ENCYCLOPEDIA OF
GEOLOGY
ENCYCLOPEDIA OF
GEOLOGY EDITED BY
RICHARD C. SELLEY L. ROBIN M. COCKS IAN R. PLIMER
ELSEVIER ACADEMIC PRESS Amsterdam Boston Heidelberg London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo
Elsevier Ltd., The Boulevard, Langford Lane, Kidlington, Oxford, OX5 1GB, UK © 2005 Elsevier Ltd. The following articles are © 2005, The Natural History Museum, London, UK: FOSSIL VERTEBRATES/Hominids Palaeontology PALAEOZOIC/Silurian PRECAMBRIAN/Overview Terranes, Overview Conservation of Geological Specimens MINERALS/Olivines MINERALS/Sulphates TERTIARY TO PRESENT/Pleistocene and The Ice Age Environmental Geochemistry Biological Radiations and Speciation PALAEOZOIC/Ordovician TERTIARY TO PRESENT/Eocene TERTIARY TO PRESENT/Paleocene FOSSIL PLANTS/Angiosperms FOSSIL PLANTS/Gymnosperms Biozones MESOZOIC/Cretaceous MESOZOIC/End Cretaceous Extinctions Stratigraphical Principles FOSSIL INVERTEBRATES/Molluscs Overview FOSSIL INVERTEBRATES/Trilobites FOSSIL INVERTEBRATES/Echinoderms (Other Than Echinoids) FOSSIL INVERTEBRATES/Echinoids TERTIARY TO PRESENT/Pliocene FOSSIL INVERTEBRATES/Bryozoans MINERALS/Feldspathoids Russia The following article is a US Government work in the public domain and not subject to copyright: NORTH AMERICA/Atlantic Margin "Earth from Space" endpaper figure reproduced with permission from Reto Stockli, Nazmi El Saleous, and Marit Jentoft-Nilsen and NASA GSFC All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. Permissions may be sought directly from Elsevier's Rights Department in Oxford, UK: phone (+44) 1865 843830, fax (+44) 1865 853333,
[email protected]. Requests may also be completed on-line via the homepage (http://www.elsevier.com/locate/permissions). First edition 2005 Library of Congress Control Number: 2004104445 A catalogue record for this book is available from the British Library ISBN 0-12-636380-3 (set) This book is printed on acid-free paper Printed and bound in Spain
EDITORS v
Editors
EDITORS Richard C. Selley
Imperial College London, UK L. Robin M. Cocks
Natural History Museum London, UK Ian R. Plimer
University of Melbourne Melbourne, VA Australia
CONSULTANT EDITOR Joe McCall
Cirencester Gloucestershire, UK
vi EDITORIAL ADVISORY BOARD
Editorial Advisory Board Jaroslav Aichler
Georg Hoinkes
Czech Geological Survey Jesenı´ k, Czech Republic
Universita¨t Graz Universita¨tplatz 2 Graz, Austria
Andrew R Armour
Revus Energy A/S Norway
R A Howie
John Collinson
Shunsho Ishihara
Delos, Beech Staffordshire, UK
Royal Holloway, London University London, UK
Geological Survey of Japan Tsukuba, Japan
Alexander M Davis
Gilbert Kelling
Infoscape Solutions Ltd. Guildford, UK
Keele University Keele, UK
Peter Doyle
Ken Macdonald
University College London London, UK Wolfgang Franke
Institut fu¨r Geowissenschaften Giessen, Germany
University of California Santa Barbara Santa Barbara, CA, USA Norman MacLeod
The Natural History Museum London, UK Stuart Marsh
Yves Fuchs
Universite´ Marne la Valle France
British Geological Survey Nottingham, UK Joe McCall
Paul Garrard
Cirencester, Gloucestershire, UK
Formerly Imperial College London, UK
David R Oldroyd
R O Greiling
Universita¨t Heidelberg Heidelberg, Germany
University of New South Wales Sydney, NSW, Australia Rong Jia-yu
Nanjing Institute of Geology and Palaeontology Nanjing, China
Gwendy Hall
Natural Resources Canada Ottawa, ON, Canada
Mike Rosenbaum
Robert D Hatcher, Jr.
Peter Styles
University of Tennessee Knoxville, TN, USA
Keele University Keele, UK
Twickenham, UK
EDITORIAL ADVISORY BOARD vii
Hans D Sues
S H White
Carnegie Museum of Natural History Pittsburgh, PA, USA
Universiteit Utrecht Utrecht, The Netherlands
John Veevers
Macquarie University Sydney, NSW, Australia
FOREWORD ix
Foreword Few areas of science can have changed as fast as geology has in the past forty years. In the first half of the last century geologists were divided, often bitterly, between the drifters and those who believed that the Earth and its continents were static. Neither side of this debate foresaw that the application of methods from physics, chemistry and mathematics to these speculations would revolutionize the study of all aspects of the Earth Sciences, and would lead to accurate and detailed reconstructions of world geography at former times, as well as to an understanding of the origin of the forces that maintain the continental movements. This change in world-view is no longer controversial, and is now embedded in every aspect of the Earth Sciences. It is a real pleasure to see this change, which has revitalized so many classic areas of research, reflected in the articles of this encyclopedia. Particularly affected are the articles on large-scale Earth processes, which discuss many of the new geological ideas that have come from geophysics and geochemistry. Forty years ago we had no understanding of these topics, which are fundamental to so many aspects of the Earth Sciences. The editors have decided, and in my view quite rightly, not to include detailed discussion of the present technology that is used to make geophysical and geochemical measurements. Such instrumental aspects are changing rapidly and become dated very quickly. They can easily be found in more technical publications. Instead the editors have concentrated on the influence such studies have had on our understanding of the Earth and its evolution, and in so doing have produced an excellent and accessible account of what is now known. Any encyclopedia has to satisfy a wide variety of users, and in particular those who know that some subject like sedimentation or mineral exploration is part of geology, and go to an encyclopedia of geology to find out more. The editors have made a very thorough attempt to satisfy such users, and have included sections on such unexpected geological topics as the evolution of the Earth’s atmosphere, the geology of Jupiter, Saturn, and their moons, aggregates, and creationism. I congratulate the editors and authors for producing such a fine summary of our present knowledge, and am particularly pleased that they intend to produce an online version of the encyclopedia. Though I have become addicted to using the Internet as my general encyclopedia, I will be delighted to be able to access something concerned with my own field that is as organized and scholarly as are these volumes. Dan McKenzie Royal Society Professor of Earth Sciences Cambridge University, UK
INTRODUCTION xi
Introduction Civilization occurs by geological consent
subject to change without notice.... Will Durant (1885 1981)
Richard de Bury, Bishop of Durham from 1333 to 1345, divided all knowledge into ‘Geologia’, earthly knowledge, and ‘Theologia’, heavenly knowledge. By the beginning of the last century, however, Geology was generally understood to be restricted to the study of rocks: according to the old dictum of the Geological Survey of Great Britain ‘If you can hit it with a hammer, then it’s geology.’ Subsequently geology has been subsumed into Earth Science. This includes not only the study of rocks (the lithosphere), but also the atmosphere and hydrosphere and their relationship with the biosphere. Presently these relationships now form a nexus in Earth System Science. The ‘Encyclopedia of Geology’ is what it says on the cover. What appealed to us when first approached to edit this work by Academic Press was a request that the encyclopedia should be rock-based. Readers are referred to the companion volumes, Encyclopedia of Atmospheric Sciences, Encyclopedia of the Solar System, Encyclopedia of Soils in the Environment and Encyclopedia of Ocean Sciences for knowledge on the other branches of Earth Science. Nonetheless we have extended our brief to include articles on the other planets and rocky detritus of our solar system, leaving others to argue, as no doubt Bishop Richard would have done, where the boundaries of earthly and heavenly knowledge might be. (His Grace would probably have charged the editors of the Encyclopedia of the Solar System with heresy.) One of the first, and most difficult, tasks of editing this encyclopedia was to decide, not only which topics merited articles, but also how these articles should be grouped to facilitate the reader. This is easy for some branches of geology, but difficult for others. It is relatively easy to logically arrange articles on mineralogy and palaeontology, since they are defined by their chemistry and evolutionary biology. Articles that describe Earth history may be conveniently arranged in a chronological order, and articles on regional geology may be presented geographically. Other topics present problems, particularly in the area of sedimentology. There is, for example, a range of inter-related topics associated with deserts. This area could be described geomorphologically, and in terms of the aeolian and aqueous processes of deserts, aeolian sedimentary structures, and aeolian deposits. All of these aspects of deserts deserve mention, but there is no obvious logical way of arranging the discrete topics into articles. To help us in this task we relied heavily on our editorial board, whose individual members had more specialized knowledge of their field than we. To the Editorial Board Members, authors and anonymous referees of each article we give heartfelt thanks. We were also, of course, constrained by the willingness of expert authorities to contribute articles. To some degree therefore, the shape of the encylopedia owes as much to the enthusiasm of experts to write for us, as for our ‘wish list’ of articles. To facilitate readers finding their way around the Encyclopedia of Geology great care has been taken in crossreferencing within and between articles, in providing ‘See Also’ lists at the end of articles, and in the index. No doubt it will be easier for readers to navigate around the online version of the work, than to manipulate the several hard copy volumes. As geological knowledge expands there is always more to learn and understand. While preparing the ‘Encyclopedia of Geology’ we have ourselves learned a great deal about geology, both within and beyond our own specialties. We invite you to read this encyclopedia and join us in the field trip of a lifetime. Richard C. Selley L. Robin M. Cocks Ian R. Plimer 1 August 2004 References to related encyclopedia published by Elsevier, Academic Press: Encyclopedia of the Solar System, 1998 Encyclopedia of Ocean Sciences, 2001 Encyclopedia of Atmospheric Sciences, 2002 Encyclopedia of Soils in the Environment, 2005
GUIDE TO USE OF THE ENCYCLOPEDIA xiii
Guide to Use of the Encyclopedia Structure of the Encyclopedia The material in the Encyclopedia is arranged as a series of entries in alphabetical order. Most entries consist of several articles that deal with various aspects of a topic and are arranged in a logical sequence within an entry. Some entries comprise a single article. To help you realize the full potential of the material in the Encyclopedia we have provided three features to help you find the topic of your choice: a Contents List, Cross-References and an Index.
1. Contents List Your first point of reference will probably be the contents list. The complete contents lists, which appears at the front of each volume will provide you with both the volume number and the page number of the entry. On the opening page of an entry a contents list is provided so that the full details of the articles within the entry are immediately available. Alternatively you may choose to browse through a volume using the alphabetical order of the entries as your guide. To assist you in identifying your location within the Encyclopedia a running headline indicates the current entry and the current article within that entry. You will find 'dummy entries' where obvious synonyms exist for entries or where we have grouped together related topics. Dummy entries appear in both the contents lists and the body of the text. Example If you were attempting to locate material on erosional sedimentary structures via the contents list: EROSION see SEDIMENTARY PROCESSES: Fluxes and Budgets; Aeolian Processes; Erosional Sedimentary Structures. The dummy entry directs you to the Erosional Sedimentary Structures article, in the SEDIMENTARY PROCESSES entry. At the appropriate location in the contents list, the page numbers for articles under Sedimentary Processes are given. If you were trying to locate the material by browsing through the text and you looked up Erosion then the following information would be provided in the dummy entry:
EROSION See SEDIMENTARY PROCESSES: Erosional Sedimentary Structures; Aeolian Processes; Fluxes and Budgets
xiv
GUIDE TO USE OF THE ENCYCLOPEDIA
Alternatively, if you were looking up Sedimentary Processes the following information would be provided:
SEDIMENTARY PROCESSES Contents Erosional Sedimentary Structures Depositional Sedimentary Structures Post-Depositional Sedimentary Structures Aeolian Processes Catastrophic Floods Deep Water Processes and Deposits Fluvial Geomorphology Glaciers Karst and Palaeokarst Landslides Particle-Driven Subaqueous Gravity Processes Deposition from Suspension Fluxes and Budgets
2. Cross-References All of the articles in the Encyclopedia have been extensively cross-referenced. The cross-references, which appear at the end of an article, serve three different functions. For example, at the end of the PRECAM BRIAN: Overview article, cross-references are used: i. To indicate if a topic is discussed in greater detail elsewhere. Africa: Pan-African Orogeny. Antarctic Asia: Central. Australia: Proterozoic Biosediments and Biofilms Earth Structure and Origins. Earth System Science.Europe: East European Craton; Timanides of Northern Russia. Gondwanaland and Gondwana. Grenvillian Orogeny. Indian Subcontinent. North America:Precambrian Continental Nucleus; Continental Interior. Precambrian: Eukaryote Fossils; Prokaryote Fossils; Vendian and Ediacaran, Russia, Sedimentary Rocks: Banded Iron Formations. Shields. Terranes, Overview. ii. To draw the reader's attention to parrallel discussions in other articles.
Africa: Pan-African Orogeny. Antarctic. Asia: Central. Australia: Proterozoic. Biosediments and Biofilms. Earth Structure and Origins. Earth System Science. Europe: East European Craton; Timanides of Northern Russia. Gondwanaland and Gondwana. Grenvillian Orogeny Indian Subcontinent. North America: Precambrian Continental Nucleus; Continental Interior. Precambrian: Eukaryote Fossils; Prokaryote Fossils; Vendian and Ediacdran. Russia. Sedimentary Rocks: Banded Iron Formations. Shields. Terranes, Overview.
GUIDE TO USE OF THE ENCYCLOPEDIA xv
iii. To indicate material that broadens the discussion. Africa: Pan-African Orogeny. Antarctic. Asia: Central. Australia: Proterozoic. Biosediments and Biofilms. Earth Structure and Origins. Earth Syatem Science. Europe: East European Graton; Timanides of Northern Russia. Gondwantand and Gendwana. Grenvillian Orogeny. Indian Subcontinent. North America: Precambrian Continental Nucleus; Continental Interior. Precambrian: Eukaryote Fossils; Prokaryote Fossils; Vendian and Ediacaran. Russia. Sedimentary Rocks: Banded Iron Formations. Shields. Terranes, Overview.
3. Index The index will provide you with the page number where the material is located, and the index entries differentiate between material that is a whole article, is part of an article or is data presented in a figure or table. Detailed notes are provided on the opening page of the index.
4. Contributors A full list of contributors appears at the beginning of each volume.
CONTRIBUTORS xvii
Contributors Abart, R University of Basel, Basel, Switzerland
Best, J University of Leeds, Leeds, UK
Aldridge, R J University of Leicester, Leicester, UK
Birch, W D Museum Victoria, Melbourne, VIC, Australia
Al-Jallal, I A Sandroses Est. for Geological, Geophysical Petroleum Engineering Consultancy and Petroleum Services, Khobar, Saudi Arabia
Bird, J F Imperial College London, London, UK
Alkmim, F F Universidade Federal de Ouro Preto, Ouro Preto, Brazil Allen, P M Bingham, Nottingham, UK Allwood, A C Macquarie University, Sydney, NSW, Australia Al-Sharhan, A S United Arab Emirates University, AI-Ain, United Arab Emirates Anderson, L I National Museums of Scotland, Edinburgh, UK Arndt, N T LCEA, Grenoble, France Arnott, R Oxford Institute for Energy Studies, Oxford, UK Asimow, P D California Institute of Technology, Pasadena, CA, USA Atkinson, J City University, London, UK Bacon, M Petro-Canada, London, UK
Black, P Auckland University, Auckland, New Zealand Bleeker, W Geological Survey of Canada, Ottawa, ON, Canada Bogdanova, S V Lund University, Lund, Sweden Bommer, J J Imperial College London, London, UK Boore, D M United States Geological Survey, Menlo Park, CA, USA Bosence, D W J Royal Holloway, University of London, Egham, UK Boulanger, R W University of California, Davis, CA, USA Braga, J C University of Granada, Granada, Spain Branagan, D F University of Sydney, Sydney, NSW, Australia Brasier, M D University of Oxford, Oxford, UK Brewer, P A University of Wales, Aberystwyth, UK
Bailey, J Anglo-Australian Observatory and Australian Centre for Astrobiology, Sydney, Australia
Bridge, M University College London, London, UK
Bani, P Institut de la Recherche pour le Développement, Noumea, New Caledonia
Brown, D Institute de Ciencias de la Tierra 'Jaume Almera' CSIC, Barcelona, Spain
Bell, F G British Geological Survey, Keyworth, UK
Brown, A J Macquarie University, Sydney, NSW, Australia
Bell, K Carleton University, Ottawa, ON, Canada
Brown, R J University of Bristol, Bristol, UK
xviii CONTRIBUTORS Bucher, K University of Freiburg, Freiburg, Germany
Cosgrove, J W Imperial College London, London, UK
Burns, S F Portland State University, Portland, OR, USA
Coxon, P Trinity College, Dublin, Ireland
Byford, E Broken Hill, NSW, Australia
Cressey, G The Natural History Museum, London, UK
Calder, E S Open University, Milton Keynes, UK
Cribb, S J Carraig Associates, Inverness, UK
Cameron, E M Eion Cameron Geochemical Inc., Ottawa, ON, Canada
Cronan, D S Imperial College London, London, UK
Carbotte, S M Columbia University, New York, NY, USA
Currant, A The Natural History Museum, London, UK
Carminati, E Universita La Sapienza, Rome, Italy
Davies, H University of Papua New Guinea, Port Moresby Papua New Guinea
Chamberlain, S A Macquarie University, Sydney, NSW, Australia
Davis, G R Imperial College London, London, UK
Charles, J A Formerly Building Research Establishment Hertfordshire, UK
DeCarli, P S SRI International, Menlo Park, CA, USA
Chiappe, L M Natural History Museum of Los Angeles County Los Angeles, CA, USA
Dewey, J F University of California Davis Davis, CA, USA, and University of Oxford, Oxford, UK
Clack, J A University of Cambridge, Cambridge, UK
Doglioni, C Universita La Sapienza, Rome, Italy
Clayton, C Eardiston, Tenbury Wells, UK
Doming, K J University of Sheffield, Sheffield, UK
Clayton, G Trinity College, Dublin, Ireland
Dott, Jr R H University of Wisconsin, Madison, Wl, USA
Cocks, L R M The Natural History Museum, London, UK
Doyle, P University College London, London, UK
Coffin, M F University of Tokyo, Tokyo, Japan
Dubbin, W E The Natural History Museum, London, UK
Collinson, J John Collinson Consulting, Beech, UK
Dyke, G J University College Dublin, Dublin, Ireland
Comerford, G The Natural History Museum, London, UK
Echtler, H GeoForschungsZentrum Potsdam, Potsdam, Germany
Condie, K C New Mexico Tech, Socorro, NM, USA
Eden, M A Geomaterials Research Services Ltd, Basildon, UK
Cornford, C Integrated Geochemical Interpretation Ltd, Bideford, UK
Eide, E A Geological Survey of Norway, Trondheim, Norway
Cornish, L The Natural History Museum, London, UK
Eldholm, O University of Bergen, Bergen, Norway
CONTRIBUTORS xix
Elliott, D K Northern Arizona University, Flagstaff, AZ, USA
Garetsky, R G Institute of Geological Sciences, Minsk, Belarus
Elliott, T University of Liverpool, Liverpool, UK
Garrard, P Imperial College London, London, UK
Eriksen, A S Zetica, Witney, UK
Gascoyne, J K Zetica, Witney, UK
Payers, S R University of Aberdeen, Aberdeen, UK
Gee, D G University of Uppsala, Uppsala, Sweden
Feenstra, A GeoForschungsZentrum Potsdam, Potsdam, Germany
Geshi, N Geological Survey of Japan, Ibaraki, Japan
Felix, M University of Leeds, Leeds, UK
Giese, P Freie Universitat Berlin, Berlin, Germany
Figueras, D BFI, Houston, TX, USA Fookes, P G Winchester, UK Forey, P L The Natural History Museum, London, UK Fortey, R A The Natural History Museum, London, UK Foster, D A University of Florida, Gainesville, FL, USA Frýda, J Czech Geological Survey, Prague, Czech Republic Franke, W Johann Wolfgang Goethe-Universitat Frankfurt am Main, Germany Franz, G Technische Universitat Berlin, Berlin, Germany French, W J Geomaterials Research Services Ltd, Basildon, UK Fritscher, B Munich University, Munich, Germany Frostick, L University of Hull, Hull, UK Fuchs, Y Université Marne la Vallée, Marne la Vallée, France Gabbott, S E University of Leicester, Leicester, UK Garaebiti, E Department of Geology and Mines, Port Vila, Vanuatu
Giles, D P University of Portsmouth, Portsmouth, UK Glasser, N F University of Wales, Aberystwyth, UK Gluyas, J Acorn Oil and Gas Ltd., Staines, UK Gorbatschev, R Lund University, Lund, Sweden Gordon, J E Scottish Natural Heritage, Edinburgh, UK Gradstein, F M University of Oslo, Oslo, Norway Gray, D R University of Melbourne, Melbourne, VIC, Australia Greenwood, J R Nottingham Trent University, Nottingham, UK Grieve, RAF Natural Resources Canada, Ottawa, ON, Canada Griffiths, J S University of Plymouth, Plymouth, UK Hambrey, M J University of Wales, Aberystwyth, UK Hancock, J M† Formerly Imperial College London, London, UK Hansen, J M Danish Research Agency, Copenhagen, Denmark Harff, J Baltic Sea Research Institute Warnemunde, Rostock, Germany †
Deceased
xx
CONTRIBUTORS
Harper, DAT Geologisk Museum, Copenhagen, Denmark
Howell, J University of Bergen, Bergen, Norway
Harper, E M University of Cambridge, Cambridge, UK
Howie, R A Royal Holloway, University of London, London, UK
Harrison, JP Imperial College London, London, UK
Hudson-Edwards, K University of London, London, UK
Hatcher, Jr RD University of Tennessee, Knoxville, TN, USA
Huggett, J M Petroclays, Ashtead, UK and The Natural History Museum, London, UK
Hatheway, A W Rolla, MO and Big Arm, MT, USA Hauzenberger, C A University of Graz, Graz, Austria Hawkins, A B Charlotte House, Bristol, UK Haymon, R M University of California-Santa Barbara Santa Barbara, CA, USA He Guoqi Peking University, Beijing, China Head, J W Brown University, Providence, Rl, USA Heim, N A University of Georgia, Athens, GA, USA Helvaci, C Dokuz Eylül Üniversitesi, Izmir, Turkey Hendriks, B W H Geological Survey of Norway, Trondheim, Norway
Hughes, N C University of California, Riverside, CA, USA Hutchinson, D R US Geological Survey, Woods Hole, MA, USA Idriss, I M University of California, Davis, CA, USA Ineson, J R Geological Survey of Denmark and Greenland Geocenter Copenhagen, Copenhagen, Denmark Ivanov, M A Russian Academy of Sciences, Moscow, Russia Jäger, K D Martin Luther University, Halle, Germany Jarzembowski, E A University of Reading, Reading, UK and Maidstone Museum and Bentlif Art Gallery, Maidstone, UK Jones, B University of Alberta, Edmonton, AB, Canada
Henk, A Universität Freiburg, Freiburg, Germany
Jones, G L Conodate Geology, Dublin, Ireland
Herries Davies, G L University of Dublin, Dublin, Ireland
Joyner, L Cardiff University, Cardiff, UK
Hey, R N University of Hawaii at Manoa, Honolulu, HI, USA
Kaminski, M A University College London, London, UK
Hoinkes, G University of Graz, Graz, Austria
Cornell University, Ithaca, NY, USA
Hooker, J J The Natural History Museum, London, UK
Kemp, A I S University of Bristol, Bristol, UK
Home, D J University of London, London, UK
Kendall, A C University of East Anglia, Norwich, UK
Hovland, M Statoil, Stavanger, Norway
Kenrick, P The Natural History Museum, London, UK
Kay, S M
CONTRIBUTORS xxi
Kogiso, T Japan Marine Science and Technology Center, Yokosuka, Japan Krings, M Bayerische Staatssammlung für Paläontologie und Geologic, Geo-Bio Center, Munich, Germany Lancaster, N Desert Research Institute, Reno, NV, and United States Geological Survey, Reston, VA, USA Lang,K R Tufts University, Medford, MA, USA Laurent, G Brest, France
Lee, E M York, UK Lemke, W Baltic Sea Research Institute Warnemünde, Rostock Germany Lesher, C M Laurentian University, ON, Canada Lewin, J University of Wales, Aberystwyth, UK Liu, J G Imperial College London, London, UK
MacLeod, N The Natural History Museum, London, UK Maltman, A University of Wales, Aberystwyth, UK Martill, D M University of Portsmouth, Portsmouth, UK Martins-Neto, M A Universidade Federal de Ouro Preto, Ouro Preto, Brazil Marvin, U B Harvard-Smithsonian Center for Astrophysics Cambridge, MA, USA Mason, P J HME Partnership, Romford, UK Massonne, H-J Universität Stuttgart, Stuttgart, Germany Matte, P University of Montpellier II, Montpellier, France Mayor, A Princeton, USA McCaffrey, W University of Leeds, Leeds, UK McCall, G J H Cirencester, Gloucester, UK
Long,J A The Western Australian Museum, Perth WA, Australia
McCave, I N University of Cambridge, Cambridge, UK
Loock, J C University of the Free State Bloemfontein, South Africa
McGhee, G R Rutgers University, New Brunswick, NJ, USA
Lowell, R P Georgia Institute of Technology, Atlanta, GA, USA
McKibben, M A University of California, CA, USA
Lucas, S G New Mexico Museum of Natural History Albuquerque, NM, USA
McLaughlin, Jr P P Delaware Geological Society, Newark, DE, USA
Liming, S University of Bremen, Bremen, Germany Luo, Z-X Carnegie Museum of Natural History Pittsburgh, PA, USA
McManus, J University of St. Andrews, St. Andrews, UK McMenamin, MAS Mount Holyoke College, South Hadley, MA, USA Merriam, D F University of Kansas, Lawrence, KS, USA
Macdonald, K C University of California-Santa Barbara Santa Barbara, CA, USA
Metcalfe, I University of New England, Armidale, NSW, Australia
Machel, H G University of Alberta, Edmonton, Alberta, Canada
Milke, R University of Basel, Basel, Switzerland
xxii CONTRIBUTORS
Milner, A R Birkbeck College, London, UK
Oneacre, J W BFI, Houston, TX, USA
Mojzsis, S J University of Colorado, Boulder, CO, USA
Orchard, M J Geological Survey of Canada Vancouver, BC, Canada
Monger, J W H Geological Survey of Canada, Vancouver, BC, Canada and Simon Fraser University Burnaby, BC, Canada
Orr, P J
Moore, P Selsey, UK
Owen, A W University of Glasgow, Glasgow, UK
Morris, N J
University College Dublin, Dublin, Ireland
The Natural History Museum, London, UK
Pälike, H Stockholm University, Stockholm, Sweden
Mortimer, N Institute of Geological and Nuclear Sciences, Dunedin New Zealand
Page, K N University of Plymouth, Plymouth, UK
Mountney, N P Keele University, Keele, UK
Paris, F University of Rennes 1, Rennes, France
Mpodozis, C SIPETROL SA, Santiago, Chile
Parker, J R Formerly Shell EP International, London, UK
Mungall, J E University of Toronto, Toronto, ON, Canada
Pfiffner, O A University of Bern, Bern, Switzerland
Myrow, P Colorado College, Colorado Springs, CO, USA
Piper, D J W Geological Survey of Canada, Dartmouth, NS, Canada
Naish, D University of Portsmouth, Portsmouth, UK
Price, R A Queens University Kingston, ON, Canada
Nickel, E H CSIRO Exploration and Mining, Wembley, WA, Australia
Prothero, D R Occidental College, Los Angeles, CA, USA
Nielsen, K C The University of Texas at Dallas, Richardson, TX, USA
Puche-Riart, O Polytechnic University of Madrid, Madrid, Spain
Nikishin, A M Lomonosov Moscow State University, Moscow, Russia
Pye, K
Nokleberg, W J United States Geological Survey, Menlo Park, CA, USA Norbury, D CL Associates, Wokingham, UK O'Brien, P J Universität Potsdam, Potsdam, Germany Ogg, J G Purdue University, West Lafayette, IN, USA
Royal Holloway, University of London, Egham, UK Rahn, P H South Dakota School of Mines and Technology Rapid City, SD, USA Ramos, V A Universidad de Buenos Aires, Buenos Aires, Argentina Rankin, A H Kingston University, Kingston-upon-Thames, UK
Oldershaw, C St. Albans, UK
Rebesco, M Istituto Nazionale di Oceanografia e di Geofisica Sperimentale (OGS), Italy
Oldroyd, D R University of New South Wales, Sydney, Australia
Reedman, A J Mapperley, UK
CONTRIBUTORS xxiii
Reisz, R R University of Toronto at Mississauga Mississauga, ON, Canada Retallack, G J University of Oregon, Eugene, OR, USA Rickards, R B University of Cambridge, Cambridge, UK Riding, R Cardiff University, Cardiff, UK Rigby, J K Brigham Young University, Provo, UT, USA Rigby, S University of Edinburgh, Edinburgh, UK Rodda, P Mineral Resources Department, Suva, Fiji Rona, P A Rutgers University, New Brunswick, NJ, USA Rose, E P F Royal Holloway, University of London, Egham, UK Rosenbaum, M S Twickenham, UK Rothwell, R G Southampton Oceanography Centre, Southampton, UK Roy, A B Presidency College, Kolkata, India
Rushton, A W A The Natural History Museum, London, UK Russell, A J University of Newcastle upon Tyne, Newcastle upon Tyne, UK Schmid, R ETH-centre, Zurich, Switzerland Scott, E National Center for Science Education Berkeley, CA, USA Scon, A C Royal Holloway, University of London, Egham, UK Scrutton, C T Formerly University of Durham, Durham, UK Searle, M University of Oxford, Oxford, UK
Searle, R C University of Durham, Durham, UK Seibold, I University Library, Freiburg, Germany Selley, R C Imperial College London, London, UK Sellwood, B W University of Reading, Reading, UK Shields, G A James Cook University, Townsville, OLD, Australia Simms, M J Ulster Museum, Belfast, UK Slipper, I J University of Greenwich, Chatham Maritime, UK Smallwood, J R Amerada Hess pic, London, UK Smith, A B The Natural History Museum, London, UK Smith, I Auckland University, Auckland, New Zealand Snoke, A W University of Wyoming, Laramie, WY, USA Soligo, C The Natural History Museum, London, UK Stein, S Northwestern University, Evanston, IL, USA Steinberger, B Japan Marine Science and Technology Center Yokosuka, Japan Stemmerik, L Geological Survey of Denmark and Greenland, Geocenter Copenhagen, Copenhagen, Denmark Stern, R J The University of Texas at Dallas, Richardson, TX, USA Stewart, I University of Plymouth, Plymouth, UK Storey, B C University of Canterbury, Christchurch, New Zealand Storrs, G W Cincinnati Museum Center, Museum of Natural History and Science, Cincinnati, OH, USA
xxiv
CONTRIBUTORS
Strachan, R A University of Portsmouth, Portsmouth, UK Suetsugu, D Japan Marine Science and Technology Center, Yokosuka Japan Surlyk, F University of Copenhagen, Geocenter Copenhagen, Copenhagen, Denmark Tait, J Ludwig-Maximilians-Universität, München, Germany Talbot, M R University of Bergen, Bergen, Norway Taylor, P D The Natural History Museum, London, UK Taylor, T N University of Kansas, Lawrence, KS, USA Taylor, W E G University of Lancaster, Lancaster, UK Tazawa, J Niigata University, Niigata, Japan Theodor, J M Illinois State Museum, Springfield, IL, USA Timmerman, M J Universität Potsdam, Potsdam, Germany Tollo, R P George Washington University, Washington, DC, USA Torsvik, T H Geological Survey of Norway, Trondheim, Norway Trendall, A Curtin University of Technology, Perth, Australia Trewin, N H University of Aberdeen, Aberdeen, UK Turner, A K Colorado School of Mines, Colorado, USA Twitchett, R J University of Plymouth, Plymouth, UK
van Geuns, L C Clingendael International Energy Programme The Hague, The Netherlands van Staal, C R Geological Survey of Canada, Ottawa, ON, Canada Vanecek, M Charles University Prague, Prague, Czech Republic Vaughan,D J University of Manchester, Manchester, UK Veevers, J J Macquarie University, Sydney, NSW, Australia Verniers, J University of Ghent, Ghent, Belgium Wadge, G University of Reading, Reading, UK Walter, M R Macquarie University, Sydney, NSW, Australia Wang, H China University of Geosciences, Beijing, China Ware, N G Australian National University, Canberra, ACT, Australia Warke, P A Queen's University Belfast, Belfast, UK Weber, K J Technical University, Delft, The Netherlands Welch, M D The Natural History Museum, London, UK Westbrook, G K University of Birmingham, Birmingham, UK Westermann, G E G McMaster University, Hamilton, ON, Canada Whalley, W B Queen's University Belfast, Belfast, UK White, N C Brisbane, OLD, Australia White, S M University of South Carolina, Columbia, SC, USA
Tyler, I M Geological Survey of Western Australia East Perth, WA, Australia
Wignall, P B University of Leeds, Leeds, UK
Valdes, P J University of Bristol, Bristol, UK
Williams, P A University of Western Sydney, Parramata, Australia
CONTRIBUTORS xxv
Wise, W S University of California-Santa Barbara Santa Barbara, CA, USA Worden, R H University of Liverpool, Liverpool, UK Wyatt, A R Sidmouth, UK Xiao, S Virginia Polytechnic Institute and State University Blacksburg, VA, USA
Yakubchuk, A S The Natural History Museum, London, UK Yates, A M University of the Witwatersrand, Johannesburg South Africa Zhang Shihong China University of Geosciences, Beijing, China Ziegler, P A University of Basel, Basel, Switzerland
CONTENTS xxvii
Contents Volume 1 A AFRICA Pan-African Orogeny A Kröner, R J Stern North African Phanerozoic S Lüning Rift Valley L Frostick
1 12 26
AGGREGATES
34
M A Eden, W J French
ALPS See EUROPE: The Alps ANALYTICAL METHODS Fission Track Analysis B W H Hendriks Geochemical Analysis (Including X-ray) R H Warden Geochronological Techniques E A Eide Gravity / R Smallwood Mineral Analysis N G Ware
43 54 77 92 107
ANDES
118
S M Kay, C Mpodozis, V A Ramos
ANTARCTIC
132
B C Storey
ARABIA AND THE GULF
/ A Al-Jallal, A S Al-Sharhan
140
VA Ramos
153
ASIA Central S G Lucas South-East / Metcalfe
164 169
ARGENTINA
ASTEROIDS See SOLAR SYSTEM: Asteroids, Comets and Space Dust ATMOSPHERE EVOLUTION
197
S J Mojzsis
AUSTRALIA Proterozoic / M Tyler Phanerozoic J J Veevers Tasman Orogenic Belt D R Gray, D A Foster
208 222 237
B BIBLICAL GEOLOGY BIODIVERSITY
E Byford
253
A W Owen
259
BIOLOGICAL RADIATIONS AND SPECIATION BIOSEDIMENTS AND BIOFILMS BIOZONES BRAZIL
P L Forey
266
M R Walter, A C Allwood
279
N MacLeod F F Alkmim, M A Martins-Neto
BUILDING STONE
A W Hatheway
294 306 328
xxviii
CONTENTS
c CALEDONIDE OROGENY See EUROPE: Caledonides Britain and Ireland; Scandinavian Caledonides (with Greenland) CARBON CYCLE
G A Shields
CHINA AND MONGOLIA CLAY MINERALS
335
H Wang, Shihong Zhang, Guoqi He
/ M Huggett
CLAYS, ECONOMIC USES
345 358
Y Fuchs
366
COCCOLITHS See CALCAREOUS ALGAE COLONIAL SURVEYS
A J Reedman
370
COMETS See SOLAR SYSTEM: Asteroids, Comets and Space Dust CONSERVATION OF GEOLOGICAL SPECIMENS CREATIONISM
L Cornish, G Comerford
E Scott
373 381
D DELTAS See SEDIMENTARY ENVIRONMENTS: Deltas DENDROCHRONOLOGY
M Bridge
387
DESERTS See SEDIMENTARY ENVIRONMENTS: Deserts DIAGENESIS, OVERVIEW
R C Selley
393
DINOSAURS See FOSSIL VERTEBRATES: Dinosaurs
E EARTH Mantle
Crust
GJH McCall
397
GJHMcCall
403
Orbital Variation (Including Milankovitch Cycles) EARTH STRUCTURE AND ORIGINS EARTH SYSTEM SCIENCE
H Palike
GJH McCall
R C Selley
410 421 430
EARTHQUAKES See ENGINEERING GEOLOGY: Aspects of Earthquakes; TECTONICS: Earthquakes ECONOMIC GEOLOGY
G R Davis
ENGINEERING GEOLOGY Overview M S Rosenbaum Codes of Practice D Nor bury Aspects of Earthquakes A W Hatheway Geological Maps / S Griffiths Geomorphology £ M Lee, J S Griffiths, P G Fookes Geophysics / K Gascoyne, A S Eriksen Seismology J J Bommer, D M Boore Natural and Anthropogenic Geohazards G J H McCall Liquefaction / F Bird, R W Boulanger, IM Idriss Made Ground / A Charles
434 444 448 456 463 474 482 499 515 525 535
CONTENTS xxix
Problematic Rocks F G Bell Problematic Soils F G Bell Rock Properties and Their Assessment F G Bell Site and Ground Investigation / R Greenwood
543 554 566 580
Volume 2 ENGINEERING GEOLOGY Site Classification A W Hatheway Subsidence A B Hawkins Ground Water Monitoring at Solid Waste Landfills ENVIRONMENTAL GEOCHEMISTRY ENVIRONMENTAL GEOLOGY
/ W Oneacre, D Figueras
W E Dubbin
P Doyle
1 9 14 21 25
EROSION See SEDIMENTARY PROCESSES: Erosional Sedimentary Structures; Aeolian Processes; Fluxes and Budgets EUROPE East European Craton R G Garetsky, S V Bogdanova, R Gorbatschev Timanides of Northern Russia D G Gee Caledonides of Britain and Ireland R A Strachan , J F Dewey Scandinavian Caledonides (with Greenland) D G Gee Variscan Orogeny W Franke, P Matte, J Tait The Urals D Brown, H Echtler Permian Basins A Henk, M J Timmerman Permian to Recent Evolution PA Ziegler The Alps O AP fiffner Mediterranean Tectonics £ Carminati, C Doglioni Holocent W Lemke, J HarffA
34 49 56 64 75 86 95 102 125 135 147
EVOLUTION
160
S Rigby, E MEharper
F FAKEFOSSILS
D I Martill
FAMOUS GEOLOGISTS Agassiz D R Oldroyd Cuvier G Laurent Darwin D R Oldroyd Du Toit / C Loock, D F Branagan
169 174 179 184 188
Hall R H Dott, Jr
194
Hutton D R Oldroyd Lyell D R Oldroyd Murchison D R Oldroyd Sedgwick D R Oldroyd Smith D R Oldroyd Steno / M Hansen Suess B Fritscher Walther I Seibold Wegener B Fritscher
200 206 210 216 221 226 233 242 246
FLUID INCLUSIONS
A H Rankin
253
xxx
CONTENTS
FORENSIC GEOLOGY
K Pye
261
FOSSIL INVERTEBRATES Arthropods LI Anderson Trilobites A WA Rushton Insects E A Jarzembowski Brachiopods D AT Harper Bryozoans P D Taylor Corals and Other Cnidaria C T Scrutton Echinoderms (Other Than Echinoids) A B Smith Crinoids M / Simms Echinoids A B Smith Graptolites R B Richards Molluscs Overview N J Morris Bivalves E M Harper Gastropods / Fry da Cephalopods (Other Than Ammonites) P Doyle Ammonites G E G Westermann Porifera / K Rigby
274 281 295 301 310 321 334 342 350 357 367 369 378 389 396 408
FOSSIL PLANTS Angiosperms P Kenrick Calcareous Algae / C Braga, R Riding Fungi and Lichens T N Taylor, M Krings Gymnosperms P Kenrick
418 428 436 443
FOSSIL VERTEBRATES Jawless Fish-Like Vertebrates D K Elliott Fish / A Long Palaeozoic Non-Amniote Tetrapods / A Clack Reptiles Other Than Dinosaurs R R Reisz Dinosaurs A M Yates Birds G / Dyke, L M Chiappe Swimming Reptiles G W Storrs Flying Reptiles D Naish, D M Martill Mesozoic Amphibians and Other Non-Amniote Tetrapods Cenozoic Amphibians A R Milner Mesozoic Mammals Z-X Luo Placental Mammals D R Prothero Hominids L R M Cocks
454 462 468 479 490 497 502 508 516 523 527 535 541
A R Milner
Volume 3
G GAIA
GJHMcCall
GEMSTONES
1
C Oldershaw
GEOARCHAEOLOGY
6
L Joyner
GEOCHEMICAL EXPLORATION GEOLOGICAL CONSERVATION GEOLOGICAL ENGINEERING
14 £ M Cameron / E Gordon A K Turner
21 29 35
CONTENTS xxxi
GEOLOGICAL FIELD MAPPING
P Canard
43
GEOLOGICAL MAPS AND THEIR INTERPRETATION GEOLOGICAL SOCIETIES
G L Merries Davies
GEOLOGICAL SURVEYS
65
G L Jones
73
S J Cribb
GEOLOGY OF WHISKY
53 60
P M Allen
GEOLOGY, THE PROFESSION GEOLOGY OF BEER
A Maltman
78
S J Cribb
82
GEOLOGY OF WINE / M Hancock† 85
85
GEOMORPHOLOGY
90
GEOMYTHOLOGY
P H Rahn A Mayor
96
GEOPHYSICS See EARTH: Orbital Variation (Including Milankovitch Cycles); EARTH SYSTEM SCIENCE; ENGINEERING GEOLOGY: Seismology; MAGNETOSTRATIGRAPHY; MOHO DISCONTINUITY; PALAEOMAGNETISM; PETROLEUM GEOLOGY: Exploration; REMOTE SENSING: Active Sensors; CIS; Passive Sensors; SEISMIC SURVEYS; TECTONICS: Seismic Structure at Mid-Ocean Ridges GEOTECHNICAL ENGINEERING GEYSERS AND HOT SPRINGS
D P Giles
100
G J H McCall
105
GLACIERS See SEDIMENTARY PROCESSES: Glaciers GOLD
MAMcKibben
118
GONDWANALAND AND GONDWANA
J J Veevers
128
GRANITE See IGNEOUS ROCKS: Granite GRENVILLIAN OROGENY
R P Tollo
155
H HERCYNIAN OROGENY See EUROPE: Variscan Orogeny HIMALAYAS See INDIAN SUBCONTINENT HISTORY OF GEOLOGY UP TO 1780
O Puche-Riart
167
HISTORY OF GEOLOGY FROM 1780 TO 1835
D R Oldroyd
173
HISTORY OF GEOLOGY FROM 1835 TO 1900
D R Oldroyd
179
HISTORY OF GEOLOGY FROM 1900 TO 1962
D F Branagan
185
HISTORY OF GEOLOGY SINCE 1962
U B Marvin
197
I
IGNEOUS PROCESSES IGNEOUS ROCKS Carbonatites K Bell Granite AIS Kemp Deceased
P D Asimow
209 217 233
xxxii
CONTENTS
Kimberlite Komatiite Obsidian
GJH McCall N TArndt, C M Lesher G / H McCall
IMPACT STRUCTURES
247 260 267
RAF Grieve
INDIAN SUBCONTINENT
277
A B Roy
285
J JAPAN
/ Tazawa
297
JUPITER See SOLAR SYSTEM: Jupiter, Saturn and Their Moons
L LAGERSTÄTTEN
S E Gabbott
LARGE IGNEOUS PROVINCES LAVA
307 M F Coffin, O Eldholm
N Geshi
315 323
M MAGNETOSTRATIGRAPHY
S G Lucas
MANTLE PLUMES AND HOT SPOTS
331
D Suetsugu, T Kogiso, B Steinberger
335
MARS See SOLAR SYSTEM: Mars MERCURY See SOLAR SYSTEM: Mercury MESOZOIC Triassic S G Lucas, M J Orchard Jurassic K N Page Cretaceous N MacLeod End Cretaceous Extinctions N MacLeod METAMORPHIC ROCKS Classification, Nomenclature and Formation Facies and Zones K Bucher PTt-Paths PJ O'Brien
344 352 360 372 G Hoinkes, C A Hauzenberger, R Schmid
386 402 409
METEORITES See SOLAR SYSTEM: Meteorites MICROFOSSILS Acritarchs K J Doming Chitinozoa F Paris, J Verniers Conodonts R J Aldridge Foraminifera M A Kaminski Ostracoda D / Home Palynology P Coxon, G Clayton
418 428 440 448 453 464
MICROPALAEONTOLOGICAL TECHNIQUES I J Slipper 470
470
MILANKOVITCH CYCLES See EARTH: Orbital Variation (Including Milankovitch Cycles) MILITARY GEOLOGY
EPF Rose
MINERAL DEPOSITS AND THEIR GENESIS
475 G R Davis
488
CONTENTS xxxiii
MINERALS Definition and Classification E H Nickel 498 Amphiboles R A Howie Arsenates K Hudson-Edwards 506 Borates C Helvaci Carbonates B Jones Chromates PA Williams Feldspars R A Howie Feldspathoids M D Welch Glauconites J M Huggett 542 Micas R A Howie Molybdates P A Williams Native Elements P A Williams Nitrates PA Williams Olivines G Cressey, R A Howie Other Silicates R A Howie Phosphates See SEDIMENTARY ROCKS: Phosphates Pyroxenes R A Howie Quartz R A Howie Sulphates G Cressey Sulphides D J Vaughan Tungstates P A Williams Vanadates P A Williams Zeolites W S Wise Zircons G J H McCall
498 503 506 510 522 532 534 539 542
548 551 553 555 557 561 567 569 572 574 586 588 591 601
MINING GEOLOGY Exploration Boreholes M Vanecek Exploration N C White Mineral Reserves M Vanecek Hydrothermal Ores M A McKibben Magmatic Ores / £ Mungall
609 613 623 628 637
MOHO DISCONTINUITY
645
P Giese
MOON See SOLAR SYSTEM: Moon
Volume 4
N NEW ZEALAND N Mortimer NORTH AMERICA Precambrian Continental Nucleus W Bleeker Continental Interior D F Merriam Northern Cordillera J W H Monger, R A Price, W J Nokleberg 36 Southern Cordillera AWSnoke Ouachitas K C Nielsen Southern and Central Appalachians R D Hatcher, Jr Northern Appalachians C R van Staal Atlantic Margin D R Hutchinson
1
8 21 36 48 61 72 81 92
xxxiv CONTENTS
o OCEANIA (INCLUDING FIJI, PNG AND SOLOMONS) I Smith, E Garaebiti, P Rodda ORIGIN OF LIFE
H Davies, P Bani, P Black,
/ Bailey
109 123
p PALAEOCLIMATES
B W Sellwood, P J Valdes
131
PALAEOECOLOGY E M Harper, S Rigby
140
PALAEOMAGNETISM
147
PALAEONTOLOGY PALAEOPATHOLOGY
T H Torsvik L R M Cocks
156
S G Lucas
160
PALAEOZOIC Cambrian N C Hughes, N A Heim Ordovician R A Fortey Silurian L R M Cocks Devonian G R McGhee Carboniferous A C Scott Permian P B Wignall End Permian Extinctions RJ Twitchett
163 175 184 194 200 214 219
PANGAEA
225
S G Lucas
PETROLEUM GEOLOGY Overview / Gluyas Chemical and Physical Properties C Clayton Gas Hydrates M Hovland The Petroleum System C Cornford 268 Exploration / R Parker Production KJ Weber, L C van Geuns Reserves R Arnott PLATE TECTONICS
R C Searle
229 248 261 268
295 308 331 340
PRECAMBRIAN Overview L R M Cocks Eukaryote Fossils S Xiao Prokaryote Fossils M D Brasier Vendian and Ediacaran MAS McMenamin 371
371
PSEUDOFOSSILS
D M Martill
382
PYROCLASTICS
R J Brown, E S Calder
386
350 354 363
Q QUARRYING A W Hatheway
399
R REEFS See SEDIMENTARY ENVIRONMENTS: Reefs ("Build-Ups") REGIONAL METAMORPHISM
A Feenstra, G Franz
407
CONTENTS xxxv
REMOTE SENSING Active Sensors G Wadge CIS P J Mason Passive Sensors / G Liu
414 420 431
RIFT VALLEYS See AFRICA: Rift Valley ROCK MECHANICS
JP Harrison
ROCKS AND THEIR CLASSIFICATION RUSSIA
440 R C Selley
A S Yakubchuk, A M Nikishin
452 456
s SATURN See SOLAR SYSTEM: Jupiter, Saturn and Their Moons SEAMOUNTS
S M White
475
SEDIMENTARY ENVIRONMENTS Depositional Systems and Fades J Collinson Alluvial Fans, Alluvial Sediments and Settings K D Jäger Anoxic Environments P B Wignall Carbonate Shorelines and Shelves D W J Bosence Contourites M Rebesco Deltas T Elliott Deserts N P Mountney Lake Processes and Deposits M R Talbot Reefs ('Build-Ups') B W Sellwood Shoreline and Shoreface Deposits J How ell Storms and Storm Deposits P Myrow
485 492 495 501 513 528 539 550 562 570 580
SEDIMENTARY PROCESSES Erosional Sedimentary Structures J Collinson Depositional Sedimentary Structures / Collinson Post-Depositional Sedimentary Structures / Collinson Aeolian Processes N Lancaster Catastrophic Floods A J Russell Deep Water Processes and Deposits D J W Piper Fluvial Geomorphology / Lewin, P A Brewer Glaciers M / Hambrey, N F Glasser Karst and Palaeokarst M J Simms Landslides S F Burns
587 593 602 612 628 641 650 663 678 687
Volume 5 SEDIMENTARY PROCESSES Particle-Driven Subaqueous Gravity Processes M Felix, W McCaffrey 1 Deposition from Suspension IN McCave Fluxes and Budgets L Frostick
1 8 17
SEDIMENTARY ROCKS Mineralogy and Classification R C Selley Banded Iron Formations A Trendall Chalk / R Ineson, L Stemmerik, F Surlyk Chert N H Trewin, S R Payers
25 37 42 51
xxxvi CONTENTS
Clays and Their Diagenesis / M Huggett Deep Ocean Pelagic Oozes R G Rothwell Dolomites H G Machel Evaporites A C Kendall Ironstones W E G Taylor Limestones R C Selley Oceanic Manganese Deposits D S Cronan Phosphates W D Birch Rudaceous Rocks / McManus Sandstones, Diagenesis and Porosity Evolution SEISMIC SURVEYS
Gluyas
M Bacon
SEQUENCE STRATIGRAPHY SHIELDS
J
62 70 79 94 97 107 113 120 129 141 151
P P Mclaughlin, Jr
K C Condie
159 173
SHOCK METAMORPHISM P S DeCarli
179
SOIL MECHANICS / Atkinson
184
SOILS Modern Palaeosols
194 203
G J Retallack G J Retallack
SOLAR SYSTEM The Sun K R Lang Asteroids, Comets and Space Dust P Moore Meteorites G J H McCall Mercury G J H McCall Venus M A Ivanov, J W Head Moon P Moore Mars M R Walter, A J Brown, S A Chamberlain Jupiter, Saturn and Their Moons P Moore Neptune, Pluto and Uranus P Moore
209 220 228 238 244 264 272 282 289
SPACE DUST See SOLAR SYSTEM: Asteroids, Comets and Space Dust STRATIGRAPHICAL PRINCIPLES
N MacLeod
295
STROMATOLITES See BIOSEDIMENTS AND BIOFILMS SUN See SOLAR SYSTEM: The Sun
T TECTONICS Convergent Plate Boundaries and Accretionary Wedges G K Westbrook Earthquakes G J H McCall Faults S Stein Folding / W Cosgrove Fractures (Including Joints) / W Cosgrove Hydrothermal Activity R P Lowell, P A Rona Mid-Ocean Ridges K C Macdonald Hydrothermal Vents At Mid-Ocean Ridges R M Haymon Propagating Rifts and Microplates At Mid-Ocean Ridges R N Hey Seismic Structure At Mid-Ocean Ridges S M Carbotte Mountain Building and Orogeny M Searle Neotectonics I Stewart
307 318 330 339 352 362 372 388 396 405 417 425
CONTENTS xxxvii
Ocean Trenches R J Stern Rift Valleys L Frostick TEKTITES
428 437
G J H McCall
TERRANES OVERVIEW
443 L R M Cocks
TERTIARY TO PRESENT Paleocene J J Hooker 459 Eocene / / Hooker Oligocene D R Prothero Miocene J M Theodor 478 Pliocene C Soligo Pleistocene and The Ice Age
459
466 472 478
A Currant
THERMAL METAMORPHISM TIME SCALE TRACE FOSSILS
455
R Abart, R Milke
F M Gradstein, J G Ogg
486 493 499 503
P J Orr
520
u ULTRA HIGH PRESSURE METAMORPHISM H-J Massonne 533
533
UNCONFORMITIES
541
A R Wyatt
UNIDIRECTIONAL AQUEOUS FLOW
/ Best
548
URALS See EUROPE: The Urals URBAN GEOLOGY A W Hatheway 557
557
V VENUS See SOLAR SYSTEM: Venus VOLCANOES
G J H McCall
565
W WEATHERING
Index
W B Whalley, P A Warke
5 81
591
AFRICA/Pan-African Orogeny 1
AFRICA Contents Pan-African Orogeny North African Phanerozoic Rift Valley
Pan-African Orogeny A Kro¨ner, Universita¨t Mainz, Mainz, Germany R J Stern, University of Texas-Dallas, Richardson TX, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The term ‘Pan-African’ was coined by WQ Kennedy in 1964 on the basis of an assessment of available Rb–Sr and K–Ar ages in Africa. The Pan-African was interpreted as a tectono-thermal event, some 500 Ma ago, during which a number of mobile belts formed, surrounding older cratons. The concept was then extended to the Gondwana continents (Figure 1) although regional names were proposed such as Brasiliano for South America, Adelaidean for Australia, and Beardmore for Antarctica. This thermal event was later recognized to constitute the final part of an orogenic cycle, leading to orogenic belts which are currently interpreted to have resulted from the amalgamation of continental domains during the period 870 to 550 Ma. The term Pan-African is now used to describe tectonic, magmatic, and metamorphic activity of Neoproterozoic to earliest Palaeozoic age, especially for crust that was once part of Gondwana. Because of its tremendous geographical and temporal extent, the Pan-African cannot be a single orogeny but must be a protracted orogenic cycle reflecting the opening and closing of large oceanic realms as well as accretion and collision of buoyant crustal blocks. Pan-African events culminated in the formation of the Late Neoproterozoic supercontinent Gondwana (Figure 1). The Pan-African orogenic cycle is timeequivalent with the Cadomian Orogeny in western and central Europe and the Baikalian in Asia; in fact, these parts of Europe and Asia were probably part of Gondwana in pre-Palaeozoic times as were small Neoproterozoic crustal fragments identified in Turkey, Iran and Pakistan (Figure 1).
Within the Pan-African domains, two broad types of orogenic or mobile belts can be distinguished. One type consists predominantly of Neoproterozoic supracrustal and magmatic assemblages, many of juvenile (mantlederived) origin, with structural and metamorphic histories that are similar to those in Phanerozoic collision and accretion belts. These belts expose upper to middle crustal levels and contain diagnostic features such as ophiolites, subduction- or collision-related granitoids, island-arc or passive continental margin assemblages as well as exotic terranes that permit reconstruction of their evolution in Phanerozoic-style plate tectonic scenarios. Such belts include the Arabian-Nubian shield of Arabia and north-east Africa (Figure 2), the Damara– Kaoko–Gariep Belt and Lufilian Arc of south-central and south-western Africa, the West Congo Belt of Angola and Congo Republic, the Trans-Sahara Belt of West Africa, and the Rokelide and Mauretanian belts along the western part of the West African Craton (Figure 1). The other type of mobile belt generally contains polydeformed high-grade metamorphic assemblages, exposing middle to lower crustal levels, whose origin, environment of formation and structural evolution are more difficult to reconstruct. The protoliths of these assemblages consist predominantly of much older Mesoproterozoic to Archaean continental crust that was strongly reworked during the Neoproterozoic. Well studied examples are the Mozambique Belt of East Africa, including Madagascar (Figure 2) with extensions into western Antarctica, the Zambezi Belt of northern Zimbabwe and Zambia and, possibly, the little known migmatitic terranes of Chad, the Central African Republic, the Tibesti Massif in Libya and the western parts of Sudan and Egypt (Figure 1). It has been proposed that the latter type of belt represents the deeply eroded part of a collisional orogen and that the two types of Pan-African belts are not fundamentally different but constitute different crustal levels of collisional and/or accretional systems. For this reason, the term East African Orogen has been proposed for the combined upper crustal Arabian-Nubian Shield and lower crustal Mozambique Belt (Figure 2).
2 AFRICA/Pan-African Orogeny
Figure 1 Map of Gondwana at the end of Neoproterozoic time (540 Ma) showing the general arrangement of Pan African belts. AS, Arabian Shield; BR, Brasiliano; DA, Damara; DM, Dom Feliciano; DR, Denman Darling; EW, Ellsworth Whitmore Mountains; GP, Gariep; KB, Kaoko; MA, Mauretanides; MB, Mozambique Belt; NS, Nubian Shield; PM, Peterman Ranges; PB, Pryolz Bay; PR, Pampean Ranges; PS, Paterson; QM, Queen Maud Land; RB, Rokelides; SD, Saldania; SG, Southern Granulite Terrane; TS, Trans Sahara Belt; WB, West Congo; ZB, Zambezi. (Reproduced with permission from Kusky et al., 2003.)
The Pan-African system of orogenic belts in Africa, Brazil and eastern Antarctica has been interpreted as a network surrounding older cratons (Figure 1) and essentially resulting from closure of several major Neoproterozoic oceans. These are the Mozambique Ocean between East Gondwana (Australia, Antarctica, southern India) and West Gondwana (Africa, South America), the Adamastor Ocean between Africa and South America, the Damara Ocean between the Kalahari and Congo cratons, and the Trans-Sahara Ocean between the West African Craton and a poorly known pre-Pan-African terrane in north-central Africa variously known as the Nile or Sahara Craton (Figure 1).
Arabian-Nubian Shield (ANS) A broad region was uplifted in association with Cenozoic rifting to form the Red Sea, exposing a large tract of mostly juvenile Neoproterozoic crust. These exposures comprise the Arabian-Nubian Shield (ANS). The ANS makes up the northern half of the East African orogen and stretches from southern Israel and Jordan south as far as Ethiopia and Yemen, where the ANS transitions into the Mozambique Belt (Figure 2). The
ANS is distinguished from the Mozambique Belt by its dominantly juvenile nature, relatively low grade of metamorphism, and abundance of island-arc rocks and ophiolites. The ANS, thus defined, extends about 3000 km north to south and >500 km on either side of the Red Sea (Figure 3). It is flanked to the west by a broad tract of older crust that was remobilized during Neoproterozoic time along with a significant amount of juvenile Neoproterozoic crust, known as the Nile Craton or ‘Saharan Metacraton’. The extent of juvenile Neoproterozoic crust to the east in the subsurface of Arabia is not well defined, but it appears that PanAfrican crust underlies most of this region. Scattered outcrops in Oman yielded mostly Neoproterozoic radiometric ages for igneous rocks, and there is no evidence that a significant body of pre-Pan-African crust underlies this region. The ANS is truncated to the north as a result of rifting at about the time of the Precambrian–Cambrian boundary, which generated crustal fragments now preserved in south-east Europe, Turkey and Iran. The ANS is by far the largest tract of mostly juvenile Neoproterozoic crust among the regions of Africa that were affected by the Pan-African orogenic cycle. It
AFRICA/Pan-African Orogeny 3
Figure 2 Pre Jurassic configuration of elements of the East African Orogen in Africa and surrounding regions. Regions in clude Egypt (Eg), Sudan (Su), Sinai Israel Jordan (SIJ), Afif ter rane, Arabia (Aa), rest of Arabian Shield (Ar), Eritrea and northern Ethiopia (En), southern Ethiopia (Es), eastern Ethiopia, Somalia, and Yemen (ESY), Kenya (K), Tanzania (T), and Madagascar (M). Numbers in italics beneath each region label are mean Nd model ages in Gy.
formed as a result of a multistage process, whereby juvenile crust was produced above intra-oceanic convergent plate boundaries (juvenile arcs) and perhaps oceanic plateaux (ca. 870–630 Ma), and these juvenile terranes collided and coalesced to form larger composite terranes (Figure 4). There is also a significant amount of older continental crust (Mesoproterozoic age crust of the Afif terrane in Arabia; Palaeoproterozoic and Archaean crust in Yemen, Figure 2) that was overprinted by Pan-African tectonomagmatic events. ANS terrane boundaries (Figure 3) are frequently defined by suture zones that are marked by ophiolites, and the terranes are stitched together by abundant tonalitic to granodioritic plutons. Most ANS ophiolites have trace element chemical compositions suggesting formation above a convergent plate margin, either as part of a back-arc basin or in a fore-arc setting. Boninites have
been identified in Sudan and Eritrea and suggest a forearc setting for at least some ANS sequences. Sediments are mostly immature sandstones and wackes derived from nearby arc volcanoes. Deposits that are diagnostic of Neoproterozoic ‘snowball Earth’ episodes have been recognized in parts of the ANS, and banded iron formations in the northern ANS may be deep-water expressions of snowball Earth events. Because it mostly lies in the Sahara and Arabian deserts, the ANS has almost no vegetation or soil and is excellently exposed. This makes it very amenable to study using imagery from remote sensing satellites. Juvenile crust of the ANS was sandwiched between continental tracts of East and West Gondwana (Figure 4). The precise timing of the collision is still being resolved, but appears to have occurred after 630 Ma when high-magnesium andesite ‘schistose dykes’ were emplaced in southern Israel but before the 610 Ma post-tectonic ‘Mereb’ granites were emplaced in northern Ethiopia. By analogy with the continuing collision between India and Asia, the terminal collision between East and West Gondwana may have continued for a few tens of millions of years. Deformation in the ANS ended by the beginning of Cambrian time, although it has locally continued into Cambrian and Ordovician time farther south in Africa. The most intense collision (i.e. greatest shortening, highest relief, and greatest erosion) occurred south of the ANS, in the Mozambique belt. Compared to the strong deformation and metamorphism experienced during collision in the Mozambique belt, the ANS was considerably less affected by the collision. North-west trending leftlateral faults of the Najd fault system of Arabia and Egypt (Figures 1 and 2) formed as a result of escape tectonics associated with the collision and were active between about 630 and 560 Ma. Deformation associated with terminal collision is more intense in the southern ANS, with tight, upright folds, steep thrusts, and strike-slip shear zones controlling basement fabrics in Eritrea, Ethiopia, and southern Arabia. These north–south trending, collision-related structures obscure the earlier structures in the southern ANS that are related to arc accretion, and the intensity of this deformation has made it difficult to identify ophiolitic assemblages in southern Arabia, Ethiopia, and Eritrea. Thus, the transition between the ANS and the Mozambique Belt is marked by a change from less deformed and less metamorphosed, juvenile crust in the north to more deformed and more metamorphosed, remobilized older crust in the south, with the structural transition occurring farther north than the lithological transition. The final stages in the evolution of the ANS witnessed the emplacement of post-tectonic ‘A-type’
4 AFRICA/Pan-African Orogeny
Figure 3 Terrane map of the Arabian Nubian Shield. (Reproduced with permission from Johnson PR and Woldehaimanot B (2003) Development of the Arabian Nubian Shield: perspectives on accretion and deformation in the northern East African Orogen and the assembly of Gondwana. In: Yoshida M, Windley BF and Dasgupta S (eds.) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications 206, pp. 289 325.)
granites, bimodal volcanics, and molassic sediments. These testify to strong extension caused by orogenic collapse at the end of the Neoproterozoic. Extensionrelated metamorphic and magmatic core complexes are recognized in the northern ANS but are even more likely to be found in the more deformed regions of the southern ANS and the Mozambique Belt. A well developed peneplain developed on top of the ANS crust before basal Cambrian sediments were deposited, possibly cut by a continental ice-sheet. The ANS has been the source of gold since Pharaonic Egypt. There is now a resurgence of mining and exploration activity, especially in Sudan, Arabia, Eritrea, and Ethiopia.
Mozambique Belt (MB) This broad belt defines the southern part of the East African Orogen and essentially consists of medium- to high-grade gneisses and voluminous granitoids. It extends south from the Arabian-Nubian Shield into southern Ethiopia, Kenya and Somalia via Tanzania to Malawi and Mozambique and also includes Madagascar (Figure 2). Southward continuation of the belt into Dronning Maud Land of East Antarctica (Figure 1) has been proposed on the basis of geophysical patterns, structural features and geochronology. Most parts of the belt are not covered by detailed mapping, making regional correlations difficult. There is no
AFRICA/Pan-African Orogeny 5
Figure 4 A diagram of the suggested evolution of the Arabian Nubian Shield.
Figure 5 A schematic block diagram showing tectonic interdigitation of basement and cover rocks in the Mozambique Belt of Kenya. (Reproduced with permission from Mosley PN (1993) Geological evolution of the Late Proterozoic ‘Mozambique Belt’ of Kenya. Tectonophysics 221: 223 250.)
overall model for the evolution of the MB although most workers agree that it resulted from collision between East and West Gondwana. Significant differences in rock type, structural style, age and metamorphic evolution suggest that the belt as a whole constitutes a Pan-African Collage of terranes accreted to the eastern margin of the combined Congo and Tanzania cratons and that significant volumes of older crust of these cratons were reconstituted during this event. Mapping and geochronology in Kenya have recognized undated Neoproterozoic supracrustal sequences that are structurally sandwiched between basement gneisses of Archaean and younger age (Figure 5). A 700 Ma dismembered ophiolite complex at the
Kenyan/Ethiopian border testifies to the consumption and obduction of marginal basin oceanic crust. Major deformation and high-grade metamorphism is ascribed to two major events at 830 and 620 Ma, based on Rb–Sr dating, but the older of these appears questionable. A similar situation prevails in Tanzania where the metamorphic grade is generally high and many granulite-facies rocks of Neoproterozoic age show evidence of retrogression. Unquestionable Neoproterozoic supracrustal sequences are rare, whereas Late Archaean to Palaeoproterozoic granitoid gneisses volumetrically greatly dominate over juvenile Pan-African intrusives. These older rocks, strongly reworked during
6 AFRICA/Pan-African Orogeny
the Pan-African orogenic cycle and locally migmatized and/or mylonitized, either represent eastward extensions of the Tanzania Craton that were structurally reworked during Pan-African events or are separate crustal entities (exotic blocks) of unknown origin. The significance of rare granitoid gneisses with protolith ages of 1000–1100 Ma in southern Tanzania and Malawi is unknown. From these, some workers have postulated a major Kibaran (Grenvillian) event in the MB, but there is no geological evidence to relate these rocks to an orogeny. A layered gabbroanorthosite complex was emplaced at 695 Ma in Tanzania. The peak of granulite-facies metamorphism was dated at 620–640 Ma over wide areas of the MB in Tanzania, suggesting that this was the major collision and crustal-thickening event in this part of the belt. In northern Mozambique the high-grade gneisses, granulites and migmatites of the MB were interpreted to have been deformed and metamorphosed during two distinct events, namely the Mozambican cycle at 1100–850 Ma, also known as Lurian Orogeny, and the Pan-African cycle at 800–550 Ma. Recent highprecision zircon geochronology has confirmed the older event to represent a major phase of granitoid plutonism, including emplacement of a large layered gabbro-anorthosite massif near Tete at 1025 Ma, but there is as yet no conclusive evidence for deformation and granulite-facies metamorphism in these rocks during this time. The available evidence points to only one severe event of ductile deformation and high-grade metamorphism, with a peak some 615–540 Ma ago. A similar situation prevails in southern Malawi where high-grade granitoid gneisses with protolith ages of 1040–555 Ma were ductilely deformed together with supracrustal rocks and the peak of granulite-facies metamorphism was reached 550–570 Ma ago. The Pan-African terrane of central and southern Madagascar primarily consists of high-grade orthoand paragneisses as well as granitoids. Recent highprecision geochronology has shown that these rocks are either Archaean or Neoproterozoic in age and were probably structurally juxtaposed during PanAfrican deformation. Several tectonic provinces have been recognized (Figure 6), including a domain consisting of low-grade Mesoproterozoic to Early Neoproterozoic metasediments known as the Itremo group which was thrust eastwards over high-grade gneisses. A PanAfrican suture zone has been postulated in eastern Madagascar, the Betsimisaraka Belt (Figure 6), consisting of highly strained paragneisses decorated with lenses of mafic–ultramafic bodies containing podiform chromite and constituting a lithological and isotopic boundary with the Archaean gneisses and granites of the Antongil block east of this postulated suture which may correlate with similar rocks in southern India.
Figure 6 A simplified geological map showing the major tectonic units of the Precambrian basement in Madagascar. Rs, Ranotsara Shear Zone; BSZ, Betsileo Shear Zone. (Reproduced with permission from Collins and Windley 2002.)
Central and northern Madagascar are separated from southern Madagascar by the Ranotsara Shear Zone (Figure 6), showing sinistral displacement of >100 km and correlated with one of the major shear zones in southern India. Southern Madagascar consists of several north–south trending shear-bounded
AFRICA/Pan-African Orogeny 7
Figure 7 Histogram of radiometric ages for the Mozambique Belt of East Africa and Madagascar. Data from Meert JG (2003) A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics 362: 1 40, with updates.
tectonic units consisting of upper amphibolite to granulite-facies para- and orthogneisses, partly of pre-Neoproterozoic age. The peak of granulite-facies metamorphism in central and southern Madagascar, including widespread formation of charnockites, was dated at 550–560 Ma. The distribution of zircon radiometric ages in the MB suggests two distinct peaks at 610–660 and 530–570 Ma (Figure 7) from which two orogenic events have been postulated, the older East African Orogeny (660–610 Ma) and the younger Kuunga Orogeny (570–530 Ma). However, the are no reliable field criteria to distinguish between these postulated phases, and it is likely that the older age group characterizes syntectonic magmatism whereas the younger age group reflects post-tectonic granites and pegmatites which are widespread in the entire MB.
Zambezi Belt The Zambezi Belt branches off to the west from the Mozambique Belt in northernmost Zimbabwe along what has been described as a triple junction and extends into Zambia (Figures 1 and 8). It consists predominantly of strongly deformed amphibolite- to granulite-facies, early Neoproterozoic ortho- and paragneisses which were locally intruded by 860 Ma, layered gabbro-anorthosite bodies and generally displays south-verging thrusting and transpressional shearing. Lenses of eclogite record pressures up to
23 kbar. Although most of the above gneisses seem to be 850–870 Ma in age, there are tectonically interlayered granitoid gneisses with zircon ages around 1100 Ma. The peak of Pan-African metamorphism occurred at 540–535 Ma. The Zambezi Belt is in tectonic contact with lower-grade rocks of the Lufilian Arc in Zambia along the transcurrent Mwembeshi shear zone.
Lufilian Arc The Lufilian Arc (Figure 8) has long been interpreted to be a continuation of the Damara Belt of Namibia, connected through isolated outcrops in northern Botswana (Figure 1). The outer part of this broad arc in the Congo Republic and Zambia is a north-eastverging thin-skinned, low-grade fold and thrust belt, whereas the higher-grade southern part is characterized by basement-involved thrusts. The main lithostratigraphic unit is the Neoproterozoic, copper-bearing Katanga succession which contains volcanic rocks dated between 765 and 735 Ma. Thrusting probably began shortly after deposition, and the main phase of thrusting and associated metamorphism occurred at 566–550 Ma.
Damara Belt This broad belt exposed in central and northern Namibia branches north-west and south-east near
8 AFRICA/Pan-African Orogeny
Figure 8 A simplified geological map of the Lufilian Arc and Zambezi Belt. (Reproduced with permission from Porada H and Berhorst V (2000) Towards a new understanding of the Neoproterozoic early Palaeozoic Lufilian and northern Zambezi belts in Zambia and the Democratic Republic of Congo.)
the Atlantic coast and continues southwards into the Gariep and Saldania belts and northwards into the Kaoko Belt (Figure 1). The triple junction so produced may have resulted from closure of the Adamastor Ocean, followed by closure of the Damara Ocean. The main lithostratigraphic unit is the Damara supergroup which records basin formation and rift-related magmatism at 760 Ma, followed by the formation of a broad carbonate shelf in the north and a turbidite basin in the south. The turbidite sequence contains interlayered, locally pillowed, amphibolites and metagabbros which have been interpreted as remnants of a dismembered ophiolite. Of particular interest are two distinct horizons of glaciogenic rocks which can probably be correlated with similar strata in the Katanga sequence of south central Africa and reflect a severe glaciation currently explained by the snowball Earth hypothesis. The Damara Belt underwent north- and southverging thrusting along its respective margins, whereas the deeply eroded central zone exposes medium- to high-grade ductilely deformed rocks, widespread migmatization and anatexis in which both the Damara
supracrustal sequence and a 1.0–2.0 Ga old basement are involved. Sinistral transpression is seen as the cause for this orogenic event which reached its peak at 550–520 Ma. Voluminous pre-, syn- and posttectonic granitoid plutons intruded the central part of the belt between 650 and 488 Ma, and highly differentiated granites, hosting one of the largest opencast uranium mines in the world (Ro¨ ssing), were dated at 460 Ma. Uplift of the belt during the Damaran Orogeny led to erosion and deposition of two Late Neoproterozoic to Early Palaeozoic clastic molasse sequences, the Mulden group in the north and the Nama group in the south. The latter contains spectacular examples of the Late Neoproterozoic Ediacara fauna.
Gariep and Saldania Belts These belts fringe the high-grade basement along the south-western and southern margin of the Kalahari craton (Figure 1) and are interpreted to result from oblique closure of the Adamastor Ocean. Deep marine fan and accretionary prism deposits, oceanic
AFRICA/Pan-African Orogeny 9
seamounts and ophiolitic assemblages were thrust over Neoproterozoic shelf sequences on the craton margin containing a major Zn mineralization just north of the Orange River in Namibia. The main deformation and metamorphism occurred at 570–540 Ma, and post-tectonic granites were emplaced 536–507 Ma ago. The famous granite at Sea Point, Cape Town, which was described by Charles Darwin, belongs to this episode of Pan-African igneous activity.
Kaoko Belt This little known Pan-African Belt branches off to the north-west from the Damara Belt and extends into south-western Angola. Here again a well developed Neoproterozoic continental margin sequence of the Congo Craton, including glacial deposits, was overthrust, eastwards, by a tectonic mixture of prePan-African basement and Neoproterozoic rocks during an oblique transpressional event following closure of the Adamastor Ocean. A spectacular shear zone, the mylonite-decorated Puros lineament, exemplifies this event and can be followed into southern Angola. Highgrade metamorphism and migmatization dated between 650 and 550 Ma affected both basement and cover rocks, and granitoids were emplaced between 733 and 550 Ma. Some of the strongly deformed basement rocks have ages between 1450 and 2030 Ma and may represent reworked material of the Congo Craton, whereas a small area of Late Archaean granitoid gneisses may constitute an exotic terrane. The western part of the belt consists of large volumes of ca. 550 Ma crustal melt granites and is poorly exposed below the Namib sand dunes. No island-arc, ophiolite or high-pressure assemblages have been described from the Kaoko Belt, and current tectonic models involving collision between the Congo and Rio de la Plata cratons are rather speculative.
West Congo Belt This belt resulted from rifting between 999 and 912 Ma along the western margin of the Congo Craton (Figure 1), followed by subsidence and formation of a carbonate-rich foreland basin, in which the West Congolian group was deposited between ca. 900 and 570 Ma, including two glaciogenic horizons similar to those in the Katangan sequence of the Lufilian Arc. The structures are dominated by east-verging deformation and thrusting onto the Congo Craton, associated with dextral and sinistral transcurrent shearing, and metamorphism is low to medium grade. In the west, an allochthonous thrust-and-fold stack of Palaeo- to Mesoproterozoic basement rocks overrides the West Congolian foreland sequence. The West
Congo Belt may only constitute the eastern part of an orogenic system with the western part, including an 800 Ma ophiolite, exposed in the Aracuaı´ Belt of Brazil.
Trans-Saharan Belt This orogenic Belt is more than 3000 km long and occurs to the north and east of the >2 Ga West African Craton within the Anti-Atlas and bordering the Tuareg and Nigerian shields (Figure 1). It consists of preNeoproterozoic basement strongly reworked during the Pan-African event and of Neoproterozoic oceanic assemblages. The presence of ophiolites, accretionary prisms, island-arc magmatic suites and high-pressure metamorphic assemblages makes this one of the best documented Pan-African belts, revealing ocean opening, followed by a subduction- and collision-related evolution between 900 and 520 Ma (Figure 9). In southern Morocco, the 740–720 Ma Sirwa-Bou Azzer ophiolitic me´lange was thrust southwards, at 660 Ma, over a Neoproterozoic continental margin sequence of the West African Craton, following northward subduction of oceanic lithosphere and preceding oblique collision with the Saghro Arc. Farther south, in the Tuareg Shield of Algeria, Mali and Niger, several terranes with contrasting lithologies and origins have been recognized, and ocean closure during westward subduction produced a collision belt with Pan-African rocks, including oceanic terranes tectonically interlayered with older basement. The latter were thrust westwards over the West African Craton and to the east over the so-called LATEA (Laouni, Azrou-n-Fad, Tefedest, and Ege´re´Aleksod, parts of a single passive margin in central Hoggar) Superterrane, a completely deformed composite crustal segment consisting of Archaean to Neoproterozoic assemblages (Figure 9). In Mali, the 730–710 Ma Tilemsi magmatic arc records oceanfloor and intra-oceanic island-arc formation, ending in collision at 620–600 Ma. The southern part of the Trans-Saharan Belt is exposed in Benin, Togo and Ghana where it is known as the Dahomeyan Belt. The western part of this belt consists of a passive margin sedimentary sequence in the Volta basin which was overthrust, from the east, along a well delineated suture zone by an ophiolitic me´lange and by a 613 My old high-pressure metamorphic assemblage (up to 14 kbar, 700 C), including granulites and eclogites. The eastern part of the belt consists of a high-grade granitoid–gneiss terrane of the Nigerian province, partly consisting of Palaeoproterozoic rocks which were migmatized at 600 Ma. This deformation and metamorphism is considered to have resulted from oblique collision of
10 AFRICA/Pan-African Orogeny
Figure 9 Diagrams showing the geodynamic evolution of western central Hoggar (Trans Sahara Belt) between 900 and 520 Ma. Stars denote high pressure rocks now exposed. (Reproduced with permission from Caby R (2003) Terrane assembly and geodynamic evolution of central western Hoggar: a synthesis.)
the Nigerian shield with the West African Craton, followed by anatectic doming and wrench faulting.
Pan-African Belt in Central Africa (Cameroon, Chad and Central African Republic) The Pan-African Belt between the Congo Craton in the south and the Nigerian basement in the north-west consists of Neoproterozoic supracrustal assemblages and variously deformed granitoids with tectonically interlayered wedges of Palaeoproterozoic basement (Figure 10). The southern part displays medium- to high-grade Neoproterozoic rocks, including 620 Ma granulites, which are interpreted to have formed in a continental collision zone and were thrust over the Congo Craton, whereas the central and northern parts expose a giant shear belt characterized by thrust and shear zones which have been correlated with similar structures in north-eastern Brazil and which are late collisional features. The Pan-African Belt continues eastward into the little known Oubanguide Belt of the Central African Republic.
Pan-African Reworking of Older Crust in North-Eastern Africa A large area between the western Hoggar and the river Nile largely consists of Archaean to Palaeoproterozoic
basement, much of which was structurally and thermally overprinted during the Pan-African event and intruded by granitoids. The terrane is variously known in the literature as ‘Nile Craton’, ‘East Sahara Craton’ or ‘Central Sahara Ghost Craton’ and is geologically poorly known. Extensive reworking is ascribed by some to crustal instability following delamination of the subcrustal mantle lithosphere, and the term ‘Sahara Metacraton’ has been coined to characterize this region. A ‘metacraton’ refers to a craton that has been remobilized during an orogenic event but is still recognizable through its rheological, geochronological and isotopic characteristics.
Rokelide Belt This belt occurs along the south-western margin of the Archaean Man Craton of West Africa (Figure 1) and is made up of high-grade gneisses, including granulites (Kasila group), lower-grade supracrustal sequences (Marampa group) and volcano-sedimentary rocks with calc-alkaline affinity (Rokel River group). Pan-African deformation was intense and culminated in extensive thrusting and sinistral strike-slip deformation. The peak of metamorphism reached 7 kb and 800 C and was dated at 560 Ma. Late Pan-African emplacement ages for the protoliths of some of the granitoid gneisses contradict earlier hypotheses arguing for extensive overprinting of
AFRICA/Pan-African Orogeny 11
Figure 10 A sketch map showing Pan African domains in west central Africa. 1, Post Pan African cover; 2, Pan African domains; 3, pre Mesozoic platform deposits; 4, Archaean to Palaeoproterozoic cratons; 5, craton limits; 6, major strike slip faults; 7, state boundaries. CAR, Central African Republic; CM, Cameroun. (Reproduced with permission from Toteu SF, Penaye J and Djomani YP (2004).)
Archaean rocks. The Rokelides may be an accretionary belt, but there are no modern structural data and only speculative geodynamic interpretations.
Gondwana Correlations The Pan-African orogenic cycle was the result of ocean closure, arc and microcontinent accretion and final suturing of continental fragments to form the supercontinent Gondwana. It has been suggested that the opening of large Neoproterozoic oceans between the Brazilian and African cratons (Adamastor Ocean), the West African and Sahara–Congo cratons (Pharusian Ocean) and the African cratons and India/ Antarctica (Mozambique Ocean) (Figure 1) resulted from breakup of the Rodinia supercontinent some 800–850 Ma, but current data indicate that the African and South American cratons were never part of Rodinia. Although arc accretion and continent formation in the Arabian-Nubian shield are reasonably well understood, this process is still very speculative in the Mozambique Belt. It seems clear that Madagascar, Sri Lanka, southern India and parts of East Antarctica were part of this process (Figure 1), although the exact correlations between these fragments are not known. The Southern Granulite
Terrane of India (Figure 1) consists predominantly of Late Archaean to Palaeoproterozoicc gneisses and granulites, deformed and metamorphosed during the Pan-African event and sutured against the Dharwar Craton. Areas in East Antarctica such as Lu¨ tzowHolm Bay, Central Dronning Maud Land and the Shackleton Range, previously considered to be Mesoproterozoic in age, are now interpreted to be part of the Pan-African Belt system (Figure 1). Correlations between the Pan-African belts in south-western Africa (Gariep–Damara–Kaoko) and the Brasiliano belts of south-eastern Brazil (Ribeira and Dom Feliciano) are equally uncertain, and typical hallmarks of continental collision such as ophiolitedecorated sutures or high-pressure metamorphic assemblages have not been found. The most convincing correlations exist between the southern end of the Trans-Saharan Belt in West Africa and Pan-African terranes in north-eastern Brazil (Figure 1). Following consolidation of the Gondwana supercontinent at the end of the Precambrian, rifting processes at the northern margin of Gondwana led to the formation of continental fragments (Figure 1) which drifted northwards and are now found as exotic terranes in Europe (Cadomian and Armorican terrane assemblages), in the Appalachian Belt of North
12 AFRICA/North African Phanerozoic
America (Avalonian Terrane assemblage) and in various parts of central and eastern Asia.
See Also Arabia and The Gulf. Australia: Proterozoic. Brazil. Gondwanaland and Gondwana. Palaeomagnetism. Tectonics: Mountain Building and Orogeny. Tertiary To Present: Pleistocene and The Ice Age.
Further Reading Abdelsalam MG and Stern RJ (1997) Sutures and shear zones in the Arabian Nubian Shield. Journal of African Earth Sciences 23: 289 310. Caby R (2003) Terrane assembly and geodynamic evolution of central western Hoggar: a synthesis. Journal of African Earth Sciences 37: 133 159. Cahen L, Snelling NJ, Delhal J, and Vail JR (1984) The Geochronology and Evolution of Africa. Oxford: Clarendon Press. Clifford TN (1968) Radiometric dating and the pre Silurian geology of Africa. In: Hamilton EI and Farquhar RM (eds.) Radiometric Dating for Geologists, pp. 299 416. London: Interscience. Collins AS and Windley BF (2002) The tectonic evolution of central and northern Madagascar and its place in the final assembly of Gondwana. Journal of Geology 110: 325 339. Fitzsimons ICW (2000) A review of tectonic events in the East Antarctic shield and their implications for Gon dwana and earlier supercontinents. Journal of African Earth Sciences 31: 3 23. Hanson RE (2003) Proterozoic geochronology and tectonic evolution of southern Africa. In: Yoshida M, Windley BF, and Dasgupta S (eds.) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Soci ety, London, Special Publications 206, pp. 427 463. Hoffman PF and Schrag DP (2002) The snowball Earth hypothesis: testing the limits of global change. Terra Nova 14: 129 155.
Johnson PR and Woldehaimanot B (2003) Development of the Arabian Nubian Shield: perspectives on accretion and deformation in the northern East African orogen and the assembly of Gondwana. In: Yoshida M, Windley BF, and Dasgupta S (eds.) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications 206, pp. 289 325. Kro¨ ner A (2001) The Mozambique belt of East Africa and Madagascar; significance of zircon and Nd model ages for Rodinia and Gondwana supercontinent formation and dispersal. South African Journal of Geology 104: 151 166. Kusky TM, Abdelsalam M, Stern RJ, and Tucker RD (eds.) (2003) Evolution of the East African and related oro gens, and the assembly of Gondwana. Precambrian Res. 123: 82 85. Meert JG (2003) A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics 362: 1 40. Miller RMcG (ed.) (1983) Evolution of the Damara Oro gen of South West Africa/Namiba. Geological Society of South Africa, Special Publications, 11. Mosley PN (1993) Geological evolution of the late Protero zoic ‘Mozambique Belt’ of Kenya. Tectonophysics 221: 223 250. Porada H and Berhorst V (2000) Towards a new under standing of the Neoproterozoic early Palaeozoic Lufilian and northern Zambezi belts in Zambia and the Demo cratic Republic of Congo. Journal of African Earth Sciences 30: 727 771. Stern RJ (1994) Arc assembly and continental collision in the Neoproterozoic East African Orogen: implications for the consolidation of Gondwanaland. Annual Reviews Earth Planetary Sciences 22: 319 351. Toteu SF, Penaye J, and Djomani YP (2004) Geodynamic evolution of the Pan African belt in central Africa with special reference to Cameroon. Canadian Journal of Earth Science 41: 73 85. Veevers JJ (2003) Pan African is Pan Gondwanaland: ob lique convergence drives rotation during 650 500 Ma assembly. Geology 31: 501 504.
North African Phanerozoic S Lu¨ ning, University of Bremen, Bremen, Germany ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction North Africa forms the northern margin of the African Plate and comprises the countries Morocco, Algeria, Tunisia, Libya, and Egypt (Figure 1). The region discussed here is bounded to the west by the Atlantic, to the north by the Mediterranean Sea, to
the east by the Arabian Plate and to the south by political boundaries. Much of the geology across North Africa is remarkably uniform because many geological events affected the whole region (Figure 2). The geological study of North Africa benefits from large-scale desert exposures and an extensive subsurface database from hydrocarbon exploration. The region contains some 4% of the world’s remaining oil (see Petroleum Geology: Overview) and gas reserves with fields mainly in Algeria, Libya, and
AFRICA/North African Phanerozoic 13
Figure 1 Location of major North African sedimentary basins. Lines indicate locations of cross sections in Figures 2, 3 and 8.
Egypt. Other natural resources that are exploited include Saharan fossil groundwater, phosphate, (see Sedimentary Rocks: Phosphates) and mineral ores.
and (vi) Oligo-Miocene rifting (see Tectonics: Rift Valleys). Infracambrian Extension and Wrenching
Structural Evolution Most of North Africa has formed part of a single plate throughout the Phanerozoic with the exception of the Atlas Mountains which became accreted during Late Carboniferous and Tertiary collisional events. North Africa can be structurally subdivided into a northern Mesozoic to Alpine deformed, mobile belt and the stable Saharan Platform (Figure 3). The latter became consolidated during the Proterozoic PanAfrican Orogeny (see Africa: Pan-African Orogeny), a collisional amalgamation between the West African Craton and numerous island arcs, Andean-type magmatic arcs, and various microplates. The Late Neoproterozoic to Phanerozoic structural development of North Africa can be divided into six major tectonic (see Plate Tectonics) phases: (i) Infracambrian extension and wrenching; (ii) Cambrian to Carboniferous alternating extension and compression; (iii) mainly Late Carboniferous ‘Hercynian’ intraplate uplift; (iv) Late Triassic–Early Jurassic and Early Cretaceous rifting; (v) mid-Cretaceous ‘Austrian’ and Late Cretaceous–Tertiary ‘Alpine’ compression,
The Late Neoproterozoic to Early Cambrian (‘Infracambrian’) in North Africa and Arabia was characterized by major extensional and strike-slip movements. Halfgrabens and pull-apart basins developed, for example, in the Taoudenni Basin (SW Algeria) and in the Kufra Basin (SE Libya). These features are considered to be a westward continuation of an Infracambrian system of salt basins extending across Gondwana from Australia, through Pakistan, Iran and Oman, to North Africa. Post-Infracambrian – Pre-Hercynian
The structural evolution of North Africa between the Infracambrian extensional/wrenching phase and the Late Carboniferous ‘Hercynian Orogeny’ is complex. Local transpressional and transtensional reactivation processes dominated as a result of the interaction of intraplate stress fields with pre-existing fault systems of varying orientation and geometry. In some areas, such as the Murzuq Basin in SW Libya, these tectonic processes played an important role in the formation of hydrocarbon traps.
14 AFRICA/North African Phanerozoic
Figure 2 Phanerozoic chronostratigraphy of petroliferous provinces in North Africa. (From MacGregor (1998).)
During most of the Early Palaeozoic the Saharan Palaeozoic basins were part of a large, interconnected North African shelf system that was in a sagging phase. Some relief, however, was locally already created associated with local uplift and increased subsidence, including, for example, late Cambrian uplift in the Hoggar and increased sagging in the SE Libyan Kufra Basin, the latter leading to thinning of Cambro-Ordovician strata towards the present-day basin margins. The Saharan basins differentiated mainly from the Late Silurian/Early Devonian onwards when ridges were uplifted, associated with a basal unconformity, that in the regional literature has often been referred to as ‘Caledonian unconformity’. This term, however, is inappropriate as tectonic events during the Silurian in North Africa were independent of those in the ‘Caledonian’ collisional zone, located many thousands of kilometres to the north, involving the continents of Laurentia, Baltica, Armorica, and Avalonia. Hercynian Orogeny
Collision of Gondwana and Laurasia during the Late Carboniferous resulted in the compressional
movements of the Hercynian Orogeny (Figure 4). In North Africa, the collisional zone was located in the north-west, leading to substantial thrusting and uplift in Morocco and western Algeria. Strong uplift associated with transpression on old faults occurred in the Algerian Hassi Massaoud region, leading to erosion into stratigraphic levels as deep as the Cambrian. The intensity of Hercynian deformation decreases eastwards across North Africa such that strong folding and erosion of anticlinal crests in the Algerian Sbaa and Ahnet basins is replaced towards the plate interior by low-angle unconformities and disconformities in the Murzuq Basin in south-west Libya. Notably, the present-day maturity levels of the main Palaeozoic hydrocarbon source rocks have a decreasing trend eastwards across North Africa (once presentday burial effects are removed) in parallel with the decrease in the intensity of the Hercynian deformation. The gravitational collapse of the Hercynian Orogenic Belt in north-west Africa was accompanied by widespread Permo-Carboniferous volcanism in Morocco. The magmatism acted here as an ‘exhaust valve’ releasing the heat accumulated beneath the
AFRICA/North African Phanerozoic 15
Figure 3 Cross section through Algeria illustrating typical the structural styles of North Africa. The Alpine deformed Atlas Mountains are separated from the Saharan Platform to the south by the South Atlas Fault. Location of section in Figure 1. (Courtesy E. Zanella.)
16 AFRICA/North African Phanerozoic
Figure 4 Hercynian compression as result of a Late Carboniferous plate collision between Laurasia and Gondwana. (After Doblas et al. (1998).)
Figure 5 Block diagrams illustrating the geological evolution of the High Atlas, including Triassic Jurassic rifting, Cretaceous and Cenozoic inversion. (After Stets and Wurster (1981).)
Pangaean Supercontinent by insulation and blanketing processes which triggered large-scale mantle-wide upward convection and general instability of the supercontinent. Mesozoic Extension
The opening of the Central Atlantic in the Triassic– Early Jurassic and contemporaneous separation of the Turkish–Apulian Terrane from north-east Africa initiated a significant extensional phase in North Africa which included graben formation in the Atlas region (Figure 5), rifting from Syria to Cyrenaica (NE Libya) and extension in offshore Libya and in the Oued Mya and Ghadames (¼Berkine) basins in
central and eastern Algeria. Rift-related Triassic volcanism occured in the northern Ghadames and Oued Maya basins. A second important Mesozoic extensional phase in North Africa occurred during the Early Cretaceous, related to the opening of the South and Equatorial Atlantic Ocean. As a result, a complex of failed rift systems originated across North and Central Africa with the formation of half-grabens in, for example, the Egyptian Abu Gharadig Basin and in the Libyan Sirte Basin. The Mesozoic extensional phase also triggered increased subsidence in several Saharan Palaeozoic basins, leading to deposition of thick, continental
AFRICA/North African Phanerozoic 17
deposits, for example, in the south-east Libyan Kufra Basin. Alpine Orogeny
The onset of rifting in the northern North Atlantic during the Late Cretaceous led to an abrupt change in the motion of the European Plate which began to move eastwards with respect to Africa. The previous sinistral transtensional movements were quickly replaced by a prolonged phase of dextral transpression resulting in the collision of Africa and Europe. The ‘Alpine Orogeny’ led to an overall compressional regime in North Africa from the mid-Cretaceous through to the Recent. Changes in the collisional process, such as subduction of oceanic crust after accretion of a seamount in the Eastern Mediterranean, produced localized stress-neutral or even extensional pulses within the overall compressive regime. An Aptian compressional event may be considered as a precursor to the ‘Alpine Orogeny’, in the narrow sense. It affected parts of North and Central Africa, inverting Early Cretaceous rift systems and reactivating older structures. Large Aptian-age anticlines occur in the Berkine Basin in Algeria and result from sinistral transpression along the N–S trending Transaharian fracture system. The post-Cenomanian ‘Alpine’ compression in North Africa resulted in folding and thrusting within the north-west African collisional zones, as well as in intraplate inversion and uplift of Late Triassic-Early Jurassic grabens. Major orogens formed during this phase include the Atlas Mountains (Morocco, Algeria, Tunisia; Figure 5) and the ‘Syrian Arc’ Fold Belt in north-east Egypt and north-west Arabia. The Cyrenaica Platform (Jebel Akhdar) in north-east Libya also is an ‘Alpine’ deformed region. The structural boundary between the Atlas Mountains and the Saharan Platform is the South Atlas Front (South Atlas Fault), a continuous structure from Agadir (Morocco) to Tunis (Tunisia). The fault separates a zone where the Mesozoic-Cenozoic cover is shortened and mostly detached from its basement from a zone where the cover remains horizontal and attached to its basement. Thrust-belt rocks north of the fault are structurally elevated by about 1.5 km above the Saharan Platform. Apatite fission track data (see Analytical Methods: Fission Track Analysis) suggests that large parts of Libya and Algeria were uplifted by 1–2 km during the ‘Alpine’ deformational phase. As a consequence, Palaeozoic hydrocarbon source rocks were lifted out of the oil window in some parts of the Saharan Palaeozoic basins, resulting in termination of hydrocarbon generation.
Oligo-Miocene Rifting
Another major rifting phase in North Africa during the Oligo-Miocene was associated with the development of the Red Sea, Gulf of Suez, Gulf of Aqaba Rift system, which is the northern continuation of the Gulf of Aden, and East African rifts. Along the north-eastern margin of the Red Sea/Gulf of Suez axis, extension was associated with intrusion of a widespread network of dykes and other small intrusions. Rifting and separation of Arabia from Africa commenced in the southern Red Sea at about 30 Ma (Oligocene) and in the northern Red Sea and Gulf of Suez at about 20 Ma (Early Miocene). Subsequently, tectonic processes in the Arabian–Eurasian collisional zone changed the regional stress field in the northern Red Sea region, causing the rifting activity to switch from the Gulf of Suez to the Gulf of Aqaba. As a consequence the Gulf of Suez became a failed rift and was in part inverted. Intense volcanic activity occurred in central and eastern North Africa during the Late Miocene to Late Quaternary. In places this had already commenced in the Late Eocene. Volcanic features include the plateau basalts in northern Libya, the volcanic field of Jebel Haruj in central Libya, the Tibesti volcanoes in south-east Libya and north-east Chad and the volcanism in the Hoggar (S Algeria, NE Mali, NW Niger). Some authors interpret this continental volcanism as related to a hot spot overlying a deepseated mantle plume while others see the cause in intraplate stresses originating from the Africa–Europe collision that led to melting of rocks at the lithosphere/asthenosphere interface by adiabatic pressure release.
Depositional History Infracambrian
The Infracambrian in North Africa is represented by carbonates, sandstones, siltstones, and shales, often infilling halfgrabens. In Morocco and Algeria, the unit includes stromatolitic carbonates as well as red and black shales, a facies similar to the Huqf Supergroup in Oman that represents an important hydrocarbon source rock there. Infracambrian siliciclastics are also known from several boreholes in the central Algerian Ahnet Basin and southern Cyrenaica (NE Libya). Infracambrian conglomeratic and shaly sandstones and siltstones occur at outcrop underneath Cambrian strata along the eastern margin of the Murzuq Basin and in some boreholes in the basin centre. In the Kufra Basin, the presence of some 1500 m of Infracambrian sedimentary rocks (of unknown lithology) is inferred for the southern basin centre, while
18 AFRICA/North African Phanerozoic
strata of similar age, including dolomites, have been reported from the eastern and western margins of this basin. Notably, salt deposits like those in Oman have not yet been confirmed from North Africa, although some features from seismic studies in the Kufra Basin may represent salt diapirs. Cambro-Ordovician
The Cambro-Ordovician in North Africa is mostly represented by continental and shallow marine siliciclastics, dominated by sandstones with minor siltstone and shale intervals (Figure 6). Deposition occurred on the wide North African shelf in a generally low accommodation setting. The sediment source was the large Gondwanan hinterland to the south, with SE-NW directed palaeocurrents prevailing. The five reservoir horizons of the giant Hassi Messaoud oilfield are located in Upper Cambrian to Arenig quartzitic sandstones, including the Lower Ordovician Hamra Quartzite. A major, shortlived (12 –1 my) glaciation occurred in western Gondwana during the latest Ordovician, with
Figure 6 Cambro Ordovician Skolithos (‘Tigillites’) in Jebel Dalma (Kufra Basin, SE Libya).
the centre of the ice sheet located in central Africa. Features commonly attributed to pro- and sub-glacial processes reported from North Africa, Mauritania, Mali, the Arabian Peninsula, and Turkey include glacial striations, glacial pre-lithification tectonics, diamictites, microconglomeratic shales, and systems of km-scale channels. Several of these features, however, may also occur in deltaic systems unrelated to glaciation, complicating detailed reconstructions of the latest Ordovician glaciation in the region. The uppermost Ordovician in North Africa represents an important hydrocarbon reservoir horizon in Algeria (Unit IV) and Libya (Memouniyat Formation) (Figure 7). Silurian
Melting of the Late Ordovician icecap caused the Early Silurian sea-level to rise by more than 100 m, leading to a major transgression that flooded the North African Shelf to as far south as the northern parts of Mali, Niger, and Chad (Figure 8). Graptolitic, hemipelagic shales represent the dominant facies, while sandstone or non-deposition prevailed in palaeohigh areas, such as most of Egypt, which formed a peninsula at that time. In Libya, the total thickness of the shales (termed ‘Tanezzuft Formation’, Figure 7) increases north-westwards from 50 m in the proximal Kufra Basin, through 500 m in the Murzuq Basin to 700 m in the distal Ghadames Basin, reflecting the north-westward progradation of the overstepping sandy deltaic system (‘Akakus Formation’, Figure 7) during the mid-Llandovery to Ludlow/Prˇ ı´dolı´ (Figure 8). The Silurian shales are generally organically lean, except for the Lower Llandovery (Rhuddanian) and Upper Llandovery/Lower Wenlock when anoxic phases occurred. During these phases, organically rich, black shales (often referred to as ‘hot shale’) with total organic carbon values of up to 16% were deposited. The older of the two black shale horizons is developed only in palaeodepressions that were already flooded in the Early Llandovery, while the upper black shale unit is restricted to areas that during the Late Llandovery/Early Wenlock had not yet been reached by the prograding sandy delta (Figure 8). Silurian organic-rich shales are estimated to be the origin of 80–90% of all Palaeozoic-sourced hydrocarbons in North Africa. The same depositional system is also developed on the Arabian Peninsula, where ageequivalent black shales exist, for example, in Saudi Arabia, Syria, Jordan, and Iraq. Characteristic limestone beds rich in ‘Orthoceras’ are interbedded with the Ludlow-Prˇ ı´dolı´ shales in Morocco and western Algeria, the most distal parts of the North African shelf (Figure 8). In more
AFRICA/North African Phanerozoic 19
Figure 7 Correlation chart of Palaeozoic formations in North Africa.
20 AFRICA/North African Phanerozoic
Figure 8 Depositional model for Late Ordovician to Early Devonian sediments in North Africa. (Modified after Luning et al. (2000).)
AFRICA/North African Phanerozoic 21
proximal shelfal locations, sand influx was already too great for limestones to develop. The ‘Orthoceras Limestone’ in some areas is organic-rich. Similar age-equivalent limestones also occur in some periGondwana terranes, such as in Saxo-Thuringia where the unit is termed ‘Ockerkalk’. Devonian
A major eustatic sea-level fall occurred during the latest Silurian/Early Devonian, resulting in a change to a shallow marine/continental facies in eastern and central North Africa. Coastal sand bar, tidal, and fluvial deposits form important hydrocarbon reservoir horizons, for example, in the Algerian Illizi Basin (unit F6, Figure 7) and the Ghadames (¼Berkine) Basin (‘Tadrart Formation’) in north-west Libya (Figure 8). On the distal side of the North African shelf towards Morocco fully marine conditions still prevailed. The Lower Devonian of Morocco is well-known for its rich trilobite horizons. A sea-level rise during the later part of the Early Devonian led to deposition of shelfal shales and sandstones in central North Africa. In Algeria significant hydrocarbon reservoirs exist in sandstones of the Emsian (units F4, F5). In western Algeria the base of the Emsian lies under a limestone bed termed ‘Muraille de Chine’ (‘Chinese Wall’), because at exposure it commonly forms a characteristic, long ridge. Due to their distal position on the North African shelf and a minimum of siliciclastic dilution Morocco and western Algeria were dominated by carbonate sedimentation during the mid-Devonian. The facies here includes prominent mud mounds, for example, in the southern Moroccan area of Erfoud and in the central Algerian Azel Matti area. Further to the east, the facies becomes more siliciclastic. Eifelian-Givetian tidal bar sandstones form the main reservoir (unit F3) in the Alrar/Al Wafa gas-condensate fields in the eastern Illizi Basin. The beginning of the Late Devonian was characterized by a major eustatic sea-level rise which resulted in deposition of hemipelagic shales, marls, and limestones over wide areas of North Africa. The Moroccan Middle to Upper Devonian typically contains rich cephalopods faunas (goniatites, clymeniids). The ‘Frasnian Event’, an important goniatite extinction event and a phase of anoxia, occurred during the Early Frasnian and led to deposition of organicrich shales and limestones in various places across North Africa. In the Algerian, Tunisian, and Libyan Berkine (¼Ghadames) Basin, Frasnian black shales contain up to 16% organic carbon and represent an important hydrocarbon source rock (Figure 9). The organic-rich unit also occurs in South Morocco and north-west Eygpt. In parts of north-west Africa, a
second organically enriched horizon exists around the Frasnian–Famennian boundary, associated with the worldwide Kellwasser biotic crisis. The deposits in southern Morocco include black limestones. A major fall in sea-level occurred during the latest Devonian, triggering progradation of a Strunian (latest Devonian–earliest Carboniferous) delta in central North Africa. These clastics form an important hydrocarbon reservoir unit (F2) in Algeria. Carboniferous
Sea-level rise during the Early Carboniferous resulted in the development of a widespread shallow marine to deltaic facies across large parts of North Africa. A carbonate platform was established in the Bechar Basin in western Algeria at this time. Early Carboniferous dolomites of the Um Bogma Formation in south-west Sinai host important Mn-Fe ores. Nondeposition and continental sandstone sedimentation occurred in southern and elevated areas, for example, in most of Egypt. In the Late Carboniferous, deposition of marine siliciclastics was restricted to north-west Africa and the northernmost parts of north-east Africa, for example, Cyrenaica and the Gulf of Suez area. Paralic coal in the Westphalian of the Jerada Basin (NE Morocco) forms the only sizable Late Carboniferous coal deposit in North Africa. In the course of the latest Carboniferous Hercynian folding and thrusting, most of north-west Africa was uplifted, resulting in a change to a fully continental environment. Only Tunisa, north-west Libya and the Sinai Peninsula were still under marine influence at this time. Permo–Triassic
Marine Permo-Triassic sedimentary rocks are restricted to the northernmost margin of central and eastern North Africa. For example, Permian marine carbonates and siliciclastics crop out in southern Tunisia representing the only exposed Palaeozoic unit in this country. Most of North Africa, however, remained subaerially exposed during the Permian to mid-Triassic. Continental red clastics (sandstones, shales, conglomerates) represent the most important lithologies. The Permian of Morocco is restricted to a series of intramontane basins located around the margin of the central Moroccan Hercynian massif. The main facies associations in the Triassic TAGI (Trias Argilo-Gre´ seux Infe´ rieur) in the eastern Algerian Berkine (¼Ghadames) Basin are fluvial channel sandstones, floodplain silts and palaeosols, crevasse splay deposits, lacustrine sediments, and shallow marine transgressive deposits. Fluvial sandstones of the TAGI are the main oil and gas reservoirs in the
22 AFRICA/North African Phanerozoic
Figure 9 Known distribution of organic rich strata of Early Silurian, Late Devonian, Cenomanian Turonian, and Campanian Maastrichtian age in North Africa.
AFRICA/North African Phanerozoic 23
Algerian Berkine and Oued Mya basins, including the super-giant gas field in Hassi’R Mel. Similar Triassic sandstones also serve as a relatively minor hydrocarbon reservoir in the Sirt Basin, sourced from Cretaceous source rocks. During the Late Triassic/Early Jurassic, evaporites were deposited in rift grabens associated with the opening of the Atlantic, and of the Atlas Gulf and with the separation of the Turkish-Apulian terrane from North Africa. Characteristic ‘salt provinces’ are located offshore along the Moroccan Atlantic coast, northern Algeria/Tunisia and offshore eastTunisia/north-west Libya. In most areas the diapiric rise commenced in the Jurassic–Cretaceous. The Late Triassic/Early Jurassic evaporites and shales in the north-east part of the Algerian Saharan Platform are up to 2 km thick and form a hydrocarbon caprock for the Triassic reservoir. In some cases, because of the Hercynian unconformity, they also form the caprock for Palaeozoic reservoirs such as at the super-giant Hassi Messaoud field in Algeria. Jurassic
Marine sedimentation during the Jurassic was restricted to the northern and western rims of North Africa, including, for example, northernmost Egypt, the Atlas region, and the Tarfaya Platform in southern Morocco. Carbonate platforms and intraplatform basins were widespread, including development of reefal limestones and oolites. In the Gebel Maghara area in northern Sinai, paralic coal was deposited during the Middle Jurassic. Locally the Lower and Upper Jurassic of North Africa contain organically enriched horizons, corresponding in age to the prominent Jurassic black shales of central Europe (e.g., Posidonia Shale in Germany and Kimmeridge Clay in England). Such Jurassic bituminous pelites occur, for example, in the Atlantic Basin, Atlas Rift of Morocco, and the Egyptian Abu Gharadig Basin. South of the North African Jurassic marine facies belt, continental redbeds were deposited (Figure 10). In the Egyptian Western Desert the Jurassic–Cretaceous contains several prolific hydrocarbon reservoir horizons. Cretaceous
Due to low eustatic sea level the Lower Cretaceous of North Africa is dominated by terrestrial clastics, termed the ‘Nubian Sandstone’ in Egypt and Libya (‘Sarir Sandstone’ in the Sirt Basin) (Figure 10). Once again, marine conditions existed only in a marine coastal belt in the north. During the Aptian to Maastrichtian, a series of transgressions gradually flooded the areas to the south. On the Sinai Peninsula, the transition phase is characterised by deltaic influenced,
Figure 10 Cross bedded fluvial ‘Nubian Sandstone’, Jurassic Cretaceous, ‘Coloured Canyon’, central East Sinai (Egypt).
mixed siliciclastic-carbonate systems that during the Albian evolved into carbonate-dominated environments. During the latest Cenomanian, large parts of North Africa became submerged following a prominent eustatic sea-level rise that is thought to be one of the most intense Phanerozoic flooding event. As a consequence, the ‘Transsaharan Seaway’ was created, connecting the Tethys in central North Africa with the Atlantic in West Africa. Similar seaways and gulfs existed in north-west Africa into the Eocene. A seaway located within the Atlas rift system, the ‘Atlas Gulf’, was restricted temporally to the Cenomanian–Turonian. The strong latest Cenomanian sea-level rise in combination with high productivity conditions in the southern North Atlantic are thought to form the basis for the Late Cenomanian–Early Turonian Oceanic Anoxic Event (OAE2) during which organic-rich strata were deposited in rift shelf basins and slopes across North Africa and in deep sea basins of the adjacent oceans. Characteristic sediments associated with this anoxia include oil shales in the Tarfaya Basin (southern Morocco), organic-rich limestones in north-west Algeria and northern Tunisia (Bahloul Formation), and black shales in offshore Cyrenaica, and the Egyptian Abu Gharadig Basin (Abu Roash Formation) (Figure 9). The unit represents a potential oil-prone hydrocarbon source rock in the region. A general decrease in peak organic richness and black shale thickness occurs in North Africa from west to east, which possibly is a result of upwelling along the Moroccan Atlantic coast and the absence of upwelling in the Eastern Mediterranean area. The organic-rich Cenomanian-Turonian deposits also play an important role in the genesis of Zn/Pb ore deposits in northern Tunisia and eastern Algeria. The origin of these Zn/Pb ores is related to hypersaline basinal brines, made of ground water and dissolved Triassic evaporites, that leached metals
24 AFRICA/North African Phanerozoic
from the Triassic-Cretaceous sediments. Ore deposition occurred when these metal-bearing solutions mixed with microbially reduced sulphate solutions that were associated with the organic carbon of the Cenomanian-Turonian strata. Due to the generally high sea-level, the marine Upper Cretaceous in North Africa is dominated by calcareous lithologies, namely dolomites/limestones, chalks, and marls (Figure 11). Lateral and vertical facies distributions are strongly related to sea-level changes of various orders as well as to the changing structural relief associated with Late Cretaceous syndepositional compression. Great variations in thickness and facies as well as onlap features, for example, are developed around the domal anticlines of the Syrian Arc Foldbelt in Sinai and within rift grabens of the Sirt Basin (N. Libya). The Campanian–Maastrichtian was characterised by very high sea-level, resulting in a widespread distribution of hemipelagic deposits, such as chalks and marls. These deposits often contain abundant foraminiferal faunas and calcareous nannofossil floras, which allow high-resolution biostratigraphic and palaeoecological studies in these horizons. As on the Arabian Peninsula, the Santonian–Maastrichtian interval in North Africa contains significant amounts of phosphorites, which are mined in, for example, Morocco/Western Sahara and Abu Tartour (Western Desert), making North Africa one of the world’s largest producers of phosphate (see Sedimentary Rocks: Phosphates). In places, the Campanian–Maastrichtian contains organic-rich intervals with total organic carbon contents of up to 16%, for example, in the Moroccan Tarfaya Basin and Atlas Gulf area, the Libyan Sirt Basin and the Egyptian southern Western Desert, Red Sea Coast and Gulf of Suez (Figure 9). Notably,
Figure 11 Contact between chalky limestones of the Early Eocene Bou Dabbous Formation (reddish) and the underlying Campanian Maastrichtian Abiod Formation (bluish) (Ain Rahma Quarry, Gulf of Hammamet area, Tunisia).
Algeria, Tunisia, and West Libya are dominated by organically lean deposition during this time. Campanian–Maastrichtian black shales form important hydrocarbon source rocks in the Sirt Basin and the Gulf of Suez. Palaeogene
Sea-level during most of the Paleocene–Eocene remained high resulting in deposition over wide areas (Egypt: Dakhla and Esna Shale) of hemipelagic marls and chalks that are rich in planktonic foraminifera. A sea-level fall occurred during the mid-Paleocene, resulting in the formation of a short-lived carbonate interbed (‘Tarawan Chalk’) in parts of Egypt. Within the Eocene, the facies typically changes here to hard dolomitic limestones with abundant chert nodules (‘Thebes Limestone’). A similar Palaeogene facies development can also be found in parts of northern Libya and Tunisia. The Eocene in Egypt, Libya, Tunisia, and Algeria includes nummulitic limestones up to several 100 metres thick, which were deposited in carbonate ramp settings. The unit forms major hydrocarbon reservoirs in offshore Libya and Tunisia. Well-exposed and continuous exposures occur in Jabal al Akhdar (Cyrenaica), where the nummulite body’s geometry can best be studied (Figure 12). Notably, the Giza pyramids in Cairo are built from Eocene nummulite limestone. The Eocene hydrocarbon play in the offshore of Tunisia is sourced by dark-brown marl and mudstone of the lower Eocene Bou Dabbous Formation. The unit contains type I and II kerogen and ranges in thickness from 50 to 300 m. Neogene and Quaternary
Marine conditions during the Miocene were again restricted to the northernmost margin of North Africa
Figure 12 High energy nummulitic bank facies, Darnah Forma tion, Middle to Late Eocene, West Darnah Roadcut, Jebel Akhdar (Cyrenaica, Libya).
AFRICA/North African Phanerozoic 25
including the Atlas, Sirte Basin, Cyrenaica, and Red Sea. Carbonate platforms and ramps were developed in northern Morocco. The Miocene Gulf of Suez in Egypt is rich in hydrocarbons, containing more than 80 oilfields. Oils in the Gulf of Suez were mostly sourced from source rocks in the pre-rift succession, including the Campanian–Maastrichtian Brown Limestone. Hydrocarbon reservoir horizons include various Miocene syn-rift sandstones and carbonates as well as pre-rift reservoirs, including fractured Precambrian granites, Palaeozoic–Cretaceous sandstones, and fractured Eocene Thebes Limestone. The thickness distribution and facies of the syn-rift strata are strongly controlled by fault block tectonics. Shales and dense limestones of the pre-rift and the syn-rift units are the primary seals, while overlying Miocene evaporites form the ultimate hydrocarbon seals. During the latest Miocene, more than 2 km thick evaporites were deposited in a deep and desiccated Mediterranean basin that had been repeatedly isolated from the Atlantic Ocean. In the near-offshore only a few tens to hundreds of metres of evaporites exist, whilst they are almost absent from the onshore area. As a consequence of the ‘Messinian Salinity Crisis’, a large fall in Mediterranean sea-level occurred, followed by erosion and deposition of nonmarine sediments in a large ‘Lago Mare’ (‘lake Sea’) basin. Cyclic evaporite deposition is thought to be almost entirely related to circum-Mediterranean climate changes. The Nile Delta system represents a major natural gas province. It was initiated during the Late Miocene with deep canyon incision into pre-existing Cenozoic/ Mesozoic substrate, allowing transportation of huge amounts of sediments into the Mediterranean. The proximal infill of these canyons is thick, coarse alluvium becoming sandier with greater marine influence northwards. The far reaches of these canyon systems have proven to be a good Plio-Pleistocene hydrocarbon reservoir linked mainly to the lowstands, when sands were conveyed to the outer belts through incised canyons in the upper slopes which led to submarine fans farther northwards. The Early Holocene (9–7 kyr BP) was a relatively humid period in North Africa. During this phase, the African Humid Period, grasslands covered the Sahara/Sahel region, and many lakes and wetlands existed here. The humid conditions at this time were associated with a strengthening of the summer monsoon circulation due to an increase in the land–sea thermal contrast under the influence of relatively high summer insolation.
See Also Africa: Pan-African Orogeny. Analytical Methods: Fission Track Analysis. Petroleum Geology: Overview. Plate Tectonics. Sedimentary Rocks: Phosphates. Tectonics: Rift Valleys.
Further Reading Ben Ferjani A, Burollet PF, and Mejri F (1990) Petroleum Geology of Tunisia. Tunis: Entreprise Tunisienne d’Acti vite´ s Pe´ trolie`res. Beuf S, Biju Duval B, de Charpal O, Rognon P, Gariel O, and Bennacef F (1971) Les gre`s du Pale´ ozoı¨que infe´ rieur au Sahara, Se´ dimentation et discontinuite´ s, e´ volution d’un craton. Publications de l’Institut franc¸ais du Pe´trole 18: 464. Coward MP and Ries AC (2003) Tectonic development of North African basins. In: Arthur TJ, MacGregor DS, and Cameron NR (eds.) Petroleum Geology of Africa: New Themes and Developing Technologies. Geological Soci ety London, Special Publication 207: 61 83. Dercourt JM, Gaetani B Vrielynck E, et al. (eds.) (2000) Atlas Peri Tethys, Palaeogeographical maps. CCGM/ CGMW, Paris. Doblas M, Oyarzun R, Lopez Ruiz J, Cebria JM, Youbi N, Mahecha V, Lago M, Pocovi A, and Cabanis B (1998) Permo Carboniferous volcanism in Europe and north west Africa: a superplume exhaust valve in the centre of Pangaea? J. Afr. Earth Sciences 26: 89 99. Hallett D (2002) Petroleum Geology of Libya. Amsterdam: Elsevier. Lu¨ ning S, Craig J, Loydell DK, Sˇ torch P, and Fitches B (2000) Lower Silurian ‘Hot Shales’ in North Africa and Arabia: Regional Distribution and Depositional Model. Earth Science Reviews 49: 121 200. Macgregor DS, Moody RTJ, and Clark Lowes DD (eds.) (1998) Petroleum Geology of North Africa. Geological Society London Special Publication 132: 7 68. Maurin J C and Guiraud R (1993) Basement control in the development of the Early Cretaceous West and Central African Rift System. Tectonophysics 228: 81 95. Pique´ A (2002) Geology of Northwest Africa. Stuttgart: Gebr. Borntraeger. Said R (1990) The Geology of Egypt. Rotterdam, Netherlands: Balkema Publishers. Schandelmeier H and Reynolds PO (eds.) (1997) Palaeogeographic Palaeotectonic Atlas of North eastern Africa and adjacent areas. Rotterdam: Balkema. Selley RC (1997) Sedimentary basins of the World: Africa. Amsterdam: Elsevier. Stampfli GM, Borel G, Cavazza W, Mosar J, and Ziegler PA (2001) The Paleotectonic Atlas of the Peritethyan Domain. Strasburg European Geophysical Society. Stets J and Wurster P (1981) Zur Strukturgeschichte des Hohen Atlas in Marokko. Geologische Rundschau vol. 70(3): 801 841. Tawadros EE (2001) Geology of Egypt and Libya. Rotterdam: Balkema.
26 AFRICA/Rift Valley
Rift Valley L Frostick, University of Hull, Hull, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The East African and Dead Sea rifts are famous examples of rifts that have played prominent parts in human evolution and history. They are both areas where the Earth’s crust has been put under tension and ripped apart to give deep valleys that snake across the landscape. They are linked tectonically, via the Red Sea–Gulf of Aden, which is an incipient ocean separating the African and Arabian plates. The differences between the two rifts are caused by differences in the relative movement of the crust. In the East African rift the tension that formed the rift is close to 90 to the rift axis, whereas the movement of Jordan relative to Israel is northwards, almost parallel to the Dead Sea, which is a small section pulled apart as a result of splaying and bending of the faulted plate boundary. The ancient crust of Africa has been subjected to rifting many times in its very long geological history. Recognizable rift basins can be identified in many locations around the continent, and they range in age from Palaeozoic to Quaternary, a time-span of over 500 Ma. In some areas there is evidence of repeated activity, and it appears that there have been at least seven phases of rifting over the past 300 Ma. The older rifts, for example the Benue trough in West Africa, have been inactive for many millions of years, but the most spectacular rift features are to be found in East Africa, where recent rifting has left a scar on the landscape that is visible from space (Figure 1). North of the zone where Africa touches Europe at the eastern end of the Mediterranean there is another famous rift, which is linked tectonically to East Africa. The Dead Sea Rift straddles the border between Israel and Jordan and is the lowest point on the surface of the Earth, reaching more than 800 m below sea-level.
in a plate-tectonic setting that is dominated by extension, particularly during the Tertiary–Quaternary period. During this time the Great or East African Rift was formed as part of a larger plate-tectonic feature that stretches from south of Lake Malawi in Africa to the flanks of the Zagros mountains and the Persian Gulf in the north (Figure 2). It changes its nature along its length, resulting in a range of geological basins and geomorphological features. In Africa it is a volcanically active continental rift hundreds of kilometres wide that contains a range of river and lake sediments. As it quits Africa it passes into an incipient ocean with a newly formed seafloor spreading centre along the length of the Red Sea and the Gulf of Aden. The deposits in these basins include thick sequences of salt, which form effective traps for hydrocarbons generated from associated organic-rich shales. North of the Red Sea the type of plate margin alters as the boundary passes through the Gulf of Aqaba/Elat and
Plate Tectonic Setting Rifting occurs when the crust of the Earth is placed under tension, pulling it apart and causing faulting. The general term for the basins so produced is ‘extensional’ but they can occur in situations where the regional sense of movement is compressional or is tearing the crust, e.g. the Baikal and Dead Sea rifts, respectively. However, the main African rift basins were formed
Figure 1 Satellite remote sensing image of the Horn of Africa and Arabia, showing the East African Rift system, the incipient ocean of the Red Sea Gulf of Aden, and the conservative plate boundary that runs through the Dead Sea. Images collected by the TERRA satellite using the MODIS instrument (moderate resolution imaging spectroradiometer) and enhanced with SRTM30 (Shuttle Radar Topography Mission 1km resolution) shaded relief.
AFRICA/Rift Valley 27
Figure 2 Diagrammatic representation of the plate tectonic setting of the area between the northern end of the East African Rift (Afar triangle) and the Zagros Mountains.
into the Levant/Areva valley between Israel and Jordan. Here, the Arabian plate is moving past the European plate without significant extension or compression. However, localized tension associated with fault bends and splays has resulted in the formation of two very well-known biblical lake basins, the Dead Sea and Lake Kinneret (otherwise known as the Sea of Galilee), both of which can be seen in Figure 3. These are also termed ‘rifts’ although the setting and geological history are different from those of their larger East African contemporary. The system terminates in the Zagros mountains, where the crust created in the new Red Sea–Gulf of Aden ocean is compensated for by the crustal shortening inherent in mountainbuilding processes.
The East African Rift Topography and Structure
Within Africa certain features of the topography and structure are common to all the basins.
Figure 3 Satellite remote sensing image of the Sinai Arabian plate boundary, showing the Dead Sea and Sea of Galilee (Lake Kinneret). Image collected by the TERRA satellite using the MODIS instrument (moderate resolution imaging spectroradiometer).
Figure 4 Stylized half graben structure typical of the basins in the East African Rift.
Topographically they comprise a central valley, often referred to as a ‘graben’, flanked by uplifted shoulders that are stepped down towards the rift axis by more or less parallel faults. Often, one flank is more faulted than the other, so that the rift valley is in fact asymmetrical and should be referred to as a ‘half graben’ (Figure 4). The width of the structure varies from 30 km to over 200 km, with the widest section at the northern extremity where the rift links to the Red Sea in the Afar region of Ethiopia. The main faulted margin alternates from one side of the rift to the other along its length, producing a series of
28 AFRICA/Rift Valley
relatively separated basins, many of which contain lakes of varying depth and character (e.g. Lakes Tanganyika, Naivasha, and Malawi). These are separated into hydrologically distinct basins by topographical barriers crossing the rift axis where the border faults switch polarity. This surface separation reflects an underlying structure, the nature of which varies from basin to basin but often includes faulting with a tearing or scissor type of movement and flexing. Geologists are not agreed on the processes going on in these areas and have given these zones different names according to their assumptions about the mechanism of formation. These include transfer, relay, and accommodation zones, as well as ramps or just segment boundaries. The distance between adjacent boundaries varies from tens to hundreds of kilometres (Figure 5). At each end of the individual border faults the displacement of the rift floor relative to rocks outside the valley reduces to zero. Displacement is greatest at the centre of the fault, and this leads to a subtle rise and fall of the rift floor along its length even without the intervention of major new cross-rift structures and processes. The rift in Kenya is characterized by numerous caldera volcanoes and at least 3 to 4 phases of faulting, the most recent forming a narrow linear
axial zone. the faulting ranges in age from Miocene to Recent. In the southern half of the rift’s 35 000 km length it divides into two distinct branches around Lake Victoria. The eastern branch contains only small, largely saline, lakes, while the western branch contains some of the largest and deepest lakes in the region, including Lake Tanganyika. Doming and Volcanicity
The East African Rift contains two large domes centred on Robit in Ethiopia and Nakuru in Kenya. These domes are over 1000 km in diameter and extend far beyond the structural margins of the rift valley. Geophysical studies of these domes have shown that they are underlain by zones of hot low-density mantle rocks and that the surface crust is thinned significantly relative to adjacent areas. The domes are centres of volcanic activity that began more than 25 Ma ago and continues to the present day (Figure 6). Volcanic features are widespread in Ethiopia and extend southwards into Kenya along the eastern branch of the rift. It is estimated that there are more than 500 000 km3 of volcanic rocks in this area, over a third of which occur in Kenya. In the branch to the west of Lake Victoria volcanism is spatially more
Figure 5 Diagrammatic representation of the river drainage close to the west shore of Lake Turkana, northern Kenya, showing the Kerio River flowing into the Lake at a transfer zone and the alluvial fans issuing from the fault scarps.
AFRICA/Rift Valley 29
Figure 6 Geyser activity in the volcanically active area around Lake Bogoria, Kenya.
limited, occurring only to the north and south of Lake Tanganyika. This contributes to the different characters of the lakes in the two branches as not only can the volcanic rocks fill the basins, leaving less space for large lakes, but also many of the rock types are rich in salts, which contribute to the salinity of the lakes once they are released by weathering. Large and active volcanoes that sit outside the rift structure are a striking feature of the landscape. Mounts Kilimanjaro and Kenya, for example, are favourite targets for climbers, and both sit on the flanks of the rift (Figure 7). Hydrology and Climate
The East African Rift system sits astride the equator, extending from 12 N to 15 S, and this dictates the overall character of the climate. Superimposed on this are the effects of the rift topography, with its uplifted domes, faulted flanks, and depressed central valleys. Rainfall is lowest in the northern parts of Ethiopia and increases southwards into northern Kenya. The region is generally desert or semi-desert with vegetation limited to sparse grasses and scrub. South of where the rift branches the rainfall is higher, with the western branch being wetter than the eastern one. The uplifted mountains that make up the margins of the rift are wetter and cooler than the valley bottom; for example, an annual figure of over 2000 mm of rainfall has been recorded in the Ruwenzori Mountains near Lake Mobutu. The doming that accompanied the rifting in East Africa has had a major impact on the present river systems. The development of the rift disrupted a
pre-existing continental drainage system in which a few large rivers with vast integrated drainage basins dominated the landscape. As the area was domed and faulted and the new valley formed, the rivers adjusted to the new landscape: some lost their headwaters, others were created, some gained new areas to drain. The overall effect was to divert much of the drainage north into the Nile system and west into the Congo drainage, with only a few small rivers now reaching the Indian Ocean. Inside the valley, the rivers are generally short and small, ending in a lake not far from the river source, but a few rivers run along the rift, often caught between faulted hills, and discharge into lakes far from their original sources, e.g. the Kerio River in Kenya has its source near Lake Baringo but discharges into Lake Turkana more than 200 km to the north (Figure 8). The segregation of the underlying structure into topographically distinct sections exerts an overriding control on the character and distribution of lakes throughout the rift. It provides the framework within which the balance between movement of water into the basin, from rainfall and rivers, and evaporation from the surface will work. The largest and deepest lake, Lake Tanganyika, is in the wetter western branch of the rift in a particularly deep section. It covers an area of over 40 000 km2 and is more than 1400 m deep at its deepest point. Lakes in the eastern branch are smaller and shallower; for example Lake Bogoria is an average of less than 10 m deep, and if the climate changes and rainfall decreases they soon become ephemeral, drying out completely during periods of drought.
30 AFRICA/Rift Valley
Figure 7 Satellite image of Mount Kilimanjaro and Mount Kenya, showing how they sit outside the main East African Rift structure. This is a shaded relief map produced from SRTM30 data with colour added to indicate land elevations.
Sedimentation and Basin Fills
As water flows into the rift basins it brings with it material dislodged and dissolved from the surrounding rocks, which is then deposited within the basin. How, where, and what is deposited depends on the shape of the basin and how surface processes work on and disperse the material. The overall shape of the basin fill is controlled by the pattern of faults and subsidence: deposits are thicker close to areas of the faults with greatest displacement (Figure 4). The geometry of the fill is therefore almost always asymmetric, thickening towards the main border fault and thinning in all other directions, giving a characteristic wedge shape. There are no marine sediments in the rift: all the deposits are terrestrial and comprise river, delta, lakecoast, and lake sediments. Wind-blown sands and dunes are rare and of only local importance. The rivers vary in character from ephemeral, flowing
only in response to seasonal rain storms, to perennial. The rivers carry and deposit sands and gravels in their beds, sweeping finer silts and clays into overbank lagoons and lake-shore deltas. The character of the lake deposits themselves depends on a variety of factors including the timing and character of river supplies, salinity, evaporation, water stratification, and animal and plant growth. In deep lakes such as Lake Tanganyika there are layered muds, which can be hundreds of metres thick and contain enough algal remains to generate oil. Shallower lakes can contain high numbers of diatoms, which leave deposits of a silica-rich rock called diatomite. Some lakes in volcanic areas of the rift have sufficiently high salt concentrations for precipitation and the development of exploitable salt deposits. One example is the trona, a complex carbonate of sodium, which is extracted seasonally from Lakes Magadi and Natron (Figure 9).
AFRICA/Rift Valley 31
Hominid Finds and Evolution
The rift forms a striking geomorphological feature cutting across the African craton, segmenting the landscape, and controlling the local geology. Along most of its length it achieves a depth of in excess of 1 km and at its deepest, in Ethiopia, it is over 3 km deep. Its striking topography generates its own set of microclimatic and hydrological conditions, which have had a major impact on plant and animal distributions and evolution. It acts as a north–south corridor for the migration of animals and birds, but equally inhibits east–west movements. During periods of climatic stress at higher latitudes, when glaciers dominated much of the European and Asian continents, the lake basins of the rift were havens for animals, including early humans. Finds of early humans (hominids) in the rift are more numerous and more complete than in almost any other part of the world, and it has been postulated that all present-day humans are derived from ancestors that migrated out of the East African Rift (see Fossil Vertebrates: Hominids).
Dead Sea Rift Topography and Structure
Figure 8 Stylized diagram of the Lake Turkana area at 3 N in the East African Rift, showing the main faults and transfer zones crossing the rift axis.
The Dead Sea Rift is superficially very similar to some of the individual lake basins in the East African Rift. It is a narrow depression in the surface of the Earth over 100 km long and only 25 km wide, reaching over 800 m deep at its lowest point (Figure 10). The Dead Sea is not, in reality, a sea at all but an enclosed salty lake, which occupies more than 80% of the surface area of the basin. It sits on the plate boundary that
Figure 9 Lake Magadi, Kenya, during the dry season, showing the surface of the lake completely encrusted with salt.
32 AFRICA/Rift Valley
Figure 10 Patterns of faulting and their influence on the development of river systems around the Dead Sea pull apart basin.
spans the 1100 km between the Gulf of Elat/Aqaba and Turkey and separates the Arabian Plate to the east from the African Plate to the west (Figure 2). Since the Miocene, a period of about 20 Ma, Arabia is thought to have moved more than 105 km northwards, a type of movement that is termed strike-slip. The Dead Sea Basin is a zone where the movement has resulted in local tension, producing faulting and leading to the sinking of a section of the crust. Such basins are termed ‘pull-apart’ basins and are characterized by very rapid subsidence and thick basin-fill sequences. The overall structure of the Dead Sea Rift is asymmetrical, not dissimilar to that of the East African Rift. The largest fault is in the eastern margin and forms the Jordanian shore of the lake. Here, the faulting exposes a spectacular rock sequence more than 1 km thick, which ranges in age from Precambrian (more than 544 Ma) to Pleistocene (less than 1 Ma). On the opposite side of the basin are a number of smaller subparallel faults, which cut the Cretaceous limestones of this margin into a series of structural steps (Figure 10). At either end of the basin are crossrift structures that link movement along the Areva fault to the south with movement on the Jordan fault to the north. A major feature of the southern part of the basin is the development of salt diapirs. These result from subsurface movements of thick deposits of rock salt,
which can push up and punch through the overlying sediments and penetrate to the surface. One famous example of such a diapir is Mount Sedom, famous for its biblical links with the doomed and ‘sinful’ cities of Sodom and Gomorrah. Rivers and Hydrology
The development of the Dead Sea Rift system disrupted a pre-existing drainage system that crossed from east to west across the Jordan plateau and drained into the Mediterranean. The headwaters of this system now run across the eastern scarp of the Dead Sea and have cut gorges over a kilometre deep to reach the lake shore (Figure 10). On the western shore a new set of rivers have evolved, which no longer drain into the Mediterranean Sea to the west but instead have been reversed and now drain from west to east. These rivers have also cut down into prerift rocks and run in gorges that are less deep than those of their eastern equivalents. The present climate of the area is desert to semidesert with rainfall of 50–200 mm year 1. Because of this all rivers except the Jordan, which has headwaters in an area of higher rainfall to the north, are ephemeral and flow only in response to winter rain storms. The lake water is renowned for its high salinity, which is 10 times that of normal seawater. Tourists
AFRICA/Rift Valley 33
are attracted to ‘swim’ in the waters, which are so buoyant that individuals can sit unsupported and read a paper. The high salinity is a result of a combination of evaporation in a closed basin and the influence of brines coming from the solution of subsurface rock salt. The lake brines are particularly rich in chlorine and bromine, which are extracted in salt ponds and exported worldwide. Climate Change and the Basin Fill
Lake levels in closed basins are very susceptible to the effects of climate change. Any increase in rainfall will upset the hydrological balance and cause lake levels to rise and salinity to fall. If rainfall decreases, lake levels will drop and evaporation will dominate, resulting in an increase in salinity. The surface of the Dead Sea shows evidence of having fluctuated between 180 m and 700 m below sea-level over the past 60 Ka in response to well-documented changes in climate. The rising and falling lake levels have a profound effect on the sedimentary deposits of the rift. High lake levels, such as those that prevailed during the deposition of the Pleistocene Lisan Formation, result in thick sequences of interlaminated chalk and silty clay (Figure 11). During periods of lower lake levels the river and fan deposits penetrate far into the basin and dominate the sequences. One surprising consequence of depressed lake levels is a change in the balance between saline and fresh groundwaters, with the latter penetrating further towards the axis of the basin. Since much of the basin axis is underlain by thick salt deposits, the fresh groundwater dissolves
the preserved layers of salt, generating subsurface caverns and solution holes. This is currently happening in response to lake levels falling as a result of over abstraction of water from the Jordan River. Earthquakes, Archaeology, and Sodom and Gomorrah
Earthquakes have been a feature of the Dead Sea Rift throughout its history. The earthquakes are generated by movement along the main fault zone and are often accompanied by the release of asphalt, gases, and tars, which are trapped in the layers of rock beneath the surface. The asphalt in particular is well documented and is found in layers within the older lake deposits. Fault movements tend to happen sporadically: long periods of quiescence are succeeded by times when earthquakes are regular events. The Dead Sea Basin has been inhabited by local peoples for many thousands of years. The alluvial plains of the valley were rendered fertile by irrigation, and trading routes to the south, east, and west allowed early settlers to exploit the mineral wealth of the area, including gathering and trading materials from oil seeps and asphalt, which have been found as far away as Egypt in the tombs of the Pharaohs. The early Bronze Age was a time when the basin was well populated and was also a tectonically quiet period when few earthquakes occurred. Towards the end of this period there was a large earthquake, which may have resulted in the destruction of two major cities, Sodom and Gomorrah. There has been speculation about precisely how and why these cities were so
Figure 11 A section through the Lisan Formation of the Dead Sea, showing layers of chalk and silt (horizontal layers at the top and bottom of the section), some of which have been disturbed by earthquake activity (folded layers in the centre of the section).
34 AGGREGATES
comprehensively demolished that they were never rebuilt. One theory is that they were built on soft sediments that became liquid (liquefaction) as they were shaken, maximizing the instability of the ground (see Engineering Geology: Liquefaction). Interestingly, the occurrence of ‘sulphurous’ fires reported in the bible corresponds well with the release of the light fractions of oil from underground reservoirs as the ground moves and slides in response to shaking. It seems likely that the myths surrounding the destruction of Sodom and Gomorrah are based in fact and are a direct consequence of the unique geology of the area.
See Also Biblical Geology. Engineering Geology: Liquefaction. Fossil Vertebrates: Hominids. Geomorphology. Sedimentary Environments: Lake Processes and Deposits. Tectonics: Earthquakes; Faults; Mid-Ocean Ridges; Rift Valleys.
Further Reading Allen PA and Allen JR (1990) Basin Analysis: Principles and Applications. Oxford: Blackwells.
Enzel Y, Kadan G, and Eyal Y (2000) Holocene earthquakes inferred from a fan delta sequence in the Dead Sea graben. Quaternary Research 53: 34 48. Frostick LE and Reid I (1989) Is structure the main control on river drainage and sedimentation in rifts? Journal of African Earth Sciences 8: 165 182. Frostick LE and Steel RJ (eds.) (1993) Tectonic Controls and Signatures in Sedimentary Successions. International Association of Sedimentologists Special Publication 20. Oxford: Blackwells. Frostick LE, Renaut RW, Reid I, and Tiercelin JJ (1986) Sedimentation in the African Rifts. Special Publication 25. London: Geological Society. Girdler RW (1991) The Afro Arabian Rift System: an over view. Tectonophysics 197: 139 153. Gupta S and Cowie P (2000) Processes and controls on the stratigraphic development of extensional basins. Basin Research 12: 185 194. Neev D and Emery KO (1995) The Destruction of Sodom, Gomorrah and Jericho. Oxford: Oxford Univer sity Press. Selley RC (ed.) (1997) African Basins. Sedimentary Basins of the World 3. Amsterdam: Elsevier. Summerfield MA (1991) Global Geomorphology: An Introduction to the Study of Landforms. Harlow: Longman.
