65,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
Invertebrate Palaeontology and Evolution is well established as the foremost palaeontology text at the undergraduate level. This fully revised fourth edition includes a complete update of the sections on evolution and the fossil record, and the evolution of the early metazoans.
- New work on the classification of the major phyla (in particular brachiopods and molluscs) has been incorporated.
- The section on trace fossils is extensively rewritten.
- The author has taken care to involve specialists in the major groups, to ensure the taxonomy is as up-to-date and accurate as possible.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 1201
Veröffentlichungsjahr: 2013
Ähnliche
Contents
Preface
PaleoBase—Macrofossils on CD-ROM
PART ONE: General Palaeontological Concepts
1 Principles of palaeontology
1.1 Introduction
1.2 Occurrence of invertebrate fossils in Phanerozoic rocks
1.3 Divisions of invertebrate palaeontology
2 Evolution and the fossil record
2.1 Introduction
2.2 Darwin, the species and natural selection
2.3 Fossil record and modes of evolution
2.4 Competition and its effects
2.5 Summary of palaeontological evolution theory
3 Major events in the history of life
3.1 Introduction
3.2 Prokaryotes and eukaryotes
3.3 Earliest metazoans
3.4 Major features of the Phanerozoic record
PART TWO: Invertebrate Phyla
4 Sponges
4.1 Phylum Porifera: sponges
4.2 Classification
4.3 Class Demospongea
4.4 Class Calcarea
4.5 Class Hexactinellida
4.6 Incertae sedis: Archaeocyatha
4.7 Geological importance of sponges
4.8 Sponge reefs
5 Cnidarians
5.1 Introduction
5.2 Major characteristics and classes of Phylum Cnidaria
5.3 Class Hydrozoa
5.4 Class Scyphozoa
5.5 Class Anthozoa
6 Bryozoans
6.1 Introduction
6.2 Two examples of living bryozoans (Figs 6.1-6.5)
6.3 Classification
6.4 Morphology and evolution
6.5 Ecology and distribution
6.6 Stratigraphical use
7 Brachiopods
7.1 Introduction
7.2 Morphology
7.3 Ontogeny
7.4 Classification (Fig. 7.16)
7.5 Evolutionary history
7.6 Ecology and distribution
7.7 Faunal provinces
7.8 Stratigraphical use
8 Molluscs
8.1 Fundamental organization
8.2 Classification
8.3 Some aspects of shell morphology and growth
8.4 Principal fossil groups
8.5 Predation and the evolution of molluscs
9 Echinoderms
9.1 Introduction
9.2 Classification
9.3 Subphylum Echinozoa
9.4 Subphylum Asterozoa
9.5 Subphylum Crinozoa
9.6 Subphylum Blastozoa
9.7 Subphylum Homalozoa, otherwise calcichordates (Fig. 9.49)
9.8 Evolution
10 Graptolites
10.1 Structure
10.2 Classification
10.3 Biological affinities
10.4 Evolution
10.5 How did graptolites live?
10.6 Faunal provinces
10.7 Stratigraphical use
11 Arthropods
11.1 Introduction
11.2 Classification and general morphology
11.3 Trilobita
11.4 Phylum Chelicerata
11.5 Phylum Crustacea
12 Exceptional faunas; ichnology
12.1 Introduction
12.2 Burgess Shale fauna (Figs 12.1, 12.4)
12.3 Upper Cambrian of southern Sweden (Figs 12.6, 12.7)
12.4 Hunsrückschiefer fauna
12.5 Mazon Creek fauna
12.6 Solnhofen lithographic limestone, Bavaria (Fig 12.10)
12.7 Ichnology
Systematic index
General index
© 1979,1986,1993,1998 by E. N. K. Clarkson
Published by Blackwell Science Ltd,
a Blackwell Publishing company
BLACKWELL PUBLISHING
350 Main Street, Malden, MA 02148-5020, USA
9600 Garsington Road, Oxford OX4 2DQ, UK
550 Swanston Street, Carlton, Victoria 3053, Australia
The right of E. N. K. Clarkson to be identified as the Author of this Work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
First published 1979 by Unwin Hyman Ltd
Second edition 1986
Third edition 1993 by Chapman & Hall
Fourth edition 1998 by Blackwell Science Ltd
10 2008
Library of Congress Cataloging-in-Publication Data has been applied for
ISBN 978-0-632-05238-7
A catalogue record for this title is available from the British Library.
The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards.
For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com
In memory of
Professor Peter Sylvester-Bradley
(1913–1978)
Preface
The first three editions of this textbook were published in 1979, 1986 and 1992, and I trust that this new one, necessitated by so many advances both in fact and theory, will retain its function as a course text for students of palaeontology from their second year onwards. I have made substantial changes to Chapters 1–3, 7, 8 and 12, and the Trace Fossils section has been transferred from Chapter 1 to the last chapter. All the other chapters have been revised to a greater or lesser extent. I have redrawn about half of the illustrations: a singularly congenial occupation for long winter evenings, and I hope that these will prove of value in helping students to understand the anatomy and terminology of fossil invertebrates.
Key words appear in bold at their first mention in the text. There is no specific index for them, but they appear in the General Index.
As before, I wish to thank all my friends, colleagues and family who have given me such assistance in preparing this new edition. Derek Briggs, Colin Scrutton and Rachel Wood gave excellent advice at the beginning of the project as to how I should proceed with the fourth edition, and I have taken on board most of their suggestions. As always, these colleagues, together with David Harper, Peter Sheldon, Susan Rigby, Liz Harper, Dick Jefferies, John Cope, Alan Owen and various others have helped me at all stages, and so have many others, too numerous to mention, in my own country and abroad, and my thanks are due to all.
I would like, above all, to record my heartfelt thanks to my true and stalwart friend Cecilia Taylor, for all the support and helpful suggestions she has given me at every stage. Had Cecilia not undertaken the vast job of rebuilding our teaching collections and preparing new course materials, I would never have had the time to complete this book before the deadline.
To Roisin Moran of University College, Galway I extend my grateful thanks for her two beautiful paintings for the Part Title pages.
I wish to thank also my editorial colleagues, Dr Ian Francis and Jane Plowman, who have given all possible assistance. From Ian came the suggestion, which we have followed up, that the new edition should be coupled with a CD-ROM of Fossil Images. This has been undertaken in conjunction with the Natural History Museum, London, and to Norman McLeod and Paul Taylor of that institution I owe a great deal. I am likewise grateful to David Hicks and Eve Daintith for their meticulous copy-editing and proofreading, respectively.
