Coral Reef Fishes: Dynamics and Diversity in a Complex Ecosystem ppt

Coral Reef Fishes
This Page Intentionally Left Blank
CO ra I R e el: Fis h es
Dynamics and Diversity
in a Complex Ecosystem
Edited by
Peter E Sale
Department of Biological Sciences and
Great Lakes Institute for Environmental Research
University of Windsor
Windsor, Ontario, Canada
Amsterdam
San Diego
ACADEMIC PRESS
An imprint of Elsevier Science
Boston London New York Oxford Paris
San Francisco Singapore Sydney Tokyo
This book is printed on acid-free paper. (~
Copyright 9 2002, 1991, Elsevier Science (USA).
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 publisher.
Requests for permission to make copies of any part of the work should be mailed to:
Permissions Department, Harcourt Inc., 6277 Sea Harbor Drive,
Orlando, Florida 32887-6777
Academic Press
An imprint of Elsevier Science
525 B Street, Suite 1900, San Diego, California 92101-4495, USA

Academic Press
84 Theobalds Road, London WCIX 8RR, UK

Library of Congress Catalog Card Number: 2001096577
International Standard Book Number: 0-12-615185-7
PRINTED IN THE UNITED STATES OF AMERICA
02 03 04 05 06 07 MB 9 8 7 6 5 4 3 2 1
Chris
Here briefly, learning, one with nature.
Memories swim ever gently.
This Page Intentionally Left Blank
Contents
Contributors
Preface xiii
xi
CHAPTER
CHAPTER
CHAPTER
CHAPTER
CHAPTER
SECTION I
Reef Fishes- A Diversity of Adaptations and Specializations
Introduction 3
1 The History and Biogeography of Fishes on Coral Reefs
David R. Bellwood, Peter C. Wainwright
2 Ecomorphology of Feeding in Coral Reef Fishes
Peter C. Wainwright, David R. Bellwood
33
Age-Based Studies
57

J. Howard Choat, D. Ross Robertson
Rarity in Coral Reef Fish Communities 81
Geoffrey P. Jones, M. Julian Caley, Philip L. Munday
5 The Ecological Context of Reproductive Behavior 103
Christopher W. Petersen, Robert R. Warner
CHAPTER
CHAPTER
SECTION II
Replenishment of Reef Fish Populations and Communities
Introduction
121
6 The Sensory World of Coral Reef Fishes
Arthur A. Myrberg Jr., Lee A. Fuiman
123
7 Larval Dispersal and Retention and Consequences for Population
Connectivity 149
Robert K. Cowen
vii
viii Contents
CHAPTER 8
CHAPTER 9
CHAPTER 10
The Biology, Behavior and Ecology of the Pelagic, Larval Stage
of Coral Reef Fishes 171
Jeffrey M. Leis, Mark I. McCormick
Biogeography and Larval Dispersal Inferred from
Population Genetic Analysis 201
Serge Planes
Numerical and Energetic Processes in the Ecology of
Coral Reef Fishes 221

Geoffrey P. Jones, Mark I. McCormick
SECTION III
Dynamics of Reef Fish Populations and Communities
Introduction
241
CHAPTER 11 Otolith Applications in Reef Fish Ecology
Simon R. Thorrold, Jonathan A. Hare
243
CHAPTER 12 Energetics and Fish Diversity on Coral Reefs
Mireille L. Harmelin-Vivien
265
CHAPTER 13 Simulating Large-Scale Population Dynamics Using
Small-Scale Data 275
Graham E. Forrester, Richard R. Vance, Mark A. Steele
CHAPTER 14 Density Dependence in Reef Fish Populations
Mark A. Hixon, Michael S. Webster
303
CHAPTER 15 Variable Replenishment and the Dynamics of Reef Fish
Populations 327
Peter J. Doherty
SECTION IV
Management of Coral Reef Fishes
Introduction
359
CHAPTER 16 The Science We Need to Develop for More Effective Management
Peter E Sale
CHAPTER 17 Reef Fish Ecology and Grouper Conservation and Management
Phillip S. Levin, Churchill B. Grimes
CHAPTER ] 8 Ecological Issues and the Trades in Live Reef Fishes
Yvonne J. Sadovy, Amanda C. J. Vincent

391
361
377
Contents ix
CHAPTER 19 Yet Another Review of Marine Reserves as Reef Fishery
Management Tools 421
Garry R. Russ
Bibliography 445
Taxonomic Index 523
SubjectIndex
527
This Page Intentionally Left Blank
Contributors
Numbers
in parentheses indicate
the pages on which the authors" contributions begin.
David
R. Bellwood
(5, 33), Centre for Coral Reef Biodiver-
sity, School of Marine Biology and Aquaculture, James Cook
University, Townsville, Queensland 4811, Australia
M.
Julian Caley
(81), School of Marine Biology and Aquacul-
ture, James Cook University, Townsville, Queensland 4811,
Australia
J. Howard Choat
(57), School of Marine Biology and Aqua-
culture, James Cook University, Townsville, Queensland
4811, Australia