AGGREGATES M A Eden and W J French, Geomaterials Research Services Ltd, Basildon, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Aggregates are composed of particles of robust rock derived from natural sands and gravels or from the crushing of quarried rock. The strength and the elastic modulus of the rock should ideally match the anticipated properties of the final product. Aggregates are used in concrete, mortar, road materials with a bituminous binder, and unbound construction (including railway-track ballast). They are also used as fill and as drainage filter media. In England alone some 250 million tonnes of aggregate are consumed each year, representing the extraction of about 0.1 km3 of rock, if necessary wastage is taken into account. Aggregates may be derived from rocks extracted from quarries and pits, or from less robust materials. For example, slate and clay can be turned, by heating, into useful expanded aggregates of low bulk density. The principal sources of aggregate are sand and gravel pits, marine deposits extracted by dredging,
and crushed rock from hard-rock quarries. As extracted, these materials would rarely make satisfactory aggregate. They need to be carefully prepared and cleaned to make them suitable for their intended purpose. The sources may also be rather variable in their composition and in the rock types present, so it is essential that potential sources are carefully evaluated. At the very least, the preparation of the aggregate involves washing to remove dust and riffling to separate specific size ranges. The classification of aggregates varies greatly. An early classification involved the recognition of Trade Groups, which were aggregates consisting of rocks thought to have like properties and which could be used for a particular purpose. A fairly wide range of rock types was therefore included in a given Group. More recent classifications have been based on petrography. Again, these groups tend to be broad, and they focus on the macroscopic properties of the materials for use as aggregate rather than on detailed petrographic variation. Because aggregates consist of particulate materials, whether crushed or obtained from naturally occurring sands and gravels, their properties are normally measured on the bulk prepared material. There are
AGGREGATES 35
therefore numerous standard tests that relate to the intended use of the material. Standard tests vary from country to country, and, in particular, collections of standard tests and expected test results are given in specific British and American Standards. Many defective materials can occur within an aggregate. It is therefore essential that detailed petrographic evaluation is carried out, with particular reference to the intended use. An example of failure to do this was seen in the refurbishment of a small housing estate: white render was applied to face degraded brickwork. At first the result was splendid, but within 2–3 years brown rust spots appeared all over the white render because of the presence of very small amounts of iron sulphide (pyrite) in the sand used in the render.
Aggregate sources Sands and gravels can be obtained from river or glacial deposits, many of which are relatively young unconsolidated superficial deposits of Quaternary age. They may also be derived from older geological deposits, such as Triassic and Devonian conglomerates (to take English examples). Flood plain and terrace gravels are particularly important sources of aggregate because nature has already sorted them and destroyed or removed much of the potentially deleterious material; however, they may still vary in composition and particle size. Glacial deposits tend to be less predictable than fluvial deposits and are most useful where they have been clearly sorted by fluvial processes. Among the quarried rocks, limestones – particularly the Carboniferous limestones of the British Isles – have been widely used as aggregate. Similarly, many sandstones have suitable properties and are used as sources of aggregate, particularly where they have been thoroughly cemented. Compact greywackes have been widely used, notably the Palaeozoic greywackes of the South West and Wales. Igneous rocks are also a very useful source of quarried stone when crushed to yield aggregates; their character depends on their mineralogy and texture. Coarsely crystalline rocks such as granite, syenite, diorite, and gabbro are widely used, as are their medium-grained equivalents. Some finer-grained igneous rocks are also used, but the very finest-grained rocks are liable to be unsatisfactory for a wide range of purposes. Reserves of rocks such as dolerite, microgranite, and basalt tend to be small in comparison with the coarse-grained intrusive plutons. Conversely, some of the high-quality granite sources lie within very large igneous bodies, which sustain large quarries and provide a considerable resource.
Regional metamorphic rock fabrics generally make poor aggregate sources. On crushing they develop an unsatisfactory flaky shape. Schists and gneisses can provide strong material, but of poor shape. On the other hand, metamorphism of some greywackes and sandstones can provide material of high quality, especially when it has involved contact metamorphism associated with the intrusion of igneous rocks, producing hornfels or marble. Such thermally metamorphosed rocks often have a good fabric and provide useful resources.
Investigation of Sources There are three levels of investigation of the potential aggregate source. The first is the field investigation, in which the characteristics and distributions of the rocks present in the source can be established by mapping, geophysics, and borehole drilling. The second concerns the specific petrography of the materials. The third involves testing the physical and chemical properties of the materials. The material being extracted from the source must also be tested on a regular basis to ensure that there is no departure from the original test results and specification. Because sources are inevitably variable from place to place, there is always the risk that certain potentially deleterious components may appear in undesirable abundance. A number of features may make the aggregate unsuitable for certain purposes; these include the presence of iron sulphide (pyrite, pyrrhotite, and marcasite). Iron sulphide minerals are unacceptable because they become oxidized on exposure to air in the presence of moisture, producing iron oxides (rust) and sulphate. This can result in spalling of material from the surface of concrete and rendering. The presence of gypsum in the aggregate is also highly undesirable from the point of view of concrete durability. Gypsum is commonly found in aggregates from arid regions. The presence of gypsum in concrete leads to medium- to long-term expansion and cracking. Other substances can create both durability and cosmetic problems.
Extraction of Aggregates The development of aggregate quarries requires the removal of overburden and its disposal, the fragmentation of rock (usually by a scheme of blasting), and the collection and crushing of the blast product (see Quarrying). Critical to the success of the operation is the stability of the size of the feed material to the primary crusher. Screening is usually necessary to ensure that the particles are suitable for the crusher regime. At this stage it is also necessary to remove
36 AGGREGATES
degraded and waste material that is not required as part of the aggregate. In sand and gravel workings, the source material is excavated in either dry or wet pit working. In marine environments, the process is based on suction and dredging using two techniques. In the first, the dredger is anchored and a pit is created in the seabed; production continues as consolidated materials fall into the excavation. In contrast, trail dredging is performed by a moving vessel, which excavates the deposit by cutting trenches in the seabed. Extracted crushed rock, sand, and gravel are then prepared as aggregates through the use of jaw, gyratory, impact, and cone crushers. The type of crusher is selected according to the individual sizes of the feed material. Grading by screening is an adjunct to comminution and is also necessary in the production and preparation of the finished aggregate in cases where the particle-size distribution of the aggregate is important. The product is also washed and cleaned. The process of cleaning often uses density separation, with weak porous rock types of low density being removed from the more satisfactory gravel materials.
Classification The classification of aggregates has changed significantly over the years but has always suffered from the need to satisfy many different interests. Most commonly aggregates are divided into natural and artificial and, if natural, into crushed rock, sand, and gravel. If the aggregate is a sand or gravel, it is further subdivided according to whether it is crushed, partly crushed, or uncrushed. It may then be important to state whether the material was derived from the land or from marine sources. Once produced, the aggregate is identified by its particle size, particle shape, particle surface texture, colour, the presence of impurities (such as dust, silt, or clay), and the presence of surface coatings or encrustations on the individual particles. Detailed petrographic examination is employed so that specific rock names can be included in the description. This also helps in the recognition of potentially deleterious substances. However, the diversity of rock names means that considerable simplification is required before this classification can be used to describe aggregates. Following recognition of the main category of rock from the field data, more specific names can be applied according to texture and mineral composition. Because aggregates are used for particular purposes, they are sometimes grouped according to their potential use. This means that they may be incorrectly named from a geological point of view. The most obvious example of this is where
limestone is referred to as ‘marble’. In 1913 a list of petrographically determined rock types was assembled, with the rocks being arranged in Trade Groups. This was thought to help the classification of road stone in particular. It was presumed that each Trade Group was composed of rocks with common properties. However, the range of properties in any one Group is so large as to make a nonsense of any expectation that the members of the Group will perform similarly, either in tests or in service. The Trade Groups were therefore replaced by a petrological group classification. However, even rocks within a single petrographic group can vary substantially in their properties. For example, the basalt group includes rocks that are not basalt, such as andesite, epidiorite, lamprophyre, and spilite. Hence a wide range of properties are to be expected from among these diverse lithologies. In the first place a classification describes the nature of the aggregate in a broad sense: quarried rock, sand, or gravel; crushed or otherwise. Second, the physical characteristics of the material are considered. Third, the petrography of the possibly diverse materials present must be established. This may require the examination of large and numerous samples. While it may be reasonable to describe as ‘granite’ the aggregate produced from a quarry in a mass of granite, that aggregate will inevitably contain a wide range of lithologies, including hydrothermally altered and weathered rocks. Whether a rock is geologically a granite, a granodiorite, or an adamellite may be less significant for the description of the aggregate than the recognition of the presence of strain within the quartz, alteration of the feldspar, or the presence of shear zones or veins.
Aggregate Grading Aggregate grading is determined by sieve analyses. Material passing through the 5 mm sieve is termed fine aggregate, while coarse aggregate is wholly retained on this sieve (Figure 1). The fine aggregate is often divided into three (formerly four) subsets – coarse, medium, and fine – which fall within specified and partly overlapping particle-size envelopes. The size range is sometimes recorded as the ratio of the sieve sizes at which 60% passes and at which 10% passes. The shapes of the particles greatly affect the masses falling in given size ranges. For example, an aggregate with a high proportion of elongate grains of a given grain size would be coarser than an aggregate with flaky particles. This can affect the properties of materials made using the aggregate for, say, concrete, road materials, and filter design. Commonly materials needed for particular purposes have standard
AGGREGATES 37
Figure 1 Aggregate grades. (A) Fine sand suitable for mortars or render (width of image: 10 mm). (B) Coarse sharp sand or ‘concreting’ sand (width of image: 10 mm). (C) Coarse natural sand (width of image: 10 mm). (D) Flint gravel 5 10 mm (width of image: 100 mm). (E) Crushed granite 5 10 mm (width of image: 100 mm). (F) Crushed granite 10 20 mm (width of image: 100 mm).
aggregate gradings. These include, for example, mortars, concrete, and road-surface aggregates. It is sometimes useful to have rock particles that are much larger than the normal maximum, for example where large masses of concrete are to be placed. Commonly, however, the maximum particle size used in structural concrete is around 20 mm. An important parameter is
the proportion of dust, which is often taken as the amount passing the 75 mm sieve. In blending aggregates for particular purposes, it is usually necessary to combine at least two and possibly more size ranges; for example, in a concrete the aggregate may be a mixture of suitable material in the size ranges 0–5 mm, 5–10 mm, and 10–20 mm.
38 AGGREGATES
The grading curve – a plot of the mass of material passing each sieve size – also determines the potential workability of mixtures and the space to be filled by binder and can be adjusted to suit particular purposes. The grading curve can be designed to reduce the volume of space to less than 10% of the total volume, but at this level the aggregate becomes almost completely unworkable.
Particle Shape Particle shape is important in controlling the ability of the aggregate to compact, with or without a binder, and affects the adhesion of the binder to the aggregate surface. Shapes are described as rounded, irregular, angular, flaky, or elongate, and can be combinations of these (Figure 2). The first three are essentially
equidimensional. The shape is assessed by measuring the longest, shortest, and intermediate axial diameters of the fragments. In the ideal equidimensional fragment, the three diameters are the same. Particles with ratios of the shortest to the intermediate and the intermediate to the longest diameters of above about 0.6 are normally regarded as equidimensional. For many purposes, it is important that the aggregate particles have equant shape: their maximum and minimum dimensions must be very similar. Spherical and equant particles of a given uniform size placed together have the lowest space between the particles. Highly angular particles and flaky particles with high aspect ratios of the same grading can have much more space between the particles. The shape of the particles can significantly affect the properties and composition of a mixture. The overall space is also determined by the grading curve. Sometimes highly flaky particles such as slate can be used in a mixture if they are accompanied by suitably graded and highly spherical particles. Flakiness Index (British Standard 812)
The flakiness index is measured on particles larger than 6.5 mm and is the weight percentage of particles that have a least dimension of less than 0.6 times the mean dimension. The sample must be greater than 200 pieces. The test is carried out using a standard plate that has elongate holes of a given size; the proportion passing through the appropriate hole gives a measure of the flakiness index. Elongation Index (BS 812)
The elongation index is the percentage of particles by mass having a long dimension that is more than 1.8 times the mean dimension. This measurement is made with a standard gauge in which pegs are placed an appropriate distance apart.
Petrography
Figure 2 Examples of particular particle shapes. (A) Well rounded spherical metaquartzite. (B) Elongate angular quartzite. (C) Rounded flaky limestone.
The petrography of the aggregate is mainly assessed on the basis of hand picking particles from a bulk sample. Thin-section analysis either of selected pieces or of a crush or sand mounted in a resin is also employed. The petrographic analysis is essential to determine the rock types present and hence to identify potential difficulties in the use of the material. It allows recognition of potentially deleterious components and estimation of physical parameters. The experienced petrographer, for example, can estimate the parameters relevant to the use of a material for road surfacing. Published standards provide procedures for petrographic description, including the standards published by the American Society for Testing and
AGGREGATES 39
Materials and the Rilem procedures. These standards list the minimum amounts of material to be examined in the petrographic examination. In BS 812, for example, it is specified that for an aggregate with a maximum particle size of 20 mm the laboratory sample should consist of 30 kg. The minimum mass of the test portion to be examined particle by particle is 6 kg. Normally the analysis would be carried out on duplicate portions. The samples are examined particle by particle, using a binocular stereoscopic microscope if necessary. Unfortunately, this procedure does not cover all eventualities, and some seriously deleterious constituents within the material may be missed. A rock particle passing a 20 mm sieve may have within it structures that give it potentially deleterious properties (Figure 3). It is therefore essential that the aggregate is examined in thin section as well as in the hand specimen. It is helpful if the aggregate sample is crushed and resampled to provide a representative portion for observation in thin section. A large thin section carrying several hundred particles is required. Some of the potentially deleterious ingredients may be present at relatively low abundance. For example, the presence of 1–2% of opaline vein silica would be likely to cause significant problems. Where a sand or fine gravel is to be sorted by hand it is first divided into sieve fractions, typically using the size ranges 5 mm. These size fractions are analysed quantitatively by hand sorting in the same way as for coarse aggregate. The stereoscopic microscope is used to help with identification. Thin sections are also prepared from the sample using either the fraction passing the 1.18 mm sieve or the whole fine aggregate. The sample is embedded in resin and a thin section is made of the briquette so produced.
Specific Tests Measuring Strength, Elasticity, and Durability For quarried rocks it is possible to take cores of the original source material and to measure the compressive and tensile strengths of that material directly. It may be necessary to take a large number of samples in order to obtain a reliable representative result. However, for sands and gravels the strength of the material can rarely be tested in this way, and so a series of tests has been developed that simulate the conditions in which the aggregate is to be used. There is often a simple relationship between the flakiness index of the aggregate and its aggregate impact value (AIV) and aggregate crushing value (ACV). In general, the lower the flakiness index, the higher the AIV and ACV. Hence, comparing the AIV and ACV values with specifications requires knowledge of the flakiness index. Consideration also needs to be given to the shape of the aggregate following the test. Density and Water Absorption
Some of the most important quantities measured for an aggregate are various density values. These include the bulk density, which is the total mass of material in a given volume, including the space between the aggregate particles. The saturated surface-dry density is the density of the actual rock material when fully saturated with water but having been dried at the surface. The dry density is the rock density after drying. In making these measurements, the water absorption is also recorded. These provide data that are essential for the design of composite mixes. Aggregate Impact Value (BS 812)
Figure 3 An alkali reactive granite coarse aggregate particle (top) with cracks filled with alkali silicate gel. The cracks run into the surrounding binder, which appears dark and contains quartz rich sand as a fine aggregate.
The aggregate impact value provides an indirect measurement of strength and involves the impaction of a standard mass on a previously well-sorted sample. The result is obtained by measuring the amount of material of less than 2.36 mm produced from an aggregate of 10–14 mm. The lower the result, the greater the resistance of the rock to impaction. It is also useful to examine the material that does not pass the 2.36 mm sieve, and it is common to sieve the total
40 AGGREGATES
product at 9.5 mm to establish whether there is an overall general reduction in particle size. Aggregate Crushing Value (BS 812)
The aggregate crushing value provides an indirect assessment of strength and elasticity in which a wellsorted sample is slowly compressed. The lower the degradation of the sample, the greater the resistance to crushing. The size ranges used are the same as for the AIV test. 10% Fines Value (BS 812)
The 10% fines value is the crushing load required to produce degradation such that 10% of the original mass of the material passes a 2.36 mm sieve, the original test sample being 10–14 mm. The samples are subjected to two different loads, and the amount passing the 2.36 mm sieve in each test is measured. Typically the two results should fall between 7.5% and 12.5% of the initial weight. The force required to produce 10% fines is then calculated. Aggregate Abrasion Value (BS 812)
In determining the aggregate abrasion value, fixed aggregate particles are abraded with standard sand, and the mass of the aggregate is recorded before and after abrasion. The reduction in mass indicates the hardness, brittleness, and integrity of the rock. The Los Angeles Abrasion Value (ASTM C131 and C535)
To determine the Los Angeles abrasion value, a sample charge is mixed with six to twelve steel balls, and together these are rotated in a steel cylinder for 500 or 1000 revolutions at 33 rpm. This causes attrition through tumbling and the mutual impact of the particles and the steel balls. The sample is screened after the rotations are completed using a 1.68 mm sieve. The coarser fraction is washed, oven dried, and weighed. The loss in mass as a percentage of the original mass is the Los Angeles abrasion value. Micro Deval test
The Micro Deval test is widely used to determine the resistance of an aggregate to abrasion. Steel balls and the aggregate are placed in a rotating cylinder. The test may be carried out either wet or dry. The Micro Deval value is calculated from the mass of material that passes the 1.6 mm test sieve, as a percentage of the original aggregate mass. Polished Stone Value (BS 812, Part 114)
To determine the polished stone value, the aggregate is mounted in resin and the exposed surface is polished
using a wheel and standard abrasive. The result is measured using a standard pendulum, with the ability of the rock to reduce the motion of the pendulum giving an indication of the potential resistance of the aggregate to skidding. The sample is small and the result can vary according to the proportions of rock that are present. This test is difficult to perform reliably, and considerable practice is required to obtain a consistent result. In practice it is found that good skid resistance is derived from a varied texture in the rock with some variation in particle quality. Wellcemented sandstones and some dolerites tend to have high polished stone values, while rocks such as limestones and chert have very low polished stone values. Franklin Point Load Strength
The Franklin point load strength can be directly assessed for large pieces of rough rock. A load is applied through conical platens. The specimen fails in tension at a fraction of the load required in the standard laboratory compressive-strength test. However, the values obtained in the test correlate reasonably well with those obtained from the laboratory-based uniaxial compressive test, so an estimated value for this can be obtained, if necessary, in the field. Schmidt Rebound Hammer Value
The Schmidt Rebound Hammer test is a simple quantitative test in which a spring-loaded hammer travelling through a fixed distance strikes the rock in a given orientation. The rebound of the hammer from the rock is influenced by the elasticity of the rock and is recorded as a percentage of the initial forward travel. A sound rock will generally give a rebound value in excess of 50%, while weathered and altered rock will tend to give a much lower value. Magnesium Sulphate Soundness Test (BS 812)
In the magnesium sulphate soundness test the degradation of the aggregate is measured following alternate wetting and drying in a solution of magnesium sulphate. The test provides a measure of the tendency of the rock to degrade through the crystallization of salts or ice formation. The result is influenced by the porosity and particularly by planes of weakness in the aggregate. Freeze–Thaw Test
In the freeze–thaw test the aggregate is subjected to cycles of freezing and thawing in water. Each cycle lasts approximately 24 h. The temperature is reduced over a period of several hours and then
AGGREGATES 41
maintained at 15 C to 20 C for at least 4 h. The sample is then maintained in water at 20 C for 5 h. The cycle is repeated 10 times, and then the sample is dried and sieved, and the percentage loss in mass is determined. Slake Durability Index
A number of small samples of known mass are placed in a wire-mesh drum. The drum is immersed in water and rotated for 10 min. The specimens are dried and weighed, and any loss in weight is expressed as a percentage of the initial weight. This is the slake durability index. Methylene Blue Absorption Test
Methylene blue dye is dissolved in water to give a blue solution. It is absorbed from the solution by swelling clay minerals, such as montmorillonite. The quantity of potentially swelling clay minerals in a sample of rock is assessed by measuring the amount of methylene blue absorbed. Chemical Tests
Aggregates are commonly tested by chemical analysis for a variety of constituents, including their organic, chloride, and sulphate contents. Organic material is readily separated from the aggregate by, for example, the alkalinity of cement paste. Its presence leads to severe staining of concrete and mortar surfaces. Sulphate causes long-term chemical changes in cement paste, leading to cracking and degradation. Chloride affects the durability of steel reinforcement in concrete, accelerating corrosion and the consequent reduction in strength. Mortar Bar and Concrete Prism Tests
The durability of concrete made with a given aggregate is evaluated by measuring the dimensional change in bars made of mortar or larger prisms of concrete containing the specific aggregate. The mortar-bar test results can be obtained in a few weeks, but the prism test needs to run for many months or even years. The tests allow the recognition of components in the rocks or contaminants (e.g. artificial glass) that take part in expansive alkali–aggregate reactions.
Aggregates for Specific Purposes Railway Track Ballasts
Railway track is normally placed on a bed of coarse aggregate. A lack of fines is required: the desirable particle size is generally 20–60 mm. The bed requires a free-draining base that is stable and able to maintain
the track alignment with minimum maintenance. The aggregate is sometimes placed on a blanket of sand to prevent fines entering the coarse aggregate layer. The aggregate layer may be up to 400 mm thick. The favoured rock types are medium-grained igneous rocks such as aplite and microgranite. Sometimes hornfels is used. Some of the more durable limestones and sandstones are also used. Weaker limestones and many sandstones are generally regarded as unsatisfactory because of their low durability and ready abrasion. The desirable qualities for an aggregate used for ballast are that it must be a strong rock, angular in shape, tending to be equidimensional, and free from dust and fines. Aggregates for Use in Bituminous Construction Materials
Aggregates for use with a bitumen binder in building construction (as used in bridge decks and in the decks and ramps of multistorey car parks) require a high skid resistance. They must also be highly impermeable, protecting the underlying construction from water and frost attack and from the effects of deicing salts. The mix design is important: there should be a high bitumen content and a high content of fine aggregate and filler in the aggregate grading. A wide range of rocks of diverse origin and a number of artificial materials are used in the bituminous mixes. The rocks must be durable, strong, and resistant to polishing. The aggregate must show good adhesion to the binder and have good shape. Skid resistance is also dependent on traffic density and, in some instances, a reduction in traffic has improved skid resistance. Visual aggregates have been developed where high skid resistance is required, and these include calcined bauxite, calcined flint, ballotini, and sinopal. Blast furnace slags yield moderately high polished stone values. The light-reflecting qualities are also important, and artificial aggregates such as sinopal, with their very high light reflectivity, are valued. Resistance to stripping, i.e., the breakdown of the bond between the aggregate and the bituminous binder, is also important. Stripping is likely to result in the failure of the wearing course and not necessarily in failure of the base course. The stripping tends to be most conspicuous in coarsegrained aggregates that contain quartz and feldspar. Basic rocks show little or no detachment. The aggregate has considerable strength, particularly in the wearing course. As an example, the aggregate crushing value for surface chasing and dense wearing courses will typically be 16 to 23, while for the base course it may be as high as 30. Similarly, the aggregate impact value might be 23 in the wearing course and 30 in the base course.
42 AGGREGATES
Aggregates in Unbound Pavement Construction
Aggregate is sometimes used in construction without cement or a bitumin binder. Examples are a working platform in advance of construction, structural layers beneath a road system, a drainage layer, and a replacement of unsuitable foundation material. Aggregates for these purposes must be resistant to crushing and impact effects during compaction and in use, and when in place they must resist breakdown by weathering or by chemical and physical processes and must be able to resist freeze–thaw processes. It is likely that recycled aggregates will become increasingly important in these situations, although levels of potentially deleterious components, such as sulphate, may point to a need for caution in the use of such material. Aggregates for unbound construction often need to resist the ingress of moisture, since moisture rise and capillary transfer can cause progressive degradation. Mortar
Mortar consists of a fine aggregate with a binding agent. It is used as a jointing or surface-rendering material. Sands for mortar production are excavated from sand and gravel pits in unconsolidated clastic deposits and are typically dominated by quartz. They are used in their natural form or processed by screening and washing. Rock fines of similar grade can also be used. The most important feature of sand for mortar manufacture is that the space between the aggregate particles must generally be about 30% by volume. The volume of binder needs to be slightly greater than this volume, and hence a relatively high proportion of cement or lime may be required. Should the space be such that voids occur in the mix, the material will commonly show early signs of degradation and will be readily damaged by penetration of moisture. The space also appears to reduce the capacity of the mortar to bond with the substrate. The workability and ease of use of the mixture also depends on the shape of the particles and the grading curve. Very uniform sand tends to have a high void space and therefore requires a high cementitious or water content and tends to develop a high voidage. On the other hand, the grading may be such that the space between the particles is too small and the mixture becomes stiff. The strength and elastic modulus of the rocks are also important because the resultant mixture of paste and aggregate must match the strength and elasticity of the material to which the mortar is applied. If it is not, then partings are liable to develop between the binder and the substrate. Similarly, the material must exhibit minimal shrinkage
because again it might become detached from the substrate. Concrete
This very widely used material has a very diverse structure and composition and serves many purposes. It is composed of aggregate graded for the specific purpose and a binder containing cement. In general, the properties of the aggregate must match the intended strength and elasticity of the product, and it must be highly durable. For many purposes a combination of coarse and fine aggregate with a maximum particle size of 20 mm is used. The grading curve is designed such that an appropriate amount of space occurs between the particles – typically around 25% by volume of the mixture. There are numerous components of aggregate that perform adversely in the medium and long term, so careful study of the material is required before use. The defective components are described in several standards, along with procedures for measuring their effects on the concrete. Some of these are described below. In the 1940s it was recognized in the USA that certain siliceous aggregates could react with alkalis derived from Portland Cement. This led to spalling of concrete surfaces and cracking, sometimes in a spectacular manner. The phenomenon occurs throughout the world, and few rock sources are immune. An enormous amount of work has been carried out to evaluate the reaction, both in the laboratory and in structures. Major international conferences on the subject have been held. The alkalis for the reaction derive from the cement and are extracted into the pore fluid in the setting concrete. The concentration of alkali in the pore fluid can be affected by external factors as well as by the internal composition of the cement matrix. The rock reacting with the alkalis is typically extremely fine grained or has extremely small strain domains. Hence, fine-grained rocks, such as opaline silica within limestone, some cherts, volcanic glass, slate, and similar fine-grained metamorphic rocks, may exhibit a high degree of strain and so be able to take part in the reaction. More recently it has been found that certain dolomitic siliceous limestones are also to be avoided, again because they react with alkalis to cause significant expansion of the concrete and severe cracking.
See Also Building Stone. Geotechnical Engineering. Quarrying. Rock Mechanics. Sedimentary Environments: Alluvial Fans, Alluvial Sediments and Settings. Sedimentary Processes: Glaciers. Sedimentary Rocks: Limestones; Sandstones, Diagenesis and Porosity Evolution.
ANALYTICAL METHODS/Fission Track Analysis 43
Further Reading American Society for Testing and Materials (1994) Annual Book of ASTM Standards (1994), Section 4, Construc tion, Volume 04.02, Concrete and Aggregates. West Conshohocken: American Society for Testing and Materials. Be´ rube´ MA, Fournier B, and Durand B (eds.) (2000) Alkali Aggregate Reaction in Concrete. Proceedings of the 11th International Conference, Quebec, Canada. British Standards Institution (1990) BS812 Parts 1 to 3: Methods for Sampling and Testing of Mineral Aggre gates, Sands and Fillers, Parts 100 Series Testing Aggregates. British Standards Institution. Dolor Mantuani L (1983) Handbook of Concrete Aggre gates: A Petrographic and Technological Evaluation. New Jersey: Noyes Publications.
(1983) FIP Manual of Leightweight Aggregate Concrete, 2nd edn. Surrey University Press (Halsted Press). Hobbs DW (1988) Alkali Silica Reaction in Concrete. Thomas Telford. Latham J P (1998) Advances in Aggregates and Armour stone Evaluation. Engineering Geology Special Publica tion 13. London: Geological Society. Popovics S (1979) Concrete Making Materials. Hemi sphere Publishing Corporation, McGraw Hill Book Company. Smith MR and Collis L (2001) Aggregates, Sand, Gravel, and Crushed Rock for Construction Purposes, 3rd edn. Engineering Geology Special Publication 17. London: Geological Society. West G (1996) Alkali Aggregate Reaction in Concrete Roads and Bridges. Thomas Telford.
ALPS See EUROPE: The Alps
ANALYTICAL METHODS Contents Fission Track Analysis Geochemical Analysis (Including X-Ray) Geochronological Techniques Gravity Mineral Analysis
Fission Track Analysis B W H Hendriks, Geological Survey of Norway, Trondheim, Norway ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Ages obtained from isotopic dating methods are based on the ratio of parent and daughter isotopes. Radioactive decay of parent isotopes causes daughter isotopes to accumulate over time, unless they decay further or are lost by diffusion or emission. In the case
of the fission track method, the daughter product is not another isotope, but a trail of physical damage to the crystal lattice resulting from spontaneous fission of the parent nucleus. When the rate at which spontaneous fission occurs is known, the accumulation of such trails, known as fission tracks, can be used as a dating tool. Analogous to diffusional loss of daughter isotopes, the damage trails in the crystal lattice disappear above a threshold temperature by the fission track annealing process. Although the physics behind the annealing process are poorly understood, the outcome is empirically well known. Annealing initially causes the length of fission tracks to decrease and may eventually completely repair the damage to the crystal lattice. The latter is known as total annealing. The rate
44 ANALYTICAL METHODS/Fission Track Analysis
at which annealing takes place is a function of both mineral properties and temperature history. Fission tracks in geological samples have been wellstudied in mica (see Minerals: Micas), volcanic glass, tektite glass; (see Tektites) titanite, and zircon (see Minerals: Zircons). However, most research has been done on fission tracks in apatite, a widely disseminated accessory mineral in all classes of rocks. Retention of fission tracks in natural minerals takes place only at temperatures well below that of their crystallization temperature. Fission track dating will, therefore, document the crystallization age of a crystal only when it has cooled rapidly to surface temperatures immediatley after crystallization (see Analytical Methods: Mineral Analysis). Fission track dating of volcanic rocks can provide an age of crystallization, while fission track dating of more slowly cooled rocks will always yield an age that is younger than the age of crystallization. The amount of fission tracks per volume and their length will be a sensitive function of the annealing process and of the cooling history of the sample being studied. A cooling history can be constrained by thermal history modelling of fission track data (fission track age and fission track length distribution). Fission track analysis and thermal history modelling of apatite fission track data provide powerful tools with which to assess regional cooling and denudation histories. Following the rejuvenation of (U-Th)/He dating in the 1990s, the technique has become an important addition to the fission track method. (U-Th)/He dating can be applied to the same minerals as those commonly used in fission track analysis. (U-Th)/He dating is unique in its capability to constrain the very low temperature part of cooling histories of rock samples; the nominal closure temperature for apatite (U-Th)/He ages may be as low as 50 C. Apatite (U-Th)/He dating today is a well-established technique in itself, but in most studies it is used in combination with apatite fission track analysis. Many fission track research groups now routinely apply (U-Th)/He dating in parallel with fission track analysis. An introduction to (U-Th)/He dating is, therefore, included here.
Fission Tracks Fission tracks are linear damage trails in the crystal lattice. Natural fission tracks in geological samples are formed almost exclusively by the spontaneous fission of 238U. Other naturally occurring isotopes, such as 235U and 232Th, also fission spontaneously, but the respective isotopes have such low fission decay rates that it is generally assumed that all spontaneous fission tracks in naturally occurring
crystals are derived from 238U. The frequency of fission events is low compared to a-particle decay events, about 1 fission event for every 2 106 a-particle decay events. During spontaneous fission an unstable nucleus splits into two highly charged daughter nuclides (Figure 1). The two fission fragments are propelled in opposite directions, at random orientation with respect to the crystal lattice. The passage of the positively charged fission fragments through the host mineral damages the crystal lattice by ionization or electron stripping, causing electrostatic displacement. The end result is a cylindrical zone of atomic disorder with a diameter of a few nanometers – known as a fission track. Detailed information on the length of fission tracks is available for apatite only. Newly created apatite fission tracks have a length of 16.3 0.5 mm. Fission tracks can be observed directly through transmission electron microscopy, but with
Figure 1 Spontaneous fission of 238U (red spheres) produces two highly charged fission fragments (red half spheres) that recoil as a result of Coulomb repulsion. They interact with other atoms in the crystal lattice by electron stripping or ionization. This leads to further deformation of the crystal lattice as the ionized lattice atoms (blue spheres with plus sign) repel each other. After the fission fragments come to rest, a damage trail (‘fission track’) is left, which can be observed with an optical microscope after chemical etching. In apatite, the fission track annealing rate is higher for tracks at greater angle (y) to the crystallographic c axis. Therefore, tracks perpendicular to the c axis are on average shorter than tracks that are parallel to the c axis.
ANALYTICAL METHODS/Fission Track Analysis 45
Figure 2 Fission tracks in apatite (left) resulting from the spontaneous fission of 238U and induced fission tracks in mica (right) produced by irradiation in a nuclear reactor. Fission tracks in the mica outline a mirror image of the polished apatite crystal with which it was in close contact during irradiation. Fission tracks are revealed by chemical etching with HNO3 (apatite) and HF (mica). Only fission tracks that intersect the polished surface, cracks (track in cleavage, TINCLE) or other tracks (track in track, TINT) can be reached and enlarged by the etchant.
an optical microscope they can only be observed after revelation by chemical etching. Seen through an optical microscope, chemically etched fission tracks appear as randomly oriented cigar-shaped features (Figure 2).
Fission Track Annealing Laboratory experiments show that residence at elevated temperatures induces shortening of fission tracks. This process of track shortening by solid state diffusion is called fission track annealing. The rate of the annealing process is dependent on mineral properties and thermal history. Pressure and stress dependency have been suggested, but the evidence is ambiguous and highly controversial. When a sample cools below the total annealing temperature, it enters the Partial Annealing Zone (PAZ; APAZ in the case of apatite, ZPAZ for zircon). As the sample cools within the PAZ, tracks shorten by lesser amounts until becoming relatively stable at low (300 100 1640 80 >1300 1600 3000 3350 4000 5000 4500 300 4540
Hindu chronology Time for natural selection Sediment thickness/deposition rate Cooling of Earth Sediment thickness/deposition rate Cooling of Earth Salinity accumulation U Pb age of a Precambrian rock Sediment thickness/deposition rate Cooling of Earth Decay of U to Pb in crust Terrestrial Pb isotope evolution Radioactive isotope abundances Terrestrial Pb isotope evolution Terrestrial Pb isotope evolution
ca. 120 150 BCE/priests 1859/Darwin 1869/Huxley 1871/Kelvin 1890/de Lapparent 1897/Kelvin 1899/Joly 1907/Boltwood 1908/Joly 1917/Holmes 1927/Holmes 1947/Holmes 1949/Suess 1953/Houtermans 1981/Tera
a In addition to these estimates, Jewish and Christian Biblical scholars from the second through seventeenth centuries suggested that the age of Earth ranged between 5000 and 7500 years, based on Julian, Gregorian, or Hebrew calendars. Some of the most well known sources for these age estimates include James Ussher, John Lightfoot, and St. Augustine. Regardless of the source, most ages of Earth published prior to the twentieth century were greatly underestimated. Research on the decay rates and processes for radioactive elements in Earth’s crust finally led to more accurate calculations for Earth’s age by the middle the 1900s. These calculations were based on the reconstruction of terrestrial Pb isotopic compositions from a primordial Pb reservoir, of composition similar to meteorites. The meteorite reference for these calculations has been the Canyon Diablo troilite.
from the radioactive decay of U was discovered at the start of the twentieth century by physicists Rutherford, Soddy, Strutt, Thomson, and Boltwood. Boltwood measured Pb–U ratios in unaltered minerals using a very rough estimate of the rate for the radioactive decay of U to Pb; he noted that the older the mineral, the greater the ratio (greater amount of the decay product, Pb). Rutherford applied the decay of U to He in a similar way to attempt to obtain ages for rock samples. At this important watershed for geochronological techniques, the realms of physics and geology became linked in a quantitative tool for measuring geological time. Through the first half of the twentieth century, great advances were made in understanding and applying radiogenic isotope geochronology to determine the ages of rocks and the age of Earth. Arthur Holmes was among those who made important contributions to the development of radiogenic geochronological techniques in this period (Table 2). Despite the progress through the middle of the twentieth century in producing absolute age constraints on Earth and its rocks, scientists lacked a cohesive Earth model in which to place the geological processes they were dating. In the 1950s and 1960s, the fundamental step was made in this regard through development of the plate tectonic paradigm and magnetic stratigraphy; plate tectonics and magnetostratigraphy also contributed significantly to development of high-fidelity time-scales and geochronological tools (see History of Geology Since 1962).
Oceanographic cruises in the 1950s identified the presence of alternating ‘stripes’ of high and low magnetic intensity on the ocean floor. This pattern was clarified in the 1960s marine geophysical work of Hess and Dietz, who proposed the theory of seafloor spreading, and Vine and Matthews, who suggested that new oceanic crust was generated at ocean ridges and became magnetized in the direction of Earth’s magnetic field. The ocean-floor stripes revealed alternating periods in Earth’s history during which the magnetic field had changed from normal to reversed polarity. When these theories were combined with new results from palaeomagnetic studies conducted on sedimentary and volcanic rocks onshore, a globally applicable pattern of periods of normal and reversed magnetic polarities was gradually defined (Figure 3). This magnetic ‘stratigraphy’ was a relative time-scale useful for global ‘pattern matching’ of magnetic anomalies and for relative geochronology. The potassium-argon (K–Ar) radiogenic isotope geochronological technique, employed since the 1950s, was used to determine ages for fine-grained basalts used in the palaeomagnetic studies and thus placed absolute age constraints on points in the magnetic anomaly stratigraphy. Through combination of palaeomagnetic and K–Ar dating methods, the magnetic stratigraphy became better defined and, eventually, globally correlatable in terms of geological time. From the 1970s to the present, ties between palaeomagnetism, radiogenic isotope geochronology,
ANALYTICAL METHODS/Geochronological Techniques 83
Figure 3 Seafloor spreading. (A) Genesis of mirror image, normal, and reversed magnetic polarity patterns in new oceanic crust, on either side of an oceanic ridge axis. The rifted continental margins yielded to new oceanic crust as seafloor spreading commenced. Alternating black (normal) and white (reversed) polarity patterns would normally be recorded by shipborne or satellite surveys. Historically, magnetic reversals were subdivided into major epochs (Bruhnes, normal; Matuyama, reversed; etc.); smaller normal and reversed ‘events’ were identified within these overall periods of normal or reversed polarity. Precise ages for these reversal epochs and, importantly, the boundaries between epochs were initially obtained with potassium argon (K Ar) geochronology. Refinements since the 1960s of the number and duration of magnetic reversals as well as their absolute ages have been accomplished by detailed comparison to biostratigraphy, the astronomically calibrated time scale, and ages from radiogenic isotope dating methods. (B)His torical refinement of the Bruhnes (B) Matuyama (M) boundary, where, in 1963, K Ar dating indicated the epoch boundary to be at 1 Ma. The Jaramillo ‘event’ close to the Bruhnes Matuyama boundary had been discovered by 1966, and more precise K Ar dating placed the age of the epoch boundary at 0.73 Ma. By 2003, the combination of several dating methods, including K Ar and 40Ar/39Ar calibrations, astronomically calibrated time scales, and geomagnetic polarity time scales (GPTS), further refined the age of the boundary to a precise 0.789 Ma. (C) The magnetic anomaly map of the northern Atlantic Ocean between northern Norway, East Greenland, and Svalbard shows a real example of the alternating striped pattern of magnetic anomaly highs (red, normal polarity) and lows (blue, reversed polarity) on either side of the mid ocean ridge axis. The mid ocean ridge axis (trace identified with the single black line) separates a relatively symmetric, mirror image anomaly pattern in this part of the seafloor. Continent ocean boundaries are schematically indicated by thick black on white lines on the Norway and Greenland margins. (C) Reproduced with permission from Eide EA (coord.) BATLAS Mid Norway Plate Reconstruction Atlas with Global and Atlantic Perspectives, pp. 8 17. Trondheim: Geological Survey of Norway.
astronomically calibrated time-scales (ATSs), and biostratigraphy have facilitated definition of the geomagnetic polarity time-scale (GPTS) (Figure 2). Because of its tight calibration with these other methods, the GPTS provides the framework for most of the integrated time-scales presently in use for Jurassic and younger times (see Plate Tectonics, Magnetostratigraphy). Today, the GTS, the GPTS, and the ATS have been intercalibrated for some geological time periods.
Continued refinement and intercalibration of these time-scales will increase the possibility to make accurate age correlations for rocks and the geological events they represent. Important to recall is the fact that different geochronological techniques have been used to generate specific features of each time-scale, and that many techniques have particular geological time periods to which they are best suited; thus, complete intercalibration of these time-scales remains a challenging objective.
84 ANALYTICAL METHODS/Geochronological Techniques
Relative Geochronological Techniques Biostratigraphy
Methodology Biostratigraphy refers to correlation and age determination of rocks through use of fossils. Determining the environment in which the fossil species lived is inherent in this type of analysis. Theoretically, any fossil can be used to make physical correlations between stratigraphic horizons, but fossils that are best suited for making precise age correlations (time-stratigraphic correlations) represent organisms that (1) had wide geographic dispersal, (2) were shortlived, and/or (3) had distinct and rapidly developed evolutionary features by which they can now be identified. Fossils fulfilling these criteria are termed ‘index’ fossils. Both evolution and changes in local environment can cause the appearance or disappearance of a species, thus the time-significance of a particular index fossil must be demonstrated regionally through distinctions made between local environmental effects and time-significant events. Environmental effects may bring about the appearance/disappearance of a species because of local conditions, whereas time-significant effects may bring about the appearance/disappearance of a species because of evolution, extinction, or regional migration. Local environmental effects are not necessarily time significant and cannot be used in time correlations between different sedimentary units.
with magnetic reversal frequencies typically between 1 and 5 My. Some rock minerals (such as hematite or magnetite) may become magnetized in the same direction as Earth’s magnetic field (normal or reversed), either when a magmatic rock cools or when sedimentary rocks are deposited. As geochronological tools, palaeomagnetism and magnetostratigraphy rely on determining the magnetic polarity, including magnetic declination and inclination, of the sample’s remanent magnetic component. Palaeomagnetism uses these parameters to calculate a palaeomagnetic pole for the sampling site. An age for the pole is determined by matching the pole to a part of the apparent polar wander path (APWP) for that continent (Figure 4). Instead of using poles, magnetostratigraphy, as outlined previously, identifies a sequence of magnetic reversals in a sedimentary or volcanic section (Figure 2). The magnetostratigraphic profile is compared and matched to similar patterns in the GPTS and a chronology for the sampled interval is established. The absolute chronology of the GPTS is tied by radiogenic isotope methods, by calibration against the ATS, and/or by calibration with a well-defined biostratigraphic zone (see Magnetostratigraphy, Palaeomagnetism).
Application Fossils from the marine sedimentary record indicate existence of primitive life perhaps as early as 2.1 By ago, although the explosion of abundant life in the seas is usually tied to the start of the Palaeozoic era 544 million years ago (Ma). The continental sedimentary record indicates existence of plants and animals by Early Palaeozoic times, with recent indications of animals making forays from the seas onto land perhaps 530 Ma. Palaeozoic biostratigraphy, especially for the marine sedimentary record, is tied to precise, absolute ages for most period and stage boundaries, but gaps in the fossil record and/or the lack of isotopically datable rocks at key boundaries leave some discrepancies yet to be resolved. Biostratigraphy and fossil zone correlation are most precisely defined for the Mesozoic and Cenozoic eras; this is largely due to the ability to calibrate biostratigraphy not only with radiogenic isotope ages, but also with the GPTS and the ATS for these time periods.
Application Palaeomagnetism and magnetostratigraphy are most successfully applied to fine-grained volcanic and sedimentary rocks; the latter include red beds, siltstones, mudstones, and limestones. Matching of palaeomagnetic poles to established APWPs yields imprecise ages for rocks, but is useful for reasonable, first-order age estimates, probably within about 10 My for Phanerozoic through Late Proterozoic rocks. The GPTS is most accurately refined through about 175 Ma because of the availability of marine magnetic anomaly profiles to which onshore data can be referenced; nonetheless, magnetic stratigraphy and the GPTS extend through the Palaeozoic to the earliest datable Cambrian sedimentary rocks (Figure 2). Well-constrained magnetostratigraphy yields very precise ages for the following reasons: (1) geomagnetic polarity reversals are rapid, globally synchronous events, and lend themselves well to global, time-significant correlations; (2) polarity reversals are not predictable and yield unique reversal patterns; (3) significant parts of the GPTS have been astronomically tuned, intercalibrated with detailed biostratigraphy, and/or constrained with absolute radiometric ages.
Palaeomagnetism and Magnetostratigraphy
Chemostratigraphy
Methodology Earth’s magnetic field, generated in the liquid outer core, undergoes periodic reversals,
Methodology Non-radiogenic chemical geochronological tools for sedimentary rocks fall into one of
ANALYTICAL METHODS/Geochronological Techniques 85
Figure 4 Palaeomagnetic poles from gabbroic sills and interleaved sedimentary rocks of initially unknown ages were obtained from a study in northern Siberia. The poles for these rocks were compared to the apparent polar wander path (APWP) for Europe in the Mesozoic. Well known ages are indicated in millions of years (Ma) for different segments of the APWP (designated with green squares). Within the uncertainty ellipses for the poles from the Siberian samples, the ages of the rocks were suggested to be between 215 and 235 My. Subsequent radiogenic isotope age determinations on the sills confirmed this suggestion and refined the ages for the rocks to lie between 220 and 234 My.
three categories: pattern matching of time-stratigraphic shifts in stable isotope (O, C, or S) values and 87Sr/86Sr ratios, identification of siderophile element anomalies (Ir, Au, Pd, Pt, etc.), and chemical dating using amino acids. The principles for stable isotope methods are based on the fractionation of heavy and light isotopes of the stable elements O, C, and S. The heavy isotopes, 18O, 13C, and 34S, are compared, respectively, to the lighter isotopes 16O, 12 C, and 32S. Stable isotopic compositions are reported as ratios (for example, 18O/16O) relative to a standard for the same isotopic ratios. Processes causing fractionation of these isotopes depend primarily on temperature, isotope exchange reactions, and, in the case of S, change in oxidation state of sulphur compounds from action of anaerobic bacteria. The isotopic composition of Sr in sedimentary rocks is characterized by the 87Sr/86Sr ratio of the water from which the sediment precipitated; the water in the catchment area or in the ocean, in turn, will have an 87Sr/86Sr ratio that represents contributions from chemical weathering of rocks. Rocks of
varying ages and different mineralogies have distinct 87 Sr/86Sr ratios that will make different contributions of Sr to the water cycle. These contributions have been shown to vary over geological time in response to changes in the exposure and weathering of different landmasses. For purposes of geochronology, the principle of ‘pattern-matching’ is also used with these isotopic methods (Figure 5). Measured isotopic ratios in a stratigraphic sample suite representing some interval of geological time yield a curve (or excursion pattern) that is compared to a global reference or supraregional curve for the same isotopes. The global reference curve must, in turn, be calibrated to an absolute timescale by some independent means, usually matching the stratigraphic section in question to another section that is tied either to the GPTS or to absolute ages. Anomalously high concentrations of siderophile elements have been identified globally at three precisely determined time intervals: the Cretaceous– Tertiary boundary (65 Ma), the Eocene–Oligocene
86 ANALYTICAL METHODS/Geochronological Techniques
Figure 5 (A) Stable isotopes used in chemostratigraphy are commonly coupled with magnetostratigraphic and biostratigraphic information. In this fictive example, the stable isotope values for O and C were acquired for an entire sedimentary sequence of Cenozoic age. Magnetostratigraphy over the same zone may have revealed a pattern similar to that shown on the bar above the stable isotope curves, and this stratigraphy could then be correlated to the geomagnetic polarity time scale and used to calibrate the ages for the sedimentary column, which in this case spanned Pliocene through latest Eocene time. Biostratigraphy over the same stratigraphic column may have revealed a predominance of three types of microfossils, with different species within each microfossil group identified (designated here with different coloured symbols). Biostratigraphy might also be used to tie together and calibrate the stable isotope curves and make fine adjustments to ages determined with the magnetostratigraphic profile. Especially interesting would be to attempt to link any significant excursions in the isotope curves, either to changes observed in the microfossil distribution or to a specific time boundary. (B) Stable isotope stratigraphy can also be used over a larger time span for more regional or global correlations. This isotope curve for sulphur shows a marked change at about 240 Ma following a steady decrease through the Palaeozoic.
ANALYTICAL METHODS/Geochronological Techniques 87
boundary (33.7 Ma), and 2.3 Ma. Other anomalies – specifically, spikes in iridium concentrations in sedimentary sequences – have been suggested at the Triassic–Jurassic boundary and at the Devonian– Carboniferous boundary. These anomalous concentrations have been associated with catastrophic events, usually meteor impacts or massive volcanic eruptions, and faunal crises or mass extinctions. Because of their global nature, limited duration, and precisely defined ages, anomalous siderophile concentrations can serve as indirect dating tools in sedimentary sequences (see Impact Structures). The amino acid racemization (AAR) method uses the asymmetry of isomeric forms of several amino acids in fossil skeletal material to determine the time since the start of racemization. Racemization is the reversible conversion of one set of amino acid isomers to another set of isomers and begins with death of the organism. Sample materials are chemically treated and the amino acid types and isomer ratios are determined through chromatography methods. These ratios are used to calculate the time since the start of racemization through a formula containing a samplesite constant for the racemization rate. Because the racemization rate depends on external factors such as temperature, pH, and moisture, the rate varies between one sample site and another and must be calibrated for each site and each sample. This usually involves calibration against other samples (from the same sites) that have been dated by other methods. Application Oxygen isotope stratigraphy may be applied to planktonic foraminiferal tests in pelagic sediments that are at least 1 My old. Sulphur isotopes are most commonly used to date marine evaporites with ages of deposition extending through 650 Ma. Carbon isotopes may be used to date marine evaporites, marine carbonates, and (metamorphosed) marbles through Neoproterozoic age. Similarly, strontium, which substitutes readily for calcium, can also be used to date marine carbonates, apatite in marine sediments, and marbles through the Neoproterozoic. All of the isotope methods generally require samples that have been relatively unaltered by postdepositional events such as erosion, bioturbation, metamorphism, or recrystallization during diagenesis. Notably, work with metamorphosed marbles has indicated that C and Sr isotopes may maintain their original sedimentary deposition ratios despite having undergone extreme changes in pressure, temperature, and deformation subsequent to deposition. Siderophile element anomalies are confined to the sedimentary rock record; the most well-documented anomaly is at the Cretaceous–Tertiary boundary (see Mesozoic: End Cretaceous Extinctions). The AAR
method is restricted primarily to dating Holocene foraminifers extracted from pelagic sediments, although ages have also been determined for coprolites and mollusc shells.
Absolute Geochronological Techniques Radiogenic Isotope Techniques
Methodology The natural decay of a radioactive isotope to a stable isotope occurs at a regular rate that is described by the decay constant (l). The decay process is defined by an exponential function represented by the decay ‘half-life’ (t1/2); the half-life is equivalent to the amount of time necessary for onehalf of the radioactive nuclide to decay to a stable nuclide form. Radiogenic isotope techniques use this principle to calculate the age of a rock or mineral through measurement of the amount of radioactive ‘parent’ isotope and stable ‘daughter’ isotope in the sample material. The parent/daughter ratio and the decay constant for that isotope series are used to calculate how much time had to elapse for all of the stable daughter isotope to have been produced from an initial reservoir of radioactive parent isotope in the material (Table 3). This calculation presumes (1) no net transfer of radiogenic parent, stable daughter, and/or intermediate radioactive isotopes in or out of the sample material (mineral or rock) since time zero, (2) no unknown quantity of daughter isotope in the sample at time zero, and (3) that decay constants have not changed over the history of Earth. Many radiogenic isotope techniques are presently used to determine the ages of geological materials; the choice of appropriate isotopic system to determine an age of a sample depends primarily on the composition of the sample material, the geological ‘event’ or ‘process’ to be dated, and the sample’s age. The latter is directly linked to the half-life of the isotope system: radionuclides with long half-lives can be used to date very old samples, whereas those with shorter half-lives are restricted to dating younger rocks. In addition to the naturally occurring radioactive isotopes, a number of nuclear reactions of cosmic rays with gas molecules will produce radionuclides, the so-called cosmogenic radionuclides. The most long-lived of these can be used for age determinations based on principles similar to those outlined for the other radioactive isotopes. Applications The methods routinely used to date terrestrial metamorphic or igneous rocks and their minerals include techniques utilizing U/Th/Pb, Pb/ Pb, Sm/Nd, Lu/Hf, Re/Os, Rb/Sr, K–Ar, and Ar/Ar (Table 3). All of these isotopes have half-lives >1 By,
Intermediate productsa
Decay scheme
Half life (years)
Sample material
Typical geological ‘events’ dated
206
From 238U: 234Th, 234 Pa, 234U, 230Th, 226 Ra, 222 Rn, 218Po, 218At, 218 Rn, 214Po, 210 Pb, 210Bi, 210Po
Chain: 238U ! 206 Pb, 235U ! 207 Pb, 232Th ! 208 Pb
238
U 4.468 109, 235 U 0.7038 109, 232 Th 14.01 109
Zircon, thorite, monazite, apatite, xenotime, titanite, uraninite, thorianite
Crystallization age (from melt or from medium to high metamorphic grade); age of Earth
U and Th are concentrated in the liquid phase and are typically incorporated in more silica rich fractions; half lives of the parent isotopes are much longer than those of intermediate products; Pb isotopes alone in rocks without U or Th can be used to calculate ‘model ages’ (with information on crustal growth)
235
207
232
208
From 235U: 231Th, 231 Pa, 227Ac, 227Th, 223 Ra, 219Rn, 215Po, 214 At, 211Bi, 211Po From 232Th: 228Ra, 228 Ac, 228Th, 224Ra, 220 Rn, 216Po, 212Pb, 212 Bi, 212Po, 208Pb
Sm/Nd
147
143
None
1.06 1011
Crystallization age (from melt or from medium to high metamorphic grade)
Lu/Hf
176
176
None
Garnet, pyroxene, amphibole, plagioclase; mafic and ultramafic igneous and metamorphic whole rocks; lunar rocks Apatite, garnet, monazite, zircon, xenotime, meteorites, lunar rocks
Ages calculated from analysis of isotopes in separated minerals or cogenetic rocks; Sm and Nd are rare earth elements that tend to be less mobile during metamorphism and weathering Can also be used for information on differentiation of the mantle and crustal growth; 176Yb branch of decay can be ignored for purpose of geochronology
Method
U/Th/Pb, Pb/Pb
Radioactive parent
Stable daughter
238
U
U
Th
Sm
Lu
Pb
Pb
Pb
Nd
Hf
Decay schemes produce alpha (4He) particles; used for (U/ Th)/He dating Simple: 147Sm ! 143 Nd (alpha decay)
Branched: 176Lu ! 176Hf (gamma ray emission); 176 Lu ! 176Yb (electron capture)
3.54 1010
Meteorite formation; high grade metamorphism; igneous crystallization
Comments
Continued
88 ANALYTICAL METHODS/Geochronological Techniques
Table 3 Common radiogenic isotope geochronological techniques
Table 3 Continued Stable daughter
Re/Os
187
187
None
Simple: Re ! 187 Os (beta particle emission)
4.56 10
Rb/Sr
87
86
None
Simple: 87Rb ! 86 Sr (beta particle emission)
4.88 1010
K Ar, 40 Ar/39Ar
40
40
None
Branched: 40K ! 40 Ca (beta emission); 40 K ! 40Ar (beta emission and electron capture)
1.25 1010
Carbon 14
14
14
14
14
a
Re
Rb
K
C
Os
Sr
Ar
N
Intermediate productsa
C produced in atmosphere by collision of thermal neutrons (from cosmic rays) with 14 N; 14C is oxidized rapidly and radioactive CO2 enters the carbon cycle; radioactive 14 C decays
Decay scheme 187
C ! 14N
Half life (years) 10
5700
Sample material
molybdenite, osmiridium, laurite, columbite, tantalite, Cu sulphides; ores, meteorites Mica, feldspar, leucite, apatite, epidote, garnet, ilmenite, hornblende, pyroxene, clay minerals, some salts; felsic whole rocks, meteorites Mica, feldspar, feldspathoids, amphibole, illite, volcanic rocks, lunar rocks, low grade metamorphic rocks, glass, salts, clay minerals, evaporites Organic matter: wood, charcoal, seeds, leaves, peat, bone, tissue, mollusc shells
Typical geological ‘events’ dated
Comments
Ore deposit formation; iron meteorite formation
Enriched in metallic and sulphide phases; relatively depleted in silicates
Crystallization age (from melt or metamorphism); cooling (after high grade ‘event’); diagenesis
Because Rb and Sr have close relationships to K and Ca, respectively, the method is especially useful for study of granitic rocks
Crystallization of quickly cooled igneous rocks; cooling of metamorphic and plutonic rocks
K Ar method involves splitting the sample to measure K and Ar; 40 Ar/39Ar uses 39Ar as a proxy for K and measures only Ar isotopes, with no sample splitting; the 40 Ar/39Ar method is commonly used today
Time since the organic material ceased to take up carbon
Dendrochronology and varve chronology are often used in carbon 14 dating to account for secular variation in the 14 C content in the atmosphere
Note that the U Th Pb decay series involves numerous intermediate radioactive isotopes with short half lives (‘chain’ decay); only the direct intermediate products are listed here (products from branched decay have not been listed).
ANALYTICAL METHODS/Geochronological Techniques 89
Radioactive parent
Method
90 ANALYTICAL METHODS/Geochronological Techniques
so the samples can be used to date Earth’s oldest geological materials and events. Lunar and cosmogenic materials have also been dated with some of the same methods. The relatively shorter half-life of the K–Ar decay series, as well as the very short half-lives of the intermediate nuclides in the U and Th decay series, allow these isotope systems to be used for dating certain geological materials of Pleistocene (the U-series nuclides) and Holocene (the K–Ar and Ar/Ar methods) ages. Of the cosmogenic radionuclides, the most well known is probably carbon-14. The carbon-14 method is used to date organic materials; 14C has a half-life of 5700 years and is restricted to materials less than about 100 000 years old (Table 3). Aside from 14C, other cosmogenic radionuclides include 10Be, 26Al, 36 Cl, 41Ca, 53Mn, 81Kr, and 129I; these can be used for dating relatively young materials (on the order of several 100 000 years for Ca and Kr and up to 1 My or more for Be, Al, Cl, Mn, and I). Though not treated in detail here, these isotopes can be applied to date a range of materials, including Quaternary sediments, ice, manganese nodules, groundwater, and soils, and to determine the age of exposure of terrestrial land surfaces and meteorites (see Analytical Methods: Fission Track Analysis). Astronomically Calibrated Time-Scales
Methodology Perturbations in the orbit of Earth about the sun are generated by gravitational interactions between Earth and the sun, moon, and other celestial bodies. These orbital perturbations cause cyclical climatic changes that are recorded in some sedimentary rocks. This principle was recognized by G K Gilbert in the nineteenth century, and he noted the potential to use this climatically driven, sedimentary cyclicity to place age constraints on certain parts of the rock record. Since Gilbert’s time, astronomically calibrated time-scales have generated astronomical solutions for these perturbations in Earth’s orbit that match sedimentary cycles recognized in nature, such as glacial varve sequences (Figure 6). These gravity-induced perturbations apply specifically to the obliquity of Earth’s orbit, Earth’s axial precession, and the eccentricity of Earth’s orbit about the sun. Obliquity refers to the angle between Earth’s axis of rotation and the orbital plane, whereas precession is the movement (‘wobble’) of the rotation axis about a circular path that describes a cone. Eccentricity is the elongation of Earth’s orbit about the sun; this varies between a circular and an elliptical shape. The main periods of eccentricity of Earth’s orbit are 100 000 and 413 000 years. The obliquity of Earth’s axis has a main period of 41 000 years and precession of the axis has a main period of 21 000 years. Because the astronomically calibrated
Figure 6 Astronomically calibrated time scales attempt to re solve the long term gravitational perturbations in Earth’s orbit about the Sun. The mathematical solutions for the cyclicity of these perturbations are projected backward in time to determine the geological age of seasonal (solar) cycles preserved in the sedimentary rock record. Most astronomical calibrations define solutions for the precession and eccentricity of Earth’s orbit. In this example, cyclical sedimentation patterns (alternating dark and light sedimentary layers) in a fictitious marine sequence were carefully logged, as on the left hand column. The log is matched to the calculated solutions for orbital precession and eccentricity that are tied to absolute time. Where possible, the stratigraphic column may also be tied to magnetostratigraphic, biostratigraphic, and/or radiogenic isotope geochronology data.
time-scales are based only on factors related to Earth’s orbit about the sun, they are the only truly ‘absolute’ time-scales, following the strict definition of this word, and are mainstays for tying together or intercalibrating the other time-scales (see Earth: Orbital Variation (Including Milankovitch Cycles)). Applications The geologically short periodicity of Earth’s orbital perturbations has allowed calibration of precise astronomical time-scales for the past 15 My. Climate changes associated with ice ages have been the most easily recognized events in the rock record and the astronomical calibration of the Plio-Pleistocene time-scale remains one of the best. Although the Miocene-and-younger time-scales have been based primarily on the marine rock record,
ANALYTICAL METHODS/Geochronological Techniques 91
continental sedimentary sections have increasingly been incorporated in these calibrations. Work with astronomically calibrated lacustrine sections of Triassic–Jurassic age has demonstrated that older rocks can also be anchored to the astronomical time-scale. Dendrochronology
Methodology and Applications Dendrochronology applies the nonsystematic, climate-dependent variations in the thickness of annual tree rings of particular tree species to determine very exact dates for young events. Although restricted to use on Holocene samples, the high precision of the method (trees produce one ring per year, and uncertainties in ages determined with the method are usually 1 year) has also been used to calibrate carbon-14 ages (see also Table 3).
Future Considerations Geochronology furnishes the temporal framework for the study of geologic processes, giving data necessary to evaluate the rates, quantity, and significance of different rocks and geological ‘events’. Both relative and absolute ages are important in this regard and should be viewed as complementary methods through which different rock types may be correlated in time. Today, a big challenge facing geochronologists is the intercalibration of the various timescales. As part of this work, geologists working with radiogenic isotopes are attempting to refine the decay constants for a number of the commonly used radiogenic isotope dating methods. Inaccurate decay constants would clearly affect the accuracy of an age for a rock determined with a particular isotope system, and would have corresponding spin-off effects for ties made to magnetostratigraphic, biostratigraphic, chemostratigraphic, and astronomically calibrated datasets. Intercalibration of the various time-scales back through Mesozoic and Palaeozoic times will probably incorporate all of these methods, with extension of astronomical calibrations to the Palaeozoic probably involving ‘floating’ astronomical time-scales intercalibrated with the continually updated and refined GPTS and GTS.
Glossary decay constant A number describing the probability that a radioactive atom will decay in a unit time. half-life The time required for half of a quantity of radioactive atoms to decay.
isotopes Atoms with the same number of protons (¼ the same element), but a different number of neutrons (¼ different mass). radioactive decay The spontaneous disintegration of certain atoms whereby energy is emitted in the form of radiation; a new, stable atom is the result. siderophile An element preferring a metallic phase, with a weak affinity for oxygen or sulphur.
See Also Analytical Methods: Fission Track Analysis. Conservation of Geological Specimens. Creationism. Earth: Orbital Variation (Including Milankovitch Cycles). Dendrochronology. Famous Geologists: Cuvier; Darwin; Hutton; Lyell; Murchison; Sedgwick; Smith; Steno. Magnetostratigraphy. Mesozoic: End Cretaceous Extinctions. Palaeomagnetism. Palaeozoic: Cambrian; End Permian Extinctions. Plate Tectonics. Time Scale.
Further Reading Butler RF (1992) Palaeomagnetism: Magnetic Domains and Geologic Terranes. Cambridge, MA: Blackwell Sci entific Publications. Cox A (ed.) (1973) Plate Tectonics and Geomagnetic Re versals. San Francisco, CA: WH Freeman and Company. Dalrymple BG (1991) The Age of the Earth. Palo Alto, CA: Stanford University Press. Dickin AP (1995) Radiogenic Isotope Geology. Cambridge: Cambridge University Press. Doyle P, Bennett MR, and Baxter AN (1994) The Key to Earth History: An Introduction to Stratigraphy. Chichester: John Wiley and Sons. Eicher DL (1976) Geologic Time, 2nd edn. Englewood Cliffs, NJ: Prentice Hall. Eide EA (2002) Introduction plate reconstructions and integrated datasets. In: Eide EA (coord.) BATLAS Mid Norway Plate Reconstruction Atlas with Global and Atlantic Perspectives, pp. 8 17. Trondheim: Geological Survey of Norway. Faure G (1986) Principles of Isotope Geology, 2nd edn. New York: John Wiley and Sons. Geyh MA and Schleicher H (1990) Absolute Age Determin ation: Physical and Chemical Dating Methods and Their Application. Berlin: Springer Verlag. Hilgen FJ, Krijgsman W, Langereis CG, and Lourens LJ (1997) Breakthrough made in dating of the geological record. EOS 78(28): 285, 288 289. Lewis C (2000) The Dating Game One Man’s Search for the Age of the Earth. Cambridge: Cambridge University Press. Renne PR, Deino AL, Walter RC, et al. (1994) Intercalibra tion of astronomical and radioisotopic time. Geology 22: 783 786.
92 ANALYTICAL METHODS/Gravity
Gravity J R Smallwood, Amerada Hess plc, London, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The law of gravitational attraction between objects was deduced by Isaac Newton in the late seventeenth century. His ‘inverse square’ law stated that the force attracting two objects was proportional to the masses of the two objects and inversely proportional to the square of the distance between them (Table 1). Since the mass of the Earth is so great relative to the mass of objects on its surface, attraction of objects towards the Earth, i.e., their response to the Earth’s gravity field, is often an important factor affecting geological processes. Measurement of the gravity field of the Earth is in itself a useful tool for investigating the sub-surface, as mass variations below the surface cause variations in the gravity field. The measurement of the shape of the Earth and its mass distribution have been important to defining the baseline gravity field from which deviations can be measured, as usually the anomaly rather than the overall field strength is useful for geological applications. There are now many ways of acquiring gravity data on land, sea, air, and from space, appropriate to the many scales on which gravity studies can be applied. Gravity variations over thousands of kilometres can be used for studies of mantle convection, variations over hundreds and tens of kilometres are relevant for studies such as lithospheric flexure, plate tectonics (see Plate Tectonics), crustal structure, and sedimentary basin development, hydrocarbon (see Petroleum Geology: Exploration) and mineral exploration (see Mining Geology: Exploration), while gravity variations over tens of metres can be used in civil engineering applications.
The Earth’s Shape and its Gravity Field The gravitational potential of a perfectly uniform sphere would be equal at all points on its surface. However, the Earth is not a perfect sphere; it is an oblate spheroid, and has a smaller radius at the poles than at the equator. Surveys in the early eighteenth century, under the direction of Ch-M de La Condamine and M de Maupertius found that a meridian degree measured at Quito, Equador, near the equator, was about 1500 m longer than a meridian degree near Tornio, Finland, near the Arctic circle.
Subsequently, various standard reference spheroids or ellipsoids have been proposed as first-order approximations to the shape of the Earth, such as the World Geodetic System 1984 (Table 1). Given such an ellipsoid, a gravity field can be calculated analytically as a function of latitude. For example, a reference gravity formula was adopted by the International Association of Geodesy in 1967 (IGF67, Table 1), and another introduced in 1984 (WGS84, Table 1). The mean density of the Earth, which is fundamental to the calculation of gravitational attraction, was first estimated following an experiment in 1775 by the Rev. Neville Maskelyne, using a technique suggested by Newton. If the Earth was perfectly spherical and of uniform density, then a plumbline would point down towards the centre of the Earth because of the force of gravity on the bob. However, any nearby mass would deflect the plumbline off this ‘vertical’. Maskelyne and his co-workers measured plumb-bob deflections on the Scottish mountain, Schiehallion (Figure 1). They discovered that the mountain’s gravitational pull deflected the plumb line by 11.7 seconds of arc. This allowed Charles Hutton to report in 1778 that the mean density of the Earth was approximately 4500 kg m 3. This density value leads to an estimate of the mass of the Earth of about 5 1024 kg, not far from the currently accepted value of 5.97 1024 kg. The Schiehallion experiment had another distinction, in that in order to calculate the mass and centre of gravity of the mountain a detailed survey was carried out, and the contour map was invented by Hutton to present the data. Since the mass of the Earth is not distributed uniformly, the real gravity field does not correspond to that calculated for an ellipsoid of uniform density. The ‘geoid’ is a surface which is defined by points of equal gravitational potential or equipotential (Table 1), which is chosen to coincide, on average, with mean sea-level. The geoid is not a perfect ellipsoid, because local and regional mass anomalies perturb the gravitational potential surface in their vicinity by several tens of metres. For example, a seamount on the ocean floor, which is denser than the surrounding seawater, will deflect the geoid downwards above it. ‘Geoid anomalies’ are defined as displacements of the geoid above or below a selected ellipsoid. The concept of the geoid as the global mean sea-level surface can be extended across areas occupied by land. This provides both a horizontal reference datum and a definition of the direction of the vertical, as a plumbline will hang perpendicular to the geoid.
ANALYTICAL METHODS/Gravity 93
Table 1 Gravity formulae Quantity
Formula
Gravitational Force between two masses, F
F
Gravitational Acceleration, a
a
GM r2
Gravitational Potential, V
V
GM r
(Vertical) Gravity anomaly above a buried sphere, dgz See Figure 6 International Gravity Formula 1967 Gravitational acceleration, gt WGS84 Ellipsoidal Gravity Formula Gravitational acceleration, gt
Constants and variables
G
GMm r2
M m r
As above
4GDrb3 h
dgz
3ðx2
2 3=2
þh Þ
g0 ð1 þ a sin2 l þ b sin4 lÞ
gt
Dr b h x
Density contrast Radius of sphere Depth of sphere Horizontal distance
g0
Mean gravititational acceleration at equator, 9.7803185 ms 2 5.278895 10 3 2.3462 10 5 Latitude
a b l g0
g0 ð1 þ d sin2 lÞ q ð1 e sin2 lÞ
gt
Gravitational or Newtonian constant, 6.67 10 11 m3kg 1s 1 Mass of body (Mass of earth approx. 5.97 1024 kg) Mass of second body Distance As above
d e l
Mean gravititational acceleration at equator, 9.7803267714 ms 2 1.93185138639 10 3 6.6943999103 10 3 Latitude
WGS Formula atmospheric correction, dgt Latitude correction for relative gravity measurements, dgL Bouguer plate correction, dgB Free air correction, dgFA Free air anomaly, gFA
dgt
0:87 105 expð 0:116h1:047 Þ
h
Elevation
dgL
8:12 105 sin2 l dl
dl l
Distance in N S direction between readings Latitude
dgB
2pr Gh
r h h
Bouguer correction density Elevation Elevation
gobs
Observed gravity
Bouguer anomaly, gB
gB
dgT
Terrain correction
Flattening factor for ellipsoid, f
f
a c
Equatorial radius of Earth, 6378.14 km Polar radius of Earth, 6356.75 km
308:6 h
dgFA gFA
gobs gobs a
c c
gt þ ðdgL þ dgFA Þ gt þ ðdgL þ dgFA
dgB þ dgT Þ
1 298:26
Measurement of Gravity The first measurements of Earth’s gravity, by timing the sliding of objects down inclined planes, were made by Galileo, after whom gravitational units were named. 1 Gal is 10 2 m s 2, and the gravitational acceleration at the Earth’s surface is about 981 Gal. For convenience in geophysical studies of gravity anomalies, the mGal is usually used, or for local surveys ‘gravity units’ (g.u.) where 1 mGal ¼ 10 g.u. Gravity may be measured as an absolute or relative quantity.
Classically, absolute gravity has been measured with a pendulum consisting of a heavy weight suspended by a thin fibre. The period of the oscillation is a function of gravitational acceleration and the length of the pendulum. H Kater designed a compound, or reverse, pendulum in 1815, that allowed some instrument-dependent factors to be cancelled out. The instrument was superceded by methods based on observations of falling objects. In a development of the free-fall method, a projectile is
94 ANALYTICAL METHODS/Gravity
Figure 1 Sketch map of the area around Schiehallion, Scotland, by Charles Hutton. Plumb line deflections measured at stations north and south of the mountain allowed the first estimate of the density of the earth. (After R.M. Sillito with permission from Hutton (1778) ß The Royal Society.)
ANALYTICAL METHODS/Gravity 95
fired vertically upwards and allowed to fall back along the same path. The gravity measurement depends on timing the upward and downward paths, which may be by light beam-controlled timers or interferometry. Gravity differences can be measured on land with a stable gravity meter or gravimeter based on Hooke’s law. A mass extends a spring under the influence of gravity and changes in extension are proportional to changes in the gravitational acceleration. More sensitive are ‘astatic’ gravity meters, which contain a mass supported by a ‘zero-length’ spring for which tension is proportional to extension. When the meter is in position, a measurement is made of an additional force needed to restore the mass to a standard position, supplied by an auxiliary spring or springs, an electrostatic system, or an adjustment of the zerolength spring itself. Gravity meters working on this principle measure differences in gravity between stations and surveys may be tied to one or more base stations at which repeated measurement can be made. Astatic gravity meters can have a sensitivity of about 0.01 mGal. For applications where the gravity meter is subject to tilting and vibration, such as on board a ship or in an aircraft, isolation of the instrument is required such as providing a moving stabilised platform for the gravity meter and damping vibrations with appropriate shock absorption. When the gravity meter is moving, accurate data on the location and trajectory of the platform is required along with the gravity measurement. For airborne application, this requirement has been greatly assisted by the advent of the global positioning system (GPS) which allows rapid, precise, and accurate positioning (see Remote Sensing: GIS). Airborne gravity surveys, whether flown using fixed wing or helicopters, can provide economic, rapid, and non-invasive geophysical reconnaissance ideal for difficult terrain such as tundra, jungles, and wildlife reserves. Deviations in artificial satellite orbits can be used to determine the long-wavelength components of the Earth’s gravity field. Altimetry tools mounted on satellites have allowed much more detailed gravity mapping over the oceans, as sea surface height data can be processed to give the marine geoid. Geoid data can then be converted to gravity data with a series of numerical operations (Figure 2). Since the mean sea-level surface is the geoid, an equipotential surface, variations in sea surface height from the reference ellipsoid reflect density changes below the sea surface, largely from the density contrast at the seabed, but also from sub-seabed changes, such as crustal thickness changes.
Adjustments to Measured Gravity Signals The first correction that can be applied to measured gravity values is the correction for latitude, to account for the centrifugal acceleration which is maximum at the equator and zero at the poles (Table 1). For gravity measurements made on land, several further corrections must be made (Table 1). The ‘free-air correction’ is made to adjust for difference in height between the measurement point and sea-level. This does not make any assumptions about the material between the sea-level datum level and the observation point and uses the inverse square law and the assumption of a spherical Earth. The ‘Bouguer correction’, named after the French mathematician and astronomer, is used to account for the gravitational effect of the mass of material between measurement point and sea-level. This requires assumptions to be made about the density of material, and the Bouguer plate or slab formula is applied (Table 1), which further assumes that this material is a uniform infinite plate. Historically a ‘density correction’ value of 2670 kg m 3 has been used as a standard density for crustal material, and this corresponds to a Bouguer correction of 1.112 g.u./m, negative above sea-level. A ‘terrain correction’ may be applied to compensate for the effect of topography, again requiring assumptions about densities. Nearby mass above the gravity measurement station will decrease the reading and any nearby topographic lows will have been be artificially ‘filled in’ by the Bouguer correction so the correction is always positive. An additional correction to gravity measurements made on a moving vehicle such as an aeroplane or boat is the Eo¨ tvo¨ s correction, which depends on horizontal speed vector, latitude, and flight altitude.
Gravity Anomalies and Derivatives Since for most geological applications the perturbations in the gravity field across an area or feature of interest are more important than the absolute gravity values, it is standard to compute gravity anomalies by subtracting the theoretical gravity value from the observed. The Bouguer gravity anomaly is the observed value of gravity minus the theoretical gravity value for a particular latitude and altitude, as outlined in Table 1. The Bouguer gravity is commonly used on land where maps of gravity anomalies can be used to view gravity data in plane view and it is convenient to have topographic effects (approximately) removed. Offshore, the free-air gravity anomaly is most useful, as the measurements are straightforward to correct to the sea-level datum.
96 ANALYTICAL METHODS/Gravity
Figure 2 Marine free air gravity anomaly map derived from satellite altimetry (Sandwell and Smith (1997)). Warm colours indicate positive gravity anomalies. The gravity anomaly primarily indicates the shape of the seafloor, due to the strong density contrast from seawater to oceanic crust. Oceanic island chains, subduction zone trenches, and mid ocean ridges form features visible on this world map. Locations of Figures 4, 8 and 13 indicated. (Image courtesy of NGDC.)
ANALYTICAL METHODS/Gravity 97
High-pass filtering or subtraction of a planar function from gravity anomaly data may be undertaken to remove a ‘residual’ or background trend if the feature of interest is known to be shallow or a subtle perturbation to a strong regional gradient. Other treatments of gravity data include upward and downward continuation, by which different observation levels can be simulated, and computation of vertical or horizontal derivatives, which may emphasise structural trends in the data.
Applications and Examples Submarine Topography
The satellite-derived free-air gravity anomaly map over the oceans (Figure 2) strongly reflects the nearest
significant density change, the seabed. There are positive gravity anomalies over seabed topographic highs such as submarine seamounts and mid-ocean ridges and negative anomalies over bathymetric deeps such as the trenches associated with subduction zones, although long-wavelength isostatically compensated structures have no gravity anomaly above them. The coverage of the marine free-air gravity anomaly data can be exploited to produce sea-floor topography data (Figure 2). For this purpose, shipboard depth surveys, usually made with sonar equipment, are used to supply the long-wavelength part of the transfer function from gravity to topography. The shipboard data is usually considered accurate but limited in global coverage due to the spacing and orientation of survey ship tracks. Bathymetry interpolation using the satellite-derived gravity data highlights isostatically
Figure 3 Model of the gravity effect of convection in the Earth’s mantle. (A) Stream function of computer modelled mantle flow (B) 100 C temperature contours (C) Variation in seafloor depth given a 30 km thick elastic lithosphere above the convecting mantle (D) Modelled free air gravity anomaly (E) Modelled geoid (sea surface height) anomaly over the convecting mantle. (Reproduced with kind permission from McKenzie et al. (1980) ß Nature Publishing Group. http:www.nature.com)
98 ANALYTICAL METHODS/Gravity
compensated topography which has no long-wavelength expression in the gravity data alone. Mantle Convection
It is commonly accepted that the Earth’s mantle convects, and the flow of mantle material gives rise to gravity anomalies. Where mantle material is anomalously hot, it has a lower density than surrounding cooler mantle, and it will give rise to a negative gravity anomaly at the surface. This effect is, however, overprinted by the positive gravity anomaly caused by the upward deflection of the lithosphere above the rising anomalously hot mantle column or sheet (see Mantle Plumes and Hot Spots). There will, therefore, be a positive gravity anomaly over rising mantle material and a negative gravity anomaly where mantle is cool and sinking (Figure 3). Isostasy and Lithospheric Strength
Not all mountains would cause a gravitational plumbline deflection such as that observed at Schiehallion. Bouguer had observed that a plumb-line was only deflected by 8 seconds of arc towards the mountains during Condamine’s Quito survey, while his calcula-
tions suggested that it should have been deflected as much as 10 4300 . This anomalous lack of deflection was attributed by R. Boscovich in 1755 to ‘compensation’ for the mass excess of the mountain by underlying mass deficiency at depth. This fed into the development of ‘isostasy’, which addresses the issue of support for topography on the Earth’s surface. Two alternative early views of isostatic theory were put forward in the 1850s. John Henry Pratt suggested that the amount of matter in a vertical column from the surface to some reference level in the Earth was always equal, and that this was achieved by the material in the column having lower density material below mountains than below topographic lows. George Biddell Airy advanced the alternative view using the analogy of icebergs, that elevated surface topography was underlain by lowdensity crustal roots which effectively displaced denser underlying material. Subsequent studies have used gravity data to investigate these alternative models in different tectonic settings and included the additional factor of the strength of the lithosphere to support loads. At wavelengths shorter than about 500 km, the relationship between the gravity anomaly and topography
Figure 4 Free air gravity anomaly (A) and topography (B) in the region of the Hawaiian islands, Pacific Ocean (see Figure 2 for location). (C) A comparison of the observed admittance along the Hawaiian Emperor seamount chain (dots) with the predictions of a simple flexure model of isostasy, with varying elastic thickness, Te (lines). The observed admittance can be best explained with an elastic thickness for the lithosphere of 20 30 km. See Watts (2001) for more details. (Reproduced with kind permission from Watts (2001) ß Cambridge University Press.)
ANALYTICAL METHODS/Gravity 99
is controlled by the mechanical properties of the lithosphere, which may be strong enough to support short wavelength loads, for example, isolated mountains. At longer wavelengths, the flexural strength of the lithosphere is commonly insufficient to support loads. The relationship between the gravity anomaly and topography can described by the wavelength-dependent ‘admittance’ function. The rate of change from flexurally-supported topography at short wavelength to topographic support by base-lithospheric pressure variations and regional density variation at long-wavelength depends on the effective ‘elastic thickness’ of the lithosphere. Figure 4 shows the topography and gravity anomaly of some of the Hawaiian island chain and the calculated admittance. For these islands, a modelled elastic thickness of about 25 km matches the admittance data. Recently, methods have been developed to also include the effect of lithospheric loads both with and without topographic expression in estimation of the elastic thickness.
Figure 5 Typical variation of (A) porosity and (B) density with depth below seafloor for sands and shales in a sedimentary basin. Increasing vertical effective stress with depth causes com paction of the rock, reducing porosity and correspondingly in creasing density. Deeply buried sedimentary rocks, therefore, have higher densities than shallower rocks of similar lithology.
Density Contrasts, Analytical Models, and Non-Uniqueness
On a smaller scale, gravity anomaly maps provide the opportunity to identify and delineate sub-surface structures, as long as there are lateral density changes associated with the structure. Rocks at and near the surface of the Earth are much less dense than the Earth’s average density of approximately 5155 kg m 3, and crustal rocks are almost universally less dense than mantle rocks. An approximate density value of 2670 kg m 3 is often taken as an average value for upper crustal rocks while values of 2850 kg m 3 and 3300 kg m 3 have been used for overall crustal rocks and uppermost mantle, respectively, although these values vary with composition and temperature. Many sedimentary rocks are less dense than metamorphic and igneous rocks. Coal (1200–1500 kg m 3) is one of the least dense rocks, while chalks and siliciclastic sedimentary rocks (1900–2100 kg m 3) are generally less dense than massive carbonates (2600–2700 kg m 3). With the exception of porous extrusive examples, crustal igneous rocks have densities approximately ranging from 2700 to 3000 kg m 3. Density is not a diagnostic for lithology and variation in parameters such as porosity, temperature, and mineralogy can give significant density variability. Rocks with the lowest densities are those with very high porosities such as volcanic pumice, and in sub-aqueous environments recently deposited sediments. Density of sediments in a sedimentary basin tends to increase with depth as grains are compacted together (Figure 5). Igneous and metamorphic rocks tend to have higher densities than sediments as they frequently have negligible porosity and consist of relatively dense minerals.
Figure 6 Modelled gravity anomaly (A) along a transect through the centre of a buried sphere (B) of varying radius b, density contrast dr, and depth of burial h. The similarity in shape between the various cases shown highlights the difficulties of interpretation of gravity anomalies, as there are no unique solutions to explain a particular gravity anomaly.
100 ANALYTICAL METHODS/Gravity
Any non-uniformity in mass distribution results in lateral variability of the gravity field. For some simple geometrical shapes, a gravity anomaly can be calculated analytically (Figure 6). The buried sphere example illustrates the observation that deep density anomalies give rise to an anomaly over a wider surface distance than otherwise similar shallow anomalies, while greater density contrasts give larger anomalies than small density contrasts. The similarity
in the gravity anomaly curves for the example of a buried sphere (Figure 6) illustrates one of the problems that arises in interpreting gravity data: there is no unique density distribution that produces a particular gravity anomaly. Gravity models tend to be constructed using additional geological or geophysical data such as seismic refraction or reflection profiles, surface geology (Figure 7), borehole density measurements, magnetic, magneto-telluric, or
Figure 7 Bouguer gravity anomaly contours overlain on geological map of part of Eastern Pennsylvania, USA. Bouguer gravity anomaly highs occur over the horst blocks of dense Precambrian material and other lows and highs in the gravity field are associated with formations of varying densities. (Geological map courtesy of the Bureau of Topographic and Geologic Survey, Pennsyvania Department of Conservation and Natural Resources, gravity data courtesy of W. Gumert and Carson Services Inc. Aerogravity Division, PA, USA.)
ANALYTICAL METHODS/Gravity 101
electromagnetic surveys as appropriate to arrive at a plausible and consistent interpretation. Crustal Observations from Satellite Gravity
Circular gravity anomalies similar in shape to that calculated for a buried sphere are observed over discrete igneous centres (Figure 8), where dense igneous rocks are inferred to have intruded less dense crustal rocks at a point of weakness in the lithosphere. Some of these ‘bulls-eyes’ in the gravity field have topographic expression but others may not have been identified without the satellite-derived gravity map. This map also allows identification of other largescale crust-mantle interactions. One example is the set of south-ward pointing ‘V-shaped’ gravity anomalies flanking the mid-ocean ridge south of Iceland (Figure 8), which are caused by ridges and troughs in the top of the igneous crust. Although partially buried by sediment, these ridges have an expression in the
gravity anomaly map because there is a significant density contrast between the igneous upper crustal rocks and the young pelagic sediments draping them. Modelling in Conjunction with Other Data
A combination of gravity data and other data types is often productive. For example, oceanic fracture zones identified in the satellite-derived gravity anomaly map are useful in conjunction with ‘sea-floor stripes’ in magnetic anomaly data to determine the relative movement between tectonic plates (Figure 8). Gravity data is commonly used to verify interpreted seismic models. Empirical relationships between seismic velocity and density can be used to convert a seismic (see Seismic Surveys) velocity model into a density model and the predicted gravity anomaly compared with observations. The example shown in Figure 9 shows a crustal velocity model along a 400 km line in the North Atlantic that has been
Figure 8 Free air gravity anomaly (A) and Magnetic anomaly (B) over the area surrounding Iceland (see Figure 2 for location). The magnetic stripes form the record of magnetic field reversals during production of oceanic crust at the spreading centre. There are gravity anomaly highs over topographic highs such as the Reykjanes Ridge (R) and Kolbeinsey Ridge (K) spreading centres, and the extinct Aegir Ridge spreading centre (A). There are also circular gravity highs over igneous centres (IC) and linear anomalies along the continental margins (CM) and ‘V shaped’ ridges (V) which flank the Reykjanes Ridge and reflect propagating pulses of anomal ously hot mantle beneath the spreading centre. Red and white circles show the position of the present day spreading centre plate boundary. Solid white line shows flowlines from present spreading centre indicating direction of paleo seafloor spreading. These are determined from reconstruction of the magnetic stripes parallel to the fracture zones seen in the gravity data. Dashed white line shows area of oceanic crust disrupted by fracture zones (FZ); outside this area, oceanic crust was formed at a spreading centre without fracture zones on this scale. Dotted black line indicates approximate line of Figure 12. Location of Figure 9 indicated. (After Smallwood and White (2002) with permission Geological Society of London.)
102 ANALYTICAL METHODS/Gravity
Figure 9 Gravity model along a profile from Iceland to the Faroe Islands (see Figure 8 for location). The crustal density model (A) was constructed by converting a wide angle seismic velocity model to density using published empirical relationships. The gravity signature of the crustal model (B) did not match the satellite derived (crosses) or ship board/land based gravity anomaly measurements. When effects of varying lithosphere thickness and mantle density variations were included (C), a good match between model and data could be achieved (D). (After Smallwood et al. (1999) by permission ß American Geophysical Union.)
ANALYTICAL METHODS/Gravity 103
converted to density. While the gravity anomaly signal expected from the crust alone does not match the observed gravity, when reasonable mantle temperature and compositional variations are included, a good match to the data can be obtained. Gravity data is increasingly being incorporated into multivariable mathematical inversion projects in which
Figure 10 Free air gravity anomaly over Faroe Shetland area derived from satellite altimetry (Sandwell and Smith 1997) and shiptrack data (see Figure 8 for location). The dominant signal is the NW SE gravity reflecting the area of deepest water between the Faroe Islands and the Shetland Isles. Shorter wavelength features arise from geological structures (see Figures 11 and 12).
multiple datasets are simultaneously modelled in order to increase confidence in a particular interpretation of the subsurface. Modelling Over Sedimentary Basins
Since there is often a significant density contrast between crustal and mantle rocks, gravity data may provide useful constraints on crustal thickness variations, which can occur in continental as well as oceanic settings. Lithospheric extension, for example, may thin the crust along with the rest of the lithosphere. As the relatively low density crust is thinned, it may isostatically subside and the resulting topographic low may form a sedimentary basin (Figure 10). If assumptions are made about rock densities, gravity anomaly data can be modelled to infer the extent of crustal thinning. Simplified models of the subsurface can be constructed and adjusted until a match or matches can be made to gravity observations. Mathematical inversion may assist by identifying a model which produces a gravity field that has a minimum misfit to observations. In the example of this, shown in Figures 11 and 12, from the UK/Faroe-Shetland Basin, the gravity data is particularly valuable as flood basalts to the west of the basin make seismic imaging difficult. Although the top of the relatively dense mantle is elevated in the position where the crystalline crust is modelled to be most highly extended, there is a free-air gravity low caused by the dominance of the relatively low density water column and sedimentary fill which are constrained by seismic data, and the long wavelength effect of the thicker continental crust on the basin
Figure 11 Modelled and observed free air gravity (A) along a profile between the Faroe Islands and the Shetland Isles (see Figure 10 for location). The seafloor and other horizons (B) were partly constrained by seismic reflection data but beneath the basalt wedge reflections were not easy to interpret and the gravity modelling along this and other intersecting lines constrains a possible crustal model. (After Smallwood et al. (2001) with permission, Geological Society of London.)
104 ANALYTICAL METHODS/Gravity
Figure 12 Modelled and observed free air gravity along a profile southeast of the Faroe Islands (see Figure 10 for location). The seafloor and other horizons were partly constrained by seismic reflection data (A) but beneath the basalt reflections were not easy to interpret and magnetic anomaly (B) and gravity anomaly (C) modelling along this and other intersecting lines constrains a possible crustal model (D). (After Smallwood et al. (2001) with permission, Geological Society of London.)
margins. Another benefit added by gravity data to the understanding of this sedimentary basin was the requirement to add a unit with elevated density approximately 1 km thick in the centre of the basin to represent an interval intruded by igneous sills. The top of this unit was imaged well by seismic data but the thickness could not be estimated without the gravity model. Figure 12 shows the value of modelling magnetic anomaly data along with the gravity to constrain basalt thickness and internal structure. Another geological structure for which gravity data provides a useful tool of investigation is the Chicxulub impact crater in the northern Yukatan peninsular of Mexico. There is no dramatic surface expression of the site, but there are concentric circular rings apparent in the gravity anomaly (Figure 13). The
gravity anomaly arises as the crater has been infilled with relatively low-density breccias and Tertiary sediments. The double humped central gravity high is thought to correspond to a central uplift buried deep within the crater. The Chicxulub crater is one example where 3D gravity modelling has proved useful to constrain crustal structures in three dimensions. Smaller Scale Surveys
Spatial deviation of gravity measurements is often used to infer lateral variations in density. If sufficiently accurate measurements can be made, then small-scale lateral variations in density can be inferred. Gravity surveying may be the best tool to identify mineral deposits if the target ores have densities contrasting with their host rocks. Massive
ANALYTICAL METHODS/Gravity 105
Figure 13 Merged free air gravity (offshore) and Bouguer (on shore) gravity anomalies across Chicxulub impact crater, Yuka tan peninsular, Mexico (see Figure 2 for location). Bouguer anomaly calculated with a reduction density of 2670 kg m 3. Grav ity anomaly over Chicxulub is a 30 mGal circular low with a 180 km diameter, with a central 20 mGal high. (Courtesy of Mark Pilkington, Natural Resources Canada.)
sulphides have densities ranging up to 4240 kg m 3, and within host rocks of densities around 2750 kg m 3, a sulphide body having a width of 50 m, a strike length of 500 m, and a depth extent of 300 m would give a gravity anomaly of about 3 mGal. On a smaller scale, ‘micro-gravity’ surveys typically involve a large number of closely spaced gravity measurements aiming to detect gravity variations at levels below 1 mGal. These surveys may be designed for civil engineering projects where underground natural cavities in limestone or disused mine workings need to be detected, or depth to bedrock needs to be established. As with any gravity interpretation, any additional available information such as outcrop geological boundaries, density values of samples, or depths to important horizons may be incorporated in order to give a more realistic model.
Gravity Gradiometry Sometimes knowledge of the magnitude of the gravity field is not sufficient to resolve between competing geological or structural models. In the example shown in Figure 14, the conventional gravity data is rather insensitive to the geometry of the salt diapir as a dominant long-wavelength gravity signal originates
Figure 14 Cross sections across two gravity models. The mod elled gravity response (A) for a cross section over a fault block and small detached salt pillow (B) is very similar to the response over a bigger salt diapir (C) offset by some other changes to the model layers. Since uncertainty and noise in marine gravity data may be at a 1 mGal level, gravity modelling of the total field may not be able to distinguish between these models. Seismic data is often poor below the top of the salt. Courtesy of A. Cunningham.
from an underlying fault block. In this case, the gradients of the gravity field may provide additional assistance. An instrument to measure the gradient of the gravity field was developed by Baron von Eo¨ tvo¨ s in 1886, and a unit of gravity gradient was named after him (1 Eo¨ tvo¨ s ¼ 0.1 mGal km 1). The concept of his torsion balance was that two weights were suspended from a beam at different heights from a single torsion fibre, and the different forces experienced by the two weights would deflect the beam. The torsion balance was accurate but somewhat cumbersome and slow, and it was superceded by the more
106 ANALYTICAL METHODS/Gravity
Figure 15 Modelled components of the gradients of the gravity field over a cube. The different tensors represent changes in the gravity field gradient in different directions, for example, Tzz is the vertical gradient of the field measured in a vertical direction. It highlights edges of subsurface density contrasts. Tz is the (directionless) full gravity anomaly field. (Courtesy of C. Murphy and Bell Geospace Inc.)
convenient astatic gravity meters. However, recently declassified military technology has seen a renaissance of gravity gradiometry, with the availability of a full tensor gradiometer consisting of 12 separate accelerometers arranged in orthogonal pairs on three separate rotating disks. This instrument has a quoted instrumental accuracy of 10 11 gal. Five independent tensor measurements are made and can be modelled, each sensitive to different aspects of subsurface density variation (Figure 15). In addition to the valuable insights gained from this multicomponent data, the components can be recombined to give a high resolution ‘conventional’ gravity map which benefits from the precision of the instrumentation. In cases similar to the salt diapir example, inversion of such precise gravity data has been used together with a correlation between seismic velocity and density to produce a modelled seismic velocity field which can be used in seismic depth processing.
Extra-terrestrial Gravity Fields Gravity fields have been computed for the moon (see Solar System: Moon), and for Venus (see Solar System: Venus) and Mars (see Solar System: Mars),
from observations of variation in artificial satellite orbits. The Doppler shift of spacecraft signals is observed, giving the spacecraft velocity in the ‘line-ofsight’ direction. The gravity field is calculated using a combination of many observations of line-of-sight acceleration. In a similar method to that outline for terrestrial studies, the wavelength-dependent relationship between gravity anomalies and topography can be used to study the internal dynamics and the support of surface loads by the lithosphere. Gravity studies on Venus show that it has a similar lithospheric rigidity to the continents on Earth, despite its higher surface temperature, and that active mantle convection is responsible for the observed volcanic rises. In contrast, large gravity anomalies on Mars for example, a maximum anomaly of 344 mGal (from a spacecraft altitude of 275 km) over the crest of the Olympus Mons volcano, have led to the suggestion that the Martian lithosphere is extremely rigid. On the moon, circular positive gravity anomalies of up to 300 mGal have been identified, associated with basaltic lava flows infilling giant impact craters. These ‘mascons’ (mass concentrations) have provided a focus for debate on isostatic lunar history.
Conclusion Gravity is a versatile tool for investigation and can provide constraints on sub-surface structure on a wide variety of scales from man-made structures to the size of an entire planet. To unlock the information contained within the gravity field, gravity observations are best used in conjunction with other types of data such as surface topography, geological mapping, borehole information, and seismic data.
See Also Mantle Plumes and Hot Spots. Mining Geology: Exploration. Petroleum Geology: Exploration. Plate Tectonics. Seismic Surveys. Solar System: Venus; Moon; Mars.
Further Reading Bott MHP (1982) The Interior of the Earth, 2nd edn. Amsterdam: Elsevier. Fowler CMR (1990) The Solid Earth: An Introduction to Global Geophysics. Cambridge, UK: Cambridge University Press. Gibson RI and Millegan PS (eds.) (1998) Geologic Applica tions of Gravity and Magnetics: Case Histories. SEG Geophysical Reference Series 8/AAPG Studies in Geology 43. Tulsa, OK: Society of Exploration Geophysicists and the American Association of Petroleum Geologists.
ANALYTICAL METHODS/Mineral Analysis 107
Gumert WR (1998) A historical review of airborne gravity. The Leading Edge 17: 113 117. http://www.aerogravity.com/carson2.htm. Hansen R (1999) The gravity gradiometer: basic concepts and tradeoffs. The Leading Edge 18: 478, 480. Hildebrand AR, Pilkington M, Connors M, Ortiz Aleman C, and Chavez RE (1995) Size and structure of the Chic xulub crater revealed by horizontal gravity gradients and cenotes. Nature 376: 415 417. Hutton C (1778) An account of the calculations made from the survey and measures taken at Schiehallion, in order to ascertain the mean density of the Earth. Phil. Trans. Royal Soc. LXVIII: 689 788. McKenzie DP, Watts AB, Parsons B, and Roufosse M (1980) Planform of mantle convection beneath the Pacific. Nature 288: 442 446. McKenzie DP and Nimmo F (1997) Elastic thickness esti mates for Venus from line of sight accelerations. Icarus 130: 198 216. Milsom J (2002) Field Geophysics, 3rd edn. Chichester, UK: John Wiley and Sons. Sandwell DT and Smith WHF (1997) Marine gravity anom aly from Geosat and ERS 1 satellite altimetry. Journal of Geophysical Research 105: 10039 10054. (www.ngdc.noaa.gov)
Smallwood JR, Staples RK, Richardson KR, White RS, and the FIRE working group (1999) Crust formed above the Iceland mantle plume: from continental rift to oceanic spreading center. Journal of Geophysics Research 104(B10): 22885 22902. Smallwood JR, Towns MJ, and White RS (2001) The struc ture of the Faeroe Shetland Trough from integrated deep seismic and potential field modelling. Journal of the Geo logical Society of London 158: 409 412. Smallwood JR and White RS (2002) Ridge plume inter action in the North Atlantic and its influence on contin ental breakup and seafloor spreading. In: Jolley DW and Bell BR (eds.) The North Atlantic Igneous Province: Stratigraphy, Tectonic, Volcanic and Magmatic Pro cesses, pp. 15 37. London: Geological Society of London, Spec. Publ. 197. Smith WH and Sandwell DT (1997) Global Sea Floor Top ography from Satellite Altimetry and Ship Depth Sound ings. Science 277: 1956 1962. Telford WM, Geldart LP, and Sheriff RE (1990) Applied Geophysics, 2nd edn. Cambridge, UK: Cambridge Uni versity Press. Watts AB (2001) Isostasy and Flexure of the Lithosphere. Cambridge, UK: Cambridge University Press.
Mineral Analysis N G Ware, Australian National University, Canberra, ACT, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Mineral Analysis Mineral analysis involves determining the chemical relationships between and within mineral grains. Microanalytical techniques are essential, and methods include X-ray spectrometry and mass spectrometry. Electron probe and laser ablation procedures are commonly used techniques for major and trace element analysis, respectively (see Analytical Methods: Geochemical Analysis (Including X-Ray)). A chemical analysis of a mineral is expressed as a table of weight percent (wt.%) of its component elements or oxides. Concentrations lower than about 0.5 wt.% are often expressed as parts per million (ppm) by weight of element. These mineral analyses are easily converted into atomic formulas and thence into percentages of the end-member ‘molecules’ within the mineral group (see Table 1). Mineral analyses are used in descriptive petrology, geothermometry, and geobarometry, and in the understanding of petrogenesis. Sometimes thousands of analyses are
collected in the completion of a single research project. Large amounts of data are presented graphically, plotting concentrations of elements or ratios of elements against each other, thus illustrating chemical trends or chemical equilibrium (see Figure 1). In addition to the chemical analysis, a complete description of a mineral requires a knowledge of its crystallography. Both chemical composition and crystallography are required to predict the behaviour of minerals, and hence rocks, in geological processes. The discovery of each new mineral involves the determination of its crystal structure as a matter of routine using X-ray and electron diffraction techniques. Thus, when a monomorphic mineral is identified from its composition, its crystallography follows. Polymorphs may be identified by optical microscopy. Whereas it is sometimes convenient to identify an unknown mineral from its diffraction pattern, and although cell parameters can be used as a rough measure of end-member composition, crystallography no longer plays a major role in quantitative mineral analysis. It was once necessary to separate a mineral from its parent rock by crushing, followed by use of heavy liquids and magnetic/isodynamic separators. Up to a
108 ANALYTICAL METHODS/Mineral Analysis
Table 1 Analysis of garnet by EMPA/WDS for major elements and by LA ICP MS for trace elementsa Oxides (wt.%)
Atoms (Oxygen ¼ 12)
Trace (ppm)
End member (%)
SiO2 TiO2 Al2O3 Cr2O3 FeO MnO ZnO MgO CaO Na2O K2O
37.12 0.03 20.80 300 Ma) make up the dark summit. These older
Figure 5 The Glarus thrust in the Tschingelhore (between Flims and Elm in eastern Switzerland). The thrust fault is visible as a sharp horizontal contact between the older rocks that form the rugged peaks and the younger rocks that form the cliffs above the snowfields.
EUROPE/The Alps 131
Figure 6 Profile through the Helvetic nappes of the eastern Swiss Alps. The Helvetic nappes were displaced along the Glarus thrust over a distance of around 50 km. But the rocks above and below the Glarus thrust were also intricately deformed, as is evident from the fold and thrust structures. The Santis thrust displaced the uppermost part, the Cretaceous strata, of the Helvetic nappes an additional 10 km to the north. Deeper down, the crystalline basement rocks of the Aar massif now form an anticlinal upwarp. NHF, North Helvetic Flysch.
Figure 7 Folded strata in the flank of Piz d’Artgas (‘peak with arcs’), overthrust by older rocks forming the summit and the yellow cliff beneath.
rocks were emplaced along a thrust fault that is located near the base of the yellow cliff. Figure 8 is a profile across the Tauern window, where the upper crust of the European margin forms a large anticlinal fold. In the centre of the upwarp, erosion has removed the higher nappes, thus providing an insight into the formerly deeper parts of the orogen. The crystalline basement rocks in the core of the upwarp
were compressed and internally shortened. From the deformed mineral grains of the rocks it is possible to determine how much horizontal shortening and vertical stretching actually occurred and to reconstruct the shape of the upwarp prior to this homogeneous deformation. The present-day shape of the Tauern upwarp (Figure 6), as well as its reconstructed geometry prior to homogeneous shortening, provide a
132 EUROPE/The Alps
reminder of the ductile behavior that granitic rocks can exhibit in the course of plate collision. The Klippen nappe, a Penninic nappe in the French–Swiss Alps, is a classic example of a style
of internal deformation characterized by fold-andthrust structures (Figure 9). The lubricating layer (evaporites) at the base of the nappe consists of a thick layer of anhydrite. This rock type, which has a particularly low shear strength, forms when very shallow areas of seawater evaporate. The great thickness of the weak evaporite layer in the northern part of the section shown in Figure 9 facilitated the formation of large-scale folds, and the anhydrite was able to flow into and fill the fold cores. In the southern part of the nappe, the anhydrite layer is thinner and the deformation style is characterized by imbricate thrusting. Each thrust fault is parallel to the strata and followed the weak anhydrite layer.
The Making of the Alps
Figure 8 Profile across the Tauern window (Eastern Alps). (A) Present day geometry; (B) retrodeformed to the configuration that existed prior to homogeneous horizontal shortening and vertical stretching.
Geologists working in the Alps had recognized early on that oceanic sediments occurred within the mountain range and were juxtaposed with rock units typical for continents. The pyramid of the Matterhorn (Figure 10), for example, is composed of crystalline basement rocks that were formed more than 300 Ma and which originated in the former (Adriatic) margin of the continental African Plate. In contrast, the base of the pyramid consists of volcanic and sedimentary rocks that formed in an ocean basin (the Piemont Ocean) 170 to 100 Ma ago. The Piemont Ocean formed in response to divergent motion between the Eurasian and African plates (see Figure 3). The Alpine Orogen evolved in a number of steps associated with relative movements between the Eurasian and African plates. The ocean basins between the two continental plates were closed in the process. The
Figure 9 Profile across the Penninic Klippen nappe of the western Swiss Alps. The Klippen nappe consists of sediments of the former Brianc¸onnais swell that have been overthrust onto sediments scraped off of the Valais basin and the Piemont ocean (the Niesen and Gurnigel nappes, respectively). The nappe internal structure of the frontal north west part of the Klippen nappe is dominated by folding, whereas in the internal south east part, imbrications stemming from thrust faulting prevail. Ga: Gastlosen thrust, He: Heiti thrust.
EUROPE/The Alps 133
Figure 10 Crystalline basement rocks pertaining to the margin of the African continent build up the Matterhorn peak and overlie the younger volcanic and sedimentary rocks that formed in the Piemont Ocean.
Figure 11 Block diagram showing the three dimensional geometry of the ancestral Alps at 90 Ma. An east dipping subduction zone in the Western Alps had consumed the Piemont Ocean. The Brianc¸onnais continental fragment was entering this subduction zone. The Valais basin and the shelf seas of the European margin were the site of ongoing sedimentation.
first basin, the Piemont Ocean, closed in Cretaceous times (100 Ma). The second basin, the Valais, closed in Tertiary times (35 Ma). Closure of these basins resulted not only from head-on collision, but also involved strike–slip movements between the European and Adriatic margins. During Cretaceous times, convergence between the Eurasian and African plates was directed east and
west. The European margin (Figure 11) was approaching the Adriatic margin, which had already formed an ancestral mountain range. The Piemont Ocean had already been subducted along an eastdipping subduction zone. Small fragments of this ocean were scraped off of the descending plate and were attached to the upper plate, a process called ‘underplating’. The Brianc¸ onnais microcontinent
134 EUROPE/The Alps
Figure 12 North south profile through the Central Alps of eastern Switzerland, reconstructed to the geometry at 32 and 19 Ma. Comparison of the two profiles reveals that the orogen grew outward with time on both sides, and that the units in the central part of the orogen were raised to a higher level by the combined action of folding and erosional denudation.
was just in the process of being subducted, but parts of it were also attached to the upper plate. The Valais basin was still the site of sedimentation, as was the shelf of the European margin. In the region of the future Central and Eastern Alps, the east–west convergence was expressed as east–west dextral strike–slip movements. At about 40 Ma, the convergence between the Eurasian and African plates changed to a north–south orientation. As a consequence, a south-dipping subduction zone evolved, into which the Valais basin gradually disappeared. Again, a number of fragments were scraped off of the descending plate and were accreted to the upper plate. About 35 Ma, the two margins started to collide. During this north–south convergence, strike–slip movements took place in the ancestral Western Alps. In the Central and
Eastern Alps, the collision phase compressed the two margins and led to the stacking of crustal pieces, horizontal shortening, and vertical stretching. Figure 12 shows two stages of this collision phase in a crosssection trough the Central Alps, reconstructed for 35 and 19 Ma. The deformation of the two continental margins pushed crustal fragments up inclined thrust faults and uplifted parts of the orogen by large-scale folding and vertical stretching. As a consequence, the land surface of the ancestral Alps was uplifted. The ensuing high elevations caused precipitation and triggered enhanced erosion. Rivers built large fan deltas in the foreland of the Alps. As far as known, denudation kept pace with uplift during mountain building. Nevertheless, deep crustal fragments were exposed in the process, bringing to the surface samples of rock
EUROPE/Mediterranean Tectonics 135
that had been at depths of several tens of kilometres during the early stages of the formation of the Alps.
See Also Europe: Mediterranean Tectonics; Variscan Orogeny; Permian to Recent Evolution. Moho Discontinuity.
Further Reading Debelmas J (1974) Ge´ ologie de la France. Paris: Doin. Debelmas J (1979) De´ couverte Ge´ ologique des Alpes du Nord. Orleans: BRGM. Debelmas J (1982) De´ couverte Ge´ ologique des Alpes du Sud. Orleans: BRGM. Krenmayr HG (ed.) (2000) Rocky Austria: A Brief Earth History of Austria. Wien: Geological Survey of Austria. Labhart TP (2001) Geologie der Schweiz, 5th edn. Thun: Ott Verlag. Labhart TP and Decrouez D (1997) Ge´ ologie de la Suisse. Lausanne: Delachaux et Niestle´ . Lemoine M, deGraciansky P C, and Tricart P (2000) De l’Oce´ an a` la Chaıˆne de Montagnes: Tectonique des Plaques dans les Alpes. New York: Gordon & Breach.
Marthaler M (2001) Le Cervin est il Africain? Lausanne: Loisir et Pe´ dagogie. Marthaler M (2002) Das Matterhorn aus Afrika: Die Entstehung der Alpen in der Erdgeschichte. Thun: Ott Verlag. Neubauer F and Ho¨ ck V (eds.) (2000) Aspects of Geology in Austria. Reports of the Austrian Geological Society, Special Issue 92(1999). Wien: Austrian Geological Society. Nicolas A, Polino R, Hirn A, Nicolich R, and ECORS CROP Working Group (1990) ECORS CROP tra verse and deep structure of the western Alps: a synthesis. In: Roure F, Heitzmann P, and Polino R (eds.) Deep Structure of the Alps, vol. 156, Me´moires de la Societe Ge´ologique de France, pp. 15 28. Paris: Geological Society of France. Pfiffner OA, Lehner P, Heitzmann P, Mueller St, and Steck A (eds.) (1997) Deep Structure of the Swiss Alps. Results of NRP 20. Basel: Birkha¨user. Roure F, Bergerat F, Damotte B, Mugnier J L, and Polino R (1996) The ECORS CROP Alpine seismic traverse. Me´moires de la Societe Ge´ologique de France 170. TRANSALP Working Group (2002) First deep seismic reflec tion images of the Eastern Alps reveal giant crustal wedges. Geophysical Research Letters 29(10): 92 1 92 4.
Mediterranean Tectonics E Carminati and C Doglioni, Universita` La Sapienza, Rome, Italy ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction It is commonly accepted that Mediterranean geology has been shaped by the interplay between two plates, the African and European plates, and possibly also smaller intervening microplates. The Mediterranean was mainly affected by rifting after the Variscan Orogeny (see Europe: Variscan Orogeny): during the Mesozoic, oceanic Tethys areas and passive continental margins developed, where widespread carbonate platforms were formed. During the Late Mesozoic, the Mediterranean area was dominated by subduction zones (from east to west, the Cimmerian, Dinarides, and Alps–Betics), which inverted the extensional regime, consuming the previously formed Tethyan oceanic lithosphere and the adjacent continental margins. The composition (oceanic or continental), density, and thickness of the lithosphere inherited from the Mesozoic rift controlled the location, distribution, and evolution of the later subduction zones. The shorter wavelength of the Mediterranean orogens relative to other belts (for example, the Cordillera and
the Himalayas) is due to the smaller wavelength of the lithospheric anisotropies inherited from the Tethyan rift. The Mediterranean basin was, and still is, a collector of sediments derived from the erosion of the surrounding continents and orogens: the best examples are the Nile and Rhone deltas. In the past, other deltas deposited sediments in the bottom of the Mediterranean, and their rivers were later disconnected or abandoned: an example is the Upper Oligocene–Lower Miocene Numidian Sandstone, which was derived from Africa, deposited in the central Mediterranean basin, and partly uplifted by the Apennines accretionary prism. It is well known that, during the Messinian eustatic lowstand, the Mediterranean dried up several times, generating a salinity crisis during which thick sequences of evaporites were deposited in the basin. This generated a pulsating loading oscillation in the Mediterranean, because the repetitive removal of the water led to significant isostatic rebound across most of the basin, particularly where it was deeper, as in the Ionian, the Provenc¸ al, and the central Tyrrhenian seas. The direction of the relative motion between Africa and Europe since the Neogene is still under debate.
136 EUROPE/Mediterranean Tectonics
Most reconstructions show directions of relative motion between north-west and north-east. Recent space geodesy data confirm this overall trend, in which Africa has a north–south component of convergence relative to Europe of about 5 mm year 1, but they also show that the absolute plate-motion directions of both Europe and Africa are north-east and not north or north-west as is usually assumed (see the NASA database on present global plate motions, http://sideshow.jpl.nasa.gov:80/mbh/series.html). The main Cenozoic subduction zones in the Mediterranean are the Alps–Betics, the Apennines– Maghrebides, and the Dinarides–Hellenides–Taurides. Closely related to the Mediterranean geodynamics are the Carpathian subduction and the Pyrenees (Figure 1). The Mediterranean orogens show two distinct signatures, which are similar to those occurring on opposite sides of the Pacific Ocean. High morphological and structural elevations, double vergence, thick crust, involvement of deep crustal rocks, and shallow foredeeps characterize eastwards- or north-eastwards-directed subduction zones (Alps–Betics and Dinarides–Hellenides– Taurides). Conversely, low morphological and structural elevations, single vergence, thin crust, involvement of shallow rocks, deep foredeeps, and a widely developed back-arc basin characterize the westwards-directed subduction zones of the Apennines and Carpathians. This asymmetry can be ascribed to the ‘westward’ drift of the lithosphere relative to the mantle, at rates of about 49 mm year 1 as computed from the hotspots reference frame. All Mediterranean orogens show typical thrust-belt geometries with imbricate-fan and antiformal-stack associations of thrusts. The main factor that varies between orogens and within single belts is the depth of the basal de´ collement. The deeper it is, the higher is the structural and morphological elevation of the related orogen. Extensional basins are superimposed on these orogenic belts: on the western side are the Valencia, Provenc¸ al, Alboran, Algerian, and Tyrrhenian basins, on the eastern side is the Aegean Basin, and to the north is the Pannonian Basin (Figures 2 and 3). The Mediterranean can be divided into western, central, and eastern basins. The western Mediterranean is younger (mainly less than 30 Ma) than the central Mediterranean and eastern Mediterranean, which are mainly relics of the Mesozoic to possibly Cenozoic Tethys Ocean. Positive gravity anomalies occur in the deep basins (the Provenc¸ al, Tyrrhenian, and Ionian seas), where the mantle has been uplifted by rifting processes. In contrast, negative gravity anomalies occur along the subduction zones.
Western Mediterranean A characteristic feature of the western Mediterranean is the large variation in lithospheric and crustal thickness (Figure 5). The lithosphere has been thinned to less than 60 km in the basins (50–60 km in the Valencia trough, 40 km in the eastern Alboran Sea, and 20–25 km in the Tyrrhenian Sea), while it is 65–80 km thick below the continental swells (Corsica–Sardinia and the Balearic promontory). The crust mimics these differences, with a thickness of 8–15 km in the basins (Valencia trough, Alboran Sea, Ligurian Sea, and Tyrrhenian Sea) and 20–30 km underneath the swells (Balearic promontory and Corsica–Sardinia), as inferred by seismic and gravity data. These lateral variations in thickness and composition are related to the rifting process that affected the western Mediterranean, which is a coherent system of interrelated irregular troughs, mainly V-shaped, that began to develop in the Late Oligocene–Early Miocene in the westernmost parts (Alboran, Valencia, Provenc¸ al basins), becoming progressively younger eastwards (eastern Balearic and Algerian basins), culminating in the presently active east–west extension in the Tyrrhenian Sea (Figures 1, 2, 3, and 4). Heat flow data and thermal modelling show that the maximum heat flows are encountered in the basins: 120 mW m 2 in the eastern Alboran Sea, 90–100 mW m 2 in the Valencia trough, and more than 200 mW m 2 in the Tyrrhenian Sea. All these sub-basins appear to be genetically linked to the backarc opening related to the coeval ‘eastwards’ rollback of the westward-directed Apennines–Maghrebides subduction zone. Extreme stretching generated oceanic crust in the Provenc¸ al (20–15 Ma), Algerian (17–10 Ma), Vavilov and Marsili (7–0 Ma) basins. Between 25 Ma and 10 Ma, the Corsica–Sardinia block rotated 60 counterclockwise (Figures 1, 2, 3, and 5). In the southern Apennines, the choking of the subduction zone with the thicker continental lithosphere of the Apulia Platform slowed the eastwards migration of the subduction hinge (Figure 6), whereas in the central and northern Apennines and in Calabria subduction is still active owing to the presence in the foreland of the thin continental lithosphere of the Adriatic Sea and the Mesozoic oceanic lithosphere of the Ionian Sea, allowing rollback of the subduction hinge. The western Mediterranean basins tend to close both morphologically and structurally towards the south-west (Alboran Sea) and north-east (Ligurian Sea; Figures 1 and 6). The eastwards migration of the arc associated with the westwards-directed subduction generated right-lateral transpression along the entire east–west-trending northern African belt
EUROPE/Mediterranean Tectonics 137
Figure 1 Present geodynamic framework. There are four subduction zones with variable active rates in the Mediterranean realm: the westwards directed Apennines Maghrebides; the westwards directed Carpathians; the north eastwards directed Dinarides Hellenides Taurides; and the south eastwards directed Alps. The Apennines Maghrebides subduction related back arc basin of the western Mediterranean stretched and scattered into segmented basins most of the products of the Alps Betics orogen.
138 EUROPE/Mediterranean Tectonics
Figure 2 Palaeogeodynamics at about 15 Ma. Note the ‘eastward’ vergence of both the Apennines Maghrebides trench and the back arc extensional wave. The Liguro Provenc¸al basin, the Valencia trough, and the North Algerian basin were almost completely opened at 10 Ma. The Dinarides subduction slowed down, owing to the presence of the thick Adriatic continental lithosphere to the west, whereas to the south the Hellenic subduction was very lively owing to the presence in the footwall plate of the Ionian oceanic lithosphere. The Carpathians migrated eastwards, generating the Pannonian back arc basin, with kinematics similar to those of the Apennines. Provencal basin (19 15) Age of the oceanic crust.
EUROPE/Mediterranean Tectonics 139
Figure 3 Palaeogeodynamics at about 30 Ma. The locations of the subduction zones were controlled by the Mesozoic palaeogeography. The Alps Betics formed along the south eastwards dipping subduction of Europe and Iberia underneath the Adriatic and Mesomediterranean plates. The Apennines developed along the Alps Betics retrobelt to the east, in which oceanic or thinned pre existing continental lithosphere was present. Similarly, the Carpathians started to develop along the Dinarides retrobelt (i.e. the Balkans). The fronts of the Alps Betics orogen were cross cut by the Apennines related subduction back arc extension.
140 EUROPE/Mediterranean Tectonics
Figure 4 Palaeogeodynamics at about 45 Ma. The Alps were continuous with the Betics to Gibraltar, consuming an ocean located to the west.
EUROPE/Mediterranean Tectonics 141
(Maghrebides) and its Sicilian continuation, whereas left-lateral transtension occurs along the same trend in the back-arc setting just to the north of the African margin. The opposite tectonic setting is found in the northern margin of the arc. Subduction retreat generated calc-alkaline and shoshonitic magmatic episodes – particularly in the western margins of the lithospheric boudins – which were followed by alkaline-tholeiitic magmatism in the back-arc to the west.
Extension partly originated in areas previously occupied by the Alps–Betics Orogen, which formed in the Cretaceous due to the ‘eastwards’-directed subduction of Europe and Iberia underneath the Adriatic Plate and a hypothetical Mesomediterranean Plate (Figure 4). If Sardinia is restored to its position prior to rotation, it can be seen that during the Early Cenozoic the Alps were probably joined with the Betics in a double-vergent single belt. The western Alps, which are the forebelt of the Alps, were connected to the
Figure 5 During the last 45 Ma, the evolution of the Mediterranean along the trace shown on the map (inset) is the result of three main subduction zones: the early eastwards directed Alpine subduction; the Apennines subduction switch along the Alps retrobelt; and the Dinarides Hellenides subduction. The last two slabs retreated at the expense of the inherited Tethyan Mesozoic oceanic or thinned continental lithosphere. In their hanging walls, a few rifts formed as back arc basins, which are progressively younger towards the subduction hinges. The slab is steeper underneath the Apennines, possibly owing to the westwards drift of the lithosphere relative to the mantle.
142 EUROPE/Mediterranean Tectonics
Figure 6 Main tectonic features of the Mediterranean realm, which has been shaped during the last 45 Ma by a number of subduction zones and related belts: the double vergent Alps Betics; the single eastwards vergent Apennines Maghrebides and the related western Mediterranean back arc basin; the double vergent Dinarides Hellenides Taurides and related Aegean extension; the single eastwards vergent Carpathians and the related Pannonian back arc basin; and the double vergent Pyrenees.
EUROPE/Mediterranean Tectonics 143
Alpine Corsica; the Alps continued south-westwards into the Balearic promontory and the Betics. The retrobelt of the Alps, the southern Alps, also continued from northern Italy towards the south-west. In a double-vergent orogen, the forebelt is the frontal part, which is synthetic to the subduction and verges towards the subducting plate; the retrobelt is the internal part, which is antithetic to the subduction and verges towards the interior of the overriding plate. The westwards-directed Apennines–Maghrebides subduction started along the Alps–Betics retrobelt (Figures 3 and 5), where oceanic and thinned continental lithosphere occurred in the foreland to the east. Subduction underneath the Apennines–Maghrebides consumed inherited Tethyan domains (Figures 5 and 6). The subduction zone and the related arc migrated ‘eastwards’ at a speed of 25–30 mm year 1. The western Late Oligocene–Early Miocene basins of the Mediterranean nucleated both within the Betics orogen (e.g. the Alboran Sea) and in its foreland (e.g. the Valencia and Provenc¸ al troughs; Figure 3). At that time the direction of the grabens (40 –70 ) was oblique to the trend of the coexisting Betics orogen (60 –80 ), indicating its structural independence from the Betics Orogeny. Thus, as the extension cross-cut the orogen and also developed well outside the thrust-belt front, the westernmost basins of the Mediterranean developed independently of the Alps– Betics orogen, being related instead to the innermost early phases of back-arc extension in the hanging wall of the Apennines–Maghrebides subduction zone. In contrast to the ‘eastwards’-migrating extensional basins and following the ‘eastwards’ retreat of the Apennines subduction zone, the Betics–Balearic thrust front was migrating ‘westwards’, producing interference or inversion structures. The part of the Alps–Betics orogen that was located in the area of the Apennines–Maghrebides back-arc basin (Figure 1) has been disarticulated and spread out into the western Mediterranean (forming the metamorphic slices of Kabylie in northern Algeria and Calabria in southern Italy). Alpine type basement rocks have been dragged up in the Tyrrhenian Sea. Similarly, boudinage of the pre-existing Alps and Dinarides orogens occurred in the Pannonian Basin, which is the Oligocene to Recent back-arc basin related to the eastwards-retreating westwards-directed Carpathian subduction zone (Figures 1, 3, and 6). In the Pannonian basin, the extension isolated boudins of continental lithosphere that had been thickened by the earlier Dinarides orogen, such as the Apuseni Mountains, which separate the Pannonian basin from the Transylvanian basin to the east. The western Mediterranean back-arc setting is comparable with Atlantic and western Pacific back-arc basins that
show similar large-scale lithospheric boudinage, in which parts of earlier orogens have been scattered in the back-arc area, like the Central America Cordillera relicts that are dispersed in the Caribbean domain. The Apennines accretionary prism formed in sequence at the front of the pre-existing Alpine retrobelt, and, therefore, the central western Apennines also contain the inherited Alpine orogen of Cretaceous to Miocene age. There was probably a temporary coexistence of opposite subductions during the Late Oligocene to Early Miocene (Figure 5). Structural and geophysical data support the presence of an eastwards-migrating asthenospheric wedge at the subduction hinge of the retreating Adriatic Plate. The subduction flip, from the Alpine eastwards-directed subduction to the Apennines westwards-directed subduction, could be reflected in the drastic increase in subsidence rates in the Apennines foredeep during the Late Oligocene to Early Miocene. Westwards-directed subduction zones, such as the Apennines, show foredeep subsidence rates that are up to 10 times higher (more than 1 mm year 1) than those of the Alpine foredeeps. The subduction flip (Figure 5) could also be reflected in the larger involvement of the crust during the earlier Alpine stages than in the Apennines de´ collements, which mainly deformed the sedimentary cover and the phyllitic basement. It has been demonstrated that the load of the Apennine and Carpathian orogens is not sufficient to generate the 4–8 km deep Pliocene–Pleistocene foredeep basins, and a mantle origin has been proposed for the mechanism (slab pull and/or eastwards mantle flow). Paradoxically, the extension that determined most of the western Mediterranean developed in the context of relative convergence between Africa and Europe. However, it appears that the north–south relative motion between Africa and Europe at the longitude of Tunisia has been about 135 km in the last 23 Ma, more than five times slower than the migration of the Apennines arc, which has moved more than 700 km eastwards during the last 23 Ma (Figures 1 and 6). Therefore, the eastwards migration of the Apennines–Maghrebides arc is not a consequence of the north–south relative convergence between Africa and Europe but is instead a consequence of the Apennines–Maghrebides subduction rollback, which was generated either by slab pull or by the ‘eastwards’ flow of the mantle relative to the lithosphere deduced from the hotspot reference frame. The western Mediterranean developed mainly after the terminal convergence in the Pyrenees at about 20 Ma, which resulted from the Late Cretaceous to Early Tertiary counterclockwise rotation of Iberia, which was contemporaneous with the opening of the Biscay Basin.
144 EUROPE/Mediterranean Tectonics
In northern Africa, south of the Maghrebides (and the related Algerian Tell and Moroccan Riff), the Atlas Mountains represent an intraplate inversion structure, in which extensional (north-north-easttrending) and left-lateral (about east–west-trending) transtensional Mesozoic intercontinental rifts were later buckled and squeezed by Cenozoic compression and right-lateral transpression in the foreland of the Apennines–Maghrebides subduction zone. This is also indicated by the Mesozoic sequences in the Atlas ranges, which are thicker than the adjacent undeformed mesetas.
The lithospheric extension was active whilst the Apennines–Maghrebides accretionary prism advanced, generating an interplay of two tectonic settings working together, with thrusts advancing over an orthogonal extending area, generating both thrusts cutting normal faults and normal faults offsetting thrusts. The rifting of the Sicily Channel seems to be physically connected north-westwards to the rift in south-western Sardinia (Campidano graben) and south-eastwards to the Sirte Basin, off the coast of Libya. One possibility is that this rift is linked through transfer zones in Egypt to the Red Sea and the East African Rift.
Central Mediterranean The Malta escarpment (Figures 3 and 4), along the eastern coast of Sicily, is a physiographic feature that has been tectonically controlled since Triassic times. Rocks dredged from the Malta escarpment range from Mesozoic to Tertiary in age. The escarpment represents a Mesozoic continental margin that has been reactivated as a transtensional feature since the Pliocene. In spite of the Apennines and Hellenides Neogene subduction zones, two conjugate passive continental margins are preserved at the margins of the Ionian Sea, along the Malta escarpment to the south-west and the Apulian escarpment to the northeast. Based on the low heat flows (18–40 mW m 2) and the 4–8 km of sedimentary cover, the Ionian Sea is probably a remnant of the Mesozoic Tethys Ocean, confined by the two conjugate passive continental margins. The transition from continental crust to oceanic crust appears to be sharper to the north-east than to the south-west. The basin between south-east Sicily and south-west Puglia was about 330 km wide. The inferred oceanic ridge could have been flattened by thermal cooling and buried by later sediments. Stratigraphic and structural constraints to the north in the Apennines belt suggest that the Ionian Ocean continued to the north-west (Figure 5). This palaeogeography is supported by the seismicity of the Apennines slab underneath the southern Tyrrhenian Sea, which implies subducted oceanic lithosphere. The adjacent absence or paucity of deep seismicity does not imply the absence of subduction but can be interpreted as a reflection of the more ductile behaviour of the subducted continental lithosphere. The Sicily Channel and the Pelagian shelf off the coast of eastern Tunisia have been undergoing extension since at least Pliocene times; in other words Africa is moving south-westwards in relation to Sicily (Figure 1). This process is responsible for the two grabens of Pantelleria and Malta deepening the seafloor and for the generation of active alkaline magmatism (e.g. the ephemeral Ferdinandea Island).
Eastern Mediterranean The Dinarides, Hellenides, and Taurides are a polyphase orogen, representing the coalescence of at least two or three subduction zones since Mesozoic times (Figures 1, 4, 5, and 6). The orogen has a part synthetic to the north-eastwards-directed subduction, i.e. the forebelt verging south-westwards. The conjugate part of the orogen is the retrobelt, which verges north-eastwards and northwards (Balkans and Pontides). The existence of three subduction zones is supported by the occurrence of two distinct oceanic sutures, preserved as the ophiolitic suites of Vardar and the Sub-Pelagonian units, which represent two separated branches of the Mesozoic Tethyan Ocean and the present oceanic subduction of the Ionian Sea. It is commonly believed that the more internal (Vardar) suture zone is the older one. The polyphase orogen exhibits a similar architecture to the Alps, but duplicated. The Rhodope–SerboMacedonian and Sakarya (northern Turkey) massifs mimic the internal massifs of the Alps, which represent the continental margin of the hanging-wall plate. On the other side, to the south-west of the Vardar oceanic suture, the Pelagonian (Macedonia–Greece) and Menderes (northern Turkey) massifs correspond to the external massifs of the Alps, representing the continental lithosphere of the footwall plate. The Pelagonian basement is at the same time the hanging-wall plate for the more external north-eastwardsdirected subduction of the Sub-Pelagonian and Pindos Ocean, which was eventually closed by collision with the eastern margin of the Adriatic Plate. However, unlike the Alps, widespread extension developed in the Dinarides–Hellenides–Taurides orogen (Figures 1 and 6). This extension resulted in the low topography of the orogen in comparison with belts such as the Alps and the Zagros or the Himalayas. In the Balkans, the Rhodope, and the SerboMacedonian massifs, structural and stratigraphic data indicate an interplay of compressional and
EUROPE/Mediterranean Tectonics 145
extensional tectonics. A Cretaceous to Eocene compressive deformation was followed by the generation of Eocene grabens. A later (possibly Miocene) compression inverted and uplifted these grabens, but it was followed by extensional tectonics that have affected the Balkan peninsula since Pliocene times, determining the north-west-trending normal faults and the related east–west right-lateral and north– south left-lateral transtensive transfer faults. Northeastwards-directed subduction is continuing along the eastern side of the Adriatic, in the Ionian Sea underneath the Mediterranean Ridge (the accretionary prism), and on the northern side of the Levantine Sea, i.e. in the eastern Mediterranean beneath Cyprus (Figure 1). The convergence rates are faster underneath the Mediterranean ridge (up to 40– 50 mm year 1), decrease eastwards along the Cyprus segment, and have minimum values along the Adriatic coast. The convergence rate appears to be controlled by the composition of the foreland lithosphere: where it is oceanic and dense, such as in the Ionian Sea, the subduction is faster than in the Adriatic and Cyprus segments, where the downgoing lithosphere is continental and transitional oceanic– continental, respectively. In the orogen, calc-alkaline and shoshonitic magmatism has accompanied most of the subduction since Cretaceous times. The later extensional process in the anomalously called ‘backarc’ is possibly responsible for the transition to the alkaline magmatic signature. One of the best-known ophiolitic sequences in the world crops out in Cyprus: a complete oceanic section is exposed (from harzburgites and peridotites of the upper mantle to gabbros, sheeted dykes, lavas, and pelagic sediments of the crust). The island is an anticline involving the whole crust, and its culmination coincides with the Erathostene seamount in the subducting foreland. The Erathostene seamount is a structural high inherited from the Mesozoic–Cenozoic rift. Since at least Miocene times, there has been an independent and presently active subduction along the northern margin of the Black Sea, generating the Caucasus. Geodynamic reconstructions of the eastern Mediterranean explain the extensional tectonics either by westwards Anatolian extrusion or by gravitational collapse of thickened lithosphere. However, these mechanisms can be ruled out because plate-velocity vectors increase from eastern Anatolia to the Aegean and Greece. This contradicts the basic rule that the velocity field decreases away from the source of the energy, i.e. the supposed squeezing of Anatolia by the Arabia indenter, or the collapse of the Anatolian orogen. Moreover, the topographic gradient between
Anatolia and the Ionian deep basin is too small (less than 1 ) to provide sufficient energy to explain the present deformation. Instead, the simplistic view of the westward Anatolian escape would close the Aegean Sea. The plates involved in the geodynamic reconstructions of the eastern Mediterranean are Africa, Greece, Anatolia, Eurasia, and Arabia. Deformation is very active in all these areas. The most prominent geodynamic factor shaping the eastern Mediterranean is the north-east-directed subduction of Africa underneath Greece and the Anatolian Plate (Eurasia). Seismic lines across the Cyprus Arc at the southern margin of the Anatolian Plate show clear active compression and deformation of the seafloor. The Aegean Sea is generally considered to be a back-arc basin resulting from the aforementioned subduction. However, the Aegean Sea is characterized by a relatively thick crust (20–25 km) in spite of longstanding subduction, which has probably been active since at least the Cretaceous. The subduction zone migrated south-westwards to the present position of the Cyprus-Hellenic subduction zone, and the associated orogen was later replaced by extension. In the Aegean Sea, Alpine-type crustal thickening with high pressures and low temperatures was followed by noncoaxial crustal-scale extension. This is consistent with the initial emplacement of thrust-sheets of basement slices, which were later cross-cut by extensional or transtensional faults. In addition, extension and associated magmatism were and are migrating southsouth-westwards, and have developed particularly since the Oligocene, while subduction began much earlier. ‘Normal’ back-arc basins (e.g. the Tyrrhenian Sea) associated with westwards-directed subduction zones opened very fast (10–20 Ma) and are always contemporaneous with the subduction. Moreover, they are characterized by oceanization and eastwards migration of extension and related magmatism, features directly surrounded by a frontal accretionary wedge. In contrast, the accretionary wedge of the Hellenic subduction zone is the south-eastern prolongation of the Dinarides thrust belt, where no back-arc rift comparable to the Tyrrhenian Sea occurs. The extension in western Turkey, the Aegean Sea, Greece, and Bulgaria appears to be the result of differential convergence rates in the north-eastwardsdirected subduction of Africa relative to the hanging wall of disrupted Eurasian lithosphere. Relative to Africa, the faster south-eastwards motion of Greece than of Cyprus–Anatolia results in the Aegean extension. The differences in velocity can be ascribed to differential decoupling with the asthenosphere. In the back-arc basins of the western Pacific the asthenosphere replaces a subducted and retreated
146 EUROPE/Mediterranean Tectonics
slab; however, the Aegean rift represents a different type of extension associated with a subduction zone, in which the hanging-wall plate overrides the slab at different velocities, implying internal deformation. According to this geodynamic scenario, during the compressive events associated with north-eastwardsdirected subduction, basement rocks (both continental and ophiolitic slices) in western Anatolia and the Aegean Sea were uplifted and eroded. Later extension caused subsidence in the area, and the basement slices were partly covered by continental and marine sediments. During its development, the Aegean extension migrated south-westwards (Figures 5 and 6). The Aegean rift affects the Aegean Sea and all of continental Greece, and it can be followed to the east, where it is widely expressed in Turkey, and to the north-west in Bulgaria, Albania, Macedonia, Serbia, and Bosnia. At the same time, from the Oligocene to the present, to the north, the Pannonian basin developed as the back-arc of the Carpathians subduction, but migrating eastwards, and affecting mainly eastern Austria, Slovenia, Croatia, Hungary, and Romania. Therefore, in the central part of the former Yugoslavia, the Pannonian and Aegean rifts meet with opposite directions of migration.
See Also Europe: Variscan Orogeny; Permian to Recent Evolution; The Alps; Holocene. Plate Tectonics. Tectonics: Convergent Plate Boundaries and Accretionary Wedges; Mountain Building and Orogeny.
Further Reading Berckhemer H and Hsu¨ KJ (eds.) (1982) Alpine Mediterra nean Geodynamics. Geodynamics Series 7. Washington: American Geophysical Union. Calcagnile G and Panza GF (1980) The main characteristics of the lithosphere asthenosphere system in Italy and surrounding regions. Pure and Applied Geophysics 119: 865 879. Carminati E, Wortel MJR, Spakman W, and Sabadini R (1998) The role of slab detachment processes in the opening of the western central Mediterranean basins: some geological and geophysical evidence. Earth and Planetary Science Letters 160: 651 665. Catalano R, Doglioni C, and Merlini S (2001) On the Mesozoic Ionian basin. Geophysical Journal Inter national 144: 49 64. Cella F, Fedi M, Florio G, and Rapolla A (1998) Gravity modeling of the litho asthenosphere system in the Central Mediterranean. Tectonophysics 287: 117 138. Christova C and Nikolova SB (1993) The Aegean region: deep structures and seismological properties. Geophys ical Journal International 115: 635 653.
Dercourt J, Gaetani M, Vrielynck B, et al. (2000) Atlas Peri Tethys, Paleogeographical Maps. Geological Map of the World. Paris: CGMW. de Voogd B, Truffert C, Chamot Rooke N, et al. (1992) Two ship deep seismic soundings in the basins of the Eastern Mediterranean Sea (Pasiphae cruise). Geophys ical Journal International 109: 536 552. Doglioni C, Gueguen E, Harabaglia P, and Mongelli F (1999) On the origin of W directed subduction zones and applications to the western Mediterranean. In: Durand B, Jolivet J, Horva´ th F, and Se´ ranne M (eds.) The Mediterranean Basins: Tertiary Extension Within The Alpine Orogen, pp. 541 561. Special Publication 156. London: Geological Society. Durand B, Jolivet J, Horva´ th F, and Se´ ranne M (1999) The Mediterranean Basins: Tertiary Extension Within the Alpine Orogen. Special Publication 156. London: Geo logical Society. Frizon de Lamotte D, Saint Bezar B, Bracene R, and Mercier E (2000) The two main steps of the Atlas building and geodynamics of the western Mediterranean. Tectonics 19: 740 761. Gueguen E, Doglioni C, and Fernandez M (1998) On the post 25 Ma geodynamic evolution of the western Mediterranean. Tectonophysics 298: 259 269. Guerrera F, Martin Algarra A, and Perrone V (1993) Late Oligocene Miocene syn / late orogenic successions in west ern and central Mediterranean chain from the Betic cor dillera to the southern Apennines. Terra Nova 5: 525 544. Huguen C, Mascle J, Chaumillon E, et al. (2001) Deform ational styles of the eastern Mediterranean Ridge and surroundings from combined swath mapping and seismic reflection profiling. Tectonophysics 343: 21 47. Kastens K, Mascle J, Auroux C, et al. (1988) ODP Leg 107 in the Tyrrhenian Sea: insights into passive margin and back arc basin evolution. Geological Society of America Bulletin 100: 1140 1156. Re´ hault JP, Mascle J, and Boillot G (1984) Evolution ge´ odynamique de la Me´ diterrane´ e depuis l’Oligoce`ne. Memorie Societa` Geologica Italiana 27: 85 96. Robertson AHF and Grasso M (1995) Overview of the Late Tertiary Recent tectonic and palaeo environmental de velopment of the Mediterranean region. Terra Nova 7: 114 127. Stampfli G, Borel G, Cavazza W, Mosar J, and Ziegler PA (2001) The Paleotectonic Atlas of the PeriTethyan Domain. CD ROM. European Geophysical Society. Stanley DJ and Wezel FC (eds.) (1985) Geological Evolution of the Mediterranean Basin. New York, USA: Springer Verlag. Vai GB and Martini P (eds.) (2001) Anatomy of an Orogen: the Apennines and Adjacent Mediterranean Basins. Dordrecht: Kluwer Academic Publishers. Wilson M and Bianchini G (1999) Tertiary Quaternary magmatism within the Mediterranean and surrounding regions. In: Durand B, Jolivet J, Horva´th F, and Se´ ranne M (eds.) The Mediterranean Basins: Tertiary Extension Within The Alpine Orogen, pp. 141 168. Special Publi cation 156. London: Geological Society.
