Biogeography of the West Indies: Patterns and Perspectives, Second Edition

BIOGEOGRAPHY of the WEST INDIES Patterns and Perspectives

S E C O N D

E D I T I O N

BIOGEOGRAPHY of the WEST INDIES Patterns and Perspectives

S E C O N D

E D I T I O N

Edited by

Charles A. Woods Florence E. Sergile

CRC Press Boca Raton London New York Washington, D.C.

Library of Congress Cataloging-in-Publication Data Biogeography of the West Indies : patterns and perspectives / edited by Charles A. Woods and Florence E. Sergile.—2nd ed. p. cm. Includes bibliographical references and index. ISBN 0-8493-2001-1 (alk. paper) 1. Biogeography—West Indies. 2. Natural history—West Indies. 3. Nature—Effect of human beings on—West Indies. I. Woods, Charles A. (Charles Arthur) II. Sergile, Florence E. QH109.A1 B56 2001 578′.09729—dc21

2001025275 CIP

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the authors and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. All rights reserved. Authorization to photocopy items for internal or personal use, or the personal or internal use of specific clients, may be granted by CRC Press LLC, provided that $1.50 per page photocopied is paid directly to Copyright clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA. The fee code for users of the Transactional Reporting Service is ISBN 0-8493-2001-1/01/$0.00+$1.50. The fee is subject to change without notice. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com © 2001 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0-8493-2001-1 Library of Congress Card Number 2001025275 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acid-free paper

Editors Charles A. Woods is Curator Emeritus at the Florida Museum of Natural History and Adjunct Professor of Zoology at the University of Florida and Adjunct Professor of Biology at the University of Vermont. He is the editor of Biogeography of the West Indies: Past, Present, and Future (1989) and Biodiversity of Pakistan (1997), and the author of many publications on various aspects of systematics, evolution, ecology, and biogeography. His interests in the West Indies have involved work on many islands, and he is currently completing a book on the mammals of the West Indies. He is also interested in the biogeography of sky islands, and has made many extended expeditions into the Himalayas and Hindu Kush and Karakoram ranges. He is completing a book on the mammals of the Western Himalayas. Dr. Woods currently lives on a farm in the mountains of northern Vermont. He is closely associated with the Vermont Leadership Center and the Bear Mountain Natural History Center. Florence E. Sergile is a visiting scientist at the Florida Museum of Natural History and the University of Florida. She is an agronomist with expertise in the fields of natural resource management and environmental education. She has spent much of her career working on the management of natural sites in Haiti. She has taught courses at the undergraduate and graduate levels in Haiti and has worked for USAID, the United Nations, and the World Bank. She was a consultant for the production of the National Environmental Action Plan (NEAP) for Haiti. During the past decade she also served as Director of Haiti-NET, a center for the conservation of the natural patrimony of Haiti. Ms. Sergile received a B.S. in agronomy from the State University of Haiti (Université d’Etat d’Haiti) in 1977 and an M.S. in Latin American studies with a major in natural resources management in 1990. She is pursuing a Ph.D. in environmental education at the University of Florida. One of her major interests is developing educational materials in the areas of conservation biology and environmental education for various target groups in Haiti and the West Indies as a whole.

Contributors Marc W. Allard Department of Biological Sciences The George Washington University Washington, D.C., U.S.A. [email protected]

Jason H. Curtis Department of Geological Sciences University of Florida Gainesville, Florida, U.S.A. [email protected]

Robert J. Asher Division of Vertebrate Paleontology American Museum of Natural History New York, New York, U.S.A. [email protected]

Daryl P. Domning Department of Anatomy Howard University Washington, D.C., U.S.A. [email protected]

Scott D. Baker Department of Biological Sciences The George Washington University Washington, D.C., U.S.A. [email protected]

Stephen K. Donovan Department of Paleontology The Natural History Museum London, U.K. [email protected]

Ross T. Bell Department of Biology University of Vermont Burlington, Vermont, U.S.A. [email protected]

Ginny L. Emerson Department of Biological Sciences The George Washington University Washington, D.C., U.S.A. [email protected]

Rafael Borroto Paéz Instituto de Ecología y Sistemática Havana, Cuba [email protected]

Julio A. Genaro Museo Nacional de Historia Natural Havana, Cuba [email protected]

Mark Brenner Department of Geological Sciences University of Florida Gainesville, Florida, U.S.A. [email protected] Jorge O. de la Cruz Grove Scientific & Engineering Orlando, Florida, U.S.A. [email protected]

Carla Ann Hass Department of Biology Pennsylvania State University University Park, Pennsylvania, U.S.A. [email protected] S. Blair Hedges Department of Biology Pennsylvania State University University Park, Pennsylvania, U.S.A. [email protected]

Donald B. Hoagland Department of Biology Westfield State College Westfield, Massachusetts, U.S.A. [email protected] David A. Hodell Department of Geological Sciences University of Florida Gainesville, Florida, U.S.A. [email protected] G. Roy Horst Department of Biology State University of New York Potsdam, New York, U.S.A. [email protected] Walter S. Judd Department of Botany and University of Florida Herbarium University of Florida Gainesville, Florida, U.S.A. [email protected]

Miriam Marmontel Sociedade Civil Mamirauá Tefé, AM, Brazil [email protected] Linda R. Maxson College of the Liberal Arts University of Iowa Iowa City, Iowa, U.S.A. [email protected] Brian K. McNab Department of Zoology University of Florida Gainesville, Florida, U.S.A. [email protected] Jacqueline Y. Miller Allyn Museum of Entomology Florida Museum of Natural History University of Florida Sarasota, Florida, U.S.A. [email protected]

C. William Kilpatrick Department of Biology University of Vermont Burlington, Vermont, U.S.A. [email protected]

Lee D. Miller Allyn Museum of Entomology Florida Museum of Natural History University of Florida Sarasota, Florida, U.S.A. [email protected]

Thomas H. Kunz Department of Biology Boston University Boston, Massachusetts, U.S.A. [email protected]

Gary S. Morgan New Mexico Museum of Natural History Albuquerque, New Mexico, U.S.A. [email protected]

Lynn W. Lefebvre U.S. Geological Survey Florida Caribbean Science Center Sirenia Project Gainesville, Florida, U.S.A. [email protected] Ross D. E. MacPhee Department of Mammalogy American Museum of Natural History New York, New York, U.S.A. [email protected]

Jose A. Ottenwalder Florida Museum of Natural History University of Florida Gainesville, Florida, U.S.A. [email protected] Roger W. Portell Division of Invertebrate Paleontology Florida Museum of Natural History University of Florida Gainesville, Florida, U.S.A. [email protected]

Pedro M. Pruna Goodgall Smithsonian Institution Academy of Sciences of Cuba Havana, Cuba [email protected]

Howard P. Whidden Department of Biology Augustana College Rock Island, Illinois, U.S.A. [email protected]

Galen B. Rathbun Department of Ornithology and Mammalogy California Academy of Sciences San Francisco, California, U.S.A. [email protected]

Jennifer L. White Department of Biology Augustana College Rock Island, Illinois, U.S.A. [email protected]

James P. Reid Biological Resources Division U.S. Geological Survey Florida Caribbean Science Center Gainesville, Florida, U.S.A. [email protected]

Laurie Wilkins Division of Mammalogy Florida Museum of Natural History Gainesville, Florida, U.S.A. [email protected]

Jonathan Reiskind Department of Zoology University of Florida Gainesville, Florida, U.S.A. [email protected]

Matthew I. Williams Department of Biological Sciences Auburn University Auburn, Alabama, U.S.A. [email protected]

Armando Rodríquez-Durán Department of Natural Sciences Inter-American University of Puerto Rico Bayamón, Puerto Rico [email protected]

Samuel M. Wilson Department of Anthropology University of Texas Austin, Texas, U.S.A. [email protected]

Florence E. Sergile Florida Museum of Natural History University of Florida Gainesville, Florida, U.S.A. [email protected] David W. Steadman Florida Museum of Natural History University of Florida Gainesville, Florida, U.S.A. [email protected] Ana E. Tejuca Museo Nacional de Historia Natural Havana, Cuba [email protected]

Elizabeth S. Wing Division of Zooarchaeology Florida Museum of Natural History Gainesville, Florida, U.S.A. [email protected] Charles A. Woods Florida Museum of Natural History University of Florida Gainesville, Florida, U.S.A. [email protected] and Bear Mountain Natural History Center Island Pond, Vermont, U.S.A. [email protected]

Acronyms and Abbreviations AMNH ATPPF BMNH BSP CM DPUH ECMU FLMNH FMNH GOH GUIA IES/ACC IGPACC IRSB ISPAN IZ JAO MARNDR MCZ MDE MNHNC MNHNH P MOE MPIH MPT MPUM mya NEAP NGO NMW NRM OA PSM rcyrbp RMNH ROUTE 2004 SEC SMF SPNS UF UMMZ UMZC UNDP USAID USNM UWI YPM ZMA ZMUH

American Museum of Natural History, New York Appui Technique à la Protection des Parcs et Forêts British Museum of Natural History Biodiversity Support Program Carnegie Museum of Natural History, Pittsburgh Departamento de Paleontología de la Universidad de la Habana, Escuela de Ciencias Biológicas Environmental Coordination and Monitoring Unit Florida Museum of Natural History, University of Florida, Gainesville Field Museum of Natural History Government of Haiti Geological Institute of the University of Amsterdam Instituto de Ecología y Sistemática, Academia de Ciencias de Cuba Instituto de Geología y Paleontología de la Academia de Ciencias de Cuba Institut Royal des Sciences Naturelles de Belgique, Bruxelles Institut de Sauvegarde du Patrimoine National Instituto de Zoología de la Academia de Ciencias de Cuba Jose A. Ottenwalder, private field collections, Santo Domingo Ministère de l’Agriculture, des Ressources Naturelles et du Développement Rural Museum of Comparative Zoology, Harvard University, Cambridge Ministère de l’Environnement Museo Nacional de Historia Natural, La Habana, Cuba Paleontological collection of the Museo Nacional de Historia Natural, La Habana, Cuba Ministry of the Environment Max-Planck-Institut für Hirnforschung, Frankfurt Most Parsimonious Tree Museum of Paleontology, University of Montana, Missoula Millions of years ago National Environmental Action Plan Nongovernmental organization Naturhistorisches Museum Wien, Wien Naturhistoriska Riksmusset, Stockholm Oscar Arredondo private collection, La Habana, Cuba Puget Sound Museum of Natural History, University of Puget Sound, Tacoma, Washington 14C years before present (i.e., before 1950, the radiocarbon datum) Rijksmuseum van Natuurlijke Historie, Leiden Project Aménagement de la Baie de Caracol à la Riviere du Massacre Sociedad Espeleológica de Cuba Forschungsinstitut und Natur-Museum Senckemberg Service des Parcs Nationaux et des Sites Naturels Florida Museum of Natural History, University of Florida, Gainesville Museum of Zoology, University of Michigan, Ann Arbor University Museum of Zoology, Cambridge, United Kingdom United Nations Development Program United States Agency for International Development National Museum of Natural History, Smithsonian Institution, Washington, D.C. Geology Museum, University of the West Indies, Mona, Jamaica Peabody Museum, Osteological Collection, Yale University Zoological Museum, Institute of Taxonomic Zoology, University of Amsterdam Zoologisches Institut und Zoologisches Museum, Universität Hamburg

Contents Chapter 1

Introduction and Historical Overview of Patterns of West Indian Biogeography ......1 Charles A. Woods

Historical Overview ...........................................................................................................................1 Acknowledgments ..............................................................................................................................6 Literature Cited ..................................................................................................................................8 Conservation Posters........................................................................................................................12 Environmental Education and Activity Books ................................................................................12 General Information.........................................................................................................................13 Conservation Exhibits ......................................................................................................................14

Chapter 2

Biogeography of the West Indies: An Overview .......................................................15 S. Blair Hedges

Introduction ......................................................................................................................................15 West Indian Biota.............................................................................................................................16 Geological History ...........................................................................................................................17 Overwater Dispersal.........................................................................................................................19 Proto-Antillean Vicariance...............................................................................................................21 The Land Bridge Model of MacPhee and Iturralde-Vinent............................................................22 Divergence Times ...................................................................................................................23 Number of Lineages Analyzed...................................................................................23 Mixture of Morphological and Immunological Data ................................................24 Taxa Are Not Discriminated in Terms of Interpretative Significance.......................24 Overrepresentation and Ambiguous Significance of Nonendemics ..........................25 Low Number of Nonendemic Lineages in the Greater Antilles ...............................25 Unknown Shaping Influence of Extinction................................................................25 Water Currents........................................................................................................................26 Inconsistencies and Problems in Model of MacPhee and Iturralde-Vinent..........................28 Evidence against a Mid-Cenozoic Land Bridge....................................................................29 Discussion and Conclusions ............................................................................................................29 Acknowledgments ............................................................................................................................30 Literature Cited ................................................................................................................................30

Chapter 3

Climate Change in the Circum-Caribbean (Late Pleistocene to Present) and Implications for Regional Biogeography............................................................35 Jason H. Curtis, Mark Brenner, and David A. Hodell

Introduction ......................................................................................................................................35 Using Oxygen Isotopes in Freshwater Carbonate Shells to Infer Past Climate.............................37 Determining the Timing of Climate Changes .................................................................................38 Late Pleistocene and Holocene Climate Change in the Circum-Caribbean...................................41 Late Pleistocene Aridity .........................................................................................................41 Early Lake Filling...................................................................................................................42

Earliest Holocene (~10,500 to ~8,500 14C yr BP) ..........................................................................43 Early to Middle Holocene (~8,500 to ~3,000 14C yr BP)...............................................................44 Late Holocene (~3,000 14C yr BP to the Present) ..........................................................................44 Summary of Circum-Caribbean Climate.........................................................................................45 Long-Term Climate Controls .................................................................................................46 Short-Term Climate Controls .................................................................................................48 Nonclimatic Controls..............................................................................................................48 Climate and Biogeography in the Circum-Caribbean.....................................................................49 Summary and Conclusions ..............................................................................................................50 Literature Cited ................................................................................................................................51

Chapter 4

Functional Adaptations to Island Life in the West Indies .........................................55 Brian K. McNab

Introduction ......................................................................................................................................55 The Adjustment of Vertebrates to Island Life .................................................................................56 Was the Fauna of the West Indies Resource Limited? ...................................................................58 Conclusion........................................................................................................................................60 Acknowledgments ............................................................................................................................60 Literature Cited ................................................................................................................................61

Chapter 5

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae) ...........................63 Walter S. Judd

Introduction ......................................................................................................................................63 Phylogenetic Relationships within Lyonia sect. Lyonia..................................................................65 Biogeographical Investigation..........................................................................................................68 Methods ..................................................................................................................................68 Results ....................................................................................................................................69 Discussion...............................................................................................................................70 Literature Cited ................................................................................................................................74

Chapter 6

Patterns of Endemism and Biogeography of Cuban Insects .....................................77 Julio A. Genaro and Ana E. Tejuca

Introduction ......................................................................................................................................77 Discussion ........................................................................................................................................77 Conclusions ......................................................................................................................................81 Acknowledgments ............................................................................................................................81 Literature Cited ................................................................................................................................81

Chapter 7

Patterns in the Biogeography of West Indian Ticks ..................................................85 Jorge O. de la Cruz

Introduction ......................................................................................................................................85 Materials and Methods.....................................................................................................................85

Host-Group Specificity...........................................................................................................86 Structural Niche......................................................................................................................86 Open Field ..................................................................................................................86 The Nest .....................................................................................................................86 The West Indies Ticks......................................................................................................................87 Distribution and Relationships.........................................................................................................94 The Cosmopolitans.................................................................................................................97 The American Species............................................................................................................97 The Caribbeans.......................................................................................................................97 The North American–Antilleans ............................................................................................98 The West Indies–South Americans ........................................................................................98 The West Indies–Central Americans ......................................................................................98 The Endemics .........................................................................................................................98 Ecological Zoogeography ..............................................................................................................100 Cuba......................................................................................................................................100 The Greater Antilles .............................................................................................................100 The Lesser Antilles...............................................................................................................100 Venezuela and Panama .........................................................................................................100 Peru .......................................................................................................................................100 Madagascar ...........................................................................................................................100 Conclusion......................................................................................................................................103 Acknowledgments ..........................................................................................................................104 Literature Cited ..............................................................................................................................104

Chapter 8

The Contribution of the Caribbean to the Spider Fauna of Florida........................107 Jonathan Reiskind

Introduction ....................................................................................................................................107 Methods..........................................................................................................................................107 Results ............................................................................................................................................108 Potential Sources of the Spider Fauna.................................................................................108 Climatic Constraints .............................................................................................................111 Several Higher Taxa .............................................................................................................112 Discussion and Conclusions ..........................................................................................................112 Literature Cited ..............................................................................................................................113

Chapter 9

Rhysodine Beetles in the West Indies......................................................................117 Ross T. Bell

Introduction ....................................................................................................................................117 Dispersal Mechanisms ...................................................................................................................118 Relationships of West Indian Genera within the World Fauna.....................................................121 Fossil Evidence ..............................................................................................................................122 Interpretations of West Indian Distributions .................................................................................122 Conclusion......................................................................................................................................123 Literature Cited ..............................................................................................................................124

Chapter 10 The Biogeography of the West Indian Butterflies (Lepidoptera): An Application of a Vicariance/Dispersalist Model ................................................127 Jacqueline Y. Miller and Lee D. Miller Introduction ....................................................................................................................................127 Previous Biogeographical Studies of Butterflies...........................................................................128 Endemism of the West Indian Butterfly Fauna .............................................................................130 The Age of Butterflies and Its Biogeographical Implications ......................................................132 Are All Butterflies Effective Dispersalists?...................................................................................134 Current Studies...............................................................................................................................134 The Dispersalists ............................................................................................................................135 A Vicariance/Dispersal Model for the Biogeography of West Indian Butterflies ........................136 Late Mesozoic to Cretaceous ...............................................................................................136 Late Cretaceous to Eocene...................................................................................................138 Oligocene to Pliocene ..........................................................................................................139 Pliocene to Holocene............................................................................................................147 The Lesser Antilles ........................................................................................................................148 Summary ........................................................................................................................................148 Acknowledgments ..........................................................................................................................149 Literature Cited ..............................................................................................................................150

Chapter 11 Relationships and Divergence Times of West Indian Amphibians and Reptiles: Insights from Albumin Immunology........................................................................157 Carla Ann Hass, Linda R. Maxson, and S. Blair Hedges Introduction ....................................................................................................................................157 Materials and Methods...................................................................................................................157 Results ............................................................................................................................................159 Bufonidae..............................................................................................................................159 Hylidae..................................................................................................................................159 Amphisbaenidae ...................................................................................................................161 Anguidae...............................................................................................................................161 Iguanidae...............................................................................................................................162 Teiidae ..................................................................................................................................162 Colubridae.............................................................................................................................163 Tropidophidae.......................................................................................................................163 Typhlopidae ..........................................................................................................................164 Discussion ......................................................................................................................................165 Bufonidae..............................................................................................................................165 Hylidae..................................................................................................................................166 Amphisbaenidae ...................................................................................................................167 Anguidae...............................................................................................................................167 Iguanidae...............................................................................................................................168 Teiidae ..................................................................................................................................168 Colubridae.............................................................................................................................168 Tropidophidae.......................................................................................................................168 Typhlopidae ..........................................................................................................................169 Conclusions ....................................................................................................................................169 Acknowledgments ..........................................................................................................................170 Literature Cited ..............................................................................................................................170 Appendix: Collecting Localities and Voucher Specimens ............................................................172

Chapter 12 The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies...175 Matthew I. Williams and David W. Steadman Introduction ....................................................................................................................................175 Brief Species Accounts ..................................................................................................................176 Macaws (Ara) .......................................................................................................................176 †Ara tricolor (Bechstein, 1811) — Cuban Macaw .................................................176 †Ara gossei (Rothschild, 1905) — Gosse’s Macaw................................................179 †Ara erythrocephala (Rothschild, 1905) — Red-headed Green Macaw................179 †Ara erythrura (Rothschild, 1907) — Red-tailed Blue-and-Yellow Macaw..........179 †Ara tricolor? or †Ara unknown sp. — Hispaniolan Macaw.................................179 †Ara autochthones (Wetmore, 1937) — St. Croix Macaw .....................................180 †Ara undescribed sp. — Montserrat Macaw ...........................................................180 †Ara guadeloupensis (Clark, 1905a) — Guadeloupe Macaw.................................180 †Ara cf. guadeloupensis — Marie Galante (Guadeloupe?) Macaw .......................181 †Ara atwoodi (Clark, 1908) — Dominica Macaw ..................................................181 †Ara martinica (Rothschild, 1905) — Martinique Macaw.....................................181 Macaws (Anodorhynchus) ....................................................................................................181 †Anodorhynchus purpurascens (Rothschild, 1905) — Guadeloupe Violet Macaw....181 †Anodorhynchus martinicus (Rothschild, 1905) — Martinique Macaw...................181 Parakeets (Aratinga) .............................................................................................................181 Aratinga euops (Wagler, 1832) — Cuban Parakeet ................................................181 †Aratinga chloroptera maugei (Souancé, 1856) — Puerto Rican/Mona Parakeet....182 †Aratinga undescribed sp. — Barbudan Parakeet...................................................182 †Aratinga labati (Rothschild, 1905) — Guadeloupe Parakeet ...............................183 †Aratinga undescribed spp. — Dominica, Martinique, and Barbados Parakeets......183 Parrots or Amazons (Amazona)............................................................................................183 Amazona leucocephala hesterna (Cory, 1886) — Cayman Parrot .........................183 Amazona leucocephala bahamensis (Bryant, 1867) — Rose-throated (Bahamas) Parrot.......................................................................................183 †Amazona undescribed sp. — Turks and Caicos Parrot .........................................183 †Amazona vittata gracilipes (Ridgway, 1915) — Culebra Parrot ..........................183 †Amazona vittata — Barbuda (Puerto Rican) Parrot ..............................................183 †Amazona vittata — Antigua (Puerto Rican) Parrot...............................................185 †Amazona undescribed sp. — Montserrat Parrot ....................................................185 †Amazona violacea (Gmelin, 1788) — Guadeloupe Parrot....................................185 †Amazona cf. violacea — Guadeloupe Parrot?.......................................................185 †Amazona martinicana (Clark, 1905c) — Martinique Parrot.................................186 ?Amazona versicolor (Müller, 1776) — St. Lucia Parrot .......................................186 †Amazona undescribed sp. — Grenada Parrot ........................................................186 Conclusions ....................................................................................................................................187 Acknowledgments ..........................................................................................................................187 Literature Cited ..............................................................................................................................187

Chapter 13 Early Tertiary Vertebrate Fossils from Seven Rivers, Parish of St. James, Jamaica, and Their Biogeographical Implications..................................................................191 Roger W. Portell, Stephen K. Donovan, and Daryl P. Domning Introduction ....................................................................................................................................191 Jamaican Tectonics and Paleogeography.......................................................................................192

Locality and Vertebrate Fauna .......................................................................................................194 Discussion ......................................................................................................................................196 Acknowledgments ..........................................................................................................................198 Literature Cited ..............................................................................................................................198

Chapter 14 The Sloths of the West Indies: A Systematic and Phylogenetic Review................201 Jennifer L. White and Ross D. E. MacPhee Introduction ....................................................................................................................................201 Brief Overview of Megalonychid Discoveries in the West Indies ...............................................202 Higher-Level Relationships............................................................................................................205 Cladistic Analysis...........................................................................................................................206 Data Set ................................................................................................................................206 Results ..................................................................................................................................207 Systematics.....................................................................................................................................210 Subfamily Choloepodinae Gray ...........................................................................................212 Tribe Acratocnini Varona .....................................................................................................213 Choloepodinae incertae sedis ..............................................................................................216 Tribe Cubanocnini Varona....................................................................................................217 Subfamily Megalocninae Kraglievich..................................................................................220 Tribe Megalocnini Kraglievich ............................................................................................220 Tribe Mesocnini Varona .......................................................................................................222 Megalonychidae, incertae sedis ...........................................................................................224 Megalonychidae, gen. et sp. indet........................................................................................225 Biogeographical Issues ..................................................................................................................225 Colonization of the Antilles .................................................................................................226 Distribution of Fauna across Islands....................................................................................227 Acknowledgments ..........................................................................................................................228 Literature Cited ..............................................................................................................................228 Note Added in Proof ......................................................................................................................232 Appendix I: Characters and Character States ...............................................................................233 Appendix II: List of Taxa Used in Cladistic Analysis..................................................................235 Outgroup Taxa ......................................................................................................................235 Extant Ingroup Taxa .............................................................................................................235 Extinct Ingroup Taxa ............................................................................................................235

Chapter 15 The Origin of the Greater Antillean Insectivorans ..................................................237 Howard P. Whidden and Robert J. Asher Introduction ....................................................................................................................................237 Recent Phylogenetic Studies..........................................................................................................238 Molecular Evidence..............................................................................................................238 Morphological Evidence ......................................................................................................239 Biogeographical Hypotheses..........................................................................................................243 McDowell (1958) .................................................................................................................246 Patterson (1962)....................................................................................................................246 Hershkovitz (1972) ...............................................................................................................246 MacFadden (1980)................................................................................................................247 The Land Span Hypothesis ..................................................................................................247

Conclusions ....................................................................................................................................248 Acknowledgments ..........................................................................................................................249 Literature Cited ..............................................................................................................................250

Chapter 16 Systematics and Biogeography of the West Indian Genus Solenodon ...................253 Jose A. Ottenwalder Introduction ....................................................................................................................................253 Evolutionary Relationships of West Indian Insectivores...............................................................254 Historical Surveys of the Solenodontidae .....................................................................................256 Materials and Methods...................................................................................................................257 Results ............................................................................................................................................261 Nongeographical Variation ...................................................................................................261 Variation with Age....................................................................................................261 Secondary Sexual Variation......................................................................................261 Individual Variation ..................................................................................................268 Specific Relationships (Geographical Variation) .................................................................268 Univariate Analyses ..................................................................................................268 Multivariate Analyses ...............................................................................................287 Variation in Cranial Morphology .........................................................................................290 Taxonomic Conclusions .......................................................................................................293 Systematic Accounts ......................................................................................................................294 Solenodon paradoxus ...........................................................................................................294 Solenodon cubanus...............................................................................................................302 Solenodon arredondoi ..........................................................................................................306 Solenodon marcanoi .............................................................................................................308 Late Quaternary and Recent Distribution of Solenodon ...............................................................315 Material and Methods ..........................................................................................................315 Results ............................................................................................................................................316 Solenodon paradoxus ...........................................................................................................316 Solenodon cubanus...............................................................................................................317 Solenodon marcanoi .............................................................................................................319 Solenodon arredondoi ..........................................................................................................319 Discussion ......................................................................................................................................320 Acknowledgments ..........................................................................................................................324 Literature Cited ..............................................................................................................................325

Chapter 17 Characterization of the Mitochondrial Control Region in Solenodon paradoxus from Hispaniola and the Implications for Biogeography, Systematics, and Conservation Management ................................................................................331 Marc W. Allard, Scott D. Baker, Ginny L. Emerson, Jose A. Ottenwalder, and C. William Kilpatrick Introduction ....................................................................................................................................331 Materials and Methods...................................................................................................................332 Results and Discussion ..................................................................................................................332 Acknowledgments ..........................................................................................................................334 Literature Cited ..............................................................................................................................334

Chapter 18 Insular Patterns and Radiations of West Indian Rodents ........................................335 Charles A. Woods, Rafael Borroto Paéz, and C. William Kilpatrick Introduction ....................................................................................................................................335 Materials and Methods...................................................................................................................338 Morphological Analysis........................................................................................................338 Molecular Analysis (Cytochrome b Gene) ..........................................................................341 Specimens Examined................................................................................................341 DNA Sequencing......................................................................................................341 Sequence Analysis ....................................................................................................342 Results ............................................................................................................................................342 Discussion ......................................................................................................................................343 Implications of Sequencing Data on Biogeographical and Evolutionary Hypotheses.................347 Summary of West Indian Evolutionary History and Biogeography of Rodents ..........................349 Acknowledgments ..........................................................................................................................351 Literature Cited ..............................................................................................................................352

Chapter 19 Biogeography of West Indian Bats: An Ecological Perspective .............................355 Armando Rodríguez-Durán and Thomas H. Kunz Introduction ....................................................................................................................................355 Biogeography of Antillean Bats ....................................................................................................355 Geography and Species ........................................................................................................355 Routes of Invasion................................................................................................................358 The Western Route ...................................................................................................358 The Northern Route..................................................................................................358 The Southern Route..................................................................................................359 Patterns in Bat Communities .........................................................................................................359 Body Size and Diet ..............................................................................................................360 Roosts ...................................................................................................................................360 Activity .................................................................................................................................362 Community Structuring..................................................................................................................364 Acknowledgments ..........................................................................................................................366 Literature Cited ..............................................................................................................................366

Chapter 20 Patterns of Extinction in West Indian Bats..............................................................369 Gary S. Morgan Introduction ....................................................................................................................................370 Methods and Materials...................................................................................................................370 West Indian Fossil Chiropteran Faunas .........................................................................................375 Cuba......................................................................................................................................375 Isla de Pinos .............................................................................................................376 Jamaica .................................................................................................................................376 Hispaniola .............................................................................................................................377 Ile de la Gonâve .......................................................................................................377 Puerto Rico ...........................................................................................................................378 Bahamas................................................................................................................................378 Abaco........................................................................................................................378 Andros.......................................................................................................................379 Exuma .......................................................................................................................380

New Providence........................................................................................................380 Grand Caicos ........................................................................................................................380 Cayman Islands ....................................................................................................................381 Grand Cayman..........................................................................................................381 Cayman Brac ............................................................................................................382 Lesser Antilles ......................................................................................................................382 Anguilla ....................................................................................................................382 Antigua .....................................................................................................................382 Barbuda.....................................................................................................................383 Taxonomic and Zoogeographical Review of West Indian Fossil Bats .........................................383 Family Noctilionidae ............................................................................................................383 Noctilio .....................................................................................................................383 Family Mormoopidae ...........................................................................................................383 Mormoops .................................................................................................................383 Pteronotus .................................................................................................................384 Family Phyllostomidae .........................................................................................................386 Subfamily Phyllostominae........................................................................................386 Macrotus....................................................................................................386 Tonatia.......................................................................................................387 Subfamily Brachyphyllinae ......................................................................................387 Brachyphylla..............................................................................................387 Subfamily Phyllonycterinae .....................................................................................388 Erophylla ...................................................................................................388 Phyllonycteris............................................................................................388 Subfamily Glossophaginae.......................................................................................389 Glossophaga..............................................................................................389 Monophyllus ..............................................................................................389 Subfamily Stenodermatinae......................................................................................389 Artibeus .....................................................................................................389 Stenoderma Group: Ardops/Ariteus/Phyllops/Stenoderma ......................................389 Subfamily Desmodontinae .......................................................................................390 Desmodus ..................................................................................................390 Family Natalidae ..................................................................................................................391 Natalus ......................................................................................................391 Nyctiellus ...................................................................................................391 Family Vespertilionidae ........................................................................................................393 Antrozous...................................................................................................393 Eptesicus....................................................................................................393 Lasiurus .....................................................................................................393 Myotis ........................................................................................................393 Family Molossidae ...............................................................................................................394 Molossus....................................................................................................394 Nyctinomops ..............................................................................................394 Tadarida ....................................................................................................394 Chiropteran Extinctions in the West Indies...................................................................................394 Causes of Extinctions ...........................................................................................................394 Island Extinction Patterns ..............................................................................................................398 Bahamas................................................................................................................................398 Cayman Islands ....................................................................................................................401 Greater Antilles.....................................................................................................................401 Northern Lesser Antilles ......................................................................................................402

Bat Extinctions Elsewhere in the Neotropics and in Florida........................................................402 Distributional Patterns....................................................................................................................403 Acknowledgments ..........................................................................................................................404 Literature Cited ..............................................................................................................................405

Chapter 21 The Mongoose in the West Indies: The Biogeography and Population Biology of an Introduced Species ..........................................................................................409 G. Roy Horst, Donald B. Hoagland, and C. William Kilpatrick Introduction ....................................................................................................................................409 Biogeography .................................................................................................................................410 History of Introduction.........................................................................................................410 Current Distribution .......................................................................................................................412 Population Biology ........................................................................................................................413 Methods ................................................................................................................................413 Sex Ratio and Age Distribution.....................................................................................................416 Population Densities and Habitat Use ...........................................................................................418 Summary and Conclusions ............................................................................................................421 Acknowledgments ..........................................................................................................................422 Literature Cited ..............................................................................................................................422

Chapter 22 Status and Biogeography of the West Indian Manatee ...........................................425 Lynn W. Lefebvre, Miriam Marmontel, James P. Reid, Galen B. Rathbun, and Daryl P. Domning Introduction ....................................................................................................................................425 Historical Distribution....................................................................................................................426 Present Distribution, Status, and Habitat Associations .................................................................428 West Indies ...........................................................................................................................428 Puerto Rico ...............................................................................................................428 Jamaica .....................................................................................................................431 Dominican Republic.................................................................................................432 Haiti ..........................................................................................................................434 Cuba..........................................................................................................................434 Bahamas....................................................................................................................435 Central America....................................................................................................................437 Belize ........................................................................................................................437 Guatemala .................................................................................................................439 Honduras...................................................................................................................440 Nicaragua..................................................................................................................440 Costa Rica.................................................................................................................441 Panama......................................................................................................................442 South America ......................................................................................................................444 Colombia...................................................................................................................444 Venezuela..................................................................................................................444 Trinidad.....................................................................................................................447 Guyana......................................................................................................................447 Suriname ...................................................................................................................448 French Guiana ..........................................................................................................449

Brazil.........................................................................................................................449 North America ......................................................................................................................451 United States.............................................................................................................451 Mexico ......................................................................................................................456 Biogeographical Patterns of Trichechus ........................................................................................459 Conclusions ....................................................................................................................................461 Acknowledgments ..........................................................................................................................463 Literature Cited ..............................................................................................................................463

Chapter 23 Historical Biogeography in Cuba: 19th-Century Interpretations and Misinterpretations ..............................................................................................475 Pedro M. Pruna Goodgall Introduction ....................................................................................................................................475 Enhancing the New World Image..................................................................................................476 Summary ........................................................................................................................................478 Literature Cited ..............................................................................................................................478

Chapter 24 Native American Use of Animals in the Caribbean ................................................481 Elizabeth S. Wing Introduction ....................................................................................................................................481 Material and Methods ....................................................................................................................482 Results ............................................................................................................................................490 Terrestrial Component of West Indian Faunal Samples................................................................490 Native Terrestrial Species.....................................................................................................490 Introduced Domestic and Captive Species ..........................................................................493 European Introductions ........................................................................................................495 Aquatic Marine Component of West Indian Faunal Samples.......................................................495 Coral Reef Habitats ..............................................................................................................497 Marine Species Living in Inshore, Estuarine, and Pelagic Waters .....................................499 Total Aquatic Fauna .......................................................................................................................503 Conclusions ....................................................................................................................................504 Land Vertebrates and Invertebrates ......................................................................................510 Captive and Domestic Animals............................................................................................512 Aquatic Fauna.......................................................................................................................514 Acknowledgments ..........................................................................................................................516 Literature Cited ..............................................................................................................................516

Chapter 25 The Prehistory and Early History of the Caribbean................................................519 Samuel M. Wilson Introduction ....................................................................................................................................519 Saladoid Migrations .......................................................................................................................521 Post-Saladoid Changes...................................................................................................................522 European Conquest ........................................................................................................................523 After the Arrival of Europeans ......................................................................................................524 Indigenous Legacies in the Caribbean...........................................................................................525 Literature Cited ..............................................................................................................................525

Chapter 26 Impact of Hunting on Jamaican Hutia (Geocapromys brownii) Populations: Evidence from Zooarchaeology and Hunter Surveys..............................................529 Laurie Wilkins Introduction ....................................................................................................................................529 The Bellevue Site...........................................................................................................................531 Methods..........................................................................................................................................531 Measurements of a Known-Age Sample of Hutias .............................................................531 Zooarchaeological Sample ...................................................................................................532 Hunter Survey.......................................................................................................................533 Results ............................................................................................................................................535 Known-Age Sample .............................................................................................................535 Zooarchaeological Sample ...................................................................................................535 Hunter Survey.......................................................................................................................538 Discussion ......................................................................................................................................538 Captive Breeding Arguments ...............................................................................................539 Age-Frequency Distribution .................................................................................................540 High-Density Populations ....................................................................................................540 Island Size and Structural Complexity ................................................................................541 Sustainable Hunting..............................................................................................................541 Theoretical Constraints.........................................................................................................541 Optimal Foraging..................................................................................................................542 Acknowledgments ..........................................................................................................................543 Literature Cited ..............................................................................................................................543

Chapter 27 Status of Conservation in Haiti: A 10-Year Retrospective......................................547 Florence E. Sergile and Charles A. Woods Introduction ....................................................................................................................................547 Physiography of Haiti ....................................................................................................................547 Biodiversity in Haiti.......................................................................................................................549 Threat to Biodiversity ....................................................................................................................549 Conservation Efforts ......................................................................................................................550 Biodiversity Conservation Strategy......................................................................................551 The NEAP ............................................................................................................................551 Protected Areas.....................................................................................................................552 Conservation Education........................................................................................................552 Management Plans................................................................................................................554 Nongovernmental Organizations ..........................................................................................554 Lessons Learned.............................................................................................................................554 Conclusions ....................................................................................................................................557 Acknowledgments ..........................................................................................................................557 Literature Cited ..............................................................................................................................558 Activity Materials...........................................................................................................................559 Conservation Posters......................................................................................................................560

Index ..............................................................................................................................................561

Map of the West Indies

and Historical 1 Introduction Overview of Patterns of West Indian Biogeography Charles A. Woods In 1989, I edited a book, Biogeography of the West Indies: Past, Present, and Future, based on a symposium held at the University of Florida March 2–5, 1987. This volume included contributions from 50 colleagues, and I attempted to make the book as comprehensive as possible. The book sold out quickly and has been out of print for many years. We discussed reprinting the book because demand has remained great; however, the field of West Indian biogeography is changing so rapidly, especially with the widespread use of molecular systematics and the availability of new fossil evidence, that a simple revision of the original volume seemed a missed opportunity. I decided instead to completely revise and reorganize the original volume into a new book with many new contributions. A few of the original chapters, such as the chapters on West Indian manatees, West Indian butterflies, and the status of introduced mongoose, remain, although they have been extensively revised and modified. The authors of these chapters are the leading authorities on their subjects and the sources of most of the recent contributions in their fields. However, in almost all other cases new chapters and new contributors are included. The original volume attempted to cover as many disciplines as possible and to provide an overview of the biogeography of the West Indies. I have chosen not to follow the same strategy in the new volume. So many new contributions have been made to the field in the last 10 years that it is impossible to cover all of them adequately in one volume. The first book was 878 pages in length, and an even larger volume seemed unworkable. Recent contributions in the area of tectonics (see Iturralde-Vinent and MacPhee, 1999) make some topics redundant, and the comprehensive review by Hedges on the historical biogeography of West Indian vertebrates (1996) touches on many concepts. Therefore, I have decided to make this new volume more focused, and to concentrate on fewer overall topics; the emphasis of the book is on “patterns.” It is hoped that the volume will pull together some of the more interesting patterns and trends in West Indian biogeography, and serve as a stimulus to future research as well as a source book on West Indian biogeography. I apologize to the many colleagues and students who have contacted me over the years hoping that we would reissue or revise the original volume. That volume was produced with the assistance of my friend and colleague, Dr. Ross Arnett, Jr., who died in 1999. It is a compliment to Ross’ skills as a publisher that the book was so attractive and has remained so sought after.

HISTORICAL OVERVIEW The original great contributors to West Indian biogeography were Glover M. Allen (1911), Harold E. Anthony (1916), Thomas Barbour (1914), and W. D. Matthew (1915). These biologists began the systematic study of the biogeography of the West Indies. Great contributions were also made by Dr. William L. Abbott, who made extensive collections of vertebrates in remote regions of Hispaniola (1916–1923) in addition to his collecting activities in other remote regions of the world,

0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

1

2

Biogeography of the West Indies: Patterns and Perspectives

such as the Himalayas. Most of these specimens are at the Smithsonian. Dr. Erik L. Ekman collected plants in Cuba, Haiti, and the Dominican Republic. The collections by Abbott and by Ekman led to the descriptions of many new species, and brought to light the realization of how diverse the flora and fauna of the West Indies are, with pockets of endemism occurring even on the same islands. The collections, field notes, and published works of these biologists provided the first attempts to formulate theories on the biogeography of the West Indies. Erik Ekman is remembered mainly for the plants that he collected, but he also wrote with humor and insight about his travels and experiences (1926, 1928). His personal correspondence and field notes in the archives in Stockholm are full of valuable information on the biogeography of the region. Ekman died in Santiago, Dominican Republic, on January 15, 1931 of complications following a severe attack of malaria. He is buried in Santiago de los Caballeros, where a monument to his memory in the Plaza Valerio is frequently visited by biologists and even tourists. Ekman (born in 1883 in Stockholm) received his Ph.D. in 1913 from the University of Lund in Sweden with a thesis on West Indian Vernonia (Ekman, 1914). That same year the famous German botanist Ignacio Urban of the Museum and Botanical Garden in Berlin convinced Dr. C. Lindman of the Imperial Museum of Stockholm to allow fellowship money to be used to support Ekman on a plant-collecting trip to Hispaniola. Ekman persuaded Lindman to allow him to stop first in Cuba to collect specimens and data on Vernonieae. Ekman arrived in Cuba in April 1914 and became fascinated with the island. He moved on to Hispaniola only after Urban threatened to cut off his funding. He made a brief trip to Haiti between May and September 1917. He collected over 3,000 plant specimens on that trip, but he missed Cuba and was plagued by fevers. He did not return to Haiti until 1924. During his years in Cuba he collected over 100,000 plant specimens including over 1,000 species previously unknown to Cuba. Professor Urban described 850 of these as new species. After returning to Hispaniola in 1924 Ekman worked throughout Haiti and the Dominican Republic, including the offshore islands of Navassa (Ekman, 1929a), Tortue (Ekman, 1929b), and Gonâve (Ekman, 1930a) collecting many species of plants and even a number of birds. For example, in the Dominican Republic he is reported to have collected 15,467 spermatophytes, 16,500 Pteridophytes, and 107 birds. Soon after the death of Ekman the well-known Danish fern specialist C. Christensen wrote a letter indicating that Ekman’s collection of 500 fern species from Hispaniola, 50 of which were new species, represented the largest and most complete collection of ferns from any location in the tropics. Ekman was a well-known figure in Hispaniola, and was the subject of a number of newspaper articles. He was even the subject of a chapter (“Portrait of a Scientist”) in the book The Magic Island by W.B. Seabrook (1929). Ekman greatly influenced James Bond, and was the subject of a chapter (“Basic Training”) in a book by Bond’s wife Mary Bond (1971). Ekman lived frugally (his entire support for 16 years of fieldwork was only about $14,000), and when he was not in the field, he frequently stayed with friends, such as the family of Wilhelm Buch, a pharmacist, in Port-auPrince. When he was in the field he often stayed with peasants and campesinos. The important specimens collected by Ekman form the bulk of the herbarium at the Ministry of Agriculture in Haiti (Damien), and they are an important contribution to the Botanical Garden in Santo Domingo. The specimens, papers, and photographs that Ekman left behind with the Buch family were bought by Dr. George Proctor of the Institute of Jamaica following the death of Wilhelm Buch. Ekman’s field books (“Dagbok”), journals, and personal correspondence are in Stockholm at the Academy of Sciences. I made photocopies of most of this material, which is available at the Herbarium of the University of Florida. In addition to his collections of plants, Ekman collected hundreds of bird specimens and made important notes on Solenodon and Plagiodontia. There is an excellent account of the life of Ekman by Hermano Alain (1954), also known as Dr. Alain H. Liogier, who went on to become the Director of the National Botanical Garden of the Dominican Republic. Publications by Ekman on his work in Haiti (1929a, 1929b, 1930a) and the Dominican Republic (1929c, 1930b) include valuable lists of plants and geographical observations.

Introduction and Historical Overview of Patterns of West Indian Biogeography

3

An extremely valuable review and compilation of the history of the biological investigation of the West Indies can be found in the work of David K. Wetherbee. Few authors have made more significant contributions to the historical record of biological exploration and comparative biology of any region of the world. For example, Wetherbee (1985) in his survey of the historical development of comparative zoology in the West Indies documents the identities of almost all collectors of vertebrate-type specimens in the West Indies up to 1850. Some of Wetherbee’s more valuable contributions include a biogeographical analysis (1989c) and book-length monographs on the zoological exploration of the following countries or regions of the West Indies: Dominican Republic (1986a), Haiti (1985b), Puerto Rico (1986b), the Lesser Antilles (1986c), Cuba (1985c), the Bahamas (1985d), and Jamaica (1985e). These large volumes, in addition to large volumes on the early history of botany in Hispaniola and Puerto Rico (1985a) and additional shorter works on St. Croix (1984), the Dominican Republic (1991c), Central America (1985f ), Hispaniola (1987a, 1988a), and more focused contributions on mollusks (1987d), beetles (1985h), butterflies and moths (1985i, 1987b, 1987c, 1988b, 1989b, 1991a, 1991b), scorpions (1989d), dragonflies (1989e), decapods (1989a), birds (1986d, 1988c), and fishes (1987e, 1988d, 1989f ), form a valuable record of historical explorations of the Caribbean as well as important original observations on West Indian natural history. His unpublished contributions represent a gold mine of information for biogeographers and systematic biologists working in the West Indies (see Wetherbee in the Literature Cited section). Dr. Ernest Williams and I had frequent conversations about the tremendous importance of Wetherbee’s contributions despite the lack of peer review or availability in widely read journals. Wetherbee’s works were “published” as Xeroxed manuscripts, “Copyright by David K. Wetherbee, Shelburne, Massachusetts,” with a date. They are available in a few major libraries, the best collection of which is in the Museum of Comparative Zoology at Harvard where Wetherbee maintained a long association. Wetherbee lived for most of his later years in the small and remote village of Restauracion in the Cordillera Central in the Dominican Republic, not far from the border with Haiti. In his 1985 summary of species types, Wetherbee explains his views on the status of his works. Their publication by xerography is viewed by the author as a medium for soliciting the editorial review that he was unable to find. It is hoped that publication (perhaps by co-authorships) of this material, in second editions, with adequate editorial review, might come to pass. Cooperation is invited.

I have great admiration for Wetherbee’s contributions, which are often overlooked by biologists and biogeographers interested in the Caribbean area. I hope that his works are not lost from the historical record, and that is why I have outlined them in some detail in this introduction. The final “publication” I have seen by Wetherbee is a 465-page tome (1996) on decapods in Hispaniola and 20 other contributions on the fauna of Hispaniola. It concludes with a complete bibliography of Wetherbee’s works. Wetherbee died soon after completing the 1996 tome, and before receiving acknowledgment of the importance of his work on the West Indies. He was about 76 years old when he died, and he had chosen a simple life in a tiny country house with a dirt floor (Andrei Sourakov, personal communication). It was a long way from Harvard, where he had once been a young curator of the bird collection. This eccentric but important contributor to West Indian natural history deserves to be remembered. The advancement of biogeographical and systematic studies in particular regions of the West Indies has often been associated with political or important historical events. For example, the occupation of Haiti by the U.S. Marines (1915–1934), and the building of programs in agriculture modeled after the ones in place in the United States, provided an ideal opportunity for a number of American biologists to work in Haiti during the 1920s and 1930s. During this period Philip Darlington (1935) journeyed to the top of Pic Macaya (October 1934); James Bond began a series of extensive field trips in the Massif de la Selle and other important regions of Haiti in 1927–1928; Gerrit S. Miller, Jr. (1930) and his associates from the Smithsonian collected fossil mammals

4

Biogeography of the West Indies: Patterns and Perspectives

throughout Hispaniola; Erik Ekman made his greatest botanical discoveries; and Alexander Wetmore traveled to some of the most remote reaches of Hispaniola. There is a fine historical review of early biological work in the Greater Antilles (especially Hispaniola) in Wetmore and Swales (1931). The plant and animal collections made during this period were deposited in museums in Berlin, Hamburg, and Leiden, as well as the Philadelphia Academy of Natural Sciences, the Harvard Museum of Comparative Zoology, and the Smithsonian Institution. More recently the United States took a second active role in the activities in Haiti. This began as an intense effort by the U.S. Agency for International Development (USAID) to promote biological conservation in Haiti and to protect watersheds. It was also a time of substantial financial support by the MacArthur Foundation of intense work in conservation and biodiversity in Haiti, Jamaica, and St. Lucia. The work led to the establishment of three national parks in Haiti, and to the collection of many specimens (Woods and Ottenwalder, 1992; Woods et al., 1992). Some of the contributions to the biogeography of the West Indies reported in Woods (1989a) and in this volume are based on these specimens and activities. These activities formed the basis for a national environmental action plan for Haiti (NEAP) and for a series of four widely distributed conservation posters featuring Hispaniolan plants and animals as well as a series of workbooks in French and Creole (see Chapter 27 by Sergile and Woods, this volume, and the list of conservation posters in the Literature Cited section of this introduction). This second spurt in biological activity ended with the political chaos of the U.S. embargo of Haiti. The events that have followed the embargo have not made most biological efforts in Haiti attractive to large numbers of biologists. These spurts in biological attention and the associated collections form the cornerstone of our present understanding of the flora and fauna of the Antilles. The analyses of these and other collections led to the first great syntheses of West Indian biogeography, such as the works of Darlington (1938, 1957) and Simpson (1940, 1943, 1956). The standard reference on the geologic history of the West Indies at the time was Schuchert (1935), who viewed the islands of the West Indies as having been relatively stable in geographical position throughout the Cenozoic. Rapid advances in our understanding of the geological history of the West Indies were made in the 1950s and 1960s. These revelations about the dynamic geological history of the Antilles led to bold new syntheses. The work by Rosen (1976) created a controversy as to how and when organisms dispersed from island to island, and what role plate tectonics played in biogeography. Biologists and geologists began to read one another’s papers with a keen interest as they searched for further evidence on the history of the Antilles. The results of the geological data are summarized in Pindell and Dewey (1982), while Pregill (1981a), Hedges (1982, 1996), and Iturralde-Vinent and MacPhee (1996, 1999) discuss the importance of the recent studies in plate tectonics to West Indian biogeography. Rosen (1985) made a second attempt to synthesize data from geology and biology into patterns that might explain West Indian biogeography. In addition, new collections of organisms from all regions of the Antilles began to bring to light new taxa of striking importance. The works of Olson (1976) on birds, Pregill (1981b) on reptiles, and Iturralde-Vinent and MacPhee (1996) on mammals suggest that the history of some vertebrates in the Antilles may be more ancient than previously anticipated. This hypothesis is supported by the findings of Roger Portell and his associates (see Chapter 13, this volume). The recent finding of fossil frogs, reptiles, and mammalian hair in amber from the Dominican Republic further supports these hypotheses (Poinar and Cannatella, 1987; Poinar and Poinar, 1999). Most importantly, a series of new biochemical techniques and the more widespread use of cladistics have made possible more rigorous testing of the various hypotheses on West Indian biogeography. For example, Hedges (1996) suggests a more recent radiation of herps in the West Indies based on his analysis of molecular data. During the past decade a similar pulse of activity has taken place in Cuba via a working agreement between the American Museum of Natural History in New York and the National Museum of Natural History in Cuba. A number of publications, many from the American Museum, have added significant new analyses based on new and old collections. Interpretations of this new material have resulted in new ideas on the flora and fauna of the region, on the time and rate of extinction of West Indian

Introduction and Historical Overview of Patterns of West Indian Biogeography

5

mammals, and on the tectonics of the Caribbean region (see MacPhee et al., 1989, 1999, 2000; MacPhee and Fleagle, 1991; Salgada et al., 1992; Biknevicius et al., 1993; MacPhee, 1993; MacPhee and Iturralde-Vinent, 1994, 1995a, 1995b; Iturralde-Vinent and MacPhee, 1996, 1999; MacPhee and Grimaldi, 1996; MacPhee and Rivero de la Calle, 1996; McFarlane et al., 1998a; 1998b; Queiroz et al., 1998; Flemming and MacPhee, 1999; Higuera-Gundy et al., 1999). The focal paper in this series of publications is the analysis of the paleogeography of the Caribbean region by Iturralde-Vinent and MacPhee (1999). They propose that the Aves Ridge was largely or totally emergent (= “subaerial”) during the latest Eocene/Early Oligocene (33 to 35 myBP). This would have formed a long peninsula or series of closely associated large islands including parts of the Virgin Islands, Puerto Rico, Hispaniola, and Cuba. They designate this large closely associated landmass “GAARlandia” (Greater Antilles + Aves Ridge), and propose that it could have provided the opportunity (and route) for the invasion of the Antilles by mammals and other organisms. Their model proposes island-to-island vicariance rather than continent/island vicariance as proposed by Rosen (1976, 1985) or overwater dispersal as proposed by Hedges (1996). This hypothesis forms the basis of much discussion, and also the opportunity to test the hypothesis using molecular systematics and the analysis of new fossil discoveries. If Iturralde-Vinent and MacPhee are correct, less divergent forms should be found in the western Greater Antilles (Puerto Rico and Hispaniola) and the more derived forms in Cuba. In the case of rodents, I have previously proposed that the center of rodent diversity was in Hispaniola and Puerto Rico (1989:774) and that the Aves Ridge may have been associated with the invasion of rodents into the Antilles from South America (Woods, 1990:659). Recent work on molecular systematics (see Chapter 18, this volume) supports this hypothesis, as do indications that Hispaniola may be a hot spot of biodiversity in the Caribbean (Hedges and Woods, 1993) and that mammalian diversity is high on Puerto Rico as well as Hispaniola (Woods, 1996). The status of West Indies rodents is revised in Woods (1989a, 1989b, 1989c, 1990) and Iturralde-Vinent and MacPhee (1996). The important hypothesis by Iturralde-Vinent and MacPhee will spawn many other comparisons, and will be the basis for many of the biogeographical analyses during the coming decade. The above publications represent the new wave of research and exploration in the West Indies. Many of these ideas are discussed in various chapters in this volume, which also includes the almost incredible new findings of ancient mammals in Jamaica (see Chapter 13, this volume). It also includes chapters on the biogeography of frequently overlooked creatures such as rhysodine beetles, insects found deep inside tree trunks and forest debris (see Chapter 9, this volume), butterflies (see Chapter 10, this volume), and West Indian plants (see Chapter 5, this volume). I also decided to include chapters by former graduate students at the University of Florida who have written dissertations on the mammals of the West Indies. Laurie Wilkins’ important study on the status of Geocapromys in Jamaica following a well-planned reintroduction coordinated by the Jersey Wildlife Preservation Trust and the University of Florida (see Chapter 26, this volume) complements the unpublished dissertation of Kevin Jordan on the Bahamian hutia (not represented in this volume). Together these two works represent a unique analysis of what happens on remote Caribbean islands when animals are introduced or “re-introduced,” as the case may be, and provide a broad-based comparison of large island vs. small island biogeography and biodiversity. The book also includes a comprehensive chapter by Jose Ottenwalder on the systematics, biogeography, and conservation of the enigmatic West Indian insectivores of the genus Solenodon (see Chapter 16, this volume). The distribution and history of Solenodon in the West Indies represents one of the most interesting conundrums in West Indian biogeography, and Ottenwalder’s contributrion is the most comprehensive analysis of this group to be published to date. A companion chapter on the smaller Antillean insectivores of the genus Nesophontes by Whidden and Asher represents an opportunity to compare and contrast these two genera (see Chapter 15, this volume). White and MacPhee pull together the biogeography of the now extinct West Indian sloths, a group that was clouded in an almost impenetrable taxonomic quagmire until this work (see Chapter 14, this volume). There are two papers on bats, a group reflecting some of the most interesting insular

6

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 1 David Klingener curating the University of Massachusetts Mammal Collection, which he did with pride and devotion.

patterns in the West Indies. Molecular systematics is used to reexamine the patterns of adaptive radiation in West Indian rodents, especially of the family Capromyidae (see Chapter 18, this volume). An intriguing short chapter by Pedro Pruna examines the historical interpretation of ground sloths and primates in Cuba in the context of the 19th-century version of biogeology and nationality (see Chapter 23, this volume). I have tried to make the book as taxonomically broad-based as possible; however, it clearly represents my interests as a mammalogist and mammalian biogeographer. There are more papers on mammals than on any other taxonomic group. For this I apologize to my friends and colleagues in other fields who crave information about other organisms more dear to their hearts and more relevant to their research interests. This book is not designed as the ultimate sourcebook on the biogeography of the West Indies, but rather as a primer on the “patterns” of biogeographical information now available. It strongly represents my bias, and I accept full responsibility for its limitations.

ACKNOWLEDGMENTS I would like to thank all of my colleagues who agreed to contribute chapters and who endured my continuing e-mails and calls for manuscripts and deadlines and in the long process for the book to reach its conclusion. I appreciate the assistance of the late Ross Arnett, Jr. for all he did to make the first volume possible, and for helping me make the “right” decision to completely revise the

Introduction and Historical Overview of Patterns of West Indian Biogeography

7

FIGURE 2 One of Dave’s cartoons. This one features a bat reading a copy of G.E. Dobson’s 1878 Catalogue of the Chiroptera in the collection of the British Museum in which the idea of separating the Microchiroptera based on the development of the shoulder lock was first expressed. Dave drew many such cartoons — most educational, some outrageous — and passed them on to students, colleagues, and friends.

book. To my editor of this volume, John Sulzycki, I express my sincere gratitude. He has been extremely helpful and insightful, and without his assistance and patience this volume would never have been possible. I express an extra measure of gratitude to my colleague, Florence E. Sergile, for her assistance. Florence is the lead author of Chapter 27, and the model of the type of new and dynamic biologist working in the Caribbean. She also played a major role in the editing of the various chapters, and in preparing the manuscript for publication. I could not have achieved this undertaking without the many talents of Florence. There are various people to whom this book could (or should) be dedicated. However, I have chosen to thank and remember Dr. David J. Klingener (Figure 1). Dave, a professor of zoology at the University of Massachusetts at Amherst for his entire professional career, was my major professor. He was also a teacher and friend of many professionals now working in the West Indies, and he touched the lives of many of the contributors to this volume. We all have our special memories of Dave, so I will include a few comments about our time together in the West Indies. For a more complete review of Dave’s life and contributions see Woods (1996a) and Woods and Combs (1996b). My initial exposure to the West Indies was 30 years ago in a remote corner of the southern peninsula of Haiti. There in the mountains above the town of Miragoâne was the facility of Reynolds Haitian Mines. Dave joined me there as we searched for “lost” island shrews of the Nesophontes, the very rare larger insectivore (Solenodon), and Hispaniolan hutias (Plagiodontia) (see Woods, 1976, 1981, 1983). We combined our search for these creatures with searches for caves where we dug for fossil remains, and a search for bats, which we mistnetted at night. I was a young assistant professor at the University of Vermont at the time, and what a pleasure it was to be in the field with Dave. It was there that we learned some of the most important lessons that would carry us through life. We were successful in finding Solenodon (Woods, 1976) and Plagiodontia (Woods, 1981, 1982, 1983), but the genus Nesophontes still remains “lost” and out there for other students and their professors to find (see Woods et al., 1986). During those years as young biologists we spent many an evening in the mosquito-filled

8

Biogeography of the West Indies: Patterns and Perspectives

lowland mangrove swamps near Miragoâne mist-netting for fishing bats (Noctilio). It was Dave who got malaria and described in scientific detail the most vivid hallucinations imaginable while in the depths of the fever spikes. Only an anatomist and accomplished cartoonist could come up with such descriptions (Figure 2). We spent many weeks together in southern Haiti (Ross Bell and his wife Joyce were there too, searching for rhysodine beetles and showing us black widow spiders to be found under each rock on the hillside where we were netting bats and birds). The work on birds revealed incredible site fidelity of migratory warblers and different site utilization by male and female warblers (Woods, 1975), but bats captured Dave’s greatest interest. He went on to write two important papers on the bats of the West Indies (Klingener et al., 1978; Griffiths and Klingener, 1988), and he was the person who taught me the most about West Indian bats. It was as a teacher that Dave made his greatest contributions, influencing the lives of many students and colleagues. He was a great communicator, lecturer, writer, and curmudgeon. We spent a lifetime exchanging letters, sharing lessons, and teaching and advising the same students. Dave’s lifetime was far too short; it ended in a heart attack at his home in South Deerfield on July 6, 1995. So, it is to Dave Klingener that this book is dedicated. He touched the lives of so many West Indian biologists. He was the best teacher I have ever been associated with. And he was a great companion in the field in the wilds of Hispaniola.

LITERATURE CITED Alain, H. 1954. Ekman, Explorador y Botanico Intrepido. Memorias de la Sociedad Cubana de Historia Natural 22(4):361–377. Allen, G. M. 1911. Mammals of the West Indies. Bulletin of the Museum of Comparative Zoology 54(6):174–263. Anthony, H. E. 1916. Preliminary report of fossil mammals from Porto Rico, with descriptions of a new genus of ground-sloth and two new genera of hystricomorph rodents. Annals of the New York Academy of Sciences 27:193–203. Barbour, T. 1914. A contribution to the zoogeography of the West Indies, with special reference to the amphibians and reptiles. Memoirs of the Museum of Comparative Zoology 44:209–359. Biknevicius, A. R., D. A. McFarlane, and R. D. E. MacPhee. 1993. American Museum Novitates 3079:25 pp. + 7 figures and 9 tables. Bond, M. W. 1971. Far Afield in the Caribbean: Migratory Flights of a Naturalist’s Wife. Livingston, Wynnewood, Pennsylvania. Darlington, P. J., Jr. 1935. West Indian Carabidae, 2: Itinerary of 1934. Forests of Haiti; new species; and a new key to Colpoides. Psyche 42(4):167–215. Darlington, P. J., Jr. 1938. The origin of the Greater Antilles, with discussion of dispersal of animals over water and through the air. Quarterly Review of Biology 13:274–300. Darlington, P. J., Jr. 1957. Zoogeography: The Geographical Distribution of Animals. John Wiley & Sons, New York. Ekman, E. L. 1914. West Indian Vernoniae. Arkiv för Botanik (Stockholm) 13(15):1–106 + 6 plates. Ekman, E. L. 1926. Botanizing in Haiti. United States Naval Medical Bulletin 24(1):483–497. Ekman, E. L. 1928. A botanical excursion in La Hotte, Haiti. Svensk Botanisk Tidskrift 22(1–2):200–219. Ekman, E. L. 1929a. Plants of Navassa Island, West Indies. Arkiv för Botanik (Stockholm) 22A(16):1–106 + 2 plates. Ekman, E. L. 1929b. Plants observed on Tortue Island, Haiti. Arkiv för Botanik (Stockholm) 22A(9): 1–60 pp. Ekman, E. L. 1929c. En busca del Monte Tina. Estación Agronomica de Moca, Series B (15):1–17. Ekman, E. L. 1930a. A list of plants from the island of Gonâve, Haiti. Arkiv för Botanik (Stockholm) 23A(6):1–73. Ekman, E. L. 1930b. Excursion botanica al nord-oeste de la Republica Dominicana. Estacion Agronomica de Moca, Series B (17):1–16. Flemming, C. and R. D. E. MacPhee. 1999. Redetermination of holotype of Isolobodon portoricensis (Rodentia, Capromyidae) with notes on recent mammalian extinctions in Puerto Rico. American Museum Novitates 3278:1–11 + 3 figures and 1 table.

Introduction and Historical Overview of Patterns of West Indian Biogeography

9

Griffiths, T. A. and D. J. Klingener. 1988. On the distribution of Greater Antillean bats. Biotropica 20(3):240–251. Guyer, C. and J. M. Savage. 1986. Cladistic relationships among anoles (Sauria: Iguanidae). Systematic Zoology 35:509–531. Hedges, S. B. 1982. Caribbean biogeography: implications of recent plate tectonic studies. Systematic Zoology 31:518–522. Hedges, S. B. 1996. Historical biogeography of West Indian vertebrates. Annual Review of Ecology and Systematics 1996, 27:163–196. Hedges, S. B. and C. A. Woods. 1993. Caribbean hot spot. Nature 364:375. Higuera-Gundy, A., M. Brenner, D. Hodell, J. H. Curtis, B. W. Leyden, and M. W. Binford. 1999. A 10,300 14C yr record of climate and vegetation change from Haiti. Quaternary Research 52:159–170. Iturralde-Vinent, M. A. and R. D. E. MacPhee. 1996. Age and paleogeographical origin of Dominican amber. Science 273:1850–1852. Iturralde-Vinent, M. A. and R. D. E. MacPhee. 1999. Paleogeography of the Caribbean region: implications for Cenozoic biogeography. Bulletin American Museum of Natural History 238: 95 pp. Klingener, D. J., H. H. Genoways, and R. J. Baker. 1978. Bats from southern Haiti. Annals of Carnegie Museum of Natural History 47(5):81–99. MacPhee, R. D. E. 1993. From Cuba: a mandible of Paralouatta. Evolutionary Anthropology 2(2):42. MacPhee, R. D. E. and J. Fleagle. 1991. Postcranial remains of Xenothrix mcgregori (Primates, Xenotrichidae) and other Late Quaternary mammals from Long Mile Cave, Jamaica. Pp. 287–321 in Griffiths, T. A. and D. Klingener (eds.). Contributions to mammalogy in honor of Karl F. Koopman. Bulletin American Museum of Natural History 206. MacPhee, R. D. E. and M. Rivero de la Calle. 1996. Accelerator mass spectrometry 14C age determination for the alleged “Cuban spider monkey,” Ateles (=Montaneia) anthropomorphus. Journal of Human Evolution 30:89–94. MacPhee, R. D. E. and D. A. Grimaldi. 1996. Mammal bones in Dominican amber. Nature 380:489–490. MacPhee, R. D. E., D. C. Ford, and D. A. McFarlane. 1989. Pre-Wisconsinan mammals from Jamaica and models of late Quaternary extinction in the Greater Antilles. Quaternary Research 31:94–106. MacPhee, R. D. E. and M. A. Ituralde-Vinent. 1994. First Tertiary land mammals from Greater Antilles: an early Miocene sloth (Xenarthra, Megalonychidae) from Cuba. American Museum Novitates 3094:1–13 + 4 figures and 2 tables. MacPhee, R. D. E. and M. A. Ituralde-Vinent. 1995a. Origin of the Greater Antillean land mammal fauna, 1: new Tertiary fossils from Cuba and Puerto Rico. American Museum Novitates 3141:1–31 + 11 figures and 3 tables. MacPhee, R. D. E. and M. A. Ituralde-Vinent. 1995b. Earliest monkey from Greater Antilles. Journal of Human Evolution 28:197–200. MacPhee, R. D. E., I. Horovitz, O. Arredondo, and O. J. Vasquez. 1995. A new genus for the extinct Hispaniolan monkey Saimiri bernensis Rimoli, 1977, with notes on its systematic position. American Museum Novitates 3134:1–21 + 6 figures and 12 tables. MacPhee, R. D. E., C. Flemming, and D. P. Lunde. 1999. “Last occurrence” of the Antillean insectivore Nesophontes: new radiometric dates and their interpretation. American Museum Novitates 3261:1–19 + 7 figures and 6 tables. MacPhee, R. D. E., White, J. L., and C. A. Woods. 2000. New megalonychid sloths (Phyllophaga, Xenarthra) from the Quaternary of Hispaniola. American Museum Novitates 3303:1–32. Matthew, W. D. 1915. Climate and evolution. Annals of the New York Academy of Science 24:171–213. McFarlane, D., J. Lundberg, C. Flemming, R. D. E. MacPhee, and S.-E. Lauritzen. 1998a. A second preWisconsinian locality for the extinct Jamaican rodent Clidomys (Rodentia: Heptaxodontidae). Caribbean Journal of Science 34(3–4):315–317. McFarlane, D., R. D. E. MacPhee, and D. C. Ford. 1998b. Body size variability and a Sangamonian extinction model for Amblyrhiza, a West India megafaunal rodent. Quaternary Research 50:80–89. Miller, G. S., Jr. 1930. Three small collections of mammals from Hispaniola. Smithsonian Miscellaneous Collections 82(15):1–9. Olson, S. L. 1981. Oligocene fossils bearing on the origins of the Todidae and Momotidae (Aves: Coraciiformes). Pp. 111–119 in Olson, S. L. (ed.). Collected papers in avian paleontology honoring the 90th birthday of Alexander Wetmore. Smithsonian Contributions in Paleontology 27.

10

Biogeography of the West Indies: Patterns and Perspectives

Paryski, P. E., C. A. Woods, and F. E. Sergile. 1989. Conservation strategies and the preservation of biological diversity in Haiti. Pp. 855–878 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Pindell, J. and J. F. Dewey. 1982. Permo-Triassic reconstruction of Western Pangea and the evolution of the Gulf of Mexico/Caribbean region. Tectonics 1:179–211. Poinar, G. O., Jr. and R. Poinar. 1999. The Amber Forest: A Reconstruction of a Lost World. Princeton University Press, Princeton, New Jersey. Pregill, G. K. 1981a. An appraisal of the vicariance hypothesis of Caribbean biogeography and its application to West Indian terrestrial vertebrates. Systematic Zoology 30:147–155. Pregill, G. K. 1981b. Late Pleistocene herpetofaunas from Puerto Rico. Miscellaneous Publications of the University of Kansas Museum of Natural History 71:1–72. Poinar, G. O., Jr. and D. C. Cannatella. 1987. An Upper Eocene frog from the Dominican Republic and its implication for Caribbean biogeography. Science 237:1215–1216. Queiroz, K. de, L.-R. Chu, and J. B. Losos. 1998. A second Anolis lizard in Dominican amber and the systematics and ecological morphology of Dominican amber anoles. American Museum Novitates 3249:1–23 + 9 figures and 2 tables. Rosen, D. E. 1976. A vicariance model of Caribbean biogeography. Systematic Zoology 24:431–464. Rosen, D. E. 1985. Geological hierarchies and biological congruence in the Caribbean. Annals of the Missouri Botanical Garden 72:636–659. Salgado, E. J., D. G. Calvache, R. D. E. MacPhee, and G. C. Gould. 1992. The monkey caves of Cuba. Cave Science 19(1):25–28. Schuchert, C. 1935. Historical Geology of the Antillean-Caribbean Region. John Wiley & Sons, New York. Simpson, G. G. 1940. Mammals and land bridges. Journal of the Washington Academy of Science 30:137–163. Simpson, G. G. 1943. Turtles and the origin of the fauna of Latin America. American Journal of Science 241:413–429. Simpson, G. G. 1956. Zoogeography of West Indian land mammals. American Museum of Natural History Novitates 1759:1–28. Seabrook, W. B. 1929. The Magic Island, Harcourt Brace, New York. Wetherbee, D. K. 1984. An instant survey of St. Croix, V.I. Natural History. Private (Xeroxed) publication, Shelburne, Massachusetts. 77 pp. Wetherbee, D. K. 1985a. Contributions to the early history of botany in Hispaniola and Puerto Rico. Private (Xeroxed) publication, Shelburne, Massachusetts. 216 pp. Wetherbee, D. K 1985b. Zoological exploration of Haiti for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 556 pp. Wetherbee, D. K. 1985c. Zoological exploration of Cuba for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 223 pp. Wetherbee, D. K. 1985d. Zoological exploration of the Bahamas for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 59 pp. Wetherbee, D. K. 1985e. Zoological exploration of Jamaica for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 212 pp. Wetherbee, D. K. 1985f. Zoological exploration of Central America for new vertebrate species. Private (Xeroxed) publication, Shelburne, Massachusetts. 69 pp. Wetherbee, D. K. 1985g. The historical development of comparative zoology in the West Indies. Private (Xeroxed) publication, Shelburne, Massachusetts. 75 pp. Wetherbee, D. K. 1985h. The two century search for beetles (Coleoptera) in Hispaniola. Private (Xeroxed) publication, Shelburne, Massachusetts. 56 pp. Wetherbee, D. K. 1985i. The sphinx moths (Sphingidae, Heterocerca) of Hispaniola and the 1775 paintings of Rabié. Private (Xeroxed) publication, Shelburne, Massachusetts. 69 pp. Wetherbee, D. K. 1986a. Zoological exploration of the Dominican Republic for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 332 pp. Wetherbee, D. K. 1986b. Zoological exploration of Puerto Rico for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 248 pp. Wetherbee, D. K. 1986c. Zoological exploration of the Lesser Antilles for endemic species. Private (Xeroxed) publication, Shelburne, Massachusetts. 128 pp.

Introduction and Historical Overview of Patterns of West Indian Biogeography

11

Wetherbee, D. K. 1986d. Les petits aras rouges: Ara tricolor — Hispaniolan macaw, and Ara cubensis — Cuban macaw. Private (Xeroxed) publication, Shelburne, Massachusetts. 119 pp. Wetherbee, D. K. 1987a. Further contributions to the history of Hispaniolan zoology. Private (Xeroxed) publication, Shelburne, Massachusetts. 114 pp. Wetherbee, D. K. 1987b. Life stages of Hamadrys amphichloe diasia in Hispaniola (Rhopalocera, Nymphalidae). Private (Xeroxed) publication, Shelburne, Massachusetts. 11 pp. Wetherbee, D. K. 1987c. Life stages of Archimestra teleboas and Dynamine egaea in Hispaniola (Nymphalidae, Papilionoidea). Private (Xeroxed) publication, Shelburne, Massachusetts. 14 pp. Wetherbee, D. K. 1987d. Catalog of the terrestrial and fluviatile mollusk fauna of Hispaniola and a history of the early Hispaniolan malacology. (Co-authored with William J. Clench, although Clench never saw the manuscript according to a handwritten notation by Wetherbee on my copy.) Private (Xeroxed) publication, Shelburne, Massachusetts. 89 pp. Wetherbee, D. K. 1987e. The endemic freshwater fishes of the Dominican Republic, and an historical outline of West Indian ichthyology. Private (Xeroxed) publication, Shelburne, Massachusetts. 237 pp. Wetherbee, D. K. 1988a. Hispaniolan geographic place-names referring to fauna and flora. Private (Xeroxed) publication, Shelburne, Massachusetts. 101 pp. Wetherbee, D. K. 1988b. Larval host-plants, newly determined, of several Hispaniolan butterflies (Rhopalocera) and notes on some early stages. Private (Xeroxed) publication, Shelburne, Massachusetts. 20 pp. Wetherbee, D. K. 1988c. The Hispaniolan versus Cuban origin of the type of Ara tricolor Bechstein (Psittacidae). Private (Xeroxed) publication, Shelburne, Massachusetts. 10 pp. Wetherbee, D. K. 1988d. Guide to marine-fishes that invade freshwaters in Hispaniola. Private (Xeroxed) publication, Shelburne, Massachusetts. 52 pp. Wetherbee, D. K. 1989a. Contributions on the decapod crustacea fauna of Hispaniola. Private (Xeroxed) publication, Shelburne, Massachusetts. 118 pp. Wetherbee, D. K. 1989b. Sixth contribution on larvae and/or larval host-plants of Hispaniolan butterflies (Rhopacera) and notice of “neoteny” — like pupa of Pyrgus oileus L. (Hesperiidae). Private (Xeroxed) publication, Shelburne, Massachusetts. 12 pp. Wetherbee, D. K. 1989c. The counterclockwise vortex of animal dispersal in the central West Indies: significance of the distribution of Diploglossus (Anguinidae, Reptilia) in Hispaniola. Private (Xeroxed) publication, Shelburne, Massachusetts. 7 pp. Wetherbee, D. K. 1989d. A brief guide to the partly-known fauna of alacanes (Scorpionida) of Hispaniola. Private (Xeroxed) publication, Shelburne, Massachusetts. 27 pp. Wetherbee, D. K. 1989e. A guide to the caballitos or libelulas (Odonta) of Hispaniola. Private (Xeroxed) publication, Shelburne, Massachusetts. 72 pp. Wetherbee, D. K. 1989f. A guide to freshwater fishes naturalized from abroad to Hispaniola or to the West Indies. Private (Xeroxed) publication, Shelburne, Massachusetts. 10 pp. Wetherbee, D. K. 1991a. Seventh contribution on larvae and/or larval host-plants of Hispaniolan butterflies and nocturnal activity of adult Hypanartia paulla (Fabricius) (Nymphalidae). Private (Xeroxed) publication, Shelburne, Massachusetts. 13 pp. Wetherbee, D. K. 1991b. Two centuries of exploration for Hispaniolan butterflies. Private (Xeroxed) publication, Shelburne, Massachusetts. 82 pp. Wetherbee, D. K. 1991c. Guayajayuco to Jarabacoa: zoological exploration of the Lamedero Massif, Cordillera Central, Republica Dominicana. Private (Xeroxed) publication, Shelburne, Massachusetts. 32 pp. Wetherbee, D. K. 1996. La Xaiba Prieta and la Xaiba Pinita (Epilobocera, Decapoda) in Hispaniola, and 20+ further contributions on Hispaniolan fauna. Private (Xeroxed) publication, Shelburne, Massachusetts. 465 pp. Wetmore, A. and B. H. Swales. 1931. The birds of Haiti and the Dominican Republic. Bulletin of the United States National Museum 155:1–483. Woods, C. A. 1975. Banding and recapture of wintering warblers in Haiti. Bird Banding 46(4):344–346. Woods, C. A. 1976. Solenodon paradoxus in Southern Haiti. Journal of Mammalogy 57(3):591–592. Woods, C. A. 1981. Last endemic mammals in Hispaniola. Oryx 16(2):146–152. Woods, C. A. 1982. Solenodon paradoxus; Plagiodontia aedium; Geocapromys brownii; Geocapromys ingrahami; Capromys nanus; Capromys melanurus. Pp. 99–100, 293–294, 297–299, 301–302, 303–305 in Thornback, J. and M. Jenkins (eds.). Red Data Book, Mammals. International Union Conservation of Nature and Natural Resources, Gland, Switzerland.

12

Biogeography of the West Indies: Patterns and Perspectives

Woods, C. A. 1983. Biological Survey of Haiti: Status of Endangered Birds and Mammals. National Geographic Society Research Reports 15: 759–768. Woods, C. A. 1989a. The biogeography of West Indian rodents. Pp. 741–798 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Woods, C. A. 1989b. The endemic rodents of the West Indies; the end of a splendid isolation. Pp. 11–19 in Lidicker, W. Z., Jr. (ed.). Rodents: A World Survey of Species of Conservation Concern. Occasional Papers of the IUCN Species Survival Commission (SSC) No. 4. Woods, C. A. 1989c. A new capromyid rodent from Haiti; the origin, evolution and extinction of West Indian rodents and their bearing on the origin of New World hystricognaths. Los Angeles County Museum, Science Series 33:59–89. Woods, C. A. 1990. The fossil and recent land mammals of the West Indies: an analysis of the origin, evolution and extinction of an insular fauna. Pp. 641–680 in Azzaroli, A. (ed.). Biogeographical Aspects of Insularity. Accadia Nazionale dei Lincei, Rome. Woods, C. A. 1996a. The land mammals of Puerto Rico and the Virgin Islands. Annals of New York Academy of Science 776:131–149. Woods, C. A. 1996b. Obituary — David John Klingener: 1937–1995. Journal of Mammalogy 77(3):898–900. Woods, C. A. and M. Combs. 1996. Obituary — David Klingener. Bulletin of the Society for Vertebrate Paleontology (February). Woods, C. A. and J. A. Ottenwalder. 1992. The Natural History of Southern Haiti. Florida Museum of Natural History, Gainesville. Woods, C. A. and F. E. Sergile. 1990. The literature of natural sciences in Haiti. Pp. 297–330 in Lawless, R. (ed.). Haiti, A Research Handbook. Garland Publishing, New York. Woods, C. A., J. A. Ottenwalder, and W. Oliver. 1986. Lost mammals of the Greater Antilles; the summarized findings of a ten weeks field survey of the Dominican Republic, Haiti and Puerto Rico. Dodo, Jersey Wildlife Preservation Trust. 22:23–42. Woods, C. A., F. E. Sergile, and J. A. Ottenwalder. 1992. Stewardship Plan for the National Parks and Natural Areas of Haiti. Florida Museum of Natural History, Gainesville.

CONSERVATION POSTERS Sergile, F. E., C. A. Woods, and L. Walz. 1992. Haiti Conservation Poster #1: Connaître et Protéger la Richesse Naturelle d’Haïti. Florida Museum of Natural History, Gainesville. Sergile, F. E., L. Walz, and C. A. Woods. 1992. Haiti Conservation Poster #2: Connaître et Protéger la Richesse Naturelle d’Haïti with descriptive text. Florida Museum of Natural History, Gainesville. Woods, C. A., F. E. Sergile, and L. Walz. 1993. Haiti Conservation Poster #3: Sauvons Haïti, Sa Nature et son Art. Florida Museum of Natural History, Gainesville. Woods, C. A., F. E. Sergile, J. A. Ottenwalder, and L. Walz. 1994. Haiti Conservation Poster #4: Veye bwa Peyi d’Ayiti. Yo inpòtan nèt, nè, nèt (in Creole). Florida Museum of Natural History, Gainesville.

ENVIRONMENTAL EDUCATION AND ACTIVITY BOOKS Sergile, F. S. and J. R.Mérisier 1993. Connaître et Protéger la Richesse Naturelle d’Haïti. Florida Museum of Natural History, Gainesville. 28 pp. Sergile, F. E. and C. A. Woods. 1995. Action Verte. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1995. Guide de Terrain des Aires Protégées en Haïti. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1995. Nou Pa Gen Tan Pou Pèdi. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1995. Nous n’avons plus de temps à perdre. Florida Museum of Natural History, Gainesville.

Introduction and Historical Overview of Patterns of West Indian Biogeography

13

Sergile, F. E. and C. A. Woods. 1995. Veye richès peyi d’Ayiti. Special Publication of the Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1996. Aux couleurs nationales: le caleçon rouge. Brochure for the national bird of Haiti. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1996. La Calebassine. Brochure for the national flower of Haiti. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1996. Ça vaut mieux qu’une mine d’or. Brochure for environmental education and exhibit on natural resources. Haiti-NET and Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1996. Notre Pin: Roi de nos montagnes. Brochure for the national tree of Haiti. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1996. Un arbre à nous: Le palmier royal. Brochure for the national tree of Haiti. Florida Museum of Natural History, Gainesville. Sergile, F. E. 1998. Anvironman. Se ki sa? Guid pou asosyasyon jèn moun nan pawòl anviwònman. ASSET/Winrock International/USAID. Sergile, F. E. and C. A. Woods. 1998. Pouki Sa Nap Plante. USFWS and Haiti-Net (Booklet). Port-auPrince. Sergile, F. E. 1999. Gestion des resources naturelles et de l’environnement. Commune de Kenscoff. Cahier d’activité pour la gestion de l’environnement dans la commune de Kenscoff. Projet pilote ASSET/Winrock/USAID, Pétionville. Sergile, F. E. 1999. Jesyon resous natirèl ak anviwònman. Komin Kenskòf. ASSET/Winrock/USAID, Petyonvil. Sergile, F. E. 1999. Jesyon resous natirèl. Bèl Fontèn. ASSET/Winrock/USAID. Sergile, F. E. 1999. Anvironman. Se ki sa? Guid pou asosyasyon jèn moun nan pawòl anviwònman. ASSET/Winrock/USAID. Sergile, F. E. 1999. Mon environnement, ma commune. Kenscoff. Cahier d’activité pour la gestion de l’environnement dans la commune de Kenscoff. Projet pilote ASSET/Winrock/USAID. Sergile, F. E. and C. Tardieu. 2000. L’environnement, C’est quoi? Guide pour les associations de jeunes sur la gestion de l’environnement. ASSET/Winrock/USAID. Sergile, F. E. 2000. Coup d’oeil sur la faune d’Haiti. Manuel pour les agents agroforestiers du programme de l’Ecole Moyenne d’Agroforesterie d’Haïti. Sergile, F. E. 2000. Coup d’oeil sur la flore d’Haïti. Manuel pour les agents agroforestiers du programme de l’Ecole Moyenne d’Agroforesterie d’Haïti. Sergile, F. E. 2000. Manuel de gestion des aires protégées. A l’usage des agents agroforestiers du programme de l’Ecole Moyenne d’Agroforesterie d’Haïti.

GENERAL INFORMATION Chevalier, F. D. and F. E. Sergile. 1999. Agenda Vert. Image Marketing: Port-au-Prince, Haiti. Dépot legal 98-12-415 Bibliothèque nationale d’Haïti. Sergile, F. E. 1996. Au nord’est du Nord’Est. Statut de l’environnement naturel d’Haiti, Numero 2. 61 pp. Sergile, F. E. 1996. Sur 1535 km: Un Océan, un golfe, une mer et un potentiel unique. Paper presented at the conference: Gestion des zones cotières en Haiti. Université Quisqueya, UNESCO. Sergile, F. E. and C. A. Woods. 1996. People, Development and Conservation. Report of the Sondeo in Macaya North. Prepared for BSP. Florida Museum of Natural History, Gainesville. Sergile, F. E. and C. A. Woods. 1998. Haiti est généreuse. Annuaire 1998 for Promo-Plus, Port-au-Prince, Haiti. Sergile, F. E. 1999. L’écologie en Haïti. Colonnes pour le Dictionnaire encyclopédique d’Haïti. Centre d’Etudes et de Culture Haitiennes. Sergile, F. E. 2000. Les aires protégées en Haïti. Colonnes pour le Dictionnaire encyclopédique d’Haïti. Centre d’Etudes et de Culture Haitiennes. 3 pp. Taylor, F. B. and F. E. Sergile. 2000. La flore d’Haiti. Colonnes pour le Dictionnaire encyclopédique d’Haïti. Centre d’Etudes et de Culture Haitiennes.

14

Biogeography of the West Indies: Patterns and Perspectives

CONSERVATION EXHIBITS Sergile, F. E. 1993. Fondation Ecologique du Nouveau College Bird. Faune et Environnement (Haiti). Sergile, F. E. 1996. Haiti-NET. Ça vaut mieux qu’une mine d’or. Sergile, F. E. 1998. L’environnement. C’est quoi? ASSET and Association des Jeunes pour la conservation de l’environnement (Jacmel, Haiti).

of the 2 Biogeography West Indies: An Overview S. Blair Hedges Abstract — The West Indies harbor a diverse flora and fauna with high levels of endemism. This, coupled with a complex geological history, has attracted interest in the historical biogeography of the region. Two major models have been proposed. The vicariance model proposes that a proto-Antillean biota connecting North and South America in the late Cretaceous was fragmented by plate tectonic movement to form the current island biotas. The dispersal model suggests that organisms dispersed over water during the Cenozoic to reach the islands. A variation on the dispersal model proposes that a dry land bridge connected the Greater Antilles with South America for a short time during the mid-Cenozoic, facilitating dispersal into the Antilles. Most biogeographical studies addressing these models have been based on well-studied groups of vertebrates. Two lines of evidence suggest that dispersal, and not vicariance or a mid-Cenozoic dry land bridge, is responsible for the origin of most lineages studied. First, most West Indian groups are characteristically depauperate at the higher taxonomic levels, yet they often have some unusually large radiations of species. This taxonomic pattern, which is reflected in the fossil record, suggests that niches left vacant by groups absent from the Antilles have been filled by other groups present. Second, times of divergence estimated by molecular clocks indicate that most lineages arrived during the Cenozoic at times when there were no continental connections with the islands. These two lines of evidence are congruent with the nearly unidirectional current flow in the West Indies that probably brought flotsam from rivers in South America to these islands throughout the Cenozoic. Despite this general pattern, a few groups appear to have arrived very early and may represent ancient relicts of the proto-Antilles. The geological history and paleogeography of the West Indies is exceedingly complex and different authors have suggested different scenarios based on the same evidence. For this reason, it is too soon to exclude any particular model of Caribbean biogeography. The geological database and fossil record will continue to improve, phylogenetic relationships will become better known, and molecular divergence time estimates soon will be available for a wide diversity of taxa. Therefore, despite shortcomings of the current models, we can look forward, in the near future, to resolving many of these long unanswered questions of Caribbean biogeography.

INTRODUCTION A significant percentage of the Earth’s known terrestrial biota is distributed on islands of the West Indies (Figure 1). Many of those species are endemic to the region, to individual islands, and even to isolated areas within some islands. Dominican amber fossils indicate with great clarity that the West Indies has been a region with high species diversity and endemism for at least 20 million years (Poinar and Poinar, 1999). In addition, the complex geological history of the region has offered many opportunities for dispersal and vicariance to affect biotas. Together, these features have made the West Indies an appealing region for the study of historical biogeography. This chapter provides a brief outline of the major hypotheses of Caribbean biogeography being debated and the current evidence bearing on them. Because vertebrates are among the best known organisms in the West Indies, they have been the focus of most biogeographical studies and will be the focus of this outline. This is not intended to be a comprehensive review of Caribbean biogeography but rather an update on the current state of the field. Williams’ (1989) earlier outline provides a useful history of the field and its personalities, and a recent review (Hedges, 1996a) is more comprehensive than this one in its treatment of West Indian vertebrates and their historical biogeography.

0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

15

16

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 1 The West Indies.

Recently, Iturralde-Vinent and MacPhee (1999) provided a detailed elaboration of their land bridge model of Caribbean biogeography that was proposed earlier (MacPhee and Iturralde-Vinent, 1994; 1995). Their model suggests that a short-lived dry land bridge in the mid-Cenozoic brought land mammals and presumably other aspects of the South American biota to the Greater Antilles. Their paleogeographical reconstructions exclude the possibility of a vicariant origin for the current biota. Moreover, while stopping short of excluding overwater dispersal altogether, they argue that “surface-current dispersal of propagules is inadequate as an explanation of observed distribution patterns of terrestrial faunas in the West Indies” (Iturralde-Vinent and MacPhee, 1999). A major focus of this outline is to examine the evidence used by Iturralde-Vinent and MacPhee (1999) to support their land bridge model and to show errors and inconsistencies in their argument. In addition, I show that their paleogeographic reconstructions of the Caribbean region have been influenced by the particular biogeographical model that they attempt to support. The current geological evidence does not exclude proto-Antillean vicariance and does not favor a dry land bridge for the mid-Cenozoic Aves Ridge any more than it favors a chain of islands. Finally, I conclude that the same biotic evidence that argues against an origin by vicariance for most lineages also argues against a mid-Cenozoic land bridge.

WEST INDIAN BIOTA Little is known of the general diversity of bacteria, fungi, and protists in the West Indies or elsewhere (Wilson, 1992; Bayuck, 1999). The flora of the West Indies has not yet undergone a comprehensive review, but there are at least 10,000 species of vascular plants, about one third of which are endemic (Adams, 1972; Gentry, 1992). It is likely that only a small fraction of the invertebrate diversity of the West Indies is known and therefore it is too soon to draw general conclusions. However, the best-known groups tend to exhibit reduced higher-level diversity and have large adaptive radiations of some taxa (Liebherr, 1988; Smith et al., 1994b; Pereira et al., 1997; Schubart et al., 1998). Vertebrates are the best-studied organisms in the West Indies; there are 1,295 described species (Table 1). Of those, endemism ranges from a low of 35% in birds to 99% in amphibians, with an average of 74%. Taxonomic diversity is poor at the higher levels, with many major groups absent,

Biogeography of the West Indies: An Overview

17

TABLE 1 Current Diversity of Native West Indian Terrestrial Vertebratesa Genera Endemic

% Endemic

Total

Endemic

% Endemic

9 4 19 49

14 6 50 204

6 1 9 38

43 17 18 19

74 174 474 425

71 172 443 150

96 99 93 35

7 9 97

32 36 342

8 33 95

25 92 28

58 90 1295

29 90 955

50 100 74

Orders

Families

Fishes Amphibians Reptiles Birds Mammals Bats Other Totals

6 1 3 15 1 4 30

b

Species

Total

Group

a

b

After Hedges (1996a), updated. Including one endemic family of birds and four of mammals.

including primary division freshwater fishes, salamanders, caecilians, marsupials, carnivores, lagomorphs, and most families of frogs, turtles, and snakes. On the other hand, some genera have undergone large radiations. For example, the frog genus Eleutherodactylus and the lizard genus Anolis each contains at least 140 West Indian species and geckos of the genus Sphaerodactylus are not far behind with approximately 85 known species. Most fossils of terrestrial organisms in the West Indies come from Quaternary deposits (Pregill and Olson, 1981; Pregill et al., 1992; Woods and Ottenwalder, 1992; Morgan, 1993) and Hispaniolan amber (Poinar and Poinar, 1999). There is not complete agreement over the dating of the amber (Poinar and Poinar, 1999), although most authors consider the major amber deposits (e.g., La Toca) to be Oligocene or Early Miocene (30 to 15 million years ago [mya]) (Grimaldi, 1995; Hedges, 1996a; Iturralde-Vinent and MacPhee, 1996). Fossils also are known from other times in the Tertiary (Cockerell, 1924; Graham and Jarzen, 1969; Graham, 1993; MacPhee and Iturralde-Vinent, 1994; 1995; Domning et al., 1997; Pregill, 1999). Dominican amber deposits contain the largest fossil assemblage of terrestrial invertebrates in a tropical environment (Poinar and Poinar, 1999). The amber ant fauna has been suggested to be more continental in taxonomic composition (Wilson, 1985) compared with the extant fauna, but such comparisons have not been made for most other invertebrate groups in amber. The fossil vertebrates found in amber are representatives of extant West Indian groups and include frogs of the genus Eleutherodactylus, lizards of the genera Anolis and Sphaerodactylus, a snake of the genus Typhlops, a capromyid rodent, a nesophontid insectivore, and a woodpecker (Poinar and Poinar, 1999). In general, these and other fossil vertebrates from the Tertiary of the West Indies reflect the same taxonomic pattern seen in the Quaternary and extant biota. Exceptions include fossil hair in Dominican amber that may have belonged to a carnivore (Poinar and Poinar, 1999) and an Eocene rhinocerotoid ungulate from Jamaica (Domning et al., 1997). The significance of the Jamaican fossil will be discussed below.

GEOLOGICAL HISTORY The Caribbean region has had a complex geological history (Dengo and Case, 1990; Donovan and Jackson, 1994). This history began when the supercontinent Pangaea separated into Laurasia (north) and Gondwana (south) in the Jurassic (~170 mya). This created the “space” for the Caribbean plate, which formed later in the mid-Cretaceous. Since that time, the Caribbean plate has been moving eastward relative to the North American and South American plates. The Antilles were formed by andesitic volcanism resulting from the subduction of the North American plate beneath the Caribbean

18

Biogeography of the West Indies: Patterns and Perspectives

plate. Initially, these were underwater volcanoes (seamounts) that enlarged with time to rise eventually above the water level as islands. It is not known precisely when the islands became emergent, but the proto-Antillean island arc more or less connected North and South America during the late Cretaceous (~100 to 70 mya). In the early Cenozoic (~60 mya) the proto-Antilles began to collide with the Bahamas platform (part of the North American plate) and fused. This initiated a transform fault south of Cuba and northern Hispaniola, adding to the geological complexity of the region. This newly defined northern edge of the Caribbean plate moved eastward carrying with it Jamaica and the southern portion of Hispaniola (south of Cul de Sac/Valle de Neiba). Eventually the two (or more) portions of Hispaniola fused in the Miocene (~10 mya). Because the Greater Antilles lie along the northern edge of the Caribbean plate where there has been mostly lateral motion during the Cenozoic, there are no active volcanoes on those islands. On the other hand, there are active volcanoes in the Lesser Antilles because they are at the leading edge of the Caribbean plate and directly above the subducting North American plate. For biogeography, it is critical to know which areas were above sea level during the geological history of the West Indies. Unfortunately, that is one of the most poorly known aspects of Caribbean geological history. This is because the exposure of dry land is the result of three interrelated factors: uplift, erosion, and sea level. Sea level fluctuations alone cannot be used as a guide, because large mountain ranges can be uplifted and eroded away in a relatively short period of time. For example, the present Blue Mountains (>2200 m) of Jamaica were uplifted only 5 to 10 mya (Comer, 1974). Although the nature of sedimentary strata provides clues to whether there was subaerial land nearby, such strata are not exposed at all locations and at all time periods. It has been claimed that no land areas in the Greater Antilles were continuously above sea level before about 45 mya (MacPhee and Iturralde-Vinent, 1994; MacPhee and Grimaldi, 1996; Iturralde-Vinent and MacPhee, 1999). However, the geological history of the region is not known in enough detail to support such speculation. In fact, other authors have claimed the opposite: “The first terrestrial (emergent) centers seem to have been in the Dominican Republic, Puerto Rico, and the Virgin Islands. In these places the date of emergence is sometime during the Albian (about 100–110 million years), and in these places emergence persisted to the present” (Donnelly, 1992). Also, the plutons of Puerto Rico were being uplifted and eroded in the early Tertiary. Larue (1994) noted that “shallow-water limestone facies are found in north- and south-central Puerto Rico, suggesting that the Central Block may have been a topographic high in the Eocene.” Even some of the best-known features of Caribbean paleogeography may need to be revised in the future. For example, it has been claimed for Jamaica that “probably no part of the island was more than a few meters above sea level at any time” between the middle Eocene and middle Miocene (Robinson, 1994). However, it seems unlikely that the major drop (160 m) in sea level at the beginning of the Oligocene (32.2 mya) (Miller et al., 1996) did not subaerially expose a similar elevation of the carbonate platform. If this happened, then most of the island would have been exposed for millions of years, at least until the platform eroded back to sea level (or subsided). Iturralde-Vinent and MacPhee (1999) make a similar point, arguing in addition that eastern Jamaica has been continuously subaerial since the Eocene and was connected at one point to southern Hispaniola. This case illustrates that paleogeographical reconstruction is difficult and that geologists with similar data can arrive at very different conclusions. The Bahamas platform has remained a relatively stable carbonate block for most of the Cenozoic (Dietz et al., 1970; Dengo and Case, 1990; Donovan and Jackson, 1994). Only barrier reefs and low islands (as seen today) are believed to have existed in the past. However, the compressional forces of the collision with the proto-Antilles during the early Cenozoic may have caused uplift along the southern margin of the Bahamas platform. Because the platform already was near sea level, any uplift would have exposed dry land for colonization by terrestrial organisms. This biogeographical possibility has not yet been explored.

Biogeography of the West Indies: An Overview

19

A large bolide (asteroid, or less likely, a comet) approximately 10 km in diameter struck the Earth in the Caribbean region at 65 mya (Hildebrand and Boynton, 1990). This well-known event almost certainly was responsible for the extinction of the dinosaurs and many other groups. Besides the global effects of the impact, the local effects are of significance to Caribbean biogeography. For example, the proto-Antilles were located only 1 to 3 crater diameters away from the impact site and apparently sustained massive waves (tsunamis) on the order of a kilometer or more in height (Maurrasse, 1991). Gigantic hurricanes (hypercanes) also would have been generated (Emanuel et al., 1994). These local effects of the bolide impact may have destroyed most or all life on the proto-Antilles at that time (Hedges et al., 1992).

OVERWATER DISPERSAL For islands that have never been connected to other landmasses (e.g., Hawaii, Galápagos), dispersal over water is the only possible biogeographical mechanism. In the case of the West Indies, the complex geological history leaves open the possibility of proto-Antillean vicariance or movement across land bridges. Nonetheless, there is evidence that overwater dispersal was the primary mechanism for the origin of the terrestrial vertebrates (Hedges, 1996a, 1996b). This evidence concerns the taxonomic composition of fauna and molecular clock estimates of divergence time between island lineages and their closest relatives on the mainland. The unbalanced taxonomic composition of the fauna (see above), with poor representation at the higher levels and enormous adaptive radiations of some groups, has been noted for over a century (Wallace, 1881; Matthew, 1918; Simpson, 1956; Darlington, 1957). This has been termed the “central problem” in Caribbean biogeography (Williams, 1989). Although it is possible to reach such a taxonomic composition by extinction of a pre-existing, diverse fauna, one would expect to see some remnants of that pre-existing complexity in the present fauna. In fact, the great radiation and morphological diversity of such groups as the ground sloths (now extinct) and hystricognath rodents, filling niches normally occupied by other orders of mammals (Morgan and Woods, 1986; Woods, 1990), supports the contention that those other orders were absent during much of the Cenozoic. A similar argument can be made for the gigantism, dwarfism, and unusual adaptations observed in many other West Indian living and extinct groups (Olson, 1978; Morgan and Woods, 1986; Pregill, 1986; Hedges, 1996a). The other evidence for dispersal as a major biogeographical mechanism comes from molecular clock studies of vertebrates. The number of amino acid differences in the protein serum albumin separating two species can be estimated using the immunological technique of micro-complement fixation (Maxson, 1992). From calibrations with the vertebrate fossil record it has been shown that such immunological distances are correlated with geological time and can be used as a molecular clock. When this method was applied to amphibians and reptiles in the West Indies (Hass, 1991; Hass and Hedges, 1991; Hass et al., 1993; Hass et al., Chapter 11, this volume) it was found that times of origin for West Indian lineages were scattered throughout the Cenozoic and not clustered during one time period (Hedges et al., 1992; Hedges, 1996b). Moreover, nearly all lineages originated more recently than would be predicted based on the vicariance model (see below). This supported an origin by overwater dispersal for most lineages of amphibians and reptiles in the West Indies. If dispersal is the predominant mechanism, then what was the source area for these lineages? The answer to this question can be obtained from phylogenies, where the source area is inferred from the location of the closest mainland relative to the West Indian lineage. Such an analysis revealed that South America was the major source area for amphibians and reptiles during the Cenozoic (Hedges, 1996b). Although the Greater Antilles, in most places, are closer to North and Central America, this South American origin agrees with the nearly unidirectional water currents in the Caribbean region, flowing from southeast to northwest (Figure 2). Thus, flotsam from rivers in South America that emptied into this current probably carried the ancestors of many Antillean

20

Biogeography of the West Indies: Patterns and Perspectives

60

40

20

AFRICA

20

10

SOUTH AMERICA 1500 KM 60

40

20

FIGURE 2 The southern half of the North Atlantic Gyre, showing the North Equatorial Current flowing from Africa to South America and the West Indies (after Hedges, 1996b). This same clockwise current flow in the North Atlantic would have operated throughout the Cenozoic because of the Coriolis force.

TABLE 2 The Origin of West Indian Terrestrial Vertebratesa Mammals Group Mechanism Dispersal Vicariance Undetermined Source South America Central America North America Other Undetermined

Fish

Amphibians

Reptiles

Birdsb

Bats

Other

Total

16 0 1

8 1 0

67 0 1

425 0 0

42 0 0

8 0 1

566 1 3

7 0 0 0 2

35 8 3 4 18

— — — — —

14 18 2 0 0

7 1 1 0 0

69 27 15 6 20

6 0 9c 2 0

a

Shown are the numbers of independent lineages (populations, species, and higher taxa), after Hedges (1996a). b The number of lineages in birds is not known, but is >300; the major source area is North America, but the specific number of lineages from each source area is not known. c Some of these lineages of fishes may have arrived from Central America.

lineages (Hedges, 1996b). In some cases, such as the endemic Cuban lizards of the genus Tarentola, flotsam probably carried them all of the way from Africa in this same current. Although molecular data generally are lacking for most other vertebrate lineages in the West Indies, some data on relationships and timing can be gleaned from the literature and fossil record. These data showed that overwater dispersal was supported for nearly all (>99%) lineages of West Indian terrestrial vertebrates (Hedges, 1996a). For nonvolant taxa, the primary source area still was South America, but most of the volant taxa (birds, bats) in the West Indies arrived from North and Central America (Table 2). How could a terrestrial vertebrate such as a frog survive a long journey (several months) across open water? Although floating mats of vegetation (flotsam) have been observed frequently (Guppy, 1917; King, 1962; Heatwole and Levins, 1972), no raft carrying an animal has been seen leaving

Biogeography of the West Indies: An Overview

21

a river in South America and later landing in the Greater Antilles. However, flotsam apparently carried green iguanas from Guadeloupe to Anguilla during September and October 1995 and a viable population was established. The entire journey was not verified but lizards were seen on the flotsam as it landed on a beach in Anguilla and circumstantial evidence suggested that the journey began in Guadeloupe at least 1 month earlier as a result of one or two hurricanes (Censky et al., 1998). Those authors elude to the importance of this observation by stating that “for overwater dispersal to be considered a realistic explanation for the distribution of species in the Caribbean, it must be demonstrated that a viable population could be established” (Censky et al., 1998). But this is not so, because many aspects of science are inferred without direct observation (e.g., existence of past life and subatomic particles). In the case of biogeography, the existence of organisms on islands (e.g., Hawaii) that never had connections with continents demonstrates that overwater dispersal must have occurred unless one evokes spontaneous generation. Whether the Greater Antillean fauna owes its origin primarily to dispersal or vicariance is another question. But the fact that dispersal is a “realistic” alternative to vicariance does not rely solely on the observation that green iguanas landed on a beach in Anguilla in 1995. If tropical storms and hurricanes have been influential in the transfer of flotsam in the Caribbean, then it is possible that the direction of transfer will not always have corresponded to the generalized water current flow. The strong winds of a hurricane, moving in a counterclockwise vortex, will move current in any direction depending on the specific track of the storm. For example, a westwardmoving hurricane passing to the north of Puerto Rico and eastern Hispaniola will bring strong winds and currents from west to east across Mona Passage. Whether this would be sufficient to carry flotsam from Hispaniola to Mona or Puerto Rico is not known, but the likelihood must be considered (also, the hurricane itself may reverse direction). Based on the number of hurricanes following such a track during the previous 50 years, it is likely that hundreds of thousands of dispersal opportunities have occurred over the last 20 million years. Some seemingly anomalous distributions of vertebrates, such as the presence of two reptiles (Anolis longiceps and Tropidophis bucculentus) with Cuban affinities on Navassa Island, may be the result of such hurricane transport. Although such phenomena may explain local distributions, it is unlikely that hurricanes would modify the direction of movement of flotsam over longer distances.

PROTO-ANTILLEAN VICARIANCE As an alternative to overwater dispersal, Rosen (1975) proposed a vicariance model of Caribbean biogeography. This model suggests that the present West Indian biota represent the fragmented remnants of an ancient biota that was continuous with those of North and South America in the late Cretaceous. Proto-Antillean vicariance cannot be eliminated on geological grounds because even the most current geological models (Dengo and Case, 1990; Donovan and Jackson, 1994) show a proto-Antillean island arc system connecting North and South America during the late Cretaceous. The question whether that island arc formed a dry land bridge or was a chain of islands has not yet been answered conclusively. Since it was proposed, the vicariance model has proven difficult to test. The original suggestion that the congruence of “tracks” (lines drawn between areas with shared faunas) supports the model is not upheld because distributional congruence could simply reflect similar patterns of dispersal as would be expected with unidirectional current patterns (Hedges, 1996a; 1996b). The same can be said of using phylogenies and area cladograms (Rosen, 1985), although the added difficulty here is that the details of land connections through time in the Greater Antilles remain poorly understood. Some cladistic biogeographers have considered dispersal to be untestable and unscientific, and have placed it in a secondary role (Nelson and Platnick, 1981; Morrone and Crisci, 1995). However, most biogeographers consider dispersal a major mechanism that cannot be ignored. The same evidence discussed above as support for overwater dispersal is the evidence that argues against vicariance as the primary mechanism explaining the origin of the West Indian fauna.

22

Biogeography of the West Indies: Patterns and Perspectives

The taxonomic composition of the current and known Tertiary fauna is depauperate at higher taxonomic levels and does not reflect a cross section of a continental biota. In addition, the times of divergence between Antillean groups and their mainland relatives suggest a more recent (Cenozoic) origin than would be predicted by vicariance. Nonetheless, several lineages of Antillean vertebrates may be quite old and could possibly date to the proto-Antilles. One is the frog genus Eleutherodactylus, which shows a time of origin in the West Indies close to the Cretaceous/Tertiary boundary (Hass and Hedges, 1991, Hedges, 1996b). Another is the xantusiid lizard genus Cricosaura that occurs in eastern Cuba. No molecular clock estimate is available for Cricosaura, contra Iturralde-Vinent and MacPhee (1999:51), but instead the older age for its lineage is inferred from mainland fossil data and the relationships of xantusiid lizards (Hedges et al., 1991; Hedges and Bezy, 1993). Even if the lineage itself is old, the relictual nature of xantusiid lizards suggests caution in using the current distribution as evidence of past distribution. Among mammals, the insectivores Solenodon and Nesophontes probably represent old lineages that might date back to the Cretaceous (MacFadden, 1980), but no molecular or fossil data have yet been offered as support of that suggestion. Even if the current West Indian fauna does not show a predominantly vicariant origin, this is not to say that a vicariant biota did not exist at earlier times. For example, the recent discovery of ungulate (rhinocerotoid) and iguanid lizard fossils from the Eocene (~50 mya) of Jamaica (Domning et al., 1997; Pregill, 1999) may be evidence of such a biota. Ungulates are not known from elsewhere in the West Indies. Whether this lineage reached Jamaica on dry land from the mainland, or dispersed across a water gap, is not known. The Oligocene submergence of Jamaica, if it occurred (see above), presumably would have eliminated most or all of the existing biota. Nonetheless, the Jamaican Eocene fossils indicate that a diverse biota may have existed on some Caribbean islands in the early Cenozoic.

THE LAND BRIDGE MODEL OF MACPHEE AND ITURRALDE-VINENT Before plate tectonics provided the mechanism for vicariance, the “land bridge” was the major alternative mechanism to dispersal. Supporters of land bridges (Scharff, 1912; Barbour, 1916; Schuchert, 1935) debated with supporters of overwater dispersal for the first half of the 20th century. The primary argument for land bridges was the seeming impossibility that some groups of organisms, such as freshwater fishes and amphibians, could disperse across salt water (see discussion above). The peninsulas of land that were erected between the islands and the mainland, based on the distributions of organisms, largely were conjectural with little or no geological evidence. After plate tectonics became accepted in the latter part of the 20th century, and paleogeography became better known, most of the proposed land bridges were not supported by geological evidence. However, the refined geological data have suggested new possibilities for past land bridges. One such possibility of a mid-Cenozoic land bridge in the Caribbean region is the Aves Ridge, now almost entirely submerged. The Aves Ridge, located just to the west of the Lesser Antilles, has long been known to have been the precursor of the present-day Lesser Antilles (Malfait and Dinkelman, 1972; Dengo and Case, 1990; Donovan and Jackson, 1994). As such, it was intimately tied to the geological evolution of the Greater Antilles and connections, in the “island arc” sense, with the adjacent continents. Biogeographers also have noted the importance of the Aves Ridge for Caribbean biogeography (Rosen, 1975; Holcombe and Edgar, 1990; Woods, 1990). In a detailed discussion of the Aves Ridge, Holcombe and Edgar (1990) stated “between middle Eocene and early Miocene time it is possible that the Aves Ridge may have been a land bridge. To have been a land bridge, the Aves Ridge would have had to have undergone about 2,000 m of subsidence. There is no direct evidence to support subsidence greater than about 1200 m, but

Biogeography of the West Indies: An Overview

23

samples of Eocene reef limestone recovered from a well (Marathon SB1) drilled on Saba Bank, which joins Aves Ridge on the north, demonstrate that the bank has subsided about 3000 m since the end of Eocene time.” Those authors show a figure of what the present Aves Ridge would look like if it were 600 m and 1000 m higher, exposing many islands, or by inference, a land bridge if subsidence had been even greater. In a separate paper in the same volume, Woods (1990) specifically proposes that this Aves Ridge land bridge (or chain of islands) provided a potential mid-Cenozoic corridor for the entry of mammals to the Greater Antilles. However, geological support for a continuous land bridge vs. a chain of islands does not exist. In a recent series of papers, MacPhee and Iturralde (1994; 1995; 1999) have championed the possibility that the Aves Ridge was a mid-Cenozoic land bridge. They refer to it as a “landspan” defined as “a subaerial connection (whether continuous or punctuated by short water gaps) between a continent and an off-shelf island (or island arc).” But for Caribbean biogeography, the distinction between a dry land bridge and an island chain is a major one. A dry land bridge will allow a cross section of the continental fauna to enter the Greater Antilles whereas an island chain will act as a filter, permitting only selected lineages to enter. Most authors discussing Caribbean biogeography have assumed that the Aves Ridge was an island chain, much like the adjacent Lesser Antilles, during the Cenozoic (Rosen, 1975; Perfit and Williams, 1989; Hedges, 1996a). This concept is not new and it fits with the taxonomic composition of the Antillean fauna. However, the suggestion of a dry land bridge would not agree with the taxonomic composition of the fauna or with molecular time estimates (see below). Although there is no geological evidence yet available to distinguish between a dry land connection and a chain of islands, the paleogeographic diagrams illustrated by Iturralde-Vinent and MacPhee clearly show a dry land connection from 35 to 33 mya, and that is the model that they emphasize. Iturralde-Vinent and MacPhee (1999) acknowledged that evidence against a dry land connection is provided by molecular clock studies and taxonomic composition of the fauna, and therefore considerable attention was given to a critique of studies supporting overwater dispersal, especially that of Hedges (1996b). The different issues that they raise will be discussed separately below.