On a wet November day, over 50 years ago, I was taken into a museum in my native Newcastle-upon-Tyne, to escape from the rain. There, in a dusty glass case were two giant ammonites (probably Titanites giganteus), and I was given to understand that they had lived millions of years ago. They excited a fascination which has continued until now, and I am still glad to escape to the hills on a fresh summer morning, to search for fossils, and then to bring them back to the laboratory for study. I hope that this book will be of value to any student wishing to explore something of the richness and diversity of ancient life, and of the methods available for its scientific study. If so, I will have achieved what I set out to do.
Euan Clarkson
Edinburgh
PaleoBase—Macrofossils on CD-ROM
The Natural History Museum London and Dr E.N.K. Clarkson have collaborated on the development of this important new initiative in paleontology teaching.
PaleoBase is a combined image library and database, containing records for 1000 key fossil genera. Each record contains a set of images, and information on stratigraphic range, hard-part mineralogy, palaeoecology, palaeobiogeography, etc. The images have been captured using the Natural History Museum’s high resolution PALAEOVISION digital imaging system, and the data and images reside in the CompuStrat database manager.
This system allows the user great flexibility in displaying data. For example, users can simply browse from record to record; pull up fossils from a taxonomic index; or select and sort records by geographic range, life habit, stratigraphic range, or by a variety of other criteria. Range charts and palaeobiographic maps are given where appropriate, and the user can print records, or export them into other applications. Within each of the major groups there are a number of labelled images and diagrams to allow the user to become familiiar with key morphological terms.
PaleoBase may be used as an adjunct to Invertebrate Palaeontology, or as a standalone product. Taxa have been selected for their relevance to Earth science teaching worldwide.
PC or MAC, minimum 8 Mb RAM, CD-ROM drive, 640 × 480 colour monitor.
For further information and ordering details, please contact: [email protected]
PART ONE
General Palaeontological Concepts
Arnioceras cf. hartmanni (Oppel), an assemblage of immature ammonites from the Lower Jurassic of Black Ven, Charmouth, England (Lower Sinemurian). These specimens were probably catastrophically buried, since the soft parts must have been in place when they died preventing sediment penetrating the chambers. Painting by Raisin Moran; original specimen in the James Mitchell Museum, University College, Galway, Ireland.
1
Principles of palaeontology
1.1 Introduction
Once upon a time …
Some 4600 million years ago the Earth came into being, probably forming from a condensing disc of particles, dust and gas, which slowly rotated round the Sun. Larger particles, or planetismals, formed from this nebular disc, and as these collided they accreted, eventually forming the planets.
Of all the nine planets in the Solar System only Earth, as far as is known, supports advanced life, though at the time of writing much interest has been generated by the discovery of organic material on Jupiter’s satellite Europa. It is, however, a striking fact that life on Earth began very early indeed, within the first 30% of the planet’s history. There are remains of simple organisms (bacteria and ‘blue-green algae’, or cyanobacteria) in rocks 3400 Ma old, so life presumably originated before then. These simple forms of life seem to have dominated the scene for the next 2000 Ma, and evolution at that time was very slow. Nevertheless, the cyanobacteria and photosynthetic bacteria were instrumental in changing the environment, for they gave off oxygen into an atmosphere that was previously devoid of it, so that animal life eventually became possible.
Only when some of the early living beings of this Earth had reached a high level of physiological and reproductive organization (and most particularly when sexual reproduction originated) was the rate of evolutionary change accelerated, and with it all manner of new possibilities were opened up to evolving life. This was not until comparatively late in geological history, and there are no fossil animals known from sediments older than about 700 Ma. Needless to say, these are all invertebrate animals lacking backbones. All of them are marine; there is no record of terrestrial animals until much later. In terms of our understanding of the history of life, perhaps the most significant of all events took place about 543 Ma ago at about the beginning of the Cambrian Period, for at this stage there was a sudden proliferation of different kinds of marine invertebrates. During this critical period the principal invertebrate groups were established, and they then diversified and expanded. Some of these organisms acquired hard shells and were capable of being fossilized, and only because of this can there be any chance of understanding the history of invertebrate life.
The stratified sedimentary rocks laid down since the early Cambrian, and built up throughout the whole of Phanerozoic time, are distinguished by a rich heritage of the fossil remains of the invertebrates that evolved through successive historical periods; their study is the domain of invertebrate palaeontology and the subject of this book.
1.2 Occurrence of invertebrate fossils in Phanerozoic rocks
Hard-part preservation
Fossil invertebrates occur in many kinds of sedimentary rock deposited in the seas during the Phanerozoic. They may be very abundant in limestones, shales, siltstones and mudstones but on the whole are not common in sandstones. Sedimentary ironstones may have rich fossil remains. Occasionally they are found in some coarse rocks such as greywackes and even conglomerates. The state of preservation of fossils varies greatly, depending on the structure and composition of the original shell, the nature and grain size of the enclosing sediment, the chemical conditions at the time of sedimentation, and the subsequent processes of diagenesis (chemical and physical changes) taking place in the rock after deposition.
The study of the processes leading to fossilization is known as taphonomy. In most cases only the hard parts of fossil animals are preserved, and for these to be fossilized, rapid burial is normally a prerequisite. The soft-bodied elements in the fauna, and those forms with thin organic shells, did not normally survive diagenesis and hence have left little or no evidence of their existence other than records of their activity in the form of trace fossils. What we can see in the rocks is therefore only a narrow band in a whole spectrum of the organisms that were once living; only very rarely have there been found beds containing some or all of the soft-bodied elements in the fauna as well. These are immensely significant for palaeontology.
The oldest such fauna is of late Precambrian age, some 615 Ma old, and is our only record of animal life before the Cambrian. Another such ‘window’ is known in Middle Cambrian rocks from British Columbia, where in addition to the normally expected trilobites and brachiopods there is a great range of soft-bodied and thin-shelled animals – sponges, worms, jellyfish, small shrimp-like creatures and animals of quite unknown affinities – which are the only trace of a diverse fauna which would otherwise be quite unknown (Chapter 12). There are similar ‘windows’ at other levels in the geological column, likewise illuminating.