Robert K. Cowen
(149), Rosensteil School of Marine and
Atmospheric Science, University of Miami, Miami, Florida
33149
Peter J. Doherty
(32 7), Australian Institute of Marine Science,
Cape Ferguson, Townsville, Queensland 4810, Australia
Graham E. Forrester
(275), Department of Biological Sci-
ences, University of Rhode lsland, Kingston, Rhode Island
02881
Lee A. Fuiman
(123), Department of Marine Science, Uni-
versity of Texas at Austin, Marine Science Institute, Port
Aransas, Texas 78373
Churchill B. Grimes
(3
77),
National Marine Fisheries Service,
Southwest Fisheries Science Center, Santa Cruz Laboratory,
Santa Cruz, California 95060
Jonathan A. Hare
(243), National Oceanic and Atmospheric
Administration, National Ocean Service, National Centers
for Coastal Ocean Science, Center for Coastal Fisheries and
Habitat Research, Beaufort, North Carolina 28516
Mireille L. Harmelin-Vivien
(265), Centre d'Ocdanologie de
Marseille, CNRS UMR 6540, Universit8 de la M~diterran~e-
Station Marine d'Endoume 13007 Marseille, France

Mark
A. Hixon
(303), Department of Zoology, Oregon State
University, Corvallis, Oregon 97331
Geoffrey P. Jones
(81,221), Centre for Coral Reef Biodiver-
sity, School of Marine Biology and Aquaculture, James Cook
University, Townsville, Queensland 4811, Australia
Jeffrey M. Lcis
(171), Ichthyology and Center for Biodiver-
sity and Conservation Research, Australian Museum, Sydney
NSW 2010, Australia
PhiUip S. Levin
(377), National Marine Fisheries Service,
Seattle, Washington 98112
Mark I. McCormick
(171, 221), School of Marine Biology
and Aquaculture, James Cook University, Townsville,
Queensland 4811, Australia
Philip L. Munday
(81), Centre for Coral Reef Biodiversity,
School of Marine Biology and Aquaculture, James Cook
University, Townsville, Queensland 4811, Australia
Arthur A. Myrberg, Jr.
(123), Rosensteil School of Marine and
Atmospheric Science, University of Miami, Miami, Florida
33149
Christopher W. Petersen
(103), College of the Atlantic,
Bar Harbor, Maine 04609

Serge Planes
(201), Centre de Biologie et d'Ecologie Tropi-
cale et Mdditerran~e, EPHE ESA 8046 CNRS, Universitd de
Perpignan, Perpignan 66860, France
D. Ross Robertson
(57), Smithsonian Tropical Research In-
stitute, Balboa, Panama
Garry
R. Russ
(421), School of Marine Biology and Aquacul-
ture, James Cook University, Townsville, Queensland 4811,
Australia
Yvonne J. Sadovy
(391), Department of Ecology & Biodiver-
sity, The University of Hong Kong, Hong Kong
Peter F. Sale
(361), Department of Biological Sciences
and Great Lakes Institute for Environmental Research,
University of Windsor, Windsor, Ontario, Canada N9B
3P4
Mark A. Steele
(275), Department of Biological Sci-
ences, University of Rhode Island, Kingston, Rhode Island
02881
Simon
R. Thorrold
(243), Biology Department, Woods
Hole Oceanographic Institution, Woods Hole, Massachusetts
02543
Richard R. Vance

(275), Department of Organismic Bio-
logy, Ecology and Evolution, University of California at
Los Angeles, Los Angeles, California 90095
xi
xii Contributors
Amanda
C. J. Vincent
(391), Project Seahorse, Department
of Biology, McGill University, Penfield, Montreal, Quebec
H3 A 1B1, Canada
Peter C. Wainwright (5, 33), Center for Population Biology,
University of California at Davis, Davis, California
95616
Robert R. Warner (103), Department of Ecology, Evolution,
and Marine Biology, University of California, Santa Barbara,
Santa Barbara, California 93106
Michael S. Webster (303), Department of Zoology, Oregon
State University, Corvallis, Oregon 97331
Pre[:ace
T
he impetus to produce this book came in a brief
phone call in 1998. Chuck Crumly, of Academic
Press, called with an invitation and a deadline. Either
The Ecology of Fishes on Coral Reefs, published in
1991, would be allowed to lapse into out-of-print sta-
tus, or I would agree to produce a second edition. Look-
ing back on all the work, I suspect it might have been
wiser to say, "Let her lapse." But I didn't. During my de-
liberations, I thought about whether a new edition was
worthwhile, whether further books on the topic were

justified, and what my colleagues would say if I came
seeking chapter authors. I told Academic Press that a
second edition was unrealistic, but that an entirely new
book, that would visit some of the same topics, was a
definite possibility. Then began the search for willing
contributors.
My intention from the beginning was to produce a
book that would speak to graduate students, to scien-
tists in the field, to reef managers and others interested
in coral reefs, and to the wider ecological and scientific
community. I am confident that this book will do so,
and will open new doors that attract new people to be-
come direct participants in this exciting field. The 30
contributors (including myself) include 15 based in the
United States, 9 in Australia, two in Canada, two in
France, one in Panama, and one in Hong Kong. (Lest
my American colleagues read this as a sign of their
preeminence, 11 of 19 chapters include authors with
significant Australian experience while just 5 have ex-
clusively American parentage. And, to keep my Aussie
friends under controlmsome of us have left your shores,
mates.) The chapters provide comprehensive coverage
of the major fields of ecology of reef fishes currently
being investigated, essential reviews in several cognate
areas, and four chapters devoted to science of manage-
ment issues. As they arrived over the last 18 months,
and I had a chance to read them, their quality provided
the spur to ensure I did the things I had to do to get
the book to press. There is some excellent work here,
and I thank each of the contributors for working hard

