Advances in the Study of Behavior, Volume 37 (Advances in the Study of Behavior)

Advances in

THE STUDY OF BEHAVIOR VOLUME 37

Advances in THE STUDY OF BEHAVIOR Edited by

H. Jane Brockmann Timothy J. Roper Marc Naguib Katherine E. Wynne-Edwards Chris Barnard John Mitani

Advances in THE STUDY OF BEHAVIOR Edited by H. Jane Brockmann Department of Zoology University of Florida Gainesville, Florida

Timothy J. Roper

Marc Naguib

Department of Biology and Environmental Science University of Sussex Sussex, United Kingdom

Department of Animal Behavior University of Bielefeld Bielefeld, Germany

Katherine E. Wynne-Edwards

Chris Barnard

Department of Biology Queen’s University Kingston, Canada

The School of Biology The University of Nottingham Nottingham, United Kingdom

John Mitani Department of Anthropology University of Michigan Ann Arbor, Michigan

VOLUME 37

AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier 84 Theobald’s Road, London WC1X 8RR, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands Linacre House, Jordan Hill, Oxford OX2 8DP, UK 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA First edition 2007

Copyright ß 2007 Elsevier Inc. 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 without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting, Obtaining permission to use Elsevier material Notice No responsibilityis assumed bythepublisher for any injury and/or damage topersons orpropertyas a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made ISBN: 978-0-12-004537-2 ISSN: 0065-3454 For information on all Academic Press publications visit our website at books.elsevier.com Pinted and bound in USA 07 08 09 10 11 10 9 8 7 6 5 4 3 2 1

Contents

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix xi

The Strategic Dynamics of Cooperation in Primate Groups JOAN B. SILK I. II. III. IV. V. VI. VII. VIII. IX. X.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (Avoiding) Definitions of Cooperation . . . . . . . . . . . . . . . . Game Theory Models of Forms of ‘‘Cooperation’’ . . . . . . Which Kinds of Games Do Primates Play? . . . . . . . . . . . . The Iterated Prisoner’s Dilemma in Primate Groups . . . . Stag Hunts in Primate Groups . . . . . . . . . . . . . . . . . . . . . . . The Battle of the Sexes in Primate Groups . . . . . . . . . . . . Games of Chicken in Primate Groups . . . . . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 2 4 7 7 18 31 32 33 34 36

Coexistence in Female-Bonded Primate Groups S. PETER HENZI AND LOUISE BARRETT I. II. III. IV. V. VI. VII.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kinship and Competition . . . . . . . . . . . . . . . . . . . . . . . . . . . Organizing Principles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Spatial Approach to Social Interactions . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

43 44 46 63 70 73 75 76

The Evolution of Sociality in Spiders YAEL LUBIN AND TRINE BILDE I. Introducing Social Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . II. Social and Subsocial Species: A Survey of Behavioral Traits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

83 89

vi

CONTENTS

III. Inbred Sociality in Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Phylogenetic Relationships Among Social Spider Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Evolution and Maintenance of Sociality in Spiders: Relevant Models . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Transitions in the Evolution of Sociality: Processes and Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Summary: From Subsocial to Inbred Social, an Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

99 113 119 128 133 135

Molecular Ecology Reveals the Hidden Complexities of the Seychelles Warbler JAN KOMDEUR AND DAVID S. RICHARDSON I. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Study Species, Study Populations, and General Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Cooperative Breeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Inbreeding and Inbreeding Avoidance . . . . . . . . . . . . . . . . . V. Mate Choice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Conclusions and Future Avenues . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

147 149 153 166 170 176 179

Mate Choice and Genetic Quality: A Review of the Heterozygosity Theory BART KEMPENAERS I. II. III. IV. V.

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mate Choice and Genetic Benefits . . . . . . . . . . . . . . . . . . . . Heterozygosity and Fitness . . . . . . . . . . . . . . . . . . . . . . . . . . Heterozygosity and Mate Choice . . . . . . . . . . . . . . . . . . . . . Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

189 191 198 221 260 262

CONTENTS

vii

Sexual Conflict and the Evolution of Breeding Systems in Shorebirds ´ S SZE´KELY, AND GAVIN H. THOMAS, TAMA JOHN D. REYNOLDS I. II. III. IV. V. VI. VII. VIII.

Sexual Conflict and Shorebird Breeding Systems . . . . . . . . Sexual Conflict Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Why Study Shorebirds? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict over Mating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Parental Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Sexual Conflict Framework for Breeding Systems . . . . . Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

279 281 283 288 297 307 308 310 330

Postcopulatory Selection in the Yellow Dung Fly Scathophaga stercoraria (L.) and the Mate-Now-Choose-Later Mechanism of Cryptic Female Choice PAUL I. WARD I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV. XV. XVI.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sexual Conflict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . What is in a Name? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Parker’s Pioneering Work . . . . . . . . . . . . . . . . . . . . . . . . . . . Female Arrival at the Dung Pat and the Evolution of Testes Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cryptic Female Choice, Sperm Competition, and Male–Female Conflict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Variation at the Phosphoglucomutase Locus . . . . . . . . . . . Ejaculate Labeling and Detailed Morphology of the Female Reproductive Tract . . . . . . . . . . . . . . . . . . . . . . . . . Cryptic Female Choice at the PGM Locus . . . . . . . . . . . . . A Selection Experiment on Sexual Conflict . . . . . . . . . . . . Field Experiments on Cryptic Female Choice . . . . . . . . . . Female Accessory Glands and Maternal Effects . . . . . . . . Checking Laboratory Results with Field Flies . . . . . . . . . . Comparative Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

343 344 346 347 347 349 352 354 356 359 360 360 361 362 363 364 365

viii

CONTENTS

The Evolution, Function, and Meaning of Marmot Alarm Communication DANIEL T. BLUMSTEIN I. II. III. IV. V. VI. VII.

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Meaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Individuality and Reliability . . . . . . . . . . . . . . . . . . . . . . . . . Applied Relevance of Alarm-Calling Behavior . . . . . . . . . . Summary and Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

371 372 381 385 388 392 393 395

The Evolution of Geographic Variation in Birdsong JEFFREY PODOS AND PAIGE S. WARREN I. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Evolution of Geographic Variation in Song: Literature Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Assessing Hypotheses of Dialect Evolution . . . . . . . . . . . . . IV. Recent Studies of Avian Vocal Evolution, and How They Support By-product Models of Vocal Geographic Divergence. . . . . . . . . . . . . . . . . . . . . . . . V. Evolution of Geographic Variation in Avian Vocal Signals: Prospectus . . . . . . . . . . . . . . . . . . . . . . VI. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

403

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

459

Contents of Previous Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

477

406 416

433 440 443 444

Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.

LOUISE BARRETT (43), Department of Psychology, University of Lethbridge, Lethbridge, Alberta T1K 3M4, Canada; School of Psychology, University of Kwazulu-Natal, Durban 4041, South Africa TRINE BILDE (83), Department of Biological Sciences, University of Aarhus, Denmark DANIEL T. BLUMSTEIN (371), Department of Ecology and Evolutionary Biology, University of California, Los Angeles, California 90095-1606, USA S. PETER HENZI (43), Department of Psychology, University of Lethbridge, Lethbridge, Alberta T1K 3M4, Canada; School of Psychology, University of Kwazulu-Natal, Durban 4041, South Africa BART KEMPENAERS (189), Department of Behavioural Ecology and Evolutionary Genetics, Max Planck Institute for Ornithology, Postfach 1564, D-82305 Starnberg (Seewiesen), Germany JAN KOMDEUR (147), Animal Ecology Group, Centre for Ecological and Evolutionary Studies, University of Groningen, 9750 AA Haren, the Netherlands YAEL LUBIN (83), Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus, 84990 Israel JEFFREY PODOS (403), Department of Biology, Graduate Program in Organismic and Evolutionary Biology, University of Massachusetts, Amherst, Massachusetts 01003 JOHN D. REYNOLDS (279), Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6 DAVID S. RICHARDSON (147), Centre for Ecology, Evolution and Conservation, School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom JOAN B. SILK (1), Department of Anthropology, University of California, Los Angeles, California 90095, USA ix

x

CONTRIBUTORS

´ S SZE´KELY (279), Department of Biology and Biochemistry, TAMA University of Bath Claverton Down, Bath BA2 7AY, United Kingdom GAVIN H. THOMAS (279), Nerc Centre for Population Biology, Imperial College London, Silwood Park, Ascot, Berkshire SL5 7PY, United Kingdom PAUL I. WARD (343), Zoological Museum of the University of Zurich, 8057 Zurich, Switzerland PAIGE S. WARREN (403), Department of Natural Resources Conservation, Graduate Program in Organismic and Evolutionary Biology, University of Massachusetts, Amherst, Massachusetts 01003

Preface

Advances in the Study of Behavior was initiated over 40 years ago to serve the increasing number of scientists engaged in the study of animal behavior. That number is still expanding. This volume makes another important ‘‘contribution to the development of the field’’ by presenting theoretical ideas and research to those studying animal behavior and to their colleagues in neighboring fields. This volume reflects many of the current themes in animal behavior, including the evolution of social behavior, sexual selection, and communication. It also reflects controversial topics on which the authors provide interesting, new insights. Joan Silk discusses the nature of cooperation and contingent reciprocity among nonkin in primate groups. Peter Henzi and Louise Barrett provide a critical review of female-bonded groups in primates and the role of coalitions and other social behaviors in favoring adaptations such as cognitive abilities. Yael Lubin and Trine Bilde review the characteristics of social spiders and their nonsocial congeners and the evidence for the evolution of sociality in spiders. Using molecular evidence, Jan Komdeur and David Richardson examine studies on cooperative breeding and mate choice in Seychelles warblers. Bart Kempenaers discusses why individual heterozygosity is related to fitness and how this affects the evolution of mate choice. Gavin Thomas, Tama´s Sze´kely, and John Reynolds discuss sexual conflict in the context of shorebird breeding systems, and Paul Ward reviews research on postcopulatory sexual selection in dung flies. Daniel Blumstein reviews studies on alarm calling in marmots, and Jeffrey Podos and Paige Warren critically examine birdsong dialects and provide a new view of the causes and consequences of geographic variation in vocal communication. Most of these studies take a highly integrative approach, examining both the proximate mechanisms and the evolution of the behavior. The chapters cover a diversity of animal taxa, including mammals, birds, spiders, and insects, and they include both laboratory and field studies and both theoretical and empirical approaches. By highlighting particularly exciting research programs that introduce important new concepts, Advances in the Study of Behavior hopes to continue its ‘‘contribution to the development of the field.’’ With this volume, we welcome Dr. John Mitani and Dr. Chris Barnard to our team of editors. Their diverse approaches to the study of behavior will contribute greatly to future volumes. Also with this volume, Dr. Peter Slater and Dr. Charles Snowdon are stepping down as editors after many xi

xii

PREFACE

years of exceptional service. We deeply appreciate all the work they have done. I remain as executive editor and Tim Roper, Marc Naguib, Kathy Wynne-Edwards, John Mitani, and Chris Barnard as editors. Together this diverse group will help maintain the intellectual diversity that has characterized this series from the beginning. H. JANE BROCKMANN

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

The Strategic Dynamics of Cooperation in Primate Groups Joan B. Silk department of anthropology, university of california los angeles, california 90095, usa

I. INTRODUCTION The evolution of cooperation has played an important role in evolutionary biology over the last 40 years. There is a broad consensus that Hamilton’s theory of kin selection (Hamilton, 1964) provides a basis for understanding the deployment of altruism among relatives in a wide range of animal taxa (Crozier and Pamilo, 1996; Dugatkin, 1997; Emlen, 1997; Sachs et al., 2004). In contrast, there is considerably less agreement about the evolutionary processes that regulate the distribution of benefits to nonrelatives. Although Trivers (1971) introduced the theory of reciprocal altruism nearly 35 years ago, researchers have produced relatively few examples of reciprocal altruism in nature (Dugatkin, 2002; Hammerstein, 2003; Noe¨, 2005, 2006). Even the most carefully documented examples, such as predator inspection in schooling fish (Milinski, 1987) and egg‐trading in simultaneous hermaphrodites (Fisher, 1988), have been disputed (Connor, 1992, 1995; Dugatkin, 1997; Hammerstein, 2003; Sachs et al., 2004). This has led some researchers to focus on other processes, such as by‐product mutualism, group augmentation, and market forces, to explain the evolution of cooperation among nonrelatives (Clutton‐Brock, 2002, 2005; Connor, 1986, 1992, 1995; Kokko et al., 2001; Noe¨, 2005, 2006). Here I consider the nature of cooperation among nonkin in nonhuman primate groups, concentrating on evidence for turn‐taking, collaboration, and coordination. Nonhuman primates are an appropriate group of animals to focus on for several reasons. First, nonhuman primates (primates, hereafter) perform a variety of services on behalf of other group members. They groom one another; provide support in agonistic encounters; collectively defend

0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37001-0

1

Copyright 2007, Elsevier Inc. All rights reserved.

2

JOAN B. SILK

access to mates, food resources, and territories; actively donate food or tolerate scrounging; provide alloparental care; and warn group members when they detect predators. Second, primates fulfill a number of the basic requirements for contingent reciprocity: they live in stable social groups, recognize group members as individuals, and have large brains and good memories. Third, there is a rich body of naturalistic and experimental data about the deployment of cooperative behavior in a number of primate species. While there is abundant evidence of kin biases in cooperative behavior (Chapais and Berman, 2004; Silk, 2002a, 2005), the role of contingent reciprocity in primates remains controversial (Barrett and Henzi, 2002, 2005; Noe¨, 2005, 2006; Stevens and Hauser, 2004; Stevens et al., 2005a). Controversy about the forces that shape cooperation among nonrelatives in primate groups persists for at least two different reasons. First, there is considerable disagreement about what the term ‘‘cooperation’’ means. Here I argue that game theory provides a way to avoid unproductive disputes about terminology and to focus our attention on the processes that favor the distribution of benefits to unrelated individuals. Second, controversy persists because it is difficult to measure critical parameters under naturalistic conditions. Thus, much of the evidence that I describe here comes from experimental studies conducted in the field and laboratory. I begin with the first experiments on cooperation which were conducted in the 1930s, and carry on to the present. This body of experimental work provides an important complement to naturalistic observations of cooperation. Well‐designed experiments constrain the range of possible explanations for cooperation; they provide a measure of control over theoretically relevant parameters, including the effort required to solve a task, the benefits acquired when a task is completed successfully, and the temporal sequence of events; and crucially, they make it possible to test whether cooperation is contingent on the previous cooperative behaviors of others. In addition, when experiments using the same protocol are conducted on different groups or species, they provide a standardized baseline for comparative analyses. This literature includes an array of valuable methods and innovative designs for apparatuses that could be adopted or adapted in new experiments.

II. (AVOIDING) DEFINITIONS OF COOPERATION Take care of the sense and the sounds will take care of themselves The Duchess to Alice in Alice in Wonderland

DYNAMICS OF PRIMATE COOPERATION

3

Cooperation means different things to different people. Boyd and Richerson (2006) define cooperation as ‘‘costly behavior performed by one individual that increases the payoff of others.’’ In contrast, Noe¨ (2006) suggests that we use the word cooperation ‘‘for all interactions or series of interactions that, as a rule (or ‘on average’), result in net gain for all participants. The term includes all other terms that have been used for mutually rewarding interactions and relationships: reciprocity, reciprocal altruism, mutualism, symbiosis, collective action and so forth.’’ Sachs et al.’s definition focuses on the behaviors that benefit others, regardless of the effect on the actor (Sachs et al., 2004). Bronstein (2003) differentiates between mutually beneficial interactions with members of the same species and members of other species. She applies the term cooperation to the former and mutualism to the latter. Brosnan and de Waal (2002) define cooperation as ‘‘the voluntary acting together of two or more individuals that brings about, or could potentially bring about, an end situation that benefits one, both, or all of them in a way that could not have been brought about individually.’’ Stevens and Hauser (2004) adopt Clements and Stephens’s (1995) definition of cooperation as ‘‘joint action for mutual benefit.’’ There is one common element of all these definitions of cooperation: benefits are provided to other conspecifics, but no consensus about the impact on the actor. There is no easy way to resolve this semantic muddle, but there is a way around it. Following the Duchess’ advice to Alice, we can make progress by focusing on the dynamics of interactions that confer benefits on other individuals, rather than worrying about what terms we should use to describe them. Game theory provides a useful way to discipline our thinking about the strategic dynamics of these kinds of interactions. There are several different games that represent situations that correspond to various definitions of ‘‘cooperation.’’ These games allow us to explore the conditions under which natural selection will favor individuals who provide benefits to unrelated partners. It should be noted that these games were mainly developed by economists and were meant to apply to real‐life situations. The names that these games were given reflect everyday scenarios which were used as exemplars of the processes involved. For example, the Prisoner’s Dilemma gets its name from a situation in which two prisoners are interviewed separately and are given the choice between informing on their partners in exchange for a lighter sentence or remaining silent. This situation is used to explore the dynamics of altruism (remaining silent) when there is a temptation to defect (informing on the partner). The games have a generality that extends beyond their names. (For more about the logic underlying game theory, see Maynard Smith, 1982; McElreath and Boyd, 2007.)

4

JOAN B. SILK

III. GAME THEORY MODELS OF FORMS OF ‘‘COOPERATION’’ A. ITERATED PRISONER’S DILEMMA The Iterated Prisoner’s Dilemma is the best known of these games. In the one‐shot version of the Prisoner’s Dilemma, players benefit if they both cooperate, but each is better off if the other cooperates and they act selfishly themselves. For example, two monkeys are threatened by a predator, like a leopard. If either one mounts a counterattack, they are likely to drive the predator away. But this is a costly thing to do, as there is some danger associated with confronting the leopard. However, if one holds back while the other chases the leopard, the one that holds back will be protected without incurring the costs of confronting the predator. The same will hold for the other monkey. So, the monkeys face a real dilemma. Not knowing what the other will do, both monkeys have a strong incentive not to confront the leopard themselves. Defectors, who hold back, will always be favored. The payoff matrix for this game is laid out in Table I. The dynamics of the Prisoner’s Dilemma are fundamentally altered when two individuals face the same situation repeatedly. Then, contingent strategies may be favored. That is, one monkey may protect her partner from the predator, as long as her partner protected her before. These kinds of contingent strategies can be sustained as long as b > c (1
1/t), where b is the benefit derived from the other’s helpful act, c is the cost of the helpful act, and t is the likelihood that they will interact again in the future. This repeated process is the foundation of the theory of reciprocal altruism (Axelrod and Hamilton, 1981; Trivers, 1971). Subsequent theoretical work has shown that a variety of contingent strategies, not just strict tit for tat, can be evolutionarily stable under the appropriate conditions.

TABLE I THE PRISONER ’S DILEMMA Player 2 Player 1 a

Cooperate Defect a

Cooperate

Defect

b
c, b
c b,
c


c, b 0, 0

Each individual has the opportunity to cooperate by helping the other individual. Helping increases the payoff of the receiver by two units and reduces the payoff of the actor by one unit.

5

DYNAMICS OF PRIMATE COOPERATION

TABLE II THE STAG HUNT Player 2 Player 1 a

Stag Hare

Stag

Hare

s, s h, 0

0, h h, h

a

Hunters can either hunt stag or hare. Hunting hares together does not affect success; they always get a small payoff, h. If they hunt stag together, they are likely to succeed, and achieve a high payoff, s. A stag hunter who hunts alone fails and receives a payoff of 0.

B. STAG HUNT The Stag Hunt describes a situation in which two individuals can profit from working together. Hunting alone, each individual is only able to catch small game, like hare. But if they hunt together, they are able to take much larger game, such as stags. One stag weighs considerably more than two hares, so it is profitable for two individuals to hunt together and share the kill. Both players earn higher payoffs if they both hunt stags than if only one hunts stag or if they both hunt hares (Table II). Of course, it will not be profitable for individuals to cooperate unless they are likely to obtain a reasonable share of the stag. It is not necessary that both partners get the same payoff, only that each partner gets a bigger payoff from participating in the Stag Hunt together than they would get from hunting hare alone. C. BATTLE OF THE SEXES The Battle of the Sexes game describes another kind of strategic interaction that can lead individuals to provide benefits to others. Imagine that two people would like to be together, but they have different preferences about what to do. One wants to go to the climbing gym, the other wants to watch monkeys at the zoo. If the benefits of being together outweigh the benefits of pursuing their own preferences, then they will make some kind of compromise about where to go (Table III). The compromise may be reached by taking turns, bargaining, or relying on some kind of conventional asymmetry (e.g., age, dominance rank, sex). The Battle of the Sexes reveals the importance of coordination. In this situation, the highest payoff is obtained when both individuals do the same thing, and neither has an incentive to provide misleading information about their intentions. It would make no sense to promise to meet at the climbing gym, and then head for the zoo.

6

JOAN B. SILK

TABLE III THE BATTLE OF THE SEXES Player 2 Player 1 a

Climbing gym Zoo

Climbing gym

Zoo

b, a 0, 0

0, 0 a, b

a

Individuals have to rendezvous at either the climbing gym or the zoo. Both players receive higher payoffs when they end up in the same place. However, player 1 prefers to meet at the climbing gym, while player 2 prefers the zoo.

TABLE IV CHICKEN GAME (ALSO KNOWN AS THE SNOWDRIFT GAME OR HAWK‐DOVE GAME) Player 2 Player 1 a

Wash Do not Wash

Wash b
c, b
c b, b
C

Do not Wash b
c, b 0, 0

a

The dishes need washing. If they are washed both players get a benefit, b. If the two players share the burden of washing, they each pay a cost, c. If only one washes, he pays a bigger cost C. However, b
C > 0, so individuals prefer washing alone to waking up to dirty dishes.

D. GAMES OF CHICKEN Finally, there is one more way that these kinds of problems may be played out. In games with a Chicken or Hawk‐Dove payoff structure, both players benefit from a given outcome, but neither wants to incur the costs associated with securing it. To see how this works, imagine the following domestic dilemma: a couple confronts a sink full of dirty dishes. There are four possible outcomes in this situation: one partner does the dishes, the other partner does the dishes, they do them together, or they leave them in the sink. Suppose that both partners would prefer to do the dishes, even if they have to do them alone, over leaving them in the sink overnight. But both would prefer that the other does the washing up by themselves than wash the dishes together. In this situation, cooperation is individually beneficial, but each partner is better off if they can persuade the other to wash the dishes by themselves (Table IV). This game differs from the Prisoner’s Dilemma in one critical way. Here, each player obtains a higher payoff from cooperating even when its partner defects than if both defect, the opposite of the ordering in the

DYNAMICS OF PRIMATE COOPERATION

7

Prisoner’s Dilemma. But even though each partner is prepared to cooperate, they can still benefit if they can induce the other to perform the altruistic act. This conflict of interest may be resolved by flipping a coin, taking turns, or relying on a conventional asymmetry, such as the person who cooks does not have to do the dishes.

IV. WHICH KINDS OF GAMES DO PRIMATES PLAY? Social life may provide animals with opportunities to play all of the games described above. However, researchers interested in the evolution of ‘‘cooperation’’ have focused mainly on situations that correspond to the Iterated Prisoner’s Dilemma and the Stag Hunt. Below, I review the substantial body of naturalistic and experimental work on reciprocity that has been conducted in the field and laboratory. Next, I describe a body of research which focuses on situations in which animals must work together to achieve joint goals, and loosely approximates the dynamics of the Stag Hunt. Researchers interested in cooperation have given considerably less consideration to coordination problems, although such situations must arise regularly in the everyday lives of group‐living animals. I describe a small body of work which focuses on the dynamics of coordination in primate groups. Finally, I sketch how Games of Chicken may be played in primate groups.

V. THE ITERATED PRISONER’S DILEMMA IN PRIMATE GROUPS A. CONTINGENT RECIPROCITY IN THE NONEXPERIMENTAL SETTINGS In the Iterated Prisoner’s Dilemma, each player cooperates as long as its partner cooperates, although there is considerable latitude in the characteristics of the strategies that may be employed. (Recall the payoff matrix in Table I). In this game, the behavior of each player is influenced by the behavior of the other player in previous moves, and cooperation is therefore contingent on previous cooperation. The first efforts to study contingent reciprocity in primates were based on correlational analyses of dyadic interactions. For example, in one of the first empirical applications of the theory of reciprocal altruism, Packer (1977) showed that male savannah baboons (Papio cynocephalus) are most likely to recruit support from the males that recruited support from them. Subsequently, it has been reported that chimpanzees (Pan troglodytes) share food most often with those that share most often with them (de Waal, 1989;

8

JOAN B. SILK

Mitani, 2005), and female savannah baboons, macaques (Macaca spp.), and capuchins (Cebus spp.) spend the most time grooming females who spend the most time grooming them (Manson et al., 2004; Silk et al., 1999). Grooming may be balanced within bouts (Barrett et al., 1999; Cords, 2002) or across bouts (Manson et al., 2004; Schino et al., 2003). Not all exchanges involve a single currency. Seyfarth (1977) was the first to suggest that support might be exchanged for grooming, and subsequently showed that female vervets (Chlorocebus aethiops) selectively support unrelated females that groom them most often (Seyfarth, 1980). However, the empirical regularity of exchanges between grooming and coalitionary support remains in dispute (Barrett and Henzi, 2002, 2005; Schino, 2001; Stevens et al., 2005). Correlational studies like these suffer from a number of shortcomings. One problem is that the observed associations could be the product of a third factor that has not been measured or taken into account, such as kinship or dominance rank (Hemelrijk and Ek, 1991). Moreover, even if relevant confounding variables can be excluded or controlled statistically, correlational analyses do not address the problem of contingency. That is, would one animal stop grooming its partner if its partner did not respond to solicitations for support or failed to groom in return? If we cannot specify what currencies are being exchanged or assess the value of the commodities being traded, we cannot tell whether the pattern of interactions satisfies the conditions for the Iterated Prisoner’s Dilemma. Even the absence of statistically significant correlations may not rule out reciprocity if one of the exchange commodities, such as coalitionary support, is rare but valuable (Schino, 2001). B. EXPERIMENTAL STUDIES OF CONTINGENT RECIPROCITY The difficulties of studying the processes underlying exchanges in nature lead us to experimental studies in which animals are able to take turns providing benefits to one another. Claims about whether primates practice contingent reciprocal strategies depend on the statistical relationship between the behavior of the two players, so it is important to scrutinize the analyses and results with some care. 1. Naturalistic Experiments on Contingent Reciprocity Seyfarth and Cheney (1984) designed the first explicit experimental test of the theory of reciprocal altruism in primates. They showed that wild vervet monkeys were more attentive to the tape‐recorded screams of unrelated group members if they had been groomed recently by the caller than if

DYNAMICS OF PRIMATE COOPERATION

9

they had not been groomed recently by the same monkey. Interestingly, contingent effects were restricted to unrelated dyads. For close kin, the likelihood of providing support was not influenced by recent grooming. Seyfarth and Cheney were able to control for variation in the quality of social bonds by using a within‐subject design in their analysis. However, because the conflicts were simulated by the experimenters, there was no opportunity for the listeners to come to the caller’s aid in a real dispute, leaving some doubt about the likelihood that they would do so. This problem was addressed in a study of captive long‐tailed macaques (Macaca fasicularis; Hemelrijk, 1994). In this experiment, unrelated monkeys from a large social group were temporarily housed in groups of three. Grooming was artificially induced by dropping a sticky treat onto the back of one of the two higher‐ranking animals. Then, aggression was induced by giving a treat to the lowest‐ranking member of the trio. Hemelrijk found that in all seven of the trios that she tested, monkeys were more likely to provide support for the lower‐ranking female if that female had previously groomed them. The magnitude of this effect ranged from a threefold increase in the likelihood of providing support to a one‐third increase in the likelihood of providing support. Hemelrijk also considered the possibility that grooming per se, regardless of its direction, increased the likelihood of subsequent support. She found that in five of the seven dyads, females were more likely to provide support to females who had previously groomed them than to females that they had groomed themselves. Chimpanzees seem to exchange grooming for access to food (de Waal, 1989), in much the same way that monkeys exchange grooming for agonistic support. To assess the contingency between grooming and food sharing more systematically, de Waal (1997a) observed captive chimpanzees for several hours before and after they were provided with a desirable food item, leafy branches. Chimpanzees that gained possession of the branches sometimes allowed others to take leaves from them. They were especially likely to allow those who had groomed them in the previous hours to share their food. Possessors were also less likely to respond aggressively when chimpanzees that had groomed them recently attempted to obtain food than when others attempted to obtain food. de Waal also examined the possibility that chimpanzees who had been groomed were uniformly more tolerant than chimpanzees who had not been groomed, but this was not the case. Chimpanzees selectively tolerated those that had groomed them. However, the relationship between grooming and tolerance only held for those who did not groom often. For pairs that groomed at higher rates, tolerance did not depend on grooming in the period before provisioning. One interpretation of these data is that some dyads keep track of exchanges over

10

JOAN B. SILK

short timescales, while others monitor exchanges across longer periods. Similar variation in the timescale of reciprocity has been reported for wild white‐faced capuchins (Cebus capucinus; Manson et al., 2004). These studies suggest that vervets, long‐tailed macaques, and chimpanzees are more likely to provide benefits to individual group members if that individual has recently groomed them. One of the main strengths of these three studies is that they assess the contingencies between grooming and other forms of altruism in fairly naturalistic settings. However, these studies were not designed to follow the sequence of interactions within dyads. If contingent reciprocity is operating in these situations, then we would expect support or tolerance by the former recipient of grooming to be associated with subsequent grooming, while the failure to provide support or tolerate sharing would lead to subsequent reductions in grooming. While these studies suggest that monkeys and apes reward cooperators, it is not clear that they subsequently withhold cooperation from noncooperators. 2. Laboratory Experiments on Contingent Reciprocity The first explicit study of turn‐taking in primates that I have found was conducted in the 1930s on monkeys and children (Wofle and Wofle, 1939). In this experiment, two monkeys were placed in adjacent cages. Each monkey could reach a rope that was attached to a lever. When the rope was pulled, the lever pivoted and delivered a grape to the monkey in the adjacent cage. Unfortunately, Wofle and Wofle tested a rather odd set of monkeys and did not always form pairs consisting of animals of the same species. All of the monkeys pulled the levers regularly and delivered rewards to their partners, but there is no information about whether their pulling was contingent on the behavior of their partners. Wofle and Wofle compared the monkeys’ performance when the levers were baited with grapes and when they were not baited, and they also evaluated the monkeys’ performance when there was another monkey in the adjacent cage and when they were alone. Half the monkeys pulled more when the lever was baited than when it was not baited. Of these four monkeys, none differentiated between the presence and absence of another monkey in the adjacent cage. (In parallel experiments on four pairs of children, children who were older than 3 years generally provided rewards for their partners, pulled more when the levers were baited than unbaited, and did not pull the levers when their partner was not present. Two children under the age of 3 years did not make these distinctions.) de Waal and his colleagues have conducted a series of studies of food exchanges in captive brown capuchins, Cebus apella. Their first set of experiments was modeled on Nissen and Crawford’s (1932) experiments with chimpanzees. Nissen and Crawford placed two chimpanzees in

DYNAMICS OF PRIMATE COOPERATION

11

adjacent enclosures and provided food (or tokens that could be fed into a food dispenser) to one of the two individuals. In this situation, possessors sometimes responded positively to begging, but did not often donate food spontaneously. Nissen and Crawford emphasized that dyads that had close social bonds were most likely to exchange food in the experimental setting. Capuchins are good subjects for these kinds of experiments because they are extremely gregarious and are very tolerant of others’ efforts to take bits of their food. de Waal (1997b) placed two capuchins in adjacent enclosures separated by wire mesh. In one series of tests, food bowls were placed at the far ends of the two enclosures, so that only one of the monkeys could reach each bowl (Fig. 1). In the first phase, one bowl was baited with cucumbers. After 20 minutes, the food bowl on the other side was baited with apples. In a second set of tests, foods were presented in the opposite order so that the first monkey got apples and the second monkey got cucumbers (nearly all capuchins prefer apples over cucumbers). When the possessor carried food close to the wire mesh, the monkey on the other side was sometimes able to obtain pieces of the possessor’s food. More than half of the time, the monkey without food waited ‘‘. . . for discarded pieces, collecting or trying to collect them from within the possessor’s reach’’ (de Waal, 1997b). In an additional 40% of cases, the monkey without food obtained food when the possessor was out of reach of the food item or its back was turned. The possessor rarely made deliberate efforts to deliver food to the other monkey, and the possessor did not often overtly resist the other monkey’s attempts to take food. Because the other

Mesh partition

Food bowl

FIG. 1. Cage configuration used in de Waal (1997b) and de Waal (2000). The monkeys are placed in adjoining enclosures divided by a wire mesh partition. The monkeys can reach through the mesh partition but cannot reach the food bowl on the far side of the adjacent cage. The dimensions of the test chamber are 144  60  60 cm3. (Drawing by Frans de Waal after an actual video still.)

12

JOAN B. SILK

monkey is only able to get food when the possessor carried it to the mesh barrier, but the possessor is otherwise fairly passive, de Waal (2000) subsequently characterized capuchin food transfers as ‘‘facilitated taking,’’ rather than food sharing. To determine whether monkeys obtain food selectively from those who obtain food from them, de Waal assessed the pattern of transfers within dyads. When he included all types of transfers in the analysis, the correlation was not significant. But when he excluded the transfers that occurred when the possessor was out of reach of the food item or its back was turned, a significant positive correlation emerged. Thus, monkeys generally obtained the most food from the monkeys who obtained the most food from them. The magnitude and direction of the correlations varied considerably across individuals (range  0.7 to
0.4), with females generally showing stronger and more positive correlations than males. This experiment shares some of the same liabilities as naturalistic correlational studies. High levels of food transfers within dyads might reflect strong mutual social tolerance, not a contingent behavioral strategy (de Waal, 1997b). To address this problem, de Waal (2000) conducted another set of experiments in which he tracked food transfers within dyads across time. In some trials, pairs of monkeys were provided with the same food at the same time, in some they were given access to the same foods in turn, and in some they were given access to different foods at different times. A number of results emerge from this set of experiments. First, although the monkeys are attracted to their partners and generally tolerated their partners’ efforts to take food from them, there are limits to their largesse. Monkeys spent more time near the partition when there was another monkey in the adjoining cage than when they were alone, but they were much less likely to drop whole pieces of food near the partition when there was another monkey in the adjoining cage. Moreover, when one monkey was given more attractive food than the other monkey, the owner of the more attractive food item spent less time near the partition. These data suggest that monkeys weigh the benefits and costs of proximity; when they are likely to lose more than they gain by being near the partition, they spend less time there. However, in the absence of evidence that the possessor would have eaten the food that its neighbor takes, it is not clear whether the loss of food represents a real cost to the possessor. de Waal’s main objective in these experiments was to determine whether food transfers were contingent. By testing the same dyad on multiple occasions, de Waal was able to examine the effects of behavior in one round on behavior in subsequent rounds, a key requirement for examining contingency. He categorized each dyad based on the behavior of the first player in the first trial. If the number of tolerant transfers in the first trial

DYNAMICS OF PRIMATE COOPERATION

13

was higher than the mean number of tolerant transfers in subsequent trials for the same dyad, the dyad was classified as having a high initial tolerant transfer rate. If the number of tolerant transfers in this trial was lower than the mean number of tolerant transfers in subsequent trials for the same dyad, the dyad was classified as having a low initial tolerant transfer rate. Then, de Waal computed the mean number of tolerant transfers in subsequent trials for both types of dyads. He found that dyads with high initial tolerant transfer rates had significantly higher transfer rates in subsequent trials than dyads with low initial tolerant transfer rates. de Waal used this procedure to categorize dyads so that he would be able to correct for differences in the overall levels of tolerant transfers across dyads, and he would be able to exclude the possibility that some dyads have high rates of tolerant transfers simply because they spend a lot of time together and were very tolerant of scrounging. However, this procedure for categorizing dyads is problematic. By definition, the dyads that were classified as having a high initial tolerant transfer rate were more tolerant in earlier trials than later ones, while those that were classified as having low initial tolerant transfer rate were more tolerant in later trials than in initial ones. de Waal’s finding that there was an absolute difference in the number of tolerant transfers among dyads with high and low initial transfer rates does not demonstrate that the amount of tolerance in one trial was positively associated with the amount of tolerance in the next trial, a crucial element required for demonstrating contingency. de Waal and Berger (2000) used a bar pull apparatus to evaluate whether capuchins trade effort for access to food. This apparatus was inspired by Crawford’s (1937) experiments on chimpanzees. de Waal’s version of the apparatus consists of a long tray with one pull bar at each end. The tray can be counterweighted so that it is too heavy for one monkey to pull forward on its own. When the bar is pulled forward, both monkeys can reach rewards placed in food cups on the tray. By manipulating the location and baiting of the food cups, a number of different questions have been addressed. In de Waal and Berger’s experiment, three different treatments were compared. In one treatment, the bar was counterweighted so that it was too heavy for one monkey to pull in alone, but only one bowl was baited. In another treatment, the bar was too heavy for one monkey to pull alone, and both bowls were baited. In a third treatment, one monkey was able to pull the bar in alone and only its own bowl was baited. de Waal and Berger found that the monkeys were considerably less successful in pulling in the bar when only one bowl was baited, but it took two monkeys to pull in the bar (food obtained in 39% trials), than they were when both bowls were baited (89%) or when one monkey was able to obtain food on its own (85%). However, more food was obtained by the

14

JOAN B. SILK

monkey who did not have access to a food bowl when both monkeys worked together to pull in the bar than when one monkey was able to pull in the bar without help. In addition, monkeys were more tolerant of others’ effort to obtain food when help had been provided. Although helping is associated with increased access to food, the magnitude of the effect is relatively small and the direction of causality is unclear. Monkeys who pulled in the food bar by themselves tolerated about 59% of all attempts to take food, while monkeys who had help pulling in the bar tolerated 65% of all attempts. This could mean that monkeys selectively rewarded helpers with tolerance. Alternatively, monkeys may have been more likely to help those that were generally tolerant of them; and tolerance may not have been contingent on prior help. Hattori et al. (2005) trained brown capuchins to complete a two‐step sequence to obtain food rewards. First, the monkeys had to pull a tab in one box, and then they had to push a block in a different box. After the sequence was completed, one reward was released in each of the boxes. Initially, each individual was taught to perform both of the necessary tasks in the appropriate order. Then, the monkeys were tested in pairs. Each individual was confined to a single box, and the two boxes were separated by a transparent window. To obtain rewards, each individual had to perform the appropriate task. Hattori and his colleagues tested three unrelated pairs; all succeeded in the task. In one treatment, Hattori and his colleagues altered the payoff structure, so that only one individual obtained a reward in each trial. The roles of the players alternated between trials. The three pairs succeeded in the majority of trials (70–90%), even though only one individual obtained a reward in each trial. Hattori and his colleagues conclude that these data provide evidence for a ‘‘primitive form of reciprocal altruism.’’ Although the monkeys frequently succeeded in the task, the data do not provide clear evidence of contingency. While successful trials were much more likely to be followed by successful trials (n ¼ 44) than unsuccessful trials (n ¼ 2), the relatively small number of unsuccessful trials (n ¼ 8) were equally likely to be followed by successful trials as unsuccessful trials. (These values do not include trials during one session which the experimenters were forced to rerun, but the pattern does not change substantially when this session is included.) This suggests that these monkeys may have been unilateral cooperators, not contingent reciprocators. Hauser et al. (2003) developed an ambitious experimental protocol to examine contingent behavioral strategies in cotton‐top tamarins, Saguinus oedipus. In these experiments, two tamarins were placed in adjacent enclosures, divided by a partition. One monkey could pull on an L‐shaped bar in front of the cages to bring food items within reach. During the training

DYNAMICS OF PRIMATE COOPERATION

15

phase, the monkeys learned to pull in the bar whenever food was placed on their own side, but never to pull the bar when food was placed on the other side. In one experiment, five tamarins were each paired with two unrelated female ‘‘stooges.’’ (Pairs were composed of members of different groups.) One of these tamarin stooges was trained to always pull the bar, the other stooge was trained to never pull the bar. The pull bar was available to nonstooge subjects and stooges on alternate trials within sessions. In every trial, the bar was baited so that the monkey that pulled the bar provided food to the other monkey, but obtained no food itself. In half the sessions, the stooge made the first move, and in half the sessions the nonstooge subject made the first move. Overall, tamarins pulled more often for the stooge that provided food to them (38%) than for the stooge that did not provide food for them (7%). These differences were apparent from the beginning of each session, suggesting that the tamarins remembered something about the behavior of their partners during previous sessions. Hauser et al. found that the rate of pulling for altruistic stooges declined across trials within each session, but there was no significant decline in the frequency of pulling across sessions. They attribute the decline within sessions to end game effects, which are sometimes observed when humans play repeated games. End game effects probably occur because players anticipate the end of a game and know that they will be able to defect without suffering retribution from their partner. However, when they are observed, end game effects are typically seen only in the very last rounds of games, creating a more precipitous decline in altruism than the tamarins displayed. Moreover, in order to be sensitive to end game effects, players need to know how exactly long the game will last, something that the tamarins are not likely to have known. Thus, this does not provide a completely satisfactory explanation for the decline in the rate of pulling across trials within sessions. Hauser et al. conducted three additional experiments that were intended to determine whether the tamarins were sensitive to the altruistic motivations of their partners. None of these experiments involved the stooges. In one of these experiments, subjects were paired in a game that consisted of 5 sessions of 24 trials. In the first three sessions, the bar was baited so that the subject provided food to its partner, but received no food itself. In the fourth session, the bar was baited so that the subject provided food to itself and its partner. In the fifth session, the bar was baited as in the first three sessions. Hauser et al. reasoned that if monkeys simply pulled when their partners had pulled, the rate of pulling would be higher in session 5 than in the first three sessions. But if the monkeys distinguished between altruistic behavior and self‐interested behavior, the rate of pulling would not

16

JOAN B. SILK

increase. In the first session, tamarins pulled about 38% of the time, and this rate declined to about 25% in the second and third sessions. In the fourth session, when both animals obtained benefits, subjects pulled in all trials. Pulling rates in session 5 were not significantly elevated over the pulling rates in the first three sessions. They conclude that ‘‘. . . to impact upon the level of altruistic food giving, it appears that prior actions must also be altruistic.’’ In the third experiment, the distribution of rewards was changed between trials. In the first trial, the monkey that pulled (Player 1) obtained one food item and the other monkey (Player 2) received three food items. In the next trial, Player 2 was given access to the pull bar. If it pulled, it would obtain no rewards, but Player 1 would obtain two food items. If both monkeys pulled in each round, both would receive three food rewards. The monkeys who took the role of Player 1 pulled in nearly all trials (97%), while the monkeys who took the role of Player 2 rarely pulled (3%). This value is considerably lower than the rate of pulling when paired with unconditional altruists (38%) and lower than the rate of pulling when paired with unconditional defectors (7%) in the first experiment. Hauser et al. interpret this difference to mean subjects are ‘‘sensitive to the costs borne by their partners when deciding whether to pull.’’ Alternatively, monkeys may be ignoring the benefits that their partner obtains, and pull only when they benefit directly. Finally, Hauser et al. evaluated the tamarins’ responses to a partner who did not actively participate in the distribution of rewards. In this case, tamarins were paired with passive partners who never had access to the pull bar. After the test subject pulled, a human experimenter flipped the bar pull around and delivered a reward to the test subject. Thus, the test subject was unconditionally rewarded in every trial, but the rewards were not provided by the partner. In these sessions, the test subjects pulled 10% of time, only slightly more often than they pulled for the stooge that never pulled. Based on these data, Hauser et al. concluded that ‘‘tamarins altruistically give food to genetically unrelated conspecifics, discriminate between altruistic and selfish actions, and give more food to those who give food back. Tamarins therefore have the psychological capacity for reciprocally mediated altruism.’’ This conclusion was modified in a subsequent paper (Stevens and Hauser, 2004) which emphasized the fact that the tamarins pulled less than half the time even when paired with unconditional altruists and that the likelihood of pulling declined across trials within sessions. For these reasons, Stevens and Hauser concluded that tamarins do not demonstrate ‘‘robust reciprocity.’’ However, Chen and Hauser (2005) reanalyzed results from the pairings among nonstooge subjects. They borrowed techniques used in economics to detect the strategies underlying the tamarins behavior. According to their analysis, the tamarins’ behavior most

DYNAMICS OF PRIMATE COOPERATION

17

closely approximates a strategy called ‘‘two tits for a tat,’’ which means that it takes two acts of cooperation by one player to induce the other player to begin cooperating again. C. IS THERE SOLID EVIDENCE OF CONTINGENCY? It is very difficult to detect the existence of contingent reciprocity in naturalistic settings, forcing investigators to rely on statistical evidence of associations between various forms of help given and received within dyads. These kinds of data are problematic because observed associations could be the product of variables that have not been measured. One solution to this problem has been to statistically control for the effects of third variables, such as rates of association. However, this does not solve the problem because it does not identify the causal factors creating the relationship between variables, no matter how many there are. Although there are many reasons to be dissatisfied with correlational evidence, it is important to remember that these correlations could be the product of contingent reciprocal strategies. This is a case in which the absence of evidence should not be mistaken for evidence of absence. Given the continuing controversy over the empirical status of contingent reciprocity in nature and the relevance of primates for understanding the evolutionary roots of cooperation in humans, it is surprising that there have been so few experimental studies of contingent reciprocity in primates. These studies are largely consistent with the prediction that altruism by one member of a dyad individual will increase the likelihood that the other member of the dyad will behave altruistically. However, none of the experiments provide entirely convincing evidence that primates reward cooperators and punish defectors in a consistent way. Being groomed apparently makes long‐tailed macaques and vervets more likely to provide support to their former grooming partners and makes chimpanzees more tolerant of efforts to take food from them. However, these studies do not tell us whether monkeys and chimpanzees also punish defectors by refusing to provide benefits to them in the future. The capuchins that Hattori et al. tested cooperated frequently, even when only one individual obtained a reward, but they did not seem to punish defectors consistently. de Waal and Berger’s conclusion that capuchins trade labor for food rewards may be correct, but their analysis does not establish the contingency between labor and payment. While Hauser’s experiments were explicitly designed to examine contingencies, even Hauser et al. have found the results somewhat difficult to interpret (references above). None of these studies have convinced skeptics that primates regularly practice contingent reciprocity.

18

JOAN B. SILK

D. DO COGNITIVE LIMITATIONS PRECLUDE CONTINGENT RECIPROCITY? Some researchers have argued that cognitive constraints and psychological biases may preclude the deployment of contingent reciprocal strategies, even in highly encephalized animals like primates (Barrett and Henzi, 2002, 2005; de Waal, 2000; Dugatkin, 2002; Stevens and Hauser, 2004; Stevens et al., 2005a; Visalberghi, 1997). For example, primates may lack the ability to assess and keep track of benefits given and received across time and currencies (Barrett and Henzi, 2002, 2005; de Waal, 2000) or wait for delayed rewards (Stevens and Hauser, 2004). Recently, researchers have begun to develop experimental procedures to investigate these capacities in primates. Primates are adept at evaluating the value of resources involved in exchanges (Brosnan and de Waal, 2003; Brosnan et al., 2005; Drapier et al., 2005; Dubreuil et al., 2006; Hyatt and Hopkins, 1998; Padoa‐Schioppa et al., 2006; Sousa and Matsuzawa, 2001). There is a considerable variation in primates’ willingness to wait for delayed rewards (Beran, 2002; Beran et al., 1999; Stevens et al., 2005b), and some species are considerably more patient than others. If such capacities have been subject to natural selection, then it is not obvious that cognitive constraints preclude the capacity for contingent reciprocity.

VI. STAG HUNTS IN PRIMATE GROUPS A. STAG HUNTS IN THE WILD In the Stag Hunt, both players derive greater benefits from working together than either could achieve alone (Table II). There are at least four kinds of interactions that can be interpreted as natural analogues of the Stag Hunt: coalition formation, group hunting, joint mate guarding, and revolutionary coalitions. In East African baboon groups, males form consortships with (or mate guard) receptive females. On the days when females are most likely to conceive, high‐ranking males monopolize access to them. In some cases, two lower‐ranking males jointly challenge the higher‐ranking male and attempt to gain control of the female. These interactions can escalate to energetically costly chases and physical confrontations. The challengers often succeed in driving the higher‐ranking male away, and one of them begins to consort with the female. Packer (1977) found that males most often solicited support from the males that most often solicited support from them, and hypothesized that these interactions were the product of reciprocal altruism. However, he did not examine the temporal sequence of solicitations and subsequent support, and did not demonstrate that support was contingent on prior aid. Subsequent studies of coalition

DYNAMICS OF PRIMATE COOPERATION

19

formation among male baboons indicated that it is often difficult (based on dominance, age, or prior events) to predict in advance which of the two coalition partners will end up with the female (Bercovitch, 1988). This suggests that these coalitions may not conform to a strict tit‐for‐tat model. The Stag Hunt provides a useful way to think about the dynamics of these coalitions. Males who are most likely to form coalitions and challenge consorting males hold middle‐ranking positions and have limited access to receptive females (Bercovitch, 1988; Noe¨, 1990, 1992). If either of the challengers attempted to disrupt a high‐ranking males’ consortship on their own, they would almost surely fail. However, when two mid‐ranking males team up, they become a formidable force. As long as each male has some probability of ending up with the female, it may be profitable for both to participate in the coalition. This may explain why males who form coalitions are typically very close in rank. These coalitions may involve short‐term reciprocal altruism, so that shirking is not possible. If this is the case, then the overall decision about whether to challenge the consorting male has a Stag Hunt payoff structure. Chimpanzees are active and skilled hunters, mainly taking arboreal red colobus monkeys. In the Taı¨ forest, chimpanzees seem to have developed a well‐organized, collaborative strategy for hunting colobus. Hunters take different roles in stalking, ambushing, and capturing prey (Boesch, 1994; Boesch and Boesch, 1989; Boesch and Boesch‐Achermann, 2000). Meat is commonly shared, although hunters receive a larger share than those who do not participate in the hunt. Males typically hunt in groups, and they are considerably more successful in capturing prey when they hunt in groups than when they hunt alone (Boesch, 1994). Chimpanzees in Taı¨ obtain the highest per capita returns when they hunt in groups of four or groups of more than six (Boesch, 1994). Thus, hunting at Taı¨ may conform to the payoff matrix of the Stag Hunt (Boesch et al., 2005). Males who hunt together achieve higher payoffs than males hunting alone and achieve higher payoffs than those who do not contribute to the hunt. At other sites, male chimpanzees do not hunt cooperatively, and hunting in groups does not seem to bring higher per capita returns than hunting alone (Gilby et al., 2006; Mitani et al., 2002; Muller and Mitani, 2005). Capuchins, which hunt squirrels, coatis, and other small animals, do not seem to coordinate their hunting tactics or adopt different roles in hunts (Rose, 1997). Like male baboons, chimpanzees often mate‐guard receptive females (Goodall, 1986; Hasegawa and Hiraiwa‐Hasegawa, 1990; Muller and Mitani, 2005; Nishida, 1979, 1983; Tutin, 1979). In the Kibale Forest, some pairs of high‐ranking males jointly defend access to females (Duffy, 2006; Watts, 1998) and rebuff other males’ attempts to mate. Cooperation allows males to fend off other males’ approaches and keep close tabs on female at the same time. Coalition partners share matings with the female that they

20

JOAN B. SILK

are guarding (Duffy, 2006; Watts, 1998). At Ngogo, participation in these coalitions depends on the level of competition that males face. For males acting alone, the likelihood of successfully controlling access to females declines as the number of males present increases. Males switch from mate guarding alone to joint mate guarding when they are in large parties. Males that switched from solo mate guarding to coalitionary mate guarding earned a ‘‘greater share of copulations than they could have expected from solo mate guarding in those parties (which would have been unsuccessful). Also, males tended to switch from solo mate guarding to two‐male coalitions, and from two‐male to three‐male coalitions, when their expected share of copulations from attempts at solo mate guarding dropped below 50% and 33.3%, respectively’’ (Watts, 1998, p. 52). Some of the coalitions that females form may also conform to the payoff matrix of the Stag Hunt. In species that form matrilineal dominance hierarchies, challenges against higher‐ranking females are extremely rare and dominance hierarchies remain stable over years and even decades (reviewed in Silk, 2002a). This suggests that it may be prohibitively costly for females to confront higher‐ranking females on their own. However, hierarchies are disrupted occasionally when lower‐ranking females mount successful challenges against members of higher‐ranking matrilines. These revolutionary events are often precipitated by collective challenges from lower‐ranking females (Engh et al., 2006; Samuels et al., 1987). While coalitionary aggression by females is normally biased in favor of close kin (Kapsalis, 2004; Silk, 2002a), these revolutionary coalitions often involve short‐term alliances among females from several different matrilines. It seems likely that females who mount collective challenges are more likely to defeat higher‐ranking females than females who act alone.

B. SOLVING COLLABORATION PROBLEMS IN THE LABORATORY A number of experiments that have been designed to probe primates’ propensity for cooperation take the form of the Stag Hunt. Although the protocols of these experiments vary considerably, they are built on the same basic logic: two animals must work together to obtain rewards. If either one does not participate, neither will profit. Because none of these experiments were designed with the Stag Hunt explicitly in mind, some of them do not conform precisely to the payoff matrix of the game. In some cases, the reward is too small to share, and in other cases not all individuals have a realistic chance of obtaining rewards. Nonetheless, these experiments provide interesting insights about the factors that influence performance in interactions that take this basic form.

DYNAMICS OF PRIMATE COOPERATION

21

The first example of this type of experiment was conducted by Meredith Crawford in the 1930s. Crawford (1937) presented five chimpanzees (combined into seven different pairs) with a heavy box that contained food rewards. Two ropes were attached to the box. The box was too far away for the chimpanzees to reach the food placed on it, and too heavy for a single individual to pull forward by themselves. After considerable training, including explicit instruction to pull at the same time, three of the pairs learned to pull the ropes simultaneously without prompting and succeeded in obtaining food regularly. One pair never succeeded in pulling the box forward. (Film footage of these experiments can be viewed at http://www. emory.edu/LIVING_LINKS/crawfordvideo.html.) Two of the four chimpanzees who were members of these four dyads learned to recruit their partners to pull on the ropes and were attuned to their partner’s position and behavior. Unfortunately, data for the three remaining pairs are not presented. Crawford also attached a balance scale to the ropes and measured the duration and force of each partner’s efforts; in most pairs, one partner pulled considerably harder than the other. Crawford also presented the same chimpanzees with two other tasks that required similar types of collaborative effort. None of the chimpanzees were able to solve the new tasks. A few years later, Warden and Galt (1943) posed the box pull task to three pairs of monkeys of different species; none succeeded in working together to obtain rewards. Povinelli and O’Neill (2000) used the box pull paradigm to examine whether experienced chimpanzees would actively recruit, instruct, or shape the behavior of naive partners. They trained two animals to work together successfully and then paired these experts with untrained partners. Only one of the five naive subjects mastered the task, and her success did not seem to be related to the behavior of her expert partners. In an early series of field experiments, monkeys were required to shift a heavy stone to obtain food rewards buried below (Guinea baboons, Papio papio; Fady, 1972; Japanese macaques, Macaca fuscata; Burton, 1977; rhesus and tonkeana macaques, Macaca tonkeana; Petit et al., 1992). In the early stages of the experiments, the stones were light enough for a single individual to move alone. This enabled the monkeys to learn that food was hidden under stones, and allowed the experimenters to calibrate the weight that individual animals could lift. As the experiment proceeded, heavier and heavier stones were used, until the monkeys were unable to shift them on their own. At this point, they could only obtain food if they worked together. The Japanese macaques, rhesus macaques, and baboons virtually never succeeded in working together to displace the heaviest stones, while the tonkeana macaques succeeded only occasionally. Petit et al. attribute the (limited) success of tonkeana macaques to their tolerance

22

JOAN B. SILK

of conspecifics which made it possible for them to work on the same stone at the same time. Although the baboons did not work together, the females and juveniles neatly solved the problem by digging tunnels under the stones, and the adult male simply took food away from them (Fady, 1972). Chalmeau and his colleagues have conducted a series of experiments in which two individuals must simultaneously pull two handles to release a food reward (chimpanzees: Chalmeau, 1994; Chalmeau and Gallo, 1996; orangutans, Pongo pygmaeus: Chalmeau et al., 1997a; brown capuchins: Chalmeau et al., 1997b). Initially, the handles were placed close together and one individual was able to manipulate the apparatus alone. Then the handles were moved apart so one animal was unable to pull both at the same time (Fig. 2). In this phase, joint action by two individuals was required to solve the task. In these experiments, only 1 small food reward was released from the apparatus at a time, but up to 11 rewards were available in a single session. Chalmeau (1994) first tested a group of six chimpanzees. One dyad, which was composed of the dominant adult male and a 2‐year‐old female, produced the great majority of successful responses in the group although others occasionally used the apparatus, the alpha male generally monopolized access to it and commandeered rewards that others managed to obtain. The alpha male and infant did not profit equally from their joint efforts. The dominant male ate 99.5% of the rewards that the infant helped him obtain. Chalmeau et al. (1997a) used the same apparatus with a pair of subadult orangutans. They became adept at manipulating the apparatus and succeeded in obtaining 86% of the available rewards. Again, rewards

FIG. 2. Device used in Chalmeau et al.’s experiments. Two handles are linked to a mechanical device that distributes small food rewards (e.g., sugar lumps or grapes). The handles are too far apart for one individual to pull both at the same time. When both handles are displaced by more than 3 cm at the same time, a food reward drops into an opaque tube and rolls to the edge of the cage. (Redrawn from Fig. 1 in Chalmeau, 1994.)

DYNAMICS OF PRIMATE COOPERATION

23

were not shared evenly within the pair; one of the orangutans obtained 92% of the rewards. Chalmeau et al. (1997b) also used the apparatus to test two small groups of capuchins. Seven of the 11 animals in these 2 groups eventually learned to operate the apparatus. Rewards were not distributed in proportion to effort. Some individuals obtained considerably more rewards than expected (if rewards were distributed evenly among cooperating partners), while others received fewer rewards than expected. In both groups, the oldest and highest‐ranking male obtained the greatest absolute number of rewards and the most rewards per unit effort. Some of the same capuchins were later retested using a different apparatus (Visalberghi et al., 2000). In this experiment, two handles were attached to a long transparent tray. When both handles were pulled at the same time, the tray tilted and released two rewards, one at each end of the tray. Visalberghi and her colleagues individually tested four closely related dyads that were tolerant enough of each other to drink peacefully in proximity. These four pairs all succeeded in obtaining rewards, although there was considerable variation in the rate of pulling within and between pairs and variation in the proportion of pulls that involved joint action by two individuals. In most pairs, one partner pulled considerably more than the other. There was no overt aggression over access to food, but one monkey sometimes took both rewards. Some of the experiments that de Waal and his colleagues have conducted on brown capuchins take the form of the Stag Hunt. In one of the bar pull experiments described earlier, the monkeys were as successful in obtaining rewards when they had to work together as when one monkey was able to obtain food on its own (de Waal and Berger, 2000). de Waal and Davies (2003) used the same device to assess the effects of competition and kinship on performance in tasks that required joint effort. They placed food cups at the distal ends of the tray (dispersed treatment) and they also placed food cups side by side (clumped treatment) in the middle of the tray. To take food from the cups when they were dispersed or clumped, the monkeys had to pull the bar in far enough that it locked into position. In this experiment, the monkeys were tested in pairs, but they were not separated from one another. de Waal and Davies tested six unrelated pairs and five mother–daughter pairs, all from the same social group. The monkeys were more likely to succeed in obtaining food when the distal food cups were baited than when the adjacent food cups were baited. The mother–daughter dyads were more successful than the unrelated dyads, particularly in the clumped trials, and the mother–daughter dyads were also more likely to obtain equal portions of the rewards in both dispersed and clumped trials. Among unrelated dyads, dominants monopolized food in clumped trials and were considerably more likely to pull in these trials than subordinates were.

24

JOAN B. SILK

Brosnan et al. (2006) used the weighted bar pull to examine how the quality of rewards influenced brown capuchins’ success in tasks that required joint effort. They compared three different distributions of rewards: both get two slices of apple, both get one grape, and one gets two slices of apple and the other gets one grape. (All of the subjects preferred one grape over two slices of apple.) Unrelated pairs of monkeys pulled more when they each got a grape or when one got a grape and the other got two slices of apple than when both got apples. Pairs that tended to divide access to valued rewards equitably were considerably more successful in pulling (65%) than pairs that did not divide rewards equitably (26%), and this difference was not an artifact of kinship. Melis et al. (2006a,b) conducted a series of experiments on chimpanzees that were intended to evaluate the effects of social conditions on collaboration in chimpanzees. All of their experiments made use of an apparatus originally designed by Hirata (2003; cited in Melis et al., 2006a) which consisted of a flat platform that slid back and forth on metal rails. A rope was threaded through two holes in the platform so that the platform could be pulled forward when both ends of the rope were pulled at the same time. If only one end of the rope is pulled, the rope comes unthreaded, and the platform cannot be moved (Fig. 3). One individual can pull the platform forward when the ropes are placed close together, but two individuals must work together when the ropes are placed far apart. Food bowls are attached to the platform, and food can only be retrieved when the platform is pulled forward. In their first set of experiments, Melis and her colleagues attempted to replicate Hare and Tomasello’s (2004) finding that competition enhances the effectiveness of cooperation among chimpanzees. They manipulated the distribution of food rewards (clumped or dispersed) and the presence of a competitor (present or absent). When a competitor was present, she was able to pull the platform in the opposite direction, and out of reach of the test pair. In this experiment, 12 individuals were tested in 6 unique dyads. All of the chimpanzees knew how to manipulate the platform when the ropes were placed close together and were strongly motivated to obtain food rewards when they were tested alone. Melis et al. found that three dyads cooperated consistently in all conditions, while three dyads failed consistently in all conditions. The presence of a potential competitor and the clumped distribution of food had no impact on chimpanzees’ performance. Moreover, success or failure did not seem to be linked to the age or sex of individuals. The inconsistency of the chimpanzees’ performance suggested that the relationship between the individuals who had been paired together might be the key to their success. To test this, Melis and her colleagues initiated a second study with a larger number of chimpanzees housed at a chimpanzee

DYNAMICS OF PRIMATE COOPERATION

25

FIG. 3. Device used by Melis et al. (2006a,b). Two food trays are mounted on a wooden platform. A rope is threaded through holes on the distal ends of the platform. If two chimpanzees pull on the rope at the same time, the platform is brought within reach. If only one chimpanzee pulls on the rope, the rope comes unthreaded, and the platform cannot be moved. This device is based on a design by Hirata (2003). (Figure reprinted from Melis et al. (2006a) with permission from Elsevier.)

sanctuary in Uganda. First, Melis et al. evaluated the extent of tolerance while feeding for each dyad, and then she and her colleagues evaluated how successful the same pairs were in retrieving rewards using the same apparatus that they used in the previous experiment. The results indicate that dyads that were most tolerant of one another were also most successful in working together. Moreover, when chimpanzees that had succeeded in the first set of trials were paired with less tolerant partners, their success rate declined. At the same time, when chimpanzees that had failed in the first set of trials were paired with more tolerant partners, they worked together more successfully. All but 1 of the 16 chimpanzees they tested was able to succeed at the task with at least 1 partner. When pairs did fail, it was often because the lower ranking of the two chimpanzees was reluctant to enter the testing room or pull on the rope when the dominant was present. Thus, chimpanzees’ success in this task seems to depend on the quality of their relationship with their partner.

26

JOAN B. SILK

In a second set of experiments, Melis and her colleagues examined how the need for collaboration and the skill of potential collaborators influenced the chimpanzees’ behavior. In these experiments, the test subject was able to admit another chimpanzee from an adjoining enclosure by removing a wooden peg from a hole in the door, which allowed the door to slide open. The experimenters manipulated the need for help by shifting the position of the ropes. In the solo treatment, the ropes were close together so that one chimpanzee could grasp and pull both ropes at the same time. In the collaborative treatment, the ropes were too far apart for one chimpanzee to pull both ropes at the same time. The chimpanzees were significantly more likely to admit the other chimpanzee in the collaborative condition than in the solo condition, suggesting that they could recognize the need for assistance. In the next phase of the project, the experimenters gave the chimpanzees a choice between two potential assistants who varied in their effectiveness in the task at hand. The two assistants were placed in rooms that adjoined the testing room, and the test subject could admit either of them by placing the peg in the appropriate door. In the first testing session, the chimpanzees did not distinguish between the two assistants, but in the second testing session, they showed strong preferences for the more effective assistant. The patterning of the chimpanzees choices across trials suggests that the chimpanzees were likely to switch from one assistant to the other after unsuccessful trials, but were likely to recruit the same assistant again after successful trials. Most of the work on collaboration has involved capuchins or apes, but Cronin et al. (2005) examined collaboration in four mated pairs of cotton‐ top tamarins. To obtain rewards, each of the monkeys had to pull on a counterweighted handle and line up two holes. When the holes were aligned on both sides, food rewards dropped down onto the floor of the enclosure. All four pairs of tamarins succeeded in coordinating their pulls and obtained rewards on nearly every trial. In all of the experiments described so far, both partners perform the same task to earn rewards. Several other protocols have been developed which require each partner to perform complementary tasks in sequence or simultaneously to obtain food rewards. Werdenich and Huber (2002) trained common marmosets (Callithrix jacchus) to pull a rope that was attached to a lever that contained a food bowl at one end. When the rope was pulled, the food bowl swung to within reach. Initially, the marmosets were trained to use the apparatus by themselves. Then, the apparatus was modified so that one had to pull the rope, and the other had to grasp the food bowl and remove the lid. Although the food bowl always contained 10 pieces of food, it was possible for one individual to monopolize the contents of the bowl. Werdenich and Huber created 16 dyads (with 8 different individuals). Inmost of these dyads, one partner consistently pulled the rope and the other partner consistently grasped the food bowl. Eight of the 16 dyads succeeded

DYNAMICS OF PRIMATE COOPERATION

27

in the task when they had to work together. The successful dyads were the ones in which the dominant partner did not monopolize the food rewards, even when it had secured the food rewards alone. As described above, Hattori et al. (2005) evaluated brown capuchin’s capacity to solve a task that required them to perform complementary actions. The monkeys were first trained to perform both parts of the task, then they were paired with a partner and confined to one part of the apparatus. The three pairs that Hattori and his colleagues tested learned to solve the task, each individual performing the appropriate action in their own box. They succeeded in obtaining rewards on virtually every trial. In another series of experiments, individuals must exchange information to obtain rewards. One individual (the Informant) knows where food is located but cannot obtain access to the food itself. The other individual (the Operator) does not know where the food is located but can gain access to the food if it knows where it is located. Both individuals must work together to obtain rewards, and neither has an incentive to lie or defect. This protocol was first used by Mason and Hollis (1962), who designed an apparatus with two trays connected by an expanding accordion‐like device. When the Operator pulled a handle connected to one of the devices, the closest tray moved to within reach of the Operator and the other tray moved to within reach of the Informant. Four parallel sets of trays were available. During the test phases of this experiment, all of the trays were covered so that only the other individual (the Informant) could see which trays contained food rewards. In order to obtain food, the Informant had to provide some kind of information to the Operator about which device was baited. Mason and Hollis found that young rhesus macaques were able to succeed in informing their partners about where food was located and achieved high levels of proficiency in this task. Povinelli et al. (1992a,b) subsequently demonstrated that both rhesus macaques and chimpanzees succeeded at a very similar task when paired with human partners. In some cases, primates can learn to exchange services for food. A young hamadryas baboon (Papio hamadryas) male learned to use an L‐shaped tool to pull a food tray within reach (Beck, 1973). In the experiment, the tool user was confined to one part of the enclosure and could not reach the tool. Other group members had access to the tool, but did not know how to use it. One group member, who was both the tool user’s mate and full sister, developed the habit of bringing the tool to the male and placing it within his reach. This greatly reduced the average amount of time it took for the male to gain access to the food. The rewards were not evenly distributed, but the cooperating pair obtained more food than others. The male ate about three‐quarters of the food, the female ate about 15%, and other group members shared the remainder.

28

JOAN B. SILK

Two language‐trained chimpanzees were trained to exchange food for tools (Savage‐Rumbaugh et al., 1978). One chimpanzee, named Sherman, was trained to name and use several different kinds of tools to obtain food rewards. Sherman used a computer keyboard to request his partner, Austin, to pass him the appropriate tool from an adjacent enclosure. Although the chimpanzees initially had to be coached to make and respond to requests, they became highly proficient at this task, eventually requesting the correct tool and providing the tool that was requested in virtually every trial. In an unspecified fraction of trials, Sherman shared the rewards that he obtained with his helper, Austin. C. WHAT ARE THE NECESSARY CONDITIONS FOR SUCCESS IN THE STAG HUNT A number of the experiments that were designed to examine joint efforts do not conform precisely to the Stag Hunt payoff matrix. These ‘‘design flaws’’ give us some insight about the necessary conditions for joint action. The data suggest that the likelihood of obtaining rewards provides an important incentive to participate in joint tasks. This depends on both how the task is structured and the monkeys’ relationships with their partners. Chalmeau’s experiments on chimpanzees and capuchins were conducted in a group setting, and only one food reward was available on each trial. In the pairs that did work together, rewards were not evenly divided. If high‐ ranking animals monopolize the apparatus or control access to rewards, then other group members may have little opportunity to learn how the apparatus works and little incentive to participate in joint action (Cronin et al., 2005; Melis et al., 2006a; Mendres and de Waal, 2000). Even when Visalberghi et al. (2000) retested some of the same animals in pairs and two rewards were released on each trial, it was possible for one individual to monopolize both rewards. de Waal and Davies (2003) showed that monkeys were more likely to pull together when rewards were dispersed than when they were clumped together and easily monopolized by one individual. Subordinate partners were particularly reluctant to pull when rewards were clumped. Perhaps it is not a coincidence that monkeys worked together more effectively when they were able to see each other, but their rewards could not be appropriated by their partners (Cronin et al., 2005; Hattori et al., 2005). Social tolerance may also facilitate success in joint tasks. Tonkeana macaques, which are well‐known for their pacific temperaments, were able to work on the same stone at the same time, permitting them to occasionally dislodge the heaviest stones (Petit et al., 1992). Similarly, marmoset and

DYNAMICS OF PRIMATE COOPERATION

29

chimpanzee pairs that shared food most equitably when they were able to obtain food without working together were the ones that succeeded most effectively when joint action was required (Melis et al., 2006a; Werdenich and Huber, 2002). Pairs of capuchins that distribute valuable rewards equitably are more likely to succeed in joint tasks than pairs in which one partner monopolizes access to the most valuable rewards (Brosnan et al., 2006). Both Brosnan and her colleagues and de Waal and Davies (2003) found that close kin cooperated more effectively than unrelated pairs and were more likely to divide resources equitably than nonkin. Differences between related and unrelated pairs were most pronounced in situations in which one partner could potentially monopolize access to rewards. Many of these types of experiments were designed to examine the cognitive capacities that underlie cooperation, and some of the variation in performance on these tasks has been linked to variation in cognitive abilities. For example, in Chalmeau’s early experiments on capuchins, the monkeys succeeded in obtaining rewards some of the time, even though they seemed to have little grasp of the requirements of the task. That is, individuals pulled on the handles without regard to the location or behavior of other group members (Chalmeau et al., 1997b). In their follow‐up experiment, Visalberghi et al. (2000) demonstrated that the capuchins were more likely to pull when their partner was on the platform and in reach of the handle than when they were on the platform, but not in reach of the handle. However, they did not time their pulls to coincide with their partners’ pulls. Thus, the capuchins learned to make an association between the task and the location of their partner, but they did not seem to focus on pulling as a critical element of the task. Even so, this was enough to allow the monkeys to pull simultaneously 6–52% of the time. The tamarins that Cronin et al. (2005) tested were more likely to manipulate the levers when their partner was present than when their partner was absent, but required considerable training to master the task. Taken together, these data suggest that monkeys may learn to adjust their own efforts in association with the presence or location of their partners. This association is sometimes sufficient to produce successful results in collaborative tasks. In contrast, the chimpanzees that Melis et al. (2006a,b) tested required very little training to master the double rope apparatus. Variation in success across tasks is sometimes attributed to the nature of the task itself. For example, Mendres and de Waal (2000) suggest that capuchins did not learn to work together in Chalmeau et al.’s (1997) experiments because the ‘‘task may not have been intuitively understandable.’’ Brosnan and de Waal (2002) provide a similar explanation of an

30

JOAN B. SILK

unpublished experiment in which capuchins did not succeed in a collaborative task. Similarly, Visalberghi et al. (2000) suggested that the monkeys might not have succeeded because the apparatus that Chalmeau designed was opaque and ‘‘did not allow the capuchins to see how it functioned.’’ Both Mendres and de Waal (2000) and Visalberghi et al. (2000) did obtain better results in their experiments with what were judged to be more ‘‘intuitive’’ apparatuses. However, in both cases, it was not just the apparatus that was changed. A number of other changes in the experimental protocols were also made. In the original experiment, the monkeys were tested in a group setting and offered a single reward on each trial. Visalberghi and her colleagues individually tested pairs known to get along well together and provided two rewards, rather than just one. Mendres and de Waal (2000) also tested animals in pairs and prevented one from monopolizing the other’s reward. In addition, they altered the structure of the experiment by requiring both animals to work together, but allowing only one to earn a reward. Thus, their experiment conformed more closely to the payoff matrix of an Iterated Prisoner’s Dilemma than a Stag Hunt. Great apes may have a more complete grasp of their partner’s role in these tasks than monkeys do. Although it took many trials before the successful pairs succeeded at Crawford’s box pull task, two of the four chimpanzees that mastered the task began to gesture to their partners to approach the apparatus and sometimes waited until they were near the ropes before they began pulling themselves. In the chimpanzee group that Chalmeau tested, the adult male sometimes tried to collect the infant and bring her to the apparatus (Chalmeau, 1994), and learned to look at the infant before pulling on his own handle (Chalmeau and Gallo, 1996). Chimpanzees reliably distinguish between tasks that require a partner and tasks that they can perform alone (Melis et al., 2006b), suggesting that they have some understanding of the role of their partners. Apes may make use of the ability to take the perspective of others in performing joint tasks. Three of the four chimpanzees that Povinelli et al. (1992a) tested in the Operator/Informant task, immediately succeeded when they switched roles, but monkeys do not seem to be able to do this as readily (Hattori et al., 2005; Mason and Hollis, 1962; Povinelli et al., 1992b). Apes may be quicker to learn how to solve tasks that require joint effort, but it is not entirely clear what exactly they learn about them. The chimpanzees in Crawford’s experiments did not immediately succeed in two other tasks that required joint action (simultaneously pulling on a single rope, simultaneously pulling two handles), suggesting that they had mastered the specific skills necessary to solve the box pull task, but could not generalize their knowledge to other tasks.

DYNAMICS OF PRIMATE COOPERATION

31

VII. THE BATTLE OF THE SEXES IN PRIMATE GROUPS Coordination problems must arise regularly in animal groups. It is likely that the dynamics of the Battle of the Sexes game are in play whenever decisions about group movements or group activities have to be made. Individuals of different age, sex, and reproductive status have quite different metabolic needs and reproductive priorities. This means that they will derive different benefits from feeding on particular foods or participating in alternative activities like resting, grooming, and feeding. The benefits of staying together, which may include the ability to defend resources or reduce vulnerability to predators (Sterck et al., 1997; Wrangham, 1980), apparently override individual preferences about where to go and what to do. For example, female baboons normally give birth overnight and are visibly tired the next day. However, the birth of an infant has no detectable influence on group movements (Altmann, 1980). Unfortunately, we have made little progress in understanding the strategic dynamics of decision making about group movements and activity patterns in the wild (Boinski and Garber, 2000), perhaps because this is a problem that cannot be easily studied in the laboratory or addressed experimentally in the field. The benefits of coordination among group members may be relevant to understanding the evolution of some forms of low‐cost signals in animal groups. Ethological analyses of communication emphasize the importance of signals in coordinating activity in social groups (Markl, 1985), and many of the signals used in these contexts are not costly to produce (Silk, 2002b; Silk et al., 2000). Mountain gorillas (Gorilla berengei berengei) and baboons give soft grunts (Rendall et al., 1999; Stewart and Harcourt, 1994) when they are ready to move from one place to another. Hamadryas baboons use an elaborate series of signals to coordinate group movement and midday reunions (Kummer, 1995; Sigg and Stolba, 1981). In situations like these, there may be conflicts of interest about movement or activities, but a strong incentive for members of the group to stay together. When animals obtain no benefit from deceiving others about their intentions, there is no need for the integrity of signals to be preserved by their high cost. There is also evidence that monkeys use low‐cost vocalizations to provide information to other group members about their disposition and motivation. Female baboons and macaques often give quiet calls as they approach others (Cheney et al., 1995; Silk et al., 2000). In baboons, these calls carry acoustic information about individual identity (Owren et al., 1997) and rudimentary referential information (Rendall et al., 1999). Grunts directed to lower‐ranking females are effective in facilitating affiliation and inhibiting supplants (Cheney et al., 1995; Silk et al., 2000). Female baboons also

32

JOAN B. SILK

grunt to former opponents in the minutes after conflicts have ended. When females grunt after conflicts, they are less likely to resume fighting and more likely to interact affiliatively (Silk et al., 1996). Playback experiments demonstrate that the calls of former aggressors reduce the former opponent’s concern about becoming the target of redirected aggression by the aggressor (Cheney and Seyfarth, 1997; Cheney et al., 1995). In both baboons and macaques, females nearly always give quiet calls before they attempt to handle other females’ infants (Bauers, 1993; Silk et al., 2000, 2003). In rhesus macaques, females who call as they approach are less likely to handle infants roughly than females who do not call (Silk et al., 2000). My colleagues and I have argued that these kinds of situations create coordination problems for individuals (Silk et al., 2000). When a high‐ranking monkey approaches a lower‐ranking monkey, there is some ambiguity about what will happen next. The high‐ranking monkey might solicit grooming from the low‐ranking monkey, attempt to handle her infant, or attack her. Given the high costs of being attacked, subordinate monkeys are likely to flee whenever high‐ranking monkeys approach. This presents a problem for high‐ranking monkeys who want to be groomed or handle infants, and for low‐ranking females who may be forced to give up desirable resting spots or interrupt feeding bouts. Reliable, predictive signals can solve the females’ dilemmas. If high‐ranking females signal that their intentions are benign, low‐ranking females will have no need to flee. Theoretical work demonstrates that honest, low‐cost predictive signals can evolve under a range of conditions (Maynard Smith and Harper, 2003).

VIII. GAMES OF CHICKEN IN PRIMATE GROUPS Games that have a Chicken payoff may also be played in primate groups, but there is little empirical evidence that bears directly on this issue. Various kinds of competitive encounters might approximate the dynamics of the Chicken game. To see how this might work, consider the following scenario. A pair of monkeys, George and Tony, is feeding in a fruiting tree. Another monkey approaches and threatens them. George and Tony are more likely to defeat the interloper if they both fight, but both would prefer to continue feeding while the other fights. If George is prepared to fight, even if Tony does not, then this is a Chicken payoff. George’s decision about whether to feed or fight may be influenced by the benefits he gains from feeding in the fruit tree; if he gets most of the fruit, he might be more willing to defend it. If neither is willing to fight alone, then a collective action problem arises.

DYNAMICS OF PRIMATE COOPERATION

33

The experiments on collaboration reviewed above suggest that animals may sometimes provide benefits to others because it is more beneficial to perform acts that benefit themselves and others than to refrain from acting altogether. For example, Crawford (1937) found that in successful pairs, one chimpanzee often pulled harder than its partner. If the harder working partner obtained more benefits from pulling hard than from matching the lackluster efforts of its partner, then it may be playing a game of chicken.

IX. CONCLUSIONS Over the last 75 years, researchers have employed a diverse set of methods and a large variety of apparatuses in their efforts to work out how, when, and why primates cooperate. Although we can find fault with some of the protocols, choice of subjects, and analytic procedures used in many of the early experiments, these experiments provided valuable methods and ideas that have been incorporated into contemporary work. For example, Meredith Crawford’s experiments with the weighted box were the inspiration for the bar pull device developed in Frans de Waal’s laboratory. The bar pull has proven to be an extremely useful experimental tool that has been adapted for a broad range of species, ranging from chimpanzees, capuchins, and tamarins, to spotted hyenas (Crocuta crocuta; Drea and Frank, 2003). These pioneering studies also contain some useful ideas that have not been widely adopted. For example, Crawford attached a balance scale to the ropes the chimpanzees pulled on and was able to measure each individual’s effort. None of the subsequent experiments based on Crawford’s design have included this valuable bit of information. No one has followed up on the idea of presenting animals in social groups with naturalistic tasks that require joint effort to solve, like moving heavy rocks to obtain buried food items. Naturalistic and experimental data are largely consistent with predictions derived from the theory of reciprocal altruism and the Iterated Prisoners’ Dilemma framework, but none of the experimental or naturalistic studies provide conclusive evidence that primates deploy contingent behavioral strategies. Researchers have presented collaborative problems to tamarins, marmosets, macaques, capuchins, orangutans, and chimpanzees using a variety of different devices. The evidence suggests that their motivation to collaborate is influenced by the likelihood of profiting from their actions, which may in turn be a function of the degree of tolerance among partners. However, not all species have succeeded in all tasks, suggesting that a failure to grasp the role of the partner or the requirements of the task

34

JOAN B. SILK

may limit the deployment of collaborative strategies. However, it is not clear that such constraints necessarily preclude the evolution of collaboration in nature. Many primates have specialized vocalizations and gestures that they use to recruit support in coalitionary aggression, to solicit grooming, and to invite others to play. Moreover, they are adept at directing these overtures to those who are most likely to respond positively (Perry et al., 2004; Schino et al., 2006; Silk, 1999). In these social contexts, both monkeys and apes act as if they understand that their goals (such as being groomed or thwarting an attack) depend on the behavior of their partners. Although it is useful to conduct experimental studies of cooperation in the laboratory, the ultimate goal is to understand the evolutionary forces that shape cooperation in nature. So, it is important to consider what the evidence from the laboratory tells us about the nature of cooperation in the wild. The lack of robust empirical evidence for contingent reciprocity could mean that reciprocal altruism is not the principal mechanism underlying cooperation among unrelated individuals in primate groups, and suggests that alternate mechanisms such as by‐product mutualism, group augmentation, and market forces may be at work (Clutton‐Brock, 2002; Connor, 1986, 1992, 1995; Kokko et al., 2001; Noe¨, 2006). At the same time, there is little evidence showing that primates consistently fail in naturalistic tasks that require them to take turns giving and receiving rewards. In fact, the naturalistic experiments on interchange in vervets, macaques, and chimpanzees provide the most compelling empirical evidence for contingent reciprocity in nature. Success in collaborative tasks is, as expected, strongly influenced by the prospects for obtaining rewards. This suggests that the expectations about the distribution of rewards may strongly influence the prospects for joint activity under natural conditions. Additional experimental work on these questions is badly needed. These data can provide an important complement to naturalistic observations of behavior. The conceptual framework outlined here may provide the basis for fruitful discussions of the evolutionary forces shaping cooperative behavior. Examination of strengths and limitations in existing empirical work may help researchers find ways to design new experiments and conduct observations that yield more conclusive findings.

X. SUMMARY The processes that influence the distribution of benefits to other group members have been central themes in evolutionary biology over the last 30 years. There is a broad consensus that kin selection underlies altruistic

DYNAMICS OF PRIMATE COOPERATION

35

behavior among genetic relatives and that nepotistic altruism is common in nature. In contrast, there is much less agreement about the processes that underlie beneficent behavior toward nonrelatives or the importance of such processes in nature. Early enthusiasm for reciprocal altruism has given way to caution, as few robust examples of contingent reciprocity, turn‐taking, and exchange have been found in nonhuman animal species. This has generated interest in a broader range of processes, such as coordination and collaboration, that may favor helpful behavior toward nonkin. Here, I review observational and experimental evidence about turn‐taking, collaboration, and coordination in primate groups. I describe four game‐ theoretic models in which individuals can benefit from providing help to others (Iterated Prisoner’s Dilemma, Stag Hunt, Battle of the Sexes, and Games of Chicken) and argue that the formal payoff structures derived from these models provide a clear and cogent framework for understanding the processes underlying various types of cooperative interactions. I use this framework to review observational and experimental evidence of cooperation among primates. Research conducted by behavioral ecologists has focused mainly on turn‐taking, and studied situations that correspond loosely to the Iterated Prisoner’s Dilemma model. Naturalistic research documents the pattern of associations between help given and received and explores the temporal span of reciprocity and the currencies exchanged. Experimental work on the Iterated Prisoner’s Dilemma has been designed to illuminate the contingencies between giving and receiving benefits. Evidence for contingent reciprocity in nature and the laboratory is inconclusive for primates, leading some to suggest that cognitive constraints on memory and psychological biases that favor immediate rewards over delayed rewards preclude contingent reciprocity. Research conducted by comparative psychologists has mainly focused on the cognitive and psychological capacities that enable animals to cooperate effectively. This work focuses on situations in which animals must work together to achieve a joint reward, a situation that corresponds loosely to the payoff structure of the Stag Hunt model. Primates succeed in some collaborative tasks in the laboratory, but not in others. Success seems to be facilitated by tolerance of partners, expectation of obtaining rewards, and having some understanding of the requirements of the task. Coordination problems, which correspond loosely to the payoff matrix of the Battle of the Sexes game, have been given considerably less attention, although animals must solve many coordination problems in the course of their everyday lives. A limited body of evidence suggests that some primates have evolved vocalizations which serve as honest signals of intent, and facilitate coordination of group movements and social interactions.

36

JOAN B. SILK

Acknowledgments The experimental work that stimulated my thinking on this topic was supported by a grant from the MacArthur Foundation Preferences Network. I thank Rob Boyd, Ernst Fehr, Dan Fessler, Josep Call, Katherine Cronin, Joe Henrich, Sarah Brosnan, Daniel Povinelli, and Jennifer Vonk for stimulating conversations about many of the ideas explored in this chapter. I thank Tim Clutton‐Brock, Keith Jensen, Alicia Melis, and Jeff Stevens for useful comments on an earlier draft of this chapter. This chapter was written while I was a Visiting Professor in the Department of Zoology and the Leverhulme Centre for Human Evolutionary Studies, and a Visiting Professorial Fellow of Magdalene College at Cambridge University. I thank my hosts for their hospitality and fellowship during my stay in Cambridge.

References Altmann, J. (1980). ‘‘Baboon Mothers and Infants.’’ Harvard, Cambridge, MA. Axelrod, R., and Hamilton, W. D. (1981). The evolution of cooperation. Science 211, 1390–1396. Barrett, L., and Henzi, S. P. (2002). Constraints on relationship formation among female primates. Behaviour 139, 263–289. Barrett, L., and Henzi, S. P. (2005). Monkeys, markets and minds: Biological markets and primate sociality. In ‘‘Cooperation in Primates and Humans’’ (P. M. Kappeler and C. P. van Schaik, Eds.), pp. 209–232. Springer, Berlin. Barrett, L., Henzi, S. P., Weingrill, T., Lycett, J., and Hill, R. A. (1999). Market forces predict grooming reciprocity in female baboons. Proc. Roy. Soc. B. 266, 665–670. Bauers, K. A. (1993). A functional analysis of staccato grunt vocalizations in the stumptailed macaque (Macaca arctoides). Ethology 94, 147–161. Beck, B. B. (1973). Cooperative tool use by captive hamadryas baboons. Science 182, 594–597. Beran, M. J. (2002). Maintenance of self‐imposed delay of gratification by four chimpanzees (Pan troglodytes) and an orangutan (Pongo pygmaeus). J. Gen. Psychol. 129, 49–66. Beran, M. J., Savage‐Rumbaugh, E. S., Pate, J. L., and Rumbaugh, D. M. (1999). Delay of gratification in chimpanzees (Pan troglodytes). Dev. Psychobiol. 34, 119–127. Bercovitch, F. B. (1988). Coalitions, cooperation and reproductive tactics among adult male baboons. Anim. Behav. 36, 1196–1209. Boesch, C. (1994). Cooperative hunting in wild chimpanzees. Anim. Behav. 48, 653–667. Boesch, C., and Boesch, H. (1989). Hunting of wild chimpanzees in the Taı¨ National Park. Amer. J. Phys. Anthropol. 78, 547–573. Boesch, C., and Boesch‐Achermann, H. (2000). ‘‘The Chimpanzees of the Taı¨ Forest.’’ Oxford University Press, Oxford. Boesch, C., Boesch, H., and Vigilant, H. (2005). Cooperative hunting in chimpanzees: Kinship or mutualism. In ‘‘Cooperation in Primates and Humans’’ (P. M. Kappeler and C. P. van Schaik Eds.), pp. 139–150. Springer, Berlin. Boinski, S., and Garber, P. A. (Eds.) (2000). ‘‘On the Move. How and Why Animals Travel in Groups.’’ University of Chicago Press, Chicago. Boyd, R., and Richerson, P. J. (2006). Culture and the evolution of human social instincts. In ‘‘Roots of Human Sociality’’ (S. Levinson and N. Enfield, Eds.), pp. 453–477. Berg, Oxford. Bronstein, J. L. (2003). The scope for exploitation within mutualistic interactions. In ‘‘Genetic and Cultural Evolution of Cooperation’’ (P. Hammerstein, Ed.), pp. 185–203. MIT Press, Cambridge, MA.

DYNAMICS OF PRIMATE COOPERATION

37

Brosnan, S. F., and de Waal, F. B. M. (2002). A proximate perspective on reciprocal altruism. Hum. Nat. 13, 129–152. Brosnan, S. F., and de waal, F. B. M. (2003). Monkeys reject unequal pay. Nature 425, 297–299. Brosnan, S. F., Schiff, H. C., and de Waal, F. B. M. (2005). Tolerance for inequity may increase with social closeness in chimpanzees. Proc. R. Soc. B. 272, 253–258. Brosnan, S. F., Freeman, C., and de Waal, F. B. M. (2006). Partner’s behavior, not reward distribution, determines success in an unequal cooperative task in capuchin monkeys. Amer. J. Primatol. 68, 713–724. Burton, J. J. (1977). Absence de comportement coope´ratif spontane´ dans une troupe de Macaca fuscata en pre´sence de pierres appaˆte´es. Primates 18, 358–366. Chalmeau, R. (1994). Do chimpanzees cooperate in a learning task? Primates 35, 385–392. Chalmeau, R., and Gallo, A. (1996). What chimpanzees (Pan troglodytes) learn in a cooperative task. Primates 37, 39–47. Chalmeau, R., Lardeux, K., Brandibas, P., and Gallo, A. (1997a). Cooperative problem solving by orangutans (Pongo pygmaeus). Int. J. Primatol. 18, 23–32. Chalmeau, R., Visalberghi, E., and Gallo, A. (1997b). Capuchin monkeys, Cebus paella, fail to understand a cooperative task. Anim. Behav. 54, 1215–1225. Chapais, B., and Berman, C. M. (Eds.) (2004). ‘‘Kinship and Behavior in Primates.’’Oxford University Press, Oxford. Chen, M. K., and Hauser, M. (2005). Modeling reciprocation and cooperation in primates: Evidence for a punishing strategy. J. Theor. Biol. 235, 5–12. Cheney, D. L., and Seyfarth, R. M. (1997). Reconciliatory grunts by dominant females influence victims behaviour. Anim. Behav. 54, 409–418. Cheney, D. L., Seyfarth, R. M., and Silk, J. B. (1995). The role of grunts in reconciling opponents and facilitating interactions among adult female baboons. Anim. Behav. 50, 249–257. Clements, K. C., and Stephens, D. W. (1995). Testing models of non‐kin cooperation: Mutualism and the prisoner’s dilemma. Anim. Behav. 50, 527–535. Clutton‐Brock, T. H. (2002). Breeding together: Kin selection and mutualism in cooperative vertebrates. Science 296, 69–72. Clutton‐Brock, T. H. (2005). Cooperative breeding in mammals. In ‘‘Cooperation in Primates and Humans’’ (P. M. Kappeler and C. P. van Schaik, Eds.), pp. 173–190. Springer, Berlin. Connor, R. C. (1986). Pseudo‐reciprocity: Investing in mutualism. Anim. Behav. 34, 1562–1584. Connor, R. C. (1992). Egg–trading in simultaneous hermaphrodites–an alternative to tit–for–tat. J. Evol. Biol. 5, 523–528. Connor, R. C. (1995). The benefits of mutualism: A conceptual framework. Biol. Rev. 70, 427–457. Cords, M. (2002). Friendship among adult female blue monkeys (Cercopithecus mitis). Behaviour 139, 291–314. Crawford, M. P. (1937). The coo¨perative solving of problems by young chimpanzees. Comp. Psychol. Monographs 14, 1–88. Cronin, K. A., Kurian, A. V., and Snowdon, C. T. (2005). Cooperative problem solving in a cooperatively breeding primate (Saguinus oedipus). Anim. Behav. 69, 133–142. Crozier, R. H., and Pamilo, P. (1996). ‘‘Evolution of Social Insect Colonies: Sex Allocation and Kin Selection.’’ Oxford University Press, Oxford. de Waal, F. B. M. (1989). Food–sharing and reciprocal obligations in chimpanzees. J. Hum. Evol. 18, 433–459. de Waal, F. B. M. (1997a). The chimpanzee’s service economy: Food for grooming. Evol. Hum. Behav. 18, 375–386. de Waal, F. B. M. (1997b). Food transfers through mesh in brown capuchins. J. Comp. Psychol. 111, 370–378.

38

JOAN B. SILK

de Waal, F. B. M. (2000). Attitudinal reciprocity in food sharing among brown capuchin monkeys. Anim. Behav. 60, 253–261. de Waal, F. B. M., and Berger, M. L. (2000). Payment for labour in monkeys. Nature 404, 563. de Waal, F. B. M., and Davies, J. M. (2003). Capuchin cognitive ecology: Cooperation based on projected returns. Neuropsychologia 41, 221–228. Drapier, M., Chauvin, C., Dufour, V., Uhlrich, P., and Thierry, B. (2005). Food–exchange with humans in brown capuchin monkeys. Primates 46, 241–248. Drea, C. M., and Frank, L. G. (2003). The social complexity of spoted hyenas. In ‘‘Animal Social Complexity: Intelligence, Culture, and Individualized Societies’’ (F. B. M. de Waal and P. L. Tyack, Eds.), pp. 121–148. Harvard University Press, Cambridge, MA. Dubreuil, D., Gentile, M. S., and Visalberghi, E. (2006). Are capuchin monkeys (Cebus apella) inequity averse? Proc. R. Soc. B. 273, 1223–1228. Duffy, K. (2006). ‘‘Social Dynamics Among Male Chimpanzees: Adaptive Significance of Male Bonds.’’ D. Phil. Thesis, University of California, Los Angeles. Dugatkin, L. A. (1997). ‘‘Cooperation among Animals.’’ Oxford University Press, Oxford. Dugatkin, L. A. (2002). Animal cooperation among unrelated individuals. Naturwissenschaften 89, 533–541. Emlen, S. T. (1997). Predicting family dynamics in social vertebrates. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), 4th ed., pp. 228–253. Blackwell Scientific, Oxford. Engh, A. L., Beehner, J. C., Berman, T. J., Whitten, P. L., Hoffmeir, R. R., Seyfarth, R. M., and Cheney, D. L. (2006). Female hierarchy instability, male immigration, and infanticide increase glucocorticoid levels in female chacma baboons. Anim. Behav. 71, 1227–1237. Fady, J. C. (1972). Absence de cooperation de type instrumental en milieu naturel chez Papio papio. Behaviour 43, 157–164. Fisher, E. (1988). Simultaneous hermaphroditism, tit–for tat, and the evolutionary stability of social systems. Ethol. Sociobiol. 9, 119–136. Gilby, I. C., Eberly, L. E., Pintea, L., and Pusey, A. E. (2006). Ecological and social influences on the hunting behaviour of wild chimpanzees (Pan troglodytes schweinfurthii). Anim. Behav. 72, 169–180. Goodall, J. (1986). ‘‘The Chimpanzees of Gombe: Patterns of Behaviour.’’ Belknap Press, Cambridge, MA. Hamilton, W. D. (1964). The genetical evolution of social behavior. I and II. J. Theor. Biol. 7, 1–52. Hammerstein, P. (2003). Why is reciprocity so rare in social animals? A Protestant appeal. In ‘‘Genetic and Cultural Evolution of Cooperation’’ (P. Hammerstein, Ed.), pp. 83–94. MIT Press, Cambridge, MA. Hare, B., and Tomasello, M. (2004). Chimpanzees are more skillful in competitive than in cooperative cognitive tasks. Anim. Behav. 68, 571–581. Hasegawa, T., and Hiraiwa‐Hasegawa, M. (1990). Sperm competition and mating behavior. In ‘‘The Chimpanzees of the Mahale Mountains’’ (T. Nishida, Ed.), pp. 115–132. University of Tokyo Press, Tokyo. Hattori, Y., Kuroshima, H., and Fujito, K. (2005). Cooperative problem solving by tufted capuchin monkeys (Cebus apella): Spontaneous division of labor, communication, and reciprocal altruism. J. Comp. Psychol. 119, 335–342. Hauser, M. D., Chen, M. K., Chen, F., and Chuang, E. (2003). Give unto others: Genetically unrelated cotton–top tamarins monkeys preferentially give food to those who altruistically give food back. Proc. R. Soc. B. 270, 2363–2370. Hemelrijk, C. K. (1994). Support for being groomed in long–tailed macaques, Macaca fasicularis. Anim. Behav. 48, 479–481.