The fossil record is, as a guide to the evolution of ancient life, unquestionably limited, patchy and incomplete. Usually only the hard-shelled elements in the biota (apart from trace fossils) are preserved, and the fossil assemblages present in the rock may have been transported some distance, abraded, damaged and mixed with elements of other faunas. Even if thick-shelled animals were originally present in a fauna, they may not be preserved; in sandy sediments in which the circulating waters are acidic, for instance, calcareous shells may dissolve within a few years before the sediment is compacted into rock. Since the sea floor is not always a region of continuous sediment deposition, many apparently continuous sedimentary sequences contain numerous small-scale breaks (diastems) representing periods of winnowing and erosion. Any shells on the sea floor during these erosion periods would probably be transported or destroyed – another limitation on the adequacy of the fossil record.
On the other hand, some marine invertebrates found in certain rock types have been preserved abundantly and in exquisite detail, so that it is possible to infer much about their biology from their remains. Many of the best-preserved fossils come from limestones or from silty sediments with a high calcareous content. In these (Fig. 1.1) the original calcareous shells may be retained in the fossil state with relatively little alteration, depending upon the chemical conditions within the sediment at the time of deposition and after.
A sediment often consists of components derived from various environments, and when all of these, including decaying organisms, dead shells and sedimentary particles, are thrown together the chemical balance is unstable. The sediment will be in chemical equilibrium only after diagenetic physicochemical alterations have taken place. These may involve recrystallization and the growth of new minerals (authigenesis) as well as cementation and compaction of the rock (lithification), and during any one of these processes the fossils may be altered or destroyed. Shells that are originally of calcite preserve best; aragonite is a less stable form of calcium carbonate secreted by certain living organisms (e.g. corals) and is often recrystallized to calcite during diagenesis or dissolved away completely.
Calcareous skeletons preserved in more sandy or silty sediments may dissolve after the sediment has hardened or during weathering of the rock long after its induration. Moulds (often miscalled casts) of the external and internal surfaces of the fossil may be left, and if the sediment is fine enough the details these show may be very good. Some methods for the study of such moulds are described in section 7.2, with reference to brachiopods. If a fossil encloses an originally hollow space, as for instance between the pair of shells of a bivalve or brachiopod, this space may either be left empty or become filled with sediment. In the latter case a sediment core is preserved, which comes out intact when the rock is cracked open. This bears upon its surface an internal mould of the fossil shell, whereas external moulds are left in the cavity from which it came. In rare circumstances the core or the shell, or both, may be replaced by an entirely different mineral, as happens in fossils preserved in ironstones. If the original spaces in the shell are impregnated with extra minerals, it is said to be permineralized, while the growth of secondary minerals at the expense of the shell is replacement. Cores may sometimes be of pyrite. Graptolites are often preserved like this, anaeorbic decay having released hydrogen sulphide, which reacted with ferrous (Fe2+) ions in the water to allow an internal pyrite core to form. Sometimes a core of silica is found within an unaltered calcite shell. This has happened with some of the Cretaceous sea urchins of southern England. They lived in or on a sediment of calcareous mud along with many sponges, which secreted spicules of biogenic silica as a skeleton. In alkaline conditions (above pH 9), which may sometimes be generated during bacterial decay, the solubility of the silica increases markedly, and the silica so released will travel through the rock and precipitate wherever the pH is lower. The inside of a sea urchin decaying under different conditions would trap just such an internal microenvironment, within which the silica could precipitate as a gel. Such siliceous cores retain excellent features, preserving the internal morphology of the shell. On the other hand, silica may replace calcite as a very thin shell over the surface of a fossil as a result of some complex surface reactions. These siliceous crusts may retain a very detailed expression of the surface of the fossil and, since they can be treed from the rock by dissolving the limestone with hydrochloric acid, individual small fossils preserved in this way can be studied in three dimensions. Some of the most exquisite of all trilobites and brachiopods are known from material such as this.
Figure 1.1 Possible processes of fossilization of a bivalve shell: (a) original shell, buried in mud (left) or carbonate (right); (b) the shell was calcite, was buried in a carbonate sediment and was preserved intact other than as a small crystallized patch; (c) shell originally of aragonite, now recrystallized to calcite which destroys the fine structure; (d) original calcite shell retained surrounding a diagenetic core of silica; (e) a silica rim growing on the outside of the shell; (f) tectonic distortion of a shell preserved in mudstone; (g) shell preserved in mud with original shell material leached away, leaving an external and an internal mould, surrounding a mudstone core; (h) a calcareous concretion growing round the shell and inside (if the original cavity was empty), with patches of pyrite in places; (if ironstone replacement of core and part of shell.
A relatively uncommon but exquisite mode of preservation is phosphatization. Sometimes the external skeleton, especially of thin organic-shelled animals, may be replaced or overgrown by a thin sheet of phosphate, or the latter may reinforce an originally phosphatic shell. In the former situation the external form of the body is precisely replicated. Alternatively a phosphatic filling of the interior of the shell may form a core, picking out internal structures in remarkable detail. Such preservation is probably associated with bacterial activity directly after the death of the animal. Many small Cambrian fossils have been preserved by phosphatization (Chapters 3 and 12), but much larger fossils may be preserved also, for example crustaceans with a fluorapatite infilling and with all their delicate appendages intact.
Fossils are often found in concretions: calcareous or siliceous masses formed around the fossil shortly after its death and burial. Concretions form under certain conditions only, where a delicate chemical balance exists between the water and sediment, by processes as yet not fully understood.
Soft-part preservation
In very rare circumstances soft-bodied organisms can be preserved as fossils, and these provide otherwise unobtainable evidence of the diversity of metazoans living at particular periods; this is discussed in Chapter 12.
1.3 Divisions of invertebrate palaeontology
Invertebrate palaeontology is normally studied as a subdivision of geology, as it is within Earth science that its greatest applications lie. It can also be seen as a biological subject, but one that has the unique perspective of geological time. Within the domain of invertebrate palaeontology there are a number of interrelated topics (Fig. 1.2), all of which have a bearing on the others and which also link up with other sciences.
Three main categories of fossils may be distinguished: (1) body fossils, in other words the actual remains of some part, usually a shell of skeleton, of a once-living organism; (2) trace fossils, which are tracks, trails, burrows or other evidence of the activity of an animal of former times – sometimes these are the only guide to the former presence of soft-bodied animals in a particular environment; (3) chemical fossils, relics of biogenic organic compounds which may be detected geochemically in the rocks.