DIVERGENCE TIMES A prediction of a dry land bridge connection is that times of divergence between Antillean groups and their mainland counterparts should cluster around 35 to 33 mya, according to the model of Iturralde-Vinent and MacPhee (1999). Molecular clock studies of West Indian vertebrates do not show this pattern, but instead show a scattering of divergence times throughout the Cenozoic (Hedges et al., 1992; Hedges, 1996b). Iturralde-Vinent and MacPhee criticize several aspects of these studies, with emphasis on the most recent study (Hedges, 1996b). None of these criticisms is valid, and I will respond to each of them below. Ironically, the evidence that they have erred in their criticisms was provided, in most cases, in the original paper (Hedges, 1996b). Number of Lineages Analyzed The first criticism of Iturralde-Vincent and MacPhee is that the number of evolutionary lineages was not correctly counted. This is not true. Information on time of origin was unavailable for 4 of the 77 lineages in my study, and the concern of Iturralde-Vinent and MacPhee (1999) was that the readers were misled into thinking that such information was available and supported dispersal. But the relevant table (Hedges, 1996b: table 3) and text are clear about information available and not available: “At least some information is available for nearly all lineages (73/77 = 95%), and of those all but one (99%) are in the Cenozoic” (Hedges, 1996b:113) (note the fraction given in the original text). Even that statement was conservative because the four lineages in question also probably arose by dispersal: “Of the four lineages for which no data on the time of origin are available (Hyla heilprini, Phyllodactylus wirshingi, Mabuya lineolata, and the Leptotyphlops

24

Biogeography of the West Indies: Patterns and Perspectives

bilenata group), all have congeneric species on mainland Central or South America and none have highly divergent morphologies that would suggest a long period of isolation. Therefore all four of those lineages likely arose in the Cenozoic by dispersal” (p. 113). In a footnote, Iturralde-Vinent and MacPhee (1999:41) comment that there were three “errors” in my tabulation of data. Again, these were not errors but confusions on the part of IturraldeVinent and MacPhee. Regarding the first purported error, they state that “Crocodylus intermedius, known from only one or two individuals, cannot be considered to be established in the West Indies.” But my criteria (Hedges, 1996b) did not consider the number of individuals — after all, some West Indian species are known only from the holotype. I excluded lineages known to be introduced by humans and C. intermedius was not in that category. Schwartz and Henderson (1991) considered C. intermedius to be part of the West Indian herpetofauna and I do not disagree with their interpretation. The second purported error involves one of the four populations of the lizard Iguana iguana. Iturralde-Vinent and MacPhee state that I. iguana “does not occur on the Cayman Islands.” This is incorrect. Schwartz and Henderson (1991) included it as part of the endemic (not introduced) Cayman herpetofauna, and its continued presence in the Cayman Islands has been confirmed (A. Echternacht, personal communication). If it is later found that it was introduced by humans (a possibility), then it would be removed from consideration as a native lineage, but in any case the statement by Iturralde-Vinent and MacPhee, that it does not occur on the Cayman Islands, is incorrect. The third purported error mentioned in the footnote concerns another lizard. Iturralde-Vinent and MacPhee state that “Mabuya bistriata is presumably a lapsus for Mabuya mabuya; M. bistriata is a Brazilian species.” There was no lapsus. As detailed in the checklist of West Indian amphibians and reptiles (Powell et al., 1996) in the same volume as my study, a taxonomic problem with M. mabuya led to the recognition of the West Indian populations as M. bistriata. Thus the use of the name M. bistriata was not an error but followed current usage. Mixture of Morphological and Immunological Data The second criticism by Iturralde-Vinent and MacPhee (1999) is that I mixed morphological and immunological (not immunodiffusion, which is another method) data, and that this obscures biogeographical inference. They state that, in the case of 40 lineages (56%), morphological data are used as a “proxy measure” of divergence time. This is not true. In a relatively small number of cases involving endemic West Indian species with congeners on the mainland, my stated assumption (see above) was that the divergence between two closely related species in the same genus (of these particular vertebrates) probably occurred in the Cenozoic and not in the Cretaceous. However, nearly all of the 40 lineages noted by Iturralde-Vinent and MacPhee involve species that have populations both in the West Indies and on the mainland. As stated in the methods, I assumed that populations of the same species most likely diverged in the Quaternary (2 to 0 mya) regardless of their morphological divergence; published support for this assumption was mentioned. Moreover, none of those time estimates was used in the figure showing times of origin (Hedges, 1996b: figure 2). Iturralde-Vinent and MacPhee were aware of this because they used this large number of nonendemics as a separate criticism (see below). Taxa Are Not Discriminated in Terms of Interpretative Significance Here, Iturralde-Vinent and MacPhee explain that different organisms disperse differently. For example, some lizards would be expected to raft rather than swim, whereas large crocodilians may not have required a raft. Of course this is true, but it is unclear why it is mentioned as a criticism because I made no claims to the contrary. However, it is worth noting that nearly all West Indian amphibians and reptiles are much smaller than a crocodilian and would most likely have dispersed by rafting.

Biogeography of the West Indies: An Overview

25

Overrepresentation and Ambiguous Significance of Nonendemics Iturralde-Vinent and MacPhee claim that I have overrepresented the number of nonendemic lineages, but they justify their claim by mentioning only three such species. However, I discussed each of the 77 lineages (including those three) separately (Hedges, 1996b) and, again, it appears that they have apparently overlooked that discussion. For Gonatodes albogularis, I mentioned that the Jamaican and Hispaniolan populations are recognized as an endemic subspecies suggesting that they are not the result of human introduction. For Hemidactylus brooki haitianus, I mentioned that the West Indian populations are considered to represent an endemic species, H. haitianus, in the accompanying checklist (Powell et al., 1996) and therefore are also not the result of human introduction. The origin of the third species in question, H. mabouia, is less clear, but that ambiguity is mentioned in the account of that species. Moreover, none of these three taxa is included in the figure of divergence times (Hedges, 1996b: figure 2). Iturralde-Vinent and MacPhee also claim that the nonendemics, in general, are overrepresented “relative to their importance.” My intention was to be objective and identify all independent lineages no matter when they arrived to the West Indies, as long as it was by natural means. A dispersal event in the Pleistocene could be just as important as a dispersal event in the Eocene. Although in my analysis these data were given equal importance, Iturralde-Vinent and MacPhee have the option not to consider them to be important. In any event, this is not an error or misrepresentation. Low Number of Nonendemic Lineages in the Greater Antilles This criticism is similar to the previous one in that Iturralde-Vinent and MacPhee place greater importance on some aspects of my analysis than others. In this case, their focus was on the Greater Antilles, so they were sensitive to the fact that Lesser Antillean lineages were included. But my study concerned the biogeography of the West Indies and therefore I was interested in the Lesser Antilles as well as the Greater Antilles. Again, there is no error or misrepresentation. Unknown Shaping Influence of Extinction The effect that the extinction of lineages has had on shaping the past and present composition of the West Indian fauna is unknown. The major problem is that there are very few Tertiary fossils. My analysis was not concerned with this question and therefore it is unclear why this was mentioned in this section of Iturralde-Vinent and MacPhee (1999). Finally, Iturralde-Vinent and MacPhee consider one possible source of error in the time estimation: phylogenetic error. This might happen when the closest mainland relative of an Antillean group is actually more distantly related, resulting in an overestimation of the divergence time. We mentioned this source of error in our original paper (Hedges et al., 1992) and noted that, because nearly all times were younger than the predicted time for vicariance, that this type of error, even if present, would not affect our conclusion. Iturralde-Vinent and MacPhee (1999) state that “it actually does matter because filling a matrix with overestimates can obscure whatever pattern — including any concentration of splits — that may exist within the phylogeny” (p. 45). Again, they have taken this out of context and misinterpreted the point. Our studies were not focused on testing a dry land bridge hypothesis in the Oligocene but rather proto-Antillean vicariance (Cretaceous) vs. dispersal (Cenozoic). So we were correct in stating that such error did not affect “our conclusion.” But at the same time, acknowledging that this source of error is a possibility is not the same as saying that our entire data set was full of this type of error. The latter is not true. Iturralde-Vinent and MacPhee further speculate that the “pre-28” mya splits might represent overestimates, in which case the absence of data points clustering at that time would not bear negatively on their model. However, all comparisons were chosen carefully, and I discussed each separately in the text (Hedges, 1996b). While a few pre-28 mya comparisons (e.g., Osteopilus,

26

Biogeography of the West Indies: Patterns and Perspectives

Typhlops, Amphisbaena) may be influenced by such phylogenetic error, most probably are not because other data were available to guide choice of sister group.

WATER CURRENTS Iturralde-Vinent and MacPhee (1999) claimed that some of the past current flow patterns “are incompatible with the history of faunal emplacement in the Caribbean region as envisaged by Hedges” (1996a, 1996b). They note that I gave “little attention to the varying paleogeographical configurations of the Caribbean region on current flow” (p. 45). This is not true, as I noted “because the Caribbean always has been north of the equator during geological history, the Coriolis Force would have produced the same clockwise current flow in the past, even while a water connection to the Pacific was in existence” (Hedges, 1996b:118). As will be seen below, the existence or not of the Aves Ridge land bridge would not alter this primary mechanism for the transport of flotsam from South America to the Antilles. Iturralde-Vinent and MacPhee (1999) present reconstructions of marine surface current patterns for four time periods during the Cenozoic (since latest Eocene) based on “slight modifications” of several primary sources. However, reference to those primary literature sources indicates that these purported slight modifications were in reality major modifications. For example, their reconstruction of 35 to 33 mya shows the Aves dry land bridge fully exposed, completely blocking current flow between the Atlantic and Pacific Oceans. However, their reference (Droxler et al., 1998) shows a continuous current flow from Atlantic to the Pacific, noting that “the Aves Swell might have been shallow enough for at least part of a 35 m.y. long interval to have modified the circulation of oceanic waters in the western North Atlantic and to have formed a partially or fully developed barrier to circulation” (p. 172). The two alternatives depend on whether there was a continuous dry land bridge (Iturralde-Vinent and MacPhee, 1999) or an island arc (Pindell, 1994). Even if the two alternatives were equally plausible (see discussion of geological evidence above), the water current flow patterns presented by Iturralde-Vinent and MacPhee are influenced by their need to explain how mammals got to the Greater Antilles. In this sense, it is circular reasoning to use such biased interpretation of surface current patterns to argue in favor of the same biogeographical model. Even Droxler et al. (1998) eluded to the influence of mammal fossils in their assessment of water current patterns: “very strong supporting evidence for this possibility [of a land bridge] comes from the islands of the Greater Antilles where fossil skeletal remains of early Miocene land mammals with South American affinities, including sloths, have been discovered” (MacPhee and Iturralde-Vinent, 1994, 1995; Iturralde-Vinent et al., 1996). However, they concluded that the part played by the exposure of the Aves Swell in “modifying the oceanic circulation and the regional and global environment is much more speculative” (p. 186). Even if the Aves Ridge formed a continuous land bridge that blocked marine current flow between the Atlantic and Pacific, this would not have prevented flotsam from reaching the Antilles. The North Atlantic Gyre would have functioned the same then as it does now, bringing currents up along the northeast coast of South America to the Caribbean (Figure 3). An equatorial countercurrent may have affected some areas along the northeast coast of South America because that region was not very far north of the equator at that time (Figure 3). However, even if this were true, at least some flotsam from northeastern South America would have been deposited on the Aves Ridge land bridge (i.e., part of the Antilles) and directly on the Greater Antilles. The attention given by Iturralde-Vinent and MacPhee to the rivers of northwestern (rather than northeastern) South America is misleading because, even today, they are less likely to be major contributors of flotsam to the Greater Antilles. Similarly misleading is the counterclockwise current direction, east of the Aves Ridge land bridge, shown by Iturralde-Vinent and MacPhee (1999: figure 10) in their water current reconstructions. Presumably this represents the Equatorial Countercurrent, but it was not illustrated in Droxler et al. (1998) — whose primary concern was paleocurrent flow in this region — and would be unlikely considering the Coriolis force (resulting in clockwise flow) and the fact that the Caribbean always has been north of the equator.

Biogeography of the West Indies: An Overview

A

27

Gulf Stream

Early Oligocene 3 6 - 3 0 M YA

Pacific Ocean Guiana Shield

B

Pacific Ocean

Gulf Stream

Pliocene/Quaternary 4 - 0 M YA

Guiana Shield

FIGURE 3 Water current patterns in the Caribbean region at two different times in the Cenozoic. Most features are based on Droxler et al. (1998), although more of the northeastern coast of South America is shown. Water current flow along the Guiana Shield (Guiana Current) is based on present-day water current patterns (Droxler et al., 1998) and inferred patterns in the past based on paleolatitude (Smith et al., 1994a). Carbonate platforms that may have affected current flow in the Caribbean are indicated with horizontal hatching. (A) Early Oligocene (36 to 30 mya). There are two possibilities. If the Aves Ridge were a dry land bridge (IturraldeVinent and MacPhee 1999; shown by dotted lines) the Guiana Current would have been deflected to the northwest along the Antillean landmasses and up to the Gulf Stream. If the Aves Ridge were a chain of islands (Droxler et al., 1998), then some current (dashed arrows) would have passed by the islands and on to the Pacific Ocean (as it did during the Miocene). In either case, rivers in northeastern South America draining into the Guiana Current would have provided a source of flotsam for the Antilles. (B) Pliocene and Quaternary (4 to 0 mya). The Guiana Current continues to flow along the northeastern coast of South America and into the Caribbean, bringing flotsam to the Antilles.

Most of northeastern South America, between the present-day Orinoco and Amazon Rivers, forms the Guiana Shield, and drainage from this region, because of its location southeast of the Lesser Antilles, is an important source of flotsam in the Caribbean (Guppy, 1917). The importance of this potential source region, and adjacent current patterns, is highlighted by the distribution and

28

Biogeography of the West Indies: Patterns and Perspectives

relationships of species occurring in northeastern South America and the Lesser Antilles (Henderson and Hedges, 1995, Hedges, 1996b: figure 4). This region also would have drained into the Atlantic during the Tertiary (Hoorn et al., 1995), but Iturralde-Vinent and MacPhee suggest that flotsam “would have been as likely to drift toward Africa as the West Indies” (p. 51). Even if true, it would only mean that about half of the millions of potential rafting organisms might be diverted elsewhere than the West Indies. However, to explain the origin of one Antillean lineage (e.g., tropidophiid snakes) requires only a single, very fortuitous rafting experience. Iturralde-Vinent and MacPhee take a similar approach in their discussion of bottle drift studies. For example, they conclude that the results “strongly imply that, given existing surface-current patterns, flotsam emitted from the Orinoco and Amazon rivers is much more likely to end up in southeastern North America or Central America than in the Greater Antilles.” But this has been known for some time (Guppy, 1917) and no one has ever claimed that all flotsam leaving South America automatically ends up in the Greater Antilles! Iturralde-Vinent and MacPhee may consider the rate of one out of every seven drift bottles released (on average) landing in the Greater Antilles to be low, but others would consider this number to be surprisingly high. In summary, Iturralde-Vinent and MacPhee do not consider that the number of rafts potentially carrying animals from South America to anywhere during the Cenozoic must have been very large (i.e., millions). This is because flotsam is quite common and animals, including amphibians, have been observed riding on flotsam (Guppy, 1917; Boyd, 1962; King, 1962; Heatwole and Levins, 1972; Censky et al., 1998). The particular destination of rafts from South America that do not land on the Greater Antilles is not of interest to understanding the origin of the Antillean fauna. It is already assumed that the vast majority of rafts and their occupants perish, and it is already known that some currents do not lead to the Antilles. For Caribbean biogeography, the most likely source of flotsam that reaches the Greater Antilles is South America, considering both past and present current patterns. The arguments given by Iturralde-Vinent and MacPhee (1999) do not change that conclusion.

INCONSISTENCIES

AND

PROBLEMS

IN

MODEL

OF

MACPHEE

AND ITURRALDE-VINENT

MacPhee and Iturralde (1994) proposed that the Aves Ridge became a land bridge in the Oligocene at 30 to 27 mya. The precise timing was based partly on uplift of the region (followed by subsidence at 27 mya) and partly on the major mid-Cenozoic sea level drop that occurred at about 30 mya (Haq et al., 1987). Presumably, this sudden drop of ~160 m fully exposed the Aves Ridge. According to their land bridge model, fauna should not have arrived prior to that time if the land bridge was the primary explanation for the origin of these endemic mammals. However, the discovery of a 34 to 33 mya sloth in Puerto Rico (MacPhee and Iturralde-Vinent, 1995) created a conundrum because it predated the land bridge. Rather than reject the land bridge as an explanation for the presence of the Puerto Rican sloth, MacPhee and Iturralde modified their model by making the land bridge an earlier event (35 to 33 mya). As an explanation, they stated “either the sea level drop is not accurately dated or was not global, or for some other reason did not affect GAARlandia [land bridge] in the way originally imagined” (MacPhee and Iturralde-Vinent, 1995:20). In the most recent version of their model, Iturralde-Vinent and MacPhee (1999:27) claim that “general tectonic uplift coincided with a major eustatic sea level drop at ca. 35 Ma” (Miller et al., 1996). However, the sea level drop shown by Miller et al. (1996) at 35 mya was not a redating of the major Oligocene drop (Haq et al., 1987) used by MacPhee and Iturralde (1994), now considered to be 32.2 mya (Miller et al., 1993), but rather another sea level drop altogether. This inconsistent use of evidence shows that their paleogeographical model was influenced by their biogeographical model (i.e., the need to have the land bridge in place before the sloth fossil date). Another inconsistency involves the definition of the land bridge itself. It is defined as a “subaerial connection (whether continuous or punctuated by short water gaps) between a continent and an off-shelf island (or island arc)” (Iturralde-Vinent and MacPhee, 1999:52). This definition is consistent with a textual description earlier in the paper (p. 31): “we argue that exposure of the ridgecrest

Biogeography of the West Indies: An Overview

29

created, for a short time ca 33–35 Ma, a series of large, closely spaced islands or possibly a continuous peninsula stretching from northern South America to the Puerto Rico/Virgin Islands Block.” However, in other places the Aves Ridge land bridge is considered to be continuous: “central to the hypothesis is the argument, sustained at length in this paper, that the Cenozoic paleogeography of the Caribbean region strongly favored emplacement over land (as opposed to over water) only once in the past 65 Ma” (p. 53). Moreover, they clearly illustrate the land bridge as a fully continuous dry land connection, with no water gaps, much like the current Isthmus of Panama (Iturralde-Vinent and MacPhee, 1999:figures 6 and 12). The difference between an island chain and a continuous land bridge is fundamental for biogeography. The former will behave as a biotic filter allowing only selected taxa to cross, whereas the latter will permit a greater diversity of terrestrial life (a cross section of a biota) to enter. But, in addition, the existence of a single water gap implies that all organisms that crossed that gap must have done so by swimming or floating on flotsam (i.e., overwater dispersal). As noted above, that the Aves Ridge was at least a chain of islands during the mid-Cenozoic is normally assumed in discussions of Caribbean biogeography and is not a new concept. The possibility that it was a continuous land bridge also has been raised previously (Woods, 1990) but, as discussed elsewhere in this chapter, the current biological evidence does not support that alternative. Finally, Iturralde-Vinent and MacPhee (1999:56) acknowledge that the taxonomic composition of the West Indian fauna, including the Tertiary mammal fossil record, supports an origin by dispersal (“low initial diversity model”) and not the transfer over land of a diverse fauna in the Oligocene. They also acknowledge that at least some sloths were adapted to marine habitats (Muizon and McDonald, 1995). This raises the question, that if the faunal evidence favors a filter and not a dry land bridge, and the geological evidence is equivocal, then why is the dry land bridge favored?

EVIDENCE

AGAINST A

MID-CENOZOIC LAND BRIDGE

As with the proto-Antillean vicariance model, evidence against a mid-Cenozoic dry land bridge connection between South America and the Antilles is the depauperate nature of the Antillean fauna and molecular clock estimates of divergence times for terrestrial vertebrates. With regard to faunal composition, Iturralde-Vinent and MacPhee (1999) concede that “all Tertiary [mammal] taxa recovered to date from these islands appear to be closely related to clades known from the Quaternary, which favors the low initial diversity model [overwater dispersal]” (p. 56). They acknowledge that the presence of a more diverse fauna on Jamaica during the Eocene (Domning et al., 1997) is not relevant to the Aves Ridge land bridge model because Jamaica was isolated and underwent submergence during the Oligocene. Concerning the available molecular clock time estimates, the data do not support a clustering of divergences around 35 to 33 mya as would be predicted by the land bridge model. Instead, divergence times are scattered throughout the Cenozoic (Hedges, 1996b). Geological data neither support nor refute the suggestion of a fully continuous dry land bridge.

DISCUSSION AND CONCLUSIONS It is tempting to consider a complex problem such as the historical biogeography of the West Indies in terms of several alternative mechanisms. However, there is no reason to exclude any of the three models discussed above based on purely geological grounds. Nonetheless, the evidence reviewed in this chapter suggests that most lineages of West Indian vertebrates arrived by overwater dispersal during the Cenozoic. If most arrived by proto-Antillean vicariance in the late Cretaceous or by a land bridge (Aves Ridge) in the mid-Cenozoic, one would expect to see a more diverse fauna resembling a cross section of the continental fauna. However, the present fauna exhibits reduced higher-level diversity, and the fossil record suggests that this pattern was similar in the past. Molecular time estimates also indicate that nearly all lineages examined arrived in the Cenozoic and not the Cretaceous. They also do not support a mid-Cenozoic land bridge because they are

30

Biogeography of the West Indies: Patterns and Perspectives

scattered throughout the Cenozoic, rather than clustered. Finally, phylogenetic evidence points to an origin from South America for most nonvolant lineages examined, and this is congruent with water current patterns in the Atlantic and Caribbean today and throughout the Cenozoic. While there is sufficient evidence now to indicate that overwater dispersal is the general pattern, it is not possible to exclude other mechanisms. For example, it is quite possible that an early Antillean fauna, now extinct (Domning et al., 1997), arose through vicariance. Also, the frogs of the genus Eleutherodactylus appear to represent an ancient lineage in the West Indies that may have originated in the late Cretaceous or early Cenozoic (Hedges et al., 1992; Hedges, 1996b). Other extant lineages such as the xantusiid lizards and insectivores also may have arrived early in the history of the Antilles. Geological data and paleogeographical reconstructions will continue to be refined and contribute to our understanding of biogeography. Nonetheless, when such reconstructions of Earth history are influenced by particular biogeographical models, that bias affects their utility. Unfortunately, the most extensive work on paleogeography of the West Indies (Iturralde-Vinent and MacPhee, 1999) falls into this category. It shows a continuous dry land bridge in the mid-Cenozoic and no land connections prior to the late Eocene. However, as discussed above, geological evidence is inconclusive with regard to both major features of their reconstruction. In this case, the paleogeographical reconstructions of Iturralde-Vinent and MacPhee, taken literally, exclude proto-Antillean vicariance and offer a dry land corridor for emplacement of a mid-Cenozoic biota. In this sense, their biogeographical model and “paleogeographical reconstruction” are one and the same. It is more useful for biogeographers to base their conclusions on unbiased reconstructions of Earth history. Although some important Tertiary vertebrate fossils have been discovered in recent years in the Antilles, these represent only a small fraction of the endemic extant lineages. In addition, fossils provide only a minimum time of origin of a lineage. The major gap in our knowledge of Caribbean biogeography is not the fossil record — which will always remain fragmentary and biased — but the phylogeny and divergence times of the extant biota. If most lineages arrived in the late Cretaceous, vicariance is a strong possibility, whereas a mid-Cenozoic arrival could be explained by a land bridge. An origin during the last 25 million years would indicate an arrival only by overwater dispersal. Unfortunately, molecular time estimates are known only for selected lineages of vertebrates, and in most of those cases, they are based on an indirect measure of time from one gene (serum albumin). Ideally, we would like to know the relationships and times of origin from multiple nuclear and mitochondrial genes for all Antillean groups of organisms. Given the limited resources for systematics, this information may not be available for all groups even in the future. Nonetheless, a major advance should come in the next decade when such sequence data become more generally available. With these data and new fossil discoveries, we can look forward in the near future to resolving many of these long-unanswered questions in Caribbean biogeography.

ACKNOWLEDGMENTS I thank the many individuals who have assisted me in the field over the years. Carla Hass offered helpful comments on the manuscript and Anthony Geneva assisted with the figures. This research was supported by grants from the U.S. National Science Foundation.

LITERATURE CITED Adams, C. D. 1972. Flowering Plants of Jamaica. The University of the West Indies, Mona, Jamaica. Barbour, T. 1916. Some remarks upon Matthew’s “climate and evolution.” Annals of the New York Academy of Science 27:1–15. Bayuck, B. 1999. Taxonomists are an endangered species in Europe. Nature 401:321.

Biogeography of the West Indies: An Overview

31

Boyd, C. E. 1962. Waif dispersal in toads. Herpetologica 18:269. Censky, E. J., K. Hodge, and J. Dudley. 1998. Over-water dispersal of lizards due to hurricanes. Nature 395:556. Cockerell, T. D. A. 1924. A fossil cichlid from the Republic of Haiti. Proceedings of the United States National Museum 63:1–3. Comer, J. B. 1974. Genesis of Jamaican bauxite. Economic Geology 69:1251–1264. Darlington, P. J. 1957. Zoogeography: The Geographical Distribution of Animals. Wiley, New York. Dengo, G. and J. E. Case (eds.). 1990. The Geology of North America. Volume H: The Caribbean Region. The Geological Society of America, Boulder, Colorado. Dietz, R. S., J. C. Holden, and W. P. Sproll. 1970. Geotectonic evolution and subsidence of Bahama Platform. Geological Society of America Bulletin 81:1915–1928. Domning, D. P., J. Emry, R. W. Portell, S. K. Donovan, and K. S. Schindler. 1997. Oldest West Indian land mammal: rhinoceratoid ungulate from the Eocene of Jamaica. Journal of Vertebrate Paleontology 17:638–641. Donnelly, T. W. 1992. Geological setting and tectonic history of Mesoamerica. Pp. 1–13 in Quintero, D. and A. Aiello (eds.). Insects of Panama and Mesoamerica. Oxford University Press, Oxford. Donovan, S. K. and T. A. Jackson (eds.). (1994). Caribbean Geology: An Introduction. University of the West Indies Publishers’ Association, Kingston, Jamaica. Droxler, A. W., K. C. Burke, A. D. Cunningham, A. C. Hine, E. Rosencrantz, D. S. Duncan, P. Hallock, and E. Robinson. 1998. Caribbean constraints on circulation between Atlantic and Pacific Oceans over the past 40 million years. Pp. 160–191 in Crowley, T. J. and K. C. Burke (eds.). Tectonic Boundary Conditions for Climate Reconstructions. Oxford University Press, New York. Emanuel, K. A., K. Speer, R. Rotunno, R. Srivastava, and M. Molina. 1994. Hypercanes: a possible link in global extinction scenarios. Eos (Supplement) 75:409 (Abstract). Gentry, A. H. 1992. Tropical forest biodiversity: distributional patterns and their conservational significance. Oikos 63:19–28. Graham, A. 1993. Contribution toward a Tertiary palynostratigraphy for Jamaica: the status of Tertiary paleobotanical studies in northern Latin America and preliminary analysis of the Guys Hill Member (Chapelton Formation, Middle Eocene) of Jamaica. Pp. 443–461 in Wright, R. M. and E. Robinson (eds.). Biostratigraphy of Jamaica. Geological Society of Jamaica, Boulder, Colorado. Graham, A. and D. M. Jarzen. 1969. Studies in neotropical paleobotany. 1. The Oligocene communities of Puerto Rico. Annals of the Missouri Botanical Garden 56:308–357. Grimaldi, D. A. 1995. The age of Dominican amber. Pp. 203–217 in Anderson, K. B. and J. C. Crelling (eds.). Amber, Resinite, and Fossil Resins. American Chemical Society, Washington, D.C. Guppy, H. B. 1917. Plants, Seeds, and Currents in the West Indies and Azores. Williams and Northgate, London. Haq, B. U., J. Hardenbol, and P. R. Vail. 1987. Chronology of fluctuating sea levels since the Triassic. Science 235:1156–1166. Hass, C. A. 1991. Evolution and biogeography of West Indian Sphaerodactylus (Sauria: Gekkonidae): a molecular approach. Journal of Zoology 225:525–561. Hass, C. A. and S. B. Hedges. 1991. Albumin evolution in West Indian frogs of the genus Eleutherodactylus: Caribbean biogeography and a calibration of the albumin immunological clock. Journal of Zoology 225:413–426. Hass, C. A., S. B. Hedges, and L. R. Maxson. 1993. Molecular insights into the relationships and biogeography of West Indian anoline lizards. Biochemical Systematics and Ecology 21:97–114. Heatwole, H. and R. Levins. 1972. Biogeography of the Puerto Rican Bank: flotsam transport of terrestrial animals. Ecology 53:112–117. Hedges, S. B. 1996a. Historical biogeography of West Indian vertebrates. Annual Review of Ecology and Systematics 27:163–196. Hedges, S. B. 1996b. The origin of West Indian amphibians and reptiles. Pp. 95–127 in Powell, R. and R. Henderson (eds.). Contributions to West Indian Herpetology: A Tribute to Albert Schwartz. Society for the Study of Amphibians and Reptiles, Ithaca, New York. Hedges, S. B. and R. L. Bezy. 1993. Phylogeny of xantusiid lizards: concern for data and analysis. Molecular Phylogenetics and Evolution 2:76–87. Hedges, S. B., R. L. Bezy, and L. R. Maxson. 1991. Phylogenetic relationships and biogeography of xantusiid lizards inferred from mitochondrial DNA sequences. Molecular Biology and Evolution 8:767–780.

32

Biogeography of the West Indies: Patterns and Perspectives

Hedges, S. B., C. A. Hass, and L. R. Maxson. 1992. Caribbean biogeography: molecular evidence for dispersal in West Indian terrestrial vertebrates. Proceedings of the National Academy of Sciences, U.S.A. 89:1909–1913. Henderson, R. W. and S. B. Hedges. 1995. Origin of West Indian populations of the geographically widespread boa Corallus enydris inferred from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 4:88–92. Hildebrand, A. R. and W. V. Boynton. 1990. Proximal Cretaceous-Tertiary boundary impact deposits in the Caribbean. Science 248:843–847. Holcombe, T. L. and N. T. Edgar. 1990. Late Cretaceous and Cenozoic evolution of Caribbean ridges and rises with special reference to paleogeography. Atti Dei Convegni Lincei (International Symposium on Biogeographical Aspects of Insularity) 85:611–626. Hoorn, C., J. Guerrero, G. A. Sarmiento, and M. A. Lorente. 1995. Andean tectonics as a cause for changing drainage patterns in Miocene northern South America. Geology 23:237–240. Iturralde-Vinent, M. A. and R. D. E. MacPhee. 1996. Age and paleogeographical origin of Dominican amber. Science 273:1850–1852. Iturralde-Vinent, M. A. and R. D. E. MacPhee. 1999. Paleogeography of the Caribbean region: implications for Cenozoic biogeography. Bulletin of the American Museum of Natural History 238:1–95. King, F. W. 1962. The occurrence of rafts for dispersal of land animals into the West Indies. Quarterly Journal of the Florida Academy of Sciences 25:45–52. Larue, D. K. 1994. Puerto Rico and the Virgin Islands. Pp. 151–165 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. The University of the West Indies Publishers’ Association, Kingston, Jamaica. Liebherr, L. K. (ed.) 1988. Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. MacFadden, B. 1980. Rafting mammals or drifting islands? Biogeography of the Greater Antillean insectivores Nesophontes and Solenodon. Journal of Biogeography 7:11–22. MacPhee, R. D. E. and D. A. Grimaldi. 1996. Mammal bones in Dominican amber. Nature 380:489–490. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1994. First Tertiary land mammal from the Greater Antilles: an early Miocene sloth (Xenarthra, Megalonchidae) from Cuba. American Museum Novitates 3094:1–13. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1995. Origin of the Greater Antillean land mammal fauna, 1. New Tertiary fossils from Cuba and Puerto Rico. American Museum Novitates 3141:1–31. Malfait, B. T. and M. G. Dinkelman. 1972. Circum-Caribbean tectonic and igneous activity and the evolution of the Caribbean Plate. Bulletin of the Geological Society of America 83:251–272. Matthew, W. D. 1918. Affinities and origin of the Antillean mammals. Bulletin Geological Society of America 29:657–666. Maurrassee, F. J.-M. R. 1991. Impacts, tsunamis and the Haitian Cretaceous-Tertiary boundary layer. Science 252:1690–1693. Maxson, L. R. 1992. Tempo and pattern in anuran speciation and phylogeny: an albumin perspective. Pp. 41–57 in Adler, K. (ed.). Herpetology: current research on the biology of amphibians and reptiles. Society for the Study of Amphibians and Reptiles, Oxford, Ohio. Miller, K. G., G. S. Mountain, the Leg 150 Shipboard Party, and Members of the New Jersey Coastal Plain Project. 1996. Drilling and dating New Jersey Oligocene-Miocene sequences: ice-volume, global sea level, and Exxon records. Science 271:1092–1095. Miller, K. G., P. R. Thompson, and D. V. Kent. 1993. Integrated late Eocene-Oligocene stratigraphy of the Alabama coastal plain: correlation of hiatuses and stratal surfaces of glacioeustatic lowerings. Paleoceanography 8:313–331. Morgan, G. S. 1993. Quaternary land vertebrates of Jamaica. Geological Society of America, Memoir 182:417–442. Morgan, G. S. and C. A. Woods. 1986. Extinction and the zoogeography of West Indian land mammals. Biological Journal of the Linnean Society 28:167–203. Morrone, J. J. and J. V. Crisci. 1995. Historical biogeography: introduction to methods. Annual Reviews of Ecology and Systematics 26:373–401. Muizon, C. D. and H. G. McDonald. 1995. An aquatic sloth from the Pliocene of Peru. Nature 375:224–227. Nelson, G. and N. I. Platnick. 1981. Systematics and Biogeography: Cladistics and Vicariance. Columbia University Press, New York.

Biogeography of the West Indies: An Overview

33

Olson, S. L. 1978. A Paleontological Perspective of West Indian Birds and Mammals. Proceedings of the Academy of Natural Sciences, Philadelphia, Special Publication 13:99–117. Pereira, L. A., D. Foddai, and A. Minelli. 1997. Zoogeographical aspects of Neotropical Geophilomorpha. Entomologica Scandinavica Supplementum 51:77–86. Perfit, M. R. and E. E. Williams. 1989. Geological constraints and biological retrodictions in the evolution of the Caribbean Sea and its islands. Pp. 47–102 in Woods, C. A. (ed.). Biogeography of the West Indies. Sandhill Crane Press, Gainesville, Florida. Pindell, J. L. 1994. Evolution of the Gulf of Mexico and the Caribbean. Pp. 13–39 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. The University of the West Indies Publishers’ Association, Kingston, Jamaica. Poinar, G., Jr. and R. Poinar. 1999. The Amber Forest. Princeton University Press, Princeton, New Jersey. Powell, R., R. W. Henderson, K. Adler, and H. A. Dundee. 1996. An annotated checklist of West Indian amphibians and reptiles. Pp. 51–93 in Powell, R. and R. Henderson (eds.). Contributions to West Indian Herpetology: A Tribute to Albert Schwartz. Society for the Study of Amphibians and Reptiles, Ithaca, New York. Pregill, G. K. 1986. Body size of insular lizards: a pattern of Holocene dwarfism. Evolution 40:997–1008. Pregill, G. K. 1999. Eocene lizard from Jamaica. Herpetologica 55:157–161. 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. Pregill, G. K., R. I. Crombie, D. W. Steadman, L. K. Gordon, F. W. Davis, and W. B. Hilgartner. 1992. Living and late Holocene fossil vertebrates, and the vegetation of the Cockpit Country, Jamaica. Atoll Research Bulletin 353:1–19. Robinson, E. 1994. Jamaica. Pp. 111–127 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. University of the West Indies Publishers’ Association, Kingston, Jamaica. Rosen, D. E. 1975. A vicariance model of Caribbean biogeography. Systematics Zoology 24:431–464. Rosen, D. E. 1985. Geological hierarchies and biogeographic congruence in the Caribbean. Annals of the Missouri Botanical Garden 72:636–659. Scharff, R. F. 1912. Distribution and Origin of Life in America. Macmillan, New York. Schubart, C. D., R. Diesel, and S. B. Hedges. 1998. Rapid evolution to terrestrial life in Jamaican crabs. Nature 393:363–365. Schuchert, C. 1935. Historical geology of the Antillean-Caribbean region. John Wiley & Sons, New York. Schwartz, A. and R. W. Henderson. 1991. Amphibians and Reptiles of the West Indies. University of Florida Press, Gainesville. Simpson, G. G. 1956. Zoogeography of West Indian land mammals. American Museum Novitates 1759:1–28. Smith, A. G., D. G. Smith, and B. M. Funnell. 1994a. Atlas of Mesozoic and Cenozoic Coastlines. Cambridge University Press, Cambridge. Smith, D. S., L. D. Miller, and J. Y. Miller. 1994b. The Butterflies of the West Indies and South Florida. Oxford University Press, Oxford. Wallace, A. R. 1881. Island Life. Harper, New York. Williams, E. E. 1989. Old problems and new opportunities in West Indian biogeography. Pp. 1–46 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Wilson, E. O. 1985. Invasion and extinction in the West Indian ant fauna: evidence from Dominican amber. Science 229:265–267. Wilson, E. O. 1992. The Diversity of Life. Harvard University Press, Cambridge, Massachusetts. Woods, C. A. 1990. The fossil and recent mammals of the West Indies: an analysis of the origin, evolution, and extinction of an insular fauna. Atti Dei Convegni Lincei (International Symposium on Biogeographical Aspects of Insularity) 85:642–680. Woods, C. A. and J. A. Ottenwalder. 1992. The Natural History of Southern Haiti. Florida Museum of Natural History, Gainesville, Florida.

Change in the 3 Climate Circum-Caribbean (Late Pleistocene to Present) and Implications for Regional Biogeography Jason H. Curtis, Mark Brenner, and David A. Hodell Abstract — We present high-resolution paleoclimate reconstructions for the circum-Caribbean region spanning the late Pleistocene to present. The stable oxygen isotope signature (δ18O) of carbonate shells from 14C-dated lake sediment cores served as a proxy for shifts in the evaporation/precipitation ratio (E/P). Inferred changes in temperature over the Pleistocene–Holocene transition were based on pollen sequences. Both continental and insular archives provide a similar, coherent picture of regional climate change since the late Glacial. Late Pleistocene conditions were cool and arid, as indicated by dry lake beds, a cold/xeric-adapted flora, and relatively positive oxygen isotopic signatures. The transition into the early Holocene was marked by warmer, moister climate. Regional lakes filled, tropical forests expanded, and δ18O values in shells declined. The middle Holocene moist period terminated about 3000 BP with the onset of drier conditions. Long-term, millennial-scale variations in moisture availability were tied to the intensity of the annual cycle and position of the Inter-Tropical Convergence Zone (ITCZ). The intensity of the annual cycle was controlled by solar insolation which was, in turn, governed by orbital forcing. Late Holocene paleoclimate records also contain evidence for pronounced climate shifts of short duration, some of which probably impacted human cultures in the region. These rapid climate changes were not orbitally driven and may have been a consequence of aerosols injected into the atmosphere during volcanic eruptions, solar input variability, changes in ocean circulation, or deforestation. Our data require that discussions of West Indian biogeography consider the role that climate changes since the last Glacial played in shaping the modern distribution and abundance of the circum-Caribbean flora and fauna.

INTRODUCTION Discussions of West Indian biogeography have been preoccupied with the influence of long-term geological processes, such as plate tectonics and land emergence/submergence, on faunal colonization and extinction in the Caribbean islands (Williams, 1989). Biogeographers have proposed several models that invoke past land bridges, cross-water dispersal, and vicariance to explain the initial arrival of insular animal taxa and their contemporary distributions on geographically isolated islands. Hypothesized modes of colonization, as well as subsequent disappearances of some taxonomic groups, are sometimes tested using the fossil record. Clearly, age-old geological changes have left their mark with respect to the modern distribution and abundance of both animals and plants in the region. Interpretations of modern West Indian biogeography must, however, consider more recent factors that have affected the Caribbean biota. Humans, for example, were responsible for biotic introductions and extinctions over the last few millennia. Late Pleistocene and Holocene climate changes in the region have also played an important role in determining the modern distribution of organisms. Climate changes have influenced biotic communities directly and indirectly, through impact on human cultures. 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

35

36

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 1 Map of the circum-Caribbean region indicating the location of lakes used in this study. The numbers on the map indicate the positions of the following lakes: (1) Lakes Punta Laguna and Coba; (2) Lake Chichancanab; (3) Lake Peten-Itza; (4) Lake Valencia; (5) Lake Miragoâne.

TABLE 1 General Information about the Circum-Caribbean Lakes Used in This Study

Latitude Longitude Area (km2) Altitude (m) Date cored (d-m-y) Core length (m) Core site Zb (m) Zb max (m) Lake water δ18O 14C dating a b c d e f

Peten-Itza

Valencia

Miragoâne

Chichancanab

Coba

Punta Laguna

16°55′N 89°50′W 100 ~100a 6-VII-93 5.45 7.55 160 c 2.6‰ terr e

10°10′N 67°52′W 350 402 16-VII-94 5.68 9.4 37 3.2‰ terr

18°24′N 73°05′W 7 20 25, 28-VII-85 7.53 42 42 —d aqua f

19°50′N 88°45′W 10 ~20a 25-VI-93 4.90 6.9 12.5 3.2‰ terr

20°30′N 87°44′W 0.55 ~15a 14, 15-VIII-80 8.80 4.6 ~8 c 0.5‰ terr/aqua

20°38′N 87°30′W 0.9 14 22-VI-93 6.30 6.3 12 c 0.93‰ terr

Altitude estimated. Z = depth. Depth estimate; lakes lack detailed bathymetry. — = no isotopic information. terr = terrestrial matter used in the construction of the age model. aqua = aquatic matter used in the construction of the age model.

We present paleolimnological evidence for climate shifts that occurred in the circum-Caribbean area from the late Pleistocene to present, discuss the possible factors that may have driven these climate changes, and provide some examples of biotic responses to the alterations in temperature and moisture availability. The evidence comes from studies of lake sediment cores collected at both insular and mainland sites. Lake sediments can provide high-resolution records of past climate because they accumulate at relatively rapid rates (~0.5 to 5.0 mm yr –1) and frequently contain proxy indicators of past climate conditions. For example, pollen grains preserved in accreting lake mud can be used to “reconstruct” past vegetation assemblages, which in turn reflect prehistoric climate. Likewise, the stable oxygen isotope (δ18O) signature of sedimented carbonate shells can be utilized to infer past shifts in the relation between evaporation and precipitation (E/P). This regional paleoclimate synthesis compares δ18O records from lake sediment cores collected at several sites throughout the circum-Caribbean. We used oxygen isotope records from multiple

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

37

sites to evaluate similarities and differences in climate evolution over a broad geographical area. Principal waterbodies examined in this study included Lakes Punta Laguna, Chichancanab, and Coba on the Yucatan Peninsula, Mexico; Lake Peten-Itza, Peten, Guatemala; Lake Valencia, northern Venezuela; and Lake Miragoâne, southwestern Haiti. Study lakes are distributed around the Caribbean Sea between about 10 and 20° north latitude and between 67 and 90° west longitude (Figure 1, Table 1). Lake Miragoâne is the only insular record reported, reflecting, in part, the limited number of appropriate, natural freshwater study lakes in the West Indies (Candelas and Candelas, 1963). The investigated lakes lie at relatively low altitude, varying in elevation from 14 to 402 m above sea level. They range in surface area from 0.55 (Coba) to 350 km2 (Valencia) (Table 1). Maximum depths for the water bodies are between about 8 (Coba) and ~160 m (Lake Peten-Itza). Lakes Miragoâne and Peten-Itza are crypto-depressions, i.e., deepest water in the lakes lies below modern sea level. Sediment cores were recovered from water depths ranging between about 4.6 and 42 m. Finally, all the lakes are effectively closed hydrologically, losing most of their water to evaporation. The exceptions, Lakes Miragoâne and Valencia, at present or historically have lost a fraction of their annual hydrologic budgets to overland outflow.

USING OXYGEN ISOTOPES IN FRESHWATER CARBONATE SHELLS TO INFER PAST CLIMATE The three naturally occurring stable isotopes of oxygen are 16O, 17O, and 18O. The lightest isotope (16O) is most common, representing 99.7630% of the isotope pool. The rarest (17O) constitutes only 0.0375% of the total, while 18O accounts for 0.1995%. The three isotopes differ in mass and therefore behave differently when they enter into physical, chemical, and biological processes in the environment. These differential behaviors among the oxygen isotopes and their consequent changing relative abundance in the environment are exploited for paleoclimate reconstructions. The stable isotope (δ18O) signal in sedimented freshwater carbonate shells can be used to infer past climate conditions. The rationale for employing this approach to paleoclimate reconstruction, discussed in detail by Talbot (1990), Chivas et al. (1993), Curtis and Hodell (1993), and Holmes (1996), is reviewed briefly here. The oxygen isotope ratio (18O/16O) in sedimented carbonate shells of freshwater ostracods and gastropods is governed by two factors: (1) the δ18O of the lake water at the time the organism lived and (2) the temperature at which carbonate precipitation occurred. There is little evidence for major changes in mean temperature over the last 10,000 years. The major determinant of δ18O in freshwater shell carbonate during the Holocene has therefore been the δ18O of the lake water from which it was precipitated. The 18O/16O ratio of lake water has, in turn, been controlled by hydrologic variables. Assuming that the δ18O of regional rainfall remained fairly constant during the Holocene, relative changes in hydrologic inputs and outputs have governed the in-lake δ18O. In tropical, closed-basin lakes, i.e., those that lack significant overland outflows, the hydrologic budget and δ18O of lake water are most influenced by the relation between evaporation (E) and precipitation (P) (Fontes and Gonfiantini, 1967; Gasse et al., 1990; Lister et al., 1991). During dry periods (high E/P), the 18O/16O ratio in the water column increases, i.e., δ18O becomes more positive because H216O, with its higher vapor pressure, is preferentially lost to evaporation. Conversely, during moist periods (low E/P), the lake water 18O/16O ratio declines, i.e., δ18O becomes more negative. Ostracods and snails preserve within their carbonate shells a record of the E/P conditions that prevailed during their lifetimes. They are short-lived and on death their shells are incorporated into the lake bottom sediments, thereby preserving a stratigraphic archive of past climate change (E/P). The oxygen isotopic signature (δ18O) in sedimented shell material can be determined by mass spectrometry. Results are presented in standard delta notation, an expression of the isotopic composition of the shell material relative to the Vienna PeeDee Belemnite (VPDB) standard: 18

16

18

16

( O ⁄ O ) sample – ( O ⁄ O ) VPDB 18 - × 1000 δ O = ---------------------------------------------------------------------------18 16 ( O ⁄ O ) VPDB

38

Biogeography of the West Indies: Patterns and Perspectives

The resulting value is basically an expression of the departure of the 18O/ 16O ratio in the sample from that of the standard, expressed on a per mil (‰) basis. More positive values indicate relatively higher E/P (drier) conditions, while more negative values indicate relatively lower E/P (moister) conditions at the time the organism lived. Generally, δ18O stratigraphies from lake sediment cores are developed using shells of the same taxon collected at numerous depths throughout a profile. Various taxa can potentially fractionate differentially, or “discriminate” among the oxygen isotopes. This phenomenon, referred to as the “vital effect,” makes it difficult to interpret a composite δ18O stratigraphy generated using multiple taxa. In situations where a single taxon is not found throughout the record, oxygen isotope ratios can be measured in multiple taxa from overlapping depths to assess or correct for “vital effect.” Multiple δ18O stratigraphies from a single lake can sometimes be generated using several taxa that occupy different ecological niches. In such cases, if stratigraphic trends for all taxa are similar, one can have greater confidence in the paleoclimatic interpretations.

DETERMINING THE TIMING OF CLIMATE CHANGES Changes in the past relation between evaporation and precipitation (E/P) can be inferred from stratigraphic shifts in the δ18O of sedimented carbonate shells. To determine the timing of climate shifts requires reliable sediment chronologies. Age/depth relations for sediment cores in this study were established primarily using accelerator mass spectrometry (AMS) radiocarbon dating. In addition, three conventional radiocarbon dates were obtained from the Lake Coba core and one conventional 14C date was run on the section from Lake Miragoâne (Table 2). Counting errors associated with individual 14C dates were ±90 years for AMS samples, while error for the four conventional 14C dates were ±160 years (Table 2). Radiocarbon dates on terrestrial material (e.g., wood, charcoal, seeds) were used preferentially to develop age/depth models. Shells and organic matter of aquatic origin, as well as bulk sediment from water bodies in karst terrain, are susceptible to hard-water-lake dating error (Deevey and Stuiver, 1964). In hard-water lakes, 14C dates on lacustrine shells and organic matter can be affected by the input of “old” bicarbonate-carbon that is derived from dissolution of ancient limestone in the watershed and is devoid of 14C. This “old” carbon can be incorporated into organic matter during photosynthesis and passed up the food chain. It may also be fixed in the carbonate of calcite and aragonite shells. Bulk sediment presents dating difficulties because it can contain “old,” detrital carbonate. Detailed discussion of the core chronologies presented here (Table 2) are found in Curtis and Hodell (1993), Curtis et al. (1996, 1998, 1999), Hodell et al. (1991, 1995), Leyden et al. (1998), and Whitmore et al. (1996). Temporal resolution among the isotopic records from the six lakes varies as a consequence of inter-core differences in sedimentation rates and stratigraphic sampling intervals. There is about a threefold difference with respect to long-term, average sedimentation rate between the water body with the highest rate of sediment accumulation (Punta Laguna = 1.6 mm yr –1) and the lake with the lowest sediment accumulation rate (Valencia = 0.49 mm yr –1) (Table 3). Cores from the study lakes were sampled at 1-cm intervals, with the exception of the Lake Coba section, which was sampled at 5-cm intervals. Carbonate shells were abundant in cores from Lakes Punta Laguna, Miragoâne, Peten-Itza, and Chichancanab. Consequently, δ18O records from those basins are relatively continuous (Table 3, Figure 2). In contrast, shell material was encountered sporadically in the Coba and Valencia cores, resulting in discontinuous, lower-resolution δ18O records. The time resolution at which climatic changes can be resolved in the records differs among cores. For example, high sedimentation rates coupled with close-interval sampling enabled us to resolve multi-decadal events in the Punta Laguna core. The Valencia, Miragoâne, Peten-Itza, and Chichancanab cores yielded results that allow us to resolve climatic shifts that occur over time frames of 50 to 100 years. In contrast, slower sedimentation rate combined with broad sampling intervals in the Lake Coba core permitted only centennial–millennial resolution of climate events.

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

TABLE 2 Radiocarbon Dates for Circum-Caribbean Cores Sample Type

Depth (cm)

Accession Number

Lake Punta Laguna, Yucatan, Mexico core 22-VI-93 Aquatic gastropods 25 OS-6549 Terrestrial wood 81 OS-6550 Aquatic gastropods 83 OS-6551 Terrestrial wood 145 OS-10009 Terrestrial wood 197 OS-10010 Aquatic gastropods 246 OS-6552 Terrestrial wood 380 OS-6553 Terrestrial wood 494 OS-6554 Aquatic gastropods 600 OS-5760 Peten-Itza, Peten, Guatemala core 6-VII-93 Aquatic shell 28–30 OS-6555 Terrestrial wood 57 OS-6556 Charcoal 73–76 OS-10004 Charcoal 107.5 OS-10005 Terrestrial wood 113.5 OS-10006 Aquatic shell 190 OS-6557 Terrestrial wood 255.5 OS-10008 Terrestrial wood 340 OS-6558 Terrestrial wood 505 OS-6559 Aquatic shell 506 OS-5761 Terrestrial wood 527 OS-6560 Valencia, Venezuela core 16-VII-94 Terrestrial wood 31 OS-8854 Ostracods 31.5 OS-8855 Terrestrial wood 69.5 OS-8856 Ostracods 69.5 OS-8857 Terrestrial wood 139 OS-8858 Terrestrial wood 171 OS-8859 Terrestrial wood 231 OS-10011 Charcoal 249 OS-10012 Ostracods 253 OS-8860 Ostracods 378 OS-8861 Terrestrial wood 486 OS-8862 Terrestrial wood 511 OS-8863 Terrestrial wood 560 OS-8864 Chichancanab, Yucatan, Mexico core 25-VI-93 Pyrgophorus 15 CAMS-12900 Terrestrial seed 65 OS-3545 Pyrgophorus 65 OS-3443 Pyrgophorus 103 OS-3446 Bivalves 142 OS-2148 Pyrgophorus 238 OS-3445 Pyrgophorus 314 OS-3444 Mixed gastropods 350 OS-2051 Mixed gastropods 385 OS-2729 Bivalves 406 OS-2055 Terrestrial charcoal 421 OS-2157 Terrestrial charcoal 421 CAMS-12780 Terrestrial charcoal 421 CAMS-12781 Land snail 472 OS-2052

Radiocarbon Age (yr BP) 1320 ± 25 610 ± 50 a 1930 ± 70 965 ± 25a 1530 ± 50 a 3160 ± 30 2440 ± 45a 2840 ± 30a 3720 ± 40 985 ± 30 75 ± 25a 815 ± 25a 1260 ± 30 a 1660 ± 30 a 4200 ± 30 4870 ± 80 a 5600 ± 35a 8480 ± 55a 10250 ± 50 8840 ± 55a 185 ± 40 a 485 ± 30 1730 ± 35a 1810 ± 35 3310 ± 35a 4830 ± 40a 7670 ± 40a 8330 ± 85a 7990 ± 45 9370 ± 80 10200 ± 55a,b 9960 ± 70 a,b 12400 ± 60 a 1550 ± 60 1140 ± 35a 1600 ± 30 3200 ± 40 5210 ± 30 7100 ± 30 9040 ± 65 8680 ± 45 9530 ± 60 9500 ± 50 7560 ± 35a 7600 ± 60 a 7460 ± 60 a 9180 ± 50 a

Corrected Age (yr BP)

39

40

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 (continued) Radiocarbon Dates for Circum-Caribbean Cores Sample Type

Depth (cm)

Accession Number

Coba, Quintana Roo, Mexico core 14, 15-VIII-80 Ostracods 75 OS-59 Ostracods 335 OS-60 Bulk sedimentc 395–400 Beta-63470 Ostracods 575 OS-61 Bulk sedimentc 620–624 Beta-63471 Ostracods 720 OS-62 Bulk sedimentc 841 –846 Beta-63472 Terrestrial wood 865 –870 OS-10014 Miragoâne, Haiti core 25, 28-VII-85 210Pb 8 Ostracods 22 AA-6703 Ostracods 216 AA-5814 Ostracods 233 AA-6704 Terrestrial wood 233 AA-6705 Ostracods 321 AA-5815 Ostracods 418 AA-5816 Ostracods 520 AA-5817 Ostracods 622 AA-5818 Ostracods 671 AA-5369 Ostracods 718 AA-5952 Bulk sedimentc 753 GX-13055 a b c

Radiocarbon Age (yr BP)

Corrected Age (yr BP)

1780 ± 65 2600 ± 50 2600 ± 100 3880 ± 70 4440 ± 80 6880 ± 40 7410 ± 80 7600 ± 35a

460a 1280a 1280a 2560a 3120a 5560a 6090a

129 ± 40 1085 ± 60 2780 ± 55 2680 ± 60 1655 ± 60 4110 ± 60 4780 ± 60 6945 ± 65 9005 ± 75 9700 ± 90 10300 ± 85 10230 ± 160

129a 85a 1780a 1680a 1655a 3110a 3780a 5945a 8005a 8700a 9300a 10230a

Used in chronology of sediment core. Averaged for 10,080 14C yr BP at 498.5 cm depth in the Lake Valencia core. Conventional 14C date.

TABLE 3 Sedimentation Rates, Sample Spacing, and Sample Interval for Circum-Caribbean Cores Lake

Sedimentation Rate (mm/year)

Sample Interval (cm)a

Sample Spacing (years/sample)a

Punta Laguna Coba Miragoâne Peten-Itza Chichancanab Valencia

1.6 1.1 0.73 0.59 0.56 0.49

1 5 1 1 1 1

5.4 90 17 15 17 27

a

Isotope samples.

Robust chronologies and high sampling resolution permit good correlation of climate records from Lakes Punta Laguna, Peten-Itza, Valencia, and Chichancanab. Dating uncertainties associated with hard-water-lake correction in the Miragoâne and Coba cores suggest that correlations with the other four sections are only possible at broad timescales.

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

41

FIGURE 2 Oxygen isotopic composition of carbonate shell material (ostracods and/or gastropods) for Lakes Valencia, Miragoâne, Peten-Itza, Chichancanab, Coba, and Punta Laguna vs. radiocarbon years BP. For the Lake Valencia record,  = Heterocypris communis, + = Cytheridella boldi, × = Cypria obtua, and ▫ = Pyrgophorus sp. For the Lake Miragoâne record, + = Candona n. sp. For the Lake Peten-Itza record,  = Candona sp. and + = Cytheridella ilosvayi. In the Lake Chichancanab record,  = Cyprinotus cf. salinus and + = Physocypria xanabanica. For the Lake Coba record,  = Physocypria xanabanica and + = Cytheridella ilosvayi. In the Punta Laguna record, + = Cytheridella ilosvayi.

LATE PLEISTOCENE AND HOLOCENE CLIMATE CHANGE IN THE CIRCUM-CARIBBEAN LATE PLEISTOCENE ARIDITY Prior to ~10,500 14C yr BP, our study basins were effectively dry, precluding recovery of lacustrine sediment cores for oxygen isotopic analysis. Nevertheless, low water levels or complete desiccation of circum-Caribbean basins until ~10.5 to 8.0 14C kyr BP points to regional, late Pleistocene aridity. Paleoenvironmental studies from many lowland sites in the Northern Hemisphere Neotropics indicate that the end of the last Glacial was much drier than present (Covich and Stuiver, 1974; Bradbury, 1979; Salgado-Labouriau, 1980; Bradbury et al., 1981; Lewis and Weibezahn, 1981; Binford, 1982; Deevey et al., 1983; Leyden, 1984; Bush and Colinvaux, 1990; Piperno et al., 1990; Bush et al., 1992; Leyden et al., 1993, 1994; Street-Perrott et al., 1993; Brenner, 1994; Holmes et al., 1995). Lacustrine δ18O records from Pleistocene-age lake deposits are available from only two circumCaribbean sites, Wallywash Great Pond, Jamaica and Lake Quexil, Peten, Guatemala. Both indicate that the evaporation to precipitation ratio (E/P) was high during the late Pleistocene (Street-Perrott et al., 1993; Leyden et al., 1993, 1994; Holmes, 1998). Dry conditions, combined with relatively colder temperatures, clearly impacted regional vegetation distribution. Pollen evidence from the 36,000-year-old Lake Quexil record indicates extremely arid conditions during marine isotope Stage 2 (~24 to 12 kyr BP), with temperature depression on the order of 6.5 to 8.0°C relative to

42

Biogeography of the West Indies: Patterns and Perspectives

the present (Leyden et al., 1993). Mesic tropical forest was completely absent in the northern Guatemalan lowlands from 36 to 10.5 kyr BP (Leyden, 1984; Leyden et al., 1993, 1994). Likewise, the pollen record from Lake Valencia, Venezuela indicates that moisture availability was severely limited in the latest Pleistocene (Leyden, 1985). Markgraf (1989, 1993) reviewed paleoenvironmental results from lowland tropical and subtropical sites that indicate arid conditions from ~13,000 to ~10,000 years BP. Dry late glacial conditions extended into the northern hemisphere subtropics, as revealed by pollen records from Florida lake basins (Watts, 1975; Watts and Stuiver, 1980; Watts and Hansen, 1988, 1994; Watts et al., 1992; Grimm et al., 1993). Low moisture availability is coupled with cool temperatures in the late Pleistocene through ocean–atmosphere interactions. Low atmospheric water vapor during the last Glacial, as reflected by dry lake basins, may have decreased atmospheric absorption of infrared energy, leading to global cooling (Broecker, 1995). Air temperature and humidity are closely tied in a positive feedback loop. Lower temperatures decrease humidity, thereby reducing the greenhouse effect and amplifying cooling. Several lines of evidence suggest that both tropical air and sea surface temperatures decreased during the last glacial maximum (LGM) by about 5 to 8°C (Markgraf, 1989; Piperno et al., 1990; Bush and Colinvaux, 1990; Seltzer, 1990; Guilderson et al., 1994; Thompson et al., 1995; Stute et al., 1995; Colinvaux et al., 1996a, 1996b). Relatively cooler late Pleistocene tropical sea surface temperatures would have reduced oceanic evaporation, resulting in lower atmospheric moisture.

EARLY LAKE FILLING Following the most arid phase of the late Pleistocene, lake basins in the circum-Caribbean began to fill with water. The timing of initial lake filling was estimated from radiocarbon dates on samples just above or below the oldest lacustrine deposits. Inception of lacustrine sedimentation in our study lakes varied from ~10,500 to ~7,600 14C yr BP. The coring site in Lake Miragoâne, located at the deepest point in the lake (~42 m), was covered with water by ~10,500 14C yr BP (Higuera-Gundy, 1991, 1999; Hodell et al., 1991; Curtis and Hodell, 1993). The Lake Valencia core site, in ~9.4 m of water and some ~28 m above the maximum depth of the lake, was submerged by ~10,000 14C yr BP (Curtis et al., 1999). Previous studies of deepwater cores from Lake Valencia showed that these profundal sites were first covered by water about 10,500 14C yr BP (Bradbury et al., 1981; Lewis and Weibezahn, 1981; Binford, 1982; Leyden, 1985), suggesting it took ~500 years for water level to rise >28 m. Relatively rapid lake-level rise is also confirmed in Lake Quexil, Guatemala, by dates from the bottoms of both deepwater and shallow-water cores. Although Quexil, with a maximum depth of ~32 m, appears to have held at least some water through the arid late Glacial, lake level rose rapidly after about 10.5 14C kyr BP (Deevey et al., 1983). A core taken in about 6 m of water has a basal age of 8.4 14C kyr BP (Vaughan et al., 1985), indicating that near-modern levels were achieved within about 2 millennia. Lakes Chichancanab and Coba and the southern basin of Lake Peten-Itza filled somewhat later, in the early Holocene. The 5.45-m core from Peten-Itza was collected in ~7.5 m of water. Dates from the base of the section indicate the core site was inundated ~9,000 14C yr BP (Curtis et al., 1998). Shallower Lake Chichancanab filled at ~8,200 14C yr BP (Hodell et al., 1995) and Coba water level rose at ~7,600 14C yr BP (Leyden et al., 1996, 1998; Whitmore et al., 1996). The timing of filling for these shallow basins on the Yucatan Peninsula coincides with filling of shallow lakes in Florida (Watts, 1969; Watts and Hansen, 1988). Two processes, operating alone or in concert, were probably responsible for regional lake level rise. First, initial lake filling in the circum-Caribbean region was likely a consequence of increased moisture availability. Pollen studies confirm that climate conditions in the region became more mesic in the early Holocene (Salgado-Labouriau, 1980; Leyden, 1984, 1985, 1987; Vaughan et al., 1985; Islebe et al., 1996a; Higuera-Gundy et al., 1999). Second, water level rise in low-elevation lakes in karst terrain also may have been controlled by rising sea level, which in turn raised water

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

43

levels in freshwater aquifers (Watts, 1975; Fairbanks, 1989; Watts and Hansen, 1994). At the end of the last Ice Age, flow of glacial meltwater to the world oceans raised eustatic sea level by 121 ±5 m (Fairbanks, 1989). Rising sea level may well have influenced water levels in low-elevation, karst basins Chichancanab, Punta Laguna, Coba, and Miragoâne. It is unlikely, however, that sea level rise directly affected water stage in Lakes Valencia and Peten-Itza, which are situated at higher elevations (Table 2). Early Holocene water level rise in Lakes Chichancanab and Coba was probably due to the combined effects of sea level rise and increased rainfall. Both lakes filled during the second stage of the most recent deglaciation (Termination 1b) that began ~10,000 BP and lasted until ~6,000 BP (Fairbanks, 1989). Using ages for the oldest lacustrine sediments in Lakes Chichancanab and Coba (Hodell et al., 1995; Leyden et al., 1996, 1998; Whitmore et al., 1996), together with Holocene sea level curves (Fairbanks, 1989), we estimate the two basins filled when sea level was ~25 and ~18 m, respectively, below current level. Oldest sediments retrieved from the shallow-water site in Lake Punta Laguna are only ~3,320 14C years old, so it is impossible to establish the timing of initial lake filling using this section. On the basis of results from other northern Yucatan lakes, including Chichancanab (Hodell et al., 1995), Coba (Whitmore et al., 1996; Leyden et al., 1998) and San Jose Chulchaca (Leyden et al., 1996), it is probable that Punta Laguna first held permanent water between 8,000 and 7,000 years BP. A core collected in 16 m water from Punta Laguna in May 2000 penetrated to bedrock. Radiocarbon dates from the base of the section will resolve the question of the timing of basin filling. It is unclear whether Lake Miragoâne filled primarily as a result of rising sea level or increasing moisture availability. The lake began to fill ~10,500 14C yr BP, when sea level was ~65 meters below current level (Fairbanks, 1989; Hodell et al., 1991; Curtis and Hodell, 1993). Filling began in Lake Miragoâne earlier than it did in low-lying karst Lakes Coba and Chichancanab. Sea level rise would be expected to have affected Lake Miragoâne earlier than the other low-elevation study lakes because Miragoâne is a cryptodepression and its bottom lies ~20 m below modern sea level. The timing of Lake Miragoâne filling may be imprecise due to dating uncertainties associated with hard-water-lake error. In any event, one consequence of lake filling in the Caribbean region was that new habitats were created for aquatic organisms.

EARLIEST HOLOCENE (~10,500 TO ~8,500 14C YR BP) Only two of our study basins, Miragoâne and Valencia, yielded isotopic records for the earliest Holocene (~10,500 to ~8,500 14C yr BP) (Figure 2). During this time period, the four other study basins remained dry. Earliest Holocene δ18O values from the Miragoâne core (~10,500 to 8,500 14C yr BP) and Valencia section (~10,000 to 8,600 14C yr BP) were relatively positive, indicating the highest E/P (driest) conditions in the records (Figure 2). Trace metal (Sr/Ca and Mg/Ca) data from the Miragoâne core reflect high lakewater salinity during the earliest Holocene (Curtis, 1992; Curtis and Hodell, 1993). Pollen reconstructions from Miragoâne (Higuera-Gundy, 1991, 1999; Hodell et al., 1991) and Valencia (Salgado-Labouriau, 1980; Leyden, 1985) confirm arid latest Pleistocene conditions. Increasing moisture availability is associated with the beginning of the Holocene. The exact timing for the onset of mesic Holocene conditions in the northern hemisphere Neotropics remains unresolved. Records from some sites indicate that wetter conditions began at the Pleistocene/ Holocene boundary or earlier (Bush et al., 1992), whereas data from other localities suggest moisture availability did not increase until ~8,000 years BP (Leyden et al., 1993; Street-Perrott et al., 1993). Wetter conditions in Panama commenced as early as 10,500 years BP (Bush et al., 1992). In Jamaica, cool, dry conditions dominated until at least 9,500 years BP (Street-Perrott et al., 1993). Pollen evidence from Lake Quexil (Leyden et al., 1993) and other lowland sites in the region suggest that

44

Biogeography of the West Indies: Patterns and Perspectives

between 10,000 and 8,500 years BP, late Pleistocene grasslands were replaced by more mesic vegetation, indicating both increased warming and moisture availability (Markgraf, 1989).

EARLY TO MIDDLE HOLOCENE (~8,500 TO ~3,000 14C YR BP) By ~7,500 14C yr BP, Lakes Valencia, Peten-Itza, Chichancanab, Miragoâne, Punta Laguna, and Coba held water and were accumulating lacustrine sediments at their respective core sites. Oxygen isotope ratios in cores from Lakes Peten-Itza, Chichancanab, and Miragoâne were lowest during the early to middle Holocene (~8,500 to ~3,000 14C yr BP) (Figure 2), indicating low E/P (relatively moist conditions) and high lake levels. Lake Valencia experienced high water levels during much of the period from ~8,200 to ~3,000 14C yr BP and overflowed at a point 25 m above the present lake level (Bradbury et al., 1981; Lewis and Weibezahn, 1981; Binford, 1982; Curtis et al., 1999). When large volumes of water exited the basin through the outflow, the lake became hydrologically “open.” During that period, minor changes in E/P had little effect on the δ18O composition of Lake Valencia water. Consequently isotope-based paleoclimatic inferences for the time span are less reliable. Nevertheless, the fact that Lake Valencia was overflowing confirms that moist conditions prevailed. This period of low E/P coincides with the early to middle Holocene moist period that has been recognized by others at sites throughout the Northern Hemisphere Neotropics, including Mexico (Covich and Stuiver, 1974), Guatemala (Deevey et al., 1983; Leyden, 1984; Islebe et al., 1996b), Panama (Piperno et al., 1990), and Costa Rica (Islebe et al., 1996a). The early to middle Holocene moist period has also been documented in records from sub-Saharan African lakes that lie north of the equator (Street and Grove, 1976, 1979; Street-Perrott and Harrison, 1985; Lezine, 1989). Intersite differences in the timing of this wet interval may reflect geographical differences in its onset and termination, but may simply be a consequence of poor dating resolution in some lake cores. In contrast to the general finding of moist early to middle Holocene conditions at lowland sites in the circum-Caribbean, two localities in the Northern Hemisphere Neotropics provide evidence for dry conditions during part or all of the early and middle Holocene. Water level in Lake La Yeguada, Panama, was low from ~8,200 to ~5,800 years BP, suggesting relatively drier conditions during this period (Bush et al., 1992). Poor dating control hampers climatic interpretation of a core from Church’s Blue Hole on Andros Island, Bahamas. Unlike the rest of the Caribbean region, however, climate conditions appear to have been dry prior to about 4,630 years BP (Kjellmark, 1996). StreetPerrott et al. (1993) and Holmes et al. (1995) reported three transgressive-regressive cycles in Wallywash Great Pond, Jamaica during the Holocene, but dating of these cycles is unreliable because of hard-water-lake error, so that the timing of climate shifts remains in question.

LATE HOLOCENE (~3,000 C YR BP TO THE PRESENT) 14

Isotope records from Lakes Chichancanab, Miragoâne, and Coba display trends toward more positive δ18O values over the last ~3,000 years, indicating gradual climate drying (Figure 2). Oxygen isotopic values from the Lake Peten-Itza core show little variation during the last ~4,800 14C yr (Figure 2), and late Holocene climate inferences based on pollen from Lake Peten-Itza and other lakes on the Yucatan Peninsula are confounded because dense Maya populations in the lowlands cleared regional vegetation for agriculture (Islebe et al., 1996b; Curtis et al., 1998). Oxygen isotopic records from Lake Valencia do not reveal a clear trend in the late Holocene (Figure 2). Other proxies from Lake Valencia, however, indicate that E/P increased in the late Holocene. Sodium content in the sediments increased and the saline-tolerant ostracod Heterocypris communis reappeared in the record at ~2,140 14C yr BP. The presence of the littoral snail Pyrgophorus sp. at ~1,960 14C yr BP may also be indicative of general drying in that it suggests lake

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

45

level had dropped sufficiently to bring the shallow-water zone nearer to the coring site. On the basis of diatom and geochemical analyses, Bradbury et al. (1981) and Lewis and Weibezahn (1981) reported that salinity in Lake Valencia increased after ~3,000 14C yr BP. Animal microfossils in Lake Valencia sediment cores suggest a more complex hydrologic history involving multiple, lateHolocene rises and falls in water level (Binford, 1982). Millennial-scale climate changes are observed in all the circum-Caribbean oxygen isotopic records (Figure 2). Multi-decadal climate variations are observed throughout the Punta Laguna δ18O record, although no periodicity is apparent in the oscillations. There is, however, evidence for a series of droughts centered at 1510, 1171, 1019, 943, and 559 14C yr BP. The earliest two dry spells occurred at 585 and 862 ± 50 calendar years A.D. They coincide with two major cultural events, the Maya Hiatus that marks the boundary between the Early and Late Classic periods (~600 A.D.) and the collapse of the Classic Maya civilization in the 9th century A.D. Sediments from Lake Chichancanab also yielded oxygen isotopic and geochemical evidence for a major drought episode between ~1300 and ~1100 14C years BP (800 to 1000 A.D.) (Hodell et al., 1995). Temporal correlation between climatic events and cultural upheavals suggests a causal linkage. If indeed drought can be held responsible for the demographic collapse of Maya civilization ca. 850 A.D., it can be argued that climate change ultimately permitted the widespread regrowth of tropical forest in the region that had been largely cleared over the previous 2000 to 3000 years. Post-collapse forest recovery is documented in numerous pollen records from the region (e.g., Vaughan et al., 1985; Leyden, 1987; Islebe et al., 1996b). Other records from the lowland Neotropics also suggest drying in the late Holocene, but the timing of onset is not always coincident (Leyden, 1985; Markgraf, 1989; Burney et al., 1994; Islebe et al., 1996a; Kjellmark, 1996). Drying around Lake Miragoâne, Haiti began ~3,200 14C yr BP. Similarly, at Lake Chichancanab, Mexico, conditions became drier ~3,000 14C yr BP (Hodell et al., 1991, 1995). In contrast, drying began at about 4,500 years BP in montane Costa Rica (Islebe et al., 1996a). Dry conditions in the Bahamas from 3,200 to 1,500 years BP may have postponed human colonization of the islands (Kjellmark, 1996).

SUMMARY OF CIRCUM-CARIBBEAN CLIMATE Oxygen isotope data from circum-Caribbean lake cores yield a generally coherent pattern of regional paleoclimate. Despite potentially moderating maritime influences near the insular Miragoâne site, its paleoclimate record is remarkably similar to those from continental sites. Long-term similarities among all the records probably result from a common forcing mechanism. Conditions were dry in the late Pleistocene, but became wetter during the earliest Holocene (~10,500 to ~8,500 14C yr BP). Maximum moisture availability occurred during the early to middle Holocene (~8,500 to ~3,000 14C yr BP) and drier conditions returned in the late Holocene, from about 3,000 14C yr BP to the present (Figure 2). There are, however, notable differences among the proxy climate records. For example, the Peten-Itza and Valencia data lack evidence for late Holocene drying that is so apparent in the other lake cores (Figure 2). This could reflect climatic stability around Peten-Itza and Valencia or may simply be a consequence of the fact that they are large, deep lakes that are “insensitive” to all but the most pronounced climate changes. With their tremendously large volumes, even lake stage declines of several meters remove only a small fraction of the total water volume and therefore have little effect on the lake water δ18O signature. The isotopic signature of water in Lakes Miragoâne, Coba, and Chichancanab is probably more responsive to climatic shifts due to the relatively small basin volumes (Figure 2, Table 1). The onset of the early to middle Holocene moist period also differs among sites. Near Lakes Chichancanab and Miragoâne, the climate became wetter around 7,000 14C yr BP, whereas greater moisture became available ~1000 years later according to the Valencia record and ~2000 years later at Lake Peten-Itza (Figure 2). Evidence for the early to middle Holocene moist period is absent

46

Biogeography of the West Indies: Patterns and Perspectives

from the Coba record. Slight temporal differences among climate reconstructions may be a consequence of inaccuracies in sediment dating. Most of the chronologies, however, are robust and differences among the records may be due to local climatic and nonclimatic factors.

LONG-TERM CLIMATE CONTROLS The generally consistent pattern of millennial-scale climate shifts in the circum-Caribbean is explained by long-term changes in the seasonal distribution of solar energy. Studies of 20th century meteorology in the Atlantic region revealed correlations between years with rainfall anomalies and strength of the annual cycle (Hastenrath, 1984). Strongly seasonal years, during which the InterTropical Convergence Zone (ITCZ) moved farther north in summer (wet season) and farther south in winter (dry season), were anomalously wet. Conversely, reduced seasonality was associated with lower rainfall (Hastenrath, 1976, 1984). Hodell et al. (1991) suggested that long-term, millennialscale E/P variations in the circum-Caribbean were controlled by intensity of the annual cycle and movement of the ITCZ. These changes were forced by orbital mechanics, namely, the precessional cycle, with a characteristic period of 19,000 and 23,000 years. The intensity of the annual cycle can be expressed as the difference in seasonal insolation at the top of the atmosphere at 10°N, between summer (August) and winter (February) (Figure 3). August and February were chosen because they are the months during which there is maximum northward and southward displacement of the ITCZ and associated weather patterns (Hastenrath, 1976). Intensity of the annual cycle in the circum-Caribbean was greatest during the early Holocene (Figure 3) because perihelion occurred during Northern Hemisphere summer and aphelion occurred during Northern Hemisphere winter. In other words, the Earth’s Northern Hemisphere was tilted on its axis toward the sun as the planet passed closest to the sun during its annual solar orbit. When Earth reached maximum distance from the sun while orbiting, the Northern Hemisphere tilted on its axis away from the sun. The consequence of this planetary orientation was that the Northern Hemisphere experienced enhanced summer insolation and reduced winter insolation relative to the present. High early Holocene seasonality in the Northern Hemisphere was associated with increased northward movement of the ITCZ during summer and southward migration during winter. These conditions generated relatively high rainfall (low E/P) in the Northern Hemisphere tropics during the early Holocene. In the late Holocene, perihelion occurred during Northern Hemisphere winter and aphelion occurred during Northern Hemisphere summer, resulting in warmer winters and cooler summers. The change in orbital geometry caused a reduction in the intensity of the annual cycle (Figure 3), which led to a restricted latitudinal range of the ITCZ and higher E/P (drying) in the circum-Caribbean region. As discussed, lack of evidence for late Holocene drying in the records from Lakes Peten-Itza and Valencia may have simply been a consequence of their large volumes and relatively poor sensitivity to climate changes. Nevertheless, the data may indeed reflect relative climatic stability at those latitudes, which is consistent with the proposed mechanism of orbital forcing. Of the six study basins, Lakes Peten-Itza and Valencia lie at the lowest latitudes and closest to the ITCZ. Late Holocene reduction in the intensity of the annual cycle would have decreased the annual north/south movement of the ITCZ and associated weather systems. More northerly sites Chichancanab, Miragoâne, Coba, and Punta Laguna would have experienced substantial rainfall reduction in the late Holocene, whereas southerly sites Valencia and Peten-Itza would have continued to receive relatively abundant rainfall because of their proximity to the ITCZ. In addition to temporal shifts in summer (August) and winter (February) insolation, spring (May) and autumn (November) insolation also changed during the late Pleistocene and Holocene (Figure 3). The timing of changes in spring and autumn insolation did not coincide with changes in summer and winter insolation (Figure 3). Maximum difference between winter and summer insolation occurred ~8,000 14C yr BP while maximum difference between spring and autumn insolation occurred ~12,500 14C yr BP. The intensity of the annual cycle was mainly driven by

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

47

FIGURE 3 Changes in insolation at the top of the atmosphere at 10°N for the months of February (= ), May (= ), August (= ●), and November (= +) from 20,000 calendar years to the present (after calculations of Berger, 1978). The distance between the February and August curves is a representation of the intensity of the annual cycle.

differences in summer vs. winter insolation, but changes in spring insolation may have affected regional vegetation (Leyden et al., 1994). The decrease in springtime insolation during the early Holocene would probably have reduced spring moisture stress (Leyden et al., 1994). Reduced spring insolation may account for why δ18O and pollen data yield contradictory climatic inferences for the early Holocene portion of the Lake Peten-Itza core. The pollen record indicates that high forest was established by ~9,000 14C yr BP, suggesting relatively moist climate conditions (Islebe et al., 1996b; Curtis et al., 1998). In contrast, the δ18O record suggests fairly high E/P conditions. This discrepancy has been attributed to high fractional loss of the water budget of the lake to evaporation during the early stages of lake filling, i.e., when it was at low volume,

48

Biogeography of the West Indies: Patterns and Perspectives

despite the fact that conditions were wet enough to support high forest (Curtis et al., 1998). Alternatively, early Holocene conditions may have been relatively dry, as inferred from the δ18O record, but mesic vegetation was able to thrive because of reduced springtime water stress.

SHORT-TERM CLIMATE CONTROLS Although millennial-scale climate conditions at the study sites were controlled largely by orbital forcing, abrupt, short-term (decade-to-century) climate shifts are attributable to mechanisms other than orbital forcing. Short-term climate events, such as the droughts detected in the Punta Laguna and Chichancanab records, occurred too rapidly to have been caused by orbital forcing (Hodell et al., 1995; Curtis et al., 1996). Similar, brief events have been observed in other lake cores, such as the stage decline detected in the Valencia record at ~3,310 14C yr BP (Curtis et al., 1999) and the short episode of low δ18O values at ~9,100 and ~8,100 14C yr BP in the Lake Miragoâne record (Curtis and Hodell, 1993). Brief climate events may be recorded in sediment profiles from some lakes, but not others, because each water body has unique characteristics (e.g., volume, watershed vegetation) that make it respond in a distinctive manner to short-term forcing. Several mechanisms have been proposed to explain short-term climate fluctuations. Some mechanisms affect climate globally, whereas others have only local effects. Brief climate events can be forced by random, natural variability in the ocean–atmosphere system that does not require external forcing. These shifts may represent the extremes of natural climate variability. Short-duration climate shifts may also be a consequence of changes in solar energy reaching Earth. For example, Lean et al. (1995) used sunspot records to show that solar variability has influenced global climate since 1610 A.D. Volcanic eruptions can also force climate by injecting large amounts of gas, especially sulfur dioxide (SO2 ), high into the atmosphere (Jonas et al., 1995). Sulfur dioxide is oxidized to sulfate aerosols, mainly H2SO4H2O, in the troposphere and stratosphere. These aerosols reflect solar radiation and cause climate cooling (Jonas et al., 1995). Climate recovery after volcanic eruptions is rapid, generally taking only 2 to 3 years, and is achieved by removal of aerosols from the atmosphere in both wet and dry deposition (Jonas et al., 1995). The circum-Caribbean region is tectonically active and regional volcanic eruptions have probably influenced climate in the past (Gill, 2000). Volcanic ash does not always preserve well in tropical lake sediments, sometimes making it difficult to document prehistoric eruptions (Ford and Rose, 1995). Brief changes in ocean circulation have also been proposed as controls of decade-to-century-scale climate shifts. Small increases or decreases in North Atlantic Deep Water production can play an important role in climate of the North Atlantic region (Keigwin et al., 1991; Lehman and Keigwin, 1992; Broecker, 1994) and Africa (Street-Perrott and Perrott, 1990). Lastly, deforestation may drive regional climate change (Lean and Warrilow, 1989). Global climate models show that reduced vegetation cover may cause a weakened hydrological cycle and hence less precipitation (Lean and Warrilow, 1989). For example, the ancient Maya began clearing vegetation on the Yucatan Peninsula more than 3,000 years ago (Leyden, 1987; Islebe et al., 1996b). By the Late Classic Period (A.D. 550 to 850), the entire May Lowlands region was effectively deforested (Deevey et al., 1979). Anthropogenic deforestation may have caused the drought that led to the Maya collapse (Hodell et al., 1995; Curtis et al., 1996).

NONCLIMATIC CONTROLS Local, nonclimatic controls on basin hydrology may be responsible for differences observed among some of the lake records. These controls may include interlake differences in filling rate, morphometry, groundwater inputs and outputs, and orography. Local, nonclimatic influences on the hydrologic balance of lakes can regulate the δ18O of the lake water and can potentially lead to erroneous paleoclimate inferences. Filling rate may have exerted some control on the isotopic composition of lake water in the study basins. For instance, in the early stages of basin filling, the δ18O of lake water is high because

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

49

the water body has a small volume and loses a large proportion of its annual water budget to evaporation. Large-volume lakes such as Peten-Itza took longer to fill than small-volume lakes such as Chichancanab. Consequently, small lakes achieve oxygen isotopic equilibrium more quickly than large lakes (Lister et al., 1991). Chichancanab reached isotopic steady state in a few hundred years, whereas Peten-Itza required thousand of years to achieve isotopic equilibrium (Figure 2). The Lake Peten-Itza oxygen isotope record may also have been influenced by basin morphology. The core was collected in the smaller, shallow south basin of the lake, which was isolated from the large, deep (160 m) north basin during initial lake filling. While filling, the shallow south basin would have had a high surface area–volume ratio and δ18O of the water would have been relatively positive. Once the north basin filled and the two basins were connected, low δ18O water of the north basin would have mixed with high δ18O water of the south basin, reducing the oxygen isotopic signal in waters near the core site. The early and middle Holocene oxygen isotope records from Lakes Valencia and Miragoâne may lack some detail concerning changing E/P conditions. Overflow in Valencia was inferred from other proxy records (Bradbury et al., 1981; Lewis and Weibezahn 1981; Binford, 1982), indicating relatively wet conditions. But overland outflows during the moist period made lake water isotopic signatures less sensitive to changes in the relation between rainfall and evaporation. Thus, δ18O measures for this period are somewhat less useful for paleoclimatic inferences. Lake Valencia lies in a valley surrounded by mountains. Orography may have influenced the inferred moisture history of the basin. When clouds rise and pass over the mountains, rainout occurs and affects both the quantity and isotopic composition of precipitation. Today, rainfall in northern Venezuela is influenced by the equatorial trough during the wet season and the Northern Hemisphere trade winds during the dry season (Snow, 1976). If the isotopic composition of the source waters differs, it is possible that the isotopic composition of Lake Valencia lake water changed through time due to shifting relative contributions of precipitation from the two regions, i.e., from the Caribbean across the Cordillera Costanera to the north, or across the continent and the Serrania del Interior to the south.

CLIMATE AND BIOGEOGRAPHY IN THE CIRCUM-CARIBBEAN Modern distributions and abundances of plants and animals in the circum-Caribbean reflect climate changes that occurred within the last 20,000 years. Antillean faunal extinctions during the late Pleistocene–early Holocene transition (40,000 to 4,500 BP) (Morgan and Woods, 1986) were probably a consequence, in part, of the transition from cool–arid to warm–moist conditions (Pregill and Olson, 1981). Xeric habitats largely disappeared in the early Holocene, accounting for the disjunct or restricted modern distributions of obligate dry-adapted species, especially reptiles and birds. Fossils of xerophilic taxa have been found at geographically widespread sites, indicating they were distributed extensively in the arid late Glacial. Cool, arid conditions in the late Pleistocene are supported by various lines of evidence, including the paucity of lake basins that held water at that time. In the few late Glacial lacustrine records that do exist in the circum-Caribbean, pollen data indicate that xerophytic, cold-tolerant plant communities occupied the landscape. In northern Guatemala, floral elements that today dominate the tropical forest in the region were completely absent from the pollen record. Throughout the region, forested habitats must have been limited in areal extent and highly fragmented. Furthermore, in regions with substantial altitudinal relief, even cold-tolerant plant species were forced to lower elevations in the late Glacial, when temperatures were as much as 8°C below the present temperature. As vegetation responded to changes in temperature and available moisture, so too did the associated fauna. In the early and middle Holocene, mesic forest expanded, leading to the greater abundance of animals that occupy wetter habitats. Forest expansion also led to contact between animal and plant populations that had been geographically isolated during the previous, arid episode. For

50

Biogeography of the West Indies: Patterns and Perspectives

instance, based on pollen from the lake Miragoâne core, Higuera-Gundy et al. (1999) suggest that the ~150-km-wide strip of lowland dry forest that today separates the Massifs de La Hotte and de La Selle in southwest Haiti, was covered by mesic vegetation some time between ~5.4 and 2.5 kyr BP. Fossil vertebrate data support this hypothesis. Three species of Nesophontes that became extinct ~100 years ago show minor morphological differentiation in the massifs and in the intervening area. This suggests that formerly isolated populations of the massifs had recent contact in the Miragoâne region. Geographic overlap among the three insectivore species probably occurred in the middle Holocene, when mesic vegetation colonized the gap that previously and at present separates the massifs. Fossils from southwestern Hispaniola indicate that some extinct mammal species, including rodents, ground sloths, and a primate, persisted until ~3,000 BP (Woods, 1989a, 1989b). Of the endemic rodents known from the island, 93% are presently extinct, but most survived until ~3,000 BP (Woods, 1989a, 1989b). Over the last few millennia, four bat species from the Massif de La Selle highlands have become extinct (Morgan and Woods, 1986). Late Holocene extinctions of some highland mesophilic animal species probably predate significant human impacts in those environments. The coincidence of these faunal extinctions, with an isotopically documented drying trend and palynological evidence for loss of mesic forests after 3,000 BP, suggests that climatic drying contributed to the demise of these taxa.

SUMMARY AND CONCLUSIONS Holocene climate variability in the circum-Caribbean was inferred using δ18O records from shell carbonate in sediment cores from six regional lakes. Temporal resolution, based on radiocarbon dating of the sediment sequences, permitted climate reconstruction at millennial to centennial timescales. Following the cool, arid late Glacial, climate ameliorated and previously dry lake basins began to fill with water. Throughout the circum-Caribbean, lowland sites generally progressed from arid in the late Glacial and earliest Holocene to moister in the early to middle Holocene. Around 3,000 BP, drier conditions returned to the region. Superimposed on millennial-scale climate variability are recent, shorter-term events that are interesting to both biogeographers and anthropologists. Of particular interest are records from the Yucatan Peninsula that point to a series of droughts that occurred since the onset of drying ~3,000 BP. Periods of extreme dry conditions center on dates of 585, 862, 986, and 1051 calendar years A.D. The drought episodes at 585 and 862 A.D. coincide with major changes in Maya cultural evolution and even suggest that drought was responsible for the 9th century A.D. Maya collapse. Long-term millennial patterns of change discerned in the proxy climate records are explained by shifts in the intensity of the annual cycle. These in turn were driven by changes in insolation that are ultimately controlled by orbital precession. Short-term climate events detected in several records were too abrupt to have been caused by orbital forcing. Mechanisms driving these short-term climate changes may have included volcanic eruptions, changes in solar variability, or ocean circulation and deforestation. Climate proxy records from several sites were probably influenced to some degree by local, nonclimatic effects such as lake-filling rate, basin morphology, and orography. This study revealed patterns of climate change in the circum-Caribbean region from the late Pleistocene to the present and demonstrated climatic variability on long and short timescales. Millennial-scale variations in E/P were probably forced by orbital mechanics that were controlled by the intensity of the annual cycle and seasonal displacement of the ITCZ. Forcing mechanisms for shortterm, centennial climate variability remain to be identified. Paleoclimatic inferences presented here suggest that analyses of West Indian biogeography must be cognizant of the potential influence that latest Pleistocene and Holocene climate change has had on the distribution and abundance of the circum-Caribbean biota.

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

51

LITERATURE CITED Berger, A. 1978. Numerical values of caloric insolation from 1,000,000 YBP to 100,000 YAP (astronomical solution of Berger, 1978). Contribution 37. Institut d’Astronomie et de Géophysique G. Lemaitre, Université Catholique de Louvain-la-Neuve, Belgique. Binford, M. W. 1982. Ecological history of Lake Valencia, Venezuela: interpretation of animal microfossils and some chemical, physical and geological features. Ecological Monographs 52:307–333. Bradbury, J. P. 1979. Quaternary diatom stratigraphy of Lake Valencia, Venezuela. Pp. 276–288 in Proceedings of the Fourth Latin American Geological Congress. Fourth Latin American Geological Congress, Port of Spain, Trinidad and Tobago. Bradbury, J. P., B. W. Leyden, M. L Salgado-Labouriau, W. M., Lewis, Jr., C. Schubert, M. W. Binford, D. G. Frey, D. R. Whitehead, and F. H. Weibezahn. 1981. Late Quaternary environmental history of Lake Valencia, Venezuela. Science 214:1299–1305. Brenner, M. 1994. Lakes Salpeten and Quexil, Peten, Guatemala, Central America. Pp. 377–380 in GierlowskiKordesch, E. and K. Kelts (eds.). Global Geological Record of Lake Basins, Vol. 1. Cambridge University Press, Cambridge. Broecker, W. S. 1994. Massive iceberg discharges as triggers for global climate change. Nature 372:421–424. Broecker, W. S. 1995. Cooling the tropics. Nature 376:212–213. Burney, D. A., L. P. Burney, and R. D. E. MacPhee. 1994. Holocene charcoal stratigraphy from Laguna Tortuguero, Puerto Rico and the timing of human arrival on the island. Journal of Archaeological Science 21:273–281. Bush, M. B. and P. A. Colinvaux. 1990. A pollen record of a complete glacial cycle from lowland Panama. Journal of Vegetation Science 1:105–118. Bush, M. B., D. R. Piperno, P. A. Colinvaux, P. E. De Oliveira, L. A. Krissek, M. C. Miller, and W. E. Rowe. 1992. A 14300-yr paleoecological profile of a lowland tropical lake in Panama. Ecological Monographs 62:251–275. Candela, G. A. and G. C. Candelas. 1963. The West Indies. Pp. 435–450 in Frey, D. G. (ed.). Limnology in North America. The University of Wisconsin Press, Madison. Chivas, A. R., P. De Dekker, J. A. Cali, A. Chapman, E. Kiss, and J. M. G. Shelley. 1993. Coupled stableisotope and trace-element measurements of lacustrine carbonates as paleoclimatic indicators. Pp. 113–121 in Swart, P. K., K. C. Lohmann, J. McKenzie, and S. Savin (eds.). Climate Change in Continental Isotopic Records. American Geophysical Union, Washington, D.C. Colinvaux, P. A., P. De Oliveira, J. E. Moreno, M. C. Miller, and M. B. Bush. 1996a. A long pollen record from lowland Amazonia: forest and cooling in glacial times. Science 274:85–88. Colinvaux, P. A., K.-B. Liu, P. De Olivera, M. B. Bush, M. C. Miller, and M. Steinitz Kannan. 1996b. Temperature depression in the lowland tropics in glacial times. Climate Change 32:19–33. Covich, A. and M. Stuiver. 1974. Changes in oxygen 18 as a measure of long-term fluctuations in tropical lake levels and molluscan populations. Limnology and Oceanography 19:682–691. Curtis, J. H. 1992. Natural Variability of Late Pleistocene-Holocene Climates in the Caribbean from Isotopic and Trace Element Analysis of Lake Sediments (10,500 Years BP to Present). M.S. thesis, University of Florida, Gainesville. Curtis, J. H. and D. A. Hodell. 1993. An isotopic and trace element study of ostracods from Lake Miragoâne, Haiti: a 10,500 year record of paleosalinity and paleotemperature change in the Caribbean. Pp. 135–152 in Swart, P. K., K. C. Lohmann, J. McKenzie, and S. Savin (eds.). Climate Change in Continental Isotopic Records. American Geophysical Union, Washington, D.C. Curtis, J. H., D. A. Hodell, and M. Brenner 1996. Climate variability on the Yucatan Peninsula (Mexico) during the last 3500 years and implications for Maya cultural evolution. Quaternary Research 46:37–47. Curtis, J. H., M. Brenner, D. A. Hodell, R. A. Balser, G. A. Islebe, and H. Hooghiemstra. 1998. A multi-proxy study of Holocene environmental change in the Maya Lowlands of Peten, Guatemala. Journal of Paleolimnology 19:139–159. Curtis, J. H., M. Brenner, and D. A. Hodell. 1999. Changes in moisture availability in the Lake Valencia Basin, Venezuela (late Pleistocene to present). The Holocene 9:609–619. Deevey, E. S. and M. Stuiver. 1964. Distribution of natural isotopes of carbon in Linsley Pond and other New England lakes. Limnology and Oceanography 9:1–11.

52

Biogeography of the West Indies: Patterns and Perspectives

Deevey, E. S., D. S. Rice, P. M. Rice, H. H. Vaughan, M. Brenner, and M. S. Flannery. 1979. Mayan urbanism: impact on a tropical karst environment. Science 206:298–306. Deevey, E. S., M. Brenner, and M. W. Binford. 1983. Paleolimnology of the Peten Lake District, Guatemala, III. Late Pleistocene and Gamblian environments of the Maya area. Hydrobiologia 103:211–216. Fairbanks, R. G. 1989. A 17,000 year glacio-eustatic sea level record: influences of glacial melting rates in the Younger Dryas event and deep-ocean circulation. Nature 342:637–642. Fontes, J. C. and R. Gonfiantini. 1967. Comportement isotopique au cours de l’évaporation de deux bassins sahariens. Earth and Planetary Science Letters 3:258–266. Ford, A. and W. I. Rose. 1995. Volcanic ash in ancient Maya ceramics of the limestone lowlands: implications for prehistoric volcanic activity in the Guatemala highlands. Journal of Volcanology and Geothermal Research 66:149–162. Gasse, F., R. Téhet, A. Durand, E. Gilbert, and J. C. Fontes. 1990. The arid-humid transition in the Sahara and Sahel during the last glaciation. Nature 346:141–146. Gill, R. B. 2000. The Great Maya Droughts: Water, Life and Death. University of New Mexico Press, Albuquerque. Grimm, E. C., G. L. Jacobson, Jr., W. A. Watts, B. C. S. Hansen, and K. A. Maasch. 1993. A 50,000-year record of climate oscillations from Florida and its temporal correlation with the Heinrich Events. Science 261:198–200. Guilderson, T. P., R. G. Fairbanks, and J. L. Rubenstone. 1994. Tropical temperature variations since 20,000 years ago: modulating interhemispheric climate change. Science 263:663–665. Hastenrath, S. 1976. Variations in low-latitude circulation and extreme climatic events in the tropical Americas. Journal of the Atmospheric Sciences 33:202–215. Hastenrath, S. 1984. Interannual variability and annual cycle: mechanisms of circulation and climate in the tropical Atlantic sector. Monthly Weather Review 112:1097–1107. Higuera-Gundy, A. 1991. Antillean Vegetational History and Paleoclimate Reconstructed from the Paleolimnological Record of Lake Miragoâne, Haiti. Ph.D. dissertation, University of Florida, Gainesville. Higuera-Gundy, A., M. Brenner, D. A. Hodell, J. H. Curtis, B. W. Leyden, and M. W. Binford. 1999. A 10,300 14C-year record of climate and vegetation change in Haiti. Quaternary Research 52:159–170. Hodell, D. A., J. C. Curtis, and M. Brenner. 1995. Possible role of climate in the collapse of Classic Maya civilization. Nature 375:391–394. Hodell, D. A., J. H. Curtis, G. A. Jones, A. Higuera-Gundy, M. Brenner, M. W. Binford, and K. T. Dorsey. 1991. Reconstruction of Caribbean climate change over the past 10,500 years. Nature 352:790–793. Holmes, J. A. 1996. Trace-element and stable isotope geochemistry of non-marine ostracod shells in Quaternary palaeoenvironmental reconstruction. Journal of Paleolimnology 15:223–235. Holmes, J. A. 1998. A late Quaternary ostracod record from Wallywash Great Pond, a Jamaican marl lake. Journal of Paleolimnology 19:115–128. Holmes, J. A., F. A. Street-Perrott, M. Ivanovich, and R. A. Perrott. 1995. A late Quaternary palaeolimnological record from Jamaica based on trace-element chemistry of ostracod shells. Chemical Geology 124:143–160. Islebe, G. A., H. Hooghiemstra, and R. Van’t Veer. 1996a. Holocene vegetation and water table history in two bogs of the Cordillera de Talamanca, Costa Rica. Vegetatio 124:155–171. Islebe, G. A., H. Hooghiemstra, M. Brenner, J. H. Curtis, and D. A. Hodell. 1996b. A Holocene vegetation history from lowland Guatemala. The Holocene 6:265–271. Jonas, P. R., R. J. Charlson, and H. Rodhe. 1995. Aerosols. Pp. 127–162 in Houghton, J. T., L. G. Meira Filho, J. Bruce, H. Lee, B. A. Callander, E. Haites, N. Harris, and K. Maskell (eds.). Climate Change 1994: Radiative Forcing of Climate Change and an Evaluation of the IPCC IS92 Emission Scenarios. Cambridge University Press, Cambridge. Keigwin, L. D., G. A. Jones, S. J. Lehman, and E. A. Boyle. 1991. Deglacial meltwater discharge, North Atlantic deep circulation and abrupt climate change. Journal of Geophysical Research 96:16811–16826. Kjellmark, E. 1996. Late Holocene climate change and human disturbance on Andros Island, Bahamas. Journal of Paleolimnology 13:133–145. Lean, J., J. Beer, and R. Bradley. 1995. Reconstruction of solar irradiance since 1610: implications for climate change. Geophysical Research Letters 22:3195–3198. Lean, J. and D. A. Warrilow. 1989. Simulation of the regional climatic impact of Amazon deforestation. Nature 342:411–413.

Climate Change in the Circum-Caribbean and Implications for Regional Biogeography

53

Lehman, S. J. and L. D. Keigwin. 1992. Sudden changes in North Atlantic circulation during the last deglaciation. Nature 356:757–762. Lewis, W. M., Jr. and F. H. Weibezahn. 1981. Chemistry of a 7.5-m sediment core from Lake Valencia, Venezuela. Limnology and Oceanography 26:907–924. Leyden, B. W. 1984. Guatemalan forest synthesis after Pleistocene aridity. Proceedings of the National Academy of Sciences U.S.A. 81:4856–4859. Leyden, B. W. 1985. Late Quaternary aridity and Holocene moisture fluctuations in the Lake Valencia Basin, Venezuela. Ecology 66:1279–1295. Leyden, B. W. 1987. Man and climate in the Maya lowlands. Quaternary Research 28:407–414. Leyden, B.W., M. Brenner, D. A. Hodell, and J. H. Curtis. 1993. Late Pleistocene climate in the Central American lowlands. Pp. 165–178 in Swart, P. K., K. C. Lohmann, J. McKenzie, and S. Savin (eds.). Climate Change in Continental Isotopic Records. American Geophysical Union, Washington, D.C. Leyden, B. W., M. Brenner, D. A. Hodell, and J. H. Curtis. 1994. Orbital and internal forcing of climate on the Yucatan Peninsula for the past ca. 36 ka. Palaeogeography, Palaeoclimatology, Palaeoecology 109:193–210. Leyden, B. W., M. Brenner, T. J. Whitmore, J. H. Curtis, D. R. Piperno, and B. H. Dahlin. 1996. A record of long- and short-term climate variation from northwest Yucatan: Cenote San Jose Chulchaca. Pp. 30–50 in Fedick, S. L. (ed.). The Managed Mosaic: Ancient Maya Agriculture and Resource Use. University of Utah Press, Salt Lake City. Leyden, B. W., M. Brenner, and B. H. Dahlin. 1998. Cultural and climatic history of Cobá, a lowland Maya City in Quintana Roo, Mexico. Quaternary Research 49:111–122. Lezine, A. M. 1989. Late Quaternary vegetation and climate of the Sahel. Quaternary Research 32:317–334. Lister, G. S., K. Kelts, C. K. Zao, J. Yu, and F. Niessen. 1991. Lake Qinghai, China: closed-basin lake levels and the oxygen isotope record for ostracoda since the latest Pleistocene. Palaeogeography, Palaeoclimatology, Palaeoecology 84:141–162. Markgraf, V. 1989. Paleoclimates in Central and South America since 18,000 BP based on pollen and lakelevel records. Quaternary Science Reviews 8:1–24. Markgraf, V. 1993. Climatic history of Central and South America since 18,000 yr B.P.: comparison of pollen records and model simulations. Pp. 357–385 in Wright, H. E., Jr., J. E. Kutzbach, T. Webb III, W. F. Ruddiman, F. A. Street-Perrott, and P. J. Bartlein (eds.). Global Climates since the Last Glacial Maximum. University of Minnesota Press, Minneapolis. Morgan, G. S. and C. A. Woods. 1986. Extinction and the zoogeography of West Indian land mammals. Biological Journal of the Linnean Society 28:167–203. Piperno, D. R., M. B. Bush, and P. A. Colinvaux. 1990. Paleoenvironments and human settlements in lateglacial Panama. Quaternary Research 33:108–116. 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. Salgado-Labouriau, M. L. 1980. A pollen diagram of the Pleistocene-Holocene boundary of Lake Valencia, Venezuela. Reviews of Palaeobotany and Palynology 30:297–312. Seltzer, G. O. 1990. Recent glacial history and paleoclimate of the Peruvian-Bolivian Andes. Quaternary Science Reviews 9:137–152. Snow, J. W. 1976. The climate of northern South America. Pp. 295–403 in Schwerdtfeger, W. (ed.). Climates of Central and South America. Elsevier, Amsterdam. Street, F. A. and A. T. Grove. 1976. Environmental and climatic implications of late Quaternary lake-level fluctuations in Africa. Nature 261:385–390. Street, F. A. and A. T. Grove. 1979. Global maps of lake-level fluctuations since 30,000 yrs BP. Quaternary Research 12:83–118. Street-Perrott, F. A. and S. P. Harrison. 1985. Lake levels and climate reconstruction. Pp. 291–340 in Paleoclimate Data and Modeling. Wiley, New York. Street-Perrott, F. A. and R. A. Perrott. 1990. Abrupt climate fluctuations in the tropics: the influence of Atlantic Ocean circulation. Nature 343:607–612. Street-Perrott, F. A., P. E. Hales, R. A. Perrott, J. C. Fontes, V. R. Switsur, and A. Pearson. 1993. Late Quaternary palaeolimnology of a tropical marl lake: Wallywash Great Pond, Jamaica. Journal of Paleolimnology 9:3–22.

54

Biogeography of the West Indies: Patterns and Perspectives

Stute, M., M. Forster, H. Frischkorn, A. Serejo, J. F. Clark, P. Schlosser, W. S. Broecker, and G. Bonani. 1995. Cooling of tropical Brazil (5°C) during the last glacial maximum. Science 269:379–383. Talbot, M. R. 1990. A review of the palaeohydrological interpretation of carbon and oxygen isotopic ratios in primary lacustrine carbonates. Chemical Geology (Isotope Geoscience Section) 80:261–279. Thompson, L. G., E. Mosley-Thompson, M. E. Davis, P.-N. Lin, K. A. Henderson, J. Cole-Dai, J. F. Bolzan, and K.-B. Liu. 1995. Late glacial stage and Holocene tropical ice core records from Huascarán, Peru. Science 269:46–50. Vaughan, H. H., E. S. Deevey, and S. E. Garrett-Jones. 1985. Pollen stratigraphy of two cores from the Peten Lake District. Pp. 73–89 in Pohl, M. (ed.). Prehistoric Lowland Maya Environment and Subsistence Economy. Peabody Museum Papers 77, Harvard University Press, Cambridge, Massachusetts. Watts, W. A., 1969. A pollen diagram from Mud Lake, Marion County, north-central Florida. Geological Society of America Bulletin 80:631–642. Watts, W. A. 1975. A late Quaternary record of vegetation from Lake Annie, south-central Florida. Geology 3:344–346. Watts, W. A. and B. C. S. Hansen. 1988. Environments of Florida in the Late Wisconsin and Holocene. Pp. 307–323 in Purdy, B. A. (ed.). Wet Site Archaeology. The Telford Press, Caldwell, New Jersey. Watts, W. A. and B. C. S. Hansen. 1994. Pre-Holocene and Holocene pollen records of vegetation history from the Florida peninsula and their climatic implications. Palaeogeography, Palaeoclimatology, Palaeoecology 109:163–176. Watts, W. A. and M. Stuiver. 1980. Late Wisconsin climate of northern Florida and origin of species-rich deciduous forest. Science 210:325–327. Watts, W. A., B. C. S. Hansen, and E. C. Grimm. 1992. Camel Lake: a 40,000-yr record of vegetational and forest history from northwest Florida. Ecology 73:1056–1066. Whitmore, T. J., M. Brenner, J. H. Curtis, B. H. Dahlin, and B. W. Leyden. 1996. Holocene climatic and human influences on lakes of the Yucatan Peninsula, Mexico: an interdisciplinary, paleolimnological approach. The Holocene 6:273–287. Williams, E. O. 1989. Old problems and new opportunities in West Indian biogeography. Pp. 1–46 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Woods, C. A. 1989a. A new capromyid rodent from Haiti: the origin, evolution and extinction of West Indian rodents and their bearing on the origin of New World hystricognaths. Natural History Museum of Los Angeles County, California, Contributions in Science Series 33:59–89. Woods, C. A. 1989b. The biogeography of West Indian rodents. Pp. 741–798 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

Adaptations to 4 Functional Island Life in the West Indies Brian K. McNab Abstract — The functional adjustments made by vertebrates to facilitate long-term persistence on islands in the West Indies are examined. Evidence of these adjustments is obscured by the extended period of human occupation of these islands, an occupation that has led to an extensive extinction of the indigenous fauna. The available evidence indicates that vertebrates living on these islands made adjustments similar to those made by vertebrates found on tropical islands in the Indian and Pacific Oceans. These adjustments include (1) a reduction in body mass; (2) the evolution of flightlessness among birds; (3) the evolution of low rates of metabolism; (4) the selective evolution of torpor; and (5) the partial replacement of endotherms by ectotherms. All of these changes reduce resource requirements, which facilitate the survivorship of species on islands where the resource base limits the size of populations and where weather instabilities decrease survivorship of resident populations. Unfortunately, relatively few surviving endemic vertebrates living on West Indies islands have been studied from this viewpoint.

