Antarctica, isolated life-rich areas in marine ecosystems, and every other kind of isolated habitat. Even though these systems are comparable in that they are isolated and different from their surroundings, their structure, origin, and evolution, as well as the factors affecting their dynamics, are widely different. Oases are more classically defined as relatively more fertile areas in a desert or wasteland, made so by the presence of water. There are numerous so-defined oases in North Africa, eastern Asia, Australia, and the southwestern region of North America (Baja, in particular), and as such, they represent important models to investigate the role of isolation for the dynamics and conservation of biodiversity in arid landscapes. This article focuses on the original meaning of “oasis” (i.e., a date palm grove in a desert), and provides information on the ecology and dynamics of these particularly poorly known island-like systems.

686

OASES AS SEMI-NATURAL CONTINENTAL ISLANDS

The world distribution of date palm oases shows that they are mainly associated with Saharan regions in Arabia, eastern Asia, the Middle East, and North Africa, where water resources are localized and where the date palm trees Phoenix dactylifera have been cultivated for thousands of years (Figs. 1–3). The structure, evolution, and functioning of these particular continental island systems depends upon a great complexity of environmental, historical, and socioeconomic factors. Their faunas and floras are also greatly shaped by those factors. The desert environment surrounding oases is characterized by harsh climatic conditions, with annual rainfalls rarely exceeding 100 to 200 mm and a summer temperature

N

1000 m

FIGURE 1 Aerial

image of Chebika oasis, southwestern Tunisia

(34°19′00′′N, 7°56′20′′ E). Modified from Google Earth© views.

environment, and because oases are directly dependent on the availability of water and on human activities for irrigation and maintenance, they can be considered as seminatural continental islands. Furthermore, given that the geographic location of an oasis is primarily constrained by the existence of a water spring, oases are not randomly distributed in the desert, but are generally concentrated in some areas where geological conditions have permitted the emergence of groundwater, leading to regional “archipelagoes” of oases.

FIGURE 2 Aerial close-up of Chebika oasis. Modified from Google

Earth© view.

N

Chott Elgharsa (Dry salty lake) Elhamma

Degache

Neflayet

Ibn Chabbat

Tozeur

FIGURE 4 Understory view within Teboulbou oasis (33°50′30′′ N,

10°07′33′′ E; southeastern Tunisia), showing the strong stratification of

Nefta Mrah El-hawar

10 km

the vegetation. Photograph by S. Selmi.

Chott Eljerid (Dry salty lake)

FIGURE 3 Aerial photography of the oasis archipelago of the Jérid area

in southwestern Tunisia. The darker and more irregular patches correspond to traditional oases; the lighter and more regular patches correspond to modern plantations. Modified from Google Earth© view.

often exceeding 40 °C. In this environment trees, are mostly absent, and the vegetation is composed of sparse steppe shrubs. Nevertheless, the particular geological conditions of some areas, in particular the existence of major faults, permitted the emergence of fossil groundwater as springs. The use of this water for irrigation by local human populations has allowed the practice of agricultural activities and the development of a thick vegetation typically composed of three distinct layers (palm trees, fruit trees, and herbaceous plants), which has induced a local microclimate that strongly contrasts with the arid climate of the surroundings (Fig. 4). This so-called oasis effect is responsible for the insular character of desert oases. Given the pronounced physical and climatic contrasts between an oasis interior and the surrounding

INSULAR SPECIFICITIES OF OASIS COMMUNITIES

In spite of their originality and the current threats to which they are exposed, the flora and animal biodiversities of oases have not been much considered by biologists until recently. For instance, within the abundant ecological literature on the biodiversity of continental insular systems, there is very little information on the dynamics of animal and floral oasis communities. The main contributions on these aspects have dealt with oasis bird communities from southern Tunisia (work by the authors). These studies have shown that the occurrence of several non–desert adapted Palearctic bird species, such as the common blackbird (Turdus merula), blue tit (Cyanistes caeruleus), chaffinch (Fringila coelebs), serin (Serinus serinus), orphean warbler (Sylvia hortensis) and woodchat shrike (Lanius senator), in the arid land of southern Tunisia is strictly related to the oasis habitat. They also highlighted the very insular characteristics of oasis breeding-bird communities, notably the fact that their species

OASES

687

richness is linked to oasis-area vegetation characteristics and degree of spatial isolation. In particular, it was shown that the number of breeding species in a given oasis is positively related to its area (Fig. 5). Such a typical species-area relationship is an important characteristic of island systems. Species richness of avian communities also depends on vegetation structure: Oases with a diversified structure (notably in terms of the presence of date palm trees, fruit trees, and herbaceous plants), and with a dense structure showing a sharp contrast with the surrounding desert, host richer bird communities than do more open oases where vegetation is less dense and diversified. In addition, it was shown that bird species richness and composition varied as a function of the degree of geographic isolation of the oasis. Oases close to each other host very similar communities, independently of their area and the quality of their habitat. This suggests that the dispersal of individuals and the exchange of species between neighboring oases play an important role in shaping oasis communities.

the regional communities will also depend on the capacities of the species to extend their area of distribution and to colonize new areas. At the scale of an oasis, the physical characteristics, notably area and the suitability of habitat for the various species, will determine the richness and composition of the oasis’s community, which will be a subset of the regional pool. The physical characteristics of oases are in fact the result of the interactions between various environmental factors, notably water availability, and socioeconomic factors linked to the agricultural system in place in the oasis. If the environmental factors are critical for determining the mere existence of the oasis and can strongly affect its area by limiting its possible extension, then the socioeconomic factors are directly responsible for the quality of the withinoasis habitat that will be available to the various species. Finally, colonization processes from nearby oases, as well as local extinction, seem to play a role. Such metapopulation processes could also largely be responsible for the presence of some species at the regional scale, and thus for the maintenance of regional richness. Comparable mechanisms are likely to be involved in the dynamics of other animal and plant communities, although rates of colonization and extinction may be very different. THREATS TO OASIS BIODIVERSITY

FIGURE 5 Relationship between the number of breeding-bird species

and oasis area in a sample of oases of southeastern Tunisia.

Overall, the work on the breeding bird communities of Tunisian oases has shown that a combination of ecological, geographical, and historical, but also socioeconomic, factors seems to determine the diversity of oasis faunas. At the scale of an oasis archipelago (a geographic area regrouping a set of oases), the sets of species constituting the local communities would be subsets of a pool of potential colonists that would have reached the region. The size and the composition of the pool of potential colonists (regional richness) are determined by the geographic location of the region of interest with regard to sources of potential colonists (notably, the closest woodland and semi-woodland areas). Those characteristics of

688

OASES

For an oasis to exist within a desert matrix, two elements are needed: water and the exploitation of that water by humans for agricultural activities. This results in the creation of a green area. An oasis is thus a sort of semi-artificial continental island, but at the same time it is a very precarious and sensitive ecosystem. It can decay or actually disappear relatively quickly if water resources decline or if the oasis farming practices change, notably when oasis farmers change their way of life. Such factors are currently threatening the mere existence of numerous old oases, which can lead to the loss of the original and sensitive biodiversity that is living in oases. Traditionally, the agricultural systems run by local oasis farmers fed a mainly self-sufficient local economy. Date palm and fruit trees, together with vegetables and food to feed animals, were grown using a stratified system of plantation that enabled the oasis farmers to optimize their use of water while producing a wide diversity of products. The traditional agricultural practice (notably based on complex local irrigation systems) was characterized by reliance on inter-family networks and by the use of traditional tools. It is those traditional agricultural activities that led oases to look like continental islands, and it is also those activities that

seem necessary for the maintenance of local diversity of oasis animal and plant communities. This system, however, has seen and is currently seeing dramatic changes, directly linked with the socioeconomic changes occurring in some oasis societies. For instance, new types of palm plantations have been established in Tunisia since the middle of the twentieth century as a means of maximizing the production and exportation of dates of the well-known Deglet Nour variety. Those modern palm plantations are actual monocultures of date palm trees, which lack the vegetation structure of traditional oases, notably the stratification of the vegetation. The agricultural production is done by employees and uses modern techniques and tools. Those modern palm plantations do not lead to a sharp climatic contrast between the oasis and the surrounding desert, and they do not host a wild biodiversity as rich as that of traditional oases. Moreover, those new oases are in competition with the traditional ones for limited fossil water. Hence, several springs that were irrigating traditional oases have dried up, leading to severe drought problems for those oases. In addition to these water-availability problems, socioeconomic issues linked with the abandonment of traditional agricultural practices are affecting oasis vegetation structure. The largest of such socioeconomic problems are the non-profitability of traditional agricultural production in the current economic context and the tendency for farmers to focus on a monocultural approach, such as date production, or to switch to more rewarding activities such as tourism and industry; the fragmentation of real estate within the oases over generations; the concurrent tendency for young people to move toward cities and to migrate to Europe; and the fast urbanization of some oases. Overall, it seems that if oases are created by humans, then their possible disappearance can also logically be caused by humans. The case of oases illustrates how the development of modern agricultural practices, done to maximize profit, can represent an important threat to the biodiversity of precarious agro-ecosystems. As with other island and island-like entities that host specific and sensitive ecosystems, special efforts should be devoted to the protection of traditional oases. In this respect, a sound knowledge of their animal and plant communities is needed to better understand the dynamics of biodiversity in such systems. Any action plan for the conservation of these very original sociohistorical and ecological islands will have to consider the human aspect of things—economic, social, and cultural issues—as well as the physical and biological aspects.

SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / Hydrology / Lophelia Oases / Metapopulations / Species–Area Relationship / Vegetation FURTHER READING

Kassah, A. 1996. Les oasis Tunisiennes: aménagement hydro-agricole et développement en zone aride. Tunis, Tunisia: Centre d’Etudes et de Recherches Economiques et Sociales. Riou, C. 1990. Bioclimatologie des oasis. Options Méditerranéennes A(11): 207–220. Rodríguez-Estrella, R., M. C. Blazquez, and J. M. Lobato. 2005. Avian communities of arroyos and desert oases in Baja California Sur: implications for conservation, in Biodiversity, ecosystems, and conservation in northern Mexico. J.-L. Cartron and G. Ceballos, eds. Oxford: Oxford University Press, 334–356. Selmi, S., and T. Boulinier. 2003. Breeding bird communities in southern Tunisian oases: the importance of traditional agricultural practices for bird diversity in a semi-natural system. Biological Conservation 110: 285–294. Selmi, S., T. Boulinier, and R. Barbault. 2002. Richness and composition of oasis bird communities: spatial issues and species-area relationship. The Auk 119: 533–539. Zaid, A. 2002. Date palm cultivation. Rome: Food and Agricultural Organization of the United Nations.

OCEANIC ISLANDS PATRICK D. NUNN University of the South Pacific, Suva, Fiji

Those of us who live close to the edges of the world’s continents are likely to be familiar with islands, often as places for recreation or retreat. In the past, they were sometimes places of refuge for people or other biota escaping continental calamities ranging from warfare to ice advance. In a geological sense, such islands are commonly slivers of continent, their connections drowned by the high-sea-level conditions in which we live today. Oceanic islands are quite different, often smaller and more remote, and to find them, the continental dweller generally has to travel much farther offshore, into the hearts of the ocean basins. DEFINING AND UNDERSTANDING OCEANIC ISLANDS

The crust of the Earth is divisible into two distinct types: continental and oceanic. Relative to the ocean surface, the less dense continental crust rests higher—and therefore forms Earth’s largest contiguous land masses—than the denser oceanic crust, most of which is covered by ocean. Only rarely does the ocean crust push above the ocean

OCEANIC ISLANDS

689

surface and form oceanic islands. Two further differences between the continental and oceanic types of crust are also important to mention. The first is their age. The continental crust is old—in places as much as 5000 million years old—but the oceanic crust is almost nowhere older than 120 million years in age. The reason for this astonishing difference has, of course, to do with plate tectonics, widely acknowledged as the driver of the evolution of the Earth’s surface. The mid-ocean ridges, mostly under water, steadily create new oceanic crust along their axes, pushing it laterally outward. At the other end of this oceanic “conveyor belt,” the old oceanic crust is pulled down into the Earth’s interior along ocean trenches, eventually perhaps to be regurgitated along mid-ocean ridges. So the oceanic crust is being continually moved sideways: pushed from one end, and pulled from the other. The second important difference between continental and oceanic crust is composition. Understandably, given that continents have been around so much longer, continental crust is far more diverse in terms of its rock types than is oceanic crust, which—at least at its surface—is rarely anything other than a stack of basalt. Oceanic islands are bits of the oceanic crust that have somehow reached above the ocean surface. The obvious difficulty of this achievement, given the comparatively short time available (120 million years maximum) and the height involved (perhaps 4 km from ocean floor to ocean surface), explains why there are so few oceanic islands. This in turn explains why, for decades, while geologists explored the continents in minute detail, oceanic islands were marginalized, typically regarded as unremarkable adjuncts to continents or—even more pejoratively—as the detritus left in the wake of drifting continents. Not surprisingly, then, the earliest ideas about the evolution of the Earth’s surface all had a continental bias, and because they effectively ignored the other 73% of the

Earth’s surface (the ocean basins), they have since been proven largely wrong. Some of the earliest investigations of oceanic islands and, more generally, of the ocean floor can today be read as full of pointers to the critical importance of these features in understanding the formation of the Earth’s surface, but it was not until after World War II, when the U.S. Navy (among others) turned some of its resources and expertise to gathering scientific information about the ocean floor, that this breakthrough in perspective occurred. The results of these investigations, which led eventually to the formulation of the theory of plate tectonics in 1967, also underlined the importance of knowing about oceanic islands. OCEANIC ISLANDS AND PLATE TECTONICS

Plate tectonics envisages the Earth’s crust as divided into huge chunks (plates) that are generally rigid, interlocking, and continually moving. Plates include both continental and oceanic crust, but it is only the latter that moves independently. To understand the variety of ways in which oceanic islands form, it is helpful to classify their origin, as in Table 1. At the highest level, this separates oceanic islands formed along plate boundaries from those—far fewer—that form in the middle of plates. Divergent Plate–Boundary Islands

In the scheme of plate tectonics, a single plate may have a divergent plate boundary—typically marked by a midocean ridge—along which (sea-floor) spreading takes place. This is therefore marked by divergence or extension, and all the world’s ocean basins have one main divergent plate boundary, on either side of which ocean floor (and the islands that rise from it) increases in age with increasing distance from the ridge axis. Islands occur in places where (part of ) a mid-ocean ridge rises above the ocean surface, an unusual occurrence best exemplified by Iceland in the North Atlantic.

TABLE 1

Genetic Classification of Oceanic Islands Level 1 Classification

Level 2 Classification

Examples

Plate-boundary islands

Divergent plate boundary (mid-ocean ridge)

Iceland (North Atlantic) Niuafo‘ou (Tonga, South Pacific) Lesser Antilles group (Caribbean Sea) Solomon Islands (western Pacific) Sunda arc (Sumatra-Java, eastern Indian Ocean) Cikobia (Fiji, South Pacific) Hawaii group (northeastern Pacific) Tristan da Cunha–Walvis ridge (South Atlantic) Réunion-Laccadive (Indian Ocean)

Convergent plate boundary (island arc)

Intraplate (mid-plate) islands

690

OCEANIC ISLANDS

Transform plate boundary Linear island groups (hotspot island chains)

Here there is a plate triple junction where high heat flow has elevated the ocean floor, causing part of it to emerge above sea level. Much of what has been learned about divergent plate boundaries comes from Iceland, but there are smaller divergent plate-boundary islands. Among these are the island Niuafo‘ou (Tonga) in the Southwest Pacific, whose doughnut shape is a result of a stretched caldera that has become filled with water and forms Lake Vai Lahi. Convergent Plate–Boundary Islands

As well as a divergent boundary, a plate may also have a convergent boundary, one type of which involves an oceanic plate pushing down into the Earth’s interior beneath another oceanic plate. In terms of island formation, these are the most productive places in the ocean basins. This ocean–ocean convergent plate boundary is generally marked by an ocean trench with parallel lines of islands, sometimes along both sides. Ocean trenches are asymmetrical, with the more gently sloping side being that of the downgoing plate, and the steeper side—along which collapses often occur, generating tsunamis—being that of the overriding plate. Along convergent plate boundaries, islands can form and emerge in one of three locations: along the volcanic island arc on the overriding plate, along the non-volcanic island arc on the overriding plate, or along a crustal flexure on the downgoing plate.

FIGURE 1 Oceanic islands forming. (A) The summit of Kavachi volcano

in Solomon Islands lies 50–100 m below the ocean surface, but when it erupts, as here in October 2002, it forms a conspicuous sight. Kavachi occasionally forms islands, but these are short-lived, being eroded by waves when the eruption ends. (Photograph by Corey Howell). In 1453 the giant Kuwae volcano in Vanuatu blew itself to pieces in one of the largest eruptions by volume in the last 10,000 years. Today, an undersea caldera lies where Kuwae once stood, and occasionally a smaller volcano named Karua that has grown up from the caldera rim erupts. The eruption shown occurred on February 22, 1971, (B) and formed

VOLCANIC ARC ISLANDS

an island one day later (C). (Photographs by Don Mallick, used with permission of the Vanuatu Cultural Centre.)

At a convergent plate boundary, the downgoing plate is pulled down into the Earth’s interior, where—because of the intense heat—it melts. The liquid rock (magma) is lighter than the solid rock, so it tries to rise back to the Earth’s surface. Where it succeeds, it will erupt on the ocean floor and may eventually build a line of volcanoes (which may grow into volcanic islands) parallel to the associated ocean trench. The composition of the volcanic rocks in these islands can be linked to the type of material being pulled down into the trench, particularly whether or not this includes significant amounts of the sediments that accumulate in the bottoms of trenches. Because ocean trenches are usually arcuate in plan, the lines of associated volcanic islands are likewise arc-shaped; hence, they are referred to as volcanic island arcs. Examples come from the Caribbean (Lesser Antilles) and western Pacific (Marianas–Izu). In youthful volcanic island arcs, there are typically many underwater islands; some of these occasionally erupt just beneath the ocean surface,

making their presence manifest (Fig. 1A), and may form islands of unconsolidated pumice that are washed away when the eruption ends (Figs. 1B and 1C). Under many volcanic island arcs, particularly when they have grown comparatively large, much of the rising magma may not reach the surface of the crust, so it solidifies below it and forms intrusive igneous rocks. Recent work, particularly in the Canary Islands, has demonstrated that the importance of intrusive rocks to the growth of oceanic islands in such locations is far greater than once suspected. ISLANDS OF NON-VOLCANIC ARCS

It is clear that, along convergent plate boundaries in the ocean basins, not only is one plate being pulled down, but the other is being thrust upward over the top of it. This overriding plate is therefore being pushed not only sideways but also upward as it rides over the downgoing

OCEANIC ISLANDS

691

one, a process that is amplified when the surface of the downgoing plate is highly irregular. The uplift of the overriding plate often causes its edge to emerge above sea level, typically producing a line of islands, parallel to both the volcanic island arc and the adjacent ocean trench (Fig. 2A). Although these islands have volcanic basements, when they emerge, the basements are draped with thick piles of ocean-floor sediments. If they emerge within the coral seas, the emergent islands will commonly exhibit a thick cover of coral reef, testimony to their journey through the uppermost layers of the ocean. Examples include parts of several larger Caribbean islands (Hispaniola, Jamaica, Puerto Rico), the Mentawai Islands of Indonesia, and islands such as Choiseul in the northern Solomon Islands in the western Pacific. Such islands are generally not visibly volcanic, and therefore form a non-volcanic arc, distinct from its volcanic counterpart, which is farther away from the trench axis. Some of the most distinctive types of non-volcanic islands of this kind are those whose form is that of a staircase of broad limestone steps, each of which represents an emerged coral reef. The highest emerged coral reef is expected to be the oldest—the first to emerge above sea level—whereas the lowest is the most recent (Fig. 2B). ISLANDS ALONG THE CRUSTAL FLEXURE ON A DOWNGOING PLATE

Oceanic crustal plates are stiff, 15-km thick slabs of solid rock that naturally resist being forced upward or downward at convergent plate boundaries in the ocean basins. One clear manifestation of this resistance to be found on the downgoing plate is the way in which it rises upward slightly before being thrust down. This upward rise produces a flexure (or bulge) in the ocean floor up which submerged islands rise and sometimes poke their heads above the ocean surface when they get close to the flexural crest. Thereafter, it is all downhill, with many formerly emergent islands being pulled down into the bottoms of the ocean trenches where they are eventually dismembered and destroyed. Examples associated with the Tonga trench in the southwestern Pacific include the emergent island Niue, which is rising up the flexure, and the underwater Capricorn seamount, which is presently on its way down the trench slope. It is rare to have a line of islands form along the crustal flexure on a downgoing plate, because usually not very many seamounts (or guyots) are appropriately positioned, but some do occur. One of the best-studied examples is the Loyalty Islands of New Caledonia in the southwestern Pacific. Of these, the largest (Maré) is close to the flexural

692

OCEANIC ISLANDS

FIGURE 2 Islands close to convergent plate boundaries. (A) Map

of part of the South Pacific showing the form of the ocean floor in the area where the Pacific plate in the east is converging with the plate in the west along the Tonga–Kermadec Trench. The non-volcanic arc is represented by a line of high uplifted reef-limestone islands from ‘Eua in the south through Vava‘u to Niuatoputapu in the north. The volcanic arc runs parallel 30–50 km to the west. To the east of the Tonga-Kermadec Trench, Niue Island is rising up the crustal bulge, whereas Capricorn seamount is on its way down into the trench. Niuafo‘ou Island formed along a small divergent plate boundary, whereas the Samoa islands are a chain of hotspot islands. (B) View of the Talava Arches in northwest Niue Island. The flat top is formed by the emergence of a fringing reef since the last interglaciation about 120,000 years ago. (Photograph by the author.)

crest, whereas the two smaller ones (Ouvéa and Lifou) are on their way upward. Transform Plate–Boundary Islands

If a rectangular plate has one divergent boundary and one convergent boundary, then its movement (from divergent

to convergent) is fixed, which means that the other two boundaries can be neither of these. They are, in fact, places where one plate slides past an adjoining plate, theoretically with no net divergence or convergence: a type of boundary termed transform (or strike-slip). These are notorious as sites of large earthquakes—the San Andreas Fault is the best-studied—but they are not generally thought of as places where islands form. Islands form along transform plate boundaries only where there are slight irregularities (kinks) in these that lead to localized convergence. The Fiji island of Cikobia may be an example of just such an island, formed at a kink in the Fiji fracture zone, a transform plate boundary in the southwestern Pacific. Linear Groups of Intraplate Islands

Away from the edges of oceanic plates, in places where crustal quiescence rather than crustal activity is the norm, islands also form, although these are generally smaller and fewer and have less complex histories than their plateboundary counterparts. Most such islands occur in approximately straight lines, something much remarked upon in early accounts of oceanic islands. Later work showed something even more remarkable: namely, that the age and size of these islands generally increased uniformly from one end to another of the island chain. And at the younger, larger end, there always seemed to be an active volcano. The combination of these observations led to the formulation of the hotspot hypothesis, the idea that lines of intraplate islands were produced when an oceanic plate passed over a hotspot—a fixed place in the Earth’s crust thin enough for underlying magma to punch its way through to the surface. The movement of the oceanic plate led to the volcano over the hotspot being gradually pulled away from it, eventually becoming extinct and being replaced by another volcano. In time, this process gives rise to a line of volcanic islands whose age increases with greater distance from the hotspot and that will slowly subside and thereby become smaller (Fig. 3). Lines of hotspot islands are common in intraplate locations and include the Hawaii–Emperor island and seamount chain in the northern Pacific and the Samoa– Tuvalu island chain in the central Pacific. Réunion Island in the Indian Ocean and Tristan da Cunha Island in the South Atlantic are both volcanic islands built on top of active hotspots. VARIETIES OF OCEANIC ISLANDS

The classification of oceanic islands given in Table 1 tells us about island origins but not necessarily about

FIGURE 3 Ages of the most recent eruptions of island volcanoes

along the Hawaii–Emperor island seamount chain (dates in millions of years ago). The upper map shows the location of the Hawaiian Ridge within the northern Pacific, and the oldest part of this hotspot chain (Meiji Seamount: 74 million years old). The lower map shows the younger, largely emerged part of the island chain from Midway Atoll (27.7 million years old) to still-active Hawai’i Island. Note the presence of newly active Lo’ihi Seamount, which is growing directly above the hotspot while Hawai’i moves away from it.

the way these islands look now. This is far less easy to generalize about systematically because various processes have caused islands to emerge or submerge relative to the ocean surface (which itself changes), irrespective of their origin. Thus, current appearance is an unhelpful guide to the origin of an island. But it is something worth knowing about, not least to help explain the nature of island biotas and—something highly topical at present—the vulnerability of particular oceanic islands to erosion, even erasure, by sea-level rise. Every oceanic island began life as an ocean-floor volcano, but not all retain an immediately recognizable volcano form. When we look at the appearance of oceanic islands, three major groups can be identified: volcanic islands, high limestone islands, and atoll islands. There are other, more complex, types, often composites of volcanic rock and limestone, but their environments are generally reflective of the dominant rock types. Volcanic islands vary in appearance depending largely on their lithology (rock type composition) and their age. Basalt volcanoes that form over very productive hotspots,

OCEANIC ISLANDS

693

for example, tend to involve huge volumes of moderately viscous material piled up in shield volcanoes—by far the most common type on Earth—in a series of comparatively low-energy, effusive eruptions. In contrast, andesite volcanoes tend to be much steeper sided—a reflection of the high viscosity of the eruptive material—and to erupt comparatively explosively. In a similar fashion, youthful— maybe even active—volcanoes form islands that generally betray that fact, whereas older (long extinct) volcanic islands may have been thoroughly disguised by post-eruptive denudation and flank collapse. Limestone is a rock that forms only beneath the ocean surface, so a high limestone island has by definition emerged. For this reason, the form of limestone islands tends to broadly reflect the flat-topped form of undersea deposits. Sharp risers (slopes) to the next flat surface indicate successive periods of island emergence. Atoll islands (motu) are distinguished by their comparative lowness—they usually rise no more than 3 m above sea level—and their transient nature. Part of this is because they are largely composed of unconsolidated sediments that are comparatively easy for the sea to remove. Oceanic islands are more numerous in the warmer parts of the world’s oceans than in the cooler parts. This is because in the former exist coral reefs, which can build up above a sunken volcanic island. It is a moot point whether living coral reefs actually constitute an oceanic island because they cannot generally grow above low-tide level, yet on almost every reef, there are accumulations of debris, swept up from below sea level by waves, that indeed reach above it—and in some parts of the world form islands large enough to be habitable by humans. Such islands—commonly called motu—are accumulations of largely unconsolidated sand and gravel, typically cemented by beachrock or phosphate rock along their fringes. LIFE CYCLES OF OCEANIC ISLANDS

For anyone who is interested in explaining the distribution of islands in the world’s ocean basins, it makes no sense to confine a survey to those islands that are currently emergent (above sea level): There are many “islands” whose summits lie below the ocean surface. Some of these islands may have once been emergent but have since subsided (sunk) and/or been drowned as a result of sea-level rise. Traditionally these islands are classified as guyots, characterized by a flat top beveled by wave erosion as the island was slowly submerged. Conversely, there are many islands that rise from the ocean floor but have not yet pushed their heads above

694

OCEANIC ISLANDS

the ocean surface. Some may do so eventually, and some not. Irrespective of whether they attain the surface, they should be included in any survey of oceanic islands. Such islands are generally referred to as seamounts, characterized by a conical form, attesting to their volcanic origins. As noted above, in contrast to continental crust, oceanic islands are transient entities, never more than 120 million years in age and rarely emergent (above sea level) for even 50% of that time. Of course, there are exceptions, including oceanic islands that have become scraped off along continental margins and now lie far above the reach of the ocean, their insides exposed for all to see. Good examples are found in the accreted Wallowa terrane (Oregon) and others along the western side of the North American continental core. There are three main external influences on the life cycles of oceanic islands: first, island tectonics—the rises and falls of the island itself; then islands and sea-level changes; and finally, island landscape evolution. Oceanic-Island Tectonics

In any part of the world, the solid Earth’s crust can rise and fall, but in the ocean basins these processes are more widespread and are a major cause of oceanic island emergence and submergence. Long-term uplift and subsidence also affect islands as a result of changing water loads on the ocean crust. Particularly following land-ice melt during deglaciation, rapid inputs of water into the oceans can cause the ocean floor to deform. Yet the principal cause of individual island emergence and submergence over shorter time periods is vertical tectonics—movements of the Earth’s crust resulting from the accommodation of stresses associated with plate movements. The process of crustal rise is known as uplift, and many oceanic islands have been uplifted, especially near convergent plate boundaries. Uplift can be continuous, in which case it is usually slow; Maré Island in the Loyalty Islands of New Caledonia in the southwestern Pacific has been climbing the crustal flexure (described previously) at rates as high as 1.9 mm/year during the last half million years or so. Uplift can also be sporadic, however. This uplift type is typified by rapid bursts of uplift during largemagnitude earthquakes, typically causing ground level to rise 1–2 m, and is termed coseismic uplift. Yet between these infrequent bursts of coseismic uplift, there is often slow subsidence, so the net uplift over long time periods may be comparatively slow. A recent example comes from Ranongga Island in the Solomon Islands in the southwestern Pacific where, early in the morning on April 2, 2007, a large earthquake raised up the entire island 2 m

exposing the surface of its fringing reefs. A similar event happened during the December 26, 2004, earthquake in Indonesia (which caused the devastating tsunami) when Simeulue Island off the coast of Sumatra was raised 1.5 m in a few minutes. Subsidence can also be rapid and abrupt, perhaps coseismic, but more often it is an expression of the gravityinduced collapse of an island’s flanks. On Hawai‘i Island, during the Kalapana Earthquake on November 29, 1975, a 60-km stretch of the south coast sank 3.5 m and moved seaward some 8 m, causing a 10-m-high tsunami. Yet far more common among the global population of oceanic islands is slow, continuous, monotonic subsidence, typically the outcome of an island being carried on a moving plate into deeper water. Thus, islands that move away from the mid-ocean ridges or from hotspots usually subside as the underlying oceanic crust cools and comes to lie at increasing depths below the ocean surface. Some of these rates of subsidence are minute but continue for extremely long periods of time; the atoll island Enewetak (Marshall Islands, northwestern Pacific) has been sinking at an average rate of 0.03 mm/year for 45 million years.

FIGURE 4 The changing geography of the southwest Pacific. (A) About

20,000 years ago during the lowest sea level (−120 m) of the last glaciation, much more land was exposed in this region. Note particularly

Oceanic Islands and Sea-Level Changes

By comparing the maps of islands in the southwestern Pacific 18,000 years ago, when sea level was around 120 m lower than it is today, with maps from the present (Fig. 4), it is possible to get a sense of just how important sea-level changes are in causing islands to alternately emerge and submerge. Over the past 2–3 million years, sea level has oscillated between glacial (ice-age) low stands and interglacial high stands every 100,000 years or so. Today, in the middle of the Holocene interglacial period, which began around 12,000 years ago, we live in a drowned world; the ocean surface is higher than it has been for around 95% of the past 150,000 years. Thus, islands are far rarer today than they were during the last glaciation, something that has implications for various types of biota (including humans) that have dispersed across the oceans, as well as for islands themselves. An island that is submerged is immune from many of the processes of erosion that affect its subaerial counterparts. Conversely, it also ceases to be a viable habitat for terrestrial biota, meaning that it has to be recolonized if it emerges again. Moreover, the process of alternate submergence and emergence may affect the stability of an oceanic island through the successive application of pressure and then the abrupt release of that pressure, which can accelerate the large-scale collapse of steep island flanks.

the large island between New Caledonia and Australia, marked by the Bellona platform from which a few isolated reefs rise today. (B) The modern geography of the region for comparison.

Landscape Evolution on Oceanic Islands

Landscape evolution on oceanic islands is—as it is on continental landmasses—controlled largely by climate and lithology (rock type), and it is therefore difficult to generalize about. That said, it is clear that oceanic islands, largely because of their discrete nature (their boundedness) and their steep-sidedness—itself the outcome of their oceanic location—are susceptible to quite different processes of landscape evolution from those that operate on continents. In terms of their discrete nature, it is the fact that most oceanic islands are comparatively small and not part of larger land masses that makes their landscapes evolve in isolation. Thus, for example, oceanic islands in the trade wind belts may have well-defined wet and dry sides, where different sets of geomorphic processes operate. Many oceanic islands are, on account of their comparative smallness, entirely coastal, which means that ocean-driven processes affect the entire island; thus, it may change far more rapidly as a result. Next, there is the issue of the steep-sidedness of many oceanic islands, something that results from their having grown upward from the deep ocean floor. Like steep

OCEANIC ISLANDS

695

slopes anywhere, those that form the flanks of such oceanic islands are more prone to failure (collapse) than are gentler slopes, a process that is exacerbated for some islands by earthquake activity. The geological record is full of incidences of island flank collapse, ranging from the uncommon gigantic ones—such as the 5000-km3 Nu‘uanu Slide on Hawai‘i Island 2.1 million years ago— to more frequent, yet smaller, ones. Contained in the sedimentary (underwater) apron that surrounds the Marquesas Islands of the central eastern Pacific, there is many times more volcanic material than there is in the modern islands, suggesting that earlier islands collapsed and rebuilt themselves several times in the past. SEE ALSO THE FOLLOWING ARTICLES

Coral / Earthquakes / Island Arcs / Island Formation / Plate Tectonics / Sea-Level Change / Volcanic Islands

FURTHER READING

Menard, H. W. 1986. Islands. New York: Scientific American Books. Nunn, P. D. 1994. Oceanic islands. Oxford: Blackwell. Nunn, P. D. 2006. Island origins and environments, in A world of islands: a physical and human approach. G. Baldacchino, ed. Malta: Agenda, 5–37. Nunn, P. D. 2009. Vanished islands and hidden continents of the Pacific. Honolulu: University of Hawaii Press. Nunn, P. D., C. D. Ollier, G. S. Hope, P. Rodda, A. Omura, and W. R. Peltier. 2002. Late Quaternary sea-level and tectonic changes in northeast Fiji. Marine Geology 187: 299–311. Sigmundsson, F. 2006. Iceland geodynamics: crustal deformation and divergent plate tectonics. Berlin, Germany: Springer. Strahler, A. N. 1998. Plate tectonics. Cambridge, MA: Geo-Books.

ORCHIDS DAVID L. ROBERTS AND RICHARD M. BATEMAN Royal Botanic Gardens, Kew, United Kingdom

Assisted by their dustlike seeds, orchids are among the first plant families to colonize islands, often speciating into the many unexploited niches on newly formed or newly disturbed islands. Reduced (or at least temporarily reduced) competition on some islands may allow more radical evolutionary shifts, as well as the establishment of new relationships between an orchid lineage and its necessary partners—animals for pollination and mycorrhizal fungi for germination and nutrition. Furthermore, the tendency of orchids to be pollinator-limited, and thus to occur as

696

ORCHIDS

small populations, has resulted in orchids frequently evolving through founder effect and genetic drift. SEED DISPERSAL

Orchids are well known for producing vast quantities of seeds, in some cases several million. This did not go unnoticed by Charles Darwin, who painstakingly recorded around 6200 seeds from a single capsule of Orchis (now Dactylorhiza) maculata. In his classic book On the Various Contrivances by which British and Foreign Orchids are Fertilised by Insects, and on the Good Effects of Intercrossing (1877) he stated whimsically that To give an idea what the above figures really mean, I will briefly show the possible rate of increase of O. maculata: an acre of land would hold 174,240 plants, each having a space of six inches square, and this would be just sufficient for their growth; so that, making the fair allowance of 400 bad seeds in each capsule, an acre would be thickly clothed by the progeny of a single plant. At the same rate of increase, the grandchildren would cover a space slightly exceeding the island of Anglesea; and the great grand-children of a single plant would nearly (in the ratio of 47 to 50) clothe with one uniform green carpet the entire surface of the land throughout the globe. But the number of seeds produced by one of our common British orchids is as nothing compared to that of some of the exotic kinds.

The exceptional dispersibility of dustlike orchid seeds in air currents over considerable distances has allowed successful colonization of islands hundreds or thousands of kilometers from the nearest seed source. Also, unusually for flowering plants, orchids have pollen that on average travels a shorter distance than the seed. However, most orchid seeds fall close to the mother plant. Orchids, particularly tropical species, are often pollinator-limited, resulting in low levels of fruiting success. Hence, tropical orchids have on average considerably lower fruiting success than do temperate species. In addition, many temperate species have several characteristics that maximize reproductive success under conditions of infrequent pollination. COLONIZATION OF ISLANDS

Seed dispersal is only the first step in the successful colonization of an island. The seed must fortuitously land in a suitable place to germinate, on a surface that provides at least one compatible mycorrhizal fungal associate. An immigrant orchid seed may face not one fungally mediated barrier to colonization but two. There is an increasing body of evidence suggesting that, in many orchids, a member of one group of mycorrhizal fungi is necessary for successful

germination of the microscopic seeds, whereas a member of a second group of fungi is needed to supply nutrition to the mature plant. There is now growing evidence that availability of suitable mycorrhizal partners is a key controlling factor of orchid populations. Recent research in Australia has suggested that rare orchids are associated with rare mycorrhizal partners. Thus, the main factor determining the success or failure of an immigrant orchid seed may be the happenstance presence, within the substrate on which it lands, of appropriate members of both cohorts of fungi. The resulting seedling must then survive long enough to flower, and if that were not a sufficient challenge, it also needs to form a relationship with a pollinator that is sufficiently competent to remove its pollinia but not so competent that it refuses to visit another compatible flower—one that has opened in the vicinity of the first and at the same moment in time. Fortunately, the vastly improbable becomes probable when extended over a geological time scale. BAKER’S LAW AND REPRODUCTIVE ASSURANCE

It is often stated, with some justification, that orchids are among the most specialized of all flowering plant families. Not surprisingly, therefore, they tend to evolve specialized pollination systems, with 60% of species supposedly being pollinated by a single species of animal. Whereas many flowering plants rely on various chemical inhibition systems to prevent self-pollination and promote outcrossing, most orchids are self-compatible. Rather than chemical compatibility barriers, they have evolved floral mechanisms that promote outcrossing through attracting, and influencing the behavior of, appropriate pollinators. This confers on orchids an advantage in colonizing new territory, as there is no longer an absolute need for a second compatible plant in order to generate viable seeds. Even pollinators, though undoubtedly helpful, may not be essential. Instead, the plant may be able to resort to vegetative reproduction or apomixis (production of viable seed without male intervention), although the latter process is rare among orchids. These advantages of self-compatibility for island colonization have become known as Baker’s Law. As a result, we find that orchid species on tropical oceanic islands exhibit either of two contrasting strategies to achieve successful reproduction. Many species emulate those of tropical rain forests in producing low frequencies of viable fruits. In contrast, the second group shows much higher levels of fruiting success through efficient selfpollination. For example, this approach to what is known as reproductive assurance is evident in the endemic species of Jumellea from the island of La Réunion in the Mascarene

Island archipelago, which are pollinated by hawkmoths. In order to succeed, some orchids have reduced the size of their floral spurs to better fit the proboscides of the moths, whereas others have become self-pollinating. THE CONSEQUENCES OF REPRODUCTIVE ASSURANCE FOR ISLAND ORCHIDS

Ultimately, reproductive assurance means that the orchid successfully passes on its genes to the next generation. However, reproductive assurance also means that characters such as floral display, scent production, and method of nectar presentation are no longer selected for by the relevant pollinator(s). What is the point in having complex, energy-consuming features that are supposed to elicit pollination if no pollinator is present to appreciate them? Thus, although the genus Angraecum is well known for producing a strong scent at dusk to attract hawkmoth pollinators, A. borbonicum (Fig. 1), endemic to La Réunion, has lost its scent and considerably reduced its nectar production. In its putative mainland ancestor, the ability to attract an effective pollinator must outweigh the considerable energetic cost of producing the scent. But if the species becomes self-pollinating, the pollinator is eliminated from the equation, and scent production thus becomes irrelevant to the long-term well-being of the orchid. Once selection has been relaxed, there is no longer pressure to purge from the population the mutations that prevent scent production. Furthermore, the energy the orchid saves by not producing a scent can usefully be reallocated toward higher-priority objectives such as producing healthy fruit. There are, of course, good reasons why the majority of orchids cross-pollinate. Notably, self-pollination allows

FIGURE 1 Angraecum borbonicum from La Réunion, Mascarenes.

ORCHIDS

697

FIGURE 2 Two species of Himantoglossum showing speciation via direc-

tional change in a lineage through anagenesis: (A) H. metlesicsianum from Tenerife; (B) H. robertianum from the mainland, Macaronesia.

much easier perpetuation of mutationally induced morphological novelties, most of which are likely to prove competitively inferior to their parents. However, ecologically mediated competition is often lower on islands, increasing the probability that such “hopeful monsters” will be perpetuated to form new evolutionary lineages. Potential examples of such monsters from La Réunion include the endemic genus Bonniera, which probably evolved through saltational evolution from species of Angraecum when the labellum (lip) was replaced by a less differentiated, petallike organ. This developmental shift has also resulted in the loss of the spur and therefore of nectar production. This transition occurred independently twice: A. conchoglossum, also found on La Réunion, gave rise to B. corrugata through cladogenesis, whereas A. arachnites from Madagascar gave rise to B. appendiculata through anagenesis. Another example of anagenesis is Himantoglossum metlesicsianum (Fig. 2A), a rare endemic of Tenerife that has a floral morphology distinctly divergent from that of its widespread mainland sister, H. robertianum (Fig. 2B). The two species also show substantial genetic differences, suggesting that the ancestor(s) of H. metlesicsianum migrated to Tenerife 1–2 million years ago. Of 13 orchid species currently occurring on the Macronesian islands, one may not merit species-level recognition, and all of the remaining 12 species could have arisen by anagenesis, demonstrating that not all orchids indulge in spectacular evolutionary radiations after colonizing an island. ALTITUDINAL ASPECTS OF ISLAND HOPPING

It is widely recognized that species richness commonly peaks at intermediate elevations. If the range of a species

698

ORCHIDS

centers on the mid-elevation level, then the species has a greater potential to occupy more of the island, as in theory it can expand to the top and bottom of the island. However, if the mid-point of a species’s range occurs at the bottom or the top of the altitudinal range offered by the island, then that species can spread within the island in only one direction—up or down, respectively. Widespread species therefore have midpoints near the center of the elevational range, whereas localized species with a narrow altitudinal range are equally likely to be found anywhere across the elevational gradient. It follows that the greater is the proportion of widespread species in a particular orchid flora, the greater is the likelihood that species richness will peak toward the middle of the altitudinal range of the island. This phenomenon is termed the “mid-domain effect.” Let us once again consider the Mascarene Islands (Mauritius, La Réunion, and Rodrigues), which lie about 900 km east of Madagascar. Morphological and DNAbased studies suggest that the lineages that have diversified to give the Mascarene Islands their present orchid flora of about 150 species largely originated from among the 960 species of orchids that currently occupy Madagascar. The similarity in altitude of Madagascar (2876 m) and La Réunion (3069 m) is reflected in the behavior of their respective orchid flora components of La Réunion (i.e., those species that are shared between Madagascar and La Réunion, and those species endemic to La Réunion), whereas there is a strong contrast between the distributions of comparable orchids on La Réunion and nearby Mauritius, which rises to only 828 m. These observations suggest that a species has difficulty escaping from its original altitudinal range, even when it colonizes an island over three times the height of the source terrain on which it evolved (Fig. 3). In contrast, no such pattern is evident when we compare the orchid floras of the Gulf of Guinea Islands (Annobon, Sao Tomé, Príncipe, and Bioko), 60–300 km off the West African coast. Like the Mascarene Islands, the Gulf of Guinea archipelago evolved from a volcanic hotspot. However, the much closer proximity of the Gulf of Guinea islands to the African mainland has permitted multiple colonization events by orchids. The effect of the proximity to major land masses can be seen in the percentage of endemic orchid species, which is 11% for Príncipe and 16% for Bioko. In contrast, the 68% endemism observed in the isolated Mascarene Islands suggests that there is little movement of orchids between the islands and the mainland but considerable movement among the islands.

SPECIATION ON ISLANDS

The most important factors determining the species richness of an island are its age, size, and distance from potential sources of immigrants. Altitudinal extent and diversity of geological substrates also strongly affect the ability of the island to differentiate habitats and thereby to multiply niches that could potentially be occupied by orchids. It has also been suggested that species diversity might itself help to drive speciation. Increasing the number of species present increases the number of likely ecological interactions, which in turn increases the number of potential niches that can then be filled by novel species—a classic positive feedback loop. Such “niche-filling” and subsequent speciation has been documented on La Réunion, where the pollination of two of the endemic species of Angraecum, A. bracteosum, and A. striatum (from the endemic Section Hadangis), has been captured on film. Pollen transfer was effected not by the expected hawkmoths, but rather by birds—the endemic white-eyes, Zosterop olivaceus and Z. borbonicus, respectively. These orchids, and a further related species endemic to the Mascarene Islands, may have evolved to fill an available niche as a result of encountering the depauperate hawkmoth fauna of the island. Why then are the orchid floras of islands not even more diverse? The factors that discourage a veritable explosion of species are (1) the failure to find an alternative coevolutionary partner on the island and (2) extinction. The latter can reflect the absence or precariousness of a particular habitat. For example, Bulbophyllum variegatum has been extirpated from Mauritius and is declining on Réunion, where it grows epiphytically on only one species of tree, Agarista salicifolia—a tree that has a regrettable predilection for an unpredictable substrate, specifically as a primary colonizer of recently extruded lava flows. However, extinction is typically caused by competition among species for finite resources such as nutrients or the attention of pollinators. In this context, Baker’s Law has important implications for conservation, because it suggests that increased pollinator specialization and/or selfincompatibility will predispose a species to extinction. Self-pollinating species may therefore be better suited to survive in the current changing climate by means of reproductive assurance and the ability of some species to maintain a dual reproductive strategy, indulging in outcrossing when conditions permit but using selfpollination as a failsafe.

FIGURE 3 Altitudinal distribution of the orchid flora of La Réunion,

Mascarenes.

SPECIATION BEYOND CONVENTIONAL ISLANDS

One of the main biological effects of island colonization is the founder effect and subsequent genetic drift. This scenario has been proposed not just for orchids on geological islands but also for the evolution of the family as a whole. Orchids are well-known for being reproductively pollinator-limited and can be even further restricted by resource constraints. Consequently, only a small proportion of the population often gives rise to the next generation. This situation is further skewed by the fact that most orchid populations, particularly in the tropics, are small, occurring in an unusually fluid ecosystem rich in underexploited niches. This leads to frequent changes in appearance or behavior among generations and subsequent diversification and speciation. In other words, low reproductive success leads to a low proportion of reproducing individuals (technically termed a small effective population size, Ne ). This often results in genetic drift being the initial cause of evolution, with Darwinian adaptation, driven by directional or disruptive selection, later imposing itself to better fit the new generation to the local pollinator population. SEE ALSO THE FOLLOWING ARTICLES

Anagenesis / Dispersal / Founder Effects / Mascarene Islands, Biology / São Tomé, Príncipe, and Annobon FURTHER READING

Arditti, J., and A. K. A. Ghami. 2000. Numerical and physical properties of orchid seeds and their biological implications. Tansley Review No. 110. New Phytologist 145: 367–421. Bateman, R. M., and W. A. DiMichele. 2002. Generating and filtering major phenotypic novelties: neoGoldschmidtian saltation revisited, in Developmental genetics and plant evolution. Q. C. B. Cronk, R. M. Bateman, and J. A. Hawkins, eds. London: Taylor & Francis, 109–159.

ORCHIDS

699

Bateman, R. M., P. M. Hollingsworth, D. Devey, and D. L. Roberts. 2005–2006. When orchids challenge an island race 1–4. Orchid Review 113: 334–337; 114: 36–41; 114: 98–102; 115: 212–217. Bateman, R. M., and P. J. Rudall. 2006. The good, the bad, and the ugly: using naturally occurring terata to distinguish the possible from the impossible in orchid floral evolution. Aliso 22: 481–496. Colwell, R. K., and D. C. Lees. 2000. The mid-domain effect: geometric constraints on the geography of species richness. Trends in Ecology and Evolution 15: 70–76. Dixon, K. W., S. P. Kell, R. L. Barrett, and P. J. Cribb. 2003. Orchid Conservation. Kota Kinabalu, Sabah: Natural History Publications. Dressler, R. L. 1990. The Orchids: natural history and classification. Cambridge, MA: Harvard University Press. Pridgeon, A. M., P. J. Cribb, M. W. Chase, and F. N. Rasmussen. 2001– 2006. Genera Orchidacearum, vols. 1–4. Cambridge: Cambridge University Press. Tremblay, R., J. D. Ackerman, J. K. Zimmerman, and R. N. Calvo. 2005. Variation in sexual reproduction in orchids and its evolutionary consequences: a spasmodic journey to diversification. Biological Journal of the Linnean Society 84: 1–54.

ORGANIC FALLS ON THE OCEAN FLOOR CRAIG R. SMITH University of Hawaii, Manoa

Organic falls are large parcels of organic matter (e.g., dead fish, marine mammal carcasses, wood and other vascular plant debris, masses of macroalgae) that sink largely intact to the sea floor. Because most of the sea floor underlies deep water and is food-limited (fed by a diffuse rain of organic material from surface waters), large organic falls on the deep-sea floor create food-rich islands in an energy-poor desert. Suites of deep-sea species rapidly consume a broad range of organic-fall types, causing these food-rich islands to be relatively ephemeral (i.e., lasting for days to decades).

WHALE FALLS

Sunken whale carcasses (or “whale falls”), ranging in mass from 103 to 105 kg, are the end-members in size and lability of large organic falls. A single 5 × 104-kg whale fall yields a massive pulse of labile proteins and lipids to the sea floor; in one moment the ocean floor underlying the whale carcass receives the equivalent of hundreds to thousands of years of background carbon flux. The assemblage of organisms exploiting a deep-sea whale fall can pass through at least three stages of community succession: 1. A mobile scavenger stage, during which aggregations of voracious, highly active scavengers (e.g., sleeper sharks, hagfish, rattail fish, and lysianassid amphipods) consume the whale’s soft tissue over time scales of months to years 2. An enrichment opportunist stage, lasting months to years, during which organically enriched sediments and exposed bones are colonized by dense assemblages of opportunistic worms and crustaceans 3. A sulfophilic (or “sulfur-loving”) stage which can last for decades, during which a large, speciesrich assemblage lives on the skeleton as it emits sulfide from anaerobic breakdown of bone lipids; the sulfide effluxing from the bones supports a chemoautotrophic assemblage of animals deriving nutrition from sulfur-oxidizing bacteria (Fig. 1).

NATURE OF ORGANIC FALLS

The rates and patterns of exploitation of large organic falls at the deep-sea floor depend on a number of factors. These include (1) parcel size (ranging from 101–105 kg), (2) lability or digestibility of the organic material (ranging from easily digested muscle protein in fish to recalcitrant cellulose and lignin in wood), and (3) the physical structure of the parcel (for example, vertebrate soft tissues, calcified whale bone impregnated with whale oil, or heavily bored wood). All three factors influence the succession of organisms utilizing a large organic fall and the persistence time of such food-rich islands at the sea floor.

700

O R G A N I C FA L L S O N T H E O C E A N F L O O R

FIGURE 1 Whale skeleton on the sea floor at 1670-m depth off the

California coast with white and red mats of sulfur-oxidizing bacteria, small white anemones, and an asteroid. This whale carcass has been on the sea floor for about 4.5 years. Photograph by Craig R. Smith.

Whale falls can harbor hundreds of species and include at least 33 species of “whale fall specialists” (i.e., animals that apparently require whale falls to maintain their populations). Whale-fall communities also contain a number

of chemoautotrophically dependent generalists found at hydrothermal vents and cold seeps. WOOD FALLS

Sunken wood and other vascular plant debris (e.g., coconut husks, palm fronds) also support specialized deep-sea communities (Fig. 2), with a total of over 200 animal species known to be associated with vascular plant material recovered from the deep-sea floor. Keystone species on vascular plant debris include “wood-eating” bivalves in the genus Xylophaga (meaning “wood eater”) that bore into sunken wood and consume it from within, much as termites reduce wood in terrestrial habitats. These woodeating clams contain endosymbiotic bacteria in their gills, which produce enzymes to break down the otherwise recalcitrant cellulose in the wood. Xylaphaga can colonize isolated wood parcels at the deep-sea floor within weeks, consuming most of the mass of 1-kg wood parcels within a year, but requiring many years to reduce large tree trunks or shipwrecks. These wood-eating clams are considered keystone species because their fecal material and biomass can support a substantial local food web of detritivores (including polychaete worms and crustaceans) and predators (e.g., nemertean worms and galatheid crabs). The tunnels formed by the boring clams, as well as other cracks and crevices in the vascular plant debris, also provide habitat for a diversity of benthic invertebrates seeking physical shelter. Still other faunal components on wood-fall islands include surface grazers such as limpets, which consume bacteria and fungi growing on the wood surface.

MACROALGAL FALLS

Sunken masses of giant kelp and other macroalgae (typically 1–103 kg in mass) provide yet another type of large organic fall. In the deep sea, kelp falls often are rapidly colonized by shrimp, galatheid crabs, limpets, and amphipods that (1) graze on the kelp and associated microbes, (2) prey on the biomass of the attracted kelp fauna, and (3) use the kelp (especially the rootlike holdfast) as an attachment substrate and as shelter. Very large kelp falls can undergo anaerobic decomposition and produce sulfide which can then foster a chemoautotrophic assemblage of bacteria and clams. Very large masses of dead kelp and seagrass accumulating in submarine canyons are known to support an extraordinary abundance of benthic invertebrates (e.g., more than 106 amphipods and other crustaceans per m2) and sustain some of the highest rates of secondary production ever measured at the sea floor. Faunal diversity on macroalgal falls can also be high, with more than 50 species exploiting a 100-kg kelp parcel at any given time. CONCLUDING REMARKS

The speed at which animals find and exploit organic falls is remarkable, especially considering the spatial rarity of food falls in the deep sea. Even in the richest deep-sea regions, research submersibles may travel many kilometers across the sea floor without encountering a dead fish, parcel of wood, or kelp fall. Nonetheless, nearly all large organic falls form oases of biomass and biodiversity, sustaining species rarely if ever encountered in the food-poor background deep sea. It is clear that these ephemeral food-rich islands have supported adaptive radiation in the deep sea, fostering specialized fauna that contribute significantly to biodiversity and evolutionary novelty in the ocean. SEE ALSO THE FOLLOWING ARTICLES

Cold Seeps / Hydrothermal Vents / Succession / Whale Falls

FURTHER READING

FIGURE 2 Wood parcel at 1670-m depth off California. The wood is

covered with the whitish siphons of the wood-eating bivalve Xylophaga protruding from holes in the wood, and by the muddy tubes of polychaete worms. A rockfish, galatheid crabs, scale worms, and a brittle star are also visible living around the wood parcel. Photograph by Craig R. Smith.

Distel, D. L., and S. J. Roberts. 1997. Bacterial endosymbionts in the gills of the deep-sea wood-boring bivalve Xylophaga atlantica and Xylophaga washingtona. Biological Bulletin 192: 253–261. Smith, C. R. 1985. Food for the deep sea: utilization, dispersal and flux of nekton falls at the Santa Catalina Basin floor. Deep-Sea Research I 32: 417–442. Smith, C. R., and A. R. Baco. 2003. Ecology of whale falls at the deep-sea floor. Oceanography and Marine Biology Annual Review 41: 311–354. Vetter, E. W. 1994. Hotspots of benthic production. Nature 372: 47. Wolff, T. 1979. Macrofaunal utilization of plant remains in the deep sea. Sarsia 64: 117–136.

O R G A N I C FA L L S O N T H E O C E A N F L O O R

701

P PACIFIC REGION ANTHONY A. P. KOPPERS Oregon State University, Corvallis

The Pacific Region is extraordinary in many aspects. It is the largest ocean on Earth, harbors the deepest trenches, has the highest abundance of islands, and underneath its sea surface it encompasses the largest tectonic plate on Earth. Its enormity is emphasized by the fact that Christmas Island in the center of the Pacific Ocean lies more than 8,000 km away from any continent. As a whole the islands in the Pacific Region are referred to as Oceania, the tenth continent on Earth. Inherent to their remoteness and because of the wide variety of island types, the Pacific Islands have developed unique social, biological, and geological characteristics. ISLAND FORMATION IN THE PACIFIC

Little was known about the formation of islands in the Pacific until the voyages of Captain Cook in 1768–1780, and until the HMS Beagle and HMS Challenger visited the region in 1835 and 1875, almost a century later. Before these famous expeditions, knowledge about the island geography in the Pacific Region was limited to the first maps of the “Quiet Ocean” or Maris Pacifici by Abraham Ortelius in 1589, based on the sparse data collected during the Magellan expedition, the very first to circumnavigate the world (Fig. 1). Today, the formation of most of the 25,000 islands in the Pacific can be explained by a singular geological phenomenon, described by the theory of plate tectonics. The first ideas for this theory date back to the early

702

work of Alfred Wegener (1915) on continental drift, yet the fact that tectonic plates move over the Earth’s surface with rates up to 100 mm/year did not get wide recognition until the late 1960s. Now it is well understood how the sometimes violent interactions at the boundaries of these moving plates generate most of the earthquakes and volcanism on Earth. In the Pacific Region the majority of these plate interactions are embodied by the so-called Ring of Fire, which is a 40,000-km-long stretch of subduction zones where the Pacific Plate is actively being destroyed (together with two smaller plates along its eastern edge) while causing many earthquakes and producing many volcanoes and volcanic islands (Fig. 2). Whereas the subduction of an oceanic plate (such as the Pacific Plate) underneath a continent does not often generate islands (Fig. 2A), subduction underneath another oceanic plate normally results in a remarkable semicircle of volcanic islands, named an island arc (Fig. 2B). The Aleutian Islands located to the north of the similarly named Aleutian Trench, and all volcanic islands located to the west of the Kurile, Japan, Izu-Bonin, Marianas, Bougainville, Tonga, and Kermadec trenches (Fig. 2) are typical examples of island arcs. New islandbuilding volcanism in these arcs has taken place as recently as August 9–11, 2006, near Home Reef in the Tonga Islands, resulting in a new volcanic island and massive pumice rafts visible by satellite for hundreds of miles in the Pacific Ocean. A few years earlier, in May 2000, a new phase of island building was witnessed for Kavachi, a submarine volcano in the Solomon Islands, which appeared above the sea surface for the first time in 1951 and which has been destroyed by wave erosion and rebuilt by volcanic eruptions several times since then. Myojinsho Island breached the sea surface for the first time in 1952–1953 along the Izu-Bonin island arc,

FIGURE 1 Map No. 12, called Maris Pacifici (of the Pacific Ocean) by Abraham Ortelius in 1589. This first printed map of the Pacific Region included

“a very new description of the peaceful sea, commonly called the South Sea with the regions lying around it, and its islands, scattered everywhere” and was created following the first expedition by Magellan. Despite the small number of prior explorations of the Pacific Region at that time, this map contains references to New Guinea (Nova Guinea), the Japan Islands (Iapan Ins.), the Galapagos Islands (Y. de Galopagos), the Hawaiian Volcanic Islands (Los Bolcanes and La Farfana), and the Solomon Islands (Insulae Salomonis).

420 km south of Tokyo. During 12 months of continuous eruption, more than 1000 phreatomagmatic explosions were recorded at this newly born volcanic island. These examples emphasize the continuous and sometimes explosive volcanic activity that is characteristic of island arcs and the cause of unremitting production of new islands in the Pacific. Only in three places do the boundaries comprising the Ring of Fire have a different character, and those are along the San Andreas fault in the western United States and Baja California, the Islands of New Zealand, and the Solomon Islands (Fig. 2). It follows that the converging motions of the tectonic plates do not everywhere result in the formation of a typical subduction zone. In some cases, convergence occurs under significant angles, causing lateral motions that form so-called transform faults or spreading centers in order to accommo-

date the non-orthogonal displacements. For example, horizontal shear motions between two plates have created the San Andreas transform fault and explain the high frequency of earthquakes observed along this active fault line. On the other hand, diverging motions caused Baja California to rift apart after a spreading center (partly) followed the Farallon Plate in its subduction underneath North America. In the case of New Zealand, the direction of subduction is reversing along an axis running through the North and South Islands, a reversal that can be accommodated only by the formation of a transform fault in between. The Solomon Islands are unusual because the Pacific plate has a hard time subducting beneath the Australian plate as a result of the 122-million-year-old and 30-km-thick Ontong Java Plateau (the biggest large igneous province on Earth) riding on top of it. This rare obstacle has been obstructing the

PAC I F I C R E G I O N

703

FIGURE 2 Ring of Fire and Island Arcs of the Pacific Region. Insets: (A)

Ocean–continent subduction zone. (B) Ocean–ocean subduction zone. Black and white base map provided by Jasper Konter.

subduction of the Pacific plate for at least the last 6 million years and resulted in the upthrusting. or obduction, of this abnormally thick oceanic crust as the Solomon Islands were formed. One of the most extraordinary aspects of the Pacific Region is that it is littered with volcanic islands that formed far away from tectonic plate boundaries and island arcs, in a so-called intraplate setting (Fig. 3). Some of these volcanoes are currently active, such as K1¯lauea on the Big Island of Hawaii. Others experienced historical eruptions, such as Savai’i Island in the Samoan Archipelago and Macdonald Seamount in French Polynesia. Their existence in the middle of the wide Pacific Ocean instantly captured the imagination of Charles R. Darwin and James D. Dana in the nineteenth century. But despite Darwin’s and Dana’s detailed explorations around and on these active islands, they could not explain why this kind of volcanism and island formation occurred in these remote locations. Neither could twentieth-century scientists when they attempted to apply the plate tectonic theory to the problem, despite the large amounts of new oceanographic data collected in the decades following World War II. In the early 1970s, and immediately following the widespread acceptance of plate tectonics, J. Tuzo Wilson and W. Jason Morgan proposed the hotspot and mantle plume models (Fig. 3A). In these models, thermally

704

PAC I F I C R E G I O N

“hot” anomalies are presumed present deep in the Earth’s mantle, causing narrow upwellings of mantle material that eventually reach the lithosphere. Decompression in these mantle plumes as they rise to depths just below the lithosphere causes melting, and the magmas that are produced penetrate the lithosphere to form intraplate volcanoes. Because tectonic plates move vast distances over geological time, each volcanic center is moved away from the hotspot and becomes inactive. At its place a new hotspot volcano or volcanic island gets produced, in a process that has generated many linear chains of volcanoes in the Pacific Region, with the Hawaii–Emperor chain being the prime example. However, today most of these extinct hotspot volcanoes exist below the sea surface as seamounts (Fig. 4). Over geological time the volcanic islands eroded, subsided, and sank several kilometers deep, because the Pacific Plate on which they ride cooled down and became more dense with age (Fig. 3A). This process typically has produced drowned flat-topped seamounts that we know as guyots. More than 50,000 of these seamounts and guyots are estimated to exist in the Pacific Region alone. They tell us of a complex history and volcanic evolution, involving many more ancient islands than we can observe above the sea surface today. ISLAND ARCS

Most islands of the Pacific Region formed along the Pacific Rim as part of the Ring of Fire (Fig. 2). Subduction of the Pacific oceanic plate underneath the surrounding North American, Philippine, and Australian plates caused the formation of numerous island arcs, most of them currently active and not older than a few million or tens of million years. Because of their location on the fringes of the Pacific Ocean, these islands have a strong continental connection in a sociohistorical sense, in their biological evolution, and in their climate. These island arcs functioned as the springboard for the early Polynesian people in their explorations of the largest ocean on Earth in search of new island habitats. They also created a buffer zone during World War II, where many combat campaigns and bombardments took place outboard of the continents and on the Solomon Islands (Tulagi, Guadalcanal, Bougainville), the Marianas and Volcano Islands (Saipan, Truk, Tinian, Guam, Iwo Jima), and the Aleutian Islands (Attu, Adak). Their proximity close to the continental land masses makes arc islands less isolated from the import of new species, allowing for more biological diversity (and less endemism) despite the rather young age of many of these islands. And, finally, as

FIGURE 3 Volcanic Islands and Atolls of the Pacific Region, most of

FIGURE 4 Seamounts of the Pacific Region. Only a selection of sea-

which are concentrated in the equatorial Pacific. (Inset A) Hotspot trail

mount trails are indicated by the light blue lines, whereas red dots indi-

formation by mantle plumes, showing the age-progressive nature of

cate the entire distribution of seamounts, estimated to total ∼50,000 in

the volcanoes formed over long periods of plate motion away from

the Pacific Ocean alone. Note that WPSP is short for the West Pacific

the hotspot and the subsequent “drowning” of each volcano as the

Seamount Province, a region mostly including Cretaceous seamounts.

oceanic crust cools down over geological time and sinks in the more

Black and white base map and seamount locations provided by Jasper

fluid or plastically behaving asthenosphere. Black and white base map

Konter.

provided by Jasper Konter.

another consequence of their location nearby the continents, many Pacific island arcs are first in line to receive a full beating of incoming typhoons or tsunamis.

pation of the Americas. At that time the Aleutian Islands were covered by extensive ice caps and glaciers. Kurile Islands

Aleutian Islands

This island arc in the northeast corner of the Pacific Ocean contains more than 300 volcanic islands and spans almost 2000 km between the peninsulas of Alaska and Kamchatka. Toward the west the Aleutian arc sharply bends southwest into the Kurile island arc segment. Fifty-seven historically active volcanoes are present on the islands, of which two (Bogoslof and Fire Island) surfaced less than two centuries ago. Its location above 52° N latitude makes this island arc most extreme in terms of its climate, compared to all other Pacific islands. With short growing seasons, generally low temperatures, and strong prevalent winds, trees remain remarkably short and only a small variety of vegetation and animal species are present. Most of the islands also may have been connected during the sea level lowstand of the last glacial period, effectively creating a land bridge between Eurasia and North America that provided a likely immigration route for the earliest occu-

The Kurile Islands form a 1300-km-long island arc that connects Japan with the Aleutian Islands in the north. This island arc contains 56 islands with about 100 volcanoes, of which 40 are active. It lies in one of the world’s most remote places and is characterized, like the Aleutians, by harsh climate. However, its northeast-southwest orientation provides a strong latitudinal difference that results in rather dissimilar subtropical weather conditions and wildlife toward the southern end of the island arc. Nonetheless, prevalent rains, snowstorms, and ice fields keep these islands in a strong weather grip most of the year. The earliest settlements date back to 7000 BC, but today no more than 17,000 inhabit the Kurile Islands. Japan Islands

Recent volcanism in the Japan Islands has been caused by the dual subduction of the Philippine Sea Plate in the south and the Pacific Plate in the northeast. The

PAC I F I C R E G I O N

705

entire island arc system has been active since the Permian (260–299 million years ago), at which time the closing of the Tethys Seaway initiated the westward subduction of the (now completely destroyed) Farallon Plate underneath the Asian continental margin. Overall, more than 3000 islands make up the Japanese Islands. Mount Fuji is Japan’s highest peak and the most famous of its more than 200 active volcanoes. Japan’s climate ranges from cold snowy peaks on the northeastern islands to subtropical conditions on its southern islands. The Japan Islands also are the most heavily populated in the Pacific Region, with more than 130 million inhabitants. This dense human population may be the cause of the (near) extinctions of four mammal species (two bat species, the Japanese wolf, and the Japanese otter), 15 bird and fish species, and more than 40 plants. Fighting in the Pacific Region during World War II ended after two nuclear weapons were detonated above the cities of Hiroshima and Nagasaki on August 6 and 9, 1945. More than 220,000 civilians perished from these bombings, the only nuclear attacks in human history. Izu-Bonin and the Marianas

Both the Izu-Bonin and Mariana island groups form a combined island arc that stretches from Guam at 12° N to the beginning of the Japan Trench at 21° N. The current island arc is the most recent product of the ongoing subduction of the Pacific Plate to the west, which began around Eocene times (34–56 million years ago). Six new volcanic islands and a similar number of seamounts formed during the past century. Behind the strongly curved Marianas island arc, an inner arc runs parallel to the outer arc and is characterized by active volcanism as well. In recent years at least 50 earthquakes with magnitudes between 6.0 and 8.1 have occurred along these arc segments. Occupation of the Izu-Bonin and Mariana Islands started in prehistoric times, resulting in a major (50%) loss among the larger and edible flightless bird species on the islands. Their extinction continued during historic times, when colonists settled on the islands and introduced cats (and other predatory invasive species) to the further detriment of the indigenous birds. Even more dramatic was the involuntary introduction of the brown treesnake (Boiga irregularis) on Guam during World War II, which resulted in the disappearance of three seabirds, ten native forest birds, one native bat, and maybe as many as five native lizards from the island. A recent estimate places more than 2500 snakes per square kilometer on Guam, making it the island with the highest snake density in the world.

706

PAC I F I C R E G I O N

Bougainville and the Solomon Islands

Bougainville and the Solomon Islands are both part of a 1500-km-long island group in Melanesia containing about 900 islands. Most islands are volcanic in origin and fringed by coral reefs. However, many small islands exist that consist of uplifted reefs (up to 800 m above current reef levels) that formed by regional tectonic forces as a result of the (failed) subduction of the Ontong Java Plateau. In fact, the downward-going Pacific Plate is bulging upward at the Bougainville Trench, and Cretaceous volcanic rocks from the Ontong Java Plateau itself are obducted at this suture to form the islands of Maliata and St. Isabel. These volcanic islands are quite different (in composition and age) from the island arcs formed elsewhere in the Pacific Region. They are seismically active as well, with an 8.1 magnitude earthquake occurring on April 2, 2007, and triggering a tsunami with waves 5 to 10 m high on the Solomon Islands. Early settlers from Papua New Guinea started to arrive as early as 30,000 BC, yet it took until 1200 to 800 BC before the ancestors of the Polynesians arrived on the islands and started to explore the wider Pacific Region. Finally, the Battle of Guadalcanal between August 1942 and February 1943 was one of the most epic (and most costly) battles in the Pacific Region during World War II, turning the strategic table in favor of the Allied Forces. Kermadec, Tonga, and Fiji

The Kermadec–Tonga island arc forms a group of small South Pacific islands (and many more seamounts) between the tropical islands of Samoa in the north and New Zealand in the south. At 29° S latitude, the two arc segments join where Osbourne Guyot (the oldest seamount in the Louisville seamount trail) is being subducted. A seismic gap appears at this location, in what seismically and volcanically otherwise is the most active plate boundary in the world. The latest volcanic episode was highlighted by the formation of a new volcanic island at Home Reef in August 2006, whereas the latest high-impact earthquake, with a magnitude of 7.9, occurred about 160 km northeast of Nuku‘alofa on May 4, 2006. The Fiji Islands lie sub-parallel and to the west of the Tonga Islands and have a more prolonged history (starting around 40 million years ago), also related to the subduction of the Pacific Plate. However, over the last 3 million years volcanic rocks erupted on Fiji have a more intraplate (similar to those on, e.g., the Hawaiian or Samoan Islands) composition, reflecting its current location inbetween the Lau and Fiji back-arc basins and away from an active subduction zone. The Kermadec– Tonga–Fiji islands were first settled by Polynesians some

3000 years ago, yet with more than 100,000 inhabitants today, Fiji has become one of the most densely-populated island groups in the Pacific Region. As a result, significant environmental pressures have been exerted on the flora and fauna of the islands. Many species declined or disappeared because of the intense consumption by the large human population, whereas introduced alien mammals (rats and cats, in particular) decimated other forest plants and animals. Exporting for the international pet industry also decreased the numbers of (endemic) Fijian shining parrots and lorikeets, up to the level of critical endangerment or extinction. The Kingdom of Tonga is the only surviving monarchy existing in the Pacific Region. Fiji’s Great Sea Reef (north of Vanualevu) is one of the largest barrier reefs in the world. CONTINENTAL ISLANDS

Many continental islands formed following the last glacial period, after the Earth warmed up and sea level started to rise steadily. Sea level rise (including the effects caused by global warming) thus accounts for the formation of many of the islands along the continental margins of the Pacific Region. Land bridges that once existed during the last glacial lowstand were relinquished to the sea, truncating entire lineages of species and civilizations from the continental hinterland. Sea level rise, for instance, helped to form the Channel Islands off the California Coast and produced the configurations of multiple Japanese and Hawaiian islands as we know them today. Continental islands are often not volcanic in origin, like most other islands existing in the Pacific Region. Nearly all are part of terranes accreted to the continents over longer periods of subduction (e.g., microcontinents, ancient island arcs). Over time some eroded to elevations only slightly higher than sea level. Others separated from the main continental masses during periods of continental rifting, which in some cases were followed by the formation of new ocean basins through sea floor spreading. Under this scenario, the continental fragment of Papua New Guinea became an independent island once it separated from the Australian continent. Channel and Farallon Islands

The Channel Islands lie just off the coast of California and, during the last ice age, were connected to each other because of the lower global sea level. These eight islands have a geology similar to the granitic Sierras that form most of Southern California. The one big island that formed during the Pleistocene is named Santarosae and provided free roaming space for many species, including

a dwarf mammoth (now extinct) and many cypress and pine species. Even though not connected to the mainland, new species arrived at Santarosae on “debris rafts” carried across the narrow seaway by storms or shifting sea currents. The island fox (Urocyon littoralis) is the smallest North American canid that evolved as a special Channel Islands species since more than 10,000 years ago, after its predecessor, the gray fox, arrived by rafting or alternatively aided by the Chumash native people, for whom the fox was sacred. With the start of the interglacial period, the Channel Islands separated from each other, allowing the island fox (and many other species) to evolve into endemic island species. The introduction of the South African iceplant in the late nineteenth century to California and the Channel Islands has been devastating to native ecosystems. This plant now covers most of the islands with a thick cover and leaches high concentrations of salt into the native soil, making a saline environment inhospitable to most native plants. New Zealand

The North and South Islands make up the majority of New Zealand and together are about 1600 km long. Because of the inclusion of the Chatham Islands, located 800 km to the east, the country has the seventh-largest Exclusive Economic Zone (EEZ) in the world, about 15 times its land area. Mount Cook in the Southern Alps is the highest peak at 3754 m, whereas the same area includes another 17 peaks taller than 3 km. Most volcanism occurs on the North Island, with Mount Ruapehu being the tallest and most active volcano. Despite its active volcanism, New Zealand is not considered to be a volcanic island, as is common in the Pacific Region. Instead, New Zealand primarily consists of continental rocks and (together with New Caledonia) forms the submerged continent Zealandia. This continental fragment separated from the supercontinent Gondwanaland around 80 million years ago, giving many of the New Zealand taxa a “Gondwanan” character that dates back to many lineages that previously evolved when Gondwana was a single landmass. The arrival of the Polynesian Maori happened only a millennium ago, yet their habituation of New Zealand changed the existing lineages in flora and fauna drastically. For example, when these Maori arrived, about 30% of the land birds were flightless, but almost none survive these days as a result of the consistent consumption by people. European settlers in the early nineteenth century in their turn altered the New Zealand landscape considerably by clearing all but 22% of the original trees and vegetation. With the colonial immigrants came

PAC I F I C R E G I O N

707

33 introduced bird species, 34 mammal species, and 20 freshwater fish, together causing a loss of 41% of the 176 endemic species. Despite these large changes in the flora and fauna, New Zealand retains a spectacular and widely varying landscape that notably inspired the filming of the Lord of the Rings trilogy there. New Guinea

The island of New Guinea is the second largest island in the world and was formed when the Torres Strait flooded as sea levels rose following the last ice age. This effectively separated New Guinea from Australia, a transgression that is recorded in the emerged reef staircases on Huon Peninsula. This geological phenomenon provided the first (tectonically corrected) eustatic records that accurately determined past sea level elevations for the New Guinea region, with lowstands and highstands that varied between −17 and +10 m over the last 10,000 years compared to present-day sea levels. The highland peaks of Mount Wilhelm (4509 m) and Puncak Jaya (4884 m) are part of an east-west mountain range in New Guinea that is one of a few regions around the equator that receives snowfall and has permanent glaciers, now disappearing because of changing climate and global warming. Even though the first humans arrived on the island about 60,000 years ago, it took more than 50,000 years until the first plant domestications were successful in the highlands. At that time, the mountain population of hunters and gatherers started to cultivate many (indigenous) plants into garden crops, including sugar cane, bananas, yams, and taro. Today, much of New Guinea is still unexplored, leaving countless plant, insect, and animal species undiscovered. New Guinea also is infamous for its ritual cannibalism or “head-hunting” that was practiced by a few ethnic groups up to a few decades ago. A diet low in protein (owing to the small size of edible indigenous animals on the island) is often theorized to have sparked this morbid habit or ritual, even though pigs were introduced a few millennia ago. Vancouver Island

This large island off Canada’s Pacific coast formed as an accretionary terrane that is now part of the Western Cordillera in the Cascadia subduction system, a convergent plate boundary stretching south to northern California. This terrane, named Wrangellia, welded itself to the North American Plate about 100 million years ago, after it traversed from south of the equator and got entangled in an ancient version of the Cascadia subduction zone. Based on fossil records and paleomagnetic

708

PAC I F I C R E G I O N

measurements, it is clear that during the Upper Cretaceous the Wrangellia terrane was located around a 25° N paleolatitude, equivalent to the location of today’s Baja California. Most of the Vancouver Island landscape was formed by strong glaciations, as was the current outline of the island itself. Because of a combination of severe winds and strong ocean currents, many ships have been stranded or were wrecked on the rocky coasts of Vancouver Island. With as many as one wreck per mile, this island is referred to as the Graveyard of the Pacific. INTRAPLATE VOLCANIC ISLANDS AND ATOLLS

A noteworthy exception in plate tectonics is the formation of intraplate volcanic islands, such as the Islands of Hawaii, the French Polynesian Islands and the Samoan Archipelago, which each may require the presence of a mantle plume originating deep in the Earth’s mantle. Overall there are more than 750 volcanic islands in the middle of the Pacific, many of them arranged in linear island chains. The highest concentration can be found in the equatorial Pacific, in particular, in the triangular area between Samoa, Easter Island, and the Marshall Islands (Fig. 3). This also explains the high proportion of fringing reefs and atolls that formed around and on top of these volcanic islands because the tropical conditions in this part of the Pacific Ocean were most favorable for prolonged coral growth. Because many of these intraplate volcanoes are located far from the tectonic plate boundaries, they also are remote from the North American and Eurasian continents on the neighboring tectonic plates. As a result, the intraplate volcanic islands and atolls of Midway, Eniwetok, Bikini, Christmas, Johnston, Mururoa, and Fangataufa gained critical strategic value in the naval campaigns during World War II and became the remote testing grounds for the development of nuclear weapons during the Cold War era. From a biological standpoint, these volcanic islands are “hotspots” of biodiversity, characterized by species unique to only certain islands or island groups (high endemism) but with an otherwise limited variability (low species diversity) because of the small size of suitable habitats. The lineages of these species are typically short-lived since the intraplate volcanic islands do not survive for more than a few tens of million years in the plate tectonic environment. Some lineages may be prolonged through sequential colonization over the lifetime of entire island chains, whereby the species hop from island to island when older islands disappear underwater and new islands

form farther up the chain by new volcanism. The extreme isolation of each island (or island chain) within the Pacific Region makes the survival of many of its unique species more difficult, particularly upon invasion by non-native predatory species, parasites, or competitors. These invading species typically have been brought onshore by the Polynesian settlers and (starting in the early eighteenth century) European immigrants and caused an increased pressure on the island species, effectively reducing enemyfree spaces on the islands. Without refuge many island species diminished quickly, many toward extinction. For example, from the 120 island birds that have become extinct since 1500 AD, it is estimated that more than 80% were caused by the introduction of predatory carnivores on the remote islands of the Pacific Region. Caroline Islands

These islands are part of Micronesia in the West Pacific and include many small coral islands and a few larger volcanic islands. The larger islands are mostly located in the eastern part of the Caroline Islands and are considered erosional remnants of extinct hotspot volcanoes. Toward the west the volcanic edifices have mostly subsided and in some cases have only a limestone platform or coral top that surfaces above the Pacific Ocean. Hotspot volcanism started around 12 million years ago and formed the Caroline Ridge between Kosrae Island and Chuuk Atoll. These are the youngest intraplate volcanoes that lie just south of the otherwise Cretaceous atolls and seamounts in the West Pacific Seamount Province. All other islands in the Caroline archipelago have more complex histories, including the formation of ancient island arcs, transform faults, spreading centers, and intraplate volcanoes. The Caroline Islands also are located close to an area of intense typhoon activity. While typhoons in the West Pacific can occur at any time, they mostly occur between February and April, with five severe typhoons per year that pass over or close by the Caroline Islands, through the so-called Typhoon Corridor. Cook-Austral Islands

The Cook and Austral Islands are both part of Polynesia and form a complex system of short volcanic chains in the South Pacific between Samoa and the Society Islands. Fifteen islands make up the Cook Islands and range in size from the large volcanic island of Rarotonga to the smallest solitary atolls in the north. Only seven islands make up the Austral Islands, which actually lie on a continuation of the Southern Cook Islands that converges on the recently active Macdonald volcano. Although the

islands generally are believed to have formed by hotspot volcanism, the age and distance systematics of this intraplate island cluster are rather complex. Alternative processes may be required to explain the complications, including tectonic control on the Pacific Plate, the presence of multiple plumes/plumelets in the mantle underneath, or even the upwelling of a 2000-km-wide hotline, rather than only a hotspot, of mantle material. The Cook–Austral Islands lie just east of the International Date Line and were inhabited between 500 and 800 AD by Polynesians crossing the Pacific Ocean from the islands of Tonga, Samoa, Tahiti, Marquesas, and Society. The extreme isolation of these islands toward the southern boundary of Polynesia makes for a low biological diversity, yet there is a significant difference in flora and fauna between the Cook–Austral islands and atolls. About 4% of the angiosperms are endemic to the islands, yet the Miti‘aro fan-palm and the Pacific fruit bat of Rarotonga and Mangaia are unique to this part of the world. Today most of the vegetation is dominated by species introduced by Polynesian and European settlers, including taro, yam, ferns, coconut palms, bananas, and papaya. Because the livable area of the combined 22 main islands is extremely small, solid waste management, land use practices, overfishing, and ecotourism are all taking a large toll on the islands’ habitat and their delicate ecosystems. For example, wetlands often are used as waste disposal sites or filled in for construction, natural drainage and groundwater systems are disturbed by extensive land cultivation, and coastal landscapes have been altered by sand mining practices. Uncontrolled fishing in the numerous lagoons of the Cook–Austral islands have had destructive effects on fish populations and the health of the coral reefs, as have numerous toxins from sewage and pesticides leached into the lagoons. Easter Island

This solitary volcanic island is one of the most remote islands in the Pacific Region, located more than 2000 km east of Pitcairn Island, the closest inhabited island in its vicinity. Its remoteness also explains its late discovery by the Marquesas or Mangareva people in 400–600 AD, making it the last inhabited island by the Polynesians. Easter Island is rather small (164 km2) and forms only one of two isolated subaerial volcanoes along the 2700 km Easter–Sala y Gomez ridge on the Pacific Plate. Plate tectonic reconstructions show that Easter also is related to the formation of the Nazca and Tuamotu ridges on the Nazca Plate. These submarine ridges formed from the same Easter hotspot when it was located directly underneath the mid-ocean spreading

PAC I F I C R E G I O N

709

center of the East Pacific Rise, simultaneously forming two hotspot trails on two tectonic plates (in a remarkable Vshaped pattern). However, Easter hotspot volcanics less than 3 million years old are rather unusual because of their continental-type igneous character. These “granitic” rocks confused many early geologists, including Alfred Wegener, who used this oddity in labeling Easter Island as a remnant of continental drift. At a site called Rano Raraku these rocks exist in the form of volcanic ash and tuff deposits that are easily workable and thus were used to carve out most of the moai, the famous stone statues or “heads” of Easter Island. Almost 900 moai statues have been found on Easter Island, and generally they are thought to have been erected between 1000 and 1700 AD. By the time the last statue was erected, overpopulation of the island and profligate consumption of trees in the transport of the statues seem to have caused a sudden drop in crop productivity and the loss of all forests and their resident birds and animals. With that, the Easter population was decimated and did not stabilize until the Europeans settled on the island. The census of 2002 shows that currently only 3791 people reside on Easter Island, one of the least dense island populations in the Pacific Region. Galápagos Islands

This archipelago of 13 islands, shown on sixteenthcentury maps as Insulae de los Galopegos (Islands of the Tortoises), was formed starting about 10 million years ago by the Galápagos hotspot. This hotspot is located close to the equator (about 965 km off the coast of Ecuador) and is still active, with the latest volcanic eruptions taking place in 2005 on the islands of Isabela and Fernandina. Before 8 million years ago the Galápagos hotspot was located on the Galápagos Spreading Center and, like the Easter hotspot, formed two concurrent volcanic traces on two oceanic plates, namely the submarine Carnegie and Cocos Ridges. Both ridges delineate the northeastern or southeastern continuation of the Galápagos Islands, except that they are entirely composed of seamounts that once were emerged volcanic islands in the same archipelago. Today this hotspot resides about 200 km south of the Galápagos Spreading Center and only continues to form the Carnegie Ridge on the Nazca Plate. Because of its close proximity to this spreading center, the Galápagos Islands have been formed on relatively thin oceanic lithosphere with increased tectonic stresses. This may explain the complex clustering of the Galápagos volcanic islands, which is contrary to the classical linear alignment of volcanic islands for the Hawaiian hotspot.

710

PAC I F I C R E G I O N

The Galápagos Islands do not have any historical settlers, so these islands were not discovered until European explorers landed in 1535. By the late eighteenth century, whalers and sailors were using Galápagos as a base station and started to kill and capture many thousands of the Galápagos tortoises for their meat, rich in protein and fat, bringing many species to the brink of elimination and some to extinction. Charles Darwin visited the islands in 1835 onboard the HMS Beagle, and even though this young naturalist spent only one month on Galápagos, his discoveries and observations on the geology and biology of the islands formed the very basis of his renowned book On the Origin of Species. On Galápagos Darwin studied, in particular, the subspecies of finches and found that there are distinct differences in these species among the 13 islands. The Galápagos Islands also are home to the marine iguana, which according to recent DNA studies evolved from a common land ancestor not more than 20 million years ago. To explain this recent observation, the drowned seamounts in both the Carnegie and Cocos Ridges must have acted as steppingstones covering the entire evolution of this species, starting more than 20 million years ago. Despite the fact that the Galápagos Islands have been occupied for only half a millennium, invasions by other (mammal) species have occurred again and again, and changes to the islands’ habitat and species inventory have been significant. For example, goats introduced on Isla Pinta in 1959 reproduced extremely rapidly, reaching a population of 20,000 by 1971. These introduced animals prevented regeneration of trees because of their intense seedling browsing. As a result, the original vegetation on this Galápagos island survived only in a few areas inaccessible to the goats. Hawaiian Islands

Formerly known as the Sandwich Islands, Hawaii appears as a linear group of 19 islands and atolls about 2600 km long and extending from the island of Hawaii (the Big Island) in the southeast to the atolls of Midway and Kure in the northwest. In this island chain the volcanoes become gradually older toward the northwest, where they are volcanically inactive, have almost no surface morphology, and are entirely capped with coral reefs. In the 1960s these observations were confirmed by radiometric age dating using the K/Ar method. These ages indeed showed a systematic increase from less than 0.6 million years on the Big Island to about 5.8 million years at Kauai (where the very last subaerial volcanic rocks could be sampled) to about 27 million years for rocks cored at Midway Atoll.

With these data in hand, geophysicists J. Tuzo Wilson and W. Jason Morgan theorized that a stationary and deep-seated mantle plume impinges on the overriding Pacific Plate, producing a trace of volcanic islands on top of that plate that becomes older toward the northwest and that directly reflects the direction and speed of past Pacific plate motions. In this model the Hawaiian hotspot must be located almost underneath the “zero-aged” Big Island, which explains the ongoing extrusion of magma at K1¯lauea crater, where lava flows described as pa¯hoehoe (low-viscosity lava that is smooth and ropey when hardened) and ‘a‘a¯ (high-viscosity lava that is rough when hardened) have been erupting from 1983 until today at a rate of ∼0.1 km3 per year. Seismic activity is not only related to the movement of magmas in the 3–7-km-deep magma chamber of K1¯lauea. On October 15, 2006, an earthquake with a 6.7 magnitude was recorded off the northwest Kona coast. Other large earthquakes are triggered by large landslides and avalanches on the (submarine) flanks of the Hawaiian Islands and have also generated local tsunamis in the past. For example, tsunami deposits have been found as high as 170 m above current sea level on the leeward side of the islands of Kohala, Maui, Lanai, and East Molokai. Tsunamis also have been caused by earthquakes happening along the outer perimeter of the Pacific Region, in the Ring of Fire. In 1960, the Great Chilean earthquake (believed to have had a 9.5 magnitude) resulted in a large tsunami that devastated Hilo with waves reaching 11 meters. Despite the remoteness of the Hawaiian Islands from Chile, this tsunami was able to kill at least 61 people. The isolation of these islands toward the north side of Polynesia has dictated the limited species inventory of Hawaii as well, which originally did not include ants, ginger plants, terrestrial reptiles, or amphibians, and only two mammal species. Most of these Hawaiian species have rather short lineages since they started to differentiate only about 6 million years ago, at the time when the volcanic island of Kauai was formed. A good example are the Hawaiian honeycreepers, representing a diverse group of 56 endemic species of cardueline finches, which evolved from a single ancestral species about 3 to 4 million years ago. More than 70% of these finch species are by now extinct, and most of the remaining species are currently endangered. Hunting, habitat modification, or the introduction of predators (such as the rat) by the early Polynesians may have caused these extinctions. Modern threats to the honeycreepers include their particular susceptibility to the avian poxvirus and malaria, spread around the oceanic islands by mosquitoes originating in Africa.

Line Islands

The Line Islands include a group of 11 atolls and low coral islands, located in extreme isolation in the central Pacific. Only three islands are inhabited, with a population of 8809 in 2005, up from 300 in the early twentieth century. Kiritimati, or Christmas Island, is the largest atoll in the world at 642 km2 and was first discovered by Captain James Cook on Christmas Eve in 1777. The Line Islands, and in particular Palmyra Atoll and Kingman Reef, contain the most pristine coral reefs remaining in the Pacific Region and the world. Geologically speaking, the formation of this island group is undecided, because they are part of a linear 4,000-km-long chain of atolls, submarine ridges, and seamounts that is rather complex, with numerous en echelon and cross-cutting seamount chains. Based on recent radiometric age dating of basaltic lavas, this island chain does not possess an obvious linear age progression, unlike the Hawaiian Islands. This requires alternate explanations, including the possibility that more than one hotspot may have been in play during the formation of the Line Islands or that extension in the Pacific Plate may have created volcanic islands and seamounts during three or four narrow eruption periods. The United States and Great Britain used Christmas Island for testing hydrogen bombs during Operation Grapple in 1957 and Operation Dominic in 1962. Marquesas, Pitcairn, and the Society Islands in French Polynesia

French Polynesia comprises several island chains over an area of 2.5 million km2 in the equatorial Pacific, including the Marquesas, Pitcairn, Society, Tuamotu, and Austral islands. Their combined land area, including 118 islands and atolls, is rather small, yet with 84 atolls French Polynesia contains the majority of all atolls on Earth. All of these island chains are volcanic in origin and are thought to have been formed by hotspots, similar to the Hawaiian Islands. Most French Polynesian volcanic islands are younger than 16 million years, except Tuamotu, which is as old as 40 million years. As an archipelago, French Polynesia is the most isolated in the world; it has a rather limited flora and fauna of which 80 endemic species are documented to have become extinct, and because of its generally low relief it is one of the Pacific island groups that will endure the most immediate consequences of any future sea level rise driven by global warming. France used the atolls of Fangataufa and Moruroa for the (underground) testing of nuclear bombs between 1970 and 1997. The Marquesas Islands are relatively young, with its oldest island dated at 3.8 million years. They contain 14

PAC I F I C R E G I O N

711

volcanic islands and one atoll about 9˚ south of the equator, with Mount Temetiu on Hiva Oa being the highest volcanic peak at 1190 m above sea level, and Nuku Hiva being the second largest island in French Polynesia. Only 8632 people were living on the islands as of the 2007 census, despite their early discovery by Polynesians around 150 BC and the fact that in the sixteenth century an estimated 100,000 Polynesians inhabited the islands. Diseases such as measles, syphilis, dysentery, smallpox, tuberculosis, malaria, and leprosy were brought in by the Europeans, and frequent epidemics decimated the population to their current level by the end of the nineteenth century. The age-distance relationships between the Marquesas Islands provide us with a 10.4 cm/year estimate for the past velocity of the Pacific Plate, which is similar to, and thus confirms, the observations for the Hawaiian hotspot. However, the 115˚ azimuth of the Marquesas island chain is ∼20˚ different from that of Hawaii and the other volcanic chains in French Polynesia, which suggests that the formation of the Marquesas islands and the emplacement of its volcanoes may have been controlled by weaknesses in the Pacific Plate induced by the Marquesas Fracture zone. The Duke of Gloucester–Moruroa–Gambier–Pitcairn island chain is 1650 km long and runs south of the Tuamotu plateau. The oldest volcanic island was formed around 12 million years ago, whereas Pitcairn itself formed around 0.45 million years ago, and active volcanism is happening 100 km southeast of Pitcairn at any one of 20 submerged seamounts in this chain. The velocity of the Pacific Plate at this location is estimated at 11.0 cm/year and thus is well attuned with other estimates in French Polynesia and the Hawaiian Islands. Pitcairn Island is not surrounded by a fringing reef. Pitcairn became uninhabited when the original Polynesian settlers died out in the cold climate interval of the fourteenth century. In 1790 it was settled again, but this time by mutineers from the HMS Bounty, a Royal Navy ship charged by King George III to sail to Tahiti to obtain breadfruit plants. The mutineers, a mixed party of nine British sailors, six Polynesian men, and 13 Polynesian women, sought refuge on the island and established a small society that today still occupies the island with less than 50 remaining descendants of the original eighteenth-century families. The Society Islands are maximally 4.3 million years old, and the plate velocity measured based on the ages and distances between the islands is 10.9 cm/year. This island chain is composed of five atolls and nine islands, of which Tahiti is the largest island and has the highest volcanic peak at 2241 m above sea level. This island consists of two coalescent eruptive systems and was active

712

PAC I F I C R E G I O N

prior to 0.87 million years ago. The youngest volcanic activity is concentrated around Mehetia, where volcanic eruptions started around 0.3 million years ago and new seamounts are forming 50 km southeast. The insect fauna species groups on the Society Islands (and French Polynesia) have adaptive radiation patterns that in many cases have resulted in 100% endemism. Good examples are 70 endemic species of Mecyclothorax on Tahiti, 35 species of Rhyncogonus in French Polynesia, and 29 endemic species of black flies Simulium in the Society Islands. Many endemic species also disappeared, such as 60 tree snails following the introduction of the carnivorous snail Euglandina rosea. Marshall Islands

The Marshall Islands consist of the oldest surviving atolls (29) and coral islands (5) on Earth that formed on top a group of hotspot volcanoes that erupted in the French Polynesia area between 97 and 67 million years ago. Today the islands are organized in the sub-parallel Ratak and Ra¯lik (sunrise and sunset) chains and were settled by Micronesians in the second millennium BC. British Captain John Marshall was the first European to visit the islands in 1788. During World War II, the Marshall Islands were occupied by the United States, who also started to use the atolls for the testing of their nuclear weapons almost directly after the war. In total, the United States tested 66 nuclear weapons between 1946 and 1958, including the 1954 Castle Bravo, the largest nuclear detonation by the United States, with a yield of 15 megatons and the first ever test of a dry-fuel thermonuclear fusion bomb. Bikini Island (and many other atolls in the Marshall Islands) are still unsafe because of the lasting effects of radiation from the contaminated soils (mostly by wind driven fallout) and thus remain uninhabited following the evacuation of the Micronesian people. The native flora and fauna are dominated by over 1000 fish species and 100 coral species, yet no endemic mammal species exist on the islands, and only one endemic bird subspecies, the Ratak Micronesian Pigeon. Samoa and American Samoa

The Samoan archipelago is divided into Western Samoa (officially Independent Samoa or Samoa) and American Samoa (an unincorporated territory of the United States) to the east. Samoa and American Samoa have populations of 214,265 and 57,291 respectively (from censuses in 2000), of whom more than 90% are native Samoans. Geologically speaking, the archipelago is a volcanic hotspot chain of

nine main islands and coral atolls located between 13˚ S and 15˚ S in the South Pacific. Savai’i (the fifth largest island in the tropical Pacific) is the oldest island in Samoa, which started to form around 5.0 million years ago, whereas Vailulu’u is the most recent volcano that still resides underwater and to the east of Ta’u Island. Hydrothermal and volcanic activity on Vailulu’u are ongoing, with the latest volcanic episode occurring in 2003, at which time a new 350-m-high volcanic cone (named Nafanua) formed inside the main crater, at approximately 1 km water depth. Rose Atoll, located farthest east in the Samoan archipelago, is much older, suggesting that its evolution began where French Polynesia is located today and from an entirely different hotspot in that area. The first settlers in Samoa arrived from Malesia in the west more than three millennia ago, and the Manu’a Islands therefore have one of the oldest Polynesian histories. Because of the proximity of Samoa to New Guinea and the Australian continent, most biodiversity in the terrestrial and marine species is derived from these regions. For example, the native Samoan flora is the largest in Polynesia and includes 550 angiosperm species and 228 pteridophyte species, 32% of which are endemic. Modern hunting, however, has brought down the extant purplecapped fruitdove (Ptilinopus porphyraceus) and manycolored fruitdove (Ptiliopus perousii) to rare and declining subspecies of manutagi. On the other hand, the Pacific pigeon (Ducula pacifica), or manume`a, is still common on Savai’i and Upolu and is important to Samoan culture with its haunting vocalizations. Now the national bird of Samoa, the pigeon is threatened by hunting and the loss of forest habitat. The recurrence of devastating cyclones in combination with increased agricultural deforestation have caused the once completely forested Samoan Islands to loose 60% of its rainforest cover in less than 40 years. SEAMOUNTS AND GUYOTS

Seamounts are underwater volcanic mountains that are taller than 500 m and form isolated edifices (or ridges) on the ocean floor. The term seamount is typically used very broadly to describe all submarine features ranging from the smallest bathymetric disturbance to the largest volcanic structure that spans up to tens or hundreds of miles across its base. On the other hand, guyots are more specifically regarded as remnants of volcanic islands that eventually got planed off by wave erosion, resulting in the formation of typically large, flat-topped seamounts. These ancient volcanic islands disappeared from the Earth’s sea surface because the tectonic plates they are riding on cool

down over geological time, making the plates more dense and allowing them to sink into the upper mantle, taking the guyots with them into the deep of the oceans. However, once drowned, the seamounts do not survive far back into geological time, because the subduction zones surrounding the Pacific have been destroying all the older ones. The oldest dated seamount is Look Seamount, located to the west of the Marshall Islands, with an age of ∼140 million years. Compared to our knowledge about the volcanic islands in the Pacific Region, we know very little about seamounts and guyots and how they formed. Knowing how many are present in the Pacific Ocean is an even more difficult question, because our maps of the ocean floor (still) are rather incomplete and not even up to par to the detailed topographic maps we have of the Moon, Mars, or any other planet body in our solar system. Not only are the seamounts poorly mapped; they have hardly been surveyed or sampled. This makes deciphering their histories a challenge, and thus it remains unclear how many islands have existed (and perished) in the Pacific Region over the last hundred million years. We also understand very little about the role of ancient seamounts (once volcanic islands) in the island-hopping of certain species, thereby significantly prolonging their lineages in certain island groups through time. Cobb Seamounts

The Cobb–Eickelberg seamount trail is one of two hotspot trails existing in the northeast Pacific, located west of the Oregon and Washington coast. This seamount chain stretches from the 33-million-year-old Patton and Murray seamounts near the Aleutian trench to Axial Seamount, the youngest volcano in the chain, with eruptions and a magnitude 4.7 earthquake occurring as recent as January 28, 1998. The flat-topped Cobb Seamount formed during the Oligocene, but drowned only recently (because of the global sea level rise following the last ice age) and thus its summit is submerged only 34 meters below the sea surface. These shallow depths sustain a rich marine environment with fish communities typical for the subarctic North Pacific and the hard rock substrate of this volcanic seamount. Rockfish (Sebastes) species seem to dominate and be isolated around Cobb, where their larvae do not spread out more than 30 km from the seamount. With its active volcanism Axial Seamount denotes the current location of the Cobb hotspot, which also intersects the Juan de Fuca Ridge at this point in time. The coincidence with this midocean spreading center has a clear effect on the evolution of Axial Seamount, which has anomalous shallow bathymetry

PAC I F I C R E G I O N

713

due to an oversupply of magma and a geochemistry with mixed hotspot and mid-ocean ridge signatures. Foundation Seamounts

The Foundation chain is unlike the Hawaii–Emperor hotspot chain, because its volcanoes are volumetrically small and because no islands exist in this hotspot chain, only seamounts. The 1350-km-long seamount chain is located about 5 degrees to the southeast of Easter Island, it includes dozens of seamounts up to 21 million years in age, falling in a narrow band maximally 200 km wide, and overall the chain itself trends parallel to most other seamount chains in the Pacific Region at 70˚ west of north. A first-order age progression is evident and gives a plate velocity of 9.1 cm/year, but in detail the age systematics are more complex. For example, over the last 10 million years the Foundation hotspot has been slowly approaching the Pacific–Antarctic spreading center, which instilled many small-scale complexities in the resulting Foundation seamount chain. Secondary mechanisms such as the “reheating” of normal oceanic crust, local stress-induced fracturing of the lithosphere, the interaction with the growing Selkirk microplate, and the creation of “extra thin” lithosphere as a result of the combined hotspot and ridge melting, may all have played a role in the formation of the Foundation Seamounts and thus cause the observed disturbances in the age-distance relationships. Hawaii–Emperor Seamounts

The Hawaii–Emperor seamount chain is 5800 km long and contains more than 80 submerged seamounts and guyots. It starts out with the Meiji Seamount, 83 million years old, located at the edge of the Aleutian Trench, and extends to Loihi Seamount, 35 km southeast of the Big Island of Hawaii. Loihi is volcanically and hydrothermally active, is a relatively small submarine volcano with a summit that lies still 980 m below the sea surface, and generally is considered to be situated directly above the Hawaiian hotspot. With that Loihi represents the “preshield” stage in the volcanic evolution of a typical Hawaiian volcano, a standard model adopted by many researchers studying hotspot volcanoes worldwide. The slightly older Big Island of Hawaii and the current eruptions at K1¯lauea crater represent the “shield” stage, which follows after a few hundred thousand years, which is the main volcanic stage responsible for building up more than 90% of the hotspot volcano structure. One of the most engaging characteristics of the Hawaiian hotspot is that its volcanoes become gradually older toward the west and that beyond Kure Atoll the hotspot chain bends

714

PAC I F I C R E G I O N

sharply to the northwest and continues underwater in the Emperor seamounts. Traditionally, these observations have been interpreted to reflect a sudden and pronounced 60˚ change in the motion of the Pacific Plate, now dated to have occurred about 50 million years ago. However, recent paleomagnetic measurements on ocean drilling samples from Detroit, Suiko, Nintoku, and Koko seamounts in the Emperors is casting some serious doubt on this interpretation. These measurements show that the paleolatitudes of these seamounts are up to 15˚ different from the latitude of the Hawaiian hotspot itself, which would be impossible if this hotspot and its associated mantle plume were to be fixed in the Pacific mantle over the last 80 million years. The Hawaiian hotspot is showing signs of significant motion prior to 50 million years ago instead, which may entirely or partly explain the bend in the Hawaii–Emperor seamount chain. Louisville Seamounts

Formerly called the Louisville Ridge, this 4300-kmlong seamount chain is the only true counterpart of the Hawaiian hotspot in the Pacific Region. In a similar fashion as Meiji, the oldest Hawaii–Emperor seamount, Osbourne Guyot (76–79 million years old) at the northwestern end of the Louisville chain is being consumed by the Tonga– Kermadec subduction zone. The likeliness to the Hawaiian hotspot continues, because the Louisville seamount chain also has a rather systematic age progression and because it has been active for more than 75 million years as well. No other seamount trail in the Pacific Region shows the same kind of prolonged volcanic history as the Hawaiian and Louisville hotspots. The outstanding question now is whether the Louisville hotspot shows the same kind of motion in its mantle plume before 50 million years ago. Tokelau Seamounts

The Tokelau seamount chain mainly consists of ancient submarine volcanoes, yet the Howland and Baker Islands appear in the north of this chain, whereas three Tokelau atolls delineate the southern most end. The chain also shows a bend, equivalent to the bend in the Hawaii– Emperor chain, toward its southern end, yet the Tokelau seamounts dated with the 40Ar/39Ar method range in age between 58 and 72 million years, making this bend at least 8 million years older than the start of the Hawaii–Emperor bend. Beside the latest evidence for moving mantle plumes and hotspots, the asynchronous bends create another enigma for the classical hotspot hypothesis, which would predict that changes in Pacific Plate motion create bends in all “active” hotspot trails at the exact same time.

West Pacific Seamount Province

The ocean floor in the West Pacific is littered with seamounts, guyots, and atolls. Most of these already formed in the early Cretaceous, yet many still have surviving coral reefs, even tens of million of years after drowning. Paleomagnetic measurements show that these seamounts formed in the tropics between 10º S and 30º S, but despite their motion to the north over time due to Pacific Plate motions, these seamounts have remained within reach of the equatorial Pacific, explaining their sustained coral growth, which was able to keep track of the gradual subsidence of these aging volcanoes. The Western Pacific seamounts also are important fishing grounds for bottomfish and seamount groundfish. However, many seamounts experience overfishing, which is especially harmful because many lutjanid, serranid, and lethrinid species (and their larvae) only appear in isolated groups that differ per seamount without many interconnections. Overfishing thus easily disrupts the ecosystems on the seamounts, making reestablishment of some fish stands slow to impossible. Protection of these seamount fish stocks from environmentally destructive fishing therefore is paramount in keeping the seamount fish populations healthy.

volcanic islands emerged approximately 30 million years ago during the late Oligocene. GEOGRAPHY

Palau, approximately 7º30′ N latitude and 134º35′ E longitude, extends northeast to southwest for 700 km, with over 500 islands and an estimated total land area of 450 km2. One island, Babeldaob, accounts for 75% of the land area. The climate is wet tropical, with annual rainfall of 350–450 cm, temperatures of 22–32 ºC, and relative humidity averaging 82%. A coral barrier reef system surrounds five central islands—from north to south, Babeldaob, Koror, Arakabesan, Malakal, and Peleliu—in a lagoon that covers over 1,000 km2. Hundreds of small raised coralline limestone islands are distributed throughout the southern lagoon. These are the famous (and photogenic) Rock Islands of Palau, which are a popular tourist attraction (Fig. 1).

FURTHER READING

Menard, H. W. 1986. The ocean of truth: a personal history of global tectonics. Princeton, NJ: Princeton University Press. Menard, H. W., and H. S. Ladd. 1963. Oceanic islands, seamounts, guyots and atolls, in The sea, Vol. 3. M. N. Hill, ed. New York: Wiley & Sons, 365–387. Nunn, P. D. 1994. Oceanic islands. Oxford: Blackwell Publishers. Wegener, A. 1915. Die Entstehung der Kontinente und Ozeane. Braunschweig: Sammlung Vieweg Heft. FIGURE 1 One of Palau’s famous rock islands. These raised limestone

formations are found throughout the southern lagoon. Photograph by Milang Eberdong.

PALAU ALAN R. OLSEN Belau National Museum, Koror, Palau

Located in the western equatorial Pacific Ocean, Palau is the westernmost group of islands in Micronesia. The archipelago rests on the eastern edge of the continental shelf of the Philippine Plate, approximately 800 km east of Mindanao. Palau is a mixture of old volcanic islands, raised limestone islands, coralline platform islands, and atolls representing an exposed crest of the now-dormant southern section of the Palau–Kyushu Ridge. It is estimated that the

To the south, the platform island of Angaur is separated from the lagoon by a deep channel. Five small remote platform islands occur 300–600 km southwest of the lagoon: Fanna, Sonsorol, Merir, Pulo Anna, and Tobi. There are three atolls: Ngeruangel and Kayangel lie north of the lagoon and Helen Reef is south of Tobi Island. Helen Reef is a protected sanctuary for coral reefs, marine turtle nests, and seabird rookeries. HISTORY AND CULTURE

According to archeological evidence, Palau was settled 1000–3000 years ago. The exact origins of modern Palauans are subject to debate. Palau’s matrilineal culture retains its language and many ancient traditions, which

PA L A U

715

are distinct from others in Micronesia. The people of the remote southwest islands are distinct from other Palauans, having cultural and linguistic affinities with the peoples of central Micronesia. The Spaniard Ruiz Lopez de Villalobos made first European contact in 1543. The Englishman Henry Wilson was shipwrecked in Palau in 1783, leading the British Empire to claim Palau. In 1885, Pope Leo XIII restored the islands to Spain, which sold Palau to Germany in 1899. At the end of World War I, Palau was ceded to Japan and then to the United States of America following World War II. Palau became independent on October 1, 1994. Currently, there are 20,000 Palauans, the majority living on the central volcanic islands. BIODIVERSITY

Palau has many types of marine habitats that support a diversity of life. In addition to 235 km of coral barrier reef, there are coral fringing reefs and mangrove forests associated with many islands. Over 50 marine lakes are located in the interiors of various rock islands. These unusual lakes are connected to the lagoon by fissures and tunnels in the limestone rock. Some of the lakes contain unusual fauna such as Palau’s golden medusa (Mastigias sp.) jellyfish, which have algae incorporated into their tissues. The algae provide a portion of the medusas’ nutritional needs through photosynthesis. Palau’s marine biodiversity is reflected in the 385 species of reef-building corals found there. Marine fauna includes an endemic nautilus (Nautilus belauensis), sea skaters (Halobates), 1450 species of fish, and breeding populations of green turtles (Chelonia mydas), hawksbill turtles (Eretmochelys imbricata), saltwater crocodiles (Crocodylus porosus), and endangered dugongs (Dugong dugon). Terrestrial habitats include a freshwater lake, rivers, tropical rain forests, swamp forests, and savannas. Palau’s terrestrial habitats are largely undisturbed, with little deforestation. For the most part, the terrestrial flora and fauna have affinities with the biota of the large land masses in the Philippines and Indonesia with minor influences from Australia, Papua New Guinea, and Polynesia. Terrestrial biodiversity encompasses an estimated 1200 species of plants, an estimated 10,000 invertebrates, 38 reptiles, and the only endemic frog (Platymantis pelewensis) in Micronesia. The avifauna numbers 148 recognized species, with 11 endemic landbirds. Palau harbors populations of the Micronesian megapode (Megapodius laperouse), an endangered mound-building forest bird extinct throughout much of its original range in Micronesia. There are two bats: the Pacific sheath-tailed bat (Emballonura semi-

716

PA L A U

caudata) and the Micronesian flying fox (Pteropus mariannus). An endemic bat, Pteropus pilosus, is believed extinct. Palau has an active conservation program that includes marine protected areas, a protected area network, ecosystem-based management initiatives, important bird areas and ecosystem management plans for the southern lagoon, and other important areas. There are active programs to monitor coral reefs, crocodiles, dugongs, forest birds, and sea turtles. The Micronesia Challenge, which was recently issued through the Convention on Biological Diversity in 2006, is a regional conservation initiative that originated in Palau. FOLKLORE: AN UNUSUAL ICON

Interestingly, a spider is identified with a benevolent demigod who is central to the folklore of Palau’s matrilineal society. Palauan folklore features a cycle of myths centered on an unusual demigod, Mengidabrudkoel, who transformed from spider into human form to introduce traditional rituals to Palau such as those involving childbirth. Among the many man-spider legends is a popular tale of a mythical island where Mengidabrudkoel planted a magical breadfruit tree that yielded a fish when a branch was broken off. Greedy villagers chopped down the tree expecting to reap a bounty of fish. Instead, a flood of ocean water poured from the tree stump to sink the entire island under the sea. Another legend in the cycle resembles the biblical account of Jonah and the whale. Spider imagery is firmly embedded in modern Palauan society, with ubiquitous icons of spiders appearing in architecture, art, handicrafts, postal stamps, textiles, and logos of government agencies and athletic organizations. Palau’s giant golden orb-weaver (Nephila pilipes) (Fig. 2),

FIGURE 2 Palau’s giant golden orb-weaver, Nephila pilipes, is found

throughout Palau except the southwest islands. Photograph by Alan R. Olsen.

a 15-cm spider, is the living symbol of the demigod and bears the vernacular name mengidarudkoel. The English name alludes to the golden hue of its 3-m web.

Orinoco An d

es

SEE ALSO THE FOLLOWING ARTICLES

Archaeology / Marine Lakes / Reef Ecology and Conservation

GH

FURTHER READING

Colin, P. L., and A. C. Arneson. 1995. Tropical Pacific invertebrates. Beverly Hills, CA: Coral Reef Press. Colin, P. L. 2007. Marine environments of Palau. Koror, Palau: IndoPacific Press. Crombie, R. I., and G. K. Pregill. 1999. A checklist of the herpetofauna of the Palau Islands (Republic of Palau), Oceania. Herpetological Monographs 13: 29–80. Force, R. W., and M. Force. 1972. Just one house: a description and analysis of kinship in the Palau Islands. Bernice P. Bishop Museum Bulletin 235. Kayanne, H., ed. 2007. Coral reefs of Palau. Koror, Palau: Palau International Coral Reef Center. McManus, Fr. E. G. 1977. Palauan-English dictionary. J. S. Josephs, ed. PALI Language Texts: Micronesia. Honolulu, HI: University Press of Hawaii. Wiles, G. J. 2005. A checklist of the birds and mammals of Micronesia. Micronesica 38: 141–189.

Amazon

FIGURE 1 Radar view of northern South America showing the placement

of the Guayana highlands (GH) region, with respect to the Orinoco and Amazon basins, and the Andean range. (Image courtesy of NASA/JPL Caltech.)

PANTELLERIA SEE MEDITERRANEAN REGION

PANTEPUI VALENTÍ RULL Botanic Institute of Barcelona, Spain

FIGURE 2 Close-up view of the Guayana highlands, showing this area’s

characteristic tabular topography, composed of several erosion surfaces and culminated by the tepuis. Lowlands (100–500 m altitude) are in green and yellow, whereas uplands and highlands (500–1500 and 1500–

Pantepui (pan, Greek for “all,” and tepui, South American indigenous name for “table mountains”) is a discontinuous biogeographical entity shaped by the assemblage of the flat-topped summits of the Guayana (northern South America) table mountains, or Guayana Highlands (Figs. 1 and 2), above 1500 m in altitude. These summits are isolated from each other and from the surrounding lowlands by spectacular vertical cliffs, and they hold a singular biota with unique adaptations and amazing levels of biodiversity and endemism. The origin of such biotic patterns is a still-unresolved evolutionary enigma. THE TEPUIS

The indigenous (Pemón) word tepui, meaning “stone bud,” has been adopted as a physiographical term to name the table mountains of the Guayana Highlands (e.g., Auyán-tepui). A typical tepui is a tabular moun-

3000 m, respectively) are in light brown. (Image courtesy of NASA/JPL Caltech.)

tain made of sandstones and quartzites (with occasional intrusive rocks, mostly diabases), with a more-or-less flat summit limited by a rim, and isolated from the surrounding lowlands by vertical escarpments in the upper part and steep talus slopes in the foothills (Fig. 3). To understand the origin of the tepuis, it is necessary to go back to the Cretaceous (145 to 65 million years ago), when Africa and South America were joined in the Gondwana supercontinent. The separation began around 80–100 million years ago and determined the initial opening of the Atlantic Ocean, which led to the formation of the huge Amazon and Orinoco basins, among others. By that time, the Guayana region was covered by extensive erosional plains modeled on the Precambrian sandstones and quartzites of the Roraima

PA N T E P U I

717

FIGURE 3 View of the Upuigma-tepui, showing the typical flat sum-

mit (2100 m altitude and more than 1 km2 in surface area), the vertical cliffs, the extensive forested slopes, and the basal level, which in this case is part of the so-called Gran Sabana, and is around 850 m in altitude. Note the characteristic savanna vegetation, spiked by clusters of “morichales” or gallery forests dominated by the palm Mauritia flexuosa. In the background, two other table mountains are present: the Angasima-tepui (left), at 2250 m in altitude, and the extensive Chimantá massif (right), composed of several tepuian summits between 2200 and 2700 m in altitude, attaining a total surface of more than 600 km2. Photograph by the author.

group (1400–2300 million years ago). These sedimentary rocks began to be denuded by the incipient fluvial systems, in a process that is still ongoing. Weathering and erosion proceeded more easily on anticlines, where water penetration has been favored by an open fracture system. Synclines, however, are more resistant to erosion, and several of them have persisted as isolated topographical remnants: the tepuis. More than 50 tepuis and tepuian massifs have been recognized as such, most of them in Venezuela—where they attain their maximum development—with a few representatives in Guyana and Brazil. The tepui summits are variable in both altitude and surface area, ranging from 1100 to 3014 m (typically 2000–2600 m) in maximum altitude and less than 1 to more than 1000 km2 (typically 200–500 km2) in area. The total Pantepui surface is ∼5000 km2 in area. The degree of physical isolation of these summits is also variable. The surrounding lowlands are situated between 100 and 1200 m elevation (commonly 100 to 400 m), and the vertical difference between them and the tepui summits ranges from 200 to 2400 m (usually 800 to 1800 m). Despite these numbers and the visual impression of the tepuian landscape, only ∼20% of the tepui summits are really isolated topographically; the others are connected to the lowlands by extensive river valleys, ridges, or eroded walls.

total annual rainfall), which allows development of dense vegetation types, most of them unique and characteristic of Pantepui. Among forests, the more emblematic are the dense high-tepui forests dominated by Bonnetia (Theaceae), associated with diabase intrusions and watercourses. The more characteristic shrublands, exclusive of one single tepuian massif (the Chimantá), are organized around a few species of the endemic Asteraceae genus Chimantaea (Figs. 4–6). The typical tepui meadows are dominated by broad-leaved plants without gramineous morphology, such as Stegolepis (Rapateaceae) and Xyris (Xyridaceae). Grasses and sedges are minor elements atop the tepuis. Characteristic pioneer communities with cyanobacteria, fungi, and incrustant lichens grow on bare rock. Vascular plants, the best-known organisms of Pantepui, are commonly used to illustrate the biodiversity and endemism patterns of the tepui summits. So far, around 2500 species (630 genera and ∼160 families) are known, of which 62% are endemic to the Guayana region, 42% are endemic to Pantepui, and 25% are local endemics (i.e., endemic to a single tepui or tepuian massif ). Local endemism can reach up to 60% in some tepuis, which is comparable to the most isolated oceanic islands. There are 23 endemic genera (∼4%) but not any endemic family. Most of the Pantepui vascular plant genera are of neotropical distribution (∼70%), followed by paleotropical (∼20%), cosmopolitan (∼5%), and temperate (∼5%) elements. Among neotropical affinities, 25% of the genera correspond to Guayana endemics, and 6% are shared with the tropical Andes; the rest are more widespread. Among the animals, the most studied are birds, followed by frogs and reptiles, and then mammals. Around 100 species of birds have been described in Pantepui, of

FIGURE 4 Inflorescences of Stegolepis ligulata (Rapateaceae) from

BIOTA

The climate atop the tepuis is mild (10–18 ºC average annual temperature) and very humid (2500–3500 mm of 718

PA N T E P U I

the summit of the Apakará-tepui (Chimantá massif), around 2200 m in altitude. This species is endemic to the summits of the Chimantá massif, where it dominates the broad-leaved meadows. Photograph by the author.

FIGURE 5 Chimantaea mirabilis (Asteraceae) from the Apakará-tepui,

around 2200 m in altitude. This species dominates the unique and characteristic “paramoid” shrublands of the Chimantá massif, to which it is endemic. Photograph by the author.

which one-third are endemic to Guayana and 10–20% are endemic to the highlands. The diversity of herpetofauna (frogs and reptiles) is about half that of birds, but the level of endemism is higher, reaching 60% in some tepuis. Mammals, most of them small, are also represented by even lower number of species, mainly of bats, rodents, and some marsupials (opossums). So far, felids, monkeys, and medium to large herbivores have not been observed atop the tepuis. Insects have not been studied systematically, but recently, a new genus of damselflies was described (Tepuibasis), with all its seven species being endemic to Pantepui. The origin of such biotic features has been debated for long time. Based on floristic data, earlier researchers (up to about 1970) explained the uniqueness of the Pantepui biota as the result of evolution in isolation since the Cretaceous. According to this hypothesis, present species would

FIGURE 6 The insectivorous plant Heliamphora minor (Sarrace-

be very old in origin. However, ecological and paleoecological evidence favoring genetic interchange among summit biotas is increasing. The proposed mechanism is related to the Quaternary (the last 2.6 million years) glaciations. Glacial cooling promoted downward migration and lowland spreading of summit taxa, whereas interglacial warming favored upward migration and colonization of new summits. Molecular phylogenetic studies on key taxa, such as Stegolepis and Myoborus, a genus of redstarts, agree with this view and favor a recent origin, probably Plio-Pleistocene (the last 5 million years), for their species. Such evolutionary processes would have promoted adaptative radiation, favored by the elevated habitat heterogeneity and ecological diversity of the tepui summits. A good example can be found in Brocchinia (Bromeliaceae), a genus of both lowlands and highlands with known morphological and functional adaptations. The issue of the origin of the Pantepui biota is a fascinating subject, which is still open to new ideas and further research efforts. CONSERVATION

Pantepui is still virtually pristine. Indigenous people do not visit the tepui summits because they consider them the homeland of gods and therefore to be sacred places. Activities such as cultivation, lodging, burning, mining, tourism, and so forth are prevented by several protection efforts, including national parks, natural monuments, biosphere reserves, and a World Heritage Site. In addition, Pantepui has been considered by the WWF/IUCN as one of the neotropical plant diversity centers (SA-2), as well as a critical ecoregion (ER-45) of the Global 2000 Project. Therefore, the tepui summits seem to be well protected against direct human intervention. However, the potential consequences of the ongoing and predicted future global warming have not been fully realized until very recently. Increasing temperatures will cause an upward displacement of suitable environmental conditions for mountain species, such that a number of them may lose their habitat. In the tepuis’ summits, this effect will be enhanced by the flat topography, which prevents further upward displacement. Preliminary estimations show that roughly 200–400 endemic vascular plant species (∼30– 50% of the total) of Pantepui are threatened with habitat loss because of the projected 2–4 ºC warming for the end of this century. Owing to the singularity of the Pantepui biota, this would be a serious danger for Guayanan, as well as for global, biodiversity. SEE ALSO THE FOLLOWING ARTICLES

niaceae) from the summit of the Eruoda-tepui (Chimantá massif), at about 2600 m in altitude. Photograph by the author.

Adaptive Radiation / Continental Islands / Global Warming

PA N T E P U I

719

FURTHER READING

Berry, P. E., and R. Riina. 2005. Insights into the diversity of the Pantepui flora and the biogeographic complexity of the Guayana shield. Biologiske Skrifter 55: 145–167. Briceño, H. O., and C. Schubert. 1990. Geomorphology of the Gran Sabana, Guayana shield, southeastern Venezuela. Geomorphology 3: 125–141. Huber, O., ed. 1992. El macizo del Chimantá. Caracas: Oscar Todtmann Ed. Huber, O. 1988. Guayana lowlands versus Guayana highlands: a reappraisal. Taxon 37: 595–614. Maguire, B. 1970. On the flora of the Guayana highland. Biotropica 2: 85–100. Rull, V. 2004. Biogeography of the “Lost World”: a palaeoecological perspective. Earth-Science Reviews 67: 125–137. Rull, V., and T. Vegas-Vilarrúbia. 2006. Unexpected biodiversity loss under global warming in the neotropical Guayana highlands. Global Change Biology 12: 1–9. Steyermark, J. A. 1986. Speciation and endemism in the flora of the Venezuelan tepuis, in High-altitude tropical biogeography. F. Vuilleumier and M. Monasterio, eds. Oxford: Oxford University Press, 317–373. Steyermark, J. A., P. E. Berry, and B. K. Holst, eds. 1995–2005. Flora of the Venezuelan Guayana. St. Louis: Missouri Botanical Garden Press.

Polynesia. Although these geographic terms continue to be widely used, except for “Polynesia,” they have little cultural or historical basis. Only Polynesia stands out as a culturally and linguistically meaningful category. More recently, historical anthropologists and archaeologists stress the distinction between Near Oceania in the western Pacific (including New Guinea, the Bismarck Archipelago, and the Solomon Islands) and Remote Oceania (which includes all of island Melanesia southeast of the Solomons, along with Polynesia and Micronesia). Near Oceania, which was first settled by Homo sapiens in the late Pleistocene, is characterized by intervisible islands with a highly diverse biota, capable of supporting hunterand-gatherer populations. The widely dispersed islands of Remote Oceania were discovered and settled only within the past 4000 years, by horticultural peoples who introduced food crops and domestic animals to these biotically more depauperate and resource-limited islands. PLEISTOCENE SETTLEMENT OF NEAR OCEANIA

PAPUA NEW GUINEA SEE NEW GUINEA

PEOPLING THE PACIFIC PATRICK V. KIRCH University of California, Berkeley

The islands of the Pacific Ocean were settled by humans in two major episodes. The earliest phase began in the Late Pleistocene, at least 40,000 years ago, and involved the movement of hunting-and-gathering populations into Near Oceania. The second major phase commenced about 4000 years ago and involved the diaspora of the Austronesian-language speakers into Remote Oceania, as well as into the Indian Ocean as far as Madagascar. The most isolated islands and archipelagoes of Remote Oceania, including Hawai‘i, Easter Island (Rapa Nui), and New Zealand (Aotearoa), were settled by Polynesians between AD 800 and 1200. GEOGRAPHIC BACKGROUND: NEAR AND REMOTE OCEANIA

The Pacific Islands, or Oceania, were classically subdivided into three main geographic regions, following the scheme of the French explorer Dumont d’Urville in the early nineteenth century: Melanesia, Micronesia, and

720

P E O P L I N G T H E PAC I F I C

During periods of glaciation in the Late Pleistocene, lowered sea levels exposed the continental shelf joining New Guinea to Australia (and Tasmania to Australia in the south). This enlarged land mass is known to biogeographers as Sahul. Similarly, the Malaysian peninsula and much of Indonesia formed another exposed land mass called Sunda. The region between Sunda and Sahul, known as Wallacea, always had water gaps that acted as barriers to biotic dispersal. However, human entry into Sahul was facilitated when these water gaps were at their narrowest, and human expansion throughout Australasia occurred rapidly once people entered the region, at least 40,000 years ago. A number of occupation sites are now radiocarbon-dated to ∼36,000 years ago, on the large island of New Guinea, and on New Britain and New Ireland in the adjacent Bismarck Archipelago. Late Pleistocene sites in the Admiralty Islands, New Ireland, and Buka (Solomons) all indicate the existence of open-ocean transport, thereby suggesting the presence of some form of early watercraft (possibly rafts, bark boats, or dugouts). There is no evidence, however, for human expansion beyond the eastern end of the main Solomon Islands chain until the Middle to Late Holocene. The earliest human colonists in Near Oceania were hunters and gatherers, who exploited tropical rain forests but also made use of inshore marine resources. The presence of simple flake tools of obsidian (originating on the island of New Britain) at these sites provides evidence for long-distance communication and exchange between com-

munities, often on separate islands. By the early Holocene (∼8000 BC), archeobotanical evidence indicates that tree, root, and tuber crops (such as the Canarium almond and various aroids) were being domesticated within Near Oceania. The Kuk Swamp site in the highlands of New Guinea has produced structural evidence in the form of drains and ditches for water control and possible taro horticulture beginning as early as 7000 BC. Such archeobotanical evidence supports long-standing ethnobotanical claims that Near Oceania was an important region for the domestication of a number of tropical root, tuber, and tree crops. Historical linguistic evidence suggests that the Pleistocene settlers of the Near Oceanic islands spoke a diversity of languages, all of which were non-Austronesian. These highly diverse languages (of which at least 900 are recorded) are often lumped together under the rubric Papuan, but in fact a number of distinct language families are involved. AUSTRONESIAN ORIGINS

Beginning around 3000 BC, a major diaspora of peoples speaking languages belonging to the Austronesian language family began to expand from island Southeast Asia into the Near Oceanic region. The immediate homeland of the Austronesians has generally been regarded as including the island of Taiwan, as well as adjacent areas of mainland China (e.g., the Fujian coast). Linguistic reconstructions of Proto-Austronesian language include a number of terms for the outrigger sailing canoe and its components, and the ability of early Austronesians to disperse rapidly is doubtless due to their invention of this technology. Linguistic evidence likewise indicates that the Austronesians were horticulturalists who transported root, tuber, and tree crops including taro (Colocasia esculenta), the true yams (Dioscorea alata and other species), coconut, and many other crop species via their canoes. They also raised domestic pigs, dogs, and chickens. Other animals, such as the Pacific rat (Rattus exulans), may have been carried as stowaways, although possibly also for food. Archaeologically, the Austronesian dispersal from Taiwan through the Philippines and onto other archipelagoes and islands of Southeast Asia and Oceania is marked by the presence of a variety of ceramic assemblages, along with related material culture including stone adzes and shell technology (fishhooks, ornaments, etc.). The oldest pottery associated with Austronesians is the Ta-p‘en-k‘eng culture of Taiwan (∼4300–2500 BC), which is followed by several local varieties of red-slipped pottery. The dispersal of Austronesian speakers into coastal regions of Near Oceania and on into the islands

of Remote Oceania is marked by sites of the Lapita cultural complex (see below). The Austronesian diaspora rapidly encompassed the major archipelagoes of island Southeast Asia; one branch of Austronesian-speakers expanded along the north coast of New Guinea into the Bismarck Archipelago. This branch is known to linguists as Oceanic, and the Oceanic languages (numbering about 450 modern languages) include most of those spoken throughout the Pacific Islands. THE LAPITA CULTURAL COMPLEX AND REMOTE OCEANIA

The movement of Austronesian speakers into the Bismarck Archipelago has been correlated with the presence of a distinctive style of pottery known as Lapita. Early Lapita pottery assemblages include vessels of various shapes, such as bowls supported with pedestal feet, flatbottomed dishes, large carinated jars, and globular jars with restricted necks. Many of these vessels were covered with lime-infilled decorations, with motifs made largely through a technique of dentate stamping, although incising was also used. Some of the pottery, such as the globular jars, was not decorated. Aside from the characteristic pottery, Lapita sites yield a variety of other portable artifacts, such as Tridacna-shell adzes and Trochus-shell fishhooks, as well as a diversity of ornaments and exchange valuables. Obsidian from sources in the Bismarck and Admiralty Islands is also common at Lapita sites and is further evidence of extensive trade or exchange between Lapita communities. The earliest phase of the Lapita cultural complex dates to ∼1500–1300 BC and is represented by sites in the Bismarck Archipelago, such as the waterlogged Eloaua Island sites of the Mussau group. Often, these early Lapita sites were hamlets or villages consisting of houses elevated on posts or stilts, situated over tidal reef flats or along shorelines (such as the large Talepakemalai site on Eloaua Island). Excavated plant and animal remains indicate a mixed economy with horticulture and marine exploitation. It is very likely that the makers of Lapita pottery can be correlated with the Proto-Oceanic stage in the history of the Austronesian language family dispersal. Similarly, genetic evidence (such as mitochondrial DNA and hemoglobin markers) suggests that the Lapita population derived from an actual demic intrusion into the Bismarck Archipelago, deriving out of island Southeast Asia. However, it is also likely that the Proto-Oceanic speakers interacted extensively with the indigenous Papuanspeaking populations who were already in place in Near

P E O P L I N G T H E PAC I F I C

721

Archeological evidence for initial Austronesian dispersal into Micronesia comes from the western Micronesian archipelagoes of Palau and the Marianas. In these islands, the earliest sites contain a red-slipped, sand-tempered pottery, some of which is decorated with lime-filled, impressed designs. Radiocarbon dates from these sites suggest that humans settled the Mariana and Palau archipelagoes no later than 1500 BC, and possibly as early as 2000 BC; the immediate homeland of these voyagers may have been in the Philippines or Molucca Islands. Around 2000 years ago, Oceanic speakers who made plainware pottery (a late form of Lapita) and who used shell adzes, fishhooks, and other implements moved northward into central Micronesia and founded settlements on several volcanic islands, including Chuuk, Pohnpei, and Kosrae. The atolls of the Marshall Islands were also colonized around this time. This two-phase settlement of Micronesia, with initial Austronesians moving in from the Philippines ∼2000–1500 BC, followed by later movement of late Lapita (Oceanic-speaking) colonists into central Micronesia, is reinforced by historical linguistic evidence. The Austronesian languages spoken in western Micronesia are distinct from those of central and eastern Micronesia, with the latter belonging to the Oceanic subgroup. However, later contact between island groups often resulted in considerable linguistic borrowing; the Yapese languages, for example, had several phases of interaction with both western and central Micronesian languages.

referred to as Western Polynesia, is considered the immediate homeland of Polynesian culture and language. A number of changes were involved in the transition from late eastern Lapita to Ancestral Polynesian culture, among them the gradual loss of the art of making pottery. A significant number of lexical changes also occurred in the language spoken by the dialect chain linking Tonga and Samoa, resulting in a distinctive Proto-Polynesian language. The final stage in the human settlement of the Pacific Islands began in the first millennium AD, with the additional expansion of Polynesian-speaking peoples eastward out of Tonga and Samoa into the archipelagoes of central Eastern Polynesia. This last great phase in the expansion of Austronesians into the Pacific was enabled by the invention and development, probably in the Tonga-Samoa region, of the double-hulled voyaging canoe. These large canoes were capable of carrying 40 to 60 people and their cargo (including plants and domestic animals to reestablish their subsistence economy on new islands), for voyages lasting a month or more. There has been considerable debate among archaeologists regarding the exact chronology and sequence of the Eastern Polynesian dispersals. Recent radiocarbon dating of sites in the Society Islands, the Cook Islands, the Marquesas, and Mangareva suggest that initial settlements were in place by around AD 900, and possibly slightly earlier in the case of the Society Islands. Remote Easter Island is likely to have been discovered by AD 900–1000. Evidence from pollen cores on O‘ahu Island likewise suggests that Polynesians were clearing lowland forests there by AD 800–1000. The large, temperate islands of New Zealand were among the last to be colonized by Polynesians, after AD 1200. As noted earlier, Polynesia is the only term of the original tripartite classification of Oceanic peoples (including Melanesia and Micronesia) that can properly be considered a monophyletic cultural and linguistic group. All of the ethnographically attested Polynesian societies can be demonstrated to have descended from a common Ancestral Polynesian culture and Proto-Polynesian language. For this reason, Polynesia is regarded as an ideal region for testing phylogenetic models of cultural differentiation from a common ancestor.

POLYNESIAN DISPERSALS

CONTACTS WITH THE AMERICAS

Ancestral Polynesian culture and Proto-Polynesian language developed in the Tonga–Samoa region between ∼900 and 500 BC, directly out of the founding Lapita cultural complex. Thus, the Tonga–Samoa region,

The question of contacts between Polynesia and the Americas has long interested scholars. The older view of Thor Heyerdahl that Polynesian populations originated in North and South America has not stood the test of

Oceania. For these reasons, the Lapita phenomenon is now interpreted as the outgrowth of simultaneous cultural processes of intrusion, integration, and innovation (the so-called Triple-I model). After the initial phase of Lapita development in Near Oceania, Lapita populations expanded eastward into Remote Oceania, first to the far eastern Solomon Islands, by at least ∼1300 BC, and soon thereafter into Vanuatu, the Loyalty Islands, and New Caledonia. The 850-km water gap between the Eastern Solomons and Fiji may have been crossed by 1100 BC. The archipelagoes of Tonga and Samoa, along with the nearby islands of Futuna and ‘Uvea, had Lapita settlements in place by 900 BC. SETTLEMENT OF MICRONESIA

722

P E O P L I N G T H E PAC I F I C

archaeological, linguistic, or human biological evidence. However, it is certain that later Polynesian voyaging canoes had the technical capability to reach the Americas, and to return. That the Polynesians did indeed reach South America and return is strongly suggested by preserved remains of the sweet potato (Ipomoea batatas), a South American domesticate, recovered from several prehistoric Polynesian sites (such as the Tangatatau rockshelter on Mangaia, Cook Islands). Further evidence of contact has recently been provided by the bones of chickens (Gallus gallus, a Polynesian domesticate) from a pre-Columbian archeological site in Chile. More controversial has been the claim that Polynesians may have introduced the technology of plank-built canoes to the Chumash of the Channel Islands, off the coast of California. SEE ALSO THE FOLLOWING ARTICLES

Archaeology / Cook Islands / Easter Island / Exploration and Discovery / Human Impacts, Pre-European / Kon-Tiki / Polynesian Voyaging / Tonga FURTHER READING

Bellwood, P. 1985. Prehistory of the Indo-Malaysian archipelago, revised ed. Honolulu: University of Hawaii Press. Irwin, G. 1992. The prehistoric exploration and colonisation of the Pacific. Cambridge: Cambridge University Press. Kirch, P. V. 2000. On the road of the winds: an archaeological history of the Pacific Islands before European contact. Berkeley: University of California Press. Kirch, P. V., and R. C. Green. 2001. Hawaiki, Ancestral Polynesia: an essay in historical anthropology. Cambridge: Cambridge University Press. Lilley, I., ed. 2006. Archaeology of Oceania: Australia and the Pacific Islands. Oxford: Blackwell Publishing. Spriggs, M. 1997. The island Melanesians. Oxford: Blackwell Publishing.

PHILIPPINES, BIOLOGY RAFE M. BROWN University of Kansas, Lawrence

ARVIN C. DIESMOS National Museum of the Philippines, Manila

The Philippines (Fig. 1) is one of the Earth’s most spectacular island archipelagoes. The country spans the Asian– Australian faunal zone interface at the sharpest biotic demarcation (Wallace’s Line) on the planet. Although collectively comprising a land mass approximately the size of the U.S. state of Arizona, the Philippines is a complex archipelago with more than 7100 distinct islands. Geographically situated on the edge of multiple colliding tectonic plates, the Philippines has an ancient and complex geological history that has only recently come to light. Ancient land mass movements, environmental gradients along steep volcanic slopes, and sea level–induced alterations of connectivity between neighboring islands have all presumably fueled the in situ evolutionary process of diversification on a magnitude seen in few other island systems. The result is a spectacular array of biodiversity, unparalleled levels of vertebrate endemism, and some of the planet’s most spectacular examples of life on islands. The Philippines is a global superpower of biodiversity that may possess one of the highest concentrations of vertebrate life on Earth. FIGURE 1 The position of the

Philippines (darkly shaded islands) in relation to surrounding

Southeast

Asian

and

Australasian land masses. The positions of Wallace’s and Huxley’s lines are indicated for reference. Wallace’s line marks the edge of the Sunda Shelf

and

the

transition

between Asian and Australian faunal regions. Huxley’s line separates Palawan and associated land bridge islands from the truly oceanic portions of the Philippines.

PHILIPPINES, BIOLOGY

723

FIGURE 2 Species exemplars from the extraordinarily diverse endemic

Philippine rodent radiation: (A) Luzon hairy-tailed rat, Batomys granti; (B) Cordillera striped earth-mouse, Chrotomys whiteheadi; (C) Luzon needle-nosed shrew-rat, Rynchomys cf soricoides; (D) Kalinga shrewmouse, Archboldomys kalinga; (E) silver earth-mouse, Crotomys silaceus; and (F) common Philippine forest rat, Rattus everetti. Photographs by Rafe M. Brown.

GEOGRAPHICAL AND GEOLOGICAL SETTING

The Philippines is situated south of Taiwan and north of Borneo and Sulawesi (between the equator and the Tropic of Cancer), separated from the Asian mainland by the South China Sea. Collectively the archipelago covers 2 million km2 with a total land mass of approximately 300,000 km2. The western part of the archipelago (Palawan, Balabac, Busuanga, Coron, and smaller associated islands) has had a stable geologic configuration and has been intermittently connected (or nearly connected) to northern Borneo via a narrow land bridge during maximum exposure of the world’s largest continental shelf (the Sunda Shelf ). The remaining portions of the Philippines are classified as oceanic islands and are believed to have risen directly from the ocean floor as a result of volcanic activity associated with subduction of the Philippine Plate at the western boundary of the Pacific Ring of Fire. BIODIVERSITY AND ENDEMISM

The Philippines is recognized internationally as a global stronghold of biodiversity. The shallow, warm seas sur-

724

PHILIPPINES, BIOLOGY

rounding the country support the richest coral reef communities on the planet and are literally the epicenter of Southeast Asian and southwestern Pacific marine diversity. Terrestrial ecosystems in the Philippines are similarly diverse, supporting a wealth of natural resources, habitat heterogeneity, and a rich array of species diversity. High levels of alpha-diversity (species richness) and beta-diversity (variation among adjacent regions) place the Philippines among the world’s most biodiversity-rich countries. Some of this diversity is truly astonishing. Highlights include dwarf water buffalos or “Tamaraw” (Bubalus mindorensis), the largest extant forest eagle in the world (Pithecophaga jefferyi), a radiation of the world’s largest flowers (genus Rafflesia), the rarest batagurid turtle in the world (Siebenrockiella leytensis), an extensive and exceptionally diverse rodent radiation (including such novelties as raccoon-sized “cloud rats” and needle-nosed shrew-rats that feed entirely on worms; Fig. 2), four endemic genera of snakes, five endemic species of spectacularly maned wild pigs, and such “living fossils” as the primitive and relictual flat-headed frog (Barbourula busuangenis). The fossil record shows that many extraordinary, large-bodied animals have gone extinct with the arrival of modern humans, including dwarf buffalos, elephants, and giant land tortoises. Terrestrial ecosystems are fantastically diverse. Recent summaries of birds recognized 593 species (32% endemic); mammal diversity currently is estimated at 175 native terrestrial mammals (65% endemic). Recent reviews of the classification of Philippine amphibians and reptiles recognized 105 species of amphibians (79% endemic) and 264 reptiles (68% endemic). These estimates emphasize conspicuous terrestrial vertebrates, but total country estimates (Table 1) are awe inspiring, with as many as 15,000 plants (and their relatives) and 38,000+ animals (vertebrates and invertebrates), for a startling total of 53,500 species. These numbers should be viewed as conservative approximations; numerous recent studies have shown that terrestrial biodiversity of the Philippines is substantially underestimated, in some cases grossly so. In poorly studied groups such as earthworms, more than a hundred TABLE 1

A Summary of Recent Estimates of Total Countrywide Philippine Species Diversity Taxonomic Group

Philippine Total

Vertebrates Invertebrates Plants Others (algae, lichens, fungi, etc.) Total

3,308–3,325 34,940–35,000 14,000–15,310 6,100+ 53,500+

FIGURE 3 Estimated Cenozoic movements of Philippine land masses

(modified from work of Hall 1998) 10, 20, and 30 million years ago.

new species have recently been discovered. Molecular phylogenetic studies have confirmed the exceptional species diversity of numerous Philippine radiations; many have uncovered suites of previously unrecognized, cryptic species. Several of these key studies of species boundaries have produced startling results and increased species diversity in selected groups by 30–50%. Virtually every molecular phylogenetic study that has been published in the last 10 years has included the discovery of new hidden species, including many groups of frogs, lizards, insects, worms, forest mice, bats, rats, and birds. One example is the case of the Philippine forest frogs of the genus Platymantis. Past studies were based solely on morphology; species numbers grew slowly from 7 to 12 named species between 1950 and 1990. It was at this point that a group of herpetologists began to focus on bioacoustic characters (analysis of the mating calls of male frogs) and applied molecular techniques (DNA sequence data) to the problem of species boundaries in this group. By emphasizing different aspects of the phenotype, these workers provided fine-scale taxonomic partitioning of

FIGURE 4 Spectacular Philippine endemic species: (A) the endemic

Philippine freshwater crocodile, Crocodylus mindorensis (photograph by Rafe M. Brown); (B) the newly discovered Calayan Island flightless rail, Gallirallus calayanensis (photograph by Marge Babon/CEAE); (C) the giant flower Rafflesia manillana (photograph by Arvin Diesmos); (D) and one of the world’s only fruit-eating monitor lizards, Varanus olivaceus (photograph by Charles Linkem).

PHILIPPINES, BIOLOGY

725

forest frogs and opened the way for a flood of discoveries and new species descriptions. It is clear that Platymantis, currently estimated at 27 described species, represents one of the Earth’s major island radiations of frogs. Several dozen species remain to be described. When the same analytical tools are applied to other frog groups throughout the archipelago, it becomes clear that the species diversity of Philippine Amphibia has been underestimated by as much as 30–40%. Despite the lack of a complete knowledge of the biodiversity of the Philippines, today’s taxonomic estimates allow researchers to calculate the number of species per unit area. When this calculation is carried out for terrestrial species, given the available land mass (300,000 km2), the end result is the highest concentration of biodiversity on Earth. BIOGEOGRAPHY AND PROCESSES OF DIVERSIFICATION

FIGURE 5 Spectacular Philippine endemic species: (A) the recently

discovered Mindoro stripe-faced flying fox, Styloctenium mindorensis (photograph by Jake Esselstyn); (B) the rarest batagurid turtle in the world, the Philippine forest turtle, Siebenrockiella leytensis (photograph by Rafe M. Brown); (C) the endemic Philippine dwarf water buffalo or Tamaraw, Bubalus mindorensis (photograph courtesy of Department of the Environment and Natural Resources–Protected Areas and Wildlife Bureau Tamaraw Conservation Program); (D) the brightly colored Igorot frog, Rana igorota (photograph by Rafe M. Brown).

726

PHILIPPINES, BIOLOGY

Several generations of biologists have converged on a startling consensus: The extraordinary biodiversity of the Philippines has likely been produced by three major processes of the geographic template (i.e., geology + climate). We consider these process to be the major “generators” of diversity in the Philippines. The first involves the complex geological history of the archipelago. Briefly, the formation of the Philippines involves more than 50 million years of collisions of multiple separate plates of the earth’s crust, resulting in a history characterized by constantly changing configurations of islands. The Philippines of 10, 20, or 30 million years ago looked almost nothing like the country today (Fig. 3). The inferred result of these ancient movements of land masses is that the distant ancestors of many of today’s Philippine endemic species may have first invaded and become isolated on Philippine paleo-islands several to many tens of millions of years before present. These events appear to have given rise to deep phylogenetic diversity and the presence of numerous “old endemics,” or highly divergent, taxonomically distinctive taxa (Figs. 4 and 5). The second generator of biodiversity in the Philippines is the process of evolutionary differentiation along dramatic elevational, atmospheric, and environmental gradients from sea level to 2000+ meters (Fig. 6). The oceanic portions of the Philippines are islands produced primarily by volcanism. Many of the smaller islands (and many isolated peaks within larger islands) were formed as active volcanoes arose from the ocean floor over the last 100 million years. Steep elevational gradients (replicated

numerous times along dozens of isolated volcanic peaks) have given rise to atmospheric variation, microclimate variability, and forest community heterogeneity—resulting in biodiversity gradients along these sheer mountain slopes. As a result, the study of Philippine biodiversity is largely the study of species succession along environmental gradients. The last 50 years of biodiversity research in the Philippines have documented the enormous impact that elevation-associated atmospheric and habitat gradients that have played a prominent role in the evolutionary processes of diversification. The third generator of Philippine biodiversity (and the one that is, for better or for worse, most often invoked) is the repeated “species pump” action of rising and falling sea levels between the mid- to late Pleistocene (350,000–12,000 years ago). Repeated cooling episodes during this recent time period resulted in the capture of the Earth’s water at the expanding polar ice caps and a concomitant lowering of sea levels throughout the globe. In the Philippines the result of these sea level oscillations was repeated episodes of connectivity and isolation between islands separated by channels of 120–180 meters (Fig. 7). Biogeographers now recognize at least eight faunal provinces that correspond to these Pleistocene Aggregate Island Complexes (PAICs), which represent maximum exposure of expanded islands. Each PAIC is a major center of biodiversity and endemism, with distinctive flora and fauna. The tracing of submarine bathymetric contours to reveal Pleistocene exposure of land provides biologists with an estimate of former paleoisland connection and serves as a basis for predictions of taxonomic affinity (in the absence of phylogenetic data) for the species inhabiting those islands. This exercise has also identified several minor subcenters of biological diversity in the form of small islands. Finally, it is clear that both in situ evolutionary diversification and repeated colonization events have contributed to the accumulation of diversity in the archipelago. The development of molecular phylogenetic methods has allowed for unprecedented study of both biodiversity in general and the historical and temporal framework for FIGURE 6 Striking elevational gradients and volcanic landscapes of

the Philippines: (A) Pagudpud area, northern Luzon Island; (B) Lake Danao, a high-elevation volcanic crater lake in northern Leyte Island; (C) the saw-toothed peaks of Mt. Guiting-guiting, Sibuyan Island; and (D) terraced fields along the slopes of the northern Cordillera Mountain range (between Bontok and Banaue). Photographs by Rafe M. Brown.

PHILIPPINES, BIOLOGY

727

FPO

FIGURE 7 A detailed map of the Philippines (light green shaded islands) with Pleistocene Aggregate Island Complexes (PAICs) indicated by tracing

of 120 m underwater bathymetric contour. Areas shaded purple indicate shallow seas that may have been exposed as many as ten times during the middle- to late Pleistocene. These land-positive connections (land bridges) may have allowed for past floral and faunal exchange between islands that today are isolated by marine channels.

the generation of Philippine biodiversity. Several modern phylogenetic studies have produced surprises that contradict earlier hypotheses and threaten to topple the prevailing view of diversification in the Philippines. The result is an emerging consensus suggesting that Philippine biodiversity is far more complex than previously thought. Recent studies of Philippine flying lizards, forest geckos, spotted stream frogs, and forest mice suggest

728

PHILIPPINES, BIOLOGY

that for some taxa, periodic sea level oscillations have had a profound impact on the production and distribution of species diversity. These studies have all demonstrated a near complete adherence of species to PAIC boundaries and Pleistocene seashores. For much of the terrestrial vertebrate life in the Philippines, these historical events have played a predominant role in shaping species distributions—much more so, in fact, than the

traditional biogeographic variables such as island size and distance from mainland. This dominant paradigm of Philippine biogeography is so prevalent, in fact, that the shared presence of species across PAIC boundaries is now viewed as an extreme outlier. For the most part, this has been a healthy development for the understanding of Philippine biodiversity. By heightened attention to the unique evolutionary histories of PAIC endemics, biogeographers are now beginning to arrive at an appreciation of Philippine biodiversity that is concordant with evolutionary history—one that recognizes distinct and cohesive lineage segments as evolutionary species. The end result is that a truly evolutionary classification of Philippine vertebrate life is now perceived as an obtainable goal. Several recent studies have dramatically upset accepted scenarios concerning the predominant processes of evolutionary diversification in the Philippines. One study, a phylogenetic analysis of Asian “Fanged Frogs” conducted by Ben Evans and colleagues, revealed a pervasive and entirely unexpected zoogeographic link between Sulawesi and the Philippines. In that study, DNA sequence data revealed that multiple over-water dispersal events have allowed for successive waves of exchange between Sulawesi and the Philippines. As might be expected, the oldest invasions gave rise to proportionately more species and have spread across more PAICs than the younger arrivals. This Philippine–Sulawesi connection stands in opposition to all previous zoogeographic evidence provided by the last 25 years of studies of Philippine vertebrates, particularly mammals. Additional studies have upset the land-bridge (Palawan) versus oceanic (remaining Philippines) dichotomy. Phylogenetic studies of fanged frogs, spiny rats, shrews, litter frogs, spotted stream frogs, slender stream frogs, emerald tree skinks, flying lizards, geckos, fresh water fish, river shrimp, and numerous groups of insects all demonstrate that a large portion of Palawan’s endemics are most closely related to the truly oceanic portions of the Philippines, to the exclusion of the (expected) species from the islands of the Sunda Shelf. In fact, many Palawan endemics are nested within the truly Philippine radiation, suggesting recent dispersal to Palawan from the oceanic portions of the Philippines, and a lack of faunal exchange with Borneo. The simple depiction of Palawan as a faunal extension of northern Borneo (based largely on mammal and bird taxonomy) is a taxon-biased expectation that is not supported by the bulk of available phylogenetic evidence. In fact, Palawan is an exceptional amalgamation of greatly divergent ancient taxa, recent dispersal events from Borneo (e.g., mammals), and much older biogeographic elements that are nested within Philippine radiations

and represent old dispersal events from the oceanic portions of the archipelago. In this sense, the slender island of Palawan represents a true biogeographic novelty, literally a crossroads spanning the most prominent biogeographic boundary in the world and one of the exceptions to expectations based on Wallace’s and Huxley’s Lines. An additional class of exceptions to the PAIC-centric tradition of biogeography in the Philippines is the increasingly prevalent pattern of micro-endemism on small land bridge islands. Because small land bridge (or even deep-water) islands next to the larger land masses in the Philippines were expected to possess a nested subset of species diversity observed on large neighboring islands, they have often gone unsurveyed and unscrutinized by biologists. Many of these tiny, seemingly irrelevant islands are now known to harbor endemics, some of which are spectacularly divergent evolutionary novelties. Examples include Lubang, Camiguin Sur, Dinagat, Siargao, Sibuyan, and the Gigante Islands. All demonstrate extensive levels of endemism, commensurate with their status as deepwater minor subcenters of diversity. Recent work suggests that our collective knowledge of these minor subcenters of Philippine biodiversity is far from complete. A final exception to the prevailing paradigm of PAIClevel partitioning of biodiversity of the Philippines is an emerging pattern of autochthonous speciation on the large islands. Numerous recent phylogenetic studies involving amphibians, reptiles, shrews, bats, rats, birds, plants, and insects have drawn attention to patterns of endemism within the large islands of Luzon, Palawan, Mindanao, Negros, Panay, and Samar-Leyte. These studies provide evidence of evolutionary and ecological processes (e.g., other than rising sea level vicariance) fueling diversification. Many of these are very interesting because they provide support for adaptive evolution, sympatric, and/or ecological speciation within larger islands. As such, these examples stand in contrast to the “passive” process of non-adaptive divergence following isolation inferred for the diversification of many other groups of species in the archipelago. HUMAN HISTORY AND IMPACT

Colonizing from mainland Asia, humans first arrived in the Philippines many thousands of years ago. Before the arrival of the Spanish (sixteenth century), the Philippines was 90–95% forested, with only scattered settlements surrounding various coastal areas. Despite a growing population and a great demand for timber, the next 300 years probably resulted in the loss of only an additional 20% of forest cover. We know that at the turn of the twentieth century, Cebu Island (Ferdinand Magellan’s capitol) was

PHILIPPINES, BIOLOGY

729

almost completely deforested and that nearby islands of Negros, Guimaras, and Panay were covered by less than half of their original forest area. But other islands maintained vast expanses of intact forest, and perhaps as much as 70% of the country’s forest persisted. The industrial and agricultural revolution, periods of American and Japanese occupation, plus the period of economic expansion following World War II took an extremely heavy toll on Philippine forests. During this the last century, it is estimated that the Philippines lost between 50% and 70% of its original forest cover. During this post-war period, many factors weighed heavily on Philippine biodiversity, including high human population growth associated with the industrial revolution (and a concomitant application of low-level pressure on forests; Fig. 8), a long tradition of colonial control and exploitation, shortcomings in the country’s (U.S.-installed) education system, crippling poverty in many undeveloped portions of the country, and chronic government graft

and corruption that have pandered to a wealthy elite and foreign hegemony. However, by far the most destructive force has been unchecked environmental destruction associated with largescale logging and mining. Large-scale logging and mining were introduced to the country during the American occupation. Over the last century, the Philippines has been the target of wholesale exploitation of natural resources combined with wide-scale plantation-style agricultural efforts (principally hemp, bananas, copra, and sugar cane), which have converted many of the country’s most fertile natural ecosystems into monoculture and barren wasteland. The end result of this unchecked colonial rule and political instability has been systematic exploitation of Philippine natural resources, and catastrophic environmental destruction at rates exceeding those almost anywhere on the planet. Originally >90% forested, the Philippines now retain only 4–8% of its original old-growth forest. CONSERVATION PROSPECTS AND CHALLENGES FOR THE FUTURE

FIGURE

8 Primary

causes of deforestation in the Philippines:

(A) Large-scale commercial logging was introduced to the Philippines by the Americans. In this image, loggers use refurbished World War II weapons carriers to skid mature hardwood trunks from lowland forest of Mt. Busa, South Cotobato Province, southern Mindanao Island. (B) Small-scale timber poaching and clearing of land for agricultural purposes. Kaingineros like this man from Mt. Malinao, Albay Province, Luzon Island, routinely clear small patches of forest, burn debris, and then plant a single crop of rapidly-growing vegetables for sustenance and sale at local markets. Photographs by Rafe M. Brown.

730

PHILIPPINES, BIOLOGY

The Philippines shares only with Madagascar the distinction of being both a megadiverse country and a Global Conservation Hotspot. Despite a greater understanding of Philippine biodiversity gained during the past 15 years, such knowledge is accumulating too slowly to conserve and stem the loss of the archipelago’s spectacular biodiversity. There can be no doubt: the Philippines is of the planet’s highest conservation priorities. Part of the roots of the global conservation crisis currently unfolding in the Philippines is related to the so-called Linnaean shortfall: the disparity between where taxonomists have focused their attention over the last 400 years (e.g., predominantly vertebrates, selected insect groups, and angiosperms) and where the greatest proportions of undescrbed biodiversity remain to be studied (invertebrates, other plants, parasites, and soil microbes, among many other groups). The Linnaean shortfall represents a failure of taxonomy and at the same time a tremendous opportunity for future generations of researchers who choose to focus their attention on these lesser-known groups. Nevertheless, because a lack of knowledge of biodiversity in part contributes to its wanton exploitation and neglect, the end result of this ignorance is biodiversity loss and extinction. Despite these bleak assessments, there is increasing cause for hope and renewed efforts towards fostering sustainable development, ecosystem restoration, and reliance on renewable natural resources (e.g., nontimber forest products such as rattan). The National Integrated Protected Areas System (NIPAS) Act has established a protocol for the establish-

ment and support of nationally protected areas, and establishment of new parks continues to this day at a growing rate. Grassroots community environmentalism has sprung up throughout the country, and a growing protectedarea system continues to spread coverage and jurisdiction over the remaining forested regions. The country’s National Commission on Indigenous Peoples has taken a strong

leadership role in protecting natural resources through management of tribal ancestral homelands, and an expanding base of young, energetic conservation biologists have enthusiastically taken up the battle to protect the country’s natural heritage. There is great cause for hope in Philippine conservation, and unique Philippine flagship species (Fig. 9) play an increasingly important role in spreading environmental awareness to the Filipino public. An understanding of Philippine biodiversity will be greatly enhanced by renewed, vigorous attention to four primary avenues of conservation-related research. First and foremost, these include a need for large-scale, countrywide, and faunistically comprehensive surveys of the remaining natural habitats of the Philippines. Great progress has been made by a handful of researchers over the past 20 years, but this progress needs to be increased by an order of magnitude in both scope and urgency. The next several decades must see a massive resurgence in biodiversity studies if the Philippine conservation crisis (and expected catastrophic extinction event) is to be averted. Second, thorough taxonomic revisionary studies will be required to avert the Linnaean shortfall in the Philippines and promote biodiversity conservation as a global priority. This field of study is surprisingly thankless and increasingly threatened by changing academic landscapes and emphasis on “high-impact” publications in science. Third, the future is very bright for molecular phylogenetic studies of endemic Philippine radiations. Many of the truly spectacular Philippine radiations have gone unstudied. A much-needed synthesis of geological evidence and time-calibrated phylogenetic studies shows tremendous promise for exposing the temporal framework for evolutionary diversification of Philippine biodiversity. Finally, for a variety of reasons, it is clear that the future of biodiversity research and conservation in the Philippines is an effort that must be led by Filipinos. For this to occur, public and government perception of biologists, societal values, and educational emphasis must all shift to endorse the preservation of natural resources and environmental quality. Aside from their inherent value, biodiversity, forested ecosystems, and environmental health all provide societal services such as clean water, food, renewable resources, and buffering from inclement weather. Filipinos have a rich cultural and historical legacy that is tightly linked to their natural heritage through these natural resources. These must all be celebrated and preserved for future generations.

FIGURE 9 Flagship species of Philippine biodiversity conservation

and symbols of an emerging surge in environmentalism: (A) the

SEE ALSO THE FOLLOWING ARTICLES

poorly-known Philippine tarsier, Tarsius syrichta; (B) the largest eagle in the world, the Philippine “monkey-eating” eagle, Pithecophaga jefferyi. Photographs by Rafe M. Brown.

Borneo / Frogs / Indonesia, Biology / Madagascar / Philippines, Geology / Sustainability / Wallace’s Line

PHILIPPINES, BIOLOGY

731

FURTHER READING

Brown, R. M. 2007. Introduction, in Systematics and zoogeography of Philippine Amphibia. R. F. Inger. Kota Kinabalu, Malaysia: Natural History Publications, 1–17. Brown, R. M., A. C. Diesmos, and A. C. Alcala. 2002. The state of Philippine herpetology and the challenges for the next decade. The Silliman Journal 42: 18–87. Catibog-Sinha, C. S., and L. R. Heaney. 2006. Philippine biodiversity: principles and practice. Quezon City, Philippines: Haribon Foundation for Conservation of Natural Resources. Collar, N. J., N. A. D. Mallari, and B. Tabaranza, Jr. 1999. Threatened birds of the Philippines. Makati City, Philippines: Bookmark, Inc. Department of Environmental Resources and United Nations Environment Programme. 1997. Philippine biodiversity: as assessment and action plan. Makati City, Philippines: Bookmark, Inc. Diesmos, A. C., R. M. Brown, A. C. Alcala, R. V. Sison, L. E. Afuang, and G. V. A. Gee. 2002. Philippine amphibians and reptiles, in Philippine Biodiversity Conservation Priorities: a Second Iteration of the National Biodiversity Strategy and Action Plan. P. S. Ong, L. E. Afuang, and R. G. Rosell-Ambal, eds. Quezon City, Philippines: Department of the Environment and Natural Resources, 26–44. Environmental Center of the Philippines Foundation. 1998. Environment and natural resources atlas of the Philippines. Manila, Philippines: ECRF. Hall, R. 1998. The plate tectonics of Cenozoic SE Asia and the distribution of land and sea, in Biogeography and geological evolution of southeast Asia. R. Hall and J. D. Holloway, eds. Leiden: Brackhuys, 99–132. Mallari, N. A. D., B. R. Tabaranza, Jr., and M. J. Crosby. 2001. Key conservation sites in the Philippines. Makati City, Philippines: Bookmark, Inc. Posa, M. R. C., A. C. Diesmos, N. S. Sodhi, and T. M. Brooks. 2008. Hope for threatened tropical biodiversity: lessons from the Philippines. BioScience 58: 231–240.

PHILIPPINES, GEOLOGY GRACIANO YUMUL, JR., AND CARLA DIMALANTA University of the Philippines, Quezon City

KARLO QUEAÑO Department of Environment and Natural Resources, Quezon City, Philippines

EDANJARLO MARQUEZ University of the Philippines, Manila

Island arc systems such as the Philippines are produced through accretion brought about by collision of geologic blocks, resulting in volcanism and emplacement of crust and mantle fragments on land. The various igneous, sedimentary, and metamorphic rock types in island arcs reflect the complex processes involved in their generation and evolution. Island arc-related processes involve interactions of geological features (e.g., trenches, volcanoes, and faults) that result in specific tectonic evolution, geologic hazards, and mineral deposits.

732

PHILIPPINES, GEOLOGY

GEOLOGIC AND TECTONIC SETTING OF AN ISLAND ARC SYSTEM

The Philippines, an arc system lying off the Asian land mass, is trapped at the margins of the Eurasian–Sundaland and the Philippine Sea Plates. The northwest-southeast oblique convergence between these plates is currently being absorbed by two oppositely dipping subduction zones, namely (1) the Manila–Negros–Sulu–Cotobato trench system along which the marginal basins (i.e., South China Sea, Sulu Sea, and the Celebes Sea) on the eastern edge of the Eurasian–Sundaland Plate are being subducted west of the Philippine arc, and (2) the East Luzon Trough–Philippine Trench system, along which the West Philippine Basin of the Philippine Sea Plate is being subducted east of the arc (Fig. 1). The subduction zones extend approximately 1500 km, delineating an approximately 400-km-wide, seismically active deformed zone known as the Philippine Mobile Belt within the archipelago. Intense deformation also affects different parts of the Philippine Mobile Belt with the activity of the Philippine Fault Zone (Fig. 1). This major left-lateral strike-slip fault system transects the archipelago for more than 1200 km, from northwestern Luzon to eastern Mindanao. It has both transpressional and transtensional components and both horizontal and vertical displacements. Several major strike-slip faults branch out from the main trace of this major fault system. These faults and fault segments can be recognized on the basis of displacements in exposed rock sequences, fault scarps, sag ponds, and pressure ridges. Linear features on aerial photographs, satellite, and remote sensing images also serve to define these faults. The Philippine Fault Zone, which formed during the Middle Miocene, accommodates the lateral component corresponding to the excess stress resulting from the oblique convergence between the Philippine Sea Plate and the Philippine Mobile Belt through shear partitioning mechanism. Data obtained from the Global Positioning System networks within the Southeast Asian region have provided measurements of the convergence rate between the Sundaland–Eurasian margin and the Philippine Sea Plate. The Sunda block to the west of the Philippine archipelago is moving with respect to Eurasia at around 10 mm/year in the direction 78° east of south along its northern margin and ∼6 mm/year toward 61° east of south along its southern portion. On the eastern side, the Philippine Sea Plate is moving northwestward at approximately 7 cm/ year in the region northeast of Luzon and around 9 cm/ year southeast of Mindanao (see inset in Fig. 1). The Manila Trench on the western part of the archipelago continues on as collision zones in the central

Legend: Magnetic lineations

20oN

Marginal basin

200 km

Subduction zone Thrust fault

zon Trou g

0

East Lu

N Manila Trench

h

1968

Philippine Sea

1990 South China Sea

e pin ilip Ph

Bicol

Mindoro

lt au eF

Samar

ne Zo

Panay Palawan

10oN EurasianSundaland Plate ~6 mm/yr

120o

Negros

Bohol

Sulu Sea

Philippine Mobile Belt

ch ren lu T

Philippine Sea Plate

Davao

Zamboanga

Su

Co tab

ato

Celebes Sea ~9 cm/yr

Trench

Palawan Microcontinental Block 10o

Cebu

125o

~7 cm/yr 15o

h nc Tre

pin ilip Ph

Philippines (e.g., Mindoro and Panay islands) and connects to the Negros Trench in the Visayas region. It merges with the collision zone in southwestern Mindanao, where fragments of Eurasian–Sundaland affinity have been accreted to the Philippine Mobile Belt (see inset in Fig. 1). These continent-derived fragments, which were rifted from the southern portion of mainland Asia, form part of the Palawan Microcontinental Block, an aseismic block composed of Upper Paleozoic to Mesozoic sedimentary and igneous bodies. A number of marginal basins surround the Philippine archipelago: the South China Sea, Sulu Sea, Celebes Sea, Molucca Sea, and West Philippine Basin (see Fig. 1). The Middle Oligocene to Miocene South China Sea oceanic crust is characterized by nearly east-west-trending magnetic lineations. This suggests a general north-south opening direction for this marginal basin. Several mechanisms that have been proposed to explain the formation of this marginal basin include an Andean-type subduction, extrusion tectonics, and large-scale shear-related strike-slip movements. The Early to Middle Miocene Sulu Sea can be divided into the northwest and southeast subbasins. The northwest subbasin is underlain by an island arc basement, whereas the southeast subbasin is made up of oceanic crust. East-west-trending (relative to present-day geographic location) magnetic lineations recognized in the Sulu Sea indicate a north-south opening. The Sulu Sea basin is postulated to have formed by back-arc spreading. The Celebes Sea basin, on the other hand, represents an oceanic crust of Early to Middle Eocene age that also opened in a north-south direction, based on observed east-west magnetic lineations. The Eocene West Philippine basin is one of the subbasins comprising the Philippine Sea Plate. This oceanic marginal basin formed from spreading along the Central Basin Spreading Center in a northeast-southwest direction (60–45 million years ago), followed by opening in a north-south direction (45–35 million years ago). Models to explain the formation of this large marginal basin include an entrapment model and a back-arc origin model. The Molucca Sea basin consists of oceanic crust that is Eocene in age. The closure of this oceanic basin has been scissor-type, with closure complete in the north and the southern portion undergoing basin closure at present. Subduction along the trench systems surrounding the Philippine archipelago produced the volcanoes and volcanic plugs, which are distributed along linear or arcuate belts. Subduction of the South China Sea crust along the Early Miocene Manila Trench is marked by a volcanic arc that extends from Taiwan to western Luzon. The nearly 1200-km-long chain of stratovolcanoes and volcanic necks

Molucca Sea

120oE

125oE

FIGURE 1 The Philippine archipelago is bounded on both sides by sub-

duction systems: Manila–Negros–Sulu–Cotabato on the west and East Luzon Trough–Philippine Trench on the east. The entire archipelago is bisected by the left-lateral strike-slip Philippine Fault Zone. Marginal basins (e.g., South China Sea, Sulu Sea, Celebes Sea, Molucca Sea, and Philippine Sea) are being consumed along the bounding trench systems (Rangin 1990 and references therein). Red stars show the location of the epicenter (and year when the event happened: 1968 in Quezon; 1990 in Nueva Ecija) of two of the most destructive earthquakes that have been experienced in the Philippine archipelago (data from the Web site of the Philippine Institute of Volcanology and Seismology). The orange triangle shows the location of Guinsaugon, the site of a massive landslide February 17, 2006. (Inset) Continent-derived fragments are juxtaposed against arc rocks as a result of the collision between the Palawan Microcontinental Block and the Philippine Mobile Belt. Arrows indicate the direction of motion of the Sundaland– Eurasian (red arrow) and Philippine Sea Plates (green arrow).

includes the active Pinatubo and Taal volcanoes (Fig. 2). In the Visayas region, the Negros arc was produced by subduction of the Sulu Sea crust along the Middle Miocene Negros Trench. The only active volcano in this nearly north-south-trending linear arc of volcanoes in Negros island is Canlaon volcano (Fig. 2). The Cotabato arc, which includes Parker volcano, was produced by subduction of the Celebes Sea crust along the Cotabato Trench, which began in the Late Miocene. The volcanic arc extending from southeastern Luzon to Leyte in the Visayas region is attributed to the subduction of the Philippine Sea crust along the Philippine Trench. Subduction along this trench began during the Pliocene, approximately 3–5 million years ago. The segment of this arc in southeastern Luzon is made

PHILIPPINES, GEOLOGY

733

120°E

125°E

Ophiolite Igneous rocks Sedimentary rocks Metamorphic rocks Active volcano

Pinatubo

15°N

Taal

Bicol Mayon Mindoro

Romblon Samar Masbate Panay Cebu

eastern and central portions of the country. Ophiolites generated during the Eocene to Oligocene found along the western side of the archipelago have also been mapped (see Fig. 2). The majority of these ophiolite complexes are believed to have formed in subduction-related marginal basins. They are mostly classified as supra–subduction zone ophiolites. The whole rock geochemistry and mineral chemistry of most of these ophiolite complexes manifest island arc signatures. Chert with terrigenous sediments caps the oceanic lithospheric fragments, consistent with generation in land-bounded marginal basins. Thick accumulations of Upper Oligocene to Holocene sedimentary units make up the onshore and offshore sedimentary basins found within the archipelago (Fig. 3). These sedimentary basins consist of shallow to deep

Leyte

Canlaon Palawan

10°N

Bohol

Negros

Surigao

116

120

124

N

Gold district Copper 0

200 km

Nickel Davao Zamboanga

N 0

Chromitite

3

2

PGM

Baguio

VMS 200 km

Parker

Marble

16

Limestone

FIGURE 2 Map showing the distribution of igneous, sedimentary,

metamorphic, and ophiolitic rocks. Red triangles show the location

1

4

5

Coal

Bicol

Sedimentary basin

of the active volcanoes found in the Philippines, which include Pinatubo, Mayon, Taal, Canlaon, and Parker. Lithologies associated with the

Masbate

6

Palawan Microcontinental Block are continent derived, whereas for 7

the Philippine Mobile Belt, the origin of the lithological suites are var-

11

Department of Environment and Natural Resources and data from the

10

Web site of the Philippine Institute of Volcanology and Seismology).

up of twelve large stratovolcanoes and smaller cones that include Mayon volcano (Fig. 2). Some areas in the Philippine archipelago are floored by a metamorphic basement. Most of the metamorphic belts recognized in the Philippine archipelago are related to regional metamorphism brought about by largescale geologic processes (e.g., collision, regional igneous activities). Some metamorphic rocks, mostly ophiolitederived lithologies, are products of ocean floor metamorphism. The metamorphic rocks are found all throughout the country (Fig. 2). Ophiolite complexes, representing oceanic crust–upper mantle sequences, have been mapped in different parts of the Philippine archipelago. A complete ophiolite sequence consists of (from bottom to top) peridotite, layered gabbro, massive gabbro, dike-sill complex, and volcanic rocks. Ophiolites recognized in the Philippine archipelago are mostly of Cretaceous age, especially those exposed in the

734

PHILIPPINES, GEOLOGY

12

8

ied (map drawn using data from the Mines and Geosciences Bureau–

Eastern Mindanao

9

12 13 8

14 16 15

FIGURE 3 Some of the metallic and nonmetallic mineral deposits found

in the archipelago include chromium, nickel, and the platinum group of metals. The major copper and gold deposits are found in northern Luzon (Baguio), southeastern Luzon (Bicol), and eastern Mindanao. There are 16 onshore and offshore sedimentary basins in the Philippine archipelago: (1) West Luzon; (2) Ilocos; (3) Cagayan; (4) Central Luzon; (5) Bicol Shelf; (6) Southeastern Luzon; (7) West Masbate–Iloilo; (8) Mindoro–Cuyo Platform; (9) Visayan; (10) Reed Bank; (11) Northwest Palawan; (12) Southwest Palawan; (13) East Palawan; (14) Sulu Sea; (15) Agusan–Davao; and (16) Cotabato (figure drawn using data from the Mines and Geosciences Bureau–Department of Environment and Natural Resources and the Department of Energy).

marine, deltaic, and fluvial clastic and carbonate rocks. The younger Neogene section in these sedimentary basins is generally undeformed, with the older Paleogene section showing more intense deformation. Most of these basins have axes trending north-south, whereas those basins near Palawan have a northeast trend. Sedimentary basins near southeastern Luzon are characterized by a more northwesterly trend. These sedimentary basins have been explored for oil and gas accumulations. The geologic features comprising the Philippine archipelago are defined by distinct geophysical anomalies. Low-gravity anomalies coincide with these deepsea trenches (Fig. 4) because the trenches are filled with water or low-density sediments instead of rock or highdensity materials. Onland, low-gravity anomalies generally characterize the sedimentary basins as a result of the low densities of sedimentary rocks. Distinct high-gravity anomalies are seen over areas underlain by ophiolitic and

120o

125o

Crustal thickness:

igneous units because these rocks have high densities. The Philippine Fault Zone and other major structural features are defined by linear contours and by displacements in the geophysical anomalies. Gravity and seismic data were used to determine the thickness of the crust. Available data show that the Philippines is generally characterized by crust with a thickness varying from ∼17 to 30 km. This is typical thickness for island arcs worldwide. The available data also suggest a thicker crust (greater than 30 km) in Central Luzon and in the Bicol–Panay–Central Mindanao area (Fig. 4). These regions experienced several episodes of arc magmatism, which served to thicken the crust. In terms of paleomagnetic investigations, results show that during the Early Miocene, Mindoro and Marinduque rotated counterclockwise whereas Panay rotated clockwise (Fig. 4). The rotations are attributed to the collision of the Palawan Microcontinental Block with the Philippine Mobile Belt. Paleomagnetic data also showed that northern Luzon occupied subequatorial latitudes throughout a significant portion of the early Cenozoic history of the Philippine archipelago.

~17 - 30 km 31 - 65 km

15o

10o

FIGURE 4 Gravity anomaly map (drawn using the map construction

tool at http://www.serg.unicam.it/Gravity.htm) shows low-gravity anomalies (dark blue–purple colors) over the deep-sea trenches surrounding the Philippine archipelago. The white arrows show the counterclockwise rotation recorded for Mindoro and Marinduque and clockwise rotation for Panay during the Early Miocene. The area defined by the white solid line corresponds to crustal thickness from ∼17 to 30 km. The brown-shaded regions have crustal thickness greater than 30 km.

GEOLOGIC HISTORY OF A COMPOSITE TERRANE

The pre-collision history of the Palawan Microcontinental Block and the Philippine Mobile Belt is evident from its contrasting stratigraphy (Fig. 5). The Palawan Microcontinental Block consists of Paleozoic or older metamorphic rocks and a chert-clastic sequence of continental affinity. The lithologic and biostratigraphic correlation of the different rock units in the Palawan Microcontinental Block with those found in China, Russia, and Japan provides evidence for a similar origin for these units—that is, these were derived from the east Asian margin. Carbonate deposition was prevalent in the Palawan Microcontinental Block from Eocene to Oligocene, as indicated by the limestone deposits that cap the older units. On the other hand, the Philippine Mobile Belt evolved mainly as a Cenozoic arc developed on a Mesozoic to Paleogene ophiolitic or metamorphic basement. Palinspathic reconstruction for the early Cenozoic puts the Philippine Mobile Belt at the subequatorial region along the proto-Philippine Sea Plate and Indo-Australian plate margin. The evolution of the Philippine archipelago began in the Mesozoic when a fragment of the Asian margin broke off to become the Palawan Microcontinental Block. The margin experienced lithospheric extension in the Late Cretaceous. This was followed by the formation of oceanic crust and opening of the South China Sea in the

PHILIPPINES, GEOLOGY

735

Palawan Microcontinental Block

Holocene Pleistocene

Clastic and carbonate rocks

Pliocene

Philippine Mobile Belt

Clastic and carbonate rocks

Miocene Oligocene

Limestones

Eocene

Ophiolitic rocks

Paleocene Cretaceous Jurassic Triassic Permian Carboniferous

Metamorphic rocks and chert-clastic sequence of continental affinity

Limestones

Carbonate rocks

Metamorphic and ophiolitic rocks

Intrusive rocks

Intrusive rocks

Intrusive rocks

No record Unconformity

FIGURE 5 The difference between the Palawan Microcontinental Block

and the Philippine Mobile Belt is evident in the contrasting lithologies observed in the two areas. The Palawan Microcontinental Block is made up of Paleozoic to Mesozoic strata of continent affinity overlain by younger carbonate rocks. The Philippine Mobile Belt is composed of metamorphic, ophiolitic, and intrusive units capped by clastic and carbonate sequences. There are three well-recognized episodes of magmatism (Cretaceous to Eocene, Oligocene to Miocene, and Pliocene), as revealed by the igneous rocks distributed in different parts of the Philippine Mobile Belt.

Late Oligocene to Early Miocene. Indentation of the oceanic leading edge of the Palawan Microcontinental Block with the Philippine Mobile Belt following its northward translation started during the Miocene, with the collision terminating by the Pliocene. The Philippine Mobile Belt, in itself, is an agglomeration of several terranes of varied origin (e.g., ophiolitic, island arc). With the collision of the Palawan Microcontinental Block with the Philippine Mobile Belt, the whole island arc system can be classified as a composite terrane. Dioritic, alkalic, and andesitic rocks distributed within the Philippine Mobile Belt represent pulses of magmatism during the Cretaceous to Paleogene, Miocene, and Pliocene times (Fig. 5). Subduction in certain parts of the Philippine archipelago commenced during the Cretaceous. Sedimentary basins within the Philippine Mobile Belt began accumulating clastic and carbonate rocks during the Oligocene. A regional Middle Miocene unconformity, indicating rapid uplift or a transgressive event, is inferred from the sedimentary sequences in these basins. Episodes of oceanic floor generation, which began during the Cretaceous within marginal basins that have long been subducted, are evident from the ophiolites that are cur-

736

PHILIPPINES, GEOLOGY

rently exposed on land (Fig. 2). The mechanisms responsible for the emplacement of these ophiolitic units include large-scale faulting, onramping, and subduction upwedging, among others. PLATE INTERACTIONS AND THE CORRESPONDING MINERAL AND ENERGY RESOURCES

Late Mesozoic to Cenozoic intrusive rocks (e.g., diorite–quartz diorite, granodiorite, tonalite, monzonite, syenites) and Plio-Pleistocene volcanic rocks (e.g., dacite) (Fig. 3) that make up the island arc system are host to the gold–copper deposits in the Philippines. More than 50% of the country’s gold production comes from epithermal veins (temperatures of 50 to 300 °C), and the remainder is derived mostly from porphyry copper–gold deposits. Volcanogenic massive sulfide deposits have been developed recently for the associated gold deposits. Most of the mineralization in the country can be grouped into two major episodes occurring during the Middle Miocene and Late Miocene–Pliocene periods. It should be noted that Cebu hosts the Atlas porphyry copper deposit, which is of Cretaceous age. The ophiolitic rocks host chromitite, nickel, and in some instances, platinum and palladium (Fig. 2, 3). Together with iron and cobalt, nickel also occurs as a secondary product within laterites, which form as a result of the weathering of ultramafic rock units. As mentioned, volcanogenic massive sulfide deposits have been recognized to be viable sources not only of base metals (e.g., lead and zinc) but, more importantly, of gold (Fig. 3). Most of these deposits are hosted by the volcanic carapace units of ophiolite complexes or the overlying sedimentary sequences that have been metamorphosed (e.g., quartz–sericite schist). Potential offshore mineral resources in the Philippines comprise mainly placer deposits that include chromitite, magnetite, silica, and, to a limited extent, gold. In terms of nonmetallic resources, Miocene and PlioPleistocene limestone deposits provide the raw materials for cement. The most significant deposits are found in northern and central Luzon, central Visayas, and central Mindanao (Fig. 3). Marble, the metamorphic counterpart of limestone, which is used in construction and building materials, is distributed in various parts of the archipelago. The most well-known marble deposits are found in the central Philippines (i.e., Romblon) (Fig. 2, 3). The Philippine archipelago has 16 sedimentary basins over an area of 700,000 km2, which have potential for hydrocarbon resources (Fig. 3). The most productive is the northwest Palawan Basin, which has, to date,

produced 54 million barrels of oil, 4 million barrels of condensate, and 67 billion cubic feet of gas. Hydrocarbon deposits were also discovered in the Cagayan Valley Basin and in the Visayan Basin. Coal resources also abound in the Philippines, the most significant of which are found in Mindoro and in Zamboanga del Sur in western Mindanao (Figs. 2, 3). GEOLOGIC HAZARDS AND ISLAND ARC DEVELOPMENT

The Philippines experiences, on the average, around 20 earthquakes daily. but most of these are hardly felt. The frequency with which earthquakes occur is due to the presence of earthquake generators such as the trenches on either side of the archipelago. Other earthquake generators are active faults, the most significant of which is the Philippine Fault Zone. The country has been devastated by 12 destructive earthquakes, with magnitudes between 6.2 and 7.9, during the period 1968–2003. This includes the earthquake that took place on August 2, 1968, with a magnitude of 7.3 and intensity VIII, reported in Casiguran, Quezon. A second devastating earthquake was the one that struck Luzon on July 16, 1990, with a magnitude of 7.9. Its epicenter was located near the town of Rizal in Nueva Ecija, but the earthquake was felt in north and central Luzon (see Fig. 1). Aside from destructive earthquakes, the Philippines has also been ravaged by volcanic eruptions from the 22 active volcanoes throughout the country. The most active volcanoes are Mayon, Canlaon, Bulusan, Taal, HibokHibok, and Pinatubo (Fig. 2). Hazards related to volcanic eruptions include lava flows, pyroclastic flows, ash fall, and lahars. The deadliest eruption of Mayon Volcano was on February 1, 1814, which was characterized as a Plinian eruption accompanied by pyroclastic flows and lahar. Another active volcano is Mount Pinatubo, which erupted on June 1991 after around 460 years of inactivity. The eruption column rose to a height of up to 30 km from the volcano’s vent. Lahars resulted when typhoon Yunya combined with the ash cloud from Pinatubo and caused most of the deaths and destruction in western Luzon. With more than 5 km3 of volcanic ash and rock fragments on the slopes of Pinatubo, lahar deposition went on for several years, filling up river channels and forming dammed lakes. Another natural hazard with which Filipinos have to deal is mass wasting or landsliding, which is usually triggered by excessive rainfall. The most recent tragic landslide happened on February 17, 2006, and buried the town of Guinsaugon, Southern Leyte, Philippines (see Fig. 1 for

location). More than 600 mm of rainfall, brought on by a La Niña event, was recorded in the weather station closest to Guinsaugon, 10 days prior to the actual landslide. The tragedy led to 154 deaths, with 973 persons still missing (and presumed dead). Considering the geographic location and geologic characteristics of the Philippines, an understanding of these natural hazards and how to mitigate the effects are critical in the development of this island arc system. SUMMARY

The Philippine archipelago is an interesting laboratory that allows the observation of features related to past and present-day active tectonic processes. Its location at the convergence between the Sundaland–Eurasian and Philippine Sea plates has produced an array of tectonic features that shaped the evolution of this island arc system. A composite terrane made up of an aseismic Palawan Microcontinental Block and the seismically active Philippine Mobile Belt, this island arc system bears witness to various geological processes and their corresponding effects. These include, among others, collision resulting to island rotations, marginal basin closure leading to ophiolite emplacement, and large-scale strike-slip fault formation as a consequence of oblique subduction. As a result, metamorphic suites are formed, volcanic arc belts are generated, and sedimentary basins receive their fill. The formation and evolution of the Philippine archipelago also saw the formation of both metallic and nonmetallic mineral resources. Gold, copper, nickel, chromium, platinum-group metals, volcanogenic massive sulfide, limestone, and marble are some of the more economically attractive resources being developed now. Oil and gas resource exploration activities are being undertaken in both onshore and offshore sedimentary basins of the country. Because the country is a composite terrane in a seismically and volcanologically active region (the Pacific Ring of Fire), various geologic hazards are present. Earthquakes, volcanic eruptions, landslides, and their related hazards are some of the challenges with which people in the Philippine archipelago have to contend. These problems are exacerbated by the meteorological and hydrological hazards that hit the country. An understanding of these hazards and how to mitigate their negative effects is important to ensure the development of the Philippines. SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Island Arcs / Landslides / Lava and Ash / Philippines, Biology

PHILIPPINES, GEOLOGY

737

FURTHER READING

Balce, G. R., R. Y. Encina, A. Momongan, and E. Lara. 1980. Geology of the Baguio District and its implications on the tectonic development of the Luzon Central Cordillera. Geology and Paleontology of Southeast Asia 21: 265–288. Dimalanta, C. B., and G. P. Yumul, Jr. 2004. Crustal thickening in an active margin setting (Philippines): the whys and the hows. Episodes 27: 260–264. Queaño, K. L., J. R. Ali, J. Milsom, J. C. Aitchison, and M. Pubellier. 2007. North Luzon and the Philippine Sea Plate motion model: insights following paleomagnetic, structural and age-dating investigations. Journal of Geophysical Research 112: B05101, doi: 10.1029/2006JB004506. Rangin, C. 1990. The Philippine Mobile Belt: a complex plate boundary. Journal of Southeast Asian Earth Sciences 6: 209–220. Yumul, G. P. Jr., C. B. Dimalanta, R. A. Tamayo, Jr., and R. C. Maury. 2003. Collision, subduction and accretion events in the Philippines: New interpretations and implications. The Island Arc 12: 77–91.

PHOSPHATE ISLANDS JAMES R. HEIN U.S. Geological Survey, Menlo Park, California

Phosphate islands host deposits of phosphate rock (also called phosphorite) of sufficient quantity and quality to be economically mined. Most of the phosphate rock deposits were derived from bird droppings (guano). The phosphate rock was mined at various times in the past, but the only extant mine in the Pacific is on

the island of Nauru in the west equatorial Pacific. The most common phosphate minerals in insular phosphate rocks are composed of calcium (Ca2+) and phosphate (PO43-), combined with charge-balancing anions such as fluoride (F-), chloride (Cl-), and hydroxide (OH-). Calcium phosphates are used predominantly in agriculture as fertilizer, although they have many other industrial applications. DISTRIBUTION OF PHOSPHATE ISLANDS

Phosphate islands are located mostly at low and middle latitudes in the Pacific Ocean, with a few exceptions such as Christmas Island in the eastern Indian Ocean and Navassa Island in the Caribbean (Fig. 1). Two types of phosphate islands exist, which are distinguished by elevation. Low islands are atolls or low-elevation (less than roughly 50 m) carbonate platforms; phosphate rock occurs on islets that rim the atoll, within the lagoon that is enclosed by the atoll reef, or on the carbonate platform. Most high islands (more than roughly 50-m elevations) that host phosphate rock deposits are carbonate platforms formed by the uplift of atolls, during which the lagoons were filled in with carbonate debris. A few insular phosphate rock deposits formed on volcanic high islands. Phosphate deposits on the low islands were mostly small and were completely mined out during the late nineteenth century, with the exception of Starbuck Island in the central Pacific, which remained in production until 1927. Other examples of these low islands include

FIGURE 1 Location of phosphate islands in the Pacific and surrounding area.