At the heart of invertebrate palaeontology stands taxonomy; the classification of fossil and modern animals into ordered and natural groupings. These groupings, known as taxa, must be named and arranged in a hierarchial system in which their relationships are made clear, and as far as possible must be seen in evolutionary perspective.
Evolution theory is compounded of various disciplines – pure biology, comparative anatomy, embryology, genetics and population biology – but it is only the palaeontological aspect that allows the predictions of evolutionary science to be tested against the background of geological time, permits the tracing of evolving lineages and illustrates some of the patterns of evolution that actually have occurred.
The rates at which animals have evolved have varied through time, but most animal types (species) have had a geological life of only a few million years. Some of these evolved rapidly, such as the ammonites, others very slowly (Chapter 2). A rock succession of marine sediments built up over many millions of years may therefore have several fossil species occurring in a particular sequence, each species confined to one part of the succession only and representing the time when that species was living. Herein lies the oldest and most general application of invertebrate palaeontology: biostratigraphy. Using the sequence of fossil faunas, the geological column has been divided up into a series of major geological time units (periods), each of which is further divided into a hierarchy of small units. The whole basis for this historical chronology is the documented sequence of fossils in the rocks. But different kinds of fossils have different stratigraphical values, and certain parts of the geological record are more closely subdivided than others. Some ‘absolute’ ages based on radiometric dating have been fixed at particular points to this relative scale, and these provide a framework for understanding the geological record in terms of real time (i.e. known periods of millions of years) rather than as just a purely relative scale. This is only possible at certain horizons, however, and for practical purposes the geological timescale based on biostratigraphy is unsurpassed for Phanerozoic sediments.
Figure 1.2 The various subdisciplines of palaeontology.
Although stratigraphy is the basis of the primary discipline of geochronology, a small facet of palaeontological study has a bearing on what may be termed ‘geochronometry’. By counting daily growth rings in extinct corals and bivalves, information has been obtained bearing upon the number of days in the lunar month and year in ancient times. This has helped to confirm geophysical estimates on the slowing of the Earth’s rotation (Chapter 5).
Since stratigraphical applications of palaeontology have always been so important, the more biological aspects of palaeontology were relatively neglected until comparatively recently. Palaeoecology, which has developed particularly since the early 1950s, is concerned with the relationships of fossil animals to their environment, both as individuals (autecology) and in the faunal communities in which they occur naturally; the latter is sometimes known as synecology.
Since the soft parts of fossil animals are not normally preserved, but only their hard shells, there are relatively few ways in which their biology and life habits can be understood. Studies in functional morphology, however, which deal with the interpretation of the biology of fossilized skeletons or structures in terms of their original function, have been successfully attempted with many kinds of fossils, restricted in scope though these endeavours may necessarily be. Ichnology is the study of trace fossils: the tracks, surface trails, burrows and borings made by once-living animals and preserved in sediments. This topic has proved valuable both in understanding the behaviour of the animals that lived when the sediment was being deposited and in interpreting the contemporaneous environment. Finally, it is only by the integration of taxonomic data on local faunas that the global distribution of marine invertebrates through time can be elucidated. Such studies of palaeobiogeography (or palaeozoogeography in the case of animals) can be used in conjunction with geophysical data in understanding the former relative positions and movements of continental masses.
All of these aspects of palaeontology are interrelated, and an advance in one may have a bearing upon any other. Thus a particular study in functional morphology may give information on palaeoecology and possibly some feedback to taxonomy as well. Likewise, recent refinements in taxonomic practice have enabled the development of a much more precise stratigraphy.
Chemical compounds of biological origin can now be recovered from ancient rocks and form the basis of biomolecular palaeontology. Such fossil molecules may help to diagnose which organisms they come from and their breakdown pathways may say something about the environment. Molecular phylogenetics based upon protein sequencing may show how far two or more related organisms have diverged from a common ancestor, and to some extent the available techniques can be applied to the recent fossil record. Immunological-determinant techniques can be used to detect proteins and polysaccharides in fossil shells but, for the moment, only shells younger than 2 Ma have proved amenable to analysis. There are also promising developments also in palaeobiochemistry and organic geochemistry which are applicable to the fossil record, though these are beyond the scope of this book.
Taxonomy
Taxonomy is often undervalued as a glorified form of filing – with each species in its folder, like a stamp in its prescribed place in an album; but taxonomy is a fundamental and dynamic science, dedicated to exploring the causes of relationships and similarities among organisms. Classifications are theories about the basis of natural order, not dull catalogues compiled only to avoid chaos.
The best monographs are works of genius … (S.j. Gould, 1990)
These words should make entirely clear the fundamental nature of taxonomy. For, as has often been said, to identify a fossil correctly is the first step, and indeed the key, to finding out further information about it. Sound classification and nomenclature lie at the root of all biological and palaeontological work; without them no coherent and ordered system of data storage and retrieval is possible. Taxonomy, or systematics as it is sometimes known, is the science of classification or organisms. It is the oldest of all the biological disciplines, and the principles outlined by Carl Gustav Linnaeus (1707–1778) in his pioneering Systema naturae are still in use today, though greatly modified and extended.
The species concept
The fundamental unit of taxonomy is the species. Animal species (e.g. Sylvester-Bradley, 1956) are groups of individuals that generally look like each other and can interbreed to produce offspring of the same kind. They cannot interbreed with other species. Since it is reproductive isolation alone that defines species it is only really possible to distinguish closely related species if their breeding habits are known. Of all the described ‘species’ of living animals, however, only about a sixth are ‘good’, or properly defined, species. Information upon the reproductive preferences of the other five-sixths of all naturally occurring animal populations is just not documented.
The differentiation of most living and all fossil species therefore has to be based upon other and technically less valid criteria.
Of these by far the most important, especially in palaeontology, is morphology, the science of form, since most natural species tend to be composed of individuals of similar enough external appearance to be identifiable as of the same kind. Distinguishing species of living animals by morphological criteria alone is not without hazards, especially where the species in question are similar and closely related. Supplementary information, such as the analysis of species-specific proteins, may be of help in some cases where there is good reason for it to be sought (e.g. for disease-carrying insects). For the rest some degree of subjectivity in taxonomy has to be accepted, though this can be minimized if enough morphological criteria are used.