EUROPE/Holocene 147
Holocene W Lemke and J Harff, Baltic Sea Research Institute Warnemu¨nde, Rostock, Germany ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The Quaternary period comprises the shortest time interval of all geological systems. Compared to the preceding climatically stable and warm Tertiary, it is characterised by a multiple alternation of large-scale glaciations and short warm intervals in between. The latest interglacial period, which is still ongoing, is called Holocene after the Greek words ‘holos’ (entire) and ‘ceno’ (new). According to recent understanding, it began ca. 11 600 calendar years before present. In contrast to other epochs of the Earth’s history, it is not defined and subdivided by certain floral or faunal assemblages but by climatic features. Another basic difference from former geological periods is the increasing human impact on the geosphere. In fact, some authors claim that the properties of the ‘system Earth’ have changed by human influence to an extent that it cannot be called natural anymore. Reconstructing the geological past, therefore, requires consideration of natural processes, as well as the results of human activity and to separate them from each other. Thus, Holocene geology is intensely interrelated not only with other natural sciences but also with human history, archaeology, and further social sciences. This adds a wealth of additional information to the data stored within geological archives. On the other hand, geological problems during the Holocene are not only a matter of actualism in the classical sense anymore. Due to the increasingly closer connection between geological processes and the development of the human society, forecasting of geological trends becomes more and more important. In this way, Charles Lyell’s (see Famous Geologists: Lyell) statement about the principle of actualism could be extended to: ‘‘The knowledge about present and past is the key to the future.’’
Dating When aiming for an accurate reconstruction of the geological past, dating becomes an essential issue. Looking back from recent times to the near past, dating of geological events is simply done on a high resolution by analysing the written historical archives. Further back in time, indirect methods (by using socalled proxy data) have to be used. Proxy data with a
yearly resolution are related to processes which result in persistent and regularly successive yearly structures within sediments (e.g., varve sequences) or organic material like wood (dendrochronology). The latter is based on the study of tree ring patterns which are controlled mainly by climatic factors. In Europe it was used particularly for oaks in central and western Europe and for pines in northern Europe. Regionally generalised curves for these two tree species cover nearly all of the European Holocene. Dendrochronological dates are highly valuable for calibrating dating results produced by other methods. This refers especially to isotopic dating by radiocarbon, which is widely used as a standard method for the dating of organic material within the Holocene. By comparing dendrochronological or varve counting dates with radiocarbon dating, inconsistencies within the later ones, particularly within the early Holocene, became obvious. Therefore, when looking at dates in the literature, it is crucial to consider if calibrated (calendar) years or radiocarbon years are referred to. Within this article calendar years before present (BP) are used (except for Figure 10). Other short-lived isotopes, such as 210Pb are used to date processes and events in the more recent past on time-scales of decades and centuries. Once the environmental history of a specific region is well known, assemblages of plants or animals might also help to assess the age of the deposits they are found in.
Climate One of the most intensively studied subjects of Holocene development is climate. Ice and marine sediment cores have been used to assess climatic changes and they provide smoothed background data to more regional, or local and mostly more dramatic, climatic variations on the European continent. The onset of the Holocene is marked by a global drastic temperature increase of about 7 C at the end of the Younger Dryas, about 11 600 calendar years BP. This climate reorganisation happened during a period of not more than a few decades. Since then, the Holocene climate has been stable by comparison with the preceding glacial period. Nevertheless, minor climate fluctuations have been reconstructed. Several periods with cooler and warmer temperatures than the last century have left their traces in the geological and biological archives and also in human history (Figure 1).
148 EUROPE/Holocene
Figure 1 Mean Holocene annual temperatures. The horizontal axis is time before present (BP) in thousands of years (Ka). (Adapted from Schonwiese C (1995) Klimaa¨nderungen Daten, Analysen, Prognosen. Berlin: Springer Verlag, with permission.)
During the first thousand years of the Holocene, the climate was possibly slightly cooler than today. Between 9000 and 8200 years BP, climatic conditions were slightly warmer and often moister than today. Cooler conditions throughout Europe have been interpreted from different proxy data for the period around 8200 years BP, when annual mean temperatures dropped by 2 C in central Europe and the Alpine timberline fell by about 200 m. This sudden cool phase lasted for about 200 years and wetterthan-present conditions in north-west Europe have been inferred. Warmer temperatures prevailed in Europe during the so-called climatic optimum in the Early Holocene (8000-4500 years BP). The Early Holocene climatic optimum was characterised by warmer summers than today in Europe. For astronomical reasons, the northern hemisphere received nearly 8% more solar radiation during summers than in recent times. A northward shift of the inner tropical convergence zone (ITCZ) forced monsoonal rainfall as far north as the Mediterranean Basin. At around 5900 years BP, a short cold episode interrupted this warm phase. Since 4500 years BP, the climatic conditions have fluctuated around a situation comparable to the recent one. Remarkable deviations occurred at about 3300 years BP, when intensified glaciation started in the Alps. About 2000 years ago BP, favourable climatic conditions promoted the development of the Roman Empire, while the Germanic migrations after its collapse went along with a cooler climate. In mediaeval times (ca. 1300–700 years BP), another warm period allowed the expansion of Scandinavian Vikings as far
as Greenland and North America. Clear indications of warm temperatures during this time interval were also reported from northern Russia, central Europe, and the Mediterranean. The youngest climatic deterioration, known as the Little Ice Age, at about 700–150 years BP, destroyed the agricultural economic basis of the Norse settlers in Greenland, and by about 500 years BP, their population in Greenland had vanished. In the mid-seventeenth century, glaciers in the Swiss Alps advanced and rivers in England and the Netherlands often froze over during the winter. Severe cold winters have been deduced from borehole data in the Czech Republic for the time slice between 300 and 400 years BP, too. Since the mid-nineteenth century, the global temperatures have risen (Figure 1) again, a process which is still going on. A general periodicity of 200 to 600 years for the whole Holocene climate can be inferred from various proxies. External processes including solar activity cycles and internal driving forces as volcanic eruptions are under debate as controlling factors of climate variability up until now. To what extent this climatic cyclicity is modified by human activity (e.g., extensive release of greenhouse gases) is a matter of current scientific discussion.
Naturally Changing Holocene Landscapes in Europe The deglaciation processes, which had started within the Late Pleistocene, were accelerated in the Early Holocene. Within the first two thousand years of the
EUROPE/Holocene 149
Holocene, the former prevailing tundra and steppe habitats (Figure 2) were nearly completely replaced by mixed deciduous forest (Figure 3). The first phase of tree invasion was characterised by birch and pine and later by hazel and elm. The local tree assemblages could differ considerably from each other and also from recent compositions. At around 10 000 years BP, in many parts of Europe the forest cover was still rather more open than at present, with more herbaceous glades. By 9000 years BP, the forest had become closed, but with conifers more abundant
Figure 2 Vegetation zones in Europe during the Younger Dryas. (A Ice; B Polar Desert; C Steppe Tundra; D Dry Steppe; E Semi Desert; F Wooded Steppe. (Adapted from Adams 2002, with permission.)
Figure 3 Vegetation zones in Europe about 9000 years BP (A Ice; B Tundra; C Open boreal woodland; D Boreal Forest; E Deciduous/Mixed Forest; F Forest Steppe; G Moist Steppe; H Woodland; K Dry Steppe; L Mediterranean Forest. (Adapted from Adams 2002, with permission.)
than at present in eastern Europe. Until that time, deciduous trees such as oaks or hornbeam were predominant or abundant, even in southern Europe (Figure 3). The typical recent Mediterranean vegetation with evergreen trees and shrubs started to develop after that time. During the climatic optimum, thermophile plants and animals extended further to the north. Possibly due to elm disease, a drastic decline in the number of elm pollen is observed in the geological records throughout Europe at about 5800 years BP. Since 4500 years BP, there has been an increasing human influence on the faunal and floral elements of the European ecosystem. Figure 4, showing the potential present vegetation without human influence, is obviously different from the coverage conditions today. Another result of the changing Holocene climate was a rising global sea-level by meltwater supply and thermal expansion of sea-water (eustatic sea-level rise). During the maximum Weichselian glaciation, the global sea-level was about 125 metres deeper than today. In the Early Holocene, large deglaciated areas started to uplift because of the vanishing ice load (glacio-isostasy). At the centre of the last glaciation, around the Bothnian Bay, an uplift of more than 280 m is recorded within Holocene sediments. This isostatic uplift was compensated by subsidence in more distal regions (Figure 5) within the southern Baltic Basin. By some authors, this process is assumed to be in the context of the collapse of an asthenospheric bulge in front of the retreating Weichselian ice shield (Figure 6). The combination of eustatic
Figure 4 Potential present vegetation zones in Europe (A Ice; B Tundra; C Open boreal woodland; D Boreal Forest; E Deciduous/Mixed Forest; F Forest Steppe; G Moist Steppe; H Woodland/Wooded Steppe; K Dry Steppe; L Mediter ranean Forest. (Adapted from Adams 2002, with permission.)
150 EUROPE/Holocene
Figure 5 Recent vertical movement of the Earth’s crust. In the northern part the map is dominated by the extensive, north east south west stretched uplift of Fennoscandia with maximum values of >8 mm year. The subsidence of a belt surrounding the Baltic Shield is less differentiated. (From Harff J, Frischbutter A, Lampe R, and Meyer M (2001) Sea level change in the Baltic Sea: interrelation of climatic and geological processes. In: Gerhard LC, Harrison WE, and Hanson BM (eds.) Geological perspectives of global climate change. Tulsa, Oklahoma, American Association of Petroleum Geologists in collaboration with the Kansas Geological Survey and the AAPG Division of Environmental Geosciences: 231 250. Reprinted by permission of the AAPG whose permission is required for further use.)
sea-level changes and isostatic uplift or subsidence, partly modified by tectonic movements, produced considerable changes in the geography of Europe during the Holocene. Depending on the geographic position of the affected area, large relative sea-level changes (positive or negative ones) have occurred (Figure 7). These changes are particularly obvious where large intracontinental basins like the recent Baltic Sea area were affected. At the beginning of the Holocene, large parts of the Baltic Basin were filled with freshwater from the Baltic Ice Lake which was fed mainly by meltwater from a large glaciated area in North and north-eastern Europe. ¨ resound (between The only important outlet in the O the recent Danish island Sealand and southern Sweden) was too narrow to serve as a sufficient spillway between the Baltic Ice Lake and the North Sea. The global sea-level was about 25 metres lower than in the Baltic Ice Lake. When the Scandinavian inland ice started
to retreat from southern Sweden, a spillway through the central Swedish Depression was opened. As a dramatic process, half of the recent Baltic Sea’s water volume drained into the Atlantic Ocean via the Kattegat and North Sea. This drainage took no longer than a few years and had an enormous impact along the former shores of the Baltic Ice Lake. Large areas previously covered by water became dry land, and southern Scandinavia became directly connected to central Europe. Saline waters of the Kattegat could enter the Baltic Basin for a time-span of a few hundred years, a stage of the Baltic Sea’s development known as the Yoldia Sea, a phase which is dated from 11 570 to 10 700 years BP (Figure 8). The connection between the Yoldia Sea and the Kattegat through central Sweden was located in a rapidly uplifting region. Therefore, the connection closed at about 10 700 years BP and a newly dammed-up freshwater lake was formed within the Baltic Basin. It
EUROPE/Holocene 151
Figure 6 Principle of glacio isostatic vertical crustal move ments: (a) Ice load causes subsidence of the Earth’s crust below and compensatory uplift beyond the ice margin as a fore bulge; (b) Uplift of the ice released Earth’s crust and related displacement of the forebulge. (Modified after Daly 1934.)
Figure 7 Schematic Holocene relative sea level curves for different European regions (blue = Oslo, Norway; yellow = south eastern Sweden; light blue = southern Baltic Sea; brown = French Atlantic coast; green = Netherlands; purple = central Mediterranean).
is called Ancylus Lake, with reference to a typical freshwater snail occurring in these waters (Ancylus fluviatilis) (Figure 9). This freshwater stage lasted more than two thousand years and was characterised
by considerable short-term water level fluctuations. As differential isostatic uplift continued, the critical thresholds between the Baltic Basin and Kattegat moved from southern Sweden to the recent-formed Danish Straits. At about 8000 years BP, the eustatic sea-level rise led to the first ingressions of marine waters into the Baltic Basin. In pace with the rapidly rising global sea-level, the thresholds were flooded and a stable connection between the Kattegat and the Baltic Basin was formed. This crucial phase of the Baltic Sea’s evolution is called the Littorina transgression, after a marine snail which is common in deposits of this period (Littorina littorea). At the onset of the Littorina transgression, the water level rose at a rate of 25 mm year within the south-western Baltic Basin (eustatic rise added to crustal subsidence, as described above), which slowed down later to about 3 mm year at about 4500 years BP (Figure 10). During a time-span of less than a thousand years, the sea-level rose by more than 20 m, implying enormous rates of coastal retreat within the southern Baltic Basin. Due to the rapidly rising sea-level at the beginning of the Littorina Stage, the glaciogenically-shaped land relief was drowned without any notable coastal erosion and longshore transport processes of sedimentary material. The resulting geographical situation is shown in Figure 11. Only after the sea-level rise slowed down at the end of the Littorina Stage (about 2000 years BP), and during the Post-Littorina Stage, was the recent spit and barrier coast formed by erosion and sediment transport. This process was mainly controlled by climatic factors, such as the wind-driven hydrographic regime which was superimposed on long-term eustatic and isostatic movements which have caused rising relative sea-levels and coastal retreat at the southern Baltic shores in recent times. Further north in Fennoscandia, isostatic uplift continuously exceeded the eustatic sea-level rise, resulting in a permanent general sea regression (Figure 7). The changing Holocene sea level within the Mediterranean basins was and is mainly controlled by eustatic processes. In contrast to northern Europe, glacioisostasy does not play a significant role. On the other hand, this region occupies the junction between the African-Arabian and the Eurasian plates which gives considerable tectonic activity in the different sedimentary basins of the Mediterranean Sea. Therefore, the general picture of the sea-level, development which reflects the eustatic curves, is superimposed on the regional and local tectonics. A matter of ongoing discussion is the possibility of the reconnection of the Black Sea and the Mediterranean Sea during the period of Holocene sea-level rise. A catastrophic flood scenario at about 7500 years BP
152 EUROPE/Holocene
Figure 8 North western Europe during the Yoldia Sea stage of the Baltic Sea’s history (about 11 000 years BP). (From Andersen BG and Borns HW (1994) The ice age world: an introduction to Quaternary history and research with emphasis on North America and Europe during the last 2.5 million years. Oslo: Scandinavian University Press, with permission.)
is questioned because of indications of a more complex and progressive transgression process over the past 12 000 years.
Interrelation of Human Activity and Natural Environment In terms of human history, the onset of the Holocene is equivalent to the beginning of the Mesolithic period, which is characterised by prevailing hunter, fisher, and gatherer societies in Europe. The temperature rise at the end of the Younger Dryas caused an accelerated deglaciation. Alpine glaciers retreated to historical dimensions and the receding inland ice in Scandinavia was followed successively by tundra, steppe, and finally forests. This was accompanied by an enhanced northward migration of animals and their Mesolithic hunters. The rapidly changing natural environment, possibly in combination with human activity, resulted in the extinction of some characteristic genera of the Pleistocene megafauna such as the mammoth (see Tertiary To Present: Pleistocene and The Ice Age). Another consequence of the rapidly changing landscapes was a very variable
migration pattern of the Mesolithic hunter, fisher, and gatherer groups in central and north-western Europe. The following Neolithic period was closely connected with the introduction of farming in the various geographical regions. The first Neolithic settlements in southern Europe (Greece) are dated to about 9000 calendar years BP. In northern Germany, southern Scandinavia and the British Isles, they are more than 3000 years younger. Coming from the Near East, the new method of food-procurement spread to the Great Hungarian Plain in a first wave from 8200–7800 years BP. A second leap entered the North European Plain at around 7400 years BP. Neolithic settlers from Southeast Europe migrating along the rivers Danube and Rhine were probably responsible for the consequent social and cultural changes. In parts of Europe (e.g., western Mediterranean and northern Europe) the native Mesolithic population adopted agricultural methods to form a transitional economy. While the hunter-fisher-gatherer societies of the Mesolithic used and manipulated the natural ecosystem without altering it considerably, the Neolithic farmers started to transform the environment according to their
EUROPE/Holocene 153
Figure 9 North western Europe during the Ancylus Lake stage of the Baltic Sea’s history (about 9500 years BP). (From Andersen BG and Borns HW (1994) The ice age world: an introduction to Quaternary history and research with emphasis on North America and Europe during the last 2.5 million years. Oslo: Scandinavian University Press, with permission.)
needs. Cultivation started within rather small (perhaps pre-existing) open land patches in the Early Neolithic. In this phase clearances remained localised and the early Neolithic farmers were still highly dependent on intact natural habitats surrounding the cultivated land patches. This situation changed in the later part of the European Holocene, when increasing demand for agricultural products induced a major transformation of the yet still mainly natural environment into an agricultural one. This process was certainly time-transgressive and affected some European regions more than others. But, far from being controlled solely by human action, natural feedback, combined with differing vulnerability of the existing ecosystems, amplified, shifted, or interfered with the initial direction of processes initiated by human activity. In order to adapt to the partly self-induced new conditions, human societies had to react by further cultural development which accelerated the general transformation process of the natural environment. This is exemplified by the domestication of animals which were advantaged by the anthropogenically influenced environment, while on the other hand natural competitors and predators became progressively extinct by hunting kill-off or simply by loss of habitat (extensive land use by farming). The same applies to floral
assemblages, as initially, their composition was governed by natural conditions but it became increasingly influenced by agriculture. This development finally produced an increasing interdependence of cultivated plants and animals with mankind. Societies like the Irish people in nineteenth century, for example, were greatly dependent on potato growth. When the Late Potato Blight ruined all the potato crop in Ireland in the 1840s, the resulting famine lead to dramatic consequences. About 1 million people died while another 1.5 million people emigrated. The stability of the cultural landscape which had evolved out of the interplay between natural and human influences, as well as that of human societies living there, depended on the state of their equilibrium. Changes of the natural component could cause considerable impact to the progressively complex human society and this is exemplified by climatic or geological influences. Some historical epochs with a prospering economy and politically stable conditions are connected with warmer periods. Colder climate deviations often were characterised by political and economical instability. This refers, for example, to the time of the Germanic migrations about 450–700 ad and the Little Ice Age between 1500 and 1800 ad. Rather short-term, but possibly catastrophic, impacts derived from events like
154 EUROPE/Holocene
Figure 10 Preliminary curve of relative sea level change for western Pomerania, shaded area: interval of ‘data confidence’, solid line: local trend for western Pomerania, dashed line: estimated trend, for the time span between 6000 and 4500 years BP tectonically controlled local uplift is assumed. Note: Ages are given here in 14C years BP. (From Harff J, Frischbutter A, Lampe R, and Meyer M (2001) Sea level change in the Baltic Sea: interrelation of climatic and geological processes. In: Gerhard LC, Harrison WE, and Hanson BM (eds.) Geological perspectives of global climate change. Tulsa, Oklahoma, American Association of Petroleum Geologists in collabor ation with the Kansas Geological Survey and the AAPG Division of Environmental Geosciences: 231 250. Reprinted by permission of the AAPG whose permission is required for further use.)
volcanism (e.g., the eruption of Vesuvius in ad 79 described by Pliny the Younger), earthquakes (e.g., the disastrous one of Lisboa in ad 1755) or floods along rivers and seashores. On the other hand, changing social conditions also induced dramatic changes of their natural environment. Once human activity stopped or declined, the surrounding environment developed depending on its natural stability. Robust ecosystems changed back to a state similar to the original one being controlled by the natural conditions. An example for such processes is the re-forestation after depopulation during the Thirty Years’ War (1618–1648 ad) in central Europe. In other cases, human activity led to irreversible effects on the ecosystem’s stability when the social structures collapsed. During the Roman period there was a land use maximum partly on metastable soils. Here, terraces were maintained, preventing soil
erosion. After the invasion of eastern nomads, the land was partly abandoned and soil erosion started at a greater extent and in the worst case, barren badlands were the final result. Therefore, maximum erosion is not connected with maximum land use, but with subsequent phases in different cultural environments. Generally, when judging human impact on the natural environment, it is often regarded only as negative. However, in contrast, the creation of new metastable ecosystems more diverse than before, particularly in the early phases of the Holocene, would not have been possible without human activity.
Human Activity and Environmental Conservation During the last 500 years, human impact on the European environment have become much more important than the natural conditions. Particularly
EUROPE/Holocene 155
Figure 11 Europe about 4500 years BP. (From Andersen BG and Borns HW (1994) The ice age world: an introduction to Quaternary history and research with emphasis on North America and Europe during the last 2.5 million years. Oslo: Scandinavian University Press, with permission.)
within the last hundred years, industrial methods in agriculture have resulted in a widespread conversion of natural habitats into farming land. Accelerated human-induced soil loss has become a major problem. Measures initiated to solve the problems of modern agriculture, for example, to protect the soil against nutrient depletion, partly result in adverse effects such as eutrophication. One of the most severe impacts on environment is connected with the beginning of industrialisation. Natural resources which had accumulated over millions of years have been exploited within decades or less. Large industrial facilities have been built on terrain of formerly less affected landscapes. Moreover, military needs have modified great parts of the terrain. Growing populations all over Europe have enhanced the conversion of natural habitats into settlement areas. At the same time, the management of industrial and municipal waste has become a major issue. The effects of industrialisation are reflected by many proxies such as the concentration of heavy metals in sub-recent deposits (Figure 12). All these tendencies have developed ideas on nature conservation, sometimes with the idealistic approach of going back to a state where human influence is negligible. Nature conservation in this context may
be regarded as a contradiction in itself because it does not recognise the vital role of human society for the natural environment. Large areas along the Netherlands’s coast would have been flooded if the coast was not protected by coastal engineering. Furthermore, the dynamic character of the environmental status must be considered. For the last 8000 years of the Baltic Sea’s history it can be shown, for example, that high nutrient levels already existed immediately after the Littorina Transgression. Without any remarkable human influence, organic substances and nutrients could accumulate in the sediments because of restricted vertical convection. This process has been intensified during the last centuries and decades by agriculture and the industrial release of nutrients (Figure 13). Thus, if the Helsinki Commission for the protection of the Baltic Sea aims to restore the eutrophication level of the 1950s, it must be stated that that was just one time slice of the Baltic Sea’s Holocene development. One might be successful in reestablishing the concentrations of certain nutrients, but it would be in a completely different new context. As human society develops further, it cannot be expected that there will be no response by the natural environment. A certain equilibrium, including the benefit for as many species as possible including
156 EUROPE/Holocene
Figure 12 Enrichment of heavy metals within the last century, as recorded in a sediment core from the Arkona Basin (western Baltic Sea). (From Th. Leipe, Baltic Sea Research Institute Warnemunde, with permission.)
Figure 13 Trends of winter phosphate concentrations in the surface layer (0 10 m) of the eastern Gotland Basin (central Baltic Sea). (From G. Nausch, Baltic Sea Research Institute Warnemunde, with permission.)
EUROPE/Holocene 157
Table 1 Summary of the climatic and historical development during the European Holocene Calendar years BP
Environmental Period
200
Subatlantic
General climatic features
Selected historical events
Stages of the Baltic Sea’s development
Modern climatic optimum, warm
World War I and II
Mya Sea
French Revolution Little Ice Age, mean annual temperature in Europe 1 C lower than today, cold winters, but pronounced fluctuations, glacier advances 400
Less brackish
Thirty Years’ War Renaissance and reformation
Mya Sea Lymnaea Sea Brackish
600 Bubonic plague kills about one third of the European population Transition to colder conditions 800 Brackish
Medieval climatic optimum, mean annual temperature 1 1.5 C higher than today, winegrowing as far north as to the British Isles, first dry, later wet Subatlantic 1000 Expansion of Norman people as far as Iceland, Greenland, North America and southern Russia 1200 Cold and wet period, many glacier advances 1400 Subatlantic
End of the Roman Empire
Slightly brackish
Invasion of the Huns forces the emigration of nations
Lymnaea Sea
1600
1800 2000
Roman climatic optimum, as warm as the Medieval optimum, mostly very wet, towards the end more dry
Foundation of the Roman Empire
First southward migrations of Gothic tribes 2200 Pronounced cold period, mean annual temperature 1 1.5 C lower than today, very cool summers, very wet, large glacier advances
Celtic La Te`ne culture in big parts of Europe
2400
2600 Subatlantic Subboreal 2800 3000
Beginning of the Greek classic period Formation of the Roman Republic Celtic Hallstatt culture in central and western Europe introduces the iron age Greek Archaic period First Celtic tribes in eastern and central Europe Brackish
3200 (Continued)
158 EUROPE/Holocene
Table 1 Continued Calendar years BP
Environmental Period
General climatic features
Selected historical events
Stages of the Baltic Sea’s development
Predominantly warm with Distinct fluctuations, less precipitation than during Subatlantic Urn field culture in central and South eastern Europe 3400 First bloom of the Mycene culture in Greece 3600 Beginning of the Bronze Age in northern Europe 3800
Subboreal Cold period with glacier advances, initially dry, later more wet
4000 4200
In Europe exists an extensive trade route network
4400 Early Minoic culture at Crete Island launches the European Bronze Age 4600 4800 5000
Increasing influence of Indo European people all over Europe Introduction of the wheel in Europe Megalithic monuments in many regions of central and western Europe
5200 5400 5600 5800 6000 6200
Brackish
Subboreal Atlantic Rapid increase of humidity in eastern central Europe
6400
Narva culture in North eastern Europe First stone buildings at the Orkney Islands First Neolithic settlements at the British Isles
Warm period, mean annual temperatures 2 3 C warmer than today, especially warm winters, very moist, former predominant pines are replaced in the forests by oaks, lime and hazel 6600 Late Mesolithic Ertebølle culture in northern central Europe 6800 7000 7200 7400
Neolithic Karanovo culture in South eastern Europe Late Mesolithic Ertebolle culture in northern Europe Neolithic ‘Bandkeramik’ culture in the Loess areas of central Europe
7600 7800
Strongly brackish Mesolithic Kongemose culture in northern Europe
Littorina Sea
8000 (Continued)
EUROPE/Holocene 159
Table 1 Continued Calendar years BP
Environmental Period
General climatic features
Selected historical events
Ancylus Lake
Short lived cold interval, drop of the mean annual temperature by 2 C 8200
Stages of the Baltic Sea’s development
First Neolithic agricultural societies in South eastern Europe and Greece
Freshwater
8400
8600 8800 9000
Slightly warmer and moister than today Atlantic Boreal
The English Channel separates the British Isles from the continent
Hunter and gatherer societies Cold phase in the Carpathian Basin
9200 During summers generally warmer than today, mainly open winters
Maglemose culture in northern Europe
9400 Azilian culture in western Europe 9600 Microliths become common 9800 Use of log boats is proven 10000 10200
Boreal Preboreal
Nomadic hunters arrive in England
10400 Ahrensburg culture in northern Germany Freshwater 10600 Summers as warm as today, but very cold winters 10800
Ancylus Lake Yoldia Sea Freshwater Regionally brackish
11000 11200 11400 11600 11800 12000
Preboreal Younger Dryas
Quick warming
Bromme culture in Denmark
Beginning of the Holocene End of the Pleistocene
Mesolithic Palaeolithic
Cold period, mean annual temperatures by 5 9 C lower than today
mankind is desirable. Closed production cycles might be one of the important targets to achieve this.
Actualism in a New Context The special character of the Holocene as a period which is not only part of the geological past but also an interface with the future, gives the principle of actualism a new dimension. In addition to regarding
Freshwater Yoldia Sea Baltic Ice Lake Freshwater
the present as being the key to the past, past and present times might be regarded as a key to the future. Predictions of future developments become more and more important to cope with possible changes in the natural environment. For this purpose, detailed knowledge about similar processes in the past is indispensable. In order to calibrate proxies from the geological record, it is necessary to analyse recent proxies by comparison with older ones, and also the
160 EVOLUTION
written record in order to separate historical trends. Improvements in dating methods, and additional information from geological, archaeological, biological, historical and other sources will help to develop scenarios which might help the recognition and response to future challenges.
See Also Engineering Geology: Natural and Anthropogenic Geohazards. Famous Geologists: Lyell. Fossil Vertebrates: Hominids. Tertiary To Present: Pleistocene and The Ice Age.
Further Reading Adams, J Europe during the last 150 000 years [online at http://www.esd.ornl.gov/projects/qen/nercEurope.html] Andersen BG and Borns HW (1994) The ice age world: an introduction to Quaternary history and research with emphasis on North America and Europe during the last 2.5 million years. Oslo: Scandinavian University Press. Bjo¨ rck S (1995) A review of the history of the Baltic Sea, 13.0 8.0 ka BP. Quaternary International 27: 19 40. Cunlifffe B (ed.) (1994) The Oxford Illustrated Prehistory of Europe. Oxford New York: Oxford University Press.
Donner J (1995) The Quaternary history of Scandinavia. Cambridge: Cambridge University Press. Emeis K C and Dawson AG (2003) Holocene palaeoclimate records over Europe and the North Atlantic: modelling and field studies. The Holocene 13: 305 464. Grove JM (1988) The Little Ice Age. London, New York: Routledge. Harff J, Frischbutter A, Lampe R, and Meyer M (2001) Sea level change in the Baltic Sea: interrelation of climatic and geological processes. In: Gerhard LC, Harrison WE, and Hanson BM (eds.) Geological perspectives of global climate change. Tulsa, Oklahoma, American Association of Petroleum Geologists in collaboration with the Kansas Geological Survey and the AAPG Division of Environ mental Geosciences: 231 250. Litt T, et al. (2003) Environmental response to climate and human impact in central Europe during the last 15 000 years a German contribution to PAGES PEPIII. Quater nary Science Reviews 22: 1 124. Pirazzoli PA (1991) World atlas of Holocene sea level changes. Elsevier Oceanography Series 58, Amsterdam, London, New York, Tokyo: Elsevier Science Publishers B.V. Roberts N (1998) The Holocene An environmental his tory. Oxford: Blackwell Publishers Ltd. Scho¨ nwiese C (1995) Klimaa¨ nderungen Daten, Analysen, Prognosen. Berlin: Springer Verlag.
EVOLUTION S Rigby, University of Edinburgh, Edinburgh, UK E M Harper, University of Cambridge, Cambridge, UK ß 2005, Elsevier Ltd. All Rights Reserved.
evolution also includes the study of patterns of diversification and extinction. Macroevolution may be the end result of microevolution working over a long time-scale or it may be a suite of emergent properties that require unique interpretations.
Introduction The theory of evolution by natural selection, put forward by Darwin in 1859 (see Famous Geologists: Darwin), is the greatest unifying theory of biology and palaeontology. In this context, evolution is the change that occurs between successive populations of organisms, due to their modification in response to selection pressures. The potential to change is provided by genetic variability within populations and by genetic change through time (mutation). The pressure for change to occur exists outside an organism and is provided by interactions within the environment. These interactions may be predominantly physical or biological effects. Small-scale changes in populations, giving rise to new species, are defined as microevolution. Larger-scale changes, such as the origin of new higher taxa – which may have new body plans or new organs – are defined as macroevolution. The study of
Historical Background The presence of large numbers of species on the Earth and the means by which they appeared were discussed throughout the Enlightenment, though the use of the word evolution did not become common until the twentieth century. The possibility that one species might change into another was of interest to Charles Darwin’s grandfather, Erasmus Darwin, for example. Studies that focused on the ways in which species transform were begun by Jean-Baptiste Lamarck and published in his Philosophie Zoologique in 1809. Lamarck argued that an ‘internal force’ caused offspring to differ slightly from their parents and also that acquired characters could be passed on to the next generation. One of his examples was that of giraffes, whose long necks were assumed to be a
EVOLUTION 161
product of successive generations reaching for higher and higher leaves. He suggested that each giraffe lengthened its neck slightly by this activity and, in turn, passed on to its descendents the capacity to grow longer necks. He visualized species as forming a chain of being, from simplest to most complicated, with each species being capable of transforming into the next in line, and all existing indefinitely. In Britain this work was disseminated by both Richard Owen, who was generally supportive of the theory, and Charles Lyell (see Famous Geologists: Lyell), who was critical of it. Charles Darwin encountered work by both of these scholars and also explored huge tracts of the natural world during his 5 years study on the Beagle (1831–1836). His work on a number of organisms, notably finches collected from the Galapagos Islands in the Pacific, persuaded him that organisms were adapted to their particular niche and that species were capable of change. The process by which this change could occur was a preoccupation of Darwin’s in the succeeding years. As early as 1838, he had read the seminal work of Malthus on populations, but he was still working on the scope and implications of his theory when he was forced to publish by correspondence from Alfred Russel Wallace. A joint paper presented to the Linnaean Society in 1858 was followed the next year by his classic work On the Origin of Species. Darwin’s theory of species originating through natural selection can be set out in a small number of propositions. First, organisms produce more offspring than are able to survive and reproduce. Second, successful organisms – those that survive long enough to breed themselves – are usually those that are best adapted to the environment in which they live. Third, the characters of these parents appear in their offspring. Fourth, the repetition of this process over a long time-scale and many generations will produce new species from older ones. The consequences of this theory are enormous. Not least, they caused scientists at the time to reconsider their assumption of a chain of life. Evolution by natural selection is a response to the local environment and is not predetermined on a grand scale. Organisms do not necessarily evolve into more complicated species over time. Amongst the general public, the theory was seen as being in conflict with a literal reading of the Bible, a view that persists amongst a religiously conservative minority. In the years after publication, the most significant weakness of Darwin’s theory was perceived to be its failure to supply a plausible mechanism for the inheritance of characters. However, this mechanism was supplied when Gregor Mendel’s (1865) work on
heredity was rediscovered in the early twentieth century. Mendel observed that characters were passed from parent to child in a predictable fashion depending on the relative dominance of the traits carried by each sexual partner. Characters did not ‘blend’ in the offspring, which is what Darwin had suggested and which astute critics had pointed out would actually have prevented evolution from occurring. These observations opened the door to the modern study of genetics. After some decades of debate, a modern consensus was reached in the 1940s, which is the basis for our current understanding of Darwin’s ideas.
Evolution and Genetics: The Living Record Evolution is possible because the genetic transmission of information from parent to offspring works as it does, in a Mendelian fashion. Subsequent work on genetics has elucidated the exact means by which this occurs and has shown how variation can be developed and sustained in a population. The information that can be passed from one generation to the next in a population is contained on strands of DNA (deoxyribonucleic acid), or occasionally RNA (ribonucleic acid), within each cell. A DNA molecule forms from a series of nucleotides, which are joined up like beads on a string. Each nucleotide has, as one of its elements, a base. The four types of base DNA are adenine, thymine, guanine, and cytosine (usually abbreviated to A, T, G, and C). Two strings of nucleotides join via base pairs to make the double-helix shape of DNA. A always joins to T, and C always joins to G. Sequences of bases are the code that stores the information needed to produce an organism. This includes information about making the various parts of the cell or set of cells and also information about the rates at which different processes should occur and their relative timings. Each piece of information that the DNA holds is called a gene. Genes can be sequences of DNA or can be little pieces of DNA separated by other sets of bases. Most of the DNA appears to have no purpose and is called non-coding DNA. A human is produced from about 30 000 genes that use about 5% of the nucleotides of our DNA (Figure 1). When sexual reproduction occurs, one copy of the DNA (carried on chromosomes) of each parent is passed to the children. The offspring therefore have two sets of instructions within their DNA. The pair of genes that share a common function are called alleles, and the combination of alleles controls the effect on the bearer. However, this effect will not be passed to
162 EVOLUTION Figure 1 A diagrammatic representation of DNA, genes, and chromosomes. (A) The molecular structure of a double strand of DNA. Each strand is made up of a chain of sugars (yellow) and phosphates (purple), linked together by a set of four bases: thymine (orange), adenine (green), guanine (blue), and cytosine (red). The shapes of these bases cause adenine to bond to thymine and guanine to bond to cytosine. This makes each strand of DNA a mirror image of the other. (B) A piece of DNA carries information in the form of sets of bases, in this example GGTCTGAAC. (C) A gene is a set of useful bits of DNA, which code for a particular protein or carry out a particular instruction. Genes may be formed in pieces separated by long intervals. (D) A chromosome is a folded up cluster of DNA found in the nuclei of eukaryotic cells.
EVOLUTION 163
the next generation, but rather one of the alleles will. This will then combine with another allele to generate another product. Although the results of allele combination can have a complicated range of expressions in a cell or a body, the alleles don’t mix, so variation is maintained. A wide range of errors can occur when the DNA strand is replicated during reproduction. These can affect the non-coding DNA or the genes and can produce mutations of varying effect depending on whether it is genes that control the production of the body or the timing or duration of elements of this production process that are affected.
Time and Narrative: The Fossil Record Biologists have explored theories of evolution in tandem with palaeontologists, who can retrieve narratives of evolutionary change from the fossil record. The ability to study change over millions of years is a great advantage of using fossils. Theoretically, it should be possible to study aspects of the morphologies of fossils collected bed-by-bed throughout a rock sequence in order to elucidate patterns of evolution. However, the preservation of individual fossils is often poor; most depositional events produce significant time-averaging, and the fidelity of long records of sedimentary sequences is often questionable. At some scale, all deposition is intermittent, and this means that there are gaps of some scale in all narratives retrieved from the fossil record. Fossils preserved in lakes or deep-sea cores may be less affected by this problem than fossils from more dynamic environments, and research has generally concentrated on these locations.
Microevolution The set of potentially interbreeding members of a population forms a species, which contains a range of variation in its appearance (the phenotype) and in its genetic codes (the genotype). In practice, most living species are defined on the basis of phenotypic characteristics rather than genetic information or reproductive potential. In fossil studies of evolution, only the phenotypes are available, and the definition of a species must be based on clusters of phenotypic characters, which are taken as proxies for the potential to interbreed. Natural selection acts on a set of individuals, so that the physical characteristics of the group and the underlying genotypes change over time. This process eventually gives rise to new species and is known as microevolution. Biologists class only gene shifts within populations as microevolution and define
anything larger, including the appearance of new species, as macroevolution. To palaeontologists, the distinction is usually between speciation and anything higher, such as the emergence of new genera or of new organs. Sometimes a species gradually changes through time until the point comes where the fossil representatives of successive populations are recognized as a different species. However, a parent species often splits into more than one offspring species or evolves into an offspring species that coexists with the parent species for some time. In this case, the original population must split into two or more subsets that cannot interbreed with one another. The two best-known methods of achieving this are called allopatric speciation and sympatric speciation. In allopatric-speciation events a single original population is split into two geographically isolated elements. This is a common phenomenon over geological time as continents fragment, mountains rise, or sea-levels change. Each geographically separated fragment of the initial population contains only a fraction of the original genetic variation, so it may tend towards difference from the original population without any active selection, although this is now regarded as a minor component in the formation of new species. More importantly, different geographical regions will tend to produce different environmental stresses from those that were experienced before separation, leading to the selection of different successful characters in the separated populations. This eventually leads to significant changes of form in the isolated populations, which may finally produce new species. An example of allopatric speciation has been recovered from the fossil record of Plio-Pleistocene (3–0.4 Ma) radiolarians, which are siliceous planktonic protists, collected in the North Pacific. A divergence in the forms of two sister species of the genus Eucyrtidium was found to have occurred at around 1.9 Ma, following a short period when the populations had been separated from one another. During sympatric speciation the emerging species share a geographical range but may become separated over time by differences in behaviour or in resource exploitation. Adaptive pressures act differently on these populations, and different characteristics will be favoured, leading to a progressive change of form and eventually to reproductive isolation. At this point a new species will have appeared. There is some doubt about the mechanism by which species first begin to diverge without becoming geographically isolated, although the generation of new species in this way has been demonstrated for a number of types of animal. Studies on cichlid fishes in African lakes show that the most closely related species of fish
164 EVOLUTION
often live in the same lake, rather than in adjacent lakes, as might be expected if allopatric speciation had occurred (Figure 2). Although it seems intuitively obvious that populations that become physically dissimilar will eventually be unable to produce offspring, the genetic basis for this change can be demonstrated in the laboratory but not yet fully explained. The fossil record can be used as a tool to help in the understanding of evolution and the formation of new species. It may be that evolution progresses gradually for most of the time, an idea known as phyletic gradualism. The classic fossil example of this slow continuous process of morphological change is the study by Peter Sheldon of Ordovician trilobites recovered from deep-water shales in central Wales. Eight different genera of trilobite, including well-known forms such as Ogygiocarella, were found to exhibit incremental changes in rib number through the duration of one graptolite zone, which probably represents significantly less than 1 Ma. Gradual change is generally difficult to observe in the imperfect fossil record. It could be argued that in a less continuous sedimentary record (or one sampled less finely) this sequence of events would appear as a series of abrupt changes. Commonly, what is preserved is a long period where little or no change is observed followed by the abrupt appearance of a new form. The theory of punctuated equilibrium attempts to explain this phenomenon not as the product of an
imperfect fossil record but as a common pattern of evolutionary change. This is done by applying the concept of allopatric speciation to the problem. Eldridge and Gould, who developed the idea of punctuated equilibrium, argue that most species probably arise in small, geographically isolated areas and that they arise rapidly as they encounter new selection pressures. At some later time the evolved offspring species may move back into areas where it encounters its parent species and may out-compete this form. In most areas where this happens, the geological record will show one species – the parent – abruptly replaced by another – the offspring – with no intermediate steps. The chance of the isolated population being represented in the fossil record during the short period of its evolution into a new species is very slim (Figure 3). It may be that Williamson, in a study of molluscs in Plio-Pleistocene sediments from Lake Turkana, found one such rare fossil example of punctuated equilibrium. Species of gastropod and bivalve both appeared to remain static in shape for long periods of time, punctuated by brief periods when their shape changed abruptly. Although some studies seem to show a punctuatedequilibrium style of evolution, others appear to show that evolution has progressed via phyletic gradualism, and a consensus has yet to emerge regarding these theories. In practice, most evolution is probably the result of a mixture of punctuated and gradual periods of change, partly depending on the scale of
Figure 2 The difference between allopatric speciation and sympatric speciation, using the example of fishes living in lakes. (A) Sympatric speciation occurs due to changes in behav iour or mode of life, in this case by a partitioning of the original population into limnetic and benthic groups. Here, descendent species are most closely related to species living in the same lake. (B) Allopatric speciation occurs following geographical sep aration of the populations, in this case caused by a fall in lake level. Descendent species are most closely related to fishes living in adjacent lakes.
Figure 3 The differences between phyletic gradualism and punctuated equilibrium models of speciation. (A) In phyletic gradualism the shape change is gradual and populations are seen to move across morphospace continuously. Periods of spe ciation are relatively long and can be recorded in the fossil record. (B) In punctuated equilibrium the shape change is inter mittant, rapid, and related to geographical separation of a part of the population. For most of the time the form of the population is static. In most areas no speciation event is seen, and the fossil record shows abrupt changes of morphology with no intermediate stages.
EVOLUTION 165
observation. Work by Johnson on Jurassic oysters (Gryphaea) from across western Europe provides a good example of this aggregate pattern. Change over approximately 6 Ma was generally slow, but rapid periods of change in isolated populations were also observed. One unfortunate result has been the suggestion that punctuated equilibrium is antithetical to Darwinian evolution. In this usage it is not, as even the rapid bursts of evolution implied by the theory would take place via a series of gradual (i.e. smallscale) changes in the form of the organism concerned.
Macroevolution Macroevolution is the study of all evolutionary events or effects larger than the appearance of a new species. This includes studies of long-term change in the geological record and of the emergence of new higher taxa, for example new phyla. Linked to both of these topics is the difficult issue of how significant new structures or organs can evolve. Palaeontology is central to this study, as it provides a measure of time and can identify the most likely dates of appearance of new characters or taxa. The single biggest and most important argument about macroevolution is whether it is a scaled-up version of microevolution or something different. If it is different, then those differences may be a reflection of the emergent properties of this complicated system and hence still reliant on microevolutionary processes occurring. More controversially, it has been argued that macroevolution includes rapid and largescale changes of form that necessitate steps that might initially produce organisms that are less successful than their ancestors. This is completely counter to Darwinian ideas of evolution. An example of these issues can be presented via a consideration of the evolution of major groups of tetrapods. All living vertebrates with pentadactyl limbs (that is mammals, reptiles, amphibians, and birds) evolved from an ancestral fish, with the process beginning in freshwater lakes and rivers in the Devonian (Figure 4). Since then a wide variety of adaptations have appeared in these higher groups, such as feathers, fur, and wings. It can be convincingly demonstrated that some lineages, or evolving lines, acquired these characters gradually, by microevolutionary processes. The classic example of this is the origin of mammals from reptile ancestors through the Triassic and Early Jurassic. Character change occurred at a relatively constant rate throughout this 100 Ma period, and intermediate forms are well known in the fossil record. However, the process by which these new characters appeared may have controls that are not seen in microevolution and which are hinted at by
Figure 4 A simplified evolutionary tree for tetrapods, those vertebrates with pentadactyl limbs. This group of organisms evolved from lobe finned fishes during the Devonian. Whilst the evolution of mammals from cynodonts was a gradual process in which macroevolution appears to conform to microevolutionary expectations, the origin of wings in pterosaurs and birds is more difficult to explain without evoking some new process unique to evolution at this scale.
the suggestion that, during this event, significant evolution tended to occur in small-carnivore groups. More controversially it has been argued that some new characters, for example wings, could not have evolved by gradual steps as they would have been useless in their early stages of development. Complicated counter arguments that invoke the possible uses of wings without flight, for example, have not cleared up the controversy. Looking at the history of life, it is clear that there have been periods of major increases in diversity and periods of major innovation. Significant increases in diversity tend to happen after mass extinctions and are called evolutionary radiations (see Biological Radiations and Speciation). Empty niches created by the extinction event are quickly filled as organisms radiate to form new species that are able to exploit the available resources. Evolutionary radiations may
166 EVOLUTION
also be facilitated by the appearance of major innovations, such as the evolution of hard parts by a variety of different taxa close to the Precambrian–Cambrian boundary. The two are not necessarily coupled. For example, eukaryotes, the complicated internally divided cells with which we are most familiar, evolved more than 2 Ga ago (possibly much more), but they did not become widespread or common until much later at around 1 Ga ago. They could not radiate until there was adequate oxygen present in the Earth’s atmosphere and oceans, as they depended on this molecule for respiration. It seems likely that evolutionary selection can work at the species level as well as at the level of an individual within a population. Species-level selection favours species that have lower extinction rates and higher origination rates than their ‘competitors’. In the long-term, these species become more abundant at the expense of their less-successful competitors. Distinguishing between the levels of selection is extremely difficult in practice, but the theory helps to demonstrate the emergent properties of species. Some types of extinction, or reductions in diversity, may also be explained by macroevolution and, in turn, throw light on the mechanisms of evolution. It is clear that the evolution of a new species will increase competition for resources and may force another species to become extinct if it is unable to compete successfully. The pattern of species extinction would be expected to be one of increasing chance of extinction with species age, but in some cases this does not seem to be so. Instead, species age does not appear to correlate with the likelihood of extinction. Van Valen has used this observation (which is itself somewhat contentious) to suggest a novel hypothesis for macroevolutionary patterns. He suggested that competition for resources produces a dynamic equilibrium between species, in which each will continue to evolve in order to survive. This is the core of the Red Queen hypothesis, which suggests that organisms evolve to keep their biological place or, to paraphrase the quotation from Alice in Alice Through the Looking Glass, they ‘run to keep still’. The characters that help organisms to survive at times of low extinction rate may be different from those that make survival of mass extinctions more likely. In other words, the criteria by which species are selected may vary with extinction rate. Specialist species tend to have greater survival potential at times when extinction rates are low and reduced survival potential when extinction rates are high. In addition, it has been suggested that small species have a greater chance of surviving mass extinctions than larger species, though the overall trend in evolution is clearly not towards smaller species.
The level of understanding of genetics is now so great that it is possible to explore macroevolution in this way. In traditional views of macroevolution, a set of ways in which different forms could be produced with small changes in the genome was known as heterochrony. The idea was that different parts of the body grew at different rates. In some examples, this might be a difference in the rate at which sexual maturity was reached relative to the rate at which the rest of the organism (the somatic portion) developed. If sexual reproduction became possible at an earlier stage in body development, this was known as pedomorphosis. The classic living example of this is the axolotl. This resembles a juvenile salamander, complete with external gills, but reproduces at this stage of development. If it is injected with extract from the thyroid gland, an axolotl will develop into an adult salamander. A genetic view of this kind of evolution is that there has been a change in the regulatory genes that switch on and off the protein-coding gene sequences within cells. If these genes start to operate at new rates, then the phenotype will change shape, in some cases dramatically. It is now known that some genes, especially a group known as Hox genes, control development by instructing the different parts of the growing embryo on which part of the body should be built. It is known that these genes are more common in vertebrates than in other groups of animals and that there was a single period when these genes duplicated (or rather, duplicated twice), so that vertebrates carry four times as many of these genes as do invertebrates. This multiplication occurred between the evolution of the cephalochordates and proper vertebrates, probably during the Cambrian period. It is tempting to assume that this evolutionary event facilitated the increase in complexity needed to produce vertebrates and may have made them more ‘evolvable’ since. Whether or not cause and effect can be proved in this example, it points to a growing understanding of the relationship between genes and macroevolution.
See Also Biodiversity. Biological Radiations and Speciation. Famous Geologists: Darwin; Lyell. Fossil Invertebrates: Trilobites. Origin of Life. Palaeozoic: Cambrian. Precambrian: Eukaryote Fossils.
Further Reading Darwin C (1859) On the Origin of Species. Penguin Books (edited by J W Burrow). Eldredge N and Gould SJ (1972) Punctuated equilibria: an alternative to phyletic gradualism. In: Schopf TJ (ed.)
EVOLUTION 167
Models in Paleobiology, pp. 82 115. San Francisco: Freeman, Cooper. Gingerich PD (1985) Species in the fossil record: concepts, trends and transitions. Paleobiology 11: 27 41. Greenwood PH (1974) Cichlid Fishes of Lake Victoria, East Africa: The Biology and Evolution of a Fish Flock. London: The British Museum (Natural History). Johnson ALA and Lennon CD (1990) Evolution of gryphae ate oysters in the Mid Jurassic of Western Europe. Palae ontology 33: 453 485. Ridley M (1996) Evolution. Oxford: Blackwell.
Sheldon PR (1987) Parallel gradualistic evolution of Ordo vician trilobites. Nature 330: 561 563. Skelton PW (ed.) (1993) Evolution: A Biological and Palae ontological Approach. Wokingham: Addison Wesley Publishing Company. Van Valen L (1973) A new evolutionary law. Evolutionary Theory 1: 1 30. Williamson PG (1981) Palaeontological documentation of speciation in Cenozoic molluscs from Turkana Basin. Nature 293: 437 443.
FAKE FOSSILS 169
FAKE FOSSILS D M Martill, University of Portsmouth, Portsmouth, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction This article considers fake fossils and the part that forgers have played in ruining scientific reputations and hindering the development of science. In some cases, financial reasons appear to have been a motive for forgery, but the impact has, nonetheless, been detrimental to science. This article is written from a certain amount of experience, the author having fallen foul of at least one fossil fabrication. Some tips are provided for those who might encounter fake fossils. Forged fossils fall into a number of categories. Some are complete fabrications and should be considered as sculptures. They may be cast from materials that resemble rock, such as plaster or cement, or carved from real rocks. Some forgeries represent embellishments of genuine fossils, and include the addition of elements from another fossil simply to make an incomplete specimen appear more complete. Other forgeries are conversions whereby a common fossil is transformed to resemble something much rarer; others are chimeras whereby two or more fossils of different animals are united to produce quite fabulous creatures. Some composites are not manufactured deliberately to deceive; rather, many simply represent attempts to fill gaps for aesthetic purposes and to make museum displays more informative. In this latter case, no deceit is intended, but when past curators have failed to keep records of which fossils were amalgamated, taxonomic problems have arisen several years later. In some unusual cases of forgery, remains of modern animals and plants are transmogrified into fossils by being embedded in resins or by being glued onto bedding planes (Figure 1). Deciding what constitutes a fossil forgery can be difficult. Purists might argue that any modification of a fossil represents an act of forgery, although a museum display specimen might be enhanced simply to demonstrate what a skeleton may have looked like when complete, or a damaged piece might be skilfully repaired to obscure an ugly scar or hole, perhaps caused by bad collecting practice. Certainly, Victorian museum curators thought it perfectly acceptable to construct a complete skeleton from the remains of a number of partial skeletons. One of the most famous examples includes the mounted skeleton of the giant sauropod
dinosaur Brachiosaurus brancai that forms the centrepiece to the Humboldt Museum in Berlin. This magnificent skeleton is thought to contain the parts of at least five different individual fossils.
Cruel Hoaxes Fake fossils represent deliberate attempts by the unscrupulous to hoodwink the unsuspecting into believing that an object is a genuine fossil. Such is human nature that as long as fossils have a financial value or can result in prestige for the discoverer or describer, then there are going to be disreputable people prepared to exploit this for their own ends, be they greed, spite, or self-betterment. This is not a new phenomenon, and has been a practice from the earliest days of palaeontology. Some faking of fossils is indeed a consequence of criminal intent to obtain money through deception, but in a number of cases, fossils have been faked in what appear to have been either jokes that have gone seriously wrong or deliberate attempts to ruin scientific reputations. Such is the case of the now famous lying stones of Eibelstadt, near Wurzburg, Germany. This is one of the oldest, well-documented cases of fossil forgery,
Figure 1 In this crude attempt to forge a fossil, a recently dead dragonfly has been glued to the surface of a piece of limestone. Such forgeries at first can appear to be examples of excellent preservation. Be alert if a thin veneer of varnish prevents direct access to the surface of the fossil.
170 FAKE FOSSILS
and concerns a certain Dr Johann Beringer. Apparently, Beringer was an extremely pompous fellow, and was despised by a number of local academics. The academics generated an enormous number of crude forgeries that they passed to Beringer via hired helpers. Despite the crude nature of the fossils, and even despite later attempts by the forgers to reveal their cruel act, Beringer published a scientific account of the fossils in 1726. The book, Lithographiae Wirceburgensis, figured hundreds of the faked specimens, which included sculptures of spiders in their webs, frogs, birds, and even shooting stars and moons. The flagrant act of forgery came to light just before publication, but the book includes a note dismissing the claims of forgery, such was Beringer’s arrogance. Perhaps the most famous example of deception is the case of the Piltdown Man forgeries. Many books have been written detailing this hoax and speculating on the identity of the perpetrator. Essentially, the lower jaw of an orang-utan was substantially altered and buried in a gravel pit along with some fragments of human cranium in Sussex. The remains were discovered by Charles Dawson in 1912 and then described by leading vertebrate palaeontologist Dr Arthur Smith Woodward, who was, at the time, the Keeper of the British Museum of Natural History, London. Although several observers had wondered if the fossil was a forgery, it was not definitively shown to be so until 1953, as a result of a fluorine analysis on the jawbone. This was a sad postscript to the scientific career of Sir Arthur, who had been a brilliant palaeontologist. The hoax was a scandal for British science, and it held back palaeoanthropology for several decades. At least the Piltdown forgery concerned genuine organic remains and Arthur Smith Woodward could perhaps be forgiven for making a mistake; after all, someone had set out deliberately to deceive. In 1966, at the age of 91, noted German vertebrate palaeontologist Professor Freiderick von Huene, of Tu¨bingen University, described a juvenile skeleton of the ichthyosaur Leptonectes (then called Leptopterygius) that had been made from cement, stained brown, and placed on a slab of rock from the Early Jurassic Posidonia Shale Formation. Not a single fossil bone was present on the specimen; it was nothing more than a sculpture, and not a very accurate one at that. Proof of this forgery came to light only when the sculpture was being cleaned by a preparator some 4 years later, and it was not revealed to the scientific world until 1976. Huene never had to face the embarrassment of this expose´ because he died in 1969, and in this particular case, the published paper of Huene was not of great scientific consequence. This is in marked contrast to the paper of Arthur Smith
Woodward, on the Piltdown ‘fossils’, which announced the presence of the oldest hominid fossils in Europe and purported to show that large human brains were an early evolutionary development. But there are some similarities in the two hoaxes. In both cases, the scientists concerned were extremely eminent and had enjoyed careers in which they had risen to the very top of their profession. It would be no surprise to learn that they had made enemies on the way up, and that some embittered rascal had sought cruel revenge. These, fortunately we hope, are rare cases.
Too Much Haste A more recent (November 1999) case of fossil forgery resulted in considerable embarrassment for North American palaeontologist Philip J Currie, artist Stephen Czerkas, and especially for the senior assistant editor of National Geographic Magazine, Chris Sloan. This sorry story concerned a strange case whereby two spectacular, and quite genuine fossils, were merged together to construct a chimera comprising the back end of a small dinosaur, Microraptor zhaoianus, and the front end of a small fossil bird, Yanornis martini, both from the famous Early Cretaceous Yixian Formation of Liaoning Province, China. The two incomplete specimens were joined together to make a single, complete feathered dinosaur. Unfortunately, so much excitement was generated over the specimen that the National Geographic Magazine printed an article on its discovery and its perceived relevance to the ‘birds are dinosaurs debate’ just before the specimen was shown to be a forgery. An even more unfortunate aspect of this case occurred because, unusually for an article in the National Geographic, the fossil chimera was given a scientific name, Archaeoraptor liaoningensis, which, according to the rules of scientific nomenclature, was valid for at least part of the specimen. Paradoxically, it turned out that both halves of the chimera represented important scientific discoveries, and both were new to science. Suspicions surrounding the nature of the fossil came to light when the specimen was scanned using computed axial tomography (CAT), and it became clear that the pieces did not fit together well. A more careful examination then revealed the forgery, and although Phil Currie highlighted some problems with the fossil, these were not relayed to National Geographic. It was only when Chinese palaeontologist Xu Xing had met with a Chinese fossil dealer that the sorry story of the forging really emerged. But by then it was too late; the article had already appeared in the November 1999 issue and the proof of the forgery came one month later. It is to the
FAKE FOSSILS 171
relief of many palaeontologists that this forgery was discovered before too much damage had been done to the scientific case being made for the bird–dinosaur hypothesis, but, sadly, the furore over the forgery has distracted from the undoubted importance of the two genuine fossils. Not all fossil forging has serious consequences for science, and indeed, some forging is undertaken in an attempt to ‘improve’ fossils for the fossil-buying public. Such forgeries usually involve adding embellishments to genuine fossils, or converting fossils from one type to another (Figure 2). Such embellishments do not enhance the value of the fossil; indeed, they destroy the scientific value, but they might make a fossil look attractive to the unwary purchaser. This type of forgery is common among the fossil dealers of Brazil who raid the spectacular fossil fish beds of the Santana and Crato formations of north-eastern Brazil. Here it is common to find forged fishes that have heads and tails belonging to different species. Fins may be added, and some specimens might be artificially lengthened by the insertion of several bodies into one example. Conversions are common, and it is frequent to find heads of large fossil fishes converted into frogs, and small specimens of the gonorhynchiform fish Dastilbe converted into lizards. To the unwary, the presence of some genuine bones is enough to encourage belief that the entire fossil is genuine. Until recently, most of the fossils available commercially from the fossil beds of Brazil were collected to supply flea markets in the tourist centres of Brazil. The fossils were often enhanced to make them visually more attractive to tourists who probably knew very little about fossils, but who wanted to have an unusual souvenir of Brazil (Figure 3). More recently, the genuine fossils have become highly sought after by museums, and many of the rarer fossils from Brazil, such as pterosaurs and dinosaurs, command very high prices. There has thus arisen a new financial incentive for the forging of fossils. Previously, forgeries were rapidly constructed, using a sharp chisel, by the addition of a crudely engraved outline of a fish. Now, elaborate constructions are made by glueing together numerous pieces of real fossils to produce such things as pterosaur bones and crocodile heads. Much time and effort goes into these constructions, but by and large they remain crude and are easy to recognise. However, in a skull of a dinosaur that had been obtained by the Museum fu¨ r Naturkunde, Stuttgart, Germany, a sagittal crest at the back of the skull was revealed to be fake only after CAT scanning. The crest, in fact, was a part of the lower jaw repositioned to make the specimen look more spectacular. There was no need for the forger to have executed this
Figure 2 Examples of genuine fossils that have been altered to resemble something rare. Both are from the Nova Olinda Member of the Crato Formation, Ceara´, Brazil. (A) The fossil gonorhynchiform fish Dastilbe has had limbs added to make it look like a lizard. (B) A fossil insect has had extra legs and claws added in ink to make it resemble something new, perhaps a spider. This fake was revealed by dropping industrial methyl ated spirit onto the fossil. The ink of the faked legs bled into the rock, whereas the real limbs remained intact. Both photographs by Robert Loveridge.
172 FAKE FOSSILS
Figure 3 A forged composite group of fossil fishes. The fishes here are all exceptionally well preserved, but were not found in this association. They are from the famous Santana Formation, Ceara´, Brazil. They have been glued together with a mix of car body filler and ground rock.
embellishment, because the skull represented a new genus and species of dinosaur and represented the most complete skull of a spinosaurid dinosaur ever found.
Amber Amber has long been famous for its fossil inclusions, and has been used in the jewellery industry with or without fossil inclusions for thousands of years. New discoveries of amber in the Dominican Republic have resulted in a large number of forgeries. A majority of these are offered to unsuspecting tourists. Most are sold cheaply, but a number of higher priced specimens containing lizards and frogs have proved to be cleverly executed forgeries. It is not always easy to distinguish forged amber from the real thing. The hot needle test, whereby a red-hot needle is pressed into the specimen, will give off a resinous smell if the specimen is genuine amber, whereas the smell will be acrid if it is a synthetic resin; however, the test inevitably marks the specimen.
Religious Zealots There have been several attempts by those creationists (see Creationism) who appear to feel threatened by palaeontological and geological evidence that runs counter to biblical interpretations that Earth is not terribly old and was created in a very short span of time. Rather than accept the findings of science, some
supporters of a biblically based creation theory have challenged the data on which certain scientific claims are made by attempting to discredit palaeontology. Attempts to do this using logical argument have proved difficult, and so some unscrupulous individuals have attempted to undermine scientific findings by forging data. Perhaps the most notable attempt was the claim that human footprints occurred alongside those of dinosaurs at the Paluxy River site in Texas, USA. The Paluxy River site is famous for lengthy trackways of footprints of Cretaceous dinosaurs and has been made into a National Park. Reports that human footprints had been found in the same layers as the dinosaur footprints had always been treated with scepticism by the palaeontological community. Wrapped in pseudoscientific jargon, photographs of the human footprints side by side with dinosaur footprints were used as ‘evidence’ that humans and dinosaurs were around at the same time, and it would therefore have followed that dinosaurs could not have been millions of years old. Despite considerable protestations by scientists, it was only later that the perpetrators of the forgeries admitted that the human prints were handmade rather than footmade.
When a Fossil is Not a Forgery One of the most important fossils, historically, is the London specimen of the small feathered bird/dinosaur Archaeopteryx lithographica. This fossil was
FAKE FOSSILS 173
widely hailed as a missing link between reptiles and birds because its exceptional preservation in the finegrained lithographic limestones of Bavaria showed it to have a dinosaurian skeleton that included a long tail with numerous vertebrae and a beak in which the jaws possessed teeth. And yet the animal was clothed in feathers, a feature known today only in birds. This was just what devotees of Darwin needed to support the theory of evolution, and indeed Archaeopteryx became the archetypal missing link; it is an animal that appears to be transitional between two groups of animals, a status that is still claimed for it today. During the 1980s, two eminent scientists, but not palaeontologists or geologists, Sir Fred Hoyle and Chandra Wickramasinghe, published a claim that the London specimen of Archaeopteryx was a forgery. If their claim had been correct, there is no doubt it would have had important implications, but such an upheld claim would have had even more dire consequences had it been made in Victorian London. However, several independent analyses of the evidence for forgery showed quite categorically that Hoyle and Wickramasinghe were out of their depth and did not understand the nature and diversity of fossilization processes. Nevertheless, the claim, coming as it did from such noted scientists, generated considerable excitement in the media, and a number of books and papers resulted from the claim. Sadly, many palaeontologists had to devote considerable time and effort to debunk these incorrect claims.
Detecting Forgeries It is advisable always to be suspicious of fossils bought commercially and to be very sceptical of any exceptional fossil that is provided by a ‘friendly’ noncolleague. Fossils traded commercially are quite likely to have been enhanced in order to increase their aesthetic appeal, but such improvements are usually easily detected by experienced palaeontologists. Some traders of ammonites increase the size of the ammonite by carving extra whorls into the rock.
Fabricated parts of fossils are often constructed using plastic-based fillers. These plastics will melt when probed with a hot needle, giving off an acrid smoke. Rock does not do this. Many of the spectacular trilobites from the Ordovician and Devonian of Morocco have been enhanced with fillers, and indeed some are simply casts made from moulds of genuine fossils. The casts are glued to blocks of limestone and coloured black with boot polish.Where there has been a real intent to deceive, the workmanship of the forgery is often very good and almost impossible to detect by casual inspection. Examination under a microscope may help, but when suspicions are raised, proof may come only after expensive CAT scanning or chemical analysis.
See Also Creationism.
Further Reading Charig AJ, Greenaway F, Milner AC, Walker CA, and Whybrow PJ (1986) Archaeopteryx is not a forgery. Science 232: 622 626. Hoyle F and Wickramasinghe C (1986) Archaeopteryx: The Primordial Bird. Swansea: Christopher Davies. Martill D (1994) Fake fossils from Brazil. Geology Today 1994: 36 40. Nield T (1986) The lying stones of Eibelstadt. Geology Today 1986: 78 82. Ross A (1998) Amber: The Natural Time Capsule. London: The Natural History Museum. Russell M (2003) Piltdown Man: The Secret Life of Charles Dawson. Stroud: Tempus Publishing. Sloan CP (1999) Feathers for T. rex. National Geographic 196(5): 98 107. Simons LM (2000) Archaeoraptor fossil trail. National Geographic 197: 128 132. Suess H D, Frey E, Martill D, and Scott D (2002) Irritator challengeri, a spinosaurid (Dinosauria: Theropoda) from the Lower Cretaceous of Brazil. Journal of Vertebrate Paleontology 22: 535 547. Wild R (1976) Eine Ichthyosaurier Fa¨lschung. Neues Jahrbuch fu¨r Geologie und Pala¨ontologie, Monatshefte 1979: 382 384.
174 FAMOUS GEOLOGISTS/Agassiz
FAMOUS GEOLOGISTS Contents Agassiz Cuvier Darwin Du Toit Hall Hutton Lyell Murchison Sedgwick Smith Steno Suess Walther Wegener
Agassiz D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Son of a clergyman, Jean Louis Rodolphe Agassiz (Figure 1) was born in the village of Moˆ tier in Canton of Fribourg, Switzerland. After schooling at Bienne and Lausanne and early acquiring an interest in natural history (particularly of fishes), he attended the universities of Zurich, Heidelberg, and Munich, intending to take a medical degree. However, at Heidelberg he began studying palaeontology under Heinrich Bronn and embryology under Friedrich Tiedemann. He also became friends with Alexander Braun and his family (later marrying his sister Ce´ cile) and Karl Schimper, and the three young men went on botanical excursions together. At Munich, he came under the influence of Friedrich Wilhelm Schelling, Lorenz Oken and German Naturphilosophie. Additionally, he studied botany under Carl Friedrich von Martius, and embryology under Ignatius Do¨ llinger. A natural philosopher should, as Agassiz represented Oken’s view in his autobiography: ‘‘[construct] the universe out of his own brain, deducing from a priori conceptions all the relations of . . . living things’’. In practice, Agassiz certainly
did not discount empirical information. Indeed, his hunt for ‘facts’ was one of his chief priorities. But this search was propelled by the desire to apprehend the activity of the Absolute Being in nature. Indeed, it was strongly influenced by the notion that he was examining the works of a divine Creator. Agassiz’s PhD (Erlangen/Munich) (Selecta Genera et Species Piscium quos in Itinere per Brasiliam Annis MDCCCXVIII–MDCCCXX [1829]) was devoted to the Brazilian fishes collected from Amazonia by the recently deceased Johann Baptist von Spix. It was tactfully dedicated to Georges Cuvier (see Famous Geologists: Cuvier). Agassiz also obtained an MD at Munich in 1830, but by then he was determined to be a naturalist not a physician. The following year Agassiz went to Paris to study comparative anatomy under Cuvier, having already examined numerous collections of fossil fish in leading museums. Cuvier was greatly impressed by the young man’s work and took him under his wing, introducing him to Alexander von Humboldt, and teaching him the principles of comparative anatomy and how to reconstruct fossil fish. So Agassiz gave up the German idea of the unity of the animal kingdom and followed Cuvier’s notion of there being four fundamental types in the animal kingdom. Cuvier was so impressed by Agassiz’s abilities that he passed on the notes, drawings, and specimens that he had collected on fossil fish for him to study. He also ensured that other institutions made their collections available to Agassiz.
FAMOUS GEOLOGISTS/Agassiz 175
Figure 1 Louis Agassiz (1807 1873).
Cuvier died of cholera in 1832, but his influence on Agassiz was strong and permanent, particularly respecting the idea of successive geological catastrophes and the creation of new species. Agassiz’s studies of fossil fish eventually yielded his great treatise Recherches sur les Poissons Fossiles (5 vols, 1833–1843), with the figures mostly drawn by the artist Joseph Dinkel (whom he employed over a long period); and Monographie des Poissons du Vieux Gre`s Rouge [Old Red Sandstone] ou Syste`me De´ vonien des Iles Britanniques et de Russie (1844–1845). Agassiz received the Geological Society’s Wollaston Medal for his ichthyological work in 1836. At a youthful 25 years of age, Agassiz was appointed Professor of Natural History at the small new Lyceum or Academy at Neuchaˆ tel, back in his home region of Switzerland, and soon began to establish that institution’s reputation. His early magnum opus made use of specimens sent to him from all over Europe, and in particular from the Old Red Sandstone of Scotland, to which country he made two visits. The later association with the amateur stonemason Hugh Miller, who arranged for Agassiz to receive specimens of Devonian fossil fish, is particularly well known through Miller’s popular book The Old Red Sandstone (1841), and his contributions were incorporated into Agassiz’s work on Devonian ichthyology. Unfortunately, Agassiz’s first marriage to Ce´ cile Braun failed, in part because he gave so much attention to his work and partly because he came under the sway of his assistant, the geologist Edouard Desor,
who pushed his way into the Agassiz household despite Ce´ cile’s objections. Moreover, Agassiz’s ambitious publishing projects led to financial problems and life became difficult for him in Neuchaˆ tel. He therefore sought the assistance of von Humboldt and Charles Lyell (see Famous Geologists: Lyell) to travel to North America, and in 1846 he went to Boston at the invitation of James Avory Lowell to give a lecture series on natural history. These were outstandingly successful, and led to his appointment as Professor of Zoology and Geology at Harvard in 1848, where he soon became one of the country’s leading scientists. In 1852, he was additionally professor at the Medical School at Charleston, South Carolina, and also at Cornell University in 1868. Declining a chair in Paris, despite the offer of most favourable terms, Agassiz committed himself to American science, pushing, with the help of endowments from Francis Calley Gray and others, for the foundation the famous Museum of Comparative Zoology at Harvard in 1858–1859 (which opened in 1860). In 1863, he helped persuade Abraham Lincoln to establish the National Academy of Sciences; the same year Agassiz was appointed a regent of the Smithsonian Institution. Agassiz had reached the top of the tree. Subsequently, Agassiz travelled widely on both land and sea and wrote numerous scientific papers in the USA, as well as popular essays, reviews, and educational works, his writings on classification being the most influential. However, while revelling in the hospitality and opportunities that America offered, he retained a belief in the superiority of European science and culture, which later alienated some colleagues. Left behind in Europe, Agassiz’s wife had died of tuberculosis in 1848. His son joined him in America, and subsequently his two daughters. In 1850 he married Elizabeth Cary, who later founded a girls’ school that later developed into Radcliffe College at Harvard. A number of Agassiz’s European epigone followed him to America, including Desor, with whom Agassiz eventually fell out, after an unpleasant episode involving accusations of plagiarism, financial malfeasance, and worse, for which Agassiz was found to be without fault. His first wife’s intuitions were more than vindicated. Desor withdrew to Europe. Apart from collecting, naming, and describing modern and fossil fish, Agassiz also proposed a scheme for fish classification, based on their scales. This was not ‘biologically’ ideal, but suited the study of fossil fish, for which in many cases the scales are the best preserved remains, the bones having been cartilaginous. Thus four main orders of fish were proposed, based on their scales, rather than their crania:
176 FAMOUS GEOLOGISTS/Agassiz
1. Having plate-like scales, often with tubercles or bony points, detached from one another and irregularly arranged on a tough skin (Placoid). 2. Having large, bony, usually shiny (enameled) platelike scales, not normally overlapping, but often interlocking in some way (Ganoid). 3. Having thin, horny, overlapping plates, each having one side with a jagged edge or comb-like projections (Ctenoid). 4. Having thin, flexible, overlapping, horny scales, smooth in outline and circular or elliptical in form (Cycloid). For the Ganoids, Agassiz was especially interested in the modern Lepisosteus, which was the sole surviving modern representative of the group. So, like Cuvier (who worked on elephants, mammoths and mastodons), he specifically sought to compare living and extinct types. Agassiz’s taxonomic system was later superseded by various others, based principally on bones rather than scales, though his introduction of the Ganoidei was a substantial contribution. His taxonomy was problematic, for, while comparing fossilized and modern forms, he saw no evolutionary connection between them. On the other hand, he offered something new by the use of fossil fish for stratigraphic purposes. Moreover, in a manner that would have appealed to Cuvier, he sought to find out something about the ‘conditions of existence’ of his specimens as to temperature, salinity, and mode of locomotion. He supposed that prior to the Cretaceous there seemed to be less distinction between fresh-water and marine forms than at present and it might be the case that these two environments were not so marked previously as they are at present. However, Agassiz’s most important contribution to geology was his advocacy of the concept of an Ice-Age (Eiszeit), fundamental to Pleistocene geomorphology and stratigraphy (see Tertiary To Present: Pleistocene and The Ice Age). Curiously, it was linked to the biological ideas that he imbibed from Cuvier. In Switzerland, the idea that the country’s glaciers were formerly of greater extent had been recognized by observers back in the eighteenth century, such as the minister Bernard Friedrich Kuhn (1787). There is a report of a manuscript by a mountaineer Jean-Pierre Perraudin (1818), which described the extent of moraines and erratic boulders, and regarded striated and polished rocks as evidence of glacial action. It was perhaps Perraudin who really initiated the glacial theory in Switzerland. The highway engineer Ignaz Venetz accepted Perraudin’s ideas and read a paper on the topic at Neuchaˆ tel in 1829. The mining engineer Jean de Charpentier, director of the salt mines at
Bex, also obtained information from Perraudin and in 1834 read a paper at Lucerne about the former greater extent of glaciers. (Agassiz met Charpentier when he was still at school and was partly inspired by him to become a naturalist.) However, Charpentier’s paper was regarded as mistaken and was mocked, Agassiz being one of the opponents. (Historians examining Agassiz’s students’ lecture notes from that period have shown that he was then critical of the theory.) But in 1836 Agassiz was in the Bex area and was shown around by Charpentier, and after calling on Venetz and examining the evidences in other parts of Switzerland he became a convert to the theory. While in Bex, Agassiz met his old student friend, the botanist and palaeontologist Karl Schimper, and the two also discussed the glacial evidence. In February 1837, Schimper gave a botanical talk at Neuchaˆ tel, at the conclusion of which he passed round a copy of a poem that introduced the new word Eiszeit. By then, Agassiz had picked up the evidences and ideas in their entirety and was running with them. He presented a first outline of his views in public at the meeting of the Socie´ te´ Helve´ tique des Sciences Naturelles at Neuchaˆ tel in July 1837, in what became known as the Discours de Neuchaˆ tel. By 1840 Agassiz published his major study on the topic, and his most important contribution to geology: E´ tudes sur les Glaciers. In publishing this, he got ahead of Charpentier’s Essai sur les Glaciers (1841), and recriminations followed, stirred up, it has been suggested, by Desor. Schimper was also annoyed with Agassiz for failing to mention him in E´ tudes (though he was mentioned in the Discours). There followed a further work on glaciers co-authored with Arnold Guyot and Desor, describing the different types of glaciers, their component parts, their motions, and a detailed account of the Aar Glacier: Syste`me Glaciaire: Ou Recherches sur les Glaciers, leur Me´ canisme, leur Ancienne Extension et le Roˆ le qu’ils ont Joue´ dans l’Histoire de la Terre (1847). The Discours was written in haste, but provided persuasive evidence for the former extension of glaciers, at least in Switzerland, and strong arguments against the floating iceberg theory favoured by Lyell in Britain, or the common idea of glacial erratics being emplaced by catastrophic floods. On the other hand, Agassiz thought that erratic boulders might have fallen into their present positions rather than being directly transported by ice. Agassiz’s E´ tudes was a sumptuous volume, beautifully illustrated, providing all the documentation necessary to convince readers of the former extension of glaciers. The theory could also explain the existence of the vast extent of superficial deposits (‘till’) over northern Europe, then
FAMOUS GEOLOGISTS/Agassiz 177
in primogeniture, or, what is the same thing, that they are identical species.
Figure 2 Sketch in Agassiz’s Discours de Neuchaˆtel showing hypothetical fall of global temperature over time.
known as ‘diluvium’ (by association with the notion of catastrophic floods) or ‘drift’ (by association with the floating iceberg theory). However, the explanatory theory advanced by Agassiz was much less persuasive than that which it was supposed to explain. It was widely held at the time (in accordance with the views of E´ lie de Beaumont, e.g.) that the earth was cooling. Agassiz arbitrarily assumed that it did so in a fashion indicated by his sketch (Figure 2), which appeared in both the Discours and the E´ tudes. This graph was supposed to represent not just a cooling inorganic planet, but one inhabited by living organisms that were wiped out, however, from time to time by ‘Cuvierian’ catastrophes, and then replaced by different sets of organisms. The abrupt falls of temperature corresponded to the sudden disappearance of life forms, with the temperatures remaining approximately constant so long as the (supposedly heat sustaining) life forms continued in existence. Agassiz supposed that the formation of the Alps themselves was an event of recent occurrence, was preceded by a ‘catastrophic’ fall in temperature, and was then followed by the establishment of modern forms of life. On this view, then, the epoch preceding the present could have been of extreme cold, producing the former extended glaciation evidenced in the Alps. The onset of cold must have been sudden, from the appearance of mammoth remains in Russia. Agassiz suggested that the glaciation could have extended from the North Pole right down to the Mediterranean and Caspian seas. Thus the Great Ice-Age. This theory was perhaps the most ‘catastrophist’ ever propounded by a ‘respectable’ geologist (other than bolide aficionados). Agassiz wrote (1838: 382): [T]he epoch of extreme cold which preceded the present creation . . . was attended by the disappearance of the animals of the diluvian epoch of geologists, as the mam moths of Siberia still attest, and preceded the uprising of the Alps, and the appearance of the animated nature of our day, as is proved by the moraines, and the presence of fish in our lakes. There was thus a complete separ ation between the existing creation and those which have preceded it; and, if the living species sometimes resemble in our apprehension those which are hid in the bowels of the earth, it nevertheless cannot be affirmed that they have regularly descended from them
Thus Agassiz set his face against transformism or evolution and offered hyper-catastrophism and the doctrine of special creations (assuming but not then stating) that they occurred by some divine means. Agassiz visited Britain in 1834 and 1835, chiefly in connection with his interests in fossil fish, but he also made the acquaintance of the ‘diluvialist’ William Buckland, who in turn visited Agassiz in Switzerland in 1838. Buckland had long been interested in the drift deposits, which he earlier has ascribed to the Noachian Flood, and introduced the distinction between ‘diluvium’ and ‘alluvium’. He was, however, converted to Agassiz new theory during the course of his 1838 visit, realizing that features of British geology that had long puzzled him could be successfully explained in terms of the land-ice theory. In 1840, Agassiz attended meetings of the Geological Society in London and the British Association in Glasgow, and presented his glacial theory, prompting much discussion in British geological circles. However, the theory, as presented in Glasgow, tried to reconcile the new doctrine with the older idea of glacial submergence, for after the melting of the glaciers flood waters could have moved boulders and gravels (thus accounting for glaciofluvial materials). Following the meeting, Agassiz and Buckland went on a tour of Scotland, and were satisfied that they could see most satisfactory evidence in favour of the land-ice theory, and successfully interpreted the ‘Parallel Roads of Glen Roy’, which Darwin had the previous year interpreted as marine shore-lines, as being due to the successive shore lines of an icedammed lake, an interpretation that was rapidly published in the newspaper The Scotsman. Following his Scottish tour, Agassiz proceeded to Ireland, where again he found ample evidence for glaciation. Returning to Scotland, he then journeyed back to London, seeing many more evidences of glaciation, and spoke at the Geological Society. Debates about the land-ice theory rumbled on in Britain for the next quarter century. Lyell was initially converted to Agassiz’s ideas, but most other influential geologists such as Roderick Murchison (see Famous Geologists: Murchison) were not. Not long after Agassiz returned to Switzerland, Lyell recanted: ‘‘he found the proposed departure from present temperature conditions too much to accept for his uniformitarianism, and he reverted to the glacial submergence theory and floating icebergs’’. It was not until the 1860s that more general acceptance of the land-ice theory began, with the suggestions of the surveyor Andrew Ramsay as to how glaciers might excavate the basins that are now
178 FAMOUS GEOLOGISTS/Agassiz
occupied by mountain lakes; and land ice could have moved uphill to deposit marine shells on the tops of hills in North Wales. Eventually, through the theoretical work of James Croll (1875), an astronomical theory of the origin of climate change was developed, and such theory has been under discussion through to the present. In America, Agassiz successfully applied his ideas to the interpretation of observations in the Great Lakes area, which he explored in 1848, finding new fish for his examination and ample evidences of glaciation (see his Lake Superior; Its Physical Character, Vegetation, and Animals, Compared with those of Other and Similar Regions [1850]), and many other regions. But most of his work in the USA was zoological rather geological. In particular, and in keeping with his long-held Cuvierian views, he was active in his criticism of Darwin’s evolutionary theory, which ran counter to Agassiz’s long-held beliefs about the special creation of life forms. Agassiz even resisted the idea that different varieties of animals of the same species could be produced through time, from which it followed that the different human races were essentially different species! Thus, he gave ‘scientific’ comfort to racial bigots. Agassiz could not comprehend how similar but different creatures of the same species could have been produced worldwide. As a special creationist, that seemed to him to be the required alternative to his ‘polygenism’, and as such had to be rejected. Agassiz’s views in fact succeeded in driving Lyell further into the evolutionist camp. Also, because he was opposed to the idea of variation over time, Agassiz was inclined to suppose that every variety of fish he encountered represented a different species. Hence his classification became inordinately unwieldy. It is interesting that a figure, published in 1844, depicting the genealogy of his four main groups of fish, looks quite like a modern evolutionary tree, yet none of the ‘branches’ are shown as linking at their bases, though they ‘lean towards’ one another, so to speak, in a way that a later evolutionist might regard as suggestive. The source of Agassiz’s anti-evolutionism can be traced to his contacts with Oken and German Naturphilosophie, and associated Platonism (fused with Christian beliefs), as well as Cuvier. Species, for Agassiz, could be regarded as ‘types’ representing the ‘thoughts’ of the Creator. Because there could be no substantial natural variation over time, events such as the Ice-Age represented catastrophes of divine origin that also offered the possibility of renewed creative activity. The ‘plan’ of Creation was, he supposed, better understood by the natural historian than the theologian.
Agassiz was not a great geologist, despite his outstanding capacity for grasping and ordering information, and his powers as a teacher. In 1865–1866, he visited South America, funded by a wealthy Bostonian, Henry Thayer, hoping to find evidences of glaciation in the tropics. Seriously perturbed by Darwin’s theory, Agassiz sought new evidence to support of his long-held ideas about catastrophes and the great Ice-Age. He wished to show that the event was of worldwide extent: so it should be possible to find evidence for it in the southern hemisphere, even in Amazonia. In Brazil, he thought he had found the evidence he sought, but he mistook boulders produced by tropical weathering for glacial erratics, and soil produced by weathering was misidentified as glacial till. His co-workers did not all agree, but Agassiz thought he had the experience and expertise to recognize glacial evidence when he saw it (though he admitted he saw no glacial striations). Agassiz’s attempt to extend his Ice-Age to equatorial regions was a failure and provided a classic example of ‘theory-laden’ observations. On the other hand, his recognition and advocacy of the concept of a glacial epoch and the land-ice theory (even if not original to him) was of fundamental importance, marking the beginning of glaciology and all that followed in the study of Pleistocene geology.
See Also Creationism. Evolution. Famous Geologists: Cuvier; Darwin; Lyell; Murchison. Fossil Vertebrates: Fish. History of Geology From 1835 To 1900. Tertiary To Present: Pleistocene and The Ice Age.
Further Reading Agassiz L (1887) Geological Sketches. New York: Houghton, Mifflin & Co. Agassiz L (1967) Studies on Glaciers Preceded by the Dis course of Neuchaˆ tel Translated and Edited by Albert V. Carozzi. New York and London: Hafner Publishing Company. (This volume contains an English translation of Agassiz’s Discours de Neuchaˆtel.) Andrews SM (1982) The Discovery of Fossil Fishes in Scotland up to 1845 with Checklists of Agassiz’s Figured Specimens. Edinburgh: Royal Scottish Museums. Brice WB and Figueiroˆ a SFdeM (2001) Charles Hartt, Louis Agassiz, and the controversy over Pleistocene glaciation in Brazil. History of Science 39: 161 184. Carozzi AV (1973) Agassiz’s Influence on Geological Thinking in America. Archives des Sciences Gene`ve 21: 5 38. Davies GL (1969) The Earth in Decay: A History of British Geomorphology 1758 1878. London: Macdonald Technical and Scientific.
FAMOUS GEOLOGISTS/Cuvier 179
Gaudant J (1980) Louis Agassiz (1807 1873), fondateur de la pale´ oichthyologie. Revue d’Histoire des Sciences 33: 151 162. Lurie E (1960) Louis Agassiz: A Life in Science. Chicago and London: Chicago University Press.
Marcou J (1896) Life, Letters, and Works of Louis Agassiz, 2 vols. New York: Macmillan (reprinted Gregg Inter national, 1971). North FJ (1943) Centenary of the glacial theory. Proceed ings of the Geologists’ Association 54: 1 28.
Cuvier G Laurent, Brest, France ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Georges Cuvier was one of the grand masters of zoology in the first third of the nineteenth century. He laid the foundations of vertebrate palaeontology, and his work led to the development of the idea of stratigraphical stages through the work of Alcide d’Orbigny. Cuvier’s rivalry with Lamarck, the founder of invertebrate palaeontology, manifested itself in their disputes over the subjects of catastrophism in the history of the Earth and transformism in the history of life.
Biography Jean-Le´ opold-Nicolas-Fre´ de´ ric (called Georges) Cuvier was born on 23 August 1769 in Montbe´ liard, which at that time belonged to the Duchy of Wurttemberg (Germany) but retained French as its language. At an early age he showed an interest in the study of the natural world. As he came from a Protestant family, his parents intended that he should become a pastor, but he failed his entrance examination to the seminary. Nevertheless, he obtained a scholarship to the Caroline Academy in Stuttgart, where, during the years 1784 to 1788, he received training appropriate for a future official in the service of the Duchy. In accordance with his personal interests, he also attended courses in natural history. He became friendly with Christian Heinrich Pfaff (1772–1852) and more particularly with Karl Friedrich Kielmeyer (1765–1844), who was similarly devoted to zoology and who became Professor of Zoology at the Caroline. It was he who taught Cuvier the art of dissection and gave him his ‘first ideas about philosophical anatomy’. After failing to obtain a post in the bureaucracy at the end of his period of training, Cuvier found employment as a tutor to an aristocratic Protestant family in Normandy, where he spent the years 1788 to 1795, the most disturbed period of the Revolution.
He devoted his leisure time to studying botany and the anatomy of animals, particularly molluscs, which he encountered in the neighbouring coastal area. Thanks to his friends Pfaff and Kielmeyer, Cuvier maintained his links with German naturalists. In April 1795, with the assistance of the physician and agronomist Abbe´ Alexandre Tessier (1742–1837), a refugee at Fe´ camp, Cuvier was able to establish himself in Paris. He was well received there, particularly by Etienne Geoffroy Saint-Hilaire, who was already a Professor at the Museum and with whom he became friendly. They collaborated with one another and coauthored some articles. Upon his arrival, Cuvier obtained a teaching position at the newly established college at the Panthe´ on. The same year he was chosen by Antoine Mertrud to fill a vacancy at the Muse´ um d’Histoire Naturelle. This marked the beginning of Cuvier’s distinguished teaching career, both there and in the university. He was named a Member of the First Class of the Institut de France (subsequently the Acade´ mie des Sciences) when it was formed in 1795. In 1800, he was appointed to Jean Daubenton’s former chair at the Colle`ge de France. In 1802, when Mertrud died, he became titular Professor of Comparative Anatomy at the Muse´ um d’Histoire Naturelle. In 1803, he became Permanent Secretary of the First Class of the Institut de France. Simultaneously, Cuvier pursued an administrative career. In 1802, he was appointed Inspector General of Public Education. In 1808, Napoleon named him Councillor of the University, which he was reestablishing, and in 1810–1811 Cuvier was one of the leading lights in the reform of higher education, first in France and subsequently in Italy, Germany, and the Netherlands. In 1813, he was a Councillor of State as ‘Maıˆtre des Requeˆtes’. The Restoration brought him still more honours. Louis XVIII appointed him Chief Councillor of Public Education and made him a Baron in 1819. The same year he was named President of the Section of the Interior in the State Council, representing the interests of non-Catholics. In 1824, Charles X conferred on him the honour of Officer of the Le´gion d’Honneur, of which he had been a Chevalier since the time of the Empire. Louis-Philippe
180 FAMOUS GEOLOGISTS/Cuvier
named him Pair de France in 1831. Cuvier was a member of three sections of the Institut de France: the Acade´ mie Franc¸ aise, the Acade´ mie des Sciences, and the Acade´ mie des Inspections et Belles-Lettres, as well as numerous foreign academies. In 1803, Cuvier married the widow of the former fermier ge´ ne´ ral Duvaucel, who had been guillotined in 1793. None of their four children survived, and their deaths caused Cuvier great distress. Cuvier died on 13 May 1832, at the height of his fame, after a short illness, the precise nature of which is unknown (although it may have been cholera or myelitis).
Cuvier’s Work and Achievements The eighteenth and nineteenth centuries were dominated by a desire to emulate the astronomical achievements of Newton in other areas of science. Cuvier aspired to be the Newton of natural history. He wanted to introduce into this field the approach that henceforth would govern all physical sciences: analyse facts, isolate them, compare them, and then try to ascertain general causes to explain the facts thus ordered according to common laws or principles. His best-known law – the one that made possible his fossil reconstructions – was the law of the correlation of organs or parts: all the parts of an organism must be suitably correlated so as to make a viable whole, capable of coping with the conditions of existence. He adopted the ‘comparative’ approach in the late eighteenth century when endeavouring to restore the remains of mastodons that had been sent to France from America for examination. The task was accomplished using anatomical analogies with modern elephants (for which he regarded the African and Indian types as being distinct, as were the remains of the Siberian mammoth). Applying this principle, Cuvier succeeded in reconstructing a large number of extinct forms. A single tooth, so to speak, told him everything about an organism, he triumphantly proclaimed apropos his reconstruction of the Mosasaurus. The immutable laws of zoology, with their wonderful constancy, which are not contradicted in any class or family, served Cuvier admirably in his arduous task of ‘resurrecting’ (his word) the past. The notion of species obviously underpinned all attitudes towards, and classifications of, animated nature. It was one of the most clearly defined concepts in Cuvier’s work. The most important concepts in nature were those of the individual and the species, and they were connected through the process of generation. Organized beings had two bases for natural classification: the individual, resulting from the common action of all the organs; and the species, resulting from the bonds created by the generation
of individuals. From his earliest publications, and particularly in his Tableau E`le´ mentaire de l’Histoire Naturelle des Animaux (published in 1797), Cuvier gave a definition to which he remained steadfast: The collection of all organized bodies born one from another, or having parents in common, and all those that resemble them in the same way as they resemble each other, is called a species. [Cuvier G (1797) Tableau E`le´ mentaire de l’Histoire Naturelle des Animaux. Paris: Baudouin. p. 11]
But, in practice, in many cases – and whenever considering the past – one cannot use the descent of forms to define species. So, they must be classified by their distinctive external, and more particularly their internal, parts. Form becomes the prime consideration in the study of living bodies, and gives anatomy a role that is almost as important as that of chemistry. Although Cuvier seems at times to have supposed that there was really nothing in nature other than the species and the individual, nevertheless the study of living forms led him to ascribe a concrete reality to another type of organization, namely that of embranchements. An embranchement was an ensemble of animal forms that had a common structural plan, which served as the basis for all external modifications. Cuvier’s four embranchements-vertebrates, molluscs, articulata (jointed or segmented animals), and zoophytes or radiata-are still well known. If there was a ‘closed system’ in Cuvier’s mind it would seem to have been in systematics, at the level of the embranchements. Each of these formed a separate whole; there was no transition or gradation from one embranchement to another. Other organisms would not be viable because they would not meet the conditions of existence. The ‘construction plans’ of the different embranchements were entirely different. There is, for example, no passage from vertebrates to molluscs. Whatever arrangement is given to animals with back bones and those without them, one can never place one of their large classes at the end of one group, and some what similar animals at the head of the other so that the two are linked together [Cuvier G (1800) Lec¸ ons d’Anatomie Compare´ e: 1. Paris: Baudouin. p. 60]
Similarly, There can be no intermediary between mollusca and articulata, nor between them and the radiata, for one cannot fail to recognise the profound interval or ‘salta tion’ there is when one passes from one construction plan to another.
It was in this spirit that Cuvier undertook the palaeontological investigations for which he became famous.
FAMOUS GEOLOGISTS/Cuvier 181
One of his first concerns was to evaluate the significance of fossils in the reconstruction of the Earth’s past. The ‘documents’ furnished by the successive layers – the ‘charters’ or ‘diplomas of the history of the globe’ – revealed that all organisms were not created simultaneously. There was a ‘definite succession in the forms of living organisms’. If there were only unfossiliferous strata, one might claim that the various terrains were created at one and the same time. But palaeontology showed that the various classes of vertebrates do not date from the same epoch. Cuvier was certain that the oviparous quadrupeds appeared much earlier than the viviparous types, for he thought that they began with the fishes, whilst the terrestrial quadrupeds appeared long after. Moreover, there is not only an order of succession between classes but also a pronounced order of the species within the stratigraphical column. In establishing this chronological sequence, thanks to the collaboration of his friend Alexandre Brongniart (1770–1847), who was more of a geologist than he was, Cuvier clearly affirmed the connection between fossils and geological strata – between palaeontology and stratigraphy – to the benefit of the ‘true’ theory of the Earth. Indeed, in 1806, he proposed to the Acade´ mie des Sciences a programme of palaeontological research that would qualify as stratigraphical. Some of the main tasks were to ascertain whether there was any regularity in the succession of fossils, to determine which species appeared first and which came later, and to discover whether these two kinds of species are never found together, or whether there are alternations in their reappearance. In his Recherches sur les Ossemens fossiles de Quadrupe`des (published in 1812), Cuvier applied himself to this programme. With Brongniart, he had proposed to resolve the following questions by means of his studies. Are there animals or plants that are proper to certain strata, and which do not occur in others? Which species appear first, and which come after? Do these two sorts of species sometimes occur together? Are there alternations in their recurrence; in other words, do the first forms recur and the second ones then disappear? Have these animals and plants perished in the places where their remains are found or have they been transported there? Cuvier noted that the fossils of the Calcaire grossier of the Paris Basin are entirely different from those of the Chalk, and he clearly enunciated the concept of ‘characteristic fossils’ in stratigraphy. If a formation could be characterized by its fossils, it was because it contained organisms that were entirely distinct from those of other formations. Fossils were the fundamental tools of stratigraphical determination, and Cuvier could use them to recognize a particular
formation in a large number of calcareous beds. A formation previously observed in some distant canton could be recognized by the nature of the fossils in each bed. Fossils were marker signals that never failed. Indeed, there was a constant relationship between the strata and the animal and plant remains found in them. Thus an immense field of observation and research was opened up, and Cuvier never doubted that reality would correspond progressively with this programme. Precisely determining fossil species and delimiting the places where bones were discovered would make it possible to compare not only the superposed strata but also strata that were juxtaposed, in a parallel geological situation, neither above nor below but adjacent to one another, in the same basin or at a distance in two separate basins. Cuvier was not content simply to study the Paris Basin: he extended his observations to other regions. He thought it was important to study the calcareous strata of other basins and to compare them with those of the Paris Basin. Applied successively to other cantons, this method would soon yield important generalizations, and palaeontology, too long fed by illusory conjectures, would evolve a rigorous progress similar to that of other natural sciences. From stratigraphy, the true history of the Earth began to emerge. When fossils were studied in situ, or in relation to strata, they ceased to be simple curiosities and became ‘historical records’. Thanks to their study according to this perspective, one could show that there had been successive epochs in the formation of the globe and that a series of different operations or processes had operated at different times. The historical key provided by palaeontology was thus well established by Cuvier. He was enthusiastic about the grand prospect of studying the past. He wanted to be able to arrange organisms in their chronological order, know about the development of life, determine precisely which forms appeared first, and recognize the simultaneous appearance of certain species and their gradual destruction. His vision provided a research programme for geologists once they had renounced their ‘just-so stories’ and begun, instead, to do the work of historians. The problem of the history of the Earth was correctly posed in palaeontological and stratigraphical terms. Even though, after Cuvier and Alexandre Brongniart, there was still much to discover, the method for discovery left nothing to be desired. Cuvier’s early works made him one of the masters of comparative anatomy and also opened up a new field to him – palaeontology – that was seemingly full of promise and in which his knowledge of German science, which was then more advanced than French
182 FAMOUS GEOLOGISTS/Cuvier
science, gave him a privileged position in France. His Recherches sur les Ossemens Fossiles de Quadrupe`des was published in 1812, but the work had been published earlier in the Annales du Muse´ um. In a communication made to the Institut de France, Cuvier had, in 1801, enunciated three hypotheses that were, even then, already known to his colleagues: former species had been entirely destroyed, or they had been modified in form, or they had been transported from one climatic zone to another. The second of these three alternative explanations had originally been suggested by Jean-Baptiste Lamarck (1744–1829), while the third was proposed by Barthe´ le´ my Faujas de St. Fond (1741–1819) and others. Cuvier preferred the first, which involved not only the disappearance but also the destruction of ancient forms. Thus, from the beginning of his scientific career, the young naturalist adopted the postulate of what came to be known as catastrophism as the basis for his palaeontological researches. With this end in view, Cuvier applied himself to distinguishing carefully between fossil and modern forms. He was certain that none of the ancient forms had ‘living analogues’. The 23 species that he had already been able to restore all appeared to have been ‘destroyed’ and to have become extinct. This assumption had direct consequences for geological theories: the lost species had ‘‘belonged to beings from a world anterior to our own and to beings that were destroyed by some revolutions of the globe’’ (Me´ moire sur les espe´ ces d’Ele´ phans vivantes et fossiles, Me´ moire de l’institut national des Sciences et des Arts, Fructidor an VII (aouˆ t-septembre, 1799), 2, 1–22: cit. p. 21). The disappearance, or, as Cuvier put it, the ‘destruction’, of former beings could only be explained by a ‘general revolution of nature’. The master palaeontologist laboured hard to establish the reality of the ‘last catastrophe’, which was related to the ideas favoured by believers of Holy Scripture. Although the picture of a total destruction of organisms did not correspond with the facts given in the Bible, Cuvier presented himself as a defender of the Noachian Deluge. He, along with Jean-Andre´ Deluc and De´ odat Dolomieu (who were defenders of the idea of the Flood as a geological agent), thought that if anything was certain in geology it was that the surface of the globe had undergone a sudden revolution no more than five or six thousand years ago. But this catastrophe was, for Cuvier, only ‘the last universal inundation’. That Cuvier was a supporter of general catastrophes does not, however, mean that he did not also uphold the idea of limited or local catastrophes. In the series of revolutions that he proposed, some were only partial. But just one universal catastrophe was enough to raise the problem that it posed
for the continuity of life. Thus, although there had been numerous revolutions, there had not been so many creations, for migrations could play a role in some cases, as he suggested could potentially occur in New Holland (Australia). If there was an inundation of Australia that destroyed all its marsupials and the continent was subsequently colonized by animals from Asia, then the stratigraphical record in Australia would show a general catastrophic flood followed by the new creation of forms. Cuvier never proposed a precise number of revolutions or creations. It was his follower Alcide d’Orbigny (1802–1857) who devoted himself to such calculations. He divided the stratigraphical column into 27 stages and therefore proposed a total of 28 creations, which came to serve as the basis for later stratigraphical work. Nowhere in Cuvier’s oeuvre do we find the expression ‘successive creations’. However, he originated the idea of repeated creations. From the moment when he envisaged general irruptions that ‘‘destroyed all the quadrupeds that they reached’’ and ‘‘caused the entire classe to perish’’ new creations were required to make the animals reappear. Cuvier’s position on this was quite explicit, and so well known that from the beginning of the nineteenth century until his death he was considered to be the catastrophists’ leader, as the eminent geologists who knew him would have agreed. Could Cuvier, who was able to speak so clearly and on occasion defend himself so well, have been misunderstood to such an extent by his contemporaries? This is inconceivable. Cuvier had other good reasons for rejecting transformism, which were doubtless less significant for him than his catastrophism, but which had greater scientific validity. He raised a palaeontological objection that was valid even from a non-catastrophist perspective, namely the absence of intermediate or transitional forms between the former creatures and those that were more recent or extant. If the most ancient forms were the ancestors of those that followed, then one would expect to find the remains of the genealogical intermediaries. But, Cuvier objected, such transitional forms are never found. Between the Paleotherium and today’s most similar species no such forms had at that time been found. Cephalopods do not lead on to fishes – a fact that even Lamarck did not dispute. However, Cuvier did not say how he thought new forms could have been created.
Influence: The ‘Domination of Cuvier’ According to David Hull, in 1860, ‘‘on the continent, especially in France, catastrophism still reigned’’. But this view is mistaken.
FAMOUS GEOLOGISTS/Cuvier 183
Cuvier died in 1832 at the height of his fame. He was, however, already a controversial figure in politics, well known for his participation in the service of a succession of different regimes, and his intellectual and scientific worth were questioned by some of his contemporaries. The judgment of Goethe is well known: ‘‘No one described a fact better than he did. But he is almost devoid of philosophy. He will produce pupils well trained but with little depth’’ Eckernonn J.P. (1948). Gespr¨ ache mir Goethe. Muncher, Deutsches Verla¨ghaus Bong, pp. 329–330. Alexander von Humboldt, who admired Cuvier’s ‘‘memorable studies of fossil bones’’, revealed his disagreements with him during a lecture Cuvier gave at the Colle`ge de France, in which Cuvier criticized the ideas of E`tienne Geoffroy Saint-Hilaire. Cuvier undoubtedly had faithful followers, especially in England; William Buckland is the best known but many others could be cited. Cuvier also had disciples in France, of whom Adolphe Brongniart is the best known and was the most devoted but there were others, for example Le´ once E`lie de Beaumont and d’Orbigny (although they had some reservations). E`lie de Beaumont followed Cuvier in his rejection of fossil man, but did not do so when it came to the general destruction of life, as Darwin noted in 1859. Even d’Orbigny questioned some of Cuvier’s conclusions about the Earth’s past. It is among the members of the thriving and internationally esteemed Socie´ te´ Ge´ ologique de France, founded in 1830 – of which, significantly, Cuvier was not a member, although Darwin (see Famous Geologists: Darwin) and Lyell (see Famous Geologists: Lyell) were – where one should look for the opponents of his catastrophist and anti-transformist opinions. Even before his death, many opposed him, including Constant Pre´ vost (who was one of Lyell’s teachers), Jean-Baptiste d’Omalius d’Halloy (the father of Belgian geology), Jules Desnoyers, Andre´ de Fe´ russac, Marcel de Serres, and Ami Boue´ (one of the Society’s founders). In 1830, Ami Boue´ , expounding Adolphe Brongniart’s work on fossil plants, maintained that Cuvier’s ideas were contested or rejected by ‘the majority of geologists’, whose names he gave. Shortly before Cuvier’s death Boue´ declared himself to be radically opposed to ‘‘the idea put out by Mssrs Cuvier and Buckland on the universal Deluge, the universality of which was shown to be false by the most obvious facts’’ and also to ‘‘other opinions of M. Cuvier, namely... his hypothesis which, contrary to the natural order and to facts, admits universal cataclysms at several epochs prior to the Noachian Deluge’’. Ge´ rard-Paul Deshayes, another of Lyell’s tutors, recalled that in 1835 French zoologists were divided into two camps, one following Cuvier
and the other following Lamarck. In 1836, Lyell himself placed Cuvier and Lamarck on the same rung. Several members of the Society, including Boue´ , openly declared themselves to be supporters of Lamarck. Informed historians will not be surprised to read the evidence of Camille Dareste, who, in 1859, before Darwin became widely known, attested that the transformist theory of Lamarck had penetrated deep into the French scientific community. The domination of Cuvier after his death is a mistaken historical legend, which conscientious historians should not perpetuate.
See Also Biblical Geology. Creationism. Evolution. Famous Geologists: Darwin; Lyell. History of Geology From 1780 To 1835. Stratigraphical Principles.
Further Reading Boue´ A (1831) Compte rendu de la traduction allemande, par No¨ ggerath, des Re´ volutions du Globe, par Cuvier. Bulletin des Sciences Naturelles et Ge´ ologie 24: 129 130. Coleman W (1964) Georges Cuvier Zoologist: A Study in the History of Evolution. Cambridge, MA: Harvard University Press. ‘Collectif’ (1932) Centenaire de Cuvier, Archives du Muse´ um National d’ Histoire Naturelle. Paris: Masson et Cie. Cuvier G (1797) Tableau E`le´ mentaire de l’ Histoire Natur elle des Animaux. Paris: Baudouin. Cuvier G (1800) Lec¸ ons d’ Anatomie Compare´ e: 1. Paris: Baudouin. Cuvier G (1801) Extrait d’un ouvrage sur les espe`ces de Quadrupe`des. Journal de Physique 52: 253 267. Cuvier G (1812) Recherches sur les Ossemens Fossiles de Quadrupe`des: 1. Paris: De´ terville. Cuvier G (1812) Essai sur la Ge´ ographie Mine´ ralogique des Environs de Paris. Paris: Baudouin. Cuvier G (1817) Me´ moire pour servir a` l’Histoire et a` l’ Anatomie des Mollusques, Me´ moire sur la Scylle´ e. Paris: De´ terville. Cuvier G (1825) Discours sur les Re´ volutions de la Surface du Globe. Paris: Dufour. Cuvier G (1827) Rapport Historique sur les Progre`s des Sciences Naturelles depuis 1789. Paris: Imprimerie Imperrales. Dareste C (1859) Biographie de Lamarck. In: Hoefer JCF (ed.) Nouvelle Biographie Ge´ ne´ rale, 29, pp. 55 62. Flourens P (1859) Histoire des Travaux de Georges Cuvier, 3rd edn. Paris: Garnier. Hull D (1973) Darwin and his Critics: The Reception of Darwin’s Theory of Evolution by the Scientific Commu nity. Chicago: Chicago University Press.
184 FAMOUS GEOLOGISTS/Darwin Laurent G (1987) Pale´ ontologie et E´ volution en France, 1800 1860: De Cuvier Lamarck a` Darwin. Paris: Comite´ des Travaux Historiques et Scientifiques. Laurent G (2000) Pale´ ontologie(s) et e´ volution au de´ but du XIXe sie`cle: Cuvier et Lamarck. Asclepio 52: 133 212. Outram D (1980) Georges Cuvier, Vocation, Science and Authority in Postrevolutionary France. Manchester: Manchester University Press.
Rudwick MJS (1997) Georges Cuvier, Fossil Bones, and Geological Catastrophes: New Translations and Inter pretations of the Primary Texts. Chicago and London: The University of Chicago Press. Smith JC (1993) Georges Cuvier: An Annotated Bibliog raphy of his Published Works. Washington DC: Smithso nian Institution Press.
Darwin D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Charles Darwin (Figure 1) is chiefly remembered for his celebrated theory of the evolution of life forms and speciation, by means of natural selection. But his considerable contributions to geology should not be forgotten. Darwin, born in 1809, was brought up at Maer, Staffordshire, UK, son of a prosperous doctor. He attended school at Shrewsbury, and at age 16 proceeded to Edinburgh University to study medicine, but withdrew from the course because of his distaste for dissections and operations conducted without anaesthetics. He then went to Cambridge to take the standard arts degree, with a view to becoming a clergyman. Darwin’s interest in natural history developed while he was still at school and was furthered in Edinburgh by studies of plankton in the waters of the Firth of Forth. He attended some of Robert Jameson’s mineralogical/geological lectures, which were presented according to the ‘geognostic’ principles of the famous eighteenth-century Freiberg teacher of ‘Neptunist’ theory, Abraham Werner; Darwin found the ideas taught unacceptable and he discontinued his attendance. However, he may have learnt more geology there than he later acknowledged in his autobiography. Both Darwin and his father described his time at Cambridge as wasted, which it was, so far as theological training was concerned, but Darwin continued his private studies in natural history (especially beetle collecting) and became an epigone (and later, a personal friend) of the botany professor John Henslow. Henslow imparted some geological understanding to Darwin, having earlier done a fair amount of geological work, notably in Anglesey.
Beagle Voyage Enthused by the writings of Alexander von Humboldt, Darwin wanted to travel. In his last year at Cambridge, he planned an informal journey with friends to Tenerife, which necessitated his brushing up on geology, having earlier largely ignored the subject at Cambridge, being ‘‘so sickened with the lectures at Edinburgh’’. Henslow taught him to use a clinometer and gave him geological advice on his project. Possibly Darwin also attended some lectures of the geology professor Adam Sedgwick (see Famous Geologists: Sedgwick), and certainly participated in the field excursions Sedgwick ran, around Cambridge. Then, in 1831, at Henslow’s suggestion, Sedgwick took Darwin along on a short field excursion in North Wales as assistant and companion, also with the idea of teaching him the rudiments of field geology. Darwin was an apt pupil, but this was essentially
Figure 1 Charles Darwin (1809 82), from a photograph (1854?). Engraved for Harper’s Magazine, October 1884. Reproduced from Darwin F (1887), 1, Frontispiece.
FAMOUS GEOLOGISTS/Darwin
all the training he had in geology. Nevertheless, when later that year Darwin joined the Beagle on its epic journey and circumnavigation of the globe, he initially regarded geology as his principal scientific interest and even skill, having also done some solo field trips in south-west England. Darwin studied Charles Lyell’s Principles of Geology (see Famous Geologists: Lyell), the first volume of which (published in 1830) he had on board the Beagle; according to Darwin’s later published accounts, he immediately sought to apply his understanding of Lyell’s ideas to the observations made at St Iago, in the Cape Verde Islands, the Beagle’s first port of call. However, Darwin’s field notes made at a village called St Domingo, situated in a wadi-like inland valley, reveal that he was initially thinking in catastrophist terms, such as he might have picked up from Sedgwick. Darwin thought that the valley had been created by some ‘‘great force’’ followed by ‘‘the agency of large bodies of water’’, which moved large boulders and seemed also to have deposited material similar to diluvium (which he had encountered aplenty in Wales) near the coast. Subsequently, when Darwin published his geological observations, his initial catastrophist ideas were set aside, his Cape Verde observations being presented as supportive of Lyellian geology. Thus, Darwin reported evidence for uplift of land, in the form of a fossiliferous limestone about 40 feet above sea-level, with similar material forming at sea-level today. A volcano there also seemed to have depressed strata, but the published comparison of past and present processes was unquestionably Lyellian. Along the east coast of South America, Darwin made important collections of megafaunal remains. He made inland excursions towards the Andes and observed what appeared to be a ‘stepped’ structure for the Patagonian plains, indicating successive elevations. Near Tierra del Fuego, he saw snow-covered mountains, and glaciers descending close to sea-level. Travelling up the Chilean coast in 1835, he experienced a major earthquake at Concepcio´ n, with considerable uplift directly evident and in seeming accord with Lyellian geology. Darwin also travelled across Andean passes and formed the idea that the earth had been ripped apart along the line of the mountains and that igneous matter had entered the huge fissure. He even prepared a sketch map (in about 1834) of the southern part of the continent, distinguishing three main units: Tertiary strata; granites, gneiss, mica, slate, quartz rock, and clay slate; and lavas, tufas, and porphyries. A more detailed, also undated, map of the more southerly part of the continent from about the same period depicted seven units: 1. Granite, Mica slate; 2. Trappean rock and Porphyries;
185
3. Purple Porphyry and Tufa. Metamorphics; 4. Clay Slates; 5. Tertiary (Pliocene?); 6. d[itt]o—Recent; 7. Basaltic Lava. Crossing the Pacific, Darwin noted that some coral islands seemed to have been elevated, as indicated by the dead coral above sea-level. Elsewhere, there seemed to have been subsidence, but the corals were growing upwards at about the same rate as the land was sinking, so that fringing reefs were formed. This evidence cohered with Lyell’s idea that some parts of Earth’s crust were rising while others were sinking. In Australia, Darwin crossed the Blue Mountains and observed its great cliffs and valleys, but having only a restricted view of the topography, he mistakenly supposed that the valleys were produced by marine rather than by fluvial erosion. In Tasmania, he quickly got the hang of the geology around Hobart, and possibly, on the basis of ‘drop-stones’, had the idea of there having been a glaciation in what he referred to as Carboniferous times (actually Permian). On his return to England, Darwin soon became acquainted with Lyell and remained a lifelong friend; Darwin was elected a Fellow of the Geological Society and was soon on its Council, then became a Secretary and, later, Vice-President. In 1838, he published a general theoretical paper, On the Connexion of Certain Volcanic Phenomena in South America, and on the Formation of Mountain Chains and Volcanos, as the Effects of the Same Power by which Continents are Elevated. Published a year after he had come to accept ‘transformism’, this work was less ‘steady state’ than Lyell’s geology envisaged. Darwin linked earthquakes, elevations, and volcanic eruptions. He thought that (for the Andes, at least) there were repeated uplifts and intrusions along the axis of the range, followed by cooling and consolidation. Rejecting global contraction, differential sedimentary loads, or interplanetary forces, he had no concrete suggestion as to the cause of the elevations. But he was convinced that the uplift proceeded in small stages and was ongoing, rather than occurring in one great catastrophic episode. He concluded that ‘‘the configuration of the fluid surface of the earth’s nucleus is subject to some change, —its cause completely unknown, —its action slow, intermittent but irresistible’’. (Modern students of plume theory and the effects of actions occurring at the mantle/core boundary may find this remarkably prescient!)
Geological Publications and ideas on Glacial Phenomena Following the Beagle voyage, Darwin published three major geological books: The Structure and Distribution of Coral Reefs (1842), Geological Observations
186 FAMOUS GEOLOGISTS/Darwin
on the Volcanic Islands Visited during the Voyage of H.M.S. Beagle (1844), and Geological Observations on South America (1846). In addition to matters previously discussed, he distinguished in 1846 between stratification, cleavage, and foliation, but the distinction between cleavage and bedding was probably imparted to him by Sedgwick. Also, and importantly, in 1844, Darwin initiated for petrologists the idea of gravity settling, based particularly on his observations of igneous rocks in the Galapagos Islands; this was based on the idea that crystals that first form from a cooling magma may separate out and thereby alter the chemical composition of the remaining fluid, thus producing magmatic differentiation. Another important piece of work undertaken by Darwin post-voyage was his attempt in 1839 to explain the strange set of markings, the so-called Parallel Roads of Glen Roy, on the sides of Glen Roy in central Scotland. These controversial parallel and horizontal markings evidently marked former shorelines of some kind. In Darwin’s view, they represented different marine shorelines, being formed (by analogy with ideas developed in South America) by a succession of land elevations. Erratic granite boulders were also to be found, and Darwin ascribed their deposition to floating icebergs. Indeed, the whole situation was seen and interpreted in terms of what he had seen in the Tierra del Fuego area, with Glen Roy being in some ways comparable to the Beagle Channel. So the observations were thought to accord with the notion of subsidence of the land, associated with cold and extended glaciation in Scotland. On subsequent elevation and amelioration of the climate, the supposed shorelines and erratics would be exposed. Subsequently, in 1842, Darwin thought the glacial submergence and floating iceberg theories could also be applied in North Wales. Darwin’s ideas about Glen Roy were later superseded by the idea that the marks were due to glacial lakes: i.e., water ponded in the valley by glaciers blocking its mouth at different altitudes, water having escaped over different passes at different altitudes, according to the size of the barriers. Thus, there were several distinct former lake margins at different levels. Darwin later acknowledged the superiority of this theory, and called his Glen Roy paper a ‘‘great failure’’ in his autobiography.
Later Years, Evolution, and the Age of the Earth By the 1840s, Darwin’s health was deteriorating and he gave up substantial fieldwork: he was beginning to focus more attention on his grand theory of
evolution by natural selection, first adumbrated in 1837. His last geological paper proper (on the geology of the Falkland Islands) appeared in 1846, and that year he turned to a taxonomic study of barnacles, both modern and fossil, continuing this work at his home for eight years. His study of modern forms led to his discovery of males living as ‘parasites’ within the female forms, and also a gradation from hermaphrodite types, through forms with females having an ‘attached’ male organ and one that was parasitic but physically detached, to types whereby there was sexual dimorphism, but in which the males were ‘parasitic’ on the females. Thus, Darwin saw, in barnacles, evidence for the evolutionary emergence of sexual dimorphism. He then turned to the study of fossil barnacles, publishing a two-volume monograph (in 1851 and 1854). In the light of hindsight, it can be seen that these works were ordered (or the organisms classified) from an evolutionary perspective. When Darwin eventually published his Origin of Species in 1859, there were two issues of principal geological interest. First, he wanted to present to the public the idea of the history of living forms as being analogous in form to a branching tree; second, he had to deal with the problem of the age of Earth. There was also the problem of the origin of life and the apparent appearance of quite new forms from time to time, particularly the appearance in the Cambrian (or Silurian as he termed it then, following Roderick Murchison (see Famous Geologists: Murchison)) of quite well-developed forms apparently without ancestors. But the actual stratigraphic record showed anything but continuity or smooth transitions. Examples of trends, with a complete presentation of the various forms in an evolving continuum, or cases of branching and speciation, were conspicuous for their absence in the fossil record. Darwin sought to answer this difficulty by appealing to the incompleteness of the stratigraphic record: many pages of the evolutionary record were missing due to weathering and erosion, or had been destroyed by metamorphism. Moreover, intermediate forms might not all have been preserved at the same locality, so it would be unreasonable to expect smooth transitions in an ascending section. Similarly, the abrupt appearance of new types in strata was a problem for Darwin (and was used as an argument against him by his contemporaries, and by critics ever since). Again he appealed to the immensity of time, the imperfection of the geological record, and geologists’ incomplete knowledge of that imperfect record. Also, he pointed out that much time might be required to evolve some particularly advantageous character (such as the ability to fly), but once
FAMOUS GEOLOGISTS/Darwin
acquired, the increase of that character would be very rapid. So the fossil record might give the appearance of sudden changes; but the reality could actually have been one of continuous change. And sometimes anticipated ancestors might be found, as in the case of fossil whales then quite recently found in the Cretaceous. Darwin long sought the occurrence of fossil sessile barnacles and was delighted when some were eventually found in the Chalk, as he had expected would be the case one day. As to the absence of fossils older than the Cambrian (Silurian), Darwin thought that metamorphism might chiefly be responsible. He did not know that soft-bodied Precambrian fossils would one day be found; but it would have been in accordance with his expectations. As to the age of Earth, comparison of the thickness of preserved sediments with the rate of deposition seemed to reveal the immensity of time. Assuming the erosion of the valley of the Weald in Kent as being chiefly due to the action of the sea, and thinking of erosion as proceeding at 1 inch per century, Darwin gave a figure of 306 662 400 years for the formation of the valley. This was obviously a crude estimate. Darwin halved this estimate in the second edition of the Origin and subsequently withdrew it all together. Modern opinion has it that Darwin’s figure was much too large, but it is evident that he had a clear vision of the immensity of geological time, and thought that it could brush away many of the objections to his evolutionary theory. In his last major work, The Descent of Man (published in 1871), Darwin boldly applied his theory to humans, but said little about fossil forms, and the book had little geological content. He did, however, speculate that humans had first evolved in Africa, because that was where our nearest animal relatives were found; this suggestion is still thought to be correct.
See Also Biological Radiations and Speciation. Evolution. Famous Geologists: Lyell; Murchison; Sedgwick. History of Geology From 1835 To 1900.
187
Further Reading Barrett PH, Gautrey PJ, Herbert S, Kohn D, and Smith S (eds.) (1987) Charles Darwin’s Notebooks, 1836 1844: Geology, Transmutation of Species, Metaphysical Enquiries. London: British Museum (Natural History); Ithaca: Cornell University Press. Darwin C (1842) Geology of the Voyage of the Beagle: The Structure and Distribution of Coral Reefs. London: Smith Elder & Co. Darwin C (1844) Geological Observations on the Volcanic Islands Visited during the Voyage of H.M.S. Beagle. London: Smith Elder & Co. Darwin C (1846) Geological Observations on South Amer ica. London: Smith Elder & Co. (Also numerous later editions and different publishers.) Darwin F (ed.) (1887) The Life and Letters of Charles Darwin, Including an Autobiographical Chapter 3 vols. London: John Murray. Herbert S (1986) Darwin as a geologist. Scientific American 254(May): 116 123. Herbert S (1991) Charles Darwin as a prospective geo logical author. British Journal for the History of Science 24: 159 192. Herbert S (2005) Charles Darwin, Geologist. Ithaca and London: Cornell University Press. In press. Pearson PN (1996) Charles Darwin on the origin and diver sity of igneous rocks. Earth Sciences History 15: 49 67. Pearson PN and Nicholas CJ (2003) Charles Darwin’s geo logical observations at Santiago (St Jago), Cape Verde Islands. International Commission on the History of Geological Sciences 28th International Symposium. Trin ity College, Dublin, Ireland . . . Programme, Abstracts & Delegates, 41. Rhodes FHT (1991) Darwin’s search for a theory of the Earth: symmetry, simplicity and speculation. British Journal for the History of Science 24: 193 229. Roberts MB (2000) I coloured a map: Darwin’s attempts at geological mapping in 1831. Archives of Natural History 27: 69 79. Rudwick MJS (1974) Darwin and Glen Roy: a ‘‘great failure’’ in scientific method? Studies in History and Philosophy of Science 5: 97 185. Secord JR (1991) The discovery of a vocation: Darwin’s early geology. British Journal for the History of Science 24: 133 157. Stoddart DR (1976) Darwin, Lyell, and the geological sig nificance of coral reefs. British Journal for the History of Science 9: 199 218.
188 FAMOUS GEOLOGISTS/Du Toit
Du Toit J C Loock, University of the Free State, Bloemfontein, South Africa D F Branagan, University of Sydney, Sydney, NSW, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Alexander Logie du Toit, South Africa’s greatest and best-known geologist, started his career in 1903 and, over the next 20 years, mapped and investigated a large area of South Africa. He collected data simultaneously on the Dwyka Tillite, other Karoo rocks (formerly spelled Karroo), vertebrate fossils, the southern Glossopteris flora, Karoo dolerites and the Drakensberg basalts. During his career he expanded his interests to include other fragments of Gondwana. This led him to propagate the idea of the former existence of a southern supercontinent. He was very much aware of the pioneering work of his predecessors, especially Wegener (see Famous Geologists: Wegener), but he carried their theories further. After a short period of consolidation and publication in the early 1920s, he settled down to write his book on the ‘wandering continents’. His Our Wandering Continents (published in 1937) cast the gauntlet at sceptics and anti-drifters. Du Toit was vindicated nearly three decades later, when geophysical studies, mainly of the ocean floors, led to the theory of plate tectonics.
families then living in Cape Town. When he passed the matriculation examination in 1893 he was among the top scholars in Cape Colony. Du Toit was sent to the University of the Cape of Good Hope, where he passed the intermediate examination with distinction in 1894 and went on to obtain the Bachelor of Arts degree in mathematics and natural science the following year, again passing with distinction. Du Toit’s descent dictated that he should study in Scotland. He qualified in mining engineering at the Royal Technical College of Glasgow in 1899 and then studied at the Royal College of Science in London. In 1901 he was appointed a lecturer at the Royal Technical College of Glasgow and also at the University of Glasgow. His studies and travels in Scotland and further afield gave him an insight into the geology and stratigraphy of the northern hemisphere and into the nature, origin, and structure of fold mountains. This newly acquired knowledge later proved crucial in his theories on continental drift. He kept his contacts with Scotland and submitted a thesis on the copper–nickeliferous layered intrusion of Insizwa in 1910, for which he was awarded a DSc degree. In Scotland all was not study. Du Toit played the oboe, a hobby that he maintained for many years. He married Adelaide Walker in Glasgow and returned
Ancestry and Opportunities Alexander du Toit (Figure 1) was fortunate in birth and ancestry. Early in the nineteenth century, Alexander Logie from Fochabers, Banffshire, Scotland, served as an officer in the 72nd Regiment in South Africa. Captain Logie married Henrietta Elizabeth Susanna du Toit, a descendant of Francois du Toit from Lille in France, who arrived at the Cape of Good Hope in 1686 as a Huguenot. As the marriage was childless, the couple adopted the infant son of Stephanus Hendrik du Toit, the brother of Mrs Logie, and his wife, Barbara Stadler. The boy was christened Alexander Logie, but he retained his du Toit surname, growing up on the family estate on the outskirts of Cape Town. Alexander married Anna Logie, daughter of Robert Clunie Logie, a brother of Captain Alexander Logie. They had four children, one of whom was Alexander Logie du Toit, born at Rondebosch near Cape Town on 14 March 1878. His relatives included some of the most prominent
Figure 1 Portrait of Alexander du Toit. (Photograph repro duced by courtesy of the Geological Society of South Africa.)
FAMOUS GEOLOGISTS/Du Toit 189
with her to South Africa late in 1902, where a new world and a long career stretched before him.
The Formative Years The Geological Commission of the Cape of Good Hope, 1903–1912
The Geological Commission of the Cape of Good Hope, established in 1896 by the Cape Parliament, consisted of a veteran politician and senior civil servants. Du Toit assumed the post of geologist at the beginning of 1903, being listed on the scientific staff as Alex L du Toit BA FGS. When he left for his first field session, he was accompanied by the Commission Director, Arthur Rogers, who introduced him to the upper beds of the Cape Supergroup (Devonian) and the overlying basal units of the Karoo Supergroup (Late Carboniferous to Early Permian). This was du Toit’s first contact with the Dwyka Tillite, which is exposed in the south-western corner of the Karoo outcrop area in a desert environment. We can picture him walking on the unweathered outcrops with a huge number of loose erratics lying around, which he identified according to rock type. Later, he was to trace these erratics in the Northern Cape to actual outcrops of Precambrian rocks. Little did he know at that stage that he was destined, three decades later, to be an international expert on the Dwyka Tillite and its equivalents in other fragments of Gondwana. During his 9 years service with the Commission, du Toit mapped and studied the rocks and the strata in three areas. In the western corner of the huge area
underlain by rocks of the Karoo Supergroup, he unravelled the stratigraphy of the two basal units, namely the Dwyka and Ecca Groups, paying special attention to the Dwyka glacials. Du Toit also had an interest in the Karoo Dolerite (Jurassic) and the diatremes that pierce the Karoo beds. In the northeastern Cape, he covered an area containing rocks ranging from Early Precambrian to Quaternary. Diatremes and kimberlite intrusions again attracted his attention. Du Toit is, however, chiefly remembered as a field geologist for his detailed and accurate maps of the north-eastern Cape, where he concentrated on the Beaufort and Stormberg Groups and the plateau basalts of the Drakensberg. An image of du Toit as a competent field geologist now emerged. He was a wiry and energetic man, who could cover long distances on foot, on a bicycle, or, on occasion, on horseback. In the more open areas to the west, his caravan-like wagon was pulled by a donkey team (Figure 2). Mapping, often in areas where largescale maps showed farm boundaries only, was done using a small plane table and alidade. He was renowned in the geological community for being able to judge distances very accurately. When mapping dolerites, he could visualize an intrusion into Karoo beds as a three-dimensional body and hence accurately predict the locations of the dolerite outcrops. One of du Toit’s ways of winning the confidence of the local inhabitants was to encourage the infirm and to dispense aspirins or coloured pills to those who feigned or claimed illness. As his fame as a ‘doctor’ spread, he found that he could rely on the locals for advice on the geography of a mountainous area
Figure 2 During his first field excursions, du Toit was provided with a wagon and a team of donkeys. (Photograph reproduced by courtesy of the Natural History Division of the South African Museum, Iziko Museums of Cape Town.)
190 FAMOUS GEOLOGISTS/Du Toit
or the route of a cattle track through bush and over mountains. The results of du Toit’s labours in the field are to be found in the Geological Commission’s annual reports, and other observations were published in scientific journals. The 12 maps with which he was involved, either on his own or mainly in conjunction with Rogers, covered an area of 180 000 km2 and were a source of wonder to his successors. One of his publications, on the evolution of the river systems in Griqualand West, is consulted to this day by geologists prospecting for alluvial diamonds. In the first edition (1905) of his book on the geology of the Cape Colony, Rogers referred to ‘Gondwanaland’, the term introduced by Suess in his Das Antlitz der Erde (1885) (see Famous Geologists: Suess). When Rogers prepared the second edition, he invited du Toit to be his co-author. The second edition contained expanded versions of the Gondwana hypothesis and of South African stratigraphy. Between 1905 and 1910, du Toit worked mainly in the more arid northern parts of Cape Colony, on unfossiliferous sediments and various igneous rocks. Here, he became friends with, and was impressed by, the Dutch geologist Gustaaf Molengraaff. Du Toit’s reports on the nature and petrology of the kimberlite pipes, which were often richly diamond bearing, were widely read and used by prospectors and mining interests. The Cape Geological Commission was disbanded at the end of 1911, but du Toit had already collected in his memory and in notebooks a vast number of facts, observations, interpretations, and ideas. He was to prepare a great synthesis in the following decade. Geological Survey, 1912–1920
Du Toit was transferred to the Geological Survey of the Union of South Africa when the four colonial surveys or commissions were amalgamated in 1912, and he suddenly found that he was free to study geological problems over the whole of South Africa. By this time, du Toit had transferred his interests to minerals and rocks. He spent more and more time on the investigation of specific sites and less time on mapping. He participated in the activities of the Geological Society of South Africa and served as President in 1918 and again in 1928. His first book, A Physical Geography for South African Schools, appeared in 1912. In 1914 he visited Australia to attend the meeting of the British Association, where he met the Sydney professor T W Edgeworth David and was able to examine the rock succession in eastern Australia and the evidence for Late Palaeozoic
glaciation. The remarkable similarities between the records of events in two widely separated southern continents, with evidence of glaciation in Australia at the same time as that indicated by the Karoo rocks in South Africa, was striking. As the First World War had broken out, du Toit was called up to serve as a geologist and was charged with finding water for the Union Defence Force. When the South Africans invaded German South-west Africa (now Namibia), du Toit had to find suitable sites at which to drill for water in the desert. His military service must be seen as a bonus because he found time to study the basement rocks, the Late Precambrian Nama beds, and, more importantly, the basal Karoo rocks. Du Toit’s last publication while a member of the Survey was his monumental compilation on the Karoo dolerites. Department of Irrigation, 1920–1927
The Department of Irrigation requested du Toit’s services because it needed his expertise. His relatively brief period of service saw both tragedy and triumph. Adelaide du Toit died in 1923, leaving her husband and a grown-up son. Two years later, du Toit married Evelyn Harvey. Du Toit’s many reports from this period, now mostly filed away and forgotten, dealt chiefly with dam sites and geomorphology. Nearly two decades had passed since du Toit had started his career, and his observations and the synthesis of the facts were ready for a wider audience. When the South African Association for the Advancement of Science met in Durban in 1921, du Toit was invited to deliver the popular evening lecture, for which he chose to speak on land connections between South Africa and other continents. He presented evidence in the form of vertebrate life, the migration of vertebrates, palaeoclimates, volcanism, and fold mountains to an audience that included sceptics. His main thrust was a resume´ and analysis of the glacial deposits at the base of the Karoo. This was augmented by a map showing Gondwana at the close of the Carboniferous. In the same year, the Geological Society of South Africa published du Toit’s summary and analyses of Carboniferous glaciation in South Africa. The references to the pre-glacial topography, the direction of flow of continental ice-sheets, and the distribution of erratics placed southern Africa in the wider context of Gondwana. A later generation of stratigraphers provided evidence that the Dwyka glacials range in age from the Late Carboniferous to the early Permian. Other adventures followed. In 1923, a grant from the Carnegie Institution in Washington enabled du
FAMOUS GEOLOGISTS/Du Toit 191
Toit to visit Brazil, Uruguay, Argentina, and Chile, which were previously poorly mapped and not well understood. He carried out a remarkable amount of field exploration in South America, meeting local geologists, including his old friend David Draper from South Africa, who was briefly managing the Boa Vista diamond mine in Minas Geraes. Du Toit mastered the difficult literature, which was mainly in Portuguese, Spanish, and German. His tour allowed him to study Devonian beds, fold ranges, and, as might be expected, the Karoo equivalents. Back in South Africa, he was able to show Edgeworth David, en route to England, the key local sites relevant to the displacement hypothesis. It was a fruitful meeting, and both geologists strengthened their support for the notion of continental drift. The Carnegie Institution published du Toit’s Geological Comparison of South Africa with South America in 1927. The book contained, inter alia, a chart showing the stratigraphic column from the Devonian to the Jurassic for selected South American countries. The boundaries of the Afro-American landmass were shown to have a bearing on the displacement hypothesis, which was becoming more widely known through the English translation of Wegener’s Origin of Continents and Oceans (1924). By this time, du Toit had made contact with the Dutch geologist Willem van Waterschoot van der Gracht, who was then working in the petroleum industry in the USA. Van der Gracht had become an apostle of continental drift and persuaded the American Association of Petroleum Geologists to organise a conference in New York in late 1926 to discuss the theory. Despite the contributions of several supportive American speakers, including Reginald Daly, a prejudiced group, led by Charles Schuchert, condemned the theory out of hand. Unable to attend, du Toit was dismayed by the intolerant attitudes of some people and their personal attacks on Wegener and others. He sent a paper to support Van der Gracht’s publication of the proceedings, which appeared in 1928, and in the following years added further publications in journals such as the American Journal of Science; even as late as 1944 he made a rejoinder to G G Simpson regarding his ideas on Tertiary mammals and continental drift. For some years, du Toit had been planning to write a textbook on the geology of South Africa. The first edition, published in 1926, was the first synthesis of its kind. A second revised edition followed in 1939. Du Toit died before he could complete his revision of the third edition, but his old friend and colleague Sidney Haughton carried on the task and the book was published in 1954.
Years of Work and Wandering In 1927, du Toit was invited to join De Beers Consolidated Mines as a consulting geologist, specializing in diamondiferous kimberlite pipes and alluvial gravels. Once again, he could travel extensively in Africa, but now in an official capacity. These travels, to areas in which De Beers had an interest, afforded him many opportunities to study the local geology. He could also visit other countries: the USA and Canada in 1932; the USSR in 1937; and India in 1938. Little concerning the areas visited survives in published form. However, from a cache of photographs and other documents discovered recently, we know that he travelled widely in the western USSR, from the Urals down to the Ukraine and the coastal area of the Black Sea. Du Toit’s greatest contribution to geology, and also his swansong, was his book Our Wandering Continents (1937), in which he assembled the observations, deductions, comparisons, and syntheses of facts and theories of a lifetime. A brief description of the features of the book is in order. After a review in which he acknowledged the work of his predecessors, he referred to tectonism, volcanism, palaeoclimates, plant fossils, and geosynclines to explain and describe his grouping of the ancient continents and his theory of drifting. The distribution of glacials formed, as before, the core of his arguments. We should note that du Toit now used the term ‘Gondwana’ for the southern supercontinent. For the northern supercontinent he introduced the term ‘Laurasia’, derived from Laurentia (the eastern North American shield) and Asia. Additionally, the book contained a discussion of Arthur Holmes’s suggestion of fracturing of continental blocks by subcrustal convection movements to explain spreading. Du Toit also included a figure showing the development of continental rifting that owed something to Holmes’s famous figure of 1929, representing convection and continental fracture. But du Toit doubted whether subcrustal convection was ‘wholly competent to account for continental drift’. The evidence for the former linkage of the southern continents was illustrated in several convincing diagrams (Figures 3, 4, and 5), and the structural correspondences between western Europe and North America were also depicted, along with a suggested pattern for the opening of the North Atlantic (see Figures 6 and 7). Throughout the book, du Toit challenged the geological community to accept his theories, but he did not live long enough to witness the acceptance of continental drift nearly two decades later.
192 FAMOUS GEOLOGISTS/Du Toit
Figure 3 Analogies between South America and southern Africa, according to du Toit. (Reproduced from Du Toit AL (1937) Our Wandering Continents: A Hypothesis of Continental Drifting. Edinburgh: Oliver and Boyd.)
Figure 4 Structural features in ‘restored’ southern continents, Late Triassic and Rhaetic, according to du Toit (1937: 93).
FAMOUS GEOLOGISTS/Du Toit 193
Figure 5 Arrangement of continents and areas of glaciation in the Late Carboniferous, according to du Toit (1937: 76).
Figure 6 Structural analogies between North America and Europe, according to du Toit (1937: 145).
Retirement and Honours, 1941–1948
Figure 7 Evolution of the Atlantic Arctic rift, according to du Toit (1937: 222).
After retiring from De Beers, Alexander and Evelyn returned to Cape Town, where du Toit embarked on a frenzied round of activity, as his diaries for 1945 and 1946 reveal. He corresponded with friends and colleagues in South Africa and other countries, made social calls, and undertook short visits to the countryside. He continued writing scientific papers and assisted aspiring authors. His final diary entry (31 December 1946) contained a vague but ominous reference to a medical problem that was to carry him away on 25 February 1948.
194 FAMOUS GEOLOGISTS/Hall
During his lifetime, du Toit was awarded many medals and honorary doctorates from five South African Universities. His richly deserved Fellowship of the Royal Society of London was not awarded until 1943, possibly delayed by English prejudice against his continental drift ideas. The greatest honour of all came when the Geological Society of South Africa instituted the biennial Alex L du Toit Memorial Lecture Series in 1949. A few years later, his image appeared on a South African stamp. A collection of his notebooks is owned by the University of Cape Town. The diaries for 1905, 1906, 1908, 1931, 1945, and 1946 are kept by the South African Museum in Cape Town. A cache of photographs and documents, together with his academic gown, was recently donated to the Museum. His awards, medals, and certificates are in the hands of a grandson.
See Also Famous Geologists: Suess; Wegener. Gondwanaland and Gondwana. History of Geology From 1835 To 1900. History of Geology From 1900 To 1962. Pangaea. Plate Tectonics.
Further Reading Branagan DF (2004) The Knight in the Old Brown Hat: A Life of Sir T W Edgeworth David, Geologist. Canberra: National Library of Australia.