DYNAMICS OF PRIMATE COOPERATION

39

Hemelrijk, C. K., and Ek, A. (1991). Reciprocity and interchange of grooming and ‘support’ in captive chimpanzees. Anim. Behav. 41, 923–935. Hirata, S. (2003). Cooperation in chimpanzees (in Japanese). Hattatsu 95, 103–111. Hyatt, C. W., and Hopkins, W. D. (1998). Interspecies object exchange: Bartering in apes? Behav. Processes 42, 177–187. Kapsalis, E. (2004). Matrilineal kinship and primate behavior. In ‘‘Kinship and Behavior in Primates’’ (B. Chapais and C. M. Berman, Eds.), pp. 153–176. Oxford, Oxford University Press. Kokko, H., Johnstone, R. A., and Clutton‐Brock, T. H. (2001). The evolution of cooperative breeding through group augmentation. Proc. R. Soc. B. 268, 187–196. Kummer, H. (1995). ‘‘In Quest of the Sacred Baboon.’’ Princeton University Press, Princeton, NJ. Manson, J. H., Navarette, C. D., Silk, J. B., and Perry, S. (2004). Time‐matched grooming in female primates? New analyses from two species. Anim. Behav. 67, 493–500. Markl, H. (1985). Manipulation, modulation, information, cognition: Some of the riddles of communication. In ‘‘Experimental Behavioral Ecology and Sociobiology’’ (B. Holldobler and M. Landauer, Eds.), pp. 163–194. Gustav Fisher Verlag, Stuttgart. Mason, W. A., and Hollis, J. H. (1962). Communication between young rhesus macaques. Anim. Behav. 10, 211–221. Maynard Smith, J. (1982). ‘‘Evolution and the Theory of Games.’’ Cambridge University Press, Cambridge. Maynard Smith, J., and Harper, D. (2003). ‘‘Animal Signals.’’ Oxford University Press, Oxford. McElreath, R., and Boyd, R. (2007). ‘‘Mathematical Models of Social Evolution: A Guide for the Perplexed.’’ University of Chicago Press, Chicago. Melis, A. P., Hare, B., and Tomasello, M. (2006a). Engineering cooperation in chimpanzees: Tolerance constraints on cooperation. Anim. Behav. 72, 275–286. Melis, A. P., Hare, B., and Tomasello, M. (2006b). Chimpanzees recruit the best collaborators. Science 311, 1297–1300. Mendres, K. A., and de Waal, F. B. M. (2000). Capuchins do cooperate: The advantage of an intuitive task. Anim. Behav. 60, 523–529. Milinski, M. (1987). TIT FOR TAT in sticklebacks and the evolution of cooperation. Nature 325, 433–437. Mitani, J. C. (2005). Reciprocal exchange in chimpanzees and other primates. In ‘‘Cooperation in Primates and Humans’’ (P. M. Kappeler and C. P. van Schaik, Eds.), pp. 107–119. Springer, Berlin. Mitani, J. C., Merriwether, D., and Zhang, C. (2002). Male affiliation, cooperation and kinship in wild chimpanzees. Anim. Behav. 59, 885–893. Muller, M. N., and Mitani, J. C. (2005). Conflict and cooperation in chimpanzees. Adv. Stud. Behav. 35, 275–331. Nishida, T. (1979). The social structure of chimpanzees of the Mahale Mountains. In ‘‘The Great Apes’’ (D. A. Hamburg and E. R. McCown, Eds.), pp. 73–121. Benjamin/Cummings, Menlo Park, CA. Nishida, T. (1983). Alpha status and agonistic alliance in wild chimpanzees (Pan troglodytes Schweinfurthii). Primates 24, 318–336. Nissen, H., and Crawford, M. (1932). A preliminary study of foodsharing behaviour in young chimpanzees. J. Comp. Psychol. 22, 383–419. Noe¨, R. (1990). A veto game played by baboons: A challenge to the use of the Prisoner’s Dilemma as a paradigm for reciprocity and cooperation. Anim. Behav. 39, 78–90. Noe¨, R. (1992). Alliance formation among male baboons: Shopping for profitable partners. In ‘‘Coalitions and Alliances in Humans and Other Animals’’ (A. H. Harcourt and F. M. B. de Waal, Eds.), pp. 285–321. Oxford University Press, Oxford.

40

JOAN B. SILK

Noe¨, R. (2005). Digging for the roots of trading. In ‘‘Cooperation in Primates and Humans’’ (P. M. Kappeler and C. P. van Schaik, Eds.), pp. 233–261. Springer, Berlin. Noe¨, R. (2006). Cooperation experiments: Coordination through communication versus acting apart together. Anim. Behav. 71, 1–18. Owren, M. J., Seyfarth, R. M., and Cheney, D. L. (1997). The acoustic features of vowel–like grunts in chacma baboons (Papio cynocephalus ursinus): Implications for production process and functions. J. Acoust. Soc. Am. 101, 2951–2963. Packer, C. P. (1977). Reciprocal altruism in Papio anubis. Nature 265, 441–443. Padoa‐Schioppa, C., Jandolo, L., and Visalberghi, E. (2006). Multi‐stage mental process for economic choice in capuchins. Cognition 99, B1–B13. Perry, S., Barrett, H. C., and Manson, J. (2004). White‐faced capuchin monkeys exhibit triadic awareness in their choice of allies. Anim. Behav. 67, 165–170. Petit, O., Desportes, C., and Thierry, B. (1992). Differential probability of ‘‘coproduction’’ in two species of macaque (Macaca tonkeana, M. mulatta). Ethology 90, 107–120. Povinelli, D. J., and O’Neill, D. K. (2000). Do chimpanzees use gestures to instruct each other? In ‘‘Understanding other minds. Perspectives from Developmental Cognitive Neuroscience’’ (S. Baron‐Cohen, H. Tager‐Flusberg, and D. J. Cohen, Eds.), pp. 459–487. Oxford University Press, Oxford. Povinelli, D. J., Nelson, K. E., and Boysen, S. T. (1992a). Comprehension of role reversal in chimpanzees: Evidence of empathy? Anim. Behav. 43, 633–640. Povinelli, D. J., Parks, K. A., and Novak, M. A. (1992b). Role reversal by rhesus monkeys, but no evidence of empathy. Anim. Behav. 44, 269–281. Rendall, D., Seyfarth, R. M., Cheney, D. L., and Owren, M. J. (1999). The meaning and function of grunt variants in baboons. Anim. Behav. 57, 583–592. Rose, L. (1997). Vertebrate predation and food–sharing in Cebus and Pan. Int. J. Primatol. 18, 727–765. Sachs, J. L., Mueller, U. G., Wilcox, T. P., and Bull, J. J. (2004). The evolution of cooperation. Q. Rev. Biol. 79, 135–160. Samuels, A., Samuels, A., Silk, J. B., and Altmann, J. (1987). Continuity and change in dominance relationships among female baboons. Anim. Behav. 35, 785–793. Savage‐Rumbaugh, E. S., Rumbaugh, D. M., and Boysen, S. (1978). Linguistically mediated tool use and exchange by chimpanzees. Behav. Brain Sci. 4, 539–554. Schino, G. (2001). Grooming, competition and social rank among female primates: A meta–analysis. Anim. Behav. 62, 265–271. Schino, G., Tiddi, B., and di Sorrentino, P. (2006). Simultaneous classification by rank and kinship in Japanese macaques. Anim. Behav. 71, 1069–1074. Schino, G., Ventura, R., and Troisi, A. (2003). Grooming among female Japanese macaques: Distinguishing between reciprocation and interchange. Behav. Ecol. 14, 887–891. Seyfarth, R. M. (1977). A model of social grooming among adult female monkeys. J. Theor. Biol. 65, 671–698. Seyfarth, R. M. (1980). The distribution of grooming and related behaviours among adultfemale vervet monkeys. Anim. Behav. 28, 798–813. Seyfarth, R. M., and Cheney, D. L. (1984). Grooming, alliances, and reciprocal altruism in vervet monkeys. Nature 308, 541–543. Sigg, H., and Stolba, A. (1981). Home range and daily march in a hamadryas baboon troop. Folia Primatol. 36, 40–75. Silk, J. B. (1999). Male bonnet macaques use information about third party rank relationships to recruit allies. Anim. Behav. 58, 45–51. Silk, J. B. (2002a). Kin selection in primate groups. Int. J. Primatol. 23, 849–875.

DYNAMICS OF PRIMATE COOPERATION

41

Silk, J. B. (2002b). Grunts, girneys, and good intentions: The origins of strategic commitment in nonhuman primates. In ‘‘Commitment: Evolutionary Perspectives’’ (R. Nesse, Ed.), pp. 138–157. Russell Sage Press, New York. Silk, J. B. (2005). Practicing Hamilton’s rule: Kin selection in primate groups. In ‘‘Cooperation in Primates and Humans: Mechanisms and Evolution’’ (P. M. Kappeler and C. P. van Schaik, Eds.), pp. 21–42. Springer, Berlin. Silk, J. B., Cheney, D. L., and Seyfarth, R. M. (1996). The form and function of post‐conflict interactions among female baboons. Anim. Behav. 52, 259–268. Silk, J. B., Cheney, D. L., and Seyfarth, R. M. (1999). The structure of social relationships among female savannah baboons in Moremi Reserve, Botswana. Behaviour 136, 679–703. Silk, J. B., Kaldor, E., and Boyd, R. (2000). Cheap talk when interests conflict. Anim. Behav. 59, 423–432. Silk, J. B., Rendall, D., Cheney, D. L., and Seyfarth, R. M. (2003). Natal attraction in adult female baboons (Papio cynocephalus ursinus) in the Moremi Reserve, Botswana. Ethology 109, 627–644. Sousa, C., and Matsuzawa, T. (2001). The use of tokens as rewards and tools by chimpanzees (Pan troglodytes). Anim. Cogn. 4, 213–221. Sterck, E. H. M., Watts, D. P., and van Schaik, C. P. (1997). The evolution of female social relationships in nonhuman primates. Behav. Ecol. Sociobiol. 41, 291–309. Stevens, J. R., and Hauser, M. D. (2004). Why be nice? Psychological constraints on the evolution of cooperation. Trends Cogn. Sci. 8, 60–65. Stevens, J. R., Cushman, F. A., and Hauser, M. D. (2005a). Evolving the psychological mechanisms for cooperation. Annu. Rev. Ecol. Syst. 36, 499–518. Stevens, J. R., Hallinan, E. V., and Hauser, M. D. (2005b). The ecology and evolution of patience in two New World primates. Biol. Lett. 1, 223–226. Stevens, J. M. G., Vervaecke, H., de Vries, H., and van Elsacker, L. (2005). The influence of steepness of dominance hierarchies on reciprocity and interchange in captive groups of bonobos (Pan paniscus). Behaviour 142, 941–960. Stewart, K. J., and Harcourt, A. H. (1994). Gorilla vocalizations during rest periods: Signals of impending departure? Behaviour 130, 29–40. Trivers, R. L. (1971). The evolution of reciprocal altruism. Q. Rev. Biol. 46, 35–57. Tutin, C. E. G. (1979). Mating patterns and reproductive strategies in a community of wild chimpanzees (Pan troglodytes schweinfurthii). Behav. Ecol. Sociobiol. 6, 29–38. Visalberghi, E. (1997). Success and understanding in cognitive tasks: A comparison between Cebus apella and Pan troglodytes. Int. J. Primatol. 18, 811–830. Visalberghi, E., Quarantotti, B. P., and Tranchida, F. (2000). Solving a cooperation task without taking into account’s the partner’s behavior: The case of capuchin monkeys (Cebus apella). J. Comp. Psychol. 114, 297–301. Warden, C. J., and Galt, W. (1943). A study of cooperation, dominance, grooming, and other social factors in monkeys. J. Genet. Psychol. 63, 213–233. Watts, D. P. (1998). Cooperative mate guarding by male chimpanzees at Ngogo, Kibale National Park, Uganda. Behav. Ecol. Sociobiol. 44, 43–55. Werdenich, D., and Huber, L. (2002). Social factors determine cooperation in marmosets. Anim. Behav. 64, 771–781. Wofle, D., and Wofle, H. M. (1939). The development of cooperative behavior in monkeys and young children. J. Gen. Psychol. 55, 137–155. Wrangham, R. W. (1980). An ecological model of female‐bonded primate groups. Behaviour 75, 262–300.

This page intentionally left blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

Coexistence in Female‐Bonded Primate Groups S. Peter Henzi*,{ and Louise Barrett*,{ *school of psychology, university of kwazulu‐natal durban 4041, south africa { department of psychology, university of lethbridge lethbridge, alberta t1k 3m4, canada

I. INTRODUCTION With few exceptions, diurnal primates are distributed in large social groups that are spatially and temporally coherent. The goal of primate socioecology since 1960s has been to understand what drives and structures this distribution. Whereas initial syntheses (Crook and Gartlan, 1966; Eisenberg et al., 1972) focused on males, this emphasis gave way during the 1970s to an increasingly articulated representation of primate sociality that was centered on females and their responses to the world (Dittus, 1977; Hinde, 1983; Seyfarth, 1976; Wrangham, 1980). This shift, to a large degree, was fueled by the coincidence of an accelerating number of detailed field studies and the emergence of sociobiological theory, including the recognition that females were the ‘‘ecological’’ sex (Emlen and Oring, 1977; Wilson, 1975). The former pointed to well‐differentiated relationships within groups, while the latter shifted the analytical emphasis from groups to individuals and promoted kin selection as the likely solution to the reproductive consequences of cooperation and coexistence (Hamilton, 1964, 1972). Given that this fieldwork concentrated on Old World monkey species (Cercopithecoidea) in which females predominantly remain in their natal groups—and hence are ‘‘female‐bonded’’ (Wrangham, 1980; where the term indicates both female philopatry and strong bonds between females)—subsequent empirical and theoretical attention was directed to the nature of the associations of female kin. Despite a well‐administered corrective to the uncritical assumption that the results of all this effort speak to the ‘‘typical’’ primate (Strier, 1994), instead of being phylogenetically circumscribed (Di Fiore and Rendall, 1994), interest in the social dynamics of female‐bonded (FB) primates 0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37002-2

43

Copyright 2007, Elsevier Inc. All rights reserved.

44

S. PETER HENZI AND LOUISE BARRETT

remains strong. There are many reasons for this. Although not representative of most primate societies, FB groups are certainly much more common within the Old‐World monkeys (and the New‐World genera, Cebus and Saimiri), compared to other families (as well as other mammals in general), and this is a distinction that promotes elaboration. There is also a well‐ developed body of socioecological theory, which has emerged from the study of FB groups and that has now developed its own impetus and connects to other central primatological concerns such as the ‘‘social intelligence’’ hypothesis (Barrett et al., 2007). Finally, a good deal of effort continues to go into field studies of FB species, generating both long‐term data from single populations, as well as comparative information across different populations of the same species, both of which allow for much richer analyses than are possible with any other primate group. Our aim here is to review these data, highlighting the strengths, insights, and shortcomings of current theoretical views of FB groups. In doing so, we hope to make a case for a more nuanced framework in which to situate studies of primate sociality and cognition.

II. KINSHIP AND COMPETITION FB societies remain interesting in their own right because they pit the explanatory power of kin selection against the understanding that, other things being equal, individual animals will behave selfishly. What then happens in a world where relatives must coexist? This question, which guides most of the work on social dynamics in FB primate groups, has derived its power primarily from the demonstration that while the advantages of group life accrue in relation to their ability to reduce predation risk, and where larger groups are therefore better, they are coupled to reproductive costs associated with local resource competition, where larger groups, consequently, are worse, at least for some group members (Dunbar, 1988; van Schaik, 1983). These demonstrations that birth rates decline with group size negate the original presumption that FB groups are primarily cooperative, in the sense of being selected in the context of defending resources against other FB groups (Wrangham, 1980). This, then, has been the springboard for everything that follows because it sets up the idea that female relatives are obliged to compete, in one way or another, for resources within a social group that they cannot readily leave. While there have been rumblings recently that this emphasis on within‐group competition discounts the generally cooperative basis of social engagement (Sussman et al., 2005) it is likely to stand, at least until the alternative is more comprehensively fleshed out (Koenig et al., 2006) and some synthesis achieved.

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

45

In the meantime, the question of how females respond to the ‘‘inevitability’’ of competition has produced a cascading set of neat, interlocking responses. These have served as powerful organizing principles for data collection and interpretation (Barrett et al., 2007) and have additional resonance because they point directly to a particular view of primate cognition (Dunbar, 1998). In this view, the canonical structure of FB social dynamics is derived from the following strategic responses displayed by females: 1. In order to defend resources against competitors within groups, it is advantageous for females to form cooperative coalitions and alliances (Harcourt and de Waal, 1992; Wrangham, 1987), where a coalition is defined as an event in which one individual aids another by actively joining forces against a third during an on‐going aggressive encounter, and where an alliance represents an enduring cooperative relationship involving repeated coalition formation. 2. Allogrooming, insofar as it provides a service to others, is used by females to persuade valuable (e.g., high‐ranking) partners to participate in coalitions (Seyfarth, 1977). Females form strong, enduring relationships with each other, which they service by grooming, as a means of ‘‘ensuring unstinting mutual support’’ (Dunbar, 1998) from their coalition partners over time, either by the exchange of grooming for support in a reciprocal fashion (Cheney and Seyfarth, 1984) or by using grooming as a signal of mutual trustworthiness (Dunbar, 1998). 3. Where such valuable relationships are damaged (e.g., by aggression between alliance partners), females will act to repair this damage via a process known as reconciliation (de Waal, 1989), so that they can continue to reap the benefits of coalition formation over time. This is, to some extent, a caricature of a more nuanced framework that also includes the effects of resource dispersion (Sterck et al., 1998; van Schaik, 1989), where these processes, associated with nepotism, are more predictive of species encountering clumped, defensible resources. Nevertheless, it is true to say that this broad structure receives support from Di Fiore and Rendall’s phylogenetic analysis (Di Fiore and Rendall, 1994) and that their results are perhaps themselves a product of the fact that most researchers have either concentrated their work on these topics or framed their analyses in this way (see, as two examples, Matsumura, 1998; Silk et al., 2003). In addition, further implicit acceptance for this focus is evinced by the widely embraced ‘‘social brain’’ hypothesis (Dunbar, 1998), which extends these premises to account for the general increase in relative brain size among the anthropoid primates relative to other mammalian groups. Here, the prospective need to predict the actions of others and to

46

S. PETER HENZI AND LOUISE BARRETT

track and service valuable relationships in order to sustain coalitions (that is, the qualitative demands of social engagement with others), all place cognitive demands on females that have selected for increased brain size; these selection pressures are assumed to have been stronger in species that, for ecological reasons, must live in larger groups, producing a quantitative component to relationship monitoring, which further selects for increased brain size. In other words, the social brain hypothesis provides the mechanistic, cognitive underpinnings that are required to sustain the long‐term, strategic relationships that structure FB groups in ways that are relevant from a functional, evolutionary perspective. In this view, therefore, the assumption being made is that females have an overt, cognitive understanding of their own relationships that they actively strive to sustain, protect, and repair over time. This framework and its corollaries are very compelling, both because the components fit together seamlessly and because they are congenial to a view of ourselves that, at least in part, helps validate research on primate sociality (Barrett et al., 2007). While there is no doubt that it has been a very powerful driver of some very good research, it is, in a sense, at the mercy of its own success. Our general argument has been that the congeniality of the argument is not matched by the data it has produced (Barrett and Henzi, 2005, 2006; Henzi and Barrett, 1999), and that there is now an opportunity—easily met, we think—to reconfigure it. In the sections that follow, we look at each of the behaviors linked to theories of female coexistence. We illustrate how each can be seen as an independent, contingent response to current need rather than as interlocking components of an overall female strategy to cultivate and enhance relationships in the long term. Having done so, we then suggest that the concept of a relationship may itself benefit from a fresh, and less anthropomorphic, assessment.

III. ORGANIZING PRINCIPLES A. COALITIONS Coalitions among adult females against other adult females of the same group provide the keystone for this conceptualization of long‐term relationships: if coalitions are not the key mechanism by which females alleviate the negative effects of competition on their reproductive success, then there is no immediate need to consider the formation and conservation of valuable long‐ term partnerships. Coalitions and alliances sprang to prominence with the rising interest in primate social cognition because they suggested a capacity to track third‐party relationships, which, at the time, appeared to be

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

47

beyond the capability of non‐primates (Harcourt and de Waal, 1992). More recent research, however, has shown that other species such as hyenas display similar abilities despite relatively smaller brains (Holekamp, 2007; Holekamp et al., 2007). What is interesting is the extent to which the deployment of coalitions as the primary means by which females achieve their competitive goals has been an article of faith for primatologists in the face of relatively little evidence that they serve such a function (Henzi and Barrett, 1999). Savannah baboons, for example, live in typical FB groups and, in some populations, experience strong within‐group competition that has clear consequences for participants. At Mkuzi in South Africa, at a time when predation pressure was particularly high, rates of female–female aggression were among the highest recorded for baboons and were a consequence of attempts to shift other females to the periphery of the group where they were more vulnerable to predation (Ron et al., 1996). Despite this, and under circumstances where one might readily imagine the advantages of conjoint action, coalitions were conspicuously absent. Similar results were obtained from another very different chacma baboon (Papio hamadryas ursinus) population (Silk et al., 1999) and the possibility has been mooted that the absence of coalitions among females might be a feature of this subspecies, since it also lacks the male–male coalitions recorded from other subspecies (Henzi and Barrett, 2003). However, subsequent data from yellow baboons (P. h. cynocephalus) indicate, too, that coalitions are a feature of only 3% of female–female aggression (Silk et al., 2004). More recently, it has been suggested that the apparent paucity of baboon coalitions does not take into account the formation of ‘‘vocal alliances’’ that obviate the need for more physical support (Wittig et al., 2007). Such alliances occur when one female gives a threat vocalization to signal the likelihood that she will intercede on another’s behalf if the dispute does not end quickly. Even with the inclusion of such vocal alliances, however, females supported each other on only 10% of occasions (dropping to 4.4% for physical coalitions alone). This suggests that, no matter how they are configured, coalitions remain rare. In addition, no observational data were presented on the actual effectiveness of these alliances in terminating aggression. The experimental playback evidence provided by Wittig et al. (2007) is also inconclusive, partly because the experimental design involved playing a threat grunt to a female who had just been threatened by another animal. While threatened females responded more strongly to calls and avoided contact with the signaler and its kin, the fact that these experimental females had been the victims of aggression means that we cannot infer that these calls have the potential to stop aggressors in their tracks. It is also hard to exclude the possibility that victims were in a state of

48

S. PETER HENZI AND LOUISE BARRETT

Genus

arousal that led them to avoid the aggressors and their relatives (who may, in any case, have been in close spatial proximity—see below) generally and not as a response to these particular calls. A stronger design, where aggressors were played the threat grunts of kin of the animal under attack, and assessing whether this terminated aggression more quickly than under control conditions, where no such vocalization was played, could perhaps provide stronger support. Admittedly, it would be much more difficult to implement since most conflicts are of short duration. This, in itself, questions the effectiveness of providing a signal of potential intervention, given that, in the time taken to produce a call, the incident is likely to have ended anyway. A larger comparative sample from the literature (Fig. 1) generates much the same impression that coalitions are not common (see also Schino, 2001, p. 270). These data are greatly constrained by the precision with which details on coalitions are reported and should be taken primarily as being illustrative; it is particularly difficult to find actual rates of coalition formation reported, to derive them from published accounts, or to determine the exact targets of coalitionary behavior. We know from Seyfarth’s otherwise comprehensive analysis (Seyfarth, 1980, p. 809) of vervet monkeys, Chlorocebus aethiops, for example, only that females form coalitions, but not how often, against whom, or to what ends. Clearly, then, vervet female–female coalitions may be underrepresented in Fig. 1, but, even so, it is not likely that they are common (see Fig. 4 in Seyfarth, 1980), an interpretation

Colobus Cercocebus Procolobus Papio Cercopithecus Presbytis Macaca Chlorocebus Erythrocebus Saimiri Cebus Ateles 0

10

20 30 40 50 Percentage of total aggression

60

70

FIG. 1. Captive and wild studies of different genera that report the extent to which aggression involves coalitions (gray: N ¼ 90 studies) and, where specified, the percentage of these that are female–female coalitions directed at other adult females (black: N ¼ 28 studies). We thank Dr. Parry Clarke for providing the data set.

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

49

corroborated by the recent estimate that coalitions occur in only 3% of aggressive interactions between females (reported as unpublished data in Wittig et al., 2007). At the same time, general reference to coalitions obscures the fact that many female–female coalitions do not target other adult females as predicted by socioecological theory. Family members, for example, assist immature offspring (Chapais, 1992; Cheney, 1977), maximizing the possible adult rank of the supported juvenile, or, as is the case for vervets or blue monkeys, Cercopithecus mitis (Henzi and Lawes, 1987; Seyfarth, 1980), females may cooperate successfully in attacks on adult males. In the case of the vervets (Fig. 2), this constitute 70% of recorded coalitions (Seyfarth, 1980). Nevertheless, it is striking that, first, the highest rates of coalitions are reported for the South American genera, for which grooming and group size do not correlate (Dunbar, 1991) and where one (Ateles) includes no FB species. Second, it is likely that some of the reports reflect captive conditions, where animals are less capable of distancing themselves from conflict. For example, wild Presbytis (now Semnopithecus entellus, the hanuman langur) appear to make little use of coalitions (Koenig, 2000). At the same time, data from free‐ranging capuchin monkeys, Cebus capucinus, show a role for coalitions in status maintenance among females that may belie their relatively low representation here (Manson et al., 1999). The data in Fig. 1 indicate, perhaps above all else, that female–female coalitions, especially when directed against other females, are generally uncommon. However, there is a difference between the importance of coalitions and their ubiquity. The fact that female–female coalitions are mostly rare in FB groups (and those against female targets even rarer) speaks to their conception as an organizing principle for current socioecological theories,

FIG. 2. The photograph shows two female vervet monkeys cooperating to keep a male at bay.

50

S. PETER HENZI AND LOUISE BARRETT

FIG. 3. The photograph illustrates an example of coalition formation in chacma baboon females. The context is as follows: B and C are sisters and, at the time (2002), the first and third ranking females, respectively. D is the older—and A the younger— daughter of B. Just prior to this photograph, C had threatened B who responded by attacking her. While they were fighting, D approached and threatened B (her mother); that is, she entered the dispute by supporting C. B then threatened D while C, instead of persisting with the coalitionary attack on B, joined her in threatening D. This new coalition was augmented by the arrival of the juvenile A. In the photograph, D is looking toward the male who is her consort partner but who does not get involved. The consequence of this episode is that the ranking B > D > C shifts to B > C > D and, later, B > A > C > D (whereas, without intervention and youngest ascendancy, it would have been B > A > D > C). The anecdote draws attention to the contingent nature of presented opportunities (Barrett and Henzi, 2005), while the photograph itself highlights the importance of spatial coherence in conjoint action (see also Fig. 2).

not to their contingent value in the lives of the individual animals that participate in such events. There is little doubt that coalitions can sometimes be very valuable for particular participants. Figure 3 shows two adult females and one immature female baboon combining forces against a fourth. The consequence of this, pursued over several days at our De Hoop study site in South Africa, was the maintenance of rank by one of the adults and the rise in rank of the immature, both at the expense of the attacked female. Despite the significance of such acts for the participants, we have seen only two such episodes in 10 years at the site. Interestingly, both were characterized by similar relative age‐ and rank‐based configurations of mothers and daughters. This suggests that some of the rarity of coalitions may be ascribed to a corresponding rarity in the combination of circumstances that allow a positive payoff for each of the participants.

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

51

In other words, demonstrating that some females in some populations sometimes engage in coalitionary behavior is not good evidence that coalitions themselves have been selected as the evolutionary response to competition, as current theory assumes. Instead, it suggests that coalitions are one element in a suite of possible responses, and it is this plasticity of response that represents the evolutionary adaptation. Instances where females do not form coalitions deserve greater empirical and theoretical emphasis because this will shed light on the full range of tactics that females can and do pursue and, as importantly, help predict more accurately the circumstances under which coalitions are expected to occur. B. GROOMING To the extent that coalitions are rare, grooming is unlikely to serve as an insurance against unpredictable future need, as postulated by some forms of the social brain hypothesis (Dunbar, 1998). Individuals would need to invest large amounts of time and effort in grooming for perhaps very little or no return, especially as coalitions are more likely to occur when both participants gain an immediate benefit rather than just the individual that needs support (Silk et al., 2004; see below for further discussion of these results). Moreover, any mutualistic coalition formation immediately obviates the need for any prior investment in a partner in order to secure support. Even from a solely evolutionary perspective, the costs of social investment would seem to outweigh the limited benefits likely to accrue. In addition, the cognitive demands of coalition formation would seem to be beyond the limited time horizons of monkeys (Barrett and Henzi, 2005; Hampton et al., 2005; Roberts, 2002). At present, we have no good evidence to show that monkeys can engage in true anticipatory planning: that is, that they can form detached representations of their future needs (Ga¨rdenfors, 1995). For example, monkeys often throw food from their cages when satiated, seemingly unable to recognize that they will be hungry again at a later point (Roberts, 2002). More rigorously, Hampton et al. (2005) found that while rhesus macaques showed robust memory for the type and location of a food reward in an open‐field test, there was no evidence that they could remember when an event occurred; as this kind of episodic‐like memory is argued to be linked causally to the ability to engage in ‘‘mental time travel’’ (Tulving, 1983), the inability to recall the temporal sequence of past events suggests that monkeys also cannot project a sequence of events into the future. We should therefore expect to find that grooming is traded either for something immediately obtainable or which does not require overt monitoring of checks and balances over time (such as tolerance around high‐quality resources). Observational data on baboons support

52

S. PETER HENZI AND LOUISE BARRETT

the notion that many social decisions are made on the basis of current need in this way rather than on the anticipation of their future needs: female–female grooming patterns tend to reflect contingent events like infant births and seasonal shifts in foraging competition (Henzi and Barrett, 2002; Henzi et al., 2003). In part, this absence of evidence may reflect only the paucity of research on anticipatory planning and the need to distinguish it carefully from immediate planning for current need (where the presence of relevant stimuli and/or the organisms’ internal state can cue behavior). It is, of course, entirely possible that further study will reveal that such abilities do exist. Nevertheless, it is telling that there is so little evidence given that the assumption that monkeys can anticipate their future social needs is central to the construction of current socioecological and, more pertinently, sociocognitive views of monkey life. Nevertheless, even if we cannot argue for the anticipatory use of grooming, it may be used more generally to cultivate dominant animals, if only to maintain spatial proximity and improve opportunities for access to resources in the vicinity of such individuals (Seyfarth, 1977, 1980). If grooming serves this purpose, we would expect to find both direct and indirect competition for access to high‐ranking females (Seyfarth, 1977). Delineating the discrepancy between intended and achieved grooming allocation is the subject of Seyfarth’s influential model (Seyfarth, 1977), and it stipulates not only that the observed allocation will be biased toward adjacently ranked females, as a consequence of competition for access, but also that the ratio of grooming received to that given will, for similar reasons, be rank related. Seyfarth’s own results (Seyfarth, 1976, 1980) together with a meta‐ analysis (Schino, 2001) provide support for grooming up the hierarchy (i.e., where high‐ranking females receive more grooming than low‐ranking females) and a bias toward grooming individuals of adjacent rank as the model predicts. In terms of the formal model, though, which is about the allocation of grooming in relation to rank, the appropriate datum for each female is really the ratio of grooming received to that given (see Henzi et al., 2003; Seyfarth, 1977), and this should be biased by rank (high‐ranking females should receive proportionately more than they give). Here the results are not as clear‐cut. A reanalysis of Seyfarth’s data indicates that while a trend is evident, there are no statistically significant correlations between this ratio and rank in his three vervet groups (data from Seyfarth, 1980, Table II. Group A: rs ¼
0.62, N ¼ 8, NS; Group B: rs ¼
0.5, N ¼ 7, NS; Group C: rs ¼
0.57; N ¼ 8; NS; Fig. 4). Schino’smeta‐analysis, similarly, indicates that high‐ranking females not only receive more grooming but also give more than low‐ranking ones (Schino, 2001).

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

53

2.5

Grooming received/grooming given

Group A Group B Group C

2.0

1.5

1.0

0.5

0.0 1

2

3

4

5

6

7

8

Rank FIG. 4. The discrepancy between the amount of grooming received by adult female vervet monkeys and the amount that they give to others is plotted against rank for each of the three vervet troops studied by Seyfarth (1980).

A detailed application of Seyfarth’s model to chacma baboons (Henzi et al., 2003; see also Sambrook et al., 1995 for olive baboons) found the same pattern of high‐ranking females both receiving and giving more grooming, while also failing to find, contrary to the model’s predictions, any effect of resource competition on grooming allocation. Specifically, during a period of high resource competition (externally imposed by habitat changes), both female grooming clique size (the number of other individuals that a given female grooms) and partner diversity (in terms of the identity and rank of grooming partners) were higher than during a period of low competition. These findings are contrary to the model’s predictions because competition over access to high‐ranking females would reduce clique size due to the exclusion of low‐ranking females by those higher in the hierarchy, and, as a consequence, result in a lower rank‐related diversity of grooming partners. The absence of such an effect suggests that grooming decisions are not driven by competition for high‐ranking partners, and that grooming ratios should not, therefore, be expected to vary by rank. In fact, given the unavoidable confound of kin and rank (adjacent ranks are likely to be occupied by close relatives), these results are as likely to indicate that high‐ranking animals, who can forage more efficiently, simply

54

S. PETER HENZI AND LOUISE BARRETT

have more time for grooming, which they provide preferentially to their close kin. Since the value of one’s rank to another individual is relative, not absolute, we expect rank distance to be more influential than high rank per se (Barrett et al., 1999) and, supporting this, Schino (2001) found no effect of rank on the intensity of preference for kin. The real issue here is not only whether Seyfarth’s model describes an accurate, intrinsic modus vivendi for females but also whether there is the expected uniformity within FB species. Our argument has been that there is too much variability to be able to conclude that this is the case (Henzi and Barrett, 1999). Some capuchin monkey species groom up the hierarchy sometimes, but show differences between populations (C. capucinus; Manson et al., 1999), while others groom down the hierarchy in both the wild (C. olivaceous, O’Brien, 1993; C. apella, Di Bittetti, 1997) and captivity (C. apella, Parr et al., 1997). Free‐ranging bonnet macaques, similarly and unusually for macaques, groom down the hierarchy (Sinha, 1998). Blue monkeys, like vervets, compete for access to grooming partners (16% of aggressive interactions) and groom up the hierarchy, but they neither form coalitions nor compete intensely for resources (Payne et al., 2003). Baboons, which often compete over resources, do not compete, directly or indirectly, for grooming partners (Henzi et al., 2003). They do, however, adjust their patterns of grooming in relation to resource competition and the manner in which they do so offers another way to configure the role of grooming. As noted above, Seyfarth’s model predicts that when resource competition increases, and if these resources are clumped and defendable (van Schaik, 1989), we should see a decrease in the grooming partner diversity of high‐ranking animals as females jockey for access to valuable partners. We find, instead, that high‐ranking chacma baboon females at De Hoop experience an increase in grooming partner diversity as food becomes more clumped and levels of aggression increase (Barrett et al., 2002; Henzi et al., 2003). High‐ranking females, in other words, are groomed by more partners when resource competition increases (Fig. 5). We explain this by reference to the theory of biological markets (BM) developed by Noe¨ and Hammerstein (1994, 1995). The goal of market theory is to explain the emergence of cooperation that evolves under the pressure of partner choice. We expect partner choice to be able to keep cooperating animals honest. For this to be so, cooperation must be tied up in the exchange of commodities, where participants have a choice of trading partners and where the value of the commodities relative to one another is set by the law of supply and demand. Allogrooming is an excellent example of a dynamic market of this kind because, as Seyfarth (1977) himself noted, grooming has intrinsic value. Since no animal is sufficiently dexterous to be

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

−11 −10 −9 −8 −7 −6 Ra−5−4−3 nk −2−1

di

st

k

Period 1: high competition

1

2

an

−10 −9 −8 −7 −6 −5 −4 Ra −3−2 n −1

di

55

3

ce4 5 6

7

8

9 10 11

1 2

7 5 6 3 4 Rank

10 11 8 9

12

Period 2: low competition 1

st

2

an

3

4

ce

5

6

7

8

9 10

1

2

3

4

5 6

7

10 11 8 9

Rank

FIG. 5. The relationship between the rank of the recipient of grooming and the rank distance of her adult female grooming partners when resource competition is high (Period 1) and when it is low (Period 2). The data come from female baboons at the De Hoop Nature Reserve (Henzi et al., 2003; Fig. 5 reproduced with permission from Elsevier). The z‐axis reflects the relative proportion of grooming received from each partner. Dark bars indicate female partners of adjacent rank. The illustration draws attention to the fact that the spread of partners beyond those adjacent in rank is marked in both periods although partner diversity is greater when resource competition is higher.

able to groom its entire surface area efficiently (Barton, 1985), allogrooming is a service that can be exchanged either for itself or for some other commodity such as tolerance at a feeding site. According to this interpretation, the observed increase in grooming diversity is a consequence of increased demand, by more females of lower rank, for greater tolerance around other females when food is clumped (Henzi et al., 2003). This is a demonstration of partner choice in operation and, instead of reflecting the shoring up of valuable relationships in case of future need, is a response by females who need to trade grooming in the here and now (Barrett et al., 2002). That is, while this short‐term contingent cultivation of partners may end up having long‐term positive benefits, there is no need to posit the latter as the proximate driver of females’ social choices in the here and now; an explanation in terms of current need is

56

S. PETER HENZI AND LOUISE BARRETT

sufficient to explain the patterns in the data. In corroboration, we have found that as demand for high‐ranking tolerance increases, so too does the price that low‐ranking females are prepared to pay; there is a significant increase in the relative amounts of time for which high‐ranking females are groomed (Fig. 6). More specifically, there is a strong positive correlation between the difference in ranks between two partners and the relative allocation of effort by the lower ranking of the two (Barrett et al., 2003). With similar results in other contexts (Barrett et al., 1999; Henzi and Barrett, 2002) and from other species (Manson et al., 2004; Payne et al., 2003), it is possible to advance the idea that grooming is a tool whose social deployment is both geared to current need and sensitive to circumstance rather than used to signal the level of mutual support once individual can expect from another. C. RECONCILIATION The term ‘‘reconciliation’’ has been used to label selective affiliative contact between individuals previously involved in an aggressive conflict. This contact can take the form of specific behaviors that signal the occurrence of reconciliation such as ‘‘kissing’’ and ‘‘embracing’’ in chimpanzees, Pan troglodytes (de Waal and van Roosmalen, 1979), but may also consist of general affiliative behaviors such as lipsmacking, contact‐sitting, and grooming (see Aureli and de Waal, 2000; Aureli et al., 2002 for a review). This behavior has been interpreted as a means by which individuals can alleviate the costs of unavoidable aggression, particularly for animals that live in permanent social groups (Aureli and de Waal, 2000). Conflict resolution fits into a view of social life which is essentially cooperative, but where aggression is seen as both inevitable and necessary for the negotiation of individual interests among nonlinguistic animals. Post‐conflict affiliative behavior is therefore crucial to ensure that the positive benefits of group life (e.g., protection from predators) are not lost due to the dispersive effects of aggression on social interaction (de Waal, 1986). However, with respect to primates in particular, the argument has been taken much further than this such that post‐conflict behavior is predicted and explained on the basis of the ‘‘valuable relationship hypothesis’’ (Aureli et al., 2002; de Waal and Aureli, 1997). This argues that when a conflict disrupts the usual pattern of interaction between individuals, it damages the relationship between them and, consequently, jeopardizes the benefits associated with it (Aureli et al., 2002). According to this hypothesis, individuals engage in post‐conflict behaviors in order to preserve the integrity of their relationships with particular individuals; post‐conflict behavior then serves to ‘‘repair the damage’’

57

A

1.5

Unstandardized residuals

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

1.0 0.5 0 −0.5 −1.0 −1.5 −12

Unstandardized residuals

B

−8

−4

0 4 Rank distance

−2

0 2 4 Rank distance

8

12

1.0

0.5

0

−0.5

−1.0

−8

−6

−4

6

8

10

FIG. 6. The impact of rank effects on timed grooming bout contributions when (A) resource competition is high (Period 1) and (B) when it is low (Period 2). The data come from female baboons at the De Hoop Nature Reserve (Barrett et al., 2002; Fig. 4 reproduced with permission from Elsevier). The x‐axis represents the rank distance between the individual participants in a grooming bout. The y‐axis represents residuals from a regression of A’s contribution on B’s contribution to a given grooming bout. The designation of Individual A was conventionally assigned to the first individual to groom in a bout, irrespective of rank. Deviations from zero indicate that one partner groomed significantly more (positive deviations) or less (negative deviations) than the other during a bout. During Period 1, extreme residual values were associated significantly with large rank distances, indicating that low‐ranking females contributed more to grooming bouts than did high‐ranking females, indicating that low‐ranking females invest more in grooming high‐ranking females under conditions of high competition. During Period 2, there was no association between the deviation from grooming equality and rank distance, indicating that rank‐related variation in grooming value was absent under conditions of low competition.

58

S. PETER HENZI AND LOUISE BARRETT

caused by aggression and leads to ‘‘reconciliation.’’ The reduction in stress‐ related, self‐directed behaviors (e.g., scratching) following reconciliation in both aggressors and victims is related to the restoration of the social bond and the associated reduction in uncertainty regarding the future of the relationship that conflict created (Aureli et al., 2002). Only the relationship repair hypothesis, it is argued, can account for why the aggressor’s, as well as the victim’s, stress‐related behavior increases after conflicts and is reduced following post‐conflict affiliation. As the former were not exposed to aggression in the same manner as the victim, it can only be the uncertainty associated with relationship status that leads them to engage in high levels of stress‐related behaviors. We show below, however, that this is not the only plausible mechanism that can account for the aggressor’s behavior. Since de Waal and van Roosmalen’s groundbreaking study (de Waal and van Roosmalen, 1979), the subject has generated a large literature in which species differences in conciliatory tendency (the probability that a particular conflict will be reconciled) have been widely documented and related to differences in the relatedness and dominance rank of individuals (Aureli and de Waal, 2000; Aureli et al., 1989, 1997; Demaria and Thierry, 2001; de Waal and Ren, 1988), as well as factors such as the intensity of aggression, the decisiveness of conflict outcome, reproductive season, and the presence of infants (Aureli et al., 2002; Call et al., 1999; de Waal, 1993; Kappeler and van Schaik, 1992; Silk, 1996). More recently, the issue of how this relates to the negotiation and functioning of individual interactions over time has been addressed, and a framework for predicting the occurrence of post‐conflict behavior has been developed (Aureli et al., 2002). As mentioned above, this framework is based on the notion that post‐ conflict behavior serves to repair damaged relationships. Consequently, the argument put forward is that post‐conflict reunions are necessary under conditions where individuals will suffer a post‐conflict loss of benefits from a relationship with their former opponents and where within‐group aggression undermines a relationship that is perceived as valuable by both partners (Aureli et al., 2002, p. 336). As a result, post‐conflict reunions should occur more often after aggressive conflicts between individuals with more valuable relationships (Aureli et al., 2002, p. 336). Observational data provide some support for the idea that animals that engage in more positive social interactions overall (taken as the measure of ‘‘value’’) also show higher rates of conflict resolution (see Aureli et al., 2002 for a review). A small number of experimental studies help establish the causality of this relationship. For example, Cords and Thurnheer (1993) manipulated the value of individuals to each other experimentally by training pairs of long‐tailed macaque females to cooperate on a task that required mutual tolerance to obtain a food reward. Reconciliatory tendencies within

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

59

pairs increased following this training, suggesting that females were more likely to reconcile following aggression when the opponent was more valuable to them. Despite this body of evidence, however, there are some outstanding issues that are not fully accounted for within this framework. For example, only a small proportion of all conflicts are actually reconciled. The ‘‘conciliatory tendency’’ ranges from 14% for rhesus macaques, Macaca mulatta, to a maximum of 41% for stump‐tailed macaques, M. arctoides (see also Kappeler and van Schaik, 1992). This means that, even among those species designated as showing a high conciliatory tendency, the majority of fights are not reconciled. Equally, while a number of studies have shown that kin reconcile at much higher rates than non‐kin, it is nevertheless the case that kin rarely reconcile any more than 50% of all disputes (Kappeler and van Schaik, 1992). This raises the question of how important reconciliation actually is to the repair and maintenance of relationships. Second, in those studies in which the context of aggression was recorded (Aureli, 1992; Castles and Whiten, 1998), the majority of reconciled conflicts were those where the observer was not able to determine any apparent context. As it seems unlikely that animals would waste time and energy attacking each other without cause, it is reasonable to assume either that we are missing something important in these interactions or, perhaps, simply looking at them in the wrong way. This suggests, in turn, that a sharper focus on the period prior to aggression, in tandem with a greater theoretical emphasis on the goal of such aggression, may help at least as much as the post‐ conflict period in identifying the underlying motivation for conflicts and their resolution. Silk (2002a), for example, has argued that ‘‘random acts of aggression and senseless acts of intimidation’’ have been selected for in social groups as a means of asserting and maintaining dominance relationships: random, unpredictable aggressive attacks exert the maximum stress on the subordinates to which they are directed, while keeping stress on the aggressor relatively low. Aggression causes the victim to mount a physiological stress response that has adaptive short‐term benefits but carries severe longer‐ term costs, including reduced fertility, if the response becomes chronic. The lack of predictability, in terms of the timing of attacks and their duration, creates these chronic conditions and can explain why selection would favor such behavior on the part of dominant animals. If this can explain patterns of aggression, at least partly, then the idea that post‐conflict behavior serves to repair valuable relationships damaged by aggression is called into question. Aggression used in this manner is, by definition, not damaging a valuable relationship, but establishing and maintaining the boundaries of

60

S. PETER HENZI AND LOUISE BARRETT

particular dominance‐related relationships. Any subsequent reconciliation under these circumstances is therefore difficult to incorporate into the relationship repair hypothesis, according to Silk (2002a). Finally, there is inconsistency in the literature as to whether the ‘‘aggressor’’ or the ‘‘victim’’ makes the first peaceful contact after conflict (de Waal, 1993; de Waal and Ren, 1988; Judge, 1991; Kappeler, 1993). This is true both between and within species. Although in the majority of cases the victim tends to make the first move (Kappeler and van Schaik, 1992), there are a number of species in which the aggressor is statistically more likely to do so, and at least one species (P. hamadryas) in which both victim and aggressor are equally likely to attempt reconciliation (Castles and Whiten, 1998; Petit and Thierry, 1994). These findings raise two interesting questions. First, if reconciliation is a straightforward act that repairs a damaged relationship, why should there be this variation? Second, and related to this, if it is the act of aggression that damages the relationship, why should the victim of aggression be more likely to reconcile? Aureli et al. (2002) suggest, on the one hand, that a more nuanced assessment of the costs and benefits of particular kinds of interactions may well account for some of this variation and, on the other, that certain relationships are so critical to social functioning that they need not be reconciled (among cooperative breeders, like tamarins, e.g., Schaffner et al., 2001). An alternative view, however, is that reconciliation may not be tied to the repair and protection of valuable relationships. Silk (1996, 1997, 2000, 2002b) has proposed that post‐conflict behaviors are ‘‘signals of benign intent.’’ These inform the recipient that the current conflict is over and that the actor’s intentions are no longer aggressive. She argues that such signals may be favored by natural selection because they enable former opponents to coordinate their interactions (Silk, 1997). This is a return to the idea that post‐conflict behaviors function to preserve the cohesion of social groups rather than repair and protect particular valuable relationships. In other words, Silk’s reading of the available data (Silk, 2002b) is that post‐conflict behavior is a contingent response to current circumstances, allowing animals to achieve short‐term objectives (e.g., access to resources, tolerance by preferred partners) and does not require a long‐term, relationship‐based component. The rise in stress‐related behavior by aggressors, for example, can be explained as a result of increased arousal following aggression and need not be associated with relationship uncertainty. The fact that redirected aggression toward uninvolved third parties can reduce self‐directed behavior as effectively as post‐conflict affiliation with a former opponent is consistent with this view (Silk, 2002b).

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

61

Silk’s proposals can also account for low rates of post‐conflict affiliation because, if these are needed only to reduce short‐term uncertainty, other tactics (e.g., increasing distance from an aggressor) can be equally effective. If the aggressor is far away, further aggression is much less likely and uncertainty reduced in equal measure. Silk (2000) also argues that the relatively higher rates of post‐conflict affiliation between kin are better explained by her benign intent hypothesis than by the relationship repair hypothesis. The latter predicts that post‐conflict affiliation should occur most often within dyads that value a relationship more and for whom damage to a relationship is more costly. This in turn is argued to be a function of how ‘‘secure’’ the bonds are between individuals. Consequently, as Silk (2002b) argues, kin should show low rates of reconciliatory behavior because, according to the relationship repair hypothesis, although equally valuable, their bonds are less likely to be damaged by conflict due to their greater security (Cords, 1988). This security is at least partly attributed to the benefits that accrue through kin selection. The fact that kin actually engage in post‐conflict behavior at equivalent, or even higher rates, than non‐kin is not good support for the relationship repair hypothesis, but does fit with the notion that individuals are signaling an end to aggression. Kin‐based reconciliation is then more common not because of a greater need to salvage damaged relationships but because, as we go into below, kin may be more strongly motivated to remain in close spatial proximity. They therefore have an active need to signal an end to hostilities and cannot use alternative responses to manage and cope with the post‐conflict period such as maintaining distance or avoiding former aggressors. The apparent lack of context of many aggressive disputes and the tendency of aggressors as well as victims to reconcile are also accommodated more satisfactorily by the benign intent hypothesis because, in Silk’s view, reconciliation is an antidote to random, intimidating aggression. Individuals subject to random attacks may be (understandably) wary of former attackers since they cannot predict the nature of any particular interaction. Reconciling reduces this short‐term uncertainty in interactions for both aggressors and victims, enabling aggressors to engage in behaviors that serve new goals (e.g., a former aggressor may now want grooming from a subordinate). Support for this comes from the findings that, among baboons, females tend to reconcile selectively with females that have young infants whom the conciliatory female goes on to ‘‘handle’’ (i.e., engage in various forms of physical contact; Silk et al., 1996). Furthermore, rates of reconciliation track rates of infant handling by non‐mothers, and females also reconcile more frequently with the mothers whose infants they are most eager to handle. All of this suggests that it is the short‐term

62

S. PETER HENZI AND LOUISE BARRETT

goal of handling infants that drives conciliatory behavior not the value or quality of the relationship with the infant’s mother (Silk, 2000; Silk et al., 1996). Silk (2002b) also points out that even those studies, like Cords and Thurnheer’s (1993), that have been taken as compelling evidence of the relationship repair hypothesis are open to an alternative interpretation. Cords and Thurnheer’s experiment (Cords and Thurnheer, 1993) assumed that the value of the social bond between two animals had been manipulated but, strictly speaking, it was the value of a partner with respect to an individual’s own success on an instrumental task that was manipulated, not the value of a social bond as such. In the cooperative feeding task, monkeys could only feed if their partner was tolerant of their presence. If one monkey was intolerant or wary, and left the feeding station, then neither monkey was able to obtain the food reward. It is possible, under such circumstances, that animals learned to use reconciliation instrumentally: females became more conciliatory because this enhanced their own success at the task rather than because they valued their relationship with their partner more highly (Silk, 2002b). As such, signals of benign intent would enable animals to communicate their intentions to behave peaceably around their partner. It has been suggested that the benign intent hypothesis, rather than being an alternative, is complementary to the relationship repair hypothesis (Cords and Aureli, 1996; de Waal, 2000). In this view, it is simply an explanation of the proximate motivation for the resumption of contact between former opponents. Ultimately, however, females are motivated to resume contact because repairing valuable relationships contributes positively to long‐term reproductive success. Silk’s counter (Silk, 2000, 2002b) is that the long‐term consequences stemming from short‐term post‐conflict behaviors do not mean that natural selection has favored the evolution of reconciliatory behaviors because they enhance long‐term social bonds. The selection pressures acting on signals of benign intent may be quite different from those that shape social bonds between females. It is, of course, possible that long‐term benefits do accrue from post‐conflict behaviors—this is, after all, the reasoning behind Grafen’s ‘‘phenotypic gambit’’ (Grafen, 1984)—but the benign intent hypothesis is complete in itself and sufficient to account both for the observed patterns and to explain the evolution of reconciliatory behaviors. In the absence of any additional explanatory power, Silk regards it as unnecessary to add an extra, long‐ term, relationship‐related benefit. Why, then, are so many researchers persuaded by the relationship repair hypothesis?

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

63

Silk’s suggestion (Silk, 2002b) is that it is because reconciliation is an anthropomorphic concept, deriving originally from a direct analogy with human conflict (de Waal and van Roosmalen, 1979). Consequently, the relationship repair model is a compelling one because it fits so well with our own folk model of how and why we resolve conflict (p. 38). This is something that we have argued is also true of the relationship concept itself (Barrett et al., 2007). In Section IV, we show how a reconsideration of the concept of a primate ‘‘relationship’’ is a logical extension of Silk’s proposition that reconciliation is a short‐term signal of benign intent (Silk, 2002b).

IV. RELATIONSHIPS Hinde’s original proposition (Hinde, 1976, 1983) was that relationships between animals could be abstracted from the frequency, quality, and patterning of individual social interactions. This further implied that, as emergent, irreducible phenomena, relationships were greater than the sum of their parts. Consequently, although relationships are, at base, quantitative measures of overt behavioral patterns such as grooming, proximity maintenance, coalitionary support, and post‐conflict behavior, a sense has developed that these behaviors are not constitutive of the relationship itself. They are, instead, only the surface reflection of a deeper underlying bond. For example, Cords (1997) states: ‘‘Validation of grooming and proximity measures as indicators of social bond strength comes from correlations between them and other types of behavior’’ (p. 27; our emphasis), before going on to admit that ‘‘there is little direct evidence that grooming is used to cultivate valuable social bonds’’ (our emphasis). Similarly, Strier (1999) writes, ‘‘Female kin in these cohesive, matrilineal groups form affiliative bonds that are expressed through proximity, grooming, and agonistic support. . .’’ (p. 300; our emphasis). Here, then, we have a clear separation between the relationship itself and the behavioral measures used to identify and assess it. Note, too, that the focus of these articles is neither on the nature of relationships per se nor are these statements presented as definitions; the existence of relationships is taken to be axiomatic and a springboard for the elaboration of primate social complexity. This reification of relationships turns them into a much more anthropomorphic construction, akin to the concept of friendship in humans. This, as Silk (2002c) suggests, following Hinde (1983), is something that we define by the kind of emotional bond that exists between individuals, and not merely by the kinds of things they do together. As Silk (2002c) points out, this makes it difficult to identify whether anything like human friendship can objectively be identified in other species, since we cannot easily gain

64

S. PETER HENZI AND LOUISE BARRETT

access to another animal’s emotions. The same applies, in our view, to current notions of all primate relationships: if a bond or relationship is constituted by something other than the behaviors used to quantify it, how do we get at the relationship itself? What would constitute good evidence to suggest that such bonds exist? A related problem is that the abstraction of relationships from a series of interactions introduces a temporal component that may also reflect our own folk‐psychological constructions rather than those of the animals themselves. The notion of grooming as something that ‘‘cultivates’’ or ‘‘maintains’’ a social bond is based on the assumption that the function of relationships is to ensure unstinting mutual support (Dunbar, 1998) from coalition partners at unknown, unpredictable future dates. This introduces a prospective element to relationships that motivates females to maintain bonds with a set of familiar, valuable partners. While an evolutionary account does not require females to make any cognitive assessment or have any conscious awareness that this is what they are doing, it is also true that this prospective element has been incorporated into the social brain hypothesis of increased brain size. There is, therefore, an assumption that females’ grooming decisions derive from cognitive assessments of the likelihood of future need. Whiten (2000), for example, in an exploration of the kinds of social complexity that would select for enlarged brains, suggests that this complexity is inherent in the fact that primate relationships vary in stability, creating a pressure to track their status, so that there is a consequent pressure to ‘‘pick up information that might not be vital at the time, but which may be utilized adaptively later on.’’ Similarly, Aureli et al. (2002) suggest that the reason why both aggressors and victims display post‐conflict anxiety is that they have some overt recognition of future costs. They argue that it is difficult to explain the stress response in former aggressors in terms of Silk’s hypothesis (Silk, 1996, 2002b) because aggressors actually face a very low risk of renewed attack. Instead, Aureli et al. (2002) state that it is the uncertainty about the future of their relationship, and the potential loss of future benefits, that creates an additional source of post‐conflict anxiety that affects aggressors as well as victims. It is difficult to construe this as a purely evolutionary argument, with no overt, cognitive component, because it suggests that aggressors understand that they have put their future relationship at risk and are not simply mounting a physiological response to the stress of aggression (which is all that a functional, evolutionary account—and the benign intent hypothesis—requires). In these examples, then, relationships are assumed to reflect ‘‘propositional knowledge’’ (Cheney and Seyfarth, 2005) on the part of the animals: the individuals concerned ‘‘know that’’ they have relationships with others (i.e., possess a declarative, explicit knowledge),

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

65

and are also assumed to know that such relationships require ongoing maintenance in order to provide benefits, and that aggression puts this at risk. The costs of conflict are therefore construed as future, rather than current, costs. Moreover, the formulation by Aureli et al. (2002) requires this prospective ability in order to distinguish it from Silk’s hypothesis: remove the temporal component (the uncertainty about the future of the relationship), and the argument reduces to the benign intent model, where animals act to reduce current uncertainty (which may or may not have long‐ term consequences). There is, therefore, a distinction to be maintained between an evolutionary account of relationships in terms of their fitness‐ enhancing properties and an evolutionary account of relationships as a selective force shaping primate cognitive capacity. As these examples illustrate, that distinction is often blurred, so that evolutionary and proximate explanations become mutually reinforcing. There is no doubt that certain individuals in FB groups—most notably close kin such as mothers, daughters, and sisters—show higher levels of affiliation than others, that they do so flexibly and contingently, and that consistent patterns of affiliation can often be detected over time. There is also recent evidence to suggest that sociability has positive fitness effects: Silk et al. (2003) have demonstrated, for female yellow baboons at Amboseli, that more sociable females rear a greater number of offspring to the age of 1 year. The unanswered question in all such studies, however, is whether females can abstract across successive interactions in a way that allows them to construe their engagement with others in terms of their relationships and then reason prospectively about how to cultivate, manage, and protect them. This, as we argue elsewhere (Barrett et al., 2007), is not something for which we currently have any good evidence. There are, on the other hand, data to suggest that short‐term, contingent, and instrumental choice of partners is common in FB primate groups. At De Hoop, measures of grooming and proximity indicate marked inconsistency across years in the degree to which females were ranked as another’s primary partner, where changes in preference were associated strongly with reproductive events (Barrett and Henzi, 2002). This, by itself, might explain why more sociable females are more reproductively successful: females with young infants attract significantly more social attention than non‐ lactating females (Altmann, 1980; Henzi and Barrett, 2002; Silk et al., 2003), with the result that those females who give birth more often experience increased levels of social interaction. If so, causality may run in the opposite direction to that assumed for Amboseli (Silk et al., 2003). There is also evidence that individuals groom and maintain proximity to others on the basis of short‐term concerns, such as access to infants, tolerance around food and water resources, access to ‘‘skilled’’ individuals and avoidance of

66

S. PETER HENZI AND LOUISE BARRETT

aggression from more dominant individuals (Barrett and Henzi, 2001; Chapais, 2006; Henzi et al., 2003; Kapsalis and Berman, 1996; Muroyama, 1994; Silk, 1992; Stammbach, 1988). We have shown, more recently, that female relationships, in both the De Hoop and Drakensberg baboon populations, differ qualitatively in different seasons (Henzi et al., in preparation). The application of new and powerful network analysis techniques (Lusseau and Newman, 2004) revealed that, in both populations, when food was plentiful, female associations, at both sites, were either only ‘‘brief’’ (measured in hours) or ‘‘casual’’ (measured in days). In contrast, when food was scarce, females formed ‘‘constant companionships’’ (the lagged association rate was constant over time). It is difficult to explain these results with reference to quantitative variation in factors like nearest neighbor distance or feeding competition since the study sites differ ecologically and the females’ ecological responses to food availability therefore differ across sites accordingly. Moreover, and more pertinently, females did not generally resume relationships with the same individuals following the period of seasonal dissolution, while the downgrading of relationships did not lead to group fission or fragmentation. This suggests that models of social life predicated on the presumption that grooming acts as a ‘‘social glue’’ may also need rethinking (Dunbar, 1992). If relationships can dissolve at certain times of year with no adverse effects on group integrity, and female dissolve and form bonds with different individuals across time, it becomes harder to see why ecologically and demographically induced reductions in grooming time lead eventually and inevitably to permanent fission. Interestingly, in his original derivation, Dunbar (1992) found no relationship between group size and social time for the Papio baboons and concluded that the relationship was masked by other factors (note that, for the purposes of these and related analyses, Dunbar treated social time as equivalent to grooming time, on the basis that grooming accounts for over 95% of social time in this genus). It now seems that his alternative suggestion, that group size is genuinely irrelevant in the case of baboons, may have been the correct one. Or rather, group size may be irrelevant in the sense that Dunbar (1992) originally assumed: a group represents a network of bonds whose functional value requires high temporal integrity. As we suggest below, however, the argument may retain its value if it is reconfigured. Female bond strength and duration have also been investigated in Amboseli yellow baboons, using an index of sociality calculated from grooming and proximity scores (Silk et al., 2006a,b). This revealed that many dyads had very weak bonds, while very few had quite strong bonds and that, on average, a female formed only 1.5 ‘‘very strong’’ bonds per year

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

67

(Silk et al., 2006a). These are similar to results obtained from Moremi, Botswana, on chacma baboons, where females groomed, on average, only 8 of the 18 available females and fully half of the female cohort devoted the majority of their grooming to just 1 other female (Silk et al., 1999). Moreover, out of a total of 1430 coresident dyads at Amboseli (coresident referring to dyads residing in the same troop at the same time), only 14 maintained a close bond for at least 5 years, which is not too surprising since the mean duration of coresidence was only 3.47 years (Silk et al., 2006b). In addition, bond duration may have been overestimated since Silk et al. (2006b) did not require high sociality index scores to be maintained across consecutive years, allowing for a 1‐year gap between ‘‘consecutive’’ years. If Female B, for example, was among Female A’s top three partners in 1992, 1993, 1995, and 1996, but not in 1994, the duration of the close social bond was still scored as 5 years (Silk et al., 2006b). In addition to overestimating bond duration slightly, this also presents a picture of social bonds as having significantly more temporal consistency than is experienced by the animals themselves. Such short coresidence times (given that females who reach adulthood can normally expect to live for another 10 years: Altmann, 1980; Altmann and Alberts, 2003), together with the low numbers of ‘‘very strong bonds’’ formed each year and the small number of dyads sustaining bond durations of 5 years or more, suggest that Amboseli females may well also display cyclicity in bond quality and that dynamic contingency in the availability of individuals then characterizes the lives of all female baboons. This perspective makes sense because the ever‐present vagaries of predation, disease, and other sources of mortality can lead to the sudden and irreparable loss of a particular social partner. Fission events can also lead to the loss of partners: the short average coresidence times at Amboseli are very much a consequence of groups splitting over time. Interestingly, these fission events often result in matrilines splitting, where a mother will leave her daughter(s), despite the existence of a strong bond between them (Silk et al. 2006a,b). This is something that has also been found in chacma baboons, where it was suggested that females distribute themselves in ways that result in an improvement in their dominance ranks (Ron et al., 1994). Given the frequency of such events over a female’s life, as the Amboseli data and our own indicate, one might expect selection on females either to sustain several partnerships at any time (which may explain the ‘‘many weak bonds’’ at Amboseli) or to recover from the loss of a partner by rapidly seeking a replacement. In this context, it is worth considering, as one example, the work of Engh et al. (2005), who showed that the loss of a close kin partner to predation raised cortisol levels in chacma baboon females. They interpreted this to be

68

S. PETER HENZI AND LOUISE BARRETT

a stress response to the loss of a close social partner, a consequent increase in social isolation and, as such, a form of bereavement. Cortisol’s effects are, however, many and varied and an acute rise in cortisol is also associated with increased alertness and readiness to act (Korte et al., 2005). While the interpretation by Engh et al. is plausible, it is not the only one consistent with their findings. The observed stress response could actually reflect a reaction to the predation event itself, and the need for increased vigilance and alertness in order to avoid suffering the same fate. The ‘‘bereavement’’ interpretation, which connects to the idea of long‐term bonds, is based on the fact that only females who lost a close relative showed this acute response. However, no control was made for the fact that close kin tend to cluster spatially (Rendall et al., 1996), and it is therefore reasonable to suggest that proximity at the time of predation induced greater stress in the survivor. The argument that the transience of the rise in glucocorticoid levels was due to an increase in grooming and partner diversity, driven by a need to reduce social isolation following the loss of a close social partner, is similarly plausible but explicable without recourse to any sense of the need to establish new, enduring bonds with other females. It is possible that females increased their grooming in response to an increase in oxytocin, as the authors suggest, but merely to increase social contact in general and not as a means of establishing new social bonds with particular individuals. Again, this could be a response to the predation event itself and an attempt to occupy more central, safer, positions in the troop where, by definition, more individuals will be encountered. The real point to be made is that interpretation in terms of social isolation rests on an anthropocentrically derived, folk‐theoretical assumption that enduring social bonds exist and are recognized by females, so that it is specifically the disruption of a social bond that causes this physiological response (here Engh et al., 2005, draw explicit parallels between the baboons’ response and that of bereaved or lonely humans). The suggestion that females are responding to their social isolation and that they attempted to cope with their loss by extending their social network, plus the juxtaposition of the baboons’ response with human psychological responses, implicitly promotes the view that relationships are ‘‘real’’ and are valued by the baboons in and of themselves. The assumption itself remains untested. The alternative view is that there is no need to interject an anthropocentric concept of social relationships in order to derive an adaptively plausible explanation. The findings can be equally well addressed by the proposition that females have been selected to respond adaptively to short‐term dynamic changes in their environment, using short‐term social tactics to reduce their own risk of predation. Given that such stressful events are

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

69

likely to be common, as Engh et al. (2005; see also Beehner et al., 2005) also point out, selection might be expected to produce females that remained relatively phlegmatic in the face of social loss, and whose ties were consequently likely to be flexible and contingent. Indeed, it is likely that this kind of social response, acting over the short term in response to specific events, leads to the correlation between sociability and offspring survival. Rather than long‐term social bonds sustained with particular individuals increasing fitness, it could be the ability to seek out appropriate short‐term social interactions with many individuals in response to stressful, reproductive, and other kinds of events that underpin offspring survival. Short‐term contingent responses of this kind, if appropriate, must then naturally translate into long‐term adaptive responses. Given the present lack of evidence to demonstrate that female monkeys possess the kinds of analogical reasoning abilities (Thompson, 1995; Thompson and Oden, 2000), prospective reasoning (Roberts, 2002), or episodic memory capacity (Hampton et al., 2005) needed to prospectively manage enduring, long‐term social relationships, combined with the short durations over which females coreside in troops, it may be that recasting both the social brain hypothesis and socioecological theories of female coexistence to account for contingency will yield real dividends. At the least, this approach proffers a competing hypothesis of greater validity than the statistical null models that are currently relied on (Barrett et al., 2007). For those who are concerned only with the evolutionary function of relationships, this will seem like something of a nonargument, since what may or may not go on in the animals’ heads is clearly irrelevant. However, the point is worth making because functional behaviors require proximate mechanisms and we need to ask what prompts and produces behavior, as well as how it contributes to fitness (Tinbergen, 1963). The separation of mechanism and function is essential because, with primates especially, it is all too easy to elide the two and slide between evolutionary and proximate explanations: an animal that acts as if it knows that its relationships are valuable and worth protecting may well be doing so in an evolutionary sense, but not necessarily at a more proximate level. Natural selection can and has produced rules of thumb that lead animals to behave in their own interests, without any cognitive assessment of what they are doing, or at least without the kinds of cognitive assessment we attribute to them. Our own folk‐ psychological understanding of female primates, and its success with respect to functional explanations of social behavior, cannot be taken to indicate that we understand anything about the ‘‘folk psychology’’ that female primates might use to understand each other. This is a point worth emphasizing because this elision between proximate and ultimate explanations then acts to circumscribe the range of possible functional hypotheses that are actually

70

S. PETER HENZI AND LOUISE BARRETT

put to the test, so that only those that address an intentional conception of primate action are considered. Owren and Rendall (2001), for example, discuss this issue in some detail with respect to primate vocal signaling studies. The argument, then, is for a specific application of Occam’s razor to theories of female coexistence: the assumption that females have an overt cognitive recognition and prospective understanding of their relationships is not needed to explain the patterns we see in the data. Note that this is not an argument for evolutionary parsimony: evolution need not be parsimonious and parsimony itself is not an inherent virtue of evolutionary explanation (Barrett et al., 2007). Rather, our argument is that this extra hypothetical construct is superfluous when attempting to account for female social engagement in a satisfactory manner. In effect, this means removing the temporal component from relationships as perceived by the females themselves. We, as humans, can, of course, continue to abstract over interactions to come up with ‘‘relationships’’ (sensu Hinde, 1981) that may be informative with respect to many evolutionary questions, but we need to recognize that such relationships do not correspond to any real‐world entity or overtly represented as such in the animals’ heads; that is, we cannot assume that the temporal component introduced by abstraction over arbitrary, human‐relative time periods has any necessary relevance to the animals under study or provides the proximate motivation for the behaviors they display.

V. A SPATIAL APPROACH TO SOCIAL INTERACTIONS Even without a human‐like representation of time, social animals are still obliged to contemplate space. The recognition of this necessity drove formulations of the functional significance of gregariousness from early on, where both the positive and negative consequences of particular spatial arrangements were assumed to underpin individual social strategies in relation to both predation and foraging (Hamilton, 1971; van Schaik, 1983). If females use social contact and proximity to jockey for position and influence within spatial confines, and monitor others to facilitate this, we can explain variability in social engagement, its contingent and instrumental nature, and its ultimate value without any assumptions about the conceptual, representational knowledge underpinning action, or any projection of this through time. In addition, if we think in terms of spatial integrity alone, Dunbar’s ‘‘social glue’’ argument retains its value, since it is the cohesion of the group in space that becomes important, not the cohesiveness of particular subgroups over

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

71

time. The inevitable spatial fragmentation of large groups, and the consequent inability to keep in contact with other group members, may be all we need to explain why large groups tend to fission. The variability in patterns of fission among baboon females seems to fit with this view: females may leave a group with the putative father of their offspring to avoid infanticide (Henzi and Barrett, 2003), they may fission in such a way that matrilines stay together (Nash, 1976), or in a way that enables each female to improve her rank (Ron et al., 1994). We can also reconfigure coalitions along these lines if we consider them to be the presentation of a spatially integrated ‘‘united front’’ by two or more females in response to current need. Such ‘‘coalitions’’ could act to reduce social stress on individuals by reducing the immediate likelihood of displacement or aggression, thus fulfilling their hypothesized socioecological function. Maintenance of spatial proximity may therefore serve as a ‘‘passive’’ form of coalitionary support, something Seyfarth (1980) originally hinted at, and reflect an immediate, dynamic response to shifts in partner location. In addition, the spacing of individuals can impede or enhance the likelihood of more active forms of coalitionary support. For example, juvenile and high‐ ranking female capuchins tend to be found in the center of the group, with the result that these females regularly support their juvenile kin in conflicts. Low‐ranking, spatially peripheral females rarely support their juvenile kin because, quite literally, they are not in a position to do so (O’Brien and Robinson, 1993). The advantage of immediate, spatially driven responses is that this kind of active coalition formation is then, as we have argued, simply an additional component in a suite of social tactics. Its rarity in the wild and the lack of evidence that participation is secured by prospective, reciprocal investment cease to be problematic because coalitions no longer need to bear the explanatory weight that past formulations placed on them. As Silk et al. (2004) have demonstrated, coalitions are best seen as mutualistic interactions that females adopt for their immediate individual benefit rather than as the foundation on which female relationships are built. Indeed, it is interesting to note that the link between coalitions, grooming, and relationships appears to have arisen initially because of an assumption that reciprocal altruism and the iterated prisoner’s dilemma (Axelrod, 1984; Axelrod and Hamilton, 1981; Trivers, 1971) were the best theoretical models to explain this behavior (Cheney and Seyfarth, 1984) rather than mutualisms. The short‐term costs that females bear when aiding their coalition partners, and the inherent problems of cheating that occur within the prisoner’s dilemma thus generated, are only avoided by the iteration and reciprocation of such acts over time (which, in effect, transforms the payoff matrix into a mutualism game). Females need to form

72

S. PETER HENZI AND LOUISE BARRETT

long‐term, temporally consistent relationships (i.e., iterated interactions with the same partner over time) because this is the only way to offset their short‐term costs. If coalitions are, in fact, mutually beneficial at the time, and there are no short‐term costs to offset, then long‐term relationships become unnecessary to explain female social functioning in the context of coalition formation. In addition to coalitions, the apparent tendency of many FB species to groom up the dominance hierarchy may involve a similar tendency to seek tolerance around higher‐ranked animals both for the benefits this may offer directly, and to reduce the likelihood of interference by third parties while in a dominant’s zone of tolerance. This spatial model also brings out the importance of kin as affiliative partners, both evolutionarily and ontogenetically. To the extent that inclusive fitness acts as a counterweight to a delay in—or absence of—reciprocity, association with kin allows females to breathe a little easier. The costs of maintaining proximity to kin in terms of feeding competition can be traded off against the evolutionary advantages accruing from kin selection, while spatial coalitions of matrilines arise naturally across development as mothers maintain close proximity to their offspring. Offspring learn the advantages of social support provided by their mothers, both passive and active, initially against their older siblings, but later against other individuals too. Close kin will, as empirical data suggest, be likely to show high rates of affiliation with respect to this kind of passive coalitionary behavior, but need not display any such fidelity with respect to other, more active, forms of engagement, where the value and competence of their partner may take precedence (Chapais, 2006). Learning the value of associating with particular individuals through time can, of course, be seen as a form of long‐term relationship. To reiterate, though, this remains a purely metaphorical usage, such a relationship is not something of which the animal itself needs to be aware nor which it is motivated to conserve. In essence, then, we are arguing for a more dynamic, contingent, action‐centered notion of primate social engagement that does not reify concepts that, as far as we know, have only human significance. Finally, this spatially oriented approach also fits with Silk’s et al. hypothesis (Silk et al., 1996) that reconciliation acts as a short‐term signal of benign intent. In all cases, such maneuvering is performed in the ‘‘here and now’’ without the need to posit any advanced form of future planning. Kin and non‐kin will reconcile, regardless of relationship status, because it is not long‐term relationship ‘‘value’’ or ‘‘security’’ that is at stake, but the need to remain within the ambit of protective others for the immediate advantages that proximity brings. Equally, the high number of nonreconciled conflicts may occur under conditions when individuals have been actively attempting to increase their distance from others (i.e., prior aggression is explicitly

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

73

aimed at decreasing proximity) and reconciliation would defeat the object of the exercise. Rejection of the notion that females have an overt cognitive concept of a relationship does not, therefore, mean rejecting entirely the notion that social engagement is complex, only that such complexity may be dynamic and emerge from ongoing spatial engagement in real‐time rather than reflecting complex cognitive cost‐benefit decisions arising from the integration of information across many social events and many social actors (Barrett et al., 2007). One very great advantage of thinking in this way is that it includes New World monkeys as a matter of course, whereas they have often been excluded under the standard ‘‘bonds serviced by grooming’’ model (Dunbar, 1995) because they tend not to display such intense grooming as Old World monkeys, and instead make use of short, discrete affiliative events such as ‘‘embracing’’ (Schaffner and Aureli, 2005). To summarize, our view is that the only true long‐term stable bonds shown by FB primates are with the group as a whole, which, after all, is what we would expect from female philopatry. Within a group, the unpredictable nature of survival and fission; the variability in births over time; and the stochastic nature of male immigration, infanticide, and other demographic events preclude the formation of life‐long enduring bonds for most females and promote a short‐term, more instrumental approach to social interactions with others. In this respect, it is interesting that Silk (2002; Silk et al., 2006a,b) argues so strongly for the existence of ‘‘relationships’’ as real entities in the life of female primates, rather than as human analytical abstractions, given her views on the short‐term value of reconciliation. This may be because she views them in purely evolutionary terms although, even then, her own argument about reconciliation would apply: the selection pressures acting on the proximate mechanisms by which females engage with each other may be very different from the evolutionary forces that have shaped fitness‐enhancing sociality in general. We argue that female ‘‘relationships,’’ as seen by females themselves, need not, and probably do not, take the long‐term, temporally consistent form that has been attributed to them, and that our argument for ‘‘expedient, quotidian cognition’’ (Barrett and Henzi, 2005), with short‐term contingent response to current need, may provide a more satisfactory evolutionary account of female coexistence and cognitive capacity.

VI. CONCLUSIONS Among the measures of success of any decent heuristic schema is that it is self‐limiting. That is, it will, sooner or later, generate information that cannot readily be shoehorned into the existing framework. Any successful

74

S. PETER HENZI AND LOUISE BARRETT

accommodation of discrepant data requires a transformation or reconfiguration of the very account that generated the data in the first place and paves the way for a new enterprise (Kuhn, 1962). Our proposition is that this is happening to the program that set out to understand local resource competition in FB groups as a social problem that is mediated by conjoint action. The linkage of observable social acts—grooming, reconciliation, and coalitions—to one another through relationships has structured our view of social life and made possible the serious consideration of natural cognition. At the same time, though, it is a formulation that is consistently situated, inadvertently or not, in the structuralist framework that has been so congenial to primatologists from the beginning. So, certain ecologies or resource bases (‘‘clumped’’/‘‘dispersed’’) beget particular social forms (‘‘uni‐male’’/ ‘‘multi‐ male’’; ‘‘nepotistic’’/‘‘egalitarian’’) that make specific demands on participants. The problem is that in a world sufficiently consistent to favor consistent response, selection for the individual propensities on which this must be built is not likely to favor intelligence, and hence larger brain size, because evolved rules of thumb will be more cost efficient (Plotkin, 1994). The reality, of course, is that a monkey’s world is full of surprises and it is this, along with a resource base that can underpin the evolution of large brains (Fish and Lockwood, 2003), that makes cognitive flexibility profitable. In the first place, surprises are ecological since resource structure and gross availability vary both temporally (Barrett et al., 2003; Koenig, 2000) and spatially (Henzi et al., 1992) within the home range. Second, surprises are social. If the value of a particular behavior depends, in large part, on what others are doing (Sutherland, 1996), then a monkey can expect to be confronted by less predictable, contingent responses. As we have argued (Barrett and Henzi, 2005), this all points to the expectation that ambitious monkeys will not or cannot rely on a suite of consistent responses but must tailor their behavior to immediate, local circumstances and affordances. It is therefore not surprising that coalitions are infrequent and that reconciliation does not always occur. For this reason, it seems to us that good research dividends can now be had by viewing female social behavior more globally, so that the deployment of all alternative responses is empirically contextualized. It may help to think of this as accentuating the negative, since the point is that the absence of reconciliation, for example, does not necessarily signal the absence of an appropriate or optimal response to local circumstances. As a final coda, this does not mean necessarily that response repertoires are constraint‐free. It is likely that founder‐effects and evolved differences among demes will tune social reaction norms (Kappeler and van Schaik, 2002) and produce inter‐ and intraspecific differences in both response range (Henzi and Barrett, 2005) and the

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

75

general tenor or style of engagement (Thierry, 1985). A more flexible account of how female monkeys manage their lives together that can incorporate these differences is the ambitious task that now presents itself. VII. SUMMARY In mammals, social groups based on female philopatry (FB societies) are common only among New and Old World monkeys. They have, for this reason, featured prominently in attempts to understand the mechanisms by which individual group members manage local resource competition without threatening group cohesion. It has been argued that this balancing act is achieved through the use of grooming to develop, sustain, and repair long‐term relationships that might prove to be valuable in the future as the basis of coalitions. This proposition, in turn, has been used to explain the evolution of large primate brains because of the cognitive demands that would be necessary to track the affiliation patterns of other group members in relation to one’s own and to plan for uncertain futures. We argue, on the one hand, that there is little evidence for either the cognitive skills or the ‘‘relationships’’ that this account requires and, on the other, that there is too little evidence to link grooming, reconciliation, and coalition formation in this way. The evidence suggests to us that patterns of grooming and reconciliation reflect responses to immediate problems and have short‐term benefits, while the general rarity of coalition formation, which may be due to a corresponding rarity of circumstances in which short‐term benefits could accrue to all participants, undercuts any role it might have as an organizing principle for evolutionary theories of female action. We conclude that there is a good deal of empirical hay to be made by incorporating observed patterns of grooming, reconciliation, and coalition formation into a larger account of the negotiation of coexistence by female monkeys. In such an account, the absence of an apparently strategic behavior will carry an analytical weight at least equal to its presence, and the central objective will be to understand how actors generate appropriate, contingent responses to immediate social problems.

Acknowledgments We would like to thank Professor Tim Roper for this opportunity to develop our ideas and Dr. Drew Rendall for his constructive and very valuable criticisms of the chapter. The NRF of South Africa has funded our long‐term empirical research.

76

S. PETER HENZI AND LOUISE BARRETT

References Altmann, J. (1980). ‘‘Baboon Mothers and Infants.’’ Harvard University Press, Cambridge, MA. Altmann, J., and Alberts, S. C. (2003). Variability in reproductive success viewed from a life‐ history perspective in baboons. Am. J. Hum. Biol. 15, 401–409. Aureli, F. (1992). Post‐conflict behavior among wild long‐tailed macaques (Macaca fascicularis). Behav. Ecol. Sociobiol. 31, 329–337. Aureli, F., and de Waal, F. B. M. (2000). ‘‘Natural Conflict Resolution.’’ University of California Press, Berkeley, CA. Aureli, F., Cords, M., and van Schaik, C. P. (2002). Conflict resolution following aggression in gregarious animals: A predictive framework. Anim. Behav. 64, 325–343. Aureli, F., Das, M., and Veneema, H. C. (1997). Differential kinship effect on reconciliation in three species of macaques (Macaca fascicularis, M. fuscata and M. sylvanus). J. Comp. Psychol. 111, 91–145. Aureli, F., van Schaik, C. P., and van Hooff, J. A. R. A. M. (1989). Functional aspects of reconciliation among captive long‐tailed macaques (Macaca fascicularis). Am. J. Primatol. 19, 39–51. Axelrod, R. (1984). ‘‘The Evolution of Cooperation.’’ Basic Books, New York, NY. Axelrod, R., and Hamilton, W. D. (1981). The evolution of cooperation. Science 221, 1390–1396. Barton, R. (1985). Grooming site preferences in primates and their functional implications. Int. J. Primatol 6, 519–532. Barrett, L., and Henzi, S. P. (2001). The utility of baboon grooming. In ‘‘Economics in Nature’’ (R. Noe¨, J. A. R. A. M. van Hooff, and P. Hammerstein, Eds.), pp. 119–145. Cambridge University Press, Cambridge. Barrett, L., and Henzi, S. P. (2002). Temporal constraints on relationship formation among female primates. Behaviour 139, 263–289. Barrett, L., and Henzi, S. P. (2005). The social nature of primate cognition. Proc. Roy. Soc. Lond. B 272, 1865–1876. Barrett, L., and Henzi, S. P. (2006). Monkeys, markets and minds. In ‘‘Cooperation in Primates and Humans: Mechanisms and Evolution’’ (P. Kappeler and C. P. van Schaik Eds.), pp. 209–232. Springer, Berlin. Barrett, L., Gaynor, D., and Henzi, S. P. (2002). A dynamic interaction between aggression and grooming among female chacma baboons. Anim. Behav. 63, 1047–1053. Barrett, L., Henzi, S. P., and Dunbar, R. I. M. (2003). Primate cognition: From ‘‘what now?’’ to ‘‘what if?’’ Trends Cogn. Sci. 7, 494–497. Barrett, L., Henzi, S. P., and Rendall, D. (2007). Social brains, simple minds: Does social complexity really require cognitive complexity? Phil. Trans. Roy. Soc. Lond. B 362, 561–575. Barrett, L., Henzi, S. P., Weingrill, A., Lycett, J. E., and Hill, R. A. (1999). Market forces predict grooming reciprocity in female baboons. Proc. Roy. Soc. Lond. B 266, 665–670. Beehner, J. C., Bergman, T. J., Cheney, D. L., Seyfarth, R. M., and Whitten, P. L. (2005). The effect of new alpha males on female stress in free‐ranging baboons. Anim. Behav. 69, 1211–1221. Call, J., Aureli, F., and de Waal, F. B. M. (1999). Reconciliation patterns among stumptail macaques: A multivariate approach. Anim. Behav. 58, 165–172. Castles, D., and Whiten, A. (1998). Post‐conflict behaviour of wild olive baboons. I. Reconciliation, redirection and consolation. Ethology 104, 126–147.

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

77

Chapais, B. (1992). Role of alliances in the social inheritance of rank among female primates. In ‘‘Coalitions and Alliances in Humans and Other Animals’’ (A. H. Harcourt and F. B. M. de Waal, Eds.), pp. 29–59. Oxford University Press, New York, NY. Chapais, B. (2006). Kinship, competence and cooperation in primates. In ‘‘Cooperation in Primates and Humans: Mechanisms and Evolution’’ (P. Kappeler and C. P. van Schaik, Eds.), pp. 47–66. Springer, Berlin. Cheney, D. L. (1977). The acquisition of rank and the development of reciprocal alliances among free‐ranging immature baboons. Behav. Ecol. Sociobiol. 2, 303–318. Cheney, D. L., and Seyfarth, R. M. (1984). Grooming, alliances and reciprocal altruism in vervet monkeys. Nature 308, 541–543. Cheney, D. L., and Seyfarth, R. M. (2005). Constraints and preadaptations in the earliest stages of language evolution. Linguistic Rev. 22, 135–159. Cords, M. (1988). Resolution of aggressive conflicts by juvenile long‐tailed macaques (Macaca fascicularis). Anim. Behav. 36, 1124–1135. Cords, M. (1997). Friendships, alliances, reciprocity and repair. In ‘‘Machiavellian Intelligence II. Extensions and Evaluations’’ (A. Whiten and R. W. Byrne, Eds.), pp. 24–49. Cambridge University Press, Cambridge. Cords, M., and Aureli, F. (1996). Reasons for reconciling. Evol. Anthrop. 5, 42–45. Cords, M., and Thurnheer, S. (1993). Reconciliation with valuable partners by long‐tailed macaques. Ethology 93, 315–325. Crook, J. H., and Gartlan, J. S. (1966). Evolution of primate societies. Nature 210, 1200–1203. Demaria, C., and Thierry, B. (2001). A comparative study of reconciliation in rhesus and Tonkean macaques. Behaviour 138, 397–410. de Waal, F. B. M. (1986). The integration of dominance and social bonding in primates. Q. Rev. Biol. 61, 459–479. de Waal, F. B. M. (1989). ‘‘Peacemaking among Primates.’’ Harvard University Press, Cambridge, MA. de Waal, F. B. M. (1993). Reconciliation among primates: A review of empirical evidence and unresolved issues. In ‘‘Primate Social Conflict’’ (W. A. Mason and S. P. Mendoza, Eds.), pp. 111–144. State University of New York Press, Albany, NY. de Waal, F. B. M. (2000). The first kiss: Foundations of conflict resolution research in animals. In ‘‘Natural Conflict Resolution’’ (F. Aureli and F. B. M. de Waal, Eds.), pp. 13–33. University of California Press, Berkeley, CA. de Waal, F. B. M., and Aureli, F. (1997). Conflict resolution and distress alleviation in monkeys and apes. Ann. N. Y. Acad. Sci. 807, 317–328. de Waal, F. B. M., and Ren, R. (1988). Comparison of the reconciliation behavior of stumptail and rhesus macaques. Ethology 78, 129–142. de Waal, F. B. M., and van Roosmalen, A. (1979). Reconcilation and consolation among chimpanzees. Behav. Ecol. Sociobiol. 5, 55–66. Di Bittetti, M. S. (1997). Evidence for an important social role of grooming in a platyrrhine primate. Anim. Behav. 54, 199–211. Di Fiore, A., and Rendall, D. (1994). Evolution of social organisation: A reappraisal for primates by using phylogenetic methods. Proc. Natl. Acad. Sci. USA 91, 9941–9945. Dittus, W. P. J. (1977). The social regulation of population density and age‐sex distribution in the toque monkey. Behaviour 63, 281–322. Dunbar, R. I. M. (1988). ‘‘Primate Social Systems.’’ Cornell University Press, Ithaca, NY. Dunbar, R. I. M. (1991). Functional significance of social grooming in primates. Folia Primatol. 57, 121–131. Dunbar, R. I. M. (1992). Time: A hidden constraint on the behavioral ecology of baboons. Behav. Ecol. Sociobiol. 31, 35–49.

78

S. PETER HENZI AND LOUISE BARRETT

Dunbar, R. I. M. (1995). Neocortex size and group size in primates: A test of the hypothesis. J. Hum. Evol. 28, 287–296. Dunbar, R. I. M. (1998). The social brain hypothesis. Evol. Anthrop. 6, 178–190. Eisenberg, J. F., Muckenhirn, N. A., and Rudran, R. (1972). The relation between ecology and social structure in primates. Science 176, 863–874. Emlen, S. W., and Oring, L. W. (1977). Ecology, sexual selection and the evolution of mating systems. Science 197, 215–223. Engh, A. L., Beehner, J. C., Bergman, T. J., Whitten, P. L., Hoffmeier, R. R., Seyfarth, R. M., and Cheney, D. L. (2005). Behavioural and hormonal responses to predation in female chacma baboons (Papio hamadryas ursinus). Proc. Roy. Soc. Lond. B 273, 701–712. Fish, J. L., and Lockwood, C. A. (2003). Dietary constraints on encephalization in primates. Am. J. Phys. Anthropol. 120, 171–181. Ga¨rdenfors, P. (1995). ‘‘Cued and detached representations in animal cognition.’’ Lund University Cognitive Studies 38. Grafen, A. (1984). Natural selection, kin selection and group selection. In ‘‘Behavioral Ecology’’ (J. R. Krebs and N. B. Davies, Eds.), pp. 62–84. Sinauer, Sunderland, MA. Hamilton, W. D. (1964). The genetical evolution of social behaviour I and II. J. Theor. Biol. 7, 1–16 and 17–52. Hamilton, W. D. (1971). Geometry for the selfish herd. J. Theor. Biol. 31, 295–311. Hamilton, W. D. (1972). Altruism and related phenomena, mainly in social insects. Ann. Rev. Syst. Ecol. 3, 193–232. Hampton, R. R., Hampstead, B. M., and Murray, E. A. (2005). Rhesus monkeys (Macaca mulatta) demonstrate robust memory for what and where, but not when, in an open‐field test. Learn. Motiv. 36, 245–259. Harcourt, A. H., and de Waal, F. B. M. (1992). ‘‘Coalitions and Alliances in Humans and Other Animals.’’ Oxford University Press, New York, NY. Henzi, S. P., and Barrett, L. (1999). The value of grooming to female primates. Primates 40, 47–59. Henzi, S. P., and Barrett, L. (2002). Infants as a commodity in a baboon market. Anim. Behav. 63, 915–921. Henzi, S. P., and Barrett, L. (2003). Evolutionary ecology, sexual conflict and behavioral differentiation among baboon populations. Evol. Anthrop. 12, 217–230. Henzi, S. P., and Barrett, L. (2005). The historical socio‐ecology of savannah baboons. J. Zool. Lond. 265, 215–226. Henzi, S. P., Barrett, L., Gaynor, D., Greeff, J., Weingrill, T., and Hill, R. A. (2003). The effect of resource competition on the long‐term allocation of grooming by female baboons: Evaluating Seyfarth’s model. Anim. Behav. 66, 931–938. Henzi, S. P., Byrne, R. W., and Whiten, A. (1992). Patterns of movement by baboons in the Drakensberg mountains: Primary responses to the environment. Int. J. Primatol. 13, 601–629. Henzi, S. P., and Lawes, M. (1987). Breeding season influxes and the behaviour of adult male samango monkeys (Cercopithecus mitis albogularus). Folia Primatol. 48, 125–136. Hinde, R. A. (1976). Interactions, relationships and social structure. Man 11, 1–17. Hinde, R. A. (1981). The bases of a science of interpersonal relationships. In ‘‘Personal Relationships 1: Study of Personal Relationships’’ (S. Duck and R. Gilmour, Eds.), pp. 1–22. Academic Press, New York, NY. Hinde, R. A. (1983). ‘‘Primate Social Relationships: An Integrated Approach.’’ Blackwell Scientific, Oxford. Holekamp, K. E. (2007). Questioning the social intelligence hypothesis. Trends Cogn. Sci. 11, 65–69.