INTRODUCTION Biologists have been profoundly interested in the faunas living on continental and oceanic islands, starting with Charles Darwin (1839) and Alfred Russel Wallace (1880) through P.J. Darlington (1957) to Robert H. MacArthur and Edward O. Wilson (1967). Of special interest is the extent to which island faunas result from the combined effects of the capacity for long-distance dispersal, the chance event of landfall, biased by the ability to attain a foothold on an island in the presence of a resident fauna, and modified by the retention or evolution of characteristics that facilitate long-term survival on islands. This chapter examines the characteristics of a fauna that facilitate persistence on islands. The characterization of island faunas today is difficult because nearly all extant faunas have been heavily modified by humans through the extensive extinction of autochthonous elements and the translocation of alien species, especially mammalian predators, onto islands (see Steadman, 1995). Most islands were free of predators. What humans have done, then, is to convert distinctive island environments and faunas into mini-continental environments and faunas, thereby radically changing the conditions in which all further evolution and survival are to operate. Consequently, the composition of island faunas today tells us little about the characteristics required for long-term survival on islands. Because we know so little about the basic biology of the surviving island endemics, we are unable to answer unequivocally the question whether island endemics are different in any consistent, substantial manner from their continental relatives. On the islands of the Caribbean the combination of an extensive extinction of endemic vertebrates and the neglect of the physiology of the living remnants has led to our ignorance of the functional bases of faunal persistence on islands in this region. The extensive extinction of the larger endemic vertebrates on Caribbean islands occurred principally because these islands have been occupied by people for 6000 to 7000 years. This is a much longer period of human presence than in outlying Polynesia (Hawaii, Easter Island, New Zealand), which were first occupied some 800 to 1000 years ago — the last places on Earth to be occupied by Homo sapiens. Only a few isolated islands in the Pacific (Galápagos Islands, Auckland Islands, Campbell Island) and Indian (Mauritius, Réunion) Oceans were first occupied by European mariners 400 to 600 years ago. In the absence of humans, the principal problems faced by island faunas are the restricted resource base of islands and the likelihood of eventually encountering severe environmental 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

55

56

Biogeography of the West Indies: Patterns and Perspectives

catastrophes. The Caribbean region is regularly subjected to hurricanes, occasionally severe enough to threaten the survival of its faunas (Askins, 1991; Waide, 1991a, 1991b; Gannon and Willig, 1994; Pedersen et al., 1996), just as encountering typhoons (= hurricanes) and El Niño and La Niña events in the Pacific. These catastrophes have the greatest impact on small, flat islands, where little shelter exists. I argue that the shortage of resources directs the principal adjustments required of vertebrates for life on islands, and it is in meeting these requirements that island faunas attain their distinctive characteristics, especially in the absence of predators, a condition that permits the most radical adjustments.

THE ADJUSTMENT OF VERTEBRATES TO ISLAND LIFE A series of adjustments by vertebrates that facilitate long-term survival on islands has been described in vertebrates (McNab, 1994a, 1994b). The common feature in these responses is a reduction in resource requirements. This reduction is required by the limited resource base. These responses, which are not possible on continents where avian, and especially mammalian, predation is ubiquitous, are permitted by the reduced predation on islands: oceanic islands have no mammalian predators and most have no avian predators. In the face of the reduced resource base on islands, birds and mammals have responded in a variety of ways. These responses include (1) a reduction in body mass; (2) the evolution of “approachability”; (3) the evolution of flightlessness; (4) a reduction in rate of metabolism independent of a change in mass or the evolution of flightlessness; (5) the evolution of torpidity; and (6) the replacement of endotherms by ectotherms. Each of these responses to island life reduces resource requirements by means of a reduction in rate of metabolism. A reduction in body mass — The change in body mass associated with island life is complex (see Lomolino, 1985). A reduction in mass reduces the amount of energy and matter required for maintenance and activity because these requirements increase with mass. Flying foxes of the genera Pteropus and Dobsonia and flightless rails have masses that are correlated with island size (McNab, 1994b). However, under some circumstances body mass in island endemics increases, but only if the resource abundance is unusually large, either because of the presence of an abundant resource, such as breeding salmon on Kodiak Island, Alaska, or because of the absence of competitors that normally exploit a particular resource, such as the absence of grazing and browsing mammals. Thus, an increase in mass occurs in the Kodiak bear (Ursus arctos middendorffi) and in various grazing and browsing birds, including gallinules (e.g., takahe [Porphyrio mantelli] on New Zealand, P. alba on Lord Howe Island), moas (Dinornithidae) on New Zealand, large, flightless “geese” on various Hawaiian islands, and elephant birds (Aepyornithidae) on Madagascar. The evolution of “approachability” — Many island birds and mammals can be approached closely without a fright reaction. This behavior is noticeable in the Galápagos Islands both in birds and mammals, but also is present on other islands, such as the Falkland Islands (Humphrey et al., 1987). Approachability probably was characteristic of island faunas, although presumably greatly reduced as a result of human predation on naive faunas (e.g., Réunion and Socorro; Humphrey et al., 1987). The absence of “flightiness” reduces the energy expenditure of island species, a behavior that is possible only in the absence of predators. In this view, the presence of approachability and the absence of flightiness is qualitatively similar to a truly flightless condition. The evolution of flightlessness — The evolution of flightlessness in island endemic birds is most marked in the family Rallidae. Steadman (1995) suggested that every island in the South Pacific for which an extensive fossil fauna exists had one to four species of flightless rails, which led to his estimate that the evolution of flightlessness in South Pacific rails probably occurred at least 2000 times. In rails, the evolution of flightlessness is correlated with a reduction in pectoral muscle mass, which in turn is associated with a reduction in basal rate of metabolism (McNab, 1994a).

Functional Adaptations to Island Life in the West Indies

57

A similar pattern has been seen in kiwis (Apteryx) and suggested in the flightless parrot (Strigops habroptilus) in New Zealand and the flightless cormorant (Phalacrocorax harrisii) in the Galápagos (McNab, 1994a, 1996; McNab and Salisbury, 1994). Ducks also have repeatedly evolved a flightless condition on islands, including Laysan, Hawaii, South Georgia, and in the New Zealand islands (North, South, Auckland, Campbell). What little we know indicates that their flightless condition is not correlated with a marked reduction in pectoral mass or basal rate, a condition that also exists in penguins (McNab, 1994a). The retention of intermediate to large pectoral muscles in flightless ducks and penguins is associated with the use of wings for flight under water (penguins), in escape from avian predators (Auckland Island teal), or in courtship (steamer ducks). A flightless condition in island ducks may, however, be associated with a reduction in field energy expenditures because of a reduced level of activity compared to flighted species. Sailer (1999) showed that species of the fruit-dove genus Ptilinopus that live on islands in the South Pacific without goshawks (Accipiter) have smaller pectoral girdles and muscle masses than species found on islands with goshawks, which in turn have smaller pectoral girdles and muscle masses than species found on continents, where predatory birds and mammals are present. Whether basal rate of metabolism in Ptilinopus is correlated with this differentiation in pectoral girdles and muscle masses is unknown because the only measurements of energy expenditure in Ptilinopus are on continental or large-island species, which have basal rates typical of continental Ducula (McNab, 2000). A reduction in rate of metabolism — Fruit-pigeons (Ducula) and flying foxes (Pteropus and Dobsonia) that are endemic to small islands have lower basal rates of metabolism than related species that are endemic to large islands and to continents (McNab, 1994b, 2000; McNab and Bonaccorso, 2001). This reduction in basal rate is independent of the ability to fly, and in flying foxes is most marked in females (McNab and Armstrong, in press). The reduction in the basal rate of Pteropus is most marked in small-sized island endemics (e.g., P. rodricensis) and in females that belong to larger island endemics (e.g., P. hypomelanus). The smallest small-island Pteropus studied, pumilus, has a basal rate that is somewhat less depressed, which raises the question whether small endotherms that are endemic to small islands might not have to adjust their basal rates. This question is unexplored; it might apply to Ptilinopus, which is much smaller than Ducula, and would be profitably studied in some small, small-island endemic passerines. Whether the reduction in basal rate in small-island endemic endotherms is related to a change in body composition, as appears to be the case in kiwis and flightless rails, is unknown. The evolution of torpor — Bonaccorso and McNab (1997) observed that several small, nectarivorous flying foxes belonging to the genera Macroglossus and Syconycteris readily entered torpor, a behavior not known to occur in nectarivorous bats of similar or smaller mass belonging to the New World family Phyllostomidae. In contrast to insectivorous bats, which tend readily to enter torpor in cool climates and sometimes have effective endothermy in tropical environments (McNab, 1969; Bonaccorso and McNab, submitted), these nectarivores are prone to use torpor in the tropical lowlands, and show more effective endothermy in the tropical highlands and warm-temperate regions (Geiser et al., 1996). These genera are found in Southeast Asia, northern Australia, and on large and small islands from the Moluccas to the Solomons. Two related genera, Melonycteris and Notopterus, are endemics on South Pacific islands from the Bismarck Archipelago to the Solomon Islands, New Caledonia, Vanuatu, and Fiji. One of these bats, M. melanops from New Britain, has been shown to be less prone to enter torpor (Bonaccorso and McNab, 1997), possibly in association with a larger body size; Notopterus is even larger. Whether torpor in these bats facilitates their persistence on small islands is unknown. The replacement of endotherms by ectotherms — On many islands, including the Galápagos, Aldabra, Fiji, and Komodo, the largest herbivorous and carnivorous niches are occupied by reptiles (Flannery, 1993; McNab, 1994b). Some of this “replacement” of endotherms may occur because

58

Biogeography of the West Indies: Patterns and Perspectives

the appropriate endotherms never arrived on these islands, or because these islands cannot sustain indefinitely the high level of resource harvesting required by endotherms. The high requirement of endotherms is produced by their high individual expenditures and the high population levels required for long-term survivorship. The success of reptiles on islands undoubtedly reflects their low individual resource requirements, compared to endotherms of the same mass, and their ability to tolerate prolonged periods of starvation because of an appreciable increase in body size.

WAS THE FAUNA OF THE WEST INDIES RESOURCE LIMITED? The adjustments that facilitate the long-term survival of vertebrates on islands have been recognized through an examination of the fauna of South Pacific islands. This section explores the extent to which these (or other) adjustments have been made to life on the islands of the West Indies. As noted, the present fauna of the West Indies is a highly biased subsample of the pre-human fauna, most of the larger endemics having become extinct, as have all, but one, of the flightless birds. Furthermore, this inquiry is limited by the few studies that have been made of the functional biology of the living species belonging to this fauna. A reduction in body mass — A change in mass by an island endemic is difficult to detect without comparison with an appropriate continental relative, but a few observations are relevant. Megalonychid ground sloths were found in the Pleistocene and Holocene on Puerto Rico, Hispaniola, and Cuba (Morgan and Woods, 1986). One of the sloths found in Hispaniola, Acratocnus comes, was the size of living tree sloths, which possibly suggests a reduction in mass in relation to life on islands under the assumption that this species was a ground sloth. Some of the capromyid rodents found on small, flat islands, e.g., Geocapromys ingrahami from the Bahamas, have a much smaller mass than relatives on larger, high islands, e.g., G. browni from Jamaica. The diversity of hummingbirds on the Antilles may be facilitated by their small masses: the smallest hummingbird, the bee (Mellisuga helenae), is a Cuban endemic. On the other hand, the extinct caproymid Amblyrhiza inundata was huge (the size of the black bear!), which may have reflected that it was the only browsing herbivore on St. Martin. The evolution of “approachability” — Few vertebrates on Caribbean islands are approachable, but that may simply mean that the approachable populations and species either learned wariness after the arrival of humans, or they became extinct. The evolution of flightlessness — Nearly all flightless species of birds that inhabited Caribbean islands are extinct. They included, at least, an ibis (Xenicibis xympithecus) from Jamaica (Olson and Steadman, 1979), some rails (e.g., Nesotrochis spp.) on the Virgin Islands, Puerto Rico, and Hispaniola (Olson, 1974), and possibly a large owl (Ornimegalonyx oterori) from Cuba (Olson, 1985). A living rail, Cyanolimnas cerverai in the Zapata swamp of Cuba, either is flightless or nearly so (Raffaele et al., 1998). A reduction in rate of metabolism — Few vertebrates native to the West Indies have had their rates of metabolism measured. One group for which we have some data are rodents of the family Capromyidae. Measurements have been made on Capromys pilorides (McNab, 1978), Geocapromys browni (Ottenwalder, personal communication), and G. ingrahami (Jordan, 1989). These species have basal rates that are 64, 82, and 67%, respectively, of the values expected from body mass in mammals generally (McNab, 1988). These low rates cannot clearly be accounted for by food habits or any other obvious ecological factor (McNab, 1989). The restriction of capromyids to islands may well be a causative factor responsible for their low basal rates (McNab, 1994b; Arends and McNab, in press). The Greater Antilles had two endemic families of insectivores. One is the Solenodontidae, of which two species survive: Solenodon paradoxus on Hispaniola and S. cubanus on Cuba. A giant species of Solenodon was also found on Cuba (Morgan et al., 1980). Unfortunately, Solenodon is little studied, but it is undoubtedly characterized by low rates of metabolism given that these species

Functional Adaptations to Island Life in the West Indies

59

are large (~1 kg), slow moving, and tropical in distribution. I suggest that they may have rates of metabolism somewhat similar to tenrecs, given that both groups are rather large, ground-dwelling insectivores that live in tropical climates, which would give them basal rates that are only about 50% of what is expected from body mass (as is the case in tenrec Setifer). If that is the case, the low rate of metabolism probably contributed to their persistence on these islands, although it may contribute to their demise in the face of imported predators as a result of the propensity of a low rate of metabolism to depress the rate of reproduction in eutherians (McNab, 1980). The second family is the Nesophontidae, which included several large, shrew-like species that belong to the (apparently) extinct genus Nesophontes. The presence of these insectivores in a tropical environment does raise the possibility that they had low rates of metabolism similar to those of crocidurine shrews from the African and Asian tropics (see McNab, 1991). The physiological ecology of Caribbean birds has been completely neglected. One group that responds to life on small islands in the South Pacific are fruit-pigeons of the genus Ducula. In the Caribbean four species of pigeons that belong in the genus Columba appear to have converged on frugivory similar to Ducula. The only species in this group that has had its rate of metabolism measured is the white-crowned pigeon (C. leucocephala). This species has a basal rate that is 100% of the value expected from the nonpasserine curve of Aschoff and Pohl (1970). Small-island specialist Ducula have basal rates that vary between 51 and 61% of the rates expected from the Aschoff–Pohl (1970) standard, whereas mainland Ducula have basal rates between 75 and 91%; an intermediate species, D. bicolor, has a basal rate equal to 71% (McNab, 2000). The whitethroated pigeon (C. vitiensis) is found on mainlands and on smaller islands in the South Pacific; it has a basal rate that is 91% of the value expected from mass. The white-crowned pigeon in parts of its range, such as the Florida Keys, Bahamas, and Lesser Antilles, is a small-island species. It is also resident on Cuba, Jamaica, Hispaniola, and Puerto Rico and occurs in the Keys only during the breeding season (Bancroft, 1992). So, is this a small-island or a large-island species, and how does one distinguish between these island-size categories? However one classifies C. leucocephala, it clearly has a high basal rate by Ducula standards and appears not to be responding to small-island life, at least as does Ducula. Other Caribbean members of Columba, two of which, the plain pigeon (C. inornata) and the ring-tailed pigeon (C. caribaea), should be examined in relation to their restriction to large Caribbean islands, as should the scalynaped pigeon (C. squamosa), which is widely distributed on large and small islands. Caribbean columbids indeed may be more nomadic than most Pacific island species and therefore treat the Caribbean islands as one fragmented continent (J. Sailer, personal communication), thereby avoiding a commitment to small-island life. This may be facilitated by the smaller scale of the Caribbean Sea compared to the Pacific Ocean. Two other groups of birds that should be examined in the West Indies are the trogons (Trogonidae) and the todies (Todidae). Trogons have not been studied anywhere, but a few data indicate a low rate of metabolism in a continental species. Whether the two species of Caribbean trogons, one in Cuba (Prioteles temnurus) and one in Hispaniola (P. roseigaster), with Prioteles a genus unique to the West Indies, have low basal rates is unknown; it may simply be that an inherently low basal rate in trogons (personal observations) facilitates their survival in the Greater Antilles. Todies (Todidae) are also unstudied from the view of energy expenditure. This family is unique to the Greater Antilles with one species each in Cuba, Jamaica, and Puerto Rico, and two in Hispaniola. Their small size (5 to 8 g), insectivorous habits, and tropical distribution make them excellent candidates for low basal rates of metabolism and the use of torpor. Merola-Zwartjes and Ligon (2000), however, have measured high basal rates in the Puerto Rican tody (Todus mexicanus). Todies are especially interesting because they have been found in North America (Olson, 1976) and Europe (Mourer-Chauviré, 1985) during the Oligocene, but at present are relictual in the Greater Antilles. Similarly, the Antilles picolet, Nesoctites micromegas, which is quite different from other picolets (D. W. Steadman, personal communication), is limited in distribution to Hispaniola.

60

Biogeography of the West Indies: Patterns and Perspectives

The occurrence of vertebrates with a relictual distribution on islands near continents is a common phenomenon, as can be seen with tenrecs and lemurs on Madagascar, Coenocorypha snipes in New Zealand, Tasmanian devil (Sarcophilus) and thylacine (Thylacine) in Tasmania, and Solenodon on Hispaniola and Cuba, as well as the todies in the Caribbean. Most of these endotherms will probably turn out to have low rates of metabolism, a condition that may facilitate their persistence on islands, especially in the absence of competition and predation, and contribute to their absence from adjacent continents. The evolution of torpor — No studies of the propensity of Caribbean endotherms to enter to torpor have been made. I am suspicious that most insectivorous bats in the Caribbean enter torpor as they do almost everywhere else. Yet, some of the larger insectivorous bats might not enter torpor, as has been seen to be the case in some tropical environments (see McNab, 1969; Bonaccorso and McNab, submitted). One possibility for the evolution of torpor in the West Indies results from an analogy with the nectarivorous pteropodids in the New Guinea region: the phyllostomid subfamily Phyllonycterinae contains three genera and nine species of small, nectarivorous bats, which in association with their endemism on the islands of the Caribbean raises the possibility that they, unlike other members of the family Phyllostomidae, readily enter torpor. Entrance into torpor by birds tends to be confined to small species, but it has not been much studied in the tropics. Hummingbirds readily enter topor, and one would suppose that Caribbean species would do so as well, but no reason exists to suspect that Caribbean species are more prone to enter torpor than North or South American species, except possibly in relation to a small size. Merola-Zwartjes and Ligon (2000) have shown that female Puerto Rican todies (Todus mexicanus) enter torpor during the breeding season, when exposed to cool temperatures. The propensity of this species to enter torpor is undoubtedly connected with its small mass (5 to 7 g), but possibly also in relation to its restriction to an island distribution. The replacement of endotherms by ectotherms — Some reasonably large ectotherms were native to the Caribbean islands. They include rather large iguanid lizards that belong to the genus Cyclura, which are (or were) found in the Greater Antilles, the Bahamas, and the Virgin Islands. Large tortoises of the genus Geochelone were present on Cuba and Hispaniola (Williams, 1950; Auffenberg, 1967; Franz and Woods, 1983).

CONCLUSION The islands of the West Indies had a native terrestrial vertebrate fauna that shared some of the characteristics that have been described for the faunas of the South Pacific: long-term persistence in the fauna was facilitated by the reduction of resource requirements. This was accomplished in the South Pacific by evolving a small body mass, a reduction in wariness, the selective evolution of flightlessness among birds, a reduction in rate of metabolism, and the acquistion of herbivory by medium- to large-sized ectotherms. The difficulty in giving a complete analysis of these phenomena in the West Indies is that its fauna has suffered a great extinction (because humans have been on these islands for several thousand years) and because the surviving fauna has been poorly studied. In the West Indies, a reduction in mass, the evolution of flightlessness, a reduction in rate of metabolism, and the occurrence of large ectotherms were elements present in the fauna. I hope that this analysis will stimulate the interest of biologists to examine the functional bases of faunal persistence in the living remnants of the native West Indian fauna.

ACKNOWLEDGMENTS I thank Charles Woods and Jeff Sailer for reviewing an earlier draft of this chapter and Charles Woods for inviting me to submit this chapter to this volume.

Functional Adaptations to Island Life in the West Indies

61

LITERATURE CITED Arends, A. and B. K. McNab. In press. The comparative energetics of ‘caviomorph’ rodents. Comparative Biochemistry and Physiology. Aschoff, J. and H. Pohl. 1970. Rhythmic variations in energy metabolism. Federation Proceedings 29:1541–1552. Askins, R. A. and D. N. Ewert. 1991. Impact of Hurricane Hugo on bird populations on St. John, U.S. Virgin Islands. Biotropica 23:481–487. Auffenberg, W. 1967. Notes on West Indian tortoises. Herpetologica 23:34–44. Bancroft, G. T. 1992. A closer look: White-crowned Pigeon. Birding 24:21–24. Bonaccorso, F. A. and B. K. McNab. 1997. Plasticity of energetics in blossom bats (Pteropodidae): impact on distribution. Journal of Mammalogy 78:1073–1088. Bonaccorso, F. A. and B. K. McNab. Submitted. Standard energetics in leaf-nosed bats (Hipposideridae): its relationship to intermittent and protracted foraging tactics in bats and birds. Physiological Zoology, submitted. Darlington, P. J. 1957. Zoogeography: The Geographical Distribution of Animals. John Wiley & Sons, New York. Darwin, C. 1839. Journal of Researches into the Geology and Natural History of the Various Countries Visited by H. M. S. Beagle, Under the Command of Captain Fitzroy, R. N. from 1832 to 1836. J. M. Dent and Sons, London. Flannery, T. 1993. The case of the missing meat eaters. Natural History 102:41–45. Franz, R. and C. A. Woods. 1983. A fossil tortoise from Hispaniola. Journal of Herpetology 17:79–81. Gannon, M. R. and M. R. Willig. 1994. The effects of Hurricane Hugo on bats of the Luquillo Experimental Forest of Puerto Rico. Biotropica 26:320–331. Geiser, F., F. K. Coburn, G. Kortner, and B. S. Law. 1996. Thermoregulation, energy metabolism, and torpor in blossom-bats, Syconycteris australis (Megachiroptera). Journal of Zoology (London) 239:583–590. Humphrey, P. S., B. C. Livezey, and D. Siegel-Causey. 1987. Tameness of birds of the Falkland Islands: an index and preliminary results. Bird Behavior 7:67–72. Lomolino, M. V. 1985. Body size of mammals on islands: the island rule reexamined. American Naturalist 125:310–316. Merola-Zwaretjes, M. and J. D. Ligon. 2000. Ecological energetics of the Puerto Rican Tody: heterothermy, torpor, and intra-island variation. Ecology 81:990–1003. MacArthur, R. H. and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, New Jersey. McNab, B. K. 1969. The economics of temperature regulation in neotropical bats. Comparative Biochemistry and Physiology 31:227–268. McNab, B. K. 1980. Food habits, energetics, and the population biology of mammals. American Naturalist 116:106–124. McNab, B. K. 1988. Complications inherent in scaling the basal rate of metabolism in mammals. Quarterly Review of Biology 63:25–54. McNab, B. K. 1989. On the selective persistence of mammals in South America. Pp. 605–614 in Redford, K. H. and J. F. Eisenberg (eds.). Advances in Neotropical Mammalogy. Sandhill Crane Press, Gainesville, Florida. McNab, B. K. 1991. The energy expenditure of shrews. Pp. 35–45 in Findley, J. S. and T. L. Yates (eds.), The Biology of the Soricidae. The Museum of Southwestern Biology, University of New Mexico, Albuquerque. McNab, B. K. 1994a, Energy conservation and the evolution of flightlessness in birds. American Naturalist 144:628–642. McNab, B. K. 1994b. Resource use and the survival of land and freshwater vertebrates on oceanic islands. American Naturalist 144:643–660. McNab, B. K. 1996. Metabolism and temperature regulation of kiwis (Apterygidae). Auk 113:687–692. McNab, B. K. 2000. The influence of body mass, climate, and distribution on the energetics of South Pacific pigeons. Comparative Biochemistry and Physiology 127A:309–329. McNab, B. K. and M. I. Armstrong. In press. The scaling of energetics in flying foxes of the genus Pteropus. Journal of Mammalogy 82. McNab, B. K. and F. J. Bonaccorso. 2001. The metabolism of New Guinean pteropodid bats. Journal of Comparative Physiology B. 171:201–214.

62

Biogeography of the West Indies: Patterns and Perspectives

McNab, B. K. and C. A. Salisbury. 1995. Energetics of New Zealand’s temperate parrots. New Zealand Journal of Zoology 22:339–349. Morgan, G. S. and C. A. Woods. 1986. Extinction and the zoogeography of West Indian land mammals. Biological Journal of the Linnean Society 28:167–203. Morgan, G. S., C. E. Ray, and O. Arredondo. 1980. A giant extinct insectivore from Cuba (Mammalia: Insectivora: Solenodontidae). Proceedings of the Biological Society of Washington 93:597–608. Mourer-Chauviré, C. 1985. Les Todidae (Aves: Coraciiformes) des Phosphorites du Quercy (France). Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen 88:407–414. Olson, S. L. 1974. A new species of Nesotrochis from Hispaniola, with notes on other fossil rails from the West Indies (Aves: Rallidae). Proceedings of the Biological Society of Washington 87:439–450. Olson, S. L. 1976. Oligocene fossils bearing on the origins of the Todidae and the Momotidae (Aves: Coraciiformes). Smithsonian Contributions to Paleobiology 27:111–119. Olson, S. L. 1985. The fossil record of birds. Pp. 79–252 in Farner, D. S. and J. R. King (eds.). Avian Biology, Vol. III. Academic Press, New York. Olson, S. L. and D. W. Steadman. 1979. The humerus of Xenicibus, the extinct flightless ibis of Jamaica. Proceedings of the Biological Society of Washington 92:23–27. Pedersen, S. C., H. H. Genoways, and P. W. Freeman. 1996. Notes on bats from Montserrat (Lesser Antilles) with comments concerning the effects of Hurricane Hugo. Caribbean Journal of Sciences 32:206–213. Raffaele, H., J. Wiley, O. Garrido, A. Keith, and J. Raffaele. 1998. A Guide to the Birds of the West Indies. Princeton University Press, Princeton, New Jersey. Sailer, J. K. 1999. Biogeography and Ecomorphology in Fruit-Doves (Columbidae: Ptilinopus). Master’s thesis, University of Florida, Gainesville. Steadman, D. W. 1995. Prehistoric extinctions of Pacific island birds: biodiversity meets zooarchaeology. Science 267:1123–1131. Waide, R. B. 1991a. The effect of Hurricane Hugo on bird populations in the Luquillo Experimental Forest. Biotropica 23:475–480. Waide, R. B. 1991b. Summary of the responses of animal populations to hurricanes in the Caribbean. Biotropica 23:508–512. Wallace, A. R. 1880. Island Life, or the Phenomena and Causes of Insular Faunas and Floras, Including a Revision and Attempted Solution of the Problem of Geological Climates. Macmillan, London. Williams, E. 1950. Testudo cubensis and the evolution of Western Hemisphere tortoises. Bulletin of the American Museum of Natural History 95:1–36.

and Biogeography 5 Phylogeny of Lyonia sect. Lyonia (Ericaceae) Walter S. Judd Abstract — The biogeographic relationships of 17 geographical regions of high endemism within the Caribbean region were assessed through a preliminary Brooks parsimony analysis, which employed as characters 53 varietal, specific, and supraspecific taxa of Lyonia sect. Lyonia. This monophyletic group is represented within the Greater Antilles by 25 species, many of which are narrow endemics, and the phylogeny of the group is fairly well understood. This analysis resulted in the discovery of two equally parsimonious area cladograms that differed only in that the positions of Puerto Rico and the Massif de la Hotte (of southeastern Hispaniola) are switched. The area cladograms indicate that all the Hispaniolan localities plus Puerto Rico and St. Thomas constitute a clade, with a sister area relationship expressed between the Cordillera Central/Massif du Nord and the Massif de la Selle/Sierra de Baoruco. All Cuban geographical regions form a monophyletic group (except for the Sierra de Trinidad, in central Cuba), with all the localities in the Oriente region constituting a distinct subclade. Within the Oriente region, the Sierra Maestra/Gran Piedra shows a sister group relationship to the Sierra de Nipe/Sierra de Cristal/Moa, Toa and Baracoa region. These biogeographical results suggest that the species of Lyonia sect. Lyonia growing on each island of the Greater Antilles are nearly always most closely related to others on the same island. It is likely that tectonic events, wind dispersal, and climatic changes have all influenced the present distribution of species of Lyonia sect. Lyonia.

INTRODUCTION Lyonia Nuttall (Ericaceae, Vaccinioideae, Lyonieae), a genus of 36 species (52 taxa) of trees and shrubs, occurs in eastern Asia (Japan to Pakistan, south to the Malay Peninsula), the Greater Antilles (including St. Thomas), and continental North America (eastern United States and Mexico) (Judd, 1995a). The genus is related to Pieris D. Don, Agarista D. Don, and Craibiodendron W. W. Smith, which together constitute the tribe Lyonieae (Judd, 1979, 1981; Kron and Judd, 1997, 1999). This tribe is diagnosed by the presence of bands of fibers in the phloem, a lignified leaf epidermis, biseriate-stalked gland-headed hairs, and seeds with a testa of elongated cells. Stamens with S-shaped filaments may be an additional synapomorphy. The monophyly of Lyonia is strongly supported in phylogenetic analyses employing morphological and/or molecular (rbcL and matK nucleotide sequences) characters (Kron and Judd, 1997, 1999). Morphological synapomorphies for the species of Lyonia include the presence of multicellular gland-headed hairs on the corolla and ovary, spurs borne on the filament, disintegration tissue extending onto the spurs, and the prominently thickened capsule sutures (see Judd, 1979, 1981); the last feature is especially distinctive. Other characters useful in identification are outlined in recent monographic treatments of the genus (Judd, 1981, 1990, 1995a), and the closely related genera Agarista, Craibiodendron, and Pieris also have been well studied (Judd, 1982, 1984, 1986, 1995b, 1995c; Judd and Hermann, 1990). Phylogenetic relationships within Lyonia have been investigated through a series of phylogenetic analyses employing morphology (Judd, 1979, 1981, 1995a), nucleotide sequences of the plastid genes rbcL and matK, and combined analyses (Kron and Judd, 1997, 1999). These analyses strongly support the placement of Lyonia sect. Lyonia, i.e., those members of the genus with ferruginous peltate scales, as the sister group of the remaining species. The monophyly of Lyonia sect. Lyonia

0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

63

64

Biogeography of the West Indies: Patterns and Perspectives

is well supported by the roughened filaments, tailed seeds, and the distinctive peltate scales (i.e., a lepidote indumentum) that cover all parts of these plants. Phylogenetic relationships among the species of Lyonia sect. Lyonia are fairly well understood (see Judd, 1981, 1995a), although at present they are based only upon morphological characters. This diverse group contains the majority of species within the genus, i.e., 28 species (Judd, 1995a), and is most diverse within the Greater Antilles with 12 species (16 taxa) occurring in Cuba, 10 species (11 taxa) in Hispaniola, 2 in Jamaica, 2 in Puerto Rico, and 1 in St. Thomas. Only three members of Lyonia sect. Lyonia occur in continental North America — L. squamulosa Martens & Galeotti (Mexico), and L. ferruginea (Walter) Nutt., and L. fruticosa (Michx.) G. S. Torrey (Florida, Georgia, South Carolina). With the exception of L. stahlii Urban and L. truncata Urban, all the species occurring in the Greater Antilles are restricted to a single island. Lyonia stahlii occurs on two islands but has differentiated into two geographically isolated varieties: var. costata (Urban) Judd on Hispaniola and var. stahlii on Puerto Rico. Lyonia truncata occurs on Hispaniola (vars. truncata and montecristina (Urban and Ekman) Judd) and on Puerto Rico (var. proctorii Judd). Species of Lyonia sect. Lyonia are rather narrowly endemic and often are limited to a particular mountain range, specific ecological situations, or restricted a elevational range (see Judd, 1981, 1995a). However, on each island a few taxa — such as L. truncata var. montecristina, L. stahlii var. costata on Hispaniola, or L. macrophylla (Britton) Ekman ex Urban, L. affinis (A. Richard) Urban, and L. latifolia (A. Richard) Griseb. on Cuba — are relatively widespread. Lyonia sect. Lyonia has two major centers of endemism in Hispaniola: the Cordillera Central/Massif du Nord, where eight taxa are indigenous and five are endemic, and the Massif de la Selle/Sierra de Baoruco, with six taxa, three of which are endemic (Judd, 1981, 1995a). The species occur there in a wide range of elevations, i.e., ~200 to 3175 m, and many species are seemingly elevationally isolated from their close relatives (Judd, 1981). In the Cordillera Central this pattern is best illustrated by three pairs of species: L. heptamera Urban – L. buchii Urban, L. tuerckheimii Urban – L. stahlii var. costata, and L. urbaniana (Sleumer) Jiménez – L. tinensis Urban; the first member of each species pair occurs at higher elevations and the second at lower elevations. Similar patterns can be found in the Massif de la Selle/Sierra de Baoruco, e.g., L. alpina Urban & Ekman occurs at much higher elevations than the related L. truncata. Other taxa of Hispaniola are isolated by their occurrence in mountain ranges that are separated by low areas of xerophytic vegetation, e.g., L. microcarpa Urban & Ekman and L. urbaniana, or L. truncata vars. truncata and montecristina. The two Jamaican species, L. jamaicensis (Swartz) D. Don and L. octandra (Swartz) Griseb., are also elevationally isolated in the Blue Mountains. In contrast to the species of Hispaniola and Jamaica, those of Cuba, with a few exceptions, are rather uniform with respect to elevational distribution but tend to be geographically isolated in different mountain ranges. Most species are limited to either the Oriente region (of eastern Cuba) or Pinar del Río (of western Cuba), although a single species, L. trinidadensis Judd, occurs in the mountains of Las Villas Province, near the center of the island. Lyonia is especially diverse in the mountains of the Oriente region (Judd, 1981, 1995a). Within this region, the Sierra de Cristal, Sierra Maestra, Gran Piedra region, Sierra de Nipe, Moa Plateau, Sierra de Moa, Sierra de Toa, and Baracoa region show the greatest diversity in species of Lyonia, and most of these areas have at least one endemic taxon (see Judd, 1981, 1995a). Some geographically isolated Cuban taxa include L. nipensis Urban var. nipensis – var. depressinerva Judd, L. obtusa Griseb. – L. longipes Urban, L. latifolia var. latifolia – var. calycosa (Small) Judd, L. affinis – L. elliptica (C. Wright ex Small) Alain, and L. glandulosa (A. Richard) Urban var. glandulosa – var. revolutifolia Judd – var. toaensis (Acuña & Roig) Berazaín. In contrast to the lepidote species (i.e., Lyonia sect. Lyonia), which are essentially a montane Antillean clade, the non-lepidote species of Lyonia occur primarily in moist forests of eastern North America and eastern Asia. Only L. lucida (Lam.) K. Koch of sect. Maria (DC.) C. E. Wood (see Judd 1981, 1995a) also grows in the Antilles, occurring on the acidic sands of the coastal plain of the southeastern United States and western Cuba. Continental species tend also to have very broad

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae)

65

geographical ranges — see especially L. ligustrina (L.) DC., L. mariana (L.) D. Don, L. lucida, L. villosa (Wallich ex Clark) Hand.-Mazz., and L. ovalifolia (Wallich) Drude (Judd, 1981). The species of Lyonia sect. Lyonia occur in a wide variety of habitats, but characteristically prefer acid soils. They may occur, however, over limestone. For example, L. truncata, L. alpina, and L. microcarpa grow in lateritic soil filling the cracks of eroded limestone rocks, and L. stahlii var. costata grows on organic soil developed over a limestone bedrock. Many Cuban and Hispaniolan species occur on red lateritic soils developed by the weathering of underlying serpentine rocks. The species of western Cuba occur on siliceous soils; some Hispaniolan species are found on soils derived from igneous rocks. Several Caribbean taxa (especially those of cloud forests, such as L. octandra and L. stahlii) may also be found in highly organic soils. The Caribbean species occur in moist montane or cloud forests, high- or low-elevation pine forests, savannas, thickets, or dry rocky scrub. They range in elevation from nearly sea level to 3175 m. Thus, L. heptamera Urban can be found at the top of the highest peak of Hispaniola (Pico Duarte, 3175 m) while L. myrtilloides Griseb. and L. ekmanii Urban grow near sea level in Pinar del Rio, Cuba, and L. macrophylla occurs just above sea level near Moa, Cuba. Within the Greater Antilles, Lyonia sect. Lyonia shows a “Western Continental” distribution pattern; genera with this pattern are missing from the Lesser Antilles, are often most diverse in Cuba, and decrease in diversity eastward to the Virgin Islands (Judd, 1981). Howard (1973) lists numerous genera exhibiting this pattern, although biogeographical patterns are not approached from a phylogenetic standpoint, limiting a precise understanding of the underlying historical processes. Because of the existence of numerous narrowly endemic species that occupy a wide range of plant communities (and elevations), Lyonia sect. Lyonia is an ideal clade to use in studying biogeographical relationships within the Greater Antilles. Well-supported phylogenies exist for very few Antillean plant groups. Some genera that have been analyzed in the past few years include Mecranium (Melastomataceae; Skean, 1993); Pictetia (Leguminosae, Faboideae; Beyra and Lavin, 1999); Poitea (Leguminosae, Faboideae; Lavin, 1993); and Sabal (Palmae; Zona, 1990). In this chapter, I use the hypothesized phylogenetic relationships among species (and varieties) of Lyonia sect. Lyonia (see Judd, 1981, 1995a) to investigate biogeographical relationships within the Greater Antilles. Biogeographical patterns derived from geographical distributions of species (and supraspecific clades) within Lyonia sect. Lyonia are, then, briefly compared with patterns seen in Mecranium, Pictetia, Poitea, and Sabal.

PHYLOGENETIC RELATIONSHIPS WITHIN LYONIA SECT. LYONIA Cladistic relationships of the numerous species within section Lyonia were initially investigated by Judd (1981) by means of a manual, Wagner-Groundplan-Divergence analysis. Later Judd (1995a) reassessed the morphological characters employed in this analysis. A total of 29 phenotypic characters (of potential phylogenetic significance) from this taxonomic treatment were selected for a second analysis, and these characters were delimited into discrete states. The phylogenetic relationships of the Antillean species, along with L. squamulosa, L. ferruginea, and L. fruticosa, then were reanalyzed using heuristic and branch-and-bound algorithms of both PAUP 2.4.1 and Hennig86, version 1.5 (see Judd, 1995a, for methodological details). The resulting trees were rooted using Lyonia sects. Pieridopsis (Rehder) Airy Shaw and Maria as functional outgroups, along with Craibiodendron and Agarista, in sequence. A representative cladogram and strict consensus tree (of all trees found in both PAUP and Hennig86 analyses) are presented in Figure 1; relationships within the subclades designated “Cuban spp.” and “Hispaniolan spp.” are shown in Figure 2. Note that L. maestrensis Acuña & Roig was deleted from the cladistic study because mature flowers of this species have not been seen; for other details relating to this analysis see Judd (1995a). The Antillean species of Lyonia sect. Lyonia may belong to two major clades. The first is called the “Cuban group” because all the taxa within it are native to Cuba (see Figures 1 and 2; Judd, 1981: figure 2). The monophyly of the Cuban clade is supported by the apomorphy of the adaxial

66

Biogeography of the West Indies: Patterns and Perspectives

6 14

L. ferruginea 17 1 (0)

2 22 11 12 (0)

L. squamulosa

2 18 22

25

L. fruticosa

“Hispaniolan” spp.

L. alainii 7

4

11 2

A

14 12 (0) 12 16 (0)

17 4

L. tuerckheimii L. rubiginosa L. stahlii var. costata L. stahlii var. stahlii L. jamaicensis L. trinidadensis

2 18 19 21

“Cuban” spp.

L. octandra 12 (0)

L. ferruginea L. fruticosa L. squamulosa “Hispaniolan” spp.

L. alainii L. tuerckheimii L. rubiginosa

B

L. stahlii var. costata L. stahlii var. stahlii L. jamaicensis L. trinidadensis “Cuban” spp.

L. octandra FIGURE 1 Representative cladogram (A) and strict consensus (B) tree resulting from PAUP and Hennig86 analyses of cladistic relationships of selected species of Lyonia sect. Lyonia (from Judd, 1995a). “Cuban spp.” = L. affinis, L. elliptica, L. ekmanii, L. myrtilloides, L. macrophylla, L. longipes, L. obtusa, L. nipensis, L. glandulosa, and L. latifolia; “Hispaniolan spp.” = Hispaniolan species with abaxially pubescent leaves, i.e., L. truncata, L. alpina, L. tinensis, L. urbaniana, L. microcarpa, L. buchii, and L. heptamera.

leaf epidermal cells with inner periclinal walls strongly thickened and lignified. The second major clade is called the Hispaniolan group (see Judd, 1981) because all the taxa within it are native either to Hispaniola or to the adjacent islands of Puerto Rico and St. Thomas. The monophyly of this clade is less well supported; the group was paraphyletic in the initial analyses of Judd (1995a) but becomes monophyletic if leaf-margin condition (i.e., strongly and irregularly toothed) is weighted by two (see Judd, 1995a, for justifications), and this condition then becomes synapomorphic for

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae)

8 (0)

21

L. microcarpa

5 8 (2)

19

13 (2)

67

L. urbaniana L. tinensis

9

19

L. alpina 2

3

13

14

L. buchii

15 5 25 (0)

16 (2)

19 (2) 26 (2)

L. heptamera

20

L. truncata 8

11 (0) 24

18 (0)

27 (0)

10

L. latifolia var. calycosa

29 14

2 23

L. latifolia var. latifolia 9

19 (0)

2

9 23

16 6

L. nipensis var. nipensis L. n. var. depressinerva L. obtusa

14

4 27

L. longipes 2 (0)

17 (0)

11

10

22 (0)

27 (0)

21 25 (0)

L. macrophylla L. myrtilloides

22 (2) 10

L. ekmanii

21 (0)

L. elliptica 2

18

19 (0) 4 11

26

28

16 6 14

L. affinis L. glandulosa var. toaensis L. g. var. revolutifolia L. g. var. glandulosa

FIGURE 2 Character state changes for Hispaniolan species of Lyonia with abaxially pubescent leaves (upper cladogram) and Cuban species of Lyonia with inner periclinal walls of epidermal cells strongly thickened and lignified (lower cladogram); see Judd (1995a) for more details. Note: Lyonia nipensis var. depressinerva is probably misplaced (see text).

these species. The phylogenetic position of L. jamaicensis, a species of low elevations of the Blue Mountains of Jamaica, is unclear. The relationships within the Cuban clade are at least partly resolved, and L. affinis, L. elliptica, and L. glandulosa are likely basal. The latter species is patristically distinctive, possessing numerous autapomorphies, e.g., 4-merous flowers with extremely small corollas, and extremely narrowly obovoid to ellipsoid capsules lacking a visible articulation with the pedicel. Lyonia latifolia vars. latifolia and calycosa form a distinctive monophyletic group, which is characterized by goldencolored peltate scales, elongated calyx lobes with the adaxial surface densely covered by peltate scales, and elongate-urceolate, densely lepidote corollas. The remaining members of the Cuban clade were considered to form the “L. obtusa – L. nipensis line” by Judd (1981), but the monophyly of this group was not supported in the computer-based analyses of Judd (1995a). Lyonia obtusa and L. longipes may be related, as indicated by their embedded leaf veins. I consider L. nipensis var. depressinerva to be most closely related to L. nipensis on the basis of their densely abaxially pubescent leaves (with nonsunken stalks of the peltate scales) and densely lepidote corollas, but

68

Biogeography of the West Indies: Patterns and Perspectives

L. nipensis var. depressinerva formed a clade with L. obtusa in the cladistic analyses (see Figure 2). I consider it likely, however, that the strongly revolute leaves and impressed secondary veins of L. obtusa and L. nipensis var. depressinerva have evolved independently, in response to the rigorous edaphic conditions occurring in the plant communities developed on the strongly lateritic soils of northern Oriente. Finally, L. trinidadensis, of the Trinidad Mountains in central Cuba, and L. octandra, of high elevations in the Blue Mountains of Jamaica, are weakly linked with the members of the Cuban clade. Within the Hispaniolan clade, L. stahlii, L. tuerckheimii, L. rubiginosa (Persoon) G. Don, and L. alainii Judd form a basal paraphyletic assemblage; these species all lack unicellular hairs on the abaxial leaf surface (except some individuals of L. rubiginosa), have obscure to laxly reticulate leaf veins, and are more or less sparsely to moderately lepidote. The derived, and clearly monophyletic group of Hispaniolan species, called the L. microcarpa – L. truncata – L. heptamera clade in Judd (1981), possess the synapomorphy of leaves with a dense layer of unicellular hairs on the abaxial leaf surface. Lyonia truncata is the sister species to the remaining members of this group. The remaining species (i.e., L. alpina, L. tinensis, L. microcarpa, L. urbaniana, L. buchii, and L. heptamera) are united by the apomorphy of the stalks of their peltate scales being not sunken into the abaxial leaf epidermis. Lyonia tinensis, L. microcarpa, and L. urbaniana are distinctive in that their leaves have a densely and finely raised-reticulate network of veins of the lower surface. Lyonia buchii and L. heptamera comprise a strongly supported monophyletic group; they share the following apomorphies: petioles lacking unicellular hairs and with medullary bundles with the xylem cylinder, leaf veins with a distinctive lignified sheath, very strongly and coarsely raised and reticulate venation of the abaxial leaf surface, 6- or 7-merous, very large, strongly carnose corollas, and very large, subglobose to shortly ovoid capsules. Lyonia rubiginosa (of St. Thomas) may be more closely related to the derived Hispaniolan species (and especially to L. truncata) than is indicated in the cladograms (Figures 1 and 2), especially if the abaxially pubescent leaves of this species are taken as synapomorphic.

BIOGEOGRAPHICAL INVESTIGATION Brooks (1981, 1985) developed an additive binary coding procedure that assumed that when several hosts were infected with one species of parasite, that group of hosts must be monophyletic, and this approach has been transferred to analogous biogeographical situations (Wiley et al. 1991; Brooks and McLennan, 1991). Zandee and Roos (1987) considered that widespread taxa should be considered indicative of monophyletic groups of areas, and this has been described as assumption 0 (in comparison with assumptions 1 and 2 of component analysis; see Forey et al., 1992). Although a single phylogeny, such as that presented here for Lyonia, cannot by itself support vicariance hypotheses, it is heuristic to consider what the distribution of various species and supraspecific clades of Lyonia suggest regarding geographical affinities within the Greater Antilles. Thus, a preliminary Brooks parsimony analysis was conducted, and it is compared with the hypothesis of the geohistory of the Greater Antilles presented by Rosen (1976, 1985). The results of this analysis are also briefly compared with published phylogenies and geographical distributions of several other genera that occur (with numerous endemics) within the Greater Antilles.

METHODS A total of 17 geographical regions were delimited and scored for the presence (or absence) of members of 53 varietal, specific, or supraspecific taxa (of Lyonia sect. Lyonia). These geographical regions (Table 1) were in nearly all cases particular mountain ranges, which were occupied by one (or more) species of Lyonia, and their delimitation from one other was not problematic (see Figures 3 and 4). The Sierra de Neiba was omitted from the analysis because only the widespread taxon L. stahlii var. costata occurs there; the Massif des Cahos was similarly omitted as it is represented

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae)

69

TABLE 1 Geographical Regions Used in Brooks Parsimony Analysis 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Outgroup (eastern North America) Mexico Massif de la Selle/Sierra de Baoruco (southwestern Hispaniola) Massif de la Hotte (southwestern Hispaniola) Cordillera Central/Massif du Nord (north-central Hispaniola) Sierra de Nipe (eastern Cuba) Sierra de Cristal/Micara (eastern Cuba) Moa/Toa region (eastern Cuba) Baracoa region (eastern Cuba) Sierra Maestra (eastern Cuba) Gran Piedra (eastern Cuba) Sierra de Trinidad (central Cuba) Pinar del Río (western Cuba) Isla de Pinos (western Cuba) Blue Mountains (Jamaica) Puerto Rico Saint Thomas

Note: For more details concerning delimitation of these regions see Figures 3 and 4, and Judd, 1981, 1995a.

only by L. buchii, a species of broad occurrence on Hispaniola. The Massif de la Hotte, although occupied only by L. stahlii var. costata, was included because the populations occurring in the region are distinctive in having mainly 4-merous flowers. Additionally, this region is a major center of endemism for some other Antillean genera and is, therefore, of more general interest. A list of the monophyletic taxa that were derived from the cladograms are presented in Figures 1 and 2, and these formed the “characters” employed in the Brooks parsimony analysis (see Table 2). The matrix of “characters” for each geographical region is provided in Table 3, and this was analyzed using the heuristic search algorithm of Hennig86, version 1.5 (Farris, 1988), using the ie-algorithm, which identifies one tree, certain to be of minimal length; extensive branch swapping (bb*) was applied to this tree.

RESULTS The Hennig86 search resulted in the discovery of two equally parsimonious trees of 57 steps (consistency index (CI) = 0.92, retention index (RI) = 0.95), which differ only in that the positions of Puerto Rico and the Massif de la Hotte (of southeastern Hispaniola) are switched. The strict consensus tree is shown in Figure 5. Inspection of the area cladogram (Figure 5) indicates that all the Hispaniolan localities plus Puerto Rico and St. Thomas constitute a clade, with a sister area relationship expressed among the Cordillera Central/Massif du Nord (of north-central Hispaniola) and the Massif de la Selle/Sierra de Baoruco (of southwestern Hispaniola). Thus a close biohistorical relationship between Hispaniola, Puerto Rico, and St. Thomas (of the U.S. Virgin Islands) is supported. Likewise, all the Cuban geographical regions form a monophyletic group (except for the Sierra de Trinidad, in central Cuba). All the localities in eastern Cuba (i.e., the Oriente region) constitute a clade, with the southern mountain ranges (Sierra Maestra/Gran Piedra) showing a sister group relationship to the northern mountain ranges (Sierra de Nipe, Sierra de Cristal, and mountains in the vicinity of Moa, Toa, and Baracoa). Within the “northern group” the geographically adjacent Moa/Toa region and Baracoa region constitute sister areas, as do the Sierra Maestra and Gran Piedra regions in the “southern

70

Biogeography of the West Indies: Patterns and Perspectives

1

2 3

21

a

c b

d e f

20

h g

j

i

77

76

75

FIGURE 3 Geographical areas of Cuba where Lyonia occurs (above, right): (1) Pinar del Río and Isla de la Juventud; (2) Sierra de Trinidad and Sierra de Sancti Spiritus; (3) mountains of the Oriente region. Geographical areas of Oriente (below): (a) Sierra de Nipe; (b) Sierra de Micara; (c) Sierra de Cristal; (d) Moa region; (e) Sierra de Moa; (f ) Sierra de Toa; (g) Sierra del Frijol; (h) Baracoa region; (i) Sierra Maestra; (j) Gran Piedra.

20 a

b

c d e

f g

18 74

72

70

FIGURE 4 Geographical areas of Hispaniola where Lyonia occurs: (a) Massif du Nord; (b) Cordillera Central; (c) Massif des Cahos; (d) Sierra de Neiba; (e) Massif de la Hotte; (f ) Massif de la Selle; (g) Sierra de Baoruco.

group.” The Oriente region (i.e., eastern Cuban localities) form the sister group to the closely related Pinar del Río and Isla de Pinos regions (of western Cuba). The localities of western Cuba, therefore, show the greatest affinity to those of eastern Cuba. Finally, the Sierra de Trinidad and Jamaica are weakly linked with the Cuban geographical areas.

DISCUSSION The pattern of biogeographical relationships expressed in Figure 5 is not surprising given that the cladograms for Lyonia sect. Lyonia strongly suggest that the species growing on each island of the Greater Antilles are nearly always more closely related to others on the same island. It is noteworthy that the most ancestral species tend to occur on St. Thomas, Puerto Rico, and Jamaica. In contrast, many species of the Cordillera Central or Massif de la Selle (of Hispaniola) or the mountains of

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae)

71

TABLE 2 Monophyletic Taxa Used as “Characters” in Parsimony Analysis of Geographical Regions Listed in Table 1 L. heptamera; L. buchii; L. urbaniana; L. alpina; L. alainii; L. tinensis; L. microcarpa; L. truncata var. truncata; L. truncata var. montecristina; L. truncata var. proctorii; L. truncata; L. tuerckheimii; L. stahlii var. stahlii; L. stahlii var. costata; L. stahlii; L. rubiginosa; L. jamaicensis; L. octandra; L. trinidadensis; L. nipensis var. nipensis; L. nipensis var. depressinerva; L. nipensis; L. affinis; L. macrophylla; L. obtusa; L. longipes; L. latifolia var. latifolia; L. latifolia var. calycosa; L. latifolia; L. ekmanii; L. myrtilloides; L. glandulosa var. glandulosa; L. glandulosa var. revolutifolia [= range of L. glandulosa]; L. glandulosa var. toaensis; L. squamulosa; L. fruticosa + L. ferruginea; L. ferruginea + L. fruticosa + L. squamulosa clade; Hispaniolan clade (with abaxially pubescent leaves); L. microcarpa + L. urbaniana clade; L. microcarpa + L. urbaniana + L. tinensis clade; L. buchii + L. heptamera clade; Cuban clade; L. obtusa + L. longipes clade; L. nipensis + L. latifolia clade; L. obtusa + L. longipes + L, macrophylla clade; L. nipensis + L. latifolia + L. macrophylla + L. obtusa + L. longipes clade; the previous + L. myrtifolia; the previous + L. ekmanii; the previous + L. affinis; Hispaniolan clade + L. alainii + L. tuerckheimii + L. rubiginosa + L. stahlii; Cuban clade + L. trinidadensis + L. octandra; Antillean clade. Note: Lyonia elliptica was not considered due to its imperfectly known range; L. maestrensis was not included in the analysis because it is incompletely known.

TABLE 3 Data Matrix for Geographical Areas Used in Brooks Parsimony Analysis Outgroup Mexico Selle Hotte Central Nipe Cristal Moa/Toa Baracoa Maestra Piedra Trinidad Pinar Rio Isla Pinos Jamaica Puerto Rico St. Thomas

10a 0000000000 0000000000 0101001101 0000000000 1110110011 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000001 0000000000

20 0000000000 0000000000 1100100000 0100100000 1100100000 0000000001 0000000001 0000000001 0000000000 0000000000 0000000000 0000000010 0000000000 0000000000 0000001100 0011100000 0000010000

30 0000000000 0000000000 0000000000 0000000000 0000000000 0101000000 0101010100 1101100000 1101100000 0010001010 0000000110 0000000000 0000000001 0000000000 0000000000 0000000000 0000000000

40 0000011100 0000100100 0000000011 0000000000 0000000011 0110000000 0010000000 0011000000 0000000000 0000000000 0000000000 0000000000 1000000000 1000000000 0000000000 0000000000 0000000000

50 0000000000 0000000000 1100000000 0000000000 1100000000 0010111111 0011111111 0011111111 0011111111 0010101111 0010101111 0000000000 0010000111 0010000111 0000000000 0000000000 0000000000

60 000 000 101 101 101 011 011 011 011 011 011 011 011 011 011 101 101

a

Two-digit numbers on top are counting devices, grouping characters in sets of ten. Note: 0 = taxon absent; 1 = taxon present.

northern Oriente (of Cuba) are morphologically distinctive, representing specialized subclades. The generalized species of each island tend to be similar morphologically and anatomically, and they usually occur in mesic to slightly xeric habitats, such as cloud forest, moist montane forest, lowelevation pine forest, or dry thickets. In contrast, the derived species of each of the islands are very different from each other and are usually adapted to more rigorous habitats, such as thickets on red lateritic soils, white-sand savannas, high-elevation pine forests on igneous soils, or pine forests/ thickets over limestone.

72

Biogeography of the West Indies: Patterns and Perspectives

Outgroup Mexico Cordillera Central Massif de la Selle Puerto Rico Massif de la Hotte Saint Thomas Sierra de Trinidad Jamaica Isla de Pinos Pinar del Rio Gran Piedra Sierra Maestra Sierra de Nipe Sierra de Cristal Moa/Toa region Baracoa region FIGURE 5 Strict consensus of two-area cladogram for Antillean regions included in Brooks parsimony analysis.

It is likely that tectonic events, wind dispersal, and climatic changes have all influenced the present distribution of species of Lyonia sect. Lyonia. Despite these diverse factors, it is worthwhile to address the fascinating biogeography of the Greater Antilles through an investigation of this plant group, which shows such a large number of species of restricted distribution in the region, and which has a fairly well-supported phylogeny. Rosen (1976, 1985) proposed the first comprehensive vicariance model for the biota of the Greater Antilles, and it is thus useful to discuss the geography of Lyonia within the framework of Rosen’s hypothesis (see also Page and Lydeard, 1994). Rosen suggested that the present biota of the region has resulted from the fragmentation of a “Proto-Antilles” archipelago that existed between North and South America in the Mesozoic. The existence of such an island archipelago is supported by the work of many geologists (e.g., Malfait and Dinkleman, 1972; Sykes et al., 1982; Perfit and Williams, 1989; Pindell and Barrett, 1990). Geological work suggests a relationship between western Cuba and southwestern Hispaniola because these two continental fragments probably were connected as a single island early in the Tertiary. Puerto Rico, north-central Hispaniola, and eastern Cuba probably also formed a single island during the same time period. Present-day areas of endemism, which presumably were involved in early Tertiary vicariant events, are eastern Cuba, western Cuba, southwestern Hispaniola, north-central Hispaniola, and Puerto Rico (see also Page and Lydeard, 1994). It is clear that Cuba and Hispaniola were formed by accretion of several different landmasses. Lyonia was probably radiating during this period, as the Ericaceae, a family of the asterid clade (within the tricolpate angiosperms), has a long fossil record, and the existence of its numerous endemics allows, at least, a preliminary testing of these general vicariance hypotheses. The results presented here strongly support the hypothesized close association of north-central Hispaniola with Puerto Rico (and St. Thomas) and also indicate a link with Cuba. Basically, the area cladogram (Figure 5) supports a sister area relationship between the two major regions: north-central Hispaniola and eastern Cuba. However, eastern Cuba shows a closer relationship to western Cuba than to southwestern Hispaniola, and the latter shows a closer relationship to north-central Hispaniola than to western Cuba. Perhaps this pattern results from the fact that Lyonia may once have been restricted to (and diversified within) the eastern Cuba – north-central Hispaniola – Puerto Rico island mass, and only later moved into southwestern Hispaniola and western Cuba, through repeated longdistance dispersal events from north-central Hispaniola and eastern Cuba, respectively. Lyonia sect.

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae)

73

Lyonia is most diverse in north-central Hispaniola and eastern Cuba, lending some support to this hypothesis. Certainly, the presence of two species of Lyonia in the Blue Mountains of Jamaica results from long-distance dispersal (since this island was submerged during the Oligocene; see Buskirk, 1985). Only one species of Lyonia occurs in the Massif de la Hotte (i.e., L. stahlii var. costata), and this occurrence likely also results from a long-distance dispersal event (because the species also occurs in the adjacent Massif de la Selle, and individuals in both regions often show 4-merous flowers, an unusual condition within this species). The very small seeds of members of Lyonia sect. Lyonia are provided with “tails” and are wind dispersed, although long-distance movement of these seeds must be quite rare, as indicated by the typical limited geographical ranges shown by these species. As with many Antillean taxa, the closest relatives of these species are in Mexico and eastern North America (see Howard, 1973, for many other examples), and this pattern fits with the hypothesis that these Antillean landmasses are continental fragments that originated in the region of southern Mexico. As mentioned above, only a few of the plant genera showing numerous endemic species in the Greater Antilles have received recent phylogenetic study. The results of cladistic analyses of Mecranium, Poitea, Pictetia, and Sabal, as they pertain to Antillean biogeography, are briefly outlined below. Like Lyonia, these are all woody taxa, but they show interesting variation in fruit type and dispersal ability. The genus Mecranium J. D. Hook. (Melastomataceae), comprising 24 species, is endemic to the Greater Antilles, and shows an extremely high level of endemism within southwestern Hispaniola (and especially the Massif de la Hotte). Of the 14 species that occur in southwestern Hispaniola (Skean, 1993; and personal communication), 11 are endemic to the region; 10 species occur in the Massif de la Hotte, and 8 of these are restricted to this mountain range. This distribution pattern contrasts strongly with that of Lyonia because most species of Mecranium occur in the Massif de la Hotte, a region with only one species of Lyonia, and that one is nonendemic. The distribution and phylogenetic relationships (developed through morphology-based cladistic analysis) of Mecranium (see Skean, 1993) suggest that Cuba and Hispaniola are composite landmasses, a conclusion not apparent from the distribution of Lyonia sect. Lyonia. The phylogeny of the genus supports the view that “the ancestral species that gave rise to Mecranium possibly was isolated on the south island of Hispaniola, underwent an extensive adaptive radiation as the island was moved toward the bulk of the Greater Antilles, and species were subsequently dispersed to Jamaica, north island Hispaniola, Cuba, and Puerto Rico” (Skean, 1993). The fruits of Mecranium are purple-black, globose berries, and are presumably bird dispersed. In contrast, Lyonia sect. Lyonia is most diverse in eastern Cuba and north-central Hispaniola, and is wind dispersed. Poitea Vent. comprises 12 species and is restricted to the Greater Antilles and Dominica (see Lavin, 1993); its phylogeny has been investigated through an analysis combining morphological and chloroplast DNA data and suggests that the genus includes two subclades, the P. galegoidesalliance and the P. florida-alliance. The former is centered primarily in southwestern Hispaniola and western Cuba, while the latter occurs mainly in eastern Cuba, north-central Hispaniola, and Puerto Rico. Thus, as discussed by Lavin (1993), these results lend support to the plate tectonic relationships postulated by Rosen (1976, 1985), Pindell and Barrett (1990), Perfit and Williams (1989), and others, i.e., a link between southwestern Hispaniola and western Cuba, and one between north-central Hispaniola, eastern Cuba, and Puerto Rico. A second genus of Fabaceae, Pictetia DC., has been studied by Beyra and Lavin (1999) and the phylogeny constructed for this genus of eight species (based on morphology and nuclear ribosomal DNA sequence data) clearly supports a biogeographical relationship between Puerto Rico, central Hispaniola, and eastern Cuba; the authors explain the presence of the genus in western Cuba and north-central Hispaniola by dispersal events. Interestingly, Puerto Rico is resolved as the sister area to a north-central Hispaniola + eastern Cuba area (Beyra and Lavin, 1999). The legumes of Poitea are explosively dehiscent, while Pictetia has reduced lomented pods.

74

Biogeography of the West Indies: Patterns and Perspectives

Finally, Sabal Adans. (Arecaceae) is diverse in the circum-Caribbean region, contains three Caribbean endemics, and recently has been monographed (Zona, 1990). The morphology-based cladogram of this genus supports an Antillean clade, with a Cuban taxon (S. maritima [Kunth] Burret) placed as the sister taxon to species of Hispaniola and/or Puerto Rico (S. domingensis Beccari and S. causiarum [Cook] Beccari). The genus is, therefore, similar to Lyonia in its support for a close biogeographical relationship between Hispaniola and Puerto Rico. The drupaceous fruits of Sabal have been documented to be dispersed by birds. Common patterns in these Antillean genera include the presence of major centers of endemism in eastern Cuba, western Cuba, southwestern Hispaniola, and north-central Hispaniola. Puerto Rico also contains endemics, and this area is biogeographically linked with north-central Hispaniola. These taxa are absent or only poorly represented in Jamaica. Some biological support is seen for the geological association between eastern Cuba and north-central Hispaniola and between western Cuba and southwestern Cuba, although long-distance dispersal events clearly have occurred and tend to obscure this pattern. It does seem likely that Tertiary Antillean vicariance events have had a major influence on the present-day geographical distributions of these taxa. Finally, Lyonia is seemingly unique in its support of a close relationship of north-central with southwestern Hispaniola and of eastern with western Cuba, suggesting possibly that longer-distance dispersal events (followed by speciation) have been relatively more common in this genus than in the others studied. Additional studies of genera showing a high degree of endemism in Cuba, Hispaniola, and Puerto Rico clearly are needed, and will undoubtedly improve our understanding of the biogeography of this complex region.

LITERATURE CITED Beyra, A. B. and M. Lavin. 1999. Monograph of Pictetia (Leguminosae-Papilionoideae) and review of the Aeschynomeneae. Systematic Botany Monographs 56:1–93. Brooks, D. R. 1981. Hennig’s parasitological method: a proposed solution. Systematic Zoology 30:229–249. Brooks, D. R. 1985. Historical ecology: a new approach to studying the evolution of ecological associations. Annals of the Missouri Botanical Garden 72:660–680. Brooks, D. R. and D. A. McLennan. 1991. Phylogeny, Ecology, and Behavior: A Research Program in Comparative Biology. University of Chicago Press, Chicago. Buskirk, R. E. 1985. Zoogeographic patterns and tectonic history of Jamaica and the northern Caribbean. Journal of Biogeography 12:445–461. Farris, J. S. 1988. Hennig86 reference, version 1.5. Published by the author, Port Jefferson Station, New York. Forey, P. L., C. J. Humphries, I. L. Kitching, R. W. Scotland, D. J. Siebert, and D. M. Williams. 1992. Cladistics: A Practical Course in Systematics. Clarendon Press, Oxford. Howard, R. A. 1973. The vegetation of the Antilles. Pp. 1–28 in Graham, A. (ed.). Vegetation and Vegetational History of Northern Latin America. Elsevier Scientific, New York. Judd, W. S. 1979. Generic relationships in the Andromedeae (Ericaceae). Journal of the Arnold Arboretum 60:477–503. Judd, W. S. 1981. A monograph of Lyonia (Ericaceae). Journal of the Arnold Arboretum 62:63–209, 315–436. Judd, W. S. 1982. A taxonomic revision of Pieris (Ericaceae). Journal of the Arnold Arboretum 63:103–144. Judd, W. S. 1984. A taxonomic revision of the American species of Agarista (Ericaceae). Journal of the Arnold Arboretum 65:255–342. Judd, W. S. 1986. A taxonomic revision of Craibiodendron (Ericaceae). Journal of the Arnold Arboretum 67:441–469. Judd, W. S. 1990. A new variety of Lyonia (Ericaceae) from Puerto Rico. Journal of the Arnold Arboretum 71:129–133. Judd, W. S. 1995a. 13. Lyonia Nuttall. Pp. 222–294 in Luteyn, J. L. (ed.). Ericaceae Part II — The SuperiorOvaried Genera. Flora Neotropica Monograph 66, New York Botanical Garden, Bronx, New York. Judd, W. S. 1995b. 14. Agarista D. Don ex G. Don. Pp. 295–344 in Luteyn, J. L. (ed.). Ericaceae Part II — The Superior-Ovaried Genera. Flora Neotropica Monograph 66, New York Botanical Garden, Bronx, New York.

Phylogeny and Biogeography of Lyonia sect. Lyonia (Ericaceae)

75

Judd, W. S. 1995c. 15. Pieris D. Don. Pp. 345–350 in Luteyn, J. L. (ed.). Ericaceae Part II — The SuperiorOvaried Genera. Flora Neotropica Monograph 66, New York Botanical Garden, Bronx, New York. Judd, W. S. and P. M. Hermann. 1990. Circumscription of Agarista boliviensis (Ericaceae). Sida 14:263–266. Kron, K. A. and W. S. Judd. 1997. Systematics of the Lyonia group (Andromedeae, Ericaceae) and the use of species as terminals in higher-level cladistic analyses. Systematic Botany 22:479–492. Kron, K. A. and W. S. Judd. 1999. Phylogenetic analyses of Andromedeae (Ericaceae subfam. Vaccinioideae). American Journal of Botany 86:1290–1300. Lavin, M. 1993. Biogeography and systematics of Poitea (Leguminosae). Systematic Botany Monographs 37:1–87. Malfait, B. T. and M. G. Dinkleman. 1972. Circum-Caribbean tectonic and igneous activity and the evolution of the Caribbean plate. Bulletin of the Geological Society of America 83:251–272. Page, R. D. M. and C. Lydeard. 1994. Towards a cladistic biogeography of the Caribbean. Cladistics 10:21–41. Pindell, J. L. and S. F. Barrett. 1990. Geological evolution of the Caribbean: a plate tectonic perspective. Pp. 450–432 in Dengo, G. and J. E. Case (eds.). The Geology of North America: The Caribbean Region, Vol. H. Geological Society of America, Boulder, Colorado. Rosen, D. E. 1976. A vicariance model of Caribbean biogeography. Systematic Zoology 24:431–464. Rosen, D. E. 1985. Geological hierarchies and biogeographic congruence in the Caribbean. Annals of the Missouri Botanical Garden 72:636–659. Skean, J. D., Jr. 1993. Monograph of Mecranium (Melastomataceae: Miconieae). Systematic Botany Monographs 39:1–116. Sykes, L. R., W. R. McCann, and A. L. Kafka. 1982. Motion of Caribbean plate during last 7 million years and implications for earlier Cenozoic movements. Journal of Geophysics Research 87:10656–10676. Wiley, E. O., D. J. Siegel-Causey, D. R. Brooks, and V. A. Funk. 1991. The Compleat Cladist: A Primer of Phylogenetic Procedures. Museum of Natural History, University of Kansas, Lawrence. Zandee, M. and M. C. Roos. 1987. Component-compatibility in historical biogeography. Cladistics 3:305–332. Zona, S. 1990. A monograph of Sabal (Arecaceae: Coryphoideae). Aliso 12:583–666.

of Endemism and 6 Patterns Biogeography of Cuban Insects Julio A. Genaro and Ana E. Tejuca Abstract — The taxonomy and biogeography of insects are poorly known for most islands of the West Indies with the exception of Cuba (8,316 species) and Puerto Rico (5,066 species). It is estimated that if all insects on Cuba could be documented that the total would be close to 10,000 species. Some groups are better known than others, and it is only in these well-known groups that real levels of endemism can be calculated. In some groups endemism is very high, such as stick insects (92.8%), mutillids (90%), Cercopoidea (82%), and Trichoptera (81%). In other groups it is surprisingly low, such as dryinids (0%), agromizids (3.8%), and mosquitoes (5.9%). Large well-known groups tend to have levels of endemism between 40 and 60%, for example, butterflies (39.9%), ants (43.6%), and bees (47.3%). Cockroaches have an endemism level of 63.5%. These data do not provide complete insight into the patterns of dispersal of insects into Cuba or between major mountain ranges or offshore archipelagos. The very different levels of endemism between the various groups of insects suggest that insects colonized Cuba in a variety of ways. Additional studies on the distribution and systematics of Cuban insects are important to help us more accurately understand biogeographical patterns of Cuban insects and how the insect fauna of Cuba relates to other islands in the Greater Antilles.

INTRODUCTION Because of its geological history and geographical location as an island, Cuba has few mammal species and few vertebrates. The highest level of species abundance (and diversity) on Cuba and its associated archipelagos is represented by invertebrates (Table 1), mainly insects. While the status of insects is poorly known in most areas of the Antilles, there is an important trend toward inventorying biodiversity to better understand their natural history and the biological potential of each habitat and biogeographical region. This is important because of the rapid loss of habitats in the region, and because some species will be lost before they are ever discovered. It may not be possible ever to know the true biodiversity of insects in Cuba. The number of insects in Cuba has been estimated from as high as 25,000 species (Berovides, 1988), downward to 17,000 (Aguayo, 1951; Alvarez Conde, 1958) and 12,000 to 15,000 species (Ferrás et al., 1995). However, they have been accurately quantified only twice. Vales et al. (1992) documented 6,384 species, although this study only evaluated 10 insect orders, and is therefore not a complete record. In the study “Proyecto Pais” (Vales et al., 1998) insects were treated in more detail resulting in a figure of 7,831 species in 29 orders. However, in our opinion, even this higher figure is not a complete picture of the total number of insects on the island.

DISCUSSION This chapter provides an estimate of the number of Cuban insect species obtained from publications that list or catalog taxa. These listings have been further updated by adding new records, new species, personal communication from various specialists, and data from specimens in collections but not yet in the public record (i.e., published). We present these results as a way of integrating scattered information and laying the foundation for future studies. The total world number of species (Table 2) in our survey was obtained from Strefferud (1952), Borror and White (1970), and Hogue (1993). Based on the data available to us there are 8,316 insect 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

77

78

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 Quantitative Biodiversity of the Main Taxa of Cuban Terrestrial Invertebrates, Excluding Insects Taxa Mollusca (land snails) Annelida (earthworms) Arachnida (spiders, scorpions, mites, ticks, harvest-bugs, and others) Decapoda (land crabs) Isopoda (sowbugs) Chilopoda (centipedes) Diplopoda (millipedes) Pauropoda (pauropods) Symphyla (symphylans) Trematoda (flatworms) Nematoda (roundworms)

Number of Genera

Number of Species

159 14

1,419 21

614 17 37 17 36 2 2 100 208

1,350 33 68 43 90 2 3 160 525

TABLE 2 Number of Families, Genera, and Species of Cuban Insects, and World Number of Species According to the Order Insect Order

Families

Genera

Species

World Species

Protura Diplura Collembola Thysanura Ephemeroptera Odonata Orthoptera Dictyoptera (Blattaria) Mantodea Phasmatodea Dermaptera Isoptera Embiidina Psocoptera Zoraptera Mallophaga Anoplura Heteroptera Thysanoptera Neuroptera Megaloptera Trichoptera Diptera Lepidoptera Siphonaptera Coleoptera Strepsiptera Hymenoptera

1 3 13 3 6 7 4 4 1 2 5 3 3 16 1 4 2 36 4 9 1 12 64 65 2 91 4 49

1 7 63 8 15 41 72 33 4 8 11 14 4 19 1 19 5 323 26 28 1 26 410 762 5 964 6 474

1 19 110 10 37 80 122 81 4 14 19 31 4 28 1 39 5 603 61 74 1 90 984 1,539 6 2,615 7 1,069

500 800 6,000 750 2,139 4,950 14,500 4,000 1,500 2,500 1,100 2,100 150 1,100 22 2,675 250 23,000 4,500 4,670 250 7,000 87,000 112,000 1,100 300,000 300 105,000

Patterns of Endemism and Biogeography of Cuban Insects

79

Coleoptera Lepidoptera Hymenoptera Diptera Homoptera Heteroptera

7.2% 7.9%

31.5%

11.8%

12.9% 18.5%

FIGURE 1 Percentages of the main Cuban insect orders.

species in Cuba belonging to 29 orders (Table 2). The largest number of species were represented by beetles, butterflies, Hymenoptera, and flies (Table 2; Aguayo, 1951; and Figure 1). We believe that the estimates for the number of Cuban insects proposed by other authors (see above) are too high. In our opinion, the maximum number of species on Cuba should be around 10,000. It is important to continue to collect insects in Cuba, and elsewhere in the West Indies, before comprehensive biogeographical comparisons will be possible. Increased financial and logistical support will be necessary to put these expeditions in the field to collect insects and to be able to curate and fully document the level of insect biodiversity in Cuba. The latter phase would require the participation of many specialists in various insect groups who are not always available because of a lack of training in the taxonomy of certain poorly known groups or because of other commitments. Many areas of Cuba are very remote and rugged, which make such faunal surveys major undertakings. The methodology of collecting insects is another major consideration. Collecting has been carried out in the past mainly by traditional methods such as insect nets. Only recently have modern insect traps such as the Malaise trap, pan traps (yellow plates), and nocturnal black-light traps been utilized. Most insect surveys have been carried out in daylight hours while we now know that night collecting offers a completely different scenario. Most new taxa are mainly small-size insects that can be found even in places disturbed by humans. The best places in Cuba for identifying new insects is deep into rugged mountain ranges, on offshore keys and islands, and in wetlands such as the Zapata Swamp. The taxonomy and distribution of Cuban insects is well known in comparison with other regions of the West Indies and South America. Nevertheless, problems arise in Cuba when trying to apply this knowledge to ecological and ethological studies or when incorporating the data into management plans focused on the conservation of ecosystems and fauna. These difficulties arise from a lack of good reference collections and taxonomic specialists, as well as scattered literature, difficult access to old publications, the deposition of holotypes in foreign institutions, and a lack of practical articles allowing for species identification through keys or field guides.

80

Biogeography of the West Indies: Patterns and Perspectives

There have been several attempts to explain how invertebrates dispersed to the Antilles (Darlington, 1938; Liebherr, 1988a, 1988b) and to present explanations for distribution patterns of invertebrates within the West Indies (Darlington, 1937; Fontenla, 1992, 1994; Fontenla and Cruz, 1992). The depth of systematic knowledge varies depending on the taxonomic group and some groups are very poorly known. Zoogeographic analyses cannot be undertaken in many groups until a more complete systematic review has been completed. The taxonomic groups that have been best studied are butterflies (Lepidoptera) (Scott, 1972; Brown, 1978; Fontenla and de la Cruz, 1986, 1992; Miller and Miller, 1989; Fontenla, 1992); beetles (Coleoptera) (Darlington, 1943, 1971; Matthews, 1966; Liebherr, 1988b; Browne and Peck, 1996); dragonflies and damselflies (Odonata) (Paulson, 1982); ants (Wilson, 1988; Fontenla, 1994); bees (Hymenoptera) (Michener, 1979; Eickwort, 1988), and to a lesser extent, mayflies (Ephemeroptera) (Edmunds, 1982), caddisflies (Trichoptera), bugs (Heteroptera), and flies (Diptera) (Liebherr, 1988b). In the West Indies, geographical isolation, wide variety of soils, and differences in altitude and climate all have combined to account for high levels of endemism in land organisms. The analysis of endemism is difficult in insects because no single publication or source integrates the systematic knowledge and geographical distributions of all species. Some taxa and groups have been better studied than others, however, and it is possible to estimate endemism in the following groups: cockroaches (63.5%) (Gutiérrez, 1995); mosquitoes (5.9%) (González and Rodríguez, 1977); sirphids (30.6%) (Garcés and Rodríguez, 1998); agromizids (3.8%) (Garcés, 1998); Odonata (62%) (C. Naranjo, personal communication); stick insects (92.8%) (Moxey, 1972); Trichoptera (81.1%) (Botosaneau, 1979; 1980); Dermaptera (15.8%) (Brindle, 1971); meloids (42.8%) (Genaro, 1996); tiger beetles (40%) (P. Valdés, personal communication); bruquids (22.2%) (Alvarez and Kingsolver, 1997); Psocoptera (52.5%); Hymenoptera, dryinids (0%), scoliids (20%); tiphiids (62.5%); ants (43.6%) (Fontenla, 1997); mutillids (90%); bees (47.3%); Cicadoidea (70%); Membracidae (63%); Cercopoidea (82%); Kinnaridae (75%) (Ramos, 1988); mirids (17%) (Hernández and Stonedahl, 1997); ligaeids (27%) (Slater, 1988); butterflies (39.9%) (Smith et al., 1994). The present status of systematic knowledge allows only for an estimate of overall endemism. In many taxa is very high, while in others it is lower, and it may be zero. On average, endemism ranges between 40 and 60%. Insect groups have different dispersion patterns; even within an order there are families with different degrees of vagileness. Fauna has been shaped by the arrival of species from several parts of the world at different geological times, and following their arrival in Cuba these species adapted and evolved under insularity conditions. This flow of insects to and from the West Indies continues. The Antilles are composed of many islands and keys dissected by open water. Air currents, especially powerful forces such as hurricanes, have played an important role in the dispersion of insects. Introductions, both accidental and intentional, have also played a role in the dispersal of insects. The dragonfly Crocothemis servilia (Libellulidae) from Asia was accidentally introduced in Florida, where it is common, and arrived recently to Cuba (Flint, 1996). The African beetle Onthophagus gazella (Scarabaeidae) was introduced on purpose in the United States and is now common in the north coast of Cuba (R. B. Woodruff, personal communication). Examples of dispersal events in the opposite direction include several species of butterflies that have become established in Florida from the West Indies within the last hundred years (Scott, 1972). These examples demonstrate that faunal exchange in the West Indies is a dynamic process. We believe that many more examples will appear if in-depth studies are carried out in Cuba and the West Indies in general. The senior author (J.A.G.) is finding many species of Aculeata (Hymenoptera) in Hispaniola that were previously thought to be exclusive to Cuba. The knowledge of insect taxonomy and systematics varies from island to island in the West Indies and has been integrated for only two islands, Puerto Rico and Cuba. In Puerto Rico, the smaller island, 5,066 species were recently quantified (Maldonado, 1996). In Jamaica, there is not a reliable count of all species, although Farr (1984) provides numbers for several groups. The most critical example of a poorly known fauna is that of Hispaniola. A little smaller than Cuba, this

Patterns of Endemism and Biogeography of Cuban Insects

81

island has both more diverse habitats and the least well-known insect fauna. What is known indicates that Hispaniolan insects have a high level of endemism, and are most similar to the insect fauna of eastern Cuba.

CONCLUSIONS The insect fauna of Cuba has been well enough documented to project that about 10,000 species occur on the island. The levels of endemism vary remarkably between the various insect groups, ranging from over 90% (mutillids and stick insects) to as low as zero (dryinids). The reasons for this high level of variability are poorly understood, and await data from more complete studies of the life history and ecology of Cuban insects. Such studies should be given a high priority. The average level of endemism is close to 50%, with endemism levels for most well-known groups in the 40 to 60% range. While the insects of Cuba are relatively well known, as are the insects of Puerto Rico, insects are poorly documented and little studied in other areas of the West Indies, with major lacunae in Hispaniola and Jamaica. Based on the preliminary data available to us, it appears that the closest relatives of insects from eastern Cuba are found in Hispaniola.

ACKNOWLEDGMENTS We thank P. Alayo (moths and other insect orders); M. Diaz (Collembola); G. Garcés and D. Rodríguez (Diptera); H. Grillo (Heteroptera); E. Gutiérez (Blattaria); C. Moxey (Phasmatodea); C. Naranjo (Odonata, Ephemeroptera); I. Fernández and S. B. Peck (Coleoptera); E. Portuondo (Hymenoptera Parasitica); R. Rodríguez-Léon (Homoptera); and A. Ruiz (Ortohoptera). For information on other invertebrata taxa, we thank G. Alayón, A. Avila, N. Cuervo, L. F. Armas, W. B. Muchmore, and A. Pérez (Arachnida); C. Rodríguez (Annelida); A. Juarrero (Decapoda, Isopoda); A. Pérez-Asso (Diplopoda, Chilopoda), J. F. Milera, L. Fernández, and A. Lomba (Gastropoda). P. Alayo provided important literature. S. B. Peck offered preliminary data that complemented our observations. We also thank G. Alayón, J. L. Fontenla, E. Gutiérez, and G. Silva for their suggestions during the critical reading of the manuscript.

LITERATURE CITED Aguayo, C. G. 1951. Los orígenes de la fauna cubana. Annales de la Academia de Ciencias, Habana 88:1–23. Alvarez Conde, J. 1958. Historia de la zoología en Cuba. Publicación de la Junta Nacional de Arqueología y de Etnologia Ediciones Lex, La Habana. Alvarez, D. and J. M. Kingsolver. 1997. A preliminary list of the Bruchidae (Coleoptera) of Cuba. Entomology News 108:215–221. Berovides, V. 1988. Orden y diversidad en el mundo viviente. Editorial de Ciencia y Técnica, La Habana. Borror, D. J. and R. E. White. 1970. A Field Guide to the Insects of America North of Mexico. Houghton Mifflin, Boston. Botosaneanu, L. 1979. The caddis-flies (Trichoptera) of Cuba and of Isla de Pinos: a synthesis. Studies of the Fauna of Curaçao and Other Caribbean Islands 59:33–62. Botosaneanu, L. 1980. Trichoptères adultes de Cuba collectés par les zoologistes cubains (Trichoptera). Mitteilungen der Münchner Entomologischen Gesellschaft 69:91–116. Brindle, A. 1971. The Dermaptera of the Caribbean. Studies of the Fauna of Curaçao and Other Caribbean Islands 131:1–75. Brown, F. M. 1978. The origins of the West Indian butterfly fauna. Pp. 5–30 in Zoogeography in the Caribbean. Journal of the Academy of Natural Sciences of Philadelphia, Special Publication 13. Browne, J. and S. B. Peck. 1996. The long-horned beetles of south Florida (Cerambicidae: Coleoptera): biogeography and relationships with the Bahama Islands and Cuba. Canadian Journal of Zoology 74:2154–2169.

82

Biogeography of the West Indies: Patterns and Perspectives

Darlington, P. J., Jr. 1937. West Indian Carabidae. III: new species and records from Cuba, with a brief discussion of the mountain fauna. Memorias de la Sociedad Cubana de Historia Natural Felipe Poey 11:115–136. Darlington, P. J., Jr. 1938. The origin of the fauna of the Greater Antilles, with a discussion of dispersal of animals over water and through the air. Quarterly Review of Biology 13:274–300. Darlington, P. J., Jr. 1943. Carabidae of mountains and islands: data on the evolution of isolated faunas and on atrophy of wings. Ecological Monographs 13:37–61. Darlington, P. J., Jr. 1971. Carabidae on tropical islands, especially the West Indies. Biotropica 2:7–15. Edmund, J. F., Jr. 1982. Ephemeroptera. Pp. 242–248 in Hurlbert, S. H. and A. Villalobos-Figueroa (eds.). Aquatic Biota of Mexico, Central America and the West Indies. San Diego State University, San Diego, California. Eickwort, G. C. 1988. Distribution patterns and biology of West Indian sweat bees (Hymenoptera: Halictidae). Pp. 231–253 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. Farr, T. 1984. Land animals of Jamaica; origins and endemism. Jamaica Journal 17:3848. Flint, O. S., Jr. 1996. The Odonata of Cuba, with a report on a recent collection and checklist of the Cuban species. Cocuyo 5:17–20. Fontenla, J. L. 1992. Biogeografía ecológica de las mariposas diurnas cubanas. Patrones generales. Poeyana 427:1–30. Fontenla, J. L. 1994. Biogeografía de Macromischa (Hymenoptera: Formicidae) en Cuba. Avicennia 1:19–29. Fontenla, J. L. 1997. Lista preliminar de las hormigas de Cuba (Hymenoprera: Formicidae). Cocuyo 6:18–21. Fontenla, J. L. and J. de la Cruz. 1986. Análisis zoogeográfico de las mariposas antillanas (Lepidoptera: Rhopalocera) a nivel subespecífico. Ciencias Biológicas 15:107–122. Fontenla, J. L. and J. de la Cruz. 1992. Consideraciones biogeográficas sobre las mariposas endémicas de Cuba. Poeyana 426:1–34. Garcés, G. 1998. Lista de los agromizidos de Cuba (Diptera: Agroniyzidae). Cocuyo 7:5–7. Garcés, G. and D. Rodríguez. 1998. Lista de los sírfidos de Cuba (Diptera: Syrphidae). Cocuyo 7:7–8. Genaro, J. A. 1996. Resumen del Conocimiento sobre los meloidos de Cuba (Insecta: Coleoptera). Caribbean Journal of Sciences 32:382–386. González Broche, R. and J. Rodríguez. 1997. Lista actualizada de los mosquitos de Cuba (Diptera: Culicidae). Cocuyo 6:17–18. Gutiérrez, E. 1995. Annotated checklist of Cuban cockroaches. Transactions of the American Entomological Society 121:65–84. Hernández, L. M. and G. M. Stonedahl. 1997. Lista anotada de los míridos de Cuba (Insecta: Heteroptera). Cocuyo 6:21–23. Hogue, C. L. 1993. Latin American Insects and Entomology. University of California Press, Berkeley. Liebherr, J. K. 1988a. General patterns in West Indian insects and graphical biogeographic analysis of some circum-Caribbean Platynus beetles (Carabidae). Systematic Zoology 37:385–409. Liebherr, J. K. (ed.). 1988b. Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. Maldonado, J. 1996. The status of insect alpha taxonomy in Puerto Rico after the scientific survey. Pp. 201–216 in Figueroa, J. C. (ed.). The Scientific Survey of Puerto Rico and the Virgin Islands. Annals of the New York Academy of Sciences 776. Matthews, E. G. 1966. A taxonomic and zoogeographic survey of the Scarabaeidae of the Antilles (Coleoptera: Scarabaeidae). Memoirs of the American Entomological Society 21:1–134. Michener, C. D. 1979. Biogeography of the bees. Annals of the Missouri Botanical Garden 66:277–347. Miller, L. D. and J. Y. Miller. 1989. The biogeography of West Indian butterflies (Lepidoprera: Papilionoidea, Hesperiodea) a vicariance model. Pp. 229–262 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Moxey, C. F. 1972. The Stick-Insects (Phasmatodea) of the West Indies: Their Systematics and Biology. Ph.D. thesis, Harvard University, Cambridge, Massachusetts. Paulson, D. R. 1982. Odonata. Pp. 249–277 in Hurlbert, S. H. and A. Villalobos (eds.). Aquatic Biota of Mexico, Central America and the West Indies. San Diego State University, San Diego, California. Ramos, J. A. 1988. Zoogeography of the Auchenorrhynchous Homoptera of the Greater Antilles (Hemiptera). Pp. 61–70 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. Scott, J. A. 1972. Biogeography of the Antillean butterflies. Biotropica 4:32–45.

Patterns of Endemism and Biogeography of Cuban Insects

83

Slater, J. A. 1988. Zoogeography of West Indies Lygacidae (Hemiptera). Pp. 38–60 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. Smith, D. S., L. D. Miller, and J. Miller. 1994. The Butterflies of the West Indies and South Florida. Oxford University Press, New York. Stefferud, A. (ed.). 1952. Insects: The Yearbook of Agriculture. U.S. Government Printing Office, Washington, D.C. Vales, M. A., L. Montes, and R. Alayo. 1992. Estado del conocimiento de la biodiversidad en Cuba. Pp. 239–249 in Halffter, G. (ed.). La diversidad biológica de Iberoamérica. Acta zoológica Mexicana. Vales, M. A., A. Alvarez de Zayas, l. Montes, and A. Avila (eds.). 1998. Estudio nacional sobre la diversidad biológica en la República de Cuba. Editora CESYTA, Madrid. Wilson, E. O. 1988. Biogeography of the West Indian ants (Hymenoptera: Formicidae). Pp. 214–230 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York.

in the Biogeography 7 Patterns of West Indian Ticks Jorge O. de la Cruz Abstract — The tick fauna of the West Indies (divided into Cuba, the Greater Antilles, and the Lesser Antilles) is compared with some continental areas (Venezuela, Panama, Peru, and Madagascar). Taxonomic inventories and host and structural niche preferences are used as zoogeographical characters. On the islands the dominants are Argasid ticks, parasites of birds, bats, and reptiles, cavernicolous and lapidicolous. In the continental conditions the dominants are Ixodid ticks, macromastophiles of open fields. The colonization of the West Indies follows three main patterns: (1) a Central America–West Indies–South America route; (2) a North America–West Indies route; and (3) an anthropogenic introduction. A fourth group with cosmopolitan distribution is not included in any of the three patterns because of lack of information.

INTRODUCTION The West Indies is a fertile ground for biogeographical research. It is amazing how much work has been devoted to study of the faunal relationships of this area. Even more amazing is the lack of agreement among the various interpretations. Although the Cuban fauna has been the best studied in the West Indies, even this database is not complete enough. More than half of the vertebrate fauna (excluding birds) has been described only in the last 30 years and invertebrates are far less well known than vertebrates. Liebherr (1988) discusses insects as a biogeographical data source and their advantage over vertebrates, but he says nothing about other invertebrates such as parasitic mites, especially ticks. Parasites have special needs that make them more complicated to collect and to analyze. They have more complex ecological constraints than most invertebrates. They not only need very specific habitats to be successful but also the presence of the “right host(s)” (an adequate systematic group within the limits of size, behavior, etc.). Not only tick systematics but also other characteristics of their natural history are clues to parasite biogeographical relationships. For these reasons, this chapter is divided into five main sections: (1) list of West Indies species of ticks; (2) West Indies tick faunal characteristics and relationships; (3) ecological biogeography; (4) historical biogeography; and (5) conclusion. Two of these terms (ecological and historical biogeography), used by Silva Taboada (1979) in his study of Cuban bats, are used differently in the following sections. A more complete discussion of the ticks of Cuba can be found in Cruz (1987). In the present discussion I provide additional information to my 1987 paper (my dissertation) and incorporate a final analysis of two other publications (Cruz, 1978, 1986).

MATERIALS AND METHODS This chapter reviews the literature of the past 40 years on the distribution and systematics of ticks. The publications on biogeography include the ticks of Panama (Fairchild et al., 1966), Venezuela (Jones et al., 1972), Cuba (Cerny, 1969), Puerto Rico (Maldonado-Carriles and Medina-Gaud, 1977), the Lesser Antilles (Morel, 1966, 1967; Kohls, 1969), Peru (Need et al., 1991), and Madagascar (Uilemberg et al., 1979). The systematic reviews include the American Ornithodorinae (Kohls et al., 1965, 1969; Jones et al., 1972) and Argasinae (Kohls et al., 1970). The list is not exhaustive in 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

85

86

Biogeography of the West Indies: Patterns and Perspectives

relation to the hosts or distribution outside the Western Hemisphere. A full list of hosts is not necessary because ticks are not really species specific. A general list is suitable to convey the idea of the relationship for biogeographical purposes. This constraint also applies to the geographical information in the sense that detailed information of the distribution of introduced species is meaningless to interpretations about patterns of West Indian biogeography. In the analysis of ecological factors two main groups were selected: the host-group specificity and the structural niche.

HOST-GROUP SPECIFICITY In this chapter, terminology of the host-group specificity varies with the host as follows. Ticks that are parasites on amphibians and reptiles are called herpetophiles; on birds they are called ornithophiles; on small mammals (rodents, insectivores, etc.) they are called micromastophiles; on large mammals they are called macromastophiles; and on bats they are called chiropterophiles. Some ticks are parasites on several kinds of hosts and so in some cases there are not clear distinctions. I follow Hoogstraal (1985) and consider the adult ticks hosts as the main ones because immature stages in almost all groups show less host specificity.

STRUCTURAL NICHE The structural niche, in this context, is the place were the ticks hide outside the host to develop some essential biological changes or process, like molts between the ontogenic stages or egg deposition. There are two main groups of structural niche for ticks: the open field and the nest. Open Field The first niche is typical for the ixodid ticks, which typically are parasites on nomadic hosts without nesting behavior. The tick biological cycle is more dependent on the hosts. The tick remains on the hosts for long periods of time. The more derivative groups have fewer stages in which the ticks can leave hosts to molt, like the so-called “one host ticks” from the genus Boophilus. In general, the ixodid ticks also have fewer developmental stages than argasids, which is an advantage for this type of biological cycle. Ixodids have only three life stages: the larval, the nymph, and the adult. There are exceptional argasid ticks, such as the Spinose ear tick, Otobius megnini (Duges, 1884), which can be part of this ecological group. They are an exception in the family because these ticks make all but the last molt (last nymph to adult) on the same host. The adults are autogenic in that they do not need any additional meals and live with the reserves built up during the younger stages. In this way the multistage, multihost argasid becomes a single-host tick. The Nest The second group, the “nest” species are typically members of the Argasidae, which are parasites on sedentary hosts with a den, nest, or place regularly visited during the year were the ticks can hide and wait. Ticks go to the host only for short periods of time to take a blood meal, and then return to their hiding places. The host contact is short and varies from a few minutes in some nymph and adult stages to a few days in the larval stage of some species. The cycles in these ticks present more developmental stages, which typically include one larva stage, two to five nymph stages, and the adult stage. By dividing the morphological changes into several nymphal stages, ticks use fewer energy reserves for each molt; therefore, it is necessary for them to hang onto the host for shorter time periods to be fed. This is an advantage since smaller meals ease the molting process. There are some deviations from these patterns. Some species have immobile first nymphs, as in some species of the subgenus Alectorobius, the autogenic species of the genus Antricola, or the coprophagous

Patterns in the Biogeography of West Indian Ticks

87

Parantricola. The deviations are in relation to the capability of the larvae to accumulate the reserves needed for the subsequent stages and their efficiency to use these reserves at the different stages. All these specialized forms have big larvae, a shorter nymphal life stage, and, in some cases, lower reproductive rates. The Antricola species have short nymph stages with short cycles and the adults probably have very low reproductive rates. The coprophagous species, P. marginatus, has meals with lower energetic contents, and thus has longer cycles, probably with more stages and larger adults. The nature of the links of these groups with their hosts and the host habitats made them more diverse and specialized. It is almost impossible to recognize differences among the “open field” species that are very clear on the “nest” species. I recognize three groups, one of which has two subgroups. The Arboricolous Their hosts use trees as nests or dens. The ticks hide under the bark or in crevices of the wood. In this group are species that are parasites on bats (that use hollow trees as refuge), on birds that nest in trees and species of the genus Argas, which are parasites on poultry and can be found on the trees used by hens to roost. The Lapidicolous They are parasites of birds, mammals, and reptiles that nest or rest on rocks or in small caves and crevices. The most common ticks of this category are Ornithodoros capensis and O. denmarki, which are parasites of terns and gulls, and O. cyclurae and O. elongatus, which are parasites of iguanas of the genus Cyclura. The Cavernicolous They are almost exclusively parasites of bats, at least in the West Indies. They represent a clearly different ecological group that I split into two catagories: the “psicrocavernicolous” and the “thermocavernicolous.” The psicrocavernicolous ticks inhabit typical caves with high humidity and low temperatures. The thermocavernicolous form a very special group that occupies “hot caves” or the “hot saloons” of some caves of the Western Hemisphere. High temperatures (over 26°C and typically around 32°C) characterize these caves, with near-saturation humidity and dense colonies of bats. These are the conditions in which the genera Antricola and Parantricola and the subgenus Ornithodoros (Subparmatus) develop their highly specialized biology.

THE WEST INDIES TICKS SUPERFAMILY IXODOIDEA MURRAY, 1877 FAMILY ARGASIDAE MURRAY, 1877 Genus Argas Latreille, 1795 Subgenus Persicargas Kaiser, Hoogstraal et Kohls, 1964 A. persicus (Oken, 1818) Hosts: Poultry, humans, Neomorphus geoffroyi(?). Distribution: Almost cosmopolitan, but many records are doubtful, especially with the resurrection of some sibling species. Recent records from Peru, U.S.A., Paraguay, Puerto Rico, Hispaniola, and Cuba. Other records from Panama, Antigua, Barbados, Trinidad, and Martinique need confirmation. Comments: A parasite of domestic poultry, with which the species spread. As the other poultry tick species, A. miniatus, it disappears for years and then can be found by the thousands in one locality, sometimes very far from the last known locality where it was found.

88

Biogeography of the West Indies: Patterns and Perspectives

A. miniatus Koch, 1844 Hosts: Poultry. Distribution: Cuba, Puerto Rico, U.S.A., Mexico and Peru. Other nonconfirmed records report Jamaica, Hispaniola, Antigua, and Martinique. Comments: See A. persicus. A. radiatus Railliet, 1893 Hosts: Poultry (including turkey), also on Coragyps atratus, Cathartes aura, herons, cormorants, warblers, and pigeons, Phalacrocorax auritus floridanus. Distribution: U.S.A., Mexico, and Cuba. Comments: As the other poultry ticks, but this species also parasitizes on wild host. Genus Ornithodoros Koch, 1844 Subgenus Alectorobius Pocock, 1907 O. cyclurae Cruz, 1986 Hosts: Cyclura nubila. Distribution: Cabo Cruz, Granma, Cuba. Comments: Known from only one larva, found on the nose of a Cuban iguana. Close to O. elongatus. O. elongatus Kohls, Sonenshine et Clifford, 1965 Hosts: Probably iguana. Distribution: Hispaniola. Comments: Known, in Miami, only from one larva found in a box from Dominican Republic, with plants and an iguana. This species is close to O. cyclurae. O. denmarki Kohls, Sonenshine et Clifford, 1965 Hosts: Puffinus lehrminieri, Sula leucogaster, Sterna anaethetus, S. fuscata, Anous stolidus. Distribution: Sand Islands, Johnston Atoll, Hawaii, U.S.A. (Florida), Mexico, Trinidad, Martinique, Dominica, Guadeloupe, Jamaica, and Cuba. Comments: A widespread species, distributed by its host, the seabirds. Probably would be found in many other localities, if collected. Studies of the distribution of seabird ornithodorine ticks, a complex of species, are greatly needed. In the West Indies, there are two species, O. denmarki and O. capensis. O. capensis Neumann, 1901 Hosts: Spheniscus demersus, Anous stolidus, Sterna fusca, Diomedea inmutabilis, Actitis macularia, Thalasseus sandvicensis, Ajaia ajaja, Sula leucogaster, Sula sp., Fregata magnificens, Stictocarbo punctatus. Distribution: Russia, Japan, New Zealand, Australia, Marshall Islands, Hawaii, Howland Islands, Sand Islands, Guam, Paget Islands, Cargado Carajos Islands, Chesterfield Islands, South Africa, U.S.A. (coast of Texas), Mexico (Revillagigedo Archipelago), Galapagos Islands, Trinidad, San Martin, Dominica, Jamaica, and Cuba. Comments: See O. denmarki. O. azteci Matheson, 1935 Hosts: Macrotus waterhousei, Artibeus jamaicensis, Brachyphylla nana, Desmodus rotundus, Phyllostomus hastatus, Trachops cyrhosus, Carollia sp., Zygmodon brevicaudata, Glossophaga soricina, G. longirostris, Artibeus sp., Macrophyllum macrophyllum, Carollia perspicillata, Peropyteryx macrotis, P. kapleri, Lonchorhina aurita.

Patterns in the Biogeography of West Indian Ticks

Distribution: Panama, Colombia, Venezuela, Trinidad, Jamaica, Cuba, Mexico, and the Lesser Antilles. Comments: The most common species on Cuban bats. O. brodyi Matheson, 1935 Hosts: Eptesicus fuscus, Myotis nigricans, Desmodus rotundus, Pteronotus parnelli, Carollia perspicillata, Carollia sp., Trachops cirrhosus, Artibeus jamaicensis, Chonopterus auriotus, Lonchorhina aurita, Pteropteryx kapleri, Rhynchonycteris sp., Natalus tumidirostris. Distribution: Mexico, Guatemala, Panama, Colombia, Venezuela, and Cuba. Comments: A species known from the Antilles after only one record from Cuba (Cerny, 1969). O. kelleyi Cooley et Kohls, 1941 Hosts: Eptesicus fuscus, Carollia perspicillata, Carollia sp., Lonchorhina aurita, Antrozous pallidus, Pipistrellus hesperus, Myotis subulatus, humans. Distribution: Canada, U.S.A., Costa Rica, and Cuba. Comments: A Nearctic species, probably distributed to Cuba with the host, E. fuscus. In Costa Rica it was found biting humans (Vargas, 1984). O. dusbabeki Cerny, 1967 Hosts: Eptesicus fuscus, Noctilio leporinus, Molossus molossus, Artibeus jamaicensis. Distribution: Isla de Pinos, Cuba Comments: Known only from the northern region of Isla de Pinos, and a record from Pilon, Mayari, Holguin province, Cuba. I reviewed one of the larva from Mexico (Dusbabek, 1970) and determined it to be an undescribed species from the O. hasei complex, but for sure not O. dusbabeki. O. tadaridae Cerny et Dusbabek, 1967 Hosts: Tadarida laticaudata, Mormoopterus minutus. Distribution: Cuba. Comments: A species very close to O. boliviensis, a South American species, also parasite of molossid bats. O. hasei Schulze, 1935 Hosts: Pteronycterix sp., Carollia sp., C. perspicillata, Uroderma bilobatum, U. magnirostris, Vampyrops helleri, Tonatia sylvicola, Rhogesia minutilla, R. tumida, Phyllostomus hastatus, Artibeus jamaicensis, A. lituratus, Brachyphylla cavernarum, Sturnira lilium, Akodon urichi, Chiroderma salvini, Mormoops megalophylla, Myotis nigricans, M. albescens, M. velifer, Mimon crenulatus, Desmodus rotundus, Neoplatymops mattogrocensis, Eumops sp., Molossus ater, M. bondae, Tadarida gracilis, Noctilio labialis, N. leporinus, Lonchorhina orinocensis, Glossophaga longirostris. Distribution: Venezuela, Brazil, Panama, Costa Rica, Bolivia, Nicaragua, Mexico, Guyana, Surinam, Martinique, Guadeloupe, Barbuda, Trinidad, Guatemala, Peru, Colombia, Dominica, Uruguay, St. Croix, and Hispaniola. Comments: It could be a complex of species (Jones et al., 1972). O. portoricensis Fox, 1947 Hosts: Rats (Rattus spp.), R. norvergicus, Sigmomys alstonis, Nectomys sp., Proechimys guyannensis, P. semispinous, Tamandua longicaudata, Marmosa robinsoni, Marmosa sp., Sylvilagus floridanus, S. braziliensis, Conepatus semistriatus, Zygmodontomys brevicaudata, Dasyprocta fuliginosa, Monodelphis brevicaudata, Mongoose (=Herpestes javanicus?), Artibeus lituratus (?), Iguana sp. (?), lizards (?), humans.

89

90

Biogeography of the West Indies: Patterns and Perspectives

Distribution: Venezuela, Panama, Colombia, Jamaica, Guadeloupe, Virgin Islands, Nicaragua, Suriname, Uruguay, Puerto Rico, Hispaniola, Argentina, Bolivia, Brazil, Paraguay, St. Croix, Guadeloupe, and Trinidad. Comments: I think this species should be present in Cuba. Its absence in collections is due to lack of investigation in the right localities (southeast coastal areas). Ornithodoros sp. (group talaje) Hosts: Eleutherodactylus cooki. Distribution: Puerto Rico. Comments: This record (from Maldonado-Carriles and Medina-Gaud, 1977) was ignored by the authors of the revisions of the New World Ornithodorinae of the last decades. No other record of Ornithodoros being a parasite on frogs is known in the Western Hemisphere. Subgenus Subparmatus Clifford, Kohls et Sonenshine, 1964 O. viguerasi Cooley et Kohls, 1941 Hosts: Phyllonycteris poeyi, Erophylla sezekorni, E. bombifrons, Brachyphylla nana, Pteronotus quadridens, P. macleayi, P. davyi, P. suapuensis, P. rubiginosa, Pteronotus sp., Mormoops megalophylla, M. blainvillei, Eptesicus fuscus, Noctilio leporinus. Distribution: Venezuela, Trinidad, Jamaica, Puerto Rico, Hispaniola, and Cuba. Comments: An important member of the hot caves community. I think all the members of the subgenus Subparmatus are specialists of this habitat, as the genera Antricola and Parantricola. Indeterminate subgenus O. natalinus Cerny et Dusbabek, 1967 Hosts: Natalus lepidus. Distribution: Isla de Pinos, Cuba. Comments: Known only from Cueva del Lago, Isla de Pinos, Cuba. Genus Antricola Cooley et Kohls, 1942 A. silvai Cerny, 1967 Hosts: Phyllonycteris poeyi, Pteronotus quadridens. Distribution: Cuba. Comments: Known only from Cueva de Colon, Caguanes, Sancti Spiritus province, Cuba. All other reviewed records of this species are misidentifications (Cruz, 1976, 1978; Cruz and Estrada-Pena, 1995). Species from this genus are another important component of the hot caves community, together with O. viguerasi, and Parantricola marginatus. A. granai Cruz, 1973 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva del Abono, Punta Judas, Sancti Spiritus province, Cuba. A. habanensis Cruz, 1976 Hosts: Phyllonycteris poeyi, Mormoops blainvillei. Distribution: Cuba. Comments: Known from Cueva del Mudo, Catalina de Guines, and Cueva de los Murcielagos, Santa Cruz del Norte, both from Havana province, Cuba.

Patterns in the Biogeography of West Indian Ticks

A. naomiae Cruz, 1978 Hosts: Phyllonycteris poeyi. Distribution: Cuba. Comments: Known from Cueva de Santa Catalina, Camarioca, Matanzas province, Cuba. A. martelorum Cruz, 1978 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva de los Murcielagos, Santa Cruz del Norte, Havana province, where it is found together with A. habanensis, the only two sympatric species of Antricola, but this appears to be a consequence of human activities (Cruz, 1978). A. cernyi Cruz, 1978 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva de Castellanos, Rodas, Cienfuegos province, Cuba. A. occidentalis Cruz, 1978 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva de los Majaes, Galalon, San Andres de Caiguanabo, Pinar del Rio province, Cuba. A. centralis Cruz et Estrada-Pena, 1992 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva del Maja, Buenaventura, Remedios, Las Villas province, Cuba. A. armasi Cruz et Estrada-Pena, 1992 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva de la Ventana, Guanahacabibes Peninsula, Pinar del Rio province, Cuba. A. siboney Cruz et Estrada-Pena, 1992 Hosts: Unknown, but certainly bats. Distribution: Cuba. Comments: Known only from Cueva de los Majaes, Siboney, Santiago de Cuba province, Cuba. Genus Parantricola Cerny, 1966 P. marginatus Banks, 1910 Hosts: Phyllonycteris poeyi, Mormoops blainvillei, Pteronotus quadridens, P. macleayi. Distribution: Cuba, Mexico, Puerto Rico, and Hispaniola. Comments: Another important member of the hot caves community, together with the species of the genus Antricola and O. viguerasi. This species was known only from Cuba and Mexico (Hoffmann et al., 1972), but I had the opportunity to review some materials from Cueva Vicenta, Santo Domingo, which became the first record for Hispaniola.

91

92

Biogeography of the West Indies: Patterns and Perspectives

FAMILY IXODIDAE MURRAY, 1877 Genus Ixodes Latreille, 1796 Subgenus Alloixodes Cerny, 1966 I. capromydis Cerny, 1966 Hosts: Capromys pilorides. Distribution: Isla de Pinos, Cuba. Comments: Known only from the southern region of Isla de Pinos, Cuba. Ixodes sp. Host: Parulidae Distribution: Cayo Caiman del Faro, north coast of Cuba. Comments: I determined that some larva from migratory warblers collected in the abovementioned locality by J. de la Cruz, Rafael Abreu, and Naomi Cuervo, in September 1984, was a member of the genus Ixodes, but it was impossible to determine the species. The material was confiscated by the Cuban Security (see Haemaphysalis leporispalustris) and it was impossible to make any other study. Regardless, the only known species of Ixodes from the West Indies is a parasite of rodents, and has a very different morphology. Genus Aponomma Neumann, 1899 A. quadricavum Schulze, 1941 Hosts: Epicrates striatus, E. angulifer, Alsophis cantherigerus. Distribution: Hispaniola and Cuba. Comments: The genus has only two species known in the New World. The second species is Aponomma elaphense Price, 1958, a parasite of Elaphe subocularis from North America (Price, 1958; Anderson et al., 1981; Keirans and Degenhardt, 1985). Genus Amblyomma Koch, 1844 A. dissimile Koch, 1844 Hosts: Reptiles and amphibians, mainly on larger species. There are some doubtful records on mammals. Distribution: Cuba, Hispaniola, Jamaica, Puerto Rico, Barbados, Grenada, St. Lucia, Antigua, Trinidad, Tobago, St. Augustine (Florida and Georgia), U.S.A., Mexico, Guatemala, Belize, Honduras, Costa Rica, Nicaragua, Panama, Colombia, Venezuela, Guyana, Brazil, Paraguay, Argentina. Comments: There might be some confusion between records of this species and those of A. rotundatum. Both species parasite the same hosts and both supposedly were introduced with the toad Bufo marinus in some of the West Indies (Maldonado-Capriles and MedinaGaud, 1977). On the other hand, this tick is recorded from Dominican amber, which shows the presence of the species in the West Indies long before human colonization. In addition, Morel (1967) mixes all records of ticks from reptiles and amphibians from the West Indies, naming them without a real review of the materials, and making it difficult to recognize the identity of many old records. A. albopictum Neumann, 1899 Hosts: Cyclura nubila, Iguana sp., C. cornuta, C. nubila, Leiocephalus carinatus, L. stictigaster, Alsophis cantherigerus, Epicrates angulifer. Distribution: Cuba, Hispaniola, Swan Islands, Cayman Islands. Comments: See comments on A. antillarum.