Du Toit AL (1921) Land connections between the other continents and South Africa in the past. South African Journal of Science 18: 120 140. Du Toit AL (1921) The Carboniferous glaciation of South Africa. Transactions of the Geological Society of South Africa 24: 188 227. Du Toit AL (1926) Geology of South Africa. Edinburgh: Oliver & Boyd. Du Toit AL (1927) A Geological Comparison of South America with South Africa. With a Palaeontological Contribution by F R Cowper Read. Washington: The Carnegie Institution. Du Toit AL (1937) Our Wandering Continents: A Hypoth esis of Continental Drifting. Edinburgh: Oliver & Boyd. Gevers TW (1949) The Life and Work of Alex L du Toit. Alex L du Toit Memorial Lecture 1. Johannesburg: Geological Society of South Africa. Haughton SH (1949 [1950]) Memorial to AL du Toit (1878 1948). Proceedings of the Geological Society of America: 141 149. Haughton SH (1949) Alexander du Toit 1878 1948. Ob ituary Notices of Fellows of the Royal Society of London 6: 385 395. Holmes A (1929) Radioactivity and Earth movements. Transactions of the Geological Society of Glasgow 18: 559 606. Rogers AW and du Toit AL (1909) An Introduction to the Geology of Cape Colony With a Chapter on the Fossil Reptiles of the Karroo Formation by Prof R Broom, 2nd edn. London: Longmans. Waterschoot van der Gracht WAJM (ed.) (1928) The Theory of Continental Drift: A Symposium. Tulsa: American Association of Petroleum Geologists.
Hall R H Dott Jr, University of Wisconsin, Madison, WI, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction James Hall (1811–1898) of New York was North America’s pre-eminent invertebrate palaeontologist and geologist of the nineteenth century. That he was a giant among early American geologists is evidenced by the facts that he served as President of the American Association for the Advancement of Science (1856), was a charter member of the National Academy of Sciences (1863), and was chosen to be the first President of the Geological Society of America (1889). Hall was also the best-known American geologist on the international scene in his time. As
early as 1837, he was elected to membership of the Imperial Mineralogical Society of St Petersburg. Later he was the Organizing President of the International Geological Congress meetings in Buffalo, New York (1876) and in Paris (1878); he was a Vice-President of the congresses in Bologna (1881) and Berlin (1885), and he was Honorary President of the Congress in St Petersburg (1897). Hall was elected Foreign Correspondent to the Academy of Sciences of France in 1884, being its first English-speaking member. It was primarily the Paleontology of New York, published in 13 volumes between 1847 and 1894, that initially brought Hall his fame. However, the broader community of geologists chiefly remembers him more for the curious theory of mountains presented in his Presidential Address to the American Association for the Advancement of Science in 1857.
FAMOUS GEOLOGISTS/Hall 195
Early Life and Education Hall was born near Boston in Hingham, Massachusetts on 12 September 1811. His parents had emigrated from England two years earlier, and James was their first of five children. The family was of modest means, but the young Hall was fortunate to have a gifted teacher in his public school, who stimulated an interest in nature. Through his teacher, James met several leading members of the Boston Society of Natural History. Having developed a strong interest in science, Hall was attracted to a new college in Troy, New York, which emphasized science and employed revolutionary new approaches to learning, with an active role for the student coupled with hands-on laboratory and field-trip instruction. This Rensselaer Plan was developed by Amos Eaton, with financial backing from his patron, Stephen van Rensselaer. Unable to afford commercial transportation, Hall walked the two hundred miles to Troy. At Rensselaer, he was instructed by Eaton and Ebenezer Emmons, and had for classmates such geologists-to-be as Douglas Houghton, Abram Sager, Eben Horsford, and Ezra Carr. Hall graduated with honours in 1832 and undertook a tour on foot to the Helderberg Mountains in south-eastern New York to collect Silurian and Devonian fossils. A job as a librarian allowed him to continue at Rensselaer for another year and to earn a Master of Arts degree with honours (1833). He then held an assistantship in chemistry for several more years. In 1838, he married Sarah Aikin, the daughter of a Troy lawyer; they had two daughters and two sons. Sarah died in 1895.
The New York Survey In 1836, the New York legislature authorized a 4-year geological and natural history survey; an extension of 2 years was later authorized. Four men – William W Mather, Ebenezor Emmons, Timothy A Conrad, and Lardner Vanuxem – were in charge of four respective districts, and Lewis C Beck was mineralogist for the survey. Botanist John Torrey and zoologist James De Kay conducted the biological survey. Hall was engaged to assist his former teacher, Emmons, in the Second District in north-eastern New York, where his first assignment was to study iron deposits in the Adirondack mountains. A year later the districts were revised; Conrad was appointed State Paleontologist, and Hall was put in charge of a new Fourth District in western New York, with former Rensselaer students Horsford, Carr, and George W Boyd as his assistants. When the survey
ended in 1841, only Hall and Emmons remained in New York. Hall became State Paleontologist, and Emmons became State Agriculturalist. Lardner Vanuxem, who had studied in France, had been instrumental in introducing to America the value of fossils for subdividing strata and correlating those of similar age from place to place based upon similar fossils. Meanwhile, Timothy Conrad had gained a reputation for studies of Cenozoic fossils of the coastal plain. Thus the survey had strength in palaeontology from the start, and its staff soon developed a New York stratigraphy that set the precedent for naming stratigraphical divisions after geographical localities. Young Hall’s career blossomed quickly after the monograph on the fossils and stratigraphy of the Fourth District was published in 1843. This and the other survey reports soon aroused much interest in Europe, where Palaeozoic fossils were being used to define stratigraphical subdivisions during the mid-nineteenth century. For example, Roderick Murchison’s Silurian System appeared in 1839 (see Famous Geologists: Murchison), John Phillips’s Palaeozoic Series was proposed in 1840, and Joachim Barrande’s monographs on the lower Palaeozoic fossils of Bohemia began to appear in 1852. These and other authors began corresponding with Hall, and European geologists began beating a path to Albany – most notably Charles Lyell (see Famous Geologists: Lyell) during several American visits in the 1840s. During a visit in 1846, Eduard de Verneuil, a close associate of Murchison, tried to persuade Hall not to introduce the name Cambrian to the New World, but rather to use only Silurian for the lowest Palaeozoic strata – a reflection of the famous Murchison–Sedgwick feud then raging in Britain. Hall, however, was not swayed, for he was a leading exponent of the widely held ‘nationalistic’ view that an American stratigraphical classification was best for America. As geological investigations in America began to mature, stratigraphical nomenclature was becoming important, especially for comparisons among different regions. Hall and others proposed that an organization be created to deal with nomenclature and other mutual problems, and so in 1838 in Albany the American Association of Geologists was created; the first formal meeting was held in Philadelphia in 1840. From this organization evolved in 1857 the American Association for the Advancement of Science, which was modelled on the British Association. Still later, the Geological Society of America was spawned in 1888 from a division of the American Association for the Advancement of Science. Hall was promptly elected President (Figure 1).
196 FAMOUS GEOLOGISTS/Hall
Figure 1 James Hall in 1856 at the peak of his career and only one year before his famous Presidential Address to the American Association for the Advancement of Science, in which he first proposed his theory of mountain formation. (From Clarke Geologist and Paleontologist, JM (1921) James Hall of Albany 1811 1898. C. Ayer Company Publishers.)
The Albany Training Ground In 1857, Hall constructed a substantial brick laboratory building, where he worked for the rest of his life. This Albany laboratory became a veritable training school for a host of budding geologists who would distinguish themselves in the history of American science. Although universities were beginning to offer formal instruction in geology during the midnineteenth century, there was practically no instruction in palaeontology. As a result, apprenticeship was the principal route into that field, and Hall’s laboratory was the place to apprentice. Among the many who profited from association with Hall were Charles E Beecher, Ezra S Carr, John M Clarke, Nelson H Darton, Grove K Gilbert, Ferdinand V Hayden, Eban N Horsford, Joseph Leidy, W J McGee, Fielding B Meek, Charles S Prosser, Carl Rominger, Charles Schuchert, Charles D Walcott, Charles A White, Robert P Whitfield, Josiah D Whitney, Charles Whittlesey, and Amos H Worthen.
Hall’s assistants learned more from him than just palaeontology, however, for they also experienced a strong, egotistical, and irascible personality. Although his sharpest attacks were reserved for his legislative enemies, most of his assistants were also treated to his notorious outbursts. In addition to throwing vituperative verbal daggers, he sometimes brandished menacingly a stout cane or even a shotgun, kept at the ready near his desk. Perhaps the most extreme self-righteous attack was upon James T Foster, a school teacher in Greenbush, New York. Foster had the audacity to publish a popularized geological chart in 1849. Hall was so outraged that he stole aboard a boat bound for New York City and threw the entire printing of the offensive chart into the Hudson River. He had quite a time fighting the subsequent libel suit, which entangled him, Louis Agassiz (see Famous Geologists: Agassiz), James D Dana, and several other notables for several years. Another celebrated example of Hall’s irascible temper involved the prominent British geologist Charles Lyell, during his first visit to America in 1841–42. At first Hall and others were greatly flattered by the attentions of their famous visitor, but Lyell’s insatiable questioning, which earned him the nickname ‘Pump’, and his copying of the Americans’ geological maps gradually provoked resentment and a fear of being pre-empted. In March 1842, an anonymous letter signed ‘Hamlet’ appeared in a Boston newspaper, charging Lyell with geological piracy. It was written by Hall after some of his compatriots criticized him for being too generous in sharing the results of his research with Lyell, especially by giving him a copy of his as yet unpublished Geologic Map of the Western and Middle United States. Needless to say, this letter cast a chill upon the Association of American Geologists’ meeting a month later, but Lyell participated as if nothing had happened. Although the charge was largely true, Hall was afterwards mortified by his rash act. For once, however, he managed to mend the damage done by his intemperate action, and he remained thereafter on good terms with Lyell. Almost as legendary as his paranoiac outbursts was Hall’s acquisitiveness for fossils. He employed every conceivable means to acquire outstanding collections. An effective technique was to flatter and invite collectors to work with him in Albany and to bring their collections. Commonly, however, when the apprentice moved on his collection did not. Hall was a workaholic who drove himself as mercilessly as he did his assistants. He could rarely say ‘no’ to even the most ridiculous schemes, and he ignored the entreaties of friends to ease his pace for the sake of his health.
FAMOUS GEOLOGISTS/Hall 197
Beyond New York As he completed his Fourth District studies, Hall decided to see how far the New York stratigraphical classification might apply beyond his state. In 1841, he made the first of several odysseys west. With geologist David Dale Owen he made a boat trip down the Ohio River to Owen’s base at New Harmony, Indiana, and, from there, he proceeded across Illinois to Missouri, Iowa, and Wisconsin. Hall was amply rewarded with evidence to support the extension of the New York stratigraphy in a broad way across the entire region. There were some significant differences, however, which he, and perhaps only he, could recognize. For example, he found that the Palaeozoic strata were much thinner to the west of New York and that there were important differences in sedimentary facies, with more clastic sediments in the east and more carbonate strata to the west. In effect, Hall had discovered the contrast between what would, much later, be termed the stable craton and the Appalachian orogenic belt. This trip also provided information to allow him to complete the Geologic Map of the Middle and Western States, which was incorporated into Hall’s Fourth District report of 1843 – the map that Lyell had used to help prepare his own geological map of the then United States, which was published in 1845 in Travels in North America. Hall’s finances were always tenuous. He was easily drawn into risky ventures and also had his salary cut, or even suspended, by a frequently hostile state legislature. At least once he had to sell some of his fossil collections in order to raise money. As his reputation grew, however, opportunities for temporary outside employment helped to tide him over his New York financial droughts. These ventures also allowed him to expand his knowledge widely. One of the first such ventures took him to the Lake Superior region in 1845 to examine copper deposits for a private company. In 1847, the Federal Government authorized a geological survey to evaluate the mineral resources of northern Michigan and Wisconsin. In 1850, Hall was engaged to provide his expertise on Palaeozoic stratigraphy and palaeontology for that survey. He made two brief trips to the area (1850 and 1851), from which he gained further insights into the stratigraphy of the Great Lakes region and added to his evergrowing fossil collections. Perhaps the most important result of his work for this survey, however, was the recognition of fossil reefs in the Silurian strata of south-eastern Wisconsin. This was the first recognition of ancient reefs in North America, and perhaps in the world. When asked to study fossils from western regions, which others had collected during various
expeditions, he willingly obliged. He recognized the first known Mesozoic fossils collected by John C Fremont in the 1840s. In 1853, he agreed to let his assistants Fielding B Meek and Ferdinand V Hayden go to the White River badlands of Nebraska Territory (now in South Dakota) to collect newly discovered Cenozoic non-marine invertebrate and mammalian fossils. Meek, whose artistic as well as collecting skills were vital to Hall’s enterprise, was glad to escape from his mentor for a few months. Eventually he extricated himself from Hall’s empire to join the new United States Geological Survey. Meek never forgave his perceived exploitation by Hall. When Iowa decided to undertake a geological survey in 1855 and needed a director, the Governor looked to New York, which had eclipsed all other states as well as the Federal Government in the calibre of its geological survey. Hall accepted the position with alacrity as his New York salary had been suspended in 1850 by a more than usually hostile legislature. Moreover, he welcomed the opportunity to obtain and study fossils from the new state. He soon suggested Amos Dean of Albany as the first Chancellor of the University of Iowa. Hall himself was identified as the first Professor of Geology but apparently he never lectured there. In fact, Hall mostly directed the survey from Albany and spent little time in Iowa. Four assistants did most of the actual work: Josiah D Whitney concentrated upon mineral resources, while Amos H Worthen of Illinois dealt with palaeontology, assisted by F B Meek and R P Whitfield. Hall knew that Worthen had the finest collection of crinoids in the country, so a condition of his employment was that Hall be allowed to describe them, which he did in the Iowa Survey report. Hall came to Iowa for the winter meetings of the legislature to lobby on behalf of the Survey, but payment of salaries was so erratic that he had to borrow money in Albany to keep the effort going. Finally in 1859 the survey was suspended, but two volumes had appeared in 1858. In 1857, Illinois undertook a geological survey, and Worthen was one of three applicants to direct it. Hall wrote a glowing endorsement of him, but he also supported the other two applicants. This lapse of judgement earned him the animosity of all three applicants, and, in the end, Hall was denied access to the fossils collected by the Survey, which was a great disappointment to him. While still working in New York and Iowa and for the Canadian Geological Survey, in 1856 Hall accepted an affiliation with Wisconsin. He joined a former Renssalaer colleague, Ezra Carr, now a professor at the University of Wisconsin, and Edward Daniels for this new effort. Hall devoted little time
198 FAMOUS GEOLOGISTS/Hall
to the Wisconsin initiative, so Carr and Daniels were really in charge. Whitney was engaged to study the lead deposits of south-western Wisconsin, and Charles Whittlesey was engaged to study the mineral deposits of northern Wisconsin. A large volume was published in 1862, but a hostile Wisconsin legislature abruptly terminated the endeavour because it judged the results to be insufficient. It cared only about potentially economic results, so a frustrated Hall and his assistant, Robert P Whitfield, published Wisconsin’s palaeontology within a New York report in 1867 (and again separately in 1871). This ingenious solution to a publication problem was typical of Hall. Much earlier he had circumvented a New York legislative edict to limit the number of expensive palaeontological monographs published simply by issuing several volumes as subdivisions of a single Part of the series, resulting ultimately in 13 separate monographs – at least twice the intended limit. Hall became involved in several other state surveys in various capacities, ranging from advising about personnel to acting as a consultant for palaeontology or the titular head of a survey. Included were surveys of Missouri (1853 and 1871) and California (1853–1856), the transcontinental railroad survey (1853–1857), and surveys of New Jersey (1854– 1857), Ohio (1854–1857), Texas (1858), Mississippi (1858), Michigan (1869–1870), and Pennsylvania (1870–1875). While this list is a testimony to his prominence, Hall’s contributions to these many surveys were minor except for the identification of fossils. Hall made his last trip to the Midwest in 1889, at the age of 77, while first President of the new Geological Society of America. His purpose was to obtain brachiopods by any and all means necessary for his latest project, namely to revise the description and classification of that great group of Palaeozoic fossils. In addition to successfully obtaining many specimens, he also met and lured to Albany a young Charles Schuchert of Cincinnati, who was destined to become his most famous prote´ ge´ and a professor at Yale. The ambitious brachiopod study culminated in the last volume, Part 8, of the Paleontology of New York, which appeared in 1894. During the completion of the brachiopod monograph, Hall had his last and sweetest wrangle with New York bureaucracy. The Executive Secretary of the Regents, which oversaw his programme, had become overly zealous in trying to impose strict accounting and efficiency procedures. Such a fuss developed that the legislature had to intervene. To resolve the fracas, it appointed crotchety old Hall as State Paleontologist and State Geologist for life, with complete managerial freedom. Doubtless the
legislators realized that Hall’s days were numbered, and, in fact, he died 3 years later. Hall must have recalled with satisfaction an earlier observation, when a particularly vicious political enemy died suddenly, that ‘‘Providence was usually on my side’’.
The Origin of Mountains Hall is most widely known for his theory of mountains, which embodied the concept of the geosyncline, a term coined not by Hall but by James D Dana in 1873. In his 1857 Presidential Address to the American Association for the Advancement of Science, Hall startled his audience with a discourse on the origin of mountains rather than speaking about palaeontology and stratigraphy. In stating that ‘‘the greater the accumulation, the higher will be the mountain range’’, he pronounced that a great thickness of strata was a prerequisite for a mountain range composed of folded strata. Hall rejected the then popular theories of mountains of Frenchman E´ lie de Beaumont – that mountains formed as a result of global cooling and contraction – and the American brothers William B Rogers and Henry D Rogers, who postulated that catastrophic wrinkling of the crust resulted from wave-like movements in a fluid subcrustal zone. Instead, Hall was influenced by a suggestion by the British astronomer John F W Herschel in 1836, which anticipated the modern theory of isostasy. Herschel argued that vertical movements of the crust are caused by changes in pressure and heat at depth, which in turn are the result of erosion and deposition at the Earth’s surface. The vertical adjustments towards gravitational equilibrium were accommodated by a pliable subcrust. The key element for Hall was the accumulation of thick sedimentary layers, which he imagined must depress the crust and, in the process, become wrinkled to form the structures seen in ranges such as the Appalachians. He envisioned compression of the upper layers and tension of the lower ones during subsidence – much as occurs when bending a ream of paper (Figure 2). In 1859, Hall published the following, in the most commonly quoted source for his theory, Part 6 of the Paleontology of New York (Volume 3: Descriptions and Figures of the Organic Remains of the Lower Helderberg Group and the Oriskany Sandstone. 1855–1859, pp. 70–73. Albany: New York State Geological Survey): ...[t]he line of greatest depression would be along the line of greatest accumulation [that is] the course of the ori ginal transporting current. By this process of subsidence ... the diminished width of surface above caused by this curving below, will produce wrinkles and folding of the [upper] strata. That there may be rents or fractures of
FAMOUS GEOLOGISTS/Hall 199
Figure 2 James Hall’s theory of down warping resulting from sedimentation. The upper layers are crumpled as their circumference diminishes, whereas the lower layers are broken by tension, which allows dykes to be intruded from below. Hall never published diagrams of his theory, so this was constructed from his verbal discussion. Modified with permission from Dott RH Jr (1985) James Hall’s discovery of the craton. In: Drake ET and Jordan WM (eds.) Geologists and Ideas: A History of North American Geology, pp. 157 167. Centennial Special, Volume 1. Boulder: Geological Society of America.
the strata beneath is very probable, and into these may rush the fluid or semi fluid matter from below, produ cing trapdykes, but the folding of strata seems to be a very natural and inevitable consequence of the process of subsidence.
A year earlier, in the report of the Iowa Survey (1858), Hall had also emphasized the contrasts in thickness between the Appalachian region and the Midwest, with detailed remarks about contrasting facies as well as thicknesses in various portions of the Palaeozoic successions of the two regions. Here, too, he included a brief summary of his theory of mountains by stating that ‘‘The thickness of the entire series of sedimentary rocks, no matter how much disturbed or denuded, is not here great enough to produce mountain features’’ (Vol. 1, p. 42). Hall was vague about the cause of mountain uplift. He simply ascribed it to continental-scale elevation, which he thought had no direct relation to the folding of strata within the mountains and which he did not attempt to explain. Contemporaries were quick to challenge him on this point, with Dana noting that Hall had presented ‘‘a nice theory of mountains with the mountains left out’’. Hall lamely denied that he ever intended to offer a complete theory of mountain building. His failure to publish the Presidential Address until 1883 may have been because of such criticisms, but his first priority was always palaeontology and he knew that the essence of his theory was to appear in both the Iowa and New York reports (as well as in an abstract in Canada) soon after his oral address. Hall’s contribution to mountain-building theory was marginal at best and was soon eclipsed by Dana’s
more profound and comprehensive contraction theory, which postulated that thick strata were a result of mountain-building processes rather than the cause. Nonetheless, Hall’s emphasis on a causeand-effect relationship between orogenic belts and very thick strata had a significant influence on three generations of geologists. By coining the term ‘geosynclinal’, which was later converted to the noun ‘geosyncline’, Dana formalized Hall’s demonstration that Palaeozoic strata are ten times thicker in the Appalachian Mountains than in the lowlands to the west (the craton). Even though Hall was wrong about the cause of mountain building, he was nevertheless the first person to underscore the profound stratigraphical contrasts between orogenic belts and what are now termed stable cratons. He drew attention at an early stage to large-scale stratigraphical patterns among some of the larger tectonic elements of the Earth’s crust and had other shrewd stratigraphical insights that were ahead of the times. By virtue of his breadth of experience in both the cratonic and the orogenic regions of eastern North America, he was uniquely equipped to see this fundamental distinction. Coupled with his prodigious contributions to palaeontology, this assured James Hall of a prominent niche in the history of geology.
See Also Analytical Methods: Geochronological Techniques. Famous Geologists: Agassiz; Lyell; Murchison; Sedgwick. Geological Surveys. Stratigraphical Principles. Tectonics: Mountain Building and Orogeny.
200 FAMOUS GEOLOGISTS/Hutton
Further Reading Clarke JM (1921) James Hall of Albany Geologist and Paleontologist, 1811 1898. Albany: Privately printed. Dana JD (1873) On some results of the Earth’s contraction from cooling, including a discussion of the origin of mountains, and the nature of the Earth’s interior. American Journal of Science, 3rd series 5: 423 495; 6: 6 14; 104 115; 161 172; 381 382. Dott RH Jr (1979) The geosyncline first major geological concept ‘Made in America’:. In: Schneer CJ (ed.) Two Hundred Years of American Geology, pp. 239 264. Uni versity Press of New England: Durham, New Hampshire. Dott RH Jr (1985) James Hall’s discovery of the craton. In: Drake ET and Jordan WM (eds.) Geologists and Ideas: A History of North American Geology, pp. 157 167. Centennial Special, Volume 1. Boulder: Geological Soci ety of America.
Fisher DW (1978) James Hall patriarch of American paleontology, geological organizations, and state geo logical surveys. Journal of Geological Education 26: 146 152. Hall J (1842) Notes upon the geology of the western states. American Journal of Science and Arts, 1st series, 42: 51 62. Hall J (1989) The Natural History of New York. Part 6. Palaeontology of New York. Vol 3. Descriptions and Figures of the Organic Remains of the Lower Helderberg Group and the Oriskany Sandstone. 1855 1859. Albany: New York State Geological Survey. Hall J (1883) Contributions to the geological history of the North American continent. Proceedings of the American Association for the Advancement of Science 31: 24 69. Hall J and Whitney JD (1858) Report on the Geological Survey of the State of Iowa. Des Moines: Legislature of Iowa.
Hutton D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The Scottish geologist, physician, farmer, philosopher, chemist, businessman, and industrialist James Hutton (1726–1797) is commonly regarded as the ‘founder of modern geology’, though a similar claim could be made for some others, and in some ways Hutton’s thinking was not at all modern by today’s standards.
Hutton’s Early Career and the Beginning of His Interest in Geology Hutton was born in Edinburgh, the son of a businessman who served for a time as City Treasurer. The young Hutton went to Edinburgh University at the age of fourteen, where he studied humanities, attended the mathematics lectures of Colin Maclaurin, and acquired a keen interest in chemistry. After his time as a student, he was briefly apprenticed to a solicitor, but eventually decided to study medicine. After taking the Edinburgh course, he went to Paris in 1747 and thence to Leiden, where he submitted a doctoral thesis in 1749. This dealt with the circulation of the blood and matters of human physiology and had a distinct chemical slant. The thesis title referred to the human body as the ‘microcosm’, which was traditionally regarded as having analogies with the Earth or with the whole
cosmos (the ‘macrocosm’). The thesis may have been the seed from which sprang Hutton’s later cyclic theory of the Earth. On returning to Britain, Hutton did not take up medicine. Instead, he went into partnership in an industrial process for extracting sal ammoniac (ammonium chloride) from soot. But Hutton sired a son, probably out of wedlock, and ‘tactfully’ left Edinburgh for several years to pursue a career in agriculture on two farms in Berwickshire, which he had inherited from his father. He wished to do his farming on a scientific basis, so he went to East Anglia to study the latest methods of agriculture, which he subsequently brought to Scotland. During his two years away Hutton travelled extensively and, as he became increasingly interested in the Earth, recognized the ubiquity and perpetuity of erosion and deposition and that sedimentary rocks were consolidated sediments. Hutton worked his farms himself and experimented with agricultural techniques. Under Maclaurin, he had become acquainted with the principles of ‘deism’, and he had apparently lost his Christian faith at an early stage of his life. According to the deist view, God had created the Earth ‘in wisdom’ as a suitable place for human habitation. The existence of God was not known by courtesy of Jesus Christ, the Bible, the Church, or any other agent of revelation, but by human reason. For Hutton, divine design was manifest in Nature itself, both in the way organisms functioned and were structured and in the way the Earth was apparently well ‘contrived’ for human existence (with air, water, soil, animals, plants, etc. all suited to us).
FAMOUS GEOLOGISTS/Hutton 201
But it was obvious that soil was constantly being washed into the sea, and, since it was essential for human well-being, it had somehow to be replenished. As a deist, rather than a biblical literalist, Hutton could take a grand view of time. The Earth could be millions of years old, but in that case the land would eventually be eroded to a plain and the good soil would end up as sediments in the seas. So Hutton asked himself how high ground could be regenerated to provide a source of new soil.
Hutton’s Theory of Cyclic Earth Processes In 1764, Hutton made a journey into the Highlands and began to collect geological information and specimens in a systematic manner. His farms were by then profitable and the sal ammoniac business was prospering. So he began to think of returning to Edinburgh, now as a gentleman–farmer. Probably in 1767, he rented out his farms and returned to Edinburgh (his old scandal had by then been forgotten or forgiven) to enjoy the pleasures of intellectual life in one of the great cities of the eighteenth-century Enlightenment. Among Hutton’s new friends were the economist Adam Smith, the chemist Joseph Black, who conducted experiments on heat, and the engineer and steam-engine inventor James Watt. It is likely that Watt’s engines encouraged Hutton to think of heat as an agent of geological change. Perhaps the Earth had a central source of heat that might somehow drive the cycle essential for a theory of the Earth that provided for a renewal of soil? The Earth’s internal heat could be analogous to the fire of Watt’s engine, which drove the complicated mechanism of the engine and the machinery of a factory. But Hutton did not imagine that the Earth’s internal heat was due to combustion. Hutton’s theory of the Earth was, then, developed as follows. The Earth, he thought, had a central reservoir of heat, the source or means of maintenance of which was unspecified. Rocks on the surface were broken down by weathering and erosion to form soils. Sediment was deposited in the seas by rivers, which also carved valleys. Sediments accumulated in layers on the ocean floors, and the lower layers were compressed and consolidated by the sediments deposited on top of them, assisted by the Earth’s internal heat. The rock-salt deposits of Cheshire seemed to Hutton to have been melted at some time. Likewise, the grains of sand in quartzites seemed to show evidence of fusion at their edges in the process of consolidation by heat.
In time, the consolidated materials, under pressure, might become so hot that they would melt. Veins of crystalline rock, dykes or sills, could be emplaced. Moreover, Hutton supposed, great masses of molten material (which we would call magma) could be intruded into the Earth’s crust, heaving it up. On cooling, this magma might crystallize to form subterranean masses of granite, which might subsequently be exposed by weathering and erosion. Thus the land would be renewed and Hutton’s Earth, ‘designed in wisdom’, would continue indefinitely as a place suited to human habitation. The upheaval of strata was confirmed by the presence of marine fossils in strata well above sea-level. However, at the time of the first public presentation of his theory, Hutton appeared to have personal knowledge only of mineral or metallic ore veins, not granitic veins, and he did not then describe any personal examinations of large granitic bodies. He went looking for these systematically only after the preliminary presentation of his ideas. Be that as it may, the whole process envisaged by Hutton was cyclic, for the upheaved strata would be eroded to form a new surface, on which other sediments might subsequently be deposited. So one might hope to find places where the lower layers were inclined to the horizontal and the overlying ones lie over them horizontally. Such a structure came to be known as an unconformity, and the subsequent discovery of unconformities was considered a triumph for Hutton’s theory, as he apparently had the idea of such structures before he actually saw them. An unconformity could be taken to mark the end of one cycle and the commencement of the next. Hutton’s cyclic Earth processes were continuous and open ended. He did not say that the Earth was infinitely old, but as he put it in a famous sentence: ‘we find no vestige of a beginning—no prospect of an end’ Hutton (1788 p. 304). His cycle has been called the ‘geostrophic cycle’ (see Figure 1, which explicates Hutton’s notion of unconformity). Hutton’s theory was formally read before the Royal Society of Edinburgh in 1785 and published in 1788. It appeared in expanded form in two volumes in his Theory of the Earth in 1795. Two further incomplete volumes remained unpublished in his lifetime, but the manuscripts were found in the nineteenth century and published in 1899 as Volume 3. This book described Hutton’s fieldwork after the presentation of his 1785 paper. Hutton’s 1785/1788 paper did not explain the Earth’s internal heat, but he tried to use field specimens to support its existence. Many of the materials that bind sediments together, such as calcareous spar, silica, etc., are not themselves water soluble.
202 FAMOUS GEOLOGISTS/Hutton
Figure 1 Representation of the ‘geostrophic cycle’. Reproduced by permission of The Geologists’ Association from Proceedings of the Geologists’ Association, Tomkeieff SI, Unconformity an historical study 1962, 73, pp. 383 417, fig. 6. ß 1962 The Geologists’ Association.
Therefore, Hutton argued, water could not have been the prime agent causing their consolidation. But heat could penetrate into bodies and, by fusion, could cause consolidation. So, he thought, heat must have been responsible for the penetration of sediments by flint nodules (which were injected while molten). Likewise, nodules containing crystalline spar that did not extend to their outer surfaces could not, Hutton supposed, have acquired the crystalline matter by transmission of solutions. Hutton also exhibited a specimen of ‘graphic’ granite, which contained crystals of quartz within feldspar within quartz. Such a texture could not, he maintained, have been produced by crystallization from aqueous solution. There were evidently gaps in the evidence for the cyclic chain of Hutton’s theory. Hutton argued that there had to be heat within the Earth and there had to be some means of elevation, even if he did not know precisely how that process worked. The hot interior was supported by the evidence of volcanoes, of course, and mines seemed to have higher temperatures at greater depths (but that was not proven by careful measurements until the nineteenth century).
Geological Evidence to Support Hutton’s Theory So Hutton’s 1785/1788 paper was not in itself sufficient to persuade all his auditors or readers. At the time of its presentation, he had not recorded observations of veins of granite penetrating other rocks, nor, so far as we know, had he discovered any unconformities (although they had been reported by others
without their theoretical significance being recognized). But following the public presentation of his ideas Hutton made excursions to various parts of Scotland to look for confirmatory field evidence. In September 1785, Hutton went into the Grampians to hunt for contacts between granite and surrounding rocks into which it might have been injected while in a fused state. He was accompanied by a friend, John Clerk of Eldin, who made excellent drawings of what they saw. Hutton thought that he would find the evidence he wanted to the west of the mass of Aberdeen granite. But he may have received some hints of where to look (possibly from Clerk), for they headed directly for the valley of the River Tilt, which runs north-east from Blair Atholl. Complicated outcrops of limestones and schists were found in the valley floor, and not far up the glen they came across fine exposures of granitic veins, which sometimes cut across the country rock and elsewhere could be seen anastomosing between, or across, the laminae of the country rock. Hutton got so excited that his guides imagined that ‘nothing less than . . . a vein of silver or gold . . . could call forth such strong marks of joy and exultation’! The granitic veins were also traced back to the large mass of granite on the north side of the glen. Hutton’s joy was, of course, due to the fact that he had found what he had predicted on the basis of his theory, and the geometry of the veins was compatible only with the granite having worked its way into the country rock from below and from the granite mass. In 1787, Hutton visited the Isle of Arran, which has a large mass of granite at its northern end, with
FAMOUS GEOLOGISTS/Hutton 203
Figure 2 Cross section of the northern part of Arran, drawn by John Clerk Jr (1787). Reproduced by permission of Sir Robert Clerk of Penecuik.
surrounding tilted-up layers of schist and beds of sandstones and other sediments that are stratigraphically above the schists (but lower in terms of altitude). A large block of schist traversed by granite veins was brought back to Edinburgh to convince critics of the virtues of his theory. John Clerk’s son (of the same name) (1757–1812) accompanied Hutton and produced a wonderful cross-section of the island, construed in terms of Hutton’s theory (Figure 2). This section is congruent with a geological map of the area, based upon modern knowledge, which shows a domed structure of schists and sedimentary rocks, disposed around a central core of granite. Also, at Loch Ranza on the northern tip of Arran, Hutton found his first unconformity, with the sedimentary strata (sandstones and limestones) lying over the inclined or almost vertical schists (Figure 3). Again he had found a state of affairs that he had predicted from his theory. A ‘swarm’ of basaltic dykes was also observed by the shore of the southern part of the island. The Loch Ranza unconformity was not, however, wholly convincing, for the rocks were obscured by vegetation. But on their way home the travellers found a much clearer example near Jedburgh, where the road ran by the banks of the River Jed and a section revealed a splendid view of Old Red Sandstone lying horizontally over the up-ended grey gritty sandstone that is now known as Silurian greywacke (which Hutton called ‘schistus’ although it was not a schist). The most famous discovery of an unconformity occurred in 1788. Hutton’s upland farm was situated on ‘schistus’, while his main farm was on soil derived from Old Red Sandstone. The contact between the two rock types ran northwards to the coast. Hutton must have been aware of the two rock types, which
Figure 3 Unconformity at Loch Ranza, Arran, as described by Hutton and figured by Sir Archibald Geikie. Reproduced from Hutton (1899). Theory of the Earth. . .Vol. III Edited by Sir Archibald Geikie, p. 235. London: Geological Society.
were similar to those that he had seen at Jedburgh. It seemed a good plan to examine the coast, where an unconformity might be exposed. Accordingly, with his friends John Playfair (1748–1819), Professor of Mathematics at Edinburgh, and a local landowner, Sir James Hall (1761–1815), a keen amateur scientist who did some of the first experimental geology, Hutton sailed along the Berwickshire coast, past the schist terrain to that of sandstone. As anticipated, they encountered an unconformable contact, at a place called ‘Siccar Point’, with Old Red Sandstone overlying the schist, as at Jedburgh, but exposed in such a way that the three-dimensional structure of the contact could be examined. The famous excursion was described in Playfair’s biography of Hutton. The three men were aware that if the Siccar Point exposure was interpreted through the lens of Hutton’s theory then it entailed the passage of a vast amount of time. The sediments of the greywacke were first deposited horizontally under the sea and consolidated by heat and the pressure of superincumbent material. Then the area was upheaved by forces acting from within the Earth, with hardening of the sediments by heat and pressure. The
204 FAMOUS GEOLOGISTS/Hutton
forces were such that the ‘schistus’ now stood almost vertically. But the strata were then subjected to weathering and erosion so that the upheaved rocks were reduced to an approximately level surface. In time, the now vertical strata subsided below the sea once more (by an inadequately explained process) and were covered by layers of sediment derived from adjacent high ground. Again there was consolidation, following which uplift occurred, exposing the strata to the elements once more, but without the overlying sandstones being folded or inclined. Thus the disposition of the rocks observed at Siccar Point could be understood – provided that time was unlimited. Playfair wrote in his biographical memoir of Hutton: Revolutions still more remote appeared in the distance of this extraordinary perspective. The mind seemed to grow giddy by looking so far into the abyss of time; and while we listened with earnestness and admiration to the philosopher who was now unfolding to us the order and series of these wonderful events, we became sensible how much further reason may sometimes go than imagination can venture to follow. Playfair (1805, p. 73)
Thus at Siccar Point Hutton provided evidence for (but not formal proof of) the Earth’s great age and the cyclic nature of geological processes. The locality has long been recognized as one of geology’s most significant field sites. Hutton acquired Playfair as a convert to his theory, and it was Playfair who popularized Hutton’s ideas – Hutton’s prolix style and confusing theory of heat (see below) did not gain him many adherents.
Hutton’s Later Work on the Theory of Heat In his old age, Hutton tried to give some kind of physicochemical explanation of the forces causing elevation and subsidence, but he had little success. He knew that bodies expanded when heated, and the kind of heat that produced this effect he called ‘sensible heat’. He also knew that when heat was applied to a solid it increased in temperature, but on reaching the melting point it would melt without changing temperature, even though it was still being supplied with heat. In the change of state, the heat supplied to produce melting was somehow hidden. Black had called this ‘latent heat’. But the nature of heat was uncertain. Hutton thought it was a kind of weightless ‘substance’. He knew that everyday objects have mass and that massive bodies are attracted to one another by
gravitation. But there also seemed to be repulsive forces at work, as for example when water is boiled: steam engines exert pressure in their cylinders. Today, we distinguish between radiant heat and heat transmitted by conduction. Hutton had no adequate concept of radiation, but he knew that heat from the sun shines on us, across space. He called it ‘solar substance’, and, though weightless, it somehow seemed to be absorbed by plants, though Hutton did not know how. Adding to the complications, Hutton accepted the old ‘phlogiston theory’ of combustion (which was collapsing at the end of the eighteenth century), according to which an inflammable material contains a weightless ‘substance’ or ‘principle’ called ‘phlogiston’, which is dispersed into the atmosphere during combustion. Hutton was inclined to suppose that ‘solar substance’ and ‘phlogiston’ were one and the same. (Actually, if one regards ‘phlogiston’ as energy, then some of the problems that Hutton was trying to understand fall into place for us.) Hutton grappled with such problems in two books: Dissertations on Different Subjects in Natural Philosophy (1792); and A Dissertation upon the Philosophy of Light, Heat, and Fire (1794). All his arguments cannot be followed here, but he tried out the idea that objects normally attracted one another according to the inverse-square law of gravitation. Thus he spoke of ‘gravitating matter’. At very close quarters, however, objects supposedly began to repel one another, according to a force law in which the distance between particles was raised to a power greater than two. The repulsive force (or ‘solar substance’) could supposedly take various guises: ‘sensible’ heat, manifested by expansion; latent heat; light; electricity; and phlogiston. So, when sediments were under extreme pressure, they might move from a compressive phase to an expansive (expanding) phase. Hence, in the geostrophic cycle, there could be alternating periods of contraction (compression or consolidation) and expansion (producing land elevation). Hutton’s theory depended on a balance of attractive (gravitational, cohesive, and concretive) and repulsive (specific, or sensible, and latent heats) forces. There could be different resultant states, arising from the forces producing elevation (expansion) and subsidence (contraction) at different times and places. But when, lacking the concept of energy, Hutton started talking about ‘solar substance’ in reference to solar radiation (as we would say) and thought that this ‘substance’ was immaterial, confusion and misunderstanding followed amongst his contemporaries; it is scarcely possible for us to make sense of his theory of heat.
FAMOUS GEOLOGISTS/Hutton 205
Hutton’s Legacy We can see, therefore, that, for all Hutton’s success in looking into the ‘abyss of time’ and his successful predictions of granitic veins and unconformities, his theory had significant lacunae, and the physical explanation of expansion and uplift was not really integrated into his Theory of the Earth and attracted little or no following. Expansion was the Achilles Heel of his theory, and the problem remained unsettled for generations. In the end, expansion and elevation were simply assumed by Huttonian cyclists. People eventually accepted Hutton’s cyclic theory, even though they could make little sense of the physical basis he envisaged. But acceptance took time, and geological theory was racked with controversy until well into the 1820s. The Professor of Natural History, Robert Jameson (1774–1854), gained control of geology teaching at Edinburgh, and even Hutton’s specimens, for many years. So Huttonian theory tended to be eclipsed in Scotland for decades, despite the best efforts of Playfair and Hall, who, after Hutton’s death, conducted experiments that sought to simulate the consolidation of sediments, aided by heat, and to show that limestone heated in a sealed gun barrel could be converted into something like marble without loss of carbon dioxide. In addition to having excellent ideas about weathering and erosion, the deposition and consolidation of sediments, rates of geological change, the immense age of the Earth, and arguments in favour of geological cycles supported by evidence of unconformities, Hutton clearly appreciated the fact that many valleys have been carved by the rivers that now flow in them. Through second-hand knowledge of the Alps, he suggested that glaciers might have been much larger in the past than at present and could have deposited large blocks considerable distances from where the rock types are found in situ. Thus it seems that he envisaged a former colder climate than at present (due to the mountains being higher and carrying more snow) and appreciated the enormous erosive powers of glaciers. In 1802, Playfair published his Illustrations of the Huttonian Theory of the Earth, which set out Hutton’s doctrines in improved literary form. Hutton had referred to the ideas of the Swiss geologist Horace Be´ ne´ dict de Saussure (1740–1799) about the transport of glacial debris by glaciers extended from the Alps to the Jura Mountains, and these ideas were given greater prominence by Playfair, who also wrote about the patterns of river drainage systems. But neither Hutton nor Playfair had the idea of an Ice Age. That came later, principally through the
advocacy of Louis Agassiz (1807–1873) (see Famous Geologists: Agassiz). Another Scottish geologist, the influential Charles Lyell (1797–1875) (see Famous Geologists: Lyell), accepted many of Hutton’s ideas and made them almost paradigmatic, handing them on to another Scot, Archibald Geikie (1835–1924), who coined the methodological maxim: ‘the present is the key to the past’. But that principle was already well established by Hutton. He used his knowledge of what he could see going on around him – on his farms and during his travels – to develop a theory about how the Earth operated as a system and how it might have been in the remote past. But Hutton’s cycles were not identical. There could be local variations from one phase to the next. The Earth had a history, while operating in a law-like manner, so as to be in a steady-state when viewed on a grand scale. We can also credit Hutton with advancing the concept known today as ‘deep time’ – and for doing so by geological reasoning.
See Also Famous Geologists: Agassiz; Lyell. History of Geology Up To 1780. History of Geology From 1780 To 1835. Igneous Rocks: Granite. Unconformities. Weathering.
Further Reading Baxter S (2003) Revolutions in the Earth: James Hutton and the True Age of the World. London: Weidenfeld & Nicolson. Dean DR (1992) James Hutton and the History of Geology. Ithaca: Cornell University Press. Donovan A (1978) James Hutton, Joseph Black and the chemical theory of heat. Ambix 25: 176 190. Gerstner PA (1968) James Hutton’s theory of the Earth and his theory of matter. Isis 59: 26 31. Gerstner PA (1971) The reaction to James Hutton’s use of heat as a geological agent. British Journal for the History of Science 5: 353 362. Hutton J (1788) Theory of the Earth; or an investigation of the laws observable in the composition, dissolution, and restoration of land upon the globe. Transactions of the Royal Society of Edinburgh 1: 209 304. Hutton J (1795) Theory of the Earth, with Proofs and Illus trations. London: Edinburgh: William Creech; London: Cadell, Junior, and Davies (republished in facsimile 1972). Hutton J (1899) Theory of the Earth. . .Vol. III Edited by Sir Archibald Geikie, p. 235. London: Geological Society. Hutton J (1997) James Hutton in the Field and in the Study edited by Dennis R. Dean: Being an Augmented Reprinting of Vol. III of Hutton’s Theory of the Earth (I, II, 1795), as First Published by Sir Archibald Geikie (1899). New York: Scholars’ Facsimiles & Reprints, Delmar.
206 FAMOUS GEOLOGISTS/Lyell
Jones J (1985) James Hutton’s agricultural research and his life as a farmer. Annals of Science 42: 574 601. McIntyre DB (1997) James Hutton’s Edinburgh: the histor ical, social and political background. Earth Sciences History 16: 100 157. McIntyre DB and McKirdy A (2001) James Hutton: The Founder of Modern Geology. Edinburgh: National Museums of Scotland (1st edn, 1997). Oldroyd DR (2000) James Hutton’s ‘Theory of the Earth’ (1788). Episodes 23: 196 202.
Playfair J (1805) Biographical account of the life of Dr James Hutton, F.R.S.Edin. Transactions of the Royal Society of Edinburgh 5: 39 99. Sengo¨ r AMC (2001) Is the Present the Key to the Past or the ) Past the Key to the Present? James Hutton and Adam Smith versus Abraham Gottlob Werner and Karl Marx in Interpreting History. Special Paper 355. Boulder: Geological Society of America. Tomkeieff SI (1962) Unconformity an historical study. Proceedings of the Geologists’ Association 73: 383 417.
Lyell D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Charles Lyell (Figure 1) was arguably the most important geologist of the nineteenth century, and his Principles of Geology (1st edn, 3 vols, 1830–1833; 11th edn, 1872) was a classic text that exerted much influence on the development of geology, as well as helping to shape the development of Charles Darwin’s thought. Lyell’s other major works were his Elements of Geology (titled Manual of Elementary Geology in some editions) (1st edn, 1838; 6th edn, 1865) and Geological Evidences of the Antiquity of Man (1st edn, 1863; 4th edn, 1873). He also published two books on his travels in North America. Lyell was born into a well-to-do family at Kinnordy House, Forfarshire, Scotland, but much of his youth was spent at the family’s second home in Hampshire (with a more agreeable climate). He attended a private school in Salisbury and then at Midhurst; thereafter he attended Exeter College, Oxford, where he studied mathematics and classics, but also became greatly interested in geology through the lectures of William Buckland, which students could attend as optional additions to their main curriculum. Lyell’s family was considerably interested in natural history, and during his vacations they travelled extensively on the continent. Lyell also made observations on the Kinnordy estate. Even while a student, he was elected Fellow of both the Linnean and Geological Societies. On leaving Oxford, Lyell started to train for the law at an office in London, but found the work uncongenial and complained of problems with his eyesight, and so, having (limited) independent means, he did not continue in this line of work. Rather, consorting with many of the leading geologists of the day, and
travelling widely, he became virtually a full-time gentleman-geologist, being elected to the Royal Society as early as 1826. Two years later when travelling on the Continent and meeting important figures in Paris, etc., he decided to give up legal work altogether. Eventually, he acquired significant income from his geological writings. In 1832, he married Mary Horner, daughter of Leonard Horner, himself a geologist and educationist, who had learned Huttonian theory in Edinburgh. The couple, who had no children, settled in London, where Lyell became established as one of its leading scientists. Buckland’s Oxford lectures were informative and entertaining. He taught the essentials of stratigraphy, and particularly William Smith’s idea that strata could be identified and correlated by their fossil contents (see Famous Geologists: Smith). But Buckland, in the religious atmosphere of Oxford, and trying to show that his science was compatible with the Bible, laid much emphasis on his studies of superficial deposits and cave remains (about which he was an authority) and sought to show that such materials
Figure 1 Charles Lyell (1797 1875).
FAMOUS GEOLOGISTS/Lyell 207
could be explained as the result of the Noachian Flood, which in terms of biblical chronology, occurred only a few thousand years ago. Thus, Buckland’s geology, which had sources in the work of Cuvier (see Famous Geologists: Cuvier), could (supposedly) lend material support to theological claims. But such a global flood would have been impossible according to the laws of nature as presently acting, and would be incompatible with geological processes seen at work today. Lyell travelled and thought much during the 1820s; in Scotland he visited such sites as Glen Tilt and Siccar Point (see Famous Geologists: Hutton). Though greatly stimulated by Buckland, he came to reject his idea on the great role ascribed to catastrophic floods, and leaned towards the geology of Hutton, his father-in-law-to-be having attended John Playfair’s ‘Huttonian’ lectures in Edinburgh. Abraham Werner’s Neptunist theories were rejected as being incompatible with the limited solvent power of water and with Hutton’s observations. In Forfarshire (which he mapped in 1824), Lyell saw marls being deposited, or already deposited, in fresh-water lakes fed by springs and associated with shells and plant remains. He knew that in the Paris area Cuvier and Brongniart had found similar fresh-water limestones, which, they thought, had no modern analogues. Thus Lyell’s thinking was turned towards the idea of explaining geological phenomena in terms of presently occurring processes. In Huttonian theory, an immense amount of time was required to make possible the great cycles of geological change that he envisaged. The question of time was one that Lyell, therefore, had to consider. Evidence for the Earth’s great antiquity was produced during Lyell’s journey to Sicily in 1828. He saw the huge still active volcano, Mount Etna, and it was evident that it was made up of successive lava flows. Historical information about recent flows gave an approximate idea of the rate of accumulation of the flows and the build-up of the mountain. The height of the mountain being known, one could thus form an approximate idea of its age. Further, Lyell examined shells in recent-looking strata lying below the lavas. Nearly all were still to be found today in the Mediterranean. So strata and shells of geologically recent appearance were in rocks that were very ancient in human terms, being older than flows from the volcano. (In a subsequent letter to his sister, Lyell offered that on a ‘moderate computation’ the shells might be 100 000 years old.) So if geologically recent rocks were ancient in human terms, rocks lower in the stratigraphic column must be exceedingly ancient. Evidently the Earth was of enormous age. In this argument, Lyell was assuming that the rate of flows
at Etna occurred at approximately equal rates. He was applying the principle that nature was uniform in her operations: what was later dubbed the principle of uniformity. Lyell was a ‘uniformitarian’ with regard to Etna, as he had been with respect to the lake deposits in Forfarshire. Also on his Italian journey of 1828, Lyell visited Pozzuoli on the coast near Naples. There he observed three standing columns of a Roman building, then thought to be an ancient temple. These had marks of the borings of marine organisms half way up, which suggested to Lyell that the land there had fallen below sea level since Roman times, and had subsequently risen; all this having happened without the columns toppling over. From this, Lyell inferred that the level of land was rising or falling in different places, as Hutton had previously proposed. Moreover, the processes were not sudden or catastrophic, but gradual. Following his return to Britain, Lyell began to write his major book, which sought to establish the working methods and procedures of geology. It was to give geology its proper method and fundamental principles: hence its title, Principles of Geology. These may be summed up by the adage (as later stated by Archibald Geikie) that ‘the present is the key to the past’. Also, for Lyell, geological processes were assumed to be ‘gradual’. Hutton’s geology envisaged grand cycles of rock formation, erosion, transport, deposition, consolidation, and subsequent elevation. The rocks of each cycle were not necessarily identical in any given place, and the geologist needed to work out the history of what had happened at each locality. But overall, the earth did not have an historical direction: it did not ‘progress’. Things were much the same in the past and present (humans excepted). Lyell’s views were much the same, but he placed more emphasis on fossils. He supposed that conditions were constantly changing at any given locality from one period to the next, because of the local changes of relative levels of land and sea. Climate could change too, according to whether more high land happened to be near the poles at a given time, or nearer the equator, the former state of affairs producing cooler conditions overall. So some forms would become extinct if they failed to meet the conditions of existence. On this basis, new types of organisms also needed to come into existence from time to time. Lyell presumed that they did so, even though he did not know how this occurred. Further, he assumed that the basic animal types had always been found on the earth. On this view, there was a gradual turnover of species. His model can be represented as shown in Figure 2.
208 FAMOUS GEOLOGISTS/Lyell
Figure 2 Diagram illustrative of Lyell’s theory of species change, with ‘random’ creations and extinctions.
According to this model, Lyell assumed that the further back the geologist explored in time, the fewer extant species there would be. In fact, nearly all species before the beginning of the Tertiary in Europe would now be extinct. Further the Tertiary could be subdivided according to its proportions of extant fossils. The subdivisions that Lyell proposed were: . . . .
Newer Pliocene 96% recent fossils Older Pliocene 52% recent fossils Miocene 19% recent fossils Eocene 3% recent fossils
Between the Secondary (later called Mesozoic) rocks and the Tertiary there was a period of nondeposition in Europe, so that Secondary fossils were now virtually all extinct. There had been a complete turnover of forms during the stratigraphic time-gap. Likewise, there was a large time-gap and turnover of forms between the Primary (Palaeozoic) forms and the Secondary types. Moreover, the further back one went through the Secondary epoch, the smaller was the percentage of forms resembling those at the top of the Secondary (i.e., in the Cretaceous); likewise, through the Palaeozoic. Lyell regarded marsupial remains found in Secondary rocks in the Stonesfield Slate near Oxford as vindication of his idea that the major animal types went back into the indefinite past.
That they could not be found in the most ancient rocks was due to the fact that they had been lost by erosion or metamorphism (an important notion that Lyell first explicitly enunciated). All this was compatible with Lyell’s uniformitarianism, but he had no modern empirical warrant for the supposition that new species could somehow come into being. Lyell’s ideas attracted much attention, though most geologists, looking at the stratigraphic record, found it hard to accept that there was no evidence of progress in the fossil record through time. On the other hand, his desire for geology to have its own procedures, with geological processes operating in accordance with the presently observable laws of nature, met with approval, as did his mastery of facts and grasp of palaeontological and stratigraphic detail. He wanted geology to be a science, sui generis, distinct from cosmology. Geologists, he thought, did not need a general ‘theory of the earth’ such as his 18thcentury predecessors had sought to offer (though Lyell’s cyclic geology was in fact fundamentally the Huttonian theory). But Lyell focused on stratigraphy and palaeontology, not ‘hard rocks’ such as granite. In 1831, following the success of Volume I of his Principles, Lyell obtained a chair in geology at King’s College, London, a newly founded Church of England establishment. By then, the putting aside of the Noachian Flood as a geological agent seemed not to concern the authorities, and Lyell’s religious view were considered ‘sound’. However, he only gave lectures there in 1832 and 1833. Ladies were allowed to be present for the first course, but were thought to present an undesirable distraction and their further attendance was terminated. In consequence, the attendance fell sharply, and Lyell decided that he was in part wasting his time there, so he resigned to return to his publishing and life as a gentleman geologist. In this he was eminently successful, continuing his extensive fieldwork, and involvement with the Geological Society and the British Association. Lyell served as President of the Geological Society in 1835–37, and again in 1849–51. Subsequently, he was knighted (1848), was awarded the Royal Society’s Copley Medal in 1858, and served as President of the BA in 1864. Lyell was seriously concerned with French geology. He acknowledged Cuvier’s mastery of palaeontology, but rejected his ‘catastrophist’ theory. Lyell’s Principles did much to counter this doctrine in contemporary Britain. In Paris in 1828, he met with the conchologist and palaeontologist Ge´ rard Deshayes, who assisted him in the identification and stratigraphic placement of the shells he collected that year. Lyell reacted negatively to the tectonic theory of Le´ once E´ lie de Beaumont (which envisaged mountain ranges as having been formed as a result of the
FAMOUS GEOLOGISTS/Lyell 209
Earth’s cooling and contraction) and significantly hindered its acceptance in Britain. Most importantly, Lyell gave close attention to the ‘transformist’ (evolutionary) theory of Jean Baptiste Lamarck in Volume II of Principles. Changing conditions cause new needs for organisms. To adjust to changing circumstances, organisms may alter their habits, and consequently their forms. These changes may be transmitted to subsequent generations, producing a gradual transformation of species. The first simple forms of life appeared naturally (without divine action) by spontaneous generation. Such ideas were rejected by Lyell over many pages. His principal objection was that the stratigraphic record did not reveal smooth transitions such as Lamarck’s theory would lead one to expect to find. But there were other objections, such as the inability to produce new species by breeding; and hybrids were sterile. Nevertheless, Lyell devoted much energy to thinking about what the concept of species meant, the ‘laws’ of distribution of species, and the extent to which they could or could not show modification due to different or changing circumstances. The problem of species and speciation was one of the main features of his book, and it set the scene for Darwin’s work, and his seeing his fundamental problem to be ‘the origin of species’ (see Famous Geologists: Darwin). A major problem for geologists in the first half of the nineteenth century was the large quantities of superficial deposits: gravel, tenacious clay containing unsorted rock fragments and fossil remains, and large boulders of rock distant from the nearest ‘solid’ outcrops of rock of that type. Such phenomena were eventually explained by the work of Louis Agassiz and his theory of an Ice Age (see Famous Geologists: Agassiz). In the early nineteenth century, these deposits were ascribed to the Noachian deluge or some like catastrophe, and William Buckland distinguished between ‘diluvium’ (Flood deposits) and ‘alluvium’ (materials deposited by rivers in the normal course of events). It was supposed that a great inundation(s) could have swept over the globe, even depositing the erratic boulders and marine shells loose at the tops of hills or mountains. But according to Agassiz, the better explanation was that there had formerly been a colder climate with the whole of northern Europe once covered by ice, which had transported boulders, ground up the underlying rock, and deposited it, along with river gravels, over the land. The ice also could have scratched the underlying rock and transported shells to hill tops. Agassiz lectured on this to the BA in 1840, and some geologists were converted to his ideas, including Lyell. His general theory was presumed to be capable of accounting for a period of extreme cold,
such as to cause widespread glaciation, if much of the high land at that time happened to be in the polar regions. But Lyell’s conversion was short lived. Agassiz’s theory seemed to take him too far from present analogies or present climatic conditions. So he adopted the theory that came to be called ‘glacial submergence’: there was a period of great cold, but not such as to produce an all-enveloping mass of land-ice. Rather, there was a general fall of land surface, causing marine submergence, accompanied by cooling, causing extension of ice-fields and the transport of boulders by drifting ice-bergs (hence the diluvial deposits are now generally called ‘drift’). During his North American trip of 1845 Lyell saw floating ice in the St Lawrence River, which modern observation seemed to account for the occurrence of erratics satisfactorily in accordance with his methodology. While Lyell later accepted Agassiz’s theory for the Alpine regions he never accepted the general land-ice theory, preferring the glacial submergence model. After Darwin returned from his Beagle voyage in 1836, he and Lyell became close friends, but during the years before the publication of The Origin of Species Darwin mostly kept his emerging transformist ideas to himself. Lyell was opposed to transformism for reasons that he developed back in the 1820s, and like many he was concerned about evolution’s implications for ‘revealed religion’ and social stability. In his Presidential Address to the Geological Society (1851), he spoke against evolutionary ideas. Man, he thought, was a very recent creation, subsequent to the mammoths. However, after Darwin revealed his ideas to Lyell about 1856, he was reluctantly converted and did his best to see the early publication of Darwin’s ideas in 1858. In his The Antiquity of Man (1863), Lyell set forth ideas on transformism and stated his acceptance of the Darwinian theory of evolution by natural selection (though he represented it as a ‘modification’ of Lamarck’s doctrine). In the 1850s, Lyell had devoted a considerable amount of travel and fieldwork to the study of ancient humans, which was consistent with his general interest in the younger parts of the stratigraphic column. By that time, considerable numbers of cave deposits and flint implements had been discovered, as well as some human-like remains, notably the Neanderthal skull, found near Dusseldorf in 1857. This seemed, according to Thomas Henry Huxley’s description, which Lyell quoted, to be intermediate between that of a modern human and a chimpanzee’s. But Lyell cautiously (and rightly) stated that ‘‘it is at present to too exceptional, and its age too uncertain, to warrant us in relying on its abnormal and ape-like characters, as bearing on the question whether the farther back we trace Man into the past, the more
210 FAMOUS GEOLOGISTS/Murchison
we shall find him approach in bodily conformation to those species of the anthropoid quadrumana which are most akin to him in structure’’ (Antiquity, p. 375). Lyell also reported on the recently discovered Archaeopteryx, which might seem to be a missing link, but he also deferred to the anatomist Richard Owen’s opinion that it was actually a bird. Thus, Lyell supported Darwin’s evolutionism in a way that was valuable to its acceptance. But at the same time he did not push all the evidence to what we might regard as its logical conclusion. On reading Lyell’s works, one is struck by his mastery of exposition and his command of the literature, especially in stratigraphy. His influence was very great, both in his own day and subsequently. There is, however, ambiguity in the concept of ‘uniformitarianism’ (gradualism, steady-statism, naturalism, and ‘actualism’ – or the idea that modern, actually observable, processes should be used to provide geological explanations). Lyell held to all these positions. Modern geologists commonly make obeisance to uniformitarianism, without making the foregoing distinctions. Modern geology does not necessarily adhere to any of them, except in its rhetoric; for Lyell convinced people that his approach was the right one to adopt for geology to be regarded as a science.
See Also Famous Geologists: Agassiz; Cuvier; Darwin; Hutton; Smith. History of Geology From 1780 To 1835. History of Geology From 1835 To 1900.
Further Reading Gould SJ (1987) Charles Lyell, historian of time’s cycle. In: Gould SJ (ed.) Time’s Arrow Time’s Cycle: Myth and Metaphor in the Discovery of Geological Time, pp. 99 179. Cambridge (Mass) and London: Harvard University Press. Hooykaas R (1963) Natural Law and Divine Miracle: The Principle of Uniformity in Geology, Biology and Theology. Leiden: EJ Brill. Lyell C (1997) Principles of Geology edited with an intro duction by James A. Secord. London, New York, Ring wood, Toronto and Auckland: Penguin Books. British Society for the History of Science (1976) The British Journal for the History of Science: Lyell Centenary Issue 9(2). Rudwick MJS (1969) Lyell on Etna, and the antiquity of the Earth. In: Schneer CJ (ed.) Toward a History of Geology, pp. 288 304. Cambridge (Mass) and London: The M.I.T. Press. Rudwick MJS (1971) Uniformity and progression: reflec tions on the structure of geological theory in the age of Lyell. In: Roller DHD (ed.) Perspectives in the History of Science and Technology, pp. 209 227. Norman: University of Oklahoma Press. Rudwick MJS (1978) Charles Lyell’s dream of a statistical palaeontology. Palaeontology 21: 225 244. Rudwick MJS (1990) ‘‘Introduction,’’ Principles of Geology, First Edition [in Facsimile] Volume I Charles Lyell, pp. vii lviii. Chicago: University of Chicago Press. Wilson LG (1972) Charles Lyell: The Years to 1841. New Haven: Yale University Press. Wilson LG (1998) Lyell in America Transatlantic Geology, 1841 1853. Baltimore and London: The Johns Hopkins University Press.
Murchison D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Roderick Murchison (Figure 1) was the eldest son of a wealthy Scottish landowner at Tarradale estate, Ross-shire. Though born in Scotland, and always emphasizing his Scottish ancestry, he spent most of his career in England and spoke with an English accent. Following his father’s death and mother’s remarriage, Roderick was sent to school at Durham, aged 7 years, soon forming the ambition to be a soldier. At 13 years old, he attended the military college at Great Marlow where his training gave him a good ‘eye for country’. He was soon involved
in the ‘Peninsula War’ in Portugal, fighting at the Battle of Vimieira, aged only 16 years. From this victory, his unit moved into Spain where things went badly for the British army in winter conditions, with forced marches, defeat in the Battle of Coruna, and withdrawal in disarray. After a spell in Sicily, Murchison was posted to Ireland, where he led a dissolute and expensive life, and later likewise in London as a half-pay captain. With the end of the wars, he was fortunate to meet a general’s daughter, Charlotte Hugonin, 3 years his senior, and they were married in 1816. They then took a leisurely tour through France, Switzerland, and Italy, and under his wife’s influence his self-education began, learning French and Italian, visiting museums and galleries and some scientists and scientific institutions.
FAMOUS GEOLOGISTS/Murchison 211
Figure 1 Roderick Murchison (1792 1871).
On their return, the Tarradale property was sold and the couple (who remained childless) resided for some years in Barnard Castle, County Durham, where Murchison devoted himself to the sporting country life. However, the two also participated in local gatherings of literary and scientific people, and met Sir Humphry Davy. Charlotte had interests in botany and mineralogy, but her husband’s foxhunting passion continued and they moved to Melton Mowbray in the hunting shires. However, urged by his wife and Davy, Murchison determined to take up geology, moving to London in 1824 and attending chemistry lectures. He joined the Geological Society, went on field trips with Buckland and others, and began serious study of the science. Work in southern England yielded Murchison’s first paper in 1825 (in which year he was elected to the Geological Society; and in 1826 to the Royal Society!), he journeyed to Scotland in 1826 and 1827, the former trip being made to Jurassic strata at Brora, following instruction on Secondary stratigraphy from William Smith (see Famous Geologists: Smith), whom he visited in Scarborough; the latter trip being with Adam Sedgwick (see Famous Geologists: Sedgwick). Murchison also visited the Continent with Lyell (see Famous Geologists: Lyell). Having gained some knowledge of how to study older rocks with Sedgwick, from 1831 Murchison and his friend
and mentor determined to study the then rather littleknown rocks of Wales; the so-called ‘Transition Series’ of Werner’s Neptunist geology. He began to work from the known base of the Old Red Sandstone in the Welsh Border country, from the Wye Valley through to Cheshire. Historians have rather detailed knowledge of Murchison’s travels and scientific work as his field notebooks have been preserved, together with much correspondence. There is also a multi-volume ‘journal’, based on the notebooks, copied out by an amanuensis in Murchison’s old age, which was intended for biographical purposes, sometimes being judiciously ‘improved’ by Murchison to give a favourable view of his accomplishments and ideas. It is known, then, that in 1831 Murchison and Sedgwick planned to work out the geology of the Welsh (and Border region) Transition Series but, Sedgwick being otherwise occupied, Murchison set out alone that year, and Sedgwick arrived later, starting in North Wales, briefly with Darwin as an assistant (see Famous Geologists: Darwin). Near Ludlow, Murchison found richly fossiliferous rocks and the structure was made out successfully. He worked in his area for the next few seasons, subdividing the ‘Upper Grauwacke Series’ into the ‘Ludlow Series’, the ‘Wenlock Limestone’, the ‘Horderley and May Hill Rocks’ (later called the Caradoc Series), and the ‘Builth and Llandeilo Flags’, the first two being placed in the Upper Silurian and the latter two in the Lower Silurian (1835). The name Silurian was coined for a new geological system after the Silures tribe that formerly inhabited that part of Britain. The arrangement Murchison envisaged is shown in Figure 2, reproduced from his great treatise, The Silurian System (1839, p. 196). This magnum opus provided immense detail concerning the different units, figures of their characteristic fossils, and a valuable map of the geology of his Silurian ‘domain’. In 1835, Sedgwick and Murchison introduced the terms Cambrian and Silurian, though the word ‘System’ was not used at that time. Moreover, the boundary between the two was not then clearly defined, though following their fieldwork of 1834 Murchison stated that the upper and lower rocks had been ‘dovetailed’ in a manner that was satisfactory to both geologists. Unfortunately, this proved later not to have been the case and in the years that followed a serious controversy developed between the two former friends (see Famous Geologists: Sedgwick, Palaeozoic: Ordovician). Murchison’s approach to geology was considerably shaped by his military background. He came to regard ‘his’ Silurian System as personal territory, and the more parts of a map that could receive
212 FAMOUS GEOLOGISTS/Murchison
Figure 2 Arrangement of Silurian Strata, according to Murchison (1839).
Silurian colours the greater was his satisfaction! Indeed, he became known as the ‘King of Siluria’. From the geological point of view it was evidently necessary to see whether the Silurian System, established in the Welsh Border area, was of general application, preferably worldwide. This grand task of spreading Silurian colours was undertaken personally by Murchison, in western and eastern Europe, the vast tracts of western Russia, and in Scotland in his old age. Other geologists were encouraged to find Silurian formations elsewhere, where Murchison’s own feet had not trodden. This expansion of Siluria also occurred in time as well as space, as Murchison sought to extend the Silurian down into the regions of Wales, where Sedgwick had established the Cambrian. So, when Murchison produced a condensed version of his stratigraphy in Siluria (1854), a name that suggested a kind of ‘kingdom’, it pushed down into Sedgwick’s territory calling his rather unfossiliferous rocks of North Wales ‘Lower Silurian’. The book was subtitled The History of the Oldest Known Rocks Containing Organic Remains. That is, Murchison claimed all the Palaeozoic rocks below the Devonian as belonging to his system. This state of affairs arose in part because, after Sedgwick had put forward his Cambrian System, he did not there and then describe its characteristic fossils, and when this was eventually done some of his types had already been classified as Lower Silurian by Murchison. (Early on, Murchison divided his system into Lower and Upper divisions.) In The Silurian System, Murchison allowed the existence of fossil-containing Cambrian rocks below the Silurian, but it was supposed that they were linked by ‘passage beds’ to the Llandeilo. So the possibility of territorial and temporal expansion was already there. It occurred again in Russia (see below). In Siluria it was stated that the fossiliferous Cambrians were lateral extensions (undulations) of Silurian strata (an opinion said to have been reached in 1841), and all that was left on the map of Sedgwick’s Cambrian was the apparently unfossiliferous rocks of the Harlech Dome area, some on the southern side of the Menai Strait between Anglesey and the rest of North Wales, and
the unfossiliferous Longmynd rocks near Church Stretton in Shropshire. The remainder of Sedgwick’s ‘Cambria’ was now depicted in Silurian colours. It was this encroachment, which had been going on through the 1840s, that so incensed Sedgwick (see Famous Geologists: Sedgwick). Murchison did well out of it all. He was awarded the Royal Society’s Copley Medal in 1849, having been knighted in 1846. But Murchison did not gain his honours lightly. After visiting German colleagues, he was in Russia in 1840 and 1841; Poland in 1843; Germany and Russia in 1844 and 1845; Scandinavia in 1844 and 1845; France and Germany in 1839 and 1843, and again in 1846 and 1847; and in Italy and Switzerland in 1847 and 1848. There were also journeys in Britain. In addition, Murchison served as General Secretary of the British Association, President of the Geological Society in 1841–1843 and of the Royal Geographical Society in 1843–1844. He also maintained an expensive but hospitable life style in Belgravia, London. Murchison’s energy was truly remarkable. Murchison’s journeys in Russia, conducted with the French stratigrapher Philippe Edouard de Verneuil and the Russian zoologist Count Alexander Keyserling, were, for their time, of epic proportions. From St Petersburg in 1840, they travelled up to Archangel, then by indirect route to Moscow, and back to St Petersburg. Though much of the terrain is covered by drift and offers few good sections, information about Carboniferous, Devonian, and Silurian strata were obtained, partly from informants with knowledge of wells, etc. The area under St Petersburg is remarkable for having Cambrian clays, which by their fossils were construed by Murchison as Lower Silurian. In The Geology of Russia (see below) he stated that the clay contained fossils ‘‘belonging to the very oldest known Silurian or protozoic type . . . [and was] the true base of the Palaeozoic series, as indicated by a gradual dwindling out of animal life in the deposit of a region, where no eruptions ha[d] taken place, and where the strata are wholly unaltered’’. A further notable discovery, agreeable to the existence of the Devonian as a System, was the discovery of Old Red Sandstone fish in rocks that otherwise resembled the Magnesian
FAMOUS GEOLOGISTS/Murchison 213
Limestone of England or the Zechstein of Thuringia, but were unlike the Old Red Sandstone rocks of Scotland. This confirmed the idea of the Devonian as a palaeontologically characterized system, which had different lithologies in different localities (see Palaeozoic: Devonian). Early in his journey of 1841, Murchison met Czar Nicholas I in St Petersburg and established a good rapport with him. He was ‘duchessed’ by the Russian aristocracy, and formed a high opinion of it and of Russia. From Moscow, the geologists travelled east to Perm, to the west of the Urals, and thence further east to the point where they could see the plains of Siberia. Turning south on the western side of the mountains, they reached Orsk, and then headed west again, crossing the Volga and reaching the Sea of Azov, before going north again to Moscow and St Petersburg. It was on the basis of the rocks in the region of Perm that, later that year, Murchison proposed a new system, the Permian, after the ancient kingdom of Permia. Silurian, Devonian, and Carboniferous rocks were confirmed. The travellers saw considerable mining activity in the Urals. A central granitic nucleus appeared to be flanked by Silurian, Devonian, and Carboniferous strata. The Urals were eventually crossed and re-crossed on seven parallels between 60 and 54 N. As he had done for the Silurian and Devonian, Murchison named the Permian after a region with strata containing characteristic fossils. On returning to St Petersburg, Murchison presented the Czar with a geological map, reports on the coal deposits of the Donetz Basin, and information about the alluvial gold of the Urals. It was a highly successful ‘campaign’ and Murchison’s self-esteem rose to new heights. It was further fortified on his return to Britain by finding that the officers of the Geological Survey were obtaining results in Wales that seemed to support Murchison’s views about the Silurian vis-a`-vis the Cambrian. In 1843, Murchison returned to eastern Europe, where he compared the Tatra Mountains of Poland with the Urals, and met von Humboldt in Berlin and the notable Palaeozoic palaeontologist Joachim Barrande in Prague. The same year Keyserling pursued the Russian researches in the northern Urals. In 1844, Murchison paid visits to Denmark, Norway, Sweden and Russia, and was able to find the basement rocks in Scandinavia underlying the Palaeozoics. In 1845, he was again in the north, receiving from Czar Nicholas the award of the Great Cross of St Stanislaus, for his notable contributions to Russian geology, and geologizing in the Baltic region. The Geology of Russia in Europe and the Ural Mountains, published in 1845 with Keyserling and Verneuil
as co-authors and incorporating information from many others, was a monumental work, providing accounts of the geologists’ journeys and descriptions of the geology of the half-continent. There were copious illustrations, including many sections and two large coloured geological maps; also descriptions and beautiful figures of fossils. Additionally, the book contained considerable theoretical discussion. It marked Murchison as the master stratigrapher and geological traveller of his age. Besides becoming a leading geologist, Murchison was also active in encouraging geographical exploration, which he saw as essential to the expansion of the British Empire. As a man of influence, he interested himself in the geological appointments being made in the colonies, and his powers of patronage were considerable. For example, he assisted in the placement of Frederick McCoy in a chair at Melbourne University, which may have seemed advantageous to Murchison as it got one of Sedgwick’s main allies out of the country! By ‘placing’ or maintaining contacts with people in different parts of the world, Murchison also gained advantage by having information channeled through his hands. He was gratified to have numerous topographical features named in his honour. Murchison’s influence became so great that he even felt qualified to offer opinions about the geology of countries that he had never visited. For example, he thought it likely that gold might be found in eastern Australia, adjacent to the hills of the ‘Australian Cordillera’ (mostly a rather inconspicuous topographic feature), on the basis of examination of some nonauriferous rock specimens shown him by the Polish explorer Paul Strzelecki, and on the grounds that the range contained rocks somewhat like those observed in the Urals (where alluvial gold had been found on the eastern flanks) and was similarly aligned, approximately N–S. In the event, this ‘prediction’ (1844) proved correct and was followed by gold rushes in the 1850s. Murchison regarded his lucky forecast as evidence that he was a ‘‘sort of authority’’ on Australian gold deposits. He recommended (1846) migration to Cornish tin miners, some of whom benefited from his fortuitously useful advice. Murchison thought that the Russian gold was emplaced by quite recent tectonic activity in the Ural region, and he suggested that the range had undergone several distinct upheavals. Nevertheless, he gave credence to the theory of E´ lie de Beaumont that mountain ranges with different alignments were of different ages. Thus, the old Palaeozoics of Scandinavia were aligned SW–NE; the supposedly postCarboniferous/Permian Urals ran N–S; while the post-‘oolitic/chalk’ of the Caucasus ran WNW–ESE.
214 FAMOUS GEOLOGISTS/Murchison
The position of Director-General of the Geological Survey of Great Britain fell vacant in 1855, upon the decease of its founder Sir Henry De la Beche, and Murchison was appointed in his place, though already in his sixties. It proved to be an astute move, so far as the Survey’s progress was concerned, for Murchison had innumerable contacts and used them to advantage to build up the organization considerably. He was indefatigably a man of organization, and competent, with his experience in the running of several societies and associations. The appointment was gratifying to Murchison as it ensured that the official maps should be constructed and coloured according to his interpretation and subdivisions of Palaeozoic geology, to the extent that the Cambrian was almost driven off the map for British geology. Much of Murchison’s geological work in his later years was focused on Scotland, where a separate branch of the Survey was established in 1867, though surveying had begun there back in 1854. Murchison’s Scottish work involved him in the last of his three great controversies, and involved the attempted expansion of Silurian colours over the greater part of northern Scotland. There is today thought to be a great thrust-plane (the ‘Moine Thrust’) that runs from the north coast near Lochs Durness and Eriboll to the south-west, terminating in the southern part of Skye. To the west, one finds ‘Fundamental [Lewisian] Gneiss’ (so called by Murchison), overlain unconformably by the unfossiliferous Torridonian Sandstone. Lying unconformably on this there is a series of sediments, including a ‘quartz rock’ and the fossiliferous Durness Limestone. Over this lies the complex unit called the Moine Schists, above the thrust plane and extending eastwards until it is itself overlain unconformably by Old Red Sandstone on the eastern side of the country. But the structure near the thrust plane is complicated, with folding, inversions, and apparent duplication or repetitions of strata; in places the gneiss ‘reappears’, both near the thrust fault and again further east. Murchison visited the north-west Highlands of Scotland in 1855 (with the Aberdeen University geologist James Nicol), in 1858 (with the local amateur naturalist Charles Peach), in 1859 (with the Survey officer Andrew Ramsay), and in 1860 (with the young surveyor Archibald Geikie). Fossils regarded by Murchison as Lower Silurian were found by Peach in the Durness Limestone (at a lower horizon than the thrust plane). All the strata appeared to dip gently to the southeast, with a strike approximately parallel with what is now thought to be the thrust-fault system. The outcome of all this work was that in the view of Murchison (and also Ramsay and
Geikie) there was an essentially simple ascending sequence (with unconformities) from ‘Fundamental Gneiss’ on the west (regarded as lying at the bottom of the whole stratigraphic column for Britain) through to the Old Red Sandstone on the east, with a repetition of quartz rock into distinct upper and lower units, and also repetition of the gneiss. This meant that the Moine Schists, lying between the supposed Lower Silurian Durness Limestone and the Devonian Old Red Sandstone, though unfossiliferous, could be regarded as Silurian. So when a geological map of Scotland was published by Murchison and Geikie in 1861, large areas of northern Scotland were represented in Silurian colours. Murchison’s empire was again expanding in a manner that he found most satisfactory. As to the Cambrian, Murchison allocated the unfossiliferous Torridonian Sandstone to that System, so Sedgwick saw some expansion his empire, but not by rocks with well-characterized fossils. But Nicol’s reading of the structure was very different from Murchison’s. He came to the view that there was repetition of the western and eastern metamorphic rocks due to a large (high-angle) fault, and that the resultant fissure had been filled in part by some kind of igneous rock. (He was perhaps mistaking some gneiss for igneous rock.) If this interpretation were correct then placement of the Moine Schists in the Silurian would be suspect. So Nicol and Murchison fell out, and Nicol thereafter conducted his work separately from the Survey chief. The issues were debated at the British Association meeting in Aberdeen in 1859, where both geologists put forward their cases. From his stronger social position, Murchison was judged the winner by most geologists, and in fact Nicol’s idea was by no means wholly correct. The results of this encounter were most satisfactory to Murchison and Geikie, who became his mentor’s advocate and eventually his sympathetic biographer. Murchison got more Silurian colour onto the geological map of Britain. In time, Geikie was appointed head of the Scottish branch of the Survey, and, when Murchison endowed a chair in geology at Edinburgh University (with Geikie’s urging), it was Geikie who moved smoothly into the position, holding it concurrently with his post in the Survey. Later he became Director-General of the Survey, President of the Royal Society, and one of Britain’s leading geologists. However, the Murchison theory of the structure of the north-west Highlands was shown to be in error by Charles Callaway’s and Charles Lapworth’s mapwork in the early 1880s, and the reputation of the deceased Nicol was restored. Lapworth showed that the structure involved folding and thrust-faulting (a
FAMOUS GEOLOGISTS/Murchison 215
term proposed by Geikie), and suggested that the Moine Schists were in fact formed by the earth movements that gave rise to the folding and faulting, while the repetitions of rock types could be attributed to the S-shaped folding. (This suggestion was eventually taken up by Geikie’s own staff, surveying in the 1880s, well after Murchison’s death.) The whole episode illustrates Murchison’s dominating personality and commanding social role towards the end of his career. The reasons underlying the Cambrian–Silurian debate have been analyzed by Rudwick (1976) in the following terms. At one level, it arose because Murchison’s structural interpretations were not always correct and because he confused the May Hill Sandstone (Wenlock) with the lithologically similar Caradoc Sandstone (Caradoc). Both geologists gained ideas about how to do stratigraphy from William Smith (see Famous Geologists: Smith). Smith himself started from the observation of superposed sections of rocks of characteristic structure and lithologies. Subsequently, he remarked that each rock suite had its own characteristic fossils, but he saw no reason in principle why one set of fossils should not graduate into another. Thus there could, in principle, be overlap between Cambrian and Silurian fossils. Murchison started off on a similar basis, but gradually shifted towards thinking that it was fossils that defined a system. Once this had happened, and he began to find ‘Silurian’ fossils in Sedgwick’s Cambrian, then annexation of territory ‘naturally’ followed (given that Sedgwick was so slow in getting his ‘Cambrian’ fossils published). It seemed to Murchison that he was dealing with a bona fide system, as it preceded land plants, had few vertebrates, and was apparently distributed widely round the world. By contrast, when Sedgwick got round to palaeontological analysis about a decade after his initial fieldwork in North Wales, he thought that the break should, if anywhere, lie between Murchison’s Lower and Upper Silurian; so that for Sedgwick the Cambrian should incorporate Murchison’s Lower Silurian. But by then the Lower Silurian was already well established, with its fossils described. Rudwick further points out that both geologists were opposed to Lyell’s ‘steady-statism’ (see Famous Geologists: Lyell). They both believed that life originated at some point in the past, and Murchison wished ‘his’ system to be the one that contained the first evidences of life with hard-bodied remains. Hence he sought to cannibalize Murchison’s Cambrian. When Barrande in Bohemia found a ‘Primordial’ fauna below Murchison’s Lower Silurian (palaeontologically defined), it could have served as palaeontological basis for a Cambrian System. But Murchison
declined to follow this path, and did not practise what he preached in the matter of the Cambrian. Like many geologists of his day, Murchison gave considerable attention to the problem of the superficial ‘drift’ deposits that blanket much of Europe, and which he saw in abundance in Scandinavia, Russia, Britain, and elsewhere. In the early nineteenth century, such materials were commonly ascribed to the Noachian Flood, or later to catastrophic floods but not necessarily universal or of divine origin. In the 1840s, there were two further contending theories: that of climatic change producing an Ice Age, with land ice as the agent for the emplacement of the ‘drift’, as advocated by Louis Agassiz (see Famous Geologists: Agassiz); and various versions of ‘glacial submergence’, with cooling and changes of sea-level relative to the land such that floating icebergs could carry detritus and deposit mud and ‘erratic’ boulders, as envisaged by Darwin and Lyell. The ‘flood theory’ received some theoretical support from the Cambridge mathematician and geologist, William Hopkins, who advocate the idea of ‘waves of translation’. A sudden uplift of the sea-floor might, it was suggested, produce not only waves at the ocean surface, but also wholesale lateral movement of masses of water, capable of transporting (‘translating’) large boulders and finer debris. It was Hopkins’ theory that Murchison favoured, in part because it was seemingly in accord with the evidences familiar to him in the Alps and elsewhere of huge earth movements, foldings, faulting, and even inversions. (Murchison had seen evidence of stratigraphic inversion in the Glarus Canton, Switzerland, when he visited the area in 1848, but subsequently disregarded it in his thoughts about the north-west Highlands of Scotland.) He accepted that retreating glaciers left moraine material in the Alpine regions, and was happy with the idea of icebergs transporting drift material. But for long he could not accept landice as being responsible for the huge tracts of drift on land of low relief that he saw in Russia. Besides, the evidence of striations did not seem to accord with the land-ice theory. For example, in the area of the Gulf of Bothnia he saw scratch-marks directed southeastwards, from an area of Sweden of low altitude. He did not imagine that glaciers could have come from further north, from the mountains of Arctic Sweden. Nor could he imagine that land-ice could on occasions travel uphill, transporting marine shells to hill tops. It was only in 1862 that Murchison conceded to Agassiz’s land-ice theory. Murchison was one of the heroes of the heroic age of geology. His contributions to stratigraphy, and the broadening of geological knowledge generally, were immense. He was extraordinarily energetic, and generally amiable. Other than Lyell, he was far
216 FAMOUS GEOLOGISTS/Sedgwick
the most influential British geologist of his day. But he exercised that influence through the hand of administrative power as much as by reasoned argument. He was a conservative in both politics and geological theory. Even his prote´ ge´ Geikie, who owed him so much, concluded that Murchison ‘‘was not gifted with the philosophical spirit which evolves broad laws and principles in science. He had hardly any imaginative power. He wanted . . . the genius for dealing with questions of theory . . .’’ Possibly things would have been different if Murchison had received a university, rather than a military, education.
See Also Famous Geologists: Agassiz; Darwin; Lyell; Sedgwick; Smith. History of Geology From 1780 To 1835. History of Geology From 1835 To 1900. Palaeozoic: Ordovician; Silurian; Devonian; Carboniferous.
Further Reading Geikie A (1875) Life of Sir Roderick I. Murchison. . . Based on his Journals and Letters With Notices of his
Scientific Contemporaries and a Sketch of the Rise and Growth of Palaeozoic Geology. London: John Murray (reprinted in facsimile by Gregg International Publishers Ltd, 1972). Oldroyd DR (1990) The Highlands Controversy: Con structing Geological Knowledge through Fieldwork in Nineteenth Century Britain. Chicago and London: Chicago University Press. Rudwick MJS (1972) Levels of Disagreement in the Sedg wick Murchison Controversy. Journal of the Geological Society 132: 373 375. Rudwick MJS (1985) The Great Devonian Controversy: The Shaping of Scientific Knowledge among Gentle manly Specialists. Chicago and London: Chicago Univer sity Press. Secord JE (1986) Controversy in Victorian Geology: The Cambrian Silurian Dispute. Princeton: Princeton Univer sity Press. Stafford RA (1989) Scientist of Empire: Sir Roderick Murchison, Scientific Exploration and Victorian Imperi alism. Cambridge, New York, Port Chester, Melbourne and Sydney: Cambridge University Press. Thackray JC (1976) The Murchison Sedgwick Controversy. Journal of the Geological Society 132: 367 372.
Sedgwick D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Adam Sedgwick (Figure 1) was born in 1785 in the village of Dent in the Yorkshire Dales, northern England, son of the local vicar and third of a family of seven. He died as a Fellow of Trinity College and Professor of Geology at Cambridge in 1873. He attended Sedbergh School near Dent, and with help from a notable local amateur mathematician, John Dawson of Garsdale, obtained a scholarship to Trinity, where he studied mathematics. He was named 5th Wrangler (fifth in the list of first-class honours students) in 1808. Following further examination, Sedgwick obtained a College Fellowship in 1810 and taught undergraduate mathematics. He was ordained in 1817. On becoming a permanent member of college, Sedgwick also committed himself to bachelorhood. During his life as a geologist, he proved to be extremely energetic in the field, covering large distances in a day. In Cambridge, he was quite often indisposed, but his chronic health
problems apparently disappeared once he got into fieldwork. Though a gifted mathematician, Sedgwick did not make a career in that discipline. From fragmentary autobiographical notes, he evidently had some geological interests from an early age, and he ‘geologized’ on the Continent in 1816. Also, he was ‘introduced’ to the Geological Society of London in 1818. Even so, it is surprising that his scientific accomplishments were thought sufficient to secure the Cambridge chair in geology that year. He was elected Fellow of the Royal Society in 1821, John Herschel heading the list of those who nominated him. Sedgwick was President of the Geological Society in 1829–31, and President of the British Association when it met in Cambridge in 1833.
Geological Work On obtaining his chair, Sedgwick threw himself into geology. He started his annual fieldwork in southern England, then worked his way northwards to Northumberland, and in the years 1822–24 he made the first systematic survey of the Lake District. He obtained topographic maps of the region, ‘recognized’ certain rock units, and systematically covered
FAMOUS GEOLOGISTS/Sedgwick 217
Figure 1 Portrait of Adam Sedgwick (1785 1873); the original is in the Sedgwick Museum, Cambridge. Copyright: The Sedgwick Museum of Earth Sciences.
the region over three seasons, colouring in his maps according to his selected lithological units. Sedgwick did not look much for fossils, but measured strata and cleavage dips and the alignments of beds, folds, faults, joints, and cleavage planes. Faults sometimes could be seen on the ground. At other times, they became apparent when the different rock types were coloured onto the maps. Sedgwick’s labour and energy were immense. Armed with hammer, acid bottle, map, compass, clinometer, and notebooks, Sedgwick tried to determine the structure of that complicated region. Trained in mathematics, the neophyte geologist was trying to ascertain whether the strata displayed any regular geometric pattern. They hardly did, but when Sedgwick published his work he sought to subsume it under the theory of Le´ once E´ lie de Beaumont, according to which, as the Earth cooled and contracted, it supposedly formed a regular fold pattern in its crust, with mountain ranges of similar age having similar alignments. The theory never exerted much influence in Britain, and Sedgwick soon gave up the idea; but the fact that he sought to deploy the French theory suggests that he was interested in a geometrical (mathematical) theory of the earth. This was consistent with the Cambridge tradition, which found expression in the activities of the Cambridge Philosophical Society, which Sedgwick helped found in 1819. He wanted a quantitative geology, with mathematically formulated laws. However, his later Lakeland work (assisted by amateur collectors) used fossils, and by the end of his career, he had worked out a stratigraphic order for the sedimentary rocks compatible with that later developed on biostratigraphic principles. Sedgwick also recognized the
rocks of central Lakeland for what they were: the relics of ancient volcanoes. He referred to waterdeposited volcanic ash deposits as volcanic mud. From his Lakeland work, Sedgwick came to understand (and publish) the distinction between bedding and cleavage, but the distinction was acquired from the local amateur Jonathan Otley, who probably got it from quarrymen. Sedgwick’s Lakeland mapping revealed the existence of large tear-faults in some of the strata, and the eastern boundary of the region was marked by a huge normal fault. Hence, he suggested that the area had been affected by great earth movements. Using terminology proposed in the 1830s by his Trinity colleague, William Whewell, Sedgwick was a ‘catastrophist’. Sedgwick soon met Roderick Murchison (see Famous Geologists: Murchison), who wanted someone to show him how to make sense of ancient rocks in mountainous regions, and together they undertook a lengthy reconnaissance journey round the north coast of Scotland in 1827, unfortunately correlating the eastern and western sandstones of northern Scotland (now regarded as Devonian and Precambrian, respectively). In 1829, they made an extensive tour of the Continent, meeting European professors and travelling to Germany, Bohemia, Hungary, Austria, Switzerland, and Italy. This greatly extended Sedgwick’s experience, especially in the ‘‘focus of Wernerian geology,’’ southern Germany, which he found to be ‘‘the most decidedly volcanic secondary country I ever saw’’. He observed granite veins and inclined or even overturned Secondary rocks (a term used in the nineteenth century as a synonym for the German term Floetz; later for the strata ranging from Silurian to Cretaceous; and later restricted to the Mesozoic Era). This demolished his earlier adherence to Neptunism and he publicly repudiated the doctrine in 1831. Observations in Italy clearly suggested uplift, which was not part of the Wernerian repertoire. On the other hand, Sedgwick was inclined to ascribe the movement of (glacial) erratic boulders to the action of catastrophic floods, but some Swiss deposits could have been emplaced by the bursting of lake barriers. He rejected the idea that ‘diluvium’ was all deposited in the Noachian Flood, but was critical of Charles Lyell’s belief that conditions on Earth were essentially similar through time (see Famous Geologists: Lyell). In the 1830s, Sedgwick collaborated with Murchison in Wales, trying to bring order to the strata there. The strata in the mountains of Snowdonia seemed to have analogy with those of the lakes. Sedgwick tackled them, working on somewhat similar lines, and making a traverse north-west to southeast across Snowdonia in 1832. He unravelled the
218 FAMOUS GEOLOGISTS/Sedgwick
structure to some extent, and although the rocks seemed to have few fossils, the Cambrian System was introduced, largely on the basis of Sedgwick’s work. Murchison, starting in the Welsh border region in gentler country with fossiliferous rocks, had an easier time and revealed what appeared to be a new system with its characteristic fossils, dubbed the Silurian. However, the line of boundary between the two systems was not established at the outset of the investigations in 1834, with the result that a bitter feud subsequently broke out between Sedgwick and Murchison, with the former trying to extend ‘his’ system upwards, the latter endeavouring to expand ‘his’ downwards, eventually to the very lowest fossiliferous rocks. (The issues were very complex; it was not until after the protagonists’ deaths that the issues were resolved, by Charles Lapworth, who, in 1879, proposed the Ordovician System to occupy the disputed territory between the Cambrian and Silurian. (see Palaeozoic: Ordovician) Sedgwick had complained, with reason, that materials he had sent to the Geological Society were changed so as to accord with Murchison’s views, without Sedgwick’s knowledge or consent. The situation grew so bad between the two that after 1853, the Geological Society declined to accept further papers by the protagonists of Siluria or Cambria. Sedgwick felt grievously ill treated and snubbed by the Society. The battle became transferred to the forum of the British Association, but after 1854, Sedgwick withdrew from that body also, so far as the Cambrian and Silurian were concerned, and continued the battle from Cambridge and in the pages of the Philosophical Magazine. Some of the stratigraphic formations and their classifications are shown in Figure 2. It should be noted that the rocks that Murchison allowed to be Cambrian in 1859 were the unfossiliferous Longmynd rocks, later classified as Precambrian. (Murchison also allowed Sedgwick the unfossiliferous Torridonian Sandstone in Scotland, also now regarded as Precambrian.) Initially, Murchison won the battle, partly because he had better fossil evidence, and placed full reliance on it. Also, he had strong influence in the Geological Society and became Director of the Geological Survey in 1855, and his classifications were used by the survey officers. Sedgwick had fewer allies, mostly at Cambridge. For the rocks he was dealing with, he had to rely on structural understanding and lithologically based mapping to a greater extent than did Murchison. (Graptolites were not regularly used for stratigraphic correlation in the mid-nineteenth century.) However, Sedgwick succeeded in showing, on palaeontological grounds, that the claimed unity of Murchison’s Silurian System was flawed. In 1852,
Sedgwick and his assistant Frederick McCoy found that one of Murchison’s Silurian formations, the Caradoc, had rocks containing two distinct faunas, as shown by the palaeontological determinations of McCoy and John Salter. There had been erroneous correlations; the same term, ‘Caradoc Sandstone’, had been applied to different series of rocks; and there should be an unconformity within the Caradoc, as Murchison then understood it. Sedgwick proposed the division of Murchison’s Caradoc into the Caradoc Sandstone, containing fossils such as Trinucleus, and an upper May Hill Sandstone, containing Pentamerus species. This eventually turned the tide against his Silurian being regarded as a coherent system. The Survey sought to retrieve the situation by adopting the terminology ‘May Hill Sandstone Llandovery’ rocks, regarding them as a kind of passage or ‘Intermediate Series’ between the Upper and Lower Silurian. There were repercussions, too, for the interpretation of Murchison’s Llandeilo. This battle was fought with extreme vehemence. Both geologists attached their names and reputations to ‘their’ system. The battle seemed to exemplify the height of the colonial era, with Sedgwick and Murchison trying to extend their empires. Murchison was popularly called the ‘King of Siluria’. Both men tried to rewrite history in their historical accounts of the events. In Sedgwick’s case, this may have been partly due to failing memory. Earlier, there had been a bitter controversy in Devonshire, where, while still friends, Sedgwick and Murchison began to unravel the structure and stratigraphy of the area, in the process becoming involved in controversy with Henry De La Beche, the first Director of the Geological Survey. From this acrimonious debate emerged the concept of the Devonian System. The so-called Old Red Sandstone was the unit well known in eastern Scotland as lying unconformably over Silurian strata. It was mapped by William Smith as ‘Red and Dunstone’. He placed it below the Coal Measures and below a limestone that cropped out in Derbyshire and elsewhere. In Devonshire, there occurred rocks with plant remains that appeared to De la Beche to belong to the old Transition/Greywacke series (Cambrian or Silurian rocks). In Murchison’s opinion, however, the plant-bearing rocks could not be so old: they must be from the Coal Measures. But he had not then been to Devon to see the rocks for himself. Murchison then combined forces with Sedgwick to combat De la Beche’s interpretation. It was an issue of more than academic significance because it bore on the question of the possible extent of coal-bearing rocks. The plant-bearing rocks overlay contorted rocks of ancient appearance, but these contained corals
FAMOUS GEOLOGISTS/Sedgwick 219
Figure 2 Classification of British Lower Palaeozoic Rocks. Reproduced from Secord JA (1986) Controversy in Victorian Geology: The Cambrian Siurian Dispute, p. 287. Copyright ß1986 by P.U. Press. Reprinted by permission of Princeton University Press.
different from those in the Silurian. In the opinion of the coral expert William Lonsdale, the corals were intermediate between Silurian and Coal Measure types. Thus, the Devon rocks might be situated between the Silurian and the Carboniferous. So, in 1839, Sedgwick and Murchison proposed the Devonian System, being one that had different facies in different localities. Here the palaeontological evidence of corals was taken to outweigh the uncertain stratigraphic reliability of plant fossils and the structural arguments advanced by De la Beche. The issues were debated with considerable rancour and again illustrate the character of stratigraphic controversies in the nineteenth century. But this time, Sedgwick and Murchison were on the same side and the debate
never got quite so out of hand as did the Silurian/ Cambrian contest.
Sedgwick as a Teacher; Other Activities, Beliefs, and Character At Cambridge, Sedgwick gave an annual course of lectures and built up the university’s geological collections, partly from his own collected specimens, but also by donations and purchases. His summer fieldwork was done at his own expense. Partly for this reason, he took a ‘second job’ in 1834, as a canon at Norwich Cathedral. This might have been a sinecure, but Sedgwick took his responsibilities seriously, and resided in Norwich for several months each year, also
220 FAMOUS GEOLOGISTS/Sedgwick
encouraging the development of a museum in the city and giving geological lectures. Sedgwick was a strong supporter of amateur science and assisted the natural history society in Kendal, near Dent. Sedgwick was a renowned orator – or preacher and lecturer. Science lectures were not a required part of the Cambridge curriculum when he gave his first course in 1819, but he attracted many students and dons to his lectures. His course was repeated until 1859, when he was 74. He spoke extempore about geological principles and his recent fieldwork, rather than about unnecessary minutiae. His lectures, utilizing specimens and maps and diagrams to explain his ideas, were popular, and he also led groups on horseback on field excursions round Cambridge. On a famous occasion at the British Association meeting at Newcastle in 1838, he spoke in the morning at Tynemouth beach to a group attending the meeting; and by the afternoon he had attracted a crowd of thousands, expounding the relationships (as he saw them) between geology, political economy, natural theology, and patriotism, reportedly drawing tears of emotion from some auditors. As John Herschel described it, Sedgwick led them on from the scene around them to the wonders of the coal country below them, thence to the economy of a coal field, then to their relations to the coal owners and capitalists, then to the great principles of morality and happiness, and last to their relation to God and their own future prospects. (Clarke and Hughes (1890).)
In this can be seen the relationship between Sedgwick’s science, social, religious, and political philosophy. Implausibly, he supposed that Britain had been specially favoured by God for its place in the world, with its deposits of coal, limestone, and iron ore. Sedgwick was, then, devoutly religious, and a preacher as well as a teacher. From a relatively obscure Anglican background, he rose to be a Cambridge Professor and Vice-Master of Trinity, and one of the leaders of the heroic age of geology. He even met with Prince Albert (whose election to the Chancellorship of the University he promoted) to discuss reforms at Cambridge, and many of Sedgwick’s suggestions were implemented. Though generally amiable, greatly liked and admired, and able to communicate with the full range of society, from quarrymen, to famous writers such as William Wordsworth or Walter Scott, to Royalty, he was uncompromising and dogmatic. He favoured Catholic emancipation, but having become an establishment figure, he did not wish to see the regular order of things upset by scientific theories that seemed to him subversive, or at odds with orthodox Anglican theology. For such
reasons, he was bitterly and publicly opposed to the transmutationist ideas expressed in 1844 by Robert Chambers in his Vestiges of the Natural History of Creation, and was privately grieved by Charles Darwin’s ideas in The Origin of Species (see Famous Geologists: Darwin). He also rejected the land–ice theory of Louis Agassiz (see Famous Geologists: Agassiz).
See Also Famous Geologists: Agassiz; Darwin; Lyell; Murchison; Smith. History of Geology From 1780 To 1835. History of Geology From 1835 To 1900. Palaeozoic: Cambrian; Ordovician; Silurian.
Further Reading Clark JW and Hughes TMcK (1890) The Life and Letters of the Reverend Adam Sedgwick, LL.D., D.C.L., F.R.S., Fellow of Trinity College, Cambridge, Prebendary of Norwich, Woodwardian Professor of Geology, 1818 1873. Cambridge: Cambridge University Press. Oldroyd DR (2002) Adam Sedgwick: a confident mind in turmoil. In: Harman P and Mitton S (eds.) Cambridge Scientific Minds, pp. 64 78. Cambridge: Cambridge University Press. Oldroyd DR (2002) Earth, Water, Ice and Fire: Two Hun dred Years of Geological Research in the English Lake District. London: The Geological Society. Rudwick MJS (1972) Levels of disagreement in the Sedgwick Murchison controversy. Journal of the Geological Society 132: 373 375. Rudwick MJS (1985) The Great Devonian Controversy: The Shaping of Scientific Knowledge among Gentle manly Specialists. Chicago and London: Chicago University Press. Rudwick MJS (1988) A year in the life of Adam Sedgwick and company, geologists. Archives of Natural History 15: 243 268. Secord JA (1986) Controversy in Victorian Geology: The Cambrian Silurian Dispute. Princeton and Guildford: Princeton University Press. Sedgwick A and Murchison RI (1835/36) On the Silurian and Cambrian Systems, exhibiting the order in which the older sedimentary strata succeed each other in England and Wales. Report of the Fifth Meeting of the British Association for the Advancement of Science held at Dublin in 1835, pp. 59 61. London: John Murray. Smith C (1985) Geology and mathematicians: the rise of physical geology. In: Harman PM (ed.) Wranglers and Physicists: Studies on Cambridge Physics in the Nineteenth Century, pp. 49 83. Manchester: Manchester University Press. Speakman C (1969) Adam Sedgwick Geologist and Dalesman, 1785 1873: A Biography in Twelve Themes. Broad Oak, London, and Cambridge: The Broad Oak Press Ltd.
FAMOUS GEOLOGISTS/Smith 221
Smith D R Oldroyd, University of New South Wales, Sydney, Australia ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction William Smith (Figure 1) is renowned in the history of geology for differentiating and listing in order the English strata, from the Chalk down to the Coal Measures, and hence enunciating the stratigraphic principle that strata have a generally regular order of superposition and may be characterized by their fossil contents. The great geological map of much of Britain that Smith published in 1815 has also brought him much posthumous fame. As Smith put it in the memoir accompanying his map: [T]here is a great deal of regularity in the position and thickness of. . . strata; and although considerable dis locations are found in collieries and mines, and some vacancies [gaps] in the superficial courses of them, yet. . . the general order is preserved; and . . . each stratum is . . . possessed of properties peculiar to itself, has the same exterior characters and chemical qualities, and the same extraneous or organized fossils throughout its course. (Smith W (1815, p.2))
Background Smith was born in 1769 in the village of Churchill, Oxfordshire, near Chipping Norton, an attractive part of Britain where Jurassic rocks (Oolitic Limestone and Lias) crop out well. His father, a blacksmith,
Figure 1 William Smith (1769 1839) aged 69, engraved by TA Dean.
died when he was only 7 years old, and he was then brought up by a farmer uncle at Over Norton, a few miles to the north-east, in similar type of country. Even at an early age, Smith was a keen collector of fossils. Not wishing for a life as a farm labourer, he began studying mathematics, geometry, and surveying techniques, and at age 18 became an assistant to a local surveyor. Smith soon became autonomous, and his work as a land surveyor evolved into the business of surveying for canal construction (initially the Somerset Coal Canal). This type of construction was then forging ahead in the days of the Industrial Revolution, and in effect, Smith became what would today be called a civil engineer. He advised on tunnel constructions, borings for coal, and mining activity in the Somerset coalfield. In 1794, he travelled to northern England on a ‘fact-finding tour’, in connection with canal work. He later advised on drainage projects for wealthy landowners who sought to develop their lands agriculturally, but these landowners were also interested in the mineral wealth that their estates might hold.
Development of Smith’s Stratigraphic Principle The varied experience Smith acquired, and especially that resulting from the canal cutting through different stratigraphic horizons, led, before 1796, to Smith recognizing the general aspects of his stratigraphic principle. He realized that the regular stratifications found within the coal mines could also be observed outside the mines. The Somerset Canal cuttings, which cut through two similar sections at two separate but neighbouring localities, revealed the lateral extents of strata, and Smith began to get the ‘feel’ of the internal structure of the earth as regards the strata of his region. In 1795, he took up residence in Bath and pondered what his work was beginning to reveal, namely, that ‘‘Nature has disposed of these singular productions [fossils] and assigned to each class its peculiar Stratum’’. In 1797, he wrote down a list of strata for the Bath district, listing 28 units, from Chalk down to Carboniferous Limestone, but without mentioning fossils (this is his earliest dated list that has survived). A revised version of this document (with fossils given) was dictated to two local clergymen/naturalists, the Reverends Benjamin Richardson and Joseph Townsend in 1799, who were also shown a circular map of the Bath district and one of
222 FAMOUS GEOLOGISTS/Smith
Somerset that Smith had geologically coloured. Not long before, Smith’s employment with the Somerset Coal Canal Company had been terminated and it is possible that he was beginning to think of finding some practical and remunerated application of his new ideas. During the next few years, he was largely involved with drainage schemes, but by 1804, he was chiefly employed in ‘mineral surveying’. He travelled great distances in these lines of work, and thus accumulated information that he later synthesized in the form of his celebrated geological map of 1815. The revised stratigraphic table of 1799 had 23 stratigraphic units, from Chalk down to Coal, with thicknesses indicated, along with localities where the rocks cropped out. ‘Fossils and Petrifactions’ were given for each unit, but the fossil categories, as stated by him, were imprecise (corals, cochleae, ostreae, impressions of ferns, etc.). Also, some strata were characterized lithologically, or the entry was stated ‘‘No fossils known’’. So Smith’s earliest table of strata was not based wholly on fossils, and appears to have been primarily a list of the lithological sequence of distinctly identifiable strata. At that time, he had no ‘scientific’ knowledge of fossils, but he was certainly collecting fossils well before 1799; his collections survive today at the British Museum. In any case, his differentiation of four blue clays and two different oolitic limestones indicates that he was making more use of fossils than is evident from the table dictated to Richardson and Townsend. They were impressed by the practical man’s revelations and encouraged Smith to continue his investigations, which he did, producing a simple geological map of England and Wales in 1801. It showed quite a clear representation of the distribution of several major stratigraphic subdivisions, notably those now known as the Carboniferous, Jurassic, and Cretaceous. With the loss of forests for shipbuilding and development of agriculture, and increasing demands for iron, there was need for coal in the years of the Napoleonic Wars, and various prospecting attempts were made in southern England, often on the illusory promise of the discovery of lignites or the occurrence of blue clays, thought by their appearance to be associated with coal measures. But Smith knew the correct order of strata by their fossils and realized that these attempts were doomed to failure. The prospectors were looking too high in the stratigraphic column. On the other hand, though Smith also gave sometimes successful advice as to where coal might be found in known coal areas, he was not always successful, due to unforeseen structural complexities, so that the coal beds present were unfortunately missed. In fact, Smith’s own entrepreneurial ventures were by no means successful. His intended book, Accurate
Delineations and Descriptions of the Natural Order of the Various Strata that are Found in the Different Parts of England and Wales, with Practical Observations Thereon, for which a prospectus was printed in 1801, was never published because of the double bankruptcy of the intended publisher in 1801 and 1804. (The text of this prospectus was published in a 1942 article by LR Cox.) Worse, in 1798, Smith purchased a small estate at Tucking Mill near Bath, where quarrying of Bath stone was later attempted. This proved a failure because of the unexpectedly poor quality of the stone, and the financial problems that flowed from this led to Smith’s subsequent financial collapse. (It is a sad irony that the experienced engineer and emerging geologist should have misjudged the stone quality and the difficulty of getting it out of the quarry.) On the other, hand, through agricultural contacts at Woburn, seat of the Duke of Bedford, Smith was introduced to the influential Sir Joseph Banks, President of the Royal Society, from whom Smith received both moral and financial support. Also, in 1801, Smith attracted, as a kind of ‘pupil’, the polymathic John Farey (likewise a practical man, but one of many accomplishments, including mathematics and music). They did fieldwork together and Farey became a constant advocate of Smith and his work. Smith also became acquainted with the map publisher and engraver John Cary, whose map of England and Wales and various county maps became the basis of Smith’s great geological map (see later). In 1802, Smith opened an office in Bath to conduct his affairs, and in 1804, he moved to London, where he displayed his fossils, arranging them on sloping shelves so that the fossils of each stratum were displayed in the order that they appeared in the English rocks, producing a kind of three-dimensional effect. But his efforts could only be spread thinly, and not necessarily systematically, because of the exigencies of his work. Although it is clear that Smith long intended to attempt to publish a geological map of England, Wales, and southern Scotland, this project was not in fact fulfilled until 1815, partly because of difficulties in reaching an agreement with a publisher and partly because he was continuing with his observations and collections. Smith’s business took him all over the country, and he developed his topographic and geological knowledge wherever he went, to the point where he was almost overwhelmed with information and specimens. The Geological Society of London was founded in 1807, with the wealthy George Bellas Greenough as its president. It might have seemed natural or appropriate for Smith to have joined the Society, but this did not happen. Smith was from a lower social class than were the Society’s founders, and his patron,
FAMOUS GEOLOGISTS/Smith 223
Banks, was at odds with the new group, which he saw as a rival to ‘his’ Royal Society. Additionally, the Society’s early Fellows were chiefly interested in mineralogical matters, and they and others of influence, such as the Board of Agriculture or Professor John Kidd at Oxford, doubted the value of Smith’s ‘biostratigraphy’. So, to an extent, Smith was on his own, and he had to carry through his project using his own uncertain financial resources. Eventually, however, in 1815, Smith issued his great map and its accompanying memoir – A Delineation of the Strata of England and Wales, with Part of Scotland and A Memoir to the Map and Delineation of the Strata of England and Wales, with Part of Scotland. The geological information was entered on a map specially engraved for the purpose by Cary. There followed Strata Identified by Organized Fossils, Containing Prints on Coloured Paper of the Most Characteristic Specimens in each Stratum, which was issued in four parts between 1816 and 1819, providing descriptions of Smith’s chosen stratigraphic units and beautiful coloured illustrations of their typical fossils, and Stratigraphical System of Organized Fossils, which was issued in 1817. (Neither publication was fully completed.) The main map (dedicated to Banks) was issued on 16 sheets, one being an index sheet. (The Banks copy at the British Library may be the ‘master’ copy, but this is not certain.) On a scale of 5 miles to the inch, the map was approximately 260 by 180 cm in size. Each copy was hand coloured (apparently using colourists employed by Cary, not always to Smith’s satisfaction) and there were five issues between 1815 and 1817 (or perhaps later). Examination of surviving copies has shown that Smith added information to the new issues as it became available to him. From 1819 to 1824, he issued also various ‘county maps’, which depicted the geology of individual counties. These lacked ‘geological rationale’, in the sense that counties were not ‘natural’ geological regions, but they were useful commercial products. Together they formed Smith’s Geological Atlas. A smaller country map on the scale of 15 miles to the inch, with revisions of the earlier map, was issued in 1820, and this map was also variously revised until at least 1828. Additionally, Smith produced several geological sections, including one from London to Snowdon in North Wales, in 1817, and one of the strata south of London, in 1817. A manuscript section from 1824, from Flamborough Head on the east Yorkshire coast to Whitehaven on the Cumberland coast on the west of England, is preserved at Oxford. The 1815 map was a mighty contribution to geology, achieved largely single-handedly. Surviving pristine copies are objects of great beauty, ingeniously
and impressively coloured, so as to convey almost a three-dimensional effect, by increasing the intensity of colouration towards the lower boundaries of the outcrops of the various units. The map depicted 23 stratigraphic subdivisions, and some of the colours that Smith chose (e.g., green for Chalk) survive into modern maps. But Smith’s position was financially precarious, as it had been ever since his unsuccessful quarrying venture near Bath, and eventually, in 1817, he was obliged to sell his fossil collections to the British Museum for £700, the catalogue for this being his Stratigraphical System of Organized Fossils. The sale only postponed Smith’s financial crisis, however, and he found himself languishing in a debtors’ prison for 10 weeks in 1819. He obtained release by sale of his property near Bath, but withdrew to the north of England, making a living by continuing his survey work and giving lectures in Yorkshire’s major towns. He had previously been assisted by his gifted nephew John Phillips, and their association continued. Phillips was perhaps chiefly responsible for the aforementioned east–west section of the north of England, which revealed an understanding of the faulted structure of the Vale of Eden, to the east of the Lake District. Phillips also did some of the lecturing. (Phillips subsequently became one of Britain’s leading geologists, a Fellow of the Royal Society and Professor at Oxford.) While in the north, Smith put both Adam Sedgwick (see Famous Geologists: Sedgwick) and Roderick Murchison (see Famous Geologists: Murchison) on the right track as to the use of fossils for stratigraphic purposes. Smith eventually settled in Scarborough and assisted in the founding of the Scarborough Philosophical Society and Museum (which was made largely to Smith’s design). Back in 1808, Smith had been visited by Greenough and other leaders of the Geological Society, but they seemed unimpressed by his work, and subsequently started compiling a collaborative Society map, based on lithological principles such as were typically used by German geologists. However, at some point during the year after Smith was imprisoned, under the influence of the publication of Smith’s map, the Geological Society group changed their approach and issued their own map, using some of Smith’s fossil-based data, though Greenough asserted that the utility of fossils had been ‘‘greatly over-rated’’. Smith, at that point in time, had been walking to the north with his nephew Phillips as companion. Smith and his supporters, such as Farey, claimed his priority rights on several occasions, but these were not fully acknowledged until there was a change of personnel in the Geological Society. Eventually, in 1831, acknowledgement was accorded Smith by making him the
224 FAMOUS GEOLOGISTS/Smith
first recipient of the Society’s Wollaston Medal (though there were objections made as to whether it was appropriate to make an award for work first announced in 1799). From the chair, the President Adam Sedgwick acknowledged his personal indebtedness to Smith’s advice and dubbed him the ‘Father of English Geology’. Smith’s reputation was thus securely sealed, and the following year he received an annuity of £100 per year from the government. Moreover, his stratigraphic subdivisions set the pattern for work in other countries: the world followed British stratigraphy. If geology had emerged in the United States, China, or New Zealand, say, the stratigraphic column would look substantially different from that which is now used.
Influence of Smith’s Work The question of Smith’s theoretical ideas in geology is important. His sections showed the strata of southern England in their correct order, where they conveniently form a ‘layer-cake’ stratigraphy. His main expertise was in the stratigraphy of these Mesozoic sediments (as they are now called). It is evident that strata of different lithologies were recognized first, and then Smith realized that each stratum had its characteristic fossils. Soon, he could reverse the argument and use fossils to identify the strata. Sometimes, however, he encountered problems. He thought, for example, that the poorly fossiliferous Magnesian Limestone of north Yorkshire (subsequently designated as Permian) and the Lias (now Jurassic) belonged to the same stratum, because they contained rather similar fossil fishes. Also, what is now thought of as Carboniferous Limestone and Lias were regarded on occasion by Smith as different facies of the same unit, there being no locality where the Carboniferous Limestone, Magnesian Limestone, and Lias appear in what is today regarded as the correct order, and some Carboniferous Limestone does occur in places reworked into Lias. This is not to blame Smith. He was pioneering, and mistakes were to be expected in those early days of biostratigraphy. Throughout his career (started as a surveyor), Smith was always primarily interested in the geometrical arrangement of rocks, because this was what counted for agricultural, mining, and engineering purposes. His livelihood depended on knowing that order. Nevertheless, though it was not his primary concern, he did ask himself why the order was the way it was, how the strata came to be formed, and how long it took for them to be deposited. Smith’s religious beliefs appear to have been conventional, or characteristic of his time, and involved use of the ‘argument from design’. So one answer (1817) was
simply that ‘‘[t]he interior of the earth. . .is formed upon the wisest and best principles’’, and that the inclinations of the strata evidenced design by making the different rock types available for human use. Fossils must ‘‘strike the admirers of nature with a degree of reverential awe and grateful admiration of the Almighty Creator’’. Earlier, in 1802, in a preface to a book that was never published, Smith had supported an older eighteenth-century idea that the inclinations of the strata were the result of Earth’s rotation when the materials were still ‘‘in soft state or of pulpy consistence’’. But Smith apparently dropped this idea, which would imply that stratigraphic order did not represent chronological order. Even earlier, according to an 1844 memoir of Smith by J Phillips, Smith thought (in about 1795) that ‘‘each stratum had been successively the bed of the sea, and contained in it the mineralized monuments of the races of organic beings then in existence’’. Another shred of evidence on this matter is provided by Farey, in a review he published in 1810 of Georges Cuvier and Alexandre Brongniart’s 1808 memoir on the geology of the Paris area. Farey claimed that soon after Smith began his investigations, he ‘‘discovered an important law regulating all the known alluvia, or that which consisted of or contained the fragments and reliquia of known strata, [namely that they] were moved from the south-east towards the south-west’’ (italics in original) for material from any particular stratum seemed to have been transported beyond its western edge. This appears to have been a reference to observations of boulder clay, or ‘drift’, which material was ascribed by Smith (or Farey?) to ‘‘vast tidal currents which have swept over all the surface from SE. to NW., since or at the time, that the deposition of regular strata ceased’’. This suggests some support on Smith’s part for the catastrophist doctrines espoused by Cuvier (see Famous Geologists: Cuvier). This could have accorded with Smith’s religious views, but would also have involved the notion of time for the emplacement of superficial materials. Additionally, HS Torrens has drawn attention to a Smith manuscript from about 1806; the manuscript indicates that Smith was then thinking of a vast extent of geological time: ‘‘the time required for the Perfection and Decay, and subsequent formation, into Strata which have evidently been formed in deep and quiet water’’. This time ‘‘would stagger the faith of Many’’. But Smith’s lectures in Leeds in 1825 referred specifically to geological proofs of the occurrence of the deluge. He seems to have been impressed with William Buckland’s recently claimed evidence for the occurrence of the Noachian Flood from cave excavations in Yorkshire. For further variety,
FAMOUS GEOLOGISTS/Smith 225
there are 40 proof sheets at Oxford of a work to be titled Abstract Views of Geology, which was apparently in press at the time of Smith’s death. He was speculating again about the formation of strata ‘‘from a chemical conversion of liquids and gases into the solid state, —the layering being the effect of an uncombinable excess of one of the ingredients in the layer then formed, and the vertical joints in that layer the effect of solidification. . ..’’ But he was not advancing geology through such suggestions. Continuing through his late lectures and this last work, he kept reverting to his great principle of identifying and ordering strata by their fossil contents, and the utility of knowledge of this kind. His speculations about time and process had little influence on the development of geology. Smith’s strata, as given in his table of 1817, with approximate modern equivalents, following JCM Fuller (1995), were as follows: 1. London Clay – Tertiary, Lower Eocene. 2. Sand – Tertiary, Lower Eocene (Woolwich and Reading Beds). 3. Crag – Tertiary, Pliocene (Shelly Sand). 4. Sand – Tertiary, Paleocene (Thanet Sand). 5. Chalk – Upper Cretaceous, Cenomanian to Senonian). 6. Greensand – Upper Cretaceous, Albian (Upper Greensand). 7. Brickearth – Upper Cretaceous, Albian (Gault Clay). 8. Sand – Lower Cretaceous, Aptan (Lower Greensand). 9. Portland Rock – Upper Jurassic, Portlandian– Purbeck. 10. Sand – Lower Cretaceous, Wealden (Ashdown Sand). 11. Oaktree Clay – Upper Jurassic (Kimmeridge Clay) and Lower Cretaceous (Wealden). 12. Coral Rag and Pisolite – Upper Jurassic, Corallian. 13. Sand – Upper Jurassic, Corallian (Lower Calcareous Grit). 14. Clunch Clay and Shale – Upper Jurassic, Oxfordian (Oxford Clay). 15. Kelloways Stone – Upper Jurassic, Callovian. 16. Cornbrash – Middle Jurassic, Bathonian and Upper Jurassic, Callovian. 17. Sand and Sandstone – Middle Jurassic (Hinton Sand). 18. Forest Marble – Middle Jurassic, Bathonian (and Wychwood Sandstone). 19. Clay over Upper Oolite – Middle Jurassic (Bradford Clay). 20. Upper Oolite – Middle Jurassic, Bathonian (Great Oolite Limestone).
21. Fuller’s Earth and Rock – Middle Jurassic, Bathonian. 22. Under Oolite – Middle Jurassic, Bajocian (Inferior Oolite). 23. Sand – Lower Jurassic, Upper Lias (Midford Sand). 24. Marlstone – Lower Jurassic, Middle Lias, Domerian. 25. Blue Marl – Lower Jurassic (Lower Lias Clay). 26. Blue Lias – Lower Jurassic (Lower Lias). 27. White Lias – Rhaetic (Lower Lias). 28. Red Marl – Triassic (Keuper Marl). 29. Redland Limestone – Permian (Magnesian Limestone). 30. Coal Measures – Pennsylvanian. 31. Mountain Limestone – Mississippian. 32. Red Rhab and Dunstone – Devonian (Old Red Sandstone). 33. Killas – Devonian and older (slates, grits). 34. Granite, Syenite, Gneiss. It is clear, then, that in its essentials, Smith’s stratigraphic order still stands to this day.
See Also Economic Geology. Famous Geologists: Cuvier; Murchison; Sedgwick. Geological Maps and Their Interpretation. History of Geology From 1780 To 1835. Palaeontology. Stratigraphical Principles.
Further Reading Cox LR (1942) New light on William Smith and his work. Proceedings of the Yorkshire Geological Society 25: 1 99. Eyles JM (1969) William Smith (1769 1839): a chronology of significant dates in his life. Proceedings of the Geological Society of London 1657: 173 176. Fuller JGCM (1995) ‘‘Strata Smith’’ and his Stratigraphic Cross Sections, 1819: A Review of Facts Worth Knowing about the Origin of Stratigraphic Geology in the Mind of William Smith (1769 1839), an English Country Sur veyor and Civil Engineer. Tulsa: American Association of Petroleum Geologists; Bath: Geological Society Publishing House. Knell SJ (2000) The Culture of English Geology, 1815 1851. Aldershot, Burlington, Singapore, and Sydney: Ashgate. Phillips J (1844) Memoirs of William Smith, LL.D. Author of the ‘‘Map of the Strata of England and Wales,’’ by his Nephew and Pupil. London: John Murray. Reprinted (1978). New York: Arno Press; and (2003) Bath: The Bath Royal Literary and Scientific Institution (with add itional essays on Smith by HS Torrens). Sheppard T (1917) William Smith: his maps and memoirs. Proceedings of the Yorkshire Geological Society 19: 75 253.
226 FAMOUS GEOLOGISTS/Steno
Smith W (1815) A Delineation of the Strata of England and Wales, with Part of Scotland; Exhibiting the Collieries and Mines, the Marshes and Fen Lands Originally Overflowed by the Sea, and the Varieties of Soil According to the Vari ations in the Substrata, Illustrated by the Most Descriptive Names by W. Smith. London: (see copy of original map at http://www.unh.edu/esci/greatmap.html). Smith W (1815) A Memoir to the Map and Delineation of the Strata of England and Wales, with Part of Scotland. London: John Cary. Smith W (1816) Strata Identified by Organized Fossils, Containing Prints on Coloured Paper of the Most Char acteristic Specimens in Each Stratum. London: Printed by W Arding; sold by W Smith; J Sowerby; Sherwood, Neely and Jones; and Longman, Hurst, Rees, Orme and
Brown. (see copy of original at http://www.unh.edu/esci/ wmsmith.html). Rudwick MJS (1996) Cuvier and Brongniart, William Smith, and the reconstruction of geohistory. Earth Sciences History 15: 25 36. Torrens HS (2001) Timeless order: William Smith (1769 1839) and the search for raw materials 1800 1820. In: Lewis CLE and Knell SJ (eds.) The Age of the Earth: From 4004 bc to ad 2002, The Geological Society, Special Publication No. 190, pp. 61 83. London: The Geological Society. Torrens HS (2002) The Practice of British Geology, 1750 1850. Aldershot and Burlington: Ashgate Variorum. Winchester S (2001) The Map that Changed the World. London: Viking.
Steno J M Hansen, Danish Research Agency, Copenhagen, Denmark ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Nicolaus Steno (Niels Stensen; Figure 1) was an anatomist, geologist, and bishop, often considered to be the founder of geology as a science. He was the first to describe the most fundamental principles of stratigraphy and crystallography, and the first to claim by rigorous arguments that fossils are the remains of former life on Earth. Steno’s principle of superposition is simple but fundamental for all geologists and belongs to the first steps of understanding that a geology student must acquire. The principle states that geological strata were originally deposited horizontally under the action of gravity, and that the upper strata are younger than the lower are. The principle also states that inclined or folded strata have been tilted or disturbed subsequent to their deposition. The principle of superposition, theoretically, builds on Steno’s statement that a crystal, sediment layer, or other kind of geological structure that takes the shape of the surface of another solid body is younger than the body from which it takes its shape. In conjunction with the principle of superposition, Steno’s principle of intersection says that a body of rock or other geological structure is younger than those rocks or structures it is found cutting through. Steno held forth yet another important stratigraphic principle, the principle of reconstruction. This states that it is possible to ‘backstrip’ a series of geological changes in reverse order, from the present to the past, having regard to the principles of superposition and
intersection. In this way it is possible to obtain knowledge about still older situations. Thereafter, with this knowledge about past situations and their order of occurrence, the geological history of a locality, from the past to the present, can be reconstructed. In mineralogy, Steno was the first to describe the principle of crystal growth, which leads to constant and specific angles between the sides of crystals of specific minerals (Steno’s Law). In the philosophy of science and natural history, Steno founded the principle of recognitive induction, which made it possible to separate palaeontology and historical geology from theology.
Career, Science, and Beliefs Steno travelled through large parts of Europe, visiting renowned scientists, academies, and universities, as if receiving scientific inspiration from the landscapes he saw. Before he was 30 years old, his anatomical studies had made him famous; in 1667, he was attached to the Medici Court in Florence, where Grand Duke Ferdinand II, impressed by Steno’s anatomical and preliminary studies on fossils, made him a member of Accademia del Cimento (‘Academy of Experiments’). During the next 2 years, Steno established the most important and permanent principles of what were to become the geological core disciplines: palaeontology, stratigraphy, and mineralogy. Steno was born in Copenhagen of a Danish mother, Anne, and a Scanian father, Sten Pedersen. Niels was a fragile child and was brought up in the orthodox Protestantism of the Copenhagen of those days. Due to illness, he was kept indoors from his third to his sixth year. Isolated from other children, he listened to his parents and their friends’ religious conversations
FAMOUS GEOLOGISTS/Steno 227
Figure 1 Steno (Danish: Niels Stensen or occasionally Steensen i.e., Niels, the son of Sten; Latin: Nicolai Stenonis or abbreviated Steno; French: Nicolas Ste´non; Italian: Niccolo` Stenone), as he most likely appeared ca.1667 69. Contemporaneous portrait, by the Dutch court painter Justus Sustermanns (1597 1681). The ori ginal is in the Uffizzi Gallery, Florence; a copy hangs in the Institute of Medical Anatomy, Copenhagen University. This ver sion is from a poster made for a Steno exhibition at Tromsø University, Norway, in 1999. In other portraits of Steno can be seen his sigilum, a stylized asymmetrical heart from which a cross rises, a suggestion that scientific knowledge provides the highest praise to God.
and became familiar with mechanical and chemical crafts in his father’s respected goldsmith shop. But shortly after Niels’ recovery from illness, his father suddenly died. Because the boy was gifted, he was sent to Vor Frue Skole, a Lutheran academy, where he learned his fluent Latin from the enthusiastic Ole Borch. Borch also took Niels and the other pupils on botanical excursions around Copenhagen. Subsequently, Borch became one of Denmark’s most distinguished intellectuals, holding professorships at the university in poetry, philology, chemistry, and botany. More than anyone else, Borch turned the young Steno towards becoming a scientist. He visited Steno in Holland and their friendship continued until the end of Steno’s life. In his eighteenth year, Steno enrolled at Copenhagen University, where he came under the influence of Thomas Bartholin, who, as head of the Faculty of Medicine, was famous as discoverer of the lymphatic vessels. At the time, Denmark and Sweden were at war. The city was besieged, the university was closed, and Steno assisted the students’ defence of the barricades. Following the cease-fire in 1659, he managed to get out of the city by taking a ship to Rostock.
Shortly afterwards, Steno turned up in Amsterdam, encountering a new world of scientific opportunities. Steno soon moved to Leiden, where he continued to study the glands, the muscles, and the heart, and where he developed friendships with Jan Swammerdam and Baruch Spinoza. Steno went to Paris in 1665, where he presented his theories on the human brain and on muscles. According to a contemporary reviewer, Steno ‘‘turned upside down what is basic in medicine’’. Thus, besides Steno’s geological and philosophical contributions, he is also famous for some important anatomical discoveries. The discovery of the duct from the parotid gland to the mouth is named ductus stenonianus, in recognition of just one of Steno’s many anatomical contributions. Also important, but hardly recognized, was his description and understanding of the threefold division of the body fluids. In 1665, Steno gave the first modern description of the human brain, contradicting the interpretations of Galen, Willis, and Descartes. Moreover, the modern understanding of the anatomy and function of muscles and muscle fibres should be attributed to Steno. After a year in Paris, Steno travelled to Italy, where his geological interest was to flourish as a member of the Medici court in Florence from 1667. On his way, he passed through southern France, where, at the University of Montpellier, he met the Englishmen Martin Lister and John Ray. Steno’s geological achievements from his years in Tuscany, and his contribution to the principles of modern science, as they developed from Bacon, Galileo, and Descartes, have hardly received the reputation they deserve. Being a Dane (writing for the most part in a beautiful Latin, which was then in decline, rather than in the up-and-coming French and Italian, and only privately in English and German), he did not contribute to the national pride and fame of any large country or court. Moreover, his contemporary reputation was hindered by his criticisms of some of his most influential scientific contemporaries and by a superficial understanding of his religious conversion to Catholicism in the year (1669) that he wrote his most important geological work, De Solido Intra Solidum Naturaliter Contento Dissertationis Prodromus (translation: The Prodromus to a Dissertation Concerning Solids Naturally Enclosed in Other Solids). Eighteenth-century writers were puzzled by Steno’s conversion from Protestantism to Catholicism and by his shift from an academic to a clerical career. His geological methods were, however, promptly applied in England, Germany, and Italy, but his name was rarely mentioned before Lyell, Humboldt, and E´ lie de Beaumont drew attention to his work in the 1830s. In 1671, Steno was recalled to Copenhagen by Christian V, who, due to Steno’s Catholic faith, could
228 FAMOUS GEOLOGISTS/Steno
not employ him as a university professor. Instead, the king made him Royal Anatomist, a title invented for the purpose. After a long journey through large parts of south-eastern Europe, Steno finally arrived in Copenhagen. However, in 1675, he obtained the king’s permission to leave, and shortly after he became a Catholic priest in Italy. Soon the Pope called him to Hannover and Steno was appointed ‘Bishop of Titiopolis’, a nowdefunct city of the Eastern Roman empire in Asia Minor. In reality, Steno was called by the Pope to lead the Catholic mission in northern Germany, Denmark, and Norway. In Hannover, Steno met the Duke’s librarian, Leibniz, who, after having read Steno’s geological work, De Solido, in 1669, was inspired to write his theory, Protogea, concerning the Earth’s origin. Leibniz became an admirer of Steno’s geology and used him as scientific mentor and ‘censor’. However, Leibniz was unsuccessful in persuading Steno to recommence his geological work. Most likely, Steno told Leibniz that he had, while in Florence, expanded his geological ‘prodromus’ (De Solido) into a more comprehensive geological dissertation, and had given it to Holger Jacobaeus, who was later Professor of Geography at Copenhagen University. After Steno’s death, Leibniz wrote to several scientists, attempting to find out what had happened to Steno’s geological papers. However, the extended version of De Solido has never been found. After the period in Hannover, Steno dealt with numerous theological matters. These writings are collected in his Opera Theologica and Letters, and Steno did not really write on science any more, except for a philosophical letter to Spinoza. In the letter, written a few days before he was made ‘Bishop of Titiopolis’, Steno criticized Spinoza for having adopted a materialistic ‘religion’. Steno died in Schwerin when he was 48 years old, weakened through several years of religiously inspired poverty and self-inflicted fasting. He was beatified in 1988.
Philosophy of Fossils and Recognition In 1667, at the request of Ferdinand II, Steno wrote a brief dissertation (Canis carchariae dissectum caput) on the similarity between the teeth of living sharks and so-called tongue-stones, or glossopetrae (glossa: tongue; petrus: stone), now interpreted as fossilized shark teeth. Through this work, Steno is considered to be the first scientist to have established a series of empirical and rigorous arguments in order to describe fossils and interpret them as the geologically preserved remains of former living organisms. In six ‘conjectures’ (conjecturae), Steno explained that solid ‘bodies’ resembling parts of marine animals are
indeed the remains of the things they resemble, provided the resemblance is found on every scale and in every visible detail. Further, Steno stated that such things do not grow in the Earth but have been deposited there by natural processes in the past, and that fossils should not be understood as inexplicable imprints of God’s finger, but as representatives of the things they resemble (Figure 2). In De Solido (1669), Steno made a general statement of his six ‘conjectures’ on the origin on fossils, formulated 2 years earlier. Now Steno enunciated a general geological, as well as a basic philosophical, principle: the principle of recognition: ‘‘If a solid body resembles another solid body in all respects, not only in the state of its surface but also in the internal arrangement of the parts and particles, it will resemble it also in the method and place of production’’. Steno mentioned that the similarity may only be structural and textural, and not necessarily chemical. He noticed that some fossil shells have been petrified, or the material substituted by other ‘smallest parts’ (elements, minerals, and sedimentary particles), different from the materials of which they were originally formed. The shape and visible structures of the original body may still be preserved even though the original material has been substituted by sediment or some type of mineral other than that of the original crystal, bone, or shell. Giving examples and descriptions, Steno further declared that the ‘principle of recognition’ is valid not only for fossils but also for geological strata, crystals, and any other solid body embedded in the earth. Nature’s laws are ‘univocal’, thus similar conditions produce similar products. A scientist should believe in direct observations and in reasoning derived from the observations, even if such reasoning implies a historical development of Earth and dramatic changes in the distribution of land and sea through time. A scientist should not trust speculations when they are contradicted by observations. The human ability to recognize things is inherent and makes possible the basic method of empirical science. In 1673, Steno further explained his conception of the senses and human reasoning: ‘‘It is not the function of the senses to display things as they are or to judge them, but to transmit to the reason those conditions of the things to be examined, which are sufficient for acquiring a knowledge of things appropriate to man’s purpose’’. In other words, Steno realized that recognition is the most fundamental cognitive capacity in humans. It is prior to recognitive induction, or generalization, in science. Recognition is a prerequisite for generalization. A priori skills are ‘above’ a posteriori skills. Thus, Steno’s principle of recognition is not only a geological principle, but also a general theory about cause and effect and regularity in nature. Furthermore,
FAMOUS GEOLOGISTS/Steno 229
Figure 2 Steno’s personal drawings of fossil shark teeth (A), compared with a contemporaneous artist’s imaginative drawing of a monster shark (B) caught near Livorno and dissected by Steno. Note how the artist has been instructed to draw the anatomically correct arrangement of the teeth, which, when they are worn off, are substituted by new teeth rolling forward (another discovery made by Steno).
his principle was probably history’s first theory of perception that was built on anatomical arguments relating to the capacity of the sensing organs and the brain to obtain relevant, albeit partial, knowledge from the evidence provided by nature.
Philosophy of Stratigraphy and Reconstruction During his travels through the mountainous regions of southern Europe and during his numerous excursions in Tuscany, Steno noticed a large number of geological structures. In Steno’s time, these structures had found no scientific explanation. At the time, most, if not all, scientists believed that landscapes and structures in Earth’s crust derived from the time of Creation or had been formed during the turmoil of the Flood. However, through his studies of fossils, Steno became convinced that Earth had a history, accessible to scientific and human understanding over and above (though not generally conflicting with) the explanations in the Scripture. His interest was directed to all scales, not simply the small-scale structures of minerals and fossils. The structure of rocks, strata, and formations and their similarity
on both sides of gorges and valleys became another philosophical and perceptual problem that Steno wanted to solve. Similarly, he wanted to find methods for exploring the history of the large-scale structures of mountain ranges. Through his work on fossils from 1667, and 2 years later in De Solido, Steno dared to formulate and apply the core of his philosophy of science. This led to his definition of the fundamental stratigraphic principles of superposition and intersection, on how to find chronological and causal clues in geological bodies, in order to reconstruct their history. Moreover, it led to understanding of the general principles of crystal growth (Steno’s Law). In consequence, De Solido is generally considered as the first scientific work on geology. The basis of Steno’s geological methods was a combined actualistic and ‘forensic’ procedure, proclaimed in the introduction of De Solido: ‘‘Given a substance endowed with a certain shape, and produced according to the laws of nature, to find in the body itself clues disclosing the place and manner of its production’’. Steno proclaimed that geological structures should be read according to the assumption that the present laws of nature were also in operation in the past.
230 FAMOUS GEOLOGISTS/Steno
Moreover, the natural structures of solid rocks and strata provided clues that could lead to ‘demonstrative certain’ understandings of how and in which environment (place) any given geological structure has been produced. De Solido is structured in five parts: (1) the aforementioned introduction to the Grand Duke, (2) a philosophical part, in which the fundamental principles and methods are explained and formulated in general terms, (3) an empirical part, with numerous examples on how to apply the fundamental principles, (4) a description and interpretation of Tuscany’s geological history, based on Steno’s fundamental principles, and (5) acknowledgements from Steno’s ‘peer reviewers’, the eminent scientists Vincenzio Vivianni and Francesco Redi. In the second (philosophical) part of De Solido, Steno summarized his geological understanding in three general ‘propositions’ about the way nature works and how it should be ‘read’. The proposition on fossils and recognition was derived from Steno’s previous work on fossils. A new proposition described the basis of stratigraphy and reconstruction: ‘‘If a solid body is enclosed on all sides by another body, the first of the two to harden [to attain a certain form] was that one which, when both touch, transferred its own surface characteristics to the surface of the other’’. From this general ‘proposition’ Steno developed a series of chronological principles, including the principles of superposition, intersection, and reconstruction. He gave a long series of examples from sediments, volcanic rocks, veins, crystals, fossils, etc., on how his principles works in practice. He realized that the principles would make it possible to reconstruct a scientifically plausible description of the historic development of Earth. He emphasized that the reconstructive method could show the succession and type of geological events, though the duration of the geological history was still unknown: ‘‘On this issue Nature is silent, only Scripture speaks’’. By the help of his general ‘propositions’ and the associated superposition principle in Part 4 of De Solido, Steno showed how the geological history of Tuscany could be separated into six stages. There were two stages when the region was flooded by water and when its geological strata were deposited, two stages when it was flat and dry land formed by crustal uplifting of the strata previously deposited in water, and two stages when it was an uneven mountainous landscape eroded by rivers and deformed so that previously horizontal strata had been tilted, and again covered by younger horizontal strata. In a cartoon-like series of didactic drawings (see Figure 3A), Steno showed how to reconstruct a region’s geological history. The reconstruction must begin with the present state of affairs. Then, by the help of the
superposition principle, it must be discovered what the situation was immediately prior to the present. When that is known, the situation immediately before this second-last situation must be discovered, and so on, with the third-last, fourth-last, etc., until it is impossible to identify any older situation. Then, when the different situations and their order of occurrence are known, the geological history can be reconstructed, beginning with the oldest known situation. By good fortune, Steno’s approval from the ‘peer reviewers’ had been easy to obtain, because he had worked in an area that could be interpreted relatively easily, and where there appeared to be no serious contradiction between Nature and Scripture. Tuscany had been flooded twice, first at the time of the Creation, before animals and plants lived on Earth (for which reason no fossils are found in the first sediments deposited by water), and again during the Flood and other marine transgressions (that is, after the creation of animals and plants, for which reason fossils are found in the sediments deposited during the Flood or later).