738

P H O S P H AT E I S L A N D S

Jarvis, Howland/Baker, and Malden islands (Fig. 1). The low-island phosphate deposits are the younger of the two types of deposit and include contemporary guano accumulations as well as somewhat older deposits formed by the interaction of guano with Holocene (last 10,000 years) carbonate sediments and rocks (see the section on origin of the phosphate deposits). Phosphate deposits on some of the high islands were mined until relatively recently, with deposits on Makatea Island exhausted in 1966 and deposits on Banaba Island (formerly called Ocean Island) depleted in 1979. Since 1979, only Nauru Island in the Pacific has produced phosphate, which is exported mostly to Australia and New Zealand. Nauru, Makatea, and Banaba islands are known as the three great phosphate rock islands. Nauru had one of the highest per capita incomes in the world soon after it gained its independence in 1968 because of its phosphate rock export, but now it is in financial difficulties, and the island itself is nearly 80% uninhabitable because of the phosphate mining (Fig. 2). Mining on Nauru began in 1907, declined significantly since the 1980s, and came to a halt in 2003. With the help of Australia, the mining infrastructure has been rebuilt; exports started again near the end of 2006, and they are

expected to continue until about 2010. Phosphate has also been mined on Christmas Island (Australia) in the eastern Indian Ocean for more than a hundred years, with a short hiatus in mining from 1987 to 1991. Several new mining operations have been recently proposed for Christmas Island. All these high-island deposits are generally older than the low-island deposits and consist of Neogene (24.1 to 1.8 million years ago) and Quaternary (1.8 to present) phosphate rocks commonly associated with karst topography. Other high-island phosphate-rock deposits were exploited mainly during the first half of the twentieth century, especially during the two World Wars. Examples of those include many western Pacific island-arc-hosted deposits, such as Kita-Daito, Rota, Fais, Ngeaur (formerly spelled Angaur), Mecherchar (formerly known as Eil Malk), and Beliliou (formerly spelled Peleliu) islands (Fig. 1). A few phosphate islands exist outside the map area of Figure 1, such as Juan de Nova Island in the western Indian Ocean, where phosphate was mined from the early 1900s until 1970. Many other islands in the global ocean host small phosphate rock or guano deposits that are not now and never have been economically minable, so those islands cannot technically be classified as phosphate islands. COMPOSITION

FIGURE 2 Aerial photograph of Nauru Island, which has an area of

21 square kilometers. Most of the interior of the island has been mined, with the exception of the area around the lake in the southwest quadrant. Courtesy of the U.S. Department of Energy’s Atmospheric Radiation Measurement Program.

Guano is an organic-rich mixture composed of uric (C5H4N4O3), phosphoric (H3PO4), oxalic (H2C2O4), and carbonic (H2CO3) acids as well as ammonia (NH3), calcium, potassium, sodium, and magnesium, among other constituents. The most commonly occurring phosphate minerals derived from the interaction of guano with calcium carbonate (limestone) are composed of calcium (Ca2+) and phosphate (PO43-), combined with chargebalancing anions such as fluoride (F-), chloride (Cl-), and hydroxide (OH-). These calcium phosphate minerals are predominantly varieties of apatite, including for example, carbonate apatite, carbonate hydroxyapatite, and carbonate fluorapatite. Other guano-derived phosphate minerals occur in insular phosphates but are rather rare; they include brushite, monetite, and whitlockite. Trace elements found in the apatites are derived from dissolution of the limestone or from seawater. Modern guano is about 5% phosphorus, which is upgraded from about 1.7% phosphorus in fresh seabird droppings containing about 22% nitrogen. With time, chemical reactions (mostly leaching) decrease the nitrogen content and increase the phosphorus content to about 9–12% phosphorus. Further upgrading to as much

P H O S P H AT E I S L A N D S

739

as 18% phosphorus can occur when the guano reacts with the carbonate rocks and sediments, thereby forming a phosphate-rock deposit of economic importance. ORIGIN OF THE PHOSPHATE DEPOSITS

The seabirds most responsible for producing large guano deposits include boobies, terns, and frigates. Two characteristics are essential for the formation of guano-derived phosphate rock deposits. The first is high primary productivity in surface waters surrounding the islands. This is essential to produce the great amount of fish (mostly anchovies) required to support the vast numbers of seabirds that deposit guano on the islands, which can range to millions of birds for a single island. It is estimated that enriched guano on Chincha Island off Peru (Figure 1) was more than 45 m thick, with pits up to 125 m thick. Primary productivity in the surface waters is supported by a process called upwelling, wherein cold, nutrient-rich waters rise from deeper levels (a few hundred meters) to the sea surface, carrying the silicate, phosphate, and other nutrients required for the flourishing of plankton, which occupy the base of the food chain. The second criterion is low rainfall, which allows for the preservation of the guano once it accumulates and also inhibits the growth of vegetation that would decrease the area available for ground-nesting birds. Further, high rainfall leaches the guano of its valuable nitrates, a highly valued component for fertilizer in addition to the phosphorus. These criteria for formation of phosphate rock—high primary productivity and low rainfall—may not describe the current conditions around some islands where older insular phosphate deposits occur, deposits that formed at a time in the past when the then extant climate was different from the modern climate. Once guano is deposited, it interacts chemically with the carbonate rock and sediment (limestone) of the underlying reef, which requires limited rainfall. The leaching solution, strongly charged with phosphate, percolates into the limestone and may form a phosphate hardground or phosphate coating on grains within the vadose zone (zone above the water table where pores contain air as well as water). Islands commonly have a lens-shaped zone of saturated rock (where the pores are filled with water) that is freshwater above and saltwater below, with mixing at the boundary. Some scientists have proposed that large insular phosphate deposits formed within the seawater part of the lens, whereas others have suggested that they formed in the freshwater lens or in the vadose zone. Phosphate rock forms when phosphate in groundwater replaces the limestone via dissolution

740

P H O S P H AT E I S L A N D S

and immediate precipitation of apatite, and precipitation of apatite in void space in turn commonly cements carbonate grains. It is clear that some chemical elements found in insular phosphates are not derived from guano—for example, the high fluorine and uranium contents—because those elements occur in very low quantities in guano. Rather, those elements must have been derived from the limestone via reactions with the phosphate-rich waters, from the dissolution of limestone during the formation of karst topography, or from seawater. Limestone and seawater are likely sources of these elements for different insular phosphate deposits; however, some deposits have never come in contact with seawater and must have acquired those elements from the dissolution of limestone. PHOSPHATE COMMODITY

Phosphorus is an essential element for plant and animal nutrition. Mined phosphate rock is usually processed to produce phosphoric acid for fertilizer and elemental phosphorus for other applications. Insular phosphates with high nitrate contents are in especially high demand, because they fulfill the need for both phosphorus and nitrogen in agriculture. A third essential element needed to fertilize crops is potassium, which is also enriched in some phosphate rock deposits. Examples of nonfertilizer applications of phosphate derivatives include food additives (such as phosphoric acid in soft drinks), detergents, herbicides, pesticides, water treatment, lubricants, matches, flares, and fireworks, to name just a few. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Island Formation / Makatea Islands / Pacific Region / Seabirds FURTHER READING

Burnett, W. C., and A. I. N. Lee. 1980. The phosphate supply system in the Pacific Region. GeoJournal 4(5): 423–436. Hein, J. R., B. R. McIntyre, and D. Z. Piper. 2005. Marine mineral resources of Pacific Islands—a review of the exclusive economic zones of islands of U. S. Affiliation, excluding the state of Hawaii. U. S. Geological Survey Circular 1286. Hutchinson, G. E. 1950. The biogeochemistry of vertebrate excretion. American Museum of Natural History Bulletin 96. Piper, D. Z., B. Loebner, and P. Aharon. 1990. Physical and chemical properties of the phosphate deposit on Nauru, Western Equatorial Pacific Ocean, in Phosphate deposits of the world, Vol. 3. W. C. Burnett and S. R. Riggs, eds. Cambridge, UK: Cambridge University Press, 177–194. Stoddart, D. R., and T. P. Scoffin. 1983. Phosphate rock on coral reef islands, in Chemical sediments and geomorphology: precipitates and residua in the near-surface environment. A. S. Goudie and H. Pye, eds. London: Academic Press, 369–400.

PIGS AND GOATS ELIZABETH MATISOO-SMITH University of Auckland, New Zealand

Pigs (Sus scrofa) and goats (Capra hircus) were two important domesticated animals taken to islands by early agriculturalists. Later introductions of these animals by European explorers and sailors, who often left them on islands as provisions for passing ships or shipwreck survivors, further extended their island distributions. Unfortunately, the impact of these animals on island ecosystems has been significant. In recent years eradication measures have been undertaken to remove these invasive species and restore the native island habitats they destroyed. ORIGINS, INITIAL DOMESTICATION, AND EARLY ISLAND INTRODUCTIONS

Pigs and goats belong to the order Artiodactyla. Both species are omnivores and particularly adaptable, which no doubt was one factor in their being two of the earliest domesticated forms of livestock. This adaptability would have also been a significant factor leading to their continued survival when introduced to many marginal or isolated islands around the world. Wild boar and goat have been important components of the human diet for tens of thousands of years, and recent molecular evidence suggests that both species were probably domesticated multiple times independently in several different locations. Archaeological and genetic evidence on goats suggest at least three major domestication events beginning around 10,500 years ago. These domestication events have been linked to the Near East and Indus regions (the source populations of later European and African lineages) and to East Asia. The fact that they provide not only meat and hides but milk and fiber in a relatively compact package that was easy to transport by boat made goats ideal animals to take on early voyages to the various islands in the Mediterranean region. Archaeological remains of domesticated goats and sheep appear from as early as 8000 years ago on Crete and Cyprus and in later Neolithic and Bronze Age settlements on other large islands in the Mediterranean and Baltic regions. The origin of Sus scrofa appears to be western island Southeast Asia. From there, populations dispersed into the Indian subcontinent and later radiated into East Asia, Eurasia, and Western Europe. Domestication appears to

have occurred independently in all of these areas beginning around 9000 years ago. Domesticated pigs were also taken to the major islands of Europe with Neolithic farmers. Despite earlier suggestions for domestication of wild boar on some islands, such as Sardinia, genetic evidence suggests that all island populations studied are descendants of pigs domesticated on the mainland. The East Asian domesticated pigs eventually were transported to the islands of the Pacific by prehistoric peoples. There is much debate about the date of pig introductions to New Guinea, which is also a center of early plant domestication. Some have argued that pigs may have been introduced to New Guinea during this early phase of agricultural development as early as 10,000 years ago. Both translocation and domestication of some of the other Southeast Asian Sus species, such as Sus celebensis, have also occurred in island Southeast Asia. The spread of Sus scrofa throughout the rest of the Pacific clearly did not occur until around 3300 years ago, when they were transported as part of the Lapita cultural complex and introduced to the islands of the Bismarck Archipelago and the Solomon Islands and out into Remote Oceania as far as Samoa and Tonga. From there they were taken to most of the Polynesian islands with the initial colonists. Pigs were not, however, successfully introduced to New Zealand or Easter Island by Polynesian colonists. Interestingly, despite their important use either as protein or for the long-term storage of surplus carbohydrates, pig populations were not always maintained on Pacific islands. Archaeological evidence suggests that on some small Pacific islands, particularly atolls, pigs were often extirpated, presumably on account of their impact on and competition for fragile and limited resources. This may explain the lack of pigs in Micronesia at European contact. Little is known for sure about the physical or behavioral characteristics of the first Pacific pigs. Early European accounts generally describe them as small, short, and dark with sharp backs, stiff bristles, and long snouts. These are indeed the characteristics of the wild pigs found in New Zealand today, often referred to as “Captain Cookers,” which are believed to be the descendents of the first pigs introduced to New Zealand by Captain James Cook. Whether these were “native” Pacific pigs he picked up in the islands he visited before arriving in New Zealand, European pigs he carried with him, or both is not clear. EUROPEAN INTRODUCTIONS

European exploration and colonialism beginning in the sixteenth and seventeenth centuries resulted in

P I G S A N D G O AT S

741

the introduction of many domesticated plants and animals to islands around the world. Pigs and goats were among the most successful introductions. Sailors, traders, whalers, and sealers would often release pigs and goats on islands, often uninhabited, so that they would establish natural feral populations, providing a guaranteed supply of fresh meat for future passing voyages and castaways. The first recorded introduction of goats to an island was in the Madeiras in 1458. When the Spanish arrived in the Canary Islands in the early 1500s, they recorded that goats and pigs were already present. These were most likely transported by the Guanches, the aboriginal peoples of Tenerife, who settled the islands from North Africa. During his second and third Pacific voyages of 1773 and 1777, Captain James Cook released pigs and goats to most of the islands he visited, including Hawaii and New Zealand. The pigs brought by the Europeans often replaced the Polynesian pig populations on islands where they were present, as the European breeds were generally larger and fatter than the original introductions. La Pérouse introduced both pigs and goats to Easter Island in 1786. Trade ships from the East Indies and the Philippines were the main sources of pig and goat introductions to Micronesia. In 1790, two ships from the British East India Company introduced eight she-goats, two rams, five sows, and two boars to Palau. Missionaries introduced pigs and goats to many Pacific islands throughout the nineteenth century. Twentiethcentury introductions of both pigs and goats to islands for game hunting, or by fishermen as in the case of Pinta in the Galápagos, resulted in the modern distribution, making these two species among the most broadly distributed mammals, after rats, worldwide.

Pinta Island in 1959 provides a dramatic example of the damage that can be done in a remarkably short period of time. Within 10 years of their initial introduction, the two females and one male released on the 60-km2 island resulted in a population of 5000 to 10,000 goats. This resulted in the loss of four endemic shrubs from the lowland areas and the dramatically reduced distribution of five other species. Unfortunately, other islands to which goats have been introduced have had similar stories. Pigs have an equally devastating impact on island ecosystems. Pig rooting not only digs up plants but significantly alters soil content, causing oxidation and leaching of key minerals. The destruction of leaf litter habitats by pigs often results in the loss of several native vertebrate and invertebrate species. In Hawaii, feral pigs destroy tree ferns, a favorite food, along with their associated epiphytes. As a secondary impact on native flora and fauna, pigs, like goats, often disperse the seeds of invasive species such as the strawberry guava, which has been particularly devastating in Hawaii. Pigs are known to eat the eggs and young of numerous ground-nesting birds and amphibians. They also feed on exposed reefs at low tide on many Pacific islands (Fig. 1), causing damage to the intertidal zone. Feral pigs transmit a number of diseases including pseudorabies, leptospirosis, and Japanese encephalitis, and they are known for carrying parasites that can be passed to humans and other animals. Pigs are known to be strong swimmers and thus can self-disperse across significant water gaps. The naturalist A. R. Wallace, in fact, described seeing pigs swimming from Singapore to the Malacca Peninsula.

ECOLOGICAL IMPACTS OF PIGS AND GOATS

Given that most oceanic island ecosystems have evolved in the absence of grazing and browsing ungulates, the introduction of both pigs and goats have a major negative impact on island floras. They are both considered to be in the top 100 of the world’s worst invading species. The grazing habits of goats result in significant loss of native plant species, particularly woody plants and shrubs. They remove leaves and young shoots and strip bark, resulting in plant death. This, in combination with trampling and other secondary impacts, leads to further devastation such as soil erosion and loss of dependent faunas. In some locations goats have been identified as the primary cause of island plant extinctions. The case of the introduction of three goats to

742

P I G S A N D G O AT S

FIGURE 1 Pacific pig feeding on a reef in Tonga. Photograph by Sean

P. Connaughton.

ERADICATION AND POPULATION CONTROL

Given the devastating impact of goats and pigs on island ecosystems, not surprisingly, eradication and population control are major issues for many governments and conservation groups. Invasive mammals can be removed from islands using a range of approaches including hunting, poisoning, trapping, or biocontrol. Previous eradication programs for both feral goats and pigs have shown that the use of a combination of approaches is most effective, particularly if a sustained effort and population monitoring is maintained. In island locations where both goat and pig removal is being attempted, it has been shown that the eradication of pigs prior to that of goats dramatically reduces the costs of removal. If goats are removed before pigs, or if removal attempts are simultaneous, the rapid recovery of dense foliage in the absence of goats makes pig removal much more difficult and time consuming. In New Zealand, pig eradication from small islands began in the mid-twentieth century and employed hunting as the primary means of removal; however, the combination of hunting and poisoning is now generally used on larger islands. The extreme mobility of both pigs and goats can make eradication efforts difficult and expensive, particularly on large islands. Once animals are removed from one area, they are quickly replaced by new populations. In Hawaii, the use of fencing has aided pig control in areas such as the Volcanoes National Park. Goats have been successfully removed from over 120 islands ranging in size from 1 hectare (Marielas Sur, in the Galapagos) to 132,867 hectares (Flinders Island, Australia). Hunting is the most common method used for goat eradications, involving game hunting, hunting with dogs, and shooting from helicopters. The gregarious nature of goats has also lead to a particularly successful hunting aid: the use of Judas goats. Judas goats are goats that are captured, fitted with radio collars, and released. These goats then search out other goats, thus identifying their location, so that hunters can then locate populations that might otherwise evade discovery. Once the identified group has been located and removed, the Judas goat is rereleased to seek out new groups. The use of Judas goats has been most successfully employed in areas where goats may only be found at low densities (for example, after major eradication programs using other methods have reduced an originally dense population) or where topography makes tracking and access difficult. The successful removal of goats and pigs from island ecosystems has been shown to have a dramatic and rapid

effect on vegetation. Unfortunately, non-native plant species are often the most rapidly recovering plants after removal of pigs and goats. The use of pig and goat exclosures to determine how the various plant communities will react to the loss of these species is recommended prior to eradication so that an environmental management scheme can be developed. Studies in Hawaii have shown that in terms of native plant recovery, lowland grasslands are perhaps the least likely to recover once pigs and goats have been removed. Native woody species have been shown to recover relatively quickly in lowland areas after removal of goats in particular. The more upland areas and rainforests recover differently depending on the degree of goat and pig impact. Where the impact has not been great because of low density or short period of pig or goat presence, native plants recover relatively well. However, where the impact has been severe, recovery even 6 to 8 years after removal has been negligible. Eradication of pigs or goats on large islands or on islands where reintroduction is likely must include some degree of community consultation, outreach, and education. Both pigs and goats have recreational and economic uses, and local communities may react negatively to eradication attempts. Unsuccessful eradications have occurred on several large islands such as Great Barrier Island in New Zealand and Lord Howe Island in Australia because of lack of local support or other political issues. Successful eradication of goats and pigs from islands requires the cooperation of biologists, ecologists, anthropologists or sociologists, educators, and local communities and government agencies. SEE ALSO THE FOLLOWING ARTICLES

Biological Control / Hawaiian Islands, Biology / Human Impacts, Pre-European / Introduced Species / Invasion Biology / New Zealand, Biology FURTHER READING

Campbell, K., and C. J. Donolan. 2005. Feral goat eradication on islands. Conservation Biology 19: 1362–1374. Giovas, C. M. 2006. No Pig Atoll: island biogeography and the extirpation of a Polynesian domesticate. Asian Perspectives 45: 69–95. Hide, R. 2003. Pig husbandry in New Guinea: a literature review and bibliography. ACIAR Monograph No. 108. Canberra: Australian Centre for International Agricultural Research. Larson, G., K. Dobney, U. Albarella, M. Fang, E. Matisoo-Smith, J. Robins, S. Lowden, H. Finlayson, T. Brand, E. Willerslev, P. RowleyConwy, L. Andersson, and A. Cooper. 2005. Worldwide phylogeography of wild boar reveals multiple centres of pig domestication. Science 307: 1618–1621. Luikart, G., L. Gielly, L. Excoffier, J.-D. Vigne, J. Bouvet, and P. Taberlet. 2001. Multiple maternal origins and weak phylogeographic structure in domestic goats. Proceedings of the National Academy of Sciences of the United States of America 98: 5927–5932.

P I G S A N D G O AT S

743

PITCAIRN

Another 160 ha are on slopes suitable for some cultivation, although the plots on such slopes are subject to erosion. CLIMATE

NAOMI KINGSTON National Parks and Wildlife Service, Dublin, Ireland

NOELEEN SMYTH National Botanic Gardens, Dublin, Ireland

Pitcairn Island is one of four islands in the Pitcairn group, the most easterly island group in Polynesia: Pitcairn, a relatively young, high volcanic island; Henderson, an atoll uplifted by the eruption of Pitcairn; and two non-uplifted atolls, Ducie and Oeno. The group is extremely remote, separated from both New Zealand and South America by over 4500 km; from Easter Island, the nearest neighbor to the east, by 1570 km; and from the Gambier Islands to the west by 450 km. The isolation of Pitcairn makes it of equal interest to those studying the island’s biota and its origins, those studying its geology, and also those studying the unique culture that has developed since its most recent phase of human settlement in 1790. GEOGRAPHIC SETTING

Pitcairn Island (25 ° 04′ S, 130° 06′ W) is very small, being only 4 × 2 km, and with the highest point at 347 m. As the island is of volcanic origin, its terrain is very rugged, with the soil being derived from volcanic ashes, and the underlying rock types formed of consolidated ash. Four different volcanic episodes account for varying ash substrates across the island. The island is not protected by a fringing reef, so the coast is surrounded by cliffs with few small-boulder beaches. On the south of the island these cliffs and steep slopes reach almost to the highest point at 347 m. From the highest point in the southwest across to the northeast, the terrain is more gradually sloping, and the cliffs below Adamstown are much lower (∼50 m, and lower again at St. Paul’s Point). The only area with low coastal cliffs (∼30 m) on the south of the island is at Tautama. A flat area known as Aute Valley covers a considerable portion of the southeast corner of the island, at an altitude of approximately 180 m. Tedside is used as a general name for the area to the west and northwest of Big Ridge and Garnets Ridge, although it is actually composed of several small valleys. It slopes from High Point right down to a beach, which is also occasionally used as a harbor and may have been the location of the main Polynesian occupation. Overall only about 10% of the island is flat land, an area of about 550 ha, the remainder being at a 20–45° slope, and much steeper in some of the remoter valleys. 744

PITCAIRN

The climate of Pitcairn is subtropical, with mean annual rainfall of about 1716 mm, but with considerable annual variation. Mean temperatures range in summer from 17 to 28 °C, and in winter from 13 to 23 °C, with winter being wetter and windier. There are no permanent springs on the island, but during wet periods streams run down the centers of several valleys. As for the rest of this region of the south Pacific, the trade winds blow from the southeast, and although Pitcairn is affected by complex climatic patterns that affect the whole of the region, it is out of the line of cyclones and only occasionally hit by them. GEOLOGY AND SOILS

Pitcairn formed comparatively recently (∼0.75–1 million years ago), and basaltic lava dating shows that the high islands of the Gambier group (5.2–7.2 million years) and the Tuamotu island of Mururoa (6.5–8.4 million years), to the west, formed at the same hotspot as Pitcairn. Farther east from Pitcairn along the same hotspot alignment there are over 20 underwater volcanic edifices (seamounts), the shallowest of which is only 60 m below sea level. The island is migrating along this alignment at a rate of 12.7 ± 5.5 cm per year. The young geological age of Pitcairn may account for the lack of fringing reef, a feature commonly associated with volcanic islands in southeast Pacific. A detailed soil survey of the island, carried out in 1958, identified three main soil suites: • • •

Pulau suite, derived from tuffs, ash, and tuff agglomerates Adamstown suite, derived from basaltic lavas Taro Ground suite, derived from iron-rich basaltic lavas and agglomerates

All of the soils are deep and fertile, but with major erosion in parts. FLORA, FAUNA, AND HABITATS

Pitcairn has a variety of habitats, which consist of rugged sea cliffs on the north and south coasts, scrub and eroding slopes, ridge vegetation, invaded Syzygium jambos forest, Homalium taypau (taypau) forest, Pandanus tectorius (thatch) forest, and Meterosideros collina (rata) forest. The flora consists of 81 native vascular plant species (including 11 endemic species) and a further 250 introduced species. Non-native species are found right across the island, but native species still dominate in the remoter valleys on the south of the island. Most of the plant communities

contain high numbers of non-native species, and even in areas where there are a high number of native species, the dominant species is often an introduced taxon. Analyses have found that 63% of the native flora is threatened on Pitcairn, while 22% is globally threatened. The combination of taxa found in the Pitcairn group are most closely related to the taxa found in the Austral group of islands to the west of Pitcairn, which were in turn derived from the southeast Polynesian biogeographic region of the Pacific. The fauna of Pitcairn is less well known than the flora. The avifauna falls into four categories: endemic breeding landbirds, migrant land- and shorebirds, the seabirds visiting the waters of the region, and the breeding seabirds. The islands are categorized as a high-priority endemic bird area by BirdLife International, with internationally significant populations of seabirds. Bird diversity is low but highly specialized; the only landbird found on Pitcairn Island is the endemic Pitcairn Reed-Warbler. The land snail fauna is also of immense interest with 16 species, eight of which are endemic. Little is known about other invertebrate groups, but they would undoubtedly show similar levels of endemicity. The fish of the Pitcairn Islands show a low degree of endemicity, and green turtles and hawksbill turtles occur around the islands. Marine mammals are in need of assessment within the Pitcairn group, and it is probable that there are many cetacean species occurring in the surrounding waters. SETTLEMENT AND GOVERNANCE

Pitcairn was settled by Polynesians from about the tenth century, at a time when world climates were warmer and calmer seas allowed easy exploration of the Pacific. There was certainly trade with other Polynesian islands, as timber and tools from Pitcairn have been found elsewhere in Polynesia, including Easter Island. Similarly, pearl shells from the Tuamotus and Gambier Islands have been found on Pitcairn Island. Contact was lost between Pitcairn and the more westerly Polynesian islands in about the fourteenth century, during a period of climatic instability, and the population probably died out soon after this. Settlement may have never been permanent, but rather the island may have been used as a stop-off point for long voyages, or as a quarry for obsidian. Pitcairn Island was rediscovered on July 2, 1767, by Carteret for Britain, but was charted incorrectly. This is probably the reason why the Bounty mutineers sought out the island as their new home and hideaway and relocated to it in 1790. The story of the mutiny on the HMS Bounty is one of the most famous in seafaring lore. The Bounty was charged by King George III to sail to Tahiti and collect breadfruit plants (Artocarpus altilis), which were to be brought to the

West Indies as a source of food for slaves. Having spent longer then was originally planned in Tahiti, the crew were not keen to go back to Britain, and they mutinied in 1789. Captain William Bligh and his supporters were cast adrift, and Fletcher Christian, reputed to be the ringleader of the mutineers, returned with the rest of the men to Tahiti. At Tahiti they took women and some native men and went to Tubuai to set up a colony. This was unsuccessful, and so, after only a few months they went to sea again, aiming for the Solomon Islands, but instead finding Pitcairn Island. The initial settlers counted of nine mutineers, six Polynesian men, and 13 Polynesian women. Difficulties soon developed between the men on the island with disputes over land and women, the result of which was a period of unrest and murder. This meant that just 10 years after landing on the island only one man, John Adams, was left with 10 women and 23 children who had been born on the island. During this time an interesting culture had been developing on the island, with a mixture of Polynesian and European influences. The Pitcairnese creole language, still in use, had begun to develop, although it was later to incorporate words learnt from American whalers and others who settled on the island. Food and cooking methods were predominantly Polynesian, while the homes were built like European dwellings, but from local timbers such as taypau (Homalium taypau), miro (Thespesia populnea), and huliandah (Cerbera manghas). Tools were similarly a mixture of those taken from the Bounty ship and from Polynesia by the women. Traditional Polynesian tapa cloth was made, as the source plant, aute (Broussonettia papyrifera), survived following the earlier period of Polynesian settlement. Fishing was performed from the rocks until in 1795 the first European style canoe was built. The curious place names that have been given to all parts of the island give an insight into the history and culture of the Pitcairnese people (Fig. 1). The islanders have always been astutely environmentally aware, as their lives have always depended on the fine balance between population size and resource availability. The Pitcairn laws through the nineteenth century reflect their concerns about environmental sustainability. A report in the 1850s noted regulations about cutting timber for enclosures, highlighting that in less than 100 years of settlement, timber resources were becoming scarce. Their complete dependence on island resources for food was apparent when a law stated “no coconuts were to be taken from T’otherside” (now Tedside) unless the collectors were accompanied by someone in authority. Erosion and drought problems were becoming evident, attributed to the loss of the island trees. In 1856, because of

PITCAIRN

745

FIGURE 1 Map of Pitcairn Island, showing the intriguing place-names. Image courtesy of Pitcairn Island Administration, modified by Mark Vity.

overpopulation, the 194 islanders were evacuated to colonize Norfolk Island, by then abandoned as a penal colony. However, in 1859, two families returned, followed in 1864 by another four families, increasing the island’s population to 43. It was from this group that the extant population of Pitcairn is now largely descended. Even today close links are kept between the families on Norfolk and Pitcairn Island. In 1887 the islanders converted to Seventh-Day Adventism, a faith they retain to this day. From the Pitcairners’ first contact with the British Admiralty in 1814, the island has been under British Rule as an Overseas Dependent Territory. The Wellingtonbased British High Commissioner to New Zealand holds the office of Governor of the Pitcairn Islands, an appointment made by Her Majesty. The Governor holds formal powers “to make laws for the peace, order and good government of the islands,” and all the laws are styled “Ordinances.” Administration is via New Zealand through the Island Commissioner based at the Pitcairn Islands Office in Auckland. The Pitcairn Island Council is responsible for the local government and administration of internal affairs within the group, and the Island Council comprises the Island Mayor (elected every three years), the Island 746

PITCAIRN

Secretary, Chairman of Internal Affairs Committee, four officers (elected annually), and two advisers, one appointed by the governor and one by the elected members. CULTURE

The population of Pitcairn declined through the 1970s, 1980s and 1990s, and today the island population is about 50. However, over one-third of the native born Pitcairners live away from the island, with most of them resident in New Zealand. Contact with the outside world is via telephone and e-mail. Many islanders are also ham-radio enthusiasts. Until recently the small population of Pitcairn lived a subsistence existence, with extensive areas of the island in cultivation for a variety of food crops. Breadfruit is a particularly important carbohydrate food stable, although, in contrast to other Polynesian islands, the breadfruit trees on Pitcairn are generally not maintained or pruned, and the small fruits are often shot down from the trees with rifles. Currently the main employment on Pitcairn is in local government and community services, such as conservation officer and postmaster. Supplementary income is provided by the sale of woodcarvings termed “curios” to passing cruise ships and, to a lesser extent, by mail order. The social

and economic status of Pitcairn Island is one area in which dramatic changes are forecast for the coming years. In the past the islands’ main income was obtained from philately. However, during the last ten years, income from this activity has been reduced by over 80%, and the islands entered into budgetary aid in 2004. Business plans for the island are currently being developed; the aim is to create a sustainable economy and self-sufficiency. The developments include a new trade link and memorandum of understanding between the Pitcairn Islands and French Polynesia, the installation of wind turbines to harness electricity, a new breakwater/harbor to encourage cruise ships to visit Pitcairn as well as to stop at the other islands in the group, and the development of international markets for the sale of local produce (carvings and honey) to provide a boost for the island’s economy and development. Supplies arrive by ship from New Zealand every 1–3 months, and a small co-op shop keeps surplus goods. Access to the island is extremely difficult, because large ships cannot dock in the small harbor. All supplies to the island currently have to be offloaded into longboats at sea and transported into Bounty Bay (Fig. 2). There is no airstrip, and plans to construct a runway on the flat part of the island have for now been shelved and a fast catamaran will be obtained, which will allow a quicker sea link to Pitcairn via Gambier Islands. CONSERVATION

The main threats affecting the island biota are posed by habitat clearance, spread of invasive species, small population sizes or restricted distributions, loss of genetic diversity within species, erosion, and exploitation. The extinction of fruit-eating birds from the island is also preventing the successful dispersal and germination of many native plant species. Conservation management programs are underway to address these threats through species-specific recovery plans

and control of invasive species. Cleared areas are being replanted with native and economically important species, resulting in an environmentally and economically sustainable resource. Planting, in turn, controls erosion and reduces unchecked exploitation of native species. In addition, three potential reserve areas have been identified (at Tautama, Big Ridge, and Down Rope), which would set aside areas for nature conservation and prevent development in these areas. An environmental management plan for Pitcairn has been produced; it sets out a series of sustainable solutions that will allow the island economy and infrastructure to develop for future generations, while maintaining the integrity of the island’s unique ecosystems. The success of future conservation measures is reliant on the involvement of the local community, and so the plan has been drawn up in consultation with them and mindful of their current and future interests. SEE ALSO THE FOLLOWING ARTICLES

Easter Island / Exploration and Discovery / Lava and Ash / Peopling the Pacific / Sustainability FURTHER READING

Benton, T., and T. Spencer, eds. 1995. The Pitcairn Islands: biogeography, ecology and prehistory. London: Academic Press. Göthesson, L-Å. 1997. Plants of the Pitcairn Islands including local names and uses. Sydney: University of New South Wales. Kallgard, A. 1991. Fut yoli noo bin laane aklen? a Pitcairn Island word list. Sweden: University of Goteborg. Kingston, N., and S. Waldren. 2003. The plant communities and environmental gradients of Pitcairn Island: the significance of invasive species and the need for conservation management. Annals of Botany 92(1): 31–40. Kingston, N., and S. Waldren. 2004. A conservation appraisal of the rare and endemic vascular plants of Pitcairn Island. Biodiversity and Conservation 14: 781–800. Nicolson, R. B. 1966. The Pitcairners. London: Angus & Robertson Ltd. Paulay, G. 1989. Marine invertebrates of the Pitcairn Islands: species composition and biogeography of corals, molluscs and echinoderms. Atoll Research Bulletin 326: 1–28.

PLANT DISEASE ULLA CARLSSON-GRANÉR, LARS ERICSON, AND BARBARA E. GILES Umeå University, Sweden

FIGURE 2 A view of the main landing point at Bounty Bay. Photograph

by Noeleen Smyth.

Diseases can increase mortality, decrease reproduction and growth of plants, and ultimately influence the sizes and genetic structures of populations and the species composition in plant communities. Plant populations situated on small and distant islands may more easily escape diseases than those on the mainland, where host and pathogen populations lie in close proximity. However, if pathogens PLANT DISEASE

747

spread to island populations that have previously evolved in absence of diseases, their effects may be severe. Studies of patterns of disease in insular systems have shown that the ages and sizes of plant populations and the distance between islands affect disease spread in archipelagoes.

spatial structure of islands (i.e., the distances between populations and the sizes of host populations) is predicted to affect the dynamics of host-pathogen interactions. Long-term studies of diseases in an archipelago in Sweden have presented interesting results.

DISEASES IN PLANTS

EFFECTS OF DISEASES ON POPULATIONS AND COMMUNITIES OF PLANTS

Plants serve as hosts to a wide range of parasites. These organisms, which can be fungi, oomycetes, nematodes, protozoa, bacteria, or viruses, live in or on their plant hosts, from which they derive their nutrients. If a parasite causes disease in its host plant and reduces the host’s fitness, we call the parasite a plant pathogen. The ways in which pathogens affect host fitness varies widely among different groups of parasites. For example, necrotic generalist pathogens among oomycetes or fungi that can also live on dead host tissue often result in extensive mortality, particularly among weakened plants suffering from waterlogging, oxygen deficit, or other stresses. Biotrophic pathogens that infect flowers (e.g., anther smut fungi, Basidiomycota) and developing fruits also have a strong effect on host fitness by preventing seed production, but these pathogens rarely affect host survival. Foliar biotrophic pathogens, including fungi such as rust, attack plant leaves and may reduce photosynthesis, ultimately decreasing resources available for growth, reproduction, and survival. However, the strength of their negative effects on plant individuals vary substantially between years (e.g., with extreme weather conditions). Due to growth of international trade and human travel around the globe, the spread of diseases that result in plant epidemics has increased over the years. This may pose a particular threat to island plant communities that may have evolved in the absence of disease or in isolation from particular diseases. Today, we have records of a number of plant pathogens that have been introduced and spread on islands. For example, an average of more than one new rust species has been found per year on New Zealand. These rusts have often been of northern temperate origin, but introductions from Australia also seem common. Increased numbers of introduced parasites causing disease have been noted on crops grown on the Maldive Islands. However, plant pathogens could also be used for controlling exotic invading species on islands. The invasion of ecosystems by alien species is actually one of the most important sources of biodiversity loss on islands. Diseases affect plant populations and communities and spread between plants and populations. Because the effects of parasitic fungi and oomycetes are better known than the effects of other groups of parasites, all examples in this article are taken from these kinds of systems. The

748

PLANT DISEASE

When a large proportion of plants are diseased and the fitness of infected individuals is reduced, the growth rate of populations may decrease, leading to reductions in the sizes and densities of plant populations. This can ultimately change the structure of plant communities, especially when large dominant keystone species are attacked. The potential numerical effects that pathogens might have on their specific hosts have been used to control plants that become invasive when introduced to new geographical areas. Such programs have been successful in controlling some exotic invasive plants on the islands of Hawaii, e.g., introduction of the host-specific fungal pathogen Septoria passiflorae (asexual Ascomycota) to Hawaii has resulted in an up to 90% decrease in its host, the invasive banana poka (Passiflora tarminianan). However, there are also examples of limited success in using fungal pathogens as control agents. For example, release of the rust fungus Puccinia chondrillina to control the Mediterranean invasive rush skeletonweed (Chondrillina juncea) in Australia reduced the abundance of the susceptible and most widespread clone of the weed, whereas the two clones that were resistant to the pathogen increased in abundance. Introduction of the pathogen resulted in selection and increased the frequencies of resistance in the skeletonweed population, leading to decreased effectiveness of biocontrol. Similar evolutionary changes might explain the varied success of other biological control programs. Evolutionary changes in host resistance can also mediate reciprocal changes in infectivity and aggressiveness in the pathogen population through frequency-dependent selection. In populations with a high frequency of resistant plants, more infective fungal isolates may be favored, and the pathogenicity of the pathogen population may increase. This, in turn, selects for rare resistant host genotypes, which then increase in frequency until a new pathogen strain manages to infect these plants, and so on. The most obvious large-scale numerical effects of diseases have been seen when pathogens migrate over large distances to new geographical regions (usually by the means of humans) and encounter new hosts for the first time. On the Seychelles, for example, a new fungal disease has caused high death rates of the native taka-

maka tree (Calophyllum inophyllum). Spores of the fungus appear to be spread by a native bark beetle, but the origin of this pathogen is still unclear. It may represent an introduced pathogen or a pathogen that has evolved to infect a new host in its home range. That hybridization between two pathogen strains may lead to novel pathogenicity has been observed on several occasions. Clearly, knowledge of the factors that allow a parasite to establish within a host population is essential for understanding and controlling diseases on islands and other natural ecosystems. DISPERSAL, INFECTION, AND ESTABLISHMENT OF DISEASES

To be able to colonize and parasitize a plant, a pathogen has to solve two problems. First, it must “find” a host plant; then it must infect that plant and grow and reproduce on or in it. Some parasites are directly transmitted by contact between susceptible hosts (e.g., from adult plants to seed offspring or between ramets of the plant). Others spread by means of wind, water, or animal vectors or through the soil. Although most pathogen spores usually land relatively close to the source of inoculum, some spores of wind-dispersed pathogens such as rusts can travel for great distances. Vector-transmitted pathogens, such as anther smut on caryophyllaceous hosts, which is spread by pollinators of the plant, spread more locally. The spread of soil-borne diseases such as Phytophthora spp. (Oomycota) is even more local, and in these systems the host usually disperses over longer distances than the pathogen. Once a parasite has dispersed to a new plant, it must be able to overcome host defenses to infect, grow, and reproduce. First of all, at least some individuals of a potential host species must be susceptible toward the particular pathogen isolate. Plants may, however, actively defend themselves biochemically by producing secondary metabolites that are toxic or otherwise inhibitory to pathogen growth. Host resistance can also be based on passive physical factors that allow the plants to avoid and escape the disease. Even when a pathogen manages to spread, overcome a host plant’s defenses, and cause an infection, the disease will not necessarily spread through the plant population. Theory predicts that the basic reproductive rate of the pathogen, R0 (the average number of new infections produced from a single infective individual), must exceed 1 to allow a pathogen to invade. It is commonly assumed that host density or population size must reach a critical threshold for a specific pathogen to persist. From agricultural situations, where plants are cultivated in homogenous stands, it is well known that increasing the density of plants can increase disease spread. In natural communities, plant populations vary from being dense, large, and continuous

to being highly subdivided in small patches. Each plant population may in turn be distinctly isolated from other populations or be a part of a network where populations exchange seeds and pollen. This variation in spatial structuring of host populations can affect disease spread. DISEASE SPREAD IN SPATIALLY STRUCTURED SYSTEMS

Theory predicts that increasing subdivision of host populations into smaller, more isolated patches slows the dispersal rate and increases the probability of local extinction of pathogens. It has been proposed that similar mechanisms could result in escape from diseases for plant species that are introduced to new, distant areas. Without diseases, these species may experience a demographic release with high recruitment, growth rates, or survival that may make them invasive in the new area (i.e., the enemy release hypothesis). When host patches are closer and more connected, pathogens may disperse frequently among populations, leading to long-term persistence of pathogen populations. In insular systems, plant populations on isolated islands are predicted to escape diseases more easily than populations that are closer to other populations. However, small, isolated populations on islands may be inbred with low diversity in resistance genes, which may limit their evolutionary response if diseases are introduced, for example, by human trade. As shown in the case of the introduced chestnut blight (Cryphonectria parasitica, Ascomycota) in North America, where all plants are more or less susceptible, a new pathogen can rapidly spread. When particular host species become rare, the pathogen may also evolve ways for persisting in an isolated situation. For example, generalist pathogen strains that are able to infect more than one species may be favored (given that these traits are genetically variable in the pathogen), which can increase the overall impact of the disease in the community. Disease dispersal and pathogen persistence are predicted to be higher in plant populations on islands found in archipelagoes. In these systems, interpopulation dispersal of both host and pathogen may occur often enough to rescue the pathogen from regional extinction, although local extinctions may occur. Theoretical studies have also shown that genetic variation in host and pathogen is easiest to maintain in such spatially structured situations. This can further increase pathogen persistence, because a few susceptible plants and a few infective pathogens are likely to occur in some sites in the system. At the same time, some host populations may escape the disease and the overall impact of a pathogen may be smaller than in large continuous populations or local isolated populations.

PLANT DISEASE

749

FIGURE 1 Islands of different ages and phases of primary succession in

the Skeppsviks Archipelago. (A) A new island just emerged above sea level. Terrestrial plants have not yet colonized the islands. (B) A young small, open island (∼50 years) with grasses and tall herbs (Valeriana sambucifolia and Filipendula ulmaria) growing in the middle part. Seedling establishment of the first tree, Alnus incana, occurs. Inundated under high-water periods. (C) A ∼100-year-old island. The central part of the island has scattered A. incana and Sorbus aucuparia bushes. Trientalis europaea may establish, and after a few more years, also Silene dioica. Inundated during autumn storms. (D) A. incana trees form closed stands in the central part on islands that are ∼150 years old. V. sambucifolia populations decline and occur only at the outer part of the Alnus border together with F. ulmaria. T. europea in dense populations and S. dioica populations begin to expand. (E) A ∼250-year-old island with an A. incana border toward the shore succeeded by S. aucuparia and then Betula pendula and spruce (Picea abies). T. europaea, S. dioica, and F. ulmaria are relatively abundant but become increasingly restricted to the shoreward parts. (F) An old island (> 300 years) fringed with a narrow A. incana border, succeeded by P. abies at the maximum high-water level. In the Alnus border, T. europaea is relatively abundant and S. dioica occurs in small patches or as single individuals. Photograph by B. E. Giles.

The theoretical framework for the importance of spatial structure in determining disease dynamics has grown alongside the development of the island biogeography and the metapopulation theories. However, detailed field studies of pathogen distributions and disease spread within and between island plant populations are restricted to a few long-term studies in Baltic archipelagoes in Sweden and Finland. DISEASE PATTERNS IN ARCHIPELAGOES

The Skeppsviks Archipelago in northeastern Sweden has been the focus for several studies of natural plant-pathogen

750

PLANT DISEASE

interactions. This area is particularly suitable for studying how various factors affect the dynamics of host-pathogen systems in island networks. The archipelago consists of about 100 islands of different ages and sizes within a 20-km2 area, and primary succession is in strikingly different phases on these islands (Fig. 1). By studying the demography of plants and patterns of disease spread in relation to the age and size of host populations and colonization and extinction rates, the long-term consequences of host-pathogen interactions can be estimated. The islands, which are composed of moraine deposited in a north-south direction by land ice during the latest glaciation, were initially left under water when the ice melted 7700 years ago. Since then, there has been an isostatic land uplift in the area, and the land still rebounds at a rate of about 0.85 cm per year. New land is continually made available for colonization, either as new islands rising above sea level or as extensions of existing islands. The height of an island above sea level is correlated with island age and also with the time since it was first colonized by plants, and it is possible to estimate the maximum population ages of plants that colonize during the early succession stages. Studies of the interactions between the plants Valeriana sambucifolia and the rust fungus Uromyces valerianae, between Trientalis europaea and the smut fungus Urocystis trientalis, and between Silene dioica and the anther smut Microbotryum violaceum in Skeppsviks Archipelago have shown that the percent of individuals infected in populations varies considerably among island populations of different ages. Since hosts must be present prior to their obligate pathogens, young populations generally show low disease levels. When island populations become older and larger, the probability of disease increases, and all intermediate-aged populations of the studied species, are diseased although the frequency of disease in different populations and host-parasite systems varies. Among old populations, fewer populations are diseased, and the percent of individuals infected is generally lower (Fig. 2). This effect is in part due to populations becoming more scattered on older islands as succession proceeds on the islands. Long-term studies of the M. violaceum/S. dioica system confirm this pattern; that is, island populations of S. dioica in which the anther smut has colonized have been larger than populations that have remained healthy. Moreover, even though M. violaceum causes a systemic infection and can live for many years in diseased plants, the pathogen has been lost from some populations. These populations have been smaller on average than diseased populations where the pathogen has persisted. The rate of disease spread is

FIGURE 3 (A) Spatial patterns in the disease occurrence of the rust

Triphragmium ulmariae among 129 populations of Filipendula ulmaria growing on islands (shown as squares) in Skeppsviks Archipelago. The FIGURE 2 Percent of diseased individuals in populations in relation

to estimated population age (years) for (A) Valeriana sambucifolia– Uromyces valerianae, (B) Trientalis europaea–Urocystis trientalis, and (C) Silene dioica–Microbotryum violaceum in Skeppsviks archipelago (from Carlsson et al. 1990).

also affected by the levels of resistance in S. dioica populations; that is, populations showing an increase in disease frequency tend to be more susceptible than populations where the disease has remained at low levels. Differences in resistances among populations are established during founding of populations; that is, by chance some populations are established by susceptible plants and others by more resistant plants. Although islands exchange seeds and pollen, populations remain differentiated because of limited gene flow between islands. Data from another long-term study in the Skeppsviks Archipelago on the interaction between the rust Triphragmium ulmariae and its host Filipendula ulmaria have similarly shown that the disease is more often found in larger host populations but also in populations that are in close proximity to larger diseased populations (Fig. 3). T. ulmaria is a nonsystemic rust that causes local lesions on its host, and this system is characterized by drastic winter bottlenecks when plants die back to an underground rootstock during the winter. It therefore shows large fluctuations in disease frequencies, both among populations and among years.

size of squares is proportional to host population size; red squares represent diseased populations, and black squares represent healthy populations (from Burdon et al. 1995). (B) Spores of T. ulmariae on F. ulmaria.

Clearly, the results obtained in these studies have confirmed some of the predictions generated from spatial hostpathogen models: the importance of population size and distance and the dynamic nature of disease. Whether these predictions also fit with the dynamics of disease in more isolated insular systems and in oceanic islands remain to be tested. Moreover, the dynamics of disease may be very different in other types of pathogens than in the biotrophic hostspecific pathogens studied in the Skeppsviks Archipelago. SUMMARY: FROM HERE AND BEYOND

Today we have a better understanding about diseases in natural plant communities than we had a mere decade ago. In particular, we have become aware that isolation and the spatial structure of host and pathogen populations can greatly affect the dynamics of plant-pathogen interactions. It has also become evident that pathogens may play dual roles in the dynamics and evolution of natural plant communities. The prevailing view is that plant pathogens are destructive organisms that reduce fitness of individual plants and cause declines of populations of particular species. However, diseases also play an important role in

PLANT DISEASE

751

maintaining genetic diversity within species and biodiversity at the community level. Their role is, however, difficult to assess without careful experimentation, but such strategies have so far been adopted only in a very limited number of studies. Further empirical work should also focus on improving our understanding about the importance of life history of hosts and pathogens for spread and the numerical and evolutionary dynamics of diseases. This knowledge is essential if we are to understand not only the risk and the challenges of plant diseases on islands and in mainland communities, but also how we conserve these often neglected organisms. SEE ALSO THE FOLLOWING ARTICLES

Biological Control / Deforestation / Dispersal / Metapopulations / Vegetation FURTHER READING

Burdon, J. J., L. Ericson, and W. J. Müller. 1995. Temporal and spatial changes in a metapopulation of the rust pathogen Triphragmium ulmariae and its host, Filipendula ulmaria. Journal of Ecology 83: 979–989. Burdon, J. J, P. H. Thrall, and L. Ericson. 2006. The current and future dynamics of disease in plant communities. Annual Review of Phytopathology 44: 19–39. Carlsson, U., T. Elmquist, A. Wennström, and L. Ericson. 1990. Infection by pathogens and population age of host plants. Journal of Ecology 78: 1094–1105. Carlsson-Granér, U., and T. M. Pettersson. 2005. Patterns of host susceptibility and disease occurrence in a metapopulation of Silene dioica. Evolutionary Ecology Research 7: 353–369. Carlsson-Granér, U., and P. H. Thrall. 2002. The spatial distribution of plant populations, disease dynamics and evolution of resistance. Oikos 97: 97–110. Gilbert, G. S. 2002. Evolutionary ecology of plant diseases in natural ecosystems. Annual Review of Phytopathology 40: 13–43. Hunter, D. G., and A. Shafia. 2000. Diseases of crops in the Maldives. Australasian Plant Pathology 29: 184–189. McKenzie, E. H. C. 1998. Rust Fungi of New Zealand: an introduction, and list of recorded species. New Zealand Journal of Botany 36: 233–271. Mill, M., D. Currie, and N. J. Shah. 2003. The impacts of vascular wilt disease of the takamaka tree Calophyllum inophyllum on conservation value of islands in the Granite Seychelles. Biodiversity and Conservation 12: 555–566. Parker, I., and G. S. Gilbert. 2004. The evolutionary ecology of novel plant-pathogen interactions. Annual Review of Ecology, Evolution and Systematics 35: 675–700. Trujillo, E. E. 2005. History and success of plant pathogens for biological control of introduced weeds in Hawaii. Biological Control 33: 113–122.

that the outer surface of the Earth is made up of thin, brittle tectonic plates, which have rigid interiors and interact only at their edges, where their relative motions produce earthquakes and volcanoes. By measuring the relative velocities of the plates in a finite number of places, their motions across the world and over the geological past can be computed using relatively simple geometric techniques. The assumption of rigid, undeformable plates breaks down to some extent in some continental areas but has proved to be a highly successful concept to describe the geology of the ocean basins. CONTINENTS, OCEANS, AND TECTONIC PLATES

The Earth can be divided into a number of concentric regions based on its composition: the crust (up to 70 km thick), the mantle (2900 km thick), and the core (3500 km thick). The crust can also be subdivided into thicker (30–70 km) continental crust, made of relatively low-density rocks such as sandstone, limestone, and granite, and thinner (3–7 km) oceanic crust, consisting mainly of denser basalt and gabbro (Fig. 1A). Thus the thicker continental crust “floats” higher than the thin but dense oceanic crust, by an average of about 4.6 km, thus forming the ocean basins. However, if we divide the Earth into layers on the basis of their mechanical strength instead of composition, we find an outer layer some 100 km thick, called the lithosphere, which constitutes the tectonic plates and comprises both crust and uppermost mantle. The lithosphere overlies a layer within the mantle that is still solid but somewhat weaker, called the asthenosphere (Fig. 1B). Lithosphere (solid, rigid, strong) Crust (O, Si, Al, Fe )

(O ,

, , Si

Mg

Lower Mantle (O, Mg, Si, Fe, in high-density forms)

Core (Fe, Ni, O)

PLATE TECTONICS

Asthenosphere (solid, ductile, weak)

s) form le ant nsity M e per -d Up in low Fe,

Mantle/Mesosphere (solid, slightly ductile)

Outer Core (liquid)

A

B

Inner Core (solid)

ROGER C. SEARLE Durham University, United Kingdom

FIGURE 1 Internal structure of the Earth. (A) Compositional subdivi-

Plate tectonics was developed in the 1960s and 1970s as the unifying, global theory of the Earth sciences. It assumes

752

P L AT E T E C T O N I C S

sions of the Earth. The most common elements in each layer are given in descending order of abundance. (B) Mechanical subdivisions of the Earth.

It is this weak asthenosphere that, over millions of years, allows the lithosphere both to adjust its depth and also to move around as described by plate tectonics. The lithosphere itself is divided into some 12 large, independently moving tectonic plates (plus a larger number of small or “micro” plates). It is important to note that a tectonic plate can contain both oceanic and continental crust; for example, the North American Plate comprises both the continental area of north America and the western half of the north Atlantic Ocean (Fig. 2). PLATES AND THEIR BOUNDARIES

The Earth’s mantle, though solid, is actually slightly ductile (like very strong toffee), and is very slowly convecting as it transports heat from the deep interior towards the surface. The tectonic plates actually represent the tops of these convection cells, where the Earth has become cool enough to behave in a brittle, rather than ductile, manner. The plates are in contact with each other everywhere on the Earth’s surface and are driven this way and that by the convection (like scum on boiling jam). Thus, the relative motions between plates may consist of one of three types of motion: divergence (at mid-ocean ridges), convergence (at subduction zones), and pure horizontal slip (at transform faults). Figure 2 shows the major plates with their boundary types and relative motions.

Mid-Ocean Ridges

At mid-ocean ridges the plates diverge, and the underlying asthenosphere is drawn up to fill the gap. As the asthenosphere material rises, the pressure drops and the melting point decreases, so that a small proportion melts to form liquid basaltic magma. This may rise to feed volcanic eruptions on the sea floor, with any residue being trapped below, where it cools to form diabase or gabbro and is incorporated into new lithosphere. Thus, new lithosphere is continually created at mid-ocean ridges. Approximately 3 km2 of new lithosphere is created this way every year. As the plates pull apart, they become fractured, and this cracking is manifested in small earthquakes. Because this new ridge is hot, it is quite buoyant, so the young lithosphere at ridges is relatively shallow. As the lithosphere moves away from the ridge, it ages, cools, becomes less buoyant, and sinks deeper, which is why divergent boundaries form ridges! Note, however, that these mid-ocean ridges are very broad features, extending the whole width of the ocean basins and deepening from less than 2900 m at their crests to over 5000 m on their flanks, which may be some 100 million years old. Most ridges are too dense to reach the sea surface, although in a few places, such as Iceland, the mantle is unusually buoyant and raises the ridges above sea level.

Eurasia North America Arabia Africa

Cocos

India

Philippine

Pacific

Pacific Nazca

South America Australia

Antarctica

FIGURE 2 Map of the world with boundaries, boundary types, and names of the major plates. Red: divergent boundaries or ridges; green: convergent

boundaries or trenches; blue: transform boundaries. Black dashed lines between North America and South America and between India and Australia indicate diffuse boundaries with very slow relative motion. Double arrows show relative motion directions across boundaries, but not relative speeds.

P L AT E T E C T O N I C S

753

The erupted lavas contain significant amounts of the naturally magnetic mineral magnetite (lodestone), and as they cool, they acquire a magnetization that is proportional and parallel to the Earth’s magnetic field. This field reverses its direction (i.e., north and south magnetic poles switch over) every few hundred thousand years, and the alternating field direction is recorded in the sea floor lavas. We can find the times at which the field reversed by dating rocks at the reversal boundaries, and can thus work out how rapidly the lithosphere moved away from the ridge axis—the so-called spreading rate—which helps us determine plate velocities.

Transform Faults

Where plates slide past one another without growing or being consumed, the boundary is called a transform fault. Many of these faults, such as the San Andreas Fault (separating the North American and Pacific plates) lie entirely within continental lithosphere. The longest ridge–ridge transform is the 1000-km-long Romanche Transform in the Equatorial Atlantic. The Alpine Fault in New Zealand is also a transform fault cutting through a continental island. Hotspots and Plumes

Subduction Zones

We have good reason to believe that the Earth is not expanding, so if new lithosphere is created at mid-ocean ridges, an equivalent amount must be lost somewhere. This occurs at convergent plate boundaries or subduction zones. Here, two plates collide and one is pushed back into the Earth’s interior; the boundary is marked by a deep trench, such as the 11-km-deep Mariana Trench. The motion of the descending plate causes earthquakes, including the largest, really destructive ones. As the plate descends, it heats up, and water that was trapped in the minerals making up the crust is released. This lowers the melting point of the mantle overlying the descending plate and causes it to start melting; the magma thus formed rises to feed volcanoes on the overriding plate. Because of the presence of the water and a somewhat different melting depth (pressure), the magma at subduction zones contains somewhat more silica than mid-ocean ridge basalts, and produces a rock type called andesite (socalled because it was first recognized in the Andes, in volcanoes formed above the Peru–Chile subduction zone). When one of the plates consists of continental lithosphere, it is too buoyant to subduct, and andesite volcanoes form on it above sea level. Examples are the Cascade volcanoes in western North America. If both plates are oceanic, the andesite will initially be erupted on the sea floor but will gradually build up on top of the plate, and eventually volcanoes will emerge above sea level. Because the subducting plate, and therefore the line of melting, is usually curved, these volcanoes form an arc referred to as an island arc. A good example is the Lesser Antilles at the eastern end of the Caribbean Sea, where the North American Plate is subducting westward beneath the Caribbean Plate and forming volcanic islands such as Guadaloupe, Dominica, and Martinique. Andesitic magma is relatively viscous and also contains large amounts of gas, so these volcanoes tend to be explosive and steep-sided—the classic volcanic cone.

754

P L AT E T E C T O N I C S

Although they are not plate boundaries, hotspots are an important cause of oceanic islands. In plate tectonic terms, a hotspot is a small geological region (as opposed to an extensive plate boundary) that is characterized by volcanic and seismic activity. They may lie on plate boundaries (Iceland is a good example), but often are far from the boundaries in mid-plate locations; the archetype of these is Hawaii. Hotspots are underlain by parts of the mantle that are unusually hot or, perhaps, have unusually low melting points so that they produce excess magma. It has been suggested that they may lie above upwelling plumes of hot mantle, but the depth from which they arise (whether a few hundred km down in the upper mantle, or perhaps 3000 km down at the boundary of the core) is still controversial. At these hotspots, the mantle melts to produce basaltic magma, similar to that produced at mid-ocean ridges (though with some subtle differences in the amounts of trace elements). This magma is eventually erupted onto the sea floor, forming volcanic seamounts that may grow large enough to rise above sea level. The Big Island of Hawaii, measured from its submarine base to the summit of Mauna Kea, is the world’s tallest volcano and is higher than Mount Everest. Basalt, being much less viscous than andesite and containing little gas, flows freely and produces shield-shaped volcanic domes with gentle slopes. If the hotspot persists for millions of years and is fixed in the Earth’s mantle, the lithospheric plate may drift over it so that the hotspot leaves a trail of volcanoes of steadily increasing age going away from it. As the plate ages and sinks, so will the volcanoes; if they were originally subaerial (i.e., above the surface) islands, they eventually sink to become underwater seamounts. This also explains the origin of atolls. Tropical volcanic islands typically develop fringing coral reefs; as the island sinks, its radius shrinks but the reef grows upwards, forming a barrier reef with a lagoon inside. Eventually, the island sinks completely, leaving just the reef as an atoll.

From the above discussion it will be realized that islands may have a number of different plate tectonic and geological origins and settings: continental fragments of all sizes, arc volcanoes at subduction zones, and hotspot volcanoes near ridges or in ocean basins.

of years. As the volcanoes age, they become eroded and, especially in the tropics, may develop steep, dramatic ridges and valleys as a result. In many hotspot volcanoes the flanks eventually collapse and fall into the sea, producing dramatic sea cliffs (such as the north coast of Molokai in Hawaii), while the process of collapse may generate tsunamis.

Continental Islands

SEE ALSO THE FOLLOWING ARTICLES

Some islands are actually whole continents, such as Australia or Greenland. Their geology will generally be a complex combination of different types of rocks, generated and altered over hundreds of millions of years, and so displaying a great deal of variability. Many continents have small islands just offshore, where a shallow waterway separates them from the mainland, with which they share a common geological structure (e.g., Long Island, New York; Sri Lanka; and countless smaller examples). Others are continental slivers that have been separated from their parent continents by plate motions and may have drifted far away from their origins. Examples are the Seychelles in the Indian Ocean (which, unusually for mid-ocean islands, display extensive outcrops of granite) and Madagascar, whose linear east coast reflects its past motion along a transform fault as it rifted away from Antarctica.

Continental Islands / Earthquakes / Island Formation / Oceanic Islands / Seamounts, Geology / Volcanic Islands

ISLANDS AND PLATE TECTONICS

FURTHER READING

Cox, A., and R. B. Hart. 1986. Plate tectonics—how it works. Oxford: Blackwell Scientific Publications, Inc. DeMets, C., R. G. Gordon, D. F. Argus, and S. Stein. 1990. Current plate motions. Geophysical Journal International 101: 425–478. DeMets, C., R. G. Gordon, D. F. Argus, and S. Stein. 1994. Effect of recent revisions to the geomagnetic reversal timescale on estimates of current plate motions. Geophysical Research Letters 21(20): 2191–2194. Isacks, B. L., J. Oliver, and L. R. Sykes. 1968. Seismology and the new global tectonics. Journal of Geophysical Research 73: 5855–5900. Kearey, P., and F. J. Vine. 1996. Global tectonics, 2nd edition. Oxford: Blackwell Science. Sella, G. F., T. H. Dixon, and A. Mao. 2002. REVEL: A model for recent plate velocities from space geodesy. Journal of Geophysical Research 107(B4): ETG 11, 1–32. Tackley, P. 1996. Mantle convection and plate tectonics: toward an integrated physical and chemical theory. Science 288: 2002–2007. Vine, F. J. 1966. Spreading of the ocean floor: new evidence. Science 154: 1405–1415.

Arc Volcanoes

Many oceanic islands are arc volcanoes above subduction zones, and all these lie near and parallel to deep-sea trenches. There are numerous examples in the Lesser Antilles, the South Sandwich Islands, and the western Pacific. All of these are based on andesite volcanoes, so they typically have active or dormant volcanoes with steep sides, explosive eruptions, and associated earthquakes. Such volcanoes, especially in the tropics, can produce rich, volcanic soils, but they also pose considerable hazards from volcanic eruptions (lava and ash flows) and mud flows.

POCKET BASINS AND DEEP-SEA SPECIATION BRUCE H. ROBISON Monterey Bay Aquarium Research Institute, Moss Landing, California

Hotspot Volcanoes

Most islands within the deep ocean basins are hotspot volcanoes. They may lie on mid-ocean ridges (e.g., Iceland), near ridges (the Azores, the Galápagos), or far from them in the plate interiors (e.g., the Hawaiian, Marquesas, and Friendly Islands in the Pacific, the Canaries and Tristan da Cunha in the North and South Atlantic, and Réunion and Mauritius in the Indian Ocean). These volcanoes are basaltic, so they have low slopes and are shield-shaped. Islands with active volcanoes may have significant areas of recent lava flows with no vegetation, although vegetation soon establishes itself in a matter

WILLIAM M. HAMNER University of California, Los Angeles

Isolation is an important agent in the evolution of new species. When two populations of a single species become sufficiently isolated that there is no exchange of genetic material, then random genetic mutations and genetic drift over time will eventually render them distinct from each other. Thereafter, if the environmental conditions that affect these two populations begin to differ, genetic separation (speciation) can proceed even faster. Island

P O C K E T B A S I N S A N D D E E P - S E A S P E C I AT I O N

755

chains at and near the sea surface and pocket basins in the deep seafloor provide the isolation necessary for speciation in the ocean. NATURAL LABORATORIES

Scientists often rely on the manipulation of natural systems or processes in order to understand them, but the vast size of oceanic ecosystems makes it clearly impossible to conduct experimental research at this scale. Instead, we must look for naturally occurring manipulations, for natural laboratories where the local variability of environmental conditions provides us with the equivalent of experimental results. The most familiar examples of these processes occur on islands like the Galápagos, where populations of related species have evolved into different forms, shaped by the particular conditions on the individual islands they inhabit. In the deep ocean, however, there are relatively few barriers to genetic exchange because the great basins of the Pacific, Atlantic, and Indian Oceans extend for thousands of kilometers and their deep-water environments are relatively homogeneous. Consistent with this global pattern is the substantial number of cosmopolitan deepsea species, those found in deep-sea basins worldwide. For shallow-water marine animals as well, speciation is clearly related to the isolation provided by islands. The area of the Indo-Pacific bounded by the Philippines, New Guinea, Indonesia, and Borneo has the highest density of islands on earth. Within this area there are more shallowwater marine species than anywhere else in the sea. These islands are correctly considered to be the “cradle of marine biodiversity.”

the rim around a basin, has an important effect as a barrier to the exchange of genetic material between the deep-living animals that occupy adjacent basins. For example, the bottom depth of the Sulu Sea is 5200 m, but the deepest place along the rim of its basin is only 420 m. This means that unless a deep-living species can tolerate the decompression associated with depth changes of perhaps several thousands of meters, it cannot exchange genetic material with a related population in a neighboring basin. DEEP-SEA LIFE CYCLES AND SPECIATION

Many deep-living species do have eggs that float upward so that their larvae can develop in the highly productive waters near the surface. But after initial development at shallow depths, the maturing juveniles move back down to the greater depth range of the adults. This ontogenetic or developmental vertical migration allows many deep-living species to broadcast their eggs widely, some of which are carried by surface currents to surrounding seas. Yet other species have eggs and larvae that remain at depth or have particular requirements that can be met only locally, and these species remain genetically isolated in their natal pocket basins.

POCKET BASINS BENEATH INDO-PACIFIC SEAS

For many kinds of deep-sea animals, the Indo-Pacific is also considered to be the center of origin and distribution. This is where species diversity is the highest, and again, speciation is clearly linked to the multitude of islands. In this case the islands and the submerged ridges between them provide barriers to the exchange of genetic material between species that inhabit deep water. Between many of the island groups there are pocket basins, areas of the seafloor that are thousands of meters deep yet isolated from each other by islands and submerged ridges. Typically, the ocean area above a pocket basin, circumscribed by islands and ridges, is called a sea. Familiar examples include the Sulu Sea of the Philippines and the Banda Sea of Indonesia. While the bottom of a pocket basin may be several thousand meters deep, the sill depth, or the deepest part of

756

P O C K E T B A S I N S A N D D E E P - S E A S P E C I AT I O N

FIGURE 1 Deep-sea fishes: (A) lanternfish, Myctophum, with small,

rounded bioluminescent organs on its flanks and along its ventral surface; (B) hatchetfish, Argyropelecus, with elaborate light organs along its underside, and upward-directed eyes.

For example, fishes of the family Myctophidae (“lanternfish” because of their light-producing organs, Fig. 1A), live in deep, dark waters during the day, hundreds of meters beneath the sea surface. At night they migrate upward to feed near the surface during the hours of darkness. Their bioluminescent organs are used to find prey, find mates, and avoid predators. Lanternfish are found worldwide, and they belong to one of the most speciose of all fish families. The center of their geographical distribution, and the place where more myctophid species have been recorded than anywhere else, is the Indo-Pacific region off Southeast Asia. Similar patterns of distribution and species richness exist for many other deep-sea fish families, including hatchetfish (Fig. 1B), dragonfish (Fig. 2A), and anglerfish (Fig. 2B). Although the fishes in each of these families have different feeding patterns, depth ranges, and life histories, they all appear to have originated in the Indo-Pacific and have speciated widely. The Sulu and Banda Seas lie among the narrow maze of twisted trenches and pocket basins that link the great basins of the Indian and Pacific Oceans (Fig. 3). These interlocking depressions in the sea floor provide the only deep waters between the continental shelves of Australia

FIGURE 3 Bathymetry of the Indo-Pacific region. Three pocket basins

beneath the Sulu, Celebes, and Banda Seas are circumscribed by islands and submerged ridges that isolate their deep-dwelling inhabitants. Satellite altimetry data: Geoware, GMT Companion CD-R Vol. 1 Version 1.9, June 2006. Image prepared by Jenny Paduan and David Clague, MBARI.

and Southeast Asia. The many seas of this region are filled with water from the western Pacific, but they are hydrographically distinct. The numerous island groups and sills that separate the seas also restrict the flow of subsurface waters between them, and these semi-isolated basins with shallow sills provide the reproductive isolation needed for the speciation of deep-sea animals. In the complex topography of the seafloor in the IndoPacific region the many deep pocket basins are the inverse of islands at the sea surface. They isolate populations of related deep-sea animals from one another, thereby leading to evolutionary speciation. SEE ALSO THE FOLLOWING ARTICLES

Cold Seeps / Hydrothermal Vents / Indonesia, Geology / Marine Lakes / Philippines, Geology FURTHER READING

FIGURE 2

Deep-sea fishes: (A) dragonfish, Idiacanthus, with a lumi-

nous chin barbel for attracting prey; (B) anglerfish, Caulophryne, with a luminous lure, small eyes, and elongate fin rays for detecting movement in the dark waters it inhabits.

Briggs, J. C. 1974. Marine zoogeography. New York: McGraw-Hill. Herring, P. J. 2002. The biology of the deep ocean. New York: Oxford University Press. Koslow, J. A. 2007. The silent deep. Chicago: University of Chicago Press. Robison, B. H. 1995. Light in the ocean’s midwaters. Scientific American 273: 60–64. Robison, B. H. 2004. Deep pelagic biology. Journal of Experimental Marine Biology and Ecology 300: 253–272.

P O C K E T B A S I N S A N D D E E P - S E A S P E C I AT I O N

757

European observations, and linguistic, archaeological, and experimental data concerning the sequence, timing, and capabilities of Polynesian voyaging.

POLYNESIA SEE PACIFIC REGION

TRADITIONS AND HISTORICAL OBSERVATIONS

POLYNESIAN VOYAGING ATHOLL ANDERSON Australian National University, Canberra

Polynesia consists of Samoa and Tonga (West Polynesia), settled initially by Lapita voyagers about 3000 years ago, and the dispersed archipelagoes of East Polynesia, especially Hawaii, French Polynesia, Easter Island, Cook Islands, and New Zealand, colonized 1100–700 years ago (Fig. 1). The term “Polynesian voyaging” refers to the means by which island colonization was effected and the extent to which interaction occurred between distant islands. One extreme of opinion envisages exclusively accidental colonization by one-way voyaging that precluded development of longrange interaction, while at the other extreme, purposeful and navigated voyaging within a strategic system of colonization, multiple contact, and interaction is proposed. The evidence at issue consists of Polynesian traditions, early

20 o

1

3

4

0o

3

2

20 o

3

5 40 o

140 o

160 o

180 o

160 o

140 o

120 o

FIGURE 1 Colonization (thick arrows) in relation to prevailing wind direc-

tions (thin arrows): 1 = West Micronesia 1500 BC, probably from Philippines; 2 = Lapita colonization, from New Guinea islands, reaches West Polynesia 1000 BC; 3 = colonization of East Micronesia and West Polynesian marginal islands 200 BC from Tonga or Samoa; 4 = colonization of East Polynesia AD 1000 from West Polynesia; 5 = colonization of South Polynesia (New Zealand region) AD 1200.

758

P O LY N E S I A N V O YA G I N G

Polynesian traditions of origin are varied, but most refer to ultimate ancestry in mythical homelands, notably Hawaiki, which is reflected in island names such as Hawaii, Havaii (Raiatea in the Society Islands), and Savaii (in Samoa). Voyaging traditions are common in Polynesia, and where they have been studied in detail, notably in New Zealand, it is apparent that spare, enigmatic early nineteenthcentury records had been elaborated subsequently by indigenous and European scholars. Stories about fleets of canoes, detailed descriptions of particular voyages, and precise navigational instructions, including methods, cannot be traced back to reliable early traditions. However, a general congruence of traditional genealogies among widely dispersed tribal populations in New Zealand and, to some extent, between marginal and central East Polynesia, does suggest that widespread colonization voyaging occurred in East Polynesia during a period 20–30 generations before AD 1800; that is, about AD 1100–1300. There are no historical records (sixteenth to early nineteenth centuries) of Polynesian voyaging; that is, no voyaging canoes were observed far at sea between archipelagoes. The records consist, instead, of numerous observations of Polynesian canoes sailing and ashore, and some accounts of Polynesian geography. Of the latter, the map constructed for James Cook in 1769 according to information from the Raiatean scholar Tupaia is the most important. It contains islands between West Polynesia and the Marquesas, although Tupaia claimed no first-hand knowledge beyond the Societies and nearer Australs. His map and other contemporary information from Polynesians suggest that by the late prehistoric era there was frequent interaction within each archipelago, infrequent interaction between them, and no contact between the central (now French Polynesian) groups and the marginal groups of Hawaii, Easter Island, and New Zealand. Historical observations record that “double canoes” were mostly used for inter-island travel. There was considerable diversity in their rigging. The oceanic lateen sail, probably derived from Micronesia and ultimately of Indian Ocean origin, was common in West Polynesia (Fig. 2). In the seventeenth century it was rigged in a rudimentary form, but by the late eighteenth century it conformed to the Micronesian style, in which, to put

FIGURE 3 A Hawaiian double canoe with an oceanic spritsail. EngravFIGURE 2 A Tongan double canoe with an oceanic lateen sail. Photo

lithograph of ink drawing by Isaac Gilsemans, 1642, at Tongatapu Island, titled, in translation, “Our Ships at Anchor in the Roadstead.” Published with permission from the Alexander Turnbull Library,

ing of drawing by John Webber, 1788, at Hawaii, titled “Tereoboo, king of Owhyee, bringing presents to Captain Cook.” Published with permission from the Alexander Turnbull Library, Wellington, New Zealand, C-131-061.

Wellington, New Zealand, PUBL-0106-001.

the vessel about, the mast was canted over and the tack point (or tack) of the sail moved from one end to the other, a maneuver called “shunting.” In East Polynesia, lateen sails were rare, and the main type was an oceanic spritsail slung, point down, between two upright spars (Fig. 3). Generally, one spar was used as, or attached to, a mast, but in New Zealand and possibly in the Marquesas the earliest records, although not definitive, suggest that the spars were free-standing, held up by wind pressure against the sheets and used only before the wind. Navigation methods were not recorded in any detail, except to note that the moon and stars were used in some way. They probably enabled estimation of latitude, to which dead reckoning and use of land-finding indicators, such as swell patterns and homing seabirds, provided some additional aid. LINGUISTIC, ARCHAEOLOGICAL, AND EXPERIMENTAL SEAFARING EVIDENCE

Comparison of Polynesian languages shows that protoPolynesian split into three groups: proto-Tongic, proto– nuclear Polynesia (mostly in the Polynesian outliers), and proto-Ellicean, which comprised mainly the Samoan language and Eastern Polynesian. The latter, in turn, split into a central eastern group (Tahitian, western Australs and Tuamotu, Cooks, Maori, and, slightly different, Marquesan and Hawaiian) and a southeastern group comprising the languages of Mangareva, Easter Island, and the eastern Tuamotu and Australs. This sequence is mirrored approximately in the archaeological data of initial island colonization. These show a considerable “pause” in colonisation between Samoa–Tonga at 3000 years ago and marginal West Polynesian islands such as Niue and

Tuvalu about 2200 years ago; another until the earliest settlement of the Societies and Cooks at 1100–1000 years ago; and then a briefer pause before colonization of the subtropical islands such as Rapa and Easter Island and of temperate New Zealand about 800 years ago. The overall pattern of island colonization is west to east, center to periphery, and clearly episodic rather than continuous. Two canoe planks and a steering oar from Huahine island, Societies, dated to about AD 1000, constitute the best, and almost the only, direct archaeological evidence of possible voyaging canoes. Notably absent is any evidence of early voyaging sails and rigging, apart from a few depictions of Polynesian spritsails in rock art, especially in Hawaii. Historical linguistic reconstructions indicate that the double canoe came into existence in West Polynesia, its cargo-carrying capacity perhaps a critical prerequisite for further voyaging to the east. However, terms for “mast” and “standing rigging” are not known for early Polynesian languages. Chemical analysis of transported adzes to establish stone sources can provide useful information about prehistoric movement, but in some cases where long-distance movement is asserted, it is apparent that the adzes are not from prehistoric contexts. At present, stone sourcing studies confirm frequency of movement across the central Polynesian islands but the isolation of marginal archipelagoes such as Hawaii, New Zealand, and Easter Island. One way of attempting to overcome the paucity of direct evidence for Polynesian voyaging is by experiment or simulation. This has taken two forms. The first consisted of building and sailing ocean-going vessels regarded as “performance-accurate” in terms of potential prehistoric types. The most famous of these was the Kon-Tiki, built in the form of a Peruvian sailing raft and sailed to Polynesia in

P O LY N E S I A N V O YA G I N G

759

1947 by Thor Heyerdahl in support of his contention that South America represented a major source of Polynesian culture. Another well-known experimental vessel is the Hokule’a, a double canoe, inspired by Ben Finney’s belief in the capabilities of prehistoric seafarers, which sailed from Hawaii to the most distant points of Polynesia. The other form of experimental sailing, increasingly preferred since its inception in 1973 by Levison, Ward, and Webb, has been voyaging by computer simulation. The core idea is to map the frequency of wind directions across the Pacific, then put numerous virtual canoes with defined sailing characteristics to sea from different islands in order to determine which potential routes were more likely used, the relative rates of successful landfall, and the probable sequence of island discovery. Experiment and simulation have shown, in general, that simple drifting would probably not have been sufficient to people the remote Pacific islands, but that a sophisticated voyaging ability, including weatherly vessels and astral navigation, should have led to faster and more continuous colonization. Prehistoric voyaging capabilities lay somewhere in between. MODELS OF POLYNESIAN VOYAGING

In the light of these various sources of evidence, it is broadly agreed that the overwhelming contribution to Polynesian populations and cultures came from the western Pacific. Claims for an American influence are not generally accepted, although close similarities in aspects of material culture between South America and Easter Island, and the existence of South American sweet potato in prehistoric Polynesia, suggest some level of contact. As large, capable sailing rafts existed in Ecuador at the time of Spanish arrival, it is quite possible that they were the agent of transport, rather than Polynesian canoes. Chicken bones discovered recently in a Chilean site are not of any distinctive Polynesian type, and as they date very close to the Spanish era, the evidence should be regarded cautiously. Claims of Polynesian voyaging to North America are implausible. Within mainstream scholarship, the various kinds of evidence have been constructed into widely differing models. The most prominent from the late nineteenth to the mid-twentieth century was “traditionalism.” Represented in the writings of Sir Peter Buck (Te Rangi Hiroa), for example, it accepted the expanded corpus of indigenous traditions at face value and assumed, as in traditional thought, that the abilities of the ancestors in seafaring exceeded those of their descendants. Advanced navigation techniques and frequent long-distance voyaging and interaction between remote islands were regarded as characterizing prehistoric voyaging. In reaction to this

760

P O LY N E S I A N V O YA G I N G

perspective, Andrew Sharp argued that many traditions were corrupted, Polynesian geographical knowledge was limited, advanced navigation was impossible, and Polynesian canoes suffered problems in sailing, seaworthiness, and seakeeping. His incisive attack served, however, to inspire a revival of traditionalism, as in the work of Ben Finney. “Neotraditionalism” uses the traditionalist assumptions to argue that Polynesian voyaging technology declined once most of the islands were colonized and that historical evidence is thus only a pale reflection of earlier prehistoric ability. Therefore, the original Polynesian voyaging canoes must have been more advanced than any recorded historically. This maxim was put into practice in building and sailing experimental canoes such as Hokule’a, which combined the most advanced of widely dispersed Polynesian technologies; Micronesian navigation techniques; European buoyancy, fastenings, and rigging; and sails that were twice as large, relative to waterline length (a proxy for displacement, which cannot be estimated independently from historical data), than on eighteenth-century Polynesian double canoes of similar size. The experimental sailing data are therefore at the most liberal extreme of probability. Nevertheless, in the absence of alternatives, they have been used in voyaging simulation studies, notably by Geoffrey Irwin, to argue that such fast, weatherly vessels were used strategically in exploration; sailing first toward the prevailing wind, which allowed easier return, and only later, as uninhabited islands became few, in more difficult directions across and before the prevailing wind. Recently, opinions have been changing again. Neotraditionalism remains the preferred model among indigenous scholars and in popular opinion, as shown in the recent Vaka Moana exhibition (Auckland Museum, New Zealand) and book, but some archaeologists have tried to rethink the problems of understanding Polynesian voyaging from more basic starting points. I have returned to the early historical descriptions to argue that they show not a supposed devolution of sailing technology following an age of exploration but rather a late evolution (for example in the introduction of the lateen sail, which encouraged greater use of fixed masts and standing rigging), and therefore that, contrary to traditionalist assumption, earlier technology might have been more rudimentary. If it was largely confined to sailing before the wind, then the episodic nature of Polynesian dispersal eastward could reflect approximately correlated periods of high El Niño frequency, notably about 3000 and 1000 years ago, in which easterly winds were reversed into tropical westerlies. Some simulation studies have also returned to a

more basic model in which only rudimentary sailing ability (drifting, paddling) is assumed. This can be shown as sufficient to reach west Polynesia but inadequate for east Polynesian conditions. Even in allowing assumptions of greater ability (e.g., weatherly sailing), recent simulations show that voyaging in East Polynesia was difficult.

Golson, J., ed. 1963. Polynesian navigation: a symposium on Andrew Sharp’s theory of accidental voyages. Wellington: The Polynesian Society. Howe, K. R., ed. 2006. Vaka Moana: voyages of the ancestors. The discovery ansd settlement of the Pacific. Auckland: David Bateman. Irwin, G. J. 1992. The prehistoric exploration and colonisation of the Pacific. Cambridge, UK: Cambridge University Press. Levison, M., R. G. Ward, and J. W. Webb. 1973. The settlement of Polynesia: a computer simulation. Minneapolis: University of Minnesota Press. Sharp, A. 1956. Ancient voyagers in the Pacific. Harmondsworth: Penguin.

CONCLUSIONS

The difficulty of understanding Polynesian voyaging arises most directly from the virtual absence of direct evidence, such as archaeological remains of boats and rigging. Were these more abundant, and the form and capabilities of seafaring, which is the critical practice of voyaging, much better known, most of the current arguments would be resolved. As it is, the burden of debate is carried by experimental and simulated sailing and the comparison of results from those against archaeological evidence of island colonization patterns. At present, this method shows that Polynesian voyaging was more difficult than envisaged in traditionalist models, except by assuming the existence of very sophisticated watercraft of types used experimentally but not recorded historically. Most East Polynesian canoes were capable off the wind but unsuited to sailing long distances against it. In that circumstance, the episodic nature of Polynesian dispersal, and the scarcity of evidence for frequent interaction over long distances, seem readily understandable. The larger questions about what actuated episodes of voyaging remain unanswered. Was it improvements in technology, periodic population growth and social pressure (as reflected in traditional references to exile), changing climate, or mere chance? These propositions need much research before it will be possible to write a more satisfactory account of Polynesian voyaging. SEE ALSO THE FOLLOWING ARTICLES

Archaeology / Exploration and Discovery / Human Impacts, Pre-European / Kon-Tiki / Peopling the Pacific FURTHER READING

Anderson, A. J. 2000. Slow boats from China: issues in the prehistory of Indo-Pacific seafaring, in East of Wallace’s Line. S. O’Connor and P. Veth, eds: Leiden: Balkema, 13–50. Anderson, A. J., J. Chappell, M. Gagan, and R. Grove. 2006. Prehistoric maritime migration in the Pacific Islands: an hypothesis of ENSO forcing. The Holocene 16: 1–6. Buck, Sir P. 1954. Vikings of the sunrise. Christchurch: Whitcombe and Tombs. Di Piazza, A., P. Di Piazza, and E. Pearthree. 2007. Sailing virtual canoes across Oceania: revisiting island accessability. Journal of Archaeological Science 34: 1219–1225. Finney, B. R. 1979. Hokule‘a: the way to Tahiti. New York: Dodd, Mead and Co.

POPULAR CULTURE, ISLANDS IN VINCENT H. RESH University of California, Berkeley

JONATHAN P. RESH Undaunted Design Co., Chicago, Illinois

Popular culture is the culture of the people. It includes the fashions, movies, television shows, advertising, and even video games that are easily accessible to individuals with a wide variety of social backgrounds. Emphasis in popular culture is on instant accessibility; no prior or profound knowledge is required. The use of islands in popular culture is an excellent example of how media can use strong images and perceptions to influence people’s taste and behavior. ISLAND IMAGES AND INFLUENCES

Island images are generally positive—they emphasize sensuality, escape, solitude, seduction, and self-sufficiency. However, negative images are there as well—islands also can be lonely, inhospitable, forbidden, or mysterious (Fig. 1). Islands can elicit stereotypical responses strictly on their location. For example, Arctic islands are typically depicted as inhospitable and isolated, while Caribbean islands are friendly, plentiful, and carefree. No islands, however, can evoke romance and allure as those of the South Seas do. Tahiti, like Timbuktu, is almost a “brand,” producing instant responses in the human imagination. Advertisers capitalize on the charms of tropical islands in the Caribbean and the South Pacific to promote travel getaways or casual clothing or just as a mechanism for escape and fantasy. The message in selling the island experience is clear—islands will do for you what continental life cannot, and people can come to islands either to find themselves or to lose themselves.

P O P U L A R C U LT U R E , I S L A N D S I N

761

Even though the participants in these more recent shows choose to be stranded, these are not island experiences for the vacation-bound. Islands have been locations for other popular television shows as well, including Hawaii Five-O, Hawaiian Eye, and Magnum PI, all based in Hawaii, with Temptation Island filmed in the Caribbean and Fantasy Island, Man from Atlantis, and Lost set on fictional islands. In all these shows, some of the benefits of island living were presented, although crime and human frailties were always present as an undercurrent. Television shows have also contributed catch-phrases to popular culture, sometimes becoming painful with their repetition but fortunately short-lived such as “Book ’em, Dano” from Hawaii Five-O, “The plane, the plane” from Fantasy Island, and “Voted off the island” from Survivor. MOVIES

FIGURE 1 Artist’s rendering of the dual nature of islands as depicted in

popular culture. Drawing by Jonathan Resh.

Island images and influences are present in many aspects of Western popular culture, including the media, food, drinks, restaurants, and the arts. TELEVISION

Clearly, the archetype of television shows featuring islands is Gilligan’s Island. Universally ridiculed by critics, this 1960s television show was produced for only four years, but reruns of its 98 episodes continue to be seen, and cartoons and even movies have revived the story. A recent book (Gilligan Unbound by Paul A. Canto) has actually maintained that this television series mirrors the beliefs of 1960s America and is a “window” into the liberal democratic culture that was present at that time. More recently, the first popular reality television show in the United States, Survivor, was first set on an uninhabited island off Borneo, and many islands, including the Marquesas, Pearl Islands, Cook Islands, Palau, Vanuatu, and Fiji, have been the location of subsequent installments. Concurrently, British television featured Castaway 2000, in which contestants spent months living on an uninhabited island off the Scottish coast. Unlike the (sometime) comedic message of Gilligan’s Island, these shows emphasize the danger of islands, where contestants are forced to be resourceful in order to “survive” in the island wild. They also serve as a populist social experiment, showing viewers how a select society may react when confronted by the forced isolation of islands.

762

P O P U L A R C U LT U R E , I S L A N D S I N

Well over a thousand movies have the word “island” appearing in their title. However, rather than drawing on real islands, many of these films draw on the metaphor of island isolation or loneliness. There are, however, hundreds of movies that do not have “island” in the title but take place on islands. These latter films evoke the idea of survival or the noble savage (Robinson Crusoe, Castaway), freedom from adults (Blue Lagoon, Lord of the Flies), an idyllic paradise interrupted by an outside event (the arrival of Europeans or missionaries as in Mutiny on the Bounty and Hawaii, respectively), and internal conflicts or disagreements (Rapa Nui and Tabu). Of course, World War II movies that recount great battles on islands of the Pacific (Guadalcanal Diary and Tora! Tora! Tora! ), and that contrast beautiful island settings with the atrocities of war, have long been a staple of Hollywood. Like other forms of popular culture, cinema has also explored the duality of island images, exploiting both the positive and the darker sides. Beyond loneliness, films set on islands have suggested the possibility of being eaten (Cannibals of the South Seas), visiting the lair of villains (Dr. No, The Island of Dr. Moreau), and even encounters with nonhuman monsters (King Kong and Jurassic Park). More recently, the Pirates of the Caribbean series has added a supernatural element to island adventures on film. The 1958 movie South Pacific gave us one of the most romantic movie images ever created—the mythical Bali Hai—which like Shangri-la (set in the Himalayas) in Lost Horizon has become synonymous with paradise, love, and transformation. The key song in that musical evokes this feeling by suggesting that people live on a lonely island and long for another, special, island. The name Bali Hai

has been applied to resorts throughout the Pacific, sometimes thousands of miles apart. LITERATURE AND ART

Although movies and television are more recent additions to the popular culture canon, islands have figured prominently in literature for centuries (The Odyssey, The Tempest). The island novel that is foremost in many readers’ minds is Daniel Defoe’s Robinson Crusoe. Based on the marooning of a Scottish seaman on Màs a Tierra (400 miles from Chile), it changed the images of islands from fearful places (albeit lonely ones) to sites of redemption and freedom. Since then, several novels set on islands that draw on traditional island motifs of romance and danger, including Ballantyne’s The Coral Island, Stevenson’s Treasure Island, Wells’ The Island of Dr. Moreau, Golding’s Lord of the Flies, Huxley’s Island, Charrière’s Papillon, Verne’s The Mysterious Island, and Garland’s The Beach. Although these stories have taken place on islands located throughout the world (e.g., The Beach was based on one of the islands off the coast of Thailand), many of the great island novels are based on stories of the South Pacific. The South Pacific has spawned a rich number of stories that have become popular lore. American writers over successive generations, from Melville, Stevenson, and London to Nordhoff and Hall to Michener, along with writers of other nationalities such as Burke (Australia) and Loti (France), have contributed to the island fantasy. A recent trend has been to revisit islands and sites from previous voyages (e.g., Horowitz’s Blue Latitudes retraces Cook’s voyages) or of those evoking strong popular culture images (Theroux’s Happy Isles of Oceania). Unfortunately, some of this genre seems to be searching for a zoolike paradise, and comments about modern life on islands range from being condescending to mean-spirited when the authors discover that DVD players have joined traditional culture. Paul Gauguin, probably the artist most closely linked to the image of island life, is also an example of the linkage between high culture and popular culture. Now recognized as one of the great impressionist painters, he went from France to Tahiti following a quest to find the “noble savage” and to live the free life he envisioned away from continental civilization. His last home on Hiva Oa in the Marquesas Islands, which he named “Le Maison de Jouir,” matches the images of a hedonistic life that drew him to these islands. Depictions of his paintings and sculptures are instantly recognizable as “Gauguins,” whether on a T-shirt or parodied in advertisements. The novel by Som-

erset Maugham and subsequent movie Moon and Sixpence, based very loosely on Gauguin’s life, helped create the popular mystique of the islands as a source of creative muse for the artist, and also the “free love and free coconuts” lifestyle that can supposedly be found there. FOOD, DRINK, AND RESTAURANTS

Just as sight and sound convey images of islands in television and movies, culinary tastes may reflect images of islands. Perhaps not unexpectedly, many of these tastes are actually continental creations. For example, from the 1930s to the present day, Polynesian-island themed bars (often referred to as “Tiki bars”) and restaurants have appeared with names like “Tiki Village,” “Don the Beachcomber’s,” and “Trader Vic’s.” By providing decorative escapism and an assortment of rum-based drinks with exotic names such as Black Widow, Zombie, Missionary’s Downfall, Samoan Fog Cutter, and Mai Tai, they have captured the exotic spirit of island mythology. Foods served in Tiki bars, which traditionally have contained non-Polynesian ingredients such as cooking sherry and Worcestershire sauce, reportedly were the origin of the Chinese “take-out” food widely consumed today. Décor of the Tiki bars included (and still includes) fishing nets and floats, A-frame design, masks, and spears (Fig. 2). While many of these Tiki bars have disappeared, reappeared, and disappeared again, others like the “Tonga Room” and “Trader Vic’s” in San Francisco continue the tradition of the Tiki bar/restaurant. One of the interesting aspects of the design of many of these island-themed restaurants is that oftentimes patrons have to cross a bridge (sometimes over a gurgling stream) to get from the entrance to the restaurant interior. Some have suggested that this represents a “symbolic crossing” and escape from reality to fantasy.

FIGURE 2 Interior of an “ultimate” tiki restaurant in Emeryville,

California. Photograph by Cheryl Resh.

P O P U L A R C U LT U R E , I S L A N D S I N

763

MUSIC AND DANCE

The often idyllic and exotic styles of “island music” also have become part of popular culture. Arguably, Jamaican music has had a greater influence on worldwide popular culture than the music of any other island. Interest and appreciation of reggae music, as well as its precursors (e.g., rocksteady, ska) and its derivatives (e.g., dancehall, dub), make reggae one of the few aspects of island popular culture that have extended into both developed and developing countries. The word “reggae” is practically synonymous with Jamaica, and the icon for the reggae movement remains the late Bob Marley. His songs are as likely to play on classic rock radio stations as they are on TV commercials or as background in elevators, and his face is as common on T-shirts as it is on music covers. Marley and the Rastafarian lifestyle associated with reggae spread the music’s popularity largely through its distinctively colorful, countercultural aspects (as portrayed by Jimmy Cliff in the cult film The Harder They Come). While the dreadlock hairstyle, open use of marijuana, and even tricolor flag—all mainstays of reggae style—had their roots in Jamaica’s cultural upheaval, they have since been widely adopted by non-Jamaicans seeking to express a sense of acceptable nonconformity. Other Caribbean islands have equally rich musical traditions, some of which developed from century’s-old African and European influences. The beginnings of calypso in Trinidad purportedly were a musical means of communicating information among working slaves because conversations in the fields were banned. But by the 1950s, modern calypso, enhanced by the steel-pan drum, became a worldwide craze, largely from Harry Belafonte’s “Banana Boat Song.” A dance/parlor game, the limbo, also became associated with music from the West Indies. In terms of culture proceeding in the opposite direction, other islands altered Western music to create their own native sound, most notably Cuba, as popularized by the film Buena Vista Social Club. Polynesian music, mostly emanating from the tourist areas of Hawaii, is also immediately recognizable. The first broad exposure to Hawaiian music was at the Panama-Pacific International Exhibition in San Francisco in 1915, introducing the slack-key guitar and the jaunty strumming of the ukulele. Hawaiian music, or at least Americanized versions of it, was an immediate hit. Hundreds of Hawaiian-type songs were penned by American musicians, with such titles as “Oh How She Could Yacki Hacki Wicki Wicki Wo,” “On the Beach at Waikiki,” and “Lovely Hula Hands.” Popular singers

764

P O P U L A R C U LT U R E , I S L A N D S I N

like Bing Crosby and Jimmie Rodgers added Hawaiian sounds to their repertoire. Hawaiian music had a particularly strong effect on country music, which incorporated the smooth tones onto steel guitar (best exemplified by masters such as Jerry Byrd and Chet Atkins), and still heard in the genre today. The ukulele was popularized in the United States by Bing Crosby and Arthur Godfrey, and it often had a pineapple shape that both reflected an island image and produced a “warmer sound.” The height of the Hawaiian Islands’ music popularity was arguably Don Ho’s “Tiny Bubbles” in 1966. By any measure, the song was a diluted offering of true Hawaiian styles, but it brought attention to the island’s sounds, with Ho himself serving as a figurehead for that style. Closely associated with Hawaiian music is the hula, a traditional dance with both ancestral and derivative versions seen throughout the Pacific Islands but universally recognized as a distinct cultural attribute of the Hawaiian Islands. Popularized variations of the dance’s image have been spread through depictions in movies, television, and tourist spectacles, as well as the ever popular hula hoop, hula costumes (e.g., grass skirt, leis), figurine dolls (often placed on car dashboards), and even hula girl tattoos. Few other islands in the world have had as much pop-culture impact as the Caribbean and Pacific islands, with two notable exceptions: Ibiza and Ireland. Although the Mediterranean islands have long had an association with music (including the irresistible song of the Sirens from The Odyssey), the Spanish island of Ibiza became a tourist destination precisely because of its music. In the 1980s, with the rise of electronic dance culture and techno/raves, Ibiza started many trendy nightclubs. By the 1990s, its reputation as a hedonistic, sun-drenched, and music-saturated paradise gave Ibiza iconic status as a popular culture sanctuary for both partying and music (as depicted in It’s All Gone Pete Tong). Over the last quarter-century, Ireland has been the island with some of the most popular and successful musical groups. Many of these groups maintain ties to the country’s older roots by drawing upon Celtic melodies and rhythms. For example, the band U2 has embedded a distinctly Celtic ethos and context in their music and has incorporated Irish symbols, which evoke a mystical, isolated romanticism of Ireland as an island. Meanwhile, the sounds of traditional jigs, reels, and drinking songs can be heard in more or less authentic renditions of Irish pubs in cities all over the world.

ADVERTISING

The allure of islands has widely been used by advertisers to attract consumers. They have learned that the top fantasy vacation for men is being marooned on a tropical island with several members of the opposite sex. Advertisers have played on sensual images of islands very effectively. Couples holding hands and sharing loving glances (often with the male considerably older than the female in magazines catering to men), sunsets, and deserted beaches have been the staple of tourist advertising to specific islands or collective, exotic island locations. A further staple of island vacation advertising is promoting the impossible combination of luxuriant tropical vegetation (which needs quite a bit of rain) and continual sunshine! However, travel is not the only product that island advertising sells: skin-care products are named “Bali Orchid Body Lotion” and “Bora Bora Sand Scrub,” rum commercials offer “slow down” images, and “Souper Star Hawaii soups” appear on television and in print advertisements. The successful clothing chain Tommy Bahama’s slogan is “purveyor of island lifestyles” and has expanded from clothing to include even home furnishings. The message in using islands in advertising is clear: islands will do for you what continental life will not. Islands have long been used as indications of personal status. The elite American families have gone to Nantucket and Martha’s Vineyard for generations, private islands in the Thousand Islands (of the St. Lawrence River) have castles built on them, and Atlantis (the lost continent originally described by Plato) has become a luxurious island resort in the Bahamas. The inaugural issue of Fortune (1930) magazine suggested that “As a great symbol of possession, the privately owned island may yet supplant the steam yacht.” Every island that is seeking tourists is also advertised as having friendly and welcoming natives. Of course, like many island images, this can be far from reality. Reportedly, one crime-ridden island even has a “Visitors’ Society” that provides clothing to tourists whose luggage has been stolen and has members visit victims of violent crime during their stays in local hospitals! VIDEO GAMES, COMICS, AND GESTURES

Video games may take place on real (Monkey Island in the Caribbean or Over the Edge in the Mediterranean) or fictional islands (Sonic the Hedgehog, Super Mario, Legend of Zelda, Myst, Yoshi’s Island, Adventure Island, and Amazing Island). The same is true for comics with X-man (Atlantic Islands), Patrouille des Castors (Carib-

bean), Corto Maltese (Pacific) appearing on actual islands and Whiz Comics, Exciting X Patrol, and Teen Titans appearing on fictional islands. The “shaka” sign is a common gesture associated with Hawaii and surfing but now widely used by teenagers to denote a carefree, “hang loose” attitude. This gesture involves extending the thumb and pinkie while keeping the middle fingers curled and rotating the wrist. This gesture even has a popular culture origin. Supposedly, it originated with a Hawaiian sugar-cane worker who lost his three middle fingers in an accident. His hand gesture of “all clear,” using only the thumb and pinkie, became the shaka! CONCLUSION

Every day we see images to remind us of islands—people in Hawaiian shirts, dreadlocked travelers, advertising posters, and beauty products. A popular parlor game that has been played for decades is “If you were stranded on a desert island, what books (or music or movies) would you want to take with you?” The connotation here is one of isolation but also of potential contentment. Through popular culture, islands provide an accessible view of the exotic—“Come to an island and your life will change, even if it’s only while you are there.” Perhaps we have made islands a romantic icon because we need something to help us to escape life’s tedium and troubles. Maybe without such a dream of going to a “better place” (even if it’s only through what we wear, taste, or see), the tedium of normal life would be overwhelming. But beware—isolation, loneliness, and especially boredom may await the island visitor as well. SEE ALSO THE FOLLOWING ARTICLES

Kon-Tiki / Missionaries, Effects of / Prisons and Penal Settlements ADDITIONAL READING

Briand, P. L. Jr. 1966. In search of paradise: the Nordhoff and Hall story. Honolulu: Mutual Publishing. Cantor, P. A. 2001. Gilligan unbound: pop culture in the age of globalization. Lanham, MD: Roman and Littlefield. Clarke, T. 2001. Searching for paradise. a grand tour of the world’s unspoiled islands. New York: Ballantine Books. Day, A. G. 1986. The lure of Tahiti: an armchair companion. Honolulu: Mutual Publishing. Day, A. G. 1987. Mad about islands: novelists of a vanished Pacific. Honolulu: Mutual Publishing. Kernahan, M. 1995. White savages in the South Seas. London: Verso. Kirsten, S. 2004. Tiki style. Cologne: Taschen. Marsh, T., and J. Sparks. 2002. The magic of the Scottish Isles. Newton Abbot, UK: David & Charles. Michener, J. A. 1951. Return to paradise. New York: Fawcett Crest. Young, L. B. 1999. Islands: portraits of miniature worlds. New York: W. H. Freeman and Company.

P O P U L A R C U LT U R E , I S L A N D S I N

765

POPULATION GENETICS, ISLAND MODELS IN JEFFREY D. LOZIER

N m migrant pool

University of Illinois, Urbana-Champaign

Species are rarely spatially continuous or homogenous in their distributions; more often they are geographically subdivided into local subpopulations, within which individuals frequently interact but between which interactions are less common. Such population structure is often complex and can alter the way in which evolutionary forces (gene flow, genetic drift, and natural selection) act in nature. Island models and their variants provide a useful framework in which to investigate genetic variation and its potential ecological and evolutionary consequences in subdivided populations. POPULATION GENETICS OF STRUCTURED POPULATIONS

In many scientific disciplines, researchers are interested in measuring demographic parameters for the organisms under study, including numbers of individuals and dispersal rates among neighboring populations. For many organisms, direct counts and observations of movement are impractical, and modern molecular methods that make use of putatively neutral genetic markers (e.g., DNA sequencing or microsatellite genotyping) provide an indirect way to estimate such parameters. However, making inferences from genetic data requires models that predict how variation is distributed within and among individuals and populations, which in turn requires assumptions about how individuals and populations are spatially arranged in nature. A few such models are highlighted in this article. “CLASSIC” MODELS OF POPULATION STRUCTURE Wright’s Island Model and F-statistics

To investigate evolutionary dynamics in a structured population, Sewall Wright developed his classic island model, which continues to be commonly used for making inferences from genetic data. The model consists of an array of discrete subpopulations each of size N that exchange genes each generation by contributing equal proportions of individuals (m) to a migrant pool, which are then redistributed randomly and equally among subpopulations (Fig. 1). Gene frequencies

766

P O P U L AT I O N G E N E T I C S , I S L A N D M O D E L S I N

FIGURE 1 A schematic representation of Wright’s island model con-

sisting of six subpopulations of size N that contribute and receive an equal proportion of migrants (m) each generation.

in the subpopulations diverge over time as a result of random genetic drift, which is dependent on the population size, while at the same time are homogenized as a result of the movement of individuals. The degree of differentiation is expressed by the standard inbreeding coefficient FST, which can be thought of as the relatedness of a pair of genes sampled within a subpopulation relative to a pair sampled from the population as a whole. Theoretically, FST can range from 0 to 1 (though in practice this can vary), where 0 represents complete random mating and 1 represents complete isolation; high levels of gene flow thus tend to make subpopulations more similar, while low levels allow populations to diverge. With an infinite number of subpopulations at equilibrium between migration and drift, FST is related to the number of migrants per generation according to the equation FST ≈ 1/(4 Nm + 1), which, theoretically, can be used to estimate migration rates. Implementing this relationship in practice can be problematic for a number of reasons. Of particular importance is that drift and migration are confounded in Nm, and FST estimated from genetic data can be highly influenced by recent demographic history as well as migration. Populations that are recently isolated, perhaps as a result of an island colonization event or habitat fragmentation, can exhibit low FST despite an absence of gene flow, because of their recent shared ancestry. Similarly, islands that receive migrants from a continental population but that never directly exchange migrants may also exhibit low FST. Interested readers should refer to the 1999 paper by Whitlock and McCauley, which critically examines the assumptions of the so-called Fantasy Island model. However, FST remains a useful measure of overall population structure, as long as researchers are cautious in its interpretation.

Steppingstone Models and Isolation-by-Distance

The steppingstone model (SSM) of population structure relaxes the particularly unrealistic assumption of equal probabilities of gene flow among all subpopulations; it is based on the empirical observation that individuals in most species are in some way dispersal limited. The SSM comprises an array of discrete subpopulations that can be arranged in one or two dimensions and is spatially explicit in that migration occurs most frequently between neighboring steppingstones (Fig. 2). This property gives rise to a positive correlation between genetic differentiation (usually estimated using pairwise FST or some similar estimator) and the geographic distance between the two subpopulations, and is known as isolation-by-distance. Isolation-by-distance can also arise in continuous populations as long as individuals are dispersal limited in some way. Tests for isolation-by-distance can be a useful way to get a general idea of the spatial scale of dispersal or to assess the equilibrium status of a population.

the assumptions and limitations of the particular models being implemented. As methods with more realistic models are developed, many of the limitations of the more basic models will be overcome, allowing for better estimation of the parameters of interest in particular situations. SEE ALSO THE FOLLOWING ARTICLES

Dispersal / Inbreeding / Island Biogeography, Theory of / Metapopulations REFERENCES

Hey, J., and C. A. Machado. 2003. The study of structured populations— new hope for a difficult and divided science. Nature Reviews Genetics 4: 535–543. Pearse, D. E., and K. A. Crandall. 2004. Beyond FST: Analysis of population genetic data for conservation. Conservation Genetics 5: 585–602. Slatkin, M. 1993. Isolation by distance in equilibrium and non-equilibrium populations. Evolution 47: 264–279. Whitlock, M. C., and D. E. McCauley. 1999. Indirect measures of gene flow and migration: FST ≠ 1/(4 Nm+1). Heredity 82: 117–125. Wright, S. 1951. The genetical structure of populations. Annals of Eugenics 15: 323–354.

N m

PRINCE EDWARD ISLAND SEE INDIAN REGION

FIGURE 2 A schematic representation of a two-dimensional stepping-

stone model.

“MODERN” MODELS OF POPULATION STRUCTURE

Several population genetic methods have made use of statistical advancements and increased availability of computational resources in recent years. Many of these methods use coalescent theory and realistic demographic models to infer parameters such as population sizes, divergence times, and migration rates. These methods relax a number of the assumptions of classical models and can estimate more complex parameters such as asymmetrical migration and populations that differ in size or change in size over time. In some situations, researchers are interested in defining which sets of sampled individuals actually belong to discrete randomly mating populations, and there are several methods that make use of genetic data for this purpose. Some of these can also be used to estimate current migration events at the individual level, which is sometimes of greater interest than the long-term estimates provided by FST. CONCLUSION

Overall, there is much insight to be gained about demographic history from the study of genetic variation in structured populations, as long as researchers are aware of

PRISONS AND PENAL SETTLEMENTS EPHRAIM COHEN Hebrew University of Jerusalem, Rehovot, Israel

Prisons are institutions in which people are confined and deprived of a large range of liberties. The obvious parallel between islands and prisons resides in isolation, whether from mainland or from society. The more distant an island is from the mainland, and the deeper, colder, stormier, and more infested with predators (e.g., sharks) its surrounding waters are, the more effective it is as a prison. Opportunities to escape from such islands are drastically reduced or even completely eliminated. Whereas prisons on the mainland are usually cell prisons, on islands convicts were not always confined but rather held in camps. On certain islands, prisoners were merely slaves of the free settlers or were subjected to harsh work in the mining or timber industries. Some ex-convicts had to serve time following release; thus, although freed, they were not allowed to leave the islands. Under such conditions

PRISONS AND PENAL SETTLEMENTS

767

the islands were regarded as penal settlements that had considerable economic benefits for the mainland. Prison islands have nowadays become tourist attractions because famous criminals or political figures were detained there. Some of the islands, which hold natural beauty, became attractive targets for tourism (including ecotourism) and for mainland vacationers. AUSTRALIAN ISLAND PRISONS

The practice of sending prisoners from England to its North American colonies stopped as the United States of America gained independence in 1783. Because of prison overcrowding, the British Crown devised the solution (initiated in 1786) of shipping convicts to the newly discovered Australia. The convicts, (both criminal and political) suffered a long and harsh 8-month journey under overcrowded conditions, being often chained and supplied with meager food rations. Throughout the eighteenth and nineteenth centuries Tasmania and some offshore islands of Australia served as prisons for criminals and political convicts. Later, for economic reasons, some islands became penal settlements. From the beginning, although provided with basic supplies from England, prisoners were expected to become self-sufficient.

Port Arthur

Port Arthur is located on the Tasman peninsula about 60 km southeast of the capital city, Hobart. The peninsula is separated from the main island of Tasmania by a narrow 30-m-long stretch of land called Eaglehawk Neck. Port Arthur served as a penal colony from 1830 until its closure in 1877. Escape of prisoners was almost impossible because the surrounding waters were shark-infested and the narrow isthmus was guarded by soldiers and dogs. Port Arthur was the preferred destination for the hardiest English convicts as well as for second offenders, those who had committed another offense after arriving in Australia. The place was also a destination for juvenile convicts, 9–18 years old, separated from adults in a special prison located at Point Puer. The first 150 convicts in Port Arthur worked as slaves to establish the timber industry. By exploiting the hard work of the convicts, Port Arthur became Tasmania’s industrial center for timber, shipbuilding, brickmaking, wheat growing, and shoe making. The convicts suffered cruel physical conditions and were punished by flogging. Later, according to new ideas of treating convicts developed in England and based on the design of Pentonville Prison in London, Port Arthur turned into a Model Prison in 1852. Physical punishments were replaced by psychological ones, using absolute silence and complete isolation in tiny cells (Fig. 1).

Cockatoo Island

Cockatoo Island, located in Sydney Harbor, New South Wales, was a prison during 1839–1869. The convicts were employed in constructing silos for storing the colony grain and in excavating rocks for construction projects in Sydney. Norfolk Island

Norfolk Island, located in the Pacific Ocean between Australia, New Zealand, and New Caledonia, is part of the external territories of Australia. In order to claim the island before the French colonized it, in preparation for commercial development, England in 1788 sent 15 convicts and 17 free men there. More convicts were later sent to create a penal settlement that would be capable of providing farm supplies to Sydney. This first penal settlement ceased to exist in 1813. However, a second penal settlement was established in 1824 with the purpose of transporting there the worst detainees, those who had committed crimes after arrival as convicts in New South Wales. Norfolk Island was notorious for exposing convicts to harsh punishment, torture, constant flogging, and scarce food. The second penal settlement was closed in 1855 when the last prisoners were transferred to Tasmania.

768

PRISONS AND PENAL SETTLEMENTS

FIGURE 1 Interior of the Model Prison, Port Arthur, Tasmania, Australia.

Photograph by Joern Brauns.

During 1830–1877 Port Arthur housed about 12,000 convicts (men and boys), and after closure it became a geriatric home and mental asylum for ex-convicts. Today Port Arthur is one of the main tourist attractions for visitors to Tasmania. Rottnest Island

Rottnest Island, which is located 19 km offshore from Perth, became a prison for Western Australian Aborig-

ines in 1838. Since the end of the nineteenth century the island has been a favorite holiday resort for Perth inhabitants.

of General Noriega. The prisoners worked long hours and were subjected to extreme torture. Today Coiba is a tourist attraction.

St. Helena Island

Dawson Island

St. Helena Island was the principal Queensland prison for men from 1867–1993. It is located 21 km east of Brisbane in Moreton Bay, a few kilometers from the mouth of Brisbane River. When an Aboriginal nicknamed Napoleon was exiled to this island, it received its present name from the better-known St. Helena in the South Atlantic, where Napoleon Bonaparte spent his final years. During the early days it was a secure prison, and the worst criminals were sent there. It was known as “the hellhole of the Pacific” and as “Queensland’s Inferno” for its physically harsh conditions. St. Helena Island was initially meant to be a self-sufficient prison, even exporting products to the mainland. Today, as a national park, the island has become a tourist destination for visitors to the Brisbane area.

Dawson Island is located in the Strait of Magellan, 100 km south of Punta Arenas city in Chile. The island became a prison for political detainees following the military coup in 1973 and was closed down in 1975.

Sarah Island

Sarah Island is located in Macquarie Harbour, west of Tasmania. The penal colony was established in 1822 and closed 11 years later, in 1833. The worst convicts who had escaped other penal systems were sent there. The prison was notorious for very harsh treatment of inmates, who suffered crowded conditions, hard forced labor, malnutrition, scurvy, and dysentery in addition to widespread severe flogging as punishment. Convicts were employed in the shipbuilding industry and logging the Huon pine trees, while in chains. Topographical conditions such as access via a narrow gate (“Hell’s Gate”) of dangerous sea and rough mountains, which separate the penal colony from the other settlements on the island, made escape dangerous and almost impossible. Aspects of the horrible British penal system have been described in several books and a play. The ruins of the penal settlement are today on the Tasmanian Wilderness World Heritage list. SOUTH AND CENTRAL AMERICA Coiba Island

Coiba is a large island (15 × 50 km) off Panama’s Pacific coast. Its penal colony was established in 1912 and closed down in 2004. The island is far from the mainland, and its surrounding waters are famous for strong currents and aggressive sharks, which make any escape quite difficult. The worst Panamanian criminals were imprisoned there, as were political opponents of the military regime

Devil’s Island

Devil’s Island, one of three islands off the coast of French Guiana, was part of a French penal colony system. It was established as a prison by Napoleon the III in 1852 and closed down in 1952 by the French government. Throughout that century the prison was used by France for sending political convicts like the anarchist Clément Duval (1938) as well as thieves and murderers. The total number of prisoners spending their term on Devil’s Island exceeded 80,000. At the timber camps on the island the convicts had to endure forced labor in malaria-infested areas. Only a few prisoners managed to escape through the surrounding dense jungle. The French authorities attempted to convert the island into a penal settlement because, according to an 1854 law, convicts who had spent more than 8 years were forced to stay permanently in French Guiana. Convicted women were sent to the territory in order to marry freed convicts (as was done in Australia). Devil’s Island became famous in history because in 1895 the Jewish French army captain Alfred Dreyfus was convicted of treason and sent to the prison. The sentence evoked a heated political reaction (J’Accuse by Émile Zola), and the case, as well as the antisemitic atmosphere in France, had a great impact on the Jewish journalist Theodor Herzl. The Dreyfus case was eventually instrumental in convincing Herzl that Jews needed to create their own homeland, and thus in his becoming the father of modern Zionism. The Devil’s Island prison was a subject of books such as Papillon by ex-convict Henri Charrière, Dry Guillotine by Rene Belbenoit, and Plan de Evasion by Adolfo Bioy Casares, as well as a movie based on Papillon and folk songs. Gorgona Island

Gorgona Island is located 50 km off the Colombian Pacific coast. The island was used as a high-security jail from the 1950s until 1984. Escape of prisoners was highly discouraged by the presence of poisonous snakes on the

PRISONS AND PENAL SETTLEMENTS

769

island as well as shark-infested waters that separated the island from the Colombian mainland. In 1985 the island became a National Natural Reservation Park, where its endemic species, subtropical forests, and coral reefs are strictly preserved. Martin Garcia

Martin Garcia is located in Rio de la Plata, 3.5 km off the Uruguayan coast. In 1765, the Spaniards installed a military prison for deserters. For about 90 years (1870–1957) the island served as prison for both political and criminal detainees. The crewmen of the German battleship Graf Spee were detained there; the Nicaraguan poet Ruben Dario lived on the island (though not as a prisoner), as did the Argentinian dictator Juan Peron. The island was a notorious site for detention and torture of political opponents to the military dictatorship in Argentina during 1976–1983. San Cristobal

On San Cristobal in the Galápagos Archipelago a penal colony was established during 1850–1860 for Ecuadorian prisoners. A second penal colony was established in 1946, turned essentially into a concentration camp, and was closed down in 1959. SOUTH AFRICA: ROBBEN ISLAND

Robben Island is located 12 km offshore from Cape Town, South Africa. Since the end of the seventeenth century the island has been used as prison for political leaders from various Dutch colonies or for isolating lepers (1836–1931). Under apartheid, Robben Island became notorious for imprisoning black political leaders, notably Nelson Mandela. The Island was declared a World Heritage Site and has been transformed into a popular tourist attraction. NORTH AMERICA Alcatraz Island

Alcatraz Island is located in the middle of San Francisco Bay, California (Fig. 2). It served as a lighthouse (first erected in 1854), a military fortification, and a military prison. The island was transformed into a federal prison in 1934, was closed in 1963, and has since become a national recreation area. A few dozen escapes were recorded, yet scores were apparently successful. Escape attempts from the penitentiary were the subject of a motion picture (Escape from Alcatraz, 1979) and a TV series.

770

PRISONS AND PENAL SETTLEMENTS

FIGURE 2 Alcatraz Island as seen from Coit Tower in San Francisco.

Photograph by Jon Sullivan, PD Photo.org.

Johnson’s Island

This prison was located in Sandusky Bay on Lake Erie, Ohio, and served for Confederate prisoners of war captured during the Civil War. The prison was opened in 1862 and was closed at the end of the war. ASIA Indonesia

Indonesia has a long history of using islands in its vast archipelago as prisons. For example, Buru Island in the Maluku region was a place for political prisoners after the attempted 1965 coup that led to the ouster of President Sukarno; Atauro Island, off the north coast of East Timor, was used as a penal colony for rebellious East Timorese. Russia: Sakhalin

Sakhalin is a large island located north of Japan, in the entrance of the Sea of Okhotsk in the Russian Far East. Being a remote and harsh place, it served as a prison island and was the largest Russian penal colony under the tsars; political opponents were also exiled there. Following a visit to Sakhalin, the famous Russian writer Anton Chekhov described the hellish life in the island in his book The Island: A Journey to Sakhalin. Today Sakhalin has become a major energy supplier because of its vast oil reserves. The Solovetsky Islands

This is a group of six islands located in the White Sea, Russia, in Onega Bay. They served as an exile location during the Russian Empire and became a prison in 1926 after the Russian Revolution. The islands were an integral part of the vast Gulag system where many prisoners were

detained. Being close to the Finnish border, the prison was closed in 1939 at the beginning of World War II. The islands have become a tourist attraction, primarily because of the famous fifteenth-century monastery, and are on UNESCO’s World Heritage List. EUROPE

second penal settlement was established in 1858 for rebels against British colonial rule in India. A prison called the Cellular Jail was built there from 1896 to 1906 and was used first by the British and, during World War II, by the Japanese occupation. Today the Andaman Islands are a territory of India, and the Cellular Jail is a national memorial.

Chateau d’If

Chateau d’If was initially a fortress and a castle built (during the period 1516–1529) by the French king François I on a small rocky island in the Mediterranean Sea about 3 km off the city of Marseilles in southern France. In 1634 the chateau became a state prison for mostly religious (e.g., several thousand Huguenots) and political (e.g., Gaston Crémieux, a leader of the Paris Commune) detainees. The notorious prison was made famous by Alexandre Dumas’s novel The Count of Monte Cristo (1844–1845). The prison was closed at the end of the nineteenth century and since 1890 has become a popular tourist attraction. ˙I mralı

Only in 1935 was a prison built on the Turkish island of ˙Imralı located in the Sea of Marmara. A number of political detainees, notably the former Prime Minister Adnan Mederes, served terms there. Since 1999 the jail has become a maximum-security prison for its only inmate, Kurdish rebel leader Abdulla Ocalan. INDIAN AND PACIFIC OCEANS Andaman Islands

The Andaman Islands are a group of Indian islands located in the Bay of Bengal, separated from the Malay Peninsula by the Andaman Sea. The first British penal settlement established there, in 1789, was closed in 1796. Dangerous convicts who broke the law in the penal settlement were sent to Viper Island, where they were put in chains while being subjected to forced hard labor. A

Mauritius

Mauritius is an island in the Indian Ocean, east of Madagascar. Primarily, the island can be regarded as a penal settlement for Indian convicts who were sent on demand, especially as a labor force for the sugar cane industry. The island was a penal settlement from 1790 until 1853, when the last convict was liberated. New Caledonia

New Caledonia became a French colony in 1853 and, since 1986, has been included under the United Nations list of Non-Self-Governing Territories. The island is located in the Melanesia region of the southwestern Pacific. A penal colony was established on the island in 1864 by Napoleon III. Most prisoners were French criminals, but inmates also included political convicts, such as 4300 socialists following the insurrection of the Paris Commune in 1871. Following a revolt of Arab and Berber tribes in Algeria, many thousand prisoners were transported to New Caledonia. It is estimated that about 22,000 prisoners were kept in the island between 1864 and 1922, when the penal colony was closed down. SEE ALSO THE FOLLOWING ARTICLES

Popular Culture, Islands in / Rottnest Island / Tasmania FURTHER READING

Charrière, H. 1971. Papillon. New York: Pocket Books. Hughes, R. 1988. The fatal shore: the epic of Australia’s founding. New York: Vintage Books.

PRISONS AND PENAL SETTLEMENTS

771

R RADIATION ZONE KOSTAS A. TRIANTIS AND ROBERT J. WHITTAKER Oxford University, United Kingdom

Species richness on island systems is a function of colonization, speciation, and extinction. In island faunas and floras, there exist radiation zones, in which phylogenesis (including by adaptive radiation) increases with distance from the major source region. Within-island speciation and within-archipelago speciation (sometimes termed simply archipelago radiation) can thus be major contributors to species richness. THE THEORETICAL BACKGROUND

In their seminal book The Theory of Island Biogeography, MacArthur and Wilson (1967) wrote, ‘‘In equilibrial biotas . . . the following prediction is possible: adaptive radiation will increase with distance from the major source region and, after corrections for area and climate, reach a maximum on archipelagos and large islands located in a circular zone close to the outermost dispersal range of the taxon.” They termed such peripheral areas the radiation zone (Fig. 1). The biogeographical circumstances in which radiations take place are reasonably distinctive. Radiations are especially prevalent on large, high, and remote islands, lying well beyond a group’s normal dispersal range. Here the low diversity of colonists, and the disharmony evident in the lack of a normal range of interacting taxa, facilitates in situ diversification for those few representatives of a group that do manage to reach the islands. For less dis-

772

persive taxa, their radiation zones may coincide with less remote archipelagoes, which have a greater degree of representation of interacting taxa than do the most remote archipelagoes. Thus, macroevolutionary aspects of island biogeography complement ecological aspects. That is: (1) within-island radiations should occur most often on distant, large islands of complex topography; (2) colonization should occur more often on near than distant islands; (3) speciation should occur more often on distant islands than on near islands. Hence, the circumstances for radiation reach their synergistic peak on the most remote islands, the epitome being the Hawaiian Island group (Fig. 2). To illustrate how the low diversity of colonists, and the disharmony evident in the lack of a normal range of interacting taxa, facilitate in situ diversification, ants dominate arthropod communities across most of the world, but in Hawaii and southeastern Polynesia they are absent (or were, prior to human interference). In their place, there have been great radiations of carabid beetles and spiders, and even caterpillars have in a few cases evolved to occupy predatory niches. NATURE’S EVOLUTIONARY RADIATION EXPERIMENTS

Examples that fit this idea of maximal radiation near the dispersal limit include, among others, birds on Hawaii and the Galápagos, frogs on the Seychelles, gekkonid lizards on New Caledonia, lemurs on Madagascar, ants on Fiji, and Anolis lizards in the Caribbean. Although the Galápagos are renowned for other endemic groups, notably the tortoises and plants, the most famous group of endemics are Darwin’s finches (Emberizinae). The radiation of the lineage has taken place in the context of a remote archipelago, presenting extensive

“empty niche” space, in which the considerable (but not excessive) distances between the islands have led to phases of inter-island exchange—but only occasionally. Differing environments have apparently selected for different feeding niches both between and within islands during a radiation process of some 3 million years, involving both allopatric and sympatric phases. Thereafter, behavioral differences between forms maintain sufficient genetic distance between sympatric populations to enable their persistence as (to varying degrees) distinctive lineages. The Hawaiian honeycreepers (or honeycreeperfinches) have shown an even greater radiation than Darwin’s finches. They are a monophyletic endemic group perhaps best considered a subfamily, the Drepanidinae. Estimates of the number of species known historically range from 29 to 33, with another 14 having recently been described from subfossil remains (and it seems more are in the process of being described); most extinctions occurred between the colonization of the islands by the Polynesians and European contact.

Isolation

Dispersive taxa Less dispersive taxa

Source pool

FIGURE 1 Dispersive taxa radiate best at or near to their effective

range limits (the radiation zone), but only moderately, or not at all, on islands near to their mainland source pools. Less dispersive taxa show a similar pattern, but their range limits are reached on much less isolated islands. The increased disharmony of the most distant islands further enhances the likelihood of radiation for those taxa whose radiation zone happens to coincide with the availability of high island archipelagoes. From Whittaker and Fernández-Palacios (2007).

those of the younger islands to the east; that is, most, but not quite all, of the inter-island colonizations have been from older to younger islands (this pattern is termed the progression rule). The general trend of colonization from old to young Hawaiian islands in Drosophila is matched by similar trends in the radiation of the silversword alliance

The Hawaiian “Experiment”

The Hawaiian endemic crickets are thought to have derived from as few as four original colonizing species, each being flightless species arriving in the form of eggs carried by floating vegetation. The ancestral forms were a tree cricket and a sword-tail cricket from the Americas and two ground crickets from the western Pacific region. Three of the successful colonists have radiated extensively, and Hawaii now has at least twice as many cricket species as the continental United States. Much later, a further eight species have colonized, but these are considered to have been introduced by humans. The tree crickets (Oecanthinae) have been calculated on phylogenetic grounds to have colonized Hawaii about 2.5 million years ago, radiating into three genera and 54 species (43% of the world’s known species), with the greatest diversification seen within the older islands, which were occupied earliest. They have radiated into habitats not occupied by their mainland relatives. Drosophila and several closely related genera in its subfamily include about 2000 known species, of which Hawaiian drosophilids (the closely related genera Drosophila and Scaptomyza) account for some 600–700 species—although, with further taxonomic work, the eventual figure could be as great as 1000 species. The ancestors of the Hawaiian drosophilids (a single species, or at most two) probably arrived on one of the older, now submerged islands and have radiated within and between islands. In general, the older islands to the west contain species ancestral to

FIGURE 2 Conceptual model showing the development of species rich-

ness on large islands or archipelagoes that experience varying rates of colonization as a result of varying degrees of isolation. As explained by Heaney, on islands near a species-rich source that initially lack the study taxon, all species will be present through direct colonization because high rates of gene flow will swamp out any potential speciation; thus, no endemic species will be present, but many non-endemics will be present. As the average rate of gene flow drops below approximately the level of Nm = 1 (gene flow equal to one individual per generation) for the study taxon (point A), anagenesis will begin to take place, and endemic species will develop, although they will be outnumbered by non-endemic species. These endemic species (between lines 1 and 2) will have their sister taxon in the source area, not on the island/ archipelago. As colonization becomes still less frequent, and as time passes, phylogenesis will produce endemic clades in which the endemic taxa have their sister taxon on the island/archipelago, not in the source area (species between lines 2 and 3). As more time passes, the oldest clades will become progressively more species-rich (radiation zone) (between lines 3 and 5). From Heaney (2000).

R A D I AT I O N Z O N E

773

(tarweeds), which is a group of 30 species in three genera (Dubautia, Argyroxiphium, and Wilkesia) in the Asteraceae. The alliance appears to have descended from a colonization event by a bird-dispersed herbaceous colonist (itself a hybrid from the genera Madia and Raillardiopsis) about 5 million years ago, matching the age of the oldest large island, Kauai. They have diversified into a range of life forms, including monocarpic rosette plants, vines, matforming plants, and trees, and they occur in habitats ranging from desert-like to rain forest. The Macaronesian “Experiment”

The Canarian invertebrates provide an outstanding example of radiation on rather less isolated islands than Hawaii and the Galápagos. There may be as many as 6000 native Canarian invertebrate species, of which half are considered to be endemic. In general, the Canarian invertebrate fauna is characterized by the absence of certain groups (e.g. scorpions, Cicadae, Solifugae, and dung beetles), but these absences have been compensated for through active speciation within colonist lineages, producing a high species/genus ratio. Additionally, 99 genera are considered exclusive to the archipelago, 58 of them being present in more than one island (multi-island genera) and 41 restricted to one island, of which 25 are found on the island of Tenerife. Radiation is particularly evident in land snails, spiders, and beetles, with at least 24 different genera producing 15 or more endemic species. The largest genera of beetles (Laparocerus), diplopod millipedes (Dolichoiulus), and spiders (Dysdera) are represented in all the major habitats, from the coast up to 3000 m, in arid and wet zones, in forested and open areas (even in lava tubes), all over the archipelago, a pattern strongly supportive of the label “adaptive radiation.” For Macaronesia, the estimated native vascular flora comprises some 3200 species, of which about 680 (~20%) are endemic. The Canaries are the largest in area, the highest, and the richest, with 570 endemics, about 40% of the native flora. Forty-four Macaronesian lineages are endemic at the generic level also, with exactly half of them restricted to the Canaries. Support for the use of the term “adaptive radiation” in several of the above Macaronesian genera comes from studies of features such as habit, leaf morphology, and habitat affinities. Evidence of similar morphology in species of the same and of different genera (termed parallel or convergent evolution) occurring in similar habitats, is supportive of a model of selection having favored particular adaptive outcomes. Within the Macaronesian flora as a whole, most genera are represented by only one or two species, and most of those with

774

R A D I AT I O N Z O N E

over four occur in the Canaries. In general, radiation of lineages has been greatest on the larger islands with the greatest diversity of habitats. The island with the greatest number of endemic plant species (320 species) is Tenerife, and it also has the largest numbers of Aeonium (12 species), Sonchus (11 species), and Echium (9 species). Most endemic plant species have a restricted distribution, 48% occurring on a single island, and a further 15% on two. EVOLUTIONARY RADIATION NOT UNIVERSAL EVEN IN THE RADIATION ZONE

Although there seem to be numerous examples of taxa radiating at or near their outermost dispersal range, there are also plenty of exceptions: taxa that have not radiated much at remote outposts. Such cases include terrestrial mammals on the Solomon Islands and snakes and lizards on Fiji, and within the invertebrates of the Canary Islands, for example, as many as 57 of the endemic genera are monotypic. Similarly, nearly 50% of the ~1000 native flowering plant species of Hawaii are derived from fewer than 12% of the ~280 successful original colonists. Most of the remaining colonists are represented by single species. Thus, it is clear from even the most remote island archipelagoes that not all lineages within a single taxon have radiated to the same degree. Such differences may reflect the length of time over which a lineage has been present and evolving within an archipelago. As MacArthur and Wilson (1963) noted, “To say that the latter [nonradiating] taxa have only recently reached the islands in question, or that they are not in equilibrium, would be a premature if not facile explanation. But it is worth considering as a working hypothesis.” Nonetheless, recent phylogenetic analysis has made it clear that some lineages persist for lengthy periods on oceanic islands without radiation. In short, although early colonization of remote, topographically complex archipelagoes may favor evolutionary radiations, these are not the only factors of significance, and such circumstances do not lead to radiations within all lineages. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Crickets / Drosophila / Galápagos Finches / Honeycreepers, Hawaiian / Lemurs and Tarsiers / Silverswords / Taxon Cycle FURTHER READING

Heaney, L. R. 2000. Dynamic disequilibrium: a long-term, large-scale perspective on the equilibrium model of island biogeography. Global Ecology and Biogeography 9: 59–74. MacArthur, R. H., and E. O. Wilson. 1963. An equilibrium theory of insular zoogeography. Evolution 17: 373–387. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton, NJ: Princeton University Press.

Ricklefs, R. E., and E. Bermingham. 2007. The West Indies as a laboratory of biogeography and evolution. Philosophical Transactions of the Royal Society B: Biological Sciences 363: 2393–2413. Schluter, D. 2000. The ecology of adaptive radiation. Oxford: Oxford University Press. Stuessy, T. F., and M. Ono, eds. 1998. Evolution and speciation of island plants. Cambridge, UK: Cambridge University Press. Wagner, W. L., and V. L. Funk, eds. 2005. Hawaiian biogeography: evolution on a hot spot archipelago. Washington, DC: Smithsonian Institution Press. Whittaker, R. J., and J. M. Fernández-Palacios. 2007. Island biogeography: ecology, evolution, and conservation, 2nd ed. Oxford: Oxford University Press.

RAFTING CHRISTOPHE ABEGG German Primate Centre, Padang, Indonesia

Natural rafting is the rare occurrence in which animals or plants of any kind succeed in crossing a sea strait using tree parts and vegetation. Although rafting has understandably been more often regarded as a means of dispersal for marine plants and animals, it also happens to terrestrial plants and non-volant animals. It is thought that animals are launched by chance from riversides (at any time for reptiles or small mammals but probably at night and during a flood for macaques; Fig. 1), or along the coast (for any animal whenever a tsunami strikes). Once the plant or animal is carried out to sea, another series of contingencies is needed for more than one individual (i.e., at

FIGURE 1 After flooding, rivers carry tree trunks and branches, which

can serve as rafts for a variety of plants and animals. The longtailed macaque, shown here, is a lightly built terminal branch feeder, whose ecological niche has raised its chances of dispersal by rafting. This species is even a good candidate for the title of world champion rafter among mammals.

least one male and one female) to reach an oceanic island and populate it. The chances of survival on a natural raft, once drifting on the sea, seem to differ according to species, judging by the terrestrial species that made their way to islands that were never connected to any land. Reptiles, for instance, are usually able to resist starvation for longer periods than are mammals. OVER-SEA DISPERSAL BY RAFTING: SPECIES AND CONTEXT

What taxa tend to use natural rafting, with what vectors, and to get where? Long-distance dispersal of reef corals by rafting has been studied extensively. Rafting on algae, pumice, and other floating materials has also been proposed for marine invertebrates. In contrast to active dispersal and to passive dispersal by wind or water, rafting is a rare phenomenon. Some reptiles, lizards, and skinks appear to have had some success with the use of rafts, as their distribution on oceanic islands attests. The distribution and diversification of Anolis lizards in the Caribbean may be explained by infrequent rafting events; indeed, anoles have been observed rafting between islands in the Lesser Antilles after a hurricane. The Neotarentola geckos seem to have reached Cuba from North Africa by following north equatorial currents across the Atlantic Ocean, a journey of 6000 km. A frog endemic to the islands of Sao Tomé and Principe, 300 km off the west coast of Africa, is thought to have colonized its habitat using rafts and oceanic currents, possibly using rotten trunks. Rafting via floating mats of vegetation has been invoked to explain the frequent occurrence of land snails on remote oceanic islands, such as the Galápagos and Hawaii. South Pacific top shells have colonized many islands of the Pacific Ocean, most likely through rafting, possibly using buoyant rafts of kelp, from Australia and New Zealand up to South America. In the same way, certain insects appear to use rafting rather than flying as a means of dispersal. Weevils (Rhyncogonus) are widespread on the remote islands of the Pacific, and their distribution is attributed to their tendency to attach their eggs to leaf phyllodes, facilitating transport on vegetation mats. The occurrence of certain ant lineages in Fiji has been attributed to their ability to nest in plant cavities. Rafting may likewise allow terrestrial plant species to colonize remote islands. Among medium-sized mammals, the genus Macaca (Mammalia: Cercopithecidae) is unique among nonhuman primates for the range of habitats it has colonized, from equatorial to temperate ecosystems, from evergreen primary forests to grassland or even habitats modified by humans, but also from continents to deepwater islands.

RAFTING

775

Eleven macaque species inhabit the Indonesian archipelago. Their ancestors colonized oceanic islands as did no other primates or even medium-sized mammals in the world, and they are found on a sufficient number of islands to allow significant assessment of dispersal and stocking determinants. For terrestrial mammals, colonization by rafting has been inferred from their presence on offshore islands that were most likely never connected to the mainland during their geological history. However, for some animals that can float, an alternative to rafting, such as floating and swimming, can also be inferred; this dispersal mode was indeed suggested for the colonization of the Galápagos by tortoises. Rafting between islands that were often connected by land during the past few million years cannot be excluded, but the most plausible hypothesis remains terrestrial dispersal—for instance, the dispersal of plants and animals between peninsular Malaysia and Sumatra. As a means of reaching most oceanic islands, rafting is an unpredictable event for terrestrial animals. However, in the Indonesian archipelago, the occurrence of rafting is connected with climate changes because the distances that had to be crossed to reach an oceanic island were shorter during cold periods, when straits’ widths were reduced by lower sea levels. Indeed, there have been climatic fluctuations over much of Earth’s history, but they intensified with polar ice accumulations from 5 to 1.6 million years ago (i.e., during the Pliocene); then the predominance of glacial intervals became characteristic of the last 1.6 million years (i.e., the Pleistocene). During the Holocene (from 10,000 years BP), the global climate entered an interglacial period and reached the present warm conditions. Interglacial periods are correlated with high sea levels, and glacial periods with low sea levels. Lands situated as low as 120 m below present sea level (BPL) were regularly exposed, and some early glacial periods might have lowered sea levels even more than 120 m, up to 200 m BPL. Quaternary climatic changes associated with periodic glaciations had a profound influence on the distribution of primates and other mammals. In such circumstances, the existence of refuges, the emergence of land bridges, and the possibility of sea rafting strongly affected the fate of populations. Because glacially lowered sea levels provoked successive exposures of the southern end of the continental shelf of Asia (the Sunda Shelf ), mammals, including a macaque progenitor, were further able to enter this region, which is today made up of islands. Because few

776

RAFTING

terrestrial mammals are believed capable of dispersing over sea channels using natural rafting, past sea-level changes represent a crucial factor in understanding the present distribution of mammals in the region. Significantly, beyond those lands that have never been exposed during the last million years, there now exists an endemic mammalian fauna that is well differentiated from mainland Asian fauna, for instance on the Philippines and Sulawesi, but also on islands off West Sumatra. The region crossed by the Wallace Line represented an effective barrier that has drastically filtered the dispersal of terrestrial mammals of Asian origin. Similarly, mammals’ dispersal to the oceanic islands found off Sumatra’s west coast was sharply restricted. At the eastern end of the Indonesian archipelago where the Australian region begins, the only Asian mammals to be found, apart from the rodents, were probably dispersed by humans. Off this 120- to 180-m sea-depth limit, where the sea floor has been regularly exposed in the past, at the periphery of the Sunda shelf, some islands are today nonetheless inhabited by terrestrial mammals of Asian origin. Judging by the high degree of endemism of these islands’ faunas, their ancestors are consequently assumed to have crossed sea barriers early on, using either a land bridge connection that later disappeared, or natural rafts. DISPERSAL BY RAFTING FOR MAMMALS: THE MACAQUE EXAMPLE

Among non-volant mammals, very few species could disperse by rafting over sea straits and populate islands successfully: only some rodents, shrews, civets, and a few macaques, each with varying degrees of success. Indeed, the only native terrestrial mammals to occur on the more isolated islands of the central Pacific are bats and seals. It has been suggested that lemurs, tenrecs, carnivores, and rodents have crossed the 400-km strait separating Madagascar from the African mainland by rafting. However, if ancient vicariance is improbable because Africa and Madagascar separated before those animals had evolved, then terrestrial migration via a transitory land bridge should not be excluded, as the gap to cross by rafting to reach Madagascar from Africa was exceptionally wide for medium-sized mammals. Among such mammals thought to have used natural rafts, the macaques are probably the biggest in size. Their ability to disperse by raft also differs according to species. For instance, in about a third of the pigtailed macaque’s time, the longtailed macaque colonized ten oceanic islands, more or less, whereas the pigtailed taxon is thought to have colo-

nized only two oceanic islands. Let us follow chronologically how the pigtailed macaque and then, much more efficiently, the long-tailed macaque used natural rafts to colonize oceanic islands otherwise unreachable by any other medium-sized primate and by very few other mammals. Separated from the continental shelf by trenches over 180 m deep, Sulawesi and the Mentawai Islands would represent refuges for the first descendants of the ancestral macaque coming from India, refuges that the long-tailed macaque was not able to reach in later times. From India, the first macaques to enter insular Southeast Asia dispersed by land, taking advantage of the glacially induced emergence of the Sunda Shelf. A macaque ancestral stock probably reached Sulawesi by island hopping and natural rafting, whereas it reached the Mentawai Islands over a transitory land bridge created by low sea levels during a maximal Plio-Pleistocene glaciation. The long-tailed macaque progenitor would have dispersed only long after. Sulawesi macaque progenitors probably migrated during the Pleistocene by natural rafting from Borneo across a water gap of about 60 km (the Macassar strait), with only a few winners. The most convincing argument for this latter possibility lies in the absence of other primate taxa in Sulawesi, unlike in the Mentawai Islands. The strait separating the Mentawai from the Sumatran shelf is 200 m deep and 30 km wide. With the exception of the Mentawai macaques, which are closely related to the Sulawesi taxa suspected to have colonized their island by raft, the other three endemic primates of the Mentawai Islands do not have any Asian congeners on oceanic islands, suggesting that overwater dispersal had extremely low chances of success for these primates. The progenitors of today’s five endemic Mentawai primates most probably arrived by land from the Sunda shelf at about the time or soon after macaques first colonized Sulawesi. It can be hypothesized that after a particularly cold glacial episode, the ancestral macaques restricted their distribution to a few refuges such as Sulawesi and the Mentawai Islands and that only afterwards did new taxa disperse toward more central ranges. Two species were able to take advantage of climate recovery and recolonize mainland insular Southeast Asia, whereas nine remained in their refuges. According to some authors, the pigtailed macaque’s ancestor, now populating central ranges, may have originated in the Mentawai or Sulawesi Islands, which acted as refuges during cooling. The Mentawai archipelago

is more plausible in this regard, as it is comparatively less isolated from the adjacent continental shelf than is Sulawesi. However, this would imply that a recolonization of the adjacent mainland, Sumatra, occurred from the Mentawai Islands either by rafting during an interglacial period or by land during a glacial maximum that exposed the 200-m-deep present sea floor. Dispersal from Siberut to the mainland via a land bridge is unlikely because if this had occurred, then the fauna of the Mentawai would not be as specific, with such a high degree of primate and mammal endemism, as it is today. More modern taxa would have taken advantage of such a land bridge and would be present on the islands now. Most likely, for the second time in their history after the colonization of Sulawesi, the ancestral pigtailed macaques dispersed by rafting, from Siberut to Sumatra by island hopping, and then succeeded in invading central ranges of Southeast Asia. THE LONG-TAILED MACAQUE: A CHAMPION OF NATURAL SEA RAFTING

Why were long-tailed macaques so successful in reaching oceanic islands, in comparison with pigtailed macaques and all other medium-sized mammal taxa worldwide? As the long-tailed macaque’s present distribution suggests, its progenitor was probably able to use natural rafting over sea straits, from 17 up to 150 km wide during maximal sea regressions, to reach such islands as Maratua, Simeulue, or Nicobar, whereas only a handful of smaller terrestrial mammals did so. There are ten M. fascicularis subspecies from mainland Southeast Asia, going even beyond the Sunda Shelf. Such diversification of subspecies is best explained by a “two-wave” hypothesis. The first wave would have promoted the evolution of the present dark pelage subspecies found on oceanic islands off the Sunda Shelf (i.e., the most differentiated long-tailed macaque populations). This supposedly happened during a recent glacial maximum (~160,000 years), when the distance to be covered by rafting dispersal would have been greatly diminished. The case of the Philippines is interesting, as it represents a composite of oceanic islands separated by more than 180-m-deep sea floors. The two-wave hypothesis suggests that the only macaque present in the Philippines, the long-tailed macaque, may have dispersed in the course of its evolution on two occasions over straits between islands, by natural rafting from Borneo to the Philippines, presumably during two glacial maxima. As a result, only populations from the northern islands are of dark pelage (M. fascicularis philippinensis), whereas the ones in the southern islands show pale to intermediate

RAFTING

777

pelage. The occurrence of an earlier dispersal also applies to the deepwater islands of Nicobar, Simeulue, and Lasia, situated off Sumatra’s northwestern coast, as well as to Maratua Island off Borneo, where the remaining dark pelage subspecies are found. The only other medium-sized primates that are thought to have dispersed by rafting in Asia to a deepwater island are the progenitor of Sulawesi macaques, and probably that of today’s pigtailed macaques. Worldwide, the only other medium-sized primate thought to have used a raft to colonize its island habitat is the vervet monkey of Pemba Island, 50 km off the East African coast. The particular riverine habits of the long-tailed macaque would have favored its dispersal by sea rafting. After examining the long-tailed macaque’s distribution and its exceptional ability to cross sea straits up to 150 km wide, one tends to favor the hypothesis of a land bridge at the origin of Madagascar’s colonization by mammals, given that a 400-km sea strait had to be crossed, unless these mammals were able to hibernate. Although long-tailed macaques rely mainly on riverine and coastal forests including mangroves, pigtailed macaques prefer inland primary forests. In Sumatra, long-tails are rarely observed more than 300 m from river banks: This placement is likely correlated with the species’s morphological features—light weight and long tail—which are adequate features for a terminal branch feeder living in a riverine habitat. From these features, it is proposed that M. fascicularis was equipped to successfully disperse over water barriers. Indeed, natural rafts originate mainly from estuaries fed by relatively wide rivers that carry diverse pieces of wood and organic matter assemblages from inland, rather than from beaches or rock-bound coasts. Long-tailed macaques often feed and usually sleep in certain preferred trees above rivers, high above ground as a protection against predators. In all likelihood, this species, because of its habitat use, has the highest probability among primates and mammals of similar size to be found on a natural raft off the coast. Because long-tailed macaques succeeded in dispersing to a higher number of deepwater islands than any other primate representatives, something other than a sea barrier must have stopped them from populating Sulawesi and the Mentawai archipelago. The absence of the longtailed macaque in Sulawesi and the Mentawai Islands does not mean that some migrants did not reach these islands but rather that they could not establish themselves there as a distinct population because these islands were already inhabited by macaque species belonging

778

RAFTING

to a more ancestral lineage. Compared with terrestrial colonization, migration by raft involves an extremely low number of migrants and an even lower number of successful ones. Two elimination mechanisms may have acted against such migrants. First, animals recovering from the state of near-starvation following long-distance rafting dispersal need to eat upon arrival. If a primate with a similar diet was already present in the new insular habitat, then competitive exclusion of newcomers for food resources must have diminished their chances of survival. Second, whenever newcomers were able to survive and gain access to food, they would have quickly intermingled with populations of a different taxon: Because species within the Macaca genus are inter-fertile (i.e., able to produce fertile offspring between species), it is likely that long-tailed macaques thereafter produced hybrids. As a consequence, they soon were absorbed by the considerably broader gene pool of the first settled taxon. Therefore, further survival and/or stocking opportunities for long-tailed macaque migrants were most likely held in check in Sulawesi and the Mentawai Islands. However, there were also many opportunities for long-tailed macaques to reach macaque-free oceanic islands because members of the first lineage, which preceded the newcomers, had considerably fewer chances to disperse over sea barriers. Because terrestrial colonization involves a relatively high number of migrants, resident species reach a demographic equilibrium with newcomers in a quite different way from that following overwater colonization. In accordance with the view expressed here is the fact that fossil primates Paralouatta found in Cuba, Jamaica, and Hispaniola did not cross sea straits wider than 150 km, in line with long-tailed macaques’ maximal performance in Southeast Asia. SEE ALSO THE FOLLOWING ARTICLES

Dispersal / Ephemeral Islands, Biology / Refugia / Wallace’s Line FURTHER READING

Abegg, C., and B. Thierry. 2002. Macaque evolution and dispersal in insular Southeast Asia. Biological Journal of the Linnean Society 75: 555–576. Ricklefs, R. E., and E. Bermingham. 2008. The West Indies as a laboratory of biogeography and evolution. Philosophical Transactions of the Royal Society of London 363: 2393–413. Thiel, M., and L. Gutow. 2005. The ecology of rafting in the marine environment. I. The floating substrata. Oceanography and Marine Biology: An Annual Review 42: 181–265. Thiel, M., and P. A. Haye. 2006. The ecology of rafting in the marine environment. III. Biogeographical and evolutionary. Oceanography and Marine Biology: An Annual Review 44: 323–429.

Coral Distribution

REEF ECOLOGY AND CONSERVATION Diversity

ROBERT H. RICHMOND

Low

University of Hawaii, Manoa

High

Mangrove Distribution

WILLY KOSTKA Micronesia Conservation Trust, Kolonia, Pohnpei Diversity

NOAH IDECHONG Low

Palau National Congress, Koror

Coral reefs are the most biologically diverse marine ecosystems on Earth, rivaling terrestrial tropical rain forests. They are ecologically, economically, and culturally valuable resources that provide billions of dollars in goods and services to millions of people. An estimated 30% of the world’s reefs have been substantially degraded during the past few decades, with predictions that nearly 60% are in jeopardy of being lost by the year 2050 if present trends continue. Anthropogenic stressors including overfishing, sedimentation, pollution, eutrophication, and warming tied to global climate change are responsible for losses and are reducing natural resilience and the potential for recovery. The bridging of research findings to the development and implementation of sound policies is critical if a legacy of intact reefs is to be left for future generations to enjoy. CORAL REEF DISTRIBUTION AND CONSERVATION NEEDS

Stony corals are distributed worldwide, with reef development being predominantly restricted to the tropical band, approximately 25° north and south of the equator. Clear coral biodiversity patterns exist globally, with the greatest number of species being found in the Indonesia–New Guinea–Philippines area, known as the coral triangle, which houses nearly 600 species (Fig. 1). Moving from west to east, Palau has over 400 species of corals, Guam has nearly 300, Hawaii drops to the mid-60s, and the eastern Pacific of Panama has approximately 12. The Caribbean is comparable to Hawaii, with around 62 species. These observed patterns and gradients are attributed to several factors including (1) the geologic age of the provinces and the time elapsed since major catastrophic events (e.g., glacial cooling of shallow coastal waters in the Caribbean), (2) tectonic events including the movement of oceanic plates with their associated fauna and the rise of the Isth-

High

Seagrass Distribution

Diversity UNEP

Low

High

PHILIPPE REKACEWICZ MAY 2002

Source : UNEP-WCMC, 2001.

FIGURE 1 Map of global coral reef and associated mangrove and sea-

grass biodiversity. Illustration courtesy of United Nations Environmental Program (UNEP).

mus of Panama that separated the Caribbean from the Pacific Ocean, with subsequent speciation and extinction events (vicariance theory), and (3) the dispersal of larvae into and out of diversity centers via oceanic currents. The high diversity found in the Coral Triangle is theorized to be the result of long-term stability, hybridization during mass spawning events, and possibly the accumulation of larvae from other regions. Scleractinian corals are found primarily in shallow tropical waters, but they can also be found in temperate areas, in the deep ocean, and across the world, including off the coast of Africa, in the Indian Ocean, and in the Red Sea. Although substantial biogeographical data are available on the distribution and abundance of coral species, little is known about genotypic diversity within species. The loss of sensitive genotypes, due to both anthropogenic and natural disturbances, has not been measured and is an increasing cause for concern for maintaining future coral populations and biodiversity. Both the quantity and quality of corals reefs are declining worldwide, and the impacts are being felt ecologically, economically, and culturally, through the loss of critical resources and ecological services. The major anthropogenic causes of coral reef loss include overfishing leading to ecosystem-level shifts to fleshy algal domination when herbivorous grazers are depleted, runoff and sedimentation associated with poor land-use

R E E F E C O L O G Y A N D C O N S E R VA T I O N

779

management, destructive fishing practices including the use of dynamite and cyanide, coastal eutrophication from human sewage and agricultural runoff, pollution from maritime activities as well as land-based sources, and temperature-related bleaching events tied to global climate change. Increases in atmospheric CO2 are responsible for ocean acidification, which is expected to become more of a problem for corals and other calcifying organisms in the future. Conservation efforts at local, regional, and global scales are needed to address the documented declines in coral reefs if these precious centers of biodiversity are to survive into the future. The establishment of community-driven marine protected areas (MPAs) in the Philippines and parts of Indonesia has increased fisheries yields outside of the boundaries and has reestablished trophic interactions supporting reef recovery. In addition to networks of MPAs, efforts are also under way to reduce land-based impacts on coastal reefs through the application of integrated watershed management practices in parts of the Coral Triangle, the Caribbean, and the Pacific Islands. Although the United States and other industrialized nations have invested financial, human, and institutional resources into coral reef conservation efforts, development and implementation of effective policies have lagged behind the available science, and the focus has often been on outputs (workshops, publications, and meetings) rather than on outcomes (improving metrics of coral reef health and resilience). Some of the best conservation efforts to date can be found in small tropical islands, where communities are actively engaged in conservation and management. Traditional conservation practices, often tied to reef tenure and resource ownership in the Pacific, provide examples of coral reef resource stewardship that serve as models for other nations aspiring to leave a legacy of functional coral reefs for future generations. CORAL REEF BIOLOGY

Coral reefs are composed of a myriad of interacting invertebrate, vertebrate, algal, and bacterial species. Although corals may not be the most abundant or most taxonomically diverse elements of a coral reef, these organisms are the trophic foundation through their symbiosis with unicellular algae called zooxanthellae and are the main framework-builders providing essential habitat for other reef creatures. These diverse biological communities are often integral components of larger ecosystems including coastal mangroves, seagrass beds, sand flats, lagoons, and deep-water oceanic systems. Coral reefs provide impor-

780

R E E F E C O L O G Y A N D C O N S E R VAT I O N

tant ecological services, are economically valuable, and provide for a variety of cultural activities central to the fabric of tropical island and coastal societies. Corals are relatively simple creatures on the evolutionary hierarchy, with a tissue level of organization. The three types of tissues found in corals and related cnidarians include the inner gastrodermis (housing symbiotic dinoflagellate algae called zooxanthellae, intracellularly within the gastrodermal cells), an outer epidermis, and a layer of mesoglea in between. Reef-building corals, also referred to as hermatypic corals, are colonies made up of interconnected flower-like polyps. The combined animal–algal association is called the holobiont. Corals receive the majority of their energy through photosynthetically derived metabolites released from their algal symbionts; hence, water clarity and associated light penetration are key determining factors affecting the state of coral reefs. The remaining energetic needs are met by the capture of small organisms, which is accomplished through the use of tentacles that are armed with adhesive or penetrating nematocycts. In order for heterotrophic feeding to occur, the surfaces of the coral must be relatively clean of sediment and debris. The living portion of a coral is a relatively thin layer covering and slightly penetrating a non-living exoskeleton composed of calcium carbonate (aragonite). Coral taxonomy is primarily based on skeletal characteristics, including the arrangement of septa within the calices (cups) or troughs housing the individual polyps. Intraspecific variations in colony morphology are often a reflection of environmental parameters including wave energy, water clarity, and depth. Coral reefs persist through the dual processes of reproduction, the formation of new individuals from prior stock, and recruitment, the process through which these new individuals become part of the reef population. Reproduction in corals can occur either asexually (e.g., through the fragmentation of colonies) or sexually through the formation and fusion of male and female gametes. Most corals are simultaneous hermaphrodites, producing both eggs and sperm within the same polyp. Larval development can occur internally, with resulting brooded planula larvae that are immediately competent to settle and metamorphose into the primary polyp of the new colony upon release. Most corals spawn their gametes, often in combined egg-sperm clusters, with external fertilization and development of embryos into the planula stage. The parental line of symbiotic zooxanthellae is usually transmitted to the larvae in brooding species, whereas many spawning species take up their

zooxanthellae from surrounding waters following settlement and metamorphosis. There are six chemically mediated steps involved in coral reproduction and recruitment: (1) synchronization among conspecifics in gamete development and release, (2) egg-sperm interactions leading to fertilization, (3) embryological development, (4) settlement, (5) metamorphic induction, and (6) acquisition of zooxanthellae if not transmitted via the parent. If any of these steps are disrupted, replenishment of populations will not occur. Freshwater runoff, coastal pollution, and sedimentation can all lead to reproduction and recruitment failure, and all have been documented to be problems on many coastal coral reefs. THE VALUE OF CORAL REEFS

Coral reefs across Australia, the Indo-Pacific, Asia, the Caribbean, the Middle East, India, and Africa are ecologically, economically, and culturally important ecosystems. Damage to and destruction of reefs can affect the structure and function of adjacent mangrove, seagrass, and lagoonal communities. Ecological benefits of coral reefs include protection of shorelines and coastal communities from erosion and damage from storm generated waves, with an estimated value of billions of dollars. Coral reefs also generate revenue in the billions of dollars annually from diving tourism and related recreational activities. Coral reefs provide habitat for marine resources of economic and cultural value, and when these resources are depleted or destroyed, social, economic, and nutritional problems often arise. Coral reefs are home to a variety of edible species, including fishes, crustaceans, holothurians, and molluscs. Some reef organisms are used for their chemical or medicinal properties (e.g., the sea cucumber Holothuria atra [Fig. 2], which contains a chemical called holothurin,

used to kill and collect fish from tidepools or to tease octopuses from their lairs). The high biodiversity of coral reefs offers a diverse source of natural chemical products of interest to the fields of medicine and pharmaceutical research. Whereas the food value of coral reef species is evident, the cultural value of traditional activities is often overlooked and is one of the most essential and important benefits of healthy and functional coral reefs to island and coastal communities. Communal fishing, sharing of resources, and the physical demands of reef fishing and gleaning are important to societies adjacent to coral reefs, and the value of these activities cannot be replaced by the provision of canned and imported foods alone. Diving-related tourism has proven to be an important economic asset for destinations possessing coral reefs, and in this context, reefs can be viewed as the “geese laying the golden eggs.” One famous dive site in Palau, the “Blue Corner,” generates over $3 million annually on dives alone, not including hotel stays and associated restaurant revenues. Recognizing the museum value of their coral reefs, Palau introduced a dive permit program and a funding mechanism for supporting a protected area network, with revenues going to coral reef protection, including the placement of mooring buoys at popular dive sites and integrated watershed management activities. Similar examples can be found in the Caribbean and Atlantic (Cayman Islands, Bermuda, and Florida) where the economic benefits of corals reefs have been recognized and serve as the impetus for conservationoriented efforts. Protecting specific reefs as museums for both the local residents and visitors is a sound management decision that provides ecological, economic, and cultural benefits. NATURAL AND HUMAN-INDUCED DISTURBANCES AFFECTING CORAL REEFS

FIGURE 2 The sea cucumber Holothuria atra, used for the toxic chemi-

cals in the skin for tidepool fishing and extracting octopuses from lairs.

Modern scleractinian corals and coral reefs date back 65 million years, to the beginning of the Cenozoic Period. Over geological time, coral reefs undoubtedly have been subjected to a variety of natural disturbances, including volcanic eruptions, changes in sea level, wave events associated with hurricanes, outbreaks of corallivores including the crown-of-thorns starfish Acanthaster planci, freshwater runoff, sedimentation, and El Niño–Southern Oscillation–related warming/bleaching events. Disturbance is important to maintaining diversity on coral reefs, as corals can aggress against one another (e.g., by using sweeper tentacles and mesenterial filaments) or can simply overgrow and/or shade slower growing and less

R E E F E C O L O G Y A N D C O N S E R VA T I O N

781

dominant species. Coral reefs have recovered from the multitude of natural disturbances over long periods of time and are particularly well adapted to rebound from acute disturbances. Anthropogenic disturbance has been affecting coral reefs for a fraction of their time on earth, but such stressors tend to be more chronic in nature, preventing recovery from occurring. Overfishing, particularly of herbivorous species that keep fleshy and filamentous algae in check, runoff and sedimentation above natural levels as a result of poor land-use practices, coastal eutrophication from sewage and agricultural activities, disease outbreaks associated with human-induced stress, marine pollution, and the frequency and magnitude of bleaching events have all been increasing in the past several decades, with documented impacts on reef populations. Although some of these stressors are associated with population centers and urbanization, even the most remote Pacific atolls have been affected by roving fishing fleets and global climate change. Runoff and sedimentation from development and poor land-use practices have had damaging consequences on coral reef communities. Turbidity reduces light penetration and affects the nutrition of corals by shading their photosynthetic zooxanthellae. Decreases in water and substratum quality interfere with critical chemically mediated events tied to coral spawning, larval development, and larval settlement. Source reefs, corridors of transport, and recipient reefs are all being affected by coastal pollution, and even the establishment of marine protected areas will not protect reefs unless integrated management practices are implemented within adjacent watersheds on land. During the past two decades, regional bleaching events tied to global warming have been responsible for the loss of hundreds to thousands of square meters of coral reefs. Elevated seawater temperatures cause the breakdown of the coral–zooxanthellae symbiosis, and many species of corals cannot recover following such bleaching episodes. Additionally, elevated levels of atmospheric CO2 are responsible for ocean acidification, which threatens the ability of corals and coralline algae to calcify. Some coral reefs damaged by regional bleaching events (e.g., in Palau following the 1998 event) have recovered over the subsequent decade, but affected reefs exposed to local anthropogenic stressors such as pollution, runoff, and sedimentation have not. This observation supports the critical value of local conservation and management efforts in supporting recovery following losses associated with global climate change, and the importance of enhanced efforts in response to what many

782

R E E F E C O L O G Y A N D C O N S E R VAT I O N

scientists and managers consider the most serious threat to the future of coral reefs. CORAL REEF CONSERVATION: EXAMPLES FROM THE PACIFIC ISLANDS Marine Conservation Practices

There are irrefutable data demonstrating that coral reefs are in decline on a global scale and that conservation efforts are essential for addressing this problem. The United States and other developed countries have only a few decades of experience in addressing coral reef management concerns. Present regulations, enforcement capabilities, and the allocation of financial and human resources are insufficient to adequately protect coral reefs. Valuable lessons can be learned from Pacific islands that have centuries to millennia of experience dealing with resource conservation and utilization in the context of limited carrying capacities, and whose traditional leaders consider the needs of future generations in their actions. Many Pacific islands still have traditional governance systems, and these lend themselves well to resource stewardship. Reef tenure or ownership is one of the most important and effective features for conservation in many of these islands, and it varies from complete and formal ownership of the reef and resources by villages to modified systems of local stewardship but central government jurisdiction. Yap State in the Federated States of Micronesia and parts of Fiji and New Guinea are among the most traditional island groups in the Pacific and have complex and sophisticated resource management frameworks. Although these jurisdictions have elected governments, their traditional chiefs still hold positions of responsibility, and villages own their coastal reefs and associated resources. Permission of the village leaders is required to access the reefs, which are typically closed to fishing or collecting by outsiders. Yap still maintains a type of caste system, which is reflected in a variety of cultural interactions including fishing techniques. Specific clans use unique tools for fishing, including breadfruit kites (Fig. 3), which employ loops of muscle bands from the pectoral/abdominal region of sharks instead of fish hooks. These bands are “danced” on the ocean’s surface and ensnare the teeth of long-nosed needlefish that bite them. Unlike nets and traps, this technique is an example of a specific type of fishing gear that is highly selective for a desired species with no unintended by-catch. On nearby Tam Tam Island, certain species of fish (e.g., the humphead parrotfish) can be eaten only by

FIGURE 3 Yapese fishing kite made from a breadfruit leaf, supported

with Pandanus spines and flown with coconut sennet. Shark muscle bands are used instead of hooks for capturing specific species of longnosed needlefish.

chiefs, reducing the fishing pressure on these ecologically important species. Large schools of these desirable fish, which are also important grazers keeping algae in check, can still be seen on many reefs where traditional leaders regulate fisheries, and community members comply. Although MPAs have gained popularity as a conservation tool in the United States and internationally during the past two decades, such restricted areas have been in place for centuries on some Pacific Islands. Fishers from Satawal still use traditional sailing canoes to travel to the uninhabited island of West Fayu to fish as a way of reducing pressure on the reefs of their home island (Fig. 4). In island systems like this, where bartering still takes place, and food is shared among families, there are no market forces pushing toward over-exploitation of coral reef fisheries, which, although diverse, cannot withstand heavy export pressure.

Palau provides one of the best examples of the value of traditional knowledge and its reapplication. Palau was part of the post–World War II Trust Territory of the Pacific Islands (TTPI) under U.S. jurisdiction. During the previous Japanese Mandate years through the TTPI period, internal governance and traditional practices within these islands were suspended, and in many cases, lost. Coral reef fisheries, which are not sustainable at high levels of exploitation, were found to be in serious decline by the 1980s as evidenced from data gathered by the Palau Marine Resources Division. Through a series of meetings with villagers and traditional leaders, the Marine Protection Act of 1994 was developed with the key provision of reestablishing the traditional system of bul, which allows traditional leaders to close sites and fisheries based on the location and timing of spawning and feeding aggregations. In addition, the commercial export of coral reef fishes was phased out. The combination of these traditionally based management approaches, now national law and part of the recently passed legislation on Palau’s Protected Area Network, is a model from which the United States and other nations can learn, with key elements of ownership, stewardship, responsibility, and legacy. American Samoa is a U.S. Territory that also has an intact traditional leadership system that approaches coral reef conservation using a village-based approach. The Matai system is one of extended family economic and political units, with the Matai as the leader. As with the other islands, modern American Samoan society is a hybrid of elected officials and traditional leaders. Through the Matai system, areas, species, and fisheries can be regulated and protected. American Samoa, using modern scientific data as well as traditional knowledge, has been proactive in protecting reef areas and large oceanic fishes. The American Samoa government is notable for recently establishing a population task force to address the key issue of island resource availability and carrying capacity as affected by immigration and local population growth. Land–Sea Connections

FIGURE 4 Traditional navigators and fishers from Satawal building a

new sailing canoe for fishing on West Fayu and for inter-island travel.

Because of the relatively small size of Pacific Islands or the steeply sloping topographic relief that defines watersheds on larger land masses, land-based activities quickly and drastically affect adjacent coastal and oceanic ecosystems. Traditional land partitioning in Hawai‘i is based on Ahupua‘a, parcels that extend from the tops of ridges to the sea. Such “ridge to reef ” systems are found in other Pacific Islands as well and reflect the understanding of the land–sea connection. Pacific Islanders are aware of and sensitive to upstream effects

R E E F E C O L O G Y A N D C O N S E R VA T I O N

783

on downstream communities, as activities often affect members of the same village. Coral reef conservation begins on land and requires an integrated watershed management approach. Village leaders in Enipein Village, Pohnpei, established a continuous protected area from the mountaintop rain forests through the coastal mangroves and out into the lagoon when studies found upland clearing for farming sakau (a plant whose root is a narcotizing agent and which is a major cash crop) was responsible for landslides and sedimentation stress on coastal coral reefs. Similarly, clearing and grading of mangroves for house lots in Airai Village, Palau, was halted when it was found that the loss of buffering mangroves was responsible for increased sedimentation impacts to adjacent coral reefs and associated fisheries. In both of these cases, data made available to traditional leaders resulted in policy development and implementation decisions within a period of weeks, rather than years as is often the case in Westernized democracies, which require legislation and in which compromise often limits the effectiveness of environmental protection measures. STEWARDSHIP VERSUS THE TRAGEDY OF THE COMMONS

Coral reef conservation practices in many of the Pacific Islands reflect a stewardship ethic that stems from ownership of the resource, as well as responsibility for addressing problems and their solutions. This is in contrast to the Western “Tragedy of the Commons” scenario, where there is universal ownership and no direct lines of responsibility by stakeholders. The advantage of traditional leadership, which still exists in a variety of forms in the Pacific, is the speed at which conservation-based decisions can be put into practice. For example, it took nearly five years for conservation policies to be developed and enforced following studies demonstrating overexploitation of a sea cucumber fishery in the Galápagos Islands. In Yap, it took only one day to move from data presentation to village chiefs to a policy decision to close the fishery. A distinguishing feature of most traditional leadership systems is community compliance without the need for legislation, enforcement activities, and legal proceedings. Recent efforts to cite “the right to fish” as part of traditional cultures to argue against the establishment of MPAs and other conservation measures are falsely based. Fishing, including the use of specific types of gear, access to particular areas, and the consumption of certain species, is a privilege granted by chiefs or master fishermen and was never a right. These privileges are granted by tradi-

784

R E E F E C O L O G Y A N D C O N S E R VAT I O N

tional leaders based on considerations including resource availability and sustainability and may be rescinded if conditions warrant. The Future of Coral Reefs

Many challenges remain in the area of coral reef conservation, especially as a result of global climate change and regional bleaching events superimposed over other anthropogenic and natural stressors. If reefs in their present (degraded) state are the legacy left to future generations, then today’s society will have failed as a steward of these remarkable ecosystems. Corals and coral reefs are resilient, and actions can be taken to reverse the present trend of increasing degradation and decline. Efforts at coral reef restoration are receiving heightened attention recently, but so far the best means of recovery is tied to restoring those conditions that allow natural recovery to occur. Select case histories have demonstrated that elements of traditional management practices that support the responsibility of communities for resource stewardship can be effective. A shift to community-based management in the Philippines, Indonesia, and other areas in the Coral Triangle has resulted in improved compliance and resource availability. Science is an important tool for guiding conservation efforts and for developing tools and metrics to determine the efficacy of mitigation measures. Emerging technologies, including the use of molecular biomarkers of stress, enable the determination of causeand-effect relationships between stressors and coral reef responses and can measure changes at the cellular and organismal level (for better or worse) within the timeframe of days and weeks, rather than the years needed using standard assessment and monitoring techniques focused on species diversity and abundance. The social sciences and economics are also critical elements of conservation program development, as the reasonable goal is to change preventable human behaviors responsible for coral reef decline, rather than to change the behavior of coral reef organisms. Conservation successes emerging in the Caribbean, Africa, and Asia have required buy-in from communities that have a vested interest in resource sustainability, as they allow for present needs to be met without compromising the future state of coral reefs and related resources. The two-, four-, and six-year electoral cycles of Western democracies often make longer-term considerations irrelevant to politicians looking to stay in office, yet it is the legacy concerns of traditional leaders and societies that provide the context for effective conservation-based policies and

their implementation. Enhanced awareness is needed to generate the political will required for policy development and implementation to insure the future of coral reefs. Bridging science to policy is a key step in moving from outputs to outcomes.

the Pleistocene, species that are endangered as a result of human actions or ongoing climate change are restricted to modern-day “refugia.” One common feature of refugia is that they may contain the greatest of relictual biodiversity, meaning that they merit special conservation attention.

SEE ALSO THE FOLLOWING ARTICLES

REFUGIA DURING GLACIAL–INTERGLACIAL CYCLES OF THE PLEISTOCENE

Coral / Fish Stocks/Overfishing / Global Warming / Marine Protected Areas FURTHER READING

Cinner J., M. Marnane, T. Clark, T. McClanahan, J. Ben, and R. Yamuna. 2005. Trade, tenure, and tradition: influence of sociocultural factors on resource use in Melanesia. Conservation Biology 19: 1469–1477. Downs, C. A., C. M. Woodley, R. H. Richmond, L. L. Lanning, and R. Owen. 2005. Shifting the paradigm for coral-reef ‘health’ assessment. Marine Pollution Bulletin 51: 486–494. Hughes T. P., and J. Connell. 1999. Multiple stressors on coral reefs: a long-term perspective. Limnology and Oceanography 44: 932–940. Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, B. J. Bourque, R. H. Bradbury, R. Cooke, J. Erlandson, J. A. Estes, T. P. Hughes, S. Kidwell, C. B. Lange, H. S. Lenihan, J. M. Pandolfi, C. H. Peterson, R. S. Steneck, M. J. Tegner, and R. R. Warner. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629–638. Johannes, R. E. 1978. Traditional marine conservation methods in Oceania and their demise. Annual Review of Ecology and Systematics 9: 349–364. Johannes, R. E. 1981. Words of the lagoon: fishing and marine lore in the Palau district of Micronesia. Berkeley: University of California Press. Johannes, R. E. 1997. Traditional coral-reef fisheries management, in Life and death of coral reefs. C. Birkeland, ed. New York: Chapman and Hall, 380–385. Pandolfi, J.M., R. H. Bradbury, E. Sala, T. P. Hughes, K. A. Bjorndal, R. G. Cooke, D. McArdle, L. McClenachan, M. J. Newman, G. Paredes, R. R. Warner, and J. B. Jackson. 2003. Global trajectories of the longterm decline of coral reef ecosystems. Science 301: 955–958. Richmond, R. H., T. Rongo, Y. Golbuu, S. Victor, N. Idechong, G. Davis, W. Kostka, L. Neth, M. Hamnett, and E. Wolanski. 2007. Watersheds and coral reefs: conservation science, policy and implementation. BioScience 57: 598–607. Wolanski, E., R. H. Richmond, and L. McCook. 2004. A model of the effects of land based human activities on the health of coral reefs in the Great Barrier Reef and in Fouha Bay, Guam, Micronesia. Journal of Marine Systems 46: 133–144.

REFUGIA ANGUS DAVISON University of Nottingham, United Kingdom

A refugium is a geographic area in which organisms survive during adverse conditions. Although the term is most frequently applied to the glacial–interglacial cycles of

The impact of the Pleistocene glacial–interglacial cycles on the flora and fauna of high latitudes has been well characterized, especially in North America and Europe. In these regions, large tracts of land were covered by ice caps during the glacial periods (or were under permafrost), so temperate species were restricted to southern refugia. In contrast, the impact of the glacial cycles at low latitudes has largely been overlooked, especially on oceanic islands. The form and extent of refugia in these regions and their role in shaping biodiversity are poorly understood, yet also controversial, because of their potential impact as “drivers” of speciation. This lack of knowledge is unfortunate in light of the traditional and continuing contribution of island species (e.g., Darwin’s finches, Partula snails, Lord Howe Island palms) to understanding speciation theory. Changes in sea level are generally considered to be one of the most important causes for islands appearing and disappearing, so one main impact of high-latitude glaciations was to increase globally the area of land above sea level. At the greatest extent of glaciation, when the sea level was about 100–130 m below that of the present day, many islands both were larger and had a greater range of elevations. Other islands became incorporated into nearby mainland (e.g., Britain), and archipelagoes became a single island (Figure 1, 2). Some generalist species may therefore have reached their greatest extent and population size during the glaciations—meaning, in effect, that their present-day distribution is refugial. A second main impact of the high-latitude glaciations was a global change in climate. At high latitudes, islands had extensive ice caps, so many species must have gone extinct, subsequently recolonizing from southerly mainland refugia. At low latitudes, it is more difficult to generalize, and the data are sparse, because climatic adjustments were more local in their action, especially on oceanic islands. In consequence, the nature of potential refugia remains undetermined for the vast majority of low-latitude island species. Because island species are often uniquely adapted to specific habitats, having undergone an adaptive radiation, then perhaps their most likely response during glaciations was to track available habi-

REFUGIA

785

FIGURE 2 Mt. Chubu on Hahajima is 463 m above sea level, with moist

forest at the summit. At the height of glaciations, the mountain would have been ~570 m above sea level, so the habitat diversity was likely greater. Genetic data indicate that snails in the region of this mountain did not undergo bottlenecks, unlike populations sampled from the low-lying peninsula in the foreground.

FIGURE 1 The Hahajima islands south of Japan are presently an archi-

more severe bottlenecks than those in the central highlands. At the opposite climatic extreme, it is claimed that some species survived the glaciations on islands in northerly refugia that happened to be free of ice. Other species may have clung on on mountain nunataks.

pelago. During the Pleistocene glaciations, they were a single island with a much greater geographic extent (gray line).

tat. If so, the geographic extant of refugia on islands was likely dependent upon the range of available elevations and consequent habitat zones. More generally, populations may sometimes become geographically restricted to refugia, so that they speciate in allopatry (e.g., in caves or in islands of forest surrounded by lava). Several methods can be used to discover places that historically served as refugia, with molecular genetic and paleobotanical studies perhaps the most powerful. In the Neotropics, one explanation for a correspondence between paleo-pollen diversity and changes in global temperature is that fluctuating climate forced plants into refugial habitat islands, directly enabling the diversification of species in allopatry, and so explaining the extraordinary diversity that is found there. In the rather few tropical island species that have been investigated, the overall impact of the Pleistocene on biodiversity is unclear. The genetic diversity of the damselfly Megalagrion xanthomelas is greatest on the Big Island of Hawaii, where habitat availability has been most constant during glacial cycling; similarly, genetic data indicate that Mandarina snails in the low lying islands of the Japanese Bonin Islands (Ogasawara) underwent much

786

REFUGIA

MODERN-DAY REFUGIA

As centers of endemism, islands harbor a high proportion of species that are found nowhere else. Island species are also particularly vulnerable to extinction and have been especially affected by introduced exotics. Nonetheless, islands are also frequently the last remaining refuges for many species, either by accident or as a consequence of their remoteness; species that were long since made extinct on the mainland, linger on remote islands. Prominent examples include the Lord Howe stick insect Dryococelus australis, recently rediscovered after being considered extinct for 70 years, and the tuatara Sphenodon, confined to a handful of offshore islands and mainland locations of New Zealand. Several Hawaiian forest birds survive in high-elevation refugia, where they cannot be reached by mosquitoes carrying avian malaria. Islands are also sometimes used as natural holding grounds (refugia) for seminatural breeding; introduced predators can be extirpated from offshore islands, and then endangered species (re)introduced. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Climate Change / Galápagos Finches / Land Snails / Lord Howe Island / Sea-Level Change / Vicariance

FURTHER READING

Davison, A., and S. Chiba. 2006. The recent history and population structure of five Mandarina snail species from sub-tropical Ogasawara (Bonin Islands, Japan). Molecular Ecology 15: 2905–2919. Hewitt, G. M. 2004. Genetic consequences of climatic oscillations in the Quaternary. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 359: 183–195. Jaramillo, C., M. J. Rueda, and G. Mora. 2006. Cenozoic plant diversity in the Neotropics. Science 311: 1893–1896. Priddel, D., N. Carlile, M. Humphrey, S. Fellenberg, and D. Hiscox. 2003. Rediscovery of the ‘extinct’ Lord Howe Island stick-insect (Dryococelus australis (Montrouzier)) (Phasmatodea) and recommendations for its conservation. Biodiversity and Conservation 12: 1391–1403.

RELAXATION

FIGURE 1 A hypothetical example of how an increase in local extinc-

tion rates following a disturbance that reduces the area of an island

KENNETH J. FEELEY

(e.g., a rise in sea level) will decrease the equilibrium number of spe-

Wake Forest University, Winston-Salem, North Carolina

cies that can be sustained. The resulting loss of species richness over time is referred to as relaxation.

Relaxation is the process by which species are lost from an island following a disturbance event that increases the rate of local extinction, decreases the rate of colonization, or both. The disturbance is typically a decrease in area or increase in isolation, but it may alternatively be a change in habitat quality or any other disturbance event that decreases the number of species that the island can support at equilibrium (Fig. 1). Relaxation is most commonly referred to in the setting of continental islands or mainland habitat fragments (e.g., sky islands). In both of these settings, habitat that was originally connected to, or embedded in, a larger “mainland” becomes isolated (e.g., in the case of continental islands, by rising sea levels). Following isolation, the newly formed island will generally be “supersaturated,” meaning that the number of species occurring on the island is greater than can be supported as based on the new conditions. In a supersaturated community, extinctions will exceed colonizations and the number of species will decrease through time until the appropriate equilibrium number is achieved. PATTERNS OF RELAXATION AMONG ISLANDS

Studies tracking the process of relaxation have observed that the rate of species loss generally mimics an exponential decay process: species loss is rapid at first and then decelerates through time until the new equilibrium rate of extinction is reached. Furthermore, it has been noted that the rate of relaxation is not constant between islands but may depend on several factors, including area and degree of isolation. Large islands (or habitat fragments) typically lose species slower than do smaller islands.

The reason for the relationship between area and rate of relaxation is poorly understood. One potential explanation is that if extinctions are primarily a passive process resulting from stochastic population fluctuations, then the reduced populations occurring on smaller islands will have elevated risks of extinction, resulting in accelerated rates of relaxation. Alternatively, the loss of area on smaller islands may lead to synergistic effects, such as changes in resource availability or predation pressures, that can in turn hasten species loss. For example, numerous studies have shown that the increased relative edge length (perimeter/area) on small islands may lead to deleterious “edge effects,” including changes in abiotic conditions (such as temperature, humidity, or wind speed), as well as changes in the biotic environment such as greater exposure to edge-frequenting predators. Conversely, a few studies have demonstrated “positive” edge effects that may actually act to decrease relaxation rates on small islands. Examples of positive edge effects include the allochthonous input of marine nutrients via seabird guano or ocean detritus. PATTERNS OF RELAXATION AMONG SPECIES

The order in which species go locally extinct during relaxation is generally not random. Rather, certain species or groups of species appear to be particularly sensitive to changes in area and hence tend to be lost first. Extinction vulnerability has been variably related to a large number of species characteristics, depending on the system and/or group of species

R E L A X AT I O N

787

studied. Examples of species attributes that have been related to extinction vulnerability include population size, trophic position, dispersal capability, degree of resource specialization, body size, and reproductive rate. As a result of selective extinction processes, newly formed archipelagoes or networks of habitat fragments often exhibit “nested” patterns of species distributions, such that species occurring on speciespoor islands also occur on all of the more diverse islands.

for scientific studies. Some island research stations are little more than specialized guest houses, but others rival the most advanced mainland laboratories. Where a station falls along this spectrum depends on a number of factors, including (1) the scientific interest of the location, (2) its socioeconomic context, and (3) its potential to support educational and public service activities. All of these are being radically transformed by (4) new information and communication technologies.

CONSERVATION IMPLICATIONS

As a result of relaxation, isolated reserves will not sustain the same number of species as they did pre-isolation. By understanding the processes and patterns of relaxation it may be possible to predict the rates at which species will be lost as well as which species are at the greatest risk of extinction. This information can in turn help improve reserve design and management. SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / Extinction / Fragmentation / Island Biogeography, Theory of / Sky Islands / Species–Area Relationship BIBLIOGRAPHY

Brown, J. H. 1971. Mammals on mountaintops: nonequilibrium insular biogeography. The American Naturalist 105(945): 467–478. Diamond, J. M. 1972. Biogeographic kinetics: estimation of relaxation times for avifaunas of Southwest Pacific. Proceedings of the National Academy of Sciences of the United States of America 69(11): 3199–3203. Karr, J. R. 1982. Population variability and extinction in the avifauna of a tropical land bridge island. Ecology 63 (6): 1975–1978. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. , Princeton, NJ: Princeton University Press. Patterson, B. D. 1987. The principle of nested subsets and its implications for biological conservation. Conservation Biology 1(4): 323–334. Pimm, S. L., H. L. Jones, and J. M. Diamond. 1988. On the risk of extinction. The American Naturalist 132(6): 757–785. Terborgh, J. W. 1974. Preservation of natural diversity: the problem of extinction prone species. BioScience 24(12): 715–722. Wilcox, B. A. 1978. Supersaturated island faunas: a species-age relationship for lizards on post-Pleistocene land-bridge islands. Science 199(4332): 996–998.

RESEARCH STATIONS NEIL DAVIES University of California, Berkeley

Most islands have limited indigenous resources (human, financial, and physical) to support research, and they rely on importing the people (skills) and equipment needed

788

R E S E A R C H S TAT I O N S

SCIENTIFIC POTENTIAL

Studies at research stations investigate natural processes (including the ecological role of humans) through experiment and observation. Sites with a field station or marine laboratory represent parts of the world where we can increase the precision of biophysical observations. A research station is thus a significant investment in a geographic area of particular scientific interest. Far from research stations, the ecological matrix is fuzzy, but it snaps into focus closer to them. Researchers consider archipelagoes to be natural laboratories, providing systems of discrete replicates facilitating the design of scientific studies. Such natural systems are particularly important for scientists who, for ethical and/or practical reasons, cannot easily manipulate their study subjects. For example, one cannot put a species (humans in the case of anthropology) in separate cages, each differing in some key environmental factor, and leave them for a million years to see what evolves. Natural experiments are messier than laboratory studies, to be sure, but researchers often find it easier to separate the signal from noise in island systems. Another advantage of smaller islands for science is that they are complete but relatively simple ecosystems and thus are scientifically more tractable for holistic study than are larger more complex regions. Major scientific breakthroughs often result from the focus on such model systems. For example, cell and molecular biology advanced tremendously by concentrating efforts on a few model species such as mice, fruit flies (Drosophila), and nematodes (Caenorhabditis elegans). Similarly, rapid progress in ecology and evolution has already come from studies of island model ecosystems, but much more potential remains to be explored. Nor is it sufficient to investigate only one model: Integration across a network of ecosystems is critical for understanding regional or global drivers of local change. Comparisons of biologically complex islands (e.g., those in the western Pacific) with simpler ones (e.g., those in eastern Polynesia) can provide insights into the role of biodiversity in ecosystem structure, function, and resilience to perturbations such

Just as they significantly influenced science in the nineteenth and twentieth centuries (e.g., inspiring Darwin), islands and their research stations are particularly well placed to advance the frontiers of science in the twentyfirst century. For example, biocomplexity addresses the causes of environmental, biological, and social changes and the interaction among them. Sustainable development relies on ecointelligence—understanding biocomplexity and managing human society’s relationship with the natural world rationally. The interacting processes underlying biocomplexity occur at multiple spatial scales, and island research stations enable the simultaneous study of local drivers (island- and archipelago-scale) as well as regional drivers such as climate change or globalization (oceanic/continental-scale). Island research stations therefore address issues of general concern to human society as well as more applied projects of primarily local importance. As the new knowledge generated by fundamental research is shared beyond the local community, so is the potential funding base commensurately larger. To succeed in attracting (national or international) support, however, island research stations must overcome significant barriers. Although all islands benefit from local investment in applied science, only some are able to attract the external resources necessary to sustain long-term fundamental research.

as climate change and globalization (e.g., invasive species). These represent some of the most important issues facing science and society in the twenty-first century. The fundamental scientific questions addressed on islands often invoke a long time scale. Research stations provide permanent facilities in relatively remote areas and represent a source of programmatic continuity as institutions that persist after individual researchers leave or particular projects end. Research stations host long-term monitoring programs as well as shorter-term, processoriented studies. Monitoring programs do not always address an explicit research question, and mechanistic studies are sometimes a series of quasi-independent projects connected more by their common geography than by an integrated research program. Exceptions exist, however, such as the U.S. National Science Foundation’s (NSF) Long-Term Ecological Research (LTER) program, which combines empirical studies with the long-term data acquisition needed to address specific questions over appropriate ecological timescales. The NSF-funded LTER network includes a number of sites based at island research stations, from the Pacific (Gump Station, Moorea) to the Caribbean (El Verde Field Station, Puerto Rico) and the Antarctic (Palmer Station, Anvers Island). The International LTER (ILTER) initiative had 32 member networks in May 2006, and as these develop, they are likely to include more island sites and associated research stations. The LTER-Europe was founded in June 2007 and could be particularly influential given the global distribution of islands associated with the European Union (Fig. 1).

SOCIOECONOMIC CONTEXT

Field research, whether marine or terrestrial, refers to those scientific studies that take place outside the confines of

Anguilla British Virginia Is. Turks & Caicos Caiman Islands Montserrat Greenland

St Pierre & Miquelon Azores Madeira Canary Islands Guadeloupe Martinique Aruba Dutch Antilles

French Guyana Ascension

Wallis & Futuna Mayotte

St Helena

Reunion

Tristan de Cunha South Georgia Falklands

Scattered Islands

New Caledonia

Pitcairn

Amsterdam St. Paul

Crozet Islands

Iles Sandwich BAT (British Antarctic Territory)

Overseas Regions (ORs)

French Polynesia

BIOT (British Indian Ocean Ter.)