Patterns in the Biogeography of West Indian Ticks

A. antillarum Kohls, 1969 Hosts: Cyclura pinguis, C. delicatissima, C. carinata. Distribution: Virgin Islands, Dominica, Bahamas (East Caicos) (Keirans, 1985). Comments: Relative to A. albopictum, but with a curious distribution. There is a population A. albopictum in Hispaniola, which is geographically between the Bahamas and the Virgin Islands. How did the species A. antillarum move from one group of islands to the other without interacting with the Hispaniolan population of A. albopictum? It is a problem that needs to be answered in the future. A. torrei Perez, 1934 Hosts: Epicrates angulifer, Cyclura nubila, C. stejnegeri, Leiocephalus macropus, L. carinatus, L. cubensis, Anolis luteogularis, A. sagrai, domestic dog (?). Distribution: Cuba, Puerto Rico, Mona Island, Cayman Islands (Sound and Little Cayman). Comments: A relative of A. arianae. A. arianae Keirans et Garris, 1986 Hosts: Alsophis portoricensis. Distribution: Puerto Rico. Comments: An endemic species from Puerto Rico. It is close to A. torrei, another parasite of reptiles also present in Puerto Rico, but probably develops a different host species preference. A. cruciferum Neumann, 1901 Hosts: Cyclura stejnegeri, C. cornuta. Distribution: Puerto Rico (Mona Island) and Hispaniola. Comments: Another species with some relations with A. torrei. A. rotundatum Koch, 1844 Hosts: Chaeromiscus minor(?), Bufo marinus, other cold-blooded vertebrates, mainly large species. Distribution: Mexico, Guatemala, Panama, Costa Rica, Jamaica, Colombia, Peru, Bolivia, Grenada, Guadeloupe, Surinam, Martinique, Trinidad, Brazil, and Venezuela. Comments: A partenogenetic species, very close to A. dissimile (see comments on this species). A. cajennense Fabricius, 1787 Hosts: Large domestic mammals, humans, and many other species of wild mammals; occasionally on birds. Distribution: Cuba, Jamaica, Trinidad, U.S.A. (Texas and Florida), Mexico, Guatemala, Honduras, Nicaragua, Costa Rica, Panama, Colombia, Venezuela, Guyana, Brazil, Bolivia. Comments: An American species, distributed with the Brazilian Zebu. A. variegatum Fabricius, 1787 Hosts: Cattle. Distribution: Puerto Rico, Antigua, Martinique, Guadeloupe, St. Kitts, French Guyana, Suriname, Venezuela, Nicaragua (Cruz, 1992), Africa. Comments: An African species introduced to America with domestic animals. Genus Anocentor Schulze, 1937 A. nitens Neumann, 1897 Hosts: Horses, mules, cattle. Other records on some wild mammals (Odocoileus virginianus) and accidental on birds.

93

94

Biogeography of the West Indies: Patterns and Perspectives

Distribution: Cuba, Hispaniola, Jamaica, Puerto Rico, Virgin Islands (St. Thomas, St. Croix), St. Kitts, Montserrat, St. Vincent, Dominica, St. Martin, Antigua, Guadeloupe, Martinique, Trinidad, U.S.A. (Texas, Florida), Mexico, Guatemala, Costa Rica, Honduras, Panama, Colombia, Venezuela, Brazil, Bolivia, Argentina. Comments: Another American species distributed with domestic animals. Genus Rhipicephalus Koch, 1844 R. sanguineus Latreille, 1806 Hosts: Domestic dogs, accidental in humans, one record from Rattus norvergicus, and other from Hydrochoerus hydrocoeris. Distribution: Cosmopolitan. Comments: An African species distributed with the domestic dogs. Genus Boophilus Curtice, 1891 B. microplus Canestrini, 1887 Hosts: Cattle, horses, and other large mammals (deer, etc.). Distribution: From the southern states of the U.S.A. (from where it was eradicated), Central and South America, West Indies (eradicated on Puerto Rico), East and South Africa, Madagascar, Australia, and much of the southern half of Asia. Comments: The common cattle ticks have a wide distribution in the West Indies since their hosts are large domestic mammals. B. annulatus Say, 1822 Hosts: Cattle, goats, Odocoileus virginianus. Distribution: Jamaica, Puerto Rico (eradicated in the mid-1940s), Guadeloupe, Africa. Comments: Another African species distributed with domestic cattle, but less successful than B. microplus. Genus Haemaphysalis Koch, 1844 H. leporispalustris Packard, 1889 Hosts: Colinus virginianus, Agelaius phoeniceus, warbler undeterminated, Sylvilagus palustris, S. floridanus, Oryctolagus cuniculus, Dasyprocta sp., Peromyscus sp. Distribution: From Canada to Argentina, but in the West Indies there is only one record from the Lesser Antilles (Kohls, 1969) and now one from Cuba. Comments: A widespread American species, never before reported from the Great Antilles. I collected (with Naomi Cuervo and Rafael Abreu) some immature stages of this tick (together with some Ixodes sp.) from migratory warblers in Cayo Caiman del Faro, Las Villas province, but the material and the records were confiscated by Cuban Security, considered “information sensitive to the enemy” and were never allowed to be published. This record is clearly accidental since migratory birds carried the ticks. Some populations of feral rabbits in Cuba that have never been studied should have a resident tick population.

DISTRIBUTION AND RELATIONSHIPS West Indies tick fauna is represented by 45 species, from 11 genera and two families (Table 1). They should represent nine different distributional patterns (Table 2).

68.57 31.43 100%

12.50 41.67 0.00 41.67 4.17 100%

18.18 9.09 9.09 36.36 9.09 0.00 9.09 9.09 100%

24 11 35

3 10 0 10 1 24

2 1 1 4 1 0 1 1 11

Argasidae Ixodidae Total Argasidae Argas Ornithodoros Otobius Antricola Parantricola Total Ixodidae Ixodes Haemaphysalis Aponomma Amblyomma Anocentor Dermacentor Boophilus Rhipicephalus Total

%

No.

Taxon

Cuba

0 0 1 9 1 0 2 1 14

2 8 0 0 1 11

11 14 25

No.

0.00 0.00 7.14 64.29 7.14 0.00 14.29 7.14 100%

18.18 72.73 0.00 0.00 9.09 100%

44.00 56.00 100%

%

Without Cuba

2 1 1 9 1 0 2 1 17

3 14 0 10 1 28

28 17 45

No.

11.76 5.88 5.88 52.94 5.88 0.00 11.76 5.88 100%

10.71 50.00 0.00 35.71 3.57 100%

62.22 37.78 100%

%

With Cuba

Greater Antilles

0 1 0 4 1 0 1 1 8

2 5 0 0 0 7

7 8 15

No.

0.00 12.50 0.00 50.00 12.50 0.00 12.50 12.50 100%

28.57 71.43 0.00 0.00 0.00 100%

46.67 53.33 100%

%

Lesser Antilles

6 2 0 26 1 0 1 1 37

1 18 1 1 0 21

21 37 58

No.

16.22 5.41 0.00 70.27 2.70 0.00 2.70 2.70 100%

4.76 85.71 4.76 4.76 0.00 100%

36.21 63.79 100%

%

Venezuela

TABLE 1 Comparative Tick Fauna Composition from Different Regions of the World

11 2 0 21 1 3 1 1 40

2 7 0 1 0 10

10 40 50

27.50 5.00 0.00 52.50 2.50 7.50 2.50 2.50 100%

20.00 70.00 0.00 10.00 0.00 100%

20.00 80.00 100%

%

Panama No.

11 1 0 14 1 0 2 1 30

5 7 1 1 0 14

14 30 44

No.

%

36.67 3.33 0.00 46.67 3.33 0.00 6.67 3.33 100%

35.71 50.00 7.14 7.14 0.00 100%

31.82 68.18 100%

Peru

7 13 0 2 0 0 1 1 24

4 3 1 0 0 8

8 24 32

29.17 54.17 0.00 8.33 0.00 0.00 4.17 4.17 100%

50.00 37.50 12.50 0.00 0.00 100%

25.00 75.00 100%

%

Madagascar No.

Patterns in the Biogeography of West Indian Ticks 95

96

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 Distribution of Ticks in the West Indies Greater Antilles Taxon Argasidae Argas persicus A. miniatus A. radiatus Ornithodoros cyclurae O. elongatus O. denmarki O. capensis O. azteci O. brodyi O. kelleyi O. dusbabeki O. tadaridae O. hasei O. portoricensis O. sp. N. talaje O. viguerasi O. natalinus Antricola silvai A. granai A. habanensis A. naomiae A. martelorum A. cernyi A. occidentalis A. centralis A. armasi A. siboney Parantricola marginatus Total Ixodidae Ixodes capromydis Ixodes sp. Haemaphysalis leporispalustris Aponomma quadricavum Amblyomma dissimile A. albopictum A. antillarum A. torrei A. arianae A. cruciferum A. rotumdatum A. cajennense A. variegatum Anocentor nitens Boophilus annulatus B. microplus Rhipicephalus sanguineus Total

Cuba

Without Cuba

With Cuba

Lesser Antilles

+ + + + – + + + + + + + – – – + + + + + + + + + + + + + 24

+ + – – + + + + – – – – + + + + – – – – – – – – – – – + 11

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + 28

+ + +

– – –

+ + +

+ + + + + + + + + + + + + + 14

+ – – + – + – + + 11

Distribution

Structural Niche

Host Affinity

– – – – – + + + – – – – + + – – – – – – – – – – – – – – 5

CO AM NA EN EN CO CO WISA WISA NAA EN EN WISA WISA EN WISA EN EN EN EN EN EN EN EN EN EN EN WISA

A A A L L L L P P P P, A A P, A L L T P T T T T T T T T T T T

O O O H H O O C C C C C C M H C C C C C C C C C C C C C

+ + +

– – +

EN NAA AM

F F F

M Z M

+ + + + + + + + + + + + + + 17

+ – – – – – + + + + + + + 9

CA AM CA CA CA EN EN WISA AM CO AM CO CO CO

F F F F F F F F F F F F F F

H H H H H H H H Z Z Z Z Z Z

Patterns in the Biogeography of West Indian Ticks

97

TABLE 2 (continued) Distribution of Ticks in the West Indies Note: Distribution is designated as cosmopolitan (CO), American species (AM), Caribbeans (CA), North American-Antilleans (NAA), West Indies-South Americans (WISA), and endemics (EN). Structural niche is designated as arboricolous (A), lapidicolous (L), psicrocavernicolous (P), termocavernicolous (T), and open fields (F). Host affinity is denoted by Herpetophile (H), Ornithophile (O), Chiropterophile (C), Micromastophile (M), Macromastophile (Z), Unknown (U).

THE COSMOPOLITANS They include parasites of domestic animals, like Argas persicus, a parasite of poultry, Rhipicephalus sanguineus, the common dog tick, and Boophilus microplus, the common cattle tick, but two species, Ornithodoros capensis and O. denmarki, are distributed by their natural hosts, terns and gulls. Many records of A. persicus have been confused with some of the numerous sibling species of the subgenus Persicargas (Kohls et al., 1970); therefore, it is impossible to give a precise distribution of any of the Persicargas species without an exhaustive review of all the old records. The tick parasites of domestic mammals are very well distributed all over the Antilles, as are their hosts. The two cosmopolitan species of Ornithodoros have been reported from Jamaica and Trinidad, but they should be found on other islands of the Antilles. Both can be considered circumtropical species (Kohls et al., 1965). Another species, mentioned by Maldonado-Carriles and Medina-Gaud (1977) as B. annulatus, from cattle in Puerto Rico, was eradicated in the mid-1940s. It is not clear to me that they are not referring to the more common species in the West Indies, B. microplus. The same species was mentioned from Jamaica.

THE AMERICAN SPECIES They are represented by four species: the common horse tick, Anocentor nitens, the Lone Star tick, Amblyomma cajennense, a cattle tick species, the toad tick, A. dissimile, and the hare tick, Haemaphysalis leporispalustris. The first two were distributed by domestic animals, as was the case in the cosmopolitan species. Amblyomma dissimile looks like an old successful species distributed with its hosts, large cold-blooded terrestrial vertebrates. In Cuba its main host are the bigger species of toads (Peltophryne spp.) but it has been reported from a wide variety of hosts outside Cuba. According to Keirans (1985) it is primarily a parasite of snakes but also parasites amphibians and is found only occasionally on iguanas. Other authors include also some mammals as occasional hosts. Amblyomma dissimile was reported from amber in the Dominican Republic, of (supposed) Eocene age (Lane, 1986; as Amblyomma sp. near testudinis). Other West Indies records are from Puerto Rico, Mona, Jamaica, Barbados, St. Lucia, Antigua, Grenada, Guadeloupe, Martinique. In the continent, it is known from Florida and Georgia to Argentina. The fourth species, H. leporispalustris, was known from migratory birds in the Lesser Antilles (Kohls, 1969) and now it is reported for the first time in Cuba. The adults are parasites on small mammals, but the immature can be found on birds (which can carry the ticks during migration). It looks as if the species is not established in Cuba or any other West Indies island.

THE CARIBBEANS The group is composed of species distributed over lands with coasts on the Caribbean Sea and the Gulf of Mexico, but never far inland. Three species represent it: Ornithodoros azteci and O. brodyi, both bat parasites, and Argas miniatus. The last species must have been distributed along with domestic chickens, but it is poorly known aside from being a chicken tick on the above-mentioned

98

Biogeography of the West Indies: Patterns and Perspectives

area (Kohls et al.,1970). In the continent, O. azteci and O. brodyi had been reported from Mexico to Venezuela. The first one has been recorded in Cuba and Jamaica (Kohls et al., 1965). The only Antillean record of O. brodyi is the Cuban one (Cerny, 1969).

THE NORTH AMERICAN–ANTILLEANS This group is represented by three species, Argas radiatus, Ornithodoros kelleyi, and Ixodes sp. The first species is a parasite of birds, including domestic species. In Cuba it has been found only in nests of cormorants. The second species is a parasite of bats, and had been reported as far south as Costa Rica (Kohls et al., 1965). The last one, Ixodes sp., is a new record from Cuba. The host was caught during migration, before it reached Cuban territory. It is a North American tick, carried by the bird. Only one species of Ixodes is reported to the West Indies, I. capromydis, but it is absolutely a different species.

THE WEST INDIES–SOUTH AMERICANS They are parasites on bats and are ecologically specific to the hot cave bats (Cruz, 1992). One species, Ornithodoros viguerasi, is distributed in the West Indies and the north coast of South America, but not Central America (Kohls et al., 1965, 1969). Other members of the same subgenus (Subparmatus), had been reported from Curaçao (O. mormoops) and Colombia and Panama (O. marinkellei) (Kohls et al., 1969).

THE WEST INDIES–CENTRAL AMERICANS As with the group cited above, they are chiropterophiles and thermocavernicolous. Parantricola marginatus is distributed in the West Indies and Mexico (Hoffmann et al., 1972), but not South America. I reviewed two collections of this species (no. RML 51193 and RML 52438), with a total of ten nymphs, two males and two females, from Cueva Vicenti, Samana, Dominican Republic, December 9, 1968, collected by F. H. Armstrong and M. L. Johnson. They were the first records of this species from Dominican Republic.

THE ENDEMICS Three species, Aponomma quadricavum, Amblyomma albopictum, and A. torrei, can be found only on Cuba and some other islands of the West Indies. Aponomma quadricavum is known from Cuba and Hispaniola; Amblyomma albopictum from Cuba, Hispaniola, and Swan Islands (Gulf of Honduras); and A. torrei from Cuba, Cayman Islands, and Puerto Rico. All are parasites on reptiles, primarily large snakes and lizards. Other species of Amblyomma, known from West Indian reptiles, are closely related to the Cuban species. The species A. antillarum is an iguana parasite in the Virgin Islands and the Bahamas (Kohls, 1969; Keirans, 1985) with a striking distribution. Its close relative A. albopictum is distributed between the two populations. Amblyomma arianae, a close relative of A. torrei, was described from Puerto Rican snakes (Keirans and Garris, 1986). Amblyomma cruciferum, a South American species that is a parasite of reptiles, has also been reported from Puerto Rico (Keirans and Garris, 1986). The Cuban endemics are the largest group, composed of 15 species: 10 species of Antricola, Ornithodoros natalinus, O. tadaridae, O. cyclurae, O. dusbabeki, and Ixodes capromydis. It is clear that almost half of the indigenous species of ticks from Cuba are local endemics. From this group, one species (O. cyclurae) is a parasite of iguanas, and one (I. capromydis) is a parasite of capromyid rodents. The remaining 13 species are parasites of bats and 10 of them, the Antricola species, are thermocavernicolous. The phylogenetic relationships of ticks at the species level have not been carefully studied. The most serious attempts have looked at generic relationships, and there is no consensus. Therefore, the relationships between Cuban and other related ticks will be considered in the sense of “appear

Patterns in the Biogeography of West Indian Ticks

99

as…” or “closer to … than to ….” I will talk only about the most evident relationships. The species with wide distributions, with many doubtful records, and those evidently distributed by humans in modern times will not be included. I will focus on the less distributed autochthonous species, which need more discussion. One species, O. cyclurae, is a nasal parasite of the Cuban iguana, Cyclura nubila. It has a close relative in Hispaniola, O. elongatus from an unknown host. Some morphological features (the shape of the dorsal setae, short, strong, and barbed; the 2/2 hypostome dentition; and the body and dorsal plate shape) and the presence of an iguana in the same container where the tick was found (Kohls et al., 1969) make me think it is also a parasite of iguanas (Cruz, 1984). No other species is close to these two species in the West Indies. Both appear highly specialized species, which is clear because of their relationship with the host and their localization (nasal parasite), together with the morphological derived characters. Another species, O. natalinus, is known from only a very special habitat of the Cueva del Lago, Nueva Gerona (Isla de Pinos). The host, the small bat Natalus lepidus, is a dweller of marginal niches, such as very small caves or caves with accumulations of CO2 , which is the situation at this locality. The tick has no clear relationship with any other species, and it was impossible to place it in any of the known subgenera (Cerny and Dusbabek, 1967; Jones and Clifford, 1972). A third species, O. dusbabeki, is also a parasite of bats on the northern region of Isla de Pinos. Its species specificity is very low and has been found in very different species of bats (Phyllostomidae, Molossidae, Noctilionidae) and in different ecological niches (hollowed trees, palm leaves, caves, human constructions). It is the only species of tick, beside O. natalinus, found on bats in the area mentioned above. In the southern region of the Isla de Pinos (south of Lanier Swamp) we found other chiropterophile ticks (O. azteci). It appears that O. dusbabeki is related to O. hasei, also a species with low host and habitat specificity. Both species had been found in the same families of bats and in the same array of habitats (tree hollows, caves, and human constructions). A fourth species, O. tadaridae, is a parasite of molossid bats, mainly from Mormopterus minutus. This bat lives in big colonies on palms (Silva, 1979) where the ticks can be found by the thousands, in all stages year round (Cruz, 1974). Of course, the tick has the same distribution in Cuba as the host. It was reported once from Mexico (Dusbabek, 1970). I reviewed Dusbabek’s material and identified it as a species very close to the Cuban one: O. hasei, a Central American species, also reported on bats from Trinidad, Martinique, Guadeloupe, and Barbuda. During a visit to the University of Mexico (January to March, 1992), I had the opportunity to study the Anita Hoffmann tick collection, and some other materials from the Laboratory of Acarology, were I found some “O. hasei” larvae. Ornithodorus hasei appears to be a complex species that needs a better review (see also Jones and Clifford, 1972). In my opinion O. tadaridae is closer to O. boliviensis, a parasite of molossid bats, found in human constructions (“huts”) in Bolivia, Venezuela, and Mexico (Jones et al., 1972). All ten species of the genus Antricola reported from Cuba are endemic, with each species inhabiting a different cave (Cruz, 1978a). There are other species of Antricola in North, Central, and South America (Cruz, 1978a; Need et al., 1991; Cruz and Estrada-Pena, 1995), and probably elsewhere in the West Indies (Cruz, 1978; Cruz and Estrada-Pena, 1995). The relationships of this group are very obscure. Some authors (Cerny, 1966; Klompen, 1989) include Parantricola in this genus, others do not (Cruz, 1974; Cruz and Dusbabek, 1989). All the representatives of the group (considered as a tribe, Antricolini, by Hoogstraal and Aeschlimann, 1982, and Hoogstraal, 1985) parasitize “hot cave” bats (Cruz, 1992). The only ixodid tick endemic to Cuba, I. capromydis, is a parasite of the endemic Cuban capromyid Capromys pilorides. I. capromydis is restricted to the southern region of the Isla de Pinos and its specific host is C. pilorides ciprianoi (considered by Woods et al., Chapter 18, this volume to be synonymous with C. p. relictus) although C. pilorides is widely distributed in Cuba and its archipelagos (Woods et al., Chapter 18, this volume). The species is the only representative of the subgenus Alloixodes, and has no clear relationship with any other members of the genus

100

Biogeography of the West Indies: Patterns and Perspectives

(Clifford et al., 1972). Ixodes capromydis shares common morphological features (cornua on all stages, palp structure of larvae and nymph and the hypostome) with the South American species, I. sigelos, a parasite of Chilean rodents. Both appear to be relicts from old species. Three other species have been reported from the West Indies but not from Cuba. The first one, Amblyomma variegatum, an African species parasitic on cattle, was introduced to Puerto Rico, some of the Lesser Antilles (Morel, 1966), and adjacent continental areas. The second species, O. portoricensis, is a parasite of rats and other small mammals. It is known from Hispaniola, Jamaica, Puerto Rico, Virgin Islands, Trinidad, St. Croix, Venezuela, Colombia, Panama, Brazil, Paraguay, Bolivia, and Argentina. It might be found in Cuba where the hosts and the appropriate ecological and colonization conditions are present. The third species is Ornithodoros sp. (group talaje) reported by Maldonado-Cariles and Medina-Gaud (1977) from Eleutherodactylus cooki from Puerto Rico. It is an unusual record since the host is a small amphibian, a host not very suitable for a tick. The account was ignored in the most recent revisions of American argasids (Kohls et al., 1965, 1969; Jones and Clifford, 1972). The species A. rotundatum mentioned by Morel (1967) is not included since he gave no convincing arguments to consider the record of this species valid. The Spinose ear tick, Otobius megnini, has been reported from different islands of the Antilles at different times by some authors, but all records were based on collections done from imported animals. No living population of this tick is known in the West Indies.

ECOLOGICAL ZOOGEOGRAPHY Ecological characters have not received the attention they deserve as biogeographical indicators. They have been mentioned as having “more or less invasive capabilities.” I found the analysis done by Henderson and Crother (1989) very useful, even if it deals with a very different subject, the predation pattern of snakes. They gave me the idea of the possible use of the ecological characters as zoogeographical indicators. I use eight different geographical units for comparison, as follows. Cuba — This island is focal to biogeographical analyses because it is the largest island in the West Indies. It is also important because of its ecological and geographical complexity, and its very wellknown vertebrate and tick faunas. The Greater Antilles — These islands are considered in two distinct groups: the Greater Antilles, sensu lato (including Cuba) and sensu stricto (excluding Cuba). The Greater Antilles include Hispaniola, Puerto Rico, Cayman Islands, Jamaica, and their satellite islands, and, of course, Cuba. I have chosen to look at the Greater Antilles in this way because some islands appear to have a distinct history and it is possible they have geological origins different from Cuba (Donnelly, 1988). Cuba was included with them to see how much Cuba influences the results. The Lesser Antilles — This group of volcanic islands is closer to continental South America than the Greater Antilles. This chain of islands offers an interesting point of comparison because of the possibility that the islands may have served as a “bridge” between the Antilles and continental South America. Venezuela and Panama — They are two units, both continental in nature and each of possible contact (directly or indirectly) of the West Indies and continental faunas (Rosen, 1975; Guyer and Savage, 1986). Peru — It is a continental area without any clear relationship with the West Indies and is a core part of the Neotropical region. Madagascar — This island has a clear relationship with its closer continental area (Africa) and has nothing in common with the West Indies, except for some aspects of historical development (Woods and Eisenberg, 1989).

Patterns in the Biogeography of West Indian Ticks

101

I recognize five groups of “habitat preferences,” as explained in Materials and Methods. Each group has special requisites for which ticks have become “preadapted.” This concept has been clearly explained by Hoogstraal and Kim (1985), but in relation to the evolution of ticks that changed host from reptiles (in the Jurassic) to mammals (in the Cretaceous). The same abilities give the ticks additional possibilities to colonize new geographical territories. One of the hardest steps of colonization for parasites is the possibility of finding an appropriate host. On “new” territories, hosts would likely be scarce. If the parasite has to leave the established host to fulfill some biological need (i.e., molts, oviposition), the lack of readily available hosts is a serious limitation to colonization and easily becomes a major cause of extinction. Then, if the host visits the same site (a den, a nest) on a regular basis (daily or seasonally), and if the ticks are “nest species,” they have greater chances for a successful colonization. Studies on the colonization on Krakatoa Atoll revealed that after volcanic eruption, seabirds were the first colonizers and soon after, their nest ticks. According to this scenario, the argasids can be considered as “pre-adapted” for overwater colonization, especially in association with the nests of birds and bats. Host preference is a complex character. First, some ticks have different host preference at different developmental stages (Hoogstraal and Kim, 1985). Second, in evolutionary time, ticks supposedly can switch from one host to another with only a few restrictions (Hoogstraal and Aeschlimann, 1982; Hoogstraal and Kim, 1985). For reasons already mentioned, ornithophiles and chiropterophiles ticks have better chances to successfully colonize offshore islands. Macromastophiles have fewer chances since their hosts have lower population densities, less nesting behavior, and few possibilities to move overseas. The most successful species of macromastophile ticks are the parasites of domestic animals, because of human actions on their distribution. Then, the “island” character can be reflected by the dominance of argasid parasites on birds and bats. Table 1 shows the dominance of argasids on islands (Cuba, 68%; Greater Antilles, with and without Cuba, 62.22 and 44%, respectively; Lesser Antilles, 46.47%) but not on continental areas (Venezuela, 36.21%; Peru, 31.82%; Panama, 20%; and Madagascar, 25%). Table 3 shows that (as would be expected) the macromastophiles dominate on continental areas (Venezuela, 41.38%; Panama, 62%; Peru, 43.18%). The situation of the Lesser Antilles (33% of macromastophiles) and Madagascar (25%) is intermediate. It is clear that the Lesser Antilles have very few species of ticks, and the five introduced species of domestic mammals became an important part of the local fauna (only 15 species in all). Madagascar has a depauperate continental tick fauna. It appears that the chiropterophiles dominate in the islands followed by the herpetophiles. Both groups, together, became the absolute dominants on islands. This pattern would be expected because of the high colonization abilities of bats and reptiles. The cavernicolous (Table 4), thermocavernicolous, and psicrocavernicolous forms have a certain level of dominance on islands also. In the Lesser Antilles the low number (two species) of psicrocavernicoles and complete lack of thermocavernicole forms can be explained by the lack of caves in this island arc as a result of their volcanic origin and geology. However, it should be noted that a lack of intensive research on the other islands of the Greater Antilles could explain the general absence of records for some species (as in the case of thermocavernicoles and some psicrocavernicoles). The thermocavernicoles (the ticks of hot caves) show a number of very interesting features. In Cuba, the hosts and the ticks are very specialized but not as much so as on adjacent continental areas. The continental bats (mormoopids and some phyllostomids) are very opportunistic and are found in very different kinds of caves. In general, the caves I have visited in Mexico were different from the close (= not large) and climatically uniform caves of Cuba. The tick populations were, as expected, less numerous in Mexico than in Cuba caves. In Cuba, the prevalent bats of hot caves are Phyllonycteris poeyi, Pteronotus macleayii, P. quadridens, and Mormoops blainvillii. Both groups, Brachyphillinae and Mormoopidae, evolved in the West Indies, from Central American ancestors (Silva, personal communication). The same pattern probably occurred for the associated ticks. Some less specialized forms of ticks came from Central America with the ancestral bats and evolved into the highly specialized thermocavernicoles. Later on, the bats returned to continental

14.29 14.29 51.43 5.71 14.29 0.00 100%

5 5 18 2 5 0 35

Herpetophile Ornithophile Chiropterophile Micromastophile Macromastophile Unknown Total 10 4 4 1 6 0 25

No. 40.00 16.00 16.00 4.00 24.00 0.00 100%

% 11 5 19 3 7 0 45

No. 24.44 11.11 42.22 6.67 15.56 0.00 100%

%

With Cuba

Greater Antilles Without Cuba

2 4 2 2 5 0 15

No. 13.33 26.67 13.33 13.33 33.33 0.00 100%

%

Lesser Antilles

6 3 12 12 24 1 58

No. 10.34 5.17 20.69 20.69 41.38 1.72 100%

%

Venezuela

b

a

5 3 5a 12 11 35

a

No.

12.86 8.57 12.86 34.29 31.43 100%

% 3 5 2b 2 14 25

b

No. 10 20 6 8 56 100%

% 6 6 6 12 17 45

No. 8.89 13.33 13.33 26.67 37.78 100%

%

With Cuba

Greater Antilles Without Cuba

Ornithodoros dusbabeki is arboricolous and psicrocavernicolous. Ornithodoros hasei is arboricolous and psicrocavernicolous.

Arboricolous Lapidicolous Psicrocavernicolous Thermocavernicolous Open fields Total

Host Affinity

Cuba

3 3 2 0 8 15

No. 16.67 20.00 10.00 0.00 53.33 100%

%

Lesser Antilles

3 7 8 3 37 58

No.

5.17 12.07 13.79 5.17 63.79 100%

%

Venezuela

TABLE 4 Comparative Tick Fauna by Structural Niche from Different Regions of the World

%

No.

Host Affinity

Cuba

8.00 4.00 10.00 14.00 62.00 2.00 100%

%

3 3 3b 2 40 50

b

No.

5.00 6.00 5.00 4.00 80.00 100%

%

Panama

4 2 5 7 31 1 50

No.

Panama

TABLE 3 Comparative Host Affinity Composition and Percentage from Different Regions of the World

4 7 2 1 30 44

No.

4 9 3 7 19 2 44

No. %

% 9.09 15.91 4.55 2.27 68.18 100%

Peru

9.09 20.45 6.82 15.91 43.18 4.55 100

Peru

3.13 12.50 6.25 53.13 25.00 0.00 100%

%

1 5 1 0 25 32

No.

3.13 15.63 3.13 0.00 78.13 100%

%

Madagascar

1 4 2 17 8 0 32

No.

Madagascar

102 Biogeography of the West Indies: Patterns and Perspectives

Patterns in the Biogeography of West Indian Ticks

103

Central America (via the Yucatan Peninsula?) carrying some of their tick parasites and became part of the actual continental populations. The genus Ornithodoros is the most diverse argasid, especially on the islands (Table 1). This might be due to some geographical influence. The ornithodorine ticks are dominant in the New World, whereas argasines are dominant in the Old World. In this analysis only islands in the Americas are included. In Ixodidae, the dominance of Amblyomma in the New World and of Haemaphysalis in Madagascar are evident. The most striking feature is the presence of Dermacentor only in Panama. The genus is clearly more diverse in the Nearctic and Palearctic regions. Its presence in Panama should represent the southern limits of a Nearctic distribution in the process of expanding its range southward.

CONCLUSION The study of West Indian ticks reveals a pattern that suggests that the West Indies is an assemblage of islands and banks of different origins, ages, and natures. It is clear that Cuba has had an independent origin and has never been connected with any continental area, or with other islands of the West Indies. This independent origin gives Cuba a more insular character in regard to its tick fauna, a character that is reflected in the composition of other taxonomic groups. The rest of the Greater Antilles appears to be characterized by a more homogenous assemblage of ticks, with many species or species groups in common with each other. At the eastern end of these islands the volcanic Lesser Antilles seem to be a bridge or series of stepping-stones between the Greater Antilles and South America, but independent of both. The different colonization patterns that influenced the composition and character of the West Indian tick fauna are as follows. 1. The invasion from Central America to the West Indies — There are species that show relationships between Central America–Cuba–the Great Antilles–the Lesser Antilles–South America. This invasion does not have a clear direction and probably went both ways. The ancestors of O. hasei-dusbabeki and O. boliviensis–tadaridae seem to have come from South America and to have become Cuban species. The genus Parantricola appears to have come from Central America to Cuba–Hispaniola. The direction of dispersal of the genus Antricola and the subgenus Subparmatus is not clear since they have representatives on both extremes and between. Other species, such as O. cyclurae-elongatus, O. natalinus, Ixodes capromydis, Amblyomma albopictum-antillarum, and A. torrei-arianae-cruciferum, have obscure relationships with the continent. They probably followed the distributions of their host species as with thermocavernicolous forms. This is certainly the oldest colonization of ticks in the West Indies and occurred between the Eocene and the Pliocene, since it escaped the invasion of the North American elements to Central America (genus Dermacentor). If the colonization were from South America through the Lesser Antilles, it would have been later than the first one from Central America. 2. The invasion from North America — The species with clear North American origin, such as Argas radiatus and O. kelleyi, arrived with their hosts, the cormorants and the bat Eptesicus fuscus, probably during some of the sea recessions of the Pleistocene. 3. Human introduction — The last colonization took place during the historical times when humans brought their domestic animals to the Antilles. In this category it is possible to include the colonization of Amblyomma rotundatum and the “redistribution” of A. dissimile and the little-known O. portoricensis. 4. Unknown origin — It is impossible at present to include the bird parasites O. capensis and O. denmarki in any zoogeographical colonizing event because of their worldwide distribution and unknown evolution.

104

Biogeography of the West Indies: Patterns and Perspectives

ACKNOWLEDGMENTS I thank everyone who helped me, directly or indirectly, to complete this difficult task. First, I thank Dr. Charles Woods for giving me the opportunity and the honor to participate in this publication and Florence Sergile for her patience and dedication revising and editing the manuscript. I also thank the staff of Grove Scientific & Engineering for allowing me to use their facilities during the elaboration of this chapter. Last, but not least, I am grateful to Dart Morales, Bruno Ferraro, and Mary Spirig, for their kind collaboration.

LITERATURE CITED Anderson, J. F., L. A. Megnarelli, and J. E. Keirans. 1981. Aponomma quadricavum (Acari: Ixodidae) collected from an infested boa, Epicrates strictus, in Connecticut. Journal of Medical Entomology 18:123–125. Cerny, V. 1966a. Nueva especie de garrapata del genero Ixodes Latreille (Ixodoidea, Ixodidae) en la Jutia Conga de la Isla de Pianos. Poeyana, seria A 24:1–9. Cerny, V. 1966b. Nuevas garrapatas (Ixodoidea) en aves y reptiles de Cuba. Poeyana, seria A 26:1–9. Cerny, V. 1969. The tick fauna of Cuba. Folia Parasitologia (Praha) 16(3):279–284. Clifford, C. M., D. E. Sonenshine, J. E. Keirans, and G. M. Kohls. 1973. Systematics of the subfamily Ixodinae (Acarina: Ixodidae). 1. The subgenera of Ixodes. Annals of the Entomological Society of America 66(3):489–500. Cooley, R. A. and G. M. Kohls. 1944. The Argasidae of North America, Central America and Cuba. The American Middland Naturalist, Monograph No. 1:1–152. Cruz, J. de la. 1974a. Notas adicionales a la fauna de garrapatas (Ixodoidea) de Cuba. I. Argasidae de las aves. Poeyana 128:1–8. Cruz, J. de la. 1974b. Notas adicionales a la fauna de garrapatas (Ixodoidea) de Cuba. II. Nuevo status para Parantricola Cerny. 1966. Poeyana 130:1–4. Cruz, J. de la. 1974c. Notas adicionales a la fauna de garrapatas (Ixodoidea) de Cuba. III. Redescripcion de Ornithorodos [sic] tadaridae Cerny y Dusbabek. 1967. Poeyana 138:1–5. Cruz, J. de la. 1976. Notas adicionales a la fauna de garrapatas (Ixodoidea) de Cuba. IV. Presencia de Argas (Persicargas persicus) (Oken, 1818). Miscellaneous Zoologia. Academia de Ciencias de Cuba 2:3. Cruz, J. de la. 1978a. Notas adicionales a la fauna de garrapatas (Ixodoidea) de Cuba. VI. Cuatro nuevas especies del genero Antricola Cooley et Kohls. 1942 (Argasidae, Ornithodorinae). Poeyana 184:1–17. Cruz, J. de la. 1978b. Composicion zoogeografica de la fauna de garrapatas (Acarina:Ixodoidea) de Cuba. Poeyana 185:1–6. Cruz, J. de la. 1984. Una nueva especie de garrapata del genero Ornithodoros (Acarina: Ixodoidea, Argasidae), parasita de la cavidad nasal de la iguana Cyclura nubila (Sauria: Iguanidae). Poeyana 277:1–6. Cruz, J. de la. 1985. Argas (Persicargas) radiatus (Acarina: Argasidae), a tick new for Cuba. Miscellaneous Zoologia. Academia de Ciencias de Cuba 25:1. Cruz, J. de la. 1986. Coeficientes de similitud zoogeografica y su aplicación a las condiciones insulares de Argasidae (Acarina) del Mediterraneo Americano. Ciencias Biologias Academia de Ciencias de Cuba 16:87–106. Cruz, J. de la. 1987. La fauna de garrapatas (Ixodoidea) de la Republica de Cuba. Doctoral dissertation, Institute of Parasitology, Ceske Budejovice, Czechoslovakia. Cruz, J. de la. 1992. Bioecologia de las Cuevas Calientes. Mundos subterraneos, UMAE, Mexico, 3:7–22. Cruz, J. de la and F. Dusbabek. 1989. Haller’s organ and anterior pit in the genera Antricola and Parantricola (Ixodoidea: Argasidae). Folia Parasitologia (Praha) 36:275–279. Cruz, J. de la and A. Estrada-Pena. 1995. Four new species of Antricola ticks (Argasidae:Antricolinae) from bat guano in Cuba and Curaçao. Acarologia 36(4):277–286. Donnelly, T. W. 1988. Geological constraints on Caribbean Biogeography. Pp. 15–37 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. Dusbabek, F. 1970. New records of parasitic mites (Acarina) from Cuba and Mexico. Mitteilungen Zoologisches Museum Berlin 46(2):273–276. Estrada-Pena, A. 1989. Indice-catalogo de las garrapatas (Acarina: Ixodoidea) en el mundo. Vol. 1: Genero Haemaphysalis. Secretariado de Publicaciones, Universidad de Zaragoza, Zaragoza.

Patterns in the Biogeography of West Indian Ticks

105

Fairchild, G. B., G. M. Kohls, and V. J. Tipton. 1966. The ticks of Panama (Acarina:Ixodoidea). Pp. 167–219 in Wenzel, R. L. and V. J. Tipton (eds.). Ectoparasites of Panama. Field Museum of Natural History, Chicago. Guyer, C. and J. M. Savage. 1986. Cladistic relationships among anoles (Sauria: Iguanidae). Systematic Zoology 35:509–531. Henderson, R. W. and B. I. Crother. 1989. Biogeographic patterns of predation in West Indian Colubrid snakes. Pp. 479–518 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Hershkovitz, P. 1966. Mice, land bridges and Latin American faunal interchange. Pp. 725–751 in Wenzel, R. L. and V. J. Tipton (eds.). Ectoparasites of Panama. Field Museum of Natural History, Chicago. Hoffmann, A., I. Bassols-Barrera, and C. Mendez. 1972. Nuevos hallazgos de acaros en Mexico. Revista de la Sociedad Mexicana de historia natural 33:151–159. Hoogstraal, H. 1985. Argasid and Nutalliellid ticks as parasites and vectors. Advances in Parasitology 24:135–238. Hoogstraal, H. and A. Aeschlimann. 1982. Tick-host specificity. Bulletin de la Société Entomologique de Suisse 55:5–32. Hoogstraal, H. and K. C. Kim. 1985. Tick and mammal coevolution, with emphasis on Haemaphysalis. Pp. 505–567 in Kim, K. C. (ed.). Coevolution of Parasitic Arthropods and Mammals. John Wiley & Sons, New York. Jones, E. K. and C. M. Clifford. 1972. The systematics of the subfamily Ornithodorinae (Acarina: Argasidae). V. A revised key to larval Argasidae of the Western Hemisphere and description of seven new species of Ornithodoros. Annals of the Entomological Society of America 65(3):730–740. Jones, E. K., C. M. Clifford, J. E. Keirans, and G. M. Kohls. 1972. The ticks of Venezuela (Acarina: Ixodoidea) with a key to the species of Amblyomma in the Western Hemisphere. Brigham Young University, Science Bulletin, Biological Series 17:1–40. Keirans, J. E. 1985. Amblyomma antillarum Kohls. 1969 (Acarina: Ixodoidea): description of the immature stages from the Rock Iguana, Iguana pinguis (Sauria: Iguanidae) in the British Virgin Islands. Proceedings of the Entomological Society of Washington 87(4):821–825. Keirans, J. E. and W. D. Degenhart. 1985. Aponomma elaphense Pine. 1959 (Acari: Ixodidae): diagnosis of the adults and nymph with first description of the larva. Proceedings of the Biological Society of Washington 98(3):711–717. Keirans, J. E. and G. I. Garris. 1986. Amblyomma arianae, n. sp. (Acarina: Ixodidae), a parasite of Alsophis portoricensis (Reptilia: Colubridae) in Puerto Rico. Journal of Medical Entomology 23(6):622–625. Keirans, J. E., C. M. Clifford, and D. Corwin. 1976. Ixodes sigelos, n. sp. (Acarina: Ixodidae), a parasite of rodents in Chile, with a method for preparing ticks for examination by scanning electron microscopy. Acarologia 18(2):217–225. Klompen, J. S. H. 1992. Comparative morphology of Argasid larvae (Acarina: Ixodida: Argasidae), with notes on phylogenetic relationships. Annals of the Entomological Society of America 85(5):541–560. Kohls, G. M. 1969a. A new species of Amblyomma from iguanas in the Caribbean (Acarina: Ixodidae). Journal of Medical Entomology 6(4):439–442. Kohls, G. M. 1969b. New records of ticks from the Lesser Antilles. Studies of the Fauna of Curaçao and Other Caribbean Islands 28:126–134. Kohls, G. M., D. E. Sonenshine, and C. M. Clifford. 1965. The systematics of the subfamily Ornithodorinae (Acarina: Argasidae). II. Identification of the larvae of the Western Hemisphere and descriptions of three new species. Annals of the Entomological Society of America 58(3):331–364. Kohls, G. M., C. M. Clifford, and E. K. Jones. 1969. The systematics of the subfamily Ornithodorinae (Acarina: Argasidae). IV. Eight new species of Ornithodoros from the Western Hemisphere. Annals of the Entomological Society of America 62(5):1035–1043. Kohls, G. M., H. Hoogstraal, C. M. Clifford, and M. N. Kaiser. 1970. The subgenus Persicargas (Ixodoidea, Argasidae, Argas). 9. Redescription and New World records of Argas (P.) persicus (Oken) and resurrection, redescription and records of A. (P.) radiatus Railliet, A. (P.) sanchezi Duges and A. (P.) miniatus Koch, New World ticks misidentified as A. (P.) persicus. Annals of the Entomological Society of America 63(2):590–606. Lane, R. S. and G. P. Poinar. 1986. First fossil tick (Acarina: Ixodidae) in New World amber. International. Journal of Acarology 12(2):75–78.

106

Biogeography of the West Indies: Patterns and Perspectives

Liebherr, J. K. 1988. The Caribbean: fertile ground for zoogeography. Pp. 1–14 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Comstock Publishing Associates, Cornell University Press, Ithaca, New York. Maldonado-Caapriles, J. and S. Medina-Gaud. 1977. The ticks in Puerto Rico (Arachnida: Acarina). Journal of Agriculture, University of Puerto Rico 61:402–404. Morel, P. C. 1966. Etude sur les tiques du bétail en Guadeloupe et Martinique. I. Les tiques et leur distribution (Acarina, Ixodoidea). Revue d’élevage et de médecine vétérinaire des pays tropicaux 19(3):307–321. Morel, P. C. 1967. Les tiques de animaux sauvages des Antilles (Acariens, Ixodoidea). Acarologia 9:341–352. Morel, P. C. and P. Fauran. 1967. Présence en Guadeloupe de l’ornithodore Alectorobius puertoricensis (Fox. 1947) (Acariens, Ixodoidea). Acarologia 9(2):338–340. Need, J. T., W. E. Dale, J. E. Keirans, and G. A. Dash. 1991. Annotated list of ticks (Acarina: Ixodidae, Argasidae) reported in Peru: distribution, hosts and bibliography. Journal of Medical Entomology 28(5):590–597. Rosen, D. E. 1975. A vicariance theory of Caribbean biogeography. Systematic Zoology 24:431–463. Schuchert, C. 1935. Historical Geology of the Antillean-Caribbean Region. John Wiley & Sons, New York. Silva Taboada, G. 1979. Los murcielagos de Cuba. Editorial Academia, Habana. Tamsitt, J. R. and I. Fox. 1970. Records of bat ectoparasites from the Caribbean region (Siphonaptera, Acarina, Diptera). Canadian Journal of Zoology 48:1093–1097. Uilemberg, G., H. Hoogstraal, and J. H. Klein. 1979. Les tiques (Ixodoidea) de Madagascar et leur rôle vecteur. Archives de l’Institut Pasteur Madagascar, Special Number:1–153. Vargas, M. M. 1984. Ocurrence of the bat tick Ornithodoros (Alectorobius) kelleyi Cooley & Kohls (Acarina: Argasidae) in Costa Rica and its relation to human bites. Revista de biologia tropical 32(1):103–107. Wilson, N. and N. W. Kale. 1972. Ticks collected from Indian River County, Florida (Acarina: Metastigmata: Ixodidae). Florida Entomologist 55(1):53–58. Woods, C. A. 1989. The biogeography of West Indian rodents. Pp. 741–798 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Woods, C. A. and J. F. Eisenberg. 1989. The land mammals of Madagascar and the Greater Antilles: comparison and analysis. Pp. 799–826 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

Contribution of the 8 The Caribbean to the Spider Fauna of Florida Jonathan Reiskind Abstract — Since most, if not all, the spider species of the Florida peninsula are relatively recent arrivals, either because the land had been fully submerged at some time during the Pleistocene or because that islands’ areas in the Florida archipelago during interglacial periods were small and local extinctions of the fauna were likely, an analysis of present-day distributions permits determination of the sources of the spider fauna. Dispersal during the Pleistocene played the dominant role in the origin of the peninsular Florida fauna. Climatic constraints to present distributions offer additional clues to origins. By using explicit criteria, most species can be placed in one of four origin groups: northern (far and near), western, southern (Caribbean), and autochthonous (originating on the islands of the Florida “archipelago”). In addition some spiders are just “neotropical,” either arriving along the Gulf Coast or through the West Indies, and others are cosmopolitan with recent anthropogenic factors likely responsible. This combination of historic contingency and ecological limitations make the spider fauna of Florida both diverse and somewhat limited. The contribution of the Greater Antilles to the spider fauna of Florida is relatively small, amounting to less than 10%.

INTRODUCTION Much of the terrestrial fauna of peninsular Florida is assumed to be of relatively recent origin since a large portion of the peninsula was submerged during the interglacial periods of the Pleistocene, including all of Florida south of 27°N latitude. During this epoch central and northern regions of Florida had periodically reduced land areas and many coastal islands. Isolated habitat islands were also likely during the cyclical climatic changes of the Ice Age. Climatic conditions varied with conditions during glacial periods, such as those in the Wisconsinan only 70,000 years ago, noticeably cooler and more xeric than present (Webb, 1990). The recency (i.e., within the last 1 million years) of the terrestrial fauna, including spiders, allows the use of present distributions of the spider species with representatives in Florida to infer their origins. Granted, the Pleistocene, as brief as it was, was quite complex with great fluctuations of sea levels as well as climatic conditions. The feasibility of such determinations is examined in this study and some rough estimates of origins made. Of course, only the presence of a species can be definitively demonstrated; its absence is only tentative. Thus all the data depend on the extensiveness and completeness of the survey of the spider fauna.

METHODS The most recent taxonomic revisions of genera and higher taxa were used for this study. The distributions of over 450 species occurring in peninsular Florida have been reviewed (from over 90 papers) and designated as found in one or more of five regions in Florida (Table 1). Then, using the regional designations in Table 1 and the criteria listed in Table 2, the probable source regions of each species were inferred. The farthest region in which a species is found was chosen as its putative source. For example, if a species is found in north and central Florida, the southeast United 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

107

108

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 Regions of Florida Regions

Geographical Position

Keys South Florida Central Florida North Florida Panhandle

Below 25°N 25° to 27°N 27° to 29°N 29° to Georgia border and east of Suwannee River West of Suwannee River

TABLE 2 Criteria of Source Regions Source Region

Criteria

Northern (far) (N+) Northern (near) (N–) Southern tropical (S) Western (W+) Western tropical (W–) Autochthony (A) Ambiguous Sources Neotropical (T) Nearctic (NW) Cosmopolitan (O) or Cosmotropical (OT) Disjunct distributions

Found north of 40°N and east of Mississippi River Found north of 31°N, but not north of 40°N South (Greater Antilles) (“the overwater route”) “Gulf coastal corridor” — west of Mississippi River in U.S. “Gulf coastal corridor” — Mexico (and Central America) Distribution restricted to peninsular Florida Both S and W– (i.e., Cuba vs. Mexico) N+ and W (+ and –) A widespread species, often anthropogenic factors contribute to its dispersal and distribution e.g., N and S

States and north into New England its source would be designated far northern (“N+”). If a species is found only in the Greater Antilles and south and central Florida, it would be designated “S” and its origin considered southern. But if it were to be found both in Mexico, along the Gulf Coast, and in Cuba its origin was considered ambiguously “tropical” (“T”). The distributions were recorded from revisionary studies of genera of four higher spider taxa (Table 3). The four higher taxa were chosen for study on the basis of the comprehensiveness of their taxonomic studies, the diversity of their ecological habits, and their high species diversity in peninsula Florida and adjacent regions. They are the Araneidae/Tetragnathidae (orb-weavers), Gnaphosidae (ground spiders), Lycosidae (wolf spiders), and Theridiidae (comb-foot or cob-web spiders).

RESULTS POTENTIAL SOURCES

OF THE

SPIDER FAUNA

Of course, all species of spiders or their immediate ancestors entered Florida either from the near north (the land route) or from the south (the overwater route). But within the Pleistocene epoch (about 1 million years) we might interpret the distributions of the fauna more broadly to reflect ultimate sources (Figures 1 and 2). Autochthony — New species arose in peninsular Florida and remain in place, i.e., autochthonous species. Of course they did not sprout from the brow of Zeus, but rather from ancestral or sister species that in turn came from elsewhere. Examination of their closest phylogenetic relatives will give a reasonable idea of their origins. For example, the sister group to the autochthonous Zelotes

The Contribution of the Caribbean to the Spider Fauna of Florida

109

TABLE 3 Citations of Revisions of the Florida Genera in the Four Taxa Used in the Survey Genus

Reference

Family Araneidae/Tetragnathidae Acacesia Levi (1976) Acanthepeira Levi (1976) Araneus Levi (1971a, 1973) Argiope Levi (1968) Azilia Levi (1980a) Cyclosa Levi (1977a) Dolichognatha Levi (1981) Eriophora Levi (1970) Eustala Levi (1977a) Gasteracantha Levi (1996) Gea Levi (1968) Glenognatha Levi (1980) Hyposinga Levi (1972) Kaira Levi (1977b, 1993) Larinia Levi (1975) Leucauge Levi (1980) Mangora Levi (1975) Mastophora Gertsch (1955) Mecynogea Levi (1980, 1997) Metapeira Levi (1977b) Metazygia Levi (1977a, 1995) Micrathena Levi (1985) Neoscona Berman and Levi (1971) Nephila Levi (1980) Nuctenea Levi (1974) Pachygnatha Levi (1980a) Scoloderus Levi (1976) Tetragnatha Levi (1981) Verrucosa Levi (1976) Wagneriana Levi (1976) Wixia Levi (1976)

Callilepis Cesonia Drassodes Drassyllus Eilica Gnaphosa Haplodrassus Herpyllus Litopyllus Micaria Nodocion Rachodrasssus Sergiolus Trachyzelotes Urozelotes Zelotes

Family Gnaphosidae Platnick (1975) Platnick and Shadab (1980b) Platnick and Shadab (1975b) Platnick and Shadab (1982) Platnick (1975b) Platnick and Shadab (1975c) Platnick and Shadab (1975a) Platnick and Shadab (1977) Platnick and Shadab (1980a) Platnick and Shadab (1988) Platnick and Shadab (1980a) Platnick and Shadab (1976) Platnick and Shadab (1981) Platnick and Murphy (1984) Platnick and Murphy (1984) Platnick and Shadab (1983)

Genus Arctosa Geolycosa Gladicosa Lycosa Pirata Trochosa Schizocosa Sossipus

Reference Family Lycosidae Dondale and Redner (1983) Wallace (1942a), McCrone (1963) Brady (1987) Wallace (1942b, 1947, 1950) Wallace and Exline (1978) Brady (1979) Dondale and Redner (1978) Brady (1962, 1972) Family Theridiidae

Anelosimus Agyrodes Achaearanea Chrysso Coleosoma Crustulina Dipoena Enoplognatha Episinus Euryopis Paidisca Latrodectus Paratheridula Pholcomma Phoroncidia Spintharus Steatoda Tekellina Theridion Theridula

Levi (1963e) Exline and Levi (1962) Levi (1963b) Levi (1962b) Levi (1959) Levi (1957b) Levi (1963a) Levi (1957a) Levi (1964a) Levi (1963a) Levi (1957a) McCrone and Levi (1964) Levi (1957c, 1966) Levi (1957c) Levi (1964b) Levi (1963d) Levi (1957b, 1962a) Levi (1957c) Levi (1957a, 1963c, 1980b) Levi (1966)

110

Biogeography of the West Indies: Patterns and Perspectives

50 Southern (S) Autochthonous (A) Near North (N-)

40

Far North (N+) Western Tropical (W-) Tropical (T)

Percentage of Species

30

20

10

0 Araneidae/Tetragnathidae

Gnaphosidae

Lycosidae

Theridiidae

FIGURE 1 Distributions and likely sources of species within four taxa using the criteria in Table 2.

ocala in central and north Florida, Z. duplex, has a far northern (N+) distribution (Platnick and Shadab, 1983). Unfortunately, most revisions do not yet identify the sister groups. Also, it is likely that some species designated near north (N–) represent species that have arisen in Florida and subsequently and recently dispersed north. Tropical sources — Spiders from the Neotropics could arrive either across water (from the Greater Antilles) or by a land route (from Mexico) along the Gulf Coast. By land — The Gulf coastal corridor is the land route from the Neotropics to Florida. This land passage had a significantly larger area during the glacial periods of the Pleistocene when the sea level was lower. It is the likely route taken by spiders is considered western tropical (W–) in origin. It is also the likely route for the majority of species classified as “tropical” (T), although finding a species in both Cuba and Mexico with continuous distribution along the Gulf Coast and in Florida presents an ambiguous situation. By sea — To arrive in Florida from the south (S) requires the ability to cross water because there was never a land connection to the Greater Antilles during this period. Spiders have that ability (via ballooning). Small but significant proportions of theridiids and araneids/ tetragnathids are southern in origin and likely arrived in this manner (see Figure 2). Temperate sources — Two continental boreal sources (W+ and N) are potential contributors to Florida’s fauna. All the spiders must have come through an area (N–) just north and west of the peninsula. Species found above 40°N latitude and not in the tropics can be considered to be northern (N) in origin. It is quite possible that the near northern area (N–) acted as a refuge for species at present distributed in the far northern (N+) areas. Nonetheless, a far northern present distribution still reflects a historic event involving a species from the north tolerant of cold climes. Few spiders show western origins. Ambiguous and other sources — Several species can be considered cosmopolitan or cosmotropical in distribution (e.g., Latrodectus geometricus, the brown widow) and often have anthropogenic factors responsible for their wide distribution. Florida species also found only in the near north (N–) are considered to have ambiguous origins as well. Few of the species examined had truly disjunct distributions (e.g., N and S). All these spiders were excluded from these analyses.

The Contribution of the Caribbean to the Spider Fauna of Florida

111

60 Tropical (S, W-, T) Temperate (N+ & W+)

50

40

Percentage of Species 30

20

10

0 Araneidae/Tetragnathidae

Gnaphosidae

Lycosidae

Theridiidae

FIGURE 2 Tropical vs. temperate sources of species within four taxa. Percentages may not add up to 100% as autochthonous species (A) and those from near north (N–) are excluded.

80 Warm (N-, W-, S, T, A) Cold (N+, W+, O, NW) 60

Percentage of Species

40

20

0 Araneidae/Tetragnathidae

Gnaphosidae

Lycosidae

Theridiidae

FIGURE 3 Climatic tolerances of species within four taxa found in tropical or moderate environments vs. colder climates.

CLIMATIC CONSTRAINTS The distribution of species is often limited by climatic constraints (Figure 3). A species may have the opportunity to disperse into an area but it may be physiologically constrained by the environmental temperature. Many can and do adapt quickly to these new conditions; others may not. Again, thanks to the recency of the fauna we may more easily detect those constraints before adaptations obscure them.

112

Biogeography of the West Indies: Patterns and Perspectives

Inability to tolerate cold restricts some to tropical and subtropical climates (species designated N–, W–, S, T, A) while others seem able to live in cooler climates (N+, W+, O, NW). Species intolerant of warm climates would not be found in Florida at all. In Figure 3, “warm” refers to those species found only in moderate to tropical areas while “cold” refers to those whose distributions include far northern regions.

SEVERAL HIGHER TAXA The species of the four higher taxa were analyzed with respect to their origins (Figures 1 and 2) and possible climatic constraints (Figure 3). Araneidae/Tetragnathidae (67 Floridian species) (Table 3) — Species are about equally “tolerant” of tropical and temperate climes (Figure 3) and their origins reflect this as well (Figure 2). Approximately 10% are clearly Caribbean in origin, reflecting their aerial dispersal abilities. Gnaphosidae (35 Floridian species) (Table 3) — Most species are of boreal origin with over half having clear temperate origins. Almost one half of the species are found north of 40°N and one fifth of the Florida fauna is autochthonous, reflecting their sedentary habits. It is not surprising that the distributions of many species show tolerance of a cooler climate. Lycosidae (42 Floridian species) (Table 3) — This family shows an even higher degree of autochthony (48%). Speciation in Florida could be still higher considering that 24% of the species show near north associations and may have arisen in Florida and spread north (Figure 1). While the Florida lycosid species appear intolerant of colder climates (Figure 3) that does not mean they have tropical origins (see Figure 2). This may just reflect characteristics of those species that have arisen in Florida recently. Theridiidae (74 Floridian species) (Table 3) — A predominantly tropical family (45% show clear tropical origins), their widespread dispersal abilities have obscured the origin of those “tropical” elements with over 70% of the species unable to be assigned a southern (Caribbean) vs. tropical western (Mexican) origin, although most are likely to have come via the land route through the Gulf coastal corridor.

DISCUSSION AND CONCLUSIONS The use of present distributions to ascertain the origin of the spider fauna of Florida allows a rough estimate of sources. Although of geologically recent origin, there has still been ample time for complex patterns to result from dispersal of extant species and speciation events with subsequent dispersal. In some cases the origins are quite clear, e.g., where a spider is widespread throughout eastern North America and found in north and central Florida (northern origin), or in cases of narrow distributions restricted to the Florida peninsula (autochthony) as in the case of Hogna ericeticola (the rosemary wolf spider), with a range of only 3000 ha in a distinctive habitat (Reiskind, 1987, 1996). But in most cases confidently ascribing sources of the fauna is at best an educated guess with severe limitations. The climatic tolerance of a species adds a constraint to dispersal opportunities, likely limiting the present-day distributions of several Florida species. Of course, there are many other factors: competition, other ecological constraints, etc. The generalizations on the distribution and potential sources of the four taxa examined merely reflect the obvious adaptations and historic distributions of those groups. Although the conclusions are limited by our knowledge of the distribution of the groups studied, I think it unlikely that common species in these well-studied groups would have been overlooked. Thus, to answer the original question: Can the origins of the spider fauna of peninsular Florida be determined using present distributions? Not easily in most cases and then only on a species-by-species basis, at best. The Caribbean (i.e., Greater Antilles) as a source is easier to estimate. Its contribution to the spiders of Florida, even southern Florida, is relative small, making up less than 10% of the theridiid and araneid/tetragnathid fauna. The picture could be significantly clarified if we had additional

The Contribution of the Caribbean to the Spider Fauna of Florida

113

collections throughout Florida and throughout North America and the Caribbean, more phenological studies of each of the Florida species, and studies of the phylogenetic relationships, especially of the autochthonous species.

LITERATURE CITED Berman, J. D. and H. W. Levi. 1971. The orbweaver genus Neoscona in North America. Bulletin of the Museum of Comparative Zoology 141(8):465–500. Brady, A. 1962. The spider genus Sossipus in North America, Mexico and Central America. Psyche 69(3):129–164. Brady, A. 1972. Geographic variation and speciation in the Sossipus floridana species group. Psyche 79(1–2):27–48. Brady, A. 1979. Nearctic species of the wolf spider genus Trochosa. Psyche 86(2–3):167–212. Brady, A. 1987. Nearctic species of the new wolf spider genus Gladicosa. Psyche 93:285–319. Dondale, C. D. and J. H. Redner. 1978. Revision of the Nearctic wolf-spider genus Schizocosa. Canadian Entomologist 110:143–181. Dondale, C. D. and J. H. Redner. 1983. Revision of the wolf spiders of the genus Arctosa in North and Central America. Journal of Arachnology 11(1):1–30. Exline, H. and H. W. Levi. 1962. American spiders of the genus Argyrodes. Bulletin of the Museum of Comparative Zoology 127(2):75–204. Gertsch, W. 1955. The North American bola spiders of the genera Mastophora and Agatostichus. Bulletin of the American Museum of Natural History 106(4):225–254. Levi, H. W. 1957a. The spider genera Enoplognatha, Theridion, and Paidisca in America north of Mexico. Bulletin of the American Museum of Natural History 112(1):1–124. Levi, H. W. 1957b. The spider genera Crustulina and Steatoda in North America, Central America and the West Indies. Bulletin of the Museum of Comparative Zoology 117(3):367–424. Levi, H. W. 1957c. The North American genera Paratheridula, Tekellina, Pholcomma and Archerius. Transactions American Microscopical Society 76(2):105–115. Levi, H. W. 1959. The spider genus Coleosoma. Breviora 110:1–10. Levi, H. W. 1962a. The spider genera Steatoda and Enoplognatha in America. Psyche 69(1):11–36. Levi, H. W. 1962b. More American spiders of the genus Chrysso. Psyche 69(4):209–237. Levi, H. W. 1963a. American spiders of the genera Audifia, Euryopis and Dipoena. Bulletin of the Museum of Comparative Zoology 129(2):121–186. Levi, H. W. 1963b. American spiders of the genus Arachaearanea and the new genus Echinotheridion. Bulletin of the Museum of Comparative Zoology 129(3):187–240. Levi, H. W. 1963c. American spiders of the genus Theridion. Bulletin of the Museum of Comparative Zoology 129(10):481–589. Levi, H. W. 1963d. The American spider genera Spintharus and Thwaitesia. Psyche 70(4):223–234. Levi, H. W. 1963e. The American spiders of the genus Anelosimus. Transactions of the American Microscopical Society 82(1):30–48. Levi, H. W. 1964a. American spiders of the genus Episinus. Bulletin of the Museum of Comparative Zoology 131(1):1–25. Levi, H. W. 1964b. American spiders of the genus Phoroncidia. Bulletin of the Museum of Comparative Zoology 131(3):65–86. Levi, H. W. 1966. American spiders of the genera Theridula and Paratheridula. Psyche 73(2):123–130. Levi, H. W. 1968. The spider genera Gea and Argiope in America. Bulletin of the Museum of Comparative Zoology 136(9):319–352. Levi, H. W. 1970. The ravilla group of the orbweaver genus Eriophora in North America. Psyche 77(3):280–302. Levi, H. W. 1971. The diadematus group of the orb-weaver genus Araneus north of Mexico. Bulletin of the Museum of Comparative Zoology 141(4):131–179. Levi, H. W. 1972. The orb-weaver genera Singa and Hypsosinga in America. Psyche 78(4):229–256. Levi, H. W. 1973. Small orb-weavers of the genus Araneus north of Mexico. Bulletin of the Museum of Comparative Zoology 145(9):473–552. Levi, H. W. 1974. The orb-weaver genera Araniella and Nuctenea. Bulletin of the Museum of Comparative Zoology 146(6):291–316.

114

Biogeography of the West Indies: Patterns and Perspectives

Levi, H. W. 1975. The American orbweaver genera Larinia, Cercidia, and Mangora north of Mexico. Bulletin of the Museum of Comparative Zoology 147(3):101–135. Levi, H. W. 1976. The orb-weaving genera Verrucosa, Acanthepeira, Wagneriana, Acacesia, Wixia, Scoloderus and Alpaida north of Mexico. Bulletin of the Museum of Comparative Zoology 147(8):351–398. Levi, H. W. 1977a. The American orb-weaver genera Cyclosa, Metazygia and Eustala north of Mexico. Bulletin of the Museum of Comparative Zoology 148(3):61–127. Levi, H. W. 1977b. The orb-weaver genera Metepeira, Kaira and Aculepeira in America north of Mexico. Bulletin of the Museum of Comparative Zoology 148(5):185–238. Levi, H. W. 1980a. The orb-weaver genus Mecynogea, the subfamily Metinae and the genera Pachygnatha, Glenognatha and Azilia of the subfamily Tetragnathinae north of Mexico. Bulletin of the Museum of Comparative Zoology 149(1):1–74. Levi, H. W. 1980b. Two new species of the genera Theridion and Achaearanea from North America. Transactions of the American Microscopical Society 99(3):334–337. Levi, H. W. 1981. The American orb-weaver genera Dolichognatha and Tetragnatha north of Mexico. Bulletin of the Museum of Comparative Zoology 149(5):271–318. Levi, H. W. 1985. The spiny orb-weaver genera Micrathena and Chaetacis. Bulletin of the Museum of Comparative Zoology 150(8):429–618. Levi, H. W. 1993. American Neoscona and corrections to previous revisions of Neotropical orb-weavers. Psyche 99(2–3):221–239. Levi, H. W. 1993. The orb-weaver genus Kaira. Journal of Arachnology 21:209–225. Levi, H. W. 1995. The Neotropical orb-weaver genus Metazygia. Bulletin of the Museum of Comparative Zoology 154(2):63–151. Levi, H. W. 1996. The American orb weavers of the genus Gasteracantha. Bulletin of the Museum of Comparative Zoology 155(3):89–157. Levi, H. W. 1997. The American orb weavers of the genera Mecynogea, Manogea, Kapogea and Cyrtophora. Bulletin of the Museum of Comparative Zoology 155(5):215–255. McCrone, J. 1963. Taxonomic status and evolutionary history of the Geolycosa pikei complex in the southeastern United States. American Midland Naturalist 70(1):47–73. McCrone, J. and H. W. Levi. 1964. North America widow spiders of the Latrodectus curacaviensis group. Psyche 71(1):12–27. Platnick, N. 1975a. A revision of the Holarctic spider genus Callilepis. American Museum Novitates 2573:1–32. Platnick, N. 1975b. A revision of the spider genus Eilica. American Museum Novitates 2578:1–19. Platnick, N. and J. Murphy. 1984. A revision of the spider genera Trachyzelotes and Urozelotes. American Museum Novitates 2792:1–33. Platnick, N. and M. U. Shadab. 1975a. A revision of the spider genera Haplodrassus and Orodrassus in North America. American Museum Novitates 2583:1–40. Platnick, N. and M. U. Shadab. 1975b. A revision of the spider genera Drassodes and Tivodrassus in America. American Museum Novitates 2593:1–29. Platnick, N. and M. U. Shadab. 1975c. A revision of the spider genus Gnaphosa in America. Bulletin of the American Museum of Natural History 155(1):1–66. Platnick, N. and M. U. Shadab. 1976. A revision of the spider genera Rachodrassus, Sosticus and Scopodes in North America. American Museum Novitates 2594:1–33. Platnick, N. and M. U. Shadab. 1977. A revision of the spider genera Herpyllus and Scotophaeus in North America. Bulletin of the American Museum of Natural History 159(1):1–44. Platnick, N. and M. U. Shadab. 1980a. A revision of the North American spider genera Nodocion, Litopyllus, and Synaphosus. American Museum Novitates 2691:1–26. Platnick, N. and M. U. Shadab. 1980b. A revision of the spider genus Cesonia. Bulletin of the American Museum of Natural History 165(4):335–386. Platnick, N. and M. U. Shadab. 1981. A revision of the spider genus Sergiolus. American Museum Novitates 2717:1–41. Platnick, N. and M. U. Shadab 1982. A revision of the American spiders of the genus Drassyllus. Bulletin of the American Museum of Natural History 173(1):1–97. Platnick, N. and M. U. Shadab. 1983. A revision of the American spiders of the genus Zelotes. Bulletin of the American Museum of Natural History 174(2):97–192.

The Contribution of the Caribbean to the Spider Fauna of Florida

115

Platnick, N. and M. U. Shadab. 1988. A revision of the American spiders of the genus Micaria. American Museum Novitates 2916:1–64. Reiskind, J. 1987. Status of the Rosemary Wolf Spider in Florida. Technical Report No. 28. Cooperative Fish and Wildlife Research Unit:1–13. Reiskind, J. and P. Cushing. 1996. A morphological study of a narrow hybrid zone between two wolf spiders, Lycosa ammophila and Lycosa ericeticola in north Florida. Revue Suisse de Zoologie, volume hors series:543–554. Wallace, H. K. 1942a. A revision of the burrowing spiders of the genus Geolycosa. American Midland Naturalist 27(1):1–62. Wallace, H. K. 1942b. A study of the lenta group of the genus Lycosa, with descriptions of new species. American Museum Novitates 1185:1–21. Wallace, H. K. 1947. A new wolf spider from Florida, with notes on other species. Florida Entomologist 30(3):33–38. Wallace, H. K. 1950. On Tullgren’s Florida spiders. Florida Entomologist 33(2):71–83. Wallace, H. K. and H. Exline. 1978. Spiders of the genus Pirata in North America, Central America and the West Indies. Journal of Arachnology 5:1–112. Webb, S. D. 1990. Historical biogeography. Pp. 70–100 in Myers, R. L. and J. J. Ewel (eds.). Ecosystems of Florida. University of Central Florida Press, Orlando.

Beetles 9 Rhysodine in the West Indies Ross T. Bell Abstract — The rhysodine beetles (Carabidae) are well represented in the West Indies by 17 species. They are of zoogeographical interest because of their natural history. The strategies for the separate invasions from North and South America are documented with fossil evidence. Relationships of the West Indian genera and interpretation to their distribution are discussed. Rafting seems to be the means of travel of these insects.

INTRODUCTION The Rhysodini comprise a group of over 300 species of highly modified ground beetles (Carabidae). They are of interest in zoogeography because the way of life of these beetles gives them an excellent chance of rafting across water barriers, while their chances of spreading by other means are unusually limited. The group is well represented in the West Indies, and there is evidence of at least four, and possibly six, separate invasions. Rhysodines are long, narrow beetles between 4 and 10 mm long. They appear red-brown in bright light and piceous to black in dimmer light; color is no help in separating the species. The antennae are relatively short for Carabidae, but moniliform (like chains of beads). A large hollow space within the head is connected to the exterior by a system of grooves, which divide the dorsal surface of the head into several lobes. The mentum, or lower lip, projects so far forward that it is visible in dorsal view, and entirely conceals the mandibles in ventral view. There is a ball-like “neck” or condyle between the head and the prothorax. The males have “calcars,” anteriorly directed processes at the apex of the hind, and usually also the middle tibiae, which make it easy to separate the sexes. Both larvae and adults are found within dead wood. The larvae live in short tunnels that are backfilled with wood chips, but the adults do not tunnel. An unusually thick exoskeleton and hyperdeveloped muscles enable them to force themselves into the wood, evidently by compressing the wood cells. The wood must be moist, but it can be surprisingly sound. It is amazing to see one of these beetles disappear into the wood without leaving a visible trace of its passage. Collecting rhysodines is very laborious and difficult, involving the use of saws, wedges, crowbars, and axes. The beetles can be deep within largely sound logs, in the centers of stumps, deep underground in large roots, or in small soft areas in large branches high in trees. Some species are encountered beneath bark on occasion, the basis for the inappropriate common name “wrinkled bark beetles.” If a single beetle is encountered, it usually pays to make a careful dissection of the log or root in which it is found, as rhysodines seem to be colonial, with up to 50 individuals sometimes found together. The larvae are often found in the vicinity of the adults. The soft-bodied larva lives in a short tunnel, which it fills in behind itself with wood chips. Sometimes more than one species of adult are found in a single log, and larvae may not be conspecific with adults found near them. Rhysodini adults have been seen to feed on slime molds (myxomycetes), using highly modified mouthparts (Bell, 1994), and attack the plasmodium stage within the wood. They are limited to forest regions where there is enough rainfall to permit the decay of wood and the growth of slime mold. In the tropics, they seem largely restricted to rain forests, and they are absent from areas with a pronounced dry season. In the temperate zones, they are less limited.

0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

117

118

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 New World Rhysodine Fauna

Genera Plesioglymmius (Ameroglymmius) Omoglymmius (Boreoglymmius) Clinidium (Arctoclinidium) Clinidium (Mexiclinidium) Clinidium (Protainoa) Clinidium (Tainoa) Clinidium (s. str.) Rhyzodiastes (s. str.) Rhyzodiastes (Rhyzostrix) Neodhysores

North America

Central America

West Indies

North Andes

South America South of the Amazon

– + + + – – – – – –

– – – + – – + – – –

+ – – – + + + – – –

+ – – – – – + – – –

+ – – – – – – + + +

DISPERSAL MECHANISMS Walking is one possible dispersal mechanism. It can be quite effective, as is shown by such carabid groups as genus Carabus and Tribe Cychrini. However, records for Rhysodini walking at a distance from logs are about as rare as flight records for fully winged species, being limited to a few pitfall captures. Walking abilities seem to be limited to moving from one log or stump to another nearby one in the same patch of forest. Examination of a live rhysodine shows just how inefficient is its walking ability. Unlike other carabids, a rhysodine cannot run or even walk moderately fast. The animal seems to have only a single ponderous, “low gear” gait. In addition, when the animal is away from wood, it is poorly balanced, and it frequently capsizes. Poor walking ability helps to account for the great number of species in the Andean region, for example, and subsequently their small ranges. Isolated areas of forest on continents are islands as far as rhysodines are concerned. Indeed, islands of forest that are not connected by rivers may be harder to reach than true islands for flightless species. There is a dramatic illustration of this in South America (Table 1, Figure 1). There are two distinct South American faunas, with only the winged genus Plesioglymmius present in both. The Andean region, from Ecuador to central Venezuela, has genus Clinidium, subgenus Clinidium, shared with Central America and the West Indies, a member of a Laurasian (Northern Hemisphere) genus, while eastern Brazil and northern Argentina have Rhyzodiastes, subgenera Rhyzodiastes s. str. and Rhyzostrix, members of a Gondwanian (Southern Hemisphere) genus (Figure 2). The two faunas meet along the Amazon River. Both genera occur on both banks, where rafting can carry them from one side to the other, but there are no records of Rhyzostrix north, away from the river, or records of Clinidium south from it. This presumably reflects the fact that logs do not raft upstream and that rhysodines spread very slowly without the help of rafting. Another common dispersal method for insects is flight. Over half the species of rhysodines have fully developed hind wings, and presumably can fly, but there are extremely few records of them flying or appearing at light traps. Such records as there are have been entirely within the forests in which the beetles live. The extremely thick, heavy exoskeleton must make them very slow, weak fliers. I suspect that flight is probably used mainly to reach dead wood in high trees. In the Southwest Pacific region, where there are many insular rhysodines, the fully winged taxa have not done better than vestigial winged taxa at reaching islands (the latter group belong to old, long flightless taxa, so the beetles could not have lost their wings after reaching the islands). In the West Indies all species except one belong to the totally flightless genus Clinidium. The exception

Rhysodine Beetles in the West Indies

119

FIGURE 1 Distribution of genera of Rhysodini in the Neotropical region.

FIGURE 2 Habitus, dorsal aspect: (A) Clinidium (Protainoa) extrarium B&B; (B) Clinidium (Tainoa) curvicosta Chev.; (C) Clinidium (s. str.) guildingii Kirby; (D) Plesioglymmius (Ameroglymmius) compactus B&B.

120

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 West Indian Species (by Islands) Species

Island

Subtribe: Omoglymmius Plesioglymmius (Ameroglymmius) compactus B&B Subtribe: Clinidiina Clinidium (Protainoa) extrarium B&B C. (Tainoa) curvicosta Chevrolat C. (Tainoa) chevrolati Reitter C. (Tainoa) xenopodium Bell C. (Tainoa) darlingtoni Bell C. (s. str.) incudis Bell C. (s. str.) planum Chevrolat C. (s. str.) smithsonianum B&B C. (s. str.) microfossatum B&B C. (s. str.) guildingii Kirby C. (s. str.) haitiense Bell C. (s. str.) corbis Bell C. (s. str.) jamaicense Arrow C. (s. str.) chiolino Bell C. (s. str.) trionyx Bell C. (s. str.) boroquense B&B

90º

80º

Eastern Cuba Western Cuba Eastern Cuba Eastern Cuba Central Dominican Republic Jamaica Puerto Rico Guadeloupe Dominica Martinique St. Vincent Haiti (South, mountains) Hispaniola (South and North) Jamaica (mountains) Jamaica Dominican Republic Puerto Rico

70º

Protainoa Tainoa Cl. s. str. jamaicense complex Cl. s. str. others Plesioglymmius Cl. s. str. guildingi group

20º

10º

200 Km 200 Mi.

80º

70º

60º

FIGURE 3 Distribution of Rhysodini in the West Indies.

is P. compactus, which is found on only one island, Cuba. All flightless species have reduced and highly modified eyes. Rafting within floating logs seems to be the one remaining possible means of long-distance transport. A floating log would seem to be an effective colonizing mechanism with a population

Rhysodine Beetles in the West Indies

121

of several to many rhysodines together with their slime mold food source. All West Indian rhysodines except one species belong to the subtribe Clinidiina, and to the genus Clinidium Kirby. This genus and its sister genus Rhyzodiastes have the hind wings reduced to minute vestiges, and have evidently been flightless for tens of millions of years. A third genus, Grouvellina Bell & Bell, is fully winged and no doubt flies, but it is confined to Madagascar. Rafting seems to be the only possible means for Clinidium to reach the West Indies. The rhysodine fauna of the West Indies was first monographed by Bell (1970). Bell and Bell (1978, 1979, 1982, 1985) treated the group on a world basis. Bell and Bell (1995) treated the Cuban fauna. Other papers of zoogeographical interest cover Australia (Bell and Bell, 1991) and Micronesia (Bell and Bell, 1981). Table 2 shows the 17 species of rhysodines found in the West Indies. Figure 1 shows representatives of the genera and subgenera of West Indian rhysodines, and Figure 3 shows their distributions.

RELATIONSHIPS OF WEST INDIAN GENERA WITHIN THE WORLD FAUNA Two of the seven subtribes of Rhysodini are present in the West Indies. Subtribe Omoglymmiina, with seven genera, is nearly worldwide in distribution, but is absent from Madagascar, New Zealand, and most of Africa and Australia. Genus Plesioglymmius B&B has three subgenera, two in and near Indonesia, and the third in South America, in Brazil and Venezuela. Subgenus Ameroglymmius has three species, one in eastern and southern Brazil, one in Amazonia and the Orinoco region, and the third in Cuba. Subtribe Clinidiina, with three genera, is also nearly worldwide, but is absent from Africa although represented by 15 species in Madagascar. Genus Grouvellina Bell & Bell is restricted to Madagascar. Genus Rhyzodiastes Fairmaire has a Gondwanian distribution from eastern and southern South America, Australia, New Zealand, Indonesia, and Indo China. Genus Clinidium is a Laurasian genus from North America, Japan, Europe, Central America, the West Indies, and the Andean region of South America. All species are vestigial-winged. The genus includes five subgenera, one in the United States and southern Europe, one in Mexico and Guatemala, two endemic in the West Indies, and one in Andean South America, Central America, and the West Indies. Subgenus Protainoa B&B has one species in western Cuba. Subgenus Tainoa Bell has five species in the Greater Antilles. Subgenus Clinidium s. str. has over 50 species in Andean South America, Greater and Lesser Antilles, and Central America from Guatemala south. Of the six species groups, two are present in the West Indies. Beccarii Group. Three species in Central America, Guatemala to Panama; two additional species with dubious locality labels; an unnamed fossil species from Hispaniola (see below, Fossil Evidence); C. incudis Bell of Puerto Rico (formerly placed in the Granatense Group). Guildingii Group. Many species from Andean South America, lower Central America (north to Costa Rica), and the West Indies. Divided into four or five sections, of which two reach the West Indies. Oberthueri Section. Six species. Greater Antilles except Cuba. Guildingii Section. Four species. Lesser Antilles.

122

Biogeography of the West Indies: Patterns and Perspectives

FOSSIL EVIDENCE Recently, through the courtesy of Dr. George Poinar, I was able to study two rhysodines in amber from the Dominican Republic, both males, and probably conspecific. The age of Dominican amber is estimated to be between 15 and 45 million years (Poinar, 1999). They are in Clinidium s. str., and appear to belong to the beccarii species group, with which they agree in having the anterior pronotal pit greatly enlarged, but without a central tubercle, and in having a long, acute antennal stylet. Two other features of the group are not visible, however: these would be the loss of the minor setal tufts on antennomeres V and VI, and the modified compound eyes, which are bilobed or divided into anterior and posterior units. Within the beccarii group, the amber species most resembles C. moldenkei Bell & Bell and C. sulcigaster Bell, both Central American species. As currently defined, the beccarii group is restricted to Central America (Guatemala to Panama), and is not known from the West Indies. However, C. incudis Bell of Puerto Rico, currently listed in the granatense group, probably belongs here. It has minor setae tufts on antennomeres VII–X. Such tufts were absent in all members of the beccarii group until the recent discovery of C. gilloglyi Bell & Bell, in which the tufts are located exactly as in C. incudis. The latter species also has a single ocelliform eye on each side, rather than the two seen in most members of the beccarii group. Clinidium incudis could have secondarily suppressed the development of the posterior ocelliform unit. Thus the fossil species and C. incudis could be related. This leaves open the question of whether or not their ancestors separated after reaching the Antilles. I am deferring naming the fossil form in hopes that more specimens will turn up, particularly one that will allow study of the eye structure.

INTERPRETATIONS OF WEST INDIAN DISTRIBUTIONS Interpreting the distribution of island faunas makes it necessary to estimate the probabilities of organisms reaching islands by various means of dispersal. Among Carabidae as a whole, flight, passive aerial transport, and rafting are all possible means of travel, but it is difficult to sort out which means were used in particular cases. With rhysodines the situation is more clear-cut, and rafting seems the only likely method. Relationships of West Indian rhysodines to mainland (Central and South America) species are listed in Table 3. Genus Plesioglymmius (Ameroglymmius) compactus. The restriction of this species to Cuba is enigmatic (Bell, 1995). On the mainland, Ameroglymmius is unknown either in Central America or in the Andean region, and the nearest record is from Suriname (Bell and Bell, 2000).

TABLE 3 Closest Mainland Taxa to West Indian Rhysodines West Indian Taxa Subtribe: Omoglymniina Plesioglymmius compactus Subtribe: Clinidiina Protainoa and Tainoa Clinidium planum, guildingii, smithsonianum, microfossatum C. incudis C. boroquense C. trionyx C. jamaicense complex

Related Mainland Taxa P. reichardti; Amazonia, Guyana, Venezuela Clinidium (Mexiclinidium); Mexico, Guatemala, 9 spp. C. rojasi and 3 related spp.; easternmost Andes of Venezuela C. moldenkei and C. sulcigaster; Guatemala to Panama C. oberthueri; Ecuador C. whiteheadi or C. alleni; Panama or C. jolyi or C. pilosum; western Venezuela C. oberthueri; Ecuador

Rhysodine Beetles in the West Indies

123

Genus Clinidium. This is a Northern Hemisphere genus, and the three subgenera found in the West Indies must ultimately be of North American origin. Subgenus Protainoa — C. extrarium, the only species, is an extraordinary relict, surviving in a minute area of suitable habitat in western Cuba, where no other rhysodines have been recorded. It may have evolved in place, and might date from before smaller islands assembled to make Cuba. Subgenus Tainoa — Endemic to the Greater Antilles, and the sister group to subgenus Protainoa. It probably originated on an island later incorporated into Cuba or Hispaniola. The absence of Tainoa from Puerto Rico is significant. The subgenus appears to have two sister groups, curvicosta–chevrolati and xenopodium–darlingtoni. The former is only in eastern Cuba, and the species appears to have separated relatively recently while the latter appears to have divided much earlier, with one species in Hispaniola and the other in Jamaica. Jamaica is believed to have been submerged for part of the Miocene, so the likely origin of C. darlingtoni is by rafting from Hispaniola to Jamaica in late Miocene or Pliocene. Subgenus Clinidium s. str. — Most closely related to Arctoclinidium of temperate North America, but separated geographically today by areas occupied by Mexiclinidium, Tainoa, and Protainoa. The point of origin is not obvious. South America has the most species, but the subgenus scarcely penetrates beyond the Andean region, and the folding of the Andes is fairly recent, especially in the east. Parts of Central America may be older, but the area is not favorably situated for rafting to the West Indies, which are predominantly upwind and upcurrent. Hurricanes, however, might transport logs contrary to the normal directions of winds and currents. A final possibility is origin in the islands with subsequent spread to areas that are now mainland. This will seem farfetched to workers on other groups. However, there is evidence of movements of Rhysodine groups from islands to the mainland in the area of Indonesia, while the restricted abilities of rhysodines to spread on continents do not appear to give any advantage to mainland species over insular ones. Clinidium (s. str.) beccarii Group. The fossil specimens cited above demonstrate that this group once inhabited Hispaniola. The living Puerto Rican species C. incudis Bell now appears to be a modified member of this group. It could have evolved from West Indian species also ancestral to the amber species, or the two could represent two separate invasions. It is also possible that the beccarii group ancestor evolved in the West Indies and later invaded Central America. Clinidium guildingii Group, guildingii Section. There are four very similar species in the Lesser Antilles, from Guadeloupe, Dominica, St. Vincent, and Martinique. They are obviously closely related to the rojasi Section, which is limited to coastal mountains of central and eastern Venezuela. The most likely scenario would be that the ancestor of the guildingii section was pushed northward in a hurricane to the islands. Clinidium guildingii Group, oberthueri Section. Many mainland species from western Venezuela to Ecuador and Costa Rica. Six species, Greater Antilles except for Cuba. (I formerly separated West Indian species as the “jamaicense section.” This no longer appears valid.) Four species form the closely related “jamaicense complex” of Jamaica and Hispaniola; two more isolated species are C. boroquense Bell of Puerto Rico and C. trionyx Bell & Bell of Hispaniola. Both appear related to oberthueri of Ecuador. If the section had a mainland origin, the Antillean species could represent one, two, or three invasions.

CONCLUSION Rhysodines have the potential to contribute insights into biogeographical studies because, unlike most other insects, they are restricted to rafting as a means of inter-island dispersal. Because of this, all West Indian rhysodine distributions can be explained by dispersal, and none needs to be attributed to vicariance. Rhysodine distributions reflect prevailing theories that the islands have been moving eastward relative to the American continents (Donnelly, 1988, 1989; Williams, 1989). West Indian rhysodines must have arrived over long intervals, as they differ from mainland forms in varying degrees, from two endemic subgenera to some species with similar, but distinct

124

Biogeography of the West Indies: Patterns and Perspectives

mainland relatives. In the most numerous group, Clinidium s. str., the situation is puzzling because the genus to which it belongs is a northern one, and the subgenus must have come from North America and is limited to the Andean region in South America. As the Andes, especially the northern part, is a relatively young mountain range, the Venezuela section may have a rhysodine fauna no older than the West Indian one, and thus be unable to serve as a source for the latter. The northern and northeastern Andes have many species of Clinidium s. str. and those mostly have very limited ranges, as though they may have differentiated and dispersed when the Andes were at least partly insular. An interesting point is the status of Puerto Rico, which lacks the two most speciose groups, Tainoa, and the Jamaicense complex of Clinidium s. str. This mirrors the situation in the more conventional Carabidae (Liebherr, 1988), where Puerto Rico lacks Platynus but has the relict genera Barylaus and Antilloscaris. Jamaica is known to have been completely inundated in the Miocene, yet it has a species of the subgenus Tainoa. This establishes the proposition that an ancestor rafted from Hispaniola since that time. It is also probable that the ancestor of the two Jamaican species of the Jamaicense complex of Clinidium s. str. moved in the same direction at a roughly similar time. More information will help establish a clearer picture in the future. Probably most of the West Indian species are already known, but there is one old species name “on the books” from Cuba, C. humeridens Chevrolat. We have been unable to locate the type specimen, and the original description is sufficient to show that the species is distinct from any currently known one, but is not sufficient to put it into the modern classification. There should also be relatives of C. guildingii on St. Lucia and Grenada. We badly need a more complete knowledge of the fauna of the Andean region, as well as Central America north of Costa Rica. At present, no rhysodines are known from Nicaragua, Salvador, Honduras, or Belize. Of course, more fossils and more definite information from geologists will also help.

LITERATURE CITED Bell, R. T. 1970. The Rhysodini of North America, Central America, and the West Indies. Miscellaneous Publications of the Entomological Society of America 6(6):289–324. Bell, R. T. 1994. Beetles that cannot bite; functional morphology of the head of adult rhysodines (Coleoptera: Carabidae or Rhysodidae). The Canadian Entomologist 126:667–672. Bell, R. T. and J. R. Bell. 1978. Rhysodini of the world. Part I. A new classification of the tribe, and a synopsis of Omoglymmius subgenus Nitiglymmius new subgenus (Coleoptera: Carabidae or Rhysodidae). Quaestiones Entomologicae 14(1):43–48. Bell, R. T. and J. R. Bell. 1979. Rhysodini of the world. Part II. Revisions of the smaller genera (Coleoptera: Carabidae or Rhysodidae). Quaestiones Entomologicae 15(4):377–446. Bell, R. T. and J. R. Bell. 1981. Insects of Micronesia. Coleoptera: Rhysodidae. B. P. Bishop Museum, Honolulu 15(2):51–67. Bell, R. T. and J. R. Bell. 1982. Rhysodini of the world. Part III. Revisions of Omoglymmius Ganglbauer (Coleoptera: Carabidae or Rhysodidae) and substitutions for preoccupied generic names. Quaestiones Entomologicae 18(1–4):127–259. Bell, R. T. and J. R. Bell. 1985. Rhysodini of the world. Part IV. Revisions of Rhyzodiastes Fairmaire and Clinidium Kirby, with new species in other genera (Coleoptera: Carabidae or Rhysodidae). Quaestiones Entomologicae 21:1–172. Bell, R. T. and J. R. Bell. 1991. The Rhysodini of Australia (Insecta: Coleoptera: Carabidae or Rhysodidae). Annals of Carnegie Museum 60:179–210. Bell, R. T. and J. R. Bell. 1995. The Rhysodini (Insecta: Coleoptera: Carabidae) of Cuba. Annals of Carnegie Museum 64(3):185–195. Bell, R. T. and J. R. Bell. 2000. Rhysodine beetles (Insecta: Coleoptera: Carabidae): New species, new data II. Annals of Carnegie Museum 69(2):69–91. Donnelly, T. W. 1988. Geologic constraints of Caribbean biogeography. Pp. 15–37 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York.

Rhysodine Beetles in the West Indies

125

Donnelly, T. W. 1989. History of marine barriers and terrestrial connections; Caribbean paleogeographic inference from pelagic sediment analysis. Pp. 103–117 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Liebherr, J. K. 1988. Biogeographic patterns of West Indian Platynus carabid beetles. Pp. 121–152 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Cornell University Press, Ithaca, New York. Poinar, G. O. and R. Poinar. 1999. The Amber Forest: A Reconstruction of a Vanished World. Princeton University Press, Princeton, New Jersey. Williams, E. E. 1989. Old problems and new opportunities in West Indian biogeography. Pp. 1–46 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

Biogeography 10 The of the West Indian Butterflies (Lepidoptera): An Application of a Vicariance/Dispersalist Model Jacqueline Y. Miller and Lee D. Miller Abstract — Previous biogeographical theories and models for West Indian butterfly biogeography are reviewed. The age of butterflies, endemism of the West Indian species, and their propensities for, or possible impediments with, dispersal are discussed. A combination vicariance/dispersal model for the evolution of the Antillean butterfly fauna originally proposed in 1989 is reviewed and updated within the general constraints of current geological evidence. Additional supportive documentation for biodiversity of butterflies in the Caribbean, and particularly for the geological history of Jamaica, the southern Hispaniolan block, the Bahama Islands, and the Lesser Antilles, is also presented.

INTRODUCTION Early biogeographical hypotheses were based on the belief that landmasses and seafloors have long been in their present positions. Wallace (1876), Matthew (1915), Simpson (1952), and Darlington (1957, 1965) provide arguments for this interpretation in detail. Accordingly, organisms were distributed on these landmasses or oceans by the action of random dispersal through time. Still other biogeographers have felt it necessary to construct land bridges to distribute faunal and floral elements that could not be easily explained by other means (Schuchert, 1935). Others were not so certain. The paleogeographical reconstructions of Wegener (1915) and du Toit (1927, 1937), which utilized movable landmasses and seafloors, spurred the interest of a number of geologists (Carey, 1958; Wilson, 1963) and biologists (Cain, 1944; Croizat, 1958; Cracraft, 1973; Raven and Axelrod, 1974; Shields and Dvorak, 1979). Croizat’s significant contribution during the 1950s and later was the realization that entire biotas evolved and might be distributed in patterns that were not necessarily as random as had been assumed previously. Croizat postulated “tracks” along which entire biotas could be shown to have moved and evolved. By carefully plotting the ranges of diverse organisms, Croizat determined that some of these tracks showed frequent congruent distribution patterns in such diverse groups as insects, trees, freshwater fish, and reptiles. Certain broad, repeated distributional patterns in fauna and flora he referred to as “generalized tracks.” Croizat et al. (1974) stressed that all species are components of biotas and that the generalized track estimates the composition and distribution of the ancestral biota before it subdivided into descendant biotas. The arguments for vicariance biogeography in its various forms are found in Hennig (1966); Brundin (1972, 1981); Raven and Axelrod (1974); Rosen (1976, 1985); Savage (1973, 1974, 1983); Patterson (1981); and elsewhere. The general vicariance rationale is well articulated by Savage (1983:491–496).

0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

127

128

Biogeography of the West Indies: Patterns and Perspectives

It is generally agreed (Coney, 1982; D. L. Smith, 1985; and others) that the opening of the Gulf of Mexico and the Atlantic dates roughly from the Jurassic, and the floors of both comprised marine crustal oceanic basalt by the end of that period (Perfit and Heezen, 1978; Case et al., 1983; Perfit and Williams, 1989). Further, North and South America were totally separated early in the Cretaceous, but a connection re-formed later in that period from a series of volcanic islands connecting southern Mexico and northern South America. Utilizing the geological data summarized by Pindell and Dewey (1982), Guyer and Savage (1986) provided a cladistic analysis of the animals and presented a very plausible explanation for the distribution of anoles in the Americas. In this model, the proto-Greater Antilles were part of the Caribbean plate that formed the early Central American connection between North and South America in the Paleocene, fragmenting away in the Eocene. These fragments drifted northeastward to become the Greater Antilles. These islands, however, may be more closely related to North than to South America by their original proximity to Yucatan and later by collision with the North American plate underlying what is now Florida and the Bahamas Rise. The Greater Antilles were subject to considerable modification during the Tertiary through the complex actions of many faults and by the independent movement of several smaller plates. This fragmentation of the proto-Greater Antilles and subsequent accretion of parts elsewhere results in at least three separate blocks forming Cuba and three or four additional separate blocks accreting to form Hispaniola, each contributing its own elements to the fauna. Parts of Hispaniola were attached either to Puerto Rico or to parts of Cuba during the Tertiary, initiating further problems in the interpretation and origin of fauna and flora. Maury et al. (1990) reviewed the geological history of the Lesser Antilles. Although originally considered to be a single volcanic arc of Eocene age, the Lesser Antilles comprised the Southern Volcanic Caribbees and the two northern arcs: “Northern Volcanic Caribbees” or inner arc from Dominica to Bass Terre, Guadeloupe and the outer arc or the “Northern Limestone Caribbees,” which includes Grande Terre, Guadeloupe to Anguilla. The latter insular group has a more ancient geological history and volcanism, and these “limestone” islands appear to have a more recent terrestrial fauna with the islands of Grande Terre and Marie Galante emerging during the Pleistocene. Although there are some endemic taxa on these islands, the majority of the biota, particularly that south of Dominica, retains greater similarity to South America than to any other landmass. The modern connection between Central and South America initiated during the Cretaceous extended southward from southwestern Mexico as a complex volcanic arc–trench system, finally closing the gap between North and South America during the Pliocene, thus facilitating the “Great American faunal interchange” (Stehli and Webb, 1985). The recent geological history of Florida and the Bahamas in association with Cuba is complex. Burke (1988) indicated that during the mid-Tertiary, Cuba overrode the Bahamas rise. During periods of the Pleistocene glacial maxima, there was a large emergent Great Bahamas Bank with subsequent reflooding of the bank in the insular pattern evident today. Similarly, Florida underwent an extension and reduction of land area (Webb, 1990) with a further exchange of the fauna and flora.

PREVIOUS BIOGEOGRAPHICAL STUDIES OF BUTTERFLIES It is informative to examine previous studies on butterfly biogeography in light of the various geological and biogeographical models. The advent of Darlington’s (1957) general treatise on zoogeography provided the impetus for several papers on the biogeography of the West Indian butterflies. Like the Darlington volume, most of the butterfly studies to 1989 were based on the dispersalist model. Even the monumental work of MacArthur and Wilson (1967) was interpreted on a dispersalist model. Fox (1963) completed one of the earliest biogeographical studies on a few species of West Indian Ithomiidae. He showed that these were most closely related to Central American species; however, since the Ithomiidae are not vagile, he resorted to Schuchert’s (1935) land bridges to transport them onto the Greater Antilles. Fox, like Schuchert, attributed these land bridges to the Tertiary and gave very little significance to the Pleistocene. Shortly thereafter, Clench (1964) published

The Biogeography of the West Indian Butterflies (Lepidoptera)

129

an analysis of the West Indian Lycaenidae. This was a purely Matthewian treatment ascribing virtually all of the West Indian fauna to dispersal events during Pleistocene interglacials with contraction and subsequent extinction of the fauna during glacial maxima and the concomitant lowering of temperature. The species themselves were largely ascribed to the Pleistocene with a similar age for the evolution of the entire family. L. Miller (1965) examined the vagile West Indian Choranthus (Hesperiidae) and its relatives. These he considered to have dispersed onto the Greater Antilles largely during the Tertiary, but he did not invoke land bridges, relying instead on rare, chance dispersals. The single Antillean species of Paratrytone was attributed to invasion from approximately present-day Honduras. L. Miller (1968) briefly mentioned the West Indies in his treatment of the zoogeography of the world Satyridae, again utilizing a dispersalist model. The only West Indian satyrid genus, Calisto Huebner, was thought to have been derived from a Middle American member of a wholly Neotropical tribe. Brown and Heineman (1972) discussed Jamaica’s butterflies on a classical dispersalist model via the Yucatan Channel to Cuba and from South America to the Lesser Antilles, but they attributed some of the dispersal to events during the Tertiary. Scott (1972) listed the West Indian butterflies and made a few comments on their distribution. This, too, was a dispersalist discussion and strongly influenced by events of the Pleistocene. Scott (1986) again listed the West Indian butterflies and commented more widely on their postulated dispersal from the American continent. Riley (1975) in his Field Guide offered some speculations on the origins of parts of the West Indian butterfly fauna, including a possible African link for one or two genera. Riley’s was not a vicariance study, however, and one is left with the impression that he was referring to long-distance dispersal, including Africa to the Antilles, to account for these anomalies. Brown (1978) presented a paper on Antillean butterfly distributions during the same symposium at which Rosen (1976) presented his classic vicariance model for the Antillean fish fauna. Brown raised the possibility of vicariance, although he did not elaborate further, hinting at it only in connection with the danaid genus Anetia. He also mentioned Calisto as a possible link to African fauna, basically repeating Riley’s information. Shields and Dvorak (1979), although interesting but controversial, proposed the first, definitive vicariance model for butterflies based on the late Jurassic to early Cretaceous separation of the Americas, Africa, and the Caribbean. This model, which ascribes most of the evolution of the butterflies to this time period, was totally at odds with the conventional dispersalist model, and since some of the taxonomic relationships that were postulated are questionable, it has not achieved much acceptance. Much of the emphasis in these earlier studies was on the Pleistocene with the exception of Fox (1963), L. Miller (1965), Shields and Dvorak (1979), and, to a lesser degree, Brown (1978). This emphasis is inherent in adoption of the Matthew–Darlington model of biogeography. It requires the long-distance transport of organisms across barriers to establish themselves in new territories. Perhaps nowhere are the difficulties of such transport more apparent than in colonization of islands, and why some species are present on certain islands and absent on others. Other variables, such as the close association of host plants for butterflies and other insects present other unique problems. Several factors need to be considered to determine the feasibility of the dispersalist model. First, one must examine the organisms themselves and their endemicity. It is also necessary to analyze the fossil record to determine the actual documented ages of butterflies. Since there are very few butterfly fossils available, what may we infer from those that are extant? Other questions that must be addressed involve dispersal itself. Are all butterflies excellent dispersalists or not? In the case of disparate groups of butterflies, are they strong fliers? What is inherent in the lifestyles of butterflies that would either facilitate or discourage dispersal? Is a sedentary butterfly much less likely to leave its localized ecological niche? Some species are well-known migrants, and their dispersal capabilities should be virtually unlimited. If one seeks to postulate dispersal by hurricanes, how do such organisms behave and survive during storms (or even at the threat of them)? We will examine these aspects in some detail.

130

Biogeography of the West Indies: Patterns and Perspectives

ENDEMISM OF THE WEST INDIAN BUTTERFLY FAUNA Endemism is a well-known phenomenon in insular butterflies (Holloway, 1979; Alayo and Hernandez, 1987; Smith et al., 1994), and the butterflies of the West Indies are no exception (Table 1). Over half of the recorded species and nearly one of every eight genera are endemic to the West Indies. This statement is based on the assumption that the southern tip of Florida and the Keys are faunistically Antillean, rather than continental (Scott, 1972, 1986; Riley, 1975; Brown, 1979; Minno and Emmel, 1993; Smith et al., 1994), because some of the Cuban and Bahamian “endemics” also occur there. Since most butterflies are closely associated with specialized larval host plants, it is necessary to examine individual food plant data for the West Indian butterflies (Brown and Heineman, 1972; Scott, 1972; Riley, 1975; Shields and Dvorak, 1979; Smith et al., 1994). It is generally accepted that modern butterflies and their associated larval host plants arose in the mid to late Cretaceous (Raven and Axelrod, 1974; Common, 1975, 1990; Powell, 1980; White, 1990) and that these associations have coevolved since that age. The food plants either had to be on the islands when the butterflies arrived or they arrived on the islands by vicariance along with their butterflies. The endemic genera and the species are clearly derived from several different faunal sources (Table 2). Most of the butterflies of the Greater Antilles are most closely related to those of Central America and Mexico. This portion of the fauna may be explained either as dispersal from the mainland (most previous butterfly biogeographical studies) and/or by invoking a model similar to that proposed by Pindell and Dewey (1982) and refined by Pindell and Barrett (1990) as a basis for vicariance. The clearly North American component of the Antillean butterfly fauna and the fauna of the southern Lesser Antilles, most closely related to South America via Trinidad, are both best explained by dispersal. The Virgin Islands and northern Greater Antilles are faunistically related to Puerto Rico in part, again probably through dispersal. To determine which biogeographical model is most plausible, one must examine the present ranges of the butterflies that inhabit the islands (Table 2) and determine the dispersal potential of the butterflies themselves. Cuba and Hispaniola harbor the greatest percentage of endemic butterflies and a number of other West Indian fauna and flora discussed in this volume, but this may reflect their status as the largest Antillean islands (Alayo and Hernandez, 1987; Schwartz, 1989; Smith et al., 1994). It may equally well be a function of the geological history of these islands as points of accretion of various small plates during the Tertiary, each harboring distinctive faunas, thus conforming to a vicariance model.

TABLE 1 Endemicity of Antillean Butterfly Fauna Family

Genera No.

Endemic No.

% Endemicity

Species No.

Danaidae Ithomiidae Satyridae Nymphalidae Libytheidae Riodinidae Lycaenidae Pieridae Papilionidae Hesperiidae Total

3 1 1 34 1 1 13 13 5 49 121

0 0 1 4 0 1 1 0 0 7 14

0.0 0.0 100.0 11.8 0.0 100.0 7.7 0.0 0.0 14.3 11.6

9 2 25+ 65 3 1 32 50 22 92 301

Endemic No. 5 2 25+ 26 3 1 22 24 15 47 170

Note: Southern Florida is included in the West Indies for purposes of establishing endemicity.

% Endemicity 55.6 100.0 100.0 40.0 100.0 100.0 68.8 48.0 68.2 51.1 56.5

The Biogeography of the West Indian Butterflies (Lepidoptera)

131

TABLE 2 Affinities of West Indian Butterfly Genera Widespread Genera

Affinity with North America

Danaus (4) Lycorea (1) Doxocopa (2) Hypna (1) Memphis (4) Marpesia (3) Colobura (1) Historis (2) Archimestra* (1) Anartia (4) Biblis (1) Vanessa (3) Euptoieta (2) Heliconius (1) Dryas (1) Agraulis (1) Libytheana (4) Strymon (11) Leptotes (2) Brephidium (2) Hemiargus (4) Ascia (1) Appias (2) Eurema (23) Anteos (2) Phoebis (6) Heraclides (11) Battus (3) Epargyreus (3) Polygonus (2) Chioides (4) Urbanus (6) Astraptes (pt.)(3) Cogia (1) Nisoniades (1) Cabares (1) Antigonus (1) Achlyodes (1) Grais (1) Gesta (1) Ouleus (1) Heliopetes (1) Pyrgus (2) Cymaenes (1) Wallengrenia (4?) Hylephila (1) Panoquina (6) Nyctelius (1) Lerodea (1)

Asterocampa (1) Basilarchia (1) Phyciodes (1) Calephelis (1) Eumaeus (1) Atlides (1) Parrhasius (1) Ministrymon (1) Pontia (1) Pieris (1) Nathalis (1) Colias (1) Eurytides (4) Pterourus (pt.) (2) Papilio (1) Parides (1) Phocides (2) Autochton (2) Erynnis (1) Oarisma (2)

Affinity with Mexico and/or Central America

Affinity with Central and South America

Anetia (4) Greta (2) Anaea (1) Siderone (1) Hamadryas (3) Dynamine (2) Lucinia* (2) Adelpha (2) Junonia (3) Atlantea* (4) Antillea* (2) Hypanartia (1) Dianesia* (1) Allosmaitia (2) Chlorostrymon (2) Nesiostrymon (1) Dismorphia (2) Ganyra (2) Kricogonia (1) Zerene (1) Pterourus (pt.) (1) Aguna (1) Polythrix (1) Astraptes (pt.) (3) Burca* (4) Timochares (1) Ephyriades (3) Pyrrhocalles (2) Perichares (1) Vettius (1) Synapte (1) Rhinthon (2) Holguinia* (1) Polites (2) Atalopedes (3) Parachoranthus* (1) Choranthus* (6) Paratrytone (1) Euphyes (3) Asbolis* (1) Hesperia (1)

Calisto (25+) Archaeprepona (1) Myscelia (1) Eunica (3) Siproeta (1) Eresia (1) Eueides (1) Philaethria (1) Pseudolycaena (1) Cyanophrys (1) Electrostrymon (4) Pseudochrysops* (1) Melete (1) Aphrissa (4) Chiomara (2) Pheraeus (1) Saliana (1)

Note: Endemic genera are set off by an asterisk (*) and placed with nearest relative(s); number of Antillean species is in parentheses after generic name.

132

Biogeography of the West Indies: Patterns and Perspectives

On a purely dispersalist model, one would expect that Cuba would have a large component of species in common with the Yucatan, since the overwater distance is only 220 km across the Yucatan Channel, but this is not so. Prevailing winds would tend to pass insects from Cuba to Yucatan, rather than vice versa. The only instance reported in recent years (Shields, 1985) of a Cuban species being taken on the Mexican (or Texas) mainland is the libytheid, Libytheana motya (Boisduval & LeConte). At the same time, there are some well-documented invasions of Cuba by Mexican species, most notably Hamadryas amphinome mexicana (Lucas), which appeared in Pinar del Rio, Cuba in the 1860s and was reported by H. W. Bates. Numerous fresh specimens were taken in western Cuba in the 1930s (M. Bates, 1936; de la Torre y Callejas, 1954), indicating that the species was well established there. Currently, both H. amphicloe and H. amphinome are recorded from Cuba with infrequent strays of H. feronia observed. Although there are some old reports of Antillean species being taken in Honduras (Evans, 1952), these are unverified and probably due to mislabeling. There should have been a great influx of Cuban species on the Florida Keys, a distance of only 150 km, no matter which model is employed; and, in fact, this has happened. A number of butterflies can best be explained by waif dispersal with the lycaenids, Electrostrymon angelia angelia (Hewitson) and Ministrymon azia (W. H. Edwards), and the skipper, Asbolis capucinus (Lucas), just a few species that have become well established in the Keys and South Florida in recent years (Klots, 1951; Anderson, 1974; Smith et al., 1994). Sightings or captures have established that Strymon limenia (Hewitson), a lycaenid; Anartia chrysopelea (Huebner), a nymphalid (both Anderson, 1974); the pierid, Aphrissa orbis (Poey) (Scott, 1986); and the swallowtail, Eurytides celadon (Lucas) (C. V. Covell, Jr., personal communication) have visited the Florida Keys recently. Many such reports involve strong-flying, often migratory, taxa. Endemic species are usually not shared between islands. In such cases, endemic species are generally shared only with nearby islands. For example, Pyrgus crisia (Herrich-Schaeffer) is found both in Cuba and Hispaniola (perhaps Puerto Rico), but not in Jamaica or Florida. Thus, most of the evidence indicates these insular faunas having evolved in situ. The question must be addressed whether these faunas are the result of chance dispersals from the mainland or passive vicariance based on the mobile history of the islands, or both.

