m JOHN G.MAIS cum of Natural Hi: *& V^^K JOHN G MAISEY Department of Vertebrate Paleontology American Museum of Natural History Illustrations by DAVID IVY MILLER RUTZKY Photography by CRAIG CHESEK DENIS FINNIN FP^TT F^HRS a PETER N N E V R A U M O N T book A Henry Holt Reference Book HENRY HOLT AND COMPANY New York contents AQUATIC A N C E S T O R S A PARADISE FOR F I S H E S FISHY O R I G I N S THE RISE AND FALL OF OUR A R M O R E D A N C E S T O R S THE FIRST PREDATORS EXTINCT RELATIVES OF M O D E R N F I S H E S S H A R K S : THE LORDS OF TIME O S T E I C H T H Y A N S : SO M U C H DIVERSITY, SO LITTLE CHANGE THE F O S S I L FOREST OF R A Y - F I N N E D 24 10 F I S H E S OUT OF WATER 11 THE WORLD OF F I S H E S Glossary 216 Acknowledgments Index 222 16 221 36 58 192 210 FISHES 70 138 To Bobb Schaeffer, colleague, mentor, and friend A Henry Holt Reference Book Henry Holt and Company, Inc Publishers since 1866 115 West 18th Street New York, New York IOOII Henry Holt ® is a registered trademark of Henry Holt and Company, Copyright © 1996 John G Maisey Illustration (reconstructions) copyright © 199(1 David W Miller Illustrations (portraits) copyright © 1996 Ivy Rutzky Photographs copyright © 1996 American Museum of Natural History Line illustrations copyright © 1996 Patricia Wynne Computer images copyright © 1996 Frank Ippolito All rights reserved Published in Canada by Fitzhcnry & Whiteside Ltd., 195 Allstate Parkway, Markham, Ontario L3R 4T8 Library of Congress Cataloging-in-Publication Data is available upon Henry Holl books are available for special promotions and premium For details contact: Director, Special Markets First Edition —1996 Book design: Jose Conde, Studio Pepin, Tokyo Printed in Italy by EBS - Editoriale Bortolazzi-Stei (Verona) 10 21 Created and Produced by Nevraumont Publishing Company New York, New York Ann J Perrini, President i AQUATI ANCESTORS 5HH»r>\vii,jt's in N i i i m ? JList HES have a unique evolutionary' history that stretches back in time more than 450 million years They are incredibly ancient, older than the dinosaurs, and include the ancestors of all the limbed vertebrates living on land, even humans Human evolution is rooted in fishes; and scien tists have discovered traces of 360-million-ycar-old fossils of transitional aquatic creatures that had both gills and limbs Not only are humans and fishes related, some fishes are more closely related to humans than they are to other fishes! We commonly think of ourselves as completely separate from our aquatic ancestors when instead we should be mar veling at the similarities PLATE Cahmopkurus, an Early Cretaceous relative of the modern bowfin about im long, lunges from concealment at a passing group of filter-feeding Vinclifer, members of the extinct teleost group known as aspidorhynchids These fishes lived in a shallow inland sea covering parts of north eastern Brazil about n o million years ago (see Plates 67 and 77) Let's see where we fit in Humans are classified as t e t r a p o d s (fourfooted; more generally, having two pairs of limbs) This same designation applies to all the air-breathing land vertebrates with which we are famil iar: all other mammals, birds, reptiles, and amphibians The tetrapods in turn are a highly specialized offshoot of a very ancient class of animals, the Sarcopterygii The only other living members of this group are the coelacanths and the lungfishes It seems that we have very important, and very fishy, ancestors! We no longer look like fishes, of course, and that is because most modern fishes evolved from different ancestors than did tetrapods Our tetrapod ancestors left the water and found a new environment on land In making this transition, they lost or modified many of their aquatic adapta tions Fishes stayed in the water, and evolved new adaptations to this environ ment [Plate 1] Both fishes and tetrapods, however, share common distant ancestors These were highly mobile, aquatic creatures with a sophisticated brain, complex senso ry systems, and a muscular tail These fea tures, which first appeared more than 450 million years ago, characterize a group known as the c r a n i a t e s {having a cranium, the part of the skull that encloses the brain) (Another name frequently used for the group is v e r t e b r a t e s , but, as we shall see, the terms vertebrate and craniate are not strictly interchangeable.) Tetrapods have superimposed their adaptations to ter restrial life upon their aquatic (craniate) fea tures; fishes have built more adaptations to aquatic life on top of theirs [Figure 1] As we go back in time, we can trace our tetrapod lineage to a common ances try with fishes When we try to trace the lineage forward from the earliest craniates, however, we find that tetrapods have been segmented, by classificatory fiat, from the fishes This poses a problem, for we see that fishes not form a natural group, because tetrapods have been excluded, but instead represent a heterogeneous assem blage The term "fish" is therefore generally used to refer to a life form that has many different representatives The basic form is a craniate that has retained some primitive features (evolved in the common ancestor of all craniates), such as gills and a tail fin, but may also have other features repre senting a long series of evolutionary adap tations Often the sense in which the word is used is merely conventional; from a practical viewpoint, fishes are what we study in fish classes! Here is a paradox, for although most of us have no difficulty rec ognizing a fish, collectively fishes have no unique features by which they can be classified Our first discovery, then, is that fishes not represent an entire natural lineage of creatures that evolved from a common ancestor, for the group is incomplete without tetrapods Why, then, write a book about an incomplete group? Precisely because there is so much misunderstand ing about the early evolution of craniates, it is more convenient to discuss various issues under a blanket title of "fishes." And let's face it—few people would have read a book entitled Discovering Primitive Fossil NonTetrapodan Craniates] ■5HIW -» I ishy Diversity From their ancient ancestors, the craniates, fishes evolved not once, in a single lin eage, but multiple times, filling countless biological niches The various classes of living and extinct fishes (for example sharks, lungfishes, coelacanths, ray-finned fishes, acanthodians and placoderms) are as different from each other as mammals are from birds Fishes are spectacularly diverse, with around 25,000 living species described so far and untold thousands more awaiting discovery Fishes make up slightly more than half of the approxi mately 48,500 described living species of vertebrates The remaining known verte brate species, about 25,500, arc represented by the tetrapods Given their long evolutionary history, it is not surprising that so many species of fishes exist today; one new fish species evolving every 18,000 years, or about 55.5 species evolving per million years, would account for the numbers Such an orderly, regular pace of evolution is unrealistic, of course, for no living species has existed for 450 million years The sum total of fishy diversity through time is far greater than now, and the evolutionary history of fishes is a vast and complex subject It is impossible to give an accurate estimate of how many species of fishes have existed through time, because the diversity of many groups has waxed and waned, and because there are significant gaps in the fossil record, but the total is likely to have been in the mil lions As more fossil fishes are uncovered, we will know better what the ancient world looked like and come to discover more of our own ancestors AQUATIC ANCESTORS FIGURE ©N0 Diagram to show the diversity of craniates Groups inside solid boxes are monophyletic (descended from a single ancestor, the line leading to box) Groups inside dashed boxes are unreal, because they include members of more than one evolu tionary line For example, gnathostomes (craniates with jaws, an advanced feature) form a mono phyletic group, but "agnathans" (craniates lacking jaws, a primitive condition) are unreal because they include two separate evolutionary lineages: hagfishes and lampreys Similarly, tetrapods (gnathostomes with limbs) are monophyletic, but the group "fishes" (all craniates lacking limbs) is unreal (Lineages number 1-4 are extinct groups of fishes discussed elsewhere in this book.) Discovering Fossil Fishes ■5>imr*,\i-