to produce a quality product, for putting up with my
demands, and for fulfilling my requests, usually in a
timely manner.
The book is divided into four sections with a brief
introduction to each. While the sections group together
chapters with thematic similarities, there are many in-
stances where chapters in one section make points of
relevance to chapters in other sections. Nevertheless, a
sequence from Chapter 1 to Chapter 19 makes reason-
able sense, and, if I used it in a graduate seminar, that's
the sequence I would follow.
I knew that growth in this field had been substan-
tial, but in finalizing the bibliography for this book, I
realized just how great it had been. When Paul Ehrlich
(1975) reviewed the population biology of reef fishes,
he did a thorough job in 36 pages and cited 313 ref-
erences going back to 1908, including a handful or
two from prior to 1950. In Sale (1980), I reviewed
the field in 54 pages, citing 318 references, nearly all
of which were from the 1960s and 1970s. The Ecol-
ogy of Fishes on Coral Reefs required 754 pages, of
which 87 pages comprised a bibliography of about
1690 citations, mostly from the 1970s and 1980s. The
present book contains over 2580 citations, of which
more than 60% are from 1990 or later, while just
14% are from the 1970s or earlier. Further, the present
book is less comprehensive than the former, and whole
fields of ecology are omitted to keep the book to man-
ageable size. There are a lot more people doing reef
fish ecology now than there were as little as 10 years

ago.
The other change in this field has been the growing
awareness by reef fish ecologists that our study animals
are not only wonderful, but valuable, rare, and be-
coming rarer. I hope that this book will encourage still
more ecologists to explore reef fishes as model organ-
isms with which to ask important and fundamental eco-
logical questions, and to this end, most chapters close
with questions for the future. But I hope, even more,
that this book will encourage ecologists to use their sci-
ence to contribute to much more effective management
of our impact on reef fish and the other components
xiii
xiv Preface
of coral reef systems. There is good, intellectually stim-
ulating science that is desperately needed if we are to
manage these systems sustainably in the future. I want
somebody to write a new book on reef fish ecology
in 10 years and to be able to keep it in the present
tense.
I have already thanked the contributors. In putting
the final manuscript together, I was helped by two
undergraduates in turn: Nick Kamenos, who worked
in my lab through the fall of 2000, and Allison
Pratt, who worked there through the spring and
summer of 2001. Each provided the careful attention
that allowed me to assemble a pooled bibliography with
minimal mistakes, and they did the work cheerfully. I
thank Caroline Lekic who came to my aid at a critical
point as we compiled the index. Finally, I cannot ade-

quately thank two special people, Donna and Darian,
who make my life worthwhile, while somehow under-
standing that I sometimes neglect them, only because I
do love what I do.
Peter F. Sale
Reel:Fishes
A Diversity of Adaptations
and Specializations
This Page Intentionally Left Blank
Introduction
E
cology is a holistic science that seeks a broad understanding of the relationships
among organisms, their environment, and, increasingly, humans. Though the science
of ecology has its own specialties, ecologists must remain well versed in the methods,
the goals, and the knowledge accumulated in other fields. In the past decade or so, rapid
developments in instrumentation, in techniques and analytical approaches, and in general
knowledge in fields as remote as molecular biology and physical oceanography have made
keeping up with progress in these allied fields ever more difficult. Yet, more than ever,
the challenges facing ecologists can be answered only by using the knowledge gained by
workers in other fields, and by building collaborations that go beyond the boundaries of
ecology. Several of these other fields are discussed in later parts of this book; however,
this first part deliberately attempts to provoke awareness and generate new questions.
The five chapters that make up Section I all focus on ecological science. Yet each chapter
also looks at an edge of ecological exploration, and will help fill in gaps in knowledge of
some cognate fields. I believe that each chapter has the capacity to stimulate new questions
and new approaches by those who will do the future work in reef fish ecology.
In Chapter 1, Dave Bellwood and Peter Wainwright begin at the beginning, with a
review of the origins of reef fishes and reef fish faunas, reminding us of how advanced
reef fish assemblages are, and how relatively recently derived are those species that now
dominate coral reef systems. (That they refer to this 50-million-year-ago history as a long