Nomenclature and identification of fossil species
In the formal nomenclature of any species, living or fossil, taxonomists follow the biological system of Linnaeus, whereby each species is defined by two names: the generic and specific (or trivial) names. For example, all cats, large and small, are related, and one particular group has been placed in the genusFelis. Of the various species of Felis the specific names F. catus, F. leo and F. pardus formally refer to the domestic cat, lion and leopard, respectively. In full taxonomic nomenclature the author’s name and the date of publication are given after the species, e.g. Felis catus (Linnaeus 1778) (see below for further discussion).
In palaeontology it can never be known for certain whether a population with a particular morphology was reproductively isolated or not. Hence the definition of species in palaeontology, as in most living specimens, must be based almost entirely on morphological criteria. Moreover, only the hard parts of the fossil animal are preserved, and much useful data has vanished. A careful examination and documentation of all the anatomical features of the fossil has to be the main guide in establishing that one species is different from a related species. In rare cases this can be supplemented by a comparison of the chemistry of the shell, as has proved especially useful in the erection of higher taxonomic categories. Within any interbreeding population there is usually quite a spread of morphological variation. On a broader scale there may be both geographical and stratigraphic variations, and all these must be carefully documented if the species is to be ideally established. Such studies may be very significant in evolutionary palaeontology.
When a palaeontologist is attempting to distinguish the species in a newly discovered fauna, say of fossil brachiopods, she or he has to separate the individual fossils out into groups of morphologically similar individuals. There may, to take an example, perhaps be eight such groups, each distinguished by a particular set of characters. Some of these groups may be clearly distinct from one another; in others the distinction may be considerably less, increasing the risk of greater subjectivity. These groups are provisionally considered as species, which must then be identified. This is done by consulting palaeontological monographs or papers containing detailed technical descriptions and illustrations of previously described brachiopod faunas of similar age, and comparing the species point by point. Some of the species may prove to be identical with already described species, or show only minor variation of a kind that would be expected in a local variant within the same species. Other species in the fauna may be new, and if so a full technical description with illustrations must be prepared for each new species, which should be published in a palaeontological journal or monograph. This description is based upon type specimens, which are always thereafter kept in a museum or research institute. Usually one of these, the holotype, is selected as the reference specimen and fully illustrated; comparative detail may be added from other specimens called the paratypes. There are various other kinds of type specimens; for example, a neotype may be erected when a holotype has been lost, or when a species is being redescribed in fuller and more up-to-date terms when no type specimen has previously been designated.
A new genus will be designated as gen. nov. by its author, this following the generic name, a new species as sp. nov. and a new subspecies as ssp. nov.
The new species must be named and allocated to an existing genus, or if there is no described genus to which it pertains then a new genus must also be erected. To appreciate the method let us consider the following historical tale. In the early nineteenth century brachiopods were poorly known and few distinct genera had been erected. One of these was the living Terebratula, named by O.F. Müller in 1776. When E.F. von Schlotheim first studied Devonian fossils from North Germany in 1820 he recognized that some of the shells were brachiopods, and he described one of the most abundant forms as the new species Terebratula sarcinulatus. By 1830, however, much more was known about brachiopods, and G. Fischer de Waldheim proposed a new genus for this species, so that it became correctly designated Chonetes sarcinulattts (Schlotheim). This is the ‘type species’ of Chonetes, a well-known Siluro-Devonian genus of the Class Strophomenata. Note that where a species was originally described under a different generic name, the original author’s name is quoted in parentheses. In 1917 F.R. Cowper Reed, then working on Ordovician and Silurian brachiopods of the Girvan district, Scotland, recognized many new species. One of these had similarities in morphology to Chonetes, but it was sufficiently different to be regarded as a species of a new subgenus; this is written Chonetes (Eochonetes) advena Reed 1917. When in 1928 the taxonomic problems of Chonetes and similar forms were addressed by O.T. Jones, the new Superfamily Plectambonitacea was erected to accommodate advena and many other related brachiopods. At a later stage Eochonetes was elevated to the rank of a full genus. In the most recent treatment, D.A.T. Harper described a large fauna from the Girvan district of Scotland, and within this he recognized two subspecies of E. advena, of somewhat different ages and distinguished by minor differences in morphology. These, in Harper’s (1989) monograph, are written Eochonete advena advena Reed 1917 (designating Reed’s original material), and Eochonetes advena Reed 1917 gracilis ssp. nov. Subsequent authors will refer to the latter, in full, as Eochonetes advena gracilis Harper 1989, or in abbreviated form as E. advena gracilis.
Where, due to indifferent preservation, or lack of an up-to-date monographic base, a species cannot be identified with certainty, it may be designated as aff. (related to) or cf. (may be compared with), an existing species (e.g. Monograptus cf. vomerinus). Where the fossil can be identified as belonging to a known genus, but cannot be ascribed to a species (as may be the case where preservation is poor or if only a fragment is preserved, the suffix sp. (plural spp.) is used (e.g. Caloceras sp.). If the specimen, in a more extreme case, can only tentatively be referred to a genus, one would write e.g.? Kutorgina sp. If only the species is dubious such an ascription as Eoplectodonta ?penkillensis might be used.
All taxonomic work such as this must follow a particular set of rules, which have been worked out by a series of International Commissions and are documented in full in the opening pages of each volume of the Treatise on Invertebrate Paleontology (a continuing series of volumes published by the Geological Society of America).
Taxonomic hierarchy
Although all taxonomic categories above the species level are to some extent artificial and subjective, ideally they should as far as possible reflect evolutionary relationships.
Similar species are grouped in genera (singular genus), genera in families, families in orders, orders in classes and classes in the largest division of the animal kingdom: phyla (singular phylum). There may be various subdivisions of these categories, e.g. superfamilies, suborders, etc., and in certain groups there is even a case for erecting ‘superphyla’. There are only about 30 phyla in the animal kingdom, and only about a dozen of these, e.g. Mollusca and Brachiopoda, leave any fossil remains.
In taxonomy higher taxa are usually distinguished by their suffix (i.e. -ea, -a, etc.). As an example, the following documents the classification of the Ordovician brachiopod Eochonetes advena gracilis, referred to earlier, according to a taxonomic scheme in which the author of the taxon and the year of publication are quoted.