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

79

Holekamp, K. E., Sakai, S. T., and Lundrigan, B. L. (2007). Social intelligence in the spotted hyena. Phil. Trans. Roy. Soc. Ser. B 362, 523–528. Judge, P. G. (1991). Dyadic and triadic reconciliation in pigtail macaques (Macaca nemestrina). Am. J. Primatol. 23, 225–237. Kappeler, P. M. (1993). Reconciliation and post‐conflict behaviour in ringtailed lemurs, Lemur catta, and redfronted lemurs, Eulemur fulvis rufus. Anim. Behav. 45, 901–915. Kappeler, P. M., and van Schaik, C. P. (1992). Methodological and evolutionary aspects of reconciliation among primates. Ethology 92, 51–69. Kappeler, P. M., and van Schaik, C. P. (2002). Evolution of primate social systems. Int. J. Primatol. 23, 707–740. Kapsalis, E., and Berman, C. M. (1996). Models of affiliative relationships among free‐ranging rhesus monkeys (Macaca mulatta) II. Testing predictions for three hypothesized organizing principles. Behaviour 133, 1235–1263. Koenig, A. (2000). Competitive regimes in forest‐dwelling Hanuman langur females (Semnopithecus entellus). Behav. Ecol. Sociobiol. 48, 93–109. Koenig, A., Borries, C., Doran‐Sheehy, D. M., and Janson, C. H. (2006). How important are affiliation and cooperation? A Reply to Sussman et al. Am. J. Phys. Anthropol. 131, 522–523. Korte, S. M., Koolhaas, J. M., Wingfield, J. C., and McEwen, B. S. (2005). The Darwinian concept of stress: Benefits of allostasis and costs of allostatic load and the trade‐offs in health and disease. Neurosci. Biobehav. Rev. 29, 3–38. Kuhn, T. S. (1962). ‘‘The Structure of Scientific Revolutions.’’ University of Chicago Press, Chicago, IL. Lusseau, D., and Newman, M. (2004). Identifying the role that animals play in their social networks. Proc. Roy. Soc. Lond. B 271, S477–S481. Manson, J. H., Navett, C. D., Silk, J. B., and Perry, S. (2004). Time‐matched grooming in female primates? New analyses from two species. Anim. Behav. 67, 493–500. Manson, J. H., Rose, L. M., Perry, S., and Gros‐Louis, J. (1999). Dynamics of female–female relationships in wild Cebus capucinus: Data from two Costa Rican sites. Int. J. Primatol. 20, 679–706. Matsumura, S. (1998). Relaxed dominance relations among female moor macaques (Macaca maurus) in their natural habitat, south Sulawesi, Indonesia. Folia Primatol. 69, 346–356. Muroyama, Y. (1994). Exchange of grooming for allomothering in female patas monkeys. Behaviour 128, 103–119. Nash, L. (1976). Troop fission in free‐ranging baboons in the Gombe Stream National Park, Tanzania. Am. J. Phys. Anthropol. 44, 63–78. Noe¨, R., and Hammerstein, P. (1994). Biological markets: Supply and demand determine the effect of partner choice in cooperation, mutualism and mating. Behav. Ecol. Sociobiol. 35, 1–11. Noe¨, R., and Hammerstein, P. (1995). Biological markets. TREE 10, 336–339. O’Brien, T. G. (1993). Asymmetries in grooming interaction between juvenile and adult female wedge‐capped capuchin monkeys. Anim. Behav. 46, 929–938. O’Brien, T. G., and Robinson, J. G. (1993). Stability of social relationships in female wedge‐ capped capuchin monkeys. In ‘‘Juvenile Primates: Life History, Development and Behaviour’’ (M. E. Pereira and L. A. Fairbanks, Eds.), pp. 197–210. Oxford University Press, New York. Owren, M. J., and Rendall, D. (2001). Sound on the rebound: Bringing form and function back to the forefront in understanding non‐human primate vocal signaling. Evol. Anthropol. 10, 58–71.

80

S. PETER HENZI AND LOUISE BARRETT

Parr, L. A., Matheson, M. D., Bernstein, I. S., and de Waal, F. B. M. (1997). Grooming down the hierarchy: Allogrooming in captive brown capuchin monkeys, Cebus apella. Anim. Behav. 54, 361–367. Payne, H., Lawes, M., and Henzi, S. P. (2003). Competition and the exchange of grooming among female samango monkeys (Cercopithecus mitis). Behaviour 140, 453–471. Petit, O., and Thierry, B. (1994). Reconciliation in a group of Guinea baboons. In ‘‘Current Primatology, Vol. II: Social Development, Learning and Behavior’’ (J. J. Roeder, B. Thierry, J. R. Anderson, and N. Herrenschmidt, Eds.), pp. 137–145. Presses de l’universite Louis Pasteur, Strasbourg. Plotkin, H. (1994). ‘‘Darwin Machines and the Nature of Knowledge.’’ Penguin Books, London. Rendall, D., Rodman, P. S., and Emond, R. E. (1996). Vocal recognition of individuals and kin in free‐ranging rhesus monkeys. Anim. Behav. 51, 1007–1015. Roberts, W. A. (2002). Are animals stuck in time? Psychol. Bull. 128, 473–489. Ron, T., Henzi, S. P., and Motro, U. (1994). A new model of fission in primate troops. Anim. Behav. 47, 223–226. Ron, T., Henzi, S. P., and Motro, U. (1996). Do female chacma baboons compete for a safe spatial position in a southern woodland habitat? Behaviour 133, 475–490. Sambrook, T. D., Whiten, A., and Strum, S. C. (1995). Priority of access and grooming patterns of females in a large and a small group of olive baboons. Anim. Behav. 50, 1667–1682. Schaffner, C., and Aureli, F. (2005). Embraces and grooming in captive spider monkeys. Int. J. Primatol. 26, 1093–1106. Schaffner, C., Aureli, F., and Caine, N. (2001). Following the rules: Why tamarins don’t reconcile. Int. J. Primatol. 54(Suppl. 1), 49–50. Schino, G. (2001). Grooming, competition and social rank among female primates: A meta‐ analysis. Anim. Behav. 62, 265–271. Seyfarth, R. M. (1976). Social relationships among adult female baboons. Anim. Behav. 24, 917–938. Seyfarth, R. M. (1977). A model of social grooming among adult female monkeys. J. Theor. Biol. 65, 671–698. Seyfarth, R. M. (1980). The distribution of grooming and related behaviours among adult female vervet monkeys. Anim. Behav. 28, 798–813. Silk, J. B. (1992). Patterns of intervention in agonistic contests among male bonnet macaques. In ‘‘Coalitions and Alliances in Humans and Other Animals’’ (A. H. Harcourt and F. B. M. de Waal, Eds.), pp. 215–232. Oxford University Press, Oxford. Silk, J. B. (1996). Why do primates reconcile? Evol. Anthropol. 5, 39–42. Silk, J. B., Cheney, D. L., and Seyfarth, R. M. (1996). The form and function of post‐conflict interactions between female baboons. Anim. Behav. 52, 259–268. Silk, J. B. (1997). The function of peaceful post‐conflict interactions among primates. Primates 38, 265–279. Silk, J. B., Cheney, D. L., and Seyfarth, R. M. (1999). The structure of social relationships among female savannah baboons in Moremi Reserve, Botswana. Behaviour 136, 679–703. Silk, J. B. (2000). The function of peaceful post‐conflict interactions among primates: An alternative view. In ‘‘Natural Conflict Resolution’’ (F. Aureli and F. B. M. de Waal, Eds.), pp. 179–181. University of California Press, Berkeley, CA. Silk, J. B. (2002a). Practice random acts of aggression and senseless acts of intimidation: The logic of status contests in animals. Evol. Anthropol. 11, 221–225. Silk, J. B. (2002b). The form and function of reconciliation in primates. Annu. Rev. Anthropol. 31, 21–44. Silk, J. B. (2002c). Using the F‐word in primatology. Behaviour 139, 421–446.

COEXISTENCE IN FEMALE‐BONDED PRIMATE GROUPS

81

Silk, J. B., Alberts, S. C., and Altmann, J. (2003). Social bonds of female baboons enhance infant survival. Science 302, 1231–1234. Silk, J. B., Alberts, S. C., and Altmann, J. (2004). Patterns of coalition formation by adult female baboons in Amboseli, Kenya. Anim. Behav. 67, 573–582. Silk, J. B., Altmann, J., and Alberts, S. C. (2006a). Social relationships among adult female baboons (Papio cynocephalus). I. Variation in the strength of social bonds. Behav. Ecol. Sociobiol. 61, 183–195. Silk, J. B., Alberts, S. C., and Altmann, J. (2006b). Social relationships among adult female baboons (Papio cynocephalus). II. Variation in the quality and stability of social bonds. Behav. Ecol. Sociobiol. 61, 197–204. Sinha, A. (1998). Knowledge acquired and decisions made: Triadic interactions during allogrooming in wild bonnet macaques, Macaca radiata. Phil. Trans. Roy. Soc. Lond. B 353, 619–631. Stammbach, E. (1988). Group responses to specially skilled individuals in a Macaca fascicularis group. Behaviour 107, 241–266. Sterck, E. H. M., Watts, D. P., and van Schaik, C. P. (1998). The evolution of female social relationships in nonhuman primates. Behav. Ecol. Sociobiol. 41, 291–309. Strier, K. B. (1994). Myth of the typical primate. Ybk. Phys. Anthropol. 37, 233–271. Strier, K. B. (1999). Why is female kin bonding so rare: Comparative sociality of New World primates. In ‘‘Primate Socioecology’’ (P. C. Lee, Ed.), pp. 300–319. Cambridge University Press, Cambridge. Sussman, R. W., Garber, P. A., and Cheverud, J. M. (2005). Importance of cooperation and affiliation in the evolution of primate sociality. Am. J. Phys. Anthropol. 128, 84–97. Sutherland, W. J. (1996). ‘‘From Individual Behaviour to Population Ecology.’’ Oxford University Press, Oxford. Thierry, B. (1985). Patterns of agonistic interactions in three species of macaque (Macaca mulatta, M. fascicularis, M. tonkeana). Aggress. Behav. 11, 223–233. Thompson, R. K. R. (1995). Natural and relational concepts in animals. In ‘‘Comparative Approaches to Cognitive Science’’ (H. L. Roitblat and J.‐A. Meyer, Eds.), pp. 175–224. MIT Press, Cambridge, MA. Thompson, R. K. R., and Oden, D. L. (2000). Categorical perception and conceptual judgments by nonhuman primates: The paleological monkey and the analogical ape. Cognitive Sci. 24, 363–396. Tinbergen, N. (1963). On aims and methods in ethology. Z. Tierpsychol. 20, 410–433. Trivers, R. (1971). The evolution of reciprocal altruism. Q. Rev. Biol. 46, 35–57. Tulving, E. (1983). ‘‘Elements of Episodic Memory.’’ Clarendon Press, Oxford. van Schaik, C. P. (1983). Why are diurnal primates living in groups? Behaviour 87, 120–144. van Schaik, C. P. (1989). The ecology of social relationships amongst primate females. In ‘‘Comparative Socioecology: The behavioral Ecology of Humans and Other Mammals’’ (V. Standen and R. A. Foley, Eds.), pp. 195–218. Blackwell Scientific, Oxford. Whiten, A. (2000). Social complexity and social intelligence. In ‘‘The Nature of Intelligence’’ (G. R. Bock, J. A. Goode, and K. Webb, Eds.), pp. 185–201. Novartis Foundation Symposium 233, Wiley Interscience, London. Wilson, E. O. (1975). ‘‘Sociobiology: The New Synthesis.’’ Belknap Press, Cambridge, MA. Wittig, R. M., Crockford, C., Seyfarth, R. M., and Cheney, D. L. (2007). Vocal alliances in chacma baboons (Papio hamadryas ursinus). Behav. Ecol. Sociobiol. (in press). Wrangham, R. W. (1980). An ecological model of female‐bonded primate groups. Behaviour 75, 262–300. Wrangham, R. W. (1987). Evolution of social structure. In ‘‘Primate Societies’’ (B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham, and T. T. Struhsaker, Eds.), pp. 282–296. University of Chicago Press, Chicago, IL.

This page intentionally left blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

The Evolution of Sociality in Spiders Yael Lubin* and Trine Bilde{ *blaustein institutes for desert research, ben‐gurion university of the negev, sede boqer campus, 84990 israel { department of biological sciences, university of aarhus, denmark

I. INTRODUCING SOCIAL SPIDERS A solitary lifestyle characterizes the vast majority of almost 40,000 known species of spiders (Platnick, 2007). Thus, the occurrence of group living in spiders begs the question: what is different about these species? Group living has arisen in spiders in basically two different forms. Cooperative or ‘‘nonterritorial permanent‐social’’ species (sensu Avile´s, 1997; also referred to as ‘‘quasi‐social,’’ Buskirk, 1981) are the main focus of this chapter. These species have family‐group territories consisting of communal nests and capture webs, which they inhabit throughout the entire lifetime of the individual, and colony members cooperate in foraging and raising young. In many ways, these species resemble the ‘‘primitively eusocial’’ wasps and bees and the cooperative breeders in vertebrate societies, where the family forms the basic unit of sociality (Brockmann, 1997; Whitehouse and Lubin, 2005). Another form of group living in spiders has been termed colonial or communal‐ territorial (Avile´s, 1997: ‘‘territorial permanent‐social’’ species). Colonial species occur in aggregations, but individuals in the colony generally forage and feed alone and there is no maternal care beyond the egg stage; thus, they lack the cooperative behaviors described below for nonterritorial permanent‐ social species (reviewed in Uetz and Hieber, 1997; Whitehouse and Lubin, 2005). Colonial species have been likened to foraging flocks of birds (Rypstra, 1979) and are described as ‘‘foraging societies’’ by Whitehouse and Lubin (2005). Colonies range from temporary aggregations under favorable conditions to large, long‐lived structures. Groups are formed parasocially, by aggregation of individuals, and their composition is variable (Lubin, 1980; Uetz and Hieber, 1997). Although colonies may include related individuals, there is no evidence of inbreeding (Uetz et al., 1987), a trait that distinguishes 0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37003-4

83

Copyright 2007, Elsevier Inc. All rights reserved.

84

YAEL LUBIN AND TRINE BILDE

them from nonteritorial permanent‐social species (see below). Coloniality is the more common form of group living in spiders: Whitehouse and Lubin (2005) list 53 species in 12 families, including both web building and non‐web building species, which exhibit temporary or permanent coloniality. The list greatly underestimates the number of colonial species, as most occur in tropical and subtropical regions, and the habits of most tropical species are totally unknown. Nonterritorial permanent‐social spiders (henceforth referred to as ‘‘social’’ spiders) inhabit colonies that last for up to a few generations, and adjacent, interconnected colonies may form a colony cluster (Avile´s, 1997). Marques et al. (1998) defined a colony as ‘‘a group of individuals that occupy a single web (colonial web) which was spun and maintained by these same individuals,’’ but there are also some social species lacking capture webs. Within colonies, individuals cooperate constructing the capture web and communal nest or retreat and in capturing prey. Young remain in the colony and accrue tasks as they mature; a proportion of females may act as helpers, but there is no division of labor in the sense of social insect castes (Franks, 1987). Unlike the vast majority of cooperatively breeding vertebrates and social insects, which have outbred mating systems, in the social spiders mating occurs within colonies among individuals of the same cohort, and adult, mated females disperse from the mother colony singly or in groups to establish new colonies. Regular inbreeding is in fact a key characteristic of this type of sociality in spiders. Exceptions may be found to one or several of the above traits in different species, but on the whole, the constellation of all or most of these traits characterizes social spider species from a wide range of families. Social species are rare, even by comparison with colonial species: fewer than 25 species are known to be social (Agnarsson et al., 2006; Avile´s, 1997; Table I). Despite the small number, social species are widely distributed across taxonomically distinct families, with social representatives found in nine genera in six families [Table I; one species, Delena cancerides, has been redefined as subsocial, making one less family and genus by comparison with Agnarsson et al. (2006)]. They are also widely distributed geographically: social species occur in the tropics and subtropics of every continent, with a particularly high concentration in the Amazonian region of South America (see Avile´s, 1997). Furthermore, there is now evidence to support as many as 18 independent origins of sociality (Agnarsson et al., 2006). The majority of social species are found within the family Theridiidae, which contains 11–12 social species in 3 genera representing 9 independent origins of sociality (Agnarsson, 2006; Avile´s, 1997; Avile´s et al., 2006). The second most abundant group, the Eresidae, contains three independent origins of sociality all found within the genus Stegodyphus (Johannesen et al., 2007; Kraus and Kraus, 1988). Thus, social spiders can be viewed as an ideal evolutionary experiment with independent replicates both within and

TABLE I BEHAVIORAL AND LIFE‐HISTORY CHARACTERISTICS OF SOCIAL SPIDERS

Family

Species

Colony compositiona

Modec

Sex ratiod (% males)

85

f

8–18%b

mf,y mf,y mf,y

? pa,a ?

f f, g ?

mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y mf,y m þ y,y

a (pm) a (pm) y?, a (pm) pa,a? ? ? a (pm) ? pa,a a (pm) ? ? ? ? ? a? ? a (pm)

a(rare?) a ? pa,a? ? ? a (rare) ? a? no (rare?) ? ? ? ? ? ? ? ?

f,p, s,g f,p, s,g f,p s,g f ? ? f,p, g ? s,g f,p s,g ? f? ? ? f s?,g? s? s?

16.5–22%b, 0–30%b 8.5%a 0–18.5%b (captive colonies) 17%a, 12%b 9%b 22%a, 8.5–33%b 50%a ? 5%b 12%a 11%b, 8%a 31%b 8%a Yes c.2%b ? 14%b 8%b 11%b 20%a 28%a

mf,y

Dictynidae

A. republicana Aebutina binotata Mallos gregalis Stegodyphus dumicola Stegodyphus mimosarum Stegodyphus sarasinorum Tapinillus sp. 1* Achaearanea disparata Achaearanea vervoorti Achaearanea wau Anelosimus domingo Anelosimus dubiosus Anelosimus eximius Anelosimus guacamayos Anelosimus lorenzo Anelosimus oritoyacu Anelosimus puravida Anelosimus rupununi Theridion nigroannulatum Diaea megagyna Diaea socialis

Thomisidae

Males ?

Agelena consociata

Oxyopidae Theridiidae

Females a; all stages (inferred) ? pa,a a?

Agelenidae

Eresidae

Colony foundationb

(Continued)

TABLE I (Continued) Group activities Family

Species

Agelenidae

Agelena consociata A. republicana Aebutina binotata Mallos gregalis Stegodyphus dumicola Stegodyphus mimosarum Stegodyphus sarasinorum Tapinillus sp. 1 Achaearanea disparata Achaearanea vervoorti Achaearanea wau Anelosimus domingo Anelosimus dubiosus Anelosimus eximius Anelosimus guacamayos Anelosimus lorenzo Anelosimus oritoyacu Anelosimus puravida Anelosimus rupununi Theridion nigroannulatum Diaea megagyna Diaea socialis

Dictynidae Eresidae

86

Oxyopidae Theridiidae

Thomisidae

Cooperatione w,p w,p w,p w,p w,d,p w,d,p w,d,p w,d w,p w,p w,p p w,p w,d,p ? w,p ? ? w,p p w w,p

Helpingf ? ? e,p ? e,p,r,g e,p,r,g? e,p,r,g? ? e p e,p r? e,p,r e,p,r ? p ? ? e (sac clusters) e ? ?

Mating behavior Aggressiong

Matingh

Sexual competitioni

Reproductive skewj

0 0 0 0 0 0 0

? ? ? ? f,s,m s,m s,m f? ? ? f,s,m ? ? ? ? ? ? ? ? ? ? s?

c? ? c? rare c,f ? c,f ? ? c,f c,f,o ? ? c ? ? ? ? ? ? ? ?

? ? 1 ? about 3 ? ? ? ? ? 1.25 ? ? about 4 ? ? ? ? 0.7–1 ? ? ?

0 ? 0 0 0 0 ? 0 ? ? ? ? ? 0

Family Agelenidae

Dictynidae Eresidae

Species Agelena consociata A. republicana Aebutina binotata Mallos gregalis Stegodyphus dumicola

Stegodyphus mimosarum Stegodyphus sarasinorum 87

Oxyopidae Theridiidae

Tapinillus sp. 1 Achaearanea disparata Achaearanea vervoorti Achaearanea wau Anelosimus domingo Anelosimus dubiosus Anelosimus eximius

Anelosimus guacamayos Anelosimus lorenzo Anelosimus oritoyacu Anelosimus puravida Anelosimus rupununi Theridion nigroannulatum

Sources Darchen, 1978; Riechert et al., 1986; D’Andrea, 1987; Riechert and Roeloffs, 1993; Avile´s, 1997 Darchen, 1967; Darchen, 1976; D’Andrea, 1987, Avile´s, 1997 Avile´s, 1993a, 1997, 2000; Avile´s et al., 2001 Jackson, 1978; Jackson and Smith, 1978; Jackson, 1979 Avile´s, 1997; Avile´s et al., 1999; Henschel et al., 1995a; Henschel, 1998; Lubin and Crouch, 2003; A. Maklakov, personal communication; M. Salomon, personal communication Wickler and Seibt, 1986; Seibt and Wickler, 1988a; Crouch et al., 1998 Jacson and Joseph, 1973; Avile´s, 1997; Seibt and Wickler, 1988a Avile´s, 1994 Darchen, 1968; Darchen and Ledoux, 1978; Buskirk, 1981 Levi et al., 1982; Lubin, 1991 Lubin, 1982, 1986, 1991, 1995; Lubin and Crozier, 1985; Lubin and Robinson, 1982 Rypstra and Tirey, 1989; Aviles, 1997; Aviles et al., 2001 Marques et al., 1998 Christenson, 1984; D’Andrea, 1987; Avile´s, 1997; Avile´s and Tufin˜o, 1998; reviewed in Agnarsson, 2006; Vollrath and Windsor, 1983 Agnarsson, 2006; Agnarsson et al., 2006a Fowler and Levi, 1979; Agnarsson, 2006 Agnarsson, 2006; Agnarsson et al., 2006a Agnarsson, 2006; Agnarsson et al., 2006a Avile´s, 1997; Aviles and Salazar, 1999; Agnarsson, 2006 Avile´s, 1997; Avile´s et al., 2001; Avile´s et al., 2006 (Continued)

TABLE I (Continued) Family Thomisidae

Species Diaea megagyna Diaea socialis

Sources Evans, 1995; Avile´s, 1997 Main, 1988; Evans and Main, 1993; Avile´s, 1997

Overlapping stages: m þ y ¼ mother and young, y ¼ young, and mf ¼ multiple females (males are not included). Dispersal stage: a ¼ adult, pa ¼ preadult, and y ¼ young. Premating or postmating adult dispersal: am ¼ premating and pm ¼ postmating. c Colony foundation (the predominant mode is shown in bold): fission (budding, splitting, or translocation) ¼ f and propagule dispersal ¼ p. Propagule size: g ¼ group and s ¼ solitary. d Sex ratio: a primary or juvenile sex ratio, b subadult or adult sex ratio. e Cooperation: w ¼ web building, d ¼ defense, and p ¼ prey capture. f Helping (allomaternal): c ¼ communal brood chambers, e ¼ guard eggsacs or young, p ¼ bringing prey to nest, r ¼ regurgitation, and g ¼ gerontophagy. g Aggression among adults: 0 ¼ no aggression to foreign females, n ¼ adult nestmates, and f ¼ foreign females. h Mating: f ¼ foreign males, s ¼ sib mating, and m ¼ multiple mating. i Sexual competition: c ¼ male courtship, f ¼ male aggression or fighting, and o ¼ opportunistic mating (during female molt or feeding). j Reproductive skew: number of females per eggsac. * Species considered intermediate‐social are noted by *. a

b

88

THE EVOLUTION OF SOCIALITY IN SPIDERS

89

across several families. Because of these repeated evolutionary events, social spiders provide a remarkable system to examine the causes and consequences of inbred sociality. The main question motivating this chapter is, why has inbred sociality arisen in spiders so many times independently, and yet there are so few social species? The generally accepted hypothesis is that nonterritorial permanent‐ sociality in spiders is derived evolutionarily from a subsocial state (Avile´s, 1997: nonterritorial periodic‐social; Wickler and Seibt, 1993). Subsocial behavior with respect to spiders refers specifically to the occurrence of two traits: (1) an extended stage of maternal care of offspring prior to their dispersal and (2) a stage of cooperation among young within the brood, usually continuing after the mother dies or leaves the nest (Avile´s, 1997). A third characteristic that distinguishes nonterritorial periodic‐sociality from permanent‐sociality is the obligatory dispersal of young or of adults of one or both sexes before mating, such that regular inbreeding is avoided (Avile´s and Bukowski, 2006; Bilde et al., 2005; Powers and Aviles, 2003). In the following sections, we review the characteristics of social species and their subsocial congeners and discuss the evidence for this transition scenario. Finally, the designation of species as nonterritorial permanent‐social and nonterritorial periodic‐social applies to groups of traits that generally occur together. Some species do not fit comfortably in a single category, and these exceptions often provide insight into important adaptive and evolutionary processes. Recently, there have been renewed attempts to find a terminology of social organization that can be applied widely across animal taxa (Costa, 2006; Costa and Fitzgerald, 2005; Crespi, 2005; Wcislo, 2005) and to define the nature of social interactions more inclusively (Rayor and Taylor, 2006; West et al., 2007). We have not attempted to fit spider sociality into these schemes, as the jury is still out as to which method will prove most durable over time. We feel that the categories erected by Avile´s (1997) suit the diversity of social organization in spiders, first because they are useful descriptors of the temporal and spatial nature of the groups in each case and second, because they imply neither function nor hierarchy and, therefore, are not teleological [a valid criticism of the classic social insect‐based terminology, in Costa and Fitzgerald (2005)]. However, as this terminology is also somewhat unwieldy, for convenience we revert to the commonly used shorthand designations of social and subsocial, for nonterritorial permanent‐social and territorial periodic‐social, respectively. II. SOCIAL AND SUBSOCIAL SPECIES: A SURVEY OF BEHAVIORAL TRAITS Tables I and II outline behavioral traits that describe the degree of social integration in the colony or nest of social and subsocial species, respectively. In assigning species to the tables, we have followed the designations of

TABLE II BEHAVIORAL AND LIFE‐HISTORY CHARACTERISTICS OF SOME SUBSOCIAL SPECIES

Species

Agelenidae Amaurobiidae Desidae Eresidae Scytodidae Sparassidae Theridiidae

Coelotes terrestris Amaurobius ferox Phryganoporus candidusi Stegodyphus lineatus Scytodes socialis, S. fusca, S. pallida Delena cancerides Anelosimus arizona (formerly A. jucundus or A. cf. jucundus) Anelosimus baeza (formerly A. jucundus) Anelosimus jabaquara* Anelosimus jucundus Anelosimus studiosus (United States‐ southern populations) Anelosimus studiosus (United States‐some northern colonies)* Anelosimus cf. studiosus (Uruguay)* Helvibis thorelli Theridion pictum Diaea ergandros

90

Family

Thomisidae

Dispersalb

Group compositiona

Females

Males

Mode

Sex ratiod (% males)

c

mþy mþy mþy mþy mf,y m þ y,y,mf mþy

y y pa,y y y a?(am) pa,a (am)

y y a (am) y y a? pa,a (am)

s s s,f s s s?,g? s

? ? 57%a No bias No bias 48%a 53%b

m þ y, y, mf?

pa

?

s

No bias

m þ y,y,mf m þ y, y m þ y,y

pa,a (am, pm) pa y,pa

pa?,a? pa a?

s,g s s

36%b No bias 48%a

mf, y

pa,a

a?

s

17–24%b

m þ y,y,mf

pa,a?

pa

s

37%b

mþy mþy m þ y,y

? y a (pm)

? y a? (am)

? s s?

? 29–36%b No bias

Group activities Family

Species

Maternal caref

Cooperatione

Aggressiong

Mating behaviorh

Agelenidae

Coelotes terrestris

p

r,p,t

?

?

Amaurobiidae

Amaurobius ferox

p

r,p,t,m

?

?

Desidae

Phryganoporus candidus Stegodyphus lineatus

w,p

?

?

f,m

w?,p

r,p,m

n,f

f,s (rare),m (both sexes),c

Eresidae 91 Scytodidaec

Scytodes socialis, S. pallida, S. fusca

p

?

?

?

Sparassidae

Delena cancerides

p

p

an (adult sisters)

f,s,m

Theridiidae

Anelosimus arizona (formerly A. jucundus or A. cf. jucundus)

w,p

Yes—no details

?

f,s?

Anelosimus baeza (formerly A. jucundus)

p

p,r?

?

?

Source Krafft et al., 1986; Gundermann et al., 1991 Kim, 2000; Kim and Horel, 1998; Kim et al., 2005a,b; Kim and Roland, 2000; Kim, Roland and Horel, 2000 Downes, 1993, 1994a, 1994b Schneider, 1997; Lubin et al., 1998; Schneider and Lubin, 1998; Bilde et al., 2005 and references therein Bowden, 1991; Li et al., 1999; Li and Kuan, 2006; J. A. Miller, personal communication Rowell and Aviles, 1995; Aviles, 1997; L. Rayor, personal communication Aviles and Gelsey, 1998; Bukowski and Avile´s, 2002; Powers and Aviles, 2003; Aviles and Bukowski, 2006 Aviles et al., 2001; Agnarsson et al., in press; Agnarsson, 2006 (Continued)

TABLE II (Continued) Group activities 92 Family

Species

Cooperatione

Maternal caref

Aggressiong

Mating behaviorh

Anelosimus jabaquara

w,p

r,p, a?

n, f

f,s?,m (both sexes)

Anelosimus jucundus

p

yes

?

?

Anelosimus studiosus (United States‐ southern populations) Anelosimus studiosus (United States‐some northern colonies)

p

r,p

n,f (adult females)

f?

p

r,p,a

?

Source Marques et al., 1998; Gonzaga and Vasconcellos‐Neto, 2001 Aviles et al., 2001; Nentwig and Christenson, 1986; Agnarsson, 2006 Brach, 1977; Agnarsson, 2006; Aviles, 1997

Furey, 1998; Jones et al., 2007

Thomisidae

Anelosimus cf. studiosus (Uruguay) Helvibis thorelli

w,p

r,p

n

f

Ghione et al., 2004; Viera et al., 2005, 2007a, 2007b

p

?

?

?

Theridion pictum Diaea ergandros

w,p w

p, r? m

? n (rare), f (females)

? f,s?,m

Coddington and Agnarsson, 2006 Ruttan, 1990 Evans, 1998, 2000; Evans and Goodisman, 2002

Overlapping stages: m þ y ¼ mother and young, y ¼ young, and mf ¼ multiple females (males are not included). Dispersing stage: a ¼ adult, pa ¼ preadult, and y ¼ young. Adult dispersal: am ¼ premating and pm ¼ postmating. c Dispersal mode: b ¼ budding, f ¼ colony fission, g ¼ group dispersal, and s ¼ solitary dispersal. d Sex ratio: a ¼ primary or juvenile sex ratio and b ¼ subadult or adult sex ratio. e Cooperation: w ¼ web building, d ¼ defense, and p ¼ prey capture. f Maternal care: e ¼ guard eggsacs or young, p ¼ bringing prey to nest, r ¼ regurgitation, t ¼ trophic eggs, and m ¼ matriphagy. g Aggression among adults: 0 ¼ no aggression to foreign females, n ¼ adult nestmates, and f ¼ foreign females. h Mating: f ¼ foreign males, s ¼ sib mating, m ¼ multiple mating, and c ¼ sexual competition. i We follow Agnarsson et al. (2006a) in assigning Phryganoporus candidus (Dictynidae) to the subsocial group (contrary to Whitehouse and Lubin, 2005). * Possible subsocial‐social transition species are noted by *. a

b

93

94

YAEL LUBIN AND TRINE BILDE

Agnarsson et al. (2006) (see also Whitehouse and Lubin, 2005): social species (Plates 1 and 2) are those in which juvenile dispersal is lacking and multiple adult females breed within the colony. Thus, social species should show an inbred population structure, and this is often associated with a female‐biased primary sex ratio (Avile´s, 1986; Hurst and Vollrath, 1992). Subsocial species (Plate 3) are those with premating dispersal. Extended maternal care and some cooperation in foraging or nest activities among young prior to dispersal distinguish subsocial from solitary species (Agnarsson et al., 2006). Data on cooperative activities of young in the nest are not available for all species listed; the prolonged stay of juveniles in the maternal nest, however, can be taken as evidence of group foraging. Species that exhibit both subsocial and social traits are noted as ‘‘transition species’’ in either Tables I or II and are discussed below. Colony composition indicates the developmental stages that overlap and interact within the colony. In most of the social species, all stages overlap; in the subsocial species, usually only the mother and her young are present together (we ignore males here). Some subsocial species, however, have multiple breeding females and their offspring (the two Scytodidae species, Anelosimus jabaquara and some populations of Anelosimus studiosus and Anelosimus cf. studiosus). Colony foundation (Table I) and dispersal (Table II) refer to the dispersal stage involved in establishment of a new colony (social species) or nest (subsocial species), whether by juveniles or by males or females A B

PLATE 1. Social Stegodyphus (Eresidae): (A) Large colony of Stegodyphus mimosarum, Spioenkop Nature Reserve, KwaZulu Natal Province, South Africa. The nest is at the tip of a branch of an Acacia tree located near water. Sticky, cribellate silk covers the surface of the nest and multiple openings are visible leading into the nest. (B) Colony of Stegodyphus dumicola (Eresidae) in a small Acacia shrub in dry savanna at Seeis, Namibia. Maximum colony size is smaller than S. mimosarum. The central nest has undergone fission to produce two smaller daughter nests (one on either side), which remain connected to the mother nest by extensive cribellate capture webs. Photos: T. Bilde.

THE EVOLUTION OF SOCIALITY IN SPIDERS

95

B

A

C

B

D

PLATE 2. Foraging groups of social spiders: (A) Females of Anelosimus eximius (Theridiidae) feeding together on a small insect, in tropical rainforest at Cuyabeno Nature Reserve, Ecuador. Photo: L. Avile´s (published in Avile´s et al., 2001; reprinted with the author’s permission). (B) Group of subadult females of Stegodyphus dumicola (Eresidae) feeding together on insect prey at Seeis, Namibia. The spiders are in the cribellate web below the communal nest. Photo: Y. Botner and M. Salomon. (C) Group of Aebutina binotata (Dictynidae) feeding together on prey at Cuyabeno Nature Reserve, Ecuador. The spiders are brightly colored yellow with dark markings and rest on a layer of silk beneath leaves of tropical forest trees. Photo: L. Avile´s (published in color in Avile´s, 1997; reprinted with the author’s permission). (D) Two female Tapinillus sp. 1 (Oxyopidae) sitting on a leaf in ambush position, Cuyabeno Nature Reserve, Ecuador. Tapinillus sp. 1 builds a loose, irregular mesh web at the tips of leaves. Photo: L. Avile´s (published in Avile´s, 1994 and Avile´s et al., 2001; reprinted with the author’s permission).

(pre‐ or postmating); the mode of colony establishment, by fission involving short‐range displacement, or propagule dispersal over larger distances; and if dispersal and colony foundation are performed by single individuals or in groups. Dispersal and colony foundation by postmating adult females, singly or in groups, is the rule in social species, whereas juveniles or premating males and females disperse singly in subsocial species.

96

YAEL LUBIN AND TRINE BILDE

B

A

C

D

PLATE 3. Subsocial spiders: (A) Large colony of Anelosimus studiosus (Theridiidae) in Knox County, eastern Tennessee, USA. The colony measured 28,280cm3 and was composed of clusters of multiple females with their eggsacs and brood and interconnected silk capture webs. Spiders from different clusters moved through the connected webs. Photo: S. E. Riechert. (B) Female huntsman spider from Australia, Delena cancerides (Sparassidae), guarding an eggsac attached to the substrate. Below the female, a group of young feed together on a fly. Young from at least two cohorts are visible above the female. Photo: L. Rayor. (C) Nest and capture web of the subsocial spider Stegodyphus lineatus (Eresidae) in the Negev Desert, Israel. The nest is conical and is covered with plant debris and prey remains; the cribellate silk web captures mainly flying insects. Photo: T. Bilde. (D) Female Stegodyphus lineatus (Eresidae) on her nest with her offspring in the Negev Desert, Israel. The mother feeds her young by regurgitation for 2–4 weeks, after which the young kill and consume her (matriphagy). Photo: T. Bilde and M. Salomon.

The sex ratio bias indicates whether nestmates inbreed regularly. Dispersal and the sex ratio define the breeding structure of populations of each of the species. Juvenile and primary sex ratios are highly female biased in the social species with the exception of Tapinillus sp. 1, which is apparently outbreeding. Group activities are cooperative if performed by multiple individuals in the colony. Colony nest and web construction include laying of silk lines and removal of debris by several individuals simultaneously. Communal

THE EVOLUTION OF SOCIALITY IN SPIDERS

97

defense behaviors against arthropod and vertebrate predators have been described in a few social species; these involve silk production and biting. In all of the social species, several spiders attack large prey jointly and feed together on it, and juveniles of subsocial species behave similarly during the stage they are together in the maternal nest. Maternal care in subsocial species may involve both feeding and defense of the brood. Cooperative brood care in the social species includes behaviors such as defense of eggsacs or young and feeding of young performed by multiple females in the nest. The helpers may be breeding females (allomaternal care) or individuals that do not produce young themselves. Subsocial species with multiple breeding females have some degree of cooperation, either in feeding young or in protection of the nest, and this differs among the species. Aggression toward nestmates as well as foreign females has been observed in multiple female colonies of subsocial species, for example, during eggsac guarding. In social species, direct aggression is generally lacking. Under mating behavior, we recorded whether females mate with nestmates or with foreign males, if multiple mating occurs, and the existence of competition among males for access to females. These are traits that determine the potential for outbreeding by means of sexual competition or mate choice, but there are very few observations of mating behavior in any of the social or subsocial species. Some form of male–male competition is recorded in five to seven of the social species, and fighting among males in four species. Finally, reproductive skew, that is, the proportion of reproducing females in a colony, reflects the balance of cooperative and competitive interactions in the colony. Three of the five species for which data are available have 1.25 to 4 females per eggsac. We suggest that this reflects skewed reproduction as females in the social species are thought to produce a single clutch in their lifetime. The exceptions to the general patterns of social and subsocial traits are illuminating. We followed Agnarsson et al. (2006) in assigning Tapinillus sp. 1 to the social species, as multiple cohorts of females and young occur together in a colony and the colonies last for more than a single generation, traits that are typical of the social species. However, dispersal is by groups of subadult or adult males and females, and the lack of a primary sex ratio bias toward females was taken as evidence of outbreeding (Avile´s, 1994). Avile´s (1997) assigned D. cancerides to the social species, but we have moved it to the subsocial category, as groups with multiple breeding females seem to be less common than single females with their brood (L. Rayor, personal communication). Nest persistence alone, however, is not a sufficient descriptor of sociality, as the social crab spider Diaea socialis (and perhaps Diaea megagyna) may also have colonies that last for a single

98

YAEL LUBIN AND TRINE BILDE

generation, yet they have an inbred mating system as evidenced by the presence of female‐biased sex ratios (Avile´s, 1997; Evans, 1995; Evans and Main, 1993; Main, 1988). Among the subsocial species (Table II), a few species (notably, A. studiosus in North America, A. cf. studiosus in Uruguay, and D. cancerides in Australia) are known to have multiple breeding females in a colony, but apparently lack regular inbreeding (A. studiosus: T. C. Jones and S. E. Riechert, personal communication; but, see Furey, 1998; A. cf. studiosus: Ghione et al., 2004; D. cancerides: Rowell and Avile´s, 1995; L. Rayor, personal communication). A. studiosus from North America and A. cf. studiosus from Uruguay are presently regarded as a single species (I. Agnarsson, personal communication), and if this is the case, the species has an extraordinary latitudinal range across the Americas. Populations studied in Uruguay by Ghione et al. (2004) and Viera et al. (2007b) as well as those studied in North America (Brach, 1977; Furey, 1998; Jones and Parker, 2000, 2002; Jones et al., 2007) have a variable social structure, from single‐female nests to multiple breeders. Most North American populations of A. studiosus have predominantly single‐ female nests (Furey, 1998; T. C. Jones and S. E. Riechert, personal communication), and there is a latitudinal gradient in the frequency of multiple female colonies, with a higher frequency occurring in lower latitudes (see Section V.C). Furthermore, in some populations, sex ratios are moderately female biased, which suggests the possibility of inbreeding (Table II). There are other species that may be transitional, but data are lacking to be able to characterize their social structure adequately. A. jabaquara from Brazil has a female‐biased sex ratio comparable to the social Anelosimus but exhibits low tolerance of female nestmates during the breeding stage and dispersal at the subadult stage, indicating possible outbreeding (Table II; Gonzaga and Vasconcellos‐Neto, 2001; Marques et al., 1998). The Australian thomisid D. megagyna also has a female‐biased sex ratio (Evans, 1995; Table I), and Avile´s (1997) includes it among species at the transition from subsocial to social as the colonies do not persist for more than one generation. In Stegodyphus tentoriicola in the Karoo region of South Africa, most nests were of single females and brood, but a small percentage of nests contained multiple adult females and males or multiple females and their broods (not included in the tables; Y. L. and T. B., personal observations). Thus, this species too may be at the transition from subsocial to social. The distinguishing characteristic of subsocial species is a prolonged period of maternal care, with the potential for interactions among the young in the maternal nest. However, many solitary species exhibit some maternal care, and the distinction between subsocial and solitary species with maternal care is somewhat arbitrary, depending mainly on the length of the

THE EVOLUTION OF SOCIALITY IN SPIDERS

99

communal stage after the end of maternal care (see above). Additionally, we lack information on most species with maternal care. Aviles (1997) indicates 16 families showing some form of maternal care of offspring beyond the egg stage. The list of subsocial species in Table II is clearly incomplete, with only seven families represented. We have included mainly those species with social congeners, and for which some behavioral information is available. Agnarsson (2006) reviewed the New World Anelosimus and indicated additional subsocial species, and collections made in Madagascar revealed several likely subsocial Anelosimus (Agnarsson and Kuntner, 2005). Observations of some of these species suggested both extended maternal care and joint prey capture and feeding by juveniles in the maternal nest. Extended maternal care and delayed juvenile dispersal occur in many other comb‐footed spider (Theridiidae) genera (e.g., Theridion, Achaearanea; Agnarsson, 2004), in all Eresidae that have been studied to date (Dorceus, Eresus, Seothyra, Stegodyphus; D’Andrea, 1987; Henschel and Lubin, 1992; Kraus and Kraus, 1988; Kullmann, 1972; Y. L., personal observations), in the Amaurobioidea families Agelenidae (Amaurobius, Coelotes; Avile´s, 1997 and references therein) and Desidae (Phryganoporus candida; Table II), and in the crab spiders, Thomisidae (Diaea ergandros; Table II). These groups need further study in order to reveal the environmental and demographic conditions that favor family‐group living and the transition to inbred sociality. III. INBRED SOCIALITY IN SPIDERS How similar are the different social spider species, spread among seven different families, in their behavioral, life‐history, and demographic characteristics? The social insects and cooperative‐breeding vertebrates each demonstrates a wide diversity of social traits, with cooperation taking different forms, reproductive patterns ranging from multiple breeding females to single breeders with helpers, and various types of dispersal and group establishment patterns (Brockmann, 1997). The constellation of traits that characterize social systems provide a window onto the selective processes that promote and maintain sociality. What traits characterize the social spiders, how variable are they, and what are the underlying mechanisms that maintain them? These questions are discussed in the next sections. A. COOPERATION VERSUS COMPETITION: A BALANCING ACT Nest construction, web building, foraging, defense, and brood care are all behaviors that give the appearance of being ‘‘cooperative.’’ Cooperative activities can be defined as collective activities of individuals that interact

100

YAEL LUBIN AND TRINE BILDE

over a period of time, and where the cooperating individuals obtain fitness benefits from their activities (Whitehouse and Lubin, 2005; see also Downes, 1995). For example, colony members cooperate in web building and removal of debris (Anelosimus eximius: Christenson, 1984; Achaearanea wau: Lubin, 1995). Cooperation, however, does not necessarily imply coordination among interacting individuals. As demonstrated in a number of studies, apparent coordination of an activity such as web building or prey capture can be obtained by means of similar responses of individuals to stimuli, with further individual adjustment of behavior in response to local information (Sendova‐Franks and Franks, 1998). Information provided in silk, pheromones, and web vibrations may thus amplify the group response (Burgess, 1979; Saffre et al., 1999; Vakanas and Krafft, 2001). One form of apparent group coordination is the synchronization of movement of individuals on the nest or web, which has been observed in several of the social species. During nest building, late instar females and males of Tapinillus sp. 1 (Oxyopidae) participate in laying silk and their ‘‘bouts of activity and quiet appeared to be synchronized’’ (Avile´s, 1994). Stegodyphus dumicola individuals move in unison toward prey trapped in the sticky capture web, starting and stopping suddenly in a synchronized, rhythmical manner (personal observations) and similar behavior was observed in A. eximius (Krafft and Pasquet, 1991). A likely explanation of this behavior is that during group prey capture, synchronized movement enables spiders to distinguish prey vibrations from those of conspecifics and thus to locate the source of prey vibrations on the web (Krafft and Pasquet, 1991). Such movements might be defensive as well, preventing potential predators from focusing on a single individual (Bertram, 1978). Group foraging occurs in all of the web building social spiders, and in all social species individuals share prey. The Australian social crab spiders, D. socialis and D. megagyna, which do not construct webs, forage individually (Main, 1988), while D. cancerides, a group‐living huntsman spider, cooperates in prey capture (Rowell and Avile´s, 1995). By hunting in a group, spiders can capture larger prey and save on per capita costs of silk production (reviewed in Avile´s, 1997). They may also benefit from sharing digestive enzymes, though here, as well as in prey capture, the question arises whether cheating occurs, in which some individuals exploit the efforts of others. While there is no direct evidence for cheating, contests over access to prey have been noted in several social species (A. eximius: Christenson, 1984; Agelena consociata: Riechert, 1985; Stegodyphus mimosarum: Ward and Enders, 1985). Competition may be over a feeding site on the prey. Willey and Jackson (1993) found in Stegodyphus sarasinorum that the first spider to begin feeding could feed on the thorax, which is the preferred site on the prey, and thus obtain more food. In S. dumicola, large individuals had a competitive advantage over smaller ones: they excluded smaller

THE EVOLUTION OF SOCIALITY IN SPIDERS

101

individuals from prey and ingested more food (Whitehouse and Lubin, 1999). Some spiders that fed in a group did not increase in mass and even lost mass, hinting at the additional possibility of enzyme exploitation by dominant individuals (Amir et al., 2000; Whitehouse and Lubin, 1999). Competition increased with feeding group size (Ward and Enders, 1985; Whitehouse and Lubin, 1999; but see Pasquet and Krafft, 1992), and the mean body mass of individuals in the colony and the proportion of breeding females were negatively correlated with colony size—likely outcomes of competition (Avile´s and Tufin˜o, 1998; Bilde et al., 2007; Seibt and Wickler, 1988b; Ward and Enders, 1985). Differential feeding success due to competition for prey is a likely mechanism underlying the variation among individuals in body size and reproduction in large colonies (Seibt and Wickler, 1988b; Ulbrich and Henschel, 1999; Ulbrich et al., 1996; Vollrath and Rohde‐Arndt, 1983; Ward and Enders, 1985). Competition may occur not just for energy, but possibly also for specific nutrients (M. Salomon, unpublished observation). Lipids are a source of energy for growth and maturation and are also presumed to be required for oogenesis (Anderson, 1978; Pulz, 1987). Field colonies of S. dumicola containing large juveniles were supplemented with lipid‐ or protein‐rich crickets. More females matured to adult, and the adults were heavier in lipid‐supplemented colonies. The proportion of adult females in protein‐ supplemented colonies actually decreased with increasing colony size, suggesting that lipids were a limiting resource in these colonies. These results were supported by a laboratory experiment in which individual spiders were given crickets that were kept on standard cricket diet: during the first 2 h of a feeding bout, they extracted mostly lipids from the prey (M. Salomon, personal communication). Cooperation in rearing young is well known in group‐living vertebrates and in social insects, where helpers feed and guard young of other females and often forgo reproduction themselves, either temporarily as in most cooperative breeding birds and mammals (Emlen, 1997) or permanently as in many eusocial insects (Wilson, 1971). In social spiders, all females in a colony are potential breeders. The occurrence of allomaternal brood care has been assumed by most students of social spider biology (reviewed in Avile´s, 1997; Whitehouse and Lubin, 2005) but is rarely tested critically. An alternative possibility is that each female cares for her own young only, in which case there is no direct cooperation. So far, allomaternal brood care has been shown unequivocally only in A. eximius (Christenson, 1984) and S. dumicola (Salomon and Lubin, 2007). For helping to be a viable strategy of cooperation, two conditions must be met, namely, that helpers increase the fitness of the females that they help and that helpers derive benefits from this behavior. Given the high relatedness among individuals within the colony, helpers are likely to gain inclusive fitness benefits from helping to raise young of other

102

YAEL LUBIN AND TRINE BILDE

females in their cohort. The costs of helping may be relatively low if helpers are females that are unable to breed due to external causes, for example lack of sufficient food to mature (Ulbrich and Henschel, 1999). Salomon and Lubin (2007) compared the growth of S. dumicola young from a single brood in experimental nests containing either a mother alone or a mother together with four or five other females (either subadult or adult). They found that young maintained with a group of females survived better and increased their body mass more than young that were raised by a single female. The young obtained food from helpers both by regurgitation and later by killing and feeding on the helpers (gerontophagy; Seibt and Wickler, 1987) as well as on their mother (matriphagy). Other forms of helping have been reported. Large juveniles of Anelosimus dubiosus fed younger individuals (Marques et al., 1998). Anelosimus rupununi females clustered their eggsacs and the clusters were tended by a single female. Avile´s and Salazar (1999) observed that on completing the construction of an eggsac, a female moved it to a cluster of sacs and assumed a position near the cluster. They observed several instances of a female moving away from this presumed guarding position and being replaced by another. Clusters of sacs were more likely to be guarded than were individual ones. This behavior has not been recorded in other social species, though eggsac guarding is reported in several (Table I). B. DO SOCIAL SPIDERS HAVE DIVISION OF LABOR? Division of labor involves the partitioning of colony tasks among different individuals in the colony and is characteristic of species with strongly skewed reproduction (Fjerdingstad and Crozier, 2006; Oster and Wilson, 1978). Activities may be divided by age (age‐based polyethism), phenotype (physical castes), or some combination of these (Wilson, 1971). While social spiders may accrue tasks as they mature, true task differentiation of either sort seems to be lacking. For example, in A. wau colonies, young remain in the nest for
11 days after emergence from the eggsac and feed on prey brought into the nest by females (Lubin, 1995). They begin to venture out and to participate in nest or web repair activities, first near the nest, and later further away. At 3–4 weeks after emergence, they begin to take part in prey capture. In A. eximius, adult females are more likely to execute repairs and remove debris from outer parts of the web than are juveniles (Christenson, 1984). Adult females of Aebutina binotata capture most of the prey, with larger juveniles joining, and the young sometimes engaging in web repair (Avile´s, 1993b). Overall, there appears to be little differentiation among adults in the tendency to hunt or repair the web (Avile´s, 1993b; Lubin, 1995). Adult males are not known to participate in colony activities.

THE EVOLUTION OF SOCIALITY IN SPIDERS

103

The frequency with which individuals engage in different activities may depend mainly on their hunger state and competitive abilities (Ainsworth et al., 2002; Ebert, 1998; Ward and Enders, 1985), but this requires further study. The South American social theridiid Theridion nigroannulatum was found to have two distinct size classes of adult females in its colonies (Avile´s et al., 2006); this begs the question if different individuals derive benefit from maturing at different instars. Several studies have addressed the question whether reproductive division of labor, or reproductive skew, occurs in social spiders (Lubin, 1995; Riechert and Roeloffs, 1993; Vollrath, 1986; Vollrath and Rohde‐Arndt, 1983). In a classic experimental study, Rypstra (1993) showed that in A. eximius colonies, when juveniles were fed large prey, some females obtained more food than others, grew larger, and produced more eggsacs, while others did not obtain enough to mature. When groups were fed only with small prey, most juveniles matured, but few had enough resources to reproduce. Competition of this sort (see Section III.A) can create an unequal distribution of resources among colony members (modeled by Ulbrich et al., 1996) and result in differential reproduction within the colony (Ulbrich and Henschel, 1999). M. Salomon (unpublished observation) found that in colonies of S. dumicola that were collected at the beginning of the egg‐laying period, but after males had largely disappeared from the colonies, on average 40% of the females were adult and the remainder were subadult. Most of these subadults would mature (if at all) too late to acquire mates and reproduce. The range of adult females, however, was very large (10–60%) and was unrelated to overall colony size (number of females). Perhaps there is a threshold group size above which competition results in reproductive skew, while the degree of skew may be density dependent or ‘‘fine‐tuned’’ by the amount or type of prey available and other environmental factors. Nonbreeding females and subadults may enhance colony productivity through brood care as well as by participation in web repairs and prey capture, while gaining fitness benefits through closely related breeders. Yet, nonbreeders also increase the competition for resources in the colony. Thus, reproductive cooperation increases colony productivity on the one hand, while on the other hand, competition for food will decrease it. These opposite effects can result in unstable colony dynamics (Avile´s, 1999), which is discussed further in later sections. C. COLONY FOUNDATION: PROPAGULE DISPERSAL VERSUS FISSION Dispersal and establishment of a new colony are critical stages in the life history of social species. In most group‐living species, juveniles or young adults of one or both sexes undergo premating dispersal to establish a new

104

YAEL LUBIN AND TRINE BILDE

group or to join another group (Emlen, 1991). The nature of dispersal events and the mode of group founding have far‐reaching consequences for population genetic structure and dynamics. Social spiders exhibit two substantially different types of dispersal, which we term ‘‘propagule dispersal’’ and ‘‘fission.’’ These correspond to potentially long‐range dispersal or emigration and short‐range colony budding or translocation, respectively (Vollrath, 1982). A colony founded by propagule dispersal begins its existence as a single adult female or a small group of females (A. wau: Lubin and Robinson, 1982; A. eximius: Vollrath, 1982; D. socialis: Main, 1988; S. dumicola and S. mimosarum: Seibt and Wickler, 1988a; Lubin and Crouch, 2003; T. nigroannulatum: Avile´s et al., 2006). These females are mostly previously mated in the mother colony. Sometimes additional individuals join an incipient colony, and these may be subadult or adult females and occasionally males (A. wau: Lubin and Robinson, 1982; A. jabaquara: Marques et al., 1998; S. dumicola, S. mimosarum: personal observation). Several modes of propagule dispersal have been observed in social species, including single or small numbers of females bridging on silk threads or mass movement of large numbers of females across the substrate and on silk‐bridging threads (A. eximius: Vollrath, 1982; Avile´s, 1997 and references therein; S. mimosarum: Crouch et al., 1998) or along silk ‘‘highways’’ (A. wau; Lubin and Robinson, 1982). Ballooning of adult females during midday thermal updrafts was observed in several species of Stegodyphus: Jacson and Joseph (1973) noted aerial dispersal, but did not distinguish it from bridging behavior, in adult females and fourth to eighth instar young of S. sarasinorum; Wickler and Seibt (1986) observed ballooning of a single adult female S. mimosarum; and Schneider et al. (2001) recorded the ballooning of multiple females of S. dumicola in Namibia. This behavior is surprising, as the body mass of a social Stegodyphus female (80–150 mg) far exceeds the estimated maximum spider mass that can be lifted on a silk line (Henschel et al., 1995b; Suter, 1991). We speculated that these cribellate spiders release multiple silk lines from the cribellum and thus acquire greater lift; indeed, ballooning S. dumicola females appeared to sail up on a broad veil of silk lines (Schneider et al., 2001). Propagule dispersal occurs at a distinct stage in colony development, when postmating females are present in the nest and before egg laying commences (A. wau: Lubin and Robinson, 1982; A. eximius: Vollrath, 1982; S. mimosarum: Seibt and Wickler, 1988a; S. dumicola: Schneider et al., 2001). In most species, the generations within a colony are discrete and development is rather synchronous. Thus, a large colony may contain numerous postmating females that will undergo dispersal over a short time period of a few days, as seen in swarming in A. wau, A. eximius,

THE EVOLUTION OF SOCIALITY IN SPIDERS

105

S. mimosarum, and S. dumicola. Indeed, only large colonies give rise to dispersers (reviewed in Avile´s, 1997; Schneider et al., 2001). Mass dispersal has been seen in several species, both from a single colony and synchronously from many colonies in a population (Table I; see Section III.F). Consequently, a large number of small nests containing a single female or a small number of females appear suddenly in the population. These dispersing propagules have many predators and the probability of survival is low (Avile´s and Tufino, 1998; Avile´s et al., 2006; Bilde et al., 2007; Henschel, 1993, 1998; Lubin, 1991; Vollrath, 1992). Colony fission (we include colony splitting or fission, budding and translocation under the general category of ‘‘fission’’; Table I) also leads to the establishment of new nests. This may be accidental, for example when falling branches or winds damage the nest or when large mammals move through the forest understory (Riechert et al., 1986). In A. binotata, the only recorded method of dispersal is by colony fission, which takes place when the colony grows and individuals expand the web onto nearby vegetation (Avile´s, 2000). The transient nature of Aebutina colonies may be related to the ephemeral quality of their two‐dimensional nests, consisting of a thin sheet of silk laid on the surfaces of leaves (Avile´s, 2000). Fission can occur at any stage of development, but is most prevalent when the colony expands in space during later growth stages of the young and when competition for food is likely to be greatest (Bilde et al., 2007). In A. wau, nests split at the stage of large juveniles and subadults, as the web is expanded to additional supports and many new leaf nests are added to the structure. Connections between parts of the nest may then be severed by wind or branch‐falls or, when spiders mature and begin to reproduce they cease to repair the web, movement between the subdivisions is reduced and the connections are discontinued (unpublished observations). In laboratory conditions, Bodasing et al. (2001) found that fission in S. mimosarum was most likely to occur in the spring when larger juveniles were present in the nest, but that the feeding regime did not influence dispersal tendency. Both dispersal and colony fission can lead to clustering of incipient nests, but only propagule dispersal produces long‐range migrants as well (but, see exceptions below). Daughter colonies established by fission have a greater chance of surviving owing to their larger size, but the greater number of individuals in such a daughter colony will also increase the amount of within‐colony competition for food. Furthermore, fission colonies contain individuals of an assortment of ages, similar to the parent colony, whereas a dispersal propagule consists of one or a few mated females ready to oviposit. Thus, for a propagule colony, the clock is reset, so to speak, and the subsequent generation will be synchronized in its development, while a fission colony is subject to the same developmental asynchrony as the

106

YAEL LUBIN AND TRINE BILDE

original parent colony. Synchronous egg laying in propagule colonies will produce young of similar size and age, thus lessening the variance in growth rate due to asymmetric contests over prey (Ulbrich and Henschel, 1999), and perhaps increasing the benefit of cooperating in colony activities (see Avile´s, 1999; Section III.E). The consequence should be that most offspring in the first generation of a propagule colony will mature and reproduce and colony growth will be maximized. By comparison, fission colonies will be subject to competition between spiders of differing sizes, colony growth rate will be retarded, and there should be a greater chance of colony failure. Data are needed to test these predictions. Propagule foundation may have other beneficial effects on colony growth and survival, such as escape from parasites or disease by dispersers (Boulinier et al., 2001; Hamilton, 1987) or access to a better habitat (Danchin et al., 2001). Avile´s et al. (2006) suggested that by undergoing ‘‘explosive’’ dispersal and propagule formation, the social T. nigroannulatum rids itself of predatory inquilines [Faiditus spp. (Argyrodinae)]. These inquilines were abundant in medium‐sized and large colonies, but absent from small ones. Several of these ‘‘guests’’ are also predators of the host spiders, as were the unidentified Argyrodinae in colonies of A. wau (Lubin and Robinson, 1982), Argyrodes projiciens in A. eximius nests in Panama (Vollrath, 1982), and members of several spider families found in nests of A. studiosus (Perkins et al., 2007). A. wau colonies suffered increasing kleptoparasite (Argyrodinae) loads with increasing colony population size and nest dimensions (Lubin and Robinson, 1982), while kleptoparasites were rarely found in propagule nests (Y. L., unpublished observations). Most of the social spiders studied to date exhibit both modes of dispersal at different life stages. The apparent exceptions are A. binotata and A. consociata, both of which form daughter colonies only by fission (Avile´s, 2000; Riechert et al., 1986). Riechert et al. (1986) suggest that segments of A. consociata colonies in equatorial African forests are carried accidentally on large mammals that disrupt the nests and disperse the spiders as they move. Aebutina undergoes repeated translocations of the entire colony [Aviles (2000) refers to a nomadic phase] at the stage of newly matured males and females (thus at the same stage that propagule dispersal should occur) in movements that can cover several hundred meters over a period of several weeks. These long‐distance translocations may compensate for the lack of propagule dispersal. In both of the above species, colony translocation is unusual in that it can result in long‐range dispersal, and thus differs from the typical budding or splitting of a colony, which produces a cluster of daughter colonies separated by only short distances of a few meters. Thus, these exceptions may prove the rule that a long‐distance

THE EVOLUTION OF SOCIALITY IN SPIDERS

107

dispersal stage is an essential component of social spider life history. We return to this theme in discussing the transition from subsocial to social group living. D. FEMALE‐BIASED COLONY SEX RATIOS: PRIMARY AND OPERATIONAL SEX RATIOS The social spiders are characterized by female‐biased sex ratios with one to two males produced for seven to eight females (Table I; reviewed in Avile´s, 1997). Such a marked departure from the Fisherian sex ratio of equal numbers of males and females in diplo‐diploid species raises questions concerning both the mechanism of the sex ratio bias and the nature of the selective forces acting on the trait (see Section V.B). Chromosomal studies of developing embryos in several species show that the female‐ biased sex ratio is caused by an overproduction of female embryos, showing a strong bias in the primary sex ratio. In S. dumicola, females appear to have control over the mean sex ratio—17% males—but no direct control over the sex of individual offspring (Avile´s et al., 1999). In contrast, the Neotropical spider Anelosimus domingo shows a precise sex ratio of at least one male per eggsac (Avile´s et al., 2000) indicating more control over the sex of individual offspring. A female‐biased primary sex ratio with a proportion of males of 0.08 was reported in A. eximius (Avile´s and Maddison, 1991) and in D. socialis with a male proportion of 0.28 (Rowell and Main, 1992). Under a heterogametic sex determination mechanism as in spiders (X1X20 with males as the heterogametic sex), an equal number of sons and daughters is expected (White, 1973). The primary sex ratio bias therefore suggests a mechanism that allows some degree of control over the two types of sperm (X1X2 and 0) in order to attain relative precision in the proportion of sons and daughters produced in each clutch (Avile´s et al., 2000; Hurst and Vollrath, 1992). The specific mechanism responsible for the sex ratio bias is not known. Field data on the sex ratio among subadult and adult spiders in a colony reveal that the female bias is maintained throughout the colony life cycle. In D. socialis, Rowell and Main (1992) found a proportion of 0.19 males in adult and subadult colonies suggesting an increased bias compared with the primary sex ratio (0.28; Table I). In contrast to this pattern, six out of seven populations of A. wau examined in Papua New Guinea had less female‐biased sex ratios in colonies with adults than in colonies of juveniles (Lubin, 1991). An explanation for either pattern could be differential mortality among the sexes or sex‐biased emigration. The dispersal of females after mating to establish new nests may explain the increase in the proportion of males.

108

YAEL LUBIN AND TRINE BILDE

Henschel et al. (1995a) found strong female bias in the secondary sex ratio—12% adult males—in colonies of S. dumicola. But, when correcting for the number of sexually mature (reproducing) females in the colony, the operational sex ratio was almost even, with 42% males. This is because competition for resources creates asymmetric growth and many females do not mature or mature only after the males have died (Henschel et al., 1995a; Ulbrich and Henschel, 1999). In the same species, colony adult sex ratios varying from extreme female biased to extreme male biased (T. B. and Y. L., personal observations) could indicate migration of one or both sexes. In field trials, we found that males left their natal nests to join nearby experimental colonies with unmated females in them (T. B. and Y. L., personal observations). Mass dispersal of females from large colonies (S. mimosarum: Crouch et al., 1998; S. dumicola: Schneider et al., 2001) may leave the nests with excess males. Hence, it is very difficult to make general conclusions about the causes of observed adult sex ratios in the social spiders. E. MATING SYSTEM: INBREEDING AND ITS POPULATION‐GENETIC CONSEQUENCES Most cooperative group‐living species go to great lengths to outbreed, with at least one sex undergoing premating dispersal and even strong behavioral taboos against mating with kin in some instances (Pusey and Wolf, 1996). By contrast, in social spiders, regular mating among nestmates of the same cohort, generation after generation, seems to be an almost general rule (Table I). In new colonies founded by a single female that mated with a male in the parental colony, mating among nestmates will be among full sibs (or partial sibs if the female mated with more than one male) already in the first generation within the colony. It would be interesting to know if females have a preference for mating with kin over nonkin; preliminary experiments in S. dumicola colonies with foreign and nestmate males suggest a lack of any preference (A. Maklakov and T. B., personal communication). Intense male–male competition over females has been observed in a few species (Table I). In the social theridiid, A. wau, males fight over access to molting females and engage in prolonged displays on the web, while females appear to be choosy; we do not know if this competition involves foreign as well as local males (Lubin, 1986). The unusual trait of intracolony mating and resulting inbreeding yields a strongly subdivided population structure (Table III). The high level of intracolony mating results in large genetic differentiation among colonies and subdivision into colony lineages (high FST values, Table III; Riechert and Roeloffs, 1993; Smith and Engel, 1994; Smith and Hagen, 1996).

TABLE III GENETIC DATA ON NESTMATE RELATEDNESS AND F‐STATISTICS OF POPULATION STRUCTURE WITHIN AND BETWEEN COLONIES OF SOCIAL SPIDERS, AND A COMPARISON WITH SOME SOLITARY AND SUBSOCIAL SPECIESa

Nestmate relatedness R

Structure within colonies relative to subpopulation FIS, FIL

Structure within colonies relative to total population FIT

Structure among colonies within total population FST

FIT ¼ 0.458

FST ¼ 0.517

109

Family

Species

Agelenidae

Agelena consociata social

0.52

FIS ¼ 0.13–0.16

Agelena aperta solitary Stegodyphus dumicola social

0.014

FIS ¼ 0.075

Eresidae

Theridiidae

Stegodyphus sarasinorum social Stegodyphus lineatus subsocial Achaearanea wau social

Structure among regions within total population FCT, FRT

Marker Allozymes

Allozymes

FST ¼ 0.92–0.99 CT ¼ 0.495b

0.25

Structure among colonies within regions FLR, FSC, FSR

FSC ¼ 0.88–0.98 CR ¼ 0.44b

FCT ¼ 0.35–0.81 RT ¼ 0.098b

mtDNA AFLPb

FIS ¼ 0.054

FIT ¼ 0.838

FST ¼ 0.818

Allozymes

FIS ¼ 0.07

FIT ¼ 0.15

FST ¼ 0.21 ’ST ¼ 0.31b

Allozymes AFLPb

FIS ¼ 0.088

FIT ¼ 0.786

FST ¼ 0.804

Allozymes

Source Roeloffs and Riechert, 1988; Riechert and Roeloffs, 1993 Riechert and Roeloffs, 1993 Johannesen et al., 2002; Deborah D. Smith, personal communicationb Smith and Engel, 1994

Johannesen and Lubin, 2001 Bilde et al., 2005b Smith and Engel, 1994

(Continued)

TABLE III (Continued)

Family

Thomisidae

110 a

Species Anelosimus eximius social Diaea ergandros subsocial

Nestmate relatedness R 0.92 (Suriname) 0.176 (Panama)d 0.44

Structure within colonies relative to total population FIT

Structure among colonies within total population FST

Structure among colonies within regions FLR, FSC, FSR

Structure among regions within total population FCT, FRT

FIS ¼ 0.075 FIS ¼ 0.083

FIT ¼ 0.886

FST ¼ 0.885

FSR ¼ 0.9

FIL ¼ 0.15

FIT ¼ 0.35

FLR ¼ 0.23

Structure within colonies relative to subpopulation FIS, FIL

Marker

Source

FRT ¼ 0.365

Allozymes

FRT ¼ 0.081

Allozymes

Smith and Engel, 1994; Smith and Hagen, 1996d Evans and Goodisman, 2002c

Hierarchical F‐statistics measures the departure from panmixia—individuals relative to total population (FIT), structure of colonies relative to total (FST), structure of colonies relative to region (FSR) or (FSC), structure of regions relative to total (FRT) or localities relative to total (FCT), and degree of nonrandom mating within colonies (FIS). b Deborah Smith (personal communication) uses an analogue to F‐statistics to show structure of colonies relative to total population (CT), structure of colonies relative to region (CR), and structure of regions relative to total (RT). c Evans and Goodisman (2002) use hierarchical F‐statistics to measure the decrease in heterozygosity at different levels of the population due to true inbreeding within locales (FIL), structure of locales within regions (FLR), structure of regions within the total population (FRT), or both inbreeding and population structure (FIT). d Disturbed roadside population might explain unusual low value (Smith and Hagen, 1996).

THE EVOLUTION OF SOCIALITY IN SPIDERS

111

Within‐colony relatedness can be higher than that of full‐sibs (r ¼ 0.5), for example, relatedness estimates in four populations of A. eximius showed values from 0.83 to 0.98 (Table III). This level of inbreeding among siblings and cousins would usually result in severe inbreeding depression and fitness loss, although regular inbreeding might purge deleterious alleles (Falconer, 1989; Roff, 1997). Colonies are very often genetically monomorphic, containing females of identical haplotype, suggesting colony foundation either by a single female or by genetically highly related individuals. The latter could occur if multiple females originate from a monomorphic colony (Johannesen et al., 2002) and establish a new nest by colony fission or by propagule dispersal (see Section III.C). A small proportion (13%) of S. dumicola colonies contained two to three haplotypes, which is the expected pattern if a ballooning female of a different haplotype joins an existing colony (Johannesen et al., 2002). Similar long‐range dispersal and introgression of different genotypes was also suggested as an explanation for colonies with multiple allozyme alleles in A. wau (Lubin and Crozier, 1985). Insufficient data are available for a range of social spider species to determine if they differ in the frequency of outbreeding, for example, depending on the predominant mode of dispersal. Another intriguing question is whether outbreeding depression occurs (as shown in bark beetles, Peer and Taborsky, 2005), though the ready acceptance of foreign spiders in social spider colonies (Table I) would argue against it. The lack of a sex ratio bias in Tapinillus sp. 1 suggests that this species is outbreeding (Avile´s, 1994; L. Avile´s, unpublished data). This is likely obtained by males moving between different colonies. Thus, Tapinillus may represent a transitional stage, where the benefits of male mating dispersal are great enough to overcome the costs of movement; this species requires further study. F. ‘‘BOOM AND BUST’’ COLONY DYNAMICS The lack of juvenile dispersal and prevalence of inbreeding and female‐ biased sex ratios discussed above have marked consequences for population and metapopulation dynamics of social spiders. After initial establishment, colony growth may be rapid due to the female‐biased sex ratio, increased reproductive potential due to cooperation, and the fact that many females reproduce (Avile´s, 1997, 1999; Bilde et al., 2007). As the colony grows, however, competition for food increases within the group and some females may not reproduce, slowing colony population growth. Dispersal by fission or by propagules could stabilize colony size at this point. Nevertheless, Aviles (1999) has shown by modeling that the initial rapid growth stages can cause

112

YAEL LUBIN AND TRINE BILDE

colonies to overshoot their carrying capacity, giving rise to a subsequent population crash. This intrinsic dynamical instability may explain the high colony extinction rate reported for some species (Avile´s, 1997, 1999). The exact causes of this instability need further clarification. In social spiders, juveniles do not disperse individually; therefore, colony fission is the only possible immediate response to competition generated by rapid expansion. The daughter colonies from a fission event should then persist (although with the disadvantages mentioned in the previous section). An alternative explanation for colony extinction is an increase in susceptibility to parasites and pathogens with increasing colony size (see below). Colony extinctions, often in conjunction with mass propagule dispersal (see above), have been described in a number of species. Three separate sets of data from South Africa show dramatic and synchronized extinctions of 80–100% of colonies over a period of 1–2 years in populations of S. mimosarum (Crouch and Lubin, 2001; Seibt and Wickler, 1988a). In one instance, the decline was accompanied by mass emigration of females and males (Crouch et al., 1998). Vollrath (1986) recorded a 90% loss of colonies of A. eximius from a population in Panama, which was also accompanied by mass dispersal, and 90% of colonies died out in a population of A. wau over a few months, while other nearby populations suffered little mortality (Lubin, 1991; Lubin and Crozier, 1985). Sixty‐three percent of colonies of A. consociata in a population in Gabon died out over a period of 3 years, possibly related to the inability of colonies to survive and reproduce during prolonged, unusually wet periods in these years (Riechert and Roeloffs, 1993). The direct causes of colony extinction are not known in most instances. Crouch and Lubin (2001) ruled out temperature, rainfall, treefalls, or other physical features of the environment as explanatory agents, but parasitoid wasps (Pompilidae, Pseudopompilus funereus) may have played a role in one S. mimosarum population, or a pathogen might spread rapidly through colonies with little genetic variation (Lubin, 1991; Roeloffs and Riechert, 1988; Vollrath, 1982). Alternatively, in a synchronized population, food may become limiting, leading to mass movement to new habitats. The latter explanation seems less likely, as both large and small colonies in a population went extinct (e.g., A. wau: Lubin and Crozier, 1985; S. mimosarum: Crouch and Lubin, 2001; Lubin and Crouch, 2003). The same factors that promote mass dispersal may be responsible also for colony extinction. Not all species show the above pattern of ‘‘boom and bust’’ at the population level. Nevertheless, several species for which there are adequate data show rapid colony turnover (Avile´s, 1997 and references therein, 2000; Avile´s et al., 2006). This is in marked contrast to colonies of some colonial (‘‘territorial permanent‐social,’’ sensu Avile´s, 1997) spiders. For example, large colonies of the araneid Cyrtophora moluccensis in Papua New Guinea are known

THE EVOLUTION OF SOCIALITY IN SPIDERS

113

to have occupied the same sites continuously for at least 12 years (or
18 generations; Lubin, 1980), whereas colonies of most social spiders have a life span of only 2–7 generations (Avile´s, 1997; Lubin and Crouch, 2003). The distribution of social spider colonies in the environment reflects the processes of colony establishment, growth, dispersal, and extinction. Some social species are rare, and colonies can be found only sporadically and not clustered, for example Achaearanea vervoorti in New Guinea cloud forest (Levi et al., 1982) and A. domingo in the understory of tropical rainforests of northern South America (Avile´s, 1997). The dispersal modes of these species are not known. In many species, colonies occur in clusters, with clusters of colonies separated by areas of apparently similar, but unpopulated habitat (A. wau: Lubin and Crozier, 1985; A. consociata: Riechert, 1985; A. eximius: Vollrath, 1982; Pasquet and Krafft, 1989; all three social Stegodyphus species, Lubin and Crouch, 2003; Seibt and Wickler, 1988a). Sometimes, colonies occupy nearly continuous stretches of vegetation along forest edge or roadsides, on fences or bridges, and along streams or rivers; in such instances, local prey abundance at ecotones or near water is a likely cause of this distribution. Colony clusters often have interconnected nests indicating that they are formed by fission; further evidence is the high relatedness among individuals within a colony cluster (Roeloffs and Riechert, 1988, for A. consociata). In a sparse population of S. dumicola at the edge of the hyperarid Namib Desert, we found a higher proportion of scattered propagule‐disperser colonies than in a dense population in the higher‐productivity savanna, where most small colonies were derived by fission and clustered around larger ‘‘mother’’ colonies (Bilde et al., 2007). These observations suggest that fission occurs mainly when conditions are favorable, perhaps as a means of reducing within‐colony competition while still remaining in a high‐productivity habitat. These clusters, however, become susceptible to increasing parasite and disease loads, which can lead to mass extinction. As suggested by Hamilton (1987), such inbred colonies would have two means of escape—long‐distance dispersal and outbreeding. The interplay between food resources, dispersal modes, and population genetic consequences is a fruitful avenue of future research.

IV. PHYLOGENETIC RELATIONSHIPS AMONG SOCIAL SPIDER SPECIES Three striking features emerge from the distribution of social species across spider families (Table I): (1) a remarkably high number of independent origins of sociality that are distributed among unrelated taxonomic clades, (2) a concentration of origins of sociality within clades where maternal care is common (Avile´s, 1997), (3) an apparent lack of diversification of the social clades (see below). These patterns prompt the questions: whether

114

YAEL LUBIN AND TRINE BILDE

there are certain conditions broadly applying to spiders of widely different taxonomic clades under which cooperation evolves, and whether there are certain common consequences of sociality for these spiders? A. COMMON FEATURES OF SOCIAL EVOLUTION The subsocial pathway to sociality proposed by Kullmann (1972) suggests that sociality should arise in clades with extended maternal care through the retention of young in the maternal nest. Based on the phylogeny by Coddington and Levi (1991), Avile´s (1997) showed that indeed the majority of social events, with the exception of Mallos (Bond and Opell, 1997; Jackson, 1978, 1979), occurred in clades where subsocial behavior (maternal care) has been reported. Phylogenies of two genera, Anelosimus and Stegodyphus, confirm both the origin of sociality within clades with maternal care, supporting the hypothesized subsocial route to sociality, and that sociality is a derived condition, suggesting multiple transitions from subsocial to social behavior (Agnarsson, 2006; Agnarsson et al., 2006; Johannesen et al., 2007). We lack knowledge of the degree of maternal care in the closest relatives or sister species of social species in the remaining genera, and we lack phylogenies for these groups; therefore, a quantitative phylogenetic test of the hypothesis that permanent‐sociality evolves through a transition from subsocial to social behavior is not yet feasible (Agnarsson et al., 2006). In addition to a transition from subsocial to social group living, permanent‐sociality in spiders is usually accompanied by a transition from outbreeding to inbreeding mating systems (Avile´s, 1997; Riechert and Roeloffs, 1993). Since each social clade contains only 1–2 species, there may have been 18 such transitions (Agnarsson et al., 2006). The causal relationships resulting in the costly transition to extreme inbreeding remain largely unknown and pose an important challenge to understanding social evolution in spiders (see Section VI). It is currently believed that factors promoting the loss of mating dispersal underlie the transition to permanent‐ sociality (Avile´s and Bukowski, 2006; Agnarsson et al., 2006; Bilde et al., 2005), implying that inbreeding is a consequence of group living (with the exception of Tapinillus) (Avile´s, 1997). A history of inbreeding in subsocial ancestors may have facilitated the transition to inbreeding mating systems through purging of the most deleterious alleles (Bilde et al., 2005). B. CASE STUDIES 1. Stegodyphus (Eresidae) Within the family Eresidae, maternal care is widespread in all known genera and appears ancestral to three derived cases of permanent‐sociality (Johannesen et al., 1998, 2007; Kraus and Kraus, 1988; Kullmann, 1972).

THE EVOLUTION OF SOCIALITY IN SPIDERS

115

The three social species of Stegodyphus are found in three distinct species groups, possibly of paraphyletic origin, supporting three independent derivations of sociality (Fig. 1; Johannesen et al., 2007; Kraus and Kraus, 1988). Interestingly, while the social species are found in tip clades, analyses of sequence divergence suggest that the social lineages are neither young nor transient. On the contrary, the large intraspecific divergence of these social lineages suggests that they are old and remarkably stable over evolutionary time. An alternative possibility is that they have undergone cryptic speciation and the three apparently old species are in fact a complex of cryptic species (Johannesen et al., 2007). It is remarkable that despite their old age, the social Stegodyphus have not diversified and evolved cladogenetically. Indeed, Avile´s (1997) suggested that social spiders are ‘‘caught in sociality’’ and that they are constrained evolutionarily by genetic or ecological factors perhaps due to the high level of inbreeding, and the combination of these factors hinders evolvability. Interestingly, low diversification applies also to clades of subsocial Stegodyphus: Johannesen et al. (2007) found no difference in the rate of lineage diversification between social and closely related subsocial lineages. However, we lack as yet a complete phylogeny for the genus, and hence, it remains unresolved whether lack of diversification is characteristic of all Stegodyphus or of social spiders in general (see discussion below). In addition, a lack of a priori expectations for speciation rates makes it difficult to assess whether diversification is constrained in the genus Stegodyphus or has to do with specific characteristics of subsocial and social clades. 2. Anelosimus (Theridiidae) Agnarsson et al. (2006, 2007) have produced a detailed phylogeny of the social Anelosimus, which led to the following conclusions: (1) there are five or six independent origins of sociality in the genus, depending on the placement of A. eximius in the phylogeny; (2) several pairs of social–subsocial sister species can be recognized (Fig. 2); and (3) there seem to be no instances of diversification within the social clades, a pattern similar to the social Stegodyphus. Thus, analyses of Anelosimus and other social theridiids support the hypothesis that maternal care is a predisposition for the transition to permanent‐sociality (Agnarsson, 2006; Agnarsson et al., 2007). Agnarsson (2006) divided the species of Anelosimus into three social categories: quasi‐social (permanent‐social), subsocial, and quasi‐solitary. In the latter category, he included species that have maternal care, but the young leave the maternal nest at an earlier stage than in the subsocial species. Thus, for example, in Anelosimus arizona subadults or even adults have premating dispersal to establish individual nests (Table II; Bukowski and Avile´s, 2002), whereas in Anelosimus pacificus and a number Madagascan species young disperse shortly after the end of maternal care,

116

YAEL LUBIN AND TRINE BILDE

99

100 100 100 100

E. cinnaberinus Rhine

E. walckenaeri Crete

100 99 100 99

100

100 100 100 100

S. lineatus Negev D S. lineatus Judea

100 99 99 99

S. mimosarum SA

S. mimosarum Mad. M S. africanus

100 100 100

100 78 100 72

S. bicolor 100 95 100 100

S. dufuori

100 100 100 100

86 100 96 100 96

S. sarasinosum 1994

S. sarasinosum 2004

S. dumicola B2

S. dumicola A5 100 99 100 100

S

D

S. tentoriicola

G. spenceri

U. durandi FIG. 1. A partial molecular phylogeny of Stegodyphus (Eresidae) [modified from Johannesen et al. (2007), with permission of the publisher]. Social species are circled. Probability levels of Bayesian inference and bootstrap scores for maximum likelihood (ML), neighbor joining (NJ), and maximum parsimony (MP) are shown in this order starting from above.

THE EVOLUTION OF SOCIALITY IN SPIDERS

117

at a much younger instar (Agnarsson and Kuntner, 2005). Three such ‘‘quasi‐solitary’’ species form a single clade that is nested within a clade of subsocial and social species, which led Agnarsson et al. (2007) to suggest a reversal from subsocial to quasi‐solitary behavior. Within the genus Anelosimus, however, there is actually a continuum of levels of subsociality, from dispersal of young at the end of maternal care, through dispersal at later instars, to groups of sisters remaining together and sharing a nest, or joining to form a new nest as in some populations of A. studiosus in North America (Furey, 1998) and in A. cf. studiosus in Uruguay (Ghione et al., 2004). In all of these instances, outbreeding is maintained through male dispersal. As we discuss below (Section VI), there appears to be considerable flexibility in subsocial life histories both among and within Anelosimus species, and ecological factors are likely to play a dominant role in determining the length of the group‐living stage. Thus, the proposed reversal more likely reflects different ecological conditions favoring early or late dispersal. C. SOCIALITY IN SPIDERS: AN EVOLUTIONARY DEAD END? A striking similarity between the Stegodyphus and Anelosimus phylogenies is the lack of diversification of the social clades, despite the old age at least of the Stegodyphus social lineages (Agnarsson et al., 2006; Johannesen et al., 2007). This characteristic has led to the suggestion that permanent‐ sociality accompanied by regular inbreeding is an evolutionary dead end (Agnarsson et al., 2006; Avile´s, 1997; Whitehouse and Lubin, 2005). The argument is that short‐term benefits of inbreeding, such as the elimination of dispersal costs and an increased rate of reproduction, are selected against in the long term because of reduced speciation rates or an increased risk of extinction (Agnarsson et al., 2006). In the case of the social spiders, continued mating among group members leads to very high inbreeding coefficients, which may result in a long‐term loss of adaptive potential (Keller and Waller, 2002). Agnarsson et al. (2006) proposed that the social clades in the Theridiidae are phylogenetically isolated and subject to high rates of extinction. While the social Stegodyphus also lack diversification, these lineages are relatively old, suggesting that they were able to track environmental changes over evolutionary time or that their environment Bayesian probability scores and bootstrap scores less than 95% and 70%, respectively, are denoted with ‐. The three Stegodyphus species‐groups based on Kraus and Kraus (1990) are denoted with respect to the social species: D (dumicola), M (mimosarum) and S (sarasinorum). The groups of Kraus and Kraus (1990) were confirmed with the exception of S. lineatus, which belongs to a fourth phylogenetic group. The relationship between S. lineatus and Eresus sp. suggests paraphyly of the genus Stegodyphus. Two species used as out‐groups are Gandanameno spenceri (Eresidae) and Uroctea durandi (Urocteidae).

118

YAEL LUBIN AND TRINE BILDE

Kochiura aulica Kochiura rosea Chrysso cf. nigriceps Tidarren sisyphoides Nesticodes rufipes 1 59 95 1 Achaearanea disparata Achaearanea tesselata 4 2 95 5 Achaearanea tepidariorum 8 Achaearanea vervoorti 56 2 92 2 Achaearanea wau 2 4 Thendula emertoni Ameridion sp. 2 99 Ameridion cf. petrum 8 Thymoites unimaculatum 3 Helvibis cf. longicaudatus 3 95 2 Theridion varians Theridion pictum 2 5 Coleosoma floridanum 3 Theridion frondeum 1 88 Theridion longipedatum 5 2 3 Theridion nigroannulatum Anelosimus sp. australia Anelosimus lorenzo 100 51 4 Anelosimus rupununi 6 2 Anelosimus dude 98 Anelosimus biglebowski 0 1 Anelosimus kohi Anelosimus nelsoni 1 Anelosimus vittatus Anelosimus pulchellus 1 80 2 Anelosimus ethicus 1 4 80 Anelosimus nigrescens 2 Anelosimus misiones 51 1 Anelosimus rabus 1 61 Anelosimus inhandava 2 2 Anelosimus may 96 Anelosimus sallee 4 Anelosimus pacificus 58 Anelosimus analyticus 1 2 Anelosimus chickeringi 3 Anelosimus domingo 61 89 5 Anelosimus dubiosus 5 5 1 Anelosimus jabaquara 6 Anelosimus eximius Anelosimus baeza 65 2 7 Anelosimus puravida 85 Anelosimus arizona 2 2 96 Anelosimus octavius 62 2 67 Anelosimus jucundus 1 8 Anelosimus guacamayos 3 Anelosimus elegans Anelosimus tosum 1 1 Anelosimus pantanal Anelosimus fratemus 2 Anelosimus studiosus 1 9 Anelosimus ontoyacu 1 Anelosimus tungurahua

99 10

FIG. 2. Interrelationships of social and subsocial Theridiidae based on molecular phylogeny [modified from Agnarsson et al. (2006a), by permission from the publisher]. Numbered circles show counts of independent social origins; arrows indicate social species; numbers above branches are bootstrap support values; below branches, Bremer support values. All the species within the clade have either documented maternal care or their behavior is unknown (see Agnarsson, 2006, Appendix).

THE EVOLUTION OF SOCIALITY IN SPIDERS

119

has remained rather stable over evolutionary time (Johannesen et al., 2007). Furthermore, the sister lineages of the social Stegodyphus species have not diversified either; hence, in Stegodyphus, we cannot draw the conclusion that speciation is reduced in the social lineages in comparison with their closest subsocial relatives. To determine if permanent‐sociality is truly an evolutionary dead end, or the social lineages need more time to diversify, we require information about the phylogenetic relationships of other social and subsocial species, and we need to establish what rate of diversification should be expected in the social and subsocial sister lineages.

V. EVOLUTION AND MAINTENANCE OF SOCIALITY IN SPIDERS: RELEVANT MODELS The subsocial route for evolution of group living points to maternal care, constraints limiting dispersal, and the transition to an inbred mating system, as common features characterizing the transition from periodic‐social to permanent‐social living. Once sociality has evolved, there may be additional selective forces promoting further development of the society. The objective of this section is to consider relevant models of social evolution which may play a role in the evolution of cooperation and the development of complex traits or social behaviors, and to review empirical evidence that support predictions from theory. The models included in this section are not mutually exclusive. It seems particularly relevant to discuss multilevel selection frameworks here since group level benefits are increasingly ascribed crucial selective forces in social evolution (Avile´s, 1993a; Leigh, 1999; Okasha, 2004; Reeve and Keller, 1999; Wade, 1978). In discussing various models, we have included relevant empirical evidence from subsocial and transitional as well as social species. A. KIN SELECTION To date, kin selection has been the most successful and widely accepted explanation for the evolution of societies and cooperation (Frank, 1998; Griffin and West, 2002, 2003; Hamilton, 1964; Maynard Smith, 1964; West‐ Eberhard, 1975). Kin selection can be defined as ‘‘the evolutionary process which occurs when individuals interact with one another in a non‐random way with respect to kinship and these interactions affect fitness’’ (from Wade and Breden, 1981). Individuals can increase fitness by their own reproduction (direct fitness) or through indirect fitness, by the reproduction of relatives with whom they share genes identical by descent (narrow

120

YAEL LUBIN AND TRINE BILDE

definition) or with whom they share the gene(s) of interest (broad definition) (West et al., 2007). Together, direct and indirect reproduction constitute an individual’s inclusive fitness. Genes coding for cooperation among individuals, which increase the reproduction of their relatives and hence their inclusive fitness, can spread in the population and lead to the evolution of more or less obligate helping behavior. The more genes identical by descent that individuals share, the higher the inclusive fitness benefits and, hence, the incentive to forego own reproduction and engage in helping behavior. This is the model known as Hamilton’s rule (Hamilton, 1964). Kin selection is most likely to occur when high relatedness within the group confers inclusive fitness benefits to individuals that cooperate, for example in nest defense, foraging, or alloparental brood care. Therefore, cooperative behavior is likely to be found in animals that live in family groups (Bourke and Franks, 1995; Emlen, 1991; Frank, 1998; Griffin and West, 2003; Koenig and Dickinson, 2004). Family structure arises through parental care, which results in overlapping generations or delayed dispersal of young, which remain in sibling groups. Cooperation and permanent group living in spiders presumably originated in family groups consisting of a female and her offspring, where extended maternal care provided sufficient benefits for siblings to remain in the nest and postpone dispersal (Agnarsson et al., 2006; Avile´s, 1997; Buskirk, 1981; Johannesen et al., 2007). This transition in life history resulted also in mating among siblings within the nest and consequently overall high relatedness among group members, thereby increasing the opportunities for kin selection to operate (Michod, 1982, 1993; but see Section VI). A classical way to infer kin selection is to identify the level of within‐ group relatedness in relation to the degree of cooperation (Emlen, 1991; Griffin and West, 2002). Since most of the social spider species studied show extremely high levels of relatedness (Table III; Avile´s, 1997; Riechert and Roeloffs, 1993; Smith and Engel, 1994; Smith and Hagen, 1996), cooperation in foraging and brood care is expected to provide inclusive fitness benefits. The identification of genetic relatedness within cooperating groups per se, however, is not necessarily sufficient to infer kin selection since cooperation may incur costs of group living such as kin competition and it is not obvious that there will be a net advantage to cooperation (Griffin and West, 2002, 2003; Hamilton and May, 1977; Queller, 1992). In cooperatively breeding birds and mammals, there is good evidence that helping behavior increases with the degree of relatedness (Crespi and Choe, 1997; Emlen, 1991; Griffin and West, 2002, 2003; Koenig and Dickinson, 2004). Experimental tests of the effect of kinship on fitness in social spiders are largely lacking. Cooperation in brood care did not differ between small experimental groups of the social S. dumicola comprising either females

THE EVOLUTION OF SOCIALITY IN SPIDERS

121

from two distant populations or from a single population in Namibia, as measured by the number and size of young raised by these females (M. Salomon, personal communication). In the subsocial Stegodyphus lineatus, growth and mass gained was higher in cooperatively feeding groups of siblings compared with nonsiblings indicating a role of kin selection in the evolution of communal feeding (Schneider, 1996b; J. Schneider and T. B., personal communication). Further experimental manipulations of relatedness and cooperation need to be performed in order to assess the role of kin selection in the evolution and maintenance of cooperation. 1. Kin Recognition Recognition in the form of group closure as seen in the social Hymenoptera has not been shown; on the contrary, social spiders appear to accept unrelated and foreign individuals into the colony with no apparent discrimination (D’Andrea, 1987; Darchen and Delage‐Darchen, 1986; Kullmann, 1968; Pasquet et al., 1997; Seibt and Wickler, 1988a). In nature, mixed groups of the subsocial D. ergandros can be found, indicating that young disperse to join other nests (Evans, 1999). While population genetic data usually show very low genetic variation within social spider colonies, which suggests colony founding by one or a few related individuals, there is evidence for the sporadic occurrence of unrelated individuals within a colony, suggesting the inclusion of long‐distance dispersers (Johannesen et al., 2002; Lubin and Crozier, 1985; D. R. Smith, personal communication). However, based on the genetic data, it seems likely that this phenomenon is the exception rather than the rule, that is, there is very little migration between colonies. If social spiders rarely experience foreign spiders immigrating naturally into their colonies, selection for mechanisms for use in kin discrimination behaviors such as group closure may be low or absent. In subsocial spider species, there are studies showing discrimination of genetic kin. In experimental groups of D. ergandros, sibs survived better than nonsibs and under conditions of starvation, cannibalism was directed at nonsibs (Evans, 1999). Similarly, groups of young siblings of the subsocial S. lineatus were less cannibalistic than groups of unrelated juveniles under experimental conditions of low food availability (Bilde and Lubin, 2001). In the latter species, there is also evidence to suggest that sib groups have a higher growth rate than groups of unrelated young (Schneider, 1996b). These studies indicate that kin recognition can occur and that kin selected benefits may accrue at least in certain situations. In the field, however, it is not known if discrimination of nonkin exists in colonies of social species. In contrast, no behavioral discrimination of sibs or nonsibs as mates was found in controlled breeding experiments of two subsocial species (S. lineatus: Bilde et al., 2005; and A. arizona: Avile´s and Bukowski, 2006).

122

YAEL LUBIN AND TRINE BILDE

2. Inbreeding and Kin Selection In models of kin selection, alleles associated with cooperative behaviors increase in frequency because cooperative individuals increase the fitness of individuals with whom they are genetically related (Wade and Breden, 1981). Inbreeding increases the degree of genetic similarity among individuals; from Hamilton’s rule, it therefore follows that kin selected traits should be favored in inbred populations. Through computer simulations, Wade and Breden (1981) showed that inbreeding favored the evolution of cooperation by changing the balance between opposing forces of individual and group level selection. Inbreeding reduces the genetic variance within the group and increases between‐group variance, which results in weaker within‐group and stronger between‐group selection, as the intensity of selection depends on the amount of genetic variance on which selection can act (Falconer, 1989). The important point is that inbreeding creates strong population subdivision through isolated breeding units, which allows selection among the isolated lineages to overcome counteracting selection within them (Wade and Breden, 1981; Wilson and Colwell, 1981). It is becoming increasingly recognized that social evolution involves multilevel selection when group‐level differences in productivity select for traits such as female‐biased sex ratios that would not be maintained without strong population subdivision (Avile´s, 1993a; Leigh, 1999; Okasha, 2004; Reeve and Keller, 1999; Wade, 1978).