one confirms they are, at heart, ecologists rather than paleontologists!)
Bellwood and Wainwright review this history in a way that facilitates understanding
of the biogeographic features of coral reef systems. Given that ecological studies are in-
creasingly being done on larger spatial scales, their discussion of how reef fish assemblages
differ from place to place is particularly helpful. Their question of whether reef fishes
have played an important role in facilitating the development of coral reefs is particularly
provocative.
Wainwright and Bellwood, in Chapter 2, shift from phylogeny to functional morphol-
ogy, developing a picture of the feeding ecology of reef fishes as driven by morphological
possibilities and constraints. For those readers brought up with classical ichthyology, this
chapter will be a refreshing update, but not a surprise. But too many students of ecology
now manage to bypass "old-fashioned" courses, and, for them, this chapter may open a
new door to appreciating that ecology is the result of interactions of real organisms that
have physical limitations and possibilities.
In Chapter 3, Howard Choat and Ross Robertson steer a more narrowly defined
ecological path, but set out a strong argument for radically revising the way reef fish
ecologists and fisheries biologists have approached demographic questions. They argue,
convincingly, that it is possible to age coral reef species, and demonstrate that the results
of so doing are going to cause some significant revisions in the "conventional wisdom"
concerning longevity and growth rates in these animals. Given that so much fisheries science
depends on knowledge of age structures and growth rates, their argument has importance
for management as well as for ecology.
4 Introduction
Geoff Jones, Julian Caley, and Philip Munday use Chapter 4 to raise a difficult ecolog-
ical questionmhow to account for the existence of rarity. Although reef fish assemblages
are noted for their high diversity relative to most other assemblages of fish, relative abun-
dance of species is typically log-normally distributed and there are always many species
that are locally rare. Many of these locally rare species also are regionally, if not universally,
rare. How do we account for the successful persistence of species that are so uncommon?
Chapter 4 stimulates thinking on a vexing problem, and is a reminder that knowing still

more about the commonest species is not going to provide all the answers.
In Chapter 5, Chris Petersen and Bob Warner turn attention to another of the bound-
aries of ecology and address the behavioral adaptations of reef fish reproduction. There
was a time 25-30 years ago when there was more behavioral than ecological research done
on reef fishes. For reasons that are not entirely clear, the quantity of behavioral research
in this system has not grown along with that of ecological research; however, studies on
a few topics (by a few particular investigators) continue to demonstrate that the reef fish
system is very manageable for sophisticated explorations of behavioral questions in field
settings. The question of evolution of behavioral processes, particularly with respect to
reproductive and parental activities, has been a fruitful area for research, and this chapter
provides an introduction to this topic from two of the leaders. Their section on applied
behavioral ecology should convince readers that behavioral science remains "relevant,"
and that there are potentially important consequences if we ignore behavioral science in
managing our impacts on reef fishes.
The History and Biogeography
o[Fishes on Coral Reels
David R. Bellwood
Centre for Coral Reef Biodiversity
School of Marine Biology and Aquaculture
James Cook University
Townsville, Queensland 4811, Australia
Peter C. Wainwright
Center for Population Biology
University of California at Davis
Davis, California 95616
I. Introduction
II. Reef Fishes: Definitions and Distribution
Patterns
III. The Origins of Reef Fishes
IV. Barriers and Vicariance Events in the Evolution

and Biogeography of Reef Fishes
V. Postvicariance Survival Patterns: Fate after
Isolation
VI. The History and Nature of the Reef-Fish
Relationship
VII. Functional Aspects of the Reef-Fish Association
VIII. Discussion and Conclusions
I. Introduction
C
oral reefs have been around since the Ordovician
(Wood, 1999), and throughout their 450-million-
year history they have shared the oceans with fishes.
Modern scleractinian-dominated coral reefs and their
associated fish faunas represent only the latest mani-
festation of a reefal ecosystem. It is almost self-evident
that history is important to coral reefs, as the reefs build
on the skeletons of past generations. But what of the
associated fauna? Today, fishes form an integral part of
reef communities, modifying benthic community struc-
ture and forming a major conduit for the movement
of energy and material. Like the reefs, reef fish faunas
have been shaped by history, but this historical influ-
ence may not be as apparent. Although it is becoming
increasingly clear that history plays an important role in
structuring local communities (Rickleffs and Schluter,
1993a), its influence on the ecology and biogeography
of fishes on coral reefs remains largely unknown.
Most studies of reef systems have addressed the
question of how biogeographic and ecological patterns
are maintained; relatively few consider how these pat-

terns arose or their consequences. However, it is the
combination of these two factors, origins and main-
tenance, that offers the clearest understanding of the
nature of biogeographic patterns in reef organisms.
Studies of the history of coral reefs have been largely re-
stricted to documenting the history of the reef builders,
which have left an outstanding fossil record (Wood,
1999). The history of associated faunas, and fish in
particular, is less clear. However, this is changing, pri-
marily as a result of phylogenetic analyses of reef fishes
and from a reappraisal of the fossil record.
Until recently, historical considerations of reef
fishes were restricted largely to studies by museum
workers (e.g., Allen, Randall, Springer, Winterbottom)
who examined the taxonomy, systematics, and bio-
geography of extant reef fishes. Paleontological infor-
mation has likewise been confined to the works of spe-
cialists in museums. Workers such as Blot, Sorbini, and
Tyler have provided a sound basis for the evaluation
of the fossil record of reef fishes. The broader appli-
cation of these findings to present-day ecology, com-
munity structure, and ecosystem function has only re-
cently begun to be considered. Ecologists are looking
increasingly at data from large temporal and spatial
scales to provide a framework within which to inter-
pret local patterns and small-scale experimental results.
It is from this integration of systematics, biogeography,
ecology, and paleontology that a new understanding of
the nature of reef fishes is arising.
In this chapter we summarize our knowledge of