TAAF (Terncs Australes ct Antartiques Françaises) Kerguelen Adélie Land

Overseas Contries and Territories (OCRs)

FIGURE 1 The European Union’s seven Overseas Regions, and 20 Countries and Territories. Reprinted with permission from IUCN.

R E S E A R C H S TAT I O N S

789

the laboratory. In other words, it describes the investigation of the natural world in vivo. In theory, field research can be conducted anywhere in the world. For practical economic and cultural reasons, however, it tends to be concentrated in the environs of large cities in the developed world within easy reach of major museums and universities. Field research is more challenging further from these core centers, if only because of the travel costs and the need to carry equipment to a makeshift base camp (which, in rural or wilderness settings, can lack even the most essential utilities such as power and water). Such forays are described as “expeditions” or “field trips.” In some (rare) cases, the base camps become permanent, and a research station is established. Given their isolation and small size, all research stations are “islands” in a socioeconomic sense, lacking the economies of scale of a large campus. The decision to invest in such permanent infrastructure is thus a trade-off between the scientific productivity of a locality and the increased costs of conducting research there. Although the decision to establish or maintain a research station might be made after a rigorous assessment of the scientific return on investment, many island research stations were not established on the basis of a carefully formulated strategic plan. Nevertheless, the subsequent development of research stations depends on their continued ability to attract researchers and funds. A process of natural selection, therefore, eventually determines whether island research stations survive or perish: Schumpeterian, or Darwinian, creative destruction! A non-exhaustive list of existing island research stations is shown in Table 1. Of course, the need to attract researchers is not unique to island research stations; mainland institutions must also struggle to recruit the best scientists and students. Research stations, however, tend to host rather than employ the majority of their scientists, and so they face challenges that are perhaps more familiar to the tourism industry than to academia. This is perhaps a fortuitous parallel, as it aligns island research stations with a major source of external revenue for island economies. Research stations are part of the tourism sector but with an important difference: Their clients, visiting researchers, not only spend hard currency in the local supermarket but also con-

tribute to the island’s heritage capital. By increasing the knowledge base of the island, researchers help to improve the management of its natural and cultural resources (a prerequisite for sustainable tourism), as well as providing even more direct benefits for ecotourism. EDUCATION AND PUBLIC SERVICE

If research stations can be seen as part of an island’s tourism sector, then students are the ultimate ecotourists. Like any visitor, students might appreciate the beaches or nightlife of an island, but their primary purpose is to learn. Other tourists might share this motivation, but it is not usually their sole reason to travel. Many research stations are financially viable thanks to the income they generate from teaching field courses. This is not just a moneymaking activity, however, as it also fulfils a core mission of many research stations, particularly those operated by universities. Combining research with advanced training programs provides increased economies of scale, helping to overcome some of the problems of being a small and remote site. It also offers richer experiences for researchers and students alike. Many research stations provide education and training for the local community in addition to their programs for visiting students. In this case, the outreach provides another crucial function, reinforcing the social contract between the research station and the local community. Research stations are a valuable tool for sustainable development and a source of conservation know-how, and such public service is another core part of their mission. Indeed, research stations sometimes represent the only scientific expertise available in some locations and represent an important link for island communities to the international network of scientific knowledge and resources. Public service can also include applied research projects commissioned (and funded) at the local level to solve local problems. This is a win-win exchange that provides the research station with more knowledge about the local ecosystem (as well as revenue toward its operations overhead) while the island receives high-quality services at lower cost thanks to the capital (human and material) already invested in the station (often from external sources).

TABLE 1

Partial Listings of Island Research Stations

World Register of Field Centers, Royal Geographical Society National Association of Marine Labs European Marine Research Stations Network Organization of Biological Field Stations Association of Marine Laboratories of the Caribbean

790

R E S E A R C H S TAT I O N S

http://www.rgs.org/ http://www.naml.org/ http://www.marsnetwork.org/ http://www.obfs.org/ http://www.amlc-carib.org/

The educational and outreach products of island research stations can have impacts far beyond the local community. Islands make particularly charismatic showcases illustrating the challenges facing modern (and prehistoric) societies. Many of these are most evident in insular systems and are being documented by today’s island research stations. For example, rising sea levels and elevated ocean temperatures are submerging tropical islands and increasing the bleaching of their coral reefs. Similar impacts will be experienced in mainland systems, but islands are the “canaries in the mine” warning us of the impending global crisis. Island research stations are not only documenting the present and modeling the future: They are also reconstructing the past. Some traditional island societies succeeded in establishing sustainable economies despite the finite natural resources available to them. Others failed, and overexploitation led to collapse with terrible consequences for the ecosystem and the well being of its human population. Such lessons provide a vital service reminding humanity that the Earth is our common “island” home. TECHNOLOGY

Advances in information and communication technologies have revolutionized the potential of islands in many economic, social, and cultural arenas, including scientific research. These technologies are particularly disruptive for island economies because they attack the root of an island’s comparative disadvantage, its isolation. It is now possible for a knowledge worker to live in the mountains of California, or the islands of Polynesia, and be (virtually) as connected to his or her colleagues and clients as if he or she were in a major metropolitan area. Interestingly, this pulls in two directions: It is becoming less necessary for scientists to physically visit a remote site in order to study it, but at the same time, it is now possible to live permanently at a remote site without being entirely disconnected from the intellectual environment of a large city. These two trends will transform research stations (and much else) on islands. Research stations occupy a key position at the beginning of the digital supply chain that delivers ecological knowledge to the global scientific community. Historically, they catered to the most basic human needs (shelter, food, etc.), but modern research stations will need to support advanced sensor technologies embedded in their field sites. Increasingly, research stations will host engineers and computer scientists as well as their more traditional clientele. Furthermore, the technologies associated with genomics and molecular biology are also being applied

outside of the laboratory, enabling whole new worlds to be studied in the field (e.g., microbial communities and their ecological role). One of the most basic requirements for scientific investigation is a detailed description of the study system. Thus, the core infrastructure of a research station is not just its buildings, equipment, and staff but also a digital rendition (preferably in real time) of the island’s physical, biological, and social characteristics. This “living encyclopedia” must be readily accessible and motivates the development of online climatic databases (weather records), maps (GIS), and biotic guides (species identification), as well as a host of sophisticated informatics tools. Just as model species such as the fruit fly or mouse were described in ever-increasing detail as technology advanced (e.g., whole-genome sequencing), island model ecosystems are also being “sequenced” (e.g., using short genetic “bar codes” that enable rapid species identification). Most research stations already maintain databases of meteorological data (technologically the most simple to collect and archive), but soon they will also participate in the acquisition and management of massive amounts of information (environmental, biological, and social). As they become more technology intensive, research stations are increasingly being compared to astronomical observatories. For the moment, however, there are no known systems in the universe as complex as those on Earth, and none more immediately important for us to understand. Ecological observatories on the “blue planet” will require (and will probably receive) unprecedented levels of investment as society wakes up to the twin crises of climate change and mass extinction. Islands and their research stations will be on the front line of the epic struggle for sustainable development. SEE ALSO THE FOLLOWING ARTICLES

Global Warming / Marine Protected Areas / Reef Ecology and Conservation / Sea-Level Change / Sustainability FURTHER READING

Check, E. 2006. Treasure island: pinning down a model ecosystem. Nature 439: 378–379. Diamond, J. 2005 Collapse: how societies choose to fail or succeed. New York: Viking, Penguin Group. ILTER (International Long-Term Ecological Research Network). http:// www.ilternet.edu/ LTER-EU (Long-Term Ecosystem Research and Monitoring, European Union). http://www.lter-europe.ceh.ac.uk/ LTER-US (Long-Term Ecological Research Network, United States) http://www.lternet.edu/ Michener, W. K., T. J. Baerwald, P. Firth, M. A. Palmer, J. L. Rosenberg, E. A. Sandlin, and H. Zimmerman. 2001. Defining and unraveling biocomplexity. Bioscience 51: 1018–1023.

R E S E A R C H S TAT I O N S

791

Vitousek, P. M. 2002. Oceanic islands as model systems for ecological studies. Journal of Biogeography 29: 573–582.

RÉUNION SEE INDIAN REGION

RODENTS DAVID TOWNS New Zealand Department of Conservation, Newton

Islands throughout the world have been modified by introduced rodents. A few highly destructive species have been deliberately released to establish a fur industry. Mice and four species of rats have reached islands as passengers during exploration, warfare, and commerce. These have become the perfect invasive species, able to spread over wide distances and with significant ecological effects wherever they colonize. DISTRIBUTION OF RODENTS ON ISLANDS

There are more species of rodents (around 2000) than of any other group of mammals. Isolation and extraordinary dispersal ability have led to considerable evolutionary radiation on less remote islands and continental fragments, with distinctive faunas in Madagascar, Sri Lanka, the Philippines, Australia, and even the Galápagos Islands. Madagascar alone has 22 species of native rodents. However, three species of rats and one of mice are distinctive—not for their evolutionary significance, but for a long history of association with people, for their ability to stow away on all kinds of seagoing craft, and for their spread across enormous geographic distances. They now occupy all continents other than Antarctica and most islands from the subantarctic Southern Hemisphere to the arctic Northern Hemisphere. In order of distances covered, these species are the Pacific FIGURE 1 The four rodent species most widely spread by people. (A)

A wild house mouse (Mus musculus) and offspring. (B) Norway rat (Rattus norvegicus). Photographs courtesy of New Zealand Department of Conservation—Rod Morris. (C) Ship rat (R. rattus) attacking a native forest bird at its nest in New Zealand. Photograph by David Mudge. (D) Pacific rat (R. exulans) in New Zealand forest. Photograph courtesy of New Zealand Department of Conservation—Dick Veitch.

792

RODENTS

rat or kiore (Rattus exulans), the Norway or brown rat (R. norvegicus), the ship or black rat (R. rattus), and the house mouse (Mus musculus) (Fig. 1). The origin of the house mouse is unclear, although current evidence points to the Indian subcontinent. They have since been spread to all inhabited parts of the world and may be the most widely distributed mammal other than people. Similarly, Norway rats have become so widespread that their center of origin is unclear, but it may be China or Siberia. They now live from South Georgia in the South Atlantic Ocean to the northernmost islands of Europe and North America. Ship rats probably originated in India, but they have since reached islands from the southern Indian Ocean to Scotland. Pacific rats are derived from Southeast Asia but have been spread across the Pacific basin as far east as Easter Island and south to New Zealand. In combination, mice and these three rats have reached at least 80% of the world’s island groups. SPREAD TO ISLANDS Origins

Rodents have been spread to islands in four ways: in cargo, by abandoning ships in port or after shipwrecks, through natural dispersal either by swimming or floating on debris, and through deliberate spreading by people. As examples, ship rats escaped onto Midway Island from military stores offloaded during World War II. The same species invaded Big South Cape Island off southern New Zealand along the mooring lines of fishing boats in about 1962. Norway rats reached Raoul Island in northern New Zealand after the wreck of the Columbia River in 1921. In 2004, a male Norway rat swam at least 400 m between islands in New Zealand while carrying a radio transmitter. In contrast, ship rats are reluctant or unable to swim more than 200 m, and Pacific rats failed to invade some islands separated by only 50 m. Therefore, the distribution of mice and of the more recently arrived rats is most likely to have been accidental through rodents abandoning vessels or being hidden in stores. However, some anthropologists have claimed that Polynesians deliberately introduced Pacific rats to islands as a source of food. Other analyses have found that Pacific rats were more likely to be present on islands with easy access by canoe than on islands where landings were more difficult. The pathways for the spread of rodents are now being determined by genetic analyses using mitochondrial DNA. This can be used to trace the origins of species spread accidentally by boats or transported by people many centuries ago. Additional species have been deliberately spread in attempts to establish fur industries. For example, Ameri-

can beavers (Castor canadensis) were introduced to the island of Tierra del Fuego in South America, and also to islands off western Canada. Speed of Invasion

The invasion biology of rodents on islands has only recently been studied. Some invasions have certainly originated from a single pregnant female rat or mouse. This raises the question of how they avoid the effects of inbreeding. In New Zealand, pregnant female Norway rats were found carrying the embryos of multiple fathers; one litter of rats can include the genetic information from several male lines. If invading rodents become established, populations can expand with great speed. For example, the ship rats that invaded Big South Cape Island (900 ha) off southern New Zealand, and the single known female Norway rat that in 1995 reached Frégate Island (210 ha) in the Seychelles, each produced sufficient offspring to spread throughout the entire island within two years of arrival. Population Densities

Once established, rodent populations on islands may periodically reach very high densities. For example, mice on Gough Island reached densities of 224 per ha, and in pasture on Mana Island, New Zealand, they became so abundant that improvised bucket traps caught over 200 per night. Elsewhere on New Zealand islands, Norway rats reached densities of 13 per ha, ship rats were recorded at densities of 50 per ha, and Pacific rats in grasslands were noted at densities of greater than 100 per ha. On many islands, the populations go through wide fluctuations in abundance. All three species decline during the cool winter in New Zealand, and Pacific rats similarly decline during the dry season on some islands in the tropical Pacific. EFFECTS ON ISLAND FAUNA AND FLORA Species That Modify Environments

Introductions of beavers and coypu (Myocastor coypus) in failed attempts to establish a fur industry have led to radically transformed landscapes. Beavers have modified watercourses by constructing ponds, and in South America they leave extensive areas of dead southern beech (Nothofagus) forest. Similarly, coypu undermined the banks of watercourses and damaged wetlands. Gray squirrels (Sciurus carolinensis) introduced to islands off Canada stripped the bark from sensitive plant species and damaged the acorns of native oaks. As a result, forest

RODENTS

793

composition has changed on islands where alien squirrel populations established. Species That Modify Biodiversity EFFECTS OF MICE

On islands, mice primarily feed on seeds and small invertebrates such as insect larvae. On Marion Island in the southeastern Atlantic Ocean, mice were found to have significant effects on the caterpillars of an endemic species of moth. On Gough Island, landbirds became confined to cliffs inaccessible to the mice. The mice also fed on the living chicks of albatrosses until the birds died of their wounds. A 90% decline of Tristan Albatross may be attributable to predation by mice. On Mana Island, populations of giant flightless crickets, geckos, and large nocturnal skinks greatly increased after mice were removed, which indicates that mice were affecting species equal to or greater than their own body weight. However, despite their widespread distribution, in general the effects of mice on native species are poorly known. EFFECTS OF RATS

Information on the effects of rats, although better documented than for mice, is still scattered, with little comparison between different species that occupy similar biogeographic regions. Much of the data available has been obtained on islands in New Zealand, but even there, the studies have been selective. Like mice, rats are omnivores and carrion feeders. Unlike mice, the diet of rats is extremely wide and includes plants, invertebrates, amphibians and reptiles, birds, and other mammals. Effects on Plants All three invasive species of rats affect forest plants largely through predation of seeds, but in rare cases, also through browsing seedlings. They may also climb trees and feed on flowers and fruit. For example, the tree-climbing abilities of ship rats led to them becoming a threat to macadamia nut production in the Hawaiian Islands. In New Zealand, Pacific rats severely suppress at least two species of forest canopy trees, which on some islands have become rare. They also suppress at least another nine and perhaps up to 17 species of forest plants, one of which is a native palm. Norway rats can have similar effects and were responsible for recruitment failure of southern beech on Breaksea Island. In the Canary Islands off northwestern Africa, ship rats consume about half of the fleshy-fruited tree species of the laurel forest and may be partly responsible for changes in forest structure. Particularly devastating effects are attributed to rats in the tropical Pacific. Heavy seed predation by Pacific rats on Rapa Nui (Easter Island) appears to have

794

RODENTS

been instrumental in the loss of palm forests. Pacific rats have also been associated with extreme forest modification on the Hawaiian Islands. Rats also have indirect effects on plants. In the Canary Islands, ship rats compete with the Canary robin for the fruits of Vibernum, reducing the distance that seeds are dispersed. In the Balearic Islands, rats affect fruit-eating lacertid lizards, which are the main seed dispersers for species of endemic plants. Effects on Invertebrates Rats feed on a very wide range of invertebrates. Numerous species of invertebrates are eaten by rats in the Hawaiian Islands, where entomologists have even discovered unknown species in rat stomachs. Some species of rats feed at husking sites where invertebrate fragments can be collected and identified. Husking sites of Pacific rats analyzed in New Zealand revealed the remains of over 60 species of native invertebrates. Such predation may not have serious effects on some species, but others have disappeared from islands invaded by the rats. Particularly vulnerable species are those that are nocturnal, large, and flightless, including one species of large (miturgid) spider, darkling (tenebrionid) beetles, and large flightless (anostostomatid) crickets. The invasion of ship rats on Big South Cape Island led to local extinction of large Hadramphus weevils, either through predation or because of heavy damage to the weevil’s host plants. Evidence of heavy predation on subfossil deposits of large Placostylus land snails in New Zealand proved useful in estimating the time of invasion by the rats. In some locations, the snails became extinct after the rats arrived. Other invertebrates affected by rats include Gecarcinus crabs in the Caribbean. Effects on Vertebrates The three invasive rat species have been associated with declines in all major groups of vertebrates, ranging from amphibians to mammals. In New Zealand, Pacific rats appear responsible for the loss of three species of endemic frogs, for range reductions of geckos, for local extinctions of at least six species of nocturnal skinks, and for breeding failure of at least four species of small seabirds. When Pacific rats were removed from islands around New Zealand, juvenile recruitment increased for the large endemic iguana-like tuatara (Sphenodon) and for three species of skinks and two species of geckos. Pacific rats also probably eliminated five species of small ground-dwelling birds in New Zealand and are also likely to have directly or indirectly instigated the demise of some species in Hawaii. Norway rats have been found to affect a wider range of vertebrates than have Pacific rats, including a range of larger ground-dwelling birds such as endemic wrens in

the Falkland Islands and seabirds in New Zealand that weigh up to 750 g. Norway rats appear to have wiped out the endemic Canary Island lark and are implicated in the declines of seabird populations in numerous island groups. Along with ship rats, Norway rats have heavily suppressed populations of burrowing seabirds around New Zealand. The most damaging species of rat is the ship rat because of its climbing abilities. In the Caribbean, ship rats almost wiped out an endemic species of racer snake. In the Canary Islands, they destroyed up to 90% of the nests of endemic pigeons through egg predation. In New Zealand, sudden irruptions of ship rats are associated with dramatic declines of a forest bird, the yellowhead, and the orange-fronted parakeet. When ship rats invaded Big South Cape Island, three species of terrestrial birds and one bat became extinct. Likewise, when ship rats escaped a wrecked ship, five species of endemic forest birds became extinct on Lord Howe Island. Groups affected by ship rats range from bats to other species of rodents, including two species of endemic rodents in the Pacific, and perhaps another in the Balearic Islands, that went extinct. Effects on Island Ecosystems In New Zealand, comparisons of islands invaded by rats with those uninvaded revealed that islands without rats had greater seabird burrow densities, higher soil fertility, greater abundance of primary consumers (herbivorous nematodes and land snails), and larger numbers of secondary consumers (enchytraeids, microbe-feeding nematodes, rotifers, and collembolans). Furthermore, plants in soil from rat-free islands grew more rapidly than did those inhabited by rats. The study revealed complex interactions between belowground food webs and aboveground plant nutrient levels and biomass related to changes in soil fertility and disturbance as an indirect result of invasions by rats. Effects on People Pacific rats were widespread throughout Polynesia by the thirteenth century, and in some parts they became a valued item of food for local people, although in New Zealand they were also a pest because they ate stored food, and Maori people built elaborate storage structures to exclude them. The spread of ship rats presented other problems. When they reached Madagascar in the fourteenth century, they carried bubonic plague, with serious effects on the local inhabitants. Disease carried by rats remains a problem in Madagascar.

include mice and all three species of rats. From such programs’ beginnings on small islands in the early 1960s, rats have been removed from over 300 islands around the world. The largest eradication was of Norway rats from 11,000-ha Campbell Island off southern New Zealand. These large eradications have been made possible by the combined use of helicopters with purpose-built bait spreaders, modern GPS systems to ensure accuracy of bait spread, and second generation toxicants that are highly effective. The eradications can be controversial, especially if it is unclear how rats have affected native species or if there are native species that might themselves be sensitive to the toxicants. One such example was Anacapa Island off California, where the eradication of ship rats in order to protect seabirds could have threatened a local subspecies of native Peromyscus deer mouse. The solution was to remove large numbers of deer mice and hold them in captivity during the eradication. Once returned to the island, Peromyscus became far more abundant than it was while the rats were present, indicating that the rats had suppressed this species, as well as the seabirds. Controversy would be reduced if the benefits of eradications were better known. Rapid responses have been reported after eradications. When Pacific rats were removed, the fledging success of Cook’s petrels immediately increased in New Zealand. After Norway rats were removed, native shrews became more abundant in islands off France, and a rare snipe recolonized Campbell Island off New Zealand. After ship rats were removed, rare seabirds successfully nested on Anacapa Island, and burrowing seabirds recolonized St. Paul Island in the South Indian Ocean. For these benefits to persist, it is necessary to prevent reinvasion, which is possible if islands are regularly checked, if there are restrictions of movement of bulk materials, and if care is taken with all luggage transported ashore. Numerous devices are now available that can detect invading rodents. Studies in New Zealand indicate that the most effective detection is with specially trained rat dogs. Unfortunately, the effects of invasions are most often only realized well after the rodents have established and unmistakeable changes to the island plants and animals have occurred. The speed and range of spread of these four species of rodents demonstrate that there is only one way to avoid the loss of additional island species: constant vigilance.

REMOVAL FROM ISLANDS

SEE ALSO THE FOLLOWING ARTICLES

Eradication programs against invasive rodents on islands have become increasingly successful, and their targets

Biological Control / Dispersal / Invasion Biology / Madagascar / New Zealand, Biology

RODENTS

795

FURTHER READING

Atkinson, I. A. E. 1985. The spread of commensal species of Rattus to oceanic islands and their effects on island avifaunas, in Conservation of island birds. P. J. Moors, ed. ICBP Technical Publication Number 3, 35–81. Fukami, T., D. A. Wardle, P. A. Bellingham, C. P. H. Mulder, D. R. Towns, G. W. Yeates, K. I. Bonner, M. S. Durrett, M. N. Grant-Hoffman, W. M. Williamson. 2006. Above- and below-ground impacts of introduced predators in seabird-dominated island ecosystems. Ecology Letters 9: 1299–1307. King, C. M. 2005. The handbook of New Zealand mammals. Melbourne, Australia: Oxford University Press. Russell, J. C., D. R. Towns, S. H. Anderson, and M. N. Clout. 2005. Intercepting the first rat ashore. Nature 437: 1107. Towns, D. R., I. A. E. Atkinson, and C. H. Daugherty. 2006. Have the harmful effects of introduced rats on islands been exaggerated? Biological Invasions 8: 863–891. Wanless, R. M., A. Angel, R. J. Cuthbert, G. M. Hilton, and P. G. Ryan. 2007. Can predation by invasive mice drive seabird extinctions? Biology Letters online: 1–4.

FIGURE 1 Longreach Bay, Rottnest Island, 2008, with view of Bathurst

Point and the smaller of the Rottnest lighthouses, built in 1900. An eroded foredune area with Spinifex and Acacia lines the beach, with dark green tea trees on higher ground. Norfolk Island pines, which have been planted along most beaches in Australia, tower above the native vegetation. Reef platforms and small islets typical of the bays lie to the left of the lighthouse. Buildings of the city of Perth on the mainland are visible in the background behind the small islets. The city lies on the Swan coastal plain, and the 400-m-high hills of Darling Scarp form the eastern boundary of the Perth sedimentary basin. The clear

ROTTNEST ISLAND

waters are host to many different seaweeds and seagrasses. Boats at anchor illustrate the pressures of modern life on the natural environment. Photograph by O. Paterson.

ANNE BREARLEY University of Western Australia, Crawley

Rottnest Island, 1900 ha in area and located 17 km off the coast of Perth, is an iconic holiday destination for many Western Australians and popular tourist venue for 500,000 visitors each year. Limestone buildings from the colonial period in the 1800s still form the heart of the settlement. Motor vehicles are used only for maintaining services, with visitors riding bicycles, walking, or using buses to explore the island. The scenic bays with clear turquoise waters are ideal for swimming, and the colorful diverse marine life is of great fascination to all (Fig. 1).

use the island as a prison for offenders of the new laws. The island was used as a prison for about 70 years but also served as a boys’ reformatory and a vice-regal summer retreat until the decision was made to convert it to a holiday resort. During World War I, the island was again used as a prison for internees and prisoners of war. Throughout the Second World War, the island, with its sweeping ocean views, was the first line of defense for protecting Fremantle Harbour, which, following the attack on Pearl

HISTORICAL PERSPECTIVE

Willem de Vlamingh, a member of the Dutch East India Company, in 1696 provided the island with its name, a derivation of “Rottenest” or “rat’s nest,” referring to the small rat-like marsupial quokka Setonix brachyurus found on the island (Fig. 2). Vlamingh’s charts also showed the mainland coast and the Swan River, named for its black swans, which was to become the focal point of British settlement of the west coast in 1829 with the foundation of Perth and Fremantle. Settlement on Rottnest commenced the following year, but farming on the island was a failure, and with growing conflict between the colonists and the aboriginal people a decision was made in 1838 to

796

ROTTNEST ISLAND

FIGURE 2 The quokka among leaves and seed pods shed by the over-

hanging native Acacia, with clumps of green onion–like Trachyandra. Photograph by H. Lambers.

Harbor and the Japanese advance through the Pacific, was a major base for Allied Forces. With the return of peace, the island resumed its holiday mode, although the Kingstown Barracks area was not relinquished until 1984, when it was handed back to the State, and the former army buildings converted to the Environmental Education Centre to foster public interest in the islands unique natural and historical heritage. The island was declared an A-class Conservation Reserve primarily for public recreation in 1917. The flora and fauna of the area have been well documented, and a research station is managed through University of Western Australia. The island is a major tourist attraction for day visitors and also features some short-term cottage and resort accommodations. CLIMATE

The island experiences a Mediterranean-type climate characterized by cool, wet winters and hot, dry summers. In winter, the westward passage of systems of low pressure bring winds from the northwest to west. In summer, winds are predominantly from the east in the morning and from the southwest in the afternoon. These sea breezes frequently exceed 50 km per hour. The tides are diurnal, occurring only once per day, and are of small amplitude, with a daily range of 0.4–1.1 m. Sea-level changes are, however, also affected by air pressure, winds, oceanic swells, and currents. In contrast to other west-facing coastlines, the marine waters are warm because of the southward-flowing Leeuwin Current, which brings warm, less saline, nutrientpoor water from the tropics in autumn and winter. As a result, the waters around the island are up to 3 °C higher than along the coast. THE ISLAND AND ITS SETTING

Rottnest, 11 km long and 5 km at the widest point, with the highest point 45 m above sea level, is the largest of a chain of small islands and reefs representing one of a series of dune lines on the coastal plain and continental shelf, which were separated from the mainland 5000–7000 years ago. The island is composed of aeolian or dune limestone (the Tamala or Coastal Limestone) of Late Pleistocene or Early Holocene age, with an intercalated Late Pleistocene coral reef (the Rottnest Limestone) exposed at Fairbridge Bluff, overlain dunes, and weakly lithified Holocene shell beds (Herschell Limestone) around the lakes, along with swamp deposits and dune sands. The coastline of rocky headlands and bays with wide sandy beaches backed by sand dunes is fringed by shallow shoreline platforms cut in the Tamala Limestone.

The shoreline platforms are undercut with overhanging visors or sloping ramps. A narrow storm bench at 3 m above the notch and visor are commonly formed below the limestone cliffs. The outer edge of the reef platform often has a raised rim, covered with coralline algae, which commonly covers much of the exposed rock surface. On some reefs, the surface is divided into a mosaic or into polygons formed by lines of brown algae. These do not appear to be associated with features in the reef surface, and their origin has been attributed to fish grazing. In other areas, mobile nodules of coralline algae or rhodoliths are abundant. Terraces up to 70 cm higher than the main platform also occur near the outer edge of some platforms. The reef platforms are covered with turfing and foliose algae and with a diversity of invertebrates including the tropical echinoid Echinometra mathei that occupy holes excavated into the reef surface. The large turban shell Turbo intercostalis, a temperate species, is also common, and broken shells and opercula deposited by Pacific gulls are found on some headlands. Salt lakes, the deepest of which is about 8.5 m, cover about 10% of the island. These are believed to overlie dolines formed during periods of lower sea level. Water levels rise with the winter rainfall and fall again in summer, when salinities may exceed 150,000 mg/L. Some of the lakes dry out completely and were a source of salt production in the late 1800s to mid-1950s. The lake waters are often pink in color because of pigments in the tissues of microscopic algae. Brine shrimp Artemia, which may have been introduced with salt harvesting equipment, provide food for a variety of birds, including transequatorial migratory species that congregate here in summer. Algal mats and stromatolites–thrombolites cover the lake bottoms. Brackish groundwater with a thin freshwater lens underlies some parts of the island, and there are a number of small fresh- and brackish-water swamps. However, a number of these are now hypersaline following excavation in the 1970s for road-building material. TERRESTRIAL VEGETATION AND FAUNA

Descriptions of the island prior to European settlement record extensive areas of Rottnest Island pine Callistris preissii, tea tree Melaleuca lanceolata, and wattle Acacia rostelliferia, forming a low woodland that probably covered about 65% of the island. By 1941 the area of forest was reduced to about 23%, and it currently covers less that 5% of the island, with an additional 6% of indigenous and exotic species. In 2003, 196 plant species were recorded, and a low heath dominated by the shrub prickle lily Acanthocarpus presissii, the tussock grass Austrostipa flavescens,

ROTTNEST ISLAND

797

and the introduced geophyte Trachyandra divaricata now covers most of the island. Changes in the vegetation are considered to be the result of human activity, initially by bushfires and woodcutting, but more recently through grazing by the marsupial quokka. The island is home to limited number of vertebrates. The wallaby-like quokka, numbering about 10,000, was at the time of European settlement also common on the mainland where it is now restricted to a few isolated areas of uncleared swamps. The dugite, a venomous snake, is smaller and darker in color than in populations on the mainland and is considered to be a separate subspecies. Lizards, geckos, and about 50 species of bird are also present on the island. Red-capped robins, golden whistlers, and singing honeyeaters inhabit the woodland and heath, and Australian shelducks, banded stilts, and migrant waders including stints, sandpipers, and turnstones are common around the lakes. Terns and gulls are also common along the shore, and ospreys nest on islets around the island.

FIGURE 3 Typical shallow-water habitat Parker Point, southern coast

of Rottnest Island. The tropical hermatypic, or reef-building, coral

MARINE ENVIRONMENT

The 12,000-km coastline of western Australia spans three marine biogeographical regions: the northern tropical, which is continuous with the Indo–western Pacific; the southern warm temperate; and the western coast, an area of overlap with tropical and warm temperate species and a small number of endemics. Rottnest, at 32° S, lies within the overlap zone and influence of the Leeuwin Current, and the waters are host to a range of temperate and tropical species (Fig. 3). About 350 species of algae, including 170 southern, 52 tropical, 50 temperate, and 59 west coast endemics, are found around the island. The largest are the kelp Ecklonia radiata and Sargassum. Nine species of seagrass are also found around the island. These include the large southern seagrasses that form extensive meadows in the shallow, clear water: the ribbon weeds Posidonia (three species), wireweeds Amphibolis (two species), and Thallassodendron. Smaller species (Halophila spp., Syringodium, and Zostera spp.) grow at the edge or within meadows and are often regarded as colonizers. Thirty species of coral have been recorded. Generally, these grow as isolated colonies; however, there are some extensive areas of Pocillopora damicornis in shallow water on the southern coast. About 360 species of fish have also been recorded. Most are temperate species, but others are typically more common among tropical corals. In response to escalating recreational fishing, four small Sanctuary Zones have been formed, and their effect

798

ROTTNEST ISLAND

Pocillopora damicornis and alga Sargassum sp. growing on limestone reef among areas of sand, and the temperate seagrass Posidonia sinuosa with the western buffalo bream Kyphosis cornelii. Photograph by G. Kendrick 2003.

is currently being investigated. Sea lions and migrating whales are also seen around the island, and the populations appear to be recovering following the nineteenthcentury closure of the sealing and whaling industries. SEE ALSO THE FOLLOWING ARTICLES

Marine Lakes / Prisons and Penal Settlements / Research Stations / Whales and Whaling FURTHER READING

Bradshaw, S. D., ed. 1983. Research on Rottnest Island. Journal of the Royal Society of Western Australia 66: 1–55. Joske, P., C. Jeffery, and L. Hoffman. 1995. Rottnest Island: a documentary history. Centre for Migration and Development Studies, University of Western Australia. Playford, P. E. 1988. Guidebook to the geology of Rottnest Island. Geological Society of Australia, Western Australian Division, Excursion Guidebook 2. Rippey, E., M. C. Hislop, and J. Dodd. 2003. Reassessment of the vascular flora of Rottnest Island. Journal of the Royal Society of Western Australia 86: 7–23. Rottnest Island Authority. 2003. Rottnest Island Management Plan 2003–2008. Saunders, D., and P. de Rebeira. 1985. The birdlife of Rottnest Island. Western Australia: DAS & CpdeR. Wells, F. E., D. I. Walker, H. Kirkman, and R. Lethbridge, eds. 1993. Proceedings of the Fifth International Marine Biological Workshop: the marine flora and fauna of Rottnest Island Western Australia 1991 (2 vols.). Perth: Western Australian Museum.

S SAMOA, BIOLOGY A. C. MEDEIROS U.S. Geological Survey, Makawao, Hawaii

The Samoan archipelago is a volcanic chain of nine main inhabited islands, high islets, and low coral islands located about 13°–15° south of the equator in the central South Pacific (though disjunct Swains Island is 11° S). The archipelago is divided into two political entities: Western Samoa (officially called Independent Samoa or Samoa) to the west and American Samoa (an unincorporated territory of the United States) to the east. Western Samoa consists of Savai‘i (1820 km2 area; 1860 m elevation) and ‘Upolu (1110 km2; 1100 m), the fifth and eighth largest islands of the tropical Pacific, and a number of sizable volcanic islets (Aleipata Islands, Manono, and Apolima). The Aleipata islands have great potential as restoration sites for native birds, some of which are now endangered on the larger islands. American Samoa consists of five high islands—Tutuila (124 km2; 653 m), ‘Aunu‘u (2.6 km2), Ofu (5 km2; 495 m), Olosega (4 km2; 640 m), and Ta‘ (39 km2; 945 m)—and two remote coral islands—Rose Atoll (2.6 km2) and Swains Island (1.5 km2). Mount Silisili, the summit of Savai‘i, is the highest point in Samoa, the tallest peak of central Polynesia, and the sixth tallest peak of the tropical Pacific.

FIGURE 1 Samoan rain forest, northern Tutuila island. Cyathea tree

fern in center. Photograph by the author.

Total rainfall varies throughout the islands because of orographic effects associated with topography and ranges from an average of 318 cm/yr on the leeward coastal plain of Tutuila to more than 500 cm/yr in the mountains. Samoa’s position in the southern trade wind zone would suggest wetter windward and drier leeward exposures, as occurs strongly in the Hawaiian archipelago. However, that is not the case, as rain forest vegetation dominates both windward and leeward exposures, presumably because of the lower elevation of Samoan mountain summits and the archipelago’s orientation, roughly parallel to southeasterly tradewinds. Larger islands are dissected by numerous short stream drainages, most flowing only intermittently.

CLIMATE AND TOPOGRAPHY

TERRESTRIAL BIOLOGY

Located well within the tropics, the Samoan islands are hot, humid, and often rainy throughout the year. There is a wet summer season (October–May) and a shorter, slightly cooler and drier winter season (June–September).

Except for the two coral islands, the Samoan islands are volcanic, created by the westward movement of the Pacific Plate over a near-stationary hotspot beginning some 2 million years ago. The Samoan biota has been

799

derived through colonization across 500 km of ocean and through in situ evolution and speciation. Like other island groups in the South Pacific, much of the ancestral terrestrial biota of Samoa arrived from Malesia in the west. Terrestrial and marine species diversity in the South Pacific is highest near New Guinea and generally declines eastward through Melanesia and Polynesia. Vegetation

The high island vegetation of Samoa has been ascribed to seven plant communities: littoral, wetland, lowland rain forest, montane rain forest, cloud forest and scrub, vegetation on recent volcanic surfaces, and modified vegetation. Rain forests are the predominant vegetation form and still dominate much of Samoa from just above the shoreline to mountain summits. Samoan rain forests are nearly as diverse as Melanesian rain forests and comprise some of the most diverse lowland rain forest in the Pacific. The native Samoan flora is the largest in Polynesia except for Hawaii, with 550 angiosperm species in 300 genera and 228 pteridophyte species. About 30% of the flowering plants of Samoa are endemic, with a single endemic plant genus, Sarcopygme. Fauna

The native fauna of Samoa consists of 24 land and freshwater birds, 20 seabirds, three mammals, seven skinks, four geckos, two marine turtles, one snake, and numerous invertebrates. Of particular interest in the latter group are 94 species of native land snails, 63% unique to the archipelago. U.S. federally listed endangered species include the humpback whale and two marine turtles. Four other vertebrate species are known as species of concern (Polynesian sheath-tailed bat, many-colored fruit dove, friendly ground dove, spotless crake). The nectar-feeding honeyeaters include one of the most common and one of the rarest forest birds of the archipelago. The common and indigenous ‘iao, or wattled honeyeater, (Foulehaio carunculata), though plainly colored, is an energetic and boisterous addition to native forest and village garden. The sweet singing segasegamau‘u, or cardinal honeyeater (Myzomela cardinalis), common around villages and gardens, is dimorphic with scarlet males and gray-olive females. One of the rarest birds of Oceania, is endemic to the Samoan archipelago: the mao, or ma‘oma‘o (Gymnomyza samoensis), one of the largest honeyeaters of the tropical Pacific, endemic to the Samoan islands of Savai‘i, ‘Upolu, and formerly Tutuila (last recorded in 1977). Described as common in certain areas of ‘Upolu in the mid-1980s, this nectarivore has declined markedly and is now a Red List Endangered species of the World Conservation Union; current population is estimated at

800

SAMOA, BIOLOGY

1000–2500 birds. In contrast, its closest living relative, the giant forest honeyeater (Gymnomyza viridis) of Fiji, is still relatively common in some areas. Of the two native starlings of Samoa, the larger endemic fuia, or Samoan starling (Aplonis atrifusca), is the more adaptable species. The wide-ranging miti vao, or Polynesian starling (A. tabuensis), is generally restricted to forested surroundings. Five species of gallinules occur in Samoa. The most easily seen include the common ve‘a, or banded rail (Rallus philippensis), and the manu ali‘i, or purple swamphen (Porphyrio porphyrio). The rare, shy puna‘e, or Samoan woodhen (Gallinula pacifica), is thought to be restricted to primary rain forests of Savai‘i. Old accounts suggest that the burrowing puna‘e was once common and was hunted with dogs and nets; it was last seen in 1908. Samoan pigeons are the subject of special cultural interest. Prior to modern hunting techniques and cyclone impacts, the manutagi, or purple-capped fruit-dove (Ptilinopus porphyraceus); the manuma, or many-colored fruitdove (Ptiliopus perousii); and the tu‘aimeo, or friendly ground dove (Gallicolumbra stairi), were once common but are now infrequent to rare and declining. The still relatively common lupe, or Pacific pigeon (Ducula pacifica), is important to Samoan culture, and in prehistory its capture influenced the construction of the large, stone platforms called tia seulupe (star-mounds), in part for pigeon catching. When first discovered in 1839, the secretive manume‘a, or tooth-billed pigeon (Didunculus strigirostris), caused quite a stir, as it was initially thought to be related to the extinct dodo of Mauritius. In fact, the genus name Didunculus means “little dodo.” Now recognized as the national bird of Samoa, the manume‘a is restricted to forests of Savai‘i and ‘Upolu. This thickbodied, short-tailed pigeon, with its strongly hooked upper bill and notched or “toothed” lower bill, is the only extant member of the subfamily Didunculinae, or toothbilled pigeons. The bill is used in feeding primarily on the tough fibrous fruits of the rain forest tree maota (Dysoxylum spp.). The only other representative of the genus is the extinct Tongan tooth-billed pigeon (Didunculus placopedetes), known only from subfossils from ‘Eua island. Losing forest habitat through agricultural deforestation as well as hunting, populations of tooth-billed pigeon were halved to an estimated 2500 birds by the 1990s and continue to decrease. The tooth-billed pigeon currently is listed as a Red List Endangered species by the World Conservation Union. Samoa has one native snake: the nonpoisonous gata, or Pacific boa (Candoia bibroni), found locally on Savai‘i,

‘Upolu, and Ta‘u islands. The only native mammals of Samoa are three bat species, including a small insectivore and two fruit bat species. The pe‘ape‘avai, or Polynesian sheath-tailed bat (Emballonura semicaudata), was once common throughout Polynesia and Micronesia but now is either extinct or nearly so in Samoa. The extremely wide-ranging pe‘a fanua, or Tongan fruit bat (Pteropus tonganus), is found from New Guinea to the central Pacific (Cook Islands). The diurnal pe‘a vao, or Samoan fruit bat (P. samoensis), still occurs in Samoa and Fiji but is extinct on Tonga. Native Samoan frugivores, especially fruit bats, pigeons and doves, and starlings, play a critical role in the islands’ ecology by dispersing seeds of many native plant species. SAMOAN MARINE BIOLOGY

Samoa has 890 species of coral reef fishes and over 200 coral species. The rate of endemism in this rich native marine biota is low, apparently because of its proximity to other South Pacific islands and the high mobility of many marine species during at least some life stages. Samoa has two marine turtle species, both of which are protected by the U.S. Endangered Species Act. Both sea turtles are declining throughout the Pacific, primarily as a result of harvesting for food and shell, incidental mortality in commercial fisheries, and disturbance of nesting beaches. Though widely distributed, the laumei uga, or hawksbill (Eretmochelys imbricata), nests in low numbers throughout Samoa and does not often migrate long distances. In contrast, the green turtle (Chelonia mydas) is highly migratory and wide-ranging (> 1000 miles); it nests primarily on Rose Atoll. Excessive harvesting of this species, the so-called “buffalo of the sea,” for its shell, cartilage, and meat has made it among the most exploited of turtle species worldwide. THREATS TO SAMOAN BIODIVERSITY

Human impacts (overfishing, pollution and accelerated erosion) and crown-of-thorns starfish (Acanthaster plana) threaten the sustainability of coral reef communities in Samoa. Although Polynesians have lived on the islands for over 3000 years, accelerated population growth and economic development in the past 50 years is leading to extensive modification of coral reef and nearshore ecosystems. The current human population of American Samoa (65,000) is increasing at a rate of 2% per annum, the vast majority living within several hundred meters of the shoreline. Climate change and consequent sea level rise obviously have potential to bring about serious damage to Samoan biodiversity. Deforestation (for timber in Western Samoa and expanding agriculture generally) and replacement by

non-native invasive plant species seem to pose the greatest threat to long-term survival of native rain forest biota of Samoa. Prior to human settlement, Samoa was almost entirely forested. Agriculture, timber harvesting, and invasion by non-native plant and animal species have reduced forest cover. In 1954, an estimated 74% of Savai‘i and ‘Upolu (93% of Samoa’s total land mass) was covered in native rain forest. That figure had declined to 40% by 1990, and forest reduction continues. Invasive species are notorious as degraders of Pacific island biodiversity. Among the most damaging of vertebrate invaders in Samoa are feral pigs and rats, including the Norway rat on Tutuila. Introduced Euglandina rosea snails prey on complexes of endemic snail species. Increasing numbers of non-native plant species are becoming established in Samoa. Currently, two of the most damaging invasive trees of Samoan rain forests are silkrubber (Funtumia elastica) and Moluccan albizia (Falcataria moluccana). Funtumia, brought from Africa and planted widely on ‘Upolu and Savai‘i, has now spread beyond plantations and is invading intact rain forest throughout the country up to 700 m elevation. Funtumia may be the worst forest invader of Western Samoa (it currently does not occur in American Samoa) and may be one of the most potentially damaging invaders of rain forests of Pacific islands. Falcataria moluccana is a large, quick-growing tree species native to rain forests of the Moluccas, New Guinea, New Britain, and the Solomon Islands. Introduced as a forestry resource, Falcataria now occurs in about 35% of Tutuila’s forests. Nitrogen-fixing abilities by Falcataria may disrupt succession in native Samoan forests by promoting establishment of non-native plant species and by deterring recruitment of natives. Other notable threats to Samoan rainforest include two Neotropical species: the shrub Clidemia hirta, widespread in Samoa, and the small tree strawberry guava (Psidium cattleianum), reported only from small naturalized populations on Tutuila and ‘Upolu. Land tracts set aside for biological and cultural diversity include several national parks in Western Samoa (Tafua and Falea¯lupo on Savai‘i and ‘O le Pupu¯-Pu‘e on ‘Upolu) and the 3600-ha National Park of American Samoa (NPSA), which includes sections of the four main islands of American Samoa. The persisting culture and tradition (fa‘a Samoa) of native Samoans can be a powerful force supportive of the management of natural ecosystems. In 2005 the NPSA joined with local villages and NGOs in eliminating Falcataria trees from more than 250 hectares on Tutuila, with over 3000 large trees killed (by girdling) as of 2007. Cooperative community-based efforts like this may serve as a model for future land management in Samoa.

SAMOA, BIOLOGY

801

SAMOA, GEOLOGY

400 km along a trend of about 290°, but the entire chain, in the form of shallow banks and submarine volcanic pinnacles, extends to the west for about another 1200 km. The islands resemble the Hawaiian Islands in many respects, but some features of them are distinct. The principal difference results from being near the Tonga Trench rather than in the middle of the basin. In plate tectonic terms, the Samoan Islands are at the edge of the Pacific plate in a most unusual spot, just to the north of where the trench turns to the west. South of there, the Pacific plate is disappearing into the trench, and the path of its subduction can be traced seismically by a dipping belt of earthquakes that extends to depths of about 700 km beneath the Tonga arc. However, because the trench changes trend just south of Samoa, the islands manage to ride the plate west past the trench and are not subducting. But it is a near thing, and the effect on Samoan volcanism is profound. The chain may even exist because of this peculiar relationship.

JAMES H. NATLAND

AGE PROGRESSION

University of Miami, Florida

From east to west, the main islands are Ta‘u and OfuOlosega (two islands separated by a narrow channel, denoted here by the hyphen) in the Manu‘a group of American Samoa; a larger island, Tutuila, also in American Samoa; and the two largest islands, Upolu and Savai‘i, in the independent country of Samoa, which was called

SEE ALSO THE FOLLOWING ARTICLES

Deforestation / Fish Stocks/Overfishing / Reef Ecology and Conservation / Samoa, Geology / Sustainability FURTHER READING

Craig, P. 2002. Natural history guide to American Samoa. A collection of articles. Pago Pago: National Park of American Samoa and American Samoa Department of Marine and Wildlife Resources. Mueller-Dombois, D., and F. R. Fosberg. 1998. Vegetation of the tropical Pacific islands. New York: Springer-Verlag. Watling, D. 2001. A guide to the birds of Fiji and Western Polynesia. Fiji: Environmental Consultants Ltd. Whistler, W. A. 2002. The Samoan rainforest: A guide to the vegetation of the Samoan Archipelago. Honolulu: Isle Botanica. Whistler, W. A. 2004. Rainforest trees of Samoa. Honolulu: Isle Botanica.

The Samoan Islands are at the eastern end of a chain of volcanoes, most of them submerged, and very near the southwestern edge of the main Pacific Basin at the Tonga Trench (Fig. 1). The islands proper span a distance of about

FIGURE 1 Bathymetry and location of the principal Samoan Islands (gray—Savai`i, Upolu, Tutuila and Ta`u are labeled), banks, and seamounts

(Muli, Malumalu, and Vailulu‘u, eastern Samoa; Papatua and Ua Mamae—also called Macias—north of the northwesterly-trending portion of the Tonga Trench. The Vitiaz lineament is shown as a continuation of the northern termination of the Tonga Trench, and is here interpreted as a trenchtrench transform fault.

802

SAMOA, GEOLOGY

Combe-Field Lineament

Shear couple

ookh lla R

SIMPLIFIED STRUCTURAL INTERPRETATION L.

La

Tonga

Pla Pos te M t-er otio osio n nal line ame nt Sideways bending Upolu into trench De

Trenc h Deeply submerged seamount

Manu’a

for me utuila d AT

p ron

Shield progagation direction since 3.2 Ma

A r c h

100 km

o f

Pacific plate bends directly into trench

b e b d

FIGURE 2 Generalized structural relationships of the principal Samoan

Islands, including lineaments linking central volcanoes in red and the great Samoan post-erosional volcanic rift system in blue. Arrows are drawn to indicate plate motion, subduction direction, propagation directions, and inferred plate bending and shear orientation near the transform system.

of this geological distinction shows that there is indeed a Hawaiian-like age progression to the underpinnings of the Samoan chain (Fig. 3). The older phase of volcanism at Savai‘i began at about 5 million years ago, at Upolu about 3.5 million years ago, and at Tutuila at about 1.5 million years ago. Ta‘u has had some barely prehistoric eruptions, and Vailulu‘u is active at the present day. The age progression along the islands is thus perfectly normal for the Pacific plate and is consistent with the hotspot model; indeed it has been extended to the submerged banks to the west as far away as Combe Bank, which is 14.1 million years old. 30 Alexa

25 20

Age (Ma)

Western Samoa until 1997. A number of smaller islands scatter about these, and several substantial seamounts rise from the seafloor near these main islands. One of them, Vailulu‘u, at the eastern end of the chain, is an active volcano. To the west are submarine banks, the nearest being Pasco and the farthest Tuscarora, which is a shallow submerged reef platform. This article will sometimes refer to the collection of islands and seamounts as Samoa, with no geopolitical connotation. American explorer James Dwight Dana first described the geology of the Samoan Islands in 1849. Dana noted many similarities to Hawaii, and in particular that the largest island, Savai‘i, has a youthful, uneroded, domed shape and is studded with many small and obviously recently erupted volcanic cones, much like Mauna Kea volcano on the island of Hawaii. The islands to the east appeared to him to be successively more dissected by erosion, and they have wide rather than narrow offshore reefs, indicating they are older. In these respects, they are similar to Oahu and Kauai in the Hawaiian chain. But whereas those older Hawaiian Islands lie to the west, in Samoa the islands most similar to these are to the east. Dana wrote, “It is hence evident that the fires were soonest extinct to the east, and burnt longest and to the latest period on the western island, Savai‘i” (Dana 1849: 335). This comparison to Hawaiian volcanism was pivotal, in my view, to Dana’s later development of the doctrine of fixity of continents and ocean basins, which, more than 60 years hence, was one among the array of ideas used in criticism of Alfred Wegener’s theory of continental drift. Dana construed island chain volcanism in the Pacific to occur along fractures and largely simultaneously along the entire lengths of the ridges at first, but then dying out toward one end or the other—to the west at Hawaii, to the east at Samoa. The Pacific crust beneath, consequently, had to be fixed in place. Nevertheless, even after plate tectonics was demonstrated and linear island chains were considered to form in consistent directions as plates passed over fixed hotspots, Samoa remained an anomaly. However, subsequent, more precise comparisons with the stages of volcanism at Hawaii cast the matter in a different light. The work of several investigators showed that Savai‘i and its nearest neighbor, Upolu, are both experiencing a second major phase of volcanism that has buried older volcanoes that should be considered separately for the purpose of establishing the age progression along the chain. Comparison to Hawaii dictated at the time that these rocks be termed “post-erosional” with respect to the eroded volcanoes they surmounted (Fig. 2). Thus, potassium–argon dating of rocks sampled in light

7 cm/year Combe Plate Speed

15 10

Lalla Rookh

Islands

(no analysis)

5 0

0

Linear Regression

500

1000

1500

2000

Distance (km) FIGURE 3 K/Ar and

39

Ar/40Ar plateau ages of Samoan shield basalts,

plus an unanalyzed sample from Lalla Rookh bank. The linear regression is nearly identical to the estimated 7 cm/year rate of motion over a hotspot at Vailulu‘u fixed with respect to an assumed fixed Hawaiian hotspot. Many more alkaline basalts and other dated samples fall below the regression (not shown), including all dated post-erosional samples.

SAMOA, GEOLOGY

803

STAGES OF VOLCANISM: A MISFIT TO HAWAII

The key to understanding the age progression turned out to be distinguishing the stages of volcanism at Samoa that are comparable to those of Hawaii: namely, preshield, shield, postshield, and rejuvenated. However, aspects of this sequence, and especially the terms, are somewhat misleading when considering the Samoan case. One can, however, see provisionally how they apply to Samoa, consider the contrasts between Samoa and Hawaii, and then try to understand those differences in evaluating the causes of Samoan volcanism. The preshield stage at Hawaii is that of nascent volcanism in which alkalic basaltic lava erupts to form a seamount at the youthful end of the chain; this is presently the active submarine volcano of Loihi, offshore of Kilauea volcano. At Loihi, volcanism has shifted through time to tholeiitic compositions, some of which erupted in the past few years. The Samoan equivalent to Loihi is Vailulu‘u, the summit lavas of which are indeed alkalic basalt; as discussed below, however, no Hawaiian-type tholeiite has been found there. The volcano has a summit crater about 2 km in diameter. Among the Hawaiian Islands, the next stage of volcanism is construction of large shield volcanoes made of tholeiitic basalt. The structures are huge, centered on principal conduit systems beneath the tallest parts of the volcanoes, but they also include enormous rift zones that are supplied laterally by magma pumping from staging areas just beneath the summit regions. Deep exposures of old rifts reveal densely spaced dikes, all with nearly the same orientation. Some of the rifts reach offshore as far as 200 km from their central conduits. The shield volcanoes overlap and structurally influence each other to the extent that rift systems nest younger against older, and dike swarms conforming to the resulting stress field are parallel. Then comes the postshield stage of waning volcanism, formation of large calderas, and eruption of alkalic basalt and affiliated differentiates, typically hawaiite, mugearite, and trachyte. Mapping reveals that most of this volcanism is centered on or at least near the older tholeiitic summits, but it is infrequent enough for some of the lavas to collect in valleys carved by rivers. At Samoa, no shield volcano is nearly as large as the typical Hawaiian shield, and they do not coalesce or combine in the same way, or even at all. No rift zone nesting occurs. Again, Hawaiian-type tholeiite has not been found on any of them. The islands of Ta‘u and Ofu– Olosega conform pretty much to the postshield Hawaiian equivalent; Ta‘u has a large caldera and has erupted alkalic basalt and some hawaiite (Fig. 4A). These are

804

SAMOA, GEOLOGY

relatively youthful structures with much of their original uneroded surfaces intact, and magmatic lineages have not produced the light-colored felsic differentiates mugearite and trachyte. The older islands of Tutuila and Upolu have substantial and now fairly deeply eroded shield volcanoes, with well-developed magmatic lineages that include alkalic olivine basalt, hawaiite, mugearite, and trachyte (Figs. 4B and C). On Tutuila the trachyte occurs as prominent plugs that intruded an elliptical caldera ring fault that has principal axial lengths of 5 km × 10 km; this now nearly surrounds Pago Pago Bay. A few similar plugs are just outside the crater. Another large, caldera-like ring fault also bounds most of Fagaloa Bay on Upolu. Trachyte cobbles have been sampled from beaches near this fault, but actual trachyte plugs there have not been mapped. Varieties of tholeiitic basalt have been obtained from deep exposures or outlying districts at both Tutuila and Upolu, but the classification of these rocks as tholeiites, and their place in the morphological development of those volcanoes, is not simple. Again, this topic is considered separately below. Structurally, the volcanoes from which these lavas erupted are not strictly comparable to the shield volcanoes of Hawaii. Thus, four separate volcanoes along the 40 km of the length of the island were mapped at Tutuila, and a fifth older one lies partly buried and without physiographic expression. However, some of the details of that early mapping amidst the jungle are now problematic, and where the mapping is still sound, the question now is whether at least two of these volcanoes may simply have been satellitic summits about a principal vent centered on what is now the caldera at Pago Pago Bay. Thus the separate summits are very closely spaced, and the total line of them would easily fit along less than half the length of the Kilauea eastern rift from the summit to the shoreline. Further, potassium–argon ages indicate that, if there were indeed separate volcanoes, four of them erupted basaltic lava simultaneously from about 1.5 to 1.2 million years ago; then mainly differentiates to about 1.1 million years ago. Late-stage trachytes are nearly completely restricted to the area around the ring fault at Pago Pago Bay and are 1.03 million years old. The most reasonable interpretation now is that there was one principal conduit system that split into separate upper-level conduits that were alternately supplied differentiated lava as the rate of magma supply diminished, and that as volcanism waned further, it became more constricted to above the vicinity of the main conduit. The final Hawaiian stage of volcanism is eruption of basaltic lava with low to very low SiO2 contents, namely alkalic olivine basalt, basanite, olivine nephelinite, and

18 Vailulu`u, Tau and Eastern Seamounts

A

Na2O+K2O (%)

15 Phonolite

12

P-N

Trachyte

P-T

ph el

in it e

9

nde

hya

c Tra

Hawaiite

Rhyolite

site

Mugearite

B+T

Dacite

Ne

6

Benmorite

B-A Andesite

3

Basalt

0 35

45

55

65

75

SiO2 (%) 18

B

Tutuila

Na2O+K2O (%)

15 Phonolite

12

P-N

Trachyte

P-T

ph el

in ite

9

Mugearite

B+T

Dacite

Andesite

B-A

3

Rhyolite

site

nde

hya

c Tra

Hawaiite

Ne

6

Benmorite

Basalt

0 35

45

55

65

75

SiO2 (%) 18

Upolu and Uo Mamae Seamount

C Phonolite

12

P-N

Trachyte

P-T

6

Benmorite

Mugearite

B+T

ph

el

in

ite

9

Rhyolite

site

nde

hya

c Tra

Hawaiite

Ne

Na2O+K2O (%)

15

Dacite

B-A Andesite

3

Basalt

0 35

45

55

65

75

SiO2 (%) 18

D Savai`i

Na2O+K2O (%)

15 Phonolite

12

P-N

Trachyte

P-T he lin

ite

9

Mugearite

B+T

Dacite

FIGURE 4 Diagrams of total alkalis (Na2O + K2O) versus SiO2 for dif-

ferent portions of the Samoan chain from literature and unpublished

Andesite

B-A

3

Rhyolite

site

nde

hya

c Tra

Hawaiite

Ne p

6

Benmorite

olivine melilitite (Figs. 4B–E). On the islands of Oahu and Kauai, these erupted chiefly as volcanic scoria at cones such as Diamond Head, but with some lava, along short fracture systems that trend across the older shield rift systems. The lavas carry numerous ultramafic xenoliths. Following eruption of trachytes on both Tutuila and Upolu, volcanism shut down at both places for nearly a million years. Then, in the last few thousand years, renewed alkalic basaltic volcanism has taken place at both islands in the form of cinder cones and lava flows. The lavas are alkalic olivine basalt, basanite, and olivine nephelinite, but not olivine melilitite. The lavas carry ultramafic xenoliths. On Tutuila, the volume of this volcanism has been small enough that it was at first strongly likened to Hawaiian rejuvenescent volcanism. But on Upolu, the extent of this volcanism was much greater, and the eruptions there are also clearly related structurally to even more substantial Holocene to historic volcanism on the western island of Savai‘i, where some of the lava flows, including two historic ones, are extensive. Almost all of this renewed volcanism has occurred along a single prominent volcanic rift system that spans the length of Savai‘i and Upolu and that only recently has reached Tutuila (see Fig. 2). At Samoa, the scale of this volcanism has been much greater that at Hawaii, especially at Savai‘i, where accumulations of the younger lavas are thousands of meters thick, and there are no exposures of older shield volcanoes on the island; those are represented only by offshore rocks obtained by dredging. Thus the archetypal Hawaiian sequence of stages does not truly fit Samoa. The concept was useful when it came to establishing the age progression. However, in detail the misfit is substantial. The nearest equivalents to shield volcanoes are much smaller than those of Hawaii and do

Basalt

data sources. The green lines divide Hawaiian shield tholeiitic (below) from alkalic basalt (above). All chemical analyses from 1902 to 2008

0 35

45

65

55

75

Ta‘u, and eastern Samoan seamounts are all given different symbols;

18

E

(B) Tutuila—shield and postshield samples are divided into red sym-

Western Banks

bols—Masafau volcano; and open squares—the coeval volcanism of Taputapu, Pago, Pago Intracaldera, Olomoana, and Asofau volcanoes

Phonolite

12

P-N P-T B+T

Rhyolite

site

lavas from Uo Mamae (Machias) Seamount. (D) Savai‘i – various col-

de yan

h

c Tra

ors besides blue for offshore seamounts and trachyte cobbles from

Dacite

the Vanu River; (E) samples from western Samoan banks analyzed by B-A

3 0 35

(C) red diamonds = the Fagaloa volcano of ‘Upolu; green triangles =

Benmorite

Mugearite

Hawaiite

ph

el

in ite

9

of Stearns (1944), here combined into the Greater Pago volcano.

Trachyte

Ne

Na2O+K2O (%)

15

6

are combined in these diagrams. On all diagrams, analyses of posterosional samples are given by half-filled blue squares. (A) Vailulu‘u,

SiO2 (%)

Andesite

Johnson et al (1986); red circles are Alexa Bank, inverted triangles are

Basalt

Combe, Lalla Rookh, and Field Banks, plus Wallis Island. Distinctive magmatic lineages leading to soda rhyolite (Greater Pago of Tutuila),

45

55

SiO2 (%)

65

75

trachyte (Fagaloa of Upolu), and phonolite (Uo Mamae Seamount) occupy islands of the central part of the Samoan chain.

SAMOA, GEOLOGY

805

not overlap or structurally interfere with each other. The postshield summit structure of Tutuila is different, having broken into four smaller centers. Hawaiian-type tholeiites are not found anywhere, and much of the shield-building stage of volcanism is alkalic, not tholeiitic, in character; and the final stage of volcanism is so prominent on two of the islands that the term “rejuvenescent” clearly is inadequate to describe it. The formerly used term “post-erosional” for this stage of volcanism at Samoa is not very good either. Later, we shall consider the substantial geochemical contrasts between Hawaii and Samoa. But given all of this, how should Samoan volcanism be described?

The general tendency for the ridges to curve toward the Tonga Trench is notable because this sort of curvature does not occur anywhere else in the Pacific, and certainly not along the Hawaiian chain. Along the eastern portion of the chain, a major younger volcanic lineament is superimposed. It is probably longer than just the combined lengths of Savai‘i and Upolu and reaches all the way to Tutuila. Hypothetically, if it additionally spans the lengths of the two ridges that align with it west of Savai‘i, this feature could be more than 500 km long. This is the great Samoan post-erosional volcanic rift zone. THE PROBLEM OF SAMOAN THOLEIITE

SAMOA CONSIDERED PHYSIOGRAPHICALLY

Consider that the second major phase of basaltic volcanism, which spans the length of Upolu and Savai‘i and reaches Tutuila, is a single straight volcanic lineament superimposed on older features (see Fig. 2). What does the rest of the Samoan chain look like? We must look at the offshore geology. It does little good just to consider island exposures, which only amount to a few percent of the entire volume of the volcanic superstructure of the chain. Bathymetry derived from satellite altimetry (see Fig. 1) combined with subaerial topography in the islands reveals several ridges along the Samoan chain that are 100–250 km long (see scale bar in Fig. 2). These are curving volcanic ridges that all probably combine several tall volcanoes. This is most obvious at the eastern end of the chain. Thus the easternmost of these combines, from west to east, Mali seamount, Ofu–Olosega, Ta‘u, and Vailulu‘u. A second links the several closely spaced summits of Tutuila. At its western end, this ridge curves toward the southwest and the westerly trending portion of the Tonga Trench; at its eastern end, the ridge curves back toward the southeast, reaching Malumalu seamount. West of Savai‘i, some of these ridges can clearly be linked end to end as longer curving ridges that are broken by short gaps. Some of the curving ridges approach each other, and their general concavity is also toward the Trench (Fig. 2). The longest curving ridge extends from Combe Bank to Field Bank. Ages for these banks are shown in Fig. 3. Two shorter curving ridges, one including Lalla Rookh Bank, are both closer to and even more strongly inclined toward the westerly trending portion of the Tonga trench. The general impression given by these trends and the radiometric ages is that the Samoan chain consists of a number of curving ridges, most of them diverging from the westerly trending part of the Tonga Trench, and with an overall progression that is younger toward the east. 806

SAMOA, GEOLOGY

Although a number of classification schemes have been used to distinguish tholeiitic from alkalic basalt, the simplest one that has been applied to Hawaiian rocks is represented as a single straight line on a plot of total alkalis (Na2O + K2O) versus silica (SiO2), separating basalts of the principal shield stages of several of the volcanoes, which all had the characteristics of tholeiitic basalt based on other classification schemes, and the alkalic basalts and differentiates of the postshield alkalic cappings. When it came to Tutuila, however, no tholeiites of the Hawaiian type were found even among what were obviously basalts of the shield stage as described for Tutuila. Presently, with many more analyses in hand, this conclusion still holds, even though a few analyzed specimens, especially from Tutuila, fall below the Hawaiian dividing line (Fig. 4). Some of these are fairly obviously lavas leached of their K2O by flow of groundwater. Several others are picritic, being highly charged with olivine phenocrysts, which serves to dilute total alkalis in a bulk-rock analysis, drawing compositions below the line. One major difference lies in the proportion of K2O in a bulk rock analysis, which is rarely below 0.8% and often more than 1% among least altered Samoan basalt of shield volcanoes not charged with olivine crystals; it is typically less than 0.5%, and in some cases not even half of that, in the most nearly similar Hawaiian tholeiite. Samoan basalts have concomitantly higher concentrations of elements with geochemical behavior similar to potassium, and most have a different isotopic signature, described by geochemists as more enriched. Put in the most general terms, although some scattered lavas might fall into the tholeiitic field on a plot of alkalis versus silica, a principal tholeiitic stage of volcanism has not been found anywhere in Samoa, whether considered looking downward into the deepest accessible stratigraphy left by erosion on the islands or upward from the structures of seamounts. Indeed, basalt at Vailulu‘u seamount is passably similar

to basalt of any part of Ta‘u, and basalt there in turn resembles much of the basalt at Tutuila. Furthermore, some portions of the shield systems of Tutuila and Upolu, and at least two of the offshore volcanoes, are distinctly more alkalic (that is, further removed from tholeiite, with higher K2O and total alkalis; e.g., Ua Mamae, Fig. 4C) than Ta‘u or Vailulu‘u in character. Even more difficult to understand is that this geochemical quality called “enrichment,” which is usually considered in terms of isotopic ratios such as 87Sr/86Sr and 206 Pb/204Pb, is inconsistent from one Samoan volcano to the next. Thus the Fagaloa shield volcano of Upolu becomes more enriched through an upward (younging) succession of basalts, whereas at Hawaii the opposite consistently occurs on almost every volcano. At Tutuila, the extent of enrichment is greatest in the oldest exposed lavas; it then starts to scatter as one proceeds upward in the succession into younger lavas. Then abruptly the geochemical bottom drops out, and everything erupted since about 1.28 million years ago is least enriched on this island and similar in 87Sr/86Sr to lavas of Ta‘u and Vailulu‘u. Yet each shield volcano also has different 206Pb/204Pb regardless of 87Sr/86Sr, and all of the second-phase alkalic basaltic lavas in turn are different isotopically from every older age-progressive shield volcano that they happen to overlie (Fig. 5).

19.6

East

West 19.4

Muli Vai

Sav Smt

206

Pb/

204

Pb

19.2

Ta`u TM Malu

19.0

TP Pas

UF

18.8 Sav PEL

18.6

Upo

TL

Average

18.4

206Pb/204Pb

Ua Mamae

18.2 500

400

300

200

100

0

Distance from Vailulu`u (km) FIGURE 5 Average

206

Pb/204Pb versus distance from Vailulu‘u for the

Samoan Islands compiled from all literature sources. Coding for central volcanoes is Vai = Vailulu‘u; Ta‘u seamount and island; Muli = Muli volcano; Malu = Malumalu volcano, TM = Tutuila Masefau volcano; TP = Tutuila Greater Pago volcano; UF = Upolo Fagaloa volcano; SavSmt = offshore Savai‘i; and Ua Mamae volcano. Coding for post-erosional lavas is TL = Tutuila Leone lavas; Upo = Upolu, Samoa; Sav PEL = Savai‘i post-erosional lavas; and Pas = ANTP 239 from Pasco Bank. From Upolu eastward, each volcano has a distinctive average

206

Pb/204Pb

increasing toward the east, and post-erosional lavas (blue band) are isotopically different from them.

THE MECHANISM OF SAMOAN VOLCANISM

Samoan volcanism has been attributed to volcanism along a series of sea floor fractures or to a mantle plume, with the plume perhaps being either generated or distorted by stresses associated with the nearby subduction of the Pacific plate, or to thermal/convective disturbances triggered by subduction in the mantle above about 700 km, which is the depth of deepest seismicity in the nearby Tonga Trench. The strongest case for a fracture is the second major phase of basaltic volcanism, that most usually termed post-erosional volcanism, which occurs along a fissure system of closely spaced volcanic cones spanning the axes of Savai‘i and Upolu and reaching Tutuila. The fissure system is too long and narrow to be a plume, its lavas are chemically distinct from everything else, and there is clear evidence for distortion of the Pacific plate, probably by bending of the plate southward between the islands of Savai‘i and Upolu toward the westward-trending portion of the Tonga Trench. In this scenario, the axis of southward plate bending is directly beneath the two islands, and the Pacific plate is moving, by being either pushed or pulled, directly along the length of the bend. The base of the bending material lies within the region of partial melting in the mantle, and here it is also in compression. A combination of ponding of relatively small partial melts of enriched source material and squeezing aggregates those partial melts and focuses their eruption along the narrow fissure system. The melts probably pond at the rheological base of the lithosphere until a combination of buoyancy forces and regional stresses acts to release them to the surface, sometimes in large volume. The older, age-progressive volcanoes are not so readily interpreted in terms of a fracture mechanism; thus, most recent workers have favored an origin for them in a mantle plume. Both the age progression and the general form of the volcanoes, which resemble those of Hawaii in so many ways (Fig. 3), are consistent with a plume model. But the plume would still have to be viewed as very circumstantially located just at this unusual tectonic juncture in the Pacific plate; the plume itself has produced no geochemical consistency from one volcano to the next or even within the history of a single island (Fig. 5); and the pattern of arcuate ridges mainly directed toward the westerly trending portion of the Tonga Trench (Fig. 2) is anomalous in the context of other volcanic chains in the Pacific. The idea that shield volcanoes are caused by thermalconvective disturbances triggered by subduction into the nearby trench, allows the chain to have mainly Hawaiian-like SAMOA, GEOLOGY

807

structural attributes and an appropriate age progression. Yet it lets portions of the chain be distorted by shear stresses so that the curving volcanic ridges tend to curve toward the trench. And, finally, it is indifferent to geochemical complexity in the mantle except to the extent that source heterogeneity is restricted to the upper 700 km of the mantle. The enriched sources sampled by the volcanoes would lie in the relatively shallow mantle. Why they are so enriched in the first place has to do with the likely ancient prehistory of mantle sources in this region of the Pacific rather than with the specific manner of their arrival in the shallow mantle beneath Samoa, where partial melting has occurred. SEE ALSO THE FOLLOWING ARTICLES

Hawaiian Islands, Geology / Lava and Ash / Pacific Region / Samoa, Biology / Volcanic Islands FURTHER READING

Daly, R. A. 1924. The geology of American Samoa. Carnegie Institution of Washington Publication 340: 95–145. Dana, J. D. 1849. U.S. exploring expedition during the years 1838–1842 under the command of Charles Wilkes, U.S.N. Geology 10: 307–336. Hart, S.R., H. Staudigel, J. Blusztajn, E. T. Baker, R. Workman, M. Jackson, E. Hauri, M. Kurz, K. Sims, D. Fornari, A. Saal, and S. Lyons. 2000. Vailulu‘u undersea volcano: the new Samoa. Geochemistry, Geophysics, Geosystems 1: 1–13, 2000GC000108. Kear, D., and B. L. Wood. 1959. The geology and hydrology of Western Samoa. New Zealand Geological Survey Bulletin 63. Macdonald, G. A. 1944. Petrography of the Samoan Islands. Geological Society of America Bulletin 56: 861–872. Natland, J. H. 1980. The progression of volcanism in the Samoan linear volcanic chain. American Journal of Science 280A: 709–735. Natland, J. H. 2003. The Samoan chain: a shallow lithospheric fracture system. www.mantleplumes.org. Stearns, H. T. 1944. Geology of the Samoan Islands. Geological Society of America Bulletin 55: 1279–1332. Workman, R. K., S. R. Hart, M. Jackson, M. Regulous, K. A. Farley, J. Blustahn, M. Kurz, and H. Staudigel. 2004. Recycled metasomatized lithosphere as the origin of the Enriched Mantle II (EM2) end-member: Evidence from the Samoan volcanic chain. Geochemistry, Geophysics, Geosystems 5.4. doi 10.1029/2003GC00623.

Bioko was connected to the continent during the last interglacial period, São Tomé, Príncipe, and Annobon are all surrounded by deep-sea trenches and are thus true “oceanic islands.” These islands are striking centers of endemism. In contrast to most oceanic islands, where habitat loss is probably the chief conservation concern, in these islands control of introduced species may play a more critical role. GEOGRAPHICAL AND GEOLOGICAL BACKGROUND

These islands are formed by shield volcanoes along the 1600-km-long Cameroon line, a linear rift zone that extends from Cameroon into the Atlantic. The volcanic chain was formed during the middle to late Tertiary and includes Mount Cameroon and the islands of Bioko, São Tomé, Príncipe, and Annobon (Fig. 1). Bioko, formerly Fernando Po, is the largest and closest to Africa, lying about 32 km from Cameroon. Príncipe, 225 km off the northwest coast of Gabon, has an area of 136 km2 and reaches 948 m at Pico de Príncipe. Several smaller islets surround Príncipe, including Ilhéu Bom Bom, Ilhéu Caroço, Tinhosa Grande, and Tinhosa Pequena. The oldest geological date of Príncipe is 31 million years ago. When sea levels were lower in the last glacial periods, Príncipe would have been consid-

SÃO TOMÉ, PRÍNCIPE, AND ANNOBON D. JAMES HARRIS University of Porto, Vila do Conde, Portugal

The islands of the Gulf of Guinea are part of one of the world’s biodiversity hotspots. Part of a volcanic chain that includes Mount Cameroon on the continent, the islands are Bioko, São Tomé, Príncipe, and Annobon. Although 808

SÃO TOMÉ, PRÍNCIPE, AND ANNOBON

FIGURE 1 Map of the Gulf of Guinea islands.

erably larger. São Tomé, 140 km southwest of Príncipe, is approximately 845 km2, about 48 km north–south and 32 km east–west. The southern tip of the island is just 2 km north of the equator, which passes through the small islet of Rolas. The highest point is 2024 m at Pico de São Tomé. The oldest dated volcanic rocks of São Tomé are 15.7 million years old, although most are basaltic lava less than 1 million years old, especially in the north, where pyroclastic cones are common. The central and southern parts of the island were exposed by a volcanic period around 3 to 8 million years ago, whereas the most southerly tip of the island and the Rolas islet are composed of pyroclastic and lava cones less than 400,000 years old. The most isolated island of the group is Annobon, about 160 km southwest of São Tomé and 350 km from the west coast of Africa. It has an area of approximately 17 km2, rising to 598 m at its peak. It is composed of a single extinct volcano with a central crater lake. Its oldest dated rocks are approximately 4.8 million years old. All the islands are within the wet tropical belt. Annual rainfall on São Tomé ranges from 1000 mm in the northeast to over 4000 mm in the southwest. Average annual temperatures range from 18 to 21 ºC minimums to 30 to 33 ºC maximums. The rainy season runs from October to May. Príncipe is very similar to São Tomé, whereas Annobon is notably drier. All are generally very steep, with lowlands at the base of the volcanoes constituting the only (relatively) flat land. The volcanic soils of basalts and phonolites are relatively fertile. There are currently estimated to be over 40 km2 of primary forest on Príncipe and 240 km2 on São Tomé. ORIGINS OF ISLAND FAUNA

Sea levels between Bioko and the continent are shallow; thus, this island would have been connected to the mainland by sea-level fluctuations during the last glacial periods. This is not the case for the islands of São Tomé, Príncipe, and Annobon, each of which is surrounded by deep sea trenches and is thus a true “oceanic island,” having never been connected to the mainland or to its neighbors. Thus, although the fauna of Bioko is essentially continental in nature, the other islands have fewer species but far more endemism. In many other Atlantic archipelagoes, such as the Canary Islands or the Cape Verde Islands, there is a strong trend for island radiations, where a colonizer reaches the islands and then radiates on to the other islands of the group. However, the islands of the Gulf of Guinea are fewer and further apart, and phylogenetic studies indicate that in several cases islands were colonized independently from the continent rather

than taking a more classic “stepping-stone” model, in which the fauna first colonize the island closest to the continent (e.g., Príncipe) and then travel successively on to the more distant islands. Many non-volant species may have reached the islands by rafting: Currents are favorable in carrying material from the continent toward the islands, and major rivers such as the Niger and Congo discharge into the area. During wet periods and flooding, sea-surface salinity can drop substantially, and this may have been critical for successful colonization by some species, such as the endemic amphibians. ENDEMISM

The Gulf of Guinea islands are a striking center of endemism. Between them, they include 29 endemic bird species, about one-third of all the endemic birds of the overall Guinea forest biodiversity hotspot and more than the number found across the famously diverse Galápagos archipelago. There are four endemic genera on São Tomé and another on Príncipe. In a global review of priority areas for bird conservation, both São Tomé and Príncipe were considered critically important because of the high number of restricted range species occurring there. Even the tiny island of Annobon has endemic birds such as the Annobon white-eye, Zosterops griseovirescens, and the Annobon Paradise flycatcher Terpsiphone smithii. As is typical in oceanic islands, most endemic mammals are bats. São Tomé has two endemic species, Myonycteris brachycephala and Chaerephon tomensis, the latter of which was only recently discovered, along with other widespread bat species. Príncipe and Annobon both have endemic subspecies. São Tomé also has an endemic shrew, Crocidura thomensis, whereas Príncipe has an endemic subspecies of Crocidura poensis. Unlike most oceanic islands, which generally have few endemic amphibians, São Tomé has four endemic frog species, and Príncipe has three. On São Tomé there is also an endemic caecilian, Schistometopum thomense. There are no amphibians on Annobon. There are large numbers of endemic reptiles to the islands, including three scolecophidian snakes, a legless skink, and multiple gecko species. Recent analyses of genetic diversity show that several island forms, currently assigned to widespread mainland species, may also deserve recognition as distinct island endemic species. This is true for the widespread skinks Mabuya and Afroblepharus, and it may also be true of the green tree snakes. Fish are essentially marine, occurring in estuarine habitats. Rates of endemism are also exceptional in invertebrate groups. For example, 75% of terrestrial gastropods are endemic to the islands, with several endemic genera and

SÃO TOMÉ, PRÍNCIPE, AND ANNOBON

809

a monospecific endemic family, Thyrophorella thomensis. Similarly, within the Geometridae (Lepidoptera), Príncipe and São Tomé have 24 and 30 species, respectively, with 15 and 26 respectively being endemics. Most of the endemics are found on single islands, with few shared across islands. Regarding plants, the forests of São Tomé and Annobon have the highest fern diversity and density in Africa. Annobon alone holds 208 species of vascular plants, of which about 15% are endemic. Approximately 601 and 314 species of vascular plants can be found on São Tomé and Príncipe, respectively, again with high levels of endemism (14% and 8%). In particular, the Rubiaceae, Orchidaceae, and Euphorbiaceae are characteristic of the islands’ flora and have large numbers of endemics. The islands are also designated as “centers of plant diversity” by the World Wildlife Fund (WWF) and the International Union for Conservation of Nature (IUCN). Only 16 of the regional endemic plants are shared by more than one island, emphasizing the extreme isolation under which each island community has evolved and further supporting the hypothesis that each island was generally colonized independently from the mainland. HUMAN HISTORY, GEOGRAPHY, AND INFLUENCE

São Tomé and Príncipe were uninhabited prior to their discovery by the Portuguese in 1470–1471, as was Annobon, which was discovered on January 1, 1473. São Tomé and Príncipe were quickly settled, with sugar cane being the dominant crop. By the mid-seventeenth century, São Tomé was an important transit point for ships engaged in the slave trade. In the nineteenth, coffee and cocoa were introduced, and extensive plantations of these crops were developed: By the early 1900s São Tomé was the world’s largest producer of cocoa. The dominant export remains cocoa, which accounts for 95% of all exports, followed by coffee, copra, and palm products. São Tomé and Príncipe became independent from Portugal in 1975. In 1778 Annobon and Bioko passed to Spanish control as part of a larger land swap, and these islands now form part of Equatorial Guinea. Annobon has no major exports. Currently, numbers of inhabitants are approximately 137,500; 6000; and 5000 for São Tomé, Príncipe, and Annobon, respectively. Following independence in São Tomé and Príncipe, many plantations were abandoned, and there has been some regeneration to secondary forest. The relatively large area of primary montane forest remaining, coupled with a low level of exploitation of this forest, means that it is a relatively secure habitat. However, many endemic

810

SÃO TOMÉ, PRÍNCIPE, AND ANNOBON

species are associated with particular areas, such as the lowland forests, which have been widely cleared for agriculture. Ongoing forest destruction of the remaining patches remains a threat to diversity. Hunting pressures, although important for some endemic bird species, are relatively light: Birds, wild pigs, monkeys, and the common, non-endemic bat Eidolon helvum are the primary targets. Medicinal plant use is similarly almost exclusively of non-endemic species. Trapping birds for the cage-bird trade may be damaging for some species, such as lovebirds and gray parrots. It is not at all clear whether current levels of exploitation are sustainable. Small-scale agricultural practices on Annobon have generally been less damaging to terrestrial biodiversity than they have been on São Tomé and Príncipe. On both islands, large areas of secondary forest are regenerating on old plantations. On Annobon, much of the lowland forest has been replaced by savanna grasslands and banana plantations, with the exceptions of the high peaks of Santa Mira and Quioveo. Primary forests of all types cover approximately 28.5% of the islands. However, lowland forest is now restricted to a few areas. Many bird species, such as the dwarf olive ibis, Bostrychia bocagei, are restricted to these areas, and further major losses of this habitat would almost certainly cause multiple species to go extinct. Montane primary forest constitutes the majority of the primary forest, with large areas on the center of São Tomé. Mist forest is limited to the Pico de São Tomé. Mature secondary and shade forest cover 32.4% of the islands. Although not as species-rich as primary forest, these areas still support substantial diversity. Along the coasts of São Tomé and Príncipe, there are small patches of mangroves. All of the islands have the usual introductions associated with humans, such as rats, cats, dogs, pigs, and so forth. Civets, weasels, and monkeys have been introduced to São Tomé. Most of these have been introduced for more than 100 years, and although there are no records of their effect on the endemic fauna rats, weasels and civets are very likely to have a deleterious effect, especially on nesting birds. Common house geckos, Hemidactylus mabouia, have also been introduced. The islands have no endemic venomous snakes, but the cobra, Naja, has been introduced to São Tomé. Many more unreported introductions are also likely and are continuing unabated. For example, despite its extreme isolation—with the lack of an airport, a port, or any other major facility—recent assessments of the herpetofauna of Annobon reported two recently introduced species, the house gecko and the blind snake Ramphotyphlops braminus, the presence of which increases

the reptile fauna from five species to seven. Similar recent surveys of every Atlantic island group, such as the Canary Islands and the Cape Verde Islands, indicate similar levels of ongoing new introductions. Probably the greatest ongoing threat to the diversity of the islands remains introduced species. It is now impossible to assess the damage caused by the various mammals, both domestic and wild, introduced over the centuries. Reports of recent introductions ranging from terrestrial gastropods to snakes indicate that extremely high levels of introduction continue. It is unlikely that all the endemic species will be able to survive in the face of this constant alien tide.

ing 16–18 species; Procellariiformes (albatross, shearwaters [Fig. 1], petrels, and storm petrels), with over 100 species; Pelicaniformes (pelicans, cormorants, boobies [Fig. 2], frigate birds, and tropicbirds), with about 55 species; and Charadriiformes (gulls [Fig. 3], terns [Fig. 4], auks, jaegers, and phalaropes), with about 125 species. Ornithologists define “seabird” broadly, including sea ducks (Anseriformes), grebes (Gaviiformes), and loons (Podicipediformes), which are primarily marine but are largely nearshore feeders. The latter three orders and phalaropes are not included in this article.

SEE ALSO THE FOLLOWING ARTICLES

Atlantic Region / Exploration and Discovery / Introduced Species / Oceanic Islands / Orchids / Rafting FURTHER READING

Measey, G. J., M. Vences, R. C. Drewes, Y. Chiari, M. Melo, and B. Bourles. 2007. Freshwater pathways across the ocean: molecular phylogeny of the frog Ptychaden newtoni gives insights into amphibian colonization of oceanic islands. Journal of Biogeography 34: 7–20.

FIGURE 1 Wedge-tailed shearwaters (Puffinus pacificus). Photograph

by Mark Rauzon.

SARDINIA SEE MEDITERRANEAN REGION

SEABIRDS MARK J. RAUZON Marine Endeavours, Oakland, California FIGURE 2 Brown boobies (Sula leucogaster), Pearl and Hermes Reef,

SHEILA CONANT

Hawaiian Islands National Wildlife Refuge. Photograph by Sheila Conant.

University of Hawaii, Honolulu

From the nearshore waters to the open ocean, seabirds are the most conspicuous component of marine systems. Soaring on brisk winds, floating buoyantly amidst cresting waves or flying underwater, this varied group of birds has adapted to the demanding life at sea, utilizing the marine environment to feed and returning to land, primarily islands, to breed. SEABIRD SPECIES

There are approximately 350 species of seabirds in seven orders. Here, we highlight four orders most characteristic of insular systems: Sphenisciformes (penguins), compris-

FIGURE 3 Red-legged kittiwakes (Rissa brevirostris), endemic to the

Bering Sea. Photograph by Mark Rauzon.

SEABIRDS

811

may spend up to 6.7 seconds underwater. The surfaceseizing gulls are among the most successful because they are opportunists. Western gulls in the San Francisco Bay eat french fries and hot dogs as well as common murre eggs, fish offal, carrion, and invertebrates. SEABIRD ADAPTATIONS TO LIFE AT SEA AND ON LAND

FIGURE 4 Brown noddy (Anous stolidus) and chick, Manana Island,

Hawaii. Photograph by Sheila Conant.

TROPHIC RELATIONSHIPS

All seabirds rely on the productivity of the ocean to one degree or another. For most, the food web begins at the edge of the continental shelf, where strong upwelling currents convey submerged nutrients to the sunlit surface. Primary productivity of phytoplankton fuels a marine food web that changes seasonally and along clines of salinity, temperature, and nutrient availability. Water regimes, major currents, and slipstreams create feeding habitats that tend to be patchy and ephemeral. Sustained marine productivity centers are important bioregions for seabird evolution, such as the Bering Sea or the Humboldt Current off South America. Water masses define how species evolve; some are actually endemic to discrete currents or to other water masses. Some tropical seabirds such as sooty terns and brown noddies (Fig. 4) depend on a commensal relationship with dolphins and, especially, tuna. Schools of tuna drive small fish and squid to the surface, where plunge-diving seabirds and surface seizers have access to prey. Without the driving force of tuna, many tropical seabirds would have less available food. Food availability forges behavioral life history strategies, energetic costs of foraging behaviors, colony visitation schedules, clutch size and offspring development, and population and colony size. Resource partitioning of patchy food supplies has shaped seabirds’ foraging ecologies, which have evolved several methods to capture food in the water column. Surface seizing, pursuit diving, and plunge diving are the main techniques that various members of seabird families may share. For example, plunge diving is well developed among the Pelicaniformes. Tropicbirds plunge from a hundred meters over the sea, and pelicans drop from tens of meters above the sea, tucking in the wings and then hitting the water and opening their large throat pouches to net their prey. These birds dive as deep as 6.3 m, and

812

SEABIRDS

There are myriad adaptations to marine life, including webbed feet and aerodynamic wings designed for soaring and floating on ephemeral winds. Seabird wings may be supported by hollow bones reinforced with internal struts similar to models for airplane wing design, or hydrodynamic wings and flippers for flying underwater, or glands for processing the toxic salt levels that confront them, or glands whose oil waterproofs plumage. Exceptions to adaptations exist in almost every family; for example, frigate birds produce very little oil, because they rarely are in contact with saltwater. With the highest wing surface to body weight ratios, they are masterful fliers and depend on flight maneuverability and speed to secure food by kleptoparasitism (theft from other birds) or outmaneuvering flying fish in the air. They cannot afford to get wet and cannot take off again if they accidentally hit the sea surface. Galápagos flightless cormorants also produce very little oil and lack functional wings because they must remain underwater and overcome buoyancy to pursuit-dive effectively after subsurface prey. On land, Pelicaniformes use their gular (throat) pouches to thermoregulate. The gular flutter is a conspicuous behavior of boobies, gannets, and cormorants, an adaptation to shed excessive heat, especially in the tropics. Seabirds must deal with harsh contingencies of living on the ocean. Albatross, fulmars, shearwaters, storm petrels, and petrels have special glands at the base of their beaks that maintain the body’s salt balance. These glands extract salt from the blood that passes through them and excrete drops of highly concentrated brine through tubes mounted on the culmen. This latter structure is the basis for referring to the order as tubenoses, which are also known as the most accomplished of fliers. The tubenoses—albatross, shearwaters, petrels, and relatives—have long, narrow, saber-like wings, and they bank and glide easily in gale-force winds. These birds spend their early adolescence—usually several years—at sea before returning to the natal colony. Other species, such as the white tern (Fig. 5) return to land during their second year, although they may not breed until several years later.

FIGURE 5 White tern (Gygis alba) incubating an egg. Photograph by

Mark Rauzon.

BEHAVIOR AND ECOLOGY OF SEABIRD GROUPS

Albatross, the largest of all seabirds, use the energy-saving flight known as dynamic soaring. The albatross glides into the wind with one wing pointed to the water, and dips its wing tip down, slicing a wake in the sea surface. Again and again, the albatross repeats the looping circles of flight without flapping its wings so long as the wind blows across the ocean. One albatross was measured flying at 55 miles per hour; another flew for six hours without flapping its wings. They fly for days without landing and may cover several million miles in their lifetimes. To test their homing ability, several Laysan albatross were taken from their nests on Midway, flown by airplane to distant airports, and then released. One albatross was taken to Washington State; it returned to Midway in 12 days, traveling over 3,000 miles—averaging over 300 miles per day. Recent advances in telemetry suggest such large-scale movements are the norm rather than the exception for many tubenoses. Hawaiian petrels regularly visit the subArctic on two- to three-week foraging excursions while provisioning chicks in the nest burrow in the subtropics. Sooty shearwaters annually migrate over 20,000 nautical miles from the Austral breeding ground to their boreal foraging grounds in the north Pacific and Atlantic Oceans to feed on seasonally abundant euphausiids, or krill. The diminutive Leach’s storm petrel covers thousands of miles from its sub-Arctic nesting colonies as far south as equatorial waters looking for food. Like their namesake, St. Peter, who was reputed to walk on water, storm petrels use their tiny, paddle-like webbed feet to patter the ocean surface, picking up the minute arthropods this attracts, including the water striders (Halobates), the only insect that lives on the open sea. The tropical storm petrels have long legs and short rounded wings whereas northern storm petrels tend to have shorter legs and more pointed

wings. Despite many species’ affinities to particular water regimes, seabirds as a group are free to wander the ocean. Indeed, some are called gadfly petrels. No other seabirds have mastered the marine realm as have the penguins of the Southern Hemisphere. The 18 species of penguins are masters of underwater flight. Propelled by wings that have evolved into flippers, and steered by feet and tail, they fly underwater after squid and fish. On land the upright stance of the penguin and comical waddle combined with the black-and-white plumage attire has helped make them one of the most popular of animals. Alcids are the northern counterparts to penguins. The family Alcidae is comprised of 23 species including puffins, murres, murrelets, auklets, and guillemots, all of which are pursuit-divers, capable of both traditional and underwater flight. But with versatility comes a price: alcids must flap continuously to stay aloft, unlike tubenoses, and they are limited to nesting on sea cliffs, where they can freefall to gain flight momentum. The great auk was flightless and swam in pursuit of prey, propelled by its small wings. Because of its flightlessness, the great auk was quickly overhunted and became one of the first North American birds to become extinct at the hands of man in 1844. Other members of the Chadriiformes include seven species of jaegers and skuas, powerfully built hawklike gulls that out maneuver other seabirds to rob them of their catch. On land they eat rodents, birds, and bird eggs. Pomarine, parasitic, and long-tailed jaegers are boreal breeders that migrate into the southern hemisphere to spend the winter, while south polar skuas migrate into the Northern Hemisphere, avoiding the Austral winter. In summer, skuas prey on penguin chicks and scavenge at seal rookeries. Sooty terns are one of the most abundant seabirds in the world. Occurring in all tropical oceans, colonies often number in the hundreds of thousands. Sooty terns are built like miniature frigate birds and can spend as long as nine months on the wing. Scientists speculate sooty terns can sleep on the wing by resting half of the brain—one hemisphere at a time. Coincidentally, their nickname, “wide-awake” tern, refers to their incessant cries at the colony. DISTRIBUTION

Although there are numerous seabird rookeries on the coastlines of the Earth’s continents, today most are found on islands. Islands in the Southern oceans, especially the Southern Pacific Ocean, have a great diversity of breeding seabirds (Table 1). A few other remote island groups,

SEABIRDS

813

TABLE 1

HUMAN IMPACTS

Numbers of Breeding Seabirds on Some of the World’s Islands Region

Breeding Species

Southern Ocean Archipelagoes

Crozet Archipelago (IO)a Kerguelen Islands (IO) Auckland Islands and Campbell Island, NZ (PO) Prince Edward Islands (AO) South Georgia (AO) Tristan da Cunha (AO) Seychelles (IO) Galápagos Islands (PO) Falkland Islands (AO) Juan Fernandez Islands (PO) Mascarene Islands (IO)

34 31 29 28 27 22 19 19 16 16 8

Northern Ocean Archipelagoes

Britain Hawaiian Islands Aleutian Islands

24 22 26

Seabird-Poor Islands

Wallacea, Indonesian Islands, and Philippines Andaman Islands South of Java (except Christmas Island) Bay of Bengal

13 4 Virtually none Virtually none

Oceans/Seas

North Pacific/Bering Sea Tropical Central Pacific (160° E to140° W) Indian Ocean Arctic North Atlantic Tropical Atlantic

35 29 23 20 21 14

sources: Croxall et al. 1984, Gaston 2004, and references therein. a Abbreviations: AO = Atlantic Ocean, PO = Pacific Ocean, IO = Indian Ocean.

such as the Galápagos Islands, the Hawaiian Islands, and the Juan Fernandez Islands, have high seabird diversity. Island groups with surprisingly low numbers of seabird species are the islands of Wallacea, the Indonesian islands, and the Philippine Islands, which comprise hundreds of islands that support only about a dozen seabird species. In the Northern Oceans, the Aleutian Islands support breeding colonies of 26 seabird species, while Britain has 24. The Pacific Ocean seabird fauna is by far the most diverse, with the tropical Pacific supporting 29 breeding species and the North Pacific 35. In contrast, the North Atlantic has 21 breeding species and the Tropical Atlantic 14. The Arctic and Indian Oceans have similar numbers: 20 and 23, respectively. Among the world’s largest seabird colonies are Kirimati Island in the Pacific Ocean, where there used to be as many as 4 to 6 million nesting birds of 18 different species, and Isla Guafo, where an estimated 4 million sooty shearwaters breed. 814

SEABIRDS

Unfortunately the activities of humans have resulted in dramatic declines in seabird populations and distributions. For example, the apparent higher level of endemism observed in the shearwaters and petrels of the tropics is largely an artifact of local extinctions as the islands were settled 1000 to 2000 years ago. Species with restricted ranges today such as the Tahiti petrel (Pterodroma rostrata) and Abbott’s booby (Sula abbotti) were much more widespread prior to human colonization of the Pacific. Virtually all islands for which paleontological surveys exist experienced dramatic reductions in numbers of species of both land birds and seabirds after humans arrived. Perhaps the most spectacular example is Easter Island, where only one species, the sooty tern, of the original 22 seabirds known from the prehistoric record still breeds. Ironically, among other things, vast flocks of seabirds helped exploring Polynesians find the islands they colonized in the first place. Climate changes in the past millennium have altered species dynamics and distribution. Recently discovered subfossil bones on Bermuda indicate that a colony of short-tailed albatross existed there 450,000 year ago. A recent eastern range expansion and increase in number of Laysan albatross in Mexico may be a result of greater food availability through ocean regime shifts. El Niño–Southern Oscillation (ENSO) events cause unpredictable changes to the seabirds’ world. The recently elucidated Pacific decadal oscillation, which increases ocean mixing approximately every other decade, can add a positive pulse to the trickle-down economy in which seabirds live. Annual hurricanes and typhoons can set seabird recovery back by causing reproductive failure, while the vagaries of anthropogenic forces such as chronic oil leaks and spills, plastics in the ocean, pesticides in the environment, and other human disturbances limit the birds’ capacity to recover from natural events. Commercial fishing poses one of the greatest threats to seabirds. It is estimated that over 100,000 seabirds are hooked as bycatch and drown each year in the long-line fishing industry. A major ecological shift observed in the trophic structure of the Southern Ocean was originally thought to be caused by climatic change but in fact was due to depletion of fish stocks. Industrial fishing removes adult fish with K-selected (long life span, late age of first reproduction, limited production of young) life histories. The resulting declines in breeding fish stocks, as well as of annually reproducing forage fish, has contributed to declines in piscivorious seabirds such as macaroni, Gentoo penguins, and Imperial shags, which seem unlikely to recover.

CONSERVATION

Seabird conservation has made great strides devising ways to limit fishery bycatch through the use of scaring devices, setting lines at night, and other techniques. Great strides have been made in eliminating introduced predators to seabird islands (Fig. 6). By removing introduced Arctic foxes, feral cats, rodents, and various other mammals and weeds, some seabird colonies have recovered to their historic population levels. Eradication programs also provide new ecological insights on a geographic scale. Introduced foxes had altered the vegetative ecology of Aleutian Islands by eating the seabirds that fertilized the islands with guano. With decreased guano, the islands’ grasslands succeeded into scrubby tundra.

Croxall, J. P., P. G. H. Evans, and R. W. Schreiber, eds. 1984. Status and conservation of the world’s seabirds. Tech. Rep. 2. Cambridge, UK: International Council for Bird Preservation. Dickinson, E. C., R. S. Kennedy, and K. C. Parkes. 1991. The birds of the Philippines. London: British Ornithologists’ Union. Gaston, A. J. 2004. Seabirds: a natural history. New Haven, CT: Yale University Press. Harrison, P. 1983. Seabirds: an identification guide. Boston: Houghton Mifflin. Lloyd, E. G., M. L. Tasker, and K. Partridge. 1991. The status of seabirds in Britain and Ireland. London: Poyser. Rauzon, M. J. 2001. Isles of refuge: the history and wildlife of the northwestern Hawaiian islands. Honolulu: University of Hawaii Press. Steadman, D. W. 2007. Extinction and biogegraphy of tropical Pacific birds. London: Oxford University Press. Stoddart, D. R. 1984. Breeding seabirds of the Seychelles and adjacent islands, in Biogeography and ecology of the Seychelles Islands. D. R. Stoddart, ed. The Hague: W. Junk, 575–592. UNEP (United Nations Environmental Programme). 2007. Global environmental outlook. Malta: Progress Press Ltd. White, C. M. N., and M. D. Bruce. 1986. The birds of Wallacea. London: British Ornithologists’ Union.

SEA-LEVEL CHANGE W. H. BERGER University of California, San Diego

FIGURE 6 Dead wedge-tailed shearwaters, killed by dogs at Ka‘ena Pt.

on O‘ahu, Hawaii. Photograph by Lindsay Young.

Seabirds are more vulnerable than ever before. During the last 20 years, the world’s human population has increased 34%, and trade is almost three times greater, thus increasing the number of seagoing vessels and consequently increasing the risk of spreading invasive species to more islands (UN 2007). Looming on the horizon is global warming, which threatens penguins in Antarctica and tropical nesting islands and brings about shifting of water masses and increased storm intensity. Seabirds will need continued and expanded conservation efforts to ensure their survival for the future. SEE ALSO THE FOLLOWING ARTICLES

Bird Disease / Bird Radiations / Climate Change / Fish Stocks/Overfishing / Midway FURTHER READING

Croll, D. A., J. L. Maron, J. A. Estes, E. M. Danner, G. V. Byrd. 2006. Introduced predators transform subarctic islands from grassland to tundra. Science 307: 1959–1961.

Sea level has been rising around the world for the last hundred years. For the second half of the last century, the overall rate of the global rise is between 1.5 and 2 mm per year, according to various compilations. When projecting this same rise forward in time, the total rise by the end of the century turns out to be between 15 cm and 20 cm, an estimate that agrees with the latest assessment offered by the Intergovernmental Panel on Climate Change (IPCC) of the World Meteorological Organisation (WMO) and the United Nations Environment Programme (UNEP). It is distinctly lower than various estimates for the total rise by the end of the century offered since 1990 by scientists considering the issue (typically 0.5 to 1 m). There are good reasons why all such predictions are subject to a high level of uncertainty. SOURCES OF UNCERTAINTY FOR ASSESSING THE PRESENT RISE

Two major problems arise when assessing present and future rise of sea level. The first is that it is difficult to establish the historical rise of global sea level because the records of tide gauges (which are at the base of the assessment) are influenced by many regional factors that must be considered in addition to global change. The second is that the mix of factors responsible for the global rise in SEA-LEVEL CHANGE

815

the last century is unlikely to persist for the current or the next century. Some of the various factors contributing to the tide gauge records are as follows: (1) changes in the effects from tides; (2) changes in the average wind field; (3) changes in the velocity or direction of offshore current flow (which results in dynamic adjustments of sea level from the changing geostrophic balance); (4) uplift or sinking of the coast; (5) local settling or shifting of the underground from mass movements (stimulated by earthquakes or changes in rainfall); (6) changes in the contribution of runoff into a harbor bearing the tide gauge (salinity effects on density); (7) global sea-level rise. It is only the last of these that is of interest. However, the changes induced by the other factors (especially dynamic adjustments to currents and winds) are very large with respect to the signal sought. To extract the signal from a plethora of noisy data, long records must be studied, and many of them must be combined from different regions on the globe. Tide gauges tend to be clumped in highly populated regions, some of which are close to formerly glaciated regions in the north, so that tectonic adjustments to the unloading of a formerly ice-covered Canada and Scandinavia are important. Tide gauges of tropical islands have other problems; for example, some are heavily influenced by El Niño–Southern Oscillation (ENSO)-related dynamic changes of sea level (Fig. 1). The weighting of the tide gauge records, depending on their geography, will influence results, naturally. SOURCES OF THE GLOBAL COMPONENT OF THE RISE

The mix of factors determining global sea-level rise includes expansion of the water column from warming, the addition of freshwater from the melting of mountain glaciers, and a possible contribution from polar ice caps, especially

FIGURE 1 Anomalous sea-level positions during El Niño conditions,

winter 1997/1998. Source: Topex/Poseidon, NASA and NOAA, as given in Ku ˝nzi (2002), modified.

816

SEA-LEVEL CHANGE

from Greenland. The first two of these factors are amenable to estimation. The temperature change can be measured and the resulting expansion can be calculated. The fate of mountain glaciers can be assessed from images taken from spacecraft. But the present and future contribution from melting in polar regions is an open question. The reason is that warming can produce increased snowfall on large ice fields. As long as the regional climate is cold enough to retain the additional snow, there is then a buildup of ice compensating for the melting at lower elevations. With further warming, melting undoubtedly wins the contest, but the point of transition is not readily determined. In the most recent Intergovernmental Panel on Climate Change (IPCC) assessment, the assumption is made that the ice on land in Antarctica and in Greenland will retain its mass for the current century, reflecting the uncertainties mentioned. It is important to realize that this assumption simply reflects a preference for a conservative approach to the problem of rising sea level, rather than the result of reliable assessment. An assessment of an increasing contribution from polar regions will be possible only after such a contribution has materialized for some time. We note, from the foregoing discussion, that the derivation of a trend in sea-level change depends on the availability of well-spaced data points over a sufficiently long time interval and on the mathematical treatment accorded such data. When contemplating the significance of observed trends, it must also be remembered that there are hardly any measurements before the late nineteenth century; that is, there are no good data for the time before the onset of the trend. This means that the timing of the onset of the modern rise of sea level is obscure. Furthermore, given the uncertainties in the determination of the average rate of sea-level rise, an increase in the rate of rise cannot be demonstrated for the twentieth century, although it is not unlikely. Available data suggest that thermal expansion of the ocean’s water column accounts for about one-half of the observed rise. The expansion comprises the upper layer of the ocean, roughly the upper 1000 m, where the warming is most noticeable. The second most important factor is thought to be addition of water from the melting of mountain glaciers. Mountain glaciers hold sufficient ice to raise sea level by 0.5 to 1 m upon melting, and they have been retreating practically everywhere since the middle of the nineteenth century. Their retreat is especially obvious in western North America and is readily recognized from the exposure of fresh rock with no soil and but little growth of lichen on such rock (Fig. 2). Records of glacier conditions

FIGURE 2 Remnant of a mountain glacier surrounded by freshly

exposed rock and moraine debris. Glacier National Park, Montana, late summer 2006. Photograph by the author.

are commonly kept in terms of the glacier length, and height of a glacier’s surface within its valley, which allows the estimation of changes in volume and hence change of volume over the period of observation. OPEN QUESTIONS REGARDING CAUSES FOR THE OBSERVED RISE

One important problem is that the values attributed to the main factors (expansion, melting of glaciers other than polar ice masses) do not add up to the sea-level rise observed for the twentieth century. Calling on melting of polar ice could solve the problem, but this hypothesis has implications for changes in the rotation of the Earth that have not been observed. Thus, it is possible that the general rise of sea level has been overestimated for the twentieth century, perhaps as a result of changes in the shape of the sea surface as it adjusted to changes in circulation. In California, for example, a slowing of the California Current produces a rise in sea level along the coast. Of course, the fact that the apparent rise of sea level for the last century is not fully explained implies that predictions of future sea-level rise are quite unreliable. A demonstration of lack of knowledge, however, does not necessarily support complacency about future developments: uncertainty works in both directions. As mentioned, the Fourth Assessment of the IPCC (whose results have been available since early in 2007) does not attempt to specify likely contributions from the melting of polar ice masses, since the dynamics of polar ice are poorly understood. LESSONS FROM POSTGLACIAL COLLAPSE OF ICE SHEETS

While a study of the more distant past (i.e., the last several ice age cycles) cannot fill the gap in such knowledge,

it can deliver some idea about the range of possible rates of sea-level rise, during the collapse of great ice masses. Large-scale collapse of ice sheets presumably helped cause the high rates of sea-level rise indicated in recent reconstructions of deglaciation; that is, the period between 16,000 and 9,000 years ago, when the Canadian and Scandinavian ice shields disintegrated. It is generally agree that the disintegration proceeded in pulses, the two major steps of deglaciation being separated by a cold spell with static sea level known as the Younger Dryas (YD) episode. The YD cold period lasted somewhat longer than one millennium. It had almost fully glacial climate conditions and it kept sea level at roughly one half of its total rise (120 m) at an intermediate position (60 m), halfway through deglaciation. The overall rate of the rise was between one and two meters per century, for a thousand years at a time. The crucial information allowing assignment of such rates is absolute dating of corals growing near sea level. Uncertainties remain because corals have a depth range for habitat, rather than a depth level. However, a rise of 2 to 3 m per century is plausible, for intervals spanning a few centuries at a time, given the rate estimates obtained. The estimates of contributions from the various ice shield sources vary; roughly two thirds of the rise is thought to be from North America and Greenland, the rest being shared by the Fennoscandian shield and Antarctica. Calculations regarding relative contributions are based on isostatic rebound statistics and have large error bars. An important question in the context of expectations for future sea-level rise concerns the vulnerability of ice sheets that remain after deglaciation to additional warming. At present, the ice remaining in Antarctica, if melted, could raise sea level by about 61 m, while the ice in Greenland could provide for a rise in sea level of 7 m. The question is difficult to decide on the basis of ice physics, because of the mass-wasting aspect of ice sheet disintegration. However, clues to answers are contained in the behavior of ice sheets and sea level over the past million years or so. PAST SEA-LEVEL RISE AS SEEN IN THE DEEP-SEA RECORD

The relevant data are oxygen isotope ratios within the shells of microscopic organisms called foraminifers, a minor but important part of the plankton of the ocean. They also have representatives living on the sea floor; that is, benthic foraminifers. The isotopic composition of their shells reflects the isotopic composition of seawater at the time of growth; records are recovered by coring the sediments on the deep seafloor.

SEA-LEVEL CHANGE

817

The isotopic composition of the benthic shells, in particular, is largely controlled by changes in the mass of polar ice sheets. The water used in making ice, which is extracted from the sea, is depleted in the heavier of the two main oxygen isotopes (18O). Thus, when ice grows, the seawater is enriched in this isotope. The range of change over the last million years, in benthic foraminifers, is roughly 1 permil (one tenth of one percent of change in the deviation from a standard). This change is readily measurable (to about 5% precision). The corresponding change in sea level can be estimated if it is assumed that the temperature effect on the composition is tightly linked to the ice effect. Also, the isotopic composition of the ice is taken as constant through time. The errors stemming from these assumptions are thought to be modest, relative to the estimates of sea-level change (on the order of 10% or less). When studying the patterns of change in the isotope values with the goal of reconstructing rates of sea-level change, one finds that rates of sea-level rise of more than 1 m per century are not highly unusual over the last million years. A similar pattern of rates (although based on few data) is obtained for values drawn from the fifth percentile of sea-level positions, on the high end, which would seem to indicate that ice sheets left over after deglaciation (the present situation) remain vulnerable to removal up to a +10-m limit. Thus, the available information from the history of sea-level change through the last million years suggests that the present sea level is not privileged but can readily move upward by 5 m or more, and that such rise could proceed at a rate of 1.5 m per century given sufficient motivation (that is, sufficient warming in high latitudes). SUMMARY

In the twentieth century, sea level has been rising between 1 mm and 2 mm each year, for a total rise of about 15 cm for that century. Estimates for the rise in recent years are between 2 and 3 mm per year, suggesting an acceleration of the rise, which is expected but not demonstrated. Precise assignment of sea-level rise is not possible, because observations are not globally distributed and are greatly influenced by local conditions. Most of the inferred rise is attributed to thermal expansion of the water column as the ocean has warmed. Some of the rise is attributed to the melting of mountain glaciers, whose ice mass is shrinking quite rapidly (which is readily verified by satellite surveys and from historical records). Forward extrapolation of observed and inferred sea-level rise yields estimates for the end of the present century with values that vary between less than 30 cm and distinctly less than 1 m. These estimates assume little or no contribution from polar ice masses. In

818

SEAMOUNTS, BIOLOGY

the recent geologic past, such ice masses have disintegrated at rates equivalent to a sea-level change of around 1 to 2 m per century when stimulated by summer warming. There is no evidence that such stimulation will not produce similar rates in the near future. Instead, the available record of ice behavior for the last million years suggests that at present there is enough vulnerable ice in existence to raise sea level by between 5 and 10 m, within a few centuries. CONCLUSIONS

The uncertainties for assessment for present and future sea-level rise are rather large. Data concerning past sealevel behavior suggest that, once started, disintegration tends to persist for some time, on the scale of centuries, which points to internal positive feedback. However, the resolution (1000 years) of the observational series for the last million years or so is insufficient to belabor the issue for time scales below that resolution. SEE ALSO THE FOLLOWING ARTICLES

Climate Change / Coral / Surf in the Tropics / Tides FURTHER READING

Berger, W. H. 2008. Sea level in the Late Quaternary: Patterns of variation and implications. International Journal of Earth Sciences 97: 1143–1150 Church, J. A., J. M. Gregory, P. Huybrechts, M. Kuhn, K. Lambeck, M. T. Nhuan, D. Qin, P. L. Woodworth, et al. 2001. Changes in sea level, in Climate change 2001: the scientific basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. J. T. Houghton, Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K. Maskell, and C. A. Johnson, et al., eds. Cambridge, UK: Cambridge University Press, 639–693. Künzi, K. 2002. Ozeane aus der Ferne gesehen [Oceans looked at from afar], in Der Ozean—Lebensraum und Klimasteuerung [The ocean— life habitat and climate control]. G. Hempel and F. Hinrichsen, eds. Bremen: Hausschild Verlag, 63. Munk, W. 2002. Twentieth century sea level: an enigma. Proceedings of the National Academy of Sciences of the United States of America 99.10: 6550–6555. Revelle, R. R., ed. 1990. Sea-level change. Studies in Geophysics. Washington, DC: National Academy Press.

SEAMOUNTS, BIOLOGY MALCOLM CLARK National Institute of Water and Atmospheric Research, Kilbirnie, New Zealand

Seamounts occur in all oceans of the world, from the tropics to the poles, and cover depth ranges from near the surface to the abyss. They provide a wide variety of

habitat types for a huge range of animals and often feature high levels of biodiversity and abundance. This can make them important components of oceanic ecosystems, yet also the target of commercial exploitation. THE SEAMOUNT ENVIRONMENT

Seamounts have three important characteristics that distinguish them from the surrounding deep-sea habitat. First, they are “islands” of shallow sea floor, surrounded by the deep ocean, and they provide a range of depths for different communities. Second, the majority of the ocean sea floor is covered with fine, unconsolidated sediments. In contrast, the often steep slopes of seamounts, combined with accelerated currents over them, can keep sediment from depositing, and bare rock surfaces can be common. Third, the physical structure of some seamounts enable the formation of hydrographic features and current flows that can restrict the dispersal of larvae and plankton and keep species and production processes concentrated over the seamount rather than dispersing them into the wider ocean system. SEAMOUNT FAUNA

The existence of seamounts has been known for a long time, but their biology received little attention until the late 1950s. In a review in 1987, it was reported that about 96 seamounts had been sampled, but recent estimates by the Census of Marine Life program on seamounts (CenSeam: http://censeam.niwa.co.nz) have the number now up to about 300 (Fig. 1). But of this total, most have been sampled opportunistically, with few specific or comprehensive biological studies having been undertaken. Geographically, the most comprehensive data on seamounts come from the northeastern Atlantic, the southwestern Pacific, and the southeastern Pacific. Other scattered seamount groups have also received study, but major gaps exist in global coverage, which makes robust estimation of biodiversity and interpretation of biogeographical patterns difficult. Nevertheless, the benthic communities of species found on seamounts are rich and varied. This is partly related to the wide variety of habitat types that seamounts can offer animals. On seamounts where currents are slow and there are sandy or muddy sediments, we can find a predominance of deposit-feeding species that utilize the matter sinking down from mid-water layers. These macrofaunal communities include polychaetes, echinoderms (ophiuroids, holothurians, echinoids), various crustacean groups, sipunculids, nemertean worms, molluscs, sponges, and nematodes. However, where faster water

FIGURE 1 The location of seamounts where biological samples have

been taken and data are available in the Seamounts Online database (as of October 2007).

flows occur over seamounts (often caused by the size and shape of the seamounts themselves), there can be greater regions of hard-bottom habitat. These rocky areas can be dominated by suspension feeders, which eat material suspended in the water by the enhanced current flow. These animal groups include corals, crinoids, hydroids, ophiuroids, and sponges. The large corals and sponges can form extensive and complex reef-like structures, which themselves provide a habitat for smaller mobile fauna. SEAMOUNT BIODIVERSITY

It is not known how many species occur on seamounts. Accounts of seamount biodiversity show a rapid increase in number of known species with research in the last few decades. A global review in 1987 recorded 449 species of fish and 596 species of invertebrates from 100 seamounts. Research in the southwestern Pacific around New Caledonia has found 730 species from just 18 seamounts at present, and 192 species of invertebrates and 171 species of fish have been recorded from a single seamount chain off South America. Worldwide, almost 800 species of fish have been recorded from seamounts. Data on seamount biodiversity is growing daily, through both national research programs and through the efforts of CenSeam, and the data are being input to a public database, Seamounts Online (http://seamounts.sdsc.edu). This database in 2007 has over 3000 taxa (although not all to species) from 250 seamounts. This number will continue to increase as more seamounts around the world, at different depths and with different ages and geological histories, are sampled. However, selectivity of sampling gear, and generally a small number of samples per seamount, means that not all species present are caught. Typically, biodiversity is not well described even after a large number of samples have been taken from a group of seamounts. It seems certain that biodiversity is not fully known for any seamount.

SEAMOUNTS, BIOLOGY

819

Seamounts can be hotspots of biodiversity. Sessile animals such as sponges and corals can attach to the hard substrate. Corals, in particular, can form large reeflike structures that provide a home for hundreds of other species. Deep-sea, or “cold-water,” corals do not need light to exist (as tropical shallow corals do) but feed on matter swept up by the currents on seamounts. These corals can be hundreds of years old and very slow growing. Despite sampling limitations, comparisons of species richness have been made between some seamounts. Results comparing seamounts with adjacent continental slopes and with deep sea floor areas have been mixed, with no consistent trend of elevated or depressed biodiversity. It is likely that various seamounts and regions differ in their comparative biodiversity levels, with some being high and others not. In many terrestrial and marine habitats, there is a relationship between habitat size and the number of species present. Hence, because of the small size of seamounts, fewer species are expected. This is likely to be the case, but the biodiversity of seamounts is, nevertheless, relatively high for their size. It is possible that seamount habitat is relatively stable over evolutionary time, which means that species may accumulate. A common feature of seamount biodiversity is low levels of species evenness, where the fauna can be dominated by a few very abundant species. The biomass of fishes and other planktivorous animals is often unusually high over seamounts. Upwelling of currents can occur, but is generally not strong enough, or permanent enough, to increase the growth of local zooplankton populations. Trophic enrichment of seamounts can occur but is caused by processes such as the entrapment of vertically migrating plankton by the sea floor, whereby plankton descending in the morning strike the seamount summit or flanks and accumulate, thus providing a rich food source for fishes. Enhanced horizontal fluxes of suspended food also occur, caused by increased current flow over and around a seamount. Hence, the biological “enrichment” of seamount fauna may not be due to increased primary production but to a bottom-up pathway that provides a greater food supply for carnivores.

regions. It is not as clear with smaller species or invertebrates. Most studies to date (on fish) have not supported the hypothesis that seamounts are highly isolated, and this is in conflict with the frequent observations of endemic species. However, benthic invertebrates may have different characteristics. For some groups, it is highly likely that distance between seamounts can drive isolation, as animals need to disperse over areas of ocean and locate the relatively small surface area of a seamount. Alternatively, reproductive mechanisms need to be adapted so that eggs and larvae remain local in the vicinity of the seamount where they were spawned. Oceanographic conditions may also play a role in retaining small animals and in promoting selfrecruitment to the same seamount. Seamount communities often have faunal affiliations with the nearest continental margin. Despite instances of high endemism and high variability of some seamount faunas, they tend generally to comprise the same broad biogeographic regional fauna as found on the continental slope. Seamounts can also occasionally yield taxa that have been thought extinct. These “relicts” or living fossils may exist on offshore seamounts long after environmental changes have occurred in continental shelf or slope waters. SEAMOUNT ENDEMISM

A feature of seamount fauna that is often referred to is the occurrence of highly endemic faunas. Studies have recorded variable levels of endemism, from 5 to 10% for fishes on some seamount groups up to levels as high as 50% for invertebrates. Despite being highly variable, endemism levels are possibly lower in fishes than invertebrates, given the wider distribution of fish ranges compared to many invertebrate groups, and also given generally broad reproductive outputs. A recent review of levels of endemism on seamounts found an overall average of about 20%. True rates of endemism are rarely known, especially on seamounts, where sampling has been so limited. Published rates of endemism are best regarded as indicative: They could increase as more cryptic species are differentiated or decrease as species ranges are better described.

SEAMOUNT ISOLATION

Seamounts close together can have markedly different faunal communities. The degree of isolation is affected by whether recruitment occurs from animals already living on the seamount, or whether recruits come in from wider areas. For large fish species and pelagic animals, there is evidence that seamounts are not isolated populations. Tuna, turtles, marine mammals, and seabirds are known to migrate between seamounts and between non-seamount

820

SEAMOUNTS, BIOLOGY

HUMAN EXPLOITATION

Dense aggregations of animals can occur on seamounts. These include fishes, among which are a number of commercial species that are the target of trawl or line fisheries: for example, orange roughy, slender armourhead, and alfonsino. Seamounts are also of increasing interest for mining for deposits of polymetallic sulfides and cobaltrich ferromanganese crusts.

However, seamount communities may be severely affected by exploitation. Deep-water coral–dominated communities are highly vulnerable to physical disturbance, with the coral matrix being fragile to impact. As well as direct removal of species and structural habitat, there can be indirect effects caused by sediment resuspension. Furthermore, changes in community structure can occur with selective removal of species, and the interactions between pelagic and benthic components of seamount ecosystems may change. Given the slow growth rates of many seamount benthic species, recovery rates may be slow, if in fact disturbed systems ever return to their original state. Natural changes are also occurring, and long-term ocean acidification will have important consequences for seamount coral species, which are affected by the chemical composition of the water at depth. However, in the short term, direct human impacts are of most concern, and careful management of seamount resources is required to balance exploitation and conservation of the habitat.

to several tens of meters in height. They generally exhibit a conical shape with a circular, elliptical, or more elongated base. Seamounts are some of the most ubiquitous landforms on Earth and are present in uneven proportions in all ocean basins. Being volcanic in nature, seamounts are mostly found on oceanic crust and to a much lesser extent on extended continental crust. They are generated near mid-ocean spreading ridges, in plate interiors over upwelling plumes (hotspots), and in island-arc convergent settings. Oceanic islands form a small subset of large seamounts that have breached sea level. THE VOLCANIC ORIGIN OF SEAMOUNTS AND ISLANDS

Most seamounts and islands are constructional aggregates of basalt, reflecting their volcanic origin. Seamounts are typically formed in one of three distinct tectonic settings, each imparting unique tectonic characteristics to its offspring.

SEE ALSO THE FOLLOWING ARTICLES

Intraplate Seamounts

Climate Change / Reef Ecology and Conservation / Seamounts, Geology

The majority of larger seamounts found in the ocean basins were formed in an intraplate setting. Because of their frequent alignment into linear, subparallel chains that correlate with the direction of past plate motions, the consensus origin of such seamounts is given by the hotspot hypothesis, which states that these seamounts formed above more or less stationary mantle plumes (or hotspots) in the Earth’s mantle. As the plates move, the seamounts thus formed are carried away from the source of magma and cease volcanic activity, building a line of extinct volcanoes that exhibits a monotonic age progression reflecting the plate motion history. Numerous hotspots have been proposed for sites of unusual volcanic activity, yet conclusive imaging of mantle plumes using seismic tomography remains elusive. Although the simple age progressions predicted by the hotspot hypothesis have been confirmed for several seamount chains, others show complex age patterns, which cast doubt on the hotspot theory as the only explanation. Seamounts formed by hotspot volcanism may grow quite large (Fig. 1A). In particular, hotspot seamounts formed on old (and hence thicker and stronger) oceanic lithosphere can in some cases reach almost 10 km (measured from their base), making Mauna Kea (one of five volcanoes that form the Big Island of Hawaii) the tallest mountain on Earth. Seamounts formed on oceanic crust must reach at least 2.5 km in height just to match the typical mid-ocean ridge depth; however, most larger seamounts were formed in even deeper water; hence, only truly large seamounts will

FURTHER READING

Pitcher, T. J., T. Morato, P. J. B. Hart, M. R. Clark, N. Haggan, and R. S. Santos, eds. Seamounts: ecology, fisheries, and conservation. Blackwell Fisheries and Aquatic Resources Series 12. Oxford, UK: Blackwell Publishing. Richer de Forges, B., J. A. Koslow, and G. C. Poore. 2000. Diversity and endemism of the benthic seamount fauna in the southwest Pacific. Nature 405: 944–947. Rogers, A. D. 1994. The biology of seamounts. Advances in Marine Biology 30: 305–350. Stocks, K., and P. J. B. Hart. 2007. Biogeography and biodiversity of seamounts, in Seamounts: ecology, fisheries, and conservation. T. J. Pitcher, T. Morato, P. J. B. Hart, M. R. Clark, N. Haggan, and R. S. Santos, eds. Blackwell Fisheries and Aquatic Resources Series 12. Oxford, UK: Blackwell Publishing, 255–281. Wilson, R. R., and R. S. Kaufmann. 1987. Seamount biota and biogeography, in Seamounts, islands and atolls. B. H. Keating, P. Fryer, R. Batiza, and G. W. Boehlert, eds. Geophysical Monograph 43, 355–377.

SEAMOUNTS, GEOLOGY PAUL WESSEL University of Hawaii, Manoa

Seamounts are traditionally defined as undersea mountains whose summits rise more than 1000 m above the sea floor; however, modern studies describe seamounts down

SEAMOUNTS, GEOLOGY

821

become islands or have a shallow-water presence. Because large seamounts often penetrate the euphotic zone, they have been the main focus of ecological studies, despite being a small subset of all seamounts globally. Mid-Ocean Ridge Seamounts

Most seamounts are small and were formed near a divergent plate boundary. Here, excess amounts of magma percolate through the thin, fractured crust to form small, sub-circular seamounts—often just a few tens to hundreds of meters tall. Occasionally, larger seamounts can be formed (Fig. 1B). It is likely that most small seamounts formed in this near-ridge environment as the thickness of the lithosphere rapidly increases away from ridges, making the ascent of small amounts of magma from an increasingly deeper source less likely. Consequently, seamount production rates decrease with increasing crustal age and lithospheric thickness, being highest close to the ridge axis. At fast-spreading ridges (e.g., the East Pacific Rise), small seamounts form on the flanks of the ridge where the crust is just 0.2–0.3 million years old, and their abundance correlates with spreading rate. At slow-spreading ridges (e.g., the Mid-Atlantic Ridge), small seamounts are produced almost exclusively within the median valley. Many new seamounts undergo extensive tectonic deformation by normal faulting, which reduces their original heights considerably. Because of increased sediment coverage on older sea floor, the smallest and most numerous seamounts, with heights less than 100 m, are likely to be buried after a few tens of millions of years. Island Arc Seamounts

FIGURE 1 (A) Intraplate seamount formation over the Hawaii hotspot.

On thick lithosphere, seamounts can grow very tall and even breach sea level to form islands. The volcano deforms the lithosphere, which responds by flexure. The hotspot feeds the active volcanoes by a network of feeder dikes; magma may pond beneath the crust as well. As plate motion carries the volcanoes away, they cease to be active and form a linear seamount chain. (B) Seamount formation near the East Pacific Rise. A thin plate cannot sustain large volcanoes, and typically only smaller cones are found. Excess magma is diverted into feeder dikes that reach the surface on the ridge flank, forming small volcanoes. (C) Island arc formation behind the Kermadec Trench. The subducting Pacific plate and its sediments will induce melt at depth, eventually erupting to create a volcanic arc that parallels the subduction zone. Note that the oldest part of the Louisville seamount chain (another intraplate chain) is currently being subducted.

822

SEAMOUNTS, GEOLOGY

Island arc seamounts form at subduction zones where one oceanic plate is being forced to subduct beneath the other. As plates descend into the mantle, the higher pressure and friction and the increasing temperatures and water content eventually cause decompressional melting that produces an ascending basaltic melt of a different magmatic composition than the basalt available at spreading centers (Fig. 1C). The magma may be more volatile, thus increasing the chance of explosive eruptions. The distribution of island arc seamounts and islands reflects the trend of the convergent plate boundaries, and the overall plate tectonic geometry places strong constraints on the evolution of such seamounts. These island arc seamounts are found in the relatively narrow collision zones between the converging tectonic plates, thus occupying a small area of the total sea floor. Like hotspot-produced seamounts, island arc seamounts can reach considerable height and often form islands. Unlike hotspot-produced seamounts, the volcanic activity along an active arc is essentially simulta-

neous, geologically speaking, with older seamounts constantly being overprinted by younger ones. MORPHOLOGY AND THE EVOLUTION OF SEAMOUNTS

Seamounts are born kilometers below the sea surface. Following a pathway of preexisting cracks or weaknesses, buoyant magma finds its way to the ocean floor. Here, emerging seamounts may be exposed to overburden pressures of 25–50 MPa. Consequently, volcanic gases within the magma cannot expand, and extrusive flows are effusive. The cooling effect of seawater affects the shape of the volcano, allowing construction of steeper flanks (greater than 10°) than would generally be possible once the volcano builds up above sea level (less than 10°). At first, the seamount is fed from a central vent, yielding an almost circular feature, and some develop summit craters. Many seamounts do not develop beyond this stage. However, if adequate magma supply is available, and the seamount is allowed to grow taller, then gravitational stresses in the flanks of the seamount, possibly enhanced by flexural stresses transmitted from the increasingly deformed subsurface, will favor the development of rift zones. These break the circular symmetry and promote construction of long ridges from fissure eruptions. As the summit of the seamount approaches sea level, water pressure can no longer keep gases locked up in the magma, and explosive eruptions become common. The extrusive products tend to be finer-grained, more vesicular, and structurally less resistant to erosion, which begins to shape the islands, augmented by

catastrophic submarine landslides. The combined effect of rift zones, erosion, and landslides is to modify the basic circular form of seamounts into stellate forms. Once the island is well established, the volcano enters the shield-building stage, during which large flows of ‘a‘a¯ and pa¯hoehoe lava are extruded. When active construction finally wanes, the island no longer regenerates to keep up with the destructive forces of erosion, which combine with long-term thermal subsidence of the sea floor and isostatic adjustments to bring the summit area back to sea level, where wave erosion forms a flat-topped guyot. Coral growth may keep up with the subsidence rate, capping many volcanic islands with a thick coral reef layer before subsidence eventually drowns the seamount. Complex interplays between eustatic sea-level changes, vertical isostatic adjustments, and latitude changes caused by plate motion result in a wide variety of seamounts, some with fringing reefs, others with lagoons with calcareous sediments, and others that never developed a coral cap and may have drowned long ago. THE DISTRIBUTION OF SEAMOUNTS

Seamounts are distributed both in space (geographically) and time (temporally), and studies of these variations have provided key insights into several factors that control the formation of seamounts. Spatial Distribution

The number of seamounts varies considerably between ocean basins (Fig. 2). Seamounts form both linear and

FIGURE 2 Distribution of seamounts inferred by satellite altimetry. Colors reflect seamount sizes (see legend). The majority of seamounts can

be found in the Pacific basin, with the remainder divided between the Atlantic and Indian Oceans. Large igneous provinces (LIPs) are outlined in orange; these are often associated with seamount provinces.

SEAMOUNTS, GEOLOGY

823

random constellations, and their sizes and distributions provide invaluable information about their origins. Studies have found that the distribution of seamounts over a wide range of sizes is well approximated by an exponential or power-law model. Such models reflect the observation that most seamounts are small, and by extrapolating the power-law, there may be perhaps as many as 100,000– 200,000 seamounts reaching heights of 1 km or more (Fig. 3A). Extrapolation further down to the smallest seamount sizes observed (a few tens of meters) would predict seamount populations reaching into the millions, but the majority of such small seamounts will likely be buried, given the typical thickness (100–200 m) of sediment in the world’s ocean basins. Consequently, the smallest seamounts are observed only on young sea floor with modest sediment cover. Seamount summit depths are normally distributed around a mean depth of ~3 km. Notably, the shallow end seems to have an additional number of shallow seamounts and islands, possibly reflecting the ability of coral reef growth to keep up with subsidence for long periods of time (Fig. 3B). A

Seamount Count

105 UNCHARTED SEAMOUNTS? GLOBAL SEAMOUNT COUNT

104 103 102 101 1

2

3

4

5

6

7

8

Predicted Seamount Height (km)

B 400

Seamount Count

ABYSSOPELAGIC

BATHYPELAGIC MESOPELAGIC

300

EPIPELAGIC

200

100

0 7

6

5

4

3

2

1

0

-1

The abundance of seamounts has been shown to vary considerably among the ocean basins. The Pacific basin is host to nearly half of the seamounts that are large enough (greater than ~ 2 km) to be mapped by satellite altimetry. The Atlantic and Indian Oceans combine to contain most of the remaining seamounts, with considerably fewer seamounts appearing on plate segments located at high latitudes (e.g., northern Atlantic on either the North American or Eurasian plates) or on relatively small plates (e.g., Cocos, Philippine Sea). It is not clear what causes seamount abundances to vary spatially. One factor may be the underlying distribution of mantle plumes, which are found in higher numbers beneath plates with numerous seamounts. However, one would still expect excess magma at the spreading center to produce the smaller and more numerous seamounts. Another factor may be systematic variations in plate stresses, with smaller plates possibly being in a compressional stress state, which would not favor the intrusion of magma. Smaller plates are also less likely to have a directional regional stress dominating the state of stress. In contrast, the large Pacific plate, in particular the equatorial region, appears to be under tension from distant slab pull forces, as evidenced by widespread extensional volcanism associated with neither hotspots nor mid-ocean ridges. Finally, plates that move the fastest appear to have the highest seamount abundances, provided they share at least one spreading plate boundary. Island arcs aside, the distribution of seamounts appears as a superpositioning of two separate processes: Divergent plate boundaries produce a near-steady stream of new, small seamounts, most of which exhibit no particular clustering pattern, whereas mantle plumes or hotspots generally create both small and large seamounts, which are often organized in linear groups by plate motions. Frequency-size analysis (Fig. 3A) of the combined seamount populations does not immediately separate out the two modes of production, but this possibly reflects the inability of satellite altimetry to detect smaller seamounts (less than 1 km) and the lack of significant spatial coverage of small-size seamount provinces using multibeam techniques.

Summit Depth (km) FIGURE 3 (A) Seamount size–frequency distribution. The green circles

Temporal Distribution

indicate the number of seamounts taller than a given size. For sea-

Seamounts are among the youngest geologic features on Earth, reflecting the youthfulness of the oceans and the regenerative processes of plate tectonics. Only a few seamounts are currently volcanically active, and they tend to be restricted to (1) the very youngest volcanoes of hotspot island chains (such as Hawaii, Samoa, Réunion, and oth-

mounts greater than 2 km tall, the data are well explained by a scaling rule (solid line). For heights less than 2 km, the trend levels off because numerous smaller seamounts fall below the resolution of altimetry data. Only ~15,000 out of a potential of ~200,000 seamounts greater than 1 km in height may have been mapped. (B) Summit depths follow a normal distribution centered on a ~3-km depth, with large and shallow seamounts appearing as outliers, possibly as a result of coral growth.

824

SEAMOUNTS, GEOLOGY

ers), (2) various places along active island arcs, and (3) newly formed smaller seamounts associated with midocean ridges. Many volcanic islands, but only a few seamounts, have been dated using radiometric techniques, yet the sparse age data, the underlying sea-floor age, and the size of seamounts imply that the production of seamounts is not steady-state. During the Cretaceous (146– 65 million years ago) the Pacific seamount production was almost twice as high, resulting in numerous large seamounts now residing in the western Pacific. This period also saw the formation of several large oceanic plateaus, such as the Ontong Java, the Manahiki, the Shatsky, and the Mid-Pacific Mountains; hence, plateau and seamount formation appear correlated.

carbon dioxide can penetrate into the deep oceans, and assumed rates of vertical mixing can considerably affect model predictions. Mineral Resources Impact

Older seamounts may accumulate a ferromanganese oxide crust enriched in the elements cobalt, copper, manganese, and sulfur, typically occurring at depths exceeding 3 km. The total cumulative amounts of such marine mineral resources might exceed the amounts currently available on land. So far, the cost of harvesting deep ocean nodules and crusts has been prohibitive. However, rising prices associated with depletion of terrestrial resources will likely make deep ocean resources more attractive, especially because the bulk of these are in international waters.

THE IMPACT OF SEAMOUNTS Geologic Impact

SEE ALSO THE FOLLOWING ARTICLES

Seamounts are windows into the mantle that allow scientists to study the nature of erupting magma. Minor changes in the chemical and isotopic composition of basaltic lavas can be used to make inferences about magma source depth and composition. Seamounts represent a significant fraction of the entire crust production, perhaps as much as 5–10%, and variations in this intraplate volcanic budget shed light on plate tectonics and how Earth gets rid of excess heat. The alignment of seamount chains provides a means to decode the motion of tectonic plates over long geologic intervals, enabling an understanding of the climatic changes experienced at islands that simply follow from latitudinal migration of plates carrying seamount provinces on their backs. Many seamounts have active hydrothermal convection systems that may have a significant effect on element cycles involving seawater, and they also participate in the dissipation of residual heat from the formation of both seamount and sea floor. Finally, seamounts and islands act as measuring sticks for relative sea-level variations, which can have both eustatic and tectonic components.

Island Arcs / Plate Tectonics / Sea-Level Change / Seamounts, Biology

Oceanographic Impact

KENNETH Y. KANESHIRO

Bathymetry influences ocean circulation in several ways. The first-order features such as ridges and plateaus steer currents and, in places, act as barriers that prevent deep waters from mixing with warmer, shallower waters. Smaller-scale bathymetry, such as seamounts, may play a largely overlooked role in the turbulent mixing of the oceans. Measurements suggest that mixing around a shallow seamount is many orders of magnitude more vigorous than in areas far from seamounts. Understanding how the climate will evolve depends on how quickly heat and

University of Hawaii, Manoa

FURTHER READING

Batiza, R. 2001. Seamounts and off-ridge volcanism, in Encyclopedia of ocean sciences. J. H. Steele, S. A. Thorpe, and K. K. Turekian, eds. San Diego, CA: Academic Press, 2696–2708. Keating, B. H. 1987. Seamounts, islands, and atolls. Washington, DC: AGU. Macdonald, G. A. 1986. Volcanoes in the sea, 2nd ed. Honolulu: University of Hawaii Press. Schmidt, R., and H.-U. Schminke. 2000. Seamounts and island building, in Encyclopedia of volcanoes. H. Sigurdsson, ed. San Diego, CA: Academic Press, 383–402. Wessel, P. 2001. Global distribution of seamounts inferred from gridded Geosat/ERS-1 altimetry. Journal of Geophysical Research 106: 19, 431–419, 441. White, S. M. 2005. Seamounts, in Encyclopedia of geology. R. C. Selley, et al., eds. London: Elsevier, 475–484.

SEXUAL SELECTION

RICHARD T. LAPOINT University of California, Berkeley

Sexual selection is defined as the differential mating success among individuals of the same sex, males in most cases. Sexual selection is viewed as a dynamic, frequencydependent process, a driver for evolutionary change, and a synergist for the formation of new species.

SEXUAL SELECTION

825

ISLANDS AND ISLAND BIOTA

Because of their isolated nature, islands are among the best places for investigating evolutionary processes, and research on specific groups of organisms that evolved on islands, or island-like habitats, has begun to shed light on processes by which species diversify and originate, processes such as sexual selection. For example, the Hawaiian Archipelago is often considered to be the world’s most outstanding living laboratory for the study of evolutionary biology, and in the most diverse lineages (e.g., Drosophila and crickets), sexual selection is hypothesized to drive evolution in combination with other factors. Likewise, the African Rift lakes have provided an outstanding context for studying the role of sexual selection in driving proliferation of cichlid fish. At the same time, in other radiations (e.g., Anolis lizards in the Caribbean and Tetragnatha spiders in Hawaii), sexual selection appears to play a secondary role to natural selection. MODELS OF SEXUAL SELECTION

Though known to be important at the population level, sexual selection is also believed to be a major force in divergence evolution. In his text on sexual selection, Darwin (1871) discussed how extreme secondary sexual characters often characterize diverse species groups. A number of models have been proposed for explaining how sexual selection might lead to rapid speciation. Runaway sexual selection was initially alluded to by Darwin and later was mathematically explored by R. A. Fisher (1930). The general concept is that one sex (usually female) develops a preference for a specific trait, or ornament, in the other sex. Quantitative genes for this preference become genetically correlated with the quantitative genes for the ornament. Over time this creates a strong positive feedback loop, where stronger preference drives a more exaggerated ornament. Very rapidly the ornament becomes more embellished to the point where the fitness benefits provided by enhanced mating success is outweighed by the selective disadvantage such an extreme trait confers. Runaway sexual selection requires the female preference to evolve for a male trait present in the population and that this trait be quantitative and heritable. Several authors, such as Russell Lande and Mary Jane West-Eberhard, have provided models showing that sexual selection, in conjunction with drift or ecological specialization, could cause speciation. After a large population with diverse mating behaviors or morphology becomes isolated, different morphologies, and subsequently female preferences, could become fixed in allopatry. When these two lineages come into contact, differences in mating behavior prevent gene flow.

826

SEXUAL SELECTION

In the specific case of Hawaiian Drosophila, studies by Hampton Carson and Kenneth Kaneshiro have shown that there is a range of mating types segregating in both sexes within a population, some males being more successful in mating than others. Similarly, there are females that are highly discriminant in mate choice and others that are not as choosy. In this situation, it appears that sexual selection occurs via a form of density-dependent selection. Many of the most diverse groups of island organisms are characterized by extreme secondary sexual characteristics and with sexual selection appearing to play a key role in diversification. Hawaiian Drosophila are a prime example, with many species that are closely related but clearly distinguishable based on exaggerated secondary sexual characters such as intricate wing patterns or modified tarsal segments and complex mating behaviors. Laupala, or Hawaiian crickets, are similarly a diverse group characterized mainly by mating calls. In the same way, New Guinea’s birds of paradise are strikingly diverse in their mating displays and plumage. In the insular habitat of the African Rift Lakes, cichlid fish have evolved a large variety of color morphs used in mate recognition. In Laupala at least, a number of attributes of the system have lead researchers to conclude that speciation is the result of forces acting on secondary sexual traits. This is because closely related species are morphologically cryptic and can only be distinguished by the pulse rate of the male courtship song, a secondary sexual trait (females prefer males with pulse rates of their own species); moreover, different species are often syntopic and synchronic. A similar argument has been used to suggest that the spectacular diversification in African cichlid fish is driven by sexual selection on male coloration and perception thereof. THE FOUNDER PRINCIPLE AND SEXUAL SELECTION

Since islands are defined by isolation, it is assumed that any new population is most likely to have been founded by a small number of individuals. Mayr first proposed and further developed the general concept of the founder principle. He argued that the reduced genetic diversity that accompanies founder events provides the mechanism by which a “genetic revolution” or new coadapted genetic system can arise and speciation could occur. Hampton Carson proposed a slightly modified version of the founder principle, suggesting that drift served to rearrange the coadapted genetic system of founding populations. His model hypothesized that the founder event is followed by rapid population growth as a result

of relaxed selection pressures in the new environment. This was referred to as the “founder-flush” model. In Carson’s model, then, the reduction in genetic variability is minimized as a result of rapid increase in population size immediately following the founder event, resulting in increased genetic variability. A paradox emerges under models of sexual selection on populations simultaneously affected by founder effects. In certain instances, if Carson is correct, the loss of variability due to founder effects can be mitigated. Sexual selection, under the classical models (e.g., runaway selection), reduces genetic variability, especially in males. Also, unless secondary sexual characters are linked to or enhance other components of fitness, these characters are energetically costly to produce and maintain in the population, and individuals possessing such traits would be strongly selected against by predation or other environmental pressures. Based on the results of mating studies of Hawaiian Drosophila species, Kenneth Kaneshiro proposed an alternative view of the sexual selection process on islands. The tremendous morphological diversity observed in the Hawaiian drosophilids, especially in the secondary sexual characters found in the males, gave a clue to early researchers that sexual selection may have played a significant role in the evolution and explosive speciation in this island fauna. Indeed, the complex courtship displays observed in the Hawaiian drosophilid fauna provided an opportunity to test some of the classical concepts of sexual selection. Kaneshiro observed that a frequent outcome of mate preference experiments among closely related species was that of asymmetrical sexual isolation between reciprocal crosses between two populations. That is, the females of, say derived population A, may readily accept the courtship displays of males from ancestral population B, resulting in successful matings. However, in the reciprocal combination, females of population B were less likely to mate with males of population A. Initially, Kaneshiro suggested that such shifts in the behavior of derived populations were the result of the severe drift conditions and the genetic revolution that accompanied founder events. Subsequently, it was suggested that the sexual selection process strongly influenced the shift in behavior in the most derived populations, which resulted in the observed asymmetrical sexual isolation between the two reciprocal crosses. It was suggested that during founder events, when population size is small, there is strong selection for females that are less choosy. Highly discriminant females are strongly selected against because they may never even encounter males that are able to satisfy their courtship

requirements. If the population bottleneck condition persists over a few generations, there will be a shift in the distribution of mating types in the population toward an increase in frequency of less choosy females. Such a shift in the mating distribution may result in a corresponding change in gene frequencies in the population, further resulting in the destabilization of the coadapted genetic systems that had evolved in the population as it adapted to the environmental conditions of the habitat in which it lives. The destabilized genetic environment of the population provides the opportunity for genetic change conducive to speciation. The breakup of coadapted gene complexes allows for novel genetic recombinants to be generated, some of which may be better adapted to the environmental conditions that led to the drastic reduction in population size. Especially when such novel genetic recombinants are linked to or correlated with the genotypes of less choosy females, it is easy to visualize how such preadaptive gene complexes can spread quickly in subsequent generations, providing an evolutionary mechanism by which populations can regenerate genetic variability that had been lost during reduced population size. Clearly, then, at least during the initial stages of colonization following the founder event, sexual selection, and especially the dynamics of sexual selection, may be playing an important role in providing a genetic environment that is most conducive to the formation of new species. In the same way, for African cichlid fish, as in the Hawaiian Drosophila, much of the sustained rapid speciation has also been explained by an interaction between drift and sexual selection in subdivided populations. SEXUAL SELECTION AND GENETIC VARIABILITY IN ISLAND POPULATIONS

The dynamics of the sexual selection process as described here indeed provide an evolutionary mechanism by which founder populations can, under small-population conditions, generate novel genetic recombinants that can be selected for adaptation to the new habitat. In fact, when viewed as a dynamic, frequency-dependent system, sexual selection can play a significant role in influencing, maintaining, and in some cases enhancing levels of genetic variability in natural populations. When the population is large and healthy, sexual selection may serve as a stabilizing force in maintaining a balanced, polymorphic mating system that maintains an appropriate level of genetic variability required to survive in a particular habitat. However, if the population undergoes a bottleneck due to some kind of environmental

SEXUAL SELECTION

827

stress, the shifts in the distribution of mating types as described above play an important role in replenishing reduced levels of genetic variability by the generation of novel recombinants (i.e., not by increasing the mutation rate). The shift toward an increased frequency of less choosy females and corresponding shift in gene frequencies toward the genotypes of the less choosy females result in the destabilized genetic environment and the breakup of coadapted gene complexes, allowing for the new recombinants to be generated. Those recombinants that are better adapted to the environmental stress conditions will be strongly selected and spread quickly throughout the population, especially if linked or correlated with the genotypes of the less choosy females. Thus, sexual selection can play an important role in mitigating the fragility of island populations that can be easily perturbed by changing environmental conditions. Sexual selection provides a mechanism by which genetic variability can be restored within relatively few generations following a bottleneck event. SEXUAL SELECTION IN CONJUNCTION WITH ECOLOGICAL ADAPTATIONS

In many well-known adaptive radiations, secondary sexual characters are remarkably diverse. These secondary characters provide no real adaptive benefit and yet are nearly ubiquitous. Sexual selection has been inferred to be a more potent force in adaptive radiations than previously thought. In cases such as the Hawaiian Drosophila, and more so Laupala, there is sometimes little ecological differentiation between the most closely related species. Sexual selection may actually increase the rate of diversification beyond that of ecological adaptation alone. Studies on cichlid fish show that while niche adaptation is an important component to this rapid radiation, sexual selection may be even more important in some cases. Intraspecific sexual selection may also increase the adaptive ability within these radiations. On islands in the West Indies, sexual dimorphism in Anolis lizards allows increased niche exploitation, as size differences allow for increases in the number of ecomorphs present in each species. This may speed up the already rapid filling of niche space for this group. On the Galápagos Islands, females of a few species of Darwin’s finches as well as those of marine iguanas were found to prefer larger males, a trait that is correlated to environmental exploitation in this case. Sexual selection may operate very quickly on an adaptive trait.

828

SEXUAL SELECTION

THE ROLE OF SEXUAL SELECTION IN CONSERVATION BIOLOGY

Although insular systems are often recognized for nature’s creativity and their high levels of endemism, they are also known for their fragility and the relatively large number of rare and endangered species. Sexual selection and recognition have been implicated directly in conservation issues in African cichlids. Here, mate recognition is critical to the dynamics of the system. However, eutrophication of the lakes has led to a breakdown of the recognition system, with resulting extensive hybridization. Clearly, understanding the nature of mate recognition and the biology of small populations is crucial for addressing conservation issues, since rare and threatened species are faced with extinction, primarily as a result of drastically reduced population size. These demographic conditions are no different from those occurring during early stages of colonization following a founder event. The sexual selection model described in this article permits the generation of novel genetic recombinants, some of which are better adapted to the new habitat or to the stress conditions that caused the population to decline. What is inferred here, then, is that if the habitat of those species that are faced with small population size can be sustained by the removal of the threats that impact on the habitat, these populations have the capacity to replenish the genetic variability that may have been lost due to drift. Furthermore, the sexual selection system may facilitate natural hybridization with related sympatric species, which can result in the introgression of genetic elements that are less susceptible to the environmental stress conditions that resulted in the population crash. Thus, when considering the management of rare and endangered species, it is important to consider the interaction with other related species with overlapping distributions, from which genetic material may be “leaking” across species barriers to maintain adequate levels of genetic variability in the population. CONCLUDING REMARKS

For some of the most diverse species found on islands, founder event speciation along with sexual selection is considered to be a likely model for the formation of new species and the explanation for the spectacular adaptive radiation among island biota. Most evolutionary biologists acknowledge the fact that natural selection can play a dominant role in directing the course of evolution within species. However, the recent research on sexual selection theory and its role in the evolutionary process suggest that the evolution of novelty is enhanced when populations are subjected to bottlenecks and that the dynam-

ics of sexual selection are extremely important, especially during the initial stages of species formation. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Cichlid Fish / Crickets / Drosophila / Founder Effects FURTHER READING

Carson, H. L. 1986. Sexual selection and speciation, in Evolutionary processes and theory. S. Karlin and E. Nevo, eds. London: Academic Press, 391–409. Darwin, C. 1871. The descent of man, and selection in relation to sex. New York: Penguin Putnam, Inc. Fisher, R. 1930. The genetical theory of natural selection. Oxford: Clarendon Press. Kaneshiro, K. Y. 1989. The dynamics of sexual selection and founder effects in species formation, in Genetics, speciation, and the founder principle. L. V. Giddings, K. Y. Kaneshiro, and W. W. Anderson, eds. New York: Oxford University Press, 279-296. Lande, R. 1981. Models of speciation by sexual selection on polygenic traits. Proceedings of the National Academy of Sciences of the United States of America 78: 3721–3725. Mayr, E. 1982. Processes of speciation of animals, in Mechanisms of Speciation. C. Barigozzi, ed. New York: Alan R. Liss, 1–19. Mendelson, T. C., and K. L. Shaw. 2005. Rapid speciation in an arthropod. Science 433: 375–376. O’Donald, P. 1980. Genetical models of sexual selection. Cambridge, UK: Cambridge University Press. Seehausen, O. 2004. Hybridization and adaptive radiation. Trends in Ecology and Evolution 19: 198–207. West-Eberhard, M. J. 1983. Sexual selection, social competition and speciation. Quarterly Review of Biology 58: 155–183.

FIGURE 1 Map of the Seychelles islands.

SEYCHELLES

The Republic of Seychelles comprises 115 islands spread over 1.3 million km2 of the western Indian Ocean. The majority of these are small and uninhabited. Of the Seychelles’ 80,000 human inhabitants, 90% live on the largest island, Mahé (153 km2), and 7% on the next largest, Praslin (28 km2). The islands can be divided into two groups: the northern islands (the granitic islands and the coral cays of Bird and Denis) and the southern coral islands (Fig. 1).

Gondwana over hundreds of millions of years finally ended with the separation of India and the Seychelles 65 million years ago. The resultant Seychelles microcontinent has gradually submerged and eroded, until today only islands representing the granitic microcontinent’s mountain tops remain. The islands are very rocky and steep, reaching a maximum height (at Morne Seychellois) of 905 m. Periods of volcanic activity and tectonic movement led to the formation of other islands, some granitic (Silhouette and North) and others volcanic. The latter have long since subsided and eroded, leaving only raised coral reefs at or near the surface. These are now represented by coral cays and atolls. These coral islands rise no more than 8 m above sea level, and most are less than 1-m high (Fig. 2).

GEOGRAPHY AND GEOLOGY

CLIMATE

The granitic islands are mostly composed of Precambrian granite (750 million years old) and represent fragments of the ancient supercontinent of Gondwana. The breakup of

The islands have an equatorial climate, with relatively small variations in temperature (24–32 °C at sea level) and constantly high humidity (over 70%). The low-lying

JUSTIN GERLACH Nature Protection Trust of Seychelles, Cambridge, United Kingdom

S E YC H E L L E S

829

FIGURE 3 Silhouette, the most natural of the high granitic islands.

FIGURE 2 Cosmoledo, a low-lying coral island.

coral islands have lower rainfall and, consequently, lower humidity. On the high islands, there may be significant local variation in climate, with high forest areas being cooler (18–24 °C) and wetter than lowlands. BIODIVERSITY AND ECOLOGY

The islands have a diverse fauna and flora and are well known for remarkable species such as giant tortoises (Dipsochelys species) and the coco-de-mer palm, or double coconut, Lodoicea maldivica. Two distinct species assemblages are found, corresponding to the origin of the islands. The coral islands, which represent land raised up from under the sea, have been colonized by organisms that have crossed large expanses of ocean, some from Asia to the north and east, some from Africa to the west, but most from the closest land mass, Madagascar, to the south. These small, flat islands have relatively few species, and almost all of those they do have are widespread outside of the Seychelles. The exception is Aldabra atoll, which supports some 1500 species, 60% of which are believed to be endemic to the atoll. The high granitic islands of the Seychelles support a diverse flora and fauna with many interesting affinities to both Asia and Madagascar (Fig. 3). The plants include some 300 species, of which 40% are endemic. These include the only member of the Dipterocarpaceae found outside of Southeast Asia (Vateriopsis seychellarum) and a whole family restricted to the islands (the monotypic family Medusagynaceae, represented by Medusagyne oppositifolia). There are populations of an endemic carnivorous pitcher plant, Nepenthes pervillei (a genus restricted to Southeast Asia, the Seychelles, and Madagascar) (Fig. 4). A major component of the natural vegetation is the palm family; in addition to the widespread tropical coconut

830

S E YC H E L L E S

FIGURE 4 The Seychelles pitcher plant Nepenthes pervillei.

Cocos nucifera, there are six endemic genera of palm: the coco-de-mer, Deckenia nobilis, Nephrosperma vanhouetteana, Phoenicophorium borsigianum, Roscheria melanochaetes, and Verschaffeltia splendida. Most of these palms have prominent spines on their stems and leaf bases, a feature generally thought to be a defense against the giant tortoises that used to roam free in the islands. Animal life is mainly small and inconspicuous, with a large proportion of the fauna being specialized to live in high forest leaf litter or in the narrow spaces between palm leaves. Unusual animal groups include a wide range of carnivorous snails, specialized carrion-feeding caddis flies, and land-dwelling diving beetles. More conspicuous are the birds, reptiles, and amphibians. The Seychelles are well known as a bird-watching destination because, although they support only a small number of species, most of these are endemic, such as the Seychelles magpie robin Copsychus seychellarum, the Seychelles black paradise flycatcher Terpsiphone corvina, and the Seychelles kestrel Falco araea. All of the endemic birds are close relatives of species found in Madagascar. The reptiles are far more diverse and are the dominant vertebrates of the islands. Most notable are the giant tortoises, which were historically the largest herbivores in the islands. Four species have been described from the Seychelles, of which the

Aldabra tortoise Dipsochelys dussumieri is still abundant in the wild, Daudin’s tortoise Dipsochelys daudinii is extinct (known from only two specimens), and the Seychelles giant tortoise D. hololissa and Arnold’s tortoise D. arnoldi were thought to be extinct. Recently (Fig. 5), individuals thought to be the few survivors of these latter two species were found and are being bred in captivity. Arnold’s tortoises were reintroduced to the wild on Silhouette Island in 2006. The islands also support declining populations of Pelusios terrapins and large breeding populations of the hawksbill turtle Eretmochelys imbricata and the green turtle Chelonia mydas. Other reptiles comprise the little-known chameleon Calumma tigris, geckos, skinks, and snakes. The geckos are easily visible, with abundant Phelsuma day geckos in coastal habitats and introduced house geckos (Gehyra mutilata and Hemidactylus frenatus) in buildings. There are also more specialized species, such as the Ailuronyx bronze geckos in palm woodlands and the strange sucker-tailed gecko Urocotyledon inexpectata, which lives in caves, often occupying potter wasp nests.

FIGURE 5 Arnold’s tortoise Dipsochelys arnoldi.

Amphibians are among the most important of groups from a biogeographical perspective, as they tend to reflect ancient land connections because their permeable skins make it very difficult for them to cross open sea. The Seychelles support four families of amphibian with differing biogeographical affinities. The most widespread is Ranidae, represented by the Mascarene frog Ptychadena mascariensis. This species was probably introduced from Mauritius. The Hyperoliidae is represented by the Seychelles tree frog Tachycnemis seychellensis, which is closely related to Malagasy species and is thought to be descended from frogs that were carried to the islands on rafts of vegetation. The last frog family in the islands is the Sooglossidae, an endemic family related to an obscure Indian species. This is an ancient family of very specialized frogs,

FIGURE 6 The minute Gardiner’s frog Sooglossus gardineri.

none of which have free-living tadpoles. The mountain streams of the Seychelles are seasonal and very fast flowing, so frogs avoid living in the water and become terrestrial or arboreal instead, either carrying their tadpoles or developing their young entirely in the eggs. This latter strategy is adopted by Gardiner’s frog Sooglossus gardineri, which is one of the smallest frogs in the world (adult size 8–11 mm), the eggs of which hatch into frogs 3-mm long (Fig. 6). Strangest of all the Seychelles amphibians are the caecilians (Caeciliidae). The Seychelles have six species of these unusual and very poorly known legless amphibians. This is another group that originated on the ancient continent of Gondwana. Almost all these amphibians form part of an adaptive radiation, with ten of the endemic species being found in five endemic genera and only one being an apparently recent arrival. Insect diversity is high, with approximately 6,000 species recorded. A high proportion of these are endemic: between 50% (Lepidoptera and Diptera) and 80% (Orthoptera). Although mollusc diversity is lower (only 80 species), 88% are endemic. These include remnants of the ancient Gondwana fauna, for example the large Stylodonta snails of the sothern African, Madagascan, Sri Lankan, and Australian family Acavidae and the Pachnodus tree snails related to species from the Congo. Pachnodus is a typical island radiation, having diverged into 12 distinct species through a combination of geographical isolation on different islands and habitat specialization. A similar radiation is found in the carnivorous snail family Strepaxidae, where ten species form an endemic radiation of six distinct genera. The remaining three species are also endemics but belong to African and Asian genera.

S E YC H E L L E S

831

The islands are fringed by coral reefs, which are diverse and productive environments. Over 1200 marine fish have been recorded (compared to only eight freshwater fish). Despite several decades of interest in the marine biology of the region, very little is known of the invertebrate communities of the reefs of Seychelles. These reefs were very badly damaged by the coral bleaching event of 1998, which resulted in the death of over 80% of corals. The corals are recovering in some areas, although the process is extremely slow. HUMAN HISTORY AND DEVELOPMENT

The Seychelles islands were uninhabited when discovered by European explorers in the fifteenth and seventeenth centuries. There have been suggestions that Melanesian sailors settled the islands on their way to colonize Madagascar, but there is no evidence of this. Similarly, suggestions that Arab traders occupied Aldabra atoll remain speculative. The first recorded sighting of the islands was by Vasco da Gama’s expedition of 1498, but the first landing did not occur until 1609 (on North Island). The first settlement of the islands was the French colonization of St. Anne in 1770. This led to the settlement of most of the granitic islands by the mid-1800s, originally by French plantation owners and their slaves. Britain took possession of the islands in 1810 but did little to change the organization of the colony beyond using the islands as a place to release slaves from captured French slave traders. Small numbers of Indian and Chinese merchants also settled the islands at this time. These people make up the ancestry of the racially mixed population of the present day. In the nineteenth century, most of the larger islands were settled and the lowland forests cleared for agriculture, especially for coconut plantations. On Mahé Island, large plantations of cinnamon Cinnamomum verum were established, and further deforestation occurred as wood was cut to supply cinnamon oil distilleries. Human impacts on the environment were high at this time, with the exploitation of giant tortoises and turtles as food sources and the deliberate extermination of the crocodile Crocodylus porosus and the Seychelles green parakeet Psittacula wardi, the latter of which was perceived as a pest of fruit plantations. By 1900 all the islands were extensively modified, with the exception of Aldabra atoll, which remained inhospitable and difficult to reach. A small settlement had been established on the atoll, but the human impact remained minimal. Economically the islands relied on agriculture throughout the nineteenth century and well into the

832

S E YC H E L L E S

twentieth century. Copra from the coconut plantations was the main export, with relatively small levels of production of whale oil, cinnamon oil, vanilla, turtle meat, seabird eggs, and fish. Periodic attempts were made to increase the value of fish exports, but this remained insignificant until the arrival of large international tuna fishing fleets in the Indian Ocean in the 1980s. Collapse of the market for copra and cinnamon oil in the mid-1900s left the Seychelles economically dependent on other sources of income, particularly tourism after the opening of the international airport in 1971. Tourism dominates the economy of the islands today, with some 120,000 visitors a year. Infrastructure development has increased dramatically in recent years, and now most coastal properties are built upon, with plans for the development of 60 new hotels in the next ten years. There are concerns that the planned rate of development is unsustainable. From 1810 the islands were governed by the British administration in Mauritius, not receiving their own governor until 1903. In 1976 the islands became an independent republic. The first president, James Mancham, led a coalition government headed by the Democratic Party for one year, after which point he was deposed by a military coup, in 1977. Power was assumed by the Seychelles People’s Progressive Front, led by Albert René, under a one-party state system until 1993. In that year, international pressure led to the adoption of a new democratic constitution. René was returned to office twice, in 1998 and 2001, and then resigned in favor of his vice president, James Michel, in 2004. Michel won his own mandate in 2006. The SPPF has dominated Seychelles politics since 1977 and retains 68% of the seats in the National Assembly, although it held only 54% of the presidential vote in the 2006 election. CONSERVATION

The islands were heavily exploited from their settlement in 1772 until the mid-1900s. Lowland forests were cleared, and all usable timber trees felled; large animals were eaten or traded. Despite this, there have been few recorded extinctions from those times: Those species that did go extinct are the crocodiles, the green parakeet, chestnutflanked white-eye Zosterops semiflava, the giant tortoise Dipsochelys daudinii, the terrapin Pelusios seychellensis, and the snails Pachnodus ladiguensis and P. curiosus. Although one of the first administrators of the islands was concerned over the rate of forest loss and established a park for the tortoises in 1787, potential problems received little attention until 1874, when eminent British scientists, led

by Charles Darwin, appealed to the governor of Mauritius to take measures to protect the declining Aldabra tortoise population. This action prompted the protection of Aldabra, and from 1891 commercial users of the atoll were required to protect the tortoises. The tortoise population on the atoll probably numbered fewer than 1000 animals in 1900 but recovered rapidly, to some 120,000 in 1974. Today, the giant tortoise is secure from exploitation, and steps are being taken to conserve the remnant populations of granitic island tortoises. Although parts of Aldabra were effectively managed as a reserve from 1955, the atoll did not receive formal protection until 1979. Other reserves were established in the late 1970s, primarily to protect water catchment areas. Today, 43% of the land area of the islands is legally protected, as are a further 228 km2 of marine habitat. These areas are managed mainly by the Seychelles government, although the most significant reserves are managed by non-governmental organizations. Management of these reserves has largely focused on conserving threatened bird populations, particularly through translocation. This has led to population recovery for the Seychelles magpie robin, the Seychelles warbler Acrocephalus sechellensis, and the Seychelles white-eye Zosterops modesta. Management of other species based on scientific research is a more recent development. This is largely at the stage of population assessment for reptiles, amphibians, invertebrates, and most plants. Science-based population management is now in place for the critically endangered Seychelles sheath-tailed bat Coleura seychellensis and for plant species such as Impatiens gordonii. The main threats to be biodiversity of the Seychelles islands have changed over the years. In the early years of human settlement, habitat loss and exploitation were the main threats, but for much of the twentieth century, invasive species have been considered the primary problem. These include predators such as cats and rats, as well as invertebrates such as the crazy ant Anoplolepis gracilipes, which affect some species. Far more widespread in effects have been the invasive plants, especially cinnamon, Chinese guava Psidium cattleianum, and Clidemia hirta. Since 2000 there has been a great expansion of development for housing and for the tourism industry. This is leading to new developments in coastal areas, and potentially to the construction of roads to remote areas. Such roads would open up new areas to invasive species and to exploitation and would also directly impact some of the few remaining fragments of primary habitat. In the future, there will be additional pressures from climate change. For small islands, sea-level rise is an

important issue. Some of the coral islands will be badly affected by any rise in sea level, although the main islands are relatively high. Despite this, the impacts of sea-level rise may be notable because most of the human population lives in coastal areas at risk from flooding. Changes in climate are expected to make the weather patterns less stable, with an increase in extreme storms on the granitic islands and a reduction in rainfall on the coral islands. Increased length and frequency of dry periods has already caused the extinction of one species, with the disappearance of the Aldabra banded snail Rachistia aldabrae in 1997. SEE ALSO THE FOLLOWING ARTICLES

Frogs / Granitic Islands / Lizard Radiations / Madagascar / Tortoises FURTHER READING

Amin M., D. Willetts, and A. Skerrett, eds. 1994. Aldabra—world heritage site. Nairobi: Camerapix Publishers International. Gerlach, J. 2004. Giant tortoises of the Indian Ocean. Frankfurt: Edition Chimaira. Gerlach, J., ed. 2007. Terrestrial and freshwater vertebrates of Seychelles. Leiden, Netherlands: Backhuys Publishers. McAteer, W. 1991. Rivals in eden. London: The Book Guild.

SHIPWRECKS JAMES HAYWARD University of California, Berkeley

Island life and exploration have always necessitated the use of ships and other vessels, be it for transportation, for exploration, for commerce, or for recreation. From rafts and canoes, to sailing ships and steamers, to modern cruise ships and tankers, there have always been some of these vessels that failed to make it safely to their destination. Every shipwreck, whether one unnamed and unknown or an icon in world history such as the Titanic, has the potential to make an impact on island life. While each shipwreck may have a variety of consequences, the most common and most influential include cultural, political, or biological impacts. ANATOMY OF A SHIPWRECK

Shipwrecks occur for a variety of reasons, inclement weather and navigation errors being among the most common. Some ships were intentionally grounded, including the famous HMS Bounty, which was grounded

SHIPWRECKS

833

on Pitcairn Island, stripped, and then burned in order to avoid sighting by British ships, which might have then arrested the mutineers. In October 1942, after striking two mines entering the Espiritu Santo Harbor in Vanuatu, the SS President Coolidge, a troop transport ship, was driven aground in order to enable all but two of the 5440 men aboard to abandon ship safely, before she settled back into the harbor waters. What we know about shipwrecks depends primarily on their causes and where they end up. Many ships break up rapidly in the powerful and dynamic environments that lead to their sinking. Some ships may settle mostly intact to the bottom, only to be covered up by sediment over many decades. In some cases rapid sedimentation can provide anoxic conditions where ship, crew, and cargo can be preserved for hundreds of years. Shipwrecks in deep freshwater, such as the Great Lakes, have been found in relatively pristine condition. CULTURAL IMPACTS

When a shipwreck occurs, the survivors who are able to find a safe harbor can bring cultural changes with them. In 1841 the slave ship Trouvador was wrecked off the Turks and Caicos in the Caribbean. While a disaster for those who died, the wreck was a mixed blessing for the slaves on board, who found themselves in an emancipated British colony instead of the slave market in Cuba, their original destination. Ironically, though free, the majority of the Africans were required to work for a year in the island salt ponds to pay for their rescue. On the other side of the world, birth rates are known to have spiked on Rapa Nui (Easter Island) after several shipwrecks in the 1800s. In some cases, cultural differences are a disadvantage to the shipwreck survivors. In 1840 the Maria was wrecked in Western Australia. After making it safely to shore, several of the crew attempted to walk back to civilization with the help of the local Aboriginals, who reportedly killed them for refusing to respect the Aboriginal culture and repeatedly harassing the Aboriginal women. POLITICAL AND FINANCIAL IMPACTS

In some cases, the greatest impact a shipwreck may have may not be felt locally or might not be appreciated for years to come. Such is the case of the Dunotter Castle, a British vessel en route from Australia to California with a load of coal in 1886. After she ran aground on uninhabited Kure Atoll in the northern Hawaiian Islands, there was nearly an international incident when British government representatives in Hawaii launched a rescue mission that

834

SHIPWRECKS

could have paved the way to claiming the island for Great Britain. A shipwreck that did ultimately change the political future of an island was the wreck of the Sea Venture in Bermuda in 1609. While leading a fleet with 600 settlers and supplies to the foundering Jamestown settlement, the Sea Venture wrecked on the coast of Bermuda during a hurricane. Although many of the survivors were able to make their way to Jamestown, a small band remained on the island until 1612, when the first ship of intended settlers arrived from England. BIOLOGICAL IMPACTS

Shipwrecks can have long-term impacts on island biology. In 1780 a Japanese shipwreck delivered Norway rats, a very destructive species, to islands off the coast of Alaska. These rats are still a problem today, feeding on the eggs of nesting sea birds. In more modern times, shipwrecks have included oil tankers such as the Exxon Valdez, which coated 2100 km of shoreline in oil after running aground on Bligh Reef, between Bligh Island and Glacier Island in the Prince William Sound area of Alaska. Fortunately, in some cases, shipwrecks may actually have a positive impact on the environment. As discussed below, the presence of shipwrecks in some environments can become the basis for an entire ecosystem. Much like a volcanic island breaking the surface of the sea for the first time, the sudden arrival of a shipwreck on the ocean floor will often lead to colonization and productivity. In some situations, the initial impact of a shipwreck may be negative, but over a number of years, decades or centuries, the shipwreck environment may recover and eventually surpass the original. SHIPWRECKS AS ISLANDS

When a vessel comes to rest on a sandy sea floor, it may offer a new fixed substrate that allows for coral settlement, sea grass bed development, or even kelp forest formation. These habitats then attract additional species of invertebrates and fish until a thriving ecosystem has taken hold in what may have been a previously barren environment (Fig. 1). Chuuk lagoon in Micronesia, the site of two major battles between the United States and Japan during World War II, now holds the remains of over 50 vessels and many more aircraft, which provide homes to a highly diverse marine ecosystem. On all coasts of the United States, and in many locations around the world, there are very successful examples of decommissioned vessels being purposely sunk to provide recreational opportunities for SCUBA divers and anglers, while giving the local marine species extra reef space to call home.

the Hawaiian silversword alliance (31 species in Argyroxiphium, Dubautia, and Wilkesia). True silverswords and greenswords (Argyroxiphium; five species), named for their narrow, sword-shaped leaves (silvery-hairy in silverswords), are spectacular rosette plants of alpine cinder slopes, forest edges, mesic scrub, and bogs on Maui and Hawai‘i (Fig. 1). These famous, young-island endemics are closely related to the bizarre, fibrous-leaved rosette plants in Wilkesia (two species), found on generally dry or exposed slopes of western Kaua‘i, and to trees, shrubs, mat-plants, cushionplants, and lianas in Dubautia (24 species), found in wet, dry, and bog habitats across the six major high islands of the Hawaiian archipelago (Fig. 2). The silversword alliance is an extreme, well-documented example of insular adaptive radiation following long-distance dispersal in plants. ADAPTIVE RADIATION OF THE HAWAIIAN SILVERSWORD ALLIANCE FIGURE 1 Marine life takes over the hull of the Benwood, in the Florida

Keys.

SEE ALSO THE FOLLOWING ARTICLES

Archaeology / Easter Island / Introduced Species / Reef Ecology and Conservation FURTHER READING

Advisory Council on Underwater Archaeology. http://www.acuaonline.org. Australasian Institute for Maritime Archaeology. http://www.aima.iinet .net.au. Ballard, R. D., L. E. Stager, D. Master, D. Yoerger, D. Mindell, L. Whitcomb, H. Singh, and D. Piechota. 2002. Iron Age shipwrecks in deep water off Ashkelon, Israel. American Journal of Archaeology 106: 151–168. Bass, G. F. 1975. Archaeology beneath the sea. New York: Walker Publishing Co. Inc. Bathurst, B. 2005. The wreckers. New York: HarperCollins. Jurisic, M. 2000. Ancient shipwrecks of the Adriatic: maritime transport during the first and second centuries AD. Oxford: Archaeopress.

The silversword alliance exemplifies adaptive radiation by exhibiting a high rate of diversification accompanied by major ecological shifts and associated morphological change. Unlike some (but not most) other prominent plant and animal examples of adaptive radiation in the Hawaiian Islands, such as the lobelioids and drosophilids, the silversword alliance radiation is evidently contemporary with the modern high islands and does not appear to date back to islands further northwest that have been reduced by erosion and subsidence to atolls, islets, or submarine seamounts. The estimated maximum age of the most recent common

SICILY SEE MEDITERRANEAN REGION

SILVERSWORDS BRUCE G. BALDWIN University of California, Berkeley FIGURE 1 Rosette plants of the Hawaiian silversword alliance. (A) The

Haleakala silversword (Argyroxiphium sandwicense subsp. macroceph-

Silverswords belong to an ecologically diverse lineage of endemic Hawaiian woody and semi-woody members of the sunflower family (Compositae) collectively known as

alum), a semelparous, thick-leaved plant endemic to dry, alpine cinder on Haleakala, East Maui. Photograph by Donald W. Kyhos. (B) The iliau (Wilkesia gymnoxiphium), a semelparous fibrous-leaved plant endemic to dry or open sites on western Kaua‘i. Photograph by Gerald D. Carr.

S I LV E R S W O R D S

835

ancestor of the silversword alliance based on molecular data (5.2 ± 0.8 million years) approximates the age of the oldest high Hawaiian island, Kaua‘i (5.1 ± 0.2 million years). The resulting minimum diversification rate estimated for the silversword alliance (0.56 ± 0.17 species per million years) falls within the range of rates estimated for other prominent examples of adaptive radiation in plants and animals. Major ecological change associated with diversification is evident in part from comparative analysis of habitats and life forms within lineages of the silversword alliance endemic to particular islands of the Hawaiian chain. On the oldest high island, Kaua‘i, for example, one endemic lineage includes rosette plants of mostly dry, open sites (Wilkesia), sprawling shrubs of wet forests (Dubautia raillardioides), and erect, relatively compact shrubs of open bogs (D. paleata). Similarly, lineages of Dubautia section Railliardia, restricted to younger islands or volcanoes (east of Kaua‘i), generally differ markedly in habitat (e.g., moisture availability) and often in life form; for example, an endemic Maui Nui lineage includes a wet-forest tree (D. reticulata), up to 8-m tall, and a shrub of dry, open scrub or barrens (D. menziesii) that occur on a single volcano, Haleakala. Studies of plant structure and function by S. Carlquist and by R. H. Robichaux and colleagues have shown that differences in leaf and wood anatomical traits among species of the silversword alliance are generally consistent with expectations based on water availability in the habitats occupied, except in species from bogs, wherein features otherwise associated with dry habitats may reflect either physiological drought (compromised root function in bogs) or retention of characteristics from ancestors of dry environments. For example, leaves of species from dry habitats, such as Dubautia menziesii, often contain copious hydrophilic extracellular “mucilage,” which confers resistance to wilting under drought stress; leaves from species of wet habitats, such as the closely related D. reticulata, contain lower or undetected amounts of these polysaccharides and wilt at levels of drought stress that are tolerated by species of dry situations. HISTORICAL BIOGEOGRAPHY AND ECOLOGY

Molecular phylogenetic data indicate that the silversword alliance in general follows the “progression rule” FIGURE 2 Ecological diversity in Dubautia. (A) D. waialealae, a cushion

plant from bogs of Kaua‘i. Photograph by Kenneth Wood. (B) D. latifolia, a liana or woody vine from mesic forests of Kaua‘i. Photograph by Bruce G. Baldwin. (C) D. scabra subsp. scabra, an often mat-forming, soft shrub from young lava of Hawai‘i and East Maui. Photograph by Bruce G. Baldwin. (D) D. menziesii, a shrub from high, dry cinder barrens and scrub of East Maui. Photograph by Susan J. Bainbridge.

836

S I LV E R S W O R D S

of Hawaiian biogeography; namely, dispersal evidently has been mostly from older to younger islands (northwest to southeast). Dispersal between islands has been uncommon; all but five species are known only from one island. In the most widespread genus, Dubautia, all species endemic to one island can be explained by diversification on that island from a single founder species. Most single-island endemic species of the silversword alliance (14 of 25) occur on the oldest modern high island, Kaua‘i, where Dubautia and Wilkesia evidently originated, based on molecular phylogenetic data and levels of genetic variation within and among species there compared to such variation in the younger-island lineages. Absence of true silverswords and greenswords (Argyroxiphium) from Kaua‘i and O‘ahu has been puzzling because, based on molecular analyses, divergence of the Argyroxiphium lineage preceded the origin of Maui Nui and Hawai‘i, where the true swords are endemic. Prehistoric extinction of Argyroxiphium on islands older than Maui Nui is consistent with loss of highelevation habitat on those islands through erosion and subsidence and the predominant occurrence of most species of Argyroxiphium at elevations that exceed those of the highest summits of the older islands. HYBRIDIZATION AND GENE FLOW

Experimental hybridization studies by G. D. Carr and D. W. Kyhos have shown that crosses between species of the silversword alliance reliably yield vigorous hybrids of low to high fertility, even when made between members of the most evolutionarily divergent lineages. Documentation of natural hybrids representing 41 different species combinations from various environments throughout the archipelago indicates that prezygotic reproductive isolation has not reached a level that is sufficient to prevent gene flow between species. Chromosomal structural mutations (specifically, one to three whole-arm reciprocal translocations) account for reduction in interfertility between some species of the silversword alliance. These chromosomal rearrangements alone do not result in strong post-zygotic reproductive barriers between species; experimental studies have shown that even the least interfertile members of the silversword alliance are capable of gene exchange under ecologically permissive conditions. For example, Carr found that natural hybrids between the closely sympatric Haleakala silversword (Argyroxiphium sandwicense subsp. macrocephalum; Fig. 1) and Haleakala kupaoa (Dubautia menziesii; Fig. 2) produce progeny (~5% seed set) by backcrossing to either of the parent species, despite highly reduced fertility (~10%) of first-generation hybrids from extensive chromosomal structural heterozygosity (the parent species differ by three

whole-arm chromosomal interchanges). Resulting backcross plants sometimes reach maturity in nature and can be highly fertile and vigorous. One additional generation of backcrossing to the same species can yield fully fertile plants that could be mistaken for that species. In summary, hybridization and gene flow remain possible between sympatric members of the silversword alliance and appear to be limited in part by post-dispersal selection against hybrid or backcross phenotypes in various natural settings (i.e., extrinsic or environment-mediated post-zygotic reproductive isolation). Differences in peak flowering time between some species (e.g., Dubautia paleata and D. raillardioides) may also contribute to reproductive isolation. The possibility that hybridization may play an important evolutionary role in the silversword alliance must be taken seriously given the likely importance of environmental factors in restricting gene flow or recombination between species of the group. During periods of major environmental change, selection favoring hybrid or backcross phenotypes could drive evolution of new, stable lineages. Phylogenetic evidence for a lasting impact of hybridization on the genetic constitution of a limited number of species in the silversword alliance has been obtained from comparison of phylogenetic data from different genes and genomes. Conversely, some completely interfertile species that occur in proximity and are known to hybridize in nature do not show convincing molecular evidence of gene exchange. ORIGIN OF THE SILVERSWORD ALLIANCE

The silversword alliance belongs to the helianthoid subtribe Madiinae, a monophyletic group that also includes ~90 species of mostly annual herbs commonly known as tarweeds (so named because of their sticky glandular exudates). All species of continental tarweeds occur in western North America, and all but two are native to the California Floristic Province, with one (Madia sativa) also indigenous to southern South America (Chile and Argentina). Multiple lines of phylogenetic data indicate that the silversword alliance belongs to one of four major lineages of the tarweed subtribe, the “Madia” lineage (including Anisocarpus, Carlquistia, Harmonia, Hemizonella, Jensia, Kyhosia, and Madia, in addition to the three Hawaiian genera), and arose well after the onset of tarweed diversification in western North America. Although tarweeds and silverswords are strikingly different in general appearance, genetic similarity between the continental and Hawaiian members of the Madia lineage remains sufficient to allow for production of vigorous (though highly sterile) tarweed × silversword-alliance hybrids. Such successful crosses have been made between arborescent or shrubby species of Dubautia,

S I LV E R S W O R D S

837

on one hand, and perennial herbaceous species or hybrids of Anisocarpus, Carlquistia, and/or Kyhosia, on the other. Multiple lines of evidence for descent of the silversword alliance from western North American tarweeds indicate that the ancestor of the Hawaiian group must have dispersed across more than 3500 km of open ocean to reach the Hawaiian archipelago. No land bridge or intervening islands would have allowed for shorter dispersal intervals. External bird dispersal appears to be the most likely means by which the ancestor of the silversword alliance reached the Hawaiian Islands; the dry, single-seeded fruits of tarweeds of the Madia lineage and silverswords have features such as sticky enveloping bracts or persistent bristles (pappus) and fruit hairs that promote adhesion to animals. Birds have also been inferred to be the probable dispersal vectors for other Hawaiian plants of temperate North American origin, such as the endemic Hawaiian sanicles (Sanicula) and mints (Haplostachys, Phyllostegia, and Stenogyne). Understanding establishment of the original colonizing ancestor of the silversword alliance in the Hawaiian Islands requires consideration of sporophytic self-incompatibility (SI), which is inferred to have been ancestral in the group based on its patterns of occurrence among continental and Hawaiian members of the Madia lineage and the theoretical difficulty of reconstituting SI once lost. Inability of self-incompatible plants to set seed from self-pollination may mean that multiple seeds of the original colonist were dispersed together. Original introduction of multiple individuals can be inferred for various other indigenous plants (and animals) of remote oceanic islands, such as ancestrally dioecious species, which bear only pollen or ovules (not both) on individual plants. Among plants of the most remote oceanic islands, the silversword alliance has been regarded as a rare exception to Baker’s Rule (that colonizing plant species tend to be self-compatible), although examples of SI plants on remote oceanic islands continue to be documented. SI is absent and presumed to have been lost in some members of the silversword alliance (e.g., Dubautia scabra) that pioneer new habitats produced by recent lava flows, in keeping with Baker’s Rule. Enforced outcrossing (from SI) and ancestral allopolyploidy (= hybrid polyploidy) of the original colonist species from North America may have helped to counter potentially deleterious consequences of excessive inbreeding and contribute to adaptive radiation of the silversword alliance. Although no polyploids with the same genomic constitution as the Hawaiian species are known among the continental tarweeds, both continental lineages that contributed a genome to the tetraploid ancestor of the silversword alliance are still extant. Vigorous hybrids have been produced between those

838

S I LV E R S W O R D S

two continental tarweed lineages (i.e., between Anisocarpus scabridus and Carlquistia muirii), and the resulting plants produced some viable, diploid pollen—the raw material for producing allopolyploid progeny—thereby verifying the biological potential for allopolyploidization between the inferred ancestral diploid lineages. Other inferences about ancestral characteristics of the silversword alliance are aided by comparison of the group to the continental tarweed members of the Madia lineage. For example, all of the continental species in the Madia lineage are herbaceous, unlike the woody or semi-woody members of the Hawaiian group; evidently, a switch from herbaceousness to woodiness and an associated increase in plant stature accompanied colonization of the islands, as has been often inferred for island plants. Invasive non-native plants and animals, habitat loss, and other human-related factors endanger diversity in the silversword alliance. Feral mammals, such as mouflon, goats, and pigs, consume or uproot plants and have been especially destructive to true silverswords and greenswords. The Haleakala greensword, Argyroxiphium virescens, was evidently driven to extinction by such impacts. Extensive fencing of habitats to exclude ungulates and an intensive, genetically informed breeding and outplanting program have been highly effective at reversing population declines of silverswords. An emerging threat is destruction of the main pollinators of silverswords—native yellow-faced bees (Hylaeus)—by introduced Argentine ants (Linepithema humile). Strategies similar to those used for silversword recovery, in addition to aggressive efforts to control invasive plants, may be necessary to prevent extinction of endangered species of Dubautia and Wilkesia as well, some of which are known from fewer than 50 surviving individuals. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Dispersal / Hawaiian Islands, Biology / Inbreeding / Invasion Biology / Pigs and Goats FURTHER READING

Baldwin, B. G. 2006. Contrasting patterns and processes of evolutionary change in the tarweed-silversword lineage: revisiting Clausen, Keck, and Hiesey’s findings. Annals of the Missouri Botanical Garden 93: 64–93. Carlquist, S., B. G. Baldwin, and G. D. Carr, eds. 2003. Tarweeds & silverswords: evolution of the Madiinae (Asteraceae). St. Louis: Missouri Botanical Garden Press. Carr, G. D. 1985. Monograph of the Hawaiian Madiinae (Asteraceae): Argyroxiphium, Dubautia, and Wilkesia. Allertonia 4: 1–123. Forsyth, S. A. 2003. Density-dependent seed set in the Haleakala silversword: evidence for an Allee effect. Oecologia (Berlin) 136: 551–557. Friar, E. A., L. M. Prince, E. H. Roalson, M. E. McGlaughlin, J. M. Cruse-Sanders, S. J. DeGroot, and J. M. Porter. 2006. Ecological speciation in the East Maui-endemic Dubautia (Asteraceae) species. Evolution 60: 1777–1792.

Lawton-Rauh, A., R. H. Robichaux, and M. D. Purugganan. 2003. Patterns of nucleotide variation in homoeologous regulatory genes in the allotetraploid Hawaiian silversword alliance (Asteraceae). Molecular Ecology 12: 1301–1313. Purugganan, M. D., and R. H. Robichaux. 2005. Adaptive radiation and regulatory gene evolution in the Hawaiian silversword alliance (Asteraceae). Annals of the Missouri Botanical Garden 92: 28–35. Robichaux, R. H., G. D. Carr, M. Liebman, and R. W. Pearcy. 1990. Adaptive radiation of the Hawaiian silversword alliance (Compositae—Madiinae): ecological, morphological, and physiological diversity. Annals of the Missouri Botanical Garden 77: 64–72.

SKY ISLANDS JOHN E. MCCORMACK, HUATENG HUANG, AND L. LACEY KNOWLES

for investigating how different evolutionary processes such as natural selection and genetic drift lead to species formation. BIOGEOGRAPHY OF SKY ISLANDS

The term “sky islands” was originally coined by Weldon Heald in his writings on the mountains of southeastern Arizona, which harbor isolated montane forest or woodland surrounded by a sea of desert (Fig. 2). Analogous to the water of oceanic archipelagoes, low-elevation habitat acts as a barrier to dispersal on sky islands, facilitating divergence of isolated populations. The general concept of sky islands has been expanded to include a variety of settings in which high-elevation habitats are separated by inhospitable lowlands, such as plateaus, ridges, páramos, and alpine meadows.

University of Michigan, Ann Arbor

Sky islands are high-elevation habitats that are geographically subdivided and isolated among different mountain ranges (Fig. 1). Because of differences in climatic history and dispersal dynamics, the ecological and evolutionary processes and patterns characterizing sky islands may not always parallel those for traditional oceanic archipelago systems. Nevertheless, like their oceanic counterparts, sky islands are generators of diversity over multiple spatial and temporal scales and offer considerable potential

FIGURE 1 Map of the sky islands from the Rocky Mountains of north-

western North America. The distribution of the most likely habitat for montane grasshoppers in the genus Melanoplus is shown in white, as determined by ecological niche-modeling based on sampled populations (blue circles).