B. MULTILEVEL SELECTION (GROUP SELECTION) In addition to selection pressures acting on individuals within groups, natural selection can also operate at the level of the group or society in a multilevel selection framework (Alexander and Borgia, 1978; Leigh, 1999; Okasha, 2004; Reeve and Keller, 1999; Wade, 1978; Williams, 1966). Within‐group selection acting on individuals will usually select against cooperation when the cost paid varies with the degree cooperation among individuals. Within‐group selection follows selfish‐genes models, where selfish types are favored at the expense of cooperative individuals who pay the costs of helping. Group‐level selection favors cooperation, when cooperation among individuals within a group yields higher overall group productivity than other groups, for example by increased survival or reproduction, or better colonization ability. Group‐level selection requires differentiation among groups (Hamilton, 1967), but does not necessarily require high relatedness within the group (Rissing et al., 1989). The conditions that must be met for selection to act on processes at the level of the group are (1) high variance and population substructure, which can be achieved if

THE EVOLUTION OF SOCIALITY IN SPIDERS

123

groups are founded by one or a few individuals; (2) limited migration among groups; and (3) high turnover rate, where some groups reproduce while others go extinct (Avile´s, 1993a; Leigh, 1999; Okasha, 2004; Reeve and Keller, 1999; Wade, 1978). Under these conditions, selection at the level of the group may override individual selection within the group. The social spiders may provide the conditions necessary for group selection (see Section III; Avile´s, 1986, 1993a; Avile´s and Tufin˜o, 1998; Bilde et al., 2007; Smith and Hagen, 1996): colonies are founded by few females (Avile´s, 1997; Jacson and Joseph, 1973; Lubin and Robinson, 1982; Main, 1988; Seibt and Wickler, 1988b; Vollrath, 1982), there is limited migration between groups (Avile´s, 1997; Main, 1988; Roeloffs and Riechert, 1988; Seibt and Wickler, 1988b; Vollrath, 1982), and there are very high extinction rates, which means that only few groups proliferate (Avile´s and Tufin˜o, 1998; Bilde et al., 2007; Crouch and Lubin, 2001; Henschel, 1998; Vollrath, 1982). Propagule dispersal and initiation of new colonies occurs mainly by individuals from large colonies (Avile´s, 1997; Lubin and Robinson, 1982; Schneider et al., 2001; Vollrath, 1982). There is an extreme degree of population structure because of the low number of founders and repeated inbreeding within colonies (see previous section, Smith and Hagen, 1996; reviewed in Avile´s, 1997). Studies of life‐history traits in colonies of increasing size classes have revealed that group living in spiders is associated with high per capita reproductive costs (Avile´s, 1997; Avile´s and Tufin˜o, 1998; Bilde et al., 2007; Riechert, 1985; Seibt and Wickler, 1988b). The benefits of group living are enhanced colony (group) survival with increasing colony size, at least until the colony reaches an intermediate optimal size. Survival benefits at the level of the colony (the group) may outweigh the individual costs paid due to a decrease in fecundity with increasing colony size (Avile´s and Tufin˜o, 1998; Bilde et al., 2007). Another interesting aspect of group selection models is the evolution of female‐biased sex ratios, which can evolve by higher‐level (between‐group) selection (Avile´s, 1986, 1993a; Wilson and Colwell, 1981). This is because higher‐level selection favors larger colonies that have a greater probability to survive and reproduce. A female‐biased sex ratio would allow the group to grow at a higher rate and reach the threshold beyond which survival benefits and group productivity are maximized. Avile´s (1993a) used a simulation model to show that a female‐biased sex ratio evolves in populations where isolated lineages grow for a number of generations, proliferate after reaching a certain threshold size, and eventually become extinct. Smaller populations, lower migration rates, and higher turnover rates resulted in an increasingly female‐biased sex ratio, while a proliferation threshold was essential, that is only colonies above a certain population size give rise to successful new colonies. These conditions appear to describe the dynamics of some social

124

YAEL LUBIN AND TRINE BILDE

spiders, for example, A. eximius and S. dumicola. Thus, while ecological benefits may give advantages to large colony size, higher‐level selection will promote a sex ratio bias that enables colonies to grow rapidly and reach the threshold for proliferation (Avile´s, 1986, 1993a). Avile´s (1993a) argued that Hamilton’s (1967) local mate competition (LMC) model for the evolution of female‐biased sex ratios in structured populations and the group selection model presented for social spiders are ‘‘two extremes of a continuum in which the same selective forces are involved.’’ Hamilton’s (1967) argument was that when mating occurs among the progeny of one or a few founders of a group, the best strategy is to produce exactly enough sons to fertilize all the daughters, which results in a departure from the sex ratio equilibrium proposed by Fisher (1930). Inseminated females then disperse to colonize new resources. Hamilton originally considered one generation of inbreeding in his model, while there was much controversy whether the LMC model can explain the maintained female‐biased sex ratio during subsequent generations of inbreeding (as is the case of social spiders) (Avile´s, 1993a; Bulmer and Taylor, 1980; Charnov, 1982; Nagelkerke and Sabelis, 1996; Nunney, 1985; Werren, 1983; Wrensch and Ebbert, 1993). Frank (1987) pointed out that in structured populations, the interaction between cooperative effects that result in higher group productivity and genetic relatedness will favor a skewed sex ratio. Avile´s (1993a) considered multigenerational groups and argued that natural selection selects acts against the female‐biasing allele within groups, so female‐biased sex ratios are maintained in the metapopulation only because they increase productivity and thus proliferation of the groups in which they occur. The social spider model becomes a LMC model if the number of generations of local mating is reduced to one before the groups disperse (Avile´s, 1993a). The main point is that female‐biasing alleles can only be maintained if groups that contain a higher proportion of females are most successful in producing dispersers (Wilson and Colwell, 1981; see also Charnov, 1982). In Avile´s model, three conditions must be met: (1) a group‐level selective advantage of biased sex ratios, (2) enough population subdivision to create between‐group variance, and (3) a nonzero rate of group turnover (Avile´s, 1993a). The underlying process of sex ratio evolution may be adaptive parental control of the sex ratio (Nunney, 1985). Yet, in the social spiders, there is no evidence for adaptive changes in the sex ratio with increasing colony size (Henschel et al., 1995a; Lubin, 1991). It may be that the social spiders are unable to alter the sex ratios of their offspring in response to environmental or genetic changes if they lack adaptive control over sex ratio determination (Hurst and Vollrath, 1992). The heterogametic sex determination system in

THE EVOLUTION OF SOCIALITY IN SPIDERS

125

spiders suggests that social spiders cannot adjust sex ratios as easily as seen in some haplo‐diploid insect systems (see Section V.D; Avile´s et al., 1999, 2000; White, 1973). The unique traits of the social spiders including constraints on dispersal, selection for rapid colony growth, and continued inbreeding over multiple generations resulting in very high intracolony relatedness suggest that the conditions for colony‐level benefits to maintain the female‐biased sex ratios are present. Hence, the female‐biased sex ratio of the social spiders prompts some very interesting questions for further studies: What is the sex ratio determination mechanism in social spiders? How is the seemingly fixed female‐biased sex ratio maintained? Is there potential for adaptive adjustments of the sex ratio with colony size and age? C. ECOLOGICAL BENEFITS Social groups can form when individuals aggregate because of inherent advantages of group living. These benefits are usually associated with defense against predators or detection and harvest of food resources (Alexander, 1974). This route to group living predicts an increase in average individual fitness as a function of group size, where costs and benefits of group living result in an optimum group size at some intermediate level (Avile´s, 1999; Emlen, 1991; Rannala and Brown, 1994). Group living in spiders may be advantageous as a defense against predators (Henschel, 1998), but there is no evidence that aggregation in response to predation pressure was the route to group living (as is proposed for the social gall thrips and aphids, coined ‘‘fortress defenders’’; Costa, 2006). Similarly, there is no evidence that social spider groups form in response to foraging benefits, on the contrary, resource competition as revealed by a reduction in body size and individual reproductive success with increasing colony size suggests a growing competition over food with group size (Avile´s and Tufin˜o, 1998; Bilde et al., 2007; Seibt and Wickler, 1988b). Nevertheless, both protection and foraging could provide net benefits to living in groups within a range of group sizes and may be factors that act to maintain sociality. The presence of multiple females in a nest may ensure that at least some of the brood will survive to maturity. This is the basis of the ‘‘fostering model’’ proposed by Jones et al. (2007). In this model, a female’s reproductive success depends on the probability that she will survive until the young become independent and is a function of the number of females in the colony, as well as the length of the period of dependence of the young on maternal care (altricial period) and the risk of mortality. The assumption is that if the mother dies during the altricial period, another female will foster

126

YAEL LUBIN AND TRINE BILDE

her young. In the case of relatively long altricial periods, this model predicts an advantage to group living. The model was developed to explain the occurrence of colonies of A. studiosus containing multiple reproducing females in cool temperate habitats, where development time is long relative to warmer, subtropical habitats (Furey, 1998). Such a scenario might provide an initial advantage to reproducing in a group independent of relatedness among its members. D. ECOLOGICAL CONSTRAINTS Constraints models assume the formation of groups by delayed dispersal of offspring, which remain in their natal group because of factors that restrict their individual dispersal and breeding attempts (Emlen, 1991; Koenig and Dickinson, 2004). Individuals that remain in the group may pay a net per capita fitness cost compared with solitary breeding. However, a low probability of breeding successfully, for example because of limited breeding sites, may shift the balance in favor of group living. Differences between years and geographical areas are likely to cause variation in the magnitude of the constraint, leading to predictable patterns in the size of groups and the benefits of helping behavior. Delayed dispersal of offspring in the natal nest leads to the formation of genetic family groups. This route to sociality is predominant in cooperatively breeding birds and mammals (Emlen, 1991; Koenig and Dickinson, 2004) and is carried to the extreme in the social spiders where the elimination of premating dispersal and formation of sibling groups further lead to mating among group members and hence inbreeding (reviewed in Avile´s, 1997). The expected benefits of delayed dispersal are reduced costs of mobility, including predation, desiccation, search costs for suitable habitat or mates, avoiding competition with nonrelatives (Andersson, 1994; Hamilton and May, 1977), and foraging benefits (saving silk, capture of larger prey, reduced variance in prey capture; see Avile´s, 1997). Some major costs are expected from group living: increased homozygosity or loss of heterozygote advantage as a consequence of inbreeding (Charlesworth and Charlesworth, 1987; Roff, 2002) and competition among kin over food and reproduction (Griffin and West, 2002; Hamilton and May, 1977). Constraints on dispersal appear to be a strong factor favoring group living in social spiders, overriding the negative effect of declining fecundity with increasing group size (Avile´s and Tufin˜o, 1998; Bilde et al., 2007). E. GAME THEORY MODELS Cooperation based on reciprocity can be an evolutionarily stable strategy (ESS) even among unrelated individuals (Axelrod and Hamilton, 1981; Trivers, 1971). Two conditions must be met: (1) an individual is not able

THE EVOLUTION OF SOCIALITY IN SPIDERS

127

to cheat without the other individual being able to retaliate, that is defectors are recognized; and (2) the probability of the same two individuals meeting must be high. Hence, game theory models usually imply individual recognition, but perhaps group recognition and repeated interactions among members of a group are sufficient for it to play a role in the maintenance of sociality. Cooperation may evolve initially either through kin selection mediated by recognition cues or clustering, or by reciprocation of cooperative actions (Axelrod and Hamilton, 1981). In the social spiders, group members usually remain in the colony throughout their life span (review in Avile´s, 1997) and therefore have repeated interactions. Thus, the second condition for cooperation based on reciprocity is met. It is unknown, however, whether specific individuals recognize and remember previous specific interactions and whether kin discrimination is under positive selection (see Section V.A.1). Packer and Ruttan (1988) suggested that there will be a temptation to cheat, in this case to join in feeding without participating in the actual capture, if the prey is large, but not if it is small enough to be monopolized by a single individual. Cooperation can be maintained in spite of cheating if a single individual is unlikely to succeed in capturing prey on its own. As group size increases, the degree of cooperation should decrease because the advantage of group capture will be outweighed by the losses to cheaters. Thus, it is not necessary to invoke reciprocity. The data on group foraging in spiders described earlier (Section III.A) are consistent with these ideas: not all individuals feeding on a prey item were involved in capturing it, the first spiders to feed tried to prevent others from joining, and joiners reduced the value of prey for the hunter (Ward and Enders, 1985; Willey and Jackson, 1993). F. BY‐PRODUCT MUTUALISM Cooperation via by‐product mutualism occurs when an individual pays an immediate cost of not cooperating (Brown, 1983; Connor, 1995). The immediate benefits outweigh the costs of cooperating, and the benefits for the group are shared so that either all individuals cooperate or none cooperate. By‐product mutualism differs from reciprocity in two fundamental ways: (1) there is no temptation to cheat and (2) no scorekeeping is required, where reciprocity requires some form of keeping track of your partner’s behaviors. The African social spider S. dumicola suffers from ant attacks, and during ant raids, the spiders take turns in spinning cribellate silk below the entries to their nests, thereby physically preventing casualties (Henschel, 1998). Comparing survival between solitary individuals and groups, all solitary females died during ant attacks, whereas some individuals survived in 85% of the larger colonies (Henschel, 1998). Presumably, the more spiders that

128

YAEL LUBIN AND TRINE BILDE

can contribute in preventive silk spinning, the greater the chance of colony survival. This may be an example where immediate survival benefits outweigh the cost of cooperation. In the context of by‐product mutualism, the net benefit of cooperation, in this case increased survival, should be greater than the benefit of not contributing in nest defense and saving silk. Another example comes from the African A. consociata where the energetic costs of maintaining individual nests during the rainy season are alleviated by cooperative building of capture webs (Riechert et al., 1986). VI. TRANSITIONS IN THE EVOLUTION OF SOCIALITY: PROCESSES AND PATTERNS The evolution of sociality in spiders was accompanied by changes in the dispersal pattern and breeding structure and in the degree of cooperation in reproduction among individuals within the group. How did these changes come about and what were the preconditions for the transition to sociality? Selection favoring the transition to sociality could differ from the mechanisms that maintain sociality. The phylogenetic evidence points to a subsocial‐to‐social transition route (though Mallos gregalis may be an exception to this, as maternal care is not known to occur in this genus), and therefore, our discussion focuses on traits found in subsocial species that may have facilitated the transition to sociality. Subsocial species provide a window onto the ways these changes might have come about. A. FROM PREMATING TO POSTMATING DISPERSAL The shift from premating dispersal of offspring in the solitary and subsocial species to postmating dispersal, or no dispersal in some generations, in social species is a seemingly unlikely and puzzling transition given the generality of the (largely) theoretical benefits to dispersal (Hamilton and May, 1977; Lambin et al., 2001). Understanding the factors that promoted a loss of premating dispersal is the key to understanding the evolution of sociality in spiders (Waser et al., 1986, 1994). Thus, the clues to the evolution of philopatry of both sexes should be sought in the behavior, life history, and demography of the subsocial species (Avile´s and Gelsey, 1998). Premating dispersal of both sexes is seen in all subsocial species (Table II), although the timing or distance of dispersal may differ between the sexes. In the subsocial A. arizona (Theridiidae) (formerly cf. jucundus), females disperse as subadults or mature in the maternal nest, but males mature and disperse before their female siblings reach adulthood (Aviles and Gelsey, 1998; Powers and Aviles, 2003). In some other subsocial species, dispersal is more flexible. In D. ergandros (Thomisidae), mated females disperse to establish individual nests, but males apparently can disperse either before or after reaching adult stage (Evans, 1995). This species is subsocial in the

THE EVOLUTION OF SOCIALITY IN SPIDERS

129

sense that the colonies last only a single generation. Yet, by having female postmating dispersal, it conforms to the typical constellation of social traits described above. Males seemingly have a mixed dispersal strategy, but unlike its social congener, D. socialis, D. ergandros has an unbiased sex ratio, suggesting that inbreeding is not common (Evans, 1995). Relatedness among young within the nest decreased with age of the young, an indication that also the young move between nests (Evans, 1999); this may be the way outbreeding is maintained. In other facultatively group‐living species such as some A. studiosus populations and A. cf. studiosus in Uruguay, males apparently retain the behavior of dispersing out of the parent nest. It is unknown, however, whether a mixed breeding strategy occurs with some inbreeding as well as outbreeding. Further study of these species that appear to be at the transition between subsocial and social is needed. There is a considerable variation in the duration of maternal care and timing of dispersal. In a subsocial agelenid (Coelotes terrestris) and amaurobiid (Amaurobius ferox), as well as in subsocial eresids (for example, S. lineatus), young disperse from the maternal nest as small juveniles, while in subsocial Anelosimus (e.g., A. jucundus, A. arizona, or A. studiosus), dispersal occurs at a later (often penultimate) instar and there is a long period during which juveniles remain together in the nest after the female dies (Table II). Agnarsson et al. (2007) distinguished between ‘‘quasi‐solitary’’ species, in which young disperse at a small body size and after a short period of maternal care, and typical subsocial, in which dispersal occurs at a later (often penultimate) instar and there is a long period during which juveniles remain together in the nest after the female dies. Rayor and Taylor (2006) refer to these two categories as ‘‘transient’’ and ‘‘prolonged’’ subsocial, respectively, and acknowledge that these reflect a continuum of subsocial behavior. The common denominator of all of these species is that juvenile dispersal follows a period of obligatory maternal care, while the stage at which dispersal occurs may be dictated by food conditions and intranest competition (Aviram, 2000) or other environmental conditions. When abundant prey were provided, S. lineatus young could remain together until they were adults (Schneider, 1995) and the addition of prey lengthened the duration that the young remain in the maternal nest in a range of species tested (Theridion pictum: Ruttan, 1990; C. terrestris: Gundermann and Krafft, 1993; A. ferox: Kim, 2000; Tegenaria atrica: Pourie´ and Trabalon, 2001). The delay of juvenile dispersal appears to be a gradual phenomenon and shows considerable plasticity both among and within species, while the loss of premating dispersal accompanying regular inbreeding may have been a more abrupt transition. Breeding site philopatry in its extreme form of remaining in the maternal nest could have benefits for both young and adult stages (Jones and Parker, 2002). For juveniles and maturing females, the nest is a costly resource, with

130

YAEL LUBIN AND TRINE BILDE

its large investment in silk and protective structure (Kim, 2005; Myles, 1988; Seibt and Wickler, 1990). In stable habitats, with low year‐to‐year spatial variation in resources, it may be advantageous for females to remain in the previously successful maternal nest. Remaining in the nest could have benefits associated with group living as well (see Section V.C). But, even if females benefit from philopatry by inheriting the nesting site and obtaining more food, they will face competition from close relatives (Griffin and West, 2002; Hamilton and May, 1977). Indeed, in subsocial A. studiosus, long‐lived mothers chase maturing female offspring from the nest (Brach, 1977; Furey, 1998; Jones and Parker, 2002). For males, the presence of females on the nest assures that they will find a mate, but why should males not disperse to find additional, and in particular unrelated females? Data on the conditions that reduce the value of mating dispersal for males are critical to understand the evolution of this inbred social system (Waser et al., 1994). Restricted opportunities for finding unrelated females can limit male dispersal. When females are clustered in kin groups, the distance to an unrelated female may be quite large. Kin clusters, risk of predator attacks, and extreme environmental conditions may all favor a reduction of mate‐searching behavior (Andersson, 1994). In the subsocial S. lineatus, males mated with one or two females on average (Maklakov et al., 2005; Schneider and Lubin, 1998) and the distribution of distances that males traveled was bimodal, such that males first visited a nearby female and later moved to a distant female (Bilde et al., 2005). As neighboring spiders are often siblings, these philopatric matings enhance population substructuring into kin groups (Bilde et al., 2005; Johannesen and Lubin, 1999, 2001). This type of male mating dispersal could represent a first step toward inbred sociality. B. FROM OUTBREEDING TO INBREEDING The systems of dispersal and breeding are linked. Social species lacking premating dispersal of either sex have regular and continuous inbreeding (Avile´s, 1997; Riechert and Roeloffs, 1993; Smith and Hagen, 1996), while subsocial species with premating dispersal of both sexes outbreed (Avile´s and Bukowski, 2006; Bilde et al., 2005). In subsocial species with multiple breeding females, there is a potential for inbreeding. In the Australian huntsman spider D. cancerides, young mature in the nest and may remain as adults (L. Rayor, personal communication) and potentially inbreed. Allozyme data suggest, however, that the young in a nest are derived from different fathers (Rowell and Avile´s, 1995). Foreign males are accepted into the colony (L. Rayor, personal communication), suggesting male movement between nests. This is supported also by the lack of female‐biased sex ratios (Rowell and Avile´s, 1995).

THE EVOLUTION OF SOCIALITY IN SPIDERS

131

The transition stage is expected to entail a loss of fitness caused by inbreeding depression (Roff, 1997, 2002); hence, many animals, including most social species (ants, social wasps, cooperatively breeding mammals, and birds) have various mechanisms to avoid or prevent inbreeding (Cook and Crozier, 1995; Greenwood, 1980; Johnson and Gaines, 1990; Pusey and Wolf, 1996). The fitness costs associated with inbreeding could act as a constraint on the evolution of sociality (Avile´s, 1997; Avile´s and Bukowski, 2006; Bilde et al., 2005). If, however, the costs of inbreeding are not large and inbreeding avoidance is lacking or weak, the loss of outbreeding may not be a severe constraint (Chesser and Ryman, 1986; Waser et al., 1986). Bilde et al. (2005) tested this hypothesis by subjecting the subsocial eresid S. lineatus to two levels of inbreeding: sib mating and within‐cluster mating. The latter test simulated the level of inbreeding found in kin‐structured populations in nature (Johannesen and Lubin, 1999, 2001). Behavioral inbreeding avoidance was lacking altogether and mild inbreeding depression was expressed only in the growth rate and adult body size. A similar sib‐mating study was conducted on the subsocial theridiid A. arizona (formerly A. cf. jucundus) by Avile´s and Bukowski (2006). In this species, inbreeding depression was evident only in postdispersal (penultimate‐instar) young; there were no obvious inbreeding effects during the stages in the maternal nest, both when the mother was present and during the sibling cohabitation stage. Avile´s and Bukowski (2006) suggested that the benefits of extended maternal care and of group living compensate for inbreeding costs, but that these benefits do not extend beyond the social stage. The impression from these two studies is that the fitness losses from inbreeding depression may have been overcome in a gradual process starting from philopatric dispersal of young, leading to kin‐structured populations with occasional inbreeding, and ending with continuous inbreeding in nests of nondispersing offspring. A history of inbreeding could allow for purging of deleterious recessive alleles thus reducing the (short‐term) cost of inbreeding (Charlesworth and Charlesworth, 1987). C. FROM MATERNAL CARE TO COOPERATIVE BREEDING Extended maternal care in the subsocial species takes the form of tending the eggsacs, for example moving them within the nest to obtain a better microclimate and protecting them from predators or parasites, and feeding of young by capturing prey for them, regurgitation of predigested prey, production of trophic oocytes, or by matriphagy (Evans et al., 1995; Gundermann et al., 1988; Jones and Parker, 2000; Kim and Horel, 1998; Kim and Roland, 2000; Kullmann et al., 1971; Marques et al., 1998; Rowell and Avile´s, 1995). The young obtain direct benefits from extended maternal care in the form of improved survival and faster growth (Salomon et al., 2005; Schneider, 2002). Maternal effects play a large role in the success of

132

YAEL LUBIN AND TRINE BILDE

the young in subsocial species: large females have both larger clutches and provide more resources during the maternal‐care stages, resulting in larger young at the stage of independence (Aviram, 2000; Gundermann, 1997; Kim and Roland, 2000; Kim et al., 2000; Salomon et al., 2005). Young that disperse at a larger body size will have an advantage in survival and growth over smaller dispersers (Aviram, 2000; Jones and Parker, 2002). Extended maternal care comes at the expense of clutch size and, indeed, clutch sizes of species with maternal care tend to be small relative to those of similar‐ size females lacking maternal care (Schneider, 1996a), reflecting an evolutionary trade‐off between number and size of progeny (Fox and Czesak, 2000; Stearns, 1992). Cooperative breeding in the social spider species provides both the general benefits of group living that accrue to offspring (see Section V.C) as well as benefits derived specifically from allomaternal care. Allomaternal care involves guarding of eggsacs of several females placed together in a cluster or brood chamber or feeding of young by multiple females, some of which may be nonbreeders. Even without the presence of helpers, the benefits of group living alone may be sufficient to select for breeding together. For example, A. jabaquara females breed in the same nest but are intolerant of one another and defend their individual eggsacs and early instar brood from one another (Marques et al., 1998). Several subsocial or transition species can be found in breeding groups, where females raise their young together (Table II; D. cancerides, A. jabaquara, A. studiosus, A. cf. studiosus). Once the young leave the vicinity of the guarded eggsac or brood, they will mix with young of other females and thus will benefit from a certain degree of allomaternal care. Jones and Parker (2002) showed direct benefits to the young and both indirect and direct fitness benefits to the reproducing females in breeding groups of A. studiosus. Thus, group living in these transition species has at least some of the components of cooperative breeding typical of inbred social species. The question is: do some females in these groups forego reproduction altogether and remain in the group as helpers, or do they disperse to breed solitarily when unable to breed in the communal nest? By extension, we need to ask: is helping a reproductive strategy in the social spiders, namely, an alternative to reproducing on one’s own, or is it only a consequence of ecological constraints on reproduction? Schneider (2002) investigated the physiological preconditions to helping behavior by cross‐fostering young of the subsocial S. lineatus to females in different reproductive stages. Unmated females and females with an eggsac provided little or no care for foster young, while females with young of their own fed the foster young both by regurgitation and matriphagy (see Bessekon et al., 1992 for a similar result with the subsocial agelenid spider

THE EVOLUTION OF SOCIALITY IN SPIDERS

133

C. terrestris). Salomon and Lubin (2007) showed, however, that in the social S. dumicola nonbreeding adult and subadult females feed young both by regurgitation and ‘‘allomatriphagy.’’ Thus, a switch in reproductive physiology and behavior must have occurred at the transition to sociality that enabled subadult or unmated females to exhibit maternal behaviors. These directed allomaternal behaviors are not present in the outbred transition species. Thus, we propose that helping in the outbred transition species mentioned above is not an alternative reproductive strategy, but rather a by‐product of maternal care where the young of different females mix in the nest and are treated equally by all reproducing females (but see Evans, 1998). Under a kin‐selection scenario, helping as a distinct reproductive strategy should develop in species where average relatedness among nest members is higher than among individuals within the breeding population at large (Michod, 1993). This could occur if sisters remain together in the maternal nest and breed together, as appears to be the case in A. studiosus and D. cancerides (Rowell and Avile´s, 1995). If helping in these transition species is favored by kin selection, it is interesting that it has not developed further. Two further considerations add to the mystery. First, helping by nonbreeders is common in the social insects as well as in birds and mammals and is widely believed to have arisen by kin selection without inbreeding. The second and perhaps more basic problem is that when inbreeding is continuous and colonies, and even local populations, are genetically homogeneous, the strength of kin selection should decline and helping may no longer be favored (Michod, 1993). This paradox may be resolved, however, if colony‐level benefits of helping are taken into account. As with the evolution of the female‐biased sex ratios, a trait that increases group productivity would enable colonies to grow more rapidly to a size at which dispersing propagules can be produced (Avile´s, 1986, 1993a; see also Section V.B). Additionally, helping behavior may have had a direct benefit during the transition to inbreeding by reducing the effects of inbreeding depression on individual reproduction, as suggested by Avile´s and Bukowski (2006). Finally, once inbred sociality was established, helping would come under positive selection as a means of lowering the costs of competition in large colonies.

VII. SUMMARY: FROM SUBSOCIAL TO INBRED SOCIAL, AN OVERVIEW A possible scenario for the evolution of sociality is the following: in a subsocial species with extended maternal care, some of the young inherit the maternal nest and breed together. Male dispersal at this stage will

134

YAEL LUBIN AND TRINE BILDE

maintain outbreeding. Care of the young is limited to one’s own offspring, at least in the early development stages. This social structure may be seen in some A. studiosus populations, A. cf. studiosus in Uruguay, A. jabaquara, S. tentoriicola, and no doubt other species whose biology is less well known. Remaining in the maternal nest and breeding together is favored by kin selection combined with constraints on dispersal and independent nesting and/or large benefits from sharing resources. Kin‐structured populations and philopatric male mating dispersal will lower the barrier to full inbreeding by purging deleterious alleles and thus reducing the negative fitness consequences of inbreeding. At the same time, selection will act on reproductive traits such as the sex ratio and helping behavior in ways that favor rapid colony growth. This stage of the transition should occur only when local populations are already significantly inbred and male dispersal away from the local population is strongly disfavored. The conditions favoring this transition remain poorly understood. It is evident, though, that this is a one‐way street: a reversal to subsocial from inbred social seems unlikely based on phylogenetic evidence, and further evolution of social clades likewise does not seem to occur. This evolutionary pathway has been portrayed as an ‘‘evolutionary dead end’’ (Agnarsson et al., 2006; Avile´s, 1997; Johannesen et al., 2007). Nevertheless, many of the inbred social species are highly successful, are widely distributed geographically, and show considerable differentiation on a geographic scale. Furthermore, some species have a long evolutionary history. This pattern could come about only by means of continued small‐scale evolutionary adjustments to the genome, namely, by microevolution at the geographic scale. Numerous puzzles remain to be solved. What is the role of long‐distance dispersal of mated females in these inbred social species? The fact that it occurs in most of the species studied so far, but by different and likely independently evolved methods, suggests that it is an essential element of this social system. The role of disease and parasitism in selecting for propagule dispersal (Hamilton, 1987) should be investigated further, both empirically and theoretically. How do the homozygous and variation‐poor genomes adapt to a wide range of ecological conditions? Is cooperative breeding a risk‐sensitive strategy that enables colonies to cope with a wider range of ecological conditions by reducing the variance in food intake within the colony? And, when do reproductive skew and helping by nonbreeders become a group‐selected strategy favoring rapid colony growth? Finally, what are the life‐history consequences of inbred sociality—for example, body size, growth rates, fecundity, and mortality schedules—and how do they affect the potential for evolutionary change? The inbred social spiders provide a rich and fertile field for understanding evolutionary processes.

THE EVOLUTION OF SOCIALITY IN SPIDERS

135

Acknowledgments We are grateful to I. Agnarsson, L. Avile´s, T. Jones, J. Miller, L. S. Rayor, S. E. Riechert, M. Salomon, and C. Viera for providing us with unpublished information and manuscripts in press; to L. Avile´s and D. R. Smith for comments on the chapter; and to L. S. Rayor and S. E. Riechert for thoughtful reviews. Above all, we are grateful to the editor H. J. Brockmann for inviting us to write this chapter and for her patience throughout. This is publication number 506 of the Mitrani Department of Desert Ecology.

References Agnarsson, I. (2004). Morphological phylogeny of cobweb spiders and their relatives (Araneae, Araneoidea, Theridiidae). Zool. J. Linn. Soc. 141, 447–626. Agnarsson, I. (2006). A revision of the New World eximius lineage of Anelosimus (Araneae, Theridiidae) and a phylogenetic analysis using worldwide exemplars. Zool. J. Linn. Soc. 146, 453–593. Agnarsson, I., and Kuntner, M. (2005). Madagascar: An unexpected hotspot of social Anelosimus spider diversity (Araneae: Theridiidae). Syst. Entomol. 30, 575–592. Agnarsson, I., Avile´s, L., Coddington, J. A., and Maddison, W. P. (2006). Sociality in theridiid spiders: Repeated origins of an evolutionary dead end. Evolution 60, 2342–2351. Agnarsson, I., Maddison, W. P., and Avile´s, L. (2007). The phylogeny of social Anelosimus spiders (Araneae: Theridiidae) inferred from six molecular loci and morphology. Mol. Phylogen. Evol. 43, 833–851. Ainsworth, C., Slotow, R., Crouch, T., and Lubin, Y. (2002). Lack of task differentiation during prey capture in the group living spider Stegodyphus mimosarum (Araneae, Eresidae). J. Arachnol. 30, 39–46. Alexander, R. D. (1974). The evolution of social behaviour. Annu. Rev. Ecol. Syst. 5, 325–383. Alexander, R. D., and Borgia, G. (1978). Group selection, altruism, and levels of organization of life. Ann. Rev. Ecol. Syst. 9, 449–474. Amir, N., Whitehouse, M. E. A., and Lubin, Y. (2000). Food consumption and competition in a communally feeding social spider, Stegodyphus dumicola (Eresidae). J. Arachnol. 28, 195–200. Anderson, J. F. (1978). Energy content of spider eggs. Oecologia 37, 41–57. Andersson, M. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton. Avile´s, L. (1986). Sex‐ratio bias and possible group selection in the social spider Anelosimus eximius. Am. Nat. 128, 1–12. Avile´s, L. (1993a). Interdemic selection and the sex ratio: A social spider perspective. Am. Nat. 142, 320–345. Avile´s, L. (1993b). Newly‐discovered sociality in the neotropical spider Aebutina binotata Simon (Dictynidae?). J. Arachnol. 21, 184–193. Avile´s, L. (1994). Social behaviour in a web‐building lynx spider Tapinillus sp. (Araneae: Oxyopidae). Biol. J. Linn. Soc. 51, 163–176. Avile´s, L. (1997). Causes and consequences of cooperation and permanent‐sociality in spiders. In ‘‘The Evolution of Social Behavior in Insects and Arachnids’’ (B. Crespi and J. Choe, Eds.), pp. 476–498. Cambridge University Press, Cambridge, Massachusetts. Avile´s, L. (1999). Cooperation and non‐linear dynamics: An ecological perspective on the evolution of sociality. Evol. Ecol. Res. 1, 459–477. Avile´s, L. (2000). Nomadic behaviour and colony fission in a cooperative spider: Life history evolution at the level of the colony? Biol. J. Linn. Soc. 70, 325–329.

136

YAEL LUBIN AND TRINE BILDE

Avile´s, L., and Bukowski, T. (2006). Group living and inbreeding depression in a subsocial spider. Proc. Roy. Soc Lond. B 270, 157–163. Avile´s, L., and Gelsey, G. (1998). Natal dispersal and demography of a subsocial Anelosimus species and its implications for the evolution of sociality in spiders. Can. J. Zool. 76, 2137–2147. Avile´s, L., and Maddison, W. (1991). When is the sex ratio biased in social spiders? Chromosome studies of embryos and male meiosis in Anelosimus species. J. Arachnol. 19, 126–135. Avile´s, L., and Salazar, P. (1999). Notes on the social structure, life cycle, and behavior of Anelosimus rupununi. J. Arachnol. 27, 497–502. Avile´s, L., and Tufin˜o, P. (1998). Colony size and individual fitness in the social spider Anelosimus eximius. Am. Nat. 152, 403–418. Avile´s, L., Varas, C., and Dyreson, E. (1999). Does the African social spider Stegodyphus dumicola control the sex of individual offspring? Behav. Ecol. Sociobiol. 46, 237–243. Avile´s, L., McCormack, J., Cutter, A., and Bukowski, T. (2000). Precise, highly female‐biased sex ratios in a social spider. Proc. R. Soc. Lond. B 267, 1445–1449. Avile´s, L., Maddison, W. P., Salazar, P. A., Este´vez, G., Tufin˜o, P., and Can˜as, G. (2001). Aran˜as sociales de la Amazonı´a equatoriana, con notas sobre seis especies sociales no descritas previamente. Rev. Chilena de Historia Natural 74, 619–638. Avile´s, L., Maddison, W. P., and Agnarsson, I. (2006). A new independently derived social spider with explosive colony proliferation and female size dimorphism. Biotropica 36, 743–753. Aviram, A. (2000). Maternal effects on dispersal decisions and fitness of juveniles in the desert spider Stegodyphus lineatus (Eresidae). M.Sc. thesis, Fac. Natural Sci., Ben‐Gurion University, Beer Sheva, Israel. Axelrod, R., and Hamilton, W. D. (1981). The evolution of cooperation. Science 211, 1390–1396. Bertram, B. C. (1978). Living in groups: Predators and prey. In ‘‘Behavioural Ecology, An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), pp. 64–96. Blackwell Scientific Publications, Oxford, England. Bessekon, D. A., Horel, A., and Gundermann, J. L. (1992). Onset and maintenance of material prey‐supply activity in the spider Coelotes terrestris (Araneae, Agelenidae). Behaviour 120, 1–10. Bilde, T., and Lubin, Y. (2001). Kin recognition and cannibalism in a subsocial spider. J. Evol. Biol. 14, 959–966. Bilde, T., Coates, K., Birkhofer, K., Bird, T., Maklakov, A., Lubin, Y., and Avile´s, L. (2007). Survival benefits select for group living despite reproductive costs in a social spider. J. Evol. Biol. (doi: 10.1111/j.1420‐9101.2007.01407.x). Bilde, T., Lubin, Y., Smith, D., Schneider, J., and Maklakov, A. A. (2005). The transition to social inbred mating systems in spiders – Role of inbreeding tolerance in a subsocial predecessor. Evolution 59, 160–174. Bodasing, M., Slotow, R., and Crouch, T. (2001). The importance of group size to dispersal of Stegodyphus mimosarum. J. Arachnol. 30, 373–382. Bond, J. E., and Opell, B. D. (1997). Systematics of the spider genera Mallos and Mexitlia (Araneae, Dictynidae). Zool. J. Linn. Soc. 119, 389–445. Boulinier, T., McCoy, K. D., and Sorci, G. (2001). Dispersal and parasitism. In ‘‘Dispersal’’ (J. Clobert, E. Danchin, A. A. Dhondt, and J. D. Nichols, Eds.), pp. 169–179. Oxford University Press, Oxford, UK. Bourke, A. F. G., and Franks, N. R. (1995). ‘‘Social Evolution in Ants.’’ Princeton University Press, Princeton.

THE EVOLUTION OF SOCIALITY IN SPIDERS

137

Bowden, K. (1991). The evolution of sociality in the spitting spider, Scytodes fusca (Araneae: Scytodidae): Evidence from observations of intraspecific interactions. J. Zool. Lond. 223, 161–172. Brach, V. (1977). Anelosimus studiosus (Araneae: Theridiidae) and the evolution of quasisociality in spiders. Evolution 31, 154–161. Brockmann, H. J. (1997). Cooperative breeding in wasps and vertebrates: The role of ecological constraints. In ‘‘The Evolution of Social Behavior in Insects and Arachnids’’ (B. Crespi and J. Choe, Eds.), pp. 347–371. Cambridge University Press, Cambridge, Massachusetts. Brown, J. L. (1983). Cooperation—a biologistst dilemma. Adv. Study Behav. 13, 1–37. Burgess, J. W. (1979). Web‐signal processing for tolerance and group predation in the social spider Mallos gregalis Simon. Anim. Behav. 27, 157–164. Buskirk, R. E. (1981). Sociality in the Arachnida. In ‘‘Social Insects’’ (H. Hermann, Ed.), Vol. 2, pp. 281–367. Academic Press, London, New York. Bukowski, C. T., and Avile´s, L. (2002). Asynchronous maturation of the sexes may limit close inbreeding in a subsocial spider. Can. J. Zool. 80, 193–198. Bulmer, M. G., and Taylor, P. D. (1980). Sex ratio theory in geographically structured populations. Heredity 56, 69–73. Charlesworth, D., and Charlesworth, B. (1987). Inbreeding depression and its evolutionary consequences. Ann. Rev. Ecol. Syst. 18, 237–268. Charnov, E. L. (1982). ‘‘The Theory of Sex Allocation.’’ Princeton University Press, Princeton. Chesser, R. K., and Ryman, N. (1986). Inbreeding as a strategy in subdivided populations. Evolution 40, 616–624. Christenson, T. E. (1984). Behaviour of colonial and solitary spiders of the theridiid species Anelosimus eximius. Anim. Behav. 32, 725–734. Coddington, J., and Agnarsson, I. (2006). Subsociality in Helvibis thorelli Keyserling 1884 (Theridiinae, Theridiidae, Araneae) from French Guiana. J. Arachnol. 34, 642–645. Coddington, J., and Levi, H. W. (1991). Systematics and evolution of spiders (Araneae). Ann. Rev. Ecol. Syst. 22, 565–592. Connor, R. C. (1995). The benefits of mutualism—a conceptual framework. Biol. Rev. Camb. Philos. Soc. 70, 427–457. Costa, J. (2006). ‘‘The Other Insect Societies.’’ Belknap Press, Harvard University Press, Cambridge, Mass, USA and London, England. Costa, J., and Fitzgerald, T. D. (2005). Social terminology revisited: Where are we ten years later? Ann. Zool. Fennici 42, 559–564. Cook, J. M., and Crozier, R. H. (1995). Sex determination and population biology in the Hymenoptera. Trends Ecol. Evol. 10, 281–286. Crespi, B. J. (2005). Social sophistry: Logos and mythos in the forms of cooperation. Ann. Zool. Fennici 42, 569–571. Crespi, B. J., and Choe, J. (1997). Explanation and evolution of social systems. In ‘‘The Evolution of Social Behavior in Insects and Arachnids’’ (B. Crespi and J. Choe, Eds.), pp. 499–524. Cambridge University Press, Cambridge, Massachusetts. Crouch, T., and Lubin, Y. (2001). Population stability and extinction in a social spider Stegodyphus mimosarum (Araneae: Eresidae). Biol. J. Linn. Soc. 72, 409–417. Crouch, T., Lubin, Y., and Bodasing, M. (1998). Dispersal in the social spider Stegodyphus mimosarum Pavesi 1883 (Araneae: Eresidae). Durban Museum Novitates 23, 52–55. D’Andrea, M. (1987). Social behaviour in spiders (Arachnida, Araneae). Ital. J. Zool. N.S. Monogr. 3, 1–156. Danchin, E., Heg, D., and Doligez, B. (2001). Public information and breeding habitat selection. In ‘‘Dispersal’’ (J. Clobert, E. Danchin,, A. A. Dhondt, and J. D. Nichols, Eds.), pp. 243–258. Oxford University Press, Oxford, UK.

138

YAEL LUBIN AND TRINE BILDE

Darchen, R. (1967). Une nouvelle araigne´e sociale du Gabon Agelena republicana Darchen (Arane´ide labidognathe). Biol. Gabonica 3, 31–42. Darchen, R. (1968). Ethologie d’Achaearanea disparata Denis, Aranea, Theridiidae, araigne´e sociale du Gabon. Biol. Gabonica 4, 6–25. Darchen, R. (1976). La fondation de nouvelles colonies d’Agelena consociata et d’Agelena republicana, araigne´es sociales du Gabon. C. R. III Col. Arach. Fr. Les Eyzies 20–39. Darchen, R. (1978). Les essaimages lointains de l’araigne´e sociale Agelena consociata Denis (Araneae, Labidognatha) dans la foreˆt gabonaise. C.r. hebd. Se´anc. Acad. Sci. Paris 278, 1035–1037. Darchen, R., and Delage‐Darchen, B. (1986). Societies of spiders compared to the societies of insects. J. Arachnol. 14, 227–238. Darchen, R., and Ledoux, J. C. (1978). Achaearanea disparata, araigne´e sociale du Gabon, synonyme ou espe`ce jumelle d’ Achaearanea tesselatta, solitaire. Rev. Arachnol. 1, 121–132. Downes, M. F. (1993). The life history of Badumna candida (Araneae: Amaurobioidea). Aust. J. Zool. 41, 441–466. Downes, M. F. (1994a). The nest of the social spider Phryganoporus candidus (Araneae: Desidae): Structure, annual growth cycle and host plant relationships. Aust. J. Zool. 42, 237–259. Downes, M. F. (1994b). Courtship and mating in the social spider Phryganoporus candidus (Araneae: Desidae). Bull. Br. Arachnol. Soc. 9, 277–280. Downes, M. F. (1995). Australasian social spiders: What is meant by ‘social’? Rec. West. Austral. Mus. Suppl. No. 52, 25–32. Ebert, D. (1998). Behavioral asymmetry in relation to body weight and hunger in the tropical social spider Anelosimus eximius (Araneae, Theridiidae). J. Arachnol. 26, 70–80. Emlen, S. T. (1991). Evolution of cooperative breeding in birds and mammals. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), 3rd ed., pp. 301–337. Blackwell Scientific Publications, Oxford, UK. Emlen, S. T. (1997). Predicting family dynamics in social vertebrates. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), 4th ed., pp. 228–253. Blackwell Scientific Publications, Oxford, UK. Evans, T. E. (1995). Two new species of social crab spiders of the genus Diaea from eastern Australia, their natural history and distribution. Rec. West. Aust. Mus. 52, 151–158. Evans, T. E. (1998). Offspring recognition by mother crab spiders with extreme maternal care. Proc. R. Soc. Lond. B 265, 129–134. Evans, T. E. (1999). Kin recognition in a social spider. Proc. R. Soc. Lond. B 266, 287–292. Evans, T. E. (2000). Male work and sex ratio in social crab spiders. Ins. Soc. 47, 285–288. Evans, T. E., and Goodisman, M. A. D. (2002). Nestmate relatedness and population genetic structure of the Australian social crab spider Diaea ergandros (Araneae: Thomisidae). Mol. Ecol. 11, 2307–2316. Evans, T. E., and Main, B. Y. (1993). Attraction between social crab spiders: Silk pheromones in Diaea socialis. Behav. Ecol. 4, 99–105. Evans, T. E., Wallis, W. E. J., and Elgar, M. A. (1995). Making a meal of mother. Nature 376, 299. Falconer, D. S. (1989). ‘‘Introduction to Quantitative Genetics,’’ 3rd ed. Longman, New York. Fisher, R. A. (1930). ‘‘The Genetical Theory of Natural Selection.’’ Clarendon Press, Oxford. Fjerdingstad, E. J., and Crozier, R. H. (2006). The evolution of worker caste diversity in social insects. Am. Nat. 167, 390–400. Fowler, H. G., and Levi, H. W. (1979). A new quasisocial spider (Araneae: Theridiidae) from Paraguay. Psyche 86, 11–18.

THE EVOLUTION OF SOCIALITY IN SPIDERS

139

Fox, C. W., and Czesak, M. E. (2000). Evolutionary ecology of progeny size in arthropods. Annu. Rev. Entomol. 45, 341–369. Franks, N. R. (1987). The organization of working teams in social insects. Trends Ecol. Evol. 2, 72–75. Frank, S. A. (1987). Demography and sex ratio in social spiders. Evolution 41, 1267–1281. Frank, S. A. (1998). ‘‘Foundations in Social Evolution.’’ Princeton University Press, Princeton. Furey, R. E. (1998). Two cooperatively social populations of the theridiid spider Anelosimus studiosus in a temperate region. Anim. Behav. 55, 727–735. Ghione, S., Costa, F. G., and Viera, C. (2004). Ability to capture prey in early instars of the subsocial spider Anelosimus studiosus (Hentz, 1850) from Uruguay (Araneae, Therdiidae). Bull. Br. Arachnol. Soc. 13, 60–62. Gonzaga, M. O., and Vasconcellos‐Neto, J. (2001). Female body size, fecundity parameters and foundation of new colonies in Anelosimus jabaquara (Araneae, Theridiidae). Ins. Soc. 48, 94–100. Greenwood, P. J. (1980). Mating systems, philopatry and dispersal in birds and mammals. Anim. Behav. 28, 1140–1162. Griffin, A. S., and West, S. A. (2002). Kin selection: Fact and fiction. Trends Ecol. Evol. 17, 15–21. Griffin, A. S., and West, S. A. (2003). Kin discrimination and the benefit of helping in cooperatively breeding vertebrates. Science 302, 634–636. Gundermann, J.‐L. (1997). Costs and benefits of maternal care in a subsocial spider, Coelotes Terrestris. Ethology 103, 915–925. Gundermann, J.‐L., Horel, A., and Krafft, B. (1988). Maternal food supply activity and its regulation in Coelotes terrestris (Araneae, Agelenidae). Behaviour 107, 278–296. Gundermann, J.‐L., Horel, A., and Roland, C. (1991). Mother‐offspring food transfer in Coelotes terrestris (Araneae, Agelenidae). J. Arachnol. 19, 97–101. Gundermann, J.‐L., and Krafft, B. (1993). Experimental manipulations of social tendencies in the spider Coelotes terrestris. Ins. Soc. 40, 219–229. Hamilton, W. D. (1964). The genetical evolution of social behaviour I and II. J. Theoret. Biol. 7, 1–16, 17–52. Hamilton, W. D. (1967). Extraordinary sex ratios. Science 156, 477–488. Hamilton, W. D. (1987). Kinship, recognition, disease, and intelligence: Constraints of social evolution. In ‘‘Animal Societies: Theories and Facts’’ (Y. Ito, J. L. Brown, and J. Kikkawa, Eds.), pp. 81–102. Japan Scientific Society Press, Tokyo. Reprinted in Hamilton, W. D. (2001) ‘‘Narrow Roads of Gene Land: The Collected Papers of W. D. Hamilton,’’ Vol. 2, pp. 393–411. Oxford University Press, Oxford, UK. Hamilton, W. D., and May, R. M. (1977). Dispersal in stable habitats. Nature 269, 578–581. Henschel, J. R. (1993). Is solitary life an alternative for the social spider Stegodyphus dumicola? J. Namibian Sci. Soc. 43, 71–79. Henschel, J. R. (1998). Predation on social and solitary individuals of the spider Stegodyphus dumicola (Araneae, Eresidae). J. Arachnol. 26, 61–69. Henschel, J. R., and Lubin, Y. D. (1992). Environmental factors affecting the web and activity of a psammophilous spider in the Namib desert. J. Arid Environ. 22, 173–189. Henschel, J. R., Lubin, Y., and Schneider, J. (1995a). Sexual competition in an inbreeding social spider, Stegodyphus dumicola (Araneae: Eresidae). Ins. Soc. 42, 419–426. Henschel, J. R., Schneider, J., and Lubin, Y. (1995b). Dispersal mechanisms by the spiders Stegodyphus: Do they balloon? J. Arachnol. 23, 202–204. Hurst, L. D., and Vollrath, F. (1992). Sex‐ratio adjustment in solitary and social spiders. Trends Ecol. Evol. 7, 326–327.

140

YAEL LUBIN AND TRINE BILDE

Jackson, R. R. (1978). Male mating strategies of dictynid spiders with differing types of social organization. Symp. Zool. Soc. Lond. 42, 79–88. Jackson, R. R. (1979). Comparative studies of Dictyna and Mallos (Araneae, Dictynidae) II. The relationship between courtship, mating, aggression and cannibalism in species with different social organization. Rev. Arachnol. 2, 103–132. Jackson, R. R., and Smith, S. E. (1978). Aggregations of Mallos and Dictyna (Araneae, Dictynidae): Population characteristics. Psyche 85, 65–80. Jacson, C. C., and Joseph, K. J. (1973). Life history, bionomics and behaviour of the social spider Stegodyphus sarasinorum Karsch. Ins. Soc. 20, 189–204. Johannesen, J., Baumann, T., Seitz, A., and Veith, M. (1998). The significance of relatedness and gene flow on population genetic structure in the subsocial spider Eresus cinnaberinus (Araneae: Eresidae). Biol. J. Linn. Soc. 63, 81–98. Johannesen, J., and Lubin, Y. (1999). Group founding and breeding structure in the subsocial spider Stegodyphus lineatus (Eresidae). Heredity 82, 677–686. Johannesen, J., and Lubin, Y. (2001). Evidence for kin‐structured group founding and limited juvenile dispersal in the subsocial spider Stegodyphus lineatus. J. Arachnol. 29, 413–422. Johannesen, J., Hennig, A., Dommermuth, B., and Schneider, J. M. (2002). Mitochondrial DNA distributions indicate colony propagation by single matri‐lineages in the social spider. Stegodyphus dumicola. Biol. J. Linn. Soc. 76, 591–600. Johannesen, J., Lubin, Y., Bilde, T., Smith, D. R., and Schneider, J. M. (2007). The age and evolution of sociality in Stegodyphus spiders: A molecular phylogenetic approach. Proc. R. Soc. Lond. Proc. R. Soc. B 274, 231–237. Johnson, M. L., and Gaines, M. S. (1990). Evolution of dispersal; theoretical models and empirical tests using birds and mammals. Ann Rev. Ecol. Syst. 21, 449–480. Jones, T. C., and Parker, P. G. (2000). Costs and benefits of foraging associated with delayed dispersal in the spider Anelosimus studiosus (Araneae: Theridiidae). J. Arachnol. 28, 61–69. Jones, T. C., and Parker, P. G. (2002). Delayed juvenile dispersal benefits both mother and offspring in the cooperative spider Anelosimus studiosus (Araneae: Theridiidae). Behav. Ecol. 13, 142–148. Jones, T. C., Riechert, S. E., Dalrymple, S. E., and Parker, P. G. (2007). Fostering model explains variation in levels of sociality in spider system. Ann. Behav. 73, 195–204. Keller, L. F., and Waller, D. M. (2002). Inbreeding effects in wild populations. Trends Ecol. Evol. 17, 230–241. Kim, K. W. (2000). Dispersal behaviour in a subsocial spider: Group conflict and the effect of food availability. Behav. Ecol. Sociobiol. 48, 182–187. Kim, K. W. (2005). L’heritage maternel d’une araigne´e subsociale: Une toile pour la pre´dation collective des jeunes. C. R. Biol. 328, 89–95. Kim, K. W., and Horel, A. (1998). Matriphagy in the spider Amaurobius ferox (Araneae, Amaurobiidae): An example of mother‐offspring interactions. Ethology 104, 1021–1037. Kim, K. W., and Roland, C. (2000). Trophic egg‐laying in the spider Amaurobius ferox: Mother‐offspring interactions and functional value. Behav. Proc. 50, 31–42. Kim, K. W., Roland, C., and Horel, A. (2000). Functional value of matriphagy in the spider Amaurobius ferox. Ethology 106, 729–742. Kim, K. W., Krafft, B., and Choe, J. C. (2005a). Cooperative prey capture by young subsocial spiders. I. Functional value. Behav. Ecol. Sociobiol. 59, 92–100. Kim, K. W., Krafft, B., and Choe, J. C. (2005b). Cooperative prey capture by young subsocial spiders. II. Behavioural mechanism. Behav. Ecol. Sociobiol. 59, 101–107. Koenig, W. D., and Dickinson, J. L. (2004). ‘‘Ecology and Evolution of Cooperative Breeding in Birds.’’ Cambridge University Press.

THE EVOLUTION OF SOCIALITY IN SPIDERS

141

Krafft, B., and Pasquet, A. (1991). Synchronized and rhythmical activity during prey capture in the social spider Anelosimus eximius (Araneae, Theridiidae). Ins. Soc. 38, 83–91. Krafft, B., Horel, A., and Julita, J.‐M. (1986). The influence of food supply on the duration of the gregarious phase of a maternal‐social spider, Coelotes terrestris (Araneae, Agelenidae). J. Arachnol. 14, 219–226. Kraus, O., and Kraus, M. (1988). The genus Stegodyphus (Arachnida, Araneae) sibling species, species groups, and parallel evolution of social living. Verh. Naturwiss. Ver. Hamburg 30, 151–254. Kraus, O., and Kraus, M. (1990). The genus Stegodyphus: Systematics, biogeography, and sociality (Araneida, Eresidae). Acta Zool. Fennica 190, 223–228. Kullmann, E. (1968). Soziale phaenomene bei spinnen. Ins. Soc. 15, 289–298. Kullmann, E., Sitterz, H., and Zimmermann, W. (1971). Erster Nachweis von Regurgitations‐ fuetterung bei einer cribellaten Spinne (Stegodyphus lineatus Latreille, 1871, Eresidae). Bonn. Zool. Beitr. 22, 175–188. Kullmann, E. J. (1972). Evolution of social behavior in spiders (Araneae, Eresidae and Theridiidae). Am. Zool. 12, 419–426. Lambin, X., Aars, J., and Piertney, S. B. (2001). Dispersal, intraspecific competition, kin competition and kin facilitation: A review of the empirical evidence. In ‘‘Dispersal’’ (J. Clobert, E. Danchin, A. A. Dhondt, and J. D. Nichols, Eds.), pp. 110–122. Oxford University Press, Oxford, UK. Leigh, E. G. (1999). Levels of selection, potential conflicts, and their resolution: The role of the ‘‘common good.’’ In ‘‘Levels of Selection in Evolution’’ (L. Keller, Ed.), pp. 15–30. Princeton University Press, Princeton. Levi, H. W., Lubin, Y. D., and Robinson, M. H. (1982). Two new species of from Papua New Guinea, with notes on other theridiid spiders (Araneae: Theridiidae). Pac. Insects 24, 105–114. Li, D., Jackson, R. R., and Barrion, A. (1999). Parental and predatory behaviour of Scytodes sp., an araneophagic spitting spider (Araneae: Scytodidae) from the Phillipines. J. Zool. (Lond.) 247, 293–310. Li, D., and Kuan, J. Y. X. (2006). Natal dispersal and breeding dispersal of a subsocial spitting spider Scytodes pallida (Araneae: Scytodidae), from Singapore. J. Zool. 268, 121–126. Lubin, Y. D. (1980). Population studies of two colonial orb‐weaving spiders. Zool. J. Linn. Soc. 70, 265–278. Lubin, Y. D. (1982). Does the social spider, Achaearanea wau (Theridiidae) feed its young? Z. Tierpsychol. 60, 127–134. Lubin, Y. D. (1986). Courtship and alternative mating tactics in a social spider. J. Arachnol. 14, 239–258. Lubin, Y. D. (1991). Patterns of variation in female‐biased colony sex ratios in a social spider. Biol. J. Linn. Soc. 43, 297–311. Lubin, Y. D. (1995). Is there division of labour in the social spider Achaearanea wau (Theridiidae)? Anim. Behav. 49, 1315–1323. Lubin, Y., and Crouch, T. (2003). Trial by fire: Social spider colony demographics in periodically burned grassland. Afr. Zool. 38, 145–151. Lubin, Y. D., and Crozier, R. H. (1985). Electrophoretic evidence for population differentiation in a social spider, Achaearanea wau Levi (Theridiidae). Ins. Soc. 32, 297–304. Lubin, Y., Hennicke, J., and Schneider, J. (1998). Natal philopatry and settling decisions of dispersing Stegodyphus lineatus (Eresidae) young. Isr. J. Zool. 44, 217–226. Lubin, Y. D., and Robinson, M. H. (1982). Dispersal by swarming in social spiders. Science 216, 319–321.

142

YAEL LUBIN AND TRINE BILDE

Main, B. Y. (1988). The biology of a social thomisid spider. In ‘‘Australian Arachnology’’ (A. D. Austin and N. W. Heather, Eds.), pp. 55–73. Misc. Publ. No. 5, The Australian Entomological Society, Brisbane. Maklakov, A. A., Bilde, T., and Lubin, Y. (2005). Sexual conflict in the wild: Elevated mating rate reduces female lifetime reproductive success. Am. Nat. 165, S38–S45. Marques, E. S. A., Vasconcellos‐Netto, J., and de Mello, M. B. (1998). Life history and social behavior of Anelosimus jabaquara and Anelosimus dubiosus (Araneae, Theridiidae). J. Arachnol. 26, 227–237. Maynard Smith, J. (1964). Group selection and kin selection. Nature 201, 1145–1147. Michod, R. E. (1982). The theory of kin selection. Annu. Rev. Ecol. Syst. 13, 23–55. Michod, R. E. (1993). Inbreeding and the evolution of social behavior. In ‘‘The Natural History of Inbreeding and Outbreeding’’ (N. W. Thornhill, Ed.), pp. 74–96. University of Chicago Press, Chicago. Myles, T. G. (1988). Resource inheritance in social evolution from termites to man. In ‘‘The Ecology of Social Behavior’’ (C. N. Slobodchikoff, Ed.), pp. 379–423. Academic Press, INC, San Diego, California. Nagelkerke, C. J., and Sabelis, M. W. (1996). Hierarchical levels of spatial structure and their consequences for the evolution of sex allocation in mites and other arthropods. Am. Nat. 148, 16–39. Nentwig, W., and Christenson, T. E. (1986). Natural history of the non‐solitary sheetweaving spider Anelosimus jucundus (Araneae: Theridiidae). Zool. J. Linn. Soc. 87, 27–35. Nunney, L. (1985). Female‐biased sex ratios: Individual or group selection. Evolution 39, 349–361. Okasha, S. (2004). Multilevel selection and the partitioning of covariance: A comparison of three approaches. Evolution 58, 486–494. Oster, G. F., and Wilson, E. O. (1978). ‘‘Caste and Ecology in the Social Insects.’’ Princeton University Press, Princeton, NJ. Packer, C., and Ruttan, L. (1988). The evolution of cooperative hunting. Am. Nat. 132, 159–198. Pasquet, A., and Krafft, B. (1989). Colony distrbution in the social spider Anelosimus eximius (Araneae, Theridiidae) in French Guiana. Ins. Soc. 36, 173–182. Pasquet, A., and Krafft, B. (1992). Cooperation and prey capture efficiency in a social spider, Anelosimus eximius (Araneae, Theridiidae). Ethology 90, 121–133. Pasquet, A., Trabalon, M., Bagne`res, A. G., and Leborgne, R. (1997). Does group closure exist in the social spider Anelosimus eximius? Behavioural and chemical approach. Ins. Soc. 44, 159–169. Peer, K., and Taborsky, M. (2005). Outbreeding depression, but no inbreeding depression in haplodiploid ambrosia beetles with regular sib mating. Evolution 59, 317–323. Perkins, T. A., Riechert, S. E., and Jones, T. C. (2007). Interactions between the social spider Anelosimus studiosus and foreign spiders that frequent its nest. J. Arachnol. 35, 146–155. Platnick, N. I. (2007). ‘‘The World Spider Catalog, Version 7.5.’’ Am. Mus. Nat. Hist. Available at http://research.amnh.org/entomology/spiders/catalog/index.html Pourie´, G., and Trabalon, M. (2001). Plasticity of agonistic behavior in relation to diet and contact signals in experimentally group‐living of Tegenaria atrica. Chemoecology 11, 175–181. Powers, K. S., and Avile´s, L. (2003). Natal dispersal patterns of a subsocial spider Anelosimus cf. jucundus (Theridiidae). Ethology 109, 725–737. Pulz, R. (1987). Thermal and water relations. In ‘‘Ecophysiology of Spiders’’ (W. Nentwig, Ed.), pp. 26–55. Springer Verlag, Berlin, Heidelberg, Germany.

THE EVOLUTION OF SOCIALITY IN SPIDERS

143

Pusey, A., and Wolf, M. (1996). Inbreeding avoidance in animals. Trends Ecol. Evol. 11, 201–206. Queller, D. C. (1992). Does population viscosity promote kin selection? Trends Ecol. Evol. 7, 322–324. Rannala, B. H., and Brown, C. R. (1994). Relatedness and conflict over optimal group size. Trends Ecol. Evol. 9, 117–119. Rayor, L. S., and Taylor, L. A. (2006). Social behavior in amblypygids, and a reassessment of arachnid social patterns. J. Arachnol. 34, 399–421. Reeve, H. K., and Keller, L. (1999). Levels of selection: Burying the units‐of‐selection debate, and unearthing the crucial new issues. In ‘‘Levels of Selection in Evolution’’ (L. Keller, Ed.), pp. 3–14. Princeton University Press, Princeton. Riechert, S. E. (1985). Why do some spiders cooperate? Agelena consociata, a case study. Fla. Entomol. 68, 105–116. Riechert, S. E., and Roeloffs, R. M. (1993). Evidence for and the consequences of inbreeding in cooperative spiders. In ‘‘The Natural History of Inbreeding and Outbreeding’’ (N. W. Thornhill, Ed.), pp. 283–303. University of Chicago Press, Chicago, Illinois. Riechert, S. E., Roeloffs, R. M., and Echternacht, A. C. (1986). The ecology of the cooperative spider Agelena consociata in equatorial Africa. J. Arachnol. 14, 175–191. Rissing, S. W., Pollock, G. B., Higgins, M. R., Hagen, R. H., and Smith, D. R. (1989). Foraging specialization without relatedness or dominance among co‐founding ant queens. Nature 338, 420–422. Roeloffs, R., and Riechert, S. E. (1988). Dispersal and population genetic structure of the cooperative spider, Agelena consociata, in west African rainforest. Evolution 42, 173–183. Roff, D. A. (2002). Inbreeding depression: Tests of the overdominance and partial dominance hypotheses. Evolution 56, 768–775. Roff, D. A. (1997). ‘‘Evolutionary Quantitative Genetics.’’ Chapman and Hall, New York. Rowell, D. M., and Avile´s, L. (1995). Sociality in a bark‐dwelling huntsman spider from Australia, Delena cancerides Walckenaer (Araneae: Sparassidae). Ins. Soc. 42, 287–302. Rowell, D. M., and Main, B. Y. (1992). Sex ratio in the social spider Diaea socialis (Araneae: Thomisidae). J. Arachnol. 20, 200–206. Ruttan, L. M. (1990). Experimental manipulations of dispersal in the subsocial spider, Theridion pictum. Behav. Ecol. Sociobiol. 27, 169–173. Rypstra, A. L. (1979). Foraging flocks of spiders: A study of aggregate behavior in Cyrtophora citricola Forskal (Araneae: Araneidae) in West Africa. Behav. Ecol. Sociobiol. 5, 291–300. Rypstra, A. L. (1993). Prey size, social competition, and the development of reproductive division of labor in social spider groups. Am. Nat. 142, 868–880. Rypstra, A. L., and Tirey, R. (1989). Observations on the social spider, Anelosimus domingo (Araneae, Theridiidae), in south‐western Peru. J. Arachnol. 17, 368–370. Saffre, F., Furey, R., Krafft, B., and Denouberg, J.‐L. (1999). Collective decision‐making in social spiders. Dragline‐mediated amplification process acts as recruitment mechanism. J. Theoret. Biol. 198, 507–517. Salomon, M., and Lubin, Y. (2007). Cooperative breeding increases reproductive success in the social spider Stegodyphus dumicola (Araneae, Eresidae). Behav. Ecol. Sociobiol. 61, 1743–1750. Salomon, M., Schneider, J. M., and Lubin, Y. (2005). Maternal investment in a spider with suicidal maternal care, Steodyphus lineatus (Araneae, Eresidae). Oikos 109, 61–622. Schneider, J. M. (1995). Survival and growth in groups of a subsocial spider (Stegodyphus lineatus). Ins. Soc. 42, 237–248. Schneider, J. M. (1996a). Differential mortality and relative maternal investment in different life stages in Stegodyphus lineatus (Araneae, Eresidae). J. Arachnol. 24, 148–154.

144

YAEL LUBIN AND TRINE BILDE

Schneider, J. M. (1996b). Food intake, growth and relatedness in the subsocial spider, Stegodyphus lineatus (Eresidae). Ethology 102, 386–396. Schneider, J. M. (1997). Timing of maturation and mating system of the spider, Stegodyphus lineatus (Eresidae): How important is body size? Biol. J. Linn. Soc. 60, 517–525. Schneider, J. M. (2002). Reproductive state and care giving in Stegodyphus (Araneae: Eresidae) and the implications for the evolution of sociality. Anim. Behav. 63, 649–658. Schneider, J. M., and Lubin, Y. (1998). Intersexual conflict in spiders (Special issue, ‘‘Costs of Reproduction,’’ H. Ylonen, Ed.). Oikos 83, 496–506. Schneider, J. M., Roos, J., Lubin, Y., and Henschel, J. R. (2001). Dispersal of Stegodyphus dumicola: They do balloon after all! J. Arachnol. 29, 114–116. Seibt, U., and Wickler, W. (1987). Gerontophagy versus cannibalism in the social spiders Stegodyphus mimosarum Pavesi and Stegodyphus dumicola Pocock. Anim. Behav. 35, 1903–1905. Seibt, U., and Wickler, W. (1988a). Bionomics and social structure of ‘family spiders’ of the genus Stegodyphus, with special reference to the African species S. dumicola and S. mimosarusm (Araneae, Eresidae). Verh. Naturwiss. Ver. Hamburg 30, 255–303. Seibt, U., and Wickler, W. (1988b). Why do ‘‘family spiders,’’ Stegodyphus (Eresidae) live in colonies? J. Arachnol. 16, 193–198. Seibt, U., and Wickler, W. (1990). The protective function of the compact silk nest of social Stegodyphus spiders (Araneae, Eresidae). Oecologia 82, 317–321. Sendova‐Franks, A. B., and Franks, N. R. (1998). Self‐assembly, self‐organization and division of labour. Proc. R. Soc. Lond. Ser. B 354, 1395–1405. Smith, D. R., and Engel, M. S. (1994). Population structure in an Indian cooperative spider, Stegodyphus sarasinorum Karsch (Eresidae). J. Arachnol. 22, 108–113. Smith, D. R., and Hagen, R. H. (1996). Population structure and interdemic selection in the cooperative spider Anelosimus eximius (Araneae: Theridiidae). J. Evol. Biol. 9, 589–608. Stearns, S. C. (1992). Trade‐offs. In ‘‘The Evolution of Life Histories’’ (S. C. Stearns, Ed.), pp. 72–90. Oxford University Press, New York. Suter, R. B. (1991). Ballooning in spiders: Results of wind tunnel experiments. Ethol. Ecol. Evol. 3, 13–25. Trivers, R. L. (1971). The evolution of reciprocal altruism. Quart. Rev. Biol. 46, 35–57. Uetz, G. W., and Hieber, C. S. (1997). Colonial web‐building spiders: Balancing the costs and benefits of group living. In ‘‘The Evolution of Social Behavior in Insects and Arachnids’’ (J. Choe and B. Crespi, Eds.), pp. 458–475. Cambridge University Press, Cambridge, England. Uetz, G. W., Kane, T. C., Stratton, G. E., and Benton, M. J. (1987). Environmental and genetic influences on the social grouping tendency of a communal spider. In ‘‘Evolutionary Genetics of Invertebrete Behavior’’ (M. D. Huettel, Ed.), Plenum Publ. Corp., New York. Ulbrich, K., Henschel, J. R., Jeltsch, F., and Wissel, C. (1996). Modelling individual variability in a social spider colony (Stegodyphus dumicola: Eresidae) in relation to food abundance. Rev. Suisse Zool. vol. hors se´rie, 661–670. Ulbrich, K., and Henschel, J. R. (1999). Intraspecific competition in a social spider. Ecol. Model. 115, 243–251. Vakanas, G., and Krafft, B. (2001). Coordination of behavioral sequences between individuals during prey capture in a social spider, Anelosimus eximius. J. Ins. Behav. 14, 777–798. Viera, C., Ghione, S., and Costa, F. G. (2005). Regurgitation among penultimate juveniles in the subsocial Anelosimus cf. studiosus (Theridiidae): Are males favored? J. Arachnol. 34, 258–260.

THE EVOLUTION OF SOCIALITY IN SPIDERS

145

Viera, C., Ghione, S., and Costa, F. G. (2007a). Me´chanisms underlying eggsac opening in the subsocial spider. Anelosimus cf. studiosus (Araneae, Theridiidae). Ethol. Ecol. Evol. 19, 61–67. Viera, C., Ghione, S., and Benamu´‐Pino, M. A. (2007b). Progeny, development and phenology of the sub‐social spider Anelosimus cf. studiosus (Araneae, Theridiidae) from Uruguay. Neotropical Fauna and Environment 42(2), 145–153. Vollrath, F. (1982). Colony foundation in a social spider. Z. Tierpsychol. 60, 313–324. Vollrath, F. (1986). Environment, reproduction and the sex ratio of the social spider Anelosimus eximius (Araneae, Theridiidae). J. Arachnol. 14, 267–281. Vollrath, F., and Rohde‐Arndt, D. (1983). Prey capture and feeding in the social spider Anelosimus eximius. Z. Tierpsychol. 61, 334–340. Vollrath, F., and Windsor, D. (1983). Subsocial and social Anelosimus: A comparison especially of nest defence. In ‘‘Proceedings of the Ninth International Congress of Arachnology’’ (W. G. Eberhard, Y. D. Lubin, and B. C. Robinson, Eds.), pp. 295–298. Smithsonian Institution Press, Washington, DC. Wade, M. J., and Breden, F. (1981). Effect of inbreeding in the evolution of altruistic behaviour by kin selection. Evolution 35, 844–858. Ward, P. I., and Enders, M. M. (1985). Conflict and cooperation in the group feeding of the social spider S. mimosarum. Behaviour 97, 213–225. Waser, P. W., Austad, S. N., and Keane, B. (1986). When should animals tolerate inbreeding? Am. Nat. 128, 529–537. Waser, P. W., Creel, S. R., and Lucas, J. R. (1994). Death and disappearance: Estimating mortality risks associated with philopatry and dispersal. Behav. Ecol. 5, 135–141. Wcislo, W. T. (2005). Social labels: We should emphasize biology over terminology and not vice versa. Ann. Zool. Fennici 42, 565–568. Werren, J. H. (1983). Sex ratio evolution under local mate competition in a parasitic wasp. Evolution 37, 116–124. West, S. A., Griffin, A. S., and Gardner, A. (2007). Social semantics: Altruism, cooperation, mutualism, strong reciprocity and group selection. J. Evol. Biol. 20, 415–432. West Eberhard, M. J. (1975). The evolution of social behavior by kin selection. Quart. Rev. Biol. 50, 1–33. White, M. J. D. (1973). ‘‘Animal Cytology and Evolution.’’ Cambridge University Press, Cambridge. Whitehouse, M. A. E., and Lubin, Y. (1999). Strategic interference competition by individuals in social spider foraging groups. Anim. Behav. 58, 677–688. Whitehouse, M. E. A., and Lubin, Y. (2005). The functions of societies and the evolution of group living: Spider societies as a test case. Biol. Rev. 80, 1–15. Wickler, W., and Seibt, U. (1986). Aerial dispersal by ballooning in adult Stegodyphus mimosarum. Naturwissenschaften 73, 628–629. Wickler, W., and Seibt, U. (1993). Pedogenetic sociogenesis via the ‘‘Sibling‐route’’ and some consequenses for Stegodyphus spiders. Ethology 95, 1–18. Willey, M. B., and Jackson, R. R. (1993). Predatory behavior of a social spider, Stegodyphus sarasinorum (Araneae: Eresidae): Why attack first? Can. J. Zool. 71, 2220–2223. Wilson, D. S., and Colwell, R. K. (1981). The evolution of sex ratio in structured demes. Evolution 35, 882–897. Wilson, E. O. (1971). ‘‘The Insect Societies.’’ Harvard University Press, Cambridge, MA. Williams, G. C. (1966). ‘‘Adaptation and Natural Selection.’’ Princeton University Press, Princeton, New Jersey. Wrensch, D. L., and Ebbert, M. A. (1993). ‘‘Evolution and Diversity of Sex Ratio in Insects and Mites.’’ Chapman and Hall, New York and London.

This page intentionally left blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

Molecular Ecology Reveals the Hidden Complexities of the Seychelles Warbler Jan Komdeur* and David S. Richardson{ *animal ecology group, centre for ecological and evolutionary studies, university of groningen 9750 AA haren, the netherlands { centre for ecology, evolution and conservation school of biological sciences, university of east anglia norwich nr4 7tj, united kingdom

I. INTRODUCTION Animals rarely act in isolation. The vast majority of animals live within a social environment, their lives affected by the presence and activities of the other individuals around them. For example, foraging, mating, rearing young, predator defense, and practically every aspect of an individual’s behavior will be influenced in some way by others around them. Many of us are genuinely fascinated by the wide spectrum of social interactions we see across the diversity of animal life. This may be because, as Trivers (1985) puts it ‘‘everybody has a social life,’’ and much of the interest probably emerges from comparisons with our own social situation. On the other hand, our fascination may be in understanding the widespread existence of social behaviors between individuals despite the apparent costs of such behavior. These social interactions, which can be in the form of cooperation or conflict, can occur over a wide range of strategies and at various levels. Individuals cooperate in hunting, food sharing, fending off enemies, and migrating from one site to another. Many animals live and breed in colonies, and males and females interact when mating and/or caring for offspring (Dugatkin, 1997; Wilson, 1975). Unlike many traits that are directly selected by the environment, social behaviors evolve in response to selection pressures created by interactions with other members of their own species. Consequently, these social behaviors are fascinating to study from an 0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37004-6

147

Copyright 2007. Elsevier Inc. All rights reserved.

148

JAN KOMDEUR AND DAVID S. RICHARDSON

evolutionary perspective. In this chapter, we are primarily concerned with how social interactions between individuals evolve as part of reproductive strategies. In the last decades, theoretical and empirical studies on cooperation and conflict over reproductive allocation have rapidly proliferated. This area is emerging as one of central importance in behavioral ecology (Barnard, 2004; Krebs and Davies, 1997). This increased interest in the patterns of reproductive allocation has, at least partially, been stimulated by advances in molecular biology. Developments since the early 1980s have resulted in range of techniques with which to investigate the molecular properties of individuals, populations, and species (reviewed in Beebee and Rowe, 2004). These relatively cheap, easy, and efficient techniques have increasingly been taken up by behavioral scientists keen to incorporate new, incisive measures within their studies. Such techniques have enabled patterns of genetic relatedness between individual to be determined (Jeffreys et al., 1985) and have allowed rapid amplification of large quantities of DNA from minute traces [using polymerase chain reaction (PCR); Mullis and Faloona, 1987; Saiki et al., 1985, 1988]. PCR‐based techniques, such as microsatellite fingerprinting, only require tiny blood or tissue samples from study organisms, thus allowing for the nondestructive study of small or rare organisms. Such molecular techniques have provided the tools to investigate a range of issues in behavioral and evolutionary ecology. So, for example, the ability to assign the genetic parents of young has revealed that genetic monogamy is relatively rare, even in taxa where social monogamy is the most common mating system. For example, genetic monogamy is found in only 14% of social monogamous passerine species studied so far (Griffith et al., 2002). Once parentage could be determined accurately, the question as to whether mate choice can increase offspring fitness through genetic benefits could also be investigated properly. Molecular techniques also now allow for the identification of differences in the underlying genes that may account for variation, both in the trait on which choice is made and in fitness variation [major histocompatibility complex (MHC) genes], which play a major role in determining the vertebrate immune system and therefore, through interactions with pathogens, have an important effect on an individual’s fitness (reviewed in Hughes and Yeager, 1998). Amplifying sections of DNA from the sex chromosomes that differ in length between the genders means that the sex of organisms that were previously difficult to sex, for example, newly hatched nestling birds can be determined (Griffiths et al., 1998). Such reliable sexing has facilitated research into concepts such as adaptive sex ratio control and sex allocation, for example, whether females adjust the sex ratio of their offspring in response to the sexual attractiveness of their mate. In addition to determining parentage, the comparisons of individual

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

149

genotypes also allow other patterns of relatedness between individuals to be estimated. This has greatly facilitated research into the evolution of cooperative breeding, since determining relatedness between helpers and those that are helped is an important prerequisite to determining kin‐selected benefits. In this chapter, we show how we have applied molecular tools to investigate and explore the patterns and consequences of social interactions between individuals in the Seychelles warbler, Acrocephalus sechellensis. The aim here is to review how we have investigated important evolutionary questions, such as the evolution of cooperative breeding, inbreeding avoidance and mate choice using this species as a model system. This chapter outlines recent observational and experimental findings, and elaborates on some of the topics published in a previous review (Komdeur, 2003). The Seychelles warbler population has many unique attributes that make it ideally suited as a study system in which to examine aspects of evolutionary ecology. First, as the original remnant population of this species was entirely confined to Cousin Island (29 ha; 04
200 S, 55
400 E), we have been able to study the entire world population. Since 1985, the Cousin Island population has been studied in intimate detail. From 1997 nearly all adults and young have had their blood sampled each year for molecular analyses. Off‐island migration by warblers is negligible (only 2 individuals over 20 years; Komdeur et al., 2004a), so any birds that disappeared could reliably be considered dead. As the population is a closed system, and all breeding attempts were monitored, the complete life history, status, and putative pedigree of nearly all birds are known. In such a system, accurate short‐ and long‐term fitness parameters can be calculated without the problems encountered by most other studies, where survival and reproductive success estimates are confounded by dispersal. This allows for the precise calculation of the total lifetime reproductive success of each individual and for quantification of how current strategies translate into future inclusive fitness. Although our chapter is focused on Seychelles warblers, we will discuss the broader relevance of our findings. We will also draw more general conclusions where appropriate and outline some of the unresolved issues within this field. Finally we will suggest future research objectives.

II. STUDY SPECIES, STUDY POPULATIONS, AND GENERAL METHODS The Seychelles warbler is a small (15–16 g) insectivorous bird which takes insect prey from leaves. The warbler is a rare endemic that is still confined to a few small islands within the Seychelles archipelago, its current IUCN

150

JAN KOMDEUR AND DAVID S. RICHARDSON

red list status is vulnerable (IUCN, 2004). It is closely related to the migratory great reed warbler Acrocephalus arundinaceus and Australian reed warbler A. australis (Leisler et al., 1997). The Seychelles warbler is assumed to have originally occurred on most of the Seychelles islands, which constituted a single large island during the last ice age (Collar and Stuart, 1985). By 1940, anthropogenic disturbance had pushed this species to the verge of extinction and less than 29 individuals remained on the island of Cousin (Crook, 1960; Loustau‐Lalanne, 1968; Vesey‐Fitzgerald, 1940). Cousin was purchased in 1968 for Bird Life International with the aim of saving the warbler. From 1968 onward coconut trees were cleared and the native vegetation was allowed to regenerate from the tiny remnants. By 1982 most of Cousin was covered with tropical forest (Bathe and Bathe, 1982). The population has recovered, and since 1981 has stabilized at around 320 adult individuals (Brouwer et al., 2006; Komdeur, 1992; Komdeur and Pels, 2005; Richardson et al., 2001). The number of territories reached its saturation level of ca. 115 territories in 1978 (Diamond, 1980). In each of the years from 1978 the island (apart from the bare hill) has been completely covered by territories with no empty spaces and there has been a surplus of nonterritorial adult birds (Komdeur, 1992). Seychelles warblers do not differ from other closely related migratory Acrocephalus warblers in the anatomy of the flight apparatus (Komdeur et al., 2004a). Despite the potential for sustained flight and the overproduction of adult warblers on Cousin in the last decades, inter‐island dispersal has been extremely rare. During the 17 years of study only 2 warblers (0.13%, n ¼ 1599) have crossed the 1.6 km stretch of sea between the islands of Cousin and Cousine unaided (Komdeur et al., 2004a). Given the vulnerability of a single isolated population, and the extremely small chance of successful establishment on unoccupied islands by unaided warblers, it was decided to establish new populations through translocations. New populations were established by introducing 29 individuals to each of the islands of Aride (68 ha) and Cousine (26 ha) in 1988 and 1990, respectively (Komdeur, 1994a), and 58 individuals to the island of Denis (140 ha) in 2004 (L. Brouwer, J. Komdeur, and D. S. Richardson, unpublished data). During the years following the translocation, with higher food availability per territory, warblers on the new islands had significantly higher reproductive success than birds remaining on Cousin. The Cousine population had increased to 130 individuals in 1996 and has stabilized at this level, the Aride population grew to ca. 1850 individuals by 2003, and the Denis population to 82 individuals in 2006 (L. Brouwer, J. Komdeur, and D. S. Richardson, unpublished data). The conservation status of the warbler has, consequently, improved from endangered to vulnerable (IUCN, 2004).

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

151

Like many other tropical bird species, the socially monogamous Seychelles warbler is long lived. Adult birds have high annual survival (84%; Brouwer et al., 2006; Komdeur, 1996a), and average life expectancy after reaching 1 year of age is 5.5 years (maximum observed life span is 21 years). Birds normally inhabit year‐round stable territories and are paired with the same mate, sometimes for as long as 9 years (Komdeur, 1992). On Cousin Island, food supplies and reproduction are highly seasonal with peaks of nest building occurring during 1 month per year from May to August. In all years there is also a smaller nesting peak from December to February, separated from the main season by a distinct gap (Komdeur, 1996a). The warblers time their nest building such that the presence of nestlings is synchronized with the peak abundance of food (Komdeur, 1996a). On Cousin Island, the warbler usually produces one clutch per season and this normally consists of just one egg, but about 20% of nests contain two or three eggs (Komdeur, 1996a; Richardson et al., 2001). Females incubate alone for about 17–19 days. Nestlings are fed in the nest for 18–20 days prior to fledging, and then for at least 6 months (and sometimes up to a year) before reaching independence (Komdeur, 1996a). This is an extremely long period of dependence for a passerine species; no other passerine has a typical fledgling period as long as 6 months, indeed the mean fledging periods for other passerine birds is only about one tenth that of the Seychelles warbler (Bennett and Owens, 2002). Although warblers can breed independently in their first year, a lack of suitable independent breeding opportunities drives some sexually mature individuals into becoming subordinates within their natal territory (Komdeur, 1992; Richardson et al., 2003a). Though not all territories support subordinates; indeed in any year only 30% of territories do (Richardson et al., 2002). Additional adult birds in the territory do not always assist in the groups’ reproductive attempt (i.e., they do not nest guard, incubate eggs, or provision chicks) hence we use the term ‘‘subordinate’’ for any additional adult (including helpers or nonhelpers) rather than ‘‘helper’’ (Richardson et al., 2002). The Seychelles warbler is atypical among cooperative breeding birds in that females are more likely to become subordinates than males. The observed percentage of subordinates that are female is 68% (Richardson et al., 2002). Females are nearly always to become subordinates on their natal territory, whereas a significant percentage of males (25%) become subordinates on nonnatal territories (Richardson et al., 2002). This results in lower levels of relatedness between male subordinates and group nestlings. Many, but not all, of these subordinates play an active role in the groups reproductive attempts (58%, Komdeur et al., 2004b; 68%, Richardson et al., 2002) by helping with nest building (mainly females), incubation (females only), nest guarding (mainly males), and feeding of

152

JAN KOMDEUR AND DAVID S. RICHARDSON

dependent young (Komdeur, 1994b). These individuals are designated as helpers and are more likely to be female than male (88% vs 12%, Komdeur, 1996b). The Cousin Island population of Seychelles warblers has been monitored since 1981, intensively so between 1985 and 2006. During this time, many of the birds (>96% of birds since 1997) have been individually color‐ringed using a unique combination of two color rings on one leg and one color ring and an individually numbered British Trust for Ornithology ring on the other leg. During the breeding seasons (May to August, December to February), all territorial groups of warblers (ca. 115) were checked for the presence of color‐ringed birds and nesting activity, at least once every 2 weeks, by following the resident female for 15 min continuously (the female alone is involved in nest building). This observation period was long enough to determine whether birds were nest building (Komdeur, 1996a). Once the nests were found, they were observed throughout the breeding cycle. Each fully built nest was checked regularly for the presence of a clutch or a nestling, either with help of a long stick and angled mirror or by observing the birds’ behavior. Observations on incubation and provisioning were made for all the breeding attempts. Observation periods typically lasted 1 h. Whether or not each nestling fledged successfully was recorded. Behavioral observations were used to determine the status of all birds within each territory. The primary male and female were defined as the dominant, pair‐bonded male and female in a territory, and the term ‘‘subordinate’’ included all other bird (>8 months old) resident in the territory. We traced all individually marked birds and monitored their breeding activity throughout their lives. Between 1993 and 1996 as many nestlings and adults as possible were blood‐sampled during the main breeding season (May to August) each year. Moreover, since 1997 until the present day, almost the entire adult population (>97%) and all the young have been blood sampled each year. Blood samples (about 15 ml) were taken by brachial venipuncture and diluted in 800 ml of 100% ethanol in a 2.0‐ml screw‐cap microfuge tube and stored at room temperature. Dead embryos were extracted from eggs that failed to hatch and stored in 100% ethanol. From 1995 onward we have attempted to sample all offspring produced in each breeding season. This allowed for the precise calculation of the total number of young produced over the entire lifetime of helpers and both members of the breeding pair. As the warblers are insectivorous, the quality of each territory depends on the abundance of insect prey available, amount of foliage and territory size. The quality of each territory was determined in each main breeding season (June to September) by measuring these variables and calculating territory quality. Territories have been classified

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

153

into three territory quality categories: low‐, medium‐, or high‐quality (for further details of territory quality measurements, see Komdeur, 1992, 1994b).

III. COOPERATIVE BREEDING The widespread occurrence of apparently altruistic cooperative behavior among animals is paradoxical for evolutionary biologists as, since Darwin (1859), the general expectation is that individual behavior will be selfish in order to promote individual fitness (Dawkins, 1976; Maynard Smith and Szathma´ry, 1995). Therefore, evolutionary theory predicts that animals will evolve strategies that maximize their fitness (Maynard Smith, 1978). A classic example of cooperation which has proven hard to explain comes from cooperative breeding. In some species individuals live and breed in bisexual groups of three or more adults and share parental care at a single nest (Brown, 1987; Stacey and Koenig, 1990). Some of these mature individuals are subordinates that do not breed independently but instead provide care to young that are not their own genetic offspring. Such helpers usually assist by feeding young, but also with defense of the young or the territory (Brown, 1987; Dickinson and Hatchwell, 2004; Emlen, 1997; Russel, 2004; Stacey and Koenig, 1990). Sometimes individuals spend all of their lives helping others to reproduce. Typically, such cooperative breeding systems comprise family groups that live together on permanent, stable territories. Among vertebrates, cooperative breeding is found in at least 3% of bird and mammal species (Arnold and Owens, 1998; Brown, 1987; Sibley and Monroe, 1990) and in some fish species (Taborsky, 1994). There is a particularly high frequency (19%) of cooperative breeding in oscine passerine species (Cockburn, 2003), in Australian birds (Arnold and Owens, 1998; Russell, 1989), and in primate species (Kappeler, 2007). Although the existence of helpers within cooperative breeding systems has been known for many years (Skutch, 1935), it was not until Hamilton (1964a,b) and Maynard Smith (1964) developed the theory now referred to as ‘‘kin selection,’’ that there was a firm foundation for the empirical study of cooperative breeding. These authors argued that the fitness of each individual is determined by the total number of genes, identical by descent to its own, that are present in subsequent generations. Consequently, helpers can increase their fitness indirectly through enhancing the reproductive success of close relatives. In most species the indirect genetic fitness benefit gain from helping is likely to be considerably less than the potential direct genetic gain from immediate independent breeding, if a territory and

154

JAN KOMDEUR AND DAVID S. RICHARDSON

mate could be obtained (Brown, 1987). Therefore, understanding why mature individuals delay dispersal is the key to understanding the evolution of cooperative breeding. As such, the evolution of the traditional vertebrate cooperative breeding system is usually viewed as a two‐step process. First, the decision by grown offspring to delay dispersal and independent breeding by staying in the natal unit, and second, the decision by those individuals who have stayed at home to become helpers (Emlen, 1982). The first step is usually attributed to the existence of ecological constraints, such as a shortage of breeding territories or mates, that prevent offspring from becoming independent breeders (‘‘ecological constraints’’ hypothesis; Emlen, 1982, 1991; reviewed in Arnold and Owens, 1998, 1999; Brown, 1987; Hatchwell and Komdeur, 2000; Koenig et al., 1992). The second step envisages that individuals that have already delayed dispersal can gain a net fitness benefit by becoming subordinates in a group (‘‘benefits of philopatry’’ hypothesis; Stacey and Ligon, 1987, 1991). Originally the ‘‘benefits of philopatry’’ hypothesis was proposed as an alternative explanation for the ‘‘ecological constraints’’ hypothesis for delayed dispersal (Stacey and Ligon, 1987, 1991). However, the distinction between these two hypotheses has been much debated (Emlen, 1991; Heinsohn et al., 1990; Koenig et al., 1992; Walters et al., 1992; Zack, 1990). One of the main reasons for initiating a study on Seychelles warblers was to unravel the relative roles of the ‘‘ecological constraints’’ and the ‘‘benefits of philopatry’’ hypotheses in the evolution of cooperative breeding. Although warblers can breed independently in their first year, some individuals remain in their natal territories as subordinates, and often help provision nondescendent offspring. In the Seychelles warbler, there are two lines of evidence against the ‘‘ecological constraints’’ hypothesis being the sole reason for cooperatively breeding. First, territory quality has a significant effect on dispersal: vacancies arising on territories are mostly filled by subordinates from territories of the same or lower quality (Komdeur, 1992). Second, individuals that delay reproduction in high‐ quality territories and later breed there have, on average, greater lifetime fitness than those that disperse and breed immediately in lower quality territories. On average 0.85 fledglings were produced each year in the high‐quality territories, but only 0.19 in the poor areas. If a young bird in a good territory stayed to help rather than going off to build its own nest, the parents’ output rose from 0.85 to 1.62 fledglings per year. In other words, subordinate helpers really helped, nearly doubling production (an extra 0.77 fledglings per year). This increase is about four times the output from the poor areas (0.19 fledglings per year), the most likely breeding site for these young if they had become independent. Subordinate helpers therefore increase their own fitness by staying home because they are

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

155

helping to propagate copies of their own genes (Komdeur, 1992). These results support the ‘‘benefits of philopatry’’ hypothesis, which emphasizes the lifetime inclusive fitness benefits from staying at home. The transfer of warblers to the unoccupied islands of Aride and Cousine allowed for a direct experimental test of this hypothesis. Directly after translocation there was no cooperative breeding, but as all high‐quality areas became occupied, young birds born on high‐quality territories began to stay as helpers rather than occupying breeding vacancies on low‐quality territories. Therefore habitat saturation and territory quality both appear to be involved in the evolution of cooperative breeding (Komdeur, 1992; Komdeur et al., 1995). These results provided strong experimental confirmation of the two previously debated hypotheses. As stated by Mumme (1992a), the Seychelles warbler results showed that the ‘‘benefits of philopatry’’ hypothesis can be accommodated successfully within the ‘‘ecological constraints’’ hypothesis by recognizing that they differ only in the emphasis they place on either the costs of leaving or the benefits of staying. Both hypotheses are based on the assumption that a cost–benefit analysis of leaving versus staying is resolved in favor of the latter option (Emlen, 1994, 1997; Koenig et al., 1992, Mumme, 1992a; but see Ligon, 1999). For the fitness calculations in the above study on the Seychelles warbler relatedness was estimated from genealogical data. Because subordinates normally remain on their natal territory and assist the breeding pair with raising the subsequent offspring (Komdeur, 1992), we assumed that subordinates were first order relatives (related by 0.5) of the breeding pair and the offspring they help to raise. However, this observational approach can be inadequate because complex patterns of shared reproduction and extra‐ group paternity may be overlooked. Without the ability to assess genetic relatedness and parentage we also assumed that the dominant male and female were the genetic parents of the young they reared, and that no reproduction was achieved by the subordinate within the group. In the next section, we describe the outcome of molecular investigations into relatedness and parentage analyses in the Seychelles warbler, and how these were used to address questions relating to the evolution of cooperative breeding. It was thought that subordinates base their helping decisions on the presence of their parents within the territory. Subordinates were more likely to help feeding nestlings when both parents were still present in the territory than when one or both parents have been replaced (Komdeur, 1994c). However, the efficiency of these cues in maximizing indirect fitness benefits had to be tested using parentage and molecular‐based coefficients of relatedness.