THE AGE OF BUTTERFLIES AND ITS BIOGEOGRAPHICAL IMPLICATIONS The question of how old the butterflies are has vexed researchers for many years. The problem becomes more complex when one realizes that there are fewer than three dozen unquestioned Rhopalocera fossils known, and these span most of the Tertiary (Common, 1975, 1990; Brown, 1976; Powell, 1980; Whalley, 1986; Murata, 1998). Because their delicate wings and chitinous exoskeletons require very special conditions for preservation, butterflies do not fossilize well. Hence, we have a few fragments and limited information in the fossil record. These are often too badly broken to even be recognizable as Lepidoptera. Even fewer more-or-less-complete adult casts exist and only one egg (Gall and Tiffney, 1983), and these are the basis of practically all of our knowledge about this insect order. Although there have been reports of Lepidoptera in amber from the Dominican Republic in recent years, none of this information thus far has been published. Since there are no fossil Lepidoptera actually published from the West Indies, it is necessary to extrapolate butterfly distributions with those of animal groups having congruent distributions and with better fossil records. The majority of the lepidopterous fossils are of Oligocene age, chiefly from the Aix formation of France and the Florissant beds of Colorado, although the latter is now considered as the latest Eocene (H. Meyer, personal communication). These fossils are mainly Satyridae, Nymphalidae, and Pieridae, the first two of which are considered among the most derived families in the Lepidoptera. Most of the fossils may be assigned with ease to extant genera (Common, 1990; Labandeira

The Biogeography of the West Indian Butterflies (Lepidoptera)

133

and Sepkoski, 1993), and in at least one instance, the fossil is almost indistinguishable from a present-day species (J. Miller and Brown, 1989). One of the other fossils, Doxocopa willmattae Cockerell, from the Florissant, is closely allied to an extant West Indian species. A fossil pierid, Oligodonta florissantensis Brown is morphologically similar to the present-day Andean genus, Leodonta Butler, and the satyrid, Prodryas persephone Scudder is a member of an extant, largely Old World tribe, the Parargini (L. Miller, in preparation). These fossils are even older than the Miocene fossil fish that are referable to a present-day species reported by Lundberg et al. (1986). Since fossils can be used only to give a minimum age for a taxon (Patterson, 1981), any species represented is therefore in fact much older than indicated by its fossils. Very few genera of terrestrial animals represented in the fossil record have persisted from at least Oligocene time to the present. Thus, the conclusion that at least some lineages of butterflies, even among the most apotypic families, are bradytelic is inescapable. However, for the Insecta, the preserved familial diversity increases steadily following the Permian–Triassic and continues to increase sharply throughout the middle Tertiary (Labandeira and Sepkoski, 1993). Although tetrapods and marine bivales have the lowest rates of extinction derived from the fossil record, it appears that insect families have even lower rates of familial turnover throughout much of their recent history. The fossil record for Lepidoptera supports this low rate of extinction and the postulated antiquity of Lepidoptera suggested by Carpenter (1930) and Forbes (1932), and presumably followed by Shields and Dvorak (1979), who assumed that the Lepidoptera arose or differentiated in the Jurassic. Common (1975, 1990) and Powell (1980) review the extant fossil Lepidoptera with an emphasis on the Microlepidoptera and larval hostplant associations through time. Whalley (1986) documents the lepidopteran fossils and describes one microlepidopteron from the Jurassic. He states that, although Triassic and earlier fossils have been reported, these have proven to be cicadas or some similar insects. However, Whalley does indicate that clearly the Trichoptera and the Lepidoptera had already diverged from their common stem sometime during the Jurassic. Similarly, based on biogeographical patterns, L. Miller (1968), in a dispersalist model, assigned the origin of the Satyridae to the Cretaceous. All of the above-cited reports place the origin of the Lepidoptera farther in the past than that of various orders of mammals or birds; in fact, the situation much more closely parallels that of reptiles, amphibians, and fish. Common (1975, 1990), Whalley (1986), Labandeira and Sepkoski (1993), and others note that the extant butterfly families and genera were well established by the mid-Tertiary. Because of herbivory, pollination, and the specific plant/butterfly associations, it has long been accepted that the origin of butterflies was contemporaneous with the angiosperms (Smart and Hughes, 1973; Raven and Axelrod, 1974; White, 1990) or previous to their origin (Common, 1975, 1990; Powell, 1980; Whalley, 1986). The angiosperms, long thought to be restricted to the early Cretaceous sediments of Laurasia, have been found recently in comparable deposits in southern South America (Romero and Archangelsky, 1986), thereby establishing their presence in at least West Gondwanaland. Another description of a primitive angiosperm from China (Sun et al., 1998) supports an even earlier origin in the Jurassic of this group, and clearly the early angiosperms were much more widely distributed than previously believed. Recently, Labandeira and Sepkoski (1993) presented an alternative hypothesis suggesting that, if indeed the rates of insect diversity were contemporaneous with the origin of the angiosperms, there should be an exponential increase in diversity within the following time interval. Their evidence suggests that insect diversification actually decreased with the radiation of the angiosperms, and that the earlier diversification of the Insecta probably was responsible for the rapid radiation of the angiosperms during the mid-Cretaceous. The above evidence documents that the Lepidoptera are much older than earlier authors once believed. They are old enough that they were present on both parts of the divided Pangaea during the Mesozoic, and at least the Satyridae are beginning to show patterns of distribution and evolution that are consistent with vicariance models proposed for other heterothermic animals (L. Miller, in preparation). It must be noted here that the satyrids are considered to be “advanced” compared to

134

Biogeography of the West Indies: Patterns and Perspectives

some other butterfly groups. Common (1990) also agrees that most modern butterfly groups were extant in the Cretaceous. This therefore suggests that butterfly distribution in the West Indies can be interpreted based on the fossil record and within the geological constraints presented by a modification of the Pindell and Dewey (1982) model as discussed by Perfit and Williams (1989) and later refined by Pindell and Barrett (1990). Guyer and Savage (1986) used these models in their biogeographical discussion for anoles, and we later employed them in our discussions on the biogeography of butterflies in the West Indies (Miller and Miller, 1989; Smith et al., 1994).

ARE ALL BUTTERFLIES EFFECTIVE DISPERSALISTS? The simple fact that butterflies have wings has led many to conclude that these insects must be accomplished fliers capable of long-distance flights over water. Arguments for such a classical dispersalist model for populating islands are eloquently discussed by Carlquist (1974). Overwater dispersal is surely true for a few species in selected, migratory genera, such as Vanessa, Phoebis, and some Danaus, and certainly this was how some Lepidoptera reached the Antilles. This explanation for populating the Antilles long has been employed by zoogeographers (Clench, 1964; L. Miller, 1965; Scott, 1972, for example), but it is necessary to examine the “life styles” of individual species to determine the accuracy of this argument. Shreeve (1992) discusses different variables involved in the migration and dispersal of butterflies, such as predators, parasitoids, and the competition and availability for host plant and nectar resources, while Endler (1982:644, table 35.1) lists a number of animals, including a few butterflies, with their estimated dispersal potential gleaned from various published sources. The Lepidoptera in the latter reference have dispersal distances calculated at between 10 m and 5 km, and the average flight distances vary with individual species. Although many butterflies are capable of longer flights and may readily disperse in this manner, other species are more localized and cannot readily traverse across water barriers. The Satyridae, for example, are very sedentary (Endler, 1982, lists the European species, Maniola jurtina [Linnaeus], with an effective dispersal distance of 10 m), as are the Ithomiidae (Fox, 1963). Fox employed land bridges from the mainland to the Antilles to account for the spread of these sedentary ithomiids, but if we reject Schuchert’s (1935) land bridges, we must explain how these nonvagile insects reached the islands. Many butterflies take refuge at the slightest hint of inclement weather, and the probability of their dispersal by hurricanes is thus minimized (Fox, 1963). While such species may disperse slowly over the relatively benign land given enough generations, it is extremely unlikely that they could have made long, overwater flights. Personal observations indicate it is unlikely that most Lycaenidae could accomplish this feat, in contradiction to the dispersalist model proposed by Clench (1964). The answer to the question, “are butterflies effective dispersalists,” then is “it depends.” Some butterflies have excellent dispersal potential and surely arrived in the Antilles by this means. In other instances, dispersal cannot have accounted for all of the Antillean distributions. For these species we searched for another model (1989), a composite model incorporating both dispersal and vicariance, to explain present Antillean lepidopteran distributions.

CURRENT STUDIES Based on our previous publications (Miller and Miller, 1989; Smith et al., 1994), recent field studies, specimens examined in the major museum collections, and information derived from Riley (1975), Clench (1977), Clench and Bjorndal (1980), J. Miller (in preparation), and Scott (1986), we have continued to examine the origin and distribution of West Indian butterfly fauna via both vicariance and dispersalist events. The vicariance model employed, presented originally in 1989, is based on the original tectonic model employed by Pindell and Dewey (1982) and refined by Perfit and Williams (1989) and later by Pindell and Barrett (1990). Classical dispersalist explanations are roughly similar to those outlined above and especially by Darlington (1957) and modified subsequently by Clench (1964).

The Biogeography of the West Indian Butterflies (Lepidoptera)

135

THE DISPERSALISTS Many butterfly species are adapted to long-range flight and migration, and these could be expected to have colonized the islands often. A number of these are well-documented migrants, whose flights in the continental landmasses cover thousands of kilometers every year. If the dispersalist model is correct, there might have been a continuous interchange of migrants that would potentially facilitate genetic exchange between insular and mainland populations and, therefore, would theoretically make such species phenotypically (and genetically) similar, whether they came from the Antilles or from the mainland. Still other, not quite so vagile, insects may have dispersed to the islands and subsequently subspeciated (perhaps speciated), but these would still be considered dispersalists. Much of the dispersal should be of Pleistocene age (Clench 1964), and their affinities ought to be clear. Based on the fossil record and the current distributions, those species listed here are assigned to geographical areas from whence they dispersed to the islands as follows: NA = those species that entered from North America; Mex = those that entered via Mexico; CA = those species that were derived from Central America; and SA = those species that entered from South America via Trinidad and/or the Lesser Antilles. The affinities of the West Indian genera are shown in Table 2. For further information on the Antillean distribution of the species listed, see Riley (1975) and Smith et al. (1994). Many danaids are well-documented dispersalists. The Antillean species of Danaus Kluk (mostly NA) quite possibly dispersed to the islands, usually subspeciated, and often are involved in at least some inter-island movement (Simon and L. Miller, 1986; Miller and Miller, in preparation). Lycorea Doubleday (NA, but see below) has undergone a similar dispersal, and the Lesser Antilles are regularly visited by the South American L. ceres atergatis Doubleday (Smith et al., 1994:41). Several nymphalids best can be ascribed to dispersal from mainland populations. Examples include Doxocopa laure (Drury) (CA), Marpesia petreus (Cramer) and chiron (Fabricius) (CA), Colobura dirce (Linnaeus) (CA), Historis odius (Fabricius) and acheronta (Fabricius) (CA), at least Hamadryas amphinome mexicana (Lucas) (CA), Dynamine species (Mex or CA), Eunica species (CA except heraclitus [Poey]), Adelpha iphicla (Linnaeus) (Mex), Hypolimnas misippus (Linnaeus) (SA?), Junonia species (NA), some Anartia species (Mex, SA), Biblis hyperia (Cramer) (SA), Siproeta stelenes (Linnaeus) (Mex or CA), Phyciodes phaon W. H. Edwards (NA), Eresia frisia (Poey) (CA), Vanessa species (NA), Euptoieta sp. (NA, Mex), Philaethria dido (Clerck) (CA), Agraulis vanillae (Linnaeus) (NA, SA), Dione juno (Cramer) (Mex or CA), possibly Dryas iulia (Fabricius) (SA or CA), perhaps Eueides melphis (Godart) (CA), and perhaps Heliconius charitonius (Linnaeus) (all of the above regions). Within the Lycaenidae, the following are almost certainly attributable to dispersal: Pseudolycaena marsyas (Linnaeus) (SA), Chlorostrymon species (Mex, CA), Ministrymon azia (SA), Leptotes cassius (Cramer) (probably SA), and Hemiargus hanno (Stoll) (CA). Most of the Pieridae, except species of Dismorphia, a few Eurema, and perhaps Melete salacia (Godart), appear to be dispersalists from various sources. Swallowtails, such as Heraclides thoas (Linnaeus) (CA) and cresphontes (Cramer) (NA, Mex, or CA), Papilio polyxenes Fabricius (NA), Pterourus palamedes (Drury) and troilus (Linnaeus) (NA), and probably the Eurytides sp. (derived from the NA E. marcellus [Cramer]) are the West Indian Papilionidae whose distributions are probably referable to dispersal. The distributions of several Hesperiidae best explained by a dispersalist model include Phocides pigmalion (Cramer) (CA), Proteides sp. (SA?), Epargyreus sp. (NA), Polygonus sp. (CA), Aguna asander (Hewitson) (CA), most Urbanus sp. (most areas), Autochton sp. (NA, Mex), Cabares potrillo (Lucas) (CA), Eantis mithridates (Fabricius) (CA), Timochares sp. (CA), Grais sp. (CA), Gesta gesta (Herrich-Schaeffer) (CA or SA), Chiomara sp. (SA), Erynnis zarucco (Lucas) (NA or Mex), Pyrgus oileus (Linnaeus) (Mex or CA), Perichares philetes (Gmelin) (Mex), Synapte malitiosa (Herrich-Schaeffer) (Mex), Polites sp. (NA or Mex), Hylephila phyleus (Drury) (any of the regions), Atalopedes species (NA or Mex), Calpodes ethlius (Stoll) (any of the regions), Panoquina sp. (NA, Mex, CA), Nyctelius nyctelius (Latreille) (CA or SA), Lerodea eufala (W. H. Edwards) (NA or Mex), and Saliana esperi Evans (CA).

136

Biogeography of the West Indies: Patterns and Perspectives

Theoretically, the West Indian butterfly fauna in general has been well documented (Scott, 1972, 1976; Riley, 1975), but some of these records are based on older specimens with some erroneous data. Through our efforts (Smith et al., 1994), those of others, and recent field studies, we have further refined our knowledge of the butterflies of the West Indies. Several new insects continue to be described each year. However, additional research is required to fully assess the fauna of the islands and its possible origin. In some families, lepidopterists often are still in an alpha taxonomic position, and biogeographical knowledge is equally fragmentary. Thus, the full extent of the geographical ranges of many butterflies has not yet been established, thereby making the task of delimiting either centers of origin or the assessment of potential vicariance difficult.

A VICARIANCE/DISPERSAL MODEL FOR THE BIOGEOGRAPHY OF WEST INDIAN BUTTERFLIES Numerous problems are associated with the biogeographical analysis of butterflies. In the light of phylogenetics, there is uncertainty about the status of the classification, although the recent efforts of Johnson (1991, 1993), Shuey (1986, 1994), Burns (1987, 1989), and others have provided insight into some groups. Much of the taxonomy above the species level requires refinement (L. Miller and Brown, 1983; Smith et al., 1994). To date, very few modern taxonomic and biogeographical revisions have been done, and even fewer have been published. The biogeographical analysis that follows is therefore presented in chronological sequence, and the examples employed are drawn from unrelated butterfly taxa from the West Indies whose geographical distributions show congruence with other animal groups that have usable fossil records. Our original evaluation of the West Indian fauna (1989) was based on the plate tectonic model of Pindell and Dewey (1982) and further refined through the efforts of Burke (1988), Perfit and Williams (1989), and Pindell and Barrett (1990) with minor modifications.

LATE MESOZOIC

TO

CRETACEOUS

During this period Africa and South America were still in contact (Brundin, 1981), but the continents were in the process of separating. The North and South American continents were connected during most of the Cretaceous (Pindell and Dewey, 1982), and the Antilles did not exist in their present form. To understand the biogeography of West Indian butterflies, one must consider the Mesozoic era even before the formation of the proto-Greater Antilles. During the Jurassic and the early Cretaceous, Pangaea split into a northern Laurasia and a southern continent, Gondwanaland, separated by the Tethys Sea. Later in the Cretaceous, Laurasia and Gondwanaland themselves began to fragment, and the parts to move toward their present positions. It is here that our story begins. Because of the probable age of butterflies (Whalley, 1986), we cannot accept the Jurassic–early Cretaceous vicariance of all butterfly groups postulated by Shields and Dvorak (1979), but upon reexamination, Whalley (1986) and Common (1990) support this time frame for the origin of some primitive moth groups. Certainly the recent review of Labandeira and Sepkoski (1993) supports an earlier origin for all insects. Butterflies are believed to have been established during the Cretaceous (L. Miller, 1968, in preparation; Whalley 1986), perhaps early to mid-Cretaceous. Some groups appear to be basically Laurasian and others Gondwanian (Common, 1975, 1990; Miller and Miller, 1997; L. Miller, in preparation). During part of this time Africa and South America were in contact (Brundin, 1981) and shared at least parts of the same fauna. Several butterfly groups fall into this category, including the satyrid sister groups Manataria (South America) + (the African Aeropetes + Paralethe) (Miller and Miller, 1997). These satyrids are too fragile to have been involved in significant overwater dispersal of the magnitude necessary to explain these distributions by more recent dispersal. There are African affinities in a very small proportion of the West Indian butterfly fauna. Four butterfly genera, about 4.5% of the Antillean fauna, definitely have their sister groups in the Ethiopian region (see below). Such affinities, while rare, are by no means unique to butterflies.

The Biogeography of the West Indian Butterflies (Lepidoptera)

137

FIGURE 1 Facies of Archimestra teleboas, (a) upper side; (b) under side, from the Dominican Republic. Neptidopsis ophione. (c) upper side; (d) under side, from Ghana. Structural features also support the resemblance as discussed in the text.

Flint (1977) commented on West Indian Odonata and Trichoptera and reported a small African influence on these, and recently Liebherr (1986) has described a genus of West Indian carabid beetles, Barylaus, from Hispaniola and Puerto Rico, whose nearest relatives are from Africa, Madagascar, and Central America. Although it has African ancestral relationships, the nymphalid genus Eunica is Neotropical with three species represented in the West Indies (Jenkins 1990). Two of these, monima (Cramer) and tatila Herrich-Schaeffer, are more or less widely distributed in Central America and given to moderate mass movements on the continent (Howe, 1975). These species appear to be candidates for dispersal to the islands and are virtually indistinguishable from continental examples. The third species, however, E. heraclitus (Poey), known only from Cuba, is aligned with the south Brazilian species, E. macris (Godart). The two species superficially (and structurally) bear similarity to African Sallya, the sister genus of Eunica (Jenkins 1990). All maintain the “primitive nymphaloid pattern” of Schwanwitsch (1924) and were reexamined more recently by Nijhout (1991) as the nymphalid ground plan in terms of phylogenetic analysis. In his revisionary studies, Jenkins derived a cladogram of Eunica and its relatives that places Eunica and the African Sallya as sister groups. This cladogram is more or less congruent with that of Liebherr (1986) for the Carabidae. This suggests a late Mesozoic vicariance separating the two genera on Africa and America followed by vicariance of E. heraclitus onto what is now Cuba rather early in the Tertiary. An even more dramatic example of this biogeographical pattern involves the endemic Archimestra teleboas (Menetries) from Hispaniola. The sister group of this monobasic genus is Neptidopsis from Africa and both are illustrated (Figure 1). The closest outgroups for these two genera are the Neotropical genera Mestra and Vila (D. W. Jenkins, personal communication). Mestra is represented on the islands by a pair of species, one in Jamaica and the other an apparent dispersalist from South America on the Lesser Antilles. The implications are clear: the ancestral member of this group must have been rather similar to teleboas, which participated in the African–South American vicariance late in the Mesozoic, and then vicariated onto Hispaniola in the earliest Tertiary. The nymphalid genera Archaeoprepona (Antilles and mainland tropical America) and Prepona (Central and South America), along with Agrias (American tropics) and Charaxes (Old World tropics), show a comparable distributional pattern. Charaxes and Archaeoprepona are sister genera and they in turn are the sister group of the more derived Prepona + Agrias (Johnson and Descimon, 1989). It is interesting that the most plesiomorphic American genus in this cluster is the one that is represented in the West Indies, a situation that is comparable to the Barylaus model (Liebherr, 1986). Archaeoprepona demophoon (Huebner), the West Indian species, has subspecies on Hispaniola, Cuba, and Puerto Rico. A similar pattern is evident in the lycaenid genus Brephidium, represented by three species: one from the southwestern United States to Venezuela and the Greater Antilles (Riley, 1975), one

138

Biogeography of the West Indies: Patterns and Perspectives

a

b

1 mm

c

d

FIGURE 2 Structural examples of the African affinities in the West Indian butterflies are found in the left lateral views of male genitalia of Oraidium and Brephidium species: (a) O. barberae, South Africa; (b) B. metophis, Africa; (c) B. isophthalma, Florida; (d) B. exilis, Arizona.

from Florida and as a probable stray on the Bahamas (Smith et al., 1994), and a third from South Africa (Eliot, 1973; Dickson and Kroon, 1978). The sister group of this genus, and the only other member of the Brephidium section of the Polyommatini of Eliot (1973), is the monobasic South African genus Oraidium. The male genitalia (Figure 2) are illustrated for Brephidium and Oraidium to demonstrate their very close relationship and other superficial characters also show their similarity. Thus, this is also a very old, now relict African–Neotropical vicariance pattern that is congruent with Liebherr’s pattern in beetles. In this genus, however, one must postulate extensive extirpation of these insects on intervening landmasses between the Cretaceous and the present. A possible relationship has been postulated between the American sister genera Eretris and Calisto, which is West Indian, and their putative African relatives, the Pseudonympha (Smith et al., 1994) and other African genera (Riley, 1975; Brown, 1978), but this relationship is much less close and dramatic as are the ones cited above. The possible African affinities of Calisto led Brown (1978:16) to state that it is “equally at home in the African tribe Dirini as in Pronophilini of the Andes.” However, reexamination of Dira and the Pronophilini in relation to Calisto does not support the close alignment of the African and West Indian insects shown in the examples given above; the latter are structurally members of the Pronophilini, a tribe comprising both Andean representatives and species widely distributed in the lowlands from Arizona to Patagonia. If there is an African element here, it is not prepossessing. Members of Calisto are all endemic to the Greater Antilles (Riley, 1975; Smith et al., 1994), and it is hoped that the studies of Sourakov (1996) and others in progress will provide further insight to the phylogenetic history of this genus.

LATE CRETACEOUS

TO

EOCENE

During late Cretaceous, the first division of the old Central American connection began with great faults cutting across the southern boundary of Yucatan and the northwest corner of South America. The resulting block formed the proto-Greater Antilles, and by the end of this period the block was

The Biogeography of the West Indian Butterflies (Lepidoptera)

139

completely separate from South America, but still in approximation to the Yucatan peninsula. The Jamaican and southern Hispaniolan blocks were farther to the west than the remainder of the protoGreater Antilles and either still adherent to Central America, or at least approximate to it (Perfit and Heezen, 1978; Case et al., 1984; Perfit and Williams, 1989; Pindell and Bartlett, 1990). Species dispersal between North and South America during the Cretaceous occurred across a new connection that was to become in turn the proto-Greater Antilles (Savage, 1983; Stehli and Webb, 1985; Guyer and Savage, 1986; Woods, 1989), resulting in an extensive feeder population of organisms for the original vicariance of the Antillean fauna. Pindell and Dewey (1982) and D. L. Smith (1985) propose that the proto-Greater Antilles began moving eastward relative to Mexico in the late Cretaceous, although significant movement of the Caribbean plate is questioned by Donnelly (1985, 1988). This proto-Antillean block was finally severed from the Yucatan during the Eocene. Several lepidopteran groups seem to have been part of this event, and these vicariants are usually species represented today on Cuba, Hispaniola, and Puerto Rico. Guyer and Savage (1986) have correlated these vicariance events with the present-day distribution of anoles, and their patterns show congruence with some patterns of butterfly distributions. The primitive anole genera, Chamaeleolis and Chamaelinorops, are found exclusively on Cuba and Hispaniola, respectively, and are most closely allied to the sub-Andean genus, Phenacosaurus (Guyer and Savage, 1986:524–528). Those authors ascribed these distributions to these late Mesozoic and early Tertiary vicariance events. A parallel vicariance pattern is evident in several butterfly genera. A few groups, however, such as Atlantea (Nymphalidae), Calisto (Satyridae), Lycorea (Danaidae, cleobaea), Heraclides (Papilionidae), Nesiostrymon, Terra (Lycaenidae), and Wallengrenia (Hesperiiidae), are recorded from all four islands and must date from the early vicariance although Jamaica was more closely positioned against Central America (Perfit and Heezen, 1978; Case et al., 1984; Perfit and Williams, 1989) at about the time that the respective genera divided from their mainland sister groups. For example, the closest relatives of the strange yet beautiful Cuban papilionid, Parides gundlachianus (C. & R. Felder) are based in South America, although the genus is distributed now from Mexico to Argentina, but not on other Antillean islands. The vicariance of Parides (Papilionidae) is postulated as a Cretaceous or Paleocene event with further dispersal taking place in the Tertiary on continental landmasses. Dating the Papilionidae from this time is not unprecedented: Durden and Rose (1978) described a recognizable papilionid from the middle Eocene. A similar scenario will explain the presence of the single member of the Riodinidae from the islands, though that species, Dianesia carteri (Holland), is also found on at least Andros and New Providence in the Bahamas (Harvey and Clench, 1980) and on Cuba (Alayo and Hernandez, 1987; Hernandez et al., 1998). While we postulate short-range dispersal for this species across the Old Bahamas Channel to emergent land now submerged on the Great Bahamas Bank, perhaps during the Pleistocene, the butterfly originally vicariated to Cuba. Almost the exact same pattern (reported once on New Providence, but present on Great Abaco) is shown within the genus Eumaeus (Lycaenidae) with some colonization, perhaps from Andros, of the southeasternmost Florida peninsula (Holland, 1931, and others). Despite the stands of available host plant, Zamia in the Dominican Republic, it is curiously absent. We doubt that biogeography of Eumaeus is as complicated as suggested by Shields and Dvorak (1979), who postulated a Jurassic vicariance of this genus. It is more easily explained with reference to its Central American sister-species, E. toxea (Godart) and vicariance during the early Tertiary onto Cuba. The Cuba–Andros–south Florida connection has been discussed elsewhere (Miller et al., 1992; Smith et al., 1994).

OLIGOCENE

TO

PLIOCENE

During this period the proto-Greater Antillean block fragmented, and the resulting small blocks moved about and accreted onto the islands as we know them today. Most of this accretion was completed by the end of the Pliocene. Both Hispaniola and Cuba are products of accretion; in one

140

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 3 Distribution of the genera Calisto (heavy stippling) and Eretris (vertical cross-hatching) in the West Indies and continental America. Eretris is widely distributed and found southward in the Andes to Bolivia.

case, both islands are the beneficiaries of the same early block (the eastern Cuba–northern Hispaniola block of Pindell and Dewey, 1982), which fragmented, one part going to Hispaniola and another to Cuba, apparently in the Oligocene (for details see Guyer and Savage, 1986). Most of Puerto Rico evidently became isolated early, but a small block that later split off that island accreted onto eastern Hispaniola, and southern Hispaniola separated from Nuclear Central America at a later date probably in the late Oligocene or early Miocene (Pindell and Dewey, 1982; Pindell and Barrett, 1990). Contrary to earlier speculations, the Blue Mountains of Jamaica and the La Selle massif of Haiti apparently were continuously emergent from the time of their separation from Central America (Case et al., 1984; Perfit and Williams, 1989) and were possible vehicles of vicariance. Because Hispaniola and Cuba are far more complex geologically than other Greater Antillean islands, they have greater diversity of the fauna than expected based on diversity of species vs. land area. For example, the complex satyrid genus Calisto is recorded on all of the Greater Antilles and on some of the Bahamian islands, with the greatest species diversity on Hispaniola followed by Cuba (Figure 3). A single species of Calisto inhabits Jamaica, another is found on Puerto Rico, three or four in Cuba, two of which are reported from the northwestern Bahamas, and more than 20 species on Hispaniola. Each island has one (two in Hispaniola) larger species with a more pronounced hind wing tornal lobe. These larger representatives of the genus Calisto are most closely allied to the basically sub-Andean Eretris, which has representatives in the mountains of Central America, one of which is found today in Guatemala and Chiapas. Calisto subdivided into two lineages on Hispaniola and Cuba. The species most like Eretris have tornal lobes on the hind wings and genitalia that are considered plesiomorphous; they inhabit the lowlands and appear to have evolved there. With the exception of C. anegadensis recently described from Anegada in the British Virgin Islands (Smith et al., 1991), only Cuba and Hispaniola have smaller, more round-winged species, which appear to be a later development and do not so closely approximate Eretris. These more apomorphic Calisto have no tornal lobe, genitalia that are similar to each other, and are usually restricted to montane habitats. Some of these species have reinvaded the lowlands, but generally they are restricted to forest–woodland habitats than are the primitive species. This species diversity leads to the

The Biogeography of the West Indian Butterflies (Lepidoptera)

141

conclusion that most of the evolution in the genus took place on the several separated plates that later accreted to form Cuba and Hispaniola. Certainly the complex geological history of Cuba and Hispaniola accounts in part for the greater species diversity on those islands, but the presence of the small, apomorphic forms, evolving on separate blocks that accreted to form both islands, indicates additional speciation must have taken place on these formerly separated islands as well. Sourakov (1996) recently has provided additional insight into the life histories of this complex group, and it is hoped that his phylogenetic studies in progress will provide further insight into the origin of Calisto. Evolution on the eastern Cuban–central Hispaniolan block could account for the present distribution of several organisms shared only by Cuba and Hispaniola. Examples in addition to the already mentioned group of Calisto include two species of Anetia (Danaidae), Lucinia sida (Huebner) (Nymphalidae), Melete salacia (Godart) and Aphrissa orbis (Poey) (Pieridae), and Astraptes xagua (Lucas) and habana (Lucas), and Polites baracoa (Lucas) (Hesperiidae). Many butterfly genera show a generalized distribution within the Greater Antilles that has been ascribed to past dispersal, but they are just as easily explained by late Cretaceous to Paleocene vicariance events (see the maps in Guyer and Savage, 1986:527, figure 10; Perfit and Williams, 1989). Such an explanation accounts for much of the genus Heraclides (Papilionidae) in the islands. We have already suggested that H. thoas and cresphontes might be dispersalists, although the former could as easily be a vicariant, perhaps dating from the Eocene (Guyer and Savage, 1986:527, figure 10, top right map). Perhaps this model addresses Riley’s (1975:143) comment that “It is curious that this widespread species should have reached none of the West Indies other than Cuba and Jamaica.” Other Heraclides species in the West Indies are congruent with a vicariance model. Heraclides aristodemus (Esper) is a species restricted to Cuba, Hispaniola, Puerto Rico, Little Cayman, the Bahamas, and the Florida Keys. Heraclides andraemon (Huebner) occurs naturally in Cuba, the Bahamas, the Caymans, and was introduced into Jamaica. In contrast, H. machaonides (Esper) is restricted to Hispaniola and Puerto Rico, H. thersites is known from Jamaica, H. aristor (Godart) occurs in Hispaniola, H. oxynius (Huebner) is from Cuba, H. pelaus (Fabricius) is found on all of the Greater Antilles, and H. caiguanabus (Poey) is known from eastern Cuba. The distributions given for these species approximate ones already mentioned. All of these insects are rutaceous feeders as larvae, and all were derived from the immigrant H. thoas (or H. cresphontes) stocks (Hancock, 1983; J. S. Miller, 1987a, 1987b). A cladogram of these insects suggests that H. machaonides, H. andraemon, and H. aristodemus represent early branchings of the Heraclides stock in the Antilles, whereas H. oxynius and perhaps H. pelaus represent a somewhat later development. Finally, the more derived H. oxynius and H. caiguanabus evolved, the former in the bulk of Cuba and the latter probably on the eastern Cuba plate after it split off northern Hispaniola. The evolution and the complex biogeographical history of this genus in the West Indies is currently undergoing further study, but the data appear congruent with a vicariance model with some relatively modern inter-island dispersal having taken place. This is especially true in H. aristodemus in the Bahamas. The papilionid species Battus polydamas (Linnaeus) basically is monomorphic on the mainland, but south Florida and the Antilles have 13 different subspecies (Figure 4; and Riley, 1975:141–143; Smith et al., 1994:165–166, pl. 25). In some parts throughout its range, this butterfly is not a good dispersalist, as witness the lack of occasional records of even any of the Antillean subspecies (for example, St. Lucia and St. Vincent have different subspecies although they are in physical proximity). Expanses of water as little as 21 to 25 km have been effective isolating barriers in the Lesser Antilles while the same subspecies, B. p. lucayanus, is shared in the Bahamas and south Florida. Apparently there appears to be a selective advantage to be a sedentary insect in these insular populations of B. polydamas similar to Dryas iulia (Fabricius) (Nymphalidae) in the West Indies. These Lesser Antillean subspecies of B. polydamas are postulated to be the result of the initial breakup of the ancient Central America, although the situation may be complicated by a possibly dispersalist South American origin of most of the Lesser Antilles subspecies. The vicariant scenario would require complete separation of the subspecies other than those in the Lesser Antilles by about the Oligocene–Miocene boundary.

142

Biogeography of the West Indies: Patterns and Perspectives

a

b c

d

e g f

h i j k l m

FIGURE 4 Distribution of B. polydamas in the West Indies. The stippled area is inhabited only by B. p. polydamas (Linnaeus). Island subspecies include (a) B. lucayus (Rothschild and Jordan); (b) B. cubensis (Du Frane); (c) B. jamaicensis (Rothschild and Jordan); (d) B. polycrates (Hopffer); (e) B. thyamus (Rothschild and Jordan); (f ) B. antiguus (Rothschild and Jordan); (g) B. christopheranus (Hall); (h) B. neodamas (Lucas); (i) B. dominicus (Rothschild and Jordan); ( j) B. xenodamas (Huebner); (k) B. lucianus (Rothschild and Jordan); (l) B. vincentius (Rothschild and Jordan); (m) B. grenadensis (Hall). See text for further discussion.

A complementary pattern is observed in another papilionid, B. devilliers (Godart), from Cuba and the Bahamas, which approaches the Chamaeleolis pattern, and B. zetides Munroe, which approximates the Chamaelinorops pattern of Guyer and Savage (1986). Both of these species are probably most closely allied to the Central and North American B. philenor (Linnaeus). Battus zetides is now found only in portions of Hispaniola that were parts of the southern Hispaniolan block of Pindell and Barrett (1990), which block split away from Central America during the Eocene and accreted to Hispaniola in the Pliocene or Pleistocene (Case et al., 1984; Burke et al., 1984). A congruent pattern is shown in the castniid moth, Ircila hecate (Herrich-Schaeffer), a species endemic to the southern Hispaniolan block and related to Mexican and Central American members of Athis (J. Miller, 1986). The same pattern is roughly that shown in Myscelia aracynthia (Dalman) and its sister taxon M. cyaniris Doubleday (Jenkins, 1984). Curiously, there is not an endemic Battus on Jamaica, other than a subspecies of B. polydamas (see above), nor is any castniid or Myscelia recorded from there. These congruent patterns originally suggested to us (1989) that perhaps the relative position of Jamaica depicted in the Pindell and Dewey (1982) reconstruction might be erroneous. Based on the evidence of Perfit and Williams (1989) that Jamaica in the Eocene and the Oligocene may have been positioned against the area that later drifted southeastward as Central America (Figure 5), this shift placed Jamaica nearer its present position relative to the southern Hispaniolan block. The biogeographical pattern associated with B. zetides is reminiscent of a similar distributional pattern of the only Antillean member of the hesperiid genus Paratrytone. Insects of this genus are montane or submontane most of their present continental range (L. Miller, 1964). Representative genitalia of Paratrytone are shown in Figure 6. The West Indian species had been placed in the Antillean genus Choranthus until separated from that genus by L. Miller (1964). The distribution suggests that Paratrytone were in the contiguous parts of Nuclear Central America and that portion which has drifted southward to become present-day Central America as shown in Figure 5, along

The Biogeography of the West Indian Butterflies (Lepidoptera)

PALEOCENE MX & CA

143

EOCENE

PG

A

CA

PG

A

J

SH

SA SA

OLIGOCENE

E. MIOCENE

W&C CU

CU

CA J

CA SH

ECU PR & & NH CH

H PR J SH

SA

SA

PLIOCENE CU CA J

PR

H

LA

SA

FIGURE 5 Modified Pindell and Dewey (1982) model of the evolution of the West Indies. Abbreviations: MX = Mexico; CA = Central America; PGA = proto-Greater Antilles; SA = South America; J = Jamaica; SH = southern Hispaniola; W&C CU = western and central Cuba; ECU & NH = Eastern Cuba and northern Hispaniola; PR & CH = Puerto Rico and central Hispaniola; H = Hispaniola; PR = Puerto Rico; CU = Cuba; LA = Lesser Antilles. Exceptions to the Pindell and Dewey model involve particularly the relative Eocene placement of Jamaica and the southern Hispaniola blocks to conform with the biogeography of butterflies.

with the isolated southern Hispaniolan block. At present P. batesi (Bell) is found in the La Selle mountains of Haiti (Riley, 1980:190) and in mountains in the southwestern Dominican Republic (Schwartz, 1989), both areas of high species diversity in a number of butterfly genera and that are associated with the southern Hispaniolan block (Case et al., 1984). The derivation of this genus seems clear if one accepts closer proximity of the southern Hispaniolan block to Nuclear Central America during the Eocene–Oligocene, as suggested above. The vicariance of the two ithomiids (perhaps subspecies) known from the Antilles, Greta diaphana (Drury) and G. cubana (Herrich-Schaeffer), as previously discussed, is somewhat different. These insects are represented on Jamaica, Hispaniola, and Cuba, but not Puerto Rico. They are extremely sedentary and not subject to dispersal. Their habitat requirements of a cool, dense, somewhat montane, moist tropical forest are significantly different from other close relatives of Greta from Mexico and northern Central America. This genus appears to have been part of the early Tertiary vicariance of the entire proto-Antillean block. Perhaps these insects were simply extirpated from Puerto Rico, but the evidence suggests that they were never present there since Puerto Rico appears to have been the southern part of the proto-Greater Antillean block.

144

Biogeography of the West Indies: Patterns and Perspectives

a b

1 mm

c

d

FIGURE 6 The potential evolutionary affinities of the hesperiid genera Paratrytone and Choranthus are apparent in the male genitalia of selected species: (a) P. rhexenor, Mexico; (b) P. niveolimbus, Guatemala; (c) P. batesi, Haiti; (d) C. radians, Cuba.

The genus Euphyes (Hesperiidae) is represented by two species in the Greater Antilles, excluding Puerto Rico, and on a few Bahamian islands. They are most closely related to the continental E. peneia (Godman) and more distantly allied to several strictly South American congeners (Shuey, 1986:104; 1994). That author ascribes the distribution of the group to “an old vicariant event between Cuba and Central America with subsequent speciation and dispersal” (Shuey 1986:111). Still other taxa are only restricted to Jamaica and Hispaniola, almost certainly derived from the late Eocene–Oligocene approximation of both the Jamaican and southern Hispaniolan blocks to the Mexican mainland; organisms congruent with this pattern include Danaus cleophile (Danaidae) and Aphrissa godartiana (Pieridae). Clench (1964) claims that the butterfly fauna of Puerto Rico has a “predominately Hispaniolan character” and postulates further that the very cold Pleistocene temperatures may have extirpated all of the Lycaenidae existing there prior to the Pleistocene. A few butterflies are apparently characteristic of the Puerto Rico–central Hispaniola block and are found nowhere else in the Antilles, including Heraclides machaonides (Papilionidae) and Dismorphia spio (Pieridae) (Riley, 1975). There may have been other species that are restricted today only on Hispaniola, and Clench’s (1964) assumptions about Pleistocene extirpation in Puerto Rico might be correct for these organisms, but it probably is unnecessary to invoke Pleistocene mechanisms. As additional biodiversity surveys are completed on Puerto Rico, additional Hispaniolan species may be recorded (Smith et al., 1994; Ramos, 1996). The apparent lack of species in common between Hispaniola and Puerto Rico may be simply seasonality or a lack of collecting in various microhabitats.

The Biogeography of the West Indian Butterflies (Lepidoptera)

145

FIGURE 7 The geographical distribution of the primitive danaid genus Anetia in Mexico, Central America, and the West Indies (stippled areas) reflects its past geological history. See text for further details.

In contrast, the distribution of Pyrgus crisia (Herrich-Schaeffer) does suggest that it was attached to Puerto Rico and subsequently dispersed over Hispaniola and to Cuba when those islands were in closer proximity than at present (Figure 5). Few other species of butterflies show this distributional pattern, and the relationships remain obscure. Recent records (Hernandez et al., 1998) have located this species in eastern Cuba so it is also possible that P. crisia was part of the Eastern Cuba Northern Hispaniola plate and simply dispersed to Puerto Rico, perhaps via Mona Island (Smith et al., 1988). The primitive danaid genus Anetia displays a distributional pattern that conforms to the pattern of Battus (Figure 7). The only mainland Anetia has two subspecies, one in Mexico and northern Central America (Nuclear Central American one) and a derived one from Costa Rica and Panama; all Anetia are associated with montane to submontane habitats. Ackery and Vane-Wright (1984) originally considered Anetia to be a sister group of Lycorea. Recent life history studies of Anetia (Ivie et al., 1990; Brower et al., 1992) from Hispaniola suggest that Anetia is perhaps a relict Antillean group and is most closely aligned with Euploea, a genus endemic to the Indomalayan region. Currently Cuba and Hispaniola have one endemic Anetia species each, and each shares two species; there is one recent record for Jamaica (Vane-Wright et al., 1992), and none known from Puerto Rico. This distribution suggests that the genus arose on the northern end of the proto-Greater Antilles, but perhaps excluded Puerto Rico. Perhaps most Anetia species were isolated on the eastern Cuban central Hispaniolan block during the Oligocene and evolved on the separate fragments when that block split. Geological and further biogeographical evidence suggests that the evolution of the Jamaican fauna is the result of its relatively late connection with Nuclear Central America. Jamaica, the west and central Cuban plates, and the southern Hispaniolan plate were all approximate or adjacent to Nuclear Central America in the Eocene (Figure 5). The evolution of the most spectacular of the Antillean butterflies, Pterourus homerus (Fabricius), is of major interest here. The sister taxon of P. homerus is P. garamas (Huebner), a butterfly found today chiefly in the mountains of western Mexico and Central America (Jordan, 1908 [1907–1909]). Pterourus homerus is found in the submontane forests of Jamaica (Brown and Heineman, 1972). The valvae of the male genitalia of

146

Biogeography of the West Indies: Patterns and Perspectives

a

b 1 mm

c

d

FIGURE 8 The possible evolutionary affinities of the Jamaican fauna are observed in the male valvae of selected Pterourus species; (a) P. homerus, Jamaica; (b) P. cleotas, Colombia; (c) P. garamas, Mexico; (d) P. esperanza, Mexico. See text for details of the relationships.

P. homerus and its near congeners (P. garamas, P. cleotus, P. esperanza) are illustrated (Figure 8). The morphological similarity among these taxa strongly suggests that the progenitors of P. homerus and its sister group were originally closely aligned toward the west with Mexico–Central America as shown in Pindell and Dewey (1982) and later refined by Perfit and Williams (1989) and Pindell and Dewey (1990) and that Jamaica actually was to the west of its current position, perhaps accreted to this western spur during Eocene–Oligocene. This is in agreement with the positions suggested by Burke et al. (1984), and it requires that at least part of Jamaica remain emergent, as suggested by Case et al. (1984) and Perfit and Williams (1989). There are no other Pterourus in the Antilles (Hancock, 1983; J. S. Miller, 1987a, 1987b) other than P. palamedes and troilus, apparent vagrants on Cuba (Riley, 1975). Pterourus homerus is at present found both east and west of the Blue Mountains and is generally associated with shales or calcareous rocks (T. Turner, personal communication), but it is quite possible that its progenitors were part of a vicariance event centered on that part of the Blue Mountains that was emergent during Jamaica’s early geological history. If this scenario were not true, one would have to require long-distance dispersal and evolution from the Miocene onward to account for P. homerus, and the question of why this insect is endemic only on Jamaica rather than Cuba must be addressed. Further evidence of the separation of Jamaican fauna from that of the rest of the Greater Antilles is presented by Johnson and Smith (1993), in Platynus (Coleoptera) (Liebherr, 1988) and in Eleutherodactylus (Amphibia) (Hedges, 1989). The hesperiine genus, Wallengrenia, is widely distributed throughout North and South America and the Western Antilles and is an example of a number of the vicariant events previously discussed. Two species, W. otho (J. E. Smith) and egeremet (Scudder) (after Burns, 1985) are commonly found in the United States. The Neotropical species, including those represented in the Caribbean are currently under revision (J. Miller, in preparation). Based on comparative morphological examination, the genus Wallengrenia includes two species in eastern South America, W. premnas (Wallengren) W. otho, which in turn is subdivided into two subspecies, W. o. curassavica (Snellen) and W. o. sapuca Evans. Three subspecies of W. otho are recognized from Central America. However, the greatest species diversity within the genus is in the West Indies, with four species and one new subspecies represented (Figure 9). The darker species of these, W. misera (Lucas) is from Cuba and the northern Bahamian Islands. Wallengrenia druryi (Latreille) is restricted to Hispaniola, Puerto Rico, and the southern Bahamas. The bright fulvous species, W. ophites (Mabille), is found in the Lesser Antilles southward to Trinidad.

The Biogeography of the West Indian Butterflies (Lepidoptera)

147

o m o

d op

v

o,p

FIGURE 9 The distribution of the hesperine genus Wallengrenia (after J. Miller, in preparation) incorporates a number of vicariant events. Wallengrenia otho (o) is widely distributed, as is W. premnas in South America. Antillean taxa include W. misera (vertical cross hatching and m), W. druryi (light stippling and d), W. vesuria (heavy stippling and v), and W. ophites (op). See text for additional discussion.

The present geographical range of Wallengrenia in Central America, and particularly in Mexico, is quite complex. Wallengrenia otho curassavica (Snellen) inhabits on the western slopes of the Sierra Madre Occidental, with nominate W. otho in the east. The other Caribbean species, W. vesuria (Ploetz), is endemic to Jamaica and morphologically more closely resembles western Mexican W. o. curassavica than any other Antillean population, thus tending to support our modified Jamaican biogeographical model. Another species in the Greater Antilles that has been somewhat of an anomaly and is now known to be of Mexican origin is Hesperia nabokovi (Bell & Williams), the only Hesperia recorded specifically from Hispaniola. Burns (1987, 1989) has recently reviewed this species and subsequently moved it from Atalopedes to Hesperia based on genitalic characters. Emmel and Emmel (1990) provided additional supportive documentation to this generic change with an illustrated life history of this unusual species. A similar, yet somewhat complex, distributional pattern is found in Nesiostrymon (Lycaenidae). This genus was initially thought to be monobasic, but it is now known to have at least one Mexican species (K. Johnson, 1991). Its sister group, Terra, contains a number of mainland species and a recently discovered Antillean species, T. hispaniola. The sister group of these two species also contains only mainland taxa. Johnson postulates a late Mesozoic vicariance of the Nesiostrymon and Terra from Central America and subsequent evolution on the islands during the Tertiary, not unlike the pattern shown for other genera.

PLIOCENE

TO

HOLOCENE

Current evidence appears to indicate no recent vicariance events. The only possible ones involve the Bahamas, and as we have indicated, the recent geological history of Florida and the Bahamas

148

Biogeography of the West Indies: Patterns and Perspectives

in association with Cuba is complex. Burke (1988) indicated that during the mid-Tertiary, Cuba overrode the Bahamas Rise. There was a large emergent Great Bahamas Bank during periods of the Pleistocene glacial maxima with subsequent reflooding of the bank. We have reviewed the current distributions of butterflies and species diversity of these islands (L. D. Miller et al., 1989, 1992, 1998). The southern Bahamas share far more taxa with Cuba than previous believed, and the origin of some of the Cuban endemic taxa is now in question (Miller and Simon, 1998; Hernandez et al., 1998). Similarly during the Pleistocene, Florida underwent a similar reduction and extension of land area (Webb, 1990) with a further exchange of the butterfly fauna with the Bahamas.

THE LESSER ANTILLES Maury et al. (1990) reviewed the geological history of the Lesser Antilles. Although this area was once considered a single volcanic arc of Eocene age, geologically it is far more complex with the “Southern Volcanic Caribbees” and the two northern arcs: “Northern Volcanic Caribbees” or inner arc from Dominica to Bass Terre, Guadeloupe and the outer arc or the “Northern Limestone Caribbees,” which includes Grande Terre, Guadeloupe to Anguilla. Although the latter have a more ancient geological history and volcanism, these “limestone” islands appear to have a more recent terrestrial fauna with the islands of Grande Terre, Marie Galante emerging during the Pleistocene. Pinchon and Enrico (1969) provided insight into the biodiversity of the Lesser Antilles with an emphasis on the “Antilles Françaises” with further discussion by Riley (1972). During the past 12 years, this area has been the focus of our field studies due to the veritable lack of documented records. The most obvious members of the fauna in the “Southern Volcanic Caribees” are some dispersalists that doubtless came from Trinidad (Riley, 1975; Scott, 1986; Smith et al., 1994), and the few possible vicariants seem to have their origin in northern South America. However, there are some taxa such as Dryas iulia, Wallengrenia ophites, Urbanus, and Ephyriades that are widespread throughout these islands and remain indicator species for the Lesser Antilles. In other cases, we suspect dispersal for much of the Lesser Antilles, although there are some endemic taxa present, especially on Dominica, which has the largest number of species (51). As with the Greater Antilles, land area and the availability of appropriate ecological niches play significant roles in the number of species present and species diversity among these islands (Miller and Miller, in preparation).

SUMMARY Previous biogeographical theories are discussed, and previous studies on the biogeography of West Indian butterflies are enumerated. Most of these models have relied primarily on a dispersalist type and heavily influenced by Pleistocene events, with the exception of the model postulated by Shields and Dvorak (1979), which referred a number of butterfly distributions to the late Jurassic–early Cretaceous opening of the Atlantic. We accept the findings of Common (1975, 1990), Smart and Hughes (1984), and Whalley (1986) that the origin of the butterflies occurred in the Cretaceous with some primitive moths present in the Jurassic. Labandeira and Sepkoski (1993) provide further supportive documentation for the earlier origin of all Insecta and present an alternative view for the coevolution of plant/insect relationships. A synopsis of the evidence on the age of butterflies is provided. This demonstrates that some lineages are much older than previously believed; for example, most of the Florissant (latest Eocene) fossils are congeneric with and/or very near existing species (J. Y. Miller and Brown, 1989; Common, 1990; Labandeira and Sepkoski, 1993). These data suggest that modern butterfly genera were well differentiated by the late Cretaceous to earliest Tertiary, and therefore more easily explained by a vicariance model than previously believed. Additional data suggesting that butterflies are not uniformly good dispersalists are presented. Other data and information presented on the endemicity of West Indian butterflies suggest that the fauna is rather old. A vicariance/dispersal model, originally proposed for the biogeography of West

The Biogeography of the West Indian Butterflies (Lepidoptera)

149

Indian butterflies (Miller and Miller, 1989), conformed in most respects to the Pindell and Dewey (1982) model and is updated based on the recent geological evidence presented by Perfit and Williams (1989) and Pindell and Barrett (1990). While there was probably some long-range dispersal, most of the butterfly stocks can be better ascribed to a vicariance model followed by shorterdistance dispersal interspersed with the vicariance events. Most butterfly groups are considered unlikely candidates for long-distance dispersal because of their fragility and lack of vagility. Many of the original butterfly stocks were on the proto-Greater Antilles and evolved in situ within the islands. In several insular butterfly species, there appears to be a selective advantage as a sedentary insect. A chronology is given for the evolution and vicariance of several Antillean butterfly groups. Based on present taxonomic relationships in different butterfly families and previous and present biogeographical studies, it is postulated that Jamaica did indeed occupy a more westerly position, against the extension of western Central America, and that the Southern Hispaniolan block might have been in closer contact with the Yucatan peninsula during the Eocene than originally believed (Pindell and Dewey, 1982; Perfit and Williams, 1989; Pindell and Barrett, 1990). Areas of continuous emergence are postulated to be the Blue Mountains of Jamaica and the Massif de La Selle of Hispaniola, potential refugia for the butterflies of the parts of their respective islands. This reconstruction is in closer harmony with the positions of these island masses during the Oligocene and the later positions of the two landmasses and provides supportive evidence for some of the more obvious similarities between the fauna of these insular areas and Central America than are shown on most other segments that formed the western Greater Antilles. The geological history of the Lesser Antilles is reviewed and is far more complex than originally thought with portions of this active volcanic arc ranging from the Eocene through the Pleistocene. As might be expected, our current knowledge of the biogeography of the butterfly fauna remains unclear. Although most of these butterflies originated through dispersal from South America toward the south and from Puerto Rico to the west, a number of endemic taxa are present, particularly on Dominica and St. Vincent and other genera that are biological indicators of only the Lesser Antilles. These latter butterflies would therefore indicate an older origin. The recent geological history of the Caribbean Basin is complex and that of Florida, the Bahamas, and Cuba is no exception. The fact that Cuba overrode the Bahamas Rise and that there was a large emergent Great Bahamas Bank during periods of the Pleistocene glacial maxima with subsequent reflooding of the bank have both played a significant role in the current distribution of butterflies for all three areas. The southern Bahamas share far more species with Cuba than previously believed, and the origin of some Cuban endemic taxa is now in question.

ACKNOWLEDGMENTS It has been gratifying to note the impact the publication of this chapter in the first edition has had on the study of biogeography of Lepidoptera of the region. Lepidopterists now look beyond the conventional geographical boundaries to review and locate closely aligned taxonomic groups. Our colleagues and associates continue to provide us with unpublished information for this and other studies in progress. We would particularly like to thank Dr. Dale Jenkins of Allyn Museum of Entomology, Florida; Museum of Natural History, University of Florida, Sarasota; John Shuey, Nature Conservancy, Indianapolis, Indiana; Dr. C. V. Covell, Jr., University of Louisville; Dr. Kurt Johnson, Environmental Affairs, Ethical Culture Society, New York; and Stuart J. Ramos and Luis Roberto Hernandez, Department of Biology, University of Puerto Rico, Mayaguez. Dr. Thomas W. Turner, Clearwater, Florida, and lately of Jamaica, provided valuable data on the distribution and ecology of Pterourus homerus and other endemic taxa. Special thanks are due Dr. Michael R. Perfit, Department of Geology, University of Florida, for discussing geological constraints with us in the original paper (1989). Special thanks to Dr. Herbert Meyer, Florissant Fossil Beds National Monument, who provided current information on the age of the Florissant deposits. Additional thanks are due Dr. Gerardo Lamas, Universidad Mayor de San Marcos, Museum Javier Prado, Lima, Peru

150

Biogeography of the West Indies: Patterns and Perspectives

and Dr. Johnson for discussions on diverse biogeographical theories. We would particularly like to thank our special colleagues, Dr. David Spencer Smith, Hope Entomological Collections, Oxford University, and Dr. Mark J. Simon and Stephen R. Steinhauser, both associates of Allyn Museum of Entomology, for their patience and for continuing to serve as sounding boards for our lengthy discussions on various projects on West Indian butterflies. To our colleague, Dr. Charles Woods, our special thanks, for convening the original symposium that brought together an extraordinary group of scientists with such widely divergent opinions and taxonomic experiences. Our research efforts in the West Indies have been supported in part through the generosity of private donors and in the Lesser Antilles in part through the National Geographic Society (No. 9726-92).

LITERATURE CITED Ackery, P. R. and R. I. Vane-Wright. 1984. Milkweed Butterflies: Their Cladistics and Biology. British Museum (Natural History), London. 425 pp. Alayo, P. D. and L. R. Hernandez. 1987. Atlas de las mariposas diurnas de Cuba (Lepidoptera: Rhopalocera). Editorial Cientifico-tecnica, La Habana. Anderson, R. A. 1974. Three new U.S. records and other unusual captures from the lower Florida keys. Journal of the Lepidopterists’ Society 28:354–358. Bates, M. 1936. The butterflies of Cuba. Bulletin of the Museum of Comparative Zoology, Harvard 78:63–258. Brower, L. P., M. A. Ivie, L. S. Fink, J. R. Watts, and R. A. Moranz. 1992. Life History of Anetia briarea and its bearing on the evolutionary relationships of the Danainae (Lepidoptera: Nymphalidae). Tropical Lepidoptera 3:64–73. Brown, F. M. 1976. Oligodonta florissantensis, gen. n., sp. nov. (Lepidoptera:Pieridae). Bulletin of the Allyn Museum (37):4 pp. Brown, F. M. 1978. The origins of the West Indian butterfly fauna. Pp. 5–30 in Gill, F. B. (ed.). Zoogeography in the Caribbean. Academy of Natural Sciences, Philadelphia, Special Publication 13. Brown, F. M. and B. Heineman. 1971. Jamaica and Its Butterflies. E. W. Classey, London. 478 pp. Brundin, L. Z. 1971. Phylogenetics and systematics. Systematic Zoology 21:69–79. Brundin, L. Z. 1981. Croizat’s panbiogeography versus phylogenetic biogeography. Pp. 94–138 in Nelson, G. and D. E. Rosen (eds.). Vicariance Biogeography: A Critique. Columbia University Press, New York. Burke, K. 1988. Tectonic evolution of the Caribbean. Annual Review of Earth and Planetary Sciences 16:201–230. Burke, K., C. Cooper, J. F. Dewey, P. Mann, and J. L. Pindell. 1984. Caribbean tectonics and relative plate motions. Geological Society of America Memoir 162:31–63. Burns, J. M. 1985. Wallengrenia otho and W. egeremet in eastern North America (Lepidoptera: Hesperiidae: Hesperiinae). Smithsonian Contributions to Zoology 423:1–39. Burns, J. M. 1987. The big shift: nabokovi from Atalopedes to Hesperia (Hesperiidae). Journal of the Lepidopterists’ Society 41:173–186. Burns, J. M. 1989. Phylogeny and zoogeography of the bigger and better genus Atalopedes (Hesperiidae). Journal of the Lepidopterists’ Society 43:11–32. Cain, S. A. 1944. Foundations of Plant Geography. Harper & Brother, New York. 556 pp. Carey, S. W. 1958. A tectonic approach to continental drift. Pp. 177–355 in Carey, S. W. (convenor). Continental Drift, A Symposium. Hobart University, Tasmania. Carpenter, F. M. 1930. A review of our present knowledge of the geological history of insects. Psyche 37:15–23. Carlquist, S. J. 1974. Island Biology. Columbia University Press, New York. Case, J. E., T. L. Holcombe, and R. G. Martin. 1984. Map of geological provinces in the Caribbean region. Geological Society of America Memoir 162:1–30. Clench, H. K. 1964. A synopsis of the West Indian Lycaenidae, with remarks on their zoogeography. Journal of Research on the Lepidoptera 2:247–270. Clench, H. K. 1977. A list of the butterflies of Andros, Bahamas. Annals of the Carnegie Museum 49:1–30. Clench, H. K. and K. J. Bjorndahl. 1980. Butterflies of Great and Little Inagua, Bahamas. Annals of the Carnegie Museum 49:1–30. Common, I. F. B. 1975. Evolution and classification of the Lepidoptera. Annual Review of Entomology 20:183–203. Common, I. F. B. 1990. Moths of Australia. Melbourne University Press, Melbourne.

The Biogeography of the West Indian Butterflies (Lepidoptera)

151

Comstock, W. P. 1944. Insects of Porto Rico and the Virgin Islands. Rhopalocera or butterflies. Pp. 421–622 in Scientific Survey of Porto Rico and the Virgin Islands, 12, Part 4. New York Academy of Sciences, New York. Comstock, W. P. and E. I. Huntington. 1943. Lycaenidae of the Antilles (Lepidoptera, Rhopalocera). Annals of the New York Academy of Sciences 45:49–130. Coney, P. J. 1983. Plate tectonic constraints on biogeographic connections between North and South America. Annals of the Missouri Botanical Gardens 69:432–443. Cracraft, J. 1973. Continental drift, paleoclimatology, and the evolution and biogeography of birds. Journal of Zoology (London) 169:455–545. Croizat, L. 1958. Panbiogeography. Private publication, Caracas, Venezuela. Croizat, L., G. Nelson, and D. E. Rosen. 1974. Centers of origin and related concepts. Systematic Zoology 23:265–287. Darlington, P. J. 1957. Zoogeography: The Geographic Distribution of Animals. John Wiley, New York. Darlington, P. J. 1965. Biogeography of the Southern End of the World: Distribution and History of the Far-Southern Life and Land, with an Assessment of Continental Drift. Harvard University Press, Cambridge, Massachusetts. Dickson, C. G. C. and D. M. Kroon (eds.). 1978. Pennington’s Butterflies of Southern Africa. A. Donker, Johannesburg. Donnelly, T. W. 1985. Mesozoic and Cenozoic Plate Evolution of the Caribbean region. Pp. 89–121 in Stehli, F. G. and S. D. Webb (eds.). The Great American Biotic Interchange. Plenum Press, New York. Donnelly, T. W. 1988. Geologic constraints on Caribbean biogeography. Pp. 15–37 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Comstock Publishing Associates, Cornell University Press, Ithaca, New York. Durden, C. J. and H. Rose. 1978. Butterflies of the middle Eocene: the earliest occurrence of fossil Papilionidae. Pearce-Sellard Series, Texas Memorial Museum 28:1–25. Du Toit, A. L. 1927. A geological comparison of South America with South Africa. Publication of the Carnegie Institute of Washington 381:1–158. Du Toit, A. L. 1937. Our Wandering Continents: An Hypothesis of Continental Drift. Oliver & Boyd, Edinburgh. Eliot, J. N. 1973. The higher classification of the Lycaenidae (Lepidoptera): a tentative arrangement. Bulletin of the British Museum (Natural History), Entomology 28:371–505. Emmel, T. C. and J. F. Emmel. 1990. The life history and ecology of Hesperia nabokovi in the Dominican Republic (Lepidoptera: Hesperiidae). Tropical Lepidoptera 1:77–82. Endler, J. A. 1982. Pleistocene forest refuges: fact or fancy? Pp. 641–657 in Prance, G. T. (ed.). Biological Diversification in the Tropics. Columbia University Press, New York. Evans, W. H. 1937. A Catalogue of the African Hesperiidae in the British Museum (Natural History). British Museum Trustees, London. Evans, W. H. 1949. A Catalogue of the Hesperiidae of Europe, Asia, and Australia in the British Museum (Natural History). British Museum Trustees, London. 502 pp. Evans, W. H. 1952. A Catalogue of the American Hesperiidae in the British Museum (Natural History), Part 2. British Museum Trustees, London. Evans, W. H. 1955. A Catalogue of the American Hesperiidae in the British Museum (Natural History), Part 4. British Museum Trustees, London. Flint, O. S., Jr. 1977. Probable origins of the West Indian Trichoptera and Odonata faunas. Pp. 215–223. in Crichton, M. I. (ed.). Proceedings of the Second International Symposium on Trichoptera, Reading, U.K. Forbes, W. T. M. 1932. How old are the Lepidoptera? American Naturalist 66:452–460. Fox, R. M. 1963. Affinities and distribution of the Antillean Ithomiidae. Journal of Research on the Lepidoptera 2:173–184. Guyer, C. and J. M. Savage. 1986. Cladistic relationships among anoles (Sauria: Iguanidae). Systematic Zoology 35:509–531. Hancock, D. L. 1983. Classification of the Papilionidae (Lepidoptera): a phylogenetic approach. Smithersia (2):48 pp. Harvey, D. J. and H. K. Clench. 1980. Dianesia, a new genus of Riodinidae from the West Indies. Journal of the Lepidopterists’ Society 34:127–132. Hedges, S. B. 1989. Evolution and biogeography of West Indian frogs of the genus Eleutherodactylus: slowevolving Loci and the Major Groups. Pp. 305–370 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

152

Biogeography of the West Indies: Patterns and Perspectives

Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana. Hernandez, L. R., L. D. Miller, J. Y. Miller, M. J. Simon, and T. W. Turner. 1998. New records and range extensions of butterflies from eastern Cuba. Caribbean Journal of Science 34:324–327. Holloway, J. D. 1979. A Survey of the Lepidoptera, Biology and Ecology of New Caledonia. W. Junk, The Hague. Howe, W. H. 1975. Eunica. Pp. 126–127 in Howe, W. H. (ed.). The Butterflies of North America. Doubleday & Co., New York. Ivie, M. A., T. K. Philips, and K. A. Johnson. 1990. High altitude aggregation of Anetia briarea Godart on Hispaniola (Nymphalidae: Danaidae). Journal of the Lepidopterists’ Society 44: 209–214. Jenkins, D. W. 1984. Neotropical Nymphalidae. II. Revision of Mycelia. Bulletin of the Allyn Museum (87):64 pp. Jenkins, D. W. 1990. Neotropical Nymphalidae VIII. Revision of Eunica. Bulletin of the Allyn Museum 131:1–177. Johnson, K. 1991. Cladistics and biogeography of two trans-Caribbean hairstreak genera: Nesiostrymon and Terra (Lepidoptera, Lycaenidae). American Museum Novitates 3011:1–43. Johnson, K. and H. Descimon. 1989. Proper generic and species status of Antillean “Prepona” butterflies with description of a new subspecies from Puerto Rico (Nymphalidae: Charaxinae). Caribbean Journal of Science 25:45–53. Johnson, K. and D. S. Smith. 1993. A remarkable new butterfly species from Jamaica (Lepidoptera: Lycaenidae), with notes on Jamaican Endemics and their sister species. Reports of the Museum of Natural History, University of Wisconsin (Steven’s Point) 24:1–14. Jordan, K. 1908 (1907–1909). Papilionidae. Pp. 11–51 in Seitz, A. (ed.). The Macrolepidoptera of the World, Vol. 5. A. Kernan, Stuttgart. Klots, A. B. 1951. A Field Guide to the Butterflies of North America, East of the Great Plains. Houghton Mifflin, Boston, Massachusetts. Labandeira, C. C. and J. I. Sepkoski. 1993. Insect diversity in the fossil record. Science 261:310–315. Liebherr, J. K. 1986. Barylaus, a new genus (Coleoptera: Carabidae) endemic to the West Indies with Old World affinities. Journal of the New York Entomological Society 94:83–97. Liebherr, J. K. 1988. Biogeographic patterns of West Indian Platynus carabid beetles (Coleoptera). Pp. 121–152 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Comstock Publishing Associates, Cornell University Press, Ithaca, New York. Lundberg, J. G., A. Machado-Allison, and R. F. Kay. 1986. Miocene characid fishes from Colombia: evolutionary stasis and extirpation. Science 234:208–209. MacArthur, R. H. and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, New Jersey. Matthew, G. A. 1915. Climate and evolution. Annals of the New York Academy of Science 24:171–318. Maury, R. C., G. K. Westbrook, P. E. Baker, P. Bouysse, and D. Westercamp. 1990. Geology of the Lesser Antilles. Pp. 141–146 in Dengo, G. and J. E. Case (eds.). The Caribbean Region, The Geology of North America, Vol. H. Geological Society of America, Boulder, Colorado. Miller, J. S. 1987a. Phylogenetic studies in the Papilioninae (Lepidoptera: Papilionidae). Bulletin of the American Museum of Natural History 186:365–512. Miller, J. S. 1987b. Host-plant relationships in the Papilionidae (Lepidoptera): parallel cladogenesis or colonization? Cladistics 3:105–120. Miller, J. Y. 1986. The Taxonomy, Phylogeny and Zoogeography of the Neotropical Moth Subfamily Castniinae (Lepidoptera: Castnioidea: Castniidae). Unpublished Ph.D. dissertation, University of Florida, Gainesville. Miller, J. Y. In preparation. A review of the Neotropical species of Wallengrenia Berg (Lepidoptera:Hesperiidae) with description of two new taxa. Miller, J. Y. and F. M. Brown. 1989. A new Oligocene fossil butterfly, Vanessa amerindica (Lepidoptera: Nymphalidae) from the Florissant formation, Colorado. Bulletin of the Allyn Museum 126:1–9. Miller, L. D. 1965. A review of the West Indian “Choranthus.” Journal of Research on the Lepidoptera 4:259–274. Miller, L. D. 1968. The higher classification, phylogeny and zoogeography of the Satyridae (Lepidoptera). Memoirs of the American Entomological Society 24:1–174. Miller, L. D. In preparation. The higher classification, phylogenetics, and biogeography of the Satyridae (Lepidoptera): a reappraisal. [tentative title].

The Biogeography of the West Indian Butterflies (Lepidoptera)

153

Miller, L. D. and F. M. Brown. 1983. Butterfly taxonomy: a reply. Journal of Research on the Lepidoptera 20:193–198. Miller, L. D. and J. Y. Miller. 1989. The biogeography of the West Indian Butterflies (Lepidoptera: Papilionoidea, Hesperioidea): a vicariance model. Pp. 229–262 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Miller, L. D. and J. Y. Miller. 1997. Gondwanian butterflies: the African-South America connection. Metamorphosis (Suppl. 3):42–51. Miller, L. D. and M. J. Simon. 1998. A preliminary annotated list of the butterflies of Cat Island, Bahamas. Bahamas Journal of Science 6:35–42. Miller, L. D. and M. J. Simon. 1998. Rediscovery of the rare, “endemic” Cuban butterfly, Achylodes munroei Bell, with notes on its behavior and possible origin. Caribbean Journal of Science 34:325–329. Miller, L. D. and S. R. Steinhauser. 1992. Butterflies of the Cayman Islands, with description of a new subspecies. Journal of the Lepidopterists’ Society 46:119–127. Miller, L. D., D. J. Harvey, and J. Y. Miller. 1985. Notes on the genus Euphyes, with description of a new subspecies (Lepidoptera: Hesperiidae). Florida Entomologist 68:323–325. Miller, L. D., M. J. Simon, and D. J. Harvey. 1992. The butterflies (Insecta: Lepidoptera) of Crooked, Acklins, and Mayaguana Islands, Bahamas, with a discussion of the biogeographical affinities of the southwestern Bahamas and description of a new subspecies by H. K. Clench. Annals of the Carnegie Museum 61:1–31. Minno, M. C. and T. C. Emmel. 1993. Butterflies of the Florida Keys. Scientific Publishers, Gainesville, Florida. Murata, Y. 1998. Notes of the evolutionary process of butterflies and butterfly fossils. Part 1. Butterflies. The Butterfly Society of Japan 20:4–17. Nijout, H. F. 1991. The Development and Evolution of Butterfly Wing Patterns. Smithsonian Institution Press, Washington, D.C. Patterson, C. 1981. Methods of paleobiogeography. Pp. 446–489 in Nelson, G. and D. E. Rosen (eds.). Vicariance Biogeography: A Critique. Columbia University Press, New York. Perfitt, M. R. and B. C. Heezen. 1978. The geology and evolution of the Cayman Trench. Geological Society of America Bulletin 89:1155–1174. Perfitt, M. R. and E. E. Williams. 1989. Geological constraints and biological retrodictions in the evolution of the Caribbean Sea and its islands. Pp. 47–102 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Pinchon, R. and P. Enrico. 1969. Faune des Antilles Françaises. Les papillons. Fort de France, Martinique. Pindell, J. and J. F. Dewey. 1982. Permo-Triassic reconstruction of western Pangea and the evolution of the Gulf of Mexico/Caribbean region. Tectonics 1:179–211. Pindell, J. L. and S. F. Barrett. 1990. Geological evolution of the Caribbean regions: a plate tectonic perspective. Pp. 405–432 in Dengo, G. and J. E. Case (eds.). The Caribbean Region. The Geology of North America, Vol. H. Geological Society of America, Boulder, Colorado. Powell, J. E. 1980. Evolution of larval food preferences in Microlepidoptera. Annual Review of Entomology 25:133–159. Preston, F. W. 1962. The canonical distribution of commonness and rarity. Ecology 43:185–215, 410–432. Ramos, S. J. 1996. An annotated checklist of the butterflies of Puerto Rico. Annals of the New York Academy of Sciences 776:217–236. Raven, P. H. and D. I. Axelrod. 1974. Angiosperm biogeography and past continental movements. Annals of the Missouri Botanical Garden 61:539–673. Riley, N. D. 1975. A Field Guide to the Butterflies of the West Indies. Collins Sons & Co., London. Romeo, E. J. and S. Archangelsky. 1986. Early Cretaceous angiosperm leaves from southern South America. Science 234:1580–1582. Rosen, D. E. 1976. A vicariance model of Caribbean biogeography. Systematic Zoology 24:431–464. Rosen, D. E. 1985. Geological hierarchies and biogeographic congruence in the Caribbean. Annals of the Missouri Botanical Garden 72:636–659. Savage, J. M. 1973. The geographic distribution of frogs: patterns and predictions. Pp. 349–445 in Vial, J. L. (ed.). Evolutionary Biology of the Anura. University of Missouri Press, Columbia. Savage, J. M. 1974. The isthmian link and the evolution of Neotropical mammals. Contributions in Science of the Natural History Museum of Los Angeles County 260:1–51. Savage, J. M. 1983. The enigma of the Central American herpetofauna: dispersals or vicariance? Annals of the Missouri Botanical Garden 69:464–547.

154

Biogeography of the West Indies: Patterns and Perspectives

Schuchert, C. 1935. Historical Geology of the Antillean-Caribbean Region or Lands Bordering the Gulf of Mexico and the Caribbean Sea. John Wiley & Sons, New York. Schwanwitsch, B. N. 1924. On the ground-plan of wing-pattern in nymphalids and certain other families of rhopalocerous Lepidoptera. Proceedings of the Zoological Society London 1924:509–528. Schwartz, A. 1983. Haitian Butterflies. Special Publication of the Museo Nacional de Historia Natural, Santo Domingo, Dominican Republic. Schwartz, A. 1989. The Butterflies of Hispaniola. University of Florida Press, Gainesville. Scott, J. A. 1972. Biogeography of the Antillean butterflies. Biotropica 4:32–45. Scott, J. A. 1986. Distribution of Caribbean butterflies. Papilio (new series) 3:1–26. Shields, O. 1979. Evidence for initial opening of the Pacific Ocean in the Jurassic. Palaeogeography, Palaeoclimatology, Palaeoecology 26:181–220. Shields, O. 1985. Zoogeography of the Libytheidae (Snouts or Breaks [sic]). Tokurana 9:1–58. Shields, O. 1989. Systematic position of the Libytheidae, diphylogeny of Rhopalocera, and heteroceran ancestry of Rhopalocera (Lepidoptera). Tyo`to Ga, Lepidopterological Society of Japan, 40:197–228. Shields, O. and S. V. Dvorak. 1979. Butterfly distribution and continental drift between the Americas, the Caribbean and Africa. Journal of Natural History 13:221–250. Shreeve, T. G. 1992. Monitoring butterfly movements. Pp. 120–138 in Dennis, R. L. H. (ed.). The Ecology of Butterflies in Britain. Oxford University Press, Oxford. Shuey, J. A. 1986. The Ecology and Evolution of Wetland Butterflies, Emphasis on the Genus Euphyes (Lepidoptera: Hesperiidae). Unpublished Ph.D. dissertation, Ohio State University, Columbus. Shuey, J. A. 1994. Phylogeny and biogeography of Euphyes Scudder (Hesperiidae). Journal of the Lepidopterists’ Society 47:261–278. Simon, M. J. and L. D. Miller. 1986. Observations on the butterflies of Great Inagua Island, Bahamas, with records of three species new to the island. Bulletin of the Allyn Museum (105):14 pp. Simpson, G. G. 1952. Probabilities of dispersal in geologic time. Bulletin of the American Museum of Natural History 99:163–176. Simpson, G. G. 1956. Zoogeography of West Indian mammals. American Museum Novitates 1759:1–28. Smart, J. and N. F. Hughes. 1973. The insect and the plant: progressive palaeoecological integration. Pp. 143–155 in van Emden, H. F. (ed.). Insect/Plant Relationships. Symposium of the Royal Entomological Society of London, No. 6. Smith, D. L. 1985. Caribbean plate relative motions. Pp. 17–48 in Stehli, F. G. and S. D. Webb (eds.). The Great American Biotic Interchange. Plenum Press, New York. Smith, D. S., S. J. Ramos, F. McKenzie, E. Munroe, and L. D. Miller. 1988. Biogeographical affinities of the butterflies of a “forgotten” island: Mona (Puerto Rico). Bulletin of the Allyn Museum (121):1–35. Smith, D. S., E. G. Munroe, and S. A. Ramos. 1989. Biogeographical affinities of the butterflies of the southwest Caribbean islands, San Andres and Providencia. Journal of the Lepidopterists’ Society 43:274–288. Smith, D. S., L. D. Miller, and F. McKenzie. 1991. The butterflies of Anegada, British Virgin Islands, with descriptions of a new Calisto (Satyridae) and a new Copeodes (Hesperiidae) endemic to the island. Bulletin of the Allyn Museum (133):1–25. Smith, D. S., L. D. Miller, and J. Y. Miller. 1994. The Butterflies of the West Indies and South Florida. Oxford University Press, Oxford. Sourakov, A. 1996. Notes on the genus Calisto. 1. Tropical Lepidoptera 7:91–112. Stehii, F. G. and S. D. Webb (eds.). 1985. The Great American Biotic Interchange. Plenum Press, New York. Sun, G., D. L. Dilcher, S. Zheng, and Z. Zhou. 1998. In search of the first flower: a Jurassic angiosperm, Archaefructus from northeast China. Science 282:1092–1095. Torre y Callejas, S. L. de la. 1954. An annotated list of the butterflies and skippers of Cuba. Journal of the New York Entomological Society 62:1–25, 113–128, 189–192, 207–249. Wallace, A. R. 1876. The Geographical Distribution of Animals. Vols. 1 and 2. Harpers, New York. Webb, S. A. 1990. Historical Biogeography. Pp. 70–100 in Myers, R. L. and J. J. Ewel (eds.). Ecosystems of Florida. University of Central Florida Press, Orlando. Wegener, A. 1915. Die Entstehung der Kontinente und Ozeane. Sammlung Vieweg, 23:F. Vieweg und Sohn, Brauschweig. 144 pp. Whalley, P. 1986. A review of the current fossil evidence of Lepidoptera in the Mesozoic. Biological Journal of the Linnean Society 28:253–271. White, M. E. 1990. The Flowering of Gondwana. Princeton University Press, Princeton, New Jersey.

The Biogeography of the West Indian Butterflies (Lepidoptera)

155

Wilson, J. T. 1963. Hypothesis of earth’s behavior. Nature (London) 198:925–929. Woods, C. A. 1989. The biogeography of West Indian rodents. Pp. 741–798 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

and 11 Relationships Divergence Times of West Indian Amphibians and Reptiles: Insights from Albumin Immunology Carla Ann Hass, Linda R. Maxson, and S. Blair Hedges Abstract — Seven families of West Indian reptiles and two families of West Indian amphibians were investigated using the immunological technique of micro-complement fixation. These data allowed the examination of relationships among members of each of these families and provided estimates of divergences times, both within the West Indies as well as between mainland and island taxa. Some of these groups are the result of a relatively recent colonization and subsequent radiation (e.g., xenodontine snakes). Other groups show deeper divergences among the West Indian species and were much earlier arrivals to these islands (e.g., amphisbaenids). When examined in conjunction with the geological history of the Caribbean, the divergence times derived from the immunological data suggest overwater dispersal as the primary mechanism for colonization of the West Indies by these terrestrial vertebrates.

INTRODUCTION The islands of the West Indies harbor a diverse array of reptiles and amphibians, ranging from the minute gecko Sphaerodactylus parthenopion to the imposing Crocodylus rhombifer, from the blind, burrowing snake Typhlops to the arboreal hylid frogs. Endemic species representing five of the six extant major reptilian lineages (turtles, crocodilians, snakes, lizards, and amphisbaenians) are found on these islands, as are members of four families of frogs (Bufonidae, Dendrobatidae, Hylidae, and Leptodactylidae). Many species are endemic to individual islands, and often are restricted to very small areas within those islands (Schwartz and Henderson, 1991). In addition to morphological studies, data obtained by looking at the variation of molecules among species, both through indirect comparisons of proteins as well as direct comparisons of nucleotide sequences, can offer insights into both the relationships among taxa as well as the timing of divergence events within groups. While there have been molecular studies of some West Indian groups (see Hedges, 1996a, for references) most groups have not been investigated. In this study, we present immunological information for seven groups of reptiles and two families of frogs. The technique of micro-complement fixation (MC’F), which allows the estimation of the number of amino acid differences between proteins, has been used to investigate relationships and divergence times in many vertebrate groups, including West Indian taxa (Hedges et al., 1992; Hass et al., 1993; Hedges, 1996a). The new data presented here extend those studies to additional groups and species of West Indian amphibians and reptiles.

MATERIALS AND METHODS Antisera were prepared against 11 species of West Indian reptiles from seven squamate families, and 9 species of amphibians, representing two anuran families (Table 1). The collection localities 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

157

158

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 Titers and Slopes for Antisera against Serum Albumin from Selected West Indian Reptiles and Amphibians Family Bufonidae

Hylidae

Amphisbaenidae Anguidae

Iguanidae Teiidae Colubridae Tropidophidae Typhlopidae

Species

No. of Rabbits

Titer

Slope

Bufo guentheri (BG) B. marinus (BM) B. peltocephalus (BP) Calyptahyla crucialis (CC) Hyla vasta (HV) Osteopilus brunneus (OB) O. dominicensis (OD) O. septentrionalis (OS) Osteocephalus taurinus (OT) Amphisbaena schmidti (AM) Diploglossus delasagra (DD) D. pleei (DP) D. warreni (DW) Wetmorena haetiana (WE) Leiocephalus schreibersi (LE) Ameiva chrysolaema (AC) A. exsul (AE) Arrhyton landoi (AR) Tropidophis haetianus (TR) Typhlops platycephalus (TY)

1 2 2 2 2 2 2 3 2 1 1 1 2 1 1 1 1 1 1 1

1/2800 1/4200 1/1400 1/2000 1/1600 1/1800 1/3000 1/4700 1/1500 1/3200 1/3000 1/4000 1/1600 1/1500 1/4200 1/1800 1/4600 1/2400 1/4400 1/2700

350 400 370 350 460 400 400 400 330 400 320 300 350 400 450 330 360 450 450 350

for these species, and the species used as antigens are listed in the Appendix. The animals were sacrificed using cryothermy (Kennedy and Brockman, 1965). In some cases, plasma samples to be used for antibody preparation were pools of multiple individuals from the same population (Appendix). Albumin was isolated from pure plasma or plasma preserved with PPS using polyacrylamide electrophoresis modified from the method of Davis (1964). Antibodies to this extracted albumin were prepared in female New Zealand white rabbits following the method of Maxson et al. (1979) as modified by Hass and Hedges (1991). When antisera were prepared in more than one rabbit, the individual rabbit antisera were pooled in inverse proportion to their titers. Hutchinson and Maxson (1986) showed that antibodies prepared using one rabbit give approximately the same estimates of immunological distance (ID) as do antibody pools. Micro-complement fixation experiments were performed using standard protocols (Maxson and Maxson, 1990). The data are reported as ID units. These data sets provide primarily one-way estimates of ID values. While reciprocal comparisons give a more accurate approximation of ID values between taxa, one-way distances are useful indicators of the degree of amino acid difference between the albumins of two species. Data sets with reciprocal ID measurements were tested for nonrandom deviations from perfect reciprocity and, when appropriate, the data were corrected by the method of Cronin and Sarich (1975); these corrected data are used in discussion of the data sets. An independent calibration of the albumin immunological clock for each group investigated in this study is not possible because of the lack of fossil information or independent geological events to use as calibration points. A “standard” calibration (1 ID unit = 0.6 million years of divergence) has been derived for a number of vertebrate groups based upon both fossil and geological information (Wilson et al., 1977; Maxson, 1992), and its consistency over diverse vertebrate lineages justifies its use in this type of study (see the more detailed discussion in Hedges, 1996a). Some of

Relationships and Divergence Times of West Indian Amphibians and Reptiles

159

the new data reported here, and their taxonomic implications, already have been discussed elsewhere (Hedges et al., 1992; Hedges, 1996a).

RESULTS BUFONIDAE We were able to examine 5 of the 11 species of Bufo endemic to the West Indies using MC’F (Table 2). The immunological data suggest that the Cuban species B. peltocephalus and B. taladai are the most closely related species examined, at an ID of 8. The other Cuban species examined, B. longinasus, was as distinct from B. peltocephalus as was the Hispaniolan species B. guentheri. Bufo guentheri and B. lemur were separated by an ID value of 18. All of the Cuban species gave ID values within the same range (31 to 38) from B. guentheri. Two mainland taxa were examined; B. granulosus appeared to be closer to the West Indian species than was B. marinus. An antiserum to B. marinus gave similar reciprocal values to the West Indian species (0 = 87 ID; approximate values were excluded from means).

HYLIDAE There were noticeable deviations in reciprocity in this data set, particularly for the Calyptahyla crucialis and Hyla vasta antisera (Table 3). Therefore, the data matrix was corrected using the method of Cronin and Sarich (1975). The most closely related species are C. crucialis and an undescribed Osteopilus species from Jamaica, with an ID value of 11. Calyptahyla crucialis and O. brunneus also showed a low mean ID value (16). These two species gave low ID values to the Jamaican species H. marianae, with higher values to H. wilderi. The Hispaniolan species O. dominicensis had an ID of 20 to both H. pulchrilineata, another Hispaniolan species, and to the Jamaican species H. marianae; the Jamaican species C. crucialis and Osteopilus sp. nov. were slightly more divergent. The Cuban species O. septentrionalis gave marginally higher average ID

TABLE 2 One-Way Immunological Distances from Bufo guentheri (BG), B. peltocephalus (BP), and B. marinus (BM) Antisera to Other West Indian Bufonids and Representative Mainland Species BG Island Hispaniola Cuba

Puerto Rico Mainland

Species Bufo guentheri (BG) B. peltocephalus (BP) B. longinasus B. taladai B. lemur B. granulosus B. marinus (BM)

BP

BM

Correction Factor 1.23 0.84 — 0 30 (37) 25 (31) 31 (38) 15 (18) — 96

44 (37) 0 46 (39) 10 (8) ~53 (~44) 62 (52) >100 (>84)

~96 79 — — — 61a 0

Note: Reciprocal ID values are in bold. Values in parentheses are corrected ID values based upon reciprocal comparisons. A dash indicates that experiment was not performed. Approximate estimates are indicated by ~. a

ID value from Maxson, 1984.

160

Biogeography of the West Indies: Patterns and Perspectives

TABLE 3 Immunological Distances from Five Antisera to Other West Indian Hylids and a Representative Mainland Hylid CC Island Jamaica

Hispaniola

Cuba Mainland

Species Calyptahyla crucialis (CC) Hyla marianae H. wilderi Osteopilus brunneus (OB) Osteopilus sp. nov. H. heilprini H. pulchrilineata H. vasta (HV) O. dominicensis (OD) O. septentrionalis (OS) Osteocephalus taurinus (OT)

OB

0.588

1.36

0 25 (18) 36 (26) 20 (15) 15 (11) — 26 (19) 73 (53) 39 (28) 43 (31) ~118 (~86)

11 (17) 13 (20) 26 (41) 0 24 (38) — 20 (32) 27 (43) 19 (30) 35 (55) 98 (156)

HV

OD

Correction Factor 1.88 0.877 21 (40) 22 (41) 42 (79) 23 (43) 29 (54) 97 (182) 24 (45) 0 26 (49) 30 (56) —a

25 (26) 19 (20) 44 (46) 34 (35) 21 (22) — 19 (20) 42 (44) 0 37 (38) ~110 (~114)

OS

OT

0.819

—

46 (46) 39 (39) 49 (49) 48 (48) 48 (48) — 40 (40) 46 (46) 37 (37) 0 78 (78)

87 — — 100 — — — —a 73 83 0

Note: Reciprocal ID values are in bold. Values in parentheses are corrected ID values following the method of Cronin and Sarich; one-way ID values also were corrected. A dash indicates that experiment was not performed. Approximate estimates are indicated by ~. a

Experiment was done but no cross reaction was seen.

TABLE 4 One-Way Immunological Distances from Amphisbaena schmidti Antiserum to Other West Indian and Mainland Amphisbaenids Island

Species

AM

Puerto Rico

Amphisbaena bakeri A. caeca A. fenestrata A. schmidti (AM) A. xera A. caudalis A. gonavensis A. innocens A. manni A. cubana Cadea blanoides A. alba Rhineura floridana

31 39 28 0 35 60 144 83 16 29 69 91 156

Hispaniola

Cuba Mainland

values to the other West Indian taxa, ranging from 37 to 49 ID. Hyla vasta, from Hispaniola, was more divergent, with IDs to the other taxa ranging from 40 to 79. Among all of the species endemic to the West Indies, H. heilprini was the most divergent. Only the H. vasta antiserum, which strongly underestimates ID values, would cross-react with H. heilprini and the adjusted ID value was 182. The West Indian taxa gave lower ID values, ranging from 73 to 156, to Osteocephalus taurinus, a mainland species (x = 96 ID).

Relationships and Divergence Times of West Indian Amphibians and Reptiles

161

TABLE 5 Immunological Distances among West Indian Members of the Family Anguidae DW Island Hispaniola

Jamaica

Cuba Puerto Rico Mainland

Species Celestus curtissi #1 C. curtissi #2 C. darlingtoni C. stenurus C. macrotus C. sp. nov. Diploglossus carraui D. warreni (DW) Sauresia agasepsoides S. sepsoides Wetmorena haetiana (WE) Celestus barbouri C. crusculus crusculus C. c. cundalli D. delasagra (DD) D. pleii (DP) Ophiodes striatus

0.72 14 (10) 13 (9) 2 (1) 14 (10) 16 (12) 10 (7) 0 0 9 (6) 12 (7) 14 (10) 13 (9) 15 (11) 15 (11) 129 (94) ~140 (~101) —

DD

DP

Correction Factor 1.3 1.06

WE

—

8 (10) 5 (6) 4 (5) 4 (5) 13 (17) 5 (6) 4 (5) 4 (5) 5 (6) 5 (6) 0 10 (13) 9 (12) 13 (17) 86 (112) 86 (112) 54

102 (108) 105 (111) 102 (108) 102 (108) 97 (103) 99 (105) 94 (100) 99 (105) 97 (103) 98 (104) 103 (109) 98 (104) 114 (121) 101 (107) 0 26 (28) —

— —a — —a — — — —a — — —a — — — 46 0 —

Note: Reciprocal ID values are in bold. Values in parentheses are corrected ID values based upon reciprocal comparisons. A dash indicates that experiment was not performed. Approximate estimates are indicated by ~. a

Experiment was done but no cross reaction was seen.

AMPHISBAENIDAE These data suggest that the closest relative to the Puerto Rican species Amphisbaena schmidti is a Hispaniolan species, A. manni (Table 4). The other Puerto Rican species ranged from 28 to 39 ID units from A. schmidti. The Cuban species A. cubana also was within this range. The Hispaniolan species A. caudalis and the Cuban species Cadea blanoides showed similar levels of divergence from A. schmidti. The Hispaniolan species A. innocens and the mainland species A. alba showed a higher level of divergence. Finally, the Hispaniolan species A. gonavensis showed a degree of divergence similar to that of a mainland species placed in another family, Rhineura floridana (Rhineuridae).

ANGUIDAE The albumin ID values show a clear dichotomy within the West Indian anguid lizards (Table 5). Two species of Diplogossus, D. delasagra (Cuba) and D. pleii (Puerto Rico), had ID values of approximately 107 units (mean of corrected values) to the other West Indian anguids examined. These two species had a mean corrected reciprocal value of 37 from each other. In contrast, the antisera against the two Hispaniolan species, Wetmorena haetiana and D. warreni, consistently gave low ID values (ranging from 0 to 17) to the species of Celestus, Sauresia, and Hispaniolan Diploglossus examined. These data do not provide sufficient resolution to examine the phylogenetic relationships among the members of this group. The mainland species Ophiodes striatus was tested against the antiserum to W. haetiana and gave an ID value of 54.

162

Biogeography of the West Indies: Patterns and Perspectives

TABLE 6 One-Way Immunological Distances from the Leiocephalus schreibersi Antiserum to Other West Indian Leiocephalus and Representative Mainland Iguanids Island Hispaniola

Cuba

Bahamas — East Plana Cay Bahamas — Great Inagua Bahamas — San Salvador Bahamas — Acklin’s Island Mainland

Species

LE

Leiocephalus schreibersi (LE) L. barahonensis #1 L. barahonensis #2 L. lunatus L. melanochlorus L. personatus L. semilineatus L. carinatus L. cubensis L. macropus L. raviceps L. stictigaster L. greenwayi L. inaguae L. loxogrammus L. punctatus Crotaphytus collaris Sceloporus spinosus Tropidurus peruvianus T. hisipidus

0 0 3 1 1 1 4 3 14 7 10 4 6 0 10 7 40 (46) 63 (58) 83 (80) 87

Note: ID values in parentheses are reciprocal values from antisera of other iguanid species to L. schreibersi.

IGUANIDAE The one-way immunological distances from the antiserum against Leiocephalus schreibersi, a Hispaniolan species, give some insights into the relationships among the members of this endemic West Indian genus (Table 6). It is clear that all of the species examined are closely related to one another. The largest immunological distance was 14 ID. There are many species that gave ID values within the error range of the technique (±2 ID units), indicating that their albumin molecules have very similar amino acid sequences. The ID values to Hispaniolan species ranged from 0 to 4 ID, while the ID values to the Cuban species ranged from 3 to 14 ID. One Bahamian species, L. inaguae, had an ID of 0 to L. schreibersi. The other Bahamian species, L. greenwayi, L. loxogrammus, and L. punctatus, had ID values within the range of those seen to Cuban species. The Leiocephalus antiserum also was used to determine ID values for some mainland species within the family Iguanidae. The lowest ID value (with a mean value of 43 for reciprocal comparisons) seen was to Crotaphytus collaris, a species that occurs in North America. Sceloporus spinosus, another North American species, gave a mean ID value of 60. The two South American Tropidurus examined, T. peruvianus and T. hispidus, gave ID values over 80.

TEIIDAE The two antisera prepared do not show any clearly defined patterns of albumin variation among the West Indian Ameiva (Table 7). Ameiva chrysolaema and A. chrysolaema #3, both from Hispaniola, are very closely related, with an ID value of 1. However, the remaining species of West Indian

Relationships and Divergence Times of West Indian Amphibians and Reptiles

163

TABLE 7 One-Way Immunological Distances from Ameiva chrysolaema (AC) and A. exsul (AE) Antisera to Other West Indian Ameiva and Representative Mainland Teiids AC Island Hispaniola

Puerto Rico Cuba Bahamas Lesser Antilles — St. Kitts Lesser Antilles — Antigua Lesser Antilles — Montserrat Mainland

Species Ameiva chrysolaema (AC) A. chrysolaema #3 A. lineolata A. taeniura A. exsul (AE) A. wetmorei A. auberi A. maynardi A. erythrocephala A. griswoldi A. pluvianotata A. ameiva Cnemidophorus uniparens Tupinambis

AE

Correction Factor 1.23 0.86 0 1 (1) 38 (47) — 44 (54) 30 (37) 50 (62) 40 (49) 37 (46) 34 (42) 42 (52) 50 (62) 53 (65) >144

63 (54) 61 (52) 47 (40) 46 (40) 0 39 (34) 64 (55) 44 (38) 70 (60) 50 (43) 58 (50) 69 (59) 73 (63) 168 (144)

Note: Reciprocal ID values are in bold. Values in parentheses are corrected ID values based upon reciprocal comparisons. A dash indicates that experiment was not performed.

Ameiva ranged from 34 to 62 ID units (corrected values). The mainland species A. ameiva was at the upper end of the range of IDs found within the West Indies. The ID value from A. exsul to A. ameiva previously has been reported to be 79 (Hedges et al., 1992); additional experiments gave a lower value (69). Two other teiid lizards were compared. Cnemidophorus uniparens gave ID values slightly higher than those seen to the mainland Ameiva, while the ID values to Tupinambis were very large, essentially at the measurement limit of the technique (Maxson and Maxson, 1986).

COLUBRIDAE All of the West Indian xenodontine snakes tested gave low ID values (20 or less) to the Arrhyton landoi antiserum (Table 8). Within the genus Arrhyton, all of the Cuban species examined gave ID values within the error limit of MC’F (±2 ID), indicating that their albumins have almost identical amino acid sequences. The Jamaican and Puerto Rican species gave slightly higher ID values, ranging from 6 to 11. However, this level of differentiation also was seen to members of three other West Indian genera, Antillophis, Hypsirhynchus, and Ialtris. The remaining three West Indian genera, Alsophis, Darlingtonia, and Uromacer, gave higher ID values (ranging from 12 to 20 ID). An ID value of 11 was reported for Darlingtonia by Hedges et al. (1992), but that value has been revised by further experiments. The South American xenodontines showed divergences ranging from 21 to 42 ID, and the Central American xenodontines were the most divergent taxa examined, with an average ID of 58.

TROPIDOPHIDAE A dichotomy in ID values among West Indian Tropidophis is obvious (Table 9). The antiserum against a T. haetianus from Jamaica gave much lower ID values (range 1 to 9) to eight species of Cuban and Bahamian Tropidophis than it did to T. haetinaus from Hispaniola. This dichotomy was

164

Biogeography of the West Indies: Patterns and Perspectives

TABLE 8 One-Way Immunological Distances from Arrhyton landoi (AR) Antiserum to Other West Indian and Mainland Xenodontine Snakes Island Cuba

Genus Arrhyton

Jamaica

Puerto Rico Cuba Hispaniola

Antillophis Hypsirhynchus Ialtris

Cuba Puerto Rico Bahamas — Nassau Lesser Antilles — Antigua Lesser Antilles — Montserrat Hispaniola

Mainland — South American Clade

Mainland — Central American Clade

Darlingtonia Alsophis

Uromacer

Liophis Oxyrhopus Thamnodynastes Xenodon Dipsas Leptodeira

Species dolichura landoi #1 (AR) landoi #2 procerum supernum #1 supernum #2 taeniatum tanyplectum vittatum callilaemum funereum polylepis exiguum andreae parvifrons ferox scalaris dorsalis #1 dorsalis #2 haetiana cantherigerus portoricensis vudii antiguae antillensis catesbyi #1 catesbyi #2 frenatus oxyrhynchus cabella melanostigma leucomelas severus catesbyi

AR 0 0 0 0 0 0 0 0 2 6 9 11 6 11 6 6 8 10 10 15 19 15 12 14 16 20 15 18 17 31 23 42 21 29 (42) 58 57

Note: The value in parentheses is a reciprocal ID value from an antiserum to Xenodon severus.

so striking, and unexpected, that multiple individuals from Hispaniola were examined to ensure that this difference was real. The ID values to those Hispaniolan snakes ranged from 23 to 29. The mainland species, T. paucisquamis, gave an ID value of 70, much larger than any values seen within the West Indies. Tropidophis does not seem to be clearly allied to any other lineage of snakes. Although the ID to Boa constrictor was lower than to the other lineages examined, that value (141) is approaching the upper limit of this technique (Maxson and Maxson, 1986).

TYPHLOPIDAE The antiserum against the Puerto Rican species Typhlops platycephalus gave low ID values (ranging from 1 to 5) to three Puerto Rican species and the two species from the Lesser Antilles (Table 10).

Relationships and Divergence Times of West Indian Amphibians and Reptiles

165

TABLE 9 One-Way Immunological Distances from Tropidophis haetianus (TR) Antiserum to Other West Indian and Mainland Tropidophiids, and Selected Representatives of Other Snake Families Island

Taxon

TR

Jamaica

Tropidophis haetianus stejnegeri #1 (TR) T. haetianus stejnegeri #2 T. haetianus stullae T. feicki T. fuscus T. maculatus T. melanurus T. pardalis T. pilsbryi T. wrighti T. canus T. haetianus haetianus #1 T. haetianus haetianus #2 T. haetianus haetianus #3 T. haetianus haetianus #4 T. haetianus haetianus #5 T. haetianus humerus T. paucisquamis Boidae: Boa constrictor Elapidae: Naja naja Pythonidae: Python molarus Colubridae: Coluber constrictor Colubridae: Arrhyton landoi

0 0 0 9 4 4 2 5 4 4 1 25 23 29 28 25 24 70 141 171 ~185 ~185 ~180

Cuba

Bahamas Hispaniola

Mainland

Note: Approximate estimated are indicated by ~.

The exception is the Puerto Rican T. rostellatus, which gave an ID value of 31. The Cuban species T. biminensis and T. lumbricalis both gave an ID value of 14. The widest range of ID values is seen among the Hispaniolan species, which ranged from 17 to 44. The Jamaican species T. jamaicensis gave an ID value within this range. Typhlops luzonensis, a species from the Phillipines, was quite divergent, at 96 ID. Species from two other scolecophidian families were included, Liotyphlops (Anomalepidae) and Leptotyphlops (Leptotyphlopidae), and both of these gave very high ID values to T. platycephalus.

DISCUSSION BUFONIDAE Two Cuban species, Bufo peltocephalus and B. taladai, shared a common ancestor approximately 5 million years ago (mya). Bufo peltocephalus diverged from the other species examined (from Cuba, Hispaniola, and Puerto Rico) about 22 to 23 mya. The Hispaniolan species B. guentheri and the Puerto Rican B. lemur diverged approximately 11 mya. The ID values from the B. guentheri antiserum give a divergence time of 19 to 23 mya from the three Cuban species examined. While the mainland affinities of this group are not yet known, the ID value to B. granulosus indicates that they had diverged from mainland taxa by about 31 mya.

166

Biogeography of the West Indies: Patterns and Perspectives

TABLE 10 One-Way Immunological Distances from Typhlops platycephalus (TY) Antiserum to Other West Indian and Mainland Typhlopids, and Selected Representatives of Other Scolecophidian Families Island Puerto Rico

Lesser Antilles Cuba Hispaniola

Jamaica Mainland

Taxon

TY

Typhlops granti T. hypomethes T. platycephalus (TY) T. richardi T. rostellatus T. guadaloupensis T. monastus T. biminensis T. lumbricalis T. capitulatus T. hectus T. pusillus T. schwartzi T. sulcatus T. syntherus T. titanops T. jamaicensis T. luzonensis Anomalepididae: Liotyphlops Leptotyphlopidae: Leptotyphlops

5 1 0 2 31 4 4 14 14 18 17 18 22 26 44 29 20 96 157 150

Maxson (1984) examined albumin ID variation in the genus Bufo and found that the Old World species (Eurasian and African) diverged from the New World species about 65 to 75 mya (~110 to 120 ID units). No West Indian toads were included in that study, but the data obtained here indicate that West Indian Bufo are much closer to New World species of Bufo than to Old World species. This does not support the recognition of a separate genus (Peltophryne) for the West Indian species. Although Graybeal (1997), in a study of cytochrome b DNA sequences in bufonids, did not significantly resolve the position of the B. peltocephalus group, she also found it to be nested among New World species of Bufo in her best supported trees.

HYLIDAE While these immunological data cannot give a detailed picture of relationships among these species of West Indian hylid frogs, general patterns are apparent. The Jamaican species, and two Hispaniolan species, Hyla pulchrilineata and Osteopilus dominicensis, appear to form a group. The Cuban species O. septentrionalis is slightly more divergent, while the Hispaniolan species H. vasta appears to be the basal taxon for this group of West Indian hylids. Within this West Indian group, the casqueheaded frogs, members of the genera Osteopilus and Calyptahyla, do not form a distinct group, as suggested by Trueb and Tyler (1974), but are interspersed among the West Indian species of Hyla. The ID values to Osteocephalus taurinus range from 73 into the 100s, with a mean ID of 96, which gives a divergence time of 58 mya, if this is the sister group to the West Indian hylid radiation. Hyla heilprini is clearly outside of the radiation of West Indian hylid frogs and most likely represents a second colonization of the West Indies by this family.

Relationships and Divergence Times of West Indian Amphibians and Reptiles

167

AMPHISBAENIDAE The results obtained for the West Indian amphisbaenids indicate that Amphisbaena schmidti last shared a common ancestor with other Puerto Rican species about 17 mya. It is more closely related to the Hispaniolan species, A. manni, and this suggests that there was a dispersal event approximately 10 mya. Details of the relationships among the species of West Indian amphisbaenids currently are being studied using sequence data from mitochondrial genes. Most of the West Indian taxa apparently diverged from mainland species by about 55 mya, based upon the immunological comparison to A. alba, a species from South America. These data led Hedges (1996a) to suggest that Cadea be synonomized within Amphisbaena, because it gave a lower ID value than did some congeners examined. Among the West Indian species, A. gonavensis is the most divergent. The level of divergence seen is close to that for a species that is placed in another family, the Rhineuridae. Until additional information is available, we cannot determine if A. gonavensis represents an earlier colonization event (about 86 mya), or is a recent colonist from a divergent mainland lineage. Although they gave different ID values from the A. schmidti antiserum, A. innocens and A. gonavensis are unusual among West Indian Amphisbaena in having a large number (2N = 50) of chromosomes; other West Indian species examined have 2N = 36 chromosomes (Cole and Gans, 1987; Hass and Hedges, unpublished). In this respect, they resemble some South American species with high numbers of chromosomes.

ANGUIDAE The recognition of the genus Celestus and the allocation of species of West Indian anguids to genera have been the subject of much debate (Boulenger, 1885; Burt and Burt, 1932; Underwood, 1959; Strahm and Schwartz, 1977; Savage and Lips, 1993). Although earlier workers considered the condition of the claw sheath to be an important character, Strahm and Schwartz (1977) primarily used the degree of development of canals in the osteoderms to allocate the diploglossine taxa to different genera. Wilson et al. (1986) determined that these osteoderm patterns are a reflection of ontogeny, with the radix more developed in older animals, and their work suggested that these patterns may be of limited use as phylogenetic characters. Savage and Lips (1993) resurrected the earlier classification based on presence (Diploglossus) or absence (Celestus) of a claw sheath. They considered the genera Sauresia and Wetmorena to be more closely related to Diploglossus because they have a claw sheath (Savage and Lips, 1993). The immunological data support the placement of the three large Hispaniolan species (anelpistus, carraui, warreni) in the genus Celestus (Savage and Lips, 1993). However, these data also indicate that the species currently recognized as Celestus, Sauresia, and Wetmorena comprise a closely related group, contra Savage and Lips (1993). Because only two antisera from species within this group were available, the immunological data cannot be used to determine relationships among the species. However, the data suggest that these species last shared a common ancestor relatively recently, within the last 10 million years. Therefore, the use of the condition of the claw sheath to determine relatedness seems to be inappropriate. Instead, a character deemed important by Underwood (1959), direct contact of the nasal and rostral scales, is in better agreement with the molecular results. Based upon the immunological data, we recommend that Sauresia and Wetmorena be synonymized within Celestus (following Hedges, 1996a). The Cuban and Puerto Rican species of Diploglossus examined appear to have diverged from each other about 22 mya; they are distantly related (64 mya) to the other West Indian species examined. The mainland species, Ophiodes striatus, appears to have diverged from the West Indian Celestus about 32 mya. Based upon the immunological data, it would appear that the West Indies were colonized at least twice by anguid lizards, assuming one colonization event for Diploglossus and another for Celestus. However, a clearer understanding of the historical biogeography of this group must await

168

Biogeography of the West Indies: Patterns and Perspectives

additional data from the mainland taxa. The fact that the West Indian Celestus are most closely related to the mainland Ophiodes than to Diploglossus, and the peculiar distribution of the mainland taxa (Savage and Lips, 1993), complicates biogeographical inferences.

IGUANIDAE Species in the genus Leiocephalus represent a relatively recent radiation. The immunological data indicate that the oldest divergences within this genus occurred less than 10 mya. Relationships among the species within this group have been investigated using data from sequences of mitochondrial genes. Those data (Hass et al., unpublished) are concordant with the patterns seen here. One Bahamian species, L. inaguae, is within the cluster of Hispaniolan species, while L. greenwayi, L. loxogrammus, and L. punctatus, also from the Bahamas, show ID values similar to those for Cuban taxa. The immunological data suggest that the closest mainland relatives of Leiocephalus are the crotaphytine lizards. This divergence dates to about 26 mya, and this group may have arisen from a colonization of the West Indies from North America, rather than South or Central America (Hedges, 1996b). These data are inconsistent with the placement of Leiocephalus within the Tropidurinae. Because of this inconsistency, and similar results obtained in other molecular studies (Macey et al., 1997), we do not follow the taxonomic recommendations of Frost and Etheridge (1989) for iguanian lizards.

TEIIDAE The ID values obtained indicate that divergences among some West Indian Ameiva are not recent. Ameiva chrysolaema and A. chrysolaema #3 diverged from each other very recently. However, the next most recent divergence for both A. exsul and A. chrysolaema is at a corrected ID value of 34 to 37, approximately 20 to 22 mya. These data do not provide a clear pattern of relationships among the West Indian members of this genus. The mainland species A. ameiva gives higher ID values to both antisera, indicating a divergence time of approximately 36 mya. However, these ID values are only slightly lower than those to the Cnemidophorus uniparens, another mainland species of teiid. The South American teiid, Tupinambis, is distantly related to the West Indian Ameiva.

COLUBRIDAE The low ID values seen indicate that divergences among the many species of West Indian xenodontines have occurred within the last 12 million years. The other Cuban species of Arrhyton could not be distinguished immunologically from A. landoi, despite morphological distinctions and sympatry among some species (Hedges and Garrido, 1992). One reciprocal ID value was available, measured from an antiserum against Xenodon severus to A. landoi. The difference in reciprocity (26 vs. 42) does suggest that the A. landoi may underestimate ID values. While these data cannot be used to determine phylogenetic relationships, they do suggest that Alsophis, Darlingtonia, and Uromacer are the most divergent genera of West Indian xenodontines. The immunological data also show that the degree of divergence to the mainland taxa is at the upper end of the divergences seen within the West Indies, supporting the hypothesis of a large monophyletic clade of West Indian xenodontines. Evidence from mitochondrial DNA sequences (Vidal et al., 2000) also supports this same monophyletic group. Both the ID data and the sequence data support a South American origin for the endemic West Indian xenodontines, perhaps as recently as 13 mya.

TROPIDOPHIDAE The immunological data suggest that the Jamaican and Cuban species of Tropidophis have diverged fairly recently, probably sharing a common ancestor within the last 6 million years. This is in sharp

Relationships and Divergence Times of West Indian Amphibians and Reptiles

169

contrast to the Hispaniolan Tropidophis, which diverged from other West Indian Tropidophis approximately 16 mya. The populations on Jamaica and Hispaniola clearly represent two different species. Because the type locality for T. haetianus is in Hispaniola, the Jamaican species should be recognized at T. jamaicensis, the oldest available name for the Jamaican populations (Schwartz and Henderson, 1991). The West Indian species last shared a common ancestor with mainland Tropidophis about 42 mya. All ID values to snakes from other families are at the upper limit of this technique and do not provide reliable information on the affinities of the Tropidophidae to other snake families.

TYPHLOPIDAE These data suggest that the majority of the Puerto Rican species and the species found in the Lesser Antilles diverged relatively recently, about 3 mya. The species on Cuba, Jamaica, and Hispaniola are more divergent, last sharing a common ancestor with the Puerto Rican species between 8 and 17 mya. Typhlops rostellata is the exception among the Puerto Rican species, and appears to have diverged from the others species on that island over 18 mya. This may represent an old Puerto Rican lineage, or that species may have colonized the island more recently, perhaps from Hispaniola. These data cannot distinguish between those alternatives. Typhlops syntherus is the most distant of the West Indian species from T. platycephalus, with a divergence time of about 26 mya. The degree of divergence between West Indian Typhlops and another member of this genus from the Philippines indicates that they have been separated for about 58 million years. The divergences between the family Typhlopidae and the other two families within the Scolecophidia are old, probably dating to the Cretaceous.

CONCLUSIONS The albumin immunological data for the different groups of West Indian amphibians and reptiles show very diverse patterns (Table 11). The data indicate that the fauna of these islands is composed of some recent radiations, such as the lizards of the genus Leiocephalus, which have diversified within the last 10 million years. This same pattern is seen within the xenodontine snakes, which diverged from mainland taxa only about 13 mya. There are some taxa where dichotomous patterns of relationships are seen. Within the anguid lizards, the West Indian members of the genus Celestus have speciated within the last 10 millon years. In contrast, the two species of Diploglossus are more distantly related, having diverged about 22 mya, and they last shared a common ancestor with the West Indian Celestus approximately 64 mya. These West Indian Celestus appear to be more closely related to some mainland taxa, with an estimated divergence time of 32 million years from Ophiodes. Within the West Indian snakes of the genus Tropidophis, the Bahamian, Cuban, and Jamaican species are closely related, having speciated within the last 6 million years. In contrast, the Hispaniolan members of this group diverged approximately 16 mya. Finally, there are groups where the majority of immunological values indicate that the divergences among the species are not recent. Within the amphisbaenians, divergences range from 17 to 41 million years. One species, Amphisbaena gonavensis, with a divergence of over 80 million years from A. schmidti, probably represents a separate colonization of the West Indies from the mainland. The lizards of the genus Ameiva also show a wide range of divergence times, from 0 to 38 mya. The oldest dates are approximately the same as the estimate divergence time from the mainland Ameiva. As first presented by Hedges et al. (1992), and discussed extensively by Hedges (1996a), these albumin immunological data support overwater dispersal during the Cenozoic as the mechanism for colonization of the West Indies by all of the groups discussed here. DNA sequence studies are now under way to elucidate the relationships within each of these groups to further enhance our understanding of West Indian biogeography.