FIGURE 2 Schematic of the altitudinal distribution of habitats and

examples of the evolutionary forces acting within and between typical Madrean sky islands of the southwestern United States and northern Mexico.

There are roughly 20 sky island complexes in the world (Table 1), each with a distinct origin, spatial arrangement, age, and climate history. As with oceanic archipelagoes, the geologic processes forming sky islands, which include volcanic activity, erosion, and mountain uplift, provide a temporal reference for determining the age of the islands. These differences have important consequences for evolution in sky islands. For example, ancient sky islands, such as those in the Pantepui region of Brazil, Guyana, and Venezuela, might be expected to show higher levels of divergence and species diversity than would younger sky islands. The Pantepui region formed through erosion of a formerly more widespread plateau that covered a large portion of the Guiana Shield. Rivers gouge this plateau, creating isolated, high-elevation mesas (tepuis), which harbor cool, humid subtropical forests among the surrounding tropical lowlands. Many of the endemic species and subspecies found in this region are thought to have evolved in situ (i.e., they are not immigrants from non–sky island habitats) through isolation among tepuis. Similarly, the sky islands of the Cameroon Line in western Africa, which harbor many endemic species, were uplifted from lowland tropical forest during the Cretaceous

SKY ISLANDS

839

TABLE I

Sky Island Complexes Discussed in the Text Name

Region

Description

Madrean archipelago

Complex of 30–40 forested mountain ranges in the desert lowlands between the Rocky Mountains and Mexican Sierra Madres

Ethiopian Highlands

Southwestern United States (Arizona, New Mexico, and western Texas); Northern Mexico (Sonora, Chihuahua, and Coahuila) Southwestern United States (Nevada and Western Utah) Western United States (Colorado, Wyoming, Idaho, and Montana); Western Canada (British Columbia and Alberta) Northeastern Africa (Ethiopia and Western Eritrea)

East African Arc

Eastern Africa (mainly Tanzania)

Cameroon Line

Western Africa (Cameroon)

Annamite Range

Southeastern Asia (Vietnam and Laos)

Western Ghats

Southern India

Pantepui region

Venezuela, Guyana, and northern Brazil

Great Basin archipelago Rocky Mountains

Complex of ~200 mountain ranges arranged in close proximity surrounded by Great Basin desert Hundreds of isolated high-elevation meadows within a matrix of lower-elevation boreal forest Mountain massif sundered by a rift valley with satellite forested mountains isolated by lowland desert and savanna Circular complex of ~20 mountains (some of volcanic origin) harboring rain forest surrounded by arid woodland Linear chain of ~10 mountain ranges of volcanic and tectonic origin that also includes oceanic islands in the Gulf of Guinea Large cordillera and nearby satellite ranges of tropical rain forest and higher-elevation evergreen forest Extremely isolated chain of ~5 mountain peaks harboring humid rain forest surrounded by arid lowlands > 25 relictual plateaus of cool subtropical forest bisected by rivers and surrounded by tropical lowlands

note: Many other sky island systems exist worldwide, as do other fragmented high-elevation habitats with characteristics similar to those of sky islands.

through tectonic and volcanic activity, and their large diversity of endemic species have apparently accumulated over this long history by in-situ speciation. Interestingly, oceanic islands in the nearby Gulf of Guinea formed concurrently through the same processes. The mountain isolates of the East African Arc, the Western Ghats in India, and the Annamite Range of Vietnam and Laos are also relatively old sky islands where differentiation among sky islands has promoted diversification. In contrast, the sky islands of North America, especially those in the Rocky Mountains, are relatively young geologically, on the order of 40–70 million years. Despite their short history, they have still promoted species diversification and have accumulated an impressive level of diversity. Species in many sky island complexes, especially those in temperate regions, have experienced multiple and frequent climate-induced distributional shifts in association with the Pleistocene glacial cycles. This dynamic history contrasts with oceanic island systems, where inter-island dispersal is fairly rare and the geographic configuration of species across islands remains relatively constant over geologic time. Glacial cycles occurred regularly throughout the Pleistocene and intensified in the last 700,000 years, causing dramatic latitudinal and elevational shifts in species’ distributions. During past glacial periods, displacement of sky island habitats to lower elevation meant that formerly isolated populations came into contact. This retreat into refugia was then followed by recolonization of

840

SKY ISLANDS

the isolated mountaintops as climate warmed during relatively short interglacial periods. Tropical sky islands—for example those in the Cameroon Line and the Ethiopian highlands—were also affected by Pleistocene glacial cycles through global cooling, although the effect was likely moderate compared to that occurring at higher latitudes. A dynamic history of climate change could potentially explain why the sky islands of the relatively high-latitude Madrean archipelago in the southwestern United States and northern Mexico have low levels of endemism despite harboring high diversity by virtue of lying at a crossroads of several biogeographic regions. In contrast, the climate regime of the sky islands of the endemic-rich East African Arc is thought to have been quite stable throughout the Pleistocene, thanks to the ameliorating influence of Indian Ocean currents. In contrast to oceanic islands, colonization of sky islands can also occur through niche shifts from neighboring low-elevation habitats or through short-distance migration, in addition to long-distance colonization from other sky islands. For this reason, the dynamics of species assemblage in sky islands differ in many ways from the equilibrium theory of island biogeography set forth by MacArthur and Wilson in 1967, which postulates that island species diversity is a balance between colonization (distance between habitat patches) and extinction (size of patch). Research by Brown and colleagues showed that species composition of mountaintop mammals in the

Great Basin was largely governed by extinction and that the type of habitat that potential colonizers had to traverse to reach a new island, rather than geographic distance per se, determined when colonization by immigration was likely. Studies on birds were more consistent with predictions of the theory of island biogeography, showing an effect of habitat area and degree of isolation on species diversity and endemism. One particularly extreme example is the Sierra del Carmen of northern Coahuila, Mexico, an isolated sky island that shows strong effects of insularity typical of oceanic islands, including reduced species diversity and resulting ecological niche expansion, and augmented density of several resident bird species.

FIGURE 3 The Chisos Mountains, a sky island in Big Bend National

Park in southwestern Texas and part of the Madrean archipelago. Chihuahuan desert gives way to desert grassland at mid-elevation and

SKY ISLANDS: WINDOWS INTO EVOLUTIONARY PROCESS

Sky islands, like oceanic islands, are ideal settings for evolutionary study. The replicated spatial arrangement of similar, isolated habitats provides a natural experiment for determining what ecological and geographic features facilitate diversification. Selection and genetic drift are both particularly powerful in island systems: The isolation between islands is conducive to divergence by genetic drift, whereas opportunities for selection may occur because of differences in ecology or factors relevant to sexual selection among islands. Additionally, divergence by natural selection might occur across altitudinal gradients within individual sky islands. For example, the Madrean sky islands are distributed among 30–40 mountain ranges of various sizes and orientations that link the Sierra Madre massifs with the Rocky Mountains in stepping-stone fashion. The present configuration of habitats within these sky islands is postglacial (less than 20,000 years ago). Whereas relatively contiguous woodlands covered the region during glacial periods, there is greater habitat contrast in the Madrean sky islands today: arid oak woodlands in low-elevation canyon, mid-elevation pine-oak forest, and boreal forest in some of the highest-elevation sky islands (Fig. 3). These ecological transitions are comparable to changes manifest at much larger geographic distances (e.g., over hundreds of kilometers in latitude), yet in sky islands they occur over just a few kilometers—and within the dispersal capacity of most species. Plant and animal populations that span these ecological transitions experience drastically different environments, both abiotic (e.g., rainfall and temperature) and biotic (e.g., food resources, competitors, and predators), which can result in divergent selective pressures. If strong enough, selection can generate observable differences between populations that are connected by

pine forest at high elevation. The nearest sky island is the Sierra del Carmen in northern Coahuila, Mexico, approximately 60 km away.

gene flow. This “gradient model” of divergence has been validated by empirical studies showing elevational diversification of ecologically divergent sibling species as well as population-level divergence in adaptive morphology. SKY ISLANDS AS GENERATORS OF DIVERSITY

Much debate surrounds how species diversification has been promoted in sky islands. The likely effect of isolation among sky islands on the formation of new species is not controversial; however, the effect of the Pleistocene glacial cycles on species divergence is a contentious issue. Because of the relatively short amount of time between glacial cycles, there are disagreements over whether diversification would have been inhibited or promoted. Although divergence is expected within sky islands from the divergent selective pressures generated from altitudinal habitat differences, as well as from opportunities for differentiation among isolated sky islands by genetic drift and/or selection, differentiation among populations that accumulated during one glacial cycle may have been lost during subsequent distributional shifts. Alternatively, if sufficient reproductive isolation evolved in any one cycle, speciation could actually have been promoted by the frequent distributional shifts, as hypothesized by the “species pump” model of diversification. Molecular data play a critical role in distinguishing among evolutionary scenarios, as neutral genetic markers provide a means for estimating the timing of divergence and reconstructing the history of divergence across the sky island landscape. In the Madrean archipelago, fossilized plant material from packrat (Neotoma spp.) nests has allowed for a thorough reconstruction of paleohabitats through the last glacial maximum. According to these

SKY ISLANDS

841

data, fragmentation of sky islands occurred approximately 10,000 years ago, offering a hypothesis that can be tested with intraspecific genetic data from sky island species. So far, molecular studies in this region have produced conflicting results on the timing of divergence, with some species showing no evidence for postglacial divergence among sky islands and other species showing evidence for divergence much older than the last glaciation. This discord between paleoecological and genetic data can be attributed in part to the difficulty of detecting and timing recent genetic divergence. Large sample sizes and thorough regional sampling will undoubtedly be necessary to thoroughly test this hypothesis. What has emerged from the many molecular phylogenetic and phylogeographic studies on sky islands is that some species show strong patterns of differentiation that are coincident with the glacial cycles, whereas divergence predates the Pleistocene in other taxa. For example, in the northern Rocky Mountains of North America, some insect groups, such as flightless montane grasshoppers in the genus Melanoplus, radiated during the Pleistocene. Genetic data indicate that divergence among multiple glacial refugia, as well as differentiation during the recolonization of sky islands, was facilitated by genetic drift. However, sexual selection has also been implicated in this diversification because of the rapid divergence in male genitalia, which can play a role in reproductive isolation. Sexual selection has also likely been important to incipient speciation among populations of a jumping spider (Habronattus pugillis) in the Madrean archipelago. Populations on different mountains have divergent phenotypes, seismic songs, and courtship displays, which result in mating incompatibilities between forms. Also in the Madrean archipelago, pollinator-mediated natural selection seems to be influencing morphological differentiation among populations of a perennial plant (Macromeria viridiflora). Thus, in North American sky island systems, natural selection, sexual selection, and genetic drift all appear to be important evolutionary processes driving Pleistocene diversification. However, the timing of speciation in other taxa from the Rocky Mountains—most notably the mammals—predates the Pleistocene, suggesting that divergence in these species may have been inhibited by the frequent displacements from their sky island habitats. A pattern of pre-Pleistocene origination also has been found in a number of tropical bird and mammal species, casting doubt on the hypothesized “species pump” model of diversification among tropical sky islands. Nevertheless, diversification of tropical taxa across sky islands, albeit

842

SKY ISLANDS

pre-Pleistocene, demonstrates how this geographic setting is conducive to divergence. More examples of divergence among sky islands in tropical regions will undoubtedly be uncovered with further study and will be important for confirming that tropical sky islands do not conform to the “species pump” model of diversification. Regardless of the timing of diversification, sky islands are clearly promoters of diversification from the level of the population to the species. Despite the fact that our most thorough descriptions of genetic and phenotypic divergence in sky islands come from temperate North America, this sampling bias should not be taken as evidence that tropical sky islands are not important centers of diversification. In the Pantepui region of South America, for example, passerine birds (genus Myioborus and Atlapetes) have differentiated at both the subspecies and species level among the isolated plateaus. Similarly, populations of poison dart frogs inhabiting sky peninsulas in central Peru have diverged rapidly in coloration despite close proximity. In a particularly thorough study of the starred robin (Pogonocichla stellata) on sky islands in the East African Arc, Rauri Bowie and colleagues showed how considerable genetic diversification has accumulated among phenotypically divergent sky island populations. Further studies promise to reveal more about patterns and timing of diversification in tropical sky islands, which have been little explored compared to their temperate counterparts. Lastly, such information will also provide an interesting counterpart to the considerable attention given to the diversification across isolated tropical habitat fragments, such as the remnants of previously widespread wet-tropical forest fragments found in North Queensland, Australia. IMPACT OF HUMAN-INDUCED CLIMATE CHANGE ON SKY ISLANDS

Climate change poses a unique threat to sky islands. Temperature increases of as little as a few degrees could push sky island habitats to higher elevations, reducing their area and potentially causing local extinction of endemic taxa and divergent populations harboring unique genetic and phenotypic diversity. Sky islands in North America and Mexico are already being affected by climate change, with increases in drought, fire, and outbreaks of invasive insects. Although these resilient systems have endured large-scale shifts in climate throughout the Pleistocene, the pace of human-induced climate change may represent an insurmountable challenge for sky islands, with potentially devastating consequences to their biodiversity and evolutionary potential.

SEE ALSO THE FOLLOWING ARTICLES

Fragmentation / Island Biogeography, Theory of / Pantepui / Refugia / Sexual Selection FURTHER READING

Bowie, R. C. K., J. Fjeldså, S. J. Hackett, J. M. Bates, and T. M. Crowe. 2006. Coalescent models reveal the relative roles of ancestral polymorphism, vicariance, and dispersal shaping phylogeographic structure of an African montane forest robin. Molecular Phylogenetics and Evolution 38: 171–188. Brown, J. H. 1978. The theory of insular biogeography and the distribution of boreal birds and mammals. Great Basin Naturalist Memoirs 2: 209–228. Heald, W. F. 1951. Sky islands of Arizona. Natural History 60: 56–63, 95–96. Lovett, J. C., and S. K. Wasser. 1993. Biogeography and ecology of the rain forests of eastern Africa. Cambridge: Cambridge University Press. Mayr, E., and W. H. Phelps Jr. 1967. The origin of the bird fauna of the south Venezuelan highlands. Bulletin of the American Museum of Natural History 136: 273–327. Pielou, E. C. 1991. After the Ice Age: the return of life to glaciated North America. Chicago: University of Chicago Press. Smith, T. B., K. Holder, D. Girman, K. O’Keefe, B. Larison, and Y. Chan. 2000. Comparative avian phylogeography of Cameroon and Equatorial Guinea: implications for conservation. Molecular Ecology 9: 1505–1516. Warshall, P. 1995. The Madrean sky island archipelago: a planetary overview, in Biodiversity and management of the Madrean archipelago: the sky islands of the southwestern United States and northwestern Mexico. L. DeBano, P. Ffolliott, A. Ortega-Rubio, G. Gottfried, R. Hamre, and C. Edminster, eds. Fort Collins, CO: USDA Forest Service Rocky Mountain Forest and Range Experiment Station, 6–18.

SNAILS SEE LAND SNAILS

SNAKES GORDON H. RODDA U. S. Geological Survey, Fort Collins, Colorado

Snakes (3000+ spp.) are a highly specialized and successful limbless form of lizard. Their low metabolic rate combined with jaw anatomy that accommodates the ingestion of relatively prodigious meals allows snakes the energetic option of long fasts between large meals, permitting them to rely on infrequent food sources such as annually nesting birds. Avian and mammalian predators require more continuous food sources and therefore cannot survive on islands lacking year-round prey. Accordingly, snakes are the top predator on many islands, especially land-bridge

islands that experience pulses of visiting birds. However, snakes are rare on most oceanic islands because they are not well suited for dispersing across saltwater. Snakes may dramatically rearrange the vertebrate ecology of formerly snake-free oceanic islands when they are introduced by humans. EVOLUTION

Snakes were the last major group of reptiles to appear, evolving from a varanid lizard–like progenitor at least 135 million years ago. Because the eyes of snakes are radically different from those of their immediate lizard ancestors, proto-snakes are believed to have passed through an evolutionary transition in which their eyes degenerated as a result of living underground or in murky water. Returning to the aboveground environs suited them well and set off an adaptive radiation yielding over 3000 modern species, covering all habitats except high arctic and cold marine environments. What are the ingredients for this evolutionary success? One is a superbly developed ability to smell prey, perhaps a benefit retained from the time spent underground or in murky water. Another is the loss of legs. Leglessness facilitates locomotion underground, in vegetative thickets, and in water. Another is their ectothermic heritage, which saved them the expense of having to feed continuously to generate heat. The low metabolic cost of ectothermy allows a predatory style that emphasizes stealth, patience, and surprise over speed and endurance. Mammals and birds are incessantly in motion, searching insatiably for each next bite of food; snakes are well suited to the complementary lifestyle: patiently conserving energy while waiting for a frenetic endotherm to wander within striking range. A less obvious asset of snakes is their very light and supple jaws, which arose in the course of ophidian evolution to permit the ingestion of extraordinarily large meals (at maximum, more than 100% of their body mass). These lightweight jaws are unsuited for biting prey into submission, so snakes developed constriction and venom (or perhaps their possession of venom and constriction enabled the concurrent evolution of lightweight jaws). Thus, snakes are often successful by virtue of visualand olfactory-targeted sit-and-wait predation from the security and concealment of water, a burrow, or a thicket. Sit-and-wait snakes generally subdue their prey by venom or constriction. Other snakes are successful by patrolling for relatively defenseless prey, such as eggs, that are large but rarely found: Having low metabolic needs, the snake can afford to wait for the rare but highly rewarding (i.e., large) prey item.

SNAKES

843

THE ROLE OF SNAKES ON ISLANDS

Snakes have limited aerobic capacity, are poor longdistance swimmers, and, of course, cannot fly, so they are not particularly well suited for dispersal to islands that have never been connected to land (i.e., oceanic islands). A snake drifting across the surface of the ocean without concealment in vegetation would also be highly vulnerable to marine predators. Most reptiles on oceanic islands are presumed to have reached their destination by rafting on flood-swept vegetation, usually in the form of eggs buried in earth or tree hollows. Should an adult succeed in making the overwater trip, a single gravid or pregnant female could start a new population. One snake species (the flowerpot snake, Ramphotyphlops braminus) is parthenogenetic, allowing each individual to clone itself on arrival to a new island; this species is pan-tropical and is found on many islands. However, most snakes are oviparous and lay eggs with leathery (semi-permeable) shells; these rarely withstand long contact with seawater. Thus, there are many remote island chains that have no native snakes. For example, only the Pacific tree boa (Candoia bibroni) colonized the central Pacific Ocean islands, and it did not manage to disperse eastward from Samoa (it may have reached even Samoa only with assistance from prehistoric humans). Should they reach an island, snakes are well suited for island living, especially if the island has suitable prey that are available infrequently. Many Mediterranean, East Indian, and West Indian islands, for example, experience annual or semi-annual pulses of migratory bird visitation. Seabirds usually visit their breeding islands at a particular time of year. Before the seabird nestlings are fledged, the chicks are often superabundant and relatively defenseless against snakes. Medium-sized mammals are often a predatory threat to snakes, but mammals disperse poorly to islands and are particularly poorly suited to survival on islands with infrequent pulses of food. Thus, snakes can often do very well on islands, if they can get there. NATIVE SNAKES ON ISLANDS

A few sea snakes take refuge on small islets to rest and breed. Most sea snakes are live bearers and have no need to ever crawl onto land, but the half-dozen species of the Australasian sea krait genus Laticauda are less fully evolved for a marine existence and—like sea turtles—must crawl up on land once a year to lay eggs. They also may come ashore to mate and rest. Sea kraits can be impressively abundant on the islands where they mate, but they are of little energetic consequence to the island’s terrestrial ecosystem because they forage exclusively offshore.

844

SNAKES

Native terrestrial snakes have been studied on a wide variety of islands with simple food chains. On large landbridge islands with complex food webs, snakes have an energetic role much as they do on adjacent continental land masses. However, on small or oceanic islands, which often have simplified food chains, snakes may take advantage of their unique adaptive advantages to attain extraordinary densities. Conditions that benefit a population may not be favorable for the constituent individuals, however: At the high densities achieved, the competition for food is intense, and each individual snake may have difficulty acquiring enough energy for rapid growth. This may lead to insular dwarfism. The proximate cause for small size on an island may be a shortage of food (phenotypic response) or a genetic disposition for reduced growth (which improves survival in the face of limited food availability). Conversely, the food that is available on an island may be of a size requiring a larger body size (e.g., if large birds such as boobies are the main food source). In this case, insular gigantism may occur. Both dwarfism and gigantism have been widely noted of snakes on islands, but in only a few cases is it known whether the observed size differences (in comparison to the nearest mainland population) are due to phenotypic or genotypic adaptation. In the best studied case, there was a strong interaction: The island snakes had a genetic disposition to allow them to take maximal advantage of large prey sizes, but they achieved large size phenotypically in response to elevated intake. Snake species tend to diversify on islands by evolving adaptations to the local environment, but those locally adapted species rarely colonize adjacent islands, as is characteristic of adaptive radiations of other taxa such as island birds or island lizards. SNAKES INTRODUCED TO ISLANDS BY HUMANS

Most human-introduced snake populations on islands probably arose accidentally. For example, the flowerpot snake has been accidentally introduced around the world in association with potted plants and soil. Because this diminutive, termite-eating subterranean snake is rarely seen—and because termites do not have many defenders—this introduction is rarely noted and less often lamented. Horticultural shipments may be an important pathway for the introduction of snakes, as plant products are probably responsible for colonization of several introduced snakes in the West Indies. Corn snakes, Pantherophis (= Elaphe) guttatus, probably concealed in ornamental plants shipped from Florida, have

become established on Anguilla, Bonaire, Curaçao, and the Cayman Islands. One of the strangest snake introductions occurred around 1971 on the island of Cozumel, Mexico, following the filming of a movie. Non-native Boa constrictor were brought to the island for the filming and were subsequently released. Twenty years later, the residents noted the extraordinary abundance of the new boas and the disappearance of a number of Cozumel’s unique wildlife species. More than a dozen endemic subspecies are believed to be at risk, including birds, mammals, and lizards. Boas have subsequently been released and are presumed established on Curaçao and Aruba in the Dutch West Indies. Perhaps the frequent traveler award for snakes should go to the rear-fanged wolf snake, Lycodon aulicus, named for its anterior teeth, which resemble those of a wolf. Introduced to several places in the Mascarene Islands during the nineteenth century and to Australia’s Christmas Island around 1987, it is also suspected of having been introduced at an unknown date to the Philippine Islands, and possibly to Java, Borneo, and Sumatra. The wolf snake is variable in size and diet, but on Christmas Island, it eats primarily lizards (of which Christmas Island has five vulnerable endemic species, as well as an endemic shrew). The wolf snake is suspected of causing multiple lizard extirpations throughout the Mascarene Islands, especially on Mauritius. The oldest reported snake introductions are of snakes apparently placed on Mediterranean islands by the Romans or Carthaginians. Scientific attention has focused on the threats to native wildlife in Spain’s Balearic Islands (Mallorca, Menorca, etc.). Two snake species are involved: Macroproton cucullatus, the false smooth snake, and Natrix maura, the misleadingly named viperine snake (a nonvenomous water snake unrelated to vipers). The false smooth snake is associated with the endangerment of an endemic lizard, Lilford’s wall lizard, Podarcis lilfordi, and perhaps other species. The viperine snake is best known as the threat to the continued survival of the endemic Mallorcan midwife toad, Alytes muletensis. The wall lizard is listed by the International Union for Conservation of Nature (IUCN) as endangered, and the toad is judged vulnerable. Both prey species persist only in highly restricted ranges in the Balearic Islands, suggesting that their ecological fates hang in the balance. This uncertainty is truly extraordinary given that the exotic predators were introduced about 2000 years ago. Presumably other prey species were eliminated more quickly. The tenuous persistence of the toad and lizard suggest that the time course of island extinctions may be vastly longer

than the usual duration of scientific studies, a cautionary tale for scientists wishing to conclude that a given introduction is harmless as no negative impacts have yet been demonstrated. Perhaps the earliest example of ecoterrorism may be manifest in the occurrence of Vipera aspis, the asp viper, on the island of Sicily. Some believe that the viper was introduced to Sicily by the Carthaginians during their conquest of the island in the years 398–368 BC. The basis for the speculation is that the Carthaginians were known to load a small boat with a collection of venomous snakes and push the boat toward enemy ships as a means of terrorizing their opponents prior to combat. In the case of an island to be conquered, the victors would then be rewarded with acquiring their new territory infested with an introduced venomous snake, a rather questionable achievement and perhaps an early example of tactical cleverness untempered by consideration of the strategic consequences. The Brown Tree Snake

The best documented snake introduction comes from the island of Guam, in the Mariana Islands of the western Pacific, and it too involves a snake translocated by naval forces (but this time inadvertently). The brown tree snake, Boiga irregularis, was accidentally introduced to Guam in war materiel salvaged from the New Guinea area immediately after World War II. For at least 35 years, the snake population growing in the southern part of the previously snake-free island was a subject of interest but not concern. The concurrent disappearance of virtually all bird life from the southern portion of the island attracted little attention until the 1980s, by which time the imminent extinction of Guam’s native forest birds set off a frantic search for the disease that was assumed to be killing the birds. A comprehensive search turned up no diseases that could account for the bird disappearances. Consideration then turned to pesticide contamination and other possible causes, again to no avail. Only when all other reasonable hypotheses were rejected did ecologists begin to accept the notion that an introduced snake could be responsible. Resistance to the snake hypothesis was based on the presumption that snakes were too rare and ecologically inconsequential to extirpate species. In this particular case, it turned out that not only was the brown tree snake responsible for the demise of virtually the entire forest avifauna of Guam (ten of 12 species have been extirpated to date, with the two other species teetering on the brink), but it was also responsible for extirpating mammals (three bats were lost,

SNAKES

845

has made snakes so successful—their invisibility and tolerance of long periods without food—is also the feature that makes them one of the most difficult invasive species to prevent or manage. SEE ALSO THE FOLLOWING ARTICLES

Cozumel / Introduced Species / Lizard Radiations / Mascarene Islands, Biology FURTHER READING

FIGURE 1 Brown tree snakes reach high densities in their introduced

population on Guam. These were collected along 0.3 km of road edge in 1989. Photograph by G. H. Rodda, USGS.

though it is still unclear what caused the first two species to disappear) and many of the island’s lizards (between one and five of the island’s nine to 12 native lizards disappeared in association with the snake). Ecologists are well aware that introduced mammals frequently wipe out island populations of birds; why then was it so hard for ecologists to accept similar culpability on the part of a snake? This is a problem for psychologists, but ecology played a role. Recall that the evolutionary success of snakes hinges on their cryptic and stealthy predatory mode. Most prey individuals never saw the snake that bit them (until it bit them). Humans are no better at spotting snakes; some humans are ophiophobic because snakes are so invisibly omnipresent. As a consequence, ecologists frequently underestimate the abundance of snakes in an area. This was the case in Guam (Fig. 1), where the brown tree snake was later found to have reached extraordinary population densities at its peak (in excess of 100/ha). This was more than enough to wipe out the island’s birds (estimated to have totaled about 30/ha for all species combined in optimal habitat), but even at their peak density the snakes were largely invisible, as they are active only at night and usually in heavily vegetated areas where the very slowmoving, vinelike, drab brown snakes are very difficult to see. Despite high snake densities persisting on the island, Guam’s visitors and residents rarely see brown tree snakes. The example of the brown tree snake on Guam revolutionized ecologists’ understanding of the potential importance of snakes in island ecosystems. All island ecosystems are at risk from the introduction of new snake species, but especially vulnerable are ecosystems such as Guam that evolved without any snakes present. Unfortunately, the difficulty of detecting snakes at low densities continues to challenge managers keen on preventing colonizations of snakes on islands. The evolutionary innovation that

846

SOCOTRA ARCHIPELAGO

Fritts, T. H., and G. H. Rodda. 1998. The role of introduced species in the degradation of island ecosystems: a case history of Guam. Annual Review of Ecology and Systematics 29: 113–140. Greene, H. W., M. Fogden, and P. Fogden. 1997. Snakes: the evolution of mystery in nature. Berkeley: University of California Press. Jaffe, M. 1994. And no birds sing. New York: Simon and Schuster. Quammen, D. 1996. The song of the dodo. New York: Scribner’s. Rodda, G. H., T. H. Fritts, and D. Chiszar. 1997. The disappearance of Guam’s wildlife: new insights for herpetology, evolutionary ecology, and conservation. BioScience 47: 565–574. Rodda, G. H., Y. Sawai, D. Chiszar, and H. Tanaka, eds. 1999. Problem snake management: the habu and the brown treesnake. Ithaca, NY: Cornell University Press. Shine, R. 1991. Australian snakes: a natural history. Ithaca, NY: Cornell University Press. Zug, G. R., L. J. Vitt, and J. P. Caldwell. 2001. Herpetology: an introductory biology of amphibians and reptiles, 2nd ed. San Diego, CA: Academic Press.

SOCIETY ISLANDS SEE PACIFIC REGION

SOCOTRA ARCHIPELAGO KAY VAN DAMME Ghent University, Belgium

The Socotra Archipelago (Yemen), situated in the Arabian Sea, consists of four main islands on an ancient microplate. Socotra, the largest island, is known as the “Galápagos of the Indian Ocean” because of its high biodiversity in both terrestrial and marine realms. A long period of isolation from the Afro-Arabian mainland and a significant geological diversity have resulted in a high endemism with remarkable relicts. “JEWEL OF THE ARABIAN SEA”

The Socotra Archipelago is located in the Arabian Sea, situated 380 km southeast from the coast of Yemen, to

which it politically belongs, and about 100 km east from the Horn of Africa (Cape Guardafui, Somalia). It consists of one major island in the west, Socotra, about 130 km long and 40 km wide; three smaller islands, Samhah, Darsa (“The Brothers”), and Abd al Kuri, to the east; and a few rocky limestone outcrops inhabited only by birds. The largest island is populated by approximately 43,000 inhabitants (census 2004) with highest concentration in two major towns, the capital Hadiboh and the coastal town Qalaansiyah. The population is rapidly increasing from immigration, resulting in strong urban expansion. Socotra’s high cultural and natural diversity and the complexity of conservation attract a global interest. Listed as a UNESCO Man and Biosphere Reserve (2003), a Global 200 Ecoregion by the World Wildlife Fund (WWF), and a Centre of Plant Diversity by Plantlife International, Socotra has become Yemen’s national pride, its “Jewel of the Arabian Sea.” The country recognizes Socotra’s importance in biodiversity. For example, Yemen designated the Detwah Lagoon on Socotra as the country’s first Ramsar site (a wetland of international importance) in 2007. Since July 2008, the Socotra Archipelago is officially a UNESCO World Heritage site, a historical event. Seventy-five percent of the land surface is indicated as terrestrial core area and consists mainly of elevated areas, the rest is terrestrial buffer zone. The marine core area is smaller, 7.6% of the total marine area, and aims to protect mainly the coral reefs of the Archipelago. GEOLOGY OF SOCOTRA

Islands of the Socotra Archipelago (Fig. 1) lie together on a submerged Precambrian basement, the Socotran Platform. This granite microplate of continental origin was formed 700–800 million years ago as part of the Afro-Arabian continent, close to Eastern Oman. At this time, Socotra was situated relatively close to India, until a series of tectonic events in the Mesozoic leading to the breakup of Eastern Gondwana. The latter events caused structural changes in the Socotra Platform and formed the onset of rifting of the Gulf of Aden, which would lead to a final separation of Socotra from the Afro-Arabian mainland. Geological and tectonic evidence suggest that the Socotran Platform remained close to Southern Arabia until oceanic rifting of the Gulf of Aden in the Oligocene–Miocene, with an estimated timing of separation about 20–18 million years ago. Sea floor spreading increased the distance between Socotra and the Arabian mainland since 18 million years ago. The region was subjected to a general uplift since this period, and the Archipelago gradually rose.

FIGURE 1 Socotra Island, geology and topography, view from the

south. Geology shows a Precambrian basement in the central eastern half of the island, surrounded by the dominant Palaeocene–Eocene limestone plateaus. The center (Zahr Plain) and coastal areas consist of lowlands, and two plateaus crop out in the east. Inset: Location of the Socotra Archipelago. Image courtesy of Kay Van Damme, satellite image based on Landsat 7 ETM (2001) draped over USGS DEM.

These tectonic events, together with several large sea transgressions, shaped the islands. Socotra consists of a granite center, cropping out in the Haggeher Mountains (1600 m), which formed as part of the Precambrian basement complex. Although there is no direct geological evidence, several authors believe that these mountains have remained above sea level since their formation and thus provided a refuge for relict taxa. Bordering the Haggeher are a series of limestone plateaus reaching up to 1000 m in altitude (Diksam Plateau), deposited during several marine transgressions in the Cretaceous and PalaeoceneEocene. These limestone plateaus from the Paleocene– Eocene transgression make up the largest surface of Socotra Island, in most areas strongly karstified and with extensive cave systems. In the center (Zahr Basin), south (Noged plain), and north of the island, these plateaus are bordered by coastal lowlands consisting of Quaternary sands and elevated coral reefs. At periods of lower sea level during the last glacial maximum, most of the Socotra Platform emerged, connecting the main island to Samha and Darsa by land bridges. Thus there is a marked variety in topography and geological features on Socotra: Precambrian granite outcrops, Paleocene–Eocene limestone plateaus with steep cliffs and

SOCOTRA ARCHIPELAGO

847

large cave systems and Quaternary coastal sandy regions. The smaller islands (Abd al Kuri and The Brothers) are built up mainly of limestone cliffs. CLIMATE

Because Socotra is situated at the margins of the subequatorial and northern tropical climate belts, its climate is governed by the Intertropical Convergence Zone and related monsoon cycles. Two seasons, the northeast monsoon (January) and the southwest monsoon (July), are most marked, and the intensity of their winds has kept the island isolated during these periods even up to the last decades. Both are quite different; the January monsoon blows dry air on land and creates upwelling of cooler waters in the sea, whereas the July monsoon brings moisture, but parts of the island (the north) are in a rain shadow and remain dry. Annual temperatures fluctuate strongly depending on region because of the differences in topography, ranging between 8 and 31 °C in the mountains (Skand) and from 28 to 43 °C at the coast (Hadiboh). Also, air humidity and precipitation strongly vary with location, some areas depending completely on permanent fog for precipitation. Overall, climate on Socotra depends strongly on place and time, and the island shows both tropical and arid features.

limited, with local hotspots of endemism such as the granite mountains, which count over a hundred endemic species. The mainly xeric vegetation has been classified into seven types: coastal mosaic, croton shrubland, succulent shrubland, woody-based herb communities, semi-evergreen woodland, submontane shrubland, and montane mosaic (Fig. 2). Extensive research has been done on the plants, in particular on the ethnobotany. The endemic flora contains several relicts from formerly widespread taxa. The Socotran dragon’s blood tree (Dracaena cinnabari) is such an example, a local representative of a genus that was more widespread in surrounding regions during the Miocene. Dracaena is considered a genus of Tethyan origin based on fossil evidence, now with fragmented distribution because of climatic shifts. Remaining species of Dracaena have survived in refugia on Socotra, parts of the Arabian Peninsula, northeastern Africa, and northwestern Africa. Known since antiquity for its red resin, the Socotran dragon’s blood tree has become a symbol for the island. Other well-known floristic elements of Socotra are the frankincense trees Boswellia (a genus that radiated on Socotra), the medicinal Aloe, and the cucumber tree Dendrosicyos socotrana, the only arborescent member of the Cucurbitaceae. Most plant species show clear xeromorphic adaptations to the dry and windy conditions governing these islands, others are restricted to wet refugia.

RESEARCH HISTORY

Biological research on Socotra Island started at the end of the nineteenth century, with main expeditions by British botanist I. B. Balfour of the Royal Botanic Gardens of Edinburgh (RBGE) and the zoologists H. Forbes and W. R. Ogilvie-Grant of the Liverpool Museums. Little coordinated research was carried out in the twentieth century until political stability increased accessibility to the island again in the late 1990s. Major expeditions, such as a multidisciplinary expedition by the RBGE and marine surveys by Senckenberg Museum in Germany since 1999. Building on the earlier works, these recent expeditions form the basis for our current faunistic and floristic knowledge of the island and provide an important framework for conservation.

848

FIGURE 2 Montane shrubland vegetation (altitude 1000 m) at Rewged,

Socotra Island, with the typical umbrella-shaped Socotran dragon’s blood tree (Dracaena cinnabari). The mountaineous areas of Socotra Island form a local hotspot of endemism.

ENDEMISM IN TERRESTRIAL FLORA

TERRESTRIAL FAUNA

Geological history, topography, and microclimates on Socotra, combined with its long period of isolation, lie at the base of a high biodiversity and endemism in terrestrial biota. Of 825 terrestrial plant species recorded, 37% are endemic to the archipelago, with 15 unique genera. In comparison, Mauritius has 31–35% endemic plant species and the Galápagos about 42%, ranking Socotra high in island biodiversity. Distributions of species are relatively

Endemism is high in terrestrial fauna, reaching up to 100% in some groups. Endemism (at species level) for spiders is ~60%, and for isopods 73%. Large groups of insects such as butterflies, dipterans, hymenopterans, grasshoppers, and beetles need taxonomic revision, and at least a third to half of the known species are endemic to the archipelago. Scorpions and Amblypygi are all endemic. Endemism and diversity in the terrestrial molluscs (Fig. 3) is exceptionally high,

SOCOTRA ARCHIPELAGO

discovered from Socotra in the last decade, including the endemic genera Paradoniscus and Dioscoridillo. As to vertebrates, primary freshwater fish and amphibians are absent. Their absence may be the result from extinctions during long periods of drought in Socotra’s history. In reptiles, about 90% of the 34 terrestrial species are endemic, with all six snake species endemic, and a radiation in the geckos (e.g., Hemidactylus and Pristurus). Indigenous mammals are absent, though the insectivores (shrews and bats) deserve closer study. For birds, the global importance of Socotra was recognized by Birdlife International, which lists 22 important Bird Areas here. Six species of birds (e.g., the Socotra sunbird) are endemic to Socotra. Several subspecies (e.g., the Socotra buzzard) are under taxonomic research and may prove to be additional true endemic species in the future. For other species, such as the Egyptian vulture and several seabirds, Socotra harbors globally significant populations. In total, Socotra counted 192 bird species in 2006. NOTE ON AFFINITIES OF THE BIOTA

FIGURE 3 “Tropidophora” socotrana on Dracaena wood. About 95%

of the land snail species in Socotra are endemic. They evolved and radiated on the archipelago, most likely after separation of Socotra from the Arabian mainland. This particular operculate species of the Pomatiidae lives in close association with the Socotran dragon’s blood tree.

with 95% of ~100 species currently described, endemic to the archipelago; 75% of the land snail genera on the archipelago are unique. A large portion was only discovered in the last decade, and more species are being described. High endemism of the Socotran molluscs provided an important argument for its World Heritage nomination. Dominating the freshwater habitats (wadis) are endemic, semiterrestrial freshwater crabs of the genus Socotrapotamon. The dominant species, Socotrapotamon socotrensis, is common on the main island. A second terrestrial species, Socotra pseudocardisoma, is restricted to rock holes in the high limestone plateaus. A separate evolution can be noted for cave invertebrates, of which seven species are unique to the island and, in several cases, to a particular cave system. The cave systems on Socotra are extensive, reaching 13 km in length (Ghiniba Cave), and harbor endemic species that have found refuge here during dry periods. Among them, a typical group of Gondwana origin, are the whip spiders (Amblypygi), with four endemic species of the genera Phrynichus and Charinus. Among the stygobionts, several new freshwater crustaceans were

The origin of the Socotran endemics depends on the group in question. Both vicariance and dispersal hypotheses have been postulated. Although for many groups a link with the Arabian mainland and Africa is clear, Oriental, Afrotropical, and Palearctic elements may also be present. The biogeography of Socotra, because of its ancient continental origin, is a complex matter and should be carefully examined jointly with dispersal capacities of each group. Only few studies have focused thoroughly on this aspect of Socotra, and relatively more work has been done on the floral links. The following are a few examples. In isopods, species composition on Socotra shows interesting biogeographical links with Afrotropical and Oriental regions, whereas important AfroArabian groups are missing. A radiation of Oriental isopod species (Serendibia and four genera of the Trachelipodidae), groups that are not present in Africa and Arabia, indicate a long evolutionary history and isolation of the isopod fauna of the Socotra Archipelago. Species of Serendibia seem to derive from a single common ancestor, and these “oriental” elements are considered survivors of relict taxa, no longer present on the surrounding mainland. The terrestrial molluscan fauna, strongly limited in dispersal, is suggested as ancient as well, of central Gondwanan origin, with virtually no species overlap in distribution ranges between the different islands. A similar (Gondwana) age is suggested for the Amblypygi. Several radiations of the latter groups have taken place on the main island. A relatively more recent (Oligocene–Miocene) age was suggested using molecular clocks for two endemic Socotran snakes belonging to the Colubridae, likely invading from Africa.

SOCOTRA ARCHIPELAGO

849

The endemics of Socotra contain a mix of biogeographical histories. Dispersal is likely in the majority of terrestrial plants, with closest relationships to the adjacent Afro-Arabian mainland (Eritreo-Arabian floristic subregion). Unraveling the connections is not simple. Relationships of the Socotran Exacum, a genus with a vicariance-like pattern, is suggested now to result from long-distance dispersal. Molecular studies confirm vicariance for a few groups. In the cucumber tree Dendrosicyos, for example, age of lineage is estimated at ~40 million years (Oligocene), predating separation of the island. Information on biogeography of Socotra is scattered in phylogenetic studies, not all equally focused, and different scenarios exist for different groups. The number of true relicts versus recent colonizers depends strongly on the group and needs further investigation. In general, true Socotran relicts are considered relatively old (i.e., Miocene or older). Furthermore, a comprehensive geological framework is lacking, with the age of separation of the island only recently becoming clear. Little has been done on the biogeographical connections between the islands of the Socotra Archipelago. Scientists are currently working jointly, across disciplines, on compiling the biogeographical data and investigating main patterns. ALOES, FRANKINCENSE, AND DRAGON’S BLOOD

Socotra is known for several products that made the island a popular trading center in antiquity. The most common products were tree resins such as frankincense and dragon’s blood (for pigment), and sap from aloe. These products are still used. In addition, the island was famous for important components in the perfume industry, such as ambergris (from whales) and musk (from the lesser civet, introduced from India). Nearly all historical sources (e.g., Marco Polo), mention one or another of these products in relation with Socotra, in particular the high-quality incense and aloes. In addition to the use of several plants for trade, Socotri traditionally maintain a vast practical etnobotanical knowledge. EXTINCTIONS

In contrast to many islands, Socotra has remained in a relatively pristine state until the last decades. The island has been strongly isolated for both political and climatological reasons (strength of the monsoons), and only since 2001 has a long period of isolation been breached. Four plant species have not been recorded since Balfour’s collections in the last century and are considered extinct. For the fauna, there is only circumstantial evidence for extinctions. A single historical source, a detailed account from the first century AD

850

SOCOTRA ARCHIPELAGO

(the Periplus), mentions the presence of crocodiles, large lizards, and three species of land turtle on Socotra. Use of the large lizards and one “giant” mountain tortoise are described, the latter being used for making Roman tableware. There is no paleontological evidence on Socotra for the presence of these vertebrates, and the evidence is, as yet, inconclusive as to whether they actually existed here. However, as the Periplus is known for being a quite accurate source, we cannot exclude the possibility that these species, including a giant land turtle, may have gone extinct as a result of human interference. Major risks for extinctions on the island have arisen only in the last decade. THREATS

Major threats to the island are many. Several that have arisen in the last decade can be expected to have a major impact on biodiversity on Socotra. Traditional practices, land-use systems, and local laws concerning the environment and use of resources are being lost. Traditionally, fisheries and livestock management of the local communities (fishermen and herders) were aimed at sustaining natural resources. Through centuries of experience, Socotri have become careful stewards of their island, well aware of the meaning of sustainable use. Disruption of traditional ways negatively impacts biodiversity through overconsumption, such as firewood collection, overgrazing, and overfishing. Overgrazing by goats is a major problem, causing severe soil erosion through loss of vegetation. Regeneration is low, in combination with the dry climate, weakening the ecosystem. Some plants such as the cucumber tree may also suffer, being used as fodder in times of scarcity. The breakdown of traditions has a direct effect on overexploitation of the Socotran resources, on which many people depend. Invasive species are another threat. Several aliens, such as the Indian house crow (Corvus splendens), the Norway rat (Rattus norvegicus), and four species of plants (e.g., Argemone mexicana, Prosopis juliflora) reproduce quickly. The number of invasive species may be relatively low but all need closer monitoring. We expect an increase in rodent populations in lowland areas following expansion of human settlements. Impacts on the terrestrial fauna of Socotra, which evolved in the absence of mammals, are likely to increase. An attempt to introduce rabbits recently was fortunately avoided by the local authorities. Pollution is caused by improper waste management in the major cities, and the use of chemicals in agriculture. With the increase of population and imports of goods, chemical pollution rapidly increases. Agricultural crops have also been imported recently. Their maintenance leads to the use of pesticides, which until now have stayed clear of Socotra.

Uncontrolled development, clearing, and habitat destruction are a major threat for the island. Infrastructure projects, particularly roads, far exceed local needs and have a direct negative impact on the environment, especially in the coastal regions, where exceedingly wide roads take up a large portion of the surface, even crossing through protected areas. The result of uncontrolled planning, road projects especially should be coordinated to minimize negative impacts on biodiversity. The number of tourists on Socotra is increasing exponentially, reaching beyond local capacities. Although all effort is put into increasing visitor awareness and attracting ecotourism, the island still attracts a significant proportion of “beach” tourists. Socotra is not that type of destination, but is rather an island with a natural and cultural heritage, globally recognized for its value and importance. A synergy of threats, together with climate effects, may cause a fast switch in the Socotra ecosystems. The recent nomination as UNESCO World Heritage site is a positive step toward long term conservation and sustainable development of the archipelago and provides an opportunity to address main biodiversity threats. SEE ALSO THE FOLLOWING ARTICLES

Caves, as Islands / Dispersal / Endemism / Ethnobotany / Sustainability / Vicariance FURTHER READING

Cheung, C., and L. Devantier. 2006. Socotra: a natural history of the islands and their people. K. Van Damme, ed. Hong Kong: Odyssey Books. Miller, A. G., and M. Morris. 2004. Ethnoflora of the Soqotra archipelago. Edinburgh, UK: The Royal Botanic Gardens. Sohlman, E. 2004. A bid to save the Galápagos of the Indian Ocean. Science 303: 1753. Thiv, M., T. Mats, K. Norbert, and L. H. Peter. 2006. Eritreo-Arabian affinities of the Socotran flora as revealed from the molecular phylogeny of Aerva (Amaranthaceae). Systematic Botany 3: 560–570.

SOLOMON ISLANDS, BIOLOGY ORLO C. STEELE University of Hawaii, Hilo

Within the South Pacific Island nation of the Solomon Islands there are hundreds of islands, ranging from large high volcanic islands to low atolls, with a total of 4023 km of coastline and an economic exclusive zone of 1,340,000

km2. The biodiversity of the Solomon Islands appears to be the richest among the Pacific Island nations, with the exception of Papua New Guinea. However, the collection of specimens and field observations has been limited, and new species are discovered with each new inventory. GEOGRAPHIC SETTING

The Solomon Islands extend 1450 km in a southeast direction from 5° S and 152° E to 12° S and 170° E. Most of the 28,785 km2 of land area is found on six large high volcanic islands, the largest being Guadalcanal with 5,310 km2. These six high islands are arranged in a double chain that extends 850 km southeastward from the Papua New Guinea island of Bougainville. The northern chain includes (from west to east) the islands of Choiseul, Santa Isabel, and Malaita, while the southern chain is composed of the New Georgia group in the west and the larger islands of Guadalcanal and Makira (San Cristobal) in the east. The maximum elevations of these islands range from 2331 m on Guadacanal to 795 m on Vella Lavella in the New Georgia group. The Santa Cruz Islands form a second group of medium-sized islands in the southeast part of the country, which are geologically and biologically more related to the Vanuatu islands (formerly known as the New Hebrides). This group is approximately 375 km east of Makira, with its major islands being Nendo, Tinakula, Utupua, and Vanikoro. In addition, the raised limestone islands of Bellona and Rennell are about 250 km to the south of Guadacanal, and the atoll of Otong Java lies approximately 450 km to the north. TERRESTRIAL BIOLOGY Vegetation and Flora

The major vegetation types of the large islands include coastal strand, mangrove forests, freshwater swamp forests and herbaceous wetlands, lowland rain forest, seasonally dry forests and grasslands, and montane rain forest. In addition, cloud forests may form along windward slopes at higher elevations. The dominant vegetation on most of the high islands is lowland rain forest, with grasslands occurring on the northern plains and foothills of Guadacanal. The flora is primarily composed of Southeast Asian elements and pertains to the same floristic province as Bougainville in Papua New Guinea, with the exception of the Santa Cruz Islands, on which the flora is more closely related to the flora of Vanuatu. There have been approximately 3571 species of flowering plants recorded for the country, with 3.5% (125 species) being endemic. The origin of the Solomon Islands flora has been mostly from short-distance migrations from westward islands, which

SOLOMON ISLANDS, BIOLOGY

851

has not fostered high levels of speciation; however, the genus Ficus is an exception, with 35% of the 63 species being endemic. Compared to the islands of Bougainville, the rain forests are relatively species-poor, with the notable absence of the Dipterocarpaceae and Fagaceae, which are important tree families further west. The lowland rainforests are typically dominated by the genera Pometia, Dillenia, Elaeocarpus, Endospermum, Campnosperma, Calophyllum, Terminalia, Canarium, Agathis, Metrosideros, and Sararanga, all of which are found on mainland New Guinea. Montane rain forests can occur as low as 700 m on wet windward slopes but more typically occur above 1000 m. Common tree genera of the the montane forest include Syzygium, Metrosideros, Ardisa, Psychotria, Schefflera, Rhododendron, and Ficus. These mountain trees are typically covered with epiphytes, primarily orchids, with an understory largely composed of pandans, gingers, and bamboo species. In freshwater swamp forests, single tree species such as Campnosperma brevipetiolata, Inocarpus fagifer, and Terminalia brassi may dominate. In contrast to the diversity of upland forests, mangrove forests are relatively rich, with 26 species (43% of the world’s mangroves), which are commonly represented by the genera Rhizophora, Bruguiera, and Lumnitzera (Fig. 1).

species of rats, most of which are large (up to 1 kg) and arboreal. Of the 44 species of bats, 26 are flying foxes (Pteropus sp.) and 18 are smaller insectivorous species. This represents one of the highest diversities of bats and rats in the world, with 26, or 50%, of these species being endemic. There are 17 species of amphibians in the Solomon Islands, all of which are frogs, with three genera and 7 species that are endemic to the country. There are 61 species of native terrestrial reptiles (lizards and snakes); three genera and 25 species are endemic. One notable endemic is the Solomon Islands’ prehensile-tailed skink (Corucia zebrata), which is a very large, arboreal skink that feeds primarily on the leaves of epiphytes. The terrestrial invertebrate fauna is poorly known, with the butterflies being the best described among the arthropods. There are reported to be 130 species of butterflies in the country with 35% of the species endemic. Many of these species are spectacular bird wings (Ornithoptera), which are farmed to supply butterfly collectors. Two of the more prominent and spectacular bird wing species are Ornithoptera allotae and O. victoriae, which depend upon specialized food plants for their reproduction. In addition, the blue emperor swallowtail (Papilio ulysses) is also found in the Solomons. Finally there are reported to be 200–270 species of land snails in the country, but this group requires considerable more study. FRESHWATER BIOLOGY

FIGURE 1 Kolombangara Island, 1981. View of mangrove forests in the

foreground and lowland rainforest ascending the slopes of this coneshaped volcano. Photograph by Dieter Mueller-Dombois.

Fauna

Among the terrestrial fauna of the Solomon Islands, the birds have been the most well studied, with approximately 173 species of resident birds and 50 species of seabirds recorded. This represents the highest avifauna diversity of all South Pacific nations. The Solomon Islands show very high levels of bird speciation, with 44% of the species and 38% of the subspecies being endemic. Three notable endemics are the fearful owl (Nesasio solomonenis), the Solomons sea-eagle (Haliaeetus sanfordi), and the megapode (Megapodius freycinet). The native terrestrial mammals of the Solomon Islands are composed entirely of bats and rats. There are eight

852

SOLOMON ISLANDS, BIOLOGY

The freshwater streams of the Solomon Islands have a diverse assemblage of catadromous fish, molluscs (Neritidae), and crustaceans (Atyidae and Palaemonidae). The icthyofauna include several species of gobies (Gobiidae), sleeping gobies (Eleotridae), river mullet (Mugilidae), and freshwater eels (Anguillidae). Because these organisms are of marine origin, they are capable of tolerating a wide range of saltwater concentrations, with the number of species decreasing further inland. In addition, there are a large number of aquatic insects that live in streams for all or part of their life cycle. Notably abundant are the dragonflies and damselflies (Odonata), caddisflies (Trichoptera), and black flies (Simuliidae). MARINE BIOLOGY Corals and Reef Fish

The Solomon Islands have a very high diversity of coral species and are within the “Coral Triangle” of the IndoPacific, which has the highest diversity of coral species worldwide. Based on recent field surveys, there are at least 485 species of corals in 76 genera found in the Solomon

Islands which is second only to Indonesia. This high diversity is thought to be due to a wide range of bathometric and current regimes. There have been 1019 nearshore reef fish species reported from 2–60-m depth in the Solomon Islands in 348 genera and 82 families. This diversity is very high and places the Solomon Islands in the top ten most diverse reef fish areas of the world. Single reef sites may contain over 200 species of reef fish, with the highest recorded on the island of Gizo in the New Georgia Province. Sixty percent of the reef fish are found in ten families, with gobies, damselfishes, and wrasses being the richest and most abundant groups. Because of the broad dispersal capacities of these fish, there are only two endemic species of reef fish in the Solomon Islands (Fig. 2).

(Dugong dugong) is also found in the shallow waters of lagoons and estuaries, where it feeds on sea grasses. Macroinvertebrates

As with their terrestrial counterparts, not much is known of the marine invertebrates, with the exception of those of commercial value. There are 19 species of sea cucumbers (holothurans), several of which are marketed as “bêchede-mer.” Pearl oysters are also important commercially, of which there are three species: blacklip (Pinctada margaritifera), goldlip (P. maxima), and brownlip (Pteria penguine). There are six species of giant clams, with Tridacna maxima, T. squamosa, and T. crocea being the most abundant. Other commercially important molluscs are the green marine snail (Turbo marmoratus) and trochus shell (Trochus niloticus). The spiny Pacific lobster (Panulirus penicillatus) and painted crayfish (P. versicolor) are harvested for food locally, along with a number of other reef and estuary crustacean species. HUMAN IMPACTS

FIGURE 2 White-bonnet clownfish, Amphiprion leucokranos, 2004.

This species is restricted to the western central Pacific and is associated with anemones. Photograph by Gary Allen.

Reptiles

There are five species of sea turtles in the Solomon Islands, namely the hawksbill, green, leatherback, loggerhead, and Olive Ridely. Five species of sea snakes are indigenous to the country, and there is one endemic species of sea krait (Laticaudata crockeri), found only in Lake Te-Nggano on Rennell Island. There is also one species of saltwater crocodile (Crocodylus porosus), which also inhabits brackish streams.

The larger Solomon Islands have been occupied by Melanesians for approximately 30,000 years, with the smaller outlying islands of Ontong Java, Bellona, Rennell, and Tikopia populated by Polynesians approximately 3500 years ago. Both of these groups brought with them their traditional crops and food animals and probably had relatively minor environmental impact as they lived in small villages and practiced subsistence agriculture and fishing (Fig. 3). In the late eighteenth century European and American whalers arrived, followed by other foreign traders in sandalwood, bêche de -mer, trochus and pearl shell. These new industries certainly affected local populations but had limited negative impacts on entire ecosystems within the archipelago.

Mammals

A total of eleven species of cetaceans in nine genera and four families have been observed in the waters of the Solomon Islands. Most common of these are the spinner dolphin, the pan-tropical dolphin, and the common bottlenose dolphin. Other dolphins also reported are the Indo-Pacific bottlenose, Risso’s, and rough-toothed dolphins and the orca. In addition the following whales have been reported: short-finned pilot whale, Mesoplodon beaked whale, blue whale, and sperm whale. The sea cow

FIGURE 3 Typical coastal village in the Solomon Islands, 2004. Houses

are made from locally available materials, with subsistence agriculture and fishing practiced nearby. Photograph by Emre Turak.

SOLOMON ISLANDS, BIOLOGY

853

Adverse human impacts on the environment have increased dramatically within the last 50 years. The major threats to biodiversity are habitat destruction and fragmentation, overexploitation, and introduced species. Although 77.6% of the Solomon Islands land area is reported to still be under forest cover (FAO 2005), extensive areas of lowland rain forests have been clear-cut for timber export. These cleared areas are then often converted into monoculture oil palm and copra plantations. This has resulted in contamination of freshwater and marine environments from increased sedimentation and chemical runoff. In addition, there has been increased pressure placed on lowland forests from expanding village gardens. The population of the Solomon Islands in 2007 was reported to be 566,842 and is growing at annual rate of 3.5%, which is one of the highest growth rates worldwide. Several species are currently overexploited for foreign markets such as crocodiles, sea turtles, pearl oysters, trochus shell, bêche de mer, and green marine snails. Finally, introduced invasive species such as pigs, cats, rats, and little red fire ants (Wasmannia auropunctata) have had devastating impacts in lowland rain forests. Feral cats, in particular, have nearly wiped out several species of native rats on Guadalcanal, and little red fire ants have locally reduced arthropod biodiversity. Because of these adverse impacts, 16 higher plants, 20 mammals, 23 resident birds, four reptiles, and two fish are threatened with extinction (IUCN 2002). SEE ALSO THE FOLLOWING ARTICLES

Coral / Freshwater Habitats / New Guinea, Biology / Solomon Islands, Geology / Vanuatu

Whitmore, T. C. 1969. The land flora: geography of the flowering plants. Philosophical Transactions of the Royal Society of London B 255: 5499–5566.

SOLOMON ISLANDS, GEOLOGY HUGH L. DAVIES University of Papua New Guinea

The Solomon Islands are part of the Outer Melanesian Arc, a discontinuous chain of islands that stretches from the Bismarck Archipelago in the northwest to Fiji and Tonga in the southeast. The islands lie within latitudes 5–12° S. PHYSIOGRAPHY

The Solomon Islands comprise more than 1000 generally mountainous islands that are distributed in two geographic areas. Most islands are in the western and central area, where they form a 1200-km-long northwest-trending double chain, founded upon a basement ridge (Fig. 1). These islands include Bougainville and Buka, which are politically a part of Papua New Guinea. Islands of the Santa Cruz Group and the Outer Eastern Islands are located 300 km to the east, at the northern end of the Vanuatu island chain. The western and central islands and the basement ridge are bounded on both sides by deep-sea trenches. The southwestern trench (Makira or South Solomons Trench, Fig. 2) partly adjoins the young ocean crust of

FURTHER READING

Green, A., P. Lokani, W. Atu, P. Ramohia, P. Thomas, and J. Almany, eds. 2006. Solomon Islands Marine Assessment: technical report of the survey conducted May 13 to June 17, 2004. TNC Pacific Islands Country Report No. 1/06. Keast, A., and S. E. Miller, eds. 1996. The origin and evolution of Pacific Island biotas, New Guinea to eastern Polynesia: patterns and processes. Amsterdam: SPB Academic. Leary, T. 1991. Survey of wildlife management in the Solomon Islands. SPREP Project PA 17. Report prepared from Solomon Islands Government, South Pacific Regional Environment Programme and TRAFFIC Oceania. Mayer, E., and J. M. Diamond. 2001. The birds of northern Melanesia. Cambridge, MA: Harvard University Press. McCoy, M. 2006. Reptiles of the Solomon Islands. Sofia and Moscow: Pensoft Publishers. Mueller-Dombois, D., and F. R. Fosberg. 1998. Vegetation of the tropical Pacific islands. New York: Springer-Verlag. Peake, T. F. 1969. Patterns in the distribution of Melanesian land Mollusca. Philosophical Transactions of the Royal Society of London B 255: 235–306. Randall, J. E., G. R. Allen, and R. C. Steene. 1990. Fishes of the Great Barrier Reef and the Coral Sea. Bathurst, Australia: Crawford House Press.

854

SOLOMON ISLANDS, GEOLOGY

FIGURE 1 Relief map of the Solomon Islands (sea floor topography

from Smith and Sandwell (1997); http://topex.ucsd.edu/marine_topo/ mar_topo.html). Spreading ridges and transforms shown in the Woodlark Basin are from B. Taylor, University of Hawaii, unpublished data. OJP = Ontong Java Plateau, SG = Santa Cruz Group, WB = Woodlark Basin.

the Woodlark Basin; in this sector the trench is poorly defined. The northeastern trench (Kiliniailau and North Solomons trenches) adjoins the Ontong Java Plateau, a vast submarine plateau that stands 2500 m above the surrounding sea floor. Atolls occur atop volcanic peaks that rise from the plateau. The climate of the Solomon Islands is wet tropical and dominated by a northwest monsoon in December to March and southeast trade winds in May to October. The capital city of the independent state of Solomon Islands is Honiara on Guadalcanal Island. The population of the islands is approximately 600,000 and is mostly Melanesian. GEOLOGY

The geology of the islands has been described in terms of three provinces: a northeastern province of Cretaceous– Paleocene oceanic basalts and pelagic sediments; a central province of more complex geology that includes metamorphic and ultramafic rocks and Eocene to Early Miocene arc-type volcanics; and a southwestern province of Late Miocene to Quaternary volcanic islands. The islands of the northeastern province are Malaita, Ulawa, a northeastern part of Makira, and the northeastern flank of Santa Isabel. On these islands Cretaceous and Paleocene submarine basalts with intercalated pelagic limestone and mudstone are overlain by younger sediments that include terrigenous material. The Cretaceous basalts on Malaita are 3.5 km thick. Those on Makira are metamorphosed. The basalts and pelagic sediments originated as part of the Ontong Java submarine plateau. On Malaita an intrusion of alnoite (a silica-poor mafic rock) is of interest because it has brought to the surface fragments of unusual mantle rocks, including garnet peridotite. The age of the intrusion is 34 million years (Oligocene). The islands of the central province are Choiseul, the southwestern part of Santa Isabel, the Florida Islands, Guadalcanal, and Makira (San Cristobal). Basement on these islands comprises Cretaceous basalt, greenschistand amphibolite-facies mafic schists, and ultramafic rocks; the metamorphic rocks have radiometric ages of around 50 million years, Early to Middle Eocene. Basement is intruded by volcano-related dioritic stocks and is unconformably overlain by arc-type volcanic rocks of Oligocene and Early Miocene age and some possibly as old as Late Eocene. A thick sequence of Miocene to Pliocene clastic sediments covers much of Guadalcanal Island. The southwestern province comprises Late Miocene to Holocene arc-type volcanic rocks and associated sediments and intrusive rocks. This association extends from Bougain-

FIGURE 2 Geological map of the central and western Solomon Islands.

Courtesy of Davies et al. (2005).

ville and the Shortland Islands in the northwest to the New Georgia Group and the near end of Guadalcanal Island in the southeast. There is an outlier of Pliocene volcanic and intrusive rocks in central Guadalcanal, at Gold Ridge. Bougainville Island combines characteristics of the central and southwestern provinces, though with minor differences in age. A basement of Middle to Late Eocene and Middle to Late Miocene arc-type volcanic rocks is partly overlain by Miocene limestone and Plio-Quaternary volcanoes and is intruded by Pliocene dioritic stocks. The islands of the Santa Cruz Group, 300 km east of Makira, comprise Miocene to Holocene volcanic rocks and sediments. They include an active volcano on Tinakula Island. ONTONG JAVA PLATEAU

The Ontong Java Plateau comprises 4–5 × 107 km3 of basaltic lava flows. Whereas normal ocean crust is 7–10 km thick, the crust beneath the plateau is more than 30 km thick. The basaltic lavas appear to have been emplaced in one major magmatic event over a period of less than 7 million years at around 122 million years ago (Early Cretaceous, Aptian) and are chemically identical to the basalts exposed on Malaita Island. They are overlain by 1 km of pelagic sediments. PLATE BOUNDARIES AND SEISMICITY

The islands are located on the boundary between the westnorthwest-moving Pacific Plate and the north-northeastmoving Woodlark and Australian plates. The net motion at the boundary is a convergence of the order of 110 mm/yr in a northeast–southwest direction. Most convergence is

SOLOMON ISLANDS, GEOLOGY

855

accommodated on the Makira Trench, but some is taken up by slow subduction at the North Solomons Trench, and some probably by deformation of crustal rocks. Shallow earthquakes coincide with the surface trace of the Makira Trench and the Woodlark Basin spreading ridges and transforms (Fig. 3). Intermediate and deep earthquakes are aligned southeastward from beneath Bougainville and indicate the existence of a steeply dipping subducted slab beneath Bougainville and beneath the open water between the two chains of islands. On April 2, 2007, a magn. 8.1 shallow earthquake located near Gizo in the New Georgia Group caused significant damage. A tsunami that followed 5 minutes after the earthquake damaged coastal villages in the New Georgia Group and as far away as the south coast of Choiseul Island. The earthquake and tsunami caused the loss of 52 lives.

The next activity was arc-type volcanism in the Eocene as a result of southwestward subduction of Pacific plate oceanic lithosphere at the Kilinailau Trench. Volcanic material accumulated on the seafloor to the point where volcanic islands were formed. Metamorphosed basalts from the Eocene subduction system and ultramafic rocks from the earth’s mantle were then exhumed (brought to the earth’s surface), where they became the platform upon which further arc-type volcanic rocks and sediments were deposited, in the Oligocene. We do not know the reason for exhumation but can speculate that it was caused by a change in plate motion resulting in an interval of crustal extension. The Oligocene and Early Miocene volcanics were generated by continuing southwestward subduction at the Kilinailau Trench. In the Middle Miocene, continuing subduction of the Pacific plate brought the Ontong Java Plateau to the Kilinailau Trench. The thick lithosphere of the plateau could not be subducted. The result was delamination and westward thrusting of the upper part of the plateau. Basalts from the upper part of the plateau were carried on thrust faults southwestward to the present location of Malaita and the other islands of the northeastern province. As a result of the emplacement of the thrust sheets, the trace of the Kilinailau Trench stepped northeastward to its present location at the North Solomons Trench. The remaining lower layers of Ontong Java Plateau continued to be slowly and steeply subducted. Because the subduction of Pacific plate was halted, or almost so, the convergence between the Pacific and Australian plates was taken up by development of a new subduction system on the southwestern flank of the islands. This required northeastward subduction of the Australian plate and the development of the Makira Trench. Volcanism associated with the Makira Trench began 8 or 10 million years ago. The Woodlark Basin developed by seafloor spreading beginning at 6 million years ago or possibly earlier.

GEOLOGICAL EVOLUTION

ECONOMIC ASPECTS

The Solomon Islands have been constructed by sea floor volcanism with no influence or input from continental crustal material. The earliest volcanism was the prolific outpouring of great volumes of basalt on to the ocean floor at the time of the development of the Ontong Java Plateau, at around 122 Ma. The volcanic activity is thought to have been associated with a mantle plume. The plateau developed at a great distance to the southeast of its present location. The plateau is thought to have remained submerged throughout the time of its development and subsequently.

A major porphyry copper–gold mine was opened at Panguna on Bougainville Island in 1972 but was closed in 1989 on account of civil unrest. A smaller gold mine at Gold Ridge on Guadalcanal opened in 1997 and closed in 2000 due to civil unrest elsewhere on the island. Alluvial gold is mined on a small scale on both Bougainville and Guadalcanal islands.

FIGURE 3 Earthquakes of magnitude 6 or greater, 1963–2007. Colors

indicate depth below Earth’s surface as follows: red, less than 50 km depth; yellow, 50–100 km; green, 100–200 km; pale blue, 200–300 km; dark blue, 300–400 km; purple, 400–500 km. The shallow earthquakes coincide with plate boundaries at the earth’s surface. The deeper earthquakes are generated by movement of the subducted slabs. Map by Emile Okal.

856

SOLOMON ISLANDS, GEOLOGY

SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Lava and Ash / Solomon Islands, Biology / Tsunamis / Volcanic Islands

FURTHER READING

Coleman, P. J. 1965. Stratigraphical and structural notes on the British Solomon Islands with reference to the first geological map. British Solomon Islands Geological Record (1959–1962) 2: 17–31. Hughes, G. W., P. M. Craig, and R.A. Dennis. 1981. Geology of the outer Eastern Islands. Geological Survey Division Solomon Islands Bulletin 4. Kroenke, L. W. 1984. Solomon Islands: San Cristobal to Bougainville and Buka, in Cenozoic tectonic development of the southwest Pacific. L. W. Kroenke, ed. Technical Bulletin 6. Suva, Fiji: Committee for Coordination of Joint Prospecting for Mineral Resources in South Pacific Offshore Areas, ch 4. Mahoney, J., G. Fitton, P. Wallace, and the Leg 192 Scientific Party. 2001. ODP Leg 192: basement drilling on the Ontong Java Plateau. JOIDES Journal 27.2: 2–6 and covers. Reagan, A. J., and H. M. Griffin. 2005. Bougainville before the conflict. Canberra: Pandanus Books/ANU. Vedder, J. G., and T. S. Bruns. 1989. The geology and offshore resources of Pacific island arcs: Solomon Islands and Bougainville. Earth Science Series 12. Houston: Circum-Pacific Council for Energy and Mineral Resources. Vedder, J. G., K. S. Pound, and S. Q. Boundy, eds. 1986. The geology and offshore resources of Pacific island arcs: central and western Solomon Islands. Earth Science Series 4. Houston: Circum-Pacific Council for Energy and Mineral Resources. REFERENCES

Davies, H. L., et al. 2005. Geology of Oceania, in Encyclopedia of Geology, vol. 4. R. C. Selley, L. R. M. Cocks, and I. R. Plimer, eds. Oxford: Elsevier, 109–123. Smith, W. H. F., and D. T. Sandwell. 1997. Global seafloor topography from satellite altimetry and ship depth soundings. Science 277: 1957–1962.

SOUTH GEORGIA SEE ATLANTIC REGION

SOUTH SANDWICH ISLANDS SEE ATLANTIC REGION

FIGURE 1 A good example of the species–area relationship is the num-

ber of land plant species on the Galápagos Islands. Top: Map of the Galápagos Islands showing the number of land plant species. Note that the largest island contains the most species and that numbers tend to be higher on large islands than on small islands. However, several devi-

SPECIES–AREA RELATIONSHIP

ations from this pattern exist, such as the “unexpected” high number of species (319) on the medium-sized island at the southern end of the archipelago, suggesting that factors other than area may sometimes be important in determining the number of species on islands. Bottom: Relationship between the number of species and island area. Regres-

DAVID A. SPILLER AND THOMAS W. SCHOENER University of California, Davis

For over a century, ecologists have been captivated by the tendency for number of species within a taxonomic group to increase with island area. This “species–area relationship” has been found for a broad range of organisms in numerous archipelagoes around the world. A partial list of studies demonstrating the relationship includes land plants on the Galápagos (Fig. 1) and Aleutian Islands; insects on

sion line is the least-squares estimate: log(number of species) = 1.46 + 0.33 log(area). Data taken from Hamilton et al. (1964).

the Tuscan Islands and on subantarctic islands; reptiles on islands in the Gulf of California and in the West Indies; birds on the Canary, Solomon, and Aegean Islands; and mammals on islands in the Philippines and North American Great Lakes. Although most studies are of higher organisms, even protozoans and diatoms have shown the relationship. To explain the occurrence of the species–area relationship, ecologists have proposed several hypotheses

S P E C I E S – A R E A R E L AT I O N S H I P

857

on causal mechanisms, which shed light on the processes that structure biological communities. Conservation biologists have applied the relationship to the design of nature reserves. STATISTICAL MODELS

The species-area relationship can be estimated statistically using the power equation S = k Az

(1)

where S = number of species, A = area of island, and k and z are fitted parameters. Logarithmic transformation of both sides of the power equation yields the linear equation log S = log k + z log A

(2)

in which z is the slope of the estimated species–area relationship (i.e., the rate of increase of log S with log A). Because the scales are logarithmic, the z-value is independent of scale, allowing comparisons between different studies even when the units for area are different. A second equation, with S instead of log S in Equation 2, is often used instead. Log-transformed data on number of land plant species and area for the Galápagos Islands (Fig. 1, bottom) show a positive linear relationship with a slope (z-value) of 0.33. Similarly, many other studies have found that the data fit the log-log model with z-values usually ranging from 0.20 to 0.40. Preston developed a mathematical explanation for the prevalence of the linear log-log relationship, with specific assumptions about the distribution of species abundances and other biological processes, in which the z-value is expected to be approximately 0.27. However, a detailed analysis of 100 studies by Connor and McCoy showed that the data often fit other statistical models just as well or better, suggesting that biological interpretation of parameters in statistical models should be made cautiously.

viduals, then the number of species should increase with island area. An analogy would be a bowl of colored marbles with ten marbles of each of ten different colors. The larger the handful of marbles you take, the more different colors you will get on average. This simple explanation may serve as a null model to compare with the other five hypotheses, all of which incorporate biological processes. Larger Populations and Less Extinction on Larger Islands

The hypothesis that populations are larger on larger islands, implying lower extinction rates, is an integral part of the equilibrium theory of island biogeography developed by MacArthur and Wilson in 1964. According to this theory, the number of species present on an island is determined by the dynamic equilibrium between the rate of immigration of species not already on the island from a source pool and the rate of extinction of species already on the island (Fig. 2A). Because larger islands

WHY DOES THE NUMBER OF SPECIES INCREASE WITH ISLAND AREA?

Although the species–area relationship is one of the surest generalizations in ecology, there has been much debate on the causal factors and processes. There are six major hypotheses, as follows.

FIGURE 2 Graphical representations of the MacArthur–Wilson equi-

librium model of island biogeography. (A) The effect of island area;

Random Sampling

Assuming that the individuals on a given island are a random sample from a nearby mainland or other source containing all species, and larger islands contain more indi-

858

S P E C I E S – A R E A R E L AT I O N S H I P

S = equilibrium number of species on a small island, L = equilibrium number of species on a large island, P = number of species in the pool on the mainland (source). (B) The effect of distance of the island from the source of immigration; N = equilibrium number of species on a near island, F = equilibrium number of species on a far island.

contain more individuals (larger population size) in each species, and extinction rate is inversely proportional to population size, extinction rate of species on small islands is higher than on large islands, so the equilibrium number of species is positively related to island area. In 1976 Simberloff provided experimental evidence for this hypothesis by showing that the number of species on mangrove islets decreased when he reduced the area of islets. In addition to the area effect, the equilibrium theory contains a distance effect; immigration rates are higher for islands near the source pool than for those farther away (Fig. 2B). Larger Interception Area and More Immigration on Larger Islands

In addition to the lower extinction rate in the equilibrium model, a larger island may also be a bigger “target” for colonists; this causes the immigration curve of a large island to be higher than that of a small island, thereby increasing the equilibrium for the large island. Higher Habitat Diversity on Larger Islands

Because the number of different types of habitats may increase with island area and different species often occur in different habitats, more species occur on larger islands. Hence, the species-area relationship may be indirect via the positive effect of area on habitat diversity. This hypothesis is most popular among ecologists familiar with the natural histories of the species in their studies. Early on, Watson found that habitat diversity was a better predictor of Aegean bird species number than by area. Very recently, Morrison found that the number of ant species on islets in the Bahamas was predicted by number of land plants species better than was area; in this case, different plant species may serve as different types of habitats or provide different types of food resources for ants. Returning to Galápagos land plants, Hamilton and collaborators showed in 1964 that elevation (a proxy for habitat diversity) was a better predictor of number of species than was area.

rium as in the MacArthur–Wilson model. In addition to the effect of episodic major disturbances, chronic noncatastrophic disturbances, such as wind and salt spray, may affect small islands more than large ones because of the greater perimeter-area ratio on small islands. Only species that can tolerate these harsh conditions could persist on small islands. Greater Speciation on Larger Islands

The larger the island, the more likely that geographic barriers exist that cause isolation between viable subpopulations of a species; this can lead to allopatric speciation. The effect may be geometric, as the opportunity for speciation is itself proportional to species number, causing an especially high species–area slope. Lomolino’s Combined Model

Lomolino proposed in 2000 a general model for the species– area relationship that integrates disturbance and equilibrium dynamics along with speciation on islands. In this model the species–area relationship has three regions (Fig. 3): (1) On small islands, the stochastic effects of disturbances are predominant, making the relationship between number of species and island area variable and unpredictable (an example is given in the later discussion of spiders in the Bahamas). (2) On medium to large islands, the deterministic effects of area on extinction/immigration and habitat diversity predominate, and number of species increases at a decreasing rate as the number of species in the source pool is approached. (3) The largest islands contain barriers isolating species populations, leading to allopatric speciation. Losos and Schluter assessed the speciation component for Anolis lizards of the West Indies. For these relatively poor dispersers, the sharp increase in the species–area slope due to speciation obliterates the flattening portion (region 2) of Lomolino’s model.

Lower Abiotic Disturbance on Larger Islands

The impact of abiotic disturbances, such as hurricanes and tsunamis, may be more devastating on small islands than on large islands, exterminating species mostly on the former and thereby causing the positive species–area relationship. Such an effect could go well beyond the effect of population size on extinction, as equivalently sized populations could be more vulnerable on smaller islands. Whittaker pointed out that such disturbances can affect island communities for decades or even centuries, during which time species would not be at a dynamic equilib-

FIGURE 3 Lomolino’s general species–area model containing three

different regions (see text for explanation). Modified from Lomolino and Weisen (2001).

S P E C I E S – A R E A R E L AT I O N S H I P

859

ECOLOGICAL CORRELATES OF VARIATION IN THE SPECIES–AREA SLOPE

Comparisons among different kinds of organisms have revealed substantial variation in the slope of the species–area relationship. Wright showed that on islands in the West Indies, the slope was greater for non-flying mammals than for bats or birds, suggesting that dispersal ability influences the relationship. A plausible explanation is that islands are more isolated for species with low dispersal ability, so most of them do not occur on small islands, whereas species with high dispersal ability are frequently found on small islands because their immigration rates are high even though they may not persist for long. Schoener suggested that z-values should be greater for species occurring at low density (e.g., territorial carnivores) than for those occurring at high density because the lowdensity species can only persist on larger islands. Holt and collaborators developed a mathematical model predicting that the z-value for species at the top of the food chain is higher than for species in lower trophic levels. SPIDERS ON BAHAMIAN ISLANDS

Studies of web spiders occurring on islands in the Bahamas by Schoener and Spiller illustrate the species–area relationship and address several of the issues discussed above (Fig. 4). Complete censuses of all web spiders occurring over the entire areas of 64 islands were conducted annually over a ten-year period. Number of species (mean over time) was positively correlated with island area for four reasons. First, larger islands tended to have larger populations, which in turn had lower extinction rates than did small populations, increasing their number of species. Second, larger islands have a higher immigration rate: the “target effect.” Third, some species are habitat generalists (e.g., Argiope argentata), occurring in areas with high or low vegetation, whereas others are more specialized (e.g., Gasteracantha cancriformis), living in only high vegetation (Fig. 5). Small islands tend to have only low vegetation, whereas large islands contain areas with both low and high vegetation. Therefore, only

FIGURE 5 (A) Argiope argentata. (B) Gasteracantha cancriformis.

Photographs by D. Spiller.

generalists occur on small islands, whereas both generalists and specialists occur on large islands with more types of habitats. Fourth, tropical storms can affect smaller islands more drastically: During several recent hurricanes, smaller Bahamian islands, which are lower, were completely inundated by high water, apparently killing all spiders. Note that variation in the numbers of spider species on small islands is relatively high as in Lomolino’s model. Another factor that can affect number of species is the presence of lizards, which are major predators of spiders and which occurred on about half of the study islands: Islands with lizards tended to have fewer spider species (Fig. 6). This “lizard effect” is more apparent for larger islands than for smaller islands, and the slope of the species-area relationship is greater for islands without lizards (Fig. 7). Hence, the z-value is greater for islands without lizards, on which spiders occupy a higher level in the food web, as in Holt’s model.

FIGURE 4 Aerial photograph showing a portion of the Bahamian spi-

der study area.

860

S P E C I E S – A R E A R E L AT I O N S H I P

FIGURE 6 Illustration of a lizard eating a spider by G. Dan.

hood, may be increased with a certain amount of reservearea fragmentation. In conclusion, several small preserves distributed over a large or varied region may sometimes be the better conservation strategy. SEE ALSO THE FOLLOWING ARTICLES

Extinction / Fragmentation / Galápagos Islands, Biology / Island Biogeography, Theory of / Spiders FURTHER READING

FIGURE 7 Species–area relationship for Bahamian orb spiders on 37

islands without lizards and 27 islands with lizards. Regression lines for islands without and with lizards are respectively: log(number of species) = -0.85 + 0.41log(area), log(number of species) = -0.28 + 0.16log(area). Unpublished data (T. Schoener and D. Spiller).

Connor, E. F., and E. D. McCoy. 1979. The statistics and biology of the species-area relationship. American Naturalist 113: 791–833. Hamilton, T. H., R. H. Barth, and I. Rubinoff. 1964. The environmental control of insular variation in bird species abundance. Proceedings of the National Academy of Sciences of the United States of America 52: 132–140. Lomolino, M. V., and M. D. Weisen. 2001. Toward a more general species-area relationship. Journal of Biogeography 28: 431–445. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton, NJ: Princeton University Press. Quinn, J. P., and S. P. Harrison. 1988. Effects of habitat fragmentation and isolation on species richness: evidence from biogeographic patterns. Oecologia 75: 132–140. Whittaker, R. J. 1995. Disturbed island ecology. Trends in Ecology and Evolution 10: 421–425.

THE DESIGN OF NATURE RESERVES

In addition to “true islands” (bodies of land surrounded by water), the species–area relationship has been well documented for many types of “island analogues” (patches of suitable habitat for species isolated by unsuitable habitat) on mainlands. Knowledge of the factors that shape the species–area relationship can be used to evaluate different strategies for designing nature reserves that maintain the highest number of species. Of course, the ideal strategy would be to have many huge preserves, but this may not be feasible. Hence, the problem that conservation biologists have pondered is whether one large reserve or several small reserves with the same total area is better. The answer to this question, sometimes referred to as the SLOSS (single large or several small) debate, is not obvious: A single large reserve will have a lower per-species extinction rate than will any smaller reserve, but the more reserves there are, the less chance there will be for a species to disappear simultaneously from all of them. In a review of land plants, insects, and vertebrates on islands, Quinn and Harrison showed that total numbers of species on groups of several small islands were higher than on a comparable area consisting of only one or a few large islands. Among other explanations, they suggest that the several small islands were distributed over a larger region and contained more types of habitats, enabling more species to exist on the entire set of islands. Similarly, genetic diversity within a species, which reduces extinction likeli-

SPIDERS MIQUEL A. ARNEDO University of Barcelona, Spain

The ability to produce silk is a distinctive feature of spiders. Silk-mediated airborne dispersal has allowed spiders to colonize even the most remote archipelagoes. Spiders on islands have served as models for the study of the evolutionary and ecological underpinnings of biodiversity. Because of their generalist predatory habits, introduced spiders may pose a serious threat to islands’ native fauna. DEFINING A SPIDER

Spiders (order Araneae) comprise a megadiverse group of arthropods that includes close to 40,000 species distributed in 109 families. The origin of spiders can be traced back to the Devonian, about 400 million years ago, representing some of the earliest evidence of terrestrial life on Earth. As the dominant non-vertebrate predators in most terrestrial ecosystems, spiders have enormous ecological importance. ACCESSING ISLANDS

Dispersal capabilities and generalist predatory habits make spiders formidable pioneers. Spiders have a unique

SPIDERS

861

method of dispersal called ballooning. During suitable meteorological conditions (high turbulence, vertical updrafts, and relatively weak horizontal winds), spiders climb to a height from which they will hang suspended from a silk thread to facilitate takeoff. Some spiders have evolved a more specialized adaptation, “tiptoeing” or “tipping behavior,” which consists of releasing a silk thread into the air while stretching their legs and raising their abdomens (Fig. 1). Ballooning behavior, which occurs mostly in spiderlings, is triggered by innate responses to food shortage and environmental conditions (i.e., temperature). Aerial dispersal allows spiders to be among the first inhabitants of devastated or newly formed areas, such as volcanic islands. The first known colonist of Krakatau (Indonesia), nine months after the eruption that exterminated all life on the island, was a spider. The 1980 eruption of Mount St. Helens created a large, barren area denudated of any resident arthropods; from 1981 to 1986, ballooning spiders represented over 23% of the windblown arthropod fallout. Native spider faunas on remote islands are generally impoverished and disharmonic because of geographic, mechanical, and ecological constraints upon dispersal. The chances of arriving on an island will ultimately depend on the distance from the source of colonization (Table 1). Only 15 out of the 109 spider families are represented in the native biota of the Hawaiian Islands, located more than 3800 km from the closest mainland. In contrast, the spider fauna of Bioko, a volcanic island situated just 32 km from the African continent, is rich, well-balanced, and shows close affinities to the nearby mainland of Cameroon. Unsurprisingly, spiders with well-known ballooning capabilities such

FIGURE 1 Two male Erigone spiders on a grass seed head preparing

for aerial take-off. The lower one is in a pre-ballooning posture, known as the “tip-toe” position, ready to disperse. Photograph by Andy Reynolds, Dave Bohan, and James Bell of Rothamsted Research.

as the orb-weaving spiders and their relatives (Araneidae, Tetragnathidae) and some non–web-building families (Lycosidae, Salticidae, and Thomisidae) are usually overrepresented among the spider faunas of oceanic islands. Medium-sized and large ground-dwelling spiders have most likely colonized oceanic islands by rafting on logs or other vegetal debris washed by rivers and waves. This seems to be especially true for spiders associated with littoral or near-tidal areas, such as the mygalomorph genus Nihoa, which includes species endemic to several Pacific islands from New Guinea to the Hawaiian Islands. Similarly, it is suggested that the woodlouse hunter spider Dysdera, usually found in damp and warm ground habitats,

TABLE 1

Total Diversity and Approximate Levels of Endemism for Spiders on Different Island Groups Distance to

862

Isolation Index

Total

Island Group

Island Type

Ocean

Continent (km)

Area (km2)

(minimum)

Altitude (m)

% Endemism

Species

Ascension Azores Balearic Islands Bioko Canary Islands Cape Verde Galapagos Islands Hawaiian Islands Kuril Islands Madagascar Madeira New Caledonia New Guinea Principe Sao Tome Selvages Tasmania

Oceanic Oceanic Continental Oceanic Oceanic Oceanic Oceanic Oceanic Oceanic Continental Oceanic Continental Continental Oceanic Oceanic Oceanic Continental

Atlantic Atlantic Atlantic Atlantic Atlantic Atlantic Pacific Pacific Pacific Indic Atlantic Pacific Pacific Atlantic Atlantic Atlantic Pacific

1700 1300 85 32 100 620 965 3000 5 370 560 1200 155 220 225 375 200

97 2,333 5,015 2,017 2,007.8 4,033 7,845 16,636 10,291.7 587,713.3 749.4 16,648.4 785,753 148.5 854.8 2.73 65,021.8

119 74 28 17 22 53 60 119 7 58 66 88 37 39 39 35

859 2,351 1,445 3,008 3,718 2,829 1,707 4,169 2,339 2,876 1,861 1,618 5,030 948 2,024 163 1,617

9 19 10 36 63 45 60 60 0 85 37 90 60 47 60 10 60

43 121 278 50 473 105 146 262 427 459 150 226 618 30 56 33 215

SPIDERS

has colonized the Canary Islands from the neighboring northeastern African coast by transporting itself on floating islands. In contrast to oceanic islands, the presence of spiders on continental or fragment islands can be attributed to vicariance in response to a changing geography rather than chance dispersal. Plate tectonics and eustatic sealevel changes have isolated spider communities formerly continuously distributed along large continental areas. For instance, the occurrence of the tree trap-door spiders of the family Migidae in South America, Africa, Australia, Madagascar, New Zealand, and New Caledonia illustrates the former connection of these land masses in the Gondwana supercontinent, which started gradual sundering about 165 million years ago. The major islands of the western Mediterranean—Corsica, Sardinia, and the Balearic Islands—are continental terranes that drifted toward their present-day location following retreat from their original position on the eastern Iberian Peninsula, about 30 million years ago. The temporal sequence of species formation in the genus Parachtes (Dysderidae), endemic to the region, closely follows the geological sequence of separation of the main terranes, suggesting that their present distribution was determined by the disjunction of the islands. SPIDERS ON OTHER ISOLATED SYSTEMS

Mountaintops and caves are among the continental areas effectively isolated by ecological barriers that show a more characteristic and richer spider fauna. Montane populations of the jumping spider Habronattus pugillis (Salticidae) became isolated on the “sky islands” of southeastern Arizona as a result of climatic changes, displaying striking amounts of phenotypic divergence between mountaintops. Similarly, species of the woodlouse-hunter genus Harpactocrates (Dysderidae) show non-overlapping distributions across major cordilleras in the Iberian Peninsula and the Alps. Many spiders have evolved striking features as a result of the adaptation to caves, including eye reduction, depigmentation, and appendage elongation. Spider families with the largest representation of troglobitic species include Pholcidae and Telemidae, worldwide; Agelenidae, Dysderidae (Fig. 2), Leptonetidae, Linyphiidae, and Nesticidae in the Holartic region; and Ochyroceratidae in the Southern Hemisphere. EVOLUTION, ECOLOGICAL ADAPTATIONS, AND COMMUNITY ASSEMBLY

Spiders that have undergone adaptive radiations on islands have served as models for examining the ecological and evolutionary processes underpinning island bio-

FIGURE 2 Dysdera unguimmanis, a troglobitic spider from the lava

tubes of the island of Tenerife, Canary Islands, illustrating some of the adaptations to cave life in spiders: eye reduction, depigmentation, and appendage elongation. Photograph by Pedro Oromí.

diversity. The Hawaiian Tetragnatha, with 37 described and over 55 estimated species, and the Canarian woodlouse-hunter spiders Dysdera, with 49 endemic species, rank at the top of the most species-rich spider lineages on islands. Other examples of island genera with a dozen or more closely related endemic species include the crab spiders Mecaphesa (Thomisidae); the cobweb spiders Theridion (Theridiidae); the Orsonwelles (Linyphiidae) and Lycosa wolf spiders (Lycosidae) in the Hawaiian Islands; the genera Spermophorides and Pholcus (Pholcidae); the wolf-spider Alopecosa; the wall spider Oecobius (Oecobiidae); and the ground-spider Scotognapha (Gnaphosidae) in the Canary Islands. Ecological adaptation seems to have played a marginal role in the diversification of some of these lineages. For example, the major force driving speciation in the Hawaiian linyphiid genus Orsonwelles has been population isolation associated with both island hopping and vicariance caused by erosion of volcanic ridges. Similarly, the diversification history of the Hawaiian jumping spiders Havaika has been shaped mostly by the successive and independent colonization down the island chain of two lineages that diverged in size and genitalic features early in the evolution of the group. Additionally, inter-island colonization within the archipelago has also play a key role in the diversification of the sixeyed pholcid Spermophorides in the Canary Islands, where endemic species from the same island have non-overlapping distributions. In other groups, however, local diversification has likely been triggered by ecological factors. Endemic Hawaiian Tetragnatha (Tetragnathidae), living in the same sites, are represented by orb weavers that construct

SPIDERS

863

webs with different architectures together with cursorial spiny-leg species that differ in color and size. Similarly, Dysdera species co-occurring in the Canary Islands show remarkable differences in body size and chelicera shape. In the three former examples, phylogenetic analyses show that coexisting species with divergent ecological traits are frequently each other’s closest relatives. Although this observation may suggest sympatric speciation, this may not be necessarily the case: Morphological differentiation may have evolved as a result of intraspecific competition following secondary contact of reproductively isolated allopatric populations. In this context, phylogeographic studies have revealed that populations of two species of Hawaiian Tetragnatha, as well as those of the Canarian species Dysdera lancerotensis, are subjected to

FIGURE 4 Orsonwelles macbeth from Molokai, Hawaiian Islands: a

remarkable example of island gigantisms in spiders. This genus belongs to the family Linyphiidae also known as dwarf spiders. Photograph by Gustavo Hormiga.

A

T. mohihi 5.1 myrs

T. kauaiensis

D

Koolaus NW Waikamoi T. tantalus

T. pilosa KAUA’I

C

B

T. quasimodo O’AHU 2.6 myrs 3.7 myrs Waianaes T. kukuiki

T. quasimodo

Waikamoi (except NW corner)

T. kikokiko T. brevignatha T. restricta T. quasimodo

T. waikamoi

T. quasimodo T. kamakou Kipahulu Valley

MOLOKA’I

MAUI 1.9 myrs 1.3 myrs T. polychromata LANA’I 1.3 myrs 0.8 myrs T. perreirai

T. restricta T. macracantha T. quasimodo

T. waikamoi

HAWAI’I

T. kamakou

T. restricta

T. brevignatha

0.4 myrs T. quasimodo

T. kamakou

T. quasimodo 0.4 myrs

0.4 myrs 0.1 myrs 0.2 myrs T. obscura

T. kukuhaa T. quasimodo

T. brevignatha

T. brevignatha

T. quasimodo

T. anuenue T. brevignatha T. quasimodo

FIGURE 3 Examples of the four ecomorphs of spiny-leg Tetragnatha

species endemic to the Hawaiian Islands: (A) green, T. waikamoi, (B) maroon, T. kamakou, (C) large brown, T. kukuhaa, and (D) small brown, T. quasimodo. Distribution of the four ecomorphs across the main Hawaiian volcanoes (gray circles), with age indicated in millions of years (myrs). Each section of a pie represents a different ecomorph (color codes as shown in the top pictures) whenever a morph is present at a site. Never are two species that share the same ecomorph found in the same locality. Phylogenetic evidence indicates that ecomorphs have evolved largely independently within islands. From Gillespie (2004). Reprinted with permission from the American Association for the Advancement of Science.

864

SPIDERS

increased genetic subdivision as a result of fragmentation by lava flows. This may ultimately drive allopatric speciation. The adaptive radiation of habitat-associated, polychromatic Hawaiian spiny-leg Tetragnatha, coupled with the snapshots of the evolutionary process provided by the chronological arrangement of the islands, has illustrated the deterministic nature of community assembly (Fig. 3). That is, at any given locality, similar sets of ecomorphs arise. This leads to a dynamic species assembly: a maximum number of species occurs in communities of intermediate age, whereas first stages are characterized by ecological release probably as a response to reduced interspecific competition. Ecological release following island colonization has been reported in the ray spider Wendilgarda galapagensis (Theridiosomatidae), endemic to the island of Cocos, which shows greater variation in terms of habitat selection and web design and construction on islands than its continental counterparts. Ecological release through decreasing levels of interspecific competition may also explain the evolution of gigantism in linyphiid spiders of the genera Orsonwelles (Fig. 4) in the Hawaiian Islands or Laminacauda in the Juan Fernández Islands. Unlike the former examples, daddy-longlegs spiders of the genus Pholcus, endemic to Madeira and the Canary Islands, do not show any somatic or ecological differentiation. In these taxa, diagnostic characters are confined to structures involved in copulation, suggesting that diversification in this group has been driven by sexual selection. The main ongoing evolutionary processes associated with continental islands are relictualization and formation of paleoendemics. These processes are evident in several spider lineages occurring in Madagascar, such

as the family Migidae. In this case, the three Malagasy genera of the family form a lineage, the closest relatives of which are taxa occurring in Australia, New Zealand, South America, and Africa. This conforms to the biogeographic predictions for the dismantling of the Gondwana supercontinent.

SPITSBERGEN MARIA WŁODARSKA-KOWALCZUK Institute of Oceanology, Polish Academy of Sciences, Sopot, Poland

CONSERVATION ISSUES

Exotic pest introductions pose a major threat to unique island faunas, including spiders. The biota of the Hawaiian Islands and southeastern Polynesia evolved without ants or other social hymenopterans. Over the last century, more than 40 species of ants have been introduced to Hawaii. Native spiders lack effective defense mechanisms against direct predation or competition from these aggressive predators. Three species have been shown to be particularly damaging in Hawaii: the argentine ant Linepithema humile, the big-headed ant Pheidole megacephala, and the long-legged ant Anoplolepis longipes. The little fire ant Wasmannia auropunctata threatens the spider biota on many of the islands of the central and southern Pacific. Spiders themselves are generalist predators and hence may become aggressive invaders. As an example, in Hawaii, the introduced araneid Gasteracantha mammosa is considered a nuisance to farmers, to residences, and potentially to endangered Hawaiian arthropods. Unfortunately, a complete assessment of the conservation status of many island native spiders is hampered by a lack of taxonomic knowledge. As a result, the only island endemic spider currently listed in the International Union for Conservation of Nature (IUCN) Red List is the Kauai cave wolf spider (Adelocosa anops), and its inclusion was motivated by the deterioration of its cave habitat. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Ants / Caves, as Islands / Continental Islands / Dispersal / Rafting / Sky Islands / Vicariance FURTHER READING

Australasian Arachnological Society. http://www.australasian-arachnology .org/ Gillespie, R. G. 2004. Community assembly through adaptive radiation in Hawaiian spiders. Science 303: 356–359. Gillespie, R. G. 2005. The ecology and evolution of Hawaiian spider communities. American Scientist 93: 122–131. Gillespie, R. G., and G. K. Roderick. 2002. Arthropods on islands: colonization, speciation and conservation. Annual Review of Ecology and Systematics 47: 595–632. Griswold, C. E. 2003. Araneae, spiders, in The natural history of Madagascar. S. Goodman and J. Benstead, eds. Chicago: University of Chicago Press, 579–587. Jocqué, R., and A. S. Dipepenaar-Schoeman. 2006. Spider families of the world. Tervuren, Belgium: Royal Museum for Central Africa.

Spitsbergen is the largest island (38,000 km2) of the Svalbard archipelago located at the northeastern edge of the Barents Sea shelf. CLIMATE

Despite its high Arctic location (74 to 81º N), the island experiences relatively mild weather conditions, with average air temperatures ranging from –12 ºC in February to 5 ºC in July. Heat is transported to these high latitudes by the Atlantic waters of the west Spitsbergen current, a distant branch of the Gulf Stream. The Greenland Sea waters off the west coast are ice-free throughout most of the year. The Barents Sea polar water masses occurring north and east of the island are colder; here the sea is more often covered with ice, and air temperatures on the east coast are usually a few degrees lower than those on west Spitsbergen. Its location at the front between the cold Arctic and warm Atlantic water masses renders the island susceptible to early signs of global climate change. LANDSCAPE

The Spitsbergen landscape was shaped during repeated Quaternary glaciations; it is dominated by steep mountains with sharply pointed peaks (“Spitsbergen” means “jagged peaks”) and large fjord systems. The two largest fjords—Isfjorden on the west and Wijdefjorden on the north—nearly cut the island in two. The broad flat plains, which measure up to 10 km in width, are partially covered by marine deposits and were formed during interglacial periods; these are referred to as “strandflats.” There are also raised marine terraces that occur commonly along the coast. Northeastern Spitsbergen is covered by large ice caps, whereas in the western and southeastern parts of the island, numerous valley glaciers flow from elevated ice fields and often terminate in the sea. Once every several decades the glaciers surge; this means that they make a massive downward movement that can advance the glacier front by several kilometers. Episodic surge events and short-term local fluctuations do not affect the general trend of Spitsbergen glacier retreat that has been observed throughout the last century. This is linked to global climate change—namely, to increases in air temperatures.

SPITSBERGEN

865

FLORA AND FAUNA

Permafrost (ground that is permanently frozen except in summer, when a layer just a few decimeters deep thaws), low temperatures, and a very short growing season (6 to 10 weeks) limit the island’s vegetation to about 170 species of vascular plants that form Arctic tundra communities. In summer, the island hosts huge populations of birds; the fjord cliffs are colonized by hundreds of thousands of auks (Brunnich’s gullemot, Oria lomvia; black guillemot, Cepphus grille; little auk, Alle alle) and gulls (kittiwake, Rissa tridactyla). Although only three species of terrestrial mammals occur on the island—the arctic fox (Alopex lagopolus), the Svalbard reindeer (Rangifer tarandus platyrhynchus), and the polar bear (Ursus maritimus)—there are a number of sea mammals, including walruses (Odobenus rosmarus), ring seals (Phoca hispida), bearded seals (Erignathus barbatus, Fig. 1), and white beluga whales (Delphinapterus leucas). Wildlife is protected in national parks, nature reserves, and bird or plant sanctuaries that together comprise more than 60% of the area of the Svalbard archipelago.

numbers. Ultimately, this led to catastrophic reductions in the natural populations. The main economic activity on the island in the twentieth century was coal mining. The Svalbard Treaty, which was ratified by the international community in 1920, established Norwegian sovereignty throughout the archipelago but also guaranteed rights for all signatory countries to conduct commercial and scientific activities. At present, about 2500 inhabitants reside year-round in a few settlements located on the west coast of the island, with most of the population in the “capital” town of Longyearbyen. Since the end of the twentieth century, the economic significance of mining has gradually declined in favor of tourism and scientific research. SEE ALSO THE FOLLOWING ARTICLES

Arctic Region / Climate Change / Whales and Whaling FURTHER READING

Conway, M. 1906. No man’s land. Cambridge: Cambridge University Press. Hisdal, V. 1998. Svalbard: nature and history. Norsk Polarinstitutt Polarhandbok Number 12. Oslo. Mehlum, F. 1989. Birds and mammals of Svalbard. Norsk Polarinstitutt Polarhandbok Number 5. Oslo. Rønning, O. I. 1996. The flora of Svalbard. Norsk Polarinstitutt Polarhandbok Number 10. Oslo.

SRI LANKA COLIN GROVES Australian National University, Canberra

KELUM MANAMENDRA-ARACHCHI Wildlife Heritage Trust, Colombo, Sri Lanka FIGURE 1 Bearded seals resting on an iceberg in front of a tidal gla-

cier in Kongsfjorden, one of west Spitsbergen’s fjords. Photograph by Wojtek Moskal.

HISTORY

The island was discovered in 1596 by Willem Barentsz, a Dutch Explorer who was leading an expedition in search of the northeast passage to southeast Asia. For the next two centuries, the island and the surrounding Greenland Sea waters were the site of intensive whaling activities. The hunting, led by Dutch and British companies, targeted primarily the Greenland right whale (Balaena mysticetus), but other marine mammals, including walruses and seals as well as cetaceans, were also taken in large

866

SRI LANKA

Sri Lanka is, for its size, the most biologically diverse of all islands. It has three distinctive climatic zones, each with its own characteristic fauna, two of which show only the most distant biological affinities with nearby India. A tropical (5°55'–9°51' N) island nation, Sri Lanka is classed as part of South Asia and is separated from southern India by the narrow (20 km wide) Palk Strait. It is teardrop shaped, with a north–south length of 432 km and a maximum east–west width of 224km. The total area is 65,610 km2. GEOLOGY AND GEOGRAPHY

Sri Lanka is geologically part of the Indian subcontinent, and as such, it was part of the supercontinent of Gond-

wana (along with South America, Africa, Australia, and Antarctica) during much of the Mesozoic (“Age of Dinosaurs”). During the latter part of this period, Gondwana began to break up, and India, with Sri Lanka, moved northward until it collided with Eurasia. After this, fauna of Eurasian origin spread into India and Sri Lanka during successive dispersal events and then differentiated in the variety of habitats found there. The highlands in the center of Sri Lanka rise to 2524 m (at Pidurutalagala); the highlands grade downward into the wide flat coastal plane in the west and southwest, but elsewhere drop down in a series of escarpments. The largest river, the Mahaweli, rises in the central highlands and initially flows west, then circles to the northeast and finally flows east. THE WET ZONE

The Wet Zone is located in the southwestern quarter of the island and extends into the highlands where lowland rain forest grades into lower montane forest (Fig. 1). Temperatures in this zone vary monthly from about 22 to 33 °C, and rainfall is high (2500–5000 mm per annum) and not strongly seasonal. The climax vegetation is evergreen tropical rain forest, dominated by Dipterocarpaceae, whose diversity here is second only to that found in Southeast Asia.

FIGURE 2 Typical vegetation of the dry zone, inside Yala National

Park, extreme southeastern Sri Lanka. The dry zone is in most respects an extension of peninsular India, and the spotted deer (Axis axis), of which a stag is seen in this photo, is one of many species that occur on both sides of the Palk Strait. Photograph by Colin Groves.

ill-defined in the lowlands, and there is quite a wide belt of intermediate climate and mosaic vegetation. THE CLOUD FOREST (OR UPPER MONTANE FOREST) ZONE

In this zone, mostly above 1500 m in altitude though as low as 800 m in places, the temperature fluctuates seasonally between 7 ºC and about 26 °C. Rainfall is as little as 1500–2000 mm, but humidity is enhanced because the low temperatures reduce evaporation, so the landscape is wreathed in moisture-bearing cloud on a daily basis. Low trees (under 10 m, often only 3–4 m high) are interspersed with tussock grassland (Fig. 3), and there are abundant FIGURE 1 Remnant rain forest in the wet zone, at the archaeological

site of Batadomba Lena, near Ratnapura in southwestern Sri Lanka. Very little rain forest vegetation now remains. Photograph by Daniel Rayner.

THE DRY ZONE

This is by far the largest climatic zone, covering the entire northern, eastern, and southeastern lowlands (Fig. 2). The vegetation varies from arid to dry forest formations. Rainfall here is much lower (less than 2000 mm per annum) and is distinctly seasonal, but the temperatures are comparable. The border between dry and wet zones is

FIGURE 3 A view of forest and tussock grassland on Horton Plains,

about 1800 m, central Sri Lanka. Photograph by Colin Groves.

SRI LANKA

867

lichens, mosses, ferns, climbers, and epiphytic orchids. In the forest, Dipterocarps are replaced by Rhododendron, Prunus, Ilex, and Berberis, whose closest relatives are in the Western Ghats, and by species of Syzygium and Eugenia related to plants in the wet zone. The largest cloud forest area, a vaguely triangular zone about 50 km from east to west and extending about the same distance to the north, is in the Central Highlands; this is separated (by the Mahaweli Valley, at only 500 m) from a smaller zone, a 30-km-long north–south trending ridge, on the Dumbara or Knuckles range to the north. The Sinharaja forest to the south of the Central Highlands incorporates the high Rakwana Hills, on the summit of which is a further cloud forest zone; there is a fourth, tiny area of cloud forest at Namunukula just to the east of the Central Highlands. The boundaries of these three zones have fluctuated in the recent past, in phase with the strengthening and weakening of the southwest monsoon. The Horton Plains, in the cloud forest zone of the Central Highlands, were a species-poor, even semiarid, environment from at least 24,000 to 18,500 years ago, becoming semi-humid as the monsoon became stronger, and then eventually becoming hyperhumid about 8700 years ago. Humidity rapidly decreased again until 3600 years ago, when the present extent of the cloud forest became established. There were, however, two short humidity events about 600 and 150 years ago. UNIQUE DIVERSITY IN A SMALL ISLAND

Sri Lanka is only slightly larger than Tasmania, somewhat smaller than Banks Island in the Northwest Territories of Canada, and smaller by at least 1000 km2 than Hispaniola in the Caribbean, Sakhalin in the Russian Far East, Hokkaido in Japan, and Ireland. None of these other islands is even remotely comparable to Sri Lanka in terms of biodiversity or endemism. Yet the Palk Strait is only 10 m deep, and during times of low sea level, most recently during the time of the last glaciation of the temperate zones (approximately 28,000–12,000 years ago), a land bridge 140 km wide joined Sri Lanka to India. The comparatively unproductive dry zone, which lies opposite the Indian mainland and was continuous with the dry country of southeastern India, forms a semiarid barrier protecting the wet zone and cloud forest, the centers of the biological hotspot, from incursions from the South Asian mainland. The wet zone and cloud forest appear to have long been isolated from comparable regions elsewhere. Moreover, the wet zone is dissected by large rivers, notably the Kalu, north and south of which many groups

868

SRI LANKA

have diverged into different species, and some of the wet zone forests, such as the Kanneliya forest, may have been partially isolated from the rest of the tropical rain forest belt over a considerable period of time. The hill forests of the wet zone change with altitude, and the ridges, slopes, and valleys have distinct vegetational assemblages. The existence of three separate cloud forest areas (excluding the tiny, and poorly known, Namunukula cloud forest) further enhances biodiversity. The flora and fauna reflect the variety of dispersal routes by which Sri Lanka has been populated. Probably the oldest stratum survives in the wet zone, the dipterocarp-rich forests of which reflect an affinity with Southeast Asia. The flora of the highlands, especially the cloud forest, has its predominant affinities with the Western Ghats of India, but there are also apparent survivors is from Gondwanan time, such as the Jade Vine, Strongylodon (Fabaceae), which is found also in northern Australia, New Guinea, New Caledonia, and Madagascar, and the spider Diallomus, whose closest relatives are found in Madagascar and the highlands of South America. This article concentrates on a few vertebrate groups, whose affinities are less ancient, but which still illustrate the amazing diversity of this small island. Amphibians

The diversity of amphibians in Sri Lanka is extraordinary (106 species recorded so far); although Sri Lankan species tend to have their closest relatives in the Western Ghats of India, they form distinct clades and have been long separated from their Indian counterparts by the intervening dry zone. Some examples will illustrate the nature and age of this amazing amphibian biodiversity and the ways in which it is divided up zonally. The species of the frog genus Philautus (Rhacophoridae), known as flying frogs, live in eastern Asia and India, but nine species groups (a greater known level of diversity than in any other part of the range of the genus) occur in the cloud forests of Sri Lanka, with most of them being entirely restricted to the cloud forests. Their greatest diversity is in the Central Highlands; some have representatives in the Knuckles Range, and fewer in the Rakwana Hills. Molecular clock dates put the separations between related species in the three areas at anything from 4–6 to 8–12 million years old; the extreme age of these clades is part of a general picture. Another rhacophorid, Polypedates, known as the whipping frog, has two species widespread in the wet zone, a third that is also from the wet zone but is restricted to the Sinharaja forests, a fourth in the cloud forest of the Central Hills, and a fifth—a dry

zone species—that is shared with India; the other species of the genus live in Southeast Asia. The tree-hole frog Ramanella (family Microhylidae) is restricted to Sri Lanka and India. It has four species in Sri Lanka, one widespread in the wet zone up to 1200 m, one in the Central Highlands at about 2000 m, and one restricted to the little-known Kanneliya Forest Reserve in the southeastern part of the wet zone; the fourth is a dry zone species that is shared with India. Finally, there are three complete genera of amphibians that are endemic to the wet and cloud zones of Sri Lanka: the dwarf toad Adenomus, the fang-bearing frog Lankanectes, and the true frog Nannophrys. Reptiles

The biodiversity of reptiles in Sri Lanka is hardly less than that of amphibians. The gecko genus Cyrtodactylus has seven species in Sri Lanka, all in the wet zone and reaching well into the lower montane forests but only regionally into the cloud forest (in the Knuckles Range and Namunukula). Oddly, the genus is not represented in peninsular India; the closest relative of the Sri Lankan group is in far northeastern India, and the genus extends through Southeast Asia to the Solomon Islands. The agamid genus Otocryptis has a species in the dry zone as well as one in the wet zone. Interestingly, these two are sister species, and a South Indian species is sister to the Sri Lankan clade. Two agamid genera, Cophotis and Lyriocephalus, are endemic to Sri Lanka. Cophotis has two species, both restricted to the cloud forest: C. ceylanica in the Central Highlands and the recently described C. dumbarae in the Knuckles Range. Lyriocephalus, the sister genus to Cophotis, has populations in the Knuckles and in the wet zone, the relationships between which have not been fully studied. Finally, 98 snake species are found in Sri Lanka, of which 45 are endemic, including five endemic genera. The majority of endemic species in this case are restricted to the lowland wet zone. Mammals

Early knowledge of the mammals of Sri Lanka was due to Kelaart in the mid-nineteenth century, Phillips in the early twentieth century, and most especially Deraniyagala in the middle years of the twentieth century. These authors were well aware of the zonation of the mammals, which they generally assessed at subspecific level. For example, Phillips distinguished at minimum one subspecies each in the wet zone, dry zone, and cloud forest for the primates Loris tardigradus (slender loris) and (what

is now referred to as) Trachypithecus vetulus (purple-faced leaf monkey), whereas Deraniyagala did the same for the giant squirrel Ratufa macroura. More recent studies have concluded that the distinctions between the taxa in different zones have been underestimated. Groves and his colleagues, for example, separated the small red lorises of the wet zone at specific level from the larger gray dry zone loris—which Groves deemed conspecific with the loris of southern India—and the wet zone and dry zone mouse deer (Moschiola) from one another, though neither is identical to the mouse deer of India. The biodiversity of mammals goes deeper than this. It is likely that most or all of the wet zone mammals are specifically distinct from their relatives, if any, in the dry zone, but the cloud forest contains many endemic genera, including the shrew Solisorex and the murid rodent Srilankamys (both of these appear to have their closest living relatives in Southeast Asia). So far, just one of the endemic genera, Feroculus, the long-clawed shrew, appears to be shared with the corresponding zone in the Western Ghats. FOSSIL HISTORY

Much remains to be discovered about the fossil history of the Sri Lankan fauna, which in the main has been recovered in piecemeal fashion during excavations for gems, especially at Ratnapura. Deraniyagala divided the Pleistocene fossils into an older Hippopotamus stage and a younger Elephas maximus stage, but could not date them given the absence of much stratigraphic control. A Pleistocene lion fossil has long been known, and it has recently been shown that there was also a fossil tiger. The two were apparently not contemporary; the tiger remains date from 16,500 years ago, whereas the lion may have become extinct before (or possibly as a consequence of ) the arrival of modern humans and dates to 37,000 years ago. Two species of rhinoceros, Rhinoceros unicornis (Indian rhinoceros, at present confined to the monsoon forests of northern India and Nepal) and R. sondaicus (Javan rhinoceros, a Southeast Asian rain forest species), are known from the Pleistocene, but whether they were contemporaries is unclear; remains of the latter date to 80,000 years ago. Another species known from the Ratnapura deposits is the gaur, Bos gaurus, a large wild ox that still occurs in mainland South and Southeast Asia. This is of special interest because of literary evidence that it survived in Sri Lanka until the seventeenth century. CONSERVATION

The conservation status of the Sri Lankan fauna and flora is alarming. Of 34 anuran species confirmed to be extinct

SRI LANKA

869

worldwide, 21 were endemic to Sri Lanka. Some 95% of the wet zone forest has been lost, because this is the area of highest human population density; only 750 km2 remains, and not a single national park or other protected area is located there. Many of the animal and plant species that survive there, most of them endemic for reasons already stated here, are endangered. The largest remaining block of lowland rain forest is the Sinharaja World Heritage Area, to the south of the Central Highlands. Matters are little better in the cloud forest; much of the forest has been cleared for tea plantations, and although there are some protected areas, the whole zone is at the mercy of climate change. Over the period 1869–1995, the average annual temperature at Nuwara Eliya (1800 m in altitude) has increased by 1.3 °C, and the average annual rainfall has decreased by 20%. Only the dry zone is reasonably well served by protected areas, including the Yala National Park in the far southeast. SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Frogs / Indian Region / Lizard Radiations / Rodents FURTHER READING

Bambaradeniya, C. N. B., ed. 2006. The fauna of Sri Lanka: status of taxonomy, research and conservation. IUCN Publications. Bossuyt, F., M. Meegaskumbura, N. Beenaerts, D. J. Gower, R. Pethiyagoda, K. Roelants, A. Mannaert, M. Wilkinson, M. M. Bahir, K. Manamendra-Arachchi, P. K. L. Ng, C. J. Schneider, O. V. Oommen, and M. C. Milinkovitch. 2004. Local endemism within the Western Ghats—Sri Lanka biodiversity hotspot. Science 306: 479–481. Pethiyagoda, R., ed. 2005. Raffles Bulletin of Zoology, Supplement 12: special issue on Sri Lanka. Phillips, W. W. A. 1980. Manual of the mammals of Ceylon, 2nd ed. Colombo, Sri Lanka: Wildlife and Nature Protection Society. Werner, W. 2001. Sri Lanka’s magnificent cloud forests. Colombo, Sri Lanka: WHT Publications (Private) Limited.

ST. HELENA PHILIP ASHMOLE AND MYRTLE ASHMOLE Peebles, Scotland, United Kingdom

St. Helena—one of the most isolated inhabited islands in the world—lies at latitude 15°58’ S and longitude 5°43’ W, about 800 km east of the Mid-Atlantic Ridge in the southern Atlantic Ocean. The island played a significant part in the development of biological concepts such as endemism, extinction, and the origins of insular biota. Less than a decade after publication of Charles Darwin’s

870

S T. H E L E N A

The Origin of Species, the botanist Joseph Hooker used his firsthand knowledge of St. Helena in a seminal lecture on insular floras to the British Association for the Advancement of Science, and the chapter on St. Helena in Alfred Russel Wallace’s Island Life (1892) includes an elegant analysis of key factors in the colonization of oceanic islands and the origin and evolution of their distinctive endemic species. In relation to current thinking, the isolated islands of St. Helena and its distant neighbor Ascension provide insights into evolutionary processes distinct from those offered by archipelagoes such as the Galápagos and Hawaiian Islands. SITUATION AND GENERAL CHARACTERISTICS

Africa is the nearest continent to St. Helena, with the coast of Angola lying 1800 km to the east and South America resting 3260 km to the west. Ascension Island, the closest land, lies 1300 km to the northwest. The island lies in a region of low marine productivity, although it is influenced by the Benguela current flowing to the northwest from the coast of southern Africa. The richer waters close inshore probably once supported seals, and there are still numerous dolphins and moderately diverse marine life, although coral reefs are absent. St. Helena has a subtropical oceanic climate with relatively slight seasonal change. The southeast tradewinds, dominant throughout the year, generate much condensation on the mountains. Annual rainfall is about 1 m on the central ridge, but less than a fifth of this in the driest areas. St. Helena is about 17 km long and 10 km wide, with an area of 121.7 km2. It is the eroded summit of a 5000-m conical volcanic pile rising from the floor of the deep ocean. The island emerged into the air some 12–14 million years ago, eventually forming a mountain much higher than the modern central ridge, which still reaches 820 m above sea level. The rocks were laid down by successive activity from two volcanic centers producing long series of eruptions, mainly of relatively liquid basaltic lavas from fissures now preserved as striking dike swarms. The northeastern volcano was active during the emergence of the island and for up to another 3 million years. Much of it has now disappeared, but the north of the island is made up of rocks generated during this phase, comprising basal breccias and the predominant mass of subaerial basaltic lavas and pyroclastics. The southwestern volcano created the greater part of the modern island between 11 and 7 million years ago. Its initial phase produced the lower shield, composed mainly of pyroclastics, the erosion of which formed the great amphitheater of Sandy Bay, open to the southeast.

The main shield, made up primarily of basalt and trachybasalt lava flows, forms most of the southern and western parts of the island. Recent analysis by Ian Baker indicates that about 9 million years ago a huge landslide removed several cubic kilometers of the eastern flank of the older (northeastern) volcano. The resultant depression was rapidly infilled by thick trachybasalt and trachyandesite flows erupting from near the modern peaks. These eastern flows (previously termed the upper shield) formed most of the relatively level areas in the northeast of St. Helena. A late intrusive phase, which occurred about 7.5 million years ago, injected massive intrusions of trachyte and phonolite into the southwestern volcano; some of these are now exposed, forming a series of landmarks. There has been no later volcanic activity, so marine and terrestrial erosion has shaped the modern island. It is ringed by sea cliffs up to 400 m high, interrupted by steep-sided valleys, and is surrounded by a broad and irregular shelf created by wave action. This was largely exposed during glacial episodes in the Pleistocene, sometimes doubling the size of the island. HISTORY

St. Helena was untouched by humans until its discovery by João de Nova in 1502. For one and a half centuries it remained free of resident people but was frequently visited, initially by Portuguese mariners and later also by the Dutch and English. Settlement was organized in 1659 by the English East India Company, and the island became an important staging point on return voyages from Asia. St. Helena came to international prominence in 1815 when Napoleon Bonaparte was exiled there, where he remained until his death in 1821. Later in the nineteenth century, the island was used as a base by the British navy for suppression of the slave trade, and since that time it has been governed from Britain. The resident population had origins in England, West Africa, Malaysia, China, the Maldive Islands, and many other places. Natural resources are few, and the economy of the island is heavily subsidized. Access is only by sea, although an airport is now planned. PLANTS AND ANIMALS

The extreme isolation of St. Helena has ensured that few kinds of land plants and animals reach it naturally; it prevents colonization by poor dispersers such as freshwater fish, amphibians, and terrestrial reptiles. However, the considerable age of the island has led to high rates of endemism among successfully colonizing taxa and to some splitting of lineages after arrival. The affinities of the indigenous biota are overwhelmingly with southern Africa.

Of the 37 endemic flowering plant species (six of which are now extinct), the four species of Commidendrum and the one species of Melanodendron (both Asteraceae) probably arose from a single dispersal event, and the three modern species of Trochetiopsis (Sterculiaceae) derive from one other colonization. Additional stocks are represented by the seven monotypic genera Trimeris (Campanulaceae); Lachanodes, Petrobium, and Pladaroxylon (all Asteraceae); Nesiota (Rhamnaceae); Nesohedyotis (Rubiaceae); and Mellissia (Solanaceae) and by several other groups with one or more endemic species. Half a dozen non-endemic flowering plants are coastal halophytes and are probably native, but almost all of the roughly 250 additional species have been introduced, from many parts of the world. St. Helena also has 30 species of native ferns (about 13 of them endemic) and a rich bryophyte flora. Fossil evidence indicates that ancestors of the modern Trochetiopsis and Lachanodes species, and also the tree fern Dicksonia and other fern genera, were already present some 9 million years ago. Also represented among the fossils are plant groups not found in the modern flora, including two lineages of palms (including Voamniola) and the genus Gunnera (Haloragaceae), which probably became extinct as a result of major volcanism more than 7 million years ago. The only land vertebrates that reached St. Helena naturally were birds. Deposits of fossil bird bones show that the island once supported large colonies of seabirds, which are now reduced to remnants. At least eight species of seabird have been lost to the island, and three of them are now extinct: Bulweria bifax, Pterodroma rupinarum, and Puffinus pacificoides (earlier extinction). Endemic landbirds known only from fossils are the two rails Aphanocrex (ex Atlantisia) podarces and Porzana astrictocarpus, a cuckoo Nannococcyx psix, a hoopoe Upupa antaios, and a dove Dysmoropelia dekarchiskos. Songbirds are not represented in the fossil record. The only surviving endemic bird is the wirebird Charadrius sanctaehelenae. Introduced landbirds comprise two gamebirds, two pigeons, and five passerines. Much of the zoological interest of the island stems from its diverse endemic invertebrates. Around 80 genera and 400 species (~38% of the fauna) of invertebrates are currently recognized as endemic and many species—as in the plants—are highly distinctive descendants from ancient colonizations. Beetles are by far the most diverse group, with about 150 endemic species (~58%) and 32 endemic genera. There have been spectacular radiations in the weevil family Curculionidae (77 endemic species) and the related Anthribidae (27 endemics). Speciation has also been striking in the Lepidoptera, with some 50 endemic moth

S T. H E L E N A

871

species including at least 20 in the genus Opogona (Tineidae). Spiders also have about 50 endemic species (~48%), but these are spread among many families. Other groups showing significant radiations include snails of the family Charopidae, in which the only survivor is the ammonite snail Helenoconcha relicta; bugs of the subfamily Phylinae (family Miridae) with at least ten species; and the homopteran family Cicadellidae with 13 endemic species probably derived from only a few colonizing stocks. The best known endemic invertebrate is the St. Helena giant earwig Labidura herculeana (Labiduridae). This is the world’s largest dermapteran, with a maximum length of over 80 mm, but it has not been seen alive for half a century. ECOLOGY, CONSERVATION, AND ECOLOGICAL RESTORATION

The original habitats of St. Helena were diverse. The fringes of the island were either semi-desert or were occupied by scrubwood formed by several endemic shrubs. Inland from these were ebony gumwood thicket and dry gumwood woodland, with the latter grading upward into moist gumwood woodland and then into cabbage tree woodland containing half a dozen tree species. The highest part of the central ridge (above 700 m) was once covered by a cloud forest of tree fern thicket with the most drought-intolerant endemic trees and shrubs. The moist mountain habitats still support a high proportion of the rich fern and bryophyte flora of the island. Invertebrate diversity is also highest in the forests on the central ridge, but surviving fragments of dry gumwood woodland and the semi-desert regions also have many endemic species. Of particular interest is an arid area in the east comprising Prosperous Bay Plain and its immediate surroundings, which unfortunately is also the most appropriate site for an airport. The area has six genera and around 40 species of invertebrates that have been recorded only in this part of the island and nowhere else in the world. The fauna of the plain includes a remarkable array of endemic nocturnal wolf spiders (Lycosidae). Discovery of St. Helena in 1502 was followed by introduction of a range of herbivores and predators. Pigs, dogs, cats, and rats decimated seabird colonies and drove vulnerable endemic land birds to extinction. Goats, released within a decade or so of 1502, were the prime destroyers of the native vegetation, although rabbits, doubtless, also played a part. Tree felling for timber and firewood, along with burning of wood in the making of lime, became major factors after settlement in 1659. By the early nineteenth century, much of the island had been denuded of native trees. Near the end of the same century, cultivation of the New

872

S T. H E L E N A

Zealand flax Phormium tenax was started and soon led to massive destruction of surviving native forest on the central ridge. Invasion of forest remnants by introduced plants (especially flax) continued through the second half of the twentieth century. Drier parts of the island also have many invasive plants, including the creeper Carpobrotus edulis, the prickly pear Opuntia spp., and Lantana camara. Within a few decades of settlement, some enlightened administrators of St. Helena instituted measures to prevent extinction of an endangered tree species (St. Helena redwood Trochetiopsis erythroxylon) and to conserve the native forests. Failure to ensure continuity of effort resulted in reduction of the original forests that once covered the greater part of the island to about 1 ha of gumwood woodland and 16 ha of cabbage tree woodland and tree fern thicket along the central ridge. The ecological devastation of St. Helena through human agency, recorded in the correspondence of the East India Company, was one of the examples that influenced the development—during the late seventeenth to early nineteenth centuries—of western philosophical and scientific ideas about the relationship of humans with the natural world. In recent years, serious efforts have been made to halt the loss of native forest and to care for relict stands of endemic plants. Establishment of the Diana’s Peak National Park in 1996 focused attention on a key area. Educational efforts have raised awareness of endemic species, and the endangered St. Helena wirebird is the subject of a special conservation initiative. There have been important conservation gains, including rediscovery and protection of the St. Helena ebony Trochetiopsis ebenus

FIGURE 1 St. Helenian conservation volunteers caring for the endemic

boxwood Mellissia begoniifolia in scree formed of trachyte/phonolite boulders (bottom left) derived from Lot’s Wife pinnacle. Note the wind screen erected to protect the plants. The boxwood was thought to have become extinct in the nineteenth century, but this small stand was found in 1998.

Grove, R. H. 1995. Green imperialism: colonial expansion, tropical island edens and the origins of environmentalism, 1600–1860. Cambridge: Cambridge University Press. Rowlands, B. W., T. Trueman, S. L. Olson, N. McCulloch, and R. K. Brooke. 1998. The birds of St Helena. Tring, UK: British Ornithologists’ Union Checklist Number 16. British Ornithologists’ Union. Weaver, B. 1999. A guide to the geology of Ascension Island and St Helena. School of Geology and Geophysics, University of Oklahoma, Norman, Oklahoma 73019, USA.

STICKLEBACKS FIGURE 2 St. Helena olive Nesiota elliptica (Rhamnaceae). The death

of the last individual of this species (the only member of an endemic genus) was witnessed by the authors in 2003.

MICHAEL A. BELL Stony Brook University, New York

and boxwood Mellissia begoniifolia (Fig. 1), both of which had been thought to be extinct. However, the recent death of the last specimen of the St. Helena olive Nesiota elliptica (Fig. 2), a generic endemic, is a poignant reminder that in attempting to preserve extremely rare species, success can only be assured in the short term, whereas failure is forever. Recent plans for construction of an airport and the development of tourism will bring new threats to the native habitats and species on the island. Conservationists on the island are continually overstretched, and resources are limited. Nonetheless, some initiatives in ecological restoration have been undertaken. These aim to re-create large areas of native forest and have sometimes involved participation by a large proportion of the population. Experience suggests that the future of the natural environment and biodiversity of St. Helena will depend on the presence of individuals who care and have relevant expertise, on funding and support by the authorities even in the face of commercial pressures, and on the maintenance of protection and restoration measures over the long term.

The stickleback fish family (Gasterosteidae) comprises five major subgroups (genera), three of which are primitively marine but commonly colonize freshwater (Fig. 1). The habitats of these freshwater colonists contrast sharply with the marine environment from which they came, and their descendants rapidly evolve behavioral, physiological, and morphological traits that adapt them to diverse freshwater environments. Glaciation and isostatic depression eliminated freshwater fishes over wide areas of the Holarctic.

SEE ALSO THE FOLLOWING ARTICLES

Ascension / Deforestation / Fossil Birds / Insect Radiations / Introduced Species / Wallace, Alfred Russel FURTHER READING

Ashmole, P., and M. J. Ashmole. 2000. St Helena and Ascension Island: a natural history. Oswestry, UK: Anthony Nelson (current distributor: www.kidstonmill.org.uk). Baker, I. 2004. St Helena—one man’s island. Windsor, UK: Wilton 65. Cronk, Q. C. B. 2000. The endemic flora of St Helena. Oswestry, UK: Anthony Nelson. Edwards, A. 1990. Fish and fisheries of Saint Helena Island. Newcastle upon Tyne, UK: Government of Saint Helena and the University of Newcastle upon Tyne.

FIGURE 1 Stickleback species that commonly colonize freshwater

habitats on islands from the ocean: (A) threespine (male with red and blue reproductive coloration), (B) fourspine, and (C) ninespine stickleback. Photographs by Joseph Ross/PhotoQuery.

STICKLEBACKS

873

abundant in freshwater. The fourspine stickleback is limited to northeastern North America. ISLANDS ON WHICH FRESHWATER STICKLEBACKS OCCUR Threespine Stickleback

FIGURE 2 Variation in the shape, size, and armor of the threespine

stickleback. The fish in the middle is a highly armored anadromous specimen, and those around the periphery indicate the range of armor and body shape among freshwater populations from western North America. All specimens were drawn at the same size, and the scale bars represent 1 cm. Important forms of armor variation include the distribution of lateral plates along the flanks, the number and length of dorsal spines, and the size and presence of the pelvis. Reprinted with permission from Bell and Foster (1994).

When those areas became exposed, sticklebacks and other fishes that move readily through the ocean but tolerate freshwater quickly colonized these newly formed, fishfree, freshwater habitats. Consequently, freshwater stickleback populations are common on islands within their Holarctic ranges. The threespine stickleback (Gasterosteus aculeatus) is ubiquitous in postglacial freshwater habitats, and isolated freshwater populations have formed spectacular adaptive radiations in Eurasia and North America (Fig. 2). Two other stickleback groups, the ninespine and fourspine sticklebacks, colonize freshwater habitats on islands. Thus, sticklebacks are abundant, ecologically important, and scientifically interesting members of insular freshwater ecosystems. ISLAND STICKLEBACKS

Threespine (Gasterosteus aculeatus species complex), fourspine (Apeltes quadracus), and ninespine (Pungitius spp.) sticklebacks are commonly anadromous (sea-run) or can pass through the ocean to enter freshwater. The fifteenspine or sea stickleback (Spinachia spinachia) of northwestern Europe never colonizes freshwater, and the North American brook stickleback (Culaea inconstans) does not occur in the ocean. The threespine stickleback is the most widespread and abundant freshwater stickleback. The ninespine stickleback comprises several species, which together are widely distributed and locally

874

STICKLEBACKS

The threespine stickleback species complex is widespread on islands around the north Pacific, Atlantic, and Arctic basins, including Japan, islands of the Russian Far East, the Aleutians, the islands of the Bering Sea, Kodiak Island, the Alexander archipelago, Haida Gwai (Queen Charlotte Islands), and Vancouver Island. They are also widespread on islands in the eastern Canadian Arctic and extend eastward into the Atlantic basin on Greenland and Iceland. They have been reported as far south as Chesapeake Bay, and in northwestern Europe, they are common in the British Isles and islands of Scandanavia and western Arctic Russia. The threespine stickleback is the only Icelandic stickleback. Ninespine Stickleback

Species of ninespine sticklebacks are also widespread. They occur throughout Japan, the Russian Far East, the Aleutian Islands, islands of the Bering Sea, and Kodiak Island in Alaska. They are widespread around the Arctic Ocean and occur in the Atlantic south to Newfoundland and Prince Edward Island, Canada; Long Island, New York; Greenland; northwestern Europe; and the British Isles. They are conspicuously absent in Iceland. Fourspine Stickleback

This species has the most limited distribution, occurring only in the northeastern United States and the adjacent Canadian Maritimes. Its insular freshwater distribution includes islands around the Gulf of St. Lawrence in Canada and Long Island, New York. CHARACTERIZATION AND IDENTIFICATION OF FRESHWATER ISLAND SPECIES

Sticklebacks are easy to identify even without capture. Threespine, fourspine, and ninespine sticklebacks rarely exceed 8 cm total in length and are usually smaller. Free dorsal spines precede the dorsal fin on the back, and each spine is followed by a membrane. The pelvis is robust and has a large spine on either side. Sticklebacks often erect these spines when they feel threatened, making identification easy. They routinely swim by sculling with the pectoral fins, holding the body straight or curving it to turn. They swim in short bursts, between which they hover to strike at food or search the bottom for prey. Thus, sticklebacks can be distinguished from other fishes by observa-

tion of the tempo of their swimming through the surface of the water or can be caught and identified. Threespine Stickleback

The threespine stickleback (Fig. 1A) is a complex of morphologically diverse populations, some of which represent separate biological species. Members of this species complex usually have two large and one small dorsal spine, and a series of large, bony lateral plates in a single row on each side. Most anadromous and marine stickleback and some freshwater populations have a continuous row of about 33 plates running from head to tail (complete plate morph). However, the plates may form separate rows near the head and tail with an unplated gap in between (partial morph) or may number fewer than ten plates restricted to the front of the body (low morph). Some freshwater populations contain all three plate morphs. Threespine stickleback range in color from silvery in open waters (including the ocean) to drab beige with dull green to brown vertical bars in shallower, more heavily vegetated habitats. During the breeding season, males typically develop red throats, but the red color may cover much of the lower body (Fig. 1A). Threespine stickleback often can be seen in the open near shore in schools or small groups. They may be mistaken for small trout or salmon, but the tempo of their swimming—stopping, darting, and swimming again—differs strikingly from the more continuous undulation of trout. Ninespine Stickleback

The ninespine stickleback (Fig. 1B) is more elongated but generally similar in shape to the threespine stickleback. Although it too may have a complete row of plates on the flanks, these plates are smaller and usually form a short row on the sides in front of the tail fin. Ninespine sticklebacks have seven to 12 dorsal spines, which are smaller than those of the threespine stickleback and which tilt alternately to the sides. Ninespines tend to be more brownish to gray in color than threespines, and reproductive males are partly or completely black with contrasting white pelvic spines. They are secretive and rarely seen in the field. Fourspine Stickleback

The fourspine stickleback (Fig. 1C) is usually less than 5 cm long, relatively deep bodied compared to the threespine, and lacks lateral plates. It usually has more than three but fewer than seven dorsal spines, and they tilt to the right and left, distinguishing it from the other sticklebacks. The body is generally brown and beige, and reproductive males have red pelvic spines.

FIGURE 3 Cladogram depicting the sequence of branching during

evolution of the genera in the stickleback family (Gasterosteidae).

DIVERSITY AND RELATIONSHIPS

The sticklebacks constitute the small family Gasterosteidae with five distinctive genera (Fig. 3). Relationships of the stickleback genera have been inferred from behavior, morphology, chromosome structure, and DNA sequences. Aside from the typical morphological features already noted, sticklebacks and a few close relatives are distinguished from most other fishes by the nests males build from plant fibers that they glue together with a kidney secretion. The cladogram (family tree) in Fig. 3 depicts the current understanding of relationships among types of sticklebacks. Only a single species of the fifteenspine, fourspine, and brook stickleback are presently recognized. Several geographically isolated species of ninespine sticklebacks are usually recognized. The threespine stickleback has one close but distinctive relative, the black-spotted stickleback (Gasterosteus wheatlandi) of northeastern North America. However, freshwater populations of both threespine and ninespine sticklebacks exhibit fascinating morphological variation, and some of the morphologically differentiated populations clearly represent unnamed biological species that are typically restricted to a single lake. Thus, although there is only one named species of threespine stickleback and a small number of named ninespine stickleback species, named species of both Gasterosetus and Pungitius may include a great deal of biodiversity that is easily underestimated. STICKLEBACK BREEDING BEHAVIOR

The elaborate breeding ritual of the threespine stickleback has been studied intensively for more than 50 years, and it can readily be observed in the field and aquaria. In the spring, a male will establish a territory that he defends against other males. He digs a pit within the territory, often near cover, brings plant fibers to the pit, and glues them together with his kidney secretion. The male probes and pulls on the fibers to shape them into a nest. When the pile of fibers is ready, he creeps through it to form a chamber. Now he is ready to court gravid (ripe) females.

STICKLEBACKS

875

Gravid females are easily recognized by their bulging bellies, and males prefer to spawn with larger females. When a male sees a gravid female, he swims in her general direction using short bursts alternating to the right and left to perform the classic zig-zag dance. If the female is receptive, she will tip her head up at about a 30° angle when the zigzagging male approaches to signal her readiness to spawn with him. The male may lead her back to the nest immediately or stick her with his dorsal spines (dorsal pricking), but he usually returns to the nest to perform nest-oriented behaviors, including gluing, fanning, or creeping through the nest, before he zig-zags back to her again. After a final bout of male zig-zagging and a female heads-up response, he leads her back to the nest, lies on his side and points his snout into the nest entrance (showing) where he wants her to enter. Now she may enter, steal a mouthful of eggs and flee, or break off the courtship. If she enters the nest, he will vibrate his snout against the base of her tail fin, which usually stimulates her to release her eggs. He may bite her tail to induce her to leave the nest after she spawns or has failed to do so, after which he creeps through the nest to release sperm and fertilize her eggs. Having achieved his goal, he chases her out of the territory. Another male (a sneaker) may observe this ritual, dash into the nest, and fertilize the eggs before the owner of the nest. The male may repeat this courtship ritual and spawn with several females before he enters the parental phase, during which his major activity is fanning water over the eggs with his pectoral fins, while he swims forward furiously with his tail fin to hold his place above the nest. His other major activities are removing dead eggs and chasing off predators, including other stickleback that may cannibalize his eggs or fry. The eggs hatch about a week after spawning, and the male tends the fry for another week, after which they become unmanageable and swim away. Only a minority of stickleback clutches survive. The male may build a new nest and repeat this process or die of exhaustion. Ninespine and fourspine sticklebacks deviate from these courtship habits but share territoriality, elaborate courtship, nest building, male parental care, and the use of glue to build the nest. However, they nest off the bottom in vegetation, and male fourspine stickleback build a separate nest for the clutch of each female and suck water through the nest instead of fanning. OTHER ECOLOGICAL PROPERTIES

Sticklebacks are used by a wide range of pathogens, multicellular parasites, and predators. They are important prey for a variety of fish, birds, aquatic insects (e.g., dragonfly larvae), and other predators. Most sticklebacks eat

876

STICKLEBACKS

any animal material they can capture, but some threespine stickleback populations specialize on plankton or benthic prey. They usually live one or two years before breeding one or more times during a single season and dying. Females ripen eggs in batches, which they spawn as a single clutch at about weekly intervals. Clutch size depends strongly on body size and ranges from a few tens of eggs to several hundred. ADAPTIVE RADIATION OF THREESPINE STICKLEBACK

There is relatively little variation in the marine or anadromous populations of any stickleback species. However, there is extensive variation within and among freshwater stickleback populations, and it is most conspicuous in the threespine stickleback. This variation was poorly understood until the late 1960s, when it became clear that many stickleback traits are adaptations to local conditions. Subsequent studies showed that variation among freshwater stickleback populations can evolve within decades, and may produce striking differences between populations only a few meters apart. For example, populations in a lake and its tributary or outlet streams may look and behave very differently. Diet (plankton versus bottom prey) and predation regime (the mix of insect, fish, and bird predators) are the major factors for stickleback adaptive radiation, but dissolved salt concentration also influences evolution of armor. Because so many Holarctic lakes contain threespine stickleback, similar conditions in multiple lakes have caused the independent evolution of similar traits in numerous geographically distant freshwater populations, enabling the detection of associations between specific environmental conditions and the evolution of specific stickleback traits. Ninespine and brook sticklebacks exhibit similar population differentiation in lakes, but they have not been studied as intensively. IMPORTANCE OF THE THREESPINE STICKLEBACK IN BIOLOGICAL RESEARCH

The threespine stickleback was among the first species named by Linnaeus in 1758. Niko Tinbergen’s research in the middle of the twentieth century established it as a classic subject for research in ethology. Research in the late 1960s led to the discovery of exceptional variation in behavior, life cycles, color patterns, skeletal traits, and other properties. Subsequent research on the evolution of the threespine stickleback revealed high levels of skeletal variation within and among populations around the Holarctic. This variation and other favorable properties of the threespine stickleback for genetic analysis (e.g., short

generation time, small body size) attracted the interest of geneticists who are beginning to use threespine stickleback to study how variation in DNA sequences affects development and phenotypic variation. In 2006, the threespine stickleback became one of the first fish to have its genome sequenced. Thus, research originally motivated by questions about adaptation and species formation has led to the development of the threespine stickleback for use in biomedical research. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Arctic Islands, Biology / Freshwater Habitats / Lakes as Islands FURTHER READING

Bell, M. A., and S. A. Foster. 1994. The evolutionary biology of the threespine stickleback. Oxford: Oxford University Press. McKinnon, J. S., and H. D. Rundle. 2002. Speciation in nature: the threespine stickleback model systems. Trends in Ecology and Evolution 17: 480–488. Östlund-Nilsson, S., I. Mayer, and F. A. Huntingford. 2007. Biology of the threespined stickleback. Boca Raton, FL: CRC Press. Wootton R. J. 1976. The biology of the sticklebacks. London, UK: Academic Press. Wootton, R. J. 1984. A functional biology of sticklebacks. Berkeley: University of California Press.

SUCCESSION BEATRIJS BOSSUYT University of Ghent, Belgium

Succession consists of the often-predictable series of changes in an ecological community over time after a disturbance. These changes occur through colonization and extinction of species. Primary succession involves the assembly of a plant community on a newly formed substrate, whereas secondary succession indicates community changes after a disturbance that has destroyed the vegetation but left the soil to a large extent intact. During primary succession, species accumulate in the plant community through dispersal, environmental selection, and biotic interactions, resulting in an increase of species richness with time. On islands, both dispersal limitations and environmental selection differ from those on the mainland. The isolated position of the island and the large distance to source populations strongly hamper seed dispersal processes, retarding the arrival of species. The newly formed substrate provides a very harsh and stressful environment for the colonizing

organisms. Much as in mainland succession situations, biotic interactions, such as facilitation and competition, impact the successional pathway. DISPERSAL LIMITATION

The degree of isolation of the island (i.e., the distance to source populations on the mainland or on other islands that may act as stepping stones) is a first filter on the potential local species pool. The first pioneers on islands are in most cases good dispersing species. These species may establish on the coastlines, although this may be hampered by the presence of cays. Islands contain fewer species, and in particular the subset of the best dispersing species, compared to the vegetation on the mainland; this results in a disharmonic flora. Because seed dispersal is to a large extent a stochastic process, and because the occurrence and sequence of seed arrival are unpredictable, the successional pathway during the first stages on islands is largely non-deterministic. The arrival of particular species on the island often depends on chance events (e.g., the establishment of a bird colony on the island, which may result in an accelerated establishment of plant species). The importance of chance events results in a divergent succession pattern, meaning that the successional pathway and the resulting species composition can strongly differ between islands with a similar area and degree of isolation. The same applies for continental islands, isolated habitat patches within a matrix of habitat that is unsuitable for the colonizing species. Also in this case, seeds of species arrive in the habitat patch from source populations at larger distances, and the stochastic nature of the probability and sequence of arrival may result in a divergent succession pattern. ENVIRONMENTAL SELECTION AND BIOTIC INTERACTIONS

New islands often arise after volcanic eruptions or because of a lowering of the lake water table. In the first stages, this substrate is very harsh and stress-imposing, unstable and with an extremely low nutrient availability, in particular of nitrogen, and a high water deficit. Cyanobacteria are known to be among the first organisms to stabilize the substrate. Some microorganisms may slightly increase the nutrient availability by nitrogen fixation. The colonization of vascular plant species and bryophytes is a slow process, because only a limited number of the stress-tolerant species that can cope with the low nutrient and water availability are able to establish at the most favorable microsites. During the early stages, bryophytes are the most important colonizers,

SUCCESSION

877

establishing on the most favorable areas and loosening the substrate. During the next stages, vascular plant species, in particular dwarf shrubs, establish in the moss carpet, often in association with scattered forbs and graminoids. Symbiosis with mycorrhizae, facilitating the uptake of nutrients by the plant, probably plays an important role in the first successional stages. Moreover, the small number of species present in pioneer populations lead to a high degree of inbreeding and a decreased sexual reproduction. In the early stages, positive interactions between neighbors (i.e., facilitation) are generally more important than negative interactions, such as competition. Facilitation by nitrogen-fixing species, which add nitrogen to the soil, can be considered an important step in the successional process. In this way, pioneer species facilitate the establishment of later successional species. However, nitrogen fixing is energy demanding. Although phosphate is released during the soil weathering process, nitrogen fixing may be hampered by phosphate shortage, such that primary succession can often be considered as nitrogen and phosphate co-limited. Some species may also act as nursery plants, protecting seedlings of later arriving species from extreme microclimate conditions. In the course of the primary succession, when the number and abundance of species gradually increase, environmental conditions change. The substrate stabilizes, and there is an increasing amount of organic matter and nutrients and a decrease in pH because of the building up and the decomposition of organic matter and because of symbiotic nitrogen fixation. An ameliorated soil structure increases the soil’s water-holding capacity. There is often a trade-off between a species’s level of stress tolerance and its dispersal capacity, which may retard succession in the case that stress-tolerant species are not able to reach the island and good dispersing species do not manage to establish a viable population. If stress-tolerant species are hampered by dispersal barriers, a long period of slow habitat change must occur before species accumulate. As conditions ameliorate, the rate of succession gradually accelerates. Species become more abundant, and total cover increases, such that competitive ability becomes a more important condition for establishment than stress tolerance. Species that can more efficiently use nutrients will outcompete less dominant species. The sequence of arrival of different species determines the outcome of these competitive interactions. When succession further continues, tree and shrub species colonize, and the canopy closes. At that point, competition for nutrients shifts toward competition for light. Early successional species will disappear from the vegetation because they can not cope with the

878

SUCCESSION

increased shade conditions. For this reason, there is often a species richness peak in the middle successional stages because of the temporal co-occurrence of early successional and late successional, shade tolerant species. Similar to isolation, habitat conditions and competitive interactions restrict the species composition to some subset of the potential species pool. In contrast to dispersal limitations, the effect of habitat conditions is to a larger extent predictable. A similar plant community can be expected on islands with equal ecological conditions, resulting in community convergence. If habit conditions and biotic interactions are the prevailing factors in determining the course of succession, then a deterministic successional pathway will result. If large-scale disturbances, such as hurricanes, destroy late successional vegetation, then a secondary succession starts. Secondary succession follows a totally different pathway than primary succession and is in most cases a much faster process because propagules are already present, and soil nutrient conditions and water availability are not restricting plant community assembly. In contrast, species can often profit from the accelerated decomposition of organic matter and the release of nutrients after a disturbance. Competition will play a much more important role in structuring plant communities during secondary succession. Parallel with the process of the colonization by plant species, animal species colonize the island. Much less is known, however, about animal successional pathways. As is the case with plant species, both distance to source populations and dispersal capacity, along with environmental factors, such as climate and soil conditions and the structure and composition of plant communities, will determine the occurrence of animal species. The first colonizers during primary succession are often almost exclusively predators, whereas herbivores and decomposers appear only in later stages. Animal species may affect the successional pathway of plant species through soil organic matter decomposition, seed dispersal, and herbivory. DIFFERENCES BETWEEN ISLANDS

Because of the importance of dispersal limitation, differences in the degree of isolation of islands are to a large extent responsible for differences in their successional pathways. Distant islands reach the equilibrium much more slowly—and with a balance between immigration and extinction—than do islands near the mainland. In addition to isolation, area determines the number of species accumulating. Large islands contain larger and more differentiated habitats, so more species can co-occur, and species accumulation is much faster. Overall, species

Surtsey Lake Hjälmaren (small)

number of plant species

120

Lake Hjälmaren (large) Lovén islands

100 80 60 40 20 0 0

50

100

150

time (years) FIGURE 1 Number of species as a function of time since the island

appeared, for Surtsey (Iceland), small and large islands in Lake Hjälmaren (average values) (Sweden), and the Lovén Islands (Norway).

accumulation and community assembly will be slower on more isolated and smaller islands, in comparison to large islands at short distances from the mainland. General ecological conditions on an island at the start of the succession (e.g., climate and initial soil conditions) have been found to strongly determine the rate of succession. Increasing temperatures and precipitation enhance rates of organic matter decomposition and soil formation. Succession and species accumulation rates on boreal islands will hence be slower than on islands at tropical latitudes, because processes of weathering and soil formation are slower. Moreover, the potential species pool is smaller at higher latitudes. Additionally, characteristics of the substrate will determine the successional rate. Parent material with a fine texture will allow a faster succession because it is likely to weather more rapidly, increasing phosphate availability and the formation of an appropriate soil structure. Fig. 1 illustrates the species accumulation rates of three island examples in the boreonemoral and boreal climate zone of the northern hemisphere. The 40 islands of Lake Hjälmaren (482 km2, south-central Sweden, 60° N) appeared after a lowering of the water table in 1886, and ranges in size between 20 m2 and 1 km2. The islands can be divided into large (greater than 0.3-ha) and small (less than 0.3-ha) islands. As can be seen, large islands reach equilibrium at a time scale of about 50 years, after which species richness slightly decreases, whereas small islands still continue to accumulate species, and the equilibrium may last for more than a century. The number of species at small islands is at every point in time always smaller compared to that of large islands. The Lovén islands (78° N), situated in Spitsbergen, Norway, were successively released from beneath the ice during the regression of a valley glacier. The islands are about 3 km from the mainland and separated by at least 1 km of

open water. Because these islands are situated in the boreal climate zone, species accumulation is very slow, and only six species were established on the island after a period of 100 years. Surtsey (63° N) covers 1.4 km2 and is situated at the southern coast of Iceland; it appeared after a volcanic eruption in 1967. The slower species accumulation rate on the Lovén islands and Surtsey, in comparison with the islands in Lake Hjälmaren, despite the similar latitudinal position, may be attributed to the much higher degree of isolation of this oceanic island in comparison with the lake islands. Plant species richness increased for the first 15 years and then stabilized. The steep increase of species richness after 20 years is due to the establishment of a seagull colony on the islands. Seeds are transported by the birds, and there is a spatially concentrated input of nutrients, allowing a larger subset of species to establish. This clearly illustrates the importance of chance events in early successional stages on islands. SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / Convergence / Dispersal / Species–Area Relationship / Surtsey / Vegetation FURTHER READING

Baldursson, S., and A. Ingadóttir. 2006. Nomination of Surtsey for the UNESCO World Heritage List. Reykjavík: Icelandic Institute of Natural History. Burns, K. C. 2005. A multiscale test for dispersal filters in an island plant community. Ecography 28: 552–560. Kadmon, R., and H. R. Pulliam. 1993. Island biogeography: effect of geographical isolation on species composition. Ecology 74: 977–981. Rydin H., and S.-O. Borgegård. 1988. Plant species richness on islands over a century of primary succession: Lake Hjälmaren. Ecology 69: 916–927. Thornton, I. W. B. 1996. Krakatau: the destruction and reassembly of an island ecosystem. Cambridge, MA: Harvard University Press. Thornton, I. W. B. 2007. Island colonization: the origin and development of island communities. Cambridge: Cambridge University Press. Whittaker, R. J., and S. H. Jones. 1997. The rebuilding of an isolated rain forest assemblage: how disharmonic is the flora of Krakatau? Biodiversity and Conservation 6: 1671–1696.

SURF IN THE TROPICS GRAHAM SYMONDS CSIRO Marine and Atmospheric Research, Wembley, Australia

THOMAS C. LIPPMANN Ohio State University, Columbus

The wave climate around islands is highly variable in space and time as a result of the level of exposure to incident waves and the frequency of local and distant storm

SURF IN THE TROPICS

879

events. The magnitude of the incident waves affects the distribution of benthic communities such as algae, corals, and fish assemblages, as well as the distribution of sediment and the accretion and erosion of shorelines. Topographic effects often provide waves highly sought by surfers. Waves breaking on coral reefs around islands can have a significant impact on sea level and mean currents in an otherwise sheltered lagoon, flushing water, nutrients, and pollutants through the system and helping to maintain a healthy marine environment. LINEAR WAVE THEORY

Field research carried out on tropical islands over the past several decades has led to fairly robust understanding of wave transformation, water levels, and circulation on fringing reef systems. The basics of the wave physics are the same as for open coast sandy beaches found worldwide; however, the importance of bathymetric irregularities and characteristic geometries of fringing reef systems on wave and current dynamics renders the circulation significantly different. The close link between the circulation and the health of the living reef ecosystem is recognized. Linking sediment and solute transport to predictive models of wave transformation and wave-driven flow with wind and tidal forcing is a difficult problem, but much progress has been made. The following sections discuss basic elements of wave physics and the dynamic consequences of waves impinging on a typical fringing reef system. Wave Characteristics

Linear wave theory adequately describes many of the observed features of wave propagation and transformation, and the reader is referred to the list of references following this article. A schematic of an idealized wave is shown in Fig. 1, where wave height is the distance Surf zone

Shoaling zone Crest

A

Wavelength

B

Direction of wave motion

Height

Trough

Still water level

Orbital motion of water particles

between wave crest and trough, wave period is the time taken for a wave crest at position B to reach position A, and wavelength is the distance between consecutive wave crests. Wave periods are typically in the range of 5 to 20 seconds, and the corresponding range of wavelengths in 20 m depth is 39 m to 270 m. Waves in the open ocean are generated by winds. In general, wave growth depends on the strength of the wind, the duration over which the wind blows, and the region (fetch) that the storm covers. As waves propagate outside their generation region, they begin to disperse, with longer-period waves traveling faster than shorter-period waves. As the waves sort themselves out, they form wave packets, or groups. In a typical group, there are anywhere from 5 to 10 wave crests that vary in height from small amplitudes at the front and rear of the group to larger amplitudes in the middle. The group pattern travels at one-half the speed of the individual waves in deep water; thus, the individual waves pass through the groups as the group pattern propagates shoreward. The energy of the ensemble of all waves in the wave field propagates with the group. In the absence of wave breaking or bottom drag, the transmission of wave energy, called the energy flux, is conserved for a given length of wave crest. It is important to recognize that waves represent a propagation of energy, not mass, and that the energy propagates with the wave group rather than the individual waves. Beneath the surface, individual water particles describe almost closed ellipses (or circles in deep water) as shown in Fig. 1. As the wave crest passes, the particles at the top of the elliptical orbit are moving in the direction of wave motion, and as the trough passes, the particles at the bottom of the elliptical orbit are moving opposite to the direction of wave motion. Fig. 1 also illustrates a number of changes that occur as waves enter shallow water. Through the shoaling zone, the speed of the wave crest decreases with depth, and because the wave period must remain the same (to conserve the number of waves in a given time span), the wavelength must also decrease. In order to conserve energy flux, there must be a corresponding increase in wave height, thereby increasing its steepness. Eventually the wave crest overturns, or breaks, onto the front face, creating turbulent vortices and dissipating energy, and thus the wave height diminishes through the surf zone, approaching zero at the shoreline. Waves with residual height are reflected off the beach face and head back out to sea.

FIGURE 1 Schematic of the transformation of ocean surface waves

approaching a shoreline. Through the shoaling zone, waves slow down, their wavelength decreases, and their height increases as they approach the surf zone. Through the surf zone, wave height is attenuated by dissipation processes and decreases to near zero at the shoreline.

880

SURF IN THE TROPICS

Wave Refraction

The dependence of the speed of the wave crests on the water depth leads to the process known as refraction.

Consider a very long crested wave approaching an island. That part of the wave crest that first encounters the shallow water around the island will slow down, whereas the part of the crest out in deeper water will be moving faster and will thus move ahead of the slower-moving part of the wave. This process is called refraction, and it causes the wave crests to bend toward the island, wrapping into what would otherwise be a shadow region behind the island. However, as the waves refract, they are effectively being stretched along the crest, and conservation of energy causes the wave height to decrease such that the wave energy flux (proportional to wave height squared times the group velocity) per unit length of wave crest remains constant. The greater the angle between the direction of the refracted wave and the corresponding deep water direction, the greater the reduction in wave height. State-of-the-art numerical models do a reasonably good job of simulating wave refraction over complex bathymetry, although care must be taken to properly represent intersecting waves refracting around opposite sides of an island. Wave Momentum

Consider a column of water with a horizontal area of 1 m2 extending from the sea surface to the bottom. Under the wave crest, this water column has a greater mass than it has under a trough. Momentum, given by mass times velocity, is thus greater under the crest than under the trough, and thus, when averaged over a wavelength, results in a flux of momentum in the direction of wave motion. This momentum flux caused by the presence of surface waves is known as radiation stress. As waves approach a shoreline or a sufficiently shallow reef, they eventually break, dissipating energy and transferring momentum to the water column. To conserve momentum, the gradient in wave momentum translates to a force that can drive a mean current in the direction of the waves or that is balanced by an equal and opposite force, such as a pressure gradient resulting from water piling up against a shoreline. On coral reefs around tropical islands, these wave-driven flows can be quite strong. A schematic representation of wave-driven flow over a submerged reef is shown in Fig. 2. Through the surf zone, the gradient in wave momentum drives a current in the direction of wave motion. However, when the waves reach the top of the reef slope, they cease to break, the gradient in wave momentum vanishes and is thus unable to provide a force to drive the flow over the reef top. In this one-dimensional case, assuming the current is uniform throughout the water column, conservation of mass requires the current to increase as

Direction of wave motion

Surf zone

Mean sea level Still water level Mean current Reef flat

Reef slope

FIGURE 2 Schematic of wave-driven mean flow (large arrows) and

corresponding mean sea level (solid blue line) over a submerged reef. Through the surf zone, the mean current is forced by radiation stress gradients and an opposing pressure gradient associated with the sea surface slope (blue line). Seaward and shoreward of the surf zone, the current is forced only by the pressure gradient caused by the sea surface slope.

the depth decreases, as illustrated in Fig. 2 by the larger arrows representing stronger currents. A pressure gradient is established by increasing mean sea level through the surf zone to a maximum at the top of the reef slope. From there, the mean sea surface slopes down to the level of water in the lagoon, providing a pressure gradient sufficient to drive the flow over the reef crest. Through the surf zone, the slope of the mean sea surface opposes the cross-reef flow such that increasing the mean sea surface elevation at the top of the reef slope increases the flow across the reef flat while decreasing the flow through the surf zone until the mass transport over the reef flat is equal to the transport through the surf zone. The magnitude of the cross-reef flow depends on the magnitude of the incident waves offshore, on the width of the reef flat, and on the depth of water over the reef flat. In the idealized case shown in Fig. 2, the mean sea level decreases to the still water level at the downstream, or lagoon, side of the reef flat. Observations have shown that in the presence of waves breaking on a reef, the mean sea level in the lagoon may also increase above the still water level, effectively reducing the cross-reef pressure gradient and cross-reef flow. It is possible for this wave-induced setup to raise the lagoon sea level sufficiently to prevent any cross-reef flow. Because of the typically narrow continental shelf around islands, wave setup can cause significantly larger sea-level fluctuations than can wind-induced storm surges that occur along continental margins. As waves wrap around an island, refraction does not always orient the wave crests entirely parallel with the

SURF IN THE TROPICS

881

bathymetry, particularly on islands where the shelf is very steep. In this case, obliquely incident waves will also drive mean flows along the reef parallel to the depth contours. WAVES ON REEFS

Natural coral reefs are significantly more complex than the idealized case shown in Fig. 2, with highly variable bathymetry across and along the reef and occasional deep passages cutting through the reef. Lord Howe Island (31°34´ S, 159°05´ E), shown in Fig. 3, is situated 630 km off the east coast of Australia, and is approximately 10 km long and up to 2.6 km wide. It is roughly crescent shaped, with a lagoon facing southwest and bounded by a fringing reef about 6 km long; it is the southernmost coral reef in the world. The lagoon and reef are shown in Fig. 3, with depth contours in gray; the white regions associated with wave breaking indicate the location of the reef crest. The reef is cut by three deep passages connecting the lagoon and open ocean, one at the northern

FIGURE 3 Mean current vectors measured along the reef and in the

lagoon at Lord Howe Island, November 27 to December 14, 2004.

882

SURF IN THE TROPICS

end of the reef (North Passage) and two at the southern end (Erscotts Passage and South Passage). Mean currents, obtained by averaging 2-Hz data over five minutes, were recorded continuously for a three-week period at a number of sites indicated on Fig. 3. The current data shown in Fig. 3 have been smoothed and decimated to 12 hourly vectors, removing most of the semi-diurnal tidal component. These data confirm the presence of a mean inflow over the reef crest as predicted by the idealized one-dimensional model described above. However, in the two-dimensional case, once the flow enters the lagoon, it may be deflected alongshore toward the deeper passages, where the absence of wave breaking allows the water to exit back out to sea. The current pattern shown in Fig. 3 indicates a persistent outflow in the passages. Current speeds in the lagoon are generally weaker, with occasional short pulses of strong currents. On a natural reef, current magnitude is also limited by bottom friction governed by the roughness of the bottom. Observations from Kaneohe Bay, Oahu, Hawaii, have shown that the dissipation of wave energy by bottom friction is similar to depth-induced breaking over a very wide reef flat. Clearly, there is variability in current speed and direction not always consistent with wave forcing, and it may reflect contributions that are the result of other processes such as wind and tidal forcing, which will certainly dominate during periods of low waves. Application of a stateof-the-art numerical model reveals the complex spatial variability of the wave-driven currents, as shown in Fig. 4. In this case, the model is forced by a constant wave height and direction at the offshore boundary with no tidal or wind forcing; thus, the spatial variability in the modeled flow field is mainly determined by spatial variability in the bathymetry. Although the model shows inflow over the reef and outflow in the major channels, qualitatively similar to the observations shown in Fig. 3, bathymetric low points along the reef crest are also sufficient to cause outflow. The magnitude of the cross-reef flows and the relative shallowness of many coral reef lagoons leads to effective flushing of the lagoon with time scales on the order of the semi-diurnal tidal period or less. Tropical waters surrounding coral reefs are generally nutrient-poor, but the rapid flushing maintains a high through-flow of lownutrient water sufficient to maintain the highly diverse coral reef ecosystem. Nutrient uptake by some corals and benthic communities has also been found to increase with increasing flow rate. Wave-driven flows also wash away fine suspended sediment from terrestrial runoff, which can have a detrimental effect on corals by reducing

Kraines, S. B., T. Yanagi, M. Isobe, and H. Komiyama. 1998. Wind-wave driven circulation on the coral reef at Bora Bay, Miyako Island. Coral Reefs 17: 133–143. Longuet-Higgins, M. S., and R. W. Stewart. 1964. Radiation stresses in water waves; a physical discussion, with applications. Deep Sea Research 11: 529–562. Monismith, S. G. 2007. Hydrodynamics of coral reefs. Annual Review of Fluid Mechanics 39: 37–55. Storlazzi, C. D., A. S. Ogston, M. H. Bothner, M. E. Field, and M. K. Presto. 2004. Wave- and tidally-driven flow and sediment flux across a fringing coral reef: Southern Molokai, Hawaii. Continental Shelf Research 24: 1397–1419. Symonds, G., K. P. Black, and I. R. Young. 1995. Wave driven flow over shallow reefs. Journal of Geophysical Research 100(C2): 2639–2648.

SURTSEY STURLA FRIDRIKSSON Reykjavik, Iceland

FIGURE 4 Numerical simulation of wave-driven currents at Lord Howe

Island using a depth-averaged two-dimensional hydrodynamic numerical model (2DBeach, http://www.asrltd.co.nz). Background shading from light to dark indicates increasing depth according to the scale on the right.

ambient light levels and smothering benthic communities. Terrestrial runoff can also contain pollutants such as fertilizers, and rapid flushing of these pollutants helps prevent toxic algal blooms from developing in the lagoon. Most fish species have a larval stage, during which time dispersal and subsequent settlement are largely governed by local and large-scale current patterns. SEE ALSO THE FOLLOWING ARTICLES

Beaches / Lord Howe Island / Reef Ecology and Conservation / Sea-Level Change / Tides / Tsunamis FURTHER READING

Brown, J., A. Colling, D. Park, J. Phillips, D. Rothery, and J. Wright. 1993. Waves, tides and shallow water processes, G. Bearman, ed. Open University, Pergamon Press. Dean, R. G., and R. A. Dalrymple. 2001. Coastal processes with engineering applications. Cambridge: Cambridge University Press. Komar, P. D. 1998. Beach processes and sedimentation, 2nd ed. Upper Saddle River, NJ: Prentice Hall.

Off the southern shore of Iceland in the North Atlantic Ocean is a group of 14 islands and a number of skerries called Vestmannaeyjar or the Westman Islands. The largest of these is Heimaey at 11.6 km2, the only populated member in the archipelago. The youngest and the outermost island is Surtsey. It was formed during a volcanic eruption that started on November 14, 1963. AN ISLAND CREATED

That morning a fishing boat was situated some 20 km southwest off Heimaey. At dawn the captain saw a column of smoke coming out of the sea and realized that a volcanic eruption had started. The pillar of smoke rose ever higher with pulsating explosions, constantly ejecting cinders and ashes, while lava bombs were being flung up and splashed into the sea. The ash particles blown into the atmosphere were positively charged, and shortly after an explosion from the volcano, they triggered great lightning displays that were spectacular to watch and were seen for miles around. It took only 24 hours of volcanic activity for an island to take form, built up from the ocean floor of 130 m in depth (Fig. 1). This island was later named Surtsey after the fire giant from Nordic mythology, Surtur the Black— a name that suited the island’s fiery origins and its black basalt lava formations. SURTSEY SECURED FOR SCIENCE

Quickly recognizing the scientific importance of the new island, the Icelandic authorities decided to protect

SURTSEY

883

FIGURE 1 A volcanic island emerges out of the sea off the southern

coast of Iceland in November 1963. Photograph by Sturla Fridriksson.

it, making it accessible only to scientists. The legislation was passed despite the enormous tourist interest in the eruption. Thus, the island became an unique laboratory for scientific studies and was secured as a typical isolate, where natural forces alone could act without any interference from humans. Surtsey has continued to awaken world attention, first and foremost because of the legal isolation of the island, but also because of the research that has been conducted on the island under the protection of law. Surtsey was an unusual geological phenomenon, and at the same time, it created an exceptional opportunity for biological research, even including the origin of life. An experiment was carried out to demonstrate that organic material could be created by the actions of volcanic eruptions. Surtsey is a small replica of Iceland, where the same elements are active and can be studied on a condensed scale. For all these reasons, Surtsey has been nominated for the World Heritage List. Those scientists who were interested in Surtsey and its development realized the need to organize their research and founded the Surtsey Research Society in 1965 to promote research in geological and biological sciences on the island. It publishes the journal Surtsey Research (www .surtsey.is).

The eruption continued from the island’s crater until early February 1964, when it shifted to a new opening, farther to the west of the island. Ashes from the two craters piled up to form two crescent-shaped cones, Austurbunki and Vesturbunki. The explosions from the new crater continued until April 4, 1964, when the volcanic shaft developed a watertight lining, and the ocean no longer flooded over the crater wall. When this occurred, floating lava emerged, and a lava fountain was formed, spouting 50- to 100-m high columns of glowing lava that bubbled and splashed up from the red-hot lake. This marked the beginning of the effusive phase. Sometimes the lava overflowed the crater rim, and floods of thin, fiery streams swept down from the crater over the south part of the island, gushing toward the sea with speeds of up to 70 km/h. The glowing lava often flowed through closed veins, retaining a temperature of 1100 °C and ending its journey by cascading off the cliffs into the ocean (Fig. 2). When the scorching lava came into contact with the cool ocean, steam whipped up, forming a white fringe around the edge of the island. The lava solidified, forming a hard dome of a lava surface that sloped gently to the sea on the eastern and southern sides of the island. During the eruption, new veins with outlets were formed from the main volcanic shaft. A few small craters opened up on the island, and two satellite islands formed off the shore of Surtsey. They consisted only of cinder cones and, lacking a lava crust, were washed away when the eruptions ceased. When the main eruption ended on June 5, 1967, it had lasted over three and a half years. Surtsey had increased in size to 2.7 km2 in area and reached a height of 175 m above

GEOLOGICAL DEVELOPMENT

The eruption experienced several distinct phases. In the first, called the explosive phase, a crater formed in the center of the newly created island. The brim of the crater was low, and the seawater would gush into the volcanic shaft. The ocean constantly came in contact with the upwelling magma, rapidly cooling it off and fracturing it into ash particles and cinder, which burst into the air with great explosions of steam.

884

SURTSEY

FIGURE 2 Lava flows over the southern part of Surtsey. Photograph

by Sturla Fridriksson.

ocean level. Because the depth of the sea was about 130 m, the total height of the island was approximately 300 m. GEOLOGICAL STUDIES

An island had formed and was secured with a cover of lava that coated the tuff cones like frosting on a layer cake. It demonstrated to geologists that the so-called table mountains or tuya of the Icelandic highlands had been formed in a similar way, as islands in glacial lakes, during the Ice Age. The total volume of material produced during the eruption was about 1.1 km3. Of this material, 70% was tephra and 30% was lava. The central part of Surtsey was made of tuff breccias, which have gradually transformed into palagonite. This process has been followed carefully by geologists on Surtsey. On the southern part of the island, two kinds of lava were formed: ropy lava, or pa-hoehoe, and rough ‘a‘a- lava. Gradually, the southwestern edge of the island was eroded by breaking waves, created by the prevailing southwest wind. In 40 years, the island has lost approximately 3 ha a year and is at present, in 2008, only 1.4 km2 in area and 155 m high. Some of the eroded material has been carried toward the leeward side at the northern edge, where it forms a spit of sand and boulder rims. The various changes in the landscape and the gradual wind and

FIGURE 3 An aerial view of Surtsey, taken in August 2007. The photo-

graph shows the gravel and sandy spit in the north, eroded water gullies (small, black furrows), the two cinder cones (light colored), the two main craters, the wind and ocean eroded edge of the island (at left), and the green spot of plant colonization. Photograph by Loftmyndir EHF, Reykjavik, Iceland.

water erosion of the island are constantly being followed, and geomorphologic evolution on Surtsey has been well recorded. The climate of Surtsey is maritime, windy and rainy with relatively warm winters (average 2 °C) and cool summers (average 10 °C) (Fig. 3). BIOLOGICAL INVESTIGATIONS

Biologists used the opportunity afforded by the formation of Surtsey to study various ecological phenomena and to understand how life would gradually colonize the island. They had to consider the location of the territory and investigate the source of available species for dispersal of life forms. A survey was made of the means of dispersal of plants and animals. The landing facilities were observed and living conditions on the island studied, as was the multiplication of various colonists. Records were made of every new introduction of plant and animal species and its performance on the island. Diaspores could be carried by air and ocean. Seeds drift in by ocean currents, and those surviving the saltwater would sometimes be able to establish themselves on the arid island. Grass knobs would fall from the nearby islands and were carried to Surtsey, taking with them insects and other small biota. Even fish aided in such transports. A number of sea purses, capsulated eggs of skate, drifted to Surtsey. A number of seeds attached themselves to the outer coat of the sea purses’ capsules when the purses lay on a distant shore. The purses acted as cargo boats, carrying their loads to this new land. Similarly, many diaspores, such as spores of molds, moss, lichens, and ferns, have been carried by air from the mainland of Iceland. Light seeds with feathers, such as cotton grass Eriophorum spp. and groundsel Senecio vulgaris, which are adapted to wind dispersal, have been found on the island, although these two species have not yet succeeded in growing there. Other species that have established themselves on Surtsey, such as dandelion and three types of willows, have probably grown from windborne seeds. Birds carry seeds attached to their feet or feathers or in their alimentary tracts. Migrating birds coming to Iceland in the spring were found to carry viable seeds long distances over the ocean. In this way the seed of the lady’s thumb plant, Persicaria, was brought by a snow bunting with the seed lodged in its gizzards. This snow bunting was of a race that lives in Scotland over the winter and migrates to Greenland in the spring. By analyzing the grit in the gizzards, it could be concluded that the minerals were of Scottish origin and that the seed had simultaneously been

SURTSEY

885

picked up in the Scottish Highlands. This was proof of a long-distance transport of seed by birds. Seagulls also carry vegetative material for their nest building, and they bring some vegetables to their young with their fish dinner, further adding to Surtsey’s ecosystem. Some insects may fly to Surtsey on their own wing support, but invertebrates are also carried by birds or can float to Surtsey on driftwood. Moths and butterflies from the continent of Europe are carried by favorable winds to Surtsey, and spiders come to Surtsey gliding on cobweb threads. Birds fly easily to Surtsey, and seals come there by swimming. Thus, a complex ecosystem has slowly evolved on the once barren island. MARINE LIFE

Monitoring of the marine life around Surtsey began in 1964 and has continued ever since.

zation of benthic animals began, and the pumice was invaded by numerous larvae. Already by November 1964, eight animals were found by a scraper from a depth of 70 m. Among them was the tubedwelling trumpet worm Pectinaria koreni. In 1966 some bottom samples were collected at a 100-m depth. The species found indicated the presence of the most common animals, such as 17 species of marine bristle worms (Polychaeta). This study has continued ever since. The marine fauna has developed normally, zonation is slowly forming, and gradually the communities on Surtsey are reaching the same balance as those of the adjacent islands. TERRESTRIAL LIFE

The terrestrial life on Surtsey has been intensely investigated. In the first years of the island’s existence, the presence of bacteria and molds was observed, and there already existed organic sources of energy on the island to support various living organisms, in the form of bird excrement as well as seaweed, kelp, and all sorts of driftwood.

Benthic Vegetation

Marine algae were among the first plants to drift to Surtsey. Knotted wrack, Ascophyllum nodosum, was occasionally washed ashore during the first summer of the island’s existence. Some marine bacteria and diatoms were pioneers of the marine benthic vegetation of Surtsey. In the summer of 1965, some filamentous green algae were added to the marine flora of the island. The rather slow process of colonization of the marine algae on Surtsey is due to the relative isolation of the island and the severe environmental conditions. In 1966, the rate of colonization increased markedly, and a noticeable zonation was formed. By 1970, the marine vegetation at Surtsey had increased to 35 species of algae and 11 of diatoms. In 1997 altogether 76 species had been recorded, omitting diatoms. The marine vegetation on Surtsey has apparently not yet reached the climax community. The kelp forests are slow in developing. However, many of the most common marine vegetation species of the adjacent areas have colonized the submarine pedestal of Surtsey. Pelagic Biota

The volcanic eruption had little effect on the pelagic biota of the ocean around the island. These waters are relatively rich in marine life, and the banks around the Vestmannaeyjar are some of the best fishing grounds in Iceland. Benthic Fauna

The benthic animals were destroyed during the eruption. While the island was still under formation, the coloni-

886

SURTSEY

Mosses

Although mosses easily occupy barren ground in Iceland, it was not until 1967 that the first moss colonies were discovered on Surtsey. The plants were all of the same species, common cord moss Funaria hygrometrica. The same year, a second location of mosses was observed at the edge of the central lava crater, where the colonies consisted of two species: cord moss and the silver moss Bryum argenteum. Since then, moss has gradually invaded the lava fields. Lichens

Lichens are common on lava fields in Iceland. Surprisingly, it was not until the summer of 1970 that the first lichen was found on Surtsey. A lava crust, northeast of the central crater, was then found occupied by small specimens of fructicose lichen Stereocaulon vesuvianum, a common pioneer lichen on Icelandic lava flows. The other habitat was situated on the outer side of the northern slope of the crater. Two species, disk lichen Trapelia coarctata and bullseye lichen Placopsis gelida, had settled this area. Lichens have since gradually occupied the lava fields. Vascular Plants

Vascular plants were earlier colonizers than most of the lower plants. A number of seeds were collected on the shore in 1964. They proved viable and capable of germinating. The first higher plant to colonize the island was found on the northern shore in 1965. These were small seedlings of the sea rocket Cakile edentula. This showed

that living seeds could be carried by sea at least 20 km, which is the distance from Heimaey to Surtsey. The second attempt of higher plants to invade Surtsey was in 1966, when four seedlings of sea lyme grass Leymus arenarius were found, and in 1967 a third species, the sea sandwort Honckenya peploides, started to grow on the island. The sea sandwort has since spread out widely over the sand-filled lava fields of Surtsey and has become the dominating plant of the island. Vascular plants are gradually colonizing Surtsey. During the first 43 years, 69 species of higher plants have been discovered on the island. There was a relatively constant addition of one new species a year in the first ten-year period, but then a ten-year stagnant period followed. The formation of a gull nesting colony from 1981 to 1985 became a turning point in the rate of arrival of new vascular plants, with a rapid increase of colonizers appearing in the following years. The species found have had a rather high percentage of survival, as 90% of these are recorded annually. Plant Communities

Since 1990 a number of permanent plots 10 × 10 m in area have been set up for studying the development of vegetation communities and soil in the various habitats on the island. At the same time, GPS instruments have been used to record locations of rare individuals discovered on Surtsey. The vegetation may be classified into four groups: (1) coastal community, dominated by the sea sandwort and lyme grass Leymus arenarius, (2) gravel flat community, which is more variable than the coastal community and has developed on a fine-textured substrate at the upper part of the lava, (3) skerry community on sand-filled lava, dominated by saltmarsh grass Puccinellia coarctata as well as scurvy grass Cochlearia officinalis, and (4) grassland community, fertilized by the colony of gulls since 1985. This community has smooth meadow grass Poa pratensis and red fescue Festuca rubra as dominant species, but it is also rich in various forbs. The succession of these communities has been followed intensely. Invertebrate Fauna

The first terrestrial invertebrate to be found on Surtsey was the midge Diamesa zernyi recorded in May 1964. A number of species were brought in by wind and birds or on various driftage during the first years of the island’s existence. In 1966 a small spider was found, which probably had glided from the mainland on its spinning thread. But there was a lack of food for the sustenance of these first settlers. With increased vegetation, the invertebrate fauna benefited, and in the first ten years, a total of 170

species had been identified on Surtsey. When the gull colony became established in 1985, there developed communities of invertebrates with their complex food chains. In 1993 the first earthworm was discovered, and the number of mites had climbed to 62 species. A total of 335 invertebrate species had been recorded on Surtsey in 2006. Bird Invasion

The bird life on Surtsey has been followed intensely. Seagulls were seen landing on the island two weeks after it was formed. The island is the southernmost landing place in Iceland for migrating birds. The place was therefore used to investigate migrants and to record the landing of stragglers. The entire island was monitored, and records were kept of visiting birds. A total of 89 species of birds have been seen on Surtsey. The black guillemot and the northern fulmar were the first birds to breed on Surtsey in 1970. Since then, 13 species of birds have been found nesting on the island. Of all these birds, the gulls have had the greatest impact on Surtsey. The lesser black-backed gulls started to nest in 1981. They began with only a couple of nests but ultimately formed a colony with 120 nests within four years. Their colony is the focus of pioneering lower life forms, worms, and insects, and to their breeding area the birds annually carry new plant material and fertilize diverse communities of plants with their droppings and offal. Mammals on Surtsey

Seals are the only mammals found on Surtsey. Since 1983 both common and gray seals have annually bred on the island. They find the surrounding ocean abundant in food, as do some whales that are often seen off the coast of Surtsey. THE FATE OF THE ISLAND

Surtsey offers more varieties of habitats than do the other outer islands. On the southeastern shore, coastal plant communities have developed. As sand blows into the vegetated area, dunes have been formed reaching a height of 1.5 m. The most conspicuous spot of vegetation on Surtsey is on the southern lava apron, where the lesser black-backed gull and the herring gull breed and fertilize the area heavily. There, the vegetation developed, and the plant community increased annually in size and density. In 20 years it covered an area of up to 10 ha. Almost 8% of Surtsey had turned green. Gradually red fescue will probably dominate the area and will eventually cover the top of Surtsey as it does on all the other islands in the archipelago. Now, as the island is

SURTSEY

887

FIGURE 4 Surtsey seen from the northern spit. The photograph shows

rounded boulders and sand that are partly overgrown by the sea sandwort. Water gullies are in the slope of the cinder cone, Austurbunki. The island already looks old. Photograph by Borgthor Magnusson.

fast diminishing in area, these habitats will start to become more uniform and to resemble those of the other outer islands that have lost all of their flatlands and only retain the hardened core. Some of the other islands of the archipelago are 6000 years old; by that measure, it can be assumed that Surtsey will last for thousands of years (Fig. 4). SEE ALSO THE FOLLOWING ARTICLES

Iceland / Island Formation / Lava and Ash / Seabirds / Vegetation / Volcanic Islands FURTHER READING

Einarsson, T. 1966. Gosið í Surtsey. Reykjavik: Heimskringla. Fridriksson, S. 1975. Surtsey: evolution of life on a volcanic island. London: Butterworths. Fridriksson, S. 2005. Surtsey: ecosystems formed. Reykjavik: Vardi, the Surtsey Research Society. Nomination of Surtsey to the UNESCO World Heritage List 2007. Reykjavik: Institute of Natural History. Surtsey Research Progress Reports 1. 1965 to 12. 2006. Reykjavik: Surtsey Research Society. Thorarinsson, S. 1967. Surtsey: the new island in the North Atlantic. New York: Viking Press Inc.