156

JAN KOMDEUR AND DAVID S. RICHARDSON

A. INDIRECT BENEFITS Indirect benefits have been suggested to be a major selective force behind the evolution of cooperative breeding (Emlen, 1997; Emlen and Wrege, 1988; Foster et al, 2006; Hamilton, 1964a,b; Koenig et al., 1992; Maynard Smith, 1964; Mumme 1992b; Mumme et al., 1989; Wilson, 2005; Wilson and Holldobler, 2005). These benefits accrue if helping obeys two requirements. First, helping should result in the improved reproductive success of the breeding pair. Many studies of vertebrate cooperative breeding systems have shown that helpers do enhance breeder fitness (reviewed in Cockburn, 1998; Emlen, 1991, 1997). In the Seychelles warbler, we demonstrated experimentally that helping behavior significantly improved the reproductive success of the breeding pair. The removal of subordinate helpers resulted in lower reproductive success for the breeding pair compared with control pairs for which the subordinates had not been removed (Komdeur, 1994b). Second, subordinates have to help related individuals. Studies on subordinate investment in species where relatedness between subordinates and nestlings has been estimated using genealogical data have provided mixed evidence for kin preferences (Emlen, 1997; Griffin and West, 2003; Komdeur and Hatchwell, 1999; Komdeur et al., 2007). Some studies have shown no relationship between the level of investment by subordinates and kinship (Clutton‐Brock et al., 2000, 2001; Cockburn, 1998). Others have shown facultative adjustment of helping by subordinates toward close kin (Arnold, 1990; Clarke, 1984; Curry, 1988; Emlen and Wrege, 1988; Komdeur, 1994c; Reyer, 1984; Russel, 1999). However, in many situations where close genetic relationships were thought to exist (i.e., subordinates helping their ‘‘social’’ parents), shared reproduction by primary and subordinates, and/or extra‐group paternity only revealed by the use of molecular techniques, have resulted in lower levels of within‐ group relatedness than originally predicted (Brooker et al., 1990; Dickinson and Akre, 1998; Mulder et al., 1994). The recent characterization of a large number of microsatellite markers for the Seychelles warbler (Richardson et al., 2000) provided us with the tools to genotype individuals and develop a protocol to assign parentage in this species (Richardson et al., 2001). We also used the software program KINSHIP (Goodnight and Queller, 1999) to calculate individual pairwise relatedness values based on microsatellite genotype similarity (Richardson et al., 2000). Using these protocols, we revealed two features of the warbler’s ecology that do not allow an accurate assessment of kinship based on social pedigrees. (1) Shared maternity. Both the dominant and subordinate females lay eggs within the nest. The incidence of female subordinate producing offspring is high in the Seychelles warbler; each year 44% of female

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

157

subordinates gained parentage within their own group, and 15% of all offspring produced on the island were attributed to subordinate females (Richardson et al., 2001). Joint nesting in female warblers is not the result of conspecific brood parasitism, in which a female from outside the territory lays eggs in a conspecific female’s nest and leaves without providing parental care (Andersson, 1984; Petrie and Møller, 1991; Yom‐Tov, 1980). Egg dumping between groups does not occur (Richardson et al., 2002). None of the male subordinates gained paternity within the group (Richardson et al., 2001). These findings are important for two reasons: (i) being able to produce offspring means that the direct benefits of being a subordinate female are much higher than previously thought and, importantly, considerably higher than the indirect benefits (see below); (ii) as all the offspring in the territory are not produced by the same female relatedness between the subordinate and nestlings (that she herself did not produce), may be considerably lower than previously thought. Basically, the social female in the territory will often not be the mother of both the subordinate helper and the new nestlings. Lower levels of relatedness will reduce the potential indirect benefits of helping such nestlings (Richardson et al., 2002). (2) Extra‐group paternity. Extra‐group paternity is relatively high in the Seychelles warbler, responsible for 40% of offspring (Richardson et al., 2001). This dramatically reduces levels of relatedness between subordinates and the offspring they may be helping. The chance that the same male fathered both the subordinate and the nestling produced at a later stage is low. These two findings, that maternity is shared and the extra‐group paternity is high, suggest that levels of relatedness originally predicted from the genealogical data will be very misleading. Furthermore, they highlight how important kin discrimination may be if individuals are to gain significant indirect benefits through helping in such systems. B. KIN DISCRIMINATION BY SUBORDINATES Those individuals that can discriminate between kin and nonkin and preferentially help only their kin will be able to maximize indirect benefits and, therefore, gain an evolutionary advantage. Using coefficients of genetic relatedness between individuals in the Seychelles warbler, we were able to test whether subordinates preferentially help more closely related nestlings. Investment in provisioning varied greatly between the different members of the cooperative groups. Female subordinates that gain parentage appear to accurately assess this fact and always provisioned. However, for female subordinates that did not gain parentage, provisioning was positively correlated to their relatedness to the dependent nestlings. This was not the case for rarer male subordinates, who also provisioned less in general

158

JAN KOMDEUR AND DAVID S. RICHARDSON

A

B Subordinate-nestling relatedness

12

Feeds per hour

10 8 6 4 2

n= 0.4

8

p = 0.002

5

0.2 0.0 −0.2 −0.4

0 −0.4

−0.2

0.0 0.2 0.4 Subordinate-nestling relatedness

Helpers Nonhelpers

FIG. 1. (A) Provisioning by subordinate female (filled triangles) and male (empty squares) nonparent Seychelles warblers in relation to subordinate‐nestling relatedness. There was a positive relationship between provisioning and subordinate–nestling relatedness for female subordinates (solid line; F1,11 ¼ 4.82, p ¼ 0.05, r2 ¼ 0.31) but not for male subordinates (broken line; F1,8 ¼ 0.04, p ¼ 0.86, r2 ¼ 0.00). (B) Subordinates–nestlings relatedness for female nonparent subordinates that helped provision compared to those that did not (helpers vs nonhelpers: 0.27  0.16, n ¼ 8 vs 0.05  0.20, n ¼ 5, t11 ¼ 3.13, p ¼ 0.01). Bars indicate means  S.E. (Adapted from Richardson et al., 2003b.)

(Fig. 1A; Richardson et al., 2003a). Compared (using one sample t‐tests) to the zero background level of relatedness between individuals within the population (set in KINSHIP; Goodnight & Queller, 1999), female subordinates that helped were significantly related to nestlings (mean r ¼ 0.27), whereas female nonhelpers were not (mean r ¼ 0.05; Fig. 1B). So nonparent female subordinates do appear to adjust their provisioning effort according to their kinship to the nestling. For male subordinates neither helpers nor nonhelpers were significantly related to nestlings (mean r ¼ 0.01 and 0.19, respectively), nor was there a significant difference in the relatedness of the two groups (Fig. 1B; Richardson et al., 2003a). For indirect benefits to accrue, there must not only be relatedness between the donor (i.e., the subordinate) and the recipient (i.e., the nestlings), but also benefits to the recipient (Hamilton, 1964a,b). In Seychelles warblers, the amount of food brought to the nestling determines fledging success and first year survival (Komdeur, 1991), and the presence of helpers increases the number of young fledged on a territory (Komdeur, 1994b). During the study for which genetic relatedness was estimated, the number of fledglings produced on a territory increased significantly (by 17%) for each subordinate present, even after excluding direct parentage

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

159

(Richardson et al., 2002). Therefore, the preferential provisioning of related young by female subordinates does increase the number of related offspring produced and so helps maximize the indirect benefits of cooperative breeding. On the other hand, male subordinates, who provision less than females, and who are not significantly related to the nestlings they do provision, do not appear to help in order to gain indirect fitness benefits. C. KIN SELECTION CUES While it is true that animal altruists tend to be related to the individuals they help, whether this is the result of active kin selection or merely a result of passive processes, such as delaying dispersal and remaining with the natal territory or group, has long been a matter of debate (Griffin and West, 2003; Komdeur and Hatchwell, 1999). There is evidence that some animals use sophisticated mechanisms (e.g., plumage characteristics, Petrie et al., 1999; Shorey et al., 1994; vocalizations, Sharp et al., 2005; or odors, Mateo, 2006; Mateo and Johnston, 2000) to discriminate between related and unrelated individuals. However, it could be that environmental or ecological factors predict relatedness so well that simple behavioral rules could offer reliable and effective means to discriminate between kin and nonkin. Typical cooperative breeding systems, where helpers assist with raising subsequent broods on their natal territory (Brown, 1987), are characterized by an extended period of association with kin, stable and strong natal philopatry, limited access to breeding sites, and the presence of long‐lived individuals (Arnold and Owens, 1998; Emlen, 1991; Hatchwell and Komdeur, 2000; Stacey and Koenig, 1990). If discrimination rules are learnt during development then periods of exclusive and prolonged association with relatives may facilitate such learning. In most cases, kin‐directed helping precedes dispersal and independent breeding. In such situations, a decision rule ‘‘care for any offspring in my natal territory’’ could serve as a reliable discriminator between kin and nonkin. Furthermore, the long period of extended care may allow the helper to develop a simple rule of thumb such as ‘‘help anyone who fed me as a nestling,’’ which may also be a good predictor of relatedness. Indeed, current evidence does suggest that learning through association is the most likely mechanism of kin discrimination in vertebrate societies (Komdeur et al., 2007). However, other behavioral, ecological, or life history traits may act to reduce the effectiveness or reliability of such acquired ‘‘discrimination through association’’ rules. First, the longer a helper remains on its natal territory the higher the probability that there will be some turnover of breeders and, consequently, a reduction in its relatedness to the helped brood. However, in theory, as long as the putative period in which discrimination cues are learnt is

160

JAN KOMDEUR AND DAVID S. RICHARDSON

relatively early on in development before any breeder turnover occurs, discrimination through association can avoid this problem. Second, shared parentage and extra‐pair paternity (EPP) within broods occur in many cooperatively breeding species (Cockburn, 2004). This may not have a major impact on a helper’s relatedness to the brood provided that parentage is shared among its relatives within the social group, as in acorn woodpeckers, Melanerpes formicivorus (Haydock et al., 2001). But, in the Seychelles warbler where paternity is often gained by nonkin from outside the social group (Richardson et al., 2002), a helper’s kinship to the helped brood may be substantially reduced. In such cases, helpers need to develop more sophisticated ways of reliably determining their true relatedness to a brood. As explained above, female Seychelles warblers do appear to effectively maximize indirect benefits (Richardson et al., 2003a,b), which suggests they have evolved some accurate means of kin discrimination. The life history parameters of this species provide a situation in which effective kin discrimination through association may develop. Breeding pairs normally remain together within a territory as long as both birds survive and offspring have a long period of dependence, remaining on the natal territory for at least 6 months (Komdeur, 1992). Evidence that kin discrimination by nonparent female subordinates is achieved through association, and not by direct assessment of relatedness, is twofold. (i) When a subordinate female had a choice of helping at two different nests built by two different females which were present during different periods on her natal territory, she did not always feed the nestling to whom she was most related, instead she only fed at those nests with breeders which had fed her as a nestling (Komdeur, 1994c). (ii) The decision to help a breeding pair appears to be based on the continued presence of the primary female (but not the male) who previously fed the subordinate in the nest (Richardson et al., 2003a,b). Female subordinates provisioned significantly more when the primary female that fed them as chicks was still present in the territory (Fig. 2). The female’s continued presence predicted the subordinate’s provisioning behavior better than did subordinate–primary female or subordinate–nestling relatedness. As the continued presence of the primary female reliably indicated relatedness to the nestling (Richardson et al., 2003a,b), this cue was effective in maximizing indirect benefits for subordinates. Levels of provisioning did not alter with the presence/absence of the primary male (Fig. 2). The fact that the primary male was not used as a cue is logical in an evolutionary sense, as high levels of female infidelity (40% extra‐group paternity, Richardson et al., 2001) mean that subordinates are often unrelated to the primary male. Consequently, his continued presence did not indicate relatedness between the subordinate and nestling (Richardson et al., 2003a,b). This study is the first to show that an associative learning mechanism could

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

n= 10

4

8 NS

6

161

7

p = 0.003

Feeds per hour

8 6 4 2 0 Primary male Primary female FIG. 2. Provisioning by subordinate female, nonparent Seychelles warblers in relation to the presence (light gray bars) or absence (dark gray bars) of their putative parents (primary males and females). Bars indicate means  S.E. (Adapted from Richardson et al., 2003b.)

have evolved to be focused only on one sex (the female attendant at the nest), the sex to which the subordinates are reliably related. However, these data were only correlative. A cross‐fostering experiment was also conducted which produced nestlings that were unrelated to the primary female that raised them and control nestlings that were closely related to their primary females. The proportion of cross‐fostered and control female offspring that stayed and became helpers on their ‘‘natal’’ territory did not differ significantly (Fig. 3; Komdeur et al., 2004b). For both groups, the chance of becoming a subordinate helper was associated only with the continued presence of the primary female; no control or cross‐fostered females helped when the primary female died and was replaced by another female (Fig. 3; Komdeur et al., 2004b). These results strongly supported the idea that helping is based on associative‐learning cues and not on the direct assessment of genetic relatedness. Although the decision rules used by female subordinates to discriminate effectively between related and unrelated nestlings are now well understood, the associative‐learning period and recognition cues (e.g., calls or plumage characteristics) used by subordinate females to distinguish between primary females have not yet been investigated. Studies like ours on the Seychelles warbler clearly demonstrate that subordinates can use kin discrimination to maximize the indirect benefits of helping in cooperative breeding systems. However, there are many other cooperative breeding societies where related subordinates do not help (Clutton‐Brock, 2002; Cockburn, 1998), or where helpers are unrelated to

162

JAN KOMDEUR AND DAVID S. RICHARDSON

0.7

18

0.5

n=

10

0.5

13

0.4

0.4

0.3

0.3

0.2

12

0.1 4

0.0

0.2 2

0.1 3

2

0.0 −0.1

Subordinate-primary female relatedness

Fraction helping

0.6

Primary female: Present Absent Present Absent Subordinate female: Control Cross-fostered FIG. 3. The fractions (gray bars) of control and cross‐fostered female subordinate Seychelles warblers that helped to feed nestlings during their second year of life in relation to the continued presence of the primary females and primary foster females (w2 ¼ 1.30, d.f. ¼ 1, p ¼ 0.253) and in relation to the new primary females that replace the primary (foster) females that raised the subordinates (absent). Only those subordinates that could have been helpers, because nests with young were present in their natal territory, were included (n ¼ 38 individuals). Circles indicate mean relatedness of subordinates (S.E.) to the primary females and primary foster females that raised the subordinates and were still present on the territory (present; t20 ¼ 3.48, p ¼ 0.002) or to new primary females that replaced the primary (foster) females that raised the subordinates (absent). The number of subordinates is indicated. (Adapted from Komdeur et al., 2004b.)

the young but still invest as heavily as close relatives (reviewed in Komdeur et al., 2007). Such behavior could result from recognition errors, the acceptance threshold being set relatively high or low, respectively. While helpers doubtless make recognition errors, these are unable to account for all instances of cooperation between nonkin. Therefore, although the majority of cooperative breeding systems involve some degree of kinship (Emlen, 1995), this may often be due to the benefits of philopatry and helping rather than the benefits of helping per se (Clutton‐Brock, 2002). In some cases, helpers actively compete for access to unrelated offspring (Clutton‐Brock et al., 2000; Dunn et al., 1995; Magrath and Whittingham, 1997). For the above reasons an adaptive explanation for care being given by helpers to nonkin is that helpers may foster ‘‘social bonds’’ with recipient young, bonds that later benefit the helper either by increasing the probability that the young will return the favor, or by promoting development of coalitions beneficial in competing for breeding positions (Emlen, 1991; Emlen et al., 1991; Ligon 1983). Thus, by feeding unrelated offspring,

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

163

helpers may parasitize a kin‐recognition mechanism based on associative learning; the deceived offspring recognize those who care for them as kin and later help rear their provisioners’ offspring (the ‘‘kinship deceit hypothesis’’; Connor and Curry, 1995). An alternative hypothesis is that helpers increase their own survival and future reproduction by cooperating with others (direct fitness benefits; see below). In these situations, the ability to discriminate kin from nonkin is not necessarily a prerequisite. To conclude, it should be clear that the selection for an effective kin discrimination mechanism will depend on the importance of kin selection in the evolution of a particular species’ cooperative system. No kin discrimination is expected if the only benefit of helping is to increase the helper’s direct fitness, and the acceptance threshold for effective kin discrimination would be expected to become more stringent as the relative importance of indirect fitness benefits increases (Griffin and West, 2003).

D. DIRECT BENEFITS The widely held view that kin selection is the explanation for the evolution of alloparental care in all vertebrate systems has recently been questioned (Cockburn, 1998; Dickinson and Hatchwell, 2004; Emlen, 1995, 1997; Komdeur, 2006). The reasons are threefold. First, several cooperative breeding societies have been described where subordinates do not appear to provide help (Cockburn, 1998). Second, there are reports of cooperative systems containing unrelated subordinates that help raising offspring (Clarke, 1989; Creel and Creel, 2002; Reyer, 1980; Rood, 1990; Sherley, 1990; Stiver et al., 2004; Whittingham et al., 1997), and many studies have found no correlation between helping effort and relatedness (Clutton‐ Brock et al., 2000; Dunn et al., 1995; DuPlessis, 1993; Magrath and Whittingham, 1997; Rabenold, 1985; Zahavi, 1990). One factor that complicates the study of benefits within cooperative breeding systems is that different members of the same group may incur different costs and benefits from helping (reviewed in Cockburn, 1998; Heinsohn and Legge, 1999). For example, studies on birds indicate that benefits differ between the sexes (Cockburn, 1998). Therefore, any study focusing on the benefits of cooperative breeding in systems with subordinates of both sexes should analyze the sexes separately. Such differences might in turn help to explain the skewed sex ratios of subordinates often seen in cooperative species (West and Sheldon, 2002). With the advent of molecular genotyping techniques to accurately determine patterns of relatedness we are now able to quantify the indirect and direct benefits accruing to different subordinates within cooperatively breeding species.

164

JAN KOMDEUR AND DAVID S. RICHARDSON

Previous work on the Cousin Island, population of Seychelles warblers showed an increased production of offspring by the recipients of help (Komdeur, 1994b). However, this study did not control for the possibility of subordinates gaining direct parentage. As shown in the previous section, many (44%) female subordinate Seychelles warblers gained direct parentage within their group. Subordinate males very rarely gained direct benefits through parentage within the group (Richardson et al., 2002). On the other hand, subordinate Seychelles warblers do not gain fitness benefits through direct reproductive success on territories other than their own. Egg dumping between groups does not occur in the Seychelles warbler and subordinate male warblers do not increase their reproductive success through extra‐group paternity, as all extra‐group offspring are sired by dominant breeding males (Richardson et al., 2001, 2002). Using parentage analyses and a coefficient of relatedness we determined, in terms of genetic equivalents, the direct breeding benefits and the indirect benefits for each subordinate within a population of the Seychelles warbler. Fitness benefits were measured using the number of nestlings produced. The study showed that the direct breeding benefits accrued by subordinates were much greater than the indirect benefits. After removing the direct offspring produced by subordinates, Seychelles warblers did still gain indirect fitness benefits through cooperative breeding. On average, the presence of a subordinate on a territory resulted in a significant increase in the number of nondescendent offspring in a territory (0.18 extra offspring produced per subordinate; Richardson et al., 2002). However, because of low levels of relatedness between subordinates and nondescendent kin (especially for males), this translated into a relatively small indirect fitness benefit gain for female subordinates (mean offspring equivalents: females ¼ 0.07  0.26, males ¼ 0.04  0.17). Furthermore, as male subordinates are not more related to the offspring in the territory than expected by chance, indirect benefits could be considered to be nonexistent for males. For female and male subordinates direct benefits were significantly higher than indirect benefits, although this difference was more extreme in females, which gained significantly more direct benefits than males (Fig. 4). The combined benefits also tended to be higher for females (p ¼ 0.07). The direct breeding benefits of cooperative breeding are, on average, six times greater than the indirect benefits (offspring equivalents; 0.36 vs 0.06, respectively; Fig. 4), suggesting that direct benefits are a stronger selective force behind the evolution of cooperative breeding than indirect benefits. It is, however, important to note that the estimates of indirect benefits are the average level for all nonparent subordinates. Variation in relatedness between the female subordinates and nestlings, combined with the adjustment of helping effort based on accurate kin

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

n=

40

165

20

Benefits (offspring equivalents)

p < 0.05 0.6

p < 0.001

0.5 0.4 0.3 0.2

NS p < 0.05

0.1 0.0 Female Male Subordinates

FIG. 4. The fitness benefits of cooperative breeding gained by female and male subordinates in the Seychelles warbler (1997–1999). Statistical significance assessed by Mann–Whitney Z‐statistic. Both female (n ¼ 43) and male (n ¼ 20) subordinates gain significantly higher direct breeding benefits (open columns) compared to indirect benefits (filled columns). Direct breeding benefits are significantly higher in females than in males, but there is no significant difference between the sexes in indirect breeding benefits. Bars indicate means  S.E. (Adapted from Richardson et al., 2002.)

discrimination, means that indirect benefits can be maximized and that they will be considerably more important for some female subordinate than others (see Section C above). Another criticism of our study is that by only measuring the production of nestlings we did not also take into account future indirect fitness benefits (Mumme et al., 1989, reviewed in Cockburn, 1998; Emlen, 1995). Postfledgling survival is greater in territories with subordinates in the Seychelles warbler (Komdeur, 1994b), but this will equally benefit subordinates’ descendent and nondescendent offspring and would not change the relative importance of direct and indirect benefits. On the other hand, future direct benefits are important in the Seychelles warbler. For example, the acquisition of parenting skills by female subordinates result in them being superior breeders when they acquire a territory (Komdeur, 1996c). Male subordinates may also gain by budding off part of a territory that has been enlarged through group augmentation (Komdeur and Edelaar, 2001). However, the inclusion of the extra direct benefits that subordinates gain would only provide further support for the predominant role of direct benefits in the evolution and maintenance of cooperative breeding in the Seychelles warbler.

166

JAN KOMDEUR AND DAVID S. RICHARDSON

Finally, the fact that females gain much higher levels of indirect benefits than do males may explain why most subordinates in the Seychelles warbler are female. Alternatively, it may be that primary birds preferentially accept female subordinates. Indeed, previous studies have shown that primary females on high‐quality females skew the sex ratio of eggs (88% of female eggs) so as to produce a greater proportion of female offspring, which will become subordinates in subsequent year (Komdeur, 1996b; Komdeur et al., 1997). This preference for female subordinates may be explained by the greater benefits accruing to primary birds through having female rather than male subordinates. With female subordinates the primary birds gain both an increase in their own productivity and also the indirect benefits associated with subordinate females breeding (J. Komdeur, D. S. Richardson, T. Burke, and S. T. Emlen, unpublished data). Indeed, primary females do not appear to control (or constrain) reproduction by subordinate females: There is no aggression among females at the nest around the time of egg laying. Primary females do not remove the other female’s egg (or experimentally introduced eggs), before or after laying their own, nor do they desert the nest because of such egg laying (Komdeur, 2005). Male subordinates do not increase the group’s productivity by ‘‘laying’’ eggs in the nest as female subordinates do, but rather compete with the dominant male for the fertilization of these eggs.

IV. INBREEDING AND INBREEDING AVOIDANCE The deleterious effects of inbreeding can be substantial in wild populations (reviewed in Crnokrak and Roff, 1999; Frankham et al., 2002) and can contribute toward driving populations to extinction (Bijlsma et al., 2000; Brook et al., 2002; Frankham, 1998; Saccheri et al., 1998). Various mechanisms have evolved to avoid the possibility of inbreeding in animals, for example, sex‐biased dispersal or reproductive suppression (reviewed in Pusey and Wolf, 1996). Kin recognition is important in relation to helping behavior and kin selection as discussed above, but may also be important for incest avoidance (Koenig and Haydock, 2004). In situations where inbreeding may still occur, that is in kin‐structured populations where individuals are likely to encounter closely related potential mates, social mate choice to avoid inbreeding should be favored by selection (Crozier, 1987; Pusey and Wolf, 1996). However, this avoidance behavior does not occur in all species (Craig and Jamieson, 1988; Keane et al., 1996; Keller and Arcese, 1998) and there are many situations, for example, social monogamy, where social mate choice is restricted and may lead to nonoptimal pairings (Kruuk et al., 2002).

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

167

Our studies on the Seychelles warblers show that this species do not appear to avoid inbreeding through social mate choice. (i) Primary females were no less related to their social mates than expected by chance across the whole population (mean r ¼ 0.01 vs 0.00; Richardson et al., 2004). Furthermore, although subordinates are not necessarily making a social mate choice when deciding to remain on a territory, it is clear that the decision to stay is not based on avoiding the possibility of being in a territory with a related male: subordinate females were more related to the primary male than expected by chance across the population (Fig. 5; mean r ¼ 0.14 vs 0.00) and were also more related to the primary male than were primary females (Richardson et al., 2004). Consequently, subordinate females that mate with the primary male in their own territory will often produce inbred offspring, though this is probably a constraint of being a subordinate rather than due to active mate choice. Such matings result in the higher than random level of inbreeding between first‐order relatives observed in this population (Fig. 6; Richardson et al., 2004). (ii) If females are aiming to avoid incestuous matings then we would expect females to gain a breeding territory further away from the natal territory than males for the following p < 0.01 p < 0.05

Pairwise relatedness

0.20

NS 19

0.15 0.10 0.05 0.00

n = 78

19

−0.05 Random pairs

Primary female Subordinate female –primary male –primary male

FIG. 5. Mean pairwise relatedness between the mother (either primary female or subordinate female) and the primary male in a territory, compared to the pairwise relatedness of random dyads on Cousin Island. Primary female–primary male pairs were not more related than expected by chance (0.008  0.22 vs 0; one‐sample t‐test; t ¼ 0.34, d.f. ¼ 77, p ¼ 0.73). Subordinate female‐primary male pairs were significantly more related than expected by chance (mean ¼ 0.14  0.20 vs 0; one‐sample t‐test; t ¼ 2.95, d.f. ¼ 18, p < 0.01) and were also more related than were primary female‐primary male pairs (mean ¼ 0.14  0.20 vs 0.008  0.22; t ¼ 2.33, d.f. ¼ 95, p ¼ 0.02). Bars indicate means  S.E. (Adapted from Richardson et al., 2004.)

168

JAN KOMDEUR AND DAVID S. RICHARDSON

20

Frequency

16 12 8 4 0 −0.6

−0.4 −0.2 0.0 0.2 0.4 Relatedness of genetic parents

0.6

FIG. 6. Inbreeding in the Seychelles warbler. The pairwise relatedness values of the genetic parents of 119 assigned offspring in the Seychelles warbler were distributed around a mean of 0.05  0.23. Mid‐points of classes are shown. This distribution displays a right skew and indicates that 5% of matings in the Seychelles warbler population were between close relatives (values centered on 0.46, the value calculated for first‐order relatives). (Adapted from Richardson et al., 2004.)

reason. The frequency of extra‐group paternity is high (40%) in the Cousin population, and extra‐group males are normally from adjacent territories (number of territories away from natal territory where 0 ¼ adjacent territory: mean 1.9  2.2, median ¼ 1, range 0–6; Fig. 7; Richardson et al., 2001). Consequently, females that disperse a short distance will have a higher chance of dispersing into the territory that holds their genetic father than females dispersing further. In contrast, as territory switching is rare (Eikenaar et al., 2007; Komdeur, 1996c) and egg dumping nonexistent in the Seychelles warbler (Richardson et al., 2001), males that disperse a short distance have an equally low chance of dispersing into the territory that holds their genetic mother as do males that disperse further. In agreement with this, the average dispersal distances between natal and breeding territory was significantly larger for females than males (females: 4.0  4.0 territories, n ¼ 69; males: 2.0  2.0 territories, n ¼ 86; Mann–Whitney: Z ¼ 2.67, p < 0.01; Eikenaar et al., 2007). Where inbreeding avoidance through social mate choice does not occur, extra‐pair mate choice may have evolved as a mechanism to avoid inbreeding (Bensch et al., 1994; Brooker et al., 1990). We investigated this possibility in the Seychelles warbler. Parentage analysis confirmed the results of an earlier study (Richardson et al., 2001) showing that a high

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

169

N

100 m

Rock Territorial boundary FIG. 7. Extra‐pair paternity in the Seychelles warbler. The map shows all territories on Cousin island in July 1999. Arrows point from the territory of the assigned extra‐group male to the natal territory of the extra‐group offspring. (Adapted from Richardson et al., 2001.)

level of EPP occurs in the Seychelles warbler (37% of offspring; Richardson et al., 2004). However, females (primary or subordinate) do not appear to use EPP to avoid inbreeding in a facultative manner, as they did not preferentially gain EPP when closely related to the primary male in the territory. Moreover, for primary females the genetic fathers of their extra‐ pair offspring were no less related to them than were the cuckolded primary males. One prediction—that subordinate females, who were often highly related to the primary male, should gain higher levels of EPP to avoid incestuous mating—was upheld. Female subordinates did have significantly higher levels of EPP than primary females (58% vs 32%, w2 ¼ 4.24, d.f. ¼ 1, p ¼ 0.04). Furthermore, the extra‐pair offspring produced by subordinate females were more heterozygous (less inbred) than the within‐pair offspring they produced (Fig. 8). Consequently, EPP in subordinate females

170

JAN KOMDEUR AND DAVID S. RICHARDSON

7

n=

58 40 NS

9

12

p < 0.05

Heterozygosity (Hz)

6 5 4 3 2 1 0 Primary

Subordinate

Female status FIG. 8. Heterozygosity of offspring produced by primary and subordinate females. For subordinate females the heterozygosity of within‐pair offspring (open columns) was significantly lower than that of extra‐pair offspring (filled columns), but there was no difference between the two groups for primary females. Bars indicate means  S.E. (Adapted from Richardson et al., 2004.)

did appear to reduce inbreeding in their offspring. However, on a case for case basis the probability of EPP was not directly influenced by how related they were to the primary male.

V. MATE CHOICE The existence and nature of indirect genetic benefits to mate choice remains a major unresolved question within evolutionary biology (reviewed in Jennions and Petrie, 2000). It has been hypothesized that females may gain indirect benefits through the acquisition of good paternal genes (Hasselquist et al., 1996; Petrie, 1994) or the enhanced genetic compatibility of maternal and paternal genomes (Jennions and Petrie, 2000; Zeh and Zeh, 1996). Several studies have provided evidence that mate choice can increase offspring fitness through genetic benefits (Hasselquist et al., 1996; Petrie, 1994; Wilkinson et al., 1998). But almost no studies have identified the underlying genes that may account for variation both in the trait on which choice is made and in fitness variation (von Schantz et al., 1996, 1997; Wilkinson et al., 1998). Such indirect genetic benefits have long been evoked to explain active mate choice in situations where direct benefits do not appear to be gained, for example, where females in socially

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

171

monogamous species pursue extra‐pair copulations (EPCs; Andersson, 1994). Males may gain direct benefits by engaging in EPCs as these can lead to extra‐pair fertilizations (EPFs), increasing a male’s reproductive output without additional paternal investment (Birkhead and Møller, 1992; Westneat et al., 1990). Since the maximum reproductive success for females is limited by the number of offspring they can produce, the benefits to females of EPFs are less clear. Extra‐pair fertilizations occur in the majority of passerine bird species (Birkhead and Møller, 1992; Griffith et al., 2002), but evidence that such mate choice can increase offspring fitness through genetic benefits remains scarce (Hasselquist et al., 1996; Petrie, 1994). Investigating the role of genetic benefits in mate choice in the Seychelles warbler has been one of our major long‐term aims. In this isolated and contained system, we have been able to mark and follow nearly all birds for a number of years. We have also been able to assign parentage to all young and determine the reproductive success and fitness of all marked individuals. Consequently, we are able to follow the fitness consequences of mate choice, be it social or extra‐pair. As discussed above, we have already shown that neither social nor extra‐pair mate choice is used to avoid inbreeding directly. In this section, we highlight the work we have done, or are currently doing, to ascertain what other genetic benefits female Seychelles warblers may gain for their offspring through their social and extra‐pair mate choice. Importantly, we also try to investigate which genes may underlie such benefits. The MHC is an important component of the vertebrate acquired immune system. Molecules encoded for by the MHC‐alleles bind peptides derived from foreign antigens and present them to T cells, thereby triggering an adaptive immune response (Hughes and Yeager, 1998). Individual differences in MHC diversity influence pathogen susceptibility (Doherty and Zinkernagel, 1975). In a population under selection pressure from pathogens, MHC‐based mate choice could increase the pathogen resistance and, consequently, the fitness of offspring (Grob et al., 1998; Jordan and Bruford, 1998; Penn and Potts, 1999). The MHC may, therefore, be the link between mate choice and the genetic inheritance of vigor in offspring. Under a good‐ genes model, if females preferentially mate with superior condition males that are relatively disease resistant they may provide their offspring with ‘‘good’’ MHC haplotypes associated with improved resistance to pathogens within the current environment (Abplanalp et al., 1992; Rulicke et al., 1998; Wedekind et al., 1996). In this scenario, males with a greater diversity of MHC alleles should, through either heterozygote advantage (Doherty and Zinkernagel, 1975) or from an increased chance of having specific resistant alleles (Bodmer, 1972), be in better condition and, therefore, be preferred by females. It is also possible that females may directly assess the MHC of

172

JAN KOMDEUR AND DAVID S. RICHARDSON

potential mates and choose to mate with diverse males, which will, in turn, sire more MHC‐diverse offspring. Alternatively, females may seek the best combination of maternal and paternal MHC genes (Brown, 1997; Tregenza and Wedell, 2000; Zeh and Zeh, 1996). In this case, MHC‐based mating preferences may be beneficial if MHC‐disassortative mating results in MHC‐diverse offspring, which respond to a greater variety of pathogens (Hedrick, 2002; Penn et al., 2002; Potts and Slev, 1995). MHC‐dissimilar mates may also provide a ‘‘moving target’’ against rapidly evolving parasites that escape immune recognition (Penn and Potts, 1999). Other studies have suggested that an intermediate, rather than the highest, level of MHC diversity is optimal (Wegner et al., 2003; but see Hedrick, 2004; Kurtz et al., 2004; Wegner et al., 2004); and the preferred male may be the one that results in an intermediate amount of MHC diversity in the offspring (Aeschlimann et al., 2003; Penn and Potts, 1999). Various studies have provided evidence that MHC‐based mate choice occurs in mammals and fish (Aeschlimann et al., 2003; Ober et al., 1997; Penn and Potts, 1999; Piertney and Oliver, 2006; Potts et al., 1991; Reusch et al., 2001; Wedekind and Furi, 1997; Yamazaki et al., 1988). However, although birds have been used extensively for studies of sexual selection and mate choice, until recently little work has focused on the role of the MHC (Bonneaud et al., 2006; Freeman‐Gallant et al., 2003; von Schantz et al., 1996, 1997; Westerdahl, 2004). We developed a protocol to screen for individual variation in the MHC class 1 exon 3, genes in the Seychelles warbler. As expected for a genetically bottlenecked species, the Seychelles warbler contains relatively limited genetic variation at the MHC (Richardson and Westerdahl, 2003). But evidences, such as the high nonsynonymous to synonymous substitutions ratio in the peptide‐binding region (PBR) and the higher levels of genetic variation in the MHC compared to neutral loci, suggest that this variation has been maintained by selection (Richardson and Westerdahl, 2003). The low number of MHC genotypes in this population means that individuals often encounter MHC‐similar potential mates and are, therefore, more likely to have evolved to discriminate against such matings. We screened the MHC of all individuals that had previously been included in the parentage analyses and tested for MHC‐dependent mating patterns. We found no influence of MHC on social mate choice (Richardson et al., 2005). First, pair males were not more MHC diverse than expected by chance. Second, MHC‐disassortative pairing did not occur; there was no difference in MHC similarity between females and their social mates than expected by chance. Nor was there any evidence that females preferred males with whom they had intermediate levels of MHC similarity. We did, however, find that MHC‐dependent extra‐pair mating occurred (Richardson et al., 2005).

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

173

A 0.6 0.5 0.4

n= 6

19

19

0.3 Proportion of extra-pair offspring

21

17

0.2 0.1 0.0

B 0.6

0.0–0.2 0.2–0.4 0.4–0.6 0.6–0.8 0.8–1.0 MHC band-sharing n = 10 14

0.5 0.4

3

37

0.3 16

0.2 0.1 0.0

2 2

3

4 5 6 Male MHC diversity

7

FIG. 9. Extra‐pair paternity and the MHC characteristic of the social male. The proportional frequency of extra‐pair paternity in relation to: (A) the MHC band‐sharing between the social male and female and (B) the MHC diversity of the social male. Figures above the columns give the number of independent females for which MHC choice was identified in each class. (Adapted from Richardson et al., 2005.)

Although there was no correlation between EPP and MHC similarity (no evidence of MHC‐disassortative extra‐pair mating; Fig. 9A), females were more likely to obtain EPP when their social mate had low‐MHC diversity (Fig. 9B). Furthermore, the MHC diversity of the extra‐pair male was significantly higher than that of the cuckolded male (Fig. 10). There was no evidence that males with an intermediate number of MHC bands were more likely to gain EPP (Richardson et al., 2005). The results point to the most MHC‐diverse male rather than the most MHC‐dissimilar male gaining EPP. As these patterns of MHC‐dependent extra‐pair matings will result in offspring of higher MHC diversity, they will provide indirect benefits to female Seychelles warblers if survival is positively linked to MHC diversity. We are presently investigating this last point (see below).

174

JAN KOMDEUR AND DAVID S. RICHARDSON

n= 4.6

53

29

29

p < 0.05 p < 0.05

MHC diversity

4.2

3.8

3.4

3.0 FIG. 10. The MHC diversity of the social and genetic mate of female Seychelles warblers; noncuckolded social males (open columns), cuckolded social male (lightly shaded columns), extra‐pair male (heavily shaded columns). Bars indicate means  S.E. Samples sizes are shown as numbers above the columns. (Adapted from Richardson et al., 2005.)

The patterns of MHC‐dependent extra‐pair matings described above could arise through various different mechanisms. They could be the result of mate choice based directly on an individual’s MHC. For example, olfactory cues may be used to assess the MHC composition of potential partners and, subsequently, to choose a mate. However, given that mating in the Seychelles warbler is biased toward males that are MHC diverse (rather than MHC dissimilar or MHC intermediate) in our opinion, the most parsimonious hypothesis is that there is an indirect link between the MHC and mate choice. For example, if mate choice is based on condition‐dependent cues and MHC diversity influences condition, then female Seychelles warblers may be mating with higher MHC‐diverse males as a result of choosing extra‐pair males with superior traits. However, analysis shows that three obvious traits that could reflect male genetic quality (tarsus length, wing length, and territory quality) were not correlated with male MHC diversity (Richardson et al., 2005). Surprisingly, mass and body condition were negatively correlated with MHC diversity. This relationship, which is in the opposite direction to that predicted based on the idea that MHC‐diverse males are of better quality, needs to be investigated further. The MHC‐based extra‐pair matings observed may, on the other hand, be a result of sperm competition (reviewed in Birkhead and Møller, 1998; Wedekind et al., 2004). Females may gain extra‐ pair copulations from males regardless of their MHC genotype. Sperm competition may then favor sperm from males of high MHC diversity. This may act through male–male sperm competition if sperm quality or quantity is influenced by male MHC diversity. Alternatively, the MHC‐based

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

175

extra‐pair mating pattern may be a result of cryptic female choice—there is evidence that MHC antigens may be expressed on sperm and that nonrandom fusion of gametes can occur (reviewed in Dorak et al., 2002; Wedekind et al., 2004). Both mechanisms could lead to a situation where EPP occurs more often in females that are mated socially to males with low‐genetic diversity, and the resulting genetic fathers will then have higher MHC diversity than the cuckolded social mates. Direct mate choice and postcopulatory sperm competition experiments are now required to resolve these issues. That social mate choice did not appear to be influenced by the MHC is not surprising, considering the extreme constraints imposed on social mate choice in the Seychelles warbler. The Cousin Island population contains a surplus of adult birds (320 birds for ca. 115 territories; Brouwer et al., 2006; Komdeur, 1992; Richardson et al., 2002), territories are maintained year round by pairs (which remain together for many years) and there is no empty space left to establish new territories (Fig. 5). Consequently, opportunities for independent breeding occur rarely and a substantial proportion of adult birds die without ever having acquired a breeding territory (Komdeur, 1992; Komdeur et al., 2004a). Under these conditions, unpaired females should pair with any male possessing a breeding territory, irrespective of his individual quality or MHC traits. Females may then have to rely on extra‐pair matings to improve the genetic quality of their offspring. The translocation of 58 warblers in 2004 to set up a new population on Denis Island did, however, provide us with the means to experimentally relax the constraints on social mate choice and observe the consequences for mate choice. A total of 27 females and 31 males were translocated from Cousin Island. No pairs of birds previously together on Cousin were translocated. All birds were released unharmed and allowed to pair up at their own discretion. The logic behind the experiment was that released females would not be constrained in their mate choice by a lack of breeding opportunities. A considerable excess of suitable habitat on the island provided the opportunity for all pairs to breed independently, while the greater number of males meant that all females would have some choice over whom to pair up with. Over the 2 years since this translocation the identity of all pairs that attempted breeding have been identified, and their reproductive success has been monitored. Work is presently underway to screen all the birds at both the MHC class I and II to determine if as predicted, the MHC would have a significant effect on social mate choice when the normal constraints limiting choice were removed.

176

JAN KOMDEUR AND DAVID S. RICHARDSON

VI. CONCLUSIONS AND FUTURE AVENUES In the above text, we have described how we have used molecular techniques to reveal some of the hidden complexities of the Seychelles warbler system. More importantly, we have tried to interpret and explain the evolutionary significance of these results, not just for the warbler but across all taxonomic boundaries. These studies show that the Seychelles warbler can be a model system in which to investigate questions in behavioral, evolutionary (and now) molecular ecology. However, there are many areas and issues which we have not yet investigated where modern molecular techniques have not yet been applied, or where our studies have merely scratched the surface and revealed how little we do know. Some of these areas have been dealt with directly in the main text where the flow of narrative has allowed. Below we outline some of the other areas that we have identified as important, or interesting, which we think we can explore using the Seychelles warbler system. This list is by no means exhaustive or exclusive but is put forward to show the directions in which we hope to take this project. We have not, as yet, quantified the long‐term fitness consequences of sex allocation and cooperative breeding. There is evidence that in the short‐ term dominant females in high‐quality territories obtain higher fitness by producing female rather than male offspring (Komdeur, 1998). With the production of female subordinates, the primary female may gain both an increase in her own productivity and indirect benefits associated with subordinate breeding females (grand offspring). However, the fact that both dominant and subordinate females may lay eggs within the nest adds complexity to the issue of sex allocation which need to be explored. In light of the subordinate maternity and EPP revealed by the studies described above, it is essential that we assess the long‐term inclusive fitness consequences of sex allocation to confirm its adaptive nature. Furthermore, as we now realize that fitness benefits may differ (or conflict) between male and female we need to investigate experimentally the fitness consequences accrued by both members of the breeding pair. While we have been able to show that MHC‐dependent patterns of mating do occur, the overall goal of determining if, and how, genetic benefits may be gained through mate choice is a long way from being reached. Prof. Terry Burke (Sheffield University, UK),with whom much of the molecular work has been done in close collaboration, is now leading a project to construct a multigeneration pedigree for the Cousin Seychelles warbler population which will span the 15 years, for which we (will) have almost complete sampling of the population (1993–2008). Accurate parentage and pedigree assignment will be determined using an enlarged suite of

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

177

genetic markers and cutting edge, Bayesian statistical software, already developed as part of this project (Hadfield et al., 2006). This pedigree, and the key parameters that are now determined for this species (e.g., MHC genotype, heterozygosity, morphometrics, survival, and reproductive success) will be incorporated into animal model analyses to produce a rigorous framework for understanding the contribution that environmental and genetic factors make toward the overall lifetime reproductive success of individuals. One major specific aim will be to further investigate the fitness consequences of mate choice. Such a study on this model system provides one of the best opportunities to determine the role of genetic benefits in mate choice in the wild. In collaboration with Terry Burke, we propose to use the pedigree based analysis to elucidate to what extent the phenotypic behavioral responses that we observe (e.g., dispersal, reproduction, and provisioning) have an underlying genetic causality. Is the behavior that we see purely facultative, or does it reflect genetic differences among individuals? Comparable studies of variance components are being pursued in other species, but they generally do not focus on behavior, as we do here. Very few previous studies have measured the heritability of behavioral traits in the natural environment (Stirling et al., 2002), though there is increasing evidence for genetically determined variation in behavioral traits of ecological significance (Dingemanse et al., 2002, 2003; Drent et al., 2002; MacColl and Hatchwell, 2003). Fundamental to our understanding of the evolution of cooperative breeding is why certain individuals refrain from reproduction. Although many studies have focused on the role of life history and ecological factors, the proximate mechanisms underlying such behaviors have received considerably less attention. In the Seychelles warbler male subordinates never gain paternity outside and only rarely within the territory (Richardson et al., 2001). We are investigating whether there are differences in testosterone levels and the amount of sperm storage between primary and subordinate male warblers, and if these differences change in relation to the period within the breeding cycle. Recent work has shown that both testosterone level and sperm storage capacity declined in primary males after egg laying of the primary female in their resident territory, but that these were constantly low in subordinate males (van de Crommenacker et al., 2004). Now we plan to determine whether subordinates are able to elevate their T levels and sperm storage capacity on experimentally being promoted to a primary male. Parasite and diseases (here termed pathogens) can have profound effects on the condition, reproductive success and survival of individuals. The MHC, described above, is a key component of the vertebrate acquired

178

JAN KOMDEUR AND DAVID S. RICHARDSON

immune system and the type and variety of MHC genes an individual carries can determine how well they can cope with pathogens. Although many studies have identified links between specific MHC characteristic and resistance to pathogens much of this work is in laboratory or domestic animals (Piertney and Oliver, 2006). Less work has been done on the role of the MHC in interactions between host and pathogens in the wild, and studies which attempt to determine, both qualitatively and quantitatively, the role of the MHC in fitness in natural populations are missing. For the last few years we have routinely screened all individuals for a range of pathogens including blood parasites (e.g., avian malaria) and gastrointestinal and parasites (e.g., coccidia). Using this data, in conjunction with individual MHC‐genotyping and survival data, means that we can assess the impact of these pathogens on their hosts and determine the role the MHC has in mitigating this. It will also allow us to follow up on the role of the MHC in mate choice study. To confirm whether the MHC‐dependent extra‐pair matings observed really do provide genetic benefits we need to determine if offspring of increased MHC diversity are fitter than those of lower MHC diversity. It is clear that many phenomena found in social species cannot be fully captured without the use of modern molecular techniques and long‐term analyses. Just as haplodiploid insects inspired much of the classic sex‐ allocation theory, as a result of their relatively simple sex ratio manipulation mechanism and life histories, the complexity of bird life histories are inspiring new theoretical developments (reviewed in Hardy, 2002). It is clear that life histories may become even more complex after the incorporation of the hidden life histories and trade‐offs, such as EPP, MHC‐based mate choice, revealed by molecular techniques. Hence, a key challenge for both theoreticians and empiricists will be to integrate the multidimensional nature of allocation decisions and the conflicts that occur over such decisions.

Acknowledgments Many people have collaborated and given invaluable help with the Seychelles warbler project over the past 18 years. We are grateful to Nature Seychelles for allowing us to work on Cousin Island and for providing accommodation and facilities during our stay. The Department of Environment and the Seychelles Bureau of Standards gave permission for fieldwork and sampling. Genetic analyses were performed in the Sheffield molecular genetics facilities of Sheffield (UK) under the guidance of Terry Burke and Deborah Dawson and of Lund (Sweden) under the guidance of Torbjo¨rn von Schantz and Helena Westerdahl. We also thank Janske van de Crommenacker, Iain Barr, Jane Brockmann, and two anonymous reviewers for their helpful comments on this chapter. We are extremely grateful for the

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

179

financial support provided by the Natural Environmental Research Council, the Netherlands Organization for Scientific Research (NWO‐WOTRO, NWO‐VICI), the Australian Research Council, and the University of Groningen and Norwich. References Abplanalp, H., Sato, K., Napolitano, D., and Reid, J. (1992). Reproductive performance of inbred congenic leghorns carrying different haplotypes for the major histocompatibility complex. Poult. Sci. 71, 9–17. Aeschlimann, P. B., Haberli, M. A., Reusch, T. B. H., Boehm, T., and Milinski, M. (2003). Female sticklebacks Gasterosteus aculeatus use self‐reference to optimize MHC allele number during mate selection. Behav. Ecol. Sociobiol. 54, 119–126. Andersson, M. (1984). Brood parasitism within species. In ‘‘Producers and Scroungers’’ (C. J. Barnard, Ed.), pp. 195–228. Croom‐Helm, London. Andersson, M. B. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton, NJ. Arnold, W. (1990). The evolution of marmot sociality. 2. Costs and benefits of joint hibernation. Behav. Ecol. Sociobiol. 27, 239–246. Arnold, K. E., and Owens, I. P. F. (1998). Cooperative breeding in birds: A comparative test of the life history hypothesis. Proc. R. Soc. Lond. B 265, 739–745. Arnold, K. E., and Owens, I. P. F. (1999). Cooperative breeding in birds: The role of ecology. Behav. Ecol. 10, 465–471. Barnard, C. J. (2004). ‘‘Animal Behaviour: Mechanism, Development, Function, and Evolution.’’ Pearson Education Limited, Essex. Bathe, G. M., and Bathe, H. V. (1982). Territory size and habitat requirements of the Seychelles brush warbler Acrocephalus (Bebrornis) sechellensis. International Council for Bird Preservation, Internal Report, Cambridge, UK. Beebee, T., and Rowe, G. (2004). ‘‘An Introduction to Molecular Ecology.’’ Oxford University Press, Oxford. Bennett, P. M., and Owens, I. P. F. (2002). ‘‘Evolutionary Ecology of Birds—Life Histories, Mating Systems, and Extinction.’’ Oxford University Press, Oxford. Bensch, S., Hasselquist, D., and von Schantz, T. (1994). Genetic similarity between parents predicts hatching failure: Nonincestuous inbreeding in the great reed warbler. Evolution 48, 317–326. Bijlsma, R., Bundgaard, J., and Boerema, A. C. (2000). Does inbreeding affect the extinction risk of small populations? Predictions from Drosophila. J. Evol. Biol. 13, 502–514. Birkhead, T. R., and Møller, A. P. (1992). ‘‘Sperm Competition in Birds: Evolutionary Causes and Consequences.’’ Academic Press, London. Birkhead, T. R., and Møller, A. P. (1998). ‘‘Sperm Competition and Sexual Selection.’’ Academic Press, London. Bodmer, W. (1972). Evolutionary significance of the HLA sytem. Nature 237, 139–145. Bonneaud, C., Federici, P., Chastel, O., Westerdahl, H., and Sorci, G. (2006). Complex Mhc‐ based mate choice in a wild passerine. Proc. R. Soc. Lond. B 273, 1111–1116. Brook, B. W., Tonkyn, D. W., Q’Grady, J. J., and Frankham, R. (2002). Contribution of inbreeding to extinction risk in threatened species. Cons. Ecol. 6, 16. Brooker, M. G., Rowley, I., Adams, M., and Baverstock, P. R. (1990). Promiscuity: An inbreeding avoidance mechanism in a socially monogamous species? Behav. Ecol. Sociobiol. 26, 191–199. Brouwer, L., Richardson, D. S., Eikenaar, C., and Komdeur, J. (2006). The role of group size and environmental factors on survival in a cooperatively breeding tropical passerine. J. Anim. Ecol. 75, 1321–1329.

180

JAN KOMDEUR AND DAVID S. RICHARDSON

Brown, J. L. (1987). ‘‘Helping and Communal Breeding in Birds.’’ Princeton University Press, Princeton, NJ. Brown, J. L. (1997). A theory of mate choice based on heterozygosity. Behav. Ecol. 8, 60–65. Clarke, M. F. (1984). Cooperative breeding by the Australian bell miner, Manorina melanophrys Latham: A test of kin selection theory. Behav. Ecol. Sociobiol. 14, 137–146. Clarke, M. F. (1989). The pattern of helping in bell miners. Ethology 80, 292–306. Clutton‐Brock, T. H. (2002). Breeding together: Kin selection and mutualism in cooperative vertebrates. Science 296, 69–72. Clutton‐Brock, T. H., Brotherton, P. N. M., O’Riain, M. J., Griffin, A. S., Gaynor, D., Sharpe, L., Kansky, M. B., Manser, M. B., and McIlrath, G. M. (2000). Individual contributions to babysitting in a cooperative mongoose, Suricata suricatta. Proc. R. Soc. Lond. B 267, 301–305. Clutton‐Brock, T. H., Brotherton, P. N. M., O’Riain, M. J., Griffin, A. S., Gaynor, D., Kansky, R., Sharpe, R., and McIrath, G. M. (2001). Contributions to cooperative rearing in meerkats. Anim. Behav. 61, 705–710. Cockburn, A. (1998). Evolution of helping behaviour in cooperatively breeding birds. Annu. Rev. Ecol. Syst. 29, 141–177. Cockburn, A. (2003). Cooperative breeding in oscine passerines: Does sociality inhibit speciation? Proc. R. Soc. B. Lond. 270, 2207–2214. Cockburn, A. (2004). Mating systems and sexual conflict. In ‘‘Ecology and Evolution of Cooperative Breeding in Birds’’ (W. D. Koenig and J. L. Dickinson, Eds.), pp. 81–101. Cambridge University Press, Cambridge. Collar, N. J., and Stuart, S. N. (1985). ‘‘Threatened Birds of Africa and Related Islands.’’ ICBP/ IUCN Red Data Book Part 1. International Council for Bird Preservation, Cambridge, UK. Connor, R. C., and Curry, R. L. (1995). Helping non‐relatives: A role for deceit? Anim. Behav. 49, 389–393. Craig, J. L., and Jamieson, I. G. (1988). Incestuous mating in a communal bird: A family affair. Am. Nat. 131, 58–70. Creel, S., and Creel, N. M. (2002). ‘‘The African Wild Dog. Behavior, Ecology, and Conservation.’’ Princeton University Press, Princeton, NJ. Crnokrak, P., and Roff, D. A. (1999). Inbreeding depression in the wild. Heredity 83, 260–270. Crook, J. (1960). The present status of certain rare land‐birds of the Seychelles Islands. Unnumbered Seychelles Government Bulletin, Department of the Environment, Victoria. Crozier, R. H. (1987). Genetic aspects of kin recognition: Concepts models, and synthesis. In ‘‘Kin Recognition in Animals’’ (D. Fletcher and C. Michener, Eds.), pp. 55–74. Wiley, Chichester, UK. Curry, R. L. (1988). Influence of kinship on helping behaviour in Gala´pagos mockingbirds. Behav. Ecol. Sociobiol. 22, 141–152. Darwin, C. R. (1859). ‘‘On the Origin of Species by Means of Natural Selection or the Preservation of Favoured Races in the Struggle for Life.’’ John Murray, London. Dawkins, R. (1976). ‘‘The Selfish Gene.’’ Oxford University Press, Oxford. Diamond, A. W. (1980). Seasonality, population structure and breeding ecology of the Seychelles brush warbler Acrocephalus sechellensis. Proc. Pan‐Afr. Orn. Congr. 4, 253–266. Dickinson, J. L., and Akre, J. J. (1998). Extra‐pair paternity, inclusive fitness, and within‐group benefits of helping in western bluebirds. Mol. Ecol. 7, 95–105. Dickinson, J. L., and Hatchwell, B. J. (2004). The fitness consequences of helping. In ‘‘Cooperative Breeding in Birds’’ (W. D. Koenig and J. L. Dickinson, Eds.), pp. 48–66. Cambridge University Press, Cambridge.

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

181

Dingemanse, N. J., Both, C., Drent, P. J., van Oers, K., and van Noordwijk, A. J. (2002). Repeatability and heritability of exploratory behaviour in great tits from the field. Anim. Behav. 64, 929–938. Dingemanse, N. J., Both, C., van Noordwijk, A. J., Rutten, A. L., and Drent, P. J. (2003). Natal dispersal and personalities in great tits (Parus major). Proc. R. Soc. Lond. B 270, 741–747. Doherty, P., and Zinkernagel, R. (1975). Enhanced immunological surveillance in mice heterozygous at the h‐2 gene complex. Nature 256, 50–52. Dorak, M. T., Lawson, T., Machulla, H. K. G., Mills, K. I., and Burnett, A. K. (2002). Increased heterozygosity for MHC class ii lineages in newborn males. Genes Immun. 3, 263–269. Drent, P. J., van Oers, K., and van Noordwijk, A. J. (2002). Realized heritability of personalities in the great tit (Parus major). Proc. R. Soc. Lond. B 270, 45–51. Dugatkin, L. A. (1997). ‘‘Cooperation Among Animals: A Modern Perspective.’’ Oxford University Press, Oxford. Dunn, P. O., Cockburn, A., and Mulder, R. A. (1995). Fairy‐wren helpers often care for young to which they are unrelated. Proc. R. Soc. Lond. B 259, 339–343. DuPlessis, M. A. (1993). Helping behaviour in cooperatively‐breeding green wood‐hoopoes: Selected or unselected trait? Behaviour 127, 49–65. Eikenaar, C., Richardson, D. S., Brouwer, L., and Komdeur, J. (2007). Sex biased natal dispersal in a closed, saturated population of Seychelles warblers. J. Avian Biol (in press). Emlen, S. T. (1982). The evolution of helping. I. An ecological constraints model. Am. Nat. 119, 29–39. Emlen, S. T. (1991). Evolution of cooperative breeding in birds and mammals. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), pp. 301–337. Blackwell Scientific Publications, Oxford. Emlen, S. T. (1994). Benefits, constraints and the evolution of the family. Trends Ecol. Evol. 9, 282–285. Emlen, S. T. (1995). An evolutionary theory of the family. Proc. Natl. Acad. Sci. USA 92, 8092–8099. Emlen, S. T. (1997). Predicting family dynamics in social vertebrates. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), pp. 228–253. Blackwell Scientific Publications, Oxford. Emlen, S. T., and Wrege, P. H. (1988). The role of kinship in helping decisions among white‐fronted bee‐eaters. Behav. Ecol. Sociobiol. 23, 305–315. Emlen, S. T., Reeve, H. K., Sherman, P. W., Wrege, P. H., Ratnieks, F. L. W., and Shellman‐ Reeve, J. (1991). Adaptive versus nonadaptive explanations of behaviour: The case of alloparental helping. Am. Nat. 138, 259–270. Foster, K. R., Wenseleers, T., and Ratnieks, F. L. W. (2006). Kin selection is the key to altruism. Trends Ecol. Evol. 21, 57–60. Frankham, R. (1998). Inbreeding and extinction: Island populations. Cons. Biol. 12, 665–675. Frankham, R., Ballou, J. D., and Briscoe, D. A. (2002). ‘‘Introduction to Conservation Genetics.’’ Cambridge University Press, Cambridge. Freeman‐Gallant, C. R., Meguerdichian, M., Wheelwright, N. T., and Sollecito, S. V. (2003). Social pairing and female mating fidelity predicted by restriction fragment length polymorphism similarity at the major histocompatibility complex in a songbird. Mol. Ecol. 12, 3077–3083. Goodnight, K. F., and Queller, D. C. (1999). Computer software for performing likelihood tests of pedigree relationship using genetic markers. Mol. Ecol. 8, 1231–1234. Griffin, A. S., and West, S. A. (2003). Kin discrimination and the benefit of helping in cooperatively breeding vertebrates. Science 302, 634–636.

182

JAN KOMDEUR AND DAVID S. RICHARDSON

Griffiths, R., Double, M. C., Orr, K., and Dawson, R. G. J. (1998). A DNA test to sex most birds. Mol. Ecol. 7, 1071–1075. Griffith, S. C., Owens, I. P. F., and Thuman, K. A. (2002). Extra‐pair paternity in birds: A review of interspecific variation and adaptive function. Mol. Ecol. 11, 2195–2212. Grob, B., Knapp, L. A., Martin, R. D., and Anzenberger, G. (1998). The major histocompatibility complex and mate choice: Inbreeding avoidance and selection of good genes. Exp. Clin. Immunogenet. 15, 119–129. Hadfield, J. D., Richardson, D. S., and Burke, T. (2006). Towards unbiased parentage assignment: Combining genetic, behavioural and spatial data in a Bayesian framework. Mol. Ecol. 15, 3715–3730. Hamilton, W. D. (1964a). The genetical evolution of social behaviour. I. J. Theor. Biol. 7, 1–16. Hamilton, W. D. (1964b). The genetical evolution of social behaviour. II. J. Theor. Biol. 7, 17–52. Hardy, I. C. W. (2002). ‘‘Sex ratios‐Concepts and Research Methods (Ed).’’ Cambridge University Press, Cambridge. Hasselquist, D., Bensch, S., and von Schantz, T. (1996). Correlation between male song repertoire, extra‐pair paternity and offspring survival in the great reed warbler. Nature 381, 229–232. Hatchwell, B. J., and Komdeur, J. (2000). Ecological constraints, life history traits and the evolution of cooperative breeding. Anim. Behav. 59, 1079–1086. Haydock, J., Koenig, W. D., and Stanback, M. T. (2001). Shared parentage and incest avoidance in the cooperatively breeding acorn woodpecker. Mol. Ecol. 10, 1515–1525. Hedrick, P. W. (2002). Pathogen resistance and genetic variation at MHC loci. Evolution 56, 1902–1908. Hedrick, P. W. (2004). Comment on ‘‘Parasite selection for immunogenetic optimality.’’ Science 303, 957a. Heinsohn, R. G., and Legge, S. (1999). The cost of helping. Trends Ecol. Evol. 14, 53–57. Heinsohn, R. G., Cockburn, A., and Mulder, R. A. (1990). Avian cooperative breeding: Old hypotheses and new directions. Trends Ecol. Evol. 5, 403–407. Hughes, A. L., and Yeager, M. (1998). Natural selection at major histocompatibility complex loci of vertebrates. Annu. Rev. Genet. 32, 415–435. IUCN (2004). ‘‘2004 IUCN Red List of Threatened Species,’’ http://www.iucnredlist.org. Jeffreys, A. J., Wilson, V., and Thein, S. L. (1985). Individual‐Specific Fingerprints of Human DNA. Nature 316, 76–79. Jennions, M. D., and Petrie, M. (2000). Why do females mate multiply? A review of the genetic benefits. Biol. Rev. Camb. Philos. Soc. 75, 21–64. Jordan, W., and Bruford, M. (1998). New perspectives on mate choice and the MHC. Heredity 81, 127–133. Kappeler, P. M. B. (2007). Genetic and ecological determinants of primate social systems. In ‘‘Ecology of Social Evolution’’ (J. Korb and J. Heinze, Eds.). Cambridge University Press, Cambridge (in press). Keane, B., Creel, S. R., and Waser, P. M. (1996). No evidence of inbreeding avoidance or inbreeding depression in a social carnivore. Behav. Ecol. 7, 480–489. Keller, F., and Arcese, P. (1998). No evidence for inbreeding avoidance in a natural population of song sparrows (Melospiza melodia). Am. Nat. 152, 380–392. Koenig, W. D., and Haydock, J. (2004). Incest avoidance. In ‘‘Cooperative Breeding in Birds’’ (W. D. Koenig and J. L. Dickinson, Eds.), pp. 142–156. Cambridge University Press, Cambridge. Koenig, W. D., Pitelka, F. A., Carmen, W. J., Mumme, R. L., and Stanback, M. T. (1992). The evolution of delayed dispersal in cooperative breeders. Q. Rev. Biol. 67, 111–150.

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

183

Komdeur, J. (1991). ‘‘Cooperative Breeding in the Seychelles Warbler.’’ University of Cambridge, Cambridge Unpublished PhD thesis. Komdeur, J. (1992). Importance of habitat saturation and territory quality for the evolution of cooperative breeding in the Seychelles warbler. Nature 358, 493–495. Komdeur, J. (1994a). Conserving the Seychelles warbler Acrocephalus sechellensis by translocation from Cousin Island to the islands of Aride and Cousine. Biol. Conserv. 76, 143–152. Komdeur, J. (1994b). Experimental evidence for helping and hindering by previous offspring in the cooperative‐breeding Seychelles warbler Acrocephalus sechellensis. Behav. Ecol. Sociobiol. 34, 175–186. Komdeur, J. (1994c). The effect of kinship on helping in the cooperative breeding Seychelles warbler (Acrocephalus sechellensis). Proc. R. Soc. Lond. B 256, 47–52. Komdeur, J. (1996a). Seasonal timing of reproduction in a tropical bird, the Seychelles warbler: A field experiment using translocation. J. Biol. Rhythms 11, 333–346. Komdeur, J. (1996b). Facultative sex ratio bias in the offspring of Seychelles warblers. Proc. R. Soc. Lond. B 263, 661–666. Komdeur, J. (1996c). Influence of helping and breeding experience on reproductive performance in the Seychelles warbler: A translocation experiment. Behav. Ecol. 7, 326–333. Komdeur, J. (1998). Long‐term fitness benefits of egg sex modification by the Seychelles warbler. Ecol. Lett. 1, 56–62. Komdeur, J. (2003). Daughters on request: About helpers and egg sexes in the Seychelles warbler. Proc. R. Soc. London B 270, 3–11. Komdeur, J. (2005). No evidence for adaptive suppression of joint laying by dominant female Seychelles warblers: An experimental study. Behaviour 142, 1669–1684. Komdeur, J. (2006). Variation in individual investment strategies among social animals. Ethology 112, 729–747. Komdeur, J., and Edelaar, P. (2001). Male Seychelles warblers use territory budding to maximize lifetime fitness in a saturated environment. Behav. Ecol. 12, 706–715. Komdeur, J., and Hatchwell, B. J. (1999). Kin recognition: Function and mechanism in avian societies. Trends Ecol. Evol. 14, 237–241. Komdeur, J., and Pels, M. D. (2005). Rescue of the Seychelles warbler on Cousin Island, Seychelles: The role of habitat restoration. Biol. Conserv. 124, 305–313. Komdeur, J., Huffstadt, A., Prast, W., Castle, G., Mileto, R., and Wattel, J. (1995). Transfer experiments of Seychelles warblers to new islands: Changes in dispersal and helping behaviour. Anim. Behav. 49, 695–708. Komdeur, J., Daan, S., Tinbergen, J., and Mateman, C. (1997). Extreme adaptive modification in sex ratio of the Seychelles warbler’s eggs. Nature 385, 522–525. Komdeur, J., Piersma, T., Kraaijeveld, K., Kraaijeveld‐Smit, F., and Richardson, D. S. (2004a). Why Seychelles warblers fail to recolonize nearby islands: Unwilling or unable to fly there? IBIS 146, 298–302. Komdeur, J., Burke, T., and Richardson, D. S. (2004b). Experimental evidence that kin discrimination in the Seychelles warbler is based on association and not on genetic relatedness. Proc. R. Soc. Lond. B 271, 963–969. Komdeur, J., Richardson, D. S., and Hatchwell, B. (2007). Kin recognition mechanisms in cooperative breeding systems: Ecological causes and behavioural consequences of variation. In ‘‘Ecology of Social Evolution’’ (J. Korb and J. Heinze, Eds.). Princeton University Press, Princeton, NJ (in press). Krebs, J. R., and Davies, N. B. (1997). ‘‘Behavioural Ecology: An Evolutionary Approach,’’ 4th ed. Blackwell Scientific Publications, Oxford. Kruuk, L. E. B., Sheldon, B. C., and Merila, J. (2002). Severe inbreeding depression in collared flycatchers (Ficedula albicollis). Proc. R. Soc. Lond. B 269, 1581–1589.

184

JAN KOMDEUR AND DAVID S. RICHARDSON

Kurtz, J., Kalbe, M., Aeschlimann, P. B., Haberli, M. A., Wegner, K. M., Reusch, T. B. H., and Milinski, M. (2004). Major histocompatibility complex diversity influences parasite resistance and innate immunity in sticklebacks. Proc. R. Soc. Lond. B 271, 197–204. Leisler, B., Heidrich, P., Schulze‐Hagen, K., and Wink, M. (1997). Taxonomy and phylogeny of reed warblers (genus Acrocephalus) based on mtDNA sequences and morphology. J. Ornith. 138, 469–496. Ligon, D. L. (1983). Cooperation and reciprocity in avian social systems. Am. Nat. 121, 366–384. Ligon, J. D. (1999). ‘‘The Evolution of Avian Breeding Systems.’’ Oxford University Press, Oxford. Loustau‐Lalanne, P. (1968). The Seychelles, Cousin Island nature reserve International Council for Bird Preservation, Cambridge. Internal report. MacColl, A. D. C., and Hatchwell, B. J. (2003). Heritability of parental effort in a passerine bird. Evolution 57, 2191–2195. Magrath, R. D., and Whittingham, L. A. (1997). Subordinate males are more likely to help if unrelated to the breeding female in cooperatively breeding white‐browed scrubwrens. Behav. Ecol. Sociobiol. 41, 185–192. Mateo, J. M. (2006). The nature and representation of individual recognition odours in Belding’s ground squirrels. Anim. Behav. 71, 141–154. Mateo, J. M., and Johnston, R. E. (2000). Kin recognition and the ‘‘armpit effect’’: Evidence of self‐referent phenotype matching. Proc. R. Soc. Lond. B 267, 695–700. Maynard Smith, J. (1964). Group selection and kin selection. Nature 201, 1145–1147. Maynard Smith, J. (1978). ‘‘The Evolution of Sex.’’ Cambridge University Press, Cambridge. Maynard Smith, J., and Szathma´ry, E. (1995). ‘‘The Major Transitions in Evolution.’’ Oxford University Press, Oxford. Mulder, R. A., Dunn, P. O., Cockburn, A., Lazenby‐Cohen, K. A., and Howell, M. J. (1994). Helpers liberate female fairy‐wrens from constraints on extra‐pair mate choice. Proc. R. Soc. Lond. B 255, 223–229. Mullis, K. B., and Faloona, F. A. (1987). Specific synthesis of DNA in vitro via a polymerase‐ catalyzed chain‐reaction. Methods Enzymol. 155, 335–350. Mumme, R. L. (1992a). Delayed dispersal and cooperative breeding in the Seychelles warbler. Trends Ecol. Evol. 7, 330–331. Mumme, R. L. (1992b). Do helpers increase reproductive success? An experimental analysis in the Florida scrub jay. Behav. Ecol. Sociobiol. 31, 319–328. Mumme, R. L., Koenig, W. D., and Ratnieks, F. L. W. (1989). Helping behaviour, reproductive value, and the future component of indirect fitness. Anim. Behav. 38, 331–343. Ober, C., Weitkamp, L. R., Cox, N., Dytch, H., Kostyu, D., and Elias, S. (1997). HLA and mate choice in humans. Am. J. Hum. Genet. 61, 497–504. Penn, D. J., and Potts, W. K. (1999). The evolution of mating preferences and major histocompatibility complex genes. Am. Nat 153, 145–164. Penn, D. J., Damjanovich, K., and Potts, W. (2002). MHC heterozygosity confers a selective advantage against multiple‐strain infections. Proc. Natl. Acad. Sci. USA 99, 11260–11264. Petrie, M. (1994). Improved growth and survival of offspring of peacocks with more elaborate trains. Nature 371, 598–599. Petrie, M., and Møller, A. P. (1991). Laying eggs in others’ nests: Intraspecific brood parasitism in birds. Trends Ecol. Evol. 6, 315–320. Petrie, M., Krupa, A., and Burke, T. (1999). Peacocks lek with relatives even in the absence of social and environmental cues. Nature 401, 155–157. Piertney, S. B., and Oliver, M. K. (2006). The evolutionary ecology of the major histocompatibility complex. Heredity 96, 7–21.

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

185

Potts, W. K., and Slev, P. R. (1995). Pathogen‐based models favoring MHC genetic diversity. Immunol. Rev. 143, 181–197. Potts, W. K., Manning, C. J., and Wakeland, E. K. (1991). Mating patterns in seminatural populations of mice influenced by MHC genotype. Nature 352, 619–621. Pusey, A., and Wolf, M. (1996). Inbreeding avoidance in animals. Trends Ecol. Evol. 11, 201–206. Rabenold, K. N. (1985). Cooperation in breeding by non‐reproductive wrens: Kinship, reciprocity and demography. Behav. Ecol. Sociobiol. 17, 1–17. Reusch, T. B. H., Haberli, M. A., Aeschlimann, P. B., and Milinski, M. (2001). Female sticklebacks count alleles in a strategy of sexual selection explaining MHC polymorphism. Nature 414, 300–302. Reyer, H.‐U. (1980). Flexible helper structure as an ecological adaptation in the pied kingfisher (Ceryle rudis rudis L.). Behav. Ecol. Sociobiol. 6, 219–227. Reyer, H.‐U. (1984). Investment and relatedness: A cost/benefit analysis of breeding and helping in the pied kingfisher (Ceryle rudis). Anim. Behav. 32, 1163–1178. Richardson, D. S., and Westerdahl, H. (2003). MHC diversity in two Acrocephalus species: The outbred great reed warbler and the inbred Seychelles warbler. Mol. Ecol. 12, 3523–3529. Richardson, D. S., Jury, F. L., Dawson, D. A., Salgueiro, P., Komdeur, J., and Burke, T. (2000). Fifty Seychelles warbler (Acrocephalus sechellensis) microsatellite loci polymorphic in Sylviidae species and their cross‐species amplification in other passerine birds. Mol. Ecol. 9, 2226–2231. Richardson, D. S., Jury, F. L., Blaakmeer, K., Komdeur, J., and Burke, T. (2001). Parentage assignment and extra‐group paternity in a cooperative breeder: The Seychelles warbler (Acrocephalus sechellensis). Mol. Ecol. 10, 2263–2273. Richardson, D. S., Burke, T., and Komdeur, J. (2002). Direct benefits and the evolution of female‐biased cooperative breeding in Seychelles warblers. Evolution 56, 2313–2321. Richardson, D. S., Burke, T., and Komdeur, J. (2003a). Sex‐specific associative learning cues and inclusive fitness benefits in the Seychelles warbler. J. Evol. Biol. 16, 854–861. Richardson, D. S., Komdeur, J., and Burke, T. (2003b). Altruism and infidelity among warblers. Nature 422, 580. Richardson, D. S., Komdeur, J., and Burke, T. (2004). Inbreeding in the Seychelles warbler: Environment‐dependent maternal effects. Evolution 58, 2037–2048. Richardson, D. S., Komdeur, J., Burke, T., and von Schantz, T. (2005). MHC‐based patterns of social and extra‐pair mate choice in the Seychelles warbler. Proc. R. Soc. Lond. B 272, 759–767. Rood, J. P. (1990). Group size, survival, reproduction, and routes to breeding in dwarf mongooses. Anim. Behav. 39, 566–572. Rulicke, T., Chapuisat, M., Homberger, F. R., Macas, E., and Wedekind, C. (1998). MHC‐ genotype of progeny influenced by parental infection. Proc. R. Soc. Lond. B 265, 711–716. Russel, A. F. (1999). Ecological constraints and the cooperative breeding system of the longtailed tit (Aegithalos caudatus) University of Sheffield, Sheffield Unpublished PhD thesis. Russel, A. F. (2004). Mammals: Comparisons and contrasts. In ‘‘Cooperative Breeding in Birds’’ (W. D. Koenig and J. L. Dickinson, Eds.), pp. 210–227. Cambridge University Press, Cambridge. Russell, E. M. (1989). Cooperative breeding: A Gondwanan perspective. EMU 89, 61–62. Saccheri, I., Kuussaari, M., Kankare, M., Vikman, P., Fortelius, W., and Hanski, I. (1998). Inbreeding and extinction in a butterfly metapopulation. Nature 392, 491–494. Saiki, R. K., Scharf, S., Faloona, F., Mullis, K., Horn, G., Erlich, H., and Arnheim, N. (1985). Enzymatic amplification of beta‐globin genomic sequences and restriction site analysis for diagnosis of sickle‐cell anemia. Science 230, 1350–1354.

186

JAN KOMDEUR AND DAVID S. RICHARDSON

Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B., and Erlich, H. A. (1988). Primer‐directed enzymatic amplification of DNA with a thermostable DNA‐polymerase. Science 239, 487–491. Sharp, S. P., McGowan, A., Wood, M. J., and Hatchwell, B. J. (2005). Learned kin recognition cues in a social bird. Nature 434, 1127–1130. Sherley, G. H. (1990). Cooperative breeding in riflemen (Acanthisitta chloris): Benefits to parents, offspring and helpers. Behaviour 112, 1–22. Shorey, L., Barnard, C. J., and Hurst, J. L. (1994). Kin‐biased behaviour in male wild house mice: Mixed paternity grouping and group member versus kin discrimination. Ethology 97, 141–160. Sibley, C. G., and Monroe, B. L. (1990). ‘‘Distribution and Taxonomy of Birds of the World.’’ Yale University Press, New Haven, CT. Skutch, A. F. (1935). Helpers at the nest. Auk 52, 257–273. Stacey, P. B., and Koenig, W. (1990). ‘‘Cooperative Breeding in Birds: Long‐term Studies of Ecology and Behaviour.’’ Cambridge University Press, Cambridge. Stacey, P. B., and Ligon, J. D. (1987). Territory quality and dispersal options in the acorn woodpecker, and a challenge to the habitat saturation model of cooperative breeding. Am. Nat. 130, 654–676. Stacey, P. B., and Ligon, J. D. (1991). The benefits of philopatry hypothesis for the evolution of cooperative breeding: Variance in territory quality and group size effects. Am. Nat. 137, 831–846. Stirling, D. G., Reale, D., and Roff, D. A. (2002). Selection, structure and the heritability of behaviour. J. Evol. Biol. 15, 277–289. Stiver, K. A., Dierkes, P., Taborsky, M., and Balshine, S. (2004). Dispersal patterns and status change in a co‐operatively breeding cichlid Neolamprologus pulcher: Evidence from microsatellite analyses and behavioural observations. J. Fish Biol. 65, 91–105. Taborsky, M. (1994). Sneakers, satellites, and helpers: Parasitic and cooperative behavior in fish reproduction. Adv. Study Behav. 23, 1–100. Tregenza, T., and Wedell, N. (2000). Genetic compatibility, mate choice and patterns of parentage. Mol. Ecol. 9, 1013–1027. Trivers, R. (1985). ‘‘Social Evolution.’’ Benjamin/Cummins, Menlo Park. van de Crommenacker, J., Richardson, D. S., Groothuis, T. G. G., Eising, C. M., Dekker, A. L., and Komdeur, J. (2004). Testosterone, cuckoldry risk and extra‐group opportunities in the Seychelles warbler. Proc. R. Soc. London B 271, 1023–1031. Vesey‐Fitzgerald, L. D. E. F. (1940). The birds of the Seychelles. 1. The endemic birds. IBIS 14, 480–489. von Schantz, T., Wittzell, H., Goransson, G., Grahn, M., and Persson, K. (1996). MHC genotype and male ornamentation: Genetic evidence for the Hamilton‐Zuk model. Proc. R. Soc. Lond. B 263, 265–271. von Schantz, T., Wittzell, H., Goransson, G., and Grahn, M. (1997). Mate choice, male condition‐dependent ornamentation and MHC in the pheasant. Hereditas 127, 133–140. Walters, J. R., Copeyon, C. K., and Carter, J. H. (1992). Test of the ecological basis of cooperative breeding in red‐cockaded woodpeckers. Auk 109, 90–97. Wedekind, C., and Furi, S. (1997). Body odour preferences in men and women: Do they aim for specific MHC combinations or simply heterozygosity? Proc. R. Soc. Lond. B 264, 1471–1479. Wedekind, C., Chapuisat, M., Macas, E., and Rulicke, T. (1996). Non‐random fertilization in mice correlates with the MHC and something else. Heredity 77, 400–409.

THE HIDDEN COMPLEXITIES OF THE SEYCHELLES WARBLER

187

Wedekind, C., Walker, M., Portmann, J., Cenni, B., Muller, R., and Binz, T. (2004). MHC‐ linked susceptibility to a bacterial infection, but no MHC‐linked cryptic female choice in whitefish. J. Evol. Biol. 17, 11–18. Wegner, K. M., Kalbe, M., Kurtz, J., Reusch, T. B. H., and Milinski, M. (2003). Parasite selection for immunogenetic optimality. Science 301, 1343–1345. Wegner, K. M., Kalbe, M., Kurtz, J., Reusch, T. B. H., and Milinski, M. (2004). Response to comment on ‘‘Parasite selection for immunogenetic optimality.’’ Science 303, 957b. West, S. A., and Sheldon, B. C. (2002). Constraints in the evolution of sex‐ratio adjustment. Science 295, 1685–1688. Westerdahl, H. (2004). No evidence of an MHC‐based female mating preference based on genetic compatibility in the great reed warbler. Mol. Ecol. 13, 2465–2470. Westneat, D. F., Sherman, P. W., and Morton, M. L. (1990). The ecology and evolution of extra‐pair copulations in birds. In ‘‘Current Ornithology’’ (D. M. Power, Ed.), Vol. 7, pp. 331–369. Plenum Press, New York, NY. Whittingham, L. A., Dunn, P. O., and Magrath, R. D. (1997). Relatedness, polyandry and extra‐group paternity in the cooperatively‐breeding white‐browed scrubwren (Sericornis frontalis). Behav. Ecol. Sociobiol. 40, 261–270. Wilkinson, G. S., Presgraves, D. C., and Crymes, L. (1998). Male eye span in stalk‐eyed flies indicates genetic quality by meiotic drive suppression. Nature 391, 276–278. Wilson, E. O. (1975). ‘‘Sociobiology: The New Synthesis.’’ Harvard University Press, Cambridge. Wilson, E. O. (2005). Kin selection as the key to altruism: Its rise and fall. Soc. Res. 72, 159–166. Wilson, E. O., and Holldobler, B. (2005). Eusociality: Origin and consequences. Proc. Natl. Acad. Sci. USA 102, 13367–13371. Yamazaki, K., Beauchamp, G. K., Kupniewski, D., Bard, J., Thomas, L., and Boyse, E. A. (1988). Familial imprinting determines H‐2 selective mating preferences. Science 240, 1331–1332. Yom‐Tov, Y. (1980). Intraspecific nest parasitism in birds. Biol. Rev. 55, 93–108. Zack, S. (1990). Coupling delayed breeding with short‐distance dispersal in cooperatively breeding birds. Ethology 86, 265–286. Zahavi, A. (1990). Arabian babblers: The quest for social status in a cooperative breeder. In ‘‘Cooperative Breeding in Birds’’ (P. B. Stacey and W. D. Koenig, Eds.), pp. 105–130. Cambridge University Press, Cambridge. Zeh, J. A., and Zeh, D. W. (1996). The evolution of polyandry I: Intragenomic conflict and genetic incompatibility. Proc. R. Soc. Lond. B 263, 1711–1717.

This page intentionally left blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

Mate Choice and Genetic Quality: A Review of the Heterozygosity Theory Bart Kempenaers department of behavioural ecology and evolutionary genetics, max planck institute for ornithology, postfach 1564, d‐82305 starnberg (seewiesen), germany

‘‘To expect selection to produce mate choices which take account of the combined result of a female’s genotype and that of her suitor is indeed to have faith’’ (Parker, 1992, p. 395) ‘‘This is an overly pessimistic view’’ (Jennions and Petrie, 2000, p. 37)

I. INTRODUCTION Mate choice has been and still is a very popular research topic in behavioral and evolutionary ecology. This is not surprising, given the role it plays in our own lives, and given its importance for understanding phenomena ranging from sperm form and function (Holman and Snook, 2006) to speciation (McPeek and Gavrilets, 2006). Research on mate choice beautifully illustrates how theoretical work in the form of verbal and mathematical models develops alongside simple and more sophisticated empirical studies (e.g., Gustafsson and Qvarnstro¨m, 2006; Kokko et al., 2006; Qvarnstro¨m et al., 2006). The study of mate choice has also created its share of scientific debate and controversy, as can be seen in the literature, and observed at scientific conferences. As a graduate student, I witnessed (from a safe distance) some surprisingly emotional exchanges about why females paired with an already mated male, and later I myself got caught up in discussions about whether and how females benefit from engaging in extra‐pair copulations. A lively research field it is indeed.

0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37005-8

189

Copyright 2007, Elsevier Inc. All rights reserved.

190

BART KEMPENAERS

Arguably the most controversial issue of all is the importance of indirect genetic benefits for the evolution of mate choice. The key questions are these. Are genetic benefits important enough to explain the evolution of more or less sophisticated (and costly) choice mechanisms? What is the nature of these genetic benefits? The main aim of this review is to consider a specific type of genetic benefit, namely the fitness gains resulting from the production of offspring with higher individual genetic diversity (heterozygosity). Research on inbreeding and on the relationship between individual heterozygosity and fitness‐related traits in free‐living populations has blossomed, particularly after molecular tools such as microsatellite markers became widely available. However, we are only beginning to understand why individual heterozygosity is related to fitness and how this affects the evolution of mate choice. To set the stage for this review, I briefly define mate choice, and discuss some of the key issues about how individuals can benefit from being choosy, including the distinction between two main types of genetic benefits (Section II). The next section focuses on heterozygosity and fitness. Here, you will find an overview of the methods that have been used to estimate individual heterozygosity or relatedness between individuals, and a summary of an ongoing debate about the importance and interpretation of heterozygosity–fitness correlations. Section III also contains a review of studies that have found correlations between individual heterozygosity and a variety of fitness‐related traits. Section IV is about mate choice and heterozygosity and consists of two parts. In the first part, I discuss mate choice based on relatedness with the partner, that is, choice to optimize offspring heterozygosity. I consider when it will pay females to mate with a more or less‐related male, and how females should trade off choice for different types of genetic benefits. Then, I emphasize the link between the evolution of promiscuity and inbreeding avoidance, and I discuss why it is important to consider the costs and mechanisms of choice. Finally, I review the evidence that social mate choice, choice of a copulation partner, and choice of sperm under multiple mating depends on relatedness. The second part of Section IV is about mate choice favoring heterozygous partners. First, I discuss why one should expect such choice and why the heterozygosity of the choosing female may affect her choice. Then, I review the evidence that individuals prefer heterozygous mates, in different contexts. This review is probably biased toward studies on birds, because those are the studies I am most familiar with. However, I attempted to review the entire recent literature, and the fact that many examples are from birds also reflects the relative amount of research on avian mate choice.

MATE CHOICE AND GENETIC QUALITY

191

II. MATE CHOICE AND GENETIC BENEFITS A. WHAT IS MATE CHOICE? There is still considerable debate about the existence and relative importance of the different mechanisms of choice, and particularly about how to define and demonstrate ‘‘cryptic choice’’ (e.g., Birkhead, 1998, 2000; Birkhead and Pizzari, 2002; Bussie`re et al., 2006; Kempenaers et al., 2000; Pilastro et al., 2004; Pitnick and Brown, 2000; Telford and Jennions, 1998). However, in general, three types of choice can be distinguished. 1. Choice of a social partner or a copulation partner (precopulatory, prefertilization). This is what we commonly understand by ‘‘mate choice,’’ that is, the process involved in selecting a mate. 2. Choice of sperm to fertilize an egg. This is a postcopulatory process, also referred to as ‘‘cryptic’’ choice, which can involve both behavioral and physiological mechanisms (Eberhard, 1996). 3. Choice to invest differentially in certain zygotes, or offspring, depending on characteristics of the mate (postcopulatory, postfertilization). This is known as differential allocation (Sheldon, 2000). It is not usually considered mate choice, but it is obviously closely linked to it. Much of what follows will be about ‘‘female choice’’ rather than ‘‘mate choice.’’ Theory suggests that male choice or mutual mate choice can also evolve, and such choice has been empirically demonstrated (Johnstone et al., 1996; Servedio and Lande, 2006). However, the focus on females reflects both the fact that females are generally expected to be choosier, and the lack of data on male choice (Andersson, 1994).

B. HOW DO FEMALES BENEFIT FROM THEIR CHOICE? In theory, choosy females can obtain two types of benefits (reviewed in Andersson, 1994). (1) Direct benefits arise when females increase their fitness because males provide better resources for them or their offspring (e.g., nest site, food, other forms of parental care). (2) Indirect or genetic benefits accrue when females increase their fitness through an increase in the genetic quality of their offspring. Although this seems straightforward, many fundamental issues remain unresolved, despite decades of study (Andersson and Simmons, 2006; Kokko et al., 2003, 2006). The question can be broken down further, as follows.

192

BART KEMPENAERS

1. What is the Relative Importance of Direct (Resources) Versus Indirect (Genetic) Benefits? In resource‐based mating systems, it is often hard to exclude that females also obtain genetic benefits. First, females that chose a male based on direct benefits (i.e., a male who provides better resources) might simultaneously have chosen a male of higher genetic quality (Kokko et al., 2003). If this quality is transferred to the offspring, females will also benefit indirectly. Second, females that have chosen a social mate for direct benefits might additionally obtain genetic benefits through their choice of an extra‐pair male (Jennions and Petrie, 2000). In nonresource‐based mating systems, on the other hand, it is often difficult to prove that there are no direct benefits. As Jennions and Petrie (2000) pointed out (p. 28) ‘‘Sometimes it is the absence of evidence, rather than the reverse, which leads to the claim that females remate for genetic benefits. When no material benefits can be found, genetic benefits are assumed by default.’’ 2. Are Indirect (Genetic) Benefits Large Enough to Explain the Evolution of Mate Choice? It has often been asserted that indirect benefits are too small to compensate for the cost of mate choice (Kirkpatrick and Barton, 1997). The theoretical argument is that if a trait is subject to directional natural or sexual selection, additive genetic variation in the trait will be reduced to a point where it does not pay anymore to choose (the so‐called ‘‘lek paradox’’). The indirect benefits are further eroded if mate choice is inaccurate and if viability indicators are unreliable (Kirkpatrick and Barton, 1997). This view is supported by recent studies that used quantitative genetics to estimate the importance of indirect selection, either within a species (Hadfield et al., 2006; Qvarnstro¨m et al., 2006), or in comparative analyses using data on extra‐pair paternity in birds (Arnqvist and Kirkpatrick, 2005). The latter study showed that the direct fitness cost of reduced paternal care outweighed the indirect fitness gain through extra‐pair offspring on average by a factor 10. These studies led Charmantier and Sheldon (2006) to conclude that ‘‘it is perhaps time to consider mechanisms other than indirect genetic benefits as driving the operation of female choice in these species and mating arenas.’’ To be fair, Charmantier and Sheldon (2006) also wrote that ‘‘it might be premature to reject indirect benefits completely,’’ when discussing the limitations of the above studies. For example, indirect benefits are usually estimated in terms of offspring survival or condition (e.g., comparing within‐ and extra‐pair offspring, as in Arnqvist and Kirkpatrick, 2005), but multigenerational fitness effects are rarely assessed

MATE CHOICE AND GENETIC QUALITY

193

(for a notable exception concerning extra‐pair paternity in birds, see Schmoll et al., 2003, 2005a). The importance of estimating long‐term fitness consequences is beautifully illustrated by a recent study on crickets, showing that the indirect benefits of increasing attractiveness of sons can outweigh substantial direct survival costs (Head et al., 2005). Furthermore, when estimates of the magnitude of direct and indirect fitness benefits are compared using meta‐analysis, it is evident that direct benefits are not always more important (Møller and Jennions, 2001). 3. What is the Nature of the Indirect Genetic Benefits? Trivers (1972) coined the term ‘‘good genes’’ and discussed three forms of genetic quality: the ability of genes to survive, the reproductive ability of genes, and the complementarity of genes. Traditional ‘‘good genes’’ models of sexual selection distinguished between a Fisherian run away process, where the benefit lies in producing more attractive offspring, and a Zahavian handicap process, where offspring viability is improved (reviewed in Andersson, 1994). The question whether both processes are fundamentally different or should be lumped in the Fisher–Zahavi process has been the focus of some recent debate (see Kokko et al., 2006, and references therein). An important issue for mate choice was raised in the late 1990s. Jeanne and David Zeh published two highly influential papers discussing how polyandry might have evolved as a mechanism to avoid negative interactions between genetic elements that cause embryo mortality or lower viability (genetic incompatibility, Zeh and Zeh, 1996, 1997; see also Olsson et al., 1996). Around the same time, Jerram Brown published his ‘‘heterozygosity theory’’ (Brown, 1997, 1999), with two key insights. First, he suggested that ‘‘what is best for one female may not be best for another.’’ This will be the case whenever offspring fitness depends on an interaction between the maternal and paternal haplotypes, so that gene effects will not be strictly additive (Zeh and Zeh, 1996, 1997). This hypothesis is now usually referred to as the genetic compatibility hypothesis. Trivers (1972) first mentioned the idea of genetic complementarity, but the hypothesis only gained widespread empirical and theoretical attention after the publication of the three above‐mentioned papers (for reviews see Jennions and Petrie, 2000; Mays and Hill, 2004; Neff and Pitcher, 2005; Tregenza and Wedell, 2000; Zeh and Zeh, 2001). Brown’s second, related insight was that some males might be superior not because they carry the classical good genes, but rather because they are heterozygous at one or more loci, implying that male quality is not simply heritable. Whether the fitness effects of the alleles are additive or nonadditive is crucial to our understanding of the evolution of mate choice, as will become clear later.

194

BART KEMPENAERS

C. GENETIC BENEFITS: DEFINITIONS Most empirical and review papers discuss genetic benefits of mate choice by contrasting ‘‘good genes’’ and ‘‘compatible genes’’ (see reviews by Jennions and Petrie, 2000; Mays and Hill, 2004; Neff and Pitcher, 2005; see also Puurtinen et al., 2005). Although this has some heuristic value, the problem is that these terms are often loosely defined or not defined at all, and that they are used in a variety of contexts (Pia´lek and Albrecht, 2005). ‘‘Good genes’’ are often thought of as genes that increase the viability or attractiveness of offspring, whereas the term ‘‘compatible genes’’ refers to male genes that are a particularly good match with the female genes. I deliberately write it this way (and so have many others), but the formulation is a bit unfortunate. What we usually mean is not genes, but variants of genes, that is, alleles. The idea is that the presence of a particular allele can be beneficial by itself (the allele has an additive effect), or in combination with another allele (a nonadditive effect; see e.g., Neff and Pitcher, 2005, p. 22). However, the more correct terms ‘‘good allele’’ and ‘‘compatible allele’’ are never used in behavioral ecology. A second problem is that the genetic mechanisms that lead to the benefit are often not specified. Consider a recent study on wild turkeys (Meleagris gallopavo) addressing the question whether females that are infected by a parasite should be more or less choosy than uninfected females (Buchholz, 2004). Previous studies had emphasized the cost of mate sampling, and predicted (and found) that infected females are less choosy, presumably because they are in worse condition (e.g., Pfennig and Tinsley, 2002; Poulin, 1994). However, Buchholz (2004) had the insight that the opposite prediction can be made. Indeed, infected females should benefit more from carefully selecting a male if this increases the probability to produce disease‐resistant offspring. Assuming that females can modify their decision rules based on their perceived susceptibility to disease, diseased females should be more choosy (which is indeed what Buchholz, 2004 found). Is this an example of selection for good alleles or compatible alleles? Buchholz (2004) uses the terms ‘‘complementary genes,’’ ‘‘good genes,’’ and ‘‘genetic compatibility,’’ without further explanation. To answer the question, let us reflect on two scenarios about how the genetic benefits of choice could arise (see Penn and Potts, 1999). Resistance against infectious disease is influenced by the major histocompatibility complex (Mhc), which is the most variable set of functional genes known in vertebrates (Piertney and Oliver, 2006). The first hypothesis is that Mhc diversity per se is responsible for rapid parasite clearance. In that case, a female would benefit most from selecting a male with complementary Mhc alleles (typically referred to as genetic compatibility; e.g., Brown, 1997, 1999). The fitness

MATE CHOICE AND GENETIC QUALITY

195

effect of an allele is then nonadditive. The second hypothesis is that a specific Mhc allele confers resistance to the disease. In that case, a female will benefit most by selecting the male that carries this ‘‘good’’ resistance allele. In this case, the fitness effect of obtaining the allele is additive. In the majority of studies on the genetic benefits of female choice, including the wild turkey study above, the specific allele or allele combinations that lead to disease resistance are unknown. The general point is that if the genetic mechanism behind the increase or decrease in the fitness‐related trait is understood, it is much easier to make predictions about optimal female choice. Is there a link between the type of genetic benefit gained and the female choice mechanism to obtain it (see Fig. 1 in Neff and Pitcher, 2005)? The term ‘‘good genes’’ is often used in the context of models of sexual selection to explain why females prefer males with exaggerated ornaments, leading to directional selection on the ornament (e.g., Colegrave et al., 2002). Such a definition implies that the type of benefit (‘‘good genes’’) is linked to the mechanism (the use of indicator traits) and that all females should prefer the same male. However, good alleles can also be obtained via different mechanisms, for example, in the absence of male ornaments (e.g., Byers and Waits, 2006), or through postcopulatory sperm competition (e.g., the ‘‘sexually selected sperm hypothesis.’’ Keller and Reeve, 1995; Pizzari and Birkhead, 2002). In contrast, even if females prefer the most ornamented males, this does not exclude a ‘‘compatible gene’’ benefit. Indeed, if heterozygous males have superior ornaments, female choice for such males might evolve (at least under some circumstances) because of the benefit of producing more heterozygous offspring (Hoffman et al., 2007; Lehmann et al., 2006; Mitton et al., 1993; see discussion below). Regarding compatibility, it is typically assumed that a pre‐ or postcopulatory mechanism exists by which females can assess their mate’s genetic constitution relative to their own (e.g., Colegrave et al., 2002), whereas under the ‘‘good gene’’ hypothesis no such mechanism is considered. I argue that we should use the terms ‘‘good allele’’ and ‘‘compatible allele’’ in a more narrow sense. We can then consider several questions in turn: 1. What would the optimal mate (genetically speaking) for a particular female be, under particular ecological circumstances, and is this due to additive or nonadditive effects? In this review, I discuss optimal mate choice in the context of the relationship between measures of individual heterozygosity or parental relatedness and fitness. 2. Is there evidence that females choose mates accordingly? Here, I review and discuss studies that have tested mate choice for genetically dissimilar males and choice for heterozygous individuals.