Philosophy of Crystals and Growth Steno’s third proposition in De Solido deals with the nature of growth. This included all kinds of natural growth, no matter whether it takes place in the inorganic or in the organic realm. Steno was inspired by Kepler’s mathematical study of dense packing of ‘atoms’ and how snow crystals become a certain shape. From his own studies of sediments and crystals, however, Steno realized that crystal growth will give rise to regular external forms that cannot be produced by sedimentary processes, but also to structures more complex than those that can be produced by the packing of identical ‘atoms’ (see Figure 3B). Steno insisted that growth must be understood as a general problem, not only for crystals. He concluded that ‘‘if a solid body was produced according to the laws of nature, it was produced from a fluid.’’ On reading De Solido and Steno’s earlier anatomical papers, it is easy to understand how he reached this perceptive view. Superficially, the third proposition may seem odd. However, it reflects deep insight into change: all changes are results of motion. Motion is expressed in three basic forms: (1) as when we make a journey or an animal is running (i.e., change of location), (2) as when water runs in a river (i.e., flow), and (3) as ‘‘the first and hitherto unknown cause of motion’’, which Steno had already (in De Thermis) described as heat, and now also described as the motion of matter’s smallest parts (i.e., diffusion). Thus, Steno envisaged three fundamental types of change: in modern terminology, this is change of location (or dislocation), flow, and diffusion.
FAMOUS GEOLOGISTS/Steno 231
Figure 3 (A) Steno’s model showing how to reconstruct the geological history of Tuscany. (B) Drawings of various crystal forms and indications of how crystals grow and dissolve.
232 FAMOUS GEOLOGISTS/Steno
All other kinds of change can be reduced to combinations of these three basic types! Now, after this third proposition, Steno explained geological types of growth. Sediments increase with the help of gravity, which adds sand grains and other dispersed particles to the bottom from a slurry, or by traction of particles along the bottom until they hit obstacles. Crystals also grow by external addition, but for different reasons. The attractor is not gravity, but some unknown force, because crystals may grow even from the roof of a cavity or cavern. Steno explained geological growth as follows: Additions made directly to a solid from an external fluid sometimes fall to the bottom because of their own weight, as is the case with sediments; sometimes the additions are made from a penetrating fluid that directs material to the solid on all sides, as is the case of incrust ations, or only to certain parts of the solid, as is the case of those bodies that show thread like forms, branches, and angular bodies.
There is no doubt that Steno held that all kinds of solid growth, whether inorganic or organic, sooner or later could be explained as being due to an external addition of dissolved or dispersed material to preexisting surfaces. The unsolved problem appeared not to be how growth takes place, but how crystals begin to crystallize and maintain a certain form as expressed by the constant angle between their crystal faces. The solution to these problems had to be found in the imperceptible smallest parts of the crystallizing matter, because neither the first ‘germ’ of a crystal nor its developing faces take shape from the substrate on which it grows. The growing crystal ‘moulds’ the substrate, but its faces are different from the latter. Kepler’s paper on the dense packing of identical ‘atoms’ did not explain this, because it would imply that crystals could only be hexagonal, trigonal, or cubic, but not rhombic, and certainly not monoclinic or triclinic, as Steno knew some crystals to be.
Philosophy of Science and the Limits of Knowledge The fundamentals of Steno’s inductive/empirical philosophy of science were formulated in 1665 in Paris, when Steno gave history’s first modern description of the human brain, including realistic drawings, completely different from those of his contemporaries Willis and Descartes. Steno opposed Descartes’ conception of the brain and showed that Descartes’ theory was built on pure speculation about God’s impact on the human will, acting through the pineal gland. In Descartes’ opinion, God controlled the soul and the human will by vibrating and rotating the pineal gland
at the centre of the brain. Then, when the pineal gland, by such ‘divine impacts’, made the gland touch various parts of the brain, the contacts supposedly made the body react correspondingly, as if the brain were a mechanical machine and God its driver. On the contrary, said Steno, the brain is so fragile, and its structures so fine and complicated, that it ‘‘cannot even comprehend itself’’. Prompted by his anatomical researches, matured through his founding of a scientific basis for the study of Earth, and made humble by his strong religious beliefs, Steno came to a clear and rigorous philosophy of science, close to that of modern scientists. During his geological studies of Tuscany and under the influence of what he had come to understand about Earth’s history and changes, Steno developed his philosophy of science in order to separate geology and medicine from theology. When encountering something that is not understood, it is necessary to find something in it that is intelligible and compare it with something that is known and can be produced. The philosophical basis for De Solido can thus be summarized as follows: Specific recognitive induction (recognition) and specific empirical investigations (experiments, dissections, fieldwork, etc.) must regulate more general deductions and speculations (generalizations), but must not overrule deductive reasoning and perspectives, which for obvious reasons cannot be observed by the human senses or comprehended by the human mind. The past must be studied through knowledge about the present, but the past and present realities are much greater than scientific knowledge about it can ever be. This should be understood to mean that it is not possible to observe the past per se, but only ‘imprints’ of past events. So, when seeking to interpret the past, primary emphasis must be placed on those clues that can actually be observed. This should be understood so that explanations about the inability to know anything directly are not neglected. However, such perspectives must, in contrast to Descartes’ misuse of deduction, always respect what is known with the aid of the senses and by rigorous reasoning. This way of thinking led to a general ‘Kantian’ (though pre-Kant) theory of human perception and interpretation of nature. In his Copenhagen lecture of 1673, Steno generalized his views on the difference between things ‘as we see them’ and things as they are ‘in themselves’. There will always be a difference between nature as it is, and nature as humans interpret it. Modesty, caution, and scientific rigour should be key in attempts to understand things. Humans must believe in their immediate sensory capacities, and in what scientifically founded investigations reveal to the senses. Finally, it is important to believe in what the senses
FAMOUS GEOLOGISTS/Suess 233
transmit to the reasoning. However, because human senses are imperfect and reasoning capacity is incomplete, the truth can only be approached, and complete understanding cannot be obtained. Steno concentrated his philosophy of science in the following sentence, which has been cited more than anything else from his hand: Pulchra sunt quae videntur, Pulchriora quae sciuntur, Longe pulcherrima quae ignorantur.
In English, this reads as follows: Beautiful is what we see, More beautiful is what we know, Most beautiful is that about which we are insensible.
This has erroneously been interpreted to mean that Steno ranked religious belief above scientific knowledge. But he explained the aphorism in this way: Yes indeed, after having rejected all the errors of the senses, who would not repeat: beautiful is what appears to the senses without dissection; more beautiful what dissection draws forth from the hidden interior; yet far the most beautiful is what, escaping the senses, is revealed by reasoning helped by what the senses have already perceived.
In Steno’s philosophy, humility about scientific principles and scientific understanding offered the highest praise to God. Scientific knowledge must never be ruled by clerical beliefs and powers. On the contrary, science will guide us towards the truth, which, however, will never be fully understood because of our limited sense capacity and imperfect intellectual resources.
See Also History of Geology Up To 1780.
Further Reading Cutler A (2003) The Seashell on the Mountaintop. A Story of Science, Sainthood, and the Humble Genius who Discovered a New History of the Earth. New York: EP Dutton. Garboe A (1954) Nicolaus Steno (Niels Stensen) and Eras mus Bartholinus: two 17th century Danish scientists and the foundation of exact geology and crystallography. Bulletin of the Geological Survey of Denmark. 4th Series 3: 1 12. Garboe A (1960) Niels Stensen’s (Steno’s) lost geo logical manuscript. Bulletin of the Geological Survey of Denmark 14: 243 246. Gould SJ (1981) The titular bishop of Titiopolis. Natural History 90: 20 24. Kardel T (1994) Steno: life, science, philosophy (with Niels Stensen’s Prooemium or preface to a demonstration in the Copenhagen Anatomical Theater in the year 1673, and Holger Jacobaeus: Niels Stensen’s Anatomical dem onstration No. XVI, and other texts translated from Latin). Acta Historica Scientiarum Naturalium et Medi cinalium 42: 1 159. Moe H (1988) Nicolaus Steno: An Illustrated Biography. Copenhagen: Rhodos. Noe Nygaard A (1986) Nicolaus Steno, paleontologist, geologist, crystallographer. In: Poulsen JE and Snorrason E (eds.) Nicolaus Steno 1638 1686. A Re consideration by Danish Scientists, pp. 167 190. Copenhagen: Nordisk Insulinlaboratorium. Rodolico F (1971) Niels Stensen, founder of the geology of Tuscany. Acta Historica Scientiarum Naturalium et Medicinalium 23: 237 243. Rudwick MJS (1972) The Meaning of Fossils. New York: MacDonald, Elsevier. Scherz G (ed.) (1969) Steno: Geological Papers (translated by AJ Pollock). Odense: Odense University Press. Steno N (1669) De Solido Intra Solidum Naturaliter Contento Dissertationis Prodromus. Florence: Ex Typo graphia Sub Signo Stellae (English translation in Scherz, 1969).
Suess B Fritscher, Munich University, Munich, Germany ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Eduard Suess (Figure 1) was the most influential European geologist of late nineteenth and early twentieth centuries. As a professor of geology, he taught at the University of Vienna for nearly 45 years. In his major works, Die Entstehung der Alpen (The Origin of the
Alps; 1875), and in the comprehensive Das Antlitz der Erde (The Face of the Earth; 1883–1909), he elaborated a ‘global tectonics’, based on the contracting hypothesis. By his works he created a new image of the structure and the formation of fold mountains and introduced basic terms of twentieth-century structural geology, such as the Laurentian and Angara Shields, the huge southern continent Gondwana, and the Tethys (as a former central sea, the precursor of the Mediterranean). Moreover, Suess was one of the pioneers of the doctrine of nappe folding in the Alps and he
234 FAMOUS GEOLOGISTS/Suess
Figure 1 Eduard Suess (1831 1914), in a portrait published in his posthumous 1916 memoir, Erinnerungen.
founded the concept of ‘eustatic’ sea-level changes. Suess is also remembered as an engineer and politician. He was a long-serving member of the Austrian national parliament and promoted and planned a new water supply for his home town, thus becoming one of the creators of modern Vienna.
Scientist, Engineer, and Politician The founder of ‘global tectonics’, Suess was born into a well-connected bourgeois family. His father, a wool merchant, born in Germany, had worked and travelled in various European countries before setting up a wool business in London in 1828. Here, Eduard Suess was born on 20 August 1831. Three years later, the family moved to Prague, where Suess, as he later recalled, arrived as a ‘complete English child’ who understood not a single German word. In 1845, his father took over a leather factory near Vienna. The young Eduard got a polyglot education from English, Belgian, and German tutors. In 1847, he entered the Polytechnic ‘High School’ (now Technical University) at Vienna, but soon left the revolutionary town of 1848, where he had participated in demonstrations, for the University of Prague. His early subjects were higher mathematics, physics, and descriptive geometry. The impressive collection of Silurian fossils at the museum in Prague roused his interest in geology, and he started to make excursions
to nearby fossil-rich areas. Back at the Vienna Polytechnic in 1849, he continued his palaeontological studies in the surroundings of Vienna. In 1850, he presented a scientific paper (published in 1851) on Bohemian graptolites to Wilhelm Haidinger (1795–1871), who was then director of the newly established Austrian Geological Survey. His very first publication, however, had been a chapter on geology for a tourist guide of the Carlsbad region; Suess had written the chapter in 1850 during a visit to the region for a ‘cure’. After returning to Vienna, Suess was imprisoned for his participation in the demonstrations in 1848. Although he was released just a few weeks later, he had to leave the Polytechnic School, thus never acquiring a doctorate or any other formal university qualification. Nevertheless, in 1852, he was appointed an assistant in the Imperial Mineralogical Collection in Vienna. Assigned to classify the brachiopods of the collection, he published some pioneering studies in this field. His efforts to become a Privatdozent (private lecturer) at the university failed for his lack of a doctorate. By his palaeontological work, however, and by early travels accompanying well-known geologists such as Franz von Hauer (1822–99), Arnold Escher von der Linth (1807–72), Paul Deshayes (1797–1875), and Ernst Beyrich (1815–96), Suess already had a name among earth scientists. Thus, in 1857, on the recommendation of leading Vienna geologists, he was appointed ‘professor extraordinary’ of palaeontology at the university, and 5 years later was appointed the same position in geology. In 1867, Suess was appointed to a full professorship in geology, which he retained until 1901. For about three decades, Suess also travelled extensively throughout Europe. An early engagement in school and university education marked the beginnings of Suess’ political career. In 1862, he published an essay on the soils and the water supply of Vienna, showing that the numerous epidemics of that time, particularly typhoid, resulted from the city’s water supply, which was at that time mainly based on wells. The following year, Suess was elected a member of the town council and was named head of a commission to study the water supply. He suggested that water should be brought by an aqueduct from mountains springs, about 70 km away; 10 years later, in 1873, the new pipeline began to operate, and the number of deaths from typhoid fever was subsequently reduced to one-fourth. Suess’ second famous engineering project was the regulation of the Danube, designed to prevent the frequent flooding of the lower lying areas of Vienna. A canal was opened in 1875, and after 1876, there were no more major floods. Suess was also a member of the Diet of Lower Austria from 1869 to 1874, and he held a seat in the Austrian
FAMOUS GEOLOGISTS/Suess 235
Parliament from 1873 to 1896, being mainly engaged in implementing liberal reforms in the school system. Suess was also an ordinary member of the Austrian Academy of Sciences from 1867, and served as its President between 1898 and 1911. The advancement of scientific cooperation between different scientific disciplines and different national Academies was one of his main goals. He also promoted the foundation of the Institute of Radium Research in Vienna in 1910. Suess was elected a member of various European scientific academies and the Geological Society of London awarded to him the Wollaston Medal in 1896. Suess died in Vienna on 26 April 1914 and was buried at Marz (Burgenland, Austria), where his family owned a small estate.
Palaeontology, Stratigraphy, and Eustatic Sea-Level Changes Following his early studies of Silurian graptolites in Bohemia, Suess’ main fields of study at the Imperial Mineralogical Collection were the palaeontology and stratigraphy of the Tertiary strata of the Vienna Basin. He entered scientific virgin territory by his brachiopod studies and was the first Austrian palaeontologist to be engaged in the classification of Tertiary mammals, thus providing an overview of the mammalian fauna of the Vienna Basin. Suess set the comparative method against the prevailing emphasis on merely taxonomic classification. Focusing on the anatomy of fossil species and comparing their modes of life with those of existing species, he created an early form of palaeobiology. As early as 1859, he discussed the ecology of brachiopods, and in 1875 he first used the term ‘biosphere’ (Biospha¨re) to denote the distinct terrestrial sphere of the living organisms, which could be regarded as the surface of the lithosphere. These approaches were continued by his longstanding assistant Othenio Abel (1875–1946), who became one of the founders of modern palaeobiology. Suess’ stratigraphic work was mainly concerned with the Alps. It brought a new structural view to alpine stratigraphy, distinguishing chronological and spatial units. Suess cleared up stratigraphic problems of the European Rhaetian and Miocene, the latter in particular in the region of Eggenburg (lower Austria), north of Vienna. Basic studies related to the correlation of Alpine Triassic, Jurassic, and Cretaceous formations with their equivalents outside the Alps, and, together with Albert Oppel (1831–65), Suess correlated the development of the Alpine and Swabian Triassic strata. The region of Eggenburg was the area where Suess first developed his concept of eustatic sea-level changes. In surveying the Tertiary beds, he remarked
Figure 2 Sketch of Eggenburg (lower Austria) by Eduard Suess, from Suess’ geological diaries, as published in Erinnerun gen in 1916. The region of Eggenburg was crucial for Suess’ ideas on large scale variations of sea level. In surveying the regularity of the ancient shorelines of this area, Suess first thought of what were later called ‘eustatic’ movements of the sea.
the regular height of the ancient shorelines of this area (Figure 2). These regularities seemed hardly explicable by an uplift of the land but, rather, by a fall of the sea-level. In 1885, after visiting Norway, he thought his ideas of the fall of the sea-level were confirmed by the stepped, horizontal terraces he had observed on the sides of the fjords and other valleys. Thus, 3 years later, he presented his theory of ‘eustatic movements’, i.e., of large-scale changes of the sea-level (separate from orogenic belts), which could be observed at approximately the same height over large parts of the earth.
The Origin of the Alps In 1865, Suess was commissioned to produce a survey of the geology of the Austrian Empire, comprising at that time Hungary, Czechoslovakia, and parts of Romania and Poland. In the following
236 FAMOUS GEOLOGISTS/Suess
FAMOUS GEOLOGISTS/Suess 237
years, he extended his field studies to the Carpathians, the Sudetes, and the Apennines. In applying the comparative method, which he had already used in his palaeontological work, he found that all these mountain ranges (including the Alps) had certain structures in common. As a result, in 1875, he published his first major book, The Origin of the Alps, a comprehensive discussion of the origin and the structure of mountain chains, anticipating most of his later ideas on tectonics. Suess’s actual entry to these ideas involved studies on earthquakes, following a visit to Calabria in 1871, where Suess witnessed the devastation of the great earthquake of 1870. This gave rise to the idea of a systematic compilation of historical accounts of earth tremors in lower Austria, i.e., in a region far away from any volcanic activity. Suess found that these Austrian earthquakes were distributed along specific lines that cut across quite different rock formations. Thus, in 1873, he published two major articles on earthquakes in lower Austria and in southern Italy, concluding that earthquakes are restricted to specific structures within Earth’s crust and are thus due to the same forces as those that gave rise to the formation of mountains (Figure 3). The revolutionary concept that Suess now set forth in The Origin of the Alps abandoned the idea of similarity of the structure of mountain chains, which had dominated geology for nearly a century. Contrary to the theory of mountain formation by vertical upheavals due to eruptive rocks, favoured by Leopold von Buch (1774–1853) and Leonce E´ lie de Beaumont (1798–1874), Suess set forth his view of horizontal movements as the essential cause of the formation of folded mountain chains, entertaining the idea of unilateral horizontal overthrustings by tangential pressures, in the case of the Alps, directed from south to north. And Suess stated a fundamental difference between the mountain chains and their older, rigid ‘forelands’ (Vorla¨ nder), which act, so to speak, as ‘earth dams’ against the mobile chains. As a further characteristic feature of developing mountain chains, he put forward the idea of ‘hinterlands’, i.e., their usually curved and relatively depressed ‘inner’ sides (Figure 4). These inner sides were the location of volcanism and earthquakes, due to the tensions caused by the movement of the newly forming mountains towards the forelands. Already during his visit
Figure 4 Sketch of the main lines of strike for the folds of the Carpathians and the Balkans, from the first volume (1885) of Suess’ Das Antlitz der Erde. The Carpathian Mountains, in particu lar, initiated Suess’ thoughts about groups of curved lines, and tangential and unilateral movements, as opposed to the prevail ing assumption of a symmetrical structure of folded mountain chains. The actual structure of the mountain chains (i.e., the disposition of the folds) depended on the form of the foreland and the resistance of its materials.
to the volcanic areas of southern Italy in 1871, Suess had thought of volcanoes and earthquakes as being associated with regions of rupture within Earth’s crust. Thus, earthquakes and volcanism were closely related to the process of mountain formation. Suess explicitly stated that volcanoes and intrusions were just the side effects of mountain formation.
A Contracting Earth In The Origin of the Alps, Suess also first introduced his ideas on the forces that could have brought about mountain ranges and their accompanying phenomena: namely, the contraction or shrinking of the earth due to its continuous cooling. In the 1840s, the American geologist James Dwight Dana (1813– 1895) had proposed the idea of unequal radial contraction of the earth due to cooling, linking this to the origin of earthquakes. In the early 1870s, Dana and other geologists extended the contraction hypothesis as the basic assumption of the formation of mountains. Thus, Suess was neither the creator nor the chief advocate of the contraction hypothesis. However, it was Suess who – referring to Dana, to Robert Mallet’s (1810–81) ideas on ‘volcanic energy’, and to Charles Darwin’s (1809–82) theory of the origin of earthquakes due to the formation of fissures – most clearly used the contraction hypothesis for more than
Figure 3 A map from Suess’ 1873 book Die Erdbeben Nieder O¨ sterreichs, showing the distribution of earthquakes in lower Austria. According to the position of the strongest effect of the respective earthquakes (years given near the names of the villages), Suess constructed three ‘earthquake lines’ (Erdbebenlinien). The Thermen Linie (A B) was named according to the thermal and sulphur springs along this line (for instance, at Brunn, Voslau, and Meidling). The Kamp Linie (C D) and the Murz Linie (E F) were both named according to rivers of the area.
238 FAMOUS GEOLOGISTS/Suess
30 years as a basis for explaining all the tectonic features of Earth’s crust. Arguing against the theory of mountain formation by volcanic elevations, Suess pointed to his studies in the Alps, which had offered only a single example that might confirm this theory: in the Euganean Hills (Italy), where a mass of Jurassic and Cretaceous limestones seemed to have been moved by trachyte. In the final chapter of The Origin of the Alps, Suess related the structures of mountain ranges to irregular earth contraction, and first used his famous phrase of ‘the face of the earth’. He emphasized that, notwithstanding the irregularity of the contraction, its direction seemed to have remained the same over large regions and extremely long periods of time. In this, Suess was anticipating his later distinction of Caledonian, Armorican, and Variscan folding.
Global View: The Face of the Earth In 1885, Suess published the first volume of his best known work, The Face of the Earth. The second volume followed in 1888, and the two parts of Volume 3 appeared in 1901 and 1909, respectively. An English translation in five volumes was published between 1904 and 1924. The most important edition became the French translation under the direction of Emmanuel de Margerie (1862–1953), in which thousands of new footnotes and about 500 figures were added. The monumental work was not simply an extension and more detailed discussion of Suess’ previously formulated ideas. Rather, The Face of the Earth provided a ‘global tectonics’. Embodying the results of his own travels and research all over Europe, Suess compiled the then-known materials relating to the tectonic structures of Earth’s crust and created the ‘language’ that made possible a global view of the planet’s tectonic features (Figure 5). At the commencement of the tectonic development of the present landmasses, Suess distinguished four ancient continents: Laurentia, Angaraland (in what is now northern/central Siberia), Gondwana, and Antarctica (Figure 6). A further ancient block was the Baltic Shield. The later term for this, Fennoscandia, however, was actually not used by Suess. Between the Eurasian and the Indo-African blocks extended a series of younger mountain chains. Suess thought of them as having originated in the Tethys, an ancient Mediterranean sea encompassing half the globe, from Central America to the Sunda Islands (the name was coined according to the Greek goddess of the sea). A second series of young mountain chains, forming festoons and garlands, encircled the Pacific Ocean. In addition to the youngest epoch of folding, the Alpine Orogeny, Suess distinguished two more
principal stages of mountain formation in Europe: the Caledonian and the Variscan (also known as the Armorican and the Hercynian) orogenies. Their ages increased in passing from south to north. Though the most recent (Alpine) chains around the Mediterranean were already subsiding, the older ones, now eroded and covered by younger deposits, are presently at rest, but could be reactivated. According to Suess’ theory of mountain formation, the subsidence and the formation of large grabens were predominant characteristics of Earth’s crust, as against horizontal dislocations, such as folds and overthrusts. With reference to Suess’ hypothesized periods of mountain formation, in 1887 the French geologist Marcel Bertrand (1847–1907) developed a tectonic classification of Earth’s history, also correlating the different periods of folding with those of increased igneous intrusions. Thus, Suess’ periods of folding became a continuous process that had built up the European continent from south to north. In addition to these basics of a ‘global tectonics’, Suess introduced a series of further large-scale features of Earth’s surface, such as his distinction of different types of coasts. According to the direction of the axes of fold mountains, he distinguished a ‘Pacific’ and an ‘Atlantic’ type. Around the Pacific Ocean, the fold lines were running more or less parallel to the coast, whereas in the case of the Atlantic Ocean, they ran approximately at right angles to the coast. In the first volume of The Face of the Earth, Suess also further elaborated his ideas on contraction. Meanwhile, the hypothesis had received strong support by the Swiss geologist Albert Heim (1849–1937), who, in 1878, had calculated the supposed reduction of the circumference of Earth due to its shrinking or contraction. For the Jura mountains and the Alps, Heim estimated a relative compression of about four-fifths and one-half, respectively (compared to the original width, which was estimated by mentally ‘smoothing out’ both chains). Relative to the full circumference of Earth, the shrinking of Earth due to the folding would be almost 1%. Such a process (for which Heim thought a cooling of 200 C would be sufficient) going on throughout the whole history of Earth was quite plausible according to the contraction hypothesis. Relying on his ideas on earthquake lines, and on further detailed discussions of the various phenomena of Earth tremors, Suess constructed a system of dislocations in rock formations, due to the reduction of the volume of the globe. The tensions produced by the process of contraction would tend to differentiate into tangential and radial tensions, thus producing both horizontal (i.e., pushing and folding) and
FAMOUS GEOLOGISTS/Suess 239
Figure 5 Map showing divergent branching (Virgation) of the Rocky Mountains, from the first volume (1885) of Suess’ Das Antlitz der Erde. To convey the complexity of the structures of folded mountain chains, Suess often used the term ‘Virgation’, meaning the rodlike spreading out, or dispersion, of the individual branches towards their forelands. Such an order also meant that a region that was the ‘hinterland’ for one branch served as the ‘foreland’ for another. Translation of the key: A, Archaean rocks and granite; p, Palaeozoic; tj, Triassic and Jurassic; cr, Cretaceous; t, Tertiary; cross hatching, younger lava; a, Quaternary and alluvium.
240 FAMOUS GEOLOGISTS/Suess
Figure 6 Section of Suess’ map (from the third volume, part 2, of Das Antlitz der Erde), showing the arrangement of the tectonic units of Earth. The rose coloured areas are Laurentia and Gondwana; brown, Caledonides and Saharides; green, Asian structures with Angaraland; pale violet, Oceaniden, Australia, and Antarctica; yellow, Cape Mountains; cross hatched, volcanic islands of Atlantic type.
vertical (or subsiding) movements. Thus, Suess distinguished two groups of dislocations, one produced by more or less horizontal movements of mountains, the other one by more or less vertical movements (i.e., by subsidence). At the end of the first volume of The Face of the Earth, Suess gave his famous statement of the history (and the future) of Earth: ‘‘What we are witnessing is the collapse of the terrestrial globe’’. He also linked Earth’s contraction to the development of life. Subsidences had made possible the accumulation of water in the deep oceans, and, at the same time, the emergence of the continents, which became the home of organisms that breathe with lungs.
This idea of the linkage of the development of life on Earth to its tectonic history was elaborated at the end of the last volume of The Face of the Earth. Suess claimed that his ancient shields (Laurentia, Angaraland, Gondwana, and Antarctica) were the essential regions for the development of life. These areas supposedly did not participate in folding and transgressions for a long time. Consequently, the development of life in these areas should show fewer disturbances than elsewhere, and therefore Suess called them ‘asylums’ (Asyle). From the asylums, the distribution of the vegetation should have started again after the great tectonic changes.
FAMOUS GEOLOGISTS/Suess 241
Figure 7 The Glarus double fold (according to a sketch by Albert Heim, as published in volume 3, part 2, of Suess’ Das Antlitz der Erde), explained as a single overthrust towards the north. m, Tertiary Nagelfluh; e, Lower Tertiary flysch; c, Cretaceous; J, Jurassic; t, Helvetic Triassic; V, Verrucano. The sea level (Meeresniveau) is indicated.
Nappe Folding The publication of The Face of the Earth covered a period of 24 years, and some of Suess’ early concepts were changed during this time. The most striking example was the doctrine of nappe fold structures, as developed from the 1880s. Suess may have implicitly accepted large overthrusts as processes of mountain formation, at least since the 1870s. And he actually played a significant role in the development of the new doctrine, which is commonly ascribed to Marcel Bertrand. As early as 1883, Suess gave a new interpretation of Heim’s ‘Glarus double fold’ (Glarner Doppelfalte) in the Glarus Canton of Switzerland (Figure 7). Whereas Heim had the idea of two folds from both north and south, Suess suggested a single overthrust toward the north. Nevertheless, it was not until publication of the third volume of The Face of the Earth, and in particular its second part, that Suess included nappe folding in his discussion of the processes of mountain formation. He discussed several of these structures, as found in the young European mountain ranges, accepting also large amounts of overthrusting, such as, for instance, in the case of the Dent Blanche massif in Switzerland, which had recently been proposed by the Swiss geologist E´ mile Argand (1879–1940).
Suess in the Twentieth Century European geological thinking was deeply influenced by Suess’ tectonics: a great number of his concepts and terms became standard in twentieth century earth sciences and are still current. On the other hand, Suess’ basic theoretical assumption, i.e., the hypothesis of the contracting Earth, was subjected to critiques, even during the publication of The Face of the Earth. In 1912, only 3 years after the publication of the last volume and 2 years before Suess’ death, Alfred Wegener (1880–1930) came up with quite new ideas on the origin of continents and oceans. Wegener was Suess’ most serious critic. Moreover, the recognition of radiogenic heat, produced within Earth, threw grave doubt on the idea of a cooling and contracting
planet. Nevertheless, it was the Viennese geologist who provided Wegener with essential information about the large-scale features of Earth, indicating former connections and movements. Thus, notwithstanding Suess’ errors, his work remains among the most impressive and comprehensive scientific theories of Earth ever written.
See Also Africa: Pan-African Orogeny; North African Phanerozoic; Rift Valley. Andes. Antarctic. Argentina. Australia: Proterozoic; Tasman Orogenic Belt. Brazil. Famous Geologists: Wegener. History of Geology From 1835 To 1900. New Zealand. Oceania (Including Fiji, PNG and Solomons). Plate Tectonics. Shields. Tectonics: Earthquakes; Folding; Mountain Building and Orogeny. Volcanoes.
Further Reading Cernajsek T, Csendes P, Mentschl C, and Seidl J (1999) ‘‘. . .hat durch bedeutende Leistungen. . .das Wohl der Gemeinde ma¨chtig gefo¨ rdert.’’ Eduard Suess und die ¨ sterrei Entwicklung Wiens zur modernen Großstadt, O chisches Biographisches Lexikon Schriftenreihe 5. Vienna: ¨ sterreichisches Biographisches Lexikon. Institut O Greene MT (1982) Geology in the Nineteenth Century. Changing Views of a Changing World. Ithaca and London: Cornell University Press. Hamann G (ed.) (1983) Eduard Suess zum Gedenken ¨ ster (20.VIII.1831 26.IV.1914). Sitzungsberichte der O reichischen Akademie der Wissenschaften, Philologisch ¨ sterrei historische Klasse 422. Vienna: Verlag der O chischen Akademie der Wissenschaften. Sengo¨ r AMC (1982) Eduard Suess’ relations to the pre ) 1950 schools of thought in global tectonics. Geologische Rundschau 71: 381 420. Sengo¨ r AMC (1998) Die Tethys: vor 100 Jahren und ) heute. Mitteilungen der O ¨ sterreichischen Geologischen Gesellschaft 89: 5 177. Sengo¨ r AMC (2003) The Large Wave Deformations of ) the Lithosphere: Materials for a History of the Evolution of Thought from the Earliest Times to Plate Tectonics, Memoir 196. Boulder: Geological Society of America.
242 FAMOUS GEOLOGISTS/Walther
Suess E (1862) Der Boden der Stadt Wien nach seiner Bildungsweise, Beschaffenheit und seinen Beziehungen zum bu¨ rgerlichen Leben: Eine geologische Studie. Vienna: W Braumu¨ ller. ¨ sterreichs. Vienna: Suess E (1873) Die Erdbeben Nieder O K Gerold’s Sohn. Suess E (1875) Die Entstehung der Alpen. Vienna: W Braumu¨ ller. Suess E (1885 1909) Das Antlitz der Erde (three volumes; vol. 3 in two parts). Prague and Leipzig: F Tempsky and G Freytag. Suess E (1897 1918) La Face de la Terre (three volumes; vol. 3 in four parts) (Traduit et annote´ sous la direction de
E de Margerie. Avec une pre´ face par M Bertrand). Paris: Librairie Armand Colin. Suess E (1904 1924) The Face of the Earth (five volumes). (Translated by HBC Sollas, under the direction of WJ Sollas.) Oxford: Clarendon Press. Suess E (1916) Erinnerungen. Leipzig: S Hirzel. Tollmann A (1981/1982) Die Bedeutung von Eduard Suess ¨ sterreichischen fu¨ r die Deckenlehre. Mitteilungen der O Geologischen Gesellschaft, (special volume to mark the 150th anniversary of E. Suess’ birth) 74/75: 27 40. Wegmann E (1976) Eduard Suess. In: Gillispie CC (ed.) Dictionary of Scientific Biography 13, pp. 143 149. New York: Charles Scribner.
Walther I Seibold, University Library, Freiburg, Germany ß 2005, Elsevier Ltd. All Rights Reserved.
Career Johannes Walther (Figure 1) was one of the early pioneers in sedimentology, introducing a modern approach that combined both lithological and biological aspects. Walther was born on 20 July 1860, the son of a vicar in Neustadt/Orla in the German province of Thuringia. In his boyhood he was handicapped by a nervous disease (probably caused by a serious fall) that prevented his regular attendance at school and as a result he did not receive a leaving certificate. But his intelligence and enthusiasm for science were so evident that he obtained special permission to study at the University of Jena and his health was restored during his period as a student. Walther was awarded his PhD in zoology in 1882, following which he went to Leipzig and Munich to study geology and palaeontology more intensively. At the same time, he started his first investigations of the sea floor in the Bay of Naples, where Anton Dohrn’s marine biology station was based. In 1886, Walther became a lecturer in Jena; in the following years he undertook extended geological expeditions abroad. During his time as lecturer Walther was almost without income, for his position had no salary, apart from the small lecture fees. He therefore had to make a living by writing (e.g., newspaper articles) and giving public lectures. However, in 1894 he was appointed associate professor (Haeckel Professor) at Jena, a post endowed by a wealthy Swiss admirer of Walther’s teacher, the famous zoologist Ernst Haeckel. In 1899, Walther married Janna Hentschel. They had two children, a son and a daughter. From 1906
until his retirement, he was Director of the Geological Institute of Halle University. His work comprised more than 120 publications, including a dozen books, most of which appeared in several editions. Walther received honorary degrees from Perth and Melbourne (the latter 2 weeks after the outbreak of World War I!). In 1928, he was Visiting Professor at John Hopkins University, Baltimore. From 1924 to 1931, he served as President of the German Academy of Scientists, Leopoldina, in Halle. He died at Hofgastein, Austria, on 4 May 1937.
Figure 1 Johannes Walther.
FAMOUS GEOLOGISTS/Walther 243
Early in his career, Walther set himself the aim of ‘modernizing’ Lyell’s uniformitarianism. Thus he undertook journeys to study as many geological conditions and environments as possible, but he focused on two fields in particular: shallow marine environments (including reefs) and deserts. At the time when he wrote his principal works, palaeontology, stratigraphy, and tectonics dominated geology in the universities, but Walther did not focus on these fields. Sedimentary rocks were chiefly studied because of their fossil content, useful for stratigraphy, and not because they illustrated former environments. Strongly influenced by Haeckel, the German apostle of Darwinism, Walther applied his ideas to geology by looking at the mode of formation of sediments, and the processes that formed rocks, not just their characteristics. His credo was: ‘‘Aus dem Sein erkla¨ren wir das Werden’’ (From the present state [of a rock] we explain its origin). But he was well aware that some past processes do not occur today. Though physical laws remain the same, geological conditions vary and may even be unique. With this dynamic approach, Walther was able to integrate climatological, sedimentological, palaeontological, and other aspects into Amanz Gressly’s ‘facies’ concept (1838), which was the basis for Walther’s comprehensive consideration of facies. His view of the succession of palaeoenvironments was expressed in the ‘Law of Facies’, which was subsequently named after him (even though it had already been found by Gressly): ‘‘Es ist ein Grundsatz von weittragender Bedeutung, dass primaer sich nur solche Facies und Faciesbezirke geologisch u¨ berlagern ko¨ nnen, die in der Gegenwart nebeneinander zu beobachten sind’’ (‘‘It is a principle of far-reaching importance that only the facies or facies areas that are at present adjacent to one another can be geologically superimposed upon one another’’) (Walther [1894], p. 979; see also Middleton [1972]). In other words, the relative horizontal distribution of sediments with their organic content will be transformed into a vertical distribution, having a chronological order. Gressly found the rule during his extended fieldwork for his admirable monograph on the Swiss Jurassic near the town of Solothurn, where he carefully studied the facies changes. Fifty years later, Walther discovered Gressly’s study of 1838 anew, when he was working for his volume on lithogenesis. He explained and discussed this idea in detail in the first of three chapters on facies in his most interesting third volume (Lithogenesis of the Present) of his fundamental work, Einleitung in die Geologie als historische Wissenschaft (1894). This discussion of the ‘law’ was only a minor part of the total corpus of his work, but it is on this that his present reputation chiefly rests.
Walther stressed the importance of organisms in geological processes and vice versa: biogeology. The dependence of biocenoses (groups of organisms living together, forming natural ecological units) on their substrates means that lithology should have priority over palaeontology. He mentioned, for example, that the empty shells of index fossils can be transported over long distances and may, therefore, give false stratigraphic results. With his zoological background of comparative anatomy Walther thus advocated ‘comparative lithology’. This concept improved stratigraphy substantially and Amadeus W. Grabau (1870–1946) paid tribute to it by dedicating his classic Principles of Stratigraphy (1913) to Walther. The first of the other two volumes of Walther’s Introduction to Geology as Historical Science (see above), Bionomie des Meeres, was a treatise on marine biogeology that dealt with the interactions between the sea floor, fauna, and flora; beyond that it also had chapters on oceanography. (He had previously published a popular booklet on general oceanography in 1893.) The second volume was mostly a collection of faunal lists. All Walther’s pioneering work was undertaken in his years in Jena. One of his later publications was the voluminous Allgemeine Pala¨ ontologie (General Palaeontology) (1927). He wanted to give a summary of all the topics he had thought and taught about, and this was a lot. The book contained a wealth of varied ideas, though regrettably a number of them were already outdated by the time the book appeared. With its 809 pages, it presents difficult, yet still inspiring reading. For Walther, theories were much more important than details.
Special Contributions Shallow Marine
Walther’s final study on the sediments of Dove Bank (Taubenbank, 1910), a shoal in the Gulf of Naples, was a classic. He compared his first maps of 1884 with the latest findings of 1910 to evaluate the changes caused by volcanic activity (Vesuvius produced a great ash-fall in 1906) and the unusual storms of the intervening years (an early look at event stratigraphy!). His former studies of the rock-forming calcareous algae (1885) enabled him to compare his old and new results. He showed that coralline algae spread rapidly, consolidating sediments within 25 years. Perhaps even more interesting was Walther’s application of biological experiments to explain bioturbation, which he stated could occur down to 15 cm. He also measured the digging velocity of mussels. Using
244 FAMOUS GEOLOGISTS/Walther
a ‘hyperactualist’ method, he explained the formation of calcareous sands by keeping crayfish and mussels together in an aquarium and finding that four crayfish of 12–18 cm length worked a load of 580 g of mussels into detritus within 12 days. The final weight was 240 g. A ‘fossil example’ was provided by his study of the fauna of the Solnhofen sediments (Upper Jurassic, Bavaria, South Germany) (1904), a contribution for a volume in honour of Haeckel’s 70th birthday. Walther tried to show all features of a biotope by determining the frequency of marine and land fossils at various localities in the Solnhofen sediments. He noted the lack of freshwater and brackish species, and localized the presence of plant debris; by the decreasing frequency of occurrence of the planktonic crinoid Saccocoma, from a centre outwards in all directions, he concluded that the greatest depth of water had been in this central area. Besides his own observations, Walther used data from all available collections and fossil lists for his demonstration of the Solnhofen biotope. Unfortunately, this exemplary study was almost hidden among the zoological articles in the Haeckel volume and in consequence was overlooked by many geologists. Reefs
During his early studies (1885) in the Gulf of Naples, Walther became interested in the growth of calcareous algae. He compared his marine findings with what could be found in Tertiary sediments in Sicily and with Triassic alpine rocks, and concluded that lithification takes place concomitantly with deposition, leading to the formation of structureless limestones. A further step towards greater understanding of reefs was achieved as a result of his voyages to the south coast of Sinai (1886) and to the Palk Strait (India) (1888–1889). Walther also emphasized that reefs are traps for considerable amounts of sediments, which form up to 60% of the whole reef complex. Further, he noted the importance of tectonic movements, sea level changes, and topography for the reef growth (elevations are preferred places), laterally or vertically. Deserts
The journey of 1886 along the Sinai coast, with a return along the Egyptian side of the Red Sea through the Galala Desert offered Walther splendid opportunities for the study of desert environments, which he was subsequently able to extend in the USA (1891), central Asia (1897), Egypt (1911), and Australia (1914). He soon recognized the importance of aeolian erosion, which previously had not been much taken
into account. Consequently, his first publication on deserts had the provocative title Die Denudation in der Wu¨ ste und ihre geologische Bedeutung (1892) (Desert Denudation and its Geological Importance) and provoked many controversies. In fact, he tended to exaggerate the notion of desert erosion and neglected the influence of periods with strong fluviatile erosion. Walther was the first to describe ventifacts and Dreikanter. The form of sand grains gave him indications of their aeolian origin, an approach that was systematically utilized by Andre´ Cailleux in the 1930s. Walther dealt with the phenomena of fossil deserts, which he classified as tropical, coastal, plantless volcanic, rain-shadow, and glacial deserts. The fact that the different types could be associated with younger or older sediments from other climatic zones was explained by changes in the Earth’s axis of rotation. With this idea, he was well ahead of his time. In a special publication, Walther described the fauna of a lake in the Buntsandstein desert in Thuringia (1904). From his study of the Nubian Sandstone in Egypt he became convinced of the desert character of the Buntsandstein. His ideas about the formation of (minor) salt deposits are still accepted, along with Ochsenius’s bar theory. Walther was the first to describe laterite profiles in the deserts of Western Australia, interpreting them as weathering products of a former climate (1915). A revised and enlarged version of his first desert book was published in 1900 and went through four editions until 1924. The last edition has recently been translated into English with commentary (1997). This book was the main reason for his being invited to Australia by the British Association for the Advancement of Science in 1914. In Germany, he was called ‘Wu¨ sten (desert) Walther’. National Education
The nineteenth century was a golden age for national science education and Walther’s teacher, Haeckel, was indefatigable (and extremely successful) in this field. Walther himself wrote a flowing, somewhat poetic, prose, which was very apt for such work. His books for a broader public enthused more than one generation and were popular with both school teachers and their students. The Vorschule fu¨ r Geologie (1905) (Elementary Course in Geology), for example, sold 22 000 copies and was translated into Russian and Czech. Its last Russian edition appeared as late as 1940. Another successful book was the Geologie von Deutschland (1910). Walther saw an important task in the training of schoolteachers in geology and held many vacation courses for them.
FAMOUS GEOLOGISTS/Walther 245
However, his extensive work in this field affected his academic reputation in later years.
Impact This may be one of the reasons why Walther’s ideas were not fully appreciated by his contemporary fellow geologists. Another reason could be that he tended to go to extremes when proposing and defending a new idea. Walther was still young when he published his pioneering works and was then of low academic status. Many of his colleagues, especially in Germany, did not recognize or appreciate the views he introduced. They did not fit into the current academic trends. The recognition of his books in Austria and Switzerland was greater than at home. But some of the outstanding German geologists of the time (Wilhelm von Gu¨ mbel, Karl von Zittel, Hermann Credner, Edmund von Mojsisovics) appreciated his revolutionary steps in the direction of biogeology, as opposed to palaeontology or stratigraphy. Walther had particular influence in Russia. His desert book was translated in 1911 and lithology became one of the main fields of Russian geology. In Britain, his reputation was mostly based on the desert book and was soon forgotten after his death. In the USA, only a few geologists, such as William Henry Twenhofel, mentioned his books. Walther’s somewhat ‘baroque’ style of writing caused language difficulties. Also the World War I may have hindered the spread of his works and ideas. Interest in them was revived during the second half of the twentieth century by the facies research in the oil industry. After some delay, Walther eventually became better known in the USA than in Europe.
See Also Fossil Plants: Calcareous Algae. History of Geology From 1900 To 1962. Sedimentary Environments: Depositional Systems and Facies; Carbonate Shorelines and Shelves; Deltas; Deserts; Reefs (‘Build-Ups’). Sedimentary Processes: Aeolian Processes. Stratigraphical Principles.
Further Reading Ginsburg RN, Gischler E, and Schlager W (eds.) (1994) Johannes Walther on Reefs. English translation with commentary. Miami: University of Miami, Rosenstiel School of Marine and Atmospheric Science (Geological Milestones II). Gischler E and Glennie KW (eds.) (1997) The Law of Desert Formation: Present and Past. English translation, with preface and introduction, of Johannes Walther
(1924). Miami: University of Miami, Rosenstiel School of Marine and Atmospheric Science (Geological Milestones IV). Gressly A (1838) Observations Ge´ ologique sur le Jura Solenrois. Nouvelles Me´ moires de la Socie´ te´ Helve´ tiques des Sciences Naturelles. Volume 2. Neuchaˆ tel. Middleton GV (1972) Johannes Walther’s law of the cor relation of facies. Bulletin of the Geological Society of America 84: 979 988. Seibold I (1992) Der Weg zur Biogeologie: Johannes Walther 1860 1937. Berlin, Heidelberg and New York: Springer. Vissotzky WR (1965) Johannes Walther and his Role in the Progress of Geology. Moscow: Nauka (in Russian). Walther J (1885) Die gesteinsbildenden Kalkalgen des Golfes von Neapel und die Entstehung structurloser Kalke. Zeitschrift deutsch Geologische Gesellschaft 37: 329 357. Walther J (1888) Die Korallenriffe der Sinaihalbinsel: Geo logische und biologische Beobachtungen. Abhandlungen der mathematisch physikalischen Classe der ko¨ niglisch Sa¨ chsischen Gesellschaft der Wissenschaften zu Leipzig 14: 435 506. Walther J (1891) Die Adamsbru¨ cke und die Korallenriffe der Palkstrasse: Sedimentstudien im tropischen Litoralge biet. Petermanns Geographische Mitteilungen 22: 40. Walther J (1891) Die Denudation in der Wu¨ ste und ihre Geologische Bedeutung. Abhandlungen der mathema tisch physikalischen Classe der ko¨ niglisch Sa¨ chsischen Gesellschaft der Wissenschaften zu Leipzig 16: 345 570. Walther J (1893) Allgemeine Meereskunde. Leipzig: Weber. Walther J (1893 1894) Einleitung in die Geologie als his torische Wissenschaft. 3 vols. Jena: Fischer. I. Bionomie des Meeres: Beobachtungen u¨ ber die marinen Lebensbe zirke und Existenzbedingungen: 1 196; II. Die Lebens weise der Meeresthiere: Beobachtungen u¨ ber das Leben der geologisch wichtigen Thiere: 200 531; III. Lithogen esis der Gegenwart: Beobachtungen u¨ ber die Bildung der Gesteine an der heutigen Erdoberfla¨ che: 535 1055. Walther J (1900) Das Gesetz der Wu¨ stenbildung in Gegenwart und Vorzeit, 1st edn. Berlin: Reimer. Walther J (1904) Die Fauna der Solnhofener Plattenkalke. Festschrift. 70. Geburtstag von Ernst Haeckel, pp. 133 214. Jena: Fischer. Walther J (1904) U¨ ber die Fauna eines Binnensees in der Buntsandsteinwu¨ ste. Zentralblatt fu¨ r Mineralogie, Geo logie und Pala¨ ontologie un numbered volume (for 1904): 5 12. Walther J (1905) Vorschule der Geologie. Jena: Fischer. Walther J (1910) Die Sedimente der Taubenbank im Golfe von Neapel. Abhhandlungen der ko¨ niglich Preussischen Akademie der Wissenschaften, Physikalisch Mathema tische Classe 3: 1 49. Walther J (1910) Lehrbuch der Geologie Deutschlands. Leipzig: Quelle & Meyer. Walther J (1915) Laterit in Westaustralien. Zeitschrift der deutschen Geologischen Gesellschaft 67: 113 132. Walther J (1927) Allgemeine Pala¨ ontologie. Geologische Fragen in biologischer Betrachtung. Berlin: Borntraeger.
246 FAMOUS GEOLOGISTS/Wegener
Wegener B Fritscher, Munich University, Munich, Germany ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Alfred Wegener (Figure 1) was the ‘father’ of the theory of continental drift, one of the most influential theories in modern earth sciences. From a geophysical point of view, Wegener constructed a new picture of a dynamic Earth, postulating large-scale, ongoing horizontal movements of the continents, contrary to the theory of the permanence of continents and oceans that prevailed in the early twentieth century. First published in 1912, Wegener’s theory had to wait for more than 50 years to become fully acknowledged, due in particular to the lack of a credible mechanism to explain, and direct empirical confirmation of, continental movement. The revival of Wegener’s theory in the late 1950s came from studies of the ocean floors, an approach that Wegener had never considered. Though modern plate tectonics differs significantly from Wegener’s original theory, there was nevertheless hardly any geological idea in the twentieth century that was subjected to greater scientific and public dispute than the idea of ‘drifting’ continents.
(1911), which became a standard textbook throughout Germany. By his studies on the chemical composition of the atmosphere, and its temperature distribution, Wegener pioneered the new science of aerology. Wegener first presented his theory of continental drift to the public at the beginning of 1912. He found little time to reply to his numerous critics, for only a few month later he was in Greenland again. Together with the Dane, Captain Johan Peter Koch (1879– 1928), Wegener became the first to winter on the icecap, and in the following spring, they undertook the longest crossing of the great ice sheet ever made up to that time. After his return from Greenland, Wegener married Else Ko¨ ppen (1892–1992), daughter of Wladimir Ko¨ ppen (1846–1940), a leading European meteorologist who became Wegener’s lifelong mentor and collaborator. In the summer of 1914, Wegener was drafted into the German army but was soon released from combat duty after being twice wounded. The fall of a meteorite in April 1916, near Marburg in Hesse,
Meteorology and Polar Research Born on 1 November 1880, Alfred Lothar Wegener studied astronomy, meteorology, and physics at Heidelberg, Innsbruck, and Berlin, earning a PhD in astronomy from the University of Berlin in 1905. Following his older brother Kurt Wegener (1878– 1964), he was appointed an assistant at the Aeronautical Observatory at Lindenberg, near Berlin. There, he became acquainted with modern methods for the study of the higher atmosphere, including free balloon riding; together with his brother, he broke the world endurance record for balloon riding in 1906 by staying aloft for more than 52 h. That same year, Wegener joined a Danish expedition to Greenland under Ludvig Mylius-Erichsen (1872–1907); the goal was to map Greenland’s north-east coast. Wegener became the first to use kites and tethered balloons to study the atmosphere in an Arctic climate. His Arctic research earned him a position at the University of Marburg, where he lectured on meteorology and practical astronomy from 1909. He published several papers on meteorological subjects, including a monograph on The Thermodynamics of the Atmosphere
Figure 1 Alfred Wegener in 1910. Reproduced from the Deutsches Museum, Munich.
FAMOUS GEOLOGISTS/Wegener 247
turned his attention to these bodies and to the origin of the craters of the moon, which he thought were formed by impacts of bodies belonging to the solar system. In 1919, following his father-in-law, he became head of the department of theoretical meteorology at the German Marine Observatory at Grossborstel near Hamburg, and a ‘professor extraordinary’ (außerordentlicher professor) at the newly founded University of Hamburg. From 1924, Wegener held a professorship in meteorology and geophysics at the University of Graz. Soon he was preparing another expedition to Greenland, for a systematic study of the great icecap and its climate; the expedition departed in 1930. Wegener died at the beginning of November 1930, a day or two after his 50th birthday, while returning from a rescue expedition that brought food to a party of his colleagues camped in the middle of the Greenland icecap. His body was eventually recovered in May 1931.
A New Image of the Earth Wegener’s famous theory of continental drift was actually the work of just a few months. In the autumn
of 1911, he became aware of a paper summarizing the evidence for the close relationship of the older fauna of South America and West Africa. He also remembered an earlier observation of the striking congruence of the coastlines on either side of the Atlantic Ocean; towards the end of the year, in 1911, he gave a preliminary account of his basic ideas in a letter to Wladimir Ko¨ ppen. On 6 January 1912, Wegener presented his new theory at a meeting of the Geologische Vereinigung (Geological Society) in Frankfurt, and promptly published a preliminary paper on his ideas in Petermann’s Geographische Mitteilungen (Figure 2). For most earth scientists of Wegener’s day, the hypotheses of wandering continents sounded rather fantastic, chiefly because of the lack of a satisfactory explanation of the moving forces. However, with the help of the German geologist Hans Cloos (1885–1951), Wegener extended his early paper and published his first book on his theory as Die Entstehung der Kontinente und Ozeane (The Origin of Continents and Oceans) in 1915, now also referring to some earlier (rather speculative) ideas on continental displacements by William Henry Pickering
Figure 2 Map from Wegener’s 1912 paper, Die Entstehung der Kontinente, showing the boundaries of the continental blocks (including the continental shelves).
248 FAMOUS GEOLOGISTS/Wegener
(1858–1938) and Frank Bursley Taylor (1860–1938). After World War I, the second and third editions of his book, both revised and enlarged, were published (in 1920 and 1922, respectively). Palaeoclimatology was an essential empirical background of his theory, and, together with Ko¨ ppen, Wegener published in 1924 a detailed discussion of the climates of the geological past (Die Klimate der geologischen Vorzeit). A fourth edition of The Origin of Continents and Oceans followed in 1929. From 1924 onwards, there were also translations (of the third edition) into English, French, and other languages, by which the theory became known internationally. Wegener’s original intention had been to give a genetic explanation of the large-scale features of Earth’s surface (the continental blocks and the ocean basins) according to a single comprehensive principle. This principle of horizontal mobility, i.e., of the splitting off and drifting apart of continental blocks, did not emerge from new experimental research, but, rather, from a ‘rearrangement’ of known geophysical and geological information. This was set against two widely accepted and closely related assumptions in the earth sciences around 1900, namely, the hypotheses of former (now sunken) land bridges between continents (postulated due to striking geological and palaeontological conformities) and the theory of earth contraction due to its general cooling, advocated by leading geologists such as Eduard Suess (1831–1914) and Albert Heim (1849–1937).
Contraction would not have been sufficient to account for the large folds of Earth’s crust. The great arching forces required to transmit the full shrinkage of a whole great-circle to one point of it have been proved to be physically impossible. The idea of a shrinking Earth, commonly illustrated by the simile of a drying apple, contradicted gravity measurements, i.e., the doctrine of isostasy (the rocks under the oceans are denser than are those under the continents, so altitudinal differences are compensated and equilibrium of pressure or ‘isostasy’ prevails; accordingly, the less dense continental blocks may be thought of as ‘swimming’ on the underlying mass, like an iceberg floating in the sea). Wegener referred to Suess’ distinction of the ‘Sial’ (silica/alumina-rich) and ‘Sima’ (silica/magnesia-rich) layers, emphasizing their different densities (2.5–2.7 for the Sial, or the continental blocks, and about 3.0 for the Sima of the ocean floors) and their different melting points (with the Sial’s melting point being 200–300 C above that of the Sima). Concerning the temperature increase towards Earth’s interior, the difference might not be sufficient to justify the assumption that solid Sialic blocks ‘swim’ in a fluid Sima (Figure 3). Rather, both layers had to be thought of as being viscous, i.e., plastic (with greater plasticity for the Sima). The characteristics of such viscous fluids are paradoxical in that the duration of the action of the forces determines whether the body behaves as a solid or a fluid. Consequently, within geological time, large horizontal displacements of the
Figure 3 Cross section along a great circle through South America and Africa, from Wegener’s 1912 paper, Die Entstehung der Kontinente, showing the ‘swimming’ continents within the Sima (silica/magnesia rich region). Also represented are the Nife (nickel/ iron rich core of the Earth), and the atmospheric layers of nitrogen (Stickstoff), hydrogen (Wasserstoff), and ‘geokoronium’ (a hypothetical gas, which Wegener had introduced to account for the typical green spectrum line of the aurora borealis) (figure in correct scale).
FAMOUS GEOLOGISTS/Wegener 249
continents would be possible, though the motion would be very slow.
Mountain Formation, Volcanism, and Rift Valleys Wegener first discussed the origin of continents and oceans and the formation of folded mountain ranges. Relating to the nappe-fault structures of the Alps, where the original areas that are now thrust up as mountains were apparently up to four or five times wider than at present, Wegener thought that mountain formation was a unilateral, irreversible process: each pressure brings about an increase of the thickness and a shortening of the surface, while, on the other hand, tension leads to splitting of the continental blocks. The individual stages of perceived as mountain formation comprised continual processes of splitting and compression, whereby the original Sialic crust (for which Wegener assumed a thickness of about 30–35 km) gradually decreased in surface area, split into separate pieces, and increased in thickness. Along with the movement of the continental blocks, a hypothesized universal ocean (‘Panthalassa’) began to divide into a shallow sea and a deep sea. Volcanism, for Wegener, was mainly related to the continental ‘fronts’. Areas where tension prevailed, such as the Atlantic Ocean, and also opening faults, seemed to be relatively poor in volcanoes as compared with areas such as the Pacific Ocean, where pressure was increasing. The fronts of moving blocks made conditions more favorable to volcanism than did the ‘backs’. Nevertheless, Wegener wondered whether the mid-Atlantic ridge might be considered as a zone where, with the continuing expansion of the Atlantic, the floor was continuously breaking up, making room for fresh, relatively fluid and high-temperature Sima from below! Moreover, increased volcanic activity in some periods of Earth history might be due to large displacements (as, for instance, during the Tertiary). Trench faults (Grabenbru¨ che), i.e., rift valleys, acquired new meaning as representing the beginnings of new continental separations. Gravity measurements had shown that beneath such lines lay material of greater density, compared to that on either side. Therefore, these lines could be seen as incipient fissures within the continental blocks (into which the denser Sima was rising according to the principle of isostasy). The best examples of such separations were provided by the East African trenches and their continuation through the Red Sea. At the majority of the trenches, the measurable mass deficit was not compensated by greater density of the matter beneath it. Thus, the trenches must be youthful disruptions of a continental block.
Wegener’s theory of mountain formation was further supported by the fact that the folding of the Andes seems to have been essentially simultaneous to the opening of the Atlantic Ocean. The American blocks, during their westward drifting, had encountered resistance at the presumably very old and relatively rigid floor of the Pacific Ocean. Thus, the extended shelf, with its mighty sediments, forming the western border of the continental block, was compressed to a range of fold mountains. For the Tertiary folds of the Himalayas, Wegener assumed that lower India had formed an extended peninsula prior to compression, the southern end of which lay next to that of South Africa. The folds had been produced by ‘impact’ of the Indian subcontinent and the main mass of Asia.
Geological and Palaeontological Evidence The palaeontological evidence indicating a former connection between the organic components of different continents had already given rise to the doctrine of former land bridges. Among the most striking findings were the distributions of the Glossopteris flora on the southern continents and the occurrence of Mesosaurus at the turn of the Permian and the Carboniferous exclusively in south-eastern South America and the western parts of Africa; both of these discoveries suggested a former connection of the two continents. Using these relationships also allowed calculations of when the continents were separated (either by horizontal displacements or by sinking of the land bridges). South America and Africa had been connected during the Mesozoic, but were separated at the end of the Eocene or Early Oligocene. The connection between Europe and North America seemed to have been maintained during the older Tertiary period, but separation occurred in the Miocene, although it might have continued in the far north (over Scandinavia and Greenland) into the Pleistocene. The connection of Lower India with southern Africa, which Wegener had postulated based on his ideas on the formation of the Himalayan range, was also confirmed by palaeontological evidence. Zoogeographers had long assumed a former elongated Indian–Madagascan peninsula (called ‘Lemuria’), separated from the African block by the Mozambique Channel. The zoogeographic concept of Lemuria had given rise to Suess’ notion of a great southern continent, Gondwana, comprising parts of South America, Africa, Lower India, Australia, and Antarctica. Assuming the unchanged positions of its present-day
250 FAMOUS GEOLOGISTS/Wegener
relics, however, required ascribing a huge extent to this continent. Wegener, by contrast, proposed a much reduced primeval continent, Pangaea. In the Permian, i.e., until some 300 Ma ago, all the continents were supposedly joined in one land mass extending from pole to pole. During the Triassic, about 200 Ma ago, Pangaea began to break up and the newly emerging continents started moving into their current positions. In the Jurassic, there were few remaining connections except at the northern and southern ends. Just as northern Europe and North America remained connected until the older Tertiary period, a connection of the southern continents seems to have persisted, running from the southern coast of Australia over Antarctica to South America. Later, the Antarctic block, like the South American block in the Tertiary, moved over from South Africa towards the side of the Pacific Ocean. Only in the Quaternary period, then, did the Australian block become detached (Figure 4).
For geological and tectonic evidence, Wegener referred particularly to Suess’ magnum opus, published in three volumes during 1885–1909, Das Antlitz der Erde (The Face of the Earth). Considering the tectonic relations, Europe/Africa and both Americas seemed to represent the edges of an immense expanded fissure. In the north, for instance, the Greenland massif was matched by Scandinavia, both consisting of gneiss, and the less mountainous North America corresponded to the likewise less mountainous Europe. The most striking example, however, was the Carboniferous mountain range, called the Armorican mountains (Suess’ ‘transatlantic Altaides’), which made the coalfields of North America appear to be the direct continuation of the European ones. Wegener’s theory of mountain formation was also confirmed by remarkable differences between the Atlantic and the Pacific hemispheres, such as the distinction between Pacific and Atlantic types of coasts (marginal chains and ocean trenches in front
Figure 4 Wegener’s reconstruction of the separation of the continents from the primeval Pangaea, from his 1926 paper Pala¨ogeo graphische Darstellung der Theorie der Kontinentalverschiebungen, showing the relative positions of the continents during the Upper Carboniferous (Jung Karbon), Eocene (Eozan), and Lower Quaternary (Alt Quartar) (in two different projections). Cross hatching represents deep seas, dotted regions represent shallow seas; rivers, recent coastlines, and outlines are shown only for orientation.
FAMOUS GEOLOGISTS/Wegener 251
of the Pacific coasts, as contrasted to the wild, irregular ‘ria’ Atlantic coastlines). There were also differences in the volcanic lavas of the two hemispheres, as emphasized by the Vienna petrographer Friedrich Becke (1855–1931) and others. The Atlantic lavas contained a greater proportion of sodium, whereas calcium and magnesium prevailed in the Pacific lavas. Such differences were intelligible according to the assumptions of continental movements. The opening of the Atlantic was matched by the general pressing of the continents against the region of the Pacific Ocean: pressure and compression prevailed at the coasts of the latter whereas tension and splitting occurred at the latter.
Palaeoclimatology Traces of glaciation during the Permian (ground moraines lying on scratched bedrock) were to be found on the southern continents, e.g., in East India and Australia. If the present-day arrangement of the land masses had prevailed at that time, this ‘Permian ice age’ would have required an icecap of seemingly impossible size. And the north pole would have been in Mexico, where no trace of glaciation during that period was recorded. Following the idea of horizontal displacements, however, all regions subjected to glaciation came together concentric to the southern margin of Africa. And one had only to place the south
pole in this much reduced glaciated area to give the Permian ice age a much more plausible form. Wegener had discussed these palaeoclimatological features since 1912. In 1924, he gave a detailed description of the climatological changes from the Carboniferous through to recent times, following the traces of glaciations, swamps, and deserts, i.e., moraines, coal, salt, and gypsum, throughout Earth’s history (Figure 5). In reconstructing the respective polar shifts, Wegener emphasized that they obviously took place along with the great displacements of the continental blocks. In particular, there was temporal coincidence of the best confirmed polar shift, in the Tertiary, and the opening of the Atlantic (Figure 6). Movement of the poles since the Pleistocene might also be related to the final separations of the continents in the north and the south.
Motive Forces Wegener was very cautious about the forces that might have caused continental displacements. First, it was necessary to demonstrate the reality and the manner of the displacements before indulging in the hope of finding their cause. Nevertheless, he tentatively suggested two candidates: centrifugal forces caused by the rotation of Earth and tidal-type waves within Earth, generated by the gravitational pull of the sun and the moon. In the 1929 revision of Wegener’s theory in
Figure 5 Wegener thought continental drift was the key to the climatic changes during Earth’s history. This map, published in the 1924 book by Koppen and Wegener, Die Klimate der geologischen Vorzeit, shows traces of glaciation, swamps, and deserts for the Carboniferous. E, Traces of glaciation; K, coal; S, salt; G, gypsum; W, desert sandstone. Dotted regions indicate arid areas, dashed lines indicate the positions (i.e., the pathways) of the poles, and the bold curved line indicates the respective position of the equator.
252 FAMOUS GEOLOGISTS/Wegener
Figure 6 Map published in the 1924 book by Koppen and Wegener, Die Klimate der geologischen Vorzeit, showing polar shifts (dashed lines) from the Carboniferous to recent, related to the African table (left, south pole; right, north pole). Bold lines outline the continental blocks; hatched lines represent the Carboniferous (Karbon) period. Perm, Permian; Jura, Jurassic; Trias, Triassic; Kreide, Cret aceous; Eozan, Eocene; Miozan, Miocene; Beginn des Quartar, beginning of the Quaternary.
Die Klimate, he also mentioned convection currents within the Sima; these had been first discussed as a cause of mountain formation by the Vienna geologist Otto Ampferer (1875–1947) in 1906. Wegener also endeavoured to calculate the recent velocity of the relative motion of the continents, though he was well aware that these values must be quite uncertain. In his 1912 paper, comparing various longitude determinations for Greenland, he had deduced an increase of the distance to Europe of 11 m year 1. Referring to the lengths of transatlantic cables, he suggested that North America was drifting away from Europe at about 4 m year 1.
From Continental Drift to Plate Tectonics The theory of continental drift was long rejected by the majority of geologists. Among Wegener’s few followers were the South African Alexander Du Toit (1878–1948), for whom continental drift provided the best explanation of the close similarities between the strata and fossils of Africa and South America, and the Swiss geologist E´ mile Argand (1879–1940), who saw continental collisions as the only means of producing the folded and buckled strata he had observed in the Alps (see Famous Geologists: Du Toit). Nevertheless, Wegener’s explanation of the PermoCarboniferous ice age impressed even his critics.
Wegener’s reputation as a meteorologist and a polar explorer contributed to keeping his theory alive. His work was immediately remembered when, around 1960, surprising data were obtained from the ocean floor: palaeomagnetic patterns alongside the mid-ocean ridges clearly suggested the spreading of the seafloor. Within about two decades, Wegener’s principle of horizontal displacements of parts of Earth’s crust became almost universally accepted, although, ironically, the process still lacked a consensus as to its causes, though convection currents in the internal mantle are most commonly advocated. It should be noted that Wegener’s original concept differed from modern plate tectonics in essential points, particularly with regard to the Sial and the Sima. According to modern theory, the (Sialic) continents do not ‘plough’ through the (oceanic) Sima. Instead, both continents and ocean floor are regarded as forming solid plates, ‘floating’ on the asthenosphere, which, due to tremendous heat and pressure, behaves like an extremely viscous liquid (as Wegener had thought the Sima did). Therefore, the older term ‘continental drift’, still often used today, is not quite appropriate for the modern concept. Notwithstanding these differences, Wegener’s basic ideas remain sound, and the lines of evidence that he used to support his theory are still valid. He first envisaged a dynamic Earth, connecting its major features and various geological processes – continental
FLUID INCLUSIONS 253
movements, folded mountain ranges, rift systems, earthquakes, volcanism, ocean transgressions, palaeoclimatological changes, etc. – on a global scale. In this sense, Wegener’s theory was a true forerunner of plate tectonics.
See Also Africa: Rift Valley. Famous Geologists: Du Toit; Suess. Gondwanaland and Gondwana. History of Geology From 1900 To 1962. History of Geology Since 1962. Palaeoclimates. Pangaea. Plate Tectonics. Tectonics: Mid-Ocean Ridges; Mountain Building and Orogeny.
Further Reading Carozzi AV (1985) The reaction of continental Europe to Wegener’s theory of continental drift. Earth Sciences History 4: 122 137. Fritscher B (2002) Alfred Wegener’s ‘The origin of contin ents, 1912’. Episodes 25: 100 106. Jacoby WR (2001) Translation of ‘Die Entstehung der Kon tinente, Dr Alfred Wegener, Petermann’s Geographische Mitteilungen, 58 (1912)’. Journal of Geodynamics 32: 29 63. Ko¨ ppen V and Wegener A (1924) Die Klimate der geolo gischen Vorzeit. Berlin: Borntra¨ ger. Lu¨ decke C (1994) Stratigraphische Methode der Rekon struktion von Expeditionsergebnissen am Beispiel des Todes von Alfred Wegener wa¨ hrend der Gro¨ nlandexpedi tion (1930 31). In: Fritscher B and Brey G (eds.) Cosmo graphica et Geographica: Festschrift fu¨ r Heribert M. Nobis zum 70. Geburtstag, Algorismus, vol. 13,
pp. 347 367. Munich: Institut fu¨ r Geschichte der Naturwissenschaften. Oreskes N (1999) The Rejection of Continental Drift: Theory and Method in American Earth Science. New York and Oxford: Oxford University Press. Runcorn SK (ed.) (1966) Continental Drift. New York and London: Academic Press. Schwarzbach M (1986) Alfred Wegener: The Father of Continental Drift. Madison. WI: Science Tech Publications. Sengo¨ r AMC (1991) Timing of orogenic events: a persistent geological controversy. In: Mu¨ ller DW, McKenzie JA, and Weissert H (eds.) Controversies in Modern Geology: Evolution of Geological Theories in Sedimentology, Earth History and Tectonics, pp. 403 473. London: Academic Press. Wegener A (1912) Die Entstehung der Kontinente. Petermann’s Mitteilungen aus Justus Perthes’ Geogra phischer Anstalt 58: 185 195, 253 256, 305 309. Wegener A (1926) Pala¨ ogeographische Darstellung der Theorie der Kontinentalverschiebungen. In: Dacque´ E (ed.) Pala¨ ogeographie, pp. 171 189. Leipzig and Wien: F Deuticke. Wegener A (1971) The Origin of Continents and Oceans. (Translation from the 4th revised German edition by J Biram, with an introduction by BC King.) London: Methuen. Wegener A (1980) Die Entstehung der Kontinente und Ozeane. (Reprint of the 1st and 4th editions, edited by A Vogel.) Braunschweig: Vieweg. Wegener E (1960) Alfred Wegener: Tagebu¨ cher, Briefe, Erinnerungen. Wiesbaden: Brockhaus. Wutzke U (1998) Kommentiertes Verzeichnis der schriftli chen Dokumente seines Lebens und Wirkens, Berichte zur Polarforschung 288. Bremerhaven: Alfred Wegener Institut fu¨ r Polar und Meeresforschung.
FLUID INCLUSIONS A H Rankin, Kingston University, Kingston-upon-Thames, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Fluid inclusions are small droplets of fluid that have been trapped within crystals either during primary growth from solution or at some later stage, usually as a result of recrystallization along healed microfractures. They are ubiquitous in both naturally occurring minerals and in laboratory-grown crystals. To the chemist or materials scientist, these gross defects cause endless obstacles in their quest to grow near-perfect crystals. However, to the geologist, they
provide a unique fossil record of the various fluids responsible for the formation and evolution of rocks and minerals throughout the history of the Earth. Despite their small size (usually less than 20 mm), their chemical composition and physical properties can be readily determined, and the data may be used to estimate the temperatures, pressures, and physicochemical nature of the fluid at the time of trapping. This information has made an immense contribution to the development of modern theories of ore genesis, petrogenesis, diagenesis, and petroleum migration and accumulation, and to our understanding of the importance of the fluid phase in a wide range of geological processes.
254 FLUID INCLUSIONS
Occurrence and General Characteristics Formation and Genetic Classification of Fluid Inclusions
Small changes in the chemical or physical properties of fluids near to a growing crystal face can lead to perturbations in the stability of crystal growth and the development of gross defects, manifested as embayments, along crystal faces. These embayments will seal over during a period of greater stability, trapping a portion of fluids to form ‘primary’ (P) fluid inclusions. In many instances, the trapped fluid will be ‘homogeneous’ at the time of trapping. In others, where immiscible fluids are present or where mechanical entrapment of other coexisting crystalline phases has occurred, trapping will be ‘heterogeneous’. At some stage after primary growth, ‘secondary’ (S) fluid inclusions can form from later fluids, particularly as a result of recrystallization along microfractures. The chemical and physical properties of these inclusions may be very different from those of the earlier mineral-forming fluids. However, if fracturing and rehealing take place during primary growth, the fluids may be indistinguishable, and the terms ‘pseudosecondary’ or ‘primary–secondary’ (PS) appropriately describe such inclusions. A schematic representation of this genetic classification of inclusions is shown in Figure 1. For most geological applications, it is necessary to establish whether the inclusions are primary, secondary, or pseudosecondary, and also whether heterogeneous trapping has occurred. Heterogeneous trapping may be recognized by the variable proportions of liquids and solids in a single group or generation of inclusions. Various criteria may be used to distinguish between P, PS, and S inclusions, but these may be difficult to apply and it may be difficult to identify primary inclusions in many samples.
Figure 1 Schematic representation of the distribution of pri mary (P), secondary (S), and pseudosecondary (PS) fluid inclu sions in a quartz crystal. Modified from Rankin AH (1989) Fluid inclusions. Geology Today 5: 21 24.
in turbid or translucent minerals, such as feldspar. Quartz is usually the preferred host. Size and Shape of Inclusions
Choice of Material for Study
The successful application of fluid inclusion studies depends partly on serendipity and partly on the type and quality of material available for study. Due to their small size, observations on fluid inclusions are carried out under a microscope using polished wafers around 1–2 mm thick. In most cases, clear, transparent minerals are needed, but it is also possible to study inclusions in some deeply coloured, semi-transparent minerals in very thin ( > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > = > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > ;
9 > > > > > = > > > > > ;
As above
As above, plus cathodoluminescence microscopy, laser Raman spectroscopy, electron probe analysis, ion probe analysis, laser ablation inductively coupled plasma spectrometry, microspectrophotometry, dating by Ar Ar and U Pb series methods As above As above, with exception of dating
As above
Optical microscopy, fluorescence microscopy, microspectrophotometry, scanning electron microscopy, X ray chemical microanalysis, Fourier transform infrared spectroscopy and microscopy, ultraviolet spectroscopy As above As above
to identify the location, red soil from the boots was examined using a combination of techniques including quantitative colour analysis, bulk sample and clayfraction mineralogy by X-ray diffraction, chemical analysis by inductively coupled plasma spectrometry, and pollen analysis. The results clearly indicated a source in a wet tropical country. Comparison was made with control samples taken from adjacent to the airport in the country (Ghana) from which the plane had last departed prior to its arrival at Heathrow, and a very high degree of similarity was obtained in terms of all comparison criteria (Figure 7). The first stage in any forensic comparison of soil or other geological samples is to determine whether or not a possible ‘match’ can be excluded. If it can, then no further attention need be given to that
sample. If it cannot, then further investigation may be warranted. A conclusion of an exact match can sometimes be drawn with virtual certainty when the samples in question make a physical fit and have the same texture and chemical composition. This may occur, for example, with two halves or several broken pieces of rock or ornamental stone. In other circumstances a physical fit may be observed, for example, between a shoe impression in mud and a shoe seized from a suspect that is of the same size and has the same tread pattern as the shoe that made the impression. However, there may be several thousand such shoes in circulation, and a specific ‘match’ with an individual shoe often cannot be made. In this instance, analysis of mud adhering to the shoe, if shown to be indistinguishable from that in which
266 FORENSIC GEOLOGY
Figure 5 House brick with soil staining recovered from a hold all containing the dismembered remains of a prostitute dumped in a canal.
Figure 6 Soil stained boot from a deceased male found in the undercarriage stowage space of a Boeing 747.
the shoe impression was found, may provide strong supportive evidence that the particular shoe under consideration made the mark. However, there are many cases where mud-stained footwear is recovered during an investigation but an exact spot at the crime scene where it may have been acquired has not been identified. In such cases, comparisons of the soil on the shoe with several different reference samples from the crime scene, and usually elsewhere, have to be made on the basis of statistical and graphical comparisons, and the results can be interpreted only in probabilistic terms. The degree of similarity between samples can be expressed in several semi-quantitative and quantitative ways, but meaningful statistics about the likelihood of such a degree of apparent ‘match’ being due to chance are often difficult to provide. This is because the full range of variation that exists in natural soils is impossible to determine
and can be estimated only on the basis of sampling. The availability of database information relating to suitable comparison samples varies greatly from one region to another, and there may be a total absence of pre-existing information in some parts of the world. In such cases, it is necessary to undertake a suitable background investigation, involving collection and analysis of a sufficiently large number of reference samples, in order to provide adequate contextual information for interpretation. Where no physical fit has been identified, the nearest thing to a definitive connection between two questioned samples is usually provided by the identification of one, or more commonly several, highly unusual (or ‘exotic’) particle types in both samples. These may be naturally occurring particles or they may be of human or animal origin. They need not be considered ‘unique’ in themselves, but should be sufficiently rare, either alone or in combination with other unusual particles in the same sample, that the chance of them occurring in any two samples under investigation is extremely low. Examples of two particles that fall into this category are shown in Figures 8 and 9. Waste-dumps, industrial premises, and roadside verges are examples of locations that often contain mixtures of particles that have a more restricted distribution than natural soils. The assemblages of particles present in such locations often show considerable local variation, and it may be possible to limit a potential source area to just a few square metres.
Persistence of Geological Evidence Geological evidence may persist for a considerable period of time after it has been picked up from the source location. For example, gravel, sand, or mud that enters
FORENSIC GEOLOGY 267
Figure 7 Comparison of X ray powder diffractograms for the 300
>1.0 0.3 1.0 0.1 0.3 0.05 0.1 575 10þ 5.00 18 Post mature (onset greenschist facies metamorphism)
%Ro vitrinite reflectance; Tmax Rock Eval Tmax; SCI spore colour index; TAI thermal alteration index; LOM Shell’s level of organic maturation. a To determine your own equivalences, produce your own cross plots.
%Ro vitrinite reflectance; Tmax Rock Eval Tmax; SCI spore colour index; TAI thermal alteration index; LOM Shell’s level of organic maturation. a To determine your own equivalencies, produce your own cross plots.
1.30
465
>9
3.50
12
PETROLEUM GEOLOGY/The Petroleum System 281
Figure 20 The measurement and interpretation of vitrinite reflectance for determining the maturity of sedimentary rocks in general and oil and gas source rocks in particular.
by the petrographer to be vitrinite are measured (Figure 20B, lower histogram): typically between 20 and 50 particles of vitrinite are measured, and an arithmetic mean value calculated. If multiple populations of vitrinite are recognized, then individual mean values are calculated for each population (Figure 20B, inset table).
If linearly polarized light is used (Figure 20A, see ‘polarizer’), rotating the microscope stage and hence the sample with produce a maximum (%Ro, max) and a minimum (%Ro,min) if the material is optically anisotropic. As a result of maturing under overburden pressure, vitrinite develops bedding-parallel anisotropy at the late oil and early gas
282 PETROLEUM GEOLOGY/The Petroleum System
generation levels ( 1.0%Ro). If the vitrinite is anisotropic and maximum and minimum reflectances are determined, then the mean of the maximum values may be reported (%Ro, mean-max). Theoretically the maximum %Ro, max is the most significant indicator of maturity. The interpretation of the mean vitrinite reflectance values is normally undertaken by plotting %Ro on a log-scale against sample depth (Figure 20C), where anomalies may be recognized as plotting away from the linear trend. The interpretation then uses the industry-standard relationship with both oil and gas generation windows and coal rank (Figure 20D). Discontinuities in the maturity trend may be seen at unconformities (where the amount of uplift may be estimated from the offset), and excursions may relate to intrusions or hydrothermal flow. As well as defining the maturity of the sampled section, the reflectance trend can be projected to predict—in the absence of major unconformities—maturity levels ahead of the drilling bit. In addition to vitrinite reflectance two other maturity parameters are commonly reported, Rock-Eval Tmax (Figure 15), and kerogen or spore colour estimated on TAI (1–5) and SCI (1–10) scales, respectively. Acronyms are defined and equivalences detailed
for oil and gas generation in Table 1 and Table 2, respectively. Although the Tmax values are often available in large numbers, in a single well they often show substantial scatter with respect to depth. In contrast, fewer visual estimates of kerogen colour are determined, but they often produce a better maturity trend with depth. Spore colour estimates are particularly reliable since reworked or caved spores and spores introduced with drilling mud or other contamination can be readily identified as being out of stratigraphic sequence. Finally the maturity involved in a petroleum system will normally be interpreted in the context of basin modelling and a burial history plot (Figure 21). The burial history plot (time and hence stratigraphy versus depth) traces the burial depth (and temperature) for each modelled strata (black lines) from deposition to the present day, including periods of uplift reflecting erosion at the palaeosurface. Compaction is modelled by various methods as the sediments are progressively buried, and the temperature grid (red lines) calculated from geothermal gradients or mantle heat flow. The heat flow calculation requires the knowledge of the thermal conductivity of each sediment unit—a function of mineralogy and porosity (compaction):
Figure 21 Burial history model showing present day and palaeotemperature and early , mid , and late mature oil windows defined by vitrinite reflectance.
PETROLEUM GEOLOGY/The Petroleum System 283
Heat flowðmW m 2 Þ ¼ Geothermal gradient
ð C km 1 Þ Thermal conductivity ðW m
1
C 1Þ
The calibration step then minimizes the difference between measured and modelled temperature (by adjusting heat flow) and maturity parameters as given in Table 1 (by adjusting uplift, etc.). From knowledge of the temperature history for each stratum, the kinetics of decomposition of the kerogens (i.e., the generation of oil and gas) are modelled using the Arrhenius equation. Once calibrated, the model is used to predict the amount, composition, and timing of hydrocarbon generated as discussed below.
Generation and Expulsion The process of generation describes the conversion of kerogen to petroleum (oil and gas): TR Kerogen ! oil þ gas þ residue Generation, as opposed to maturation, is measured in terms of the extent of kerogen conversion (transformation ratio, TR) and is used to define the oil and gas windows (shown as yellow, green, and red areas in Figure 21).
The generation process can be monitored by laboratory measurement of the volumes (or masses) of generated oil or gas extracted from source rocks (Figure 22), or liberated during pyrolysis (Figure 23). Although the former is constructed from solvent extraction of the materials within source rocks as a fraction of the TOC (units of milligrams of extract per gram of TOC), the latter is reported as the Rock-Eval pyrolysis production index [PI ¼ S1/ (S1 þ S2)]. The PI is the ratio of the free hydrocarbon (S1, kg t 1) as a function of the remaining pyrolysate (S2, kg t 1). In both cases, the onset of generation (early generation) is indicated by an increase in yield. While the early generation phase starts the build-up of petroleum in the source rock, the continuation of the process saturates the source rock, leading to expulsion and eventually the cracking of oil to gas. Assuming a typical geological heating rate (e.g., 1 C My 1), the following maturity zones can be defined with respect to oil generation, expulsion, and cracking from Type II oil-prone kerogens: Immature, 330 F): No remaining oil generative capacity exists in the source rock kerogen, and unexpelled or reservoired oil is cracking down to condensate and wet gas.
The temperatures given above relate to corrected wire-line log temperatures and Tertiary effective burial, as in the North Sea. Studies related to DST temperatures quote higher temperatures (up to 12 C higher) for these boundaries, while absence of temperature correction can produce substantially lower temperature boundaries. Lower temperature (e.g., 10 C lower overall) may be expected where a Mesozoic burial event is the controlling factor, while higher temperatures (up to 25 C higher) are reported where rapid Neogene– Quaternary burial controls generation. Both extract yields and production index discussed above are measured parameters. The theoretical and hence modellable measure of kerogen degradation is the transformation ratio on a scale of 0 ! 1: TR ¼ Generated petroleum= Original petroleum potential where: Generated petroleum ¼ Original petroleum potential Residual petroleum potential
Figure 23 Oil generation as indicated by the Rock Eval production index (PI) where the maturity trend depends on both generation and expulsion (retention).
The original petroleum potential is determined from immature source rocks of the same organofacies, whereas the residual petroleum potential is determined from the source rocks as it is buried through the oil and gas window. The equivalence of the transformation ratio as a measure of generation to the maturity parameters discussed in the previous section (Tables 1 and 2) is given in Table 3 for oil generation from Type II kerogen and Table 4 for gas generation from Type III kerogen. Generation is a kinetically controlled process, being a function of the effects of both temperature and time on the breakage of chemical bonds in the kerogen present in the source rock. It can be modelled using simple reactions or networks of competing or sequential reactions, expressed in terms of the Arrhenius equation. The Arrhenius equation effectively expresses the ease of degradation of the kerogen to petroleum as a distribution of chemical bond strengths reflecting the C–C, C–H, C–O–C, C–S–C, and C–N–S bonds holding the kerogen network
Table 3 The generation defined oil window approximately related to some common maturity parameters Parameters (oil & gas)
Temperature ( C)
Transformation ratioa (0 1)
Vitrinite reflectance (%)
Spore colour index (1 10)
Rock Eval Tmax( C)
Early mature Mid mature Late mature Post mature
80 115 115 145 145 165 >165
0.05 0.15 0.15 0.65 0.65 0.95 >0.95
0.5 0.7 0.7 1.0 1.0 1.3 >1.3
4.0 5.0 5.0 7.0 7.0 9.0 >9.0
432 442 442 455 455 465 >465
a
Transformation ratio of Type II kerogen (LLNL).
PETROLEUM GEOLOGY/The Petroleum System 285
together. Each of these chemical bonds has a unique strength (activation energy), depending on the adjacent atoms and functional groups, together with stereochemistry. Using the information shown in a burial history plot (Figure 21) to solve the Arrhenius equation, the extent of kerogen degradation can be modelled. The understanding of generation in both space and time is at the centre of the use of the petroleum system concept to explore for oil and gas, and this requires
the use of computer calculation and modelling. The calculation of a theoretical transformation ratio using the Arrhenius equation can be undertaken using industry standard kinetics and basin modelling software, and can be used to calculate generation as a function of depth (Figure 24) and geological time (Figure 25). The generation versus depth plot (Figure 24) is a simplification in that it assumes that all units within the modelled section contain the same kerogen type
Table 4 The generation defined gas window approximately related to some common maturity parameters
Parameters (gas)
Temperature ( C)
Transformation ratioa (0 1)
Vitrinite reflectance (%)
Spore colour index (1 10)
Rock Eval Tmax( C)
Early mature Mid mature Late mature Post mature
115 145 145 220 >220 ??
0.05 0.45 0.45 0.85 0.85 0.95 >0.95
0.7 1.0 1.0 2.2 2.2 >3.0 >3.0
5.0 7.0 7.0 9.0 10 10
455 465 465 525 525 575 >575
a
Transformation ratio of Type III kerogen (LLNL).
Figure 24 Computer modelled depth trend for the potential generation of oil and gas, assuming all rock units contain a common kerogen type (compare with Figure 22 based on measured data).
286 PETROLEUM GEOLOGY/The Petroleum System
Figure 25 Computer modelled generation and expulsion of oil and gas through time from a single source rock unit containing kerogen.
(unlikely). In this case the kerogen is modelled to be broken down to oil and then oil potentially cracked to gas in a two-step procedure: Kerogen ! Oil þ Primary gas Oil ! Secondary gas Considering generation in terms of geological time, a single source rock unit will progressively generate a range of products, grouped somewhat arbitrarily as methane plus wet gas, light oil, and main oil ranges of molecules in Figure 25. The subdivision of oil into these molecular ranges, though subjective, is useful for estimating compositional risk in oil exploration. In the source rock, these molecules form a single petroleum phase (monophasic mix), which can be modelled as being expelled from the source rock once the pore space in the source rock is saturated. Expulsion is poorly understood in terms of process, but well understood in terms of efficiency. Possible mechanisms for petroleum expulsion include movement: 1. From high to low pressure potential—with pressure also causing rock fracturing; 2. From high to low concentrations (diffusion); 3. Under capillary forces induced at mineral pore throats;
4. Under capillary forces through a continuous kerogen ‘wick’; and 5. In aqueous solution during decompaction and clay dehydration. A combination of these mechanisms possibly contributes to expulsion from the source rock, the relative contributions being the basis for the current debate. The favoured view envisages the dominant process as being based on pressure, where compaction coupled with the generation of hydrocarbon fluids from solid kerogen and temperature-induced expansion of pore fluids produces overpressure in the source rock. The overpressure eventually fractures the source rock, sporadically releasing the fluids into the migration pathway. The expelled hydrocarbon yields shown in Figure 25 are calculated from the total generated less that required to saturate (a fraction of) the porosity of the source rock. The modelled porosity can be calculated using one of a number of possible compaction equations. A quantitative understanding of expulsion efficiency is essential for successful prospect evaluation. Expulsion efficiency (EE) is defined in mass units as: EE ¼ Petroleum expelled=Petroleum generated
PETROLEUM GEOLOGY/The Petroleum System 287
where Petroleum expelled ¼ Petroleum generated Petroleum retained and Petroleum generated ¼ Initial potential Transformation ratio Within a basin such as the North Sea, which contains a uniform oil-prone source rock organofacies in the Upper Jurassic Kimmeridge Clay Formation, the initial potential is determined by analysis of immature source rock samples from the rim of the basin. The retained (i.e., generated but not expelled)
petroleum can be determined by solvent extraction or pyrolysis (S1 peak—see Figure 15). The transformation ratio can be determined from modelling, equated with vitrinite reflectance or pyrolysis Tmax (e.g., Table 3) or from ratios of the initial to measured hydrogen index (e.g. initial HI ¼ 600, present HI ¼ 300, therefore TR ¼ 0.5). Examples of the range of expulsion efficiencies calculated from North Sea source rock data are shown in Figure 26, where values rise from about 20% at 3.0–3.5 km to about 50% at 4.0 km with the deepest samples exhibiting values in the 75–95% range. Considering dry gas (methane) generation in, and expulsion from seam coals requires an understanding of additional factors, in particular the very high
Figure 26 Expulsion efficiency as a function of depth in the North Sea based on solvent extraction (green trend line, from Larter (1988)) and Rock Eval pyrolysis data (histograms).
288 PETROLEUM GEOLOGY/The Petroleum System
surface area (macro-, meso-, and microporosity) of coal in general and the vitrinite maceral in particular. The developing understanding seems to suggest a three-step process: 1. Generation of the methane from humic kerogen; 2. Adsorption of the methane on the abundant porosity of vitrinite; and 3. Expulsion of the methane at higher maturity as the porosity collapses. In all cases the expelled monophasic petroleum leaves the source rock, but will only form commercial accumulations if migration provides a focus towards a trap. The processes controlling migration are discussed in the next section.
Migration How petroleum travels long distances in the subsurface is a matter for speculation: facts are hard to come by. Certainly oil being more buoyant than water, and gas being more buoyant than both, there is a natural tendency for the upward movement of separate gas and oil phases in a water-saturated permeable sediment. The fluids are essentially separating out within the porous media of sedimentary rocks. Attempting to understand the processes contributing to migration requires some knowledge over a large range of scales of the properties of the solid rock and the fluid phases flowing therein (Figure 27). The scales are not only
Figure 27 Our understanding of how oil and gas move through water saturated rock requires a study of flux over a large range of scales. Modified after Mann et al. (1991).
PETROLEUM GEOLOGY/The Petroleum System 289
those from sedimentary grains to basins, but also from oil molecules to accumulations containing billions of barrels of oil. At deposition, sedimentary rock is initially porous, but compaction and diagenesis rapidly reduces the porosity. Once sediments are indurated (compacted and cemented), fracture porosity may take over as the dominant conduit allowing fluid flow. Thus fluids can move through both intergranular or fracture porosity. Once expelled into a more porous conduit such as a sandstone bed, a fault plane, or diagenetically altered carbonate, petroleum will start to move upwards under the force of buoyancy (Figure 28, inset). The buoyancy derives from the lower density of oil (typically 0.86 g cc 1 or 36 API) relative to formation water (typically 1.03 g cc 1), as in the kitchen where fat rises to the surface of milk. Once within the migration conduit the petroleum will rise to the roof of the rock layer and if the layers are tilted, the ‘river of oil’ will move up-dip and along zones of highest permeability from high point to high point. When seen at outcrop, this produces a braided network of oil-saturated rock, running under the top-seal of the migration conduit (Figure 28). Migration can also be influenced by any water flux sharing the same migration route (hydrodynamic control) or by pore pressure gradients associated with the progressive increase in pore throat diameters (permeability control). The pore throat diameters together with wettability essentially control the permeability of the rock with respect to oil or gas. Migration distances can vary widely. In situations where mature source rock is interbedded with reservoir sands the migration path may only be a number of metres. Large accumulations, however, demand the
gathering of oil and gas over a wide area, and hence long-distance migration pathways. In the North Sea migration distances are mainly in the range 10–20 km with 65 km as a maximum. In large basins such as the West Canada Basin and West Siberian Basin migration pathways of many hundreds of kilometres have been proposed. As the expelled monophasic petroleum moves upwards under buoyancy, the pressure and temperature regime will change and the bubble-point or dew-point may be reached (Figure 29). Separation into a liquid (oil) phase and a gas phase will result, and these may migrate separately due to differential buoyancy and relative permeabilities. Studying a petroleum system demands an understanding of the efficiency of the migration process. This is probably the least well-quantified element of a petroleum system, and is normally determined by difference: Accumulation ¼ Generated Expulsion efficiency Migration efficiency Therefore Migration efficiency ¼ Accumulated= ðGenerated Expulsion efficiencyÞ The above equation can be used to determine the migration efficiency if the size of an accumulation, the amount generated, and expulsion efficiency are all known. Calibrated against known accumulations, the migration efficiency can then be used to predict the petroleum charge to undrilled prospects. Migration efficiencies are not widely discussed or reported, but values seem to fall in the range of 50% (maximum) to 0 (i.e., no oil or gas reached the drilled prospect). It may be helpful to consider the inefficiency of migration as the residual oil left behind along the migration path. The recognition that oil migration is a focussed process affecting only a small percentage of the rock between kitchen and accumulation fits well with the calculated migration efficiencies.
Accumulation and Survival
Figure 28 Migration occurs as ‘rivers of oil’ running up dip mainly under the influence of density driven buoyancy, though hydrodynamic flux and capillary pressure may exert ancillary controls. Modified after IES (1985).
Once charged to the trap, the survival of the petroleum depends on a good seal and avoiding the effects of various oil alteration processes. An approach to use bulk, molecular, and isotopic characterization of oil and gas to identify such reservoir processes is shown in Figure 30. As emphasized in the figure, the analytical approach to differentiating between oil alteration processes requires knowledge of the geometry and history of the trap.
290 PETROLEUM GEOLOGY/The Petroleum System
Figure 29 Phase separation of upward migrating petroleum depending on the initial gas/oil ratio (GOR) and reductions in temperature and pressure.
In terms of a seal, efficiencies range from near perfect in the case of intact salt (Halite) to ‘highly leaky’ in the case of thin and poorly compacted mudstone containing a significant portion silt and fine sand. Diagenetic cement can greatly enhance the sealing efficiency of more porous strata, even when relatively shallow. In essence the efficiency of a seal must be measured against the charge rate if generation continues, or the time since charge ceased if in the past. A more specific and dramatic way of breaching the seal of the trap is faulting. For structural traps, the prediction of whether a fault is a seal or a migration conduit is complex, the controlling factors being only poorly understood at present. Some unusual seals comprise a hydrodynamic trap, where a counterflow of formation water opposes the buoyancy of the oil, and shallow seals in the Arctic involving the permafrost layer or gas hydrate (see Petroleum Geology: Gas Hydrates).
The concept of the ‘half-life’ of an oil or gas field has been proposed, this being defined as the time (millions of years) required for half of the charge volume to escape through the seal. Gas sealed by mudstone has a half-life in the range 30–60 My, and oil about an order of magnitude greater. A continuous and stable halite seal can be considered near perfect with an infinite half-life. A number of intrareservoir processes are listed towards the base of Figure 30, and of these the most common are bacterial degradation and thermal cracking. These occur when the reservoir is unusually cool (165 C), respectively. Of the listed intrareservoir processes, bacterial degradation is arguably the most destructive. In essence it occurs when petroleum accumulates in a reservoir that is at a temperature allowing bacterial degradation of the oil or gas. As reported, the maximum
PETROLEUM GEOLOGY/The Petroleum System 291
Figure 30 Steps required to identify and interpret vertical and lateral compositional gradients resulting from intrareservoir processes.
temperature boundary is in the range 65–75 C, the range reflecting different oil chemistries, formation waters, and hence active bacterial communities, together with different methods of measuring reservoir temperatures. In nature, viable bacteria will be preserved if the reservoir has never been buried to temperatures hotter than about 90 C. If hotter and then uplifted to a cooler regime, meteoric water influx into the reservoir is required to reintroduce bacteria. The bacterial alteration of the oil progresses with the selective removal of various families of molecules (Table 5). The sequence of removal is broadly n-alkanes > branches alkanes > cyclic alkanes > aromatics, this reflecting the ‘digestibility’ of hydrocarbon molecules by the bacteria. Bacterial degradation of a gas accumulation removes the wet gas components (C2–C5) and in particular propane (C3) with preferential removal of
molecules with the 12C isotope. Thus an initial wet gas is reduced to a dry gas accumulation of >95% methane with isotopically heavy residual propane. Bacterial activity is often confirmed by the thin rim of biodegraded heavy oil below the dry gas. Most reservoirs are cooler than the source rocks from which they were filled. If the petroleum-filled reservoir is buried to higher temperatures, the oil will crack to condensate and gas (Table 6). This process of intrareservoir cracking starts at about 160 C and black oil is normally destroyed by 175 C. This has been termed the base of the oil preservation (as opposed to oil generation) window. An oil trap that has suffered intrareservoir cracking will be recognized as containing gas, with black flecks of pyrobitumen seen occupying some of the remaining porosity: 2 oilðCH2 Þ ! pyrobitumenðCÞ þ methaneðCH4 Þ
292 PETROLEUM GEOLOGY/The Petroleum System
Table 5 The degree of bacterial degradation of oil on a 1 10 scale based on progressive loss of specific chemical groups Degree of bacterial degradation
Loss of named molecules indicates stage of biodegradation
C15þ fraction
1
C15 C7
Gasoline (C4 C7) n Alkanes
AEC ð1 POSÞ where NPV is the net present value (monetary return in the case of a successful project), POS is the probability of success (chance that the project will be successful), AEC is the abortive exploration cost (the lost investment in the case of failure), and (1 POS) is the probability of failure (complement of the probability of success). An exploration programme is, therefore, a carefully planned series of activities aimed at discovering whether oil and gas fields are present in a particular area and, if present, whether they are sufficiently large and productive to be economical. Based on an analysis of the geological information, often involving sophisticated computer modelling of geological processes and an analysis of analogous information from petroleum provinces around the world, the probabilities of finding reserves of certain magnitudes can be assessed. In a well-defined area with fully appraised discoveries, for instance, there may be a high degree of certainty about the current reserves but little chance of finding major additions. In a speculative venture in a littleknown area, on the other hand, the chances of finding any hydrocarbons may be low but there is an outside chance of making a very large discovery. However, even with the most sophisticated analysis, the discovery of any quantity of oil and gas cannot be anticipated with certainty, and petroleum exploration remains a classic example of ‘decision-making under uncertainty’.
Exploration Methods Exploration methods are the techniques employed in the search for oil and gas. Their primary purpose is not directly to find oil and gas but to provide evidence about the geology of the subsurface, the interpretation of which may eventually lead to drilling and the discovery of hydrocarbons. Geological Analysis
The first stage in any exploration programme is to define a ‘play’ concept to explain how oil and gas may have accumulated in the basin. A play is defined
296 PETROLEUM GEOLOGY/Exploration
Figure 1 Geological conditions for hydrocarbon accumulation, showing the migration of oil and gas from mature source rock into an initially water filled reservoir. GOC, gas oil contact; OWC, oil water contact. Reproduced with permission of Nautilus Limited.
as a group of prospects (sites of potential hydrocarbon accumulation) and any related fields having common oil or gas sources, migration relationships, reservoir formations, seals, and trap types, thus sharing common elements of geological risk (Figure 2). The analysis of hydrocarbon plays involves the synthesis and mapping of all the key geological parameters controlling the occurrence of oil and gas, which are principally source, maturation and migration, reservoir, trap, timing, seal, preservation, and recovery (Figure 3). Geochemical techniques can help to establish the presence of a suitable source rock, while sedimentary geological models can predict the distribution of reservoir and seal rocks. Data are derived from previously drilled wells in the basin (although not necessarily testing an identical play), from traditional field geology (examining surface outcrops within or on the margins of the basin), and from remote sensing (Figure 4) (which may elucidate the geological structure of the area). For previously unexplored areas lacking in ‘hard’ data, concepts can be tested against analogous basins elsewhere in the world. Play analysis helps to estimate the risks and potential volumes of prospects, as the absence of any one of the factors listed above invariably means that there will be no hydrocarbons in the prospect; a poor development of source rock will mean a limited generation of hydrocarbons, and a poor development of reservoir rock will mean only limited opportunities for entrapment. Geophysical Analysis
Several geophysical techniques are used to enhance the geological understanding of a basin. Gravimetric (see Analytical Methods: Gravity) and magnetic surveys (both surface and airborne) (Figure 5) can be used in a reconnaissance mode to delineate the
deeper parts of the basin and the major highs before undertaking much more expensive seismic surveys (see Seismic Surveys). However, seismic surveys are the most important geophysical tool for obtaining a detailed understanding of the subsurface structure, including identifying likely migration paths, inferring the relative timing of trap formation and charge, delineating the geometry and size of potential traps, and even establishing the presence of hydrocarbons themselves. Seismic surveys involve recording artificially generated shock waves that are reflected or refracted from the different rock strata, and they require a suitable sound source and an appropriate set of detectors. Dynamite has largely been replaced on land by special trucks called vibrators, which have metal plates on their undersides to shake the ground in a controlled manner (Figure 6). At sea, powerful air or water guns are the main energy sources. The reflections are picked up by receivers called geophones on land and hydrophones offshore (Figure 7), which are laid out in such a way as to form directional antennae, termed arrays. Geophones and hydrophones work in a similar way to seismographs, which register earthquakes, but all data are recorded in digital rather than analogue form. When gathering data in new or little-explored areas, two-dimensional seismic surveys are generally acquired first; with lines spaced as much as 20 km apart, these can cover a great deal of ground at comparatively low cost. If these surveys show interesting structures, then more closely spaced two-dimensional lines can be shot to provide a more detailed picture. However, for an enhanced understanding of the structural and reservoir detail of a prospect, three-dimensional seismic surveys employing a dense grid of lines are commonly used, even in the exploration phase before any wells have been drilled or hydrocarbons discovered. This is
PETROLEUM GEOLOGY/Exploration 297
Figure 2 North Sea play types: (A) a truncated tilted fault block, typical of the northern North Sea; (B) a drape or pinchout trap, typical of the central graben; (C) a salt induced structure typical of the central graben; and (D) a horst block capped by salt, typical of symbol, source rock; pink, salt; blue, chalk reservoir; red, gas; the southern North Sea. Yellow, sandstone reservoirs; brown, with green, oil. Reproduced with permission of Nautilus Limited.
particularly true in geologically complex areas, where the cost of a three-dimensional survey can be considerably less than the cost of a mislocated well. A traditional two-dimensional seismic survey is shot along single lines to provide vertical slices through the Earth’s crust; from a series of parallel and intersecting lines, the three-dimensional structure of the subsurface can be interpreted (Figure 8). However, the assumption is that the reflections seen on each line
originate within the plane of the vertical slice, whereas in reality many of the reflections originate from inclined rock layers that may be hundreds of metres outside the plane of the seismic line; also, the gaps between the lines may hide faults and other geological complexities. These limitations have been overcome by the use of three-dimensional seismic surveys, which employ multiple sound sources and simultaneous recording by lines of detectors laid out in a dense grid to
298 PETROLEUM GEOLOGY/Exploration
Figure 3 Elements of hydrocarbon play analysis, combining topographic/bathymetric data, source rock maturity, reservoir distri bution and structure, and seal. Reproduced with permission of Nautilus Limited.
Figure 4 Landsat image of south eastern Iran, showing large, eroded anticlinal structures. The area shown is 45 km 30 km. Image from Landsat 7, data available from US Geological Survey, EROS Data Center, Sioux Falls, SD, USA.
PETROLEUM GEOLOGY/Exploration 299
Figure 5 Gravimetric and magnetic profiles (above) which mirror the deeper parts of the basin and the major highs shown on the geological section (below). Note the effect of iron rich rocks, such as lava flows and igneous dykes on the magnetic profile. Yellow, (sandstone) reservoir. Reproduced with permission of Nautilus Limited.
Figure 6 Principles of seismic surveying: (A) onshore, with diagrams showing the various methods of producing the signal, and (B) offshore. Reproduced with permission of Nautilus Limited.
300 PETROLEUM GEOLOGY/Exploration
Figure 7 Aerial view of an offshore seismic survey, showing the array of hydrophones being towed behind the ship. Repro duced with permission of Millenium Atlas Company Limited.
acquire a true three-dimensional picture (Figure 9). This development has been made possible by an almost exponential increase in computer power over the years coupled with a concomitant decrease in cost, which has enabled the vast amount of data acquired by three-dimensional surveys to be quickly processed, separating true signals from interference and providing a clearer picture by accurately locating all reflections in their true subsurface positions. Increasing sophistication in both seismic acquisition and seismic processing techniques allows the direct identification of hydrocarbons in the subsurface. Each interface between separate rock layers will reflect some of the sound waves back towards the surface, and the amplitude of this reflection depends on the nature of that interface. For a porous rock, such as a sandstone, the reflectivity will depend in part on the fluid present in the rock pores, and from this it may be possible to identify oil, gas, and water in the subsurface (Figure 10). An extension of this concept that is being increasingly applied is the so-called four-dimensional or timelapse seismic survey. A three-dimensional seismic survey is re-acquired at regular intervals (say, every 3–4 years) over a producing field. The changes in the positions of the reflections indicating the positions of the oil–water contact and the gas–oil contact are used
Figure 8 Interpretation of a two dimensional seismic survey. A geological horizon is traced around all the lines of the grid and then plotted in map view. Reproduced with permission of Nautilus Limited.
PETROLEUM GEOLOGY/Exploration 301
to monitor the efficiency with which oil and gas are being produced from the field and can identify areas that are not being drained for future infill drilling (Figure 11). The increase in computer power that has allowed the acquisition and processing of the vast amounts of data resulting from three-dimensional seismic surveys
has also revolutionized the interpretation of these data. Computer-based interpretation systems, termed workstations, allow the profiles to be viewed on colour screens and key geological horizons to be identified (Figure 12). Once they have been identified through the use of appropriate algorithms, these horizons can be automatically ‘picked’ by the computer by recognizing their ‘character’ throughout the rest of the dataset. Information from wells and on the regional geological setting can also be integrated. The resulting interpretation can then be viewed and manipulated in a variety of ways: vertical sections can be viewed in any direction; time and horizon slices can be generated that can be horizontal, dipping, or undulating; maps of horizons can be produced; and the frequency and amplitude of the seismic reflectors can be analysed to give clues as to the nature of the lithology and the fluid content of the subsurface layers. The threedimensional survey can be considered as a cube of data, and computer visualization techniques allow the data to be viewed from any direction and, indeed, to be rotated on screen (Figure 13). The latest development is the use of a so-called immersive ‘visionarium’ environment, in which observers are surrounded by the data, viewing it in actual three dimensions using special spectacles (Figure 14).
Prospect Appraisal Systems
Figure 9 Correct positioning of reflections by the use of three dimensional seismic surveys. (A) A two dimensional seismic survey may contain spurious information generated from out of the plane of the profile being acquired. (B) A three dimensional seismic survey and its subsequent processing, enables all in formation recorded to be correctly located in the subsurface. Reproduced with permission of Nautilus Limited.
A number of methods have been developed to attempt to translate complex geological data, derived from the geological and geophysical studies described above, into the set of numbers needed for decisionmaking. The particular appraisal method used depends on the size of the unit to be assessed (from global to part of a structure) and the amount and type of information available (from sparse regional data to detailed drilling results, or data of a more statistical nature such as the number of wells drilled and the amount of reserves found). The methods can be broadly grouped under three main headings: subjective methods, statistical methods, and deterministic models.
Figure 10 Gas effect seen on seismic showing the change in amplitude of the signal (so called ‘bright spot’). There is a phase reversal at the gas water contact (GWC) and a time sag. Reproduced with permission of BG Group PLC.
302 PETROLEUM GEOLOGY/Exploration
Figure 11 Time lapse (four dimensional) seismic survey over a North Sea field; the changes in amplitude result from a significant depletion of the oil due to extraction. OWC, oil water contact. Reproduced with permission of Millenium Atlas Company Limited.
Subjective Methods
Subjective methods result from the implicit intuitive thought processes of a single individual or group of individuals and can be formidable tools when a substantial body of expertise is available. However, there is the problem of bias: usually people are more certain about the unknown than they can afford to be on the basis of the information available, and some people are consistently pessimistic, while others are optimistic. There is also the problem of consistency of knowledge: local knowledge can override possibly more relevant but less well-known worldwide information. Statistical Methods
Statistical methods attempt to extrapolate past experience using a variety of statistical techniques. Some depend on the validity of an essentially linear extrapolation, for example a hydrocarbon richness per unit volume of sediments in explored basins, which can then be applied to unexplored basins. Other methods attempt to improve this extrapolation by considering past exploration results, such as the decline of success ratio through time and the decline of field size as a function of exploration effort. In the example shown in Figure 15, the larger fields are discovered first (the so-called ‘creaming’ effect), but, although the average
size of the fields found is decreasing, the number found is still increasing. One of the best known of such studies is that of M King Hubbert, who in 1956 predicted that US oil production would peak in the early 1970s; this proved to be correct (the so-called ‘Hubbert’s Peak’; Figure 16). Disadvantages of these methods are that a certain degree of exploration maturity is required for the necessary historical data to be available, the results may be strictly applicable only to the particular basin or province analysed, and the interrelationships of the variables chosen are not always known and correlations may be fortuitous. Deterministic Models
Deterministic models attempt to integrate all the relevant geological concepts and research findings on the processes of hydrocarbon generation, migration, accumulation, and retention into a coherent model that will provide quantitative estimates of the hydrocarbon volumes likely to be present and the risks involved. These methods are now widely employed in the industry. The problem is that the information available is usually insufficient for precise estimates of the required variables to be made, and, in such
PETROLEUM GEOLOGY/Exploration 303
Figure 12 (A) Geological interpretation of a seismic line, and (B) depth converted to show the true fault geometry. Reproduced with permission of Millenium Atlas Company Limited.
cases, a single numerical estimate is often more misleading than helpful. Hence, the inputs will almost always be in the form of probability distributions, and the use of repetitive calculations, each time using an equally likely choice (the so-called ‘Monte Carlo’ technique; Figure 17) for the input parameters,
results in a series of equally likely outcomes, reflecting the uncertainties of both the inputs and the applied formulae. In addition, despite many advances in the understanding of the geological processes involved, any model, although logically consistent and based on the best of present knowledge, is
304 PETROLEUM GEOLOGY/Exploration
likely to be only partially correct because there are variables and processes that are unknown, cannot be measured with sufficient precision, or cannot be accommodated in the model. Hence, it is important to calibrate the output from these models against real-life situations.
Exploration Drilling
Figure 13 Three dimensional seismic cube which can be re viewed from any direction, or rotation, on the computer workstation screen. Reproduced with permission of Ikon Science Limited.
Assuming that the geological and geophysical work outlined above has identified a ‘prospect’ – a possible hydrocarbon-bearing structure – it has then to be decided whether to drill an exploratory well, which is the only way to prove the actual presence of hydrocarbons. Factors involved in this decision are the likelihood of hydrocarbons being present in the prospect, the extent – and hence economic viability – of any hydrocarbons found, the cost of development if hydrocarbons are found, and political and economic considerations relating to the concession and the country in which the prospect is situated.
Figure 14 Three dimensional seismic interpretation system (so called ‘visionarium’).
Figure 15 Discovery by (A) field size classes, showing that the larger fields are discovered first (so called ‘creaming effect’), and (B) average size and number of fields showing that although the size of fields found is still decreasing. North Sea data, both plotted against time.
PETROLEUM GEOLOGY/Exploration 305
Figure 16 Hubbert’s Peak: predicted and actual US oil production.
Figure 17 Monte Carlo technique, combining input probability distributions describing total gross rock volume and porosity (illustrated in the top diagrams) together with those describing net to gross ratio, hydrocarbon saturation, formation volume factor and recovery factor (not illustrated) to give an expectation curve (cumulative frequency distribution; bottom diagram). Reproduced with permission of Nautilus Limited.
306 PETROLEUM GEOLOGY/Exploration
In addition to confirming the presence of hydrocarbons, an exploration well provides information from which the size of the accumulation can be assessed, the development of the field planned, and further exploration in the area undertaken. Rock cuttings brought to the surface by the drilling mud and specially taken core samples contribute to an understanding of the geological history of the sedimentary basin and, specifically, of the nature and characteristics of the reservoir rocks. Wireline logging (Figure 18), using special sondes to measure the electrical, acoustic, and radioactive properties of the rocks, allows key physical properties of the rocks, such as porosity and the fluid content – of oil, gas, or water – within the pore space, to be determined. Other logging tools can detect the presence of
fractures (which can be important for the flow characteristics of a reservoir), the dip (or inclination) of the rock layers, and sedimentary features (such as cross-stratification). If this information indicates a potentially commercial accumulation, then a flow test can be undertaken, in which the oil or gas is flowed to the surface: from flow and pressure measurements, further insights into the porosity and, particularly, permeability of the reservoir and the likely performance of wells under operational conditions can be gained. Further surveys and additional drilling are then undertaken to appraise the size of the accumulation and reduce the geological uncertainties; this information is used to plan the development of the field.
Exploration Costs Onshore, the exploration costs depend very much of the nature of the terrain: operations in remote mountainous or jungle areas are much more expensive than those in flat desert country. Operations in urban and agricultural areas and in areas of environmental sensitivity are expensive, as access can be difficult and strict regulations relating to noise, pollution, and environmental protection must be adhered to. Offshore, costs are less variable, though remote and less climatically favourable locations inevitably attract higher costs (Table 1).
Petroleum Agreements
Figure 18 Wireline log response of a hydrocarbon bearing sandstone reservoir, with a shale seal above and water saturated reservoir below the oil. DT, interval transit time; GR, gamma ray. Reproduced with permission of Millenium Atlas Company Limited.
Apart from the onshore USA, where individuals can have title to the mineral rights, in most countries these rights are vested in the state, and exploration for oil and gas can be carried out only under licences or contracts granted by the state. Petroleum agreements can be divided into two main types: licences (including leases, permits, and concessions) and production-sharing contracts. Under a licence agreement, a company is granted exclusive rights to explore particular areas. The company finances the exploration campaign and, if the exploration is successful, the development cost and in return is entitled to dispose of the production, sometimes subject to a deduction of a royalty in kind and an obligation to supply the domestic market. The state obtains revenues from the royalty and from taxation, often through special petroleum taxes. A variation on this type of agreement is a joint-venture agreement, where the state shares with the company the risk and expenses of the development phase; generally the company will carry the project at its own cost through the exploration phase.
PETROLEUM GEOLOGY/Exploration 307
Table 1 Indicative costs of exploration (for the year 2003) (all values are in US dollars) Onshore jungle terrain
Seismic cost per two dimensional kilometre Seismic cost per three dimensional square kilometre Well cost (to 3500 m)
7 000 12 000 (but can be significantly higher) 3 5 million
Onshore desert terrain
Seismic cost per two dimensional kilometre Seismic cost per three dimensional square kilometre Well cost (to 3500 m)
2 000 4 000
1.5 2 million
Onshore urban or agricultural environment
Seismic cost per three dimensional square kilometre
12 000
Offshore
Seismic cost per two dimensional kilometre Seismic cost per three dimensional square kilometre Well cost (to 3500 m)
600 5 000
6 8 million (but can be significantly higher for deep wells encountering high pressure and high temperature, or in remote locations; up to 25 million)
Postscript The development of new exploration techniques continues to improve geologists’ and geophysicists’ understanding of petroleum geology and to increase the efficiency of exploration by allowing wells to be sited more accurately and with a greater chance of success. However, even if the geological conditions for the presence of hydrocarbons are promising, exploration remains a high-risk business and investments are made in exploration many years before there is any prospect of producing the oil: those ventures that are successful must generate sufficient profit to pay for the unsuccessful ventures, both past and future. Thus, the fiscal framework established by states is vital to the commercial success of any exploration venture. The interests of governments wishing to develop their petroleum resources and the interests of companies as risk takers have much in common; petroleum-exploration strategies must take these mutual interests into account.
See Also Analytical Methods: Gravity. Petroleum Geology: Overview; Chemical and Physical Properties; The Petroleum System; Production; Reserves. Seismic Surveys.
Further Reading In a production-sharing agreement, the company carries the exploration risk and funds all the development and operating costs. These costs are then recovered from part of the production, known as ‘cost oil’. The remaining ‘profit oil’ is then split in a pre-determined manner between the state and the company. A further type of agreement is the service contract, in which the state contracts for a service from the company for which the company receives a fee; this can be, for example, a set fee per barrel produced or a percentage of the hydrocarbons produced while providing the service. Petroleum agreements take many forms, are increasingly complex, and are often a combination of elements from the different types. The main considerations for the company are the investment at risk, its ability to manage the operations, access to oil in the event of success, and the economic return on the investment.
Deffeyes KS (2001) Hubbert’s Peak: the Impending World Oil Shortage. Princeton: Princeton University Press. Gluyas JG and Swarbrick RE (2004) Petroleum Geoscience. Oxford: Blackwell. Johnson HD and Fisher MJ (1998) North Sea plays: geo logical controls on hydrocarbon distribution. In: Glennie KW (ed.) Petroleum Geology of the North Sea: Basic Concepts and Recent Advances, pp. 463 547. Oxford: Blackwell. Poelchau HS, Baker DR, Hantschel Th, Horsfield B, and Wygrala B (1997) Basin simulation and the design of the conceptual basin model. In: Welte DH, Horsfield B, and Baker DR (eds.) Petroleum and Basin Evolution, pp. 5 70. Berlin: Springer Verlag. Selley RC (1996) Elements of Petroleum Geology, 2nd edn. San Diego: Academic Press. Shell Briefing Service (1994) Upstream Essentials. London: Shell International Petroleum Company Limited. Steinmetz R (ed.) (1992) The Business of Petroleum Ex ploration. Treatise of Petroleum Geology. Handbook of Petroleum Geology. Tulsa: American Association of Petroleum Geologists.
308 PETROLEUM GEOLOGY/Production
Production K J Weber, Technical University, Delft, The Netherlands L C van Geuns, Clingendael International Energy Programme, The Hague, The Netherlands ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The geological activities following the discovery of a petroleum accumulation commence with a thorough analysis of the find in terms of structural definition, reservoir rock characteristics, and accumulation conditions. This is followed by the selection of appraisal well locations and the study of the results. If considered commercially viable, a multidisciplinary team, including specialists of all disciplines in petroleum engineering, designs an initial development plan. At this stage, the volume of petroleum-in-place is estimated and a dynamic model is built to test the production capacity and producible reserves of the field, which may be generated by a series of alternative development plans. After executing the selected initial development plan, the reservoir model is updated with new data. The increasing information on field performance is used to calibrate the model by history matching. In certain favourable cases, fluid movements can be observed on additional three-dimensional (3D) seismic data (see Seismic Surveys). Further development activities, such as infill drilling, side tracking, recompletions, or fluid injection schemes, are planned on the basis of the updated model. Detailed field studies by integrated teams of petroleum engineers and production geologists are carried out periodically until the ultimate stages of the field life (Figure 1). Reservoir performance is a function of the reservoir characteristics and the petroleum properties. The study of reservoir characteristics is the primary task of the production geologist. An overview is given of the influence of various reservoir heterogeneities on fluid flow. To analyse these characteristics, and to quantify the results, requires a wide range of techniques. These cover such a large field of subjects that the work has to be carried out by a group of specialists in seismology, sedimentology, structural geology, geostatistics, and petrophysics. In addition, a good understanding of the principles of reservoir and production engineering is required to be able to function successfully in a multidisciplinary team.
The production geological activities at the various stages of field development have a number of different purposes which all require geological input of some kind (Figure 2). This is illustrated by the different techniques and modelling methods that are employed at each stage (Figure 3). The pace of development of new techniques, in particular integrated computer systems, is very fast. As a result, the production geological activities are centred around workstations being served by a database containing all relevant information (Figure 4).
Historical Development of Production Geology Production geology as a separate petroleum engineering discipline was only generally recognized in the 1950s. Early field development, with shallow, closely spaced wells and only cuttings and surface outcrop observations, provided little scope for reservoir studies. In the USA and Russia, where grid drilling was and still is in fashion, most companies did very little production geology work in the appraisal stage. In Indonesia, where complicated structures were encountered, the need was felt for more geologist involvement at an early stage. Around 1915, the Bataafse Petroleum Company (Royal Dutch) started to engage mining engineers with the particular task to improve cooperation between production and geology to achieve a more efficient exploitation of the field. The reports from that time show that structural interpretation and detailed mapping of the producing horizons constituted the main tasks. In these pre-logging and pre-seismic days, correlations were carried out with the aid of cuttings glued to narrow planks. The wireline logs developed by the Schlumberger brothers in 1927 progressively replaced these early methods. Just before World War II, effective reflection seismic methods came into use, although any sophisticated processing had to wait for the availability of more powerful computers after 1960. This is also the time that production geology, also called ‘development geology’, began to be more generally employed, although, in most companies, the geologists were still firmly attached to the exploration departments. In the meantime, sedimentology had made great strides forwards and more attention was given to the internal architecture of the reservoir. Core descriptions and analysis were improved. Research studies of outcrops and recent sedimentation of both
PETROLEUM GEOLOGY/Production 309
Figure 1 Field study cycles.
Figure 2 Purposes of reservoir characterization. Courtesy of Shell.
310 PETROLEUM GEOLOGY/Production
Figure 3 Reservoir modelling techniques during successive development stages. Reproduced with permission from Weber KJ (1995) Visions in reservoir management what next? In: Reservoir Characterisation: Integration of Geology, Geophysics and Reservoir Engineering, The Third JNOC TRC International Symposium, February 2 23, Technical Research Centre, Chiba, Japan, pp. 1 15. Chiba: Japan National Oil Corporation.
Figure 4 High level workflow for computerized reservoir modelling. Courtesy of Shell.
clastics and carbonates were carried out by several major oil companies, in combination with laboratory studies of diagenetic processes. Rock classification systems such as the ‘Archie classification’ appeared,
providing the link between rock texture and permeability. The introduction of induction-, latero-, formation density- and dip-meter logs greatly improved the quantitative assessment of hydrocarbon volumes
PETROLEUM GEOLOGY/Production 311
Figure 5 Penetration log correlation.
and distributions. Between 1960 and 1980, most of the presently used production geological methods were developed. As a typical example of a classical production geological method, Figure 5 shows a penetration log correlation. The different rates of penetration of sandstones and shales (whilst drilling with constant speed and weight) can often be related to spontaneous potential (SP) or gamma-ray (GR) log curves, and thus the progress of wells could be monitored prior to the introduction of modern logging-while-drilling (LWD) much later. Further refinements in structural and reservoir imaging came with the spectacular advance of seismic acquisition, processing, and interpretation. Seismic attribute analysis is now one of the major production geological techniques. Sophisticated borehole imaging logs formed another new source of detailed information. The increase in
computer capacity also led to reservoir simulation based on complex heterogeneous reservoir models comprising very large numbers of grid cells, which, in turn, required realistic well-calibrated geological models. This promoted the use of geostatistical methods and stochastic modelling. The present trend is to use a combination of all relevant reservoir data in integrated databases from which visualization can be generated, simulation models can be constructed, and well trajectories can be planned.
Reservoir Characteristics Influencing Fluid Flow and Storage Capacity Controlling Reservoir Characteristics
The quality of a reservoir is directly related to the available pore space (porosity) and the capacity to
312 PETROLEUM GEOLOGY/Production
conduct fluid flow through the pores (permeability). Porosity depends on texture (grain size distribution, grain shape, and packing) and the degree of diagenesis (compactions, leaching, cementation). Permeability is a function of the size, shape, and distribution of pore spaces and their interconnectedness. Rocks with a similar texture that have been subjected to similar diagenetic processes will generally show a good correlation between porosity and permeability. However, for the same porosity, significantly different permeabilities may be found in the same reservoir because of facies-related textural differences and/or diagenetic variations.
There are a series of geological features that influence fluid flow and hydrocarbon recovery (Figure 6). Faults, fractures, reservoir architecture, internal zonation of genetic units, low permeability intercalations, sedimentary structures, and pore shape all have an impact on a different scale. The drive mechanism can also be influenced by these factors because they control the pressure support from the aquifer surrounding the reservoir. An accurate evaluation of a reservoir has to take into account all of the relevant heterogeneities at the right scale. The resulting 3D reservoir model forms the basis for volumetric estimates of the trapped hydrocarbons and simulations of
Figure 6 Reservoir heterogeneity influencing fluid flow and hydrocarbon recovery, and relevant data for analysis. ROS, Residual Oil Saturation. Modified with permission from Weber KJ and van Geuns LC (1990) Framework for constructing clastic reservoir simulation models. Journal of Petroleum Technology 42(10): 1248 1253, 1296 1297. ß Society of Petroleum Engineers.
PETROLEUM GEOLOGY/Production 313
reservoir performance. In this way, the appraisal campaign can be planned and the optimum development scheme can be determined. Faults and Fractures
Faults have a significant influence on hydrocarbon migration and trapping. The sealing capacity of a fault can be controlled by cataclasis and diagenesis in the fault zone, juxtaposition against a tight layer, or smearing of clay along the fault. In deltaic settings, where multiple reservoirs separated by shales are common, the determination of the clay smear potential along faults is particularly useful. Fractures can have a positive or a negative effect on fluid flow. Cemented fractures compartmentalize the reservoir and reduce recovery. Swarms of shear fractures are common along many faults and can significantly reduce productivity over a wide zone. Open fractures, which are mainly associated with folded carbonate reservoirs, can increase productivity enormously, although they may also be the cause of early water breakthrough to the wells. Clastic Reservoirs
Clastic reservoirs occur in a wide variety of sedimentary settings, ranging from almost homogeneous beds to an intricate 3D network of sinuous sand bodies. From an observation of the existing reservoir configurations, it appears that most clastic reservoirs can be classified into one of three basic architectural types – layer-cake, jigsaw, or labyrinth – which represent a decreasing order of connectivity (Figure 7). Correlation is comparatively easy in layer-cake reservoirs, such as barrier bars or shallow marine sheet sands. Thus, such reservoirs can already be modelled quite realistically in the appraisal stage. At the other end of the scale are the labyrinth-type reservoirs, such as low-sinuosity distributary channel complexes, which require a dense well spacing to achieve a deterministic correlation. If we add common internal permeability distributions and discontinuity patterns, the impact of the reservoir types on sweep efficiency can be shown (Figure 8). In layer-cake reservoirs, not only the sands but also the shales can be very continuous, preventing vertical flow. High permeability at the base of point bars can lead to early water breakthrough, while the reverse, as in barrier bars, can have the opposite effect. Jigsaw reservoirs, with interlocking sand bodies with contrasting properties and local baffles to vertical and horizontal flow, often show uneven drainage and erratic flood fronts. Labyrinth reservoirs reflect their discontinuous nature in very uneven sweep efficiency and bypassing of major oil volumes.
Carbonate Reservoirs
Carbonate reservoirs are generally more heterogeneous than clastic reservoirs, because the original organic sediments can differ widely in their grain size and composition, but also because of the large influence of diagenesis. The original sedimentary setting can be split into four main types: the deeper chalks and turbidites; the shallow-water build-ups; platform/shoal sediments; and ramp/nearshore packages (Figure 9). These reservoirs are sometimes left with rock characteristics closely related to their original properties, but often considerable changes are effected by the various diagenetic processes. However, in the majority of cases, the original material with a similar texture and composition will be transformed into the same rock type. There may be a complete reversal of the original porosity and permeability distribution. A porous reef may become tight and a low-permeability, fine-grained back-reef may change into a permeable sucrose dolomite (Figure 9). Several diagenetic processes can increase the heterogeneity markedly. Karstic leaching can be pervasive or restricted to certain zones. Some carbonate reservoirs consist of virtually tight rock in which leached fractures and voids comprise the only effective pore space. Another complicated type, which is very difficult to evaluate, is formed by reservoirs submitted to a mixed process of leaching and dolomitization, forming a network of moldic sucrose dolomite nodules and tunnels through a tight matrix. For nearly all carbonate reservoirs, detailed microscopic studies are required to unravel their diagenetic history and to identify facies and composing organisms. The rock types have to be classified to be able to analyse and predict the permeability distribution, which, in carbonates, can cover an extremely large range even in a single reservoir (Figure 10). Small-Scale Heterogeneity
Although the larger scale heterogeneities usually control the production performance, the sedimentary structures and rock texture can cause anisotropy of the reservoir and have a strong influence on residual oil distribution. In cross-bedded reservoirs which have been subject to severe diagenesis, it may be necessary to measure the permeability contrast between the laminae and to take the resulting anisotropy into account in the dynamic fluid flow model. It has also been shown that, even with permeability contrasts between the laminae as low as two, capillary forces can trap of the order of 10% additional oil in comparison with more homogeneous rock. Detailed studies of the capillary pressure curves and relative
314 PETROLEUM GEOLOGY/Production
Figure 7 Clastic reservoirs classified into three architectural types: layer cake, jigsaw, and labyrinth. N/G ratio, Net /Gross. Modified with permission from Weber KJ and van Geuns LC (1990) Framework for constructing clastic reservoir simulation models. Journal of Petroleum Technology 42(10): 1248 1253, 1296 1297. ß Society of Petroleum Engineers.
PETROLEUM GEOLOGY/Production 315
Figure 8 Connected permeable pathways through the major reservoir types. Courtesy of Shell.
permeability to oil and gas for each rock type are necessary. Major lateral differences in apparent oil– water contact can result from variations in rock type.
Techniques Seismic Methods
The range of methods, techniques, and tools used in production geology has become very large. Seismic techniques are of primary importance in the appraisal stage, but are also used for reservoir monitoring (Figure 11). The traditional use of seismic techniques for structural definition has been much improved by 3D surveys and better processing techniques. Where seismic resolution is good and acoustic impedance contrasts are sufficiently large, amplitude measurements along the beds can reveal depositional patterns as well as oil and gas accumulations (Figure 12). Besides acoustic impedance mapping, amplitude versus offset (AVO) techniques and shear-wave analysis are now being employed. These techniques are powerful methods to determine lithology, while shear waves are affected by open fractures. Repeat 3D surveys (so-called 4D seismic) can reveal movements of fluid interfaces and saturation changes (Figure 12). Analysis of changes in acoustic impedance and
velocity can also be used to estimate pressure changes in the reservoir resulting from hydrocarbon production. This can reveal compartmentalization caused by sealing faults and uneven aquifer drive. Rock stresses can be derived from measurements of shear-wave polarization. In favourable cases, porosity and pore-fill analysis is carried out via acoustic impedance measurements, calibrated with well data. In this way, volumetric estimates of hydrocarbon-in-place may be made. Of particular importance is to relate the 3D seismic reservoir models to models made from well data correlations with geostatistical methods. Improved calibration of seismic data, attributed using neural network techniques, can provide estimates of facies types and related properties throughout the reservoir. Seismic methods are frequently used for planning well trajectories (Figure 12). Three-dimensional displays and interactive virtual reality techniques are employed for this purpose. The developments in seismic methods for reservoir analysis have been so spectacular in the recent past that further research is likely to result in important advances. Core Description and Analysis
Facies and rock types have to be identified and characterized from cores and wireline logs. Systematic
316 PETROLEUM GEOLOGY/Production
Figure 9 Scheme of the main carbonate depositional settings and common pattern of permeability distribution. Courtesy of Shell.
PETROLEUM GEOLOGY/Production 317
Figure 10 Generalized trends for porosity versus permeability of the main primary and secondary carbonate pore types. Courtesy of Shell.
geologically steered core analysis is the key to the proper understanding of the reservoir composition and the basis for log calibration, static and dynamic modelling, detailed seismic interpretation, and the identification of key heterogeneities. An overview of core analysis methods and applications is given in Figure 13. Core observations under normal light can be augmented by ultraviolet (UV) and infrared photography to reveal remaining oil and sedimentary structures. Standard petrophysical analysis should be carried out with a sample choice and spacing reflecting the reservoir heterogeneity. Probe permeability measurements or cube-shaped samples may be required to analyse permeability anisotropy (Figure 14). The influence of in situ stress on reservoir properties and rock compressibility has to be measured.
The core data are used to establish rock types with similar characteristics, such as porosity/permeability relationship, capillary pressure, and relative permeability curves. Each rock type usually represents a specific facies type having a similar lithology and texture. The core data are correlated with the logs to identify typical characteristics that allow the recognition of rock types in uncored wells. This is necessary to delineate facies distribution for well-to-well correlation, as well as to estimate permeability via porosity– permeability transforms. High-resolution borehole imaging logs are very useful for the identification of structural features, sedimentary structures and dips, and trends of individual sand bodies (Figure 15). A series of electric logs is used to identify rock types in uncored wells. Linear multiple regression of, for example, GR, neutron and density log readings of
318 PETROLEUM GEOLOGY/Production
Figure 11 Overview of seismic techniques for reservoir delineation, analysis, and monitoring. AVO, amplitude versus offset; EOR, enhanced oil recovery; LWD, logging while drilling; VSP, Vertical Seismic Profile. Modified with permission from Weber KJ (1995) Visions in reservoir management what next? In: Reservoir Characterisation: Integration of Geology, Geophysics and Reservoir Engineering, The Third JNOC TRC International Symposium, February 2 23, Technical Research Centre, Chiba, Japan, pp. 1 15. Chiba: Japan National Oil Corporation.
known rock types in cored well is used to set up a diagnostic system for uncored wells. Alternatively, neural networks can be trained to carry out this identification. The advantage of the neural network techniques is that the influence of the input parameters is non-linear. Permeability Distribution
Permeability measurements are amongst the most difficult to derive from well data and yet comprise one of the most useful and relevant measurements in production geology. The permeability distribution in the well is derived from cores and via rock type identification in uncored wells. However, this only
provides local horizontal permeability data around boreholes. Cores and logs in horizontal holes can provide lateral permeability development, but only along narrow zones. There are a large number of techniques to augment the permeability database both horizontally and vertically on a range of scales (Figure 16). Production geologists must play an active role in identifying opportunities to measure the permeability, which provides an important control on fluid flow in the reservoir. Vertical permeability on the medium and large scale is particularly difficult to estimate from cores and logs. For this purpose, multiprobe well testers, pulse testing, production logging, and pressure build-up analysis are used.
PETROLEUM GEOLOGY/Production 319
Figure 12 Integrated use of three dimensional (3D) seismic measurements for static modelling purposes.
Fault and Fracture Analysis
Faults are mainly analysed from seismic data, although well-to-well correlation can reveal fault cutouts. Dip-meter logs can also indicate fault positions by showing drag patterns. The fault types have to be analysed from the deformation pattern seen on seismic measurements and the structural history of the sedimentary basin. The associated sealing capacity of the faults is closely related to the type of fault, fault throw, reservoir lithology, and depth. Cataclastic fault zones tend to be cemented and impermeable after burial to a depth of 1–2 km. Clay smearing along normal faults is an important process in many deltaic reservoirs. Estimates of the sealing potential can be based on the shale thickness that passed along a fault plane (Figure 17). This can be achieved by a log analysis of wells near the fault and
the fault throw derived from the cut-out observed in a well intersecting the fault. If good quality seismic data are available, however, the fault throw distribution can be established. With 3D seismic data, it is possible to sample the acoustic impedance values along both flanks of a fault and to translate these data into synthetic lithological logs. By comparing and correlating the hanging- and footwall stratigraphy, both the 3D distribution of the fault throw and the thickness of the shales that moved across any point of the fault surface can be obtained. The method requires local calibration because the ductility of the shales at the time of faulting can differ markedly. Fractures are seen on borehole imaging logs. By the use of Stoneley waves (shear waves), open fractures can be identified and the fracture width can be estimated. This provides estimates of the usually very low
320 PETROLEUM GEOLOGY/Production
Figure 13 Facies definition from cores and logs. Courtesy of Shell.
PETROLEUM GEOLOGY/Production 321
Figure 14 Probe permeability measurements showing small scale heterogeneities. Reproduced with permission from Weber KJ and van Geuns LC (1990) Framework for constructing clastic reservoir simulation models. Journal of Petroleum Technology 42(10): 1248 1253, 1296 1297. ß Society of Petroleum Engineers.
fracture porosities. Flowmeter-type production logs can also show open fractures. The shear zones around faults can be analysed with borehole imaging logs in horizontal wells. Fracture density, as a result of tectonic deformation, can be estimated by determining the rate of change of the dip. In fault and fracture analysis, extensive outcrop studies have added enormously to the knowledge of fault zone compositions and fracture distributions (Figure 18). Correlation and Use of Analogues
The correlation from well to well requires a detailed knowledge of the facies distribution and expectation values of the typical shape and dimensional relationships of genetic reservoir bodies. For this purpose, there is a need for a data bank of such information based on outcrop and field data, complete with their sedimentological setting, including type of basin and climate, to be able to choose a proper analogue example. Such data banks have been compiled by several oil companies, while many universities are
engaged in reservoir analogue outcrop studies. The stratigraphy forming the major framework of the reservoir can be interpreted with sequence stratigraphical principles. These principles and, especially, the analysis of sea-level fluctuations can be used to delineate the sedimentary successions, and are a great help in correlating the wells (Figure 19). Sea-level changes have a large influence on the sedimentary processes along time slices of the reservoir and, in carbonates, they also have a strong impact on diagenesis (e.g., leaching and dolomitization). The well-known techniques of comparing the floral and faunal contents of discrete intervals remain very useful, not only for stratigraphical purposes, but also for facies analysis. The palaeontological and palynological studies can be carried out on drill cuttings whilst drilling, and thus can provide important indications of drilling progress through a known stratigraphical sequence. In combination with LWD techniques, the correct casing setting depth can be determined with sufficient safety margins.
322 PETROLEUM GEOLOGY/Production
Figure 15 High resolution acoustic borehole images compared with core slab showing aeolian cross bed set. Reproduced with permission from Frikken HW (1999) Reservoir Geological Aspects of Productivity and Connectivity of Gasfields in the Netherlands. PhD thesis, TU Delft.
Another useful technique is oil typing via gas chromatography. Detailed comparison of oil samples has proven to be a sensitive indicator of reservoir continuity. Oils that are not in communication either vertically or laterally will show distinctive differences in their chemical fingerprints.