SUSTAINABILITY R. R. THAMAN University of the South Pacific, Suva, Fiji

Sustainability, or “sustainable development,” is the ability of island nations or communities to acquire the income needed to purchase material and non-material goods

888

S U S TA I N A B I L I T Y

from the modern cash economy that are needed to make life healthier, safer, more productive, and more enjoyable, but, at the same time, doing so without destroying the local natural and cultural capital needed for the material and cultural survival of future generations. Sustainability is about balancing these two, often conflicting, objectives: achieving the right balance between cash and subsistence economies, between self-reliance and dependency. For many traditional and indigenous peoples, sustainability is not so much about “production,” but rather about the “reproduction” and survival of the cultures and communities that have proved relatively sustainable for millennia. It is suggested that, for islands, their biodiversity and ethnobiodiversity constitute the most important natural and cultural capital, which must be managed as the living bank account and foundation for sustainability. Such diversity has been the foundation for “subsistence affluence” and self-reliance for millennia, a foundation that is now seriously threatened. EROSION OF BIODIVERSITY-BASED SUSTAINABILITY ON ISLANDS AND THE CBD

The accelerating erosion of the Earth’s biodiversity, along with climate change, population growth, poverty, and overconsumption (which are closely linked), constitute perhaps the most serious obstacles to sustainable human occupation of the Earth. The recognition of this mainly human-induced biodiversity crisis resulted in the ratification by most nations, including most island nations, of the United Nations Convention on Biological Diversity (CBD), launched at the United Nations Conference on Environment and Development (UNCED), the “Earth Summit,” held in Rio de Janeiro in 1992. Islands, particularly small islands, with their limited and fragile biodiversity inheritances, are among the most seriously affected. Although most island nations are signatories to the CBD, the evidence is that most of the island “developmental elite,” when asked what “biodiversity” is, neither understand nor can define it in any detail, nor do they recognize that we have a “biodiversity crisis” that threatens the very sustainability of the “development” they promote. They also do not seem to understand or appreciate the extreme uniqueness and fragility of island biodiversity, which has become the focus of the most recent work program of the CBD, the Work Program on Island Biodiversity, approved in Curitiba, Brazil, in April 2006. For most small island nations and communities, there are few options for modern, market-driven industrial or export-oriented development, and the conservation, sustainable use, and equitable sharing of the benefits from

biodiversity, the three main objectives of the CBD, constitute the most important foundation for sustainable development. For most small island developing states (SIDS), the sustainability of their island cultures and economic well being is dependent on their island biodiversity inheritances. Although the main examples used here are the islands of the tropical Pacific Ocean, where biodiversity inheritances are relatively intact, the importance of biodiversity as a foundation for sustainability holds true for most islands. Sadly, despite the current apparent state of well-being and the absence of real poverty and associated social breakdown in the Pacific Islands, relative to other areas of the “developing” world, there is clearly a “biodiversity crisis” of unprecedented proportions” that undermines the sustainability of island life. There are three simple messages: (1) The conservation and sustainable use of biodiversity and the protection of what has been referred to as “subsistence affluence” constitute, perhaps, the single most important foundation for short- and long-term poverty alleviation and sustainability on islands; (2) there is clearly a “biodiversity crisis of unprecedented proportions,” both on land and in the seas surrounding islands, which undermines the sustainability and the integrity of island cultures; and (3) if not addressed as an integral component of all development thinking, the ultimate result of the erosion of island biodiversity will be the abject poverty that we have come to associate with continental societies that have pillaged their biodiversity, leaving behind degraded and life-depleted deserts, scrublands, polluted lakes and rivers, dying reefs, squatter settlements, and associated communities that are now among the poorest of the poor. To achieve sustainability on islands, there is a need for rethinking the allocation of aid and funding for education, research, business development, and so forth toward the realization of a much more environmentally and culturally benign form of development, a development that builds on existing biocultural foundations rather than eroding or destroying these foundations. If such a strategy were to be adopted, it might just be possible to avoid the inevitable tightening of the vicious circle of resource depletion, environmental degradation, loss of biodiversity, and resultant economic and cultural breakdown that has doomed so many developing and “developed” (read “biodiversity destroyed”) countries to a future of abject poverty and has marginalized their time-tested biodiversity-use traditions. ISLAND BIODIVERSITY: A DEFINITION

Island biodiversity is defined as including (1) diversity of island types, (2) diversity of island and associ-

ated marine ecosystems and habitats, (3) species and taxonomic diversity within these ecosystems, (4) genetic diversity within and between species populations, and (5) ethnobiodiversity. Island diversity includes the almost unbelievable diversity of island types found in the tropical Pacific Ocean and elsewhere on Earth. The diversity among islands affects the richness of respective biodiversity inheritances and the ability of governments, nongovernmental organizations (NGOs), and local communities to mount systematic programs for conservation and sustainable use. Island ecosystem diversity includes all natural and “cultural” terrestrial, freshwater, and marine ecosystems and habitat types found on or around these islands and the “ecosystem services” they provide. These include terrestrial and freshwater ecosystems such as tropical lowland and upland forests; swamp and riverine forests; mangrove and coastal beach forests; woodlands and savannas; meadows; scrublands; deserts; marshes; rivers, streams, and lakes; marine ecosystems such as algal and seagrass beds; beaches; a range of reef and lagoon types; estuaries; offshore slopes, terraces, shelves, canyons, sea mounts and abyssal plains; and subsets of these, such as seabird rookeries, sea turtle nesting areas, and upwelling systems in the ocean. Also included are cultural or humanized ecosystems, such as secondary and fallow forests, shifting and permanent agricultural lands, grazing lands and livestock production systems, fish ponds and reservoirs, urban gardens, and towns. Table 1 is a simplified classification of Pacific Island ecosystems that can be used at the community or landowner, national, and regional levels, and in schools, to promote biodiversity conservation and serve as a basis for systematically gathering traditional knowledge (ethnobiodiversity) about these ecosystems. Species and taxonomic diversity includes all species and taxonomic groupings (e.g., vascular and non-vascular plants, vertebrates and invertebrates, and microorganisms) found in island ecosystems and their surrounding seas. Table 2 attempts to summarize some of the main taxonomic groups from both scientific and indigenous Pacific Island perspectives. It is stressed that almost all categories are known to, and have recognized economic, cultural, and ecological utility for, Pacific Island societies. Genetic diversity includes all subspecies, genetic types, breeds, cultivars, or varieties of wild and domesticated or cultivated plants and animals found in these ecosystems (e.g., cultivars of yams, taros, sweet potatoes, sugarcane, coconuts, breadfruit, mangoes, pandanus, etc., as well as breeds of pigs, chickens, dogs, and other animals), and, for the purpose of protecting intellectual property rights, all chemical extracts

S U S TA I N A B I L I T Y

889

TABLE 1

Terrestrial, Freshwater, and Marine Ecosystems of the Pacific Islands Terrestrial/freshwater ecosystems

Lowland native forest Upland or montane rain forest Mature fallow forest Plantation forest Grassland/woodland Scrubland/scrub-fernlands Shifting agricultural land Permanent/semi-permanent agricultural land Plantations Pasture Houseyard/urban gardens Intensive livestock holdings Ruderal sites Wetlands/swamps Rivers/streams/lakes/ponds Fish ponds/aquaculture Mangrovesaa Coastal strand vegetation Beaches and dunes Bare rock Caves Built/urban areas Marine ecosystems

Mangrovesa Estuaries Intertidal zone Lagoons/bays Fish ponds/maricultural areas Coral reefs Island shelf/reef platform/ocean floor Open ocean note: Ecosystems listed are those that (1) constitute major resource-use zones and (2) could serve as the focus for community-based, national, and regional biodiversity conservation in the Pacific Islands. Adapted from Thaman 1994a. a Mangroves are listed as both terrestrial and marine ecosystems.

from these organisms. In most traditional polycultural Pacific Island shifting agricultural systems, for example, there are invariably many distinct cultivars of staple food plants, all of which have differential utilities and levels of resistance to drought, tropical cyclones, diseases, floods, and so forth. Genetic diversity also includes human lineages. “Ethnobiodiversity” is defined as the knowledge, uses, beliefs, resource-use systems, management systems and conservation practices, taxonomies, and language that a given society (or ethnic group), including the modern scientific community, has for their islands, ecosystems, species, taxa, and genetic diversity. It is stressed that this final category or “level” of biodiversity is central to the definition of biodiversity itself because, in the Pacific Islands and in most other traditional or indigenous island societies, people and their knowledge, traditions, and spirituality are seen as inseparable from their terrestrial, freshwater, and marine ecosystems

890

S U S TA I N A B I L I T Y

(e.g., in western Melanesia as embodied in the Melanesia pidgin concepts of kastom/custom or ples/place; in Fiji in the concepts of vanua/land and iqoliqoli/fisheries; in Polynesia under the all-encompassing pan-Polynesian concept of land/ fonua, fanua, fenua, whenua, henua or ‘enua, depending on where you are; or the concepts of te aba and bwirej, in Kiribati and the Marshall Islands, respectively). ISLANDS AS BIODIVERSITY HOTSPOTS AND COOLSPOTS

Although biodiversity is the foundation for sustainability on most islands, there is great diversity among and disparity between biological inheritances. The larger continental islands (e.g., the islands of New Guinea, New Caledonia, and Madagascar) and the isolated high volcanic islands (Hawai‘i and the Galápagos) have among the richest biodiversity inheritances and highest levels of endemism on Earth and are considered global “biodiversity hotspots.” Many, such as Hawai‘i and the Galápagos, have historically served as laboratories of evolution and extinction. Many of these “hotspots” also have considerable natural resource endowments and potential for modern economic development. Other islands with reasonably high levels of endemism and relatively rich biodiversity, but with less modern development potential, include the Solomon Islands, Vanuatu, and Fiji in Melanesia; Samoa, Tahiti, and some of the other islands of French Polynesia; and Palau and Pohnpei in Micronesia. Conversely, many small, low-lying islands, particularly atolls, have among the poorest or least diverse terrestrial and freshwater floras and faunas on Earth, with virtually no endemic species. They, in effect, constitute the Earth’s “biodiversity coolspots.” Moreover, their terrestrial floras and faunas are among the most highly degraded and threatened on Earth. However, this fragile, very limited biodiversity inheritance is, for most isolated small island communities and nations, their only foundation for survival! In terms of marine biodiversity, the disparity between large and small islands is not so great, although there is still an attenuation or dropoff in the number of species from west to east as we move from the center of diversity in Malesia (the tropical island world between Asia and Australia including Indonesia, the islands of Malaysia, the Philippines, New Guinea, and Taiwan) and the Indo–western Pacific to the small, more isolated, islands of the central Pacific. In summary, although the larger “hotspot” high islands have far greater diversity of ecosystems than the smaller, low-lying “coolspot” islands, all islands have terrestrial and marine ecosystems, and in the case of larger islands, mon-

TABLE 2

Biological Resources of Community-Level Ecosystems in the Pacific Islands Class

Subclass

Lower life forms

Specific Type

Utility

Bacteria Viruses

E,s,c E,s,c

Plants

Indigenous Aboriginal introductions Recent introductions Wild plants Domesticated plants Food plants Non-food plants Terrestrial Freshwater Marine Trees

Phytoplankton Algae Fungi Mosses Other lower plants Ferns Herbs/forbs Grasses/sedges Vines Shrubs E,C,C

E,s,c E,S,C E,s,c E,s E,s,c E,S,C E,S,C E,S,C E,S,C E,S,C

Animals

Indigenous Aboriginal introductions Recent introductions Wild animals Domesticated animals Food species Non-food species Terrestrial Freshwater Marine

Protozoa Zooplankton Sponges Corals Jellyfish Worms Molluscs Insects Crustaceans Echinoderms Other invertebrates Finfish Amphibians Reptiles Birds Non-human mammals Humans

E,s,c E,s,c E,s,c E,S,c e,s,c E,s,c E,S,C E,s,c E,S,C E,S,C E,s,c E,S,C E,s,c E,S,C E,S,C E,S,C E,S,C

note: Classes, subclasses, specfic types, and the utility of terrestrial, freshwater, and marine resources that constitute the pool of ecologically important and functionally useful biological resources of community-level ecosystems in the Pacific islands. Under “Utility,” E, S, and C = direct major ecological, subsistence, or commercial/export utility to people at the community and national level in Melanesia, Polynesia, or Micronesia, and e, s, and c = minor or indirect ecological, subsistence, or commercial/export importance (e.g., amphibians are found on very few islands and plankton is of indirect importance to commercial tuna fishing in terms of its importance in marine food chains). It must be stressed that taxa in some categories may also be harmful or have a negative impact on sustainable development (e.g., pathogenic viruses or bacteria, mosquitoes, etc.).

tane freshwater ecosystems, that provide for most of the subsistence and cash needs of their resident rural communities. It is stressed that, regardless of island type, size, or degree of isolation, all islands have ecosystems, plants, animals and microorganisms that are critical to the continuing health and survival of each island’s biodiversity itself and to the human communities that depend on the islands for their ecological, economic, and cultural survival. For many of the smaller, more isolated islands, their dependence on biodiversity and associated ethnobiodiversity is an “obligate dependence” because there are no other options. The “pyramid of sustainable development” (Fig. 1) based on the conservation of biodiversity is an attempt to illustrate the critical importance of biodiversity and associated ethnobiodiversity as factors that underpin all forms of development, including subsistence production (the main protection

against poverty) and production for local sale and export, both of which provide an economic, cultural, and material base for sustainability. CURRENT STATUS OF ISLAND BIODIVERSITY

Although all island societies have critically important terrestrial, freshwater, and marine biodiversity inheritances, including associated ethnobiodiversity, as we begin the twenty-first century, there are frightening signs of the loss or endangerment of this living inheritance that has supported island societies for millennia. These include a wide range of terrestrial, freshwater, and marine ecosystems and species that are now rare or endangered. This is not a new phenomenon, but rather a phenomenon that began long before European contact with the islands, when the early Pacific Islanders severely deforested many islands,

S U S TA I N A B I L I T Y

891

ALL URBAN ACTIVITIES (Politics, Business, Govemment, Banking, Law, Teaching. etc.)

EXPORT PRODUCTION (Timber, Crops, Fish, Minerals and Tourism)

PRODUCTION FOR LOCAL SALE

diversity, larger areas of strategic ecosystems, larger populations of individual organisms, and lower human population densities, there is, nevertheless, a serious need to protect designated ecosystems and specific endangered organisms throughout the Pacific. Even in the larger island countries, there exist many small outer island communities with limited resources and high population densities that experience the same trends of degradation and loss of biodiversity as the communities in the smaller “coolspot” islands of the eastern Pacific. This assessment is based on over 15 years’ field observation and in-depth interviews with local communities. THREATS TO ISLAND BIODIVERSITY

(Food, Fish, Handicrafts, etc.)

SUBSISTENCE PRODUCTION (Food, Fuel, Medicines, Construction Materials, etc)

SPECIES, TAXONOMIC AND GENETIC DIVERSITY (Plants, Animals and Micro-Organisms) ECOSYSTEMS (NATURAL AND CULTURAL) (Terrestrial, Freshwater and Marine) TRADITIONAL AND MODERN ETHNOBIOLOGICAL KNOWLEDGE (Uses, Knowledge Beliefs, Management Systems, Taxonomies and Language that a Culture or Society has for its Biodiversity)

FIGURE 1 Pyramid of Island Sustainable Development illustrating the

critical importance of (1) biodiversity as a foundation for most forms of development, including subsistence production and production for local sale and export, which provide the economic and material base for sustainability, and (2) “ethnobiodiversity”—the knowledge that underpins the conservation and sustainable use of limited island biodiversity inheritances.

including islands in the Marquesas, the Austral and Hawaiian Islands, and Easter Island (Rapa Nui). They also brought many bird species to extinction or extirpation (local extinction), extinguished much of the lowland biota throughout much of Polynesia, and brought to extirpation some species of giant clams in Fiji, Tonga, and Samoa. The process has, however, intensified, and the identification of the ecosystems, species, and genetic resources that are rare, endangered, or in need of protection or re-establishment is critical to the success of biodiversity conservation efforts and for the maintenance of cash and non-cash incomes (subsistence affluence), at all levels (regional, national, and local). An attempt to do this for the entire Pacific is shown in Table 3. As can be seen from Table 3, there is a wide range of ecosystems and organisms that are in need of protection. Although the need for the protection of these organisms is slightly less in the larger “biodiversity hotspot” islands of Melanesia, where there is greater ecosystem and biotic

892

S U S TA I N A B I L I T Y

Table 4 is an attempt to classify the most significant threats to the conservation and sustainable use of island biodiversity. They include (1) direct threats, which seriously degrade and upset the stability of natural and cultural ecosystems and their biodiversity and (2) social, institutional, or infrastructural activities or phenomena that indirectly threaten or undermine capacity to conserve or sustainably use island biodiversity. Most of these threats are of both global and local concern and can be addressed, in some way, at all levels. Some, however, such as global warming, the depletion of stratospheric ozone, the international trade of endangered or potentially invasive plants and animals, and pest and disease infestations and epidemics, are best dealt with regionally or internationally, whereas the protection of endangered or threatened species and ecosystems are, perhaps, best dealt with at the national or community levels. Similarly, many of these threats overlap or feed into each other and, if not addressed in some way, could lead to a dangerous negative synergistic effect and to the collapse of entire island ecosystems or biological communities and the countries and cultures that depend on them. Conversely, if a number of threats are addressed simultaneously, the result could be very positively synergistic and could lead to significant gains in the mainstreaming of conservation and sustainable use of biodiversity in the Pacific Islands. It is beyond the scope of this article to discuss each of these threats. Suffice it to say that all of these threats, many of which are interrelated, are of serious concern in varying degrees throughout the Pacific Islands and other islands globally. If efforts to promote island biodiversity conservation are to be successful, these threats need to be systematically addressed at a range of different levels (government, school systems, local communities), by different agencies (e.g., NGOs, private enterprises, funding/aid agencies), and in a range of different international forums.

TABLE 3

Pacific Islands Taxa in Need of Protection Category

Melanesia

Polynesia

Micronesia

++ ++ ++ ++ +++ ++ ++ ++ ++ +++ ++ ++ +++

+++ +++ +++ ++ +++ +++ +++ ++ ++ +++ +++ +++ +++

+++ +++ +++ ++ +++ +++ +++ +++ +++ +++ +++ +++ +++

++ ++ ++ ++ ++ ++ ++ ++ +++ +++ +++ +++ +++

+++ +++ +++ +++ +++ +++ +++ +++ NP ++ +++ +++ +++

+++ +++ +++ +++ +++ +++ +++ +++ NP ++ +++ +++ +++

++ ++ ++ +++ ++ ++ +++

+++ +++ ++ +++ +++ NP +

+++ +++ ++? +++ +++ NP +

++ +++ +++ +++ +++ ++ ++ ++ +++ ++ +++ +++ +++ +++ +++

++ +++ +++ +++ +++ +++ ++ ++ +++ +++ +++ +++ – +++ +++

++ +++ +++ +++ +++ +++ ++ ++ +++ +++ +++ +++ + +++ +++

Ecosystem

Uninhabited islands Coastal littoral and mangrove forests Lowland forests Montane/cloud forests Rivers and lakes Wetlands/swamps Shifting agroforestry lands and agroforests Semi-permanent/intensive agricultural areas Houseyard and village gardens Selected productive reefs Intertidal zone and seagrass beds Reef passages Coral reefs Terrestrial organisms

Native coastal and mangrove plants Native inland trees and plants Cultivated trees and plants Plant cultivars/varieties Native insects/arthropods Land crabs Native molluscs Other native invertebrates Native amphibians Native reptiles Native birds Native mammals Humans (ethnobiological knowledge) Freshwater organisms

Freshwater plants Crustaceans Shellfish Insects Finfish/eels Amphibians Reptiles Marine organisms

Seaweeds (marine macro-algae) Sea grasses Stony reef-forming corals Shellfish (giant clams, trochus, turban snail, pearl oyster, triton) Bêche-de-mer/holothurians Crabs, lobsters, mantis shrimp Reef and lagoon fish Eels (conger, moray) Large demersal finfish (rockcods, wrasses, parrotfish) Sharks and rays Billfish Turtles Crocodiles Seabirds Mammals (whales, dolphins, dugongs)

note: Ecosystems and groups or taxa of terrestrial, freshwater, and marine plants and animals that are rare, endangered, or in short supply and in need of protection in the Pacific Islands (+++ = of serious widespread concern and in need of immediate protection; ++ = of some widespread concern or of serious concern in specific areas; + = of limited or localized concern; – = of no concern; NP = not present).

TABLE 4

Threats to Biodiversity in the Pacific Islands Direct Threats to Biodiversity

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

High frequency of extreme events/natural disasters Global warming/eustatic sea-level rise Stratospheric ozone depletion and increasing UV-B radiation Breakdown and simplification of the species composition and trophic structure of terrestrial, freshwater, and marine ecosystems and ecosystem functions Degradation of uninhabited islands Upland and inland deforestation and forest degradation Coastal and mangrove deforestation and degradation Degradation of freshwater resources and ecosystems Agricultural simplification and degradation, agrodeforestation, and the loss of biodiversity in agricultural systems Overgrazing and degradation of biodiversity by domestic livestock Destruction caused by feral animals Alien invasive plants and animals Pest and disease infestations and epidemics Soil degradation and accelerated soil erosion Fire Destruction and degradation of productive marine ecosystems and disruption or change in the dynamics of marine ecosystems Overuse/overexploitation/unsustainable use of terrestrial plant and animal resources Overfishing/overexploitation/unsustainable use of marine resources Use of destructive fishing technologies Illegal fishing Pollution of freshwater resources Air pollution Marine pollution Indiscriminate and increasing use of pesticides Hazardous/toxic waste disposal Nuclear/radioactive pollution and contamination

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Uncontrolled population growth Loss of traditional ethnobiological/ethnobiological knowledge Breakdown in traditional diversified subsistence economy Inadequate modern scientific baseline knowledge of the nature and status of biodiversity Inadequate systems of marine and terrestrial conservation areas Inadequate capacity to deal with terrestrial, freshwater, and marine invasive species Inadequate legislation/legal instruments Inadequate infrastructure/capacity for biodiversity conservation Inappropriate modern education and curricula Rapid and uncontrolled urbanization Unforeseen large-scale developments Free trade/globalization and increasing international free trade in biodiversity Poverty and economic deterioration Gender inequity in the control, use, and management of biodiversity Political instability and political ignorance or lack of political will to commit to conservation

Social, Institutional, and Infrastructural Threats

note: Table lists significant reasons for the loss of biodiversity or threats to biodiversity and biodiversity conservation in the Pacific Islands that can be addressed in the mainstreaming of biodiversity conservation at the regional, national, and local community levels.

One threat that should be singled out, however, is the loss or lack of ethnobiodiversity, because the loss of traditional and indigenous knowledge about the uses, beliefs, management systems, taxonomies, and language, and the lack of modern scientific knowledge related to biodiversity, could be among the most serious obstacles to successful biodiversity conservation and sustainability on islands. As suggested diagrammatically in the “Pyramid of Sustainable Island Development,” good knowledge, both traditional and modern, about biodiversity is the basic requirement

894

S U S TA I N A B I L I T Y

for all island biodiversity conservation and sustainability. At the local level, site-based biodiversity conservation will be problematic if local people cannot marry traditional conservation strategies with modern scientific models and findings as part of co-management systems. If local people no longer know the local names, uses, and management systems for their biodiversity, chances are that they will not place a priority on its preservation. At the same time the modern scientific community is bemoaning the lack of financial support for the training of good modern

taxonomists to replace those who pass away, many of the best traditional Pacific Island men and women scientists and taxonomists are dying and are not being replaced by a younger generation that is less interested in the natural world. Along with them dies traditional ethnobiodiversity that has been accumulated over thousands of years in close contact with the island environment. MAINSTREAMING BIODIVERSITY CONSERVATION AS A FOUNDATION FOR SUSTAINABLE ISLAND LIFE

The conservation and sustainable use of island biodiversity as a foundation for sustainability is, perhaps, the single most important obligation we have, as the current generation, to future generations of islanders. The challenge before us is clearly to get this message out to all island stakeholders, at all levels, and from all walks of life, and to get them to understand its real meaning. Hopefully, as a result, they will also take on board, as individuals and as groups or institutions, the protection and wise use of this living inheritance as central to their philosophies of life and theories of island development. This should be the objective of “mainstreaming” island biodiversity conservation. Fortunately, there are an increasing number of local, national, and international initiatives that are spreading this message and implementing programs, particularly community-based programs, to promote the conservation and sustainable use of biodiversity in the Pacific Islands. Many of these initiatives are led by consortia or networks of NGOs, regional organizations, national and local government agencies, the private sector, and local landowners and resource users and are funded by an increasing number of entities including the MacArthur and Packard Foundations and Australian, New Zealand, French, American, and Japanese funding agencies. Prominent among these include international conservation NGOs, such as the WWF— World Wide Fund for Nature (known in the United States as the World Wildlife Fund), The Nature Conservancy (TNC), Conservation International (CI), the Wildlife Conservation Society (WCS), Birdlife International, the Foundation for the Peoples of the South Pacific International (FSPI) and its national counterparts, the World Conservation Union (IUCN), the United Nations Education, Social and Cultural Organization (UNESCO), as well as Pacific Island regional organizations, such as the Secretariat of the Pacific Regional Environment Programme (SPREP), the Secretariat of the Pacific Community (SPC), the Secretariat for the South Pacific Applied Geoscience Commission (SOPAC), the University of the South Pacific (USP), the University of Hawaii, the University of Guam, the

Bishop Museum, and the Institute for Research and Development (IRD) in the French Territories. A number of local NGOs are also participating, such as Nature Fiji–Mareqeti Viti and Micronesians in Island Conservation (MIC). Both alone and together, with funding from bilateral and multilateral sources and foundations, these entities have, over the last decade, established regional, national, and local community–based networks of terrestrial and marine protected areas, produced educational materials, assessed the status of biodiversity, and championed programs and projects designed to address the major threats to biodiversity conservation and sustainable use and to create alternative sources of revenue to take pressure off threatened biodiversity. In an increasing number of cases, governments have sanctioned and become integral partners in these initiatives, and in the many successful cases the initiatives have focused on the full involvement of local communities and resources users and owners in the planning, implementation, and monitoring of the initiatives. Some of the more notable initiatives include the Fiji and Asia-Pacific Locally Managed Marine Areas Networks (FLMMA and APLMMA), the Pacific-Asia Biodiversity Transect Network (PABITRA), the Marine Aquarium Council (MAC), the South Pacific Regional Initiative on Forest Genetic Resources (SPRIG), the Micronesia Conservation Trust (MCT), the Micronesians in Island Conservation Network (MIC), the Micronesian Challenge, the accession to the Ramsar Wetlands Convention by five Pacific countries, the Pacific Invasives Initiative (PII), and the Pacific Invasives Learning Network (PILN). Under the FLMMA and wider APLMMA, over the past decade, some 300 locally managed marine areas with associated management plans have been established, over 200 of which are in Fiji. The main partners in this initiative are Fiji, Solomon Islands, the Federated States of Micronesia, Palau and Papua New Guinea in the Pacific, and the Philippines and Indonesia in Asia. Results indicate that fisheries stocks have increased in most areas and that local communities are taking biodiversity conservation into their own hands. Much of the success has been based on the sharing of success stories and methodologies for communitybased conservation. Also in Fiji, prohibited fishing zones have been set aside in Fiji’s Great Sea Reef (known locally as Cakau Levu) to conserve Fiji’s most extensive area of coral reef, which covers more than 200,000 km2 and is home to thousands of marine species, including marine turtles, dolphins, sharks, and 43 new hard coral species. The reef is also an important fishing ground for local communities. For over a decade, PABITRA has promoted comparative studies and built local research capacity for the study

S U S TA I N A B I L I T Y

895

of island ecosystems across the Pacific, with a focus on local people’s perceptions and multiple ecosystem-use systems along the lines of the Hawaiian ahu pua‘a, “summit to sea” or “ridge to reef” integrated ecosystem and land-use approach. Under PABITRA, stress has also been placed on the importance of the protection, recording, and application of indigenous knowledge about island biodiversity and ethnobiodiversity. The Marine Aquarium Council (MAC) has, over the past ten years, established a program to certify entities involved in the trade of aquarium fish, coral, and other marine organisms. The program has been very successful in a number of areas of the Pacific in promoting best practices in this area. There are similar, but so far less successful, initiatives addressing certification of timber exploitation from indigenous rain forests in some Pacific countries, and the SPRIG program has had, as its main focus, the protection, collection, and propagation of forest genetic diversity as a basis for the development of sustainable plantation forestry and agroforestry systems in the Pacific Islands. The Micronesian Challenge, which was launched in 2006 by the president of Palau, involves committing at least 30% of nearshore marine and 20% of forest resources across Micronesia to conservation by 2020. Similarly, in 2006, the Kiribati government designated the atolls and marine area of the Phoenix Islands as the world’s third largest marine protected area, and uninhabited Ant (Ahnd) Atoll in Pohnpei State of the Federated States of Micronesia has been recently designated as a UNESCO Biosphere Reserve, as has the Lake Tenggano area of East Rennell, in the Solomon Islands. Also, as of 2008, five Pacific countries (Papua New Guinea, Fiji, Samoa, Palau, and the Marshall Islands) have ratified the Ramsar Wetland Convention and designated “Wetlands of International Importance” that are now under some form of conservation. The Pacific Invasives Initiative (PII) and the Pacific Invasives Learning Network (PILN) are focused on reducing the impacts of invasive species on island economies and ecoystems through prevention and management of invasive species and through building capacity at the national level to deal with invasive species. To mainstream biodiversity conservation on islands is to ensure that all individual and institutional stakeholders involved in island development (local communities, private enterprises, governments, NGOs, the aid community, and other international agencies) include the conservation and sustainable use of biodiversity as a priority concern, if not as the most important precondition, for our own individual or institutional well being. To make

896

S U S TA I N A B I L I T Y

this vision a reality, there is an urgent need for all stakeholders to clearly understand (1) what island biodiversity really is, its current status, and why biodiversity conservation should be a priority concern as a basis for sustainable development and the well being of all and (2) what we can do as individuals and institutions to promote its conservation. Once this level of awareness of the issues is achieved (and there are, as outlined above, a number of very positive developments in this direction), it is hoped that the incentives and reward systems will serve to motivate all stakeholders to address the increasingly serious threats to island biodiversity, even if it requires sometimes drastic changes in the way we conduct, and with whom we conduct, international and local business. If efforts to mainstream the conservation of this “living foundation” are successful, then those who live on islands, their children, and their children’s children will hopefully be able to walk along island mountain trails, streams, shores, and reefs and through island towns, and continue to marvel at the cowries, fishes, trees, flowers, birds, and other island plants and animals that, for thousands of millennia, have provided the cultural, economic, and ecological foundation for the rich (not poor) biodiverse island cultures of today. If these efforts are not successful, then widespread impoverishment of future generations and their environments and nature-based cultures and widespread unsustainability will be the result. SEE ALSO THE FOLLOWING ARTICLES

Ethnobotany / Marine Protected Areas / Reef Ecology and Conservation FURTHER READING

Clarke, W. C., and R. R. Thaman, eds. 1993. Pacific Island agroforestry: systems for sustainability. Tokyo: United Nations University Press. Kay, E. A. 1999. Biogeography, in The Pacific Islands: environment and society. M. Rapaport, ed. Honolulu, HI: The Bess Press, 353–365. Steadman, D. W. 1995. Prehistoric extinctions of Pacific Island birds: biodiversity meets zooarchaeology. Science 267: 1123–1131. Stoddart, D. R. 1992. Biogeography of the tropical Pacific. Pacific Science 46: 276–293. Thaman, R. R. 1999. Pacific Island biodiversity on the eve of the 21st century: current status and challenges for its conservation and sustainable use. Pacific Science Association Information Bulletin 51: 1–37. Thaman, R. R. 2004. Sustaining culture and biodiversity in Pacific Islands with local and indigenous knowledge. Pacific Ecologist Autumn–Winter: 43–48. Thaman, R. R. 2005. Biodiversity is the key to food security. Spore 117 (June): 1–3.

SVALBARD SEE ARCTIC REGION

T TAIWAN, BIOLOGY MAN-MIAO YANG National Chung Hsing University, Taichung, Taiwan

KUANG-YING HUANG Yangmingshan National Park, Taipei, Taiwan

Taiwan, an island of approximately 36,000 km2, is situated in the western Pacific Ocean and separated from China by the Taiwan Strait. It is also known as Formosa (Fig. 1), meaning “the beautiful island,” a name given to it by Portuguese mariners who encountered it in the sixteenth century. Around the main island of Taiwan, there are more than 80 smaller islands, including the graniteorigin continental islands, volcanic oceanic islands, cubic basalt-formed Pescadores (Penghu archipelago) (Fig. 2A), and coral-reef based islands (e.g., Pratas atoll [Fig. 2B]). This article focuses on the main island of Taiwan. LANDSCAPE AND CLIMATE

Lying directly in the center of the East Asian islands, Taiwan belongs to the Pacific Ring of Fire, an area of dense volcanic action. The island of Taiwan is located at the place where the Philippine plate is subducting under the Eurasian plate. Consequently, the elevated Central mountain range, stretching from north to south in the middle of Taiwan, forms the spine of the island and is a natural watershed for the rivers. In addition, there are four additional parallel mountain ranges: to the west is the Hsuehshan (Snow Mountain) range in the north, the Yushan mountain range in the middle, and the Alisan

FIGURE 1 There are more than 80 islands in the area of Taiwan, and

they have diverse origins. This map of the main island Taiwan (Formosa) in 1901 clearly shows its relative position in East Asia and the distribution of various indigenous people living on it. Courtesy of SMC Publishing Inc., after James W. Davidson (1901).

897

TABLE 1

Altitudinal Vegetation Zones and Their Temperature Ranges in Central Taiwan Altitudinal zone

Vegetation zone

Elevation (m)

Alpine Subalpine Upper montane Montane Submontane Foothill

Juniperus–Rhododendron zone Abies zone Tsuga–Picea zone Quercus zone Machilus–Castanopsis zone Ficus–Machilus zone

3600–4000 3100–3600 2500–3100 1500–2500 500–1500 0–500

Mean annual temperature (°C)

23

source: Su (1984).

mountain range in the southeast; to the east is the lower Coastal mountain range. More than 200 mountain peaks in Taiwan exceed an altitude of 3000 m, with Yushan, or Mt. Jade, the highest peak in East Asia, reaching 3952 m. Some vestiges of the glacial age may be seen in the alpine areas of these mountains, such as a glacier cirque on Snow Mountain. Short and rushing rivers, including 19 primaries, 32 tributaries, and 100 minors, cut through these steep mountains, with the longest one, Chuo-Shuei River, flowing 186 km into the Taiwan Strait. Although two-thirds of the island is covered by mountains, the remaining area, which mainly lies along the western coast, consists of hills, tablelands, coastal plains, and basins. The length of coastal line around the main island is 1250 km, and the southern part features rich coral reefs. Lying on the Tropic of Cancer and extending from 21°45’25’’ to 25°56’20’’ N latitude, Taiwan features a climate that is mainly subtropical, with the southern part being tropical; areas at higher elevations are considered temperate. Snow may be seen on high mountain caps, but the lowland rarely has temperatures below 10 °C. The average temperature is 28 °C in summer and 14 °C in winter, and annual precipitation is higher than 2000 mm in most areas. In general, Taiwan has an oceanic and subtropical monsoon climate, evidently affected by its topography. Spring monsoon is also known as “plum rain,” and the length varies from year to year. Although typhoons often attack Taiwan in the summer and autumn, causing floods and damage, they also bring plentiful rainfall to fill the reservoirs for supporting farming and other livelihoods. FIGURE 2 (A) Aerial photograph of Jishan Islet of the Pescadores

archipelago, which reveals the typical geology of andesite volcanism with plateau basalt forming flat-topped tablelands. Photograph by PoLin Chi. (B) Satellite photograph of Pratas Atoll with its main island situated on the west of the reef ring. The 350,000-ha area was newly established as Dongsha Marian National Park in 2007. Courtesy of ©CSRSR/CNES 1994.

898

TA I WA N , B I O L O G Y

FLORA

The nearly 4000 recorded vascular plants denote the high floristic diversity of Taiwan, considering its relatively small area. As a consequence of the high annual precipitation and the lofty mountains, forests are the dominant natural vegetation types, and various forest vegetation zones are characterized by different dominant plant groups (Table 1), with temperature gradients changing with elevation. In general, lowland broadleaf forests are

dominated by Moraceae, Euphorbiaceae, and some Lauraceae. Between altitudes of 500 and 2500 m, Lauraceae and Fagaceae are the dominant groups. Above 2500 m, conifer forests are found in the region with various zones being dominated by different evergreen trees; the ecotone may be obvious in high mountain areas. The recorded numbers of families for ferns, gymnosperms, and angiosperms in Taiwan are 38, 8, and 228, respectively, representing more than 50% of world flora in each category. The flora of Taiwan is not only abundant for its high diversity but is also remarkable for its high endemism (Fig. 3A). There are about 4000 species of indigenous vascular plants in Taiwan, and nearly a quarter of the plant species are endemic. Additionally, because of its transitional position biogeographically, Taiwan has become the home of many relic species, such as the Taiwan red cypress (Chamaecyparis formosensis), Taiwan cypress (Chamaecyparis obtusa var. formosana), the Formosan redwood or Taiwan cedar (Taiwania cryptomerioides), the wheelstamen tree (Trochodendron aralioides), the Chinese sweet gum (Liquidambar formosana), Taiwan Pieris (Pieris taiwanensis), the fern Dipteris conjugata (Fig. 3B), and so forth. FAUNA

The known fauna of Taiwan includes more than 32,000 species and many more surely remain to be explored. Recorded species (TaiBNET 2008) include 113 species of mammals, 461 birds, 99 reptiles, 36 amphibians, nearly 3000 freshwater and marine fishes, and about 20,000 insects. Located on the boundary of the tropics and subtropics, Taiwan’s transitional position is also depicted by its separation from mainland China by the Taiwan Strait on the west and its division from its southeast island neighbor Lanyu by Kano’s extension of Wallace’s Line on the east. These factors, along with Taiwan’s diverse landscapes, have created a variety of ecological habitats for wildlife and have nurtured many endemic species, such as the Formosan macaque (Macaca cyclopis), Swinhoe’s pheasant (Lophura swinhoii), the Formosan blue magpie (Urocissa caerulea) (Fig. 4A), the Taipei tree frog (Rhacophorus taipeianus), the Formosan greater horseshoe bat (Rhinolophus formosae) (Fig. 4B), the snake Rhabdophis swinhonis (Fig. 5A), and the highland red-belly swallowtail butterfly (Atrophaneura horishana) (Fig. 5B), as well as many endemic subspecies, such as the Formosan mole (Mogera insularis insularis), the Formosan hare ( Lepus sinensis formosus), the crested serpent eagle (Spilornis cheela hoya) (Fig. 6A), the Formosan land-locked salmon (Oncorhynchus masou formosanus), and the Formosan long-armed scarab (Cheirotonus macleayi formosanus) (Fig. 6B). The high mountain ranges in Taiwan play an important role as a natural barrier for

FIGURE 3 Endemic plants and animals of Taiwan. (A) The Taiwan lily,

Lilium formosanum, a widely distributed endemic plant among four indigenous lilies of Taiwan; (B) Dipteris conjugata, a relic fern that survived through ice ages and has lived on the Earth since the Jurassic Period. Photographs by Kuang-Ying Huang.

TA I WA N , B I O L O G Y

899

FIGURE 5 Endemic animals of Taiwan. (A) Rhabdophis swinhonis, an FIGURE 4 Endemic animals of Taiwan. (A) The Formosan blue magpie,

endemic snake named after Robert Swinhoe, a British consul on Tai-

Urocissa caerulea; (B) the Formosan greater horseshoe bat, Rhinolo-

wan in the nineteenth century who made great contributions to the

phus formosae. Photographs by Kuang-Ying Huang.

natural history of Taiwan; (B) the highland red-belly swallowtail butterfly, Atrophaneura horishana. Photographs by Kuang-Ying Huang.

many species. Species assemblages in the west and the east of the island may differ, as is the case with some freshwater fishes and insects. The endemism is as high as 64% in mammals, 31% in reptiles, 27% in amphibians, and 18% in birds (TaiBNet 2007). The Estimate of insect endemism in Taiwan is more than 60%. The unique location of Taiwan also provides roosting places for migratory birds. The population of the endangered black-faced spoonbill, Platalea minor (Fig. 7), numbers only slightly over a thousand worldwide, and Taiwan has served as an overwinter area for more than two-thirds of this population. The gray-faced buzzard Butastur indicus is a spring and fall migrant in Taiwan. Magnificent views can be seen when large numbers of these buzzards utilize rising air currents to gain lift and then soar into the sky. BIOGEOGRAPHIC AFFINITIES

The flora and fauna of Taiwan originated in the Northern Hemisphere and dispersed to the mid-high altitude of Taiwan during several ice ages, the latest of which was between 7000 and 10,000 years ago. During these periods, a land bridge connected Taiwan and mainland Asia, which

900

TA I WA N , B I O L O G Y

allow organisms to disperse between the two land masses. Organisms arrived in Taiwan via two routes: one from the west, through the Himalayas, and the other from the northern continental region. Evidence is found in many birds, insects, and other organisms that affinity species distributed in a disjointed manner from their center of origin to Taiwan. Moreover, the montane cloud forests, often covered by cloud or mist, nurture magnificent virgin cypress forests in Taiwan (Fig. 8), accounting for the largest known mass of old growth conifers in the subtropical regions. HUMAN HISTORY AND IMPACTS

Located in a biogeographical transitional zone, Taiwan encompasses a remarkable biodiversity in both biota and culture. There are more than 13 indigenous Austronesian tribes, who have been lived on the island for more than 5000 years; additionally, many immigrants from southeastern China have settled on the island in the last 400 years. Linguistic and some other archeological evidence support Taiwan as the center of origin of the Austronesians, who later dispersed between Easter Island and Madagascar.

and fauna. Some native species, such as the Formosan clouded leopard (Neolelis nebulosa) and the Formosan sika deer (Cervus Nippon taiouanus), were reported to be extinct in the wild because of habitat loss in recent centuries. As the twentieth century ends and the twenty-first begins, both government and non-government organizations have made concerted efforts to promote nature conservation. Pertinent laws have been enacted, such as the Cultural Heritage Preservation Act in 1981 and the Wildlife Conservation Act in 1989. To protect the island’s diverse biological resources and ecosystems, approximately 20% of the land is now officially designated as protected areas, including six terrestrial national parks, one marine national park, 19 nature reserves, eight forest reserves, 15 wildlife refuges, and 29 major wildlife habi-

FIGURE 6 Endemic animals of Taiwan. (A) the crested serpent eagle

(Spilornis cheela hoya), a snake predator commonly seen in the lowland area; (B) the Formosan long-armed scarab (Cheirotonus macleayi formosanus), an endemic subspecies of large-sized beetle that is on the protected species list. Photographs by Kuang-Ying Huang.

FIGURE 8 The fog forests of the cypress oldgrowth in the Magao area

of Taiwan. Cloud fall and fog forest are typical phenomenon seen on

The heavy population pressure in Taiwan, which currently houses 23 million people, has resulted in immense forest clearing in the lowlands for several centuries. No virgin forest is left on the plain area, and only a few patches remain on the foothills. Many temperate old growth forests in the montane zone, at 1500–2500 m, faced extensive logging over the last 100 years. Excessive farming in this area during the last several decades has also caused severe erosion. The change of habitats certainly impacts the biota

the mountains in the altitude range between 1500 and ~2500 m. Photograph by Kuang-Ying Huang.

FIGURE 9 A wildlife underpass, part of the corridor system in Yang-

mingshan National Park, reconnects habitats divided by a road to preFIGURE 7 Two-thirds of the world’s population of the endangered

vent roadkills. (A) Guiding net; (B) an endemic Formosan gem-faced

black-faced spoonbill, Platalea minor, overwinter in the estuarine wet-

civet, Paguma larvata taivana, emerging from the underpass (B). Pho-

lands of southwestern Taiwan. Photograph by Chieh-Te Liang.

tographs by Kuang-Ying Huang.

TA I WA N , B I O L O G Y

901

tats. Many long-term biodiversity research projects and hotspot surveys have been initiated. Furthermore, some mitigation projects to reduce human impact on wildlife have been implemented, such as the wildlife underpass system, which has successfully reduced 40% of roadkills in Yangmingshan National Park (Fig. 9). SEE ALSO THE FOLLOWING ARTICLES

Hurricanes and Typhoons / Island Formation / Pacific Region / Taiwan, Geology / Wallace’s Line FURTHER READING

Blundell, D. 2000. Austronesian Taiwan: linguistics, history, ethnology, prehistory. Phoebe A. Hearst Museum of Anthropology and Shung Ye Museum of Formosan Aborigines. Kuo, C. M., ed. 2002. Discovery of green Taiwan. Taiwan Forestry Bureau and BCSD-Taiwan, R. O. C. Lai, Y. M., ed. 2000. Vanishing dancers. The special issue on the conservation of rare and endangered animals in Taiwan. Council of Agriculture and BCSD-Taiwan, R. O. C. Peng, C. I., ed. 1992. The biological resources of Taiwan: a status report. Institute of Botany, Academia Sinica. Su, H. J. 1984. Studies on the climate and vegetation types of the natural forests in Taiwan. (II). Altitudinal vegetation zones in relation to temperature gradient. Quarterly Journal of Chinese Forestry 17: 57–73. Su, H. J. 1994. Species diversity of forest plants in Taiwan, in Biodiversity and terrestrial ecosystems. C. I. Peng and C. H. Chou, eds. Institute of Botany, Academia Sinica. Monograph Series 14: 87–98. Taiwan Biodiversity International Network (TaiBNET). 2008. http://taibnet .sinica.edu.tw/ajaxtree/allkingdom.php.

TAIWAN, GEOLOGY YUE-GAU CHEN

FIGURE 1 Map showing the tectonic environment of Taiwan and its

adjacent region. The Philippine Sea plate is moving northwestward and subducting under the Eurasian plate along the Ryukyu trench. Accordingly, the Luzon arc, generated by the subduction of South China Sea crust under the Philippine Sea plate, collides with the continental margin to create the island of Taiwan.

The “arc” is a volcanic arc (i.e., the Luzon arc) formed by subduction of South China Sea oceanic crust beneath the Philippine Sea plate (Fig. 1). However, the Philippine plate moves northwesterly toward the Eurasian plate and dives under it along the Ryukyu trench, conveying the Luzon arc into the subduction zone once it approaches the trench. The Luzon arc is a huge mass sitting on the top of the plate, so during subduction, it inevitably causes a collision between the arc and the margin of the “continent.” In response to such collisional compression, the crustal materials are thickened to form an accretionary prism, which is the reason for the island of Taiwan’s existence.

National Taiwan University, Taipei

Taiwan, located in the western margin of the Pacific with a center at 24° N, 121° E, is an olive-shaped island with a length and width of ~400 km and 200 km, respectively. The highest mountain peak, Jade Mountain, rises ~4000 m above sea level. Approximately two-thirds of the island’s area is characterized by mountains, hills, and tablelands. Only the remaining one-third is settlement-suitable plain, an area that is mainly distributed along the western coast. MOUNTAINOUS ISLAND AND ARC-CONTINENT COLLISION

Taiwan is dominated by mountain ranges because of its tectonic setting. Based on geologic and seismologic lines of evidence, Taiwan has been defined as one of the modern examples representing an arc–continent collision.

902

TA I WA N , G E O L O G Y

WESTWARD DEVELOPMENTS OF NORTH– SOUTH-DISTRIBUTED MOUNTAIN RANGES

Under such a geologic situation, the rocks exposed on land consist of materials from both the arc and continent sides, which are divided by the longitudinal valley (LV) in eastern Taiwan (Figs. 2 and 3). To the east of the LV, a parallel mountain range is composed of a rock association related to the arc system (i.e., andesitic volcanics, limestones, and deep-sea sedimentary turbidite sequences). To the west of the LV, a suite of mountain ranges are distributed westward all the way to the western coastal plain (Fig. 3), which has been regarded as the deformed continental margin. Additionally, the metamorphic grade of the rocks decreases from the LV westward. The eastern flank of the Central Range, located immediately next to the LV in the west,

FIGURE 3 A cross-island section showing west-vergent development

of the fold-and-thrust belt of Taiwan. The easternmost mountain range (i.e., the Coastal range) is situated as a back-stopper to westerly plow the continental margin and to incrementally form mountains by developing the major thrust faults toward the western coast.

of the Hsuehshan Range is a hilly zone with a width of ~30 km, called the Western Foothills, which are mainly composed of non-metamorphic sedimentary rocks. These have a similar origin as the Hsuehshan Range but have not undergone deep burial. In the western margin of these foothills and further west in the coastal plain, terraces and tablelands are widely distributed and are mainly composed of fluvial sands and cobbles, indicating the early stage of the mountain front depositional systems. However, they soon became the early stage of mountain building because the tips of the deformation belt migrated westward into the plain area. The entire mountain-building history and topographic geometry have been studied for several decades, and a model of thin-skinned critical wedge taper first proposed in the early 1980s can still explain most of what we observe. HAZARDS UNDER HIGH PRECIPITATION, HIGH MOUNTAINS, AND HIGH SEISMIC OCCURRENCE RATES

FIGURE 2 Simplified geological map of Taiwan showing north-north-

east–south-southwest trending geological entities in response to the stress caused by the tectonic collision.

is mainly constituted of schists, gneisses, and marbles, which are believed to be the preexisting internal materials of the continental crust. The western flank of the Central Range and the Hsuehshan Range, which is connected with the Central Range in the west, are dominated by low-grade metamorphic rocks including slates and meta-sandstones, which were originally sedimentary rocks that were deposited on the continental shelf prior to the tectonic collision mentioned above, but that were then transformed and deformed as a result of their deep burial during the collision. On the west

Because the neighboring subduction and mountain building associated with arc-continent collision are still occurring, Taiwan and its surrounding regions are seismically active. Seismic hazards are certainly one of the threats to people living on this island. Despite the numerous earthquakes generated by subduction processes offshore, onshore large earthquakes are in fact the major seismic hazard source. This is because when the epicenter is located on land, the shaking damage will be obvious, and surface ruptures may further extend the degree of the hazards. Theoretically, land earthquakes are caused by active fault systems, which are major structural elements of the accretionary prism and are widely distributed on land in Taiwan. In the past century, five large earthquakes (1906, 1935, 1946, 1951, and 1999) occurred with surface ruptures along the surface trace of active faults. Each time, the earthquake event seriously damaged vital systems where the epicenter and rupture were located.

TA I WA N , G E O L O G Y

903

On the other hand, the genetic mountain-building process produces a generally high rate of uplift in most of the Taiwan region, which induces an increase of the mass of debris on the mountain slopes. Unfortunately, for six months per year (during typhoon season), Taiwan undergoes tropical storms because of its geographic location. The abrupt high-mountain relief results in extremely high precipitation occurring within a day or less. Hence, landslides, floods, and debris often strike the colonial or other urban areas. Such hazards are significantly enhanced in the event that a typhoon quickly follows a large earthquake. Being better prepared to mitigate natural hazards requires having better knowledge about all these geologic processes. Such mitigation is urgently needed because Taiwan is one of the few places in world where the rates of geologic processes are extraordinarily high.

aboriginal people for at least 30,000 years and by Europeans for just over 200 years. LOCATION

Tasmania is a continental island of about 68,400 km2, lying between latitudes 40 and 43° S, about 200 km off the coast of southeastern Australia, from which it is separated by Bass Strait. Several substantial offshore islands are associated with the mainland of Tasmania, notably Bruny and Maria Islands off the east coast, King Island halfway between Tasmania’s northwest tip and the Australian mainland, and the Furneaux Group (Cape Barren, Flinders, etc.), which forms a chain linking the northeast tip with Wilson’s Promontory, the southernmost point of the Australian mainland. Tasmania is a state of the Commonwealth of Australia, with a human population of about 500,000.

SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Landslides / Plate Tectonics / Taiwan, Biology / Volcanic Islands FURTHER READING

Angelier, J. 1986. Geodynamics of the Eurasia-Philippine sea plate boundary: preface. Tectonophysics 125, ix–x. Davis, D., J. Suppe, and F. A. Dahlen. 1983. Mechanics of fold-andthrust belts and accretionary wedges. Journal of Geophysical Research 88: 1153–1172. Ho, C. S. 1986. A synthesis of the geologic evolution of Taiwan. Tectonophysics 125: 1–16. Suppe, J. 1984. Kinematics of arc-continent collision, flipping of subduction, and back-arc spreading near Taiwan. Memoir of the Geological Society of China 6: 21–33. Teng, L. S. 1986. Late Cenozoic arc-continent collision in Taiwan. Tectonophysics 183(1990): 57–76. Tsai, Y. B. 1986. Seismotectonics of Taiwan. Tectonophysics 125: 17–38.

TARSIERS SEE LEMURS AND TARSIERS

TASMANIA ALASTAIR M. M. RICHARDSON

904

GEOLOGICAL AND POSTGLACIAL HISTORY

As part of the Australian tectonic plate, Tasmania’s history can be traced back to the supercontinent of Gondwana, and it shared Australia’s long isolation following the separation from Antarctica (80 million years ago) that lasted until the collision with southeast Asian plates about 14 million years ago. As the most southern part of the Australian land mass, Tasmania has always been farthest from the source of animals and plants colonizing from the north. Its southern location and relatively high topography meant that Tasmania felt the effects of the Pleistocene glaciations more strongly than anywhere else in Australia. At the glacial maximum, almost half the land mass was covered by ice, and periglacial influences extended to sea level in some places. The proximity of the continental shelf edge to the east and west coast allowed only a small increase in land area as sea levels fell, but Bass Strait disappeared and was re-flooded several times as the glaciers advanced and retreated. During the glacial maxima, the Bassian plain was probably dry, cold, and treeless, providing quite a strong filter to potential colonizers. After the last glacial maximum, about 18,000 years ago, sea levels rose for the last time, and by about 10,000 years ago Tasmania was separated from the mainland.

University of Tasmania, Hobart, Australia

GEOLOGY AND TOPOGRAPHY

Tasmania is a medium-sized continental island lying southeast of the Australian mainland. It is topographically diverse, largely forested, and supports a number of relictual and endemic species. It has been occupied by

In many ways, Tasmania is an island of two halves: the cool, wet west and the drier east. This pattern results from the interaction between the topography and the westerly weather patterns of the “roaring forties” latitudes. Western Tasmania is formed of old, hard, folded, and sometimes

TA S M A N I A

mineralized rocks, whereas the east is mostly composed of softer rocks of Permian and Triassic age, into which were injected large sills of dolerite during the Jurassic. Extensive faulting, accompanied by weathering of the softer rocks to expose the dolerite surfaces, has produced in the east a landscape of plateaus and grabens, whereas the folded rocks of the west give a much more rugged topography. There are some exceptions to this general pattern, notably the granites of the northeast, but the east–west divide is striking. The mountainous nature of the west, with its high rainfall and low fertility, has largely protected the region from European settlement, with the result that it is now mostly in the Tasmanian Wilderness World Heritage Area. CLIMATE

Westerly winds blowing in from the southern ocean rise over the western mountains depositing over 2000 mm of rain annually in some places, but leaving a rain shadow to the east, where annual rainfall may be below 500 mm per year. Annual temperatures also increase from west to east, with mean January maxima and July minima of around 23 °C and 5 °C, respectively, at sea level. Although Tasmania’s climate can generally be described as cool maritime, its location to the south of a large dry continent, which extends into low latitudes, allows hot, dry air to affect the island when northerly winds blow. Tasmania’s weather is largely the product of a series of high and low pressure systems that move across the island from west to east. As an area of high pressure moves to the east of the island, winds become northerly or northwesterly; in summer, they bring hot, dry air, and temperatures can exceed 35 °C but usually for only a short period. This pattern may be sharply terminated by a cold front, after which winds turn southerly or southwesterly, bearing squalls of rain, or snow at higher altitudes. Above 300 m, frosts may be experienced in any month, but snow does not lie for long periods, even at the highest elevations. This typical pattern has been modified in recent years by the increasing strength and frequency of El Niño events. During strong El Niños, anticyclones pass south of Tasmania, breaking the westerly pattern and increasing rainfall in the east, as warm, moist easterly or northeasterly winds blow in from the Tasman Sea. VEGETATION AND SOILS

The east–west divide in topography and climate is reflected in the vegetation and soils. The climax vegetation is forest over most of the island below 1000 m; in the west the climax is rain forest, dominated by south-

ern beech, Nothofagus cunninghamii, in areas where the annual rainfall exceeds 50 mm per month. In the east, the forests are dominated by various species of Eucalyptus: very tall closed forest in the wetter areas, grading to lower open savanna forest in the driest areas. These forest climaxes are often diverted by edaphic conditions or by fire. Large areas of western Tasmania are covered by a tussock sedgeland growing on acidic peat soils, often dominated by buttongrass, Gymnoschoenus sphaerocephalus, and maintained by a fire frequency of 20–30 years. At the eastern boundary of the range of southern beech, the forest is often in the form of “mixed forest,” (i.e., an overstory of tall, old eucalyptus and an understory of rain forest species, including southern beech). In the absence of fire, these forests become rain forest when the eucalyptus trees, unable to regenerate without fire, reach the end of their lives at 350 or more years; fire in this forest type is catastrophic, although very infrequent, and it results in large areas of even-aged regeneration of eucalyptus from seed shed from standing trees after the fire. Non-forest vegetation is found at altitudes above 1000 m, although endemic pines may form patchy woodlands; the alpine vegetation is highly sensitive to fire. Deciduous beech, Nothofagus gunni, the only native deciduous tree, is found in subalpine regions. Around the coast, wind and salt deposition allow the development of a florally rich coastal heath of shrubs and herbs. Poa grasslands are also found at low altitudes in dry or frost-hollow situations. FAUNA

The Tasmanian fauna shows the characteristics that might be expected of a continental island in its particular location: The fauna is depauperate compared to that of the adjacent mainland, but it includes several groups with relictual members and species that have, at least until recently, found a refuge there from introduced species on the Australian mainland. The list of breeding landbirds, for example, includes about 104 species, compared to 176 in equivalent habitats in Victoria, the adjacent mainland state. This number fits quite well the species-area curve for island avifaunas, and the number of endemic species (12) is in the same range as that of other large continental islands close to the source. The presence of one endemic flightless bird, the Tasmanian native hen (Gallinula mortierii) (Fig. 1A), reflects the island status. Tasmania is perhaps best known for two members of its marsupial fauna, although one of these (the thylacine, Thylacinus cynocephalus) is extinct, and the other (the Tasmanian devil, Sarcophilus harrisii) (Fig. 1B) is rapidly declining as a result of a highly unusual communicable

TA S M A N I A

905

facial tumor. The condition was first observed in 1996, and the tumors are spread directly, as cancerous cells, when the animals bite each other—for example during competition when feeding at carcasses. The island is also a refuge for two other large dasyurids, the eastern quoll (Dasyurus viverrinus) and the spotted-tailed quoll (D. maculatus), and supports good populations of the bettong (Bettongia gaimardi), pademelon (Thylogale billardierii), long-nosed potoroo (Potorous tridactylus), and barred bandicoot (Perameles gunni), which are extinct or threatened on the Australian mainland. Other common marsupials include the brush-tail possum (Trichosurus vulpecula) and Bennett’s wallaby (Macropus rufogriseus). Wombats (Vombatus ursinus) are widely distributed and are common in sedgelands, heaths, and grasslands. Two monotremes, the echidna (Tachyglossus aculeatus) and platypus (Ornithorhynchus anatinus), are widespread and common, but there is concern about an outbreak of a necrotic fungal disease, first observed in 1982, in the platypus populations in some northern rivers. The reptile fauna is modest by Australian standards, comprising 17 skinks, one agamid, and three snakes; however, several of the skinks are short-range endemics, with highly restricted populations on isolated mountain ranges and, in the case of one (the pedra branca skink, Niveoscincus palfreymani), on a small off-shore island. The frog fauna (11 spp.) is also limited compared with the rest of Australia but includes three endemics. Chytrid fungus, which attacks cartilage and is implicated in worldwide frog declines, has been identified in Tasmania, and there is evidence that the populations of several frog species have decreased. The native fish fauna (25 native species) is dominated by members of the Galaxiidae (15 spp.), and once again levels of endemism are high (12 spp.). The galaxiids are all small species and generally suffer from predation by the introduced salmonids, principally brown and rainbow trout. It is difficult to do justice to Tasmania’s invertebrate fauna, but the non-marine crustaceans serve as an example of the relictual nature of the fauna. Most widely known, FIGURE 1 Endemic animals from Tasmania. (A) The flightless Tas-

manian native hen (Gallinula mortieri) (photograph by Dave Watts). (B) The largest extant marsupial carnivore, the Tasmanian devil (Sarcophilus harrisii) (photograph by Menna Jones); (C) Mountain shrimps (Anaspides tasmaniae), syncarid crustaceans (photograph by Niall Doran); the largest specimen is about 60 mm long. (D) The giant freshwater crayfish, or tayatea (Astacopsis gouldi), the world’s largest nonmarine invertebrate; this specimen is about 40-cm long and weighs about 2 kg (photograph by Niall Doran).

906

TA S M A N I A

at least to specialists, are the Syncarida, or mountain shrimps, primitive malacostracans with a body plan that suggests the morphology of the earliest members of that group. Anaspides tasmaniae (Fig. 1C), which can exceed 50 mm in length, is common in mountain streams and lakes where trout are absent or where there are good refuges. Paranaspides lacustris is another large syncarid, an inhabitant of weed beds, and is convergent with decapod shrimps. The lakes and streams of Tasmania also support diverse faunas of amphipods and phreatoicoid isopods. On land, the leaf litter in the wetter forest is often dominated by very large numbers of talitrid amphipods, sometimes exceeding 5000 animals per square meter. At least 15 species of these relatives of the familiar coastal beachfleas are found on the island. The most spectacular member of the crustacean fauna is the giant freshwater lobster, Astacopsis gouldi (Fig. 1D), known to aboriginal people as tayatea. It can exceed 1 m and a weight of 4 kg, making it the world’s largest freshwater invertebrate. Very large animals are now rare as a result of fishing pressure, and although fishing is now illegal, poaching and habitat deterioration from land clearance remain problems for this species. This giant is one of over 30 species of freshwater crayfish found in Tasmania; these include species living in the acid peaty sedgelands, and others that have become highly terrestrial, living in burrows in clay soils of the rain forests, where their water supply is collected from the surface runoff and stored in underground chambers. The crustacean fauna is not the only invertebrate group that shows special features. Many insect orders include

primitive endemics, such as the Gondwanan dragonfly Archipetalia auriculata. Other endemic arthropods include the enigmatic centipede Craterostigmus tasmanianus and the Tasmanian cave spider Hickmania troglodytes, one of a very small number of hypochilid spiders found in the Southern Hemisphere that represent the basal form of the araneomorph spiders. ENDEMISM AND GIGANTISM

Endemism in the Tasmanian fauna can be related directly to the vagility of each group (Table 1); almost all the groups with very low powers of dispersal show 80% or more endemism. Gigantism, either globally or locally to Australia, is a feature of several invertebrate groups: for example, syncarids, parastacid crayfish, collembolans, and stoneflies. Among the birds, some species are larger than their mainland counterparts (e.g., the masked owl Tyto novaehollandiae, the wedge-tailed eagle Aquila audax, the superb blue wren Malurus cyanaeus), whereas others are smaller (e.g., the yellow-tailed black cockatoo Calyptorhynchus funereus, the tawny frogmouth Podargus strigoides, the eastern spinebill, Acanthorhynchus tenuirostris); several show colors darker than those of equivalent mainland species. ABORIGINAL COLONIZATION

Aboriginal people have lived in Tasmania for at least 35,000 years. The earliest 14C dates for human occupation, obtained from fire hearths, are around 33,850 years old, but estimated sea levels suggest that the Bass Strait was dry and would have offered a passage to Tasmania as

TABLE 1

Examples of Various Animal Groups Occurring in Tasmania: The Relationship between Vagility and Level of Endemism Vagility

Group

Species in Tasmania

Endemic species

Endemism (%)

High to medium

Land mammals Landbirds Reptiles Frogs Water beetles (Dytiscidae) Dragonflies Water bugs

26 104 21 11 26 28 37

3 14 7 3 1 6 4

12 13 33 27 4 21 11

Low

Caddisflies Stoneflies Terrestrial snails

157 46 49

116 40 33

74 87 67

Very low

Freshwater crayfish Terrestrial amphipods Mountain shrimps Torrent midges Trechine beetles

33 19 8 6 63

31 18 7 6 62

94 95 88 100 98

TA S M A N I A

907

long as 130,000 years ago. Pollen records from sediment cores suggest an increase in fire frequency at about that time, which may have been the result of aboriginal burning. Although debate continues about the date of aboriginal colonization, cave deposits provide good evidence that the island was occupied during the last glacial maximum. At the time of European colonization, the aboriginal population was probably between 4000 and 6000. It is likely that the fire regime in Tasmania changed with the establishment of aboriginal people, and fires became more frequent throughout the island. After Europeans arrived, the aboriginal population declined sharply as a result of disease, displacement, and direct conflict with the colonists. Attempts to establish aboriginal settlements on Flinders Island and later at Oyster Cove in the southeast did nothing to prevent the decline, and the woman considered to be the last fullblooded aboriginal, Truganinni, died in 1876. However, the aboriginal community in Tasmania remains strong and has recently been involved in a vigorous worldwide campaign to repatriate aboriginal skeletal material held in museums. EUROPEAN COLONIZATION

There is no evidence to suggest that coastal vessels from Southeast Asia, or Arab traders, penetrated as far south as Tasmania, and the first non-aboriginal group to discover the island was almost certainly Abel Tasman and the crew of the ships Heemskerck and Zeehaen in 1642. French and English explorers, including Du Fresne, D’Entrecasteaux, and Bligh, visited the island, but the first European settlement was not established until 1803. European sealers, however, had already discovered the seal colonies in Bass Strait, and they established informal settlements on the islands, taking aboriginal women as partners. After the sealers, the main economic use of the island was as a penal colony, and prisons were set up in various locations, with the largest at Port Arthur. Free settlers began a small pastoral industry, but economic growth began in earnest toward the end of the nineteenth century with the discovery of metal ores, especially copper, in the western mountains. Extraction and smelting of zinc, aluminum, and other metals continues today, and these metals still lead the list of the state’s exports, but forestry, both for lumber and paper production, follows closely. Charles Darwin spent ten days in Hobart during the voyage of the HMS Beagle in 1836; he climbed Mt. Wellington and collected fossils near the town but made few other observations. Scientific observations and collections flourished in the nineteenth century, and the Royal Society of Tasmania was the first of its type in Australia when

908

TA S M A N I A

TABLE 2

Summary of Species Listed under the Tasmanian Threatened Species Protection Act 1995 Vertebrates

Extinct Endangered Vulnerable Rare Total

5 35 19 13 72a

Invertebrates

4 17 16 89 126b

note: A further five species are listed under the Commonwealth of Australia’s Environment Protection and Biodiversity Conservation Act 1999. a Includes 31 species from oceanic or Antarctic waters, or Macquarie Island. b Includes 43 species of hydrobiid freshwater snails with highly localized distributions.

it was established in 1843. The University of Tasmania was founded 1890, making it the fourth oldest in Australia. CONSERVATION ISSUES

Over 40% of the Tasmanian land mass is protected in national parks and other reserves, including the Tasmanian Wilderness World Heritage Area established in 1982, but the eastern half of the island is underrepresented in reserves, and land clearance or conversion to non-native vegetation continues. Just three vertebrates (the thylacine, the Tasmanian emu Dromaius novaehollandiae diemenensis, and the King Island emu D. minor) are known to have become extinct since European colonization; four invertebrates (two caddisflies, a beetle, and a spider) are listed as extinct under the Tasmanian Threatened Species Protection Act 1995, but the ranges of many more have been severely fragmented (Table 2). A number of European plants, fish, birds, and mammals were deliberately introduced following European settlement, although Tasmania did not suffer as badly in this regard as New Zealand. More recently, several species have invaded the island with human help, all of which have potentially devastating effects. European carp, Cyprinus carpio, have appeared in two lakes, and the arrival of the European red fox, Vulpes vulpes, may result in the elimination of the mid-weight range species of marsupials, for which Tasmania has been such a valuable refuge. Vigorous efforts are being made to eradicate both of these invaders. Several birds have colonized Tasmania naturally in the last 50 years: The kelp gull, Larus dominicanus, and the cattle egret, Ardea ibis, are two examples. Other Australian native birds have become established, probably as a result of deliberate introductions or aviary escapes: Examples include the kookaburra (Dacelo novaeguineae), the superb lyrebird (Menura novaehollandiae), the galah (Cacatua rosiecapilla), the little and long-billed corellas (C. sanguinea and C. tenuirostris), and the rainbow lorikeet (Trichoglossus haemotodus).

SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Gigantism / Prisons and Penal Settlements / Vegetation / Voyage of the Beagle FURTHER READING

Doran, N., A. M. M. Richardson, and R. Swain. 2001. The reproductive behaviour of Hickmania troglodytes, the Tasmanian cave spider (Araneae, Austrochilidae). Journal of Zoology 253: 405–418. Hamr, P. 1990. Rare and endangered: Tasmanian giant freshwater lobster. Australian Natural History 23: 362. Parks and Wildlife Service Tasmania: Nature of Tasmania website. http:// www.parks.tas.gov.au/nature.html Reid, J. B., R. S. Hill, M. J. Brown, and M. J. Hovenden. 1999. Vegetation of Tasmania. Canberra: Australian Biological Resources Study. Smith, S. J., and M. R. Banks, eds. 1993. Tasmanian wilderness—world heritage values. Hobart: Royal Society of Tasmania. Williams, W. D. 1974. Biogeography and ecology in Tasmania. The Hague: Dr. W. Junk.

TATOOSH EGBERT GILES LEIGH, JR. Smithsonian Tropical Research Institute, Balboa, Panama

ROBERT T. PAINE University of Washington, Seattle

Tatoosh (Fig. 1) is a set of islets covering 17 to 18 ha at 48°24´ N, 124°44´ W, 0.6 km off Cape Flattery, the northwest tip of the Olympic Peninsula. A lighthouse on the largest islet marks the southern lip of the strait of Juan de Fuca, which leads from the Pacific Ocean to Seattle and Vancouver. These islets’ rocky shores support a luxuriant community of intertidal organisms. Research there by Robert Paine and his students and colleagues has revealed many of the factors that govern what lives where on rocky shores, and has shed light on other, very different ecosystems.

its many long processions of great swells. These “weather shores” support a particularly lush intertidal community, whose barnacle zone can extend over 5 m above mean lower low water. In a relatively sheltered area between two islets, the barnacle zone extends less than half as high. These differences in wave exposure govern the distribution and abundance of many organisms. Tatoosh has gloomy, rainy winters and drier summers. Hard freezes are relatively rare. The main intertidal herbivores and predators are slow-moving invertebrates— sea urchins, chitons, snails, starfish—which crawl over the rock. Tatoosh also has a large gull colony. Like the recently returned sea otters, gulls eat a wide variety of invertebrates. The ecological roles of many species of animals and plants have been assessed by using cages, strips of poisonous paint, or frequent manual removal to exclude consumers or competing plants, and quantifying the impacts of the exclusion. An intertidal zone is alternately submerged by the sea and exposed to the air. This zone separates the marine world from the land. At Tatoosh, nearly vertical walls rise from lush stands of kelps exposed only at the lowest tides. Above is a barer zone with brightly colored crusts, lowstatured invertebrates, and scattered algae, followed by dense mussel beds; sometimes a zone of short, springy algae; a barnacle zone; and, finally, lichen-encrusted rock (Fig. 2). More gently sloping surfaces reveal subtle variations on the same theme: a few meters’ walk upward from the low tide level leads through lush stands of several kinds of kelp to a broad band of mussels, followed by a barnacle zone, sometimes separated by a zone of springy algae, Mazzaella. These “zones,” some dominated by plants, others

THE SITE

Native Americans almost certainly had occupied Tatoosh for millennia. They were progressively displaced after its lighthouse was built in 1857, and they rarely visited after the early 1900s. The U.S. Coast Guard automated the lighthouse and abandoned Tatoosh in 1976, and ownership of Tatoosh has been restored to the Makah Nation. Its intertidal zone has been relatively free of human disturbance for a century. Tatoosh is at a strait’s mouth. The continental margin (the 100-fathom line) is only 2.5 km distant, so wave action is heavy. Its different shores experience very different conditions. The western shores face the ocean, with

FIGURE 1 An aerial view of Tatoosh, facing roughly south, at or near

low tide. The lighthouse is visible on the largest of this set of islets. The western side of Cape Flattery appears at the upper left. Photograph by Alan Trimble.

TAT O O S H

909

FIGURE 2 Zonation along an east-facing wall of a surge channel at

Tatoosh. From the top, the greenish tinge of the green alga Prasiola, then a broad band of the barnacle Balanus glandula, separated from the bed of mussels, Mytilus californianus, by a thin band of the red alga Endocladia. Below the mussels is a mixture of the barnacle Semibalanus cariosus and the gooseneck barnacle Pollicipes. The band closest to the water is occupied by a mixture of sea anemones, Anthopleura; sponges; kelps; and the occasional bright orange hydrocoral, Allopora. Photograph by Robert T. Paine.

by sessile animals, differ more fundamentally than the zones of montane forest, cloud forest, elfin forest, and paramo separating lowland rain forest from the bare rock or perpetual snow of a high tropical mountain. A moderate tidal range (maximum of 3.8 m) and an abundance of waves and swells make the zonation at Tatoosh particularly clear. The factors maintaining these striking patterns, and the variation in distribution and abundance of this community’s species, have been our primary focus. Discussed below are four themes of research at Tatoosh, some of which were initiated on nearby, ecologically similar, mainland weather shores. INTERTIDAL KELPS AND OTHER SEAWEEDS

A visitor cannot help being impressed by the variety of form among Tatoosh’s hundreds of species of seaweeds: sheets, tubes, straps on stalks, miniature “palms,” and crusts that are more limestone than living plant. High in the intertidal are extensive carpets of the springy, 2–3-cm high Mazzaella cornucopiae. Paine installed marker screws in various places for recording how this carpet’s sharp upper limit varied through time. Surprisingly, this limit has receded 30 to 40 cm since 1978. Mazzaella supports about 7 m2 of fronds per square meter of rock. Lower down, in gaps within the most wave-beaten mussel beds, grow sea palms, Postelsia palmaeformis, which support a thatch of fronds atop a stout, hollow, 50-cm stalk. Sea palms need waves to sweep away adjacent, encroaching mussels and protect the palms from grazers.

910

TAT O O S H

Near the low-water mark is the laminarian zone, populated by large kelps and an abundance of grazers. The canopy of these kelps shelters a diverse flora of understory seaweeds and many species of algal crust. On most shores Laminaria, with an upright stalk 45–50 cm tall carrying a long, smooth broad frond, dominates the lowermost intertidal. Where they are present, sea urchins, Strongylocentrotus purpuratus, eliminate Laminaria: it returns if the sea urchins are removed. These kelps grow where waves restrict urchin activity or natural enemies keep urchins out. On gentle, less exposed shores, another perennial kelp, Hedophyllum, flourishes above the Laminaria. Hedophyllum’s broad, heavy fronds grow from a flat holdfast. Hedophyllum grows where chitons, Katharina, eat their competitors. The faster-growing annual kelp Alaria marginata occupies plots cleared of both Hedophyllum and chitons, but Hedophyllum recolonizes plots with chitons. Settings sheltered enough to allow sea urchins to aggregate, however, lack Hedophyllum. Instead, these “urchin barrens” favor coralline crusts, which are hard to eat because of their limestone content. Experiments show that the fastest-growing coralline algae are most susceptible to grazers. This trade-off between fast growth and effective defense helps maintain the diversity of corallines, and probably of many other groups. The influence of herbivores on what plants grow where is especially clear on these lower shores, where herbivores so clearly influence zonation, species composition, and productivity. WHO EATS WHOM AND WHY IT MATTERS

Another theme is who eats how much of whom, where, and how these eating habits organize the intertidal community. To answer this question one must observe who eats whom, and remove consumers to see how they affect the species they eat. In 1963, Paine removed the musseleating starfish, Pisaster ochraceus from a plot on a weather coast 10 km south of Tatoosh. Removing this starfish allowed mussels to spread lower in the intertidal, overgrowing the scattered kelps, coralline crusts, sponges, and the like that grew where starfish had kept the mussels out. This experiment, since repeated twice at Tatoosh, spotlighted Pisaster as a “keystone species” whose impact on the community was far greater than its numbers suggested. Similar studies by Paul Dayton showed how limpets affected the algae they ate, how some limpets influence barnacle species composition by “bulldozing” young barnacles of some species, and how large starfish, Pycnopodia, that live from the low tide downward protect

Laminaria by eating, and chasing off, the sea urchins that would destroy them. Large green sea anemones, which eat many of the sea urchins fleeing from Pycnopodia, also aggregate where they can catch mussels dislodged by foraging Pisaster. In 1974 Kenneth Sebens learned that the normally sessile anemones can move—slowly—to where they can catch more food. Timothy Wootton, Jennifer Ruesink, and Paine have used a variety of experimental methods to assess who eats how much of whom, who replaces whom, and how fast they do it. This knowledge is essential for understanding how the ways different species affect each other organize the community to which they belong. MUSSEL BEDS

Mussels, Mytilus californianus, form a conspicuous zone on gently sloping, rather wave-beaten shores. In 1978 Thomas Suchanek showed that a thick, old mussel bed may shelter over 300 macroscopic species. Mussels grow faster lower down, but starfish, Pisaster ochraceus, limit how far down they spread. Where waves pound hardest, limiting the activity of these starfish, or where starfish are regularly removed, the mussels spread downward. When mussels spread among Hedophyllum or a shrubby, manyfronded kelp of more exposed shores, Lessoniopsis, the mussels replace the kelps because waves shred the fronds of the kelps against the many sharp edges of the mussel bed. Moreover, long-term data on the upper limit of the mussel bed has shown how it has tracked the 18.5-year cycle in gravitational influence on tides. Waves unravel mussel beds that have grown too thick and multilayered, clearing gaps of several square meters or more, exposing bare rock. How often do gaps open? How long does it take gaps of given size to close again? In prime mussel territory, waves strip mussels from a given point every 7–10 years on the average. Gaps of various ages provide a dynamic patchwork of variety within the mussel bed. The centers of big gaps, beyond the reach of grazers sheltering under the mussels, are colonized by fast-growing kelps—Postelsia palmaeformis on the most wave-beaten angles, Alaria nana elsewhere. Mussel beds, with their rich array of associated species, are the biological centerpiece of these exposed shores. THE IMPORTANCE OF WAVES

A fourth research theme is how exposure to waves influences intertidal life. The intertidal zone is wider and more luxuriant on more exposed shores, where sea spray reaches higher. Starfish, sea urchins, limpets, and snails cannot forage where waves are breaking too vigorously,

so they are less able to control their food species. As we have seen, a shore’s zonation reflects its degree of wave exposure. Where their surrounding water moves faster, plants and filter-feeding animals extract nutrients and food from it more easily, and the wastes they shed disperse more quickly. Kelps of wave-beaten shores, where wave damage is the primary threat, have smooth, sleek fronds. Sheltered waters house other species whose crinkled or ruffled fronds are designed to excite turbulence in the surrounding water, helping nutrients to penetrate and wastes to leave the boundary layer of still water around the frond. On sheltered coasts, Hedophyllum is cabbage-like, with short, wrinkled upstanding fronds quite unlike the broad smooth Hedophyllum fronds of more exposed shores. Starting in 1979, Mark Denny measured the force of breaking waves on Tatoosh. He calculated the force needed to dislodge various sea urchins, limpets, and snails when they were foraging, and showed how the threat of breakage or dislodgement by waves limited the size of intertidal organisms, especially those of weather shores. On the most exposed angles waves allow sea palms, Postelsia, to support over 15 m2 fronds/m2 rock by stirring their fronds enough to bring light to each. Sea palms can produce as much dry matter in five months as wheat carefully cultivated for maximum production in a mock spaceship. In mature rain forests, tall trees shade their neighbors: canopy leaves are flooded with light, while forest floor herbs receive barely enough light to survive. Unequal sharing of light means that a hectare of mature rainforest carries less than ten hectares of leaves. By limiting maximum height, thereby enforcing the Roman maxim debellare superbos, parcere subjectis (beat down the proud, spare the meek), waves allow a sufficiently even distribution of light among fronds to enable Postelsia’s extraordinary productivity. LESSONS FROM TATOOSH

What lessons emerge from this story? First, the research of a group of biologists attracted to a site of singular beauty and fascination, each working on a project of his or her own design, provides a coherent picture of how that site’s community functions. “Unity of place” focuses the research, enabling each project to provide context for the others, raise new questions, and provide a basis for answering them. Second, we cannot infer process from pattern. Organisms do not always live where they grow best, because predators or competitors may keep them out.

TAT O O S H

911

Third, certain keystone species exert an impact utterly disproportionate to their numbers. For example, by restricting the downward spread of mussels, the starfish Pisaster makes room for a diversity of kelps, sponges, coralline algae, and their consumers. Fourth, we learn how the degree of wave action modulates the processes that organize the community. Kelps that require ceaseless waves to stir their many fronds are restricted to the most wave-beaten shores; sea urchins are more active, and more devastating, where they are safer from waves. Finally, Tatoosh represents an extreme of interdependence. Any weather coast collects energy from swells stirred anywhere upwind in the vast expanses of ocean they face. At Tatoosh, these waves enhance intertidal productivity. The ocean near Tatoosh is also a rich source of food. Its abundant plankton allows filter feeders to compete on even terms with primary producers. This ocean feeds planktonic larvae; how dispersal of these larvae enhances the genetic variation of their species or restricts their capacity for local adaptation is an open question. This ocean’s fish feeds the gulls and puffins, cormorants and auklets, that abound on Tatoosh. Excrement of these birds fertilizes plankton in the surrounding water. The gulls supported by this ocean eat the sea urchins they can reach, helping to protect kelp. Compared to most rain forests, Tatoosh is an emporium that depends utterly on the wealth it collects from a remarkably wide area. SEE ALSO THE FOLLOWING ARTICLES

Barr0 Colorado / Succession / Tides FURTHER READING

Dayton, P. K. 1971. Competition, disturbance and community organization: the provision and subsequent utilization of space in a rocky intertidal community. Ecological Monographs 41: 351–389. Denny, M. W. 1987. Life in the maelstrom: the biomechanics of waveswept rocky shores. Trends in Ecology and Evolution 2: 61–66. Leigh, E. G., Jr., R. T. Paine, J. F. Quinn, and T. H. Suchanek. 1987. Wave energy and intertidal productivity. Proceedings of the National Academy of Sciences of the United States of America 84: 1314–1318. Paine, R. T. 1974. Intertidal community structure: experimental studies on the relationship between a dominant competitor and its principal predator. Oecologia 15: 93–120. Paine, R. T. 2002. Trophic control of production in a rocky intertidal community. Science 296: 736–739. Paine, R. T., and S. A. Levin. 1981. Intertidal landscapes: disturbance and the dynamics of pattern. Ecological Monographs 51: 145–178. Rosenfeld, A. W., and R. T. Paine. 2002. The intertidal wilderness. Berkeley: University of California Press. Wootton, J. T. 1993. Size-dependent competition: effects on the dynamics vs. the end point of mussel bed succession. Ecology 74: 195–206.

912

TA X O N C YC L E

TAXON CYCLE MANDY L. HEDDLE Center for Environmental Education, Ahmedabad, India

The taxon cycle was originally conceived to explain species distributions on islands where, within a particular species complex, some species are widespread, abundant, and found in coastal habitats, whereas others are fragmented, rare, and occupy montane habitats. The mechanism for the cycle is driven by new colonists (early-stage species) that establish populations on coastal, marginal habitats, outcompeting established populations, which experience shifts in ecological adaptations as they are forced toward more interior habitats. As more colonists arrive, older taxa (late-stage species) become further specialized for particular ecological conditions and occupy increasingly interior habitats, such that their populations become fragmented and more susceptible to extinction. These species finally go extinct, completing the cycle (Fig. 1). The taxon cycle is controversial as a theory to explain species distribution. DEFINING THE TAXON CYCLE

The taxon cycle was first defined by E. O. Wilson in 1959 to describe island populations of Melanesian ants. Although the concept of species proceeding through predictive stages, similar to the life cycle of an individual, had been proposed by several authors prior to Wilson, the taxon cycle specifically describes the characteristics of these stages by their geographical distribution and ecological adaptations. REFINEMENTS OF THE TAXON CYCLE

Several authors have proposed modifications and refinements to the taxon cycle. In 1972, Ricklefs and Cox explored the concept of competition as it pertains to the taxon cycle proposing that competitive ability will change at different stages of the cycle. For example, habitat shifts could result in changes in coevolutionary relationships with predators, and as species become marginalized and rare, they escape density-dependent predation. Erwin (1985) proposed an alternate, non-cyclical model of progressive specialization that he termed a “taxon pulse,” in which taxa make adaptive shifts along deterministic pathways depending on the taxon. As with the refinements offered by Ricklefs and Cox (1972), competition and

1. Colonization

2. Ecological shifts in response to new colonists

3. Further specialization as new colonists arrive

4. Extinction of fragmented specialized species

resulting from extrinsic ecological events such as natural disturbances or climate change. If the taxon cycle is an intrinsic evolutionary process, then species lineages will reflect a history consistent with the cycle. The taxon cycle predicts that clades will form within islands and contain representatives of diverse ecological phenotypes, whereas an extrinsic process predicts simultaneous evolutionary events across multiple lineages. Phylogenetic studies permit testing of these hypotheses but have produced variable results. Although some researchers have found that periods of expansion or contraction for particular species do not necessarily correspond to extrinsic environmental events, and that the stages of the taxon cycle do show a historical progression, with correlation between increased specialization and geographical distribution, other studies have uncovered evolutionary histories that show that species occupying similar ecological niches on multiple islands are more closely related to each other than are species within an island. The taxon cycle therefore remains a controversial theory for explaining species distribution and abundance on islands. SEE ALSO THE FOLLOWING ARTICLES

FIGURE 1 Stages of the taxon cycle.

Ants / Extinction / Fragmentation / Lizard Radiations

habitat shift are predicted as primary mechanisms driving the shifts. Roughgarden and Pacala (1989) have applied the taxon cycle and competitive displacement concepts to body size of lizards on islands rather than to ecological specialization. PROPOSED MECHANISMS FOR THE TAXON CYCLE

Various mechanisms by which the taxon cycle operates have been theorized, but all stem from the impact of competition between established and colonizing species both directly and indirectly. However, critics of the taxon cycle theory have contended that competition between colonizing species and established species is unlikely to result in the successful establishment of the newly arrived species. Roughgarden and Pacala’s (1989) model of competitive displacement based on body size has resulted in a range of outcomes, only some of which are consistent with a taxon cycle scenario.

FURTHER READING

Erwin, T. L. 1985. The taxon pulse: a general pattern of lineage radiation and extinction among carabid beetles, in Taxonomy, Phylogeny and Zoogeography of Beetles and Ants. G. E. Ball, ed. The Hague: Dr. W. Junk, 437–472. Liebherr, J. K., and A. E. Hajek. 1990. A cladistic test of the taxon cycle and taxon pulse hypotheses. Cladistics 6: 39–59. Pregill, G. K., and S. L. Olson. 1981. Zoogeography of West Indian vertebrates in relation to Pleistocene climatic cycles. Annual Review of Ecology and Systematics 12: 75–98. Ricklefs, R. E., and E. Bermingham. 2002. The concept of the taxon cycle in biogeography. Global Ecology and Biogeography 11: 353–361. Ricklefs, R. E., and G. W. Cox. Taxon cycles in the West Indian avifauna. American Naturalist 106: 195–361. Roughgarden, J., and S. Pacala. 1989. Taxon cycles among Anolis lizard populations: review of the evidence, in Speciation and its Consequences. D. Otte and J. A. Endler, eds. Sunderland, MA: Sinauer Associates, 403–432. Wilson, E. O. 1959. Adaptive shift and dispersal in a tropical ant fauna. Evolution 13: 122–144.

TEPUI SEE PANTEPUI

PHYLOGENETIC EVIDENCE FOR THE TAXON CYCLE

Another criticism of the taxon cycle is whether it exists at all as an evolutionary process or is instead a pattern

TIDAL WAVES SEE TSUNAMIS

TIDES

913

humans, who predict the depth of water in a harbor in order to dock a ship, but also by the many living organisms in the sea that make use of tidal currents to transport a regular flux of nutrients or suspended food particles past them in order to feed. The regular cycle of exposure and inundation of beaches, tidal pools, and tidal flats is essential for the survival of many oceanic life forms.

TIDES MARLENE NOBLE U.S. Geological Survey, Menlo Park, California

Tidal sea-level fluctuations, which are fundamentally caused by the gravitational attractions between the rotating Earth and the moving positions of the sun and the moon, not only cause the sea level to regularly cover and expose beaches over most of the Earth but also transport water, suspended material, nutrients, larvae, and debris onto or off of the adjacent landmass. Because of their regularity, these tidal fluctuations are well known at any particular location not only by

A axis of Earth’s rotation

equator Moon tidal bulge

B

axis of Earth’s rotation north pole

4 1

equator tidal bulge

north pole

Moon

3 2

C axis of Earth’s rotation

Moon equator tidal bulge

FIGURE 1 Equilibrium tide on a water-covered Earth, (A) if the Earth

and moon were not rotating around a common center of mass and (B) if the Earth and moon are rotating around a common center of mass as viewed from the side and from above the north pole of the Earth. The red dot denotes a specific location on the rotating Earth. Note that in a single day, two high (and low) tides are experienced as the red dot moves through positions 1 and 3 (2 and 4). (C) The equilibrium tide when the moon’s orbit is inclined to the Earth’s equator.

914

TIDES

SURFACE AND INTERNAL TIDES: GENERAL CHARACTERISTICS

Tidal sea-level fluctuations and tidal currents are generated by rhythmic interactions among the gravitational forces between the Earth, the moon, and the sun. Because the moon is much closer to the Earth than the sun is, and because the gravitational attraction force scales as one over the distance squared between objects, the strength of the moon’s gravitational attraction on the Earth is about twice that of the sun, even though the sun is much larger and more massive. Hence the largest tides are caused by the gravitational attraction between the Earth and the moon. One might think that because (1) the moon’s force on the Earth is largest on the side of the Earth that is closest to the moon, and (2) the moon’s position with respect to the Earth remains relatively constant in a 24-hour period, the ocean on an entirely water-covered Earth would bulge out toward the moon on the side of the Earth that is closest to the moon (Fig. 1A). Because the Earth spins on its axis once a day, this would lead to a daily (diurnal) tidal cycle in water-level fluctuations as a particular location on the Earth moves under, then out from under, the tidal bulge in sea level. However, because the Earth and moon jointly revolve around a common center of mass once every 27.3 days, the centrifugal force on the Earth caused by that rotation is directed approximately opposite to the gravitational force from the moon. The total centrifugal force in the Earth– moon system balances the gravitational forces at the center of their joint rotation. But these forces are not everywhere the same. The centrifugal force is less than the gravitational force on the side of the Earth facing the moon; it is larger than the gravitational force on the side of the Earth farthest from the moon. Hence, sea level on a water-covered Earth has two bulges, one toward and one away from the moon (Fig. 1B). Therefore, a particular point on a rotating Earth generally experiences two highs and two lows in sea level each day (i.e., a semidiurnal tide). The semidiurnal tides would be of similar magnitudes if the moon revolved directly around the Earth’s equator. However, because the moon’s orbit is inclined to the equator, the tidal bulge is not located on the axis of the Earth’s rotation (Fig. 1C). A

Amplitude

particular spot on the Earth usually experiences two high tides a day, but often the amplitude of one tide is much larger than that of the other. Both diurnal and semidiurnal tidal cycles would be observed on a water-covered Earth-Moon system. Given that the Earth–sun system has analogous gravitational forces, although different rotational cycles, the interactions of all these planetary bodies cause some of the complexity in the tidal cycles seen on the Earth. Based on the strength of the astronomical forcing, the largest tidal constituent is the semidiurnal principal lunar tide, M2, which has a period of 12.42 hours. The other large semidiurnal constituent is the principal solar tide, named S2, which has a period of 12.00 hours. The diurnal tides, which are the other dominant astronomical tidal constituents, are the principal lunar diurnal tide, O1, with a period of 25.82 hours, and the luni-solar diurnal tide, K1, with a period of 23.93 hours. An additional layer of complexity is added when one realizes that the Earth is not a water-covered sphere. The Earth’s ocean basins are separated by large continental land masses that prevent the tidal bulges in sea level from moving freely around the Earth. In addition, the combination of basin topography and the influence of Coriolis force on the ocean currents causes the tides in ocean basins to travel around amphidromic points within each basin. Tidal elevations are nearly zero at the amphidromic points and increase with distance from that point. Hence tidal fluctuations at islands in the same ocean basin may be zero if the island is close to an amphidromic point, such as Tahiti or Madagascar, and be much larger if the island is farther from that point. The amplitudes of tidal sea-level fluctuations at the major frequencies are typically 0.5 to 2 m. At any one location, the tidal range for any particular constituent, such as M2, is fairly constant because the causes of the tidal fluctuations are linked to the movements of the Earth, sun, and moon and the large-scale topography of that site. However, because there are several different tidal constituents in the major tidal bands, such as M2 and S2 or O1 and K1, the individual tidal constituents move in and out of phase with each other, adding or subtracting to the amplitude of the observed sea-level fluctuations. In the semidiurnal band, the principal lunar tide, M2, usually beats against the principal solar tide, S2, moving in and out of phase with each other every 14.8 days (Fig. 2). This is denoted the Spring/Neap cycle. The principal diurnal tides beat against each other every 13.6 days. Only twice a year are the two principal diurnal and the two principal semidiurnal tides in phase. This is typically when one observes the highest, and lowest, tides of the year.

Days

FIGURE 2 The Spring/Neap cycle in tidal heights created by using

simulated semidiurnal M2 (12.42 hour) and S2 (12.00 hour) tidal constituents. In this example, the amplitude of M2 is twice as large as S2.

The currents intrinsically caused by the tidal forcing tend to be classified into two types of flows: barotropic and baroclinic. The barotropic, or surface, tidal current is linked to the tidal fluctuations in sea level. They are oscillating flows at tidal periods that, theoretically, have uniform amplitude with depth and time. Barotropic tidal currents are the only tidal current present in a body of water that has a constant density. Baroclinic, or internal, tidal currents exist because there are vertical gradients in the temperature and salinity, and therefore in density, in the ocean. They are usually generated when barotropic tidal currents flow over submerged topographic features in the ocean. Baroclinic tidal currents account for most of the vertical structure in observed tidal currents and are not connected to the tidal sea-level fluctuations. The amplitude of the internal tides varies with time, partly because local density gradients vary with time. TIDES, BEACHES, AND BIOTA

In the shallow regions that fringe an island’s beaches, the most obvious effect of tidal sea-level fluctuations is that they regularly raise and lower sea level nearly simultaneously around the entire island. The beaches can be covered at high tide (Fig. 3A), whereas tidal flats are exposed at low tide (Fig. 3B). A rising, or flooding, tide will fill lagoons adjacent to the beach with salty ocean water that may be nutrient-rich and contain suspended material such as plankton. If the tidal lagoon is river-dominated, the seawater will mix with the freshwater. The strength and extent of mixing will depend on the amplitude of the

TIDES

915

FIGURE 3 The (A) rising and (B) falling tide on the south coast of

Molokai, Hawaii. Figures courtesy of Mike E. Field, U. S. Geological Survey.

of this resuspended sediment is then carried alongshore by the prevailing currents or offshore on the subsequent ebbing of the tide. This moving zone where sediment is resuspended, and then transported on a tidal cycle, can be critical to the development of shallow coral reefs because coral prefers to grow in clear water. The corals can expand their habitat because the flooding tide allows waves to carry sediment deposited in these shallow waters out of the region. Wind-driven shore-parallel currents, normally reduced in the shallow water off the beach by friction at low tide, can also be enhanced at high tide. These shore-parallel currents move larvae, suspended materials, and pollutants over the shallow water just off the beach. Although tidal fluctuations obviously affect the shallow regions around island margins, they can also alter the current flow in narrow straits between adjacent islands. For example, because sea-level fluctuations do not rise and fall synchronously on both ends of the strait between Maui and the Big Island in Hawaii, an along-strait pressure gradient is generated that drives strong tidal currents up and down this strait. These strong currents can enhance local wind-generated waves, delay ships sailing against the current, and enhance the transport of water and other materials along the islands. TIDES AND BARRIER ISLANDS

flooding tide and the strength of the river flow. Several hours later, the falling, or ebb, tide will remove much of this mixed water from the lagoon. This regular cycle of flooding and ebbing tidal currents is essential for the health of the lagoon, for it tends to enhance the exchange of larvae, plankton, and other suspended materials between the lagoon and the coastal ocean. The raising and lowering of sea level just off the island beaches, or over the fringing coral reefs, also changes the physical processes in the coastal ocean that affect these regions. During low tide, surface waves propagating toward the shore from the surrounding ocean feel the seabed at some distance from the shoreline. Typically a surface wave will feel the bed and begin to shoal or break when the water depth is less than one-half the wavelength of the wave. On a rising tide, the water depth increases, and deep-water surface waves can travel farther toward the shore. Larger, locally generated, wind-driven waves can also develop in this deeper water. These waves shoal, or break, closer to the shoreline on a rising tide, thus moving the location where the energy from those waves can resuspend fine sediment closer to the beach. Some

916

TIDES

Barrier islands are low-relief islands that lie just offshore of continental land masses, usually where one or more large rivers supply sediment to the coastal ocean. The barrier islands are typically grouped as a series of small landmasses elongated parallel to a continental shoreline. A typical set of barrier islands is found in the Gulf of Mexico, near the mouth of the Mississippi River. Barrier islands react to tidal fluctuations similarly to other islands, but because they lie close to a continental shoreline, they are even more sensitive to interactions between tidal fluctuations and other processes that change sea-level height. The local alongshore winds can be strong enough to raise sea level at the continental and barrier island shoreline by tens of centimeters. When this wind-driven increase in sea level is associated with a rising, rather than a falling, tide, the potential for flooding on that barrier island is substantially increased. TIDES AND SEAMOUNTS

Seamounts, which are found throughout the ocean basins, are essentially submerged islands. Their tops can lie anywhere from 100 to 2000 m below the sea surface, so they

are not affected by the relatively small rise and fall of the surface tides. Nor do the currents associated with the surface tide, which have amplitudes of 2–4 cm/s in the deep ocean, directly affect them. However, when surface tidal currents flow over the very gently sloping summit of a large seamount, they can generate much larger internal tidal currents, with speeds of 10 cm/s and higher. Depending on the location of the seamount, these internal tidal currents either propagate away from the seamount or are trapped to that seamount. The frequencies of both the freely propagating and trapped internal tidal currents depend on the shape of the seamount, the density profile of the surrounding ocean water, and the latitude of the seamount. The amplification of tidal currents around seamounts is locally important in that the enhanced internal tidal currents tend both to increase the supply of particulate food to benthic organisms and to disperse their larvae over and possibly off the seamount. It is clear that the internal tidal currents do reach speeds strong enough to sweep detritus and fine sediment off the seamount, thus allowing ferromanganese crusts to precipitate out of the cold ambient seawater onto the hard rock substrate of that seamount. These crusts, which occur over vast areas of the sea floor, are rich in economically important minerals, such as cobalt and platinum. SEE ALSO THE FOLLOWING ARTICLES

Barrier Islands / Beaches / Sea-Level Change / Seamounts, Geology

TIERRA DEL FUEGO MATTHEW J. JAMES Sonoma State University, Rohnert Park, California

JOHN M. WORAM Rockville Centre, New York

Tierra del Fuego is the extensive archipelago of large and small islands at the southern tip of South America, separated from the mainland by the Strait of Magellan. Its total area is 73,746 km2, two-thirds of which is owned by Chile, one-third by Argentina. The largest island within the archipelago is Isla Grande de Tierra del Fuego (Fig. 1). GEOGRAPHIC SETTING

The archipelago is further divided by the Beagle Channel running along the southern coast of Isla Grande. Along the Argentine territory, it forms the border with the Chilean islands to the South. The Chilean territory contains Cape Horn and False Cape Horn (both located on islands). Cape Horn is located on Isla Hornos in the Hermite Islands group, a small archipelago at the very southern extent of Tierra del Fuego. The cape was not named for its shape, but rather for the city of Hoorn in the Netherlands, the birthplace of Dutch navigator Willem Cornelisz Schouten, who first sailed around the cape in 1616. Although Cape Horn is considered the south-

FURTHER READING

Grant, S. B., B. F. Sanders, A. B. Boehm, J. A. Redman, J. H. Kim, R. D. Mrse, A. K. Chu, M. Gouldin, C. D. McGee, N. A. Gardiner, B. H. Jones, J. Svejkovsky, G. V. Leipzig, and A. Brown. 2001. Generation of Enterococci bacteria in a coastal saltwater marsh and its impact on surf zone water quality. Environmental Science and Technology 35: 2407–2416. Holloway, P. E., and M. A. Merrifield. 1999. Internal tide generation by seamounts, ridges, and islands. Journal of Geophysical Research 104: 25,937–25,951. Noble, M., D. A. Cacchione, and W. C. Schwab. 1988. Observations of strong mid-Pacific internal tides above Horizon Guyot. Journal of Physical Oceanography 18: 11,300–11,306. The Open University. 2006. Waves, tides and shallow-water processes, 2nd ed. Jointly published by the Open University, Milton Keynes, and Butterworth Heinemann. Storlazzi, C. D., E. K. Brown, and M. E. Field. 2006. The application of acoustic doppler current profilers to measure timing and pattern of larval dispersal. Coral Reef 25: 369–381. Storlazzi, C. D., A. S. Ogston, M. H. Bothner, M. E. Field, and M. K. Presto. 2004. Wave- and tidally-driven flow and sediment flux across a fringing coral reef: southeastern Molokai, Hawaii. Continental Shelf Research 24: 1397–1419.

FIGURE 1 Isla Grande and other islands constituting the Tierra del

Fuego archipelago.

TIERRA DEL FUEGO

917

ern tip of South America, Islas Diego Ramirez are farther south, with Islote Aguila being the southernmost in that group. False Cape Horn is a headland at the southern tip of Isla Hoste and is the southernmost point of one of the large islands that constitute Tierra del Fuego, but it is quite often mistaken for Cape Horn itself, especially by sailors approaching from the west. Also in Chilean territory, Cape Froward is the southernmost extremity of the South American continental land mass, located on the Brunswick Peninsula on the north shore of the Strait of Magellan, south of Punta Arenas. FLORA AND FAUNA

The region supports an attenuated southern flora and fauna, as well as dominant introduced species such as the North American beaver, the European rabbit, and sheep, as well as the native guanaco. Forests within the Tierra del Fuego National Park contain tree species of Lenga, Guindo, and Ñire. In his Journal and Remarks (later, Voyage of the Beagle), Charles Darwin mentions “vegetation thriving most luxuriantly, and large woody stemmed trees of Fuchsia and Veronica in full flower” that was noted on a previous expedition. Darwin himself claims to “have seen parrots feeding on the seeds of the winter’s bark, south of latitude 55°.”

route.” Adding to the negative experiences from tempestuous passages around Cape Horn were the remarks of Philo White (1789–1883), who rounded the Horn in 1841 in the U.S. sloop of war Dale and wrote: “It is now going on four weeks, that we have been off Cape Horn! buffeting strong gales of contrary winds,—wearing and tacking ship, in endeavors to make progress against tremendous head swells, and strong adverse currents,—amidst furious snow squalls, and hail and sleet storms: There is consequently much suffering amongst the crew.” Tierra del Fuego has the world’s southernmost city (Ushuaia), national park (Parque Nacional Cabo de Hornos), highway (Argentina’s RN 3), and brewery (Cervecería Fueguina). SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / Juan Fernandez Islands / Voyage of the Beagle FURTHER READING

Armstrong. P. 2004. Darwin’s other islands. New York: Continuum International Publishing Group. Olivero, E. B., and D. R. Martinioni. 2001. A review of the geology of the Argentinian Fuegian Andes. Journal of South American Earth Sciences 14: 175–188.

TOKELAU ISLANDS HUMAN HISTORY

Events that took place in Tierra del Fuego during the first of three surveying voyages of HMS Beagle include the suicide of Royal Navy Captain Pringle Stokes in 1828, the appointment of Flag Lieutenant Robert FitzRoy as his replacement, and FitzRoy’s subsequent kidnapping of four Fuegian Indians in 1830, which ultimately led to Charles Darwin’s participation in the second Beagle circumnavigation-surveying voyage in 1831–1836. It was on this voyage that Captain FitzRoy returned three of the Fuegian Indians—a fourth had died of smallpox in England—to their home on Isla Navarino, on the Chilean side of the Beagle Channel. Tierra del Fuego in general is known for its harsh weather, and Cape Horn in particular is renowned in the history of sailing for the difficulty experienced by vessels and their crews when “rounding the Horn.” Buccaneer William Ambrosia Cowley described his experience in 1684: “The weather in the lat. of 60 deg. was so extream cold that we could bear drinking 3 quarts of Brandy in 24 hours each Man, and be not all the worse for it.” With perhaps less brandy on hand, Captain David Porter of the U. S. frigate Essex offered this recommendation after his own passage in February 1813: “I would advise those bound into the Pacific, never to attempt the passage of Cape Horn, if they can get there by any other

918

TONGA

SEE PACIFIC REGION

TONGA DONALD R. DRAKE University of Hawaii, Manoa

Tonga is an archipelago of small, tropical, Pacific islands consisting of limestone or volcanic rock. The natural vegetation is tropical rain forest, and the native vertebrate fauna includes bats, birds, and lizards. The terrestrial environment has been strongly modified by humans, resulting in deforestation, animal extinctions, and invasion by alien species, although significant remnants of the native biota still exist. GEOGRAPHY AND PHYSICAL ENVIRONMENT

The Kingdom of Tonga is a South Pacific nation consisting of about 700 km2 of land divided among 170 scattered islands (Fig. 1). The archipelago lies just west of where the Pacific tectonic plate is being subducted beneath the Indian-Australian plate. This geological activity has produced the 10,000-m-deep Tonga Trench in the east and

nificant contact with Europeans through Dutch explorer Abel Tasman in 1643. The current population of approximately 110,000, two-thirds of whom live on Tongatapu, is spread across 45 islands. The combination of long human occupation, high population density, and small island size has resulted in significant environmental modification, including extensive deforestation and extinction. Tonga has always been an independent nation with indigenous governance, and for much of its history it was ruled by a paramount chief. In 1875, the ruler established a constitutional monarchy and became the first in a hereditary line of five kings and queens that extends to the present. Tonga is the last Polynesian country to be ruled by a hereditary monarch. Since 1875, the government has largely been controlled by members of the royal family and the traditional nobles. In recent years, however, a pro-democracy movement has been advocating political change. Tonga’s economy is somewhat constrained by the country’s small land area, which limits its natural resources. Land is widely distributed among the population, and most families cultivate a diverse range of crops for home consumption or sale. Many people are subsidized by remittances from relatives living overseas. There is a modest tourist industry. BIOTA Biogeography

FIGURE 1 The Kingdom of Tonga. From Steadman (2006), courtesy of

University of Chicago Press.

the Tongan Islands in the west. The islands form two north–south lines paralleling the trench: raised limestone islands in the east and volcanoes in the west. The limestone islands range from small cays to high, terraced landforms (Fig. 2). The limestone islands are covered with 1–4 m of fertile soil derived from volcanic ash. The volcanic islands are all either active or dormant and, in many places, have relatively young surfaces with poorly developed soil. Although there are no permanent streams, several islands have freshwater lakes, ponds, or marshes. The climate is tropical to subtropical. The prevailing winds are the southeast trade winds. Mean annual rainfall is approximately 1900–2300 mm, with November to April being slightly warmer and wetter. Cyclones occasionally strike the islands during the wet season.

Because Tonga has never been connected to a continent, its native biota is limited to those organisms that were able to colonize via long-distance dispersal. Hence, the biota is less diverse than that of Australasia, but more diverse than that of the distant island groups of eastern Polynesia. Tonga’s biota is largely a subset of that of its larger neighbors, Fiji and Samoa. Although few species are endemic to Tonga, many are endemic to the Tonga–Fiji–Samoa biogeographic

HUMAN HISTORY

Tonga was originally colonized by the ancestors of modern Polynesians 2800–3000 years ago and first made sig-

FIGURE 2 Small, raised limestone islands in the Vava‘u Group. Note the

extensive fringing reef around Taunga (left) and the terraced topography on ‘Euaiki (right).

TONGA

919

FIGURE 3 Diverse rain forest covers the upper slopes of the national

park on ‘Eua, a 312-m-high island consisting of a volcanic core covered by limestone terraces.

region. In addition, this region represents the furthest natural penetration into the Pacific for many groups of organisms, including gymnosperms, mangroves (Rhizophoraceae), snakes, and honeyeaters (Aves: Meliphagidae). Terrestrial Flora and Vegetation

Tonga’s native flora has about 340 species of angiosperms, two species of gymnosperms, and 83 species of ferns and fern allies. Only eight angiosperms, one gymnosperm, and one fern are endemic. The natural vegetation for nearly all of Tonga is tropical rain forest (Fig. 3). Mangrove vegetation occurs in some lagoons and other sheltered, coastal sites. Forests on coral-sand beaches are dominated by coastal trees common throughout the tropical South Pacific, whereas inland rain forests are dominated by species with narrower geographic ranges. Today, most variation in rain forest composition on the limestone islands is determined by a site’s history of disturbance by humans or cyclones rather than by variation in the physical environment. Only the highest islands exhibit significant altitudinal variation in species composition. Vegetation on the sparsely inhabited volcanic islands is poorly known. Terrestrial Fauna

Tonga’s native vertebrate fauna is limited to those groups of animals most capable of long-distance dispersal across seawater: bats, birds, and reptiles. The invertebrate fauna is less well known. 920

TONGA

Tonga once supported two species of large, fruit-eating bats (flying-foxes) and one small, insectivorous bat (Emballonura semicaudata). One of the flying foxes (probably Pteropus samoensis, based on subfossil remains) is locally extinct but still occurs in Fiji and Samoa. The other, P. tonganus, is a widespread South Pacific species that is still common in Tonga. With the extinction of many of Tonga’s large, native birds, P. tonganus plays a key role in forest dynamics as one of the few seed dispersers of large-seeded trees. Birds are Tonga’s most diverse native vertebrates. There are 21 species of land birds in 20 genera today; 26 more species are known from the subfossil record but disappeared following human colonization. The only extant endemic birds are the Tongan whistler (Pachycephala jacquinoti) of Vava‘u and the Niuafo‘ou megapode (Megapodius pritchardii) of Niuafo‘ou. Pigeons and fruit doves (Columbidae) appear to have been major components of the prehistoric bird fauna, but many are extinct. In addition, about 20 species of seabirds occur in and around the islands, though not all breed there. Tonga has many species of reptiles but no amphibians. Although sea turtles and sea snakes use coastal habitats, the only truly terrestrial reptiles in Tonga today are lizards, including four genera of geckos and five of skinks. Tonga also has extinct and extant iguanas (Brachylophus spp.), a tropical American group of reptiles otherwise known in the South Pacific only from Fiji. The subfossil remains of the Pacific boa (Candoia bibronii) suggest this snake was once native to Tonga. Marine Biota

Tonga’s limestone islands are surrounded by extensive coral reefs that support a diverse marine flora and fauna. During the winter, humpback whales (Megaptera novaeangliae) migrate from Antarctica to Tonga to give birth and mate. CONSERVATION

Tonga’s natural environment has been strongly impacted by the direct and indirect effects of humans and the hundreds of alien plant and animal species they have introduced. Today, most of the land is used for villages or agriculture, and few islands retain more than 10% of their original forest cover. Most remaining forest is on steep slopes or sparsely inhabited volcanic islands. However, the government has also set aside several areas, such as ‘Eua National Park (Fig. 3), in an effort to conserve Tonga’s natural heritage. Alien mammals, such as rats (Rattus spp.) and cats (Felis catus) pose significant threats to plants and animals that evolved in the absence of mammalian predators.

SEE ALSO THE FOLLOWING ARTICLES

Fiji, Biology / Fossil Birds / Makatea Islands / Peopling the Pacific / Samoa, Biology FURTHER READING

Fall, P. 2005. Vegetation change in the coastal-lowland rain forest at Avai‘o‘vuna Swamp, Vava‘u, Kingdom of Tonga. Quaternary Research 64: 451–459. Flannery, T. 1995. Mammals of the south-west Pacific and Moluccan Islands. Chatswood, Australia: Reed Books. Franklin, J., S. K. Wiser, D. R. Drake, L. E. Burrows, and W. R. Sykes. 2006. Environment, disturbance history and rain forest composition across the islands of Tonga, Western Polynesia. Journal of Vegetation Science 17: 233–244. McConkey, K. R., and D. R. Drake. 2006. Flying foxes cease to function as seed dispersers long before they become rare. Ecology 87: 271–276. McConkey, K. R., D. R. Drake, H. J. Meehan, and N. Parsons. 2003. Husking stations provide evidence of seed predation by introduced rodents in Tongan rain forests. Biological Conservation 109: 221–225. Meehan, H. J., K. R. McConkey, and D. R. Drake. 2002. Potential disruptions to seed dispersal mutualisms in Tonga, Western Polynesia. Journal of Biogeography 29: 695–712. Mueller-Dombois, D., and F. R. Fosberg. 1998. Vegetation of the tropical Pacific Islands. New York: Springer-Verlag. Nunn, P. D. 1994. Oceanic islands. Oxford, UK: Blackwell Publishers. Steadman, D. W. 2006. Extinction & Biogeography of Tropical Pacific Birds. Chicago: University of Chicago Press. Wiser, S. K., D. R. Drake, L. E. Burrows, and W. R. Sykes. 2002. The potential for long-term persistence of forest fragments on a large island in western Polynesia. Journal of Biogeography 29: 767–787.

species (including those extinct within historic times) and appears first in the fossil record around 60 million years ago. Tortoises are found on all continents except Australia and Antarctica. Giant continental forms disappeared from mainland ecosystems at the close of the Pleistocene, perhaps coincident with the spread of humanity out of Africa. Most land tortoises possess a suite of characteristics that distinguish them from all other turtles, such as a domed carapace, elephantine feet with two phalanges or fewer in each digit, and several skull characters. A xeric lifestyle, common among tortoises, may be one reason why tortoises survived and became gigantic on island ecosystems prone to climatic and resource variability and lengthy drought. Phylogenetic studies of comparative anatomy and more recent studies of molecular systematics have confirmed both the monophyly (shared common ancestry) of tortoises and the independent origin (polyphyly) of insular giant tortoises.

TORTOISES CHARLES R. CRUMLY University of California, Berkeley

Size change is one of the most common patterns of evolution on islands—dwarfism in some cases and gigantism in others. Land tortoises become giants. Three independently evolved lineages of giant tortoises survived until historic times (Figs. 1–4). The Galápagos Islands tortoise populations include a dozen or so species, some extinct and others with small but recovering populations. Two other lineages of giants evolved on different island groups of the Indian Ocean. Only one species of these two lineages survives, mostly on the isolated atoll of Aldabra. Land tortoises represent, in the public conscience, both the pattern of insular gigantism and a vivid example of the process of evolution.

FIGURE 1 Chelonoidis nigra porteri from Santa Cruz. Photograph

by Adalgisa Caccone and gratefully used here with permission.

THE TESTUDINIDAE

All land tortoises belong to the Testudinidae, a monophyletic lineage of turtles that includes approximately 50 living

FIGURE 2 Chelonoidis nigra hoodensis from Española. Photograph

by Adalgisa Caccone and gratefully used here with permission.

TORTOISES

921

du Jardin des Plantes, Paris.

FIGURE 4 Cylindraspis vosmaeri from the Muséum National d’Histoire Naturelle (a female specimen, number AC-A5222).

THE PATTERN IN VERTEBRATE EVOLUTION

INDEPENDENT EVOLUTION

Size change is common in vertebrate evolution, especially on islands. Some lineages are well known for including species much larger, whereas others include lineages much smaller than close mainland relatives. Although it often seems obvious that a particular radiation of organisms evolved by changing size, it can be difficult to accumulate evidence sufficient to support a hypothesis regarding the cause of size increase or decrease. Even determining the size of the ancestor can be complicated by the absence of a fossil record, a poorly studied phylogenetic history, and an inadequate number of specimens of intermediate sizes. Thus, evolutionary size change both is perceived as common and, paradoxically, is difficult to corroborate because affirming evidence is scant. This was certainly true for land tortoises until the recent publication of reliable and corroborated hypotheses of tortoise phylogeny. Whether island animals become giants or dwarfs seems partly influenced by physiology. Those animals with body temperatures that do not vary with ambient temperatures, referred to as endotherms, can become gigantic or can become dwarfs, such as the smaller elephants, rhinoceroses, and hippopotami. In contrast, vertebrates such as tortoises, whose body temperatures match the ambient temperature of the environment (ectotherms), tend to become gigantic. Examples of this pattern include several lineages of lizards as well as land tortoises. Many, but not all, of these groups are herbivores and capable of living through lengthy periods of drought and starvation. Insular ectothermic giants often possess long life spans, show sporadic and unpredictable levels of recruitment, and inhabit islands usually free of predators and competitors (Table 1).

In land tortoises, there have been many instances of size change; dwarfism on continents is just as common as gigantism on islands. Indeed, because the phylogenetic history is relatively well documented, it is easier to confirm that island lineages are giants. It is not as easy to determine whether the first tortoises to reach these islands were as large as they are today. Indeed, it is likely that all of these cases involve size increase and that they evolved independently. And in every case, the tortoises reached their island refuges by over-water dispersal. In fact, there are many reports of tortoises found at sea after being pitched overboard accidentally or intentionally during storms or conflicts, or naturally in the central lagoon of Aldabra. Because giant tortoises can survive lengthy periods of over-water transport, dispersal is acknowledged as the likely means of colonization.

FIGURE 3 A captive Aldabrachelys gigantea from the Ménagerie

922

TORTOISES

Galápagos

Tortoises of the Galápagos are the most diverse surviving lineage of insular giant tortoises, and they are most closely related to the tortoises of South America. As many as 13 taxa are classified as subspecies of Chelonoidis nigra. Five subspecies share Isabela, each restricted to a volcanic cone separated from neighboring cones by impassable lava fields. Pinzon, San Cristóbal, Santa Cruz, and Santiago each harbor a separate unique species. A single adult male (“Lonesome George”) is the last of the Pinta population, and he now lives at the Charles Darwin Research Station on Santa Cruz. Three species—two named (one on Fernandina, another on Santa Maria) and one unnamed (on Santa Fe)—are extinct. In addition, genetic studies have revealed unnamed populations that might warrant taxonomic recognition.

TABLE 1

Giant Tortoises on Islands Listed by Fritz and Havas (2007) Taxon

Distribution

Estimated Population Size

Galápagos

Chelonoidis nigra abingdonii Chelonoidis nigra becki Chelonoidis nigra chathamensis Chelonoidis nigra darwini Chelonoidis nigra ducanensis Chelonoidis nigra guntheri a Chelonoidis nigra hoodensis Chelonoidis nigra microphyesa Chelonoidis nigra nigra Chelonoidis nigra phantastica Chelonoidis nigra porteri Chelonoidis nigra vandenburghia Chelonoidis nigra vicina

Pinta Isabela, Volcán Wolf San Cristóbal (introduced onto Rábida) San Salvador (or Santiago) Pinzón Isabela–Sierra Negra Española Isabela–Volcán Darwin Santa María (or Floreana) Fernandina Santa Cruz Isabela–Volcán Alcedo Isabela–Cerro Azul

1 (extinct in the wild) 1000–2000 500–700 500–700 150–200 100–300 15 (native) 500–1000 Extinct Extinct 2000–3000 3000–5000 400–600

Western and central Madagascar Aldabra and Granitic Seychelles Southwestern Madagascar

Extinct 85,000 Extinct

Réunion Mauritius Rodrigues and Île aux Aigrettes Mauritius Rodrigues

Extinct Extinct Extinct Extinct Extinct

Aldabra, the Seychelles, and Madagascar

Aldabrachelys abrupta Aldabrachelys gigantea Aldabrachelys grandidieri Mascarenes

Cylindraspis indica Cylindraspis inepta Cylindraspis peltastes Cylindraspis trisserata Cylindraspis vosmaeri

note: Galápagos estimated population sizes from Caccone et al. 2002. a Some experts have suggested that guntheri, microphyes, and vandenburghi are conspecific with vicina. Genetic studies support a close relationship among all the taxa on Isabela (see Fig. 7), but these populations are geographically isolated from one another.

These are among the largest of the living land tortoises, approaching 300 kg and 1.2 m long. Although the maximum age in nature is not known, it is estimated that Galápagos tortoises often live to be at least 200 years old. Galápagos tortoises are informally divided on the basis of shell shape (Fig. 5). Some have domed carapaces similar to those of familiar mainland tortoises. But some possess a carapace whose anterior perimeter is raised, creating a shape like a medieval Spanish saddle. Tortoises with saddlebacked carapaces are smaller than their domed relatives and usually occur on lower, drier islands. A raised anterior carapacial edge may confer a competitive advantage. Thomas Fritts reported observations of male/male contests; the individual whose head is raised highest wins. In his observations of mixed captive herds, large domed forms lose contests with smaller saddlebacked opponents, because the raised anterior edge of the shell permits the smaller form to raise its head further. Furthermore, gaining access to higher branches of trees and shrubs is an advantage conferred by a saddlebacked carapace. Examples of the saddlebacked form

include forms from Pinzon, Española, and Pinta (Fig. 6). Domed forms are found on Santa Cruz and on Isabela. Intermediate forms also occur. During the first decade of the twenty-first century a team of researchers at Yale University, including Adalgisa Caccone, Jeffrey Powell, and Michael Russello, studied the genetic relationships among and between Galápagos tortoises. This team was able to corroborate the earlier hypothesized relationship between Galápagos tortoises and the Chilean tortoise Chelonoidis chilensis. They also proposed the first explicit hypothesis of the phylogeny of Galápagos tortoises (Fig. 7) and discovered that some populations are related to nearby populations—as would be expected. For example, all isolated populations on Isabela are more closely related to each other than to any tortoise population found elsewhere in the archipelago; in fact, there is so little genetic variation that some believe nearly all the Isabela populations are the same form. But Caccone and co-workers also discovered unexpectedly close relationships between geographically distant popu-

TORTOISES

923

lations (e.g., Chelonoidis nigra hoodensis from Española in the far southeast of the archipelago share genetic markers with C. n. abingdonii from Pinta in the far north). Finally, by interpreting mitochondrial DNA evidence, the researchers were able to hypothesize instances of humans transporting tortoises from one island to another. Seychelles and Aldabra

FIGURE 5 The shells of Galápagos tortoises can be domed, like those

of most other tortoises, or saddlebacked. Here are three views of the domed condition of the Santa María tortoise Chelonoidis nigra nigra with the front of the shell to the left: (A) ventral view of the plastron; (B) dorsal view of the carapace; (C) lateral view of the shell. This is Plate 18 from Garman (1917), drawn from Museum of Comparative Zoology specimen number 4479.

FIGURE 6 Three views of the saddlebacked condition of Galápagos tor-

toise shells as exemplified by the Pinta tortoise Chelonoidis nigra abingdonii: (A) ventral view of the plastron; (B) dorsal view of the carapace; (C) lateral view of the shell. This is Plate 40 from Garman (1917) drawn from Günther’s (1877) plate 40 and 41. It is the type specimen of C. n. abingdonii and is stored in the Natural History Museum in London.

924

TORTOISES

Once present on several islands of the Indian Ocean (Madagascar and islands of the Seychelles group), this lineage is now represented by a single species, mostly restricted to the atoll of Aldabra. The ecosystem on Aldabra is dominated by tortoises – one of the rare cases wherein a herbivore is the dominant taxon. Indeed, the ground vegetation on Aldabra is referred to as “tortoise turf.” Individuals of this species (Aldabrachelys gigantea) have also been introduced elsewhere and captive populations are maintained, partly against the event of the catastrophic loss of the only remaining native populations on Aldabra. Although several different scientific names have been proposed for living tortoises in this lineage, unambiguous genetic evidence documents variation typical of a single living species that is most closely related to tortoises of Madagascar. Composed of several smaller islets (Grande Terre, Malabar, and Picard), the atoll of Aldabra supports three separate populations, each with slight differences in size, abundance and certain demographic factors such as fecundity. About 90% of Aldabra tortoises occupy Grande Terre, the largest part of the atoll. These tortoises lay about a third the number of eggs laid by Malabar tortoises and about a fourth the number of eggs laid by Picard tortoises. The average egg size and the typical maximum body size also vary from islet to islet. The largest individuals occur on Picard, whose tortoises lay the greatest number of the smallest eggs. All members of this lineage, extinct as well as the one living species, possess a feature unique among all turtles: a vertically expanded external narial opening. This feature is coincident with an unusual drinking behavior. On Aldabra infrequent rainfall gathers in shallow depressions. In most tortoises, drinking is done by submerging the head in water and oscillating the gular region of the neck to pump water into the esophagus. This behavior requires water sources deep enough to permit full emergence. But the puddles on Aldabra are too shallow to allow this. Instead, Aldabra tortoises dip their nostrils into puddles and suck water up through their nasal passages. Specialized internal soft tissues prevent the fouling of the olfactory epithelium.

abingdoni

Española

possess a single, rather than paired, gular scute and, unlike Aldabrachelys, lack a nuchal scute.

San Cristóbal

A CONFUSION OF NAMES

Pinta

hoodensis chatamensis chatamensis

SANTA CRUZ (Cerro Fatal)

porteri becki1

Isabela

becki2 darwini

Santiago

ephippium

Pinzón

undescribed

SANTA CRUZ (Cerro Monturra)

guntheri1 guntheri2 guntheri3 vicina1 vicina2 guntheri4

Isabela

vicina3 vicina4 vicina5 vandenburghi microphyes Floreana

undescribed

SANTA CRUZ (La Caseta)

porteri

0.005 substitutions/site FIGURE 7 The evolutionary relationships among Galápagos tortoises

(modified slightly from Russello et al. 2005) confirm that they are very closely related. Other studies (Caccone et al. 2002) propose a shared common ancestor from around 2–3 million years ago. Most parts of this diagram are well supported by evidence from mitochondrial DNA. The relationships among very closely related Isabela populations are not as well supported. The numbers after species names are used to document populations of the same taxon with almost identical and yet still distinctive types of mitochondrial DNA.

Mascarenes

The most poorly known of the giant island tortoises are members of the Cylindraspis lineage, which includes five named species, all of which are extinct. They disappeared before significant museum collecting in the mid-1700s to early 1800s. Those specimens that did make it into museum collections were often fragmentary or without reliable locality data. Thus, there are too few specimens with too little information on provenance to allow for detailed systematic analysis. One feature that unifies these tortoises is the tendency toward shell ankylosis when approaching adult size. In addition, some Cylindraspis

There are many names associated with giant island tortoises and where they have lived. Early studies of Galápagos tortoises used English names and, for some islands, there are multiple Spanish names. Thus, linking specimens to localities can be difficult and is sometimes impossible. Human transport of specimens, without any record or reliable documentation, further confounds efforts to name and classify tortoises. This is true for both Galápagos tortoises and the tortoises of the Indian Ocean. Limited samples of Galápagos tortoises with poor locality data contributed to the same species being named more than once. An inadequate appreciation of variability also generated invalid names. Studies of the phylogeny of testudinids have required changes in the classification of all tortoises including Galápagos species. Once most tortoises of the neotropics were allocated to Testudo, then to Geochelone, and are now referred to as Chelonoidis. For technical nomenclatural reasons, the species epithet elephantopus was replaced first by nigrita and then by nigra despite many years of consistent usage of elephantopus. The application of different species concepts generated both species and subspecies names for the Galápagos forms. Most races are essentially isolated reproductively, a prerequisite for species recognition using the biological species concept. But genetic evidence suggests that, in some races, interbreeding has occurred naturally as well as resulting from human transport of tortoises from one population to another. The capacity to interbreed is, under the biological species concept, an indication that individuals are the same species. Thus, recognition of the species versus subspecies status of Galápagos tortoises depends upon the importance one places on observations that can be interpreted differently. The name used for the Aldabra tortoise is even more confused. Stable and often used for nearly a century, the epithet gigantea was rejected by some well-meaning but misguided experts based on disputable or ambiguous interpretation in favor of the strict adherence to nonbiological codes of nomenclature. Over the past 20 years, questionable nomenclatural choices have compounded this confusion. And in some instances, definitive evidence has been ignored or discounted in order to cling to falsified hypotheses. Thus, in the most recent checklist of turtles compiled by Uwe Fritz and Peter Havaš, nearly 40 names, employed during the last 30 years, were included in the synonymy of Aldabrachelys gigantea.

TORTOISES

925

HUMANS AND INTRODUCED SPECIES

All insular giant tortoise populations evolved and survived on islands free of predators and relatively free of competition until human exploration and invasive species began processes of decline and extinction. Rats escaped sailing ships and ate eggs and hatching tortoises. Goats, brought on ships for meat and milk, became feral and ate vegetation that might have sustained tortoises. And, of course, sailors loaded their ships with tortoises, whose flesh sustained crews during long ocean voyages. The tortoises of the Indian Ocean were driven extinct by the combined impact of all these factors. And these same factors have been in operation in the Galápagos; why have the tortoises of Galápagos fared better? The difference seems partly due to when disturbances began. In the case of Indian Ocean tortoises, sailing ship visitations, colonization, hunting, and introduced animals occurred several hundred years earlier than in Galápagos. In addition, human visits to the Mascerenes and Seychelles were more frequent than to the Galápagos. Furthermore, permanent settlements were established on Indian Ocean islands long before any settlements were established in Galápagos. Aldabra tortoises avoided extinction because the atoll is remote and outside regular sailing routes, and there is no permanent water. CONSERVATION

The success of most conservation programs depends on timing and effort. For Cylindraspis, no effort was made because conservation was not a priority in the eighteenth and nineteenth centuries. Aldabrachelys on Madagascar became extinct between 750 and 1250 years before present, well after the first appearance of humans on the island. Today, active conservation efforts are ongoing in Galápagos and on Aldabra. The Charles Darwin Research Station was built by the Charles Darwin Foundation and inaugurated in 1964. The Station is headquartered on Santa Cruz and manages continuing efforts to help in recovery of Galápagos tortoises. For races rare in their native range, the Station raises hatchlings until they are large enough to be released back to their native habitat. Other activities include the eradication of goats and rats. Financial support comes from organizations and institutions, as well as individuals. The government of Ecuador, which exercises sovereignty over the archipelago, established the Galápagos National Park Service and deserves special praise for the commitment made to preserve Galápagos biodiversity. The work of the Station is one of the success stories in conservation of biodiversity and habitat restoration. Aldabra has been vigorously protected through a variety of programs and with considerable international participa-

926

T R I N I DA D A N D TO B AG O

tion. Many of the same institutions and organizations that support the activities of the Charles Darwin Research Station also support conservation efforts on Aldabra. SEE ALSO THE FOLLOWING ARTICLES

Adaptive Radiation / Galápagos Islands, Biology / Gigantism / Madagascar / Seychelles FURTHER READING

Austin, J. J., E. N. Arnold, and R. Bour. 2003. Was there a second adaptive radiation of giant tortoises in the Indian Ocean? Using mitochondrial DAN to investigate speciation and biogeography of Aldabrachelys (Reptilia, Testudinidae). Molecular Ecology 12: 1415–1424. Caccone, A., G. Gentile, J. P. Gibbs, T. H. Snell, H. L. Snell, J. Betts, and J. R. Powell. 2002. Phylogeography and History of Giant Galápagos Tortoises. Evolution 56.10: 2052–2066. Fritts, T. H. 1984. Evolutionary divergences of giant tortoises of Galapagos. Biological Journal of the Linnean Society 21: 165–176. Fritz, U., and P. Havaš. 2007. Checklist of the Chelonians of the world. Vertebrate Zoology 57: 149–368. Garman, S. 1917. The Galapagos tortoises. Memoirs of Museum of Comparative Zoology 30.4: 261–296. Günther, A. 1877. Gigantic land-tortoises (living and extinct) in the collections of the British Museum. London: Taylor and Francis. Le, M., C. J. Raxworthy, W. P. McCord, and L. Mertz. 2006. A molecular phylogeny of tortoises (Testudines: Testudinidae) based on mitochondrial and nuclear genes. Molecular Phylogenetics and Evolution 40: 517–531. Pritchard, P. C. H. 1996. The Galápagos tortoises: nomenclatural and survival status. Chelonian Research Monographs 1: 1–85. Russello, M. A., S. Glaberman, J. P. Gibbs, C. Marquez, J. R. Powell, and A. Caccone. 2005. A cryptic taxon of Galápagos tortoise in conservation peril. Biology Letters 1.3: 287–290. Van Denburgh, J. 1914. The gigantic land tortoises of the Galapagos archipelago. Proceedings of the California Academy of Sciences 4th Ser., 2.I: 203–374.

TRADE WINDS SEE CLIMATE ON ISLANDS

TRINIDAD AND TOBAGO CHRISTOPHER K. STARR University of the West Indies, St. Augustine, Trinidad and Tobago

Trinidad and Tobago are two small islands with a combined land area of about 5100 km2, lying just off the northeast edge of the South American continent (Fig. 1) at 10°02´–11°21´ N and 60°31´–61°55´ W. Southwest Trinidad is separated from the mainland by an 11-km strait, whereas in the northwest there are steppingstone islands between Trinidad and the mainland. Tobago is separated from Trinidad by a 36-km strait. Trinidad’s Northern

FIGURE 1 Trinidad and Tobago, position and topography. The present -125-m line approximates the coastline at the height of the most recent gla-

ciation about 20,000 years ago. Map by Bheshem Ramlal.

Range and Tobago are eastern extensions of Venezuela’s long Coastal Range. CLIMATE AND TOPOGRAPHY

The islands are characterized by moderate topography—maximum elevation 940 m for Trinidad, 576 m for Tobago—and by a climate typical of their tropical latitude. Mean annual rainfall varies from about 125 to about 325 cm, according to locality, with a moderately distinct dry season from about mid-January to late May. Mean daily temperature fluctuation is estimated at 10.4 °C, with very little seasonal difference. Trinidad and Tobago lie south of the usual path of Atlantic hurricanes and have not been significantly affected by them in most decades. PEOPLE AND GOVERNMENT

The two islands, together with various associated islets, form the Republic of Trinidad and Tobago. About 80% of the populace of 1.3 million is of Indian and African descent in equal proportions, with small minorities of people of other races and of mixed descent. English is the language of public affairs, and no other language is spoken by large numbers. The government of this former British colony, independent since 1962, is a parliamentary democracy in the British model. The economy is semi-industrialized and is heavily reliant on the petro-

leum industry (Trinidad) and on tourism (Tobago) and hardly at all on agriculture. Per-capita GDP is variously estimated at US$15,500–17,500. Life expectancy at birth is 74 years for women, 68 years for men. ENVIRONMENT

The tectonic history of Trinidad and Tobago is complex and controversial. However, they appear to have undergone no significant movement or other gross disturbance since the Tertiary. Although it is difficult to plot Quaternary sea-level changes, they are thought to have caused several cycles of isolation and reunification with the mainland. The age of present isolation is generally estimated at 10,000 years for Trinidad and 14,000 years for Tobago, although a minority view holds that a land bridge connected Trinidad to the mainland at least intermittently until much more recently. These fluctuations in land area were presumably accompanied by cyclical changes in gross habitat type, as throughout northern South America. The greatest extent of savanna, relative to forest, occurred during glacial maxima (most recently about 20,000 years ago), and it is estimated that seasonal evergreen forest came to cover about 75% of the land surface by 10,000 years ago and to remain at about that figure through pre-Columbian times. Forest cover is now reduced to about 20–30%, depending on

T R I N I DA D A N D TO B AG O

927

definition, although the decline of agriculture over about the last century has slowed the pace of deforestation. The predominant natural land habitat is evergreen seasonal forest, found in wetter areas up to about 250 m. Other habitats of note include swamp forest (most notably on the east coast of Trinidad), mangrove (on the east and west coasts of Trinidad and in southwest Tobago), savanna (in central and southwest Trinidad), and lower montane forest (above about 250 m on both islands), with some elements of montane forest in the highest parts of Trinidad’s Northern Range. Coastal habitats include many sand beaches, the major Buccoo Reef at the southwest end of Tobago, and several lesser coral reefs in Tobago and northeast Trinidad. Each island has a great many streams, but no significant rivers or natural lakes.

resentation in the Guianas or eastern Venezuela appears to be absent from Trinidad and Tobago. As rough estimates, these islands harbour about 3% of the world’s land and freshwater animal species and about 2% of plant species. It is expected that over an extended period of time, a continental island will increasingly partake of the biotic features of an oceanic island: decreased diversity, increased disharmony, and increased endemism. We can refer to these outcomes col-

BIOTA

Whereas the rest of the West Indies—the Antilles—are oceanic islands, Trinidad and Tobago are typical continental islands. That is, they show only slight endemism, and they closely resemble comparable nearby mainland habitats in their (harmonic) biotic composition and diversity (Fig. 2). In addition, they are relatively resistant to invasive species and their effects. Endemism among the approximately 6600 species of seed plants, for example, is estimated at 2.1%. To cite some other well-studied examples, the corresponding figure for land vertebrates is 2 of 521 species (0.4%) (Fig. 3), for butterflies (sensu stricto, excluding Hesperiidae) is 5 of 387 species (1.3%), and none of the 42 known species of freshwater fishes is endemic. In line with this trend, no family of plants or animals with strong rep-

FIGURE 3 The golden treefrog, Phyllodytes auratus (A), one of Trini-

dad and Tobago’s very few putative endemic species, known only from the upper reaches of Trinidad’s two highest peaks. It breeds in

928

FIGURE 2 Like much of Trinidad and Tobago’s biota, the social wasp

the water that accumulates among the bracts of Glomeropitcairnia

Mischocyttarus alfkeni is very broadly distributed in South America.

erectiflora (B). This tank bromeliad, although not rare, is known only

It nests in a variety of lowland habitats on many different substrates.

from high elevations in Trinidad and nearby parts of Venezuela. Photo-

Photograph by Allan W. Hook.

graphs by Daniel G. Thornham.

T R I N I DA D A N D TO B AG O

lectively as the “island effect.” The earliest of these features to appear is likely to be the first, a lowering of diversity as a result of uncompensated local extinction, or “relaxation,” which may be the engine of the island effect as a whole. To what extent is an island effect manifest in Trinidad and Tobago? This question is only now coming to be addressed, by way of floristic and faunistic comparisons between Trinidad’s Northern Range and similar habitat in Venezuela’s Paria Peninsula. After some 10,000 years of separation, it is predicted that the magnitude of Trinidad’s island effect will vary in a meaningful way among taxa. Preliminary results suggest, for example, that the diversity of social wasps (Polistinae) is much the same in Trinidad as in comparable habitats on the mainland, whereas that of stingless bees (Meliponini) is markedly lower. CONSERVATION ISSUES

Trinidad and Tobago are a signatory of several international agreements relating conservation and the environment, including CITES, the Convention on Wetlands (Ramsar), the Convention on Biological Diversity, and the Cartagena Convention. Furthermore, a relatively high proportion of land area is under public ownership, and much of this remains in a natural or semi-natural state. A contributing factor here is undoubtedly the heavy dependence of the national economy on petroleum and, to a lesser extent, tourism, which limits pressure on the land for agricultural purposes. At the same time, legal protection remains weak. Much of the country’s conservation policy and infrastructure dates back to colonial times. There is still no formal system of national parks and protected areas that meets today’s international standards, and the few designated conservation areas enjoy little real protection. Even in these areas, poaching and logging are relatively unchecked. However, the growth of ecotourism, together with the presence of a number of active conservation-related NGOs federated under a national umbrella body, are promising signs. Allied with this latter factor is a perceptible, ongoing shift in government toward an increased local participation in management of the natural environment. The most striking conservation success story of recent times is the rise of community-based patrolling of sea-turtle nesting beaches in both Trinidad and Tobago. This earns substantial revenue from both domestic and foreign ecotourism and has reduced poaching of adult turtles and eggs to a fraction of its former level. Another promising development is a move toward formal designation of a well-preserved, 90km2 forested area in northeastern Trinidad as the Matura National Park, again with community involvement.

SEE ALSO THE FOLLOWING ARTICLES

Antilles, Biology / Endemism / Island Biogeography, Theory of / Relaxation / Sea-Level Change FURTHER READING

Brereton, B. 1981. A history of modern Trinidad, 1783–1962. Kingston, Jamaica: Heinemann. Living World, journal of the Trinidad and Tobago Field Naturalists’ Club. http://livingworldjournal.googlepages.com/home. Murphy, J. C. 1997. Amphibians and reptiles of Trinidad and Tobago. Malabar, FL: Krieger. Woods, C. A., and F. E. Sergile, eds. 2001. Biogeography of the West Indies: patterns and perspectives, 2nd ed. Boca Raton, FL: CRC.

TRISTAN DA CUNHA AND GOUGH ISLAND PETER G. RYAN University of Cape Town, South Africa

Renowned for supporting the most remote human community, the Tristan archipelago and Gough Island are small, cool-temperate, volcanic islands in the central South Atlantic. The islands range in age from 0.2 to 18 million years, resulting in a wide diversity of topography. Their isolation has led to high levels of endemism among the biota. Despite being discovered more than 500 years ago, Tristan was settled only in the early 1800s and is the only permanently inhabited island. The other islands have been relatively little impacted by humans. Currently, the main threats to native species are introduced rodents as well as a suite of introduced plants. LOCATION AND PHYSICAL STRUCTURE

Tristan da Cunha is an archipelago of three main islands located almost midway across the Atlantic Ocean between the southern tip of Africa and South America. Gough Island lies 350 km south-southeast of Tristan (Fig. 1). They are the only cool-temperate islands in the South Atlantic; the nearest other islands are St. Helena to the north, and frigid Bouvet Island to the south. The islands are volcanic in origin, rising up steeply from the abyssal plain. Despite being only 20–30 km apart, they are separated by deep trenches, with water that is more than 500 m deep between Inaccessible and Nightingale, and more than 2000 m deep between these islands and Tristan. The islands differ greatly in age and, as a consequence, in size and height (Table 1).

T R I S TA N D A C U N H A A N D G O U G H I S L A N D

929

Edinburgh Tussock grass Bogfern heath Fern bush Wet heath Sphagnum bog Feldmark

30 km

Tristan

Gough 32 km

Inaccessible

20 km

5 km

5 km

Nightingale FIGURE 1 Map of the Tristan archipelago and Gough Island showing

main vegetation types. Scale bars apply to the islands only.

The youngest and largest island, Tristan, is still an active volcano, whereas the oldest and smallest island, Nightingale, has been eroded until only the core of tough trachyte rocks remains. Marine erosion outstrips fluvial erosion, resulting in steep sea cliffs and hanging valleys. Soils are generally shallow and poorly developed, with a mantle of peat in many areas. Peat slips are frequent, especially at Gough Island. There are numerous perennial streams on Gough and Inaccessible, but not on the other two islands. On Tristan, rain rapidly soaks into the porous lava flows, emerging as springs around the base of the island. The low elevation of Nightingale results in lower rainfall, but swampy ponds occur in depressions on the island’s summit (Fig. 2). CLIMATE

The climate is cool-temperate and oceanic, with relatively little seasonal variation in temperature or rainfall. Gough Island is distinctly colder, wetter, and windier than the Tristan group, lying as it does on the edge of the roaring

forties. The average temperature near sea level at Gough is 12 °C (–3 to 25 °C), compared with 15 °C (2 to 25 °C) at Tristan. Rain falls year round, usually associated with the passage of cold fronts, averaging 1670 mm per year at Tristan but closer to 3000 mm at Gough. Temperatures decrease and rainfall increases with elevation. Snow is regular on the peak of Tristan and on the highlands of Gough in winter, and orographic clouds are often found over the islands. The prevailing winds are from the west but veer to the northeast prior to the passage of a cold front, then back steadily to the south or southwest as the front moves through. Average wind speed is 36 km/h at Tristan and 44 km/h at Gough, with a tendency for stronger winds in winter. Tristan and Gough were not glaciated during the last ice age, and analysis of pollen cores suggests that the vegetation (and hence the climate) has remained fairly constant for at least the last 20,000 years. However, over the last 40 years, average air temperatures have increased by 0.6 ºC, and climate change models predict further increases of 1–5 ºC over the next century. A warmer climate is likely to favor alien, introduced species that outcompete the native biota. BIOTA AND TERRESTRIAL HABITATS

Tristan and Gough have never been connected to a continent, so all terrestrial animals and plants, and the shallow-water marine biota, have had to disperse across the ocean. Most immigrants arrived from South America and adjacent islands, thanks to the prevailing westerly winds and currents. However, some species have arrived from southern Africa, and some species are shared with Amsterdam and St. Paul, temperate islands in the central Indian Ocean. Once they reached the islands, many evolved to adapt to their new island home, resulting in endemic species. These include all seven landbirds, four of 22 breeding seabirds, 27 of 50 native flowering plants, 16 of 35 ferns, and close to 100 invertebrates (total diversity unknown). Levels of endemism are high among some marine groups too, notably seaweeds and bivalves. In some cases, the new colonists underwent adaptive radiations to exploit the many vacant niches. At Tristan, the best-studied example is among the endemic

TABLE 1

The Area, Height, and Age of the Tristan da Cunha and Gough Islands

Tristan Inaccessible Nightingale Gough

930

Area (km2)

Height (m)

Age (million years)

Most recent eruption

96 14 4 65

2060 600 350 910

0.2 3–4 18 3–5

1961 50,000 years ago 200,000 years ago 100,000 years ago

T R I S TA N D A C U N H A A N D G O U G H I S L A N D

HUMAN HISTORY

FIGURE 2 Swampy “ponds” on the summit of Nightingale Island, with

Middle Island and Tristan in the background.

Nesospiza buntings, with small-billed dietary generalists and large-billed specialists evolving to exploit the woody fruits of Phylica trees. This has resulted in two well-segregated species at Nightingale Island and three only partly segregated ecomorphs at Inaccessible Island. Other radiations have occurred among Scaptomyza flies (with two flightless, strap-winged species), Tristanodes weevils (11 species), Balea land snails (nine species), Agrostis grasses (seven species) and Nertera chicken berries (three species). Unfortunately, the relative paucity of competitors and predators on the islands renders their biota highly susceptible to extinction when new species are introduced. Terrestrial habitats tend to be segregated by altitude. Coastal areas are dominated by tussock grassland. Spartina arundinacea dominates the coastal lowlands and cliffs up to 500 m at the Tristan islands, and is joined by the smaller Paridochloa flabellata at Gough Island. Fern bush is a diverse community found above coastal tussock, up to around 800 m at Tristan and 500 m at Gough. It is characterized by two large and distinctive species: the cycad-like fern Blechnum palmiforme and the island trees Phylica arborea. Wet heath is a fairly short, transitional vegetation type, containing elements of other vegetation types. It occurs from the upper limit of fern bush to above 800 m and contains fewer ferns than fern bush, with a higher proportion of mosses, grasses, sedges, and other flowering plants. At even higher elevations and on more exposed ridges, wet heath gives way to feldmark, an assemblage of dwarf, cushion-forming plants. Bogs are widespread at the islands, forming in hollows where drainage is impeded. There are two main types: Some support floating mats of big bog grass Scirpus sulcatus, whereas others are dominated by Sphagnum moss.

The islands were first discovered by Portuguese explorers pioneering a sailing route around Africa: Gough in 1505 and Tristan in 1506. Despite plentiful water, fish, seals, and seabirds, the islands remained uninhabited because they lacked safe anchorages. A proposal to establish a British penal colony on Tristan was rejected in favor of Australia, and it was only when commercial sealing started in the late eighteenth century that protracted visits were made to the islands. Gangs of sealers were put ashore from 1790, killing thousands of seals for their skins and oil. Vegetable gardens were established at Tristan, and goats, pigs, and poultry were introduced. The first attempt to settle the islands was led by Jonathan Lambert, a Yankee whaler, in 1810, but this foundered when Lambert drowned in 1813. The islands were annexed by Britain in 1816, when a garrison was stationed at Tristan to prevent the French from using the islands as a base from which to free Napoleon from his exile on St. Helena. When the garrison withdrew in 1817, William Glass was given permission to remain at Tristan. The fledgling community, augmented by castaways and crew from passing ships, flourished until the 1870s, thanks to the many vessels, especially whalers, calling to trade for fresh produce. But thereafter the number of ships dwindled, thanks to the switch to steamships, the opening of the Suez Canal, and collapsing whale stocks. Tristan’s isolation was greatest during the early twentieth century, resulting in an increasing reliance on seabird populations for food and guano. Links with the outside world increased during the Second World War, when a naval garrison was stationed at Tristan. The island was evacuated in 1961, when a volcano erupted next to the settlement, causing the entire community to flee to the United Kingdom. Most residents returned to the island in 1963. Today, the islands form the U.K. Overseas Territory of Tristan da Cunha, led by an Administrator and elected Island Council. The community of some 270 people is largely self-sufficient, generating income from fishing and the sale of stamps, and meeting most of their food needs from farming sheep, cows, poultry, potatoes, and other crops (Fig. 3). Careful stewardship of marine resources is crucial to Tristan’s economy. HUMAN IMPACTS AND CONSERVATION

Early visitors and colonists exploited natural resources, leading to the near extinction of seals, whales, and some seabird species. Subantarctic fur seals (Arctocephalus tropicalis) found a refuge on the remote western coast of Gough Island, and their numbers have recovered since the cessation of sealing. However, southern elephant seals (Mir-

T R I S TA N D A C U N H A A N D G O U G H I S L A N D

931

FIGURE 3 Potato patches on the settlement plain at Tristan, backed by

the steep cliffs that lead to the Base. The Spartina tussock that once covered the plain has been replaced by a mix of introduced pasture plants, and of the native landbirds, only a few Tristan thrushes persist in the deep gulches.

ounga leonina) have not recovered, and only a small, relict population persists on Gough’s northeast coast. Human impacts on seabirds were most severe at the main island of Tristan, where Tristan albatrosses (Diomedea dabbenena) and southern giant petrels (Macronectes giganteus) disappeared, and populations of other seabirds were greatly reduced, exacerbated by predation by rats, which arrived on a shipwreck in 1882. Fortunately, the other islands have remained free of rats, but house mice (Mus musculus) were introduced by sealers to Gough Island in the 1800s and now pose a serious threat to the chicks of Gough buntings (Rowettia goughensis) and winter breeding seabirds, as well as to many native invertebrates (Fig. 4). At Inaccessible Island, feral pigs almost wiped out spectacled petrels and reduced the last population of Tristan albatrosses breed-

ing at the northern islands to just a few pairs. Luckily, the pigs died out before they managed to finish the job. Introduced plants outnumber native species at Tristan, and several species have reached the other islands. At Tristan the native vegetation has been almost entirely replaced by introduced pasture species on the coastal plains, and pasture species are also widespread on the island’s plateau or Base. There are also many introduced invertebrates at the islands, including representatives of many groups that had not reached the islands prior to the arrival of humans. Their impacts are not well known, but some native species have become quite rare. Fortunately, the need for environmental protection was recognized. In 1976, a conservation ordinance proclaimed Gough Island a nature reserve and provided some protection for seabirds at Tristan. Inaccessible Island subsequently was made a reserve, and together with Gough forms one of only two British Natural World Heritage Sites. The traditional harvesting of seabirds is now confined to Nightingale Island, where only great shearwaters and rockhopper penguin eggs may be collected. Tristan has protected almost half of its land area and has taken active steps to conserve its marine heritage. The community recently adopted a biodiversity action plan and appointed a conservation officer. Controls on imported goods are being tightened to reduce the risk of further accidental introductions to the islands, and eradication programs for invasive alien species are under way, targeting New Zealand flax (Phormium tenax) at Inaccessible and Nightingale and procumbent pearlwort (Sagina procumbens) at Gough Island. Plans are also being drawn up to eradicate rats and mice from Tristan and mice from Gough. SEE ALSO THE FOLLOWING ARTICLES

Atlantic Region / Biological Control / Introduced Species / Rodents / Seabirds / St. Helena FURTHER READING

FIGURE 4 The Gough bunting Rowettia goughensis is confined to

Gough Island. Its population has decreased alarmingly in recent years, apparently as a result of predation on eggs and chicks by house mice that were inadvertently introduced to the island by early sealing expeditions.

932

T R I S TA N D A C U N H A A N D G O U G H I S L A N D

Baker, P. E., I. G. Gass, P. G. Harris, and R. W. le Maitre. 1964. The volcanological report of the Royal Society Expedition to Tristan da Cunha, 1962. Philosophical Transactions of the Royal Society of London A 256: 439–578. Crawford, A. B. 1982. Tristan da Cunha and the roaring forties. Edinburgh: Charles Skilton. Munch, P. A. 1945. Sociology of Tristan da Cunha. Results of the Norwegian Scientific Expedition to Tristan da Cunha 1937–1938 13: 1–331. Preece, R. C., K. D. Bennett, and J. R. Carter. 1986. The Quaternary palaeobotany of Inaccessible Island (Tristan da Cunha group). Journal of Biogeography 13: 1–33. Ryan, P. G., ed. 2007. Field guide to the animals and plants of Tristan da Cunha and Gough Island. Newbury: Pisces Publications. Wace, N. M., and M. W. Holdgate. 1976. Man and nature in the Tristan da Cunha Islands. International Union for the Conservation of Nature and Natural Resources Monograph 6: 1–114.

TSUNAMIS EMILE A. OKAL Northwestern University, Evanston, Illinois

Tsunamis are gravitational oscillations of the entire body of water of an ocean basin, following a disruption in the bottom (or exceptionally the surface) of the ocean. They differ from more conventional swells by their much longer periods (typically from 10 min to 1 hr) and relatively faster speeds over deep ocean basins (typically 220 m/s, or the speed of a modern jetliner). Tsunamis are capable of exporting death and destruction across entire ocean basins, their propagation being limited only by continental masses. TSUNAMI SOURCES

Although tsunamis were once called “tidal waves,” they are not caused by tides. Most tsunamis are generated when very strong earthquakes deform the ocean floor. Such a mechanism can move extremely large amounts of water (the fault rupture reached 1200 km in the 2004 Sumatra event), but only over relatively short distances (at most 20 m in the largest earthquakes), resulting in long waves of considerable energy (> 1022 erg for the Sumatra tsunami). Secondary, less frequent, tsunami sources include landslides, which can be either submarine (e.g., Papua New Guinea, 1998; Storrega, Norway, 6000 BC) or aerial, falling into the water (e.g., Stromboli, 2002; Aysen, Chile, 2007). Finally, catastrophic volcanic eruptions in the marine environment (e.g., Santorini, 1630 BC; Krakatau, AD 1883) and bolide impacts at sea (Yucatan, 65 million years ago) can also give rise to major tsunamis. However, non-earthquake sources obey different source scaling laws, which in lay terms means that they displace water over considerable distances (up to hundreds of km), but involve more contained volumes (rarely exceeding 50 km in linear dimensions). As a result, their tsunamis, which can be devastating in the near field (less than 1000 km from the source), feature shorter wavelengths and experience more efficient dispersion while propagating over large distances, resulting in generally benign amplitudes in the far field. As the floor of the ocean has finite rigidity, it reacts to the passage of the tsunami by deforming elastically, resulting in a small, but significant, coupling of the tsunami to the solid Earth. Conversely, an earthquake source embedded in the solid Earth can excite a tsunami in an overlying ocean, but because the coupling is weak, appreciable transoceanic tsunamis are generated only by truly great earthquakes (of

moments ≥ 5 × 1028 dyn cm or so-called “moment magnitude” ≥ 8.7). The resulting tsunami wave remains in all cases relatively small on the high seas: Even for the catastrophic 2004 Sumatra tsunami, satellite altimetry provided a direct measurement of only 70 cm zero-to-peak in the Southern Bay of Bengal. Such low amplitudes are also spread over considerable wavelengths (~300 km), giving the tsunami a flat aspect ratio and rendering the wave undetectable on the open sea by classical (visual or optical) means. THE INTERACTION OF TSUNAMI WAVES WITH COASTLINES

When approaching a shoreline, a tsunami undergoes shoaling; that is, the wave slows down in the shallower water, while its amplitude increases considerably, with run-up heights at the shoreline having reached, during the Sumatra event, 32 m in the near field and up to 12 m in the far field. If the structure of the wave remains stable at the shoreline, it can continue to propagate over initially dry land and inundate the coastal areas in the form of a progressively rising swell over distances having reached, again in 2004, 10 km in the near field and 3 km in the far field. Otherwise, the wave breaks like surf and hits the shore as a wall of water or “bore,” particularly destructive but unable to propagate far inland. In very general terms, tsunamis with shoreline run-up of decimetric amplitudes (< 1 m) are generally benign; amplitudes of a few meters will result in significant destruction of individual structures and in instances of loss of life; and dekametric amplitudes (≥ 10 m) in total eradication of infrastructure and population. The exact run-up amplitude at a given shore location is a very complex function of the shape of the coastline and of the small-scale bathymetry and topography at the receiving beach. In particular, bays, harbors, and coves can resonate at typical tsunami frequencies, and their nonlinear responses can locally increase the wave amplitude. While such effects can be successfully modeled, they must be addressed on a case-by-case basis. In this general context, the following properties have been regularly observed, and justified theoretically. Near-Field Scaling

In the near field, the maximum run-up observed from a nearby seismic source along a smooth, linear coastline featuring no substantial indentation does not exceed twice the amplitude of the seismic slip on the fault. This simple rule of thumb (known as “Plafker’s law”) expresses a general scaling of the near field to the seismic source, verified by numerical simulations and in the field during the 2004 Sumatra

TSUNAMIS

933

tsunami: its local run-up, exclusive of splashes on cliffs, reached 32 m for a seismic slip estimated at 15 to 20 m. Any departure from the Plafker law is a proxy for the presence of an ancillary source, such as a submarine landslide, as was the case in Papua New Guinea (1998), Riangkroko, Flores, Indonesia (1992) or Unimak, Aleutian Islands (1946). Effect of Island Geometry

The response of an island to a distant tsunami depends crucially on the geometry of its structure, from the ocean floor up. In this respect, atolls, characterized by small dimensions, and steep underwater structures (with slopes reaching 40°) offer an overall smaller cross-section to the onslaught of the tsunami than do traditional islands sloping more gently to the ocean floor. As a result, all other parameters being equal, run-up on atolls has generally been smaller than on high islands. In practical terms, the tsunami is able to flow unimpeded around the structure, largely ignoring it, while the gently dipping, and necessarily larger structure of a high island would provide a physical barrier against which the wave has to abut, leading to a substantial transfer of momentum. This property was observed during the 2004 Sumatra tsunami in the Maldives, where the run-up was relatively contained (≈ 2 m), while it reached 9 m along the coast of Somalia, essentially in the same azimuth but at double the distance along the same ray paths. These results are a scaled-up expression of the well-known value of pillared structures (houses, etc.) in tsunami mitigation: the water flows effortlessly around the pillars whereas it would take down a continuous wall at the same location. Effect of Fringing Reefs

Coral structures fringing high islands provide some degree of protection to the shorelines. Although a tsunami can penetrate a lagoon and reach the shore of a reefed island, it propagates very inefficiently over the irregular and extremely shallow topography of the lagoon, resulting in a significant loss of energy before it reaches the high ground. As a result, the type of island most vulnerable to a tsunami is the unreefed high island. Numerous examples exist of this difference in vulnerability; for example, in Polynesia the Marquesas Islands, which are young, reefless volcanic high islands, have traditionally suffered much larger run-ups from distant tsunamis (Aleutian, 1946; Chile, 1960) than the nearby Society or Austral Islands, of comparable geological structure and age, but protected by substantial reef systems. Similarly, Mauritius (reefed) was much less affected by the 2004 Sumatra tsunami than its sister island of Réunion (unreefed).

934

TSUNAMIS

Effect of Small-Scale Topography

Small-scale island topography also plays a significant role in controlling the run-up of a tsunami wave at a coastline. In this respect, river beds and gulches are known to function as efficient channels of tsunami flow, often doubling or trebling the local amplitude of run-up. For example, the river valleys in the Marquesas Islands recorded up to 20 m of run-up during the 1946 tsunami, while nearby overland locations were typically 5 to 7 m. Similarly, during the 1993 Japan tsunami, run-up at a gulch on the West Coast of Okushiri Island reached 32 m, in rough numbers double its values along most of the nearby coastline. Numerical modeling at Okushiri has explained such high run-up as a result of the concentration of tsunami energy in the cove formed by the estuary of the gulch. Unfortunately, human settlement usually favors estuaries, which provide freshwater and communication routes to the hinterland, as well as locales featuring gaps in coral reefs, which provide easy access to the high seas, but considerably restrict the mitigating effect of the reef. Refraction Around Islands

Circular island structures with dimensions comparable to tsunami wavelengths can result in refraction of the wave along the island and focusing on the lee side of the island. This situation, which may go against common-sense intuition, was demonstrated dramatically at Babi Island during the 1992 Flores, Indonesia, tsunami. Even though this local tsunami approached the island from the north, the southern shore of the island suffered more devastation. This “Babi Island effect” was later both reproduced in the laboratory and explained theoretically. Wave Energy Spectrum

Even though the most perceptible energy in a tsunami is usually in the millihertz (mHz) frequency range (typical periods from 10 minutes to one hour), large earthquake sources contribute tsunami energy throughout a broad spectrum. Higher frequencies (typically in the 5–15 mHz range) have shorter wavelengths and no longer qualify as shallow-water waves. As a result, they are considerably dispersed; that is, their velocities across the ocean basins can be reduced to as little as 70 m/s, and this portion of the tsunami can reach distant coastlines as much as five hours after the main components of the tsunami. This “tail” to the tsunami wave was identified for the first time on hydrophone and seismic records of the 2004 Sumatra tsunami. While it carries comparatively less energy than the more traditional (and more rapidly propagating) components at longer periods, it can trigger oscillations

of large amplitudes in specific harbors when the tsunami spectrum matches their resonance frequencies. During the 2004 Sumatra tsunami, this has led to incidents in which large vessels broke their moorings in harbors of the Western Indian Ocean (Réunion, Madagascar) several hours after the passage of the more traditional tsunami waves. Although these phenomena can be simulated numerically given an adequate model of the harbor, they raise very sensitive issues regarding the duration of tsunami alerts and, in particular, the issuance of an “all clear” message to harbor communities.

lous character in real time from their seismic waves, they remain a formidable challenge, notably because they are poorly felt by the local population, which may then not be receptive to the issuance of a tsunami alert. Such a dramatic scenario occurred in Java on July 17, 2006, where waves ran up to 20 m and 700 people were killed despite a warning issued by the Pacific Tsunami Warning Center in Hawaii, which remained largely ignored by authorities along a section of shoreline where the earthquake, distant only 200 km, had hardly been felt. TSUNAMI MITIGATION

TSUNAMI WARNING

Because tsunamis propagate much more slowly than seismic waves (typically 220 m/s rather than 3–10 km/s), it is possible, at least in principle, to issue a warning to coastal communities, based on the interpretation of seismic waves in terms of earthquake source, and on the evaluation of the potential of the source for tsunami genesis. The full description of tsunami warning procedures transcends the scope of this article; however, the remaining challenges in this field are fundamentally of two kinds. First, it remains difficult to accurately quantify truly gigantic earthquakes in real time. In the case of the 2004 Sumatra event, the true size of the earthquake took about six weeks to assert, through a study of the free oscillations of the Earth. The major problem in this respect is that all real-time evaluation algorithms were by necessity (and to a large extent, continue to be) designed, implemented, and tested on earthquakes of lower magnitude, and the adjustment of their parameters to mega-events is far from a trivial task. Note, however, that the triggering of a tsunami alert is fundamentally a matter of overcoming a threshold, beyond which the exact size of the earthquake source is not crucial. In this respect, the Sumatra earthquake had been evaluated as having widespread tsunami potential within about 40 minutes of its source; the failure to issue adequate warnings for the far field had more to do with communications than with pure science. A second challenge is that of anomalous earthquakes disobeying scaling laws, whose rupture proceeds more slowly than in conventional sources, resulting in a significant deficiency of seismic release at the high frequencies (1 Hz) typical of shaking and damage to property (and to some extent at the intermediate frequencies (0.1 to 0.01 Hz) traditionally recorded on seismometers), while lowfrequency waves such as tsunamis are vigorously excited. The geological context in which these so-called “tsunami earthquakes” can take place remains obscure, and while we are making progress toward identifying their anoma-

Efforts to minimize the effect of tsunamis can take several forms. Passive mitigation relies on the building of structures designed to absorb or reflect the wave’s momentum before it can reach coastal infrastructure, houses, or individuals. Among them, tsunami walls have long been used, as in Japan, and are now engineered to optimize the reflection of the wave back towards the sea. However, they remain only as good as their height in relation to that of the incoming wave. The role of vegetation (mangroves or forest) has also been researched both in situ and through scaled experiments in the laboratory. The relocation of critical facilities (hospitals, schools, fire houses) is also necessary in low-lying areas prone to inundation during a tsunami. Finally, building construction can lessen tsunami damage, in particular through the use of stilts or pillars that provide a free flow through an empty first floor offering no cross section to the wave’s momentum. Active mitigation by individuals consists essentially of taking refuge at a combination of altitude and distance from the shore that the wave is not expected to reach. This evacuation depends crucially on the amount of time available before the arrival of the tsunami. In the near field, this may be as short as a few minutes, and the value of a centralized warning becomes marginal; the responsibility for evacuation away from the shore must be borne by the individual, as soon as shaking is felt or an anomalous behavior of the sea, most notably a regression exposing a normally submerged beach, is observed. (Note, however, that automatic systems not requiring human intervention, including closing sluices and stopping trains, can be successfully implemented in the near field.) A generally appropriate rule of thumb is to evacuate to an altitude of 15 m and to remain there at least three hours after anomalous wave activity has ceased. Motorized vehicles should be avoided in the near field, as they will almost certainly contribute to traffic jams. In low-lying areas providing no adequate relief, vertical evacuation must be used. During the 1998 Papua New Guinea tsunami (2200 deaths), some villagers

TSUNAMIS

935

survived by quickly climbing trees; high-rise buildings can serve (and are occasionally built for) the same purpose in developed communities, but the use of elevators during evacuation should be avoided. Evacuation platforms standing on pillars have been built in Japanese ports to provide harbor workers with a means of vertical evacuation at the workplace. Tsunami evacuation drills are regularly conducted in countries at risk, such as Japan and Peru. In the far field, tsunami alerts may benefit from several hours’ advance notice, which can be used for a more profound level of orderly evacuation over greater distances. In both fields, a critical aspect of a successful evacuation is some advance knowledge of the geometry of the expected flooding. This is achieved by running, before the fact, numerical simulations of the extent of flooding for a given community under various scenarios of local or distal tsunamis. These simulations use models of expectable sources, and computer codes solving the equations of hydrodynamics under the relevant initial conditions to map the inundation of the wave down to the scale of a city block, based on available small-scale bathymetry and topography. Their output is made available to civil defense and law enforcement officials, who can then review zoning, optimize evacuation procedures, and conduct drills. For example, the entire west coast of the United States is presently undergoing a systematic program of inundation mapping for all coastal communities.

from the hinterland shrugged off the earthquake and were swept by the waves as they tended to crops in the delta of the Camana River. Similarly, a number of tsunami-aware tourists, mostly from Japan, but also an 11-year-old British girl who had been taught about tsunamis at school, escaped the catastrophic 2004 tsunami on Thai beaches by recognizing anomalous down-draws as harbingers of disaster and immediately evacuating the beaches. Unfortunately, tsunami awareness inherited from ancestral tradition will fade after an estimated four or five generations in the absence of a recurring event; it is estimated that the recurrence time of an event of the size of the 2004 one is at least 400 years in Sumatra, and thus the local populations were not educated to this hazard (with the possible exception of the Moken people of the Surin, Andaman Islands, who live in complete isolation and may have been able to preserve their heritage longer). In this context, it is crucial to emphasize both the value of, and the need for, permanent education of populations at risk. The fundamental messages are simple: (1) Tsunamis are a natural phenomenon associated with the dynamic nature of the Earth, as opposed to supernatural occurrences; hence they must and will recur. (2) Upon feeling any kind of shaking along a shore line, or noticing an anomalous behavior of the sea, and in particular a strong down-draw, one should immediately evacuate to higher ground. Such simple precautions have repeatedly been proved to save lives.

THE VALUE OF EDUCATION

Above all, a number of recent occurrences have repeatedly shown that tsunami fatalities can be significantly reduced among a population educated to this kind of hazard. For example, following the Papua New Guinea tsunami of 1998, an informative video was developed, translated into many local languages, and shown in neighboring countries, including on battery-operated televisions in remote villages. When a large earthquake hit the island of Pentecost in Vanuatu just a few months later in 1999, the village chief immediately ordered its evacuation. Minutes later, the village was destroyed by the tsunami, with all but a handful of residents unharmed. In addition to formal education of children in the classroom, tsunami awareness can come as part of a community’s cultural heritage through parental or ancestral education in regions regularly affected by tsunamis. For example, fishing communities in Southern Peru suffered no casualties during the 2001 tsunami, as the villagers took to the hills as soon as they felt the earthquake and noticed a down-draw of the sea. By contrast, farm workers hired

936

TSUNAMIS

SEE ALSO THE FOLLOWING ARTICLES

Earthquakes / Eruptions / Landslides / Surf in the Tropics FURTHER READING

Geist, E. L., V. V. Titov, and C. E. Synolakis. 2005. Tsunami: wave of change. Scientific American 294: 56–63. Okal, E. A. 2008. The excitation of tsunamis by earthquakes, in Tsunamis. E. N. Bernard and A. R. Robinson, eds. The Sea 15. Cambridge, MA: Harvard University Press, 137–177. Okal, E. A., and C. E. Synolakis. 2004. Source discriminants for nearfield tsunamis. Geophysical Journal International 158: 899–912. Synolakis, C. E., E. A. Okal, and E. N. Bernard. 2005. The mega-tsunami of December 26, 2004. The Bridge 35.2: 26–35.

TUAMOTU ISLANDS SEE PACIFIC REGION

TYPHOONS SEE HURRICANES AND TYPHOONS

V VANCOUVER MARTIN L. CODY University of California, Los Angeles

Vancouver Island, 48–51° N latitude, is the largest island off the Pacific coast of North America, part of the western Canadian province of British Columbia, and the location of its provincial capital Victoria. The island is renowned for its soaring mountains, abundant lakes and waterfalls, spectacular coastal scenery, and imposing coniferous forests especially along the cooler and wetter western coastlines. It measures 450 km on its long axis (Cape Scott in the northwest to Victoria in the southeast), about four times its maximum width, with an area of 32,000 km2; the 2200-m Mt. Golden Hinde, in Strathcona Provincial Park, is at the highest point along the island’s rugged backbone. ORIGINS: GEOGRAPHY AND GEOLOGY

Vancouver Island is a continental shelf island isolated from the mainland by shallow straits: the Juan de Fuca Strait in the south separates the island from Washington State’s Olympic Peninsula, whereas mainland British Columbia lies east across the Georgia Strait, which narrows abruptly northward into the Johnstone Strait before opening into the Queen Charlotte Strait. The first recorded circumnavigation of the island in 1792 by Captain George Vancouver on the British Navy’s ships Discovery and Chatham undoubtedly drew on navigational talents honed on his earlier voyage to the region under Captain James Cook. The eponymous commander’s mission: to counter Spanish influence in the region, secure

the fur trade (in sea otter pelts), and settle the question of a northwest passage. Most of Vancouver Island, like other segments of the continent’s western coastline, is part of a tectonic microplate or “wandering terrane” termed Wrangellia, which dates from Devonian times (400 million years ago) and originated in the southern paleo–Pacific Ocean 10,000 km from its present position. Volcanics, marine carbonates, and intruded granites are mostly souvenirs of the terrane’s northward drift. Permian and Mesozoic fossil corals and ammonites preserved in the old sediments reveal the historical legacy in their close affinity with ammonites in southern Asia rather than to those on the North American plate. Accretion of Wrangellia to the North American plate occurred in the early Cretaceous (130 million years ago). Present-day island topography, resulting from platelet deformation, subduction, mountain uplift, deposition of new sediments, and most recently glaciation, evolved long since Vancouver Island docked with the continent; it was almost completely ice-covered during the successive glacial episodes of the Pleistocene, when contiguous ice sheets extended from the mainland over the Straits, the island, and 50–60 km beyond the present western coast. The last glacial recession (after ~18,000 years ago) signaled a reconstituted island status and recolonization via northern advances of forest, woodland, and coastal vegetation and associated fauna from southern refuges, including the ice-free southern half of the Olympic Peninsula. VEGETATION

The current diversity of habitats on Vancouver Island owes much to the climate gradient from outer to inner coast, determined largely by the mountainous interior. Points on the outer, western coast receive rainfall in excess of 3 m per year, about three times that of corresponding inner-coast

937

stations; the west has cooler summer temperatures (by ± 4 °C, with 30 rather than 80 days a year with daily maximum temperatures over 20 °C and less sunshine); winters are cooler on the inner coast. This climate shift corresponds to a turnover from cool and damp coniferous forests dominated by cedar Thuja plicata, hemlock Tsuga heterophylla, and spruce Picea sitchensis on the outer coast to drier, more open forest and woodlands of madrone Arbutus menziesii, Douglas fir Pseudotsuga menziesii, and Garry oak Quercus garryana on the inner coast. Overall, most of the southeastern and mainland-facing parts of the island enjoy moderate and benign climates unknown elsewhere in Canada, and the profusion of formal gardens and gardeners around the provincial capital attests to this. Some 2717 plant species are listed for British Columbia, 22.6% of them aliens; of these, 1604 are known to occur on Vancouver Island, and 150 or more are absent from the mainland. The dominant trees of the island’s magnificent coniferous forests range from northern California to Alaska, but many bog, alpine, or shoreline plants have much broader distributions (e.g., Circumboreal, Palearctic, or Nearctic). There are no plant endemics, although Vancouver Island is the only provincial locale for several British Columbian plant species whose ranges extend further north or further south. Two threatened near-endemics are Macoun’s meadowfoam Limnanthes macounii (Limnanthaceae) and Vancouver Island beggarticks Bidens amplissima (Asteraceae) of southeastern coastal areas, both now known to occur in a few sites off the island, with the former having recently been discovered (1998) in California. The Brooks Peninsula, a provincial park located on the northwestern coast of Vancouver Island, remained ice-free at the last glacial maximum and represents a refugium for alpine plants shared with the Haida Gwai (Queen Charlotte Islands) and Alaskan mountains to the north. FAUNA

With few exceptions, the terrestrial vertebrates of mainland and island are very similar. The island’s coasts abound in marine mammals (ten species of whales, dolphins, porpoises, sea lions, and seals) and bald eagles (Haliaeetus leucocephalus), and support diverse seabird colonies of gulls, cormorants, and alcids on the outlying rocks. The threatened marbled murrelet (Brachyramphus marmoratus) is an inland-breeding seabird still common around inlets that have not yet been logged. There is one spectacular mammal endemic, a large (up to 7 kg) ground squirrel, the Vancouver Island marmot (Marmotus vancouverensis), which is related to the similarly restricted Olympic

938

VA N C O U V E R

marmot (M. olympus), and to the more widespread Hoary marmot (M. caligula). Although listed as “endangered” since 1979, a captive breeding program has been successful; reintroductions to the wild have tripled the population size from its 1998 low of around 70 animals. HUMAN HISTORY

With the last glacial recession, ice-free coastal areas likely permitted the continent’s first humans, crossing from northeast Asia to North America via the Bering Strait land bridge, to expand south rapidly, settling the productive island coasts by as early as perhaps 12,000 years ago. When European interest and activity in the region intensified in the latter half of the eighteenth century, explorers found the coastal regions densely populated by First Nations peoples, such as the Nuu-chah-nulth (Nootka) on the west coast, living in complex organized and sophisticated societies. They had found ingenious ways to utilize a wide variety of marine and forest resources, especially salmon and cedar. Energetic territorial defense necessitated a near-constant state of internecine warfare, the lethality of which European arms, along with European diseases, enhanced. CONSERVATION ISSUES

The old-growth forests are now largely harvested out (about three-quarters gone, and only 10% left in the valley bottoms where the tallest trees occur); some of the remainder is preserved in the few parks and reserves (6% of the island area); the rest is being harvested at levels much above those sustainable over the longer term. The old growth is replaced by managed timber that cannot support the previously diverse canopy ecosystem (e.g., habitat-specific insects, spiders, and orobatid mites; voles; long-eared bats; hole-breeding owls and swifts; nesting murrelets) and subcanopy biota (e.g., ferns, slugs, lichens, the many endemic forest floor rove beetles, northern red-legged frogs [Rana aurora], and several salamander species), various components of which are considered endangered, threatened, or at risk. Forest streams and their denizens survive logging operations poorly; the salmon runs become sparse and sporadic. There is a strong popular voice for banning raw log exports and for conserving the remaining sparse acreage of old-growth forest. In that event, tourism based on the spectacular landscape, seascape, natural history, and more persistent natural resources will offer a better, more economically sound future. SEE ALSO THE FOLLOWING ARTICLES

Climate on Islands / Deforestation / Frogs / Vegetation

167

168

169

–13

166

170 N

–14

–14

Torres Islands

–15

–15

Banks Islands

Maewo Espiritu Santo Pentecost

–16

–16

Ambae (Aoba)

Ambrym

Malekula

–17

–17

Epi

Efate (Vate) –18

–18

Vanuatu

–19

–19

Erromango

–20

New Caledonia Ouvéa

Loyalty Islands 165

166

167

Anatom (Aneityum) Lifou 0 37,5 75 168

169

150 Km

–21

–20

Tanna

–21

Cannings, R., and S. Cannings. 1996. British Columbia: a natural history. Vancouver, BC: Greystone Books, Douglas & McIntyre. Cody, M. L. 2006. Plants on islands: diversity and dynamics on a continental archipelago. Berkeley: University of California Press. Cody, M. L., and J. McC. Overton. 1996. Short-term evolution of reduced dispersal in island plant populations. Journal of Ecology 84: 53–61. Douglas, G. W., G. B. Straley, D. E. Meidinger, and J. Pojar. 1998–2002. Illustrated flora of British Columbia. 8 vols. Victoria, BC: Ministry of Environment, Lands and Parks, Ministry of Forests, British Columbia Provincial Government. Duff, W. 1997. The Indian history of British Columbia: the impact of the white man. Victoria, BC: Royal British Columbia Museum. Krajina, V. J. 1973. Biogeoclimatic zones of British Columbia. Victoria, BC: British Columbia Ecological Reserves Committee. Ludvigson, R., and G. Beard. 1994. West Coast fossils: a guide to the ancient life of Vancouver Island. Vancouver, BC: Whitecap Books. Pojar, J. 1980. Brooks Peninsula: possible Pleistocene refugium on northwestern Vancouver Island. Botanical Society of American Miscellaneous Series Publication 158: 89. Vancouver: British Columbia Ministry of Forests and Lone Pine Publishing. Winchester, N. N. 1998. Severing the web: changing biodiversity in converted northern temperate ancient coastal rainforests, in Structure, process, and diversity in successional forests of coastal British Columbia. J. A. Trofymow and A. MacKinnon, eds. Northwest Science 72 (special issue #2), Washington State University Press. Yorath, C. J., and H. W. Nasmith. 1995. The geology of southern Vancouver Island: a field guide. Victoria, BC: Orca Book Publishers.

165

–13

FURTHER READING

170

FIGURE 1 Map of the position of Vanuatu. (From the Millenium Coral

Reef Mapping Project.)

VANUATU JÉRÔME MUNZINGER Institut de Recherche pour le Développement, Nouméa, New Caledonia

Vanuatu, an archipelago in the southwestern Pacific, is famous for its active volcanoes, both emergent and under the ocean’s surface. Largely because of its relatively modest terrestrial biodiversity, Vanuatu was recently included in the newly expanded East Melanesian Islands hotspot. The archipelago’s recent volcanic origin, the result of interaction between the subducting Australian plate and the Pacific plate, is often proposed as an explanation of its relative low level of distinctiveness. However, Vanuatu has also been poorly investigated, and some islands have never been explored; thus, it may provide surprises in the future. GEOGRAPHY OF THE TERRITORY

Vanuatu, officially the Republic of Vanuatu, is a Melanesian archipelago in the southwestern Pacific Ocean, comprising 83 islands and islets, between 13–21° S and 166–170° E (Fig. 1), with a total land area of ~12,220 km2

and an exclusive economic zone (EEZ) of 680,000 km2. The archipelago forms a Y shape, with the longest branch extending over 900 km, with a northwest–southeast orientation, and located some 1150 km east of the northeastern Australian coast. The base of the Y (Anatom Island) is only 220 km northeast of the Loyalty Islands (New Caledonia). Fourteen islands have surface areas exceeding 150 km2, with the largest island, Espiritu Santo, reaching ~3900 km2 and culminating in Vanuatu’s highest mountain, Tabwemasana, at 1879 m (Fig. 2). Several islands are quite high, and eight of them bear active volcanoes, such as Yasur on Tanna Island and Marum and Benbow on Ambrym Island. The archipelago’s climate is generally hot and humid, with oceanic characteristics, but there are some differences between the islands, ranging from hot, very humid, and with little seasonality in the north to warm, humid, and with well-marked wet and dry seasons in the south. The climate variation follows three gradients: latitudinal, from south toward the equator; altitudinal, from sea level to mountain summits; and longitudinal, from east to west. The latter gradient results from the effects of the trade

VA N U AT U

939

FIGURE 2 Tabwemasana Mountain range. Photograph courtesy of

Y. Pillon (IRD, Nouméa).

winds (“Alizés”) and their interaction with relief, creating a windward and a leeward side of each island. Mean annual temperature is greater than 23 °C; temperature varies less than 5 °C between seasons; average annual rainfall is greater than 2000 mm (from 2000 mm in the south to 4000 mm in the north); and usually there is no month with a rainfall deficit. Most of the soils in Vanuatu are derived from volcanic rocks, of both submarine and aerial origin, and from sediments derived from the latter. Some calcareous substrates of reef origin are also found. The oldest rocks occur in the western (Malekula, Espiritu Santo, Torres) and eastern arcs (Pentecost, Maewo), where some Oligocene and Miocene substrates occur, without recent volcanic forms. Some ultramafic rocks have been reported on Pentecost Island. Vanuatu’s population is modest in size with only 220,000 habitants for the entire country, although the annual rate of increase is about 2.8%. The capital, and largest town, Port-Vila, on Efate (Vate) Island, has nearby 40,000 inhabitants. BIOLOGICAL CHARACTERISTICS

Vanuatu’s flora was recently included in a broad East Melanesian Islands hotspot, which also encompass the Bismarck Archipelago, the Solomon Islands, and the Santa Cruz Islands (Temotu). The vegetation of Vanuatu can be divided into six main categories: (1) lowland rain forest, (2) montane cloud forest and related vegetation, (3) seasonal forest, scrub, and grassland, (4) vegetation on new volcanic surfaces, (5) coastal vegetation, including mangroves, and (6) secondary and cultivated woody vegetation. The first of these is subdivided into six variants, such as the remarkable Agathis–Calophyllum forest or mixed-species forests

940

VA N U A T U

lacking gymnosperms and Calophyllum. Montane cloud forests are typically dominated by species of the genera Metrosideros, Syzygium, Weinmannia, Geissois, Quintinia, and Ascarina. In humid summit areas, the trees are covered with epiphytes, mostly filmy ferns and liverworts, but also with epiphytic shrubs such as Vaccinium. The seasonal forest, scrub, and grassland can also be divided into three variants, including the Acacia spirorbis forest, locally referred to as “gaiac forest,” where sandalwood can grow; in drier locations, introduced shrubs are dominant, such as Leucaena leucocephala, Acacia farnesiana, and Psidium guajava. Intact forest landscapes would cover ~710 km2, or about 6% of the country. The Torres Islands appear to be particularly well preserved, without damage from logging and with few introduced species. Compared to its neighbors such New Caledonia or Fiji, the biological diversity of Vanuatu is modest, with 870 native vascular plant species in 534 genera, and local species endemism that is quite low (~17%); a single genus is endemic to Vanuatu (the monotypic palm Carpoxylon), and only four species of gymnosperms are recorded, of which one, the kauri (Agathis macrophylla), is an emergent forest tree that is imposing in appearance and provides a special physiognomy to vegetation where it occurs. With exception of bats (including flying foxes), no native mammals are present, although rats, mice, pigs, cows, and horses have been introduced. About 121 bird species have been recorded in Vanuatu, nine of which are endemic, and one, the buff-bellied monarch (Neolalage banksiana), belongs to a monotypic endemic genus; ten introduced bird species are known. Amphibians are represented by a single species, introduced during the 1970s, whereas reptiles are slightly more diverse, with only two snakes known (including just the Pacific boa as indigenous), but two families of lizards (Gekkonidae and Scincidae) being present and having several species each, including some endemics. A Fijian iguanid was also recently introduced to Efate Island only. Vanuatu remains poorly studied, with the exception of the work of a few individual researchers (such as Kajewski or Guillaumin), a large international botanical expedition led by the Royal Society of London in 1971, a Japanese botanical expedition on Espiritu Santo in 1996–1997, and more recently, a large multidisciplinary expedition conducted in 2006, focusing just on Espiritu Santo but studying both marine (lagoon) and terrestrial habitats. All of these endeavors yielded many discoveries in various groups such as plants (including large trees), molluscs, and arthropods.

BIOLOGICAL AFFINITIES

Although the closest territory to Vanuatu is New Caledonia, especially the Loyalty Islands, this is not correlated with the dominant biogeographic relations. Several studies of species richness in various plant groups, such as palms, figs, and ferns, have indicated closer relationships with the Solomons and Fiji. Molecular studies of kauri trees also show the species present in Vanuatu are more closely related to taxa in Fiji, Malaysia, and Australia than to species of New Caledonia. Exceptions do, however, exist, such as the genus Tinadendron (Rubiaceae), only known from calcareous substrates on New Caledonia and also on Erromango and Anatom Islands in Vanuatu, or Megastylis gigas (Orchidaceae), recorded only on the main island in New Caledonia and on Anatom. Nevertheless, as a whole, Vanuatu’s flora is considered to represent an extension of that of Malaysia, through Bougainville and the Solomons. Based on information from several zoological groups, the fauna of Vanuatu is typical of an oceanic island and does not require land bridges or former continents to explain its origin. Closer affinities are observed with Fiji than with New Caledonia, just as with plants, although there are also exceptions to this, such as the genus Emoia (scincid lizards), which has two species in New Caledonia, which are restricted to the Loyalty Islands and are presumed to be the result of an introduction from Vanuatu. Finally, the insect fauna is especially poorly known, and several papers dealing with biogeography of insects in the Pacific lack samples from Vanuatu. The first humans arrived in Vanuatu about 3300 years ago; they are identified with the culture called “Lapita” by the very distinctive pottery that they left behind them and that marks out their migration routes. Their impact was direct, by changing natural space for agricultural use, multiplying the incidence of fires, and introducing invasive species (voluntarily and accidentally). THREATS TO THE BIODIVERSITY

Logging appears to be an important threat to the forests of Vanuatu, with special pressure being particularly high on kauri (Agathis) and kohu (Intsia) species. Sandalwood (Santalum) has been also strongly used in the past and was cut for decades; it is still under exploitation today. Invasive species also represent a major threat to biodiversity, and invasives occur in every group. A notable invasive plant is the liana Merremia peltata, one of the worst pests of the world, which is said to have been introduced during the Second World War to hide military infrastructures (a story whose status as myth or reality is unknown), and now cover large areas of secondary forest, killing the remaining trees and preventing regeneration (Fig. 3). For

FIGURE 3 Slopes of the volcano of Vanua Lava Island (Banks), with

forest completely covered by introduced invasive liana. Photograph courtesy of Michel Lardy (IRD, Nouméa).

insects, the recent discovery (in 2006) of the tiny fire ant, Wasmannia auropunctata, in Luganville, is terrible news for the biodiversity in Santo Island, given the impact of this insect in New Caledonia, where it destroys the native entomofaune and has a heavy impact on reptiles as well. SEE ALSO THE FOLLOWING ARTICLES

Ants / Deforestation / Fiji, Biology / Invasion Biology / New Caledonia, Biology FURTHER READING

Bouchet, P., H. Le Guyader, and O. Pascal, eds. 2007. Santo2006 Expedition Progress Report. Santo2006/Gamma, Paris. Corner, E. J. H., and K. E. Lee. 1975. A discussion on the results of the 1971 Royal Society–Percy Sladen expedition to the New Hebrides. Philosophical Transactions of the Royal Society of London B 272: 267–486. [This special edition includes numerous review papers on the subject]. Iwashina, T., T. Hashimoto, and E. Bani, eds. 1998. Contributions to the flora of Vanuatu: scientific results of the botanical expedition to Vanuatu and adjacent countries in 1996 and 1997. Tsukuba Botanical Garden, National Science Museum, Tsukuba. Mueller-Dombois, D., and R. F. Fosberg. 1998. Vegetation of the tropical Pacific Islands. New York: Springer-Verlag. Quantin, P. 1992. Les sols de l’archipel volcanique des Nouvelles-Hébrides (Vanuatu). Editions de l’ORSTOM, Paris. Tardieu, V., and L. Barnéoud. 2007. Santo: les explorateurs de l’île-planète. Belin.

VEGETATION DIETER MUELLER-DOMBOIS University of Hawaii, Manoa

Vegetation is, simply, the plant cover of landscapes. It can be subdivided into plant communities on the basis of differences in structure (such as forest, shrubland, savanna,

V E G E TAT I O N

941

open forest, closed forest, summer deciduous forest, etc.), species dominance and composition (for example, eucalyptus forest, pine forest, softwood/hardwood forest, etc.). Vegetation is the most obvious and important biological component of terrestrial ecosystems and provides basic environmental services: Through its role as an absorber of carbon dioxide, an oxygen emitter, and a primary producer, it provides for life in terrestrial environments; it acts as an air filter or air conditioner in the Earth’s biosphere; and it filters, cleans, and regulates the flow of water in and on the soil. The forest recycles rainwater by evapotranspiration into the atmosphere from layers deeper in the soil than can be reached by evaporative power alone. The “wick action” of a forest can result in greater circulation of water from evapotranspiration per ground area than from an open water surface. VEGETATION CONCEPTS

Vegetation can be a natural plant cover, such as a rain forest. It can also be an artificial (i.e., under human control) plant cover, such as a field of sugar cane. Vegetation is also a hierarchical concept. There are such broad categories as island versus continental vegetation, and there are progressively narrower categories. For example, within island vegetation, there are biomes or ecosystems such as montane rain forests, lowland dry forests, freshwater swamp forests, mangroves, and seagrass beds (Fig. 1). Such biomes are also found in tropical continental areas and along their coasts. A major building block of vegetation is the flora of an area. The flora usually differs greatly among plant covers on islands versus on continents, and it differs also among island areas. Thus, although different islands may support

the same biome, such as a tropical rain forest, these rain forests differ from island area to island area in their structure and floristic composition. We may distinguish “dominance type” forests from “multi-species type” forests on the basis of structure and floristic composition of their canopies. Below these categories, we can usually recognize more narrowly defined community types by differences in forest undergrowth patterns of recurring species groups. VEGETATION DEVELOPMENT

Although the flora of a region is the major structural component of vegetation, it is not the only component. The flora is the broad matrix of species growing in an area. From this matrix, only certain species form assemblages at a specific geoposition or site. Floras are usually treated in taxonomic books or in checklists with descriptions, keys, diagrams, and pictures for identifying the species. Vegetation, in contrast, is built from a spectrum of environmental and biotic factors. These are summarized as a function of six factors with two overriding dimensions as follows: Vegetation development = ƒ (cl, g, d, fl, ac, e) where cl = climate (regional, local, and microclimate); g = geoposition (geographic position, geology, geomorphology, and ground = soil); d = disturbances including human interventions; fl = flora of the region, ac = access potential of a species to the geoposition in question; e = ecological/ evolutionary properties of species coming together in a vegetation or plant community (such as growth form and function, e.g., shade intolerance versus tolerance). The overriding dimensions of time and space determine the state of vegetation development after disturbance. In harsh geopositions, vegetation development

FIGURE 1 Typical volcanic high island. Reproduced with permission from Imamura and Towle (1987).

942

V E G E TAT I O N

takes much longer than in ameliorated sites, and large disturbed areas are generally much slower to recover in vegetation than are small disturbed areas. THE ISOLATION FACTOR WITH HABITAT RESTRICTIONS

The study of island ecology was highly stimulated by the island biogeography theory of MacArthur and Wilson. These authors developed a model on the natural assemblage of island biota based on island size and distance from biotic source areas (Fig. 2). The model makes two predictions. First, it predicts that the rate of invasion of species in islands near biotic source areas will be faster than in similarly sized islands further removed. Secondly, it predicts that the rate of species extinction will be greater on small islands than on larger ones.

habitat size and diversity. Certainly, smaller islands have more restricted habitats. With regard to species extinction through habitat loss, smaller islands can be interpreted as habitat-reduced larger islands. This also means that they offer smaller target areas for invasion. Generally, there is much greater species packing on islands than there is species extinction. This is particularly true because the isolation factor has been broken by introduction of species by humans. Naturally invaded and successfully established species prior to human interference are the native species, known as indigenous. In islands they are the progenitors of the other group of native species known as endemics. The isolation factor, prior to being broken by humans, had the distinct effect of facilitating the evolution of additional species, the endemics. FLORISTIC IMBALANCES

FIGURE 2 The island biogeography model of MacArthur and Wilson

A small archipelago, such as Palau, which is near a biotic source area (the Philippines), may have the same number of native plant species (~1000) as a large archipelago far removed from any biotic source area, such as the Hawaiian Islands. But among Palau’s native plant species, only 10% may be endemic, versus 90% in Hawai‘i. However, the fraction of endemics among an archipelago’s native species is not simply a function of isolation in terms of distance and time. It is also a function of the propensity among indigenous species to evolve into different species. For example, in the Hawaiian Islands, only 10% of the approximately 280 indigenous species gave rise to the native flora of about 1000 species. The isolation factor had a distinct effect on primary exclusion and secondary enrichment of taxa.

(1967), slightly modified. Reproduced with permission from Princeton University Press.