196

BART KEMPENAERS

3. What are the mechanisms of choice? I briefly discuss why it is important to consider mechanisms in order to understand patterns of choice with respect to individual heterozygosity (see Section IV). In short, there is a need for clear‐cut and generally accepted definitions. I here adopt the definitions proposed by Neff and Pitcher (2005) and by Puurtinen et al. (2005), with a few modifications (see also Hunt et al., 2004; Pia´lek and Albrecht, 2005). 1. Genetic Quality Genetic quality is defined as the contribution of an allele or a genotype (both alleles) to an individual’s fitness (lifetime reproductive success, LRS). The genetic quality of an individual is defined relative to that of individuals with other alleles or genotypes. This definition is useful, because it encompasses effects of both additive (good allele) and nonadditive (compatible allele) effects. Whether female choice for indirect benefits evolves will depend on the overall effect of the choice on the genetic quality of the offspring, and should be independent of the type of genetic benefit. Hunt et al. (2004) defined genetic quality of an individual as the sum of the additive effects of an individual’s genes on its fitness. This is only useful when you are interested exclusively in the long‐term heritable component of genetic quality, or when nonadditive effects on fitness are negligible or nonexistent, which seems unlikely. 2. Good Alleles A good allele is defined as an allele that increases fitness independent of the architecture of the rest of the genome (additive effect). It is important to be explicit about whether you assume that all females in the population are the same, that is, use the same choice rules (as is usually the case in population or quantitative genetic models), or whether choice depends on the genetic makeup of the individual female. In the latter case, only an allele with an additive effect should be considered a good allele. Under complete dominance, it is beneficial for an individual to carry at least one copy of the good allele A1 (the fitness of A1A1 equals the fitness of A1A2). Thus, a homozygous A1A1 female cannot increase the fitness of her offspring by being choosy, unless one considers effects on grandchildren (everything else being equal, A1A2 offspring would have to be more choosy to avoid producing offspring that lack the A1 allele). In other words, the benefit of choice depends on the female’s own genetic makeup. At the population genetic level, any fitness‐related trait that shows additive genetic variance will confer a fitness benefit to the offspring. Even if a beneficial allele has a purely dominant effect, females choosing

MATE CHOICE AND GENETIC QUALITY

197

males with the (largest expression of the) trait (either A1A1 or A1A2 males) will on average do better than nonchoosing females. Female choice for good alleles thus causes directional selection and—under constant environmental conditions—the allele would typically go to fixation. Note that the definition makes no implicit assumptions about the traits that are involved. A good allele can be an allele that increases survival (e.g., through disease resistance), competitiveness (e.g., through general vigor or strength), fertility (e.g., sperm quality), or attractiveness to females. The benefit (increased fitness) can be expressed in both sexes, in males only, or in females only. The gene can be neutral in the other sex, or it can even decrease fitness of the opposite sex (sexually antagonistic alleles; Rice, 1996). In the latter case, female choice for the genetic benefit might evolve, for example, when it goes hand in hand with adaptive sex allocation, or not evolve, depending on the relative costs and benefits to daughters and sons. Further note that selection on females to obtain good alleles for their offspring (i.e., an allele with a beneficial effect) is similar to selection on females to avoid bad alleles (i.e., a disadvantageous allele, e.g., resulting from a deleterious mutation). 3. Compatible Alleles A compatible allele is defined as an allele that increases fitness in a specific genetic context, either when paired to a particular homologue (overdominance, i.e., heterozygote advantage), or when paired to an allele at another locus (epistasis, i.e., gene–gene interactions). The trait involved therefore shows nonadditive genetic variance. Note that choice for compatible alleles also causes directional selection, not on the trait, but on the mechanism that allows females to acquire the compatible alleles. A compatible allele will never go to fixation as a result of intersexual selection, because it is only beneficial in combination with another allele. Again, the definition does not make assumptions about the type of traits that are involved. Like good alleles, compatible alleles can affect survival, competitiveness, fertility, or attractiveness in the carrier. Just as there will be selection on females to acquire compatible alleles, there will also be selection to avoid incompatible alleles. The most obvious example is avoidance of inbreeding, because of the increased probability that recessive deleterious alleles are expressed (Keller and Waller, 2002). There are many examples of other mechanisms that lead to genetic incompatibility, such as agents that cause intra‐ or intergenomic conflict (for reviews and examples of these mechanisms see Tregenza and Wedell, 2000; Zeh and Zeh, 2001).

198

BART KEMPENAERS

Pia´lek and Albrecht (2005) suggested use of the term genetic complementarity when referring to benefits of heterozygosity, and the term genetic (in)compatibility when referring to the special case of sterility or inviability of hybrids as a result of epistatic interactions between different genomes. I suggest a hierarchical approach instead: first distinguish between additive (good allele) or nonadditive (compatible allele) effects, and then—if known—specify the mechanism (e.g., heterozygote advantage, genome incompatibility).

III. HETEROZYGOSITY AND FITNESS The heterozygosity theory suggests that individuals benefit from choosing a mate that will maximize heterozygosity at some or many loci in the offspring (Brown, 1997). Such benefit would arise if a positive correlation between individual heterozygosity and fitness exists. A special case of heterozygote advantage is referred to as hybrid vigor or heterosis. This stems from the observation that offspring from crosses between two breeding lines (e.g., in agriculture or animal breeding) often show higher fitness (Lynch and Walsh, 1998), probably because of increased heterozygosity in the offspring. Often the opposite, that is, the negative fitness consequences of homozygosity, is emphasized, as in inbreeding depression (Lynch and Walsh, 1998). Either way, there is a vast amount of literature showing evidence for inbreeding depression, and for a positive relationship between individual heterozygosity at specific (e.g., Mhc) or genome‐wide loci and fitness‐related traits. However, there is also much skepticism about heterozygosity–fitness correlations. The generality and magnitude of the effect has been repeatedly questioned, and there is an ongoing debate about whether the correlation reflects inbreeding or something else. Below, I discuss these issues in detail.

A. METHODS TO ESTIMATE INDIVIDUAL HETEROZYGOSITY, INBREEDING AND RELATEDNESS Table I summarizes the most common estimators that have been proposed to describe either the level of individual heterozygosity, the level of inbreeding, or the relatedness between parents. They are grouped in five categories based on the type of data (method) used: allozymes, multilocus minisatellite markers, microsatellite markers, Mhc markers, and pedigree data. Discussing the pros and cons of each of these estimators is beyond the scope of this chapter.

MATE CHOICE AND GENETIC QUALITY

199

TABLE I METHODS AND ESTIMATORS OF INDIVIDUAL HETEROZYGOSITY OR PARENTAL RELATEDNESS Method Allozymes Multilocus minisatellite markers Single‐locus microsatellite markers

Multilocus Mhc markers Single‐locus Mhc markers

Pedigrees a

Estimator

Referencesa

Number or proportion of heterozygous loci, Hb Homozygosity index, H

1– 4 5

Band‐sharing coefficient, D Number or proportion of heterozygous loci, Hb

6–8 9, 10

Standardized heterozygosity, SHc Internal relatedness, IRd Homozygosity by loci, HL Mean d2, MD2e Relatedness, Rf Paternal relatedness, IPR Possession of rare allele(s) Inbreeding coefficient (Ritland’s estimator), F Band‐sharing coefficient, D Allele diversity (number of bands), N Number or proportion of heterozygous loci, Hb Number of shared alleles, N Genotypic distance Possession of rare allele(s) Inbreeding coefficient, F

9, 10 11 12 10, 13 14, 15 16 17 18, 19 20 20, 21 22 23, 24 25 26 27, 28

The study that first described the method, or examples of studies that have used the method to estimate heterozygosity–fitness correlations. b SLH, single‐locus heterozygosity, or MLH, multilocus heterozygosity; sometimes individuals are simply classified as homozygous or heterozygous (at one or more loci). c The ratio of the heterozygosity of an individual to the mean heterozygosity of the loci at which the individual is typed (reference 9); avoids bias due to individuals being typed at different sets of loci. d Measure of heterozygosity based on allele sharing that takes allele frequencies at each locus into account (sharing of rare alleles weighs more than sharing of common alleles, based on reference 14; see reference 11). e The squared difference in repeat units between alleles at a microsatellite locus, averaged over all loci (reference 10); also used are standardized mean d2, and restricted mean d2 (reference 13); note that the use of this estimator has been criticized (Goudet and Keller, 2002; Hedrick et al., 2001; Tsitrone et al., 2001; see also reference 10). Mean d2 seems a better predictor of fitness than multilocus heterozygosity only if two large, divergent populations have mixed. f There are many ways to calculate relatedness (reference 15), but the most frequently used is described in reference 14. 1. Zouros et al., 1988; 2. Weatherhead et al., 1999; 3. Carchini et al., 2001; 4. Kark et al., 2001; 5. Aparicio et al., 2001; 6. Bensch et al., 1994; 7. Kempenaers et al., 1996; 8. Schmoll et al., 2005b; 9. Coltman et al., 1999; 10. Slate and Pemberton, 2002; 11. Amos et al., 2001; 12. Aparicio et al., 2006; 13. Coulson et al., 1998; 14. Queller and Goodnight, 1989; 15. Van de Casteele et al., 2001; 16. Summers and Amos, 1996; 17. Masters et al., 2003; 18. Ritland, 1996; 19. Lens et al., 2000; 20. Richardson et al., 2005; 21. Hansson et al., 2004a; 22. Sauermann et al., 2001; 23. Roberts et al., 2005b; 24. Sommer, 2005; 25. Landry et al., 2001; 26. Thornhill et al., 2003; 27. Kruuk et al., 2002; 28. Reid et al., 2003.

200

BART KEMPENAERS

B. THE MAGNITUDE OF HETEROZYGOSITY–FITNESS CORRELATIONS Even though many studies have reported significant correlations between marker‐based estimates of individual heterozygosity and a variety of fitness‐related traits (Table II), the generality of the effect and its biological significance have often been questioned (reviewed in Coltman and Slate, 2003; David, 1998; Hansson and Westerberg, 2002; Lynch and Walsh, 1998). First, there is evidence of publication bias in favor of significant results (e.g., Coltman and Slate, 2003). Furthermore, even when significant, the average effect size is usually small. Meta‐analyses or reviews reported a mean r of only .13 for growth rate, and .17 for fluctuating asymmetry (Britten, 1996), r ¼ .10–.22 (David, 1998), r ¼ .09 for fluctuating asymmetry (Vøllestad et al., 1999), mean r < .10 for a variety of traits (Coltman and Slate, 2003). In contrast, there seems to be little doubt about the biological significance of inbreeding depression (Lynch and Walsh, 1998, p. 269), also in natural populations (Keller and Waller, 2002; Table II). Note that effect sizes reported here for heterozygosity–fitness correlations are only slightly lower than the average effect size in a large sample of ecological and evolutionary studies (r ¼ .18, Møller and Jennions, 2002). Coltman and Slate (2003) noted that effect sizes are highly variable between studies, and it is a major challenge to understand this variation (see below). To predict or understand patterns of mate choice, it therefore remains necessary to clarify whether heterozygosity affects fitness‐related traits in a particular population, during a particular time period. The caveat is that it is hard to interpret nonsignificant effects because low expected r‐values increase the probability of Type II errors. Several suggestions have been made to avoid Type II errors: (1) increasing the number of individuals (N), (2) increasing the number of loci (L), and (3) using more variable loci (Smith et al., 2005). Coltman and Slate (2003) suggested that a minimum of 600 individuals are needed to obtain sufficient power. Slate and Pemberton (2002) suggested that at least 10,000 genotypes (¼N
L) should be generated; their data indicated that the variability of the loci was not important, a conclusion that was confirmed by Lieutenant‐Gosselin and Bernatchez (2006). In contrast, Hansson et al. (2004b) found that the strength of the association between heterozygosity at a locus and survival correlated positively with the number of alleles or with the expected heterozygosity at that locus. The opposite problem, the risk of a Type I error, should also be considered when many estimators of heterozygosity are tested for associations with many fitness‐related traits. However, our database suggests that a surprisingly high proportion of studies in a wide variety of species show

TABLE II TRAITS AFFECTED BY INDIVIDUAL HETEROZYGOSITY OR PARENTAL RELATEDNESS IN NATURAL OR SEMINATURAL POPULATIONSa References by method Trait category Survival

201

Health

Development Body size Reproduction

Status Sexually selected signals

Trait Embryo survival (egg hatchability) Offspring survival (recruitment) Adult survival (age, lifespan) Disease severity Parasite load Immune response Fluctuating asymmetry Offspring size Adult size Sperm abnormality Testes mass Laying date Territorial status Territory size Male ornament size Male song characteristics

Allozyme

Multilocus minisatellite

Microsatellite

Pedigree‐based

5, 6, 7b

10

40–42

8

10–21

17, 40, 42, 43

16, 46c, 47

17, 22–26

40, 43

48b,d

8 1, 2

3

27 22, 28, 29 27 30, 31b 11, 12 24, 25 32 32 33e 25, 26, 34 35

Mhc

47, 49b, 50b–52 44, 45 42

48b,d

41

9 35, 36

45 (Continued)

TABLE II (Continued) References by method Trait category Reproductive success

a

Trait Male mating success Male RS Male LRS Clutch size Female (L)RS

Allozyme 2, 4

Multilocus minisatellite

Microsatellite

Pedigree‐based

Mhc

9

34

41

53

9b

23, 31b, 35, 37, 38 21, 24, 39 23, 33e 21, 37

b

54

40

The list is not exhaustive. For additional examples, see Keller and Waller (2002), Table I, p. 236, and Piertney and Oliver (2006) for Mhc genes. Quadratic relationship. c Homozygote deficiency in 1‐year‐old individuals suggests survival advantage of heterozygotes. d Body size and age are strongly correlated. e Heterozygosity calculated based on a mix of four allozyme and eight microsatellite loci. All relationships between heterozygosity and fitness‐related traits are positive and linear, unless otherwise indicated. LRS ¼ lifetime reproductive success. 1. Zouros et al., 1988; 2. Carchini et al., 2001; 3. Weatherhead et al., 1999; 4. Rola´n‐Alvarez et al., 1995; 5. Bensch et al., 1994; 6. Kempenaers et al., 1996; 7. Cordero et al., 2004; 8. Stockley et al., 1993; 9. Aparicio et al., 2001; 10. Van de Casteele et al., 2003; 11. Coltman et al., 1998; 12. Coulson et al., 1998; 13. Hansson et al., 2001; 14. Rossiter et al., 2001; 15. Bean et al., 2004; 16. Hansson et al., 2004b; 17. Markert et al., 2004; 18. Richardson et al., 2004; 19. Halverson et al., 2006; 20. Oh and Badyaev, 2006; 21. Duarte et al., 2003; 22. Coltman et al., 1999; 23. Foerster et al., 2003, 24. Merila¨ et al., 2003; 25. Lieutenant‐Gosselin and Bernatchez, 2006; 26. Hoffman et al., 2004; 27. Hawley et al., 2005; 28. Acevedo‐Whitehouse et al., 2003; 29. MacDougall‐Shackleton et al., 2005; 30. Lens et al., 2000; 31. Neff, 2004; 32. Gage et al., 2006; 33. Tomiuk et al., 2006; 34. Ho¨glund et al., 2002; 35. Seddon et al., 2004; 36. Marshall et al., 2003; 37. Amos et al., 2001; 38. Zedrosser et al., 2007; 39. Slate et al., 2000; 40. Keller, 1998; 41. Marr et al., 2006; 42. Kruuk et al., 2002; 43. Keller et al., 2002; 44. Reid et al., 2003; 45. Reid et al., 2005; 46. Von Schantz et al., 1996; 47. Paterson et al., 1998; 48. Madsen and Ujvari, 2006; 49. Wegner et al., 2003a,b; 50. Kurtz et al., 2004; 51. Harf and Sommer, 2005; 52. Westerdahl et al., 2005; 53. Sauermann et al., 2001; 54. Widdig et al., 2004. b

202

MATE CHOICE AND GENETIC QUALITY

203

significant effects for key fitness traits such as survival, and the majority of nonsignificant correlations are in the expected direction (see also Balloux et al., 2004). C. THE INTERPRETATION OF HETEROZYGOSITY–FITNESS CORRELATIONS 1. Hypotheses Much of the debate focuses around the meaning of correlations between an individual’s heterozygosity as inferred from molecular markers and fitness (Balloux et al., 2004; David, 1997; Hansson and Westerberg, 2002; Lynch and Walsh, 1998; Pemberton, 2004). Three hypotheses have been proposed (reviewed in Hansson and Westerberg, 2002; see also Slate and Pemberton, 2002; Tiira et al., 2006). a. The general effect hypothesis This hypothesis proposes that the observed heterozygosity–fitness relationship is due to negative effects of homozygosity at genome‐wide distributed functional loci (i.e., inbreeding depression). Inbreeding depression can be caused: (1) by the expression of (partly) recessive deleterious alleles in homozygous individuals, or (2) by functional overdominance, that is, higher fitness for the heterozygous state relative to both homozygous states. Which of these two effects is most important is not clear. Charlesworth and Charlesworth (1999) concluded that functional overdominance is not important in most cases. However, Lynch and Walsh (1998) argued that even if functional overdominance occurs at only a few loci, it can contribute substantially to inbreeding depression. The general effect hypothesis assumes that the marker loci (usually microsatellites) are selectively neutral, but that there is associative overdominance with functional loci. Associative overdominance occurs when heterozygosity at the neutral marker locus is linked to heterozygosity at one or more functional loci (Zouros, 1993). The general effect hypothesis suggests that the associative overdominance arises because of identity disequilibria. This simply means that the probability that alleles are identical by descent is correlated across loci even if these loci are physically unlinked. This occurs under inbreeding, which results in genome‐wide linkage disequilibrium. The general effect hypothesis thus predicts that individual heterozygosity only correlates with fitness if it also correlates with individual inbreeding level. An association between individual heterozygosity and fitness is then only expected in partially inbred populations, but not in large, randomly mating populations.

204

BART KEMPENAERS

b. The local effect hypothesis This hypothesis suggests that the association between heterozygosity and fitness is due to negative effects of homozygosity at functional loci that are in linkage disequilibrium with marker loci. Thus, the heterozygosity–fitness correlation is also due to associative overdominance, but only requires linkage disequilibrium between particular markers and particular functional loci. This is expected in natural populations: (a) when the marker and the functional locus are closely physically linked, (b) when populations that originated from a few founders or that went through a bottleneck rapidly expand again, (c) when different populations intermix, (d) when genetic drift occurs in small populations, or (e) as a result of selection (see Bierne et al., 2000b; Hansson and Westerberg, 2002). The associative overdominance mechanism can work both if the linked loci show functional overdominance or via partially recessive deleterious alleles (Lynch and Walsh, 1998). The local effect hypothesis predicts that heterozygosity at particular microsatellite loci will correlate with fitness‐related traits.

c. The direct effect hypothesis This hypothesis suggests that the association between heterozygosity and fitness results from functional overdominance at the scored loci themselves. Thus, heterozygotes at the typed markers would have greater fitness than homozygotes. It is usually suggested that this is more likely for allozymes (Mitton, 1997; Van Oosterhout et al., 2004; but see Zouros et al., 1988) and for Mhc loci than for microsatellites, which are assumed to be selectively neutral. However, this assumption might not always hold. Polymorphic microsatellites can be involved in forming particular DNA structures, can influence DNA replication and gene expression, and can be hotspots for recombination, depending on the type of microsatellite and on their location (reviewed in Chistiakow et al., 2006). 2. Evidence There is compelling evidence against the general effect hypothesis as a general explanation for the observed heterozygosity–fitness correlations, that is, against the idea that heterozygosity across a relatively small number of microsatellite markers simply reflects inbreeding. First, theory predicts and empirical data show that estimates of individual heterozygosity correlate only weakly with the individual inbreeding coefficient, even if a large number of microsatellites are used that are evenly spread over the genome (Fig. 1; Balloux et al., 2004; Coulson et al., 1998; Markert et al., 2004; Pemberton, 2004; Slate et al., 2004; but see Hansson, 2004).

205

MATE CHOICE AND GENETIC QUALITY

A 1.4 1.3

SH

1.2 1.1 1.0 0.9 0.8 0.7 B 0.8 0.7

H

0.6 0.5 0.4 0.3 0.2 C 1.6 1.4 1.2

SH

1.0 0.8 0.6 0.4 0.2 0.0 0.00

0.05

0.10

0.15

0.20

0.25

Inbreeding coefficient (f ) FIG. 1. Three examples of the relationship between the inbreeding coefficient f derived from pedigree data and (standardized) multilocus heterozygosity (H or SH). (A) r ¼ .18, N ¼ 590 Coopworth sheep (O. aries) typed at 101 loci; adapted from Fig. 3 in Slate et al. (2004) by permission from Macmillan Publishers Ltd.: Heredity, copyright 2004. (B) r ¼ .16, N ¼ 211 Darwin’s ground finches (Geospiza fortis) typed at 13 loci; adapted from Fig. 1A in Markert et al. (2004) by permission from Macmillan Publishers Ltd.: Heredity, copyright 2004. (C) r ¼ .21, N ¼ 241 Soay sheep (O. aries) typed at 18 loci; adapted from Fig. 1a in Overall et al. (2005) by permission from the authors and Blackwell Publishing.

206

BART KEMPENAERS

Second, the general effect hypothesis predicts that estimates of individual heterozygosity based on different sets of markers should be strongly positively correlated, but this is often not the case. For example, Lieutenant‐ Gosselin and Bernatchez (2006) emphasized that the mean correlation coefficient between individual heterozygosity computed with 2 random subsets of 15 loci was only .23, and Tiira et al. (2006) even found a negative mean value of r ¼ .11 when comparing subsets of a total of 13 loci. Similarly, heterozygosity at 5 loci did not correlate with heterozygosity at 14 different loci in a study on great reed warblers (r ¼ .14; Hansson et al., 2004b). The strongest correlations have been reported in a study on inbred populations of rabbits Oryctolagus cuniculus (mean r ¼ .31, range .20–.50, comparing random subsets of a total of 29 loci, Gage et al., 2006), and in a study on common frogs Rana temporaria (r ¼ .26–.51, comparing two random sets of four loci, Lesbarre`res et al., 2005). So, if not inbreeding depression, how should we interpret the many significant correlations between individual heterozygosity and fitness‐ related traits observed in outbred populations? Currently, there are two explanations. Markert et al. (2004) suggested that the inbreeding coefficient is not that good an estimate of the total proportion of an individual’s alleles that are identical by descent. This is easy to see when considering a full‐sib mating. Offspring from such a pair will on average share 50% of their genes, but the range goes (theoretically) from 0% to 100%. Markert et al. (2004) further suggested that individual heterozygosity might capture some of the inbreeding depression that is not accounted for by the inbreeding coefficient. This could explain why heterozygosity–fitness correlations are found even within groups of individuals with the same inbreeding coefficient (Hansson et al., 2004b; Markert et al., 2004). Alternatively, the observed effects could reflect local effects due to linkage disequilibrium between marker loci and loci with a fitness effect. The evidence for such local effects comes from the many studies that have shown significant heterozygosity–fitness correlations at single loci (e.g., Bean et al., 2004; Da Silva et al., 2006; Ferreira and Amos, 2006; Hansson et al., 2004b; Heath et al., 2002; Lieutenant‐Gosselin and Bernatchez, 2006; Merila¨ et al., 2003; Van Oosterhout et al., 2004). However, these results should be interpreted with some caution, because calculating correlations between many individual markers and many fitness‐related traits increases the probability of Type I statistical errors. Nevertheless, Lieutenant‐Gosselin and Bernatchez (2006) showed that some loci display strong overdominance effects, with heterozygosity at a given locus explaining up to 22% of the variance in a fitness‐related trait. They also showed that for some loci the correlations are significantly negative, that is, the homozygotes do better than the heterozygote

MATE CHOICE AND GENETIC QUALITY

207

(underdominance). Thus, the apparent multilocus overdominance effects, which are usually relatively weak, might actually be the summed effect of a suite of stronger single‐locus over‐ and underdominance effects. Ferreira and Amos (2006) took the study of heterozygosity–fitness correlations one step further. They first identified loci that showed strong heterozygote advantage in inbred and outbred Drosophila. Then, they selected several additional microsatellite markers that flanked these loci to show a decreasing effect with increasing distance from the original locus. This implies that the local effect is due to physical linkage between the marker locus and a gene that is under balancing selection. The fact that most studies use only few markers, this clearly suggests that there are more polymorphisms with major fitness effects in the genome than commonly assumed. If linkage disequilibrium is not only due to physical linkage, then general heterozygosity–fitness correlations are only expected in specific populations (e.g., small, recently bottlenecked, recently mixed). The fact that positive correlations are also found in large, outbred populations suggests that we might have to reconsider the extent of linkage disequilibrium in natural populations. To further our understanding of the cause of heterozygosity–fitness correlations, it will be important to obtain information about the location of the microsatellites in the genome, and about linkage disequilibrium (Hansson et al., 2004b), and ultimately to identify the genes with fitness effects. 3. Heterozygosity–Fitness Correlations and the Mhc Some studies have suggested that heterozygosity–fitness correlations based on microsatellite markers could reflect heterozygosity effects of Mhc genes (e.g., Hawley et al., 2005; Neff, 2004). However, there are two problems with this hypothesis. First, it requires that multilocus heterozygosity (MLH) at microsatellite markers reflects inbreeding, or that marker loci are in linkage disequilibrium with Mhc loci. As expected, significant positive correlations between Mhc band‐sharing and overall genetic similarity are found in pairs of birds from small island populations where inbreeding is common (r ¼ .44 for 33 pairs of Savannah sparrows, Passerculus sandwichensis, Freeman‐Gallant et al., 2003; r ¼ .24 for 97 pairs of Seychelles warbler, Acrocephalus sechellensis, Richardson et al., 2005). In contrast, heterozygosity at an Mhc locus was unrelated to heterozygosity estimated from 10 microsatellite markers in a study on rhesus macaques Macaca mulatta (Sauermann et al., 2001). Similarly, allelic diversity at the Mhc did not correlate with heterozygosity estimated from microsatellite markers in great reed warblers, Acrocephalus arundinaceus (r ¼ .06, p ¼ .58, N ¼ 100 individuals typed at 19 microsatellite loci, Hansson et al., 2004b;

208

BART KEMPENAERS

r ¼ .04, p ¼ .38, N ¼ 340 individuals typed at 18 loci, Westerdahl et al., 2005), in three‐spined sticklebacks Gasterosteus aculeatus (r ¼ .02, p ¼ .8, N ¼ 124 individuals typed at seven loci; Reusch et al., 2001), or in humans (Carrington et al., 1999). The second thing to note is that the fitness effects of heterozygosity at Mhc loci are probably not due to simple overdominance effects. First, numerous studies have demonstrated significant associations between the presence of specific Mhc alleles and the susceptibility to a particular disease (e.g., Bonneaud et al., 2006b; De Eyto et al., 2007; Langefors et al., 2001; Westerdahl et al., 2005; reviewed in Piertney and Oliver, 2006; for data on humans see the online Supplementary Table I from de Bakker et al., 2006). Effects on fitness are thus a consequence of the presence of a resistance allele (a ‘‘good allele’’ type of benefit; Paterson et al., 1998; Pitcher and Neff, 2006). This may lead to an apparent heterozygote advantage in that the fitness of heterozygotes is greater than the average fitness of the homozygotes, but similar to the fitness of the homozygote that carries the resistance allele (Penn et al., 2002). Second, the apparent heterozygote advantage may be due to selection for rare alleles, which are more likely to occur in heterozygous individuals (Apanius et al., 1997). Third, heterozygous individuals might be more fit than any homozygote if there are multiple pathogens or parasites, simply because each specific resistance allele will protect the individual against a specific pathogen. Note that most loci of Mhc genes are duplicated, and that these duplicated loci are most likely functionally equivalent (Klein, 1986), which allows individuals to carry more than two resistance alleles. Considering the relation with fitness, it is thus the total number of alleles at these duplicated loci that counts, and not the heterozygosity at a single locus (Piertney and Oliver, 2006; Wegner et al., 2003a,b). Hence, the detection of a heterozygote advantage at Mhc loci is more likely in natural environments, where individuals will often be faced with multiple pathogens (Wegner et al., 2003a, 2004). It has also been suggested that Mhc‐based mate choice could facilitate inbreeding avoidance and increase overall offspring heterozygosity (Piertney and Oliver, 2006). However, this would also require linkage disequilibrium between Mhc loci and genome‐wide loci (see also Reinhold, 2002). D. IS THE HETEROZYGOSITY–FITNESS RELATIONSHIP LINEAR OR QUADRATIC? Optimal mate choice with respect to relatedness will also depend on whether the relationship between heterozygosity and fitness is linear (i.e., the more heterozygous, the better) or quadratic (i.e., individuals with intermediate levels of heterozygosity will have the highest fitness). There are good theoretical reasons to expect a quadratic relationship between fitness and heterozygosity.

MATE CHOICE AND GENETIC QUALITY

209

First, there is probably an optimal degree of outbreeding (Bateson, 1983). Individuals in a population might carry coadapted gene complexes, that is, there are epistatic interactions with positive fitness effects. This might occur under local adaptation by natural selection or due to sexually antagonistic coevolution (Rice and Chippindale, 2002). If an individual mates with a partner that is too distantly related, these coadapted gene complexes would break up, leading to reduced offspring fitness. This phenomenon, referred to as outbreeding depression, is observed in nature when closely related species interbreed (e.g., Veen et al., 2001), or when individuals from distant populations are crossed (Lynch and Walsh, 1998). In natural panmictic populations, outbreeding depression is probably unimportant, but it could become an issue when individuals mate with a rare immigrant from a distant population. Second, a special case where extreme heterozygosity is almost certainly detrimental is at the Mhc, because this leads to autoimmunity (Doherty and Zinkernagel, 1975). Nowak et al. (1992) presented a model showing that an optimal, intermediate number of Mhc alleles maximizes the probability of mounting an immune response against a large number of foreign peptides. Thus, one would expect a quadratic relationship between Mhc allelic diversity (heterozygosity) and fitness‐related traits. E. REVIEW OF FITNESS EFFECTS OF HETEROZYGOSITY Table II provides an overview of a variety of fitness‐related traits for which evidence exists that they are affected by inbreeding depression, by overall individual heterozygosity, or by heterozygosity at specific loci. Although there are also many studies showing negative results (not cited), there can be little doubt that in a wide variety of species individual genetic diversity has an influence on fitness. 1. Effects on Survival One of the most commonly found effects of inbreeding or reduced genetic diversity is a decreased probability of survival, often at a very early stage (embryo), but also later in life. For example, in the collared flycatcher (Ficedula albicollis), inbreeding caused a substantial decrease in hatching success and in postfledging juvenile survival, so that the number of recruits from a brood with an inbreeding coefficient of f ¼ .25 was 94% lower (confidence interval: 81–98%) compared to a noninbred brood (Kruuk et al., 2002). Similarly, in great tits (Parus major) pair relatedness had a strong negative effect on hatching success (effect size: r ¼ .41, N ¼ 111), and on fledging success (r ¼ .47, N ¼ 109; Van de Casteele et al., 2003). Studies on great reed warblers and blue tits (Cyanistes caeruleus)

210

BART KEMPENAERS

showed a negative correlation between hatching success and genetic similarity between the pair members (Fig. 2), but no effects on postfledging survival were found (Bensch et al., 1994; Hansson, 2004; Kempenaers et al., 1996; Krokene and Lifjeld, 2000). The relationship between hatching success and genetic similarity was nonlinear, with a strong decline in hatching success at higher relatedness values (Fig. 2, see also Hansson, 2004). It is noteworthy that for a great reed warbler population that went through a recent bottleneck there might be an optimal intermediate relatedness (Fig. 2A), whereas for an outbred blue tit population hatching success decreased with increasing relatedness (Fig. 2B). Another study on blue tits found an inbreeding effect 1.0

A

0.8 0.6

B

Proportion of hatched eggs

0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 0.1

0.2

0.3

0.4 0.5 Band-sharing D

0.6

0.7

FIG. 2. Hatching success of a clutch in relation to genetic similarity between the parents (estimated as the band‐sharing coefficient, D). (A) Great reed warbler (A. arundinaceus), r ¼ .33, p < .01, N ¼ 80, second order regression line drawn for illustrative purposes only; adapted from Fig. 3 in Bensch et al. (1994) by permission from the authors and Blackwell Publishing. (B) Blue tit (C. caeruleus), r ¼ .27, p < .01, N ¼ 103; adapted from Fig. 1 in Kempenaers et al. (1996).

MATE CHOICE AND GENETIC QUALITY

211

on hatching success in an island population, but not in a nearby mainland population (Krokene and Lifjeld, 2000). At the interspecific level, the mean hatching success in a population was also negatively correlated with the mean genetic similarity between pair members or between random adults in the population (Spottiswoode and Møller, 2004). The patterns based on correlational data are confirmed by experiments where the degree of inbreeding is manipulated. For example, the proportion of eggs hatching in two species of crickets (Gryllus bimaculatus and Teleogryllus commodus) was significantly reduced when females were only allowed to mate with siblings (Fig. 3A; Jennions et al., 2004; Tregenza and

A 0.6

B

Proportion of eggs hatching (mean ⫾ SE)

0.5

0.4

0.3

0.6

0.5

0.4

0.3 SS NN SN NS Female mating treatment FIG. 3. Hatching success for female crickets in relation to mating treatment (S ¼ full sibling male, N ¼ unrelated male; note the order of mating). (A) Field cricket (G. bimaculatus), F3,75 ¼ 6.01, p < .001; adapted from Fig. 1 in Tregenza and Wedell (2002) by permission from Macmillan Publishers Ltd.: Nature, copyright 2002. (B) Field cricket (T. oceanicus), F3,84 ¼ .47, p > .70; adapted from Fig. 1 in Simmons et al. (2006) by permission from the authors and Blackwell Publishing.

212

BART KEMPENAERS

Wedell, 2002). Interestingly, the effect was not found in a similar study on another cricket species (Teleogryllus oceanicus, Fig. 3B; Simmons et al., 2006). Several studies found a significant effect in one sex only. For example, in wild greater horseshoe bats (Rhinolophus ferrumequinum; Rossiter et al., 2001) only male offspring survival was affected by individual heterozygosity. In contrast, in blue tits, only female adult survival correlated with individual heterozygosity (Foerster et al., 2003). In the latter study, it is possible that the survival cost of reduced heterozygosity influenced males at an earlier age, because offspring that were recruited to breed were more heterozygous than their nonsurviving nestmates and 72% of 53 recruits were males (Foerster et al., 2003). There is also a study on red deer (Cervus elaphus) where the effect of heterozygosity on the survival of calves is opposite in both sexes (Coulson et al., 1999). The study suggests that female survival is most influenced by recent inbreeding, whereas male survival is influenced by the degree of outbreeding. In the above‐mentioned studies, egg hatchability was affected by the relatedness between the pair members, and not by male or female qualities. This was confirmed in blue and great tits by showing that egg hatchability was repeatable for pairs, but not for individual males or females (Kempenaers et al., 1996). However, some studies have shown that it is the female’s inbreeding coefficient or heterozygosity that has the strongest effect on egg hatchability (Cordero et al., 2004; Keller, 1998), whereas another study suggested that there is an additive genetic sire effect (Garcı´a‐Gonza´lez and Simmons, 2005). Several studies have shown evidence for outbreeding depression. A more extreme example is the hybridization between collared and pied (Ficedula hypoleuca) flycatchers. Compared to the parental species, F1 hybrids show reduced hatching and fledging success (Veen et al., 2001). In this case, there is again a marked difference between the sexes, with a much stronger negative effect on females. Hatching success was reduced by 89% relative to the parental species in F1 females, but only by 8% in F1 males, and F1 females never recruited any offspring (Veen et al., 2001). Outbreeding depression within a species has been suggested in a study on ornate dragon lizards (Ctenophorus ornatus), where the proportion of surviving offspring decreased with increasing mean d2 value of the female (LeBas, 2002). However, the sample size was small (N ¼ 11 females), and there was no evidence that individual genetic diversity (heterozygosity or mean d2) affected juvenile survival. In the spotless starling (Sturnus unicolor), Cordero et al. (2004) found a linear (negative) relationship between the genetic similarity of the pair and hatching success, and an additional quadratic relationship between female heterozygosity and hatching success,

MATE CHOICE AND GENETIC QUALITY

213

suggesting both in‐ and outbreeding depression. Simultaneous in‐ and outbreeding depression was also suggested in a reintroduced population of the Arabian oryx (Oryx leucoryx): juvenile survival correlated positively with individual heterozygosity, and negatively with mean d2 when both variables (H and mean d2) were entered together in the model (Marshall and Spalton, 2000). The effect of mean d2 might be a consequence of the fact that the population came from a small number of founders that were taken from different sources (Marshall and Spalton, 2000; Tsitrone et al., 2001). Relatively few studies have directly investigated the relationship between Mhc diversity and survival, and those that have usually found additive effects of specific Mhc alleles that confer resistance against a particular pathogen (e.g., Paterson et al., 1998; Pitcher and Neff, 2006; see also De Eyto et al., 2007). Surprisingly, the study on wild‐caught Chinook salmon (Oncorhynchus tshawytscha) also showed an underdominance effect at one locus, which would lead to a reduction in Mhc variability (Pitcher and Neff, 2006). In a study on great reed warblers where dyads of full siblings were compared (Hansson et al., 2004b), surviving offspring tended to have a higher Mhc allelic diversity (effect size r ¼ .27). Interestingly, individual heterozygosity at several presumably neutral microsatellite loci showed stronger relationships with offspring survival (largest effect size r ¼ .40) than Mhc diversity (Hansson et al., 2004b). 2. Effects on Parasite Load and the Immune System Many studies have found significant correlations between individual genetic diversity and measures of disease resistance, and this might be one of the underlying causes of the effects on juvenile and adult survival discussed above. For example, in a sample of stranded California sea lions (Zalophus californianus), more inbred individuals (estimated based on 11 microsatellite markers) had more different types of helminth infections, and when sick individuals were treated rehabilitation time was longer for inbred individuals (Acevedo‐Whitehouse et al., 2003). Similar effects have been shown in island populations or in populations of endangered species, where inbreeding is common. For example, in the Gala´pagos hawk (Buteo galapagoensis), genetic diversity decreases with island size, and there is a strong negative correlation between average parasite load (louse abundance) and average heterozygosity in eight island populations (Whiteman et al., 2006). In Soay sheep (Ovis aries) and in Cuvier’s gazelle (Gazella cuvieri) fecal egg counts—a reliable estimate of parasite load—decreased with increasing heterozygosity or with decreasing inbreeding (Cassinello et al., 2001; Coltman et al., 1999). Reid et al. (2003) showed that the cell‐mediated immune response in an island population of

214

BART KEMPENAERS

song sparrows (Melospiza melodia) strongly depended on the individual inbreeding coefficient for juvenile and adult birds. However, it was only the mother’s inbreeding coefficient that affected the immune response in chicks, suggesting that inbred mothers were lower quality parents (Reid et al., 2003). Even in outbred populations, there is now evidence that individual genetic diversity affects immunocompetence. For example, in mountain white‐ crowned sparrows (Zonotrichia leucophrys oriantha), individuals infected with a blood parasite were less heterozygous than uninfected individuals, and parasite load decreased with increasing mean d2 (MacDougall‐ Shackleton et al., 2005). In an experimental study, Hawley et al. (2005) inoculated house finches (Carpodacus mexicanus) with a recently emerged disease (Mycoplasma gallisepticum) and showed that individual heterozygosity predicted the severity of the disease (r2 ¼ .19). They also showed that heterozygosity positively affected a measure of the cell‐mediated, but not of the humoral immune response. As mentioned above, disease resistance is often directly linked to the presence of specific Mhc alleles (additive effects), and optimal immunity is expected at intermediate levels of Mhc diversity. This is beautifully illustrated by two studies on three‐spined sticklebacks. When fish were experimentally exposed to multiple parasites, individuals with an intermediate number of Mhc class II alleles had lower parasite loads (Wegner et al., 2003a) and suffered less from the exposure (Kurtz et al., 2004) than either more or less Mhc diverse individuals. A quadratic relationship between the number of blood parasites and allelic diversity of Mhc class I genes was also found in free‐living populations of great reed warblers (Westerdahl et al., 2005) and water pythons (Liasis fuscus; Madsen and Ujvari, 2006). There is also some evidence for a maximum heterozygote advantage at Mhc loci. For example, in humans, maximum heterozygosity at three HLA class I loci delayed the onset of AIDS in individuals that were infected with the HIV‐1 virus (Carrington et al., 1999). A study on Chinook salmon also showed a heterozygote advantage at an Mhc class II locus (Arkush et al., 2002). In this study, inbred (f ¼ .25) and outbred (f ¼ 0) offspring from the same female were experimentally exposed to one of three pathogens. For one of those pathogens, inbred individuals were more likely to become infected and the infection was more severe. For another pathogen, Mhc heterozygotes had an increased probability of survival compared to homozygotes. However, the effect was marginal and most heterozygotes were individuals that carried one of three rare alleles (see Table I in Arkush et al., 2002).

MATE CHOICE AND GENETIC QUALITY

215

3. Effects on the Competitive Ability of Males and Females If individual heterozygosity or inbreeding affects immunocompetence and survival, and also influences growth rate, and developmental stability (e.g., Bierne et al., 2000a; Britten, 1996; Mitton and Grant, 1984; Vøllestad et al., 1999; see also Ru¨licke et al., 2006), it seems likely that an individual’s ability to compete for access to resources will also depend on its genetic diversity. This was directly demonstrated in an elegant experiment with the butterfly Bicyclus anynana, where three males with different inbreeding coefficients competed for one unmated, outbred female (Joron and Brakefield, 2003). Outbred males had higher mating success, particularly when they were allowed to compete under more natural‐like conditions (Joron and Brakefield, 2003). There is also ample indirect evidence from a variety of systems. 1. Individual heterozygosity was positively correlated with measures of social status, such as dominance (in brown trout, Salmo trutta; Tiira et al., 2006), aggressive behavior (in salmon Salmo salar; Tiira et al., 2003), territorial status (in males of the lekking black grouse Tetrao tetrix; Ho¨glund et al., 2002), and territory size (in a cooperatively breeding bird, the subdesert mesite, Monias benschi; Seddon et al., 2004). 2. Heterozygosity correlated positively with the development of male ornaments, for example in the spotless starling (length of the throat feathers; Aparicio et al., 2001) and in the blue tit (UV reflection of the crown feathers; Foerster et al., 2003). In the guppy (Poecilia reticulata), close inbreeding led to a decrease in the area covered with orange and black spots (Van Oosterhout et al., 2003). This is particularly interesting because the genes that determine these color patterns are probably Y‐linked and hemizygous. This means that the observed effect cannot be due to inbreeding depression at those genes. Instead, the trait might be condition‐ dependent and inbreeding might have reduced condition, or the effect might be the result of epistatic interactions between autosomal loci and the hemizygous Y‐linked genes (see Van Oosterhout et al., 2003). 3. Heterozygosity or inbreeding affects courtship behavior. For example, Van Oosterhout et al. (2003) also found a reduction in courtship activity in inbred guppy males, particularly in the time males spent intensively courting and pursuing females. However, it is unclear whether such an effect could have been linked to the reduction in ornamentation, which might have affected how females responded to the males. Other studies showed that song repertoire size increased with decreasing inbreeding coefficient in the song sparrow (Reid et al., 2003) and with increasing heterozygosity in

216

BART KEMPENAERS

the sedge warbler Acrocephalus schoenobaenus (Marshall et al., 2003). Individual heterozygosity also affected song structure (Seddon et al., 2004), and song frequency (in male Drosophila montana, Aspi, 2000). The latter study also found evidence for outbreeding depression on the same trait, by crossing the two most dissimilar inbred strains from opposite ends of the distribution range. 4. A reduction in individual heterozygosity is related to a decrease in sperm quality and in testis size in wild rabbits (Gage et al., 2006), suggesting that inbred males would do worse under sperm competition. A positive correlation between heterozygosity and relative gonad size (gonadosomatic index) was also found in male domestic Chinook salmon (Heath et al., 2002). Gage et al. (2006) suggested that their findings could explain why inbreeding depression often causes decrease in fertilization and hatching rate. This seems rather unlikely, however, because male heterozygosity does not correlate with hatching success, at least not in the studies reviewed above. However, experimental work on the butterfly B. anynana showed that brother–sister matings strongly increased infertility in sons, but not in daughters (Saccheri et al., 2005). The underlying reasons for this sex‐ difference remain elusive. Only few studies have investigated secondary sexual traits in relation to Mhc diversity. Von Schantz et al. (1996) found that spur length in wild ring‐ necked pheasants Phasianus colchicus varied significantly depending on the Mhc genotype, but homozygote and heterozygote individuals did not differ. 4. Effects on Male and Female Reproductive Success Individual heterozygosity or inbreeding can have significant effects on the reproductive success of an individual (Table II). In most cases, individual heterozygosity correlates positively with the number of offspring males and females produce in a season (e.g., Amos et al., 2001) or over their lifetime (e.g., Slate et al., 2000). In the latter study on red deer, the effect was stronger in males. Furthermore, some studies have shown that more heterozygous females produce larger clutches (e.g., in domestic Chinook salmon, Heath et al., 2002; in blue tits, Foerster et al., 2003), or lay earlier in the season (e.g., in great tits, Tomiuk et al., 2006). In contrast, a study on bluegill sunfish (Lepomis macrochirus) showed that individuals with intermediate levels of heterozygosity had the highest reproductive success (Neff, 2004). Neff (2004) suggested that the effect might be linked to variation in immunocompetence (cfr. Wegner et al., 2003a), which would require a link between mean d2 (estimated based on 11 microsatellite loci) and diversity at the Mhc. In this particular

MATE CHOICE AND GENETIC QUALITY

217

example, the observed outbreeding depression may be caused by the disruption of coevolved gene complexes, either due to hybridization with the closely related pumpkinseed sunfish (Lepomis gibbosus), or due to a disadvantage of disassortative mating with respect to foraging morph (there are two morphs, with different foraging strategies; Neff, 2004).

F. DETERMINANTS OF THE PRESENCE/ABSENCE OF HETEROZYGOSITY–FITNESS CORRELATIONS There are many studies that did not find the expected positive relationships between individual heterozygosity and fitness‐related traits (e.g., Duarte et al., 2003). Should this be taken as an argument to doubt the general importance of individual genetic diversity or inbreeding? A more fruitful approach may be to ask under which circumstances heterozygosity–fitness correlations can be expected. Independent of whether it is a general or local effect, it can be hypothesized that the effect depends on the environment, and this may serve to explain variation among years, populations, and studies (see Coltman and Slate, 2003). Armbruster and Reed (2005) reviewed the literature to compare inbreeding depression in more or less stressful environments. They found that in 76% of the cases, inbreeding depression was more severe under stressful conditions (statistically significant in 48% of cases), but in the remaining cases the opposite was found (although it is unclear whether this was ever significant). Given the obvious difficulties in comparing studies with different (often low) sample sizes, different experimental procedures, different definitions of stress, and different fitness components measured, Armbruster and Reed (2005) emphasized that it will be ‘‘difficult to impossible’’ to accurately predict the response of populations to inbreeding. However, they suggested that progress could be made in three ways. (1) More studies on inbreeding under natural conditions are needed, because the reviewed studies often used artificial stress factors. (2) The specific loci that contribute to inbreeding depression should be identified, so that their environmental sensitivity can be experimentally measured. (3) A theoretical framework needs to be developed. Studies using microsatellite markers (or long‐term pedigree data) to test heterozygosity–fitness correlations have much to contribute to the first two issues mentioned by Armbruster and Reed (2005). These studies generally support the hypothesis that the effects of inbreeding or individual heterozygosity on fitness are stronger in more stressful environments (Table III).

TABLE III INTERACTIVE EFFECTS OF ENVIRONMENTAL FACTORS AND HETEROZYGOSITY/INBREEDING ON FITNESS‐RELATED TRAITS Species Common name

Latin name

Methoda

Environmental factor

Trait

218

Natterjack toad

Bufo calamita

C

I

Intraspecific competition

Survival

Darwin’s finch

Geospiza scandens

C

F

Rainfall

Juvenile survival

Rainfall, population density

Adult survival

Great tit

P. major

C

R

Laying date

Song sparrow

M. melodia

C

F

Temperature

Nestling survival, fledging success Laying date

Rainfall

Hatching success

Spotless starling

S. unicolor

C

R

Year

Hatching success

Taita thrush

Turdus helleri

C

R

Habitat disturbance

Fluctuating asymmetry

Heterozygosity effect

References

More pronounced under competition with outbred individuals Only in years with low rainfall More pronounced in very dry and crowded conditions More pronounced in late clutches

1

More pronounced at higher temperatures More pronounced during periods of rain Stronger in year with higher mean hatching success More pronounced under high levels of disturbance

4

2 2

3

4 5 6

Alpine marmot

M. marmota

C

H

Year

Juvenile survival

Earthworm

Eisenia fetida

E

H

Moisture

Growth

Butterfly

B. anynana

E

F

Mating success

Termite Common frog

Zootermopsis angusticollis R. temporaria

E E

I H

Cage versus seminatural conditions Fungus infection Food

Survival Survivalb

Temperature

Survivalb

Field versus laboratory

Survival

Wood frog

Rana sylvatica

E

H

More pronounced in bad (wet) years (interaction: p ¼ .072) Only in low to moderate moisture treatment Stronger in seminatural conditions

7

Only at high dose Only in restricted food treatment Negative effect at low temperatures, positive effect at high temperaturesc Only in field conditions

10 11

8 9

11

12

219

a C ¼ correlational; E ¼ experimental; F ¼ individual inbreeding coefficient; H ¼ individual heterozygosity; I ¼ inbred versus outbred; R ¼ relatedness between pair members (genetic similarity). b Only effects that were significant after Bonferroni correction are shown here. c Heterozygosity calculated as mean d2. 1. Rowe and Beebee, 2005; 2. Keller et al., 2002; 3. Van de Casteele et al., 2003; 4. Marr et al., 2006; 5. Cordero et al., 2004; 6. Lens et al., 2000; 7. Da Silva et al., 2006; 8. Audo and Diehl, 1995; 9. Joron and Brakefield, 2003; 10. Calleri et al., 2006; 11. Lesbarre`res et al., 2005; 12. Halverson et al., 2006.

220

BART KEMPENAERS

For example, a field study on great tits showed that the negative effects of genetic similarity between pair members on nestling survival and fledging success were more pronounced in clutches that were laid later in the season, presumably when feeding conditions for the offspring were worse (Van de Casteele et al., 2003; Fig. 4). In an exemplary experimental study on common frogs (R. temporaria) coming from four populations along a 1600 km latitudinal gradient, Lesbarre`res et al. (2005) tested the effect of heterozygosity on tadpole survival, developmental rate, and growth rate. Adults from the different populations were crossed in the laboratory, eggs and tadpoles were kept at three temperature treatments, and tadpoles were kept under two food treatments (restricted or ad libitum). In this way, interactions between estimates of heterozygosity or genetic similarity between the parents and population, temperature and food treatment could be tested. The study showed that the effect of heterozygosity on survival was independent of the population, but most pronounced in stressful environments, that is, at low temperatures and in the restricted food treatment (Lesbarre`res et al., 2005). Although not all studies have found interactions between inbreeding and environmental effects on fitness (e.g., Armbruster et al., 2000; Edly‐Wright et al., 2007; Kruuk et al., 2002), and not all fitness‐related traits are similarly affected by these interactions (e.g., Van de Casteele et al., 2003), the results

Nestling survival

1.0 0.8 0.6 0.4 0.2 0.0 −0.2

Early broods Late broods −0.1

0.0

0.1

0.2

0.3

0.4

Kinship coefficient FIG. 4. Nestling survival in great tit (P. major) broods in relation to the genetic similarity between the parents. There is a significant interaction between kinship and standardized laying date (w2 ¼ 34.53, p < .001). For early broods (N ¼ 52), the relationship is not significant (p > .6). For late broods (N ¼ 57) the relationship is highly significant (r ¼ .57, p < .0001; second order regression line drawn on the figure). Nestling survival is the proportion of hatchlings that fledged. The genetic similarity between the parents was estimated as the kinship coefficient, based on nine microsatellite markers. Data from Van de Casteele et al. (2003).

MATE CHOICE AND GENETIC QUALITY

221

from recent studies (Table III) clearly indicate that heterozygosity–fitness correlations can be sensitive to environmental conditions. This is important for two key reasons. First, it suggests that the strongest effects can be expected during periods when selection is harshest (e.g., Keller et al., 1994). This might explain age‐ as well as sex‐specific effects. For example, the negative consequences of reduced genetic diversity might be amplified in the more competitive sex (e.g., Bonneaud et al., 2006b; Meagher et al., 2000; Merila¨ et al., 2003). Second, it suggests that parents can buffer the negative consequences of low genetic diversity in their offspring, for example through increased parental care (e.g., Avile´s and Bukowski, 2006).

IV. HETEROZYGOSITY AND MATE CHOICE A. MATE CHOICE TO OPTIMIZE OFFSPRING HETEROZYGOSITY 1. Theoretical Considerations If individual heterozygosity generally correlates with fitness under circumstances that are often found in natural populations, females (and males) would clearly benefit from maximizing (or optimizing) the heterozygosity of their offspring. Many empirical studies have attempted to test this, and I review them below after first discussing some other issues that have an impact on the evolution of mate choice for heterozygosity. a. ‘‘Altruistic’’ inbreeding As mentioned above, outbreeding avoidance might have evolved because of the negative fitness consequences of the break up of coadapted gene complexes. This should lead to females preferring males of intermediate relatedness. However, another reason why females could benefit by preferring a related mate is what Lehmann and Perrin (2003) referred to as ‘‘altruistic’’ inbreeding: females would help a relative spread genes that are identical by descent. Inclusive fitness theory predicts that females should accept a related male as breeding partner as long as the fitness loss as a result of inbreeding d is smaller than r/(1 þ r), where r is the relatedness (Kokko and Ots, 2006; Lehmann and Perrin, 2003). Both studies emphasized that the magnitude of inbreeding depression has to be surprisingly high before inbreeding should be avoided (at least a 33% reduction in fitness for full‐sib and 25% for half‐sib matings). This led Kokko and Ots (2006) to state ‘‘it remains a mystery why preferences for incestuous matings are hardly ever reported in animals.’’ Kokko and Ots (2006) give several explanations, of which three are of particular interest here.

222

BART KEMPENAERS

(1) The magnitude of inbreeding depression could be underestimated, for example because only some fitness components are measured, or because they are measured under too benign environmental conditions (see previous section). (2) Researchers have ignored evidence or alternative explanations that do not ‘‘fit’’ the general expectation that inbreeding is bad and should be avoided. (3) If both sexes invest heavily and approximately equally in parental care, and if mating opportunities are not that limited, then it might be better to avoid inbreeding. Indeed, the above models are based on the assumption that the relative would otherwise be unable to reproduce. Thus, in birds, one could predict that inbreeding avoidance should be more pronounced in socially monogamous species, whereas mating with relatives might pay in lekking species. There is currently no evidence supporting this idea, but few studies on lekking species are available (see below). b. How should females trade off the choice between good and compatible alleles? This question was first addressed by Colegrave et al. (2002) using a modeling approach. Colegrave and colleagues assumed that females could either be choosy, based on secondary sexual traits, which are indicators of additive genetic quality, or they could mate promiscuously, and benefit through a postcopulatory mechanism that selects the most compatible sperm to fertilize the eggs. If we more generally define genetic quality as the combination of additive (good allele) and nonadditive (compatible allele) effects, then the model shows that the optimal choice will depend on the relative importance of additive and nonadditive fitness effects, and on the relative costs to obtain them (Colegrave et al., 2002). Further, it is important to consider how well the secondary sexual traits reflect additive genetic quality (precision of information). This led Neff and Pitcher (2005) to develop a verbal model, in which they proposed that two evolutionary scenarios could emerge. In the first scenario, additive and nonadditive genetic variation would occur at intermediate levels, with only small fluctuations over time. This could work via an evolutionary stable strategy, where a mechanism has evolved that allows individual females to optimize choice for good and compatible alleles simultaneously. Alternatively, it could work via an evolutionary stable state, where different female types have evolved, specializing in choice for good or for compatible alleles, with negative frequency dependent selection on the two types (Neff and Pitcher, 2005). The second scenario proposes that populations would cycle between a state where most genetic variation is additive, and a state where most of the variation is nonadditive. In the former state, female choice for good alleles would be favored, which would deplete the additive genetic variation (and hence the information conveyed

MATE CHOICE AND GENETIC QUALITY

223

by indicators of additive genetic quality), and move the population to the other state; selection will then favor choice for compatible alleles, and this will increase the additive genetic variation again, thus completing the cycle (Neff and Pitcher, 2005). These scenarios are intuitively appealing, because they explain how additive and nonadditive genetic variance is maintained. A few formal models have been developed that shed further light on the issue. Using individual‐based computer simulation models, Howard and Lively (2004) showed that both condition‐dependent mate choice and choice for genetic dissimilarity at Mhc loci can invade a population with random choice, but only choice for genetic dissimilarity is evolutionarily stable. However, the advantage of alleles leading to choice for genetic dissimilarity decreases with increasing frequency of such alleles in the population until it becomes selectively neutral. The models thus suggest that in populations where individuals are at risk of infectious disease, different types of mate choice, including random mating, can coexist (Howard and Lively, 2004). Lehmann et al. (2006) came to similar conclusions using population genetic modeling. Their results also suggest that disassortative mating would first spread, but the frequency of the ‘‘choice allele’’ cannot exceed that of the ‘‘random mating allele.’’ Empirical data to evaluate the above scenarios are scarce. In pioneering work on three‐spined sticklebacks, Milinski and colleagues showed that females use olfactory cues to optimize Mhc allelic diversity, and visual signals (red color) to obtain direct or good allele benefits (Milinski, 2003; see also below). The complexity of the relationships between male mating success and a variety of traits that might relate to choice for direct or indirect benefits is also illustrated by a study on a natural population of sticklebacks (Blais et al., 2004). A study on blue tits showed that some females performed extra‐pair copulations with high quality neighbors, presumably obtaining good allele benefits, whereas others engaged in extra‐ pair copulations with males outside the study site, thereby increasing the heterozygosity of their offspring (Foerster et al., 2003; Kempenaers et al., 1992; Fig. 5). The question remains whether these different ‘‘extra‐pair tactics’’ depend on opportunities (e.g., availability of high‐quality extra‐ pair males), or are specific to individual females. Only one experimental study has directly addressed the trade‐off between good and compatible allele benefits (Roberts and Gosling, 2003). Female mice (Mus musculus) were allowed to choose between males that differed in scent‐marking rate (presumably an indicator of genetic quality) and in Mhc dissimilarity with the female (a genetic compatibility cue). As expected, the study showed that females preferred males with the highest scent‐marking rate and males that were more dissimilar. More interestingly, however, it could be shown that marking rate always predicted choice,

224 Difference in heterozygosity (SH) between extra-pair and within-pair young

BART KEMPENAERS

0.10

p < 0.05

NS

p < 0.10

p < 0.001

0.08 0.06 0.04 0.02 0 −0.02 −0.04 −0.06

101

58

15

44

All extrapair fathers

Close neighbours

Local nonneighbours

Nonlocal males

FIG. 5. The mean difference (S.E.M.) in standardized heterozygosity (SH) between extrapair young and their within‐pair nestmates for blue tit (C. caeruleus) broods with mixed paternity. Extrapair fathers are categorized as local neigbors (breeding one to two territories, or 50–200 m from focal nest), local nonneighbors (breeding three to eight territories, or 150–800 m from the focal nest), or nonlocal males (unidentified, not breeding in the study area). Broods containing extrapair young from multiple males are included more than once if the males differed in category. Sample size (number of broods) is indicated above the x‐axis. Adapted from Fig. 1 in Foerster et al. (2003) by permission from Macmillan Publishers Ltd.: Nature, copyright 2003.

independent of variation in Mhc similarity, whereas choice for Mhc dissimilar males was only obvious when the variation in scent‐marking rate among potential mates was limited (Roberts and Gosling, 2003). Thus, in this experiment, choice for a ‘‘good allele’’ indicator was more important than choice for compatibility. Clearly, much more work is needed, ideally on systems where the costs of inbreeding (or the benefits of heterozygosity) and the ‘‘good allele’’ benefits can be quantified (see e.g., Oh and Badyaev, 2006; Wedekind et al., 2001). c. Promiscuity and inbreeding avoidance It has often been proposed that female promiscuity, defined as mating with more than one male during a single reproductive event, is a strategy to reduce inbreeding depression or increase offspring genetic compatibility (e.g., Brooker et al., 1990; Sillero‐ Zubiri et al., 1996; Stockley et al., 1993; Tregenza and Wedell, 2000). Although there are other hypotheses for the evolution of promiscuity that are not mutually exclusive, there is substantial empirical support for the idea. Zeh and Zeh (2001) suggested that female promiscuity as a strategy to avoid incompatibility should be more important in viviparous than in oviparous species, because in the former embryonic development in the

MATE CHOICE AND GENETIC QUALITY

225

mother ‘‘creates a postfertilization arena for genomic conflicts’’ (Zeh and Zeh, 2001). However, studies on multiple mating in insects (Simmons, 2005) and on extra‐pair paternity in birds (see below) also suggest that females might benefit by increasing the heterozygosity of their offspring. There are at least three types of indirect evidence that suggest the plausibility of the hypothesis. 1. In systems where the social mate provides resources for breeding, females might maximize the gain of direct benefits through social mate choice, and the gain of indirect benefits through extra‐pair mate choice. Several studies on birds have indeed shown that extra‐pair young do better in terms of growth, immune response, or survival than the within‐pair young in the same brood (e.g., Johnsen et al., 2000; Kempenaers et al., 1997; Schmoll et al., 2005a). Obviously, this could be due to additive or nonadditive genetic effects. However, in several species extra‐pair young also showed a stronger immune response than their paternal half‐sibs in the nest of the extra‐pair male (Garvin et al., 2006; Johnsen et al, 2000), which suggests a nonadditive effect. 2. Under the hypothesis that female choice evolved to increase offspring heterozygosity, one would expect that the female decision to engage in extra‐pair copulations, or even the success of (extra‐pair) copulations, depend on the particular male–female combination (nontransitivity of male success). This is suggested by the observation that the between‐year repeatability for the proportion of extra‐pair young in a brood is often significant for pairs, but not for individual males or females (see Dietrich et al., 2004 and references therein). Also, in artificial insemination experiments in domestic fowl, male success in gaining paternity depended on the female environment (Birkhead et al., 2004). This can be explained by female choice favoring sperm of more compatible males, or by differential embryo mortality. 3. Inbreeding or decreased heterozygosity typically leads to a reduction in survival, often at an early embryonic stage (see previous section). Thus, the observation that female promiscuity is linked to increased hatching success, or offspring survival can be seen as support for the above hypothesis (correlational studies: e.g., Kempenaers et al., 1999; Madsen et al., 1992; Olsson et al., 1994, 1996; experimental studies: reviewed in Table I in Neff and Pitcher, 2005; Engqvist, 2006; Fisher et al., 2006; Zeh and Zeh, 2006). For example, in the pseudoscorpion Cordylochernes scorpiodes, females that received a spermatophore from each of two different males produced 32% more offspring over their lifetime, compared to females that received two spermatophores from the same male (Newcomer et al., 1999).

226

BART KEMPENAERS

A comparative study on mammals provides further support: species in which females are promiscuous had significantly lower rates of early reproductive failure than species in which females are monandrous (Stockley, 2003), whereby early reproductive failure was defined as the proportion of eggs that were lost due to failed implantation or because of early embryo mortality. Note that there was no difference in late reproductive failure between the two species categories (Stockley, 2003). Similarly, offspring survival was higher in experiments where females mated with their preferred mate rather than with a nonpreferred male, even when different females preferred different males (e.g., Bluhm and Gowaty, 2004). Of course, the observation that female promiscuity increases offspring quality or survival is not sufficient evidence for the genetic compatibility hypothesis. In some cases, the results can also be explained through a different process, usually referred to as the ‘‘good sperm’’ hypothesis (Yasui, 1997). The hypothesis proposes that females can gain good allele (additive genetic) benefits from mating with multiple males, because the sperm from high quality males will do better in competition with sperm from lower quality males. Evidence for this hypothesis stems from experimental work showing that a male’s success in sperm competition correlates positively with offspring survival or quality, independent of the female he mates with (Fisher et al., 2006; Hosken et al., 2003; but see Engqvist, 2006; Simmons, 2001). Studies that have explicitly tested the hypothesis that promiscuity is linked to inbreeding avoidance or increasing offspring heterozygosity are reviewed below. d. Costs and mechanisms of choice We should not necessarily expect to find female choice based on criteria such as relatedness or heterozygosity, even if there are benefits of producing more heterozygous offspring or of avoiding inbreeding. This is because: (a) both the costs and the benefits of choice have to be considered, and (b) there has to be a mechanism that allows such choice. Discussing the costs of mate choice and the variety of—often surprisingly intricate—mechanisms that have been discovered (e.g., Milinski et al., 2005) is beyond the scope of this chapter. Here, I want to provide a few examples illustrating why it is important to consider costs and mechanisms, and how failing to do so can lead to different (and perhaps incorrect) predictions about patterns of mate choice. The most obvious way to avoid inbreeding depression is by avoiding mating with kin. Even in the absence of any kin recognition mechanism, if male and female offspring differ in dispersal distance, the likelihood that they end up as a pair might be very small. A few studies have explicitly

MATE CHOICE AND GENETIC QUALITY

227

tested whether the observed frequency of close inbreeding differs from what would be expected under random mating. For example, Keller and Arcese (1998) showed that in an island population of song sparrows the expected and observed frequency of close inbreeding (f  .125), as well as the expected and observed distribution of inbreeding coefficients did not differ. Thus, despite evidence for inbreeding depression, there is no evidence for inbreeding avoidance through mate choice. In this context, Wheelwright et al. (2006) had an interesting further insight: a female should have better information about her relatedness to her father, than about her relatedness to brothers or sons. This is because as a chick, she can observe the phenotypic characteristics of her father (e.g., song, plumage features, size), but she will only see the juvenile phenotype of her brothers and her sons. Wheelwright et al. (2006) thus predicted that father–daughter matings should be less common than sister–brother or mother–son pairings. Indeed, in their 17‐year study on an island population of Savannah sparrows, they found that father–daughter pairs were significantly less common than expected under random mating (in fact, no such cases have ever been found on the island), and significantly underrepresented compared to mother–son or brother–sister pairs. Furthermore, females were detected to have extra‐pair young with their sons (N ¼ 2, in 3 years), but never with their father, nor with their brothers (Wheelwright et al., 2006). Given that extra‐pair paternity is common and that brood parasitism is rare or absent (in the savannah sparrow, and in many other birds), a female is always related (f ¼ .25) to her sons, and to her brothers (although on average less, because they can be full or half‐sibs), but not necessarily to her social father. Hence, when only considering the average levels of relatedness, one would predict mother–son pairs to be less common than father–daughter pairs, which is the opposite of what Wheelwright et al. (2006) found. Wheelwright et al. also observed that mother–son pairs almost always involved first‐time breeding males and older females. For reasons outlined above, females might actually increase their reproductive success by mating with their son, if otherwise the son has to forego breeding (Kokko and Ots, 2006), which seems not unlikely for juvenile male Savannah sparrows (see Wheelwright et al., 2006, who provide an alternative explanation for why older females are more likely to ‘‘accept’’ inbreeding). The hypothesis that female promiscuity has evolved to increase offspring heterozygosity also makes some assumptions about mechanisms. One possibility is that females can actively select a male that is genetically less similar than their social mate as extra‐pair mate. That does not imply that females have to be able to accurately assess relatedness. Simple rules of thumb, such as ‘‘choose a male that is breeding far away’’ (Foerster et al., 2003; Waser and DeWoody, 2006), or ‘‘mate with a male outside the social

228

BART KEMPENAERS

group’’ (Brooker et al., 1990; Sillero‐Zubiri et al., 1996) might suffice. Several studies have now shown evidence for a small‐scale genetic structure within a population, which would facilitate such simple rules (e.g., Double et al., 2005; Foerster et al., 2006; Shorey et al., 2000; Stow and Sunnucks, 2004; Woxvold et al., 2006). Alternatively, a postcopulatory choice process might take place after females copulated with multiple males, so that the most compatible sperm has a higher probability to fertilize the eggs (Birkhead and Pizzari, 2002). There are several lines of evidence suggesting that this is indeed possible. (1) The female reproductive tract often has multiple sperm storage organs, which might facilitate female control of paternity (e.g., Briskie, 1996; Snow and Andrade, 2005). (2) It is known that certain polymorphic genes are expressed on the sperm surface, for example, bindin (Levitan and Ferrell, 2006), and olfactory receptors (Fukuda and Touhara, 2006; Spehr et al., 2006; Vanderhaeghen et al., 1993). Together with Mhc genes they are probably involved in sperm–egg communication and might provide a mechanism for cryptic female choice (Fukuda et al., 2004; Levitan and Ferrell, 2006; Ziegler et al., 2005). (3) Experiments where sperm velocity of different male Arctic charr (Salvelinus alpinus) was measured in the ovarian fluid of different females, showed a highly significant male–female interaction, suggesting that females can differentially stimulate sperm depending on male characteristics (Urbach et al., 2005). Surprisingly, the idea of the existence of postcopulatory sperm choice remains controversial, despite general agreement that such processes clearly work in relation to inbreeding in plants (Bernasconi et al., 2004), and in the context of hybridization with other species or geographic races (outbreeding) in animals and plants (Howard, 1999). What if there is no pre‐ or postcopulatory mechanism to select a less‐ related male, or his sperm? Could females benefit from mating multiply, thereby increasing the chance to have at least some young fathered by a more compatible male? Several studies have addressed the issue, and the answer seems to be no, in most cases (Colegrave et al., 2002). Theoretically, the average female has a 50% chance to mate with a more related male, and a 50% chance to mate with a less‐related male. Females mating with multiple males will increase the variance in fitness, but the mean fitness will only be higher if the fitness increase resulting from mating with a less‐ related male is higher than the fitness reduction due to mating with a more related male. Thus, the relationship between fitness and mate relatedness must be nonlinear and convex (Hosken and Blanckenhorn, 1999). In reality, mating with multiple males may benefit those females that are closely related to their social partner, or to their previous mating partner, but it is unlikely that it will benefit all females in the population.

MATE CHOICE AND GENETIC QUALITY

229

2. Review of the Evidence What is the current evidence that mate choice maximizes or optimizes offspring heterozygosity, generally, or at specific loci such as the Mhc? Below, I discuss this in the context of both social and extra‐pair mate choice, and with respect to multiple mating. a. Can females increase offspring heterozygosity through mate choice? Some studies explicitly tested whether females increase the heterozygosity of their offspring by mating with a genetically dissimilar male. These studies demonstrated that offspring heterozygosity correlates negatively with parental relatedness (e.g., in the white‐toothed shrew, Crocidura russula, r ¼ .68, N ¼ 81 pairs, p < .001, Duarte et al., 2003; in Seychelles warblers, r ¼ .51, N ¼ 97 pairs, p < .001, Richardson et al., 2004; in barn swallows, Hirundo rustica, r ¼ .51, N ¼ 279 parent–offspring combinations, p < .001, Kleven et al., 2005; in house finches, r ¼ .74, N ¼ 46 pairs, p < .01, Oh and Badyaev, 2006). b. Choice of social mate or copulation partner Eleven studies have used multilocus minisatellite or microsatellite markers to test whether the social mate is less related to the female than expected under random mating (Table IVa). Only two studies found evidence for this hypothesis, at least under specific conditions. Oh and Badyaev (2006) showed how choice for ornamented males traded off with choice for genetic dissimilarity. Early in the season, female house finches found high variation in ornamentation among potential mates, and the average relatedness to those males was relatively low. Hence, mate choice was based on ornament size, not on relatedness. However, for females that arrived later in the season, the variance in ornamentation among available males had decreased, whereas the mean relatedness to unpaired males was higher. As a consequence, females selected among all available males those to which they were less related (Oh and Badyaev, 2006). In Australian sleepy lizards (Tiliqua rugosa), females were also less related to their social mate than expected under random mating, but only in comparison to the other males within the female’s home range (Bull and Cooper, 1999). Presumably, those are the males she could choose from. The other studies did not find evidence for nonrandom mating, often despite large sample sizes and known fitness effects of heterozygosity (Table IVa). A study on blue tits even found that breeding pairs on an island (but not in a mainland population) were more related than expected under random mating, despite a cost in terms of reduced hatching success (Krokene and Lifjeld, 2000).

TABLE IV TESTS OF THE HYPOTHESIS THAT FEMALES ARE LESS RELATED TO THEIR MATE THAN EXPECTED UNDER RANDOM MATING

Species Barn swallow Blue tit

230

Great reed warbler Great snipe House finch House sparrow

Savannah sparrow

Markera

Number of locib

Estimatec

Sample size

Effectd

p

Comment

(a) Choice of social mate or copulation partner (chosen male vs other males in the population) Micro 9 R 188 No .61 Multi D 42 Neg <.001 Pairs versus random dyads (island) Multi D 28 No >.25 Pairs versus random dyads (mainland) Micro 5–7 R 219 No .68 Mhc D 279 No .99 Mhc D 32 No .62 Pairs versus other males on the lek Micro 14 R 42 No .18 Early pairs Micro 14 R 54 Pos .04 Late pairs Multi D 98 No .60 Pairs versus random dyads Micro 7 R 37 No >.05 Also compared to nonbreeding males Mhc D 37 No .31 Mhc D 37 Neg .006 Pairs compared to nonbreeding males Micro 5 R 192 No >.46 Analyzed for 4 years separately Mhc D (rank) 23 Pos <.005 Yearling females Mhc D (rank) 23 No >.50 Adult females

H‐fitness correlation (yes/no)e

References

– Yes

1 2

No

2

Yes – –

3 4 5

Yes Yes No

6 6 7

–

8

Yes Yes

8 8

Yes

9

– –

10 10

Seychelles warbler

Pedigree

1110

Pos

<.05

Pedigree

1110

No

.63

Micro

14

R

78

No

.73

Micro

14

R

19

Neg

<.01

Mhc Pedigree

F

82 479

No No

.41 >.17

Pedigree

F

15

No

.85

Pedigree

No

.13

Tree swallow

Pedigree

Neg

<.001

Tu´ngara frog

Micro

7

R

153

No

.61

Australian sleepy lizard

Micro

4

R

17

No

>.90

Micro

4

R

17

Pos

<.05

Micro Mhc Mhc

5 1 1

R N Nnew

24 65 22

No No Neg

.33 .55 .09

Song sparrow

231 Ornate dragon lizard Malagasy giant jumping rat

Father–daughter pairs (obs vs expected) All inbred pairs (obs vs expected) Primary female and social male (pairs) Subordinate female and social male Analyzed for years separately Pair versus nearest neighbor (mean of 15 years) Parent–offspring matings (obs vs expected) Full or half‐sib matings (obs vs expected) Pairs versus other males on the lek Male partner versus all other males Male partner versus other males in home range

New male mate versus random

Yes

11

Yes

11

Yes

12

Yes

12

– Yes

13 14

Yes

14

Yes

14

Yes

15

–

16

–

17

–

17

– – –

18 19 19 (Continued)

TABLE IV (Continued) Number of locib

Estimatec

Mhc

1

A

65

No

.63

Reindeer

Micro

17

R

88

No

p > .40

Atlantic salmon

Micro Mhc Mhc

5 1 1

R N A

221 221 221

No No Pos

.45 .22 <.05

Wood frog

Micro

9

R

27?

No

>.35

Cunningham’s skink

Micro

10

R

55

Pos

<.001

Ornate dragon lizard

Micro

7

R

20

No

.20

Northern water snake

Micro

5

R

?

No

>.16

Species

Markera

Sample size

Effectd

p

Comment Genotypic distance at the peptide‐ binding region

232

H‐fitness correlation (yes/no)e

References

–

19

–

20

Yes Yes Yes

21 21 21

Yes

22

–

23

–

18

No

24

(b) Choice of sire (sire vs all other males in population)

Genotypic distance at the peptide‐ binding region Sire versus average pairwise R of the population Breeding pairs versus potential mates Sire versus all other males Tested at two study sites

Banner‐tailed kangaroo rat

Micro

4–11

R

54

Pos

.02

Micro

4–11

R

220f

Pos

.05

Micro

4–11

R

169f

Pos

<.001

D

11

No

>.05

North American pika

Multi

Soay sheep

Mhc

5

N

518

No

>.17

White‐toothed shrew

Micro

12

R

65

Neg

<.05

a

Only significant for sedentary females Neighbor sire versus neighbor nonsire Sire versus neighbor nonsire More sires with intermediate D than expected Tests for five microsats linked to Mhc Not sign. when sire compared to neighbors

–

25

–

25

–

25

–

26

–

27

No

28

233

Marker used to measure relatedness: Mhc, microsatellite, multilocus minisatellite, pedigree. If blank: multilocus, unknown number. c A ¼ genotypic distance (sum of all pairwise amino acid differences of the four possible alleles), D ¼ band‐sharing coefficient (see Table I), F ¼ inbreeding coefficient, N ¼ number of shared alleles, Nnew ¼ number of new alleles obtained through mating, R ¼ relatedness. d Pos: expected effect (p < .10); Neg: opposite effect (p < .10); No: no effect (p > .10 or unknown). Bold when significant. e Yes: evidence for a significant relationship between heterozygosity and fitness, or inbreeding depression (in bold when tested in the same study), No: no significant relationship between heterozygosity and fitness (or inbreeding depression), – ¼ no information. f Analyses based on number of sired offspring, not on number of sires. 1. Kleven et al., 2005; 2. Krokene and Lifjeld, 2000; 3. Foerster et al., 2003; 4. Westerdahl, 2004; 5. Ekblom et al., 2004; 6. Oh and Badyaev, 2006; 7. Edly‐Wright et al., 2007; 8. Bonneaud et al., 2006a; 9. Freeman‐Gallant et al., 2006; 10. Freeman‐Gallant et al., 2003; 11. Wheelwright et al., 2006; 12. Richardson et al., 2004; 13. Richardson et al., 2005; 14. Keller and Arcese, 1998; 15. Shutler et al., 2004; 16. Lampert et al., 2006; 17. Bull and Cooper, 1999; 18. LeBas, 2002; 19. Sommer, 2005; 20. Holand et al., 2006; 21. Landry et al., 2001; 22. Halverson et al., 2006; 23. Stow and Sunnucks, 2004; 24. Weatherhead et al., 2002; 25. Waser and DeWoody, 2006; 26. Peacock and Smith, 1997; 27. Paterson and Pemberton, 1997; 28. Duarte et al., 2003. b

234

BART KEMPENAERS

Similarly, only one out of five studies showed evidence that females prefer more Mhc dissimilar males as social partner (Table IVa). In the island population of Savannah sparrows (discussed above), Freeman‐ Gallant et al. (2003) showed that yearling females paired almost exclusively with Mhc dissimilar males, whereas older females paired at random with respect to Mhc similarity. To explain this age difference, they argued that older females are more constrained in their mate choice, because they are highly philopatric and breeding dispersal may be costly. In contrast, Bonneaud et al. (2006a) found that female house sparrows (Passer domesticus) shared more Mhc alleles with their social mate than with available, nonbreeding males. They argued that females might have avoided males that were too dissimilar (no shared alleles), supporting the hypothesis that selection favors an intermediate number of Mhc alleles. Three pedigree‐based studies are included in Table IVa that tested whether inbreeding occurs more often than expected by chance. Two of those have been discussed above; additionally, Shutler et al. (2004) identified three tree swallow (Tachycineta bicolor) pairs consisting of former nestmates (either full or half‐sibs, this was unknown). They calculated that the probability of this happening by chance is exceedingly small. No negative consequences of inbreeding could be detected, but unfortunately no information about paternity was available for these inbred pairs. It is not unlikely that the social male did not father any of the offspring, because a paternity study showed that full loss of paternity occurred in almost one nest out of five (Kempenaers et al., 1999). It is perhaps less likely to find social mate choice based on relatedness, because of trade‐offs with direct benefits. The latter do not play a role in lekking mating systems, but the two studies that tested it failed to find that males that obtained copulations were genetically less similar to the female than a random male on the lek (Ekblom et al., 2004; Lampert et al., 2006; Table IVa). Similarly, direct benefits are probably less important in harem‐polygynous reindeer (Rangifer tarandus), but there is no evidence for nonrandom mate choice with respect to relatedness (Holand et al., 2006). c. Divorce and second broods Several studies investigated whether remating decisions are influenced by pair relatedness. Even if females (or males) are not able to assess the relatedness with their social mate directly, they could use the presence of unhatched eggs as a cue. Kempenaers et al. (1998) first tested this in blue and great tits. Although there was no evidence that the decision to divorce depended on the presence of unhatched eggs, egg hatchability increased after divorce for both male and female blue tits, which was not the case for faithful pairs (Kempenaers et al., 1998). Because the analysis was restricted to pairs that had at least one unhatched egg in

MATE CHOICE AND GENETIC QUALITY

235

their clutch in the year before the divorce, it suffers from the problem of ‘‘regression to the mean’’ (Kelly and Price, 2005). Meanwhile, three studies on tits have tested the hypothesis using molecular markers to estimate relatedness. In coal tits, Parus ater, multilocus band‐sharing values between pair members failed to explain the probability of divorce (N ¼ 24, p ¼ .55), or the probability that a pair produced a second brood (N ¼ 61, p ¼ .75; Schmoll et al., 2005b). In blue tits, females were less related to a partner they stayed with across years than to a partner they divorced, but the difference was clearly not significant (N ¼ 25, p > .30; Foerster et al., 2006; based on five to seven microsatellite markers). Furthermore, the relatedness with the new partner was lower, but not significantly so (for males: N ¼ 12, p > .60; for females: N ¼ 12, p > .20). In great tits, Van de Casteele et al. (2003) also found that divorced females were less related to their new mate than to their previous mate, but here the difference was significant (N ¼ 13, p < .02; estimates based on nine microsatellite markers). In the latter two studies, but not in the first, negative effects of homozygosity on fitness‐related traits were found. Another study, on the cooperatively breeding long‐tailed tit (Aegithalos caudatus), tested the hypothesis that divorce is linked to inbreeding avoidance (Hatchwell et al., 2000). In cooperative breeders the risk of inbreeding is higher when offspring (here, males) stay in the territory of their parents until they are reproductively active. Females could then avoid inbreeding with their sons by leaving the family group. Hatchwell et al. (2000) found that successful pairs (those that presumably produced sons) were more likely to divorce than pairs that failed to produce offspring, and that females remated outside the family flock, supporting the hypothesis. d. Choice of sire In many study systems, the identity of the social mate is unknown, or there is no social mate, and behavioral information on copulation partners is often missing or does not reliably reflect paternity. As long as potential mates can be sampled, paternity analysis can then be used to test whether the female is less related to the male that sired the offspring compared to other males in the population. Because in these systems direct benefits are often absent, choice for additive and nonadditive genetic benefits might become more important. Some evidence in favor of the hypothesis was found in three out of nine studies that used either microsatellite markers, Mhc markers or both (Table IVb). The strongest effect was found in a study on the long‐term socially monogamous Cunningham’s skink (Egernia cunninghami), where the relatedness of breeding pairs was about 10 times lower than the relatedness of nonbreeding pairs (Stow and Sunnucks, 2004). This is an

236

BART KEMPENAERS

exceptional case of inbreeding avoidance, probably via direct mate choice. The results might be explained by low levels of adult dispersal leading to a high number of close relatives in the environment. A study on Atlantic salmon showed that sires did not differ from nonsires, either in terms of relatedness (estimated using microsatellites) or by the number of shared Mhc alleles (Landry et al., 2001). However, Landry et al. (2001) noted that females might prefer partners that maximize the complexity of the Mhc genotype of their offspring, that is, the number of alleles that encode proteins that can bind different peptides. Therefore, they calculated a genotypic distance value for each pair, which is the sum of all the pairwise amino acid differences. In support of their hypothesis, they found that the genotypic distance value for sires was significantly higher than expected under random mating, when genotypic distance was calculated based on the amino acids at the peptide‐binding region (reflecting functionality). Finally, Waser and DeWoody (2006) found that female banner‐tailed kangaroo rats (Dipodomys spectabilis) were less related to the male that sired their offspring than to a random male in the population. However, this effect was only significant for a subset of sedentary females. Furthermore, they reported (a) that nearest neighbors that sired offspring were less related to the female than nearest neighbors that did not sire offspring, and (b) that when the nearest neighbor was not the sire, then the sire was less related to the female than the nearest neighbor. Contrary to expectation, females of the greater white‐toothed shrew were more related to their mate than expected under random mating (Duarte et al., 2003). However, there was no evidence for nonrandom mating within neighborhoods, and the population‐wide effect is probably due to limited dispersal. Interestingly, no negative consequences of inbreeding could be detected in this population (Duarte et al., 2003). A study on pikas (Ochotona princeps) showed no difference between multilocus band‐sharing values of reproducing pairs and those of potential mates. However, the variance in band‐sharing values was much lower among mated pairs than among potential pairs ( p < .003), suggesting a preference for males of intermediate genetic similarity (Peacock and Smith, 1997). Unfortunately, information on the fitness consequences of in‐ or outbreeding is not available for this population. e. Extra‐pair mate choice There is a rich literature on extra‐pair paternity, particularly in birds, and many researchers have used their genetic marker data to test whether extra‐pair mate choice may function to increase offspring heterozygosity. Three separate, but closely linked predictions have been tested (reviewed in Table V).