In the above section we have seen the divisions of the taxonomic hierarchy, but in defining these groupings how do taxonomists actually go about it? The basic principle is that morphological and other similarities reflect phylogenetic affinity (homology). This is always so unless, for other reasons, similarity results from convergent evolution. But in assessing ‘similarities’ how does one decide upon which characters are important? How should they be chosen to minimize subjectivity and produce natural order groups? There is no universally accepted method of facilitating these ends and taxonomists have used different methods. In recent years three sharply contrasting schools have emerged: these are the schools of (1) evolutionary taxonomy, (2) numerical taxonomy and (3) cladism.
Until the 1970s most palaeontologists, especially those working with fossil material which they have collected in the field, were evolutionary taxonomists. In erecting a hierarchical classification, such classical taxonomists used a traditional and very flexible combination of criteria. First there is morphological (or phenetic) resemblance, the extent to which the animals resemble one another. Second, phylogenetic relationships are along with phenetic resemblance considered important. By this is meant the way (as far as can be determined) in which animals are actually related to each other, i.e. in terms of recency of common origin, which of course grades into evolutionary taxonomy. The order of succession in the rock record and geographical distribution may play an important part in deciding relationships. This practical approach to taxonomy, which took all factors into consideration, has for a long time been the backbone of palaeontological classification, and is still considered to be the best method by stratophenetic palaeontologists, who place much emphasis on time in seeking ancestor-descendant relationships (Henry, 1984; Gingerich, 1990).
For some taxonomists, however, the uncertainties and subjectivity which are almost inescapable in any kind of classification seemed to be particularly acute in classical taxonomy, as did the limitations of the fossil record in terms of preservation. The numerical taxonomists tried to escape from this problem by opting for quantified phenetic resemblance as the only realistic guide to natural groupings. It was their view that if enough characters were measured, quantified and computed and represented by the use of cluster statistics, the distances between clusters could be used as a measure of their differences. Numerical taxonomy has been found very useful in some instances, but subjectivity cannot be eliminated since the operator has to choose (subjectively) how best to analyse the measurements made, and may need to ‘weight’ them, giving certain characters more importance (again subjectively) in order to obtain meaningful results. Hence the objectivity of numerical taxonomy is less than it might appear.
The third school relies upon phylogenetic criteria alone, emphasizing that features shared by organisms manifest, in nature, a hierarchial pattern, evident in the distribution of characters shared amongst organisms. It is known as cladism or phylogenetic systematics – a school founded by the German entomologist Willi Hennig (1966) and was soon applied vigorously to palaeontology (e.g. Eldredge and Cracraft, 1980). In the early days of cladistics there were many doubters (including, as I have to admit, the present author). But, as the method itself evolved, cladistics has come to be recognized as by far the most effective method for reconstructing phylogeny. Smith’s (1994) explanation of cladistic methodology is so comprehensive that only brief comments are given here.
Hennig was of the opinion that recency of common origin could best be shown by the shared possession of evolutionary novelties or ‘derived characters’. Thus in closely related groups we would see ‘shared derived characters’ (synapomorphies) which would distinguish this group from others. Hennig’s central concept was that in any group characters are either ‘primitive’ (symplesiomorphic) or ‘derived’. Thus all vertebrates have backbones; the possession of a backbone is a primitive character for all vertebrates and is not, of course, indicative of any close relationship between any group of vertebrates. What is a primitive character for all vertebrates is of course a derived character as compared to invertebrates – synapomorphy and symplesiomorphy thus delineate the relative status of particular characters with respect to a specific problem.
Hennig endeavoured to provide an objective methodology for determining recency of common origin in related taxa, based upon primitive and derived characters. Such relationships are expressed in a cladogram (Fig. 1.3), in which dichotomous branching points are arranged in nested hierarchies.
Here taxa A and B share a unique common ancestor. They are said to be sister groups. They both share an evolutionary novelty or synapomorphy, not possessed by taxon C. Now C is the sister group of the combined taxa A and B, and D is the sister group of combined taxa A, B and C. In performing a cladistic analysis, therefore, a taxonomist assumes that dichotomous splitting had occurred in each lineage and compiles an (unweighted) character data matrix. The more characters and character states there are available the larger the database, and large databases are often routinely processed and the construction of a cladogram speeded up by using one of several computer programs. The PAUP (Phylogenetic Analysis Using Parsimony) program, for example, is a technique which makes the fewest assumptions in ordering a set of observations.
Figure 1.3 A cladogram (for explanation see text).
How can we distinguish symplesiomorphic (shared primitive) from synapomorphic (shared derived) character states? The most useful way is ‘outgroup comparison’. Here an ‘ingroup’ (of which the relationships are under investigation) is specifically designated, and compared with a closely related ‘outgroup’. Any character present in a variable state in the ‘ingroup’ must be plesiomorphic if it is also found in the outgroup. Likewise, apomorphic characters are only present in the ingroup.
Hennig distinguished three kinds of cladistic groupings. Monophyletic groups contain the common ancestor and all of its descendants (D, C, B, A in Fig. 1.3); paraphyletic groups are descended from a common ancestor (usually now extinct and known as the stem group) but do not include all descendants (B and C, for example, in Fig. 1.3); polyphyletic groups on the other hand, are the result of convergent evolution. Their representatives are descended from different ancestors and hence, although these may look superficially similar, any polyphyletic group comprising them is artificial.
A cladogram is not an evolutionary tree; it is an analysis of relationships. As such it is a valuable and rigorous way of working out and showing graphically how organisms are related, and it forces taxonomists to be explicit about patterns and groups. The methodology of cladistics is especially good when dealing with discrete groups with large morphological and stratigraphic gaps and to these it brings the potential for real objectivity. A cladogram shows how sister groups are hierarchically related on the basis of shared-derived homologies, but although it portrays taxonomic relationships in terms of recency of common origin, the order of succession in the rock record is not taken into account (though implicitly cladograms have a time axis).
So where does that leave the potential contribution of stratigraphy in reconstructing phylogeny? Whilst a few ‘transformed cladists’ negate the value of the fossil record altogether (a view vigorously opposed by Ridley, 1985), the successive appearance of taxa in stratigraphy cannot be denied as an essential source of data, however imperfect the fossil record actually is. Thus as Gingerich (1990) comments, ‘Time is a fundamental dimension in evolutionary studies, and a major goal of palaeontology should continue to be the study of the diversification of major groups in relationship to geological time’. So, having constructed an appropriate cladogram, the next stage in exploring relationships is to combine this information with biostratigraphic data. For this, Smith (1994) discusses methodology in detail. The result is a phylogenetic tree, which shows the splitting of lineages through time and is effectively ‘a“best estimate” of the tree of life’. Very commonly there is an excellent correspondence between the cladogram and the rock record; on the other hand the combined cladistic and biostratigraphic approach may throw up unexpected patterns.