Although the first prediction is generally true, the second prediction assumes native species extinction to be related to the invasion of new native species at certain saturation points. These species saturation points are assumed to be lower in remote versus near-source islands. This is an unrealistic prognosis, because there is no known state of species saturation. Even if such were to exist, there is no reason to assume remote islands to have a lesser capacity for species diversity than islands near biotic source areas. Moreover, competition is not the main cause of species extinction. Species extinction is primarily due to loss of habitat. In this respect, smaller islands are more prone to species loss than larger islands when subjected to natural as well as human disturbances. This is a relationship of disturbance intensity, frequency, and size to

Primary Exclusion

Certain families of plants present on other tropical Pacific islands never reached Hawai‘i naturally. They include the terminalia family, the cunonia family, the mahogany family, the melastome family, and others. In contrast to the southwestern Pacific islands, Hawai‘i never received gymnosperms naturally. Likewise, mangrove stands, which form important shore stabilization communities in Melanesia and Micronesia, did not reach Hawai‘i naturally. Thus, there is an inherent primary impoverishment of taxonomic groups on oceanic islands. Secondary Enrichment

The initial impoverishment is taxonomically counterbalanced by a most remarkable secondary enrichment of species through endemism. In Hawai‘i for example, the pantropical African violet family has only one genus,

V E G E TAT I O N

943

Cyrtandra. But this single genus has proliferated into 51 endemic species. The cosmopolitan bellflower family, which is considered by some botanists to be the “crown jewel” of Hawai‘i, has five endemic genera. The bellflower genus Cyanea is the most prolific with 52 endemic species. The only palm genus in Hawai‘i, Pritchardia, has speciated into 19 endemic species. This is more than half of all species recognized in this palm genus. Because of this primary exclusion and secondary enrichment, island biotas have been characterized as “disharmonic” by some. Others have argued that island forests are not in balance because they lack certain important life forms, such as mammalian herbivores and their predators. No doubt, disharmony applies to island floras and biota in general when compared to continental floras. But does this floristic disharmony also result in disharmonic or imbalanced island vegetation? This question will be further elucidated in the following sections. THE BIODIVERSITY FACTOR WITH FUNCTIONAL SURPRISES

In spite of floristic disharmonies or imbalances, island forests are functionally sound. Healthy forests are found on all Pacific islands, where climate and soil are amenable for tree growth. In this respect, there is no difference between island and continental forests. Most island forests have also continuously renewed themselves in spite of their great distances from continental source areas. Limitations in self-maintenance or sustainability may have occurred in the initial stages of island forest development until trees arrived that were able to become established through generational turnover. Indications for inefficiencies in establishing new forests are present in Hawai‘i today. For example, the swamp mahogany (Eucalyptus robusta) from northeastern Australia has been planted in wet areas as replacement cover for native ‘o¯hi‘a lehua (Metrosideros polymorpha) forest; yet stands of swamp mahogany do not seem to be self-sustainable in spite of seemingly favorable environmental conditions. Similarly, fig tree species, whose seed was spread in abundance on stumps in the early twentieth century, largely failed to become established in the target areas. Instead, the alien paperbark tree (Melaleuca quinquenervia) established itself after planting and now is in the process of forming self-maintaining forests. The three species (E. robusta, M. qinquenervia, M. polymorpha) are members of the same family (Myrtaceae). Their tiny seeds are emitted from dry capsules and are distributed by wind. In spite of this similarity in seed size and seed dispersal mechanism, these three species have quite different eco-

944

V E G E TAT I O N

logical properties. This has become abundantly clear from an operational experiment to fix what has become known in Hawai‘i as the “Maui Forest Trouble.” A native Metrosideros rain forest on the lower windward slope of East Maui was noticed to deteriorate rapidly in the early twentieth century. Initially, this was thought to be the result of a new forest disease. Yet after a decade of research, no disease agent was found. Instead, the research conclusion was that the native Metrosideros rain forest could not persist because it was a pioneer vegetation unable to adapt to aging soils. This reasoning argued for the introduction of non-native trees to save the Hawaiian watersheds. The idea was put into practice in the Depression years of the 1930s. Plantations of Eucalyptus robusta and Melaleuca quinquenervia were established in half of the area with deteriorating native forest. Both these alien tree species grow well on this wet and swampy soil. Metrosideros is still present and reproducing on that site, but it grows only in the form of dwarf trees or shrubs. Melaleuca quinquenervia trees, on the other hand, are now invading this area in large numbers (Fig. 3).

FIGURE 3 Photograph of alien paperbark trees invading the “Maui

Forest Trouble” area. Photograph by Dieter Mueller-Dombois, 2006.

This example shows that the native rain forest has floristic-functional and structural limitations: The tree-to-shrub size reduction in the recovering Metrosideros population is an adaptation of this species to bog formation. Indeed, in more advanced bogs, Metrosideros has formed new dwarf varieties or ecotypes. There are no native trees available to grow tall on hydromorphic or swampy soils, except perhaps the hala tree (Pandanus tectorius). But this tree lacks an efficient seed disperser in terrestrial environments, as do many other native tree species, which can be interpreted as another limitation in biodiversity function. The process of forest deterioration on East Maui was not simply a response to soil aging but also a response

to a complete landscape change from normally drained wet rain forest soil to incipient bog and stream formation. On that long time scale of geomorphologic aging, involving breakdown of the volcanic shield, we now see little justification for introducing alien trees to slow down the natural process of bog formation. Bogs may be just as efficient in watershed protection as are forests. THE SIMPLIFICATION FACTOR WITH ADDED COMPLEXITIES

Metrosideros polymorpha still rules today as the dominant tree species in the wet rain forests from the youngest island (Hawai‘i) to the oldest high island (Kaua‘i). This can be attributed to a floristic-functional simplification. Vitousek made great use of this simplification by emphasizing Hawai‘i as a model for ecosystem research. In most primary chronosequences associated with soil aging, a change in dominant forest trees is expected in continental environments. The simplification in structure and species composition of forests as evolved in isolation has eluded their functional interpretation in the past. This is best shown by the example of forest dieback and succession as researched in Hawai‘i.

more or less synchronously like a stand of planted trees. If not cut, burned, or otherwise destroyed, the cohort stand will eventually reach senescence and break down in the form of canopy dieback. Synchrony is inherent in such a stand because the trees are mostly of the same generation and are similarly stressed through habitat constraints and old age. A tropical storm can be the dieback trigger in an aging or otherwise stressed cohort forest when such a storm has removed a large proportion of canopy foliage. Senescing stands lack the reserves of non-structural starch to replace their canopy. Biotic agents then can hasten tree death, with the result that new openings are created that allow for stand rejuvenation. A new cycle of cohort stand development may follow because canopy dieback itself is a major disturbance in an otherwise evergreen forest. A model developed for the theory of cohort senescence with subsequent rejuvenation is shown in Fig. 4.

Forest Dieback

During the early 1970s a major decline of native Metrosideros rain forest in the form of canopy dieback was discovered on the island of Hawai‘i. As in the “Maui Forest Trouble,” the general assumption of research foresters was that a new disease was killing the forest, and the same prognosis was made that the native Hawaiian rain forest was doomed. After a decade of intensive disease and insect pest research, no biotic disease agent could be established as causing the dieback. Causal research was continued for environmental stresses in soil, climate variability, and extreme weather events. All of these factors were found to be involved in the canopy dieback, but the answer for the principal cause remained elusive until the early 1980s. Eventually, the principal cause was found in the Metrosideros forest itself, in its population structure and landscape mosaic. Individual stands of Metrosideros forest display a simplified uniform cohort structure. There is a canopy tree cohort and a cohort of small seedlings present in mature stands. Saplings are lacking or are very scarce. This indicates a common generational origin. Metrosideros stands become reestablished as generational or cohort stands after catastrophic events that destroy an existing forest (such as volcanic explosions) or after a clearing of the former canopy (such as resulting from canopy dieback). Once a new cohort stand is formed, it develops to maturity

FIGURE 4 (A) The cohort stand model; (B) stand rejuvenation upon

canopy dieback (Mueller-Dombois 1987). Reprinted with permission from BioScience 37, by American Institute of Biological Science. DBH = diameter at breast height.

V E G E TAT I O N

945

Canopy dieback that fits the cohort senescence theory has been found on other islands, such as New Zealand, New Guinea, Lord Howe Island, Norfolk Island, and the Galápagos. Cohort senescence, rather than single-tree senescence, can be considered a consequence of structural and functional simplification in island forests. Permanent plot research has given evidence of stand recovery after dieback, with Metrosideros being the leading species again after 30 years in most cases. Invasion of new species can change this pattern.

phase may not be surprising, but in the continental tropics, tree diversity evolved in such a way that even soils with low nutrient contents support relatively high biomass. Speciation in the island flora has rarely resulted in successional replacers of the dominants. Most endemic trees other than the leading dominants are distributed spatially in different locations, often separated by topographic barriers that prevent efficient gene flow. Some of them are associated with the leading trees, but they do not take over as dominants.

Succession

Invasive Species, an Added Complexity

Another related simplification is the peculiar succession of island forests. Many islands lack successional trees. Forest turnover is generally a simple process known as auto-succession or direct succession, whereby the same species resume dominance after some initial recovery with less tall species such as grasses and shrubs. Auto-succession differs from what has been termed obligatory or normal succession, where different tree species dominate one another in succession. The latter is typical in continental temperate environments. The succession model of Hawaiian rain forest (Fig. 5) illustrates the concept of auto-succession along a long-term chronosequence with dieback-induced secondary successions. After the soil fertility peak, the height, diameter, and biomass of the leading canopy tree, Metrosideros polymorpha, declines with decreasing soil fertility. This regression

The virtual absence of native successional species provides resources for two types of tree life forms not found among the native trees. First, there are the quick responders, the species with r-type strategies, which include many of the cosmopolitan second-growth species that make up the secondary forests in the continental tropics. Competitive replacement occurs when alien species overtop the native shade-intolerant pioneer trees. A second group of successional invaders providing threats of competitive displacement are slow-growing shade-tolerant trees that grow taller than the native forest trees, the so-called K strategists. Because they usually require specific animal seed dispersers, they have not yet been generally observed as threats in native Hawaiian rain forests. But some of them have been dispersed by humans, thereby causing displacement of native tree species. FUTURE OUTLOOK

FIGURE 5 The Hawaiian rain forest succession model with five Metro-

sideros dieback types on the two main volcanic substrates. Checkmark-like symbols indicate episodic stand breakdowns with secondary successions (Mueller-Dombois 1986). Displacement dieback refers to suppression of Metrosideros regeneration from competition by other plants, such as tree ferns (Cibotium spp.); gap formation dieback refers to small groups of trees dying in the form of patches; bog formation dieback is a form of stand reduction relating to landscape change through geomorphic aging; dryland dieback is related to welldrained substrates, wetland dieback to poorly drained and firm substrates. Reprinted with permission from the Annual Review of Ecology and Systematics 17 © 1986 by Annual Review www.annualreviews.org.

946

V E G E TAT I O N

The demise of the native Hawaiian rain forest has been predicted twice in earlier scientific publications, first by Lyon (1918) and thereafter by Petteys et al. (1975). In both cases the argument was that the native Metrosideros rain forest could not sustain itself naturally. The same argument could be made today, but for another reason. Invasive species have become the modern threat. There is now a new, unstoppable, natural dynamic on account of human-introduced invasive species. Because this problem is caused by human interference, it also now requires human interference to steer the problem into a more amenable direction. It is therefore necessary to apply silvicultural measures based on proper ecological knowledge. This means that the new dynamic trajectories are carefully assessed, and control measures are taken that divert them into more desirable directions. This requires willingness to compromise, because one cannot expect original conditions to be maintained, given that vegetation and plant communities are living systems that change with changing environmental constraints. A more general problem in island vegetation that applies to all Pacific islands is loss of indigenous forests. As already discussed, this is the major cause of native

species extinction. Measures to counteract it will come only through ecological research, education, and training for mutual capacity building in cooperation with the Pacific islanders. Mutual capacity building is needed to approach traditional island cultures with respect and for bridging indigenous knowledge together with modern scientific understanding of the dynamics of local ecosystems and landscapes. To be successful, such approaches must also be integrated with conservation management and decision making for environmental policy. SEE ALSO THE FOLLOWING ARTICLES

Hawaiian Islands, Biology / Island Biogeography, Theory of / Mangrove Islands / Succession / Sustainability FURTHER READING

Cuddihy, L. W., and C. P. Stone. 1990. Alteration of native Hawaiian Vegetation: effects of humans, their activities and introductions. Honolulu: University of Hawaii Press. Holt, R. A. 1983. The Maui Forest Trouble: a literature review and proposal for research. Hawaii Botanical Science Paper No. 42. www.botany .hawaii.edu/pabitra Huettl, R. F., and D. Mueller-Dombois, eds. 1993. Forest decline in the Atlantic and Pacific regions. New York: Springer-Verlag. Imamura, C. K., and E. Towle. 1987. Integrated renewable resource management in U.S. insular areas (Island study 1987). Paper commissioned for the U.S. Office of Technology (OTA), Pacific Basin Council Research Institute, Honolulu. Lyon, H. L. 1918. The forests of Hawaii. Hawaiian Planter’s Record 20: 276–281. MacArthur, R., and E. O. Wilson. 1967. The theory of island biogeography. Princeton, NJ: Princeton University Press. Mueller-Dombois, D. 1987. Natural dieback in forests. BioScience 37.8: 575–583. Mueller-Dombois, D. 2006. Long-term rain forest succession and landscape change in Hawai‘i: The “Maui Forest Trouble” revisited. Journal of Vegetation Science 17: 685–692. Mueller-Dombois, D., and F. R. Fosberg. 1998. Vegetation of the tropical Pacific Islands. NY: Springer-Verlag. Petteys, E. Q. P., R. E. Burgan, and R. E. Nelson. 1975. Ohia forest decline: its spread and severity in Hawaii. USDA Forest Service Research Paper PSW-105. Pacific SW Forest and Range Experiment Station, Berkeley, California. Vitousek, P. 2004. Nutrient cycling and limitation: Hawai‘i as a model system. Princeton, NJ: Princeton University Press.

VICARIANCE MICHAEL HEADS Ngaio, Wellington, New Zealand

In plants and animals, closely related species and groups of species often occur in different areas, and this pattern is called vicariance. Its origin can be explained by a pro-

cess, also termed vicariance, in which a widespread common ancestor differentiates and breaks up, more or less in situ, into related descendants. For example, consider a variety of plant or animal that is found only on (i.e., is endemic to) a certain island. It may have its closest relative on a nearby mainland. Isolation and differentiation of the island population from the mainland one could have arisen by vicariance if the island and its life were separated from the mainland by a geological process, such as erosion or subsidence. In an alternative explanation, dispersal theory, the island is colonized after it forms by long distance, overwater dispersal of an individual or seed from the mainland. Colonization then stops, and the new population differentiates to become an island endemic. The well-known “equilibrium theory” of island biogeography is based on dispersal theory and suggests that immigration from a mainland center of origin and extinction together determine biogeography. DISPERSAL AND VICARIANCE

In dispersal theory, a related group, or taxon, evolves in a restricted area—its “center of origin”—and attains its distribution by dispersing out from there. Each taxon comes into existence on its own, not with others. In contrast, during vicariance of a widespread ancestor, the population of each different sector evolves into a new taxon where it is, often over a wide area. There is no center of origin and no “dispersal,” although subsequent range expansion or contraction may take place. In dispersal theory, a “founder” population is established through normal dispersal, but at some point migration stops, and the founder becomes isolated from its parent population. It has never been explained why exactly dispersal would stop, but this is a critical question. In dispersal theory the end of dispersal is attributed to chance rather than to geological or climatic change, as in vicariance. Normal, “ecological” dispersal, the physical movement of plants and animals, is observed every day. It does not involve the differentiation of new taxa, unlike long-distance, chance dispersal, which is a theory of speciation. All organisms move, and given the amount of geological time available, they should be able to migrate to areas of suitable habitat around the world. However, in fact, most groups show marked local and regional endemism. This creates an apparent paradox: The movement that exists everywhere does not seem related to the most fundamental aspect of distribution. In addition, distribution patterns are usually shared by many taxa with different ecology and means of dispersal. These observations conflict with the dispersal theory of speciation and biogeography and have led to

VICARIANCE

947

ongoing debate about biogeographic processes. Normal “ecological” dispersal and vicariance are accepted by all workers. The center of origin/founder dispersal mode of speciation is much more controversial. DATING TAXA

Attempts have been made to distinguish between origin by vicariance or by dispersal through dating taxa, because dispersal theory generally proposes a geologically younger age for a taxon than does vicariance. There are difficulties with this approach, however. Fossils provide only minimum ages for groups, and most groups are much older than their oldest known fossil. Fossil-based molecular clock calculations also give minimum ages for taxa, but these are often misrepresented as maximum dates and used to rule out earlier vicariance events as irrelevant. Simplistic correlations with paleogeography, such as calibrating all Pacific-Atlantic divergence with the rise of the Isthmus of Panama at 3.5 million years ago, can also give ages for taxa that are much too young. Dating studies usually assume that evolutionary differentiation is more or less continuous over time and so roughly clock-like. Degree of differentiation would then be more or less proportional to the time since divergence. However, evolution probably proceeds in distinct phases with long periods of stasis, and so taxa may show little or no differentiation despite having been separated for many millions of years. CENTERS OF ORIGIN

The center of origin for a particular group has often been deduced from the arrangement of taxa in a cladogram or phylogenetic tree. However, the “basal” branch or clade is not a primitive, ancestral group occurring at the center of origin—just a small sister-group, and no more primitive, just less diverse, than the main clade. A series of vicariant taxa that branch off sequentially in a cladogram represents a geographic sequence of differentiation in a widespread ancestor, not a series of chance dispersal events. VICARIANCE IN THE PACIFIC The Dispersal Model

Dispersal and vicariance models of Pacific biogeography were vigorously debated by the Victorian naturalists, but after the First World War, dispersal theory became almost universally accepted. According to the theory, all Pacific taxa derive from ancestors that originated in Asia or America and migrated into the Pacific. The dispersalists recognized the existence of prior land in the Pacific, as they knew that many volcanic islands had sunk there,

948

VICARIANCE

each leaving only an atoll as a trace. Nevertheless, they interpreted distributions among the islands of the Pacific as the result of dispersal among extant, rather than former, islands. For example, the distinctive Pacific clade of Cyrtandra shrubs is endemic to rain forest on islands between the Carolines and southeastern Polynesia. The very wide range has been taken as proof that the plants are highly vagile and have a remarkable capacity for long-distance dispersal. However, this proposal overlooks the many single-island endemic species in the group, the clear-cut vicariance between the group and its relatives (which are absent from the central Pacific), and the volcanism that has been widespread and continuous in the Pacific basin since its formation. Volcanism and Metapopulations

Dispersal theory for islands involves random volcanism, a center of origin, and long distance dispersal, whereas vicariance emphasizes recurrent volcanism, normal migration among unstable local populations any of which may go extinct, and regional persistence of taxa despite changing geography. The dispersal model assumes that the Pacific region was originally devoid of islands and island life, but this is very unlikely. Volcanism does not occur at random, but takes place around particular sectors over periods that are much longer than the age of the individual volcanoes. Oceanic volcanic islands form at subduction zones, spreading centers, hotspots, and propagated fissures, which have always been active in and around the Pacific. The individual islands are relatively short-lived, but new islands are constantly being formed in the vicinity. These new islands are colonized by “ordinary dispersal” from nearby islands, not by “long-distance dispersal,” and there is no speciation involved. The taxon originates and survives in the region as a population of populations, a metapopulation, and whether or not the current islands have ever been joined to a mainland or not (the distinction between “oceanic” and “continental” islands) is not relevant for their biogeography. Populations on volcanic islands and atolls, and their reefs, survive and evolve in the same way that they do on any other recurring habitat “islands,” such as termite mounds or forest gaps. Establishing the age of an individual volcanic island is not straightforward and involves more than just dating the exposed strata. An island may be composed of very young volcanics or limestone but could be very old as an island, if new material is added as old material is removed by subsidence, erosion, or burial, which is usually the case. In practice, the age of an island’s rocks is not nearly as important for biogeographic analysis as the age and

history of the associated subduction zone, fissure, or hotspot that has been generating the volcanism. In Hawaii and the Galápagos, for example, biologists have suggested that endemic taxa are much older than the rocks that currently form the islands. The populations have survived in the region by constantly dispersing from older, now largely eroded islands to nearby younger islands. In one model, island chains form by plate movement over hotspots, and so archipelagoes may eventually join with others. This process might explain the Hawaii–southeastern Polynesia connections seen in many marine and terrestrial taxa. However, some geologists now suggest that linear island chains are formed not by hotspots but by propagated fissures resulting from plate tectonics processes. The fissures, like subduction zones, may be much older than the individual islands. Vicariance Model

Recently, several authors have discussed a model for the Pacific which assumes that plants and animals have always occurred there. The fact that the central Pacific is a large, well-marked center of endemism that includes smaller areas of endemism is not well accounted for in traditional biogeographic theory. If the Pacific had been populated by dispersal from a western center of origin, a simple dropping out of individual species across the Pacific from west to east might be predicted. But regional areas of endemism occur throughout the Pacific and often involve parts of one archipelago and parts of another. Across the Pacific, there is a west-to-east drop in total diversity of terrestrial and shallow-water marine species, and this is sometimes cited as evidence for dispersal. However, the decline is simply due to the islands of Indonesia and Melanesia being larger than the islands of Polynesia. The rain forest trees of Metrosideros (Myrtaceae) are a typical example of central Pacific endemism. The distributions of the five main clades are mostly vicariant (Fig. 1), with overlap only in Vanua Levu, Fiji (although clade 3 is only known there from one mountain), and North Island, New Zealand (where clade 1 has a very restricted range). Each of the Metrosideros clades occupies an area of endemism also held by many other, very different taxa. Southeastern Polynesia, an important biogeographic sector that is often overlooked, is illustrated here by the land snail Tubuaia (Fig. 1). Metrosideros has close relatives endemic to New Guinea, the Philippines, South Africa, and Chile, indicating that it originated by vicariance, as the central Pacific representative in a South Africa–Pacific–Chile group. This group, in turn, has vicariant relatives in Australia.

FIGURE 1 Distribution of Metrosideros (Myrtaceae), endemic to the

central Pacific. The five clades (1, 2a, 2b, 2c, and 3) are mainly vicariant. Distribution of the land snail genus Tubuaia (Achatinellidae) is also shown (stippled). A = Auckland Is., AUS = Australia, B = Bonin Is., C = Campbell I., F = Fiji, H = Hawaii, M = Marquesas Is., NZ = New Zealand, Rp = Rapa I. (Austral or Tubuai Is.), Rt = Rarotonga (Cook Is.), Sa = Samoa, K = Kermadec Is., L = Lord Howe I., NI = New Ireland, P = Pitcairn I. (Tuamotu ridge), So = Solomon Is., T = Tahiti (Society Is.), V = Vanuatu.

Atolls, Extinction, and Survival

The atoll zone of the central Pacific comprises a vast area from the Carolines to southeastern Polynesia—an area that has undergone subsidence and massive extinction. Land snails characteristic of high, forested islands occur as fossils on the atolls of the Marshall Islands and Midway Island, west of Hawaii. These atolls are the remains of former high islands, which subsided at least 1500 m over the Cenozoic. As the high islands collapsed, the peaks no longer caught the mists, and this wiped out the moisture-dependent snails and other forest taxa, perhaps including Metrosideros. On some Pacific islands, relic species in groups with a high survival coefficient, such as flies and certain birds, still survive on the young limestone covering the sunken volcanics. PLATE TECTONICS

Rifting of the Earth’s crust is a well-known mode of vicariance. Geologists have proposed that the Solomons, Vanuatu, Fiji, and Tonga once formed a continuous island arc, which was rifted apart into separate chains, and zoologists at the Smithsonian Institution have interpreted disjunct distributions in reef fishes and terrestrial lizards as a direct result of this. A geological terrane is a fault-bounded block of the Earth’s crust that has had its own independent history.

VICARIANCE

949

Accretion, or the docking and fusion of terranes at convergent plate margins, can also modify and create vicariance patterns. New Guinea, New Caledonia, and New Zealand are larger than the other Pacific islands and include older, continental-type rock. They are often described as “fragments of Gondwana” but are in fact geological and biological composites (like the Greater Antilles in the Caribbean). Each comprises an older, Gondwanic part plus many other terranes, including island arcs, which have accreted to the Gondwana fragments after arriving from the Pacific side with the encroaching Pacific plate. The regions of accreted terranes, like others in Indonesia and the Philippines, show high diversity in many groups and indicate evolution by juxtaposition rather than by radiation from a center of origin. Geological and biological accretion has also occurred in the archipelagoes of Indonesia and the Philippines, and accreted island arc complexes now form vast areas of western North, Central, and South America. In the central Pacific, very large igneous plateaus formed in the Cretaceous and then moved south, west, and east, colliding with countries like the Solomons, New Zealand, and Colombia. The plateaus are now largely submarine but include many formerly emergent seamounts up to 24-km across, as well as sediment layers with fossil wood. In summary, the plants and animals of the Pacific islands are probably the result not of Neogene founder dispersal from Asia or America but of long-term survival and evolution in the Pacific basin since its Jurassic origin. Despite enormous extinction, metapopulations in different parts of the Pacific have preserved patterns of endemism and vicariance that reflect plate tectonic rifting and convergence. OTHER ISLANDS

Outside the Pacific, fossil-calibrated dating studies have inferred dispersal from mainland Africa to Madagascar, but fossils of many groups are scarce there and Mesozoic–Early Cenozoic vicariance better explains the close biogeographic relationship among eastern Africa, the southwestern Indian Ocean islands, and India/Sri Lanka. The Galápagos–West Indies connection was discussed in some early vicariance analyses and illustrates the predictive power of the method. Although molecular dating is generally misleading, molecular cladogram topologies are more reliable and are revealing a vast amount of previously hidden vicariance in all large clades. Recent studies show that Galápagos finches (Geospiza, etc.) and Galápagos mockingbirds (Nesomimus) each have their sister group in the Caribbean, not

950

VOLCANIC ISLANDS

on nearby mainland South America. Some geologists now trace the origin of the Caribbean plate to the Gálapagos hotspot, and this would provide an explanation for the distribution pattern. SEE ALSO THE FOLLOWING ARTICLES

Convergence / Dispersal / Endemism / Island Biogeography, Theory of / Metapopulations / Plate Tectonics FURTHER READING

Arbogast, B. S., S. V. Drovetski, R. L. Curry, P. T. Boag, G. Suetin, P. R. Grant, B. R. Grant, and D. J. Anderson. 2006. The origin and diversification of Galapagos mockingbirds. Evolution 60: 370–382. Craw, R. C., J. R. Grehan, and M. J. Heads. 1999. Panbiogeography: tracking the history of life. New York: Oxford University Press. Fitton, J. G., J. J. Mahoney, P. J. Wallace, A. D. Saunders, eds. 2004. Origin and evolution of the Ontong Java plateau. Geological Society of London, Special Publication 229: 1–369. Foulger, G. R., and D. M. Jurdy, eds. 2007. Plates, plumes, and planetary processes. Geological Society of America, Special Paper 430: 1–998. Heads, M. 2006. Seed plants of Fiji: an ecological analysis. Biological Journal of the Linnean Society 89: 407–431.

VOLCANIC ISLANDS JOHN M. SINTON University of Hawaii, Honolulu

In the broadest sense, volcanic islands include all islands that form by volcanic processes, even those islands that represent ancient volcanic terranes that have been tectonically exposed, such as Gorgona Island offshore of Colombia or Macquarie Island in the southwestern Pacific. It also should be noted that most atolls represent a calcareous carapace surmounting oceanic volcanoes, the volcanic part of which has subsided below sea level, a concept originally proposed by Charles Darwin and subsequently confirmed by drilling through shallow carbonate atolls into the underlying volcanoes of the Hawaiian and Tuamotu island chains. However, attention here will be restricted to those islands primarily formed by volcanic processes more or less in situ, and where portions of the original volcano are still exposed above sea level. This restriction still leaves a subject that encompasses hundreds of islands and island groups scattered around the world’s oceans. ISLAND ARCS AND OCEANIC ISLANDS

Volcanic islands can be broadly divided into two principal categories, reflecting fundamentally different geo-

logic processes responsible for the volcanism. One class of islands forms in regions of plate convergence above subduction zones, whereas the second class is associated with anomalous upwelling zones of upper mantle, not associated with subduction zones. Differences in mantle sources and melting processes in these two environments result in contrasting chemical and mineralogical compositions of the volcanic rocks that form. The recognition of two categories of oceanic islands predates modern concepts of plate tectonics. For example, in 1912 Patrick Marshall distinguished the high-SiO2 basalts, andesites, and rhyolites of islands of the easternmost South Pacific from the dominantly basaltic lavas of the Pacific basin lying to the west of what he called the Andesite Line. The development of the plate tectonic concept more than 50 years later provided a framework for the reinterpretation of the Andesite Line to coincide with subduction zones associated with plate convergence in the western Pacific. The rocks that make up the two main classes of volcanic islands are now generally known as island arc volcanics (IAV) and ocean island basalts (OIB). IAV are dominated by SiO2-rich lavas, whereas OIB span a considerable range in silica and alkali contents, although generally basaltic compositions are by far the most common. MELTING PROCESSES

The geologic processes responsible for the formation of volcanic islands depend on their tectonic setting. Melting beneath island arcs is initiated in the mantle wedge overlying subducted lithosphere as a consequence of the addition of water and other volatiles derived by dehydration of the downgoing plate. This melting is commonly referred to as flux melting, because the addition of water reduces the melting temperature of the overlying mantle in this region. Mantle melting in this environment produces basaltic magmas that carry the imprint of relatively shallow, hydrous melting that is manifest in generally higher silica and volatile contents compared to basaltic magmas elsewhere. In addition, the common presence of very high-silica magmas (dacites and rhyolites) in some arc islands is generally ascribed to direct melting of the lower crust, which can be heated above the melting point when voluminous basaltic magma ponds above the crust– mantle interface. The formation of oceanic islands away from subduction zones signifies the presence of melting anomalies in the ambient mantle. In this case, melting is a direct consequence of mantle upwelling, commonly referred to as decompression melting. There is considerable debate about the origin of these melting anomalies or hotspots,

mainly centered around the relative contributions from thermal and compositional anomalies in the underlying mantle. Several oceanic island provinces appear to be associated with thermal upwelling from the deep mantle that can contain the chemical signature of recycled oceanic lithosphere. Other island provinces may be associated with chemical heterogeneities in the mantle that require neither an origin in the deeper mantle nor unusually high mantle potential temperatures. Although the presence of melting anomalies is independent of plate boundary processes, some melting anomalies coincide with divergent plate boundaries or mid-ocean ridges, most notably in Iceland, the Azores, and the Galápagos. This coincidence probably reflects the weakening of the lithosphere in the presence of melting anomalies, such that spreading ridges tend to localize around them, rather than the other way around. Whether or not melting anomalies are spatially fixed with respect to one another is another issue of considerable debate. For example, the geometry of some linear island chains have been used to define the motion of lithospheric plates over the deeper mantle. Such “absolute” plate motion models, using a relatively fixed hotspot reference frame, have been used successfully to predict the locations and azimuths of linear island provinces and other major bathymetric anomalies of the ocean basins, such as the Hawaiian–Emperor chain (Hawaiian hotspot), the Iceland–Faeroe Ridge (Iceland), the Walvis Ridge and Rio Grande Rise (Tristan da Cunha hotspot), and the Carnegie and Cocos ridges (Galápagos hotspot). However, the fixity of deep-mantle melting anomalies has been challenged by numerical modeling constraints on the stability of thermal plumes in convecting mantle and also by paleomagnetic studies of island and seamount lavas that suggest significant variation in latitude of formation. The evidence for crustal melting and the presence of rhyolitic magmas in ocean islands is much more limited that in island arcs, although notable examples have been reported from Iceland, Easter Island (Rapa Nui), Ascension Island, and Pitcairn. Quartz-normative lavas with SiO2 contents greater than 55 wt% are extremely rare in the Hawaiian Islands, with the notable exception of the Wai‘anae volcano on O‘ahu. ERUPTIVE ACTIVITY AND VOLCANIC PRODUCTS

The nature of volcanic activity is directly related to the composition of magmas in different environments. The generally high volatile content of arc magmas promotes relatively explosive eruptions in that environment. The

VOLCANIC ISLANDS

951

solubility of volatiles is inversely related to pressure so that as volatile-rich magmas ascend beneath arc volcanoes, volatile components exsolve as gases, which can lead to large increases in pressure under confining conduit conditions. High conduit pressures promote explosive eruptions and lava fragmentation, two characteristics that commonly distinguish arc volcanic processes from those in oceanic islands. Although there are many exceptions to this generality, typical island arc volcanoes contain a much higher proportion of tephra (ash and other ejecta) to lava, compared to typical ocean island volcanoes. In contrast, most ocean island volcanic eruptions are relatively quiet, generally effusive, and dominated by lava rather than tephra, although again there are notable exceptions. Some of these exceptions can be related to the interaction of shallow magma with meteoric water in the

FIGURE 1 Perspective views of selected ocean islands (top) and arc

islands (bottom); note different scales. Oceanic islands tend to be much larger than typical arc islands. Hawai‘i Island comprises five separate shield volcanoes; Pagan Island comprises two separate volcanoes. Lava flows on Hawai‘i emanate from along linear rift zones, radiating away from the summits, whereas Fernandina lacks prominent rift zones, and lava flows mainly emanate from near the summit region. The caldera of Fernandina is several times larger in diameter and deeper than that on Mauna Loa. The diameter of the caldera of Anatahan is more than 50% of the entire island.

952

VOLCANIC ISLANDS

edifice substructure, but others may be related to high primary magma volatile contents, such as some of the eruptions in the Azores and Iceland provinces. VOLCANO STRUCTURES

Volcanoes that produce significant amounts of tephra tend to build composite volcanoes with relatively steep slopes and an internal stratification consisting of mixed layers of lava and tephra; such volcanoes are sometimes referred to as stratovolcanoes. In contrast, volcanic islands dominated by lava eruptions tend to have much gentler slopes and can generally be described as shield volcanoes. Arc islands are almost exclusively composite volcanoes, whereas oceanic islands tend to be dominated by shield volcanoes. Individual islands can be composed of single volcanoes or multiple, coalesced volcanoes in both environments (Fig. 1). However, the number of coalesced volcanoes in arc islands rarely exceeds two, whereas single oceanic islands can comprise five or more individual volcanoes, such as the Island of Hawai‘i in the Hawaiian archipelago or Isabella in the Galápagos archipelago. Iceland is by far the largest and most complex volcanic island, consisting of tens of volcanic systems and hundreds of individual volcanoes. The largest individual volcano on Earth is Mauna Loa on the island of Hawai‘i, with an estimated total volume of ~40,000 km3. Although this volcano clearly is atypical, it is certainly true that most ocean island volcanoes tend to be significantly larger that most arc volcanoes. The diameters of ocean island volcanoes commonly exceed 10 km, whereas most arc volcanoes have diameters less than 5 km. Although most ocean island volcanoes can generally be described as shield volcanoes, there can be considerable variation in the morphology and structure of them, particularly with respect to the development of rift zones, the overall shape, and the nature of calderas (Fig. 1). Hawaiian volcanoes commonly develop rift zones that radiate away from a summit region characterized by the presence of a caldera during the main growth phase of activity (see below). Eruptions can occur either from the summit region or from along the rift zone. Rift zones of Hawaiian volcanoes can be more than 50-km long, in some cases with a submarine extension that is longer than the subaerial rift. The Hawaiian example appears to be exceptional, however, as many other oceanic island volcanoes are much more circular in plan form, and evidence for highly elongated rift zones is much less common. For example, most eruptions on Fernandina volcano in the Galápagos archipelago occur from circumferential fissures near the summit region, whereas eruptions on Mauna Loa can occur far down the

rift zones away from the summit (Fig. 1). Highly elongated rift zones are rare in arc volcanoes, which mainly have a nearly conical shape with eruptions that are strongly concentrated near the central summit region. Summit depressions, or calderas, characterize most volcanic islands while they are active. However, there can be considerable variation in the dimensions of calderas and probably also in the processes that are responsible for their formation. In arc volcanoes, cataclysmic explosive eruptions can result in calderas with dimensions up to ~80% of the total volcano diameter. Perhaps the most well-known such eruption is the AD 1883 eruption of the Indonesian volcano Krakatau. The 1883 eruption, however, was only the latest giant eruption of this region, occurring within a 7-km-wide caldera that had formed during a previous large event around AD 416. Such cataclysmic processes are much rarer in ocean island volcanoes. Calderas that characterize the summit regions of active island volcanoes tend to be smaller features formed by collapse or down-sagging above relatively shallow, quasi-steady-state magma reservoirs within the volcanic edifice. Even within ocean island volcanoes, dramatic differences in size, shape, and depth of calderas can be present, as exemplified by the differences in Galápagos and Hawaiian volcanoes (Figs. 1, 2). ISLAND GROWTH

The initiation of volcanic islands begins below sea level. Deep submarine volcanism mainly produces pillow lavas where water pressures are sufficient to inhibit lava fragmentation. Below ~1000 m, H2O, which is by far the most abundant volatile component of most magmas, remains dissolved in the liquid magma. At lower pressures and shallower depths of eruption, volatile exsolution increases the amount of fragmentation, and most volcanic islands probably have variable amounts of hyaloclastite (literally, broken glass) within the volcanic edifice. As the volcano emerges from the sea, violent explosions are characteristic, but once the magmatic conduits are insulated from interaction with seawater, the nature of volcanic eruptions becomes progressively less explosive and more effusive. Hawaiian volcanoes are known to proceed through a series of volcanic stages that reflect variations in the composition of magma and the nature of volcanic eruptions. Subaerial Hawaiian volcanoes are dominated by a shield stage of evolution, characterized by the eruption of basalts relatively low in alkali elements. This stage can last for a million years or more. Many Hawaiian volcanoes are known to have evolved into a later stage of activity, characterized by less frequent eruption of lavas that form by lower extents of melting of the underlying mantle, and by evolution in

FIGURE 2 Vertical profiles of the islands shown in Fig. 1. Although the

two classes of islands are shown at different scales, both are displayed with 3D vertical exaggeration. It is apparent that ocean island shield volcanoes tend to be much larger and have lower slopes than do typical arc volcanoes.

much deeper magma chambers within the volcanic edifice. This later stage is called the postshield stage of volcanism. In most Hawaiian volcanoes, this stage is known to last approximately 200,000 years, although there are examples where it was much longer lived, and yet others in which this stage of activity is entirely lacking. The details of other oceanic island provinces are less well known than in Hawai‘i, but the concepts of shield and postshield volcanic stages have been extended to the geological evolution of several volcanoes in the Samoan, Society, Marquesas, and Austral island chains, although the characteristic variations in lava chemistry are not always identical to those in Hawai‘i. One of the most intriguing and least understood aspects of many oceanic islands is the presence of rejuvenation volcanism (i.e., volcanic activity that recurs after a period of quiescence lasting up to a million years or more). During the quiescent period, erosion, subsidence, and reef growth dominate the evolution of the island. Although rejuvenation volcanism was first recognized in the Hawaiian Islands, geological and geochronological investigations have confirmed that it also characterizes some Samoan and Marquesan islands, as well as most of those in the CookAustral chain, and the island of Wallis, north of Fiji. ISLAND EVOLUTION BY EROSION, MASS WASTING, AND SUBSIDENCE

Important processes that affect the morphology of volcanic islands include those that modify the original landscapes. Among these are erosion, mass wasting, and subsidence. Erosion occurs throughout the history of a VOLCANIC ISLANDS

953

FIGURE 3 Variation in elevation along the Hawaiian–Emperor volca-

nic chain, illustrating the importance of subsidence in island evolution. The effect of subsidence is shown schematically by the dashed line. The uniform elevation of volcanoes near sea level ~700–2500 km from Kilauea indicates the region of prominent reef growth. Most islands in this province have subsided at least 2 km since formation. Figure modified from Moore (1987).

volcanic island, but it has the greatest effect at times of relative volcanic inactivity. Evidence for erosion of oceanic volcanoes mainly takes the form of stream valleys incised into the volcanic flanks and sea cliffs cut into the volcano during relative high stands of the sea. One of the most important processes affecting island morphology is mass wasting, which can occur at a variety of scales. Mass wasting in the form of rock falls, landslides, and debris avalanches is the principal process by which valleys widen. However, the growth of volcanic edifices surrounded by the sea produces edifices that are inherently unstable. Large-scale flank collapses can have a dramatic effect on volcanic islands. Evidence from the Hawaiian chain indicates that these large collapse events can occur at any stage of volcanic activity, although they are most likely to occur when the volcano is actively growing and the edifice is least stable. Mass wasting of volcanic flanks of island arc volcanoes is less well known than in ocean island volcanoes but almost certainly also occurs in such environments. High-resolution mapping of the submarine flanks of some arc volcanic islands shows large sedimentary aprons that probably represent debris flows associated with mass wasting events. Several historical tsunamis originating in various island arcs are thought to be related to submarine landslides in those regions. The important role of subsidence in affecting island topography was fully appreciated only after the advent of

954

V O YA G E O F T H E B E A G L E

detailed studies of the submarine flanks and deep drilling of some islands. These studies documented the transition from subaerial to submarine lava sequences and other evidence for ancient shorelines lying up to several kilometers below present sea level. Studies of submerged shorelines around oceanic islands have contributed to this understanding by determining the rates of subsidence on some islands. It is now clear that the dominant reason why the Hawaiian Islands have become gradually lower in elevation with age has more to do with subsidence than with erosion (Fig. 3). Islands subside because the underlying lithosphere is not rigid but behaves somewhat elastically. Thus, subsidence is inevitable following loading during volcanic growth. However, the resistance of the underlying lithosphere to flexure and subsidence depends on its age and on the amount of reheating that has taken place during volcanic growth. In general, subsidence should be much greater where ocean islands are forming on young lithosphere, such as the Galápagos Islands and Iceland, than for those forming on older and stronger oceanic lithosphere. SEE ALSO THE FOLLOWING ARTICLES

Darwin and Geologic History / Hawaiian Islands, Geology / Island Formation / Krakatau / Lava and Ash / Oceanic Islands FURTHER READING

Clague, D. A., and G. B. Dalrymple. 1987. The Hawaiian Emperor volcanic chain, part I. Geologic evolution. U. S. Geological Survey Professional Paper 1350: 5–54. Darwin, C. 1837. On certain areas of elevation and subsidence in the Pacific and Indian Oceans, as deduced from the study of coral formations. Proceedings of the Geological Society of London 2: 552–554. Marshall, P. 1912. The structural boundary of the Pacific basin. Australasian Association for the Advancement of Science 13: 90–99. Moore, J. G. 1987. Subsidence of the Hawaiian ridge. U.S. Geological Survey Professional Paper 1350: 85–100. Morgan, W. J. 1972. Plate motions and deep mantle convection. Geological Society of America Memoir 132: 7–22. Volcano World website. http://volcano.und.edu/.

VOYAGE OF THE BEAGLE JERE H. LIPPS University of California, Berkeley

HMS Beagle, with Capt. Robert FitzRoy and young Charles Darwin aboard, left the island of Britain for hydrographic surveying, particularly in the Southern

Hemisphere, on December 27, 1831, on a voyage spanning nearly five years, to October 2, 1836. Over 40 islands were visited or closely approached, including the Galápagos Islands, which Darwin later made famous. Darwin observed and recorded information about the geology, biology, and people of the many places he visited, which was then used in his later books and papers. CAPTAIN FITZROY, HMS BEAGLE, AND CHARLES DARWIN

Robert FitzRoy had been an officer on the earlier voyage of the Beagle (1828–1831) to southern South America, and he was made captain of the ship after Captain Pringle Stokes committed suicide. Continuing the hydrographic survey of Tierra del Fuego, FitzRoy held four Fuegian natives hostage in exchange for a boat that had been stolen and then decided to take them back to England, and he made anomalous and confusing magnetic measurements, which he attributed to the possible presence of magnetic minerals in the mountains nearby. He determined then that he should return to Tierra del Fuego to return the Fuegians and to take a geologist with him on his next voyage to investigate the magnetism problem. In mid-1831, the British Admiralty agreed to his request to return the Fuegians, but also provided him with voluminous instructions on additional surveying in South America and the acquisition of navigational, magnetic, and astronomical measurements at meridians around the world. Beagle was recommissioned on July 4, 1831, and FitzRoy began planning, including a search for a naturalist to accompany him.

Charles Darwin had the good fortune to do geological fieldwork during the summer of 1831 with Reverend Adam Sedgwick, his professor of geology at Cambridge University. Chiefly for this reason, he was chosen as naturalist on board HMS Beagle. After responding to an inquiry from the Reverend John Henslow, Darwin was interviewed by FitzRoy. At this time Darwin thought of himself more as a geologist than a biologist. Indeed, although he often called himself a “naturalist”; the only time he ever referred to himself as something else, he chose “geologist.” Thus Darwin was invited aboard Beagle by FitzRoy as an intellectual companion as well as a naturalist. His father initially opposed the idea but later relented. FitzRoy received the approval of the Lords of the Admiralty to include Darwin on the voyage. FitzRoy’s mission and needs dictated where Darwin would go and what he would see on the voyage of the Beagle. He would take good advantage of this arrangement. Darwin, like many people, was fascinated by islands and even had plans to visit the Canary Islands before he was offered the position on Beagle. Islands attracted Darwin because they are isolated, often contain different or unusual biotas, are generally simple in ecological structure, and can be neatly circumscribed and thus appear to be more readily understood than the contiguous geology and biology of much larger continents. So Darwin looked forward to the voyage with great anticipation. ISLANDS VISITED BY THE BEAGLE

Beagle left England from Devonport (Fig. 1, no. 1) and headed south through the Atlantic Ocean (Fig. 1) with

FIGURE 1 Route of the HMS Beagle, December 27, 1831, to October 2, 1836, showing the major islands visited by the ship and Darwin. Points visited,

in the order of visit, were (1) Britain; (2) Canary Islands; (3) Cape Verde Islands; (4) St. Paul’s Rocks; (5) Fernando Noronha Island; (6) Tierra del Fuego, Straits of Magellan, and Beagle Channel; (7) East Falkland Island; (8) Chonos Archipelago; (9) Chiloé; (10) Galápagos Islands; (11) Tahiti; (12) Bay of Islands, New Zealand; (13) Sydney, Australia; (14) Hobart, Tasmania; (15) King George’s Sound, Western Australia; (16) Cocos (Keeling) Atoll; (17) Mauritius; (18) St. Helena; (19) Ascension Island; (3) Cape Verde Islands again; (20) Azores.

V O YA G E O F T H E B E A G L E

955

FitzRoy intending to put in at various islands to make detailed measurements for the Admiralty. After ten miserable days at sea, during which time Beagle passed Madeira and Piton Rock from a distance and searched in vain for the Eight Stones, Darwin was excited to approach Tenerife in the Canary Islands (Fig. 1, no. 2). He had read Humboldt’s description of the island and its biota, and made much about the book and its descriptions. He thus was anxious to see Humboldt’s island for himself. But Beagle had been blocked from making that visit by port authorities, although Darwin could clearly see the brightly colored houses of Santa Cruz as Beagle moved to within a half mile of the island. Beagle had been quarantined because the authorities feared she might harbor cholera, and so the ship would have to lie at anchor for 12 days to demonstrate she was free of the disease. FitzRoy would have none of that, however, and set sail immediately south along Africa for the Cape Verde Islands, passing through the Canary Islands close to Gran Canaria. Darwin could see from the ship the volcanic nature of the islands. Modern geology considers these islands to be hotspot volcanoes, although not with certainty. Although greatly disappointed by not visiting Tenerife, Darwin was still excited by his imagination of what that island and the disappearing Eight Stones islands held. Heading south along the African coast, Darwin was disappointed by the Cape Verde Islands and St. Jago (now Santiago) Island especially, also now best considered hotspot volcanoes, because they were so desolate and miserable. On St. Jago and adjacent tiny Quail Island, he began to examine and compare the volcanic rocks to those at home, the larger corals living there with the small solitary forms near Edinburgh, and the various sea levels he could recognize on the islands. He recalled his geological instruction from Sedgwick and began to understand the geology of these islands. The tropical plants impressed him, and he devoted much space in his notes to the people living there. After three weeks in the Cape Verde group (Fig. 1, no. 3), Beagle sailed southwesterly to the middle of the Atlantic Ocean nearly at the equator and found tiny St. Paul’s Rocks (Fig. 1, no. 4). These rocks, not large and not high, were uninhabited except for a huge number of birds, as they are today, and Darwin’s stay lasted only a few hours. FitzRoy described his crew throwing stones at the birds to kill them for food, and noted that Darwin successfully used his geological pick in a similar fashion. Later Darwin used it appropriately to collect samples of green and black rock types; these are now known to be pieces of an intrusion from deep in the earth’s mantle

956

V O YA G E O F T H E B E A G L E

on the Mid-Atlantic Ridge upon which St. Paul’s Rocks lie. Although Darwin did not know this origin then, his description of the rock types was accurate. He also was impressed by the simplicity of the biota—two birds (a booby and a noddy) plus associated insects. Beagle continued southwesterly and put in for a day at another small island in Fernando de Noronha group (Fig. 1, no. 5), about 370 km off the Brazilian coast. These islands were volcanic as well, forested and contained an array of insects, all noted by Darwin in just a few hours. After stopping for two weeks at Bahia on mainland Brazil, FitzRoy took the ship to the Abrolhos Islands lying in shallow waters 32 km off the coast to commence lead-line bathymetric surveys. Here Darwin observed a “saurian” under every rock, vivid green vegetation consisting of just a few species, and many spiders and rats. He also observed the relationship of these volcanic islands to their coral development and recalled the ones he had previously seen. From these islands, Beagle continued south to the tip of South America, landing along the way at various places on the east coast. At last, on December 16, 1832, the ship was off Tierra del Fuego and the crew could see fires on the shore lit by the people there. The southern part of South America includes a large number of islands, many of different sizes, and all pieces of continental South America isolated during the past glaciations and sea level rise. Darwin spent more time in these islands (Fig. 1, no. 6) than anywhere else on the voyage, for Beagle and FitzRoy had particular aims to accomplish. FitzRoy was determined to return the three Fuegians (one had died in England) he had taken on his previous voyage to their homelands and of course to finish surveying. Foul weather interfered with these objectives. Nevertheless Beagle or its whaleboats and yawl sailed among this archipelago for 75 days (Fig. 2), putting ashore in various places. The area, of course, is huge,

FIGURE 2 HMS Beagle at Tierra del Fuego (painted by Conrad Mar-

tens). From Leakey (1979).

and Darwin could see only glimpses of any of it, and he had little time to study the people, geology and biology of any particular place. In Beagle’s smaller boats, FitzRoy, Darwin and members of the crew traversed the length of the Beagle Channel, named by FitzRoy on his previous journey to this region, twice, once in each direction. Since Darwin’s actual views of the region were incomplete, he once again drew general conclusions, but they have remained more or less valid up to the modern times. He observed that the western and southern parts of the group were made largely of igneous and metamorphic rock with trends like those of the Andes and that the northeastern part had sedimentary rocks identical to those in Patagonia. Darwin was piecing together a much bigger picture than he had observed. His biological observations, focused on birds, insects, marine invertebrates, and the Nothofagus forests, dealt with a broader range of conclusions, including some about the behaviors and associations of the animals and of the plants. His observations of the penguins and ostriches (rheas) led him to make comparations of the life styles of the two birds that he retained even in the Origin of Species. He was not impressed with the people, whom he characterized as “miserable creatures, stunted in their growth, their hideous faces daubed with white paint & quite naked”; and later he would refer them to the lowest rung of all the peoples he met on Beagle’s voyage, a rather characteristic Eurocentric view at the time, but one no longer accepted. Upon reaching the Beagle, they sailed northward to the Falkland Islands and Patagonia for surveying work and replenishment and repair of the ship. The Falkland Islands, in the southwestern Atlantic east of Tierra del Fuego (Fig. 1, no. 7), were visited twice, once from December 1832 to February 1833 and again from March to April 1834. In between these times, the ship returned to the east coast of South America to continue surveying and to sail south down the Straits of Magellan before turning back to exit the Strait, and then to explore the eastern and southern (again) coast of Tierra del Fuego. Both times at the Falklands, Beagle spent her time in Berkeley Sound anchored near the English settlement of Port Luis on the eastern extremity of East Falkland Island. For 10 weeks, Darwin was able to study a relatively small area longer than any other place on the voyage, while FitzRoy was compelled to deal with shipwrecked French sailors, American whalers’ destruction of British property, sovereignty issues about the islands, and unrest among the inhabitants resulting in mistrust and murder, as well as his chief mission of surveying the area. Although not happy there, Darwin kept busy documenting the geology,

marine life and domesticated animals of Berkeley Sound. He made detailed geologic notes with cross sections of the folded strata he encountered, wondered about the “stone runs” of angular rocks, ranging from a few centimeters to meters in size, that formed long, narrow lineaments in the valleys, and noted that fossils he found suggested a different environment in much earlier times. Darwin was confronted by evidence of catastrophe while still holding his gradualistic, Lyellian views of geology. He collected more animals and plants, marine and terrestrial, continuing his comparative approaches by noting that the extremely abundant kelp forests were similar to those of Tierra del Fuego and to the tropical rain forests in abundance he had seen earlier in South America. On leaving the Falklands in April 1834, more than halfway through her voyage, Beagle made for and then passed through the Straits of Magellan to the myriad of islands just offshore of southwestern South America. Beagle struggled north along the islands, passing the Archipélago de los Chonos, which she would later visit in the next summer. As winter was upon it, Beagle’s passage was rough, and all on board were relieved when the ship landed first on June 28, 1834, at San Carlos on the north end of Chiloé Island (Fig. 1, no. 9), southwest of Puerto Montt, Chile. FitzRoy stayed only two weeks at San Carlos, while Darwin walked inland and along the coast. Then the ship headed north to Valparaiso to present papers and gain permission from authorities in Santiago to survey coastal Chile, as well as for resupply, refitting, and surveying in the region. At Valparaiso, Darwin climbed the Andes and found fossil plants and clams. Returning to Chiloé for November and December 1834, Darwin, along with the ship’s crew, explored the island. During January 1835, Beagle went further south to survey the Chonos Archipelago, before going back in February and March to Chiloé for a third time to complete surveying there. At Chonos, Darwin collected barnacles, perhaps part of the incentive to monograph them later. Darwin spent much of time on Chiloé commiserating with the crew about the overcast and depressing weather, although he considered Chiloé a “fine island” in terms of its geology and temperate rain forests. On the western islands, Darwin saw evidence of sea level changes he attributed to the rise of the land and of active volcanism in the Andes that could be seen from the islands. Using the small boats and horses, Darwin explored Chiloé and the nature of the biota and human populations. His impressions from Chiloé lasted the rest of his life, perhaps more so than those from the Galápagos, and he incorporated them into his books and correspondence repeatedly.

V O YA G E O F T H E B E A G L E

957

After finishing his surveys of the intricate coastal areas, FitzRoy headed north along Chile and Peru. On September 15, 1835, the ship began survey work in the Galápagos Islands (Fig. 1, no. 10) for five weeks. Darwin, however, was mainly restricted to Beagle as it moved among the islands, and he spent only 19 days or parts of a day ashore on only four of the islands (Fig. 3) although he passed closely by eight others. Indeed he did not like the Galápagos because they were too hot and barren, nor did he achieve any great insights at the time of his visit. He was awed by the evidence of the very recent volcanic activity, and wondered about the giant tortoises and their future given the intense predation on them by ships’ crews, including that of Beagle. Although Darwin was impressed by the volcanism and low diversity of the fauna and flora, he still made large collections that served him and others well later. These included examples of the finches and other birds, tortoise shells, molluscs, insects, fish, plants, and rocks. Many of these were studied by others, such as John Gould, who later corresponded with Darwin about their significance.

FIGURE 3 Darwin’s view of the Galápagos Islands. Darwin saw the

Galápagos as a series of young volcanoes on which unusual varieties of organisms had later developed; among them, the land iguana, which he called “ugly animals” with a “singularly stupid appearance” but that “when cooked, yield a white meat.” Image taken by J. H. Lipps on South Plazas Island, Galápagos Islands, 1982.

From the Galápagos, Beagle sailed directly to the Society Islands, passing by the eastern low islands and finally anchoring in Matavai Bay, Tahiti (Fig. 1, no. 11), on September 15. The sequence of islands that Darwin saw from the ship intrigued him because he was formulating his own theory of reef formation. FitzRoy set about making magnetic and navigational measurements at Point Venus, where Cook had done so in 1769; this gave Darwin time to climb the adjacent mountain high enough (above 800 m) to see clearly the nearby island of Moorea, then known as Eimeo. He saw this view of Moorea like a picture, the reef being the frame, the smooth lagoon the mat, and the island itself the picture (Fig. 4). This

958

V O YA G E O F T H E B E A G L E

FIGURE 4 Darwin saw Moorea as a picture in a frame from Tahiti. The

barrier reef with waves crashing on it formed the frame, the rather small and narrow lagoon was the mat around the picture, and the island itself was the picture. Image of the reef, lagoon, and island taken by J. H. Lipps at Viare, Moorea, 2002.

view, together with his observations of the other Society Islands, came together to confirm his ideas, first formulated off South America, about the formation of atolls. He carefully studied the details of the reefs of Tahiti while wading and canoeing over them. He determined that the corals grew best in the upper 35 m of the ocean and not as well below. The seaward edge of the reefs, he discovered, fell away sharply into much deeper water. His descriptions again were perfect, and his conclusions well drawn. He placed every coral island, whether low or high, that he encountered into his own grand picture of the process of fringing, barrier, and atoll reef formation. So sure of his new idea, he wrote a manuscript proposing it while on board Beagle in the month after he left Tahiti. Darwin also noted that the volcanic rocks of Tahiti were well weathered, and from that observation he concluded that the island had been active at some time in the distant past, unlike the rocks at the Galápagos. He was impressed by the Tahitians, who he found happy and generally quite religious, although he said little about the lack of dress of the women. After just 11 days, Beagle left Tahiti bound for New Zealand, where FitzRoy would take more measurements at the Bay of Islands (Fig. 1, no. 12). As the voyage extended well beyond the two years it was planned for, FitzRoy and the crew began to move more quickly and do less extensive surveying work. Thus, Beagle found anchorage on the northern tip of North Island in the Bay of Islands for only seven days. This was the only place Darwin visited in New Zealand, and he was not impressed because of the accumulation of whalers and seamen from other places. He was interested in the Maoris and what he regarded as their lowly life style, adding yet another rung to his ladder of human advancement. The missionaries, he thought, were doing fine work among them by raising their expectations, however.

The ship then proceeded to Australia, arriving 12 days later for restocking of food and supplies. Darwin would spend a total of 37 days at Port Jackson (now Sydney), New South Wales; Hobart Town, Tasmania; and King George’s Sound, Western Australia (Fig. 1, no. 13, 14, 15). At Port Jackson, Darwin undertook an excursion inland. He became very much aware of the dissimilarity of the fauna and flora of Australia relative to the previous places he had visited. He remarked on the nature of eucalyptus forests, and on marsupials and the platypus. As impressed by those differences as he was, he did find similarity between an ant lion he watched and those of England. His biogeography was developing into a tool for understanding evolution, although he was a long time from enunciating it. Even in 1836, Darwin recorded that fire was a prevalent feature of the landscape. He also noted the major geological features of the area. Leaving Port Jackson 18 days after arriving, Beagle set sail to Tasmania. There Darwin engaged in his usual studies of geology around Hobart, also making observations on the native people, climbing Mt. Wellington to view the spectacular large eucalyptus trees, and collecting organisms. Again he dwelt on the geology, in particular his discovery of some “Devonian or late Carboniferous” fossils, and the evidence of possible former sea levels and uplift of the land. He compared the local geology to that he saw in New South Wales, using his comparative approach to try to understand the rock types and sequences he described. He was also concerned about the conflict between the Aboriginals and the Europeans, concluding that the native population would become extinct unless they were removed to an area remote from the Europeans. After 10 days, Beagle left Hobart for Port Williams on Australia’s southwestern coast. He felt that his stay at Port Williams was the most “dull and uninteresting time” of the entire voyage. Yet Darwin expressed interest in the sparse plants, such as the grass tree, and he was fascinated by smooth domes of granite penetrated by numerous veins. A group of Aborigines called the “White Cockatoo men” visited and were persuaded to put on a dance that evening. Darwin described the dancing, often imitating emus or kangaroos, in negative terms: “all moving in hideous harmony, formed a perfect display of a festival amongst the lowest barbarians.” Overall, Darwin was not impressed, writing, “he who thinks with me will never wish to walk again in so uninviting a country.” Indeed, he did not find Australia in general appealing and thought the idea of being served by convicts repulsive. “Farewell, Australia! You are a rising child, and doubtless some day will reign a great princess in the South: but you are too

great and ambitious for affection, yet not great enough for respect. I leave your shores without sorrow or regret.” Although Darwin had finished his essay on the formation of atolls after leaving Tahiti, he had never actually set foot on one. Cocos Keeling Islands, lying nearly 2000 km northwest of Perth, Australia, had originally been suggested to FitzRoy as a landing place by the Admiralty to determine their exact position. However, Beagle was now far to the south off southeastern Australia, and FitzRoy wrote a letter to the Admiralty that he would take the ship directly across the Indian Ocean, then to England. Darwin, however, may have influenced FitzRoy to visit the islands, as originally suggested, because of his own interest in atolls. In any case, FitzRoy set sails for those islands, much to Darwin’s delight. The Cocos Keeling Islands (Fig. 1, no. 16) consist of two atolls with many small islets on the coral rims. The Beagle anchored off the southernmost of the two where the only population of Europeans and Malays lived. Darwin visited at least five of the islands, examining the vegetation, beaches with pumice and plant debris tossed up by waves, the spiders, insects, and large coconut crabs (Birgus latro), the lagoon with its variety of marine animals including many giant clams (Tridacna), and the communities of people. He collected much from the rather depauperate flora and fauna, and he thought he had representatives of every plant (20) save two, which existed as single trees whose seed were likely tossed ashore by the waves. The first tree to occupy newly formed land in the atolls, he noted, was Pemphis acidula. Darwin was probably satisfied that his atoll theory seemed reasonable in view of what he saw, and he might have wondered how far below his feet was the volcano that supported this atoll. He was quite happy about his visit and seemed to enjoy his stay. Continuing the homeward trip, Beagle next anchored at Port Louis on the northern side of Mauritius (Fig. 1, no. 17). Although Darwin spent much time walking about town and watching the society at work, he also examined the relationship of the coral reefs to the volcanic rocks of the island, the apparent recent elevation of the land, and the nature of volcanic structures observed in the interior; he was able to place this island into his theory of reefs through comparisons to the other islands he had visited. Darwin did not collect much, but he did find a frog, which puzzled him since amphibians were nonexistent on islands, presumably because they could not cross saltwater. Later, the frog was discovered to have been introduced. From Mauritius, FitzRoy steered around Cape Horn into the Atlantic with stops at St. Helena, Ascension, and,

V O YA G E O F T H E B E A G L E

959

finally, Terceira in the Azores (Fig. 1, nos. 18–20). The first two are hotspot volcanoes close to the Mid-Atlantic Ridge of basaltic composition, younger than 15 million years, and well eroded by the sea. Darwin thought these islands and St. Jago, which Beagle had visited early in its voyage, had rims raised by volcanic forces that surrounded the interiors of the islands. Each was experiencing volcanism, uplift, and denudation eroding them lower. On Ascension, he saw that the lava flows were far fresher than on St. Helena, concluding that it was also far younger. In fact, later geologic studies have confirmed this. The Beagle arrived at Terceira, after it had revisited Bahia in Brazil, to complete FitzRoy’s measurements on the meridians, on September 19. Darwin seemed less than enthusiastic about his short stay, but in spite of that he made three excursions on the island, one to see fumaroles. From these, he concluded that the lava flowed from the central part of the island outwards toward the shores. On October 2, 1836, after nearly five years, the Beagle, FitzRoy, and Darwin arrived at Falmouth on their home island. Darwin never again left the British Isles, the last islands he would ever see and the islands of his birth, until his death April 19, 1882. WHAT DARWIN LEARNED ON THE VOYAGE OF THE BEAGLE

Darwin’s scientific observations came from a vast experience on the islands and two continents (South America and Australia) on the voyage and provided him with a strong grasp on both his geological and biological hypotheses. He developed his comparative research style, and focused more on geology of these places than on their biology. Overall, he absorbed much about both fields and integrated them into his thinking. Islands contributed a great deal to his books, articles, and correspondence. He saw volcanic islands, continental islands, very isolated islands, tropical and subpolar islands, and those in close proximity to their neighbors. These differences in environment, Darwin realized, accounted for the islands’ differing biotas. Darwin’s interest in geology comes through in his notes and theory development while on the Beagle. He wrote 1383 pages of geological notes, only 368 pages of zoological ones, and none on botany. He developed the theory of atoll formation as he sailed, noting that the coral rings could easily form by subsidence of the volcano and growth of the corals. He explained each stage from corals growing closely together and against the shore of an island to barrier reefs and finally atolls. His theory, widely adopted by geologists and biologists everywhere, did not include a

960

V O YA G E O F T H E B E A G L E

mechanism for subsidence. Darwin then surmised, based on his observations of submergence of islands in the middle of oceans, the Pacific in particular, and of continents such as South America rising, that when some part of the Earth’s surface submerges, another must rise to balance it. Only since the 1960s and the development of plate tectonics did a proper mechanism became available for the submergence of volcanoes as they are carried away from hotspots of rising magma by the moving plates on which they lie. Darwin, of course, had no idea of this, but his theory of atolls remains valid. Darwin wrote of conservation through his observations on many of the islands he visited. He was concerned, for example, that the harvesting of Galápagos tortoises by ships’ crews for food might lead to their extinction. Likewise, in the far more complex situation of the Tasmanian Aborigines, whose conflicts with the European population Darwin well understood from his other island observations, led him to suggest that the only way to preserve them as a people was to remove them from Tasmania and place them in a locale far from Europeans. Without this move, he predicted that they would go extinct as a pure race, which indeed they did less than 40 years after his statement. Darwin was also concerned that the introduction of alien species to islands caused destruction and extinction of native species. At the Bay of Islands on New Zealand, he admired the many fine English plants that grew around the houses, but he saw that many weeds and the common black rat had also been introduced. These he regarded as a threat to the native species, especially the rat, which Darwin knew had already eliminated the Polynesian rat. At St. Helena, he noted the same situation—a large number of introduced plants crowding out the natives. There he also noted that the introduced goats had demolished the trees and much of the other vegetation, and that this resulted in a cascade of extinction of a number of species of land snails and probably impacted the insects too. On Ascension, he saw many feral cats, which he condemned as a plague on the land. He even took note of the impact by humans through removal of the trees on the formerly well-forested island. While recording his observations, Darwin commonly referred to “struggles” between various natural elements. This concept developed early in the voyage and he returned to it often. He referred to the struggle of native species to resist those recently introduced, of corals to resist waves, of one species in competition with another, and of waves against the shore. Darwin had seen enough, chiefly on islands, during the voyage of the Beagle to write

many books and articles that changed biology and geology forever. SEE ALSO THE FOLLOWING ARTICLES

Atolls / Cape Verde Islands / French Polynesia, Geology / Galápagos Islands, Geology / St. Helena / Tasmania / Tierra del Fuego / Tortoises FURTHER READING

Armstrong, P. 2004. Darwin’s other islands. London: Continuum. Barlow, N., ed. 1958. The autobiography of Charles Darwin, 1809–1882, with original omissions restored. London: Collins. Browne, J. 1996. Charles Darwin voyaging. Princeton, NJ: Princeton University Press. Darwin, C. 1842. The geology of the voyage of the Beagle. Part 1. The structure and distribution of coral reefs. London: Smith, Elder and Co.

Darwin, C. 1844. The geology of the voyage of the Beagle. Part 2. Geological observations on volcanic islands. London: Smith, Elder and Co. Darwin, C. 1846. The geology of the voyage of the Beagle. Part 3. Geological observations on South America. London: Smith, Elder and Co. Darwin, C. 1859. On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. London: John Murray. Darwin, C. 1899. The descent of man and selection in relation to sex, 2nd ed. London, United Kingdom: John Murray. FitzRoy, R. 1839. Narrative of the surveying voyages of His Majesty’s Ships Adventure and Beagle between the years 1826 and 1836, describing their examination of the southern shores of South America, and the Beagle’s circumnavigation of the globe. Proceedings of the second expedition, 1831–36, under the command of Captain Robert Fitz-Roy, R.N. London: Henry Colburn. Herbert, S. 2005. Charles Darwin, geologist. Ithaca, NY: Cornell University Press. Leakey, R., abr., illus. 1979. The illustrated Origin of Species by Charles Darwin. London: Faber and Faber.

V O YA G E O F T H E B E A G L E

961

W WALLACE, ALFRED RUSSEL ELIN CLARIDGE University of California, Berkeley

Alfred Russel Wallace (1823–1913) is best known as the co-author of the theory of evolution by natural selection, with Charles Darwin. However, he was also an influential British scientist and social thinker of his time. As a young man, he traveled extensively in the Amazon River basin, and then in the Malay Archipelago, working as a professional collector, naturalist, and explorer. Besides his contributions to the development of evolutionary theory, Wallace is considered a founding father of the field of biogeography. He published articles and books covering a wide array of topics in both the natural and social sciences. WALLACE’S THEORY OF EVOLUTION

Wallace developed his theory of evolution by natural selection independently of Darwin. In his autobiography, Wallace recounts that the idea of natural selection came to him as he was lying in bed delirious with a malarial fever, pondering Malthus’s ideas about checks on populations. At the time he was traveling in the Malay Archipelago (present-day Malaysia and Indonesia). After recovering from his fever, he wrote a short essay outlining his ideas and sent a copy to Darwin, an esteemed colleague, without realizing that Darwin himself had been developing these same ideas over 20 years prior to Wallace’s discovery but had not yet published on the subject because of its radical implications. Wallace’s essay was an unpleasant 962

shock for Darwin but provided the impetus he needed to start writing up his own research. WALLACE’S EARLY LIFE

Wallace was an unconventional and somewhat controversial scientist with broad interests. Unlike most of the gentleman scientists of his time, he came from a modest background and although he received some formal schooling, he was primarily self-educated. He was also a radical socialist who championed unpopular causes, without consideration of the impact it might have on his reputation or career. He was one of the first scientists to raise concerns about the environmental impacts of human activities. He was highly critical of the economic and social injustices that he saw in nineteenth century Britain. His advocacy of spiritualism and his belief that natural selection could not explain the origin of consciousness or human intellect put him in direct conflict with many of the other proponents of evolutionary theory. Wallace was born in Wales in the village of Llanbadoc, near Usk in Monmouthshire. He was the eighth of nine children. Wallace’s father had a law degree but never actually practiced law because he had inherited some property that generated sufficient income to support the family; however, he, like his son, had a knack for making bad investments, and a series of failed business ventures led the family into financial decline. The Wallace family moved to Hertford, north of London, when Alfred was five years old, and there he attended Hertford Grammar School until, at the age of 13, financial difficulties meant that he had to withdraw. He initially moved to London to work with his brother John, who was an apprentice builder. In 1837, he left London to work for his older brother William, who was a surveyor. Over the following years, he became an experienced

surveyor working primarily in the west of England and Wales. Much of the surveying work that they undertook was associated with the Tithe Act and General Enclosure Act, which modified land rights, in restricting the small farmers’ rights to graze the common lands. Wallace was outraged by the social injustice. He considered that these Acts were nothing more than the persecution of poor farmers by rich landowners and lawmakers. This sense of social injustice never left him; he was a socialist at heart, who believed that everyone had a fundamental right to share the world’s resources. When in 1843 his brother was unable to find sufficient work for them both, Wallace left the surveying business and found a position as a schoolmaster at the Collegiate School in Leicester, teaching technical drawing and surveying. In Leicester, he continued to read widely and frequented the library, where he read Malthus’s pivotal work on populations. He also met the young entomologist Henry Walter Bates. They became good friends, and Wallace was inspired by Bates to start collecting insects. They shared a common passion for natural history and insect collecting, and they exchanged frequent written correspondence, including lists of insects they had collected and discussions of their own scientific ideas and the most recent literature. Wallace also became interested in mesmerism (hypnotism) and phrenology while he was in Leicester. He even discovered that he was able to mesmerize some of his students. In 1845 when his brother William died, Wallace left his teaching position in order to deal with his brother’s business affairs in Neath. He was unable to re-launch the business, but instead found work as a civil engineer, working on a survey for a proposed railway line in the Vale of Neath. The railway line was never built, but Wallace’s surveying work, hiking through the remote Welsh countryside, fed his passion for insect collecting and natural history. In early 1846 Wallace was able to persuade his brother John to join him in starting another architecture and civil engineering firm in Neath. They rented a small cottage, and Wallace’s mother and sister came to live with them. Their firm carried out a number of projects including designing a building for the Mechanics Institute of Neath. William Jevons, the founder of that institute, was a family friend and persuaded Wallace to lecture on science there for two winters. During this period Wallace was still reading avidly and exchanging letters with Bates; in particular, they were discussing the anonymous evolutionary treatise Vestiges of the Natural History of Creation, which propounded the radical idea of species transmu-

tation; Charles Darwin’s journal and remarks from his adventures on the Beagle; and Charles Lyell’s Principles of Geology. TRAVEL AND EXPLORATION

When Bates came to visit Wallace in Wales in 1847, inspired by the travel chronicles of earlier naturalists, they hatched a plan to embark on their own big adventures. Having read William Edwards’s “Voyage Up the Amazon,” they decided that they too would travel to South America, where they could work as professional collectors in the Amazon rain forest, sending exotic and unusual specimens back to England for sale. They would also be able to indulge their own passion for natural history and might even find evidence for the transmutation of species. In 1848, the companions left for Brazil on the sailing barge Mischief. Wallace and Bates spent most of their first year collecting near Belém do Pará, before deciding to part company and explore the more remote reaches of the rain forest separately. Wallace spent a total of almost four years charting the Rio Negro, collecting specimens for sale and for his personal collection, as well as writing prodigious notes on the natural history, people, and places he visited. In 1849, he met up with another young explorer, who would become a lifelong friend, the botanist Richard Spruce, as well as Wallace’s younger brother Herbert, who planned to work as Wallace’s assistant. After some time together, Herbert decided that he was not cut out to be a collector, so he and Wallace parted company. Sadly, Herbert died of yellow fever while he was in Pará waiting for his passage back to Liverpool. Wallace probably felt partly responsible for this family tragedy, and it seems likely that this incident inspired, at least in some small part, his later interest in spiritualism. In July 1852, Wallace decided to return home on the brig Helen. This was a fateful decision, because the brig’s cargo of balsam caught fire three weeks away from Brazil. The vessel eventually sank, and Wallace lost all his personal collections and most of his notes, not to mention some of the live animals he was hoping to sell to collectors upon his return. However, he escaped with his life and was even able to save part of his diary and some sketches. The journey was an ordeal: Wallace and the crew spent ten days at sea in an open life boat before being picked up by another brig, the Jordeson, which was in a sad state of disrepair and running low on rations and water. Fortunately, his agent Samuel Stevens had insured Wallace’s collections, on his behalf, and Wallace spent the

WA L L A C E , A L F R E D R U S S E L

963

next 18 months in London living off the insurance payment and the profits he had made by selling the collections that he had sent back during his stay in the Amazon. During this time, he wrote six academic papers and two books describing his experiences in the Amazon. One of the papers dealt with the monkey species found in the Amazon. He had noticed that closely related monkey species tended to occur in close proximity, often found on neighboring sides of a river, and he was still actively thinking about species transmutation. Despite the rigors of his Amazon expedition, Wallace decided, almost as soon as he returned, to plan another expedition, this time to the even less explored Malay Archipelago. He was able to get his passage to Singapore arranged by the Royal Geographic Society, so in 1854 he left England once more. From Singapore he traveled to Borneo; it was the start of an eight-year-long journey. During that time, he collected more than 125,000 specimens (more than 80,000 beetles alone); of these, more than a thousand were species new to science. WALLACE’S GREAT DISCOVERY

Wallace’s travels around the many islands of the Malay archipelago also got him thinking about geographic distributions. He was struck by the zoological differences between the islands and island groups that he visited. He was amazed by the faunal discontinuity he found across the narrow Lombock Strait, dividing Bali and Lombock. He later proposed this as a zoogeographical boundary, which is now known as Wallace’s Line, that divides the Australasian and Asian fauna. He also had ample opportunity to refine his thoughts on transmutation. In September 1855, he published a paper in the Annals and Magazine of Natural History “On the Law Which Has Regulated the Introduction of Species.” In this paper, he did not discuss the mechanistic aspect of species transmutation, but he concluded that all species have come into existence coincident, in both space and time, with a closely allied species. His next paper went much further. In February 1858, Wallace sent a copy of his essay “On the Tendency of Varieties to Depart Indefinitely from the Original Type” to Darwin, with a request that he review it and pass it on to Charles Lyell, if he thought it worthwhile. This may sound like a strange request, as Darwin and Wallace had met only once and then only very briefly, but during his time in the Malay Archipelago, Wallace had begun corresponding with Darwin, and he clearly respected Darwin’s opinions.

964

WA L L A C E , A L F R E D R U S S E L

Wallace’s essay outlined the means by which species could diverge as a result of environmental pressures, although he did not use Darwin’s term “natural selection.” Darwin was shocked to receive such a concise synthesis of a theory that was so very similar to the one he had been working on for 20 years but had not yet published. Darwin asked for advice from Charles Lyell and Joseph Hooker, both respected scientists and close friends of Darwin’s. It was decided that Wallace’s essay would be published in a joint presentation, together with some unpublished writings that would highlight Darwin’s priority. Wallace’s essay was presented to the Linnean Society of London on July 1, 1858. The reading of the essay generated remarkably little reaction. It was not until the publication of Darwin’s On the Origin of Species in 1859 that the radical implications of this theory generated the public scrutiny Darwin had so much feared. Wallace had little choice but to accept the arrangement after the fact; he may even have been grateful to receive any credit at all. Wallace had neither the social nor the scientific status that Darwin had, and it is unlikely that his views on evolution and the transmutation of species would have received much attention without being linked to Darwin’s name. When Wallace returned to England in 1862, he found that he now had access to elite British scientific circles. Much to Darwin’s delight, Wallace turned out to be a staunch defender of natural selection. During the 1860s, he wrote several outspoken and eloquent articles, demolishing the arguments put forward by opponents of the theory. WALLACE’S VS. DARWIN’S THEORY OF EVOLUTION

Wallace and Darwin did not always share the same ideas about natural selection; in fact, there are fundamental differences in the emphasis of Darwin’s and Wallace’s theories. Wallace was always struck by geographic differences between neighboring islands or habitat islands, and he emphasized the role of the physical environment in driving adaptation and evolution, whereas Darwin emphasized the role of competition between individuals of the same species, and in particular the role of sexual selection. This fundamental difference in emphasis accounts for the frequent scholarly disagreements they had. WALLACE’S LIFE AFTER HIS TRAVELS

Back in England, Wallace moved in with his sister Fanny Sims and her husband Thomas, who owned a photographic company. Wallace occupied himself giving

lectures, writing articles about his travels and scientific findings, and dealing with both his personal and family business affairs, which took a lot of his time and energy. In 1864 Wallace became engaged to a young woman named Marion Leslie, whom he had been courting for some time. However, she soon broke off the engagement, for unclear reasons, leaving Wallace heartbroken and unable to concentrate on his scientific work. The return of his good friend Richard Spruce from his South American travels was the perfect distraction. Through Spruce, Wallace met William Mitten, who was a pharmacist by trade and bryologist by passion. Mitten had four daughters; the eldest, Annie, became a good friend of Wallace’s, and in 1866 they were married. SPIRITUALISM

During the summer of 1865, about the same time that he first met Annie Mitten, Wallace began to investigate spiritualism. His sister Fanny Sims had been involved in it for some time. Wallace reviewed the literature on the topic and attended séances; he became convinced that at least some of the phenomena he saw were real. He was not alone; spiritualism was very popular at the time. Many educated Victorians found that strict religious doctrine was at odds with their new understanding of the world, but they were not ready to adopt the completely materialistic and mechanical views that were increasingly emerging from nineteenth-century science, in no small part associated with Darwinian philosophy. Unfortunately, Wallace’s public advocacy of spiritualism seriously damaged his scientific reputation and strained his friendships with Henry Walter Bates, Thomas Huxley, and Darwin. Other scientists were more publicly hostile. At about the same time that Wallace began to become interested in spiritualism, he also began to vocally maintain that natural selection could not account for human intellect and artistic talents. Darwin disagreed vehemently, arguing that sexual selection could easily explain such apparently non-adaptive mental phenomena. Wallace’s views were at odds with the fundamental Darwinian idea that evolution is neither teleological nor anthropocentric. THE MALAY ARCHIPELAGO

In 1869, Wallace finally published a popular account of his travels and observations in The Malay Archipelago. His book became a bestseller; it was kept in continuous print by its original publisher for over 40 years. The book brought in some much needed income, particularly as Annie was now pregnant with their second child, Vio-

let. Despite having accumulated a considerable amount of money from the sale of specimens while he was away traveling, Wallace had squandered most of this money on bad business investments. FINANCIAL STRUGGLES

Financial difficulties were a recurring theme throughout Wallace’s life. Despite his scientific prowess, Wallace was never able to secure a permanent salaried position, possibly because of some of his more controversial and outspoken opinions. He relied almost entirely on income generated by writing to support his family. He graded government exam papers, wrote articles for commission, edited scientific works, and gave lectures. Finally in 1881, Darwin was able to secure a government pension of £200 per year for Wallace, in recognition of his contributions to natural history and science, but it required much hard lobbying on Darwin’s part. The modest income that the pension provided helped Wallace’s uncertain financial situation considerably. ZOOGEOGRAPHY

In 1874, tragedy struck the Wallace family: Wallace’s eldest son Herbert (named after his deceased uncle) died, after a long period of sickness. Wallace was devastated, and he threw himself headlong into his new project on the geographic distributions of animals. The Geographical Distribution of Animals was a monumental work that was published in 1876; in it, Wallace dealt with the patterns and causes of the global distribution of animal species. He emphasized the role that both historical and physical factors played in shaping the observed distributions of animals, and his exhaustive treatment of the patterns he observed provided a basis for defining the zoogeographic regions that are still recognized today. This two-volume work was quickly adopted as a classic text for students and was widely used for over 80 years after its initial publication. His next book, Island Life (1880), also addressed the processes shaping the geographic distributions of animals, but focused exclusively on islands. Wallace classified islands into three different types. Oceanic islands, such as the Hawaiian Islands and Galápagos, were categorized as those that had been formed in the middle of the ocean and had never been part of any large continent. As such, their biota are the result of transoceanic dispersal and subsequent in situ diversification. Such islands are notable for the almost complete absence of terrestrial mammals or amphibians. Continental islands,

WA L L A C E , A L F R E D R U S S E L

965

those that had once been part of a larger landmass, were divided into two separate classes depending on whether the connection was recent or ancient. Islands that have been isolated for a long time retain elements of ancient continental faunas that may no longer be abundant or even extant on the continents themselves. Wallace discussed the roles that isolation and climate change would have on island biota. He devoted a significant portion of the book to a discussion of the causes and impacts of glacial cycles on species distributions. It was an innovative and interesting publication, which received wide public acclaim. SOCIAL AND POLITICAL INTERESTS

From the late 1870s until his death, Wallace became increasingly interested in social and political issues, writing on a wide array of topics from environmental issues, to land reform, to women’s rights, to a critique of militarism, to opposing obligatory smallpox vaccination. His broad interests and talent for writing concise, clear prose meant that he was constantly being asked to contribute articles to popular journals and newspapers. THE AMERICAN TOUR AND DARWINISM

In 1886 Wallace was invited to give a lecture series at the Lowell Institute in Boston, and given that he was also in need of a financial boost, he decided to make the invited lecture series part of a ten-month lecture tour of the United States. He was a gifted speaker and had an undeniable knack for making complex scientific ideas easily accessible. Most of the lectures he gave were on Darwinism (evolution and natural selection), but he also lectured on biogeography, spiritualism, and social/economic reform. During the trip he traveled to California, to Oakland, to meet up with his brother John, who had emigrated there more than 30 years previously. He also got the opportunity to meet John Muir and visited the mighty redwood forests; he was much impressed by their grandeur and dismayed by humanity’s thoughtless destruction of these majestic trees. Before he left America, he made sure to spend a week exploring the Rocky Mountains with the American botanist Alice Eastwood as his guide. He had developed a passion for alpine floras. While in the States, he also met many other prominent American naturalists and viewed their natural history collections. Despite all the new and exciting places and people, Wallace was very happy to arrive back in Liverpool in August 1887. He had been troubled with minor ailments throughout the trip and was beginning to feel he was too old for these kinds of things.

966

WA L L A C E , A L F R E D R U S S E L

When he returned home, he set to work adapting the information he had compiled for the lecture series and other evidence he had accumulated during his American trip for his 1889 book Darwinism. It was one of Wallace’s most scholarly books, in which he explained and defended evolution by natural selection and developed carefully argued discussions of evolutionary phenomena. In this book, he presented a mechanism by which natural selection could drive reproductive isolation and thus result in the formation of new species. He argued that once two populations had become adapted to their own particular environments, hybridization could be disadvantageous, if the hybrid offspring had intermediate characters that were less adapted to their environment than the parental forms. Natural selection would select against these hybrids, whereas mechanisms that prevented hybridization from occurring would be favored, reinforcing the reproductive isolation of these two species. This idea is sometimes known as the Wallace Effect. Wallace also wrote at length about the adaptive significance of animal coloration, in particular the evolution of warning coloration in unpalatable insects. He and Darwin had corresponded at length about this topic. Darwin considered that sexual selection could explain most bright color patterns and ornamentation in the animal kingdom; Wallace disagreed, and even suggested that there might be alternative hypotheses to explain some of the examples of sexual selection Darwin had proposed. He also went on at some length about his ideas about humanity and evolution, once more reiterating his belief that human intellect was beyond the realms of natural selection. A LIFETIME OF ACHIEVEMENT HONORED

Throughout the 1890s, Wallace continued to write prolifically, tackling projects such as The Wonderful Century, reviewing the scientific achievements and failures of the nineteenth century. His lifetime of achievement was recognized and awarded; he was offered an honorary doctorate from Oxford University in 1889, and in 1890 he was awarded the Darwin Medal by the Royal Society. In 1892 he was awarded a founder’s medal by the Royal Geographic Society and was given the Royal Medal at the Linnean Society. Always a modest and unassuming character, Wallace was rather embarrassed by all of this recognition. Even his autobiography, My Life, which he published in 1905, is extremely self-effacing. The year 1908 marked the 50th anniversary of the jointreading of Darwin’s and Wallace’s papers, and the Linnean Society produced a medal in honor of the event. He also received the prestigious Copley Medal that same year and

much to his surprise and amusement, he was presented with an Order of Merit by the British Sovereignty. This was quite extraordinary, given that he was a self-acclaimed radical socialist who had openly opposed the Boer Wars and had campaigned for land nationalization.

WALLACE’S LINE AND OTHER BIOGEOGRAPHIC BOUNDARIES

WALLACE’S DEATH

Toward the end of his life, Wallace became more reclusive and spent most of his time at home with his family and his beloved garden. He died at home, peacefully, on November 7, 1913. His death was widely reported in the press. Some of Wallace’s colleagues even suggested that he be buried in Westminster Abbey, like Darwin; however, following his personal wishes, his family had him buried in the small cemetery at Broadstone, Dorset, near their home. In lieu of this honor, a committee of prominent scientists lobbied to have a medallion of Wallace placed in Westminster Abbey, near Darwin’s burial site. The medallion was uncovered in 1915. CONCLUDING REMARKS

During his lifetime, Wallace published at least 747 papers and 22 books, on a wide variety of subjects. He made significant contributions to evolutionary theory and biogeography and was deservingly honored as an exceptional scientist. His early exploration of both the Amazon and the islands of the Malay archipelago defined him as one of the foremost explorers and naturalists of his time. Despite all of his achievements, after his death, Wallace received relatively little recognition for his role in the development of evolutionary theory, at least in comparison with Darwin. This may be due to his vocal support of ideas that cannot be rationalized by evolutionary thinking, or it may be because of his modest and unassuming manner: He had always been willing to stand in Darwin’s shadow. SEE ALSO THE FOLLOWING ARTICLES

Continental Islands / Oceanic Islands / Sexual Selection / Voyage of the Beagle / Wallace’s Lines FURTHER READING

Fichman, M. 2004. An elusive Victorian: the evolution of Alfred Russel Wallace. Chicago: University of Chicago Press. Raby, P. 2002. Alfred Russel Wallace: a life. Princeton, NJ: Princeton University Press. Severin, T. 1997. The Spice Islands voyage: the quest for Alfred Wallace, the man who shared Darwin’s discovery of evolution. New York: Carroll and Graf. Slotten, R. A. 2004. The heretic in Darwin’s court: the life of Alfred Russel Wallace. New York: Columbia University Press. Wallace, A. R. 1905. My life. Google Books.

JEREMY D. HOLLOWAY The Natural History Museum, London, United Kingdom

Wallace’s Line represents the first attempt to establish a boundary between two major biotic regions of the world within an entirely archipelagic context. A prerequisite is the recognition of such biotic regions: gross areas that support distinctive but generally distributed assemblages of plants and animals. These regions are all continentbased, with climate being a secondary factor in the Northern Hemisphere: temperate versus tropical. A regional boundary must represent the most significant transition between two regions in terms of losses and gains of components. This transition should also be relatively abrupt: a discontinuity. HISTORICAL BACKGROUND

The eighteenth century saw a significant increase in exploration of the world by European navigators, traders, and scientists, and an increasing realization that other continents supported plants and animals of form and diversity that often contrasted very strongly with the European experience. The impact of the biotas of Australia and New Zealand, coming soon after the Linnaean revolution in taxonomy, were of particular force relative to the better-known ones of the “East Indies.” Concepts of distinct biogeographic regions were promoted and, inevitably with this, a perceived need emerged to establish the boundaries between them. Only the boundary between the Oriental and Australian regions would be likely to be discovered amid a geography that was entirely archipelagic. Alfred Russel Wallace provided the initial stimulus for such a boundary. Debate over its position and nature continued for the next century, as did further biological exploration, yielding an ever more detailed picture of plant and animal distributions. The very weight of this information resource led in the latter half of the twentieth century to a much more analytical approach to the classification of the organisms themselves, through cladistics, and of their distributions, through several distinctive and often antagonistic approaches to biogeography. These new approaches coincided with better understandings of a much more dynamic Earth surface than had previously

WA L L A C E ’ S L I N E A N D O T H E R B I O G E O G R A P H I C B O U N D A R I E S

967

been agreed upon and included the recognition of the dynamism brought about by the processes of plate tectonics, processes that have shaped the Indonesian Archipelago and its relationship with major continents in a most dramatic manner. WALLACE AND HIS LINE

By the time Wallace undertook his voyages in the “Malay Archipelago” in the mid-nineteenth century, much biological information and material had already been brought to Europe by the disciples of Linnaeus and by Dutch, British, and French explorers, including expedition scientists and traders. Wallace was aware of this work and so went out with at least some background knowledge. The stage was effectively set for his particular contribution: to make zoological collections intensively on a transect from west to the east through the archipelago, visiting the major islands of Sundaland and many of the Lesser Sundas, the extremities of Sulawesi, many of the Moluccas, and various places and offshore islands in western New Guinea. In this he had the support of two similarly indefatigable assistants, Charles Allen and Ali, recruited as teenagers, the latter from Sarawak in Borneo. Wallace’s combination of toughness, enthusiasm, curiosity, and perspicacity led him to accumulate and realize the significance of an immense amount of data on the distribution of vertebrates and insects through the archipelago, and, from his background knowledge and later research project, to recognize elements that were Australian in more westerly localities, particularly among the better known bird fauna. Wallace had a clear concept of what he considered to be the Indo-Malayan fauna, grouping the Philippines with this fauna, so he was particularly struck by the disappearance of many key elements of this fauna, especially bird groups, coupled with the appearance of elements he considered to be Australian, when he traveled from Bali to Lombok. The “Australian” taxa were not noted on his visits to Sulawesi, where he found a high level of peculiarities to the island, together with similar impoverishment of Indo-Malayan elements. These were the major influences in his first publication on the zoogeography of the archipelago, a work that appeared in 1860 (with an ornithological note in 1859) when he was still on his travels, as related in his more extensive account of them published in 1869, some years after his return. In 1860 he stated that the Strait of Lombok marked the limits and abrupt separation of two of the great zoological regions of the globe, but in 1869 his views appeared more measured on reflection, as he implied that it was only the Indo-

968

Malayan fauna proper that terminated at a line running between Bali and Lombok (the Strait of Lombok), north between Borneo and Sulawesi (the Makassar Strait), and thence northeast between Sulawesi and the Philippines. Areas to the east were either distinctive like Sulawesi or had an increasingly strong Australasian influence as shown by the Lesser Sundas (his Timor group of islands) and lands from the Moluccas eastward. By this time, T. H. Huxley had termed this boundary Wallace’s Line, but he proposed that its northern course should run to the west of the main Philippine archipelago, but to the east of Palawan. FURTHER LINES

Although Wallace’s Line initially met with general acceptance, the accumulation of data in the decades following led to alternative proposals (Fig. 1), the most significant of which was Weber’s Line of Faunal Balance, sited further to the east. This proposal was premised on the grounds that Weber’s Line marked the transition between islands where the fauna is predominantly Oriental and those where the fauna is predominantly Australasian. This line runs between the Sulawesi group of islands and the Lesser Sundas on one hand and the Moluccas on the other, with its southern extremity running to the east of the Tanimbar group. Wallace’s Line was seen increasingly as the eastern boundary of the Oriental region, with an equivalent boundary, Lydekker’s Line, marking the western limits of the Australo-Papuan region, which extends along the continental shelf of Australia and New Guinea. Geological information on the area was also becoming more detailed, and it became apparent that Huxley’s version of Wallace’s Line marked the extent of lands that had been connected to the Asian mainland during periods of low sea level during the glaciations. This included Bali but not Lombok, and also Palawan but not the rest of the Philippines. The intervening areas, particularly Sulawesi, the Moluccas, and the Philippines, were regarded more as a zone of transition, often termed Wallacea. The value of such lines in themselves has also come under scrutiny, a process that commenced in the middle of the last century, as exemplified in an essay by Ernst Mayr in 1944 from a zoological viewpoint, and in a review by Scrivenor and a group of zoologists and botanists in 1943; botanical and zoological views were found to differ considerably. THE IMPACT OF PLATE TECTONICS AND NEW BIOGEOGRAPHIC METHODS

The last half-century has seen major advances in two areas. The first is in our understanding of plate tectonics,

WA L L A C E ’ S L I N E A N D O T H E R B I O G E O G R A P H I C B O U N D A R I E S

100°

110°

120°

130°

140°

ASIA Wallace’s Line (of Huxley) 20°

20°

10°

10° Wallace’s Line (Original) Weber’s Line (Faunal Balance) Lydekker’s Line

BORNEO 0°

0°

NEW

GUI

NEA

10°

10°

Wallace’s Line (of Huxley) Wallace’s Line (Original) Weber’s Line (Faunal Balance) Lydekker’s Line 200-m. Line (Edge of Continental Shelves) 20°

100°

AUSTRALIA

110°

120°

130°

140°

20°

FIGURE 1 Zoogeographic boundaries proposed between the Oriental and Australian regions. The shaded areas are the continental shelves; these

would have been mostly exposed in periods of low sea level during the glaciations. Adapted from Mayr (1944).

a process leading to movement of the major continents relative to each other, and to extensive island arc formation and uplift in zones of plate convergence and subduction. A glance at a map of Indonesia and the western Pacific reveals the influence and complexity of such processes in the area. The movement of Australia northward on the Indian Ocean plate, and the shearing influence of the Pacific plate moving from east to west across the north of this, can be witnessed in the swirls and whorls of the emergent lands, the massive mountains in central New Guinea, the arcs of volcanoes, and the deep ocean basins and trenches. Thus, land areas have emerged, submerged, and moved relative to each other, leading to separation or convergence, often with fusion. Many of today’s islands are composites formed from different island systems, some incorporating continental fragments detached from Asia or Australia; Sulawesi incorporates all of these. This process has been operative incrementally through most of the Tertiary and has most recently been augmented by

the effects of sea-level fluctuations during the Quaternary glaciations. It is often said that Earth and life have evolved together. It is probable, therefore, that the mere drawing of lines on maps to explain the complexity of life in the archipelago between Asia and Australia is inevitably overly simplistic, given that the evolution of both Earth and life has involved significant mobility. The second major advance has been in the methodology used to study the distribution of plants and animals on a geographic scale, although this has involved some dichotomy of purpose and approach that reflects the evolutionary dichotomy just mentioned as well as a perceived need to separate the analysis of pattern from the interpretation of that pattern in terms of process. The basic data to hand are much more comprehensive today, and the taxonomy of many groups has been modernized by application of cladistic methodology to suites of morphological and molecular data to provide strong phylogenetic

WA L L A C E ’ S L I N E A N D O T H E R B I O G E O G R A P H I C B O U N D A R I E S

969

hypotheses for the groups concerned, including some still controversial attempts to date events in the phylogeny, particularly for groups with poor fossil records. The phylogenetic hypotheses can be combined with distributional data, especially where most individual taxa are endemic to islands or parts of islands, to provide hypotheses of the interrelationships of islands as an indication of Earth’s history (cladistic biogeography, also known as vicariance biogeography). The more extreme proponents of this approach have yet to attract more than fleeting attention from geologists, particularly given the difficulties of dating just mentioned and the difficulties of separating components of the patterns revealed, which owe their origins primarily to geological mobility rather than to the mobility of the organisms themselves. The potential for organisms to disperse across water gaps between islands and thus to reach remote and recently emergent islands, such as in the Hawaiian archipelago, is essential if such islands are to develop biotas with a degree of endemism but problematic if the preferred methodology postulates that patterns derived from dispersal are uninformative. The very nature of the Indo-Australian archipelago and what we know of its geological evolution makes it likely that such patterns will predominate. The alternative biogeographic approach in such a situation is to place more reliance on geological hypotheses and dating and to use biogeographic patterns to explore processes of dispersal, colonization, and speciation, while still allowing for reciprocal illumination between geology and biogeography. But, whereas geologists identify the lack of dating of biological events as a deficiency, biologists can also be frustrated by the reluctance of geologists to identify the extent and timing of their structures as dry land. LINES AS DISTRIBUTIONAL DISCONTINUITIES

So do Wallace’s Line or any of the others proposed have any reality? Even in the absence of modern taxonomic treatments, the distributions of species and higher taxa can be used to identify discontinuities in such distributions: major changes between islands in terms of components of their fauna or flora. Such analyses, usually involving some sort of “phenetic” clustering methodology applied to similarity coefficients, identify several major discontinuities in the archipelago coincident with all or parts of the lines of Weber and Wallace. That of the former is usually recaptured in entirety, but that of the latter is confirmed between Sulawesi and Borneo, and to some extent on either side of the Philippines; however, varying links between the Philippines and Sulawesi are not uncommon and may predominate over those between the

970

former and Borneo once the effects of faunal size disparities have been factored out. The Lesser Sundas from Lombok to Timor can show some integrity as a group, but usually they are revealed as an attenuation of the Javan fauna, with an increasing Australian element as one passes eastward. Bali was united with Java and the rest of Sundaland during Pleistocene sea-level falls, with the Strait of Lombok persisting throughout. Hence, the striking contrast between the two islands noted in the bird faunas by Wallace and others was mostly due to a sharp decline in the diversity of IndoMalayan elements, which rendered the small number of Australian elements more conspicuous. There is an even greater contrast in less dispersive vertebrate groups such as the mammals and freshwater fish, although the diversity of the former was greater in the recent past, as revealed by fossil faunas of Indo-Malayan character in the Lesser Sundas and southwest Sulawesi. The major discontinuities show variation in their importance from group to group, with Weber’s Line being emphasized for butterflies and birds, but the northerly part of Wallace’s Line being stronger for flowering plants and hawkmoths. In most such analyses, Sulawesi occupies a relatively isolated position, and it is often unclear whether its closest link is with the Philippines, the southern Moluccas, or even Java and the Lesser Sundas. THE POSSIBILITY OF TWO AUSTRALASIAN BIOTAS

Whereas the Oriental region is predominantly tropical in character, a significant part of the area and biota of Australasia can be considered temperate. During its northward drift, the original uniform Gondwanan biota of the Australian continent became more polarized into tropical and temperate elements, combined with increasing aridity in the interior. The tropical archipelago that developed to the north of Australia into the major island of New Guinea, with the Moluccas to the west and the Bismarck and Solomon groups to the east, now has a flora and a fauna that in some groups are more Indo-Malayan in character than Australian. This is particularly true of the insects, where the Indo-Malayan influence also extends strongly down through Queensland to northern New South Wales, leading J. L. Gressitt to suggest that the boundary between the Oriental and Australian regions for most insect groups occurred within Australia, rather than at Wallace’s Line! There is a contested possibility also that all the Australian butterflies were derived from the north by dispersal. It has also been suggested, from fossil pollen evidence, that India contributed significant Gondwanan

WA L L A C E ’ S L I N E A N D O T H E R B I O G E O G R A P H I C B O U N D A R I E S

elements to the Indo-Malayan flora when it made contact with Asia in the Early Tertiary, and that these have also spread eastward, probably via Sulawesi. The distributional discontinuities are evident in all these patterns, but their role in them is variable. For example, the two major lineages of cicadas in the Australasian tropics have an Indo-Malayan origin rather than originating within the temperate Australian fauna, yet they barely extend west of Wallace’s Line. The monitor lizard genus Varanus has an Australian lineage (including the Komodo dragon) that extends to Wallace’s Line through the Lesser Sundas, but also has a separate tropical Australasian lineage that is related to an Indo-Malayan lineage, which it replaces east of Wallace’s Line. MONTANE FLORAS AND FAUNAS

The previous sections have focused on the very rich lowland biotas of the archipelago, but it must not be forgotten that there are also distinctive ecosystems at higher altitudes on many of the islands, forming an archipelago of montane habitats within the greater archipelago. The recency of tectonic uplift on the majority of islands, except, perhaps, for parts of Sundaland (especially Borneo), means that distribution patterns of montane plants and animals can be very different from those of the lowlands, transgressing the discontinuities observed in the lowland flora and fauna. The oak-laurel forests of middle altitudes can support a diversity of animal life that is as rich as, or richer than, that of the lowlands, and the zones above this, with conifers and Ericaceae dominating and elements of a herbaceous alpine flora sometimes present, can also show significant diversity. North-temperate elements extend both from the Himalayan regions down though Southeast Asia to the mountains of Sumatra, Java, and the Lesser Sunda chain of islands, and from Taiwan to the Philippines, northern Borneo, and Sulawesi. But the major central archipelago of montane habitat in Borneo, Sulawesi, the Moluccas, and New Guinea shows strong representation of southtemperate taxa as well as some intrinsic elements that have radiated particularly within New Guinea. Thus, on these central mountains, there is an intriguing mixture of southern coniferous and myrtaceous shrubs, Himalayan ericaceous shrubs, and general alpine herbs in genera such as Ranunculus and Euphrasia. SEE ALSO THE FOLLOWING ARTICLES

Borneo / Indonesia, Geology / Island Arcs / New Guinea, Geology / Philippines, Geology / Plate Tectonics / Vicariance / Wallace, Alfred Russel

FURTHER READING

Gressitt, J. L. 1974. Insect biogeography. Annual Review of Entomology 19: 293–321. Hall, R., and J. D. Holloway, eds. 1998. Biogeography and geological evolution of South East Asia. Leiden, Netherlands: Backhuys. Holloway, J. D., and N. Jardine. 1968. Two approaches to zoogeography: a study based on the distribution of butterflies, birds and bats in the IndoAustralian area. Proceedings of the Linnean Society of London 179: 153–188. Mayr, E. 1944. Wallace’s Line in the light of recent zoogeographic studies. Quarterly Review of Biology 19: 1–14. Scrivenor, J. B., T. H. Burkill, M. A. Smith, A. S. Corbet, H. K. Airy Shaw, P. W. Richards, and F. E. Zeuner. 1943. A discussion of the biogeographic division of the Indo-Australian archipelago, with criticism of the Wallace and Weber Lines and of any other dividing lines and with an attempt to obtain uniformity in the names used for the divisions. Proceedings of the Linnean Society of London 154: 120–165. van Steenis, C. G. G. J. 1965. Plant geography of the mountain flora of Mt. Kinabalu. Proceedings of the Royal Society of London B 161: 7–38. Wallace, A. R. 1860. On the zoological geography of the Malay archipelago. Proceedings of the Linnean Society of London 4: 172–184. Wallace, A. R. 1869. The Malay archipelago. New York: Dover Edition (1962). Whitmore, T. C., ed. 1981. Wallace’s Line and plate tectonics. Oxford, UK: Clarendon Press. Whitmore, T.C., ed. 1987. Biogeographical evolution of the Malay archipelago. Oxford, UK: Clarendon Press.

WARMING ISLAND KURT M. CUFFEY University of California, Berkeley

Warming Island, situated at the north end of Liverpool Land in eastern Greenland (71o29’ N; 21o51’ W), is not particularly significant as a physical feature. Instead this island is noteworthy because of its connection to two major themes: the formation of new islands and the impacts of climate warming on glacial landscapes. Warming Island became an island only recently—between 2002 and 2005—when the glacial isthmus connecting it to the mainland was destroyed by melt and disintegration of ice, a consequence of climate warming (Fig. 1). CONFIGURATION, FORMATION, AND DISCOVERY

Quaternary glaciation has sculpted the island into three narrow parallel ridges, whose sides plunge steeply to footings below sea level. The highest elevation is about 520 m. The island is about 7 km in length and is separated from the mainland by a narrow strait. This strait is becoming wider as the glaciers emanating from the mainland continue to retreat. The terrain is mostly barren rock, with several small glaciers filling the heads of valleys. The bed-

WA R M I N G I S L A N D

971

FIGURE 1 Aerial view of Warming Island, showing the newly formed strait separating it from the glaciers flowing off the mainland, seen at left.

Photograph by Jeff Shea, taken during the 2006 expedition led by Dennis Schmitt.

rock consists of very old (Proterozoic) metamorphosed sedimentary sequences that were deformed in a tectonic collision with Eurasia around 400–450 million years ago. Igneous intrusions are also exposed in this region. Warming Island does not appear as a separate landmass in the first detailed aerial surveys of the east Greenland coast, in the mid-twentieth century. A partial map of the region from 1957 seems to show Warming Island separated from the mainland, but this map is missing numerous features of the region—including a whole island and a large part of Liverpool Land—and cannot be regarded as complete or accurate. Satellite images show that in the mid-1980s the island was still connected to the mainland by glaciers flowing from the highlands of northern Liverpool Land. At this time, the ice surface close to the island appears to have been nearly flat, suggesting that this was a floating ice shelf. Like many ice masses in eastern Greenland, this ice shelf subsequently thinned, and its margin retreated; by 2002, the ice connecting Warming Island to the mainland was a narrow neck, no more than 1 km in width. In September 2005, a group of explorers led by Dennis Schmitt of California discovered that the ice neck had vanished. The explorers sailed through the open waters of the new strait. Schmitt named the newly isolated island Uunartoq Qeqertoq, an Inuit-language phrase for “the warming island.” This discovery coincided with new scientific reports of increased melt and ice flow on the nearby Greenland Ice Sheet, and with increased general concern about global warming. Warming Island thus became an object of public interest, a symbol of Earth’s transformation in a warming climate. It was featured in media reports, including an article in The New York 972

WA R M I N G I S L A N D

Times. Editors of the Oxford Atlas of the World designated Warming Island as the “Place of the Year” in 2007. THE ISLAND AS A SYMBOL OF GLOBAL WARMING

Does it make sense to use Warming Island as a symbol of the transformative effects of climate change—and of human-caused warming in particular? There is essentially no specific information available about the glacier flow and the climate history in the immediate vicinity of Warming Island. And it is inherently problematic to use events at a single place (or a few places) to represent a complex, diverse, and global-scale phenomenon. But diminishing glacial ice is without doubt a central issue of climate warming. Diminishing glaciers provide visually captivating evidence of the warming itself and, because they contribute to sea-level rise, are one of the warming’s most important consequences. The majority of glaciers in the world’s high mountain ranges are retreating because of climate warming. This is occurring, for example, in the Himalayas, the Andes, the Canadian Rockies, and the ranges of southern Alaska. Retreat in eastern Greenland is thus part of a global phenomenon. In many locations, retreat began in the late nineteenth century, before human-caused warming was significant. Brightening of the sun was likely the most important cause of this initial warming. By the first decade of the twenty-first century, however, most of the cumulative warming—including essentially all of the warming after 1970—was due to human-caused increases of atmospheric greenhouse gases. Thus ongoing worldwide glacier retreat can be attributed, in large part, to human causation.

In Greenland, warming of air and ocean waters has increased melt of glaciers and thinning of floating ice shelves. On the main Greenland Ice Sheet, warming in the 1990s and the 2000s expanded the zone of melting. The ice sheet shrank as a result of both increased melt and increased glacier flow to marine margins where icebergs form. The glaciers of Liverpool Land and Warming Island are not part of this main ice sheet, but are subject to the same climatic and oceanic influences. Measurements of air temperature show that eastern Greenland is now about 2 °C warmer than at the start of the twentieth century. Similarly warm temperatures prevailed in the 1930s. The intervening decades were cooler, but not as cold as the first years of the twentieth century. Given this history, it is likely that the glacier retreat in eastern Greenland began in the first warm period (the 1930s) and was rejuvenated by the more recent warming. It is thus reasonable to regard the glacier retreat here as partly natural and partly human-caused. Over the next century, the human-caused warming is expected to become very much larger than natural variations. Glacier retreat around Warming Island is not documented nearly as well as glacier retreat in some other locations, such as the European Alps. Nor has the retreat at Warming Island been as extensive and dramatic as that in places like coastal Alaska and Patagonia. But the Warming Island case is culturally compelling. Not only did retreat here lead to the formation of a new coastal island, but it also occurred at 71° latitude, in the heart of the Arctic, adjacent to one of the world’s great ice sheets. The attention given to Warming Island as a symbol of climate change seems fully justified, provided it is recognized that glacier retreat in this particular location is caused by local climate changes whose relation to global-scale warming is complex and not fully elucidated.

ing the weight of the ice leaves a low-pressure zone in the underlying mantle, which is gradually filled by lateral flow of mantle rock. This causes the surface above to rise. This process—isostatic rebound—creates islands when topographic summits emerge from the surface of the ocean or a large lake. Second, when large ice sheets melt, the resulting rise of sea level drowns continental watersheds and converts topographic summits into islands. Many islands were formed by this process as sea level rose by approximately 130 m at the end of the last ice age. SEE ALSO THE FOLLOWING ARTICLES

Global Warming / Island Formation / Sea-Level Change FURTHER READING

Box, J. E., D. Bromwich, B. A. Veenhuis, L.-S. Bai, J. C. Stroeve, J. C. Rogers, K. Steffen, T. Haran, and S. Wang. 2006. Greenland Ice Sheet surface mass balance variability (1988–2004) from calibrated polar MM5 output. Journal of Climate 19: 2783–2800. Henriksen, N., A. K. Higgins, F. Kalsbeek, and T. C. R. Pulvertaft. 2000. Greenland from Archaean to Quaternary. Descriptive text to the geological map of Greenland, 1:2,500,000. Geology of Greenland Survey Bulletin 185. (Geological Survey of Denmark and Greenland.) Kaser, G., J. G. Cogley, M. B. Dyurgerov, M. F. Meier, and A. Ohmura. 2006. Mass balance of glaciers and ice caps: consensus estimates for 1961–2004. Geophysical Research Letters 33: L19501. Oerlemans, J. 2005. Extracting a climate signal from 169 glacier records. Science 308: 675–677. Rignot, E., and P. Kanagaratnam. 2006. Changes in the velocity structure of the Greenland Ice Sheet. Science 311: 986–990. Rudolf, J. C. 2007. A peninsula long thought to be part of Greenland’s mainland turned out to be an island when a glacier retreated. The New York Times, January 16, 2007. Solomon, S., et al. Technical summary, in Climate change 2007: the physical science basis. Contribution of Working Group 1 to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press. Steffen, K., S. V. Nghiem, R. Huff, and G. Neumann. 2004. The melt anomaly of 2002 on the Greenland Ice Sheet from active and passive microwave satellite observations. Geophysical Research Letters 31: L20402.

ISLAND FORMATION BY GLACIER RETREAT

Warming Island is also interesting as an example of the formation of a new island, an event that is rarely witnessed. The best known new islands, like Surtsey in Iceland, owe their birth to volcanic eruptions. But the majority of new islands formed in the last 15,000 years are the result of glacier retreat. Glaciers can scour bedrock to well below sea level. They scoured deep valley systems near the edges of the Quaternary ice sheets in Scandinavia, Patagonia, and several regions in North America (the Pacific Northwest, Canadian Arctic, and North Atlantic). Retreat then produced numerous islands as ocean water replaced ice. The formation of Warming Island is a good modern example of this process. Two other processes form new islands as ice sheets retreat. First, retreat causes uplift of the Earth’s crust; remov-

WATER RESOURCES SEE HYDROLOGY

WHALE FALLS AMY BACO Associated Scientists at Woods Hole, Massachusetts

“Whale fall” is the general term for a sunken whale carcass resting on the sea floor. Whale falls pass through a series of successional stages ranging from scavenging of soft tissue through to a highly diverse chemoautotrophic

W H A L E FA L L S

973

assemblage, fueled by sulfides from the anaerobic breakdown of lipids contained in the whale bones. Whale falls have species overlap with hydrothermal vents and cold seeps and thus have been hypothesized to have played a role in the dispersal and evolution of vent and seep fauna. In addition to these shared species, whale fall communities have components of endemic and specialized fauna that are specific to whale falls. HISTORY

Scientists have theorized that sunken whale carcasses might be important to deep-sea fauna since the 1930s, and whale bones with attached fauna had been brought up in fisheries’ trawls for over 150 years. The first observed whale fall was discovered accidentally off southern California in 1987, during other deep-sea studies with the submersible Alvin. A second whale fall was discovered nearby during military mapping work in 1995. In addition to these two natural whale falls, a number of whale carcasses, from animals that died of natural causes, have been sunk for scientific study off southern California, central California, Japan, Sweden, and New Zealand. Fossilized whale bones with associated fauna have also been found, dating to 30–40 million years ago. SUCCESSIONAL STAGES

When a fresh whale fall reaches the sea floor, it represents a food bonanza to an otherwise food-poor deep-sea fauna. The soft tissue of the carcass will be rapidly consumed by sharks, hagfish, crabs, and amphipods, at rates as high as 50 kg/day. This “mobile scavenger stage” slowly transitions to an “organic enrichment stage” in which millions of smaller animals and bacteria, in the sediments and on the bone surfaces, continue to break down and consume the putrid remnants of the whale soft tissues, eventually leaving the bones exposed (Fig. 1).

FIGURE 1 A whale fall that has been on the sea floor for 18 months.

Hagfish swarm among the bones, looking for leftover bits of soft tissue. Photograph courtesy of Craig Smith and Mike DeGruy.

974

W H A L E FA L L S

In addition to the highly labile carbon provided by the whale blubber, organs, and other soft tissues, the bones of whales contain large volumes of lipids (fats and oils), up to 60% by volume, which are slowly broken down by heterotrophic bacteria with inorganic sulfate as oxidizer, producing sulfides. These sulfides in turn fuel chemosynthetic production by various free-living microbes on the bones and endosymbiotic bacteria hosted by various invertebrate taxa living on or near the bones. This stage is termed the “sulfophilic stage” and is characterized by a highly diverse suite of fauna with a complex food web. COMPARISON TO OTHER CHEMOSYNTHETIC ECOSYSTEMS

Whale falls, like other types of deep-sea chemosynthetic ecosystems (hydrothermal vents, cold seeps, sunken wood), are a type of deep-sea habitat island. Throughout each of the successional stages, the fauna found on the whale falls are very distinct in species composition and abundance from that of the surrounding sea floor. Abundant taxa in the sulfophilic stage include vesicomyid clams, bathymodiolin mussels, and siboglinid worms. These are the same taxa that characterize many vent and seep habitats. In fact, much of the fauna found at whale falls is similar to what is found at vents and seeps, with many of the same families and genera, and even some of the same species, being represented. Unlike vents and seeps, whale falls are not restricted to certain types of geological features. They are also potentially abundant on the sea floor, estimated at one whale fall every 12–36 km, based on abundance and mortality of the nine largest whale species. Whale falls from the largest species of whales are estimated to harbor chemosynthetic fauna for up to 100 years. These factors, combined with their faunal similarities to vents and seeps, led to the hypothesis that whale falls may act as dispersal stepping stones for vent and seep fauna. There has been DNA sequence–based confirmation of species overlap in several taxa between whale falls, vents, and seeps, supporting this hypothesis. An interesting spinoff of these studies is that whale falls, along with other types of organic remains such as sunken wood, may have also played some role in the evolution of vent and seep fauna. Although whale falls share species with vents, seeps, and sunken wood, they have much higher levels of biodiversity on a local scale than do vent and seep habitats, and also more than any other type of deep-sea hard-substrate habitat, with nearly 200 species being found on a single skeleton in the sulfophilic stage. Large whales have existed for sufficient time (30–40 million years) to plausibly have evolved a whale-fall specialist or endemic fauna, which may contribute to higher

diversity levels. In fact there are close to 40 species (approximately 10% of the known species from whale falls) that have so far been found only on whale falls and in no other type of habitat. Sulfide, which fuels the whale fall communities, is also toxic to most animals. Whale falls have lower levels of sulfide than do vents and seeps, which may allow more of the background fauna to colonize, again contributing to the high levels of biodiversity. SEE ALSO THE FOLLOWING ARTICLES

Cold Seeps / Hydrothermal Vents / Organic Falls on the Ocean Floor / Succession / Whales and Whaling FURTHER READING

Baco, A. R., and C. R. Smith. 2003. High biodiversity levels on a deep-sea whale skeleton. Marine Ecology Progress Series 260: 109–114. Baco, A. R., C. R. Smith, A. S. Peek, G. K. Roderick, and R. C. Vrijenhoek. 1999. The phylogenetic relationships of whale-fall vesicomyid clams based on mitochondrial COI DNA sequences. Marine Ecology Progress Series 182: 137–147. Distel, D. L., A. R. Baco, E. Chuang, W. Morrill, C. M. Cavanaugh, and C. R. Smith. 2000. Do mussels take wooden steps to deep-sea vents? Nature 403: 725–726. Rouse, G. W., S. K. Goffredi, and R. C. Vrijenhoek. 2004. Osedax: boneeating marine worms with dwarf males. Science 305: 668–671. Smith, C. R. 1992. Whale falls. Oceanus 35: 74–78. Smith, C. R. 2007. Bigger is better: the role of whales as detritus in marine ecosystems, in Whales, Whaling and Ocean Ecosystems. J. Estes, ed. Berkeley: University of California Press. Smith, C. R., and A. R. Baco. 2003. Ecology of whale falls at the deep-sea floor. Oceanography and Marine Biology Annual Review 41: 311–354. Smith, C. R., H. Kukert, R. A.Wheatcroft, P. A. Jumars, and J. W. Deming. 1989. Vent fauna on whale remains. Nature 341: 27–28.

WHALES AND WHALING JOE ROMAN University of Vermont, Burlington

Whaling has impacted all species of great whales. It has also changed island ecosystems and cultures, spread invasive species and disease, and brought some island species, such as the Galápagos tortoise, to the brink of extinction. As hunting has declined, whale watching has risen as an important economic activity on many islands. WHALE DISTRIBUTION

There are currently about 80 recognized species of cetaceans, divided into two suborders, the Mysticeti, or filterfeeding baleen whales, and the Odontoceti, or toothed whales, a category that includes dolphins, porpoises, and

sperm whales. The great whales, a group based on the cultural history of whaling rather than systematics, are mostly composed of mysticetes—the gray whale (Esrichtiidae), right whales (Balaenidae), and rorquals (Balaenopteridae) such as humpbacks and blue whales—and one odontocete, the sperm whale Physeter macrocephalus. The distribution of whales is linked to environmental features such as temperature and underwater topography. Many species of cetaceans are attracted to the coastal margins of islands and continents to feed, where highly productive habitats are found. Physical processes such as transverse circulation and tidal mixing can bring nutrients into the euphotic zone in these areas, enhancing primary and secondary productivity. Such productivity concentrates zooplankton, fish, and squid on shelf fronts, attracting whales, seals, and seabirds. Whales may also exploit island wakes, feeding on predictable aggregations of plankton and nekton along localized upwellings and fronts. Islands provide shallow breeding areas for whales. For coastal species such as humpback and right whales, such waters are especially important. Many northern Pacific humpback whales, for example, migrate from high-latitude feeding grounds to the Hawaiian Islands in the central Pacific, where they gather to breed. In the North Atlantic, humpbacks travel from summer feeding areas in northeast North America, northern Europe, and Iceland to breeding grounds in the West Indies and Cape Verde Islands. Many species exhibit site fidelity to their nursery grounds, indicating that these are important, and perhaps limited, areas for reproduction. Such fidelity to feeding and calving areas has made whales vulnerable to human hunting and disturbance. TRADITIONAL AND ABORIGINAL WHALING

Humans have hunted large cetaceans for millennia, and whaling has played an important role in the culture and ecology of several island groups around the globe. Archaeological evidence indicates that whaling cultures arose about 2000 years ago along the Bering and Chukchi Seas and on the coasts of Korea and Norway. In the North Pacific, Arctic and temperate aboriginal whalers used dugout boats or umiaks (large, open boats covered with skin), and harpoons and lances. Whaling for bowhead and gray whales changed societies around the Pacific Rim, supporting the establishment of permanent villages and trade networks. The hunt continued for centuries along the islands around the rim of the North Pacific, from Japan to Vancouver Island. Some islanders developed their own traditions of hunting. Poison darts were used to hunt right whales off

WHALES AND WHALING

975

the Aleutians, and flax nets were employed by Japanese whalers to capture humpbacks, right whales, and other species. In the tropics, aboriginal hunting of humpbacks and sperm whales began in the Philippines and Indonesia around the 1600s, employing open boats and handheld harpoons or hooks. In at least one case, prehistoric hunting continues to affect island ecology. Thule Inuit whalers in the High Arctic thrived on whale meat and built the structures of their winter settlements out of bowhead-whale bones (Fig. 1). Somerset Island, in what is now Nunavut, Canada, has the greatest concentration of these structures, which were occupied between AD 1200 and 1600. The towing of whales ashore, where they were flensed (stripped of their blubber), and the concentration of human activity changed the water quality and planktonic assemblages of the island’s ponds. Although whalers abandoned the area more than four centuries ago, the legacy of these human disturbances is still evident in the present-day limnology of the island’s ponds, characterized by elevated nutrient concentrations and atypical biota. Subsistence hunting continues on some of these islands: Bowheads are hunted in Saint Lawrence Island in the North Pacific, humpbacks in the West Indies and in Tonga, and Bryde’s whales in the Philippines. A small hunt for sperm whales and other odontocetes continues, with wooden boats and hand-delivered harpoons, on the Indonesian islands of Solor and Lembata. COMMERCIAL WHALING

The development of whaling as a commercial industry is generally credited to the Basques, with the earliest known

FIGURE 1 The direct and indirect effects of whaling have altered the

ecology of many islands. This house on Somerset Island, Nunavut, built from the bones of bowhead whales, dates back at least 400 years. The island’s ponds retain high nutrient levels and unusual planktonic assemblages as a result of the centuries-old hunt of Thule Inuit whalers. Photograph by John P. Smol, Queen’s University.

976

WHALES AND WHALING

records from around AD 1000. The reach of whaling in the centuries that followed would touch almost every island: tropical and polar, continental and oceanic, those whose inhabitants had a history of whaling, and those who had only seen a cetacean as a distant blow or a windfall of meat and when it cast ashore. Svalbard, the Azores, Cape Verde, Saint Helena, South Georgia, Nantucket, the West Indies, the Galápagos, the Hawaiian Islands, New Zealand, and the Seychelles are a few of the islands central to the whaling trade. In Basque-style whaling, open boats were deployed from shore or coastal ships, typically manned by six oarsmen, a captain, and a harpooner. Whalers hunted with hand-held harpoons and lances, and much of the processing of the oil was typically done onshore. Basque whalers were hired by other European nations, and their technique was employed around the rim of the Atlantic. Mysticetes were the first commercially exploited whales, first the North Atlantic right and then extensively the “common whale” or bowhead. One of the most productive whaling grounds in history was the region surrounding the Arctic island of Spitsbergen in the Svalbard archipelago. In 1599, the Dutch expedition of Willem Barents reported the presence of numerous whales in the area. Within a decade, the Basques and the British, Dutch, and Danes, often with Basques onboard, were competing for the bowhead whale, attractive for its long and valuable baleen (“whalebone”), and high oil yields. On Spitsbergen, as on islands before and after, shorelines became littered with the debris of the fishery—piles of blubber, bones, and guts—and lined with buildings to support the shore-based effort. Before the hunt for bowheads began, there may have been 500,000 of these polar whales in the Arctic. There are now probably less than 10,000, with numbers in the tens around Svalbard. The removal of this species from the marine ecosystem altered the food web in the area, from plankton to cod and seabirds. The development of onboard tryworks in the eighteenth century enabled American and European whalers to process and cask oil at sea, opening up whaling grounds in the Pacific. They still used hand-thrown harpoons and petal-shaped lances or “killing irons” to finish off their quarry, but the primary target of American-style pelagic whaling was the sperm whale, with its clean-burning oil and spermaceti, which was made into candles. Pelagic hunting soon brought commercial hunters to remote islands around the world in search of provisions, shore leave, and a chance for some unhappy crewmembers to desert a luckless voyage or overbearing captain. Whalers collected firewood and seabirds and their eggs on isolated

atolls in the central Pacific. They also helped spread infectious disease. Epidemics of tuberculosis, typhoid, influenza, and smallpox followed colonization and commercial shipping, including whaling, throughout Polynesia. Dysentery hit Tahiti in 1807 after the passage of the whale ship Britannia. As a result of these diseases, many Polynesian islands lost more than 80% of their population. The Hawaiian Islands were an important stopover for nineteenth century whalers, providing freshwater, fruits, and vegetables. The presence of kanaka (Hawaiian men) willing to fill empty berths, and brothels full of welcoming wahine, made stops at Honolulu and Lahaina attractive to captain and crew alike. Further south, American, British, and French whalers set up hundreds of stations along the bays and open coasts of New Zealand to hunt for right whales. The most famous whaling grounds in the Pacific in the nineteenth century were the Offshore Grounds (5–10º S, 105–125º W), west of the Galápagos Islands. Discovered in 1818, the grounds were visited by whalers in all seasons. The Galápagos were attractive for the provision of food and freshwater, but they were no tropical paradise. Herman Melville remarked on the “emphatic uninhabitableness” of the islands. “No voice, no low, now howl, is heard. The chief sound of life here is a hiss.” There were some inhabitants: iguanas, birds, and, most prized to whalers, turtles. Galápagos tortoises were easy prey, supplied much-needed fresh meat, and could be stowed aboard ship for months. In 1848, the Daniel Lincoln took 273 tortoises from Chatham Island in just five days. With tens of thousands of tortoises removed by whalers, many islands were soon emptied of adults. Rats, introduced accidentally, removed the remaining eggs. Goats were intentionally introduced to the Galápagos, probably by New England whalers, to provide a larder for ships when they provisioned on the islands. Long after American whaling declined in the late nineteenth century, goats remain a problem, though eradication efforts are under way. In the twentieth century, commercial whalers employed diesel-powered catcher boats and deck-mounted cannons that could kill even the swiftest whale. One of the first whaling stations used to exploit the Southern Ocean was founded in Grytviken on South Georgia in 1904. The island had already been emptied of seals in the eighteenth century, but Carl Anton Larsen, a famous Norwegian mariner and whaler, saw the potential as a whaling station. Not only did large populations of whales surround the island, but its shores also remained ice-free in the winter. The coastal humpback was soon depleted; fin whales and blues followed. To make the island more like home, Larsen had reindeer released on the islands, and

several herds persist there to this day. When he visited the island in 1950, British surgeon R. B. Robertson described the station as “the most sordid, unsanitary habitation of white men to be found the world over.” The abandoned station still causes environmental and health hazards, with asbestos and petroleum remaining among the ruins. After the rise of factory ships and the near depletion of many coastal species, whaling became almost entirely pelagic in the mid-twentieth century, with stern slipways allowing the carcasses to be hauled quickly onboard for processing. We see the ramifications of whaling in the oceans today. Centuries of exploitation have greatly reduced cetacean populations in the Northern Hemisphere. Four hundred years after the Svalbard fishery began, the horizon lines off these Arctic islands are bereft of the once-abundant V-shaped blows of bowhead whales. Right whales in both the North Atlantic and North Pacific number just a few hundred individuals. Gray whales in the east Pacific now number more that 25,000, yet only a remnant population of about 100 whales survives along the western rim of the ocean. In the Southern Hemisphere, more than 2 million whales were killed in the twentieth century. Humpbacks and southern right whales remain rare, and blue whales show little sign of recovery. We also see the impacts of whaling on islands, in the extinction of endemic species such as turtles, in the persistence of intentionally and unintentionally introduced species, and in the declining health of Hawaii’s birds. The crew of the British whaling ship Wellington has frequently been blamed for introducing the first Culex mosquitoes to Hawaii, in drinking water from Mexico in 1826. The subsequent epidemics of avian malaria and poxviruses that broke out among native birds may have been initiated with this ship. Although several authors challenge the specificity of the event, there is no doubt that the frequent trafficking between islands by whale ships inevitably spread species and disease. American whalers have also been implicated in the spread of mosquitoes to Australia and many other Pacific islands. (In one case, whales played a minor role in the eradication of an invasive species. When the Asian citrus blackfly, Aleurocanthus woglumi, was found in Key West, Florida, in 1934, an emulsion of whale-oil soap, paraffin oil, and water was sprayed on infected trees. The fly never reached the other keys.) Industrialized whaling also had impacts on the ocean ecosystems themselves. One of the most highly productive areas in the Northern Hemisphere was the Aleutian arc, an oceanographic hotspot that attracted whales and later whalers. More than 60,000 fin, sei, and sperm whales were killed in this area in the 1940s through 1960s, while they

WHALES AND WHALING

977

were on their summer feeding grounds. (Right whales, bowheads, humpbacks, gray whales, and blues had already been depleted decades earlier.) The near extirpation of most whale species may have prompted a dietary shift in the largest mammal-eating odontocete, the killer whale. As killer whales switched from a diet of offshore whales to sea otters and Steller sea lions, they may have caused the decline of both nearshore species in the Aleutians. Since the moratorium on commercial whaling was imposed in 1986, direct impacts from hunting have been greatly reduced. Only a few island nations continue industrial whaling. Iceland resumed commercial whaling in 2006, despite protests from Great Britain, the United States, and other nations. Japan uses a loophole in the International Whaling Commission’s bylaws, selling meat that has been collected for what it claims to be a scientific hunt. Both countries have a long tradition of whaling, dating back at least a thousand years in surrounding waters. Japan’s contemporary, and controversial, whale hunt is now largely conducted in the Southern Hemisphere. Once viewed as goods, whales are widely recognized for the ecosystem services they provide, largely through tourist operations. Hawaii, many islands in the West Indies, the Canary Islands, the Galápagos, and Iceland all have thriving whale-watching industries. Despite the shift away from commercial whaling, humans continue to impact cetacean populations. Whales are killed in collisions with ships and after being entangled in fishing gear. Seismic air guns and sonar can cause internal hemorrhages and gas-bubble disease, or the bends. Persistent anthropogenic noise from ships and active sonar reduces the ability of whales to communicate. Even whale watching itself can change whale behavior, feeding patterns, and mating activity. WHALE ISLANDS

One persistent and fascinating connection between whales and islands is the mythology of whales as islands. The whale-island motif dates back at least 2000 years and can be found among the folklore of maritime cultures around the world. Two of the most famous tales involve St. Brendan the Navigator and Sinbad. On his epic voyage in search of Tir na nÓg (translated as the Enchanted Isles or the Land of Promise), Brendan and his fellow monks set foot on a barren island on Easter Sunday to say Mass. As he prayed at the altar, the monks lit a fire to cook breakfast, and the ground stirred beneath them. When they reached their boat, the island swam away. The deceptive island was in fact an enormous whale, who would later offer his back to help the monks on their Atlantic journey (Fig. 2). On Sinbad’s first voyage, he and his fellow merchants also land

978

WHALES AND WHALING

FIGURE 2 Stories of whale islands date back at least 2000 years. Dur-

ing St. Brendan’s legendary Atlantic voyage in the sixth century, he and his fellow monks stopped at a barren island to say Mass. After they lit a fire, the island—actually a whale—awoke, tossing the pilgrims into the sea. Image courtesy of the New Bedford Whaling Museum.

on a small island and light a fire. But for Sinbad, who loses his treasure when the waking whale plunges into the ocean, the elusive cetacean is an obstacle to his pursuit of wealth. Sinbad’s mistrusting view, and the whaler’s dogged pursuit of cetaceans, would dominate the relationship between humans and whales for centuries. By the late twentieth century, however, as perceptions of cetaceans changed, Brendan would become the patron saint of whales. Such tales are more fancy than fact, of course. Although right whales can be seen logging at the surface in calm days at sea, it is hard to imagine anyone mistaking the blubbery slab for an actual island. Yet there are real whale islands. After death, whale falls in the deep sea attract scavenger assemblages that recycle soft tissue over a short time. These carcasses serve as habitats for numerous endemic species, including polychaetes, molluscs, and other chemoautotrophic organisms. They also provide nutrients for scavenging crustaceans, fish, and sharks. A large whale can support successional communities for at least 80 years before it decomposes completely. Just as whaling changed island ecosystems, it may have endangered these communities before they were even discovered. SEE ALSO THE FOLLOWING ARTICLES

Bird Disease / Galápagos Islands, Biology / Iceland / Spitsbergen / Whale Falls FURTHER READING

Ellis, R. 1991. Men and whales. New York: Knopf. Estes, J. A., D. P. Demaster, D. F. Doak, T. M. Williams, and R. L. Brownell Jr., eds. 2006. Whales, whaling, and ocean ecosystems. Berkeley: University of California Press.

Perrin, W. F., B. Würsig, and J. G. M. Thewissen, eds. Encyclopedia of marine mammals. 2002. San Diego, CA: Academic Press. Reeves, R. R. 2002. The origins and character of ‘aboriginal subsistence’ whaling: a global review. Mammal Review 32: 71–106. Roman, Joe. 2006. Whale. London: Reaktion. Tønnessen, J. N, and A. O. Johnsen. 1982. The history of modern whaling. R.I. Christophersen, trans. Berkeley: University of California Press.

WIZARD ISLAND DAVID W. RAMSEY U.S. Geological Survey, Vancouver, Washington

Rising steeply above the water like a sorcerer’s pointed hat, Wizard Island is the most prominent and recognizable feature in Crater Lake, Crater Lake National Park, Oregon. Crater Lake, the deepest lake in the United States and the seventh deepest lake in the world (594 m depth relative to the shoreline), partially fills the caldera that formed approximately 7700 years ago by the eruption and subsequent collapse of an approximately 3700-m volcano called Mount Mazama. Since the climactic eruption of Mount Mazama, there have been several less violent, smaller post-caldera eruptions within the caldera itself. Wizard Island is one of four known volcanic vents within the caldera and is the only postcaldera volcano visible above the surface of Crater Lake. MOUNT MAZAMA AND CRATER LAKE

Mount Mazama, a major, ~3700-m andesite-dacite stratovolcano in the Cascade Range, collapsed during a climactic eruption approximately 7700 years ago, leaving an 8 × 10-km caldera, which is now partially filled by Crater Lake (Fig. 1). Prior to the climactic event, Mount Mazama had a 400,000 year history of cone-building activity compa-

rable to that of other long-lived Cascade volcanoes such as Mount Shasta. Since the climactic eruption, volcanism has been confined to the caldera, where most of the products of post-caldera eruptions are obscured beneath Crater Lake’s surface. Exploration of the lake floor by remotely operated vehicles (ROV) and the manned submersible Deep Rover, dredged and cored samples, and a recent high-resolution multibeam bathymetric survey of the entire lake nearly to its shoreline have revealed the geology, geomorphology, and post-caldera eruptive history of Crater Lake. These surveys also discovered evidence that Crater Lake filled rapidly, has remained at or near its present level for an extended period of time, and maintains this level through a balance of precipitation, evaporation, and leakage into the subsurface, especially through permeable glacial sands and gravels found along the waterline in the northeast caldera wall. GEOLOGY AND FORMATION OF WIZARD ISLAND

Wizard Island is one of four known volcanic vents within the caldera and is the only post-caldera volcano visible above the surface of Crater Lake. However, the 1.6-km2 visible cinder cone and blocky lava flows of andesitic Wizard Island volcano represent only 2.4% of the volume of the total edifice, which rises around 750 m above the lake floor, has an areal extent of 9.0 km2, and has a volume of at least 2.6 km3 (Fig. 2). Based on recent modeling of the history of lake filling, it is believed that Wizard Island began erupting within a few decades of caldera collapse and ceased erupting about 200 to 500 years later. The subaqueous flanks of Wizard Island consist of relatively flat slopes (2–10°) of lava that transform abruptly to steep talus slopes (29–36°) of broken lava fragments, representing former shorelines where subaerial lava flows entered water, chilled, and fractured, forming slopes of subaqueous breccia (talus) (Fig. 2). The transition from

FIGURE 1 Panoramic photograph of Crater Lake and Wizard Island taken from the visitor overlook on The Watchman by Peter Dartnell. Digital

photographic processing by Eleanore Ramsey. View is to the east.

WIZARD ISLAND

979

FIGURE 2 Digital perspective view of generalized geologic map of the lake floor draped over shaded-relief image of 2-m bathymetry (after Ramsey

et al., 2003). Light greens in foreground highlight lava flows and breccia slopes of Wizard Island volcano that are obscured beneath Crater Lake.

subaerial lava to subaqueous breccia is called a passage zone. The presence of passage zones on Wizard Island volcano indicates that the edifice was actively growing as Crater Lake was filling. Preservation of successive passage zones on the flanks of Wizard Island that would otherwise have been overridden by younger lava implies changes in the source vent location for the edifice or a decrease in eruption rate in comparison to lake-level rise. ECOLOGY OF WIZARD ISLAND

Wizard Island is separated from the crater walls on its west side by narrow Skell Channel, only about 300 m

wide and less than 100 m deep (Fig. 3). The channel is easily breached by avian and aquatic fauna as well as the seeds of vascular plants that are blown to Wizard Island by the wind or carried there by seed-eating birds. Yet only around 100 species of the nearly 600 vascular plant species found in the rest of Crater Lake National Park are found on Wizard Island. Unstable or youthful substrate development on a relatively young volcano, temperature extremes, a short growing season, and soil moisture availability limit plant development and stratify plant communities on the sparsely vegetated island. FIGURE 3 Shaded-relief image

of the lake floor color-coded by depth (after Gardner et al., 2001). Narrow Skell Channel separates Wizard Island from the caldera walls.

980

WIZARD ISLAND

Limitations on the vascular plant community are reflected in the wildlife found on the island, especially the birds. Absent or scarce are birds that feed in meadows, such as juncos, robins, and sparrows, because of the lack of substrate suitable for the development of meadow flora. Common are forest-inhabiting species such as Clark’s nutcracker, Stellar’s jay, nuthatches, and chickadees. The island is forested by eight conifers: mountain hemlock, Shasta red fir, lodgepole pine, western white pine, ponderosa pine, sugar pine, subalpine fir, and whitebark pine. The hemlocks and firs dominate the lower flanks of the island, where soil moisture is greatest, whereas the crater rim is dominated by whitebark pine. Harsh conditions at the crater rim contribute to high mortality rates in whitebark pine and may leave the stressed trees susceptible to attack by the mountain pine beetle, dwarf mistletoe, and white pine blister rust, which have reduced whitebark pines on Wizard Island in recent years. Herbs and shrubs on the island exhibit similar stratification, with Davis’s knotweed being the most abundant plant on the unstable cinder slopes, but being conspicuously absent from the lower flanks, where Crater Lake currant thrives. The flowering vegetation on the island feeds nectar-eating rufous hummingbirds. Wizard Island is home to many insects, including ants, bees, butterflies, dragonflies, and spiders. Frogs and toads also inhabit the island, as do garter snakes. Mice, pikas, minks, golden-mantled ground squirrels, two species of chipmunk, and bats are the only mammals living on Wizard Island. Deer and bears have been reported to have visited the island. Non-winged mammals most likely arrived on the island by swimming across Skell Channel

or by crossing to the island by land during a time of lower lake level when Wizard Island was erupting subaerial lava flows across Skell Channel. SEE ALSO THE FOLLOWING ARTICLES

Eruptions / Frogs / Insect Radiations / Lakes, as Islands / Lava and Ash FURTHER READING

Applegate, E. I. 1934. The flora of Wizard Island. Crater Lake Nature Notes 7: 7–8. Bacon, C. R., J. V. Gardner, L. A. Mayer, M. W. Buktenica, P. Dartnell, D. W. Ramsey, and J. E. Robinson. 2002. Morphology, volcanism, and mass wasting in Crater Lake, Oregon. Geological Society of America Bulletin 114: 675–692. Bacon, C. R., and M. A. Lanphere. 2006. Eruptive history and geochronology of Mount Mazama and the Crater Lake region, Oregon. Geological Society of America Bulletin 118: 1331–1359. Campbell, B. 1934. The birds of Wizard Island. Crater Lake Nature Notes 7: 6. Count, E. W. 1932. Wizard Island. Crater Lake Nature Notes 5: 4–5. Gardner, J. V., P. Dartnell, L. Hellequin, C. R. Bacon, L. A. Mayer, M. W. Buktenica, and J. C. Stone. 2001. Bathymetry and selected perspective views of Crater Lake, OR. U.S. Geological Survey Water Resources Investigation Report 01-4046. Huestis, R. R. 1937. Mammals of Wizard Island. Crater Lake Nature Notes 10: 35–36. Jackson, M. T., and A. Faller. 1973. Structural analysis and dynamics of the plant communities of Wizard Island, Crater Lake National Park. Ecological Monographs 43: 441–461. Jones, J. G., and P. H. H. Nelson. 1970. The flow of basalt lava from air into water: its structural expression and stratigraphic significance. Geological Magazine 107: 13–19. Nathenson, M., C. R. Bacon, and D. W. Ramsey. 2007. Subaqueous geology and a filling model for Crater Lake, Oregon. Hydrobiologia 574: 13–27. Ramsey, D. W., P. Dartnell, C. R. Bacon, J. E. Robinson, and J. V. Gardner. 2003. Crater Lake revealed. U.S. Geological Survey Geologic Investigations Series I-2790.

WIZARD ISLAND

981

Z ZANZIBAR N. D. BURGESS University of Cambridge, United Kingdom

R. A. D. BURGESS Cambridge Regional College, United Kingdom

The tropical archipelago of Zanzibar comprises two large islands and 53 smaller ones. It is located in the Indian Ocean at longitude 39° E and latitude 6° S. The larger island has traditionally been referred to as Zanzibar Island but is known locally as Unguja. It is 35 km offshore of the Tanzanian mainland, from which it is separated by the Zanzibar Channel, and it is distinct from its sister island of Pemba, which is located further north across the deep-water Pemba Channel (Fig. 1). These islands are a part of a global center of species diversity in the terrestrial and marine realms, with significant local endemism. This diversity reaches its peak in the forests on land and in the coral reefs in the ocean. Species diversity is strongly influenced by the process of island formation and a long history of human habitation, land holding, development, and use. These islands’ historical importance as an entrepôt and their contemporary engagement in tourism defines their remaining habitats, endemic species, and natural resources and will influence their survival. BACKGROUND

Land coverage of Unguja (hereinafter referred to as Zanzibar) is 1666 km2; the main settlement area (Zanzibar Stone Town), developed around a sheltered harbor on the west, is now a World Heritage site. The main language spoken is Kiswahili, with English used in commerce and business

982

FIGURE 1 Map indicating the position of Unguja (Zanzibar) and Pemba

Islands off the coast of mainland Tanzania and showing the location of the key sites for conservation and human history on these islands.

organizations. The origin of the name is disputed but could well have come from the Persians (Zanzibar meaning the “coast land of the black”; Zenj or Zangh meaning Negro), although Omani Arabs state that Zayn Zal Barr (“Fair is the Island”) is more apt. The name could also be a reference to the production of ginger (genus Zingiber) traditionally grown on this “spice” island. There is no doubt of Zanzibar islands’ importance as a main trading port throughout history on the East African equatorial coast. The official human population (2002 census) is 981,754; of these, 97% are Muslim (predominantly Sunni), and the remaining 3% are Christian, Hindu, or Sikh. Zanzibar’s people, with their distinct features, owe much to Arab, Indian, and African influences. The Swahili language is primarily a blend of African tribal languages, especially Bantu, but around onethird of its words are derived from Arabic. GEOLOGICAL HISTORY

Zanzibar is composed of uplifted sedimentary rocks, including coral reef limestone, and marine and fluviatile sediments. The uplifting that formed Zanzibar Island is associated with the rifting of East Africa, which started in the Middle Tertiary, some 30 million years ago. The sea level of eastern Africa has fluctuated considerably since then, with low sea levels at the end of the Oligocene around 26 million years ago and in the past 1 million years, when Zanzibar was connected to the mainland as recently as 100,000 to 10,000 years before present. Since the last period of uplifting, the coral limestone has been eroded by rainfall, creating karst features, such as underground caverns, sinkholes, and jagged rocky surfaces. CLIMATE

Zanzibar has a tropical oceanic climate with little seasonal variation in day length (11–13 hours per day), temperature (between 15 and 39 °C, with a mean of 27 °C), or annual average humidity (71%), but with significant variation in rainfall, due to the northern and southern movements of the inter-tropical convergence zone (ITCZ), bringing rain in November and December (short and lighter rains) and March to June (long and heavier rains). Annual rainfall is between 800 and 1600 mm per annum, and although most falls during the rainy seasons, tropical storms can occur in any month.

bar–Inhambane regional mosaic, which extends from southern Somalia through the coastal regions of Kenya and Tanzania into Mozambique. This distinct biogeographical region has been defined scientifically as a center of global endemism for plants and animals and has thus been prioritized for conservation attention by international and national conservation organizations. Some species in this biogeographical region have affinities to those further west in the Congo basin and in West Africa; others have affinities with other Indian Ocean Islands, but there are also many local endemics. Species with affinities to those in the western and central African rain forests indicate the former existence of a tropical forest belt across the entire African continent; this belt has now disappeared in East Africa as the forest has largely been replaced by savanna–woodland habitats except along the moister coast. Species with affinities to those of other Indian Ocean islands are often mobile, such as fruit bats or birds, or are plants that have been dispersed either by the ocean or perhaps by people. Zanzibar would have originally been covered by various types of forest habitat. On the deeper and better soils, higher-canopy eastern African coastal forest would have predominated, with stunted scrub (coral rag) forests being found on the outcropping limestone and along the exposed eastern seaboard, and swamp forests occurring in freshwater wetlands. In sheltered marine bays, mangroves would have dominated. The most biologically important habitats are the forests. However, the coastal forest habitats of Zanzibar have fairly low biological values when compared to the similar habitats on the Tanzanian mainland, or those of Pemba Island. The flora is rather impoverished, with four island endemic plants and few regional endemics. The island supports around 2500 individuals of the unique Zanzibar red colobus monkey (Procolobus kirkii) and under 1000 individuals of the dwarf antelope Aders’s duiker (Cephalophus adersi), which is also known from coastal Kenya. Three butterflies are also confined to this island, as is the newly described amphibian Kassina jozani, and there are four endemic subspecies of birds (three shared with other islands) and mammals (including the Zanzibar leopard, which has not been seen in recent years). Marine

BIOGEOGRAPHY AND BIOLOGICAL IMPORTANCE Terrestrial

The islands offshore of the Tanzanian mainland, including Unguja and Pemba, are part of the so-called Zanzi-

The marine habitats of Zanzibar are a part of an eastern African marine region recently named the East African Marine Ecoregion, which extends from Somalia through Kenya, Tanzania, Mozambique, and northern South Africa. It is the most species-rich marine region of Africa

ZANZIBAR

983

and one of the most diverse marine areas in the world, but there are only a few locally endemic marine species. The main oceanographic influence on Zanzibar is the South Equatorial Current, which hits the East African mainland in northern Mozambique and then flows north as the East Africa Coastal Current and south as the Mozambique Current. The seas around Zanzibar and Pemba contain a variety of habitats. These include expansive areas of coral reef and seagrass habitats, interspersed with areas of sand and mud. These marine habitats are among the most diverse in Africa and are part of one of the global centers of coral reef diversity. The waters around Zanzibar also support populations of green turtle (Chelonia mydas), hawksbill turtle (Eretmocheles imbricata), loggerhead turtle (Caretta caretta), olive ridley turtle (Lepidochelys olivacea), and leatherback turtle (Dermochelys coriacea). There are also healthy numbers of sailfish, blue and black marlin, dolphins, and migrating humpback whales. HUMAN HISTORY

Archaeological excavations in Unguja Ukuu, in the south of Zanzibar island during the early 1990s, date evidence of settlement occupation (glass beads, pottery, worked metal objects) from between AD 400 and 600. Seen in the wider context of East Africa, Zanzibar Island may have been settled for millennia by the mainland African tribes especially the Hadimu and Tumbatu. It is known that the West Central African tribe, the Nyamwezi, pioneered routes to the Zanzibar coast and were very important carriers and traders, as documented early in the nineteenth century. By the mid-nineteenth century, Zanzibar Arabs financed by Indian traders had monopolized the trade in ivory and slaves and exchanged guns, small arms, cloth, and manufactured goods to areas beyond the shores of Lake Tanganyika and Lake Victoria. Zanzibar’s location was the prime reason for its increasing importance as a trading center. It is claimed that the Zanzibar Island was known to the classical world; the Greeks referred to it, and Roman merchant seaman sailing under Arab captains are thought to have provisioned and exchanged goods circa AD 80. The Omani Arabs from Muscat are documented as having arrived in the eighth century, and there is a tradition that relates how Abi Ben Sultan Hazan of Shiraz in Persia sailed from Bushehr in the Persian Gulf with seven dhows and landed unexpectedly on Zanzibar in 975. The Portuguese (Vasco de Gama, 1499) visited Zanzibar on the journey from the Cape of Good Hope and were escorted by Arabs on the trade route to India. It is

984

ZANZIBAR

documented how they were astounded at the wealth of the Zanzibaris; from then on, the Portuguese began to dominate the trade routes by constructing forts along the mainland coast and on Zanzibar during the sixteenth and seventeenth centuries. It is also apparent that the Chinese traded with Zanzibar from well before this time. Later, the East Indian Trading Company capitalized on the knowledge of trade in this area. Philanthropic and Christian efforts to stop the slave trade led to legislation by 1822, but this remained ineffective until the end of the nineteenth century, when this trade was finally halted. Contemporary evidence of this trade, which centered on the market in Stone Town, is now an integral part of cultural tourism of the area. Historically, goods were traded to and from Zanzibar by sailing boats that took advantage of the seasonal wind patterns (trade winds or monsoons; Swahili: kazkazi), using the island as an emporium for various commodities and goods. From November to March, the hot northeast monsoon blows down the East African coast from Asia, a weather pattern that assisted sailing passage for oceangoing dhows from Aden, Hadhramaut, Muscat (Persia), and India laden with barter goods. At the end of March, the winds change to the southwest, and this assisted ships bound back to these ports laden with ivory, timber, charcoal, building poles, gum copal, spices, grain, and slaves. This trade, and the need for agricultural lands, decimated the forests of Zanzibar and also resulted in the depletion of elephant populations and forest lands well into the Tanzanian mainland. Even today, Zanzibar is a major transit port for logs cut from the coastal forests and woodlands of the mainland, for charcoal burned from the same forests and woodlands, and for building poles cut from mangroves. The explosion of interest in natural sciences and exploration of mainland Africa by colonial powers in Europe and America (particularly Johann Krapf, David Livingstone, Richard Burton, John Speke, and Henry Morton Stanley) meant that in the late nineteenth century, Zanzibar was unrivalled in its importance as a starting and finishing point for expeditions into East and East Central Africa. A symbiosis developed between the Sultans of Zanzibar, the traders such as Tippo Tip (with their caravans, porters, and the knowledge of local tribes and provisioning skills), and the explorers. This cooperation led to the international advances in knowledge of the natural science of Zanzibar and mainland Africa and the economic resources of this region, and it also furthered the work of missionary societies in converting mainland Africans to the Christian religion.

Archaeological evidence from existing buildings illustrates many facets of Zanzibar’s existence and development: the Shirazi Dimbani Mosque ruins at Kizimkazi dated 1107, the Old Slave Market and Anglican cathedral, Livingstone House, Kirk House, Guliani Bridge, Dhow Harbour, the Beit el Ajaid (House of Wonders), Beir al Sahel (People’s Palace), Jamhuri Gardens, Mbweni Palace ruins, Portuguese Mvuleni ruins, Hamamni Persian baths at Kidichi, and the Bububu Railway. All point to the importance and wealth of this area, which was the greatest exporter of slaves, ivory, and cloves.

Up until the early 1990s, Zanzibar’s separate status ensured that mainland Tanzanians could visit only with a passport and could not own property; now there is a greater influence from the Tanzanian mainland. However, Zanzibar is entitled to keep all its foreign exchange earnings, and so it remains, in part, economically independent. It is only in the past decade that conservation has become a focal area of work for the Zanzibar government, with a network of protected areas being developed to cover marine and terrestrial habitats and key endemic or severely threatened species.

POLITICAL HISTORY AND LEGAL FRAMEWORK

CURRENT ECONOMY

The political history of this island has influenced the development of protected areas for the conservation of the biodiversity (marine and terrestrial, especially forests) of Zanzibar. The dominance of the sultanate, the importance of trade, and the emergence of the main European powers in the late nineteenth century (French, German, and British) with the scramble for economic resources from Africa led to Zanzibar’s status of British protectorate in the 1890s. The existence of the protectorate ensured that the British resident of Zanzibar was responsible to the London-based Colonial Secretary, a system that lasted until Zanzibar obtained independence in 1963. During this period, no protected areas were established on Zanzibar, whereas many were declared on mainland Tanganyika. On December 9, 1961, Julius Nyerere obtained independence for Tanganyika (the mainland), and there was a growing political awareness on Zanzibar. On January 12, 1964, the government was overthrown, and this resulted in the flight of the Sultan (Jamshid ibn Abd Allah). By April 26, 1964, the Revolutionary Government of Zanzibar signed an Act of Union with Tanganyika to form a united republic; this was later renamed the United Republic of Tanzania, and the Revolutionary Government of Zanzibar is an integral part. However, the Revolutionary Government of Zanzibar has a distinct and separate legal system with its own assemblies, and it shares only the Court of Appeal of the United Republic with mainland Tanzania. Because of longstanding agreements, it is represented by an entitlement of 30% of the seats at the Union Assembly. Zanzibar’s high court, zadhis courts, and magistrates courts serve the five district areas, which are further subdivided into ten administrative districts, including Pemba. The president of Zanzibar is also the executive of the United Republic of Tanzania and has several distinct political and statutory roles.

The economy of Zanzibar is dominated by agriculturalrelated activities, but tourism has developed rapidly in recent years. Local people rely on subsistence agriculture and on harvesting natural resources for survival: wildlife, forestry, and fishing. Commercial crops include spices (cloves, vanilla, cinnamon, peppers, nutmeg, ginger) and tropical fruit (especially coconuts). There is some smallscale manufacturing, especially for the tourist trade and for trade with the mainland. Zanzibar’s exotic image, location, beaches, ocean, wildlife, and history have led to its growing importance for international tourism (particularly ecotourism) and associated services in retail and hospitality, located mainly in Stone Town and the coastal fringes. Current estimates (2006) place the number of visitors in excess of 100,000 per year, of which more than 20% visit the Jozani– Chwaka Bay National Park. Tourism activities include deep-sea fishing (especially for tuna, marlin, and shark), wildlife watching (dolphins, Zanzibar red colobus, and birds), and excursions to neighboring islands. Zanzibar is developing its tourism experience to include visiting its protected areas, globally important habitats, and unique species. This strategy aims to give value to these areas and thus ensure their conservation in the long term. HUMAN IMPACTS ON HABITATS AND SPECIES

Over the past 2000 years of human expansion, natural habitats have been reduced to fragments and replaced by areas cultivated for food production and spice export. Zanzibar and Pemba Islands are famous for their plantations of spices such as cloves, cinnamon, nutmeg, pepper, ginger, and cardamom and tree crops such as coconut, jackfruit, banana, and citrus. Today, the largest patches of natural habitat on Zanzibar are found in the Jozani–Chwaka Bay region (southwest), in the Kiwengwa forest (northeast), and in areas of coral rag thicket as these cannot be cultivated. Because of the small remaining areas of habitat and

ZANZIBAR

985

986

the high levels of hunting, which have affected some species, populations of several of the key species on Zanzibar are either threatened by extinction or declining.

SEE ALSO THE FOLLOWING ARTICLES

CONSERVATION

FURTHER READING

Forest conservation on Zanzibar is the responsibility of the Department of Commercial Crops and Forestry, and marine reserves are managed by the Department of Fisheries and Marine Resources. The protected area network is newly established and still expanding. It includes two marine reserves (Menai Bay, 476 km2, and Mnemba Island, 0.6 km2), one national park (Jozani–Chwaka Bay, 50 km2), one forest reserve (Kiwengwa–Pongwe, 32 km2), and one private island/marine reserve (Chumbe, 0.3 km2). There have also been efforts to conserve the remaining populations of Zanzibar red colobus and Aders’s duiker that are found in farmland areas unprotected by official reserves.

Burgess, N. D., and G. P. Clarke. 2000. The coastal forests of eastern Africa. Cambridge, UK: IUCN Forest Conservation Programme. Burgess, N. D., J. D’Amico Hales, E. Underwood, E. Dinerstein, D. Olson, I. Itoua, J. Schipper, T. Ricketts, and K. Newman. 2004. Terrestrial ecoregions of Africa and Madagascar: a continental assessment. Washington, DC: Island Press. Dale, G. 1969. The peoples of Zanzibar: their customs and religious beliefs. New York: Negro Universities Press. Horton, M., and K. Clark. 1985. Archaeological survey of Zanzibar. Azania: The Journal of the British Institute in Eastern Africa 20: 167–171. Ingrams, W. H. 1967. Zanzibar, its history and its people. London: Frank Cass. Pakenham, T. 1991. The scramble for Africa. London: Abacus Books. Sheriff, A. 1987. Slaves, spices and ivory in Zanzibar. London: James Currey.

ZANZIBAR

Archaeology / Coral / Exploration and Discovery / Indian Region / Marine Protected Areas / Missionaries, Effects of

GLOSSARY

The glossary that follows defines over 900 specialized terms that appear in the text of this encyclopedia. Included are a number of terms that may be familiar to the lay reader in their common sense but that have a distinctive meaning within these fields of study. Definitions have been provided by the encyclopedia authors so that these terms can be understood in the context of the articles in which they appear. Basaltic lava flow of typically higher viscosity, which cools to produce a very rough surface. (A term originating from the Hawaiian language.) abiotic disturbance A non-living cause, such as wind or earthquake, that results in a change in the environment (perturbation). aboriginal introduction A plant or animal intentionally or unintentionally introduced to an area by the original colonizers (as in the case of Oceania, by the Micronesians, Melanesians, and Polynesians). accretionary wedge/prism/complex A mass of sediments that is scraped off the subducting plate at a convergent plate boundary and accreted or stuck to the non-subducting plate to form a mound-shaped mass. adaptive radiation Evolution of ecological diversity within a rapidly multiplying lineage. adaptive trait A characteristic of an organism that allows it to maximize its fitness. adventive species Species in a particular location that arrived from elsewhere. agamids A family of lizards that is spread worldwide and occurs especially in tropical and subtropical regions; mostly characterized by their very long legs and fast movement. aggressiveness Of plant pathogens, the ability to invade and establish within the host fitness. ‘a‘a ¯

See NON-INDIGENOUS SPECIES. alkaline magmas Magmas that are relatively poor in silicon and rich in sodium and potassium. allele One of two or more alternative forms of a gene. allelic drift See GENETIC DRIFT. allochthonous Formed elsewhere than its present place. allopatric Occurring in separate, non-overlapping geographic areas. allopatric speciation/allopatry The formation of new species that results when an extrinsic geographic barrier (e.g., a mountain or a sea strait between islands) prevents the interbreeding of two or more populations of a species. allopolyploidy The condition of having two or more sets of chromosomes as a result of combining genomes from evolutionarily distinct lineages. allospecies A pair of species that are each other’s closest relatives occupying non-overlapping (i.e., allopatric) distributions. alpine tundra An ecozone that does not contain trees because the climate (generally with no month having an average temperature in excess of 10 °C) limits tree growth. altimetry Satellite-borne radars measuring small, permanent sea-surface topography caused by the gravitational attraction of the seafloor relief. These undulations may be inverted to yield information about seamount shapes. ambergris A fatty substance secreted in sperm whale intestines. After long exposure to seawater and sunlight, it develops a dark gray color and pleasant odor and may wash up on island shores. Because it retains fragrance molecules, ambergris has been used since antiquity as a basis for perfumes. ammonoids A group of extinct (since the end of the Cretaceous) cephalopod molluscs with an external shell alien species

987

divided internally into chambers that control flotation; related to the nautilus. amphidromic point A location within a tidal system where the tidal range is near zero. Amphidromic points occur because the Coriolis effect alters current paths on a rotating Earth and because the topography of oceanic basins interferes with the free propagation of the tidal wave. The tidal wave pattern within an oceanic basin or bay rotating around an amphidromic point tends to have the appearance of spokes in a wheel. AMS radiocarbon dates A method of radiocarbon dating in which an accelerator-based mass spectrometer is used to count all radiocarbon atoms found in a sample, thereby reducing the amount of material needed for a radiocarbon date. amygdule A secondary deposit of minerals found in spherical, elongated, or almond-shaped cavities (vesicles) in igneous rock. The cavities are created by the expansion of gas bubbles or steam within lava. anadromous Describing or referring to a life cycle in which fish breed in freshwater, and the progeny run to the sea where they live most of their life before returning to freshwater to breed. Also, SEA-RUN. anagenesis Speciation by gradual genetic and morphological divergence due largely to genetic drift. anastomosing Connecting in a network pattern, as, for example, branches of rivers, leaf veins, or blood vessels. anatexis The partial melting of rock in the Earth’s crust. anchialine Describing or referring to marine or brackish water bodies that lack surface connections to the sea but with subsurface hydrologic connections; influenced by both marine and terrestrial ecosystems. andesite A gray to black volcanic rock with between about 52 and 63% silica (SiO2); produced at subduction zones by melting of wet mantle. angiosperms Flowering plants. anhydrobiosis See DESICCATION TOLERANCE . ankaramite A basalt containing abundant (more than 10%) olivine and clinopyroxene. Anomura An infraorder within the suborder Pleocyemata of the order Decapoda (crabs and shrimps). anoxic Lacking oxygen. Antarctic polar frontal zone/Antarctic convergence A circumpolar oceanic front in the Southern Ocean where cold, dense, north-flowing Antarctic waters sink beneath the relatively warm, less dense subantarctic waters. The zone is up to 100 km wide and variable in latitude. anthropogenic Derived from human activities, as opposed to effects or processes occurring in natural environments without human influences.

988

GLOSSARY

A convex fold in rock, the central part of which contains the oldest section of rock. anurans The group of amphibians that includes frogs and toads. aphotic Lacking light. apomictic Of plants, producing seeds without pollination. apozooxanthellate Of corals, ordinarily possessing zooxanthellae but having lost them as a result of environmental stress. apterous Lacking wings. arboreal Living in trees. arc-continent collision A condition in which two tectonic plates, one of continental lithosphere and the other of volcanic arc lithosphere, converge and, because neither can be subducted, collision ensues. archaeology The study of past lifeways by analysis of material remains. archaeophyte A plant species that is not native to a region but that was introduced prior to AD 1500. archipelago A group of geographically or geologically related islands. Arctic The area north of the northern tree-line or north of the 10 °C July isotherm. arc-type volcanic rocks Volcanic rocks that have the physical and chemical features characteristic of arc volcanoes. arc volcanism A process creating arcuate island chains at the boundaries of convergent tectonic plates. Area of Special Scientific Interest (ASSI) See SITE OF SPECIAL SCIENTIFIC INTEREST (SSSI) . Armorica A Roman name for the Brittany region of northwestern France. arthropods The most diverse animal phylum (Arthropoda), which includes segmented animals with a chitinous exoskeleton and a series of paired articulated appendages. artifact Any object made by humans, such as a tool or piece of pottery. Artiodactyla Even-toed ungulates (hoofed animals). Members of this order, also known as the cloven-hoofed animals, have two hooves on each foot. Examples are goats, pigs, deer, and cows. ash In volcanology, particles of volcanic glass, crystals, and rock that are less than 2 mm in size and are either ejected from a volcanic vent or formed by secondary fragmentation of volcanic material. asthenosphere A weak layer of solid but ductile rock underlying the lithosphere, and on top of which the Earth’s lithospheric tectonic plates move. It extends from the bottom of the lithosphere to depths of several hundred kilometers into the Earth’s upper mantle. anticline

A ring-shaped oceanic reef formation surrounding a lagoon, caused by coral reef growth around a volcanic island that subsides into the ocean. Australasian Pertaining to Australasia, the area including Indonesia, Papua New Guinea, and Australia, where Asian faunas may mix with Australian faunas across Wallace’s Line. Australo-Melanesians A diverse group of peoples with shared physical traits presently found in the Andaman Islands, in parts of Southeast Asia, in Australia, and in Melanesia. They include Australian aborigines, Melanesians, and Negrito populations scattered across the Philippines, the Malay Peninsula, and the Andaman Islands and are believed to be the descendants of the first wave of modern humans to populate the region, arriving from Africa via the southern coast of Asia around 65,000 years ago. Austronesian 1. The family of languages (including more than 1200 modern languages) that extends from Madagascar to Easter Island and includes all of the Polynesian, Micronesian, island Melanesian, and Indonesian languages; believed to have originated in Taiwan. 2. A person who speaks any of the Austronesian languages. authigenic carbonate Carbonate rock created in situ through chemosynthetic activity associated with cold seeps. autochthonous Formed in the place where it is currently found. Autochthonous species do not result from dispersal, but rather, have evolved in situ where they exist today. avifauna The birds of a given region. azooxanthellate Of corals, naturally without zooxanthellae; non-photosynthetic. back-arc basin A small oceanic basin on the side of an island arc opposite to the trench and subduction zone; commonly a site for minor seafloor spreading. Also, RETRO - ARC BASIN . backcross A cross between a hybrid and one of the parental types from which it was formed. Baker’s Law The theory that it is more likely for selfcompatible than self-incompatible species to establish sexually reproducing colonies after long-distance dispersal. (Formulated by American evolutionist G. Ledyard Stebbins and named for British ecologist Herbert G. Baker.) balsa 1. A Spanish colloquial term for a species of tree (Ochroma lagopus) with extremely light wood. 2. A raft made from balsa wood. 3. A pond or pool. barrier reef A reef that is separated from the shore by an open-water lagoon. atoll

A common gray to black volcanic rock with less than about 52% silica (SiO2); produced when magma erupts into either air or water. basanite Dark-colored lava containing the minerals olivine, pyroxene, calcium-feldspar, and nepheline; a type of alkaline basalt. basement Underlying or deeper rocks; a term used to distinguish cover rock sequences from underlying rocks. Typically, basement rocks are igneous and metamorphic rocks found beneath a sedimentary cover. basin A feature of seafloor topography; a low part of the lithosphere lying between continental masses. bathymetry 1. The measurement of underwater depth or topography. 2. Terrain, or relief, of the seafloor. beach nourishment An engineering technique used to slow beach erosion that involves placing a significant volume of sand derived from elsewhere on the eroding beach. beach profile A cross section of the beach surface, commonly used to compare different beach types. Repeated beach profile surveys are one method used to measure changes in beach morphology. beach ridge A relict inland shore ridge of a similar alignment to the modern beach. Beach ridge complexes are indicative of prograding coastlines. beachrock A sedimentary rock formed in the intertidal zone through cementation of calcareous grains and skeletal fragments. Benioff (or Benioff–Wadati) zone A deep active seismic area in a subduction zone. Differential motion along the zone produces deep-seated earthquakes, the foci of which may be as deep as ~700 km. They develop beneath volcanic island arcs and continental margins above active subduction zones. Also, WADATI-BENIOFF ZONE. benthic Of or relating to the sea floor, including organisms living in or on the seabed. Bergmann’s Rule An ecogeographic phenomenon predicting that within species of homeothermic vertebrates (e.g., birds and mammals), large-bodied individuals will inhabit higher latitudes, or colder areas, whereas smaller conspecifics will inhabit lower latitudes, or warmer areas. (Formulated in 1847 by German biologist Christian Bergmann.) berm A shore-parallel ridge or platform on the upper beach that marks the break in slope between the foreshore and backshore. biodiversity The variety of living things in an area, measurable from genetic to ecosystem levels. biodiversity coolspot An area (such as an atoll or other small island) that has very limited biodiversity inheritances, little or no endemism, and often basalt

GLOSSARY

989

with high cultural dependence on this limited biodiversity. biodiversity hotspot An area with exceptional endemism of plants and/or animals as well as significant levels of habitat loss, as first defined by British ecologist Norman Myers in 1988 and 1990. Thirty-four hotspots are now recognized worldwide. biogeographic disjunction The discontinuous distribution of closely related taxa. biogeography The study of the geographic distribution of organisms. bioherm A carbonate rock formation consisting of the fossilized remains of marine organisms such as coral, algae, and molluscs. biological control/biocontrol The use of living organisms to control pest species or diseases. biological species Groups of interbreeding populations that are reproductively isolated from other such groups, allowing them to evolve independently and accumulate trait differences. bioluminescence The production of light by an organism through a chemical reaction, usually involving a protein (luciferin) and a catalyst (luciferase). biomass The total mass of living matter, such as fish in a stock, within a given area; a biological measurement used to establish the importance of certain groups of living things in an ecosystem, as opposed to numbers of individuals. biosecurity The protection and maintenance of ecological integrity. biosphere reserve A UNESCO designation for a dedicated conservation area. biota The entire assemblage of plants, animals, and other living organisms in a given region. biotic disharmony A condition in which major highorder groups of plants or animals are absent from a particular biota; observed frequently in islands. biotroph An organism that feeds on living hosts. blackbirding Recruitment of laborers through kidnapping and trickery, especially for work on sugar plantations in Queensland and Fiji and in Peruvian mines; a practice that occurred primarily in the last half of the nineteenth century. bleaching Loss of color in corals as they expel their symbiotic algae (zooxanthellae) under stress from increased water temperatures or changes in salinity, light availability, or sedimentation. blueschist Metamorphic rock containing the mineral glaucophane; produced at low temperatures and high

990

GLOSSARY

pressures when basalt or gabbro is subducted into the Earth’s mantle. bottleneck See GENETIC BOTTLENECK . bovids Cloven-hoofed mammals of the family Bovidae, which includes cattle, sheep, goats, and buffaloes. brachypterous Having small, nonfunctional, straplike wings. Brachyura An infraorder within the suborder Pleocyemata of the order Decapoda (crabs and shrimps). breccia A coarse-grained rock composed of angular rock fragments held together by a fine-grained matrix or cement. breeding migration Migration by terrestrial crabs en masse over several kilometers from the land to the ocean for breeding, usually triggered by the first rains of the wet season. broadcast spawning The release of gametes (eggs and sperm) into the water column. Bronze Age A period in human cultural development when the most advanced metalworking led to a bronze alloy by melting copper and tin together, and casting them into bronze artifacts. In Greece and China the Bronze Age began before 3000 BC, in Britain around 1900 BC. bycatch The nontarget species (e.g., unwanted marine species, juveniles of target species, and other marine wildlife, including seabirds) caught in fishing gear. calcarenite A sedimentary rock formed on land by cementation of sand-sized fragments of calcium carbonate, derived by weathering of shells, calcareous algae, and coral. caldera A wide, basin-shaped structure that forms abruptly at a volcano’s summit as a result of collapse following a major eruption of magma. California Current The southward flow of cool water from British Columbia to Mexico, extending offshore up to about 500 km; the current is the eastern edge of the North Pacific Gyre, the clockwise-circulating pool of water occupying most of the northern Pacific Ocean. California Floristic Province A region of Mediterranean climate (winter-wet, summer-dry) and high floristic endemism in western North America (western California and southwestern Oregon in the United States and northwestern Baja California in Mexico). canoe plant A plant, generally a useful one, that was carried by indigenous peoples, usually in a canoe (in the case of Oceania and the Caribbean), for planting in new areas that were being colonized. canopy dieback See FOREST DIEBACK.

A type of scavenging hawk found in South and Central America. carrying capacity The maximum number of individuals that an ecosystem can support. Stocks above carrying capacity decrease in abundance, whereas stocks below it increase in abundance. Caste War A rebellion of the Maya of the Yucatán Peninsula against the economically and politically dominant European-descended Yucatecos. The war lasted from 1847 to 1901, with skirmishes continuing until 1933. As a result, an independent Maya state developed in the southeastern part of the peninsula. catadromous Of fish, living in freshwater and breeding in the sea. catchment An area that collects and drains rain water. Cathaysia The continental region assembled as a single landmass in tropical latitudes during the Carboniferous and Permian. The term was originally used to indicate an area of Sino-Malaya with a distinctive flora that includes Gigantopteris, but is now also used to identify a particular East Asian tectonic block that includes South China. cay A small, low island composed primarily of coral or sand. (Called a “key” in American English.) cay sandstone Lithified reef island sediments cemented by phosphate-rich cements, commonly associated with bird droppings (guano). cenote A natural freshwater-filled sinkhole or dolina of karst origin. These geomorphologic structures are typical of limestone platforms in the Yucatán and are ritually significant to the Maya, representing the passage to the underworld. (From the Maya dzonot, meaning “sacred well.”) Cenozoic The most recent geological era, extending from ~65 million years ago to present; includes the Tertiary and Quaternary epochs. centrifugal force A pseudo- or “fictitious” force that appears when a rotating reference frame is used for analysis of motion. The (true) frame acceleration is substituted by a (fictitious) centrifugal force that is exerted on all objects and directed away from the axis of rotation. Chamorros The native Pacific island people of Guam. chaparral The often dense evergreen plant community of drought-adapted shrubs and scrub oaks typical of Mediterranean-type climates in California and northern Baja. Similar growth forms occur in all Mediterraneantype climates. character displacement An effect of competition or predation that results in morphological change within animals. caracara

The anteriormost pair of appendages in a spider, each one comprising a large basal part and a fang used to inoculate venom produced by modified salivary glands. chemoautotroph An organism that uses chemosynthesis to make organic matter. chemosynthesis The biological conversion of carbon dioxide to organic matter, as in photosynthesis, but using sulfide or some other inorganic molecule as the energy source, rather than sunlight. chicle The natural sap of the chicozapote or sapodilla tree (Manilkara zapota; Sapotaceae), obtained in tropical rain forests of Mexico, Belize, and Guatemala by repeated tapping of trees. Chicle was the original source of chewing gum, replaced now by synthetic gum. chuckawalla A large, herbivorous desert lizard found in the southwestern United States and in Baja California. cinder cone A conical hill formed by accumulation of solidified bubble-rich droplets and clots of lava that fall around the vent during a single eruption. circulation In the ocean, the movement of water masses over a reef system. circumboreal Referring to plant or animal distributions at high, subarctic latitudes across both North America and Eurasia. cirque A deep, steep-walled, half-bowl-like recess, located high on the side of a mountain. clade A group of organisms descended from a common ancestor (monophyletic). cladistics The classification of species into a hierarchy where groups are characterized by shared characteristics derived from a common ancestor (homologies), optimizing the recapturing of phylogeny (genealogy). cladogenesis Speciation by splitting of the original gene pool and subsequent divergence, as commonly occurs during adaptive radiation in islands. cladogram A branching diagram or family tree that groups species accordingly to recently evolved, shared traits. Only the branching pattern is represented; branch lengths do not represent time. clast A single sediment unit in sedimentary rock. climax For vegetation, the community representing the endpoint of succession in a particular climatic zone. cloud forest A tropical or subtropical habitat type that occurs within a relatively narrow altitudinal zone, generally between 2000 and 3500 m on continents, though reduced to about 1000 m on some islands (e.g., Hawai’i), or even lower on small steep islands (e.g., Kosrae and Pohnpei in Micronesia), and is usually covered in mist and fog. The high moisture level and cool temperatures chelicerae (singular chelicera)

GLOSSARY

991

promote the growth of abundant mosses covering both ground and vegetation. Other characteristics include reduced tree stature and very high endemism. Also, FOG FOREST, MIST FOREST , MOSSY FOREST . Cnidaria A phylum of predominantly marine invertebrate animals, all of which are carnivores; formerly termed Coelenterata. coadapted genetic system The system of tightly linked genes that interact to allow high fitness within a local population. coalescence time The total number of generations back to the common ancestor of gene copies in a present-day sample. coalescent theory or process An approach that models the genealogical history of a set of sampled genes backward through time. Coalescent theory can be applied to many demographic models, including structured populations with migration or populations that change size over time, and can be used to make inferences about the historical processes giving rise to observed data. coastal plain Low-lying and gently seaward-sloping plain along the coast, extending from the sea to elevated land; commonly depositional in nature and formed by shoreline progradation under relatively stable tectonic conditions. coccolithophores Microscopic one-celled marine plants with calcium carbonate plates that live in the plankton. coevolution Evolution of multiple species caused by interactions between the species. cohort A generation of individuals having a statistical factor (such as age or size-class membership) in common in a demographic study. cold seep A seafloor habitat where high concentrations of hydrocarbons are seeping out of the ocean floor; characterized by unusual communities of organisms dependent on chemoautotrophic production. As opposed to hydrothermal vents, cold seeps do not have a temperature anomaly compared to the surrounding water. collective evolution A mechanism through which moderate levels of gene flow between differentiating species may increase the rate of divergence between them. collisional orogenesis The process by which the Earth’s crust is deformed and thickened, often resulting in the formation of mountains, as a result of the collision of continental and/or oceanic crustal material in a subduction zone. colonization The successful establishment of a reproducing population of a species in a new environment. colonization curve The change through time of numbers of species found together on an island.

992

GLOSSARY

Describing or referring to a relationship between two living organisms in which one benefits and the other is not significantly harmed or helped. 2. A member of such a relationship. competitive exclusion An ecological theory stating that two species competing for the same resources cannot stably coexist, and one will therefore exclude the other. composite volcano A steep volcano built by both lava flows and pyroclastic eruptions. Also, STRATOVOLCANO. Conservation International The international environmental non-governmental organization that has identified 34 regions worldwide as biodiversity hotspots, covering 2.3% of the Earth’s surface. conservation of mass A law of physics that describes the balance of mass through convergences and divergences of water movements. conspecific Belonging to the same species. constraint In evolution, any impairment in the anticipated course of evolution as a result of the phylogenetic history of a lineage. In fact, species inherit from their ancestors developmental pathways that predetermine most of their design and functioning in a strict sense. continental crust The basement rock for all continental land areas, normally between 35 and 50 km thick and characterized by granite, an igneous rock that is relatively enriched in the light elements silicon (Si) and aluminum (Al), and is therefore less dense than oceanic crust. continental island An island on the continental shelf that is geologically part of the continent but, as a result of changes in sea level, is surrounded by water; if sea level drops, a continental island can become reconnected to the continental land mass. continental plate One of the basic elements of the plate tectonics hypothesis. These plates underlie continents, whereas oceanic plates underlie major ocean basins. Because the material making up the continental plate contains light elements, it tends to ride up over a converging, heavier oceanic plate, creating young mountain ranges along the line of collision (its leading edge). The trailing edge of the plate is usually sinking slightly, which allows wide deltaic and coastal plains to form along that margin of the plate. continental rocks The mixture of sedimentary, igneous, and metamorphic rocks that make up the continental crust. continental shelf The region of submerged topography surrounding a continental island and leading to deep water. commensal 1.

An actively deforming region where two tectonic plates move toward each other, usually involving one being thrust down into the Earth’s interior and the other being thrust upward. In the oceans, the oldest (and densest) plate will subduct beneath the other plate, resulting in the formation of an island arc. convergent evolution The development of similar adaptations by distantly related species living in different locations under similar environmental conditions. Cope’s Rule A theory that lineages of animals evolve to larger body sizes over their evolutionary histories; hence, descendant species will usually be larger than the species they evolved from. (Named after American paleontologist Edward Drinker Cope.) coral rag Uplifted and eroded coral that supports desiccation-resistant thicket vegetation. Coriolis force In motions observed in a rotating, rather than fixed, frame of reference, an apparent force that causes moving objects on the rotating Earth to be deflected from a straight path. Hence, freely moving objects on the surface of the Earth veer to the right in the Northern Hemisphere and to the left in the Southern Hemisphere. The Coriolis force does not appear when the motion is analyzed in an inertial frame of reference. Cretaceous The geologic time span extending from about 145 million to 65 million years ago. crown-of-thorns A sea star (Echinodermata) that preys on living coral, causing high levels of destruction of coral reefs when in outbreak phase. crust The solid outer shell of the Earth, divisible into oceanic crust and continental crust. cryoprotectors Substances synthesized within the animal body, such as glycerol, that have a protective function at low temperatures. They form stable hydrogen bonds with water molecules and thereby decrease the freezing point of any solution in which they are included. cryptoturnover Turnover of species (extinction followed by immigration) that remains undetected because it occurred between two surveys. cultural inheritance Traits that are learned from other organisms in the course of development and are passed on to future generations by learning without genetic encoding. cumulate An igneous rock formed by the accumulation of crystals from a magma. customary marine tenure The traditional rights of communities to regulate and manage access and use of marine resources. convergent boundary/plate boundary/margin

Light-colored volcanic rock made essentially of plagioclase and lesser quartz plus minor hornblende or biotite, and containing about 63 to 68% silica (SiO2). Dacite generally erupts at temperatures between 800 and 1000 °C. Darwinian fitness See FITNESS. debris avalanche A sudden and rapid movement of rock and other debris, such as vegetation, driven by gravity; a fast-moving debris flow. May result from the collapse of the side of an oversteepened volcano or gravitational collapse of unconsolidated sediments. debris flow A landslide of soil, sediment, or fragmented rock, saturated with water, that moves rapidly and over long distances with pervasive internal deformation and development of flow structures. demersal Found on or near the bottom of an ocean or lake. demographic stochasticity The variation in population growth rates arising from random differences among individuals in survival and reproduction. density compensation On an island, compensation for the absence of mainland species by niche expansions and higher abundances such that the total population density of individuals of all species on islands equals the total mainland densities. density dependence An effect (e.g., per capita birth rate or death rate) that increases in intensity as population density increases. depauperate Having limited biodiversity. depositional coasts Shorelines dominated by deltaic and coastal plains; the location of a majority of the major barrier islands of the world. desiccation tolerance The ability of certain organisms to survive long periods of drought and to rehydrate fully when conditions improve. Also, ANHYDROBIOSIS. deterministic extinction Extinction that occurs when the birth rate of a species becomes less than the mortality rate. Typically, the habitat is destroyed (by humans) or changes (by natural succession), the number of predators or competing species increases, or the species’ food or prey decreases. Devensian The last glacial stage within the Pleistocene epoch of the British Quaternary from about 110,000 to about 10,000 years ago. Equivalent to the North American Wisconsin. diabase A coarse-grained intrusive rock of basaltic composition. diadromous Migrating between freshwater and marine environments during a fish’s life cycle. dacite

GLOSSARY

993

Fragmental rock with angular clasts in a mud matrix interpreted to be formed by rocks dropped from melting ice into marine muds. diaspores Seeds, spores, or buds that can disperse and form new plants. differentiated lavas Lavas formed by partial crystallization of basaltic magma. Segregation of the crystallizing minerals yields a residual (i.e., differentiated) magma. Typically, residual melts have higher SiO2, Na2O, and K2O, but lower MgO than does the parental basalt. Examples of differentiated lavas are hawaiite, phonolite, and trachyte. differentiation series Igneous rocks that are closely linked in space and time and are related by cooling and partial crystallization of a basaltic magma. dike A vertical or steeply inclined sheet of intrusive rock. dipterocarp Any member of the tree family Dipterocarpaceae, characterized by seeds having two “wings,” which enable them to be dispersed by air currents. direct development In frogs, abbreviated or truncated larval development such that embryos develop (usually terrestrially) entirely in the egg capsule and emerge as fully formed froglets (thus omitting the aquatic larval stage or “tadpole” stage of development). disharmony The different balance of species composition on islands when compared to similar patches of mainland. Thus, disharmonic biota, disharmonic fauna, and disharmonic flora are included in this comparison. disjunction Geographically outlying populations, separated from the majority of populations within a species range. dispersal Movement of species away from an existing population, such that the process either maintains or expands the species’ distribution. In many cases organisms have evolved adaptations for dispersal that take advantage of various forms of environmental kinetic energy, such as water flow in rivers or ocean currents, or wind. dispersal ability The ability of a species to diffuse over new habitats. divaricating habit A distinctive, tangled plant growth form consisting of many flexible, interlacing branchlets and small leaves. Some biologists interpret this growth form as a response to moa browsing, although the consensus of evidence points to a climatic reason. divergence The accumulation of differences (in genes and in traits) among isolated lineages resulting from distinct evolutionary trajectories. diamictite

994

GLOSSARY

A boundary where two tectonic plates move away from each other. In oceans, such zones are generally marked by spreading ridges where new ocean floor is created. diversification rate The rate of origin of new, evolutionarily divergent lineages (speciation) minus the rate of extinction of such lineages. diversity The relative degree of abundance of wildlife species, communities, or habitats per given area. DNA (deoxyribonucleic acid) sequence The order of the four nucleotide bases A (adenine), C (cytosine), G (guanine), and T (thymine) within a given strand of DNA. DNA sequencing Analysis of the order of bases of a particular region in order to reconstruct evolutionary relationships and history of the species or group of species. dome Lava extruded as a dome-like feature. dominance 1. In ecology, the ranked order in terms of relative abundance (number or dry weight per unit area, usually in a square meter of habitat) of each species in a community. This ranking is often related to the importance of each species’ role in an ecosystem process. 2. In genetics, a relationship in which one of a pair of alleles suppresses the expression, or dominates the effects, of the other (recessive) allele. Drepanidiinae The subfamily name for the Hawaiian honeycreepers, a single group or clade of evolutionarily related bird species living in the Hawaiian Islands that are derived from cardueline finch ancestors, family Fringillidae. Some authors still prefer to categorize this group as a family, Drepanididae. drift See GENETIC DRIFT. drip pool A small pool underneath dripping water. drusy crystals Minute crystals that form a continuous coating over a surface. dugong A marine mammal closely related to manatees (order Sirenia). dunite A coarse-grained rock consisting primarily of the mineral olivine (Mg,Fe)2SiO4. dynamic equilibrium 1. A condition of balance between opposing forces. 2. In the theory of island biogeography (developed by Robert H. MacArthur and Edward O. Wilson in 1967), the balance between rates of immigration (plus speciation) on the one hand and extinction on the other. Where these rates precisely balance, species richness remains constant through time although species composition is continually varying. early-stage species Species newly colonizing an island, and therefore in the early stages of the taxon cycle. divergent plate boundary

Defined as having a widespread distribution in marginal, coastal habitats. eclogite A type of rock consisting mostly of garnet and pyroxene, formed from basaltic rocks when they are subducted into the Earth’s mantle. ecological niche See NICHE. ecological release Expansion of range, habitat, and/or resource usage by an organism when it reaches a community from which competitors, predators, and/or parasites may be lacking. ecological sorting Determination of the membership of an ecological community through immigration, interaction, and extinction of species. ecomorphology The study or classification of species on the basis of their ecological preferences and overall morphology. Repeated evolution of similar but unrelated ecomorphs on islands is a common theme in literature pertaining to the evolutionary biology of island life. ecomorphs Populations or species whose appearance is determined by ecological factors. ecosystem-based management A process that integrates biological, social, and economic factors into a comprehensive approach to ensure the sustainability, diversity, and productivity of an ecosystem. ecotone A boundary between ecological zones. ecotourism Specialized tourism where the primary attraction is the species or habitats of the area to be visited. ectothermy Reliance by certain animals on the external environment for temperature control, as in most invertebrates, fish, amphibians, and reptiles. edaphic Relating to the soil conditions. edge In ecology, the amount of patch habitat that is in contact with contrasting environments. At the forest edge, there is an effect of physical (e.g., wind, sun) and biological conditions (e.g., exposure to predators and parasites) as compared to those found in the interior of the habitat patch. edge effects Changes in abiotic and biotic conditions along habitat edges. effective population size The number of breeding individuals in a typical population, usually smaller than the actual or census population size; an important statistic in population genetic studies. effectively neutral Not showing an effect from the deterministic nature of natural selection (with beneficial alleles spreading and deleterious alleles declining) because natural selection is swamped by the effects of random sampling.