TABLE V EXTRA‐PAIR PATERNITY AND FEMALE CHOICE FOR INCREASED OFFSPRING HETEROZYGOSITY

Species

Markera

Number of locib

Estimatec

Sample size

Effectd

p

Comment

H‐fitness correlation (yes/no)e

References

237

(a) Prediction: relatedness with the social mate predicts the presence of extra‐pair young (yes/no, unless otherwise mentioned) 92 No .11 – Black‐throated blue Micro 5 Cd2 warbler Blue tit Multi D 103 No >.30 Yes Multi D 46 No .17 Island population Yes Multi D 28 No .65 Mainland population No Micro 4–7 R 177 No .85 – Micro 5–7 R 202 No >.80 Yes Coal tit Multi D 202 No .27 Similar result with No proportion of EPY Common sandpiper Multi D 15 Pos .014 Includes cases of – quasi‐parasitism House sparrow Micro 10 R 57 No .78 Proportion of EPY – Multi D 103 No .27 Proportion of EPY No Kentish plover Multi D 65 Pos <.005 Includes cases of – quasi‐parasitism Mexican jay Multi D 31 Pos <.025 Similar result with Yes proportion of EPY Pied flycatcher Multi D 36 Neg <.001 In‐ and outbreeding Yes depression Reed bunting Micro 9 R 53 No .27 Similar result with No proportion of EPY

1 2 3 3 4 5 6

7 8 9 7 10

11 12

(Continued)

TABLE V (Continued)

Species Savannah sparrow

Markera

Number of locib

Mhc Multi

Estimatec

Sample size

Effectd

p

D (rank) D

43 33

Pos Pos

<.007 .09

Micro Micro

5 14

R R

144 98

Pos No

.009 >.9

14

Splendid fairy‐wren

Micro Mhc Micro

6

R D R

82 82 114

No No Pos

.80 .54 .01

Micro

6

R

68

Pos

.052

Micro

6

R

29

Pos

.009

Multi Multi

D D

72 37

No No

.85 .45

Multi

D

72

Neg

.059

Multi

D

37

Neg

.007

D R

25 68

Pos Pos

<.001 .002

238

Seychelles warbler

Tree swallow

Western sandpiper Alpine marmot

Multi Micro

5–12

Comment Mhc and multilocus D pos. correlated Similar result for subordinate females

Higher R in broods with only EPY R with social mate higher than with random male for nests with EPY R with social mate higher than with random male for nests with only EPY All broods Broods where all eggs were typed Proportion of EPY, all broods Proportion of EPY, all eggs typed Quadratic function

H‐fitness correlation (yes/no)e

References

– Yes

13 13

Yes Yes

14 15

Yes Yes No

16 16 17

No

17

No

17

Yes Yes

18 18

Yes

18

Yes

18

– Yes

7 19

Common mole‐rat

Micro

7

R

21

No

.61

Presence of extra‐ colony paternity

–

20

– –

21 21

–

1

Yes – Yes No Yes

2 4 22 6 23

No

24

Yes – – –

25 8 26 26

No No

12 12

(b) Prediction: female is less related to extra‐pair than to within‐pair male Barn swallow

Black‐throated blue warbler Blue tit

239

Coal tit Great reed warbler

Micro Micro

9 9

R R

90 90

Neg Neg

.058 .024

Micro

5

Cd2

25

No

.31

D R R D D

18 61 96 63 5

No No No No Pos

>.50 .71 .54 .49 –

D

–

No

–

7 11 20 20

Pos No No Pos

.04 .17 .16 .028

20 20

No No

.67 .24

Multi Micro Micro Multi Multi

4–7 5–7

Mhc House finch House sparrow House wren

Reed bunting

Micro Micro Micro Micro

14 10 5–7 5–7

R R R

Micro Micro

9 9

R R

Extra‐pair sire versus other breeding males

In four cases out of five Mentioned in paper (no details)

Extra‐pair sire has more rare alleles Extra‐pair sire versus nearest neighbor

(Continued)

TABLE V (Continued)

Markera

Number of locib

Seychelles warbler

Micro

Splendid fairy‐wren

Species

240

Estimatec

Sample size

Effectd

p

Comment

14

R

40

No

.22

Similar result for subordinate females

Mhc Micro

6

D R

29 107

No No

.61 .19

Micro

6

R

48

Pos

.022

Alpine marmot

Micro

5–12

R

9

No

.51

Common mole‐rat

Micro

7

R

18

Pos

<.05

All females with EPY For females with only EPY Not less related, not ‘‘optimally related’’ Extra‐colony male versus social male

H‐fitness correlation (yes/no)e

References

Yes

15

Yes No

16 17

No

17

Yes

19

–

20

–

1

Yes

22

–

8

No

12

(c) Prediction: extra‐pair young are more heterozygous than within‐pair young Black‐throated blue warbler Blue tit

Micro

5

d2

154

No

.32

Micro

5–7

SH

101

Pos

.032

House sparrow

Micro

10

H

21

Neg

.07

Reed bunting

Micro

9

SH

30

No

.77

Due to nonlocal extra‐pair males (N ¼ 44, p ¼ .001) EPY compared to expected H of WPY

Seychelles warbler

Micro Micro

14 14

Mhc Splendid fairy‐wren

H H

98 21

No Pos

.33 .039

N

84

No

.76

Bluegill sunfish

Allo Micro Micro

9 6 11

H SH Mean d2

7 51 1799f

Pos No Pos

.016 .15 <.001

Common mole‐rat

Micro

7

SH

37

Pos

.05

a

For primary females For subordinate females Population‐wide, not pairwise Similar result for IR Parental versus cuckolder offspring Not pairwise (litters with different sires)

Yes Yes

15 15

Yes

16

– No Yes

27 17 28

–

20

Marker used to measure relatedness: allozyme, Mhc, microsatellite, multilocus minisatellite. If blank: multilocus, unknown number. c Cd2 ¼ mean squared difference of all four pairwise comparisons in d2 of a female’s allele sizes to that of her mate, D ¼ band‐sharing coefficient, H ¼ heterozygosity, IR ¼ internal relatedness, SH ¼ standardized heterozygosity, N ¼ number of bands as estimate of within‐individual allelic diversity (see Table I), R ¼ relatedness. d Pos: expected effect (p < .10); Neg: opposite effect (p < .10); No: no effect (p > .10 or unknown). Bold when significant. e Yes: evidence for a significant relationship between heterozygosity and fitness, or inbreeding depression (in bold when tested in the same study), No: no significant relationship between heterozygosity and fitness (or inbreeding depression), – ¼ no information. f Individual offspring used as the unit of analysis. 1. Smith et al., 2005; 2. Kempenaers et al., 1996; 3. Krokene and Lifjeld, 2000; 4. Charmantier et al., 2004; 5. Foerster et al., 2006; 6. Schmoll et al., 2005b; 7. Blomqvist et al., 2002; 8. Stewart et al., 2006; 9. Edly‐Wright et al., 2007; 10. Eimes et al., 2005; 11. Ra¨tti et al., 1995; 12. Kleven and Lifjeld, 2005; 13. Freeman‐Gallant et al., 2003; 14. Freeman‐Gallant et al., 2006; 15. Richardson et al., 2004; 16. Richardson et al., 2005; 17. Tarvin et al., 2005; 18. Barber et al., 2005; 19. Cohas et al., 2006; 20. Bishop et al., 2007; 21. Kleven et al., 2005; 22. Foerster et al., 2003; 23. Bensch et al., 1994; 24. Westerdahl, 2004; 25. Oh and Badyaev, 2006; 26. Masters et al., 2003; 27. Brooker et al., 1990 (calculated from data in Appendix); 28. Neff, 2004. b

241

242

BART KEMPENAERS

Prediction 1: The relatedness with the social mate predicts the presence or proportion of extra‐pair young in the brood. This prediction was clearly supported in 7 out of 16 species in which it was tested (Table Va). In these studies, females that had extra‐pair young were significantly more related to their social mate than females without extra‐pair young. In the Savannah sparrow population, this conclusion was supported independent of the type of marker used (Table Va; Freeman‐Gallant et al., 2003, 2006). In a study on Alpine marmots (Marmota marmota), the relationship was quadratic, so that the probability of extra‐pair paternity was lowest in pairs of intermediate relatedness (Cohas et al., 2006). This suggests that excessive outbreeding also has negative fitness consequences, although the only available evidence in the marmot is for inbreeding depression (Da Silva et al., 2006). Two other studies support the idea of outbreeding avoidance: in tree swallows, the proportion of extra‐pair young in a brood decreased with increasing band‐sharing between the social pair (Barber et al., 2005). The same study did not find a relationship between the proportion of eggs that hatched and the band‐sharing coefficient in 22 nests without extra‐pair young. Ra¨tti et al. (1995) reported a similar result in the pied flycatcher. Here, reproductive success was highest at intermediate band‐sharing values. Interestingly, there is evidence that female collared flycatchers that are paired to male pied flycatchers engage in extra‐pair copulations with conspecific males to avoid or attenuate the negative effects of hybridization (Veen et al., 2001). Alternatively, a female collared flycatcher’s decision to engage in extra‐pair copulations may be independent of her pairing status (conspecific or heterospecific male), but conspecific sperm may be more likely to fertilize eggs (Veen et al., 2001). Indeed, conspecific sperm precedence would also explain why broods of heterospecifically mated females end up with more extra‐pair young. Prediction 2: The female is less related to the extra‐pair than to her social mate. This was clearly supported in only 4 out of the 17 studies that tested it (Table Vb). In one study (on great reed warblers) the data are suggestive, but the sample size was too small to support it statistically (Bensch et al., 1994). In a study on house wrens (Troglodytes aedon), the extra‐pair males were not significantly less related, but they had significantly more alleles that were rare in the population compared to the within‐pair males (Masters et al., 2003). The authors suggest that mating with males with rare alleles will increase offspring heterozygosity, but they do not show evidence that this is indeed the case. It might seem surprising that so few studies find the predicted effect, because Prediction 2 can be tested with a pairwise test, which is more powerful compared to the nonpaired tests of prediction 1. However, the

MATE CHOICE AND GENETIC QUALITY

243

sample sizes were usually smaller (compare Table Va with Vb), and for 9 out of the 11 studies that tested both the first and second prediction, the results were similar. Furthermore, it is seldom possible to identify all extra‐ pair fathers, and the ones that are sampled might be a biased subset with respect to relatedness (see Foerster et al., 2003). Prediction 3: Extra‐pair young are more heterozygous than within‐pair young. There is some support for this prediction in three out of the six bird species in which it has been tested, and in the common mole‐rat Cryptomys hottentotus hottentotus (Table Vc). In blue tits, the effect was significant when all broods were considered, but it was much stronger when the analysis was restricted to broods where the extra‐pair young were fathered by nonlocal extra‐pair males (Foerster et al., 2003; see Fig. 5 above). In the Seychelles warbler, the effect was only significant when within‐ and extra‐pair young of subordinate females were compared. These females sometimes reproduce with the territorial male, to whom they are more closely related than primary females (Richardson et al., 2004). In the study on bluegill sunfish, Neff (2004) tested whether offspring differed in genetic diversity (estimated using mean d2) depending on whether they were fathered by a cuckolder or a parental male. Remember that in this species there was stabilizing selection on genetic diversity (see previous section). Parental offspring had lower mean d2 values than expected under random mating and values closer to the optimum, whereas the opposite was true for cuckolder offspring. Parental offspring thus had significantly lower genetic diversity than cuckolder offspring (Neff, 2004). Note that individual offspring were used as the unit of replication (Neff, 2004), but accounting for nest (as a proxy for family) gave similar results (Bryan Neff, personal communication). In any case, in this system, mate choice only takes place between females and parental males, whereas cuckolders spawn opportunistically. The study thus provides evidence that mate choice optimizes offspring genetic diversity in this species. In humans there is also evidence that extra‐pair mating is related to genetic diversity at the Mhc: women had more extra‐pair partners when they shared more Mhc alleles with their social partner (Garver‐Apgar et al., 2006). f. Relatedness‐dependent paternity bias Females might use multiple mating in combination with pre‐ or postcopulatory processes to ensure that their eggs are fertilized by the most compatible sperm. This hypothesis has been tested repeatedly, in correlational and experimental studies, and support for it is substantial (Table VI). In birds and other animals, single broods often contain offspring that are sired by two (or more) males. Even if the copulation history of a female is unknown, as is often the case, broods with multiple paternity offer a great

TABLE VI TESTS OF PATERNITY BIAS IN RELATION TO GENETIC SIMILARITY BETWEEN MATES Species

Methoda

Experiment/observationb

244

Diplosoma listerianum (ascidian) Cricket, decorated field

Multi

Pairings between clones

Pedigree

Double matings: SN or SS

Cricket, field

Pedigree

Cricket, field

Pedigree

Fruitfly

Pedigree

Red flour beetle

Strain

Double matings: SS, SN, NS, NN Double matings: SS, SN, NS, NN Double matings: SN, S1/2N, CN, NN Double matings: same or different strain to female

Guppy

Pop

Artificial insemination with sperm from native or foreign male

Resultc

References

Mating success correlates negatively with genetic similarity P1 lower in SN than in SS matings P1 higher than 50% in NS treatment, random in SN P1 higher than 50% in NS treatment, random in SN P1 lowest in SN matings

1

P1 higher when first male of same strain as female; no effect on P2 (outbreeding avoidance) Paternity bias in favor of sperm from native males (outbreeding avoidance)

2 3 4 5 6

7

Alpine newt

Micro

Sand lizard

Multi Mhc

Female allowed to mate with two males Female mated with two males

Blue tit

Micro

Broods with multiple paternity

Mallard

Pedigree

Ruff

Micro

Artificial insemination with sperm of S and N Broods with multiple paternity

Agile antechinus

Micro

Female mated with two males

a

Paternity bias in favor of less‐related male Paternity bias in favor of the less related and less Mhc related male Paternity bias in favor of the less‐related male No paternity bias Paternity bias in favor of less‐ related male Paternity bias in favor of less‐ related male

8 9

10 11 12 13

Field populations (Pop), laboratory strains (strain), microsatellite (micro), multilocus minisatellite (multi), pedigree‐based (pedigree). C ¼ cousin, N ¼ nonsibling (unrelated), S ¼ sibling (full‐sib), S1/2 ¼ half‐sib. c P1 ¼ paternity obtained by the male that mated first. 1. Bishop et al., 1996; 2. Stockley, 1999; 3. Bretman et al., 2004; 4. Simmons et al., 2006; 5. Mack et al., 2002; 6. Pai and Yan, 2002; 7. Ludlow and Magurran, 2006; 8. Garner and Schmidt, 2003; 9. Olsson et al., 1996, 2004; 10. Foerster et al., 2006; 11. Denk et al., 2005; 12. Thuman and Griffith, 2005; 13. Kraaijeveld‐Smit et al., 2002; 14. Ru¨licke et al., 1998. b

245

246

BART KEMPENAERS

opportunity to test whether the relatedness with the female influences the proportion of offspring sired by each male. So far, a nonexperimental test of the hypothesis has been reported in two bird species, and in the sand lizard (Lacerta agilis). In all three, the result was similar: the male that was less related to the female sired a larger proportion of her eggs. In the sand lizard, sample size was small, but the result was confirmed in controlled laboratory matings (Fig. 6A), and further experimental work showed that Mhc relatedness also negatively, and independently, influenced a male’s probability to sire offspring (Olsson et al., 2004). In a study on blue tits (Fig. 6B), the difference in relatedness between the female and the two males explained a significant, but small amount (only 4.8%) of the variation in the paternity bias. However, this is not surprising, given that: (a) the estimate of relatedness was based on only few (five to seven) microsatellite markers, and (b) unknown variation in the number and timing of copulations with the social and extra‐pair male(s) might have influenced the outcome. A study on the ruff (Philomachus pugnax) also showed a clear‐cut result: in 12 out of 13 broods where paternity was unequally divided between 2 males, the more distantly related male sired the larger proportion of the brood (effect size r ¼ .48, N ¼ 13, p ¼ .007; based on 7 microsatellite markers; Thuman and Griffith, 2005). Table VI lists 11 studies that have experimentally addressed the question whether paternity bias after multiple mating is affected by genetic similarity, usually by mating females with 2 males that differ in relatedness. Three studies on crickets have shown that when females mated with a sibling (S) and a nonsibling (N), the nonsibling sired a disproportionately larger number of the offspring (Bretman et al., 2004; Simmons et al., 2006; Stockley, 1999), suggesting that postcopulatory sperm selection might occur. However, the story is more complicated. An earlier study on crickets with a similar double mating design (as described above) already showed that egg hatchability was higher if at least one of the mated males was unrelated to the female (Tregenza and Wedell, 2002; Fig. 3A). This is strong support for the idea that females can avoid the negative effects of inbreeding by mating with multiple males, and that a postcopulatory mechanism is involved. However, subsequent studies on different cricket species did not support the earlier finding, that is, they did not find that NS or SN matings led to higher hatching success compared to SS matings (Jennions et al., 2004; Simmons et al., 2006; Fig. 3B). Nevertheless, both Simmons et al. (2006) and Bretman et al. (2004) found a paternity bias in favor of the unrelated male, but only when the unrelated male mated first. These studies illustrate how difficult it is to find patterns that have a more general validity. The cricket system is very well studied, and highly amenable to experimentation. Even then, the results vary substantially. In a similar experiment with

247

C

Sired relative to the social male (%)

B

Sired relative to male 2 (%)

A

Sired relative to other males (%)

MATE CHOICE AND GENETIC QUALITY

60 40 20 0 −20 −40 −60 −30

−20 −10 0 10 Band-sharing relative to other male

20

60 40 20 0 −20 −40 −60 −0.8

−0.6 −0.4 −0.2 0.0 0.2 0.4 Relatedness relative to social male

0.6

−10 −5 0 5 10 15 Allele-sharing relative to male 2

20

20 0 −20 −40 −60 −15

FIG. 6. Paternity bias (the proportion of offspring sired by male x relative to male y) in mixed paternity broods is explained by the difference in relatedness of the males to the female. (A) Sand lizards (L. agilis); filled circles: field data from natural matings (r ¼ .85, N ¼ 7, p < .02), open circles: experimental data from controlled laboratory matings (r ¼ .84, N ¼ 7, p < .02); relatedness estimated as band‐sharing coefficient from multilocus minisatellite fingerprinting. Adapted from figure in Olsson et al. (1996) by permission from Macmillan Publishers Ltd.: Nature, copyright 1996. (B) Blue tits (C. caeruleus); effect size r ¼ .22, N ¼ 99, p ¼ .03; relatedness estimated using five to seven microsatellite markers (see Foerster et al., 2006). (C) Agile antechinus (A. agilis); effect size r ¼ .57, N ¼ 13, p ¼ .054; relatedness estimated as proportion of alleles shared, based on five microsatellite markers. Adapted from Fig. 3 in Kraaijeveld‐ Smit et al. (2002) by permission from the authors and Blackwell Publishing.

248

BART KEMPENAERS

Drosophila, Mack et al. (2002) did not only use full siblings, but also half sibs and cousins in combination with an unrelated male. Interestingly, the predicted effect, lower siring success of related males, was only significant for full‐sibs (Fig. 7). This is in contrast with the more subtle effects of relatedness suggested by studies in natural populations (see above). Despite the issues raised above, there are several good reasons to believe that paternity bias with respect to genetic similarity is a general phenomenon. 1. Only one of the studies listed in Table VI did not find the predicted effect, despite a carefully designed experiment. Denk et al. (2005) artificially inseminated female mallards (Anas platyrhynchos) with a mixture of equal numbers of sperm from a full‐sib and an unrelated male. They showed that relatedness did not explain any variation in the success in obtaining paternity, whereas sperm motility and swimming speed did (Denk et al., 2005). 2. In two studies the experiments were not specifically designed to test the hypothesis, that is, the experimenters used males of different size classes instead of different classes of genetic similarity. Nevertheless, both studies found an effect of genetic similarity (Fig. 6C, Garner and Schmidt, 2003; Kraaijeveld‐Smit et al., 2002). 3. A similar effect in support of avoidance of outbreeding depression was observed when males of different strains or populations were used in experiments, so that the ‘‘local’’ male was more successful (Ludlow and Magurran, 2006; Pai and Yan, 2002). Furthermore, under sperm competition between closely related species, there is convincing evidence for conspecific sperm precedence (e.g., Rugman‐Jones and Eady, 2007 and references therein). 4. Even without multiple mating, females might be able to increase offspring heterozygosity through sperm selection. Marshall et al. (2003) reported evidence for such a process in a study on sedge warblers. They used nine microsatellite markers to calculate all possible mean d2 values based on the parental genotypes. They then showed that offspring in 12 out of 19 broods had a higher genetic diversity (mean d2) than expected when fertilization would be random with respect to sperm haplotype. Further evidence for postcopulatory processes favoring sperm from less‐ related males comes from a study on red junglefowl (Gallus gallus; Pizzari et al., 2004). Pizarri and colleagues counted the number of sperm that reached the outer perivitelline membrane on eggs laid between 1 and 10 days after successful insemination by either a brother or an unrelated male. They found much lower sperm numbers on the eggs when females had been inseminated with a brother (Fig. 8). This is an impressive result, which clearly suggests that females can select against sperm from a related male.

MATE CHOICE AND GENETIC QUALITY

A

249

Proportion of eggs surviving to adults

1.0

0.9

0.8

0.7 0.4 Proportion sired by male 1

B

0.3

0.2

0.1 0 1/8 1/4 1/2 Female relatedness to male 1 FIG. 7. Results of double mating experiments on Drosophila melanogaster. Virgin females were mated first to a male of an experimentally determined relatedness (0 ¼ unrelated, 1/8 ¼ cousin, 1/4 ¼ half‐sibling, 1/2 ¼ full sibling), and then to an unrelated male. (A) The proportion of eggs that survived to the adult stage depended on mating treatment; effect size r ¼ .33, N ¼ 112, p < .0001. (B) The proportion of offspring sired by the first male depended on mating treatment; effect size r ¼ .19, N ¼ 226, p ¼ .0057. Shown are mean and standard errors. Adapted from Table III and Fig. 2 in Mack et al. (2002) by permission from the authors and Blackwell Publishing.

5. There is further evidence from studies on mice that fertilization is not random with respect to Mhc haplotypes. Wedekind et al. (1996) performed in vitro fertilizations with two inbred mice strains that were only heterozygous at an Mhc locus. They found that selection took place at two levels. First, there was evidence that eggs selected sperm based on Mhc haplotype. Second, Wedekind et al. (1996) found evidence that the second meiotic division of an egg was influenced by the Mhc type of sperm that entered it. This resulted in more Mhc homozygous offspring in an infection‐free situation, and more Mhc heterozygous offspring during an (unplanned) period of viral infection. These results were corroborated by a further experiment

250

BART KEMPENAERS

PVL-sperm number

160 Brother Unrelated

80

0 0

5 Day after insemination

10

FIG. 8. Sperm numbers (hydrolysis points) on the outer perivitelline layer of eggs laid over a period of 10 days after a female red jungle fowl (G. gallus) was inseminated by an unrelated male or by a brother. The effect of relatedness is highly significant; General Linear Mixed Model: F1,303 ¼ 43.04, N ¼ 24 females, p < .0001. Shown are mean  S.E.M. Adapted from Fig. 1 in Pizzari et al. (2004) by permission from the authors and the Royal Society Publishing.

where infected and sham‐infected mice were compared (Ru¨licke et al., 1998). Note that an excess of Mhc heterozygotes in offspring is often reported, but can also be due to differential mortality (e.g., Dorak et al., 2002). That the double mating experimental design is a very useful tool is also demonstrated by an interesting study on viviparous pseudoscorpions (Zeh and Zeh, 2006). As in the cricket studies discussed above, females were mated with two males, either a sibling (S) or an unrelated male (N), in the combinations SS, NS, SN, NN, and N (two copulations with the same male), and the number of nymphs that were born were counted. The result was clear‐cut: as expected under the hypothesis that females can avoid inbreeding depression by mating with multiple males and selecting the most compatible sperm, SS females had a significantly lower reproductive success compared to all other females (whereby SN, NS, N had similar success, but lower than NN). However, paternity analyses revealed that in the SN and NS treatments, the majority of offspring were sired by the sibling male (on average 69%, after correcting for inbreeding‐related mortality). Zeh and Zeh (2006) argued that the reason why inbreeding depression is reduced in the NS and SN treatment is that in a viviparous species the noninbred embryos can ‘‘rescue’’ their inbred half‐sibs, perhaps by ensuring that sufficient nutrients enter the communal brood sac, or by buffering a maternal immune response against inbred embryos. Although the

MATE CHOICE AND GENETIC QUALITY

251

mechanism remains unclear, this is an interesting hypothesis, and one that gives a different twist to the idea of interactive environmental effects on inbreeding depression (discussed above). g. Experimental mate choice studies In a now classical experiment, Bateson (1982) tested 22 male and 13 female Japanese quail (Coturnix japonica) in a choice chamber and let them choose between a familiar full sibling, a novel full sib (never seen before), a novel first cousin, a novel third cousin, and an unrelated bird, all of the opposite sex. The results showed that both males and females preferred to spend time with birds of intermediate relatedness (first cousins, Fig. 9). Bateson (1982) compares the choice of the quail with human mate choice, and suggests that the preference for an individual of intermediate genetic similarity leads to optimal outbreeding. Several subsequent studies confirmed that females (and males) prefer individuals of intermediate relatedness (e.g., in mice, Barnard and Fitzsimons, 1988; Keane, 1990). The latter study and Barnard and Fitzsimons (1989) also investigated the fitness consequences, and found evidence for inbreeding depression in terms of reduced litter size at weaning. However, Barnard and Fitzsimons (1988) finding that their experimental females preferred full or half‐sib males was hard to match with the observed inbreeding depression. Similar experiments have been carried out with crickets, and Simmons (1991) also reported a preference for cousins, although he referred to it as ‘‘a weak tendency,’’ and suggested that ‘‘female choice is simply a mechanism of inbreeding avoidance.’’ Two recent choice studies have used microsatellite markers to estimate the relatedness between the female and potential partners. In an experiment with tu´ngara frogs (Physalaemus pustulosus), Lampert et al. (2006) let females choose between the calls of two males, and tested whether relatedness determined the outcome. It did not, which was not surprising, because Lampert et al. (2006) also showed that male call similarity did not correlate with genetic similarity. The other experiment was conducted with the agile antechinus (Antechinus agilis). In this Australian marsupial, the only resource females obtain from mating is sperm, and in natural populations, larger males sire more offspring. Hence, Parrott et al. (2006) let 20 females choose between a large and a small male. Instead of finding a preference for size, however, they showed that the preferred male was less related to the female than the nonpreferred male (based on 15 females that made a choice, p ¼ .002). Note that here the choice is based on much subtler differences in relatedness than in the experiments described above. The most impressive example of the sophistication of mate choice comes from studies on the three‐spined stickleback. Work on the Mhc showed that an intermediate number of class‐IIB alleles is optimal (as discussed in the

252

BART KEMPENAERS

Females Males

40

% time

30

20

10

0 Sib Familiar

Sib

First Third Unrelated Cousin Novel

FIG. 9. Classical choice experiment with Japanese quail (C. japonica). Shown is the percentage of time males (N ¼ 22) and females (N ¼ 13) spent near members of the opposite sex that differed in relatedness and familiarity with the choosing individual. Each individual could see five birds (one from each category) simultaneously. Adapted from Fig. 2 in Bateson (1982) by permission from Macmillan Publishers Ltd.: Nature, copyright 1982.

previous section). Reusch et al. (2001) tested two alternative mate choice strategies. (a) If females attempt to avoid inbreeding depression, they should choose males with dissimilar Mhc genotypes. (b) If females aim to optimize their offspring’s resistance to parasites, they should prefer males with a higher diversity of Mhc alleles. In elegant experiments, they showed that females indeed preferred the odor of males with a large number of Mhc alleles, and that females with more alleles are less choosy than those with fewer alleles. In contrast, there was no choice for Mhc dissimilarity (Reusch et al., 2001). Such choice was found in a follow‐up experiment, where a small effective population size was simulated (Aeschlimann et al., 2003). Under those circumstances, inbreeding depression might have been more important, and females seemed to respond accordingly. However, females (which had few alleles themselves) would also have to choose Mhc dissimilar males in order to optimize the number of alleles for disease resistance. Thus, the study by Aeschlimann et al. (2003) does not provide unequivocal evidence that female sticklebacks avoid inbreeding. Such evidence comes from another study (Frommen and Bakker, 2006). Here, females could choose between a familiar brother and an unfamiliar unrelated male, when both males were similar in body size and in the intensity of the red coloration. Twelve out of 16 females spent more time with the unrelated male (p ¼ .046). Previous work suggested a similar result in guppies, where females

MATE CHOICE AND GENETIC QUALITY

253

preferred to court and mate with unfamiliar (potentially unrelated) males (Hughes et al., 1999; Kelley et al., 1999). However, when female guppies were allowed to choose between a brother and an unrelated male, based on olfactory cues alone, or in combination with visual cues, no preference for the unrelated male could be detected (N ¼ 40, all p > .5; Viken et al., 2006). Mate choice experiments in relation to Mhc similarity have been popular in other species, particularly mice and humans (reviewed in Eklund, 1999; Piertney and Oliver, 2006). In humans, odor‐based choice for Mhc‐dissimilar individuals was first reported by Wedekind et al. (1995), and has been confirmed in subsequent work (e.g., Wedekind and Fu¨ri, 1997, no gender effect; Thornhill et al., 2003, only for males choosing females; Santos et al., 2005, only for females choosing males). Wedekind and Fu¨ri (1997) experiments also suggested that the observed preferences did not aim for particular Mhc combinations that might be beneficial under particular environmental conditions (e.g., presence of pathogens), but would increase offspring heterozygosity. In contrast, experiments on visual‐based choice showed that faces were rated more attractive if the female rater was Mhc‐similar (Roberts et al., 2005a,b). Males are expected to be less choosy, also with respect to relatedness, but there is experimental evidence that they do respond to the relatedness of potential mates. For example, in behavioral trials with red junglefowl, males waited longer before they copulated with a sister than with an unrelated female (Pizzari et al., 2004).

B. MATE CHOICE FOR HETEROZYGOUS INDIVIDUALS 1. Theoretical Considerations a. Should females prefer heterozygous males? If heterozygosity correlates with individual quality (see previous section), there are two reasons why females might benefit from choosing heterozygous males. First, through direct benefits, for example, because they will obtain better paternal care, or because they can reduce the risk of male infertility. For example, Marr et al. (2006) showed that fledglings of highly inbred fathers survived less well. It seems worthwhile to investigate whether homozygous individuals provide less care, or lower quality care, perhaps particularly under stressful environmental conditions. Second, although heterozygosity is not heritable, there are conditions under which females produce more heterozygous offspring by mating with heterozygous males (Hoffman et al., 2007; Mitton et al., 1993). How does this work? A highly heterozygous male will have parents that are unusually dissimilar. This means that one or both of the males’ parents will be dissimilar to an average local individual. Therefore, when a local female mates with a heterozygous male, the

254

BART KEMPENAERS

dissimilarity will be passed on as increased heterozygosity in her offspring, whereby the strength of the effect will depend on the population structure (Joe Hoffman, personal communication). Given a single overdominant locus where the two homozygotes differ in fitness, choice for the heterozygote would lead to a higher frequency of the least fit genotype compared to random choice (Irwin and Taylor, 2000). Thus, Lehmann et al. (2006) concluded that female choice for heterozygous males can only evolve under rather special (and perhaps unrealistic) conditions. However, when Irwin and Taylor (2000) included changing environmental conditions in their model so that the least fit of the two homozygotes can become the fittest, choice for heterozygous males can evolve. Their conclusion thus sounds very different: choice for heterozygous males will be favored under a wide range of biologically plausible conditions. Reinhold (2002) also used population genetic modeling to compare the evolution of three choice strategies under the presence of recessive deleterious alleles. The first strategy assumed that females choose dissimilar males, that is, males with the lowest number of deleterious alleles in common with the female. The second strategy assumed that females mated with males that had the lowest number of deleterious alleles. Under the third strategy, females would choose the male with the lowest number of loci that are homozygous for the deleterious alleles (choice for heterozygous males). The results suggested that the advantages of choice are higher for the first than for the second strategy, and negligible for the third. Note that the existing empirical results are hard to match with the theoretical predictions. For example, Lehmann et al. (2006) suggested that if heterozygous males are more ornamented (for which there is empirical evidence, see above), it is hard to understand why female choice for ornamentation would persist (see also Edvardsson and Arnqvist, 2006). Recently, Roberts et al. (2006) reported on relationships between estimates of individual heterozygosity and estimates of genetic similarity between individuals (mean number of alleles shared and mean relatedness), using two datasets. In humans, individual heterozygosity at three Mhc loci correlated positively with the mean number of alleles shared, and negatively with mean relatedness (calculated at these Mhc loci). In peafowl (Pavo cristatus), typed at 13 microsatellite loci, heterozygosity was negatively correlated with both mean alleles shared and with mean relatedness, but only for loci with low polymorphism (4 loci with mean expected heterozygosity of .15). Roberts et al. (2006) admit that these correlations are at least in part due to the way in which relatedness is calculated. However, they also suggest that if these are biologically meaningful relationships (which

MATE CHOICE AND GENETIC QUALITY

255

remains to be shown), there are consequences for female choice. For example, if heterozygous males are more ornamented, then females might benefit from choosing ornamented males, simply because this lowers the genetic similarity with their partner. Roberts et al. (2006) also make the interesting suggestion that heterozygous females should be more selective with regard to genetic similarity, and that female heterozygosity should be positively linked with the incidence of extra‐pair mating. There is some evidence in support of the latter (see below). However, an alternative explanation is that individual heterozygosity reflects quality, and that high‐ quality females are more likely to engage in extra‐pair mating (Gowaty, 1996). This clearly needs further investigation. Reid et al. (2006) showed a (unexpected) positive correlation between the inbreeding coefficient of parents and the inbreeding coefficient of their offspring (r ¼ .34, N ¼ 315, p < .001), again suggesting that parents would do better to avoid less heterozygous individuals. However, whether this is more generally true remains to be seen. The observed correlation can result from random mating in an island population with mostly inbred residents and some outbred immigrants (Reid et al., 2006), and might thus be particular to this type of population. b. Should males prefer heterozygous females? A female’s heterozygosity can influence her condition, fecundity, and survival probability (see previous section). For example, Reid et al. (2003) suggested that inbred females were worse parents, and Richardson et al. (2004) showed that low heterozygosity of mothers had a negative impact on offspring survival, probably because of maternal effects (nestlings were cross‐fostered). These observations lead to the prediction that males should prefer heterozygous females as social partners. If heterozygosity reflects quality in both sexes, one could expect assortative mating for heterozygosity. A few studies have tested this (see below). c. Should heterozygous females choose differently? There are several reasons why one can expect heterozygous females to differ in their preference or in their choosiness. (1) If less heterozygous females are generally of lower quality, they might invest less in costly mate search, and they might be less successful in outcompeting other females for the best breeding opportunities. In that sense, they should be less choosy. (2) Alternatively, one could argue that inbred females should be more choosy, for two reasons. First, inbred females need to compensate for their lower quality, and might be prepared to pay the costs of increased choosiness for higher marginal benefits (Mazzi et al., 2004). Second, if a female is homozygous at key loci (with negative fitness consequences), it is even more important to avoid a

256

BART KEMPENAERS

similar homozygous male, because all the offspring of such a mating would be homozygous as well. (3) In the context of the Mhc, if there is an optimum number of alleles at multiple loci, heterozygous females will already carry many alleles, and they might thus prefer homozygous males carrying fewer alleles (see discussion above; Aeschlimann et al., 2003).

2. Review of the Evidence a. Choice of social mate or sire In total, three out of eight studies that tested it found evidence that females prefer heterozygous males as social mate or as sire (Table VIIa). In the house sparrow, mated males (but not females) showed a higher allelic diversity at Mhc loci, compared to unmated males, but no difference in heterozygosity at seven microsatellite loci was found (Bonneaud et al., 2006a). Similarly, in a seminatural population of rhesus macaques, heterozygosity at an Mhc class II locus was the best predictor of male success at siring offspring (Sauermann et al., 2001; Widdig et al., 2004). No effects of particular alleles were found (Sauermann et al., 2001). Finally, in Atlantic salmon, sires (but not breeding females) were more heterozygous than the population mean (Garant et al., 2005). In these and other studies that show that heterozygous or less outbred males or females had a higher reproductive success (Table II), it is unclear whether the effects are due to intra‐ or intersexual selection, or both (e.g., Marr et al., 2006). For example, in the black grouse, heterozygous males had a higher copulation success, but this may also be due to male–male competition for the best territories (Ho¨glund et al., 2002). However, a recent study on Antarctic fur seals (Arctocephalus gazella) presents strong evidence that females actively choose males with respect to heterozygosity and relatedness (Hoffman et al., 2007). In this species, males are extremely sedentary during the breeding season, and females move across the colony to copulate with a particular male. Hoffman et al. (2007) measured the distance between the females’ pupping location and the location of the sire, and showed that females that moved further conceived to more heterozygous males, and to males to which they are less related. They also show that this leads to the production of offspring with increased heterozygosity, because male (but not female) heterozygosity predicted offspring heterozygosity (the latter indirectly estimated as parental relatedness; r ¼ .17, p ¼ .003, N ¼ 310). Only four studies tested whether there was assortative mating for heterozygosity or for Mhc allelic diversity, and the latter was shown in only one of the studies (Bonneaud et al., 2006a; Table VIIa).

TABLE VII EVIDENCE FOR CHOICE FAVORING HETEROZYGOUS MATES Markera

Species Great reed warbler House sparrow

257

Reed bunting Seychelles warbler

Number of locib

Mhc Micro Mhc Mhc Mhc Micro Micro

Estimatec N

7

9 14

Mhc

Sample size

Effectd

p

(a) Choice of social mate or sire 279 No >.14

H N N N SH H

30 30 30 37 68 97

No Pos No Pos No No

.84 .01 .38 .024 .77 .73

N

82

No

.82 .23 .035 .83

Atlantic salmon

Mhc Micro Micro

5 5

N IR IR

82 31 31

No Pos No

Tu´ngara frog Rhesus macaque

Micro Mhc

7 1

H H

? 114

No Pos

>.5? <.05

Mhc

1

H

283

Pos

.06

Hypothesis tested/comments Social male versus random male (or average male) Mated versus unmated male Mated versus unmated males Mated versus unmated females Assortative mating Assortative mating Assortative mating Social male versus population mean Assortative mating Sire versus population mean Breeding female versus population mean Mated versus unmated males Males siring offspring (yes/no) Number of offspring sired by male

References 1 2 2 2 2 3 4 5 5 6 6 7 8 9

(b) Prediction: male or female heterozygosity predicts the presence of extra‐pair young (yes/no, unless otherwise stated) Black‐throated blue warbler

Micro

5

d2

92

Neg

.03

Male

10 (Continued)

TABLE VII (Continued) Species Blue tit

Great tit House sparrow

258 Reed bunting

Savannah sparrow Seychelles warbler

Number of locib

Estimatec

Sample size

Effectd

p

Micro Micro

5–7 5–7

SH SH

117 117

No No

.63 .44

Micro Micro Micro Micro

5–7 5 5 10

SH d2 d2 H

95 23 23 57

Pos Pos No No

<.05 .02 >.05 .59

Micro

10

d2

57

No

.11

Micro

10

H

34

No

.82

Micro

10

d2

34

No

.60

Micro Micro

9 9

SH SH

53 53

No No

.20 .54

Micro Micro

9 9

SH SH

62 59

Pos No

.07 .39

Micro

9

SH

68

No

.48

Micro

9

SH

68

No

.17

Micro

4–6

H

116

No

>.20

Mhc

N

82

Pos

.02

Male

17

Mhc

N

82

No

.56

Female

17

Markera

Hypothesis tested/comments Male Male gains extra‐pair paternity (yes/no) Female Male Female Proportion of young male sired in a brood Proportion of young male sired in a brood Female; proportion of extra‐ pair young in a brood Female; proportion of extra‐pair young in a brood Male Proportion of young male sired in a brood Female Female; proportion of extra‐pair young in a brood Male gains extra‐pair young (yes/no) Number of extra‐pair young sired by male Male

References 11 11 12 13 13 14 14 14 14 15 15 15 15 15 15 16

Alpine marmot Common mole‐rat

Micro Micro

5–12 7

SH SH

68 31

No No

.21 .65

Black‐throated blue warbler Blue tit House wren

Micro

5

d2

92

No

.38

Micro Micro

5–7 5–7

SH SH

117 20

No No

.11 >.20

Micro Micro

9 9

SH SH

20 20

No No

.57 .44

Micro

4–6

H

56

Neg

.077

N

29

Pos

.05

Male Male

18 19

(c) Prediction: extra‐pair male is more heterozygous than within‐pair male

Reed bunting

259

Savannah sparrow Seychelles warbler Splendid fairy‐wren Alpine marmot Common mole‐rat a

Mhc Micro

6

SH

48

Pos

.047

Micro Micro

5–12 7

SH SH

9 17

Pos No

.01 .83

10 Similar result for mean d2 ( p > .15) Extra‐pair sire versus nearest neighbor Significant in 1 year ( p ¼ .05)

11 20 15 15 16 17

Same effect with IR

Extra‐colony male versus social male

21 18 19

Marker used to measure relatedness: Mhc, microsatellite, multilocus minisatellite. If blank: multilocus, unknown number. c H ¼ heterozygosity, IR ¼ internal relatedness, N ¼ number of bands as estimate of heterozygosity (see also Table I), SH ¼ standardized heterozygosity. d Pos: expected effect ( p < .10); Neg: opposite effect ( p < .10); No: no effect ( p > .10 or unknown). Bold when significant. 1. Westerdahl, 2004; 2. Bonneaud et al., 2006a; 3. Kleven and Lifjeld, 2005; 4. Richardson et al., 2004; 5. Richardson et al., 2005; 6. Garant et al., 2005; 7. Lampert et al., 2006; 8. Widdig et al., 2004; 9. Sauermann et al., 2001; 10. Smith et al., 2005; 11. Foerster et al., 2003; 12. Kempenaers, unpublished data; 13. Otter et al., 2001; 14. Stewart et al., 2006; 15. Kleven and Lifjeld, 2005; 16. Freeman‐Gallant et al., 2006; 17. Richardson et al., 2005; 18. Cohas et al., 2006; 19. Bishop et al., 2007; 20. Masters et al., 2003; 21. Tarvin et al., 2005. b

260

BART KEMPENAERS

b. Extra‐pair mate choice Whether the probability that a brood contains extra‐pair young depends on the heterozygosity of the social male has been tested in nine species (Table VIIb). In the Seychelles warbler, males with a higher allelic diversity at Mhc loci were less likely to lose paternity (Richardson et al., 2005), and in the great tit more genetically diverse males (estimated as mean d2) were less likely to have extra‐pair young in their brood (Otter et al., 2001). In the black‐throated blue warbler (Dendroica caerulescens), the exact opposite was found, that is, males with higher d2 values were more likely to lose paternity (Smith et al., 2005). Interestingly, two studies (out of five that tested it) showed that females with extra‐pair young were on average more heterozygous than females without extra‐pair young, although the effect was small and nonsignificant in one study (p ¼ .07). In three out of nine studies that tested it, the male that sired the extra‐ pair young was significantly more heterozygous (two studies) or had a higher Mhc allelic diversity than the female’s social partner (Table VIIc). c. Choice by inbred females This has rarely been studied. However, in an experimental study on three‐spined sticklebacks, inbred females showed a stronger preference for symmetrical male models, than outbred females (Mazzi et al., 2004). In contrast, Frommen and Bakker (2006) did not find a difference in female choice patterns between inbred and outbred females: both preferred unrelated males rather than brothers in a choice experiment.

V. CONCLUSIONS AND OUTLOOK In this review, I addressed the hypothesis that mate choice is linked to genetic compatibility. Synthesizing the evidence reviewed above, it is clear that mate choice processes can lead to increased offspring heterozygosity, and that this can be beneficial. However, the current data are still insufficient to make sweeping statements about the generality of the heterozygosity hypothesis. It is noteworthy that many studies have tested multiple predictions, but found evidence for only one (if at all), and often for only a subset of the data. It should also be noted that the effects are usually small, even if the potential fitness benefits seem substantial. On the other hand, the tools used (a few microsatellite markers) are often crude. The increasing possibilities of typing individuals at many more loci at low cost, might help to obtain better estimates of relatedness. I also expect that more information will become available on specific genes that are relevant for mate choice, such as Mhc genes.

MATE CHOICE AND GENETIC QUALITY

261

The overall evidence that precopulatory mate choice is related to genetic similarity appears rather weak, but the evidence that multiple mating or extra‐pair copulations lead to increased offspring heterozygosity is more compelling. Data such as that on the red junglefowl (Fig. 8) strongly suggest that postcopulatory sperm selection is possible, but the relevance in situations of sperm competition remains to be shown (Denk et al., 2005). Looking ahead, I expect that progress will first come in our understanding of the mechanisms behind heterozygosity–fitness correlations. This is a prerequisite for making better predictions about optimal mate choice. An exciting development is that we can start using information from molecular genetics (or genomics) to learn more about the reasons behind the additive and nonadditive genetic effects on fitness. For example, markers are becoming available that allow researchers to investigate the heterozygosity at particular genes with known functions, which might soon lead to a better understanding of the reasons for overdominance (e.g., Gemmell and Slate, 2006). Two of the most pressing questions are whether individual heterozygosity indeed reflects individual quality, and whether female choice for ornamented males equals choice for heterozygous males rather than choice for good alleles. The evidence that females choose heterozygous males is about as strong as the evidence that they choose males that are genetically distinct from themselves, and both can lead to increased offspring heterozygosity (see Hoffman et al., 2007). With respect to the Mhc there is evidence that females choose males to maximize or optimize offspring heterozygosity, but relatively little evidence that offspring heterozygosity per se is beneficial. There is also evidence that allelic divergence is important, that is, the degree of dissimilarity in functional Mhc proteins (e.g., Sommer, 2005). Finally, several studies suggest that specific Mhc alleles protect against specific (common) diseases or parasites (additive effects). However, the described type of benefit and the observed patterns of choice rarely seem to match. Studies on mate choice and individual genetic diversity will continue to provide answers to more fundamental questions. Traits closely related to fitness seem to have low heritabilities. However, this does not mean that nonadditive genetic effects cannot have a strong effect on fitness (e.g., Merila¨ et al., 2003). Is balancing selection the key process that maintains genetic polymorphisms in populations? What is the role of temporal or spatial variation in selection, or of frequency‐dependent selection? Humans are rapidly changing the environment, and populations of many species are becoming smaller due to habitat loss or fragmentation. Thus, we need to better understand the consequences of decreasing genetic variability (e.g., Spielman et al., 2004).

262

BART KEMPENAERS

The idea that females obtain genetic benefits from choosing particular males as sires for their offspring remains controversial. However, I assert that the key question is not whether, but when, which and how.  When are direct and indirect benefits more or less important?  Which genetic benefits are more important, additive or nonadditive

effects, in which circumstances?

populations,

and

under

which

ecological

 How are females (and males) optimizing their choice for direct, addi-

tive, and nonadditive benefits? If we want to better understand the adaptive significance of mate choice, and its consequences, we will have to address these questions. This is particularly important because there might be conflicts between different types of benefits (e.g., Oneal et al., 2006), and, it is unlikely that investigating only one type of benefit will generally be sufficient to understand the selective forces that shape mating strategies (Ivy, 2007; Wedekind et al., 2001). Tackling these questions is going to be a major challenge.

Acknowledgments I sincerely thank Axel Krikelis and Renate Alton for library assistance, Heike Gorny‐ Leimpeters and Anke Hundrisser for secretarial assistance, Theo Weber for redrawing the figures, and the members of my group, particularly James Dale, Emily DuVal, Wolfgang Forstmeier, Jakob Mueller, and Mihai Valcu for stimulating discussions over breakfast, coffee, lunch, tea and dinner, and for insightful comments on an earlier version of the manuscript. I am grateful to Tom Van de Casteele, who kindly provided the raw data on which Fig. 4 is based, to Marc Naguib and an anonymous reviewer for constructive comments on the manuscript, to Joe Hoffman and Bryan Neff for answering my questions, and to Tom Pizzari and Leigh Simmons for comments on a figure. The generous support of the Max Planck Society, the patience and support of Christina Muck, and the help of the current administrative director Manfred Gahr with institute matters, gave me the freedom and time to complete this review.

References Acevedo‐Whitehouse, K., Gulland, F., Greig, D., and Amos, W. (2003). Disease susceptibility in California sea lions. Nature 422, 35. Aeschlimann, P. B., Ha¨berli, M. A., Reusch, T. B. H., Boehm, T., and Milinski, M. (2003). Female sticklebacks Gasterosteus aculeatus use self‐reference to optimize MHC allele number during mate selection. Behav. Ecol. Sociobiol. 54, 119–126. Amos, W., Wilmer, J. W., Fullard, K., Burg, T. M., Croxall, J. P., Bloch, D., and Coulson, T. (2001). The influence of parental relatedness on reproductive success. Proc. Roy. Soc. Lond. B 268, 2021–2027. Andersson, M. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton.

MATE CHOICE AND GENETIC QUALITY

263

Andersson, M., and Simmons, L. W. (2006). Sexual selection and mate choice. Trends Ecol. Evol. 21, 296–302. Apanius, V., Penn, D., Slev, P. R., Ruff, L. R., and Potts, W. K. (1997). The nature of selection on the major histocompatibility complex. Crit. Rev. Immunol. 17, 179–224. Aparicio, J. M., Cordero, P. J., and Veiga, J. P. (2001). A test of the hypothesis of mate choice based on heterozygosity in the spotless starling. Anim. Behav. 62, 1001–1006. Aparicio, J. M., Ortego, J., and Cordero, P. J. (2006). What should we weigh to estimate heterozygosity, alleles or loci? Mol. Ecol. 15, 4659–4665. Arkush, K. D., Giese, A. R., Mendonca, H. L., McBride, A. M., Marty, G. D., and Hedrick, P. W. (2002). Resistance to three pathogens in the endangered winter‐run Chinook salmon (Oncorhynchus tshawytscha): Effects of inbreeding and major histocompatibility complex genotypes. Can. J. Fish. Aquat. Sci. 59, 966–975. Armbruster, P., and Reed, D. H. (2005). Inbreeding depression in benign and stressful environments. Heredity 95, 235–242. Armbruster, P., Hutchinson, R. A., and Linvell, T. (2000). Equivalent inbreeding depression under laboratory and field conditions in a tree‐hole‐breeding mosquito. Proc. Roy. Soc. Lond. B 267, 1939–1945. Arnqvist, G., and Kirkpatrick, M. (2005). The evolution of infidelity in socially monogamous passerines: The strength of direct and indirect selection on extrapair copulation behavior in females. Am. Nat. 165, S26–S37. Aspi, J. (2000). Inbreeding and outbreeding depression in male courtship song characters in. Drosophila montana. Heredity 84, 273–282. Audo, M. C., and Diehl, W. J. (1995). Effect of quantity of enviromental‐stress on multilocus heterozygosity growth relationships in Eisenia‐Fetida (Annelida, Oligochaeta). Heredity 75, 98–105. Avile´s, L., and Bukowski, T. C. (2006). Group living and inbreeding depression in a subsocial spider. Proc. Roy. Soc. Lond. B 273, 157–163. Balloux, F., Amos, W., and Coulson, T. (2004). Does heterozygosity estimate inbreeding in real populations? Mol. Ecol. 13, 3021–3031. Barber, C. A., Edwards, M. J., and Robertson, R. J. (2005). A test of the genetic compatibility hypothesis with tree swallows, Tachycineta bicolor. Can. J. Zool. 83, 955–961. Barnard, C. J., and Fitzsimons, J. (1988). Kin recognition and mate choice in mice: The effects of kinship, familiarity and social interference on intersexual interaction. Anim. Behav. 36, 1078–1090. Barnard, C. J., and Fitzsimons, J. (1989). Kin recognition and mate choice in mice: Fitness consequences of mating with kin. Anim. Behav. 38, 35–40. Bateson, P. (1982). Preferences for cousins in Japanese quail. Nature 295, 236–237. Bateson, P. (1983). Optimal outbreeding. In ‘‘Mate Choice’’ (P. Bateson, Ed.), pp. 257–278. Cambridge University Press, Cambridge. Bean, K., Amos, W., Pomeroy, P. P., Twiss, S. D., Coulson, T. N., and Boyd, I. L. (2004). Patterns of parental relatedness and pup survival in the grey seal (Halichoerus grypus). Mol. Ecol. 13, 2365–2370. Bensch, S., Hasselquist, D., and von Schantz, T. (1994). Genetic similarity between parents predicts hatching failure: Nonincestuous inbreeding in the great reed warbler? Evolution 48, 317–326. Bernasconi, G., Ashman, T.‐L., Birkhead, T. R., Bishop, J. D. D., Grossniklaus, U., Kubli, E., Marshall, B., Schmid, B., Skogsmyr, I., Snook, R. R., Taylor, D., Till‐Bottraud, I., et al. (2004). Evolutionary ecology of the prezygotic stage. Science 303, 971–975.

264

BART KEMPENAERS

Bierne, N., Beuzart, I., Vonau, V., Bonhomme, F., and Be´dier, E., and AQUACOP (2000a). Microsatellite‐associated heterosis in hatchery‐propagated stocks of the shrimp Penaeus stylirostris. Aquaculture 184, 203–219. Bierne, N., Tsitrone, A., and David, P. (2000b). An inbreeding model of associative overdominance during a population bottleneck. Genetics 155, 1981–1990. Birkhead, T. R. (1998). Cryptic female choice: Criteria for establishing female sperm choice. Evolution 52, 1212–1218. Birkhead, T. R. (2000). Defining and demonstrating postcopulatory female choice—again. Evolution 54, 1057–1060. Birkhead, T. R., and Pizzari, T. (2002). Postcopulatory sexual selection. Nat. Rev. Genet. 3, 262–273. Birkhead, T. R., Chaline, N., Biggins, J. D., Burke, T., and Pizzari, T. (2004). Nontransitivity of paternity in a bird. Evolution 58, 416–420. Bishop, J. D. D., Jones, C. S., and Noble, L. R. (1996). Female control of paternity in the internally fertilizing compound ascidian Diplosoma listerianum. II. Infestigation of male mating success using RAPD markers. Proc. Roy. Soc. Lond. B 263, 401–407. Bishop, J. M., O’Ryan, C., and Jarvis, J. U. M. (2007). Social common mole‐rats enhance outbreeding via extra‐pair mating. Biol. Lett. 3, 176–179. Blais, J., Rico, C., and Bernatchez, L. (2004). Nonlinear effects of female mate choice in wild threespine sticklebacks. Evolution 58, 2498–2510. Blomqvist, D., Andersson, M., Ku¨pper, C., Cuthill, I. C., Kis, J., Lanctot, R. B., Sandercock, B. K., Sze´kely, T., Wallander, J., and Kempenaers, B. (2002). Genetic similarity between mates and extra‐pair parentage in three species of shorebirds. Nature 419, 613–615. Bluhm, C. K., and Gowaty, P. A. (2004). Social constraints on female mate preferences in mallards, Anas platyrhynchos, decrease offspring viability and mother productivity. Anim. Behav. 68, 977–983. Bonneaud, C., Chastel, O., Federici, P., Westerdahl, H., and Sorci, G. (2006a). Complex Mhc‐based mate choice in a wild passerine. Proc. Roy. Soc. Lond. B 273, 1111–1116. Bonneaud, C., Pe´rez‐Tris, J., Federici, P., Chastel, O., and Sorci, G. (2006b). Major histocompatibility alleles associated with local resistance to malaria in a passerine. Evolution 60, 383–389. Bretman, A., Wedell, N., and Tregenza, T. (2004). Molecular evidence of post‐copulatory inbreeding avoidance in the field cricket Gryllus bimaculatus. Proc. Roy. Soc. Lond. B 271, 159–164. Briskie, J. V. (1996). Spatiotemporal patterns of sperm storage and last‐male sperm precedence in birds. Funct. Ecol. 10, 375–383. Britten, H. B. (1996). Meta‐analyses of the association between multilocus heterozygosity and fitness. Evolution 50, 2158–2164. Brooker, M. G., Rowley, I., Adams, M., and Baverstock, P. R. (1990). Promiscuity: An inbreeding avoidance mechanism in a socially monogamous species? Behav. Ecol. Sociobiol. 26, 191–199. Brown, J. L. (1997). A theory of mate choice based on heterozygosity. Behav. Ecol. 8, 60–65. Brown, J. L. (1999). The new heterozygosity theory of mate choice and the MHC. Genetica 104, 215–221. Buchholz, R. (2004). Effects of parasitic infection on mate sampling by female wild turkeys (Meleagris gallopavo): Should infected females be more or less choosy? Behav. Ecol. 15, 687–694. Bull, C. M., and Cooper, S. J. B. (1999). Relatedness and avoidance of inbreeding in the lizard, Tiliqua rugosa. Behav. Ecol. Sociobiol. 46, 367–372.

MATE CHOICE AND GENETIC QUALITY

265

Bussie`re, L. F., Hunt, J., Jennions, M. D., and Brooks, R. (2006). Sexual conflict and cryptic female choice in the black field cricket, Teleogryllus commodus. Evolution 60, 792–800. Byers, J. A., and Waits, L. (2006). Good genes sexual selection in nature. Proc. Natl. Acad. Sci. USA 103, 16343–16345. Calleri, D. V. II, McGrail Reid, E., Rosengaus, R. B., Vargo, E. L., and Traniello, J. F. A. (2006). Inbreeding and disease resistance in a social insect: Effects of heterozygosity on immunocompetence in the termite Zootermopsis angusticollis. Proc. Roy. Soc. Lond. B 273, 2633–2640. Carchini, G., Chiarotti, F., Di Domenico, M., Mattoccia, M., and Paganotti, G. (2001). Fluctuating asymmetry, mating success, body size and heterozygosity in Coenagrion scitulum (Rambur) (Odonata: Coenagrionidae). Anim. Behav. 61, 661–669. Cassinello, J., Gomendio, M., and Roldan, E. R. S. (2001). Relationship between coefficient of inbreeding and parasite burden in endangered gazelles. Conserv. Biol. 15, 1171–1174. Carrington, M., Nelson, G. W., Martin, M. P., Kissner, T., Vlahov, D., Goedert, J. J., Kaslow, S., Buchbinder, S., Hoots, K., and O’Brien, J. (1999). HLA and HIV‐1: Heterozygote advantage and B*35‐Cw*04 disadvantage. Science 283, 1748–1752. Charlesworth, B., and Charlesworth, D. (1999). The genetic basis of inbreeding depression. Genet. Res. Camb. 74, 329–340. Charmantier, A., and Sheldon, B. C. (2006). Testing genetic models of mate choice evolution in the wild. Trends Ecol. Evol. 21, 417–419. Charmantier, A., Blondel, J., Perret, P., and Lambrechts, M. M. (2004). Do extra‐pair paternities provide genetic benefits for female blue tits Parus caeruleus? J. Avian Biol. 35, 524–532. Chistiakow, D. A., Hellemans, B., and Volckaert, F. A. M. (2006). Microsatellites and their genomic distribution, evolution, function and applications: A review with special reference to fish genetics. Aquaculture 255, 1–29. Cohas, A., Yoccoz, N. G., Da Silva, A., Goossens, B., and Allaine´, D. (2006). Extra‐pair paternity in the monogamous alpine marmot (Marmota marmota): The roles of social setting and female mate choice. Behav. Ecol. Sociobiol. 59, 597–605. Colegrave, N., Kotiaho, J. S., and Tomkins, J. L. (2002). Mate choice or polyandry: Reconciling genetic compatibility and good genes sexual selection. Evol. Ecol. Res. 4, 911–917. Coltman, D. W., and Slate, J. (2003). Microsatellite measures of inbreeding: A meta‐analysis. Evolution 57, 971–983. Coltman, D. W., Bowen, W. D., and Wright, J. M. (1998). Birth weight and neonatal survival of harbour seal pups are positively correlated with genetic variation measured by microsatellites. Proc. Roy. Soc. Lond. B 265, 803–809. Coltman, D. W., Pilkington, J. G., Smith, J. A., and Pemberton, J. M. (1999). Parasite‐mediated selection against inbred soay sheep in a free‐living, island population. Evolution 53, 1259–1267. Cordero, P. J., Aparicio, J. M., and Veiga, J. P. (2004). Parental genetic characteristics and hatching success in the spotless starling, Sturnus unicolor. Anim. Behav. 67, 637–642. Coulson, T., Albon, S., Slate, J., and Pemberton, J. (1999). Microsatellite loci reveal sex‐dependent responses to inbreeding and outbreeding in red deer calves. Evolution 53, 1951–1960. Coulson, T. N., Pemberton, J. M., Albon, S. D., Beaumont, M., Marshall, T. C., Slate, J., Guinness, F. E., and Clutton‐Brock, T. H. (1998). Microsatellites reveal heterosis in red deer. Proc. Roy. Soc. Lond. B 265, 489–495. Da Silva, A., Luikart, G., Yoccoz, N. G., Cohas, A., and Allaine´, D. (2006). Genetic diversity‐ fitness correlation revealed by microsatellite analyses in European alpine marmots (Marmota marmota). Conserv. Genet. 7, 371–382.

266

BART KEMPENAERS

David, P. (1997). Modeling the genetic basis of heterosis: Tests of alternative hypotheses. Evolution 51, 1049–1057. David, P. (1998). Heterozygosity‐fitness correlations: New perspectives on old problems. Heredity 80, 531–537. de Bakker, P. I. W., McVean, G., Sabeti, P. C., Miretti, M. M., Green, T., Marchini, J., Ke, X., Monsuur, P., Whittaker, P., Delgado, M., Morrison, J., Richardson, A., et al. (2006). A high‐resolution HLA and SNP haplotype map for disease association studies in the extended human MHC. Nat. Genet. 38, 1166–1172. De Eyto, E., McGinnity, P., Consuegra, S., Coughlan, J., Tufto, J., Farrell, K., Megens, H.‐J., Jordan, W., Cross, T., and Stet, R. J. M. (2007). Natural selection acts on Atlantic salmon major histocompatibility (MH) variability in the wild. Proc. Roy. Soc. Lond. B 274, 861–869. Denk, A. G., Holzmann, A., Peter, A., Vermeirssen, E. L. M., and Kempenaers, B. (2005). Paternity in mallards: Effects of sperm quality and female sperm selection for inbreeding avoidance. Behav. Ecol. 16, 825–833. Dietrich, V., Schmoll, T., Winkel, W., Epplen, J. T., and Lubjuhn, T. (2004). Pair identity—an important factor concerning variation in extra‐pair paternity in the coal tit (Parus ater). Behaviour 141, 817–835. Doherty, P. C., and Zinkernagel, R. M. (1975). A biological role for the major histocompatibility antigens. Lancet 1, 1406–1409. Dorak, M. T., Lawson, T., Machulla, H. K. G., Mills, K. J., and Burnett, A. K. (2002). Increased heterozygosity for MHC class II lineages in newborn males. Genes Immun. 3, 263–269. Double, M. C., Peakall, R., Beck, N. R., and Cockburn, A. (2005). Dispersal, philopatry, and infidelity: Dissecting local genetic structure in superb fairy‐wrens (Malurus cyaneus). Evolution 59, 625–635. Duarte, L. C., Bouteiller, C., Fontanillas, P., Petit, E., and Perrin, N. (2003). Inbreeding in the greater white‐toothed shrew, Crocidura russula. Evolution 57, 638–645. Eberhard, W. G. (1996). ‘‘Female Control: Sexual Selection by Cryptic Female Choice.’’ Princeton University Press, Princeton. Edly‐Wright, C., Schwagmeyer, P. L., Parker, P. G., and Mock, D. W. (2007). Genetic similarity of mates, offspring health and extrapair fertilization in house sparrows. Anim. Behav. 73, 367–378. Edvardsson, M., and Arnqvist, G. (2006). No apparent indirect genetic benefits to female red flour beetles preferring males with intense copulatory courtship. Behav. Gen. 36, 775–782. Eimes, J. A., Parker, P. G., Brown, J. L., and Brown, E. R. (2005). Extrapair fertilization and genetic similarity of social mates in the Mexican jay. Behav. Ecol. 16, 456–460. Ekblom, R., Saether, A., Grahn, M., Fiske, P., Kalas, A., and Ho¨glund, J. (2004). Major histocompatibility complex variation and mate choice in a lekking bird, the great snipe (Gallinago media). Mol. Ecol. 13, 3821–3828. Eklund, A. C. (1999). Use of the MHC for mate choice in wild house mice (Mus domesticus). Genetica 104, 245–248. Engqvist, L. (2006). Females benefit from mating with different males in the scorpionfly Panorpa conata. Behav. Ecol. 17, 435–440. Ferreira, A. G. A., and Amos, W. (2006). Inbreeding depression and multiple regions showing heterozygote advantage in Drosophila melanogaster exposed to stress. Mol. Ecol. 15, 3885–3893. Fisher, D. O., Double, M. C., Blomberg, S. P., Jennions, M. D., and Cockburn, A. (2006). Post‐mating sexual selection increases lifetime fitness of polyandrous females in the wild. Nature 444, 89–92.

MATE CHOICE AND GENETIC QUALITY

267

Foerster, K., Delhey, K., Johnsen, A., Lifjeld, J. T., and Kempenaers, B. (2003). Females increase offspring heterozygosity and fitness through extra‐pair matings. Nature 425, 714–717. Foerster, K., Valcu, M., Johnsen, A., and Kempenaers, B. (2006). A spatial genetic structure and effects of relatedness on mate choice in a wild bird population. Mol. Ecol. 15, 4555–4567. Freeman‐Gallant, C. R., Meguerdichian, M., Wheelwright, N. T., and Sollecito, S. V. (2003). Social pairing and female mating fidelity predicted by restriction fragment length polymorphism similarity at the major histocompatibility complex in a songbird. Mol. Ecol. 12, 3077–3083. Freeman‐Gallant, C. R., Wheelwright, N. T., Meiklejohn, K. E., and Sollecito, S. V. (2006). Genetic similarity, extrapair paternity, and offspring quality in savannah sparrows (Passerculus sandwichensis). Behav. Ecol. 17, 952–958. Frommen, J. G., and Bakker, T. C. M. (2006). Inbreeding avoidance through non‐random mating in sticklebacks. Biol. Lett. 2, 232–235. Fukuda, N., and Touhara, K. (2006). Developmental expression patterns of testicular olfactory receptor genes during mouse spermatogenesis. Genes Cells 11, 71–81. Fukuda, N., Yomogida, K., Okabe, M., and Touhara, K. (2004). Functional characterization of a mouse testicular olfactory receptor and its role in chemosensing and in regulation of sperm motility. J. Cell Sci. 117, 5835–5845. Gage, M. J. G., Surridge, A. K., Tomkins, J. L., Green, E., Wiskin, L., Bell, D. J., and Hewitt, G. M. (2006). Reduced heterozygosity depresses sperm quality in wild rabbits, Oryctolagus cuniculus. Curr. Biol. 16, 612–617. Garant, D., Dodson, J. J., and Bernatchez, L. (2005). Offspring genetic diversity increases fitness of female Atlantic salmon (Salmo salar). Behav. Ecol. Sociobiol. 57, 240–244. Garcı´a‐Gonza´lez, F., and Simmons, L. W. (2005). The evolution of polyandry: Intrinsic sire effects contribute to embryo viability. J. Evol. Biol. 18, 1097–1103. Garner, T. W. J., and Schmidt, B. R. (2003). Relatedness, body size and paternity in the alpine newt, Triturus alpestris. Proc. Roy. Soc. Lond. B 270, 619–624. Garver‐Apgar, C. E., Gangestad, S. W., Thornhill, R., Miller, R. D., and Olp, J. J. (2006). Major histocompatibility complex alleles, sexual responsivity, and unfaithfulness in romantic couples. Psychol. Sci. 17, 830–835. Garvin, J. C., Abroe, B., Pedersen, M. C., Dunn, P. O., and Whittingham, L. A. (2006). Immune response of nestling warblers varies with extra‐pair paternity and temperature. Mol. Ecol. 15, 3833–3840. Gemmell, N. J., and Slate, J. (2006). Heterozygote advantage for fecundity. PLoS ONE 1, e125. doi:10.1371/journal.pone.0000125. Goudet, J., and Keller, L. (2002). The correlation between inbreeding and fitness: Does allele size matter? Trends Ecol. Evol. 17, 201–202. Gowaty, P. A. (1996). Battles of the sexes and origins of monogamy. In ‘‘Partnerships in Birds’’ (J. M. Black, Ed.), pp. 21–52. Oxford University Press, Oxford. Gustafsson, L., and Qvarnstro¨m, A. (2006). A test of the ‘‘sexy son’’ hypothesis: Sons of polygynous collared flycatchers do not inherit their fathers’ mating status. Am. Nat. 167, 297–302. Hadfield, J. D., Burgess, M. D., Lord, A., Phillimore, A. B., Clegg, S. M., and Owens, I. P. F. (2006). Direct versus indirect sexual selection: Genetic basis of colour, size and recruitment in a wild bird. Proc. Roy. Soc. Lond. B 273, 1347–1353. Halverson, M. A., Skelly, D. K., and Caccone, A. (2006). Inbreeding linked to amphibian survival in the wild but not in the laboratory. J. Hered. 97, 499–507.

268

BART KEMPENAERS

Hansson, B. (2004). Marker‐based relatedness predicts egg‐hatching failure in great reed warblers. Conserv. Genet. 5, 339–348. Hansson, B., and Westerberg, L. (2002). On the correlation between heterozygosity and fitness in natural populations. Mol. Ecol. 11, 2467–2474. Hansson, B., Bensch, S., Hasselquist, D., and Akesson, M. (2001). Microsatellite diversity predicts recruitment of sibling great reed warblers. Proc. Roy. Soc. Lond. B 268, 1287–1291. Hansson, B., Bensch, S., and Hasselquist, D. (2004a). Lifetime fitness of short‐ and long‐ distance dispersing great reed warblers. Evolution 58, 2546–2557. ˚ kesson, M., and Bensch, S. (2004b). Does Hansson, B., Westerdahl, H., Hasselquist, D., A linkage disequilibrium generate heterozygosity‐fitness correlations in Great Reed Warblers? Evolution 58, 870–879. Harf, R., and Sommer, S. (2005). Association between major histocompatibility complex class II DRB alleles and parasite load in the hairy‐footed gerbil, Gerbillurus paeba, in the southern Kalahari. Mol. Ecol. 14, 85–91. Hatchwell, B. J., Russell, A. F., Ross, D. J., and Fowlie, M. K. (2000). Divorce in cooperatively breeding long‐tailed tits: A consequence of inbreeding avoidance? Proc. Roy. Soc. Lond. B 267, 813–819. Hawley, D. M., Sydenstricker, K. V., Kollias, G. V., and Dhondt, A. A. (2005). Genetic diversity predicts pathogen resistance and cell‐mediated immunocompetence in house finches. Biol. Lett. 1, 326–329. Head, M. L., Hunt, J., Jennions, M. D., and Brooks, R. (2005). The indirect benefits of mating with attractive males outweigh the direct costs. PLoS Biol. 3, 289–294. Heath, D. D., Bryden, C. A., Shrimpton, J. M., Iwama, G. K., Kelly, J., and Heath, J. W. (2002). Relationships between heterozygosity, allelic distance (d2), and reproductive traits in chinook salmon, Oncorhynchus tshawytscha. Can. J. Fish. Aquat. Sci. 59, 77–84. Hedrick, P. W., Fredrickson, R., and Ellegren, H. (2001). Evaluation of d2, a microsatellite measure of inbreeding and outbreeding, in wolves with a known pedigree. Evolution 55, 1256–1260. Hoffman, J. I., Boyd, I. L., and Amos, W. (2004). Exploring the relationship between parental relatedness and male reproductive success in the Antarctic fur seal Arctocephalus gazella. Evolution 58, 2087–2099. Hoffman, J. I., Forcada, J., Trathan, P. N., and Amos, W. (2007). Female fur seals show active choice for males that are heterozygous and unrelated. Nature 445, 912–914. Ho¨glund, J., Piertney, S. B., Alatalo, R. V., Lindell, J., Lundberg, A., and Rintama¨ki, P. T. (2002). Inbreeding depression and male fitness in black grouse. Proc. Roy. Soc. Lond. B 269, 711–715. Holand, Ø., Askim, K. R., Røed, K. H., Weladji, R. B., Gjøstein, H., and Nieminen, M. (2006). No evidence of inbreeding avoidance in a polygynous ungulate: The reindeer (Rangifer tarandus). Biol. Lett. 3, 36–39. Holman, L., and Snook, R. R. (2006). Spermicide, cryptic female choice and the evolution of sperm form and function. J. Evol. Biol. 19, 1660–1670. Hosken, D. J., and Blanckenhorn, W. U. (1999). Female multiple mating, inbreeding avoidance, and fitness: It is not only the magnitude of the costs and benefits that counts. Behav. Ecol. 10, 462–464. Hosken, D. J., Garner, T. W. J., Tregenza, T., Wedell, N., and Ward, P. I. (2003). Superior sperm competitors sire higher‐quality young. Proc. Roy. Soc. Lond. B 270, 1933–1938. Howard, D. J. (1999). Conspecific sperm and pollen precedence and speciation. Annu. Rev. Ecol. Syst. 30, 109–132.

MATE CHOICE AND GENETIC QUALITY

269

Howard, R. S., and Lively, C. M. (2004). Good vs complementary genes for parasite resistance and the evolution of mate choice. BMC Evol. Biol. 4, 48. Hughes, K. A., Du, L., Rodd, F. H., and Reznick, D. N. (1999). Familiarity leads to female mate preference for novel males in the guppy, Poecilia reticulata. Anim. Behav. 58, 907–916. Hunt, J., Bussie`re, L. F., Jennions, M. D., and Brooks, R. (2004). What is genetic quality? Trends Ecol. Evol. 19, 329–333. Irwin, A. J., and Taylor, P. D. (2000). Heterozygous advantage and the evolution of female choice. Evol. Ecol. Res. 2, 119–128. Ivy, T. M. (2007). Good genes, genetic compatibility and the evolution of polyandry: Use of the diallel cross to address competing hypotheses. J. Evol. Biol. 20, 479–487. Jennions, M. D., and Petrie, M. (2000). Why do females mate multiply? A review of the genetic benefits. Biol. Rev. 75, 21–64. Jennions, M. D., Hunt, J., Graham, R., and Brooks, R. (2004). No evidence for inbreeding avoidance through postcopulatory mechanisms in the black field cricket, Teleogryllus commodus. Evolution 58, 2472–2477. Johnsen, A., Andersen, V., Sunding, C., and Lifjeld, J. T. (2000). Female bluethroats enhance offspring immunocompetence through extra‐pair copulations. Nature 406, 296–299. Johnstone, R. A., Reynolds, J. D., and Deutsch, J. C. (1996). Mutual mate choice and sex differences in choosiness. Evolution 50, 1382–1391. Joron, M., and Brakefield, P. M. (2003). Captivity masks inbreeding effects on male mating success in butterflies. Nature 424, 191–194. Kark, S., Safriel, U. N., Tabarroni, C., and Randi, E. (2001). Relationship between heterozygosity and asymmetry: A test across the distribution range. Heredity 86, 119–127. Keane, B. (1990). The effect of relatedness on reproductive success and mate choice in the white‐footed mouse, Peromyscus leucopus. Anim. Behav. 39, 264–273. Keller, L. F. (1998). Inbreeding and its fitness effects in an insular population of songs sparrows (Melospiza Melodia). Evolution 52, 240–250. Keller, L. F., and Arcese, P. (1998). No evidence for inbreeding avoidance in a natural population of song sparrows (Melospiza melodia). Am. Nat. 152, 380–392. Keller, L., and Reeve, H. K. (1995). Why do females mate with multiple males? The sexually selected sperm hypothesis. Adv. Study Behav. 24, 291–315. Keller, L. F., and Waller, D. M. (2002). Inbreeding effects in wild populations. Trends Ecol. Evol. 17, 230–241. Keller, L. F., Arcese, P., Smith, J. N. M., Hochachka, W. M., and Stearns, S. C. (1994). Selection against inbred song sparrows during a natural population bottleneck. Nature 372, 356–357. Keller, L. F., Grant, P. R., Grant, B. R., and Petren, K. (2002). Environmental conditions affect the magnitude of inbreeding depression in survival of Darwin’s finches. Evolution 56, 1229–1239. Kelley, J. L., Graves, J. A., and Magurran, A. E. (1999). Familiarity breeds contempt in guppies. Nature 401, 661–662. Kelly, C., and Price, T. D. (2005). Correcting for regression to the mean in behavior and ecology. Am. Nat. 166, 700–707. Kempenaers, B., Verheyen, G. R., Van den Broeck, M., Burke, T., Van Broeckhoven, C., and Dhondt, A. A. (1992). Extra‐pair paternity results from female preference for high‐quality males in the blue tit. Nature 357, 494–496. Kempenaers, B., Adriaensen, F., van Noordwijk, A. J., and Dhondt, A. A. (1996). Genetic similarity, inbreeding and hatching failure in blue tits: Are unhatched eggs infertile? Proc. Roy. Soc. Lond. B 263, 179–185.

270

BART KEMPENAERS

Kempenaers, B., Verheyen, G. R., and Dhondt, A. A. (1997). Extra‐pair paternity in the blue tit (Parus caeruleus): Female choice, male characteristics and offspring quality. Behav. Ecol. 8, 481–492. Kempenaers, B., Adriaensen, F., and Dhondt, A. A. (1998). Inbreeding and divorce in blue and great tits. Anim. Behav. 56, 737–740. Kempenaers, B., Congdon, B., Boag, P., and Robertson, R. J. (1999). Extrapair paternity and egg hatchability in tree swallows: Evidence for the genetic compatibility hypothesis? Behav. Ecol. 10, 304–311. Kempenaers, B., Foerster, K., Questiau, S., Robertson, B. C., and Vermeirssen, E. L. M. (2000). Distinguishing between female sperm choice versus male sperm competition: A comment on Birkhead. Evolution 54, 1050–1052. Kirkpatrick, M., and Barton, N. H. (1997). The strength of indirect selection on female mating preferences. Proc. Natl. Acad. Sci. USA 94, 1282–1286. Klein, J. (1986). ‘‘Natural History of the Major Histocompatibility Complex.’’ Wiley, New York. Kleven, O., and Lifjeld, J. T. (2005). No evidence for increased offspring heterozygosity from extrapair mating in the reed bunting (Emberiza schoeniclus). Behav. Ecol. 16, 561–565. Kleven, O., Jacobsen, F., Robertson, R. J., and Lifjeld, J. T. (2005). Extrapair mating between relatives in the barn swallow: A role for kin selection? Biol. Lett. 1, 389–392. Kokko, H., and Ots, I. (2006). When not to avoid inbreeding? Evolution 60, 467–475. Kokko, H., Brooks, R., Jennions, M. D., and Morley, J. (2003). The evolution of mate choice and mating biases. Proc. Roy. Soc. Lond. B 270, 653–664. Kokko, H., Jennions, M. D., and Brooks, R. (2006). Unifying and testing models of sexual selection. Annu. Rev. Ecol. Syst. 37, 43–66. Kraaijeveld‐Smit, F. J. L., Ward, S. J., Temple‐Smith, P. D., and Paetkau, D. (2002). Factors influencing paternity success in Antechinus agilis: Last‐male sperm precedence, timing of mating and genetic compatibility. J. Evol. Biol. 15, 100–107. Krokene, C., and Lifjeld, J. T. (2000). Variation in the frequency of extra‐pair paternity in birds: A comparison of an island and a mainland population of blue tits. Behaviour 137, 1317–1330. Kruuk, L. E. B., Sheldon, B. C., and Merila¨, J. (2002). Severe inbreeding depression in collared flycatchers (Ficedula albicollis). Proc. Roy. Soc. Lond. B 269, 1581–1589. Kurtz, J., Kalbe, M., Aeschlimann, P. B., Haberli, M. A., Wegner, K. M., Reusch, T. B. H., and Milinski, M. (2004). Major histocompatibility complex diversity influences parasite resistance and innate immunity in sticklebacks. Proc. Roy. Soc. Lond. B 271, 197–204. Lampert, K. P., Bernal, X. E., Rand, A. S., Mueller, U. G., and Ryan, M. J. (2006). No evidence for female mate choice based on genetic similarity in the tu´ngara frog Physalaemus pustulosus. Behav. Ecol. Sociobiol. 59, 796–804. Landry, C., Garant, D., Duchesne, P., and Bernatchez, L. (2001). ‘‘Good genes as heterozygosity’’: The major histocompatibility complex and mate choice in Altantic salmon (Salmo salar). Proc. Roy. Soc. Lond. B 268, 1279–1285. Langefors, A., Lohm, J., Grahn, M., Andersen, O., and von Schantz, T. (2001). Association between major histocompatibility complex class IIB alleles and resistance to Aeromonas salmonicida in Atlantic salmon. Proc. Roy. Soc. Lond. B 268, 479–485. LeBas, N. R. (2002). Mate choice, genetic incompatibility, and outbreeding in the ornate dragon lizard, Ctenophorus ornatus. Evolution 56, 371–377. Lehmann, L., and Perrin, N. (2003). Inbreeding avoidance through kin recognition: Choosy females boost male dispersal. Am. Nat. 162, 638–652. Lehmann, L., Keller, L. F., and Kokko, H. (2006). Mate choice evolution, dominance effects, and the maintenance of genetic variation. J. Theor. Biol. 244, 282–295.

MATE CHOICE AND GENETIC QUALITY

271

Lens, L., Van Dongen, S., Galbusera, P., Schenk, T., Matthysen, E., and Van de Casteele, T. (2000). Developmental instability and inbreeding in natural bird populations exposed to different levels of habitat disturbance. J. Evol. Biol. 13, 889–896. Lesbarre`res, D., Primmer, C. R., Laurila, A., and Merila¨, J. (2005). Environmental and population dependency of genetic variability‐fitness correlations in Rana temporaria. Mol. Ecol. 14, 311–323. Levitan, D. R., and Ferrell, D. L. (2006). Selection on gamete recognition proteins depends on sex, density, and genotype frequency. Science 312, 267–269. Lieutenant‐Gosselin, M., and Bernatchez, L. (2006). Local heterozygosity‐fitness correlations with global positive effects on fitness in threespine stickleback. Evolution 60, 1658–1668. Ludlow, A. M., and Magurran, A. E. (2006). Gametic isolation in guppies (Poecilia reticulata). Proc. Roy. Soc. Lond. B 273, 2477–2482. Lynch, M., and Walsh, B. (1998). ‘‘Genetics and Analysis of Quantitative Traits.’’ Sinauer Associates Inc., Sunderland, Massachusetts, USA. MacDougall‐Shackleton, E. A., Derryberry, E. P., Foufopoulos, J., Dobson, A. P., and Hahn, T. P. (2005). Parasite‐mediated heterozygote advantage in an outbred songbird population. Biol. Lett. 1, 105–107. Mack, P. D., Hammock, B. A., and Promislow, D. E. L. (2002). Sperm competitive ability and genetic relatedness in Drosophila melanogaster: Similarity breeds contempt. Evolution 56, 1789–1795. Madsen, T., and Ujvari, B. (2006). MHC class I variation associates with parasite resistance and longevity in tropical pythons. J. Evol. Biol. 19, 1973–1978. Madsen, T., Shine, R., Loman, J., and Ha˚kansson, T. (1992). Why do female adders copulate so frequently? Nature 355, 440–441. Markert, J. A., Grant, P. R., Grant, B. R., Keller, L. F., Coombs, J. L., and Petren, K. (2004). Neutral locus heterozygosity, inbreeding, and survival in Darwin’s ground finches (Geospiza fortis and G. scandens). Heredity 92, 306–315. Marr, A. B., Arcese, P., Hochachka, W. M., Reid, J. M., and Keller, L. F. (2006). Interactive effects of environmental stress and inbreeding on reproductive traits in a wild bird population. J. Anim. Ecol. 75, 1406–1415. Marshall, R. C., Buchanan, K. L., and Catchpole, C. K. (2003). Sexual selection and individual genetic diversity in a songbird. Proc. Roy. Soc. Lond B (Suppl.)270, S248–S250. Marshall, T. C., and Spalton, J. A. (2000). Simultaneous inbreeding and outbreeding depression in reintroduced Arabian oryx. Anim. Conserv. 3, 241–248. Masters, B. S., Hicks, B. G., Johnson, L. S., and Erb, L. A. (2003). Genotype and extra‐pair paternity in the house wren: A rare‐male effect? Proc. Roy. Soc. Lond. B 270, 1393–1397. Mays, H. L., and Hill, G. E. (2004). Choosing mates: Good genes versus genes that are a good fit. Trends Ecol. Evol. 19, 554–559. Mazzi, D., Ku¨nzler, R., Largiade`r, C. R., and Bakker, T. C. M. (2004). Inbreeding affects female preference for symmetry in computer‐animated sticklebacks. Behav. Genet. 34, 417–424. McPeek, M. A., and Gavrilets, S. (2006). The evolution of female mating preferences: Differentiation from species with promiscuous males can promote speciation. Evolution 60, 1967–1980. Meagher, S., Penn, D. J., and Potts, W. K. (2000). Male‐male competition magnifies inbreeding depression in wild house mice. Proc. Natl. Acad. Sci. USA 97, 3324–3329. Merila¨, J., Sheldon, B. C., and Griffith, S. C. (2003). Heterotic effects on fitness in a wild bird population. Ann. Zool. Fenn. 40, 269–280. Milinski, M. (2003). The function of mate choice in sticklebacks: Optimizing Mhc genetics. J. Fish Biol. 63, 1–16.

272

BART KEMPENAERS

Milinski, M., Griffiths, S., Wegner, K. M., Reusch, T. B. H., Haas‐Assenbaum, A., and Boehm, T. (2005). Mate choice decisions of stickleback females predictably modified by MHC peptide ligands. Proc. Natl. Acad. Sci. USA 102, 4414–4418. Mitton, J. B. (1997). ‘‘Selection in Natural Populations.’’ Oxford University Press, Oxford. Mitton, J. B., and Grant, M. C. (1984). Associations among protein heterozygosity, growth rate, and developmental homeostatis. Annu. Rev. Ecol. Syst. 15, 479–499. Mitton, J. B., Schuster, W. S. F., Cothran, E. G., and de Fries, J. C. (1993). Correlation between the individual heterozygosity of parents and their offspring. Heredity 71, 59–63. Møller, A. P., and Jennions, M. D. (2001). How important are direct fitness benefits of sexual selection? Naturwissenschaften 88, 401–415. Møller, A. P., and Jennions, M. D. (2002). How much variance can be explained by ecologists and evolutionary biologists? Oecologia 132, 492–500. Neff, B. D. (2004). Stabilizing selection on genomic divergence in a wild fish population. PNAS 101, 2381–2385. Neff, B. D., and Pitcher, T. E. (2005). Genetic quality and sexual selection: An integrated framework for good genes and compatible genes. Mol. Ecol. 14, 19–38. Newcomer, S. D., Zeh, J. A., and Zeh, D. W. (1999). Genetic benefits enhance the reproductive success of polyandrous females. Proc. Natl. Acad. Sci. USA 96, 10236–10241. Nowak, M. A., Tarczy‐Hornoch, K., and Austyn, J. M. (1992). The optimal number of major histocompatibility complex‐molecules in an individual. Proc. Natl. Acad. Sci. USA 89, 10896–10899. Oh, K. P., and Badyaev, A. V. (2006). Adaptive genetic complementarity in mate choice coexists with selection for elaborate sexual traits. Proc. Roy. Soc. Lond. B 273, 1913–1919. Olsson, M., Gullberg, A., Tegelstro¨m, H., Madsen, T., and Shine, R. (1994). Can female adders multiply? Nature 369, 528. Olsson, M., Madsen, T., Ujvari, B., and Wapstra, E. (2004). Fecundity and MHC affects ejaculation tactics and paternity bias in sand lizards. Evolution 58, 906–909. Olsson, M., Shine, R., Madsen, T., Gullberg, A., and Tegelstro¨m, H. (1996). Sperm selection by females. Nature 383, 585. Oneal, E., Connallon, T., and Knowles, L. L. (2006). Conflict between direct and indirect benefits of female choice in desert Drosophila. Biol. Lett. 3, 29–32. Otter, K. A., Stewart, I. R. K., McGregor, P. K., Terry, A. M. R., Dabelsteen, T., and Burke, T. (2001). Extra‐pair paternity among Great Tits Parus major following manipulation of male signals. J. Avian Biol. 32, 338–344. Overall, A. D. J., Byrne, K. A., Pilkington, J. G., and Pemberton, J. M. (2005). Heterozygosity, inbreeding and neonatal traits in Soay sheep on St Kilda. Mol. Ecol. 14, 3383–3393. Pai, A., and Yan, G. (2002). Female mate choice in relation to heterozygosity in Tribolium castaneum. J. Evol. Biol. 15, 1076–1082. Parker, G. A. (1992). Snakes and female sexuality. Nature 355, 395–396. Parrott, M. L., Ward, S. J., and Temple‐Smith, P. D. (2006). Genetic similarity, not male size, influences female mate choice in the agile antechinus (Antechinus agilis). Aust. J. Zool. 54, 319–323. Paterson, S., and Pemberton, J. M. (1997). No evidence for major histocompatibility complex‐ dependent mating patterns in a free‐living ruminant population. Proc. Roy. Soc. Lond B 264, 1813–1819. Paterson, S., Wilson, K., and Pemberton, J. M. (1998). Major histocompatibility complex variation associated with juvenile survival and parasite resistance in a large unmanaged ungulate population (Obis aries L.). Proc. Natl. Acad. Sci. USA 95, 3714–3719.

MATE CHOICE AND GENETIC QUALITY

273

Peacock, M. M., and Smith, A. T. (1997). Nonrandom mating in pikas Ochotona princeps: Evidence for inbreeding between individuals of intermediate relatedness. Mol. Ecol. 6, 801–811. Pemberton, J. (2004). Measuring inbreeding depression in the wild: The old ways are the best. Trends Ecol. Evol. 19, 613–615. Penn, D. J., and Potts, W. K. (1999). The evolution of mating preferences and major histocompatibility complex genes. Am. Nat. 153, 145–164. Penn, D. J., Damjanovch, K., and Potts, W. K. (2002). MHC heterozygosity confers a selective advantage against multiple–strain infections. Proc. Natl. Acad. Sci. USA 99, 11260–11264. Pfennig, K. S., and Tinsley, R. C. (2002). Different mate preferences by parasitized and unparasitized females potentially reduces sexual selection. J. Evol. Biol. 15, 399–406. Pia´lek, J., and Albrecht, T. (2005). Choosing mates: Complementary versus compatible genes. Trends Ecol. Evol. 20, 63. Piertney, S. B., and Oliver, M. K. (2006). The evolutionary ecology of the major histocompatibility complex. Heredity 96, 7–21. Pilastro, A., Simonato, M., Bisazza, A., and Evans, J. P. (2004). Cryptic female preference for colorful males in guppies. Evolution 58, 665–669. Pitcher, T. E., and Neff, B. D. (2006). MHC class IIB alleles contribute to both additive and nonadditive genetic effects on survival in Chinook salmon. Mol. Ecol. 15, 2357–2365. Pitnick, S., and Brown, W. D. (2000). Criteria for demonstrating female sperm choice. Evolution 54, 1052–1056. Pizzari, T., and Birkhead, T. R. (2002). The sexually‐selected sperm hypothesis: Sex‐biased inheritance and sexual antagonism. Biol. Rev. 77, 183–209. Pizzari, T., Løvlie, H., and Cornwallis, C. K. (2004). Sex‐specific, counteracting responses to inbreeding in a bird. Proc. Roy. Soc. Lond. B 271, 2115–2121. Poulin, R. (1994). Mate choice decisions by parasitized female upland bullies, Gobiomorphus breviceps. Proc. Roy. Soc. Lond B 256, 183–187. Puurtinen, M., Ketola, T., and Kotiaho, J. S. (2005). Genetic compatibility and sexual selection. Trends Ecol. Evol. 20, 157–158. Queller, D. C., and Goodnight, K. F. (1989). Estimating relatedness using genetic markers. Evolution 43, 258–275. Qvarnstro¨m, A., Brommer, J. E., and Gustafsson, L. (2006). Testing the genetics underlying the co‐evolution of mate choice and ornament in the wild. Nature 441, 84–86. Ra¨tti, O., Hovi, M., Lundberg, A., Tegelstro¨m, H., and Alatalo, R. V. (1995). Extra‐pair paternity and male characteristics in the pied flycatcher. Behav. Ecol. Sociobiol. 37, 419–425. Reid, J. M., Arcese, P., and Keller, L. F. (2003). Inbreeding depresses immune response in song sparrows (Melospiza melodia): Direct and inter‐generational effects. Proc. Roy. Soc. Lond. B 270, 2151–2157. Reid, J. M., Arcese, P., Cassidy, A. E., Marr, A. B., Smith, J. M., and Keller, L. F. (2005). Hamilton and Zuk meet heterozygosity? Song repertoire size indicates inbreeding and immunity in song sparrows (Melospiza melodia). Proc. Roy. Soc. Lond. B 272, 481–487. Reid, J. M., Arcese, P., and Keller, L. F. (2006). Intrinsic parent‐offspring correlation in inbreeding level in a song sparrow (Melospiza melodia) population open to immigration. Am. Nat. 168, 1–13. Reinhold, K. (2002). Modelling the evolution of female choice strategies under inbreeding conditions. Genetica 116, 189–195.

274

BART KEMPENAERS

Reusch, T. B. H., Ha¨berli, M. A., Aeschlimann, P. B., and Milinski, M. (2001). Female sticklebacks count alleles in a strategy of sexual selection explaining MHC polymorphism. Nature 414, 300–302. Rice, W. R. (1996). Sexually antagonistic male adaptation triggered by experimental arrest of female evolution. Nature 381, 232–234. Rice, W. R., and Chippindale, A. K. (2002). The evolution of hybrid infertility: Perpetual coevolution between gender‐specific and sexually antagonistic genes. Genetica 116, 179–188. Richardson, D. S., Komdeur, J., and Burke, T. (2004). Inbreeding in the Seychelles warbler: Environment‐dependent maternal effects. Evolution 58, 2037–2048. Richardson, D. S., Komdeur, J., Burke, T., and von Schanz, T. (2005). MHC‐based patterns of social and extra‐pair mate choice in the Seychelles warbler. Proc. Roy. Soc. Lond. B 272, 759–767. Ritland, K. (1996). Estimators of pairwise relatedness and individual inbreeding coefficients. Genet. Res. 67, 175–185. Roberts, S. C., and Gosling, L. M. (2003). Genetic similarity and quality interact in mate choice decisions by female mice. Nat. Genet. 35, 103–106. Roberts, S. C., Hale, M. L., and Petrie, M. (2006). Correlations between heterozygosity and measures of genetic similarity: Implications for understanding mate choice. J. Evol. Biol. 19, 558–569. Roberts, S. C., Little, A. C., Gosling, L. M., Jones, B. C., Perrett, D. I., Carter, V., and Petrie, M. (2005a). MHC‐assortative facial preferences in humans. Biol. Lett. 1, 400–403. Roberts, S. C., Little, A. C., Gosling, L. M., Perrett, D. I., Carter, V., Jones, B. C., Penton‐Voak, I., and Petrie, M. (2005b). MHC‐heterozygosity and human facial attractiveness. Evol. Hum. Behav. 26, 213–226. Rola´n‐Alvarez, E., Zapata, C., and Alvarez, G. (1995). Multilocus heterozygosity and sexual selection in a natural population of the marine snail Littorina mariae (Gastropoda: Prosobranchia). Heredity 75, 17–25. Rossiter, S. J., Jones, G., Ransome, R. D., and Barratt, E. M. (2001). Outbreeding increases offspring survival in wild greater horseshoe bats (Rhinolophus ferrumequinum). Proc. Roy. Soc. Lond. B 268, 1055–1061. Rowe, G., and Beebee, T. J. C. (2005). Intraspecific competition disadvantages inbred natterjack toad (Bufo calamita) genotypes over outbred ones in a shared pond environment. J. Anim. Ecol. 74, 71–76. Rugman‐Jones, P. F., and Eady, P. E. (2007). Conspecific sperm precedence in Callosobruchus subinnotatus (Coleoptera: Bruchidae): Mechanisms and consequences. Proc. Roy. Soc. Lond. B 274, 983–988. Ru¨licke, T., Chapuisat, M., Homberger, F. R., Macas, E., and Wedekind, C. (1998). MHC‐genotype of progeny influenced by parental infection. Proc. Roy. Soc. Lond. B 265, 711–716. Ru¨licke, T., Guncz, N., and Wedekind, C. (2006). Early maternal investment in mice: No evidence for compatible‐genes sexual selection despite hybrid vigor. J. Evol. Biol. 19, 922–928. Saccheri, I. J., Lloyd, H. D., Helyar, S. J., and Brakefield, P. M. (2005). Inbreeding uncovers fundamental differences in the genetic load affecting male and female fertility in a butterfly. Proc. Roy. Soc. Lond. B 272, 29–46. Santos, P. S. C., Schinemann, J. A., Gabardo, J., and da Graca Bicalho, M. (2005). New evidence that the MHC influences odor perception in humans: A study with 58 Southern Brazilian students. Horm. Behav. 47, 384–388.

MATE CHOICE AND GENETIC QUALITY

275

Sauermann, U., Nu¨rnberg, P., Bercovitch, F. B., Berard, J. D., Trefilov, A., Widdig, A., Kessler, M., Schmidtke, J., and Krawczak, M. (2001). Increased reproductive success of MHC class II heterozygous males among free‐ranging rhesus macaques. Hum. Genet. 108, 249–254. Schmoll, T., Dietrich, V., Winkel, W., Epplen, J. T., and Lubjuhn, T. (2003). Long‐term fitness consequences of female extra‐pair matings in a socially monogamous passerine. Proc. Roy. Soc. Lond. B 270, 259–264. Schmoll, T., Dietrich, V., Winkel, W., Epplen, J. T., Schurr, F., and Lubjuhn, T. (2005a). Paternal genetic effects on offspring fitness are context dependent within the extrapair mating system of a socially monogamous passerine. Evolution 59, 645–657. Schmoll, T., Quellmalz, A., Dietrich, V., Winkel, W., Epplen, J. T., and Lubjuhn, T. (2005b). Genetic similarity between pair mates is not related to extrapair paternity in the socially monogamous coal tit. Anim. Behav. 69, 1013–1022. Seddon, N., Amos, W., Mulder, R. A., and Tobias, J. A. (2004). Male heterozygosity predicts territory size, song structure and reproductive success in a cooperatively breeding bird. Proc. Roy. Soc. Lond. B 271, 1823–1829. Servedio, M. R., and Lande, R. (2006). Population genetic models of male and mutual mate choice. Evolution 60, 674–685. Sheldon, B. C. (2000). Differential allocation: Tests, mechanisms and implications. Trends Ecol. Evol. 15, 397–402. Shorey, L., Piertney, S., Stone, J., and Ho¨glund, J. (2000). Fine‐scale genetic structuring on Manacus manacus leks. Nature 408, 352–353. Shutler, D., Hussell, D. J. T., Horn, A. G., Leonard, M. L., Shutler, R. W., and Lepage, D. (2004). Breed ng between tree swallows from the same brood. J. Field Ornithol. 75, 353–358. Sillero‐Zubiri, C., Gottelli, D., and Macdonald, D. W. (1996). Male philopatry, extra‐pack copulations and inbreeding avoidance in Ethiopian wolves (Canis simensis). Behav. Ecol. Sociobiol. 38, 331–340. Simmons, L. W. (1991). Female choice and the relatedness of mates in the field cricket, Gryllus bimaculatus. Anim. Behav. 41, 493–501. Simmons, L. W. (2001). The evolution of polandry: An examination of the genetic compatibility and good‐sperm hypotheses. J. Evol. Biol. 14, 585–594. Simmons, L. W. (2005). The evolution of polyandry: Sperm competition, sperm selection, and offspring viability. Annu. Rev. Ecol. Syst. 36, 125–146. Simmons, L. W., Beveridge, M., Wedell, N., and Tregenza, T. (2006). Postcopulatory inbreeding avoidance by female crickets only revealed by molecular markers. Mol. Ecol. 15, 3817–3824. Slate, J., and Pemberton, J. M. (2002). Comparing molecular measures for detecting inbreeding depression. J. Evol. Biol. 15, 20–31. Slate, J., Kruuk, L. E. B., Marshall, T. C., Pemberton, J. M., and Clutton‐Brock, T. H. (2000). Inbreeding depression influences lifetime breeding success in a wild population of red deer (Cervus elaphus). Proc. Roy. Soc. Lond. B 267, 1657–1662. Slate, J., David, P., Dodds, K. G., Veenvliet, B. A., Glass, B. C., Broad, T. E., and McEwan, J. C. (2004). Understanding the relationship between the inbreeding coefficient and multilocus heterozygosity: Theoretical expectations and empirical data. Heredity 93, 255–265. Smith, S. B., Webster, M. S., and Holmes, R. T. (2005). The heterozygosity theory of extra‐pair mate choice in birds: A test and a cautionary note. J. Avian Biol. 36, 146–154. Snow, L. S. E., and Andrade, M. C. B. (2005). Multiple sperm storage organs facilitate female control of paternity. Proc. Roy. Soc. Lond. B 272, 1139–1144. Sommer, S. (2005). Major histocompatibility complex and mate choice in a monogamous rodent. Behav. Ecol. Sociobiol. 58, 181–189.

276

BART KEMPENAERS

Spehr, M., Schwane, K., Riffell, J. A., Zimmer, R. K., and Hatt, H. (2006). Odorant receptors and olfactory‐like signaling mechanisms in mammalian sperm. Mol. Cell. Endocrinol. 250, 128–136. Spielman, D., Brook, B. W., Briscoe, D. A., and Frankham, R. (2004). Does inbreeding and loss of genetic diversity decrease disease resistance? Conserv. Genet. 5, 439–448. Spottiswoode, C., and Møller, A. P. (2004). Genetic similarity and hatching success in birds. Proc. Roy. Soc. Lond. B 271, 267–272. Stewart, I. R. K., Hanschu, R. D., Burke, T., and Westneat, D. F. (2006). Tests of ecological, phenotypic, and genetic correlates of extra‐pair paternity in the house sparrow. Condor 108, 399–413. Stockley, P. (1999). Sperm selection and genetic incompatibility: Does relatedness of mates affect male success in sperm competition? Proc. Roy. Soc. Lond. B 266, 1663–1669. Stockley, P. (2003). Female multiple mating behaviour, early reproductive failure and litter size variation in mammals. Proc. Roy. Soc. Lond. B 270, 271–278. Stockley, P., Searle, J. B., MacDonald, D. W., and Jones, C. S. (1993). Female multiple mating behaviour in the common shrew as a strategy to reduce inbreeding. Proc. Roy. Soc. Lond. B 254, 173–179. Stow, A. J., and Sunnucks, P. (2004). Inbreeding avoidance in Cunningham’s skinks (Egernia cunninghami) in natural and fragmented habitat. Mol. Ecol. 13, 443–447. Summers, K., and Amos, W. (1996). Behavioral, ecological, and molecular genetic analyses of reproductive strategies in the Amazonian dart‐poison frog, Dendrobates ventrimaculatus. Behav. Ecol. 8, 260–267. Tarvin, K. A., Webster, M. S., Tuttle, E. M., and Pruett‐Jones, S. (2005). Genetic similarity of social mates predicts the level of extrapair paternity in splendid fairy‐wrens. Anim. Behav. 70, 945–955. Telford, S. R., and Jennions, M. D. (1998). Establishing cryptic female choice in animals. Trends Ecol. Evol. 13, 216–218. Thornhill, R., Gangestad, S. W., Miller, R., Scheyd, G., McCollough, J. K., and Franklin, M. (2003). Major histocompatibility complex genes, symmetry, and body scent attractiveness in men and women. Behav. Ecol. 14, 668–678. Thuman, K. A., and Griffith, S. C. (2005). Genetic similarity and the nonrandom distribution of paternity in a genetically highly polyandrous shorebird. Anim. Behav. 69, 765–770. Tiira, K., Laurila, A., Peuhkuri, N., Piironen, J., Aikio, S., Ranta, E., and Primmer, C. R. (2003). Aggressiveness is associated with genetic diversity in landlocked salmon (Salmo salar). Mol. Ecol. 12, 2399–2407. Tiira, K., Laurila, A., Enberg, K., Piironen, J., Aikio, S., Ranta, E., and Primmer, C. R. (2006). Do dominants have higher heterozygosity? Social status and genetic variation in brown trout, Salmo trutta. Behav. Ecol. Sociobiol. 59, 657–665. Tomiuk, J., Stauss, M., Segelbacher, G., Fietz, J., Ko¨mpf, J., and Bachmann, L. (2006). Female genetic heterogeneity affects the reproduction of great tits (Parus major L., 1758) in low‐quality woodlands. J. Zool. Syst. Evol. Res. 45, 144–150. Tregenza, T., and Wedell, N. (2000). Genetic compatibility, mate choice and patterns of parentage: Invited review. Mol. Ecol. 9, 1013–1027. Tregenza, T., and Wedell, N. (2002). Polyandrous females avoid costs of inbreeding. Nature 415, 71–73. Trivers, R. L. (1972). Parental investment and sexual selection. In ‘‘Sexual Selection and the Descent of Man, 1871–1971’’ (B. Campbell, Ed.), pp. 136–179. Aldine Press, Chicago. Tsitrone, A., Rousset, F., and David, P. (2001). Heterosis, marker mutational processes and population inbreeding history. Genetics 159, 1845–1859.