the phylogenetics, paleontology, and biogeography of
fishes on coral reefs and examine how these data, along
with geological evidence, can aid our understanding of
the role of historical factors in shaping modern coral
reef fish faunas and their ecological attributes. In par-
ticular, we wish to address several specific questions:
Coral Reef Fishes
Copyright 2002, Elsevier Science (USA). All rights reserved.
6 Bellwood and Wainwright
1. What are coral reef fishes, when did they appear,
and where did they come from?
2. Are Caribbean and Indo-Pacific reef fish
assemblages comparable, and how do we explain
major differences in reef fish assemblages across
the Indo-Pacific?
3. How tight is the reef fish-coral reef association,
and how do we evaluate the interaction between
fishes and coral reefs?
4. What role have fishes played in the evolution of
coral reefs, and is there any evidence of a change
in this role over time?
II. Reef Fishes: Definitions and
Distribution Patterns
Reef fishes are often seen as a distinctive and easily
characterized group of fishes. However, though nu-
merous texts and papers refer to "reef fishes," the
uniting characteristics of these assemblages are rarely
defined. Although there have been several attempts to
characterize the essence of a reef fish, none of these
descriptions has proved to be diagnostic. Bellwood

(1988a) provided a classification based on the degree
of ecological association between the fish and reef, in
terms of the reef's role in providing food and/or shelter.
A broader overview was given in Choat and Bellwood
(1991), who described the ecological and taxonomic
characteristics of reef fishes. In this scheme, they noted
the abundance of small-gaped deep-bodied fishes on
reefs, and the numerical dominance of a few families,
including labrids, pomacentrids, chaetodontids, and
acanthurids. Later Bellwood (1996a) established a
more specific "consensus list" of reef fish families. This
list comprised all families that one would find on a
coral reef irrespective of its biogeographic location (i.e.,
Acanthuridae, Apogonidae, Blenniidae, Carangidae,
Chaetodontidae, Holocentridae, Labridae, Mullidae,
Pomacentridae, and Scaridae). These 10 families were
regarded as characteristic reef fish families, the essence
of a reef fish fauna; all are abundant and speciose on
coral reefs (Fig. 1, but see Section VI below).
However, these studies have all looked at the sim-
ilarities among reef fish faunas. They provide only a
description of a reef fish fauna and are not diagnostic
(Bellwood, 1998). Further examination of reef and non-
reef areas has found that many of the characteristics
of reef fish faunas may apply equally well to nonreefal
fish faunas (Bellwood, 1998; Robertson, 1998b). In this
chapter therefore, the term "reef fish" refers to those
taxa that are found on, and are characteristic of, coral
reefs (i.e., the consensus list plus taxa characteristic of
reefs in specific areas).

An understanding of the nature of the differences
among reef fish faunas is critical to our understanding
of the evolution of reef fishes and the role of history
in determining the structure of modern reef fish assem-
blages. The dissimilarity between reef fish faunas can
be seen in Fig. 1, which contrasts the species richness in
a number of fish families at four biogeographically dis-
tinct reefal locations. Several features are immediately
apparent:
1. Despite a more than threefold decrease in species
numbers between the Great Barrier Reef (GBR) and
the Red Sea, the basic pattern remains broadly com-
parable. The Red Sea reef fish fauna appears to be a
random subset of a comparable high-diversity Indo-
Pacific system such as the GBR. Indeed, there is no sig-
nificant difference between the two faunas in terms of
the distribution of species in families (x 2 16.9; p, 0.46;
df, 17).
2. Although overall the data for species/familial
diversity are similar in the Caribbean and Red Sea
(277/45 and 281/40, respectively), the familial com-
position and patterns of familial species richness vary
markedly. In the Caribbean, the Lethrinidae, Pseu-
dochromidae, Siganidae, Nemipteridae, and Caesion-
idae (Caesioninae) are absent. Together these fami-
lies comprise approximately 7% of the species in the
GBR fish fauna. However, several families are relatively
well represented in the Caribbean, including the Ser-
ranidae, Haemulidae, and Sparidae and the regional
(East Pacific/Caribbean) endemics, the Chaenopsidae

and Labrisomidae.
3. Many of the characteristic reef fish families
(e.g., Labridae, Pomacentridae) are present and abun-
dant in New Zealand, a temperate region devoid of
coral reefs. A similar pattern is seen in South Africa,
South Australia, and western North America. Thus,
although we readily recognize them as coral reef fish
families, most of these characteristic reef fish fam-
ilies do not disappear when coral reefs stop. These
taxa are characteristic of, but not restricted to, coral
reefs.
If comparable data sets collected from a range
of reefal and subtropical/temperate locations are ex-
amined using a Principal Component Analysis [PCA;
modified after Bellwood (1997)] clear regional group-
ings are apparent (Fig. 2A, C), with high-, medium-,
and low-diversity, low-latitude Indo-Pacific sites laying
along the first axis. The decreasing diversity at these
sites generally tracks a longitudinal shift away from
History and Biogeography of Fishes on Coral Reefs 7
120
'~176
It
80 Great Barrier Reef
40
20 Nflnnnlnnnnnnnnn
o
4o-iit1,,,,3o
20
o iHnnfloRn

nlnnn~_~ ~n~n
r.~ o
40 - I
~ 30
20
,o II ,,,,hnno no noon
lO
I,! n nHlO
nHn nnn
0 I
9
? ~,~ J J
5 10 15 20 25 30 35 40 45
Red Sea
Caribbean
_,,_ _ _n H
New Zealand
n Flnnr, ~
H
I t i t
50 55 60 65
nNooH
I I
70 75
Family
FIGURE 1 Species richness, by family, at four sites. The ranking of families at each
site follows the Great Barrier Reef. Characteristic reef fish families are indicated by
solid bars (after Bellwood, 1996a, 1997). Families: 1, Labridae; 2, Pomacentridae;
3, Serranidae; 4, Blenniidae; 5, Apogonidae; 6, Chaetodontidae; 7, Acanthuridae; 8,
Scaridae; 9, Holocentridae; 10, Lutjanidae; 11, Pomacanthidae; 12, Scorpaenidae;