A volcanic eruption that occurs when magma reaches the Earth’s surface and erupts passively, producing lava flows and lava domes; generally occurs when the gas content of the magma is low. Rocks formed during such an eruption are called effusive rocks. El Niño A sea-surface temperature anomaly of greater than 0.5 °C across the southern tropical Pacific Ocean, occurring at irregular intervals of two to seven years and typically lasting one to two years. As warm water moves from the Indo-Pacific to the eastern Pacific, rainfall and storms increase along the eastern Pacific Ocean margin, and droughts become more prevalent in the western Pacific. (Originally named in the late 1890s by Peruvian fishermen who observed the changes in currents and weather associated with a small, warm coastal current that intensified during Christmas time and so named it “The Little Boy” for the Christ Child.) El Niño Southern Oscillation (ENSO) The combination of the periodic warming of the sea surface in the central and eastern Pacific Ocean, El Niño, and the coincident negatively correlated variations in the sea-level pressures between the eastern South Pacific and the western equatorial Pacific, called the Southern Oscillation. emergence In geology, the exposure of a coastline because of a relative drop in sea level. emerging disease A disease circulating in a natural ecosystem that transfers to humans when the environment is perturbed. endangered species Species that are considered in danger of imminent extinction and are usually formally listed as such by the U.S. Fish and Wildlife Service or other government agencies. endemic Native to, and restricted to, a particular area, such as a mountain, island, or continent; found only there. endemism The fraction of an area’s biota that is found nowhere outside that area. Also, LOCAL ENDEMISM, MICROENDEMISM , REGIONAL ENDEMISM , SHORT - RANGE ENDEMISM . endolithic Living inside rock. endosymbiont Any organism that lives within the body or cells of another organism. endosymbiotic Living inside the cells of another organism, typically in a mutually beneficial relationship. endothermy The ability of some animals, such as mammals, birds, and some insects, to generate their own body heat from internal metabolic reactions. enemy release hypothesis A hypothesis that attributes the success of introduced invasive species to the absence of natural enemies in the introduced range. effusive eruption

GLOSSARY

995

The principle that smaller organisms, with reduced energetic needs, exhibit increased density relative to larger organisms with greater energetic needs. enrichment opportunist An animal adapted to utilize conditions of organic enrichment, typically at the sea floor. Enrichment opportunists typically have high rates of colonization and production and are physiologically tolerant to habitat conditions associated with organic enrichment (e.g., low oxygen concentrations). environmental stochasticity Variation in population growth rates arising from the influence of factors external to the population, such as weather, predation, disease, and competition. Eocene The geologic epoch in the Paleocene period from about 56 million to 34 million years ago. epicenter The point on the Earth’s surface directly above the hypocenter of an earthquake. epidemic disease A disease that develops and spreads rapidly over a short period of time. epikarst The interface zone between soil and rock, characterized by fractures and small solution pockets; an ecotone between surface and subterranean water in karst. epilithic Growing on stone. epiphyte A plant that grows upon another plant, using it for support. epithermal Referring to shallow depths (from surface down to 2 km) in the Earth, characterized by temperatures varying from 50 to 300 °C. epithermal mineralization Mineralization associated with volcanic rocks usually forming within 1–2 km of the surface and typically dominated by gold in quartz vein systems. epizootoic A disease outbreak among wild animals. eradication In pest management, elimination from a site of all individuals of a species. estuary A flooded river valley (lowstand valley) where freshwater and saltwater mix under the influence of rising and falling tides. ethnobiodiversity The knowledge, uses, beliefs, management systems, classification systems (taxonomies), and language that a given culture, including modern scientific culture, has for biodiversity. ethnobotany The study of the relationship between plants and people, including how plants are used and managed as food, medicine, housing materials, cordage, textiles, cosmetics, dyes, and other artifacts and practices that are a part of all cultures. ethology The study of behavior in its natural context. energetic equivalence rule

996

GLOSSARY

The upper layer of the ocean, where sufficient sunlight penetrates to drive photosynthesis. eustatic Referring to global changes in sea level attributed to the ice ages. evolution In biology, the change of inherited traits in successive generations, resulting from natural selection. evolutionary attractor A value toward which things tend to evolve. If there is, for example, an evolutionary attractor at a body mass of 1 kg, then successive generations will be closer to 1 kg than their ancestors were. exotic species See NON-INDIGENOUS SPECIES . explosive eruption A volcanic eruption that involves the rapid expansion of gas, causing the surrounding rock or magma to fragment explosively. There are three types of explosive eruptions: magmatic, phreatomagmatic, and phreatic. extensional tectonism The formation of structures associated with the stretching and thinning of the Earth’s crust. extinction In conservation biology, the total disappearance of all individuals throughout the range of a species. extinction debt A situation in which, following habitat loss, conditions have become insufficient for survival of some species; these species are destined for extinction, yet are still extant because of the time delay in their response to the habitat loss. extinction filter A reduction in the vulnerability of an assemblage to further species extinction as a consequence of previous exposure of that assemblage to the same threatening process. extinction rate The number of species in a given area that become extinct per unit time. extirpation Complete loss of a population of a species at a specific location, with the species continuing to survive elsewhere (as opposed to extinction, in which all populations of a species are lost). extratropical cyclone A cold-core cyclonic storm of middle and high latitudes. Such storms include attendant frontal systems and are much larger than tropical cyclones. extrusion The passive eruption of magma on the Earth’s surface, either in the air or under water. Rocks and magmas formed by this process are termed extrusive. facilitation In ecology, the enhancement of a process by other species or conditions. facultative Able to exist with or without a given set of conditions. fault A planar fracture in bedrock resulting from brittle failure under stress. A strike-slip fault is one in which euphotic zone

the dominant displacement is horizontal. A thrust fault is low-angle and is the result of compression, allowing rocks to override one another. A normal fault is a fault in which the dominant displacement is vertical or high-angle. fellfield A plant community characteristic of harsh cold climatic conditions where vegetation has less than 50% ground cover. Vegetation is sparse, low, and mainly composed of bryophytes, lichens, and small herbs. feral Describing or referring to an individual of a domestic species that has shifted into a wild state and is no longer under the control of humans. fernbrake Thick vegetation dominated by one or more fern species. fetch The spatial region over which the wind blows when generating wave motion in the ocean. fish In the context of fish stock management, not only vertebrate fish (finfish) but also molluscs, crustaceans, and all other forms of marine animal and even plant life. fitness An individual’s genetic contribution to the gene pool of the next generation relative to the average for the population. Also, DARWINIAN FITNESS. floating islands Large portions of soil and vegetation removed and dragged away by the action of rivers and subsequently carried to neighboring islands by marine currents. flocculation The process by which individual, minute, suspended particles are bound together to form aggregations, such as the settling out of suspension of clay particles in salt water. flying fox A large diurnal bat of the family Pteropidae. flysch Sediment deposited in deep water at a subduction zone. focus See HYPOCENTER. fog forest See CLOUD FOREST. folivory Subsistence on a diet dominated by leaves. foraminifera Amoeboid protists with a calcium carbonate shell. They may be either planktonic or benthic and are typically about 1 mm in diameter. forb A broad-leafed non-woody plant. forearc or forearc basin A depression on the sea floor located between the subduction zone and its associated volcanic or island arc. It is typically filled with sediments from the adjacent landmass and island arc in addition to trapped oceanic crustal material. forest dieback or decline Stand-level loss of canopy foliage out of season; a vegetation process involving a structural dynamic change conditioned by chronic stress caused by aging and/or habitat constraints, and

initiated by a trigger that usually remains elusive. Also, CANOPY DIEBACK . Fossa Magna A rupture zone traversing the middle part of the Japanese islands and separating the Southwest Japan and Northeast Japan arcs. founder effect Loss of genetic variation due to a colonizing population carrying only a small fraction of the total genetic variation of the parental population. Can lead to intense genetic drift and, in extreme cases, to speciation (founder effect speciation). founder event Formation of a new population by one or a few individuals that are genetically, geographically, or behaviorally isolated from a source population; effective population size is much smaller than source population. May subsequently lead to genetic changes, yielding a founder effect. founder flush A type of founder effect speciation that proposes a reduction in intraspecific competition and an increase in population size following a bottleneck. fracture zone Linear fractures that extend lateral to transform faults on the ocean floor. frankincense An aromatic resin derived from plants, used for creating a pleasant scent when burned. The frankincense tree, Boswellia, produces high-quality frankincense and has several endemic species on Socotra. frequency-dependent selection The principle that a phenotype will experience variable selection pressure (e.g., predation), with fitness declining (negative frequency-dependent selection) or increasing (positive frequency-dependent selection) as the relative frequency of that phenotype increases. fringing reef A reef system attached to a mainland or continental island shoreline. frugivory Subsistence on a diet dominated by fruits. fundamental niche The entire range of biological and physical environmental conditions under which a species can reproduce and survive. gabbro A coarse-grained rock formed by intrusion of basaltic magma. gene flow The movement of genes among populations as a consequence of organismic dispersal. genetic bottleneck An event associated with a large decrease in the numbers of breeding individuals of a population. Also, BOTTLENECK, POPULATION BOTTLENECK. genetic differentiation The accumulation of genetic differences between populations. genetic drift Random changes in the frequency of alleles in a gene pool as a result of chance rather than natural selection; usually occurs with greater force in small

GLOSSARY

997

populations. Also, DRIFT .

DRIFT, ALLELIC DRIFT, RANDOM GENETIC

New genetic combinations generated by the process of recombination via reassortment of chromosomes during meiosis and crossing-over. genetic revolution A condition, generally associated with founder event speciation, in which coadapted sets of genes are broken apart and allowed to rearrange as a result of a change in the genetic environment of the population. genetic transilience A type of founder effect speciation where reshuffling of alleles within an outbred founding population can lead to a shift in selective forces. genetic variance The variance among individuals in a population resulting from the presence of different genotypes. genotype The internally coded, heritable information of an organism. ghost of interaction past A trait that is exhibited by an extant species and that evolved or was evolutionarily maintained by the interaction of that species with one or more other species that are now extinct. gigantism An evolutionary trend toward increasing body size; common on islands where it is thought to be a result of reduced competition. glacial An interval of colder temperatures and glacier advances (glaciation), in which the passage of large masses of slow-moving ice (glaciers) causes changes in the Earth’s surface by erosion or deposition. Gondwana A supercontinental configuration of Southern Hemisphere land masses that existed between Permian and Cretaceous times. All the continents of the hemisphere are fragments of the Gondwana continent that have been separated from each other by continental drift. (Named after the common occurrence of the distinctive “Gondwana sequence” in parts of India, Southeast Asia, Australia, Antarctica, South America, and Africa.) graben A depressed block of land between two parallel faults. gradient In climatology, the rate of change of a variable with distance; mathematically expressed as the rate of decrease. granite Silica-rich intrusive igneous rock consisting mainly of alkali-feldspar and quartz. granulite A metamorphic mineral assemblage formed when crustal rocks are metamorphosed at temperatures greater than 650 °C. gravity anomaly A change in the strength of the Earth’s gravitational field resulting from lateral differences in the density of materials. genetic recombinants

998

GLOSSARY

Atmospheric gases that contribute to global warming; they are, in order of abundance, water vapor, carbon dioxide, methane, nitrous oxide, and ozone. Gruiformes A bird order (lit. “cranelike”) containing the coots, cranes, and rails. guano Droppings from birds, bats, or other vertebrates, rich in nitrogen and phosphorus. Guiana shield A more-than-1-billion-year-old geological formation that underlies much of northeastern South America. guyot An undersea mountain (seamount) with a flat top, considered to be an extinct volcano. (Named after Swiss-born U.S. geologist Arnold Henri Guyot.) gymnosperms Non-flowering seed plants, including conifers and cycads. habitat A place with a suitable environment (vegetation, food availability, and the like) for a particular species. habitat island A distinct area of suitable habitat, surrounded by areas unsuitable for the species in question. Examples include forest patches in an agricultural landscape, mountaintops (for montane species), or coral reefs. halieutic Of or referring to fishing. halophilous Describing or referring to organisms that live in areas of high salt concentration; the organisms have special adaptations to permit them to survive in these environments. halophytes Plants that are capable of surviving and growing in high salt concentrations. hardground A limestone that has been cemented at a sea bed or lake bed, forming a hard and commonly irregular surface. Hardy–Weinberg ratios The genotypic ratios (expressed in terms of the allele frequencies) expected in a large, isolated population given random mating, no mutation, and an equal fitness of all genotypes. harmonic biota An island biota that has much the same composition as the nearest continental area. harvest rate The fraction of a fish stock that is harvested in a given time period. hawaiite A volcanic rock, a differentiated type of basalt with moderate sodium and potassium. head Poorly sorted, poorly stratified deposits of locally derived angular rock fragments produced by solifluxion (i.e., downslope movement of soil material in which water acts as a lubricant rather than as an agent of transportation) under periglacial conditions; these deposits may mantle high ground or may occur on slopes or in valley bottoms. greenhouse gases

A plant (Agave fourcroydes; Agavaceae), native to the Yucatán, whose leaves produce a fiber used primarily in cordage (ropes, cords, and twine). Synthetic fibers have now largely replaced its use. 2. The natural fiber derived from agave plants. hermatypic A descriptive term for reef-building corals. herpetofauna The reptiles and amphibians of a given region. heterochrony Evolutionary changes in size and appearance caused by the differential timing of development of features. Highland Clearances Forced displacements of people from the Scottish Highlands in the eighteenth century. high-pressure/low-temperature metamorphics The metamorphic rocks formed in unusually low geothermal gradients, characteristic of metamorphism in subduction zones. hinge rollback Movement of the subduction zone as subducted lithosphere descends into the mantle under the force of gravity. Subduction is often viewed as the result of plate convergence but can equally be considered as the result of one slab falling into the mantle with the upper plate extending to fill the space created. Holarctic The biogeographic realm comprising arctic, boreal, and temperate regions of Eurasia and North America. Holocene The geologic time span from about 11,500 years ago to the present. holomictic lake A lake that is mixed completely from top to bottom. holotype The specimen or specimens designated as representative of a newly described species. hominin Any member of the tribe Hominini in the subfamily Homininae, which comprises all creatures believed to be modern humans or human ancestors; includes all species within the Homo and Australopithecus genera. Homo erectus Latin for “upright man”; an extinct member of the human lineage that first evolved in Africa around 2 million years ago and lasted until about 400,000 years ago. horizon A distinct spatial pattern in material culture. horst A block pushed upward between faults. host In biology, an animal or plant on or in which a parasite (including those that cause disease) or commensal lives. hotspot 1. In geology, a small region of increased volcanic activity, thought to be caused by locally increased mantle melting due to increased temperature or lowhenequen

1.

ered melting point, which produces a chain of islands (hotspot archipelago). 2. See BIODIVERSITY HOTSPOT. Hoxnian Interglacial An interglacial period between 300,000 and 200,000 years ago. Humboldt Current The cool current flowing northward along the west coast of South America; first scientifically described by German naturalist and geographer Alexander von Humboldt. husking sites Sheltered sites used by rodents to strip inedible material (husks) from seeds and other collected food such as invertebrates. hyaloclastite Rock made of fragmented volcanic glass. hybridization Interbreeding between two different species (or populations), resulting in the production of offspring (a hybrid); may be either artificial or natural. hydroexplosive zone The upper 600 m or so of the ocean, within which overlying water pressure is insufficient to prevent underwater volcanic eruptions from being explosive. hydrology The scientific study of the behavior, distribution, and movement of global water (both liquid and solid) in the atmosphere and on and under the Earth’s surface. hydrothermal mineralization Mineral deposits created by the circulation of hot, watery fluids through the uppermost part of the Earth’s crust. hydrothermal vent A hot-water vent on the sea floor, occurring near volcanically active places and characterized by unusual, high-biomass communities dependent on chemoautotrophic production. hypocenter The point at which a sudden breakage of rocks within the Earth starts, leading to an earthquake. Also, FOCUS. Iapetus Ocean An ancient ocean, the opening and closing of which preceded the opening of the Atlantic Ocean. (Named for Iapetus, the mythical Greek father of Atlantis.) ichthyofauna The fishes of a given region. igneous rock Rock produced by solidification of magma on or beneath the Earth’s surface. ignimbrite A deposit formed as a result of an explosive eruption and subsequent deposition of hot clastic materials, such as ash. immigration In the theory of island biogeography, the process of arrival of a propagule on an island. The fact of an immigration implies nothing concerning the subsequent duration of the propagule or its descendants. immigration rate The number of new species arriving on an island per unit time.

GLOSSARY

999

On an island, the reduction in the number of species or higher taxonomic ranks per unit area compared to the mainland. indigenous Describing or referring to species occurring naturally within a specific geographic area, though they may also occur elsewhere. Also, NATIVE. insectivory Subsistence on a diet dominated by arthropods. intensity In seismology, a measure of the effect of an earthquake based on its macroseismic or “felt” impact; commonly reported according to the modified Mercalli intensity scale. A single earthquake can be associated with a wide range of effects and intensities that vary primarily as a function of distance away from the earthquake. interfertility The capability to produce fertile progeny through interbreeding; often expressed at different levels depending on the proportion of viable spores or gametes produced. interglacial An interval between glacial periods when climates are warmer, sea level is higher, and tree lines are higher on mountains. internecine Relating to mutual violence within a larger group. Intertropical Convergence Zone The boundary (convergence zone) between the northeast and southeast trade winds, seasonally shifting 10° north to south of the Equator. intraplate Situated within the interior of a crustal tectonic plate. intrinsic growth rate The rate at which a [fish] stock changes in size over a given time period. introduced species See NON-INDIGENOUS SPECIES. introgression The introduction of genes of one species into the gene pool of another through hybridization. intrusion The emplacement of bodies of molten rock within the Earth’s crust. Rocks and magmas formed by this process are termed intrusive. invasional meltdown A process whereby two or more introduced species enhance one another’s probability of establishment or impact on native species. invasive Describing or referring to a species, usually non-indigenous, that expands its range, often causing negative environmental, economic, and/or human health impacts. inversion An atmospheric layer in which temperature increases with increasing height; an exception to the normal decrease (or lapse) of atmospheric temperature with height. inverted barometer The response of sea level to changes in atmospheric pressure (10 mb of pressure decrease corimpoverishment

1000

GLOSSARY

responds to 0.1 m of sea level increase); a component of storm surge. Iron Age A period in human cultural development, preceded by the Bronze Age, and marked by the development of tools and weapons made of iron. The adoption of this material coincided with other changes in some past societies, often including differing agricultural practices, religious beliefs, and artistic styles. In the ancient Near East and Greece, the Iron Age began in the twelfth century BC; in Great Britain it lasted from about the seventh century BC until the Roman conquest. irruption Rapid and irregular increases in a species’ abundance, which are often subsequently followed by rapid declines in their populations. island arc An arcuate group of islands and seamounts formed by magma produced during the subduction of an oceanic plate beneath another oceanic plate. The subducted slab adds water, and decompressional melting generates basaltic magma that penetrates the overriding plate and forms volcanoes. island biogeography A field of biogeography that attempts to establish and explain the factors that affect species richness in archipelagoes. The field began in the 1960s and was established by the ecologists Robert MacArthur and E. O. Wilson, whose theory of island biogeography attempted to predict the number of species that would exist on a newly created island. More recently, this theory has been extended to terrestrial fragmented systems. islet Any small island, including any of many small islands comprising an atoll. isometry A relationship in which one variable changes with another one while keeping the same proportion between the two (e.g., if heart weight doubles when body weight is doubled, then the relationship between them is isometric; but when body length doubles, surface area increases four-fold, and body weight increases eight-fold, the relationship between these variables is allometric rather than isometric). isostasy The state of equilibrium of vertical forces acting on the Earth’s crust. isostatic rise or subsidence Changes in the elevation of a tectonic plate as its thickness or density changes (e.g., as by erosion or cooling) so as to maintain the overall gravitational equilibrium between the Earth’s lithosphere and asthenosphere. isotopic age An age expressed in years and calculated from the quantitative determination of radioactive elements and their decay products and known rates of decay.

The World Conservation Union, formerly the International Union for the Conservation of Nature and Natural Resources. karst Carbonate rock terrain with complex erosional features, such as sinkholes and caves, formed through dissolution by water. karstification The process by which an area of irregular limestone is eroded, producing fissures, sinkholes, underground streams, and caverns. key innovation In evolution, any modification in structure or function that permits a lineage to exploit the environment in a more efficient or novel way and thereby leads to a comparatively rapid diversification of the lineage. keystone species A species that usually plays a large role in community or ecosystem function relative to other species. kı¯puka An island-like area of older substrate, often supporting forest, that is surrounded by younger lava flows. kı¯puka systems Kı¯puka, their surrounding lava flows, and their inhabitants. lagoon A shallow body of water, separated from the sea by coral reefs or sandbars. lahar An Indonesian term for water combined with volcanic rock debris deposited on the slope of a volcano that flows rapidly downslope. lamprophyre Dark-colored igneous rock with large crystals of minerals such as hornblende, biotite, and alkali feldspar. land-bridge island An island once connected to the adjacent land mass but later isolated by rising sea levels. Landnám A Norse term (lit. “taking land”) usually applied to the process of preparing the landscape for farming, such as forest clearance. landscape A distinct area of land that may be occupied or unoccupied by natural vegetation; a geological landform with built-up environmental or human-constructed features; the geographical equivalent of ecosystems or interconnected ecosystems. Lapita An archaeological complex marked by a distinctive style of earthen-made pottery decorated with dentate-stamped designs and found from New Guinea to Samoa during ~1500–500 BC. Makers of Lapita pottery are thought to have been the speakers of proto-Oceanic language and to have been direct ancestors of the Polynesians. large organic fall A common term for a large parcel of organic matter, such as a dead whale, log, or detrital kelp mass, that has sunken to the sea floor. IUCN

The maximum extent of ice sheet coverage during the most recent glaciation. Names for this period differ according to geographic distributions, the most intensively studied being the Wisconsin in North America, generally thought to have occurred 70,000–20,000 years ago, and the Devensian in the British Isles. lateral collapse A very large landslide that removes a sector of the flank of a volcano to a depth of 0.5–3 km below the surface, and often also removes the summit. lateral plates Modified scales that occur as a single row along the flanks of most sticklebacks. The plates mechanically link the spines on the back and pelvis of the threespine stickleback, producing a formidable defense structure. late-stage species Species that have passed through several stages of the taxon cycle; they tend to have a narrow and fragmented distribution in interior or montane habitats. laurel forest/laurisilva A previously widespread forest type that, in the Tertiary, occupied the area that is presently Europe and southwestern Africa. lava A term for magma that has passively erupted onto the Earth’s surface. lava domes Roughly circular mounds of lava formed by the eruption of viscous (sticky) magma. Lazarus phenomenon The reappearance of a species thought to be extinct, usually as a very small population, and sometimes in a previously undocumented place or habitat. lecithotrophic Feeding on yolk supplied by an egg. leptospirosis A disease caused by the bacterium Leptospira that can infect humans as well as wild and domestic animals, causing flulike symptoms. It is generally spread through contamination of soil or water supplies by the urine of infected animals. Levallois The archaeological name for a type of flint knapping developed by humans during the Paleolithic. limestone A sedimentary rock composed primarily of calcium carbonate (CaCO3 ), usually as the mineral calcite. Linnaeus, Carl The scientist who laid the foundations for binomial nomenclature (genus name followed by species name) and modern taxonomy in the mid-eighteenth century. lithified Describing or referring to sediment that has been changed (over time and under certain conditions) into rock. lithology The types and characteristics of the rocks of a region. last glacial maximum

GLOSSARY

1001

Early flying birds with a paleognathous palate, generally similar to tinamous. lithosphere The outer, rigid part of the Earth, including the crust and part of the mantle to depths of about 100 km, forming the tectonic plates. living fossils Species or species groups most closely related to a now totally extinct archaic lineage. Groups of living fossils usually have low taxonomic diversity. local endemism See ENDEMISM. loess A widespread, homogeneous, commonly nonstratified, porous, friable, slightly coherent, usually highly calcareous, fine-grained blanket deposit consisting predominantly of silt with secondary grain sizes ranging from clay to fine sand; formed as a windblown dust of Pleistocene age, and carried from land surfaces unprotected by a vegetative cover, such as glacial or glaciofluvial deposits uncovered by glacial recession. lowstand valley A valley carved across coastal plains and the continental shelf when sea level was lower during the ice ages (Pleistocene epoch). Macaronesia The biogeographic region comprising the archipelagoes of Azores, Canaries, Madeira, Selvagens, and Cape Verde and some parts of Morocco and Mauritania. machair The land between a beach and the area where sand encroaches on peat bogs; former beach. macrotidal Experiencing tides with a vertical range over 2 m. mafic Composed chiefly of one or more ferromagnesian minerals such as olivine and pyroxene. magma Molten rock that may include crystals and exsolved gases (bubbles); when magma erupts at the Earth’s surface, it loses most of its gas and is called lava. magmatic eruption An explosive volcanic eruption that occurs when dissolved gases in rising magma expand to form gas bubbles that then burst as the magma nears the Earth’s surface, leading to explosive fragmentation of the magma. magnetic anomaly A change in the strength of the Earth’s geomagnetic field resulting from lateral differences in the magnetic mineral content of rocks. magnetic lineation A linear pattern of magnetic anomaly that is preserved in rocks on the ocean floor. magnitude A measure of the size of an earthquake, derived from seismic signals recorded on calibrated seismic instruments that are corrected for the distances between the earthquake and the recording instruments; the magnitude is a property of the earthquake rather than of its effects or intensities. Lithornithiformes

1002

GLOSSARY

The whole set of species inhabiting a mainland that can potentially colonize a given island. makatea 1. Fossilized coral, referring to a raised coral reef that encircles an island. 2. A type of island consisting of a volcanic center surrounded by an emerged (or uplifted) fringing coral reef. 3. A composite island (volcanic and limestone). malacology The biological subdiscipline that deals with molluscs. Malesia The geographic area that includes the countries of Indonesia, Malaysia, the Philippines, Papua New Guinea, Singapore, and Brunei Darussalam. mammal-niche species Species other than mammals that occupy a similar ecological niche in their absence. mangrove The tropical or subtropical vegetation that grows in the intertidal zone; often used more narrowly to refer to species in the family Rhizophoraceae, which frequently dominate mangrove vegetation. mantle The ~2900-km thick part of the Earth between the crust and the core. It is divided on the basis of seismic wave velocities into upper mantle and lower mantle, separated by a transition zone. mantle plume An upwelling of abnormally hot rock within Earth’s mantle. Where such plumes interact with the lithosphere, they form volcanic provinces called hotspots. mantle wedge Part of the Earth’s convecting upper mantle (asthenosphere) overlying a subducted slab, composed of peridotite (olivine-pyroxene rock) but likely with heterogeneities established by prior melt loss and ingress. maquis Evergreen xerophyllous and sclerophyllous shrubs in the Mediterranean region (analogous to chaparral in California, fynbos in South Africa, and mallee in Australia). marginal basin A small oceanic basin that is adjacent to a continent and is separated from larger oceans by an island arc. marine advection The dominant atmospheric process on islands because of their oceanic context. When the atmospheric circulation is constant all around the year, this process can lead to a drastic contrast between the wet windward part and the dry leeward part of a mountainous island. marine carbonates Rock derived from calcium carbonate fixed by marine animals, of submarine origin. marine protected area (MPA) Any intertidal or subtidal area, together with its associated flora, fauna, and historical and cultural features, that has been set aside by mainland species pool

law or other effective means to protect part or all of the designated environment. Maritime Continent The region encompassing Indonesia, Southeast Asia, and the southwestern Pacific. The numerous large islands influence the atmosphere in a manner similar to the equatorial continents of Africa and South America. marl A soft, impure limestone that commonly makes fertile soils. massif A massive topographic and structural feature, usually formed of rocks more rigid than its surroundings. massive sulfide mineralization Mineral deposits with a high percentage of sulfide minerals, typically containing copper, lead, and zinc; commonly interbedded with marine volcanic horizons, and forming generally through exhalation of hydrothermal fluids through vents on the sea floor (“black smokers”). masting A group phenomenon in which plants within a population correlate their reproductive activity in both time and size of crop. mating asymmetry Mating behavior in which one gender of a population or species is more likely (relative to the opposite gender) to mate with members of another population or species. matrix The most extensive and connected land cover type in a landscape, which therefore plays the dominant role in the functioning of the landscape. mean sea level The average sea level (including waves, currents, and tides) determined over time. Mediterranean-type ecosystem (MTE) An ecosystem defined by a climate that resembles those of the lands bordering the Mediterranean Sea, with wet winters and dry summers. MTEs are found in five widely-separated regions of the world, namely Australia, California, Central Chile, South Africa, and the Mediterranean Sea. megaherbs Spectacular herbaceous perennial flowering plants found on subantarctic and cool temperate islands in the Pacific and Indian Ocean regions. The plants are characterized by very large leaves and large, unusually colored flowers, and are well adapted to the harsh climatic conditions. melanephelinite Dark-colored low-SiO2 basaltic lava with large crystals of the minerals olivine and pyroxene, and also the mineral nepheline. Melanesian Referring or belonging to Melanesia, those islands from New Guinea south to New Caledonia and east to Fiji. melange A mappable unit composed of a heterogenous mixture of deformed rocks, typically with a pervasively sheared muddy matrix.

A lake that is not mixed seasonally from top to bottom but remains stratified for long periods of time because of density. mesocosm An experimental container that is used to simulate the natural environment while allowing researchers to manipulate the local conditions under replicated conditions. Mesolithic period A period in the development of human technology starting about 11,500 years ago (toward the end of the Pleistocene) and ending with the introduction of farming (around 5000 BC in northern Europe). Sometimes referred to as the Middle Stone Age (between the Paleolithic or Old Stone Age and the Neolithic or New Stone Age). mesotidal Experiencing tides with a vertical range of 2–4 m. Mesozoic The geological era in the Phanerozoic eon from ~250 million to 65 million years ago that includes the Triassic, Jurassic, and Cretaceous periods. Messinian Salinity Crisis The period when the Mediterranean Sea evaporated partly or completely during the Messinian period of the Miocene epoch, approximately 6 million years ago. metabasite Metamorphosed mafic rock, such as metamorphosed basalt. metacommunity A set of local communities that are linked by dispersal of multiple species. metallogeny Origin of mineral deposits. metamorphism A change of mineral assemblage by heat and pressure. Deep in the Earth’s crust, this action produces metamorphic rock (formed from sedimentary and other kinds of rocks). The resulting rock types are considered metamorphosed (e.g., marble is metamorphosed limestone, schist is metamorphosed shale). metapopulation A collection of populations connected through immigration that undergo periodic extinction and recolonization. metasediment A metamorphosed sedimentary rock. microarray A microscopic array of single-stranded pieces of DNA fixed on a piece of glass or plastic used to screen a biological sample for the presence of specific genetic sequences. Fluorescent tags reveal actively transcribing genes that contain the sequence being probed. In this way, relative expression levels can be measured for a very large number of different genes. micro-endemism See ENDEMISM. microfossils Preserved remains such as foraminifera shells, diatoms (photosynthetic protists with silica skeletons) frustules, and plant pollen grains, which are too meromictic lake

GLOSSARY

1003

small to see with the naked eye but provide detailed information to scientists who study past climates. Micronesia A group of islands in the Pacific Ocean that includes the Republic of Palau, the Mariana Islands (Guam and the Commonwealth of the Northern Mariana Islands), the Federated States of Micronesia (Yap, Chuuk, Pohnpei, and Kosrae states), and the Republic of the Marshall Islands. microsatellites Highly variable, putatively neutral, dominant allelic markers located in the nuclear genome that consist of variable numbers of short, repeated base pair sequences. microtidal Experiencing tides with a vertical range less than 2 m. mid-ocean ridge A divergent plate boundary, where new crust is created from basaltic magma by volcanic and plutonic processes. Miocene The geologic epoch in the Neogene period extending from about 23 million to 5 million years ago. miogeocline Continental margin. mist forest See CLOUD FOREST. modern introduction A species that has been intentionally or accidentally relocated by humans in the Modern Era, or since approximately 1500 AD. molasse Sediment deposited in shallow water at a subduction zone. molecular markers Specific segments of the genome that are screened for mutations and used to gauge relatedness or associations with phenotypic traits. molecular phylogenetics The use of macromolecular (usually DNA) data to infer relationships among evolutionarily divergent lineages. molluscs A phylum of animals that comprise a monophyletic lineage originating early in the explosive Cambrian metazoan diversification more than 500 million years ago, including invertebrates such as snails, slugs, oysters, clams, octopuses, and squids, and characterized by bilateral symmetry, a well-developed foot, a mantle, a mantle cavity, a complete gut, and well-developed kidneys (metanephridia). monophyletic Derived from a common ancestor. monophyletic group An exclusive group containing all of the descendants from a unique common ancestor. monotypic family A single species that is the sole member of the family to which it belongs. monotypic genus A genus in which there is only one species. monsoon A seasonal prevailing wind. The term is often applied to rainy seasons, but not all prevailing seasonal winds are wet (e.g., the winter monsoon of East Asia).

1004

GLOSSARY

A climate characterized by an annual cycle of prevailing winds, typically featuring a rainy season and a dry season. Moravian A denomination that was the first large-scale Protestant missionary movement. It has roots in the late fourteenth-century Catholic Church reform movement of Jan Hus and was revived in the eighteenth century in what is today Germany (but takes its name from a part of the Czech Republic). MORB Mid-ocean ridge basalt, erupted at a seafloor spreading axis. morphology The study of the structure and form of organisms and their parts. mossy forest See CLOUD FOREST. MPA network A series of marine protected areas (MPAs) connected by larval dispersal or juvenile and adult migration. MTE Mediterranean-type ecosystem. multibeam bathymetry A technique in which highresolution bathymetry data are collected by ship echosounders that transmit multiple beams of sound simultaneously to collect a swath of data. murids Members of the Muridae (order Rodentia), the largest family of mammals; the group of rodents that includes rats and mice. mycorrhiza A mutual symbiosis between a fungus and the roots of a plant, in which the fungus extracts carbohydrates from the plant, and the plant uses the fungal mycelium to more efficiently extract nutrients from the soil. mysticete A member of the whale suborder Mysticeti (order Cetacea), the baleen, or filter-feeding, whales, including right whales, rorquals, and grey whales. mythicomyiid Belonging to the family Mythicomyiidae (order Diptera), a group of tiny humpbacked flies. myxomatosis A ravaging disease of European rabbits (Oryctolagus cuniculus), caused by the myxoma virus and spread by fleas and mosquitos. Purposely introduced to Australia to control rabbit populations, it reached the United Kingdom in 1953. nappe A sheet of rock that has moved a considerable distance as a result of tectonic activity and has been emplaced on or adjacent to rocks of different character. National Wildlife Refuge System Lands and waters managed by the U.S. Fish and Wildlife Service, Department of Interior, for the conservation, management, and (where appropriate) restoration of wildlife (plants and animals) and their habitats for the benefit of present and future generations of Americans. native See INDIGENOUS. monsoon climate

An organism that feeds on another; a term used in biological control. naturalized species An intentionally or unintentionally introduced species that has adapted to and reproduces successfully in its new environment without human help. natural selection The process by which favorable heritable traits become more common in successive generations, because individuals bearing these favorable traits are more likely to survive and reproduce than those bearing less favorable traits. nautilus A primitive cephalopod with an external spiral shell. Nazca plate An oceanic tectonic plate in the eastern Pacific Ocean basin off the west coast of South America. (Named after the Nazca region of southern Peru.) Neanderthal Homo neanderthalensis, a species of humankind extinct by about 30,000 years ago, distinguished from modern Homo sapiens by features of the skull and postcranial skeleton, and typically associated with a Mousterian stone-tool culture and Eurasian geographic range. Nearctic The high arctic latitude zone of North America. Near Oceania The part of Oceania lying west of Santa Ana Island; includes New Guinea, the Bismarck Archipelago, and the Solomon Islands. Parts of this region were settled by humans as early as ~40,000 years ago. necrotroph An organism that kills the host and then lives as a saprobe on dead host tissue. nematocyst A microscopic capsule unique to Cnidaria that is synthesized by a single cell (the nematocyte) and can be used just once; there are about 30 kinds (although animals of a given species typically possess no more than four to five kinds), some of which function offensively in prey capture, some defensively, and, in a few cases, for locomotion. neo-endemic A species endemic to an area by reason of having evolved in situ, originating from an ancestor with origin on another island or on the mainland. Neogene The geologic period in the Cenozoic era from about 23 million years ago to present that includes the Miocene, Pliocene, Pleistocene, and Quaternary epochs. Neognathae The suborder of birds (lit. “new jaws”) (order Aves) containing the vast majority of extant species. Neolithic period A period that ran in northern Europe from around 5000 BC to 1700 BC (the beginning of the Bronze Age). natural enemy

The retention of juvenile body characters in the adult state as a result of maturation of the reproductive system in a larval form that fails to undergo metamorphosis. neotropics An ecozone that includes more tropical rainforest (tropical and subtropical moist broadleaf forests) than any other ecozone, extending from southern Mexico through Central America and northern South America to southern Brazil, including the vast Amazon rainforest. nested distribution/nestedness A distribution of species in which the species occurring within species-poor communities are subsets of those occurring in richer communities. niche 1. The particular environmental habitat to which a species is adapted. 2. The functional role of a species in an ecosystem, including the habitat in which it lives and the resources it uses. Also, ECOLOGICAL NICHE. nitrogen fixation The process by which plant species interact with bacteria to incorporate atmospheric nitrogen (N2) into compounds such as ammonia and nitrate that are useful for various chemical processes, including physiological processes of plants. non-adaptive radiation Species proliferation that has not been attended by diversification of ecological roles. nonequilibrium A condition of instability in ecological communities that change in structure and composition in response to the environment. non-indigenous or non-native species A species found in an area in which it does not occur naturally. Also, ALIEN SPECIES, EXOTIC SPECIES , INTRODUCED SPECIES. non-volant Unable to fly; among mammals, only bats are volant. Norse Describing or referring to the people and cultural traditions of communities living in, or migrating out from, Norway, mostly beginning around 700 AD. Notogean Realm The zoogeographical division of the Earth’s surface that includes Australia, New Guinea, New Zealand, and the southwestern Pacific islands and which is based on shared geographical distributions of distinctive faunal elements (e.g., four of the six orders of marsupials). null hypothesis A hypothesis that is set to be tested against the theory being developed. It usually involves the distribution of values to be obtained if only random processes were operating. nunatak A mountain peak that protrudes above a glacier. obligate Restricted to a particular way of life. neoteny

GLOSSARY

1005

A large group of islands in the south Pacific including Melanesia, Micronesia, and Polynesia (and sometimes Australasia and the Malay Archipelago). oceanic crust The mafic (magnesium- and iron-rich) basement rock beneath all oceans, normally about 7 km thick; it is thinner and denser than continental crust. oceanic island An island (usually of volcanic origin) that begins its above-water existence in isolation from the mainland, without a prior land connection. odontocete A member of the whale suborder Odontoceti (order Cetacea), toothed whales, including porpoises, dolphins, and sperm whales. Oligocene The geologic epoch in the Paleogene period that extends from about 34 million to 23 million years ago. omnivorous Eating both plants and animals. ontogenetic Referring to the developmental stages that occur in the sequential transformation of a fertilized egg into an adult organism. oolites Spherical-shaped sand grains composed mostly of calcium carbonate precipitated out of seawater in the form of spherical layers around a nucleus (commonly a small shell fragment). ophidian 1. Pertaining to snakes. 2. A member of the suborder Ophidia, generally considered an alternate name for Serpentes. ophiolite An assemblage of ultramafic and mafic rocks, widely thought to represent oceanic crust. ophiolite complex or suite A fragment of ancient ocean floor emplaced onto a continent. The idealized ophiolite sequence consists of deep-sea sediments, underlain by basaltic (pillow) lavas, next to a sheeted-dike complex, and next to coarse gabbroic plutonic rocks, which together form the oceanic crust. Ophiolites also often contain slivers of the underlying mantle. optimal body size A body size that serves as an evolutionary attractor under some theories of size evolution. orogen A geological mountain belt of deformed, intruded, altered rocks. orthopteran An insect of the order Orthoptera, a group with incomplete metamorphosis that includes grasshoppers, crickets, and locusts. osteology Characterization of skeletal elements, often used to identify and classify species. outcrossing The movement of genetic material between two separate individuals. outrigger A section of a canoe’s rigging that helps to stabilize the craft, especially when turning. overriding See SUBDUCTION. Oceania

1006

GLOSSARY

An oscillation phenomenon in the Pacific Ocean on a time scale of 20 to 30 years, in contrast to 6 to 18 months for the El Niño–Southern Oscillation (ENSO). The climatic fingerprints of the PDO are most visible in the North Pacific/North American sector, whereas secondary signatures exist in the tropics. In contrast, ENSO signatures are global. paedomorphosis A condition in which the body shape of a descendant adult species resembles the juvenile shape of its ancestor species; one of a class of phenomena relating to phyletic alteration of growth trajectories known as heterochrony. pa ¯ hoehoe Basaltic lava flow of typically lower viscosity, which cools to produce a smooth and ropy surface. (A term originating from the Hawaiian language.) palagonite Clays resulting from hydration of volcanic glass, a line of evidence that a lava flow may have quenched in contact with water or ice to form the volcanic glass. Palearctic The high, arctic latitude zone of Eurasia. paleobotany The study of fossil remains of plants. paleoclimatology The study of climates of the past through indirect methods such as the study of sedimentary records. paleoecology The study of past environments through examination of the fossil records of organisms living in the environment at that time. paleo-endemic A species endemic to an area by reason of other populations or close relatives elsewhere becoming extinct; has changed little after colonization. Paleocene The geologic epoch in the Paleogene period from about 65 million to 56 million years ago Paleogene The geologic period in the Cenozoic era from about 65 million to 23 million years ago, which includes the Paleocene, Eocene, and Oligocene epochs. Paleognathae The suborder of birds (lit. “old jaws”) containing the primitive tinamous and ratites. Paleolithic The Old Stone Age, a period in human technological development during which stone tools were introduced and the climate changed frequently. The Lower Paleolithic (or Palaeolithic) ran from about 2.5 million years to 100,000 years ago; the Middle Paleolithic, 300,000 to 30,000 years ago; and the Upper Paleolithic 40,000 and 10,000 years ago. The period was followed by the Mesolithic. paleomagnetism The direction and intensity of the Earth’s magnetic field as recorded in igneous rocks when they cool. paleontology The study of fossils. Pacific decadal oscillation (PDO)

The zoogeographical division of the Earth’s surface that includes Africa, Madagascar, India, and Indo-Malaya through the Lesser Sundas and Moluccas, and which is based on shared geographical distributions of distinctive faunal elements (e.g., elephants, pangolins, rhinoceroses, civets, and mongooses). Paleozoic The geologic era in the Phanerozoic eon from about 540 million to 250 million years ago and including the Cambrian, Ordovician, Silurian, Devonian, Carboniferous, and Permian periods. pandanus Tree and/or fruit of the monocotyledonous species Pandanus tectorius, sometimes referred to as screw pine. Pangea The ancient supercontinent that included North America, Scandinavia, Europe, and Africa; produced by the closing of the Iapetus Ocean. panmictic Describing or referring to a population in which all individuals are potential partners; there are no restrictions to mating. pantepui A biogeographic region in Venezuela and elsewhere in the north central Guyana shield that includes all the isolated tepui summits. The region is noted for the distinctive biota adapted to the typically cold, wet, rocky, and windy conditions on these summits. Panthalassa Ocean A name given to the vast ocean configuration that existed during the time of the Gondwanaland and Laurasia supercontinents (much of Paleozoic and Mesozoic time). Papuan The large group of 900 or more non-Austronesian languages (including many different and ill-defined language families) centered on the island of New Guinea and neighboring islands. paramo A high-elevation habitat found only in South America, associated with the Andes but also with some outlying ranges, dominated by grasses and herbs (notably bromeliads). parasitoid An arthropod that completes its development attached to or inside a single host organism, which it ultimately kills (and often consumes) in the process. Parks Australia North An agency of the Australian Government’s Department of Environment and Water Resources, locally responsible for natural resource management on Christmas Island. passage zone The surface that separates subaerial lava from underlying subaqueous breccia. pathogen An agent, such as a bacterium or fungus, that causes disease. pathogenicity The ability to cause disease. pathway In invasion biology, a route or method along which an invasive species spreads. Paleotropical Realm

In or on the open ocean, far from land. peridotite An olivine plus pyroxene rock that is abundant in the upper mantle; it partially melts to form basaltic magma. periglacial Around the margin of a glacier; said of the processes, conditions, areas, climates, and topographic features near the margins of former and existing glaciers and ice sheets influenced by the cold temperature of the ice. permafrost Earth materials (soil, sediment, or rock) that maintain a temperature at or below 0 °C for a period of at least two years. phenetic clustering 1. A statistical method for grouping objects. 2. In taxonomy, grouping of objects based on overall similarity, without consideration of whether characters are unique to the group or also occurred in the ancestral species; for analysis of evolutionary relationships, this method has largely been replaced by cladistics. 3. In biogeography, grouping of objects based on shared attributes (e.g., islands in terms of shared species). phenotype The outward physical appearance of an organism. phenotypic adaptation Change that allows an organism to become better suited to the environment; can occur through either developmental plasticity (frequently involving parental effects) or adaptive plasticity, allowing directional selection to act and thereby enhance fitness. Such plasticity allows adaptive evolution to occur on an ecological time scale and has been characterized colloquially as “nurture versus nature.” phenotypic plasticity Under different environmental conditions, the ability of organisms of a given genotype to either change their phenotype or produce different phenotypes as a result of developmental plasticity. philopatry The tendency of an animal to return to its birthplace to breed (natal philopatry) and to return to the same nesting site in successive years (breeding philopatry). phonolite A gray, fine-grained differentiated lava with high silica content (~60%) and consisting of alkali feldspars and nepheline. The name relates to its resonance when struck. phreatic eruption or explosion An explosive volcanic eruption that occurs when confined, sub-surface geothermal waters are heated to temperatures above their boiling point and flash to steam, thereby expanding to form an explosion. No molten rock is involved. phreatomagmatic eruption An explosive eruption that occurs when magma comes into contact with water. pelagic

GLOSSARY

1007

Desert trees and shrubs with deep tap roots to the water table. phylogenetics The classification of organisms based on their degree of evolutionary relatedness. phylogenetic tree A branching diagram or family tree showing the evolutionary relationships among species or other entities relative to a common ancestor. Usually represented as a phylogram or cladogram. phylogeny The pattern of relationships of species, usually represented in the form of a tree or hierarchy indicating commonality of descent. phylogeography The study of the geographic distributions of genealogical lineages (populations or species) across the geographic landscape, and of the processes that have shaped these patterns. phylogram A phylogenetic tree in which branch lengths represent time. phytophagous Feeding on plants, including shrubs and trees; said especially of certain insects. phytophagous filter An agent that affects the recruitment of plant species by eating some of them. (For example, land crabs can be phytophagous filters because, by differentially consuming seeds or seedlings of many rain-forest species, they can cause the community of established seedlings to be a nonrandom draw of all seeds falling to the forest floor.) phytoplankton Algae that occur as single cells or in colonies in the open water of lakes, streams, and oceans. picrite Basaltic rock rich in the mineral olivine. Picts Ancient people of northern Britain, undefeated by the Romans. The Picts joined with the Scots from Ireland to form a kingdom (later to become Scotland) in the ninth century AD. Pikermi fauna Fossil vertebrate fauna of Upper Miocene (Turolian) age, intermediate in form between European, Asiatic, and African types, and first found at Pikermi (Attiki, Greece) by Jean Albert Gaudry between 1855 and 1860. Representatives of this fauna have since been found in many other sites around the Mediterranean. Hipparion is a typical fossil genus of this fauna. pillow lavas Elongated lava mounds formed by repeated oozing and quenching of hot basalt lava extruding underwater, and typically having a glassy crust. pinnacle A pointed formation arising from the sea floor. pinniped A marine mammal of the order Carnivora, including seals (Phocidae) and sea lions (Otariidae). plankton Passively floating, drifting, or weakly motile organisms in a body of water, primarily comprising microscopic algae and protozoa. planktotrophic Feeding on materials from the plankton. phreatophytes

1008

GLOSSARY

Large preserved plant remains that have formerly grown in a locality. plate A rigid section of the Earth’s lithosphere, usually containing a proportion of oceanic crust and of continental crust. plate tectonics The study of the dynamics of the surface of the Earth; the movement and marginal interaction of the rigid tectonic plates that form the lithosphere. platform reef An uplifted reef consisting of a flat-topped bench with or without a very shallow lagoon. Pleistocene The geologic epoch in the Neogene period that began about 1.8 million years ago and ended about 11,500 years ago. The period was marked by the repeated formation of extensive ice sheets and other glaciers because of the cooling of the climate. Pleistocene Aggregate Island Complexes (PAICs) Groups of islands separated by shallow seas (depths less than 120 m) that were connected during Pleistocene glaciations when the sea level was lower than it present is. Pliocene The geologic epoch in the Neogene period from about 5 million to 1.8 million years ago. plug Lava intruded as a cylinder-like body. plutonism Intrusion of magmatic rocks such as granite. pneumatophores Small roots arising from the cable root system of some mangroves (e.g., in genera such as Avicennia, Sonneratia, Heritiera, and Xylocarpus), which extend upward into the air as small conical projections that facilitate the movement of oxygen into the root system of mangroves growing in anoxic conditions. polar frontal zone The area between two high-velocity regions of the Antarctic Circumpolar Current: the subAntarctic Front to the north and the Antarctic Polar Front to the south. polje A valley formed by the collapse of the roof of a limestone cave tunnel, through which a stream flows from a cave mouth at one end to a cave mouth at the other. polyp The body form of corals, which is a cylinder with only one body opening—the mouth—that is at one end and is surrounded by tentacles; in some cnidarians, the life cycle alternates between polyp and a jellyfish-like medusa. population Individuals of the same species in a given area that interbreed. population bottleneck See GENETIC BOTTLENECK. porphyry copper–gold Low-grade copper and gold mineralization disseminated through a large mass of granitic to dioritic intrusive rock that typically shows porphyry texture (larger crystals in a finer groundmass). plant macrofossils

Mineralization usually associated with subvolcanic intrusive rocks with a porphyritic texture, forming at depths of around 3–5 km and typically dominated by copper sulphide minerals in an intense network of quartz veins (stockwork). portmanteau biota The assemblage of plants and animals transported by humans, both purposively and inadvertently, during human population dispersals and expansions. Also, SYNANTHROPIC BIOTA. post-caldera Occurring after caldera formation. post-zygotic reproductive isolation Lack of gene flow between evolutionarily distinct lineages in which mating and fertilization have occurred, because of intrinsic factors (hybrids are either developmentally unsound or sterile) or extrinsic factors (hybrids are sound and fertile but unfit in the current environment, especially in comparison with parental phenotypes). Precambrian The geologic supereon from formation of the Earth 4.567 × 109 years ago to 540 million years ago; includes 90% of geologic time and the Hadean, Archean, and Proterozoic eons. pressure gradient The slope of isobaric surfaces; most commonly associated with mean sea levels. pre-zygotic reproductive isolation Lack of gene flow between evolutionarily distinct lineages because of factors that either prevent mating or prevent fertilization after mating. primary productivity In marine systems, a measure of the amount of phytoplankton (chlorophyll-bearing plankton, “plant plankton”) produced in surface waters, which occupy the base of the food chain; usually quantified as grams of carbon produced per square meter per unit time. progradation The seaward or lakeward development of a shoreline by accumulation or deposition of sediments. progression rule Colonization and divergence of a clade within an archipelago, occurring in a stepwise manner, from oldest to youngest island. propagule 1. An individual or group of individuals that arrive at a site. 2. In the theory of island biogeography, the minimal number of individuals of a species required to achieve colonization of a habitable island. propagule pressure A composite measure of the number and rate of arrival of propagules at a site. protists A diverse group of largely unicellular eukaryotic organisms formerly considered to comprise the kingdom Protista. pseudorabies A highly contagious swine disease caused by a herpes virus that can also affect horses, cattle, sheep, porphyry mineralization

goats, dogs, and cats. The pseudorabies virus (PRV) causes reproductive problems, including miscarriages and stillbirths, and can lead to death. PRV is generally spread through animal-to-animal contact. pseudoturnover The apparent loss and gain of particular species from an island resulting from incomplete census data, when they have actually been residents throughout or, alternatively, never properly colonized. pumice A highly vesicular, light-colored fragment or its aggregate produced by volcanic eruptions with low bulk density. pyroclastic Formed from fragments of volcanic origin. (From Greek pyro, meaning “fire,” and clastic, meaning “broken.”) Volcanic rock layers formed in this way are termed pyroclastic deposits (or pyroclastics). Examples of pyroclastic rocks are volcanic ash, pumice, and ignimbrites. pyroclastic density current A current composed of hot fragments and gases produced by volcanic eruption and driven by gravity or its bulk density. quartzite A metamorphic rock derived from pure sandstone and consisting primarily of quartz. Quaternary The geologic period in the Cenozoic era that extends from about 1.8 million years ago to the present; includes the Pleistocene and Holocene epochs. Quechua The native Peruvian language spoken by the Incas; spoken widely in Peru, Bolivia, and neighboring countries. radiation In evolutionary biology, an increase in biodiversity by divergence, especially involving the formation of new species. radiometric dating The dating of igneous rocks utilizing the constant decay of isotopic series elements such as uranium–lead (U–Pb series) and potassium–argon (K–Ar series). rafting The passive transport of floating objects and their associated fauna to distant habitats. rain shadow An area of reduced rainfall on the leeward side of a mountain range, caused by the cooling-induced precipitation as air is forced up over the range. Ramsar List of Wetlands of International Importance An inventory of sites selected within the scope of the Ramsar Convention by the presence of representative, rare, or unique wetland types of international importance for conserving biological diversity. (Formerly the emphasis of the Ramsar Convention was the conservation and wise use of wetlands primarily as habitat for water birds, but over the years the Convention has broadened its scope of implementation to cover all aspects of wetland conservation and sustainable use.)

GLOSSARY

1009

A wetland included in the Ramsar List of Wetlands of International Importance of the Convention on Wetlands, an intergovernmental treaty signed in Ramsar, Iran, in 1971. random genetic drift See GENETIC DRIFT . realized niche A narrower subset of the fundamental niche to which a species is confined as a result of interactions with other organisms. recent introduction Another term for modern introduction, but it typically refers to very recent times. reciprocal translocation Chromosomal mutation resulting in exchange of chromosomal segments between non-homologous chromosomes. recruitment The addition of new individuals to a habitat, generally as a result of reproduction and growth of young to the adult population. Red List An inventory of the conservation status of species in a particular region, according to criteria defined by the International Union for the Conservation of Nature and Natural Resources (IUCN). Categories of conservation status include: extinct, extinct in the wild, critically endangered, endangered, vulnerable, near threatened, least concern, data deficient, and not evaluated. The IUCN Red List Program aims to identify and document those species most in need of conservation attention if global extinction rates are to be reduced and to provide a global index of the state of degeneration of biodiversity. reef A strip of coral that rises close to the surface of the water. refugia (singular refugium) Locations of relictual populations of species that were once more widespread. regional endemism See ENDEMISM. reinforcement In evolutionary biology, the development of prezygotic barriers in response to selection against interspecific mating. relaxation A decrease in species diversity over time on a continental island as a result of local species extinction without compensating immigration, associated with the decrease in area subsequent to island formation. relic A human artifact or tradition that remains from a past era. relict A surviving member of a lineage or other natural phenomena. A relict endemic is equivalent to a paleoendemic. relictual Describing or referring to those organisms originally occurring in a location prior to fragmentation, which subsequently form part of the fragment biota; one of three classes of taxa inhabiting patchy landscapes. Ramsar site

1010

GLOSSARY

A mixture of endemics within an island biota with differing degrees of phylogenetic divergence from non-island relatives as a result of different colonization times and lineage histories. Remote Oceania The islands of Oceania east of Santa Ana in the eastern Solomon Islands, including Vanuatu, New Caledonia, Fiji, Micronesia, and Polynesia. rescue effect The prevention of extinction of a small insular population (of a species) by the occasional influx of individuals from another (e.g., mainland) area. research station A scientific facility to support research and educational programs focusing on the natural world and human interactions with it. Unlike institutions on the mainland and larger islands (e.g., universities or museums) that have permanent scientific staff, research stations (field stations and marine laboratories) provide permanent infrastructure typically for visiting researchers and students. resistance The ability to be unaffected by something; for example, a host plant may resist the activity of a pathogen. resource depression An effect of heavy predation pressure on one or more natural resource species, typically indicated by decreases in the abundance of larger or more desirable prey, or by changes in the demographic structure of prey populations (e.g., decreased number of older adults). retro-arc basin See BACK-ARC BASIN. rhyolite A light-colored, typically glassy, differentiated volcanic rock made of alkali feldspar and quartz with few iron-bearing minerals; erupted at temperatures of 700 to 850 °C. Rim of Fire The ring of volcanoes around the Pacific Ocean, which is also the scene of much earthquake activity. Rodinia supercontinent The oldest known supercontinent, which contained most or all of Earth’s land mass at the time. It was formed during the Precambrian supereon between 880 and 1100 million years ago through an event known as the Grenville orogeny and began to rift apart no later than 750 million years ago. runaway sexual selection An idea first proposed by the British geneticist R. A. Fisher to explain the evolution of extreme traits, generally in a single sex (e.g., the tail of a male peacock); these traits may result from sexual preference that thus confers reproductive success. sacbé A raised causeway built by the ancient Maya civilization. The limestone used in the construction of these roads gave them the white appearance for which they relictual series

were named. (From the Maya sac, meaning “white,” and bé, meaning “road.”) Saladoids The earliest ceramic-making horticulturalists to settle islands in the Caribbean. Known by their distinctively styled pottery, with occupations that lasted from about 400 BC to AD 600. saltation In genetics, a modification that is expressed as a profound phenotypic change across a single generation and results in a potentially independent evolutionary lineage. sandstone A consolidated sedimentary rock composed of sand particles. saprobe An organism that derives its nourishment from decomposing organic matter. scarp A steep erosional feature in the beach face, resulting from rapid beach retreat during high-wave activity. scleractinian A stony coral (order Scleratinia). sclerites Needle-like calcium carbonate skeletal elements secreted by the tissues of octocoral polyps and tissues connecting the polyps of a colony; scattered in the tissues or fused to form a solid structure around or under the polyps and connecting tissues, they serve for protection and/or support. sclerophyll forest Woodland that develops in regions that receive less than 1100 mm of precipitation per year and where the dry season can last for several months in a row. Also, TROPICAL DRY FOREST. sclerophyllous Of plants, having hard leaves and short distances between leaves along the stem. scoria cones Volcanic cones formed by explosive eruptions. SCUBA Acronym for self-contained underwater breathing apparatus. seabird Any of several groups of birds specifically evolved to life at sea, including penguins, albatrosses, shearwaters, petrels, storm petrels, pelicans, cormorants, boobies, gannets, frigate birds, tropicbirds, gulls, terns, auks/ alcids (including auks, auklets, dovekies, guillemots, murres, murrelets, puffins, and razorbills), and jaegers. seafloor spreading Gradual divergent movement of tectonic plates at mid-ocean ridges resulting in the formation of new oceanic crust through volcanic activity. sea ice Ice created when the surface of the ocean freezes. Through the process of brine rejection, sea ice is largely freshwater ice, although the initial freezing of saltwater requires temperatures at or below –1.8 °C. seamount An underwater mountain, generally of volcanic origin, that arises from the sea bottom but does not reach the ocean surface; after sustained growth, may produce an oceanic island volcano.

See ANADROMOUS. sea skater A pelagic marine insect, family Gerridae. sector collapse The collapse of a portion of a volcano, typically resulting from an earthquake, rising magma, or abundant precipitation in combination with high relief, steep slopes, and unstable or altered rock. sedimentary breccia Poorly sorted rock dominated by angular fragments not greatly abraded during their deposition (as opposed to conglomerate, which has rounded or subrounded fragments). sedimentary rock Rock that forms either as a result of erosion of pre-existing landscapes and deposition of the resulting detritus in a range of settings, or from precipitates and organic remains in seas and lakes. seed dispersal The scattering of seeds away from each other and from the parent plant. seedling recruitment In plant ecology, a population-level process beginning with seed germination and continuing through an early period of high seedling mortality, to the establishment of survivors with a relatively high probability of survival. seismic tomography The use of seismic waves to identify velocity variations and, hence, three-dimensional mantle structures within the Earth. semelparous Having only one set of progeny in a lifetime; in angiosperms, flowering once. semi-fossorial Describing or referring to a species that forages by burrowing but that also spends some time on the surface of the ground. senescence 1. The condition or process of aging. 2. In plants, the growth phase from full maturity to death that is characterized by accumulation of metabolic products, increase in respiratory rate, and loss of tree foliage out of season. sessile Permanently attached to the substrate. sexual dimorphism Any differences in morphology between the male and the female of the same species. sexually dimorphic characters Morphological, behavioral, or other differences between males and females of animal species. shamanism An animistic religion of northern Asia in which the shaman mediates between the visible and spiritual worlds. (From a Siberian word for “a person possessed by spirits who has mastered them.”) shear See WIND SHEAR. sheeted-dike complex A rock unit that consists almost entirely of near-vertical, parallel dikes (i.e., vertical planar fissures up to several meters wide filled with igneous rock, typically diabase). Sheeted-dike complexes are sea-run

GLOSSARY

1011

locally found in ocean crust and are very characteristic of seafloor spreading environments. shield volcano A gently sloping volcano with the shape of a flattened dome, built almost exclusively of lava flows. shifting baseline In conservation, the concept that each human generation believes that current populations or ecosystems are at a natural state despite declines that have occurred over many generations. shingle A type of sediment that is common to coral reef islands and is made up mostly of fragments of branching corals. short-range endemism See ENDEMISM. sill A flat sheet of igneous rock injected between existing rocky strata. sister species Distinct species that descended from a single common ancestor. Site of Special Scientific Interest (SSSI) A nationally designated site providing statutory protection for flora, fauna, or geological or physiographical features. In Ireland, such a site is called an Area of Special Scientific Interest (ASSI). skerry A rocky islet or reef (Scottish). skylight A hole in the roof of a lava tube or lava tube cave, usually formed by collapse. SLOSS debate A debate in conservation biology as to whether it is preferable, for the conservation of biodiversity in a fragmented landscape, to protect a single large or several small (SLOSS) reserves. socioecosystems Managed landscapes that have become inextricably linked with the human societies that inhabit them, through dynamically coupled human-natural processes. spawning The often synchronous release of eggs and sperm by numerous coral species during specific lunar phases and seasons. Special Protected Area (SPA) A site designated in accordance with the Birds Directive set up by the European Union in 1979, which aimed to conserve all species of naturally occurring birds and their habitats. speciation The evolutionary process through which new species originate by descent and divergence from an ancestral species. species–area relationship The relationship of area to number of species. As the area of an island or other potential (isolated) habitat increases, the number of species present also increases. species richness The number of species in an ecological community; one of several important measures of biodiversity. Other measures of biodiversity

1012

GLOSSARY

include genetic diversity and individual behavioral attributes. species turnover The replacement over space or time of one species or species assemblage by another. speciose In evolutionary ecology, describing or referring to species-rich taxa. speleothem A natural formation inside a cave. spermaceti A wax from the barrel-shaped organ in a sperm whale’s head, used to make fine candles that were the standard for the candlepower, a measure of artificial light. spillover The emigration of adult or juvenile fishes or other marine organisms outside a marine protected area as a result of density-dependent effects such as competition or aggression. spiritualism A religious movement that was popular in the 1840s–1920s, which centered around the belief that the spirits of the dead can be contacted by mediums and can provide guidance to those inhabiting the living world. sporophytic self-incompatibility (SI) The inability to form progeny from self-fertilization or from mating with plants that share either allele at the SI locus (called S-alleles), regardless of which allele is carried by the pollen. stable isotopes Versions of elements that are not radioactive but have differing numbers of neutrons and therefore slightly different weights; studies of the ratios of these elemental forms, such as oxygen, nitrogen, and carbon, provide useful measures of variations in past climates. stalactite A cylindrical or conical protrusion from the roof of a cave. stalagmite A cylindrical or conical protrusion from the floor of a cave. state factors Factors that control the characteristics and dynamics of terrestrial ecosystems, such as time, substrate parent material, substrate age, relief, topography, and regional flora and fauna. statistical model A model containing parameters that are varied to fit a set of observations. still water level The level of the water in the absence of wave motions. stiltroots Prominent arching, generally branched roots growing from the trunk of some mangroves into the substrate, facilitating the movement of oxygen into the root system. stochastic extinction Extinction of a species because of random events; the opposite of deterministic extinction. stock In conservation management, a species, subspecies, or geographical grouping that can be managed as a unit.

The abnormal elevation of sea level caused by the pressure fall and winds associated with a tropical cyclone. stratocone A volcanic cone consisting of both lava and pyroclastic rocks. stratovolcano See COMPOSITE VOLCANO. structure In geology, the arrangement and disposition of rocks in the Earth’s crust related both to Earth movements and to intrinsic morphological features such as joints. stygobiont Of aquatic species, inhabiting underground or cave waters and adapted to life in permanent darkness. subaerial In volcanology, erupted on the land surface, as opposed to subaqueous. Subantarctic A biogeographic region in the Southern Ocean, north of the Antarctic region and south of the cool temperate region, characterized by large expanses of ocean with scattered small islands and strong connections between marine and terrestrial ecosystems. Its terrestrial vegetation has no trees or shrubs. subaqueous Formed underwater. subduction The process in which one of the plates that make up the Earth’s crust is carried down into the Earth’s interior beneath another plate with which it is converging. Disturbance to the rocks of the mantle by the release of volatiles or melt from the subducted slab may cause partial melting of the mantle and may trigger volcanic activity. Also, OVERRIDING. subduction complex A mappable terrane composed of deformed sediments and associated volcanic rocks scraped off a downgoing oceanic plate, accreted to the overlying plate, and metamorphosed under relatively high pressures and low temperatures. subduction zone An area where two tectonic plates meet and one slides beneath the other, carrying materials from the Earth’s surface deep into its interior. subfossil Bones or other materials from organisms that have been dead for some time but whose fossilization process is not complete; usually of relatively recent geological origin. subsidence Sinking of crust, typically caused by thermal contraction or crustal thinning. subsidized population A population that is artificially increased either directly or indirectly through human agency (e.g., as from feeding on an introduced species or by consuming garbage from a landfill). substrate The base on which an organism lives. succession Predictable changes in a community over time. Development of complex communities on new substrates is termed primary succession, whereas secondstorm surge

ary succession is the process of ecosystem development on previously vegetated sites that have been cleared. sulfide A general term for chemical compounds containing sulfur in it lowest oxidation state (e.g., hydrogen sulfide [H2S]). sulfophilic “Sulfur-loving” (i.e., thriving in habitats containing high concentrations of sulfide). Sundaland The area of land formed by sea-level falls during the glaciations, uniting Borneo, Palawan, Sumatra, Java, and Bali with the Malay Peninsula and mainland Asia. Sunda Shelf The southeasternmost extension of the Eurasian continental shelf, the eastern boundary of which is marked by Wallace’s Line. sunderbans The largest mangrove forest in the world, situated at the mouth of the Ganges River in India. It is crisscrossed by networks of tidal waterways that drain mudflats and create islands within the forest. supercolony In ants, a group where aggression between worker offspring of conspecific queens is absent, contributing to high worker density over very large areas. superplume A large-scale cold downwelling or a hot upwelling flow in the mantle. supersaturated Having a larger number of species than can be maintained at equilibrium. Supersaturation generally occurs prior to relaxation. suspect terrane A geologic entity of unknown origin. sustainable development Development that generates sufficient income to purchase material and non-material goods from the modern cash economy to make life healthier, safer, more productive, and more enjoyable, without destroying the local natural and cultural capital needed for the material and cultural survival of future generations. sverdrup A unit of river and ocean current flow, equal to 1 million m3 of water per second. swath mapping Multibeam sonar mapping of the bathymetry and acoustic backscatter of the ocean floor. sweepstakes route A rare, chance colonization after a long-distance dispersal. Only a small number of individuals arrive and successfully establish on isolated islands. Their success is often followed by adaptive radiation. syenite A coarse-grained rock with abundant feldspar; formed by intrusion of trachyte magmas. sympatric 1. Having broadly overlapping geographic ranges. 2. Specifically, living in the same local community and able to interact. sympatric speciation/sympatry Species formation occurring when populations diverge within the same geographic region.

GLOSSARY

1013

See PORTMANTEAU BIOTA. A concave fold in rock, the central part of which contains the youngest section of rock. tacking A maneuver in sailing in which the bow of the boat is turned into the wind, causing it to fall off on the opposite site, which then allows the craft to change course. target area effect The effect of island area on the rate of immigration, by which larger islands provide easier targets for passively dispersing propagules, thus enhancing colonization. taxon (plural taxa) A clade, usually a named one. taxon cycle Sequential phases of expansion and contraction of the ranges of species, associated generally with shifts in ecological distribution and differentiation. Stages range from expanding (Stage I), to differentiating (Stage II), to fragmenting (Stage III), and to endemic (Stage IV). tectonic earthquake A rupture in the stiff, outermost part of the Earth (lithosphere). Tectonic earthquakes are triggered by the movement of tectonic plates relative to one another. tectonic plates Pieces of the Earth’s lithosphere (crust and upper mantle) that move in relation to one another on the Earth’s surface. Earthquakes and volcanoes commonly occur along the boundaries between plates. tectonic processes Large-scale movements of the Earth’s crust, during which rocks are folded and faulted. tectonics The study of how movements of the Earth’s surface relative to its center shape the forms observable on the surface. tectonostratigraphic terrane A group of rocks that have a common history in terms of both their tectonic and their sedimentary evolution. teleseism A distant earthquake, as opposed to a “local” earthquake. tephra Fragments of volcanic rock and lava of any size expelled from a volcano (e.g., ash, bombs, cinders). tephrochronology A method of using ash layers (tephra) of known age as marker horizons in soil or sediment sections. tepui A steep-sided mesa or table-shaped mountain, characteristic of Venezuela and elsewhere in the north central Guyana shield. terrane A part of the Earth’s crust that is bounded by faults on all sides and that differs from adjacent parts of the Earth’s crust by the nature of its geology and geological history. terrigenous Derived from the land; in most cases, terrigenous sediment is siliciclastic (i.e., noncarbonate). synanthropic biota syncline

1014

GLOSSARY

The geologic subera extending from about 65 million to 1.75 million years ago, including the Paleocene to Pliocene epochs. Tethyan Pertaining to the age when the Tethys Ocean existed between Laurasia and Gondwana during the Mesozoic era until its closure in the Tertiary (Miocene). In relation to terrestrial vegetation, the term is used with reference to a wide “Madrean– Tethyan belt” of sclerophyllic taxa that, based on fossil evidence, occupied a large subhumid region across the Holarctic in this period (mainly Eocene– Miocene); a hypothesis often used to explain strong disjunctions. tholeiitic magmas Magmas that are relatively rich in silicon and poor in sodium and potassium, usually generated by larger extents of melting of the mantle. Tholeiite or tholeiitic basalt is the rock name. thrashers Large, noisy, mostly-terrestrial birds, family Mimidae, with a chiefly southwestern North American species radiation. threatened species A species that faces a risk of extinction in the near future. The best-known international list of threatened species is the IUCN Red List. tillite The rock equivalent of till, a sediment laid down by glaciers. trachyandesite A fine-grained differentiated volcanic rock with relatively high silica content and consisting mainly of sodium-rich plagioclase and alkali feldspars. trachybasalt A dark, fine-grained volcanic rock with higher silica content than basalt, which consists of calcium-rich plagioclase and alkali feldspars in equal proportions. trachyte/trachytic rock A gray, fine-grained volcanic rock with high silica content (~60%) and consisting largely of alkali feldspars. It usually has a rough and gritty surface. trade wind inversion The inversion layer associated with the trade winds, characterized by increasing temperature and sharply decreasing relative humidity; it represents the boundary between heated surface air and sinking air aloft. trait Any distinct or quantifiable feature (structure, behavior, or physiological or developmental process) of an organism. transform fault/transform plate boundary A plate boundary at which two plates slide past one another horizontally. transmutation The alteration of one species into another. Tertiary

A deep depression on the ocean floor, produced when a tectonic plate sinks upon colliding with another tectonic plate. trilobite An extinct fossil marine arthropod characterized by an exoskeleton consisting of three sections. troglobionts Terrestrial species that are obligate subterranean dwellers. troglomorphic Pertaining to the morphological, behavioral, and physiological characteristics that are convergent in subterranean species. trophic structure The ratio of the numbers of species in different dietary categories, such as herbivores, scavengers, nectar feeders, predators, parasites, and so on. tropical dry forest See SCLEROPHYLL FOREST. trough A feature of seafloor topography; a narrow depression that is less steep than a trench. tryworks The equipment used to render oil from the blubber of whales. tsunami A series of waves that travel across the surface of the ocean, generated by impulsive disturbances of the sea floor. Tsunamis are most often associated with submarine earthquakes or landslides. tuff A consolidated rock composed of pyroclastic fragments and fine ash. turbidite A sediment or rock deposited from a turbidity current, a bottom-flowing density current of suspended sediment moving quickly down a subaqueous slope and spreading out at the base of the slope, with deposits characterized by a fining-up grain-size distribution and changes in cross-stratification type that indicate an upward waning of current velocity. turnover The product of opposing rates of species immigration to, and species extinction from, an island. Where these rates precisely balance, a dynamic equilibrium is formed, with species richness remaining constant through time, but with composition continually changing. turnover rate The number of species eliminated and replaced per unit time. tussock Grass that grows in clumps, not evenly. tuya A flat-topped, steep-sided volcano formed when lava erupts through a thick glacier or water. ultrahigh-pressure metamorphism (UHPM) A type of metamorphism observed at many plate collision zones. Metamorphics contain relic grains of coesite or microdiamond, indicating formation at depths as great as 100 km, commonly found in mountains uplifted by continent– continent collision. ultramafic Composed entirely of mafic (magnesiumand iron-rich) minerals. trench

A metalliferous soil resulting from the weathering of ultramafic rocks and characterized by low nutrient and (on New Caledonia) high nickel concentration. underground water Accumulation of water underground, due to fast percolation of rainfall and surface water into porous soils and bedrocks. Occurs on islands where sandy deposits, limestone rocks (including coral), or volcanic layers are dominant. UNESCO The United Nations Educational, Scientific and Cultural Organization, whose mission is to encourage the identification, protection, and preservation of cultural and natural heritage around the world considered to be of outstanding value to humanity. ungulates Hooved animals such as pigs, deer, or cattle. uplift The rise of part of the Earth’s crust relative to its center. upwelling A process by which warm, less dense surface water is drawn away from a shoreline by offshore currents and replaced by cold, denser water brought up from the subsurface. vagile Highly mobile with good dispersal potential; the opposite of sessile. vagility The capacity of animals to disperse, either under their own power or passively (e.g., as eggs or cysts). varanid A lizard of the family Varanidae, the monitor lizards. The Komodo dragon is the world’s largest varanid. vascular plants Ferns, flowering plants, and other plants that circulate water and nutrients in conductive (vascular) tissue called phloem and xylem. vector 1. An organism that transports another organism from one place to another. 2. A species such as a mosquito that transmits a disease between hosts. vegetable sheep A prostrate, mat-forming perennial New Zealand herb with woolly, hoary leaves. vegetative reproduction Asexual plant reproduction in which new individuals (ramets) arise without the production of seeds. VEI See VOLCANIC EXPLOSIVITY INDEX. vein In geology, a fissure, crack, or channel in rock or ice that has been filled with a mixture of minerals. velamen radicum In desiccation-tolerant vascular plants, a specialized outer layer of mostly aerial roots, consisting primarily of dead cells that rapidly absorb water. vent The opening at the Earth’s surface through which volcanic materials are ejected. vesiculation Exsolution of dissolved volatile components (water, carbon dioxide, etc.) from magma to form ultramafic soil

GLOSSARY

1015

bubbles; the energy provided by bubble expansion provides much of the energy of volcanic eruptions. vicariance The separation of closely related organisms by a natural barrier or a vicariant event, such as a body of water or the rise of a mountain range isolating drainages, resulting in differentiation of the original group into new varieties, species, or other taxa. volant Capable of flight. volcanic arc A row of volcanoes formed where one of the plates that make up the Earth’s outer shell is subducted below another. The cause of the volcanic activity is the release of volatiles or melt from the subducted slab. This causes partial melting in the adjacent mantle, and the melt rises to the earth’s surface. The volcanoes typically lie along an arc when seen in plan view. volcanic bombs Molten rock fragments larger than 65 mm in diameter that are ejected during a volcanic eruption; bombs may acquire aerodynamic shapes during transport through the air. volcanic dome A rounded, steep-sided mound formed of very viscous magma, usually either dacite or rhyolite. Such magmas are typically too viscous to move far from the vent before cooling and crystallizing. volcanic earthquake An earthquake characterized by high-frequency seismic signals thought to be generated by the fracturing of rock in response to the intrusion and migration of magma. Volcanic earthquakes often occur in swarms and usually precede the onset of volcanic activity, although they do not always culminate in a volcanic eruption. volcanic explosivity index (VEI) A method for measuring the scale of an eruption by means of a 0 to 8 index of increasing explosivity. Wadati–Benioff Zone See BENIOFF ZONE. Wallace’s Line A biogeographic boundary that separates Asian from Australasian species. It runs through the Malay archipelago between Borneo and Sulawesi with various northern extensions, but its precise location has been modified a number of times. (First proposed by Alfred Russel Wallace when he observed the sudden changes in bird families between the islands of Bali and Lombok.) washover fan A fan-shaped sheet of sand deposited on top of a barrier island during an unusually high tide (either astronomically- or storm-induced). washover terrace A continuous sheet of sand formed where several washover fans merge together. wave The profile of the sea surface elevation between two successive downward zero-crossings of the mean sea surface.

1016

GLOSSARY

The process by which organized kinetic and potential wave energy is converted to turbulence and dissipated. wave-dominated coast A depositional coast where waves are the dominant process shaping the geomorphology of the coast (e.g., long barrier islands, numerous parallel beach ridges on arcuate-shaped deltas, or continuous sandy shores), usually in areas with small tides. wave-driven flows Currents that are driven by gradients in wave momentum, usually associated with wave dissipation by breaking or bottom friction. wave-formed bar A large, ridgelike bedform that develops in the surf zone seaward of sand beaches. wave group The association of several to many individual waves that vary in height from small to large to small again. wave group velocity The speed and direction at which a wave group propagates. wave refraction The bending of a wave as it moves into shallow water so that wave crests become parallel to bottom contours. wave transformation The process by which waves change as they move from deep to intermediate to shallow water. weatherly Capable of sailing upwind and relatively close to the direction of the wind. West Indies The Antilles island arc in the Caribbean Sea extending from Cuba south and east as far as, but excluding, Trinidad and Tobago. weta Any of a number of species of orthopterans belonging to the Anostostomatidae (king crickets) and Rhaphidophoridae (camel or cave crickets), most endemic to New Zealand. wetlands Areas with sufficient water to support vegetation adapted for life in saturated soils, including swamps, marshes, bogs, and similar areas. whale fall A common term for a whale carcass resting on the ocean floor. wildlife translocation A wildlife management technique in which a population of plants or animals is moved to a site where it is not present currently. wind shear Variation of the wind vector over a distance. Also, SHEAR. World Conservation Union See IUCN. World Heritage Site A site (such as a forest, mountain, or lake) that is on the list maintained by the international World Heritage Program administered by the UNESCO World Heritage Committee. wrack Seaweed floating in the sea or growing on the shoreline. wave breaking

In geology, an exotic fragment, typically coarsegrained, included in a volcanic rock (e.g., an upper mantle peridotite included in basalt). xeric Of, marked by, or adapted to, a very dry habitat. xeromorphic Adapted to drought and/or arid conditions, using strategies to store water in the leaves or stem (e.g., succulents). xerophytic Describing or referring to plants adapted to cope with limited water supply. Zealandia The continental crust of which New Zealand and some other islands form the modern emergent parts. Some 93% of Zealandia is today under the sea. zonation Living in zones; the restriction of particular species to one of the life zones (e.g., wet zone, dry zone, cloud forest) instead of being found all over an island. zooarchaeology The study of animal remains (bones, shell, teeth, etc.) to understand human diet, subxenolith

sistence practices, and other phenomena of ancient times. zoogeography A branch of biogeography that studies the distribution of animal species over space and time. zoonosis A disease of wild animals that can be transmitted to humans. zooplankton Small, free-swimming animals that drift in lake or ocean currents. Considered a pivotal component of food webs because they graze on algae and are a source of food for higher trophic levels such as aquatic insects and fish. zooxanthellae Unicellular dinoflagellate algae that live intracellularly within the gastrodermal cells of corals as symbionts and which provide much of the daily energy needs for the host coral through the translocation of photosynthetically derived metabolites. zooxanthellate Of corals, possessing zooxanthellae.

GLOSSARY

1017

This page intentionally left blank

INDEX

Boldface indicates main articles. Italics indicate illustrations and tables. For information about specific islands, see groups or regions to which they belong, and vice versa. For information about specific taxa, see other taxa to which they belong, and common names. ‘a‘a- , 543, 545, 711, 823 aboriginal introductions, 272, 273, 274, 275, 891 aboriginal peoples Australia, 114, 768 French Polynesia, 332 Rottnest Island, 796 Tasmania, 904, 907–908, 958, 959, 960 Tenerife, 742 whaling and, 975–976 Abrolhos Islands, 956 acclimatization societies, 473, 672 accretion Antarctic, 18 Antilles, 29, 30, 31 barrier islands and, 86 Borneo, 657 Canadian Shield and, 77 faults and, 903 Iceland, 429, 430 island arcs and, 483 Japan, 502, 503, 504, 505 Maldives and, 587 mangrove islands and, 592 New Guinea and, 653, 657, 661, 662, 664 New Zealand, 677, 678 Philippines, 738 Taiwan and, 902, 903 Vancouver Island, 708, 937 vicariance and, 950 waves and, 880 achenes, 227 acidification, ocean, 70, 379, 780, 782, 821. See also pH acoustical signaling birds, 106, 110, 111, 355, 399, 413, 713 crickets, 6, 207, 208, 209, 210, 211, 463, 465, 826 fishes, 826

flies, 6, 465 frogs, 27, 725 humans, 762, 764 mammals, 589 spiders, 842 tarsiers, 553 Acrididae, 670 Adamson, 541 adaptation, local, 52, 224, 226, 353, 355, 594, 911 adaptive convergence. See convergence, evolutionary adaptive radiation, 1–7. See also convergence, evolutionary; insular radiation; specific islands; specific organisms anagenesis vs., 8 autochthonous, 157, 465 competition and, 89 endemism and, 257 environmental factors and, 353 evidence for, 251–253, 257 genes and, 465, 477, 526 geology and, 221 gigantism and, 373 inbreeding and, 437 invasive species and, 475 isolation and, 13 lakes and, 527–528, 531, 606 sexual selection and, 828 taxon cycles and, 912 zones, 772–774 Aders’s duiker, 983, 986 Admiralty Islands, 349, 471, 569, 720, 721 Adriatic islands, 471, 564, 622, 623, 625, 626, 628 advertising, 765 adzes, 42, 43, 45, 46, 47, 247, 721, 722, 759 Aegean islands, 482, 623, 626–627, 628, 857, 859

Aegean Sea, 628 Aeonium, 4, 128, 129, 774 Africa. See also cichlid fish; North Africa; South Africa; specific locations ants and, 36, 38, 40 barrier islands, 83 bird divergence, 108–109 birds, 122 Canary Islands and, 129, 131 Comoros and, 178, 179, 180 conservation, 784 crickets, 208 dispersals, 560 Great Rift Valley lakes, 7, 165 mammals, 121 missionaries and, 637 monocotyledons, 467 Newfoundland and, 650 rabbits, 121 rock sequences, 66 spiders, 863 succulents, 468 vicariance, 950 African plate, 64, 65, 71, 142, 144, 215, 388, 392, 394–395, 482, 620, 622, 624, 753 African violets, 943–944 agamids, 180, 618, 869, 906 age of islands. See also progression rule ants and, 36, 37, 39 Azores, 71 beaches and, 94 Cooks, 191 cricket diversity, 209 deforestation and, 222 distribution of species and, 181 diversity and, 73, 320, 840 drowned islands and, 370 endemic taxa and, 948–949 endemism and, 254

1019

age of islands (continued) fossils and, 319 fragmentation and, 186–187, 330 French Polynesia, 342 freshwater species and, 343–344, 345 Gough Island, 930 granitic vs. oceanic, 381 Hawaii and, 398, 406, 407, 408 hotspot theory and, 339 insects and, 463, 464 landslides and, 340 Madagascar, 577 mantle plumes and, 368 motu, 642 New Zealand, 676 plant disease and, 750 pre-European impacts and, 417 radiation zone and, 773–774 rate of succession and, 879 Samoa, 802–803 speciation rates and, 210 spiders and, 864 Surtsey Island, 888 timing of species formation and, 412 Tristan da Cunha, 929 age of taxa, 319 agoutis, 90 agriculture. See also domestication; ethnobotany Britain and Ireland and, 117–118 Canary Islands, 130, 132 Cape Verde, 144, 146 Caroline Islands and, 148 Channel Islands (British), 155 Comoros, 180 conservation and, 224 Cook Islands, 195, 196 coral reefs and, 784 crickets and, 209 deforestation and, 222 Faroe (Faeroe) Islands, 291, 292 fragmentation and, 330 French Polynesia, 336 Greek Islands, 391, 393 Gulf of Guinea, 810 lava and, 543 Mediterranean, 629 Mesoamerica and, 184 missionaries and, 636 motu and, 642 New Guinea, 658 New Zealand and, 672 oases and, 688–689 overviews, 942 Phosphate Islands and, 740 plant disease and, 749 pre-European, 417 Samoa, 801 Seychelles, 832 Socotra, 850 Solomons and, 853, 854 St. Helena, 873 sustainability and, 889–890 Taiwan, 901 Tonga, 919

1020

INDEX

Tristan da Cunha/Gough, 931, 932 Vanuatu, 941 Zanzibar, 984, 985 Agrihan, 594, 600 Aguijan, 594 Ailsa Craig, 125 Ainu people, 522 air flow, 172–173 air transportation, 477, 479 Aitutaki Island, 192, 193, 194, 195, 196, 342 Alaid, 484 Alamagan, 598, 599, 601 Alaska ash, 543 global warming and, 973 landslides, 536 missionaries and, 634, 637 plants, 938 sea levels and, 48 shipwrecks, 834 albatrosses, 10, 12, 16, 104, 358, 574, 632, 672, 794, 811, 812, 813, 814, 932 Albertine Rift, 513 Alcatraz, 770 Alcids, 813 Aldabra, 179, 577, 830, 831, 832, 833, 921, 922, 924, 926 Alderney, 154 Aleipata, 799 Aleutian Islands, 52, 481, 483, 702, 705, 814, 874, 934, 977–978 Alexander archipelago, 52 Alexander Island, 11, 17, 18, 19, 874 algae Antarctic, 12 arctic, 50, 55, 56 atolls and, 68 Bermuda, 96 Canary Islands, 131 Cape Verde, 146 Cook Islands, 194 coralline, 199 ephemeral islands, as, 259 Farallon Islands, 296 French Polynesia invasive, 337 Japan, 497 jellyfish and, 716 Macquarie, 574 New Guinean beetles and, 656 organic falls of, 701 overfishing and, 310 Philippines, 724 Rottnest Island, 797, 798 Surtsey Island, 886 sustainability and, 891 Tatoosh, 909, 910, 911 waves and, 880 algal ridges, 68, 205, 206 algarrobo trees, 23, 27 Ali, 968 alien species. See also exotic species; introduced species; invasive species Antilles, 23, 26 Ascension, 62 Azores and, 71, 74

Bermuda and, 97, 98 Britain and Ireland and, 118 climate change and, 15 Darwin on, 959 dispersal and, 228 freshwater species and, 346 Hawaii and, 36, 944 lakes and, 531 Macquarie, 574 sub-Antarctic, 14–15 Tonga, 920 Vancouver, 938 alkalinity, 176–177 Allee effects, 478 Allen, Charles, 968 allopatric speciation. See also isolation caves, 151 climate change and, 786 fish, 166 Galápagos, 773 Indonesia, 452 insect, 81, 123, 462 lizards, 561, 563, 564 moa, 640 neotropical, 786 refugia and, 786 sexual selection and, 826 size of island and, 859 snails, 539 spiders, 864 Wallacea, 452 weta, 669 allopolyploid speciation, 255, 838 de Almeida, Fernando, 298 aloes, 132, 850 Alpine Fault, 674, 754 altitude. See elevation altruism, 225, 226 aluminum ore, 149 Amanu, 334 Amazon, 964 amber, 23, 27, 38, 561 Amblypygi, 849 Ambon, 448, 660 Amborellaceae, 644 American Samoa, 104, 378, 637, 712–713, 783, 799, 801, 802. See also Samoa Amerindians, 23, 24, 25 Amirante Island, 177 ammonites, 937 Ampère seamount, 64 amphibians Antilles, 27 Bermuda, 98 Borneo/New Guinea and, 658 Britain and Ireland and, 122 Channel Islands (California), 159 Comoros, 179 Fijian, 301 Gulf of Guinea, 809 Indonesia, 449, 450 Japanese, 499 Kurile Islands, 525 Madagascar, 580 New Guinea, 653, 654, 656

New Zealand, 667 overviews, 347 Philippines, 724, 729 pigs and goats and, 742 Seychelles, 830, 831, 833 sustainability and, 891 Taiwan, 899 Vanuatu, 940 Wallacea, 452 Zanzibar, 983 amphipods, 151, 701, 907, 907 Amsterdam Island, 320, 441, 789 Amundsen-Scott station, South Pole, 15 Anacapa Island, 156, 157, 158, 161, 162, 163, 795 anagenesis, 8–10, 698, 773 Anak Krakatau, 206 analogues, 619 Anatahan Island, 598, 598, 599, 599, 602, 952 Anatolian plate, 388, 392 Andaman Islands, 114, 439, 771 Anderson, Edgar, 415 Andes, 217 Andesite Line, 951 andesites, 31, 694, 754, 755 Adriatic islands, 471 anemones, 700, 853, 909, 910 anerobic oxidation of methane (AOM), 176–177 Ángel de la Guarda Island, 80, 81, 82 angiosperms, 145, 146, 193, 360, 466, 644, 667, 671, 709, 713, 800, 899, 920 Anglesey, 125 Anguilla, 22, 24, 29, 31 Anjouan Island, 177, 178, 179, 180 Annamite range, 840 Annobón Island, 65, 808–811 Anolis (anoles) amber, in, 23 colonizations, 560 convergence, 189, 190 dwarfism, 236 insular radiation, 3, 5, 7, 27, 531, 561–564, 773 invasions, 477 isolation, 859 rafting, 775 sexual selection and, 826, 828 Anson, George, 509 Antarctic adaptive radiation, 13 biology, 10–16 bird disease and, 105 endangered species, 908 flightless insects, 227 flightlessness and, 317 geology, 17–20, 504 global warming and, 815 island arcs and, 481 research stations, 789 rock sequences, 66 antelope, 51, 120, 983 anthropogenic disturbances. See human impacts Antigua, 23, 24, 31, 32, 35, 278 Antilles adaptive radiation, 24, 27

biology, 20–29, 36, 190 colonization, 278–279 convergent evolution, 7 geology, 29–35 Antipodes, 11, 16, 675 ants, 35–41 adaptive radiation, 37–38 Antilles, 23 Barro Colorado, 88, 91 Britain and Ireland and, 122, 123 crickets and, 211 Fiji, 36, 38, 300, 775 Hawaii and, 865 invasive, 39, 40, 472, 473, 477, 479, 708 Madagascar, 580 maximal radiation, 773 Melanesia, 38, 253, 912 New Caledonia, 645 radiation zone and, 772 silverswords and, 838 Vanuatu, 941 Anvers Island, 17, 18, 19 Anzhu Island, 51 aphids, 49 apomictically reproducing plants, 118 apomixis, 697 Appalachians, 152, 649, 650, 652 aquaculturists, 346, 347 aquatic species. See also fish; marine life and environments; seabirds cave, 150, 151 dispersal, 528 invetebrates, 131 New Guinea, 656 plants, 101 vertebrates, 22 Arabia, 104, 146, 215 Arabian plate, 753 Arabs, 832, 984 arachnids, 130, 146, 300–301. See also spiders Arago sea mount, 342–343 Arakamchechen Island, 51 Arapawa Island, 675 arboriculture, 224 archaeology, 41–47. See also fossils ants and, 40 Bermuda and, 97 Britain and Ireland and, 119, 120, 121, 124, 126 Caribbean, 279 Caroline Islands, 149 Channel Islands (California), 160, 161 Cozumel, 204 Europe, in, 117 Faroe (Faeroe) Islands, 287–288 languages and, 759 Micronesia, 722 Oceania, 42 pigs and, 741 Polynesian voyaging and, 759 pre-European impacts and, 417 Sarawak caves, 114 seafaring and, 276–277 Sri Lanka, 867 whaling and, 975 Zanzibar, 984, 985

archeophytes, 119 archipelagoes. See also specific archipelagoes adaptive radiation and, 3 anagenesis and, 9 biodiversity and, 617 caves and, 153 glaciations and, 785 insect radiations and, 460, 462–463 mammals and, 588 plant disease and, 749 refugia and, 786 architecture Faroe (Faeroe) Islands, 292 Polynesian, 46–47 Rapanui, 246 Solomons, 853 whaling and, 976 arcs. See island arcs Arctic fishes, 874 seabirds, 814 whales, 976 Arctic islands stereotypes, 761 arctic islands, biology, 47–54 arctic islands, geology, 55–59 Arctic Ocean, 55, 58, 86 arctic region, 59–61 birds, 126 arctic vegetation Britain and Ireland and, 124 area factors. See also size of islands; species-area relationship (SAR) equilibrium and, 487 island rule and, 495 oases and, 688 Argentina, ants and, 40 Argyranthemum, 4, 129, 583 Arizona, 839 armor, 874, 876 Armorican Massif, 154 Arran Island, 116, 118, 125, 255 art, 763 arthropods, 3, 13, 907 ants and, 39, 40 arctic, 51, 53 Azorean, 71, 73, 74 Borneo, 115 Canary Islands, 130, 132 Cape Verde, 145 cave, 548–549 Fijian, 300 French Polynesian endemics, 335 gigantism, 375 Hawaii, 36 lava cave, 548 Madeira, 584 matrix-derived taxa, 182 Midway Island, 632 New Caledonia, 645 Solomons, 854 sub-Antarctic, 15 Vanuatu, 940 volcanism and, 862 Aruba, 29

INDEX

1021

Aru Islands, 448, 653 Ascarina, 940 Ascension Island, 61–63, 66, 218, 320, 789 geology, 951 introduced species and, 471 invasive species, 959 research stations, 789 spiders, 862 volcanism, 959 asexual reproduction, 285 ash, volcanic, 543–544, 664, 744, 884 Asia ants and, 36, 38 arctic and, 51 biota, 114 birds, 122 conservation, 784 crickets, 208 frogs, 7 mammals, 121 missionaries and, 637 prisons, 770–771 Asian plate, 482 Asians, 149 asteroids, 700 asthenosphere, 752–753 Asuncion, 595, 598, 599, 600 Atiu Island, 192, 193, 194, 196, 197 Atkinson, Carter, 414 Atlantic Ocean, 63, 288, 379, 426, 482, 650 Atlantic region, 63–67 bird radiation, 106 Canary Islands and, 131 coral reefs, 200, 781 fishes, 874 flightless birds, 314, 316–317 mammals and, 588 overfishing and, 311 research stations, 789 seamounts, 819, 823, 824 vents, 424 atmospheric features, 172 atoll lagoons, 219 atolls, 67–70 Caroline Islands, 149 Cook Islands, 193, 194 Darwin and, 220 Darwin on, 219, 958 flora, 276 French Polynesian, 711 geology, 754 Hawaiian, 631 Indian Ocean region, 442 Maldives, 587 Marshall Islands, 712 motu and, 641 overviews, 694 Pacific region, 704, 708–709 snails, 949 Solomons, 855 tsunamis and, 934 volcanism and, 950 western Pacific, 715 Auckland Islands, 11, 12, 14, 16 geology, 675, 679

1022

INDEX

weevils, 13 auklets, 813 Channel Islands (California), 159 Farallon Islands, 297 auks, 124, 811, 813 Britain and Ireland and, 124 Aure Fold Belt, 664 aurocks, 120–121 Britain and Ireland and, 120 Australasia, 971 Gondwana and, 669 Austral Fracture Zone, 340, 343 Australia aborigines, 114 barrier reef, 200 beetles, 7 biocontrol and, 102 biotas, 114 bird disease and, 105 birds, 6, 304, 312 crickets, 208 Darwin and, 958 extinctions, 170 flies, 571 frogs, 348 geology, 447, 448, 458, 504, 755 global warming and, 380 Indonesia and, 453 inselbergs diversity, 469 introduced species, 470, 477 island arcs and, 485 island rule and, 493, 495 lava, 545 monitor lizards, 513–514 New Guinea and, 656 New Zealand and, 677 orchids, 697 pigs and goats and, 743 plant disease and, 748 prisons, 769 seafaring and, 277 shipwrecks, 834 snakes, 844 spiders, 863 Wallace and, 968 Wallace’s Line and, 970 wet forest, 842 Australian Island Territories, 572 Australian plate, 448, 753 Indonesia and, 459 Macquarie Island and, 575, 576 New Caledonia and, 648 New Guinea and, 653, 660, 664 New Zealand and, 674, 675, 676, 679 overviews, 705 Solomons and, 703 Tasmania and, 904 Australian region lizards, 560 Austral Islands, 709 architecture, 47 Cook Islands and, 192 endemism, 336 geology, 342–343 languages, 759

overviews, 332 volcanism, 953 Australo-Melanesians, 114 Austronesians, 42, 43 Borneo and, 114, 115 Near Oceania and, 721 seafaring and, 277 Taiwan and, 900 authigenic carbonates, 176–177 autochthonous, 660 autochthonous differentiation, 255 autochthonous speciation, 729 Avars, 54 Aves Ridge, 21, 30 avian influenza, 105 avian malaria and pox, 103–104 avifauna. See birds Axel Heiberg Island, 50, 58, 59 Axial Seamount, 713 Azores archipelago, 64, 70–74, 127 birds, 106 coral conservation, 567 Darwin and, 220 Drosophila, 234 eruptions, 952 geology, 755, 951 invasive species, 346 spiders, 862 volcanism, 959 whaling, 976 Azores-Gibraltar fracture zone, 64, 133 Azores Plateau, 424, 426 Babeldaob Island, 149 back-arc basins Antarctic, 19 Antilles, 30 Atlantic region, 66 Caroline Islands, 149 Fiji, 306, 307 Indonesia, 459 Japan, 486, 502, 506 Mediterranean, 626 Pacific region, 481, 483, 484, 485, 486, 706 Philippines, 733 vents and, 424 back-colonizations, 254, 464, 585 Backer, C. A., 517 bacteria, 974 inselbergs and, 467 lakes and, 530 natural selection and, 6 organic falls and, 700 Pantepui, 718 plant disease and, 748 succession and, 877 Surtsey Island, 886 sustainability and, 891 badgers, 120, 375 Baeropsis, Asteraceae, 78 Baffin Bay, 55, 58 Baffin Island, 48, 50, 54, 58, 60, 76–79 Bahamas ants, 859 climate change and, 170

geology, 651 global warming and, 378 hydroclimate, 423 rodents, 23 spiders, 860 watercraft, 279 Bahamian Bank, 22 Baja California, 78–82, 81, 157, 162, 703 Baker, Ian, 871 Baker’s Law, 697, 699, 838 balance of species richness, 486–487 Hawaiian, 944 Balcones Escarpment, 152 Balearic Islands, 559, 624, 628, 629 rats and, 794, 795 spiders, 862 Balfour, I. B., 848 Bali, 450–451, 970 Ballard, R., 424 Balleny Islands, 11, 17, 20, 423 Ball, Henry Lidgbird, 571 Baltica, 56, 504 Baltic archipelagoes, 750–751 Banda arc, 459 Banda Sea, 756, 757 bandicoots, 906 Bangladesh, 591 Banks Island, 50, 51 Banks, Sir Joseph, 219 Barbados, 24, 29, 31, 35, 483 Barbuda, 23, 24, 31, 35 Barentsz, Willem, 866 Barlavento group, 65 barnacles, 426, 909, 910, 957 Barnes Ice Cap, 77 Barombi Mbo (Cameroon), 166 Barra, 123–124 barrage lakes, 331 Barren Island, 484 barrier islands, 82–87, 92, 93, 211, 420, 474, 536, 673, 675, 743, 915, 916 barrier-reef islands, 333 barrier reefs, 69, 200 Cook Islands, 194, 195 Fijian, 299, 707 overviews, 219 western Pacific, 715, 716 barriers, insect diversity and, 463 Barro Colorado Island, 88–91, 181 Barton, C., 416 basalts adzes, 45 Antarctic, 19, 20 Antilles, 31 arctic, 55, 58 Atlantic, 64, 65, 66, 67 Baffin Island, 76 Baja California, 80 beaches and, 93 Borneo, 459 Canary Islands, 135, 136, 137, 138, 139, 140 Cape Verde, 144 Caroline Islands, 149 Channel Islands (British), 154 Channel Islands (California), 163, 164

Comoros, 177 Cook Island, 193 coral and, 200 Cyprus, 213, 214 Darwin and, 218, 220 Faroe (Faeroe) Islands, 288 Fernando de Naronha, 298 Fiji, 306, 307, 308 French Polynesia, 338, 339, 340, 342 Galápagos, 370, 371 Hawaii, 404, 406, 409, 807 Iceland, 428, 429, 430, 431, 432, 433, 434 Indian region, 437, 438, 439, 440, 441, 442, 443, 444, 445 Isla de Guadalupe, 377 island arcs and, 483, 484, 485 Japan, 503, 505 Kick ‘em Jenny, 510, 511 Kuriles, 522, 523 Laki vs. Tambora, 270 lava flows, 542, 543, 544, 545, 546, 547 Line Islands, 554, 711 Lord Howe Island, 568–569 Macquarie Island, 576 Madeira, 582 Mariana Islands, 598, 600, 601 marine species and, 378 Mascarenes, 613, 620, 622 Mediterranean Islands, 625, 626, 627 Mid-Atlantic Ridge islands, 959, 981 New Caledonia, 647, 648 New Guinea, 663, 664 New Zealand, 675, 679 North Atlantic, 288–289 oceanic islands and, 380, 381, 690, 693, 951 overviews, 690, 754 perched beaches and, 93 Pitcairn, 744 plate tectonics and, 752, 753, 754, 755, 822–823 Samoa, 803, 803, 804, 805, 806–807 São Tomé Island, 809 seamount, 821, 822, 825 Solomons, 855, 856 St. Helena, 870, 871 Surtsey Island, 883 Taiwan, 897, 898 volcanism and, 339, 693–694, 951, 953 basins. See also back-arc basins; specific basins East Asian, 505–506 overviews, 752, 756 Philippines, 735 seamounts and, 824 Solomons, 854, 856 Basque people, 976 Bates, Henry Walter, 963 bathymetry anomalous, 951 arctic islands, 55 Azores, 71 Canary and Madeira Islands, 133 Canary and Selvagen Islands, 141 Caroline Islands, 148 Channel Islands (California), 156 Crater Lake, 979, 980

French Polynesia, 339, 341 Galápagos, 367 Indonesia, 447, 454, 455 New Zealand, 674, 678 ocean circulation and, 825 Philippines, 727, 728 Samoa, 802, 806 seamounts, 714 Sunda shelf, 112 tsunamis and, 933 waves and, 882 bathyspheres, 97 bats, 79, 81 Antilles, 23, 26 Azorean, 72 Barro Colorado, 89, 90, 91 Britain and Ireland and, 126 Canary Islands, 129, 130–131 Cape Verde, 144 cave/islands, 152–153 Channel Islands (California), 158 Comoros, 179 Cook Islands, 194 dispersal, 158 dwarfism and, 236 extinctions, 845 Farallon Islands, 295 Gulf of Guinea, 809 island rule and, 492 Krakatau and, 519 Lord Howe Island, 569, 570 mangroves and, 593 Mascarene Islands, 615, 617 New Guinea, 655 New Zealand, 670–671 overviews, 588 Pacific region, 776 Palau, 716 Philippines, 726 Rarotonga/Mangaia, 709 Samoa, 800, 801 Seychelles, 833 Socotra, 849 Solomons, 852 Taiwan, 899, 900 Tonga, 920 Vanuatu, 940 Wizard Island, 981 Zanzibar, 983 bauxite, 149 Bay of Bengal, 420 beaches, 91–94. See also dunes Bermuda, 97 Britain and Ireland and, 126 Canary Islands, 127 Cape Verde, 146, 147 Channel Islands (California), 154 Cook Islands, 194 Cook Islands forests, 193 geology, 642 Great Barrier Reef, 385 Hawaiian Island, 400 Midway Island, 632 seawalls and, 261 tides and, 914, 915

INDEX

1023

beachrock, 86 Beagle, HMS, 702, 918, 954–961 beaks character displacement and, 106 divergence and, 108 finches, 352, 355–356 isolation and, 108 kiwi, 670 Madagascar, 579 New Zealand, 667 reproductive isolation and, 110–111 Samoan bird, 800 bears, 981 Baffin Island, 78 Britain and Ireland and, 119, 120 island rule and, 492 Kurile Islands brown, 525 polar, 54 Beaufort Island, 17 beavers, 471, 793, 918 Britain and Ireland and, 120 Beebe, William, 97 Beeches Bay formation, 163 beech trees, 472, 670, 905 New Zealand, 672 bees, 3, 23, 53, 472, 667, 672, 838, 929, 981 introduced species, 472 New Zealand, 672 beetles, 462, 902 ants and, 39 arctic, 51, 52, 53 Azorean, 72, 73 Borneo, 114, 964 Britain and Ireland and, 122–123 Canary Islands, 128–129, 130, 774 Cape Verde, 145, 146 Comoros, 179 Cratopus, 2, 615 endangered species, 285 Fijian, 300 flightless, 318 Galápagos, 361 Hawaiian, 772 Miocalles, 3, 335 New Guinea, 656 New Zealand, 672 Polynesian, 772 Rynchogonus, 3 scarabs, 7, 460, 899, 902 single-island radiations, 462 St. Helena, 871 Tasmania, 907 taxon cycles, 253 Wallacea, 452 behavior adaptive radiation and, 5 Barro Colorado research, 88–89 dispersal and, 225 diversification and, 462 diversity of, 166 ecology and, 412–413 extinction and, 286 Hawaiian Drosophila and, 233 learned, 109 Belau, 148

1024

INDEX

Beliliou Island, 149, 738, 739 bellflowers, 944 Bellinghausen, 337 Bellona Platform, 695 Bellwood, Peter, 45 Benioff zone, 30, 32 Bequ Island, 308 Berau, marine lakes and, 605 Beringian islands, 48, 50, 51 Bering Sea fishes, 874 seabirds and, 812 Berkner Island, 17 Bermuda, 95–98. See also oceanic islands ants and, 40 shipwrecks, 834 Bermuda Islands, 65 Bermuda Rise, 65 bet-hedging strategies dispersal and, 226 bettongs, 906 Big Ben volcano, 440–441 Big Island, Hawaii crickets, 211 Big South Cape Island, 673, 674, 675, 793, 794, 795 Bikini (Pikini) atoll, 69, 610, 611, 680–681, 683, 684, 712 Billefjorden, Svalbard, 58 bills. See beaks biocontrol. See biological control biodiversity biogeography and, 107 fragmentation and, 329 influences on, 29 kı-puka and, 512 overviews, 888, 889–890 threats to, 894 biodiversity hotspots Borneo, 112, 113 coral, 566 Florida snails, 542 French Polynesia, 335 Gulf of Guinea, 808, 809 Indian region, 2 Japan, 500 Lord Howe Island, 568 Madagascar, 577, 578 Mascarene Islands, 612 New Caledonia, 643 New Guinea, 652 overviews, 890, 892 Pacific region, 538 seamounts, 820 Socotra, 848 Sri Lanka, 868 Sundaland/Wallacea, 447, 449, 450, 452 Taiwan, 902 Vanuatu, 939, 940 biogeography, 486–490. See also Equilibrium Theory of Island Biogeography Antarctic, 12–13 Antilles, 27, 28 Baja California Islands, 82 Barro Colorado and, 91

cladistic, 970 Darwin and, 220–221 distribution of species and, 947–948 diversity and, 107 endemism and, 254 fragmentation and, 328–329, 330 French Polynesia, 332, 336 geology and, 968–969, 970 Greek Islands, 389 overviews, 329–330, 527, 943 pre-European impacts and, 417 snails and, 538 Socotra, 849–850 vicariance and, 949 Wallace and, 962 biogeomorphic agents, 197 bioherms, 565 Bioko Island, 65, 808 spiders, 862, 862 biological control, 99–102 Bermuda, 98 freshwater habitat and, 346–347 Hawaiian insects, 479 invasive species, 478 New Zealand, 672 plant disease and, 748 The Biology of the Amphibia (Noble), 348 biota, Greek Islands, 389 biotic environment. See communities biotic stressors, 787 biotic variables, freshwater species and, 343 bird fossils, 318–326 birds, 3, 189, 236. See also bird fossils; flightlessness; landbirds; migrants; seabirds; songbirds; specific birds adaptive radiations, 105–111 Antarctic, 12 Antilles, 21, 23, 26 ants and, 39, 40 arctic, 50, 51, 52, 53–54 Ascension, 62 Azorean, 72 Baffin Island, 77, 78 Baja California Islands, 78, 79, 80, 81 Barro Colorado, 90, 91 Barro Colorado biogeography and, 91 Bermuda, 96, 98, 98 biogeography and, 841 Borneo, 114 breeding colonies/Fernando de Noronha archipelago, 298 Britain and Ireland and, 118, 120, 122, 123–124 Canary Islands, 131, 132 Canary Islands flightless, 131 Cape Verde, 144, 146, 147 Channel Islands (California), 159 Comoros, 178, 179, 180 convergence of, 189 Cook Islands invasives, 195 deforestation and, 222 disease, 103–105 dispersal and, 227, 539 dwarfism and, 238 dwarfs, 237

endangered Javan, 451 endemism and, 256 extinctions, 228–231, 282, 283, 892, 920 (see also flightlessness) Faroe (Faeroe) Islands, 288–289 food and, 843 fragmentation and, 186 French Polynesia, 335 freshwater species and, 345 Galápagos, 368 (see also Galápagos finches) gigantism and, 373–374 Gould on, 220 Guam, 845 Gulf of Guinea, 809, 810 Hawaiian, 480, 786, 977 Hawaiian extinctions, 283 humans and, 44 Indonesia, 449, 450 introduced species and, 470, 471, 472, 473 island rule and, 493, 494 Japanese, 499 Japanese extinctions, 500 Krakatau and, 519 Kurile Islands, 525 Lord Howe Island, 569 mammals and, 321–326 Marianas, 594 Mascarene Islands, 615 Mauritius, 229 maximal radiation, 773 New Caledonia, 644 New Guinea, 653, 654–655 New Zealand, 99–100, 183, 638–641, 667, 669, 670, 671, 672, 673 oases, 687, 688 Pantepui, 718–719 Philippines, 724, 729 pigs and goats and, 742 Pitcairn, 745 Pitcairn extinctions, 747 radiations, 105–111 Rapanui climate change and, 245 Rottnest Island, 798 Samoa, 713, 800, 801 Seychelles, 830, 833 Seychelles extinctions, 832 sky islands, 842 Solomons, 852 species-area relationship and, 857 Spitsbergen, 866 St. Helena, 873 St. Paul’s Rocks, 956 Surtsey Island, 885, 886, 887 sustainability and, 891 Taiwan, 899, 900 Tasmania, 907, 908 taxon cycles, 253 Tonga, 920 Vanuatu, 940 Wallacea, 452 Weber’s Line and, 970 Wizard Island, 981 Zanzibar, 983 birds of paradise, 6, 303, 453, 655, 656, 657, 826 New Guinea, 655, 655, 656, 657

sexual selection and, 826 Biscoe Islands, 17 Bismarck Archipelago frogs, 349 geolgoy, 664 human colonization, 720, 721 humans and, 42, 43 lizards, 560 pigs, 741 bison, 51 bitterns, 122 Bjørnoøya, 59 Black and White Islands, 17 blackbirds, 122 blackbutt forest, 331 black corals, 198 Black Hills, 152 Black Sea, 82 Blake plateau, 565 Blanca formation, 162–163 Blanco Transform Fault, 425 bleaching (coral), 202, 336, 378, 379 human impacts, 782 Seychelles, 832 blocks, crustal, 457 blue tits, 122 bluffs, and insects, 318 boars, Britain and Ireland and, 120 boats. See watercraft Boatswain Bird Island, 62 Boavista Island, 143, 144, 145, 146, 147 body size. See also dwarfism; gigantism; island rule evolution and, 922 extinction and, 285, 788 food and, 640 Galápagos, 361 Komodo dragons, 514 lemurs, 550 spiders, 864 sticklebacks, 874, 877 body temperatures, 922 bogs, 836, 930 alien trees and, 945 Hawaiian, 401 Hawaiian trees and, 944 Holocene and, 433 succesion and, 946 Vancouver, 938 Bonaparte, Napoleon, 871, 931 Bonin, 277, 307, 481–486, 499, 501, 506, 522, 545, 702 Bonnetia, 718 boobies, 197, 556, 812 Great Barrier Reef, 383, 384 human impacts, 814 Lord Howe Island, 570 overviews, 811 St. Paul’s Rock, 956 Bora Bora, 333 snails, 540 volcanism and, 340 boreal forests, 185 fragmented, 185 Borneo, 36, 111–116

climate, 971 climate change and, 171 frogs, 348, 349, 350 geology, 448, 457, 458–459 Indonesia, 451–452 Indonesian, 451–452 macaques, 778 mammals, 553 New Guinea and, 657–658 overviews, 447 snakes, 845 Bornhardt, Wilhelm, 466 bottlenecks bird disease and, 103 birds and, 107 Galápagos, 372 invasive biology and, 477 refugia and, 786 sexual selection and, 827–828 Bougainville, 485, 706, 854, 855, 856 Boulenger, G.A., 348 boundaries, biogeographic, 967–971. See also Wallace’s Line Bounty, HMS, 11, 16, 712, 745, 833 Bouvet Island, 67 Bouvetøya, 11 bowerbirds, 655 Bowie, Raurie, 842 Bowman, Bob, 361 boxwood, 872, 874 Brabant Island, 17, 19 brachiopods, vent, 426 bracken, 124 Bransfield Strait, 19 Brava Island, 65, 143, 144 Brazil, 963 caves, 151 inselbergs, 466, 467, 469 introduced species from, 101, 195 sustainability, 888 tepuis, 718, 839, 840 Brazilian islands, 62, 65–66, 297, 298, 560, 956 Brecquou, 154 breeding islands, 320 whales and, 975 breezes, 173 Bridgeman Island, 19 brids, Socotra, 849 Britain and Ireland, 116–126 age of, 428 birds, 120 Channel Islands (British), 154–155 fishes, 874 fishing, 310 glaciations and, 785 introduced species and, 472–473 missionaries and, 634, 637 overviews, 183 Pitcairn and, 746 rat eradication, 474 seabirds, 814 British Columbia, Canada, 4 British people, and Borneo, 115 Brocchinia, 719

INDEX

1025

Broken Ridge, 439–440 bromeliads, 152 Bronze Age, Mediterranean and, 741 Brook, B.W., 282 Brown, James H., 185, 840–841 brown noddies, 812 brown rats. See Norway rats Brunei, 112 frogs, 349 overviews, 115 Brunet, A.K., 152 bryophytes, 12, 13, 14 arctic, 51, 52, 54 Atlantic, 72 Azorean, 73, 74 Britain and Ireland and, 124 Cape Verde, 146 Japan, 497 Macquarie, 574 Madeira, 583 succession and, 877–878 temperatures and, 48 Buck, Peter, 760 buckwheat, 130, 157 buffalos, 724, 726 Bugis people, 115 bugs arctic, 53 biocontrol and, 101 Hawaiian, 462 St. Helena, 872 Tasmania, 907 Buka, 720, 854 bulls, Baffin Island, 78 Bunger Hills, Antarctica, 10 buntings, 108, 930–931, 932 buoyancy, 258, 425 island arcs and, 481 mantle plumes, of, 440 plate boundaries and, 753 Burton, Richard, 984 bustards, 122 Bute, 125 butterflies, 122 Antilles, 23, 28 arctic, 53 Australia, 970 Barro Colorado, 91 biocontrol and, 102 Britain and Ireland and, 118, 119, 122, 126 Channel Islands (California), 159 introduced species and, 473 New Caledonia, 644 New Guinea, 656 New Zealand, 667 novel adaptive zones and, 2 Solomons, 852 Surtsey Island, 886 Taiwan, 899, 900 Trinidad and Tobago, 928 Wallacea, 452 Weber’s Line and, 970 Zanzibar, 983 buttongrass, 905

1026

INDEX

cabbages, 125 Caccone, Adalgisa, 923, 923–924 cacti, 81 Galápagos, 359, 361, 362 Cadomian Orogeny, 154 caecilians, 831 cahows, 96, 97 Caicos Islands, 22 Cairn Peak, Antarctica, 10 Caithness, 292 calcareous animals and plants, 70 calcium carbonate coral reefs and, 202 overviews, 203 calcium, coral and, 199 calderas, 952, 953. See also specific volcanoes Kurile Islands, 523–524 Caledonia orogeny, 55, 56, 57 California, avian diseases, 104 California Channel Islands, 5 California Continental Borderland, 162 California tarweeds, 4–5 Callitris forests, 331 Cambrian-Ordovician boundary, 649 Cameroon, 839 inselbergs, 467 Cameroon line, 65, 840 Campbell, I.H., 485 Campbell Island, 11, 16 eradications, 673 geology, 675, 677, 679 gigantism, 671 rat eradication, 474, 795 Canada coral, 565 fishes, 874 Greenland and, 50 ponds, 526 rodents and, 793–794 Canadian Arctic Archipelago (CAA), 47, 48, 50–51 diversity of, 49 human impacts, 54 sedimentation and, 57–58 Canadian Shield, 77, 651 Canary Islands adaptive radiations, 4, 774 anagenesis, 9 Beagle and, 956 biology, 72, 127–133 birds, 104, 106, 132 Cape Verde and, 146 coral, 567 Drosophila, 234 endemic species, 73 flora, 146 geology, 63, 64–65, 133–142, 691 invertebrates, 462, 774 kı-puka and, 513 landslides, 536 lizards, 557, 561, 563 mammals and, 588 pigs and goats and, 742 rats, 794, 795

snails, 538 species-area relationship and, 857 spiders, 862, 863, 864 volcanism, 220 whales, 978 de Candolle, Augustin Pyramus, 254 canoe plants, 272, 275, 276 canoes, 758–759, 760 Canterbury Plains, 676 Canton Island, 173–174 Cape Horn, 918 capercaillie, 118, 122 Cape Romain, South Carolina, 84 Cape Verde archipelago, 64, 65, 127, 143–147 bird divergence, 108–109 Darwin and, 956 Drosophila, 234 gigantism, 373 landslides, 536 spiders, 862 whales, 975, 976 Capian Sea, 83 Capricorn seamount, 692 carbon Antilles, 32 coral and, 198, 199 diversity and, 13 carbon dioxide, 70 coral reefs and, 202 Caribbean, 36 adaptive radiation of lizards, 531 adaptive radiations, 3 Anolis lizards, 775 ants and, 37, 40 Bermuda and, 96 climate change and, 779 conservation, 784 coral reefs, 200, 378–379, 780, 781 crickets, 209, 210 Darwin’s finches and, 359 Drosophila, 233–234 dwarfism and, 236 ecological release and, 253 ethnobotony and, 271, 273 extinctions, 170 flightless birds, 313, 316 geckos, 559 introduced species, 471 invasive species, 347 lizards, 5, 7, 561–562, 561–563, 773 popular culture and, 764 pre-Columbian seafaring, 278–279 primates, 778 rats and, 795 reefs, 98, 779 research stations, 789 shipwrecks, 834 snails, 538 snakes, 845 stereotypes, 761 vicariance, 950 Caribbean plate, 950 caribou, 51, 78 Carlquist, Sherwin, 78, 836

Carmen Island, 80, 81 Carnegie ridge, 370, 951 carnivores extinct dwarfs, 237 gigantism and, 375 introduced species, 470 island rule and, 494 Madagascar, 588, 776 New Guinea, 655 carnivorous plants, 830 Caroline Islands, 148–149 dispersal and, 594 geology, 709 Caroline Ridge, 149 carp, 908 Carpenter, C. R., 88 Carr, G. D., 837 Carson, Hampton, 233, 327, 826 Carteret, 745 Carthaginians, 845 cascades of extinction, 284 Cascadia subduction system, 708 Caspian Sea, 82 cassowaries, 639, 655 casuarinas, 98, 633 Catalina schist, 162 catastrophes, 630, 957 caterpillars, 2, 15, 398, 472, 773, 794 Cathaysia, 504, 505 cats Antilles, 23, 24, 26 Ascension, 61, 62 Baja California Islands, 79 biocontrol and, 102 Britain and Ireland and, 120, 125 Canary Islands, 132 Channel Islands (California), 159, 161 extinction and, 283 Southern Ocean island, 15 Tonga, 920 cattle Britain and Ireland and, 124, 126 Hawaiian crickets and, 211 Caucasus, 121 causeways, 261 cave art, 119 caves, 150–153. See also karst features Antilles, 24 Azorean, 72, 74 Bermuda, 95, 96 Canary Islands, 130–131 Cook Islands, 192 crickets, 207, 210 freshwater species and, 344 insect convergence and, 463 islands, as, 181 lava tube, 545, 546–548 Mascarene Islands, 613 matrix-derived taxa, 182 Mediterranean, 628 overviews, 182 refugia, as, 786 Sarawak, 114 Seychelles geckos, 831

Socotra, 847, 849 spiders, 863, 865 Caygill, David, 301 Cayman Islands, 21, 23, 25, 27, 28, 235, 278, 374, 781, 845 cays (keys, quays). See also islets Antilles, 21, 22 Cook Islands, 193 Great Barrier Reef, 383, 384–385, 385–386, 387, 388 islets and, 200 motu vs., 641 sand, 93 Tonga, 918 cedars, 95, 97, 98, 899 centipedes, 385, 390, 399, 671, 907 Central America, 23, 24, 769 Central Indian Ridge, 424 central Pacific, 693, 949, 950, 977 cephalopods, 145 Cerralvo Island, 81, 82 cetaceans, 745, 866 Chadriiformes, 811, 813 chaffinches, 106, 129 Chagos Archipelago, 442, 613 Challenger, 702 Challenger plateau, 568 chameleons, 178–179, 236, 559, 560, 831 Chamorros people, 634, 636 chance (stochastic) events dispersals and, 35, 36, 775, 863, 877, 879, 947, 948 extinctions and, 187, 284, 285, 329, 478, 527, 630, 787 founder effects, 6, 326, 327, 437, 751 immigrations, 107, 206, 386 invasive species and, 478 Komodo dragons and, 514 New Zealand geology and, 679 Polynesian dispersal and, 761 species-area relationship and, 859 succession and, 879 Channel Islands (British Isles), 154–155 Channel Islands (California) adaptive radiation, 5 biology, 155–161 dwarfs, 236, 237 formation of, 492 geology, 154–155, 161–164, 707 introduced species, 472 chaparral shrubs, 78 Channel Islands (California), 157 character displacement, 106, 108 charcoal, 33, 44, 132, 221, 291, 416, 984 Charriére, Henri, 769 Chateau d’If, 771 Chatham, 937 Chatham Islands, 322, 324, 324, 461, 666, 667, 674, 675, 679, 707 whaling and, 977 Chekhov, Anton, 770 chemosynthetic biological communities (CBC), 174, 175, 176 cherry trees, 157

Chiapas, Mexico, 328 chickens, 43, 103, 104, 180, 223, 246, 337, 414, 415, 595, 721, 723, 760, 889 Chile, 957 geology, 482 tsunamis, 934 Chiloé, 423, 957 China biocontrol and, 100 Borneo and, 115 climates, 423 earthquakes, 242 eruptions and, 266, 269 geology, 504, 505 patches, 466 seafarers, 277 tarsiers, 552 watercraft, 277 Zanzibar and, 984 Chincha Island, 738, 740 Chinese, 832 Chinese sweet gum tree, 899 Chios, 395 chitons, 910 Christianity, 293. See also missionaries Christman, M. C., 151 Christmas Atoll, nuclear tests and, 684 Christmas Island, 738, 738, 739 geology, 439, 711 invasive ants, 39, 40, 472, 479, 708 isolation of, 702 lagoons, 554 land crabs, 235, 532–535 parallel evolution, 235 chuckawallas, 82 Chukchi Plateau, 55 Chumash people, 158, 160, 707 Polynesians and, 723 Chuuk, FSM, 94, 149 cicadas, 101, 300, 465, 465, 499, 570, 667, 672, 774, 872 971 Fijian, 300 cichlid fish, 165–169 Africa, 3, 6, 7, 165–169 ecomorphs, 7 isolation and, 531 sexual selection, 6, 826, 827, 828 size of lakes and, 531 cichlids, adaptive radiation, 2, 3, 166, 463 Cierva Point, Antarctica, 14 Cikobia, 693 civets, 179, 850, 902 Civil War, 770 cladogenesis, 8, 9 clams, 174, 176, 194, 424, 426, 557, 701, 853, 892, 893, 957, 959, 974 Clarence Islands group, 17, 18 Clarión Fracture Zone, 80 Clarión Island, 80 clastic rocks, Philippines, 736 clay, 149 Clermontia, 2, 3, 4, 399 cliffs, 261 erosion and, 262

INDEX

1027

cliffs (continued) Faroe (Faeroe) Islands, 288 French Polynesia, 334 geology, 755 Hawaiian Island, 400–401 Pitcairn, 744 Rottnest Island, 797 seabirds and, 813 climate, 171–174. See also rain; temperatures; weather Antarctic plants and, 13 barrier islands and, 85–86 dispersal and, 210 endemism and, 257 extinction and, 282 flightlessness and, 312 hydrology and, 421–422, 423 invasive biology and, 477 perhumid, 112, 113 climate change, 169–170. See also glacialinterglacial cycles; global warming; ice ages; sea levels Antarctic, 15 Antilles, 23, 26 archaeology and, 44 arctic, 54, 58 atmosphere and oceans, 174 Baffin Island and, 76 beaches and, 263 Britain and Ireland and, 119, 120 Channel Islands (British) and, 154 coral and, 565 defined, 172 erosion and, 261 extinction and, 282–283 Faroe (Faeroe) Islands and, 292 fragmentation and, 182 French Polynesia, 332 Galápagos and, 365–366 Gough Island, 930 Greek Islands and, 389 lakes and, 531 Mediterranean and, 628 Mesoamerican forests and, 186 montane remnants and, 184–185 motu and, 642 neotropics and, 786 rafting and, 776 Rapanui vegetation and, 245 reefs and, 779 refugia and, 785 reproductive strategies and, 699 research and, 41 Samoa and, 801 seabirds and, 814 seabirds research, 295 seamounts and, 825 Seychelles and, 833 sky islands and, 842 Spitsbergen and, 865 Sri Lanka and, 870 Tristan da Cunha, 930 vegetation change and, 185 Wallace on, 966 climate effects, volcanos and, 270

1028

INDEX

climate fluctuations, Iceland and, 435 climax vegetation, Tasmania, 905 clinkers, 543 cloth, 274, 745 cloud forests Ascension, 61, 62 climate change and, 171 Fijian, 299 French Polynesia, 334, 335, 337, 338 Lord Howe Island, 570, 571 Mascarene Islands, 614, 616 Mesoamerican, 182 overviews, 184, 312 Philippines, 589 Polynesian, 194 relictual taxa, 182 Samoa, 800 Solomons, 851 Sri Lanka, 867–868 Taiwan, 900, 902 Vanuatu, 940 clouds, 172–173 navigation and, 280 clownfish, 853 CO2 (carbon dioxide), 376, 379, 548, 625 coral reefs and, 780, 782 forests and, 942 geology and, 485 coal, 58, 866 Borneo, 112 Caroline Islands, 149 Greek Islands, 393 Philippines, 737 coastal communities Surtsey Island, 887 Trinidad and Tobago, 926 Vanuatu, 940 coastal plains, Sri Lanka, 867 coastal processes arctic, 61 cichlids, in, 166 coral reefs and, 781 erosion, 261–263 Krakatau, 519 overviews, 695 tides and, 915–916 coastal shrub, 331 coastal vegetation, 335 Madeira, 582 coastlines Britain and Ireland and, 124, 125 climate change and, 171 Cook Islands, 193 Mediterranean, 215 coasts. See also beaches; shorelines Bermuda, 97 Canary Islands, 128–129, 132 Channel Islands (California), 155, 157 Cook-Australs, 709 French Polynesian, 336 Hawaiian Island, 400–401 Macquarie, 574 Madeira, 583 Pacific Islands, 94 Singapore, 263

Socotra, 848 tsunamis and, 933–934 coatis, 90 Cobb-Eikelberg seamount trail, 713 Cockatoo Island, 768 cockatoos, 907 cockroaches, 97, 667 coconuts, 68, 273 crickets and, 208 Seychelles, 830 Cocos Island bird radiation, 106 finches, 352 spiders, 864 Cocos-Keeling Islands, 439, 958 Cocos plate, 357, 753 Galápagos, 367–368, 370 Cocos Ridge, 370, 951 cod, 976 coevolution Barro Colorado, on, 88 dispersal and, 226 Drosphila, 235 New Zealand birds, 325 rabbits and virus, 121–122 cohort stand model, 945–946 Coiba Island, 769 cold seeps, 174–177 cold temperate zones northern hemisphere, 14 Coleoptera, 113 Cape Verde, 145 Coley, Phyllis, 90 collapses, 954 overviews, 696 sea levels and, 695 collapses, volcanoes Kurile Islands, 524 collective evolution, 6 colonialism, Borneo and, 115 colonization. See also community assembly; founder effects and events; propagules; rafting adaptive radiation and, 5 ants and, 41 Azores and, 72, 73–74 Bermuda, 96 biogeography and, 487–488 bird, 81, 107 crickets, by, 206 dispersal and, 227 diversity and, 617 endemism and, 254 Faroe (Faeroe) Islands, 292–293 Galápagos, 357 Galápagos age and, 372 Gulf of Guinea, of, 809, 810 Hawaiian Islands, 398 human, 42 human of Pacific Islands, 44–47 insects, by, 460–461 lizard, 560–561 Madagascar, 580 Marianas and, 594 Mascarene Islands, 613

New Zealand, 669 oases, 688 orchids, 696–697 radiation zone and, 772 coloration frog, 842 insect, 966 sexual selection and, 826 Colquhoun, D. J., 87 Columbian Exchange, 280 Columbus, Christopher, 139, 280 comics, 765 Commander Islands, 52 commercial activity, Cook Islands, 195 communities (biotic environment) convergence and, 188–189, 191 coral, 199–120 dwarfism and, 236 community assembly, 6–7 Comoros Archipelago, 107, 177–180, 444–445 frogs, 348 Comoros, lizards, 560 competition adaptive radiation and, 6 ants and, 38 archaeology and, 47 Baja California Islands, 82 body size and, 514 Channel Islands (California), 160 character displacement and, 106–107 co-evolution and, 189 coral reefs and, 781–782 cricket morphology and, 211 dispersal and, 225, 226, 227, 228 divergence and, 108 diversity and, 168, 550 dwarfism and, 236, 238 ecological release and, 251, 252 extinction and, 282, 283–284, 943 gigantism and, 376 groups and, 88 Hawaiian Drosophila and, 233 insect speciation and, 461 introduced species and, 472 invasive species and, 346, 479 island rule and, 494, 495 macaques and, 778 metacommunity paradigms and, 528 place and, 911 spiders, 864 succession and, 877, 946 taxon cycles and, 912, 913 competitive exclusion, 89 condition-dependent dispersal, 225 “conies,” 23, 24 coniferous trees, 644, 937, 971 Borneo, 114 Kurile Islands, 525 Taiwan, 900 Wizard Island, 981 conservation, 28–29. See also biological control; eradication programs; reserves, nature African cichlids and, 169 Antarctic, 16 Antilles, 27

arctic, 54 Ascension and, 61, 63 Azores and, 71, 74 beach erosion and, 261, 262–263 Bermuda, 97–98 biogeography and, 489–490 Borneo, 115 Britain and Ireland and, 118, 119, 124, 125 Canary Islands, 132 Cape Verde, 146–147 Channel Islands (California), 158, 161 Christmas Island crabs, 534 Cook Islands, 196–197 coral, 567 Cozumel, 203 crabs, 534 Cuba, 22 Darwin on, 959 Darwin’s finches, 356 Farallon Islands sharks, 296 Fijian, 305 forests, of, 224 fragmentation and, 187 Frazer Island, 332 French Polynesia, 332, 337–338 frogs, 351 Galápagos, 364–365 Greek Islands, 390, 391 Hawaiian, 403–404, 413–414 inbreeding management, 437 Indonesia, 452 introduced species and, 474 Japan, 500 Juan Fernandez islands, 508–509 Komodo dragons, 514–515 kı-puka and, 513 Krakatau, 519 Line Islands, 557 Lord Howe Island, 568, 571–572 Madagascar, 581–582 Madeira, 585 mammal radiations and, 590–591 Marianas, 596–597 Mascarene Islands, 617–618 metapopulations and, 629, 630–631 Midway Island, 632 New Caledonia, 645 New Guinea, 658 New Zealand, 673 organizations and initiatives, 895 overfishing and, 310, 311 Pacific Islands, 102 Palau, 716 Pantepui, 719 pest control and, 99 Philippines, 730–731 Pitcairn and, 747 reefs, 779–785 relaxation and, 788 research and, 41 seabird, 295–296, 815 sexual selection and, 828 Seychelles, 832 snails, 540–542 solenodons, of, 25

spiders, 865 Sri Lanka, 869–870 St. Helena, 872–873 Taiwan, 901–902 Tasmania and, 908 Tonga, 920 tortoises, 926 Trinidad and Tobago, 928–929 Vancouver, 938 Zanzibar, 985, 986 Zanzibar and Pemba, 982, 983 continental biota, 168 crickets, 207, 210 Hawaii and, 234 seamounts and, 820 continental collision, 55–56 continental islands, 180–185. See also granitic islands; oases biogeography and, 948 biota, 271 freshwater species and, 344, 345 geology, 707, 755 global warming and, 376, 377 Great Barrier Reef, 383, 386 lizards, 559 New Zealand, 679 oases, 688 overviews, 490, 665, 707–708 South America, 956 spiders, 863, 864 succession and, 877 Taiwan, 897 Trinidad and Tobago, 926–927 Wallace on, 965–966 continental lithosphere, 754 continental shelf islands, 9, 180–187 arctic, 76 barrier islands, 87, 93 convergence, 191 endemism and, 254 Sunda shelf, 112 Vancouver, 937 continental shelves, 289 continents, 690 climatic effects, 172 collisions with, 902 ephemeral islands and, 260 geology, 481 island arcs and, 481, 485 continuous differentiation, 948 Convention on Biological Diversity (CBD), 888 convergence, evolutionary, 188–191, 774, 876 caves and, 151 continental islands and, 186 flightlessness and, 315 gigantism and, 375 Hispaniola/Mauritius, 109 honeycreepers and, 412, 413 insects and, 463 Macaronesia and, 774 mammals and, 588, 590 overviews, 1, 7 succession and, 877, 878 Tasmania and, 906

INDEX

1029

convergence, plate. See plate convergence convergent-plate boundaries, 691 Cook, James, 42, 195 Christmas Island discovery, 711 Frazer Island and, 332 Vancouver, G. and, 937 Hawaii and, 403, 759 New Zealand and, 672 overviews, 280, 702 pigs and, 741 pigs and goats and, 742 Polynesian voyaging and, 758 popular culture and, 763 Rapanui and, 250 Tahitians and, 222 Cook-Austral chain, 953 Cook Islands, 191–197 freshwater species, 345 geology, 342, 343 human colonization, 44, 759 missionaries and, 636–637 motu, 641 overviews, 709 coots, 374 copepod fauna, 151, 152, 153, 153 Cope’s Rule, 239 copra, 68 coral, 197–203. See also coral islands; coral reefs arctic, 52 Atlantic, 65 atolls and, 69 Bermuda, 96, 98 boulders, 554, 555 Cape Verde, 145, 147 Caroline Islands, 149 Darwin on, 217 dispersal, 779 French Polynesian, 336, 337 global warming and, 378, 379–380 human impacts, 821 hydrology and, 421 Lesser Antilles, 22 Mascarene Islands, 614 overfishing and, 310 overviews, 565 Pacific region, 708 Rapanui, 247 Rottnest Island, 798 sea level and, 817 seamounts and, 819, 820 shipwrecks and, 834 Solomons, 852–853 sustainability and, 891 Tatoosh, 909, 910 Vancouver, 937 volcanism and, 956 waves and, 880 coral islands, 641–642, 712, 799, 830, 833 coralline algal microatolls, 205 coral reefs. See also atolls; fringing reefs; reefs atolls, 68–69 biology, 780–781 Caribbean, 200–201 conservation, 895

1030

INDEX

Cook-Australs, 709 Cook Islands, 192, 193, 195–196, 196–197 Darwin on, 219, 958, 959 ecology and conservation, 779–785 French Polynesia, 332 geology, 692, 754 global warming and, 70, 377, 378–379 Hawaiian Islands, 400, 408 Japan, 500 landslides, 536 Line Islands, 555, 556, 557, 711 Lord Howe Island, 571 Maldives, 587 mangroves and, 591 Mascarene Islands, 614–615, 620 Midway Island, 631 motu, 641–642 northernmost, 95 overviews, 200–202, 566, 694 rafting and, 775 Rottnest Island, 797 Samoa, 801 seamounts and, 823, 824 Seychelles, 831 Solomons, 853 Taiwan, 897, 898 temperature and, 68 tides and, 916 Tonga, 920 waves and, 882 western Pacific, 715, 716 west Pacific, 715 Zanzibar, 984 Coral Sea Basin, 664 cordage plants, 273, 274 core, Earth’s, 752 corellas, 908 Corliss, J., 424 cormorants, 140, 159, 294, 295, 296, 313, 359, 374, 525, 811, 812, 912 938 corn, 184 Corsica, 624, 629 Corvo Island, 64, 71 cotton, 357, 359 cottontails, 79, 81 Coulman Island, 17 The Count of Monte Cristo (Dumas), 771 cowbirds, 329 Cowley, William Ambrosia, 918 Cox, G. W., 912 coyotes, 81 coypu, 121, 474, 793 Cozumel Island, 203–205, 236, 845 crabs ants and, 39, 40 Ascension, 62 Cape Verde, 145 Christmas Island, 479 Cocos Keeling Islands, 958 cold seeps and, 174 dispersal, 425 drosophilids parallel evolution, 235 Line Islands, 556 New Zealand, 669 Socotra, 849

crakes, 320, 326 cranes, 122, 313, 321 Crater Lake, 979 crayfish, 346 Crémieux, Gaston, 771 Cretaceous-Tertiary mass extinction, 612 Crete biota, 237, 390, 391, 626, 741 geography, 623 geology, 389, 392, 393, 394–395, 492, 626, 628 human colonization, 629 crickets, 206–212 Britain and Ireland and, 123 Channel Islands (California), 159 Hawaiian, 3, 6, 461, 462–463, 465, 548, 773 Lord Howe Island, 569, 571 mice and, 794 New Zealand weta, 318 sexual selection and, 826 crinoids, 819 crocodiles, 27, 304, 325, 654, 670, 716, 726, 832, 850, 853, 854, 893 Cromwell current, 366 Crosby, Alfred, 280 Crosby, Andrew, 415 crossbills, 26 cross breeding, 108 Cross Island, Alaska, 93 Crozet Archipelago, 11, 12, 16, 444, 789 Crozet plateau, 444 crustaceans. See also specific crustaceans Cape Verde, 145, 146 Cook Islands, 194 seamounts and, 819 Socotra, 849 Solomons, 852, 853 sustainability and, 891 Tasmania, 907 crustal processes. See also blocks, crustal; granitic islands; plate tectonics; subduction; specific plates arctic, 55, 58 Atlantic, 65, 66, 67 Borneo and, 113 Canary Islands, 133 Caroline Islands, 149 Cook Islands, 191–193 Cyprus and, 215 drowned islands and, 370 Fiji and, 306 French Polynesia and, 339 Galápagos, 371 Greek Islands and, 395 Hawaiian, 408 Iceland, 429–430, 434 Indian Ocean region, 445 Kurile Islands, 523 New Caledonia, 647 New Zealand and, 675 Pacific, 484 sea level and, 261 seamounts and, 714 crust, continental. See also crustal processes; specific plates

island arcs and, 482 Kerguelen Plateau and, 679 New Zealand, 676, 678 overviews, 689–690, 752, 754 seamounts and, 821 Taiwan and, 903 water and, 485 crust, oceanic. See also crustal processes; specific plates Borneo/New Guinea and, 657 French Polynesia, 340–341, 342 island arcs and, 482–483 Kuriles and, 522 Newfoundland, 650, 651 New Guinea, 664 New Zealand, 676, 678 North Atlantic and, 652 oceanic islands and, 951 overviews, 689–690, 692, 694, 752, 754 Philippines and, 733, 735 Samoa, 803 seamounts and, 821, 825 Solomons, 855 thickness of, 735 Cruz, Eliecer, 366 cryptic species, 725, 820 cryptogams, 13, 14, 467 Cuba ants, 36, 37, 38 bats, 26 biogeography, 20, 21–22, 28–29 birds, 26, 321 cave deposits, 26 colonization, 278 crickets, 210 Drosophila parallel evolution, 235 fish, 28 geckos, 775 land mammals, 25 lizard ecomorphs, 562 reptiles/amphibians, 27 rodents, 23, 24 sloths, 24 snails, 537 cuckoos, 109, 322, 450, 669, 871 Culebra Island, 22 Culver, D.C, 151 Curaçao, 474 curlews, 611, 633 currents, 61. See also rafting; specific currents colonization and, 357 coral and, 565 dispersal and, 425 fish and, 525 freshwater species and, 344 global warming and, 376, 379 Juan Fernandez islands and, 507 Line Islands and, 555 sea levels and, 817 tides and, 915, 916, 917 cuscus, 453 Cyclades, 394, 626 cyclones, 178, 299, 418, 420. See also hurricanes; typhoons birds and, 800

mangroves and, 592 Samoa and, 713 cypress, 899 Taiwan, 900 Cyprus, 212–216 geology, 627 human colonizations, 629 daisies, 157 Dale, 918 damselflies, 74, 88, 123, 300, 402, 656, 719, 786, 852 Dana, James D., 200, 704, 803 dance, 764, 958 dandelions, 50 Daniel Lincoln, 977 Daphne Major, 106–107, 108, 355 Dario, Ruben, 770 darters, 593 Darwin, Charles, 2–3, 217–221. See also Beagle, HMS Ascension and, 61, 62 atolls and, 68–69, 950 coral reefs, on, 200 endemism and, 254 Fernando de Noronha archipelago, 297 flightlessness, on, 227, 311–312 Galápagos and, 352, 710 geology and, 367, 587, 704 missionaries, on, 638 molluscs, on, 538 orchids, on, 696 sexual selection, on, 826 Tahiti, on, 334 Tasmania and, 908 Tierra del Fuego and, 918 tortoise conservation and, 833 Darwinian islands, 228 Darwinism (Wallace), 966 Darwin point, 201 Darwin’s finches (Galápagos finches), 2–3, 106–107, 108, 110–111, 352–356, 365, 461, 773, 828, 950 Dawson Island, 769 Dayak people, 112, 115 Dayton, Paul, 910 decapods, 145, 146, 345, 906, 345 Deccan Plateau, 441, 442, 443, 444, 445, 612, 620, 622 Deception Island, 19, 485 deciduous habitat Channel Islands (California), 157 Indonesia, 453 Madagascar, 578, 579 New Guinea, 657 deciduous woodland, 155 deep corals, 198–199, 820 deep-sea speciation, 755–757 deer, 901, 981 Baja California Islands, 79, 81 Britain and Ireland and, 118, 120, 121, 124, 125 Channel Islands (British Isles), 155 Channel Islands (California), 158, 160 dodo and, 230

dwarfism and, 236, 237, 238 Marianas, 596 New Caledonia, 645 Sri Lanka, 869 defense mechanism, loss of, 361 Defoe, Daniel, 508, 763 deforestation, 221–224. See also timber Borneo, of, 115 Britain and Ireland and, 117–118, 123 Channel Islands (California), 155 climate change and, 170 Comoros, 177, 180 Cook Islands, 195 Indonesia, 448, 450 introduced species and, 471 Line Islands, 557 Pacific region, 891–892 Philippines, 730 pre-European impacts, 416, 417 Samoa, 713, 800, 801 Seychelles, 832 Solomons, 854 St. Helena, 873 Taiwan, 901 Tonga, 920 deglaciation, 4 atolls and, 68 Faroe (Faeroe) Islands, 291 overviews, 694 sea levels and, 817 Deinandra, 5, 79, 157 De-Longa Island, 51 Denny, Mark, 911 de novo formation, 180–181, 183, 187, 460, 462 density, extinction and, 284–285 Deperet, Charles, 239 depth. See bathymetry Desertas Islands, 64 deserts, 686–687 development. See also sustainability Bermuda, 97, 98 biodiversity hotspots and, 890 Canary Islands, 132 Cook Islands, 196 coral reefs and, 782 erosion and, 261 hydrology and, 423 Madagascar, 581 New Guinea, 658 Pitcairn, 747 Seychelles, 832, 833 Socotra, 851 St. Helena, 874 Devil’s Island, 769 Devon, 50 diagenetic processes, 176–177 Diamond, 106 diatoms, 857 Surtsey Island, 886 dicotyledons, Fijian, 299 Diego Alvarez Island, 66 Diksam Plateau, 847 dimorphism birds, 667 dispersal and, 227

INDEX

1031

dimorphism (continued) Hawaiian Drosophila, 232 moa, 639 New Zealand, 671 Samoa, 800 dinosaurs dwarf, 237, 239 New Zealand, 322 Diomede Island, 48, 51 diploids, dispersal and, 226 Diptera, Fijian, 300 Dipterocarpaceae, 830, 867 Sri Lanka, 868 dipterocarp forest, 114 Disco, 423 discovery, 276–281. See also individual explorers Discovery, 937 diseases. See also specific diseases Channel Islands (California), 158 coral reefs and, 782 Cozumel, 205 extinction and, 282, 283 Galápagos, 356 Hawaiian bird, 414, 977 honeycreepers and, 711 introduced species and, 472, 473, 474 invasive species and, 478 Marquesas and, 712 missionaries and, 636 pigs and goats and, 742 plant, 747–752 rats and, 795 refugia and, 786 Tasmania, 905, 906 Tierra del Fuego and, 918 whaling and, 977 disharmony, 944 radiation zone and, 772 dispersal, 224–228. See also specific methods; specific organisms alien species and, 228 altruism and, 225, 226 aquatic species, 528 behavior and, 225 bet-hedging strategies and, 226 biogeography and, 970 chance events and, 35, 36, 761, 775, 863, 877, 879, 947, 948 climate, 210 and coevolution and, 226 colonization and, 227 competition and, 225, 226, 227, 228 condition-dependent, 225 conservation and, 619 currents and, 425 dimorphism and, 227 diploid, 226 distribution of species and, 947–948 diversity and, 167 dormancy and, 226 ecological succession and, 228 endemism and, 360, 452, 907 equatorial forests and, 210 Equilibrium Theory of Island Biogeography and, 947

1032

INDEX

extinction and, 225, 226, 284 faulting and, 425 flightlessness and, 460–461 fragmentation and, 330 fragmentation (patches) and, 226, 227–228 freshwater species and, 345–346 generalist species and, 225 global warming and, 380 habitat factors and, 224–225 Holocene period, 776 human impacts, 528, 529 inbreeding and, 225, 226, 227 inter-patch, 182 introduced species and, 472 invasive species, 347, 478, 742 lakes and, 527, 528–530, 531 local extinctions and, 225, 226 longevity and, 226 metacommunity paradigms and, 528 mortality rates and, 227–228 new islands and, 228 nutrient availability and, 531 ocean, 367–368 plant disease and, 749 Pliocene period, 776 pollen and, 226–227 predation and, 225, 531 Quaternary period, 776 radiation zone and, 772 random, 225 sexual difference and, 226 size of islands and, 228, 527 space, ecological and, 225, 226, 228 specialist species and, 225 species-area relationship and, 860 stages of life and, 226 storms and, 350 succession and, 877 time and, 225 vicariance vs., 560, 561, 947–948 volcanism and, 948 wings and, 227 dispersal ability ants, 37 Azores and, 72–73 biogeography and, 12 diversification and, 5 endemism and, 256 extinction and, 286, 788 fragmentation and, 282–283 freshwater species, 344 frogs, 348, 349–350 Hawaiian Islands and, 398 immigration and, 488 isolation and, 463 lakes and, 527–528, 529 loss of, 361 open space and, 479 succession and, 877 dissolution theory, 69 distance, 13, 14, 38, 209, 859 distribution of species, 181, 947–948, 969, 970, 971 disturbances, 942–943. See also fires; storms divaricating species, 671

divergence, evolutionary, 826. See also diversification; diversity bird, 105, 108–109 cricket morphology, 211 cricket songs and, 209 ecological release and, 252 sky islands and, 841–842 stochastic influences, 877 divergence, plate. See plate divergence (spreading) divergent natural selection, 4 divergent plate boundaries, 690–691 divergent selection, 166 divers (birds), Britain and Ireland and, 123 diversification. See also anagenesis; divergence; speciation; species-area relationship (SAR) adaptive radiation and, 6 ants and, 36 Channel Islands (California), 157 cichlids, of, 166 constraints on, 462 ecological release and, 252 fragmentation and, 182 insects, 461, 462 overviews, 220 Philippines, 729 plasticity and, 6 sky islands and, 841 snails and, 538 diversity, 4, 23. See also divergence adaptive radiation and, 5 ant, 37–38 Antarctic, 11–12 ants and, 40 arctic, 52 Azores and, 73 Barro Colorado, 89–90 Borneo, 113–114 Britain and Ireland and, 117–118, 122, 124, 125–126 Canary Islands, 128–129 Cape Verde, 145 Channel Islands (California), 157, 160 cichlid fish, of, 165 colonization/extinction and, 529 Comoros, 179 convergence and, 189 Cook Islands low, 193 cricket, 209 endemism and, 254, 256 French Polynesian, 335 genetic, 178, 889–890 human, 42 influences on, 29 inselbergs hotspots, 468–469 island arcs and, 705 morphological, 8 Oceanian, 42 overviews, 708, 889–890 Philippines, 724–725, 726–727 plant disease and, 751–752 Samoan, 800 temperatures and, 48–49 variations over time, 9–10

divisions (fragments), 220–221 DNA, Hawaiian birds and, 412, 413 Dodecanese Islands, 395–396, 626 dodos, 228–231, 316, 619 dogs, 43 Antilles, 26 Baja California Islands, 79 Canary Islands, 132 dolphins, 812, 853, 975 Britain and Ireland and, 120, 125 Canary Islands, 132 Cape Verde, 145 Channel Islands (California), 159 Line Islands, 557 river, 113 St. Helena, 870 Vancouver, 938 domestication. See also specific animals and plants Britain and Ireland and, 120–121 Faroe (Faeroe) Islands, 292 Mediterranean, 629 Near Oceania, 721 New Guinea, 741 Polynesians and, 723 Dominica, 29, 32, 35 lizards, 563 Dominican Republic, 22, 23 amphibians, 27 biodiversity of, 23 birds, 26, 27 lizards, 561 protection in, 29 shrews, 25 Domnica lizards, 564 dormancy, 224 dispersal and, 226 dormice, 126 Britain and Ireland and, 120 Dorset peoples, 76 doves Baja California Islands, 80 bird disease and, 105 Fijian, 303 Samoa, 800 Tonga, 920 dragonflies, 852, 907, 907 Japan, 499 New Guinea, 656 dragon’s blood trees, 848, 849, 850 Drake, Francis, 294 Dreyfus, Alfred, 769 drift, genetic adaptive radiation and, 6 bird colonization and, 107 bird disease and, 103 conservation and, 828 endemism and, 256 founder effects and, 110, 327 isolation and, 108 orchids, 696, 699 overviews, 766 sexual selection and, 826–827 sky islands and, 841 speciation and, 166

Driftless Area, 152 dripping water, 152–153 drip pools, 152 Drosophila, 232–235 adaptive radiation, 3, 6, 461, 462, 463, 616 colonization, 465, 616 island age and, 464, 773–774 isolation and, 402, 512 science and, 788 sexual selection, 826, 827 droughts freshwater species and, 345 Great Barrier Reef, 385–386 Madagascar, 579 oases and, 689 Pitcairn, 745 silversword adaptations to, 836 drowned islands, 678 Galápagos, 370–371, 372 Makatea, 585 overviews, 483 seamounts, 823 drowned reefs, 201 Ducie Atoll, 45 ducks, 671, 811 Britain and Ireland and, 122, 124 Galápagos, 359 Hawaiian, 326 introduced species and, 472 Midway Island, 633 Dumas, Alexandre, 771 dunes barrier islands and, 83, 85, 86, 87 Bermuda, 95 Britain and Ireland and, 125 Cape Verde, 144, 147 Channel Islands (California), 157 Frazer Island and, 331 isolation and, 187 Surtsey Island, 887 dung beetles, 7 Dunotter Castle, 834 D’Urville Island, 675 Dutrou-Bournier, Jean-Baptiste Onesime, 250 Dutson, Buy, 302 dwarfism, 151, 235–239. See also body size; island rule Channel Islands (California), 158, 160 Cozumel, 205 Galápagos, 361 Hawaiian, 944 island area and, 494 Juan Fernandez islands, 508 lemurs, 550 lizard, 559 mammoths, 707 Mediterranean, 628 Philippines, 724, 726 snakes, 844 succession and, 878 tarsiers, 553 tortoises, 922 Zanzibar, 983 dye plants, 274–275 Dystaenia, 8

eagles, 122, 671 Britain and Ireland and, 124, 125 Channel Islands (California), 158, 159, 161 gigantism, 374 Philippines, 724, 731 Solomons, 852 Taiwan, 899, 902 Tasmania, 907 Vancouver, 938 earthquakes (seismic activity), 240–244 Antilles, 30, 32, 34–35 arctic, 55 Azores and, 72 Chile, 711 Darwin on, 217 faults and, 481 Greek Islands and, 392, 394, 395, 396 Hawaiian, 711 Iceland, 430–431, 435 Indonesia, 454–455, 455–456 Japan, 485, 502–503 Kurile Islands, 522 Macquarie Island and, 576 Mediterranean, 625 New Guinea, 661, 662, 664–665 New Zealand, 676 overviews, 481, 483, 484, 485, 693, 703, 753, 754 Pacific region, 706, 802 Philippines, 732, 737 seamount, 713 Solomons, 706, 856 Taiwan, 903 tsunamis and, 933, 934, 935 uplift and, 694–695 Earth Summit, 888 earwigs, 872 East African arc, 840, 842 East African beetles, 7 East African islands, 778 East Asian plate, 481 East Azores Fracture Zone, 64 Easter Island (Rapanui), 244–251 archaeology, 47 bird diversity, 320 geology, 951 human colonization, 41, 720, 722 introduced species, 471 overviews, 709–710 rodents, 794 seabirds, 814 settlement of, 760 shipwrecks and, 834 Eastern Plateau, 659 East Malaya-Indochina block, 456 east Pacific, 892, 977 East Pacific Rise, 78, 164, 425 Eberhard, Mary-Jane West, 5 ebony, 873 echidnas, 906 echinoderms, 194, 379, 476, 605, 797, 819, 891 ecological factors, 8, 462, 488, 564, 729, 772. See also adaptive radiation; convergence, evolutionary; specific factors

INDEX

1033

ecological release, 5, 211, 251–253, 375, 864 ecological specialization, 233–234 ecological succession, 228 ecology and behavior, 412–413 ecomorphs, 7, 190, 191, 349, 531, 562–563, 931 economies. See also sustainability Cook Islands, 195 coral reefs and, 781 Cozumel, 205 Pitcairn, 746–747 research stations, 789–790 Seychelles, 832 Solomons, 856 sustainability and, 891 Tonga, 919 Trinidad and Tobago, 927, 928 Vancouver, 938 Vanuatu, 939 Zanzibar, 985 ecoregions, Indonesia, 448, 451 ecosystem diversity, 889–890, 893 vegetation, 942 ecoterrorism, 845 ecotherms, 922 ecotourism, 790 bird disease and, 105 Cook-Australs, 709 Cook Islands, 197 Socotra, 851 Trinidad and Tobago, 929 Zanzibar, 985 edge effects, 787 Edwards Plateau, 152 Edwards, William, 963 eels, 124, 159, 669 Cook Islands, 194 egg colors, 109, 110 egging, 294 Egg Island, Alaska, 86 egrets, 908 Egypt, 591 Ehrlich, Paul, 1 Eiao Island, 223–224, 337, 341 eiders, 125 Eigg Island, 124 Eil Malk Island, 149 elastic rebound theory, 240 Eldgja eruption, 434 elephant birds, 639 Elephant Islands group, 17, 18 elephants, 628, 922, 984 Borneo pygmy, 113 dwarf, 236, 237, 238 island rule and, 493, 494 elephant seals, Farallon Islands, 296 Elephas, 869 elevation. See also inselbergs; mountains; plateaux, mountain anagenesis and, 9 Antarctic, 11, 13–14 Antilles, 22 ants and, 38 arctic diversity and, 51 Borneo, 112, 113–114 Borneo forests and, 114

1034

INDEX

Canarian insects and, 130 Canary Islands, 128, 132 Cape Verde, 143, 146, 147 Caroline Islands, 149 climate change and, 15 Comoros, 178 deforestation and, 222 divergence and, 841 diversity and, 10, 186 endemism and, 257, 551 extinctions and, 171 finches and, 352 flightlessness and, 312, 317 fragments and, 186 freshwater species and, 344 global warming and, 377, 378 Gough Island, 929 Great Barrier Reef, 384, 386 groundwater and, 420–421 Hawaiian birds and, 414 Hawaiian colonization and, 398 hydrology and, 421 insect diversity and, 463 inselbergs and, 466 island-like systems and, 235 Kurile Islands flora, 525 Madeira and, 582, 583 Marianas and, 594 Mascarene Islands, 613 montane remnants, 184 New Caledonia, 643, 645, 648 Philippines, 727 Samoa, 799 silverswords and, 837 sky islands and, 839, 840 snails and, 541 species richness and, 698 storms and, 419–420 Taiwan, 898 timber and, 274 Tristan da Cunha, 929, 930 Vanuatu, 939 vegetation and, 578, 579 wind and, 172 elks, Britain and Ireland and, 120 Ellef Ringnes Island, 51 Ellesmere Island, 50, 51, 53, 54, 57 zooplankton, 530 Ellice Islands, 69 El Niño, 81, 82 climate change and, 174 Cook Islands reefs and, 196 Fiji and, 299 Galápagos and, 365 global warming and, 379–380 Line Islands and, 553, 554 Polynesian voyaging and, 760 Tasmania and, 905 El Niño-Southern Oscillation (ENSO), 93, 171, 173–174, 345 coral reefs and, 781 corals and, 378 Galápagos and, 355 hydrology and, 423 Line Islands and, 555

Palau marine lakes and, 605 seabirds and, 814 storms and, 419 tides and, 816 emergent islands, 692, 694, 695 Emperor seamounts, 201 emus, 639, 908 endangered and threatened species. See also conservation; introduced species; reserves, nature amphibians, 27 Antilles, 25, 26 arctic, 54 birds, 845 body size and, 285 Britain and Ireland and, 122 Cape Verde, 145, 146–147 Channel Islands (California), 161 Comoros, 179 Cook Islands birds, 194 Cozumel, 205 Fijian, 299, 302, 304, 707 finches, 356 Frazer Island, 331 French Polynesian, 332, 335, 337 frogs, 351 Galápagos, 365 Greek Islands, 390–391 Hawaiian, 282, 403, 413, 414, 711, 838 Indonesia, 450, 451, 452 introduced species and, 473 Juan Fernandez islands, 508–509 lizards, 845 Lord Howe Island, 572 New Zealand, 325, 673 overviews, 891–892 Pitcairn, 745 Puerto Rican, 472 reefs, 779 Samoa, 799, 800, 801 Solomons, 854 spiders, 865 Sri Lanka, 870 Taiwan, 902 Tasmania, 908 Vancouver, 938 vegetation, 643 Zanzibar, 986 Endemic Bird Areas of the World, 147 endemic diseases, 104 endemism, 253–258. See also biodiversity hotspots; biogeography Antilles, 20, 25, 26, 27, 28, 234 arctic, 51 Azorean, 72, 73, 74 Bermuda, 96, 98 Borneo, 114 Britain and Ireland, 118 Canadian Arctic Archipelago, 49 Canary Islands, 129, 130 Cape Verde, 145, 147 caves, 151 Channel Islands (California), 155, 157, 159, 161 climate change and, 171 Comoros, 177, 178, 179

Cook Islands, 194 Cozumel, 204, 205 extinction and, 319–320 Farallon Islands, 296 Fijian, 299, 301–304 French Polynesia, 332, 711 Greek Islands, 390 Gulf of Guinea, 809–810 Hawaii, 232, 233, 234 Hawaiian Island, 400 Indian Ocean, 107 Indonesia, 448 isolation and, 320, 335, 359, 773 Madagascar, 580 Marianas, 596, 597 Marianas (hotspot), 594 marine lakes and, 605 Marshall Islands, 712 Mascarene Islands, 612, 615–616 New Zealand, 183–184, 666, 669 oceanic islands, 808 overviews, 9, 708, 943 Pantepui, 718–719 Philippines, 724–725, 729 Rapanui, 245 Samoa, 713, 800 seamount, 820 Society Islands, 712 Southern Ocean Islands and, 13 spiders, 862 Tasmania, 907 Vanuatu, 940 vulnerability of, 35, 39 whale falls and, 974–975 Endler, 209 endotherms, 922 energetic equivalence rule, 514 energy, species richness and, 13 Enewetak (Eniwetok) atoll, 69, 680, 682, 684 coral and, 200 geology, 219 subsidence, 695 environmental factors. See also convergence, evolutionary; habitat factors adaptive radiation and, 353 Darwin’s finches, 355 lakes and, 529, 530 reproductive isolation and, 327–328 succession and, 877–878 environmental protection, 16 Antilles diversity and, 29 ants and, 40 environments, novel, 1 Eocene Jolla Vieja formation, 162 Eocene Poway conglomerate, 162 ephemeral islands, 258–260, 700 geology, 483–484 organic falls, 700–701 ephemeral vegetation, inselbergs and, 468 Epi-Jomon people, 522 epikarst, 150–151, 152, 153 episymbiosis, 656 equatorial forests, dispersal and, 210 equatorial Pacific, 708, 754 equilibrium

overviews, 947 relaxation and, 787 size of islands and, 879 species-area relationship and, 858 Surtsey Island and, 886 Equilibrium Theory of Island Biogeography conservation and, 28 continental islands and, 185 dispersal and, 947 evidence, 384, 517, 519 Krakatau and, 451, 517 overviews, 858, 943 sky islands and, 840–841 species-area relationship and, 858–859, 943 eradication programs, 618 Aleutian foxes, 815 Galápagos, 926 pigs and goats, 741, 743 refugia and, 786 rodent, 795 Tasmania, 908 Tristan da Cunha/Gough, 932 Ericaceae, 971 ericaceious plants, 579 ermine, 52 erosion. See also landslides atolls and, 68 barrier islands and, 83, 85, 87 beaches and, 92 Borneo and, 112, 113 Britain and Ireland and, 125 Canary Islands, 134, 135, 142 Cape Verde, 144 Channel Islands (California), 161 climate and, 85–86 climate change and, 642 coastal, 261–263 Comoros, 180 coral reefs and, 782 Cyprus, 216 deforestation and, 221–222 exotic species and, 223 Fernando de Noronha archipelago, 297–298 Frazer Island, 331 freshwater species and, 343, 344 granitic islands and, 381 Great Barrier Reef, 385 Greek Islands, 393 guyots and, 713 Hawaiian Islands, 397, 398 Indonesia and, 458 Japan and, 502 landslides and, 535 Lord Howe Island, 569 mangroves and, 592 Marquesas, 223–224 Mascarene Islands, 613 Mediterranean, 627–628, 629 motu and, 641–642 New Caledonia, 646 New Zealand, 675, 676 pigs and goats and, 742 Pitcairn, 744, 745, 747 Samoa, 803

seamounts and, 823 St. Helena, 871 tepuis and, 717, 718 Tristan da Cunha, 930 Uniformitarianism and, 218 volcanic islands and, 953–954 volcanoes and, 755 waves and, 880 eruptions. See also lava; lava pioneer communities; magma; volcanism; specific eruptions Greek Islands, 394, 396 Iceland, 434 magma and, 952 Marianas, 600–602 Mascarene Islands, 613 overviews, 544, 545, 755 rifting and, 952–953 seamount, 713 tsunamis and, 933 escalation/diversification hypothesis, 1–2 Española, 921 Espíritu Santo Island, 80, 81 Essex, 918 Estanque Island, 283 Ethiopian Highlands, 840 ethnobiodiversity, 890, 892, 894–895 ethnobotany, 271–276 Eucalyptus, 905, 944, 958 Eurasia birds, 122 squirrels, 119 Eurasian plate, 71, 392, 448, 482, 624, 753 Cyprus and, 215 Indonesia and, 454 Japan and, 501, 502, 506 seamounts and, 824 Taiwan and, 897, 902 Eurasian-Sundaland plate, Philippines and, 732 Euraud, Eugène, 249 Europe Azores and, 73 barrier islands and, 83 birds, 122, 126 Borneo and, 115 Britain and Ireland and, 126 Canary Islands and, 131 Channel Islands (California) and, 155 fishes, 874 glacial-interglacial cycles, 785 Greenland and, 50 Newfoundland and, 650 prisons, 771 research and, 789 tectonism and, 55, 56, 57 volcanism and, 58 Europeans. See also missionaries; individual Europeans Caroline Islands and, 148 Cook-Australs and, 709 explorers, 280 Marianas and, 595 Marshall Islands and, 611 New Zealand and, 672 overviews, 709

INDEX

1035

Europeans (continued) pigs and goats and, 741–742 Tasmania and, 907–908 Vancouver and, 938 eustatic processes, 261. See also sea levels Evans, Ben, 729 evergreen and semi-evergreen forests, 128 Canary Islands, 132 Cozumel, 205 Java, 451 Madagascar, 578, 578–579 Madeira, 582, 583 New Zealand, 671 Socotra, 848 Trinidad and Tobago, 930 everwet (evergreen) forests, 113, 114 evolution Azores and, 73–74 caves and, 151 evolutionary radiation, 253–254 evolutionary sorting, 189, 191 evolutionary trees, 412 chance and, 948 New Zealand, 640 Philippine rodents, 589 tortoises, 924 Evvoia, 393, 626 Exacum, 850 exotic species. See also alien species; introduced species; invasive species birds and, 230 Canary Islands, 132 Cape Verde, 145, 146 caves and, 151 Channel Islands (California), 156, 160 Comoros, 177, 179–180 Cook Islands, 195 Cozumel, 205 deforestation and, 222–223 eradication of, 618 ethnobotony and, 272 Farallon Islands, 295 freshwater, 343 Galápagos, 365 Lord Howe Island, 571 New Zealand, 673 Rapanui, 245–246 expeditions Caroline Islands and, 148 overviews, 790 Socotra and, 848 Vanuatu, 940 Zanzibar and, 984 exploration, 276–281. See also individual explorers; specific explorers boundaries and, 967 Farallon Islands and, 294 Juan Fernandez islands, 508 Seychelles and, 832 tortoises and, 926 Tristan da Cunha/Gough and, 931 Zanzibar and, 984 Explorer ridge, 426 explorers, 203

1036

INDEX

extinctions, 281–286. See also bird fossils; endangered and threatened species; local extinctions; metapopulations; relaxation adaptive radiation and, 2 Africa, 129 African cichlids and, 169 Antilles, 23, 24, 25 Antilles birds, 26 anuran (amphibian), 869–870 Ascension, 62 Australia, 905 Baja California Islands, 78, 79, 80 Barro Colorado, 91 Bermuda, 96, 98 biocontrol and, 101–102 biogeography and, 487, 488 birds, 103, 228–231, 314, 324, 871, 920 (see also bird fossils) Borneo and, 115 Britain and Ireland and, 118, 119, 120, 120, 122, 124 Canary Islands, 128, 131, 132 Cape Verde, 146 caves and, 151, 152, 153 Channel Islands (California), 158, 160 climate change and, 170, 171 continental islands and, 183, 254 Cozumel, 205 dwarfism and, 236–237 Fijian, 302, 304 flowering plants, 871 forests, 947 fragmentation and, 187, 329, 330 fragments and, 185 French Polynesia, 332, 335, 337, 711 Galápagos, 364–365 Galápagos finches, 773 Greek Islands, 389–390, 391 Hawaiian, 403, 404, 413–414, 711 Hawaiian birds, 410, 411 honeycreeper, 773 human-caused, 24, 25, 44, 107 inbreeding and, 437 Indonesia, 450, 451 Indonesian biota, 448–449 insect radiations and, 461 inselbergs and, 466 intraplate islands, 709 introduced species and, 470, 471, 473 invasive species and, 347 island rule and, 495 Japan, 500, 706 Juan Fernandez islands, 508, 509 lakes and, 529–530 lizard, 559, 561 logical, 630 Lord Howe Island, 569, 570 Madagascan lemurs, 549 Madagascar, 581 mammal endemics, 590 Marianas, 706 Marquesan birds, 223 Mascarene Islands, 615, 617, 618 Mauritanian ants and, 39

Mediterranean, 628 metapopulations and, 630 Micronesian, 716 Midway Island, 633 New Caledonia, 644–645 New Caledonian plants, 255 New Zealand, 669, 671, 672, 673 New Zealand bats, 671 New Zealand birds, 314 oases and, 688 pigs and goats and, 742 Pitcairn birds, 747 pollination and, 699 prehistoric, 595 Rapanui, 245–246 rats and, 795 Samoa, 801 seabirds, 814 seamounts and, 820 Seychelles, 832 Seychelle tortoises, 831 size of islands and, 859, 860 sky islands, 842 snakes and, 845 Socotra, 849, 850 species-area relationship and, 858, 943 species richness variation and, 13 St. Helena, 959 Taiwan, 901 Tasmania, 959 Tonga, 920 tortoises, 921, 922, 924, 926 trees, 874 Tristan da Cunha/Gough, 931 vent, 426 weevils, 794 whaling and, 977 extinctions, local, dispersal and, 225, 226 extinctions, overviews, 220 extrusion, overviews, 491 extrusion tectonics, Philippines and, 733 Exxon Valdez, 834 eyes, 151 caves and, 548 Hawaiian crickets, 461 lemur, 552 snake, 843 spider, 863, 863 Eysturoy Island, 288, 290, 291, 292 Faroe (Faeroe) Islands, 50, 63–64, 104, 287–293, 291, 310 Faial Island, 64, 71, 72 falcons, 159 Falkland Islands, 11, 12, 66. See also continental islands Beagle and, 957 continent and, 381–382 rats and, 795 research stations, 789 Falkland Plateau, 382 Fangataufa atoll, 336, 339, 341, 685, 708, 711 Faraday/Vernadsky station, Antarctica, 15 Farallon Islands, 293–297, 377

geology, 707 Farallon plate, 78, 164 Baja California and, 703 Japan and, 503, 505, 706 Farallon Ridge, 426 Farne Islands, 116, 125 faros, 68, 587 Farquhar Island, 177 Fatu Hiva, 342 faulting. See also fracture zones; specific faults arctic, 56, 57 Asian strike-slip, 506 Baja California Islands, 80, 81 beaches and, 92 Canary Islands, 136, 142 Caroline Islands and, 149 CBCs and, 175 Channel Islands (California) and, 161–162, 164 Cyprus, 214, 215 dispersal and, 425 earthquakes and, 242 Fiji and, 308 Galápagos and, 370 Greek Islands and, 392–393, 393–394, 395 Hawaiian, 408 Iceland earthquakes and, 435 Indian region, 444 Indonesia, 458, 459, 460 island arcs and, 481 Japan and, 504 Kurile Islands, 523 Macquarie Island, 577 Mediterranean, 626 New Caledonia, 647, 648 New Guinea, 663, 664, 665 New Zealand, 676 overviews, 240 Pyrenean, 624 ridges and, 425 Samoa, 807 seamounts and, 822 Solomons, 856 strike-slip, 454, 457 Taiwan, 903 Tasmania and, 905 vicariance and, 949 fauna arctic, 57 Atlantic, 67 Baja California Islands, 80 Britain and Ireland and, 119–122 Channel Islands (California), 155, 157–158 Comoros, 178 competition and, 284 Great Barrier Reef, 385 lava flows and, 547 Samoan, 800–801 Socotra, 848 succession and, 878 Vanuatu, 941 fecundity, 102, 225, 226, 285, 672, 924 Federated States of Micronesia (FSM), 94, 148, 641, 681, 770, 782, 895, 896

fellfield vegetation, 11, 14, 289, 574 female choice, 6, 106, 110–111, 233, 656, 826, 827, 828 Fernandez, Juan, 509 Fernandina Island, 313, 356, 357, 359, 364, 366, 367, 368, 369, 370, 371, 710, 952, 952 Fernando de Noronha archipelago, 65, 66, 297–298, 560, 955, 956 Fernando Nerhona, 218 Fernando Po Island, 65 fernbrakes, 14 ferns Bermuda, 98 biocontrol and, 101 Britain and Ireland and, 118 deforestation and, 221 Fijian, 299 Gough Island, 931 Gulf of Guinea, 810 inselbergs and, 468, 469 Juan Fernandez islands, 507 Madeira, 583 St. Helena, 871 sustainability and, 891 Taiwan, 899, 899 Tonga, 920 Tristan da Cunha, 931 Vanuatu, 940 Ficus, 852 fig trees, 88–89, 617, 944 Fiji amphibia, 305 ants, 36, 38, 773, 775 bats, 194, 801 biocontrol and, 101–102, 305 biology, 298–305 birds, 302–304, 305, 325–326 conservation, 305, 782, 895, 896 crickets, 208 endangered species, 285 extinctions, 304–305 fishes, 301, 305 frogs, 348, 349 geology, 305–309, 690, 693, 706 humans and, 42 introduced species, 305, 471, 477 mammals, 304 missionaries and, 634, 637 reptiles, 301–302 skinks, 560 snails, 539 traditional practices, 890 Fiji Fracture Zone, 307, 309, 693 finches, 106, 131, 159, 326, 412, 477. See also Galápagos finches; honeycreepers Finney, Ben, 760 fires Borneo forests and, 115 Channel Islands (California), 159 deforestation and, 222 extinctions and, 171 introduced species and, 471 Juan Fernandez islands, 509 New Caledonia, 645

New Zealand and, 672 Tasmania, 905, 907 Vanuatu, 941 Firth of Clyde, 125 fish. See also fish, freshwater; marine life and environments; shellfish arctic, 52, 57 Baffin Island, 77 Bermuda, 97 Britain and Ireland and, 124 Canary Islands, 131 Cape Verde, 145 Channel Islands (California), 159 Comoros, 179, 180 conservation, 783 Cook Islands, 194, 197 deep-sea, 756–757 Faero Islands, 289 French Polynesian endemics, 335, 336 Gulf of Guinea, 809 humans and, 43 insular, 310–311 introduced species and, 470 Mascarene Islands, 615 Midway Island, 632 migratory, 311 New Caledonia, 644 New Guinea, 653–654 Pitcairn, 745 rifting and, 949 Rottnest Island, 798 seamounts and, 819, 820 Seychelles, 832 shipwrecks and, 834 Solomons, 853 Surtsey Island, 885 sustainability and, 891 Taiwan, 899 Tasmania, 906 waves and, 880, 884 Zanzibar, 984 Fisher, John, 87 Fisher, R.A., 826 fish, freshwater Antilles, 28 Borneo, 114 Borneo/New Guinea and, 658 Japan, 499–500 lakes and, 530 New Guinea, 654 New Zealand, 669, 673 Philippines, 729 Seychelles, 832 Solomons, 852 sticklebacks, 873–875, 876 Taiwan, 899 Trinidad and Tobago, 927 fishing, 310–311 Bermuda and, 98 Cape Verde, 147 Comoros, 180 conservation and, 782, 784, 895 Cook Islands, 196 coral and, 564, 567, 781

INDEX

1037

fishing (continued) Farallon Islands, 296 Galápagos, 366 Kurile Islands, 525 mangroves and, 593 marine protected areas and, 607, 608 poison for, 275 Rapanui, 245 Rottnest Island, 798 seabirds and, 814 seamounts and, 820 Solomons and, 853 Tasmania and, 907 Zanzibar, 985 fission, 37 fissures, 949 fitness, 108, 225, 226, 437, 748, 751, 826, 827. See also natural selection FitzRoy, Robert, 918, 954, 955 fjords, 54, 56, 57, 58, 60, 77 Icelandic, 432, 436 Indian Ocean region, 440 Spitsbergen, 865 flamingoes, 358, 359 flatworms, 124 fleas, 121, 906 flies alien species, 475, 478–479 biocontrol and, 101–102 Britain and Ireland and, 125 Cape Verde, 145 eradication of, 474 French Polynesian endemics, 335 New Guinea, 656 Tasmania, 907 whales and, 977 flight, 813 flightlessness, 311–318 Channel Islands (California), 160 dispersals, 460–461 Drosophila, 234 extinction and, 813 Fiji, 325 fossils and, 319 Galápagos, 359, 361, 812 gigantism and, 375, 376 global warming and, 378 Hawaiian Islands, 326, 398–399, 773 mammals and, 322, 323, 325 Marianas, 595, 706 New Caledonia, 644 New Guinea, 655 New Zealand, 184, 324, 671, 672, 707 pre-European impacts, 416 Tasmania, 905, 906 Tristan da Cunha/Gough Island insects, 931 Flinders, Matthew, 332 floating objects, 258–259 flooding Taiwan, 904 tides and, 916 tsunamis and, 936 floor mats, 273–274 Flores Island, 64, 71, 236, 237, 267, 276, 375, 439, 446, 448, 452, 454, 459, 495, 514, 934

1038

INDEX

Florida, 152, 175, 477, 542, 565, 781, 855, 977 Florida Keys, 236, 378, 419, 420, 591, 835, 855, 977 flowering plants. See also ornamental plants Borneo, 113 Britain and Ireland and, 118, 124, 126 Cook Islands, 194 endemism and, 255 Galápagos, 365 Gough Island, 930 Hawaiian, 774 Indonesia, 450 Philippines, 724, 726 Solomons, 851–852 St. Helena, 871 Tristan da Cunha, 930 Wallace’s Line and, 970 flycatchers, 106, 159, 194, 197, 204, 303, 304, 335, 336, 337, 413, 597, 809, 830 flysch, 212, 213, 214, 625, 626, 647 fodder, 146, 147 Fogo Island, 65, 143, 147 folding. See also specific fold belts Falklands, 957 Fiji and, 308 New Guinea, 663, 664, 665 Tasmania and, 905 Fonualei, 484 food (nutrients). See also ethnobotany; nutrient availability; resource exploitation; resource limitation; trophic structure coral and, 781 intertidal habitats, 910, 912 island rule and, 494 moa and, 640 popular culture and, 763 seabirds and, 812 seamounts and, 820 snakes and, 843, 844 succession and, 877 tides and, 914, 915, 916 waves and, 911 foothills, 898, 903 foraminifers, 817–818 Forbes, H., 848 forbs, 12, 81, 146 147, 632, 878, 887, 891 fore-arc, 483 fore-reefs, 556, 557 forests. See also deforestation; trees; specific kinds of forests Antilles pine, 27 arctic, 52, 58 Ascension, 61 Atlantic island, 72 Baja California Islands, 78 Barro Colorado, 89, 89–90 Bermuda, 95, 98 Borneo’s, 115 Britain and Ireland and, 117–118, 124 Canary Islands, 128, 130, 132 Channel Islands (California), 155, 157 climate and, 613 climate change and, 170, 171 Comoros, 178, 179 Cook Islands, 194

dry, 335 fragmentation, 182, 328, 329 Gulf of Guinea, 809, 810 humans and, 44 Juan Fernandez islands, 509 kı-puka and, 513 Krakatau, 519 Kurile Islands, 525 macaques and, 778 Madagascar, 582 maritime, 83 Mediterranean, 628 New Caledonia, 645 New Guinea, 658 New Zealand, 666, 666–667 overviews, 941–942 Pacific region, 944 Pacific region extinctions, 946–947 Panamanian, 88 Pitcairn and, 744 Rapanui, 245 refugia, as, 786 rodents and, 793–794 Rottnest Island, 797 Samoan, 713 snails and, 541 Solomons, 851 St. Helena, 872–873 Taiwan, 898 Tierra del Fuego, 918 Tonga, 920 Trinidad and Tobago, 930 Zanzibar, 984 formation of islands, 490–492 Formentera, 237, 321, 624 Forrest, Jessica, 530 fossil fuels. See also hydrocarbons arctic, 54, 58 Borneo, 112, 115 Channel Islands (California), 161 coral and, 566, 567 Iceland, 433 Madagascar and, 582 New Guinea, 665 Philippines, 737 Sakhalin, 770 seabirds and, 814 shipwrecks and, 834 South Georgie, 977 Trinidad and Tobago and, 929 fossils Antarctic, 19 Antilles, 20, 25 Antilles bat, 26 arctic tree, 58 Ascension bird, 62 Bermuda, 96 Canary Islands, 127, 131 Cape Verde, 144, 145 Channel Islands (California), 157, 158, 160 cold seep, 177 Darwin and, 221 dodo, 230–231 dwarf, 237, 238 endemism and, 256

Greek Islands, 395 Hawaiian bird, 411 Lord Howe Island, 569 Newfoundland, 649 New Zealand, 670 Philippines, 724 primates, 778 Sri Lanka, 869 St. Helena, 871 Tasmania, 908 taxon age and, 948 whale, 974 Foster, John Briston, 493 Foster, Robin, 89–90 Foundation seamount chain, 714 founder effects and events, 5–6, 326–328 African lakes, in, 165 Antarctic, 13 bird, 107 cricket, 209 DNA and, 110 Hawaiian Drosophila and, 233 inbreeding and, 437 ontogeny and, 9 orchids, 696, 699 plant disease and, 751 sexual selection and, 826–827 silverswords and, 836 foxes arctic, 49–50, 51, 52, 54 Baffin Island, 78 Britain and Ireland and, 120 Channel Islands (California), 158, 160, 161, 707 island rule and, 492 Kurile Islands and, 525 Tasmania and, 908 fracture zones. See also faulting; specific fracture zones Antilles, 20 Baja California, 79 Canary Islands, 142 Cook Islands, 192 Galápagos, 368 hydrothermal vents and, 425 Mascarene Islands and, 612 Mid-Ocean Ridges and, 753, 822 Pantepui and, 718 Samoa, 803, 805, 807 fragmentation (patches), 328–330. See also island-like systems; metapopulations biogeography and, 489–490 dispersal and, 226, 227–228 Drosophila, 235 endemism and, 254 ephemeral islands, 259 extinction and, 282, 788 Mascarene Islands, 618, 619 speciation and, 585 spiders, 864 Sulawesi and, 553 Tasmania, 908 Zanzibar, 985–986 France bird disease and, 104

Channel Islands (California) and, 155 Comoros and, 180 François I, 771 frankincense, 850 Franklin Island, 17 Franz Josef Land, 47–48, 50, 51, 52 Fraser Island, 330–332 freezing strategies, 14, 16 French Caribbean, 421, 422 French Polynesia, 332–343 adaptive radiation, 3, 332, 712 biocontrol and, 101 Cook Islands and, 192 geology, 708 invasive species, 478 missionaries and, 635 motu, 641 overviews, 711–712 Polynesian voyaging and, 758 research stations, 789 snails, 537, 542 volcanism, 704 freshwater habitats and organisms, 343–347. See also aquatic species; fish, freshwater; hydrology; lakes Greek Islands, 391 Hawaii, 402 Line Islands, 557 Madagascar, 580, 582 motu, 642 Rottnest, 797 Socotra, 849 Solomons, 851, 852, 854 Tasmania, 906, 907–908 threatened, 893, 894 frigatebirds, 62, 556, 811 fringing reefs Cook Islands, 192, 193, 194, 195 Cyprus, 215 Darwin on, 69, 219, 823 French Polynesia, 334 Hawaii, 420 island arcs and, 481 Marianas, 601, 602 Mascarenes, 614 oceanic islands and, 692, 695 Palau, 149, 716 Pitcairn, 744 seamounts and, 823 Tonga, 919 tsunamis and, 934 waves and, 880, 882 Fritz, Uwe, 925 Frobisher, Martin, 77 frogmouths, 907 frogs, 347–351 Antilles, 27 Baja California Islands, 78 Borneo, 114 Borneo/New Guinea and, 658 Comoros, 179 Indonesia, 453 introduced species and, 471 Mauritius, 958 maximal radiation, 773

New Caledonia, 644 New Guinea, 654, 656 New Zealand, 670 Palau, 716 Pantepui, 718–719 Philippines, 724, 725, 726, 727–728, 729 ranid, 7 Seychelles, 831 sky islands, 842 Solomons, 852 South American, 655 Sri Lanka, 868–869 Taiwan, 899 Tasmania, 906, 907 Wizard Island, 981 Fuerteventura, 64, 127, 128, 129, 132, 135, 140, 141 fulmars, 50, 104, 124, 289, 812, 887 fumaroles, 67, 960 Funafuti atoll, 69, 420 fungi arctic, 54 Barro Colorado, 89 Britain and Ireland, 124 Canary Islands, 130 Drosophila and, 235 Macquarie, 574 Madeira, 583–584 orchids and, 696–697 Pantepui, 718 Philippines, 724 plant disease and, 748–749 sustainability and, 891 Swedish archipelagoes and, 750 Tasmania, 906 GAARlandia theory, 21 gabbros Arctic, 58 Atlantic region, 67 Canary Islands, 140 Channel Islands (British Isles), 154 Channel Islands (California), 162 Cyprus, 213 Fiji, 306, 307, 308 Iceland, 429, 430, 431 Indian region, 440, 443, 445 island arcs and, 482 Macquarie Island, 576 New Caledonia, 647 New Guinea, 663, 664 overviews, 576, 752, 753 Philippines, 734 Tahiti, 340 Gabrieliño people, 158, 160 Gadow, Hans, 410 galah, 908 Galápagos. See also Galápagos finches adaptive radiation, 361–364 biology, 357–366 bird disease and, 104, 105 crickets, 206, 210 Darwin and, 218, 220, 955, 958 dispersal methods, 776 endemism and, 257

INDEX

1039

Galápagos (continued) extinctions, 170 flightlessness and, 318 geology, 220, 357, 367–372, 710, 755, 951 human impacts, 710 invasive species, 474, 479 kı-puka and, 513 lizards, 560, 561 mammals and, 588 penal colony, 770 pigs and goats and, 742, 743 plants, 848, 859 seabirds, 814 snails, 538, 775 species-area relationship and, 857 spiders, 862 subsidence and, 954 tortoises, 921–926, 922, 923, 925, 926 vicariance, 950 volcanism, 953 whaling and, 975, 976, 977, 978 Galápagos finches, 2–3, 106–107, 108, 110–111, 352–356, 365, 461, 773, 828, 950 Galápagos Rift, 425 Galápagos Spreading Center, 424 gallinules, 184, 800 Galveston Hurricane, 420 da Gama, Vasco, 832, 984 Gambiers, 333, 341 game species, 117, 120, 126, 132, 160, 289, 295, 403, 742, 743. See also hunting gannets, 122, 124, 125, 383, 672, 812 Gardiner’s frog, 831 garnet peridotite, 855 gas, natural, 54, 58, 737. See also fossil fuels gastropods Azorean, 73 Bahamas, 378 Canary Islands, 130 Cape Verde, 146 cold seeps and, 174 Fijian, 300 French Polynesia, 337 freshwater, 344, 345 human impacts, 416 marine lakes, 605 overviews, 537 São Tomé Island, 809, 811 vents, 426 Gatun Lake, Panama, 181 Gauanches, 742 Gaugin, Paul, 763 Gau Island, 308 geckos Antilles, 23, 27 Baja California Islands, 82 Canary Islands, 131 Caribbean, 559 Comoros, 179 dwarf, 27 Gulf of Guinea, 809, 810 introduced species, 472 Mascarene Islands, 616, 617 maximal radiation, 773 mice and, 794