170

Biogeography of the West Indies: Patterns and Perspectives

TABLE 11 Summary Table of Times of Origin for the West Indian Groups Examined in This Study, and of the Earliest Divergence within Each Group Time of (mya) Group

Origin

Earliest Divergence within the West Indies

Bufonidae Hylidae Amphisbaenidae Anguidae — Celestus Anguidae — Diploglossus Iguanidae Teiidae Colubridae (Xenodontinae) Tropidophidae Typhlopidae

31 ± 4.3 58 ± 5.8a 55 ± 5.6 32 ± 4.3 Comparison not available 26 ± 4.4 36 ± 4.4 13 ± 5.1 42 ± 4.6 58 ± 5.8

23 ± 4.5 34 ± 4.4 b 41 ± 4.6 c 10 ± 5.4 22 ± 4.5 8 ± 5.5 38 ± 4.4 12 ± 5.2 16 ± 4.9 26 ± 4.4

Note: The estimates of time of origin were based upon the ID value to the most closely related non-West Indian taxon (lowest ID value). If multiple estimates to a single taxon were available, the mean value was used (excluding any approximate estimates, as indicated by ~ in the previous tables). The earliest divergence time among the West Indian species was based on the highest ID value seen from an antiserum against a West Indian species to another West Indian species. Calibration error estimates were obtained following the methods described in Hedges et al. (1994). a

Hyla heilprini is considered to represent a separate lineage (see Discussion). The ID values to H. wilderi were significantly different from those to other taxa and therefore they were not used in this estimate (see Table 3). c Amphibaena gonavensis and A. innocens appear to represent a lineage that is distinct from the other West Indian amphisbaenians, so they were not used in this estimate (see Discussion). b

ACKNOWLEDGMENTS We would like to thank Sandra Buckner, Herndon G. Dowling, Richard Thomas, Ron Heyer, Robert Henderson, and Charles Ross for providing us with specimens. Tanya Miller and Joyce Stohler gave technical assistance in the lab. Collecting and export permits were obtained from authorities in each of the countries where specimens were collected. All experimental protocols involving animals were approved by the University of Maryland Institutional Animal Care and Use Committee, approval code R-86-039, and The Pennsylvania State University Institutional Animal Care and Use Committee, protocol number 1418. Financial support was provided by grants from the National Science Foundation.

LITERATURE CITED Boulenger, G. A. 1885. Catalogue of the Lizards in the British Museum (Natural History). 2nd ed. Taylor and Francis, London. Burt, C. E. and M. D. Burt. 1932. South American lizards in the collection of the American Museum of Natural History. Bulletin of the American Museum of Natural History 61:1–597. Cole, C. J. and C. Gans. 1987. Chromosomes of Bipes, Mesobaena, and other amphisbaenians (Reptilia) with comments on their evolution. American Museum Novitates 2867:1–9. Cronin, J. E. and V. M. Sarich. 1975. Molecular systematics of the New World monkeys. Journal of Human Evolution 4:357–375.

Relationships and Divergence Times of West Indian Amphibians and Reptiles

171

Davis, B. J. 1964. Disc electrophoresis II. Method and application to human serum proteins. The Annals of the New York Academy of Sciences 121:404–427. Frost, D. E. and R. E. Etheridge. 1989. Phylogenetic analysis and taxonomy of iguanian lizards (Reptilia: Squamata). University of Kansas Museum of Natural History Miscellaneous Publications 81. Graybeal, A. 1997. Phylogenetic relationships of bufonid frogs and tests of alternate macroevolutionary hypotheses characterizing their radiation. Zoological Journal of the Linnean Society 119:297–338. Hass, C. A. and S. B. Hedges. 1991. Albumin evolution in West Indian frogs of the genus Eleutherodactylus (Leptodactylidae): Caribbean biogeography and a calibration of the albumin immunological clock. Journal of Zoology (London) 225:413–426. Hass, C. A., S. B. Hedges, and L. R. Maxson. 1993. Molecular insights into the relationships and biogeography of West Indian anoline lizards. Biochemical Systematics and Ecology 21(1):97–114. Hedges, S. B. 1996a. The origin of West Indian Amphibians and Reptiles. Pp. 95–128 in Powell, R. and R. W. Henderson (eds.). Contributions to West Indian Herpetology: A Tribute to Albert Schwartz (Contributions to Herpetology, Vol. 12). The Society for the Study of Amphibians and Reptiles, Ithaca, New York. Hedges, S. B. 1996b. Historical biogeography of West Indian vertebrates. Annual Review of Ecology and Systematics 27:163–196. Hedges, S. B. and O. H. Garrido. 1992. Cuban snakes of the genus Arrhyton: two new species and a reconsideration of A. redimitum Cope. Herpetologica 48(2):168–177. Hedges, S. B., C. A. Hass, and L. R. Maxson. 1992. Caribbean biogeography: molecular evidence for dispersal in West Indian terrestrial vertebrates. Proceedings of the National Academy of Sciences, U.S.A. 89:1909–1913. Hedges, S. B., C. A. Hass, and L. R. Maxson. 1994. Towards a biogeography of the Caribbean. Cladistics 10:43–55. Hutchinson, M. N. and L. R. Maxson. 1986. Immunological evidence on relationships of some Australian terrestrial frogs (Anura: Hylidae: Pelodryadinae). Australian Journal of Zoology 34:575–582. Kennedy, J. P. and H. L. Brockman. 1965. Open heart surgery in Alligator mississippiensis Daudin. Herpetologica 21:6–15. Macey, J. R., A. Larson, N. B. Anajeva, and T. J. Papenfuss. 1997. Evolutionary shifts in three major structural features of the mitochondrial genome among iguanian lizards. Journal of Molecular Evolution 44:660–674. Maxson, L. R. 1984. Molecular probes of phylogeny and biogeography in toads of the widespread genus Bufo. Molecular Biology and Evolution 1:345–356. Maxson, L. R. 1992. Molecular perspectives on tempo and pattern in amphibian evolution. Pp. 41–57 in Adler, K. (ed.). Herpetology: Current Research on the Biology of Amphibians and Reptiles: Proceedings of the First World Congress of Herpetology. Society for the Study of Amphibians and Reptiles, Ithaca, New York. Maxson, L. R., R. Highton, and D. B. Wake. 1979. Albumin evolution and its phylogenetic implications in the plethodontid salamander genera Plethodon and Ensatina. Copeia 1979:502–508. Maxson, L. R. and R. D. Maxson. 1990. Immunological techniques. Pp. 127–155 in Hillis, D. M. and C. Moritz (eds.). Molecular Systematics. Sinauer Associates, Sunderland, Massachusetts. Maxson, R. D. and L. R. Maxson. 1986. Micro-complement fixation: a quantitative estimator of protein evolution. Molecular Biology and Evolution 3:375–388. Savage, J. M. and K. R. Lips. 1993. A review of the status and biogeography of the lizard genera Celestus and Diploglossus (Squamata: Anguidae), with a description of two new species from Costa Rica. Revista de Biologia Tropical 42(3):817–842. Schwartz, A. and R. W. Henderson. 1991. Amphibians and Reptiles of the West Indies: Descriptions, Distributions, and Natural History. University of Florida Press, Gainesville. Strahm, M. H. and A. Schwartz. 1977. Osteoderms in the anguid lizard subfamily Diploglossinae and their taxonomic importance. Biotropica 9:58–72. Trueb, L. and M. J. Tyler. 1974. Systematics and evolution of the Greater Antillean hylid frogs. Occasional Papers of the Museum of Natural History, The University of Kansas, Lawrence 24:1–60. Underwood, G. 1959. A new Jamaican galliwasp (Sauria, Anguidae). Museum of Comparative Zoology Breviora 102:1–13. Vidal, N., S. G. Kindl, A. Wong, and S. B. Hedges. 2000. Phylogenetic relationships of xenodontine snakes inferred from 12S and 16S ribosomal RNA sequences. Molecular Phylogenetics and Evolution 14:389–402.

172

Biogeography of the West Indies: Patterns and Perspectives

Wilson, A. C., S. S. Carlson, and T. J. White. 1977. Biochemical evolution. Annual Review of Biochemistry 46:573–639. Wilson, L. D., L. Porras, and J. R. McCranie. 1986. Distributional and taxonomic comments on some members of the Honduran herpetofauna. Milwaukee Public Museum Contributions in Biology and Geology 66:1–18.

APPENDIX: COLLECTING LOCALITIES AND VOUCHER SPECIMENS Numbers refer to tissue samples in the following collections: (LM = Linda Maxson; RH = Richard Highton; SBH = S. Blair Hedges). An asterisk (*) indicates species for which an antiserum was made. Bufonidae. Bufo: Hispaniola: *guentheri, Dominican Republic, Independencia, 12.2 km W Cabral (SBH 101227). Cuba: *peltocephalus, Guantánamo Bay U.S. Naval Station, Golf Course/ Nursery (SBH 161934); longinasus, Sancti Spiritus, north slope of Pico Potrerillo (LM 2782); taladai, Santiago de Cuba, La Esmajagua (SBH 190537). Puerto Rico: lemur (SBH 190648-50). Mainland: granulosus, Brazil (LM 329); marinus, Costa Rica (LM 206). Hylidae: Jamaica: *Calyptahyla crucialis, St. Elizabeth, Mandeville, Marshall’s Pen (LM 2570); Hyla marianae, Trelawny, Quick Step (RH 56382); Hyla wilderi, Trelawny, Quick Step (RH 56379); *Osteopilus brunneus, Trelawny, Quick Step (LM 1190); Osteopilus sp. nov., Trelawny, 5 mi WNW Quick Step (RH 60045). Hispaniola: Hyla heilprini, Dominican Republic, La Vega, 10.5 km W Hayaco (SBH 101105-06); Hyla pulchrilineata, Dominican Republic, El Seibo, 4.1 km S Sabana de le Mar (SBH 101086); *Hyla vasta, Haiti, Dept. de l’Ouest, Furcy (SBH 160414, 160417); *Osteopilus dominicensis, Dominican Republic, Barahona, 15.8 km S Cabral (SBH 101244). Cuba: *Osteopilus septentrionalis, United States: Florida (LM 1768). Mainland: *Osteocephalus taurinus, Peru, Cuzco Amazónico (LM 1866). Amphisbaenidae. Amphisbaena: Puerto Rico: bakeri, 5.8 km S Mora (SBH 172208); caeca, 6.8 km S Mamey (SBH 172233); fenestrata, USVI, St. Thomas, Dorothea Estate (SBH 161375); *schmidti, 12 km SSE Arecibo (SBH 172169, SBH 172171, SBH 172173); xera, Playa de Tamarindo (SBH 101727). Hispaniola: caudalis, Haiti, Grande’Anse, 11.8 km S Pestel (SBH 191845); gonavensis, Dominican Republic, Pedernales, Hoyo de Pelempito (SBH 192635); innocens, Haiti, Sud, 11 km N Camp Perrin (SBH 103823-24); manni, Dominican Republic, Hato Mayor, 9.5 km W Sabana de la Mar (SBH 102373). Cuba: cubana, Guantánamo Bay U.S. Naval Station, Nursery (SBH 161959); Cadea blanoides, Pinar del Rio, Viñales, Cueva de San Jose Miguel. Mainland: alba, Peru, Cuzco Amazónica (LM 1988); Rhineura floridana, Florida, Hillsborough, Plant City (SBH 172913). Teiidae. Ameiva: Hispaniola: *chrysolaema #1, Dominican Republic, Independencia, Tierra Nueva (SBH 102872, SBH 102874, SBH 102878); chrysolaema #2, Dominican Republic, Pedernales, 2 km S Oviedo (SBH 102628); chrysolaema #3, Dominican Republic; Barahona, vicinity of Barahona (SBH 101429); lineolata, Haiti, l’Artibonite, 1.1 S Colminy (SBH 191673); taeniura, Haiti, Sud’Est, 9.5 km E Jacmel (SBH 104391). Puerto Rico: *exsul, 12 km radius of Arecibo (SBH 172203-204); wetmorei, Isla Caja de Muertos (SBH 190731). Cuba: auberi, Guantánamo U.S. Naval Station, South Toro Cay (SBH 161973). Bahamas: maynardi, Great Inagua (SBH 192970). Lesser Antilles: erythrocephala, St. Kitts, Godwin Gut (SBH 172748); griswoldi, Antigua, Great Bird Island (SBH 192785); pluvianotata, Montserrat, St. Peter, Spring Ghut (SBH 192779). Mainland: ameiva, Peru, Cuzco Amazónico (LM 1993); Cnemidophorus uniparens (LM 2997); Tupinambis teguixin, Peru, Cuzco Amazónico (LM 2421). Anguidae. Hispaniola: Celestus curtissi #1, Dominican Republic, Pedernales, Juancho (SBH 102707); Celestus curtissi #2, Dominican Republic, Pedernales, 6.4 km SW and 0.7 km SE Juancho (SBH 102610); Celestus darlingtoni, Dominican Republic, La Vega, ca. 37 km SE Constanza (SBH 161687); Celestus macrotus, Haiti, Sud’Est, ca. 15 km W Gros Cheval (SBH 104405); Celestus stenurus, Dominican Republic, Independencia, 1 km E Tierra Nueva (SBH 102917); Celestus sp.

Relationships and Divergence Times of West Indian Amphibians and Reptiles

173

nov., Dominican Republic, Independencia, 23.9 km SE Puerto Escondido (SBH 192480); Diploglossus carraui, Dominican Republic (SBH 191573); *Diploglossus warreni, pet trade (SBH 172914); Sauresia agasepsoides, Dominican Republic, Barahona, 13.7 km (airline) E Canoa (SBH 160188-90); Sauresia sepsoides, Dominican Republic, Hato Mayor, 9.5 km W Sabana de la Mar (SBH 102369); *Wetmorena haetiana, Dominican Republic, Barahona, 15.3 km S and 6.7 km E of Cabral (SBH 102565-566). Jamaica: Celestus barbouri, Trelawny, vicinity of Quick Step (SBH 161120); Celestus crusculus crusculus, Hanover, 3.2 km SE Content (SBH 101572); Celestus crusculus cundalli, Portland, 1.3 km WSW Section (SBH 172465). Cuba: *Diploglossus delasagra, Guantánamo, 1 km SW San Luis de Potosí (SBH 191015). Puerto Rico: *Diploglossus pleii, 5.5 km N Rio (SBH 161370), about 5 km SE Maricao and 6 km NW Sabana Grande (SBH 172199-200). Mainland: Ophiodes striatus (MVZ 191047) Iguanidae. Leiocephalus: Hispaniola: *schreibersi, Dominican Republic, Independencia, Tierra Nueva (SBH 102721, SBH 102879-880, SBH 102889); barahonensis #1, Dominican Republic, Pedernales, about 2 km S Oviedö (SBH 102645); barahonensis #2, Dominican Republic, Independencia, 5.1 km SE Puerto Escondido (SBH 192536); lunatus, Dominican Republic, Altagracia, 1 km W Boca de Yuma (SBH 160123); melanochlorus, Haiti, Grande’Anse, about 3 km N Bois Sec (SBH 103721); personatus, Dominican Republic, Maria Trinidad Sanchez, 4 km SE Nagua (SBH 103024); semilineatus, Haiti, l’Ouest, 11.7 km E Thomazeau (SBH 191661-63). Cuba: carinatus, Guantánamo, Guantánamo Bay U.S. Naval Station, pump station and water tower on leeward side of bay (SBH 161965); cubensis, Matanzas, Soplillar (SBH 172490); macropus, Guantánamo, Guantánamo Bay U.S. Naval Station, pistol range on leeward side of bay (SBH 161984); raviceps, Guantánamo, Guantánamo Bay U.S. Naval Station, pistol range on leeward side of bay (SBH 161980); stictigaster, Guantánamo, Tortuguilla (SBH 190161). Bahamas: greenwayi, East Plana Cay (SBH 192972); inaguae, Great Inagua (SBH 192973); loxogrammus, San Salvador (SBH 192971), punctatus, Acklin’s Island (SBH 192975). Crotaphytus collaris (LM 2534). Sceloporus spinosus (LM 2264). Tropidurus peruvanius (LM 1556B). Tropidurus hispidus (LM 2795). Trophidophidae. Tropidophis: Jamaica: *haetianus stejnegeri #1, Trelawny, vicinity of Quick Step (SBH 103592); haetianus stejnegeri #2, Trelawny, 0.3 km W Duncans (SBH 101580); haetianus stullae, Clarendon, Portland Ridge (SBH 103593). Cuba: feicki, Pinar del Río, Soroa (SBH 172745); fuscus, Guantánamo, Minas Amores (SBH 190300); maculatus, Pinar del Río, Soroa (SBH 191543); melanurus, Pinar del Río, Soroa (SBH 172610); pardalis, La Habana; Narigon (SBH 191545); pilsbryi, Santiago de Cuba, Simpatía (SBH 191368); wrighti, Guantánamo, 2 km N La Munición (SBH 191066). Bahamas: canus, Andros (RH 54403). Hispaniola: haetianus haetianus #1, Haiti: Sud (RH 54404); haetianus haetianus #2, Domican Republic, El Seibo, 4.1 km S Sabana de la Mar (SBH 101398); haetianus haetianus #3, Dominican Republic, Samana, 6 km SSW Las Galeras (SBH 103121); haetianus haetianus #4, Haiti, Sud, 8.6 km SW Carrefour Joute on the Prequille de Port Salut (SBH 192361); haetianus haetianus #5, Dominican Republic, El Seibo, Nisibon (SBH 192455), haetianus hemurus, Dominican Republic, La Altagracia; 28 km NW Higuey (SBH 192454). Mainland: paucisquamis, Brasil, São Paulo, Boraceía (LM 908). Other taxa: Boa constrictor, pet trade (RH 54430). Naja naja, pet trade (RH 58101). Python molarus, pet trade (RH 56048). Coluber constrictor (#4). Arrhyton landoi, Guantánamo, Guantánamo Bay U.S. Naval Station, vicinity of John Paul Jones Hill (SBH 161893-895). Typhlopidae. Puerto Rico: Typhlops granti, Bosque Estatel de Guanica (SBH 172210); Typhlops hypomethes, University of Puerto Rico campus at Rio Piedras (SBH 161807); Typhlops hypomethes, University of Puerto Rico campus at Rio Piedras (SBH 172150); *Typhlops platycephalus, 12.3 km SSE Arecibo (SBH 172180); Typhlops richardi, British Virgin Islands; Guana Island (SBH 172759); Typhlops rostellatus, 12.3 km SSE Arecibo (SBH 172174). Lesser Antilles: Typhlops guadeloupensis, Guadeloupe, Pointe de la Grande’Anse (SBH 102276), Typhlops monastus, Nevis, 0.3 km N Cotton Ground (SBH 172760). Cuba: Typhlops biminensis, Guantánamo, Playites de Cajobabo (SBH 190234); Typhlops lumbricalis, Havana, National Botanical Garden (SBH 172600). Hispaniola: Typhlops capitulatus, Haiti, l’Ouest, Soliette (SBH 103826); Typhlops hectus, Dominican Republic,

174

Biogeography of the West Indies: Patterns and Perspectives

Barahona, 13.5 km SW Barahona (SBH 102665); Typhlops pusillus, Dominican Republic, Azua, 18 km NNW Azua (SBH 160284); Typhlops schwartzi, Dominican Republic, El Seibo, Nisibón (SBH 192458); Typhlops sulcatus, Dominican Republic, Pedernales, SW of Enriquillo (SBH 102438); Typhlops syntherus, Dominican Republic, Pedernales, SW of Enriquillo (SBH 102437); Typhlops titanops, Dominican Republic, Pedernales, 20 km N Pedernales (SBH 160293). Jamaica: Typhlops jamaicensis, St. Mary, 6.2 km W Oracabessa (SBH 172445). Philippines: Typhlops luzonensis, Negros Island, Negros Oriental Province, Valenica Municipality, Bong Bong Bario, Camp Lookout (SBH 194117). Mainland: Leptotyphlops, Trinidad (SBH 175446); Liotyphlops albirostris, Venezuela, Caracas (SBH 172151). Colubridae (Xenodontinae). Bahamas: Alsophis vudii, New Providence, Sandy Port (SBH 192985). Cuba: Arrhyton dolichura, Havana, National Botanical Garden (SBH 172601); *Arrhyton landoi #1, Guantánamo Bay U.S. Naval Station (SBH 161893-95, SBH 161985); Arrhyton landoi #2, Guantánamo, 3.5 km E Tortuguilla (SBH 191258); Arrhyton procerum, Matanzas, Playa Giron (SBH 191526); Arrhyton supernum #1, Guantánamo, SW slope of El Yunque de Baracoa (190230); Arrhyton supernum #2, Guantánamo, Monte Libano, ca. 20 km SSE La Tagua (SBH 191153); Arrhyton taeniatum, Guantánamo Bay U.S. Naval Station, ca. 0.2 km E Windmill Beach (SBH 171002); Arrhyton tanyplectum, Pinar del Río, 4 km NW San Vincente (SBH 191492); Arrhyton vittatum, Pinar del Río, Cueva de San Miguel (SBH 191491), Antillophis andreai, Pinar del Río, Soroa (SBH 172603); Alsophis cantherigerus, Pinar del Río, 2.0 km W Vinales (SBH 172602). Hispaniola: Antillophis parvifrons protenus, Dominican Republic, Barahona, 19.5 km SW Barahona (SBH 103086); Darlingtonia haetiana, Haiti, Grande’Anse, ca. 2–3 km S Castillon (SBH 103806-10); Hypsirhynchus ferox, Dominican Republic, Barahona, vicinity of Barahona (SBH 101393); Hypsirhynchus scalaris, Haiti, Grande’Anse, 7.2 km S Roseaux (SBH 191992); Ialtris dorsalis #1, Haiti, Grande’Anse, ca. 3 km N Bois Sec (SBH 103702); Ialtris dorsalis #2, Haiti, Grande’Anse, 7.5 km N Beaumont (SBH 192360); Uromacer catesbyi #1, Dominican Republic, Monte Plata, 2.8 km N Yamasa (SBH 101397); Uromacer catesbyi #2, Dominican Republic, La Altagracia, 4.4 km W Canada Honda (SBH 192456); Uromacer frenatus frenatus, Haiti, Grande’Anse, ca. 6 km E Jeremie (SBH 104668); Uromacer oxyrhynchus, Dominican Republic, La Altagracia, 4.4 km W Canada Honda (SBH 192457). Jamaica: Arrhyton callilaemum, St. Mary, 2.9 km N Port Maria (SBH 172463); Arrhyton funereum, St. Mary, Port Maria, 2.9 km N Port Maria (SBH 172462); Arrhyton polylepis, Portland, 0.3 km S Alligator Church (SBH 101581). Lesser Antilles: Alsophis antiguae, Antigua, Great Bird Island (SBH 192790); Alsophis antillensis, Montserrat (SBH 192791). Puerto Rico: Arrhyton exiguum, 1.9 km NE Vista Alegre (SBH 160050); Alsophis portoricensis, 1.5 km W Playa de Tamarindo (SBH 160062). Mainland: Dipsas catesbyi, Peru, Pasco, 1.5 km NW Cacazu (SBH 171139); Leptodeira sp., Panama (LM1145); Liophis cabella, Peru, Pasco, Oxapampa (SBH 171143); Liophis melanostigma, Brazil, São Paulo, Boraceia (LM 904); Oxyrhopus leucomeles, Peru, Pasco, Oxapampa (SBH 171142); Thamnodynastes sp., Peru, Madre de Dios, Tambopata Reserve (LM 1104); Xenodon severus (RH 68185).

Historic and Prehistoric 12 The Distribution of Parrots (Psittacidae) in the West Indies Matthew I. Williams and David W. Steadman Abstract — If not for human impact, three genera of psittacids (Ara, Aratinga, and Amazona) would be represented today throughout the Greater and Lesser Antilles. The Cayman Islands and Bahamas are the only regions lacking evidence of Ara and Aratinga. Guadeloupe is the only island with possible evidence for the occurrence of a fourth genus, Anodorhynchus. The growing body of information from paleontology, zooarchaeology, and post-Columbian history further suggests that multiple sympatric species of Amazona were widespread. In the other two genera, a single species was typically confined to one major island or a cluster of nearby islands. We suggest that as many as 50 to 60 endemic species of psittacids would occupy the West Indies in the absence of human influence, as compared to the 12 species (3 of Aratinga, 9 of Amazona) that survive today.

INTRODUCTION Parrots (Order Psittaciformes, Family Psittacidae) are one of the most successful groups of land birds on tropical islands. In the West Indies, indigenous species of parrots (usually endemic to a single island or island cluster; see Snyder et al., 1987) are or were represented by three genera: Ara (macaws), Aratinga (parakeets), and Amazona (amazons or simply “parrots”). A fourth genus, Anodorhynchus, also may have occurred, but the evidence currently available is inadequate to support that hypothesis. The goal of this chapter is to review briefly the past distribution of psittacids in the West Indies during the historic era (the past 500 years) and especially prehistoric times (i.e., pre-Columbian). Our geographical coverage includes the entire West Indian faunal region (see map in Raffaele et al., 1998:12). Excluded, therefore, are Trinidad, Tobago, Margarita, Aruba, Bonaire, Curaçao, and other smaller islands off the northern coast of South America. The avifaunas of these islands have only a minor West Indian influence. We do not include introduced populations, whether of species that are indigenous on other West Indian islands (such as the populations of Aratinga chloroptera introduced to Puerto Rico and Guadeloupe; see Raffaele et al., 1998:308) or of various non-West Indian psittacids that have been released over the past century in the Bahamas, Cayman Islands, Jamaica, Hispaniola, Puerto Rico, Virgin Islands, Guadeloupe, Dominica, Martinique, Barbados, and perhaps elsewhere. We believe that all of the certain or possible extinctions of indigenous West Indian psittacids are anthropogenic. Some of these losses clearly occurred in historic times, such as that of Ara tricolor in Cuba and Isla de Pinos (Isla de la Juventud). Other losses are more likely to have occurred during prehistoric human occupation of the islands (see Keegan, 1994, 2000, for a review of West Indian prehistory). In some cases we cannot be certain if the population in question was indigenous or had been transported by prehistoric peoples to the island. Such uncertainty is removed when fossils document the presence of a species before the arrival of people, as on Barbuda (see below). While acknowledging the inter-island exchange of psittacids by Amerindians at European contact (see Oviedo, 1959; Wilson, 1990), we believe it likely that most or all West Indian islands did sustain their own sets of indigenous, if not endemic, species of macaws, parakeets, and parrots.

0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

175

176

Biogeography of the West Indies: Patterns and Perspectives

We hope that this chapter will help keep alive the memory of extirpated populations or species of West Indian psittacids among those who study extant Antillean birds. With the blending of ornithology and birdwatching over the last decade or two, interest has waned in species that no longer can be seen with binoculars. Most of the species listed in Tables 1 to 4 are seldom mentioned in the ornithological literature, with Snyder et al. (1987) and Butler (1992) being conspicuous and important exceptions. The validity of some of the names or records is difficult to evaluate based on evidence in hand; we cannot be certain that every one of these species existed, although we are confident that most of them did. Our compilation of distributional data on parrots is in some measure a response to the close of the Bondian era in West Indian ornithology. Bond (1971, and various “checklists” such as those in 1950 and 1956) often ignored prehistoric and early historic records in compilations of the West Indian avifauna. Many living species of Antillean birds were also lumped with little justification by Bond. Raffaele et al. (1998) has recognized once again many of these endemic species, although we would like to point out that even this excellent new standard reference for the modern distribution of West Indian birds gives very little indication of the extent to which species of birds, parrots and otherwise, have been lost since human arrival in the Antilles. Some appreciation for these losses can be gleaned from Olson (1978, 1982), Steadman et al. (1984a), and Pregill et al. (1994).

BRIEF SPECIES ACCOUNTS All West Indian species of psittacids are listed in Tables 1 to 4, proceeding west to east in the Greater Antilles and north to south in the Lesser Antilles (Figure 1). Our brief species accounts will cover only extinct species, extirpated populations, or prehistoric records of extant populations. The current distribution and status of extant forms are summarized in Snyder et al. (1987), Butler (1992), and Raffaele et al. (1998); the former two papers also review historical distributions. Our primary contribution here is to summarize the prehistoric records, several of which have not been published before. A dagger (†) indicates an extinct taxon. We will name the extinct, undescribed species from archaeological and paleontological sites in a separate publication. Synonyms for genera are from Ridgway (1916).

MACAWS (ARA) West Indian synonyms Anodorhynchus (in part), Arara, Macrocercus, Psittacus, Sittace No species of macaws (Ara spp.) still exist anywhere in the West Indies (Table 1). Only one species of Ara, the Cuban A. tricolor, is known from whole specimens (Walters, 1995). Evidence for the others comes from prehistoric bones or from written accounts of the 17th through 19th centuries. We believe that each Greater Antillean and Lesser Antillean island once sustained one or two indigenous if not endemic species of Ara. Macaws survived into historic times on at least Cuba, Isla de Pinos, Jamaica, Hispaniola, Guadeloupe, Dominica, and Martinique, although the species-level systematics of these macaws often is poorly resolved, as detailed below. †Ara tricolor (Bechstein, 1811) — Cuban Macaw According to Bangs and Zappey (1905), the last known pair of Cuban macaws was shot in 1864 at La Vega on the Zapata Peninsula. Gundlach (in Cory, 1886; Greenway, 1958) believed that A. tricolor persisted in the swamps of southern Cuba in 1876. Wetmore (1928) identified a carpometacarpus from an undated cave deposit in Cuba as A. tricolor. There are skins of A. tricolor in the British Museum and Liverpool Museum (Salvadori, 1891, 1906b; Walters, 1995), but no modern skeletal specimens of A. tricolor exist. Wetmore’s identification was based on extrapolation from skins and the relative size of the carpometacarpus in living species of macaws; the fossils were larger than those in A. severa, a relatively small species of Ara.

FIGURE 1 The past and present distribution of macaws (Ara), parakeets (Aratinga), and parrots (Amazona) in the West Indies. † = extinct species, subspecies, or population. Based on data in Table 4.

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies 177

178

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 Prehistoric, Historic, and Modern Records of Ara in the West Indies Species

Island

Present Status

Prehistoric

Historic

Carpometacarpus from Ciego Montero cave — Wetmore, 1928

Last recorded in 1864 — Salvadori, 1906a; Ridgway, 1916:136–137; Snyder et al., 1987 Last recorded in 1860s — Bangs and Zappey, 1905 Clark, 1905d Salvadori, 1906a; Rothschild, 1905, 1907; Ridgway, 1916:137–138; Wetmore, 1937; Snyder et al., 1987. Rothschild, 1905, 1907; Ridgway, 1916:140; Wetmore, 1937; Snyder et al., 1987 Salvadori, 1906a; Rothschild, 1907; Ridgway, 1916:125; Wetmore, 1937 Clark, 1905d; Wetmore and Swales, 1931; Snyder et al., 1987; Raffaele et al., 1998 —

Extinct

—

Extinct

Clark, 1905a; Salvadori, 1906a; Ridgway, 1916:131–132; Wetmore, 1937; Snyder et al., 1987; Butler, 1992

Extinct

†Ara tricolor

Cuba

†A. tricolor

Isla de Pinos

—

†A. tricolor* †A. gossei

Jamaica Jamaica

— —

†A. erythrocephala

Jamaica

—

†A. erythrura*

Jamaica?

—

A. tricolor? and/or †Ara unknown sp.

Hispaniola

—

†A. autochthones

St. Croix

†Ara undescr. sp.

Montserrat

†A. guadeloupensis

Guadeloupe

†Ara cf. guadeloupensis

Marie Galante

†A. atwoodi

Dominica

†A. martinica * and/or †Ara undescr. sp.

Martinique

—

†Anodorhynchus purpurascens*

Guadeloupe

—

†A. martinicus*

Martinique

—

Tibiotarsus from Amerindian midden — Wetmore, 1937; Olson, 1978; Wing, 1989 Coracoid from Trants site — this chapter —

Ulna from Folle Anse site — this chapter —

Extinct

Extinct Extinct Extinct

Extinct

Extinct

Extinct

Extinct Clark, 1905a; Wetmore, 1937; Snyder et al., 1987; Butler, 1992 Rothschild, 1907; Ridgway, 1916:125; Wetmore, 1937; Snyder et al., 1987; Butler, 1992 Salvadori, 1906a; Rothschild, 1907; Ridgway, 1916:119; Wetmore, 1937 Rothschild, 1905; Salvadori, 1906a

Extinct Extinct

Extinct

Extinct

Note: † = Extinct species; * = validity of species, or of the species on this particular island, needs to be corroborated and may be doubtful; these records, therefore, are not included in Table 4.

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies

179

The juvenile plumage of Ara tricolor may have been predominantly green, which might account for several early reports of A. militaris on Cuba and Jamaica (Clark, 1905d). The overall plumage pattern of A. tricolor suggests that its nearest mainland relative may be A. macao, the Scarlet Macaw. The distribution of red and blue in the plumage is similar, as is the presence of a white facial patch that is featherless except for small crescentic lines of tiny red feathers. Ara tricolor differs from A. macao in that it lacks the yellow shoulder patch, has an all-black bill, and is much smaller. The modern range of A. macao is in lowland forest from southern Mexico through Central America to much of tropical South America. Among Caribbean islands it has been recorded, but is not currently resident, on Trinidad, a continental rather than oceanic island (ffrench, 1991:183). Thus the modern range of A. macao encompasses much of the western and southern margins of the Caribbean Sea. †Ara gossei (Rothschild, 1905) — Gosse’s Macaw Found in “the mountains” of Jamaica, one specimen was shot about 1765 near the Montego Bay area (Gosse, 1847). The fate of this specimen is unknown, but Greenway (1958:318) reported that “Such a bird was described by a Dr. Robinson, who saw a stuffed specimen.” No exact date was given for Dr. Robinson’s report. Ara gossei probably looked very similar to A. tricolor. The major difference was in the forehead, described as yellow in A. gossei and red in A. tricolor. Robinson described the preserved specimen as: “forehead, crown, and back of neck bright yellow; sides of face around eyes, anterior and lateral part of the neck, and back a fine scarlet; wing coverts and breast deep sanguine red; winglet [sic] and primaries an elegant light blue; basal half of the upper mandible black, apical half ash colored; lower mandible black; tail and feet were missing” (Greenway, 1958:318). †Ara erythrocephala (Rothschild, 1905) — Red-headed Green Macaw Greenway (1958:320) called this an “almost mythical bird” given the circumstances surrounding its description. Ara erythrocephala was said to have been found in the mountains of Trelawney and St. Anne’s parishes, Jamaica (Rothschild, 1905). The head was red, the body bright green, and the wings and greater coverts blue. The tail was scarlet and blue on top, whereas the tail and wings were intense orange-yellow underneath (Rothschild, 1905; Salvadori, 1906a; Greenway, 1958). Snyder et al. (1987) suggested that A. erythrocephala may represent A. militaris or A. ambigua, both Central American species. Especially given that two endemic species of Amazona occur in Jamaica, we see no reason why multiple species of Ara could not also have inhabited this large island of diverse habitats. †Ara erythrura (Rothschild, 1907) — Red-tailed Blue-and-Yellow Macaw Rothschild (1907:53) named Ara erythrura from the report of two large, blue and yellow parrots observed by a Reverend Comard of Jamaica in the early 1800s. Greenway (1958:319) regarded Rothschild’s description of A. erythrura as not credible because it was based on de Rochefort (1658), who had not visited Jamaica but “seems to have taken his account from du Tertre.” Greenway (1958:319) suggested that, if anything, A. erythrura is a synonym of A. martinica, a poorly documented form supposedly from Martinique (see below). †Ara tricolor? or †Ara unknown sp. — Hispaniolan Macaw Among the three species of psittacids noted by Casas (1876) on Hispaniola at the end of the 1400s was a macaw that differed from those on other islands in that it had a white forehead, not red as is seen in Ara tricolor. Macaws were said to have been common formerly in Hispaniola but rare by 1760 (Clark, 1905d). Buffon (1779 [not seen by us; in Greenway, 1958]) reported a macaw on the south coast of Hispaniola. Gosse (1847) mentioned that a small macaw reported to be A. tricolor

180

Biogeography of the West Indies: Patterns and Perspectives

was found on Haiti, although he himself had not seen it. He suggested that it represented a species of Ara other than those already known from Cuba and Jamaica (from Rothschild, 1905). †Ara autochthones (Wetmore, 1937) — St. Croix Macaw A tibiotarsus excavated from a prehistoric archaeological site on St. Croix is the basis for Ara autochthones (Wetmore, 1937). This bone was from an adult-sized, immature individual intermediate in size between A. macao and A. severa and slightly larger than in A. tricolor. Olson (1978) confirmed that the tibiotarsus is of an immature macaw and is not referable to any living species, noting further that A. autochthones was not necessarily indigenous to St. Croix because prehistoric West Indian peoples were known to keep and trade live psittacids, a particular concern with material excavated from a cultural site. Wing (1989) also suggested that A. autochthones may have been traded to St. Croix. While this is possible, there also is no reason why St. Croix could not have sustained an indigenous species of Ara, especially given the substantial evidence of indigenous macaws in both the Greater and Lesser Antilles. †Ara undescribed sp. — Montserrat Macaw A nearly complete coracoid from the Trants archaeological site on Montserrat represents a small, presumably undescribed species of Ara. The specimen is smaller than in A. ararauna and larger than in A. severa or A. manilata, although closer in size to the last two. This specimen was recovered from excavations at Trants by D. R. Watters subsequent to the recovery of numerous bird bones from this rich site reported by Steadman et al. (1984b) and Reis and Steadman (1999). †Ara guadeloupensis (Clark, 1905a) — Guadeloupe Macaw This species was superficially similar to A. macao, but smaller and with the tail entirely red (Salvadori, 1906a). du Tertre (1654; in Clark, 1905a) gave the following description: “the head, neck, underparts, and back are flame color. The wings are a mixture of yellow, azure, and scarlet. The tail is wholly red, and a foot and a half long.” The tail in A. guadeloupensis was much larger than in A. tricolor, a relatively small macaw with a tail length of ~12 in. (290 to 305 mm; Ridgway, 1916:136), although Greenway (1958:318) incorrectly claimed that A. guadeloupensis had a shorter tail than A. tricolor. Labat (1742:II:211) observed a macaw on Guadeloupe with similar plumage, stating further that the macaws and parrots of Guadeloupe were generally larger than those from other islands, although the parakeets were smaller. du Tertre (1654:294) mentioned that this species was long-lived (“live longer than a man”) but that they were “almost all subject to a falling sickness.” Thus perhaps a disease outbreak, combined with hunting pressure, could account for the extinction of A. guadeloupensis. Macaws were becoming rare in the Lesser Antilles (and presumably throughout the West Indies) even in the 1700s (Clark, 1905a). We find no evidence for the suggestion by Clark (1905a) that A. guadeloupensis also occurred on Dominica and Martinique. Based on what is attributed to Labat (Clark, 1905a:269), it would seem more likely that the Lesser Antillean macaws were endemic to each island or set of nearby islands. Christopher Columbus reported red parrots that were called “Guacamayos” by the Caribs on Guadeloupe (Clark, 1905a). Because these Caribs were able to tell Columbus the direction of the mainland, Greenway (1958:319) suggested that the parrots could have been imported to Guadeloupe. We admit this possibility, but see it as no more likely than that they were indigenous. de Rochefort (1658) mentioned three plumage patterns for macaws of the Lesser Antilles but, being pre-Linnean, did not refer to any binomial or other diagnostic names. The first had a pale yellow head, back, and wings, with the tail entirely red [Ara?]. In the second the whole body was flame and the wings were yellow, blue, and red [A. guadeloupensis]. The third pattern was “a mixture of red, white, blue, green, and black with a body size similar to a pheasant (Phasianus colchicus) [similar to A. macao?].”

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies

181

†Ara cf. guadeloupensis — Marie Galante (Guadeloupe?) Macaw An ulna from the Folle Anse archaeological site on Marie Galante represents a species of Ara that most likely, although far from certainly, is from A. guadeloupensis, an extinct species for which no skeletal (or skin) specimen exist. This ulna is slightly smaller than that in A. macao, and substantially smaller than that in A. ararauna. †Ara atwoodi (Clark, 1908) — Dominica Macaw Thomas Atwood (1791) noted a macaw from Dominica that was larger than the two local species of parrots (Amazona arausiaca, A. imperialis) and in great abundance. This species was said to have green and yellow plumage “with a scarlet coloured fleshy substance from the ears to the root of the bill.” The “chief feathers” of the wings and tail were scarlet as well. While macaws are characterized by their patch of bare skin on the face, no extant macaw (or other described extinct macaw) has a red facial patch (Clark, 1908). †Ara martinica (Rothschild, 1905) — Martinique Macaw Greenway (1958:319) and Snyder et al. (1987) both suggested that this putative species likely pertained to A. ararauna, a mainland species that could have been traded to Martinique. Snyder et al. (1987) noted, however, that an unnamed and poorly known but distinctive macaw once lived on Martinique.

MACAWS (ANODORHYNCHUS) †Anodorhynchus purpurascens (Rothschild, 1905) — Guadeloupe Violet Macaw Rothschild (1905) based his description on a paper by de Navaret (1838), which neither Greenway nor we were able to locate. Greenway (1958:320) and Snyder et al. (1987) suggested that the species was based on either a poor description of Amazona violacea (now extinct but formerly found on Guadeloupe) or of the Brazilian Anodorhynchus leari, which must have been imported to Guadeloupe. The plumage was described as entirely violet, which suggests a species of Anodorhynchus. †Anodorhynchus martinicus (Rothschild, 1905) — Martinique Macaw Rothschild (1905) described this species from an account by Bouton (1635, which we have not seen) of a macaw on Martinique that was blue above with orange underparts. Salvadori (1906a) regarded Anodorhynchus martinicus to be based on Ara ararauna. We regard both supposed species of Anodorhynchus in the Lesser Antilles as requiring further corroboration.

PARAKEETS (ARATINGA) West Indian synonyms Conurus, Euopsitta, Psittacara, Psittacus Parakeets (Aratinga) are long-tailed, often rather small psittacids that are proportionally similar to macaws of the genus Ara. Modern records of West Indian forms of Aratinga are confined to the Greater Antilles (Table 2). According to Clark (1905b) the West Indian parakeets were too small to attract much attention from early writers and, as a result, the accounts of Aratinga from the 17th and 18th centuries are brief and often lack in diagnostic information. Aratinga euops (Wagler, 1832) — Cuban Parakeet The Cuban parakeet, which still exists on Cuba itself, was once abundant on Isla de Pinos, where it was bordering on extirpation a century ago (Bangs and Zappey, 1905) and was lost shortly thereafter.

182

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 Prehistoric, Historic, and Modern Records of Aratinga in the West Indies Species

Island

Prehistoric

Historic Ridgway, 1916:160–161 Extirpated ca. 1900 — Bangs and Zappey, 1905; Marien and Koopman, 1955; Snyder et al., 1987 Ridgway, 1916:174–175 Ridgway, 1916:153–154

Aratinga euops †A. euops

Cuba Isla de Pinos

— —

A. nana A. chloroptera chloroptera †A. chloroptera maugei

Jamaica Hispaniola and offshore islands Puerto Rico and Mona

— —

†Aratinga undescr. sp.

Barbuda

†A. labati

Guadeloupe

†Aratinga undescr. sp.

Dominica

—

†Aratinga undescr. sp. †Aratinga undescr. sp.

Martinique

—

Barbados

—

—

Palatine — Pregill et al., 1994; sternum — this chapter —

Last specimen 1892, believed to be extinct shortly thereafter; perhaps extant on Mona in 1905 — Clark, 1905c; Salvadori, 1906a; Ridgway, 1916:155, Rothschild, 1905; Marien and Koopman, 1955; Snyder et al., 1987 —

Already rare before 1760 — Rothschild, 1905; Salvadori, 1906; Ridgway, 1916:175; Snyder et al., 1987 No description, exterminated before 1878 — Clark, 1905b, 1911; Snyder et al., 1987; Butler, 1992 Clark, 1905b, 1911; Snyder et al., 1987; Butler, 1992 Clark, 1905b, 1911; Snyder et al., 1987; Butler, 1992

Present Status Threatened Extinct

Common Locally common but declining Extinct

Extinct

Extinct

Extinct

Extinct Extinct

Note: Subspecies follow Ridgway (1916); † = extinct species, subspecies, or population.

†Aratinga chloroptera maugei (Souancé, 1856) — Puerto Rican/Mona Parakeet Compared to Aratinga c. chloroptera, this extinct form (recognized as a full species in Raffaele et al., 1998) was smaller, with a darker-colored bill, lighter red under primary coverts, and completely green lesser primary coverts (Ridgway, 1916:155). Only three specimens of A. c. maugei exist, all from Mona, the last collected in 1892 (Greenway, 1958:321). It was lost from Puerto Rico in the late 1800s. Rothschild (1905) regarded A. c. maugei as still living on Mona, whereas Bond (1950) reported that A. c. maugei probably was gone from Mona. †Aratinga undescribed sp. — Barbudan Parakeet A quadrate of a very large, undescribed species of Aratinga was reported from Barbuda II, a paleontological cave locality on the east coast of Barbuda, by Pregill et al. (1994). Larger than in any extant species of Aratinga, this quadrate also represented the first specimen of Aratinga from

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies

183

anywhere in the Lesser Antilles. A sternum of Aratinga recently discovered in the Vertebrate Paleontology collections at the Florida Museum of Natural History, from the nearby site of Barbuda I, is also very large and must be from the same extinct species. †Aratinga labati (Rothschild, 1905) — Guadeloupe Parakeet Thought to be endemic to Guadeloupe, Aratinga labati was named by Rothschild (1905) based on a description by Labat (1742:II:211). Aratinga labati, for which no specimen exists, was small and green overall, with a small patch of red on the crown and a pale bill (Clark, 1905b). Greenway (1958:322) believed that the species probably existed because du Tertre (1654:299, 1667:251) mentioned a third species of parrot on Guadeloupe that was “all green and big as magpies.” Hughes (1750) also noted a small, green “Parakite” [sic] on Guadeloupe. †Aratinga undescribed spp. — Dominica, Martinique, and Barbados Parakeets Known only from early travelers’ accounts (summarized in Clark, 1905b; Snyder et al., 1987), distinctive forms of parakeets once occurred on these three islands and undoubtedly all other islands in the Lesser Antilles.

PARROTS

OR

AMAZONS (AMAZONA)

West Indian synonyms Androglossa, Chrysotis, Oenochrus, Onochrus, Psittacus. Amazona leucocephala hesterna (Cory, 1886) — Cayman Parrot This subspecies has been extirpated in historic times on Little Cayman Island (Bradley 1995). It survives on Cayman Brac. Amazona leucocephala bahamensis (Bryant, 1867) — Rose-throated (Bahamas) Parrot Fossils and historic records indicate that this species, now restricted in the Bahamas to Abaco and Great Inagua, was once widespread in the island group, including the Turks and Caicos Islands (Brodkorb, 1959; Olson and Hilgartner, 1982; Carlson, 1999; Table 3, this chapter). Columbus noted flocks of parrots that would “obscure the sun” in the Bahamas (Dunn and Kelley, 1989:105). †Amazona undescribed sp. — Turks and Caicos Parrot This extinct parrot is known only from a palatine and scapula from the Coralie archaeological site on Grand Turk, where it was sympatric with the smaller A. leucocephala (Carlson, 1999). †Amazona vittata gracilipes (Ridgway, 1915) — Culebra Parrot This endemic subspecies perished sometime early this century, but we have been unable to find any details. †Amazona vittata — Barbuda (Puerto Rican) Parrot A nearly complete rostrum is from Barbuda I, an undated (but almost certainly precultural) paleontological cave locality on Barbuda. We recently discovered this specimen, collected in 1962, in the Vertebrate Paleontology collections at the Florida Museum of Natural History. It agrees with the rostrum of modern Amazona vittata from Puerto Rico.

184

Biogeography of the West Indies: Patterns and Perspectives

TABLE 3 Prehistoric, Historic, and Modern Records of Amazona in the West Indies Species Amazona leucocephala bahamensis

†A. leucocephala bahamensis

†Amazona undescr. sp.

Island

Present Status

Prehistoric

Historic

Abaco, Bahamas

—

—

Common

Great Inagua, Bahamas Crooked Island, Bahamas New Providence, Bahamas

—

—

Common

Premaxilla — Wetmore, 1938; Olson and Hilgartner, 1982 Tarsometatarsus, radius — Brodkorb, 1959; ulnae, radius, carpometacarpi, femur, tarsometatarsi — Olson and Hilgartner, 1982 Todd and Worthington, 1911 — — Six bones — Carlson, 1999

—

Extinct

—

Extinct

— Bond, 1956 Bond, 1956 —

Extinct Extinct Extinct Extinct

Acklins, Bahamas Long, Bahamas Fortune, Bahamas Grand Turk, Bahamas Grand Turk, Bahamas Cuba

Palatine, scapula — Carlson, 1999 —

—

Extinct

—

Isla de Pinos

—

—

Grand Cayman

—

—

Locally common Low but recovering Common

Cayman Brac

—

—

Common

Little Cayman

—

Bradley, 1995

Extinct

Jamaica

—

Ridgway, 1916:267–269; Pregill et al., 1991

A. agilis

Jamaica

—

Ridgway, 1916:262–263; Pregill et al., 1991

A. ventralis

Hispaniola, Grande Cayemite, Gonâve, Saona, Beata Puerto Rico Culebra

—

Ridgway, 1916:265–267

Locally common and widespread Threatened but locally common Uncommon, local

— —

Ridgway, 1916:263–265 Ridgway, 1916:265; Snyder et al., 1987 — —

A. leucocephala leucocephala A. leucocephala palmarum A. leucocephala caymanensis A. leucocephala hesterna †A. leucocephala hesterna A. collaria

A. vittata vittata †A. vittata gracilipes †A. vittata

Barbuda Antigua

†Amazona undescr. sp.

Montserrat

Rostrum — this chapter Two bones (as Amazona sp.) — Steadman et al., 1984a, Pregill et al., 1988; as A. vittata — Pregill et al., 1994 Humerus — Reis and Steadman, 1999; coracoid, humerus, ulna, femur — this chapter

—

Endangered Extinct Extinct Extinct

Extinct

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies

185

TABLE 3 (continued) Prehistoric, Historic, and Modern Records of Amazona in the West Indies Species

Island

Prehistoric

Historic

—

Rare in 18th century — Clark, 1905c; Ridgway, 1916:224; Olson, 1978; Snyder et al., 1987; Butler, 1992 — Ridgway, 1916:229–232; Butler, 1992 Ridgway, 1916:222–224; Butler, 1992 Ridgway, 1916: 231; Olson, 1978; Snyder et al., 1987; Butler, 1992 Ridgway, 1916:227–229; Butler, 1992 Ridgway, 1916:225–227; Butler, 1992 Butler, 1992

†A. violacea

Guadeloupe

†Amazona cf. violacea A. arausiaca

Marie Galante Dominica

A. imperialis

Dominica

—

†A. martinicana

Martinique

—

A. versicolor

St. Lucia

—

A. guildingii

St. Vincent

—

†Amazona undescr. sp.

Grenada

—

Tibiotarsus — this chapter —

Present Status Extinct

Extinct Endangered Endangered Extinct

Rare Rare Extinct

Note: Subspecies follow Ridgway (1916); † = extinct species, subspecies, or population.

†Amazona vittata — Antigua (Puerto Rican) Parrot Two bones were reported from the Indian Creek and Mill Reef archaeological sites on Antigua as Amazona sp. (Steadman et al., 1984a; Pregill et al., 1988). In Pregill et al. (1994) these bones were identified more precisely as A. vittata. While it has been suggested that this parrot may have been brought to the island by early human colonizers (Steadman et al., 1984a), the precultural fossil from Barbuda shows that this species is indigenous to Barbuda and therefore, presumably, to Antigua as well. It is possible that this species or species-complex once ranged from the northern Lesser Antilles to Puerto Rico. †Amazona undescribed sp. — Montserrat Parrot This small species is about the size of Amazona ventralis or A. agilis, smaller than all other West Indian species of Amazona. It is represented by five specimens (coracoid, two humeri, ulna, and femur) from the Trants archaeological site on Montserrat. One of the humeri was reported as Amazona sp. (smaller than any living Lesser Antillean species) by Reis and Steadman (1999). The other four specimens have come to light only recently, and demonstrate that an undescribed, extinct species of Amazona once inhabited Montserrat. †Amazona violacea (Gmelin, 1788) — Guadeloupe Parrot Although specimens are lacking, the early descriptions and observations of Amazona violacea (summarized in Greenway, 1958:222, 328; Snyder et al., 1987; Butler, 1992) provide a solid basis for believing that this large species did indeed exist but has been extinct since the early 1700s. †Amazona cf. violacea — Guadeloupe Parrot? A tibiotarsus from the Folle Anse archaeological site on Marie Galante represents a large species of Amazona. The fossil is much larger than in A. arausiaca, and most similar to that in A. imperialis,

186

Biogeography of the West Indies: Patterns and Perspectives

TABLE 4 Summary of Distribution of West Indian Psittacids on Major Islands or Island Groups Island Bahamas Cuba Isla de Pinos Cayman Islands Jamaica Hispaniola Puerto Rico Mona Culebra St. Croix Barbuda Antigua Montserrat Guadeloupe Marie Galante Dominica Martinique Barbados St. Lucia St. Vincent Grenada

Ara — †tricolor †tricolor — †gossei †erythrocephala †tricolor? and/or †unknown sp. — — — †autochthones — — †undescr. sp. †guadeloupensis †cf. guadeloupensis †atwoodi †martinica and/or †undescr. sp. — — — —

Number of Genera/Species

Aratinga

Amazona

—

leucocephala, undescr. sp. leucocephala leucocephala leucocephala agilis, collaria ventralis

1/2 3/3 3/3 1/1 3/5

v. vittata

euops †euops — nana c. chloroptera

3/3

†chloroptera maugei †chloroptera maugei — — †undescr. sp. — — †labati — †undescr. sp. †undescr. sp.

— †vittata gracilipes — †cf. vittata †cf. vittata †undescr. sp. †violacea †cf. violacea arausiaca, imperialis †martinicana

2/2 1/1 1/1 1/1 2/2 1/1 2/2 3/3 2/2 3/4 3/4

†undescr. sp. — — —

— versicolor guildingii †undescr. sp.

1/1 1/1 1/1 1/1

Note: † = Extinct species, subspecies, or population.

but with a slightly shorter overall length. The modern avifauna of Marie Galante shares many species with nearby Guadeloupe, so it is possible that the prehistoric tibiotarsus represents A. violacea, a large species known historically but now extinct (see above). †Amazona martinicana (Clark, 1905c) — Martinique Parrot Extinct since the 18th century, this large species was rather similar to A. violacea and the other very large Lesser Antillean species of Amazona, such as A. arausiaca (Ridgway, 1916:231; Greenway, 1958:328). ?Amazona versicolor (Müller, 1776) — St. Lucia Parrot A psittacid carpometacarpus from the Grand Anse archaeological site on St. Lucia is too fragmentary to be referred a genus. It is similar in size to the carpometacarpus in both Ara severa (extralocal) and Amazona versicolor, which is endemic to St. Lucia where it survives in low numbers (Keith, 1997). †Amazona undescribed sp. — Grenada Parrot This apparently large species is poorly known from a description by du Tertre (1667), as mentioned by Snyder et al. (1987) and Butler (1992).

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies

187

CONCLUSIONS Although many distributional gaps remain, the overall conclusion from the data presented is that at least one species in each of the three widespread West Indian psittacid genera (Ara, Aratinga, Amazona) once occurred on each major island in the Greater and Lesser Antilles (Table 4). In the Bahamas and Cayman Islands, only Amazona is known. Extinct are all species of West Indian macaws (Ara spp.), Lesser Antillean parakeets (Aratinga spp.), and parrots (Amazona spp.) between Puerto Rico and Dominica. The various extinctions probably occurred in prehistoric as well as historic times. Sympatric congeneric pairs of species may have occurred, at least locally, in Amazona and perhaps in Ara. Details of many of the species-level issues remain unresolved, although this situation can be improved through additional historical, zooarchaeological, and paleontological research on the fascinating but badly fragmented psittacid fauna of the West Indies. In the absence of both prehistoric and historic human impact, perhaps 50 to 60 species of macaws, parakeets, and parrots would inhabit the West Indies today. At least three fourths of these species already are gone.

ACKNOWLEDGMENTS We thank S. Scudder, F. Sergile, A. V. Stokes, A. Van Doorn, E. S. Wing, and C. A. Woods for their help with various aspects of our research. For access to specimens under their care, we also thank the curatorial staffs at the American Museum of Natural History, Field Museum of Natural History, Florida Museum of Natural History, National Museum of Natural History (Smithsonian Institution), and University of Michigan Museum of Zoology.

LITERATURE CITED Atwood, T. 1791. The History of the Island of Dominica. J. Johnson, London. Bangs, O. and W. R. Zappey. 1905. Birds of the Isle of Pines. American Naturalist 39:179–215. Bond, J. 1950. Check-List of Birds of the West Indies, 3rd ed. Academy of Natural Sciences, Philadelphia. Bond, J. 1956 [+ 25 supplements, 1956–1984]. Check-List of Birds of the West Indies, 2nd ed. Academy of Natural Sciences, Philadelphia. Bond, J. 1971. Birds of the West Indies, 2nd ed. Houghton Mifflin, Boston. Bradley, P. E. 1995. Birds of the Cayman Islands. Caerulea Press, Italy. Brodkorb, P. 1959. Pleistocene birds from New Providence Island, Bahamas. Bulletin of the Florida State Museum, Biological Sciences 4:349–371. Butler, P. J. 1992. Parrots, pressures, people, and pride. Pp. 25–46 in Beissinger, S. R. and N. F. R. Snyder (eds.). New World Parrots in Crisis; Solutions for Conservation Biology. Smithsonian Institution Press, Washington, D.C. Carlson, L. A. 1999. Aftermath of a Feast: Human Colonization of the Southern Bahamian Archipelago and Its Effects on the Indigenous Fauna. Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville. Casas, B. de las. 1876. Historia de las Indias, 5 vols. Madrid. Clark, A. H. 1905a. The Lesser Antillean macaws. Auk 22:266–273. Clark, A. H. 1905b. The genus Conurus in the West Indies. Auk 22:310–312. Clark, A. H. 1905c. The West Indian parrots. Auk 22:337–344. Clark, A. H. 1905d. The Greater Antillean macaws. Auk 22:345–348. Clark, A. H. 1908. The macaw of Dominica. Auk 25:309–311. Clark, A. H. 1911. A list of the birds of the island of St. Lucia. West Indian Bulletin 11:182–192. Cory, C. B. 1886. The birds of the West Indies, including the Bahama Islands, the Greater and Lesser Antilles, excepting the islands of Tobago and Trinidad. Auk 3:454–472. Dunn, O. and A. E. Kelley, Jr. 1989. The Diario of Christopher Columbus’s First Voyage to America 1492–3: Abstract by Fray Bartolomé las Casas. University of Oklahoma Press, Norman. ffrench, R. 1991. A Guide to the Birds of Trinidad and Tobago, 2nd ed. Cornell University Press, Ithaca, New York. Gosse, P. H. 1847. The Birds of Jamaica. Van Voorst, London.

188

Biogeography of the West Indies: Patterns and Perspectives

Greenway, J. C. 1958. Extinct and Vanishing Birds of the World. American Committee for International Wild Life Protection Special Publication 13. Hughes, G. 1750. Natural History of Barbadoes. London. Keegan, W. F. 1994. West Indian archaeology. 1. Overview and foragers. Journal of Archaeological Research 2:255–284. Keegan,W. F. 2000. West Indian Archaeology. 3. Ceramic age. Journal of Archaeological Research 8:135–167. Keith, A. R. 1997. The birds of St. Lucia, West Indies. British Ornithologists’ Union Check-list No. 15. Labat, J. B. 1742. Nouveau voyage aux isles de l’Amérique, contenant l’histoire naturelle de ces pays, l’origine, les moeurs, la religion & le gouvernement des habitants anciens et modernes. T. Le Gras, Paris. Marien, D. and K. F. Koopman. 1955. The relationships of the West Indian species of Aratinga (Aves, Psittacidae). American Museum Novitates 1712:1–20. Olson, S. L. 1978. A paleontological perspective of West Indian birds and mammals. Pp. 99–117 in Zoogeography of the Caribbean. Academy of Natural Sciences of Philadelphia, Special Publication 13. Olson, S. L. 1982. Biological archaeology in the West Indies. The Florida Anthropologist 35:162–168. Olson, S. L. and W. B. Hilgartner. 1982. Fossil and subfossil birds from the Bahamas. Pp. 22–56 in Olson, S. L. (ed.). Fossil Vertebrates from the Bahamas. Smithsonian Contributions to Paleobiology, No. 48. Oviedo, G. F., de. 1959. Natural History of the West Indies, edited and translated by S. A. Stoudemire. University of North Carolina Press, Chapel Hill. Pregill, G. K., D. W. Steadman, S. L. Olson, and F. V. Grady. 1988. Late Holocene Fossil Vertebrates from Burma Quarry, Antigua, Lesser Antilles. Smithsonian Contributions to Paleobiology, No. 463. Pregill, G. K., R. I. Crombie, D. W. Steadman, L. K. Gordon, F. W. Davis, and W. B. Hilgartner. 1991. Living and late Holocene fossil vertebrates, and the vegetation of the cockpit country, Jamaica. Atoll Research Bulletin 353:1–19. Pregill, G. K., D. W. Steadman, and D. R. Watters. 1994. Late Quaternary vertebrate faunas of the Lesser Antilles: Historical components of Caribbean biogeography. Bulletin of Carnegie Museum of Natural History, No. 30. Raffaele, H., J. Wiley, O. Garrido, A. Keith, and J. Raffaele. 1998. A Guide to the Birds of the West Indies. Princeton University Press, Princeton, New Jersey. Reis, K. R. and D. W. Steadman. 1999. Archaeology of Trants, Montserrat. Part 5. Prehistoric avifauna. Annals of the Carnegie Museum 68:275–287. Ridgway, R. 1916. The birds of North and Middle America (Part VII). Bulletin of the United States National Museum 50(part 7):1–543. Rochefort, C. C., de. 1658. Histoire naturelle et morale des Isles Antilles de l’Amérique. Chez Arnould Leers, Rotterdam. Rothschild, W. 1905. Untitled. (Notes on extinct parrots from the West Indies). Bulletin of the British Ornithologists’ Club 16:13–15. Rothschild, W. 1907. Extinct Birds. Hutchison, London. Salvadori, T. 1891. Catalogue of the Psittaci or Parrots in the Collection of the British Museum. British Museum of Natural History, London. Salvadori, T. 1906a. Notes on the parrots (Part V). Ibis, Series 8, 6:451–465. Salvadori, T. 1906b. Notes on the parrots (Part VI). Ibis, Series 8, 6:642–659. Snyder, N. F. R., J. W. Wiley, and C. B. Kepler. 1987. The Parrots of Loquillo: Natural History and Conservation of the Puerto Rican Parrot. Western Foundation of Vertebrate Zoology, Los Angeles, California. Steadman, D. W., G. K. Pregill, and S. L. Olson. 1984a. Fossil vertebrates from Antigua, Lesser Antilles: evidence for late Holocene human-caused extinctions in the West Indies. Proceedings of the National Academy of Sciences, U.S.A. 81:4448–4451. Steadman, D. W., D. R. Watters, E. J. Reitz, and G. K. Pregill. 1984b. Vertebrates from archaeological sites on Montserrat, West Indies. Annals of Carnegie Museum 53:1–29. Tertre, J. B., du. 1654. Histoire générale des isles St. Christophe, de la Guadeloupe, de la Martinique et autres dans l’Amérique. Chez Jacques Langlois et Emamanuel Langlois, Paris. Tertre, J. B., du. 1667. Histoire générale des Antilles habitées par les François, 3 vols. Chez Thomas Jolly, Paris. Todd, W. E. and W. W. Worthington. 1911. A contribution to the ornithology of the Bahama Islands. Annals of the Carnegie Museum 7:388–464. Walters, M. 1995. On the status of Ara tricolor Bechstein. Bulletin of the British Ornithologists’ Club 115:168–170.

The Historic and Prehistoric Distribution of Parrots (Psittacidae) in the West Indies

189

Wetmore, A. 1928. Bones of birds from the Ciego Montero deposit of Cuba. American Museum Novitates 301:1–5. Wetmore, A. 1937. Ancient records of birds from the island of St. Croix with observations on extinct and living birds of Puerto Rico. The Journal of Agriculture of the University of Puerto Rico 21:5–15. Wetmore, A. 1938. Bird remains from the West Indies. Auk 55:51–55. Wetmore, A. and B. H. Swales. 1931. The birds of Haiti and the Dominican Republic. United States National Museum Bulletin 155:1–483. Wilson, S. M. 1990. Hispaniola: Caribbean Chiefdoms in the Age of Columbus. University of Alabama Press, Tuscaloosa. Wing, E. S. 1989. Human exploitation of animal resources in the Caribbean. Pp. 137–152 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

Tertiary Vertebrate Fossils 13 Early from Seven Rivers, Parish of St. James, Jamaica, and Their Biogeographical Implications Roger W. Portell, Stephen K. Donovan,* and Daryl P. Domning Abstract — The Seven Rivers vertebrate site, parish of St. James, western Jamaica, is the oldest known land mammal site in the Antillean region (late early or early middle Eocene). The site represents a shallow-water, nearshore, marine paleoenvironment, most probably in an estuarine/deltaic setting, and is at least 12 million years older than the next oldest comparable site, situated in Puerto Rico. The vertebrate fauna from Seven Rivers is dominated by fully aquatic and amphibious taxa, including chondrichthyian and osteichthyian fishes, a crocodilian (possibly ?Charactosuchus kugleri), a pelomedusoid pleurodiran (side-necked) turtle, and two new species of prorastomid sirenian. However, it also includes three fully terrestrial taxa, the rhinocerotoid Hyrachyus sp., an iguanian lizard, and a possible archontan (most likely a plesiadapiform or a primate). Tectonic reconstruction suggests that in the Paleocene to early middle Eocene, the western portion of Jamaica, situated toward the eastern tip of the Nicaraguan Rise, was subaerially exposed and either very close to or connected to Central America. During this time, North American terrestrial tetrapods may have populated western Jamaica across a land bridge connection that was subaerially exposed. The first direct paleontological evidence of this comes from the discovery of terrestrial tetrapods at Seven Rivers. Indeed, these are the first Stage I (pre-inundation) terrestrial tetrapods to have been identified from the Antillean region.

INTRODUCTION Tertiary terrestrial tetrapods are very rare in the fossil record of the Antillean region. Those finds that have been made all occur in the Greater Antilles. These uncommon fossils are of particular importance for testing theories of island biogeography and Caribbean tectonics (both of which have been heavily debated for much of the 20th century). Recent significant finds have included a fascinating array of mammals from the Oligo-Miocene of Cuba, the Dominican Republic, and Puerto Rico (see, for example, MacPhee and Iturralde-Vinent, 1994, 1995a, 1995b; MacPhee and Grimaldi, 1996). These fossils are notable for their antiquity and their diversity, and include sloths, platyrrhine monkeys, rodents, and insectivores. Williams (1989:24) proposed a twofold division of Antillean biogeographical history into Stages I and II. Terrestrial faunas of Stage II are those which colonized the Antillean islands post-inundation, that is, after the islands rose above sea level for the last time. These faunas would obviously include those flightless mammals that have inhabited the Greater Antilles in the Pleistocene and Holocene, consisting solely of ground sloths, monkeys, rodents, and insectivores (Williams, 1989:table 1), while lacking survivors of the pre-inundation fauna (= Stage I). The known Oligo-Miocene taxa of the Greater Antilles represent a post-inundation, Stage II assemblage. As such, they potentially support a vicariant model for the origin of this mammal fauna (see, for example, discussion in MacPhee and Wyss, 1990:3–7), at least for these islands, although data are admittedly sparse. While at least some * Stephen K. Donovan’s contribution is © The Natural History Museum, London. 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

191

192

Biogeography of the West Indies: Patterns and Perspectives

of these taxa may also have been present in pre-inundation, Stage I assemblages, they would have been extirpated by inundation and would have had to repopulate the islands following final emergence. Thus, a true Stage I terrestrial mammal in the Greater Antilles would have to be one of a taxon that does not occur in the Quaternary biota and predates an undoubted “final” major period of inundation. The first Stage I mammal to be recognized from the Greater Antilles was recently collected from western Jamaica by the senior author (see also Domning et al., 1997). This specimen is a right dentary and dp3-m3 from the rhinocerotoid perissodactyl Hyrachyus sp. Herein, we document this fossil and the associated vertebrate fauna, and explain their importance as supporting evidence for tectonic and paleogeographical interpretations of Jamaica’s Paleogene evolution. Vertebrate specimens (and artificial casts of the same) collected from this excavation have been, and will be, deposited in the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM), the Florida Museum of Natural History, University of Florida, Gainesville, Florida (UF), and the Geology Museum, University of the West Indies, Mona, Jamaica (UWI). According to Domning and Clark (1993), the only Tertiary vertebrates reported from Jamaica, prior to our work, were Prorastomus sirenoides Owen, 1855 from the Eocene Stettin and Guys Hill members of the Chapelton Formation; ?Charactosuchus kugleri Berg, 1969 and unidentifiable fragments of turtle from the Guys Hill Member of the Chapelton Formation; and a supposed needlefish jaw, cf. Platybelone argalus (LeSueur, 1821) from the Pliocene Bowden Formation (Caldwell, 1965). The latter record, however, has been shown to be based on the misidentification of a shrimp claw (see Collins and Portell, 1998). Additionally, Fitch and Barker (1972) mentioned the occurrence of one “Miocene” species of fish otolith (Family Moridae) from an unstated locality (very possibly the Pliocene Bowden Formation); Fitch (1969) noted more than 110 kinds of fish otoliths from the Bowden Formation; and Clarke and Fitch (1979) reported that a 225-kg matrix sample from the Bowden Formation, collected for their study of teuthoid (cephalopod) statoliths, contained 25,000 fish otoliths. Recently, Stringer (1998) formally documented the Bowden shell bed otoliths which contained 68 species of teleost fish. Furthermore, Domning (1999) recorded the first Oligocene sea cow remains from the Browns Town Formation. Therefore, in addition to the Seven Rivers material discussed here, the depauperate Jamaican Tertiary vertebrate record comprises Eocene, Oligocene, doubtfully Miocene, and Pliocene marine taxa.

JAMAICAN TECTONICS AND PALEOGEOGRAPHY Draper (1987, 1998) proposed a model that divided the geological and tectonic evolution of Jamaica into four distinct phases (see also Robinson, 1994). During the early Cretaceous to early Cenozoic, Jamaica formed part of the Greater Antillean island arc; volcanism migrated from the central to the eastern part of the island in the late Cretaceous (Phase 1). The island was largely emergent in the latest Cretaceous–early Eocene (Phase 2), a time of intrusion and rifting (Figure 1). In the Eocene, the island became a submerged carbonate bank, perhaps similar to the Bahamas Bank at the present day (Phase 3). This phase of near-continuous limestone deposition in shallow to deeper water settings persisted throughout the mid-Cenozoic. The island was again uplifted about 10 mya and has remained tectonically active throughout the late Cenozoic (Phase 4). The early-middle Eocene was a time of marine transgression, which flooded western Jamaica (= early Phase 3). Volcanism continued, but was waning. Deposition of the Yellow Limestone Group commenced in the west (Robinson, 1988a:figure 3), spreading eastward in the early-middle Eocene. Deeper-water lithofacies were developed off the north coast and in the Wagwater graben, which flanks the southwestern Blue Mountain block in eastern Jamaica (Eva and McFarlane, 1985:figures 5, 6). The Yellow Limestone Group is a mixed sequence of limestones, evaporites, lignites, and siliciclastic rock units. It represents a succession of marginal marine to marine rocks, divided into a number of members of contrasting lithologies (Robinson, 1988a:figure 3), that were deposited as

Early Tertiary Vertebrate Fossils and Their Biogeographical Implications

193

Y PE UCA NI TA NS N UL A

MEXICO

NICARAGUA RISE CHORTIS BLOCK

CO ST A

RI CA

-P AN AM A

JAMAICA

AR C

N FIGURE 1 Reconstruction of the Central American region during the late early–early middle Eocene, redrawn after Robinson (1988b:fig. 6) and Pindell (1994:fig. 2.6k). The probable direction of migration of Hyrachyus is indicated by the arrows. White = land areas; v v v = active volcanic arc;    = shallow-water marine; stipple = deep-water marine;  = subduction zone;  = major thrust faults.

Jamaica sank below sea level. Quartzo-feldspathic sandstones with limestone lenses occur close to the paleoshoreline, with impure limestones occurring farther offshore. The succession that includes the Seven Rivers vertebrate site forms part of the Guys Hill Member of the Chapelton Formation, Yellow Limestone Group, of late-early or early-middle Eocene age. Diagnostic features of this member include the presence of dominant sandstones and associated siliciclastic sedimentary rocks, and locally abundant oysters, Carolia, and carbonized plant remains (Robinson, 1988a). Sedimentological and paleontological evidence favors an estuarine/deltaic environment of deposition for the Seven Rivers locality. Decrease in the proportion of clastic impurities through the sequence of the Yellow Limestone Group reflects the progressive submergence of the landmasses. In the late-middle Eocene, continued erosion and submergence led to the complete disappearance of exposed land areas. Total submergence led to sedimentation dominated by pure limestones of the White Limestone Group, totaling about 2.75 km in thickness (Robinson, 1994:121), which persisted until the middle-late Miocene, that is, over 30 million years. Although Perfit and Williams (1989:70) speculated that Jamaica may have been, in part, subaerially exposed during the mid-Cenozoic, available evidence supports near-continuous deposition of the White Limestone Group (for further discussion, see Domning et al., 1997:638).

194

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 2 Outline map of Jamaica showing the Seven Rivers area in relation to the cities of Montego Bay and Kingston.

LOCALITY AND VERTEBRATE FAUNA Sirenian rib fragments were first observed at Seven Rivers by J. Bryan and the senior author in 1990 (Figure 2). The exposed section is approximately 7 m thick, consisting of a sequence of mudrocks, siltstones, and fine- to medium-grained sandstones. Some units are gypsiferous or contain calcareous nodules. The associated fossil biota contains plant debris (including carbonized wood), moderately diverse benthic mollusks, benthic foraminifers, rarer asteroid ossicles, fragmentary echinoids and crustaceans, and abundant bones, particularly at certain stratigraphic horizons. Excavations at Seven Rivers have been motivated primarily by the desire to obtain further and more complete specimens of the earliest sirenians. The search for these remains has focused on the Chapelton Formation, Yellow Limestone Group, in western Jamaica, because this unit previously yielded the unique type specimen of Prorastomus sirenoides Owen, 1855. Owen’s type specimen, which comprises only the skull, mandible, and atlas vertebra, is the world’s oldest and most primitive sirenian (Savage et al., 1994). Although other, roughly coeval Jamaican localities had yielded additional sirenian remains (see Donovan et al., 1990), these provided only minimal evidence of the postcranial skeleton of the animal. It was hoped that the Seven Rivers site would yield more adequate material when annual excavations commenced in 1994. This hope of finding postcranial (as well as further cranial; Figure 3) material has now been abundantly realized and the Seven Rivers excavation continues to produce bones in gratifying numbers (Domning et al., 1995). It appears that these remains do not represent P. sirenoides, but instead they indicate the presence of at least one, and probably two, taxa of the same family that are slightly more derived. However, these are very close to Prorastomus in their stage of evolution and give a clear picture of the sort of creatures that were the prorastomid sea cows. Perhaps remarkably, these fossils confirm the accuracy of the speculative reconstruction by R. J. G. Savage (in Dixon et al., 1988), that was based only on the holotype skull of P. sirenoides that was originally described by Owen (1855). This reconstruction depicts a pig-sized, barrel-chested animal with four stout legs, stubby toes, and a long and muscular, but slender, tail that was not much modified as a swimming organ. Although these prorastomids had well-developed legs, and at least the earliest of them could support their bodies on land, it is clear that they spent most of their time in the water, where they probably fed on aquatic plants in rivers, estuaries, and lagoons. This aquatic habit is indicated by their enlarged and retracted nasal openings (which facilitated breathing at the water’s surface) and

Early Tertiary Vertebrate Fossils and Their Biogeographical Implications

195

FIGURE 3 Partial skull (rostrum missing) of undescribed species of prorastomid sirenian from the late early or early middle Eocene of Seven Rivers, parish of St. James, western Jamaica (to be deposited at the USNM): (A) dorsal view; (B) ventral view.

their typically sirenian thick, dense ribs (which provided ballast for submerging). They evidently swam by forcefully extending the spinal column, kicking backward and upward with the hind feet used simultaneously, and, perhaps, deriving some additional thrust from the tail. This is the swimming style hypothesized for early whales, such as Ambulocetus (Thewissen et al., 1994), and is also observed in otters today. Later sirenians and cetaceans, in contrast, lost the hind legs and strengthened the tail, adding a horizontal caudal fin for optimal tail-only propulsion. The lower and upper bone beds at Seven Rivers have produced sirenian fossils that represent two distinct stages in this evolutionary transition from terrestrial to fully aquatic locomotion; these exactly parallel the stages through which primitive whales were evolving at the very same time. The earlier stage is reflected in a type of sacrum that consists of at least four vertebrae, rigidly articulated or fused together like those of land mammals and strongly connected to the pelvis. In the later stage,

196

Biogeography of the West Indies: Patterns and Perspectives

the sacrum has been reduced to a single vertebra, which alone joins with the pelvis. This indicates a reduced ability to support the body’s weight out of water and an increased flexibility of the spine in the pelvic region. Together with the features of the ribs and nasal opening cited above, these modifications of the sacral vertebrae testify that the Jamaican prorastomids had already made an irrevocable commitment to aquatic life. This is not surprising, considering where they were found. Sirenians, as close relatives of such groups as proboscideans, embrithopods, desmostylians, and hyracoids, surely shared with them an origin in the Old World, presumably along the shores of the former Tethys Seaway between Eurasia and Africa. The anomalous fact that the most primitive sea cows have instead turned up in the West Indies is apparently an accident of paleontological sampling. A possible prorastomid vertebra has recently been found in the Eocene of Israel (Goodwin et al., 1998). It appears that as soon as ancestral sirenians gained some enhanced ability to swim, they were able to disperse rapidly westward along the tropical shores of Tethys to what are now Central America and the West Indies. Rare fossils from the middle and late Eocene of Florida may represent prorastomids (Savage et al., 1994), and further collecting in Eocene, or even late Paleocene, marginal marine deposits elsewhere in the Tethyan realm should eventually reveal the true history of their evolution and dispersal. Among the hundreds of sirenian bones collected from Seven Rivers, a single right dentary and dp3-m3 of a further large mammal has been collected, attributable to the rhinocerotoid perissodactyl Hyrachyus (Figure 4), close to H. affinis (Marsh, 1871). This specimen was described in detail by Domning et al. (1997), where its affinities and relationships were also discussed. This specimen is considered to be at least 12 million years older than the next oldest Antillean mammal, an early Oligocene sloth from Puerto Rico (MacPhee and Iturralde-Vinent, 1995b). The paleogeographical implications of the Jamaican Eocene Hyrachyus are discussed above and below. The Seven Rivers site has also yielded fossils of fishes, crocodilians, turtles, a lizard, and most recently a possible archontan. The fishes have yet to be studied in detail, but include sharks and rays. The crocodilians are eusuchians, but have not been more precisely identified. They may represent ?Charactosuchus kugleri Berg, 1969, described from the Dump Limestone lenticle of the Guys Hill Member, Chapelton Formation (that is, the same member as the Seven Rivers site), in the parish of Manchester, western Jamaica. The genus is otherwise known only from the Miocene of the Caribbean region, but ?C. kugleri may instead be synonymous with Dollosuchus dixoni (Owen, 1850), from the middle Eocene of England and Belgium (Domning and Clark, 1993). The turtle remains, currently under study by E. Gaffney and R. Weems, represent a pelomedusoid pleurodiran (side-necked) turtle. It is probably a podocnemidid and is similar to members of the shweboemydine group, which are known from Tertiary nearshore/fluvial or marine sediments in Venezuela, Puerto Rico, Cuba, North Africa, and Asia. The lizard (USNM 489192), an iguanian and possibly an anoloid (?Polychrotidae), consists of four dentary fragments, two of which bear a single pleurodont tooth. It represents the oldest terrestrial reptile known from the Caribbean (Pregill, 1999). During the 1999 Seven Rivers field season a second land mammal, a possible archontan currently under study by R. MacPhee et al., was discovered. It is most likely a plesiadapiform or a primate and is represented by an incomplete right petrosal (MacPhee et al., 1999).

DISCUSSION As succinctly stated by Donnelly (1988:15–16), “the Caribbean has remained one of the most controversial areas in the world for geologic reconstructions.” The complex geological history of the Caribbean region has been explained by a plethora of published evolutionary models. These models have relied on a variety of interpretations of the many sources of data available, both plate tectonic (such as Donnelly, 1988, and references therein) and otherwise (Morris et al., 1990, and references therein). Herein, we favor the interpretation of Pindell and co-workers (see, for example, Pindell and Barrett, 1990; Pindell, 1994), which has received wide acceptance, at least in its broad details. It also permits the most reasonable explanation of the occurrence of Hyrachyus, and the

Early Tertiary Vertebrate Fossils and Their Biogeographical Implications

197

FIGURE 4 Right dentary and dp3-m3 of Hyrachyus sp. (USNM 489191) from the late-early or early middle Eocene of Seven Rivers, parish of St. James, western Jamaica (after Domning et al., 1997:figure 1): (A) medial view; (B) lateral view; (C) occlusal view. Scales in cm (A and B) and mm (C).

other associated terrestrial tetrapods, in Jamaica, which we regard as strong, and independent, supporting evidence for the essential correctness of an important aspect of Pindell’s model. The crust of the Caribbean Plate is one of the largest Phanerozoic oceanic plateau basalt provinces in the world (Donnelly, 1994). Available evidence supports an origin for this plate in the Pacific (Pindell, 1990); to grossly simplify a complex process (see Pindell, 1994 for greater detail), the plate was pushed between North and South America, resulting in the subduction of the socalled “Proto-Caribbean Plate.” Perhaps the most compelling evidence for such an origin of the Caribbean Plate is that the best date we have for the opening of the Caribbean seaway is late-middle Jurassic, about 165 mya. However, obducted seafloor sedimentary rocks in southwest Puerto Rico contain microfossils that are 30 million years older, indicating that the Caribbean Plate was formed before the opening of the seaway (Montgomery et al., 1994). During the mid to late Cretaceous, an island arc, including the terranes that constitute the Greater Antilles, formed at the leading edge of the Caribbean Plate as the “Proto-Caribbean Plate” was subducted beneath it. This island arc system was disrupted and fragmented from the late

198

Biogeography of the West Indies: Patterns and Perspectives

Cretaceous onward, as the Caribbean Plate was transported eastward between the North and South American Plates. From the Paleocene to the early Eocene, at least the western portion of Jamaica, at the northeastern edge of the Nicaragua Rise (itself a topographic extension of northern Central America) was subaerially exposed and situated adjacent to the Yucatan Peninsula, from which it was separated by a transform fault (Pindell, 1994:figure 2.6j; Figure 1, this chapter) (= Phase 2 of Draper; see above). This subaerial exposure is interpreted as due to the Yucatan block preventing simple east–northeast movement of the Chortis Block–Nicaragua Rise–Jamaica, resulting in uplift and block faulting of the Nicaragua Rise. As was recognized by Donnelly (1988:28), this formed a terrestrial connection between Central America and Jamaica via the Nicaragua Rise. This connection was to be severed by the middle Eocene following submergence of the Nicaragua Rise. Only after it had moved past this “obstruction” formed by the Yucatan did Jamaica and the Nicaragua Rise enter a period of tectonic quiescence, and become a Bahamas Bank-like carbonate platform (= Phase 3 of Draper). However, during Phase 2, immigration of the North America terrestrial biota would have been facilitated by the continuous landmass of North America–Mexican Arc–Chortis Block–Nicaragua Rise–Jamaica, as indicated by the trail of arrows in Figure 1. Hyrachyus sp. from the Seven Rivers site predates the Phase 3 inundation of Jamaica and is thus part of the Stage I terrestrial mammal fauna of the island, based on the evidence of both taxonomic assignment and geology. Except for the iguanian lizard and possible archontan, all other vertebrates thus far discovered from Seven Rivers are obligate aquatic or, at best, amphibious organisms, the latter group including ?C. kugleri and prorastomid sirenians. Hyrachyus is well known from the Eocene of Eurasia and North America (Radinsky, 1969). Such a large terrestrial mammal (the size of a large dog) is unlikely to have been dispersed across broad water barriers, so its presence in Jamaica must be recognized as strong supporting evidence for an ancient landbridge connection with North America (Figure 1). Although MacPhee and Wyss (1990:3) correctly considered that the “land-bridge argument … suffers from an acute lack of supporting geological fact” for explaining the evolution of the Stage II fauna, it must now be considered a plausible explanation for at least part of the Stage I fauna, that of Jamaica.

ACKNOWLEDGMENTS Fieldwork in Jamaica was supported by National Geographic Society grants nos. 5116-93, 5327-94, and 5562-95. The additional financial support provided by Barbara and Reed Toomey is gratefully acknowledged. We thank the many collaborators who made vital contributions to the Seven Rivers excavation, including Hal Dixon, Trina MacGillivray, Simon Mitchell (all University of the West Indies, Mona), Kevin Schindler, Barbara and Reed Toomey, Kaffie Commins, Douglas Jones, Craig Oyen, and Debra Krumm (all Florida Museum of Natural History, Gainesville). Steve and Suzan Hutchens skillfully prepared many of the sea cow, turtle, and crocodile bones. Fred Grady helped to prepare the Hyrachyus mandible, which was photographed by P. Kroehler. Trevor A. Jackson (University of the West Indies, Mona) and Gary S. Morgan (New Mexico Museum of Natural History, Albuquerque) are thanked for critically reading parts of an early draft of this chapter. S. K. D. thanks Angela Milner and Jerry Hooker (both of The Natural History Museum, London) for invaluable service tracing key references. This is University of Florida Contribution to Paleobiology 507 and a contribution to Natural History Museum project #298, “Palaeoecology and Climatic History of Cainozoic Land Biota.”

LITERATURE CITED Berg, D. M. 1969. Charactosuchus kugleri, eine neue Krokodilart aus dem Eozän von Jamaica. Eclogae Geologicae Helvetiae 62:731–735. Caldwell, D. K. 1965 (1966). A Miocene needlefish from Bowden, Jamaica. Florida Academy of Sciences Quarterly Journal 28(4):339–344.

Early Tertiary Vertebrate Fossils and Their Biogeographical Implications

199

Clarke, M. and J. E. Fitch. 1979. Statoliths of Cenozoic teuthoid cephalopods from North America. Palaeontology 22:479–511. Collins, J. S. H. and R. W. Portell 1998. Decapod, stomatopod and cirripede crustacea from the Pliocene Bowden Shell Bed, St. Thomas Parish, Jamaica. Pp. 113–127 in Donovan, S. K. (ed.). The Pliocene Bowden Shell Bed, Southeast Jamaica. Contributions to Tertiary and Quaternary Geology 35(1–4). Dixon, D., B. Cox, R. J. G. Savage, and B. Gardiner. 1988. Macmillan Illustrated Encyclopedia of Dinosaurs and Prehistoric Animals. MacMillan, London. Domning, D. P. 1999. Oligocene sirenians of the Caribbean region. Appendix 1, p. 29 in Dixon, H. L. and S. K. Donovan. Report of a field meeting to the area around BrownsTown, parish of St. Ann, northcentral Jamaica, 21 February, 1998. Journal of the Geological Society of Jamaica 33:24–30. Domning, D. P. and J. M. Clark. 1993. Jamaican Tertiary marine Vertebrata. Pp. 413–415 in Wright, R. M. and E. Robinson (eds.). Biostratigraphy of Jamaica. Geological Society of America, Memoir 182, Boulder, Colorado. Domning, D. P., S. K. Donovan, H. L. Dixon, R. W. Portell, and K. S. Schindler 1995. The world’s most primitive seacow: a new sirenian site in western Jamaica. Geological Society of America, Abstracts with Programs 27(6):A386. Domning, D. P., R. J. Emry, R. W. Portell, S. K. Donovan, and K. S. Schindler 1997. Oldest West Indian land mammal: rhinocerotoid ungulate from the Eocene of Jamaica. Journal of Vertebrate Paleontology 17:638–641. Donnelly, T. W. 1988. Geologic constraints on Caribbean biogeography. Pp. 15–37 in Liebherr, J. K. (ed.). Zoogeography of Caribbean Insects. Comstock Publishing, Cornell University Press, Ithaca, New York. Donnelly, T. W. 1994. The Caribbean sea floor. Pp. 41–61 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. University of the West Indies Publishers’ Association, Kingston, Jamaica. Donovan, S. K., D. P. Domning, F. A. Garcia, and H. L. Dixon. 1990. A bone bed in the Eocene of Jamaica. Journal of Paleontology 64:660–662. Draper, G. 1987. A revised tectonic model for the evolution of Jamaica. Pp. 151–169 in Ahmad, R. (ed.). Proceedings of a Workshop on the Status of Jamaican Geology. Journal of the Geological Society of Jamaica, Special Issue 10. Draper, G. 1998. Geological and tectonic evolution of Jamaica. Contributions to Geology, UWI, Mona 3:3–9. Eva, A. and N. McFarlane. 1985. Tertiary to early Quaternary carbonate facies relationships in Jamaica. Transactions of the 4th Latin American Geological Congress, Port-of-Spain, Trinidad, 7th–15th July, 1979, 1:210–219. Fitch, J. E. 1969. Fossil lanternfish otoliths of California, with notes on fossil Myctophidae of North America. Los Angeles County Museum, Contributions in Science 173:1–20. Fitch, J. E. and L. W. Barker. 1972. The fish Family Moridae in the eastern North Pacific with notes on morid otoliths, caudal skeletons, and the fossil record. Fishery Bulletin 70(3):565–584. Goodwin, M. B., D. P. Domning, J. H. Lipps, and C. Benjamini. 1998. The first record of an Eocene (Lutetian) marine mammal from Israel. Journal of Vertebrate Paleontology 18(4):813–815. LeSueur, C. A. 1821. Observations on several genera and species of fish, belonging to the natural family of the Esoces. Journal of the Academy of Natural Sciences of Philadelphia 2(1):124–138. MacPhee, R. D. E. and D. A. Grimaldi. 1996. Mammal bones in Dominican amber. Nature 380:489–490. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1994. First Tertiary land mammal from Greater Antilles: an early Miocene sloth (Xenarthra, Megalonychidae) from Cuba. American Museum Novitates 3094:1–13. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1995a. Earliest monkey from Greater Antilles. Journal of Human Evolution 28:197–200. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1995b. Origin of Greater Antillean land mammal fauna, 1: New Tertiary fossils from Cuba and Puerto Rico. American Museum Novitates 3141:1–30. MacPhee, R. D. E. and A. R. Wyss. 1990. Oligo-Miocene vertebrates from Puerto Rico, with a catalog of localities. American Museum Novitates 2965:1–45. MacPhee, R. D. E., C. Flemming, D. P. Domning, R. W. Portell, and B. Beatty. 1999. Eocene ?primate petrosal from Jamaica: morphology and biogeographical implications. Fifty-ninth Annual Meeting of Society of Vertebrate Paleontology — Abstracts of Papers. Journal of Vertebrate Paleontology 19(3):61A. Marsh, O. C. 1871. Notice of some fossil mammals from the Tertiary formation. American Journal of Science and Arts (3)11, 8:36–37.

200

Biogeography of the West Indies: Patterns and Perspectives

Montgomery, H., E. A. Pessagno, Jr., and J. L. Pindell. 1994. A 195 Ma terrane in a 165 Ma sea: Pacific origin of the Caribbean Plate. GSA Today 4(1):1, 3–6. Morris, A. E. L., I. Taner, H. A. Meyerhoff, and A. A. Meyerhoff. 1990. Tectonic evolution of the Caribbean region; alternative hypothesis. Pp. 433–457 in Dengo, G. and J. E. Case (eds.). The Geology of North America, Vol. H, The Caribbean Region. Geological Society of America, Boulder, Colorado. Owen, R. 1850. Monograph of the fossil Reptilia of the London Clay, and of the Bracklesham and other Tertiary beds. Part 2. Crocodilia (Crocodilius, etc.) Monograph of the Palaeontographical Society, London 5–50. Owen, R. 1855. On the fossil skull of a mammal (Prorastomus sirenoides, Owen), from the island of Jamaica. Quarterly Journal of the Geological Society 11:541–543. Perfit, M. R. and E. E. Williams. 1989. Geological constraints and biological retrodictions in the evolution of the Caribbean Sea and its islands. Pp. 47–102 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Pindell, J. L. 1990. Geological arguments suggesting a Pacific origin for the Caribbean Plate. Pp. 1–4 in Larue, D. K. and G. Draper (eds.). Transactions of the 12th Caribbean Geological Conference, St. Croix, Virgin Islands, 7th–11th August, 1989. Miami Geological Society, Miami, Florida. Pindell, J. L. 1994. Evolution of the Gulf of Mexico and the Caribbean. Pp. 13–39 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. University of the West Indies Publishers’ Association, Kingston. Pindell, J. L and S. F. Barrett. 1990. Geological evolution of the Caribbean region; a plate-tectonic perspective. Pp. 405–432 in Dengo, G. and J. E. Case (eds.). The Geology of North America, Vol. H, The Caribbean Region. Geological Society of America, Boulder, Colorado. Pregill, G. K. 1999. Eocene lizard from Jamaica. Herpetologica 55(2):157–161. Radinsky, L. B. 1969. The early evolution of the Perissodactyla. Evolution 23:308–328. Robinson, E. 1988a. Late Cretaceous and early Tertiary sedimentary rocks of the Central Inlier, Jamaica. Journal of the Geological Society of Jamaica 24 (for 1987):49–67. Robinson, E. 1988b. Early Tertiary larger foraminifera and platform carbonates of the northern Caribbean. Pp. 5:1–5:12 in Barker, L. (ed.). Transactions of the 11th Caribbean Geological Conference, Dover Beach, Barbados, July 20–26, 1986. Energy and Natural Resources Division, National Petroleum Corporation, Barbados. Robinson, E. 1994. Jamaica. Pp. 111–127 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. University of the West Indies Publishers’ Association, Kingston. Savage, R. J. G., D. P. Domning, and J. G. M. Thewissen. 1994. Fossil Sirenia of the West Atlantic and Caribbean region. V. The most primitive known sirenian, Prorastomus sirenoides Owen, 1855. Journal of Vertebrate Paleontology 14:427–449. Stringer, G. L. 1998. Otolith-based fishes from the Bowden Shell Bed (Pliocene) of Jamaica: systematics and palaeoecology. Pp. 147–160 in Donovan, S. K. (ed.). The Pliocene Bowden Shell Bed, Southeast Jamaica. Contributions to Tertiary and Quaternary Geology 35(1–4). Thewissen, J. G. M., S. T. Hussain, and M. Arif. 1994. Fossil evidence for the origin of aquatic locomotion in archaeocete whales. Science 263:210–212. Williams, E. E. 1989. Old problems and new opportunities in West Indian biogeography. Pp. 1–46 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

Sloths of the West Indies: 14 The A Systematic and Phylogenetic Review Jennifer L. White and Ross D. E. MacPhee Abstract — Megalonychid sloths are well known from the Quaternary of many West Indian islands, and some isolated remains date back as far as the early Oligocene. Phylogenetic relationships among these sloths have been unclear and controversial for numerous reasons, thus hindering biogeographical interpretations. As part of a complete systematic revision of West Indian megalonychids, we performed a cladistic analysis on 17 ingroup taxa using 69 osteological and dental characters. The analysis discovered three most-parsimonious trees that differ only in the placement of one incompletely known taxon. Two well-supported clades are identified, one of which includes the extant two-toed sloth Choloepus. Each major clade contains two or more genera, and within each genus, sister species are distributed across islands. Our results allow us to address two major biogeographical issues: the colonization of the Greater Antilles by megalonychid sloths, and the inter-island relationships of these sloths. An initial emplacement in the early Oligocene is consistent with overland dispersal across a short-lived land span connecting the developing Greater Antilles with northwestern South America. Our data suggest at least two separate invasions, and therefore a diphyletic origin of Antillean megalonychids. Inter-island distributions are explained most parsimoniously by island–island vicariance sometime before the end of the Miocene.

INTRODUCTION Until the middle Holocene or perhaps somewhat later, megalonychid sloths formed a significant component of the land mammal fauna of the insular Neotropics. All of these “Antillean sloths” are now extinct (MacPhee, 1997a; MacPhee et al., 1999). Fortunately for science, their remains — often in some abundance — have been found in Cuba, Puerto Rico, Hispaniola, La Gonâve, Ile de la Tortue, Curaçao, and, most recently, Grenada (MacPhee et al., 2000a). Many smaller islands in the Caribbean region have not been adequately surveyed (e.g., Vieques and other islands on the Puerto Rican shelf; Venezuelan Antillas Menores), and the possibility that these sloths had an even wider distribution within the West Indies needs to be seriously considered. Although there is no doubt that Antillean sloths are proximally related to the extant two-toed sloth (Choloepus) and the famous Neogene megalonychids of Argentina (Kraglievich, 1923; Scillato-Yané, 1979; Pascual et al., 1985), the history of this family in northern South America during the latter part of the Cenozoic is exceedingly obscure (Iturralde-Vinent and MacPhee, 1999). Paleontologically speaking, we know the climax and denouement of the story of sloth evolution in the West Indies, but of the early chapters, beginning with the first arrival of phyllophagans on the islands at least 32 million years ago (mya), we know next to nothing. That we are so well informed about the terminal phases of sloth evolution on these islands is largely due to the excellence of conditions for Quaternary fossil preservation in the caves that are one of the main features of Antillean karst landscapes. The trouble is, of course, that caves are evanescent structures, particularly in the tropics. Although it is possible that older karst regions on some islands may eventually yield demonstrably older faunas, we are not aware of any suitable candidates (MacPhee, 1997b). Apart from cave contexts, terrestrially derived basinal sediments in which land mammal fossils might be found are extremely rare in the West Indies. Nevertheless, things have improved in the last decade or so, with the discovery of pre-Pleistocene deposits on several islands that have yielded evidence 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

201

202

Biogeography of the West Indies: Patterns and Perspectives

of earlier faunas — regrettably not yet enough to fill many blank pages in the story, but enough to know that the plot will be interesting (e.g., MacPhee and Wyss, 1990; MacPhee and IturraldeVinent, 1994, 1995; MacPhee and Grimaldi, 1996; Domning et al., 1997; Iturralde-Vinent and MacPhee, 1999; MacPhee et al., 2000a, 2000b). It is widely accepted that all phyllophagans that have been identified in the West Indies are members of family Megalonychidae, but their phylogenetic relationships inter se have long been controversial. For example, there is much disagreement among specialists concerning the number and composition of subfamilies, tribes, genera, and species. For many taxa, hypodigms (i.e., the totality of specimens assigned to a particular species) are severely muddled as a result of numerous reassignments of material and confusing name changes. A relatively large number of species have been named on the basis of isolated cranial fragments or single limb elements, leading, in our opinion, to an unnecessary proliferation of nomina and noncomparable type specimens. Few associated skeletons have been found (or at least reported as such), and because of the number of named taxa in play it has proven difficult to connect crania with postcrania in many cases. An additional problem is that, for island megafaunal species in general, the influence of intraspecific variation in shape and size (particularly in postcrania) has been greatly underappreciated (cf. White, 1993a, 1993b; McFarlane et al., 1998). Prior to the senior author’s cladistic investigation of Antillean phyllophagans (White 1993a, 1993b; White et al., 1996), little attention had been given to the phylogenetic relationships of sloth species living on different West Indian islands. Indeed, one could examine the existing plethora of names and come to the conclusion that each island must have been separately colonized by its own sloth propagules coming directly from South America, as scarcely any islands shared the same genus-level taxa. Although the biogeographical improbability of this was obvious to some (e.g., Varona, 1974), the fact remains that the complex systematic history of Antillean sloths has never been adequately surveyed. In this chapter we attempt to resolve some of the major issues that need to be confronted in revising Antillean megalonychids, recognizing that some solutions will not be forthcoming until more basic revisory work has been completed.

BRIEF OVERVIEW OF MEGALONYCHID DISCOVERIES IN THE WEST INDIES The first discovery of an Antillean sloth to receive public notice occurred in 1861, when a partial mandible was recovered from the Chapepote casimba (i.e., hot spring) at Ciego Montero in Sierra de Jatibonico, near Cienfuegos in south-central Cuba (Figure 1). Although some observers thought the jaw might represent a hippo or a giant rodent, Joseph Leidy (1868) recognized that it was incontestably that of a sloth related to Megalonyx, which he named Megalocnus rodens. Despite this promising find, the pace of discovery of Antillean sloths was slow until the early decades of the 20th century, when, in rapid succession, a host of new finds were made in Cuba, Puerto Rico, and Hispaniola. The first intimation that a variety of megalonychid sloths had once existed in the Greater Antilles came with the publication of a paper on the sloths of Cuba by Carlos de la Torre and William D. Matthew (1915), who briefly discussed collections that had just been made (in 1911) by de la Torre and Barnum Brown at the same series of hot springs at Ciego Montero that had yielded the type jaw of M. rodens. In addition to the latter species, these authors recognized three new kinds of sloths in this material, provisionally giving them the genus-level names Microcnus, Mesocnus, and Miocnus. (To the unending frustration of later students of Antillean sloths, these names technically remained nomina nuda until 1931, when Matthew in a posthumous paper formally named and described type species for each.) Matthew (1919) emphasized that all of these taxa were closely related to the Miocene megalonychids of South America, confirming Leidy’s original insight. The next discovery of importance was by Harold E. Anthony (1916), who described the distinctive Puerto Rican species Acratocnus odontrigonus on the basis of a large quantity of material

The Sloths of the West Indies: A Systematic and Phylogenetic Review

North and Central American megalonychids

ISLA DE PINOS

Megalocnus rodens Neocnus major Neocnus gliriformis Parocnus browni Acratocnus antillensis Imagocnus zazae Species B Species C ILE DE LA TORTUE CUBA

203

Megalocnus zile Acratocnus ye Parocnus serus Neocnus comes Neocnus dousman Neocnus toupiti PUERTO RICO

GONAVE

H I S PA NIOLA Acratocnus odontrigonus Species A

Paulocnus petrifactus

Species D GRENADA CURAÇAO

500 KILOMETERS

South American megalonychids

FIGURE 1 Sketch map of the Caribbean region, showing general distribution of megalonychids on the mainland and in the West Indies (cf. Table 2). Megalonychids arose in South America, presumably in the early Paleogene. Their pre-Miocene history is poorly known, but there is evidence that by the early Oligocene they had penetrated the Antilles (species A, Puerto Rico). The oldest Cuban evidence (Imagocnus zazae) is early Miocene. By 9 mya, and perhaps well before, megalonychids were present in Central and North America, but these species were different from the ones that had penetrated the West Indies. As time went on, significant radiations of megalonychids occurred in Cuba and Hispaniola, leading to a multiplicity of species on those islands by the end of the Pleistocene. Single species are also known from the Quaternary of Puerto Rico and Curaçao, and from the late Pliocene or early Pleistocene of Grenada (species D). All mainland megalonychids (except Choloepus) are thought to have become extinct by the end of the Pleistocene; Antillean taxa held on until the middle-late Holocene (at least on some islands), but all were gone well before European arrival. For more detailed coverage of Antillean sloth localities: Cuba and satellites (Matthew and Paula Couto, 1959; MacPhee and Iturralde-Vinent, 1994); Hispaniola and satellites (Miller, 1929; MacPhee et al., 2000b); Puerto Rico (Anthony, 1918); Curaçao (Hooijer, 1964); Grenada (MacPhee et al., 2000a).

recovered from cave sites on that island. Although it was obvious that Acratocnus was a near relative of Megalocnus (and the other, then incompletely named, species identified by Matthew), it was also obvious that it was rather different. Anthony’s discovery raised the possibility that a major radiation of sloths, heretofore completely unsuspected, had occurred in the Greater Antilles.* * It should be noted that Anthony (1920, 1940) thought that there should have been sloths on Jamaica as well, and in 1919–1920 he conducted the first paleontological reconnaissance of that island in order to find out. Contrary to his expectations, he found no sloths — and neither has anyone else, leaving Jamaica as the only Greater Antillean landmass to lack phyllophagans (for discussion, see Iturralde-Vinent and MacPhee, 1999).

204

Biogeography of the West Indies: Patterns and Perspectives

This possibility was confirmed shortly thereafter by Gerrit S. Miller (1922, 1929) and his collectors, who discovered remains of medium-sized and large sloths in numerous cave sites in Haiti and the Dominican Republic. At first Miller (1922) assigned these remains to Megalocnus, albeit with a question mark. Later, realizing that two quite different species were represented in the Hispaniolan sample and that neither conformed to Megalocnus as known from Cuba, he proposed the new taxa Parocnus serus and Acratocnus comes for their reception (Miller, 1929). Although these discoveries were highly significant, very few collecting expeditions were mounted in Hispaniola after Miller’s time, and even fewer resulted in publications (e.g., Hoffstetter, 1955; Hooijer and Ray, 1964). In the late 1950s, Carlos de Paula Couto undertook an extensive study of all Cuban sloth material then housed in the AMNH and several other museums. As part of this effort he completed a manuscript, published subsequently as Matthew and Paula Couto (1959), which Matthew begun decades earlier but never finished in his lifetime. This comprehensive review recognized four genera of Cuban sloths: Megalocnus, Mesocnus, Microcnus, and Acratocnus (into which last taxon he placed all fossils previously assigned to Matthew’s Miocnus). Shortly thereafter, Oscar Arredondo proposed the erection of two new Cuban genera, Neomesocnus and Neocnus, including within the latter the new species N. major and N. minor (Arredondo, 1961). At that time Arredondo considered Microcnus to be a good genus, distinct from Neocnus. In the following year, Dirk A. Hooijer named a new genus and species, Paulocnus petrifactus, based on a partial skull and some limb bones from Curaçao (Hooijer, 1962, 1964, 1967). In 1967 Paula Couto published a review of all West Indian sloth taxa. While this comprehensive investigation built upon earlier work, it presented several new taxonomic interpretations. First, whereas Matthew and Paula Couto (1959) placed all Miocnus material in the genus Acratocnus, in his later paper Paula Couto (1967) reestablished the validity of Miocnus as a separate genus. Second, he suggested that Neomesocnus might be a synonym of Megalocnus, and that both species of Neocnus were synonyms of Microcnus gliriformis. Third, he erected the new genus Synocnus to receive the Hispaniolan material previously relegated to Acratocnus comes by Miller (1929). Without making reference to Paula Couto’s proposed synonymy of Microcnus and Neocnus, in the following year Milos Kretzoi (1968) proposed Cubanocnus as a replacement name for Microcnus s.s. (which Kretzoi had shown to be preoccupied by Microcnus, proposed as a subgenus of Botaurus [Aves] in 1877). Nevertheless, a few years later, Karl-Heinz Fischer (1971) employed the name Microcnus in his detailed morphological description of newly recovered sloth fossils from Cuba. At first, Luis S. Varona (1974) recognized Cubanocnus as the valid name for material that had been assigned previously to Microcnus and Neocnus, but shortly thereafter he abandoned this position and placed the combined hypodigms of these taxa under a single species name, N. gliriformis (Varona, 1976). In other respects Varona’s (1974) taxonomy of Cuban sloths essentially agreed with that of Paula Couto’s (1967), except that he recognized only one valid species of Cuban Megalocnus. In 1978, Néstor A. Mayo (1978a) named a new genus and two new species of sloths based on three femora collected from two localities in Cuba: Habanocnus paulacoutoi and H. hoffstetteri. Reversing Paula Couto (1967), Mayo (1978b, 1980b) rehabilitated N. major and N. minor and proposed yet another species for inclusion in this genus (N. baireiensis; Mayo, 1980a). Because of the large number of named forms from Cuba, it might be assumed that this island had a much greater diversity of Antillean sloths than did the other islands. New discoveries and new synonymies (see below) indicate that neighboring Hispaniola was also a center of sloth diversity in the Quaternary (cf. Woods, 1990). In the late 1970s and early 1980s, Charles A. Woods and collaborators from the Florida Museum of Natural History undertook several collecting trips to Haiti for the purpose of recovering Quaternary mammals. Their efforts were rewarded with the discovery of several rich sink-hole sites that yielded, in addition to other important finds (e.g., Woods, 1989), numerous remains of sloths (including several associated skeletons). This material was sorted preliminarily by Margaret A. Langworthy, but for various reasons no analyses were completed or published. As a preliminary to this chapter and various other studies now in progress,

The Sloths of the West Indies: A Systematic and Phylogenetic Review

205

the new taxa represented in the Haitian material have been named and briefly described by MacPhee et al. (2000b). Detailed description of this material, however, is reserved for a future publication. In the 1990s, R. D. E. MacPhee and Manuel A. Iturralde-Vinent (1994, 1995, in press) recovered sloth remains from Tertiary localities in Cuba and Puerto Rico. The Cuban material, all of which comes from the Early Miocene locality of Domo de Zaza, has been referred to Imagocnus zazae, a species that is in some (possibly size-related) ways strongly reminiscent of Pleistocene Megalocnus and Parocnus (= Mesocnus). The Puerto Rican find, consisting of a single proximal femur from a locality near Yauco, is not diagnostic as to family affiliation. However, parsimony considerations suggest that it is probably a megalonychid, inasmuch as no other phyllophagan family had Antillean representation as far as is now known. The Yauco locality has yielded invertebrates consistent with an early Oligocene age (MacPhee and Iturralde-Vinent, 1995; Iturralde-Vinent and MacPhee, 1999). Discoveries of apparently new sloths in western Cuba (O. Arredondo and O. Vasquez Jimenez, personal communication) have been briefly announced but not yet reported in the literature. The newest Antillean territory in which sloth fossils have been discovered is Grenada. The fossils, consisting of isolated phyllophagan teeth from the Locality 12° North, a Plio-Pleistocene lahar deposit on the southern end of the island, are described in a paper by R. D. E. MacPhee and collaborators (2000a). Although the limited material compares reasonably well in size with Paulocnus petrifactus from Curaçao, distinctive features are few and the teeth are currently allocated only to Megalonychidae, gen. et sp. indet. The importance of the Grenadian faunule (which also includes a new capybara) is that it underlines the incredibly wide distribution of Megalonychidae in the insular Neotropics during the latter part of the Cenozoic.