13, Lethrinidae; 14, Monacanthidae; 15, Pseudochromidae; 16, Balistidae; 17,
Microdesmidae; 18, Tetraodontidae; 19, Mullidae; 20, Syngnathidae; 21, Siganidae;
22, Cirrhitidae; 23, Haemulidae; 24, Nemipteridae; 25, Ostraciidae; 26, Pinguipedi-
dae; 27, Synodontidae; 28, Caesionidae; 29, Antennariidae; 30, Diodontidae; 31,
Plesiopidae; 32, Sphyraenidae; 33, Tripterygiidae; 34, Callionymidae; 35, Ephippi-
dae; 36, Malacanthidae; 37, Pempheridae; 38, Kyphosidae; 39, Priacanthidae; 40,
Bythitidae; 41, Caracanthidae; 42, Gobiesocidae; 43, Mugilidae; 44, Opistognathi-
dae; 45, Plotosidae; 46, Solenostomidae; 47, Trichonotidae; 48, Acanthoclinidae; 49,
Aploactinidae; 50, Aulostomidae; 51, Batrachoididae; 52, Carapidae; 53, Centrisci-
dae; 54, Centropomidae; 55, Chandidae; 56, Creediidae; 57, Dactylopteridae; 58,
Echeinidae; 59, Eleotridae; 60, Fistulariidae; 61, Sparidae; 62, Teraponidae; 63, Ura-
noscopidae; 64, Xenisthmidae; 65, Zanclidae; 66, Albulidae; 67, Aplodactylidae; 68,
Berycidae; 69, Chaenopsidae; 70, Cheilodactylidae; 71, Clinidae; 72, Cynoglossidae;
73, Labrisomidae; 74, Odacidae; 75, Ogcocephalidae; 76, Pentacerotidae.
the Indo-Australian Archipelago. Examination of the
family-vectors (Fig. 2B) suggests that the first axis is
associated primarily with total species richness. How-
ever, principal component 1 (PC1) does not just mea-
sure species richness. The scores reflect similar numbers
of species in those families exhibiting greatest varia-
tion in the data set. The strong correlation with total
species richness reflects the congruence among fami-
lies in the decrease in familial species richness. This
pattern is seen in the relatively uniform orientation of
family-vectors around PC1, which also suggests that
differences between high- and low-diversity sites are
a result of the absence of taxa at low-diversity sites,
i.e., there is no replacement. Low-diversity, low-latitude
sites merely contain a lower number of species in the
families found at high-diversity sites (as in Fig. 1). There

are no "new" families that are characteristic of low-
diversity sites (cf. Bellwood and Hughes, 2001).
The second axis explains only 12.3 % of the varia-
tion but it appears to reflect changes in the relative com-
position of the assemblages in terms of temperate vs.
tropical taxa (Fig. 2B). This axis separates high-latitude
8 Bellwood and Wainwright
A
PC1
(48.8%)
i
-4.0
3.0
E. Pacific 9
9 Caribbean
9 New Zealand
iBermuda
lSea of Cortez
Oshima Is Lord Howe Is
- 9
~Ascension I~' 9 Is
Clipperton 9 9 Rapa 9
Atoll Red Sea
Ducie Atoll9 _ 9 Is
9 9 Pitcairn Is
Henderson
Is Chri.,
-2.0
PC2
(12.3%)

9 Southern Africa
9 SE. Australia
Taiwan
escadores Is
New
9
Caledonia
i i i I
9 Samoa 9 4 0
/aii ChesterfieldO9 s 9 Micronesia
) Rowley 9 Marshall Is
.3hagos9 Shoalst 9 Bay
mals is 9 French
Polynesia
B
0.4-
38
20
C 3.o- PC2
(12.3%)
Te rn pe rate
Tropical / ~ ~ t/'~
PC1 Atlantiy / ~ /Sunhb-tl~ pical,
(48.8%) / 4" ,,,~ ~ ~ high latitude
-;.0
, ,
Low
'ow"'itu~ t
. I low latitud2. 0 .J e
I

-0.1
4.0 -0.2
PC2 (12.3%)
@
33
10
49
4
36
44 75 70
~75~1 ~ 72 69
1 7
0 46
) 1 314350 53 63
1_^ 51 15 40
18 " ~66 .68 PC1
(48.8%)
?2.75. "1~
,~z~ ,415 ' 7
377 4 623674 0:3
78 57
6~4 12
56
45 42 21
24 59
16 39
2 35
FIGURE
2 Principal Component Analysis of reef and nonreef sites. (A) Plot of sites on the first two axes.
(B)