Volumetric Estimates
The static modelling carried out as the basis for volumetric estimates and dynamic modelling requires 3D infilling of the space between the wells. In an early stage, hydrocarbon-in-place estimates are made by
PETROLEUM GEOLOGY/Production 323
Figure 16 Overview of methods to determine or estimate permeability distribution on a range of scales. Modified with permission from Weber KJ (1995) Visions in reservoir management what next? In: Reservoir Characterisation: Integration of Geology, Geophysics and Reservoir Engineering, The Third JNOC TRC International Symposium, February 2 23, Technical Research Centre, Chiba, Japan, pp. 1 15. Chiba: Japan National Oil Corporation.
combining the tentative probability distribution of the controlling parameters: reservoir volume, net-togross, porosity, petroleum saturation, and formation volume factor (FVF), the ratio between the net petroleum volume in situ and under standard conditions. The cumulative probability distribution curve or expectation curve for the petroleum-in-place is generated with Monte Carlo techniques. The probability distributions of the parameters are randomly sampled and the resulting petroleum-in-place volumes are plotted until the graph hardly changes by additional values (Figure 20). As an interim step towards 3D modelling, Net-OilSand (NOS) maps can be constructed, which yield better estimates than combining the field averages of all parameters. For thin reservoirs, such maps almost represent 3D models, if the properties of the individual sand bodies can also be plotted (Figure 21). Geostatistics
Classic statistical methods for the analysis of numerical data sets to determine distributions, standard deviations, and correlations have been used in production geology for a long time. There are, however, a number of problems for which these methods are not suitable. Firstly, one is frequently confronted with sparse data sets which do not allow reliable
contouring or correlation. Secondly, there is a need to quantify uncertainty in spatial distributions and also to estimate the possible reduction in uncertainty resulting from drilling an additional well in a given place. Thirdly, one wants to use spatial relationships established for specific types of reservoir body to constrain the models based on well-to-well correlations. The method which has been developed to handle these problems is called ‘kriging’, after the originator Daniel Krige, a South African mining engineer. It is based on the reasonable assumption that the unknown spatial distribution of a geological property can be predicted on the basis of the spatial distribution of the measurement of that property. The main tool in this method is the ‘variogram’, which quantifies the spatial continuity of a property. It is a graph of the variance of the difference of two measurements as a function of their spacing. One can use variograms made on the basis of a data set. If the number of data points is insufficient to define a variogram, one can select a best-fitting variogram or use a variogram that is typical for a modelled property. Known trends can be honoured by using different variograms in different directions. An example of the above is shown in Figure 22. A map is made of the thickness variation of a reservoir using a best-fitting variogram. Next, the residual
324 PETROLEUM GEOLOGY/Production
Figure 17 Clay smearing along fault as seen in outcrop. Note discontinuous clay smear of thin (red) shale.
variation between the variogram and the thickness map is computed to show the uncertainty at any point of the map. Finally, one can place a grid over the map and compute the reduction of uncertainty resulting from drilling a fictitious well in any of the grid points. This method can be used to select a next appraisal well. Further applications of ‘kriging’ methods include the combination of data from different sources and with different accuracy to estimate a property. In this way, reservoir depth data from wells and from seismic measurements can be merged into one map. Probabilistic reservoir modelling also makes use of variograms to constrain the extent of the bodies within the reservoir. A series of equally probable models can be generated. Here lies one of the most
useful applications of analogue data derived from outcrops and densely drilled fields. Static Modelling
The particular technique used for 3D reservoir modelling depends on the reservoir type. For layer-cake reservoirs, deterministic models can often be made directly from the well-to-well correlations and the rock properties can be interpolated. For labyrinth reservoirs, on the other hand, deterministic modelling is rarely possible without a very close well spacing. Computer systems have been developed to generate a series of equally probable 3D models through ‘probabilistic’ modelling techniques. Most systems work in three steps. Firstly, correlatable reservoir bodies are determined and incorporated in the
PETROLEUM GEOLOGY/Production 325
Figure 18 Faults and fractures in anticlinal structure of a thick limestone formation.
Figure 19 Sequence stratigraphical correlation in reservoir modelling. Courtesy of Shell.
model. Where extrapolations beyond control points are necessary, use is made of an analogue database for the relevant genetic type. Secondly, the uncorrelatable bodies are considered, for which the dimensions are also derived from the database, while body orientations are taken from borehole imaging logs or estimated on the basis of the general geological model. Characteristic variograms for the thickness distribution of genetic sand body types in different orientations relative to their expected trend are also used. Thirdly, especially when the well spacing is rather large, there are smaller or narrower reservoir
bodies not penetrated by wells. Such bodies are added using statistical estimates of their occurrence from the wells. Their position and dimensions are conditioned by geological modelling rules and the analogue database (Figure 23). Jigsaw-type reservoirs are also difficult to correlate in the appraisal stage and probabilistic modelling is required. In a later stage, a large part of the architecture may be determined and only limited recourse has to be taken to probabilistic techniques. The resulting architectural models can be compared with the seismic models and, in favourable
326 PETROLEUM GEOLOGY/Production
Figure 20 Approach to determine the cumulative probability (expectation) curve for the oil in place volume. RBV, Rock Bulk Volume.
Figure 21 Net Oil Sand (NOS) map of deltaic reservoir.
cases, quantitative correlations can be made. This reduces the spread of possible model configurations. In any case, the permutation of the probabilistic models generated has to be carefully ranked in order of probability by screening them with regard to the presence of geological anomalies. A limited number of 3D models are selected as representing the likely
range of variation in the reservoir with respect to hc-volume (hydrocarbon volume), connectivity, architecture, and permeability distribution. Dynamic Modelling
The profitability of a hydrocarbon reservoir depends crucially on how its development is planned in
PETROLEUM GEOLOGY/Production 327
Figure 22 Geostatistical methods in mapping and well plan ning. (A) Estimated thickness map made with the best fitting variogram. (B) Thickness uncertainty (66% confidence limit) computed from the residual variation between the variogram and thickness map. (C) Expected reduction in estimation variance of the mean thickness after drilling well in location A.
the context of recovery mechanism(s) and infrastructure environment. A key element from the outset of planning is a conceptual model of the hydrocarbon reservoir (ranging from a simple analytical model to a multitude of equiprobable realizations). The reservoir model allows the evaluation of alternatives by providing an estimated range of hydrocarbon volumes that are potentially recoverable under various development schemes. The majority of the field development plans are based on reservoir simulation models with varying degrees of complexity. Integrated dynamic reservoir models are constructed by reservoir engineers and require input on geological structure and architecture,
rock and fluid properties and their distribution, displacement characteristics, and description of the wells and the surface facilities. Static and dynamic modelling programs differ from each other in terms of the number of volume elements they can handle. Like the 3D static modelling packages, the dynamic modelling programs require the reservoir to be segmented into unit cells, each of which has uniform properties, before it can perform computations. Static modelling computations involve the categorization, sorting, and counting of voxel cells; dynamic modelling computations involve the balancing of the fluid masses and pressures within and between ‘gridblock’ cells. Upscaling programs enable one to aggregate the geological detail (¼3D static model) into the bigger elements of a reservoir engineering model (¼3D dynamic simulation model) by using a method of permeability averaging (e.g., harmonic, arithmetic, geometric). These programs also produce a corresponding gridblock model of porosity through conventional averaging of voxel porosity values. It is important that the 3D modeller identifies the key geological characteristics that should not be lost during upscaling (e.g., thin-bedded shales that may act as permeability barriers or baffles) (Figure 24). Not all geological architecture will have equal weighting in terms of its influence on reservoir performance, and therefore the geologist (with the help of a reservoir engineer) must filter through the geological detail in order to determine what is/is not important for reservoir simulation. Iteration between the static and dynamic models is recommended in order to ensure that the relevant detail (flow units and barriers) is not compromised during the upscaling procedure.
Conclusions The scope of production geological activities and methods has become very wide, requiring an equally diversified group of specialists to cover all topics. Advanced seismic techniques are employed at every stage, from discovery to additional development planning late in the life of a field. The backbone of the profession is formed by expert sedimentologists who can understand and model the architecture of reservoirs, including the features that influence fluid flow. There is clearly a division in clastic and carbonate reservoir expertise because of the difference in sedimentological and diagenetic processes and relevant analysis techniques. Structural geology is another speciality that should be handled by experts. Again, there are large differences between faults and fractures in clastic and carbonate reservoirs. The advances in seismic resolution
328 PETROLEUM GEOLOGY/Production
Figure 23 Summary of three dimensional (3D) reservoir modelling. Courtesy of Shell.
on the large scale and borehole imaging logs on the small scale have greatly improved the analysis of faults and fractures. Geostatistical methods are widely used in constructing maps and reservoir models, which reflect the level of uncertainty, but in which analogue information can be incorporated. The present approach to production geology is heavily dependent on computer systems, and much of the time is spent behind a workstation or personal
computer. Modern interactive computer techniques and integrated data systems are used to design models and well trajectories. Visualization plays an important part in this work and virtual reality techniques are already being used. Typical of the modern reservoir management approach is the integrated team composed of various production geology specialists, together with petrophysicists, reservoir engineers, and production technologists.
PETROLEUM GEOLOGY/Production 329
Figure 24 Building a reservoir model with measured data, expert knowledge, and statistics. ß Schlumberger Limited, used with permission.
330 PETROLEUM GEOLOGY/Production
The future will undoubtedly bring important further improvements in seismic techniques. There is a need for more analogue data and sedimentological knowledge to aid in reservoir modelling. The reservoir models will benefit more and more from integration with seismic information, both with respect to reservoir architecture and to fluid content and movements. The high costs of deep-water field appraisal and development will place a premium on the timely construction of realistic reservoir models. On land, there is a large scope for enhanced oil recovery that equally requires reliable detailed reservoir models. Thus, it may be expected that there will be a significant further development in production geological techniques.
See Also Diagenesis, Overview. Petroleum Geology: Overview; Chemical and Physical Properties; Exploration; Reserves. Sedimentary Environments: Depositional Systems and Facies. Sedimentary Rocks: Dolomites; Limestones; Sandstones, Diagenesis and Porosity Evolution. Seismic Surveys.
Further Reading American Association of Petroleum Geologists (AAPG) (1993) Development Geology Reference Manual. AAPG Methods Series No. 10. Tulsa, OK: American Association of Petroleum Geologists. Barwis JH, McPherson JG, and Studlick JRJ (eds.) (1990) Sandstone Petroleum Reservoirs. New York: Springer Verlag. Bishop CM (1995) Neural Networks for Pattern Recogni tion. Oxford: Clarendon Press. Brown AR (1999) Interpretation of Three Dimensional Seismic Data. AAPG Memoir 42. Tulsa, OK: American Association of Petroleum Geologists. Dickey PA (1986) Petroleum Development Geology, 3rd edn. Tulsa, OK: PennWell Publishing Company. Dikkers AJ (1985) Geology in Petroleum Production. Developments in Petroleum Science 20. Amsterdam New York: Elsevier Science.
Flint SS and Bryant ID (eds.) (1993) The Geological Modelling of Hydrocarbon Reservoirs and Outcrop Analogues. Special Publication No. 15 of the IAS (Inter national Association of Sedimentologists). Oxford: Blackwell Scientific Publications. Goovaerts P (1997) Geostatistics for Natural Resources Evaluation. New York: Oxford University Press. Jahn F, Graham M, and Cook M (1998) Hydrocar bon Exploration and Production. Developments in Pet roleum Science 46. Amsterdam New York: Elsevier Science. Jensen JL, Lake LW, Corbett PWM, and Goggin DJ (1997) Statistics for Petroleum Engineers and Geoscientists. Upper Saddle River, NJ: Prentice Hall. Laudon R (1996) Principles of Petroleum Development Geology. Petroleum Engineering Series. Upper Saddle River, NJ: Prentice Hall. Lowell JD (1985) Structural Styles in Petroleum Explor ation. Tulsa, OK: OGCI Publications, Oil and Gas Consultants Inc. Moller Pedersen P and Koestler AG (eds.) (1997) Hydrocar bon Seals Importance for Exploration and Production. Norsk Petroleum Forening/NPF Special Publication No. 7. Singapore: Elsevier Science. Nelson RA (1985) Geological Analysis of Naturally Fractured Reservoirs. Houston, TX: Gulf Publishing Company. Reading HG (ed.) (1996) Sedimentary Environments and Facies, 3rd edn. Oxford: Blackwell Scientific Publica tions. Van Wagoner JC, Mitchum RM, Campion KM, and Rahmanian VD (1990) Siliciclastic Sequence Stratigraphy in Well Logs, Core and Outcrops for High Resolution Correlation of Time and Facies. AAPG Methods in Ex ploration Series No. 7. Tulsa, OK: American Association of Petroleum Geologists. Weber KJ (1995) Visions in reservoir management what next? In: Reservoir Characterisation: Integration of Geology, Geophysics and Reservoir Engineering, The Third JNOC TRC International Symposium, February 2 23, Technical Research Centre, Chiba, Japan, pp. 1 15. Chiba: Japan National Oil Corporation. Weber KJ and van Geuns LC (1990) Framework for con structing clastic reservoir simulation models. Journal of Petroleum Technology 42(10): 1248 1297.
PETROLEUM GEOLOGY/Reserves 331
Reserves R Arnott, Oxford Institute for Energy Studies, Oxford, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Fossil fuels can be broadly categorized as either resources or reserves. Resources include all fuels, both those identified and those as yet unknown. Reserves are that portion of the identified resources which can be economically extracted and exploited using current technology. Petroleum reserves can be labelled under a wide variety of physical, chemical, and geological circumstances. For example, the boundaries between crude oil as a liquid and condensates have long been the subject of controversy. In addition, there are issues of definition as to what to include or exclude from a particular production forecast, as there are as to what can and cannot be reported as reserves because of legal and political considerations. The only near-certainty on the supply side is the actual volume of oil that has been produced. This is because the definitions of petroleum reserves often include assumptions with regard to existing technology and present economic conditions. However, there is no uniformity or stated policy as to the time period over which the existing technology and present economic conditions are anticipated to prevail. As a consequence, there is often fierce debate about how long existing petroleum reserves are likely to last and the economic consequences if future production cannot meet demand.
Definitions of Reserves Because of the considerable uncertainty surrounding the definitions of reserves, most authors prefer a probabilistic rather than a deterministic approach (Figure 1). An initial declaration of recoverable oil that can, with reasonable certainty, be recovered in the future under existing economic and operating conditions is the usual definition of the ‘proven’ reserves. However, all fields will also be declared as having additional volumes of ‘probable’ and ‘possible’ reserves. The definition of ‘proven’, ‘probable’, and ‘possible’ reserves varies from country to country. ‘Proven’ reserves are usually defined as being P90 reserves, indicating that there is a greater than 90% chance that the actual proven reserves base will be higher, and a 10% chance that it will be less. Similarly, ‘probable’ and ‘possible’ reserves can be defined
as P50 and P10 reserves, respectively. The important point here is that, when oil reserve numbers are quoted in the literature, it must be realized that the numbers are probabilistic and, in the case of proven reserves, are more than likely to be exceeded. Therefore, it is wrong for commentators to argue that the volumes of discovered reserves are a fixed entity as, in the case of proven reserves, there will be a 90% chance that the initial reserve number will be exceeded. Declarations of proven reserves are often only for a specific reservoir. Therefore, when such information is summed, eventually, to the regional and national level, the simple arithmetical addition of a large number of independent values, each representing the 90% probability of a specified volume in a specific reservoir, produces a higher joint probability of the total. It is for this reason that field growth and rates of field growth are well documented in practice. Resources can be defined as existing reserves plus all of the accumulations of reserves that may eventually become available. These additional reserves might have already been discovered but be uneconomic, or may not yet have been discovered. The distribution of fossil fuels can therefore be viewed as a pyramid, with a small amount of higher quality resource at the top, but with increasingly large amounts of lower grade resource as we move down the pyramid (Figure 2). The costs of retrieving the resource increase lower down in the pyramid, making a larger amount of the resource available at higher prices. The issue as to what defines the total volume of petroleum resources depends on where the pyramid is sliced, and this is a very subjective decision. There is often a curious circularity in the estimates of undiscovered and ultimate reserves. Undiscovered reserves are the difference between estimates of the discovered and ultimate reserves. However, ultimate resource estimates obviously depend on an estimate of undiscovered reserves. Although there are a number of ways in which to try and surmount this difficulty, the most commonly described method involves an examination of the discovery pattern. In general, there is some regularity to this pattern, with the largest fields being discovered first with attention then switching to the smaller fields. Therefore, in any one basin, the curve (commonly referred to as the creaming curve) relating cumulative discoveries to wildcat wells is usually hyperbolic with the asymptote revealing the level of ultimate reserves. However, this method is not immune to error, especially if larger
332 PETROLEUM GEOLOGY/Reserves
Figure 1 Expectation curve used to find approximate values of proven, probable, and possible reserves. bbl, barrels.
Figure 2 Resources visualized as a pyramid, with a small volume of prime resources that are of high quality and easy to extract, and a large volume of resources of lesser quality that are more difficult or expensive to extract.
fields are discovered late in the life of a particular basin. In the case of the UK sector of the North Sea (Figure 3), the asymptote has moved over time as the older fields have grown in size through better recovery techniques and reservoir definition. As a consequence, estimates of the total resource base have almost doubled in size from the initial values.
Predictions of Ultimate Recoverable Reserves Numerous authors have attempted to predict the magnitude of the ultimate recoverable resource base. What is interesting about these estimates is that they all essentially identify those crude oil reserves that will be produced over the next 40–50 years. In other words, every assessment tends to slice lower through the resource pyramid (Figure 2). Estimates of the ultimate resource vary significantly depending upon the relative optimism or pessimism of the subsurface
geotechnical assessment, as well as the definition of what is the minimum size of economic importance. Recent estimates of the ultimate recoverable reserves of oil (Figure 4) vary between 1.5 and 3.8 trillion barrels. It is interesting to note that, despite the range of estimates, the mean estimate has remained stable at 2 trillion barrels for at least the past 20 years, suggesting that technology has reached an asymptote. Of course, not all of the estimates are based on the same set of data, as different estimates slice the resource pyramid at different levels. The most pessimistic estimate only includes conventional oil reserves and excludes all other oil types (for example, heavy oil, tar sands, secondary recovery). In contrast, the optimistic estimate by the United States Geological Survey (USGS) only excludes from its estimates oil from tar sands, oil from polar regions, and oil shales. However, nearly all estimates rely on existing technology and do not attempt to predict the impact of future technological developments. As a result, the estimates can only be regarded as static estimates at a fixed point in time under a particular set of technological and economic circumstances. The actual volume recovered will depend on a number of factors, not least whether the reserves are economic to extract. Political, economic, and technological constraints all play a role in deferring or accelerating production. In addition, there is interplay with other sorts of fuel, which is likely to have a significant impact on determining the rate of extraction. Historical evidence from producing fields highlights the fact that reserve estimates are dynamic as circumstances change and knowledge and technology progress. An understanding of the potential importance of the growth of reserves can be gained from considering the current estimate of the global oil resource base and the current average recovery factor. Using consensus estimates that the global average
PETROLEUM GEOLOGY/Reserves 333
Figure 3 Creaming curves for the UK sector of the North Sea showing the impact of reserves creep on the older fields. The upper curve includes all fields that had been discovered and were producing by 1998. The symbols represent estimates of the same 25 fields made at different times (1980, 1985, 1990, and 1998). The figure shows the change in estimated size of these discoveries. The full line shows the addition of a further 100 discoveries made up to 1998 and which are currently in production.
Figure 4 Published estimates of world ultimate oil recovery (trillions of barrels). This chart shows the various published estimates of ultimate oil recovery through time. Earliest estimates made in the 1940s suggested an ultimate recovery of around 0.5 trillion barrels. However, the average estimate since Halbouty (1981) has been around 2.0 trillion barrels. USGS, United States Geological Survey.
oil recovery factor is around 35% (Figure 5), and that the current resource base of reserves is around 2 trillion barrels, original oil in place would be nearly 6 trillion barrels. Therefore, every 1% increase in global oil recovery would lead to an additional 55 billion barrels of oil being produced. Moving the average global recovery factor up to 45%, a not
unrealistic target, could increase global recoverable reserves by around 550 billion barrels. Over the past 50 years, estimates of global recoverable oil reserves have risen annually, except for two small falls in the late 1970s, despite annual increases in oil production (Figure 6). This has led to a great deal of attention being paid to reserve reporting by
334 PETROLEUM GEOLOGY/Reserves
Figure 5 Distribution of recovery factors in all currently producing oil fields.
Figure 6 Global estimated reserves and reserves to produc tion (R/P) ratios by year. OPEC, Organization of Petroleum Exporting Countries. R/P, reserves to production ratio.
companies and countries. Individual Organization of Petroleum Exporting Countries (OPEC) members have been accused of inflating reports of reserves because export quotas were originally based on estimates of recoverable reserves. The huge reserve revisions reported by many OPEC companies in 1987 have been used by some to undermine confidence in published numbers. In reality, it is possible that previous OPEC estimates were too conservative, having been inherited from private companies before nationalization. Although the absolute level of reserves has risen only slightly over the past 5 years, within that total, upward revisions to existing fields and discoveries have exceeded 150 billion barrels, 30% more than actual consumption. The USGS recently published a mean estimate of 612 billion barrels from increased recovery, significantly increasing their estimate of the world’s ultimate recoverable reserves. In the UK sector of the North Sea, reserves have grown by an average of 35% from the time of field approval (Annex B) to the present day or abandonment (Figure 7).
Estimates of the total resource base remaining depend not just on new discoveries being made and brought on stream, but also on the amount of reserves used due to depletion of existing fields. Over 70% of current world oil supply comes from oil fields that were discovered prior to 1970, and 14 of these fields produce over 20% of the world’s total supply. Even though data for many of these fields are very hard to come by, the decline rates for these fields are likely to become of increasing importance in the future. However, the effect of depletion is often overstated as forecasters often assume a depletion rate of 10–20%, and no further investment. As an example, estimates of non-OPEC production have always tended to indicate that production is at or near its peak, and yet overestimation of depletion rates has led to successive estimates being proven to be incorrect (Figure 8).
The Peak Oil and Depletion Debate Ever since oil was first produced in significant quantities over 140 years ago, debate has taken place on whether petroleum will run out within the foreseeable future. Historically, near-term supply concerns arise when the relative rate of production capacity growth falls short of expected rates of demand growth. Through time, however, changing attitudes to oil supply and demand have led to varying perceptions of whether the resource is in short supply or not. It is also noticeable that changes in perception can take place over a very short period of time depending on a particular event on either the supply side or the demand side. The discovery of super giant oil fields in Texas in the 1930s gave rise to perceptions of a huge glut in oil. This perception was sustained by the discovery of major reserves in the Middle East. However, by the early 1970s, there was a clear paradigm
PETROLEUM GEOLOGY/Reserves 335
Figure 7 Growth of reserves in UK oil fields. This chart shows the change in reserve estimates of UK oil fields with time, since their ‘Annex B’ approval. Some field estimates have decreased, but most have increased very substantially. The average of all the fields (full line) has grown by 35% over 20 years.
Figure 8 Forecast of non Organization of Petroleum Exporting Countries (OPEC) Third World oil production over time made by the United States Department of Energy. The chart shows that estimates of non OPEC production have consistently proved to be conservative through time. For example, the forecast made in 1990 for the year 2000 is now some 4 million barrels per day lower than actual production. MB/D, million barrels per day.
shift as part of a trend towards pessimism about resource availability as studies began to emerge that oil production outside the Middle East had peaked. This was triggered when US production levels started to fall from peak levels by the end of the 1980s. In the 1990s, the introduction of new development techniques to boost recovery rates, coupled with
the revolution in computers which transformed seismic acquisition and processing, the technical innovations that enabled deep-water development, and the opening up of new petroleum provinces that were previously closed due to political constraints, all led to a period of optimism that weakened concerns about the exhaustibility of reserves. More recently, these
336 PETROLEUM GEOLOGY/Reserves
concerns have been raised again as the oil industry worldwide struggles to replace its reserves and achieve targeted rates of production growth. Much of the published literature on the depletion debate falls into two camps. On the one hand, there are those who warn of the imminent danger of a collapse in oil supply and the economic consequences of that occurring. On the other hand, there are those who argue to varying degrees that there is no limit to oil supply in the near future. What is common to both camps is the extensive reference to the work of King Hubbert. Hubbert gained prominence in the American geological community because he anticipated the peak and subsequent fall of US oil and gas production. At the heart of his prediction was the assumption that all resources are finite, and that eventually all resources will be depleted and exhausted over a period of time that is determined by the rate of production. Hubbert argued that the complete cycle of exploitation must have the following characteristics. Beginning at zero, the rate of production tends initially to increase exponentially. Then, as difficulties of discovery and extraction increase, the production rate slows in its growth, passes one or more maxima and, as the resource is progressively depleted, declines eventually back to zero. He argued that this cycle of production would be bell shaped when plotted against time. The approximate rule here is that peak production will
correspond to the mid-point of depletion and that this usually occurs about 20–25 years after the discovery of the mid-point. Although some production curves do have a shape that approximates such a ‘normal’ distribution, there is no inherent reason why production should follow such a pattern, and some production curves show very strong asymmetry. For example, when production is dominated by a small number of large fields, peak output tends to precede the depletion mid-point. Where the discovery pattern is more dispersed, or where offshore fields are significant, the production peak usually comes after the mid-point. Hubbert’s basic assumption was that, if known past and prospective rates of production are combined with a reasonable estimate of the amount of fuel initially present, one can calculate the probable length of time that the fuel can be exploited. In other words, the area under the complete production curve would be equal to the size of the resource. However, estimating the amount of oil and gas that will ultimately be discovered and produced in a given area is full of uncertainty, as we have seen. The Hubbert curve might have fitted US production, but, on a global scale, the modelled curve no longer fits the historical data (Figure 9). Up until the mid-1970s, the modelled production matched the actual production, but, after the supply shocks, which led to sharp oil price rises, annual increases in demand for oil fell from around 7% to just 2% per annum.
Figure 9 World production and the Hubbert peak. This chart shows a theoretical Hubbert curve based on production up to the oil price shocks of the 1970s. The resulting high oil prices led to a decrease in demand for oil. As a result, the slope of the oil supply curve departs from the predicted Hubbert curve after 1973. Assuming oil demand growth of 2% from 2003, and an ultimate reserve base of 2.25 trillion barrels, the peak in oil production could be reached in 2018. The chart highlights the impact that changes in demand can have on predictions of when global oil production will peak. mmbbl, billion barrels; mmbbpd, million barrels per day.
PETROLEUM GEOLOGY/Reserves 337
Models of Resource Depletion There are, in essence, two main models that have been applied to the estimations of the volume of remaining reserves and the rate of depletion. On the one hand, there is the model for an open market and, on the other, the model for a closed market. The closed market model, in essence, treats the global oil reserves as a finite or ‘fixed stock’, whereby market depletion would ultimately result in higher prices. In the open market model, the cost of new energy sources is forecast to fall because of technological innovation, economies of scale, and the development of efficient systems of transportation. With lower costs, new energy sources become competitive with existing energy sources and ultimately lead to their replacement. The open market model could ultimately lead to lower energy costs through time through the potential substitution of fuels, in particular natural gas. The open market model has stronger affinities to actual market examples. The US energy market is an open market, which has benefited from a long-term reduction in energy prices despite rising demand. Coal substituted for firewood; petroleum substituted for whale oil for illumination purposes, and eventually coal for transportation; gas substituted for coal in domestic heating; and gas and coal are currently competing in the electrical generation sector (Figure 10).
The End of the Petroleum Era? With any finite resource, there will come a time when unfavourable economics will not permit further extraction. Worries concerning the long-term supply of petroleum began shortly after production started in 1859. The huge growth in energy markets after World War II also led to renewed fears that fossil fuels would be exhausted. In the 1970s, many people argued that the growth in world population and industrial production would lead to the total depletion of oil within
50 years of the forecast. The fact that the deadlines for these scenarios have now passed, and oil production and reserves continue to rise, shows that these forecasts were all overly pessimistic. The resource exhaustion spectre was initially raised largely on the basis of a comparison of the current year’s oil production with the so-called proven reserves of oil. However, this type of evaluation is inadequate as proven reserves represent nothing but the working stock of the oil industry. For example, the ‘optimum’ reserves life of 10 years is merely an expression of confidence from private and state oil companies as to where oil will come from in the future in order to fund investment in new infrastructure. Therefore, the current reserves life of oil is deemed to be irrelevant, especially in the context of a global reserves life that has risen to around 40 years from 1940 to 1975 despite the significant increase in global production. The debate has now moved on from reserves life to a focus on the point in time when global production starts to decline. Supporters of this ‘peak oil’ school of thought forecast serious economic and social consequences in the aftermath of this occurring. They highlight the fact that most of the world’s major oil discoveries were made during the 1960s and 1970s and that, since that time, the discovery rate has declined (Figure 11). Using examples from various basins around the world, not least the USA, it is argued that peak oil production usually occurs around 25–30 years after peak exploration success. Using this as a proxy, it is predicted that oil production will peak within the next 10 years and, given that there are no near-term substitutes to oil, an economic crisis will ensue.
The Economic Viewpoint One key assumption made by the supporters of ‘peak oil’ is that the resource base is finite. However, the amount of oil that can be recovered depends not only
Figure 10 United States primary energy consumption by fuel, 1860 2000.
338 PETROLEUM GEOLOGY/Reserves
Figure 11 Discovery volumes and world oil production, 1900 2000. mmbbl, billion barrels.
Figure 12 The effect of changing the resource base on the timing of peak oil. bnbbl, billion barrels; mmbopd, million barrels oil per day.
on the total resource base of oil, but also on dynamic variables, such as price, infrastructure, and technology. The level of peak production will also increase if recoverable reserves rise and, as a consequence, the timing of peak oil will be pushed further into the future (Figure 12). Economists argue that oil reserves are not a ‘fixed stock’ and that energy has not been getting scarcer in basic economic terms, but rather has been getting more plentiful. The fall in the long run cost of oil production has been used in support of this argument. It is also argued that there is no such thing as an exhaustible natural resource and that the
total petroleum reserves in the Earth are an irrelevant non-binding constraint. If oil becomes uneconomic to produce, the industry will disappear and whatever is left in the ground will remain unknown: a geological fact of no economic interest. On the demand side, it has been proven wrong to project current rates of growth well into the future. For example, in the late 1970s, the history of nearly 30 years of an average 7% per annum rate of growth in oil demand still played a powerful role in determining attitudes to the longevity of oil resources. This was at a time when growth rates had already fallen to around 1% per annum and when most
PETROLEUM GEOLOGY/Reserves 339
sufficiently to enable energy substitution. Volatility in the price of oil will sharpen the debate and could actually cause near-term unjustified changes in investment patterns by industry and governments alike.
See Also Geological Surveys. Petroleum Geology: Overview; Chemical and Physical Properties; The Petroleum System; Exploration; Production. Seismic Surveys.
Further Reading Figure 13 Conventional oil supply model showing the timing of peak oil with four rates of oil demand growth. R/P, reserves to production ratio; USGS, United States Geological Survey.
economists were forecasting a maximum growth rate of 3%. Changes in forecast to oil demand therefore have a major impact on the timing of peak oil; a difference of just 3% in average annual demand growth for oil gives a 46 year range in the timing of peak oil (Figure 13). Similar arguments can be applied to changes in the level of investment and the rate of depletion.
The Real Debate The amount of fossil fuel on our planet is finite, certainly within the time-frame of civilization. However, does this necessarily mean that we will run out of petroleum reserves imminently or at all? Or will we develop alternative sources of energy long before the planet’s petroleum reserves have been extracted? The changing pattern of world oil supply has meant that many of the predictions that oil would run out by the turn of the twentieth century have been proven wrong. Fears of impending scarcity of non-renewable energy have all been proven to be groundless as important changes on both the supply and demand sides have occurred. The history of those who predict future trends in the depletion of petroleum reserves is one of persistent tension between those who assert that the best decisions are based on quantification and numbers, determined by the patterns of the past, and those who base their decisions on more subjective beliefs about the uncertain future. Ultimately, oil is a finite resource, but the real debate is not so much whether we are likely to significantly diminish the petroleum reserve base within the next few decades, but whether prices will fluctuate
Adelman MA (1990) Mineral depletion, with special reference to petroleum. The Review of Economics and Statistics LXXII(1): 1 10. BP (2003) BP Statistical Review of World Energy 2003. London: BP plc. Campbell CJ (1997) The Coming Oil Crisis. Brentwood: Multi Science Publishing Company and Petroconsul tants. Campbell CJ and Laherrere JH (1998) The end of cheap oil. Scientific American March: 80 86. Deffeyes KS (2001) Hubbert’s Peak: The Impending World Oil Shortage. Princeton, NJ: Princeton University Press. Hubbert MK (1956) Nuclear energy and the fossil fuels. Drilling and Production Practice: 7 25. Hubbert MK (1971) The energy resources of the Earth. Scientific American February: 31 40. Laherrere JH (1999) World oil supply what goes up must come down, but when will it peak? Oil and Gas Journal 97: 57 64. Lynch MC (1996) The analysis and forecasting of pe troleum supply: sources of error and bias. In: El Mallakh DH (ed.) Energy Watchers VII, pp. 51 71. Colorado: International Research Center for Energy and Economic Development. Lynch MC (2001) Forecasting oil supply: theory and practice. Quarterly Review of Economics and Finance July: 1 28. McCabe PJ (1998) Energy resources cornucopia or empty barrel? American Association of Petroleum Geologists Bulletin 82(11): 2110 2134. Odell PR (2001) Oil and Gas: Crises and Contro versies 1961 2000. Brentwood: Multi Science Publishing Company Ltd. Simmons MR (2001) The World’s Giant Oilfields. Energy Service Industry Research. Houston: Simmons and Com pany. Smith AJ and Lidsky BJ (1993) King Hubbert’s analysis revisited: update of the lower 48 oil and gas resource base. The Leading Edge 12: 1082 1086. Weeks LG (1958) Fuel reserves of the future. American Association of Petroleum Geologists Bulletin 42: 431 438.
340 PLATE TECTONICS
PLATE TECTONICS R C Searle, University of Durham, Durham, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction In the theory of plate tectonics, it is assumed that the Earth has a thin, brittle, outer layer, or ‘lithosphere’, which is broken up into a small number of tectonic ‘plates’. These plates are further assumed to be rigid and undeformable, but capable of independent motion relative to each other on the surface of the Earth. The different relative motions of the plates are considered to be responsible for much of the varying geological activity on the Earth. The plate motions are describable in simple geometric terms and, once they have been determined, usually from a limited number of observations; they can be used to predict the type, magnitude, and direction of relative motions across plate boundaries anywhere in the world, both present and past. The power of plate tectonics theory arises both from its unifying view of global tectonics and from its powerful predictive ability. The subject of plate tectonics, sensu lato, is sometimes divided into ‘plate kinematics’, which deals with plate motions and their geometry, and plate tectonics sensu stricto, which deals with the geological consequences of the motions.
Tectonic Plates, Lithosphere, and Asthenosphere The fundamental assumption underlying plate tectonics is that the outer part of the Earth consists of a small number of thin, rigid plates which are curved to fit the spherical shape of the Earth. The plates are considered to suffer no internal deformation, and only to deform and interact with each other along their boundaries, which are considered to be very narrow compared with the plate dimensions. This view arose in the 1960s, largely as a result of the newly observed pattern of global seismicity provided by the World Wide Standard Seismograph Network. This showed that, over much of the world, earthquakes occur in narrow bands no more than a few tens of kilometres wide, whilst the large areas between these bands are essentially aseismic (Figure 1). The assumption was made that the aseismic areas are largely devoid of deformation and behave rigidly, and the bands of seismicity mark the plate boundaries. In simple plate tectonics theory, the plate boundaries are considered to have negligible width, although in
practice they are associated with narrow but finite bands of geological activity (see History of Geology Since 1962). Early workers on plate tectonics theory divided up the Earth’s surface into about 12 major plates (Figure 2), although since then considerable numbers of smaller ‘microplates’ have been recognized. The rigid outer layer of the Earth, which comprises the plates, is known as the ‘lithosphere’, from the Greek ‘lithos’ meaning stone. Below this is a weaker layer, the ‘asthenosphere’ (Greek ‘asthenia’, weakness), which deforms relatively easily and allows the plates to slide around. It is important to understand that these are terms describing the mechanical behaviour of the Earth and not its composition. In particular, the lithosphere does not normally correspond to the crust, but comprises both the crust and the uppermost part of the upper mantle, which usually behave together in a brittle manner (see Earth: Mantle; Crust). The lithosphere ranges in thickness from a few kilometres at mid-ocean ridge axes (where it may be entirely crust) to around a hundred kilometres in old ocean basins, and perhaps several hundred kilometres under continental cratons (where it is predominantly mantle). The asthenosphere constitutes a relatively weak layer of mantle below the lithosphere, capable of plastic deformation. The thickness of the asthenosphere is less well defined than that of the lithosphere, but may range from around 50 km in some places to several hundred kilometres, and may even encompass the whole upper mantle. Furthermore, except in the case of some subduction zones (see below), tectonic plate boundaries do not follow the boundaries between continents and oceans. An individual plate typically comprises regions of both continental and oceanic crust. For example, the South American plate consists of the continent of South America plus the western half of the South Atlantic Ocean (Figure 2). The different mechanical behaviours of the lithosphere and asthenosphere are explained by the different deformation mechanisms that occur at different temperatures and pressures. At relatively low temperatures, most rocks deform elastically and suffer brittle failure (faulting) when their elastic limit or ‘strength’ is exceeded. This strength increases with pressure and hence depth (Figure 3). However, as the temperature increases, solid-state creep becomes increasingly important. Deformation by this mechanism becomes easier at higher temperatures, and the rock weakens exponentially (Figure 3). The depth at which brittle deformation gives way to ductile creep
PLATE TECTONICS 341
Figure 1 Distribution of shallow (0 33 km, red), intermediate (33 70 km, yellow; 70 300 km, green), and deep (300 700 km, blue) earthquakes of magnitude 4.0 or greater recorded between 1990 and 1996. Reproduced from D. Sawyer, http://www.geophysics. rice.edu/plateboundary/intro.html.
depends mainly on the composition of the rock, the rate of strain, and the geothermal gradient. Laboratory measurements of the mechanical properties of rocks and minerals predict depths of the brittle– plastic transition in the range suggested for the base of the lithosphere by observations of the Earth. Evidence for the existence of the lithosphere and asthenosphere comes from a variety of sources. Seismic waves generated by earthquakes, particularly shear or S waves, show very little attenuation of their energy whilst travelling through the lithosphere, but are attenuated somewhat more in the asthenosphere (see Tectonics: Earthquakes). The velocities of these waves through the Earth show a marked decrease in the region of the asthenosphere. The effective viscosity of the Earth’s mantle, as inferred from, for example, the rate of uplift of continental areas following the retreat of glaciers, shows a decrease in viscosity in the asthenosphere.
In general, crustal rocks are weaker than mantle, and, because oceanic crust is much thinner (around 6 km) than continental crust (approximately 35 km), oceanic lithosphere has a much greater component of mantle rocks and is correspondingly stronger. For this reason, oceanic areas tend to behave in a more rigid way than continents. In particular, continental mountain belts often display deformation over quite broad zones compared with oceanic plate boundaries (Figure 1).
Plate Tectonics Although plate boundaries are primarily defined by the earthquakes along their boundaries, they are also marked by distinctive morphological features, narrow bands of tectonic activity, and, especially at transform boundaries (see below), by individual major faults. The active plate boundary may be only
342 PLATE TECTONICS
Figure 2 Major plates of the world and their boundaries. Reproduced from D. Sawyer, http://www.geophysics.rice.edu/platebound ary/intro.html, after D. Mueller, University of Sydney, Sydney, Australia.
a few kilometres wide, and rarely more than a few tens of kilometres. Three types of plate boundary are recognized, corresponding to the three types of relative motion (Figure 4): ‘divergent’ (also ‘ridge’ or ‘accretionary’) plate boundaries, ‘transform’ (or conservative) plate boundaries, and ‘convergent’ (‘subduction’ or destructive) plate boundaries. In general, any type of plate boundary can join any other type. Divergent (Ridge) Boundaries
Divergent boundaries occur only at mid-ocean ridges, and are the sites of creation of new lithosphere by seafloor spreading (see Tectonics: Mid-Ocean Ridges). The Mid-Atlantic Ridge and East Pacific Rise are typical examples. These boundaries are characterized by extensional tectonics dominated by normal faulting and extensional earthquake focal mechanisms. Generally, a divergent plate boundary is approximately orthogonal to the plate separation
or spreading direction, but this is not essential, and ridges with greater or lesser extents of obliquity are fairly common. The mid-ocean ridges associated with these plate boundaries arise because new plate is created at the boundary; thus, the plate gets older with increasing distance from the plate boundary and, as it does so, it cools, becomes denser, and thermally contracts. The plate boundary is thus the youngest, hottest, and shallowest region, forming the crest of the mid-ocean ridge. The detailed morphology of the plate boundary appears to depend on the temperature of the underlying mantle, which is mainly controlled by the spreading rate. For ‘slow spreading ridges’ (with a plate separation rate of less than about 60 km per million years), there is usually a ‘median valley’, a few kilometres deep and a few tens of kilometres wide, formed by rifting (Figure 5). At ‘fast spreading ridges’ (separation faster than 70 km per million years), the plate is too hot and weak to support rifting, and instead there is a volcanically built axial high.
PLATE TECTONICS 343
Ridges are formed by melting of dry mantle at depths of 60 km, producing basaltic magma. Basaltic volcanism is widespread at mid-ocean ridge plate boundaries. Plate tectonics theory allows for asymmetrical spreading (one plate accreting faster than the other). However, although temporary asymmetrical spreading is common, the net effect averaged over millions of years is usually approximately symmetrical. The precise reasons for this are not fully understood. Transform Boundaries
Figure 3 Calculated strength envelopes for material subject to brittle, semi brittle, and plastic deformation with increasing depth. Reproduced from Searle RC and Escartı´ n J (2004) The rheology and morphology of oceanic lithosphere and mid ocean ridges. In: German C, Lin J, and Parson L (eds.) Thermal Regime of Ocean Ridges and Dynamics of Hydrothermal Circulation. American Geophysical Union, Geophysical Monograph, Fig. 8. Washington, DC: American Geophysical Union.
Transform boundaries are those in which plates slide past one another with essentially no convergence or divergence; they thus conserve the areas of the adjacent plates. Because of this, they have the important property that they are exactly parallel to the direction of relative plate motion, and can be used for its estimation. Transform boundaries are characterized by strike-slip faulting and earthquakes with strike-slip mechanisms. A typical example of a transform boundary is the San Andreas Fault Zone in California, USA (Figure 6). In simple plate tectonics theory, a transform boundary consists of a single fault called a ‘transform fault’, although, in practice, there is often a zone of such faulting some tens of kilometres wide. Because there is no large-scale convergence or divergence at these boundaries, there is also relatively little vertical relief,
Figure 4 Diagram illustrating the three principal types of plate boundary (ridge, transform, and subduction), their general morph ology, and the distribution of earthquakes. Large arrows show relative plate motion. Circles with double arrows indicate the general distribution of earthquakes. Modified from Davidson JP, Reed W, and Davis PM (2002) Exploring Earth, Fig. 8.13. Upper Saddle River, NJ: Prentice Hall.
344 PLATE TECTONICS
Figure 5 Oblique view of the topography of the Mid Atlantic Ridge Median Valley near 24 N, based on multibeam echosoun der measurements. The view is towards the south, with the cross cutting Kane Transform Fault in the distance. The width of the distant view is approximately 50 km, width in the foreground about 20 km, and distance from the foreground to the horizon about 100 km. Depths less than 2500 m, red; greater than 4200 m, blue. The median valley is approximately 30 km wide. There is evi dence of fault scarps on the valley walls. The ridge in the centre foreground is a linear volcano sitting above the plate boundary and representing the current axis of plate accretion. Image produced by M. Jones, University of Durham, Durham, UK, using data from Purdy GM, Sempere JC, Schouten H, Dubois DL, and Goldsmith R. (1990) Bathymetry of the Mid Atlantic Ridge, 24 N 31 N Amap series. Marine Geophysical Researches 12: 247 252; and Pockalny RA, et al. (1988) Journal of Geophysical Research 93: 3179 3193.
although small valleys and scarps usually mark the trace of the transform fault itself. Transforms link other plate boundaries, and can thus ‘transform’ one type of tectonic boundary (e.g., extensional) to another (e.g., compressional), hence the name. Many transform faults offset mid-ocean ridges, producing (in map view) a staircase pattern of alternating ridge and transform boundaries (Figure 2). Such offsets range from a few tens of kilometres to over a thousand kilometres in length. Transform boundaries occur in both continental and oceanic lithosphere. Convergent Boundaries
Convergent plate boundaries in the strict plate tectonics sense occur only at the deep-sea trenches and related ‘subduction zones’ (see Tectonics: Convergent Plate Boundaries and Accretionary Wedges). Areas of active mountain building on continents (e.g., the Alpine– Himalayan zone) are zones of plate convergence, but are characterized by broad and almost continuous
zones of deformation. This is partly because continental lithosphere is weaker and more easily deformed than oceanic lithosphere, and partly because it is less dense and so is not readily removed by subduction (see below). Plate tectonics does not provide a very useful description of such broad continental convergence zones. Where one of the converging plates consists of oceanic lithosphere, it will be overridden by the other and pushed down into the asthenosphere in a process known as subduction (Figure 7). This can occur because oceanic lithosphere (unlike continental lithosphere) has a similar density to the underlying asthenosphere. Where both plates are oceanic, either one may be subducted under the other, and sometimes, depending on local conditions, the ‘polarity’ of subduction may reverse (i.e., the subducted plate breaks off and subsequently becomes the overriding one, whilst that which was overriding begins to subduct). The plate boundary at a subduction zone is marked by a deep trench, the deepest and probably most famous of which is the Mariana Trench in the western Pacific, whose base is over 11 km below sea-level (see Tectonics: Ocean Trenches). As the slab subducts, it is deformed and generates earthquakes. These occur in a narrow band, known as the Wadati–Benioff zone, that follows the position of the subducting slab, and constitutes some of the best evidence for the existence of subduction zones. Subduction zones are very complex, and contain local regions of both compressional deformation (mainly at the actual plate boundary and in the leading edge of the overriding plate) and extensional deformation (mainly where the subducting plate bends to begin its descent). Earthquake mechanisms reflect this, and also show variation in mechanism with depth in the subducting plate. Some of the world’s largest earthquakes are associated with subduction zones. As the slab descends, it heats up and gives off water trapped in the crustal rocks. The presence of this water lowers the melting point of the surrounding mantle and, at a depth of some 100 km, magma is generated that rises through the overriding plate to create a volcanic arc, a hundred or so kilometres behind the trench (depending on the angle at which the slab is descending). As with ridges, oblique subduction is allowed by plate tectonics theory, and is relatively common.
Plate Kinematics Rotation Poles
The great strength of plate kinematics is that it can describe plate motions in terms of simple geometry,
PLATE TECTONICS 345
Figure 6 Map of major faults (red) and movement measured over 1 year by a Global Positioning System (GPS) (arrows) over part of the San Andreas Fault Zone in California, USA. Reproduced from Davidson JP, Reed W, and Davis PM (2002) Exploring Earth, Fig. 8.24. Upper Saddle River, NJ: Prentice Hall.
Figure 7 Diagram of a typical ocean ocean subduction zone. Reproduced from Davidson JP, Reed W, and Davis PM (2002) Exploring Earth, Fig. 10.3a. Upper Saddle River, NJ: Prentice Hall.
and hence make precise predictions of relative motions anywhere on the globe. At the heart of this geometry is a concept called Euler’s theorem, which states that any displacement of a rigid body on the surface of a sphere can be described in terms of a
single rotation about a specified axis (Figure 8). Such axes cut the Earth’s surface at pairs of points called ‘rotation poles’ (or ‘Euler poles’). Once the poles and the angular rotations are specified, the whole motion is completely determined. Thus, the
346 PLATE TECTONICS
Like the geographic poles, we can imagine the Euler poles as centres of coordinate grids (Figure 8). ‘Small circles’ centred on the poles are the equivalent of lines of latitude, and plate relative motions are everywhere parallel to these small circles. Along any given small circle, the angular separation rate is constant, and increases with the distance from the rotation pole to the circle. ‘Great circles’ represent the shortest distance between two points on the surface of a sphere. Great circles that pass through the pole of rotation are equivalent to lines of longitude in the geographic system, and cut the small circles at right angles. Measuring Plate Motions
Figure 8 A Euler rotation pole describing the motion between Block 1 and Block 2. Plate boundary shown by double lines (mid ocean ridge) and bold single lines (transform faults). Circles (medium lines through transforms) and great circles drawn orth ogonal to them and passing through the pole are analogous to latitude and longitude in the geographic coordinate system. Light lines show the geographic coordinate system for comparison. Reproduced from Lowrie W (1997) Fundamentals of Geophysics, Fig. 6.30. Cambridge: Cambridge University Press.
motion of a given plate is specified in terms of its Euler pole and a corresponding angular rotation rate. In practice, determining the motions of individual plates relative to some common reference frame is not easy, but determining ‘relative’ motions between pairs of plates is quite straightforward. These relative motions can also be described by rotation poles, although then the pole position must be specified relative to one or other of the pair of plates. Such relative rotation poles are the basis of most descriptions of plate kinematics. As Euler poles define the ‘directions’ of the rotation axes, and the angles or rates of rotation define their ‘magnitudes’, rotations can also be described as vectors. Vector and matrix algebra can then be used to calculate plate motions, greatly simplifying and speeding up such calculations, which can be performed easily on computers. It is important to distinguish between so-called ‘instantaneous poles’, which describe motion at an instant only, and ‘finite rotation poles’, which may describe the net result of motion over long periods of time. In descriptions of current plate motions, ‘instantaneous’ is usually taken to be about the past 1–3 million years. Motions over longer periods can be approximated by successions of so-called ‘stage poles’, each of which may describe the motion over a period of a few million years.
Plates move at average rates of a few tens of millimetres per year. Thus, relatively indirect methods have usually been used to determine plate motions, although in recent years various geodetic measurements have been developed that are sufficiently precise to provide direct measurements of plate motions. The most common way of determining plate separation rates is to use the linear magnetic anomalies (Vine–Matthews anomalies) produced during seafloor spreading to determine plate ages. These anomalies are produced as the Earth’s magnetic field episodically reverses its direction, and the reversing field direction is recorded in the basalts of the oceanic crust as they cool after eruption at the mid-ocean ridge axis. This varying rock magnetization causes small variations, or ‘anomalies’, in the magnetic field above the seafloor, and these are readily observed by towing a magnetometer behind a ship or low-flying aircraft. Because the reversal process is irregular, the resultant magnetic anomalies have a characteristic pattern, which has been calibrated against crustal age through radiometric dating of samples of the magnetized rock. If the pattern of reversals can be recognized, the crust that is the source of the anomalies can thus be dated. The magnetic anomalies mark isochrons of crustal creation; therefore, by measuring the distance between a recent magnetic isochron on one plate and its conjugate on the other, the ‘instantaneous’ divergence rate, and hence the rotation rate, can be determined. (The commonly quoted ‘spreading rate’ is the rate at which a single plate accretes; for symmetrical spreading, it is half the divergence rate.) As the spreading or divergence rate varies with the distance from the Euler pole, in principle, the distance to the pole can be determined by measuring the spreading rate at several places along the plate boundary. The best measure of the direction of relative plate motion is the azimuth of transform faults which, as stated above, are exactly parallel to the relative motion direction. Transform faults thus follow small
PLATE TECTONICS 347
Figure 9 Map of plate boundaries showing the relative velocities across them from the data of DeMets C, Gordon RG, Argus DF, and Steins (1990). Reproduced from Lowrie W (1997) Fundamentals of Geophysics, Fig. 1.11. Cambridge: Cambridge University Press.
circles about Euler poles. A great circle at right angles to a small circle passes through the pole. Thus, if the azimuths of several transforms along a plate boundary are determined, great circles can be constructed normal to them, and should intersect at the Euler pole (Figure 8). Transform faults are readily recognized by their morphology (for example, a narrow linear valley for ridge–ridge transforms along mid-ocean ridges). In practice, both spreading rates and transform azimuths are used together to solve for pole positions, sometimes supplemented by other data, such as earthquake focal mechanisms, to estimate relative motion directions. Euler poles may be calculated for individual plate pairs, for groups of plates, or for the global plate system (Figure 9). The most recent determination of global plate motions was performed by C. DeMets and others from Northwestern University, Illinois, USA, and provides a remarkably precise and self-consistent description of these motions (see ‘Further Reading’). As stated above, an important attribute of plate kinematics is its ability to predict plate motions. An interesting example of this is the possibility of determining convergence rates across subduction zones. Even at ocean–ocean subduction zones, one plate is destroyed, together with the record of magnetic lineations carried on it. Thus, there was no direct way of measuring such motion until the recent development of sufficiently precise geodetic methods. However, the relative motions of the plate pair can be determined by global fits as described above, and then the motion at
any point on the common plate boundary, including subduction zones, can be calculated from the Euler pole data. In recent years, geodetic methods have been developed to the level at which they can begin to measure plate motions directly. Where plate boundaries exist on land (such as the Mid-Atlantic Ridge in Iceland or the San Andreas Fault in California, USA), standard geodetic methods, such as electronic distance measurement, can be used at a local scale (over ranges of a few kilometres). On a slightly larger scale of tens to hundreds of kilometres, precise relative position determinations (to precisions of a few millimetres) can be made by careful use of the Global Positioning System satellite network. Relative positions between widely separated continents can be determined by Very Long Baseline Interferometry, in which the variation in phase of radio signals from distant quasars is used. Repeat measurements by these methods over times of a few years can now resolve plate motions, and give results that, in general, agree well with the more traditional determinations. The rotation rates of the major plates about their Euler poles range from about 2.1 per million years for the Cocos–Pacific pair, to about 0.1 per million years between Africa and Europe or Africa and Antarctica, and only 0.03 per million years for India–Arabia. Many of the minor plates (so-called ‘microplates’) rotate much faster than this, at tens of degrees per million years. In terms of linear rates, the fastest plate divergence rate at present is on the East Pacific Rise between the Pacific and Nazca plates, at
348 PLATE TECTONICS
Figure 10 Map of plate boundaries and absolute motion vectors. Arrow lengths are proportional to the plate speed, with the longest (in the western Pacific) corresponding to speeds of about 100 mm per year. Reproduced from Fowler CMR (1990) The Solid Earth: An Introduction to Global Geophysics, Fig. 2.20. Cambridge: Cambridge University Press.
about 160 km per million years (or 160 mm per year). At the slow end, the North America–Eurasia plate boundary passes very close to its Euler pole in northern Siberia, where the relative motion becomes essentially zero. Absolute Plate Motions
So far, we have dealt only with relative motions, which are fairly easy to determine. There is also interest in determining so-called ‘absolute’ plate motions, where the motions of all plates are related to some common reference frame. The possibility of doing this arises from the proposed existence of mantle plumes, which rise as narrow columns of relatively hot rock from deep in the mantle, possibly from the core– mantle boundary. They reach the Earth’s surface in so-called ‘hotspots’ where they are manifest by clusters of intense volcanic and seismic activity. Wellknown examples occur in Iceland and Hawaii, but there are thought to be many tens of such hotspots and associated plumes. Hotspots leave clear trails on the Earth’s surface, which comprise lines of volcanoes or volcanic seamounts and zones of thickened volcanically produced crust. The Hawaii–Emperor seamount chain, trending north-west from Hawaii, is an excellent example, but there are many other subparallel seamount trails in the south-western Pacific which are thought to have
resulted from plumes. If points along the hotspot trails are dated (e.g., by radiometric dating of volcanic products), the relative motion between the plumes and the plate, and between given plumes, can be determined. We can also calculate the motions of individual plates relative to the average plume motion. When this is performed, the relative motion between many of the hotspots turns out to be quite small, and significantly less than the average relative motions between plates. From this, and the fact that plumes are thought to rise through the mantle, it seems reasonable to assume that the average plume motion relative to the mantle is quite small, although it should be admitted that this has remained difficult to test in detail, and there are those who even question the existence of plumes. Nevertheless, if it is assumed that the average motion of hotspots relative to the mantle is zero, plate motions can then be given in the supposedly fixed mantle reference frame. These are referred to as absolute plate motions. They can also be described in terms of Euler poles, and are shown in Figure 10.
Mechanisms and Plate Driving Forces Plates as Parts of the Mantle Convection Cycle
Plate motions are ultimately driven by the Earth’s heat energy, and they are intimately related to the
PLATE TECTONICS 349
mantle convection that is driven by this heat. One view of plates is that they simply represent the surficial parts of mantle convection cells: as hot, ductile mantle rises to the surface, it cools and becomes brittle – a plate – and then moves as a rigid block over the surface before being subducted, gaining temperature and becoming ductile again. Recent results from seismic tomography suggest that, around the rim of the Pacific, sheets of cold material descend below subduction zones deep into the lower mantle, implying a strong coupling of mantle motion and subducted plates. However, the coupling is not perfect. There are some parts of the mid-ocean ridge (divergent plate boundaries) where it seems that the deeper mantle (below the asthenosphere) may be descending rather than rising. One such place is the so-called AustraloAntarctic Discordance south of Australia. Moreover, some plates, such as Africa, are almost entirely surrounded by ridges and have very few subduction zones on their boundaries. In such cases, a rigid coupling of plates to convection cells would imply the unusual scenario of upwelling along an expanding ring, with a downwelling column inside it. In fact, one of the advantages of plate tectonics is that it allows partial decoupling of plate motions from deeper mantle flow via the ductile asthenosphere. The Forces Acting on Plates
Another way to look at the problem of the driving mechanism is to consider the forces acting directly on the plates. There are many possible forces, but amongst the most important are ridge push, slab pull, trench suction, and mantle drag. Ridge push arises from the tendency of the plates on a mid-ocean ridge flank to slide down the slopes of the wedge of thermally expanded asthenosphere that lies beneath the ridge. Slab pull arises from the negative buoyancy of the subducted plate which tends to drag the rest of the plate down with it. Trench suction is an additional force tending to pull plates together at subduction zones as a result of local convection driven by the subduction. Mantle drag is the frictional force between the base of the plate and the underlying asthenosphere. It is possible to estimate some of these forces, at least approximately, and their relative importance can also be assessed by considering the observed stresses in plates and inferred absolute plate velocities. The latter is particularly instructive. Absolute velocities are largely independent of the total area of the plate, and so it is unlikely that mantle drag is an overall driving force, as was once thought (i.e., plates do not ride as passive passengers on top of mantle
convection cells). However, plate velocities are inversely correlated with the area of continental lithosphere, suggesting that large areas of continent act as a brake (perhaps because such lithosphere is very thick or because subcontinental asthenosphere is cold and rather viscous). The fastest plates are those, mainly in the Pacific, which have large lengths of subduction zone along their boundaries, implying that slab pull and/or trench suction are important driving forces, as also suggested by calculation. There is a modest correlation with the effective length of ridge on the plate, indicating that ridge push is a driving force, but less strong than trench pull; this also tends to be backed up by calculations. Observations of intraplate stresses are also consistent with these conclusions.
Tests of Plate Tectonics There have been numerous tests of plate tectonics theory. Its self-consistency, direct measurements of predicted plate motions, earthquake focal mechanisms, and distributions of earthquakes have all played their part in confirming the theory.
See Also Earth: Mantle; Crust. History of Geology Since 1962. Tectonics: Convergent Plate Boundaries and Accretionary Wedges; Earthquakes; Mid-Ocean Ridges; Mountain Building and Orogeny; Ocean Trenches.
Further Reading Cox A and Hart RB (1986) Plate Tectonics How It Works. Oxford: Blackwell Scientific Publications. DeMets C, Gordon RG, Argus DF, and Stein S (1990) Current Plate Motions. Geophysical Journal Inter national 101: 425 478. DeMets C, Gordon RG, Argus DF, and Stein S (1994) Effect of recent revisions to the geomagnetic reversal timescale on estimates of current plate motions. Geophysical Research Letters 21(20): 2191 2194. Isacks B, Oliver J, and Sykes LR (1968) Seismology and the new global tectonics. Journal of Geophysical Research 73: 5855 5900. Kearey P and Vine FJ (1996) Global Tectonics, 2nd edn. Oxford: Blackwell Science. Le Pichon X (1968) Sea floor spreading and continental drift. Journal of Geophysical Research 73: 3661 3697. McKenzie DP and Parker RL (1967) The North Pacific: An example of tectonics on a sphere. Nature 224: 125 133. Morgan WJ (1968) Rises, trenches, great faults and crustal blocks. Journal of Geophysical Research 73: 1959 1982. Wilson JT (1965) A new class of faults and their bearing on continental drift. Nature 207: 343 347.
350 PRECAMBRIAN/Overview
PRECAMBRIAN Contents Overview Eukaryote Fossils Prokaryote Fossils Vendian and Ediacaran
Overview L R M Cocks, The Natural History Museum, London, UK Copyright 2005, Natural History Museum. All Rights Reserved.
Introduction Since the Precambrian–Cambrian boundary can now be dated at 543 Ma ago, and since the Earth is estimated to have been formed at about 4500 Ma ago, it follows that the Precambrian represents about seven eighths or 88% of geological time. However, that fact was not at all obvious to early geologists who, 200 or so years ago, had no idea of the age of the Earth. Between about 1820 and 1845, Earth history was divided into various named systems, with the Cambrian the oldest, based on the successive assemblages of distinct fossils contained in the sedimentary rocks and on the Law of Superposition, which states that rocks are older than other rocks now above them (assuming that they have not been structurally overturned). Igneous rocks were either undated or dated relatively as younger than the sedimentary rocks through which they had been intruded. Since no unambiguous fossils were then known from rocks below the Cambrian, all such rocks were and are simply termed the Precambrian. It was not until the invention and progressive refinement of radioisotopic dating in the twentieth century that the true ages of both the Precambrian and the rocks above it began to be understood.
Divisions of the Precambrian Unlike the Phanerozoic (Cambrian–Holocene), there are no formal subdivisions of the Precambrian. However, the Precambrian is normally divided into two: the earlier Archaean and the younger Proterozoic. The Proterozoic has itself been divided into three,
and the Precambrian is thus divided as follows (the ages are obviously approximate): Neoproterozoic – 1000 Ma–543 Ma, Mesoproterozoic – 1600 Ma–1000 Ma, Palaeoproterozoic – 2500 Ma–1600 Ma, and Archaean – ca. 4000 Ma–2500 Ma No Earthly rocks are known that are older than about 4000 Ma, so there is no universally used name for the period between the formation of the Earth at about 4500 Ma and the formation of the oldest known rocks of the Archaean, although the Hadean is a term used by some. The Neoproterozoic is divided into the Riphean (1000–600 Ma) and the Vendian (600–543 Ma), and the Vendian is sometimes termed the Ediacaran.
The Precambrian in this Encyclopaedia A large number of articles in this encyclopaedia deal in various ways with aspects of the Precambrian, and thus this article is devised to help the reader to locate the appropriate entry through cross-referencing. Thus there follows a brief guide to the key aspects of the Precambrian, under the headings the origin of the Earth, major Precambrian outcrops, Precambrian sediments and climate, the origin of life and Precambrian fossils, Precambrian orogenies and Precambrian terranes and palaeogeography.
The Origin of the Earth Our galaxy and the universe appear to have been formed about ten thousand million years (10 Ga) ago, but our star – the Sun – and its Solar System do not seem to have been formed for a long time after that, at about 4.5 Ga. The article on Earth structure and origins (see Earth Structure and Origins) describes how this process is thought to have occurred. The oldest known individual minerals are found within
PRECAMBRIAN/Overview 351
zoned zircons from Western Australia, which give an age of nearly 4.4 Ga. However, the oldest known rocks are the Acasta Gneisses of Canada, which are about 4.0 Ga old; this age is closely followed in other areas, for example the 3.9 Ga Napier Complex of Antarctica. The oldest cratons so far identified form parts of the South African and West Australian shields (see Shields) and are dated at about 3.2 Ga.
Major Precambrian Outcrops Although slivers of Precambrian rocks are found in many terranes consisting mainly of younger rocks (see Terranes, Overview), the great majority of Precambrian outcrops are found in the old shield areas of North America (principally the Canadian Shield, North American Precambrian Continental Nucleus (see North America: Precambrian Continental Nucleus; Continental Interior), South America (largely underlying Brazil and the adjacent areas), Africa (several areas) (see Africa: Pan-African Orogeny), Northern Europe (the East European Craton (see Europe: East European Craton)), Siberia (the Angara Craton, Central Asia (see Asia: Central), and Russia (see Russia)), India (see Indian Subcontinent), Antarctica (see Antarctic), and Australia (see Australia: Proterozoic). Several of these shield areas are made up of more than one shield unit; for example, three individual shields can be identified within Australia, and another three are found within Antarctica. From late in the Proterozoic and for all of the Palaeozoic, the South American, African, Indian, Antarctic, and Australian shields were grouped together to form the supercontinent of Gondwana (see Gondwanaland and Gondwana).
Precambrian Sediments and Climate The Earth’s atmosphere in Archaean and preArchaean times was a reducing one, consisting largely of carbon dioxide, nitrogen, water vapour, and inert gases, with subsidiary amounts of hydrogen, methane, and ammonia. A critical event in Earth history was the change just before about 2.2 Ga in the atmosphere from reducing to oxidizing, termed the Great Oxidation Event. The sediments reflect that profound change; for example, a characteristic sediment of many Archaean and Early Palaeoproterozoic shields is the banded iron formation (see Sedimentary Rocks: Banded Iron Formations). This sediment type is first known from the Isua Supracrustal rocks of Greenland, dated at 3.8 Ga, and these chemically precipitated deposits, which contain up to 35% iron, reach their maximum distribution between 2.8 Ga and
2.5 Ga, the end of the Archaean. They are often associated with greywackes deposited by turbidity currents in ‘greenstone belts’, which are extensive sedimentary mudstones and coarser rocks (usually metamorphosed) with high percentages of iron and other minerals. Following the change in the atmosphere from reducing to oxidizing, there are no true banded iron formations, although comparable granular iron formations are known from a few areas, notably Lake Superior, Canada (dated at about 1.9 Ga), and, much later (Late Neoproterozoic), the Yukon and Namibia. Each grain of clastic sediment at the bottom of the oceans today has been through several (an estimated average of five) cycles of erosion and sedimentation during Earth history. Since there was no vegetation on the land and soils only developed progressively with time, Precambrian erosion rates were much higher than those of today. Nearly all Early Archaean sediments were volcanogenic, usually basic or intermediate in chemical composition, but today these account for only about 20% of sediments, most of the earlier ones having broken down into their individual mineral components. Since quartz is more resistant to weathering than most other minerals, the proportion of quartz in sedimentary rocks has progressively increased, reaching today’s figure of approximately 50%. Limestones became common for the first time in the Proterozoic and often contain algal stromatolites. These Precambrian limestones had a higher magnesium content than average modern limestones, which resulted in a higher proportion of dolostones. A form of coal termed schungite is found for the first time in rocks aged about 2.0 Ga, and, although now metamorphosed, it probably formed initially from local accumulations of blue-green algae. Chert is known from Archaean rocks onwards and occurs in the form of layers within banded iron formations and as nodules within dolomites; however, these cherts are of inorganic or secondary origin, and primary cherts, formed from siliceous organisms such as radiolaria, are not known until the Late Neoproterozoic. Precambrian climates have formed the topic of many research projects. As can be seen from the desert sediments that many Precambrian rocks contain, the climates were often very hot; however, at other times, glacigenic deposits were widespread. These Precambrian glaciations occurred in several phases, and some workers believe that the Earth was at one time completely covered in glaciers, the socalled ‘snowball Earth’, but that concept is controversial, although palaeomagnetic evidence indicates that some glacigenic rocks in Australia must have been close to the then equator.
352 PRECAMBRIAN/Overview
The Origin of Life and Precambrian Fossils It is not known when life originated here on Earth; however, as far as can be deduced, life originated only once. Prokaryotes, which include bacteria (see Biosediments and Biofilms), are apparently simpler than eukaryotes, since they lack a nucleus, and were probably the first organisms. Because early organisms were largely soft-bodied, the early fossil record is tantalizingly incomplete. As detailed in the article on Precambrian prokaryotes (see Precambrian: Prokaryote Fossils), the changes in overall carbon isotopes seen after 3.5 Ga indicate that biological systems were in place by that time. This was well before the Great Oxidation Event at about 2.2 Ga, and thus early life forms must have lived in a reducing environment. Laminated structures known as stromatolites have been described from rocks as old as 3.4 Ga in Australia and South Africa; however, it appears likely that the earliest stromatolites were inorganic in origin. In contrast, stromatolites deposited after about 3.0 Ga were apparently organic in origin. Although the range and diversity of these complex organisms peaked in the Middle Mesoproterozoic and Neoproterozoic, their descendants are still to be found living today, most famously at Shark Bay, Western Australia. The role of stromatolites and other early organisms in taking carbon dioxide out of the atmosphere and replacing it with free oxygen played a major part in the change of the atmosphere during the Great Oxidation Event at about 2.2 Ga. As the Precambrian progressed, life forms became more diverse (see Precambrian: Eukaryote Fossils), although they rarely exceeded 1 mm in size. A major evolutionary event took place in the later Proterozoic Vendian period (see Precambrian: Vendian and Ediacaran), when much larger macroscopic fossils, albeit still without hard parts, evolved; these are known as the Ediacara fauna. That fauna has been found on most of the larger Late Precambrian terranes, but apparently became extinct at or near the Precambrian– Cambrian boundary.
Precambrian Orogenies In the Archaean there were many orogenies, but in most cases their detailed ages and lateral extents have not yet been fully determined, and long periods of orogeny are often conflated under a single name. For example, various orogenies have been recognized as occurring during the formation and subsequent fusion of the three cratons that make up present-day Australia. However, the 2.6–2.4 Ga Sleafordian
Orogeny of Australia marked the assembly of the South Australia craton and spanned the Archaean– Palaeoproterozoic boundary (see Australia: Proterozoic). In central Canada the Trans-Hudson Orogeny extended from 2.2 Ga to 1.7 Ga, peaking at 1.9 Ga (see North America: Precambrian Continental Nucleus). Some workers correlate the 1.9–1.8 Ga Barramundi Orogeny of the North Australian craton with the Trans-Hudson Orogeny of North America, and the two together form a key element in the assembly of the Rodinia supercontinent (see below). In Antarctica there are at least three different Archaean– Palaeoproterozoic cratons that formed between 3 Ga and 1.6 Ga (see Antarctic). The best-known Precambrian orogeny in the Mesoproterozoic is the Grenvillian Orogeny (see Grenvillian Orogeny), which occurred between 1.3 Ga and 1.0 Ga and probably resulted in the assembly of the superterrane of Rodinia. The Grenvillian Orogeny is principally represented in eastern North America and consists of an earlier accretionary stage between 1.3 Ga and 1.2 Ga, widespread magmatism from 1.2 Ga to 1.1 Ga, and a continent–continent collision at about 1.0 Ga. The 1.3–1.1 Ga Albany Fraser Orogeny of the South and West Australian cratons has been interpreted as part of the Grenvillian Orogeny. During the final part of the Neoproterozoic, the Timanide Orogeny (see Europe: Timanides of Northern Russia) affected today’s northern Europe and resulted in the accretion of a substantial number of terranes to the old craton of Baltica. In addition, the Cadomian Orogeny affected most of the terranes now in southern Europe; this continued until about 530 Ma, in the Early Cambrian.
Precambrian Terranes and Palaeogeography It is not known exactly when continental and oceanic crust became differentiated on the Earth’s surface, but the process is thought to have started before the Archaean. The oldest known cratons are dated at about 3.2 Ga, and it is not until after the formation of these cratons that we can be sure that wide shallow continental-shelf seas actually existed. Although there are glimpses of early terranes, for example between 3.0 Ga and 1.6 Ga (Archaean–Palaeoproterozoic) several Antarctic cratons were stabilized, there is no clear image of global palaeogeography prior to the Late Mesoproterozoic. The existence of a Precambrian superterrane was suggested in the 1970s, as geologists noted a number of 1.3–1.0 Ga mountain belts now on different continents, and later the name Rodinia was adopted for
PRECAMBRIAN/Overview 353
Figure 1 Palaeogeographical reconstruction at 750 Ma, just after the initial breakup of the Rodinia superterrane. The dark green shading represents terranes with good palaeomagnetic data, in contrast to the light green from with palaeomagnetic data is poor or absent. The red areas are the tectonic belts. Modified by Trond Torsvik, Trondheim, from Torsvik (2003).
Figure 2 Palaeogeographical reconstruction at 550 Ma, near the end of the Neoproterozoic. New diagram by Trond Torsvik, Trondheim.
354 PRECAMBRIAN/Eukaryote Fossils
this terrane. There is now a progressively convincing scenario emerging of the slow assembly of the superterrane of Rodinia during the Mesoproterozoic between 1.3 Ga and 1.1 Ga. After relative stability for a long period, Rodinia started to break up at about 850 Ma. That breakup is documented by several events; for example, the 760 Ma Areyonga event in Australia may represent rifting between Australia and the new terrane of Laurentia. Similarly, Laurentia appears to have separated from Antarctica at about 750 Ma (Figure 1). However, South America, Africa, peninsular India, Australia, and Antarctica stayed together and, with further smaller accretions, remained together to form the superterrane of Gondwana in the Late Neoproterozoic, a process that was complete by the Early Cambrian. Between 580 Ma and 555 Ma there was massive plutonism in Scandinavia, and flood basalts in the Ukraine occurred at 576 Ma; these events may document the separation of Baltica from Laurentia. Gondwana remained a united superterrane until after the breakup of Pangaea in the Early Mesozoic. Figure 2 shows a probable palaeogeography for the Late Neoproterozoic. The progressive rifting between Laurentia (North America) on the one hand and Gondwana and Baltica on the other resulted in the opening and steady widening of the Iapetus Ocean between the two sets of terranes, an opening that continued until about the end of the Cambrian. In addition the opening of the Ran Ocean marked the division between Baltica and Gondwana, and the
Aegir Sea separated Baltica from Siberia. Most of the other side of the globe was occupied by the vast Panthalassic Ocean.
See Also Africa: Pan-African Orogeny. Antarctic. Asia: Central. Australia: Proterozoic. Biosediments and Biofilms. Earth Structure and Origins. Earth System Science. Europe: East European Craton; Timanides of Northern Russia. Gondwanaland and Gondwana. Grenvillian Orogeny. Indian Subcontinent. North America: Continental Interior. North America: Precambrian Continental Nucleus. Precambrian: Eukaryote Fossils; Prokaryote Fossils; Vendian and Ediacaran. Russia. Sedimentary Rocks: Banded Iron Formations. Shields. Terranes, Overview.
Further Reading Cocks LRM (ed.) (1981) The Evolving Earth. Cambridge: Cambridge University Press. Edwards K and Rosen BR (2000) From The Beginning. London: The Natural History Museum. Hancock PL and Skinner BJ (2000) Oxford Companion to the Earth. Oxford: Oxford University Press. Hartz EH and Torsvik TH (2002) Baltica upside down: a new plate tectonic model for Rodinia and the Iapetus Ocean. Geology 30: 255 258. Torsvik TH (2003) The Rodinia jigsaw puzzle. Science 300: 1379 1381. Windley BF (1995) The Evolving Continents, 3rd edn. New York: John Wiley.
Eukaryote Fossils S Xiao, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Eukaryotes, archaeobacteria, and eubacteria are the three domains of living organisms (Figure 1). Most life forms that we can see with the naked eye are multicellular eukaryotes, which include animals, fungi, plants, and seaweeds. Many eukaryotes (for example, dinoflagellates, ciliates, and amoebae) are, however, single-celled and microscopic. Eukaryotes are cytologically distinct from prokaryotes (archaeobacteria and eubacteria). A typical eukaryotic cell contains membrane-bound intracellular structures such as a nucleus, mitochondria, and, for
photosynthetic eukaryotes, chloroplasts. Eukaryotes are also distinctively characterized by DNA-associated histone, eukaryotic gene regulation, and tubulinand actin-based structures known as cytoskeletons, which help to maintain and manipulate the shape of the cell. Nuclei, mitochondria, chloroplasts, histone, and cytoskeletons are rarely preserved in the fossil record, making it a serious challenge to recognize fossil eukaryotes, particularly single-celled ones. It has been suggested that cell size may be a useful guide: eukaryotes tend to have larger cells than prokaryotes. The size range of eukaryotic cells, however, overlaps considerably with that of prokaryotic cells. Cell size alone therefore is only indicative, not conclusive, evidence of eukaryotic affinity. Morphological complexity provides more reliable evidence. Because of the supportive function of cytoskeletons,
PRECAMBRIAN/Eukaryote Fossils 355
Figure 1 Phylogenetic relationships among the three major domains (eubacteria, archaeobacteria, and eukaryotes) and phylogen etic structure within the eukaryote domain. Shaded clades include photosynthetic members. Simplified from Woese CR, Kandler O, and Wheelis ML (1990) Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya. Proceedings of the National Academy of Sciences of the USA 87: 4576 4579 and Baldauf SL, Roger AJ, Wenk Siefert I, and Doolittle WF (2000) A kingdom level phylogeny of eukaryotes based on combined protein data. Science 290: 972 977.
many eukaryotes maintain complex cell morphologies that can be preserved in the fossil record. For multicellular organisms, eukaryotic morphologies extend to cell organization and differentiation. Through intercellular interaction (e.g. via plasmodesmata and pit connections), controlled cell division, and genetic regulation, multicellular eukaryotes can have well-organized cell arrangements and some degree of cellular differentiation, which distinguishes them from prokaryotes. In addition, eukaryotic biochemistry may leave fingerprints in ancient rocks in the form of molecular fossils or biomarkers. Steranes, for example, are good eukaryote biomarkers. Steranes are derived from steroids, and the biosynthesis of steroids is almost exclusively a characteristic of eukaryotes. Not only the size range but also the morphological disparity of eukaryotes overlaps with that of prokaryotes. Many eukaryotes, particularly single-celled ones, can be morphologically simple and volumetrically
small. It can be very difficult to distinguish such eukaryotes from prokaryotes. This is a significant challenge because the majority of Precambrian microfossils are morphologically simple. Many of them are featureless filamentous or spheroidal structures, less than a few tens of micrometres in size, preserved in cherts or shales. The default interpretation of these microfossils is that they are prokaryotes. This is not an unreasonable interpretation for many, particularly if they can be shown to be microbial mat builders. For others, however, a eukaryote interpretation is equally plausible; there is simply not enough morphological detail to allow an unambiguous interpretation. These fossils are not the focus of this article. The discussion below takes a more conservative approach, focusing only on the Precambrian fossils that can be reasonably interpreted as eukaryotes. In addition, the discussion will concentrate on marine sediments; there are no known convincing eukaryote fossils in Precambrian terrestrial sediments.
356 PRECAMBRIAN/Eukaryote Fossils
Figure 2 Palaeproterozoic (A) and Mesoproterozoic (B F) eukaryote fossils. (A) Coiled ribbons from the Negaunee Iron Formation, Michigan, USA. Specimen is about 2 cm in maximum dimension. Photograph courtesy of Bruce Runnegar. (B) Dictyosphaera delicate from the Ruyang Group, northern China. Notice the polygonal pattern (ca. 1 mm in diameter) on the vesicle surface. (C) Shuiyou sphaeridium macroreticulatum from the Ruyang Group, northern China. Arrow points to spines on the vesicle surface. This taxon also shows a polygonal pattern on the vesicle surface. Reproduced with permission from Xiao S, Knoll AH, Kaufman AJ, Yin L, and Zhang Y (1997) Neoproterozoic fossils in Mesoproterozoic rocks? Chemostratigraphic resolution of a biostratigraphic conundrum from the
PRECAMBRIAN/Eukaryote Fossils 357
Archaean (3800–2500 Ma) Eukaryotes Phylogenetic relationships among the three domains (Figure 1) indicate that the eukaryotes may have a history that extends back as far as those of the other two domains. Since there are microbial fossils in the Archaean, the possibility of Archaean eukaryotes is intriguing. Indeed, we can follow the footprints of eukaryotes into the Archaean. Steranes, which are eukaryote biomarkers derived from steroids, occur abundantly in the 2600 Ma Marra Mamba Formation and the 2715 Ma Maddina Formation in northwestern Australia. But so far no eukaryote microfossils have been found in Archaean rocks. The presence of steranes in Archaean rocks, however, does not necessarily imply that all of the features (mitochondria, cytoskeletons, and histone) that collectively define living eukaryotes evolved in the Archaean. It is possible that the biosynthesis of steroids appeared early in eukaryote history. These Archaean organisms may represent ancient branches that are more closely related to the eukaryotes than to the other two domains but that diverged before the last common ancestor of all living eukaryotes evolved. In other words, they may represent stemgroup eukaryotes.
Palaeoproterozoic (2500–1600 Ma) Eukaryotes Eukaryotic biomarkers continue to occur in Palaeoproterozoic and younger rocks, for example in the McArthur Group (approximately 1600–1700 Ma) in northern Australia. But it is in the Palaeoproterozoic rocks that the earliest morphological evidence for eukaryotes has been found. Such morphological evidence appears in two forms: macroscopic carbonaceous compressions and organic-walled microfossils. Among Palaeoproterozoic carbonaceous compressions, the coiled ribbons (Figure 2A) of millimetric width and centimetric length from the ca. 2000 Ma Negaunee Iron Formation of Michigan are probably the most famous because of the ancient age of the formation. The Negaunee ribbons resemble the Mesoproterozoic fossil Grypania spiralis. Millimetre-sized discoidal to elliptical compressions resembling Chuaria and Tawuia are known from the ca. 1800–1900 Ma
Changzhougou Formation and Chuanlinggou Formation in northern China. The slightly younger (ca. 1700 Ma) Tuanshanzi Formation in northern China contains millimetre- to centimetre-sized carbonaceous ribbons and blades. Although the cellular details of these fossils are not preserved, their macroscopic and stable morphologies suggest that they are probably the earliest eukaryotic fossils known so far. The Changzhougou Formation and Chuanlinggou Formation in northern China also contain organicwalled microfossils (or acritarchs) and multicellular eukaryotes. The Chinese acritarchs are spherical vesicles with simple morphology but relatively large size (about 100 mm in diameter). They are commonly referred to the genus Leiosphaeridia. They are interpreted as single-celled resting cysts of ancient eukaryotes. The relatively thick and resistant vesicles of the resting cysts allowed their preservation in carbonaceous shales. The cellular details of another fossil, Qingshania magnifica, described from the Chuanlinggou Formation, are preserved, and the organism shows evidence of cellular differentiation – the expanded terminal cell at one end of the clavate filament was probably a reproductive cell. Qingshania magnifica may well be a multicellular eukaryote. Palaeoproterozoic eukaryotic fossils do not have distinct morphological features that would allow them to be placed into extant eukaryotic clades. Like the Archean molecular fossils, the Palaeoproterozoic fossils may also reflect stem-group divergence in the early history of eukaryotes.
Mesoproterozoic (1600–1000 Ma) Eukaryotes Morphologically simple acritarchs such as Leiosphaeridia continue to dominate Mesoproterozoic microfloral assemblages, and Grypania spiralis, Chuaria, and Tawuia are also known in Mesoproterozoic successions. However, several morphologically complex acritarchs first appeared in the Mesoproterozoic. In addition, the first crown-group eukaryotes (that is, fossils that can be phylogenetically resolved to an extant eukaryote clade) also occur in the Mesoproterozoic. This indicates that crown-group eukaryotes began to diversify in the Mesoproterozoic. By the
North China Platform. Precambrian Research 84: 197 220. (D, E) Bangiomorpha pubescens, interpreted as a bangiophyte red alga, from the Hunting Formation, arctic Canada. Arrow in (E) points to basal holdfasts. Specimen in (D) shows the transition from uniseriate to multiseriate growth, suggesting multiple ontogenetic phases and probably sexual reproduction in Bangiomorpha pubescens. Repro duced with permission from Butterfield NJ (2000). Bangiomorpha pubescens n.gen., n.sp.: Implications for the evolution of sex, multi cellularity, and the Mesoproterosoic/Neoproterozoic radiation of eukaryotes. Paleobiology 26: 386 404. (F) Palaeovaucheria clavata, interpreted as a xanthophyte alga, from the Lakhanda Group, south eastern Siberia. Photograph courtesy of Andrew H. Knoll.
358 PRECAMBRIAN/Eukaryote Fossils
end of the Mesoproterozoic, several algal groups had already diverged. Acanthomorphic acritarchs (or acritarchs bearing spines) and acritarchs with patterned vesicles are known in the 1400–1500 Ma Roper Group of northern Australia and the >1000 Ma Ruyang Group of northern China. The Roper and Ruyang assemblages include such acanthomorphs as Tappania plana and Shuiyousphaeridium macroreticulatum (Figure 2C), as well as acritarchs with polygonally patterned (Dictyosphaera delicate and Shuiyousphaeridium macroreticulatum; Figure 2B and 2C) or striated (Valeria lophostriata) vesicles. Some of these fossils have wide geographical and long stratigraphical ranges. For example, Valeria lophostriata and Tappania occur in Mesoproterozoic and Neoproterozoic rocks in Laurentia, Australia, India, and northern China. As no prokaryotes are known to have comparable levels of morphological complexity, these acritarchs are probably eukaryotic. Some of these acritarchs (for example Dictyosphaera delicate) preserve an organic d13C signature that is consistent with eukaryotic photosynthetic biochemistry, but it is unclear which algal group (e.g. chlorophytes, rhodophytes, or stramenopile algae – a group that includes chrysophytes, xanthophytes, diatoms, and brown algae; Figure 1) they belong to. One of the earliest eukaryotic fossils that has been confidently attributed to a modern algal group is Bangiomorpha pubescens (Figures 2D and 2E) from the 1200 Ma Hunting Formation in arctic Canada. This is a multicellular filamentous fossil that shows evidence of holdfast differentiation and sexual reproduction. It is interpreted as a benthic bangiophyte red alga. Another phylogenetically resolved eukaryotic fossil is Palaeovaucheria clavata (Figure 2F), interpreted as a xanthophyte alga, from the upper Mesoproterozoic Lakhanda Group in south-eastern Siberia. Xanthophyte algae are members of the photosynthetic stramenopiles whose plastids were derived from a secondary endosymbiont (probably a red alga). The occurrence of Bangiomorpha pubescens and Palaeovaucheria clavata in Mesoproterozoic rocks suggests that not only must crown-group eukaryotes such as red algae have diverged but also the secondary endosymbiotic event leading to stramenopile algae must have occurred by the end of the Mesoproterozoic.
Neoproterozoic (1000–540 Ma) Eukaryotes The Neoproterozoic era includes several major milestones in eukaryote evolution. The diversity and morphological complexity of eukaryotes increased
appreciably in the Neoproterozoic, and several phylogenetically and ecologically important eukaryotic groups make their first appearance in the Neoproterozoic fossil record. These include heterotrophic protists, biomineralizing protists, and, towards the end of this era, animals. Molecular-clock estimates also indicate that land plants and fungi may have diverged in the Neoproterozoic, but so far this has not been confirmed by palaeontological evidence. Major environmental crises occurred in the middle Neoproterozoic. Between about 720 and 600 Ma, the Earth experienced at least two global glaciations (also known as ‘snowball Earth events’), during which glaciers reached the tropical oceans. More glaciations may have occurred in the Neoproterozoic, but these were not nearly as extreme. It is therefore convenient to divide the Neoproterozoic into three intervals: Early (1000–720 Ma), Middle (720–600 Ma), and Late (600–543 Ma) Neoproterozoic. Early Neoproterozoic
A quick look at several fossiliferous units of Early Neoproterozoic age gives us a broad picture of eukaryote diversity at that time. The Early Neoproterozoic Little Dal Group (850–780 Ma) in northwestern Canada, the Chuar Group (>742 Ma) in the Grand Canyon, and the Huainan Group and Huaibei Group (ca. 740–900 Ma) in northern China contain some of the best-preserved carbonaceous compressions in the Early Neoproterozoic. Chuaria and Tawuia (Figures 3A and 3B) are abundant in these successions. In addition, Longfengshania stipitata – a benthic alga with an ellipsoidal head, a stipe, and a simple holdfast – has been described from the Little Dal Group and from the Changlongshan Formation (ca. 800–900 Ma) in northern China. Individuals of L. stipitata sometimes occur in clusters (Figure 3C), like a bunch of inflated balloons tethered together. These carbonaceous compressions are probably multicellular eukaryotes, although Chuaria and Tawuia have been interpreted as colonial cyanobacteria by some palaeontologists. Acritarchs, particularly those with more complex morphologies, became more diverse in the Early Neoproterozoic (Figure 3D). More than 20 acritarch species with complex morphologies have been described from Early Neoproterozoic successions. Except a few (e.g. Tappania and Valeria lophostriata), they are not known in the Mesoproterozoic. The Wynniatt Formation (ca. 800–900 Ma) of arctic Canada, the Svanberfjellet Formation (ca. 700–800 Ma) in Spitsbergen, and the Mirojedikha Formation in Siberia, for example, contain some of the best-preserved acritarchs in the Early Neoproterozoic. The Svanberfjellet
PRECAMBRIAN/Eukaryote Fossils 359
Figure 3 Early Neoproterozoic eukaryote fossils. (A) Chuaria circularis from the Huaibei Group, northern China. (B) Tawuia dalensis from the Huaibei Group, northern China. (C) Longfengshania stipitata from the Little Dal Group, north western Canada. Reprinted with permission from Butterfield NJ, Knoll AH, and Swett K (1994) Paleobiology of the Neoproterozoic Svanbergfjellet Formation, Spitsbergen. Fossils and Strata 34: 1 84. (D) Trachyhystrichosphaera polaris from the Svanberfjellet Formation, Spitsbergen. Reproduced from Hofmann HJ (1985) The mid Proterozoic Little Dal macrobiota, Mackenzie Mountains, north west Canada. Palaeontology 28: 331 354 by permission of Taylor & Francis AS. (E) Proterocladus major, interpreted as a coenocytic green alga, from the Svanberfjellet Formation, Spitsbergen. Reproduced from Butterfield NJ, Knoll AH, and Swett K (1994) Paleobiology of the Neoproterozoic Svan bergfjellet Formation, Spitsbergen. Fossils and Strata 34: 1 84 by permission of Taylor & Francis AS. (F) Cyclocyrillium simplex, a vase shaped microfossil interpreted as a testate amoeba, from the Chuar Group, Grand Canyon. Reproduced with permission from Porter SM and Knoll AH (2000) Testate amoebae in the Neoproterozoic era: evidence from vase shaped microfossils in the Chuar Group, Grand Canyon. Paleobiology 26: 360 385. (G) Protoarenicola baiguashanensis, probably a benthic alga, from the Huaibei Group, northern China. Arrows point to holdfast structures. Photograph courtesy of Xunlai Yuan.
Formation has also yielded the earliest known green algal fossil–Proterocladus (Figure 3E). Biologically controlled mineralization and the formation of skeletons is an evolutionary event that significantly enhances fossil preservation. All fossils
discussed in the preceding sections are non-biomineralizing eukaryotes. The earliest known biomineralizing protists are probably from the Early Neoproterozoic. Silicified scale-shaped microfossils (less than 100 mm in diameter) from the Tindir Group (age poorly
360 PRECAMBRIAN/Eukaryote Fossils
constrained, but possibly 620–780 Ma) of north-western Canada can be compared to chrysophyte skeletons. Vase-shaped microfossils (Figure 3F) from the Chuar Group and other Early Neoproterozoic successions (e.g. the Visingso¨ Formation in Sweden and the Draken Conglomerate Formation in Spitsbergen) have been interpreted as testate amoebae whose mineralized tests are typically preserved in casts and moulds. The Chuar testate amoeba fossils add another dimension to our consideration of the Neoproterozoic biosphere, and that is heterotrophy. Because testate amoebae are heterotrophic protists, the Chuar vaseshaped microfossils suggest that the Early Neoproterozoic biosphere was ecologically complex. Of course, heterotrophic eukaryotes must have evolved earlier. In fact, the earliest eukaryotes may be heterotrophic, given that eukaryotic autotrophy evolved through primary and secondary endosymbiotic events. However, evidence for heterotrophy in the fossil record is scarce. Vase-shaped microfossils and ciliate biomarkers from the Chuar Group are probably the earliest known evidence for heterotrophic eukaryotes. The best-known heterotrophic eukaryotes are perhaps the animals. There have been many reports of animal fossils from Early Neoproterozoic and Mesoproterozoic successions, but their interpretation has been controversial. Sinosabellidites huainanensis, Pararenicola huaiyuanensis, and Protoarenicola baiguashanensis (Figure 3G), from the Huainan Group and Huaibei Group (ca. 740–900 Ma) of northern China, are some of the often-cited Early Neoproterozoic animal fossils. They are carbonaceous compressions of tubes of millimetric diameter and centimetric length with transverse annulations. The transverse annulations superficially resemble animal metameric segmentation. A few specimens bear poorly defined terminal structures that have been interpreted as proboscis-like structures. However, recent study has shown that these carbonaceous compressions are probably benthic tubular algae. Middle Neoproterozoic
The Middle Neoproterozoic is characterized by multiple global glaciations, and is unofficially labelled as the Cryogenian Period by some Precambrian geologists. On a broad scale, acritarchs and other eukaryotes suffered significant losses of diversity in the Middle Neoproterozoic. The documented diversities of several Middle Neoproterozoic assemblages are extremely low, and such assemblages are typically dominated by Sphaerocongregus variabilis (or Bavlinella faveolata). This Cryogenian drop in eukaryote diversity may be a true evolutionary pattern that was related to the glaciation events. Despite the
loss of diversity, the occurrence of red algae, green algae, photosynthetic stramenopiles, and testate amoebae in Mesoproterozoic and Early Neoproterozoic rocks suggests that some members of these groups must have survived the Middle Neoproterozoic glaciations. Late Neoproterozoic
Eukaryote diversity rose sharply in the Late Neoproterozoic. Both acritarchs and multicellular algae reached unprecedented levels of complexity and diversity in the Late Neoproterozoic. Some of the multicellular algae are preserved in anatomical detail, allowing them to be placed within the red algae. One of the most important landmarks in Late Neoproterozoic eukaryote evolution is the emergence of animals and animal biomineralization. Molecularclock studies suggest that the deepest (protostomes– deuterostomes) divergence within the crown-group bilaterian animals probably occurred in the Mesoproterozoic or Early Neoproterozoic. But, as discussed above, there is no convincing palaeontological evidence to support these molecular-clock estimates. Some have suggested that perhaps the earliest animals were microscopic in size and would not be well preserved in the fossil record. The Doushantuo Formation and Dengying Formation of South China provide several taphonomic windows onto the Late Neoproterozoic biosphere. Carbonaceous shales, cherts, and phosphorites of the Doushantuo Formation (ca. 600–550 Ma) preserve some of the most extraordinary eukaryote fossils in the Neoproterozoic. More than 20 taxa of macroscopic carbonaceous compressions have been reported from the Yangtze Gorges area and elsewhere in South China. Most of these compressions can be unambiguously interpreted as multicellular algae (Figure 4A). Some of them show clear evidence of holdfast anchoring, dichotomous branching, apical meristematic growth, and specialized reproductive structures. A few of these compressions (for example Calyptrina striata and Sinospongia typica; Figure 4B), however, have been interpreted as sponges or cnidarians, but such interpretations are not unique and an algal interpretation cannot be falsified conclusively. In any case, it is safe to conclude that none of the Doushantuo compressions can be interpreted as macroscopic bilaterians (bilaterally symmetrical animals). Multicellular algae also occur in the Doushantuo cherts and phosphorites. Cellular features are preserved (Figures 4C–F), so anatomical detail can be deduced from these fossils. Many of these silicified or phosphatized algal fossils show pseudoparenchymatous thallus construction, apical meristematic
PRECAMBRIAN/Eukaryote Fossils 361
Figure 4 Late Neoproterozoic eukaryote fossils from the Doushantuo Formation, South China. (A) Konglingiphyton erecta, a dichot omously branching alga from the Doushantuo shales. (B) Sinospongia typica, possibly a benthic tubular alga from the Doushantuo shales. A, B, reproduced from Xiao S, Yuan X, Steiner M, and Knoll AH (2002) Macroscopic carbonaceous cpmpressions in a terminal Proterozoic shule: A systematic reassessment of the Miaohe biota, South China Journal of Paleontology 76: 345 374 with permission of The Paleontological Society. (C) A florideophyte red alga with medulla cortex thallus differentiation from the Doushantuo phosphorites. (D, E) Florideophyte red algae with possible reproductive structures from the Doushantuo phosphorites: (D) larger and darker cells arranged in clusters are interpreted as possible carposporangia; (E) tetrads and octads embedded in algal thallus are interpreted as possible tetraspores and octospores. (F) Wengania globosa, a possible stem group florideophyte red alga from the Doushantuo phosphorites. (G) Meghystrichosphaeridium reticulatum, an acanthomorphic acritarch from the Doush antuo phosphorites. (H) Tianzhushania spinosa, a large acanthomorph from the Doushantuo cherts. (A, C F) are courtesy of Xunlai Yuan.
growth, thallus differentiation (Figure 4C), and specialized reproductive structures (Figures 4D and 4E). Some of them have been interpreted as stem-group florideophyte red algae or stem-group coralline algae, suggesting the presence of advanced red algae in Late Neoproterozoic oceans. In addition, the Doushantuo cherts and phosphorites contain an assemblage of large (several hundred micrometres in diameter) acanthomorphs that are morphologically complex and taxonomically diverse (Figure 4G and 4H). This assemblage includes nearly 30 species, some of which are also known to occur in other Late Neoproterozoic successions, for example
the Pertatataka Formation of central Australia, the Scotia Group of Svalbard, the lower Krol Group (Krol A) of Lesser Himalaya, and lower Vendian rocks in eastern Siberia. In the Yangtze Gorges area, elements of this acritarch assemblage (for example Tianzhushania spinosa) first appear just metres above Cryogenian glacial deposits, suggesting that the eukaryote recovery occurred shortly after the last Cryogenian glaciations. The Doushantuo diversification of complex acritarchs appears to be ephemeral, however. Available evidence suggests that most of the Doushantuo acritarchs disappeared when Ediacaran animals
362 PRECAMBRIAN/Eukaryote Fossils
Figure 5 Late Neoproterozoic eukaryote fossils. (A D) Animal eggs and blastulas at (A) one , (B) two , (C) four , and (D) many cell cleavage stages, Doushantuo phosphorites, South China. Scale bar in (A) applies to (A D). Reproduced from Xiao S, Zhang Y, and Knoll AH (1998) Three dimensional preservation of algae and animal embryos in a Neoproterozoic phosphorite. Nature 391:553 558 with permission from Nature Publishing Group, and from Xiao S and Knoll AH (2000) Phosphatized animal embryos from the Neoproterozoic Doushantuo Formation at Weng’an, Guizhou, South China. Journal of Paleontology 74:767 788 with permission from The Paleontological Society. (E) Sinocyclocyclicus guizhouensis, interpreted as a stem group cnidarian fossil, from the Doushantuo phosphorites, South China. (F) Dickinsonia costata, an Ediacaran fossil from the Flinders Ranges of South Australia. Photograph of a cast replicate preserved at the Botanical Museum, Harvard University. (G) Cloudina riemkeae, a biomineralized tubular animal fossil, from the Dengying Formation, South China. Photograph courtesy of Hong Hua.
began to diversify about 575 Ma ago. The disappearance is unlikely to be a preservational artefact: slightly younger phosphorites in the Dengying Formation contain no Doushantuo-type acritarchs, and basal Cambrian acritarchs are entirely different from those in the Doushantuo Formation. It is possible that the extinction of Doushantuo-type acritarchs is related to yet another small-scale glaciation in the Late Neoproterozoic. Perhaps the most exciting fossils from the Doushantuo Formation are the submillimetric globular microfossils preserved in phosphorites at Weng’an in Guizhou Province. These globular microfossils have been interpreted as animal blastulae at successive cleavage stages, containing 2n blastomeres within an envelope of roughly constant size (Figure 5A–D). The geometry of these microfossils is consistent with blastula cell division. It has not been determined, however, to which animal clade(s) these fossil embryos belong. Their chimeric combination of features that individually occur in crown-group sponges, cnidarians, and bilaterians suggests that these embryos may belong to stem groups
at the animal, eumetazoan, or bilaterian levels. There are other microfossils in the Weng’an phosphorites that have been interpreted as adult sponges, adult cnidarians (Figure 5E), and putative bilaterian gastrulas. However, none of these are more than a few millimetres in size. Perhaps the earliest animals, at least bilaterian animals, were indeed microscopic. Macroscopic bilaterian animals probably first evolved in the latest Neoproterozoic Ediacaran time. Among the best-known Ediacaran fossils, those in the Newfoundland assemblage (ca. 575–565 Ma) are probably the oldest, those in the White Sea and South Australia assemblages (ca. 555 Ma) are younger, and those in the Namibia assemblage (ca. 549–543 Ma) are the youngest. There do not appear to be any unambiguous macrobilaterians in the Newfoundland assemblage. The White Sea and South Australia assemblages, on the other hand, include body and trace fossils of macrobilaterians (e.g. Kimberella quadrata), as well as classic Ediacaran fossils such as Dickinsonia (Figure 5F) whose phylogenetic interpretations are still controversial. The Namibia assemblage and its equivalents (for
PRECAMBRIAN/Prokaryote Fossils 363
example the Dengying Formation in South China) include biomineralizing animals such as Cloudina (Figure 5G), Namapoikia, and Namacalathus. These Ediacaran assemblages, therefore, record a succession of evolutionary events leading to the rise of macrobilaterians and animal biomineralization. Most Ediacaran fossils, however, disappeared near the Precambrian– Cambrian boundary, closing the last chapter in Precambrian eukaryote evolution.
Acknowledgments I would like to thank the National Science Foundation for funding my research on Proterozoic palaeontology. Nicholas J Butterfield, Hans J Hofmann, Hong Hua, Andrew H Knoll, Susannah M Porter, Bruce Runnegar, Leiming Yin, and Xunlai Yuan kindly provided the photographs used in this article.
See Also Fossil Plants: Calcareous Algae. Microfossils: Acritarchs. Origin of Life. Precambrian: Prokaryote Fossils; Vendian and Ediacaran. Sedimentary Rocks: Chert. Trace Fossils.
Further Reading Baldauf SL, Roger AJ, Wenk Siefert I, and Doolittle WF (2000) A kingdom level phylogeny of eukaryotes based on combined protein data. Science 290: 972 977. Brocks JJ, Logan GA, Buick R, and Summons RE (1999) Archean molecular fossils and the early rise of eukary otes. Science 285: 1033 1036. Butterfield NJ, Knoll AH, and Swett K (1994) Paleobiol ogy of the Neoproterozoic Svanbergfjellet Formation, Spitsbergen. Fossils and Strata 34: 1 84.
Fedonkin MA and Waggoner BM (1997) The late Pre cambrian fossil Kimberella is a mollusc like bilaterian organism. Nature 388: 868 871. Hofmann HJ (1985) The mid Proterozoic Little Dal macro biota, Mackenzie Mountains, north west Canada. Palaeontology 28: 331 354. Knoll AH (1996) Archean and Proterozoic paleontology. In: Jansonius J and McGregor DC (eds.) Palynology: Prin ciples and Applications, pp. 51 80. Salt Lake City: American Association of Stratigraphic Palynologists Foundation, Publishers Press. Narbonne GM (1998) The Ediacara biota: a terminal Neo proterozoic experiment in the evolution of life. GSA Today 8: 1 6. Porter SM and Knoll AH (2000) Testate amoebae in the Neoproterozoic era: evidence from vase shaped microfossils in the Chuar Group, Grand Canyon. Paleobiology 26: 360 385. Schopf JW and Klein C (1992) The Proterozoic Biosphere: A Multidisciplinary Study. Cambridge: Cambridge University Press. Vidal G and Moczydlowska Vidal M (1997) Biodiversity, speciation, and extinction trends of Proterozoic and Cambrian phytoplankton. Paleobiology 23: 230 246. Waggoner B (2003) The Ediacaran biotas in space and time. Integrative and Comparative Biology 43: 104 113. Woese CR, Kandler O, and Wheelis ML (1990) Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya. Proceedings of the Na tional Academy of Sciences of the USA 87: 4576 4579. Xiao S, Zhang Y, and Knoll AH (1998) Three dimensional preservation of algae and animal embryos in a Neopro terozoic phosphorite. Nature 391: 553 558. Yuan X, Xiao S, Yin L, et al. (2002) Doushantuo Fossils: Life on the Eve of Animal Radiation. Hefei: China University of Science and Technology Press. Zhang Y, Yin L, Xiao S, and Knoll AH (1998) Per mineralized Fossils from the Terminal Proterozoic Doushantuo Formation, South China. Memoir 50. The Paleontological Society, Lawrence, Kansas, USA.
Prokaryote Fossils M D Brasier, University of Oxford, Oxford, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Conditions for the development and survival of life on the young planet Earth are thought to have been extremely harsh. Reconstruction of those conditions requires multidisciplinary and interdisciplinary research (combining astronomy, planetary science, microbial and molecular biology, genetics, geochemistry, petrology, palaeobiology, and geology). These
disciplines are united in the emerging field of astrobiology (exobiology), which covers the phenomenon of life within the solar system and beyond. Astrobiology takes a particular interest in modern and ancient prokaryotic ecosystems on Earth and in the prebiotic–biotic transition and its definition. Meteoritic and lunar research shows that accretion of planet Earth took place around 4550 Ma ago (see Earth Structure and Origins). This accretion was soon followed by melting of the planetary surface owing to the energy released through gravitational collapse. A cataclysmic collision with a Mars-sized planet is
364 PRECAMBRIAN/Prokaryote Fossils
thought to have produced the Moon within the first 100 Ma or so. The Moon played a major role in setting the equable conditions for life on Earth through its influence in stabilizing our planetary motion about the rotational axis and by providing stable tidal and seasonal oscillations. Earth’s surface had cooled and the oceans had probably condensed by about 4000 Ma, as shown by ca. 4200 Ma old (hydrous melt) zircons in the oldest rocks of Canada and by putative sediments in the 3800 Ma Akilia and Isua Groups of Greenland. Even so, the study of
craters on the Moon and Mars shows that their surfaces were bombarded by large asteroids between 4550 Ma and 3800 Ma. Similar bombardment is likely to have affected the surface of the Earth, although any evidence has been destroyed by the mobility of the planetary surface. These bombardments would have had a devastating effect on any early biosphere (see Figure 1; the so-called ‘impact frustration of the origin of life’). For example, the impact of an asteroid more than 350 km across would release enough energy to boil off the oceans, while that of
Figure 1 The main succession of events inferred for the evolution of the biosphere alongside geological evidence for changing levels of atmospheric oxygen and carbon dioxide during the Precambrian. The evidence for the first isotopic indicators (more than 3700 Ma) and the oldest prokaryotes (ca. 3450 Ma) is currently in dispute.
PRECAMBRIAN/Prokaryote Fossils 365
one about 150 km across could destroy the photic zone. Any life-forms synthesized prior to 3800 Ma could therefore have been repeatedly destroyed by episodic catastrophes. The only forms that could have survived such a holocaust would have been heat-tolerant hyperthermophile bacteria, which today are found living around volcanic and hydrothermal vents or deep in the Earth’s crust. Such lifeforms arguably originated from amino acid to RNA and DNA synthesis around hot alkaline hydrothermal vents. This is suggested not only by the antiquity of hyperthermophiles, as shown by molecular phylogeny (see below), but also by the ready availability of carbon sources (methane, carbon dioxide, carbon monoxide, aliphatics) and electron donors (metals, hydrogen sulphide, hydrogen) around such vents today, together with a plentiful supply of the phosphorus and transition metals that are needed to synthesize nucleic acids and enzymes. Against this, it has been argued that hydrothermal temperatures are too high to allow the necessary stability of complex organic molecules. But, on the early Earth, such hydrothermal vents were rich in metals (most of the world’s metal resources come from the Archaean to Palaeoproterozoic crust) and clay minerals. These metals and clay minerals could arguably have allowed the stabilization of nucleic acids, even at relatively high temperatures. No other site on Earth could supply these essential building blocks in so many forms and so readily. Asteroid bombardment of the Earth is thought to have declined progressively between about 3800 Ma and 3000 Ma. The hot, thick, and largely oceanic crust of this age can be envisaged as mantled by oceanic waters of moderate depth with a mean temperature of approximately 60 C. This means that aquatic habitats on Earth prior to about 3000 Ma may have been essentially hydrothermal in nature. It was not until about 3000 Ma that the wide shallow continental shelf seas, deep-sea troughs, and land areas began to develop. Crustal sources of free energy may therefore have begun to dwindle after about 3000 Ma, and the emerging biosphere may have been forced to switch to using solar energy, with one group perhaps converting from a heat-seeking catalyst to a light-seeking (photosynthetic) one. Prokaryotes are considered to be the most primitive organisms on Earth and are assumed to have emerged early in the comparatively hot world of the Archaean. The cells of still-living forms are extremely small, generally less than 1 mm in diameter, and they lack a nucleus (hence the name prokaryote, which is derived from the Greek pro ¼ before and karyos ¼ nucleus). They may be single or colonial, the latter enclosed within a mucilaginous sheath called a capsule. Some
living bacterial cells bear a whip-like thread (flagellum), and a few contain chlorophyll pigments for photosynthesis. Prokaryotes are important today in the formation of microbial sediments (see Biosediments and Biofilms), such as cyanobacterial mats and stromatolites, iron and manganese ores, carbonate concretions, and sulphide and sulphate minerals. They also yield important information about the early evolution of the cell and the histories of methanogenesis, photosynthesis, and biogeochemical cycles.
Molecular and Biochemical Evidence Contrasts between the ribosomal RNA sequences of diverse bacteria, protists, fungi, plants, and animals indicate that life on Earth can be divided into three primary domains: the Archaea (or ‘Archaebacteria’), which includes the methanogenic and sulphur bacteria; the Bacteria (or ‘Eubacteria’), which includes cyanobacteria and other forms with photosynthetic pigments; and the Eucarya (or ‘Eukaryota’), which includes all the protists, fungi, plants, and animals (Figure 2). It is notable that all the most deeply rooted branches of the tree of life are occupied by modern hyperthermophilic bacteria, which live at temperatures of 80–110 C and are seldom able to grow below 60 C. This has been taken to suggest that the
Figure 2 The three branches of the tree of life, in which all the deep seated branches are hyperthermophilic bacteria (shown in bold). Time increases along the branches, but not necessarily in a linear fashion or at the same rate in each branch. Longer branches relate to faster evolution.
366 PRECAMBRIAN/Prokaryote Fossils
last common ancestor of all living organisms was a hyperthermophile, adapted to hot hydrothermal springs or to life deep in the Earth’s crust. One explanation for this is that life originated in such conditions. Another explanation is that hyperthermophiles were pre-adapted to survive the catastrophic period of meteorite bombardment between 4500 Ma and 3800 Ma. Biochemical evidence can be taken to suggest the following evolutionary sequence of autotrophic prokaryotes, each of which used carbon dioxide as their sole source of carbon. 1. Anaerobic chemolithotrophic prokaryotes, which mainly use hydrogen produced from inorganic reactions between rock and water as their main electron source. 2. Anaerobic anoxygenic prokaryotes such as green and purple sulphur bacteria, which use photosynthesis to reduce carbon dioxide to form organic matter, with hydrogen sulphide as the electron source, in the absence of oxygen. 3. Oxygenic cyanobacteria, which use photosynthesis to reduce carbon dioxide to form organic matter, with water as the electron source, releasing oxygen. These must have had an enormous impact on Earth surface processes and the biosphere, and considerable interest has been focused by astrobiologists upon their first appearance in the rock record. At the time of NASA’s Viking missions to Mars in 1976, it was such photosynthetic autotrophy that scientists were hoping to find. Heterotrophic prokaryotes do not synthesize organic matter. Like us, they use preformed organic matter as their source of carbon and can use a range of oxidants to break it down and release the energy bonds. Methanogenic Archaea are among the most primitive heterotrophs alive today, living in highly reducing sediments (such as peat bogs) and releasing methane gas. Sulphate-reducing bacteria use seawater sulphate (SO4) ions in the absence of oxygen, but require a highly oxidized form of sulphur (SO4), which may not have been widely available in the early ocean. Aerobic heterotrophic bacteria use freely available atmospheric oxygen and are unlikely to have radiated before the so-called Great Oxygenation Event, 2450–2200 Ma ago, when various indicators of the weakly reducing planetary surface (banded iron formations (see Sedimentary Rocks: Banded Iron Formations), detrital pyrite, uraninite, and siderite) begin to disappear from the rock record and red beds start to appear. This oxygenation event may relate in part to increasing rates of carbon burial in expanding cratonic basins and subduction zones and
in part to the irreversible loss of hydrogen to space from the upper atmosphere. While oxygen producers and consumers could have existed prior to 2450 Ma, they were probably restricted to rather local oases of oxygenation. This inferred evolutionary sequence of methanogenic to sulphate-reducing to aerobic heterotrophic prokaryotes is likely, on the basis of evidence from living bacteria, to have been accompanied by an increasing yield of energy from the same amount of carbonaceous ‘food’. Significantly, this evolutionary succession closely resembles the modern distribution of prokaryotic populations within marine muds, with methanogenic Archaea lying deep within the sediment pile, aerobic heterotrophs and photoautotrophs in the upper layers of the sediment, and sulphate reducers in between.
Evidence for the Earliest Biosphere Biogeochemistry
The fossil evidence for life on Earth gets increasingly scarce as the age of rock units increases. This is because older rocks have suffered more exposure to erosion and have experienced a greater degree of alteration by metamorphism. Hence, the oldest rocks on Earth (approximately 3800–3700 Ma), from Isua and Akilia in Greenland, have been too heavily metamorphosed to yield morphological evidence of life. Possible traces of life must therefore be explored using biogeochemical techniques. Stable isotopes of carbon from Isua and Akilia, for example, are somewhat lighter than usually expected from an inert world (ca. 18% d13 CPDB cf. Pee Dee Belemnite standard). This has been taken to imply that life was able to self-organize and survive the period of catastrophic meteorite impacts before about 3800 Ma (Figure 1). Such a view is now controversial for a variety of reasons. The sedimentary origin of the carbonaceous grains is questionable: fractionation of carbon compounds could also have resulted from abiogenic processes or even from carbonaceous meteoritic debris. The carbon may also be younger than claimed. While light carbon isotopes (ca. 40% to 25% d13 PDB) are commonly encountered in rocks younger than about 3500 Ma, some of these hydrocarbon compounds may also have an abiogenic origin, and precise discriminators, such as C–H ratios, H isotopes, and aliphaticity, are needed to discount this possibility. Even so, most scientists assume that this 25–40% difference in carbon isotopes of carbonates and organic matter seen after about 3500 Ma provides key evidence that biological metabolic pathways (i.e. autotrophic fractionation of carbon isotopes) were in place by this time.
PRECAMBRIAN/Prokaryote Fossils 367
A trend towards highly negative carbon isotope ratios about 2800–2200 Ma ago has been explained as the result of a bloom in heterotrophic methanogenic and methanotrophic Archaea at that time, prior to the Great Oxidation Event. Sulphur isotopes can also be used to trace the history of prokaryotic metabolism. In this case, 32S is preferentially taken up by sulphate-reducing bacteria, leaving the water column enriched in the heavier isotope, 34S. Recent studies of the ratio between 32S and 34S in sedimentary pyrite and in supersaturated barite, gypsum, and anhydrite have been used to argue that bacterial sulphate reduction was not in place before about 2450 Ma. Prior to this, sulphur isotope fractionation seems to have been of the abiogenic mass-independent kind, more like that seen in the Martian atmosphere. This lack of bacterial sulphate reduction before about 2450 Ma may be explained by a lack of sufficient atmospheric oxygen to form the sulphate ions and/or by surface water temperatures that were too high in the Archaean to produce a measurable fractionation. Organic geochemical ‘biomarkers’ can also be sniffed out in well-preserved rocks, at least as far back as 2700 Ma. Biomarkers called 2-methylhopanes have been reported from rocks of this age in Western Australia and have been taken to indicate the
presence of cyanobacteria at that time. This line of uniformitarian reasoning assumes, of course, that such biomarkers were not present in any other prokaryotic group, living or extinct, which is rather difficult to refute. Unfortunately, most rocks older than 2700 Ma appear to be too ‘cooked’ to preserve complex organic molecules. Stromatolites
Laminated domical structures known as ‘stromatolites’ (Figure 3A) have been described from carbonate rocks as old as 3450 Ma in the Pilbara Supergroup of Western Australia (Figure 3C) and in the coeval Swaziland Supergroup of South Africa. Although an origin from the accretion of prokaryotic and even cyanobacterial mats has often been inferred for these early Archaean examples, they do not contain microfossils and they show some features that render their biogenicity rather questionable: an association with, and continuation down into, hydrothermal dyke systems; close association with epigenetic crystal fans (e.g. after aragonite, barite, gypsum) and directly precipitated carbonates; intergradation with ripple-like forms having rotational symmetry (i.e. they look the same upside down); lack of multiple
Figure 3 Stromatolites and Precambrian prokaryotes. (A) Living cyanobacterial diatom stromatolites from the hypersaline Shark Bay in Western Australia; compass is ca. 6.5 cm wide. (B) The coccoid cyanobacterium Eoentophysalis preserved in cherts of the ca. 800 Ma old Boorthanna Chert of Australia; scale bar below 100 mm. (C) A 30 cm wide exposure of 3400 Ma old stromatolites from the Strelley Pool Chert of Western Australia; it is unclear whether these are of microbial origin or are an abiogenic crystalline precipitate. (D) The 1900 Ma old filaments of cyanobacterial or iron bacterial origin preserved in the Gunflint Chert of Mink Mountain, Ontario, Canada; scale bar (main picture) 40 mm; scale bar (inset) 100 mm.
368 PRECAMBRIAN/Prokaryote Fossils
fractal dimensions; continuous laminae of barely varying thickness (typical of crystal growth); lack of wrinkle fabrics typical of biofilms; and lack of fenestrae or other signs of biological processing. Such simple stromatolites seem likely to have formed from largely physical processes such as direct chemical precipitation from seawater. It is only after about 3000 Ma that fractally complex stromatolites are found in carbonate platform settings, for which the presence of biofilms seems more plausible, if not always demonstrable. These tend to have wrinkle mat fabrics, discontinuous laminae, and laminar fenestrae (e.g. those from the 2700 Ma old Belingwe Group of Zimbabwe and the Fortescue Group of Western Australia). Microfossils associated with Archaean stromatolites remain questionable, however, until about 2530 Ma and are not really diverse until the 1900 Ma old Gunflint Chert of Canada (Figure 3D). Even here, the biogenicity of the associated stromatolites is open to question because they resemble abiogenic sinters. Silicified Microbiotas
Precambrian oceans appear to have been supersaturated with silica (SiO2) because it was not being removed from the water column by groups that evolved later, such as the diatoms, radiolaria, and sponges. In environments where early diagenetic silica was able to engulf prokaryotic populations, such as in peritidal hypersaline bacterial-mat settings, prokaryotic sheaths and even cell walls were sometimes (but still very rarely) well preserved in three dimensions. This prevalence of hypersaline settings may have produced an unfortunate bias towards prokaryotic assemblages in the early fossil record. In other words, the inferred dominance of prokaryotic microfossils within Precambrian cherts may be due to their restricted hypersaline setting rather than to evolutionary factors. Silicified microbiotas are usually studied by means of standard (30 mm) to thick (300 mm) petrographic thin sections, at magnifications of up to about 400. This thin-section technique is paramount because of the way in which it provides for contextual analysis, including three-dimensional morphology, mineralogy, rock fabric, and rock history. Other techniques, such as maceration (digestion in strong acids), etching of rock chips, scanning electron microscopy, and atomic force microscopy, are also used, but these do not provide the requisite information on context and are prone to the inclusion of structures that are later contaminants or ‘artefacts’ of the preparation process.
A classic example of a silicified Precambrian microbiota is the 1900 Ma old Gunflint Chert, which preserves about 12 taxa of prokaryotes, including forms that superficially resemble coccoid and filamentous cyanobacteria (Figure 3D and insert) but may be more closely allied to extant iron bacteria. The putative cyanobacteria Eoentophysalis (Figure 3B) and Archaeoellipsoides are thought to be present in the approximately 2100 Ma old cherts of West Africa and the 2000 Ma old Belcher Group cherts of Canada, respectively. The latter is claimed to preserve the specialist heterocyst cells used by cyanobacteria to help fix nitrogen in an otherwise oxidizing atmosphere. As one moves back into the Archaean, microfossils become both extremely rare and highly questionable, despite the great abundance of carbonaceous cherts and tufa-like carbonates. This may be explained partly by the inference that Archaean cherts were laid down in largely hydrothermal conditions that were often acidic and reducing, and partly by the scarcity of large and resistant cellular materials at that time. Bundles of silicified filaments and tiny calcified holes from stromatolites in the 2530 Ma old carbonates of the Transvaal Supergroup of South Africa may be the casts of coccoid and filamentous cyanobacteria but little of diagnostic significance is preserved. Intriguingly, such encrusted cyanobacterial filaments are rarely seen before about 1000 Ma, and endolithic microborings are not reported prior to 1500 Ma. A single microfossil-like structure from the 2700 Ma old Fortescue Group of Western Australia has been compared with a cyanobacterial filament but its biogenicity and context awaits full documentation. Skeins of pyritic filaments found within carbonaceous cherts from 3200 Ma old black smokers of the Sulphur Springs Formation in Western Australia may be the remains of anaerobic hyperthermophile bacteria, though the indigenous and biogenic nature of these intriguing structures has yet to be demonstrated beyond question. The oldest cherts containing a supposed diverse prokaryotic microflora (Figure 4A) come from the 3450 Ma old Apex Cherts, which are intimately associated with ultrabasic and basaltic lava flows of the Apex Basalt in the Warrawoona Group of Western Australia. At least eleven different kinds of filamentous microfossil have been described from these rocks, some of which have been compared with cyanobacteria (Figure 4A). This has been taken to suggest that photosynthesis had begun to release oxygen into the atmosphere by 3450 Ma and that a substantial amount of evolution had taken place by this time. A critical re-examination of the context and fabric of these cherts suggests, however, that all these
PRECAMBRIAN/Prokaryote Fossils 369
Figure 4 Pseudofossils. (A) Montage of the pseudofossil Archaeoscillatoriopsis disciformis holotype from the 3450 Ma old Apex Chert of Western Australia, showing (in black and white) its original interpretation and (in colour) its much more complex morphology when montaged (with topographic map) and its proximity to crystal growths. (B) Spherulites formed by the recrystallization of glassy hydrothermal silica with iron oxide impurities; recrystallization causes the formation of a spectrum of forms that range from abiogenic to microfossil like pseudofossils; scale bar in B 40 mm. (C) Complex carbonaceous artefacts formed in the laboratory; scale bars 40 mm. (D) Complex stromatolitic growths can easily be formed artificially by over spraying a surface with layers of paint; width of view 12 cm.
structures occur deep within a cross-cutting hydrothermal vein. They comprise a continuous spectrum of arcuate to complexly branched carbonaceous reaction rims, formed around the edges of spherulitic silica botryoids, that appear to have formed during the recrystallization of hydrothermal silica glass to chalcedony and around rhombic crystal margins. Doubt is also cast on other ‘microfossils’ of about this age from the Warrawoona Group of Australia and from the Barberton mountains of South Africa. Their origin as mineralic pseudofossils or contaminants remains an open question.
Criteria for Biogenicity The search for the signals left in rocks by very ancient prokaryotes requires a checklist of criteria to help establish biogenicity, not least because prokaryotic
cells are so easily mimicked by physical artefacts and so readily introduced by contamination. A previous formula for the recognition of bona fide Archean microfossils has been that the putative microfossil structures must be unquestionably biogenic remnants that are indigenous to and syngenetic with the primary deposition of a sedimentary rock, which is itself of known stratigraphic and geographical source and of established Archaean age (i.e. greater than 2500 Ma old). Until recently, the above criteria have been used to contrast the questionable ‘microfossils’ in Martian meteorite ALH 84001 with the 3450 Ma old supposedly bona fide ‘microfossils’ from the Apex Chert. It is now necessary to acknowledge, however, that complex structures of organic composition, like those in the Apex Chert, can be assembled very easily (e.g. Figure 4C), including abiogenic artefacts of
370 PRECAMBRIAN/Prokaryote Fossils
isotopically light carbon formed through late-stage remobilization of hydrocarbons with hydrothermal silica. Complexity per se is not a valid criterion for the recognition of life. Biological populations may be distinguished from abiogenic ones by the lack of a continuous spectrum of morphological variation, limited by the DNA-controlled genetic constraints. This narrow range can be contrasted with complex abiogenic artefacts, which will tend towards a ‘symmetry-breaking cascade’ of forms, ranging from a few biological look-alikes with rotational symmetry (Figure 4C) to highly information-rich forms (Figures 4A and 4B) with poor symmetry. Research into the earliest biosphere is now served by a marvellous range of new tools, including molecular (RNA and DNA) analysis of living-prokaryote phylogeny, biogeochemical analysis of ancient Earth rocks, and digital-image analysis of putative microfossils. These techniques should not be allowed to prosper, however, at the expense of more traditional techniques such as mapping and thin-section petrography. The latter are also vital for providing a context for early-life studies at a range of scales, from satellite images of greenstone belts, through field mapping, to microfabric mapping of thin sections. Future research will also require a much better understanding of the nature of complexity itself and of the multitudinous ways in which seemingly complex structures can self-organize. (Figure 4 shows a range of microfossil-like and stromatolite-like objects of inferred non-biological origin.) This better understanding of complexity is needed, not only to understand the ways in which pseudofossils and pseudo-stromatolites can form, but also to comprehend the conditions in which life-forms and their building blocks were able to assemble and develop on the early Earth, and perhaps beyond.
See Also Biosediments and Biofilms. Earth Structure and Origins. Origin of Life. Precambrian: Eukaryote Fossils. Pseudofossils. Sedimentary Rocks: Banded Iron Formations. Solar System: Mars.
Further Reading Brasier MD, Green OR, Jephcoat AP, et al. (2002) Ques tioning the evidence for Earth’s oldest fossils. Nature 416: 76 81. Farquhar J, Bao HM, and Thiemens M (2000) Atmospheric influence of Earth’s earliest sulfur cycle. Science 289: 756 768. Furnes HR, Banerjee NR, Muehlenbachs K, et al. (2004) Early life recorded in Archean pillow lavas. Science 304: 578 581. Garcia Ruez JM, Hyde ST, Carnerup AM, et al. (2003) Self assembled silica carbonate structures and detection of ancient microfossils. Science 302: 1194 1197. Hofmann HJ, Grey K, Hickman AH, and Thorpe RI (1999) Origin of 3.45 Ga coniform stromatolites in Warrawoona Group, Western Australia. Bulletin of the Geological Society of America 111: 1256 1262. Kazmierczak J and Altermann W (2002) Neoarchean bio mineralization by benthic cyanobacteria. Science 298: 2351. Kerr R (2004) New biomarker proposed for earliest life on earth. Science 304: 503. Knoll AH (2003) Life on a Young Planet. Princeton, NJ: Princeton University Press. Mojsis SJ, Arrenhius G, McKeegan KD, et al. (1996) Evi dence for life on Earth before 3,800 million years ago. Nature 384: 55 59. Rasmussen B (2000) Filamentous microfossils in a 3,250 million year old volcanogenic massive sulphide. Nature 405: 676 679. Rothman DH and Grotzinger JP (1996) An abiotic model for stromatolite morphogenesis. Nature 383: 423 425. Schopf JW (1999) The Cradle of Life. New York: Princeton University Press. Simpson S (2003) Questioning the oldest signs of life. Scientific American April 2003: 70 77. Westall FM, de Wit MJ, Dann J, et al. (2001) Early Archean fossil bacteria and biofilms in hydrothermally influenced sediments from the Barberton greenstone belt, South Africa. Precambrian Research 106: 93 116. Woese CR, Kandler O, and Wheelis ML (1990) Towards a natural system of organisms: proposals for the domains of Archaea, Bacteria and Eucarya. Proceed ings of the National Academy of Sciences USA 87: 4576 4579.
PRECAMBRIAN/Vendian and Ediacaran 371
Vendian and Ediacaran M A S McMenamin, Mount Holyoke College, South Hadley, MA, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The geological, palaeobiological, and earth systems (i.e., global geochemical) events associated with the Vendian period set the stage for all subsequent Earth history. Unprecedented change took place in the biosphere at this time, and the Vendian witnessed the appearance of the first Ediacarans, the first shelly fossils, and the earliest known animals. The Vendian saw the continuation of breakup of the earliest known giant supercontinent (Rodinia) and, at the end of the period, continuing into the Cambrian, the amalgamation of the supercontinent Gondwana. The Vendian also witnessed the termination of the worst series of glaciations known. The climate change was accompanied by dramatic fluctuations in the records of carbon and strontium isotopes. Linkages between supercontinent breakup and extreme climate change, and between biotic diversification and ecological change (e.g., as driven by metazoan disturbance of the marine substrate), are being actively explored by researchers following diverse avenues of investigation. The Vendian period and system (called by some authors Lipalian or Ediacaran) spans the interval from 600–543 Ma. It was the final period of Proterozoic time (Neoproterozoic). Some accounts place the beginning of the Vendian as far back as 670–620 Ma, but this would push it into what should be the preceding period (the Sinian, 680–600 Ma). In any case, the Vendian represents the last interval of Precambrian time, and its end marks the most important division in the geological time-scale, the Proterozoic–Cambrian boundary. The importance of this boundary has led to escalating interest amongst researchers, and great advances have recently been made in understanding the Vendian. Many questions remain, however, and amongst them are some of the most contentious issues in contemporary earth science. The primary questions of interest are threefold. First, what caused global climate to warm after the worst glaciation on record? Second, what are the Ediacarans? Third, what events triggered the so-called Cambrian Explosion (see Palaeozoic: Cambrian) and the appearance of familiar animal types? A host of unsolved secondary questions follow. These include: What was the makeup of supercontinent Rodinia? What was the timing sequence and geometry
of its breakup? What were the controls on eustatic sealevel change during the Vendian? Was there a mass extinction at the Proterozoic–Cambrian boundary? If so, what was its cause? What was happening to global geochemistry during this time? The implications of these questions are far reaching, and the discussion that follows is divided into three sections: Geological Events, Palaeobiological Events, and Earth System Events.
Geological Events It can be said that the prelude to the Vendian world began at 1000 Ma with the amalgamation of the supercontinent Rodinia. The continental collisions that led to the formation of the supercontinent are generally called the Grenville Orogenic Event (see Grenvillian Orogeny). Named for 1000 Ma rocks that record an ancient episode of mountain building and continental collision in North America, Grenvillian rocks have now been recognized in places as distant as northwest India and Antarctica. Composed of all the large continents of the Precambrian planet, Rodinia began to split apart beginning at about 750 Ma. Some geologists argue for a failed attempt at breakup at 850 Ma. Rodinia is thus a remarkably long-lived supercontinent, lasting some 250 million years. The rifting event was a drawn out affair that seems to have proceeded in two stages. Stage one began with the opening of the Pacific Ocean, as parts of what now constitute eastern Gondwana (Australia, Antarctica) split apart from what is now the west coast of North America. In stage two, fragments connected to the north-eastern and eastern parts of North America split off and began to collide with the continental blocks already set free by continental drift. The net result of all this continental motion, which was not complete until about the time of the Cambrian boundary, was the reorganization of the continental plate geometry of the planet with the sundering of Rodinia and the amalgamation of Gondwana. Some geologists believe that a short-lived global supercontinent (Pannotia) existed between the times of Rodinia and Gondwana, but evidence for this supercontinent is tenuous at best. Block faulting and volcanicity associated with tectonic sundering characterize many Vendian stratigraphical sections. Deposition of the Vendian Tindir Group of east-central Alaska was influenced by block faulting in North American basement rocks. The Vendian La Cie´ nega Formation of northern
372 PRECAMBRIAN/Vendian and Ediacaran
Mexico is punctuated by layers of porphyritic basalt erupted from fissures presumably related to tectonic extension. Palaeomagnetic evidence suggests that the rates of continental motion near the Cambrian boundary were amongst the fastest ever measured. At one point near the boundary, the continents appear to have been moving synchronously, prompting what has been called inertial interchange theory, in which the Earth’s crust is thought to have detached from the mantle for a relatively short time. As the fragments of Rodinia went their separate ways, they set into motion a sequence of events that ultimately resulted in the drowning by seawater of all or most continental margins. As Rodinia split apart, the Rodinia’s counterpart superocean, Mirovia, began to vanish beneath the subduction zones along the leading edges of the dispersing continental fragments. This led, overall, to an exchange of old, cold Mirovian seafloor crust for the relatively hot, less dense, and hence more buoyant ocean crust of the newly developing rift basins on the trailing edges of the continents opposite the subduction zones. Seawater was consequently displaced on a massive scale by buoyant, new mid-ocean ridges, and this led to the great marine transgression that began the Palaeozoic era. The transgression began very slowly, as the early stage of breakup involved only a single (albeit long) continental margin, namely the split between Australia–Antarctica and North America. This gradual rise was then suddenly punctuated by a rapid fall in eustatic sea-level due to the onset of the first of several (Sturtian and Marinoan) Late Proterozoic glaciations. These were the worst glaciations known, leading many to believe that the Earth must have passed through what has been called ‘White Earth’ or ‘Snowball Earth’ conditions. The timing of these glaciations is somewhat uncertain, although two groups, an older (Sturtian, including the Rapitan, Chuos, and Stuartian glaciations) and a younger (Marinoan, including the Ice Brook, Ghaub, and Elatina glaciations), appear to each consist of roughly synchronous glacial events. Compelling evidence exists during both glacial episodes for glaciers on land at sea-level at the equator, and some geologists suggest that the ice cap extended into the equatorial ocean as well. Regardless of the true extent of the ice cap, the glaciation was tremendously severe. The climatic aberration is made all the more curious by the character of the sedimentary layers deposited directly above the glacial deposits. Also, sedimentary (banded) iron formations reappeared during these ice ages after having virtually disappeared from the record for 1.5 Ga. These and
other geochemical anomalies are interpreted to suggest almost incredible changes in the oxygen and carbon dioxide cycles of the planet – anoxia under the marine ice cap, leading to banded iron deposition; build-up of sufficient carbon dioxide in a life-depleted Earth to eventually trigger greenhouse conditions and melt the glaciers; and supergreenhouse conditions and supersaturation, leading to massive deposition of cap carbonates under unusual carbonate depositional conditions. The cap carbonates consist primarily of abiogenically precipitated calcite and dolomite. Carbonate rocks of this nature are usually associated with deposition under very warm climatic conditions. Therefore, the juxtaposition of these sediments directly above deposits of the worst known glaciation is unusual in the extreme. The last Proterozoic cap carbonates were deposited at approximately 600 Ma and define the base of the Vendian system. With the final melting of the ice, the transgression resumed its flooding of continental shelf areas. The transgression continued essentially unabated until well into the Cambrian. In many stratigraphical sections throughout the world, the base of the Cambrian is marked by a basal unconformity, although this is by no means universal, particularly in regions with a more or less complete Vendian section as well. The Vendian is thus defined in many places by a cap carbonate at its base (a pronounced aid to lithostratigraphical correlation when not associated with an unconformity), but an unconformity at its top (a decided hindrance to correlation efforts). Nevertheless, a combination of radiometric dates, lithostratigraphy, biostratigraphy, palaeomagnetic stratigraphy, and carbon and strontium isotope stratigraphy has rendered preliminary correlations possible between Vendian sections on different continents. Much work remains to be done to refine these correlations. The Vendian is not characterized by major extraterrestrial impact events, but this may be an artefact of a less well-understood record of mass extinction during the period. The Acraman impact site (570 Ma) of South Australia, associated with a 160 km diameter crater, shock metamorphism, and shattercone development, is comparable in size to the 214 Ma, Manicouagan impact structure of Canada.
Palaeobiological Events Although eukaryotic organisms are thought to have existed for more than 1000 million years before the beginning of the Vendian, they are rather inconspicuous until the Vendian begins. Thus, the Vendian marks the beginning of the Phanerozoic, the age of
PRECAMBRIAN/Vendian and Ediacaran 373
visible life. The reason for the rapid expansion of eukaryotic life at this time is unclear, but some researchers have linked it to the climatic amelioration following the Proterozoic glaciations. Biogenic stromatolites, known from the oldest rocks bearing microbial fossils, dramatically changed texture during the Vendian. The concentric lamination that characterized more ancient stromatolites gave way to a clotted, thrombolitic texture. This transition from stromatolite to thrombolite has been attributed to the burrowing and lamination-disturbing activities of early animals. Biological inferences regarding the origin of metazoa, based on molecular clock data, indicate that metazoa appeared by at least 1000 Ma, but these inferences are beset by controversy. The date of origin of the animal kingdom is controversial as well, and the rock record before the Vendian does not provide many unambiguous clues to metazoan origins. One thing is clear, however; by the Middle to Late Vendian, animals were present, as indicated by their trace fossils (see Trace Fossils). These trace or ichnofossils show evidence of peristaltic burrowing and displacement of sediment by burrowing activity, indicating that, by the end of the Vendian, animals with hydrostatic skeletons (coelomic spaces) were well established in the marine biosphere. A difficulty with the study of Vendian animals is the fact that, apart from some phosphatized embryos that cannot be identified confidently to phylum, and some fossil sponges with preserved spicules, actual body fossils of these animals are rare. A comb-like structure from Russia, called Redkinia, may represent the flexible, filter-feeding mouthpart of a Vendian animal.
However, apart from this, there are very few Vendian animal body fossils, particularly body fossils of the burrowing tracemakers. The Cloudinidae consists of Vendian shelly fossils. The cone-in-cone tubular shells of genera such as Cloudina, Sinotubulites (Figure 1), and Wyattia are presumed to have been formed by worm-like animals. Cloudinids are occasionally associated with a bizarre, weakly calcified, stalked, goblet-shaped organism called Namacalathus. The biological affinities of cloudinids and namacalathids are not well understood, but they do seem quite unlike the more familiar skeletonized animals of the Early Cambrian. An algal affinity for the two has been suggested and cannot be ruled out. Large pores in the calyx of Namacalathus were probably filled with soft tissue in life; the pores might have served to admit light into the organism to sustain photosymbionts. Ediacarans are the most puzzling part of the Vendian biota. These bizarre creatures grew to enormous sizes by Vendian standards, with some of the frondose forms reaching 2 m or more in length. The Ediacaran body is non-skeletal and seems to have been formed of a tough integument, in some cases partitioned into modules of similar shape. Unlike most other soft-bodied creatures, Ediacarans were capable of being preserved in sandstones. Ediacarans appear to be multicellular, but even this inference has been subjected to dispute. The concept of metacellularity (i.e., a body composed of uni- or polycellular partitions called metacells) has been applied to Ediacarans with some success. Many palaeontologists have assigned these forms to conventional animal phyla, such as the Cnidaria,
Figure 1 Sinotubulites cienegensis. An early shelly fossil, this cloudinid is from the La Cie´nega Formation of Sonora, Mexico. Holotype specimen. Length of largest tube, 12 mm.
374 PRECAMBRIAN/Vendian and Ediacaran
Echinodermata, or Annelida, but the arguments supporting these assignments have not convinced everyone. Sceptics point out that not a single uniquely animalian trait has been identified on any of the thousands of Ediacaran fossils collected so far. A consensus for at least some of the fossils may be emerging, however, with recent new evidence showing trilobitoid arthropod features in Spriggina, and with the association of Kimberella with grazing traces. In addition to the difficulties with systematic placement, there is the thorny problem of Ediacaran preservation. None of the Ediacaran fossils have skeletons, with the possible exception of the cloudinids and the namacalathids (whose relationship to the other Ediacarans is not known). Their cuticle also appears to have been relatively soft and flexible. Many of the Ediacaran fossils are preserved in relatively coarse sandy sediment, not ordinarily considered to be a good substrate for the preservation of non-skeletal organic remains. Thus, the preservation of Ediacarans must be explained either by unusual properties of the cuticle itself, or by anactualistic processes on the Vendian seafloor that could account for the preservation of soft tissue in sandstone. The latter possibility has evoked what is called the ‘death mask’ hypothesis, namely the idea that microbial mats in some way hardened the surface of dead Ediacarans, allowing them to be preserved as fossils. Unfortunately, the death mask theory is invoked by its supporters to explain all aspects of Ediacaran taphonomy – for
example, the death mask mats supposedly formed beneath layers of storm sand and both above and below living Ediacarans, three highly uncertain propositions. A key bedding plane surface from Newfoundland provides a new perspective on the death mask controversy (Figure 2). This bedding surface was smothered by a volcanic ash fall, entombing specimens of an unnamed spindle-shaped Ediacaran and a stalked form called Charniodiscus. The Charniodiscus was superimposed over the spindle form, and yet the morphology of the spindle creature shows clearly through the cuticle of the Charniodiscus. Thus, the death mask hypothesis is falsified, as there was no space for a post-mortem microbial mat to form between the Charniodiscus and the spindle. If the death mask hypothesis were correct, only the overlying Charniodiscus morphology should be preserved, but this is clearly not the case in this example. Consequently, we are left with the idea that Ediacarans bore an unusually resilient cuticle. There is much going for this concept, as a link between Ediacarans and a problematic group known as conulariids (first suggested in 1987) has been dramatically confirmed by the discovery of the probable conulariid Vendoconularia triradiata in Vendian strata of the Ust’-Pinega Formation of the Onega River region, Russia. Conulariids appear to be a group of Ediacarans that survived well into the Palaeozoic. As with the Ediacarans themselves, most attempts at classifying
Figure 2 Cast of specimen of Charniodiscus superimposed over a spindle shaped form from the Mistaken Point assemblage in Newfoundland. Fibreglass cast (Pratt Museum, Amherst College). Width of view, approximately 11 cm.