The ultimate problem, not only for cladism, but for all taxonomic methodology, results from convergent evolution. Resemblances in characters or character states may have nothing to do with recency of common origin, but from convergence, and it may not always be possible to disentangle the results of the two. Thus Willmer (1990) and Moore and Willmer (1997) in considering the relationships between major invertebrate groups argue that ‘cladistic analysis based on parsimony will tend to minimize and thus conceal convergence’, and they contend that convergence at all levels is far more important than has generally been believed. There is certainly a problem here. Even so, cladistic methodology, coupled with biostratigraphy seems to be the best way forward, and an essential prerequisite for drawing up meaningful phylogenetic trees.
Use of statistical methods
Inevitably palaeontological taxonomy carries a certain element of subjectivity since the information coded in fossilized shells does not give a complete record of the structure and life of the animals that bore them. There are particular complications that cause trouble. For instance, palaeontological taxonomy can do little to distinguish sibling species, which look alike and live in the same area but cannot interbreed. Polymorphic species, in which many forms are present within one biological species, may likewise be hard to speciate correctly. In particular, where sexual dimorphism is strong the males and females of the one species may be so dissimilar in appearance that they have sometimes been described as different species, and the true situation may be hard to disentangle (as with ammonites; Chapter 8).
This is only one of a whole series of possible tests, and more elaborate techniques of multivariate analysis are becoming increasingly important in taxonomic evolutionary studies.
With the advent of microcomputers and the provision of specialist software packages designed specifically to meet the needs of palaeontologists (PALSTAT; Bruton and Harper, 1990; Ryan et al., 1995), the use of numerical analysis is becoming standard. Statistical methods are likewise essential in defining palaeocommunities, in ‘undeforming’ populations of deformed fossils and comparing them to unaltered material.
Palaeobiology
Various categories may be included in palaeobiology: palaeoecology (here discussed with palaeobiogeography), functional morphology and ichnology, each of which requires some discussion.
Palaeoecology
Since ecology is the study of animals in relation to their environment, palaeoecology is the study of ancient organisms in their environmental context. All animals are adapted to their environment in all of its physical, chemical and biological aspects. Each species is precisely adapted to a particular ecological niche in which it feeds and breeds. It is the task of palaeoecology to find out about the nature of these adaptations in fossil organisms and about the relationships of the animals with each other and their environments; it involves the exploration of both present and past ecosystems (Schäfer, 1972)
Although palaeoecology is obviously related to ecosystem ecology, it is not and cannot be the same. In modern community ecology much emphasis is placed on energy flow through the system, but this kind of determination is just not possible when dealing with dead communities. Instead palaeontologists perforce must concentrate on establishing the composition, structure and organization of palaeocommunities, in attempting from here to work out linkage patterns in food webs and in investigating the autecology of individual species.
Many attempts have been made to summarize categories of fossil residue, to provide the background for interpreting original community structure. The scheme proposed by Pickerill and Brenchley (1975) and amended by Lockley (1983) is used here:
Palaeoecology must always remain a partial and incomplete science, for so much of the information available for the study of modern ecosystems is simply not preserved in ancient ones. The animals themselves are all dead and their soft parts have gone; the original physics and chemistry of the environment is not directly observable and can only be inferred from such secondary evidence as is available; the shells may have been transported away from their original environments by currents, and the fossil assemblages that are found may well be mixed or incomplete; post-depositional diagenetic processes may have altered the evidence still further. Despite this palaeoecology remains a valid, if partial science. Much is now known about the postmortem history of organic remains (taphonomy; Chapter 12), of which an additional dimension involves burial processes (biostratinomy). This helps to disentangle the various factors responsible for deposition of a particular fossil assemblage, so that assemblages preserved in situ, which can yield valuable palaeoecological information, can be distinguished from assemblages that have been transported.
Biostratinomy or preservation history has both pre-burial and post-burial elements. The former include transportation, physical, chemical or biological damage to the shell, and the attachment of epifauna. Post-burial processes may involve disturbance by burrowers and sediment eaters (bioturbation), current reworking, solution and other diagenetic preservation changes.
Modern environments and vertical distribution of animals
Figure 1.4 shows the main environments within the Earth’s oceans at the present day and the nomenclature for the distribution of marine animals within the oceans.
Modern marine environments are graded accordingly to depth. The littoral environments of the shore grade into the subtidal shelf, and at the edge of the shelf the continental slope goes down to depth; this is the bathyal zone. Below this lie the flat abyssal plains and the hadal zones of the deepocean trenches. There is often a pronounced zonation of life forms in depth zones more or less parallel with the shore. In addition there is a general decrease in abundance (number of individuals) but not necessarily diversity (number of species) on descent into deeper water from the edge of the shelf. The faunas of the abyssal and hadal regions were originally derived from those of shallow waters but are highly adapted for catching the limited food available at great depths. These regions are, however, impoverished relative to the shallow-water regions.
Animals and plants that live on the sea floor are benthic; those that drift passively or swim feebly in the water column are planktic () since they are the plankton. Nekton (nektic or nektonic fauna) on the other hand comprises active swimmers. Neritic animals belong to the shallow waters near land and include demersal elements which live above the continental shelves and feed on benthic animals thereon. Pelagic or oceanic faunas inhabit the surface waters or middle depths of the open oceans; bathypelagic and the usually benthic, abyssal and hadal organisms inhabit the great depths.
Figure 1.4 Modern marine environments. A, B and C in the inset refer to supralittoral, littoral and sublittoral environments. (Based on a drawing in Laporte, 1968.)
Only the shelf and slope environments are normally preserved in the geological record, the trench sediments rarely so. The abyssal plains are underlain by basaltic rock, formed at the mid-oceanic ridges and slowly moving away from them to become finally consumed at the subduction zones lying below the oceanic trenches; it moves as rigid plates. The ocean basins are very young geologically, the oldest sediments known therein being of Triassic age. These are now approaching a subduction zone and are soon to be consumed without trace. Hence there are very few indications of abyssal sediment now uplifted and on the continents.