Family-vectors, with families as listed below (temperate families are encircled). (C) Plot of sites on the first two
axes with tropical, subtropical, and temperate sites delineated. Solid dots and solid lines indicate Indo-Pacific sites;
open dots and dashed lines indicate Atlantic sites. Families: 1, Acanthoclinidae; 2, Acanthuridae; 3, Albulidae;
4, Antennariidae; 5, Aplodactylidae; 6, Aploacinidae; 7, Apogonidae; 8, Aulostomidae; 9, Balistidae; 10, Batra-
choididae; 11, Berycidae; 12, Blenniidae; 13, Bythitidae; 14, Caesionidae; 15, Callionymidae; 16, Caracanthidae;
17, Carapidae; 18, Centriscidae; 19, Centropomidae; 20, Chaenopsidae; 21, Chaetodontidae; 22, Chandidae;
23, Cheilodactylidae; 24, Cirrhitidae; 25, Clinidae; 26, Creediidae; 27, Dactylopteridae; 28, Diodontidae; 29,
Echeinidae; 30, Eleotridae; 31, Ephippidae; 32, Fistulariidae; 33, Gobiesocidae; 34, Haemulidae; 35, Holocen-
tridae; 36, Kyphosidae; 37, Labridae; 38, Labrisomidae; 39, Lethrinidae; 40, Lutjanidae; 41, Malacanthidae; 42,
Microdesmidae; 43, Monacanthidae; 44, Mugilidae; 45, Mullidae; 46, Nemipteridae; 47, Odacidae; 48, Ogco-
cephalidae; 49, Opistognathidae; 50, Ostraciidae; 51, Pempheridae; 52, Pentacerotidae; 53, Pinguipedidae; 54,
Plesiopidae; 55, Plotosidae; 56, Pomacanthidae; 57, Pomacentridae; 58, Priacanthidae; 59, Pseudochromidae; 60,
Scaridae; 61, Sciaenidae; 62, Scorpaenidae; 63, Serranidae; 64, Siganidae; 65, Sillaginidae; 66, Solenostomidae;
67, Sparidae; 68, Sphyranidae; 69, Syngnathidae; 70, Synodontidae; 71, Teraponidae; 72, Tetraodontidae; 73,
Trichonotidae; 74, Triglidae; 75, Tripterygiidae; 76, Uranoscopidae; 77, Xenisthmidae; 78, Zanclidae.
vs. low-latitude low-diversity assemblages in the Indo-
Pacific. As one moves away from the center of diversity
in the Indo-Australian Archipelago, total species diver-
sity decreases steadily with changes in both latitude and
longitude. In both cases, characteristic reef fish fami-
lies remain consistently well represented, whereas less
speciose families are progressively lost. However, the
latitudinal and longitudinal changes are not the same;
high-latitude sites have a marked temperate influence.
This temperate influence is even clearer in the
tropical Atlantic and tropical East Pacific sites. These
sites are united by the presence of endemic families
History and Biogeography
of Fishes
on Coral

Reefs 9
(Chaenopsidae, Labrisomidae), the absence of sev-
eral speciose Indo-Pacific families (e.g., Lethrinidae,
Nemipteridae, Siganidae), and an increase in the di-
versity of other families (Haemulidae), including some
with a strong representation in temperate waters (e.g.,
Sparidae, Monacanthidae). This similarity probably re-
flects a common history of the two areas prior to the
closure of the Isthmus of Panama and a shared period of
faunal loss (see Sections IV and V). The analyses sug-
gest that the Caribbean, despite being a low-latitude
tropical region with strong coral reef development, has
a reef fish fauna that is more similar to those of high
latitude or temperate Indo-Pacific sites than to tropical
Indo-Pacific sites. The Caribbean reef fish fauna has a
distinct temperate component.
The similarity between the patterns described in
reef fishes and corals are striking (Bellwood and
Hughes, 2001). The two groups have markedly differ-
ent life histories, approaching the extremes seen in ma-
rine benthic faunas. If the biogeographic patterns seen
in fish and corals reflect a common mechanism, then the
processes may be operating at the regional or ecosys-
tem level and at large temporal scales. If this is the case,
then one may expect to see congruent patterns in other
benthic marine taxa.
III. The Origins of Reef Fishes
A. Major Lineages
Fishes and corals both have a long tenure in the fos-
sil record. However, at what point in the past did events

begin to have a direct bearing on the ecology and distri-
bution patterns of modern reef fish taxa ? Devonian fish
certainly have a legacy that passes through to modern
times, but when did the history of modern reef fishes
begin? The answer, it seems, is that these groups were
already in place by the early Tertiary [50 million years
(Ma) ago], with origins spreading back to at least the
late Cretaceous (70 Ma), and possibly even to the early
Cretaceous ( 100-130 Ma).
Most reef fish families have been placed in the order
Perciformes. This order contains approximately 9293
species, and represents about 63% of all marine fish
species (Nelson, 1994). The order encompasses about
75% of the fish species found on coral reefs (Randall
et al.,
1990), including all of the characteristic reef
fish families (Fig. 1). Unfortunately this order is proba-
bly paraphyletic (Johnson and Patterson, 1993). How-
ever, the Perciformes along with the Scorpaeniformes,
Pleuronectiformes, and Tetraodontiformes may form a
monophyletic group, the Percomorpha
(sensu
Johnson
and Patterson, 1993).
Estimates of the ages of major fish groups are based
on fossils or inferences from cladograms and biogeo-
graphic patterns. Fossil evidence ranges from isolated
fragments, predominantly otoliths, to complete, fully
articulated skeletons. Age estimates based on otoliths
are consistently older than those based on complete