HIGHER-LEVEL RELATIONSHIPS In the last 15 years a number of differing taxonomic arrangements of megalonychids within Phyllophaga have been proposed, but there is no clear consensus in sight (e.g., Engelmann, 1985; Mones, 1986; Arredondo, 1988; Pascual et al., 1990; White, 1993a, 1993b; Gaudin, 1995; McKenna and Bell, 1997). Inasmuch as a serious revision of the higher-level phylogeny of megalonychids implies a revision of all of Phyllophaga, in this section we can only touch on leading issues and interpretations. Among phyllophagan specialists there is no disagreement that Antillean sloths are indisputably members of Megalonychidae, exhibiting such defining features (where known) as dental formula 5/4; premaxilla, if present (Hapalops), does not bear teeth; molars quadrangular or elliptical; diastema long; first maxillary tooth usually caniniform; mandibular symphyseal region usually elongated; limbs relatively gracile and pentadactyl; calcaneal tuberosity mediolaterally expanded; and femoral third trochanter present (Paula Couto, 1979; De Muizon and McDonald, 1995). However, disagreement is otherwise rife concerning family content and interrelationships. Formerly, the subfamily Choloepodinae (nec Choloepinae), originally proposed as a tribe by Gray (1871) for reception of the extant two-toed sloth Choloepus, was often placed in Bradypodidae. However, current taxonomic practice restricts Bradypodidae to Bradypus and places Choloepodinae (or taxon of similar rank) in Megalonychidae (Scillato-Yané, 1980; Webb, 1985; Webb and Perrigo, 1985; Wetzel, 1985; White, 1993b; Gaudin, 1995; McKenna and Bell, 1997; but see Engelmann, 1985). In addition to Choloepodinae, six other nominal subfamilies of Megalonychidae are in varying degrees of systematic usage at present: Ortotheriinae, Megalocninae, Xenocninae, Nothrotheriinae, Megalonychinae, and Ocnopodinae (Mones, 1986). The Antillean diversity has consistently been placed in two subfamilies (usually, Ortotheriinae and Megalocninae or variants thereof), suggesting a diphyletic origin for West Indian sloths. However, the content and internal organization of all megalonychid groupings vary significantly from author to author. At one extreme, Engelmann (1985:60) restricts Megalonychidae to the Antillean taxa and several genera known from North America (e.g., Megalonyx), but includes no strictly South American

206

Biogeography of the West Indies: Patterns and Perspectives

forms on the ground that they “do not exhibit those features that unite the North American megalonychids.” Carried to its logical conclusion, this position would require that all Antillean sloths derive from a proximate North American ancestry, which is at odds with the emerging paleogeographical picture as well as the reconstructed faunal history of virtually all other Antillean land mammals (cf. Iturralde-Vinent and MacPhee, 1999). At the other extreme, various authors (e.g., Kraglievich, 1923; Patterson and Pascual, 1968; Scillato-Yané, 1977, 1979; Pascual et al., 1990) have claimed that, not only are Antillean sloths South American by origin, but that the cladogeny of specific terminal branches can actually be traced back to specific Miocene taxa from the southern end of the continent (e.g., Late Miocene Amphiocnus seneum, favorably viewed as a close ancestor or sister-taxon of Megalocnus by Kraglievich, 1923, and more recently by Pascual et al., 1990). If carried to its logical conclusion, this interpretation implies that the lower-level clades known from the Antillean Quaternary must have already differentiated in South America before their propagules reached the West Indies. On its face this scenario, requiring several separate invasions of phyllophagans, is less parsimonious than one requiring only one initiator (cf. Iturralde-Vinent and MacPhee, 1999). However, actual testing of these hypotheses using real character evidence (White, 1993a, 1993b) indicates that more than one colonization apparently occurred (see Cladistic Analysis: Results). Simpson (1945) grudgingly adopted Kraglievich’s (1923) subfamilial allocation of Acratocnus and Parocnus to Ortotheriinae, and Megalocnus and Microcnus to Megalocninae, both of which contain South American genera. If interpreted cladistically, this classification would certify a diphyletic origin for Antillean sloths, a possibility that has recently been endorsed on quite different grounds (and in a different arrangement) by White (1993a, 1993b). Engelmann (1978, l985) and Webb (1985) have provided some information useful for phylogeny reconstruction, but their cladograms are not rich in characters. Webb’s (1985) cladogram shows the Greater Antillean genera as comprising a single group, but the node bearing the terminal taxa has no supporting characters. Paulocnus is included with a separate group of more derived South, Central, and North American taxa on the basis of two characters (deepened parabolic symphysis, caniniforms aligned with molariforms), but its position within this clade is not clarified. Engelmann (1985) considers Acratocnus to be the sister of the other Antillean genera on the ground that these latter possess a greater flexure of the basicranialbasifacial angle, and a “characteristic meniscoid cross-section of the lower caniniforms.” Both of these authors place Parocnus next to Megalocnus rather than Acratocnus.

CLADISTIC ANALYSIS DATA SET We developed a set of 75 characters to explore phylogenetic relationships among megalonychids (see Appendix I for characters and character states; Appendix II for taxon set). We used our own observations as well as characters from the literature; all characters derived from the literature were thoroughly checked against real specimens. Gaudin (1995) recently collected data on 85 basicranial characters of Phyllophaga. However, we were able to use only nine of these characters (eight of which provided informative results). Gaudin’s other characters were not included, either because they applied to taxa we did not study, or because we lacked appropriate material, or because Gaudin’s definitions of the morphological underpinnings (and therefore the homology) of certain characters differed from ours. Six characters were uninformative because they were constant in the ingroups surveyed; these were excluded from analysis. Interpretation and evaluation of the cladistic results are based on the remaining 69 parsimony-informative characters. We tried to confine ourselves to those continuous characters that displayed gaps in their distributions among taxa. However, as is common in fossil investigations, we often had to work with small sample sizes or fragmentary material. Therefore, it is possible that some of the gap boundaries that we have specified will have to be revised after

The Sloths of the West Indies: A Systematic and Phylogenetic Review

207

A

Bradypus tridactylus Choloepus didactylus Acratocnus odontrigonus Acratocnus antillensis Acratocnus ye Paulocnus petrifactus† Neocnus gliriformis Neocnus toupiti Neocnus dousman Neocnus comes Neocnus major Hapalops longiceps Paramylodon harlani Megalocnus rodens Megalocnus zile Parocnus browni Parocnus serus Dasypus novemcinctus Tamandua tetradactyla

B

Bradypus tridactylus Choloepus didactylus Acratocnus odontrigonus Acratocnus antillensis Acratocnus ye Paulocnus petrifactus† Neocnus gliriformis Neocnus toupiti Neocnus dousman Neocnus comes Neocnus major Hapalops longiceps Paramylodon harlani Megalocnus rodens Megalocnus zile Parocnus browni Parocnus serus Dasypus novemcinctus Tamandua tetradactyla

C

Bradypus tridactylus Choloepus didactylus Acratocnus odontrigonus Acratocnus antillensis Acratocnus ye Neocnus gliriformis Neocnus toupiti Neocnus dousman Neocnus comes Neocnus major Paulocnus petrifactus† Hapalops longiceps Paramylodon harlani Megalocnus rodens Megalocnus zile Parocnus browni Parocnus serus Dasypus novemcinctus Tamandua tetradactyla

D

Bradypus tridactylus Choloepus didactylus Acratocnus odontrigonus Acratocnus antillensis Acratocnus ye Neocnus gliriformis Neocnus toupiti Neocnus dousman Neocnus comes Neocnus major Hapalops longiceps Paramylodon harlani Megalocnus rodens Megalocnus zile Parocnus browni Parocnus serus Dasypus novemcinctus Tamandua tetradactyla

FIGURE 2 (A–C) The three most-parsimonious trees resulting from a branch-and-bound search on a data matrix of 69 characters and 19 taxa (length = 225, CI = 0.613, HI = 0.413, RI = 0.682, RC = 0.419); trees are identical except for position of Paulocnus petrifactus (taxon marked by †). (D) Single most-parsimonious tree resulting from a branch-and-bound search on same data matrix, but with Paulocnus omitted (length = 222, CI = 0.622, HI = 0.405, RI = 0.689, RC = 0.428). See text for discussion.

discovery of additional fossil material. All characters were weighted equally and were treated as unordered. Missing characters were scored as question marks and polymorphic characters were scored using the CS-&-CS convention. Phylogenetic analysis was performed using the program PAUP, version 4.0b2 (Swofford, 1998), by applying a branch-and-bound search. Tamandua and Dasypus were designated as outgroup taxa, and trees were rooted at a basal node with a basal polytomy.

RESULTS The branch-and-bound search yielded three most-parsimonious trees (MPTs) (Figures 2A–C) with tree lengths of 225 steps (CI = 0.613; HI = 0.413; RI = 0.682; RC = 0.419). Variation in the three MPTs was entirely due to the variable placement of Paulocnus, and a strict consensus tree is shown

208

Biogeography of the West Indies: Patterns and Perspectives

Bradypus tridactylus Choloepus didactylus

D

Acratocnus odontrigonus

B

Acratocnus antillensis

Paulocnus petrifactus

C

Neocnus gliriformis

A

Neocnus toupiti

Choloepodinae

Acratocnus ye

Neocnus dousman

E

Neocnus comes Neocnus major Hapalops longiceps Paramylodon harlani

Megalocnus zile

G Parocnus browni Parocnus serus

Megalocninae

Megalocnus rodens

F

Tamandua tetradactyla Dasypus novemcinctus FIGURE 3 Strict consensus of the three most-parsimonious trees depicted in Figure 2A–D. Lettered portions of the branching sequence are discussed in text. Terminal taxa included in boxes comprise all valid Antillean sloth species discussed in this chapter (plus Choloepus didactylus, one of two Recent species in this genus which together form the mainland sister-taxon of Acratocnus).

in Figure 3. When Paulocnus (which is missing 51 of the 69 informative characters) is removed, only one MPT (Figure 2D) emerges at 222 steps (CI = 0.622, HI = 0.405, RI = 0.689, RC = 0.428). In all MPTs, taxa resolve into two major groupings. One includes ((Megalocnus, Parocnus) Paramylodon). All remaining taxa fall into the second major clade, which includes ((Choloepus, Acratocnus) (Neocnus)), with Paulocnus being variably placed as sister group to (Choloepus, Acratocnus), Neocnus, or both. The sister group of this assemblage is Bradypus, followed by Hapalops. The data set obviously contains a substantial amount of homoplasy. Nevertheless, the support for many of the groupings is impressive. In the following discussion, important nodes (labeled in Figure 3) are presented along with the number of characters that define them. Monophyly of sloths is well supported and assumed here, so we do not discuss characters supporting the ingroup. Total numbers of characters defining nodes are presented for both ACCTRAN (A) and DELTRAN (D) optimization methods (Maddison and Maddison, 1992). We also note how many characters are unambiguously placed, and how many of those are unique. Identities of unambiguously placed characters and their state transformations are presented in Table 1.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

209

TABLE 1 Character State Transformations of Unambiguously Placed Characters at Nodes Labeled in Figure 3 Node A Node B Node C a Node D Node E Node F Node G

21 (2→0) 25 (1→0) 3 (0→1) 18 (0→1) 6 (0→1) 2 (1→0) 5 (0→1)

31 (0→1) 29 (0→1) 4 (0→1) 40 (0→1) 7 (0→1) 27 (2→1) 8 (1→0)

36 (0→1) 33 (0→1) 6 (3→0) 69 (0→1) 9 (0→1) 44 (1→2) 13 (0→1)

53 (0→1) 38 (0→1) 9 (3→0)

56 (0→1) 40 (2→0) 12 (0→1)

74 (0→1) 47 (0→1) 17 (1→0)

26 (0→1) 51 (0→2) 16 (0→1)

53 (1→0) 57 (1→0) 17 (1→0)

18 (0→1)

70 (0→1)

72 (2→1)

75 (0→2)

Note: Identities of characters and character states are listed in Appendix I. Character states in boldface are unique states. a Characters 6, 9, 17: unambiguous when Paulocnus included; characters 3, 4, 12, 17, 70, 72, 75: unambiguous when Paulocnus excluded.

Node A — This node embraces what we consider to be the Choloepodinae, plus Bradypus and Hapalops. It is defined by 7(A)/6(D) characters (six unambiguous, two unique). Node B — This node includes the Choloepodinae, plus Bradypus. It is supported by 13(A)/6(D) characters (six unambiguous, two unique). Node C — This node includes Acratocnus, Choloepus, Neocnus, and Paulocnus, considered to form Choloepodinae in our classification. It is defined by 10(A)/11(D) characters (three unambiguous). If Paulocnus is excluded from the analysis, the number of unambiguous characters increases to seven (one unique). Node D — This node supports Acratocnini + Choloepus, and is supported by 6(A)/5(D) characters (three unambiguous, one unique). Node E — Cubanocnini is defined by 10(A)/9(D) characters (five unambiguous, one unique). Node F — This node combines the Megalocninae and Paramylodon, and is defined by 18(A)/8(D) characters (five unambiguous, one unique). Node G — Megalocninae is defined by 9(A)/13(D) characters (six unambiguous, one unique). Two aspects of our topology require special mention: the position of the extant three-toed sloth Bradypus, and the deep division within the Antillean megalonychid group. These results raise phylogenetic and biogeographical issues that will be discussed more extensively elsewhere, in conjunction with a detailed character analysis that is beyond the scope of this chapter. Originally, Bradypus was selected as an outgroup, but specifying it as such resulted in a nonmonophyletic ingroup in all of the most-parsimonious trees. In the 18 trees that are one step longer (length 226), Bradypus occupies the same position within the ingroup. This result was surprising, in light of the accumulating evidence that the tree sloths are diphyletic (Webb, 1985; Naples, 1987; White, 1993a, 1993b; Gaudin, 1995), and Gaudin’s (1995) suggestion that Bradypus is sister taxon to all other sloths. As a firmly embedded member of the ingroup, Bradypus associates closely with the choloepodines, which includes the extant Choloepus. Since many of the characters utilized in this investigation are postcranial — which is unusual for a study of sloth phylogeny — we may be sampling convergences related to locomotion. Indeed, the six unambiguous characters that support this node are all postcranial, and may be functionally related to arboreality. However, Bradypus is not the sister taxon to Choloepus (the only other fully suspensory taxon). Instead, Choloepus is more closely related to Acratocnus, Neocnus, and Paulocnus, none of which displays the extreme degree of suspensory adaptation seen in the two extant tree sloths. Plus, node C (our Choloepodinae) is supported by numerous unambiguous characters that are all cranial in nature and clearly unrelated

210

Biogeography of the West Indies: Patterns and Perspectives

to locomotor function. Therefore, we are confident in the unity of the Choloepodinae, whereas the position of Bradypus as a close relative to that clade may be driven by functional convergence to some extent. Similarly, the Megalocninae (node G) is well supported by largely cranial characters unrelated to locomotor function. In contrast, node F (uniting the mylodontid sloth Paramylodon with the Megalocninae) is largely supported by postcranial characters related to large size and terrestriality, and therefore seems to be driven partly by functional convergence. It is likely that the addition of more sloth taxa and more characters would help to resolve these issues. The wide separation between the two major Antillean subgroups Choloepodinae (node C) and Megalocninae (node G) is another result that warrants further study. Both of these clades are well supported by numerous strong characters, postcranial as well as cranial. However, in the present analysis each of them is associated with a different nonmegalonychid sister-group (Hapalops and Paramylodon, respectively). In McKenna and Bell’s (1997) classification, Hapalops is listed as a schismotheriine megatheriid and Paramylodon as a lestodontine mylodontid. No doubt these associations would be substantially modified if many more South American taxa were added to the matrix, but the basic point seems clear enough: a diphyletic origin of the Antillean sloths implies at least two separate invasions of the Antillean regions by phyllophagans. Previous cladistic analyses (White, 1993a, 1993b) have supported the same conclusion. Whether these findings will eventually require a restructuring of the classic family Megalonychidae remains to be seen; in any case, this is not a job that we can undertake here. Notwithstanding the complications presented by the placement of Bradypus and the division within the Megalonychidae, the distal branches are secure in their membership (nodes C, D, E, and G) and well supported. Variation in major branching patterns among trees that are one or two steps longer than the MPTs, or among trees with some taxa deleted, usually indicates that results are not very robust and are easily affected by the composition of the data set. However, in the present case the topology of more distal branches is very consistent across trees, suggesting that these have been characterized accurately. This gives us some confidence that the classification (Table 2) adopted here on the basis of these results is reasonably natural.

SYSTEMATICS As a basis for this presentation, we utilize McKenna and Bell’s classification for family-level (and higher) names. For a recent discussion of the higher-level affinities of Megalonychidae (which will not be discussed here), see Gaudin (1995). Below the family level, taxonomic organization and diagnoses mostly comply with our cladistic results (q.v.). In general, diagnoses for taxa at the same hierarchical level are differential, and should therefore be read in conjunction with one another. Magnorder XENARTHRA Cope, 1889 Order PILOSA Flower, 1883 Suborder PHYLLOPHAGA Owen, 1842 Superfamily MEGATHERIOIDEA Gray, 1821 Family MEGALONYCHIDAE Gervais, 1855 Since we have covered only a fraction of the taxa included in Megalonychidae by most workers, we will not take up the issue of defining this family or differentially diagnosing it from other sloth families (see Gaudin, 1995). However, it is important to repeat in this context that our cladistic results indicate that Antillean megalonychids are deeply cleaved into megalocnin and nonmegalocnin lineages — so deeply, in fact, that it seems that Antillean sloths had a diphyletic origin, arising from quite different progenitors. In the absence of an equivalently deep fossil record on the islands

The Sloths of the West Indies: A Systematic and Phylogenetic Review

211

TABLE 2 Classification of Antillean Megalonychid Slothsa,b Magnorder XENARTHRA Cope, 1889 Order PILOSA Flower, 1883 Suborder PHYLLOPHAGA Owen, 1842 Infraorder MEGATHERIA McKenna and Bell, 1997 Superfamily MEGATHERIOIDEA Gray, 1821c Family MEGALONYCHIDAE Gervais, 1855:44d [= “Famille des Mégalonycidés” Gervais, 1855:44] Megalonychidae incertae sedis Imagocnus MacPhee and Iturralde-Vinent, 1994:3e Imagocus zazae MacPhee and Iturralde-Vinent, 1994:3; Early Miocene; Cuba; Zaza sloth Gen. et sp. indet. Species Af; Early Oligocene; Puerto Rico Species B; Quaternary; Cuba Species C; Quaternary; Cuba Species Dg; ?Late Pliocene and/or Quaternary; Grenada Subfamily Choloepodinae Gray, 1871:430 [= Choloepina Gray, 1871:430; Choloepodinae Gill, 1874:24; nec Choloepidae (see Honacki et al., 1982); nec Choloepinae (see Gardner, 1993:63)h]; Quaternary; Central America, South America; Puerto Rico, Cuba, Hispaniola Tribe indet. Paulocnus Hooijer, 1962:47 Paulocnus petrifactus Hooijer, 1962:47; Quaternary; Curaçao; Curaçao sloth Tribe Choloepodini Gray, 1871:430 [= Choloepina Gray, 1871:430]; Quaternary; Central America, South America Choloepus Illiger, 1811:108; two-toed sloths, unaus Choloepus didactylus Linnaeus, 1758:35; Quaternary; South America Choloepus hoffmanni Peters, 1858:128; Quaternary; Central America, South America Tribe Acratocnini Varona, 1974:49; Quaternary; Puerto Rico, Cuba, Hispaniola Acratocnus Anthony, 1916:195; acratocnuses [= Miocnus Matthew, 1931:3; Habanocnus Mayo, 1978a:688] Acratocnus odontrigonus Anthony, 1916:195 [= Acratocnus major Anthony, 1918:412]; Quaternary; Puerto Rico; Puerto Rican acratocnus Acratocnus antillensis Matthew, 1931:4, new combination [= Miocnus antillensis Matthew, 1931:4 (introduced as nomen nudum by de la Torre and Matthew, 1915); Habanocnus hoffstetteri Mayo, 1978a:688; Habanocnus paulacoutoi Mayo, 1978a:689]; Quaternary; Cuba; Cuban acratocnus Acratocnus ye MacPhee, White, and Woods, 2000:11; Quaternary; Hispaniola; yesterday’s acratocnus Tribe Cubanocnini Varona, 1974:48 [= Neocnini Paula Couto, 1979:193]; Quaternary; Cuba, Hispaniola Neocnus Arredondo, 1961:29; neocnuses [= Cubanocnus Kretzoi, 1968:163] Neocnus gliriformis Matthew, 1931:4 [= Microcnus gliriformis Matthew, 1931:4; Cubanocnus gliriformis Kretzoi, 1968:163 (nec Microcnus Reichenow 1877, a subgenus of bird)]; Quaternary; Cuba; small Cuban neocnus Neocnus major Arredondo, 1961:32 [= Neocnus minor Arredondo, 1961:33; Neocnus baireiensis Mayo, 1980a:224]; Quaternary; Cuba; large Cuban neocnus Neocnus comes Miller, 1929:26, new synonymy [= “Acratocnus (?)” comes Miller, 1929:26; Synocnus comes Paula Couto, 1967:36]; Quaternary; Hispaniola; Miller’s Hispaniolan neocnus Neocnus dousman MacPhee, White, and Woods, 2000:13; Quaternary; Hispaniola; slow neocnus Neocnus toupiti MacPhee, White, and Woods, 2000:15; Quaternary; Hispaniola; least neocnus

212

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 (continued) Classification of Antillean Megalonychid Sloths Subfamily Megalocninae Kraglievich, 1923:54 [= Megalocni McKenna and Bell, 1997:101] Tribe Megalocnini Kraglievich, 1923:54; Quaternary; Cuba, Hispaniola Megalocnus Leidy, 1868:180; megalocnuses [= Myomorphus Pomel, 1868:665; Neomesocnus Arredondo, 1961:21] Megalocnus rodens Leidy, 1868:180 [= Myomorphus cubensis Pomel, 1868:665; Megalocnus rodens rodens Leidy, 1868:179; Megalocnus rodens casimbae Matthew, 1959 in Matthew and Paula Couto, 1959:27; Megalocnus ursulus Matthew, 1959 in Matthew and Paula Couto, 1959:30; Megalocnus junius Matthew, 1959 in Matthew and Paula Couto, 1959:30; Megalocnus intermedius Mayo, 1969:15; Neomesocnus brevirrostris Arredondo, 1961:21]; Quaternary; Cuba; Cuban megalocnus Megalocnus zile MacPhee, White, and Woods, 2000:7 [= Megalocuus? [lapsus calami] sp.? in parte, Miller, 1922:6]; Quaternary; Hispaniola; Hispaniolan megalocnus Tribe Mesocnini Varona, 1974:46; Quaternary; Cuba, Hispaniola Parocnus Miller, 1929:28, new synonymy; parocnuses [= “Megalocuus?” [lapsus calami], in parte Miller, 1922:6; Mesocnus Matthew, 1931:2] Parocnus serus Miller, 1929:29 [= “Megalocuus? [lapsus calami] sp?”, in parte Miller, 1922:6]; Quaternary; Hispaniola; Hispaniolan parocnus Parocnus browni Matthew, 1931:2 [= Mesocnus browni Matthew, 1931:2; Mesocnus torrei Matthew, 1931:2; Mesocnus herrerai Arredondo, 1977:2]; Quaternary; Cuba; Cuban parocnus a

For fuller synonymies for certain taxa, see McKenna and Bell (1997) and Gardner (1993). With the exception of the living two-toed sloth, Choloepus, sloth taxa lacking representation on islands in the Caribbean Basin are not included in this list. Bradypus, the three-toed sloth or ai, is relegated to its own (nominally megatherian) superfamily Bradypodoidea in most recent classifications. There are three extant species (Gardner, 1993): B. tridactylus (S. A. only), B. torquatus (S. A. only), and B. variegatus (also Central America). c See Gaudin (1995) and Greenwood et al. (2001) for provisional evidence that Megalonychidae is more closely related to mylodontan rather than megatherian sloths. d Although authorship of Megalonychidae is often attributed to Ameghino (1889) (cf. Gardner, 1993), Gervais’ (1855:44) “Famille des Mégalonychidés” precedes it by many decades and is available under ICZN art. 11f (iii) (see McKenna and Bell, 1997). e Monotypic/monogeneric taxon (distribution of species/genus is therefore as for genus/tribe). f See MacPhee and Iturralde-Vinent (1995). g See MacPhee et al. (2000a). h Choloepod-, not Choloep-, is the proper combining form for family-group names based on genus Choloepus. b

that might be used to test this notion, we are uncertain whether diphyly is the best explanation of our results. However, this is the conclusion to which the data we have gathered have led us. We hope to have additional insights into this problem resulting from the larger investigation we are at present undertaking. Subfamily Choloepodinae Gray, 1871:430 Diagnosis — Cranium domed; first maxillary and mandibular tooth triangular in cross section; glenoid posterior shelf present and glenoid shelf mediolaterally wide; symphyseal spout present; rostrum flared; diastema long; plane of mandibular condyles located just dorsal to toothrow; coronoid superior to condyle; femoral head spherical and extends rostral to greater trochanter; fovea for ligamentum teres centric if present (i.e., located in middle of articular surface on femoral head); third trochanter of femur a distinct lateral crest; lesser trochanter conspicuous; tibial and fibular shafts bowed; tarsus alternate; astragalar neck long; fibular facet of astragalus concave; calcaneal tuberosity mediovolarly expanded; inferior scapular angle acute; deltoid and pectoral crests of

The Sloths of the West Indies: A Systematic and Phylogenetic Review

213

humerus non-confluent; ulna gracile and anteriorly bowed. (Extracted from Anthony, 1916, and Paula Couto, 1967, 1979, with additional observations by authors.) Comment — Placing Acratocnus (tribe Acratocnini) with Neocnus (tribe Cubanocnini) in the same subfamily is similar to concepts developed by several other workers (e.g., Ameghino, 1889; Varona, 1974; Paula Couto, 1979; Mones, 1986), but differs substantially from McKenna and Bell’s (1997) tribal and subtribal divisions of Megalonychidae. The third tribe that we recognize, Choloepodini (containing Choloepus only), is not discussed here because it has been thoroughly examined in the neontological literature (e.g., Wetzel, 1985). Note — In our cladistic analysis, Bradypus is the sister-group of Acratocnini + Cubanocnini (+ Paulocnus). This association is supported by six unambiguously placed characters. Such a high degree of allegedly homoplastic similarity is indeed remarkable. Clearly, the last word on the relationship of three-toed sloths to the megalonychids has not been written. Tribe Acratocnini Varona, 1974:49 Diagnosis — As for type genus (Acratocnus). Comment — McKenna and Bell’s (1997) subtribe Acratocnina contains Synocnus (i.e., Neocnus, in parte) and is therefore paraphyletic. Acratocnus Anthony, 1916:195 Diagnosis — Pterygoid inflation absent; cranium relatively tall and domed with prominent postorbital constriction, sagittal crest, pronounced rostral mediolateral flare, and moderate airorrhynchy; glenoid ventral to superficies meatus; first maxillary tooth spike shaped, trigonal, anteriorly projecting, and curved (i.e., caniniform); last maxillary molariform convex and narrowest lingually; first mandibular caniniform straight, trigonal, and lacking posterointernal groove; last mandibular molariform convex lingually; symphyseal spout pointed and short; femoral shaft cylindrical; quadriceps femoris tubercle of tibia a long scar; prominent rectus femoris tubercle on pelvis; tibial surface of astragalus parallel sided, not divided, and posteriorly squared; fibular facet on astragalus deeply concave and funnel shaped; calcaneal tuberosity volarly expanded; humeral head globular; radius with ovoid head and long, well-developed, abrupt pronator quadratus flange; anterior border of scapula rounded. (Extracted and amended from Varona, 1974; Paula Couto, 1979). Type species = A. odontrigonus Anthony, 1916: 195. Distribution — Puerto Rico, Cuba, Hispaniola. Synonyms — Miocnus Matthew, 1931; Habanocnus Mayo, 1978a. Comment — The nomen Acratocnus has been widely used as a shelter for taxa that are, in fact, not intimately related phylogenetically (e.g., Miller, 1929; Arredondo, 1961). As reorganized here, Acratocnus strictly defined had a multi-island distribution that included Puerto Rico, Cuba, and Hispaniola — i.e., each of the three Greater Antillean islands known to have had sloths. Acratocnus odontrigonus Anthony, 1916:195 Holotype and type locality — Rostrum retaining caniniform tooth (AMNH 14170) collected at Cueva de la Ceiba, near Utuado, Puerto Rico. Diagnosis — As for genus, but differs from other Acratocnus species in exhibiting the following combination of features: pronounced airorrhynchy; femoral head foveate and proximally inclined; femoral shaft straight and cylindrical; significant gap in acetabular rim; fibula with posteriorly projecting proximal articulation; humerus with double bicipital groove; humeral trochlea flares distally; entepicondylar foramen not visible posteriorly. (Extracted from Anthony, 1916, and Paula Couto, 1967, with additional observations by authors.) Distribution — Mainland Puerto Rico only (i.e., not yet known from Vieques or other nearby islands on Puerto Rico shelf). Synonyms — A. major Anthony, 1918.

214

Biogeography of the West Indies: Patterns and Perspectives

Comment — Aeratocnus major is based on a partial skeleton including an incomplete cranium and mandible (AMNH 17169), also from Utuado area but from a different cave site. Hypodigm consists of the holotype only, no other specimens having been attributed to it. In comparing the skeletons of the two nominal species, Anthony (1926) stated repeatedly that bones of A. major were larger than those of A. odontrigonus but were not substantially different in structure. Current consensus is that A. odontrigonus is the only valid Puerto Rican sloth species from the Quaternary (Paula Couto, 1967, 1979; Varona, 1974), and that AMNH 17169 simply represents a large, aged member of that species. Acratocnus antillensis (Matthew, 1931:4), new combination Holotype and type locality — Edentulous mandible (AMNH 16680) from “the Casimba in the Sierra Jatibonico,” Cuba (Matthew, 1931:4; see also Brief Overview of Megalonychid Discoveries). Diagnosis — Agrees with A. odontrigonus for characters that define the genus, but differs from other Acratocnus species in exhibiting the following combination of features: femoral head very large and globular, proximally or anteriorly inclined; fovea for ligamentum teres absent or anteriorly located; femoral shaft with torsion; third trochanter thick and padlike; fibula with posteriorly projecting proximal articulation; distinct ridge separating cuboid and sustentacular facets of calcaneus; humeral trochlea mediolaterally flat with no distal flare; bridge over entepicondylar foramen with distinct knob; entepicondylar foramen present but not visible posteriorly. Distribution — Cuba. Synonyms — Miocnus antillensis Matthew, 1931 [introduced as nomen nudum by de la Torre and Matthew, 1915]; Habanocnus hoffstetteri Mayo, 1978a; H. paulacoutoi Mayo, 1978a. Comment — In describing the short, robust mandible of A. antillensis, with its abbreviated symphyseal spout and triangular caniniforms, Matthew (1919) drew attention to evident similarities to Puerto Rican Acratocnus, and even alluded to the possibility that they were closely related. Paula Couto advocated synonymizing Miocnus and Acratocnus (see Hoffstetter, 1955), but preserved a species-level distinction for the Cuban material (as A. antillensis; Matthew and Paula Couto, 1959:41). This was the first more-or-less explicit recognition of the fact that closely related species, nomenclaturally brigaded within the genus Acratocnus, had been present on each of the northern Greater Antilles (A. odontrigonus, including A. major, on Puerto Rico; A. antillensis on Cuba; and “A.” comes on Hispaniola). However, after completing a more detailed study of the type mandible and examining an unprepared skull attributed to Miocnus in the private collection of Oscar Arredondo, Paula Couto eventually decided to reestablish Miocnus as distinct from Acratocnus (Paula Couto, 1967). Since then the status of Miocnus as a separate genus has been widely accepted (e.g., Varona, 1974; Paula Couto, 1979; Woods, 1990). Identified material that actually belongs to A. antillensis is scarce in collections; worse, many specimens have been misidentified. To the confusion of later workers, Paula Couto (1967) in his figure 21 labeled the antillensis holotype as that of Synocnus comes, while his figure 22, said to show the mandible of M. antillensis, actually depicts a mandible of Parocnus serus. Also, three femora (AMNH 49944, 49945; MCZ 4442) that Matthew and Paula Couto (1959) included in the antillensis hypodigm should rightly be assigned to Neocnus (see below and discussion by Mayo [1978b]). To further complicate the issue, a femur (AMNH 49919) that should be assigned to A. antillensis has been repeatedly identified as anything but this taxon. Although Matthew and Paula Couto (1959, plate 35) listed it as representing Mesocnus (i.e., Parocnus). Paula Couto (1967) later stated that it probably belonged to Miocnus (i.e., Acratocnus). In a subsequent reevaluation, Mayo (1978a) stated that this femur might instead belong to Habanocnus (i.e., Acratocnus). AMNH 49919 is here assigned — one hopes definitively — to A. antillensis. Our synonymization of Habanocnus requires some defense, as this genus has been accepted by some workers (e.g., McKenna and Bell, 1997). Mayo (1978a) named a new genus and two new

The Sloths of the West Indies: A Systematic and Phylogenetic Review

215

species based on three femora collected from two localities in Cuba. Habanocnus hoffstetteri (type species) was based on a complete femur (IGPACC Nr. 421-119) from Cueva de Paredones, Prov. de La Habana, Cuba. The only other material attributed to this species is a proximal femur from the same locality (IGPACC 421-120). Habanocnus paulacoutoi was erected for a single femur (IGPACC Nr. 417-4) from Pio Domingo, Prov. de Pinar del Río, Cuba. No other newly collected material has been assigned to this species (although, as discussed above, Mayo [1978a] suggested that AMNH 49919 might represent H. paulacoutoi rather than Acratocnus [= Miocnus]). Although he provided a detailed account of species differences between H. hoffstetteri and H. paulacoutoi (e.g., in size of head, presence of fovea, shape of third trochanter, width of patellar groove), the differences he cited appear to represent nothing more than intraspecific individual variation. Similarly, although Mayo, 1978a, listed several ways in which the Habanocnus femora differ from those of other Cuban sloths, in our view they do not differ, quantitatively or qualitatively, from Hispaniolan sloth femora that we have assigned to Acratocnus. In summary, the material assigned (and sometimes diversely reassigned) over the years to A. antillensis, M. antillensis, H. hoffstetteri, and H. paulacoutoi can be comfortably accommodated within the bounds of a single genus and species (which must be A. antillensis by operation of the rule of priority). There is no hypodigmatic material other than femora to support Habanocnus, and on close study these specimens cannot be separated from those assigned to A. antillensis. Furthermore, the Cuban species is notably similar to its Hispaniolan and Puerto Rican close relatives. Acratocnus ye MacPhee, White and Woods, 2000:11 Holotype and type locality — Cranium and associated mandible (UF 170533) from Trouing Vapè Durand, Plain Formon, Département du Sud, Haiti. Hypodigm also includes numerous postcranial elements found in several different caves. Diagnosis — Agrees with A. odontrigonus for characters that define the genus, but differs from other Acratocnus species in exhibiting the following combination of features: superior aspect of cranium extremely domed along sagittal crest, forming a significant angle with rostrum; postorbital constriction not extreme; palatine foramina consistently prominent and abundant; symphyseal spout laterally pinched, extremely ventrally pinched on either side of ventral keel, and projecting anteriorly at an angle significantly different from that of anterior border of mandible; rectus tubercle of pelvis very prominent and laterally projecting, creating a right angle in posterior view; acetabular rim nearly closed and pit partly or completely filled in; femoral neck anteriorly projecting; femoral head extremely large, globular, and afoveate; femoral shaft with great torsion and reduced third trochanter; tibial shaft with prominent anteromedial muscle scar; tibia with posterolaterally projecting proximal articulation; distinct ridge separating cuboid and sustentacular facets of calcaneus; calcaneal tuberosity waisted and relatively symmetrical; humeral head extremely large; humeral trochlea mediolaterally flat with no distal flare; bridge over entepicondylar foramen with distinct knob; entepicondylar foramen slightly visible posteriorly; pectoral crest markedly projecting medially; forelimb greatly elongated; ulnar shaft with prominent anterolateral ridge (see MacPhee et al., 2000b, for discussion of A. ye). Distribution — Hispaniola. Synonyms — None. Acratocnus ye is completely distinct from “Acratocnus” comes (Miller, 1929), which is properly a member of Neocnus (see below). The “Acratocnus” metapodial described by Hooijer and Ray (1964) is probably assignable to N. comes. Comment — Yesterday’s acratocnus is very similar to A. odontrigonus in nearly all morphological respects, indicating a very close relationship (White et al., 1996). It is also noteworthy that the large, afoveate femoral head, reduced third trochanter, and long, gracile femoral shaft, all features utilized by Mayo (1978a) to distinguish Habanocnus (see A. antillensis), also occur on A. ye.

216

Biogeography of the West Indies: Patterns and Perspectives

Choloepodinae incertae sedis Paulocnus Hooijer, 1962:47 Diagnosis — As for type species. Paulocnus petrifactus Hooijer, 1962:47 (type and only species) Holotype and type locality — Skull preserving one tooth (GUIA R8222) from Tafelberg Santa Barbara, eastern Curaçao. Diagnosis — Size intermediate between that of Megalocnus and Acratocnus; zygomatic arch open; sagittal crest absent; first maxillary tooth caniniform and triangular (see below); last upper molariform trigonally shaped; symphyseal spout spatulate; first mandibular molariform (i.e., m2) subquadrate with anteroposteriorly long inner face; manus generalized; ungual phalanges laterally compressed (Hooijer, 1962; Paula Couto, 1967). Paulocnus also differs from Acratocnus in the following respects: astragalar trochlea wedge-shaped with distinct separation between medial and lateral sides; quadriceps femoris tubercle reduced; calcaneal tuberosity not volarly expanded (Hooijer, 1962; J. White, personal observation). Distribution — Curaçao. Synonyms — None. Comment — Other material in the original hypodigm as described by Hooijer (1962) includes a radius, femur, tibia, fibula, calcaneus, astragalus, and several hand and foot bones, but no further material seems to have been collected (or, at least, none has been reported). Curiously, Paulocnus is not known from Venezuela or elsewhere on the continent, despite the fact that Curaçao (which lies immediately off the South American continental shelf ) would have been separated from the mainland by only a small stretch of water from mid-Wisconsinan through early Holocene time. Perhaps megalonychid remains of late Quaternary age recently collected by John Moody and Greg McDonald in a cave in the Perija Mountains (on border between Venezuela and Colombia) will shed some much-needed light on the fate of this sloth family in northern South America (G. McDonald, personal communication). Paulocnus has considerable biogeographical importance because it is related to sloths from islands lying deeper within the Caribbean. However, the affinities of Paulocnus within Megalonychidae have not been treated in any real depth. Hooijer (1962, 1964, 1967) insisted that Paulocnus differed from other Antillean forms more than they differed among themselves (surely an overstatement when viewing Megalocnus against Neocnus, for example), and he avoided placing it anywhere in particular within the family. However, no one has properly reexamined the rather questionable characters on which Hooijer’s claims of distinctiveness are based. We note that MacPhee et al. (2000a), in reinterpreting Hooijer’s (1964) photograph of the only snout region of Paulocnus, concluded that the maxillary caniniform of this species corresponded in detail to that of other nonmegalocnin Antillean megalonychids, contra Hooijer (1964). Without citing any character evidence, Webb and Perrigo (1985) placed Paulocnus outside their own (unnamed) grouping of Antillean sloths, as the sister of a clade otherwise composed of Pliometanastes, Meizonyx, and Megalonyx. In complete contrast, McKenna and Bell (1997) placed Paulocnus in the coordinate subfamily Ortotheriinae, which implies that the Curaçao sloth is distantly, rather than closely, related to megalonychines. The badly preserved elements comprising the hypodigm of Paulocnus admit no certainty about its relationships. In our view it is probably reasonably closely related to Acratocnus, although it shows some important distinctions (e.g., elongated spout). It is incidentally important to mention that Ortotheriinae has traditionally included Antillean representation since Kraglievich (1923) first proposed the idea. However, his concept has been trimmed down over the years. At present, the only Antillean taxa regarded as ortotheriine by McKenna and Bell (1997) are Paulocnus and Habanocnus. The former, as noted, may or may not be a relative of Acratocnus; the latter is certainly so. No ortotheres from southern South America

The Sloths of the West Indies: A Systematic and Phylogenetic Review

217

were included in our investigation, so the position of this subfamily with respect to our concept of Choloepodinae is uninvestigated. Tribe Cubanocnini Varona, 1974:48 Diagnosis — As for type genus (Neocnus). Comment — Although Neocnus Arredondo, 1961 has priority over Cubanocnus Kretzoi 1968, Varona’s (1974) tribe Cubanocnini (containing Neocnus only) is valid and available (ICZN art. 40[a][i]). As the first suprageneric name to be proposed for this clade (ICZN art. 24), Cubanocnini has priority over more logical constructions (e.g., Neocnini, proposed by Paula Couto [1979:193]), even though Cubanocnus is now reduced to the rank of junior synonym of Neocnus. Neocnus Arredondo, 1961:29 Diagnosis — Small body size; cranial flexion absent; postorbital constriction weak; rostral flare slight; pterygoid inflation absent; sutural attachment area for jugal on maxilla very small (?jugal reduced); cranial glenoid at or above superficies meatus; second maxillary molariform anterolaterally concave; last maxillary molariform broadest on lingual side; mandibular caniniform grooved posterointernally; last mandibular molariform with deep lingual groove; rectus femoris tubercle inconspicuous; femoral head fovea centric; femoral shaft anteroposteriorly flat and medially bowed with prong on anterior aspect; tibial articular surface with slight separation; astragalar head well defined; astragalar trochlea wedge shaped; fibular facet of astragalus truncated and crescent shaped; humeral head spherical and skewed laterally; entepicondylar foramen visible in posterior view; humerus short and slender but with prominent and squared supracondylar ridge; deltoid and pectoral crests nonconfluent; pronator quadratus flange distally confined. (Extracted from Paula Couto, 1979, Webb and Perrigo, 1985, and Gaudin, 1995, with additional observations by authors.) Type species = N. gliriformis (Matthew, 1931). Distribution — Cuba (N. major, N. gliriformis) and Hispaniola (N. comes, N. dousman, N. toupiti). Synonyms — Cubanocnus Kretzoi, 1968. Comment — The taxonomic history of this genus is convoluted. While Paula Couto (1967) and Fischer (1971) made a case for lumping Arredondo’s N. major and N. minor into N. gliriformis, Mayo (1978b, 1980b) summarized evidence suggesting that they should remain separate species. We provisionally recognize N. gliriformis (as originally conceived) as separate from N. major (including N. minor) because of the former’s extremely small size and gracility of obviously adult specimens. There seems to be no compelling reason to recognize any additional species of Neocnus in Cuba, however. Neocnus gliriformis Matthew, 1931:4 Holotype and type locality — Partial mandible (AMNH 16882) from “the Casimba in the Sierra Jatibonico,” Cuba (see Brief Overview of Megalonychid Discoveries). Diagnosis — As for genus, but differs from other Neocnus species in exhibiting the following combination of features: cranium flattened; palatine grooves absent; ventral aspect of mandibular body convex; mandibular second molariform subtriangular; mandibular fourth molariform flat posteriorly; femoral shaft with slight torsion and reduced anterior prong; quadriceps femoris tubercle reduced; supracondylar ridge of humerus reduced; pronator quadratus flange of radius gentle and reduced; sigmoid notch of ulna unsegmented and shallow. Distribution — Cuba. Synonyms — Microcnus gliriformis Matthew, 1931. Comment — Matthew’s original concept of Microcnus (de la Torre and Matthew, 1915; Matthew, 1918, 1919) was based on a small collection consisting of a mandible and some foot bones. These elements were considered by Matthew to be similar to, but smaller than, their counterparts in living tree sloths, and hence distinctive among megalonychids generally. When

218

Biogeography of the West Indies: Patterns and Perspectives

M. gliriformis was finally validly named (Matthew, 1931), lack of a symphyseal spout was given as its main distinguishing feature. Arredondo (1961) 30 years later created the new genus Neocnus for two new species, N. major and N. minor, noting that while similar to N. gliriformis in most respects, they differed in possessing a small but definite symphyseal spout. In Paula Couto’s (1967) opinion, the spout had been present in life on the N. gliriformis type, but was later lost. Arredondo’s distinction therefore lacked a difference, and Paula Couto sank major and minor into Microcnus gliriformis. However, we regard N. gliriformis as distinguishable from the other Cuban species we recognize, N. major, by the features listed below. Neocnus major Arredondo, 1961:32 Holotype and type locality — Arredondo (1961) did not specifically identify a holotype for this species, although his descriptions repeatedly emphasized that he was founding N. major on two left mandibles (which must therefore be regarded as syntypes): IGPACC 417-5 (formerly SEC P. 318) from Cueva de Pio Domingo, near Sumidero, Prov. de Pinar del Río, Cuba; and an unnumbered specimen from Cueva de Paredones, in Alquízar, La Habana. Mayo (1980b) in effect designated the Paredones specimen as the name-bearing type (i.e., lectotype) for N. major. (The Pio Domingo specimen thus becomes a paralectotype under ICZN art. 73-74.) Diagnosis — Agrees with N. gliriformis for characters that define the genus, but differs from other Neocnus species in exhibiting the following combination of features: cranium slightly domed; second mandibular tooth subquadrate; fourth mandibular tooth convex posteriorly; symphyseal spout pointed and short; femoral shaft with slight torsion and well-developed anterior prong; distal tibial articular surface divided; astragalar trochlea tapered posteriorly; pronator quadratus flange forming abrupt lateral crest. Further descriptions may be found in Arredondo (1961) and Mayo (1978b, 1980b). Distribution — Cuba. Synonyms — Neocnus minor Arredondo, 1961; N. baireiensis Mayo, 1980a. Comment — Specimens previously assigned to N. minor and N. baireiensis (the latter consisting of a single femur) appear to fall within the range of individual variation for N. major. It may be eventually warranted to recognize N. gliriformis as the only Cuban species of the genus, if additional material indicates that all relevant differentiae of N. major are simply clinal. Neocnus comes (Miller, 1929:26), new synonymy Holotype and type locality — Proximal femur (USNM 253178, renumbered USNM 299642) collected from a cave near St.-Michel-de-l’Atalaye [or Atalye], Haiti. For comment on hypodigm, see below. Diagnosis — Agrees with N. gliriformis for characters that define the genus, but differs from other Neocnus species in exhibiting the following combination of features: sagittal crest double; cranium slightly domed; second mandibular tooth subquadrate; symphyseal spout narrow; proximal facet of tibia laterally oriented; femoral shaft with well-developed anterior prong; quadriceps femoris tubercle forming hook, with associated groove; distal tibial articular surface divided; astragalar trochlea tapered posteriorly; pronator quadratus flange forming abrupt lateral crest; ulnar shaft straight with medially hooked olecranon. Distribution — Frequent in cave faunules in Haiti and Dominican Republic. Synonyms — “Acratocnus (?)” comes Miller, 1929; Synocnus comes Paula Couto, 1967. Comment — Miller (1929, 1930) assigned a variety of elements to the hypodigm of “Acratocnus (?)” comes. Paula Couto (1967) included additional material that he believed belonged to the same taxon (skull fragment [USNM 293837], anterior portion of a mandible [USNM 293836], edentulous ramus [MNHN 1881-28]). On the basis of the now-widened hypodigm, he went on to conclude that differences from Puerto Rican Acratocnus were sufficient to warrant the creation of a new genus, Synocnus, for the Haitian fossils.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

219

In reviewing this material ourselves, we have determined that Paula Couto’s (1967) hypodigm for S. comes includes more than one taxon. Although the skull and mandible unfortunately cannot now be located, Paula Couto’s (1967:40–46) description of them could well serve as a description of the homologous elements of Parocnus. In addition, the photographs in his figures 21 and 22 are reversed: the photograph presented as figure 22 (of Parocnus) depicts USNM 293836, which is actually described in the caption of figure 21 (which has instead an image of an A. antillensis mandible). Removing from consideration fossils that actually belong to Parocnus, Paula Couto’s (1967) original hypodigm is reduced to the specimens which were described as “Acratocnus?” by Miller (1929). With the much larger Hispaniolan sample now available, it is clear that Miller’s specimens correspond much more closely to Neocnus than to Acratocnus. We confirm that UF 25702, a skull from Trou Wòch Dadier assigned by Webb (1985) to “Synocnus” comes is indeed N. comes. In addition, the “Acratocnus” metapodial described by Hooijer and Ray (1964) is probably assignable to N. comes. Neocnus dousman MacPhee, White, and Woods, 2000:13 Holotype and type locality — Skull (UF 76363) from Trouing de la Scierie, Morne La Visite, Haiti. Hypodigm also includes mandible from the same locality, and postcranial elements from diverse localities in Haiti and Dominican Republic. Diagnosis — Agrees with N. gliriformis for characters that define the genus, but differs from other Neocnus species in exhibiting the following combination of features: sagittal crest consistently present; cranium flattened; lateral groove of pterygoid present; second mandibular tooth subtriangular; symphyseal spout long, narrow, and untapered; proximal fibular facet of tibia oval and posteriorly oriented; femoral shaft with slight anterior prong; quadriceps femoris tubercle forming hook with deep groove; pronator quadratus flange forming abrupt lateral crest; bicipital tuberosity anteriorly placed (see MacPhee et al., 2000b, for discussion of N. dousman). Distribution — Hispaniola. Synonyms — None. Comment — While there are discrete traits that define this new species as distinct from the larger N. comes and the smaller N. toupiti (q.v.), this species can be distinguished from its closest relatives morphometrically, thanks to the large sample sizes housed at UF. Neocnus toupiti MacPhee, White and Woods, 2000:15 Holotype and type locality — Skull (UF 156892) and various elements comprising an associated skeleton of a single individual (all from Trouing Jeremie #5, Plain Formon, Département du Sud, Haiti). Hypodigm also includes material from several other sites in Haiti. Diagnosis — Agrees with N. gliriformis for characters that define the genus, but differs from other Neocnus species in exhibiting the following combination of features: very small, extremely gracile postcranial skeleton; upper caniniform with very deep lingual groove; second mandibular tooth subtriangular; symphyseal spout long, narrow, and untapered (partly broken on holotype); angular process of mandible projecting far posteriorly; femur lacking third trochanter; femoral shaft cylindrical with reduced anterior prong; femoral head tiny; distal tibial articular surface very narrow and divided only at anterior edge; astragalus tiny with relatively long neck; ectal and sustentacular facets very close together; calcaneal tuberosity triangular; ectal facet distinctly humped; most anterior aspect of glenoid fossa of scapula pointed; supracondylar ridge reduced; pronator quadratus flange gentle and reduced; ulnar shaft extremely laterally compressed; sigmoid notch of ulna unsegmented and shallow; proximal fibular facet of tibia round, reduced, and posteriorly oriented (see MacPhee et al., 2000b, for discussion of N. toupiti). Distribution — Hispaniola. Synonyms — None.

220

Biogeography of the West Indies: Patterns and Perspectives

Comment — While morphologically very similar to N. gliriformis, this new species is much smaller — indeed, significantly smaller than both genera of extant tree sloths, and quite possibly the smallest sloth on record from any age or place. Neocnus toupiti can be distinguished from other Neocnus by a suite of unique discrete traits as well as by morphometric statistical analyses. Elements assigned to this species are clearly adult, and are smaller than obviously immature elements belonging to the larger Hispaniolan Neocnus species. We rule out sexual dimorphism as an explanation for the presence of multiple similar species of Neocnus on Hispaniola on two grounds: the three species can be distinguished by discrete morphological traits that are not associated with sexual dimorphism in other mammalian groups, and there is no evidence for sexual dimorphism in either extant sloths or in other megalonychid taxa for which there are large sample sizes. Subfamily Megalocninae Kraglievich, 1923:318 Diagnosis — Postorbital constriction absent; cranium long, of relatively uniform width, and flattened superiorly; jugal expanded; pterygoid inflation present; paroccipital process greatly enlarged and free standing (Gaudin, 1995); lateral groove of pterygoid absent (Gaudin, 1995); pronounced airorrhynchy (posterior palate flexed ventrally); mandibular coronoid process not superior to condyle, and condyles well above tooth row; last maxillary molariform medially narrow; last mandibular molariform convex; femur with nonspherical head and anteroposteriorly deep distal end; fovea for ligamentum teres posterior and eccentric (i.e., located on periphery of articular surface on femoral head); rectus femoris tubercle prominent; shaft of tibia and fibula straight; tarsus serially arranged; calcaneal tuberosity symmetrical and thick with lateral foramen and lacking volar expansion; astragalar neck very short; astragalar articular surface distinctly divided; fibular facet of astragalus flat; deltoid and pectoral crests of humerus confluent; ulnar shaft straight; coronoid process of ulna extensive and shelf-like. Type genus = Megalocnus Leidy, 1868. Comment — The traditional concept of this subfamily as defined by Kraglievich (1923) includes both Megalocnus and Parocnus (= Mesocnus; for synonymy and choice of valid name, see below) and has been supported by a number of authorities (e.g., Varona, 1974; Arredondo, 1977; Paula Couto, 1979; Mones, 1986; Pascual et al., 1990; McKenna and Bell, 1997). However, at least as many investigators have inserted Parocnus within the other major grouping of extinct megalonychids, Ortotheriinae (e.g., Simpson, 1945; Aguayo and Rivera, 1954; Hoffstetter, 1955; Matthew and Paula Couto, 1959; Arredondo, 1960; Paula Couto, 1967; Fischer, 1971). Cutting the knot, but not resolving any phylogenetic issues thereby, Arredondo (1988) erected the new subfamily Mesocninae exclusively for this genus. The possible association of Parocnus with Ortotheriinae sensu McKenna and Bell (1997) is, of course, not testable with our data because none of the members of this subfamily was included in our cladistic analysis (Acratocnus [= Habanocnus] and Paulocnus are not considered members by us). A considerably more involved problem, which cannot be discussed in detail here, is McKenna and Bell’s (1997) close association of Megalonyx and Megalocnus (to the exclusion of all other Antillean taxa) within the same subtribe (Megalonychina, within tribe Megalonychini). Parocnus is relegated to a separate subtribe in the same collocation. Although Megalonyx is certainly related to Megalocnus, and may even be the sister-group of a clade consisting of Antillean megalonychids plus Choloepus (cf. Gaudin, 1995), it seems highly improbable that the former is actually deeply embedded within the latter. At present, we know of no character evidence to support this relationship. Tribe Megalocnini Kraglievich, 1923:318 Diagnosis — As for type genus.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

221

Megalocnus Leidy, 1868:180 Diagnosis — Maxillary teeth pseudorodentiform or incisiform rather than caniniform (i.e., broad and anteroposteriorly compressed); femoral head not flat; acetabular rim with large gap; scapular spine divergent at vertebral border; prescapular fossa much larger than postscapular fossa; second scapular spine near inferior angle prominent; fossa for teres major expanded into a blade; anterior scapular border ventrally concave and smoothly curved; glenoid fossa greatly angled and anteroposteriorly concave. Type species = M. rodens Leidy, 1868. Distribution — Cuba (M. rodens) and Hispaniola (M. zile). Comment — Megalocnus, the first sloth taxon identified from the West Indies, is one of the most frequently recovered mammal fossils in Cuba. MacPhee et al. (2000b) have presented evidence that true Megalocnus occurred on Hispaniola and its satellite Ile de la Tortue as well. Megalocnus rodens Leidy 1868:180 Holotype and type locality — Partial mandible from Ciego Montero, Sierra de Jatibonico, Cuba, “presumably in the collections of the Madrid museum” (Paula Couto, 1967:9), catalog number unknown. Hypodigm includes numerous specimens listed by Matthew and Paula Couto (1959). Diagnosis — Symphyseal spout absent; mandibular incisiform meniscoid; femoral neck proximally oriented and above greater trochanter; femoral shaft anteroposteriorly flat and of relatively uniform width; third trochanter absent; lesser trochanter forming low, gentle bump; distal tibial articular surface with slight separation; proximal fibular facet of tibia posteriorly located; medial trochlea of astragalus short and odontoid; navicular facet concave; calcaneal tuberosity with very slight medial expansion; humerus with large entepicondylar foramen, not visible posteriorly; humeral head spherical; distal radial articular surface smooth. (Extracted from Matthew and Paula Couto, 1959, and Paula Couto, 1967, with additional observations by authors.) Distribution — Widespread in western and central Cuba during the Quaternary, possibly with many local populations. Synonyms — Myomorphus cubensis Pomel, 1868 (based on same material as Leidy’s, whose name has priority), Megalocnus rodens rodens Leidy, 1868, Megalocnus rodens casimbae Matthew, in Matthew and Paula Couto (1959), Megalocnus ursulus Matthew, in Matthew and Paula Couto (1959), Megalocnus junius Matthew, in Matthew and Paula Couto (1959), Megalocnus intermedius Mayo, 1969, Neomesocnus brevirrostris Arredondo, 1961. Comment — Matthew thought that several species of Megalocnus existed in Cuba during the Quaternary, and had developed manuscript names and at least partial diagnoses for them before his death. Paula Couto edited and published these in their joint paper (Matthew and Paula Couto, 1959), somewhat cumbersomely, as “Matthew in schedis.” As Paula Couto noted, he chose to underscore Matthew’s authorship of these names because he doubted their validity — even as he enshrined them in the literature by officially publishing them. To preserve everyone’s good name and actual intentions, the authorship of these taxa should be given as “Matthew, in Matthew and Paula Couto (1959).” None of Matthew’s Megalocnus taxa have much current support, and their details can be quickly summarized. Megalocnus rodens rodens is based on the type mandible for the species and genus. Megalocnus rodens casimbae was founded on a different mandible (AMNH 49987) from Casimba, but is not otherwise meaningfully distinguished. Megalocnus ursulus is based on an edentulous mandible fragment (AMNH 49996), also from Casimba, that is roughly two thirds the average mandible size of (adult) M. rodens. Paula Couto (1967) suggested that that mandible is probably that of a juvenile M. rodens, a view with which we concur. Megalocnus intermedius Mayo, 1969 is based on a nearly complete cranium (DPUH 1201) from Cueva del Vaho, Prov. de La Habana. Other material attributed by Mayo to this species includes several molariform teeth, a partial radius and ulna, and a partial skeleton from Caverna de Pio Domingo, Prov. de Pinar del Río. Mayo (1969) argued that these remains represented a new species because, compared to M. rodens, they are smaller and less robust, and because dental and diastema

222

Biogeography of the West Indies: Patterns and Perspectives

dimensions are slightly different. However, he also noted that morphometric differences in the limbs of the two species were negligible, suggesting to us that the distinctions he defined to separate the two species are not taxonomically meaningful. Neomesocnus brevirrostris Arredondo, 1961 consists of a mandibular fragment (symphyseal region and alveoli of two teeth) which was found at Cueva de Paredones, Prov. de La Habana, Cuba (currently identified as No. 51 in O. Arredondo’s private collection). Although Varona (1974) maintained N. brevirrostris as a separate genus and species, the mandibular fragment is generally thought to represent a juvenile individual of Megalocnus (Paula Couto, 1967, 1979; Mayo, 1969; Fischer, 1971). Arredondo (1988) himself noted that the distinctiveness of some of the mandible’s characters may be questioned, which seems to us to put an end to the matter. In summary, current consensus recognizes that M. rodens exhibits a substantial range of intraspecific morphometric variation (Fischer, 1971; Varona, 1974; Paula Couto, 1979), and that other named forms are invalid, as they are based on individuals displaying extremes of variation. The extensive range of intraspecific variation evident in other Antillean sloth taxa that are represented by large sample sizes supports this view (White, 1993a; White et al., 1996). Megalocnus zile MacPhee, White, and Woods, 2000:7 Holotype and type locality — Scapula (left side, UF 169930) from Trou Gallery, Ile de la Tortue, Département du Nord-Ouest, Haiti. Other material referred to this species includes several molariforms and postcranial elements from Trou Gallery. A scapular fragment from the Dominican Republic confirms that this species also lived on the mainland. Diagnosis — With respect to known elements, agrees with M. rodens, but can be distinguished from the latter in that the fossa for teres major on the caudal border of the scapula is more capacious and expands abruptly beneath secondary scapular spine. Also, the femoral head of M. zile is more spherical, nearly to the degree seen in the subfamily Choloepodinae (see MacPhee et al., 2000b, for discussion of M. zile). Distribution — Hispaniola. Synonyms — “Megalocuus? sp?” [lapsus calami], in parte (Miller, 1922). Comment — In his 1922 paper, Miller tentatively suggested that some fragmentary postcranials from a cave near St.-Michel-de-l’Atalaye (Haiti) could represent Megalocnus, but later decided that they represented a distinct form (eventually denominated Parocnus serus; see below). Nevertheless, Megalocnus did in fact exist in Hispaniola, as the newly described material from Trou Gallery makes clear (MacPhee et al., 2000b). The Trou Gallery specimens correspond in size and shape to their better-known counterparts in M. rodens. Unfortunately, although large molariforms have been recovered at Trou Gallery, no incisiforms are known. It is possible that the highly specialized front teeth of M. rodens were autapomorphic, but this possibility cannot be usefully evaluated in the absence of relevant remains of the Hispaniolan species. Why M. zile was apparently a much rarer animal than M. rodens is not known. Tribe Mesocnini Varona, 1974:46 Diagnosis — As for type genus (= Parocnus Miller, 1929:28). Comment — Varona (1974) proposed Mesocnini as a tribe containing Mesocnus, Neomesocnus, and Parocnus. Neomesocnus is a synonym of Megalocnus (see above). Mesocnus and Parocnus are so similar for all important diagnostic features (see below) that a generic separation between them can no longer be sustained. The appropriate genus-level name for the two species that we recognize (P. browni and P. serus) is Parocnus (Miller, 1929; see below); however, the name of the tribe remains Mesocnini (ICZN art. 24, principle of first reviser). Nearly as many workers have preferred to place Parocnus and its allies in Megalocninae as in Ortotheriinae. Our analyses support its placement in subfamily Megalocninae, together with Megalocnus.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

223

Parocnus (Miller, 1929:28), new synonymy Diagnosis — Sagittal crest double; symphyseal spout greatly elongated and spatulate; anterior maxillary teeth small and triangular; mandibular caniniform teeth with deep inner groove; molariform teeth subquadrate; mandibular mental foramina very large; femoral head medially oriented, nonspherical, and flattened; femoral shaft wide and anteroposteriorly compressed for the upper two thirds, and distally narrowed and rounded; greater trochanter level with head, with deep posteromedial concavity, and confluent with prominent third trochanter; lesser trochanter absent or inconspicuous; distal tibial articular surface with distinct separation; medial side of astragalar trochlea concave; calcaneal tuberosity ventrally excavated; scapula with rounded borders and oval-shaped glenoid fossa; pre- and postscapular fossae approximately equal in size; humeral head flattened; entepicondylar foramen of humerus absent; supracondylar ridge gentle; distal radial articular surface irregular. Type species = P. serus Miller, 1929: 29. Distribution — Cuba (P. browni) and Hispaniola (P. serus). Synonyms — “Megalocuus? sp?” [lapsus calami], in parte (Miller, 1922); Mesocnus Matthew, 1931. Comment — It might be remarked that, in justice, the genus-level name for the species accepted here should be Matthew’s Mesocnus. Unfortunately, Matthew used this name several times in print as a nomen nudum, without identifying a type species or providing a diagnosis. By the time Matthew validly published it (posthumously, in 1931), Miller (1929) had already named P. serus. Hard luck, but if browni and serus belong in the same genus, as we argue here, that genus has to be Parocnus by virtue of priority. Parocnus serus Miller, 1929:29 Holotype and type locality — Immature partial right femur, lacking epiphyses (USNM 253228), from St.-Michel-de-l’Atalaye, Haiti. Diagnosis — P. serus differs from P. browni in the following respects: acetabular rim lacks gap and is surmounted by prominent rectus femoris tubercle; navicular facet of astragalus convex; medial trochlea of astragalus less concave, less dorsally elevated; trochlear articular surface grooved and divided but continuous; ectal facet of calcaneus very concave; distal articular surface of radius irregular but single faceted. Distribution — Haiti and the Dominican Republic, including the islands of La Tortue and La Gonâve. Synonyms — Megalocuus? sp? (in parte) Miller, 1922. Comment — Miller (1929) referred a humerus, a proximal tibia and fibula, an astragalus, three calcanei, an atlas fragment, and several foot bones to the hypodigm of P. serus. Reviewing this material a quarter century later, Paula Couto, in a letter cited by Hoffstetter (1955), concluded that Miller’s Parocnus material actually represented Megalocnus (as M. serus). Matthew and Paula Couto (1959) listed part of Miller’s Parocnus material as a synonym of Megalocnus, but they also listed part of it as a synonym of Mesocnus. In Paula Couto’s (1967) later review, however, Parocnus was maintained as a separate genus, and he even added a mandibular ramus (USNM 293831) to the hypodigm. Interestingly, he speculated that, had Matthew’s nomen Mesocnus been properly published prior to 1929, Miller might have assigned his Haitian material to Mesocnus rather than coin a new genus for its reception. Varona (1974) recognized the extensive similarities between Parocnus and Mesocnus, but advocated maintaining generic distinction until more materials were found. Parocnus browni Matthew, 1931:2 Holotype and type locality — Anterior half of cranium (AMNH 16877) from Ciego Montero, Cuba. Diagnosis — Full descriptions of this material may be found in Matthew (1931), Matthew and Paula Couto (1959), and Paula Couto (1967). Features that differentiate P. browni from P. serus

224

Biogeography of the West Indies: Patterns and Perspectives

include: navicular facet of astragalus flat/concave; medial trochlea of astragalus very concave and dorsally elevated, with articular surface widely separated from that of lateral trochlea; humeral head very flattened; pronator quadratus flange of radius laterally very abrupt; distal articular surface of radius forming two distinct facets. Distribution — Various localities in Cuba, primarily Ciego Montero, Casimba (Sierra de Jatibonico), and Cueva de los Niños (Cayo Salinas). Synonyms — Mesocnus browni Matthew, 1931; M. torrei Matthew, 1931; M. herrerai Arredondo, 1977. Comment — Although M. browni and M. torrei have frequently been accepted as valid species (Varona, 1974; Paula Couto, 1979; Arredondo, 1988), Paula Couto (1967) noted that such differences as there are between them relate to size rather than morphology. In particular, he noted that M. torrei may in fact represent immature or female members of M. browni. Likewise, Fischer (1971) felt that there was not enough material available to allow the species to be definitively distinguished: their size ranges overlap, and there are no consistent discrete traits that distinguish the two. Mesocnus herrerai Arredondo, 1977 is based on a mandibular ramus from Cueva Funeraria de los Niños (IZ 6995). Arredondo noted that the mandible differs from those of browni and torrei in being narrow and in lacking ventral convexity, but it is difficult to accept the validity of this species because it is represented by so little material. Although in our classification the type of herrerai is regarded as a synonym of Parocnus browni, it is worth noting that the herrerai mandible markedly resembles that of Hispaniolan P. serus. Arredondo (1977), who noted this resemblance originally, actually speculated that Mesocnus and Parocnus may turn out to be synonymous — a prescient thought that we have now taken to its logical conclusion. Megalonychidae, incertae sedis It is convenient to list several sets of sloth remains as Megalonychidae, incertae sedis. Only one has a formal binomen at present (Imagocnus zazae). Some of the others probably represent new species and even genera, but which of them rate this distinction will have to await proper study and evaluation by the investigators concerned. Imagocnus MacPhee and Iturralde-Vinent, 1994:3 Diagnosis — As for type species. Imagocnus zazae MacPhee and Iturralde-Vinent, 1994:3 (type and only species) Holotype and type locality — Edentulous palate (MNHNH P 3014); Domo de Zaza, Prov. Sancti Spiritus, Cuba. Diagnosis — Distinctively megalonychid organization of alveoli; differs from all Antillean sloths (except Acratocnus and Parocnus) in lacking large palatal palatine foramina situated at transverse level of first molariforms. Differs from Acratocnus in being larger, in possessing a midline torus, and other features. Differs from Parocnus in having greater interalveolar breadth at level of first molariform. Distribution — Central Cuba, Early Miocene. Synonyms — None. Comment — Despite many seasons of work at Domo de Zaza, only a small number of sloth fossils have been recovered at this locality (MacPhee and Iturralde-Vinent, in press). The material now available includes a partial pelvis that is larger than any known pelvic specimen attributable to Megalocnus rodens, usually regarded as the largest known Antillean megalonychid. However, whether the size of the Zaza pelvis implies that there was another (larger) species in Cuba during the Miocene or that body size in I. zazae was simply highly variable cannot be usefully assessed at this time.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

225

MacPhee and Iturralde-Vinent (1994) did not speculate in detail on the nearer affinities of I. zazae, although they noted some possibilities. In size this species must have resembled Quaternary megalocnines, although size is a very poor indicator of relationships among phyllophagans. Its upper caniniform may have been trigonal (see MacPhee and Iturralde-Vinent, 1994), but this is a primitive feature within Megalonychidae (although it is one critical way in which Imagocnus differs from Megalocnus, whose caniniforms are uniquely reniform in section). Until better material is forthcoming, it is appropriate to leave the Zaza sloth as incertae sedis (cf. McKenna and Bell, 1997). Megalonychidae, gen. et sp. indet. The remaining entities to discuss under this heading can be conveniently grouped as “gen. et sp. indet.” (Figure 1). The specimens come from different islands and vastly different time periods, and there is no reason at all to think that they represent the same lower-level taxa (although most are certainly megalonychid). Species A (Yauco sloth) MacPhee and Iturralde-Vinent (1995) recently reported the discovery of a phyllophagan at AMNH 1994/1, an early Oligocene locality near the town of Yauco in southwestern Puerto Rico. Unfortunately, the only attributable specimen is the proximal end of a femur (AMNH VP 129883), which, although diagnostically phyllophagan, is not certainly megalonychid. (However, this remains the most plausible allocation, in view of the known mammal diversity in the Greater Antilles.) Although some resemblances to the femur of Acratocnus odontrigonus are evident, the Yauco specimen is in the size range of Neocnus, indicating that very small sloths were part of the faunal picture of the Greater Antilles as early as 33 to 34 mya. Species B and C (Cuban sloths) Oscar Arredondo and Osvaldo Vasquez Jiménez (personal communication) are in the process of proposing new taxa of megalonychids based on isolated bones from ?Quaternary cave sites in the province of Pinar del Río. We omit them from discussion here because names and descriptions have not yet been formally published. Species D (Grenadian sloth) Several sloth teeth recovered from a late Cenozoic locality in southern Grenada are not sufficiently diagnostic to determine their affiliation beyond the clear fact that they are megalonychid (MacPhee et al., 2000a). The authors place the age of the specimens, with a question mark, as latest Pliocene or early Pleistocene on the basis of preliminary K/Ar dates and taphonomic considerations. A molariform in the sample is comparable in size to that of Parocnus browni, while one of the caniniforms is reminiscent of the equivalent tooth in Paulocnus petrifactus. Comparisons with the latter genus are of special interest because of the proximity of Curaçao to Grenada (as compared to the Greater Antilles), and because — uniquely in the West Indies — both islands also supported capybaras.

BIOGEOGRAPHICAL ISSUES Megalonychid sloths are known from Cuba, Hispaniola (including the islands of La Tortue and La Gonâve), Puerto Rico, Curaçao, and Grenada (Anthony, 1926; Hooijer, 1962, 1964, 1967; Paula Couto, 1967; MacPhee et al., 2000a, 2000b). The majority of fossils are probably or certainly Pleistocene, but recent discoveries of Tertiary remains indicate that sloths entered the Antilles at least as early as the Early Oligocene (MacPhee and Iturralde-Vinent, 1995). The biogeographical history of Antillean land vertebrates has been the subject of decades of debate, controversy, and new interpretations (e.g., Rosen, 1975; Williams, 1989; MacPhee and Wyss, 1990; Hedges et al., 1992; MacPhee and Iturralde-Vinent, 1994, 1995; Hedges, 1996;

226

Biogeography of the West Indies: Patterns and Perspectives

Iturralde-Vinent and MacPhee, 1999; Pregill and Crother, 1999). The cladistic analysis and systematic revision presented above bear on two major biogeographical issues relevant to sloths and other Antillean land mammals: the origin and initial colonization of terrestrial environments in the Greater Antilles, and the significance of patterns of inter-island relationships.

COLONIZATION

OF THE

ANTILLES

The last ancestor of Antillean sloths lived in South America, as did the proximal ancestors of most other Antillean land mammal clades within the orders Primates, Chiroptera, and Rodentia. The only possible exception is Insectivora (Whidden and Asher, Chapter 15, this volume). The discovery of a femur (Species A in our classification) from the early Oligocene of southwestern Puerto Rico (MacPhee and Iturralde-Vinent, 1995) gives a minimum age for land mammal emplacement on the Antillean islands of 33 to 34 mya. Other Tertiary sloth material (Imagocnus zazae) has been found in early Miocene deposits on Cuba (MacPhee and Iturralde-Vinent, 1994). Three major mechanisms have been proposed to explain colonization of the Antilles by land mammals: continent–island vicariance (e.g., Rosen, 1975), overwater dispersal (e.g., Hedges et al., 1992), and dispersal over a short-lived land span* connecting South America with the nascent Greater Antilles via the Aves Rise (e.g., MacPhee and Iturralde-Vinent, 1995). Continent–island vicariance has been largely refuted as an explanation for the initial emplacement of mammals because modern mammalian lineages had not differentiated at the time of the hypothesized vicariance event (Williams, 1989; Hedges et al., 1992; MacPhee and Iturralde-Vinent, 1995; IturraldeVinent and MacPhee, 1999). In addition, Iturralde-Vinent and MacPhee (1999) have argued that permanently terrestrial environments were not present within the Caribbean basin until the late Eocene. Therefore, any mammals that had managed to colonize any Caribbean landmasses prior to the late Eocene would not have survived. Several recent workers, especially Hedges and coworkers, have argued that overwater dispersal was a major factor in colonization of the Greater Antilles. While this mechanism may offer an apparently parsimonious explanation for the presence of some taxa currently known from the islands, its random and fortuitous nature precludes useful testing (Page and Lydeard, 1994; MacPhee and Iturralde-Vinent, 1999; Pregill and Crother, 1999). An initial emplacement date in the early Oligocene is in good agreement with the land span model proposed by MacPhee and Iturralde-Vinent (1994, 1995) and Iturralde-Vinent and MacPhee (1999), although until the fossil record improves, the argument must be carried mostly by geology and paleogeography rather than biology. During the Eocene–Oligocene transition (ca. 35 mya), the developing northern Greater Antilles (i.e., proto-Cuba, Hispaniola, and Puerto Rico) and northwestern South America were briefly connected by a land span centered on the emergent Aves Ridge. This structure (Greater Antilles Ridge + Aves Ridge) has been named GAARlandia (MacPhee and Iturralde-Vinent, 1995). The massive uplift event that apparently permitted these connections was spent by 32 mya; a general subsidence followed, ending the GAARlandia land span phase. Thereafter, Caribbean neotectonism resulted in the subdivision of existing land areas. Under this scenario, a land span lasting from 35 to 33 mya would have enabled faunal emplacement by overland dispersal during a relatively short interval of geological time. However, just because the means to disperse to the Greater Antilles existed does not demonstrate that it was utilized by the taxon of interest. To show this in an inductively plausible manner, real biological evidence is required. Although it is permissible to argue that at least one kind of sloth was in the Greater Antilles by the early Oligocene, our phylogenetic evidence does not support the idea of a unified clade arising from one megalonychid invasion. Instead, this analysis suggests at least two separate colonization events by megalonychids. As early as 1923, Kraglievich divided the Antillean genera into two separate subfamilies, both of which also included South American taxa. Kraglievich (1923) and later Pascual et al. (1990) * Subaerial connection between a continent and one or more off-shelf islands.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

227

proposed that Megalocnus was a close relative of the South American Miocene taxon Amphiocnus. While neither of those claims was based on cladistic analysis, the phylogeny we present in this chapter is, and it strongly supports a diphyletic origin of Antillean megalonychids (Figure 3). Choloepodinae (including Choloepus, Acratocnus, Paulocnus, and Neocnus) is separated from Megalocninae (including Megalocnus and Parocnus) by the South American Miocene taxon Hapalops. That Choloepodinae shares a common ancestry with a Miocene taxon from South America implies that the subfamily diverged from Megalocninae well before the Miocene, and that its ancestor must have reached the Greater Antilles in a separate event. Whether there were two separate migrations across GAARlandia, presumably closely spaced in time, cannot be determined at present. Unfortunately, the Tertiary sloth material known to date cannot shed much light on the issue of diphyly. The oldest Antillean sloth fossil, a proximal femur from the early Oligocene of Puerto Rico, appears to resemble members of the Choloepodinae in that it is similar in size to the smallest species of Neocnus (MacPhee and Iturralde-Vinent, 1995). However, there is so little of it present that we do not feel comfortable allocating it to a taxon any more specific than Megalonychidae gen. + sp. indet. (see classification). Imagocnus zazae, from the Miocene of Cuba, is more certainly megalonychid. Its large size and some aspects of its morphology may link it with members of the Megalocninae rather than the Choloepodinae, but again, there is too little material present to make a definitive taxonomic allocation. Further resolution of this issue will require recovery of additional early fossil sloth material from the Greater Antilles as well as the analysis of a much larger comparative data set like the one employed by Gaudin (1995).

DISTRIBUTION

OF

FAUNA

ACROSS ISLANDS

A separate biogeographical issue, addressed in the second part of the model proposed by IturraldeVinent and MacPhee (1999), concerns how the known inter-island distribution of mammalian taxa arose. Assuming that ancestral sloths reached the landmass in the mid-Tertiary, their descendants must have become distributed throughout the islands either through island–island vicariance or by overwater dispersal. At the proposed time of emplacement, central and eastern Cuba, northern Hispaniola, and Puerto Rico were all connected (Iturralde-Vinent and MacPhee, 1999). This period coincides with fossil evidence for first appearances of the major clades of nonvolant Antillean mammals. Paleogeographical evidence suggests that eastern Cuba did not separate from northern Hispaniola until the Windward Passage was formed in the late Oligocene to early Miocene, and that Hispaniola and Puerto Rico were probably connected into the late Miocene (MacPhee and Iturralde-Vinent, 1994, 1995; Iturralde-Vinent and MacPhee, 1999). Therefore, the Antillean islands did not achieve their present geographical arrangement until the end of the Miocene, well after the earliest records of known Antillean mammal clades. Similarities in distributions of sister taxa in different taxonomic groups suggest that these distributions may have been formed by a common cause (i.e., island–island vicariance), rather than a series of random overwater dispersal events. While there are some exceptional cases of taxa whose presence cannot be explained in this manner (see Iturralde-Vinent and MacPhee, 1999), island–island vicariance seems to be the most parsimonious way to explain current faunal distributions. The phylogeny and classification presented here offers strong support for island–island vicariance as the mechanism that created the known distribution of sloth taxa across the Greater Antilles. Each of the four genera is represented on more than one island, and Acratocnus is present on three islands. Such a balanced arrangement would have required at least five separate overwater dispersal events, which we regard as highly unlikely. A much more reasonable scenario is that the four genera had already differentiated and spread before the northern Greater Antilles assumed their current configuration. As predicted by vicariance theory, the (biological) branching pattern of taxa should accord with the (geological) branching pattern of the landmasses on which they lived. One rather enigmatic result of our analysis and previous cladistic analyses (White, 1993a, 1993b; Gaudin, 1995) is the placement of Choloepus. It is now well accepted that Choloepus is

228

Biogeography of the West Indies: Patterns and Perspectives

not the closest relative of Bradypus, and most recent workers place the two-toed sloth within the Megalonychidae (e.g.; Scillato-Yane, 1980; Webb, 1985; Wetzel, 1985; White, 1993b; Gaudin, 1995; McKenna and Bell, 1997). Recent cladistic analyses, including the one presented here, not only support this placement, but also deeply embed Choloepus within the Antillean megalonychids as the sister taxon to Acratocnus, to the exclusion of other Antillean megalonychids. Choloepus is only known as a Recent taxon from South and Central America, and it has not been found in the Greater Antilles or indeed anywhere else in the Caribbean Basin. In addition, Acratocnus has not been found from the mainland. Such a close relationship between the two-toed tree sloth and Acratocnus in particular suggests that their divergence occurred after the divergence between Acratocnus and Neocnus. Whether this relationship implies that Acratocnus and Neocnus dispersed separately to the Antilles from South America while Choloepus remained on the mainland (necessitating at least three sloth invasions) or that Choloepus dispersed to South America from the Antilles cannot be determined with available material. It is our hope that larger data sets including more taxa and more characters, combined with a more extensive fossil record, will bring this issue to resolution in the future.

ACKNOWLEDGMENTS We thank Charles Woods for granting us access to the Hispaniolan sloth collection at the Florida Museum of Natural History (FLMNH), and for inviting us to contribute to this book. Loans of material were approved by S. David Webb and sent to us by Marc Frank and Brian Beatty (FLMNH). J. L. W. thanks Erika Simons (FLMNH) for help with photography and translation of German text. Patricia Wynne and Clare Flemming (AMNH) helped to prepare figures. Howard Whidden made helpful suggestions for improving earlier drafts of the manuscript.

LITERATURE CITED Aguayo, C. G. and L. H. Rivera. 1954. Sinopsis de los mamíferos cubanos. Circulación del Museo y Biblioteca Zoologicos, La Habana:1283–1324. Ameghino, F. 1889. Contribución al conocimiento de los mamíferos fósiles de la República Argentina. Actas Academia Nacional de Ciencias Córdoba 6:1–1027. Anthony, H. E. 1916. Preliminary report of fossil mammals from Porto Rico, with descriptions of a new genus of ground sloth and two new genera of hystricomorph rodents. Annals of the New York Academy of Sciences 27:193–203. Anthony, H. E. 1918. The indigenous land mammals of Porto Rico, living and extinct. Memoirs of the American Museum of Natural History new series 2:333–435. Anthony, H. E. 1920. New Mammals from Jamaica. Bulletin of the American Museum of Natural History 42:469–476. Anthony, H. E. 1926. Mammals of Porto Rico, living and extinct — Rodentia and Edentata. Scientific Survey of Porto Rico and the Virgin Islands, 9(2):97–241. Publication of the New York Academy of Sciences. Anthony, H. E. 1940. Summary of fossil land mammals of the West Indies. Proceedings of the 8th American Scientific Congress 4:359–363. Arredondo, O. 1960. Origen, evolución, y extinción del “Megalocnus rodens.” El Cartero Cubano 9–12. Arredondo, O. 1961. Descripciones preliminares de dos nuevos géneros y especies de edentados del Pleistoceno Cubano. Boletin del Grupo de Exploraciones Científicas 1:19–40. Arredondo, O. 1977. Nueva especie de Mesocnus (Edentata: Megalonychidae) del Pleistoceno de Cuba. Poeyana 172:1–10. Arredondo, O. 1988. Nueva subfamilia de Megalonychidae (Mammalia: Edentata). Miscelanea Zoologica 39:1–2. Domning, D. P., R. J. Emry, R. W. Portell, S. K. Donovan, and K. S. Schindler. 1997. Oldest West Indian land mammal: rhinocerotoid ungulate from the Eocene of Jamaica. Journal of Vertebrate Paleontology 17(4):638–641. Engelmann, G. F. 1978. The Logic of Phylogenetic Analysis and the Phylogeny of the Xenarthra. Ph.D. dissertation, Columbia University, New York.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

229

Engelmann, G. F. 1985. The phylogeny of the Xenarthra. Pp. 51–64 in Montgomery, G. G. (ed.). The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Fischer, K. 1971. Riesenfaultiere (Megalonychidae, Edentata, Mammalia) aus dem Pleistozän der Pio-DomingoHöhle in Kuba. Wissenschaftliche Zeitschrift der Humboldt-Universität zu Berlin MathematischNaturwissenschaftliche Reihe 20(4/5):609–673. Gardner, A. L. 1993. Order Xenarthra. Pp. 63–68 in Wilson, D. E. and D. M. Reeder (eds.). Mammal Species of the World: A Taxonomic and Geographic Reference. 2nd ed. Smithsonian Institution Press, Washington, D.C. Gaudin, T. J. 1995. The ear region of edentates and the phylogeny of the tardigrada (Mammalia, Xenarthra). Journal of Vertebrate Paleontology 15(3):672–705. Gervais, F. L. P. 1855. Mammifères. Animaux nouveaux ou rares recueillis pendant l’expédition dans les parties centrales de l’Amérique de Sud. P. Bertrand, Paris. Greenwood, A. D., J. Castresana, G. Feldmaier-Fuchs, and S. Paabo. 2001. A molecular phylogeny of two extinct sloths. Molecular Phylogenetics and Evolution 18:94–103. Hedges, S. B. 1996. Vicariance and dispersal in Caribbean biogeography. Herpetologica 52(3):466–473. Hedges, S. B., C. A. Hass, and L. R. Maxson. 1992. Caribbean biogeography: molecular evidence for dispersal in West Indian terrestrial vertebrates. Proceedings of the National Academy of Sciences, U.S.A. 89:1909–1913. Hoffstetter, R. 1955. Un mégalonychidé (édenté gravigrade) fossile de Saint-Domingue (Ile d’Haïti). Bulletin du Muséum National d’Histoire Naturelle, Paris, Serie 2, 27(1):100–104. Honacki, J. H., K. E. Kinman, and J. W. Koeppl (eds.). 1982. Mammal species of the world: a taxonomic and geographic reference. Allen Press and The Association of Systematic Collections, Lawrence, Kansas. Hooijer, D. A. 1962. A fossil ground sloth from Curaçao, Nertherlands Antilles. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Series B 65(1):46–60. Hooijer, D. A. 1964. The snout of Paulocnus petrifactus (Mammalia, Edentata). Zoologische Mededelingen 39:79–84. Hooijer, D. A. 1967. Pleistocene vertebrates of the Netherlands Antilles. Pp. 399–406, in Martin, P. S. and H. E. Wright (eds.). Pleistocene Extinctions: The Search for a Cause. Yale University Press, New Haven, Connecticut. Hooijer, D. A. and C. E. Ray. 1964. A metapodial of Acratocnus (Edentata: Megalonychidae) from a cave in Hispaniola. Proceedings of the Biological Society of Washington 77:253–258. Iturralde-Vinent, M. A. and R. D. E. MacPhee. 1999. Paleogeography of the Caribbean region: implications for Cenozoic biogeography. Bulletin of the American Museum of Natural History 238:1–95. Kraglievich, L. 1923. Descripción de dos cráneos y otros restos del género “Pliomorphus” Ameghino, procedentes de la formacion entrerriana de las Barrancas del Río Paraná. Anales del Museo Nacional de la Historia Natural, Buenos Aires 33:1–56. Kretzoi, M. 1968. New generic names for homonyms. Vertebrata Hungarica 10(1–2):163–166. Leidy, J. 1868. Notice of some vertebrate remains from the West Indian islands. Proceedings of the Academy of Natural Sciences of Philadelphia 20:178–180. MacPhee, R. D. E. 1997a. Digging Cuba: the lessons of the bones. Natural History 106(11):50–54. MacPhee, R. D. E. 1997b. Vertebrate paleontology of Jamaican caves. Pp. 47–56 in Fincham, A. (ed.). Jamaica Underground. 2nd ed. The Press University of the West Indies, Kingston, Jamaica. MacPhee, R. D. E. and D. A. Grimaldi. 1996. Mammal bones in Dominican amber. Nature 380:489–490. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1994. First Tertiary land mammal from Greater Antilles: an early Miocene sloth (Xenarthra, Megalonychidae) from Cuba. American Museum Novitates 3094:1–13. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1995. Origin of the Greater Antillean land mammal fauna, 1: new Tertiary fossils from Cuba and Puerto Rico. American Museum Novitates 3141:1–30. MacPhee, R. D. E. and M. A. Iturralde-Vinent. In press. Domo de Zaza: paleontological and geological investigations of an early Miocene vertebrate locality in south-central Cuba. American Museum Novitates. MacPhee, R. D. E. and A. R. Wyss. 1990. Oligo-Miocene vertebrates from Puerto Rico, with a catalog of localities. American Museum Novitates 2965:1–45. MacPhee, R. D. E., C. Flemming, and D. Lunde. 1999. “Last occurrence” of the Antillean insectivore Nesophontes: new radiometric dates and their interpretation. American Museum Novitates 3261:1–19. MacPhee, R. D. E., R. Singer, and M. Diamond. 2000a. Late Cenozoic fossil mammals from Grenada, Lesser Antilles Island-Arc. American Museum Novitates 3302:1–20.

230

Biogeography of the West Indies: Patterns and Perspectives

MacPhee, R. D. E., J. L. White, and C. A. Woods. 2000b. New megalonychid sloths (Phyllophaga, Xenarthra) from the Quaternary of Hispaniola. American Museum Novitates 3303:1–32. Maddison, W. P. and D. R. Maddison. 1992. MacClade: Analysis of Phylogeny and Character Evolution. Version 3.0 manual. Sinauer Associates, Sunderland, Massachusetts. Matthew, W. D. 1918. Affinities and origin of the Antillean mammals. Bulletin of the Geological Society of America 29:657–666. Matthew, W. D. 1919. Recent discoveries of fossil vertebrates in the West Indies and their bearing on the origin of the Antillean fauna. Proceedings of the American Philosophical Society 58:161–181. Matthew, W. D. 1931. Genera and new species of ground sloths from the Pleistocene of Cuba. American Museum Novitates 511:1–5. Matthew, W. D. and C. de Paula Couto. 1959. The Cuban edentates. Bulletin of the American Museum of Natural History 117:1–56. Mayo, N. A. 1969. Nueva especie de Megalonychidae y descripción de los depósitos cuaternarios de la Cueva del Vaho, Boca de Jaruco, La Habana. Memorias de la Facultad de Ciencias de la Universidad de La Habana 3:1–58. Mayo, N. A. 1978a. Ein neues Genus und zwei neue Arten fossiler Riesenfaultiere von der Familie der Megalonychidae (Edentata: Mammalia) aus dem Pleistozän Kubas. Eclogae Geologicae Helvetiae 71/3:687–697. Mayo, N. A. 1978b. Revision of Neocnus minor, Arredondo, 1961 (Edentata: Megalonychidae) from the Pleistocene of Cuba. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Series B 81(3):322–338. Mayo, N. A. 1980a. Nueva especie de Neocnus (Edentata: Megalonychidae de Cuba) y consideraciones sobre la evolución, edad, y paleoecología de las especies de este género. Actas II Congreso Argentino de Paleontología y Bioestratigrafía y I Congreso Latinoamericano de Paleontología, Buenos Aires, T. III:223–236. Mayo, N. A. 1980b. Revision de Neocnus major Arredondo, 1961 (Mammalia: Edentata del Pleistoceno de Cuba) con descripciones de un cráneo y algunos huesos postcraneales. Estúdios Geológicos 36:427–440. McFarlane, D. A., R. D. E. MacPhee, and D. C. Ford. 1998. Body size variability and a Sangamonian extinction model for Amblyrhiza, a West Indian megafaunal rodent. Quaternary Research 50:80–89. McKenna, M. C. and S. K. Bell. 1997. Classification of Mammals above the Species Level. Columbia University Press, New York. Miller, G. S. 1922. Remains of mammals from caves in the Republic of Haiti. Smithsonian Miscellaneous Collections, 74(3):1–8. Miller, G. S. 1929. A second collection of mammals from caves near St. Michel, Haiti. Smithsonian Miscellaneous Collections 81(9):1–30. Miller, G. S. 1930. Three small collections of mammals from Hispaniola. Smithsonian Miscellaneous Collections 82(15):1–10. Mones, A. 1986. Palaeovertebrata sudamericana: catálogo sistemático de los vertebrados fósiles de America de Sur. Parte I. Liste Preliminar y Bibliografía. Courier Forschungsinstitut Senckenberg 82:1–625. Muizon, C. de and H. G. McDonald. 1995. An aquatic sloth from the Pliocene of Peru. Nature 375:224–227. Naples, V. L. 1987. Reconstruction of cranial morphology and analysis of function in the Pleistocene ground sloth Nothrotheriops shastense (Mammalia, Megatheriidae). Contributions in Science 389:1–21. Publication of the Natural History Museum of Los Angeles County. Page, R. D. M. and C. Lydeard. 1994. Towards a cladistic biogeography of the Caribbean. Cladistics 10:21–41. Pascual, R., M. G. Vucetich, G. J. Scillato-Yané, and M. Bond. 1985. Main pathways of mammalian diversification in South America. Pp. 219–247 in Stehli, F. and S. D. Webb (eds.). The Great American Biotic Interchange. Plenum Press, New York. Pascual, R., M. G. Vucetich, and G. J. Scillato-Yané. 1990. Extinct and Recent South American and Caribbean edentates and rodents: outstanding examples of isolation. Atti Convegni Lincei 85:627–640. Publication of Accademia Nazionale dei Lincei, Rome. Patterson, B. and R. Pascual. 1968. Evolution of mammals on southern continents: the fossil mammal fauna of South America. Quarterly Review of Biology 43:409–451. Patterson, B., W. Segall, W. D. Turnbull, and T. J. Gaudin. 1992. The ear region in xenarthrans (=Edentata, Mammalia). Part II. Pilosa (sloths, anteaters), palaeanodonts, and a miscellany. Fieldiana Geology n. s. 24:1–79. Paula Couto, C. de 1967. Pleistocene edentates of the West Indies. American Museum Novitates 2304:1–55.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

231

Paula Couto, C. de 1979. Tratado de Paleomastozoologia. Academia Brasileira de Ciências, Rio de Janeiro. Pregill, G. K. and B. I. Crother. 1999. Ecological and historical biogeography of the Caribbean. Pp. 335–356 in Crother, B. I. (ed.). Caribbean Amphibeans and Reptiles. Academic Press, San Diego, California. Rosen, D. E. 1975. A vicariance model of Caribbean biogeography. Systematic Zoology 24:431–464. Scillato-Yané, G. J. 1977. Nuevo Megalonychidae (Edentata, Tardigrada) de edad Chasiquense (Plioceno temprano) del sur de la provincia de Buenos Aires (Argentina). Su importancia filogenetica, bioestratigrafica y paleobiogeografica. Revista de la Asociacion de Ciencias Naturales del Litoral 8:45–54. Scillato-Yané, G. J. 1979. El más antiguo Megalonychinae (Edentata, Tardigrada, Megalonychidae) conocido: Holomegalonyx menendezi n. gen., n. sp., del Colhuehuapense (Oligoceno tardio) de Gaiman (Provincia de Chubut, Argentina). Ameghiniana 14:311–312. Scillato-Yané, G. J. 1980. Nuevo Megalonychidae (Edentata, Tardigrada) del “Mesopotamiense” (Mioceno tardio-Plioceno) de la Provincia de Entre Ríos. Ameghiniana 17:193–199. Simpson, G. G. 1945. The principles of classification and a classification of the mammals. Bulletin of the American Museum of Natural History 85:1–350. Swofford, D. L. 1998. PAUP* 4.0. Phylogenetic Analysis Using Parsimony (*and other methods). Sinauer Associates, Sunderland, Massachusetts. Torre, C. de la and W. D. Matthew. 1915. Megalocnus and other Cuban ground sloths. Bulletin of the Geological Society of America 26:152. Varona, L. S. 1974. Catálogo de los mamíferos vivientes y extinguidos de las Antillas. Academia de Ciencias de Cuba, La Habana. Varona, L. S. 1976. Reemplazo de Cubanocnus por Neocnus (Mammalia: Edentata). Miscelanea Zoologica 2:4. Webb. S. D. 1985. The interrelationships of tree sloths and ground sloths. Pp. 105–112 in Montgomery, G. G. (ed.). The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Webb, S. D. and S. Perrigo. 1985. New megalonychid sloths from El Salvador. Pp. 113–120 in Montgomery, G. G. (ed.). The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Wetzel, R. M. 1985. The identification and distribution of recent Xenarthra (= Edentata). Pp. 5–21 in Montgomery, G. G. (ed.). The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. White, J. L. 1993a. Functional and Phylogenetic Implications of the Postcranial Skeleton of Fossil Sloths for the Evolution of Arboreality in Tree Sloths. Ph.D. dissertation, State University of New York at Stony Brook, New York. White, J. L. 1993b. Phylogenetic implications of the postcranial skeleton in sloths. Journal of Vertebrate Paleontology, 13(Supplement 3):62A. White, J. L., R. D. E. MacPhee, and C. A. Woods. 1996. Hispaniolan sloths: phylogeny and functional morphology. Journal of Vertebrate Paleontology 16(Supplement 3):72A. Williams, E. E. 1989. Old problems and new opportunities in West Indian biogeography. Pp. 1–46 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Woods, C. A. 1989. A new capromyid rodent from Haiti: the origin, evolution, and extinction of West Indian rodents, and their bearing on the origin of New World hystricognaths. Pp. 59–90 in Black, C. C. and M. R. Dawson (eds.). Papers on Fossil Rodents in Honor of Albert Elmer Wood. Scientific Series of the Natural History Museum of Los Angeles County 33. Woods, C. A. 1990. The fossil and recent land mammals of the West Indies: an analysis of the origin, evolution, and extinction of an insular fauna. Atti Convegni Lincei 85:641–680. Publication of Accademia Nazionale dei Lincei, Rome.

232

Biogeography of the West Indies: Patterns and Perspectives

NOTE ADDED IN PROOF Since this chapter was sent to press, it has come to our attention that two new genera of megalonychids from Cuba have recently been formally named: Galerocnus jaimezi Arredondo and Rivero, 1997,* and Paramiocnus riveroi Arredondo and Arredondo, 2000.** Each new taxon is based on a single element. We will not attempt to assess the phylogenetic position of these taxa until we are able to study the original material.

* Arredondo, C. and M. Rivero. 1997. Nuevo genero y especie de Megalonychidae del Cuaternario Cubano. Revista Biologia 11:105–112. ** Arredondo, C. and O. Arredondo. 2000. Nuevo genero y especie de perezoso (Edentata: Megalonychidae) del Pleistoceno de Cuba. Revista Biologia 14(1):66–72.

The Sloths of the West Indies: A Systematic and Phylogenetic Review

233

APPENDIX I: CHARACTERS AND CHARACTER STATES All characters are unordered and weighted equally. Characters marked with an asterisk (*) are uninformative and were excluded from the analysis. Characters 67 through 75 were defined by Gaudin (1995). 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. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.

sagittal crest: (0) absent; (1) single; (2) double/bifid postorbital constriction: (0) none; (1) moderate; (2) prominent cranial height: (0) flat; (1) domed; (2) significantly and irregularly domed rostrum mediolateral flare: (0) none; (1) slight; (2) very flared airorrhynchy (basicranial flexion): (0) none; (1) posteriorly flexed ventrally upper C1 (=M1): (0) strongly triangular; (1) mildly triangular; (2) rodentiform; (3) peglike; (4) lost upper M2: (0) anterolaterally convex; (1) anterolaterally concave; (2) none upper M5: (0) medially narrow; (1) medially wide; (2) oval; (3) none lower c1 (=m1): (0) triangular; (1) lingually concave; (2) meniscoid; (3) peglike; (4) none lower m2: (0) subtriangular; (1) subquadrate; (2) oval; (3) meniscoid; (4) none lower m4 lingual surface: (0) convex; (1) grooved; (2) multilobed; (3) none spout: (0) none; (1) pointed and short; (2) long and narrow; (3) spatulate; (4) huge condyles relative to toothrow: (0) slightly above; (1) very high caniniform relative to edge of maxilla: (0) at edge; (1) not at edge tooth row convergence: (0) parallel; (1) posterior; (2) anterior; (3) no teeth coronoid relative to condyle: (0) superior; (1) not superior; (2) absent maxillary/mandibular diastema: (0) long; (1) none; (2) short rectus femoris tubercle: (0) not prominent; (1) prominent; (2) forming a lip that projects laterally at nearly a right angle to the acetabulum acetabular rim: (0) significant gap; (1) tiny or absent gap femoral head: (0) spherical; (1) nonspherical; (2) large and globular femoral neck orientation: (0) proximal; (1) anterior; (2) medial femoral head fovea: (0) centric; (1) on posterior margin; (2) absent femoral shaft torsion; (0) none; (1) slight; (2) extreme third trochanter: (0) none; (1) forms lateral crest; (2) huge and confluent with greater trochanter greater trochanter: (0) inferior to head; (1) level with head; (2) superior to head well-developed anterior prong on femoral shaft: (0) no; (1) yes lesser trochanter: (0) nearly absent; (1) low, gentle bump; (2) conspicuous protrusion femoral shaft: (0) anteroposteriorly flat; (1) cylindrical; (2) proximally flat, distally round tibial shaft: (0) straight; (1) bowed separation of tibial articular surface: (0) anterior edge only; (1) slight; (2) distinct; (3) none proximal fibular facet of tibia: (0) posterolateral; (1) posterior; (2) lateral quadriceps femoris tubercle: (0) bump; (1) long scar; (2) hook with groove fibular shaft: (0) straight; (1) bowed calcaneal tuberosity expansion: (0) posterior; (1) posteromedial; (2) mediovolar; (3) volar calcaneal tuberosity ventral excavation: (0) no; (1) yes calcaneal tuberosity shape: (0) symmetrical and thick; (1) J-shaped; (2) symmetrical and waisted; (3) triangular tarsus: (0) serial; (1) alternate astragalar neck: (0) very short; (1) long and narrow navicular facet of astragalus: (0) concave; (1) flat; (2) convex astragalar trochlea: (0) wedge; (1) parallel and single surfaced; (2) well divided asymmetrically; (3) well divided symmetrically fibular facet of astragalus: (0) flat; (1) crescentic and slightly concave; (2) funnel-shaped ball and socket; (3) complete ball and socket medial trochlea of astragalus: (0) long and convex; (1) short and convex; (2) short and odontoid; (3) concave; (4) reduced anterior scapular border: (0) rounded; (1) ventrally concave inferior scapular angle: (0) round/obtuse; (1) acute; (2) has extra flange; (3) square

234

45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. *58. *59. *60. *61. *62. 63. 64. 65. 66. 67. *68. 69. 70. 71. 72. 73. 74. 75.

Biogeography of the West Indies: Patterns and Perspectives

scapular spine: (0) does not diverge at vertebral border; (1) diverges at vertebral border prescapular (=supraspinous) vs. postscapular (=infraspinous) fossa: (0) larger; (1) equal; (2) smaller deltoid and pectoral crests: (0) confluent; (1) do not join humeral trochlea: (0) flares distally; (1) not as distal as capitulum humeral head: (0) spherical; (1) flat; (2) globular; (3) mediolaterally narrow entepicondylar foramen: (0) present; (1) absent entepicondylar foramen visible posteriorly: (0) yes; (1) no; (2) absent entepicondylar bridge knob: (0) present; (1) absent supracondylar ridge: (0) prominent; (1) gentle pronator quadratus flange of radius: (0) abrupt lateral crest; (1) minimal crest; (2) posterior; (3) gentle, reduced lateral crest radial distal articular surface: (0) smooth; (1) irregular ulnar shaft: (0) straight; (1) bowed sigmoid notch of ulna: (0) wide with huge coronoid process; (1) segmented and shallow; (2) unsegmented and shallow dentition: (0) more than 7 upper and lower teeth; (1) 5 upper and 4 lower or fewer; (2) absent coracoscapular foramen: (0) absent; (1) present zygomatic arch: (0) complete; (1) complete but not fused; (2) incomplete tibia and fibula fusion: (0) yes; (1) no premaxilla: (0) large with extensive contact with nasals; (1) reduced with small nasal contact; (2) reduced with no nasal contact mandibular symphysis: (0) not well fused; (1) well fused optic foramen within sphenorbital fissure: (0) no; (1) yes acromion and coracoid form complete arch: (0) no; (1) yes pterygoid inflation: (0) absent; (1) present tympanic external surface: (0) smooth; (1) rugose entotympanic participation in tympanic cavity floor: (0) rudimentary or absent; (1) weak participation in medial portion of floor; (2) strong, forming almost entire medial half of floor glenoid position relative to superficies meatus [defined as “the groove on the ventral surface of the squamosal lateral and dorsal to the tympanum” by Patterson et al. (1992)]: (0) at or above meatus; (1) ventral to meatus glenoid posterior shelf: (0) absent; (1) present direction of root of zygoma: (0) anterior; (1) anterolateral; (2) lateral entotympanic participation in sulcus for internal carotid artery: (0) forms lateral wall of sulcus; (1) forms lateral wall and at least part of the roof; (2) forms lateral wall, roof, and has medial ridge forming at least part of medial wall paroccipital process: (0) weakly developed or rudimentary; (1) well-developed; (2) greatly enlarged, free-standing process pterygoid lateral groove: (0) absent; (1) present shape of glenoid: (0) elongate anteroposteriorly; (1) hemispherical; (2) widened mediolaterally

The Sloths of the West Indies: A Systematic and Phylogenetic Review

235

APPENDIX II: LIST OF TAXA USED IN CLADISTIC ANALYSIS For the Antillean sloth taxa, specimens were included in accordance with the systematic revision presented in this chapter. Therefore, some specimens were reallocated by us from their initial designations, and taxonomic names reflect our reorganization. Distributions are presented for Antillean fossil sloth taxa only. See Gardner (1993) for distributions of extant taxa and McKenna and Bell (1997) for distributions of other fossil taxa.

OUTGROUP TAXA Dasypus novemcinctus (Family Dasypodidae) Tamandua tetradactyla (Family Myrmecophagidae)

EXTANT INGROUP TAXA Bradypus tridactylus Choloepus didactylus

EXTINCT INGROUP TAXA Acratocnus antillensis (Cuba) A. ye (Hispaniola) A. odontrigonus (Puerto Rico) Megalocnus rodens (Cuba) M. zile (Hispaniola) Neocnus major (Cuba) N. gliriformis (Cuba) N. comes (Hispaniola) N. dousman (Hispaniola) N. toupiti (Hispaniola) Parocnus browni (Cuba) P. serus (Hispaniola) Paulocnus petrifactus (Curaçao) Hapalops longiceps Paramylodon harlani

Origin of the Greater 15 The Antillean Insectivorans Howard P. Whidden and Robert J. Asher Abstract — The origin of the endemic Antillean insectivorans Solenodon and Nesophontes has been a subject of considerable debate. We evaluate the main biogeographical hypotheses that have been proposed for these taxa in the light of recent phylogenetic, paleontological, and geological evidence. Recent phylogenetic analyses conflict with one another in many ways, but they provide little support for origin from a North American soricid ancestor. At least four hypotheses appear viable: origin by (1) overwater dispersal of a species related to some early Tertiary North American insectivoran (such as Apternodus or Centetodon); (2) vicariance of an early Tertiary North American insectivoran on a Western Jamaica Block, with subsequent overwater dispersal to the other islands of the Greater Antilles; (3) dispersal of a Gondwanan zalambdodont insectivoran directly from Africa to the Greater Antilles; and (4) dispersal of a Gondwanan zalambdodont insectivoran across a GAARlandia land span from northwestern South America, with subsequent vicariance as GAARlandia broke up. The relationship between Solenodon and Nesophontes is unclear, and it is possible that their distributions in the Greater Antilles are the results of different mechanisms.

INTRODUCTION The islands of the West Indies are home to two unusual and endemic genera of insectivorans, Solenodon Brandt, 1833 and Nesophontes Anthony, 1916. Solenodon contains two extant species, S. paradoxus from Hispaniola and S. cubanus from Cuba (Hutterer, 1993). In addition to these extant species, Patterson (1962) described S. marcanoi on the basis of fossil remains from Hispaniola, and Morgan and Ottenwalder (1993) described S. arredondoi on the basis of fossil remains from Cuba. The eight currently recognized species of Nesophontes are known from Puerto Rico, Cuba, and Hispaniola (Hutterer, 1993); two additional undescribed species are reported from the Cayman Islands (Morgan, 1994). Although there have been suggestions that Nesophontes survived into the 20th century (Nowak and Paradiso, 1983; Morgan and Woods, 1986), a recent study concluded that the genus has probably been extinct for several hundred years (MacPhee et al., 1999b). Both the degree of affinity between Solenodon and Nesophontes and their relationships to other insectivorans have never been well established. Despite some osteological similarity between Solenodon and Nesophontes, differences in their molar cusp patterns have led many zoologists to place them in taxonomically disparate groups. Because Solenodon has zalambdodont upper molars, like tenrecids and chrysochlorids (Figures 1a–c; Maier, 1985), most early workers (e.g., Peters, 1863; Dobson, 1882–1890; Allen, 1910; Winge, 1941) considered it to be allied with these Old World zalambdodonts. In contrast, Nesophontes had dilambdodont upper molars, like soricids and talpids (Figures 1d–f), and this led Anthony (1916) to argue that it had affinities with the Soricidae rather than with Solenodon; he placed in its own family, the Nesophontidae. However, Allen (1918) noted several similarities in the skulls of Solenodon and Nesophontes, and he argued that the two genera were closely related. Allen also disputed Anthony’s claim of soricid affinities for Nesophontes, and held instead that both genera were probably related to tenrecids and chrysochlorids. In a detailed monograph on the two Antillean genera, McDowell (1958) also stressed their osteological similarity, and he argued that they were closely related. Like Allen, McDowell placed the two genera in the same family (Solenodontidae), but he allied them with the Soricidae rather than with the Tenrecidae. Most recent classifications (e.g., Simpson, 1945; Miller and Kellogg, 1955; Hutterer, 1993; McKenna and Bell, 1997) have placed these genera in separate families. Van Valen (1967) went so 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

237

238

Biogeography of the West Indies: Patterns and Perspectives

a. Solenodon

b. Chrysochloris

c. Microgale

d. Nesophontes

e. Uropsilus

f. Sylvisorex

FIGURE 1 Molar cusp patterns of insectivorans: (a–c) zalambdodont, (d–f) dilambdodont. (a, d, e, and f after Butler, 1988; b after Maier, 1985; and c after Gregory, 1910.)

far as to place the Nesophontidae near the Soricidae in the order Insectivora, and the Solenodontidae near the Tenrecidae in the order Deltatheridia (although elsewhere in the same paper, p. 276, he remarks that a special relationship between Solenodon and Nesophontes is equally likely). In this chapter, we use the terms “insectivoran” and “lipotyphlan” synonymously. Both terms refer to the six extant families of insectivorans — chrysochlorids, erinaceids, solenodontids, soricids, talpids, and tenrecids — as well as such extinct lineages as Nesophontes, apternodontids, and geolabidids. However, we recognize that several recent studies have argued against the monophyly of these taxa (e.g., Springer et al., 1997; Stanhope et al., 1998; van Dijk et al., 2001; Madsen et al., 2001; Murphy et al., 2001).

RECENT PHYLOGENETIC STUDIES Until recently, interpretations of the phylogenetic affinities of Solenodon and Nesophontes were largely subjective, without explicit analyses to back them up. However, in the past several years, a number of studies have addressed the issue of insectivoran relationships more rigorously and more explicitly. These studies have employed cladistic and other modern phylogenetic methods of analysis, and they have utilized both molecular and morphological data.

MOLECULAR EVIDENCE Several recent studies have used DNA sequence data to examine the phylogenetic relationships of insectivorans. These studies have been concerned both with the placement of insectivorans in Mammalia and with the interrelationships of the insectivoran families. Sequence data are available primarily for extant taxa and and have not been used to address either the relationship between Solenodon and Nesophontes or possible affinities between these taxa and such extinct lineages as apternodontids or geolabidids. However, a number of the molecular studies have included an array of insectivoran taxa and therefore are relevant to the biogeographical history of the Greater Antillean taxa.