What is preserved in the geological record is therefore only a fraction, albeit the most populous part, of the biotic realm of ancient times. The sediments of the continental shelf include those of the littoral, lagoonal, shallow subtidal, median and outer shelf realms. Generally sediments become finer towards the edges of the shelf, the muddier regions lying offshore. There may be reefs close to the shore or where there is a pronounced break in slope.
Horizontal distribution of marine animals
The main controls affecting the horizontal distribution of recent and fossil animals are temperature, the nature of the substrate, salinity and water turbulence. The large-scale distribution of animals in the oceans is largely a function of temperature, whereas the other factors generally operate on a more local scale. Tropical shelf regions carry the most diverse faunas, and in these the species are very numerous but the number of individuals of any one species is relatively low. In temperate through boreal regions the species diversity is less, though the number of individuals per species can be very large.
Salinity in the sea is of the order of 35 parts per thousand. Most marine animals are stenohaline, i.e. confined to waters of near-normal salinity. A few are euryhaline, i.e. very tolerant of reduced salinity. The brackish water environment is physiologically ‘difficult’, and faunas living in brackish waters are normally composed of very few species, especially bivalves and gastropods belonging to specialized and often long-ranged genera. These same genera can be found in sedimentary rocks as old as the Jurassic, and their occurrence in particular sediments which lack normal marine fossils is a valuable pointer to reduced salinity in the environment in which they lived.
Water turbulence may exercise a substantial control over distribution, and the characters of faunas in high-and low-energy environments are often very disparate. Robust, thick-shelled and rounded species are normally adapted for high-energy conditions, whereas thin-shelled and fragile forms point to a much quieter water environment, and it may be possible to infer much about relative turbulence in a fossil environment merely from the type of shells that occur.
MODERN AND ANCIENT COMMUNITIES
In shallow cold-temperature seas marine invertebrates are normally found in recurrent ecological communities or associations, which are usually substrate related. In these a particular set of species are usually found together since they have the same habitat preferences. Within these communities the animals either do not compete directly, being adapted to microniches within the same habitat, or have a stable predator–prey relationship.
Community structure is normally well defined in cold-temperature areas, but in warmer seas where diversity is higher it is generally less clear.
Petersen (1918), working on the faunas of the Kattegat, first studied and defined some of these naturally occurring communities. He also recognized two categories of bottom-dwelling animals: infaunal (buried and living within the sediment) and epifaunal (living on the sea floor or on rocks or seaweed). It was soon found that parallel communities occur, with the same genera but not the same species, on the opposite sides of the Atlantic. Since this pioneer work a whole science of community ecology has grown up, having its counterpart in palaeoecology. Much effort has been expended in trying to understand the composition of fossil communities, the habits of the animals composing them, community evolution through time and, as far as possible, the controls acting upon them (e.g. Thorson, 1957, 1971). This is perhaps the most active field of palaeoecology at present, as a host of recent original works testifies (e.g. Craig, 1954; Ziegler et al., 1968; Boucot, 1975, 1981; Scott and West, 1976; McKerrow, 1978; Skinner et al., 1981; Dodd and Stanton, 1990; Bosence and Allison, 1995; Brenchley and Harper, 1997).
As Fürsich (1977) makes it clear, however, most fossil assemblages ‘lack soft-bodied animals. Where possible trace fossils can be used to compensate for this but they are no real substitute’. Hence assemblages, associations and even palaeocommunities must not be considered as directly equivalent to the sea-floor communities of the present day. Using biostratinomic and sedimentological data the ‘degree of distortion’ from the original community can in some cases be estimated.
FEEDING RELATIONSHIPS AND COMMUNITIES
All modern animals feed on plants, other animals, organic detritus or degradation products. The tiny plants of the plankton are the primary producers (autotrophs), as are seaweeds. Small planktic animals are the primary consumers (herbivores and detritus eaters); there are secondary (carnivores) and tertiary consumers (top carnivores) in turn. Each animal species is therefore part of a food web of trophic (i.e. feeding) relationships wherein there are a number of trophic levels. In palaeoecology it is rarely possible to draw up a realistic food web (though this is one of the more important aspects of modern ecology), but most fossil animals can usually be assigned to their correct feeding type and so the trophic level may be estimated reasonably.
Of primary consumers the following types are important:
Secondary and tertiary consumers, the carnivores, may prey on any of these, but it is the communities of the primary trophic level that are most commonly preserved because of their sheer number of individuals.
In many living communities most of the ‘biomass’ is actually contributed by very few taxa, usually not more than five (the trophic nucleus), but there may be representatives of a number of other species in small numbers. In this system competition between the species concerned seems to be minimized. It is thus mutually beneficial since the different taxa are exploiting different resources within the environment. Living communities are therefore generally well balanced, the number of species and individuals of particular species being controlled by the nature and availability of food resources. Fossil assemblages may be tested according to this concept. If they are ‘unbalanced’ then either (1) there may have been soft-bodied unpreservable organisms which originally completed the balance or (2) the assemblage has been mixed through transportation and thus does not reflect the true original community.
FAUNAL PROVINCES
Marine zoogeography (Ekman, 1953; Briggs, 1974; Hallam, 1996) is primarily concerned with the global distribution of marine faunas and with the definition of faunal provinces. These are large geographical regions of the sea (and most particularly the continental shelves) within which the faunas at the specific, generic and sometimes familial level have a distinct identity. In faunal provinces many of the animals are endemic, i.e. not found outside a particular province. Such provinces are often scparated from neighbouring provinces by fairly sharp boundaries, though in other cases the boundaries may be more gradational.
Figure 1.5 shows the main zoogeographical regions of the present continental shelves as defined by Briggs (1974).
Tropical shelf, warm temperate, cold temperate and polar regions can be distinguished, whose limits are controlled by latitude but also by the spread of warmer or colder water through major marine currents. Tropical shelf faunas occur in four separate provinces. Of these the large Indo-West Pacific province, extending from southern Africa to eastern Australia, is the richest and most diverse and has been a major centre of dispersal throughout the Tertiary. Smaller and generally poorer provinces of tropical faunas are found in the East Pacific, West Atlantic and (least diverse of all) East Atlantic. These are separated both by land barriers (e.g. the Panama Isthmus) and by regions of cooler water (e.g. where the cold Humboldt Current sweeping up the western side of South America restricts the tropical fauna to within a few degrees south of the equator).