1040

INDEX

New Zealand, 671 nonadaptive radiations, 561 Philippines, 727–728 rafting and, 775 Rottnest Island, 798 Samoan, 800 Seychelles, 831 Socotra, 849 Tonga, 920 geese, 49, 51, 78, 122, 671, 672 Geissois, 940 gene flow, 6, 766–767, 773, 826, 827, 828, 837, 841, 946 gene pools, 6, 108, 778 generalist species diseases and, 748 dispersal and, 225 extinction and, 283, 285 glaciations and, 785 overviews, 461 species-area relationship and, 860 genetic revolution model, 6, 235, 826 genotypic responses, 844 gentian, 126 geochemistry, 135–136, 141 The Geographical Distribution of Animals (Wallace), 965 geography. See biogeography; spatial considerations; specific geographical factors geology. See also specific islands, processes and structures biogeography and, 970 Darwin and, 217 global warming and, 376 overviews, 421 radiations and, 2, 464 time scales, 56 turnover and, 38 geophytes, 798 Georgia coast, 83 geothermal energy, 435, 524 Gerald Island, 51 Germany, 148, 150 gestrues, 765 Ghyben-Herzberg lens, 421 gigantism, 372–376. See also body size; island rule; tortoises barrier islands, 82 birds and, 237 Canary Islands lizards, 131 Caribbean, 24 Channel Islands (California), 160 community composition and, 236 Comoros, 178 Cozumel, 205 cricket, 211 dwarfism and, 238–239 Galápagos, 361 Japan, 499 Kurile Islands, 525 Line Islands clams, 557 lizard, 559 Mascarene Islands, 619 Mascarene Islands tortoises, 614

New Zealand, 638–641, 671 Philippines, 724 Seychelles, 830–831 snakes, 844 spiders, 864 Tasmania, 907 tortoise, 921, 922 Gilsemans, Isaac, 759 Gjalp eruption, 432 glacial-interglacial cycles. See also deglaciation; glaciers; ice sheets Borneo and, 113 Channel Islands (California) foxes and, 707 climate change and, 169 diversification and, 841 formation of islands and, 492 Greek Islands and, 390 Gulf of Guinea and, 809 Iceland, 432, 885 Indonesia and, 448 Kurile Islands and, 524 macaques and, 777 marine lakes and, 603 motu and, 641 refugia and, 785 sea levels and, 695 sky islands and, 840, 841–842 Spitsbergen and, 865 Taiwan and, 898 Tasmania and, 907 Trinidad and Tobago and, 927, 927 Vancouver and, 937 Wallace on, 966 glaciations. See also glacial-interglacial cycles; glaciers; ice ages; ice sheets Antarctic, 19 arctic, 47–48, 49, 51, 52, 58–59 arctic diversity and, 50 arctic evidence, 56–57 Atlantic, 65, 66 atolls and, 69 Baffin Island, 77 Baltic archipelagoes and, 750 birch and, 433 Borneo and, 113 Britain and Ireland and, 117–118, 122–123, 125 coasts and, 83 Faroe (Faeroe) Islands and, 287, 288 Greenland, 58 Hawaiian, 409 Iceland, 431, 434 New Zealand, 322, 666 Pantepui and, 719 South America and, 956 southwest Pacific, 969 Spitsbergen and, 865 springtails and, 13 Sri Lanka and, 868 sticklebacks and, 873 Tasmania and, 904 Vancouver and, 708, 937 glaciers. See also glacial-interglacial cycles; ice sheets arctic, 60

Atlantic, 64, 67 Baffin Island, 77, 78 climate change and, 15, 171, 971–973 formation of islands and, 973 Iceland, 432, 434–435 Indian Ocean region, 444 Montana, 817 New Guinea, 657, 708 sea level and, 816–817 Spitsbergen, 865 succession and, 879 Glass, William, 931 glassy-winged sharpshooter, 478, 479 global warming, 376–380. See also climate change arctic and, 54 atolls and, 69 barrier islands, 85, 85–86 Britain and Ireland and, 118 coral reefs and, 202, 782 Europe and, 117 French Polynesia and, 336, 711 Galápagos, 356 hurricanes and, 419 New Guinea, 708 Pantepui and, 719 seabirds and, 815 sustainability and, 892 vent fauna and, 426 Warming Island and, 971–973 gneiss, 56, 66, 67, 154, 395, 466, 626, 627, 663, 677, 903 goats, 741–743 Baja California Islands, 78 Canary Islands, 132 Cape Verde, 146 Channel Islands (California), 155 Galápagos, 365, 926 Marquesas, 223 Socotra, 850 St. Helena and, 873, 959 tortoises and, 926 whalers and, 977 gold, 149, 459, 524, 856 Golden whistlers, 106 Golson, Jack, 43 La Gomera Island, 64, 128, 130, 132, 134, 135, 136–137, 138 Gonave Island, 22 Gondwana biota, 325 birds, 312 Falklands and, 66 fauna, 831 flora, 868 fragments and, 187, 299 geology, 677 Indian region and, 445 Indonesia and, 447, 453, 457 moa and, 639 New Caledonia and, 644, 645, 646, 647 New Zealand and, 322, 665, 666, 669–670, 678, 707 overviews, 504, 580 Seychelles and, 829

Socotra and, 847, 849 South America and, 717 southwestern Pacific islands and, 950 spiders, 863, 865 Sri Lanka and, 866–867 Tasmania and, 904 tectonic activity and, 113 Gorda Ridge, 426 Gorgona Island, 769–770, 950 Gough archipelago, 11, 12, 13, 15, 16, 234, 316 Gough Island, 66, 793, 794, 929–932 Gould, John, 220, 352, 361, 958 groundwater, Cook Islands, 193 Graciosa Island, 71 Graf Spee, 770 Gran Canaria, 132, 134, 135, 138, 139–140, 557 Grand Comoro Island, 177, 178, 179 Grande Comoro, 107 Grande Terre, 646 granitic islands, 380–382 Farallon, 377 Seychelles, 438, 829, 830, 832, 833, 922 Socotra, 847 Taiwan, 897 granitic rock Aegean, 393, 394, 395 arctic intrusions, 57, 58 Atlantic region, 66, 67 Australia, 959 Baffin Island, 77, 78 Britain and Ireland, 125, 126 Channel Islands (British), 155 Channel Islands (California), 293, 707 continental islands, 755 earthquakes and, 503 Easter Island, 710 Fiji, 306, 307, 308, 309 Indian region, 338, 440, 445 Indonesia, 456, 457, 459 inselberg, 466 Japan, 505, 506 Mediterranean, 624, 627 Newfoundland intrusions, 650, 651, 652 New Guinea, 663 New Zealand, 675, 677, 679 overviews, 485 Tasmanian intrusions, 905 Vancouver intrusions, 937 Grant, Rosemary and Peter, 3, 106, 108, 353, 361 grasses. See also grasslands; sea grasses; tussocks Antarctic, 12, 14, 15 arctic, 51 Ascension, 62 Baffin Island, 78 Baja California Islands, 80 Canary Islands, 132 Cape Verde, 146 Channel Islands (California), 157, 160 conservation and, 590 convergence, 189 Cook Islands, 194 deforestation and, 221, 222 diseases, 750 dispersal and, 862, 885

Easter Island, 221 ecological release and, 252 Faroe (Faeroe) Islands, 289, 290 Fiji, 330 grazing and, 265, 743 Great Barrier Reed, 384 Gulf of Guinea, 810 Hawaii, 401, 471, 513 invasive, 470, 471, 480, 513 Krakatau, 517, 518 Kurile Islands, 520, 525 Macquarie, 574 Madagascar, 579 Mascarene Islands, 614 matrix-derived taxa, 182 Midway Island, 632 New Guinea, 653, 657 Pantepui, 718 pigs and goats and, 743 Rapanui, 245, 246 Rottnest Island, 798 Solomons, 851 spiders and, 862 Sri Lanka, 867 succession and, 946 Surtsey Island, 887 sustainability and, 891 Tasmania, 905 Tristan da Cunha, 931 Vanuatu, 940 grasshoppers, 130, 206, 361, 452, 460, 464, 667, 839, 842, 848 grasslands. See also grasses Aleutians, 815 Annobon, 810 Britain and Ireland, 117, 118, 123, 124, 126 Cape Verde, 146 Channel Islands (British), 155 Channel Islands (California), 157 convergence, 189 Cook Islands, 194 Easter Island, 245 Faroe (Faeroe) Island, 210, 290 fire and, 222 Krakatau, 518, 519 Macquarie Island, 574 Madagascar, 579 mammals and, 781 Mascarenes, 614 Mediterranean, 628 moa and, 640 New Guinea, 657, 658, 666 patchy, 182, 183, 184 rats and, 793 sky island, 841 Solomon Islands, 851 Surtsey Island, 887 Tasmania, 905 Vanuatu, 940 gravel flats, 887 gravity, 30, 735, 911, 914, 932 grazing Canary Islands, 132 Channel Island (California), 472 intertidal habitats, 910, 911, 912

INDEX

1041

grazing (continued) introduced species, 470, 471, 474 isolated islands and, 473 Marquesas and, 222–224 Socotra, 850 Great Barrier Reef Islands (GBR), 382–388, 474, 592 Great Basin, North America, 185–186, 840, 841 Greater Antilles adaptive radiation, 531 amphibians, 27 ants, 38 biogeography, 20–22 fish, 28 geology, 485 lizards, 3, 7, 190, 562 mammals and, 588 Greater Kurile Ridge, 523 Greater Timor, 448 Great Lakes, North America, 104 Great Rift Valley lakes, 165 Great Sea Reef, 707 grebes, 811 Greek Islands, 388–396 greenhouse gases, 126, 376, 426, 435, 972 Greenland age of, 428 climate effects, 47 diversity of, 49 fishes, 874 geology, 58, 755 glaciation, 59 global warming and, 972, 973 human impacts, 54 insects and plants, 51 Newfoundland and, 650 plants, 50, 51 research stations, 789 size of, 59, 111 tectonism, 55, 56, 57 Greenland Ice Sheet, 60, 972, 973 Green Mountain, Ascension, 61, 62 Green, Roger, 42, 44 greenswords, 401, 835, 837, 838 Greenwich Island, 19 Grenada, 31, 32, 35 Grenada basin, 29 Grenada trough, 30 Grenadines, 26, 35, 510 Grenvillian orogens, 504 Gressitt, J.L., 970 Gressitt Line, Antarctica, 12 de Grijalva, Hernando, 80 groundwater arctic, 61 atoll, 68 Channel Islands (British), 155 coral and, 201 Frasier Island, 332 Hawaiian, 409 magma and, 33 overviews, 420–421 grouse, 51 groynes, 261, 262, 263 Gryllidae, 670

1042

INDEX

Gryllotalpidae, 670 Guadaloupe, 24, 31, 35, 422 Guadalupe Island, Mexico, 5, 157 Guadalupe Mountains, 152 Guadeloupe, 23, 27, 31, 32 Guadeloupe archipelago, 421 Guam biology, 94 extinctions, 706 geology, 585 human impacts, 595 introduced species, 471, 473, 474, 479, 595, 845–846 missionaries and, 633, 634 overviews, 148, 593, 594 species richness, 594 guanacos, 918 guano, 104, 222, 557, 739–740, 912 Guayana, 717, 718, 719 Guernsey, 154, 155 Guguan, 595, 598, 599, 601 Guide to the Birds of Fiji & Western Polynesia (Watling), 302 Guillaumin, 940 guillemots, 125, 158, 289, 294, 295, 297, 813, 866, 887 Gulags, 770–771 Gulf Coast, U.S., 87, 916 Gulf of California, 164, 857 Gulf of Guinea Islands, 65, 560, 699, 808–811, 840 Gulf of Mexico, 175, 565–566, 567 Gulf Stream, 71, 95, 96, 116, 124–125 gulls Britain and Ireland, 122 Channel Islands (California), 159 food and, 812 Galápagos, 360 New Zealand, 322 succession and, 879 success of, 812 Surtsey Island, 866, 887 Tasmania, 908 Tatoosh, 909, 912 guyots, 134, 142, 201, 219, 339, 381, 490, 568, 610, 692, 694, 704, 706, 713–715, 713–715, 823, 917 gymnosperms, 299, 644, 667, 899, 920, 940 habitat factors. See also competition; predation anagenesis and, 9 biogeography and, 489 conservation and, 619, 630 dispersal and, 224–225 diversity and, 320 endemism and, 256–257 extinctions and, 282–283, 943 Fijian, 299 Indonesia, 448 metacommunity paradigms and, 528 size and, 187 specialist species and, 285 species-area relationship and, 859 succession and, 877 temporal, 330

Hahajima, 786 Hainan, 423 Haiti, 21, 23, 24, 25, 26, 27, 28, 29, 108, 313, 634 Hale, Marie, 119 Halmahera Islands, 446, 448, 453, 458, 459, 460, 482 H na Ridge, 405 Hoa, 334 haplo-diploidy, 226 Hardy, Alister, 61 hares, 52, 53, 78, 121, 238, 289, 493, 672, 899 Harmattan (wind), 144 Hatuta’a, 224 Hatutu, 337, 338 Hava , Peter, 925 Hawaiian-Emperor chain, 951, 954 Hawaiian Islands. See also specific volcanoes adaptive radiations, 399 age of, 71, 464 alien species, 36, 99 anagenesis and, 9 ants and, 36, 39 arthropods, 36 beaches, 93 bees, 3 beetles, 7 biocontrol and, 100, 101, 102, 748 biology, 397–404 bird disease and, 103, 104, 283, 977 bird extinctions, 283, 315–316, 324, 326 birds, 106, 322, 480, 773, 813 climate, 172, 173 climate change and, 170 coastline, 94 collapses, 696, 954 conservation, 865 Cook Islands and, 193 coral reefs, 783 crickets, 206, 208, 209, 210, 211, 773 Drosophila, 3, 6, 773 earthquakes and, 242, 243, 244 ecological release and, 252 elevation and, 594 endemism, 254, 255–256, 257, 943–944 ethnobotony and, 272, 273 extinct/endangered species, 282 fauna, 320 flightlessness and, 313, 314–315, 317 flora, 3, 4, 78, 274, 275 fragmentation and, 329 freshwater species, 346 geology, 201, 404–418, 621–622, 690, 695, 704, 707, 708, 710, 754, 950, 951, 954 gigantism, 374 honeycreepers, 2, 3, 4 humans and, 41, 42, 46, 416, 417–418, 632, 720 hurricanes and, 345, 420 introduced species, 470, 471, 472, 473, 477 invasive species, 346, 347, 478, 480 kı-puka and, 512, 513 lava, 542–543, 547 lava caves, 548 lava tubes, 545 marine endemism, 96

name of, 758 popular culture and, 762, 764, 765 radiation zone, 772, 773 rats, 794 refugia and, 786 seabirds, 811, 812, 814 seamounts, 822, 824–825 shipwrecks, 834 silversword alliance, 4–5, 835 snails, 537, 538, 540, 541, 542, 775 source biotas, 461 species overviews, 475 spiders, 5, 7, 188, 531, 826, 862, 863–864 tides and, 916, 916 volcanism, 134, 135, 140, 704, 755, 804–805, 952, 953 whales, 975, 976, 977, 978 Hawaii Division Line, 434 Hawaii-Emperor chain, 693, 714 hawkmoths, 970 hawkweeds, 50 head hunting, 708 Heald, Weldon, 839 Heaney, L.R., 773 Heard Island, 11, 13, 16, 440–441 heat dissipation, 825 heat generation, 843 heather, 49, 119, 178, 290 heath forests, 114, 157 Hebrides, 116, 120, 123–124, 292 hedgehogs, 120, 132 Hekla, 433 Helen, 963 Hellenic Arc, 394 Henderson Island, 44, 45, 320, 490, 744 Henslow, John, 955 herbfields, 14 herbivores, 453, 462, 472, 559, 878. See also grazing herbs, 891, 905, 981 Hercynian-Appalachian orogenies, 505 Herm, 154 Heron Island, 374 herons, 62, 147, 256, 320 Herre, Allen, 88 Herzl, Theodor, 769 Hesperelaea, Oleaceae, 78 heterochrony, 239, 312 heterosis, 225 Heyerdahl, Thor, 250, 515, 722, 760 El Hierro Island, 132, 133, 134, 135, 136 high limestone islands, 694 Hilo, 207 Hilton Head Island, South Carolina, 85 Hindus, 115, 445, 633, 983 hippopotami, 236, 237, 321, 322, 628, 869, 922 Hirta, island of, 121 Hispaniola adaptive radiations, 24 ants, 36, 37, 38 bats, 26 biogeography, 20, 22, 23 birds, 26, 108–109, 110 butterfly diversity, 28 colonization, 278

crickets, 210 fish, 28 geology, 692 introduced species, 471 invasive species, 346 land mammals, 24, 25 lizards, 559, 562 reptiles/amphibians, 27 rodents, 23 shrews, 25 size of, 29 HMS Challenger, 97 hoa, 68 hobbits, 237, 495 Hokkaido, 498, 499, 500, 501, 503, 504, 520, 523, 524, 525, 536, 868 Hokule’a (watercraft), 760 Holarctic, 497, 863, 873, 874, 876. See also North America holothurians, 174 Holt, 860 Holy Island, 125 Homo erectus, 237, 276 Hondo, 492, 499 honeycreepers adaptive radiation, 2 Hawaii, 401, 402, 410–414, 711, 773 overviews, 639 radiation, 106 radiation zone, 773 honeydew, 472, 478, 479, 535, 672 honeyeaters, 800, 920 Honshu, Japan, 423, 482, 484, 486, 497, 498, 499, 500, 501, 506 Hooker, Joseph Dalton, 61, 870, 964 hoopoes, 316, 871 horses, 51, 120, 155 hotspot islands. See also mantle plumes Ascension/ St. Helena, 66, 959 Canary Islands, 133, 134, 956 Caroline Islands, 709 Easter Island, 709–710 French Polynesia, 333, 335, 339, 341–342, 342–343 geology, 553 Iceland, 428 insects, 462, 463, 464 Juan Fernandez islands, 507 Mascarene Islands, 620–622 overviews, 339, 340, 342, 690, 693, 704 Pacific region, 704, 709–713 Samoa, 799–800, 803 hotspots, biodiversity. See biodiversity hotspots hotspots, geological Atlantic, 66 Caroline Islands, 149 Comoros, 177 Cook Islands and, 191–192 French Polynesia, 341 Galápagos, 367, 950 Hawaiian, 404 Hawaiian/Louisville Seamount, 714 Indian Ocean region, 438, 439, 441–442, 444, 445

Mascarene Islands, 612 Mediterranean region, 147 oceanic islands and, 951 overviews, 201, 714–715, 754 Pitcairn, 744 seamounts and, 821, 824 hotspot volcanoes, 755 house mice, 793 Hual lai volcano, 405 Hubbell, Stephen, 89–90 huia, 672 human colonizers. See also Austronesians; Europeans; Lapita; Polynesians Bermuda, 834 Kurile Islands, 705 Madagascar, 577, 579, 580–581 Marianas, 594–595 Marshall Islands, 611 Mediterranean, 626, 629 motu, 642 New Guinea, 658, 708 Pacific region, 720–723 Palau (western Pacific), 715–716 Pitcairn, 745–746 Seychelles, 832 Solomons, 706 St. Helena, 871 Tasmania, 907 Tatoosh, 909 Tristan da Cunha, 929 West Indies, 590 Zanzibar, 984 human dwarfs, 237 human impacts. See also archaeology; dodos; ethnobotany; human colonizers; introduced species; specific activities; specific peoples Antarctic, 14–15 ants and, 36, 40, 41 aquatic species dispersal, 528, 529 arctic, 54, 59 Ascension, 62, 959 Ascension Island, 61, 62 atolls, on, 68 Azores and, 71, 72, 74 Baja California Islands, 79–80, 82 barrier islands and, 83 beaches and, 94 Bermuda and, 96–98 biodiversity, on, 29, 107 birch, on, 433 bird disease and, 104 birds and, 103, 230, 319, 320 body size and, 285 Borneo and, 115 Britain and Ireland and, 117–118, 120, 121, 123, 125, 126 Canary Islands, 128, 131, 132 Cape Verde, 145, 146 Caroline Islands, 149 Channel Islands (California), 155, 157, 158, 159, 160–161 Christmas Island crabs, 534 Comoros, 177, 180 Cook Islands, 195, 196

INDEX

1043

human impacts (continued) coral reefs, on, 202–203, 566–567, 779–780 coral reefs and, 782 Cozumel, 205 crickets and, 208, 209 Darwin and, 220 diversity, on, 169 Easter Island, 710 ephemeral islands and, 258, 259 erosion, 261, 262 Europe, 117 extinction and, 170, 171, 282, 284, 286, 773 Faroe (Faeroe) Islands, 289, 291–292 Farallon Islands, 294–295, 297 Fernando de Noronha archipelago, 297 Fiji, 304–305, 707 flightless birds, 314, 316, 322 fragmentation and, 187, 328, 330 Frazer Island, 332 French Polynesia, 332, 336, 337, 338 freshwater species, 343 freshwater species and, 346 Galápagos, 356, 364, 773, 958 global warming and, 972 Great Barrier Reef, 385 Greek Islands, 388, 391 Gulf of Guinea, 810 Hawaii, on, 403, 711 Hawaiian birds, 326 Indian Ocean region, 442 Indonesia, 451 inselbergs and, 466–467 invasive species vectors, 478 isolated islands and, 99–100 Japan, 500, 706 Juan Fernandez islands, 508, 509 Kurile Islands, 522, 525 lakes and, 531 Line Islands, 557 lizards and, 560 Lord Howe Island, 570, 571, 572 Macquarie, 575 Madagascar, 549, 579 Madeira, 584 mangroves and, 377–378 Marianas, 706 Marshall Islands, 611–612 Mascarene Islands, 613, 617–618 Mediterranean, 628 Midway Island, 632 New Caledonia, 325, 645 New Guinea, 658–659 New Zealand, 323, 639, 672–673, 707–708 Panama, 88 plant disease, 748 predation and, 283 rain forests and, 946 Rapanui, 245, 248 rats and, 793 reefs and, 779 refugia and, 785 richness variation and, 13 Rottnest Island, 797, 798 Samoa, 713, 801 Sarawak caves, 114

1044

INDEX

seabirds and, 814 Seychelles, 833 Socotra, 850–851 Solomons, 853–854 St. Helena, 873 sustainability and, 888, 894 Taiwan, 900–901 Tierra del Fuego, 918 Tonga, 918, 919, 920 tortoises, 924, 926 Tristan da Cunha/Gough, 931–932 vent communities and, 427 Wallace on, 962, 966 whales, 978 Zanzibar, 985 human impacts, pre-European, 414–418 Humboldt Current, 357, 368 seabirds and, 812 humidity crabs and, 533 crickets and, 210 endemism and, 257 frogs and, 301 lemurs and, 550 Madeira and, 582, 583 Mediterranean climates and, 389 overviews, 421 Sri Lanka and, 868 hummingbirds, 91, 159, 236, 303, 507, 981 hunting, 115, 118, 743, 800. See also game species; whaling Hurricane Allen, 202 Hurricane Hugo, 84 Hurricane Iniki, 418, 420, 474 hurricanes, 418–420. See also specific hurricanes Bahaman spiders and, 860 barrier islands and, 87 Bermuda and, 95 bird eradication and, 474 Cozumel, 205 crickets and, 211 freshwater species and, 345 seabirds and, 814 succession and, 877 Huxley’s Line, 723, 729 Huxley, T.H., 968 hybridization adaptive radiation and, 6 African lakes, 165 Britain and Ireland, 118, 120, 125 cichlid fish, 828 coral, 779 Darwin’s finches, 354–355 extinction and, 285 Greek Islands, 390 introduced species and, 472 lizards, 561–562 macaques, 778 plant disease and, 749 reproductive isolation and, 110 selection and, 108 silverswords and, 837 Wallace on, 966 hydrocarbons, 174, 175, 458, 736–737. See also fossil fuels

hydrocorals, 199 hydroids, 819 hydrology, 420–424 hydrothermal areas, 521, 524 hydrothermal vents, 424–427, 425, 700, 974 hyenas, 119 Hymenoptera, 145 Iapetus Ocean, 57, 504, 505, 649, 650, 651, 652 Iberia, 118, 863 Iberian subplate, 624 Ibiza, 237, 321, 764 ice, 13, 49, 50, 59, 60, 77–78, 93 ice ages, 23, 56–57, 58–59, 78, 170, 248 icebergs, 60, 181 Iceland, 428–436. See also Surtsey Island biology; specific eruptions bird disease and, 104 dispersal and, 50 eruptions, 952 fishes, 874 fishing, 310 formation of, 490 geology, 55, 58, 690, 690–691, 753, 754, 755, 885, 951 human impacts, 291 hydroclimate, 423 overviews, 63–64 subsidence and, 954 volcanism, 952 whaling, 978 Iceland-Faroe (Faeroe) Ridge, 951 Iceland Plateau, 63, 428 ice sheets, 17, 60, 254, 255, 817–818. See also glaciation; glaciers igneous rocks. See also granitic rock Antilles, 30 arctic, 55, 56, 58 Atlantic region, 65 Canary Islands, 133, 135 Cape Verde, 144 central Pacific, 950 Channel Islands (British), 154 Channel Islands (California), 162, 163 chemistry, 433 Cyprus, 214 Darwin on, 220, 957 Easter Island, 710 Fernando de Noronha, 298 Galápagos, 371 Iceland, 431 Indian region, 437, 439, 441, 442 island arcs and, 482, 483, 485, 486 island formation and, 491, 492 Japan, 503 Mediterranean region, 624, 625, 626 New Caledonia, 646, 647 New Guinea, 662 New Zealand, 675, 677 oceanic islands and, 691 overviews, 381, 691 Pacific region, 703, 710 Philippines, 732, 733, 734, 734, 735, 736, 736 plutonic vs., 220 seamounts and, 823

Tierra del Fuego, 957 volcanic islands and, 950 Warming Island, 972 iguanas Antilles, 27 Comoros, 179 Fijian, 301–302 Galápagos, 359, 360, 365, 372, 559, 710, 828, 958 Galápagos endemism and, 257 Tonga, 920 Vanuatu, 940 iguanids, 179 Ikhotsk plate, 522 I-Kiribati people, 557 immigration, 489, 526, 527, 529–530, 860. See also colonization; introduced species; propagules species-area relationship (SAR) and, 859 immune systems, 103 Important Birds Areas in Fiji (Masibalavu and Dutson), 302 mralı, 771 Inaccessible Island, 66, 108 inbreeding, 436–437 coefficient of, 766 dispersal and, 225, 226, 227 extinction and, 284 rats and, 793 silverswords and, 838 India Borneo and, 115 drift, 177 frogs, 348–349 geology, 458 Gondwana and, 580, 669 mammals, 121 missionaries and, 637 snakes, 844 India-Australian plate, 454, 918 Indian Ocean plate, 439, 969 Indian Ocean region ants, 37, 39 atolls, 67 biogeography and, 107 climate, 173 continental effects, 172 crickets, 208, 209 currents, 840 earthquakes, 242 fishing, 832 flightless birds, 313, 316 geology, 437–446 human colonization, 720 islands of, 177 prisons, 771 research stations, 789 seabirds, 814 seamounts, 823, 824 tortoises, 924 vents, 426 volcanism, 755 weevils, 13 Indian plate, 440, 441, 454, 620, 753 Indians, 832

indigenous species, 15, 475, 943 Indo-Australian archipelago, 970 Indo-Australian plate, 306 Indochina-East Malaya block, 457 Indo-Malaysia, 274 Indonesia. See also Maritime Continent; Sunda; specific eruptions biology, 446–453 Borneo and, 115 conservation, 895 coral reefs, 200, 780 dwarfs, 237 fishing, 310 forests, 452 frogs, 348, 349 geology, 437, 454, 536, 695 global warming and, 380 invasive species, 347 mammals, 776 snails, 537 tsunamis, 934 whaling, 976 Indonesian Borneo, 451–452 Indonesian volcanic arc, 439 Indo-Pacific region, 200, 277, 756, 757 influenza, 104–105 infrared radiation, 258 Inner Hebrides, 116, 124–125, 292 insectivores, 23 insects. See also flightlessness; specific types of insects adaptive radiation and, 4 Antarctic, 12, 15 Antilles, 28 ants and, 40, 941 arctic, 50, 53 Australia, 970 Azorean, 73, 74 Baja California Islands, 78 Barro Colorado coevolution, 90 Bermuda, 98 biocontrol and, 100 Britain and Ireland and, 122–123 Canary Islands, 130, 131 Cape Verde, 145 Channel Islands (California), 159 dispersal, 227, 460 endangered, 285 extinctions, 500 fecundity, 226 introduced species and, 473 Japan, 499 Lord Howe Island, 569, 571, 786 Marianas, 594 Midway Island, 632–633 New Caledonia, 645 New Guinea, 655–656 New Zealand, 667, 669, 672 open sea, 813 Philippines, 729 radiations, 460–465 rafting and, 775 Seychelles, 831 sky islands, 842 Society Islands, 712

Socotra, 848 Solomons, 852 species-area relationship and, 857 Surtsey Island, 886 sustainability and, 891 Taiwan, 899, 900 Tasmania, 907 temperatures and, 14 trade-offs and, 89 Wizard Island, 981 inselbergs, 182, 186, 466–469 in-situ speciation, 839–840, 947 in-situ vocanoes, 950 insular communities, 321, 328, 384, 687–688, 749 insular dwarfism. See dwarfism insular radiation, 3–4, 326, 348, 561–562 integration, 44 Interior Low Plateau (U.S.), 152 inter-patch dispersers, 182 intertidal habitats, 160, 909, 912 Intertropical Convergence Zone (ITCZ), 144 intraplate islands Antarctic, 19–20 Fiji, 706 Indian region, 437, 439, 444, 445 Line Islands, 554 New Zealand, 675 oceanic islands, 690 overviews, 693, 708–710 Pacific region, 704 seamounts, 821–822 introduced species, 469–475 Channel Islands (California), 707 Fiji, 707 Galápagos, 356, 365 Gulf of Guinea, 810 Hawaiian, 414, 773, 865 honeycreepers and, 413 Juan Fernandez islands, 508–509 Kurile Islands, 525 Line Islands, 556 Madagascar, 579 Madeira, 585 Marianas, 595–596 Marshall Islands, 611 Midway Island, 632 New Guinea, 658–659 New Zealand, 671, 672, 708 Pitcairn, 744–745 Rottnest Island, 797 seabirds and, 815 snakes, 843, 844–846 St. Helena, 873 Tasmania, 906, 908 Tierra del Fuego, 918 Tristan da Cunha/Gough, 929, 932 Vanuatu, 940 whaling and, 977 intrusive rocks arctic, 57, 58 Newfoundland, 650, 651, 652 New Guinea, 664, 665 overviews, 491, 691 Philippines, 736

INDEX

1045

intrusive rocks (continued) Solomons, 855 St. Helena, 871 St. Paul’s Rocks, 956 Tasmanian, 905 Vancouver, 937 Inuit people, 54, 76, 77, 266, 976 invasional meltdown, 472, 479 invasive species, 475–480. See also alien species; exotic species; human impacts; human impacts, pre-European; introduced species; Polynesians; specific invasives ants, 39, 40, 472, 479, 708 dispersal of, 742 ecological release and, 253 French Polynesia, 336 freshwater species and, 346–347 Galápagos, 710 Hawaiian, 403 Juan Fernandez islands, 508, 509 kı-puka and, 512, 513 Lord Howe Island, 571–572 Madagascar, 581 Marianas, 706 Mascarene Islands, 618, 619 Midway Island, 633 New Zealand, 673 niaouli, 645 overviews, 742, 748 plant disease and, 749 Samoa, 801 Seychelles, 833 silverswords and, 838 size of islands and, 943 snails and, 541–542 Socotra, 850 Solomons, 854 succession and, 946 sustainability and, 896 Tasmania and, 908 tortoises and, 926 Vanuatu, 941 invertebrates Antarctic, 14, 19 ants and, 41 arctic, 52–53, 54 Ascension, 62 Azorean, 73 biocontrol and, 100 Britain and Ireland and, 122–123 Canary Islands, 130–132, 774 Cape Verde, 144–145 Comoros, 179, 180 extinctions, 540–541, 908 Gough/Tristan da Cunha, 930 Great Barrier Reef, 383 Gulf of Guinea, 809–810 introduced species and, 470 Japan, 499 lava flows and, 547 Lord Howe Island, 569 Macquarie, 574 Madagascar, 580 Madeira, 584

1046

INDEX

mangroves and, 593 New Caledonia, 644 New Guinea, 655–656 New Zealand, 667, 669, 670, 672 Palau, 716 Philippines, 724 rafting and, 775 rats and, 794 reefs and, 201–202 Rottnest Island, 797 Samoa, 800 seamounts, 819, 820 Seychelles, 833 shipwrecks and, 834 Solomons, 852, 853 St. Helena, 871–872 Surtsey Island, 886, 887 sustainability and, 891 Tasmania, 906, 907 Tatoosh, 909 Tristan da Cunha/Gough, 932 Ionian Islands, 393–394, 625 Ireland, 292, 567, 764. See also Britain and Ireland Irian Jaya, 660 irrigation, 224 Irwin, Geoffrey, 42, 760 Isla de Cedros, 79 Isla de Guadalupe, 377 Isla de Mona, Puerto Rico, 93 Isla Grande, 471, 472 Isla Guadalupe, 78 Isla Guafo, 814 Isla Isabela (Galápagos), 356, 357, 358, 359, 361, 363, 364, 365, 366, 369, 370, 474, 710, 923, 952 Islam, 115, 633 The Island. A Journey to Sakhalin (Chekhov), 770 island arcs, 481–486. See also specific arcs Antilles, 29–31 Baffin Island and, 76 Borneo and, 112, 113 ephemerals and, 260 Fiji and, 306–308 geology, 950–951 Greek Islands, 388 Indian Ocean region, 438, 439 Japan, 500–501 Java, 450–451 lava tubes and, 545 Lesser Sunda, 452 Lyell and, 220 mass wasting, 954 overviews, 691, 705, 754 seamounts and, 821, 822–823, 825 volcanism, 951, 952 island effect, 167–168, 172, 928–929 island hopping, 107, 594, 639, 698, 713, 777, 863 Island Life (Wallace), 870, 965–966 island-like systems, 248, 258–260, 460, 613. See also caves; continental islands; fragmentation (patches); hydrothermal vents; kı-puka; marine lakes; oases;

organic falls; Pantepui; patches; seamounts; shipwrecks; sky islands; whale falls island rule, 237, 238, 374, 375, 492–496, 606 Isla Pinta, 360, 362, 363, 364 Isla Porcia, 423 Isla San Benedicto, 80, 204 Isla San Pedro Mártir, 81, 82 Isla Santa Cruz, 367 Isla Santiago, 367, 368 Islas Revillagigedo, 80, 82 Isle of Anglesey/Arran/Islay/Man/Skye/Wight/ Scilly, 116, 124–126 Isle of Youth, 21, 27 islets. See also cays (keys, quays) Aldabra, 924 Antilles, 22 Atlantic region, 65, 66 Bermuda, 95 Canary Islands, 132 Cape Verde, 143, 147 Caroline Islands, 149 Cocos, 959 Cook Islands, 192, 193 Fernando de Naronha, 297, 298 French Polynesia, 333, 334 Greek, 388, 390, 391 Gulf of Guinea, 808 Hawaiian, 835 Indian region, 441, 442, 445 Juan Fernandez, 507 Kuriles, 521, 525 Lord Howe Islands, 568, 570, 571 Macquarie Island, 573, 576 Madeira, 582 Marianas, 600 Marshalls, 610, 611 Mascarenes, 614, 618 Midway, 631 New Zealand, 674 North Farallones, 293, 296–297 Phosphate Islands, 738 Rottnest Island, 796, 798 Samoa, 637, 799 snakes and, 844 South America, 926 species-area relationship and, 859 Tatoosh, 909 Vanuatu, 939 isolation. See also allopatric speciation; ecological isolation; fragmentation (patches); pocket basins; reproductive isolation adaptive radiation and, 352–353, 531 alien species and, 99 Azores and, 72 Baja California Islands, 81 biogeography and, 107, 488 birds and, 103 Borneo and, 113 caves and, 150, 151, 152, 153 Channel Islands (California) fauna and, 159 characterized, 3 continental islands and, 184, 185 distance, by, 767

diversity and, 186, 272, 329 ecological, 210–211 endemism and, 52, 254, 256, 257, 320, 335, 359 evolution and, 2, 124, 234 extinction and, 787, 943 finch speciation and, 355 fragmentation and, 329 French Polynesia, 333 freshwater species and, 343, 344, 345, 346 frogs and, 350 Greek Islands and, 391 habitat loss vs., 330 Hawaii and, 209, 398, 400, 711 human, 43, 44–45 index of, 243 insect radiations and, 461, 463–464 introduced/invasive species and, 473, 475, 479–480, 513, 709 lakes and, 166, 169, 529–530 lemurs and, 550 Line Islands and, 555 Madagascar and, 577 mammals and, 588 Marianas and, 594 moa and, 640 New Guinea, 663 oases and, 688 plant disease and, 749 seamount, 820 snails and, 540 Socotra and, 850 speciation and, 292, 527, 585, 773 species-area relationship and, 859 St. Helena and, 871 succession and, 877, 878, 879, 945 traditional practices and, 936 vent biota and, 425 Wallace on, 966 isopods, 390, 399, 570, 848, 849, 907 isostatic processes, 68, 70, 78, 254, 261, 308, 381, 432, 434, 436, 603, 817, 823, 873, 973 Isthmus of Panama, 426, 948 Italy, 151, 188, 237, 392, 423, 536, 543, 544, 623, 624, 625 Iturralde-Vinent, Manuel, 21 ivory, 52 Izanagi plate, 503, 505 Izu, 484, 501, 503, 504, 506, 522, 702, 704, 706 Izu-Bonin arc, 702, 706 jackrabbits, 81 J’Acuse (Zola), 769 jaegers, 811, 813 Jamaica ants, 38 bats, 26 birds, 26, 106 caves, 151 coral reefs, 202 geology, 22, 692 introduced species, 471 land mammals, 24 lizards, 190 missionaries and, 635, 637

Onychophora, 670 reptiles, 27 rodents, 23, 24 size of, 29 James, Helen, 411 James Ross Island group, 17, 19 Jamshid ibn Abd Allah, 985 Jan Mayen, 50, 55, 58, 59, 423 Janzen, D.H., 210 Japan archipelago, 497–506. See also Old World tropics Caroline Islands and, 148, 149 extinct dwarfs, 237 fishes, 874 fishing, 310 formation of, 492 geology, 484, 485, 500–506, 702–703, 705–706, 707 gigantism, 373 introduced species, 471, 474 invasive species, 346 landslides, 535, 536 mountain floras, 114 tsunamis, 934, 9365 weta, 670 whaling, 978 Japanese, 522, 525, 936 Japan Sea, 8 Japan Trench, 501 Jaussen, Tepano, 249 Java biodiversity, 113, 451 Borneo and, 115 climate, 971 Drosophila, 234 geology, 112, 448, 450, 457, 458, 484 hydroclimate, 423 New Guinea and, 660 snakes, 845 tsunamis and, 935 Java Plateau, 306 jaws, 843 jays, 159, 160 Jean de Fuca Ridge, 713–714 jellyfish, 199, 604, 605, 606, 716, 891 Jersey, 154, 155 Jevons, Willaim, 963 João de Castro seamount, 64 Johnson’s Island, 770 Johnston atoll, 680, 684–685 Joinville Island group, 17, 18 jökulhlaups, 64, 433, 434 Jomon people, 522 Jordeson, 963 Juan de Fuca Ridge, 425, 426 Juan de Nova, 739 Juan Fernandez Islands, 507–509, 704, 814 juniper, 78, 128 Jura Island, 116, 124–125 “jutias,” 23–24 Kadavu, 302, 303, 308, 309 kagu, 644 Kaiapit landslide, 536 Kajewski, 940

kakapos, 314 Kalapana earthquake, 695 Kalimantan, 115, 451–452, 457, 458, 459 Kalogrea-Ardana Flysch, 212–213 Kamchatka current, 521 Kaneshiro, K.Y., 233, 826, 827 kangaroos, 331, 373, 374, 453, 655, 959 Kapingamarangi Atoll, 148 Kapiti Island, New Zealand, 184 Karpathos, 237, 391, 626 karst features. See also caves Bermuda, 95, 96 Cook Islands, 192 endemism and, 151 marine lakes and, 603 Mediterranean, 625, 626, 628 Phosphate Islands, 739 Socotra, 847 theory, 69 Zanzibar and, 983 Kaua’i age of, 710 birds, 326, 413 cliffs, 400 coastline, 94 crickets, 208, 211 evolution and, 397 forests, 401 freshwater habitats, 402 geology, 405, 406 hurricanes and, 474 lava tube cave spider, 548 mesic forests, 401 plants, 774, 835, 836, 837 rainfall, 173 snails, 539 volcanoes, 408 kauri, 667, 941 Kavachi volcano, 702 Kayangel, 149 Keeling Atoll, 218, 219 kelps Channel Islands (California), 156, 159 Falklands, 957 Macquarie, 574 organic falls, 701 shipwrecks and, 834 Surtsey Island, 886 Tatoosh, 909, 910, 911, 912 Kerguelen archipelago, 11, 12–13, 15, 16, 440, 471, 475, 789 Kerguelen hotspot, 439, 440, 441 Kerguelen Plateau, 11, 440, 443, 673, 679 Kermadec, 483, 485 Kermadec-Tonga arc, 481, 706–707 kestrels, 618, 830 Ketoy Island, 522 key innovations, 1–2, 169, 252 keys. See cays (keys, quays) keystone species, 15, 36, 479, 535, 701, 748, 910, 912 Kiawah Island, South Carolina, 83, 85, 87 Kick ‘em Jenny, 32, 34, 510–511 Kikihia cicadas, 465 Kili atoll, 682

INDEX

1047

King George Island, 14, 17, 19 King Island, 51 kin selection theory, 225, 226 Kiribati, 380, 641, 896 Kirimati Island, 814 Kiritimati Island, 335, 554, 555, 556, 558, 684, 711, 814 kites (birds), 122, 132 kittiwakes, 811 kiwis, 314, 323, 639, 670, 671 K lauea, 207, 243, 402, 405, 406, 407, 512, 542, 543, 545, 704, 711, 714 Knibb, William, 635 knotweed, 124, 981 Kodiak Island, 423, 483, 492, 635, 874 Kohala, 408 kohu trees, 941 kokako, 472 Kolbeinsey Ridge, 428 Kolguyev Island, 51, 54 Kolombangara Island, 852 Kol’tsevoye lake, 521 Kolumadulu Atoll, 67 Komodo dragons, 373, 452, 513–515, 559, 971 Kon-Tiki, 515–516, 759, 759–760 kookaburra, 908 Koro, 309 Kosrae, FSM, 94 kı-puka, 402, 404, 512–513 Krafla Fires, 434 Krakatau, 206, 439, 451, 454–455, 517–519, 862, 953 Krapf, Johann, 984 Kula plate, 503 Kurile Islands, 481, 483, 484, 501, 520–525, 536, 705, 862 Kuroshio current, 500 Kwajalein, 642, 682, 685 Kyhos, D.W., 837 Kythrea Flysch, 213, 214 Kyushu, 149, 484, 485, 486, 497, 498, 499, 500, 502, 506, 715 Laccadive archipelago, 442, 690 Lack, David, 105, 361 lagoon reefs, 614 lagoons atoll, 219 barrier islands and, 82, 85, 86, 87 Bermuda, 95, 96 Cook-Australs, 709 Cook Islands, 193, 194, 195, 196, 197 Cozumel, 205 French Polynesia, 334, 336, 338 geology, 68, 219, 754 Hawaiia, 631 Line Islands, 557 Maldives, 587 Mascarene Islands, 621 nuclear tests and, 680, 681, 682, 683, 684 pre-European impacts, 417–418 seamounts and, 823 tides and, 915 Tonga, 920 tsunamis and, 934

1048

INDEX

Vanuatu, 940 waves and, 880, 881, 882 Lake Malawi (Africa), 3, 165, 166, 167 Lake, P., 481 lakes. See also specific lakes Britain and Ireland, 124, 125 Comoros, 179 Cook Islands, 194 deglaciated, 4 fishes, 876 Frazer Island, 331 French Polynesia, 335 freshwater species and, 345–346 islands, as, 165, 169, 181, 526–531 Kurile Islands, 521 Marianas, 601 marine, 603–606 Palau, 716 Philippines, 727 rate of succession, 879 Rottnest Island, 797 Solomons snakes and, 853 Lake Tanganyika, 165, 167 Lake Victoria (Africa), 3, 165, 166, 169 Laki eruption, 263–266, 270, 434 Lakshadweep, 442 Lambert, Jonathan, 931 Lambir, Sarawak, 90 L na’i, 94, 374, 408–409, 711, 864 landbirds. See also birds Antarctic, 13 Cook Islands, 194 diversity of, 318 Farallon Islands, 295 Gough Island, 932 Great Barrier Reef, 383 New Zealand, 673 Palau, 716 St. Helena, 871 Tasmania, 905, 907 Tristan da Cunha, 932 Tristan da Cunha/Gough, 932 land-bridge islands Aleutians, 705 Baja California Islands, 80, 81 Borneo, 657–658 lizards and snakes, 557–558, 561, 562 Mediterranean, 628 Pacific region, 707 Philippines, 723, 724, 727 predators, 843 snakes and, 844 land bridges Antilles, 22 Britain and Ireland and, 117, 118 frogs and, 350 human colonization and, 720 Indian Ocean region, 440 Indonesia and, 448 Kurile Islands and, 524 Madagascar and, 776, 778 mammals and, 776 Philippines, 728 Socotra, 847 Sunda shelf islands, 113

Taiwan and, 900 Tasmania, 904, 907 Trinidad and Tobago and, 927 land crabs, 532–535 Lande, Russell, 826 landscape evolution, 695–696 landslides, 535–537 age of islands and, 340 diversification and, 464 French Polynesia and, 341 Hawaiian, 405, 408, 409, 711 Indian Ocean region, 439, 441 Marianas and, 599 Mediterranean, 627–628 Philippines, 732, 737 seamounts and, 823 St. Helena, 871 Taiwan, 904 tsunamis and, 933 land snails, 537–542 land-use practices, 123, 196, 329, 416, 446, 473, 632, 779, 782, 850, 896 languages Borneo’s, 115 Caroline Islands and, 148–149 missionaries and, 634 Oceania, 721–722 Pitcairn, 745 Polynesian, 759 Rapanui, 251 langurs, 867 La Niña, 355, 737 Lanzarote, 65, 127, 128, 129, 131, 132, 134, 135, 140, 141, 423 Lapita, 43–44, 277, 415, 721–722, 741, 758, 941 Lapithos Formation, 212 large igneous provices (LIPs), 11, 441, 703, 823, 950 larks bird disease and, 104 Cape Verde, 144 Channel Islands (California), 159 Larsen, Carl Anton, 977 Lasia, 778 Latin America, 637 latitudinal effects, 14 flightlessness, 317 Maldives, 586, 587 sky islands, 840 Vanuatu, 939 Laupala, 3, 6, 207, 209, 211, 461, 463, 465, 826, 828 Laurasia, 187, 504–505 laurels, 72, 98, 128, 274, 582, 583, 585, 794, 971 Laurentian shield, 56 laurisilva, 72 lava, 542–544. See also kı-puka; volcanism; specific eruptions Cape Verdes, 218 crickets and, 206, 207, 208, 210 Cyprus, 214–215 Fernando de Noronha archipelago, 298 fragmentation and, 329 Galápagos, 370 Mascarene Islands, 612–613

oceanic islands and, 952 spiders and, 864 Surtsey Island, 884, 885 Laval, Honoré, 635 lava pioneer communities, 402, 512, 886 lava tubes, 72, 140, 141, 151, 207, 208, 209, 410, 411, 461, 463, 464, 543, 544–549, 774 Lawlor, Tim, 185 Laysan fever, 104 Laysan Island, 474 Leeuwin Current, 797, 798 Leeward Islands, 22, 143 leeward regions Cape Verde, 146 erosion and, 93 Fiji, 299 freshwater species and, 344 hydrology and, 421 lava and, 543 predators and, 557 Rapanui, 245 reefs, 556 Samoa, 799 Tikehau, 68 Vanuatu, 940 weather, 172 leglessness, 831, 843 legumes, 50, 130 Leibold, M.A., 528 lemmings, 51, 52, 53, 78, 120 Lemnos and Imroz, 395 lemurs, 549–553 adaptive radiation, 2 Comoros, 180 gigantism, 375 Madagascar, 580, 588, 589, 776 maximal radiation, 773 parallel radiation of, 462 Lengguru Fold Belt, 664 leopards, 451, 901, 983 Lepidoptera, 113, 114, 124, 126, 300, 317, 336, 390, 810, 831, 871–872 Lesser Antilles biology, 7, 24, 26, 27, 28 colonization, 278 earthquakes, 32, 35 geochemistry, 31, 32 geology, 22–23, 29, 690, 691, 754, 755 island rule and, 493 tectonism, 30 volcanism, 32–33, 34–35 lesser sheathbills, 12 Lesser Sundas, 448, 452, 970, 971. See also Wallacea Lesvos Island, 391, 393, 395, 627 Levallois people, 117 Levins, Richard, 629 Lewis and Harris, 116, 123–124 liana, 941 lice, 23, 462 lichens Antarctic, 12 arctic, 51, 52 Atlantic, 67 Baffin Island, 78, 78

Canary Islands, 130 Cape Verde, 146 inselbergs and, 467 Macquarie, 574 Madeira, 583–584 Pantepui, 718 Philippines, 724 Surtsey Island, 886 Tatoosh, 909 liestone, Rottnest Island and, 797 light, 200, 202, 884, 911 light houses, 294 lignite, 149 L ’ihi, 404–405 lilies, 899 limestone Cape Verdes, 218 Cyprus, 212, 214, 215 New Guinea, 663 New Zealand, 677 Philippines, 735, 736 phosphates and, 740 reefs, 798 Tonga, 918, 919 western Pacific, 715 Limestone Caribbees, 29, 31 limpets, 425, 910, 911 Lindisfarnre, 125 linear island groups (hotspot island chains), 690, 693, 708 biogeography and, 949 Line Islands, 173–174, 553–558, 632, 680, 711 Linnaen shortfall, 730 Linnaeus, 876 lions, 52, 119, 869 literature, 763 Lithic/Archaic peoples, 278 lithosphere. See also subduction Galápagos and, 710 Hawaii, 407 Indonesia and, 456 New Caledonia, 648 oceanic islands and, 951 overviews, 481, 485, 752–753, 754 Philippines, 734 seamounts and, 822 Solomons and, 856 subsidence and, 954 Lítla Dímun Island, 288 Little Cumbrae, 125 Little Ice Age, 433 Iceland and, 435 Little Swan Island, 23 littoral regions Samoa, 800 liverworts, 12, 940 Canary Islands, 130 livestock Cape Verde, 146 Channel Islands (California), 156, 160 Livingstone, David, 984 Livingston Island, 17, 19 lizards, 3, 557–564 adaptive radiation, 561 Antilles, 27

ants and, 40 Baja California Islands, 80, 81, 82 Bermuda, 96 body size, 913 Borneo, 114 Borneo/New Guinea and, 658 Canary Islands, 131, 132 Cape Verde, 144 Caribbean Anolis, 5, 7, 531 Caribbean convergence, 189 Channel Islands (California), 159, 161 Comoros, 179 convergence, 190 dwarfism and, 236, 238 endangered, 845 extinctions, 846 Fiji, 774 Galápagos, 359, 362, 363 gigantism, 922 Greater Antilles, 531 Indonesia, 452 island rule and, 493, 494 Madagascar, 580 Madeira, 584 Marianas, 594, 596 maximal radiation, 773 monitor, 513–514 New Caledonia, 773 New Guinea, 654, 655, 656, 657 New Zealand, 670 Philippines, 726, 727–728 rafting and, 775 rifting and, 949 Rottnest Island, 798 Socotra, 850 species-area relationship and, 859 spiders and, 860–861 Tonga, 920 Vanuatu, 940, 941 Wallace’s line and, 971 La- na’i, 94, 374, 408–409, 711, 864 lobelioids, 2, 3, 4, 400, 401, 835 lobsters, 907 Cape Verde, 145 Channel Islands (California), 159 coral andanemones, 566 local adaptation, 52, 224, 226, 353, 355, 594, 911 local extinctions. See also relaxation Channel Islands (California) eagles, 158 climate change and, 842 continental islands and, 183 defined, 281 dispersal and, 225, 226 ecological sorting and, 189, 531 endemism and, 814 human impacts and, 814, 892 island effect and, 928–929 isolation and, 282, 285, 466 metapopulations and, 630 oases and, 688 overviews, 527 pathogens, of, 749 Polynesians and, 892 skinks, 794 Tonga, 920

INDEX

1049

local participation, 929–930. See also traditional practices locusts, 206 Lofoten Islands, 59 logging, 115, 332, 769, 941. See also timber Lohachara Island, 380 Lo- ’ihi, 404, 408, 693, 714, 804, 864 Lo- ’ihi Seamount, 714 Lombok, 447 Lomolino, Mark, 493 Lomolino, Mark V., 859 Lomolino’s Combined Model, 859, 860 longevity, dispersal and, 226 Long Island, 345, 755 fishes, 874 longitudinal effects, 939–940 Look Seamount, 713 loons, 811 Lophlia, 564–567 lorakeets, 908 Lord Howe Island, 210, 568–572 endangered species, 285 insects, 786 pigs and goats and, 743 rats and, 795 waves and, 484, 882 lorikeets, 98, 194, 303, 305, 335, 337, 707 Losos, Johathan, et al., 7, 859 Losos, Johathan B., 561 Louisville Seamounts, 714 low reef islands, 386, 387, 388 Loyalty Islands, 491, 643, 646, 647, 648, 692, 694, 722, 939 Lunday Island, 255 Lusitania, 118 Luzon arc, 902 Lyakhovsky Island, 51 Lydekker’s Line, 447, 968 Lyell, Charles, 218, 219, 220, 221, 964 Lyme disease, 104 lynx, 120 Lyon, H. L., 946 lyrebirds, 908 Ma’anyan speakers, 115 macaques, 775–776, 776–778, 899 Macaronesian Islands biogeography, 72–73, 127 endemic species, 130 fauna, 74 geology, 64–65 lizards, 560 orchids, 698 plants, 129 radiation zone, 774 MacArthur, Robert H., 271, 629, 772, 774. See also Equilibrium Theory of Island Biogeography Macdonald, Gordon, 342 Macdonald hotspot, 343 Macdonald seamount, 342 Macdonald volcano, 709 MacKenzie, Neil, 635 MacPhee, Ross, 21

1050

INDEX

Macquarie Island, 11, 15, 16, 573–577, 908, 950 macroalgae, 701 macroevolutionary factors, 772 macropods, 453 Madagascar, 577–582. See also Gondwana adaptive radiation, 578–579 age of, 319 ants, 36, 37, 38, 39 birds, 106, 319, 321 chameleons, 559 climate change and, 170 Comoros and, 178, 179, 180 deforestation and, 170 drift, 177 dwarfs, 237 freshwater species, 345 frogs, 7, 348, 349 geology, 445, 446, 755 gigantism, 374 Gondwana and, 669 humans, 115, 720 inselbergs, 469 island rule and, 493 lemurs, 773 lizards, 559, 560 mammals, 2, 322, 549–550, 588, 590 monocotyledons, 467 orchids, 698 overviews, 183 rodents, 792, 795 snakes, 179, 559 South American dispersals, 560 spiders, 862, 863, 864–865 succulents, 468 tides, 915 topography, 551 tortoises, 922, 926 tsunamis and, 935 Madagascar plateau, 444 madder, 157 Madeira archipelago, 582–585 Cape Verde and, 146 conservation, 567 coral conservation, 567 fauna, 72, 73, 106, 129, 130, 559, 862, 864 flightlessness and, 311–312, 317 flora, 145, 146 geology, 64, 127, 141, 142 map, 133 pigs and goats and, 742 research stations, 789 spiders, 862, 864 Madrean Archipelago, 235, 839–842, 840, 841, 841–842 Mafia Island, 179 Magellan, Ferdinand, 280, 595, 702, 703 magma corners, 485 eruptions and, 951–952 Galápagos, 371 Hawaii and, 405 island arcs and, 481 melting, 951

overviews, 543, 691, 753, 754 Philippines and, 736 seamounts and, 823, 825 subduction and, 822 water and, 953 magmatic rocks, 19 magnetic fields, 754, 955 magpies, 899, 900 mahogany, 157, 944 Maiao, 333 mainland effects, 123 Maio Island, 65, 143, 144, 145, 147 Majuro, RMI, 94 Makah Nation, 909 Makatea, 333, 334, 336, 339, 739 makatea islands, 192, 193, 196, 333, 335, 585–586 Makira, 855 Makira Trench, 856 Malagasy songbirds, 106 Malaita, 855 malaria, 103, 104, 414 Malay archipelago, 446, 964, 965 The Malay Archipelago (Wallace), 446, 965 Malay/Muslim peoples, 115, 276 Malaysia, 112, 113, 114, 115, 208, 310, 347, 349. See also Sunda Malden, 680, 684, 738, 739 Maldive Islands, 586–587 climate change and, 171 crickets, 208, 209 geology, 442, 613 hydrology and, 423 photo, 67 plant disease, 748 wind and, 92–93 Male islet, 423 Malesians, 713 Malesian subkingdom, 447 mallards, 472, 572 Mallorca, 321, 559, 623, 624, 845 Malta, 629 Malthus, 962 mammals, 588–593 Antilles, 21, 22, 23, 24, 26 ants and, 40 arctic, 51, 52, 53, 53–54 Asia, 121 Baffin Island, 76 Baja California Islands, 80, 81 Barro Colorado, 90 Bermuda, 98 bird fossils and, 320–326 boreal forests fragments, 185–186 Borneo/New Guinea, 114, 658 Britain and Ireland, 119–120, 124 Canary Islands, 129, 131 Cape Verde, 144, 145 Channel Islands (California), 155, 159–160 convergence and, 188 Cook Islands, 194–195 dwarfism/gigantism, 236, 238, 375 extinctions, 282, 330, 500, 845–846 Fiji, 305

food and, 843 Great Barrier Reef, 383 Hawaii, 470 Indonesia, 447, 449, 450, 451, 452, 453, 513 island rule and, 493 Japanese, 499 kiwis and, 670 Komodo dragons and, 513, 514 Kurile Islands, 525 Macquarie, 574, 575 Madagascar, 580 Marianas, 595 Mascarenes, 618 Mediterranean, 629 New Caledonia, 644 New Guinea, 653, 654, 655 New Zealand, 99–100, 472, 667, 670, 671, 672, 673 overviews, 112, 184, 793 Pantepui, 719 Philippines, 724 rafting and, 775–776 Samoa, 800, 801 seamounts and, 820 silverswords and, 838 sky islands, 840–841, 842 snakes and, 844 Socotra, 849 Solomons, 774, 853 species-area relationship and, 857, 860 Spitsbergen, 866 Sri Lanka, 869 sustainability and, 891 Taiwan, 899 Tasmania, 907 Tonga, 920 Vancouver, 938 Wizard Island, 869, 981 mammoths, 51, 52, 120, 158, 160, 237, 707 Mamonia Terrane, 214 Mana Island, 794 Mancham, James, 832 Mandela, Nelson, 770 Mangaia Island, 44, 192, 193, 194, 196, 342, 417 Mangareva, 45–46, 250, 417, 635 Mangareva people, 709 mangrove islands, 591–593 mangroves Bermuda, 95, 96, 97 Borneo, 114 Comoros, 178 crickets and, 208 dispersals, 367, 368 erosion and, 261–262 ethnobotony and, 274–275 Frazer Island, 331 Gulf of Guinea, 810 Hawaiian, 943 Madagascar, 579 Marshall Islands, 611 New Caledonia, 644 New Guinea, 657 sea levels and, 377–378 Solomons, 851, 852

South Atlantic, 298 Tonga, 920 Trinidad and Tobago, 928 Vanuatu, 940 Zanzibar, 983, 984 Manihiki Plateau, 191, 195 mantle. See also intrusive rocks Cook Islands and, 709 hotspots and, 754 Newfoundland, 651 New Zealand and, 675, 679 North Atlantic and, 652 overviews, 752 Samoa and, 808 seamounts and, 825 Solomons and, 855 volcanic islands and, 951 mantle plumes anomalies and, 951 Antarctic, 17, 20 Antilles, 32 Arctic, 63, 64 Atlantic, 63, 64, 66, 67 Canary Islands, 135, 142 Cape Verde, 144 Cook Islands and, 709 French Polynesia, 340 Galápagos, 367, 368–369, 370, 371 Hawaiian, 406, 407, 711, 714 Iceland and, 428 Indian Ocean region, 439, 440, 445 intraplate islands and, 708 Laurasia and, 504–505 line islands and, 553 Madeira and, 582 Marquesas, 342 Mascarene Islands, 622 moving, 714–715 overviews, 481, 544, 704, 754 Pacific region, 704 Samoa and, 807 seamounts and, 821, 824 Solomons, 856 mantle rocks, 66 Manuae Island, 192, 194, 195, 341 manuring, 14 Manus basin, 664 Manx breeds, 125 mao (ma’oma’o), 800 Maore Island, 177 Maori people, 273, 470, 672, 707, 795, 958 Map Island, 149 Maratua, 777, 778 marble, 393, 394, 395, 736 Maré, 694 Mare-aux-Songes marsh, 230–231 Margarita, 23 Maria, 834 Maria Cleofas Island, 79 Maria Madre Island, 79 Maria Magdalena Island, 79 Mariana arc, 502, 691, 706 Marianas, 148, 483, 484, 485, 593–603, 705 Mariana Trench, 149, 754

Marie Galante, 24, 31 marine lakes, 603–606 marine life and environments. See also specific types of marine life archaelogy and, 43, 44 arctic, 54, 57 Baja California Islands, 78, 81 Bermuda, 96 Borneo/New Guinea and, 658 Britain and Ireland and, 125 Canary Islands, 131–132, 140 Cape Verde, 145, 146 Channel Islands (California), 155, 159–160, 161 Cook Islands, 194, 195 eastern African, 983 endangered, 893 Faroe (Faeroe) Island, 288 French Polynesia, 336, 338 freshwater species and, 346 Galápagos, 368, 710 Galápagos endemism, 359–360 global warming and, 378 Gough Island, 930 Gulf of Guinea, 809 Japan, 500 Kurile Islands, 525 Lord Howe Island, 569, 571 Macquarie, 574, 575 Madeira, 584 mangroves and, 592–593 Marianas, 594 marine lakes and, 606 Mascarene Islands, 615 New Guinea, 654 organic falls and, 700–701 Palau, 716 Pitcairn, 745 pre-European impacts, 417 rafting and, 775 Rapanui, 245 Rottnest Island, 798 Samoa, 801 seamounts, 713 Seychelles, 832 shipwrecks and, 834 South American, 69 speciation of, 756 St. Helena, 870 Surtsey Island, 886 sustainability and, 896 Tonga, 920 Tristan da Cunha/Gough, 932 Vancouver, 938 Zanzibar, 983–984, 986 marine protected areas (MPAs), 98, 196, 403, 427, 567, 607–609, 716, 780, 783, 895, 896 marine reserves, Socotra, 847 Marion Island, 13, 14, 15, 16, 444, 471, 794 Marion mantle plume, 445 Maritime Continent, 172, 173. See also Indonesia Marlborough Sounds, 675

INDEX

1051

marmots, 938 Marotiri, 342 Marquesa Fracture Zone, 339, 340, 341, 342, 712 Marquesan people, 195, 709 Marquesas deforestation, 222 Drosophila, 234 endemism, 335, 336 freshwater species, 345 geology, 333, 341–342, 696, 711–712 grazing and, 223 landscapes, 334 marine life, 336 missionaries and, 636 snails, 538, 539, 540, 541 tsunamis and, 934 volcanism, 755, 953 Marshall Islands, 610–612 conservation, 896 motu, 641, 642 overviews, 69, 148, 680, 681, 712 Marshall, John, 712 Marshall, Patrick, 951 marshes, 97, 194, 335 marsupials, 43, 188, 453, 655, 796, 798, 905, 906, 908, 958 martens, 120 Martin Garcia, 770 Martinique, 27, 28, 31, 32, 35, 108–109, 564 Martinique Passage, 23 Martín Vaz Island, 65, 66 Mascarene Plateau, 612, 613, 620, 622 Mascarenes biogeography and, 107 biology, 612–619 birds, 179, 319, 320, 322 extinctions, 170, 922 flightless birds and, 316 geology, 442–443, 620–622 introduced species, 472 lizards, 560, 561 orchids, 698, 699 snakes, 845 tortoises, 922, 925 weevils, 2 Masibalavu, Vilikesa, 302 mass effects, 527, 528, 530 mass extinction event, 612 Massif de la Hotte, 24, 27 massifs, 182 mass wasting, 134, 953–954. See also landslides Matai people, 783 material plants, 273–274 mating behaviors, 233, 413, 656. See also specific organism mating signals, 108, 110, 465, 548. See also acoustical signaling matrix contrast, 181 matrix-derived taxa, 182 Maug, 595, 598, 599, 600 Maui birds, 413 climate, 397, 398 climate change and, 171

1052

INDEX

coastline, 94 crickets, 211 dry forests, 401 forests, 944–945 silverswords, 835 snails, 540 tides and, 916 volcanoes, 408 Maui Nui islands, 408 Mauke Island, 192, 193, 194, 196, 342 Mauna Kea, 134, 172, 398, 402, 405, 406, 407, 409, 542, 754, 803, 821 Mauna Loa, 134, 172, 173, 405, 406, 407, 408, 409, 542, 547, 620, 952 Maupihaa, 341 Maupiti, 333, 341 Mauritius. See also dodos ants, 37, 39 beetles, 2 biodiversity, 617 biogeography and, 107 bird divergence, 108–109 conservation, 618 Darwin and, 218, 958 ecology, 614 geology, 442, 443–444, 612, 613, 621–622 gigantism, 373 introduced species, 471 orchids, 698, 699 phylogeography, 616 plants, 848 prisons, 771 Maya people, 203 mayflies, Fijian, 300 Mayotte Island, 177, 178, 180, 560 Mayr, 106, 826 Mayr, Ernst, 327, 968 McCauley, D.E., 766 McDonald Islands, 11, 16, 441 McMurdo Dry Valleys, Antarctica, 10, 13 meadows, 72, 78, 471, 524, 652, 718, 798, 839, 840, 887, 889, 938, 981. See also grasses mealybugs, 39 Mecherchar Island, 149, 604, 738, 739 Medellín, R.A., 152 Mederes, Adnan, 771 medicines, 90, 91, 146, 275, 781, 810, 848 Medioeuropean region, 127 Mediterranean plate, 392 Mediterranean region, 622–629 birds, 321 Canary Islands and, 131 Cape Verde and, 145, 146 dwarfs, 237 extinct dwarfs, 236–237 extinctions, 170 island arcs, 482 lizards, 564 Macaronesia and, 127 mammals, 321 pigs and goats and, 741 plants, 128, 129 popular culture and, 764 prisons, 771 seafaring and, 277–278

snakes, 844, 845 spiders, 863 Mediterranean Sea, 83 Medusagynaceae, 830 megafauna, 581 megaherbs, 574, 671 megapodes, 304, 314, 325, 374, 416, 716, 920 Mehetia, 333, 341, 712 melaleuca, 944 Melanesia ants, 38, 253, 912 bird radiation, 106 conservation, 892, 893 diversity, 800 ethnobotony and, 272, 274, 275 frogs, 348, 349 humans and, 114, 277 missionaries and, 634 overviews, 42, 722, 890, 891 snails, 537, 538 sustainability and, 892 traditional practices, 890 Melanesian arc, 300, 306, 309 Melanesians, 180, 634, 637, 660, 832, 853, 855 melting glaciers, 816 melting of mantle, 951, 953 Melville, Herman, 977 memertean worms, 819 Mendelson, T.C., 209 Mendocino Fracture Zone, 425 Menorca, 321, 559, 624, 845 Mentawai Islands, 446, 451, 491, 692, 777, 778 Mercalli scale, 241 Mesaoria Plain, 214–215 mesas, 182, 601, 646, 839 mesic forests, 335, 401, 836 Mesoamerica, 182, 186, 205 Mesopotamia, 121 mesosphere, 752 Messinian salinity crisis, 624, 625 metacommunities, 527–531 metamorphic rocks Antarctic, 18 Antilles, 31 arctic, 55, 56, 57 Channel Islands (British), 154 Channel Islands (California), 162 Darwin and, 957 Greek Islands, 393, 394, 395 Indian region, 445 Indonesia, 456, 457 island arcs and, 481, 484 Japan, 501, 504, 505, 506 Mediterranean region, 624, 625, 626, 627 New Caledonia, 646, 647–648 Newfoundland, 649 New Guinea, 661, 662, 663, 664, 665 New Zealand, 675, 677 Philippines, 732, 733, 734, 735, 736 Solomons, 855, 856 Taiwan, 902–903 Warming Island, 972 metapopulations, 227, 527, 629–631, 688, 948 methane, 175–176 Metrosideros, 940, 949

Mexican Caribbean Sea, 203 Mexico, 38, 168, 544, 593, 840, 841 mice Antarctic, 15 Baja California Islands, 79, 81 biodiversity and, 794 Britain and Ireland and, 120 Channel Islands (California), 158, 161 endemism and, 255 extinctions, 283 Farallon Islands, 295 Gough Island, 932, 932 Philippines, 589, 727–728 rats and, 474 sub-Antarctic, 15 wide-spread species, 792 Wizard Island, 981 Michel, James, 832 microallopatric speciation, 166 microbes, 11–12, 89 microevolution, 108, 538, 563 Micronesia climate change and, 171 human colonization, 722 humans and, 42 motu, 641 pigs and, 741 pigs and goats and, 742 shipwrecks and, 834 snails, 538 sustainability and, 896 Micronesians, 712 microorganisms, 54, 62, 118, 174 Mid-Atlantic Ridge, 58, 63, 67, 426, 428, 435, 490, 565, 822, 956 Middle East, 121, 482 mid-domain effect, 698 midges, 12, 13, 14–15, 23, 887, 907 mid-ocean ridges. See also specific ridges Antarctic islands and, 18, 19 CBCs and, 175 Galápagos and, 368–369 hotspot volcanoes and, 755 melting anomalies, 951 overviews, 690–691, 753–754 seamounts and, 714, 822, 825 mid-Pacific continent, 538 Midway Island, 104, 105, 206, 631–633, 710, 793 migmatites, 56 migrants. See also dispersal; human colonizers birds, 26, 50, 78, 122, 124, 125, 126 Canary Islands, 132 Cape Verde, 144, 147 cave, 152–153 Christmas Island crab, 533 Comoros, 179 conservation and, 630 endemism and, 254 Farallon Islands, 295 fishes, 311 freshwater species, 344–345 gene flow and, 766, 767 Greek Islands, 391 jellyfish, 605

Kurile Islands and, 522 mangroves and, 593 Marshall Islands, 611 metapopulations and, 629 New Zealand, 669 Samoan, 801 seabirds, 813 Surtsey Island, 887 Tonga, 920 vertical, 756, 757 Milankovitch Theory, 376 millipedes, 300 mimicry, 6, 258, 655 mineral deposits. See also mining; specific minerals Fiji, 308 Indonesia, 457, 460 lava tubes, 546 Mediterranean, 625, 627 New Caledonia, 647 Philippines, 734, 736 seamounts and, 825 subduction and, 483 tides and, 917 mining. See also mineral deposits; Phosphate Islands arctic islands, 54 Baffin Island, 77 Borneo, 112, 115 Caroline Islands, 149 Christmas Island, 534 Cook-Australs, 709 Fiji and, 309 Frazer Island, 332 Greek Islands, 395 Madagascar, 581–582 New Caledonia, 645, 647 New Guinea, 665 Philippines, 730 seamounts and, 820 Solomons, 856 Spitsbergen, 866 Tasmania and, 908 vent fauna and, 427 minks, 124, 525 Minoan eruption, 394 Miogeoclines, 651 mires, 14 Mischief, 963 Misima Island, 665 missionaries, 633–638, 742, 958, 984 mist forest, 810 mites, 12, 13, 53, 100, 130, 887 Mitiaro Island, 192, 193, 194 Mittermeier, Russell, 449 mixed forests, 905 Mnaihiki, 195 moa-nalos, 324, 326 Moana, Vaka, 760 moas, 106, 184, 315, 319, 322, 323, 324, 325, 373, 416, 638–641, 667, 670, 671, 672 mockingbirds, 80, 359, 950 Moheli Island, 177, 178, 179, 180 Mohotane Island, 223 Mohotani, 337

Mojave Desert taxa, 81 molds, 44, 479, 885, 886 molecular research, 7, 8 moles, 899 molluscs Azores, 72, 73 Bermuda, 98 Canary Islands, 130, 132 Cape Verde, 145, 146 conservation, 540–541 Fijian, 299–300 French Polynesia, 336–337 Greek Islands and, 393 Madeira, 584 Seychelles, 831 Socotra, 848–849 Solomons, 852, 853 sustainability and, 891 Vanuatu, 940 vent, 426 Moloka’i, 46–47, 94, 400, 408–409, 540, 635, 864, 915 Moluccas, 448, 453, 459, 968, 971. See also Wallacea mongooses, 23, 24, 25, 26, 179, 305, 471, 473 monkeys, 21, 25, 88, 90, 113, 144, 236, 778, 983 monocotyledons, 467, 468, 469 monocultures, 99 monogamy, 413 monophylism, 1, 26, 52, 73, 165, 234, 411, 463, 464, 513, 561, 579, 639, 722, 773, 837, 921 monotremes, 655, 656 Monserrat, 80 monsoons Australasian and African, 172 beaches and, 92 Borneo, 112 Cape Verde, 144 climate change and, 174 Comoros, 178 forests, 452 Kurile Islands, 521 latitudinal effects, 587 New Guinea, 657 Socotra, 848 Sri Lanka, 868 Taiwan, 898 Zanzibar, 984 Montagne Pelée, Martinique, 32, 35 Monteray shale, 163 Monterey Bay, 176 Montserrat, 23, 24, 32, 35, 543, 544 Morgan, W. Jason, 704, 711 Morocco, 104, 130 morphological evolution, 211, 233–234, 353–354, 355–356, 563–564, 923 Morrison, 859 mortality rates, 227–228 Moruruo, 336, 341 Moslow, Tom, 87 mosquitoes, 88, 103, 104, 105, 283, 472, 711, 977 mosses Antarctic, 12 arctic, 50

INDEX

1053

mosses (continued) Atlantic, 67 Bermuda, 98 Gough Island, 931 Lord Howe Island, 571 Surtsey Island, 886 sustainability and, 891 Tristan da Cunha, 931 Mother Lode, 152 moths. See also caterpillars Antarctic, 15 Antilles, 23, 28 arctic, 49, 51, 53 biocontrol (Fiji) and, 101–102 Borneo, 114 Britain and Ireland and, 123 cave, 548 Channel Islands (California), 159 Fiji, 101–102, 300 French Polynesia, 336 Galápagos Galagete, 362–363 Hawaii, 398, 400, 548 Japan, 336 Madeira, 584 Midway, 632 New Guinea, 656 New Zealand, 318 orchids and, 697 St. Helena, 871–872 Surtsey Island, 886 Wallace’s Line and, 970 motu, 67, 68, 641–642, 694 Motu One, 333, 337, 341 mountains. See also elevation; sky islands arctic, 61 Borneo’s, 113, 114 breezes, 173 Britain and Ireland, 116, 125 Canary Islands and, 127 climate and, 172 Cook Islands, 193 Cyprus geology, 212 endemism and, 257 Guayana, 717 Indonesia, 460 Japan, 501 New Zealand, 666–667 North Atlantic island, 287 Solomons, 852 Tahiti, 334 Taiwan, 897, 899, 902–903 Tasmania, 9054 Vancouver, 937–938 Vanuatu, 940 Mount Cameroon, 808 Mount Cook, 707–708 Mount Desert Island, 380 Mount Etna, 220, 549, 625 Mount Fuji, 706 Mount Kinabalu, 112, 113, 114 Mount Mazama eruption, 979 Mount Pagan, 600–601 Mount St. Helens, 543, 862 mourning doves, 80 movies, 762–763

1054

INDEX

Moynihan, Martin, 88 Mozambique Channel, 177 Mt. Chubu, 786 Mt. Erebus, 19–20 Mt. Etna, 220 Mt. Haddington, 19 Mt. Siple, 20 Mt. Vesuvius, 220 Muck Island, 124 mulberry, 274, 275 Mull, 116, 124, 125 Müllerian mimicry, 6, 655 mullets, 124 Mull Island, 124–125 multiple colonization, 477, 478 murrelets, 158, 161, 813, 938 murres, 813 Baffin Island, 78 Farallon Islands, 294, 296–297 Mururoa atoll, 69, 680, 685, 744 music, 764 musk oxen, 51 mussels, 425, 426, 529, 909, 911, 974 mustelids, 672 mutations, 6, 103, 125, 151, 225, 238, 355, 437, 563, 697, 698, 755, 828, 837 mutualism, 89, 91, 189, 519, 535, 619, 655, 920 Mwali Island, 177 My Life (Wallace), 966 Myojinsho Island, 702–703 Myrdalsjökull, 435 myxoma virus, 121–122 The Naked Island (Shindo), 423 Nankai Trough, 501, 502 Nansen-Gakkel Ridge, 58 Nantucket, 976 Nason, John, 89 national parks. See also marine protected areas (MPAs); reserves, nature Antilles, 29 Borneo, 115 Canary Islands, 129, 132 Channel Islands (California), 161 Christmas Island, 534 Cook Islands, 197 Egypt, 591 Fraser Island, 330, 332 Galápagos, 365, 366, 926 Greek Islands, 391 Komodo, 514 Mascarene Islands, 618 Midway Island, 632 Newfoundland, 649, 651 prisons, 769, 771 Samoa, 801 Sri Lanka, 867 St. Helena, 873 Taiwan, 898, 901–902, 902 Tierra del Fuego, 917, 918 Tonga, 919, 920 Trinidad and Tobago, 929 U.S., 979, 980 Vancouver, 937, 938 Zanzibar, 985

Native Americans, 909 natural enemies. See predation natural gas, 112, 627 natural selection. See also fitness divergence and, 108 drift and, 327 Mascarene Islands, 617 plasticity and, 5, 6 speciation and, 166 Wallace on, 220, 964, 965, 966 nature reserves. See reserves, nature Nauman, Edmund, 501 Nauru Island, 585, 738, 738, 739, 739 Navidad Bank, 22 navigation, 204, 279–281, 280, 281, 759, 760 Nazca plate, 357, 367–368, 753 Ndzuani Island, 177 Near Oceania, 720 needlefish, 783 Negrito populations, 114 Neilson bank, 342 nematocysts, 199 nematodes, 10, 12, 14, 748, 788, 795, 819 Nenets, 54 neoendemics, 73, 74, 183, 184, 185, 253, 255, 256, 524 neoteny, 548 neotropical regions, 7, 131, 234, 467, 468, 508, 560, 718, 786, 801, 925 nesting Line Islands, 556 mangroves and, 593 New Zealand, 669 seabirds and, 813–814 snakes and, 843, 844 stickleback, 875–876 Surtsey Island, 887 turtle, 801, 929 Netherlands Antilles, 23 neuston, 258 neutral view, 528 Nevis Island, 23, 34 New Amsterdam archipelago, 11 New Britain, 485 New Caledonia, 643–648. See also Gondwana age of, 319 biota, 324 birds, 303, 304, 322 crickets, 208 endemism and, 254 forests, 645 freshwater species and, 345 geology, 492, 644, 666, 679, 692, 707 Gondwana and, 669 humans and, 42 lizards, 773 Lord Howe Island and, 571 missionaries and, 634 overviews, 183 paleoendemics, 255 pre-European impacts, 416, 417 prisons, 771 research stations, 789 seamounts, 819 skinks, 560

snails, 538 spiders, 862, 863 Newcastle disease, 105 Newfoundland, 649–652, 874 New Guinea ants, 36 bananas, 272 biology, 652–659 bird radiation, 106 Borneo biota and, 114 climate, 971 climate change and, 170 coral reefs, 200 diversity, 450 ecoregions, 448 extinctions, 170 frogs, 348, 349, 350–351 geology, 447, 448, 460, 656, 659–665, 708 human colonization, 720, 721 humans and, 42, 43 insects, 571 mammals and, 588 missionaries and, 634 mountain floras, 114 overviews, 453 pigs, 741 pitohui birds, 5 seafaring and, 277 shape and climate, 172 size of, 111 snails, 538 spiders, 862, 862 Wallace and, 968 New Guinea Trench, 664–665 New Hebrides, 306, 309, 481, 482, 485, 648 New Ireland, 485, 486, 720, 949 new islands, 228 New, T.R., 206 newts, 122 New World, 7, 40, 52 New Zealand. See also Gondwana age of, 319 alien species and, 99–100 architecture, 47 avian diseases, 104 biocontrol and, 102 biology, 665–673 biota, 322–325 bird disease and, 105 birds, 106, 304, 314, 322, 323–324, 638–641 conservation, 474 Cook Islands and, 195 Darwin and, 958 deforestation and, 170, 221 diversity and, 320 ethnobotony and, 272 Fijian skinks from, 302 fishing, 310 flightlessness and, 314, 317–318 frogs, 348 geology, 647, 665–666, 669, 673–680, 703, 707–708, 754 gigantism, 373, 374, 375 global warming and, 380 humans and, 42, 707–708, 720, 722

insects, 465 introduced species, 470, 471, 472, 477, 959 invasive species, 474, 480 island rule and, 492, 494 landslides and, 535 lizards, 559 Lord Howe Island and, 571 mammals, 322–323, 672 mice, 793 millipedes, 300 overviews, 183 pigs and goats, 741, 742, 743 plant disease, 748 Polynesians and, 758 pre-European impacts, 416 rats, 792, 793, 794, 795 settlement of, 41 spiders, 863 sub-Antarctic islands, 11, 16 threatened species, 285 tuatara, 786 weed costs, 99 whaling, 976, 977 Ngeaur Island, 149, 739 niaouli tree, 644, 645 Nias, 483 Niau Island, 333, 337, 642 niches competition and, 106 divergence and, 110 empty, 233 extinction and, 285 novel, 252 speciation and, 699 nickel, 647 Nicobar Islands, 439, 777, 778 Nightingale Island, 66, 108 Nihoa Island, 208, 211, 375, 411, 412, 862 Ni’ihau Island, 208 Ni’ihau shield, 406 Ninety East Ridge, 439–440 Nisyros, 396 nitrogen, 13, 14, 470–471, 472, 475, 801, 877, 887, 888 Niue (Tonga), 692 Noble, G.K., 348 noddies, 956 non-adaptive radiations, 1, 461, 561 nonequilibrium, 489 non-native species. See alien species nonradiating taxa, 774 nonscientific projects, 102 nontarget effects, 102 non-volcanic arcs, 691–692 Nordvestfjord, Greenland, 56 Norfolk Island, 210, 571, 572, 768 Normandy, 155 Normans, 121 North Aegean fault zone, 392 North Africa, 627, 686 North America. See also specific locations Antilles and, 21, 25 arctic and, 51 barrier islands, 83, 85 Bermuda and, 96

birds, 26 Britain and Ireland and, 117, 126 extinctions, 813 fish, 28 fishes, 874, 875 geology, 504, 694, 754 glacial-interglacial cycles, 785 human colonization, 722–723 introduced species, 473 island formation, 973 plant disease, 749 Polynesian voyaging and, 760 prisons, 770 raccoons, 26 silverswords and, 837 sky islands, 839, 840 species-area relationship and, 857 tectonism, 56, 57 volcanism and, 58 North American Plate. See also Okhotsk plate Antilles and, 22, 30 Azores and, 64, 71 Baja California and, 78 Bermuda and, 95 Channel Islands (California) and, 161, 164 Iceland and, 429 Japan and, 501 Kuriles and, 522 overviews, 482, 705, 753, 754 seamounts and, 824 Vancouver/Wrangellia and, 708, 937 vents, 425, 426 North Andros Island, 211 North Atlantic region. See also Iapetus Ocean coral, 565 plate system, 429 seabirds, 814, 814 tectonics, 428, 651 volcanism, 755 whales, 975, 976, 977 North Carolina coast, 87, 565 northeast Pacific region, 705, 713 North Equatorial Current, 172 northern hemisphere, 14, 977 Northern Ocean Archipelagoes, 814 North Moluccas, 459 North Pacific, 55, 158, 418, 419, 713, 813, 814, 814, 874, 975, 976, 977 North Pole, 59 North Sea, 117, 258, 266, 287–288, 289, 566, 567 North Slope, Alaska, 58, 85–86 Norway, 54, 289, 292, 423, 565, 567, 866, 879, 933, 975 Norway rats, 471, 472, 792, 792, 793, 794, 795, 801, 834, 850 Norwegian-Greenland Sea, 55, 58 Nostostomatidae, 670 Notogean Realm, 446, 447, 448 de Nova, João, 871 Novaya Zemlya, 47, 48, 51, 52, 54, 55, 57, 59, 60 novel environments, 1, 2–4 novel niches, 252 Novosibirskiye Ostrova Islands, 48, 51, 59

INDEX

1055

nuclear weapons, 336, 341, 612, 680–685, 706, 708, 711, 712 Nukuoro Atoll, 148 nunataks, 10, 12, 13, 17, 19, 48, 433, 786 Nunn, Patrick, 248 nutria, 474 nutrient availability caves and, 547, 548 dispersal and, 531 diversity and, 13 edge effects, 787 introduced species and, 471, 513 island rule and, 494 rats and, 795 seabirds and, 812 succession and, 877–878 waves and, 882 nutrients. See food (nutrients) Nuu-chah-nulth people, 938 Nyerere, Julius, 985 O’ahu adaptive radiation and, 412 beaches, 92 climate, 172 coastline, 94 geology, 406, 951 human colonization, 722 insects, 317 introduced species, 471 landslides, 536 plants, 3 silverswords and, 837 snails, 539, 540, 541 waves and, 882 oak forests, 157, 182, 184, 186, 389, 525, 628, 793, 841, 938 oak-laurel forests, 971 oaks, 129 oases, 174, 686–689 Oaxaca Valley, 184, 186, 187 obsidian, 43, 46, 247, 277, 278, 394, 625, 626, 629, 720, 721, 745 Ocalan, Abdulla, 771 ocean circulation, 172, 288. See also Gulf Stream ocean floor, 174, 175, 176, 177, 576, 933. See also organic falls; vents; whale falls Oceania, 41–43, 313–314, 720. See also Lapita oceanic islands, 689–696. See also de novo formation age of, 319 anagenesis and, 9 basalts, 951 biogeography and, 948 Comoros, 178 diversity and, 10 dwarfs, 238 endemism, 129, 254, 257, 870 erosion and, 954 flightlessness and, 312 frogs and, 350 geology, 951 global warming and, 376–377 insects, 130

1056

INDEX

introduced species and, 473 overviews, 490 seamounts and, 821 snakes, 843, 844 speciation on, 8 spiders, 862 subsidence and, 954 Taiwan, 897 volcanism, 950–951, 952, 953 Wallace on, 965 oceans. See acidification, ocean; specific oceans octopuses, 145, 160, 176, 635, 781 Oeno Atoll, 45 Ogasawara Islands, 234, 498, 499, 504, 538 Ogilvie-Grant, W.R., 848 ohia trees, 402 oil. See fossil fuels oil palm, 115 Okhotsk cultures, 522 Okhotsk plate, 506, 522 Okinawa, 474 Okinawa Trough, 502 Old Crow Flats, 526 old growth, 88, 901, 938. See also timber Old World flora and fauna, 52 Old World tropics, 7 olfactory cues, 210, 235, 670, 843, 924 olive trees, 874 olivewood bark, 95 Olson, Storrs, 411 One Tree Island, 385–386 “On the Law Which Has Regulated the Introduction of Species” (Wallace), 964 On the Origin of Species (Darwin), 710, 964 “On the Tendency of Varieties to Depart Indefinitely from the Original Type” (Wallace), 964 On the Various Contrivances by which British and Foreign Orchids are Fertilised by Insects, and on the Good Effects of Intercrossing (Darwin), 696 ontogeny, 9–10, 239, 515 Ontong Java Plateau, 306, 659, 660, 664, 703–704, 706, 825, 854, 855, 856 Onychophora, 670 oomycetes, 748 oopu, 402 ophiolites, 113 Cyprus, 214, 215 New Caledonia, 646, 647 Newfoundland, 649 New Guinea, 663, 664 Philippines, 734, 735, 736 ophiuroids, 819 opportunists, 700 Öræfajökull Zone, 429, 433, 434 orangutans, 113 orbit, Earth’s, 376 orchids, 467, 469, 571, 653, 696–699 Oregon, 694 organic falls, 700–701. See also whale falls Orient, 114 Origin of Species (Darwin), 2–3 Orkney Islands, 116, 120, 123, 292 ornamental plants, 275

ornithosis, 104 orogenies, 55–56, 57, 77, 112, 154, 308, 646 orographic effects, 172, 422 Ortelius, Abraham, 702, 703 Orthoptera, 670 Osborn, 1 Osbourne Guyot, 706, 714 osprey, 122, 798 ostriches, 639, 957 otters, 120, 159, 160, 294, 296 500, 525, 909, 937, 978 “Our Ships at Anchor in the Roadstead” (Gilsemans), 759 Outer Banks, North Carolina, 87 Outer Hebrides (Western Isles), 116, 120, 123–124, 292 outrigger canoes, 277 outwelling, 592–593 Ovalau, 302, 305, 308 overfishing, 709, 715, 779, 782 overwash forests, 591 Owen, Richard, 230–231, 638 owls, 26, 179, 295, 374, 852 oxen, 155, 869 oystercatchers, 122, 132, 295 oysters, 195, 336, 853, 854, 893 Ozarks, 152 ozone, 892, 894 Pacala, S., 913 pacas, 90, 491 Pacific Ocean, 55, 172, 173, 566 Pacific-Philippine Sea plate, 454 Pacific plate Bougainville and, 706 Channel Islands (California) and, 164 Cook Islands and, 191, 192, 709 Easter Island and, 709 Farallon Islands and, 287 Fiji and, 306 French Polynesia and, 339, 340, 341, 342, 712 Galápagos and, 357 Guam and, 585 Hawaii and, 404, 406, 711, 714 Japan and, 503, 506, 705 Kurile Islands and, 522 Line Islands and, 711 Macquarie Island and, 575, 576 Marianas and, 593, 598, 706 New Guinea and, 660, 664 New Zealand and, 674, 675, 676, 678, 679 overviews, 692, 702, 704, 705, 753, 754, 969 Palau and, 149 Samoa and, 802, 807 seamounts and, 715 Solomons and, 703–704, 856 Tonga and, 918 vicariance and, 950 Pacific rats, 415, 416, 470, 471, 472, 672, 721, 792, 793, 794 Pacific region (Oceania), 702–715. See also Line Islands archaeology, 41 bird diversity, 320–321

coral conservation, 782, 784 coral reefs, 783–784 crickets, 208–209 earthquakes, 240 ethnobotany and, 271–276 extinctions, 286 flightlessness, 312, 317 forests, 946–947 formation of, 490 frogs, 348 introduced species, 472 invasive species, 478 mammals and, 588 overfishing and, 311 pre-European impacts, 416, 891–892 rafting and, 775 rats and, 794, 795 reefs, 779 research stations, 789 seafaring, 276–277, 280 seamounts, 819, 823 snails, 537 snakes, 844 sustainability, 890–896 trees, 667 vents, 424, 425, 426 whaling, 976 Pacific ridge system, 426 Pacific slab, 502 pademelons, 905 Pagalu Island, 65 Pagan, 594, 598–599, 600–601, 952 pa- hoehoe, 543, 545, 711, 823 Paine, Robert, 909, 910 Palaearctic region, 120 Palau Islands, 715–716 archaeology and, 44 architecture, 47 conservation, 895, 896 coral reefs, 781, 782, 783, 784 frogs, 349 marine lakes, 603, 604, 605, 606 overviews, 148, 149 pigs and goats and, 742 vegetation and, 943 Palau Ridge, 484, 485 Palawan Island, 349, 729 Palawan Microcontinental Block, 736 Palearctic origination, 72–73, 144 Paleoasia (Laurasia), 504–505 paleoendemics, 73, 183, 184, 253, 255, 256, 390, 864. See also relictual taxa paleofauna, 158 paleorelicts, 157 Paleotropical Realm, 447 La Palma Island, 64–65, 128, 131, 132, 134, 135, 136, 537 Palmerston, 192, 194, 195 palmettoes, 95 palms Canary Islands, 130 Cook Islands, 193 date, 686, 689 extinctions, 245 Fijian endemics, 299

Hawaiian, 944 intraplate islands and, 709 introduced species and, 471 Juan Fernandez islands, 508 Lord Howe Island, 571 Mare aux Songes, 231 Mascarene Islands, 613 New Guinea, 653 Rapanui, 247 rats and, 794 Seychelles, 830 St. Helena, 871 Vanuatu, 940 Palms of the Fiji Islands (Watling), 299 palm worms, 425 Panama, 88 Panama current, 357 pandanus, 611 Pangaea, 63, 504, 505, 624 Pangea, 650 Pantepui, 717–719, 839, 840, 842 Panthalassa, 505 paperbark trees, 944 Papillon (Charriére), 769 Papuan Basin, 661 Papua New Guinea. See also Gondwana biocontrol and, 101 bird plasticity, 6 conservation, 895, 896 ecology, 657 freshwater species, 345 geology, 485, 662, 707 humans and, 42 island arcs, 484 landslides, 535, 536 marine lakes and, 605, 606 missionaries and, 637 oil, 665 overviews, 854 tsunamis, 935–936, 936 vent communities, 427 parakeets, 618, 832 para-littoral forests, 335 parallel evolution. See coevolution parapatric speciation, 166, 564 parasitism, 109, 472, 479, 742, 748. See also specific parasites parasitoids, 101, 102 parrotfish, 782–783 parrots colonization by, 320–321 Fijian, 707 introduced species, 472 New Guinea, 655 New Zealand, 184, 322, 667, 670, 671 Tierra del Fuego, 917 parthenogenesis, 74, 844 Partida Norte Island, 80 Partida Sur Island, 80 partitioning, 106 Patagonia, 973 patches. See fragmentation (patches); islandlike systems pathogens, 252, 748 PCR (polymerase chain reaction), 412

pea family, 157 Pearl Archipelago, 251 peat, 97, 114, 123, 124, 290, 291, 592, 905, 931 peccaries, 91 pelagic animals, 820 pelagic birds, 12, 13, 557, 594 pelicans, 81, 158, 159, 161, 593, 811, 812 Pemba Island, 778 penal colonies, 767–771, 908 Penan people, 115 Penguin Bank, 408 Penguin Island, 19 penguins Antarctic, 12 Atlantic, 67 bird disease and, 104 Darwin on, 957 diseases and, 104 Galápagos, 357, 359, 361 global warming and, 815 human impacts, 814 Lord Howe Island, 569 Macquarie, 574, 575 New Zealand, 322 overviews, 811, 813 Tristan da Cunha/Gough, 932 Peninsular Ranges, Mexico, 162, 163 Penrhyn, 195 Pentadaktylos Range, 212 peppers, 98 Perapedhi Formation, 214 perch, 169 perched lakes, 331 Periplus, 850 perissodactyls, 22 Perkins, R.C.L., 411 permafrost, 60, 86, 866 Peron, Juan, 770 La Pérouse, 742 Persian Gulf, 180, 421 persisters, 81 Peru, 104, 482, 936 Peru-Chile subduction zone, 754 Pescadores, 897, 898 pesticides, 100, 161, 709, 850 pests, 89, 90, 99–102, 121, 160, 556. See also invasive species Peter I Island, 11, 17, 20 Peter the Great, 423 petrels Bermuda, 96 Britain and Ireland, 124 Channel Islands (California), 158 human impacts, 814 introduced species and, 471 Mascarenes, 619 overviews, 811, 812, 813 rats and, 795 Tristan da Cunha/Gough, 932 petroglyphs, 543 petroleum. See fossil fuels pet stores, 346 Petteys, 946 pH, 379, 521, 602, 642, 878. See also acidification, ocean

INDEX

1057

phalaropes, 811 Phanerozoic revolutions, 2 phasmids, 570 phenotypic responses, 844 pheromones, 210. See also olfactory cues Philippine Fault Zone, 732, 733, 735, 737 Philippine Mobile Belt, 736 Philippine plate, 149, 585, 593, 598, 705, 715, 724, 753, 897 Philippines adaptive radiations, 3 biology, 723–731 Borneo and, 114–115 climate, 971 conservation, 895 coral conservation, 780 corals, 379 deforestation, 221 dwarfs, 237 fishing, 310 frogs, 348, 349 geology, 725, 732–738 humans and, 43 invasive species, 346, 347 island rule and, 492 macaques, 777 mammals, 553, 588–589, 590, 776 missionaries and, 634 snakes, 845 species-area relationship and, 857 whaling, 976 Philippine Sea, 149 Philippine Sea plate Indonesia and, 454, 458 Japan and, 501, 502, 503, 506, 705 Philippines and, 732, 733, 735 Taiwan and, 902 pa- hoehoe, 543, 545, 711, 823 phosphate, 149, 336, 387, 439, 532, 534, 694, 877, 878. See also Phosphate Islands Phosphate Islands, 738–74 phreatophytes, 81 phylogenesis, 773 phylogenetic analysis ants, 3 birds, 108, 183, 353, 412, 616 Canary Islands, 129 cichclids, 165, 168 dispersal/vicariance and, 528, 539, 561, 948 distribution of species and, 969–970 Drosophila, 232, 234 endemism and, 255 flightlessness, 312, 317 frogs, 348, 350 gigantism, 376 Gulf of Guinea, 809 invasion biology and, 480 lizards, 562, 563 Mascarenes, 617 New Caledonia, 645 New Zealand, 639 overviews, 464, –465 Pantepui, 719 Philippines, 725, 727, 729, 731 planthoppers, 252

1058

INDEX

Polynesia, 722 radiation zones and, 772, 773, 774 rodents, 589 silverswords, 836, 837 snails, 363 Socotra, 850 spiders, 864 taxon cycles and, 913 tortoises, 359, 921, 922, 923, 925 vent organisms, 424, 426 phylogenetic trees, 412, 464 phylogeography, 13, 616 phytoplankton, 289, 527, 528, 812, 891 picaflor, 507 Pico de Teide, Tenerife, 65 Pico Duarte, 22 Pico Island, 64, 71, 72 piculet, Antilles, 21, 23 pigeon guillemots, 158 pigeons Canary Islands, 131, 132 colonization by, 320–321 diversity and, 319 dodos, 231 Mascarene Islands, 618 New Guinea, 655 New Zealand, 672 Samoa, 800 Tonga, 920 pigments, 151, 234, 258, 461, 468, 797, 863 pigments, dye, 274–275 pigs, 741–743 archaeology and, 43 Baja California Islands, 80 Canary Islands, 132 Channel Islands (California), 155, 158, 160–161 Philippines wild, 724 pre-European impacts, 417 Samoa, 801 Tristan da Cunha/Gough, 932 Pikas, 120 Pikini (Bikini) atoll, 69, 610, 611, 680–681, 683, 684, 712 pillow lava, 491, 953 Pimm, S.L., 282 pine forests, 27, 79, 128, 129, 130, 132, 182, 614, 841, 905, 942 pine-oak forests, 182, 184, 186, 841 pine trees, 27, 78, 128, 157, 288, 296, 331, 395, 450, 614, 628, 707, 769, 797, 981 pingos, 60 pinnipeds, 79, 159, 161, 294, 296, 297, 378 pioneer communities adaptive radiation and, 361 ants, 36–37 dispersal and, 877 gigantism and, 372 Great Barrier Reef, 384, 385 invasive species and, 946 Kerguelen, 15 Krakatau, 518 lava, 402 lava tubes, 547 Pantepui, 718

silverswords, 838 soils and, 944 spiders, 861 succession and, 877, 878 Surtsey Island, 887 pirates, 203, 762, 918 Pisonia grandis, 556 Pitcairn, 538, 712, 744–747, 789, 834, 951 Pitcairn-Gambier, 341 Pitcairn Group, 44–46, 492 pitcher plant, 830 pitohui birds, 5, 6, 655 Piton de la Fournaise, 443, 613, 620, 622 Piton des Neiges, 443, 613, 620, 622 plankton. See also zooplankton dispersal and, 357, 528 Faroe (Faeroe) Islands, 288, 289, 290 Galápagos, 360 global warming and, 379, 817–818 lake, 527, 528 Marianas, 594 Phosphate islands, 740 seabirds and, 812 seamounts and, 819, 820 snails and, 539 sticklebacks and, 876 sustainability and, 891 Tatoosh, 911, 912 tides and, 916 whales/whaling and, 975, 976 plant diseases, 747–752, 749, 751 planthoppers, 252, 463, 548 plants, seed, 129 plants, vascular. See also flowering plants alien species, Antarctic, 15 Antarctic, 12, 14, 15 arctic, 50, 51, 52 Atlantic, 72 Azorean, 71, 73, 74 Canary Islands, 129, 132 Fijian, 299 global warming and, 719 Gulf of Guinea, 810 Hawaiian, 774 Indonesia, 448, 449, 450 inselbergs and, 467, 468, 469 Juan Fernandez islands, 508 Kurile Islands, 524 Lord Howe Island, 571 Macquarie, 574 Madagascar, 580 Madeira, 583 Mascarene Islands, 616, 617 New Guinea, 653 New Zealand, 673 organic falls and, 701 Pitcairn, 744–745 Southern Ocean Islands and, 13 succession and, 877 Surtsey Island, 886–887 Taiwan, 898, 899 temperatures and, 48 Vanuatu, 940 Wallacea, 452 Wizard Island, 980–981

plants (flora). See also ethnobotany; flowering plants; plants, vascular; pollination; seeds; succession adaptive radiation and, 2, 3 age of, 750 Antilles, 27 ants and, 40 arctic, 52, 57 Ascension, 61, 62 Bahamas, 859 Baja California Islands, 80, 81 Bermuda, 98, 98 Borneo/New Guinea and, 658 Britain and Ireland and, 118, 119, 124 Canary Islands, 128–130 Cape Verde, 145–146 Channel Islands (California), 155, 156–157 competition and, 283 convergence, 189 dispersal, 556, 775, 776 dodo and, 231 endemism and, 256 extinctions, 500 Faroe (Faeroe) Islands, 291 French Polynesian endemics, 336 Galápagos, 848 Great Barrier Reef, 384–385 Gulf of Guinea, 810 Hawaiian Islands, 2, 399–400 intertidal, 911 introduced species, 44, 470, 673 invasive, 253, 474 lava flows and, 547 Mauritius, 848 Mediterranean, 628, 629 Midway Island, 632 New Guinea, 653, 656 New Zealand, 322, 667, 671, 672 non-vascular, 130, 889 ornamental, 44, 195, 275, 541, 844 Philippines, 729 rats and, 794, 795 Samoa, 801 Seychelles, 830, 833 sky islands, 842 snakes and, 844–845 species-area relationship and, 857 sub-Antarctic, 15 Surtsey Island, 885 sustainability and, 891 Tenerife, 774 Trinidad and Tobago, 927–928 Tristan da Cunha/Gough, 932 Vanuatu, 941 vegetation vs., 942 waves and, 911 Zanzibar, 983 plasticity, 5, 6, 15, 477–478, 480, 514 plateaux, mountain. See also specific plateaux Africa, 165 Antilles, 20 Arctic, 58 Atlantic, 66 Baja California, 78 Britain and Ireland, 124, 154

Cook Islands, 192, 193 deforested, 223 Faroe (Faeroe) Islands, 287, 288, 289 Fernando de Noronha, 297, 298 French Polynesia, 334, 335, 339 Galápagos, 369 glaciers, 434, 435 Haiti, 24, 25 Kaui, 208 Macquarie Island, 573 Madagascar, 469, 550, 551 makatea, 585, 586 Marianas, 594, 601 Mauritius, 621 Mediterranean, 626, 628 New Caledonia, 646 Pantepui, 839, 840, 842 Socotra, 849 Tasmania, 905 Tristan da Cunha, 931 U.S., 152 plateaux, oceanic. See also specific plateaus Atlantic, 64 Azores, 71 basalts, 431, 898 French Polynesia, 339, 340 Japan, 505 New Guinea, 653, 654 Pacific region, 950 seamounts and, 825 plate-boundary islands, 690, 692–693 plate convergence. See also uplifts extrusion and, 491 Indian region, 437–438 Indonesia and, 454 Kurile Islands and, 523 Macquarie Island and, 576 Marianas and, 598 New Caledonia, 646 New Guinea and, 660 New Zealand, 676, 678 overviews, 481, 690, 691–693, 703, 753, 754 Philippines, 733 seamounts and, 822 Vancouver Island and, 708 vicariance and, 950 plate divergence (spreading) Baja California and, 703 Galápagos, 371, 424 Iceland and, 428, 429, 490 Indian region, 437 Macquarie Island, 575, 576 melting anomalies and, 951 overviews, 481, 484, 690, 690–691, 753 seamounts and, 824 Seychelles and, 382 vents and, 424 plate tectonics, 752–755. See also earthquakes; plate convergence; plate divergence (spreading); rifting; volcanism; specific plates age of crusts and, 690 Antarctic, 18 Antilles, 22, 29–30, 32 arctic, 55–56, 57, 59

Atlantic, 63, 64, 66, 71–72 Baffin Island, 77 Baja California Islands, 78, 79 barrier islands and, 82 beaches and, 92 biogeography and, 949, 968–969 Borneo/New Guinea and, 112, 657 Channel Islands (British) and, 154 Comoros, 178 coral reefs and, 201 Cyprus, 214 earthquakes and, 240 evolution and, 220 fragments and, 187 French Polynesia, 339 freshwater species and, 343 frogs and, 348–349 Galápagos, 367–369 granitic islands and, 381 Greek Islands, 388, 396 Iceland, 428–430, 429, 435 Indian Ocean region, 437–446 Indonesia and, 447 Japan and, 501–502 Kurile Islands and, 522 Lord Howe Island and, 571 Madagascar, 580 Makatea Islands, 585 Mediterranean, 625, 626, 627, 628 moa and, 639 New Caledonia, 646, 647, 648 Newfoundland, 649 New Guinea, 664–665 New Zealand, 674, 676–677 overviews, 690–693, 694, 702, 713 reefs and, 779 Samoa, 803 seamounts and, 714, 821–822, 824, 825 Seychelles and, 829 sky islands and, 840 Socotra and, 847 Solomons and, 855–856 spiders and, 863 Vancouver and, 937 Warming Island and, 972 platform islands, 715 platforms, 642, 738, 797, 803 platypuses, 906, 958 Pleistocene Aggregate Island Complexes (PAICs), 727, 728, 729 plovers, 78, 106, 159 plumes. See cold seeps; mantle plumes plumes, volcanic, 544 plutonic rock Antarctic, 19 Antilles, 30, 31 Canary Islands, 138, 140, 141 Channel Islands (British), 154, 162, 163 Cyprus, 213 Darwin on, 220 Fiji, 306, 307, 308, 309 French Polynesia, 339 Iceland, 430, 433 Japan, 503, 504 Kerguelen, 440

INDEX

1059

plutonic rock (continued) Macquarie Island, 576 Newfoundland, 652 pocket basins, 755–757 Podocarpus, 580 Pohnpei, FSM, 94, 211, 784 Poinar, George and Roberta, 23 Point Bennett formation, 163 polar deserts, 14, 48, 52, 78 polarity reversals, 459 polar regions, 85–86, 104, 171, 421–422, 816 polecats, 120 pollen, 44, 226–227, 291, 838, 907, 970–971 pollination arctic, 52 conservation and, 619 evolution of, 227 extinction and, 284, 286 orchid, 696, 697, 699 selection and, 842 pollution, 54, 266, 782, 884 Polo, Marco, 850 polychaetes, 426, 819 Polynesia ants and, 36 diseases, 977 ethnobotony and, 273 gods, 515 humans and, 42, 43 Japan and, 498 overviews, 758 pigs, 741, 742 popular culture and, 764 radiation zone, 772 spider conservation, 865 timber trees, 274 trade and, 44–45 traditional practices, 890 tsunamis and, 934 Polynesians, 758–761. See also Lapita; Maori people Americas and, 722–723 Comoros and, 180 Cook-Australs and, 709 Cook Islands, 195 deforestation and, 221, 222 Easter Island and, 709 extinctions and, 282 flightless birds and, 316 Hawaii and, 403, 413–414, 711 homeland, 722 impacts, 891–892 lizards and, 560 Marquesas and, 712 New Zealand and, 672 Norfolk Island and, 572 overviews, 705, 706–707, 709 Pacific region and, 720 Pitcairn and, 745 pre-European impacts, 415–416, 516 rats and, 793, 795 Samoa and, 713, 801 seabirds and, 814 Tonga and, 920 ponds, 400, 526

1060

INDEX

pools, 467–468 poppies, 132, 157 popular culture, 761–765, 769, 770, 771 population density, 417, 514, 559–560, 793, 826, 846, 912 population genetics, 766–767 population size, 788 porpoises, 120, 125, 159, 938, 975 Port Arthur, 768 Porter, David, 918 Porter’s hypothesis, 359 portmanteau biota, 415 Porto Santo Island, 64 Portuguese, 180, 984 position of islands, 377 positive gravity anomalies, 339 Possession Island, 12, 15 possums, 673 potoroos, 905 pottery, 43, 721, 722, 941 poverty, 888, 889, 891, 896 Poway formation, 163 Powell, Jeffry, 923 Powell Island, 18 Pratas Atoll, 898 praying mantis, 129, 130 Precambrian-Cambrian boundary, 649 precipitation. See also rain beaches and, 93 elevation and, 173 flowering and, 90 lagoons and, 69 predation body size and, 514 Channel Islands (California), 159 co-evolution and, 189 cricket morphology and, 211 dispersal and, 225 diversity and, 168–169 dwarfism and, 236, 238 ecological release and, 251, 252 extinction and, 169, 282, 283 flightlessness and, 317 Great Barrier Reef, 383 Indonesia, 451 introduced species and, 471–472 invasive species and, 479 island rule and, 494, 495 lizard adaptations and, 564 place and, 911 radiations and, 206 predators birds, 325, 326 coloration and, 655 dispersal and, 531 eradication of, 474 flightlessness and, 312 food and, 843 gigantism and, 376 Hawaiian Islands and, 398–399 introduced, 596, 672 introduced species and, 473 Mascarene Islands, 618 succession and, 878 taxon cycles and, 912

preserves, nature. See reserves, nature President Coolidge, SS, 834 Président Thiers bank, 342 Preston, Chris, 118 prey, 596. See also predation; predators Price, Trevor, 108 prickly pear, 132 primates, 23, 25, 238, 375, 494, 549, 775–776, 778. See also specific primates Prince Edward Islands, 11, 12, 16, 444 Principe, 65, 474, 775, 808–811, 862 Principle of Geology (Lyell), 963 prisons, 767–771, 796, 908 progression of life, 385 progression rule Azores, 73–74 Drosophila and, 773 endemism and, 254 Hawaiian biota and, 233, 255 Hawaiian Islands, 399 insects and, 464 silverswords and, 835–837 propagules, 475–479, 487, 531 protozoa, 62, 857, 891 pseudoscorpions, 62, 130 ptarmigans, 50, 53 pteridophytes, 146, 644, 713 Puercos Island, 251, 252 Puerto Rican Bank, 22, 472 Puerto Rico amphibians, 27 ants, 38 beaches, 93 birds, 26 fish, 28 geology, 22, 29, 692 hydroclimate, 423 invasive species, 346 rodents, 24 shrews, 25 sloths, 24 puffins, 124, 125, 158, 813 Pukapuka, 193 Pulo Anna Island, 149 pumas, 91 pumice rafts, 260 purging, 437 quail, 132 Quammen, David, 220 quays. See cays Queen Charlotte Islands, 874, 938 Queen Charlotte Sound, 675 Queen Maud Land, Antarctica, 66 Quintinia, 940 quokka, 796, 798 quolls, 905 rabbits Antarctic, 15 Baja California Islands, 81 Britain and Ireland and, 121 Canary Islands, 132 Cape Verde, 146 Channel Islands (California), 159

dwarfism and, 238 Farallon Islands and, 295 introduced species and, 473 Macquarie, 575 Mediterranean, 628 Tierra del Fuego, 917 raccoons, 26, 79, 236 radiation, nuclear test, 683 radiations. See also adaptive radiation African, 168 chance and, 206 Comoros, 179 Hawaiian crickets, 208 overviews, 461 preludes to, 108 radiation zone, 772–774 Rae Craton, 77, 78 rafting, 775–778 Azores and, 72 Bermuda and, 96 crickets, by, 206 ephemeral islands and, 259 Fijian iguanas, of, 302 Galápagos and, 368 Gulf of Guinea islands and, 809 lizard, 560 pumice, 260 snails, 539 ragwort, 124 Raiatea, 641, 758 rails Ascension, 62 colonization by, 320–321 diversity and, 319 extinctions, 286, 320 flightless, 312, 313, 314, 314, 316, 321, 325 gigantism, 374 Great Barrier Reef, 383 introduced species and, 471 Madagascar extinctions, 322 Midway Island, 633 New Zealand, 184, 322, 667, 671 Philippines, 726 Samoa, 800 St. Helena, 871 rain. See also precipitation arctic, 51 atolls and, 68 climate and, 172 coral reefs and, 201 deforestation and, 222 dunes and, 95 estimation of, 421 freshwater species and, 345 guano and, 740 Gulf Stream and, 116 heat and, 172–173 mountains and, 172 prehumid climates, 112 storms and, 420 vegetation shifts and, 185 rain forests Borneo, 111–112, 114, 115 Christmas Island, 532, 533–534 climate change and, 171

Cook Islands, 193 diversity and, 106 endemism, 949 Falklands, 957 Fijian, 299 fragments, 181 Frazer Island, 331 French Polynesia, 335 Hawaiian, 945 Indonesia, 451, 453 Mascarene Islands, 614 Maui, 944 New Caledonia, 643 New Guinea, 657 New Zealand, 667 overviews, 942 Samoa, 800, 801 Solomons, 851, 852 Sri Lanka, 867, 868, 870 succession and, 945–946 Tasmania, 905 Tonga, 918, 920 Vanuatu, 940 Zanzibar and, 983 Raivavae Island, 338, 342 Raja Ampat, 653 ranching, 157, 160 Rand, Stanley, 90 random dispersal, 225 range size, 284, 285 Rangiroa, 333, 334, 642 Rangitoto eruption, 676 ranid frogs, 7 Ranongga Island (Solomons), 694–695 Rapa, 342, 462, 538 Rapa Island, 3, 335–336 raptors, 124, 182 Raratonga Treaty, 685 rarity and restriction, 284–285 Rarotonga, 191–193, 194, 195, 196, 197, 335, 339, 342, 416, 636, 709, 949 Rasa Island, 80, 81 ratites, 639, 670, 671 rats adaptive radiations, 3 Antarctic, 15 Antilles, 25 Ascension, 62 Baja California Islands, 79, 81 Bermuda, 97 biodiversity and, 794–795 Britain and Ireland and, 120, 121, 124 Canary Islands, 131, 132 Cape Verde, 146 Channel Islands (California), 161 Comoros, 179 dispersal, 793 eradication of, 474 Galápagos, 926 Gulf of Guinea, 809 Hawaiian birds and, 326 Indonesia, 453 introduced species, 470, 471 Lord Howe Island, 570 mammals and, 590

Marianas, 595, 596 New Zealand, 673, 959 Norway, 792 overviews, 742 Philippines, 588–589, 729 Rapanui, 245, 248 Samoa, 801 Solomons, 852 sub-Antarctic, 15 Tonga, 920 tortoises and, 925, 977 Tristan da Cunha, 932 Raven, Peter, 1 Ravolomanana, M. Marc, 581 rays, 194 razorbills, 125 Reao, 334, 339 rear-arc chains, 484 recolonization, 465, 517–519, 777, 840 red colobus, 986 Red Sea, 421 red-tailed tropic birds, 103–104 redwood, 899 reef ecology and conservation, 779–785 reef fishes, 310, 500, 615, 632, 654, 658, 781, 783, 801, 852–853, 949 reef flats, 614 reef islands, 587 reefless islands, 934 reefs. See also coral reefs age of, 803 beaches and, 93, 94 Bermuda, 95, 98 Caroline Islands and, 149 Cook Islands, 192–193, 194 Cyprus, 214, 215 French Polynesia, 334, 338 mangroves and, 592 pigs and goats and, 742 pre-European impacts, 417–418 Trinidad and Tobago, 927 waves and, 881 reefs, non-coral, 819 refugia, 182, 463, 632, 645, 785–786, 795, 847, 849, 905, 937, 938 regoliths, 433 reindeer, 15, 49, 51, 52, 53, 54, 120, 236, 470, 471, 472, 473, 866, 977 Reitoru atoll, 334 rejuvenation volcanism, 953 relaxation, 548, 697, 787–788, 929 relaxation models, 151, 183, 185, 312, 489–490 relictual taxa endemism and, 255 examples, 157, 181, 183 Fiji, 299 fragments and, 181, 185, 187 frogs, 348 Japan, 499 Madagascar, 578 Mascarene Islands, 614 New Caledonia, 643, 645 overviews, 182, 184 Pacific region, 949 Socotra, 847, 850

INDEX

1061

relictual taxa (continued) spiders, 864 Taiwan, 899 Tasmania, 905, 906 religious architecture, 46–47 The Reluctant Mr. Darwin (Quammen), 220 remnants, 329. See also fragmentation (patches) Remote Oceania, 720, 721, 741 René, Albert, 832 reproduction corals, 780–781 effort, 226, 238, 239 extinctions and, 788 insect radiations and, 465 pressures and dwarfism, 236 seamounts and, 820 sticklebacks, 875–876 strategies, 285, 560, 697, 831 success, 283, 285, 361, 607 reproductive assurance, 697–698, 699 reproductive isolation birds, 105, 106, 108–111 deep-sea animals, 757 fish, 169 founder effects and, 327 plants, 837 sky islands, 841, 842 spiders, 864 tortoises, 925 Wallace on, 966 reptiles Antilles, 27 Baja California Islands, 78, 80, 82 Bermuda, 98, 98 Borneo/New Guinea and, 658 Britain and Ireland and, 122 Canary Islands, 131 Cape Verde, 144, 146 Channel Islands (California), 155, 159 Comoros, 178, 179, 180 Cook Islands, 194 dwarfism and, 236, 238 dwarfs, 237 fire ants and, 941 Great Barrier Reef, 383 Gulf of Guinea, 809, 811 Indonesia, 449, 450 Japanese, 499 Kurile Islands, 525 Mascarene Islands, 615 New Caledonia, 644 New Guinea, 654, 656 New Zealand, 184, 667, 670 overviews, 843 Palau, 716 Pantepui, 718–719 Philippines, 724, 729 rafting and, 774 Seychelles, 830, 833 Socotra, 849 Solomons, 852, 853 species-area relationship and, 857 Sri Lanka, 869 sustainability and, 891

1062

INDEX

Taiwan, 899 Tasmania, 906, 907 Tonga, 920 Vanuatu, 940 rescue effect, 488 research. See also scientific operations ants and, 39, 41 biocontrol and, 102 Farallon Islands, 293, 294–295, 296–297 Fijian, 299 forest, 91 frogs, 350–351 invasive species, on, 480 Krakatau and, 519 overviews, 756 Panama and, 88 red deer, 124 Socotra, 848 Spitsbergen, 866 Tatoosh, 909 research stations, 125, 788–791, 884, 926 resemblance. See convergence, evolutionary reserves, nature. See also marine protected areas (MPA’s); national parks Antarctic, 16 Antilles, 29 ants and, 39 arctic, 54 Atlantic, 64, 66, 67 Azorean, 74 Baja California Islands, 78, 81 barrier islands, 83 beaches, 93 Bermuda, 97–98 Borneo, 112 Britain and Ireland, 124, 125, 126 Canary Islands, 132 Cape Verde, 147 Channel Islands (California), 161 Cook Islands, 196–197 coral reefs, 780, 782 Cozumel, 205 Farallon Islands, 293–294 Fernando de Noronha archipelago, 297, 298 Frazer Island, 330 French Polynesian, 337–338 Great Barrier Reef, 386 Haiti, 27 Hawaiian, 403 Indian Ocean region, 440, 441 Indonesia, 449, 450 Juan Fernandez islands, 508 Lord Howe Island, 568, 571 Macquarie, 575 Madeira, 585 Micronesia, 896 Norway, 567 Panamanian, 88 Pitcairn, 747 Rottnest Island, 797, 798 Seychelles, 832–833 species-area relationship and, 861 Spitsbergen, 866 Taiwan, 901–902

Tasmania, 908 Tristan da Cunha/Gough, 932 vent communities, 427 resource exploitation. See also fishing; mining; oil; specific activities archaeology and, 44 coral reefs, 783 Gulf of Guinea, 810 Juan Fernandez islands, 508 Madagascar, 581–582 Philippines, 730 research on, 790 Samoan turtles, 801 Seychelles, 832 Solomons, 854 sustainability and, 889, 891 Tristan da Cunha/Gough, 931 Vancouver, 938 Wallace on, 963 resource limitation, 236, 237–238 resource management, 417 restaurants, 763 restriction and rarity, 284–285 Resurrection plants, 469 La Réunion Island ants, 41, 755 biogeography and, 107 birds, 108–109, 229, 313, 316, 615, 617 conservation, 618–619 ecology, 614 geology, 441, 442–443, 444, 445, 536, 612, 613, 620, 621, 622, 690, 693 gigantism, 373, 374 orchids, 697, 698 research, 789 research stations, 789 seamounts, 824–825 storms, 420 tortoises, 922 tsunamis and, 935 weevils, 2 Reykjanes Ridge, 428 rheas, 639 rhinoceroses, 22, 51, 113, 285, 389, 395, 447, 450, 869, 922 Rhodes, 237, 388, 393, 396, 623, 626 rhododendron, 124, 852, 868, 898 Rhyncocephalia, 184 Rhynochetidae, 644 rhyolites Antarctic, 19 Antilles, 31 Atlantic, 64, 65, 66, 67 Canary Islands, 139 Channel Islands (California), 162, 163, 164 Iceland, 432, 433 island arcs, 484, 485 Mediterranean, 625, 627 New Zealand, 676 Pacific islands, 309, 523, 805, 951 Richter scale, 241 Ricklefs, R.E., 912 ridge-hotspots, 71, 441 ridges, 424–425. See also specific ridges ridge spreading, 425, 441

rifting continental islands and, 707 East Asian, 505–506 Gulf of Guinea, 808 Hawaii, 804 island arcs and, 484 Japan and, 485, 504 Mediterraean, 624 New Guinea, 665 New Zealand and, 675, 678 North Atlantic and, 652 Pacific region vicariance and, 949–950 Samoa, 806 Socotra and, 847 volcanism and, 952–953 rift zones Canary Islands, 136 Fiji, 308, 309 Galápagos, 370 Hawaii, 405 Iceland, 429, 434 Indonesia, 458, 460 seamounts and, 823 volcanic, 64 Rimatara Island, 335, 342 Ring of Fire, 240, 523, 545, 676, 702–703, 704, 711, 897 ringtail lemurs, 552 Rio Grande Rise, 951 riparian forests, 157, 182, 335, 345 Ritter Island, 537 rivers. See also granitic islands barrier islands and, 916 Comoros, 180 isolation and, 187 Kurile Islands, 521 mangroves and, 592 rafting and, 775 Taiwan, 898 tides and, 915 tsunamis and, 934 Robertskollen nunatak group, 10 Robertson, R.B., 977 Robichaux, R.H., 836 robins, 122, 667, 830, 833, 842 Robinson Crusoe (Defoe), 508, 763 rockfish, 159, 701, 713 rock outcrops, 466 rock pools, 467–468 Rocky Mountains, 840, 841, 842 rodents, 792–795. See also specific rodents Antilles, 23–24 arctic, 49, 51 extinctions, 205 Galápagos, 365 gigantism, 374–375 Greater Antilles, 21 Madagascar, 588, 776 New Guinea, 655 Philippines, 724 sigmodontine, 24 Sri Lanka, 869 Rodinia supercontinent, 504 Rodrigues Island, 107, 231, 313, 316, 373, 374, 444

birds, 615 ecology, 614 geology, 442, 444, 612, 613, 621, 622 phylogeography, 616 Roggeveen, Jakob, 249–250 Romanche Transform, 754 Rongelap, 683 Rongerik, 680, 682, 683 Roosevelt Island, 17 Roosevelt, Theodore, 632 Rosario group, 163 Ross Island, 17 Ross Sea region, 20 Rota, 474, 538, 594, 595, 597, 598, 599, 602, 738, 739 rotation African plate, 624 Canary Islands, 142 Channel Islands (California), 162, 164 Cyprus, 215 Iceland, 435 overviews, 485 Pacific islands, 113, 308, 309, 458, 675, 735, 737 Philippines, 735 rotation, Earth’s, 817 Rothera station, Antarctica, 14 Roth, Louise, 238 Rothschild, Walter, 410 rotifers, 12, 298 Rottnest Island, 768, 796–798 Roughgarden, J., 913 Roussel, Hippolyte, 249, 250 Royal Society Fiord, 77 rubber, 115 Rubiaceae, 653 Rubinoff, Ira, 88 Ruesink, Jennifer, 910 Rum, island of, 120 Rurutu, 333, 342, 343 Russello, Michael, 923 Russian Arctic islands, 49, 51–52, 54, 59, 874 Russians, 522 rust fungus, 748, 750, 751, 752 Ryan, Peter, 108 Ryukyu Islands, 474, 482, 497, 498, 499, 500, 501, 502, 503, 506, 522, 902 Saba, 26, 32 Sabah, 115 Sahara desert, 72, 95, 140, 144, 146, 208 Sahul, 720 sailing, 42, 43, 103, 276, 918, 984. See also discovery; exploration; seafaring sailors, 228–229, 570, 710, 741, 742 Saint Cuthbert, 125 Saint Peter and Paul Rocks, 66 Saipan, CNMI, 94, 483, 593, 595, 596, 597, 598, 599, 602, 603, 634, 705 Sakhalin, 770 salamanders, 159, 161, 182, 499 salinity, 202, 344, 345, 346, 347 salinity crisis, 624, 625 Sal Island, 65, 143, 144, 147 Salix herbacea, 290

salmon, 124, 899 Salsipuedes Island, 80 salt, 14, 797, 812 salt marshes, 331 Samoa, 799–808 bats, 194 birds, 303, 304 conservation, 896 Cook Islands and, 195 crickets, 208 ethnobotony and, 274 geology, 708, 712–713 Lapita and, 43 medicinal plants, 275 missionaries and, 636 seamounts, 824–825 snails, 539 snakes, 844 volcanism, 704, 953 Samoa-Tuvalu island chain, 693 Samos, 395 Samothrace, 395 San Andreas Fault, 78, 161, 164, 693, 703 San Clemente Island, 156, 157, 159, 162, 163 San Cristobal (Galápagos), 360, 361, 363, 365, 770, 922, 923, 925 Sand Adreas Fault, 754 sandalwood, 941 sand and bare ground, 632 Sanday Island, 124 San Diego, 162, 163 San Diego Island, 80, 82 Sandoy Island, 292 sandpipers, 78, 106 sandstones, 650, 651, 677 Sandwich Islands, 423, 755 San Estéban Island, 80, 82 San Francisco Island, 80 Sangihe arc, 459 San José Island, 80, 81 San Juanito Island, 79 San Lorenzo Island, 80 San Marcos Island, 80 San Miguel Island, 156, 157, 158, 160, 160, 161, 162, 163 San Nicolas Island, 156, 159, 162, 163 San Onofre breccia, 162, 163 Santa Barbara Island, 156, 158, 159, 161, 162, 163 Santa Catalina Island, 81, 156, 157, 159, 162, 163 Santa Cruz Island, California, 80, 82, 156, 157, 158, 159, 161, 162, 163, 257, 474 Santa Cruz Island, Galápagos, 111, 359, 924 Santa Cruz Island fault, 162 Santa Cruz Islands, Solomons, 851, 854, 855, 921 Santa Luzia Island, 143 Santa Maria Island, 64, 71, 73, 74, 923 Santarosae, 158, 160, 492, 707 Santa Rosa Island, 156, 157, 158, 159, 160, 162, 163 Santiago (St. Jago) Island, 143, 144, 146, 147, 217, 218, 474 Santo Antão Island, 65, 143, 144 São Jorge Island, 64, 71, 72 São José Islet, 65

INDEX

1063

São Miguel Island, 64, 71 São Nicolau Island, 143, 147 São Tomé Island, 65, 108–109, 775, 808–811, 862 São Vicente Island, 65, 143, 147 Sarah Island, 769 Sarawak, 90, 115 sardines, 296, 525 Sardinia, 624, 629, 741 Sarigan, 598, 599, 601 Sark, 154 Saronic Islands, 626 Satawal, 783 Sator, 82 saturation, 384 Sauromalus, 82 Savai’i, 758, 952 savannas, 644, 657, 718, 927, 983 scales, 98, 472 Scandinavia, 117, 121, 650, 874, 973 scarab beetles, 7 scarabs, 899, 902 scavengers, 700, 974 scent, 697 Schluter, Dolph, 1, 4, 859 Schmitt, Dennis, 972 Schnierla, T.C., 88 Schoener, T., 860 Schouten, Willem Cornelisz, 917 science, 228–230, 784–785. See also research; research stations scientific expeditions, 427, 542. See also Beagle, HMS scientific operations alien species and, 16 Bermuda and, 97 biocontrol and, 101 Britain and Ireland, 124 Channel Islands (California), 161 cold seeps and, 175 Indian Ocean region, 440, 441, 443, 444 outdated, 186 Scilly Isles, 116, 121, 126, 337, 338 scleractinian coral, 96, 195, 198, 199, 200, 201, 202, 614, 779, 781 sclerophyllous forests, 72, 128, 132, 156, 299, 578, 579, 614, 618, 643 Scotia Arc, Antarctica, 10, 66 Scotia plate, 482 Scotia Sea formation, 18 Scotia Trench, 67 Scotland, 567 Scott Island, 17, 20 scribbly gum/wallum banksia forest, 331 Scrivenor, J.B., 968 scrub Baja California Islands, 81 Cape Verde, 146 Cook Islands, 193 Maui, 836 Mesoamerican, 184 New Caledonia, 643 Rapanui, 245 Samoa, 800 Vanuatu, 940

1064

INDEX

Zanzibar, 983 seabirds, 811–815. See also birds; nesting Antarctic, 12, 13, 14 arctic, 50 Ascension tropical, 62 Atlantic, 67 Baja California Islands, 80, 81 Bermuda, 96 bird disease and, 105 Britain and Ireland and, 122, 124, 125, 126 Cape Verde, 144, 147 Channel Islands (California), 158–159, 161 Cook Islands, 194, 197 extinctions, 931 Farallon Islands, 293–294, 295–296, 296–297 global warming and, 379 Gough Island, 930 Great Barrier Reef, 383, 384–385 humans and, 43 Indonesia, 453 introduced species and, 471 Line Islands, 556 Lord Howe Island, 570 Macquarie, 574 Midway Island, 632, 633 navigation and, 280 New Zealand, 672 phosphate and, 740 pre-European impacts, 416 rats and, 795 Rottnest Island, 798 seamounts and, 820 Seychelles, 832 Socotra, 849 Solomons, 852 St. Helena, 871 Tatoosh, 911 Tonga, 920 Tristan da Cunha, 930 Tristan da Cunha/Gough, 931 Vancouver, 938 whaling and, 976 sea cows, 22, 853 sea cucumbers, 781, 853 sea depths, 491 seafaring, 276–281, 292, 629. See also sailing sea floor, 847, 854, 856. See also deep-sea speciation; vents seamounts and, 825 sea grasses Bermuda, 96 coral and, 780, 781 Fiji, 302 global distribution, 779 Japan, 500 Mascarene Islands, 615 organic falls, 701 patchy, 182 Raratonga, 194 Rottnest Island, 796, 798 shipwrecks and, 834 Solomons, 853 threatened species, 893 Zanzibar, 984

sealers, 742, 908, 931 sea levels, 815–818. See also intertidal habitats; land bridges; tides barrier islands and, 83, 87 Bermuda and, 95, 96 biodiversity and, 113 Borneo and, 112 Britain and Ireland and, 117, 125, 126 Canary Islands and, 127 Channel Islands (British) and, 154 Chatham Islands and, 323 climate change and, 170, 171 coasts and, 83 coral and, 201 coral reefs and, 219 Darwin on, 957 dispersals and, 776 diversity and, 320 erosion and, 261 Fernando de Noronha archipelago and, 297 Frazer Island and, 331 French Polynesia and, 336, 711 freshwater species and, 345 glaciations and, 524, 785 global warming and, 377–378 granitic islands and, 381 Great Barrier Reef and, 388 Greek Islands and, 388, 389 Gulf of Guinea, 808–809 Indian Ocean, 178 Indonesia and, 447, 448 island formation and, 491–492, 973 Line Islands and, 554 Lord Howe Island and, 569, 570 Maldives and, 587 mangroves and, 592 marine lakes and, 603 Mascarene Islands and, 613 Mediterranean, 628 moa and, 640, 642 motu and, 641, 642 New Guinea and, 665, 708 New Zealand, 666, 666, 675, 678, 679 oceanic islands and, 695–696 overviews, 693 Pacific region, 707 Philippines, 727, 728 prediction of, 262 seamounts and, 825 Seychelles and, 833 Socotra and, 847 South America, 956 southwest Pacific and, 969 Tasmania, 904, 958 Trinidad and Tobago and, 927 volcanoes and, 268 Wallace’s Line and, 968 waves and, 880, 881 sealing, 15, 54 sea lions, 79, 159, 161, 296, 358, 359, 368, 525, 672, 798, 938, 978 Seal Nunataks, 17, 19 seals Antarctic, 12, 13, 14 Atlantic, 67

Baffin Island, 76, 78 Baja California Islands, 78 Britain and Ireland and, 125, 126 Canary Islands, 132 Channel Islands (California), 159, 160, 161 currents and, 357 Farallon Islands, 294, 296 Galápagos, 368 human impacts, 14 Juan Fernandez islands, 508 Kurile Islands, 525 Macquarie, 574, 575 Midway Island, 632 Pacific region, 776 predators, 50 South Georgia, 977 Spitsbergen, 866 St. Helena, 870 Surtsey Island, 886, 887 Tristan da Cunha/Gough, 931–932 Vancouver, 938 seamounts, 818–825. See also specific seamounts age of, 824–825 Atlantic, 63, 64, 65 Bermuda, 95 Canary Islands, 133, 136, 140–141, 142 Caroline Islands, 149 central Pacific, 950 Galápagos, 371 Galápagos, 710 geology, 951 Line Island chains, 711 North Atlantic, 652 old, 706 overviews, 490, 491, 694, 704–705, 706, 713, 754 Pitcairn, 744 Samoa, 802, 803 Society Islands, 712 tides and, 916–917 Sea of Cortés Islands, 80–82 Sea of Japan, 536, 505–506, 537 sea palms, 910, 911 Searle, Jeremy, 124 seas, defined, 756 seasonal forest, 940 sea urchins, 566, 910, 911, 912 Sea Venture, 834 seawalls, 261, 262 seaweeds Cape Verde, 146 cold seeps and, 175 Cook Islands, 194 ephemeral islands, as, 258, 259 Macquarie Island, 574 Rottnest Island, 796 Surtsey Island, 886 Tatoosh, 910 threathened, 893 Tristan da Cunha/Gough, 930 Sebens, Kenneth, 910 secondary sexual characteristics, 826, 827 sedges, 53, 78, 905, 930 Sedgwick, Adam, 955

sedimentary rocks. See also sedimentation Antarctica, 19 New Caledonia, 646 Newfoundland, 650, 651, 652 New Zealand, 677 Philippines, 734 South American, 718 Taiwan, 903 Tierra del Fuego, 957 Warming Island, 972 Zanzibar, 983 sedimentation. See also flysch; sedimentary rocks Antarctic, 18–19 Antilles, 30, 31 archaeology and, 44 arctic, 55, 56–57, 57–58 Atlantic, 67 atolls and, 68 Baffin Island, 77 barrier islands and, 82, 83, 84–85, 86, 86, 87, 93 beaches and, 91, 92, 93, 94 Borneo and, 113 Channel Islands (British), 154 Channel Islands (California), 162, 163, 164 Cook Islands, 194 coral and, 200 Cyprus, 212, 214, 214–215 deforestation and, 222 erosion and, 261 Fernando de Noronha archipelago, 298 Fiji and, 308 Frazer Island and, 331 freshwater species and, 344 global warming and, 379–380 Iceland, 436 Indonesia and, 456, 457, 458, 459 island arc volcanoes and, 954 landslides and, 535 Lord Howe Island, 569 Macquarie Island and, 576 Madagascar and, 578 Maldives and, 587 marine lakes and, 606 methane and, 175 motu and, 641 Newfoundland, 649 New Guinea, 661, 663, 664 New Zealand, 680 overviews, 481 seamounts and, 491, 819, 822, 824 subduction and, 482 tundra and, 60 Vancouver and, 937 volcanic arc islands and, 691 waves and, 880, 882–883 seeds, 225, 226–227, 299, 877, 885–886, 887, 920, 944 seeps, 426, 974 Segeberger Höhle, 150 seismic activity. See earthquakes (seismic activity) selection pressures dwarfs and, 237–238

ecological release and, 252 island rule and, 494–495 New Zealand, 671 orchids and, 699 sky islands, 841 self-incompatibility, 838 self-pollination, 697–698 Selkirk, Alexander, 508 Selvagen Islands, 64, 127, 133, 140–142, 862 semi-deciduous forests, 193 Senckenberg Museum, 848 sensu stricto, 483 Seri Indians, 82 Sespe formation, 163 Severgin volcano, 524 Severnaya Zemla Ostrova, 47, 51, 52, 59 sewage, 196, 197 sexual difference, 226 sexual dimorphism, 232 sexual selection, 825–829 adaptive radiation and, 3 diversification and, 6 Drosophila, 232 Hawaiian crickets, 463 mental phenomena and, 965 sky islands and, 842 Wallace on, 966 Seychelles, 829–833 ants, 39, 41 bats, 179 bird disease and, 104 continent and, 381–382 endemic amphibians, 348 frogs, 348, 773 geckos, 179 geology, 177, 445–446, 755 introduced species, 472 lizards, 561 plant disease and, 748–749 rats and, 793 tortoises, 922, 924 whaling, 976 Seychelles Plateau, 177 Seymour Island, 17, 19 shags, 814 shales, 67, 214, 393, 439, 457, 651, 663 shape of islands, 172 Shark Bay, Australia, 56 sharks, 78, 145, 194, 296, 301, 557 Sharp, Andrew, 760 Shaw, K.L., 209 shearwaters, 132, 811, 812, 813, 815, 932 sheep Britain and Ireland and, 118, 121, 124, 125, 126 Canary Islands, 132 Channel Islands (California), 160–161 Faroe (Faeroe) Islands, 289, 290 Marquesas, 223 Mediterranean, 629 Rapanui, 245–246, 250 Tierra del Fuego, 917 sheeted-dyke complexes, 213, 214, 576 Shekelle, Myron, 552 shellfish, 194, 196

INDEX

1065

Shetland Islands, 116, 123, 292 shields/shield volcanoes Antarctic, 19 Arctic, 55, 56 Atlantic, 64, 65 Baja California and, 78 Canadian, 77 Canary Islands, 134, 135, 136, 137–139, 140, 141, 142 French Polynesia, 339, 341 Galápagos, 369–370 Gulf of Guinea, 808 Hawaiian stages, 404–409, 953 Iceland, 432 Iceland Holocene, 434 Indian Ocean region, 444, 445 inselbergs and, 466 Kurile Islands and, 522 landslides and, 536 Lord Howe Island, 568, 569–570 Mascarene Islands, 621–622 overviews, 694, 754, 755, 952, 953 Samoa, 803, 804, 805–806, 807–808 seamounts and, 823 St. Helena, 870–871 Shikoku, 484, 485, 497, 498, 499, 500, 502, 506 Shindo, Kaneto, 423 ship rats, 792, 793, 795 ships. See trade; watercraft shipwrecks, 833–835 shorelines. See also beaches; coasts Bermuda, 98 birds, 106 colonization and, 206 subsidence and, 954 shrews Antilles, 21, 23, 25 Baja California Islands, 81 Britain and Ireland and, 126 Canary Islands, 129, 131, 132 Greek Islands, 390 Philippines, 729 rats and, 795 Socotra, 849 Sri Lanka, 869 shrikes, 159 shrimps, 425, 426, 729, 797, 906, 907 shrubs Baja California Islands, 81 Channel Islands (California), 156, 157 Hawaiian, 836, 944 introduced species, 470–471 Mediterranean, 628 Midway Island, 632 New Caledonia, 644 New Guinea, 657 Pacific region, 891, 948 Pantepui, 718, 719 pigs and goats and, 742 Rottnest Island, 797–798 Samoa, 801 Socotra, 848 St. Helena, 873 succession and, 878 Tasmania, 905

1066

INDEX

Vanuatu, 940 Wizard Island, 981 Siamese people, 115 Siberian Arctic, 51, 58 Siberut, 483, 774 Sibumasu block, 456, 457 Sicily, 183, 238, 423, 492, 494, 625, 629 Signy Island, 14–15 Silver Bank, 22 silverswords, 4, 401, 402, 773–774, 835–838. See also tarweeds Simberloff, D., 859 Simeulue, 483, 777, 778 Sinbad, 978 Sindia, 146 Singapore, 263, 282 single-island radiations, 462, 463, 464 sinkholes, 20, 24, 25, 150, 394, 983 sinks, 528, 630 SiO2, 406, 804–805, 806, 951 Siple Island, 17, 20 sipunculids, 819 sister taxa, 359 size of islands. See also area factors; space, ecological; species-area relationship (SAR) adaptive radiation and, 531 Antilles, 28 ants and, 36 biodiversity and, 107 biogeography and, 488–489 deforestation and, 222 dispersal and, 228 distribution and, 949 diversity and, 186, 320, 329 erosion and, 262 evolution and, 124 extinction and, 283, 285, 329 fragments and, 187 Galápagos biodiversity and, 371–372 Gough Island, 930 groundwater and, 420–421 introduced species and, 473–474 lakes and, 529 Line Islands, 555 overviews, 9 research and, 788 saturation and, 384 seamounts, 820, 824 snails and, 540 species-area relationship and, 858–859 species richness and, 259, 527 succession and, 878–879 Tristan da Cunha, 930 size of lakes, 166 size of population, 750 size responses. See dwarfism; gigantism skerry communities, 887 skinks Bermuda, 96 colonizations, 560 Comoros, 179 Gulf of Guinea, 809 Mascarene Islands, 618 mice and, 794

New Zealand, 671 nonadaptive speciation, 561 Philippines, 729 Samoan, 800 Seychelles, 831 Solomons, 852 Tasmania, 906 Tonga, 920 Skokholm Island, 255 Skottsberg, Carl, 509 skuas, 813 skunks, 158 Skúvoy Island, 289 sky islands, 839–842, 839–843. See also Pantepui Hispaniola, 22 insects and, 460 relaxation and, 787 spiders, 863 U.S., 6, 235, 460, 863 skylark, 126 slabs, 484–485. See also subduction slash and burn, 416 slip, plate, 753 sloths, 21, 23, 24–25, 90, 236, 321, 375, 590 Slovenia, 150, 151, 152, 153, 623 slugs, 118, 145, 671, 938 small-island effect, 384 Small Isles, 116, 124 small populations, 284, 285 Smith, A.C., 299 Smith Island, 17, 18 Smithsonian Institution, 88, 89, 411 smut fungus, 748, 749, 750 Snæfellsjökull, 434 Snæfellsnes Zone, 429, 433 snails adaptive radiation, 540 Azores, 73 Baja California Islands, 78 Bermuda, 95, 96, 98 biocontrol and, 100, 101 Canary Islands, 130, 774 conservation, 540–541 dispersal, 539 endangered species, 908 extinctions, 283 French Polynesian, 335–336, 337 Galápagos, 363–364, 365 Gondwana and, 670 introduced species, 470, 473 invasive species, 346–347 Japan, 499, 786 lakes and, 529–530 Lord Howe Island, 569, 571, 572 Marianas, 594 Mascarene Islands, 617 New Zealand, 672 Pacific region, 670 Pitcairn, 745 Polynesian, 949 rafting and, 775 rats and, 794, 795 refugia and, 786 Samoa, 800, 801

Seychelles, 830, 831 Society Islands, 712 Socotra, 849 Solomons, 852, 853 St. Helena, 872 Tasmania, 907 Tristan da Cunha, 930 waves and, 911 snakes, 843–846 Antilles, 27 Baja California Islands, 80, 82 body size, 373, 559 Borneo, 114 Borneo/New Guinea and, 658 Britain and Ireland, 122 Channel Islands (California), 159 colonization and, 560 Comoros, 179, 180 dispersal, 844 dwarfism and, 238 Fijian, 302, 774 Guam, 596, 706 Gulf of Guinea, 809, 810 introduced species and, 471 invasive species, 479 island rule and, 493, 494 land bridge islands and, 558–559 New Guinea, 654 Philippines, 724 Rottnest Island, 798 Samoa, 800–801 Seychelles, 831 Socotra, 849 Solomons, 853 Sri Lanka, 869 Taiwan, 899, 900 Tasmania, 906 Tonga, 920 Vanuatu, 940 Wizard Island, 981 Snares, 16 Snares Island, 11, 16 snipe, 795 Snowball Earth, 57 Snow Hill Island, 19 social facilitation, 285 societies, human, 41–42, 46–47 Society Islands adaptive radiation, 712 biocontrol and, 101 birds, 335 Cook Islands and, 195 ethnobotony and, 273 freshwater species, 345 geology, 340–341, 712 human colonization, 759 insects, 712 marine life, 336 missionaries and, 634 snail, 283 snails, 538, 540 volcanism, 953 socioecosystems, 416 Socotra, 421, 423, 847 Socotra archipelago, 846–851

soft corals, 198–199 soils aging, 944, 945 alien species and, 99 Borneo, 114 Britain and Ireland and, 123, 124 Britain’s, 117 Cape Verde, 144, 146 Cook Islands, 193 coral reefs, on, 201 dwarfism and, 238 Hawaiian Islands, 398 hydrology and, 421 introduced species and, 471 lava flows and, 547 Line Islands, 555–556 Mediterranean, 628 movement, 13 New Caledonia, 643, 645 pigs and goats and, 742 Pitcairn, 744 plant disease and, 749 Rapanui, 251 rats and, 795 succession and, 877, 879, 946 Tasmania, 905 temperature and, 172 Tonga, 919 Tristan da Cunha, 930 ultramafic rocks and, 663 Vanuatu, 940 vegetation and, 942 volcanic, 755 Solander Island, 676 solar radiation, 109 solenodons, 21, 25 solid waste management, 196, 197 Sollas, W.J., 481 Solomons, 851–857 conservation, 895 crickets and, 208 forests, 851 frogs, 349 geology, 306, 485, 690, 691, 702, 703 human colonization, 720 humans and, 42, 43 landslides, 536 lizards, 560 mammals, 774 overviews, 705, 706 pigs, 741 snails, 538, 539 species-area relationship and, 857 Solomon Sea basic, 664 Solovetsky Islands, 770–771 Somalia, 934 Somali plate, 178, 445 Somerset Island, 976 songbirds Channel Islands (California), 159 introduced species, 472 mating, 413 radiation, 106 reproductive isolation and, 110 St. Helena, 871

song, cricket, 210, 211 songes, 891 songs, bird, 110–111, 413 songs, cricket, 209 Sonoran taxa, 80, 81, 82 Sonsorol Island, 149 soricomorphs, 21, 25 Sorong Fault, 454, 459, 663 Sorong fault, 459 sorting, 189, 191, 220, 528 sorting, species, 527, 528, 530, 531 Sotavento group, 65 Soufrière, Montserrat, 31, 32, 33–34, 35 South Aegean volcanic arc, 396 South Africa, 7, 16, 104, 477, 669, 770 South America. See also specific regions African dispersals, 560 Antilles and, 21, 23 ants from, 36 beavers and, 793 Darwin on, 69, 217 fish, 28 Galápagos and, 358–359, 360, 361 Gondwana and, 669 human colonization, 722–723, 760 Juan Fernandez Islands and, 508 land mammals, 24 prisons and, 769 rats and, 793 seamounts, 819 spiders, 863 tectonism, 69 South American plate, 30, 31, 67, 482, 753 South Atlantic, 755 South Carolina coast, 83, 84, 87 South China, 113, 501, 504, 505, 506 South China Sea, 446, 458, 506, 724, 732, 733, 735, 902 Southeast Asia frogs, 348, 349 human colonization, 721 macaques, 777 mammals, 552 missionaries and, 634, 637 New Guinea and, 656 pigs, 741 primates, 549 threatened species, 285 southern hemisphere, 14 Southern Hemisphere whales, 977, 978 Southern Ocean islands, 10, 11, 12, 13, 15, 16, 814 Southern volcanic zone, 429, 433, 601 South Georgia, 10–11, 13–14, 16, 66–67, 104, 471, 472, 789, 793, 814, 976, 977 South Iceland Seismic Zone (SISZ), 428, 429, 435 South Korea, 545, 546 South Orkney Islands, Antarctica, 10, 11, 17, 18, 66–67 South Pacific geology, 951 literature and, 763 missionaries and, 635, 636 South Pacific Superswell, 554

INDEX

1067

South Sandwich Islands, Antarctica, 10, 66–67 South Sea Islands, 42, 761 South Shetland Islands, Antarctica, 10, 11, 17, 18, 485 southwest Pacific, 695 Soya current, 521 space, ecological. See also isolation adaptive radiation and, 1 caves and, 152 continental islands and, 183 convergence and, 191 coral and, 200 dispersal and, 225, 226, 228 diversity and, 13 rivers vs. lakes, 169 space, open, 479 Spain, 121, 148 Spanish explorers, 203, 729 sparrows, 144, 159 spatial considerations, 529–530, 749, 751, 766, 809, 820, 824. See also isolation; size of islands; species-area relationship (SAR) spawning areas, 607 specialist species. See also ecomorphs Barro Colorado insects, 90 dispersal and, 225 ecological release and, 252, 253 extinction and, 283, 285, 788 Galápagos birds, 110 insects, 461 inselbergs and, 469 Krakatau, 519 New Zealand alpine, 667 orchids, 697 overviews, 461 Seychelles, 831 whale falls and, 700 speciation. See also adaptive radiation; anagenesis adaptive radiation and, 5, 7 Azorean, 73 Cape Verde, 146 caves and, 151, 153 cichlids, of, 165–166 distance and, 772 founder effects and, 327 freshwater species, 347 genetic propensities to, 168–169 Greek Islands, 390 Hawaiian Islands, 399 isolation and, 108 mechanisms of, 166–167 New Zealand, 669 overviews, 755 rapid, 3 sexual selection and, 826, 827 temporal separation and, 209 species-area relationship (SAR), 73, 527, 562, 688, 857–861, 943 species boundaries, 725 species pump, 747, 841, 842 species richness biogeography and, 486–487 elevation and, 698 equilibrium theory and, 384

1068

INDEX

extinction and, 284 Hawaiian, 711 island rule and, 494 lakes and, 526 Madagascar, 577, 578 oases and, 688 orchids, 699 Philippines, 724 seamounts and, 820 succession and, 877 species-sorting, 527, 528, 530, 531 Speke, John, 984 spiders, 861–864 arctic, 50, 51, 52, 53 Australian, 671 Barro Colorado, 90 Bermuda, 98 Canary Islands, 130, 131, 774 Cape Verde, 145 Channel Islands (California), 159 convergence and, 188, 191 dispersal, 861–862 Hawaiian, 5, 7, 257, 531, 548 Hawaiian convergence, 189 Hawaiian Drosophila and, 233 kı-puka and, 512 Lord Howe Island, 570 Madagascar, 580 New Zealand, 671 Palau, 716–717 predators, 235 sexual selection and, 6, 826 sky islands, 842 Socotra, 848, 849 species-area relationship and, 860–861 Sri Lanka, 868 St. Helena, 872, 873 Surtsey Island, 886, 887 Tasmania, 907 spiderworts, 132 Spiller, 860 spinebills, 907 spiny bush, 578, 579 spiritualism, 965 spit elongation, 87 Spitsbergen Island, 58, 172, 423, 865–866, 879, 976 sponges, 52, 819, 820, 909, 911 spoonbills, 902 Sporades, 626 spots, 110 spreading. See also plate divergence (spreading) continental islands and, 707 Easter Island and, 709 Galápagos and, 710 Mediterraean, 624 New Guinea, 665 New Zealand and, 678 overviews, 703 seamounts and, 714, 821, 822 spreading, ridge, 854, 856 spreading, sea floor, 847, 856 springtails Antarctic, 12, 13 arctic, 48, 51, 52, 53

climate change and, 15 temperatures and, 14 Spruce, Richard, 963 spurges, 653 squirrels, 91, 119, 121, 125, 126, 132, 472, 793–794, 981 Sri Lanka, 346, 349, 445, 755, 866–869 St. Anne, 832 St. Barthelemy, 23, 31 St. Brendan the Navigator, 978 St. Christopher, 23 St. Croix, U.S. Virgin Islands, 92 St. Eustatius, 23, 24, 31, 32 St. Helena, 789, 870–873 Drosophila, 234 endangered species, 285 endemism and, 256 flightless birds, 316 forests, 873 introduced species and, 470, 959 island rule and, 492 rainfall, 173 research stations, 789 volcanism, 959 whaling, 976 St. Helena (Australia), 769 St. Helena Island, 66 St. Helens, Mount, 543 St. Jago (Santiago), 217, 218 Darwin and, 956 volcanism, 959 St. John, 395 St. Kilda, 121, 123–124, 126, 635 map, 116 St. Kitts, 24, 31, 32, 34 St. Kitts Bank, 23 St. Lawrence Island, 48, 51 St. Lucia, 24, 31, 32, 35 St. Martin, 35 St. Martin Bank, 22, 24, 31 St. Paul Fracture Zone, 66 St. Paul Island, 441, 789, 795 St. Paul/New Amsterdam archipelago, 11 St. Paul plateau, 441 St. Paul’s Rocks, 218, 956 St. Vincent, 24, 31, 32, 35 stages of life, 226 Standley, Paul, 89–90 Stanley, Henry Morton, 984 starfish, 910–911 starlings, 800 static equilibrium, 489 Steadman, David, 314, 320, 594, 595 steep-sidedness, 695–696, 754, 755, 783, 823, 871, 952, 953 Stephens Island, 283, 673 stepping-stone islands, 384, 767 stereotypes, 761 Stevens, Samual, 963 stickleback fish, 4, 5, 124, 873–877 stochasticity. See chance (stochastic) events Stokes, Pringle, 917, 955 stonechats, 129, 131 stonecrop, 157 Stóra Dimun Island, 288

storms. See also specific types of storms climate change and, 171 Comoros, 178 coral and, 200 erosion and, 262, 263 freshwater species and, 344, 345 frog dispersal and, 350 global warming and, 379–380 Great Barrier Reef, 385 island arcs and, 705 motu and, 641, 642 Tierra del Fuego, 918 waves and, 779–880 Strait of Sicily islands, 625 stratigraphy, 522–523 streams, 94 Strecker, Angela, 531 stress tolerance, 877 Streymoy Island, 292 strike-slip faults (transform convergence), 454, 661, 693, 732, 733 stromatolites, 56, 797 stygobionts, 151, 152, 153 sub-Antarctic, 15 subantarctic region, 14, 359 subduction. See also specific plates Caroline Islands and, 149 Channel Islands (California) and, 161, 164, 162 cold seeps and, 175 Cyprus, 215, 216 Fiji and, 306 Indian Ocean region, 437 Indonesia and, 454, 456, 457, 458, 459 initiation, 484, 485 island arcs and, 702 Japan and, 502, 503, 504 Kurile Islands and, 523 lava tubes and, 545 Macquarie Island and, 576 Marianas and, 598 Mediterraean, 625, 626 New Caledonia, 648 New Caledonia metamorphic rock and, 648 New Guinea, 664–665 New Zealand, 676, 678 overviews, 424, 481–486, 691–692, 754 Philippines and, 733 seamounts and, 713 Solomons, 856 Vancouver and, 937 subduction-related islands, 17–19, 22, 30, 31, 112, 113. See also island arcs subpopulations, 766 subsidence atolls and, 69 Darwin and, 220 Darwin on, 219 Enewetak and, 219 erosion and, 261 granitic islands and, 381 Hawaiian Islands, 397, 408 Indian Ocean region, 440 Maldives and, 587 overviews, 484, 694, 695

sea levels and, 695 Uniformitarianism and, 218 volcanic islands and, 953–954 subterranean species. See caves succession, 877–879 Krakatau, 518, 519 Mediterranean forests, 628 organic falls and, 700 plant disease and, 750, 750 rain forests and, 945–946 Surtsey Island, 886, 887 succession, geologic, 650 succulents, 468, 469 Sudan, 146 Sulawesi. See also Wallacea Borneo and, 115 climate, 971 distribution and, 970 fauna, 553 freshwater species, 346 frogs, 349 geology, 457, 458, 459, 969 macaques, 777, 778 mammals, 776 New Guinea and, 660 overviews, 114, 446, 452 Philippines and, 729 shape and climate, 172 tarsiers, 549 Wallace and, 968 sulfophilics, 700 sulfur, 176, 524 Sulu Sea, 756, 757 Sumatra biodiversity, 113 Borneo biota and, 114 climate, 971 earthquakes, 242 endangered species, 450 endemism, 450 geology, 439, 448, 456–457, 458 Krakatau and, 517 macaques, 778 mammals, 553, 776 mountain floras, 114 overviews, 112 snakes, 845 tsunami, 933, 935 Sumatran fault, 454 Sumbawa island, 270, 454, 484 Sunda. See also Wallacea age of, 690 diversity, 450 geology, 455–456, 457, 458, 459 human colonization and, 720 volcanism, 439 Wallace and, 968 Sundaland plate, 732 Sunda Shelf, 112–113, 448, 455, 724, 729, 776, 777 Sunderbans, 380 sunflowers, 132, 157, 359, 362, 363, 835 sunflower star fish, 159 sunken continents, 673–674, 679 surface water, 421

surf in the tropics, 879–883 Surtsey eruption, 63–64, 429, 432, 491, 511, 545, 883–885 Surtsey Island biology, 879, 885–888 sustainability, 888–896. See also conservation overviews, 789 Pitcairn and, 745 rain forests and, 946 Seychelles, 832 successes, 784 vegetation and, 944 Suwarrow, 195 Svalbard archipelago, 48–49, 50, 51, 53, 54, 55, 56, 57, 58, 59, 60, 976 Swahili, 983 swans, 122, 123, 796 swash bars, 85, 86 Sweden, 122, 750–751, 879 sweepstakes routes, 36 sweet potatoes, 246, 723, 760 swells, topographic, 340 swiftlets, 182 swifts, 144 Swinhoe, Robert, 900 Sylvianoris, 325 sympatric speciation amphibians, 27 birds, 6 Galápagos, 352, 355, 773 insect, 461, 462 lakes and, 166, 169, 531 lizards, 561, 562, 563 mammals and, 589 Philippines, 729 rats, 589 sexual selection and, 828 silverswords, 837 snails, 672 spiders, 864 Syowa station, East Antarctica, 14 Syzygium, 940 Tachigali, 89 Tahiti adaptive radiation, insects, 712 Banks and, 219 biocontrol and, 101 Cook Islands and, 195 Darwin and, 218, 958 diseases, 977 endemics, 335, 337 forests, 222 freshwater species, 345 geology, 341, 341, 585, 712 humans and, 42 insect diversity, 463 invasive species, 478, 479 missionaries and, 634, 636, 637 ornamental plants, 275 popular culture and, 763 population of, 337 Rapanui and, 250 snails, 539, 540, 541 tides, 915 volcanoes, 333

INDEX

1069

Taiaro, 334, 337 Taiwan, 897–904. See also continental islands; Old World tropics Borneo and, 114–115 climate, 971 fishing, 310 human colonization, 721 humans and, 43 invasive species, 346, 347 landslides, 535 seafaring and, 277 takahes, 184 Takapoto, 334 Takutea Island, 192, 193, 197 Tamang Bank, 149 tambalacoque trees, 619 Tambora eruption, 263, 267–270, 439, 454, 484 Tam Tam Island, 782–783 tanagers, 26–27 tardigrade (water bear) species, 12, 14 target effect, 488, 489 tarsiers, 549–553, 731 tarweeds, 78, 157, 837, 838. See also silverswords Tasman, Abel, 672, 908, 919 Tasmania, 904–909 Darwin and, 958, 959 freshwater species and, 345 hydroclimate, 423 overviews, 181, 183 spiders, 862 Tasmanians, 114 Tasmanian Sea, 678 Tatoosh, 909–912 Taupo Volcanic Zone (TVZ), 675, 676 Taveuni Island, 300, 301, 302, 309 taxon cycles, 5, 38, 253, 912–913 taxonomic diversity, 889–890 Taylor, S.R., 485 Taymyr Peninsula, 51 tea trees, 797 technology, 791 tectonism. See plate tectonics television, 762 temperatures. See also climate change; global warming caterpillars and, 15 coral and, 200, 202, 565 diversity and, 13, 14 Drosophila and, 235 elevation and, 173 endemism and, 257 extinctions and, 171 extreme, 578, 579, 656–657 flightlessness and, 317 fluctuations, 555 Holocene, 433 insects and, 227 overviews, 173, 421 seabirds and, 81 Templeton, Alan, 233, 327 temporal separation, 209 Tenerife adaptive radiation, 774 birds, 314

1070

INDEX

lizards, 558, 563 orchids, 698 overviews, 127, 138 pigs and goats and, 742 species, 129, 130, 131, 132 spiders, 863 volcanism, 65, 134, 135, 137–139 tenrecs, 179, 550, 580, 588, 589–590, 776 tephra, 80, 139, 222, 263, 264, 265, 288, 432, 433, 511, 599, 601, 676, 885, 952. See also ash, volcanic Surtsey Island, 885 tephrochronology, 433 tepuis, 182, 717, 718 Terceira Island, 71, 72, 959 Terceira Rift, 64 “Tereoboo, king of Owhyee, bringing presents to Captian Cook” (Webber), 759 terns, 62, 81, 125, 248, 556, 611, 811, 812, 813, 814 terrapins, 831 Testudinidae, 921 Tethys Ocean, 215, 392, 426, 624, 627 Tettigonidae, 670 Texas coast, 83, 87 Thailand, 347 Thasos, 395 thatch, 273–274 The Theory of Island Biogeography (Macarthur and Wilson), 772 Thera, 394 Thera eruption, 626 thermal expansion, 816 thistles, 227 tholeiites, 803, 804, 806–807 Thomas, Jeremy, 118 Thompson Glacier, 59 Thorarinsson, S., 433 Thornton, W.B., 206 Thorpe, Roger S, 563 thrashers, 81, 472 threatened species. See endangered and threatened species thrombolites, 797 thrushes, 26, 104, 122, 931 Thule people, 76–77, 976 Thurston Island, 17 Tiburón Island, 80, 81 ticks, 23, 104 tidepools, 181 tides, 914–917 arctic and, 61 barrier islands and, 83, 84, 85, 86, 87 beaches and, 91 Bermuda, 95 Britain and Ireland and, 125 Channel Islands (California), 155 marine lakes and, 603, 605 Mediterranean, 628 mussels and, 911 sea levels and, 815–816 Tierra del Fuego, 917–918, 955, 956–957 tigers, 869 Tikehau atoll, 68 Tikopia, 417

timber. See also logging Borneo and, 115 Canary Islands, 130 ethnobotony and, 274 Pitcairn, 745 Vancouver, 938 time, dispersal and, 225 time dwarfs, 495 Timor, 459 Tinbergen, Niko, 876 Tinian, 594, 596, 597, 598, 599, 602, 952 Tip, Tippo, 984 Tiputa, 642 Tiree, 116 Tjörnes Fracture Zone (TFZ), 428, 429, 435 toads, 27, 118, 122, 477, 478, 845, 981 Toba eruptions, 455 Tobago, 23, 35 Tobi Island, 149 Tofua, 484 Tokelau, 256, 273, 333, 637, 705, 714–715 Tokelau seamount chain, 714 tomatoes, 125 Tonga, 918–921 bats, 194 birds, 303, 304 Cook Islands and, 195 ethnobotony and, 274 extinctions, 801 Fiji and, 299 geology, 306, 483, 485, 491, 690, 692, 702 Lapita and, 43 missionaries and, 634, 635, 637, 638 overviews, 707 pigs, 742 plants, 273 plate tectonics, 691 watercraft, 759 whaling, 976 Tonga-Kermadec subduction zone, 714 Tonga-Kermadec Trench, 692 tortoises, 920–926 Bermuda, 96 Canary Islands, 131 dispersal, 922 dispersal methods, 776 extinct, 231 Galápagos, 359, 361–362, 364–365, 372, 710, 958 Galápagos endemism and, 257 Mascarene Islands, 614, 615, 617, 619 Seychelles, 830–831, 832–833 Tortuga Island, 82 tourism. See also ecotourism advertising and, 765 Antarctic, 14 arctic, 54 atolls and, 68 Azorean, 74 Baja California Islands, 78 beach renourishment and, 262 Bermuda and, 97 Channel Islands (California), 155 Cook Islands, 195, 196 coral reefs and, 781

Cozumel, 205 Frazer Island, 332 Galápagos, 356 Line Islands and, 555 Lord Howe Island, 571 Madeira, 584 Midway Island, 633 Pitcairn, 746 prisons and, 768, 769, 771 Rapanui, 250, 251 Rottnest Island, 796 Seychelles, 832, 833 Spitsbergen, 866 Tonga, 919 Vancouver, 938 western Pacific, 715 whales and, 978 towhees, 159 Townsend’s shearwater, 80 trachytics, 66 trade ancient, 44–45 Bermuda and, 97 bird disease and, 104, 105 Borneo and, 115 Cook Islands and, 195 dodo and, 229 Farallon Islands and, 294 Gulf of Guinea and, 810 invasive biology and, 476–477, 815 invasive species and, 346, 478–479 Kurile Islands and, 525 lakes and, 531 pigs and goats and, 742 plant disease and, 748 Rapanui, 249–250 Seychelles and, 832 St. Helena and, 871 sustainability and, 892 Tristan da Cunha/Gough and, 931 Zanzibar and, 984 trade winds Canary Islands and, 128, 131 Cape Verde and, 144, 146 climate and, 172 climate change and, 174 clouds and, 172 Cook Islands and, 193 Hawaii and, 173 heat and, 173 landscape evolution and, 695 Vanuatu and, 939–940 traditional practices, 609, 782–783, 784, 850, 890, 894–895, 936, 947. See also ethnobiodiversity Tragedy of the Commons, 784 Traill, Greenland, 60 transform faults, 753, 754, 755, 802 transform plate boundaries, 703 transform plate-boundary islands, 690, 692–693 Trans-Hudson Orogeny, 77 transilience, 233 translocation, 120 transported landscapes, 415–416

tree ferns, 13 treefrogs, 654, 655 trees. See also conifers; deforestation; forests; specific trees Antarctic, 13 Antilles, 27 ants and, 40 arctic, 58 Ascension, 62 Atlantic island, 72 Baja California Islands, 78, 80, 81 Barro Colorado, 91 Barro Colorado coevolution, 88–90 Britain and Ireland and, 118, 125, 126 Cape Verde, 145 Comoros, 178 Cook Islands, 194 Dominican Republic,of, 23 Gough Island, 931 New Guinea insects and, 656 New Zealand, 667 Pitcairn, 745 Rapanui, 245 Rottnest Island, 796 Samoa invasives, 801 Socotra, 848, 850 Solomons, 852 succession and, 877 Tristan da Cunha, 931 Vancouver, 938 Vanuatu, 941 Trematolobelia, 2, 3, 4, 400 trenches, 691, 702, 704, 854. See also specific trenches Las Tres Marías, 79 trilobites, 651 Trindad Island, 23, 26, 35, 36, 65–66, 278, 298 Trinidad and Tobago, 926–929 Trinity Peninsula group, 18 Tristan da Cuhna archipelago, 929–932 anagenesis, 9 biotas, 11 birds, 108 Drosophila and, 234 geology, 66, 690, 693, 951 research stations, 789 Tristan/Gough plume, 66 Tristan (volcano), 66 Trobriand Islands, 664 troglobionts, 151, 152, 863 Troodos Range, 213–214, 216 trophic structure, 385–386 tropical Atlantic, 814 tropical Pacific, 814, 889 Tropical Pacific Decadal Variability (TPDV), 345 tropical wilderness areas, 453 tropicbirds, 556, 811, 812 trout, 124 Trouvador, 834 Truman, Harry, 682 tsetse flies, 474 tsunamis, 933–936 2004, 439 earthquakes and, 240–244, 695

geology, 755 Hawaiian, 408–409, 711 Kick ‘em Jenny and, 511 Kurile Islands and, 522, 524 landslides and, 537, 954 Mediterranean, 626 overviews, 484, 691 Solomon Islands, 706, 856 Tuamotu Archipelago, 46, 68, 333, 335, 336, 337, 340, 492, 553, 642, 711, 950 Tuamotu-Gambier archipelago, 334 Tuamotu plateau, 341, 712 tuataras, 184, 670, 786 tube worms, 174, 176, 424, 425, 426 tuna, 311, 338, 812, 820, 832, 891 tundra, 48, 50, 51, 52, 53, 60, 77, 78, 155, 291, 524, 657, 866 Tunisia, 687, 688 Tupaia, 758 Turkish subplate, 626 turnover of species ants, 38, 41 Krakatau, 519 New Zealand, 670 rates, 488, 489 vegetation, 944 Turtle Island, 22 turtles Ascension, 62 Baja California Islands, 80 beaches and, 91, 92 beach renourishment and, 262 Bermuda, 96, 97 Borneo, 114 Canary Islands, 131–132 Cape Verde, 145, 146, 147 Cook Islands, 194, 197 Fernando de Noronha archipelago, 298 Great Barrier Reef, 383, 385 Indonesia, 452 island rule and, 493 Marshall Islands, 611 Mauritius, 229 Midway Island, 632 Pelagie Islands endangered, 625 Philippines, 724, 726 Pitcairn, 745 Samoa, 800, 801 seamounts and, 820 Seychelles, 831 Socotra, 850 Solomons, 853 Tonga, 920 Trinidad and Tobago, 929 whaling and, 977 Zanzibar, 984 Tuscan Islands, 623, 625, 857 tussocks, 14, 52, 108, 184, 401, 471, 574, 667, 798, 867, 905, 929, 931, 932 Tuvalu, 256, 333, 420, 591, 641, 681, 693, 759 typhoons, 418–420, 709, 814, 898, 904 Tyrrhenian Sea islands, 625 Uist, 116, 123–124 Ullung Island, Korea, 8, 9

INDEX

1071

ultraviolet radiation, 14, 15, 151, 258 ungulates, 132. See also specific ungulates Unified Neutral Theory of Biodiversity and Biogeography (Hubbell), 488 Uniformitarianism, 218 United States Caroline Islands and, 148 caves, 150, 151, 152 coral conservation, 567, 780, 782 coral reefs and, 783 introduced species, 473 mosquitoes and, 977 shipwrecks and, 834 snails, 542 uplifts coseismic, 694 Darwin and, 220 glaciers and, 973 Indian Ocean region, 439 Kurile Islands, 522 Macquarie Island and, 576–577 Madagascar and, 578 Makatea Islands and, 585 Marianas and, 594, 599 Mediterranean, 624 New Caledonia, 648 Newfoundland, 651 New Guinea, 657, 665 New Zealand, 675, 676 overviews, 491–492, 691–692, 694 Philippines, 736 sky islands and, 839 Socotra, 847 Solomons, 692, 704, 706 Taiwan, 903 Tasmania, 958 Vancouver, 937 Zanzibar and, 983 Uracas (Farallon de Pajaros), 599, 600 Urals, 52, 57 Urup Island, 522 d’Urville, Dumont, 42, 720 Utirik, 683 vagrants (birds), 12, 122, 124, 126, 489 Vaigach Island, 51 Vancouver, 333, 375, 423, 704, 708, 874, 937–938 Vancouver, George, 937 vangas, 579 van Riper, Charles, 414 Vanua Levu, 299, 300, 301, 302, 304, 305–306, 308, 309 Vanuatu, 306, 380, 491, 834, 854, 936, 939–941 Van Valen, Lee, 493 Vatnajökull glacier, 64, 435 vectors, 103, 478, 527, 749 vegetation, 941–947. See also forests; plants Antarctic, 13–14 arctic, 49 Atlantic, 72 barrier islands and, 83, 85 beaches and, 91 climate and, 85–86

1072

INDEX

climate change and, 44 Cook Islands, 193 diversity and, 186 endangered, 643 extinction and, 282 Gough Island, 930–931 Japan, 500 Midway Island, 632–633 motu and, 642 New Guinea, 657 New Zealand, 707 oases, 687 overviews, 171 pigs and goats and, 743 Samoan, 800 Spitsbergen, 866 Surtsey Island, 887 Taiwan, 898 Tonga, 918 Tristan da Cunha, 930–931 volcanism and, 940 vegetative reproduction, 697 Venezuela, 21, 30, 35, 36, 71, 839, 840, 926–927 vents, hydrothermal, 424–427, 425, 700, 974 Verbeek, R.D.M., 517 vertebrates Antarctic fossils, 19 Antilles fossils, 20, 22 ants and, 41 Azorean, 72 biocontrol and, 100 Canary Islands, 129, 131 Cape Verde, 144 Cozumel, 205 distribution, 970 dwarfism and, 236 evolution of, 922 extinction and, 284 extinctions, 581, 908 food-limited populations, 121 founder events and, 6 humans and, 3 Indonesia, 448, 453 Madeira, 584 Marianas and, 594 New Guinea, 653–654, 656 Philippines, 724, 729 rats and, 794–795 Rottnest Island, 798 Socotra, 849 Tonga, 918, 920 Trinidad and Tobago, 927 Vancouver, 938 world, 654 Vespucci, Amerigo, 297 Vestfold Hills, Antarctica, 10 Vestiges of the Natural History of Creation, (anonymous), 963 de Veuster, Damien, 635 vicariance, 947–950 Antilles, 28 dispersal vs., 560, 561 diversity and, 182 fragments and, 187 Gondwana and, 669, 670

Madagascar, 580 overviews, 970 reefs and, 779 snails and, 539 Socotra, 849 Victoria Island, 50 Victoria Land, Antarctica, 13 video games, 765 Vidoy Island, 290 Vieques Island, 22 Vikings, 291–293 village weaverbirds, 108–109 de Villalobos, Ruiz Lopez, 716 vireos, 96 Virgin Islands, 22, 92 viruses, 121–122, 891 visual signaling, 211 Viti Levu, 299, 300, 301, 302, 303, 304, 305, 305–306, 307, 308, 319, 322, 325–326 Vitousek, 945 de Vlamingh, Willem, 796 vocalizations. See acoustical signaling volcanic arcs. See also island arcs collisions with continents, 902, 903 geology, 755 magma and, 952 overviews, 953 Volcanic Caribbees, 29, 31 volcanic islands, 950–954. See also island arcs; volcanic arcs age of, 825 geology, 951 Gough Island, 929 growth of, 953 landslides, 536–537 lava tubes and, 544–545 overviews, 693–694 Pacific region, 704, 708 Samoa, 799–800 Seychelles, 829 shield volcanoes and, 952 Taiwan, 897 Tristand da Cunha, 929 volcanism. See also ash, volcanic; hotspot islands; island arcs; kı-puka; lava; mantle plumes; seamounts; specific eruptions; specific volcanoes Antarctic, 11, 17, 19–20 Antilles, 22, 29, 30, 31, 32–35 ants and, 38 arctic, 50, 55, 56, 58 Ascension and, 61 Atlantic, 58, 63–66, 67, 71–72 Azorean, 74 Baffin Island, 77 Baja California Islands, 78, 79, 80, 81 beaches and, 92, 93, 93, 94 Bermuda and, 95 Borneo, 112 Britain and Ireland, 125 Canary Islands, 127, 133–142 Cape Verdes, 144, 218 Caroline Islands, 149 Channel Islands (British), 154 Channel Islands (California), 162, 163, 164

climate effects, 266–267 Comoros, 178 Cook Islands, 191–193 coral and, 200, 956 Cyprus, 214 Darwin and, 217, 218, 219–220, 958, 959 deforestation and, 222 dispersal/vicariance and, 948 diversification and, 464 earthquakes and, 240–241 endemism and, 254 ephemeral islands and, 260 extinctions and, 500 Faroe (Faeroe) Islands and, 288 Fernando de Noronha archipelago, 297, 298 Fiji, 305 French Polynesia, 338–343 freshwater species and, 344, 345 geothermics and, 435 global warming and, 376 Greek Islands and, 388, 389, 393, 394, 395–396 Gulf of Guinea, 808, 809 Hawaiian Islands, 397, 404–408 Iceland, 430–434, 433, 434, 435 Indian Ocean region, 437–446, 441, 442–444 Indonesia, 454–455, 458, 459 isolation and, 187 Japan, 501, 502 Kurile Islands, 521, 522, 523–524 Lesser Antilles, 510 Lord Howe Island, 568, 569 Macquarie Island and, 576 Madagascar, 177, 578 Makatea Islands, 585 Marianas and, 593–594, 598–599, 706 Mascarene Islands, 612, 620–622 Mediterranean, 624, 625, 626, 627 navigation and, 280 New Caledonia, 648 Newfoundland, 649, 650, 651, 652 New Guinea, 661, 663, 664 New Zealand, 666, 674, 675, 676, 677, 679 North Atlantic, 652 northeast Pacific, 705 ocean floor, 490–491 overviews, 219, 483, 753, 754, 755 Pacific region, 706, 707 Philippines, 725, 727, 727, 733–734, 734, 737 Rapanui and, 243 Samoa, 802, 803, 804–808 seamounts and, 821, 822–823, 824–825 sky islands and, 840 Solomons, 855, 856 spiders and, 862 St. Helena, 870–871 Surtsey Island and, 484 Taiwan, 898 tectonism and, 339 Tonga, 919 Tristan da Cunha eruptions, 931 Uniformitarianism and, 218 Vancouver and, 937 Vanuatu, 939 voles, 49, 52, 53, 120, 126

“Voyage Up the Amazon” (Edwards), 963 Vulcão de Paredão, 65 vultures, 181 Wadati-Benioff zone, 32, 484–485 Wadati, K., 481 Waipounamu Erosion Surface (WES), 680 walking stick insects, 6 wallabies, 182 Wallace, Alfred Russel, 114, 220, 446, 742, 870, 962–967. See also Wallace’s Line Wallacea, 114, 446, 447, 448, 450, 452, 453, 720, 814, 968 Wallace effect, 966 Wallace’s Line, 114, 447, 458, 656, 723, 729, 776, 899, 964 Wallis Island, 953 walruses, 76, 78, 866 Walvis Ridge, 951 warblers Antilles, 26 Canary Islands, 131 Cape Verde, 144 Channel Islands (California), 159 continental islands, 186 Cook Islands, 194 Cozumel, 204 Fiji, 303 French Polynesia, 335 Galápagos, 352, 353, 354, 360 Madagascar, 106 Marianas, 597 New Guinea, 657 oases, 687 Pitcairn, 745 Rodrigues, 107 Seychelles, 833 Warming Island, 971–973 Warner, Richard, 414 washover fans/terraces, 83, 84, 85 wasps, 23, 88–89, 90, 100, 102, 123, 672, 927, 929 water, 172, 409, 749. See also groundwater; tides; waves water availability. See also hydrology aliens species and, 99 arctic, 51 Ascension, 62 atolls, on, 68 Cape Verde, 146 Channel Islands (British), 155 climate change and, 15 Cook Islands, 192, 193 coral reefs and, 782 Cozumel, 205 diversity and, 13, 14 factors in, 201 forests and, 128 freshwater species and, 343 geology, 485 Hawaii, 409 inselbergs and, 466, 467–468 Kurile Islands, 521 lava and, 543, 547, 548 Madagascar, 579

oases, 687, 688, 689 seabirds and, 812 silverswords and, 836 weaverbirds and, 109–110 watercraft. See also Kon-Tiki; shipwrecks ancient, 277, 278, 279–280 Austronesian, 721 human colonization and, 720 Polynesian, 722, 723, 758–759 Satawal traditional, 783 whaling, 975, 976, 977 water density, 421 waterfowl, 124, 323, 325, 326. See also seabirds water movement, 13, 14, 131, 200. See also cold seeps Watling, Dick, 299, 302, 304 Watson, 859 wattle, 797 waves. See also storms; surf in the tropics; tsunamis barrier islands and, 83, 85, 87 beaches and, 91, 93, 262 climate and, 86 ephemeral islands and, 259 erosion and, 261, 491, 885 freshwater species and, 345 Hawaii and, 409 intertidal habitats and, 909, 910, 912 Krakatau and, 517 Maldives and, 587 navigation and, 280 reefs and, 386 sea levels and, 916 shorelines and, 93 tsunami, 933–934 weakness of islands, 473 weather, 11, 92, 93, 193. See also climate; rain; temperatures weaverbirds, 108–109, 110 Webber, John, 759 Weber’s Line, 968, 970 weeds, 100, 330, 479 weevils adaptive radiation and, 2, 3 arctic, 51 Barro Colorado, 89 Bermuda, 97 biocontrol and, 101 Canary Islands, 130 French Polynesian endemics, 335 Galápagos, 361, 363, 368 Gough Island, 931 New Zealand, 671 rafting and, 775 Rapa, 461 single-island radiations, 462 Southern Ocean Islands and, 12–13 St. Helena, 871 Tristan da Cunha, 931 Wegener, Alfred, 702, 710 Weichselian glaciation, 59 Weinmannia, 940 Weisler, M.I., 45 wekiu bugs, 402 Wellington, 977

INDEX

1073

west Africa, 651 West Bismarck Arc, 485 West-Eberhard, Mary Jane, 826 Western Ghats, 840 Western Isles (Outer Hebrides), 116, 120, 123–124 western Pacific, 755, 825 West Indies ants and, 40 bird radiation, 106 birds, 321 frogs, 348, 349 introduced species, 471 lizards, 828 mammals, 321, 590 snakes, 844, 845 species-area relationship and, 857, 860 vicariance, 950 whales, 975, 976, 978 West Nile virus, 105, 479 west Pacific, 715 West Pacific Seamount Province (WPSP), 705, 715 West Virginia caves, 151 weta, 211, 285, 318, 667, 668, 669, 670, 671, 672 wet forests and bogs, 331, 401–402, 842, 870, 907 wetlands Britain and Ireland and, 125 Cape Verde, 147 conservation, 896 French Polynesian, 335 global warming and, 377–378 Greek Islands, 391 Midway Island, 632 Samoa, 800 Solomons, 851 Zanzibar, 983 whale falls, 973–975, 978 whales, 975–978 Antarctic, 195 Baffin Island, 77, 78 Bermuda, 97 Britain and Ireland and, 125 Canary Islands, 132 Cape Verde, 145 Channel Islands (California), 159–160 cold seeps and, 175 falls, 700 Farallon Islands, 296 human impacts, 14 Kurile Islands, 525 Line Islands, 557 Rottnest Island, 798 Samoa, 800 Socotra, 850 Solomons, 853 Surtsey Island, 887 Tonga, 920 Vancouver, 938 wintering areas, 22 Zanzibar, 984 whaling, 975–978 Antarctic, 15, 54 Beagle and, 956

1074

INDEX

Bermuda, 97 Farallon Islands, 296 Galápagos and, 710 pigs and goats and, 742 Pitcairn, 745 Rottnest Island, 798 Seychelles, 832 Solomons and, 853 Spitsbergen and, 866 Tristan da Cunha, 931 Whataroa virus, 104–105 wheelstamen tree, 899 white-eyes, 617 White Island, 423, 674 White, Philo, 918 White, Thomas, 124 Whitlock, M.C., 766 Whitsunday Islands, 387 wild cats, 120 Williams, Ernst, 562 Wilson, Edward O., 38, 253, 271, 629, 772, 774, 912. See also Equilibrium Theory of Island Biogeography Wilson, Henry, 716 Wilson, J. Tuzo, 704, 711 wind ants and, 36, 37 arctic, 49, 52, 61 Ascension and, 61 Azores and, 72 Bermuda and, 95 Borneo biota and, 114 Canary Islands and, 127 climate and, 172 climate change and, 15 dispersal and, 227, 368 diversity and, 13 erosion, 885 flightlessness and, 317 freshwater species and, 344, 345 Line Islands and, 555 nutrients and, 295 plant disease and, 749 reefs and, 68 seafaring and, 279 shorelines and, 92–93 size of island and, 859 waves and, 880 window lakes, 331–332 Windward Islands, 23, 143 windward regions Fiji, 299 freshwater species and, 344 hydrology and, 421 lava and, 543 Rapanui, 245 reefs, 556 Samoa, 799 Vanuatu, 940 winglessness, 37, 130, 211, 318, 373, 375, 476, 499, 548 wings, 227, 812 wireweeds, 97 Wisconsin glaciation, 59, 83 within-island radiations, 772

Wizard Island, 979–981 wolverines, 120 wolves, 52, 78, 120 The Wonderful Century (Wallace), 966 wood-eating clams, 701 woodhens, 572 Woodlark Basin, 664, 665 woodpeckers, 21 Wootton, Timorth, 910 World War II intraplate islands and, 708 Marshall Islands and, 712 movies, 762 nuclear weapons, 706 Philippines and, 730 Phosphate Islands and, 739 prisons and, 771 rats and, 793 Rottnest Island and, 796–797 shipwrecks, 834 snakes and, 845 Tristan da Cunha/Gough and, 931 Vanuatu and, 941 worms Antarctic, 15 Antarctic nematode, 12 cold seeps and, 174 Cook Islands, 194 flat, 124 palm, 425 Philippines, 724–725 Surtsey Island, 886, 887 sustainability and, 891 tube, 176 whale falls and, 974 Wrangel Island, 48, 49, 51, 59, 423 Wrangellia, 708, 937 wrens, 159, 322, 671, 672, 907 Wright, 209, 860 Wright, S. Joseph, 90 Wright, Sewall, 436, 766 Wright’s island model, 766 xenoliths, 31, 805 Yap, FSM, 784 Yap Islands, 94, 148, 149, 277, 722, 782, 783, 784 yellow crazy ants, 39, 472, 479, 534–535 Yemen, 100 Yttygran Island, 51 “zagoutis,” 23, 24 Zambia, 146 Zanzibar, 39, 445, 982–986 Zealandia, 322, 323, 325, 665, 666, 670, 673, 674, 677–680, 707 Zemlya Frantsa Iosifa, 59 Zetek, James, 88 Zola, Émile, 769 zoogeography, 965–966, 969 zooplankton, 295, 527, 528–529, 530, 531, 565, 820, 891 zooxanthellae, 780, 781, 782 zooxanthellate corals, 198, 199, 201, 202, 378

This page intentionally left blank

Spitsbergen

Arctic Region

ds

lan

Is rile

Ku

Mediterranean Greek Islands

Region

an

Jap

Cyprus

Maldives

Mari New Guinea

lom

on

Isla

nd

ia

s GrNew Caledonia ea t Re Ba ef rrie r

car agas Mad

s

Comoros

So

Indones

Krakatau

nd

Borneo

Isla

Caroline Islands

Sri Lanka

Seychelles

all

rsh

Zanzibar

Palau

es

São Tomé, Príncipe, and Annobon

pin

Socotra Archipelago

Ma

ilip

Ph

anas

Taiwan

Mascarene Islands

Indian Region Rottnest Island Tasmania

Vanuatu

Fraser Island Lord Howe Island nd ala e Z ew N Macquarie

Antarctic Region

fin

Warming Island

Isla

nd

Iceland Surtsey Faroe Islands

Britain Ireland Newfoundland Channel Islands

Vancouver Tatoosh

Wizard Island Azores

Farallon Islands Midway

Channel Islands

Haw

aiian

Islan

ds

Bermuda

B Ca aja lifo rn

Atlantic Region Cozumel

ia

Madeira Archipelago Canary Islands Cape Verde Islands

Antilles Kick 'em Jenny Trinidad and Tobago

Pacific Region Barro Colorado Island

Fernando de Noronha

Galápagos Islands

s

nd

sla

eI

Lin

u

Baf

Samoa

Ascension Fre

nc hP Fiji oly Cook ne Tonga Islands sia

St. Helena

Pitcairn Easter Island Tristan da Cunha

Juan Fernandez Islands

Gough Island

Tierra del Fuego