skeletons (cf. Patterson, 1993), possibly reflecting the
abundance of otoliths in the fossil record and the fact
that otoliths do not require the exceptional conditions
necessary for preservation of the complete fish skeleton.
Identifying a fish taxon based on otoliths can be diffi-
cult because they have a limited range of characters,
often of unknown phylogenetic significance. Further-
more, fossil otoliths are often worn, and considerable
subjectivity may arise in character-state designations.
The taxonomic utility of otoliths also varies widely be-
tween taxa (Nolf, 1985). In contrast, complete skele-
tons often permit fossil taxa to be incorporated into ex-
isting cladograms, providing estimates of the minimum
age of specific lineages along with a great deal of infor-
mation on changes in functional capabilities through
time. However, complete fossil skeletons of reef fishes
are rare and minimum ages based on complete skele-
tons are likely to underestimate the actual age of the
group.
The biogeographic patterns of reef fishes observed
today are the result of a long and complex history,
which has probably involved a number of vicariance,
dispersal, and extinction events (Fig. 3). When trying
to disentangle this convoluted history, fossils provide a
unique series of reference points. The utility of fossils in
the study of phylogeny and biogeography has been crit-
ically appraised by Patterson (1981) and Humphreys
and Parenti (1986). Fossils provide neither ancestors
nor absolute ages of taxa. However, accurately dated
fossils, when combined with phylogenies, can provide

the minimum age of a lineage, its sister group, and all
of the more basal lineages. Given this age one may be
able to identify the vicariance events (i.e., environmen-
tal changes leading to the separation of populations)
that were associated with the origin and subsequent
diversification of lineages. Fossils also pinpoint a taxon
in a location at a given time. This is particularly valu-
able when this location lies outside the geographic
range of living forms.
The earliest record of the Perciformes is based on
otoliths from the late Cretaceous (Cenomanian, 97.0-
90.4 Ma) (Patterson, 1993), with the first full skeleton,
Nardoichthys,
being recorded from the upper Campa-
nian/lower Maastrichtian (c. 74 Ma) of southern Italy
(Sorbini and Bannikov, 1991 ). Of the remaining perco-
morph groups the oldest fossil, to date, is a tetraodon-
tid
Plectocretacicus
(Sorbini, 1979) from the late
1 0 BeUwood and Wainwright
-r-
i
I
I
I
I
I
i
I

I
>,
-T-
( )
0
t'q
0
HOLOCENE =0. !
PLEISTOCENE
-I .64
PLIOCENE
MIOCENE
OLIGOCENE
EOCENE
PALEOCENE
CRETACEOUS
Innundation Persian Gulf 8000 yrs
0 ~ T Current sea level reached 8000 vrs
1 plce Ages- Rapid sea level changes
~
3-
4-
-5.2 5 -
-5.2 5 I
lg-
2O
-23.3
3O
-35.4
40-

5O
-56.5
60
-65
70
~- Closure of Isthmus of Panama
~Red Sea Indian Ocean opens
to
Mediterranean dries -
Messinian Salinity Crisis
I
Chaetodon, Chromis,
'B
_ ~g olbometopon
~k i First Scarid - Calotomus (Austria)
-Tcrminai Tethyan Event
/
-
~ Tethys connects
]Indian and Atlantic
Oceans
- Development of Circum-
Antarctic Current
F India Asian continent
impacts
F Labridae recorded from Antarctica
-
~ Monte Bolca (Italy) - First Labridae,
Acanthuridae, Pomacentridae,
Siganidae, Apogonidae, Ephippidae.

F irst Acropora coral
(N. Somalia)
K/T Mass Extinction Event
First Perciform fish -
Nardoichthys (Italy)
; Periodic loss of extensive reef
I
!
, areas - GBR dry at 18000 yrs
',
I
Extensive faunal turnover or
I extinction in isolated locations
[ incl. E. Pacific and Caribbean
. Marine tropics physically
[ divided into two regions
t______ Increasing reef area in Indo-
Australian Archipelago
i Global tropics divided:
W. Tethys = lndo-Pacific,
E. Tethys = E. Pacific,
Caribbean, Atlantic
Diverse coral assemblages in Tethys,
especially in S. Europe
t__ Poles cool, tropical conditions
increasingly restricted to low
latitudes
~- Last remnants of Mesozoic fishes -
A few Pycnodontids remain near
I

reefs
Extinction of rudist bivalves,
atnmonites, dinosaurs and most
Mesozoic fish groups
u_._ Rudists dominate carbonate platforms
I - Scleractinian corals minor
I
, components
I Fishes Mesozoic forms incl.
I ~
I
,, Semionotids, Pycnodontids, etc.
I
80-
~r
145.6
FIGURE 3 Major events in the history of fishes on coral reefs, showing the relationship between the
appearance or loss of fish groups, changes in the status of coral reefs, and major biogeographic events.
Earliest records of fish groups refer to identifications based on complete skeletal remains. Ages given
in Ma. See text for details.