PRECAMBRIAN/Vendian and Ediacaran 375
conulariids have tried to assign them to the Cnidaria, and have met with failure. Part of the problem seems to be that conulariids display, as do other Ediacarans, both triradial and fourfold radial symmetry. This translates in metacellular terms to the unipolar iteration of six, eight, twelve, or sixteen founding metacells. The phosphatic nature of the (also somewhat flexible) conulariid cuticle may provide an important clue to the nature of the Ediacaran cuticle. The most difficult Ediacaran problem is that of body geometry. Attempts to classify Ediacarans as animals have been hindered by their unusual constitution, which can show combinations of triradial symmetry and glide symmetry (as in the genus Pteridinium; Figure 3); these have proven to be impossible to fit into a conventional metazoan body plan model. Some researchers have tried to break the impasse by arguing that Ediacarans are colonial communities of individual animals. This explanation is not adequate to the task either, as each Ediacaran seems to be a well-integrated individual holobiont rather than a loose collection of individuals in a colony. Furthermore, there is no evidence for the loss of morphological features in specialized individuals, as is the case in the highly integrated colonies of modern hydrozoans. The concept of metacellularity may help to solve this morphological problem. Some Ediacarans are apparently modular partition creatures, and the
individual units or partitions, referred to as metacells, are either uni- or polycellular pods that may be flattened, stretched, inflated, or repeated (to form an iterated chain of metacells) as required. Metacellular creatures of this style are known amongst the modern biota (e.g., characean pondweeds, such as Nitella, can have individual cells measuring up to 15 cm in length), but no other type of organism has explored the potential of metacellularity as thoroughly as have the Ediacarans. The concept of metacellularity seems to apply best to the Ediacarans of the Mistaken Point biota in Newfoundland, the morphologies of which (spindle forms, branch forms, pectinate forms) are strange even by Ediacaran standards. The metacellularity concept does not solve the problem of whether Ediacarans are animals; for example, hexactinellid (glass) sponges can undergo a coenocytic or syncytial (plasmodial) stage that is quite uncharacteristic for animals. Metacellularity is unknown in living animals and, for this reason, a number of palaeontologists prefer to place at least some of the Ediacarans into their own Kingdom, the Vendobionta. These Ediacarans could conceivably bear a closer relationship to several protoctist groups (such as the xenophyophores) than they do to animals. A common ancestry with the protoctist group that gave rise to the animals is still plausible, though, and may even be likely. Even Ediacarans that would seem to have a more animalian character, such
Figure 3 Pteridinium simplex. Cast of specimen from the Nama Group, Namibia. The fossil is viewed from its underside; the undersides of two vanes are visible, and the basal part of the third vane is visible between them as a stitch like chain of tube ends. Scale bar in centimetres.
376 PRECAMBRIAN/Vendian and Ediacaran
Table 1 Ediacaran body forms, classified by metacellular grade Original number of metacells
Founding metacell(s) only, no iteration
Unipolar iteration of metacells
Bipolar iteration of metacells
One Two Three Four Five
Cyclomedusa, Evandavia, Parvancorina (?) Gehlingia Tribrachidium, Anfesta, Albumares Conomedusites Arkarua
Charniodiscus, Charnia Dickinsonia, Phyllozoon Swartpuntia Spriggina (?), Marywadea (?) Rangea (?)
Windermeria, Ernietta Pteridinium simplex Pteridinium carolinaensis
Spindle form of Newfoundland
Unknown
Figure 4 Evandavia aureola. This hiemalorid Ediacaran occurs in the Clemente Formation, Sonora, Mexico. It is perhaps the oldest known (approximately 600 Ma) specimen of a complex life form. Scale bar in centimetres.
as Yorgia waggoneri from the White Sea region in Russia, show strange asymmetries in the ‘head’ region that challenge the animal interpretation. Many Ediacarans may be placed in the metacellular grouping (Table 1). These forms manifest both a smooth central metacell surface and a surface formed of radial (and often bifurcating) tubes. Evandavia aureola is a non-iterated discoidal form from Sonora, Mexico, with a single metacell and radial tubes in its outer ring (Figure 4). Figures 5 and 6 show Parvancorina minchami from Australia, displaying a single trifold metacell with numerous branches coming from each branch of the initial metacell. Parvancorina serves as the subunit counterpart to Tribrachidium and Gehlingia, with the ‘thumb structures’ in the bilaterally symmetric Gehlingia and in the triradiate Tribrachidium homologous to the main medial branch in Parvancorina (Figure 7). Similar metacellular rearrangements may be observed in another Ediacaran clade, which
includes the Newfoundland spindle-shaped form (Figure 8) and a 2 m long Newfoundland frond (‘Charnia’ wardii) shaped like an extremely elongate primary feather of a bird’s wing. An interesting variation on the discoid body theme is a Chinese fossil from the Xingmincun Formation, southern Liaoning Province, consisting of a single, elongate metacell helically coiled into a flat disc. These specimens reach up to 4 cm in diameter.
Earth System Events Eukaryotic phytoplankton are known from strata approaching 2 Ga in age, but, by the Vendian, these organisms began to resemble the modern phytoplankton at least in terms of diversity of tests, skeletons, and sheaths. Siliceous chrysophyte algae are known from Vendian strata of the Tindir Group, Alaska, and genera such as Chilodictyon and Characodictyon share a superficially diatom-like aspect.
PRECAMBRIAN/Vendian and Ediacaran 377
Figure 5 Parvancorina minchami. A cast of a specimen from the Pound Supergroup of South Australia. Greatest dimension of specimen, 10 mm.
Figure 6 Parvancorina minchami. This Ediacaran poses difficul ties for the arthropodan interpretation of this genus. Greatest dimension of specimen, approximately 10 mm.
In contrast, however, the Vendian seafloor had a distinctly primitive character. Microbial mats still blanketed the seafloor, as they had been doing since the Archaean, and stromatolites remained abundant even as burrowing began to alter their internal texture to the thrombolitic state. A wide variety of unusual sedimentary structures (with names such as
Arumberia, Kinneya, and ‘elephant-skin’ texture) are known from fine clastic rocks of the Vendian. Lozenge-shaped structures (Figure 9) are often found on Vendian bedding surfaces; whether or not such structures might have use as a basis for interbasinal correlations is unknown. These primary sedimentary structures are puzzling, leading sedimentologists to invoke anactualist sedimentary processes to explain them. It seems quite reasonable to do so, as microbial mat carpeting of the seafloor would certainly influence the nature of marine sedimentation. Preston Cloud once called stromatolites ‘organosedimentary structures’, and the shelf, slope, and rise sediments of the Proterozoic may be thought of as a gigantic, connected, organo-sedimentary structure. Judging from the diversity of bedding plane texture types, never before or since have microbes had such a direct and intimate association with the basic processes of clastic sedimentation. Ediacarans evidently adhered to the surface of these mats or (in the case of forms with holdfasts and possibly in the case of Pteridinium) lived beneath them. Metazoan burrowers apparently began their excavations beneath this mat surface, and evidence suggests that the earliest ichnofossil makers were submat burrowers. A good example is the Vendian trace fossil, Vermiforma, from the Carolina Slate Belt. This enigmatic and relatively large trace fossil, associated with other types of submat burrowers, consists of ten specimens that all follow the same rather tortuous track. As the traces are separated from one another by some distance, all underneath the mat, it is hard to
378 PRECAMBRIAN/Vendian and Ediacaran
Figure 7 Ediacaran homology. Homologous structures (‘thumb structures’) linking the Ediacaran genera Parvancorina (left: length, 1 cm), Gehlingia (right: length, 8 cm), and Tribrachidium (centre: diameter, 1 cm). In spite of the vast differences in body symmetry type, the three genera are seen to be closely related.
Figure 8 Spindle shaped form from the Mistaken Point assemblage in Newfoundland. Fibreglass cast (Pratt Museum, Amherst College). Scale bar in centimetres.
imagine how and why the track paths were coordinated in shape. Arguments that Vermiforma is a pseudofossil are called into question by the presence of other types of trace fossils on the same bedding plane surface. Throughout its duration the Vendian is marked by steadily increasing levels of bioturbation. The top of the Vendian system is currently defined by the appearance of the three-dimensional trace fossil Trichophycus pedum. The maker of this trace was able to excavate vertically as well as horizontally, and its activities (along with those of other vertical burrowers, such as the makers of Skolithos) tended to homogenize seafloor sediments. It would also presumably have punctured and shredded any microbial mats in the immediate vicinity. It is thought that, towards the end of the period, the seafloor mat seal began to break down due to the intensity of metazoan burrowing activity. Such uncapping of the seafloor had dramatic consequences, both for the carbon budget of the planet (e.g., buried
carbon was put immediately back into circulation) and for marine nutrient levels (e.g., sediment grains in suspension make wonderful substrates for nutritious bacterial growth). At about the same time, there was a tremendous flux of mineral nutrients to the oceans resulting from the fact that the Vendian saw a rare tectonic coincidence in Earth’s history: the simultaneous occurrence of both divergent tectonics (final breakup of Rodinia) and convergent tectonics (formation of Gondwana, in an event known as the Pan-African Orogeny (see Africa: Pan-African Orogeny)), on a massive scale. As suggested by marine strontium isotopes of the Vendian, huge amounts of siliciclastic debris were shed from rift valley margins into the Vendian ocean. This would have been added to the volcaniclastic sediments derived from weathering of rift-associated basalts. The enhanced pool of igneous rock debris contributed mightily to an oceanic fertilization event that has been implicated in the emergence of skeletonized animals and the Cambrian explosion.
PRECAMBRIAN/Vendian and Ediacaran 379
Figure 9 Lozenge shaped structures from the Clemente Formation, Sonora, Mexico, found in association with Evandavia aureola and other Ediacarans. Such structures are known from Ediacaran bearing strata on other continents. Scale bar in centimetres.
It appears that a massive palaeoecological reorganization of the marine biosphere took place at the end of the Vendian. With the exception of some very Late Vendian evidence for predatory activity in Sonora, Mexico, no evidence for large predators has been recognized in Vendian strata. In contrast, the Early Cambrian exhibits evidence for a great intensity of macropredation, involving forms such as the 2 m long predatory animal Anomalocaris. The Vendian biosphere has thus been referred to as a uniquely peaceful Garden of Ediacara, where large creatures, many with flattened bodies, partook of sunlight (via photosymbiosis), hydrogen sulphide (via chemosymbiosis), or osmotrophy (direct absorption of dissolved nutrients in seawater) on a mat-covered seafloor surface free of large predators. However, it was too good to last, and animals in the role of burrowers from below and predators from above triggered a rapid end to the Garden and the end of the Vendian. The question of whether or not there was a mass extinction at the end of the Vendian is unresolved at present. A number of Ediacarans appear to have survived the Cambrian boundary, occurring with Cambrian fossils in Australia, Ireland, and elsewhere. Studies based on borehole data from the oilproducing strata in Oman have recently shown that cloudinids vanish from the record without a trace, and that this disappearance is not associated with any discernible lithological change. Thus, the disappearance of cloudinids from Oman cannot be attributed to environmental or facies change. Their loss has
thus been interpreted as a major extinction event. Whether or not this localized loss of cloudinids represents a mass extinction, or an extinction event at all, is not known. Cloudinid-like fossils have been reported from the Cambrian Tornetra¨ sk Formation in northern Sweden. Secular variations in secular isotopes in the Vendian are rather difficult to interpret, but, after a slight decrease from 3% d13C to 5% d13C after the Marinoan glaciation (Canadian and Namibian sections), the d13C values during the Vendian appear to undergo a fairly steady rise to a value of over þ3% right before the Cambrian boundary (Dvortsy section, Aldan River, Siberia). The Vendian–Cambrian boundary itself is marked by a sudden þ3% to 1.4% drop in d13C over a very short stratigraphical interval, as measured at the Dvortsy section. This excursion has been linked to a variety of factors, including some sort of global environmental perturbation that also triggered the extinction of the cloudinids and many of the Ediacarans. Another way to look at it, though, is that the boundary excursion represents that moment in geological time when marine burrowing intensity crossed a threshold. At this critical point, the microbial mat seal on the seafloor may have been breached, resulting in previously immobile sediments (and their associated organic matter) becoming mobilized and injecting huge amounts of biogenic carbon into the water column. Owing to biogenic isotopic fractionation, this detrital organic matter was significantly depleted
380 PRECAMBRIAN/Vendian and Ediacaran
in the heavy isotope of carbon and, as it went into circulation in marine water, caused the precipitous drop or boundary excursion in the d13C value. It therefore seems reasonable to interpret the Vendian carbon isotope curve as a record of gradually increasing biotic productivity, with sequestering of much of the organic matter within and below the seafloor microbial mats, followed by a relatively sudden release of part of this organic matter deposited at the end of the Vendian as a result of increased burrowing intensity. Other factors, such as destabilization of gas hydrates in seafloor sediments, may also have been involved in these isotopic excursions. A number of other phenomena can perhaps be traced to what has been called the Cambrian Substrate Revolution. The Vendian and Cambrian both saw an increase in the proportion of calcified filamentous microbes (such as Girvanella), which were perhaps less palatable to mat grazers than filamentous cyanobacteria and unprotected algae. With all the new sediment and organic matter in suspension, filter feeding probably became more possible throughout the water column, leading to the evolution of the first tiered filter feeders in the Cambrian. The only Vendian organisms that were likely to engage in suspension feeding were the cloudinids, which lived close to the sediment–water interface. Finally, assuming that we are interpreting the secular carbon isotopic curve correctly, it is entirely possible that oxygen levels increased in the Vendian due to the sequestration of organic matter. Whether or not increasing oxygen levels influenced metazoan evolution is not known, although it seems fair to say that early burrowing animals would not have required high levels of oxygen. The earliest animal habitat appears to have been the submicrobial mat environment, where oxygen levels would probably have been rather low considering the relative abundance beneath the mats of hydrogen sulphide and other reduced compounds.
Glossary abiogenically A term applied to rocks formed by processes not directly influenced by living organisms. anactualistic processes Processes that occurred at one time in the Earth’s past, but which are no longer operational today. cloudinid A late Vendian calcareous shelly fossil consisting of closely nested, thin-walled tubes or cones. Thought to represent one of the earliest examples of a shelly animal fossil. Includes the genera Cloudina and Sinotubulites.
coelomic spaces The compartments that house the rigid, fluid-filled body cavity present in many animals. The coelom serves as a hydrostatic skeleton. conulariid Any member of an enigmatic group of Vendian/Cambrian to Triassic shelly organisms. They formed conical, often pyramidal, tapering cones with transverse ribbing, composed of calcium phosphate. Ediacaran Any member of a group of marine, megascopic fossils with a metacellular growth pattern. Found primarily in strata deposited before the Cambrian period. Assigned to extinct Kingdom Vendobionta. frondose forms Ediacarans with a leaf, palm, or frond body form. Garden of Ediacara A palaeoecological theory that holds that the marine ecosystems of the Vendian were largely free of megascopic predators and thus allowed organisms such as Ediacarans to survive unmolested using photosymbiotic, chemoautotrophic, and osmotrophic life styles. holobiont A single integrated organism, as opposed to a colonial organism. hydrostatic skeleton A fluid-filled internal organ or support structure within an animal’s body that can be kept rigid or made limp by control of internal water pressure. metacell A single or isolated modular unit of a metacellular organism; usually consists of a single enlarged cell. metacellularity Term applied to organisms that are either multicellular (such as animals and plants) or consist of clusters or metacells (such as Ediacarans and certain types of aquatic algae). Mirovia The Precambrian superocean that surrounded Rodinia. molecular clock Any gene or gene sequence used by biologists in an attempt to determine the evolutionary time of divergence from a common ancestor between two or more groups of organisms belonging to different species. osmotrophy A feeding strategy utilizing osmosis or direct absorption of nutrients. peristaltic burrowing A burrowing strategy in metazoans that consists of rhythmic muscular contractions along the length of the body. Rodinia A supercontinent consisting of all or nearly all of the continents. Consolidated one billion years ago (in an event referred to in North America as the Grenville Orogeny), this supercontinent broke up into smaller continents by the process of plate tectonics and continental drift before the Cambrian. Sinian The Precambrian geological period immediately preceding the Vendian period.
PRECAMBRIAN/Vendian and Ediacaran 381
Snowball Earth An extreme phase of glaciation in which glaciers reached tropical latitudes. thrombolitic texture A texture in sedimentary rocks characterized by disrupted bedding lamination and a clotted fabric. Stromatolites develop this texture when influenced by burrowing metazoa. Vendian The latest Precambrian geological period, immediately preceding the Cambrian period. Synonymous with Lipalian or Ediacaran. Vendobionta The extinct kingdom to which Ediacarans are assigned.
See Also Africa: Pan-African Orogeny. Australia: Proterozoic. Biosediments and Biofilms. Grenvillian Orogeny. Palaeozoic: Cambrian. Precambrian: Overview; Eukaryote Fossils. Trace Fossils.
Further Reading Bottjer DJ, Hagadorn JW, and Dornbos SQ (2000) The Cambrian substrate revolution. GSA Today 10: 1 7. Crimes TP (1999) Review of Garden of Ediacara. Palaeo geography, Palaeoclimatology, Palaeoecology 150: 357 358. Crimes TP and Fedonkin MA (1996) Biotic changes in platform communities across the Precambrian Phanerozoic boundary. Rivista Italiana di Paleontologia e Stratigrafia 102: 317 332. Crimes TP, Insole A, and Williams BPJ (1995) A rigid bodied Ediacaran biota from Upper Cambrian strata in Co. Wexford, Eire. Geological Journal 30: 89 109. Donovan SK and Lewis DN (2001) The Ediacaran biota. Geology Today 17: 115 120. Glaessner MF (1984) The Dawn of Animal Life. Cam bridge: Cambridge University Press. Hoffman PF and Schrag DP (2002) The snowball Earth hypothesis: testing the limits of global change. Terra Nova 14: 129 155. Ivantsov AYu (1999) A new dickinsoniid from the Upper Vendian of the White Sea Winter Coast (Russia, Arkhan gelsk Region). Paleontological Journal 33: 211 221. Ivantsov AYu and Fedonkin MA (2002) Conulariid like fossil from the Vendian of Russia: A metazoan clade across the Proterozoic/Palaeozoic boundary. Palaeon tology 45: 1219 1229. McMenamin MAS (1987) The fate of the Ediacaran fauna, the nature of conulariids, and the basal Paleozoic
predator revolution. Geological Society of America Abstracts with Program 19: 29. McMenamin MAS (1990) Vendian. In: Briggs DEG and Crowther PR (eds.) Palaeobiology: A Synthesis, pp. 179 181. Oxford: Blackwell Scientific Publications. McMenamin MAS (1996) Ediacaran biota from Sonora, Mexico. Proceedings of the National Academy of Sciences (USA) 93: 4990 4993. McMenamin MAS (1998) The Garden of Ediacara: Dis covering the First Complex Life. New York: Columbia University Press. McMenamin MAS (2000a) The antiquity of life: From life’s origin to the end of the Vendian Period. In: Margulis L, Matthews C, and Haselton A (eds.) Environmental Evo lution, 2nd edn. pp. 158 169. Cambridge, MA: MIT Press. McMenamin MAS (2000b) Out of the shadows. Notes and Records of the Royal Society of London 54: 407 408. McMenamin MAS (2001a) The Garden of Ediacara and the appearance of complex life. In: Guerzoni S, Harding S, Lenton T, and Ricci Lucchi F (eds.) Pro ceedings of the International School of Earth and Plan etary Sciences, Siena, Italy, 2001, pp. 61 68. Siena, Italy: Consiglio Nazionale delle Richerche, University of Siena. McMenamin MAS (2001b) Paleontology Sonora: Vendian and Cambrian. South Hadley, MA: Meanma Press. McMenamin MAS (2003) Origin and early evolution of predators: The ecotone model and early evidence for macropredation. In: Kelley P, Kowalewski M, and Hansen T (eds.) Predator Prey Interactions in the Fossil Record, Topics in Geobiology Series 20, pp. 379 400. New York: Plenum Press/Kluwer. McMenamin MAS (2003) Spriggina is a trilobitoid ecdy sozoan. Geological Society of America Abstracts 35: 105. McMenamin MAS and McMenamin DLS (1990) The Emergence of Animals: the Cambrian Breakthrough. New York: Columbia University Press. McMenamin MAS and Weaver PG (1992) Proterozoic Cambrian paleobiogeography of the Carolina Terrane. Southeastern Geology 41: 119 128. Seilacher A (1997) Fossil Art. Drumheller, Alta: Royal Tyr rell Museum of Paleontology. Seilacher A and Pfluger F (1997) From biomats to benthic agriculture: A biohistoric revolution. In: Krumbein WE, Paterson DM, and Stal LJ (eds.) Biostabilization of Sediments, pp. 97 105. Oldenburg, Germany: Bib liotheks und Informationssystem der Carl von Ossietzky Universitat Oldenburg (BIS) Verlag.
382 PSEUDOFOSSILS
PSEUDOFOSSILS D M Martill, University of Portsmouth, Portsmouth, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction This entry discusses those objects that appear to be fossilized organic remains – often called pseudofossils – but which in reality are natural structures that merely resemble real fossils. Such structures occur commonly in sedimentary rocks and can easily deceive inexperienced, or even experienced, geologists and palaeontologists. There are many types of pseudofossil, and they often occur associated with real fossils. They may be formed by both physicochemical and biochemical processes, some of which may be the same processes that are responsible for generating real fossils (see Diagenesis, Overview). Pseudofossils are highly variable in form, and the resemblance of a pseudofossil to a real fossil can be quite remarkable. To some degree an object’s categorization as a pseudofossil is a reflection of the experience of the discoverer, and, as such, there is no strict definition of what is or is not a pseudofossil. Those with little or no experience of genuine fossils can easily believe that rocks and pebbles with striped or mottled patterning or unusual shapes might be genuine fossils; indeed, some pseudofossils are real fossils. Some calcareous algae can assume growth forms that resemble miniature versions of macro plant remains. A famous example of this is the so-called Landscape Marble from the Upper Triassic of the Bristol District of England. Here, intermixed arborescent and laminated growths of calcareous algae resemble scenes of forested hills (Figure 1). Objects that are commonly confused with genuine fossils include both three-dimensional objects that represent some form of in situ mineral growth – called concretions or nodules – and largely twodimensional entities occurring on rock surfaces that assume patterns resembling organic remains. In the latter category are the banded mineral growths called Liesegang bands (Figure 2A), dendritic mineral growths (Figure 2B), and various types of fracture surface (Figures 3 and 4).
Liesegang Banding Liesegang banding (Figure 2A) is the result of the diffusion of soluble chemicals through micropores in a rock in differing concentrations and/or oxidation
states and their precipitation to produce colourful (often orange, red, and brown) and variable banded patterns. Trace fossils (see Trace Fossils), such as filled burrows, can be prone to such diffusion and may resemble segmented worms and arthropods. In fact, such a structure would be both a genuine trace fossil (the burrow) and a pseudofossil (the imagined worm). Liesegang banding occurs very commonly in chert nodules in the Cretaceous chalk formations of Europe. It also occurs commonly in many rocks where iron and manganese in a reduced state are weathering from the fresh rock to form oxides, such as goethite (Fe2O3.H2O) and pyrolusite (MnO2). Liesegang rings may penetrate the rock or may be restricted to surfaces such as bedding planes, joints, and microfracture planes.
Dendrites Dendrites are arborescent patterns produced by the fractal growth of minerals as they precipitate from solutions that migrate through fractures in rocks. Commonly they occur as black traces of manganese dioxide (the mineral pyrolusite) or orange-brown precipitates of hydrated iron oxides such as goethite, although other mineral types may also occur. Such structures are commonly mistaken for the remains of soft foliage such as ferns and liverworts. Particularly beautiful examples occur in the Jurassic Solnhofen Limestone (Figure 2B) of Bavaria, Germany, and are often sold as ornaments.
Fracture Surfaces The fractured surfaces of rocks may exhibit patterns that can be confused with organic remains. Very fine-grained and glassy rocks may split with conchoidal fracture – a series of concentric ripples emanating from the initial point of fracture (Figure 3A). Indeed, the very name suggests that the fracture resembles a shell. It is common for flint to fracture in this way, and such conchoidal fractures in flint may resemble the bivalve Inoceramus (Figure 3B), which is a genuine fossil that often occurs in Cretaceous flints. Some fracture surfaces may exhibit feather-like traces, while others may show regular banding resulting from the intersection of cleavages or bedding planes with the fracture surfaces.
PSEUDOFOSSILS 383
Figure 1 A real fossil resembling another. A cut surface through the Late Triassic ‘Landscape Marble’: both the laminated and the arborescent patterns were formed by calcium carbonate precipitation influenced by benthic photosynthetic microbes (probably blue green bacteria). Such structures are known as stromatolites. In former times, small village scenes were painted onto these slabs to create picturesque landscapes.
Figure 2 Superficial pseudofossils on bedding planes of the Late Jurassic Solnhofen Limestone Formation from Bavaria, Germany: (A) Liesegang banding as a series of iron oxide (goethite) diffusion fronts, and (B) dendritic patterns in pyrolusite and goethite. Photographs by Robert Loveridge.
In an unusual example (Figure 4), a piece of highly compacted fossil wood has numerous parallel and orthogonal evenly spaced cracks that have divided the wood into a series of small cubes. Such a blocky fracture pattern is common in fossil woods and can be seen in most bituminous coals. In the specimen shown in Figure 4, each small coaly cube generated a conchoidal fracture when the rock was split to reveal the fossil wood (the split propagated through the fossil wood rather than around it). As a consequence, each cube has a near-spherical surface with concentric ripples. This circular structure resembles the leaf-scar pattern seen on many stems of fossil plants, but the structures are an artefact of splitting.
Cone-In-Cone Structures Cone-in-cone structures are the result of an unusual growth of minerals in which fibrous crystals assume a cone-like growth form (Figure 5A). They are frequently mistaken for fossils, particularly fossil corals, because the outer surface appears as a series of concentric rings or raised discs (Figure 5B). Conein-cone structures are usually composed of fibrous calcite crystals, which nucleate on the surfaces of limestone bands, shell beds, or even large fossils and grow orthogonal to that surface into fine-grained strata, such as clays and shales.
384 PSEUDOFOSSILS
Figure 3 (A) Conchoidal fracture in flint. (B) The bivalve Inoceramus. Inoceramus occurs commonly in flint, and so conchoidal fractures may be easily confused with this bivalve. Photographs by Robert Loveridge.
Figure 4 Conchoidal fractures in small cuboid segments of a fossil tree stem from the Lower Jurassic Posidonia Shale Formation of Dottenhausen, Germany. The conchoidal fractures resemble leaf scars but are a consequence of fracturing of the hard amorphous coal like material. Photograph by Robert Loveridge.
Nodules and Concretions Many of the common diagenetic minerals (those minerals that form in sediment and cement that sediment into rock) form nodular growths that can easily be mistaken for organic remains. Such mineral growths may assume bizarre shapes, influenced by the rate
and direction of diffusion of solutes through the sediment and fluctuations in pH and eH. They may contain genuine fossils, and in some cases the external shape of the nodule assumes the shape of the enclosed fossil as it grows around it. The external appearance of nodules and concretions can be
PSEUDOFOSSILS 385
Figure 5 Cone in cone structures: (A) a cut section of a cone in cone that has grown on the surface of a shell bed, and (B) surface view each cone appears as a circle or disc that can easily be mistaken for a fossil coral at first glance. Photographs by Robert Loveridge.
Figure 6 Nodules and concretions. (A) A spherical concretion of mudstone can easily be mistaken for a fossil egg. (B) The concretion in Figure 6A is split in half to reveal internal septarian cracking. Such cracks are often mistaken for fossils. Photographs by Robert Loveridge.
highly irregular, and they frequently assume shapes resembling isolated bones or even whole animals when there is no fossil enclosed. Highly spherical concretions (Figure 6A) are often mistaken for fossil eggs. Internally, concretions may have radial cracks or a honeycomb network of crystal-lined cracks (Figure 6B). Such concretions are called septarian concretions, and they are frequently mistaken for fossils. Common mineral species that occur as concretions and nodules include the iron sulphide minerals pyrite and marcasite (nodules of the latter are often mistaken for meteorites), silica in the form of chert and flint, and calcite where it cements clay. Nodules sometimes grow in bands or strings, and, en-masse,
they can resemble rows of vertebrae or ribs of large skeletons. A nodule of ironstone found within the rib cage of a dinosaur was mistakenly thought to be a fossil heart by a team of Canadian palaeontologists. Very recently, experiments in which crystals of barite have been grown in gels under laboratory conditions have produced morphologies and sizes that are remarkably similar to structures found in Precambrian rocks that have previously been interpreted as fossil bacteria. The laboratory-grown crystals also resemble structures found in Martian-derived meteorites that were interpreted as evidence of extraterrestrial life. It looks as though these ancient and extraterrestrial microfossils might also prove to be pseudofossils.
386 PYROCLASTICS
See Also
Further Reading
Biosediments and Biofilms. Diagenesis, Overview. Fossil Plants: Calcareous Algae. Sedimentary Rocks: Chert. Trace Fossils.
Garcia Ruiz JM, Hyde ST, Carnerup AM, et al. (2003) Self assembled silica carbonate structures and detection of ancient microfossils. Science 302: 1194 1197.
PYROCLASTICS R J Brown, University of Bristol, Bristol, UK E S Calder, Open University, Milton Keynes, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Pyroclasts are formed by the explosive fragmentation of magma (molten rock) during volcanic eruptions (see Volcanoes). They are carried away from the vent by buoyant eruption plumes, extensive umbrella clouds, or by destructive ground-hugging pyroclastic density currents. The largest explosive eruptions can produce pyroclastic deposits many 1000’s km3 in volume, which can be emplaced on a regional scale in a matter of hours. The study of the physical characteristics of pyroclasts and pyroclastic deposits can reveal much about the dynamic processes involved during explosive eruptions and in pyroclast dispersal and deposition. Recognizing, observing, and understanding pyroclastic deposits are vital first steps in assessing and mitigating volcanic hazard (see Engineering Geology: Natural and Anthropogenic Geohazards). This chapter summarizes the physical characteristics of the principal types of pyroclastic deposits and presents an introduction to their generation and emplacement mechanisms.
Generation of Pyroclastic Material Explosive fragmentation of magma during volcanic eruptions can occur by two main mechanisms. The first involves the rapid exsolution of dissolved magmatic gases during rapid decompression events (magmatic eruptions), and the second results from the interaction of hot magma with external water sources (phreatomagmatic eruptions). Pyroclastic material can also be generated by rapid decompression and by autobrecciation processes during lava dome collapses. Magma comprises three separate materials or phases: a viscous silicate melt (of varying composition), variable amounts of crystals (phenocrysts), and gas (volatiles) such as H2O, CO2, S, F, and Cl. There is a general positive correlation between the
silica content of a magma and the degree of explosivity (Table 1). However, it is the quantity and behaviour of the gas phases that are critically important in determining the eruption style, because it is the rapid expansion of gas during decompression that drives explosive volcanic eruptions. Magma is stored at depth in magma chambers, under high temperatures and pressures. Magmatic eruptions are preceded by an increase in pressure and volume in the magma chamber. This is often attributed to the arrival of new magma into the chamber. The upper parts of many magma chambers are thought to contain a small volume fraction of gas bubbles (vesicles) due to supersaturation with volatiles, and seismic disturbance of these pre-existing bubbles can also lead to increases in magma chamber pressure (Figure 1). Crystallization, which enriches the melt in volatiles, can also act as a trigger. Once a critical point is reached, mechanical failure of the magma chamber roof occurs, allowing magma to rise, decompress, and exsolve gas in a runaway process (vesiculation) that can rapidly drive magma up the conduit at speeds of 200–400 m/s. Vesicle growth is controlled by the volatile content and by the physical properties of the magma (diffusivity rate, density, viscosity, and surface tension). The diffusivity rate is particularly important, and controls the rate at which gas bubbles escape from the magma: where escape is fast (in hot basic lavas), eruptions tend to be effusive or weakly explosive, but where escape is inhibited by high viscosities and low diffusivity rates (in intermediate and rhyolitic magmas), the exsolution of gas can explosively, and very violently, disrupt the magma. The expansion and coalescence of these bubbles forms a magma foam with radically different physical properties to that of the parent magma. During ascent, this rising vesiculated magma is fragmented into discrete particles and transforms into a gas-particle mixture, which accelerates up the conduit and is discharged into the atmosphere (Figure 1). Phreatomagmatic fragmentation is driven by the volumetric expansion of external water after it has been rapidly heated by contact with magma. This mechanism is not restricted by magma type or vent type and it encompasses a spectrum of eruption styles
Table 1 Summary characteristics for the major types of explosive volcanic eruptions Eruption type and examples
C
H (km)
V (km3)
D (km2)
Common eruption processes
Pyroclast types
Duration
Deposits
Ba
1
1
15 >7
Fractures
All rocks
>0.1
Pseudotachylite formation
All rocks
>0.1
>170 >100 >50 >20, possibly as low as 7 >15, possibly as low as 7
Comments
Melts on release of pressure Melts on release of pressure Melts on release of pressure Energy increase on shock compression much greater for porous materials Diaplectic glass forms by solid state transformation. It is amorphous, but retains original crystal form and usually has a higher refractive index than melt glass. Lower bound pressure from PDF formation Stishovite, hollandite, polymorphs of quartz, and feldspar found in impact craters and meteorites. Lower bound pressure from PDF formation Coesite found in impact craters in association with diaplectic glass, implying that it formed on release of pressure. Could conceivably be found in a pseudotachylite that solidified under pressure Found in meteorite melt veins; pseudotachylite like structures that were quenched at high pressure Found in meteorite melt veins; pseudotachylite like structures that were quenched at high pressure Found in meteorites; P 100 GPa from graphite in iron meteorites. Found in impact craters; P 30 GPa from graphite in gneiss. Made in laboratory shock experiments Made in laboratory shock experiments PDFs in quartz are a primary diagnostic for impact. A PDF is a lamellar feature aligned with a low index crystallographic plane. A number of different orientations may appear in the same grain. There is evidence that the lamellae contain high pressure phases that invert to low pressure forms during electron microscopy Laboratory shock experiments show dynamic fracture strength comparable (1.5 times) to static strength Pressure estimate based on observation of pseudotachylites in the fractured zone
Numerous other high pressure minerals have been observed in meteorites and impact craters. The most common and readily observed are listed. The book by French (see Further Reading) contains numerous micrographs of shock metamorphosed quartz.
velocity. Melosh has suggested one such mechanism: entrainment of the rock in the vapour plume formed by strongly shocked material (see Table 1). Finally, there are long-standing controversies over the peak pressures associated with various metamorphic effects. Shock metamorphic effects in rocks and minerals have been studied in numerous laboratory shock experiments over the past 45 years. It was initially hoped that a peak shock pressure calibration could be established based on the presence of various metamorphic effects. The assumption that the peak shock pressure is the only significant parameter seems to be incorrect, as may be inferred by analyses of apparent conflicts in research reports. These conflicts can usually be resolved by considerations of experimental differences between the experiments. Samples loaded to the same peak pressure via different loading paths (single shock vs. a sequence of
shock reflections) often show marked differences in metamorphic effects. Although the pressure duration of laboratory shock experiments is in the range of a microsecond, the shock pressure duration for a large natural impact may exceed a second. The interpretation of metamorphic effects on the basis of laboratory static high-pressure data may be more appropriate in this regime. Some controversies over the precise peak pressure to which a given natural sample has been exposed may not be resolvable on the basis of present knowledge. However, there is usually no argument about whether a given sample has been shock metamorphosed at all.
See Also Impact Structures. Solar System: Meteorites; Mercury; Moon; Mars. Tektites.
184 SOIL MECHANICS
Further Reading Desonie D (1996) Cosmic Collisions, A Scientific American Focus Book. New York: Henry Holt and Co. French BM (1998) Traces of Catastrophe: A Handbook of Shock Metamorphic Effects in Terrestrial Meteorite Impact Structures. LPI Contribution 954. Houston, TX: Lunar and Planetary Institute. Koeberl C and Martinez Ruiz F (eds.) (2003) Impact Mar kers in the Stratigraphic Record. Berlin, Heidelberg: Springer Verlag.
McCall GJH (2001) Tektites in the Geological Record: Showers of Glass from the Sky. Bath: Geological Society Publishing House. Melosh HJ (1989) Impact Cratering: A Geologic Process. Oxford: Oxford University Press and Oxford: Clarendon Press. Rubin AE (2002) Disturbing the Solar System: Impacts, Close Encounters, and Coming Attractions. Princeton: Princeton University Press.
SOIL MECHANICS J Atkinson, City University, London, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Soil and Mechanics Engineering Soils
Soil mechanics describes the mechanical behaviour of granular materials. Mechanical behaviour covers strength, shear stiffness, volumetric compressibility, and seepage of water. Granular materials include powders, grain, and other foodstuffs, mineral ores and concentrates, as well as natural soils. The simple theories of soil mechanics are intended for collections of grains which are uncemented or only very slightly cemented and which contain fluid, usually water or air, in the pore spaces. This covers dense and loose sands and soft and stiff clays. Rock mechanics describes the behaviour of strongly bonded grains whose overall behaviour is governed by joints and fractures. There is a range of materials between these, including weathered rocks, weak rocks, and cemented soils for which simple theories of soil mechanics have limited application. The theories of soil mechanics apply equally to sands (coarse-grained soils) and clays (fine-grained soils). Figure 1 shows samples of sand and clay under load in unconfined compression. In each case the strength arises from suctions in the pore water. The clay is stronger than the sand because it can sustain larger suctions: otherwise their behaviour is fundamentally the same. In describing theories for the behaviour of materials some mathematics is unavoidable. In the following, the mathematics is kept as simple as possible and does not extend beyond simple algebra. Only the most basic and fundamental equations and parameters are included.
Mechanics: Strength, Stiffness, Compressibility, and Permeability
Soils are highly compressible. The volume decreases significantly as it is compressed under an isotropic stress state. This is illustrated in Figure 2C. Soils also change volume when they are sheared and distorted. Strength is basically the maximum shear stress a soil can sustain before it fails. Stiffness is the distortion which occurs as the soil is loaded before it fails. These are illustrated in Figure 2(D). G is the shear modulus and describes stiffness: tf is the shear stress after large distortion and it is the strength. In soils both strength and stiffness increase with increasing mean stress. The frictional nature and the coupling between shear and volume change are the two main differences between the mechanical behaviour of granular materials and the mechanical behaviour of metals and other similar materials.
Figure 1 Unconfined compression of sand and clay.
SOIL MECHANICS 185
s0 ¼ s u
½2
The Terzaghi effective stress equation has been found to apply for a very wide range of loadings and soils, and it is used universally for geotechnical analysis of saturated soils. Plasticity and Cam Clay
In the 1960s Andrew Schofield and Peter Wroth were lecturers at Cambridge University. They applied the then relatively new theories of plastic flow to frictional materials and created a complete stress-strain theory for soils. The model they developed they called Cam Clay and this remains the basis for many of the current constitutive equations for soils. These theories of frictional strength, effective stress, and plastic flow are the basic buiding blocks for modern soil mechanics.
Figure 2 Compression and distortion.
Effective Stress and Drainage
A Brief History of Soil Mechanics
Principle of Effective Stress
Coulomb and Soil Strength
Theories for soil mechanics originated around the middle of the eighteenth century. Coulomb was a military engineer and he was concerned with calculating soil loads on masonry retaining walls. He carried out experiments on the strength of soils and he found that the resistance of soil to shear loading had two components, one cohesive and the other frictional. Coulomb tested unsaturated samples and his analyses were in terms of forces, not stresses. Methods for analysis of stress discovered later by Mohr were incorporated into Coulomb’s results and this is the basis of the well-known Mohr–Coulomb strength equation: tf ¼ c þ s tan f
½1
Neither Coulomb not Mohr had a clear understanding of the importance of pore pressures and effective stresses and the original Mohr–Coulomb equation is in terms of total stress. It is now known that the Mohr–Coulomb equation for soil strength is limited but it is still widely used. Terzaghi and Effective Stress
Karl Terzaghi was an Austrian civil engineer. The major contribution which he made to soil mechanics was to set out a clear theory in the 1920s for accounting for the influence of pore pressure on soil strength and deformation. He proposed an effective stress s0 which controls all soil behaviour and he discovered that for saturated soil this is related to total stress s and pore pressure u by:
The Principle of effective stress first proposed by Terzaghi in 1923, states that the stress which is effective in determining strength, stiffness, and compressibility, the effective stress, s0 is given by eqn [2]. Total stresses arise from external loads due to foundations and walls and loads from self weight. Pore pressures are the pressures in the fluid in the pore spaces. For dry soils the pore pressure is the pressure in the air in the pores. For saturated soil it is the pore water pressure. For soils which are not fully saturated and which contain both air (or gas) and water in the pores the equivalent pore pressure is some combination of the air and water pressure. At present there is no simple and robust theory for determining the equivalent pore pressure and effective stress in unsaturated soils. So far as is known, the principle of effective stress and the effective stress equation (eqn [2]) holds for all dry or saturated soils over a very wide range of stress and pore pressure up to several tens of MPa. The strength and stiffness of soil 1 m below the bed of the deep ocean, where the depth of water may exceed 5 km, will be the same as that of soil 1 m below the bed of a duck pond. Drainage and Consolidation
Because water is relatively incompressible in comparison with soil, volume changes in soil can occur only if water can flow into or out from the pore spaces. Whether or not this happens depends on the rate of drainage and the rate of loading.
186 SOIL MECHANICS
If water cannot drain from the soil it is said to be undrained: its volume must remain constant but pore pressures will change in response to the loading. If water has time to drain freely from the soil it is said to be drained: pore pressures remain constant and volume changes occur. Hence: Undrained loading : dV ¼ 0 and u changes Drained loading : du ¼ 0 and volume changes where the symbol d means ‘a change of.’ In many cases soil in the ground is neither fully drained not fully undrained but simple soil mechanics theories are applicable only for fully drained or fully undrained cases. If soil is loaded undrained the resulting pore pressures will not be in equilibrium with the long-term groundwater pressures. As the excess (out of balance) pore pressures dissipate under constant total stress there will be changes of effective stress and volume changes. This process is known as consolidation. Because the rate of drainage during consolidation depends on hydraulic gradient, which decreases as excess pore pressures dissipate, the rate of consolidation diminishes with time.
Description and Classification of Engineering Soils There are standard schemes for description and classification of soils for engineering purposes. These essentially classify soils under the two main headings: the nature of the grains and how they are packed together. For natural soils, descriptions are added for structure including bonding, bedding, and discontinuities. The Nature of the Soil: Characteristics of the Grains
The most important characteristic is the grain size or grading. Figure 3 shows the range of grain sizes commonly found in natural soils and their descriptions (e.g., sand is 0.06 m to 2 mm). The range is very large. Clay grains are of the order of 1000 times smaller than coarse sand grains. Since permeability is related to the square of the size; sands are of the order of 1 million times more permeable than clays. If a soil is essentially
Figure 3 Grain size descriptions.
single sized it is poorly graded (or well sorted). If it contains a range of sizes it is well graded (or poorly sorted). In a well graded soil it is usually the 10% smaller than (D10) size which governs drainage. Grains of silt size and larger normally consist of rock fragments. They may be rounded or angular, rough or smooth. Grains of clay size are normally made of a clay mineral belonging to one of the major families which are kaolinite, illite, and montmorillonite (smectite). These may be distinguished by their Atterberg Limits and Activity (see below). The characteristics of the grains do not effect the fundamental behaviour of soils but they do influence numerical values of strength and stiffness parameters. Rates of Loading and Drainage
In soil the rate of drainage depends primarily on the permeability which itself depends on the grading of the soil. The Hazen formula for coefficient of permeability k is k / D210
½3
where D10 is the size of the grains with 10% smaller. Typical values for coefficient of permeability range from greater than 10 2 ms 1 (approx. 1.5 m in a minute) for coarse-grained soils to less than 10 8 ms 1 (approx 1 m in 3 years) for fine-grained soils. This very large difference (more than 1 million times) in rate of drainage between coarse-grained and finegrained soils accounts for many of the differences in observed behaviour of sands and clays. The rate of loading also varies widely. Some natural processes, such as deposition and erosion, occur relatively slowly (over decades) while others, such as earthquakes, occur relatively rapidly (over a few seconds). In construction, a shallow trench might be dug in a few hours and a large dam built in a few years. In determining whether a certain event applied to a certain soil is drained or undrained, it is necessary to consider both the rate of drainage and the rate of loading. During earthquakes, coarse-grained sandy soils may be undrained causing liquefaction failure. In construction it is usual to take fine-grained clay soils as undrained and coarse-grained sand soils as drained. Atterberg Limits
If a clay soil has a very high water content it will flow like a liquid; if it has a low water content it will become brittle and crumbly. For intermediate water contents it will be plastic. The Atterberg Limits, the Liquid Limit, Wl, and the Plastic Limit, Wp, define the range of water content over which a clay soil is plastic. The Plasticity Index, IP , is the difference between the Liquid and Plastic Limits:
SOIL MECHANICS 187
Table 1 Typical values for some characteristic soil parameters
Parameter
Liquid Limit Plastic Limit Plasticity Index Activity Maximum specific volume Minimum specific volume Coefft of compressibility Coefficient of swelling Specific volume on NCL at s0 1 kPa Specific volume on CSL at s0 1 kPa Critical state friction angle Very small strain shear modulus at s0 100 kPa on the NCL
Symbol and units
Kaolinite clay (China clay)
London clay
Alluvial sand (Thames)
Carbonate sand
Decomposed granite (Dartmoor)
Wl Wp Ip A Vmax Vmin Cc Cs Vn Vc f0 c degrees G0 o MPa
65 35 30 0.4 2.72 1.92 0.44 0.11 3.26 3.14 25 40
75 30 45 1 2.98 1.80 0.37 0.14 2.68 2.45 23 15
2.2 1.5 0.37 0.03 3.17 2.99 32 60
3.2 2.0 0.78 0.01 4.8 4.35 40 60
0.21 0.01 2.17 2.04 39 60
(Data from research at City University, London.)
IP ¼ Wl Wp
½4
For a natural clay soil, which may contain silt and sand sized grains, the Activity is A¼
IP %clay
½5
and this is related to the mineralogy of the clay, as shown in Table 1. Many numerical values for soil parameters are related to the clay mineralogy and to the Atterberg Limits. State: Liquidity Index and Relative Density
Grains in a soil may be densely packed or loosely packed or in an intermediate state of packing. The packing influences strength and stiffness, as shown in Figure 4. In a clay soil, the loosest packing corresponds to the Liquid Limit and the densest to the Plastic Limit. Intermediate states are described by the Liquidity Index: I1 ¼
w Wp IP
½6
where w is the water content. At the Liquid Limit, the Liquidity Index is 1.0 and at the Plastic Limit it is 0, as shown in Figure 4. In a coarse-grained soil the loosest packing corresponds to the maximum water content, wmax, and the densest to the minimum water content, wmin. Intermediate states are described by the relative density: wmax w Id ¼ wmax wmin
½7
At the loosest state the Relative Density is 0 and at the densest state it is 1.0, as shown in Figure 4.
Figure 4 Packing: plasticity index and relative density.
Because soil strength and stiffness are essentially frictional they depend on the current effective stress. Packing, described by Liquidity Index or Relative Density, is not sufficient itself to describe soil behaviour. Soil state will be defined by a combination of packing and stress, as discussed later.
Behaviour in Compression: Change of Size Isotropic Compression and Swelling
As saturated soil is loaded and unloaded under drained conditions water flows from and into the soil as it compresses and swells, rather like a sponge. The change in volume with changing effective stress is
188 SOIL MECHANICS
overconsolidated; it has experienced a greater stress, s0m. The overconsolidation ratio is R0 ¼
illustrated in Figure 5(A). The soil is first loaded from A to B and it compresses. The compression of the soil skeleton is due mostly to particles rearranging and also to weak coarse grains fracturing or clay grains bending. The soil is unloaded from B to C and reloaded back to B. Some strains are recovered and there is a hysteresis loop. Volume changes in coarsegrained soils will be small because grains do not ‘unrearrange’ or ‘unfracture’ but may be significant in clay soils as the grains can unbend. In Figure 5B, effective stresses are plotted to a logarithmic scale and the compression and swelling curves have been idealised. The volume axis is the Specific Volume defined as V Vs
½8
where Vs is the volume of soil grains in a volume V of soil. For many soils v will range from about 1.2 if the soil is dense to over 2 if it is loose. The linear normal compression line ABD is given by v ¼ vn Cc log s0
½9
and the linear swelling and recompression line CB is given by v ¼ vs Cs log s0
½11
Equations [9] and [10] relate volume to stress for isotropic loading and unloading. Since the stress scale is logarithmic, the stress-strain behaviour is non-linear; the bulk modulus is not a constant but varies with both stress and overconsolidation. The idealization of the hysteresis loop in Figure 5A to the line CB in Figure 5(B), common in simple soil mechanics theories, is unrealistic for many soils. Soil stiffness will be discussed later.
Figure 5 Isotropic compression and swelling.
v¼
s0m s0
One-dimensional Compression in the Ground
Below level ground, the state of stress is not isotropic but one-dimensional, with zero horizontal strain during deposition and erosion; the vertical and horizontal effective stresses s0v and s0h are related by the coefficient of Earth pressure, Ko, given by Ko ¼
The Compression Index, Cc, the Swelling Index, Cs, and the Specific Volume, vn, at unit stress are material parameters and are related to the characteristics of the grains. Typical values are given in Table 1. The location of a swelling line is given either by the maximum stress, s0m, or the specific volume, vs, at unit stress. A soil whose state lies on the line ABD is said to be normally compressed and ABC is the normal compression line (NCL). Soil whose state is on the NCL has not experienced a larger stress. A soil whose state is on a swelling line, such as CB, is said to be
½12
For normally consolidated soil (Ro ¼ 1) Konc is given by Konc ¼ ð1 sin f0c Þ
½13
where f0cs is the critical state friction angle. Ko increases with overconsolidation ratio. Horizontal effective stresses given by eqn [12] are for level ground with zero horizontal strain. Near slopes, foundations, and other underground construction stresses will be modified by the stresses imposed by the slope and the structure. Calculations of settlement in the ground are often carried out in terms of a coefficient of compressibility, mv, or a one-dimensional modulus, M, given by M¼
½10
s0h s0v
1 Ds0v ¼ mv Dv=v
½14
where Dv is the change of specific volume observed in a laboratory test on a soil sample with initial specific volume, v, when subjected to an increment of vertical stress, Ds0v. Since soil stiffness is non-linear, M is not a soil constant and the increment of stress applied in the test should correspond to the expected change of stress in the ground. State: Stress and Packing
The behaviour of a particular soil depends on both the current effective stress and on the Relative
SOIL MECHANICS 189
stress is very large. Similarly, a soil which has a relatively high specific volume and the grains are relatively loosely packed will dilate if the effective stresses are very small. Sediments at great depth deform plastically. Nearsurface soils often behave in a brittle manner and crack. Relative Density, or Liquidity Index, on its own is not sufficient to predict the behaviour on subsequent shearing; the effective stress must be taken into account as well.
Strength of Soil
Figure 6 States and state parameters.
Behaviour of Soil During Shearing
Density or Liquidity Index. These may be combined into a state parameter. Figure 6A, which is similar to Figure 5B, shows the state of an overconsolidated sample at X and the normal compression line. All samples with states on the broken line through X parallel to the NCL will behave in a similar way. These states can be described by a stress state parameter Ss given by Ss ¼
s0x s0e
½15
where s0e is the equivalent stress on the NCL at the same specific volume as that at X. The state parameter describes the distance of the state from the NCL. If the swelling index Cs is small, Ss is approximately equal to the overconsolidation ratio, R0. The concept of state is of fundamental importance for soils which are both frictional and which change volume during loading, as it combines both relative density and stress into a single parameter.
Figure 7A shows a block of soil with a constant normal effective stress s0 subjected to an increasing shear stress t0 . The soil is drained and it distorts with a shear strain g and a volumetric strain ev. If the soil is undrained, there are no volume changes but the pore pressures change. The block of soil represents conditions inside a slip zone in the slope illustrated in Figure 7B or in a foundation illustrated in Figure 7C. If the slope is created by excavation or erosion the normal stress decreases and, since soil strength is frictional, it will weaken, whereas below the loaded foundation the normal stress increases and the soil becomes stronger. The behaviour of soil initially loose and initially dense of critical is illustrated in Figure 8. The loose soil (marked L) compresses during shearing even
Dense and Loose States
After shearing, soils reach ultimate or critical states in which they continue to distort at constant state (i.e., at constant stress and volume). The relationship between specific volume and effective stress gives a critical state line (CSL) parallel to the normal compression line, as shown in Figure 6(B). The critical state line is given by v ¼ vc Cc log s0
½16
Soil states which are above the CSL are known as ‘loose of critical’ and the soil will compress on shearing. Soil states which are below the CSL are known as ‘dense of critical’ and the soil will dilate on shearing. The CSL separates regions of fundamentally different behaviour of the same soil. A soil which has a relatively low specific volume and the grains are relatively closely packed will compress if the effective
Figure 7 Shearing of soil.
190 SOIL MECHANICS
though the normal stress remains constant and the dense soil (marked D) dilates. The rate of dilation is given by an angle of dilation c, given by tan c ¼
dev dg
½17
(The negative sign is required as c is positive for negative (dilation) volumetric strains.) Critical State Strength
The samples shown in Figure 8 have the same effective stress and they reach the same critical shear stress and the same critical specific volume after relatively large strains. Figure 9 shows critical states for a number of samples. There are unique relationships between the critical shear stress t0f , the critical normal stress sf0 , and critical specific volume vf, given by
tf ¼ s0f tan f0c
½18
tf ¼ vc Cc log s0f
½19
These equations define a critical state line and the parameters fc0 , Cc, and vc are material parameters. (Critical state lines are usually shown as double lines, as in Figure 9). Typical values are given in Table 1. During shearing distortions, all soils will ultimately reach a critical state; if they did not they would continue to change state indefinitely, which is impossible. In simple soil mechanics theories, the critical states reached by a particular soil, given by eqns [18] and [19], are independent of the starting state and whether the soil is drained or undrained. Undrained Strength
Figure 9B shows that the shear stress at failure, which is the shear strength, decreases as the specific volume at failure increases. If soil is undrained the water content and the undrained strength remain unchanged for any changes in total normal stress. The undrained strength. tf ¼ su
½20
depends on the water content. In practice, samples are taken from the ground and tested without change of water content. The undrained strengths measured can be used for design so long as the water content in the ground does not change. It is common knowledge that soils become weaker as their water content increases. This is shown in Figure 10 in which the undrained strength, with a logarithmic scale, decreases linearly with water content. The strength of soil at its Liquid Limit is approximately 1.5 kPa and the undrained strength Figure 8 Stress and volume change in shearing soil.
Figure 9 Critical states.
Figure 10 Undrained strength and water content.
SOIL MECHANICS 191
of soil at its Plastic Limit is approximately 150 kPa (i.e., the strength of soil changes by about 100 times as the water content changes from the Liquid Limit to the Plastic Limit). Peak Strength
Soils whose initial states are dense of critical have a peak strength before they reach a critical state, and they dilate during drained shear, as shown earlier in Figure 8. The peak strengths vary with effective normal stress and specific volume, as shown in Figure 11. Samples which reach their peak states at the same specific volume have peak strengths on an envelope shown in Figure 11A. The envelope is often approximated by a straight line, shown in Figure 11A given by tp ¼ c0p þ s0p tan f0p
½21
The peak friction angle, f0p , is a material parameter and, from Figure 11A f0p < f0c . The cohesion intercept, c0p , is not a material parameter and its value depends on the specific volume. Moreover c0p is not
the strength at zero effective normal stress, as this must be zero for an uncemented granular material. The linear approximation for peak strength given by eqn [21] is applicable only within the range for which data are available. Figure 11B shows additional data at smaller normal effective stresses; there the envelope is now distinctly curved and passes through the origin. The curved peak failure envelope, shown in Figure 11C, can be represented by a power law of the form. 0
tp ¼ As b
½22
where b is a material parameter and A depends on the specific volume. From analyses of the stresses and strains in the soil block, shown in Figure 7A, peak shear strength is given by tp ¼ s0 tan ðf0c þ cÞ
½23
At the critical state, c ¼ 0 and t0c is given by eqn [18]. At the peak state, the angle of dilation is at a maximum. The maximum rate of dilation is governed by the state parameter so the peak strength increases as the initial state moves away from the critical state line. Equations [21, 22 and 23] are alternative theories for the peak strength of soils. They all contain a combination of material parameters and state dependent parameters. Equations [22 and 23] correctly give zero strength at zero effective stress. Equation [21] is most commonly applied in practice.
Stiffness of Soil Figure 5A shows non-linear isotropic unloading and reloading behaviour. Similar non-linear behaviour occurs during shearing, as shown in Figure 12A. The tangent shear modulus G0 is the gradient of the stress-strain curve given by G0 ¼
dt dg
½24
At the start of shearing near the origin the shear modulus is G0o and at failure the shear modulus is zero.
Figure 11 Peak strength.
Figure 12 Stiffness and shear modulus.
192 SOIL MECHANICS
Figure 12B shows the variation of shear modulus G0 with the progress of loading. There is a very small range up to a shear stress to, in which G0o is constant and the soil is linear, but over the remainder of loading the shear modulus decays with loading. For a particular soil the value of G0o and the shear modulus at a particular strain, vary with the effective stress and with the state parameter. For modest compression the bulk modulus, K0 , and the one-dimensional compression modulus, M, both decay with normal stress in a manner similar to the decay of shear modulus with shear stress, shown in Figure 12B. At large compressive stresses the stiffness is the modulus corresponding to states on the NCL. At very large compressive stresses, the stiffness becomes very large as the specific volume approaches 1.0.
Consolidation As soil is loaded or unloaded undrained, there are no volume changes but there are changes of pore pressure. These create excess pore pressures which are not in equilibrium with the surrounding pore pressures and so they dissipate with time. As they dissipate, under constant total stress, there are changes of effective stress which cause volume changes accompanied by drainage of water. The basic theories for consolidation are for onedimensional loading and drainage, illustrated in Figure 13A, in which all movements of soil and water are vertical. In practice this corresponds to conditions below a wide foundation or embankment. Solutions for the rate of consolidation are given in terms of the degree of consolidation, Ut, and the time factor, Tv, given by Ut ¼
rt r1
½25
Tv ¼
cv t H2
½26
Figure 13 Consolidation.
where rt is the settlement at time t, r1 is the settlement after a very long time, H is the length of the drainage path, and cv is the coefficient of consolidation given by cv ¼
Mk gw
½27
where M is the one-dimensional modulus, k is the coefficient of permeability, and gw is the unit weight of water. The relationship between degree of consolidation and time factor is shown in Figure 13B. The rate of consolidation depends on soil characteristics of stiffness and permeability and also on the geometry of the consolidating layer. This is given by the drainage path length H which is the greatest distance water must move to reach a drainage layer. Consolidation times can be significantly reduced by installing drains into the ground to reduce H. Consolidation is the principal cause of the settlement of foundations and embankments on clays long after construction is complete.
Normalization and a State Boundary Surface Figure 14A and B shows some different soil states. There are peak states corresponding to two different specific volumes; these are the same as those shown in Figure 11. There are paths for shearing of normally consolidated loose samples: path LD is for drained shearing and path LU is for undrained shearing. These soil states involve three parameters, shear stress t, normal stress s0 , and specific volume v. Soil states can be represented by a three-dimensional surface using these axes. They may be represented on a two-dimensional graph using an appropriate normalizing procedure. There are several possibilities and one is to divide the shear and normal stresses by the equivalent stress s0e , shown in Figure 6A. Figure 13C shows the states normalized by the equivalent stress. The NCL and the CSL reduce to single points. The peak states fall on a unique curve. The state paths for drained and undrained shearing of normally consolidated samples fall on a unique curve. The full curve represents a boundary to all possible states, known as a state boundary surface. The concept of a state boundary surface is employed in advanced soil mechanics theories to develop complete constitutive relationships for soils. The surface is taken to be a yield surface and as a plastic potential surface from which plastic strains are determined. For states inside the boundary surface, the behaviour is taken to be elastic. One such theory is known as Cam Clay, for which the state boundary surface is represented by a logarithmic spiral curve.
SOIL MECHANICS 193
If the soil is assumed to be fully drained, pore pressures can be determined and effective stresses calculated. Analyses are then carried out using effective stresses with effective stress strength and stiffness parameters. If the soil is assumed to be undrained, there are no changes in volume but there are changes in pore pressure which cannot be easily determined. In this case analyses have to be carried out using total stresses with undrained strength and stiffness parameters. The critical state strength should be used to investigate ultimate failures. The peak strengths, with appropriate factors, should be used to investigate designs which are required to limit movements. Simple analyses of foundation settlement are often carried out assuming one-dimensional conditions using the one-dimensional modulus, M, or using simple elastic theories using a shear modulus, G, and a bulk modulus, K. In all cases, it is necessary to take account of non-linear stress-strain behaviour and the appropriate drainage conditions. Simple analyses of rate of settlement due to consolidation can only be carried out assuming one-dimensional conditions. The advanced soil mechanics theories, such as Cam Clay, are not used in simple analysis and design except for extremely simplified cases. Instead they form the basis for analyses using finite element or other comparable numerical methods.
See Also Figure 14 A state boundary surface.
Engineering Geology: Liquefaction; Made Ground; Problematic Soils; Subsidence. Soils: Modern; Palaeosols.
Further Reading Applications The simple theories presented above for granular materials form the basis for analysis and design of engineering works which interact with the ground such as foundations, slopes, tunnels and retaining walls. The basic theories for the mechanical behaviour granular materials are applicable equally to coarsegrained soils (sands and gravels) and fine-grained soils (clays). The principle factor to consider is the relative rate of loading and drainage. For routine analysis a particular case must be taken to be either fully drained or fully undrained.
Atkinson JH (1993) The Mechanics of Soils and Founda tions. London: McGraw Hill. Goodman RE (1999) Karl Terzaghi: the Engineer as Artist. American Society of Engineering Press, Reston, Virginia. Heyman J (1972) Coulomb’s Memoir on Statics. Cam bridge: Cambridge University Press. Lancellotta R (1995) Geotechnical Engineering. Balkema, Rotterdam. Muir Wood DM (1990) Soil Behaviour and Critical State Soil mechanics. Cambridge: Cambridge University Press. Powrie W (2004). Soil Mechanics, 2nd edn. Spon Press: London. Schofield AN and Wroth CP (1968) Critical State Soil Mechanics. McGraw Hill.
194 SOILS/Modern
SOILS Contents Modern Palaeosols
Introduction
distinct from the minerals of most soils, and they bestow high fertility and low bulk density on some volcanic soils (the process of andisolization). Waterlogging in low-lying parts of the landscape prevents the rusting of iron minerals and imparts a grey-green colour to the soil (the process of gleization). Leachates from highly acidic vegetation, such as pine forest, create soils in which clays are destroyed but quartz and haematite accumulate (the process of podzolization). Finally, climate is also an important factor in
There are many soil-forming processes, which in varying combinations create the large array of soils forming at the surface of the Earth. The study of soils is aided by the observation that soil-forming processes are slow and seldom go to completion. The parent materials of soils are modified over thousands of years by physical, chemical, and biological influences. However, few of these processes can be observed directly. Podzolization is one of the few soilforming processes that is rapid enough to be recreated in the laboratory. Soil-forming processes that operate over thousands of years are studied using a space-fortime strategy (that is, studying soils of differing ages that are subject to the same soil-forming regime). A set of soils of different ages with comparable climates, vegetation, topographical positions, and parent materials is called a chronosequence (Figure 1). Mathematical relationships between the development of particular soil features and time are called chronofunctions, and include the increased clayeyness produced by the soil-forming process of lessivage (Figure 2). While specifying the rate and progress of soil formation, chronofunctions can also be used to infer the ages of landscapes from undated soils by comparison with dated soils. Such estimates of soil age can be important in the study of the neotectonic deformation of landscapes and their suitability for long-term installations such as dams and nuclear power plants. Soil fertility also varies with soil age, and chronofunctions can guide agricultural use and rehabilitation of soils. Soil-forming processes vary not only with time but also with parent materials, topographical relief, vegetation, and climate. For example, the fragments of volcanic glass in certain kinds of air-fall tuff are
Figure 1 Soil development stages involving progressive calci fication (top), lessivage (middle), and paludization (bottom). Re produced with permission from Retallack GJ (2001) Soils of the Past. Oxford: Blackwell.
Modern G J Retallack, University of Oregon, Eugene, OR, USA ß 2005, Elsevier Ltd. All Rights Reserved.
SOILS/Modern 195
derived from a Russian term for the grey clay of swamps and bogs. Waterlogged peat-covered stagnant groundwaters allow the preservation of ferrous iron in clay minerals, such as grey smectite, carbonates, such as the siderite of freshwater bogs, and sulphides, such as the pyrite of mangrove swamps and salt marshes. In normally drained soils these minerals rust to produce red and brown clays, hydroxides such as goethite, and oxides such as haematite (Table 3). Goethite and haematite also form within gleyed soils when a short-term depression of the water table allows the atmospheric penetration of oxygen. Despite these red nodules and concretions, the dominant colour of gleyed soils is bluish or greenish grey (Figure 4).
Paludization
Figure 2 Chronofunctions for the progress of lessivage in soils of the Coastal Plain and Piedmont of south eastern USA over time: (A) solum thickness; (B) thickness of the argillic horizon; and (C) the amount of clay in the solum. The solum is the A and B horizons; the argillic horizon is the Bt horizon; and the total profile is the A, B, and C horizons as defined in Table 2. Reproduced with permis sion from Retallack GJ (2001) Soils of the Past. Oxford: Blackwell, using data from Markewich HW, Pavich MJ, and Buell GR (1990) Contrasting soils and landscapes of the Piedmont and Coastal Plain, eastern United States. Geomorphology 3: 417 447.
soil-forming processes, encouraging deeper and more thorough weathering in wetter and warmer climates (Figure 3). The study of soil-forming processes has informed both soil taxonomy (Table 1) and soil-profile terminology (Table 2). The following outlines of soil-forming processes are presented in the order in which they would be encountered from warm wetlands to cold arid lands.
Paludization is literally ponding, but a pond would not be commonly understood as a soil. Paludization is soil flooding that is tolerated by swamp trees but not by most soil decomposers. Paludization is thus an accumulation of undecayed plant debris as peat in the waterlogged surface layer (O horizon of Table 2) of Histosols (Table 1). This process requires a balance between plant production and decomposition. If ponding is intermittent and the soil is moderately oxidized, usually because of a low subsidence rate, then fungal and other decay prevents the accumulation of plant debris. If, on the other hand, ponding is too deep or prolonged, because of high subsidence rates, then soil stagnation kills the roots of woody plants, thus cutting off the supply of vegetation for further peat accumulation. As swamp forests die from anoxia at the roots, peaty soils become overwhelmed by lakes, bayous, or lagoons. The rate of subsidence and accumulation of woody peats is generally between 0.5 mm and 1 mm per year, because of constraints on the growth rate of woody plants in low-fertility peaty substrates and the depth of penetration of air and decomposers within woody peats. Herbaceous plants and mosses are less constrained in their growth rates and form domed peats that rise well above the water table. Peat accumulation in both cases involves addition from the top, in the same way as sediment accumulation, and thus differs from soil-forming processes that modify pre-existing materials. The progress of paludization leads to progressively thicker peat (Figure 1).
Podzolization Gleization Gleying or gleization is a process that produces and maintains unoxidized minerals in soils and is a term
Podzol in its original Russian means ‘under ash’ and refers to the light-coloured quartz-rich (E) horizon immediately beneath the humus. Many podzolic
196 SOILS/Modern
Figure 3 Selected common soil forming processes arranged along a climatic gradient. The ecosystems depicted are (from left to right): bald cypress swamp, spruce forest, oak forest, tropical rain forest, Acacia savannah, and saltbush scrub. Horizon nomenclature is described in Table 1, and the large arrows indicate the movement of key soil components. Reproduced with permission from Retallack GJ (2001) Soils of the Past. Oxford: Blackwell.
Table 1 Outline of soil taxonomy Order
Description
Entisol
Very weakly developed soil with surface rooting and litter (A horizon) over weathered (C horizon) sediment with relict bedding or weathered igneous or metamorphic rock with relict crystals Weakly developed soil with surface rooting and litter (A horizon) over somewhat weathered (Bw horizon) clayey (Bt horizon) or calcareous (Bk horizon) subsurface Soil composed of volcanic ash with low bulk density and high fertility Peat (O horizon) over rooted grey clay (A horizon) Quartz rich clay poor soil with bleached subsurface (E horizon) above a red black iron aluminia organic cemented zone (Bs horizon) Very clayey profile with common swelling clay (smectite), laterally variable thickness of surface (A horizon), and strongly slickensided subsurface (Bt horizon) Grassland soil with thick crumb textured carbon rich surface (A horizon) Permafrost soil with frost heave and other periglacial features Desert soil with a shallow subsurface accumulation of pedogenic carbonate (Bk horizon) and soluble salts (By horizon) Fertile forest soil with clay enriched subsurface (Bt horizon) and high amounts of Mg, Ca, Na, and K Infertile forest soil with clay enriched subsurface (Bt horizon) and low amounts of Mg, Ca, Na, and K Deeply weathered tropical soil, often highly ferruginous and aluminous, but with very low amounts of Mg, Ca, Na, and K
Inceptisol Andisol Histosol Spodosol Vertisol Mollisol Gelisol Aridisol Alfisol Ultisol Oxisol
For technical limits of soil orders see Soil Survey Staff (2000) Keys to Soil Taxonomy. Blacksburg: Pocahontas Press.
soils are now included in the USDA (United States Department of Agriculture) soil order Spodosol (Table 1), which refers to the red, brown, or black (Bs) horizon below the light coloured near-surface
layer. This striking differentiation between white near-surface and dark subsurface horizons is created by podzolization, which effectively leaches iron and organic matter from the upper horizons and
SOILS/Modern 197
reprecipitates them in a lower horizon. The resulting effect is as striking as the chromatographic separation of organic compounds, and podzolization is one of the few soil-forming processes that is rapid enough to have been recreated under controlled laboratory conditions. The process is particularly helped by highly acidic soil solutions (with a pH of less than 4) in welldrained soils of humid climates under acid-generating litter such as that of conifer forest (Figure 3). Under highly acidic conditions clay minerals are destroyed, so Podzols and Spodosols usually have a sandy texture.
Ferrallitization The term ferrallitization is derived from iron (Fe) and aluminium (Al), which become enriched in minerals such as haematite, kaolinite, and gibbsite during
intense weathering of well-drained tropical soils such as Oxisols (Figure 3). Much of the loss of major cations (Ca2þ, Mg2þ, Naþ, Kþ) by hydrolysis requires carbonic acid derived from the carbon dioxide of soil respiration, yet the soil pH remains above 4, so that clays are not destroyed. Mitigation of acidity and deep oxidation of these soils may in part be due to the activity of termites and tropical trees, as ferrallitization is primarily found in soils under tropical rainforest. The broad-leaved trees of tropical rainforests produce less acidic litter than conifers and other plants, and litter decomposition rates are high on humid and warm forest floors. Furthermore, ferrallitic soils commonly contain abundant microscopic (125–750 mm) spherical to ovoid pellets of oxidized clay, like the faecal and oral pellets of termites. Some ferrallitic soils appear to have passed through the guts of termites many times. Termites are unique in having extremely alkaline midguts (with a pH of 11–12.5).
Table 2 Standard acronyms for soil horizon description
Biocycling
Acronym
Description
O A
Surface accumulation of peaty organic matter Surface horizon of mixed organic and mineral material Subsurface horizon rich in weather resistant minerals, e.g. quartz Subsurface horizon enriched in washed in clay Subsurface horizon enriched in organic matter, or iron or aluminium oxides Subsurface horizon enriched in pedogenic carbonate Subsurface horizon with domed columnar structure and sodium clays Subsurface horizon enriched in salts such as gypsum and halite Subsurface horizon deeply depleted of Ca, Mg, Na, and K Subsurface horizon mildly oxidized and little weathered Mildly weathered transitional horizon between soil and substrate Unweathered bedrock
E Bt Bs Bk Bn By Bo Bw C R
For technical limits of soil orders see Soil Survey Staff (2000) Keys to Soil Taxonomy. Blacksburg: Pocahontas Press.
Biocycling includes a variety of processes in which nutrient elements are exchanged by soil biota without reincorporation into soil minerals. In tropical soils such as Oxisols (Table 1) this is a very efficient process in which the decay of leaves and wood is orchestrated by waves of bacteria, fungi, ants, and termites, which excrete and die to feed a copious network of epiphytes and tree roots. Effective biocycling explains the spectacular luxuriance of tropical-rainforest ecosystems despite their extremely nutrient-depleted and humus-poor mineral soils (Oxisols). Comparable mechanisms operate in swamp forests growing in peat (Histosols), which also experience severe mineralnutrient limitations. These mineral nutrients include the major cations (Ca2þ, Mg2þ, Naþ, Kþ), but these are seldom as limiting as nitrogen, which is derived largely from the microbial recombination of atmospheric nitrogen, or phosphorous, which is derived largely from the weathering of apatite. Biocycling of
Table 3 Common kinds of chemical reactions during weathering Reaction
Example
Hydrolysis
2NaAlSi3 O8 þ 2CO2 þ 11H2 O ! Al2 Si2 O5 ðOH4 Þ þ 2Naþ þ 2HCO 3 þ 4H4 SiO4 albite þ carbon dioxide þ water ! kaolinite þ sodium ions þ bicarbonate ions þ silicic acid 2Fe3þ þ 4HCO 3 þ 1=2O2 þ 4H2 O ! Fe2 O3 þ 4CO2 þ 6H2 O ferrous ions þ bicarbonate ions þ oxygen þ water ! haematite þ carbon dioxide þ water 2FeOOH ! Fe2 O3 þ H2 O goethite ! haematite þ water CaCO3 þ CO2 þ H2 O ! Ca2þ þ 2HCO 3 calcite þ carbon dioxide þ water ! calcium ions þ bicarbonate ions
Oxidation Dehydration Dissolution
198 SOILS/Modern
Figure 5 Light brown near surface (E) and dark brown subsur face (Bt) horizons of an Alfisol produced by lessivage near Killini, Greece. Hammer handle upper right is 25 cm long.
Figure 4 Red and brown mottles of goethite in the upper part of the profile and dark stains of pyrite formed by gleization in the lower part of the profile of a gleyed Inceptisol, excavated as a soil column from a salt marsh on Sapelo Island, Georgia, USA. Hammer handle is 25 cm long.
nitrogen is especially important during the early development of soils such as Entisols and Inceptisols, which are developed over decades or centuries. Biocycling of phosphorous becomes increasingly important in very old soils such as oxisols and ultisols, which are depleted in apatite over thousands or millions of years.
Lessivage Lessivage or argilluviation is the process of clay accumulation within a subsurface (Bt or argillic) soil horizon (Figures 1, 2 and 3). This is a common and widespread soil-forming process in the forested soils of humid climates, particularly Alfisols and Ultisols (Figure 5). The clay is primarily derived from a hydrolytic weathering reaction in which clays remain as a residuum and dissolved cations are removed in groundwater during the incongruent dissolution of feldspars and other minerals by carbonic acid
(Table 3). Driving the reaction are abundant rainfall and high soil respiration rates fuelled by high primary productivity. Clay forms rinds around mineral grains of the sedimentary, igneous, or metamorphic parent material, but is also washed down cracks in the soil created by desiccation, roots, and burrows. This washed in or illuvial clay has a very distinctive banded appearance, which is obvious in petrographic thin sections. The clay is not washed any lower than the water table, where percolating rainwater ponds. Clay is less common near the surface of the soil, where unweathered grains are added by wind and water, and grains are leached of clay by plant acids. The net effect is a subsurface clayey horizon that becomes more clayey over time (Figures 1, 2 and 3).
Lixiviation Lixiviation is a process of leaching of major cations (Ca2þ, Mg2þ, Naþ, Kþ) from soil minerals and their loss from the soil in groundwater. Lixiviation is a component of ferrallitization, podzolization, and lessivage, and represents the progress of the hydrolysis chemical reaction, in which hydronium ions (Hþ) of a weak acid (usually carbonic acid) displace cations into solution and thus convert primary minerals such a feldspars into soil minerals such as clays (Table 3). The term lixiviation is primarily used to describe the beginnings of this process in soils such as Entisols and Inceptisols that have developed over
SOILS/Modern 199
only decades or centuries. Such young soils have not yet acquired the distinctive deeply weathered and oxidized horizons produced by ferrallitization in Oxisols, the distinctive leached (E) and enriched (Bs) horizons produced by podzolization in Spodosols, or the distinctive clay-enriched subsurface (Bt) horizons produced by lessivage in Alfisols and Ultisols.
Melanization Melanization is a process of soil darkening due to the addition of soil organic matter. The process is best known in Mollisols, the fertile dark crumb-textured soils of grasslands (Figure 6). In these soils melanization is largely a product of the activities of grasses and earthworms. Earthworms produce faecal pellets rich in organic matter and nutrients such as carbonate. Earthworms also produce slime, which facilitates their passage through the soil. Root exudates from grasses are also added to soil crumbs. Many soils have dark humic near-surface horizons, but a peculiarity of grassland soils is that dark organic fertile crumbtextured soil extends to the base of the rooting zone, which can be more than a metre deep in soils under tall-grass prairie. Melanization also occurs in swamp and marsh soils (gleyed Inceptisols and Entisols), where the decay of humus is suppressed by poor oxidation and waterlogging. Unlike the alkaline crumbtextured melanized surface of grassland soils, the melanized surface of wetland soils is nutrient-poor, acidic, and has a massive to laminated fabric. Melanization is not usually applied to the precipitation of
Figure 6 Dark organic rich surface (mollic epipedon) of a Mol lisol formed by melanization near Joliet, IL, USA. The shovel handle is 15 cm wide at the top.
amorphous Fe–Mn oxides (birnessite) in gleyed soils, which can also produce dark soil. The creation of these Fe–Mn-stained (placic) horizons is a process of gleization rather than melanization.
Andisolization Andisolization is the formation of fertile mineralogically amorphous low-density horizons within soils of volcanic ash (Andisols). Many volcanic ashes are composed largely of small angular fragments (shards) of volcanic glass. Unlike soil minerals such as feldspar, volcanic glass weathers, not to crystalline minerals such as clay, but to non-crystalline compounds such as imogolite. The loosely packed angular shards and colloidal weathering products create a soil of unusually low bulk density (1.0–1.5 g cm 3, compared with 2.5–3.0 g cm 3 for most common minerals and rocks). Furthermore, these colloidal compounds contain plant-nutrient cations, and particularly phosphorous, in a form that is more readily available to plants than those of other kinds of soils dominated by crystalline minerals such as apatite. Andisolization is not sustainable for more than a few thousand years unless there are renewed inputs of volcanic glass, because glass and other colloids (such as imogolite) weather eventually to oxides and clay minerals.
Vertization Vertization is the physical soil overturning and mixing by means of the shrink–swell behaviour of clays. It occurs mostly in Vertisols but also in Entisols, Inceptisols, Mollisols, and Alfisols. It is especially characteristic of soils rich in swelling clays (smectites), which swell when wet and shrink when dry. Also characteristic is a climate with a pronounced seasonal contrast in precipitation. During the wet season the clays swell and buckle under the pressure of their inflation. During the dry season they open up in a system of cracks, which are then partly filled by wall collapse. This fill exacerbates the buckling in the next wet season so that the soil develops ridges or mounds with intervening furrows or pits, called gilgai microtopography. In a soil pit, the cracks of mounded areas divide areas of festooned slickensides under the furrows and pits in a distinctive arrangement called mukkara structure (Figure 7). Vertization is mainly a phenomenon of semiarid to subhumid regions. Soils of arid regions are generally not sufficiently clayey, whereas soils of humid regions are generally too deeply weathered to contain abundant smectite and are also stabilized by massive plant and animal communities.
200 SOILS/Modern
Figure 7 Gilgai microrelief (low to left, high to right) and its subsurface mukkara structure (festooned intersecting slicken sided cracks) produced by vertization in the Branyon clay soil, a Vertisol, near New Braunfels, TX, USA. Scale to left shows 50 cm and 100 cm; red and white bands on pole to right are 10 cm wide.
Anthrosolization Anthrosolization is the alteration of soil by human use, such as buildings, roads, cesspits, garbage dumps, terracing, and ploughing. Archaeological ruins and artefacts are important clues to prior occupation of a site, but many sites also contain impressive amounts of mollusc shells and mammal and fish bones. A distinctive soil structure of subsoil pockets of laminated clay between large soil clods is produced by moldboard ploughs. The primitive or ard plough also tends to disrupt the natural crumb structure to a fixed depth (plough line). Phosphorous content is an indicator of human use. Many soils have trace amounts of phosphorus (10–20 ppm by weight), but occupation floors and long-used garden soils and middens have large amounts of phosphorous (1000– 2000 ppm). Anthrosolization is locally common worldwide in cities and fields, both ancient and new, but is scattered and local in deserts, polar regions, and high mountains.
Calcification Calcification is the accumulation of calcium and magnesium carbonates in the subsurface (Bk) horizons of soils (Figures 1 and 3). The carbonate is usually obvious, appearing as soft white filaments, hard white nodules, and thick white benches within the soil. Calcification is largely a soil-forming process of dry climatic regions, where evaporation exceeds precipitation. It is characteristic of Aridisols but is also found in some Mollisols, Andisols, Vertisols, Inceptisols, and Alfisols. The source of the carbonate is the soil respiration of roots, soil animals, and microorganisms. Calcification requires soil respiration at
levels greater than those in hyperarid soils, where halite and gypsum formed by salinization prevail, and less than those in humid soils, where lessivage prevails. The source of the cations of calcium and magnesium, which create the soil minerals calcite and dolomite, respectively, is the weathering of soil minerals by hydrolysis (Table 3). Some of these cations originate from feldspars and other minerals of the parent material, but dry regions of calcification have open vegetation and are often dusty, so that carbonate and feldspar dust is an important source of cations. Dissolved cations from hydrolytic weathering are commonly lost downstream in the groundwater in humid regions, but in arid lands the water table is commonly much deeper than the soil profiles, which are seldom wet much beyond the depth of rooting. The subsurface zone of groundwater evaporation and absorption is where the wisps of soil carbonate form, then coalesce into nodules and, eventually, thick layers.
Solonization Solonization is a process by which clays rich in soda are formed within the soils of dry climates (Aridisols), where the hydrolytic mobilization of major cations (Ca2þ, Mg2þ, Naþ, Kþ) is weak. Hydrolysis removes cations from soils by lixiation in humid climates, but in dry climates the acidity created by soil respiration after rain storms is sufficient to remove cations from minerals such as feldspar without leaching them from the profile. Solonized soils commonly contain carbonate nodules of dolomite or low-magnesium calcite, formed by calcification, as well as salts of halite and gypsum, formed by salinization. Solonized soils have illitic clays rich in potassium and smectite clays rich in sodium, and the progress of solonization can be assessed by measuring the pH (which is usually around 9–10), by chemical analysis, or by X-ray diffraction to determine the mineral composition. A field indicator of solonization is the presence of domed columnar peds that run through most of the subsurface (natric or Bn) horizon of the soil (Figure 8). The sodium-smectite clays of solonized soils have some shrink–swell capacity, meaning that they form prismatic cracks as the soil dries out and swelling or domed tops to the prisms when the soil is wet. Solonization is common around desert playa lakes and salinas and in coastal soils affected by saltwater spray.
Solodization Solodization is intermediate between solonization and lessivage, and creates profiles with acidic-toneutral near-surface horizons but alkaline subsurface
SOILS/Modern 201
thick benches, or vertical cracks depending on the local climatic conditions. Soil mixing results from the expansion of water to ice during winter freezing and the relaxation of the deformation on summer melting. Ice-wedge polygons, for example, are wide polygonal cracks that are filled with ice in winter but can be filled with layered sediments in water during the summer in climates where the mean annual temperature is less than 4 C. Sand-wedge polygons form in colder climates where the mean annual temperature is less than 12 C; here, summer melting of ice is limited and sediment fills cracks between the ice and soil in a series of near vertical layers.
Conclusion
Figure 8 Domed columnar peds produced by solonization in an inceptisol near Narok, Kenya. Hammer handle is 25 cm.
horizons dominated by sodium-smectite. Solodized soils have domed columnar clayey peds in a subsurface (Bn) horizon, but these are sharply truncated by a granular leached (E) horizon. Solodization occurs in desert soils (Aridisols) with better vegetative cover and a more humid climate than solonized soils.
Salinization Salinization is the precipitation of salts in soils (Figure 3) and is found mostly in desert soils (Aridisols). The most common salts are halite and gypsum, which can form either as clear crystals within soil cracks or as sand crystals that engulf the pre-existing soil matrix. Salts are easily dissolved by rain and so accumulate in regions where there is a marked excess of evaporation over precipitation, which is generally less than 300 mm per year. There is a strong relationship between mean annual precipitation and the depth of leaching of salts in soils. Salinized soils are sparsely vegetated or lack vegetation, and occur in playa lakes, sabkhas, and salinas. Although these are commonly regarded as depositional environments, they are significant soil environments as well.
Cryoturbation Cryoturbation is the mixing of soils by the freezing and thawing of ground ice. The ice can form disseminated crystals, hair-like threads, thin bands,
Soil-forming processes are varied and complex, and our understanding of them guides the classification, description, and management of soils. The processes are also of interest in simplifying the vast array of chemical reactions, biological processes, and physical effects that create soil. Some processes are more common and widespread than others. Lixiviation and its underlying hydrolysis chemical reaction is perhaps the most important weathering process on Earth, affecting geomorphology, sedimentation, ocean chemistry, and climate. Other processes are restricted to more specific climatic, biotic, geomorphological, geological, and temporal environments, but are no less important in their local environments.
Glossary Alfisol A fertile forested soil with subsurface enrichment of clay. Andisol A volcanic-ash soil. Andisolization A soil-forming process that creates low-density non-crystalline fertile soil from volcanic ash. Anthropic epipedon A soil surface modified by human use. Anthrosolization A soil-forming process involving modification by human activities. Argillic horizon A subsurface horizon of soil enriched in clay. Argilluviation A soil-forming process that involves creating clay and washing it into a subsurface clayey horizon. Aridisol A soil of arid regions, usually containing carbonate nodules. Biocycling The recycling of nutrient elements by biota. Birnessite A non-crystalline mixture of iron and manganese oxides. Entisol A very weakly developed soil.
202 SOILS/Modern
Ferrallitization A soil-forming process involving intense weathering that removes most elements other than iron and aluminium. Gelisol A soil of permafrost regions, usually containing ground ice. Gibbsite An aluminium hydroxide mineral (Al(OH)3). Gilgai A soil microrelief consisting of ridges or mounds alternating with furrows or pits. Gleization A soil-forming process involving chemical reduction of the soil due to waterlogging. Halite A salt mineral (NaCl). Haematite An iron oxide mineral (Fe2O3). Imogolite A colloidal weathering product of volcanic-ash soils. Inceptisol A weakly developed soil. Lessivage A soil-forming process that creates clay and washes it into a subsurface clayey horizon. Lixiviation A soil-forming process that involves leaching nutrient cations from the soil. Melanization A soil-forming process that involves darkening the soil with organic matter. Mollic epipedon A humic fertile crumb-textured soil surface typical of grassland soils. Mollisol A grassland soil with a humic fertile crumbtextured surface. Mukkara A soil structure consisting of festooned and slickensided cracks between uplifted parts of the soil; the subsurface structures below gilgai microrelief. Natric horizon A subsurface horizon of soil enriched in sodium-clay. Oxisol A deeply weathered soil of tropical humid regions. Paludization A soil-forming process involving peat accumulation in waterlogged soils. Ped A clod, a unit of soil structure. Placic horizon Iron- and manganese-stained bands and nodules in soils. Plaggen epipedon A ploughed surface horizon of soils. Podzol A sandy soil with a bleached near-surface horizon. Podzolization A soil-forming process in which acid leaching creates a bleached sandy upper horizon and an iron- or organic-rich subsurface horizon. Siderite An iron carbonate mineral (FeCO3). Solonization A soil-forming process that creates soda-rich clays and domed columnar peds in arid regions. Spodosol A sandy clay-poor soil with an iron- or organic-rich subsurface horizon.
Ultisol A deeply weathered forest soil with subsurface enrichment in clay. Umbric epipedon A humic acidic clayey massive-tolaminar soil surface found in wetland soils. Vertisol Swelling clay soil. Vertization A soil-forming process involving deformation and mixing due to the shrink–swell behaviour of clay during drying and wetting cycles.
See Also Carbon Cycle. Clay Minerals. Engineering Geology: Ground Water Monitoring at Solid Waste Landfills. Sedimentary Environments: Deltas; Deserts. Sedimentary Processes: Glaciers. Soils: Palaeosols. Weathering.
Further Reading Bockheim JG and Gennadiyev AN (2000) The role of soil forming processes in the definition of taxa in soil tax onomy. Geoderma 95: 53 72. Bohn H, McNeal B, and O’Connor G (1985) Soil Chemis try. New York: Wiley. Eisenbeis G and Wichard H (1987) Atlas on the Biology of Soil Arthropods. Berlin: Springer. Jenny H (1941) Factors of Soil Formation. New York: McGraw Hill. Lu¨ ndstrom US, Van Breeman N, and Bain D (2000) The podzolization process: a review. Geoderma 94: 91 107. McFadden LD, Amundson RG, and Chadwick OA (1991) Numerical modelling, chemical and isotopic studies of carbonate accumulation in arid soils. In: Nettleton WD (ed.) Occurrence, Characteristics and Genesis of Car bonate Gypsum and Silica Accumulations in Soils, pp. 17 35. Special Publication 26. Madison: Soil Science Society of America. Markewich HW, Pavich MJ, and Buell GR (1990) Con trasting soils and landscapes of the Piedmont and Coa stal Plain, eastern United States. Geomorphology 3: 417 447. Marshall TJ, Holmes JW, and Rose CW (1996) Soil Physics. Cambridge: Cambridge University Press. Paton TR, Humphreys GS, and Mitchell PB (1995) Soils: A New Global View. London: UCL Press. Retallack GJ (1997) A Colour Guide to Paleosols. Chiches ter: Wiley. Retallack GJ (2001) Soils of the Past. Oxford: Blackwell. Richter DD and Markewitz D (2001) Understanding Soil Change. Cambridge: Cambridge University Press. Sanford RI (1987) Apogeotropic roots in an Amazon rain forest. Science 235: 1062 1064. Soil Survey Staff (2000) Keys to Soil Taxonomy. Blacks burg: Pocahontas Press. Washburn AL (1980) Geocryology. New York: Wiley.
SOILS/Palaeosols 203
Palaeosols G J Retallack, University of Oregon, Eugene, OR, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Palaeosols are ancient soils, formed on landscapes of the past. Most palaeosols have been buried in the sedimentary record, covered by flood debris, landslides, volcanic ash, or lava (Figure 1). Some palaeosols, however, are still at the land surface but are no longer forming in the same way that they did under different climates and vegetation in the past. Climate and vegetation change on a variety of time-scales, and the term relict palaeosol for profiles still at the surface should be used only for such distinct soil materials as laterites among non-lateritic suites of soils (Figure 2). Thus, not all palaeosols are fossil soils or buried soils. An alternative spelling of paleosol has been adopted by the International Quaternary Association. Other terms such as pedoderm and geosol refer to whole landscapes of buried soils. These soil stratigraphical units are named and mapped in order to establish stratigraphical levels. The terms pedotype and soil facies are more or less equivalent and are used to refer to individual palaeosol types preserved within ancient buried landscapes. These terms are used to distinguish one type of palaeosol from another in environmental interpretations of palaeosols. Pedolith, or soil sediment, describes a sediment, as indicated by bedding and other sedimentary features, with distinctive soil clasts, such as ferruginous concretions. Pedoliths are uncommon in sedimentary sequences, because soils are readily eroded to their constituent mineral grains, which retain few distinctive soil microfabrics.
discoloured haloes or mineralized alteration (Figure 3). Both fossilized roots and root traces show the downward tapering and branching of roots. Soils also contain fossil burrows, but these are usually more sparsely branched and parallel-sided than root traces. The distinction between burrows and roots can be blurred in cases where soil animals feed on roots and where roots find an easier passage through the soft fill of burrows. For very old rocks, predating the Early Devonian evolution of roots, the criterion of root traces is of no use in identifying palaeosols.
Figure 1 The subtle colour banding in these cliffs is the result of a sequence of 87 Eocene and Oligocene palaeosols in 143 m of nonmarine silty claystones exposed in the Pinnacles area of Badlands National Park, South Dakota, USA.
Recognition of Palaeosols Palaeosols buried in sedimentary and volcaniclastic sequences can be difficult to distinguish from enclosing sediments, tuffs, or lavas and were not widely recognized before about 20 years ago. Three features of palaeosols in particular aid their identification: root traces, soil horizons, and soil structure. Soil is often defined as the medium of plant growth. Geological and engineering definitions of soil are broader, but fossilized roots and traces of their former paths through the soil are universally accepted as diagnostic of palaeosols. Not all palaeosol root traces are permineralized or compressed original organic matter: some are tortuous infillings of clay with
Figure 2 The red rock exposures to the left on the beach are a lateritic palaeosol of Middle Miocene age. Even though these horizons are at the surface, they are considered to be palaeosols because soil horizons of this type are not currently forming in this area. The red rock in the background is a sequence of Early Triassic palaeosols in Bald Hill Claystone, near Long Reef, New South Wales, Australia.
204 SOILS/Palaeosols
Figure 4 Two successive palaeosols overlain sharply by vol canic grits show crumb structured organic surfaces (A horizon) over calcareous nodule studded subsurfaces (Bk horizon). In the upper right corner is a comparable modern soil (Middle Miocene fossil quarry near Fort Ternan, Kenya).
Figure 3 The sharply truncated top and abundant drab haloed root traces (A horizon) petering out downwards into red clays tone (Bt horizon) are soil horizons of a palaeosol (Long Reef clay palaeosol, Early Triassic, Bald Hill Claystone, near Long Reef, New South Wales, Australia).
Palaeosols also have recognizable soil horizons, which differ from most kinds of sedimentary bedding in their diffuse contacts downwards from the sharp upper truncation of the palaeosol at the former land surface. Palaeosol horizons, like soil horizons, are seldom more than a metre thick and tend to follow one of a few set patterns. Subsurface layers enriched in clay are called Bt horizons in the shorthand of soil science (Figure 3). Unlike a soil, in which clayeyness can be gauged by resistance to the shovel or plasticity between the fingers, clayeyness in palaeosols that have been turned to rock by burial compaction must be evaluated by petrographic, X-ray, or geochemical techniques. Subsurface layers enriched in pedogenic micrite are called Bk horizons in the shorthand of soil science and are generally composed of hard calcareous nodules or benches in both soils and lithified palaeosols (Figure 4). A final distinctive feature of palaeosols is soil structure, which varies in its degree of expression and
replaces sedimentary structures such as bedding planes and ripple marks, metamorphic structures such as schistosity and porphyoblasts, and igneous structures such as crystal outlines and columnar jointing. Because they lack such familiar geological structures, palaeosols are commonly described as featureless, massive, hackly, or jointed. Palaeosols, like soils, have distinctive systems of cracks and clods. The technical term for a natural soil clod is a ped, which can be crumb, granular, blocky, or columnar, among other shapes. Peds are bounded by open cracks in a soil and by surfaces that are modified by plastering over with clay, by rusting, or by other alterations. These irregular altered surfaces are called cutans, and they are vital in recognizing soil peds in palaeosols that have been lithified so that the original cracks are crushed. The rounded 3–4 mm ellipsoidal crumb peds of grassland soils and palaeosols (Figure 4) are quite distinct from the angular blocky peds of forest palaeosols (Figure 3). Common cutans in soils and palaeosols include rusty alteration rinds (ferrans) and laminated coatings of washed-in clay (argillans). Cutans and other features of lithified palaeosols are best studied in petrographic thin sections and by electron microprobing and scanning electron microscopy. Some petrographic fabrics, such as the streaky birefringence of soil clays when viewed under crossed Nicols or sepic plasmic fabric, are diagnostic of soils and palaeosols.
Alteration of Soils after Burial Palaeosols are seldom exactly like soils because of alteration after burial or exposure to additional weathering, and this can compromise their interpretation and identification with modern soils. Palaeosols,
SOILS/Palaeosols 205
like sediments, can be altered by a wide array of burial processes: cementation with carbonate, haematite, or silica; compaction due to pressure or overburden; thermal maturation of organic matter; and metamorphic recrystallization and partial melting. These high-pressure and high-temperature alterations of palaeosols are not as difficult to disentangle from processes of original soil formation as are three common early modifications: burial decomposition, burial reddening, and burial gleization. Some soils are buried rapidly by chemically reducing swamps or thick lava flows, preserving most of their organic matter. In contrast, many palaeosols are covered thinly by floodborne silt or colluvium, and their buried organic matter is then decomposed by aerobic bacteria and fungi deep within the newly forming soil of the palaeosol sequence. For this reason many palaeosols have much less organic carbon (fractions of a weight per cent) than comparable modern soils (usually 5–10% by weight of carbon at the surface). Thus palaeosol A horizons are seldom as dark as soil surface horizons, and must be inferred from the abundance of roots rather than from colour and carbon content. Soils vary considerably in their degree of redness, but most palaeosols are red to reddish brown from haematite (iron oxide) or occasionally yellowish brown from goethite (iron hydroxide). Soils become redder from the poles to the tropics, from moderately drained to well drained sites, and with increasing time for development, as iron hydroxides are dehydrated to oxides. The dehydration of iron hydroxides continues with the burial of soils, so that red palaeosols are not necessarily tropical, unusually well drained, or especially well developed. In river-valley and coastal sedimentary sequences with abundant palaeosols, formerly well-drained soils can find themselves subsiding below the water table with root traces and humus largely intact. Burial gleization is a process in which organic matter is used by microbes as a fuel for the chemical reduction of yellow and red iron oxides and hydroxides. Comparable processes of biologically induced chemical reduction are common in swamp soils, but superimposition of this process on the organic parts of formerly well-drained soils produces striking effects in some palaeosols. The whole A horizon is turned grey, with grey haloes extending outwards from individual roots, which diminish in abundance down the profile (Figure 3). Burial gleization is especially suspected when the lower parts of the profile are highly oxidized and have deeply penetrating roots, as in welldrained soils, and when there is no pronounced clayey layer that would perch a water table within the soil. The combined effect of burial decomposition, dehydration, and gleization can completely change the
appearance of a soil. The gaudy grey-green Triassic palaeosol shown in Figure 3, for example, was probably modified by all three processes from an originally dark brown over reddish brown forest soil.
Palaeosols and Palaeoclimate Many palaeosols and soils bear clear marks of the climatic regime in which they formed. The Berkeley soil scientist Hans Jenny quantified the role of climate in soil formation by proposing a space-for-climate strategy. What was needed was a carefully selected group of soils, or climosequence, that varied in climate of formation but were comparable in vegetation, parent material, topographical setting and time for formation. He noted that mean annual rainfall and the depth in the profile to calcareous nodules decline from St Louis west to Colorado Springs, in the mid-western USA, but that temperatures and seasonality at these locations are comparable. Also common to all these soils is grassy vegetation on postglacial loess that is about 14 000–12 000 years old. From these soils he derived a climofunction or mathematical relationship between climate and soil features. A 1994 compilation of comparable data showed a clear relationship between the depth from the surface of the soil of carbonate nodules (D in cm) and the mean annual precipitation (P in mm) according to the formula: P ¼ 139:6 þ 6:388D 1:01303D2 Such climofunctions can be used to interpret palaeoclimate from the depth within palaeosols of calcareous nodules (Figure 4), once allowance is made for reduction in depth due to burial compaction. Climatic inferences also can be made from ice deformation features, concretions, clay mineral compositions, bioturbation, and chemical analyses of palaeosols. The thick clayey palaeosol shown in Figure 5 is riddled with large root traces of the kind found under forests and is very severely depleted in elemental plant nutrients such as calcium, magnesium, sodium, and potassium. Comparable modern soils are found at mid-latitudes, yet this palaeosol formed during the Triassic at a palaeolatitude of about 70 S. This palaeoclimatic anomaly indicates pronounced global warming, in this case a postapocalyptic greenhouse effect following the largest mass extinction in the history of life at the Permian-Triassic boundary.
Palaeosols and Ancient Ecosystems Just as soils bear the imprint of the vegetation and other organisms they support, so many aspects of
206 SOILS/Palaeosols
ancient ecosystems can be interpreted from palaeosols. The palaeosols shown in Figure 4, for example, have a dark crumb-textured surface horizon with abundant fine (1–2 mm) roots, comparable to the modern grassland soil seen forming on the outcrop to the upper left. Forest soils, in contrast (Figure 3), have large woody root traces, a blocky structure, and thick subsurface clayey horizons (Bt). In some cases root traces in palaeosols are identifiable, although the species Stigmaria ficoides (Figure 6) is a form genus for roots of a variety of extinct tree lycopsids and not a precisely identified ancient plant. The tabular form of the roots of Stigmaria indicates a poorly drained soil, because roots do not photosynthesize, but rather respire using oxygen from soil air. Tabular, rather than deeply
Figure 5 An unusually warm palaeoclimate is indicated by this palaeosol, which is unusually thick, clayey, and deeply weathered for its palaeolatitude of 70 S and is comparable to soils now forming no further south than 48 S (Early Triassic Feather Conglomerate, Allan Hills, Victoria Land, Antarctica).
reaching, root traces (Figure 3) are characteristic of swamp palaeosols. Some palaeosols also contain fossil leaves, fruits, wood, stones, bones, and teeth. These are direct evidence of soil ecosystems. Unlike fossils in deposits of lakes and shallow seas, fossil assemblages in palaeosols have the advantage of being near the place where the organisms lived. However, the preservation of fossils in palaeosols is seldom as ideal as complete skeletons in river-channel deposits or compressed leaves in carbonaceous shales. The carbon and carbonate contents of palaeosols can be used to evaluate the Eh and pH, respectively, of the palaeosol preservational environments of the fossils.
Palaeosols and Palaeogeography Just as soils vary from mountain tops to coastal swamps, so do palaeosols give clues to their ancient topographical setting. Many palaeosols within sedimentary sequences show clear relationships with deposits of palaeochannels and levees, so that their depositional subenvironment can be inferred from context. Water tables are close to the ground surface in many sedimentary environments, and palaeosols yield important information on their position relative to ancient water tables. Palaeosols formed below the water table include peats and are grey with chemically reduced minerals such as pyrite and siderite. Burrows of crayfish and other aquatic organisms are locally common in waterlogged soils, but burrows of most rodents and beetles are not. Root traces also do not penetrate deeply into waterlogged soils or palaeosols (Figure 6). Deeply penetrating roots and burrows and red oxidized minerals of iron or aluminium are common in formerly well-drained palaeosols (Figure 3). Palaeosols may also reveal upland sedimentary environments such as alluvial and colluvial fans, glacial moraines, river terraces, and erosional gullies (Figure 7). Major geological unconformities often mark erosional landscapes of the past. Rocky cliffs and bedrock platforms are found along geological unconformities, but so are upland palaeosols. For example, the hilly erosional landscape of Lewisian Gneiss in northern Scotland had 1 km of relief (Figure 8).
Palaeosols and their Parent Materials
Figure 6 Swamp forests of tree lycopsids (Stigmaria ficoides) grew in waterlogged soils, in which lack of oxygen forced the roots to form planar mats rather than reaching deeply into the soil (Carboniferous Lower Limestone Coal Group, Victoria Park, Glasgow, Scotland).
The parent material of a soil or palaeosol is the substance from which it formed and can usually be inferred from the less-weathered lower parts of the profile. The parent material may be precisely known if the palaeosol is on metamorphic or igneous rocks (Figure 8), because pedogenic minerals are easily
SOILS/Palaeosols 207
glass than minerals. Volcanic glass weathers to noncrystalline amorphous substances such as imogolite, which confer high fertility from loosely bound phosphorous, potassium, and other plant nutrients. Such soils also have low bulk density and good moisture-retaining properties. Such soils around tropical volcanoes support intensive agriculture, despite the hazards of the nearby active volcano, because they are so much more fertile than surrounding soils. Comparable palaeosols are commonly associated with volcanic arcs of the past (Figure 1).
Figure 7 A palaeogully in a strongly developed sequence of palaeosols (dark coloured) is filled with alluvium including weakly developed palaeosols (Late Triassic Chinle Formation, Petrified Forest National Park, Arizona, USA). The hill in the foreground is 11 m high. Photograph courtesy of Mary Kraus.
Figure 8 The bleached pink palaeosol formed on gneiss to the right (Sheigra palaeosol) is thicker and more deeply weathered than the light green palaeosol formed on amphibolite to the left (Staca palaeosol). Both palaeosols are overlain by red quartz sandstones of the Torridonian Group (Late Precambrian, near Sheigra, Scotland).
distinquished from igneous and metamorphic minerals. Parent material is more difficult to find in palaeosols that are developed from sedimentary parent materials, especially if sedimentary facies reveal erosional relief (Figure 7). In such settings, the sediment is derived from pre-existing soils, whose degree of weathering can be quite varied. The kinds of soils of sediment and rock also can be very different. If soil were a commercial product, economy would dictate manufacturing it from materials that are already similar in chemical composition and physical characteristics. Soils form more readily from sediments than from rocks. Perhaps the most distinctive of parent materials is volcanic ash, because it may consist of more volcanic
Palaeosols and their Times for Formation Soils develop their profiles over time, although some soils, such as peats, also accumulate layer-by-layer in the manner of sediments. Each palaeosol within a sedimentary or volcanic sequence represents a short break in sedimentary accumulation, or diastem, whose duration can be calculated from key features of the soil. The peats that become coal seams in the geological record, for example, cannot accumulate at rates of more than 1 mm year 1 because the roots will be suffocated by stagnant water. Nor can they accumulate at rates of less than 0.5 mm year 1 because aerobic decay will destroy the organic debris as fast as it accumulates. Thus, the durations of coal-bearing palaeosols can be calculated from coal thickness, once compaction is taken into account. Calcareous soils and palaeosols accumulate carbonate at first in wisps and filaments, and later in nodules, which become larger and larger (Figure 4). The size of the nodules thus gives us an idea of the time over which they formed. The development of clayey subsurface horizons is comparable (Figure 3) in that clay becomes more and more abundant over time. The amount of washed-in clay can thus be a guide to the time over which palaeosols formed. From the times for palaeosol formation and the thickness of rock for successive palaeosols it is possible to calculate rates of sediment accumulation. In the badlands of South Dakota, for example, the clayey lower part of the section accumulated at a slower rate than the ashy and silty upper part of the section (Figure 1). Variations in the rate of sediment accumulation can be used to address a variety of tectonic, volcanic, and sequence stratigraphical problems using palaeosols.
Glossary Argillan Clay skin, a kind of planar feature in a soil or cutan formed of clay.
208 SOILS/Palaeosols
Burial decomposition An early diagenetic modification of a palaeosol in which buried organic matter is decayed microbially. Burial gleization An early diagenetic modification of a palaeosol in which buried organic matter fuels microbial chemical reduction of iron oxides and oxyhydraes to ferrous clays, siderite or pyrite. Climofunction A mathematical relationship between a soil feature and a measure of climate. Climosequence A set of soils formed under similar vegetation, topographic setting, parent material and time, but varied climate. Concretion A seggregation of materials in a soil, harder or more cemented than the matrix, with prominent internal concentric banding, for example iron-manganese concretion. Cutan A planar feature within a soil formed by enrichment, bleaching, coating or other alteration, for example a clay skin (argillan). Ferran Ferruginized surface, a kind of planar feature in a soil (cutan) formed by chemical oxidation. Geosol A mappable land surface of palaeosols, a soil stratigraphic unit in the North American Code of Stratigraphic Nomenclature. Nodule A segregation of materials in a soil, harder or more cemented than the matrix, with massive internal fabric, for example caliche nodule. Palaeosol A soil of a landscape of the past: a past surficial region of a planet or similar body altered in place by biological, chemical or physical processes, or a combination of these. Ped A natural aggregate of soil: stable lumps or clods of soil between roots, burrows, cracks and other planes of weakness. Pedoderm A mappable land surface of palaeosols, a soil stratigraphic unit in the Australian Code of Stratigraphic Nomenclature. Pedolith Soil sediment: a seadimentary rock dominated by clasts with the internal microfabrics of soils. Pedotype A kind of palaeosol: an ancient equivalent of soil series of the United States Soil Conservation Service. Perched water table Level of water ponded in a soil by an impermeable subsurface layer. Sepic plasmic fabric Birefringence microfabric: appearance of the fine grained part of a soil or palaeosol in petrographic thin sections viewed under crossed Nicols of wisps or streaks of highly
oriented and highly birefringent clay in a less organized dark matrix.
See Also Carbon Cycle. Clay Minerals. Palaeoclimates. Sedimentary Environments: Depositional Systems and Facies; Alluvial Fans, Alluvial Sediments and Settings. Sedimentary Processes: Karst and Palaeokarst. Sedimentary Rocks: Evaporites. Soils: Modern. Weathering.
Further Reading Delvigne JE (1998) Atlas of Micromorphology of Mineral Alteration and Weathering. Canadian Mineralogist Special Publication 3. Ottawa: Mineralogical Association of Canada. Follmer LR, Johnson GD, and Catt JA (eds.) (1998) Revisitation of concepts in paleopedology. Quaternary International 51/52: 1 221. International Subcommission on Stratigraphic Classifica tion (1994) International Stratigraphic Guide. Boulder: Geological Society of America. Jenny HJ (1941) Factors in Soil Formation. New York: Wiley. Martini IP and Chesworth W (eds.) (1992) Weathering, Soils and Paleosols. Amsterdam: Elsevier. Ollier C (1991) Ancient Landforms. London: Belhaven. Ollier C and Pain C (1996) Regolith, Soils and Landforms. Chichester: Wiley. Reinhardt J and Sigleo WR (1988) Paleosols and Weathering through Geologic Time: Principles and Ap plications. Special Paper 216. Boulder: Geological Society of America. Retallack GJ (1983) Late Eocene and Oligocene Paleosols from Badlands National Park, South Dakota. Special Paper 193. Boulder: Geological Society of America. Retallack GJ (ed.) (1986) Precambrian paleopedology. Precambrian Research 32: 93 259. Retallack GJ (1991) Miocene Paleosols and Ape Habitats of Pakistan and Kenya. New York: Oxford University Press. Retallack GJ (1997) A Colour Guide to Paleosols. Chiches ter: Wiley. Retallack GJ (2001) Soils of the Past. Oxford: Blackwell. Retallack GJ, Bestland EA, and Fremd TJ (2000) Eocene and Oligocene Paleosols of Central Oregon. Special Paper 344. Boulder: Geological Society of America. Thiry M and Simon Coinc¸ on R (eds.) (1999) Palaeo weathering, Palaeosurfaces, and Related Continental Deposits. Oxford: Blackwell. Wright VP (ed.) (1986) Paleosols: their Recognition and Interpretation. Oxford: Blackwell.
SOLAR SYSTEM/The Sun 209
SOLAR SYSTEM Contents The Sun Asteroids, Comets and Space Dust Meteorites Mercury Venus Moon Mars Jupiter, Saturn and Their Moons Neptune, Pluto and Uranus The Sun’s Effective Temperature
The Sun K R Lang, Tufts University, Medford, MA, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Physical Characteristics of the Sun Distance to the Sun
The mean distance of the Sun from the Earth sets the scale of our Solar System and enables us to infer, from other observations, the luminosity, radius, effective temperature, and mass of the Sun. This distance is called the astronomical unit, or AU for short, with a value of 1 AU ¼ 1.49597870 1011 m. At that distance, light from the Sun takes 499.004 782 s to travel to the Earth. By way of comparison, light from the Sun’s nearest stellar neighbour, Proxima Centauri (part of the triple star system Alpha Centauri), takes 4.29 years to reach us. Absolute Solar Luminosity
The Sun’s absolute, or intrinsic, luminosity is designated by the symbol L, where the subscript denotes the Sun. We can infer the Sun’s luminosity from satellite measurements of the total amount of solar energy reaching every square centimetre of the Earth every second, obtaining L ¼ 3.854 1026 W, where a power of 1 W ¼ 1 J s 1.
We can use the Stefan–Boltzmann law, together with the Sun’s size and luminous output, to determine an effective temperature of 5780 K. The temperature of the Sun increases below and above the visible disk (Table 1). Mass of the Sun
The Sun’s gravitational pull holds the solar system together. That is why we call it a solar system: governed by the central Sun with its huge mass. This gravitational attraction keeps the planets in orbit around the Sun, with longer orbital periods at increasing distances from the Sun. And since we know the Earth’s orbital period and mean distance from the Sun, we can weigh the Sun from a distance, obtaining its mass M ¼ 5.9165 1011 (AU)3/P2 ¼ 1.989 1030 kg, where the constant is equal to 4p2/ G, the universal constant of gravitation is G, the semi-major axis of the Earth’s orbit is 1 AU ¼ 1.4959787 1011 m, and the orbital period of the Earth is P ¼ 1 year ¼ 3.1557 107 s. The Sun does not just lie at the heart of our solar system; it dominates it. Some 99.8% of all the matter between the Sun and halfway to the nearest star is contained in the Sun. It is 332 946 times the mass of the Earth. All the objects that orbit the Sun—the planets and their moons, the comets, and the asteroids—add up to just 0.2% of the mass in our solar system.
Radius of the Sun
The Sun’s radius, which can be inferred from its distance and angular extent, has a value of R ¼ 6.955 108 m. That is about 109 times the radius of the Earth.
Composition of the Sun
When the intensity of sunlight is displayed as a function of wavelength, in a spectrum, it exhibits numerous fine dark gaps of missing colours called
210 SOLAR SYSTEM/The Sun
Table 1 The Sun’s physical propertiesa Mean distance, AU Light travel time from Sun to Earth Radius, R Volume Mass, M Escape velocity at photosphere Mean density Solar constant, f Luminosity, L Principal chemical constituents
Age Temperature (center) Temperature (effective) Temperature (photosphere) Temperature (chromosphere) Temperature (corona) Rotation period (equator) Rotation period (60 latitude) Magnetic field (sunspots) Magnetic field (polar)
1.4959787 1011 m 499.004782 s 6.955 108 m (109 Earth radii) 1.412 1027 m3 (1.3 million Earths) 1.989 1030 kg (332 946 Earth masses) 617 km s 1 1409 kg m 3 1366 J s 1 m 2 1366 W m 2 3.854 1026 J s 1 3.854 1026 W (By number (By mass of atoms) fraction) Hydrogen 92.1% X 70.68% Y 27.43% Helium 7.8% Z 1.89% All other 0.1% 4.566 billion years 15.6 million K 5780 K 6400 K 6000 to 20 000 K 2 million to 3 million K 26.8 days 30.8 days 0.1 to 0.4 T 1000 to 4000 G 0.001 T 10 G
Mass density is given in kilograms per cubic metre (kg m 3); the density of water is 1000 kg m 3. The unit of luminosity is joules per second, power is often expressed in watts, where 1.0 W 1.0 J s 1. a
absorption lines. Each chemical element, and only that element, produces a unique set, or pattern, of wavelengths at which the dark absorption lines fall. So these lines can be used to determine the chemical ingredients of the Sun. They indicate that hydrogen is the most abundant element in the visible solar gases. Since the Sun is chemically homogenous, except for its core, a high hydrogen abundance is implied for the entire star, and this was confirmed by subsequent calculations of its luminosity. Hydrogen accounts for 92.1% of the number of atoms in the Sun, and it amounts to 70.68% by mass. Helium, the second-most abundant element in the Sun, accounts for 7.8% of the number of atoms in the Sun, and it amounts to 27.43% by mass. Helium is so rare on Earth that it was first discovered on the Sun. All of the heavier elements in the Sun amount to only 0.1% of the number of atoms, and just 1.89% by mass. Rotation of the Sun
The Sun rotates, or spins, around a rotational axis whose top and bottom mark the Sun’s north and
south poles. Like Earth, the Sun rotates from west to east when viewed from above the north pole, but unlike Earth, different parts of the Sun rotate at different rates. We know from watching sunspots that the visible disk of the Sun rotates faster at the equator than it does at higher latitudes, decreasing in speed evenly towards each pole. Also, because the Earth orbits the Sun, we observe a rotation period that is about a day longer than the true value. The synodic rotation period of the visible solar equator, as observed from Earth, is 26.75 days, while the equatorial region of the visible solar disk is intrinsically spinning about the Sun’s axis once every 25.67 days. Scientists have used sound waves, generated inside the Sun, to show that the differential rotation of the Sun persists to about one-third of the way down inside the Sun, or 220 000 km from the visible disk. Lower down the rotation speed becomes uniform from pole to pole and the rotation rate remains independent of latitude. The Sun’s magnetism is probably generated at the interface between the deep interior, which rotates with one speed, and the overlying gas that spins faster in the equatorial middle. Solar Magnetic Fields
Detailed scrutiny indicates that the visible solar disk often contains dark, ephemeral spots, called sunspots, which can be as large as the Earth. The sunspots appear and disappear, rising out from inside the Sun and moving back into it. Most sunspots remain visible for only a few days; others persist for weeks and even months. Sunspots contain magnetic fields as strong as 0.3 T, or 3000 G, thousands of times stronger than the Earth’s magnetic field. The intense sunspot magnetism chokes off the upward flow of heat and energy from the solar interior, keeping a sunspot thousands of degrees colder than the surrounding gas. The total number of sunspots visible on the Sun varies over an 11-year cycle. At the maximum in the cycle we may find 100 or more spots on the visible disk of the Sun at one time; at sunspot minimum very few of them are seen, and for periods as long as a month none can be found. Since most forms of solar activity are magnetic in origin, they also follow an 11-year cycle. Thus, the sunspot cycle is also known as the solar cycle of magnetic activity. Sunspots are usually found in pairs or groups of opposite magnetic polarity. The magnetic field lines emerge from a sunspot of one polarity, loop through the solar atmosphere above it, and enter a neighbouring sunspot of opposite polarity. The highly magnetized realm in, around, and above bipolar sunspot pairs or groups is a disturbed area called an active
SOLAR SYSTEM/The Sun 211
region; it consists of sunspots and the magnetic loops that connect them. Sunspots are usually oriented roughly parallel to the Sun’s equator, in the east–west direction of the Sun’s rotation. Moreover, sunspot pairs in either the northern or southern hemisphere have the same orientation and polarity alignment, with an exact opposite arrangement in the two hemispheres. The Outer Solar Atmosphere
The visible photosphere, or sphere of light, is the level of the solar atmosphere from which we get our light and heat, and it is the part that we can see with our eyes. The thin chromosphere and extensive corona lie above the visible sharp edge of the photosphere. They can both be seen during a total solar eclipse, when the Moon blocks the intense light of the photosphere. Telescopes called coronagraphs allow us to see the corona by using occulting disks to mask the Sun’s face and block out the photosphere’s glare. Modern solar satellites, such as the Solar and Heliospheric Observatory (SOHO), use coronagraphs to get clear, edge-on views of the corona.
The solar corona has a temperature of millions of degrees kelvin, hundreds of times hotter than the underlying visible solar disk whose effective temperature is 5780 K. Very hot material—such as that within the corona—emits most of its energy at X-ray wavelengths. Also, the photosphere is too cool to emit intense radiation at these wavelengths, so it appears dark under the hot gas. As a result, the million-degree corona can be seen all across the Sun’s face, with high spatial and temporal resolution, in X-rays. Since X-rays are totally absorbed by the Earth’s atmosphere, they must be observed through telescopes in space. This has been done using a soft X-ray telescope on the Yohkoh spacecraft (Figure 1). Yohkoh’s soft X-ray images have demonstrated that the corona contains thin, bright, magnetized loops that shape, mold, and constrain the million-degree gas. Wherever the magnetism in these coronal loops is strongest, the coronal gas in them shines brightly at soft X-ray wavelengths. Not all magnetic fields on the Sun are closed loops. Some of the magnetic fields extend outward, within regions called coronal holes. These extended regions
Figure 1 The Sun in X rays. The bright glow seen in this X ray image of the Sun is produced by ionized gases at a temperature of a few million degrees kelvin. It shows magnetic coronal loops which thread the corona and hold the hot gases in place. The brightest features are called active regions and correspond to the sites of the most intense magnetic field strength. This image of the Sun’s corona was recorded by the Soft X ray Telescope (SXT) aboard the Japanese Yohkoh satellite on 1 February 1992, near the maximum of the 11 year cycle of solar magnetic activity. Courtesy of Gregory L Slater, Gary A Linford, and Lawrence Shing, NASA, ISAS, Lockheed Martin Solar and Astrophysics Laboratory, National Astronomical Observatory of Japan, and University of Tokyo.
212 SOLAR SYSTEM/The Sun
have so little hot material in them that they appear as large dark areas seemingly devoid of radiation at X-ray wavelengths. Coronal holes are nearly always present at the Sun’s poles, and are sometimes found at lower solar latitudes. The open magnetic fields in coronal holes do not return directly to another place on the Sun, allowing charged particles to escape the Sun’s magnetic grasp and flow outwards into surrounding space.
Explosions on the Sun Solar Flares
Sudden and brief explosions, called solar flares, rip through the atmosphere above sunspots, releasing an incredible amount of energy, amounting to as much as a million, billion, billion (1024) joules in just a few minutes. All of this power is created in a relatively compact explosion, comparable in total area to an Earth-sized sunspot. For a short time, usually about 10 minutes, a flare is heated to tens of millions of degrees kelvin. The explosion floods the solar system with intense radiation across the full electromagnetic spectrum, from the shortest X-rays to the longest radio waves, and hurls high-energy electrons and protons out into interplanetary space. Despite the powerful cataclysm, most solar flares are only minor perturbations in the total amount of emitted sunlight. Routine visual observations of solar explosions are only made possible by tuning into the red emission of hydrogen alpha, designated Ha, at a wavelength of 656.3 nm, and rejecting all the other colours of sunlight. Since solar flares are very hot, they emit the bulk of their energy at X-ray wavelengths, and for a short while, a large flare can outshine the entire Sun in X-rays. The energetic electrons that produce the impulsive, flaring X-ray emission also emit radio waves known as a radio burst to emphasize its brief, energetic, and explosive characteristics. A solar flare can also outshine the entire Sun at radio wavelengths. There are more flares near the peak of the 11-year cycle of magnetic activity, but this does not mean that sunspots cause solar flares. They are instead energized by the powerful magnetism associated with sunspots. When these magnetic fields become contorted, they can suddenly and explosively release pent-up magnetic energy as a solar flare, with a main energy release in the corona just above sunspots. The energy is apparently released when magnetized coronal loops, driven by motions beneath them, meet to touch each other and connect. If magnetic fields of opposite polarity are pressed together, an instability takes place and the fields partially
annihilate each other, releasing energy to power the explosion. Coronal Mass Ejections
A coronal mass ejection (CME) is a giant magnetic bubble that rapidly expands to rival the Sun in size. Each time a mass ejection rises out of the corona, it carries away up to 50 billion tons (5 1013 kg) of coronal material. Its associated shocks also accelerate and propel vast quantities of high-speed particles ahead of them. CMEs release about as much energy as a solar flare. However, most of the energy of a mass ejection goes into the kinetic energy of the expelled material, whereas a flare’s energy is mainly transferred into accelerated particles that emit intense X-ray and radio radiation. Coronal mass ejections are detected during routine visible-light observations of the corona from spacecraft such the SOHO. With a disk in the centre to block out the Sun’s glare, the coronagraph is able to show huge pieces of the corona blasted out from the edge of the occulted photosphere (Figure 2). Like sunspots, solar flares, and other forms of solar activity, coronal mass ejections occur with a frequency that varies in step with the 11-year cycle. A few coronal mass ejections balloon out of the corona per day, on average, during activity maximum, and the rate decreases by about an order of magnitude by sunspot minimum. The triggering mechanism for CMEs seems to be related to large-scale interactions of the magnetic field in the low solar corona. This magnetism is continuously emerging from inside the Sun, and disappearing back into it, driven by the Sun’s 11-year cycle of magnetic activity. The release of a coronal mass ejection appears to be one way that the solar atmosphere reconfigures itself in response to these slow magnetic changes.
The Sun’s Winds Basic Properties of the Solar Wind
The tenuous solar atmosphere expands out in all directions, filling interplanetary space with a ceaseless wind that is forever blowing from the Sun. This solar wind is mainly composed of electrons and protons, set free from the Sun’s abundant hydrogen atoms, but it also contains heavier ions and magnetic fields. This perpetual solar gale brushes past the planets and engulfs them, carrying the Sun’s atmosphere out into interstellar space at the rate of a million tons (106 tons ¼ 109 kg) every second. The relentless wind has never stopped blowing during the more than three decades that it has been
SOLAR SYSTEM/The Sun 213
Figure 2 Coronal mass ejection. A huge coronal mass ejection is seen in this coronagraph image, taken on 27 February 2000 with the Large Angle Spectrometric Coronagraph (LASCO) on the Solar and Heliospheric Observatory (SOHO ). The white circle denotes the edge of the photosphere, so this mass ejection is about twice as large as the visible Sun. The black area corresponds to the occulting disk of the coronagraph that blocks intense sunlight and permits the corona to be seen. About one hour before this image was taken, another SOHO instrument, the Extreme Ultraviolet Imaging Telescope (EIT), detected a filament eruption lower down near the solar chromosphere. Courtesy of the SOHO LASCO consortium. SOHO is a project of international cooperation between ESA and NASA.
Table 2 Mean values of solar wind parameters at the Earth’s orbita Parameter
Mean value
Particle density, N
N 10 million particles m
Velocity, V Flux, F Temperature, T Magnetic field strength, H
3
(5 million electrons and 5 million protons) V 375 000 m s 1 and V 750 000 m s 1 F 1012 to 1013 particles m 2 s 1 T 120 000 K (protons) to 140 000 K (electrons) H 6 10 9 T 6 nT 6 10 5 G
a
These solar wind parameters are at the mean distance of the Earth from the Sun, or at one astronomical unit, 1 AU, where 1 AU 1.496 1011 m.
The charged particles in the solar wind drag the Sun’s magnetic fields with them. While one end of the interplanetary magnetic field remains firmly rooted in the photosphere and below, the other end is extended and stretched out by the radial expansion of the solar wind. The Sun’s rotation bends this radial pattern into an interplanetary spiral shape within the plane of the Sun’s equator. This spiral pattern has been confirmed by tracking the radio emission of highenergy electrons emitted during solar flares (Figure 3), as well as by spacecraft that have directly measured the interplanetary magnetism. Origin of the Sun’s Winds
observed with spacecraft. Two winds are always detected—a fast, uniform wind blowing at about 750 km s 1 and a variable, gusty slow wind moving at about half that speed. By the time it reaches the Earth’s orbit, the solar wind has been diluted by its expansion into the increasing volume of space to about 5 million electrons and 5 million protons per cubic meter (Table 2).
The million-degree coronal gas creates an outward pressure that tends to oppose the inward pull of the Sun’s gravity. At great distances, where the solar gravity weakens, the hot protons and electrons in the corona overcome the Sun’s gravity and accelerate away to supersonic speed. Instruments aboard the Ulysses spacecraft conclusively proved that a relatively uniform, fast wind pours out at high latitudes near the solar poles, and
214 SOLAR SYSTEM/The Sun
Figure 3 Spiral path of interplanetary electrons. The trajectory of flare electrons in interplanetary space as viewed from above the polar regions using the Ulysses spacecraft. As the high speed electrons move out from the Sun, they excite radiation at successively lower plasma frequencies; the numbers denote the observed frequency in kilohertz (kHz). Since the flaring electrons are forced to follow the interplanetary magnetic field, they do not move in a straight line from the Sun to the Earth, but instead move along the spiral pattern of the interplanetary magnetic field, shown by the solid curved lines. The squares and crosses show Ulysses radio mea surements of type III radio bursts on 25 October 1994 and 30 October 1994. The approximate locations of the orbits of Mercury, Venus, and the Earth are shown as circles. Courtesy of Michael J Reiner. Ulysses is a project of international collaboration between ESA and NASA.
that a capricious, gusty, slow wind emanates from the Sun’s equatorial regions at activity minimum. Comparisons with Yohkoh soft X-ray images showed that much, if not all, of the high-speed solar wind flows out of the open magnetic fields in polar coronal holes, at least during the minimum in the 11-year cycle of magnetic activity. In addition, instruments aboard SOHO have shown that the strongest high-speed flows gush out of a magnetic network at the bottom of coronal holes near the Sun’s poles. Comparisons of Ulysses data with coronagraph images pinpointed the equatorial coronal streamers as the birthplace of the slow and sporadic wind during the minimum in the 11-year cycle.
the heliosphere—from the Greek word ‘helios’ for ‘Sun’. Within the heliosphere, physical conditions are dominated, established, maintained, modified, and governed by the magnetic fields and charged particles in the solar wind. The solar wind’s domain extends out to about 150 times the distance between the Earth and Sun, marking the outer boundary of the heliosphere and the edge of our solar system. Out there, the solar wind has become so weakened by expansion that it is no longer dense or powerful enough to repel the ionized matter and magnetic fields coursing between the stars.
The Heliosphere
Radiation from the Sun
A solar gale carries the Sun’s rarefied atmosphere past the planets and out to the space between the stars, creating a large cavity in interstellar space called
The Sun emits radiation that carries energy through space as waves. Different types of solar radiation differ in their wavelength, although they propagate
The Sun–Earth Connection
SOLAR SYSTEM/The Sun 215
at the same speed—299 792 458 m s 1, the velocity of light. Our eyes detect a narrow range of these wavelengths that is collectively called visible radiation. From long to short waves, the colours of visible sunlight correspond to red, orange, yellow, green, blue, and violet. Red light has a wavelength of about 7 10 7 m, or 700 nm, and violet waves are about 400 nm long. Although the most intense radiation from the Sun is emitted at visible wavelengths, it emits less luminous radiation at invisible wavelengths that include the infrared and radio waves, with wavelengths longer than that of red light, and ultraviolet (UV), X-rays, and gamma (g) rays, which have wavelengths shorter than that of violet light. Radio waves have wavelengths between 0.001 and 10 m, and they pass easily through the atmosphere, even on cloudy days or in stormy weather. The infrared part of the Sun’s spectrum is located between the radiowave region and the red part of the visible region. Atmospheric molecules, such as carbon dioxide and water vapor, absorb infrared radiation from the Sun. The short-wavelength, ultraviolet radiation from the Sun is sufficiently energetic to tear electrons or atoms off many of the molecular constituents of the Earth’s atmosphere, particularly in the stratosphere where ozone is formed. The X-ray region of the Sun’s spectrum extends from a wavelength of one-hundred billionth (10 11) of a meter, which is about the size of an atom, to the short-wavelength side of the ultraviolet. X-ray radiation is so energetic that it is usually described in terms of the energy it carries. The X-ray region lies between 1 and 100 keV (kiloelectron volts) of energy, where 1 keV ¼ 1.6 10 16 J. The atmosphere effectively absorbs most of the Sun’s ultraviolet radiation and all of its X-rays. Varying Solar Irradiance of Earth
The total amount of the Sun’s life-sustaining energy is called the ‘solar constant’, and it is defined as the total amount of radiant solar energy per unit time per unit area reaching the top of the Earth’s atmosphere at the Earth’s mean distance from the Sun. Satellites have been used to accurately measure the solar constant, obtaining a value of f ¼ 1366.2 W m 2, where the power of one watt is equivalent to one joule per second and the uncertainty in this measurement is 1.0 W m 2. The total power received at any square metre of the Earth’s surface, known as the solar insolation, is much less than the solar constant. This is due to the absorption of sunlight in the terrestrial atmosphere, as well as the time of day.
The solar constant is almost always changing, in amounts of up to a few tenths of a per cent and on time-scales from 1 s to 20 years. This inconstant behaviour can be traced to changing magnetic fields in the solar atmosphere, and its variation tracks the 11-year cycle of magnetic activity (Figure 4). There are enormous changes in the Sun’s radiation at the short ultraviolet and X-ray wavelengths that contribute only a tiny fraction of the Sun’s total luminosity. The ultraviolet emission doubles from the minimum to maximum of the 11-year cycle, while the X-ray brightness of the corona increases by a factor of 100. Global Warming and Cooling by the Sun
The brightening and dimming of the Sun dominated our climate for two centuries, from 1600 to 1800. Cooling by hazy emission from volcanoes next played an important role, but the Sun noticeably warmed the climate for another century, from 1870 to 1970. After that, heat-trapping gases apparently took control of our climate. Global warming by the greenhouse effect is probably responsible for this recent, unprecedented rise in temperature. If current emissions of carbon dioxide and other greenhouse gases go unchecked, the average surface temperature of the globe will rise by about 2 C, making the Earth hotter than it has been in millions of years. The varying Sun may offset some of this warming. Observations of other stars indicate that the Sun luminosity could vary by much more than that observed by satellites so far, producing dramatic changes in the Earth’s climate on time-scales of hundreds and thousands of years. Radioactive isotopes found in both tree rings and ice cores indicate that the Sun’s activity has fallen to unusually low levels at least three times during the past one thousand years, each drop apparently corresponding to a long, cold spell on Earth of roughly a century in duration. Further back in time, during the past one million years, our climate has been dominated by the recurrent ice ages, each lasting about 100 000 years. These glaciations begin and end in a relatively short interval of unusual warmth, called an interglacial, lasting 10 000 or 20 000 years, when the glaciers retreat. We now live in such a warm interglacial interval. The rhythmic alteration of glacial and interglacial intervals is related to periodic alterations in the amount and distribution of sunlight received by Earth over tens of thousands of years. Our Sun-Layered Atmosphere
Our thin atmosphere is pulled close to the Earth by its gravity, and suspended above the ground by
216 SOLAR SYSTEM/The Sun
Figure 4 Variations in the solar constant. Observations with very stable and precise detectors on several Earth orbiting satellites show that the Sun’s total radiative input to the Earth, termed the solar irradiance, is not a constant, but instead varies over time scales of days and years. Measurements from five independent space based radiometers since 1978 (top) have been combined to produce the composite solar irradiance (bottom) over two decades. They show that the Sun’s output fluctuates during each 11 year sunspot cycle, changing by about 0.1% between maximums (1980 and 1990) and minimums (1987 and 1997) in magnetic activity. Temporary dips of up to 0.3% and a few days’ duration are due to the presence of large sunspots on the visible hemisphere. The larger number of sunspots near the peak in the 11 year cycle is accompanied by a rise in magnetic activity that creates an increase in luminous output that exceeds the cooling effects of sunspots. Here the total irradiance just outside our atmosphere, called the solar constant, is given in units of watts per square metre, where 1 W 1 J s 1. The capital letters are acronyms for the different radiometres, and offsets among the various datasets are the direct result of uncertainties in their scales. Despite these offsets, each dataset clearly shows varying radiation levels that track the overall 11 year solar activity cycle. Courtesy of Claus Frohlich.
molecular motion. The atmosphere near the ground is compacted to its greatest density and pressure by the weight of the overlying air. At greater heights there is less air pushing down from above, so the compression is less and the density and pressure of the air falls off into the near vacuum of space. The temperature of the air falls and rises in two full cycles at increasing altitudes, and the temperature increases are caused by the Sun’s energetic radiation (Figure 5). The temperature above the ground tends to fall at higher altitudes where the air expands in the lower pressure and becomes cooler. The average air temperature drops below the freezing point of water (273 K) about 1 km above the Earth’s surface, and bottoms out at roughly 10 times this height. The temperature increases at greater heights, within the stratosphere, where the Sun’s invisible
ultraviolet radiation warms the gas and helps make ozone. This ozone layer protects us by absorbing most of the ultraviolet and keeping its destructive rays from reaching the ground, where it can cause eye cataracts and skin cancer. Due to the Sun’s variable ultraviolet radiation, the total global amount of ozone becomes enhanced, depleted, and enhanced again from 1 to 2% every 11 years, modulating the protective ozone layer at a level comparable to human-induced ozone depletion by chemicals wafting up from the ground. Monitoring of the expected recovery of the ozone layer from outlawed, man-made chemicals will therefore require careful watch over how the Sun is changing the layer from above. The temperature declines rapidly with increasing height just above the stratosphere, reaching the lowest levels in the entire atmosphere, but the temperature
SOLAR SYSTEM/The Sun 217
Figure 5 Sun layered atmosphere. The pressure of our atmos phere (right scale) decreases with altitude (left scale). This is because fewer particles are able to overcome the Earth’s gravi tational pull and reach higher altitudes. The temperature (bottom scale) also decreases steadily with height in the ground hugging troposphere, but the temperature increases in two higher regions heated by the Sun. They are the stratosphere, with its critical ozone layer, and the ionosphere. The stratosphere is mainly heated by ultraviolet radiation from the Sun, and the ionosphere is created and modulated by the Sun’s X ray and extreme ultraviolet radiation.
rises again within the ionosphere, reaching temperatures that are hotter than the ground. The ionosphere is created and heated by absorbing the extreme ultraviolet and X-ray portions of the Sun’s energy. This radiation tears electrons off the atoms and molecules in the upper atmosphere, thereby creating ions and free electrons not attached to atoms. At a given height in the ionosphere, the temperature, the density of free electrons, and the density of neutral, unionized atoms all increase and decrease in synchronism with solar activity over its 11-year cycle. The Earth’s Magnetosphere
Invisible magnetic fields, produced by currents in the Earth’s molten core, emerge out of the Earth’s south geographic polar regions, loop through nearby space, and re-enter at the north polar regions. The surface equatorial field strength is 0.000031 T, or 31 000 nT, and the field strength decreases at greater distances from the Earth. Yet, the Earth’s magnetism is strong enough to deflect the Sun’s wind away from the Earth, forming the magnetosphere (Figure 6). The magnetosphere of the Earth, or any other planet, is that region surrounding the planet in which its magnetic field dominates the motions of energetic charged particles such as electrons, protons, and other ions. It is also
the volume of space from which the main thrust of the solar wind is excluded. The solar wind pushes the terrestrial magnetic field towards the Earth on the dayside that faces the Sun, compressing the outer magnetic boundary and forming a bow shock at about 10 times the Earth’s radius. Also the Sun’s wind drags and stretches the Earth’s magnetic field out into a long magnetotail on the night side of our planet. The magnetic field points roughly towards the Earth in the northern half of the tail and away in the southern. The field strength drops to nearly zero at the centre of the tail where the opposite magnetic orientations lie next to each other and currents can flow. Some of the energetic particles outside the magnetosphere do manage to penetrate it, especially in the magnetotail. When the solar and terrestrial magnetic fields touch each other in the magnetotail, it can catapult the outer part of the tail downstream and propel the inner part back towards Earth. The inner magnetosphere is always filled with electrons and protons, trapped within two torus-shaped belts that encircle the Earth’s equator but do not touch it. These regions are often called the inner and outer Van Allen radiation belts, named after James A Van Allen (1914–) who discovered them in 1958. The inner belt is about 1.5 Earth radii from planet centre, and the outer belt is located at about 4.5 Earth radii, where the Earth’s radius is 6378 km. Intense Geomagnetic Storms
Significant variations in the Earth’s magnetic field, lasting seconds to days, are known as geomagnetic storms. The great, sporadic geomagnetic storms, which shake the Earth’s magnetic field to its very foundations, can produce magnetic fluctuations as large as 1.6% at mid-terrestrial latitudes, or 500 nT, compared with the Earth’s equatorial field strength of 31 000 nT. Solar wind disturbances driven by exceptionally fast coronal mass ejections produce the most intense geomagnetic storms. The Earth intercepts about 70 coronal mass ejections per year when solar activity is at its peak, and less than 10 will have the punch needed to produce large, geomagnetic storms. These mass ejections plow through the solar wind, driving a huge shock wave far ahead of them. When directed at the Earth, these shocks ram into the terrestrial magnetic field and trigger the initial phase, or sudden commencement, of an intense geomagnetic storm a few days after the mass ejection leaves the Sun. Strong interplanetary magnetic fields are also generated by fast coronal mass ejections (see Magnetostratigraphy). It is their intense magnetism and high speed that account for the main phase of a powerful
218 SOLAR SYSTEM/The Sun
Figure 6 Magnetosphere. The Earth’s magnetic field carves out a hollow in the solar wind, creating a protective cavity, called the magnetosphere. A bow shock forms at about 10 Earth radii on the sunlit side of our planet. Its location is highly variable since it is pushed in and out by the gusty solar wind. The magnetopause marks the outer boundary of the magnetosphere, at the place where the solar wind takes control of the motions of charged particles. The solar wind is deflected around the Earth, pulling the terrestrial magnetic field into a long magnetotail on the night side. Plasma in the solar wind is deflected at the bow shock (left), flows along the magnetopause into the magnetic tail (right), and is then injected back towards the Earth and Sun within the plasma sheet (centre). The Earth, its auroras, atmosphere, and ionosphere and the two Van Allen radiation belts all lie within this magnetic cocoon.
geomagnetic storm, provided that the magnetic alignment is right. The Earth’s field is generally directed northwards in the outer dayside magnetosphere, so a fast coronal mass ejection is more likely to merge and connect with the terrestrial field if it points in the opposite southward direction. Moderate Geomagnetic Activity
Moderate mid-latitude magnetic fluctuations of about 0.1%, or tens of nanoTesla, last a few hours, and they are most noticeable near the minimum of the 11-year solar activity cycle. They have a 27-day repetition period, corresponding to the rotation period of the Sun at low solar latitudes when viewed from the moving Earth. The recurrent activity is linked to long-lived, highspeed streams in the solar wind that emanate from coronal holes. When this fast wind overtakes the slow-speed, equatorial one, the two wind components interact, producing shock waves and intense magnetic fields that rotate with the Sun, and periodically sweep past the Earth, producing moderate geomagnetic activity every 27 days. The Auroras
The northern or southern lights, named the ‘aurora borealis’ and ‘aurora australis’ in Latin, are one of the most magnificent and earliest-known examples of
solar–terrestrial interaction. They illuminate the cold, dark Arctic and Antarctic skies with curtains of green and red light that flicker across the night sky far above the highest clouds. Spacecraft look down on the auroras from high above, showing an oval centred at each magnetic pole (Figure 7). An observer on the ground sees only a small, changing piece of the aurora oval. The reason that auroras are usually located near the polar regions is that the Earth’s magnetic fields guide energetic electrons there. The high-speed electrons move down along the Earth’s magnetic field lines into the upper polar atmosphere, colliding with oxygen and nitrogen. The pumped-up atoms or molecules fluoresce, giving up the energy acquired from the electrons and emitting a burst of light. The electrons that cause the auroras come from the Earth’s magnetic tail and are also energized locally within the magnetosphere. The rare, bright, auroras seen at low terrestrial latitudes only become visible during very intense geomagnetic storms. Space Weather
Down here on the ground, we are shielded from the direct onslaught of solar explosions and the solar wind by the Earth’s atmosphere and magnetic fields, but out in deep space there is no protection. Energetic charged particles hurled out from intense solar flares
SOLAR SYSTEM/The Sun 219
Figure 7 The auroral oval. The POLAR spacecraft looks down on an aurora from high above the Earth’s north polar region on 22 October 1999, showing the northern lights in their entirety. The glowing oval, imaged in ultraviolet light, is 4500 km across. The most intense aurora activity appears in bright red or yellow. It is typically produced by magnetic reconnection events in the Earth’s magnetotail, on the night side of the Earth. Courtesy of the Visible Imaging System, University of Iowa and NASA.
can seriously damage satellites, including their solar cells and electronic components, and even kill an unprotected astronaut. The high-speed protons and electrons follow a narrow, curved path once they leave the Sun, guided by the spiral structure of the interplanetary magnetic field, so they must be emitted from active regions near the west limb and the solar equator to be magnetically connected with the Earth. Solar flares emitted from other places on the Sun are not likely to hit Earth, but they could be headed towards interplanetary spacecraft, the Moon, Mars, or other planets. The most energetic flare particles can travel from the Sun to the Earth in just 8 minutes, moving at nearly the velocity of light. Coronal mass ejections move straight out of the Sun, energizing particles over large regions in interplanetary space. Mass ejections are most likely to hit the Earth if they originate near the centre of the solar disk, as viewed from the Earth, and are sent directly towards the planet. They take about 4 days to travel from the Sun to the Earth, moving at a typical speed of about 400 km s 1. The strong blast of X-rays and ultraviolet radiation from a solar flare alters the Earth’s atmosphere, transforming the ionosphere, which reflects radio waves to distant locations on Earth. During moderately intense flares, radio communications can be silenced over the Earth’s entire sunlit hemisphere, disrupting contact with airplanes flying over oceans or remote countries.