The Origin of the Greater Antillean Insectivorans

239

Studies using sequence data have been nearly unanimous in concluding that the order Lipotyphla (= Insectivora) is not monophyletic (e.g., Springer et al., 1997; Stanhope et al., 1998; Emerson et al., 1999; Liu and Miyamoto, 1999; Mouchaty et al., 2000). These studies have also generally held that chrysochlorids and tenrecids are part of an African mammal clade (the Afrotheria) that also includes elephants, hyraxes, sirenians, elephant shrews, and aardvarks. In addition, molecular clock estimates based upon the sequence data indicate that the Afrotheria has been separate from other eutherian lineages for more than 90 million years (Springer et al., 1997; Stanhope et al., 1998). This implies that tenrecids and chrysochlorids have been separate from Solenodon and Nesophontes for at least as long (although divergence estimates based on a molecular clock have been disputed, e.g., Foote et al., 1999). The two published molecular analyses that have incorporated sequences from Solenodon have come to different conclusions about its affinities. Analyzing mitochondrial genes for 12S rRNA, 16S rRNA, and valine tRNA, Stanhope et al. (1998; fig. 1) found weak support (51% of bootstrap replicates in a neighbor-joining analysis) for a clade in which Solenodon is sister taxon to soricids and talpids (Figure 2a). In contrast, an analysis with greater taxonomic sampling but based strictly upon sequences of the mitochondrial gene for 12S rRNA (Emerson et al., 1999) found no support for a soricid-talpid-Solenodon clade, and instead discovered weak support for Solenodon as sister to a clade of rodents (Figure 2b). Despite the apparent strength of the molecular support for Afrotheria, some authors have been cautious in their acceptance of the group. In their analyses of sequences from nuclear (IRBP and vWF) and mitochondrial (12S rRNA) genes, and also in combined analyses of molecular and morphological data, Liu and Miyamoto (1999) found support for Afrotheria, but they also noted that the inclusion of African insectivorans was the weakest part of the group’s definition. Emerson et al.’s (1999) strict consensus tree from 12S rRNA sequences supported an African clade that included not only chrysochlorids and tenrecids, but also the Southeast Asian primate Tarsius, an arrangement that few zoologists would take seriously. Emerson et al. (1999) went on to note that the 12S rRNA data set was able to recover relatively few interordinal groupings, and that most of the ones that were recovered were not well supported. Consequently, they concluded that 12S rRNA sequences might lack the ability to resolve basal relationships within Eutheria. Also, when Stanhope et al.’s (1998) data (alignment ds34832 available at ftp://ftp.embl-heidelberg.de/pub/databases/embl/align/) are analyzed using other sequence alignments and tree-reconstruction techniques (including parsimony), the support for Solenodon being a sister taxon to soricids and talpids disappears (Asher, 2000).

MORPHOLOGICAL EVIDENCE Asher (1999b) presented a cladistic morphological analysis that directly addresses the phylogenetic placement of Solenodon and Nesophontes. Asher used 193 morphological character states from 35 living and extinct taxa, and his study was designed to test hypotheses of insectivoran relationships under different sets of assumptions for character weighting, treatment of missing data, and character ordering. The results of the analysis were ambiguous about lipotyphlan monophyly but unambiguous about rejecting the African clade: neither the set of most-parsimonious trees (MPTs) nor the 256,000 suboptimal trees that Asher evaluated contained a topology supporting an African clade. The morphological analysis also did not support McDowell’s (1958) proposed association of soricids as the sister group to Solenodon and Nesophontes; this grouping was not present in any of the MPTs, and it was found in only a few suboptimal trees. However, in the most parsimonious result for one of Asher’s eight assumption sets, Nesophontes appears alone as the sister group to soricids. Also, although Nesophontes and Solenodon were not sister taxa in any of the MPTs, for some assumption sets such a relationship required only one additional step. A topology consistently present among Asher’s MPTs, but without strong branch or bootstrap support, places Solenodon and Apternodus (and in most cases Nesophontes) as successively distant sister taxa to the Tenrecidae and usually Chrysochloridae (Figure 2c). This arrangement is in general

240

Biogeography of the West Indies: Patterns and Perspectives

a

Solenodon mole shrew Perissodactyla Carnivora Chiroptera Cetacea Ruminantia pangolin Xenarthra aardvark elephant shrew tenrec golden mole elephant manatee hyrax Primates hedgehog rabbit myomorph rodents caviomorph rodents Metatheria platypus

c

b

Solenodon myomorph rodents pangolin Megachiroptera mole flying lemur Primates* Microchiroptera dormice rabbit tree shrew armadillo caviomorph rodents Perissodactyla Suiformes Cetacea Ruminantia shrew Carnivora aardvark tarsier elephant shrew tenrecs golden mole Proboscidea Sirenia hyrax hedgehogs Metatheria

tenrecs golden moles

†Apternodus Solenodon †Nesophontes †Centetodon moles shrew hedgehogs

†Leptictis tree shrew hyrax Carnivora aardvark Xenarthra Metatheria

FIGURE 2 Cladograms from recent phylogenetic analyses that have included Solenodon, modified to emphasize the positions of Solenodon and other insectivorans. Taxa from generally accepted clades (for example, those recognized in Anderson and Jones, 1984) have been lumped together; the names of these clades begin with a capital letter. (a) Majority-rule neighbor joining bootstrap tree from Stanhope et al. (1998); (b) Strict consensus tree from Emerson et al. (1999); (c) Majority rule tree of the eight MPTs from Asher (1999b). *Primates is paraphyletic in this cladogram. † = extinct.

The Origin of the Greater Antillean Insectivorans

241

FIGURE 3 Reconstructed ventral view of the ear region of Apternodus. Dotted line represents suture between petrosal and basisphenoid/basioccipital. Rostral is at top, medial is at right. (Adapted from McDowell, 1958.)

agreement with claims that the Caribbean insectivorans have an African sister taxon. In addition, it adds the interesting twist that Apternodus, an early Tertiary North American taxon, may also be related to the extant Afro-Malagasy zalambdodonts (i.e., tenrecids and chrysochlorids). The placement of Apternodus in a clade with Solenodon and other lipotyphlans challenges McDowell’s (1958) conclusion that Apternodus played no part in the evolutionary history of Caribbean insectivorans. McDowell had argued that living insectivorans possess a conservative arterial pattern in the middle ear. Based largely on his interpretation that it departed from this pattern, he concluded that Apternodus is more likely to be related to the extinct carnivoran-like Creodonta than to living insectivorans. Specifically, “… the course of the internal carotid artery through the tympanic cavity is quite characteristic of the Lipotyphla … Indeed, although at one time [McDowell] searched diligently for differences between families in tympanic arterial pattern, he has been unable to discover any notable departures from this pattern among the Lipotyphla … Apternodus, however, shows a very different arterial pattern from that of the Lipotyphla, being perhaps more like that of creodonts in this regard” (p. 168). Many aspects of soft-tissue reconstruction in McDowell’s paper are accurate — an impressive feat, given that his reconstructions were based primarily on dry skulls. However, McDowell overestimated the conservatism of the arterial pattern through the middle ear of insectivoran-grade mammals. Furthermore, there is now better material with which to reconstruct middle ear vasculature in Apternodus than he had available in the 1950s. Figure 3 shows McDowell’s arterial reconstruction of Apternodus, based on the type specimen of A. brevirostris (AMNH 22466). He is correct in stating (p. 168) that the promontory of AMNH 22466 is “quite smooth,” and he is probably also correct in noting a very shallow groove for the stapedial artery. Compared to the vasculature of Solenodon paradoxus, reconstructed by MacPhee (1981) from a histologically sectioned individual (Figure 4), the promontory artery of Apternodus as reconstructed by McDowell travels much more medially, near the petrosal-basisphenoid suture. This medial course led McDowell to compare the arterial pattern of Apternodus to that of the

242

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 4 Reconstructed ventral view of the ear region of Solenodon. Dotted line represents suture between petrosal and basisphenoid/basioccipital. Bifurcation of the proximal stapedial artery actually takes place within the braincase in this specimen of Solenodon. Rostral is at top, medial is at right. (Adapted from MacPhee, 1981.)

oxaenid creodont Patriofelis (Denison, 1938), and to conclude that Apternodus exhibited an arterial pattern outside the range of variation seen among insectivorans. In fact, the type specimen of A. brevirostris shows very slight grooves for middle ear vasculature, and these grooves are also visible in several well-preserved Apternodus skulls with basicrania that have been discovered since McDowell’s paper was published (e.g., AMNH 74941, 74943; FMNH 1690; MPUM 6855). These grooves permit only the general conclusion that Apternodus possessed promontory and stapedial arterial distributaries. Details on the spatial relations of specific arteries based on these specimens are much more tentative. Another Apternodus basicranium, AMNH 74942, shows more clearly than other specimens a groove coursing medially from the vestibular fenestra (housing the footplate of the stapes) to the ventral apex of the promontory bone, presumably for the proximal stapedial artery. Close to the apex, this groove is met by another traveling anteriorly (for the promontory artery), and continues a short distance medially (for the internal carotid proximal to the promontory-stapedial bifurcation). The groove for the promontory artery courses farther lateral to the petrosal-basisphenoid suture than that seen in McDowell’s reconstruction for the A. brevirostris type specimen. This indicates either that Apternodus was polymorphic in the course of its promontory artery along the petrosal, or that McDowell’s reconstruction was incorrect. Since McDowell’s 1958 work, considerable progress has been made in understanding the morphology and development of the cranial vasculature in mammals. For example, several researchers (e.g., Presley, 1979; MacPhee, 1981; Wible, 1984, 1987) have questioned the argument (made by Gregory, 1910) that primitive eutherians possessed both medial and lateral entocarotid arteries. These recent workers have recognized considerable vascular diversity among mammals, and they have also noted that middle ear vessels vary in course throughout the ontogeny of an individual. Bugge (1974), for example, reported that Tenrec ecaudatus lacks an inferior stapedial ramus altogether; this is in direct contrast to McDowell’s assertion (p. 168) based on dry skulls that “in Tenrec the ramus inferior follows the normal lipotyphlan course.” Asher (1999a, 2000) has also documented the lack of a ramus inferior in multiple individuals of Tenrec and Hemicentetes, confirming previous work by Bugge (1974) on Tenrec and Hemicentetes and by MacPhee (1981) on Hemicentetes. Additional

The Origin of the Greater Antillean Insectivorans

243

arterial variants among tenrecids include the reduced superior stapedial artery of Potamogale and the intracranial stapedial bifurcation of Geogale (Asher, 1999a, 2000). Furthermore, among extant insectivorans the course of the promontory artery varies from a medial course, near the petrosal-basisphenoid suture (e.g., Microgale; MacPhee, 1981; Asher, 2000), to a lateral one, entering the braincase lateral to the anterior pole of the petrosal bone (e.g., Chrysochloris; Asher, 1999a, 2000). In sum, it is unlikely that Apternodus possessed a promontory artery that traveled any more medially than that of extant tenrecids; also, extant insectivorans are much more variable in their cranial arterial supply than McDowell estimated. Thus, the main reason given by McDowell (1958) for removing Apternodus from consideration in insectivoran phylogeny is invalid.

BIOGEOGRAPHICAL HYPOTHESES Over the years, a variety of hypotheses have been proposed to account for the presence of Solenodon and Nesophontes in the West Indies. These hypotheses have been concerned primarily with the phylogenetic affinities and possible ancestors of the Antillean taxa, but they have also speculated on the routes that the animals may have used to get to the Antilles and on the time of their arrival in the islands. We depict these hypotheses graphically in Figure 5a–d, and summarize them as follows: 1. McDowell (1958) rejected possible affinities between the Antillean taxa and either AfroMalagasy tenrecids or North American Tertiary taxa such as Apternodus and Palaeoryctes, and he argued instead for a relationship with the Soricidae. He held that Solenodontidae (including both Solenodon and Nesophontes) was derived from an unknown tropical North American Tertiary “soricoid” ancestor (his “Soricoidea” did not include talpids). McDowell did not specify how this ancestor arrived in the Antilles; we assume that he envisioned overwater dispersal (Figure 5a). 2. Patterson (1962) held that Solenodon derived from a relatively unspecialized apternodontid living on the Central American peninsula during the first half of the Tertiary (Figure 5b). Rafting to the Antilles may have taken place in the later Eocene, at roughly the same time as the hypothesized rafting of caviomorphs and platyrrhine primates from South America to the Antilles. Patterson apparently did not consider Solenodon to be closely related to Nesophontes, and he did not discuss the biogeographical origin of Nesophontes. Simpson (1956) had put forth a similar explanation, although he was less specific in his statement of relationships, suggesting only that the Antillean genera had affinities to unspecified North American Tertiary insectivorans. 3. Hershkovitz (1972) claimed that both Nesophontes and Solenodon are zalambdodont insectivorans (despite the dilambdodont molars of Nesophontes), and that they are most likely related to tenrecids. These presumed African affinities, plus the absence of near relatives in the Americas, suggested to him a late Cretaceous to early Tertiary Gondwanan origin for the lineage, with overwater dispersal to the Antilles. He may have envisioned a direct crossing from Africa to the Greater Antilles, but he is vague enough to leave open the possibility of dispersal from Africa to South America and from there to the Greater Antilles (Figure 5c). As further support for this hypothesis, Hershkovitz noted the strong African affinities and presumed African origin of the South American primates and hystricomorphous (caviomorph) rodents. 4. MacFadden (1980) built upon the ideas of Rosen (1975) to present a vicariance explanation for the Antillean presence of Solenodon and Nesophontes. MacFadden (1980) accepted McDowell’s (1958) hypothesis that the Caribbean insectivorans comprise sister taxa that were derived from a continental North American taxon. However, he differed from McDowell in considering the Recent Antillean insectivorans to be close relatives of North American fossil forms such as Apternodus, and he explained the presence of

244

Biogeography of the West Indies: Patterns and Perspectives

Early Cenozoic

a North America Nuclear Central America overwater dispersal of shrew-like form

Eurasia

Greater Antilles

South America

Africa

Early Cenozoic

b

later extinction of North American zalambdodont forms

North America Nuclear Central America overwater dispersal

Eurasia

Greater Antilles

South America

Africa

FIGURE 5 Graphic representations of four biogeographical hypotheses for the origin of the Greater Antillean insectivorans. (a) McDowell (1958); (b) Patterson (1962); (c) Hershkovitz (1972); (d) MacFadden (1980). (After MacFadden, 1980, and Mouchaty, 1999.)

the living taxa in the Antilles as being the result of plate tectonic movements. Under this hypothesis, at some point during the early Tertiary, one or more proto-Antillean islands carried an Apternodus-like ancestor eastward from southern North America (Figure 5d). Once in the Antilles, this form then dispersed between islands and underwent allopatric speciation, resulting in Solenodon and Nesophontes.

The Origin of the Greater Antillean Insectivorans

245

Early Cenozoic

c North America

Eurasia

Nuclear Central America Greater Antilles

South America

Late Mesozoic

dispersal of zalambdodont when South Atlantic was narrower

Africa

Early Cenozoic

d North America Nuclear Central America ProtoAntilles

South America

North America Nuclear Central America Greater eastward movement of Antilles Caribbean Plate

South America

Later Cenozoic extinction of North American zalambdodont forms

North America Nuclear Central America Lower Central America

Greater Antilles

South America

FIGURE 5 (continued )

We do not believe that the evidence is currently at hand to resolve the question of the biogeographical history of Solenodon and Nesophontes. Despite the recent attention from both molecular and morphological systematists, higher-level relationships among lipotyphlans remain largely unresolved, and the phylogenetic affinities of Antillean insectivorans remain a mystery. However, not all of the proposed biogeographical hypotheses are consistent with the recent phylogenetic, paleontological, and geological data. In the section below, we use the new data to assess the four hypotheses of origin stated above, and we also propose an additional hypothesis.

246

Biogeography of the West Indies: Patterns and Perspectives

MCDOWELL (1958) McDowell’s hypothesis, which calls for derivation of both Solenodon and Nesophontes from a North American shrewlike ancestor, received little support from the previously cited phylogenetic analyses. Neither molecular nor morphological analyses found a special relationship between Solenodon, Nesophontes, and soricids. Stanhope et al. (1998) did find weak support for an association between Solenodon and soricids, but Solenodon was sister taxon to a soricid-talpid clade, not just to soricids as McDowell had argued. Also, as mentioned previously, even this support disappears with a parsimony analysis of their data set. Therefore, we believe that McDowell’s hypothesis for the origin of the Antillean insectivorans can be rejected.

PATTERSON (1962) Asher’s (1999b) morphological analysis found support for affinities between Solenodon and Nesophontes and extinct North American insectivorans such as Apternodus and Centetodon, and therefore there is phylogenetic evidence that is consistent with this hypothesis. Reconstructions of paleoceanographic currents suggest that the most plausible mode of overwater dispersal from North America or Central America to the Greater Antilles would be rafting in the Loop Current of the Gulf of Mexico. In the mid-Tertiary, the Loop Current flowed north across the tip of the Yucatan Peninsula, circled in the eastern Gulf of Mexico, and then flowed down the western coast of the Florida peninsula and through the Straits of Florida to join the Gulf Stream (Mullins et al., 1987; IturraldeVinent and MacPhee, 1999). The most likely arrival point in the Greater Antilles for organisms transported by this current would therefore be the landmasses at present associated with the northern coast of Cuba. If this mechanism is correct, it is somewhat surprising that other North American mammals did not reach the Greater Antilles in a similar fashion. Other mammalian lineages that are common in the North American Tertiary, such as carnivorans, ungulates, and marsupials, are completely absent from the known record for the Antilles (with the exception of the recent discovery of Hyrachyus; see below). In addition, muroid rodents were well diversified in North America by the end of the Miocene (Baskin, 1986; Korth, 1994). This group of rodents is noted for its overwater dispersal abilities (Darlington, 1957; Baskin, 1986), yet apparently none of them dispersed to the Greater Antilles until sigmodontine rodents made it to Jamaica in the Pleistocene (Morgan and Woods, 1986; Woods, 1989). Also, despite the fact that apternodontids are present in the early Tertiary of western and central North America, there are no relevant fossils from Central America or the southeastern United States, the most likely points of departure.

HERSHKOVITZ (1972) We noted above that molecular phylogenetic analyses support an association of chrysochlorids and tenrecids with an African mammal clade that also includes elephants, sirenians, hyraxes, elephant shrews, and aardvarks. If this placement is correct, it would preclude any direct association between Solenodon and the Old World zalambdodonts, and it would therefore contradict Hershkovitz’s hypothesis that the Antillean insectivorans were derived from a zalambdodont taxon that crossed the Atlantic from Africa. However, tenrecoid affinities for Solenodon and Nesophontes do receive support from the morphological analysis: most of the MPTs in Asher’s analysis included Apternodus, Solenodon, and Nesophontes as successively distant sister taxa to tenrecids. Therefore, we believe that a Gondwanan origin for Solenodon and Nesophontes remains a viable hypothesis. Hershkovitz held that this African zalambdodont insectivoran crossed the South Atlantic roughly contemporaneously with the hypothesized dispersal of caviomorph rodents and platyrrhine monkeys (e.g., Hoffstetter, 1980; Lavocat, 1980; Aiello, 1993; George, 1993). We feel that this explanation cannot be ruled out.

The Origin of the Greater Antillean Insectivorans

247

MACFADDEN (1980) Although MacFadden’s hypothesis of affinities between the Antillean insectivorans and Tertiary North American taxa receives support from Asher’s (1999b) analysis, classic continent–island vicariance, originally proposed by Rosen (1975) and hypothesized specifically for Solenodon and Nesophontes by MacFadden (1980), is not supported by the recent geological evidence. Recent interpretations of the geological history of the Caribbean indicate that permanent subaerial landmasses were not present in the area until the beginning of the late Eocene, which is long after elements of the Caribbean plate had separated from North America (Robinson, 1994; IturraldeVinent and MacPhee, 1999). Thus, even if ancestral insectivorans had been carried out into the Caribbean on a “proto-Antilles” island in the late Mesozoic, they almost certainly could not have survived there because repeated subsidence and transgression events completely submerged any islands that may have existed prior to the late Eocene (Iturralde-Vinent and MacPhee, 1999). However, a variant of this vicariance hypothesis may provide a viable explanation for the origin of Solenodon and Nesophontes. Domning et al. (1997) recently described the rhinocerotoid Hyrachyus from late early or early middle Eocene deposits of western Jamaica. This specimen is the oldest known land mammal from any of the West Indian islands, and it is also the first Tertiary land mammal from these islands that has definite North American affinities (Domning et al., 1997). The discovery of Hyrachyus in western Jamaica is consistent with geological reconstructions that place a Western Jamaica Block close to or in contact with Central America in the early Tertiary; this western block is thought to have moved east and become incorporated into modern Jamaica at some point in the Miocene (Iturralde-Vinent and MacPhee, 1999). This raises the possibility that a North American insectivoran used Jamaica as a conduit to emigrate from North America to the other Great Antilles (MacPhee et al., 1999a). However, if insectivorans arrived in the Caribbean via western Jamaica, they must have quickly dispersed to newly emergent islands to the east because western Jamaica was submerged for most of the time between the late Eocene and the Miocene (Robinson, 1994; Iturralde-Vinent and MacPhee, 1999).

THE LAND SPAN HYPOTHESIS The possibility of a land bridge connection between the American mainlands and the Greater Antilles has been proposed at various times over the years (see the discussion in Williams, 1989). However, this idea has not received much support recently because there has been little geological evidence for such a bridge (MacPhee and Wyss, 1990). In addition, many zoogeographers considered the Antillean vertebrate fauna to be depauperate and unbalanced, and therefore more consistent with the filtering effects of overwater dispersal than with the wholesale faunal movement that would likely accompany a land bridge (e.g., Simpson, 1956; Williams, 1989). In a recent series of papers, MacPhee and Iturralde-Vinent have developed a modification of the land bridge hypothesis (MacPhee and Iturralde-Vinent, 1994, 1995; Iturralde-Vinent and MacPhee, 1999). They refer to their model as the land span hypothesis, and they make a distinction between a land bridge, which connects two continental landmasses, and a land span, which connects a continent and an off-shelf island or group of islands. In support of this hypothesis, MacPhee and Iturralde-Vinent present evidence that during the Eocene–Oligocene transition the developing Greater Antilles were connected to northwestern South America by a land span. At that time, the islands on the northern part of the Greater Antillean Ridge (central and eastern Cuba, north-central Hispaniola, Puerto Rico, and the Virgin Islands) were either joined into a single island or formed a series of islands separated by narrow water gaps. MacPhee and Iturralde-Vinent hypothesized that during the Eocene–Oligocene transition, orogenic effects and a eustatic drop in sea level exposed the Aves Ridge, a ridge in the Caribbean basin that lies between the Greater Antilles and northern South America. When the Aves Ridge became subaerial, it connected the eastern end of the Greater Antilles Ridge with northwestern South America, which was then a

248

Biogeography of the West Indies: Patterns and Perspectives

microcontinent and partly or completely separated from southern South America by shallow water barriers. This land span connection, which MacPhee and Iturralde-Vinent named “GAARlandia” (for Greater Antilles Aves Ridge), may have existed for just 1 or 2 million years, and it was probably gone by the late Oligocene. MacPhee and Iturralde-Vinent also noted that the time frame for their hypothesized land span agrees well with the appearance in the Antilles of sloths, rodents, and primates, all of which have clear South American affinities. For example, the oldest non-Jamaican Antillean land mammal fossil, a femur from what is believed to be a megalonychid sloth, was collected from early Oligocene deposits on Puerto Rico (MacPhee and Iturralde-Vinent, 1995). In addition, there are fossils of platyrrhine primates, capromyid rodents, and megalonychid sloths from the early Miocene of Cuba (MacPhee and Iturralde-Vinent, 1994, 1995). MacPhee and Grimaldi (1996) also described a small, Nesophontes-sized mammal represented by a partial axial skeleton embedded in a piece of late Oligocene/early Miocene Dominican amber. MacPhee and Grimaldi identified a number of morphological features that are consistent with this animal being an insectivoran, and on the basis of size and morphology they also ruled out other mammalian groups known from the West Indies (bats, primates, rodents, and sloths). The age and location of this fossil suggest the possibility that insectivorans may also have arrived in the Greater Antilles via a GAARlandia land span. This hypothesis offers the advantage of invoking a common mechanism to explain the origin of all known Tertiary mammalian lineages in the Antilles. However, the hypothesis requires that the Antillean insectivorans had at least a proximate origin in South America, and there are no relevant insectivoran fossils from South America prior to the PlioPleistocene interchange with North America (Simpson, 1945; McKenna and Bell, 1997). Scott (1905) had suggested that the Miocene Necrolestes from Argentina might be related to chrysochlorids. However, Patterson (1958; following Winge, 1941) noted that, among other features, this genus has five upper incisors and a robust zygoma, indicating its status as a marsupial. It is still possible that the ancestors of the Antillean insectivorans originated in Africa and dispersed across a narrower South Atlantic to South America, as has been proposed for caviomorph rodents and platyrrhine primates (see Figure 5c). If South America served just as a temporary way station on the route from Africa to the Greater Antilles, then it would not be surprising that they did not leave a record of their presence.

CONCLUSIONS We consider there to be at least four viable biogeographical hypotheses for the origin of the Greater Antillean insectivorans: 1. Origin from a Tertiary North American insectivoran that colonized the Greater Antilles by overwater dispersal from Central America or the southeastern United States. 2. Origin from a Tertiary North American insectivoran that was carried from Central America into the Caribbean on the Western Jamaica Block, with subsequent overwater dispersal to other islands of the Greater Antilles. 3. Origin from a Tertiary African zalambdodont insectivoran that dispersed across a narrower South Atlantic directly from Africa to the Greater Antilles. 4. Origin from a Tertiary insectivoran that dispersed to the Greater Antilles from the northwestern South American microcontinent across a GAARlandia land span. This would have occurred roughly contemporaneously with the dispersal of sloths, primates, and caviomorph rodents. Subsequent breakup of GAARlandia was then at least partly responsible for the present distributions. The view that seems to be most commonly accepted today — origin from a North American Tertiary form with overwater dispersal to the Greater Antilles — appears to be largely a default position, with no strong evidence in its favor but without the apparent conflicts of alternative

The Origin of the Greater Antillean Insectivorans

249

explanations. It relies on the assumption that Solenodon and Nesophontes are related to such extinct North American forms as Apternodus and Centetodon, and it receives significant support from the absence of insectivoran fossils from South America. Prior to Asher (1999b), there were no explicit phylogenetic analyses supporting affinities between the Antillean taxa and extinct North American forms such as Apternodus, and Asher’s analysis does not provide unequivocal support because it also suggests affinities to the Afro-Malagasy tenrecids. In addition, except for the special case of Hyrachyus, there is no concrete evidence that North American land mammals reached the Greater Antilles at any time in the Tertiary. If this biogeographical hypothesis is correct, we would expect future phylogenetic analyses to support the derivation of Solenodon and/or Nesophontes from Apternodus, Centetodon, or some other North American Tertiary form. Also, we might expect to find related fossil forms in Central America or the southeastern United States. The modified vicariance hypothesis would receive considerable support if insectivorans are ever discovered in the western Jamaican Eocene sediments that yielded Hyrachyus. Again, future phylogenetic analyses would be expected to link such new fossils to Solenodon and Nesophontes, and also to extinct North American taxa such as Apternodus and Centetodon. The discovery of relevant insectivoran fossils in early Tertiary deposits from Central America would be consistent with this hypothesis as well as with the overwater dispersal hypothesis. Direct dispersal from Africa to the Greater Antilles would be supported if future phylogenetic analyses provide compelling support for close affinities between the Antillean insectivorans and AfroMalagasy tenrecids — especially if the Tertiary North American forms are excluded from this relationship. The hypothesis might also be strengthened if paleogeographical and paleoceanographical discoveries make an early Tertiary crossing of the South Atlantic seem more plausible. For example, future paleogeographic reconstructions might determine that the distance to be crossed was less than previously thought, or that the direction of prevailing oceanic currents would have carried rafting animals from western Africa directly to the Antilles. One of the main arguments against possible Gondwanan affinities for the Antillean insectivorans has been the absence of insectivoran-grade placental fossils from South America that predate the Plio-Pleistocene land bridge interchange with North America. The GAARlandia hypothesis would be supported if zalambdodont insectivorans are ever discovered in early Tertiary deposits from northwestern South America. This hypothesis would also be indirectly supported if future phylogenetic analyses confirm the association between Solenodon, Nesophontes, and tenrecids that appears in many of Asher’s (1999b) MPTs — with or without affinities to such North American taxa as Apternodus and Centetodon. In conclusion, although recent years have seen much research activity in areas relevant to the origin of Solenodon and Nesophontes, the answer to this biogeographical problem remains elusive. Since there are major conflicts in the three phylogenetic analyses to specifically address the affinities of Solenodon — Stanhope et al. (1998), Asher (1999b), and Emerson et al. (1999) — there is clearly a need for additional broad-based analyses that incorporate a diversity of lipotyphlan taxa, and in particular the fossil forms. Also, phylogenetic analyses of intrageneric relationships within Solenodon and Nesophontes may find patterns of distribution that favor one or another of the biogeographical hypotheses. And there is a need for additional relevant fossils, both from the Greater Antilles and also from areas in Central and South America that are claimed as departure points by the different biogeographical hypotheses. Finally, we note again that the nature of the relationship between Solenodon and Nesophontes remains unclear. If these two lineages are not closely related, then it is quite possible that their Antillean distributions may be explained by different mechanisms.

ACKNOWLEDGMENTS We thank Ross MacPhee, Malcolm McKenna, Jennifer White, and Charles Woods for discussions of Caribbean biogeography and in particular the puzzle of Solenodon and Nesophontes. Ross MacPhee and Jennifer White provided suggestions for improving the manuscript. H. P. W.’s work

250

Biogeography of the West Indies: Patterns and Perspectives

on this project was supported in part by a Research Opportunity Award supplementing NSF Grant DEB 9020002 to Ross MacPhee, and R. J. A. was assisted by NSF Grant DEB 9800908 to Callum Ross. We also thank Charles Woods for inviting us to contribute to this volume.

LITERATURE CITED Aiello, L. C. 1993. The origin of the New World monkeys. Pp. 100–118 in George, W. and R. Lavocat (eds.). The Africa – South America Connection. Clarendon Press, Oxford. Allen, G. M. 1910. Solenodon paradoxus. Memoirs of the Museum of Comparative Zoology 40:1–54 + 9 plates. Allen, G. M. 1918. Fossil mammals from Cuba. Bulletin of the Museum of Comparative Zoology 62:133–148 + 1 plate. Anderson, S. and J. K. Jones, Jr. 1984. Orders and Families of Recent Mammals of the World. John Wiley & Sons, New York. Anthony, H. E. 1916. Preliminary diagnosis of an apparently new family of insectivores. Bulletin of the American Museum of Natural History 35:725–728. Asher, R. J. 1999a. Cranial arteries in living and extinct insectivoran mammals. Journal of Vertebrate Paleontology 19:30A. Asher, R. J. 1999b. A morphological basis for assessing the phylogeny of the “Tenrecoidea” (Mammalia, Lipotyphla). Cladistics 15:231–252. Asher, R. J. 2000. Phylogenetic History of Tenrecs and Other Insectivoran Mammals. Ph.D. dissertation, State University of New York, Stony Brook. Baskin, J. A. 1986. The late Miocene radiation of Neotropical sigmodontine rodents in North America. Pp. 287–303 in Flanagan, K. M. and J. A. Lillegraven (eds.). Vertebrates, phylogeny, and philosophy. Contributions to Geology, Special Paper No. 3, University of Wyoming, Laramie. Brandt, J. F. 1833. De Solenodonte. Novo mammalium insectivorum genere. Mémoires de l’Académie des Sciences de St. Petersbourg 6:459–478. Bugge, J. 1974. The cephalic arterial system in insectivores, primates, rodents and lagomorphs, with special reference to the systematic classification. Acta Anatomica 87:1–160. Butler, P. M. 1988. Phylogeny of the insectivores. Pp. 117–141 in Benton, M. J. (ed.). The Phylogeny and Classification of the Tetrapods, Vol. 2: Mammals. Clarendon Press, Oxford. Darlington, P. J., Jr. 1957. Zoogeography: The Geographical Distribution of Animals. John Wiley & Sons, New York. Denison, R. H. 1938. The broad-skulled Pseudocreodi. Annals of the New York Academy of Sciences 37:163–256. Dobson, G. E. 1882–1890. A Monograph of the Insectivora, Systematic and Anatomical. J. Van Voorst, London. Domning, D. P., R. J. Emry, R. W. Portell, S. K. Donovan, and K. S. Schindler. 1997. Oldest West Indian land mammal: rhinocerotoid ungulate from the Eocene of Jamaica. Journal of Vertebrate Paleontology 17:638–641. Emerson, G. L., C. W. Kilpatrick, B. E. McNiff, J. Ottenwalder, and M. W. Allard. 1999. Phylogenetic relationships of the order Insectivora based on complete 12S rRNA sequences from mitochondria. Cladistics 15:221–230. Foote, M., J. P. Hunter, C. M. Janis, and J. J. Sepkoski, Jr. 1999. Evolutionary and preservational constraints on origins of biologic groups: divergence times of eutherian mammals. Science 283:1310–1314. George, W. 1993. The strange rodents of Africa and South America. Pp. 119–141 in George, W. and R. Lavocat (eds.). The Africa – South America Connection. Clarendon Press, Oxford. Gregory, W. K. 1910. The orders of mammals. Bulletin of the American Museum of Natural History 27:1–524. Hershkovitz, P. 1972. The Recent mammals of the Neotropical region: a zoogeographic and ecological review. Pp. 311–431 in Keast, A., F. C. Erk, and B. Glass (eds.). Evolution, Mammals, and Southern Continents. State University of New York Press, Albany. Hoffstetter, R. 1980. Origin and deployment of New World monkeys emphasizing the southern continents route. Pp. 103–122 in Ciochon, R. L. and A. B. Chiarelli (eds.). Evolutionary Biology of the New World Monkeys and Continental Drift. Plenum Press, New York. Hutterer, R. 1993. Order Insectivora. Pp. 69–130 in Wilson, D. E. and D. M. Reeder (eds.). Mammal Species of the World: A Taxonomic and Geographic Reference, 2nd ed. Smithsonian Institution Press, Washington, D.C.

The Origin of the Greater Antillean Insectivorans

251

Iturralde-Vinent, M. A. and R. D. E. MacPhee. 1999. Paleogeography of the Caribbean region: implications for Cenozoic biogeography. Bulletin of the American Museum of Natural History 238:1–95. Korth, W. W. 1994. The Tertiary Record of Rodents in North America. Plenum Press, New York. Lavocat, R. 1980. The implications of rodent paleontology and biogeography to the geographical sources and origin of platyrrhine primates. Pp. 93–102 in Ciochon, R. L. and A. B. Chiarelli (eds.). Evolutionary Biology of the New World Monkeys and Continental Drift. Plenum Press, New York. Liu, F.-G. R. and M. M. Miyamoto. 1999. Phylogenetic assessment of molecular and morphological data for eutherian mammals. Systematic Biology 48:54–64. Liu, F.-G. R., M. M. Miyamoto, N. P. Freire, P. Q. Ong, M. R. Tennant, T. S. Young, and K. F. Gugel. 2001. Molecular and morphological supertrees for eutherian (placental) mammals. Science 291:1786–1789. MacFadden, B. J. 1980. Rafting mammals or drifting islands? Biogeography of the Greater Antillean insectivores Nesophontes and Solenodon. Journal of Biogeography 7:11–22. MacPhee, R. D. E. 1981. Auditory regions of primates and eutherian insectivores: morphology, ontogeny, and character analysis. Contributions to Primatology 18:1–282. MacPhee, R. D. E. and D. A. Grimaldi. 1996. Mammal bones in Dominican amber. Nature 380:489–490. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1994. First Tertiary land mammal from Greater Antilles: an early Miocene sloth (Xenarthra, Megalonychidae) from Cuba. American Museum Novitates 3094:1–13. MacPhee, R. D. E. and M. A. Iturralde-Vinent. 1995. Origin of the Greater Antillean land mammal fauna, 1: New Tertiary fossils from Cuba and Puerto Rico. American Museum Novitates 3141:1–31. MacPhee, R. D. E. and A. R. Wyss. 1990. Oligo-Miocene vertebrates from Puerto Rico, with a catalog of localities. American Museum Novitates 2965:1–45. MacPhee, R. D. E., C. Flemming, D. P. Domning, R. W. Portell, and B. Beatty. 1999a. Eocene ?primate petrosal from Jamaica: morphology and biogeographical implications. Journal of Vertebrate Paleontology 19:61A. MacPhee, R. D. E., C. Flemming, and D. P. Lunde. 1999b. “Last occurrence” of the Antillean insectivoran Nesophontes: new radiometric dates and their interpretation. American Museum Novitates 3261:1–20. Madsen, O., M. Scally, C. J. Douady, D. J. Kao, R. W. DeBry, R. Adkins, H. M. Amrine, M. J. Stanhope, W. W. de Jong, and M. S. Springer. 2001. Parallel adaptive radiations in two major clades of placental mammals. Nature 409:610–614. Maier, W. 1985. Zalambdodontic teeth of mammals as a morphological paradigm. Fortschritte der Zoologie, 30:253–256. McDowell, S. B., Jr. 1958. The Greater Antillean insectivores. Bulletin of the American Museum of Natural History 115:113–214. McKenna, M. C. and S. K. Bell. 1997. Classification of Mammals above the Species Level. Columbia University Press, New York. Miller, G. S., Jr. and R. Kellogg. 1955. List of North American Recent mammals. Bulletin of the United States National Museum 205:1–954. Morgan, G. S. 1994. Late Quaternary fossil vertebrates from the Cayman Islands. Pp. 465–508 in Brunt, M. A. and J. E. Davies (eds.). The Cayman Islands: Natural History and Biogeography. Kluwer Academic, Boston. Morgan, G. S. and J. A. Ottenwalder. 1993. A new extinct species of Solenodon (Mammalia: Insectivora: Solenodontidae) from the late Quaternary of Cuba. Annals of Carnegie Museum 62:151–164. Morgan, G. S. and C. A. Woods. 1986. Extinction and the zoogeography of West Indian land mammals. Biological Journal of the Linnean Society 28:167–203. Mouchaty, S. K. 1999. Mammalian Molecular Systematics with Emphasis on the Insectivore Order Lipotyphla. Ph.D. dissertation, Lund University, Sweden. Mouchaty, S. K., A. Gullberg, A. Janke, and U. Arnason. 2000. The phylogenetic position of the Talpidae within Eutheria based on analysis of complete mitochondrial sequences. Molecular Biology and Evolution 17:60–67. Mullins, H. T., A. F. Gardulski, S. W. Wise, Jr., and J. Applegate. 1987. Middle Miocene oceanographic event in the eastern Gulf of Mexico: implications for seismic stratigraphic succession and Loop Current/Gulf Stream circulation. Geological Society of America Bulletin 98:702–713. Murphy, W. J., E. Eizirik, W. E. Johnson, Y. P. Zhang, O. A. Ryder, and S. J. O’Brien. 2001. Molecular phylogenetics and the origins of placental mammals. Nature 409:614–618. Nowak, R. M. and J. L. Paradiso. 1983. Walker’s mammals of the world, 4th ed., Vol. 1. The Johns Hopkins University Press, Baltimore, Maryland.

252

Biogeography of the West Indies: Patterns and Perspectives

Patterson, B. 1958. Affinities of the Patagonian fossil mammal Necrolestes. Breviora 94:1–14. Patterson, B. 1962. An extinct solenodontid insectivore from Hispaniola. Breviora 165:1–11. Peters, W. 1863. Über die Säugethier-Gattung Solenodon. Abhandlungen der Königl. Akademie der Wissenschaften, Berlin. Presley, R. 1979. The primitive course of the internal carotid artery in mammals. Acta Anatomica 103:238–244. Robinson, E. 1994. Jamaica. Pp. 111–127 in Donovan, S. K. and T. A. Jackson (eds.). Caribbean Geology: An Introduction. University of the West Indies Publishers’ Association, Kingston, Jamaica. Rosen, D. E. 1975. A vicariance model of Caribbean biogeography. Systematic Zoology 24:431–464. Scott, W. B. 1905. Reports of the Princeton University Expedition to Patagonia, 1896–1899, Pp. 365–388 in Vol. 5: Paleontology etc., pt. 2. Insectivora. Princeton University, Princeton, New Jersey. Simpson, G. G. 1945. The principles of classification and a classification of mammals. Bulletin of the American Museum of Natural History 85:1–350. Simpson, G. G. 1956. Zoogeography of West Indian land mammals. American Museum Novitates 1759:1–28. Springer, M. S., G. C. Cleven, O. Madsen, W. W. de Jong, V. G. Waddell, H. M. Amrine, and M. J. Stanhope. 1997. Endemic African mammals shake the phylogenetic tree. Nature 388:61–64. Stanhope, M. J., V. G. Waddell, O. Madsen, W. de Jong, S. B. Hedges, G. C. Cleven, D. Kao, and M. S. Springer. 1998. Molecular evidence for multiple origins of Insectivora and for a new order of endemic African insectivore mammals. Proceedings of the National Academy of Sciences 95:9967–9972. van Dijk, M. A. M., O. Madsen, F. Catzeflis, M. J. Stanhope, W. W. de Jong, and M. Pagel. 2001. Protein sequence signatures support the African clade of mammals. Proceedings of the National Academy of Sciences 98:188–193. Van Valen, L. 1967. New Paleocene insectivores and insectivore classification. Bulletin of the American Museum of Natural History 135:217–284. Wible, J. R. 1984. The Ontogeny and Phylogeny of the Mammalian Cranial Arterial Pattern (Internal Carotid Artery). Ph.D. dissertation, Duke University, Raleigh, North Carolina. Wible, J. R. 1987. The eutherian stapedial artery: character analysis and implications for superordinal relationships. Zoological Journal of the Linnean Society 91:107–135. Williams, E. E. 1989. Old problems and new opportunities in West Indian biogeography. Pp. 1–46 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida. Winge, H. 1941. The interrelationships of the mammalian genera, Vol. 1: Monotremata, Marsupialia, Insectivora, Chiroptera, Edentata. C. A. Reitzels Forlag, Copenhagen, Denmark. Woods, C. A. 1989. The biogeography of West Indian rodents. Pp. 741–798 in Woods, C. A. (ed.). Biogeography of the West Indies: Past, Present, and Future. Sandhill Crane Press, Gainesville, Florida.

and Biogeography 16 Systematics of the West Indian Genus Solenodon Jose A. Ottenwalder Abstract — The study of the geographical variation of Solenodon indicates that this Greater Antillean insectivore genus is represented by four species: two living, S. cubanus from Cuba and S. paradoxus from Hispaniola, and two extinct, S. marcanoi from Hispaniola and S. arredondoi from Cuba. A new geographical population of S. paradoxus from southern Hispaniola is described. The diagnosis of S. marcanoi is revised and specimens of the original type series are re-assigned to S. paradoxus. Solenodon marcanoi shares characters of both S. paradoxus and S. cubanus, and is considered an intermediate lineage. The two extinct species were restricted, respectively, to western Cuba and to southwestern Hispaniola. It is hypothesized that elucidation of both S. marcanoi relationships and southern Hispaniola paleogeographical reconstruction might hold a key role in the interpretation of the biogeographical history of Antillean insectivores. The late Quaternary distribution of S. paradoxus in Hispaniola is insufficiently known. The discovery of new extant populations in the Dominican Republic is presented. Results of recent and previous field surveys indicate that the species is still widely dispersed in this country, but extant populations are fragmented in distribution and low in numbers. In Haiti, the species appears to survive only in the Massif de la Hotte, in the southwestern end of Peninsula of Tiburon. No single fossil or Recent records are yet known from northern Haiti, north of the Cul-deSac. Paleontological and archaeological evidence suggest that S. cubanus was well distributed throughout western and eastern Cuba. The species is now extirpated in the western and central portions of the island, and only survives in the eastern mountain areas.

INTRODUCTION The West Indian insectivores Solenodon and Nesophontes are endemic to the Greater Antilles and probably the most ancient members of the West Indian mammalian fauna. The genus Solenodon is restricted to the islands of Cuba and Hispaniola and contains the only surviving members of the Insectivora in the region. The closely related Nesophontes comprises eight extinct species known from the Holocene and late Pleistocene of Cuba, Hispaniola, the Cayman Islands, and Puerto Rico (Figure 1). The Puerto Rican species, N. edithae, which is intermediate in size between the larger Solenodon and the much smaller remaining species of Nesophontes, has been discovered in a kitchen midden in Vieques Island (Morgan and Woods, 1986; E. Wing, personal communication). Although they have fared better than other groups such as edentates and primates, insectivores have also suffered a high extinction rate recorded among other West Indian mammals (Morgan and Woods, 1986; Woods, 1989, 1990). Two of twelve species of insectivores have survived until today. Pleistocene climatic events, human exploitation, and predation pressure from exotics have been indicated as major causes of extinction of the Antillean vertebrate fauna. Increasing support for the latter two factors have been presented (Steadman et al., 1984; Woods et al., 1986; Woods, 1989). Most West Indian mammals were still extant at the time Amerindians arrived on the islands (Morgan and Woods, 1986). Association of Nesophontes with rats in cave deposits led Miller (1929) to suggest all three species of Hispaniolan Nesophontes might have survived to the beginning of this century. Both Nesophontes and Solenodon have been found in archaeological sites throughout their 0-8493-2001-1/01/$0.00+$1.50 © 2001 by CRC Press LLC

253

254

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 1 Map of the West Indies showing distribution of Greater Antillean insectivores. Shaded areas = Solenodon and Nesophontes;  = Nesophontes.

historical ranges. Their presence in archaeological deposits is, however, virtually insignificant compared to the abundance of other groups, and they do not seem to have represented an important source of human food. Evidence from cave deposits and owl pellet accumulations indicate that Nesophontes were clearly very abundant, but not Solenodon. Therefore, Nesophontes might have been somewhat neglected as food by the Amerindians because of their small body size. Although Solenodon species were much larger, they do not seem to be more common in archaeological sites, and they are found infrequently in cave deposits. Today, Cuban and Hispaniolan Solenodon are among the few native West Indian land mammals that still survive. They have been considered among the most endangered mammals, and probably are the most threatened of all insectivores (Thornback and Jenkins, 1982; Thornback, 1983).

EVOLUTIONARY RELATIONSHIPS OF WEST INDIAN INSECTIVORES The early evolutionary history of the group was comprehensively summarized by McDowell (1958). The relationships of Solenodon and Nesophontes, among themselves and within the Insectivora, are not yet well understood. This is, in part, a reflection of the continuing problems of insectivoran classification. The group, which continues to be found among the least understood mammalian orders, has been an assemblage classically regarded as stem eutherians. In fact, the group was long considered to include elephant shrews (Macroscelididae), tree shrews (Tupaiidae), many early Tertiary mammals (Gregory, 1910), and to be related to primates (Szalay, 1975; Novacek, 1982). These conclusions were made mainly due to shared primitive resemblance, and for which the Insectivora have been regarded as a “taxonomic wastebasket” (McKenna, 1975) and “Eutheria incertae sedis” (Novacek, 1990). Following the exclusion of the Menotyphla (Butler, 1972; McKenna, 1975; Novacek et al., 1983), the Insectivora was restricted to the Lipotyphla or Recent insectivores (Butler, 1988).

Systematics and Biogeography of the West Indian Genus Solenodon

255

Separation of lipotyphlans (Insectivora sensu stricto) from tupaiid insectivores (Scandentia) (Miyamoto and Goodman, 1986), supported the traditional views for Lipotyphla to be regarded as monophyletic comprising two clades of subordinal rank: the Erinaceomorpha (hedgehogs) and the Soricomorpha (the other five families) (Saban, 1954; Butler, 1956, 1972; McKenna, 1975). Within the Soricomorpha, Butler (1988) suggested that golden moles and tenrecs form a clade and that moles and shrews cluster together, followed by solenodons. MacPhee and Novacek (1993), however, proposed three clades of subordinal rank: Soricomorpha (Soricidae, Talpidae, Tenrecidae, and Solenodontidae); Erinaceomorpha (Erinaceidae); and Chrysochloromorpha (Chrysochloridae). Rejecting a monophyletic Insectivora, Springer et al. (1997) concluded that golden moles are not part of the Insectivora, but instead belong to an “African clade” that includes hyraxes, elephants, sirenians, aardvarks, and elephant shrews. In a second paper (Stanhope et al., 1998), the same group of researchers not only confirmed earlier findings in Springer et al., but also proposed a new partitioning of Insectivora, placing golden moles and tenrecs in a new order within an African superordinal clade, and out of the classical order Insectivora (Soricomorpha), with no association with solenodons, moles, and shrews. In addition, these authors suggested that the hedgehogs, shrews, moles, and solenodons form a monophyletic group to be retained in the order Insectivora. Another recent study (Emerson et al., 1999) argued that molecular and morphological data are currently in conflict over the possible monophyly of the living members of the Insectivora (Lipotyphla sensu Butler, 1988), and that the relationships within the group remain largely unresolved, since available data are insufficient and current evidence is as yet inconclusive. Butler (1956) and McDowell (1958) included Solenodon and Nesophontes in the Soricomorpha, together with the living shrews (Soricidae), moles (Talpidae), tenrecs (Tenrecidae), and chrysocholorids (Chrysochloridae), as well as several fossil taxa, such as apternodontids (Apternodontidae) and geolabidids (Geolabididae). As such, Nesophontes and Solenodon are often considered to represent a monophyletic group derived from Eocene or Oligocene North American soricomorphs belonging to either the Apternodontidae or the Geolabididae (Matthew, 1910, 1918; Schlaikjer, 1934; Van Valen, 1967; Butler, 1972; McKenne, 1975; MacFadden, 1980; Lillegraven et al., 1981). They may have reached the Greater Antilles in the early Tertiary, either through vicariance by way of a proto-Antillean archipelago (MacFadden, 1980) or by dispersal from nuclear Central America. Van Valen (1967) considered the apternodontids as possibly ancestral to all of the extant zalambdodont lipotyphlans: solenodons, tenrecids, and chrysochlorids. However, whether the zalambdodont condition of the dentition (triangular upper molar teeth with V-shaped cusps and prominent outer styles) in these groups is homologous or convergent is a problem that as yet remains unsolved. McDowell (1958) rejected any special relationships between Antillean insectivores and apternodontids and, based on cranial similarities, suggested closer affinity between Solenodon and Nesophontes within the Soricidae than to any other soricomorph insectivore. He concluded this despite the fact that Nesophontes has a fully dilambdodont dentition (upper molar teeth with W-shaped cusps). However, Van Valen (1966) has suggested the possibility that Nesophontes may be secondarily dilambdodont. In the opinion of McKenna (1975), McDowell’s conclusions reflected a small sample and poor preservation of the material then available. Although unable to separate ancestral from derived characters, McDowell’s work represents to date the most serious attempt to clarify the affinities of the West Indian insectivores. More recently, Butler (1988) suggested the possibility that Centetodon (Geolabididae), Solenodon, and Nesophontes had a common ancestor and that Solenodon is probably not especially related to either Apternodus or to the Soricidae. Solenodon may be the only survivor of a North American branch that includes Centetodon, Nesophontes, and possibly Apternodus. In short, one Solenodontidae (McDowell, 1958; Findley, 1967; Yates, 1984) or two families, Solenodontidae and Nesophontidae (Hall, 1981; Honacki et al., 1982), have been recognized. I follow the latter arrangement in this discussion and treat West Indian Insectivores in two distinct families.

256

Biogeography of the West Indies: Patterns and Perspectives

HISTORICAL SURVEYS OF THE SOLENODONTIDAE The genus Solenodon was described in 1833 by Brandt from a single Hispaniolan specimen with an incomplete skull. Although the existence of a solenodontid in Cuba was discovered in 1836 (Poey, 1851), the animal was considered conspecific with the type from Hispaniola, S. paradoxus (Poey, 1851), until it was finally named (Peters, 1861) and critically described as a distinct species, S. cubanus, 27 years later (Peters, 1863). Whereas Peters’ separation of the two species in the same genus was generally adopted (Gundlach, 1866–1867, 1872, 1877, 1895; Dobson, 1884; True, [1884] 1885; Flower and Lydekker, 1891; Elliot, 1905; Leche, 1907; Allen, 1908, 1911; Beddard, 1909; Gregory, 1910; Miller, 1924; Webber, 1928). Dobson (1882) considered both species to represent geographical forms of one species. Disagreement concerning their generic status arose thereafter. Allen (1908) pointed out that certain characters were different enough to justify subgeneric condition, whereas Cabrera (1925) created the genus Atopogale for the Cuban species. With few exceptions (Miller and Kellogg, 1955; Hall and Kelson, 1959; Findley, 1967), most authors disregarded Cabrera’s criteria, recognizing but a single genus for the two species, and either relegating Atopogale to subgenus (Aguayo, 1950; Arredondo, 1955; Moreno, 1966; Cave, 1968; Varona, 1974; Hall, 1981; Nowak and Paradiso, 1983) or simply considering it a synonym of Solenodon (Winge, 1941; Allen, 1942; Simpson, 1945, 1956; Westermann, 1953; Vrydagh, 1954; Eisenberg and Gould, 1966; Eisenberg, 1975; Walker, 1975; Kowalski, 1976; Paula Couto, 1979; Lawlor, 1979; Corbet and Hill, 1980; Honacki et al., 1982; Yates, 1984). The validity of Atopogale was discussed by Podushcka and Podushcka (1983). Essentially, their conclusions agree with the placement of the Cuban form under Solenodon as used by most authors since the description of cubanus last century. In their evaluation of Cabrera’s characters, these authors also expressed serious doubts concerning the consistency of most characters accepted until now to distinguish Cuban from Hispaniolan solenodons. A second form of Solenodon from the northeastern mountainous region of Cuba, S. poeyanus (Barbour, 1944), was described exclusively based on external characters (coloration and claw length) of a single specimen. Aguayo (1950) and Koopman and Ruibal (1955) argued that at most this proposed form be considered as a subspecies. In agreement with these authors, Patterson (1962) expressed doubts of the validity of poeyanus beyond subspecies level (see also Arredondo, 1970a), if any, whereas Varona (1974) stated that this proposed form cannot be separated from cubanus even at subspecific rank. Some authors, however, have retained Atopogale as subgenus (Hall, 1981; Nowak, 1991), and poeyanus as a distinct geographical population (Hall, 1981). A new genus and species of a somewhat smaller solenodontid, Antillogale marcanoi, was described from late Pleistocene to Recent fossil deposits of the Dominican Republic (Patterson, 1962). But the generic validity of Antillogale was questioned by Van Valen (1967) and relegated to subgenus by Varona (1974) who placed marcanoi under Solenodon. The existence of another extinct species of Solenodon was mentioned by Arredondo (1970a), based on a femur from a late Quaternary fossil site in western Cuba. Morgan et al. (1980) illustrated and described this femur and mentioned the existence of two additional large fossil femora from western Cuba. The discovery in 1991 by Oscar Arredondo of a large partial skull in the Museo Nacional de Historia Natural de Cuba, in La Habana, allowed for detailed taxonomic comparisons of the new material with all three nominated species of Solenodon. These comparisons established the much larger extinct Cuban solenodontid as a distinct species, clearly distinguishable from any of the species already described for the genus, living or extinct (Ottenwalder, 1991), and later described as S. arredondoi by Morgan and Ottenwalder (1993). Despite the varied views and proposals concerning the taxonomic status of the different nominated species and genera in the literature, the group has not been the subject of systematic revision. In part, taxonomic studies might have been prevented in the past due to paucity of Solenodon material in collections. Furthermore, the majority of the specimens available in systematic collections until now were collected in the beginning of the century and lack adequate collecting data.

Systematics and Biogeography of the West Indian Genus Solenodon

257

During this research, new Hispaniolan material was obtained from the Dominican Republic, including the fresh remains of several specimens of very small body size. These specimens are smaller than the known Hispaniolan species, S. paradoxus, and, in fact, resemble in size the animal described by Patterson (1962) as Antillogale marcanoi. This has led some authors (Woods and Einsenberg, 1989) to suggest that A. marcanoi, thus far assumed to be extinct, appears to be alive. During the past 15 years, a number of Hispaniolan specimens have also been secured from the Massif de la Hotte, in the southwestern end of Haiti (Woods, 1986). Both fossil and Recent Solenodon specimens are represented in this material, including four skulls of S. marcanoi, until now known only from partial mandibles and limb bones.

MATERIALS AND METHODS A total of 247 Recent specimens was examined. Specimens were conventional museum specimens preserved as skins, skulls, skeletons, fluid, and/or taxidermy mounted specimens. These specimens are deposited in the following collections of Recent mammals: American Museum of Natural History, New York (AMNH); Carnegie Museum of Natural History, Pittsburgh (CM); Field Museum of Natural History, Chicago (FMNH); Florida Museum of Natural History, University of Florida, Gainesville (UF); Instituto de Ecología y Sistemática, Academia de Ciencias de Cuba, La Habana (IES/ACC); Institut Royal des Sciences Naturelles de Belgique, Brussels (IRSB); Jose A. Ottenwalder private field collections, Santo Domingo (JAO); National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM); Museo Nacional de Historia Natural de Cuba, La Habana (MNHNC); Museum of Comparative Zoology, Harvard University, Cambridge (MCZ); Yale Peabody Museum Osteological Collection, Yale University, New Haven (YPM); Puget Sound Museum of Natural History, University of Puget Sound, Tacoma (PSM); Rijksmuseum van Natuurlijke Historie, Leiden (RMNH); Zoologisches Institut und Zoologisches Museum, Universität Hamburg (ZMUH). In this chapter, specimens will be referred to by their collection acronyms. All specimens were assigned to three age classes; Age I, juvenile; Age II, subadult; and Age III, adult. Age was established based on tooth wear and the following criteria: Juvenile — last cheektooth is not fully erupted; temporal ridges not joined to form a sagittal crest; lambdoidal crest is not well defined; basioccipital and basisphenoid are not fused. Subadult — all cheekteeth are fully erupted; temporal ridges are joined to form a weakly developed sagittal crest; lambdoidal crest is not well developed; basioccipital and basisphenoid are not completely fused; maxilla and pre-maxilla are not completely fused. Adult — all cheekteeth are fully erupted; sagittal and lambdoidal crests are well defined; basioccipital and basisphenoid are completely fused; maxilla and pre-maxilla are completely fused; traces of labial reentrant angles are usually gone. Older adults often have a more massive cranium, with more pronounced sagittal and lambdoidal crests, interorbital region, and occipital region. Five external measurements (total length, TL; head–body length, HBL; tail length, TL; ear length, EA; and hind foot length, HF) were taken directly from live animals (Dominican Republic only) and museum specimens preserved in fluid. External measurements were also obtained from labels of specimens preserved as standard museum skins, and, in the case of missing types or otherwise unavailable critical specimens, from the literature. Fifty-eight (58) cranial, dental, and postcranial measurements divided into lengths (L), breadths or widths (B), and heights or depths (H) were taken. All internal measurements were taken with dial calipers to the nearest 0.05 mm. Needlepoint dial calipers were utilized in dental measurements. There is disagreement concerning the missing premolar of Solenodon, as to whether it is the P2/2 or the P3/3. The criteria of McDowell (1958), who tentatively regarded the missing premolar as the P3/3, are followed here.

258

Biogeography of the West Indies: Patterns and Perspectives

All measurements are given in millimeters. Definitions of internal measurements and their abbreviations are given below. GLS: Greatest length of skull — Greatest distance between the posteriormost part of the skull above the foramen magnum (supraoccipital processes) and the anteriormost part of the premaxilla CBL: Condylobasal length — Greatest distance from the posteriormost part of the exoccipital condyles to the anteriormost part of the premaxillary PL: Palatilar length — Greatest distance from the anteriormost point on the border of the palate to a line connecting the posteriormost margins of the alveoli of the upper incisors PPL: Postpalatal length — Greatest distance from the anteriormost margin of the foramen magnum to the posterior border of the palate AMTR: Alveolar length of upper molar toothrow — Least distance from posterior point of alveolar margin of last molar to anterior point of alveolar margin of first molar MMTR: Length of upper molar toothrow — Least distance measured at the crowns LMTR: Length of maxillary toothrow — Least distance between the anteriormost and posteriormost margins of the alveoli of the maxillary teeth (C1-M3) MTRW: Breadth across maxillary toothrow — Least width of palate (from M1M2 to M1M2) taken at the labial margins of each toothrow AC: Anteorbital constriction — Least distance between lower anteriormost part of the fossae, taken over the opening of the canale infraorbitale ZB: Zygomatic breadth — Greatest width across zygomatic arches, measured at right angles to the longitudinal angles of cranium IC: Interorbital constriction — Least width across postorbital constriction, measured between the orbits at right angles to the long axis of the cranium SB: Squamosal breadth — Least width across the lateral margins of the squamosal bones, measured at right angles to the long axis of the cranium MB: Mastoid breadth — Greatest width across the mastoid processes, measured at right angles to the long axis of the cranium BB: Breadth of the braincase — Greatest width across braincase, measured at right angles to the long axis of the cranium CB: Condylar breadth — Greatest width across external margin of occipital condyle SH: Skull height — Perpendicular distance from a plane going through the most inferior part of the post-glenoid processes, to the highest point on cranium LC1: Maximum length of C1 WC1: Maximum width of C1 LP1: Maximum length of P1 WP1: Maximum width of P1 LP2: Maximum length of P2 WP2: Maximum width of P2 LP4: Maximum length of P4 WP4: Maximum width of P4 LM1: Maximum length of M1 WM1: Maximum width of M1 LM2: Maximum length of M2 WM2: Maximum width of M2 LM3: Maximum length of M3 WM3: Maximum width of M3 GML: Greatest mandible length — Least distance from most posterior part of condyle to anterior (lowest) point of the first incisor at its alveolus (= tip of the dentary)

Systematics and Biogeography of the West Indian Genus Solenodon

259

MTR: Mandibular toothrow — Length from anterior edge of alveolus of canine to posterior edge of alveolus of last molar P4-M3: Alveolar length of P4-M3 — Posterior point of alveolar margin of last molar to anterior point of alveolar margin of last premolar DCP: Depth through coronoid process — Least vertical height between tip of coronoid process to highest edge of lunate notch ACH: Angular-condylar height — Least distance from lowest point on angular process to highest point on condyle LC1: Maximum length of C1 WC1: Maximum width of C1 LP1: Maximum length of P1 WP1: Maximum width of P1 LP2: Maximum length of P2 WP2: Maximum width of P2 LP4: Maximum length of P4 WP4: Maximum width of P4 LM1: Maximum length of M1 WM1: Maximum width of M1 LM2: Maximum length of M2 WM2: Maximum width of M2 LM3: Maximum length of M3 WM3: Maximum width of M3 LF: Maximum length of femur MWF: Maximum width of femur, at proximal end FMW: Minimum shaft width of femur LH: Maximum length of humerus MWH: Maximum width of humerus, at distal end HMW: Minimum shaft width of humerus LU: Maximum length of ulna MWU: Maximum width of ulna, at olecranon UMW: Minimum shaft width of ulna, at lower section of diaphisis Specimens of Recent Solenodon were grouped into seven reference samples throughout the geographical range of the genus (Figure 2) as follows: (1) Peninsula de Samana–Promontorio de Cabrera, northeastern Dominican Republic (North Hispaniola); (2) Los Haitises–Sierra de Seibo–Caribbean Coastal Plain, eastern Dominican Republic (North Hispaniola); (3) Cordillera Central–Cibao Occidental Valley, central north–central Dominican Republic (North Hispaniola); (4) Peninsula de Barahona, southwestern Dominican Republic (South Hispaniola); (5) Sierra de Baoruco, southwestern Dominican Republic (South Hispaniola); (6) Massif de la Hotte, southwestern Haiti (South Hispaniola); (7) Eastern Cuba, including both the southern (Sierra Maestra) and northern ranges (Sierra de Nipe, Sierra del Cristal, Cuchillas de Moa, Toa, and Baracoa). A total of 110 specimens of the late Pleistocene, early Holocene, and Amerindian times from Cuba, Dominican Republic, and Haiti were examined. These fossil, subfossil, and kitchen midden specimens are housed at the following collections: Carnegie Museum of Natural History, Pittsburgh (CM); Florida Museum of Natural History, University of Florida (UF); Instituto de Ecología y Sistemática, Academia de Ciencias de Cuba, La Habana (IES/ACC); Museo Nacional Historia Natural, La Habana (MNHNC); Museum of Comparative Zoology, Harvard University (MCZ); National Museum of Natural History, Smithsonian Institution (USNM); Oscar Arredondo private collection, La Habana (OA). The sites from where these specimens were recovered are discussed in the section “Late Quaternary and Recent Distribution of Solenodon.”

260

Biogeography of the West Indies: Patterns and Perspectives

FIGURE 2 Map showing geographical samples of extant Solenodon in Cuba and Hispaniola. (1) Promontorio de Cabrera–Peninsula de Samana, northeastern Dominican Republic (North Hispaniola); (2) Los Haitises–Sierra de Seibo–Caribbean Coastal Plain, eastern Dominican Republic (North Hispaniola); (3) Cordillera Central–Cibao Occidental Valley, central north-central Dominican Republic (North Hispaniola); (4) Peninsula de Barahona, southwestern Dominican Republic (South Hispaniola); (5) Sierra de Baoruco, southwestern Dominican Republic (South Hispaniola); (6) Massif de la Hotte, southwestern Haiti (South Hispaniola); (7) Eastern Cuba.

Statistical analyses were performed using the NCSS Statistical System (Version 5.0), and the Statistical Analysis System (SAS Institute, 1985). Descriptive statistics (mean, range, standard deviation, standard error, variance, and coefficient of variation) were calculated with the MEANS routine. Univariate analyses of variation with age, individual variation, secondary sexual variation, and geographical variation were performed using a single classification analysis of variance (ANOVA). The specimens from central Hispaniola, the largest sample available, were selected to study the influence of variation of age and sex on the populations. Although small, the sample from Eastern Cuba was also tested for secondary sexual variation, but analysis of variation with age in this population was prevented by insufficient sample size. The General Linear Model (GLM-ANOVA) was used to test for significant differences among or between means for each character. Subsequently, a Duncan’s Multiple Range Test was used to determine maximally nonsignificant subsets, if means were found to be significantly different. Because solenodons are very rare in collections and endangered in the wild, samples of Recent specimens available for examination are limited in number. Furthermore, most subfossil specimens had missing measurements. To maximize sample size, characters were analyzed separately in three data sets (cranial, mandibular, and limb bones) to assess multivariate relationships. The multivariate technique used was discriminant function analysis. Stepwise discriminant analysis performs a multiple discriminant analysis in a stepwise manner, selecting the variable entered by finding the variable with the greatest F value. The F value for inclusion was set at 0.01, and the F value for deletion was set at 0.05. The program also classifies individuals, placing them with the group to which they are nearest on the discriminant functions. The weight of five cranial characters was evaluated for diagnostic consistency in separating S. paradoxus from S. cubanus, and for usefulness in assessing specific relationships among and between known Solenodon species. A total of 115 specimens of living and extinct Solenodon representing all three nominated taxa plus the undescribed skull of a suspected distinct species were individually examined. Characters were not polarized but treated as having equal weight. The following character states were evaluated for analysis:

Systematics and Biogeography of the West Indian Genus Solenodon

261

Character 1. Para-nasal (os proboscis) bone support: 0 = absent; 1 = present. Character 2. Diastema I3-C1: 0 = absent; 1 = present. Character 3. Accessory cusp C1: 0 = absent; 1 = present; 2 = vestigial. Character 4. Shape P2: 0 = triangular; 1 = simple, oval or conical; 2 = intermediate. Character 5. Mesopterygoig fossa: 0 = wider posteriorly than anteriorly; 1 = wider anteriorly than posteriorly; 2 = parallel. Cranial measurements, collecting data, or photographs of 53 additional specimens also were examined. These data were not included in the statistical analysis. These specimens are found in the following mammal collections: Museum of Zoology, University of Michigan (UMMZ); British Museum (Natural History), London (BMNH); Forschungsinstitut und Natur-Museum Senckemberg, Frankfurt (SMF); Max-Planck-Institut für Hirnforschung, Frankfurt (MPIH); Naturhistorisches Museum Wien, Wien (NMW); Naturhistoriska Riksmusset, Swedish Museum of Natural History, Stockholm (NRM); University Museum of Zoology, Cambridge University, U.K. (UMZC); Zoological Museum, Institute of Taxonomic Zoology, University of Amsterdam, Amsterdam (ZMA).

RESULTS NONGEOGRAPHICAL VARIATION Three kinds of nongeographical variation were investigated: variation with age, secondary sexual variation, and individual variation. Variation with Age Age categories used in this study are referred to as Age I, juveniles; Age II, subadults; and Age III, adults. These categories are based on the criteria described above (see Materials and Methods) and on dental wear, and do not reflect reproductive age. The influence of age was tested using GLM-ANOVA. Because of insufficient sample, the Cuban population was not tested for age variation. In the sample from Hispaniola, adults, subadults, and juveniles form nonoverlapping subsets in only 3 out of 41 measurements (zygomatic breadth, maximum width of P4, minimum shaft width of humerus). Adults and subadults form an overlapping subset that differs significantly from the juvenile subset in 30 measurements. Adults averaged the largest in most measurements, except in 16, in which subadults were slightly larger. Nevertheless, only adult specimens were used in subsequent analyses. Variation with age is discussed in more detail in Ottenwalder (1991). Secondary Sexual Variation Forty-one cranial and postcranial measurements of adult males of two samples (Eastern Cuba and Central Hispaniola) were tested against those of adult females, utilizing GLM-ANOVA, to establish the existence of any significant differences in size between the sexes. The results are shown in Table 1. Although females averaged larger than males in most measurements, no significant (p < 0.05) differences were observed between males and females of S. cubanus in any of the internal and most external measurements tested. Only in hind foot length were females different from males in the Cuban sample. In the sample from the Cordillera Central–Cibao Occidental Valley region, in central Dominican Republic, females proved significantly larger than males in only two measurements (breadth across maxillary toothrow and anteorbital constriction). As in the Cuban Solenodon sample, females from Hispaniola were also somewhat larger than males in most measurements, but again, variation in size between the sexes was only slightly different. For instance, all measurements of

262

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 Secondary Sexual Variation in Cranial, Dental, and Postcranial Measurements of Recent Samples of Solenodon from Central Dominican Republic, Hispaniola (Sample 3) and Eastern Cuba (Sample 7) Sample

Sex

N

Mean ± SD

Hispaniola

M F M F

21 27 5 4

Greatest length of skull 86.1 ± 2.92 81.0–91.5 86.8 ± 1.91 82.6–90.9 77.7 ± 4.12 71.4–82.4 78.8 ± 3.03 75.5–82.6

2.2 3.4 5.3 3.9

M F M F

20 26 5 4

Condylobasal length 81.2 ± 2.88 76.0–86.6 81.0 ± 1.68 77.2–84.6 73.0 ± 3.33 68.5–77.1 75.0 ± 2.59 71.8–77.8

2.1 3.6 4.6 3.5

M F M F

21 27 5 5

37.6 37.4 33.9 34.5

Palatal length ± 1.42 34.8–40.4 ± 0.91 35.9–39.2 ± 1.43 31.8–35.7 ± 0.94 33.2–35.6

2.4 3.8 4.2 2.7

M F M F

20 25 5 4

Postpalatal 30.3 ± 1.18 30.7 ± 0.86 26.9 ± 1.51 27.6 ± 1.36

Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

M F M F

Range

length 27.6–32.3 29.0–32.6 25.0–28.8 26.0–29.3

Alveolar length of upper 20 10.3 ± 0.72 25 10.4 ± 0.66 5 8.0 ± 0.64 5 8.1 ± 0.28

CV

2.8 3.9 5.6 5.0

molar toothrow 9.1–12.4 6.3 9.5–12.9 7.1 7.3–8.8 8.1 7.6–8.4 3.5

M F M F

Length of upper molar toothrow 20 10.9 ± 0.49 9.9–11.8 24 11.2 ± 0.42 10.3–12.0 5 8.7 ± 0.62 8.0–9.6 5 8.6 ± 0.34 8.3–9.2

3.7 4.5 7.1 3.9

M F M F

21 24 5 5

Length of maxillary toothrow 26.4 ± 0.79 24.9–27.6 26.3 ± 0.60 25.2–27.3 23.5 ± 1.21 21.7–24.7 23.7 ± 0.76 22.6–24.5

2.3 3.0 5.2 3.2

M F M F

Breadth across maxillary toothrow 20 23.6 ± 0.79 22.1–25.6 3.3 25 24.2 ± 0.81 22.5–25.9 3.4 5 21.5 ± 1.44 20.3–23.9 6.7 5 21.2 ± 1.15 20.4–23.3 5.4

F

P

0.83

0.36

0.21

0.67

0.08

0.78

0.96

0.36

0.41

0.52

0.55

0.48

2.00

0.16

0.51

0.50

0.88

0.35

0.03

0.88

3.35

0.07

0.26

0.62

0.07

0.79

0.15

0.71

5.10

0.02*

0.08

0.79

Systematics and Biogeography of the West Indian Genus Solenodon

263

TABLE 1 (continued) Secondary Sexual Variation in Cranial, Dental, and Postcranial Measurements of Recent Samples of Solenodon from Central Dominican Republic, Hispaniola (Sample 3) and Eastern Cuba (Sample 7) Sample

Sex

N

Hispaniola

M F M F

M F M F

Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Mean ± SD

Range

CV

5 4 5 4

Maximum length of C1 4.6 ± 0.38 4.3–5.1 4.4 ± 0.10 4.3–4.6 4.5 ± 0.26 4.1–4.7 4.5 ± 0.33 4.1–4.8

2.3 8.2 5.8 7.4

5 4 5 4

Maximum width of C1 2.4 ± 0.10 2.3–2.6 2.4 ± 0.58 2.4–2.5 2.9 ± 0.24 2.6–3.2 2.9 ± 0.11 2.8–3.1

2.4 4.4 8.1 3.7

of WM3 6.0–7.1 6.1–7.5 4.6–5.1 4.3–5.3

5.0 4.3 4.4 7.9

M F M F

20 26 5 5

Maximum width 6.6 ± 0.28 6.7 ± 0.33 4.8 ± 0.21 4.7 ± 0.37

M F M F

21 27 5 5

Anteorbital constriction 14.0 ± 0.68 13.2–15.9 14.5 ± 0.50 13.5–15.5 14.9 ± 0.67 14.3–15.9 15.1 ± 0.79 14.5–16.4

3.5 4.9 4.5 5.3

M F M F

16 21 3 4

Zygomatic breadth 34.3 ± 1.62 31.8–37.4 35.0 ± 1.70 32.0–39.0 31.4 ± 0.74 30.5–31.9 32.5 ± 0.71 31.7–33.4

4.9 4.7 2.4 2.2

M F M F

21 27 5 5

Interorbital constriction 14.9 ± 0.60 13.9–16.5 14.8 ± 0.57 13.6–16.3 15.0 ± 0.80 14.4–16.4 15.4 ± 0.51 15.0–16.3

3.9 4.1 5.3 3.3

M F M F

21 27 5 4

Squamosal breadth 32.2 ± 1.34 30.1–34.5 32.2 ± 1.20 29.2–34.0 30.8 ± 0.76 29.8–31.9 30.9 ± 0.99 29.5–31.8

3.7 4.2 2.5 3.2

M F M F

20 27 5 4

Mastoid breadth 25.9 ± 1.01 24.0–27.6 26.0 ± 1.01 23.7–28.4 24.5 ± 0.54 23.8–25.0 24.8 ± 0.30 24.4–25.1

3.9 3.9 2.2 1.2

F

P

1.02

0.34

0.00

0.95

0.26

0.62

0.13

0.72

0.11

0.75

0.35

0.57

6.95

0.01*

0.07

0.80

1.36

0.25

4.52

0.08

0.51

0.48

0.89

0.37

0.01

0.93

0.03

0.87

0.11

0.75

0.67

0.44

264

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 (continued) Secondary Sexual Variation in Cranial, Dental, and Postcranial Measurements of Recent Samples of Solenodon from Central Dominican Republic, Hispaniola (Sample 3) and Eastern Cuba (Sample 7) Sample

Sex

N

Mean ± SD

Range

CV

Hispaniola

M F M F

21 26 5 4

Breadth of the 25.0 ± 0.96 24.9 ± 0.63 25.3 ± 0.86 24.5 ± 0.52

braincase 23.4–26.5 23.7–26.2 24.3–26.4 23.9–25.2

2.5 3.8 3.4 2.1

M F M F

20 26 5 4

Condylar breadth 17.0 ± 0.72 15.7–18.0 16.9 ± 0.73 15.3–18.4 15.9 ± 0.68 15.0–16.6 16.4 ± 0.79 15.7–17.2

4.3 4.2 4.3 4.8

M F M F

21 26 5 4

19.7 20.3 19.0 19.1

Skull height ± 1.05 17.3–21.3 ± 0.92 18.3–22.2 ± 1.11 17.8–20.6 ± 0.63 18.1–19.6

4.5 5.3 5.9 3.3

M F M F

21 27 5 4

Greatest mandible length 54.1 ± 1.94 50.9–58.1 54.5 ± 1.19 52.4–56.7 48.8 ± 2.77 44.6–51.9 49.0 ± 1.55 47.1–50.4

2.2 3.6 5.7 3.2

M F M F

21 26 5 5

Mandibular toothrow 27.3 ± 0.91 25.4–28.9 27.4 ± 0.64 26.1–28.8 24.8 ± 1.25 22.9–26.1 25.4 ± 0.84 24.5–26.3

2.3 3.3 5.1 3.3

M F M F

21 24 5 5

Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

21 27 5 4

P

0.16

0.68

2.47

0.16

0.58

0.45

1.01

0.35

3.69

0.06

0.01

0.92

0.74

0.39

0.01

0.91

0.22

0.64

0.65

0.44

2.00

0.16

0.05

0.83

process of mandible 22.3–26.0 4.2 22.2–25.7 4.9 0.82 21.4–23.9 4.2 22.2–24.2 3.7 2.00

0.37

Alveolar length 17.1 ± 0.67 17.4 ± 0.57 14.1 ± 0.53 14.1 ± 0.50

Depth through coronoid M 19 23.9 ± 1.17 F 27 24.2 ± 1.01 M 5 22.4 ± 0.94 F 5 23.2 ± 0.85

M F M F

F

of P4-M3 15.8–18.5 16.2–18.4 13.5–14.7 13.6–14.8

Angular-condylar height 15.1 ± 0.75 13.9–16.4 15.1 ± 0.77 13.4–16.9 13.2 ± 1.15 12.0–14.7 13.6 ± 0.65 13.2–14.6

3.3 3.9 3.8 3.5

5.1 4.9 8.7 4.8

0.20

0.05

0.83

0.46

0.52

Systematics and Biogeography of the West Indian Genus Solenodon

265

TABLE 1 (continued) Secondary Sexual Variation in Cranial, Dental, and Postcranial Measurements of Recent Samples of Solenodon from Central Dominican Republic, Hispaniola (Sample 3) and Eastern Cuba (Sample 7) Sample

Sex

N

Hispaniola

M F M F

20 26 5 5

Maximum length of Pm4 4.3 ± 0.21 4.0–4.9 4.3 ± 0.23 3.7–4.7 3.9 ± 0.53 3.3–4.7 3.9 ± 0.14 3.7–4.1

5.4 5.0 13.6 3.7

M F M F

20 26 5 5

Maximum width of Pm4 3.3 ± 0.15 3.1–3.7 3.3 ± 0.15 3.1–3.6 3.2 ± 0.13 3.1–3.4 3.2 ± 0.28 3.0–3.7

4.4 4.4 3.9 8.5

M F M F

19 26 5 5

Maximum length of M1 4.6 ± 0.28 4.0–5.0 4.6 ± 0.31 4.0–5.1 3.8 ± 0.15 3.6–4.0 3.7 ± 0.35 3.2–4.1

6.9 6.2 3.9 9.5

M F M F

19 26 5 5

Maximum width of M1 4.3 ± 0.15 4.1–4.6 4.3 ± 0.15 4.1–4.7 3.9 ± 0.23 3.7–4.2 3.7 ± 0.14 3.5–4.0

3.5 3.6 5.7 3.8

M F M F

20 25 5 5

Maximum length of M2 4.6 ± 0.29 4.2–5.4 4.6 ± 0.22 4.3–5.1 3.6 ± 0.10 3.4–3.7 3.7 ± 0.27 3.4–4.0

4.7 6.3 2.8 7.4

M F M F

20 25 5 5

Maximum width of M2 4.3 ± 0.17 4.1–4.6 4.3 ± 0.14 4.1–4.7 3.7 ± 0.15 3.5–3.9 3.6 ± 0.16 3.3–3.8

3.2 3.9 4.2 4.5

M F M F

20 25 5 5

Maximum length of M3 5.3 ± 0.28 4.7–5.8 5.3 ± 0.26 4.7–5.7 4.3 ± 0.16 4.1–4.5 4.1 ± 0.20 3.9–4.4

4.9 5.3 3.8 4.8

M F M F

20 25 5 5

Maximum width of M3 3.5 ± 0.15 3.3–3.8 3.5 ± 0.18 3.1–3.9 2.8 ± 0.19 2.5–3.0 2.7 ± 0.31 2.2–3.0

5.1 4.2 6.7 11.5

Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Mean ± SD

Range

CV

F

P

0.18

0.68

0.01

0.94

0.56

0.46

0.02

0.90

0.09

0.77

0.45

0.52

1.25

0.27

2.89

0.13

0.02

0.88

0.14

0.72

3.10

0.08

2.37

0.16

0.15

0.69

5.26

0.051

0.20

0.66

0.55

0.48

266

Biogeography of the West Indies: Patterns and Perspectives

TABLE 1 (continued) Secondary Sexual Variation in Cranial, Dental, and Postcranial Measurements of Recent Samples of Solenodon from Central Dominican Republic, Hispaniola (Sample 3) and Eastern Cuba (Sample 7) Sample

Sex

N

Mean ± SD

Hispaniola

M F M F

10 17 2 1

M F M F

11 17 3 1

Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Range

CV

F

P

Total length of femur 46.4 ± 1.76 43.3–48.6 47.0 ± 1.52 44.9–50.4 46.6 ± 1.41 45.6–47.6 46.5

3.2 3.8 3.0

1.12

0.30

Maximun width of femur 13.4 ± 0.53 12.3–13.9 13.5 ± 0.42 12.8–14.3 12.8 ± 0.86 12.2–13.8 12.6

3.1 4.0 6.8

0.45

0.51

M F M F

Minimum shaft width of femur 11 5.2 ± 0.24 4.9–5.6 4.4 17 5.3 ± 0.23 5.0–5.8 4.5 3 4.4 ± 0.46 4.1–4.9 10.5 1 4.5

1.05

0.32

M F M F

10 17 3 1

3.1 3.4 4.3

0.16

0.69

M F M F

11 17 3 1

3.1 3.4 4.4

0.17

0.68

M F M F

Minimum shaft width of humerus 11 5.1 ± 0.25 4.6–5.4 6.0 17 5.0 ± 0.30 4.6–5.8 5.0 3 3.9 ± 8.90 3.8–4.0 2.3 1 4.0

0.72

0.40

M F M F

6 11 2 2

2.50

0.13

1.30

0.37

M F M F

6 11 2 2

0.48

0.50

0.28

0.34

Total length of humerus 47.9 ± 1.61 45.5–49.7 48.1 ± 1.48 45.6–50.7 42.5 ± 1.84 41.1–44.6 42.6 Maximum width 17.9 ± 0.61 18.0 ± 0.56 15.0 ± 0.66 15.9

Total length 54.2 ± 2.72 52.5 ± 1.73 48.6 ± 1.15 51.0 ± 2.62

of humerus 16.7–18.5 17.0–18.8 14.4–15.7

of ulna 50.3–58.0 50.3–56.2 47.8–49.5 49.1–52.8

Maximum width of ulna 7.1 ± 0.21 6.8–7.4 7.0 ± 0.41 6.2–7.6 6.6 ± 0.56 6.1–7.2 6.9 ± 0.63 6.4–7.3

3.3 5.0 2.4 5.1

5.8 3.1 8.6 9.2

Systematics and Biogeography of the West Indian Genus Solenodon

267

TABLE 1 (continued) Secondary Sexual Variation in Cranial, Dental, and Postcranial Measurements of Recent Samples of Solenodon from Central Dominican Republic, Hispaniola (Sample 3) and Eastern Cuba (Sample 7) Sample

Sex

N

Hispaniola

M F M F

6 11 3 2

M F M F

27 22 4 6

529 549 462 426

Total length ± 27.9 485–580 ± 43.2 498–715 ± 47.4 429–530 ± 80.1 325–530

5.3 7.9 10.3 18.8

M F M F

13 8 4 6

Head-body length 296 ± 13.8 273–320 325 ± 67.3 286–490 301 ± 47.9 260–360 253 ± 55.8 195–340

4.7 20.7 15.9 22.0

M F M F

26 22 4 6

224 227 161 162

Tail length ± 14.3 202–254 ± 11.7 196–242 ± 14.4 140–170 ± 23.9 130–190

6.4 5.1 8.9 14.8

M F M F

20 17 4 6

Hind foot length 63.5 ± 4.9 56–72 64.7 ± 4.1 57–70 52.0 ± 5.0 45–56 54.2 ± 2.6 50–56

7.8 6.3 9.6 4.7

M F M F

22 15 4 6

Ear length 28.5 ± 2.2 22–31 28.9 ± 4.1 21–38 24.0 ± 8.2 15–31 28.8 ± 2.7 25–32

7.8 14.2 34.4 9.4

M F M

17 8 2

Body weight (g) 801.4 ± 88.1 620–1080 860.7 ± 165. 726–1166 769.0 ± 55.1 730–808

11.0 19.2 7.1

Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Hispaniola Cuba

Mean ± SD

Range

CV

Minimum shaft width of ulna 2.2 ± 0.16 2.0–2.4 8.5 2.0 ± 0.17 1.7–2.3 7.2 1.6 ± 0.26 1.3–1.9 16.2 1.7 ± 9.89 1.6–1.7 5.9

F

P

3.08

0.09

0.12

0.75

3.82

0.05

0.65

0.55

2.28

0.14

1.39

0.30

0.69

0.41

0.19

0.83

0.68

0.42

5.08

0.03

0.12

0.72

1.08

0.39

1.40

0.25

Note: Statistics given are number (N), mean ± standard deviation, range, coefficient of variation (CV), and F value. Means for males and females that are significantly different at P < 0.05 are marked with an asterisk.

268

Biogeography of the West Indies: Patterns and Perspectives

the lower dentition were identical for males and females from Hispaniola. Measurements of the upper teeth were also very close for both sexes, including that of the canines, often an important character in secondary sexual dimorphism in mammals. Males were then tested against females from the Dominican Republic (as one sample), and no significant differences were observed between males and females in any of the measurements. Considering these results, males and females were not treated separately in subsequent analyses. Individual Variation The majority of the internal measurements examined revealed a relatively low degree of individual variation as expressed by the coefficient of variation (Tables 1 through 3). Cranial and mandibular measurements in all populations usually had coefficients of variation of less than 5, whereas dental and limb bone measurements ranged mostly between 2 and 15. All external measurements, except hind foot length and body weight, had higher coefficients of variation, and therefore were excluded from geographical variation analysis.

SPECIFIC RELATIONSHIPS (GEOGRAPHICAL VARIATION) To establish the specific relationships of the Solenodon populations from Cuba, the Dominican Republic, and Haiti, univariate and multivariate analyses were utilized to compare the geographical samples. Univariate Analyses Standard statistics for each geographical sample of living solenodons and the results of Duncan’s multiple range test for determination of the maximally nonsignificant subsets of 41 variables are given in Table 2. The GLM-ANOVA analysis yielded highly significant differences between the seven geographical samples in all measurements with the exception of one (interorbital constriction). Results of Duncan’s test revealed geographical samples from North Hispaniola (samples 1, 2, 3) grouped alone in one subset differed significantly from all other samples in the following eight measurements: greatest length of skull, condylobasal length, palatal length, length of maxillary toothrow, length of mandibular toothrow, alveolar length of P4-M3, maximum length of P4, and total length of humerus. The samples from North Hispaniola, also assembled separately, differed significantly from the rest of the samples in two nonoverlapping subsets for two measurements (maximum width of M3 and angular-condylar height) and in two overlapping subsets for one measurement (greatest mandible length). Furthermore, the three samples from North Hispaniola grouped together with the sample from southwestern Haiti (sample 6, South Hispaniola) in a single subset, differing significantly from all other samples in the following eight measurements: length of upper molar toothrow, breadth across maxillary toothrow, maximum width of C1, maximum width of P4, maximum length of M1, maximum width of M1, maximum width of M2, maximum width of M3. The samples from South Hispaniola (samples 4, 5, 6) grouped together in 11 measurements with the Cuban population (7) differed significantly from North Hispaniolan samples, in one (angular-condylar height, maximum length of P4, total length of humerus) or in two or more overlapping subsets (greatest length of skull, condylobasal length, palatal length, length of maxillary toothrow, greatest mandible length, length of mandibular toothrow, maximum width of femur, total length of humerus). All three South Hispaniolan samples also clustered in one subset in three measurements (maximum width of M3, maximum width of P4, and total length of femur), and in two significantly different subsets in one (alveolar length of P4M3). Whereas the Haitian sample showed an intermediate position between the North and Dominican Republic south samples (4, 5) in eight measurements, the latter populations differed significantly from all other samples in length of upper molar toothrow, maximum width of C1, and maximum length of M1. These two samples also separated from the others with the Cuban population in breadth across maxillary toothrow (in one subset) and in maximum width of M1, maximum width of M2, and maximum width of M3 (in two subsets). The sample from Sierra de Baoruco (5) isolated from all other samples in one subset in P4-M3 and maximum length of M2.

Systematics and Biogeography of the West Indian Genus Solenodon

269

TABLE 2 Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Range

CV

F and P Values

Sample No.

N

Mean ± SD

3 1 2 4 6 5 7

55 4 16 11 16 10 12

86.5 85.8 85.3 80.7 80.5 79.8 78.2

± ± ± ± ± ± ±

2.36 3.51 2.25 2.21 2.18 2.90 3.51

Greatest length of skull 81.0–91.5 2.7 82.7–90.9 4.1 80.9–88.4 2.6 76.3–83.1 2.7 76.2–83.5 2.7 72.3–82.6 3.6 71.4–82.8 4.5

3 1 2 6 4 5 7

53 4 16 16 11 10 12

81.1 80.3 79.9 76.4 75.7 75.2 73.9

± ± ± ± ± ± ±

2.30 3.22 2.16 1.93 2.53 3.15 2.87

Condylobasal length 76.0–86.6 2.8 77.0–84.2 4.0 75.5–82.7 2.7 73.4–79.4 2.5 71.2–78.4 3.3 67.1–78.9 4.2 68.6–77.8 3.9

3 2 1 6 4 7 5

55 16 4 19 12 14 10

37.4 37.1 37.0 35.5 34.5 34.3 34.1

± ± ± ± ± ± ±

1.15 1.05 0.89 0.94 1.10 1.07 1.44

Palatal length 34.8–40.4 3.1 35.0–38.6 2.8 36.4–38.3 2.4 34.0–37.0 2.7 32.8–35.8 3.2 31.8–35.7 3.1 30.8–35.7 4.2

1 3 2 4 6 5 7

4 52 18 11 17 10 12

31.0 30.5 30.1 29.1 29.0 28.8 27.1

± ± ± ± ± ± ±

2.05 1.02 1.22 1.4 0.97 1.36 1.27

Postpalatal 28.2–32.6 27.6–32.3 27.9–32.0 26.8–31.2 27.6–31.0 25.6–30.2 25.0–29.3

1 3 2 6 4 5 7

4 52 18 21 12 10 14

Alveolar length of upper molar toothrow 10.5 ± 0.11 10.4–10.6 1.0 10.3 ± 0.67 9.1–12.9 6.5 10.9 ± 0.38 10.4–11.7 3.5 9.7 ± 0.50 8.7–10.6 5.2 26.37*** 9.6 ± 0.68 7.9–10.7 7.1 0.0001 9.5 ± 0.60 8.4–10.4 6.3 8.2 ± 0.62 7.3–9.8 7.6

length 6.7 3.4 4.1 4.7 3.4 4.7 4.7

Results Duncan’s I I I

33.36*** 0.0001

I I I

I I I

I I I 25.94*** 0.0001

I I I

I I I

I I I 31.49*** 0.0001

I I

I I I 17.69*** 0.0001

I I

I I I

I I I I

I I I

I I I

I I I I

270

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Range

CV

F and P Values

Sample No.

N

Mean ± SD

3 2 1 6 4 5 7

51 18 4 18 12 10 12

Length of upper molar toothrow 11.1 ± 0.45 9.9–12.0 4.1 10.9 ± 0.38 10.4–11.7 3.5 10.9 ± 0.30 10.6–11.3 2.8 10.7 ± 0.39 10.0–11.6 3.6 57.05*** 10.3 ± 0.41 9.6–11.1 4.0 0.0001 10.1 ± 0.55 9.3–11.0 5.5 8.6 ± 0.44 8.0–9.6 5.1

3 1 2 6 4 7 5

52 4 18 21 12 14 10

26.3 25.7 25.6 24.5 24.2 23.7 23.5

± ± ± ± ± ± ±

Length of maxillary toothrow 0.76 24.5–27.6 2.9 0.75 25.1–26.8 2.9 0.93 23.6–27.5 3.6 0.75 23.5–25.7 3.1 38.08*** 0.81 23.0–25.7 3.3 0.0001 0.87 21.8–24.7 3.7 0.95 21.6–25.1 4.1

3 1 2 6 4 7 5

52 4 17 21 12 14 10

23.9 23.2 23.2 23.1 21.7 21.5 21.2

± ± ± ± ± ± ±

Breadth across maxillary toothrow 0.87 21.7–25.9 3.7 0.50 22.7–23.9 2.2 1.04 21.3–25.3 4.5 0.73 22.1–24.9 3.2 23.07*** 1.24 19.6–23.8 5.7 0.0001 1.16 20.3–23.9 5.4 1.01 19.2–22.5 4.8

7 3 1 2 6 4 5

14 55 4 17 21 12 10

15.0 14.2 13.8 13.8 13.6 13.5 13.5

± ± ± ± ± ± ±

0.72 0.65 0.34 0.81 0.61 0.40 1.08

Anteorbital constriction 14.3–16.4 4.8 12.9–15.9 4.6 13.4–14.2 2.5 12.6–15.1 5.9 12.6–14.6 4.5 12.7–14.3 3.0 11.6–14.9 8.1

3 2 1 5 7 4 6

43 15 4 9 9 11 18

34.5 33.4 33.2 32.8 32.4 32.4 32.2

± ± ± ± ± ± ±

1.68 1.79 1.15 1.06 1.31 1.26 0.89

Zygomatic breadth 31.5–39.0 4.9 30.3–35.8 5.4 31.9–34.4 3.5 31.2–34.5 3.2 30.5–35.2 4.0 30.1–34.9 3.9 30.1–34.2 2.8

Results Duncan’s I I I I I I I

I I I I I

I I I

I I I I I I I

I I I I

9.90*** 0.0001

I I I 8.07*** 0.0001

I I I I I I

I I I I I

Systematics and Biogeography of the West Indian Genus Solenodon

271

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Sample No.

N

Mean ± SD

Range

CV

7 1 3 2 6 4 5

13 5 55 17 20 11 10

15.2 15.1 14.9 14.9 14.7 14.6 14.5

± ± ± ± ± ± ±

0.62 0.38 0.58 0.50 0.44 0.43 0.48

3 1 7 6 2 4 5

55 5 12 18 17 11 9

32.1 31.4 30.9 30.3 30.2 30.2 30.2

± ± ± ± ± ± ±

1.29 0.92 0.79 0.92 1.55 0.65 0.83

Squamosal 29.2–34.5 30.3–32.7 29.6–31.9 28.9–31.6 28.2–33.3 29.0–31.3 28.6–31.6

3 1 2 6 5 7 4

54 4 18 16 9 12 11

25.9 25.6 25.5 25.0 24.6 24.6 24.2

± ± ± ± ± ± ±

1.01 1.25 0.94 0.75 1.13 0.67 0.54

Mastoid breadth 23.7–28.4 3.9 23.8–26.6 4.9 23.5–27.1 3.7 23.7–26.1 3.0 22.6–26.6 4.6 23.0–25.6 2.7 23.4–25.1 2.3

7 3 1 6 2 4 5

12 54 5 19 18 11 9

25.1 24.9 24.5 24.4 24.3 24.2 24.0

± ± ± ± ± ± ±

0.82 0.79 0.59 0.64 0.75 0.51 0.93

3 2 1 6 5 7 4

53 18 4 17 10 12 11

17.0 16.8 16.7 16.3 16.2 16.1 16.0

± ± ± ± ± ± ±

0.70 0.65 0.51 0.49 0.77 0.70 0.55

Interorbital constriction 14.5–16.4 4.1 14.7–15.7 2.6 13.6–16.5 3.9 13.6–15.9 3.4 13.7–15.4 3.0 14.0–15.4 3.0 13.6–15.6 3.3 breadth 4.0 2.9 2.6 3.0 5.1 2.2 2.8

Breadth of the 24.0–26.4 23.4–26.6 23.6–25.3 23.1–25.2 23.0–25.7 23.7–25.2 22.0–25.3

F and P Values

2.02 0.06

I I I I I

I I

I I I I I

I I

11.46*** 0.0001

I I I 9.29*** 0.0001

braincase 3.3 3.2 2.4 2.7 4.85*** 3.1 0.0002 2.1 3.9

Condylar breadth 15.3–18.4 4.1 15.6–18.4 3.9 16.2–17.4 3.0 14.8–16.8 3.0 15.2–17.4 4.8 15.0–17.2 4.4 15.4–16.9 3.4

Results Duncan’s

I I I

I I I 6.95*** 0.0001

I I I

I I I I

I I I I

I I I I I

I I I I I

I I I I

I I I I I

I I I I I

272

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Sample No.

N

Mean ± SD

3 7 2 1 6 4 5

54 11 18 5 17 11 10

20.0 19.2 19.0 18.9 18.8 18.6 18.2

3 7 6 2 1 5 4

7 3 1 2 6 5 4

12 12 22 9 4 9 9

12 12 4 9 22 9 9

4.5 4.5 4.2 4.2 4.2 3.7 3.5

3.0 2.4 2.3 2.3 2.3 2.1 2.1

± ± ± ± ± ± ±

± ± ± ± ± ± ±

± ± ± ± ± ± ±

0.96 0.89 1.38 1.19 1.35 1.03 1.19

Range

CV

Skull height 17.3–22.2 4.8 17.8–20.6 4.7 17.0–21.3 7.3 17.4–20.5 6.3 15.2–20.7 7.2 17.3–20.5 5.6 16.5–20.4 6.6

0.37 0.27 0.28 0.19 0.27 0.20 0.24

Maximum length of C1 3.9–5.2 8.3 4.1–4.8 6.1 3.8–4.7 6.7 3.9–4.5 4.5 3.8–4.3 6.6 3.3–4.0 5.6 3.0–3.9 6.9

0.16 0.97 0.19 0.85 0.12 0.59 0.59

Maximum width of C1 2.6–3.2 5.7 2.2–2.6 4.1 2.1–2.6 8.1 2.1–2.4 3.7 2.0–2.5 5.6 2.0–2.2 2.8 2.0–2.2 2.8 Maximum width of M3 5.2–7.5 5.4 6.1–6.4 2.9 5.4–7.7 9.3 5.2–6.7 6.8 4.8–6.4 8.6 4.4–7.2 15.4 4.2–5.3 6.6

3 1 2 6 5 4 7

53 4 18 19 10 12 12

6.6 6.2 6.2 5.7 5.5 5.4 4.7

± ± ± ± ± ± ±

0.35 0.17 0.58 0.38 0.47 0.84 0.31

3 1 2 6 4 5 7

55 6 17 23 11 9 13

54.2 53.6 52.8 50.6 50.2 49.2 48.9

± ± ± ± ± ± ±

1.58 1.35 1.39 1.48 1.31 1.58 1.97

F and P Values

Results Duncan’s I I

6.76*** 0.0001

I I 19.64*** 0.0001

I I I I I I

I I I

I I I I I

I I I I I

69.66*** 0.0001

I I

I I I 39.06*** 0.0001

Greatest mandible length 50.9–58.1 2.9 51.8–55.3 2.5 50.9–55.3 2.6 47.7–52.6 2.9 38.99*** 47.4–51.6 2.6 0.0001 45.2–50.5 3.2 44.7–51.9 4.0

I I I I

I I

I I I I

I I

I I

Systematics and Biogeography of the West Indian Genus Solenodon

273

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) N

Mean ± SD

3 1 2 6 4 7 5

54 6 18 24 11 14 9

27.3 26.9 26.7 25.7 25.4 25.2 24.6

± ± ± ± ± ± ±

Length of mandibular toothrow 0.79 25.1–28.9 2.9 0.63 25.9–27.7 2.3 1.31 23.3–28.2 4.9 0.76 24.3–27.4 3.2 25.28*** 0.76 24.1–26.3 3.0 0.0001 0.93 23.0–26.3 3.7 0.85 23.5–26.3 3.5

1 3 2 6 4 5 7

6 52 18 24 11 9 14

17.3 17.2 16.9 16.1 16.0 15.4 14.2

± ± ± ± ± ± ±

0.46 0.68 0.74 0.51 0.53 0.59 0.44

3 1 2 4 7 6 5

53 6 17 11 14 23 9

24.1 23.3 23.0 22.9 22.5 22.5 22.2

± ± ± ± ± ± ±

Depth through coronoid process 1.09 22.2–26.2 4.5 1.08 22.0–24.5 4.7 1.31 20.6–25.0 5.7 0.79 21.5–23.9 3.5 10.96*** 1.13 20.6–24.2 5.0 0.0001 0.82 20.6–23.6 3.7 0.89 21.1–24.1 4.0

3 1 2 7 4 6 5

55 6 17 13 11 24 9

15.0 14.7 13.8 13.2 13.1 12.9 12.8

± ± ± ± ± ± ±

0.81 0.68 0.86 0.90 1.00 0.52 0.62

1 3 2 6 4 7 5

5 53 17 23 11 13 9

4.4 ± 0.27 4.3 ± 0.23 4.2 ± 0.23 3.9 ± 0.29 3.9 ± 0.38 3.9 ± 0.32 3.7 ± 0.24

Range

CV

F and P Values

Sample No.

Alveolar length of P4-M3 16.7–17.9 2.7 14.9–18.5 4.0 15.8–18.2 4.4 15.4–17.3 3.2 55.22*** 15.3–16.8 3.3 0.0001 14.6–16.7 3.8 13.5–14.8 3.1

Angular-condylar height 13.4–17.1 5.4 13.9–15.6 4.6 12.2–15.6 6.2 12.1–14.7 6.8 33.76*** 12.6–15.1 7.6 0.0001 12.0–14.0 4.0 11.9–13.8 4.8 Maximum length of P4 3.9–4.7 6.3 3.7–4.9 5.6 3.8–4.7 5.5 3.5–4.9 7.4 3.5–4.6 9.8 3.4–4.7 8.4 3.5–4.2 6.6

Results Duncan’s I I I I I I

I I I I I I I

I I

I I I I I

I I I I I

I I I I I I I

I I I 12.33*** 0.0001

I I

I I I I

274

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Sample No.

N

Mean ± SD

Range

CV

3 1 7 2 6 4 5

53 5 13 18 23 10 9

3.3 ± 0.17 3.3 ± 0.15 3.2 ± 0.18 3.2 ± 0.31 2.8 ± 0.20 2.7 ± 0.18 2.7 ± 0.28

Maximum width of P4 2.7–3.7 5.6 3.1–3.4 4.6 3.1–3.7 5.7 2.8–3.8 9.6 2.3–3.1 7.1 2.6–3.2 6.4 2.2–3.2 10.4

1 3 2 6 4 5 7

4 52 18 23 11 9 13

4.8 4.6 4.6 4.5 4.3 4.2 3.7

± ± ± ± ± ± ±

0.19 0.30 0.19 0.20 0.27 0.24 0.29

Maximum length of M1 4.6–5.0 4.1 4.0–5.1 6.5 4.1–4.8 4.1 4.2–5.0 4.5 4.0–4.9 6.1 4.0–4.7 5.6 3.2–4.1 7.9

3 1 2 6 7 4 5

52 4 18 23 13 10 9

4.3 4.2 4.2 4.1 3.9 3.8 3.7

± ± ± ± ± ± ±

0.17 0.85 0.22 0.18 0.19 0.29 0.26

Maximum width of M1 3.8–4.7 4.1 4.1–4.3 2.1 3.9–4.7 5.3 3.8–4.4 4.3 3.6–4.2 5.0 3.5–4.6 7.8 3.3–4.2 7.1

3 1 2 6 4 5 7

52 5 18 22 11 8 13

4.6 4.6 4.6 4.5 4.5 4.2 3.6

± ± ± ± ± ± ±

0.26 0.19 0.25 0.20 0.20 0.16 0.18

Maximum length of M2 4.0–5.4 5.6 4.4–4.9 4.1 4.1–5.0 5.4 4.1–5.0 4.6 4.2–4.8 4.5 3.9–4.4 3.9 3.4–4.0 5.2

3 1 2 6 4 5 7

52 6 18 22 10 8 13

4.3 4.2 4.2 4.1 3.9 3.7 3.6

± ± ± ± ± ± ±

0.17 0.11 0.20 0.17 0.25 0.23 0.15

Maximum width of M2 3.8–4.7 4.0 4.1–4.4 2.7 3.9–4.7 4.8 3.8–4.5 4.0 3.7–4.5 6.5 3.5–4.3 6.1 3.4–3.9 4.3

F and P Values

28.65*** 0.0001

23.81*** 0.0001

Results Duncan’s I I I I I I I

I I I I I I I

I I I 22.17*** 0.0001

35.19*** 0.0001

I I I I I

I I

I I I I I I I

33.59*** 0.0001

I I I I I I

I I

Systematics and Biogeography of the West Indian Genus Solenodon

275

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Mean ± SD

Range

N

3 1 2 6 4 5 7

52 5 18 18 11 9 13

5.3 5.3 5.1 5.0 4.9 4.7 4.2

± ± ± ± ± ± ±

0.27 0.10 0.25 0.23 0.26 0.28 0.22

Maximum length of M3 4.7–5.8 5.2 5.2–5.4 1.8 4.7–5.9 4.9 4.4–5.4 4.7 4.5–5.3 5.3 4.3–5.3 6.0 3.9–4.5 5.2

3 1 2 6 4 5 7

52 5 18 18 11 9 13

3.5 3.5 3.5 3.3 3.1 3.0 2.9

± ± ± ± ± ± ±

0.19 0.17 0.19 0.11 0.30 0.19 0.39

Maximum width of M3 3.0–4.0 5.5 3.3–3.7 5.0 3.1–3.9 5.6 3.1–3.5 3.5 2.9–3.9 9.6 2.7–3.4 6.4 2.2–3.9 13.5

1 3 7 2 6 4 5

3 29 5 10 24 10 8

47.1 46.8 46.7 46.3 44.8 43.8 43.6

± ± ± ± ± ± ±

1.35 1.61 0.78 1.40 1.44 1.40 1.43

3 1 2 7 4 5 6

30 4 10 6 9 8 25

13.4 13.3 13.3 12.7 12.6 12.5 12.4

± ± ± ± ± ± ±

0.45 0.49 0.80 0.61 0.42 0.55 0.56

1 2 3 5 4 6 7

4 10 30 8 10 25 6

5.5 5.4 5.3 5.1 5.1 4.7 4.5

± ± ± ± ± ± ±

Minimum shaft width of femur 0.59 5.2–6.4 10.7 0.35 4.8–5.9 6.6 0.24 4.9–5.8 4.7 0.43 4.6–6.0 8.4 11.96*** 0.40 4.5–5.7 8.0 0.0001 0.24 4.3–5.4 5.1 0.31 4.1–4.9 6.9

Total length 45.7–48.3 43.3–50.4 45.6–47.6 44.3–48.8 42.4–47.6 41.4–45.5 41.0–45.8

CV

F and P Values

Sample No.

of femur 2.9 3.4 1.7 3.0 3.2 3.2 3.3

Results Duncan’s I I I

36.03*** 0.0001

I I I

I I I

22.99*** 0.0001

10.94*** 0.0001

Maximum width of femur 12.3–14.3 3.4 12.8–14.0 3.7 12.3–14.6 6.1 12.1–13.8 4.8 9.90*** 11.9–13.3 3.3 0.0001 11.6–13.3 4.4 11.4–13.4 4.5

I I I I I I

I I

I I I I I I I

I I I I I I I

I I I

I I I I

I I I I

276

Biogeography of the West Indies: Patterns and Perspectives

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Range

CV

F and P Values

Sample No.

N

Mean ± SD

3 1 2 4 6 5 7

29 3 11 8 21 8 5

48.0 47.1 46.8 44.2 44.0 43.9 42.6

± ± ± ± ± ± ±

1.45 0.52 1.54 1.23 1.36 1.30 1.35

3 1 2 6 5 4 7

30 3 11 22 8 10 6

17.9 17.9 17.8 17.3 17.0 16.8 14.9

± ± ± ± ± ± ±

Maximum width 0.59 16.7–18.8 0.35 17.6–18.3 0.53 16.8–18.5 0.57 16.1–18.2 0.78 16.0–18.3 0.65 15.8–18.1 0.75 14.1–15.9

1 6 3 2 5 4 7

3 21 30 11 8 10 6

5.2 5.0 5.0 5.0 4.9 4.9 3.9

± ± ± ± ± ± ±

Minimum shaft width of humerus 0.28 4.9–5.4 5.5 0.23 4.5–5.4 4.5 0.29 4.6–5.8 5.7 0.35 4.4–5.5 7.1 15.25*** 0.34 4.5–5.5 7.0 0.0001 0.30 4.3–5.4 6.2 0.70 3.8–4.0 1.8

1 3 2 6 4 7 5

3 19 8 17 7 6 6

53.6 53.2 52.2 51.1 51.1 49.6 49.4

± ± ± ± ± ± ±

1.13 2.09 1.39 1.13 1.59 1.90 4.53

3 1 2 4 7 5 6

19 4 8 7 7 6 16

7.0 6.9 6.8 6.8 6.7 6.7 6.4

± ± ± ± ± ± ±

0.33 0.39 0.47 0.33 0.47 0.44 0.35

Total length of humerus 45.5–50.7 3.0 46.7–47.7 1.1 43.9–49.0 3.3 42.2–45.4 2.9 27.90*** 40.6–46.8 3.1 0.0001 41.3–45.6 3.0 41.2–44.6 3.2

Total length 52.3–54.3 50.3–58.0 50.0–54.0 49.0–53.4 49.3–53.8 47.8–52.8 40.6–53.8

of humerus 3.3 2.0 3.0 3.3 23.51*** 4.6 0.0001 3.9 5.0

of ulna 2.1 3.9 2.7 2.2 3.1 3.8 9.2

Maximum width of ulna 6.2–7.6 4.7 6.4–7.2 5.7 6.0–7.4 6.9 6.3–7.1 4.9 6.1–7.3 7.0 5.9–7.2 6.7 5.9–7.1 5.6

4.66*** 0.0006

4.49*** 0.0008

Results Duncan’s I I I I I I I

I I I I

I I I I

I I I I I I I

I I I I I

I I I I I I

I I I

I I I

Systematics and Biogeography of the West Indian Genus Solenodon

277

TABLE 2 (continued) Geographical Variation in Cranial and Postcranial Measurements of Seven Samples of Extant Solenodon Populations from North Hispaniola (Samples 1, 2, and 3, Dominican Republic), South Hispaniola (Samples 4, 5, Dominican Republic; and 6, Haiti), and Eastern Cuba (Sample 7) Sample No.

N

1 3 6 2 4 5 7

3 19 17 8 7 6 7

Mean ± SD

2.2 2.1 2.0 2.0 1.8 1.8 1.6

± ± ± ± ± ± ±

Range

CV

F and P Values

Minimum shaft width of ulna 0.28 1.9–2.5 12.8 0.19 1.7–2.4 9.1 0.19 1.7–2.3 9.5 0.13 1.9–2.2 6.4 8.90*** 0.12 1.6–2.0 6.5 0.0001 0.20 1.5–2.1 11.6 0.16 1.4–1.9 9.9

Results Duncan’s I I I

I I I

I I

I I

I I

Note: Statistics given are number (N), mean ± standard deviation, range, coefficient of variation (CV), F and P values, and results of Duncan’s multiple range test (