The enhanced ultraviolet and X-ray radiation from solar flares also heats the atmosphere and causes it to expand, and similar or greater effects are caused by coronal mass ejections that produce major geomagnetic storms. The expansion of the terrestrial atmosphere brings higher densities to a given altitude, increasing the drag exerted on a satellite, pulling it to a lower altitude, and causing a premature and fatal spiral towards the Earth. When a coronal mass ejection slams into the Earth, the force of impact can push the bow shock, at the dayside of the magnetosphere, down to half its usual distance of about 10 times the Earth’s radius, compressing the magnetosphere below the orbits of geosynchronous satellites and exposing them to the damaging effects of the full brunt of the gusty solar wind. During an intense geomagnetic storm, associated with a colliding coronal mass ejection, strong electric currents flow in the ionosphere. They induce potential differences in the ground below them, and produce strong currents in any long conductor such as a power line. These currents can blow circuit breakers, overheat and melt the windings of transformers, and cause massive failures of electrical distribution systems. A coronal mass ejection can thereby plunge major urban centres, like New York City or Montreal, into complete darkness, causing social chaos and threatening safety. Our technological society has become so vulnerable to the potential devastation of these storms in space that national centres employ space weather
220 SOLAR SYSTEM/Asteroids, Comets and Space Dust
forecasters to continuously monitor the Sun from ground and space to warn of threatening solar activity.
See Also Earth: Orbital Variation (Including Milankovitch Cycles). Gaia. Magnetostratigraphy. Palaeoclimates. Tertiary To Present: Pleistocene and The Ice Age.
Further Reading Bone N (1996) The Aurora: Sun Earth Interactions. New York: Wiley. Calowicz MJ and Lopez RE (2002) Storms from the Sun: The Emerging Science of Space Weather. Washington, DC: Joseph Henry Press.
Golub L and Pasachoff JM (2001) Nearest Star: The Sur prising Science of Our Sun. Cambridge, MA: Harvard University Press. Kaler JB (1992) Stars. Scientific American Library. New York: WH Freeman. Lang KR (1995) Sun, Earth and Sky. New York: Springer Verlag. Lang KR (2000) The Sun from Space. New York: Springer Verlag. Lang KR (2001) The Cambridge Encyclopedia of the Sun. New York: Cambridge University Press. Lang KR (2003) The Cambridge Guide to the Solar System. New York: Cambridge University Press. Odenwald S (2001) The 23rd Cycle: Learning to Live with a Stormy Sun. New York: Columbia University Press. Phillips KJH (1992) Guide to the Sun. New York: Cambridge University Press.
Asteroids, Comets and Space Dust P Moore, Selsey, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Asteroids and comets must be regarded as minor members of the Solar System. Compared with planets they are of very low mass, and they have even been referred to as cosmic debris. The asteroids, dwarf worlds most of which are well below 1000 km in diameter, are found mainly between the orbits of Mars and Jupiter, though some stray from this ‘main belt’; comets have been described as ‘dirty snowballs’, and though they may become very conspicuous in the sky they are very insubstantial. This article reviews the asteroids and comets, together with the large amount of thinly-spread material lying in the Solar System.
Some small asteroids can leave the main belt, and swing closer to the Sun; they may even approach the Earth, and collision cannot be ruled out (it may even be that the impact of an asteroid, some 65 million years ago in Mexico, caused a climatic change and mass extinction, which included the dinosaurs). All of these Near Earth Approach (NEA) asteroids are very small indeed. There are asteroids known as Trojans which share the orbits of major planets; others have very eccentric orbits which take them into the far reaches of the Solar System, and recently it has been found that there are asteroid-sized bodies near and
Distribution of the Asteroids The Solar System is divided into two well-defined parts. There are four relatively small, rocky planets: Mercury, Venus, the Earth, and Mars. Then come the four giants: Jupiter, Saturn, Uranus, and Neptune. Between the orbits of Mars and Jupiter thousands of asteroids, otherwise known as minor planets, make up what is known as the main belt (Figure 1). Of the main belt asteroids, only one (Ceres) is as much as 900 km in diameter, and only one (Vesta) is ever visible with the naked eye. Some of the larger main belt asteroids are listed in Table 1.
Figure 1 Distribution of asteroids.
SOLAR SYSTEM/Asteroids, Comets and Space Dust 221
Table 1 Some of the larger Main Belt ateroids
Asteroid
1 2 3 4 5 6 7 8 9 10 72 87 253 153 279 511 704
Ceres Pallas Juno Vesta Astrea Hebe Iris Flora Metis Hygeia Feronia Sylvia Mathilde Hilda Thule Davida Interamnia
q
Q
Period, years
Orbital eccentricity
Orbital inclination
T
Diameter, km (max)
M
2.55 2.12 1.98 2.15 2.08 1.94 1.84 1.86 2.10 2.76 1.99 3.19 1.94 3.10 4.22 2.61 2.61
2.77 2.77 2.67 2.37 2.57 2.43 2.39 2.20 2.39 3.13 2.67 3.48 3.35 3.97 4.27 3.18 3.06
4.60 4.62 4.36 3.63 4.13 3.77 5.51 3.27 3.69 5.54 3.41 6.50 5.61 7.91 8.23 5.66 5.36
0.078 0.234 0.258 0.090 0.190 0.202 0.229 0.156 0.121 0.120 0.120 0.083 0.262 0.142 0.011 0.177 0.148
10.60 34.80 13.00 7.14 5.36 14.79 5.51 5.89 5.59 3.84 5.42 10.87 6.70 7.83 8.23 15.93 17.30
C CU S V S S S S S C U P C P D C D
960 571 288 525 120 204 208 162 158 430 96 282 66 222 130 324 338
7.4 8.0 8.7 6.5 9.8 8.3 7.8 8.7 9.1 10.2 12.0 12.6 10.0 13.3 15.4 10.5 11.0
q perihelion distance, in astronomical units. Q aphelion distance, in astronomical units. M mean magnitude at opposition. T type (see Table 2).
well beyond the orbits of Neptune and Pluto. These make up what is known as the Kuiper Belt.
Discovery A mathematical relationship, known as Bode’s Law, led astronomers to believe that there should be another planet moving between the orbits of Mars and Jupiter. From 1800, a systematic search was carried out by a group of observers who called themselves the ‘Celestial Police’, and on 1 January 1801, the first asteroid, Ceres, was discovered by G Piazzi (who was not then a member of the group, though he joined later). Three more small bodies were found during the next few years: Pallas, Juno, and Vesta. It was not until 1845 that the next asteroid, Astræa, was found; others followed, and by now many thousands are known. When a new asteroid is discovered, it is given a temporary designation and then, when its orbit has been reliably worked out, a number. At first mythological names were used, but the supply of these names soon ran out; today the discoverer is entitled to suggest a name, which is then ratified by the International Astronomical Union.
Origin and Orbits It is no longer thought that the asteroids are fragments of a large planet which broke up. It seems that no planet of appreciable size could form in this part of the Solar System, because of the disruptive
influence of Jupiter. The asteroids in the main belt are not evenly distributed; Jupiter’s gravitational pull tends to produce groups or ‘families’, made up of numbers of asteroids moving at around the same distance from the Sun (Figure 2). A family is named after one of its most prominent members, and does seem to be due to the disruption of a larger body, The Flora family has at least 400 members. There are also gaps in the main belt (the Kirkwood Gaps) which are almost empty, because of regular gravitational interactions with Jupiter. For instance, there is a gap at a distance of 375 million km from the Sun, where an asteroid would complete three orbits for every one orbit of Jupiter (Figure 3).
Types of Asteroids Asteroids are divided into various types, according to their physical and surface characteristics. The main types are listed in Table 2 (Figure 4). There is certainly a link between comets and small asteroids; thus a tailed comet discovered in 1951 (WilsonHarrington) was lost for years before being recovered in 1979 in the guise of an asteroid. It was given a number (4015) and now shows no sign of cometary activity.
Asteroid Surfaces and Composition Details on some asteroids have been recorded. 3 Vesta has been imaged by the Hubble Space Telescope, and
222 SOLAR SYSTEM/Asteroids, Comets and Space Dust
Figure 2 Orbits of some asteroids.
Table 2 Types of asteroids C (Carbonaceous)
S (Silicaceous)
M (Metallic) E (Enstatite) D
A P V Figure 3 Sizes of same asteroids compared with British Isles Diameters: Ceres 970 km Vesta 288 230, Flora 204.
is geologically of great interest; there are two distinct hemispheres, covered with different types of solidified lava, and there is one huge impact crater. Some asteroids have been imaged from passing space-craft; 253 Mathilde (Figure 5) is very dark and irregular, and has been described as ‘a heap of rubble’, while 243 Ida is cratered and is accompanied by a tiny satellite, Dactyl. 216 Kleopatra has two lobes of similar size, and looks remarkably like a dog’s bone!
U
Most numerous, increasing in number from 10% at 2.2 a.u. up to 80% at 3 a.u. Low albedo; spectra resembles carbonaceous chondrites Most numerous in the inner part of the main zone. Generally reddish, spectra resemble those of chondrites Moderate albedoes; may be the metal rich cores of larger parent bodies Relatively rare, high albedos; enstatite (MgSiO3), is a major constituent Low albedo; reddish; surfaces are 90% clays, with magnetite and carbon rich substances Almost pure olivine Dark and reddish; not unlike Type B. Igneous rock surfaces, very rare; 4 Vesta is the only large example Asteroids which are regarded as unclassifiable
Asteroids closer-in than the Main Belt These are of various types. Details are given in Table 3. All are small, usually only a few kilometres across, and are irregular in shape. The first to be discovered (in 1898) was 433 Eros; it is an Amor asteroid, so that its orbit crosses that of Mars but not that of the Earth. It can approach Earth at a distance of 23 million km. On 12 February 2001,
SOLAR SYSTEM/Asteroids, Comets and Space Dust 223
Table 3 Asteroids closer in than the Main Belt Apohele type Aten type
Figure 4 511 Davida, a Main Belt asteroid 320 km in diameter. This sequence of images was taken at the WM Keck Observatory on 28 December 2002, The rotation period is just over one hour; here Davida is seen from above its north pole as it spins counter clockwise.
Apollo type Amor class
Orbit entirely within that of the Earth, only one example is known, the tiny 2003 CR20 Average distance from the Sun less than 1 a.u., though they may cross the Earth’s orbit. All very small Orbits cross that of the Earth; average distance from the Sun over 1 a.u. Orbits cross that of Mars, but not that of the Earth
no danger of collision as its orbital inclination is 61 . It is no more than 15 km in diameter. The ‘Centaur’ asteroids remain well beyond the Main Belt; the first to be found (in 1977) was 2060 Chiron, which moves mainly between the orbits of Saturn and Uranus, in a period of 50 years. It shows traces of a coma at times, but seems much too large to be classed as a comet, even though it has been given a cometary number.
The Kuiper Belt
Figure 5 Asteroid 253 Mathilde, imaged by NEAR space craft on 27 June 1997, from a range of 2400 km. There are large craters. The asteroid is very dark, with an average albedo of 4%. Mathil de’s diameter is 50 50 70 km, rotation period 418 hours.
the space-craft NEAR-Shoemaker made a controlled landing on it; Eros proved to be a very primitive body, (Figure 2) and very ancient. Craters were plentiful, as well as rocks and boulders of all kinds, and superficial ‘landslides’ in the surface material were recognized. Some small asteroids pass between the Earth and the Moon, and collision cannot be ruled out, and there are more potentially hazardous asteroids (PHAs) than used to be thought.
Many asteroidal bodies have been found near and beyond the orbits of Neptune and Pluto; the existence of such a belt was suggested by GP Kuiper (and earlier, less positively, by K. Edgeworth). Some are larger than any Main Belt asteroids; 50 000 Quaoar has a diameter of about 250 km, more than half that of Pluto. Other large Kuiper Belt objects are 28 978 Ixion (1200 km), 20 000 Varuna (900 km), and 38 093 Rhadamanthus (320 km). There are also asteroidsized bodies which recede to immense distances from the Sun, and have orbital periods of hundreds of years. There are excellent reasons for suggesting that Pluto should be regarded as merely an exceptionally large Kuiper Belt object rather than as a bonafide planet. The Kuiper belt also includes some comets.
Comets Asteroids Beyond the Main Belt The Trojan asteroids move in the same orbit as Jupiter, though they keep either well ahead of or well behind the Giant Planet and are in no danger of being engulfed. Mars has several Trojans, and Neptune one. No true Earth Trojans are known, though 3753 Cruithne has almost the same orbital period and describes a curious sort of ‘horseshoe’ path with respect to the Earth. There are also asteroids, such as 944 Hidalgo and 5335 Damocles, with very eccentric orbits, very like those of comets. For example, Damocles has a period of 40.9 years; its orbit crosses those of Mars, Jupiter, Saturn, and Uranus, but is in
Comets are the most erratic members of the Solar System. They were once regarded as unlucky, and descriptions of them go back for thousands of years. Certainly a brilliant comet may look really spectacular, but by planetary standards all comets are of very low mass. They are true members of the Solar System, but in general their orbits are very eccentric, and their movements are quite unlike those of the planets. Nature of Comets
The only fairly substantial part of a comet is the nucleus, made up of rocky fragments held together by frozen ices such as H2O methane, carbon dioxide,
224 SOLAR SYSTEM/Asteroids, Comets and Space Dust
and ammonia. When a comet is warmed as it approaches perihelion the rise in temperature leads to evaporation, so that the comet develops a head or coma, often with a tail or tails. Cometary tails always point away from the Sun, and are of two main types ion and dust tails. A gas or ion tail is due to particles being repelled by the solar wind, while with a dust tail the particles are driven out by the pressure of sunlight; this means that when a comet is moving outward, after perihelion, it travels tail-first. However, not all comets develop tails of any kind, and even a large comet will lose its tail when it has receded into the far part of the Solar System. Nomenclature
Traditionally, a comet is named after its discoverer or discoverers; thus the brilliant comet seen in 1995 and 1996 was known as Hale-Bopp, since it was found independently by two American observers, Alan Hale and Tom Bopp. Occasionally the comet is known by the name of the mathematician who first computed its orbit, as with Comets Halley and Encke. There is also an official numbering system which relates to the date of discovery. Orbits
Many comets have short periods – for example 3.3 years for Comet Encke. These short-period comets can be predicted, and some can be followed all round their orbits. Many have their aphelia near the distance of the orbit of Jupiter, making up what is termed Jupiter’s comet family. Most of them are faint, and few attain naked-eye visibility. The only reasonably bright comet with a period of less than 100 years is Halley’s (76 years), which last returned to perihelion in 1986–1987. Long-period comets recede to great distances, and since their periods amount to many centuries they
cannot be predicted, Hale-Bopp (Figure 6) will be back in about 2350 years, but for the next return of Comet Hyakutake, which was bright for a few weeks in 1996, we must wait for around 14 000 years. These orbits are almost parabolic, and indeed some comets are thrown into parabolic orbits after passing perihelion, so that they will never return. Arend-Rola`nd, the bright comet of April 1957, is one example of this. Origin of Comets
It seems that short-period comets come from the Kuiper Belt. In general, their orbits are not highly inclined to the ecliptic, though some, notably Halley’s Comet, have retrograde motion. Comets of much longer period are thought to come from the Oort Cloud, a huge spherical cloud of debris surrounding the Sun at a distance of over one light-year; its existence was suggested in 1950 by the Dutch astronomer JH Oort. It is, of course, unobservable from Earth. If an Oort Cloud comet is perturbed for any reason, it may swing in towards the Sun; it may then be perturbed into a short-period orbit, it may fall into the Sun and be destroyed, or it may simply return to the Oort Cloud. The orbital inclinations may be very high, and many long-period comets have retrograde motion. It may be that the Oort Cloud comets were formed closer to the Sun than the Kuiper Belt objects. Lowmass objects formed near the giant planets would have been ejected by gravitational encounters. While Kuiper Belt objects, formed further out, were not so affected. Details of some notable comets are given in Table 4 (see Solar System: Meteorites).
Comets and Meteors As a comet moves, it leaves a ‘dusty trail’, and if the Earth passes through such a trail we see a meteor shower. In many cases the parent comets are identifiable, for example the Orionid meteors, seen every October, come from Halley’s Comet, while the August Perseids come from Comet Swift–Tuttle. Some comets have been seen to disintegrate; thus Biela’s Comet, which had a period of 6.6 years, broke in two during its return in 1846, and has not been seen since 1852, though for many years meteors appeared from the position where the comet ought to have been. Other periodical comets have been lost because their orbits have been so violently perturbed by planetary encounters. One comet, ShoemakerLevy 9, in captured orbit around Jupiter, was destroyed in 1994 when it impacted Jupiter. Halley’s Comet
Figure 6 Comet Hale Bopp, 1997. Note the straight ion tail, and the curved dust tail. This was the most spectacular comet for many years.
Named for Edmond Halley, who observed it in 1682 and was the first to realize that it was periodical
SOLAR SYSTEM/Asteroids, Comets and Space Dust 225
Table 4 Some Notable Comets Periodical Comets Comet
2 26 10 46 9 7 6 21 19 15 4 36 8 27 13 1 109 153
Encke Grigg Skjellerup Tempel 2 Wirtanen Tempel 1 Pons Winnecke D’Arrest Giacobini Zinner Borrelly Finlay Faye Whipple Tuttle Crommelin Olbers Halley Swift Tuttle Ikeya Zhang
P
q
Q
E
I
M
3.28 5.10 5.47 5.46 5.51 6.37 6.51 6.61 6.80 6.95 7.34 8.53 13.51 27.4 69.6 76.0 135.0 341
0.33 0.99 1.48 1.07 1.50 1.26 1.35 1.03 1.37 1.09 1.59 3.09 0.997 0.74 1.18 0.59 0.96 0.51
2.21 2.96 3.10 3.10 3.12 3.44 3.49 3.52 3.59 3.64 3.78 4.17 5.67 17.4 32.6 35.3 51.7 60
0.850 0.664 0.552 0.657 0.502 0.634 0.614 0.706 0.623 0.699 0.578 0.239 0.824 0.919 0.930 0.967 0.964 0.99
11.9 6.6 12.0 11.7 10.5 22.3 19.5 31.9 30.2 3.7 9.1 9.9 54.7 19.1 44.6 162.2 113.4 28.1
11 12 10 16 9 14 6 10 13 13 8 9 8 11 5 4 4 5
q perihelion distance, astronomical units. Q aphelion distance, astronomical units. E orbital eccentricity. I orbital inclination, degrees. M absolute magnitude (the magnitude which the comet would have if seen from a distance of 1 a.u. from the Sun and 1 a.u. from the Earth.) P period, years.
(Figure 7). It was probably record by the Chinese as early as 1059 bc, and since 240 bc it has been seen at every return; it came to perihelion in 1066, and is shown in the famous Bayeux Tapestry. During the 1986 return several space–craft were sent to it, and one of these, Giotto, passed within 605 km of the nucleus. The nucleus was shaped rather like a peanut, and measured 15 8 8 km. Over 60 000 million comets of this mass would be needed to equal the Earth. The nucleus was dark-coated, and made up largely of water ice; dust-jets were active, though only from a small area on the sunward wide (Figure 8). The comet is now too faint to be detected, though it should be recovered before the next perihelion passage, due in 2061. Great Comets
Really brilliant comets were seen fairly frequently during the nineteenth century, but were less common in the twentieth century (Figure 9). The brightest comet of near-modern times was probably that of 1843, which cast shadows and was visible in broad daylight. The last two really spectacular comets were those of 1910 – the Daylight Comet, seen shortly before Halley’s – and 1965 (Ikeya–Seki), which faded quickly. Its period has been given as 880 years. Some Great comets are listed in Table 5.
Figure 7 Halley’s Comet, March 1986, (Photo by Tom Polaks with a 100 mm lens at f/2.8.) The faint globular cluster M75 is also shown. From the most left of the three conspicuous stars left and above Halley’s head, go to the fainter star above and left. This star forms a fainter, nearly rectangular triangle with the other stars above and left of it. On the line connecting with the far left edge a star like spot anneals; this is M75.
Comet Hale-Bopp was not so brilliant as these, but was exceptionally beautiful, and was visible with the naked eye for over a year, from July 1996 to October 1997. It was enormous by cometary standards, with a 40 km nucleus, but unfortunately it did not come close to the Earth. There were both ion and dust tails, plus a third inconspicuous tail made up of
226 SOLAR SYSTEM/Asteroids, Comets and Space Dust
sodium. It was last at perihelion about 4200 years ago, but planetary perturbations mean that it should return in about 2350 years, though of course all periods of this kind of length cannot be given accurately. Its orbital inclination is 89 , so that its path lies at almost a right angle to that of the Earth. During its period of visibility there were marked changes in the tails, and a spiral structure developed in the coma. Comet Ikeya-Zhang of 2002 was much less striking – it became no brighter than the fourth magnitude – but is notable because it was found to be a return of the
comet of 1661, and is therefore the longest-period comet to have been seen at more than one apparition. It will be back once more in 2343.
Figure 8 Head of Halley’s Comet, imaged from the Giotto space craft. The dark coating and the active dust jets are well seen. (Photograph from the HMC [Halley Multi colour Camera]), Giotto passed 605 km from the nucleus on the night of, 13 14 March 1986.
Figure 9 Comet Hyakitake, C/1996 B2. This beautiful comet was conspicuous object briefly in April May 1996; it was obvi ously greenish, and had a long tail. It was in fact a small comet, but made a fairly close approach to the Earth. It will next come to perihelion in 14 000 years! time, look out for it then.
Life in Comets?
The ‘panspermia’ theory was due to the Swedish scientist Svants Arrhenius, whose work won him the Nobel Prize for Chemistry in 1903. Arrhenius believed that life on Earth was brought here in a meteorite, but the theory never became popular, because it seemed to raise more problems that it solved. The same sort of theme has been followed up much more recently by Sir Fred Hoyle and C Wickramasinghe,
Table 5 Some Great Comets Year
Name
1744 1811 1843 1858 1882 1910 1927 1947 1965 1976 1996 1997
de Che´seaux Flaugergues Great Comet Donati Great Comet Daylight Comet Skjellerup Maristany Southern Comet Ikeya Seki West Hyakutake Hale Bopp
Multi tailed comet; max. magnitude 7 Mag. 0; 24 degree ion tail. Period 3096 years Mag. 6. Sun grazer. Period 517 years Mag, 1. Most beautiful of all comets, with ion and dust tails. Period 1951 years Reached mag. 4. Period 760 years Magnitude 4. Immensely long period Magnitude 6; 35 degree tail Magnitude 5, 25 degree tail Magnitude 10; seen very near the Sun Magnitude 2. Multiple dust tail Briefly reaches magnitude Q. Very long tail. The green comet Magnitude 0.5; naked eye object for over a year
SOLAR SYSTEM/Asteroids, Comets and Space Dust 227
Another glow due to cosmic dust is the Gegenschein, seen as a faint patch exactly opposite to the Sun in the sky. It is extremely elusive, and is visible only under near-ideal conditions. The best opportunities occur when the anti-Sun position is well away from the Milky Way, from February to April and from September to November. Generally it is oval in shape, measuring about 10 by 22 , so that its maximum diameter is 40 times that of the full moon. The Zodiacal Band is a very dim, parallel-sided band of radiance which may extend to either side of the Gegenschein, or prolonged from the apex of the Zodiacal Light Cone to join the Zodiacal Light with the Gegenschein. It also is due to sunlight being reflected from interplanetary particles near the main plane of the Solar System.
See Also Solar System: The Sun; Meteorites; Mars; Jupiter, Saturn and Their Moons; Neptune, Pluto and Uranus.
Further Reading Figure 10 The Zodiacal Light. A typical display, photographed on 19 November 1998 over the Qinghai Radio Observatory near Delinghom Qinghai, Central China. (M Langbroek).
who claimed that comets can actually deposit harmful bacteria in the Earth’s upper atmosphere, causing epidemics. Again there has been little support.
Space Dust There is a great quantity of ‘dust’ in the Solar System, particularly near the main plane. It is the cause of the Zodiacal Light, (Figure 10) which may be seen as a cone of light extending upwards from the horizon for a fairly brief period either after sunset or before sunrise. Since it extends along the ecliptic, it is best seen when the ecliptic is nearly vertical to the horizon, in February to March and again in September–October. Cometary debris is a major contributary factor. It was first correctly explained by the Italian astronomer, GS Cassini, in 1683.
Bone N (1986) Meteors. London: Philip. Bhandt G and Chapman D (1982) Introduction to Comets. Cambridge: Cambridge University Press. Burnfam R (2000) Great Comets. Cambridge: Cambridge University Press. Krishna S (1997) Physics of Comets. Singapore: World Scientific. Kronk G (1988) Comet Catalogue. Enslow: Hillside NJ and Aldershot. Kronk G (1988) Meteor Showers. Enslow: Hillside NJ and Aldershot. Moore P (2001) Astronomy Data Book. London: Institute of Physics, Publishing. Moore P (2003) Atlas of the Universe. London: Philip. Norton CR (1992) Rooks from Space. Montana: USA Mountain Press Publishing. Schmadel L (2002) Dictionary of Minor Planet Names. Berlin, Heidelberg, New York: Springer verlag. Kowal CT (1996) Asteroids. Wiley. Whipple FL (1985) The Mystery of Comets. Cambridge: Cambridge University Press. Yeomans K (1991) Comets. New York: Wiley Science Editions.
228 SOLAR SYSTEM/Meteorites
Meteorites G J H McCall, Cirencester, Gloucester, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Meteorites are bodies of metal or stony material mixed with metal which fall to Earth in sporadic and random arrival events, characterized by entry of a fireball or bolide streaking, often with punctuated explosive bursts, through the sky on their frictional passage through the Earth’s atmosphere (Figure 1). The history of the gradual scientific acceptance of the reality of such events is followed by a brief description of the classification of various types of meteorite; the four age and time interval measurements significant for any meteorite; and the known or likely provenance in the bodies of the Solar System of the various types are then considered. After a brief mention of impact cratering and tektites, and also ‘fossil’ meteorites enclosed in ancient rocks, an account is given of
Figure 1 A painting by P.V. Medvedev of the fireball accom panying the Sikhot Alin fall of 1949 (reproduced from McCall 1973).
the revolution in ‘Meteoritics’ (essentially an extension of geology, geochemistry, metallurgy, and physics into the realms of astronomy and planetology) during the latter half of the twentieth century. This is a result of space exploration and the recognition of hitherto unknown optimum collection regions (icebound Antarctica; the Nullarbor Plain, Australia; and other desert regions). Examples of some extensions of research into meteoritics in modern state-of-the-art science are listed.
Historical: the Fall of Stones and Metal from the Sky Records of shooting stars, bright objects seen to dart across the night sky, go back to Egyptian papyrus writings of ca. 2000 bc and records of actual meteorites falling to Earth out of the sky go back almost as far – the fall of a black stone in the form of a cone, circular below and ending in an apex above, was reported in Phrygia about 652 bc, the familiar image of a stony meteorite such as the Middlesborough Meteorite (Figure 2) coming to us from the distant past. The Parian chronicle records falls of stones in Crete in 1478 bc and in 1168 bc of iron. In 618 bc, a fall of stones is reported to have broken
Figure 2 The Middlesborough, England, stone (fell 1881) show ing the dark fusion crust and anterior surface in flight, the apex of the cone being in the direction of flight and the radiating flutings being produced by atmospheric ablation (from McCall 1973).
SOLAR SYSTEM/Meteorites 229
several chariots and killed ten men, a unique fatality. The sacred stone built into the Kaaba at Mecca is reported to have been long known prior to Islam and to have fallen from the sky. Such falls were given a religious significance, and officers of the Geological Survey in India had to go hot-foot to the site of a fall or the mass was either enshrined or broken into pieces to release evil spirits. American Indians confused later scientists by transporting masses long distances and burying them in cysts. Particularly pleasing is the custom in mediaeval France of chaining meteorites up to prevent them departing as swiftly as they arrived or from wandering at night. The earliest material from a fall preserved in western Europe is believed to be at Ensisheim, Alsace, stored in the local church since it fell in ad 1462. Despite all these early records (and there are many more, in particular from Russia and China), scientists were slow to accept the process of rocky or metal material falling from the sky. Though there are records of the finds of irons and the falls of stones much earlier and the problem had been solved – Diogenes of Apollonia wrote ‘‘meteors are invisible stars that fall to Earth and die out, like the fiery, stony star that fall to Earth near the Egos Potamos River (in 465 bc): and natives in northern Argentina had led the conquistadors to buried masses of exotic iron, of supposed celestial origin in 1576 – scientific acceptance was widely achieved only in the last years of the eighteenth century Age of Enlightenment and the earliest years of the nineteenth century, with natives leading the explorer Pallas in Siberia to a buried stony-iron mass reputedly fallen from the sky; also falls were followed by material recovery at Wold Cottage, near Scarborough, Yorkshire and L/Aigle France. The fall at Albareto, Italy, in 1766, had been well described by the Abbe´ Dominico Troili, but dismissed as the product of a subterranean explosion which hurled it high in the sky from a vent in the Earth. The stone which fell at Luce´ , France in 1768, the first to be chemically analysed, was dismissed as neither from thunder, nor fallen from the sky, but as a piece of pyritiferous sandstone by a panel of august scientists! So it was the Pallas stony-iron meteorite (700 kg, ‘Krasnojarsk’), the subject, together with the Otumpa iron from South America, of a book published by E.F.F. Chladni in Riga in 1794, which really established the scientific reality of meteorite falls. Both were exotic, being found far from any known volcanic province, and by a process of elimination, he reached a single possible answer and further connected them with the phenomenon of fireball meteors. Russian
scientific circles were distant from western Europe, and the English were really convinced only by the fall of a stone at Wold Cottage near Scarborough in 1795. This came into the possession of Joseph Banks, who recognized the similarity of the black fusion crust to the Siena fall material of 1794 in his possession. Edward Howard studied both and the presentation of his findings to the Royal Society in 1802–1803 convinced sceptics in England. Presentation to the Institut de France convinced several important scientists, but resistance to the idea was not finally overcome in that country until 3000 stones showered down on L’Aigle, Normandy and were described by Biot. Chladni’s work then received belated international acknowledgement, but decades would elapse before the connection with fireballs was completely established and a century before the origin of most of them through impacts between asteroids would be established.
Classification The classification of meteorites has developed over the years and some new types and revisions of the system have inevitably arisen in the last half of the twentieth century with the prolific collection from optimum Antarctic and desert regions; despite this, the system remains workable though some revision might in time be necessary. There are three principal classes: Irons (Figures 3, 4) Stony-irons (Figure 5) Stones (Figure 2) The latter are subdivided into (i) Chondrites, which display rounded bodies (chondrules) (Figures 6, 7),
Figure 3 The Haig, Western Australia, iron (find 1951, 480 kg, III AB) with typical hackly markings on the surface.
230 SOLAR SYSTEM/Meteorites
Figure 6 The Cocklebiddy, Western Australia, ordinary chon drite (fall 1949, 0.794 kg), cut face showing specks of light grey nickel iron disseminated in a dark grey silicate matrix: the rounded chondrules are microscopic and thus not visible (from McCall 1973).
Figure 4 Cut and etched surface of the Mount Edith iron, West ern Australia (find 1913, 160 kg, III AB) showing the Widmanstat ten pattern and dark troilite (sulphide) nodules.
Figure 7 View in a microscope thin section across a chondrule (2 mm diameter) showing elongated olivine crystals and dark glass, within the rounded chondrule, which is set in an aggregate of olivine, pyroxene, and feldspar grains, opaque nickel iron, sulphide and products of secondary weathering (Mulga South ordinary chondrite, Western Australia (from McCall 1973)). Figure 5 Cut surface of the Brenham, Kansas, pallasite stony iron (find 1962, 22 and 9 kg), showing nickel iron (light grey) and olivine (dark) (from McCall 1973).
believed to be relics of a very early accretionary stage in the formation of the asteroidal parent bodies (the chondrules may be wholly obliterated by recrystallization); and (ii) Achondrites, without chondrules, having textures resembling those of terrestrial igneous rocks (Figure 12). The classification used worldwide, as at 2003, is shown in Table 1a and 1b and the statistics of meteorite falls and finds in Table 2.
Meteorites within Meteorites
Many meteorites are brecciated, probably mainly due to shock processes through collision with other meteorites in space, but some also carry other meteorite types as fragments within them. Chondrites may occur as fragments within dissimilar host chondrites. Even more spectacular are shocked eucrite achondrite bodies within the Mount Padbury stony iron (mesosiderite) and enstatite and carbonaceous and ordinary chondrite bodies within the Bencubbin stony iron meteorite, both found in Western Australia.
SOLAR SYSTEM/Meteorites 231 Table 1b Differentiated meteorites
Table 1a Undifferentiated meteorites Class
Symbol
Example
Class
Symbol
Example
Carbonaceous chondrites
CI CM CO CV CK CR CH R K LL L H EL EH
Orgueil Murchison Ornans Allende Karounda Renazzo ALH 85085 Rumuruti Kakangari Saint Mesmin L’Aigle Wiluna Eagle Saint Sauveur
Irons
I AB
Campo del Cielo
Stony irons
IC II AB II C II D II E II F III AB III CD III E III F IV A IV B Mesosiderites
Rumurutiites Kakangari type chondrites Ordinary chondrites
Enstatite chondrites
Carbonaceous chondrites: characterized by sparse to abundant
chondrules set in a dark, friable matrix of carbon rich compounds, phyllosilicates, mafic silicates, Ni Fe metal, and glass. The letter symbols separate groups based on different mineralogy, relative abundance of different lithophile and siderophile elements, relative abundance and size of chondrules, and oxygen isotope signatures. Numerical suffixes 3, 2, and 1 denote progressive aqueous alteration and 3, 4, 5, and 6 progressive thermal alteration. Rumurutiites: a new rare group of chondrites. Kakangari type chondrites: a small group of chondrites now separately defined. Ordinary chondrites: chondrules are embedded in a finely crystalline matrix of mafic minerals, pyroxene, and olivine, together with NI Fe metal and glass. Some are recystallised thermally and lose the definition of chondrules and the glass. The H, L, and LL groups differ in the magnesian/iron ratio in the ferromagnesian silicate minerals. The number suffixes 3 7 denote degree of thermal alteration (loss of original texture and recystallization). Enstatite chondrites: these are chondrites with the Mg rich pyroxene enstatite. The EL and EH groups have different relative abundances of silicates and metals. The numerical suffixes above (3 6) may be applied.
Age There are four periods of time that are significant in the history of any meteorite:
Terrestrial age: the time spent on Earth since fall. Obviously, the material from an observed fall has an immediately known terrestrial age. Cosmic-ray induced isotopes are used to obtain this age from such finds. We know from observed fall meteorites how much of these isotopes are in a meteorite when it arrives. A meteorite found later will have less isotopes because the Earth’s atmosphere protected it after arrival, and unstable products of cosmic radiation, such as 14C will decay, so that the difference between the normal content on arrival and that
Stones (Achondrites)
Pallasites Eucrites Diogenites Howardites Angrites
Primitive achondrites
Ureilites Aubrites SNC Meteorites (Mars sourced?) Shergottites Naklites Chassignite (Orthopyroxenite) Basaltic and anorthositic achondrites (Lunar sourced) Brachinites Winonaites
Sikhot Alin
Cape York
Gibeon Mount Padbury Krasnojarsk Camel Donga Johnstown Kapoeta Angra dos Rios Novo Urei Aubres
Shergotty Nakhla Chassigny ALH 84001 ALH 85085
Brachina Winona
The primitive achondrites have igneous textures with no chondrules, but their mineralogy and bulk chemistry shows little difference from ordinary chondrites. They are supposed to have undergone igneous processes but with no fractional crystallization, but partial melting and segregation of the phases to varying degrees. The irons were formerly separated into octahedrites (kamacite plus taenite; on etching yield criss crossing Windmanstatten patterns) (Figure 4): hexahedrites (mostly kamacite, yield only narrow thin Neumann lines on etching) and ataxites (no etch pattern). The Symbol classification above which replaced this metallurgical classification is still being modified and I AB and III CD have recently been grouped as I AB III CD. These symbols reflect the differences in chemistry (nickel, gold, iridium content, etc.). The eucrites have basaltic textures. Many meteorites defy classification and are listed as unclassified. For example, the Bencubbin (find, Australia) meteorite appears to be a stony iron but is in fact a mixture of four types, an iron, an enstatite achondrite, and two chondrites, one carbonaceous. It would seem to be the result of more than one collision, the first mixing occurring very early in its history (ca. 4500 Ma) and causing heating and melting.
232 SOLAR SYSTEM/Meteorites
Figure 8 The distribution of the Mundrabilla irons on the Nullarbor Plain, Western Australia (rediscovered 1964 onwards) showing the typical dispersion ellipse. Below left: the M1 mass (est. 11 tonnes), as found, showing the space capsule shape with striations on anterior surface in atmosphere descent: also the curved face where the M2 mass separated. Below right, the M2 mass (est. 5 tonnes), showing the 10 cm pad of iron shale below, the product of a million years weathering by surface agents since fall (from McCall 1999, reproduced with permission of Palgrave Macmillan).
SOLAR SYSTEM/Meteorites 233
Table 2 The total of known meteorites up to the end of 1999 Class
Fall
Find
Total
Stones Stony irons Irons Unknown Total
940 12 48 5 1005
20574 104 817 7 21502
21514 116 865 12 22507
(After MM Grady (2002)).
Cosmic ray exposure age: the time spent as a metrescale meteoroid orbiting the Sun. Cosmic rays react with some atoms in iron or stony meteoroids and the quantity of gases formed depends on the chemical nature of the meteoroid and the duration of exposure to cosmic rays in space. The most usual measurements are of the quantity of neon gas resulting from this cosmic ray exposure. The evidence suggests that few stony meteorites survive in space without further collisional destruction and pulverisation for more than 40 million years, but iron meteorites are more robust, surviving up to 1000 million years. Formation age: the age between the last high temperature episode in the parent body and the present. In the case of basaltic achondrites, this represents the time of crystallization from the liquid in a magma: chondrites, which have slightly greater formation ages, did not melt but were hot and recrystallised as solids soon after formation. The method involves the normal radioactive ‘clocks’ used by geologists, such as uranium-lead, the amount of lead produced by radioactive decay being an indicator of formation age. Values for chondrites are near to 4550 million years; some parent bodies were then heated and melted with fractional crystallization during the next 100 million years.
Figure 9 (A) Terrestrial age distribution for meteorites from the Allan Hills main icefield, Antarctica. (B) Terrestrial age distribu tion for 280 Antarctic meteorites sorted by stranding site. (A) from AJT Jull, S Cloudt, and E Cielaszyk; and (B) from ME Zolensky, in Grady et al. (1998). Published with the permission of the Geological Society Publishing House, Bath.
measured after the find can be used to determine the terrestrial age. As meteorites decay through natural weathering processes, these ages are usually values of tens of thousands of years, but in arid regions such as the Nullarbor Plain they are likely to be more, even a million years in the case of the large Mundrabilla iron (Figure 8); and in the Antarctic the ages taper off about 300 000 years though a very few have ages of one to three million years (Figure 9).
Formation interval: the time of the formation of the elements in stars (where almost all the elements except H and He were formed) and their incorporation in the parent body. This is done by measurement of the decay products of plutonium, an element which, because of its short half life, does not occur naturally. Plutonium was formed in a star about 150 million years before the formation of the asteroidal parent bodies of meteorites, but other elements were formed at different times.
Provenance Asteroidal
Meteorites are nowadays accepted as fragments of strays from the asteroid belt between Mars and Jupiter. Prior to the mechanism being established of producing (due to collisions) eccentric elongated orbits for asteroids – replacing their quasi-circular orbits beyond Mars – the nucleii of comets, impoverished in volatiles by repeated passage round the Sun, were long favoured as their source, but petrological and mineralogical evidence is against this. The Farmington fall in Kansas in 1890 seems to have heralded the firm establishment of asteroidal source. Sixty reports of visual observations of this fireball, at 12.50 pm on
234 SOLAR SYSTEM/Meteorites
a midsummer day and reportedly rivalling the Sun, were selected by scientists who deduced an orbit indicating that the parent asteroid was 1862 Apollo, Hermes, or 1865 Cerberus. Direct observation of fireballs by astronomers of the Sikhot-Alin, Siberia, 1949 and Pribram, Czechoslovakia, 1959 fireballs again strongly supported asteroidal sources and there have been many further supporting observations since (Figure 10). In recent years there have been numerous attempts to use optical and spectrographic methods to equate the reflectance and chemistry of asteroids with different classes of meteorites, but results seem to be inconclusive, possibly because of the operation of little understood space-weathering processes which affect the regolith surface of asteroids. Even a direct exploration mission to Eros in 2000–2001 (Figure 11) yielded no correlation and it must be borne in mind that there must be asteroids
Figure 10 Orbits crossing that of the Earth derived photo graphically from the falls of the Pribram (Czechsoslovakia), Innis free (Canada), and Lost City (USA) meteorites. (New figure, after Hutchison and Graham (1992).)
Figure 11 Asteroid 433 Eros (NEAR Shoemaker multispec tral NASA image). The large crater, Psyche, has a diameter of 5.3 km.
of classes never sampled by meteorites falling on the Earth. Several thousand asteroids are now known and it is estimated that there may be as many as 10 000 out there. Even in these small parent bodies, though some did not reach 100 C, others heated to more than 1200 C, the temperature needed to form a basaltic-textured eucrite. The heat sources in these small bodies are not known for certain, but a source in extreme early heating of the Sun or internal short-lived radioactive isotopes such as 26Al is favoured. Martian Achondrites?
Some meteorites apparently do not originate from asteroids. The ‘SNC’ group of achondrites (Shergottites, Nakhlites, Chassignite) (Figures 12 and 13)
Figure 12 The Nakhla achondrite (fell 1911, Egypt, one of 40 stones, totalling 40 kg); one of the SNC (?Mars sourced) meteor ites (from McCall 1999, reproduced with permission of Palgrave Macmillan).
Figure 13 Thin section view of the microtexture of the Nakhla meteorite, a typical achondritic texture resembling that of terres trial igneous rocks, formed by diopside pyroxene, olivine, and a few plagioclase crystals (10) (from McCall 1999, reproduced with permission of Palgrave Macmillan).
SOLAR SYSTEM/Meteorites 235
were first thought to come from Mars because of the presence of oxidised iron and hydrated minerals. Later in the twentieth century, entrapped gases in these meteorites were found to be similar to the Martian atmosphere sampled by Viking missions. The ages of formation of these meteorites (see below) are not those of the asteroidal meteorites (ca. 4550 Ma), but fall into two groups – Nakhlites 180 and Shergottites 1300 Ma (equivalent to Earth’s Jurassic and midProterozoic). The widely accepted source of these meteorites is Mars – the source must surely be a planet, and the mechanism the spalling off the surface by large impacts (there are theoretical objections to volcanic ejection). However there are problems: the trapped atmosphere should be the planet’s atmosphere 180 and 1300 Ma ago, not the present atmosphere, and atmosphere’s change with time: also, why are the 26 SNC meteorites recovered to date all a limited range of familiar igneous rocks – Mars is a very diverse surfaced planet? A hypothetical geological history of Mars has been built up by scientists on the basis of these 26 meteorites, an edifice which direct exploration may surely demolish? The joker in the pack is the famous ALH 84001 from Antarctica, a unique orthopyroxenite, which has a formation age similar to the asteroidal meteorites and contains the famous putative microfossils, the evidence about which seems now to favour inorganic rather than organic origin. Lunar Achondrites
Lunar achondrite meteorites (Figure 14) so completely match lunar surface rock samples obtained by Apollo and Luna missions that there is no doubt as to their provenance. First found in Antarctica, they have been later recognized in an existing collection from Western Australia and also new finds in the Libyan desert. Volcanic ejection can be ruled out; isotopic evidence suggests that all were spalled off by geologically quite recent and relatively minor impacts on the surface of the Moon, but here there is a glaring unresolved problem. There is widespread scientific acceptance of a major impact bombardment of the Moon 3.9 Ma ago, forming innumerable and immense craters: this must have hurled vast volumes of rock out into space, sampling deep below the regolith and surficial breccia (which is all that has yet been directly sampled), there is no trace of this material in the varied log of meteorites. Where has it gone?
Cratering and Tektites Meteorites normally land with little effect on the ground – even the 11 tonne Mundrabilla iron left no
Figure 14 Lunar sourced achondrite meteorite, ALH 81005 from Antarctica, discovered in 1981, after the first such discovery in 1979 by Japanese scientists in the Yamato Mountains. The structure is that of the lunar regolith breccias and a large white fragment of highlands anorthosite is visible. The cube has sides of 1 cm length (from McCall 1999, reproduced with permission of Palgrave Macmillan).
dent in the limestone surface – but multiple showers may produce small, simple craters (the 1947 SikhotAlin shower produced 106 associated with nickel-iron fragments). Larger masses have, in the quite recent geological past, produced kilometre-scale simple craters associated with nickel-iron (e.g., Canyon Diablo, Arizona; Wolfe Creek, Western Australia) and about 170 larger simple craters and more complex ring structures in the geological record are attributed to impact explosion processes involving larger masses, even asteroids. The largest, at 180 km diameter (Chicxulub, Yucatan, Mexico) has been associated with the Cretaceous-Tertiary boundary extinction of life (see Impact Structures). Only geochemical traces of the impactor have been discovered at such sites. Tektite showers were associated with a very small minority of such structures, but tektites are not meteorites, but are glassy objects melted from the impacted surface rocks, and spread over strewn fields at long distances from the impact sites (see Tektites).
Fossil Meteorites The only recorded case of a meteorite being recorded in ancient rocks relates to limestone strata at a quarry near Goteborg, Sweden, where there are 12 horizons crowded with ordinary chondrite meteorites, which must have been derived from rains of stones 480 Ma ago, in the Ordovician, the stones falling onto the limey mud bottom of shallow sea. Meteorites do not fall repeatedly at the same place because of the Earth’s rotation and this repetition is astonishing, as it implies repeated globally spread rains of meteorites over a period of about 1.75 million years.
236 SOLAR SYSTEM/Meteorites
The Twentieth-Century Revolution in Meteoritics Up to the orbital flight of Sputnik heralding the space age in 1957, the study of meteorites was a quiet museum occupation. The scientific interest in meteorites then exploded because of what they might tell us about planets, satellites, and asteroids. By coincidence, the year of Apollo XI, 1969, saw a Japanese party find nine meteorites on an area of bare ice in the Yamato Mountains. Antarctica. This 5 10 km area subsequently yielded 1000 meteorites. Blue ice areas and moraines in Antarctica have now yielded approximately 30 000 specimens representing some 20 000 falls. Two principle factors produce the optimum conditions for recovery: weathering is virtually nil in the arid climatic conditions and low temperatures prevailing, and the ‘conveyor belt’ situation on the ice sheet, snow falling and being buried and compacted to ice together with any meteorites on the surface, the snow moves coastwards and where nunataks (rocky peaks) obstruct its passage, the entrained and buried meteorites are excavated by wind action which removes the ice above (Figures 15 and 16).
By coincidence again, in the 1960s, rabbit trappers kept bringing in meteorite finds strewing the limestone surface of the arid Nullarbor Plain in Western Australia, and the writer of this entry, then working at the Western Australian Museum, wrote prophetically, ‘‘that the Nullarbor Plain must be littered with meteorites of all types’’. This was indeed so and systematic collection has so far yielded about 300 individual meteorites including two shower groups of more than 500 meteorite masses. Other desert areas of the world were then searched and Libya, Algeria, Morocco, and Oman have yielded several hundred finds, while desert areas in Roosevelt County, New Mexico have also proved productive. Neither Antarctica nor the desert areas are ‘worked out’ and many more finds will undoubtedly be made in the next years of this century. There are some desert areas in Asia, including the Gobi, that are not even searched so far, but a reconnaissance in the Gobi proved disappointing.
State-of-the-Art Research Meteoritics is a major area of scientific research nowadays and as many as 500 scientists may attend the yearly meetings of the Meteoritical Society. Research topics are extremely varied and besides such related topics as impact processes; tektites; planetary, lunar, satellite, cometary, and asteroid exploration, topics bearing directly on meteorites may include: . Ca-Al rich inclusions (CAIs) in meteorites, believed to be survivals from the accretion of the Solar System . Isotope fractionation in pre-solar graphite in carbonaceous chondrites . Isotope studies of chondrules and CAIs . Modelling conditions for the launch-window of ?Martian meteorites . Aqueous alteration of carbonaceous chondrites . Presolar nano-diamonds in meteorites . Xenon isotopes in nano-diamonds . Trapped gases in ordinary chondrites . Trace elements trapped in lunar meteorites
Figure 15 A meteorite as found on blue ice, its position flagged, Antarctica (from McCall 1999, reproduced with permis sion of Palgrave Macmillan).
This random sample illustrates the diversity of research: the revolution in meteoritics described above has produced enough subject material to keep science busy for many decades, if not centuries, and more keeps coming in. The important point to remember that meteorites come in free of charge – they have been called the poor man’s ‘space probe’. Even the cost of searching after major bolide events, searching Antarctica and systematic searching of the
SOLAR SYSTEM/Meteorites 237
Figure 16 Diagram showing how ice, moving very slowly towards the coastal ice front, is arrested by a rock nunatak, is stripped by wind action while stationary, to reveal entrained meteorites (new figure, after Hutchison and Graham 1992).
Nullarbor are infinitesimal when compared with the costs involved in direct space exploration.
See Also Impact Structures. Solar System: Asteroids, Comets and Space Dust. Tektites.
Further Reading Bevan AWR and Deha`eter JR (2002) Meteorites: A Journey Through Space and Time. Washington DC: Smithschian Institution Press. Grady MM (2002) Catalogue of Meteorites, 5th edn. London: Natural History Museum. Grady MM, Hutchison R, McCall GJH, and Rothery DA (1998) Meteorites: Flux With Time and Impact Effects, Special Publication No. 140. Bath: Geological Society Publishing House. Hey MH (1966) Catalogue of Meteorites, 3rd edn. London: British Museum (Natural History). Hutchison R and Graham A (1992) Meteorites. London: Natural History Museum.
Krinov EL (1960) Principles of Meteoritics. Oxford, London, New York, Paris: Pergamon Press. Mason B (1962) Meteorites. New York, London: John Wiley & Sons. McCall GJH (1973) Meteorites and their Origin. Newton Abbot: David and Charles. McCall GJH (1999) The Mundrabilla iron meteorite from the Nullarbor Plain, Western Australia: an update. In: Moore P (ed.) 2000 Yearbook of Astronomy, pp. 156 166. London: Macmillan. McCall GJH (1999) Meteoritics at the millennium. In: Moore P (ed.) 2000 Yearbook of Astronomy, pp. 153 177. London: Macmillan. McCall GJH (2001) Tektites in the Geological Record. Bath: Geological Society Publishing House. McCall GJH and de Laeter JR (1965) Catalogue of Western Australian Meteorite Collections, Special Publication No. 3. Perth: Western Australian Museum. Norton OR (2002) The Cambridge Encyclopedia of Meteorites. Cambridge: Cambridge University Press. Olson RJ and Pasachoff JM (1998) Fire in the Sky. Cambridge: Cambridge University Press. Zanda B and Rotaru M (2001) Meteorites. Cambridge: Cambridge University Press.
238 SOLAR SYSTEM/Mercury
Mercury G J H McCall, Cirencester, Gloucester, UK ß 2005, Elsevier Ltd. All Rights Reserved.
Historical Mercury, the closest planet to the Sun, was something of a mystery to ancient watchers of the sky, being visible to the naked eye only low down on the horizon close to sunset or sunrise – it is never seen more than 28 of arc from the Sun and is never seen against a dark sky. It was also some time before ‘morning Mercury’ and ‘evening Mercury’ where identified as the same planet. Nothing was known of its physical appearance until the advent of telescopes. Its phases and the blunting of its ‘horns’ (an optical effect) were then recognized (Figure 1). Johann Schroter (1745–1815) and W F Denning (1848–1931) claimed to have detected light and dark configurations, but their sketches bear no resemblance to the real surface as revealed by Mariner 10 in 1974. Denning also claimed to have detected a 25 h rotation period, now known to be erroneous. In 1953 A Dollfus recorded the presence of a tenuous atmosphere, which was later confirmed by Mariner 10, although it is even more tenuous than he supposed. The largest telescope cannot show Mercury as well as the Moon can be seen with the naked eye. Thus, accurate representation of a large part of its surface had to await Mariner 10, which reached a distance of 470 miles from the planet and transmitted images with a resolution of approximately 100 m showing a surface remarkably like that of the Moon, predominantly cratered with scarps, ridges, and plains.
planet, closer to the Sun than Mercury. He had earlier found the movements of Mercury to suggest that such a planet existed, but in fact the anomalous movements have since been explained, and it is certain that ‘Vulcan’, the putative inner planet, does not exist, although some asteroids may pass closer to the Sun than Mercury on their orbits oblique to the ecliptic.
Statistics
‘Vulcan’: An Inner Neighbour Planet?
Mercury is situated within the Solar System 57 850 000 km from the Sun. Its orbital eccentricity is 0.206, as determined by Antoniadi (1870–1943), the largest eccentricity of any planet except Pluto. It is, unlike Venus, brightest when gibbous. The equatorial diameter is 4880 km, intermediate between those of the Moon and Mars, more or less equal to that of Jupiter’s satellite Callisto, and less than those of Ganymede (Jupiter) and Titan (Saturn). The escape velocity is 4.3 km s 1, meaning that very little atmosphere is likely to be retained. Its density is surprising, at 5.4 g cm 3; this high value compared with the Moon requires that a heavy iron-rich core takes up a higher relative proportion of the globe than in the case of the Earth. The mass of Mercury is 0.055 times that of the Earth, and its volume is 0.056 times that of the Earth. Its orbital period of 87.969 Earth days is not, as in the case of the Moon, synchronous with its rotation around its axis, which takes 58.65 Earth days. On Mercury there is no area of permanent daylight or night and no twilight zone. It has no satellite. There is a suggestion in the literature that it may have once been a satellite of Venus – the diameter ratio is not unlike that of the Earth and Moon.
In 1958, Le Verrier received a report that a French amateur astronomer had discovered an innermost
Mariner 10 Mission Technical Summary
Figure 1 Phases of Mercury showing the optical effect of blunted ‘horns’. Reproduced from Cross CA and Moore P (1977) The Atlas of Mercury. London: Mitchell Beazley Publications.
All we know in any detail of Mercury is derived from the remarkable Mariner 10 mission, which visited both Venus and Mercury in 1974 on a gravity-assist trajectory. The mission lasted 17 months, and the same instruments were used throughout to obtain information about Earth, Moon, Venus, and Mercury – an advantage in making comparisons. There were two daylight-side encounters with Mercury as well as a night-side encounter, for orbital-change reasons, which allowed measurement of the night-surface temperature, the atmosphere, and the magnetic field. During 17 days of encounter only 17 h were spent
SOLAR SYSTEM/Mercury 239
close enough to obtain high-resolution images: 647 pictures were taken during the first daytime encounter and 300 during the second. The peculiar relation between the rotation period and the orbital period of Mercury meant that the same hemisphere was studied during both encounters. The sun rises and sets once during a Mercury day, which is two Mercury years long. During a Mercury day the planet rotates three times with respect to the stars. Results
Mariner 10 imaged 40% of the lunar-like landscape, covering virtually a complete hemisphere. Despite the startling similarity, the high density of Mercury means that similarity to the Moon is only skin deep. In the 1970s the Moon was considered to supply a ‘paradigm’ for the understanding of other planets, but the high density and geochemical properties (volatiles, refractory minerals, FeO content in the crust) of Mercury revealed by Mariner 10 suggested that Mercury is the end member of an inner–outer progression of planets, whereas Mercury is anomalous. Of course, if the suggestion that Mercury is a displaced satellite of Venus is correct, then both the Moon and Mercury are anomalous. Only further missions to Mercury will answer this question. The surface revealed by Mariner 10 has all the features of the Moon, except that it lacks extensive dark smooth plains (e.g. Imbrium) but there are quite substantial areas of lighter smooth-plain terrain, and the circular Caloris Planita feature (the largest single feature so far recognized at 1300 km in diameter) is of comparable extent to some maria and does show resemblance to lunar maria (Figure 2). The smoothplain material does appear to have lapped over, obscured, and infilled large as on the Moon craters (Figure 3), which were formed in an older, rougher surface formation, analogous to the lunar highlands, although probably not of the same composition. Craters dominate the entire mapped surface, and, as on the Moon, when one crater interferes with another the smaller crater is usually the intruder. Beethoven, the largest crater on Mercury, has a diameter of 625 km. Tolstoj (Figure 4), at 400 km, is about the same size as Mare Crisium on the Moon and is larger than any lunar crater. Some craters are double or have distorted circular outlines – Bach (225 km in diameter) shows both these features (Figure 5). Other craters have double walls. There are crater-sized rings outlined by annular grooves in otherwise flat plains. There are even circlets of small craters. There are prominent rayed craters like those of the Moon, which are apparently late-introduced features (e.g. Copley, Kuiper, Snori, Mena): Copley (Figure 6) is clearly later than the smooth plains; Mena (Figure 7)
Figure 2 Caloris Planitia (dashed line) showing the concentric ridge pattern. The basin has a diameter of 1300 km. Mountain blocks at the margin rise to 1 2 km above the surrounding terrain and the peripheral linear ridge terrain extends to 100 km from the outer edge. Photograph from NASA image bank.
displays two anomalous features – one sector totally lacks rays and the rays are both curved in places and do not all emanate from a shared point focus, a characteristic seen at other rayed craters. These features are more consistent with rays being due to deposits along fracture lines than being ejection rays. Central peaks are common and may be single central peaks or off-centre single and clustered peaks. There are few ‘Montes’ on Mercury, the only such feature recorded so far being the edge of Caloris Planita. Linear scarps called ‘Rupes’ are, however, widespread. The albedo is different from that of the Moon – on the Moon iron-rich plain basalts and light feldspathic highland anorthosites make for a dark–light contrast, whereas the surface rocks of Mercury are all relatively light coloured because of their
240 SOLAR SYSTEM/Mercury
Figure 3 Part of Tir Planitia, on Mercury, showing the flooding of older large craters by smooth plain material.
Figure 5 Bach, a double ring crater with plain material in its floor; the shape of the outer ring is subpolygonal and one side has a wall formed by an almost straight groove.
Figure 6 The rayed crater Copley, which is clearly younger than the smooth plain material. The rays extend out into the south east sector for 400 km. Note the irregularity and curved trace of the rays and the fact that rays overprint smooth plain terrain. Figure 4 The large crater Tolstoj (outlined by the dashed line), which is comparable in size to the lunar Mare Crisium.
iron-poor nature (Figure 8). At the Mercury conference in Chicago in 2001 there was a consensus that the FeO percentage in the rocks of Mercury averages around 3%. This is consistent with models in which
the planet was assembled from planetesmals that were formed near the planet’s current position. The magmas of Mercury may be similar in composition to the aubrites (enstatite achondrite meteorites), though these are demonstrably from their isotopic character asteroidal, not Mercurian, in provenance.
SOLAR SYSTEM/Mercury 241
non-metal such as sulphur is required to lower the crystallization temperature and density. Volcanism on Mercury?
Figure 7 The rayed crater Mena, the rays of which neither emanate from a single focal point nor are straight; they extend outwards for more than 250 km. Photograph from NASA image bank.
Figure 8 The contrast between the dark floored Mare Crisium on the Moon and the similar smooth Rudaki plains of Mercury; both have embaying boundaries (arrowed). Both images were taken by Mariner 10 and are reproduced from Robinson MS and Taylor GJ (2001) Ferrous oxide in Mercury’s crust and mantle. Meteoritics and Planetary Science 36: 842 847. ß 2001 by the Meteoritical Society.
There are lobate scarps on the surface that may be due to shrinkage (thermal models predict 4–6% shrinkage), and these have been suggested to be the result of thrust faulting. One entire side of Mercury remains to be seen, and this may be either similar to or very different from the cratered known surface. The magnetic field was the biggest surprise revealed by Mariner 10. Though only amounting to 1% of the strength of our own planet’s field, it is enough to indicate the existence of a core dynamo. Only the strength of the dipole component is at present known, but a solid inner core and liquid outer core are required by the present evidence. Convection in the outer core becomes more complex as the inner core grows. Thermal models suggest that the inner core of Mercury, if it exists, cannot be pure metal, and a
The impact-cratering paradigm is part and parcel of NASA’s exploration and interpretation philosophy. For example, a presentation by Potts and others in 2002 made the assumption that the overall cratering results from ‘bombardment time’. It remains possible, however, that many supposed impact craters, especially simple craters and some very complex structures, on surfaces of space bodies may have been too summarily dismissed as due to impact. Past volcanism is manifest on Mars, the Moon, and Venus, and there is active volcanism on Io. The widespread plains material on the surface of Mercury, as revealed by Mariner 10, though not as extensive as the larger lunar Maria, could be the result of primary volcanic flows or lobate crater ejecta. Study of the theoretical possibilities by Milkovich and others, published in 2002, indicates that widespread volcanism or no volcanism whatever or something in between are all possible. No volcanic features can be identified in the Mariner 10 images, although at the same image resolution few, if any, volcanic features would be identified on the Moon without the prior knowledge obtained by Apollo on-the-ground examination. High-resolution low-sun-angle images from Mariner 10 do show what appear to be flow fronts on Mercury; these could be volcanic lava flow fronts or ejecta flows. It seems likely that Mercury may show the same ‘freezing’ of the surfaces of lava flows as seen on the Moon, with lava of a basaltic (sensu lato) or a peculiar Mercurian petrological character preserving a plains surface among the craters and flooding some craters that were formed in the early history of the planet, perhaps around 4 Ga ago, with, as in the case of the Moon, very little having happened since then except for minor-scale impacts. This is, however, no more than informed speculation – Milkovich and others are correct in concluding that a clear assessment of the role of volcanism and whether it is primary or secondary (impact generated) must wait for better data. Nevertheless, there are many features in the Mariner 10 database that appear to be incompatible with impact origin – for instance, the sprinkling of an area about 400 km across centred on the crater Zeami with a myriad of small craters, all virtually the same size. These have been dismissed as ‘secondaries’, but this explanation appears facile. The scalloped walls of the largest crater, Beethoven (Figure 9), are anomalous in an impact crater, and it lies in the centre of a cluster of similarly scalloped-walled craters that include double and triple rings. Unfortunately,
242 SOLAR SYSTEM/Mercury
until 2001 when the two new ‘Messenger’ and ‘Bepi Colombo’ missions were proposed.
The Future: ‘Messenger’ and ‘Bepi Colombo’
Figure 9 The area around Beethoven, the largest Mercurian crater, showing its scalloped walls and its position within a cluster of scalloped walled craters including doubles and triples. Photograph from NASA image bank.
rigorous analysis of cratering features appears to have lapsed amongst planetologists, with the convenient assumption that impact craters can be secondarily modified in every conceivable way – and so all can be dismissed as products of the ‘Great Bombardment’. It must be concluded at the post-Mariner 10 state of knowledge that Mercury probably has had no volcanic activity, like the Moon, for nearly 4 Ga, but that the heavily cratered ‘lunar-like’ terrain cannot, as yet, be entirely dismissed as impact generated, and volcanism (either primary and endogenous or secondary, exogenous, and impact-generated) may have contributed significantly to the early formation of the crater dominated surface of Mercury. Future Mercury-directed missions will hopefully resolve this problem. The success of Mariner 10
Mariner 10 told us just enough for us to realize how interesting it would be to plan return missions, more technically equipped and specifically designed, and building on Mariner 10, to answer outstanding questions, but Mercury remained the elusive planet
It requires considerable energy to put a spacecraft into orbit around the innermost planet. NASA’s ‘Messenger’ will use multiple gravity-assist encounters when it is launched in 2004 to reach Mercury in 2009. The European Space Agency and Japan Institute of Space and Astronomical Science’s ‘Bepi Colombo’, to be launched in 2009 and arrive in 2012, will use propulsion technology, which is costlier and riskier but reduces the transit time from 5 years to 2.5 years. Messenger will study the nature of the surface, geochemistry, the Space environment, and ranging. Bepi Combo’s remit is not fully worked out but will include geochemistry. It will deploy two orbiters and a lander. Thus, the nature of the unexplored side of the planet at least will be revealed when the orbiters send back the data. The thermal environment provides the biggest challenge, with the Sun being more than 11 times as intense as on Earth and temperatures reaching up to 400 C on the sunlit side. The high temperature causes stress in equipment and may inhibit full uptake of geochemical data. The collection of mineralogical data is similarly inhibited by the blocking out of some infrared bands. Bepi Colombo will carry an actively cooled infrared spectrometer to counter this. Use of conventional solid-state detectors is impossible without power-hungry active cooling. Solar panels decay under such high temperatures. Despite these constraints, both missions will use photon (gamma, X-ray, optical) and neutron spectrometers to provide impressive geochemical information. These two missions will research the magnetic field and its implications for the core configuration. Experiments will determine whether there is an inner core, decoupled from the rest of the planet. These experiments are important in understanding the geophysical properties of the planet and its volatile inventory, sulphur being a volatile element. Space weathering by micro-impacts is likely to be greatly enhanced on Mercury compared with the Moon because of the flux of incoming particles close to the Sun (ca. X 20% has been suggested) but, because of the magnetic field, this effect may be limited to an equatorial belt. Mercury has a long comet-like sodium tail, which is probably caused by particle sputtering. Ground-based radar observations suggest that there may be water-ice in the polar regions – high
SOLAR SYSTEM/Mercury 243
radar reflectivity suggests ice, possibly mantled by dust. There is similar evidence from craters where the surface is in ‘shadow’ from direct solar heating. Thermal models, however, predict that ice would not be stable there. The two missions will investigate this problem. The tenuous atmosphere will also be the subject of investigation. Ground-based spectroscopy has detected a high sodium to potassium ratio. This is the case in the Moon’s exosphere, and there it is related to the ratio found in the returned lunar-surface rocks. However, the Mercurian ratio is very large and shows diurnal variations; it appears to be related to the magnetic field rather than to the surface rocks. Data return to Earth will also be constrained, because Mercury and the Sun interfere with radio transmission during part of the duty cycle. Antenna design is critical and constrained by weight limits and the fact that pointing antennae will tend to fail owing to the thermal cycling. Two fixed-phase array antennae on Messenger will limit the data return rate, and the same constraints will affect Bepi Colombo’s planetary orbiters. The two missions will overlap in their remits. Unfortunately, there seems to be little scope for short-term adjustment of the remit of the later Bepi Colombo mission on the basis of Messenger findings as the launch of the former in 2009 will coincide with the duty of the latter. The orbits of both missions will, of necessity, be highly eccentric, and periapsis for Messenger will be over the north pole, so the southern hemisphere will be less well mapped. Thus, it would be ideal if Bepi Colombo had its periapsis at the south pole.
Like the ejecta from the Moon’s supposed Great Bombardment, this is as yet unrecognized among Space bodies. Because the Mercury ejecta was sent out closer to the Sun, total spiralling into the Sun is more conceivable as a reason for this absence, but it is still to be expected that some of it would have formed breccias by collision with asteroids. Once we know more about the nature of Mercurian rocks it will be possible to search for such foreign material in asteroidally sourced brecciated meteorites. The renewed interest in Mercury is welcome, for the astonishing resemblence of its surface to that of the Moon revealed by Mariner 10 does suggest that when we know more about Mercury we will be able to extend this knowledge to the prime conundrum of the Earth–Moon system, and we may have to reject out of hand models for the Earth–Moon system that are at present widely supported (as we had to throw out the concept of lunar tektites after the Apollo and Lunar missions). Geologists would value more than anything a Luna-style retrieval and return of a rock sample from Mercury, but, alas, it appears that the technical obstacles are overwhelming. Yet there seem to be no limits to the ingenuity of Man. One thing is certain, there will never be a ‘one step for mankind’ on Mercury, such are the physical constraints.
See Also Earth Structure and Origins. Impact Structures. Solar System: Meteorites; Venus; Moon; Mars; Jupiter, Saturn and Their Moons. Volcanoes.
Further Reading Conclusion The idea that Mercury is a displaced satellite of Venus, though perhaps unlikely for astronomical reasons, must, unlike the ‘Vulcan’ concept, be taken seriously, in view of the surficial similarity to our Moon. If true, it would relegate the popular but criticized ‘big planet collision’ model for the origin of our Moon to the outer limits of credibility, for two such collisions of like dimensions are beyond the realms of probability. The surface rocks of Mercury have been likened to the aubrite meteorites (enstatite achondrites) in their low FeO content (though we know from isotopic evidence that these achondrites do not come from Mercury). If the surface of Mercury was moulded by a giant bombardment as is widely supposed for the surface of the Moon, then the vast amount of internal rock material ejected into Space must be somewhere.
Cross CA and Moore P (1977) The Atlas of Mercury. London: Mitchell Beazley Publications. Hunten DM and Sprague AL (2002) Diurnal variation of sodium and potassium at Mercury. Meteoritics and Planetary Science 37: 1191 1195. Koehn PL, Zurbuchen TL, Gloeckler G, Lundgren RA, and Fisk LA (2002) Measuring plasma environment at Mercury: the fast plasma spectrometer. Meteoritics and Planetary Science 37: 1173 1189. Kracher A (2002) Mercury 2001 conference Field Museum, Chicago, 2001, October 4 5. Illinois. Meteoritics and Planetary Science 37: 307 309. McCall GJH (2000) The Moon’s origin: constraints on the giant impact theory. In: Moore P (ed.) 2001 Yearbook of Astronony, pp. 212 217. London: Macmillan. McCall GJH (2002) Back to the elusive planet. Geoscientist 12: 19. Milkovich SM, Head JW, and Wilson J (2002) Identifica tion of Mercurian volcanism. Meteoritics and Planetary Science 37: 1209 1222.
244 SOLAR SYSTEM/Venus
Moore P (1961) Astronomy. London: Oldbourne. Peale SJ, Phillips RJ, Solomon SC, Smith DE, and Zuber MT (2002) A procedure for determining the nature of Mercury’s core. Meteoritics and Planetary Science 37: 1269 1283. Potter AE, Killen RM, and Morgan TH (2002) The sodium tail of Mercury. Meteoritics and Planetary Science 37: 1165 1172. Potts LV, von Freese RRB, and Shum CK (2002) Crustal properties of Mercury by morphometric analysis of
multi ring basins on the Moon and Mars. Meteoritics and Planetary Science 37: 1197 1207. Robinson MS and Taylor GJ (2001) Ferrous oxide in Mercury’s crust and mantle. Meteoritics and Planetary Science 36: 842 847. Sprague AL, Emery JP, Donaldson KL, et al. (2002) Mer cury: mid infra red (3 13.5 mm) observations show het erogeneous composition, presence of intermediate and basic soil types, and pyroxene. Meteoritics and Planetary Science 37: 1255 1268.
Venus M A Ivanov, Russian Academy of Sciences, Moscow, Russia J W Head, Brown University, Providence, RI, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Venus, similar to Earth in many ways, also shows many differences and provides insight into different paths of evolution that can be taken by Earth-like planets. The atmosphere of Venus is predominantly carbon dioxide; surface temperatures exceed the melting point of lead and surface pressures are almost 100 times that of Earth’s atmosphere. The crater retention age of the surface of Venus is very young geologically, similar to that of the Earth; however, plate tectonics does not seem to be recycling the crust and lithosphere at present. The surface is dominated by regional volcanic activity and vertical crustal accretion, and regional tectonism appears to have been much more pervasive in the earliest part of the preserved stratigraphic record, dating from less than a billion years ago. The characteristics and distribution of superposed impact craters suggest that a major resurfacing event, perhaps catastrophic in nature, occurred on Venus in its relatively recent geological history. Venus may thus be characterized by relatively recent episodic heat loss, rather than the more monotonic loss thought to be typical of the other Earth-like planets. Despite the fact that the majority of the preserved geological record on Venus dates from the last 20% of its history, Venus may provide insight into processes, such as the formation of continents, that operated in the first half of Earth history. Venus is the second largest terrestrial planet by size after Earth, and in major characteristics is close to our planet: The radius of Venus is 6051.8 km (0.95 of Earth’s radius), its mass is 4.87 1027 g
(0.81 of Earth’s mass), bulk density is 5.24 g cm 3 (0.95 of Earth’s density), and surface gravity is 8.87 m s 2 (0.91 of Earth’s gravity). For decades, Venus was considered as a ‘twin’ planet to Earth. Current knowledge of Venus geology is derived from several interplanetary missions, including landers and orbiters, as well as Earth-based observations. In the mid-1970s, the former Soviet Union conducted a series of successful landings; the Soviet landers transmitted panoramas of the surface of Venus, in addition to data on the near-surface environment, on surface rocks, and the chemistry of the atmosphere. The Pioneer Venus was the first American orbiter of Venus; launched by the United States in 1978, the Pioneer Venus collected data on global topography and gravity. The fundamental findings of this mission were that the global Venus hypsogram, in contrast to that of Earth, is characterized by one peak corresponding to the mean planetary radius (MPR), about 6051 km (Figure 1A), and that the gravity and topography of Venus are highly correlated. Three major topographic provinces characterize the surface of Venus (Figure 1B): lowlands (below MPR, 11% of the surface), midlands (0–2 km above MPR, 80% of the surface), and highlands (>2 km above MPR, 9% of the surface). The spatial resolution of the Pioneer Venus imaging radar was too low to describe morphology of the surface in detail. The systematic photogeological study of Venus began when high-resolution radar images were collected by the Soviet Venera-15/16 orbiters and by Earth-based radar observations from Arecibo Observatory. At a resolution of 1–2 km, Venera-15/16 mapped the surface in the northern hemisphere above 30o N; images from the Arecibo telescope covered a large area between 65o S–65o N and 270o E–30o E. In the early 1990s, the United States Magellan orbiter provided almost complete coverage (97% of the surface) of Venus, providing high-resolution images (100–200 m) and medium-resolution altimetry
SOLAR SYSTEM/Venus 245
Figure 1 Characteristics of the global altimetry of Venus. (A). The Venus hypsogram (the fraction of surface area vs. elevation) has one peak, implying the absence of the surface elevation dichotomy that characterizes the distribution of the surface elevation on Earth (high standing continents and low lying ocean floor). The mean planetary radius (MPR) of Venus is 6051.84 km. (B) The map in simple cylindrical projection, showing the areal distribution of the three major topographic provinces of Venus. Lowlands (light grey, below 0 km), midlands (white, 0 2 km), and highlands (dark grey, above 2 km). The majority of the surface is within the midlands.
(20-km footprint). Magellan also collected data on the Venus gravity field.
Surface Conditions and Rock Composition In contrast to Earth, a very dense atmosphere (the pressure at the surface is 95 bar) consisting primarily of CO2 (Table 1) blankets Venus. The relative role of three major contributors to the atmosphere,
primordial nebular material, volcanic outgassing, and influx of volatiles by comets, in the formation of the present atmosphere is an open question. Although the current atmosphere is very dry, a minute quantity of water is still detectable. An important feature of water in the Venusian atmosphere is that the deuterium/ hydrogen (D/H) ratio is 150 30 times higher than is found in terrestrial water. If water on Earth represents a sample of primitive water on Venus, the Venusian D/H ratio suggests that, depending on the
246 SOLAR SYSTEM/Venus
escape flux of hydrogen and deuterium, originally Venus had 260 to 7700 times the current amount of water. Such an amount is equivalent to a global layer of water 4 to 115 m deep. The dense and dry atmosphere on Venus results in a strong greenhouse effect that governs the conditions on the surface, leading to very high near-surface temperatures (740 K) and equalizing the temperatures over the planet. Important consequences are the absence of both free water and water erosion, along with significantly reduced wind activity. Thus, the principal factors resurfacing Venus are volcanism and tectonics. Volcanism is the main mechanism driving the growth of the Venusian crust. The chemical compositions of the surface rocks have been measured at seven landing sites (Tables 2 and 3). The rock chemistry correlates with the compositions of terrestrial basalts, suggesting that volcanism on Venus is mostly basaltic.
Surface Population of Impact Craters A study of the spatial density and distribution of impact craters is the principal means of understanding the age of the surface and the history of resurfacing of planetary bodies. There are 968 impact craters on the Venusian surface, making the mean crater density 2 craters per 106 km2. This value
Table 1 Composition of the atmosphere of Venus Elevation above the surface (km) Atmosphere component
From
To
Content
CO2 N2 H2O H2O H2O O2 CO SO2 Ar
1.5 1.5 45 25 0 0 0 0 1.5
63 63 54 45 15 42 42 42 24
97 4 vol. % 1.35 4.5 vol. % 500 10000 ppm 500 ppm 20 ppm 4T
½8a ½8b
where T is the tensile strength of the material. The geometrical relationships between the principal stresses and the fractures they produce (i.e., a conjugate set of shear fractures symmetrically about s1 and a single set of tensile fractures at right angles to s3) is shown in Figure 1 and, as noted below, the understanding of these relationships provides a powerful tool in fracture analysis. It follows therefore that the orientation of the fractures that form in response to a stress field is determined by the orientation of the principal stresses (Figure 1), and the type of fracture (shear or tensile) by the magnitude of the differential stress.
The Effect of a Fluid Pressure on Fracturing Fluid-Induced Failure
The state of stress in the crust is dominantly compressional. For example, in a nontectonic environment, the stress at any depth is generated by the overburden which produces a compressive vertical stress which induces a compressive horizontal stress. Thus, at any depth the Mohr stress circle will plot in the compressive regime in Figure 7B and there will be no possibility of tensile failure. Geologists were, therefore, perplexed to find that large numbers of tensional
fractures occur in the crust. This paradox was resolved when the importance of fluid pressures within a rock was understood. Pore fluid pressure within a rock increases as the rock is buried. (see Tectonics: Hydrothermal Activity). The stress state within the pores is hydrostatic and the pressure acts so as to appose the lithostatic stress caused by the overburden. This effect can be shown diagrammatically by representing the lithostatic stress as an ellipse with the stress acting compressively and the fluid pressure as a circle with the pressure acting outwards (Figure 8A). The fluid pressure reduces all the lithostatic stresses by an amount Pfluid to give an effective stress. Thus, the principal stresses s1 and s3 become (s1 Pfluid) and (s3 Pfluid). This new stress field can be plotted as a Mohr stress circle (Figure 8B). It can be seen that the original lithostatic stress circle is moved towards the tensile regime but that the diameter of the circle, i.e., the differential stress, remains unchanged. The amount of migration of the stress circle is determined by the magnitude of the fluid pressure. Thus, as the fluid pressure gradually increases during burial, the stress circle is pushed inexorably towards the failure envelope. When it hits the envelope, failure occurs. Such failure is termed fluid induced or hydraulic fracturing. In this way an originally compressional stress regime can be changed so that one or more of the principal stresses becomes effectively tensile and the conditions for tensile failure can be satisfied. The Expression of Fluid-Induced Failure
In the example shown in Figure 8B the lithostatic stress had a small differential stress (i.e., less that 4T (see eqn [8]) and as a result the induced hydraulic fractures were tensile fractures. If it had been greater than 4T, shear fractures would have formed. The Organization of Tensile Fractures
The Mohr circles shown on Figure 9 all intersect the failure envelope in the tensile regime, i.e., the differential stresses are all less than 4T and will all therefore result in tensile failure. Their differential stresses vary from just less than 4T (circle (i) Figure 9), to zero (circle (iv), Figure 9). Note that when the stress state is hydrostatic, the Mohr circle is reduced to a point. As noted above, tensile fractures form normal to the minimum principal compressive stress s3 (Figure 1B), i.e., they open against the minimum compressive stress. The stress state represented by Mohr circle (i) in Figure 9, has a relatively large differential stress and there is, therefore, a definite direction of easy opening for the fractures. The fractures would
TECTONICS/Fractures (Including Joints) 357
Figure 8 (A) Diagramatic representation of the effect of a fluid pressure (the circle with the outwardly acting stress) on the stress state in a rock (the ellipse with the inwardly acting stresses). All normal stresses are reduced to an effective stress (s pfluid) but the differential stress (s1 s3) remains unchanged. The effect is to cause the Mohr stress circle to move to the left by an amount equal to the fluid pressure; (B). Thus depending on the magnitude of the differential stress the induced fractures will be either shear (stress state (i)) or tensile (stress state (B)) (see Tectonics: Folding).
Figure 9 (A) Mohr stress circles (i) (iv) representing a range of stress states, all of which will lead to tensile failure. NB the Mohr circle (iv), that represents hydrostatic stress is a point. (B) Pat terns of tensile failure generated by the corresponding stress states shown in (A).
therefore exhibit a marked alignment normal to this direction (Figure 9B (i)). However, for the stress states represented by the Mohr circles (ii–iv), the differential stress becomes progressively smaller until, for the hydrostatic stress represented by circle (iv), the differential stress is zero. In a hydrostatic stress field the normal stress across all planes is the same and there is, therefore, no direction of relatively easy opening for the fractures. Thus, they will show no preferred orientation and, if they are sufficiently closely spaced and well developed, will produce a brecciation of the rock (Figure 9B (iv)). It can be seen that as the differential stress becomes progressively lower so the tendency for the resulting tensile fractures to form a regular array normal to s3 decreases. Tensile fracture systems ranging from well-aligned fractures to randomly
358 TECTONICS/Fractures (Including Joints)
Figure 11 Polygonal arrays of tensile fractures cause by the desiccation of a silt layer.
Figure 12 Polygonal arrays of tensile fractures cause by the cooling of a lava flow. (Giant’s Causeway Northern Ireland).
Figure 10 (A) A regular array of tensile fractures exposed on a bedding plane in Carboniferous sandstone, Millook, Cornwall, England. (B) Less well organized tensile fractures cutting Devon ian sandstones, St. Anne’s Head, South Wales. (C) Carboniferous sandstone cut by randomly oriented tension fractures.
In both these examples, the fractures are organized into polygonal arrays showing that the tensile stresses generated were the same in all directions.
Fracture Sets
oriented fractures are to be expected in rocks, and field observations support this idea, Figure 10. As noted above, the problem of forming tensile fractures in the compressive stress field that generally characterizes the Earth’s crust can be solved by appealing to high fluid pressures. However, tensile failure can occur in rocks without the aid of a high internal fluid pressure, for example, during the contraction of a layer as a result of desiccation of a sediment (Figure 11) or the cooling of an igneous body (Figure 12).
Generally, the state of stress in the Earth’s crust is not hydrostatic. Consequently a single episode of deformation is likely to generate a set of fractures with the same orientation. However, most rocks experience several different stress regimes during their history with the result that several fracture sets are frequently found superimposed on each other to produce a fracture network (Figure 13). The interaction of late fractures with early fractures is illustrated in Figure 14. The effect of early fractures on later ones is to arrest their propagation and to modify their orientation. It can be seen from Figures 14A and B that one
TECTONICS/Fractures (Including Joints) 359
fracture often ends abruptly against another. This abutting relationship gives the relative age of the fractures, i.e., the later fracture abuts against the earlier fracture. If an early fracture is an open fracture
then it will represent a free surface within the rock and, as discussed in the above section on classification of faults, will be unable to support a shear stress. Consequently, the principal stresses will reorient as they approach it into a position either normal or parallel to the fracture. This effect can be clearly seen in Figure 14, where later fractures curve into an orientation at right angles to the earlier fracture as they approach it.
Fracture Networks
Figure 13 A fracture network in a Liassic limestone bed from Lilstock, North Somerset, England. It was produced by the super position of individual fracture sets.
Structural geologists study the cross-cutting relationships of different fracture sets in order to determine their relative age. A variety of rules have been established to help in this task. It is found that early fractures tend to be long and relatively continuous and, as noted above, later fractures abut against these and are consequently shorter. Some of these features can be seen in Figure 15, which shows a fractured limestone pavement containing several fracture sets. The
Figure 14 Details of the limestone pavement shown in Figure 13 illustrating the interaction of late fractures with early frac tures. The effect of early fractures on later ones is to arrest their propagation and to modify their orientation. It can be seen that the later fractures are deflected by and abut against the earlier fractures.
Figure 15 Fracture patterns in a limestone pavement at Lil stock, North Somerset, SW England. The older fracture sets are the most continuous and, as the sets become progressively younger, they become less continuous and less well oriented.
360 TECTONICS/Fractures (Including Joints)
longest and most continuous runs approximately N–S. These are the oldest fractures and are crosscut by several younger sets which become progressively less continuous and less aligned as the regional stress fields responsible for their formation becomes progressively modified by the pre-existing fractures. The fracture set trending approximately NW–SE, the second set to form, shows a remarkable degree of continuity, being only affected by the N–S fractures; its orientation is related directly to the regional stress field. However, as more fracture sets develop in the rock mass, modification of the stress orientation by the pre-existing fractures may result in there being a poor correlation between the fracture orientation and the regional stress field responsible for its formation. This is well illustrated in subarea A in Figure 15 which has been enlarged in the bottom left-hand corner of the figure. The influence of the pre-existing fractures on the orientation of the late fracturing is so marked that the later fractures display a polygonal organization and cannot be linked directly to the regional stress field responsible for their formation.
Fracture Analysis A fracture analysis is the study of a fractured rock mass in order to: (i) establish the detailed geometry of the fracture network; (ii) determine the sequence of superposition of the different fracture sets that make up the fracture network; and (iii) deduce the stress regime associated with the formation of each fracture set. The reason why a detailed knowledge of the geometry of the fracture network is so important is that the bulk properties (e.g., strength, permeability) of a fractured rock mass (and most natural rocks are fractured) are generally determined by the fractures they contain rather than by the intrinsic rock properties. Stages (ii) and (iii) of a fracture analysis are carried out using the principals outlined above relating to the interaction of fractures and the relationship between the stress field and fracture orientation (Figure 1).
vertical minimum stress (Figure 5C). Divergent plate margins result in the formation of oceans and the separation of plates. The initial stage of this process is the fracturing of the lithosphere and the formation in the upper crust of major rift systems such as the East African Rift (see Tectonics: Rift Valleys). The stress regime of a horizontal minimum principal compressive stress and a vertical maximum stress is appropriate for the formation of normal faults (Figure 5A). When plates move parallel to each other at different velocities, conditions are appropriate for the formation of major wrench (strike-slip) faults (Figure 5B) such as the San Andreas Fault zone of California which separates the Pacific and North American plates. Thus it can be seen that each of the three types of plate margins is characterized by a different types of fault.
Scale of Fracturing Fractures occur on all scales within the Earth’s crust, ranging from major faults that define plate margins, through faults that can be seen on seismic sections (see Tectonics: Seismic Structure At Mid-Ocean Ridges), down to faults that can be observed directly in the field, e.g, Figure 4, to microscopic fractures only visible under the microscope. Detailed studies of the microfractures in rocks at different stages of the evolution of tensile fractures show, as predicted by Griffith’s theory of stress magnification (1925) outlined above, that the microfractures grow by tensile failure at the crack tips and that suitably located microfractures link to form larger fractures oriented normal to s3, the minimum compressive stress (see Tectonics: Faults). More remarkably, when the growth of shear fractures are studied in the same way, it is found that
Types of Faults a Plate Margins The ‘type’ of plate margin is controlled by the relative motion of the two adjacent plates. They can be subdivided into three classes, convergent, divergent, and strike-slip. Convergent margins lead to compressional regimes at the plate margins which results in the formation of mountain belts. The stress regime is that appropriate for thrusts to form, namely a horizontal maximum principal compressive stress and a
Figure 16 Randomly oriented micro fractures within a material and their growth by tensile failure and subsequent linkage to form (A) macroscopic tensile fractures and (B) macroscopic shear fractures.
TECTONICS/Fractures (Including Joints) 361
Figure 17 A block diagram illustrating the different types of surface structures (patterns) and fracture trace architecture. (Based on Kullander et al., 1990.) (1) Main joint face, (2a) abrupt twist hackle fringe, (2b) Gradual twist hackle fringe, (3) Origin of fracture, (4) Hackle plume, (5) Plume axis, (6) Twist hackle face, (7) Twist hackle step, (8) Rig marks (front line of the fracture), (9) hooking, (10) En echelon fractures.
initially fracturing occurs by the growth of microfractures at their tips and in an orientation normal to s3. However, the macroscopic shear fractures are formed by the linking of offset microfractures, as shown in Figure 16. Thus, it can be seen that despite the two types of fractures having independent failure criteria and different orientations with respect to the principal stresses, they are nevertheless fundamentally linked on a microscopic scale. They are both the result of the growth of microfractures by tensile failure and differ only in the way in which these fractures are linked to a macroscopic fracture.
Surface Features of Fractures Fractures display a variety of surface features and Figure 17 is a summary diagram showing some of these. Fractography is the science which deals with the description, analysis, and interpretation of fracture surface morphologies and links them to the causative stresses, mechanisms, and subsequent evolution of the fractures. It has been demonstrated that the diverging rays of the plumose structures (Figures 17 and 5) always remain parallel to the direction of propagation of the fracture. Thus, by constructing lines at right angles to these rays, the position and shape of the fracture front at different times of its evolution can be determined (Figures 17 and 8).
When the exposures are sufficiently good, it is found that the fracture fronts form a series of concentric ‘ellipses’, the centre of which marks the site of fracture initiation.
See Also Tectonics: Earthquakes; Faults; Folding; Hydrothermal Activity; Seismic Structure At Mid-Ocean Ridges; Rift Valleys.
Further Reading Ameen MS (1995) Fractography: fracture topography as a tool in fracture mechanics and stress analysis. Special Publication Geological Soc. Of London, No. 92; p 240. Anderson EM (1951) The dynamics of faulting and dyke formation with application to Britain, 2nd edn. Oliver & Boyd. Griffith AA (1925) The theory of rupture. 1st Inter national Conference of Applied Mechanics Proceeding Delft. P55. Kullander BR, Dean SL, and Ward BJ (1990) Fracture Core Analysis: Interpretation, Logging, and Use of Natural and Inducted Fractures in Core: Methods in Exploration Series, No. 8. Tulsa, Oklahoma, USA: American Associ ation of Petroleum Geologists. Mandl G (1999) Faulting in Brittle Rocks. Springer. Price NJ (1966) Fault and joint development in brittle and semi brittle rock. Pergamon press.
362 TECTONICS/Hydrothermal Activity
Hydrothermal Activity R P Lowell, Georgia Institute of Technology, Atlanta, GA, USA P A Rona, Rutgers University, New Brunswick, NJ, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction Hydrothermal activity results from the complex interplay of heat transfer, fluid–rock chemical reactions, and fluid circulation within Earth’s continental and oceanic crust. Hydrothermal circulation redistributes heat energy in the crust, often giving rise to regions of concentrated thermal output that lead to the emplacement of economically important mineral deposits and that serve as geothermal energy resources. Hydrothermal activity is thus an important component of Earth’s global heat engine whereby heat transferred to the lithosphere by mantle convection is transferred to Earth’s surface by thermal conduction, volcanic extrusion, and hydrothermal venting. Lithospheric plate motions, volcanic and tectonic activity, and earthquakes are manifestations of Earth’s global heat engine. There is a close connection between tectonic plate boundaries and sites of hydrothermal activity (Figure 1). In the following sections of this article we compare some aspects of terrestrial and submarine hydrothermal activity, describe the basic physics and chemistry of hydrothermal circulation, briefly discuss the importance of two-phase flow, and suggest some directions for future study.
Comparison between Terrestrial and Submarine Activity Although hydrothermal activity in terrestrial and submarine settings has many similarities there are significant distinctions. Part of the reason for this distinction stems from the manner in which heat is transported in continental and oceanic lithosphere, respectively (Table 1). In terrestrial settings, nearly 40% of the heat flux stems from radiogenic heat production in the crust, and 60% is conducted from the underlying mantle. Terrestrial hydrothermal activity accounts for less than 1% of Earth’s thermal budget. On the other hand, the process of plate creation and seafloor spreading along the roughly 60 000-km ocean ridge system dominates the thermal regime of the oceanic lithosphere. Conductive heat flux from the spreading lithosphere decreases as t 1/2, where t is the lithospheric age. Hydrothermal circulation transports a significant fraction of the lithospheric heat
advectively, leading to lower than expected conductive heat flow in young lithosphere (Figure 2). Nearly 25% of Earth’s global heat loss and 33% of the heat loss from oceanic lithosphere result from hydrothermal activity. Most seafloor hydrothermal heat loss occurs at low temperature. High-temperature hydrothermal activity, which accounts for less than 10% of the total seafloor hydrothermal heat loss (Table 1), appears to occur only in lithosphere less than 1 My old. Another interesting distinction between terrestrial and submarine hydrothermal activity has been their role in human endeavours. Warm and hot springs on the continents have been used for bathing and medicinal purposes since antiquity. Thermal springs were utilized throughout the Roman empire, and early descriptions of springs in Europe appear in seventeenth-century writings. Terrestrial hydrothermal systems have also long been used as an energy resource. Geothermal waters in Iceland have been used for heating for centuries, and by the 1930s a centralized heating system was established for Reykjavik. Geothermal steam has been produced at Lardarello, Italy, since the latter half of the nineteenth century, and the Geysers geothermal field in California was first exploited in the 1920s. It is remarkable that commercial development of geothermal resources occurred long before measurements of geothermal heat flux and without detailed geophysical exploration. On the other hand, direct detection of submarine hydrothermal activity did not occur until the 1960s (Red Sea) and the 1970s at mid-ocean ridges of the Atlantic and Pacific. As a result of these discoveries the understanding of biogeochemical processes on Earth was revolutionized. It became clear that submarine hydrothermal circulation significantly impacts global geochemical cycles of both the lithosphere and the ocean. Chemical transport of certain major and trace elements to the ocean by hydrothermal discharge equals or exceeds river inputs. Moreover, hydrothermal fluids also serve as an energy resource for complex chemosynthetic biological ecosystems. The discovery of chemosynthetic ecosystems at seafloor hydrothermal vents has led to a new awareness of life in extreme environments and has stimulated the discussion of the origin of life on Earth and other planetary bodies in the solar system.
Physics and Chemistry The fundamental components of hydrothermal activity are a heat source and a fluid circulation system
TECTONICS/Hydrothermal Activity 363
Figure 1 Plate tectonic map of the world showing locations (1 50) of selected submarine and terrestrial high temperature hydro thermal sites as follows: (1) Krafla; (2) Namafjall; (3) Svartsengi; (4) Rainbow; (5) Lost City; (6) TAG; (7) Snake Pit; (8) Logatchev; (9) Larderello; (10) Mt. Amiata; (11) Travale; (12) Kizildere; (13) Afyon; (14) Atlantis II Deep; (15) Olkaria; (16) Puga; (17) Kawah; (18) Kamodjang; (19) Dieng; (20) PACMANUS; (21) North Fiji Basin; (22) Lau Basin; (23) Brothers; (24) Kawerau; (25) Rotorua; (26) Broadlands; (27) Wairakei; (28) Tiwi; (29) Mariana Trough; (30) Okinawa Trough; (31) Otake; (32) Sunrise; (33) Matsukawa; (34) Paratunka; (35) Pauznetska; (36) Magic Mountain; (37) Main Endeavour; (38) Sea Cliff; (39) Escanaba; (40) Yellowstone; (41) Geysers; (42) Imperial Valley; (43) Cerro Prieto; (44) Guaymas Basin; (45) East Pacific Rise 21 N;(46) Pathe; (47) East Pacific Rise 9 N; (48) Galapagos; (49) Rapa Nui; (50) El Tatio.
Table 1 Heat flux from the Earth (1012 W)a Type
Continental Crust (a) Crustal radiogenic heat production (b) Conductive heat flux from the mantle (c) Extrusion of lavas (d) Hydrothermal flux Total Oceanic Crust (a) Conduction (b) Extrusion of lavas (c) Axial high temperature hydrothermal flux (d) Axial low temperature hydrothermal flux (e) Off axis low temperature hydrothermal flux Total hydrothermal flux Total Global Heat flux
Value
4.6 6.8 0.03 0.1 11.5 20.3 0.3 0.3 2.7 7 10 30.6 42.2
a Compiled from Sclater JG, Jaupart C, and Galson D (1980) and Elderfield and Schultz (1996); modified after Lowell (1991).
Figure 2 Observed mean heat flow for oceanic spreading centres compared with theoretical curve for conductive cooling of lithosphere. Reproduced from Anderson RN, Langseth MG Jr, and Slater JG (1977) The mechanisms of heat transfer through the floor of the Indian Ocean. Journal of Geophysical Research 82: 3391 3409.
364 TECTONICS/Hydrothermal Activity
Figure 3 (A) Cartoon of single pass hydrothermal circulation model at an ocean ridge crest. The major single pass segment refers to a deep circulation cell in which seawater recharge penetrates through sheeted dykes to near the top of a magma body, takes up heat and undergoes water rock chemical reactions while flowing quasihorizontally, and ascends through faults or fractures to the seafloor as high temperature focussed black smoker flow. Mixing of the deep circulation with shallow cooler circulation in the basaltic pillow lavas may result in diffuse discharge. Reproduced from Germanovich LN, Lowell RP, and Astakhou DK (2000) Stress dependent
TECTONICS/Hydrothermal Activity 365
(see Geysers and Hot Springs). It is the nature of the heat source that generally determines whether hydrothermal activity occurs at high (>150 C) or low temperature. The circulation system consists of a recharge zone through which fluids enter the crust, a region in which the fluid takes up heat from its surroundings, and a discharge zone, through which the heated hydrothermal fluid emerges at the surface as a hot spring or hydrothermal vent. Although fluid may sometimes recirculate several times before exiting the system, it is often convenient to describe circulation in terms of a simple single-pass circulation model. Figure 3 shows cartoons of single-pass models envisioned for high-temperature terrestrial and submarine systems and a low-temperature warm spring system. In addition, all hydrothermal activity exhibits temporal variability, and chemical reactions between the circulating fluid and rock are often important. Heat
Geothermal gradient Conductive heat flux, H, is related to the geothermal gradient by H ¼ l dT/dz, where l is the thermal conductivity. For rocks, l ranges from approximately 1.8 to 5 W (m C) 1, with most igneous and metamorphic rocks falling into a narrower range between 2.0 and 2.5 W (m C) 1. In older, stable continental cratons, the geothermal gradient may be as low as 10 C km 1, whereas in active volcanic regions it may be more than 100 C km 1. A typical geothermal gradient of 25 C km 1 gives a conductive heat flux of 60 mW m 2. In terrestrial low-temperature hydrothermal activity, fluids driven by a topographic head circulate to a depth of 1–3 km in the crust where they are heated by the geothermal gradient. The fluids emerge through faults at the surface as warm or hot springs with temperatures ranging from a few tens of degrees above ambient to the local surface boiling temperature (Figure 3C). Such springs are found worldwide in areas of both normal and elevated heat flow. Low-temperature hydrothermal circulation in oceanic crust occurs from ridge axes to a lithospheric age of 60 My. This circulation is partially controlled seafloor topography in combination with the geothermal gradient, with discharge occurring at highs and recharge occurring at topographic lows. Type and thickness of sediment cover also influences this circulation. More than 90% of all hydrothermal heat loss from the seafloor occurs at low temperature.
This circulation impacts geochemical cycles as the equivalent of an ocean volume approximately evens 106 years. Magmatic heat High-temperature hydrothermal activity (typically classified as > 150 C) is associated with active volcanism. In these settings, shallow magmatic intrusions provide the heat source. Part of this heat comes from the latent heat of crystallization and part of the heat is derived from the cooling pluton. Thermal buoyancy differences between the colder and hotter parts of the system drive convective fluid motions. As volcanism is associated with ocean ridges, hot spots, and island arc systems (fore-arc, arc, and back-arc settings) at subduction zones, it is not surprising that essentially all high-temperature hydrothermal activity occurs in these regions (Figure 1). In terrestrial settings, boiling hot springs and geysers provide the surface expressions high-temperature hydrothermal activity. Reservoir temperatures of these systems typically lie between 200 and 350 C. In oceanic settings vigorous high-temperature hydrothermal activity is exhibited as ‘‘black smoker’’ venting at temperatures between 300 and 400 C (Figure 4). Lower temperature ‘‘white smokers’’ with temperatures 150–200 C are also common. Because of the high pressure (250 bars) at the seafloor, these hightemperature vents lie below the boiling temperature. As discussed later, however, boiling and phase separation appear to occur in the subsurface of both terrestrial and submarine high-temperature hydrothermal systems. Chemical heat It has long been recognized that hydration of peridotite is an exothermic reaction that produces heat, that alters the chemistry of the rocks and hydrating solutions involved, and that expands the volume of the rocks (40%). It is only now emerging how widespread this process called the ‘‘serpentinization reaction’’ may be beneath ocean basins and possibly continents. The reaction involves peridotite, the characteristic ultramafic rock type of the Earth’s upper mantle, and either seawater or meteoric water. Serpentinization is commonly observed in ultramafic rocks recovered from the seafloor and in slices of ancient oceanic mantle exposed on land as ophiolites. This reaction yields distinctive chemical solutions characterized by high alkalinity, high ratios of Ca to Mn and other metals, and abiogenic
permeability and the formation of seafloor event plumes. Journal of Geophysical Research 105: 8341 8354. (B) Analogous cartoon representing single pass flow in a high temperature terrestrial system. Adapted from White (1973) Characteristics of geothermal resources. In: Kruger P and Otte C (eds.) Geothermal Energy. Stanford, CA: Stanford University Press. (C) Cartoon of a terrestrial low temperature warm spring system. Reproduced from Lowell (1992) Hydrothermal systems. In: Encyclopedia of Earth System Science, vol. 2, pp. 547 557. San Diego, CA: Academic Press.
366 TECTONICS/Hydrothermal Activity
Figure 5 Schematic of dD d18O relationship in meteoric waters. Reproduced from Craig (1961). Horizontal arrows indi cate the d18O shift generally found in hydrothermal fluids that results from isotopic exchange with d18O enriched igneous and metamorphic rocks.
Figure 4 Black smoker vent at the East Pacific Rise 21 N hydrothermal field. (ß Woods Hole Oceanographic Institution, Woods Hole, Massachusetts.)
generation of methane (CH4) and hydrogen (H2) gas. The heat released, depending on the volume and rate of serpentinization, may be sufficient to drive hydrothermal circulation over a range of fluid temperatures, typically low to intermediate (degrees to tens of degrees Celsius), and possibly up to several hundred degrees Celsius. Serpentinization is favoured by conditions that facilitate access of water to large volumes of the upper mantle. In ocean basins the conditions include a low magma budget, which produces thin ocean crust, and tectonic extension and volume expansion that creates permeability through fractures and faults and that exposes rocks of the upper mantle on the seafloor. Such conditions generally occur at sections of slowspreading ocean ridges in the Atlantic, Indian, and Arctic oceans. For example, fluids with the chemical signatures of serpentinization reactions are common along the mid-Atlantic ridge where several hightemperature (to 360 C) seafloor hydrothermal fields (Logatchev at 14 450 N, 44 580 W and Rainbow, 36 140 N, 33 540 W) at least partially situated in serpentinized ultramafic rocks of the upper mantle have been found. Only one of these fields appears to be an end member of a hydrothermal system entirely driven by serpentinization reactions (Lost City field, near
30 N, 42 W) located about 15 km west of the eastern intersection of the rift valley with the Atlantis Fracture Zone, where the field is apparently isolated from magmatic heat sources. There serpentinizationderived fluids are discharging at temperatures up to 75 C and precipitating calcium carbonate and magnesium hydroxide chimneys, which have grown up to 60 m high. Thermal and chemical fluxes from such serpentinization-driven seafloor hydrothermal systems have yet to be determined, but may be a significant fraction of global hydrothermal mass and heat budgets. Seawater and upper mantle rocks are ubiquitous in ocean basins, although the sites of serpentinization may be localized. Fluid Sources
The aqueous fluid involved in hydrothermal activity can, in principle, have several different origins. Meteoric waters are the predominant fluid, but metamorphic or magmatic fluids may also contribute. The origin of the fluid is generally determined by examining their oxygen and hydrogen isotopic ratios (Figure 5). Meteoric waters are defined by a characteristic linear relationship between dD and d18O (MWL), whereas metamorphic and magmatic rocks and waters tend to be enriched in d18O relative to the MWL, and hence lie to the right of the curve. Ocean waters occupy a small range of dD–d18O space near the MWL (denoted by SMOW). Hydrothermal source waters are typically meteoric (or seawater) and hence lie somewhere along the MWL. At temperatures greater than 200 C, hydrothermal waters may be enriched in d18O as a result of isotopic exchange during water–rock reactions. The presence of a magmatic or metamorphic component may also move the
TECTONICS/Hydrothermal Activity 367
isotopic signature of the fluid to the right of the MWL (Figure 5). Isotopic evidence of a magmatic component in active hydrothermal systems is generally inconclusive, but the presence of CO2 in some hydrothermal fluids points to the presence of magmatic volatiles. Seawater is by far the predominant fluid in submarine hydrothermal systems. The Circulation System
k ¼ Cb2 fn
Heat transport by fluid flow through the rock requires interconnected fluid pathways and a driving force for fluid flow. The relationship between the driving ˆ force, the gradient of hydraulic head @ H=@x j , and volumetric flow rate per unit area per unit time, or specific discharge qi, is generally given by the empirical relationship called Darcy’s Law. ˆ qi ¼ ðgkij =nÞð@ H=@x jÞ
desirable for two reasons. First, porosity is scale independent and therefore laboratory-based measurements of porosity are meaningful; secondly, in situ porosity can be estimated from both electrical and seismic data. Mathematical models relating effective porosity f to a scalar bulk permeability k are generally of the form
½1
where g is the acceleration due to gravity, n is the kinematic viscosity of the fluid, and kij is the intrinsic rock permeability tensor, respectively. Subscripts, i, j refer to the Cartesian coordinate directions. In many applications kij is treated as a scalar k; the units of kij are m2. Rock permeability is the single most important physical parameter that affects hydrothermal circulation. This parameter is a measure of the interconnectivity of pore spaces and fractures; however, these features and their interconnectivity may depend upon physical and chemical processes related to the flow itself. Consequently, in any given hydrothermal environment, rock permeability may be a complex function of time and space that is difficult to determine in situ at the field scale. Moreover, permeability is often heterogeneous and anisotropic; it is a scaledependent parameter that may vary over several orders of magnitude on relatively small spatial scales. Considerable research effort has been devoted to the determination of permeability and its temporal evolution during hydrothermal activity. In the following subsections we discuss some approaches to describing permeability in hydrothermal systems and its temporal evolution. Porous medium permeability The percentage of rock volume that may be occupied by fluid is termed the porosity. To the extent that this porosity is interconnected it may give rise to permeable pathways for fluid flow. Such porosity-related permeability is termed primary permeability. There have been several mathematical models attempting to relate effective, or interconnected, porosity, f, with permeability, but these have had limited success. A mathematical relationship between porosity and permeability is highly
½2
where b is the average grain size of the medium and C is a numerical constant, respectively. The exponent n ranges between 2 and 3 in most formulations. The well-known Carmen-Kozeny relation is similar to eqn [2]. Equations of the form [2] often fail in practice because in most systems, even those that have significant primary permeability, field-scale permeability is controlled by fractures. Fracture- and fault-related permeability Permeability in essentially all hydrothermally active regions is controlled by fractures and faults. Such permeability is termed secondary. In igneous and metamorphic rocks, which host most high-temperature hydrothermal activity, cracks must provide the main permeability because porosity and, hence, primary permeability is low. When permeability is controlled by fractures, large permeability can exist in the presence of very low interconnected porosity. In fracture-controlled systems, permeability is related to crack density, the abundance of crack intersections and the cube of the crack aperture. A generalized formulation can be written as k ¼ C0
l3 Na2 h
½3
where C0 is a dimensionless coefficient describing the degree of crack interconnectivity, l is the mean crack aperture, h is the crack spacing, a is the crack length, and N is the number of cracks per unit area. As special case of [3], one may consider a set of planar parallel cracks of aperture l and spacing h. In this case the permeability is k¼
l3 h2 3 ¼ f 12h 12
½4
where the porosity f ¼ l/h. Figure 6 depicts k vs f for selected values of h; the results show that large values of crack permeability can exist for f 1%. These values are several orders of magnitude greater than 10 18–10 20 m2, which are typical laboratory values for unfractured granite. Estimates of permeability for hydrothermally active regions have been determined from borehole
368 TECTONICS/Hydrothermal Activity
Figure 6 Graph of permeability versus porosity at a number of given fracture spacings for rock permeability resulting from planar parallel fractures. The curves show that high fracture permeability can occur in low porosity rocks.
measurements and from field measurements of crack distributions in fossil systems. Field-scale permeability has also been estimated from mathematical modelling of hydrothermal heat output. Measurements made in Deep Sea Drilling Project and Ocean Drillng Project boreholes give permeability values ranging from 10 18 m2 in sheeted dykes to as high as 10 10 m2 in pillows. Values in continental hydrothermal systems often fall between 10 12 and 10 15 m2. Crack spacing and apertures in ophiolites yield permeability values ranging between 10 13 and 10 8 m2. Mathematical modelling studies of high-temperature venting at ocean ridge crests give a similar range. The high values of permeability and the broad range of values estimated from field and modelling studies further indicate that permeability in hydrothermal regions is fracture-controlled. Fracture-controlled permeability is typically heterogeneous and possibly anisotropic. Fracture concentration and orientation may result from tectonic stresses as well as processes related to magma emplacement and volcanic eruptions and thermal stresses. Zones of high fracture-controlled permeability are associated with dikes. In both continental and submarine hydrothermal systems discharge zones are often focussed along tectonic faults. Recharge zones are more problematical, but faulting could be important there as well. Although fractures and faults control hydrothermal circulation patterns, one does not often consider flow in discrete fractures. Mathematical models usually treat flow in fractured rock as an equivalent porous medium and apply Darcy’s Law. It is important in this regard, however, to recognize that fracture permeability can exist on several spatial scales and that the permeability may depend on time. Temporal variations in permeability Temporal changes in permeability can result from several
mechanisms. In addition to changes in response to tectonic and magmatic processes, both thermal and chemical processes can be significant. Because the processes have not been quantified in great detail, their relative importance is uncertain. As circulating aqueous fluids encounter different pressure and temperature environments dissolution and precipitation of chemical constituents may occur. During water–rock reactions, hydrothermal fluids reach thermodynamic equilibrium with quartz. At pressures of a few hundred bars, the solubility of quartz reaches a maximum between 350 and 400 C. Thus if the hydrothermal solution is heated above 400 C quartz will precipitate and clog fractures and pore spaces. Similarly as the hydrothermal solution ascends towards the surface, both lower pressures and temperatures in the environment will foster quartz precipitation. Both quartz and amorphous silica are common vein minerals in hydrothermal systems, and precipitation of these phases may exert a strong influence on hydrothermal circulation over time. The development of a low permeability barrier as a result of quartz precipitation is likely an important factor in the evolution of vapour-dominated hydrothermal systems. In submarine hydrothermal systems precipitation of anhydrite may be important in both recharge and discharge zones. Because the solubility of anhydrite decreases with increasing temperature, heating of seawater during hydrothermal recharge to T 150 C results in precipitation of anhydrite in recharge zones (Figure 3A). Sulphate is removed from seawater by precipitation of anhydrite and reaction with crustal rocks; however, upon ascent, mixing of sulphate-poor, hot hydrothermal fluid with cold sulphate-rich seawater may again result in the precipitation of anhydrite. Mixing in the subsurface may contribute to focussing seafloor venting into black smokers; whereas mixing above the seafloor contributes to the formation of chimney structures. Thermoelastic stresses result from the passage of either cold fluids through initially hotter rock or hot fluid through cooler rocks. In the former case, cooling of rock surfaces leads to thermal contraction and the enhancement of permeability. In the latter case, heating of the rock leads to thermal expansion and reduction of permeability. The dependence of permeability on temperature can be expressed as k ¼ k0 ½1 gðT T0 Þ3 þ kres
½5
where k0 is the permeability of the main crack network, kres is a finer-scale residual permeability, and g is factor expressing the strength of the thermoelastic effect.
TECTONICS/Hydrothermal Activity 369
At temperatures exceeding 350–400 C, rocks begin to exhibit ductile behaviour. Although this behaviour depends upon the rate which stresses are applied, ductile behaviour will tend to seal cracks. Thus permeable pathways that may initially be opened by stresses resulting in brittle failure may gradually close. This process may limit the depth to which cracks remain open in the crust and the extent to which hydrothermal circulation may approach magma bodies. Water–Rock Chemical Reactions
As aqueous fluids pass through hot subsurface rocks, chemical reactions occur. Some chemical constituents may be removed from the fluid, others may be extracted from the rock. The reactions may also involve isotopic exchange between the fluid and rock. These reactions are complex functions of temperature, pressure, lithology, permeability structure, duration of activity, and other factors. A detailed discussion of this topic is beyond the scope of this article; however, the use of geochemical thermometers and the formation of hydrothermal ore deposits are discussed briefly. Geochemical thermometers The strong temperature dependence of solubility of certain chemical constituents in hydrothermal fluids, the temperature dependence of elemental partitioning between rock and solution, and the temperature dependence of isotopic partitioning between mineral and fluid phases have led to the development of a variety of geochemical thermometers to deduce subsurface conditions from surface samples. The quartz geothermometer utilizes the strong temperature dependence of quartz solubility and the slow kinetics of quartz precipitation at low temperature. As hydrothermal solutions in equilibrium with quartz at high temperature rise to the surface and cool, the high degree of disequilibrium in the measured quartz concentration permits a calculation of the equilibrium temperature at depth Other common geothermometers include: 1. Na/K, which makes use of the temperature dependence of partitioning of these elements between aluminosilicate rocks and hydrothermal fluid; 2. Na–K–Ca, which includes the effect of Ca in the partitioning; and 3. ratios of stable isotopes such as d13C, d18O, dD, and d34S. Various factors affect the resolution and reliability of each of these geothermometers, so often many independent ones are used.
Ore deposits–fossil hydrothermal systems As a result of water–rock chemical reactions at intermediate to high temperature, trace metallic ore-forming metals such as Fe, Cu, Zn, Sb, Au, Ag, and Pb are transferred from the rock to the hydrothermal solution. Because most of these metals form metallic sulphides that are highly insoluble in water, solubility is achieved by the formation of bisulphide or chloride ion complexes. Various mechanisms may cause local precipitation of these metal–ion complexes, resulting in a concentrated accumulation of metallic ore. A rapid drop in the solution temperature because of thermal conduction or mixing with cooler fluids, a change in solution pH, and boiling can all lead to ore deposition. Many types of ore deposits in the geological record, such as porphyry ore deposits associated with silicic volcanism, Mississippi Valley-type lead-zinc deposits, and volcanically hosted massive sulphide deposits are linked to hydrothermal activity. Such ore deposits thus present an integrated fossil record of hydrothermal activity, and provide a window into subsurface heat transfer and fluid flow processes. By coupling this integrated fossil record with studies of active oreforming processes on the seafloor and in other active hydrothermal environments, one can obtain a more complete picture of hydrothermal activity (see Mining Geology: Hydrothermal Ores). Temporal Variability in Hydrothermal Activity
Temporal variability on a range of time-scales is a fundamental characteristic of hydrothermal activity (Table 2). Some of this variability is linked to episodicity in magmatic and tectonic activity, or climate changes. The occurrence of these processes ranges from scales of plate reorganization of 106–107 years to magma replacement times of 101–104 years at fast and slow spreading ridges, respectively. Temporal variability related to the fluid circulation system, mainly resulting from changes in crustal permeability, occurs on time-scales 1–102 years. Seafloor hydrothermal activity is known to change on time-scales of hours to months following earthquakes, igneous intrusions (e.g., dykes), or volcanic eruptions. Climate changes may alter precipitation patterns, and hence fluid recharge, on time-scales of 10–103 years; ice ages and glaciation may affect high-altitude systems on similar time-scale.
Two-Phase Flow Boiling and phase separation commonly occur in high-temperature hydrothermal systems. For pure water, boiling is defined by the boiling point curve as a function of pressure. Liquid phase occurs below
370 TECTONICS/Hydrothermal Activity
Table 2 Time scale of events and processes related to hydrothermal activitya Time scale 6
10 106 105 101 103 101 103 103 101 100
7
10 years 107 years years 106 years 104 years 103 years 104 years 106 years 103 years 102 years
100 101 years (hours to decade) 100 101 years 104 105 105 105 103 0.1
107 s 107 s 107 s 106 s 106 s 3s
Activity or process
Plate reorganization Episodes of seafloor spreading Magnetic polarity interval Duration of ore formation processes Eruption cycle on slow spreading ridges Eruption cycle on fast spreading ridges Glacial episodes Duration of hydrothermal activity Episodes of climate change Duration of individual seafloor hydrothermal vent Duration of volcanic eruption
Residence time of hydrothermal fluid in oceanic crust Transit time of upwelling hydrothermal fluid Duration of earthquake swarms Duration of dyke emplacement event Duration of seafloor event plume Period of tidal signals Precipitation of sulphide particles during mixing of high T hydrothermal fluid with ambient seawater
a
Modified from Lowell RP, Rona PA, and Von Herzen RP (1995) and Rona (1988).
the boiling point and the vapour phase occurs above it. The two phases are in equilibrium along the curve, and the volume fractions of each phase along the curve depend upon the enthalpy of the system. The phase diagram for water (Figure 7) most readily shows these relations. A key feature of the pure water phase diagram is the critical point defined by Pc ¼ 218 bars, T c ¼ 374 C. Above the critical point only a single-phase ‘‘water substance’’ exists. Most hydrothermal aqueous fluids contain some amount of dissolved salts; however, the presence of these dramatically alters the two-phase behaviour of water. First of all, both Pc and T c increase as the salt content increases. For seawater, which can be represented by salinity x 3.2% NaCl solution, Pc 300 bars, T c 405 C. Secondly, two-phase flow is defined by a region of P-T-x space rather than a curve. Thirdly, because fluid density is a function of salinity, the fractionation of salt between the liquid and vapour phases affects the dynamics of the phases. Figure 8 depicts part of the phase diagram for NaCl– water solution. In terrestrial systems subsurface boiling results in either liquid- or vapour-dominated systems. In liquid-dominated systems, fluid pressures are near hydrostatic. Hot spring fluids are generally neutral to alkaline pH and chloride rich. Liquid and vapour phases are intermingled within the two-phase zone,
Figure 7 Boiling point curve for pure water compared with that for seawater. Note that the pure water curve ends at the critical point. The region above the pure water curve is pure vapour and the two phases only exist along the boiling curve. Reproduced from Bischoff and Rosenbauer (1984) The critical point and two phase boundary of seawater, 200 500 C. Earth and Planetary Sci ence Letters 68: 172 180.
and the two-phase zone overlies a single-phase fluid. By contrast, in vapour-dominated systems low, nearly uniform, vapour-static fluid pressure occurs over a considerable thickness. Fluid discharge occurs at low pH and low chloride. The presence of a region of underpressure implies a permeable barrier between the vapour-dominated zone and the surrounding cold recharge. Vapour-dominated systems act as a heat pipe with near zero net mass flux; heat is carried upwards by high enthalpy vapour while small amounts of low-enthalpy liquid flow downwards. Most systems are liquid-dominated, including the geyser basins of Yellowstone National Park, USA, Wairakei and Broadlands, NZ, and Ahuachapan, MX; vapourdominated systems include Geysers, USA, Lardarello, IT, Kamodjang, IND, and Matsukawa, JP (Figure 1). In submarine systems venting black smoker fluids are mostly in the liquid phase; however, the chlorinity of vent fluids seldom corresponds to seawater (540 mmol kg). Rather it ranges from 30 to 1200 mmol kg (Table 3). The departure of vent salinity from that of seawater is attributed to phase separation. The lowest chlorinity values often occur shortly after magmatic eruptions or diking events and thus are indicative of active phase separation.
TECTONICS/Hydrothermal Activity 371
Figure 8 Three dimensional perspective of the NaCl H2O phase diagram between 300 and 500 C. Reproduced from Bischoff and Pitzer (1989) Liquid vapor relations for the system NACL H2O: Summary of the P T x surface from 300 to 500 C. American Journal of Science 289: 217 248.
Table 3 Chlorinity of selected high temperature seafloor hydrothermal Ventsa
Vent site
East Pacific rise 9 10 N 21 N Juan de Fuca ridge North cleft South cleft Endeavour Axial volcano Mid Atlantic ridge TAG MARK Lau basin
Year(s) sampled
Value (mmol kg 1)b
1991 1979, 81, 85
32 860 489 579
1990 92 1984 1984 88 1986 88
730 896 253 176
1986 1986 1989
659 559 650 800
1245 1087 505 624
a
Modified from Von Damm (1985). Normal seawater chlorinity 540 mmol kg 1.
b
Phase separation also results in the formation of saline brines that, because of their high density, sink towards the base of the system. Later mixing of these saline brines with seawater may result in vent
chlorinity greater than seawater. The formation of a brine layer at the base of a hydrothermal system may act as a thermal conductive barrier between the overlying hydrothermal circulation and the magma body (Figure 3A) and be a salinity source for saline vent fluids.
Future Directions Hydrothermal activity represents an exciting dynamic area for future research. This is particularly true for submarine systems because of their links to studies of the origin of life, life in extreme environments, and the continued discovery of novel types of hydrothermal activity. The detailed sampling and data analysis and continued exploration for serpentinization-driven hydrothermal activity will likely grow during the next decade. At ridge crest and volcanic island arc systems, advances in ocean drilling technology, remote and autonomous sensing devices, long-term monitoring, integrated interdisciplinary experiments at various well-characterized seafloor
372 TECTONICS/Mid-Ocean Ridges
sites, and improvements in mathematical modelling techniques will stimulate the science over the next decade. Finally, we believe that analysis of seafloor hydrothermal activity over geological time, and attempts to discern the importance of hydrothermal activity elsewhere in the solar system will emerge as an important endeavour. Such studies are necessary to understand the links between hydrothermal activity and life. In terrestrial systems continued exploitation as an energy resource will be important. Moreover, climatically induced precipitation changes, because of the link to hydrothermal recharge, may alter warm spring and geyser behaviour. Utilization of hydrothermal activity as a climate monitor has yet to receive attention.
See Also Geysers and Hot Springs. Igneous Processes. Mining Geology: Hydrothermal Ores; Magmatic Ores. Origin of Life. Plate Tectonics. Tectonics: Faults.
Further Reading Anderson RN, Langseth MG Jr, and Sclater JG (1977) The mechanisms of heat transfer through the floor of the Indian Ocean. Journal of Geophysical Research 82: 3391 3409. Bischoff JL and Pitzer KS (1989) Liquid vapor relations for the system NACL H2O: Summary of the P T x surface from 300 to 500 C. American Journal of Science 289: 217 248. Bischoff JL and Rosenbauer RJ (1984) The critical point and two phase boundary of seawater, 200 500 C. Earth and Planetary Science Letters 68: 172 180. Craig H (1961) Standard for reporting concentrations of deuterium and oxygen 18 in natural waters. Science 133: 1833 1834. Elder J (1981) Geothermal Systems. San Diego, CA: Aca demic Press.
Elderfield H and Schultz A (1996) Mid ocean ridge hydro thermal fluxes and the chemical composition of the ocean. Annual Reviews of Earth and Planetary Science 24: 191 224. Germanovich LN, Lowell RP, and Astakhov DK (2000) Stress dependent permeability and the formation of sea floor event plumes. Journal of Geophysical Research 105: 8341 8354. Humphris SE, Zierenberg RA, Mullineaux LS, and Thompson RE (eds.) (1995) Seafloor Hydrothermal Systems: Physical, Chemical, and Biological Interactions, AGU Geophysical Monograph 91. Washington, DC: American Geophysical Union. Kelley DS, Baross JA, and Delaney JR (2002) Volcanoes, fluids and life at mid ocean ridge spreading centers. Annual Review of Earth and Planetary Science 30: 385 491. Kruger P and Otte C (eds.) (1973) Geothermal Energy. Stanford, CA: Stanford University Press. Lowell RP (1991) Modeling continental and submarine hydrothermal systems. Reviews of Geophysics 29: 457 476. Lowell RP (1992) Hydrothermal systems. In: Nierenburg WA (ed.) Encyclopedia of Earth System Science, vol. 2, pp. 547 557. San Diego, CA: Academic Press. Lowell RP, Rona PA, and Von Herzen RP (1995) Seafloor hydrothermal systems. Journal of Geophysical Research 100: 327 352. Rona PA (1988) Hydrothermal mineralization at oceanic ridges. Canadian Mineralogist 26: 431 465. Sclater JG, Jaupart C, and Galson D (1980) The heat flow through oceanic and continental crust and the heat loss of the Earth. Review of Geophysics 18: 269 311. Von Damm KL (1995) Controls on the chemistry and tem poral variability of seafloor hydrothermal fluids. In: Humphris SE, Zierenberg RA, Mullineaux LS, and Thompson RE (eds.) Seafloor Hydrothermal Systems: Physical, Chemical, and Biological Interactions, AGU Geophysical Monograph 91, pp. 222 247. Washington, DC: American Geophysical Union. White DE (1973) Characteristics of geothermal resources. In: Kruger P and Otte C (eds.) Geothermal Energy. Stan ford, CA: Stanford University Press.
Mid-Ocean Ridges K C Macdonald, University of California–Santa Barbara, Santa Barbara, CA, USA ß 2005, Elsevier Ltd. All Rights Reserved.
Introduction The mid-ocean ridge system is the largest mountain chain and the most active system of volcanoes in the solar system. In plate-tectonic theory, the ridge is
located between plates of the Earth’s rigid outer shell that are separating at speeds of approximately 10–170 mm year 1 (up to 220 mm year 1 in the past). The ascent of molten rock from deep within the Earth (ca. 30–60 km) to fill the void between the plates creates new seafloor and a volcanically active ridge. This ridge system wraps around the globe like the seam of a baseball and is approximately 70 000 km long (including the lengths of ridge offsets, such
TECTONICS/Mid-Ocean Ridges 373
as transform faults). Yet the ridge itself is only about 5–30 km wide, very small compared with the plates, which can be thousands of kilometres across (Figure 1). Early exploration showed that the gross morphology of spreading centres varies with the rate of plate separation. At slow spreading rates (10–40 mm year 1) a rift valley 1–3 km deep marks the axis, while for fast spreading rates (more than 90 mm year 1) the axis is characterized by an elevation of the seafloor of several hundred metres called an axial high (Figure 2). The rate of magma supply is a second factor that may influence the morphology of midocean ridges. For example, a very high rate of magma supply can produce an axial high even where the spreading rate is slow; the Reykjanes Ridge south of Iceland is a good example. Also, for intermediate spreading rates (40–90 mm year 1) the ridge crest may have either an axial high or a rift valley depending on the rate of magma supply. The depth to the seafloor increases from a global average of approximately 2600 m at the spreading centre to more than 5000 m beyond the ridge flanks. The rate of deepening is proportional to the square root of the age of the seafloor because it is caused by the thermal contraction of the lithosphere. Early mapping efforts also showed that the mid-ocean ridge is a discontinuous structure, which is offset at right angles to its length by
numerous transform faults that are tens to hundreds of kilometres long. Maps are powerful: they inform, excite, and stimulate. Just as the earliest maps of the world in the sixteenth century ushered in a vigorous age of exploration, so the first high-resolution continuous-coverage maps of the mid-ocean ridge system stimulated investigators from a wide range of fields, including petrologists, geochemists, volcanologists, seismologists, tectonicists, and practitioners of marine magnetics and gravity, as well as researchers outside the Earth sciences, including marine ecologists, chemists, and biochemists. Marine geologists have found that many of the most revealing variations are observed by exploring along the axis of the active ridge. This alongstrike perspective has revealed the architecture of the global rift system. The ridge axis undulates up and down in a systematic way, defining a fundamental partitioning of the ridge into segments bounded by a variety of discontinuities. These segments behave like giant cracks in the seafloor, which can lengthen or shorten and have episodes of increased volcanic and tectonic activity. In fact, elementary fracture mechanics can be used to explain the interaction between neighbouring ridge segments. Another important change in perspective came from the discovery of hydrothermal vents by marine geologists and geophysicists. It became clear that,
Figure 1 Shaded relief map of the seafloor showing parts of the East Pacific Rise, a fast spreading centre, and the Mid Atlantic Ridge, a slow spreading centre (courtesy of the National Geophysical Data Centre).
374 TECTONICS/Mid-Ocean Ridges
Figure 2 Topography of spreading centres. (A) Cross sections of typical fast intermediate and slow spreading ridges based on high resolution deep tow profiles. The neovolcanic zone (the zone of active volcanism) is indicated and is several kilometres wide; the zone of active faulting extends to the edge of the profiles and is several tens of kilometres wide. (Reproduced from Macdonald KC (1982) Mid ocean ridges: fine scale tectonic, volcanic and hydrothermal processes within the plate boundary zone. Annual Review of Earth and Planetary Sciences 10: 155 190.) EPR, East Pacific Rise; MAR, Mid Atlantic Ridge. (B) Shaded relief map of a 1000 km stretch of the East Pacific Rise extending from 8 N to 17 N. Here, the East Pacific Rise is the boundary between the Pacific and Cocos plates, which are separating at a ‘fast’ rate of 120 mm year 1. The map reveals two kinds of discontinuity: large offsets, about 100 km long, known as transform faults, and smaller offsets, about 10 km long, called overlapping spreading centres. Colours indicate depths of 2400 m (pink) to 3500 m (dark blue). (Reprinted from Encyclopedia of Ocean Sciences, Steele J, Thorpe S, and Turekian K (eds.), Macdonald KC, Mid ocean ridge tectonics, volcanism and geomorphology, pp. 1798 1813, Copyright (2001), with permission from Elsevier.) (C) Shaded relief map of the Mid Atlantic Ridge. Here, the ridge is the plate boundary between the South American and African plates, which are spreading apart at the slow rate of approximately 35 mm year 1. The axis of the ridge is marked by a 2 km deep rift valley, which is typical of most slow spreading ridges. The map reveals a 12 km jog of the rift valley, a second order discontinuity, and also shows a first order discontinuity called the Cox transform fault. Colours indicate depths of 1900 m (pink) to 4200 m (dark blue). (Reprinted from Encyclopedia of Ocean Sciences, Steele J, Thorpe S, and Turekian K (eds.), Macdonald KC, Mid ocean ridge tectonics, volcanism and geomorphology, pp. 1798 1813, Copyright (2001), with permission from Elsevier.)
in studies of mid-ocean ridge tectonics, volcanism, and hydrothermal activity, the greatest excitement is in the linkages between these different fields. For example, geophysicists searched for hydrothermal activity at mid-ocean ridges for many years by towing arrays of thermistors near the seafloor. However, hydrothermal activity was eventually documented more effectively by photographing the distribution
of exotic vent animals. Even now, the best indicators of the recency of volcanic eruptions and the duration of hydrothermal activity are found by studying the characteristics of benthic faunal communities. For example, during the first deep-sea mid-ocean-ridge eruption witnessed from a submersible, divers did not see a slow lumbering cascade of pillow lavas, as observed by divers off the coast of Hawaii. What they
TECTONICS/Mid-Ocean Ridges 375
saw was completely unexpected: white bacterial matting billowing out of the seafloor, creating a scene much like a mid-winter blizzard in Iceland, covering the freshly erupted glassy black lava with a thick blanket of white bacterial ‘snow’.
Ridge Segmentation The most recognizable segmentation of mid-ocean ridges is that defined by transform faults. These plate boundaries are usually perpendicular to the ridge segments they offset and are tens to hundreds of kilometres long, although some exceed 1000 km in length (e.g. the Romanche and San Andreas faults). In plate tectonics, a transform fault traces a small circle about the Euler pole of opening between any pair of plates. Thus the transform fault and its off-axis fracture zone traces may be used to determine the pole of opening as well as changes in the pole of opening. At a ridge–transform intersection, normal spreading processes are truncated. Normal faulting predominates on mid-ocean ridges, while strike-slip faulting dominates along transform faults. The transition can be very complex, with normal faults and strike-slip faults occurring along trends that are affected by shear stresses on the transform fault. Crustal accretionary processes are also affected by the juxtaposition of thick cold lithosphere against the end of a spreading segment. This effect increases with the age and thickness of the lithosphere that is sliding past the ridge–transform intersection. Transverse ridges occur along the length of some of the largest transform faults; some of these ridges have been elevated above sea-level for part of their history. Between major transform faults, the axial depth profile of mid-ocean ridges undulates up and down with a wavelength of tens of kilometres and an amplitude of tens to hundreds of metres at fast-spreading and intermediate-spreading ridges. This pattern is also observed for slow-spreading ridges, but the wavelength of undulation is shorter and the amplitude is larger (Figure 3). In most cases, ridge-axis discontinuities occur at local maxima of the axial depth profile. These discontinuities include transform faults, as discussed above (first order), overlapping spreading centres (second order), and higher order (third and fourth order) discontinuities, which are increasingly short-lived, mobile, and associated with smaller offsets of the ridge (see Table 1 and Figure 4). A much-debated hypothesis is that the axial depth profile (Figures 3 and 5) reflects the magma supply along a ridge segment. According to this idea, the magma supply is enhanced along shallow portions of ridge segments and is relatively starved at segment ends (discontinuities). In support of this hypothesis
Figure 3 Axial depth profiles for (A) slow spreading, (B) fast spreading, and (C) superfast spreading ridges. Discontinuities of orders 1 and 2 typically occur at local depth maxima (discontinu ities of orders 3 and 4 are not labelled here). The segments at faster spreading ridges are longer and have smoother lower amplitude axial depth profiles. These depth variations may reflect the pattern of mantle upwelling. (Reprinted from Encyclopedia of Ocean Sciences, Steele J, Thorpe S, and Turekian K (eds.), Macdonald KC, Mid ocean ridge tectonics, volcanism and geo morphology, pp. 1798 1813, Copyright (2001), with permission from Elsevier.)
is the observation at ridges with an axial high (fastspreading ridges) that the cross-sectional area or axial volume varies directly with depth (Figure 6). Maxima in the cross-sectional area (more than 2.5 km2) occur at minima along the axial depth profile (generally not near ridge-axis discontinuities) and are thought to correlate with regions where magma supply is robust. Conversely, small cross-sectional areas (less than 1.5 km2) occur at local depth maxima and are interpreted to reflect minima in the magma-supply rate along a given ridge segment. On slow-spreading ridges characterized by an axial rift valley, the crosssectional area of the valley is at a minimum in the midsegment regions, where the depth is at a minimum. In addition, there are more volcanoes in the shallow midsegment area, and fewer volcanoes near the segment ends. Studies of crustal magnetization show that very highly magnetized zones occur near segment ends; these are most easily explained by a local restriction of magma supply resulting in the eruption of highly fractionated lavas that are rich in iron. Multichannel seismic and gravity data support the axial volume–magma supply–segmentation hypothesis (Figure 6). A bright reflector, which is phasereversed in many places, occurs commonly (>60%
376 TECTONICS/Mid-Ocean Ridges
Table 1 Characteristics of segmentation, updated from Macdonald KC, Scheirer DS, and Carbotte SM (1991) Mid ocean ridges: discontinuities, segments and giant cracks. Science 253: 986 994 (see references therein). This four tiered hierarchy of segmentation probably represents a continuum in segmentation Order 1
Order 2
Order 3
Order 4
Segment longevity (years)
600 300a (400 200)b >5 106
Rate of segment lengthening (long term migration) (mm y 1) Rate of segment lengthening (short term propagation) (mm y 1)
0 50 (0 30) 0 100 (?)
140 90 (50 30) 0.5 5 106 (0.5 30 106) 0 1000 (0 30) 0 1000 (0 50)
20 10 (15 10?) 104 105 (?) Indeterminate: no off axis trace Indeterminate: no off axis trace
75 (7 5?) 30 >0.5 106 (>2 106) 300 600 (500 2000) Fracture zone
Devals, offsets of axial summit caldera (intravolcano gaps)