MATE CHOICE AND GENETIC QUALITY

277

Urbach, D., Folstad, I., and Rudolfsen, G. (2005). Effects of ovarian fluid on sperm velocity in Arctic charr (Salvelinus alpinus). Behav. Ecol. Sociobiol. 57, 438–444. Van de Casteele, T., Galbusera, P., and Matthysen, E. (2001). A comparison of microsatellite‐ based pairwise relatedness estimators. Mol. Ecol. 10, 1539–1549. Van de Casteele, T., Galbusera, P., Schenck, T., and Matthysen, E. (2003). Seasonal and lifetime reproductive consequences of inbreeding in the great tit Parus major. Behav. Ecol. 14, 165–174. Vanderhaeghen, P., Schurmans, S., Vassart, G., and Parmentier, M. (1993). Olfactory receptors are displayed on dog mature sperm cells. J. Cell Biol. 123, 1441–1452. Van Oosterhout, C., Trigg, R. E., Carvalho, G. R., Magurran, A. E., and Hauser, L. (2003). Inbreeding depression and genetic load of sexually selected traits: How the guppy lost its spots. J. Evol. Biol. 16, 273–281. Van Oosterhout, C., Van Heuven, M. K., and Brakefield, P. M. (2004). On the neutrality of molecular genetic markers: Pedigree analysis of genetic variation in fragmented populations. Mol. Ecol. 13, 1025–1034. Veen, T., Borge, T., Griffith, S. C., Saetre, G.‐P., Bures, S., Gustafsson, L., and Sheldon, B. C. (2001). Hybridization and adaptive mate choice in flycatchers. Nature 411, 45–50. Viken, A., Fleming, I. A., and Rosenqvist, G. (2006). Premating avoidance of inbreeding absent in female guppies (Poecilia reticulata). Ethology 112, 716–723. Vøllestad, L. A., Hindar, K., and Møller, A. P. (1999). A meta‐analysis of fluctuating asymmetry in relation to heterozygosity. Heredity 83, 206–218. Von Schantz, T., Wittzell, H., Go¨ransson, G., Grahn, M., and Persson, K. (1996). MHC genotype and male ornamentation: Genetic evidence for the Hamilton‐Zuk model. Proc. Roy. Soc. Lond. B 263, 265–271. Waser, P. M., and DeWoody, J. A. (2006). Multiple paternity in a philopatric rodent: The interaction of competition and choice. Behav. Ecol. 17, 971–978. Weatherhead, P. J., Dufour, K. W., Lougheed, S. C., and Eckert, S. G. (1999). A test of the good‐genes‐as‐heterozygosity hypothesis using red‐winged blackbirds. Behav. Ecol. 10, 619–625. Weatherhead, P. J., Prosser, M. R., Gibbs, H. L., and Brown, G. P. (2002). Male reproductive success and sexual selection in northern water snakes determined by microsatellite DNA analysis. Behav. Ecol. 13, 808–815. Wedekind, C., and Fu¨ri, S. (1997). Body odour preferences in men and women: Do they aim for specific MHC combinations or simply heterozygosity? Proc. Roy. Soc. Lond. B 264, 1471–1479. Wedekind, C., Seebeck, T., Bettens, F., and Paepke, A. J. (1995). MHC‐dependent mate preferences in humans. Proc. Roy. Soc. Lond. B 260, 245–249. Wedekind, C., Chapuisat, M., Macas, E., and Ru¨licke, T. (1996). Non‐random fertilization in mice correlates with the MHC and something else. Heredity 77, 400–409. Wedekind, C., Mu¨ller, R., and Spicher, H. (2001). Potential genetic benefits of mate selection in whitefish. J. Evol. Biol. 14, 980–986. Wegner, K. M., Kalbe, M., Kurtz, J., Reusch, T. B. H., and Milinski, M. (2003a). Parasite selection for immunogenetic optimality. Science 301, 1343. Wegner, K. M., Reusch, T. B. H., and Kalbe, M. (2003b). Multiple parasites are driving major histocompatibility complex polymorphism in the wild. J. Evol. Biol. 16, 224–232. Wegner, K. M., Kalbe, M., Schaschl, H., and Reusch, T. B. H. (2004). Parasites and individual major histocompatibility complex diversity—an optimal choice? Microb. Infect. 6, 1110–1116. Westerdahl, H. (2004). No evidence of an MHC‐based female mating preference in great reed warblers. Mol. Ecol. 13, 2465–2470.

278

BART KEMPENAERS

Westerdahl, H., Waldenstro¨m, J., Hansson, B., Hasselquist, D., von Schanz, T., and Bensch, S. (2005). Associations between malaria and MHC genes in a migratory songbird. Proc. Roy. Soc. Lond. B 272, 1511–1518. Wheelwright, N. T., Freeman‐Gallant, C. R., and Mauck, R. A. (2006). Asymmetrical incest avoidance in the choice of social and genetic mates. Anim. Behav. 71, 631–639. Whiteman, N. K., Matson, K. D., Bollmer, J. L., and Parker, P. G. (2006). Disease ecology in the Gala´pagos Hawk (Buteo galapagoensis): Host genetic diversity, parasite load and natural antibodies. Proc. Roy. Soc. B 273, 797–804. Widdig, A., Bercovitch, F. B., Streich, W. J., Sauermann, U., Nu¨rnberg, P., and Krawczak, M. (2004). A longitudinal analysis of reproductive skew in male rhesus macaques. Proc. Roy. Soc. Lond. B 271, 819–826. Woxvold, I. A., Adcock, G. J., and Mulder, R. A. (2006). Fine‐scale genetic structure and dispersal in cooperatively breeding apostlebirds. Mol. Ecol. 15, 3139–3146. Yasui, Y. (1997). A ‘good‐sperm’ model can explain the evolution of costly multiple mating by females. Am. Nat. 149, 573–584. Zedrosser, A., Bellemain, E., Taberlet, P., and Swenson, J. E. (2007). Genetic estimates of annual reproductive success in male brown bears: The effects of body size, age, internal relatedness and population density. J. Anim. Ecol. 76, 368–375. Zeh, J. A., and Zeh, D. W. (1996). The evolution of polyandry. 1. Intragenomic conflict and genetic incompatibility. Proc. Roy. Soc. Lond. B 263, 1711–1717. Zeh, J. A., and Zeh, D. W. (1997). The evolution of polyandry. 2. Post‐copulatory defences against genetic incompatibility. Proc. Roy. Soc. Lond. B 264, 69–75. Zeh, J. A., and Zeh, D. W. (2001). Reproductive mode and the genetic benefits of polyandry. Anim. Behav. 61, 1051–1063. Zeh, J. A., and Zeh, D. W. (2006). Outbred embryos rescue inbred half‐siblings in mixed‐ paternity broods of live‐bearing females. Nature 439, 201–203. Ziegler, A., Kentenich, H., and Uchanska‐Ziegler, B. (2005). Female choice and the MHC. Trends Immunol. 26, 496–502. Zouros, E. (1993). Associative overdominance: Evaluating the effects of inbreeding and linkage disequilibrium. Genetica 89, 35–46. Zouros, E., Romero‐Dorey, M., and Mallet, A. E. (1988). Heterozygosity and growth in marine bivalves: Further data and possible explanations. Evolution 42, 1332–1341.

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

Sexual Conflict and the Evolution of Breeding Systems in Shorebirds Gavin H. Thomas,* Tama´s Sze´kely,{,1 and John D. Reynolds{ *nerc centre for population biology, imperial college london silwood park, ascot, berkshire sl5 7py, united kingdom { department of biology and biochemistry, university of bath claverton down, bath ba2 7ay, united kingdom { department of biological sciences, simon fraser university, burnaby british columbia, canada v5a 1s6

I. SEXUAL CONFLICT AND SHOREBIRD BREEDING SYSTEMS Sexual conflict emerges from the different evolutionary interests of males and females over reproduction (Arnqvist and Rowe, 2005; Chapman et al., 2003). Males usually gain more from polygyny than do females from polyandry (Bateman, 1948; Queller, 1997). The interests of the sexes also differ over care provisioning since parental care is costly, so that each parent often does better by shunting care provisioning to its mate (Houston et al., 2005; Lessells, 1999). Recent theoretical models (Gavrilets and Waxman, 2002; McNamara et al., 2000; Parker and Partridge, 1998) and experimental studies (Arnqvist and Rowe, 2002; Holland and Rice, 1999; Royle et al., 2002) suggest that sexual conflict has profound implications for the evolution of many morphological, behavioral, and life history traits. Our central thesis in this review is that sexual conflict is a powerful paradigm to understand breeding systems. We illustrate this thesis with shorebirds (Aves: Charadriiformes), which have played a central role in the development of breeding system theory. Shorebirds have long been noted for their extreme diversity in breeding systems. We follow Reynolds (1996) and define breeding system as a description of both social mating system (the form and duration of 1 Present address: Museum of Comparative Zoology, Harvard University, 26 Oxford Street, Cambridge, Massachusetts 02138, USA.

0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37006-X

279

Copyright 2007, Elsevier Inc. All rights reserved.

280

GAVIN H. THOMAS ET AL.

pair‐bonds) and parental care (the form and duration of care by each sex) of both sexes. Shorebird mating systems encompass social polygyny (polygyny hereafter; including lekking and territorial polygyny), social polyandry (polyandry hereafter; including classical polyandry and rapid multiclutching) and social monogamy (monogamy hereafter; including double‐ brooding and lifelong monogamy). This variation has attracted intensive research since the late 1960s (reviewed by Erckmann, 1983; Pitelka et al., 1974; Oring, 1986; Sze´kely et al., 2006). Initially, researchers focused on describing the mating patterns of individual populations or species, and putting this into an ecological context. Later research adopted experimental manipulations, genetic markers, and phylogenetic comparative methods. There is still controversy over explanations for why some taxa are socially monogamous with biparental care, whereas others are polygynous or polyandrous with uniparental care (Andersson, 1994, 2005; Bennett and Owens, 2002; Ligon, 1999). While the concept of sexual conflict is not new (Bateman, 1948; Davies, 1992; Lessells, 1999; Parker, 1979; Trivers, 1972), the last decade has seen an upsurge of interest in sexual conflict as a driver of evolutionary change, including impacts on breeding systems (Tregenza et al., 2006). This has led to a burst of theoretical, experimental, and phylogenetic studies (see references above). Although it is increasingly clear that some of the conflicts are played out at the genomic level (e.g., genomic imprinting, Haig, 2004; selfish genetic elements, Burt and Trivers, 2006), these topics have not been investigated in birds. Thus, shorebirds (or in birds), thus as a necessity we are focusing on nongenomic issues. Our main objectives are to review both how sexual conflict theory can be used to advance our understanding of the evolution of shorebird breeding systems, and how studies of shorebirds are advancing our understanding of sexual conflict theory. We begin by summarizing sexual conflict theory (Section II). We then review the diversity of shorebirds, focusing on the distribution of mating systems and parental care (Section III). In Section IV, we discuss the evidence for conflicts over mating optima in shorebirds. In particular, we highlight field studies of behaviors in shorebirds that enable each sex to exploit, or coerce, the other and consider the evidence for conflict over reproductive rate and clutch size. We also discuss the links between sexual conflict and sexual size dimorphism (SSD). We then consider the role of sexual conflict over parental care in driving breeding system evolution (Section V). We argue that mating systems and parental care both influence, and are influenced by, mating opportunities and are involved in a coevolutionary feedback loop (Section VI). Finally, we highlight possible avenues for future research (Section VII), and provide the

PARENTAL CONFLICT IN SHOREBIRDS

281

shorebird data in the Appendix A that were used in our comparative analyses. We anticipate that access to these data will stimulate further research on evolutionary ecology of shorebirds.

II. SEXUAL CONFLICT THEORY Males and females rarely have identical evolutionary interests over reproduction (Arnqvist and Rowe, 2005; Chapman et al., 2003; Wedell et al., 2006). Since reproduction is costly, conflict between the parents will arise because each parent will be selected to exploit the other (Lessells, 1999). Thus, except in species that breed once and then die (or have lifelong obligate monogamy), at least some form of conflict is inevitable (Rice, 2000; Wedell et al., 2006). Precise definitions of sexual conflict are controversial and somewhat ambiguous (Arnqvist and Rowe, 2005; Tregenza et al., 2006). In the context of breeding system evolution, we use the division of pre‐ and postzygotic conflicts introduced by Parker et al. (2002). Prezygotic conflict arises from different mating and fertilization optima for males and females, whereas postzygotic conflict involves differences between the sexes in care provisioning. Many authors also distinguish between intra‐ and interlocus conflicts (Arnqvist and Rowe, 2005; Chapman et al., 2003; Wedell et al., 2006). Intralocus conflict arises where the same allele has different effects on the fitness of the two sexes, thus selection favors different values for a single phenotypic trait in males and females (Arnqvist and Rowe, 2005; Fisher, 1930; Lande, 1980; Wedell et al., 2006). In interlocus conflict the optimal outcome of a male–female interaction differs for the average male and the average female. For the vast majority of organisms we do not know whether mating and parental conflicts are driven by intra‐ or interlocus conflicts, therefore, these issues are not explored further here. A. PREZYGOTIC CONFLICT Bateman’s (1948) classic experiments on Drosophila demonstrated that the benefits of multiple mating are not always equal for males and females: males increased their reproductive success by increasing the number of mates but females did not. In general, males stand to gain more by multiple mating than females (Parker, 1979, 2006; Queller, 1997). Indeed, females may be harmed physically if they exceed their mating optima (Arnqvist and Rowe, 2005). For example, seminal fluid of male Drosophila melanogaster reduces both the competitive ability of sperm from other males and female survival due to toxic side effects (Chapman et al., 1995; Rice, 1996).

282

GAVIN H. THOMAS ET AL.

The conflict over mating optima can drive rapid evolutionary change. According to the ‘‘chase‐away’’ model, preexisting sensory bias (Basolo, 1990, 1995; Ryan and Rand, 1998) in females selects for some form of display trait that enhances male attractiveness (Holland and Rice, 1998). The female response to the most attractive males is to mate in a suboptimal manner, for example by increasing their mating rate. Because of the deleterious effects of mating on female fitness, selection should favor resistance to male attractiveness by increasing the threshold for attractiveness. Selection then favors further exaggeration of the display trait above the new threshold. The cycle of male persistence and female resistance is known as sexually antagonistic coevolution (SAC). Although the chase‐away hypothesis refers directly to female choice over a display trait, it has also been suggested that SAC can result in the evolution of male coercion and female avoidance through both behavioral and morphological adaptations in both sexes. Examples of coercion associated with SAC include the evolution of structures for grasping (in males) and avoiding grasping (in females) in water striders (Arnqvist and Rowe, 2002), diving to resist males in some diving beetles (Dytiscidae; Bergsten et al., 2001), and traumatic insemination in bedbugs (Schuh and Stys, 1991). The burgeoning evidence for SAC suggests that it is widespread in nature (see Arnqvist and Rowe, 2005 for some fascinating cases of SAC).

B. POSTZYGOTIC CONFLICT Postzygotic sexual conflict is an emergent property of the costs of providing parental care against the benefits of deserting. Parental care is beneficial for the survival and development of offspring, but it is costly for the parents because it demands time and energy, and it reduces the chances of exploiting additional mating opportunities (Balshine et al., 2002; Clutton‐Brock, 1991). Each sex is therefore selected to shift the burden of care to its mate (Houston et al., 2005; Lessells, 1999). Sexual conflict over care can be resolved by one parent abandoning the brood to its mate, or by a truce whereby both parents remain with their brood and cooperate to raise the young (Chapman et al., 2003; Houston and Davies, 1985; Parker et al., 2002). A more extreme form of resolution may arise where one parent deserts and the other parent is at a disadvantage in rearing the young by itself, so ultimately, both parents abandon, and thus doom, the brood (e.g., Eurasian penduline tit Remiz pendulinus, Franz, 1991; Szentirmai et al., 2005). Note, however, that resolution in this sense does not mean that there is no longer a conflict, rather that the phenotypic expression of conflict is in some way a compromise in the fitness optima of the two sexes.

PARENTAL CONFLICT IN SHOREBIRDS

283

The outcomes of sexual conflict over care have important implications for breeding system evolution. Recent studies suggest that parental care and social mating systems both influence, and are influenced by, mating opportunities such that mating systems and parental care are involved in a conflict driven by a negative feedback relationship (Section VI; Alonzo and Warner, 2000; Kokko and Jennions, 2003; Sze´kely et al., 2000a). III. WHY STUDY SHOREBIRDS? Shorebirds have a long history of interest among evolutionary biologists and ornithologists alike (Hockey et al., 1995; Kam et al., 2004; Pringle, 1987). Recent phylogenetic analyses (Fig. 1; Baker et al., 2007; Fain and Houde, 2004; Paton and Baker, 2006; Paton et al., 2003 Thomas et al., 2004a,b) show that shorebirds have evolved along three main lineages consisting of approximately 350 species found on all continents: Scolopaci (sandpipers, jacanas, painted‐snipes, seedsnipes, and plains‐wanderer), Lari (coursers and pratincoles, gulls, alcids, terns, skimmers, and skuas), and Charadrii (plovers, oystercatchers, stilts and avocets, sheathbills, and magellanic plover). There is also compelling molecular evidence that buttonquails (Turnicidae), which have previously been included in either the Gruiformes, the Galliformes, or as their own order (Turniciformes; Monroe and Sibley, 1993), are most likely basal Lari (Baker et al., 2007; Fain and Houde, 2004; Paton and Baker, 2006; Paton et al., 2003). Across these three clades, shorebirds display exceptional diversity of life histories, ecology, and breeding systems (Erckmann, 1983; Oring, 1986; Pitelka et al., 1974; Sze´kely and Reynolds, 1995). A. SHOREBIRD BREEDING SYSTEMS The diversity of mating systems and parental care strategies in shorebirds has made these birds a popular group for studies of sexual selection. Indeed, Darwin (1871) illustrated his theory of sexual selection by reference to sex‐ role reversal in greater painted‐snipe Rostratula benghalensis and to the lekking behavior of ruff Philomachus pugnax: The polygamous ruff (Machetes pugnax) is notorious for his extreme pugnacity; and in the spring, the males, which are considerably larger than the females, congregate day after day at a particular spot, where the females propose to lay their eggs.

In shorebirds, as in most animals, mating systems tend to match closely with parental care. Polygynous species are defined as those in which males form multiple pair‐bonds within a breeding season, or have sequential

Burhinidae

Chionidae

Pluvianellidae

Charadriidae

Recurvirostridae

Haematopodidae

Scolopacidae

Thinocoridae

Pedionomidae

Rostratulidae

Jacanidae

Turnicidae

Glareolidae

Alcidae

Stercorariidae

Laridae

GAVIN H. THOMAS ET AL.

Sternidae

Rynchopidae

284

Fig. 1. A phylogenetic tree of shorebird families based on Ericson et al. (2003), Paton et al. (2003), and Paton and Baker (2006). Note that the monotypic family Dromadidae is missing because it has not been included in any molecular phylogenetic study to date. The differential shading illustrates the three major clades: light gray—Charadrii; dark gray—Scolopaci; and black—Lari. This is a hypothesis, undergoing perpetual improvement. Photographs (left to right): dolphin gull Larus scoresbii (Tama´s Sze´kely); African jacana (Aron Sze´kely); and Magellanic oystercatcher Haematopus leucopodus (Tama´s Sze´kely).

copulations while displaying on leks with multiple females. In monogamous species one male and one female form a pair‐bond that lasts the duration of at least one breeding season. In polyandrous species, females form pair‐ bonds with more than one male in a breeding season. Shorebird mating systems range from full polygyny (e.g., territorial polygyny: pectoral sandpiper Calidris melanotos, northern lapwing Vanellus vanellus; lekking: ruff, buff‐breasted sandpiper Tryngites subruficollis), through monogamy (oystercatchers, many plovers, gulls, alcids), to classical polyandry (e.g., phalaropes Phalaropus spp., African jacana Actophilornis africanus). Parental

PARENTAL CONFLICT IN SHOREBIRDS

285

care associated with these mating systems ranges from female‐only care (e.g., jack snipe Lymnocryptes minimus, sharp‐tailed sandpiper Calidris acuminata) to male‐only care (e.g., plains‐wanderer Pedionomus torquatus, Eurasian dotterel Eudromias morinellus). The diversity of mating systems in the three major clades of shorebirds is shown in Fig. 2. Note that in socially monogamous species, the duration of care provided by the sexes may be fully biparental, but is often biased toward predominantly male care (e.g., dunlin Calidris alpina, semipalmated plover Charadrius semipalmatus). Shorebirds are frequently cited as the most appropriate group for studying polyandry and male‐only care. Overall, the proportion of polyandrous shorebirds is marginally higher than that of birds as a whole (around 5% in shorebirds whereas probably less than 5% of all bird species; Bennett and Owens, 2002). Yet, among the clade that includes sandpipers and allies, 10–15% of known species are polyandrous (Fig. 2).

Fig. 2. The distribution of social mating system in major shorebird clades, including 11 polygynous species (female care), 199 monogamous species (biparental care), and 17 polyandrous species (male care). Note that data are not available for some species including the family Turnicidae which are not included.

286

GAVIN H. THOMAS ET AL.

The Lari clade, to which the Turnicidae evidently belong, is normally regarded as being conservative with regard to breeding system since most species are typically monogamous with biparental care. The addition of the Turnicidae to this clade is intriguing since at least one species of buttonquails (Turnix silvatica) is likely polyandrous and the chicks are raised by the male only (Cramp and Simmons, 1980). The distribution of polyandry in shorebirds is particularly striking when one considers that polyandrous species occupy a wide range of habitats around the world (Erckmann, 1983). There are polyandrous shorebirds from the tropics (e.g., pheasant‐tailed jacana Hydrophasianus chirurgus, bronze‐winged jacana Metopidius indicus) to northern temperate zones (spotted sandpiper Tringa macularia, Wilson’s phalarope Steganopus tricolor) to the Arctic (e.g., red phalarope Phalaropus fulicaria, red‐necked phalarope Phalaropus lobatus). Polyandrous species may breed in lowland grasslands (e.g., mountain plover Charadrius montanus), alpine heathland (e.g., Eurasian dotterel), or in tropical marshes (greater painted‐snipe, African jacana). This raises the question, why do shorebirds, more than other birds, show such propensity for divergence in breeding systems? In Section IV, we argue that this divergence is driven by the high potential for conflict among species in which the need for biparental care is relaxed. Rates of multiple paternity (MP) and extra‐pair paternity (EPP) are also variable in shorebirds. Among socially monogamous species rates of MP are low: 2.9
2.8% (SD) of chicks and 4.6
3.4% of broods were extra pair [n ¼ 5 and 11 species, using data from Griffith et al. (2002) and Ku¨pper et al. (2004)]. These frequencies are lower than reported from socially monogamous birds other than shorebirds (chicks: 12.0
13.5%, n ¼ 89 species, Mann‐Whitney U test, z ¼ 2.296, p ¼ 0.022; broods: 20.4
21.8%, n ¼ 83 species; z ¼ 2.272, p ¼ 0.023). In contrast, rates of MP are higher among polyandrous shorebirds (Table I; Emlen et al., 1998; Oring et al., 1992), and higher still among lekking species such as the ruff (Lank et al., 2002) and buff‐breasted sandpiper (Lanctot et al., 1997). The different duration of male and female care (Fig. 3) and the variety of mating systems suggest the costs and benefits to desertion are not identical for the sexes in different species. Thus, shorebirds are likely to express a wide continuum of sexual conflict over mating and care. Taken together, the immense natural variation in shorebirds with respect to mating and parental behavior makes them an ideal group to investigate the role of sexual conflict in breeding system evolution.

TABLE I MULTIPLE PATERNITY IN SHOREBIRDSa

287

Species

Multiple paternity (broods, n) (%)

Multiple paternity (chicks, n) (%)

MF

CA

FC

ST

SS

References

Spotted sandpiper: Tringa macularia Wattled jacana: Jacana jacana Wilson’s phalarope: Steganopus tricolor Red phalarope: Phalaropus fulicarius Red‐necked phalarope: Phalaropus lobatus Eurasian dotterel: Charadrius morinellus Kentish plover: Charadrius alexandrinus Ringed plover: Charadrius hiaticula Semipalmated plover: Charadrius semipalmatus Western sandpiper: Calidris mauri Purple sandpiper: Calidris maritima Common; sandpiper: Tringa hypoleucos Common; sandpiper: Tringa hypoleucos Eurasian Oystercatcher: Haematopus ostralegus Great snipe: Gallinago media Ruff: Philomachus pugnax Ruff: Philomachus pugnax Buff‐breasted sandpiper: Tryngites subruficollis Brown skua: Catharacta Antarctica South Polar skua: Catharacta maccormicki Western gull: Larus occidentalis Common gull: Larus canus Common murre: Uria aalge Common tern: Sterna hirundo Little auk: Alle alle

11.1 (1/9) 24.3 (18/74) 0 (0/17) 8.6 (6/70) 6 (4/63) 9.1 (2/22) 3.4 (3/89) 0 (0/21) 4.2 (1/24) 8 (3/40) 3.7 (1/27) 6.7 (1/15) 18.5 (5/27) 3.8 (1/26) 28.6 (2/7) 50 (17/34) 51.5 (34/66) 40.4 (19/47) 0 (0/16) 7.7 (2/13) 0 (0/22) 8.3 (2/24) 7.8 (6/77) 0 (0/10) 0 (0/26)

2.9 (1/34) 1.3 (24/235) 0 (0/51) 33.3 (6/18) 1.7 (4/226) 4.6 (2/44) 1.3 (3/229) 0 (0/57) 4.7 (4/85) 5 (5/98) 1.2 (1/82) 1.8 (1/53) 15.7 (13/83) 1.5 (1/65) – –

N N N – Y Y – N Y Y N – – – – –

N N N Y Y Y – N Y Y Y – – – – –

– Y N – Y N – N Y – N – – – – –

Y – – – Y Y – – – – N – Y – – –

Y N Y Y Y – – – – – – – – – – Y

– 0 (0/45) 7 (2/14) 0 (0/33) 3.6 (2/55) 7.8 (6/77) – 0 (0/26)

– – – – – – – –

– – – – – – – –

– – – – – – – –

– – – – – – – –

– – – – – – – –

Oring et al., 1992 Emlen et al., 1998 Delehanty et al., 1998 Dale et al., 1999 Schamel et al., 2004 Owens et al., 1995 Ku¨pper et al., 2004 Wallander et al., 2001 Zharikov and Nol 2000 Blomqvist et al., 2002b Pierce and Lifjeld 1998 Blomqvist et al., 2002a Mee et al., 2004 Heg et al., 1993 Cited in Lank et al., 2002 Lank et al., 2002 Thuman, 2003 Lanctot et al., 1997 Millar et al., 1994 Millar et al., 1997 Gilbert et al., 1998 Bukacinska et al., 1998 Birkhead et al., 2001 Griggio et al., 2004 Lifjeld et al., 2005

a

Paternity assurance behaviors include: mate following (MF); close association with female (CA); frequent copulation (FC); strategic timing of copulation (ST); and sperm storage (SS). Studies in which observations of paternity assurance behaviors were recorded (Y); studies in which behaviors were not observed (N); and no information (–).

288

GAVIN H. THOMAS ET AL.

Fig. 3. Distribution of parental care in shorebirds (updated from Sze´kely and Reynolds, 1995). ‘‘Male biased’’ means that the male contributes all care either until the chick fledge (‘‘All care’’), or the majority of care with females deserting before hatching (‘‘Eggs’’), or before fledging (‘‘Chicks’’). The same terminology applies to ‘‘Female‐biased’’ care. In biparental taxa both parents provide care until the chicks fledge.

IV. CONFLICT OVER MATING A. CONFLICT OVER MATING OPTIMA 1. Multiple Mating Bateman’s classic experiments on Drosophila suggested that males should gain more by multiple mating than females (Bateman, 1948). However, this theory is no longer taken for granted; Sutherland (1985) showed that higher variance in mating success in males than females may occur as a result of random mating rather than any reflection of the relative investment of each sex. Furthermore, it now seems clear that both males and females can gain from multiple mating (e.g., the polyandrous pipefish Syngnathus typhle; Jones et al., 2000) and promiscuity (Jennions, 1997; Tregenza and Hosken, 2005). In shorebirds, female gains from multiple mating are perhaps most obvious in polyandrous species, but there is also evidence of females holding the upper hand in conflict over mating in lekking species. The highest rates of multiple paternity among shorebirds are found in

PARENTAL CONFLICT IN SHOREBIRDS

289

lekking species: ruff, buff‐breasted sandpiper, and great snipe. Gowaty (1996) proposed that where females are largely independent of males, rates of MP are expected to be high (the ‘‘constrained female’’ hypothesis). Lekking shorebirds fit well into Gowaty’s hypothesis since males do not contribute to parental care and consequently have little leverage over female choice. In these lekking shorebirds, males do attempt to disrupt copulations, although rates vary across species. The ruff has very low rates of copulation disruption by males (2.7%; Widemo, 1997) despite showing the highest recorded degree of multiple mating (Table I). Rates of copulation disruption are highest in the buff‐breasted sandpiper (43.2%; Lanctot et al., 1998) and intermediate in the great snipe (28.6%; Saether et al., 1999). Disruption attempts by males may simply be due to male–male competition, but it is also likely that some are by males trying to prevent females with whom they have already mated copulating with other males. Among sex‐role reversed shorebirds, it is the female that gains most from nesting with multiple partners. For instance, in the polyandrous spotted sandpiper female reproductive success increased with the number of mates, although polyandrous females also tended to be older and more experienced than monogamous ones. Males were limited to only one clutch per year, regardless of age or experience although some males increased reproductive success through sperm storage and subsequent fertilization of eggs in later clutches (Oring and Lank, 1986). Similarly, red‐necked phalarope females may have successful clutches with two consecutive males per year whereas males never raise young from more than one clutch (Reynolds, 1987). This dichotomy sets the stage for strong sexual conflict over mating opportunities. For example, female red‐ necked phalaropes often approach incubating males, and are driven off with conspicuous squeaking and fluttering displays called ‘‘driving flights’’ (Reynolds, 1987). These displays (also observed in red phalaropes; Kistchinski, 1975) could not be explained by an alternative explanation of prevention of infanticide by females, which is considered in the context of sexual conflict in Section IV.A.3. In both spotted sandpipers and red‐necked phalaropes, the operational sex ratio (OSR; ratio of sexually active females to males; Emlen and Oring, 1977) becomes more female biased in the population as males take up incubation and thereby drop out of the mating pool (Lank et al., 1985; Reynolds, 1987). Lank et al. (1985) showed that within a breeding season the reproductive rate of females declined as the sex ratio became more female biased, and suggested that this was due to the increase competition between females for males. It has been noted that the frequency of polyandry in small populations may be highest in locations and years when there are male‐biased sex ratios (e.g., northern jacanas Jacana spinosa,

290

GAVIN H. THOMAS ET AL.

Jenni and Collier, 1972; red phalaropes, Schamel and Tracy, 1977; red‐ necked phalaropes, Reynolds, 1987). We can therefore expect the extent of sexual conflict over multiple pair‐bonds to show similar variation within species, although we acknowledge that the evidence linking OSRs to mating systems comes primarily from polyandrous species. The availability of mates appears to be important in determining the direction and strength of sexual conflict over mating, as the examples discussed above illustrate, and it is therefore particularly relevant to our understanding the evolution of polyandry. This goes against a previous hypothesis proposed for shorebirds that in species with fixed clutch sizes, the only way for females to increase their reproductive output is to seek multiple mates (Alcock, 2005; Davies, 1991; Erckmann, 1983; Oring, 1986). The latter hypothesis also ignores an obvious conflict between the sexes over which parent should desert the brood—which is the central focus of this review. The key assumption of the clutch‐size hypothesis is that shorebirds have a fixed upper limit on the number of eggs that can be laid and incubated. This upper limit is generally considered to be four eggs (Maclean, 1972), although across species clutch size ranges from one to four (Appendix A), and they even vary within species. In our opinion, the clutch‐size constraint hypothesis lacks generality. First, several bird taxa with ‘‘fixed’’ (i.e., less variable) clutch sizes do not exhibit sex‐role reversal (e.g., bustards, albatrosses, and many seabirds). Furthermore, polyandry is absent (or rare) among Lari and Charadrii, even though these taxa lay clutches that are as ‘‘fixed’’ as those of sandpipers Scolopaci. Second, the argument focuses on females only, assuming that males are widely available and they will ‘‘automatically’’ assume the full care of eggs. Given that the males in general can gain more by being polygamous than females (Queller, 1997), it is not trivial to explain why the males ‘‘sacrifice’’ themselves for the sake of their mate. Third, polyandry with multiple mates occur in birds that have variable clutch sizes, such as tinamous, coucals, the penduline tit, and several passerines (Andersson, 2005; Owens, 2002; Szentirmai et al., 2005). Taken together, the fixed clutch size may contribute to the evolution of mate change and polyandry, because females in species with fixed clutch size may gain more from multiple mates than females in species with variable clutch sizes, but it cannot fully explain polyandry. 2. Multiple and Extra‐Pair Paternity Westneat and Stewart (2003) proposed that extra‐pair copulations (EPCs) result from sexual conflict in a three‐way interaction between the female, pair male, and extra‐pair male. It is in the best interests of both sexes to produce offspring with a high chance of survival and future reproductive success, but males gain an additional benefit from fertilizing as

PARENTAL CONFLICT IN SHOREBIRDS

291

many eggs as possible. Conflict arises if the females’ criteria are not fully met by mating with a particular male (Schamel et al., 2004; Westneat and Stewart, 2003). Thus, females may often benefit from mating with an extra‐ pair male, for example through guarding against infertility of the social mate (Wetton and Parkin, 1991), increased genetic diversity of the offspring (Blomqvist et al., 2002a; Westneat et al., 1990; Williams, 1975), enhanced genetic compatibility between the parents (Kempenaers et al., 1999; Tregenza and Wedell, 2000), obtaining good genes (Birkhead and Møller, 1992; Hamilton, 1990; Westneat et al., 1990), or direct benefits to the offspring (Burke et al., 1989; Colwell and Oring, 1989). However, extra‐ pair mating itself results in conflict between the pair due to selection on pair males to reduce the risk of cuckoldry. Indeed, we suggest that sexual conflict over extra‐pair mates is consistent with a form of chase‐ away selection: first male paternity assurance behaviors evolve (e.g., high rates of within‐pair copulations), then females counteract with sperm storage or cryptic mate choice, males escalate the conflict by further paternity assurance behaviors (e.g., delayed copulation) and the cycle continues. There is ample scope for conflict over multiple and extra‐pair mating in shorebirds, judging from the numerous reports of MP (Table I). The highest levels of MP occur in lekking species such as the ruff (50% of broods with at least one extra‐pair chick) and the buff‐breasted sandpiper (40% of broods). In general, MP appears to incur greater costs for pair males in species with polygamous mating systems. For example, in the wattled jacana Jacana jacana, MP was 0% in monandrous pairs, whereas in polyandrous groups extra‐pair males sired 41% of broods and 17% of chicks (Emlen et al., 1998). In bronze‐winged jacanas, females copulated with all males in their harem before laying their first clutch, leading to a high risk of cuckoldry for all males within the group (Butchart, 1999). In the wattled jacana, the main source of MP was from co‐mates with whom females copulated at exceptionally high rates (1.3 per hour; Emlen et al., 1998). High copulation rates have also been recorded in the polyandrous red phalarope (Schamel et al., 2004), and in the socially monogamous semipalmated plover (Zharikov and Nol, 2000). In some species, notably the red‐necked phalarope, high copulation rates occur within pairs rather than with extra‐pair mates: this is presumably a form of paternity assurance (Reynolds, 1987). The risk of being cuckolded is particularly high for males breeding late in the season if females store sperm from earlier copulations. For example, male spotted sandpipers that pair early in the breeding season cuckolded their mate’s subsequent mates via stored sperm (Oring et al., 1992). Females of all three species of phalarope are thought to use sperm stored from earlier mates to fertilize eggs late in the breeding season (Dale et al., 1999;

292

GAVIN H. THOMAS ET AL.

Delehanty et al., 1998; Schamel et al., 2004; see Table I). This sets up a strong sexual conflict between females and mates late in the breeding season and favors male choice against females that have already mated. This behavior was noted in the red‐necked phalarope by Whitfield (1990) but was not detected in the same species by Schamel et al. (2004). It is not yet known how males may distinguish between already mated and unmated females. Nonetheless, if males can detect females that have already mated, then females should be selected to conceal their mating status. This response by males to sexual conflict could place a strong cap on the frequency of polyandry in this species. Males can minimize sexual conflict over paternity by strategically timing their copulation attempts. Owens et al. (1995) suggested that in the Eurasian dotterel, males only attempt to copulate with females after having been paired for several days. By strategically delaying copulation, males can gauge whether the clutch is likely to have been sired by a previous mate. If the clutch is laid too early then the male may desert because the clutch is unlikely to be his. The costs of desertion may be lower for males than females, as female‐biased OSRs in this species imply that males will be able to find new mates more readily than can females. Such tactics suggest that the benefits of EPCs for females may be offset by the possible loss of male care (Delehanty et al., 1998). Females, however, may be able to reduce the risk of male desertion by maintaining a short distance from their mate, thereby providing greater paternity assurance (Dale et al., 1999; Owens et al., 1995; Schamel et al., 2004). Strategic timing of copulations has also been reported in the spotted sandpiper, where male copulation attempts coincide with the female fertile period (Oring et al., 1994). Similarly, the copulation rate of paired males in the common sandpiper increases just before and during egg laying (Mee et al., 2004). Individual‐based focus on sexual conflict over EPP can be scaled up to species‐specific differences in morphology. Both sperm length (Johnson and Briskie, 1999) and testis size (Cartar, 1985) are greater in nonmonogamous species than in monogamous ones, even after controlling for allometry with body size. This is consistent with the observation that rates of MP tend be highest in polygamous species (Table I), implying intense sperm competition. Indeed, the largest testes (Cartar, 1985; Johnson and Briskie, 1999) and longest sperm (Johnson and Briskie, 1999) are found in the ruff, a species that breeds on leks, and is also notable for high rates of EPP (50% of broods, Lank et al., 2002; 51.5% of broods, Thuman, 2003) and for sperm storage by females. It is interesting to compare how the interpretations of these results differed over the 14 years that spanned the 2 studies. Cartar (1985) suggested that larger testes had evolved in response to selection on males in polygynous species to be able to avoid running out of sperm. In the 1980s,

PARENTAL CONFLICT IN SHOREBIRDS

293

few people were thinking about sperm competition in birds, and Cartar’s pioneering comparative study was therefore not framed in terms of sexual conflict. With the advent of DNA paternity analyses through the 1990s, it became clear that extra‐pair fertilizations could be common, leading to the emphasis on sperm competition that was evident in Johnson and Briskie’s study (Johnson and Briskie, 1999). While shorebirds have been supportive models for advancing general theories of conflicts over EPP, several issues remain unresolved. While it seems possible that sperm storage could be widespread in shorebirds, there is little hard evidence about this, and we do not know the extent to which stored sperm may be swamped by subsequent mates or used preferentially by females. Sperm storage is clearly an important part of the three‐player interaction identified by Westneat and Stewart (2003). Furthermore, data are needed on the costs and benefits of EPCs for females. Do females use extra‐ pair mating as a means for assessing and acquiring future mates, as has been suggested in the spotted sandpiper (Colwell and Oring, 1989)? Alternatively, if females form pair‐bonds with relatives they may avoid the cost of inbreeding by copulating with unrelated males (Blomqvist et al., 2002a). Why pair‐bonds between close relatives should occur in the first place is contentious, although it may simply reflect a lack of alternatives (Blomqvist et al., 2002a). Intensive field studies including paternity analysis and behavioral observations of the kind carried out by Schamel et al. (2004) would help resolve these issues. 3. Infanticide Infanticide is an extreme form of sexual conflict that was first highlighted among polygynous mammals in which males are behaviorally dominant over females (e.g., rodents, primates, and carnivores; Hrdy, 1979; Packer and Pusey, 1984). Males in these species kill unrelated offspring, leading to faster reproduction by their new mates. Emlen et al. (1989) hypothesized that in sex‐role reversed jacanas the females, which are larger and behaviorally dominant over males, may commit infanticide for similar reasons (Stephens, 1982). To test this hypothesis, Emlen et al. experimentally removed harem‐holding female wattled jacanas, and followed the consequences. As predicted, new females quickly overtook the vacant territories, they killed or evicted the broods of usurped males, and solicited copulations within two days of driving off or killing the males’ offspring. In three of four broods, the males intervened by attacking the female or through distraction behavior. In one brood the male did nothing and allowed the female to drive the brood away. This brood was the oldest of those in the study (35–40 days old compared to a maximum of 30 days in the other 3 broods), suggesting that those chicks may have had some chance of survival. This study convincingly showed female‐driven infanticide in an experimental

294

GAVIN H. THOMAS ET AL.

situation. The frequency of infanticide in natural populations and its implications for reproductive success of males and females, however, remain to be quantified. It would be interesting to perform tests analogous to those of Emlen et al. (1989) but with a male‐territorial polygynous species such as the pectoral sandpiper, to see whether males are infanticidal in species with this breeding system. B. MATE CHOICE AND SEXUAL SIZE DIMORPHISM Shorebird species span nearly the entire range of sexual size dimorphism (SSD) shown by the world’s birds (Sze´kely et al., 2004). This includes extremes in both male‐ and female‐biased dimorphism. Sexual dimorphism can lead to intralocus sexual conflict since one or the other sex may not be at its optimum body size (Arnqvist and Rowe, 2005; Fairbairn, 1997; Wedell et al., 2006). Thus, the extreme range of SSD in shorebirds implies a variety of conflicts over body size optima. These are difficult to discern, because body size is a complex trait controlled by multiple loci and linked to fitness in many ways. Here, we focus on the best‐understood links between SSD and sexual selection in each sex. Recent evidence, notably from a series of comparative studies, suggests that sexual selection is the main driver of SSD in shorebirds, including species in which males are the most competitive sex but are smaller than females (Figuerola, 1999; Lindenfors et al., 2003; Sze´kely et al., 2000b, 2004). The hypothesis that SSD results from sexual selection was proposed by Payne (1984) and Jehl and Murray (1986). They suggested female‐biased SSD evolves as a response to selection for male display agility. The size‐agility‐ sexual dimorphism hypothesis requires first that sexual selection favors more acrobatic flight displays, second that small male size enhances flight agility, and third that this results in the evolution of males that are smaller than females in species with acrobatic displays. Grønstøl (1996) confirmed the first assumption that display agility is important for mating success. He analyzed video recordings of song‐flights in male northern lapwings and found that an energetically demanding component of the flight display, the roll angle in the alternating flight, significantly predicted both the number of mates attained by the male and the clutch laying date of his mate. Lanctot et al. (2000) suggested that song‐ flights in the western sandpiper Calidris mauri are more likely used to attract mates than to defend territories (i.e., they are effectively display flights) since the frequency of song‐flights by males peaked during the pair formation period, rather than earlier when territories were established. The next step to confirming the size‐agility‐sexual dimorphism hypothesis was provided by Blomqvist et al. (1997), who showed that both display rate and the time spent displaying decreased with male size in the dunlin.

295

PARENTAL CONFLICT IN SHOREBIRDS

The final step to confirmation of the hypothesis has been provided by comparative studies among species. For example, in a study of socially monogamous shorebirds, Figuerola (1999) found that female‐biased SSD was associated with more acrobatic displays by males. Sze´kely et al. (2000b) showed that males were larger (relative to females) in polygynous taxa, compared with monogamous or polyandrous taxa, whereas male size was smaller (relative to female size) in polyandrous taxa. However, in taxa in which males perform acrobatic flight displays, males are smaller relative to females than in taxa with nonacrobatic or ground displays (Fig. 4). The interaction between social mating system, a proxy for mating competition, and selection for display agility has been argued to drive a pattern between body size and SSD, termed Rensch’s rule (Sze´kely et al., 2004). Rensch’s rule is an allometric trend in body

Male biased

0.12

0.08

3 Courtship agility Low Medium High

3 0.04 17

Female biased

0.00

−0.04

15 24 5

−0.08 3 −0.12

5 Low

Medium Intensity of sexual competition

High

Fig. 4. Sexual dimorphism in body mass [mean
SE log10(male mass)  log10(female mass)] in relation to the intensity of sexual competition and male agility. Intensity of sexual competition refers to competition between males and is based on the social mating system of each species (low ¼ polyandrous; medium ¼ monogamous; high ¼ polygynous). The numbers of species are given below (or above) each bar. From Sze´kely et al. (2004; copyright # 1993–2005 by The National Academy of Sciences of the United States of America, all rights reserved).

296

GAVIN H. THOMAS ET AL.

size in which the degree of SSD increases with body size when males are the larger sex but decreases with body size when females are the larger sex. Taken together, the evidence is consistent with the hypothesis that mate choice is an important factor in determining the direction and extent of SSD in shorebirds. However, one important caveat is that many shorebirds spend long periods away from the breeding grounds and we know little about how body size is affected by selection outside the breeding season (Sze´kely et al., 2000b). Future analyses would benefit from taking into account data on, for example, sex differences in resource use on the nonbreeding grounds and differential migration (Nebel, 2007). There is some evidence that the extent of SSD is mediated by both natural selection and sexual selection. In a comparative study of 71 shorebird species in which the males incubate the eggs, Lislevand and Thomas (2006) demonstrated that egg size increases as the degree of female‐biased size dimorphism decreases. Lislevand and Thomas (2006) imply that this pattern is driven primarily by increases in male size, since female size also increases with egg size in the same model. Large males could be more efficient incubators, resulting in selection for increased male size. This implies that selection on incubation efficiency operates in opposition to selection for display agility. An alternative explanation is that male body size is being dragged along by sexual selection that favors larger females. In polyandrous species, males may be close to their natural selection optima, and pull females toward a smaller size than they would achieve if sexual selection for large female size were unconstrained by genetic correlations. Lindenfors et al. (2003) refer to this as the chauvinistic rubber band hypothesis since it focuses on selection acting on males with the females dragged along through a genetic correlation. The authors expressed a degree of skepticism about their own hypothesis, since it is not clear why only females would be subject to this effect. Sexual selection and mate choice over traits other than display agility may also mediate the extent to which conflict of body size occurs. In particular, sexual conflict over body size may be minimized where sexual selection favors contrasting pigmentation in males with acrobatic flights. Graul (1973a) argued that male displays could be amplified by dark, or contrasting pigmentation. Bo´kony et al. (2003) found strong support for Graul’s hypothesis using 101 species of Charadrii. After controlling for phylogenetic nonindependence, they found that aerial flight displays were associated with both more extensive plumage melanization and increased melanin dichromatism. Similar relationships occur among bustards (Dale, 2006). These comparative studies suggest that plumage characteristics play an important role in mate choice or territorial displays and may consequently

PARENTAL CONFLICT IN SHOREBIRDS

297

influence body size evolution. Sexual selection on agility may be less intense in species with contrasting pigmentation because of the amplification effect of pigmentation contrast on display flights (Bo´kony et al., 2003; Graul, 1973a). Thus, based on Graul’s logic, we predict that in species where males perform acrobatic displays, the degree of female‐biased SSD will decrease with increases in the degree of contrast in pigmentation. This hypothesis might also be extended to traits other than pigmentation that could exaggerate the complexity of flight displays. To date only a few studies have sought to test explicitly the role of other traits such as color patterns and plumes in mate choice in shorebirds. For example, female golden plovers Pluvialis apricaria and Eurasian dotterels prefer males with brighter or more extensively melanized plumage (Edwards, 1982; Owens et al., 1994). Reynolds (1987) measured seven characteristics of plumage of male and female red‐necked phalaropes, and found evidence of assortative mating where large females mated with dull males. The interactions between different putative sexually selected traits warrant further investigation, particularly with respect to the mediating effects they may have on sexual conflict over body size.

V. PARENTAL CARE A. TUG‐of‐WAR OVER CARE Since it is in the best interests of each parent to have its partner pay the costs of raising the young (Houston et al., 2005; Lessells, 1999), the patterns of parental care at the population level might be considered as the resolution of a ‘tug‐of‐war’. In general, we would expect that offspring benefit most from biparental care. Erckmann (1981) used removal experiments to show that neither males nor females could (or would) raise a clutch alone in western sandpiper C. mauri, although males attended the clutch for longer than males. Lessells (1983) found consistent results in Kentish plover Charadrius alexandrinus by removing either the male or the female during incubation: single parents struggled to continue incubating the clutch, and males persisted incubation longer than females. More recent studies have shown that survival of Kentish plover chicks is higher in biparental families than uniparental families (Sze´kely and Cuthill, 1999; Sze´kely and Williams, 1995). The tug‐of‐war leaves its imprint in two ways. First, either parent can try to stop its partner from deserting by interfering with mate attraction or using direct aggression (Chapman et al., 2003). Second, a parent may increase its parental effort following desertion by its partner. Houston and Davies (1985) showed that parental cooperation in rearing young is

298

GAVIN H. THOMAS ET AL.

only stable if, when one parent reduces its parental effort, the other parent increases its effort to compensate. However, compensation is predicted to be incomplete (Houston et al., 2005). Behaviors used by parents to manipulate one another may involve sophisticated signals, or sheer force. Males of the polyandrous bronze‐winged jacana use a call, termed a ‘‘yell,’’ to attract females to mate (Butchart et al., 1999). Yelling for sex may signal male quality since the males that yell at the highest rate receive most copulations. Males with the highest yell rates were also subject to intrusions by other females. The territorial female’s response can be interpreted as an attempt to prevent other females from copulating with males in their territory. By preventing harem males from copulating with other females, the males are coerced into remaining with the territory holder and raising her young. In northern lapwings, which commonly are polygynous, females were aggressive toward males when presented with a female dummy near to their nest (Liker and Sze´kely, 1997). Liker and Sze´kely suggested that this female aggressive behavior is an attempt to monopolize paternal care by actively preventing the males from exploiting potential mating opportunities. Field studies and phylogenetic comparative analyses suggest that compensation for the loss of the partner occurs in shorebirds. Kosztola´nyi et al. (2003) experimentally removed one parent (the male or the female) from their nests in the Kentish plover, a species with natural variation in care, ranging from fully biparental to either sex deserting soon after hatching. When the female was removed, her mate increased its time spent incubating, but not sufficiently to compensate for the loss of the female. Females did not compensate for the loss of their mate. One interpretation is that males may normally withhold care whereas females operate at their investment maxima. Such behavior may coerce females into remaining with the brood, and also afford males the opportunity to desert more readily than their partner. However, the balance of power over coercion may vary between species since female magnificent frigatebirds Fregata magnificens nearly doubled their feeding rate after male desertion, presumably to compensate for the loss of care (Osorno and Sze´kely, 2004). The tug‐of‐war over care also manifests itself at the phylogenetic level. Using phylogenetically independent differences between related taxa, Reynolds and Sze´kely (1997) demonstrated that evolutionary increases in the duration of care in one sex were associated with evolutionary decreases in the duration of care by the other sex (Fig. 5). Further studies are required to understand how conflicts over care are resolved, at both the phylogenetic and individual level. First, most comparative studies have used fairly crude scored variables based on the duration of

PARENTAL CONFLICT IN SHOREBIRDS

299

Fig. 5. The relationship between the duration of female care and the duration of male care in shorebirds. The independent axis shows phylogenetically independent contrasts in the duration of female care. The dependent axis shows phylogenetically independent contrasts in the duration of male care (as in Reynolds and Sze´kely, 1997 but updated with recent informaton; see Appendix for details parental care scores). Fitted line is the major axis regression through the origin (slope ‐1.309, lower confidence interval ‐2.246, upper confidence interval ‐0.570).’’

care relative to the total fledging period, as an index of parental effort (Reynolds and Sze´kely, 1997; Sze´kely and Reynolds, 1995; Thomas and Sze´kely, 2005). Ideally, quantitative data on a range of parental care activities including incubation, brooding, antipredator behavior and (in some taxa) feeding would be incorporated into cross‐species studies. Second, at the individual and population level, it would be informative to quantify how the contributions of each sex to these different aspects of care influence both offspring survival and mortality of the parents. Third, experimental tests, including removal experiments in several shorebird species, are required to examine the response of parents to the removal of their mates. It is likely that these responses will not be linear, thus designing and executing experiments at sufficient resolution of parental manipulation (or handicapping, see Houston et al., 2005) will be challenging. B. CONSTRAINTS ON THE DURATION OF CARE Parents may benefit from avoiding (or at least minimizing) their contribution to care, if their partner is willing, or can be coerced into, taking on the workload. Yet biparental care is still the most common resolution to sexual

300

GAVIN H. THOMAS ET AL.

conflict over care in shorebirds (Fig. 3). Why do parents rarely succeed in fully exploiting their partners in most species? The most likely explanation is that the requirements of the progeny are high so that both parents are required to successfully raise the brood (Erckmann, 1981; Lack, 1968). This is particularly important if parents provision food to their young, and food availability is poor (Ens et al. 1992). In precocial species both parents may lose weight during brood rearing such as in the hooded plover Charadrius rubricollis, implying that the burden of care is high and must be shared (Weston and Elgar, 2005). The requirements of the offspring may include incubation, brooding, defense against predation and feeding. We envisage two main ways in which the demands of the young vary among species. First, demand may vary according to the developmental mode of the young, specifically their feeding ability. Second, there may be variation due to environmental and ecological conditions including temperature and predation risk. Some shorebird species have chicks that feed themselves within hours of hatching but are dependent on the parents for brooding and defense against predation (precocial young; Scolopaci, Charadrii except oystercatchers, thick‐knees, and sheathbills), whereas in other species the young additionally depend on the parents for feeding until they fledge (semiprecocial young; Lari, oystercatchers, thick‐knees, and sheathbills). In precocial species the parents are relieved from feeding the young. Consequently, uniparental care and social polygamy are expected to occur more frequently in precocial than semiprecocial species (Lack, 1968; Orians, 1969). An alternative view is that the evolution of uniparental care results in selection on chicks to be more independent and thus lead to precocial offspring, thus implying a reversal in the direction in cause and effect. However, developmental mode has changed only a handful of times across the shorebird phylogeny (i.e., it is not a labile trait). Thomas and Sze´kely (2005) found that evolutionary changes across the shorebird phylogenetic tree toward precocial young preceded evolutionary changes toward reduced parental care. This result is important for two reasons. First, it suggests that developmental mode is more likely to be the causative factor in the relationship with parental care. Second, it implies that the resolution of sexual conflict toward desertion is facilitated by the evolution of precocial young (i.e., that uniparental care is more likely to occur in precocial species). The latter result was corroborated by a subsequent analysis comparing rates of trait evolution in precocial and semiprecocial species, because among precocial species both parental care and mating systems diverged more rapidly relative to semiprecocial taxa (Thomas et al., 2006). Taken together, these phylogenetic analyses suggest that the feeding capability of the offspring has a major influence on whether one parent has the opportunity to desert. We note, however, that species that share the same developmental mode

PARENTAL CONFLICT IN SHOREBIRDS

301

could also be similar with respect to other ecological and life history traits, so that we cannot be certain of a cause and effect relationship between developmental mode and desertion. The main alternative hypothesis based on environmental harshness predicts that in extreme environments (e.g., very low or very high temperature regimes, high risk of predation), the need for parental care is high and sexual conflict can only be resolved through biparental care. However, despite the fact that many shorebirds inhabit extreme locations and that they do incur costs as a result, there is little evidence to corroborate the environmental harshness hypothesis. For example, the daily energy expenditure (DEE) of incubating shorebirds in the Canadian and Eurasian Arctic (little stint Calidris minuta, white‐rumped sandpiper Calidris fuscicollis, dunlin, ringed plover Charadrius hiaticula, sanderling Calidris alba, purple sandpiper Calidris maritima, ruddy turnstone Arenaria interpres, red knot) is about 50% higher than in temperate breeding birds when measured using doubly labeled water techniques (Piersma and Morrison, 1994; Piersma et al., 2003). One might expect uniparental care to be constrained by such high metabolic requirements of adults. Yet there is nothing to suggest that these costs constrain desertion as a resolution to sexual conflict, since in several Arctic‐breeding species care is uniparental (e.g., buff‐breasted sandpiper, red and red‐necked phalaropes, and pectoral sandpiper). The Kentish plover nests in areas where ground temperatures are high enough to cook the eggs (50  C in the shade, Amat and Masero, 2004, or beyond 60  C in exposed sites Kosztola´nyi, A., personal communication) if left unattended for only a few minutes. Here we might expect both parents to be required for incubation and brood care although no such constraint is obvious, because either the male or the female may desert shortly after the eggs hatch (Sze´kely et al., 1999). Studies of predation by gull‐billed terns Gelochelidon nilotica on Kentish plover chicks showed that the duration of care increased in years when predator numbers where high (Amat et al., 1999). While this implies that the resolution of sexual conflict could potentially be restricted in favor of biparental care as a result of predation risk, it also indicates that the environmental harshness hypothesis is rather idiosyncratic and does not generalize across species. An additional alternative is that low food availability may limit the potential for the evolution of uniparental care. For example, biparental (monogamous) shorebirds lay their eggs earlier than uniparental (polyandrous) species in the Arctic (Whitfield and Tomkovich, 1996). Whitfield and Tomkovich suggested that the availability of food early in the breeding season constrains the outcome of conflict over care: only biparental species are evidently able to exploit the relatively low abundance of food early in

302

GAVIN H. THOMAS ET AL.

the breeding season. If variation in food abundance is unrelated to latitude or climate, then this may partially explain the apparent lack of widespread support for general environmental constraints on incubation and care. C. DESERTION AND MATING OPPORTUNITIES Identifying whether desertion of the brood and subsequent remating is possible is only part of the story of sexual conflict over care. The second major challenge is in predicting which sex should desert. Two major hypotheses have been proposed. The differential cost of parental care hypothesis argues that the deserting parent should be the sex that incurs the largest direct costs of care (e.g., high energy expenditure or high mortality costs). In contrast, the differential mating opportunities hypothesis posits that the sex that has more opportunities to remate following desertion should desert (e.g., if there are many adults of the opposite sex in breeding condition available in the population). Below, we discuss the weight of evidence for these two hypotheses and argue that there is little supporting evidence for the differential cost of parental care hypothesis, and that differential mating opportunities are a more viable predictor for the resolution of sexual conflict in breeding systems. 1. Differential Costs of Parental Care Do males and females differ in the costs that they pay for parental care? Such differences could determine which sex likely to desert. However, current evidence is inconclusive. In a comparative analysis across all birds (including several shorebird species), Liker and Sze´kely (2005) showed that male mortality increased with the duration of male care only after controlling for the degree of mating competition (i.e., mating system). In contrast, female mortality was unrelated to the duration of female care. This implies that males incur greater costs of care and should be more likely to desert than females. Nonetheless, the differential mortality costs of care cannot fully explain the outcome of sexual conflict over parental care since either the male or the female may desert. Furthermore, the relationship between male care and male mortality was significant only after controlling for mating competition and the full model explained only 18% of the variation in male mortality. Few studies to date have considered how the division of parental care might affect the outcome of sexual conflict. For example, in both black‐ tailed godwits Limosa limosa and northern lapwings, males are more involved in nest defense, whereas females spend more time on incubation (Hegyi and Sasva´ri, 1998; Liker and Sze´kely, 1999). Consequently, the opportunity for either parent to desert may depend on the type of external

PARENTAL CONFLICT IN SHOREBIRDS

303

threats, for instance, high predation risk or climate, that most influnces offspring survival. There are two general outcomes of sexual conflict for each parent if males and females perform different tasks in parental care. First, if the need for the main male component of care (e.g., nest defense) is low then the male parent has the opportunity to desert. Second, if the need for the male component of care is high then that sex may not have the opportuntity to desert and is constraints on care. The latter explanation is equivalent to a relaxation of the constraints on care discussed above, except that in this case the constraints apply only to one parent. We do not envisage this explanation alone could generally determine which sex deserts for two resons. First, the division of sex roles in components of parental care may be liable in many species. Second, where there are low constraints on parental care in both sexes, the division of labour hypothesis makes no prediction on which sex is more likely to desert. To determine whether the division of labour hypothesis is plausible, new experimental manipulations on the division of parental duties between the sexes are needed. 2. Differential Mating Opportunities Mating opportunities of individuals within a population may vary for two main reasons. First, sex ratios could be biased, influencing mating opportunities of all individuals within a population. Second, mating could be state dependent (i.e., mating opportunities vary from individual to individual). Mating opportunities appear to be male biased in several shorebirds, which should favor female ‘‘emancipation’’ from care. Male‐biased sex ratios, and presumably female‐biased mating opportunities, are well known from some polyandrous species, with examples in the wattled jacana (Emlen and Wrege, 2004), red phalarope (Whitfield, 1995), and bronze‐winged jacana (Butchart, 2000). Adult sex ratio appears to be male biased in Kentish plover populations too. Over a 10‐year period adult sex ratios were male biased in breeding populations in Sweden (Jo¨nsson, P. E., personal communication; Sze´kely and Williams, 1995) and California (Warriner et al., 1986). This male bias favors female desertion and the evolution of polyandry, and suggests a causative explanation for the rapid remating of female Kentish plovers. The mating opportunities for male and female plovers were experimentally compared by Sze´kely et al. (1999) by removing either males or females from breeding pairs, and recording the time taken for the abandoned parent to remate. Males took on average 25.4 days to find a new mate whereas females took only 5.3 days. Furthermore, the time taken to renest increased for both sexes as the breeding season progressed. However, biased sex ratio per se is not necessarily a prerequisite for the evolution of polyandry. Erckmann (1981) found that the male to female ratio in the polyandrous Wilson’s phalarope was 0.61. Nonetheless, and consistent with

304

GAVIN H. THOMAS ET AL.

the sexual conflict framework that we describe in Section VI, he pointed out that local sex ratios can change rapidly in phalaropes and polyandry in these species is facultative. Exploitation of mating opportunities may also be state dependent. In spotted sandpipers, older (and presumably more experienced) females are able to outcompete younger individuals in the search for mates (Oring and Lank, 1986). Young females that attain a mate are typically monogamous and provide parental care. In contrast, older females often abandon the clutch. This behavior is probably most frequent in high‐density populations, implying that abandonment by older females is contingent on there being males available for mating. In some species, males adjust their behavior in order to seek out meeting opportunities. The buff‐breasted sandpiper, an Arctic‐breeding sandpiper, displays lekking behavior and Lanctot and Weatherhead (1997) observed that while the sex ratio of an Alaskan population was almost constant throughout a breeding season, the absolute number of males at a lek increased with the number of fertile females. They suggested that the Arctic breeding season, where snow accumulation and melt is highly changeable between sites and years, is notable for the variability of mating opportunities. Consequently, males moved opportunistically between sites in search of females (although it is also possible that females moved to leks with more males), a behavior also noted in the ruff (Lank and Smith, 1987). We speculate that the ability to move between sites is state dependent and may be largely restricted to the best quality individuals. Mathematical models (Houston et al., 2005) and evidence from field studies suggest that mating opportunities, rather than differential costs of care, are the likely determinant of desertion. This has been corroborated by a phylogenetic comparative analysis of birds. Female‐only care occurs in taxa where remating opportunities are abundant for both sexes, whereas male‐only care occurs in taxa where remating opportunities are rare, particularly for males (Owens, 2002). The implication is that to predict the mating system of a species we need some measure of mating opportunities, such as the time to remating for males and females in a population (Sze´kely et al., 1999), or the OSR (Emlen and Oring, 1977). However, OSR and the time to remate are tied to one another, and may be as much an outcome of the breeding system as they are an explanation for it (see below). It will be valuable to test differential mating opportunities in more shorebird species to complement work on the Kentish plover. We predict that in monogamous biparental species there will be less bias in mating opportunities between males and females than in the Kentish plover, in which females desert more often than do males.

PARENTAL CONFLICT IN SHOREBIRDS

305

3. Feedback Between Parental Care and Social Mating System To assess the gains (i.e., payoffs in game‐theoretic parlance) from caring and deserting, one needs models in which the behavior and the number of mated and unmated individuals in the population are consistently followed throughout the breeding season (McNamara et al., 2002; Webb et al., 1999). The emphasis is placed on a feedback loop between mating strategies (to mate or not), mating opportunities, and parental care strategies (desert or care, Sze´kely et al., 2000a). The essence of the aforementioned game‐theoretic models can be summarized as follows. The quality and availability of mates in a population depends on the mating decision of all other members of the population, and the best strategy may differ for each individual. Once mated, the parent is then faced with the dilemma: ‘‘should I stay or should I go?.’’ To stay could enhance the chances of offspring survival, but to go could afford new mating opportunities and more offspring. The payoff from caring is likely to be high if the brood is of high quality or if staying results in mate‐retention. But desertion also has high payoffs if mating opportunities are abundant. If a parent deserts, it is returned to the pool of unmated individuals, and the mating optimum is shifted across the population. Mating opportunities are determined by the mating and parental care strategies of each individual in the population. At the same time, the decision of each individual determines the mating opportunities, and consequently the optimal mating and parental care strategy for the rest of the population. Does this scenario have any relevance for natural populations? We argue that it may. For instance, in Temminck’s stint Calidris temminckii many females lay two clutches, each for a different male. After the female has laid a clutch, the pair splits and both the male and the female attempt to attract a new mate and produce a new clutch. If both of them are successful in finding a new mate, then the male starts incubating the first clutch, whereas the female incubates the second one (Hilde´n, 1975). There are, however, deviations from this pattern: some females lay three successive clutches leaving successive partners to deal with the first two clutches; some males mate with two females, for whom this is their first clutch. We argue that these deviations are produced by the race for new mates. The race for mates can be wasteful. Hilde´n (1975) noted that both the male and female abandoned some clutches. These wasted clutches may be the outcome of intense sexual conflict: both parents are seeking new mates while assuming (wrongly) that their former mate will take up parental duties. Parental care patterns and social mating system are emergent outcomes of the mating opportunity‐mediated feedback loop. We argue that these emergent traits are observable and leave their imprint at the phylogenetic level. The correlation between parental care and social mating system is

306

GAVIN H. THOMAS ET AL.

well known: male‐only care is often associated with social polygyny, biparental care with social monogamy and female‐only care with social polyandry (Bennett and Owens, 2002; Lack, 1968; Ligon, 1999). The relationship is strong across species in shorebirds, after accounting for phylogenetic relationships (Fig. 6). However, it is usually assumed that the direction of causation runs from parental care to social mating system. That is, the sex that invests less in offspring care competes more intensely for mates. But the feedback model above suggests that this need not be the case, since mating and parental behaviors may mutually influence one another (Alonzo and Warner, 2000; Reynolds, 1996; Sze´kely et al., 2000a). At the phylogenetic level, the feedback model predicts that changes in the mating system of a species are just as likely to precede changes in parental care as the reverse. According to parental investment theory (Trivers, 1972) evolutionary transitions toward social polygamy should arise only after transitions toward uniparental care and the reverse transitions should be absent, or very rare. In contrast, sexual conflict theory predicts that changes back to biparental care from uniparental care in

Fig. 6. The relationship between parental care and social mating systems in shorebirds. The independent axis shows phylogenetically independent contrasts in parental care bias between the sexes (duration of male care—duration of female care). The dependent axis shows phylogenetically independent contrasts in social mating system (polygyny ! monogamy ! polyandry). Note that social mating system refers to the type of pair‐bond, whereas parental care refers to the duration of care. Fitted line is the major axis regression through the origin (slope 0.077, lower 95% confidence interval 0.067, upper 95% confidence 0.091). Adapted from Thomas (2004).

PARENTAL CONFLICT IN SHOREBIRDS

307

socially polyandrous species are likely, as are changes to social monogamy from social polygamy in uniparental species. Recent phylogenetic methods using continuous‐time Markov models allow the most likely order and direction of trait evolution to be evaluated (Pagel, 1994). Using this approach, implemented in the computer program DISCRETE (Pagel, 1994), Thomas and Sze´kely (2005) showed that the phylogenetic patterns of transition of parental care and social mating system in shorebirds are consistent with the predictions of the feedback model. Thus Trivers’s dictum that ‘‘what governs the operation of sexual selection is the relative parental investment of the sexes in their offspring’’ appears to be incorrect, at least in shorebirds (Trivers, 1972). VI. A SEXUAL CONFLICT FRAMEWORK FOR BREEDING SYSTEMS The classical views of shorebird breeding system evolution focused on ecological hypotheses (Erckmann, 1983, Oring, 1986; Pitelka et al., 1974). Yet only two ecological traits have been shown to correlate with breeding systems in comparative studies that include shorebirds, and both have relatively low explanatory power. Owens (2002) showed that nesting density was associated with desertion in polyandrous versus nonpolyandrous birds, a result consistent with the predictions made by Emlen and Oring 25 years earlier. Reynolds and Sze´kely (1997) found that the duration of care declined with increasing migration distance in shorebirds, although the latter result only applied to males. Reynolds and Sze´kely (1997) suggested that it is likely that migration distance evolved in response to parental care, since migratory patterns must have evolved since the last glacial retreat 12,000 years ago (Alerstam, 1990). This need not necessarily be true though, because while the absolute migration distance will have undoubtedly changed for many species, the relative migration distance across species, along with its correlation with parental care, may have its roots much deeper in evolutionary time. The dynamic nature of the feedback between parental care and mating strategies may explain why general correlates of social mating system and parental care have been difficult to identify (Ligon, 1999; Owens, 2002). We suggest that it is time to move on, and explicitly consider interactions between individuals in addition to the more general factors that make desertion possible. In outlining their hierarchical view of avian mating systems, Bennett and Owens (2002) recognized three levels of mating system determinants: phylogenetic constraint, ecological facilitation, and social interactions. We concur with this broad concept. However, the forms of social interactions were not detailed by Bennett and Owens (2002). We emphasize the significance of sexual conflict, particularly with respect to its influence on mating opportunities for each sex.

308

GAVIN H. THOMAS ET AL.

The general framework for breeding system evolution we proposed above is based on sexual conflict, starting from the viewpoint of a mated pair. Whether one parent has the potential to desert depends on the survival prospects of the young. The evidence reviewed above suggests that the most important constraint on the potential to desert is the developmental mode of the offspring (e.g., precocial or semiprecocial), rather than a specific aspect of the environment. Phylogenetic comparative approaches are likely to provide further insight into these constraints. Similarly, long‐term data on changes in parental care in relation to seasonal variation in predation risk, food availability, and climate variables would be helpful. However, we stress that relaxation of these constraints does not determine whether desertion actually occurs, nor which sex is more likely to desert. We believe that relative mating opportunities for each sex, and thus population sex ratio, are key factors in determining which sex deserts. Emlen and Oring (1977) also stressed the importance of mating opportunities in relation to the spatial distribution of mates and the OSR. The fundamental difference between Emlen and Oring’s view and the sexual conflict one that we are presenting here is that we consider mating opportunities and parental behaviors as being in a dynamic state. In particular, the decisions of mated and unmated individuals of each sex are both a cause and a consequence of mating opportunities and parental behaviors. In this view, the parental care and mating patterns of a species are ultimately the result of conflicts between the sexes over mating and parental investment in the offspring. Testing the links between mating opportunities, parental care, and mating system will be challenging. In particular, data on mating opportunities are lacking for most shorebirds. Further experimental manipulations, particularly removal studies, are needed. We believe that the sexual conflict model of breeding system evolution has the potential to explain the patterns of parental care and social mating system observed across species in shorebirds, and can be readily adapted to other groups. Furthermore, by explicitly accounting for variation in mating opportunities for each sex, it also indicates which sex should pursue mating more vigorously and thus informs on the potential for extra‐pair fertilizations.

VII. FUTURE DIRECTIONS Throughout this chapter we have argued that sexual conflict is an important concept for understanding breeding systems of shorebirds. However, we acknowledge that much of the current evidence is qualitative, observational, and indirect. To corroborate (or falsify) the influence of sexual conflict, a new generation of experiments, advanced comparative analyses, and theoretical

PARENTAL CONFLICT IN SHOREBIRDS

309

models are needed. Given that several extensive field projects have been carried out with shorebirds, it is somehow surprising that only a few of them have carried out experiments. In addition, sexual conflict theory potentially has impacts beyond reproductive behaviors, particularly with respect to large‐ scale evolutionary processes including speciation and extinction. Below we briefly outline areas of research that are ripe for future development. A. DISPLAY TRAITS In Section IV, we highlighted the link between male display behavior and sexual conflict over body size evolution. As Graul (1973a) noted, male displays could be amplified by contrasting pigmentation. To fully understand how mate choice might influence conflict in body size, future research should expand the analyses of sexually dimorphic traits, specifically plumage and display behavior, in three major ways. First, display agility has been quantified in only a few shorebirds, and plumage colors and ornaments in fewer still. Methods for measuring these traits are well established (e.g., plumage color reflectance, Bennett and Cuthill, 1994; Olson and Owens, 2005; and agility of display behavior: Blomqvist et al., 1997; Grønstøl, 1996). Quantitative descriptions would be particularly valuable in those taxa that exhibit large intraspecies and/or interspecific variation in these traits (Reynolds, 1987). Second, experimental alteration of these traits would be very illuminating. For instance, no study has yet manipulated flight maneuverability in shorebirds to test the effect of agility on mate attractiveness and male–male competition. Finally, it is important to assess effects of these studies beyond behavioral response (e.g., mating time, number of mates). Measuring reproductive success over the full breeding season, and possibly the reproductive success of the offspring produced by manipulated and control individuals would be very valuable. Given the highly mobile nature of many shorebirds, these will be challenging tasks. B. DIVERSIFICATION Sexual selection by mate choice may result in the coevolution of attractive traits and choice for those traits (Fisher, 1930). Thus, secondary sexual characters can evolve rapidly in a positive feedback runaway process. According to the sexual selection hypothesis, rapid divergence in female mate choice is likely to result in the formation of reproductive barriers in isolated populations and thus promote speciation (Barraclough et al., 1998; Lande, 1981; Panhuis et al., 2001; West‐Eberhard, 1983). Sexual conflict has been identified as an alternative driver for speciation as a result of chase‐away selection on male and female reproductive traits (Arnqvist et al., 2000; Gavrilets, 2000; Gavrilets and Waxman, 2002; Martin and Hosken, 2003; Parker and Partridge, 1998). Testing these hypotheses in

310

GAVIN H. THOMAS ET AL.

any group will be challenging, in particular quantifying sexual conflict and sexual selection. Previous studies on species richness and speciation used variables such as mating system and SSD as proxies for sexual conflict (Arnqvist et al., 2000; Gage et al., 2002). More direct measures of both sexual selection and sexual conflict are required (Mills et al., 2007). C. EXTINCTION RISK AND POPULATION DECLINES While interlocus conflicts are likely to be the primary source of sexual conflict‐driven speciation, heightened extinction risk could arise through either inter‐ or intralocus conflicts. Theory developed for sexually selected traits suggests that the accumulation of exaggerated male traits through female choice may be energetically costly (Tanaka, 1996). Energetic costs from traits subject to sexually antagonistic coevolution (SAC) may also occur, and indeed be exacerbated by direct harm from the traits themselves. The effect of selection load may directly influence survival and therefore drive population declines. Alternatively, it may make some species more susceptible to external threats. As we discuss in Section IV, intralocus conflicts may prevent one (or both) sexes from reaching their fitness optima. Comparative analyses of population trends or threat status will be fruitful. VIII. SUMMARY We argue that sexual conflict theory is an excellent conceptual framework for understanding the diversity of breeding systems. In this review we focus on shorebirds (Charadriiformes), although the theoretical framework should be applicable beyond this clade. Shorebirds are an excellent group to understand breeding system evolution, because they exhibit unusual ranges of mating systems and parental care among birds. First, we review cases in which the interests of males and females are different over mating. This includes mate choice, female–female competition, and infanticide. Second, we review experimental and phylogenetic studies that suggest conflict between males and females over parental care. The latter studies, along with game‐theoretic models, suggest that conflict resolution over care has implications for mating behavior. In turn, the resolution of mating conflict may have implications for parental behavior. Thus a key element of sexual conflict theory, unlike traditional sexual selection theory, is the dynamic view of mating and parental behaviors. To test the predictions and implications of sexual conflict theory, we need thorough experiments in the field or aviary, and advanced phylogenetic comparative analyses. We anticipate that shorebirds will continue providing challenges and solutions to central issues of evolutionary biology, such as breeding systems.

APPENDIX. TABLE 1. SHOREBIRD DATA SETa

311

Male Female wing wing length length

Male Female bill bill length length

Social mating

Male display

Male

Female

Egg mass

Clutch

Chick

Breeding

Male

Female

(mm)

(mm)

system

type

care

care

(g)

size

feeding

habitat

polygamy

polygamy

–

0

4

1;1;1;1;_;_;1;1; 2;2;3;4;1; _;1;1

2

–

0

0

–

2

2

0

0

–

–

2

–

–

–

5;5;5;5;5;5;5;6;5;5;5; _;5; _;6;6 _;_:5;5;5;5;5;5; 5;5;6;_;5; 7;5;5 _;_;8;8;_;_:6;_; _;_;_;_;6; _;_;

7 7

– 28

– 1

2 2

– 1.5

0 0

0 0

7

7

12.4

2

0

–

1

0

–

7

7

25.1

1

2

–

0

0

1

AA

7

7

1

2

2

0

0

39.2 26.5

– 1

– AA

7 7

7 7

17.4 22.4

1 4

2 –

– 1

– –

– –

22.8

1

AA

7

4

17.9

3.5

0

2

0

0

Male mass

Female mass

(g)

(g)

Actophilornis africanus

143.2

260.7

145.8

167.5

–

–

3

GD

7

0

8.6

4

0

Aethia cristatella

267.7

251.4

143

140.3

12.6

11.8

1

GD

7

7

54.2

–

Aethia pusilla

–

–

89.9

90.8

9.3

9.3

1

GD

7

7

17.4

Aethia pygmaea

–

–

107.8

109.2

–

–

1

–

–

–

Alca torda Alle alle

734 154

700 151

209.7 123.3

207.6 120.7

35.7 16.1

34.5 15.5

1 1

– GD

7 7

Anarhynchus frontalis

58.6

56.7

119.3

119

29.9

28.8

1

GD

Anous minutus

–

97.5

226.5

221.2

44.1

41.3

1

Anous stolidus

196.4

172.3

283

273

40.8

38.1

Anous tenuirostris Aphriza virgata

92 186.3

114 216.3

210 171.8

213.3 178.8

41.9 24.1

Arenaria interpres

108

113

155

157

22.4

Arenaria melanocephala

113.6

124.2

147

(mm)

151.9

22.5

(mm)

23

1

ANA

7

7

17.3

4

0

–

0

0

References

9 9;9;9;9;9;9;9;9; 9;9;9;6;9; 9;9;9 3;3;10;10;10;10;11;7;11; 11;3;11;11;_;10;10 _;12;12;12;12;12;12;_; 12;12;6;6;12;_;12;12 12;12;12;12;12;12;6;12; 12;12;_;1;12;9;1;1 12 6;6;6;6;6;6;6;6;6;6;_;6; _3;_;_ 13;13;13;13;13;13;14;15; 14;14;3;13;13;13; 13;13 16;16;6;6;6;6;17;6; 17;17;18;3;19;_;6;6

(Continued)

APPENDIX. TABLE 1. (Continued)

312

Male mass (g)

Female mass (g)

Male Female wing wing length length (mm) (mm)

Male Female bill bill length length (mm) (mm)

Social mating system

Attagis gayi

–

–

191

191.3

23.8

24.3

1

Bartramia longicauda Brachyramphus brevirostris

151 –

164 –

166 –

170 –

28.9 –

28.8 –

Brachyramphus marmoratus Burhinus capensis Burhinus grallarius

217

222.7

134.2

132.6

15.5

– 671.5

440.5 625.3

231 278.8

231 272

Burhinus oedicnemus

475

449

241

Burhinus senegalensis

–

–

Burhinus vermiculatus

297

Calidris acuminata

Male Female care care

Egg mass (g)

Clutch size

Chick Breeding Male Female feeding habitat polygamy polygamy

–

–

–

–

–

–

1

–

–

1 1

ANA –

7 –

7 –

23.5 –

3.99 –

0 2

– –

– –

– –

15.3

1

GD

7

7

59.8

1

2

1

–

–

36.8 49.2

36.8 48.7

1 1

– GD

7 7

7 7

– 41.8

2 2

2 2

– 1

0 0

0 0

240

40.1

39.3

1

GD

7

7

42

1.9

2

–

0

0

221

221

47.2

46.8

1

GD

7

7

32

2

–

–

0

0

314.3

205

205

44

41.3

1

–

–

–

–

2

–

0

0

70.3

63.5

140

130

26.3

24.7

2

ANA

0

7

13.7

3.8

–

1

4

4

Calidris alba

52.8

55.4

125

130

25.6

26.4

4

AA

7

7

11.2

3.9

0

0.5

0

1

Calidris alpina

41

45.1

112

116

26.1

29.5

1

AA

6

4

10.7

3.9

0

2

1

1

Calidris bairdii

39.3

39.7

125

129

22.3

23.8

1

Male display type

AA

7

4

9.6

4

0

1

0

0

References _;_;20;20;20;20;7;_;_;_; _;_;_;21;_;_ 6 6 5;5;5;5;5;5;5;6; 5;5;5;6;5;7;_;_ 1 10;10;10;10;10;10;22;10; 22;22;22;22; 10;7;10;10 13;13;13;13;13;13;23;13; 23;23;13;13;13; _;13;13 _;_;1;1;1;1;1;13; 24;24;18;1;_;_;1;1 25;25;25;25;25;25;_;_; _; _;_;25;_;_;1_;_ 3;3;13;13;15;15;26;21; 26;26;3;3;_;7;13;13 13;13;13;13;13;13;27;13; 27;27;3;13; 13;13;13;13 13;13;13;13;13;13;28;13; 28;28;18;13; 13;13;13;13 13;13;13;13;13;13;26;15; 26;26;3;3;3;7;6;6

313

Calidris canutus

126

148

169

173

32.6

34.4

1

AA

7

6

19.3

3.7

0

2

0

0

Calidris ferruginea

63.2

63.3

132

133

36.2

40.1

2

AA

0

7

12

4

–

2

–

–

Calidris fuscicollis

39.7

45.8

122

125

23.2

24.4

2

AA

0

7

10.8

4

0

1.5

3

0

Calidris maritima

67.6

76.3

127

132

27.5

32

1

AA

7

3

13.3

3.9

0

2

0

0

Calidris mauri

28

31

97.1

101

23.1

26.7

1

AA

7

4

7.5

3.9

0

1

0

0

Calidris melanotos

97.8

65.1

144

131

29.6

27.7

2

ANA

0

7

13.1

4

0

1

–

–

Calidris minuta Calidris minutilla

24 20.3

27.1 22.2

96.4 89.8

99.5 91.6

18.1 18.2

18.5 19.5

4 1

AA AA

7 7

0 5

6.3 6.4

3.8 3.9

– 0

– 2

1 0

2 0

Calidris ptilocnemis

76.3

83

121.3

125.5

26.8

29.8

1

AA

7

4

14.2

4

0

1.5

0

0

Calidris pusilla

25

27

95.9

100.1

18.6

20.2

1

AA

6

5

6.9

4

0

2

0

0

Calidris ruficollis

25.7

26.6

103.5

106.2

17.5

18.7

1

AA

7

4

8.3

4

–

2

0

0

Calidris subminuta

29

32

93.1

95

17.8

18.8

1

AA

7

4

7.5

4

–

1.5

–

–

Calidris temminckii

24.3

27.8

98.2

99.3

16.9

17.2

4

AA

7

0

5.8

4

0

–

3

3

Calidris tenuirostris

156

174

185

192

42.1

43.8

1

AA

7

3

22

4

–

1

0

0

Catharacta maccormickii Catharacta skua

1277 1735

1421 1935

410 411.9

415 420.6

49.4 54.7

50.9 55.3

1 1

AA AA

7 7

7 7

– 111.7

1.8 2

2 2

2

0 0

0 0

13;13;13;13;15;15;13;13; 29;29;18;13; 13;13;13;13 13;13;13;13;15;15;3;15; 30;30;3;3;_;13_;_ 13;13;13;13;15;15;31;6; 31;31;3;3;3;7;6;6 13;13;13;13;13;13;13;15; 13;13;3;13; 13;13;13;13 12;12;13;13;15;15;32;15; 32;32;3;33;32;3;6;6 13;13;13;13;15;15;34;15; 34;34;3;3;6;13;_;_ 13 13;13;13;13;13;13;35;6; 35;35;3;3;36;13;6;6 16;16;37;37;37;37;26;26; 26;26;3;3;13;7;6;6 13;13;13;13;15;15;38;15; 38;38;3;3;6;13;6;6 13;13;13;13;15;15;26;15; 26;26;3;3;_13; _;26;26 13;13;13;13;13;13;26;26; 39;39;3;3;_;7;_;_ 13;13;13;13;13;13;40;13; 40;40;3;13;13; _;13;13 13;13;13;13;13;13;26;26; 26;26;3;3;_;13;26;26 12 12

(Continued)

APPENDIX. TABLE 1. (Continued) Male

Female

Male

Female

Male mass (g)

Female mass (g)

wing length (mm)

wing length (mm)

bill length (mm)

bill length (mm)

Social mating system

Male display type

Male Female care care

Egg mass (g)

Clutch size

Chick Breeding Male Female feeding habitat polygamy polygamy

314

Catoptrophorus semipalmatus Cepphus carbo Cepphus columba

273

301.4

205.2

213.5

59.4

64.1

1

AA

7

5

39.5

4

0

–

0

0

510 487

480 483

195 187.5

189.5 187.5

– –

– –

– 1

– –

– 7

– 7

– –

1 –

– 2

– –

– 0

– 0

Cepphus grylle

376

386

164

165

33.3

33.2

1

GD

7

7

50

1.43

2

2

1

0

Cerorhinca monocerata Charadrius alexandrinus

510 48.2

456 47.1

182.9 111

177.7 112

18.3 15.4

18.5 15.2

1 1

GD ANA

7 7

7 4

77.7 9

1 3

2 0

– 2

– 2

– 1

Charadrius asiaticus Charadrius bicinctus

75.6 58.8

76.2 58.4

150 130.2

150 127.2

19.8 17.2

20.1 17

1 1

ANA ANA

7 7

7 7

– 11.5

– 2.93

0 0

– –

– 0

– 0

Charadrius dubius

38.3

39.2

117

116

12.7

12.9

1

ANA

7

4

7.7

3.9

0

2

0

0

Charadrius forbesi Charadrius hiaticula

– 63.5

– 64.7

– 132

– 135

– 14.1

– 14.5

– 1

– ANA

– 7

– 6

– 10.9

– 3.8

– 0

– –

0 0

0 0

Charadrius leschenaultii

–

–

141

144

23.3

23.2

–

ANA

–

–

–

–

–

–

–

–

Charadrius marginatus Charadrius melodus Charadrius mongolus Charadrius montanus

48.3 54.9 – 102

48.3 55.6 – 114

110 121 130 146

110 120 131 149

17 – 18.6 20.5

17 – 18.6 22

1 1 – 4

– AA – GD

– 7 – 7

– 7 – 0

– 9.4 – 16.5

2 3.3 – 3

0 0 – 0

– – – –

0 0 – 0

0 1 – 0

Charadrius obscurus

160

110

159.3

159.1

29

27.5

1

GD

7

7

–

–

–

–

0

0

References 3;3;3;3;3;3;41;41; 41;41;3;3;41;_;6;6 8 6;6;6;6;_;_;7;_; 7;7;_;_;7; _;6;6 5;5;5;5;5;5;5;9; 5;5;5;9;5;9;9;9 6 13;13;13;13;13;13;42;13; 42;42;43;13; 13;13;13;13 13 10 13;13;13;13;15;15;13;13; 44;44;13;13; 13;13;13;13 1 13;13;13;13;15;15;13;13; 13;13;3;13;13; _;13;13 _;_;10;10;10;10;_;13;_;_; _;_;_;_;_;_ 1 6 10 3;3;3;3;15;15;45;6; 45;45;3;3;46;_;6;6 16;16;13;13;13;13;13;7; 13;13;_;_;_;_;10;10

– – –

– – –

– 106 143.2

– 106 144.4

– 15 17.7

– 15 17.6

1 1 1

GD GD

7 7 7

7 7 7

– – –

– – –

– – –

– – –

0 0 0

0 0 0

Charadrius ruficapillus

37.3

37.6

105.4

105

13.8

13.9

–

ANA

7

7

–

2

–

–

–

–

Charadrius

–

–

–

–

–

–

1

–

7

7

10.8

2

0

–

–

–

sanctaehelenae Charadrius semipalmatus Charadrius tricollaris Charadrius veredus Charadrius vociferus

46.1 – – 92.1

47.8 – – 101

– – 166.5 167

– – 164.5 167

– – 22.5 20.2

– – 22.3 20.3

1 1 – 1

– GD – –

7 7 – 7

5 7 – 7

– – – 14.5

3.94 – – 4

0 0 – 0

– – – –

0 0 – 0

0 0 – 0

Charadrius wilsonia

59

63

116.4

117.6

20.7

20.8

1

–

7

2

12.4

3

0

–

–

–

Chionis alba

735

638

253

242

34.4

32

1

GD

7

7

45.5

2.2

2

2

0

0

Chionis minor

625

551

240

231

–

–

1

GD

7

7

43.1

2.5

2

–

0

0

Chlidonias albostriatus Chlidonias hybridus

– 90

– 86

– 242

– 232

– 31.6

– 28.5

1 1

– AA

– 7

– 7

– 16

2 2.74

2 2

– –

– 0

– 0

Chlidonias leucopterus

–

–

215

212

–

–

1

–

7

7

10.5

2.8

2

–

0

0

Chlidonias niger

61.8

58.7

218

213

27.8

26.5

1

ANA

7

7

11

2.83

2

2

0

0

Cladorhynchus leucocephalus Coenocorypha aucklandica

262.8

261

201.5

196.2

71.4

66.3

–

GD

7

7

–

–

0

–

–

–

101.2

116.1

106.2

109.2

57.5

60.9

1

AA

7

7

23.7

2

2

1

2

0

315

Charadrius pallidus Charadrius pecuarius Charadrius rubricollis

1 13 _;_;10;10;10;10;10;7; 10;10;_;_;_;_;10;10 10;10;10;10;10;10;_;7; 10;10;_;10;_;_;_;_ _;_:_;_;_;_;47;_; 47;47;18;47;19;_;_;_ 6 1 10 1;1;1;1;15;15;48;_; 48;48;3;3;3;_;6;6 3;3;3;3;15;15;13;_; 49;49;3;3;3;_;_;_ 10;10;10;10;10;10;50;10; 50;50;50;50; 10;7;10;10 51;51;15;15;_;_;52;10; 52;52;18;10;10; _;10;10 12 9;9;9;9;9;9;9;9; 9;9;9;53;5;_;1;1 _;_;9;9;_;_;9;_; 9;9;9;13;9;_;9;9 13;13;13;13;13;13;13;13; 13;13;9;13;13;9;9;9 10;10;10;10;_;_;10;7; 10;10;_;_;10;_;_;_ 12;12;12;12;12;12;12;12; 54;54;54;54;12; 7;12;12

(Continued)

APPENDIX. TABLE 1. (Continued)

316

Male mass (g)

Female mass (g)

Male wing length (mm)

Female wing length (mm)

Male bill length (mm)

Female bill length (mm)

Social mating system

Coenocorypha pusilla

75.9

85.4

99.9

100.5

43.6

44.4

1

Cursorius

–

–

153

153

–

–

coromandelicus Cursorius cursor

131

128

158

156

–

Cursorius rufus Cursorius temminckii

– 68.6

– 74.3

135 –

135 –

Cyclorhynchus psittacula

292.3

285.7

147.4

Dromas ardeola

–

–

Elseyornis melanops

28.5

Erythrogonys cinctus

Male display type

Male care

Female care

Egg mass (g)

Clutch size

Chick feeding

Breeding Male Female habitat polygamy polygamy

AA

7

7

16.1

2.1

2

1

0

0

1

–

7

7

–

–

1

–

–

–

–

1

AA

7

7

13.9

2

1

–

–

–

22.9 –

22.9 –

– 1

– –

– –

– –

– 8

1.9 2

– –

– –

0 0

0 0

145.2

15.8

15.6

1

GD

7

7

53.8

1

2

–

0

0

215

215

58.1

56.8

–

GD

–

–

45

1

2

1.5

–

–

31.5

110

112.5

15.4

15.4

1

ANA

7

7

–

–

–

–

0

0

46.8

50

111.5

112.2

21.2

20.7

–

–

7

7

–

–

–

–

0

0

Eudromias morinellus

100

117

151

155

15.9

16.8

3

ANA

7

0

17

2.9

0

1

2

3

Eurynorhynchus pygmeus

31

34.6

104

106.9

21.1

22.3

1

AA

7

4

8

4

–

–

0

0

Fratercula arctica

391

361

164

162

45.7

43.1

1

GD

7

7

60

1

2

2

0

0

Fratercula cirrhata

815.9

753.3

203.8

199.7

–

–

1

–

7

7

87.9

1

2

–

–

–

Fratercula corniculata

529.9

512.5

190.2

185.7

–

–

1

–

–

–

75.9

2

–

0

0

References 12;12;12;12;12;12;54;12; 54;54;54;54;12;7; 12;12 25 13;13;13;13;_;_;13;1; 13;13;18;13;1;_;_;_ 1 1;1;_;_;_;_;1;_; _;_;18;1; _;_;1;1 5;5;5;5;5;5;5;6; 5;5;5;6;5; _;6;6 _;_;13;13;13;13;_;15;_; _;1;13;7;13;_;_ 16;16;10;10;10;10;10;7; 10;10;_;_;_;_;10;10 10;10;10;10;10;10;_;_; 10;10;_;_;_;_;53;53 13;13;13;13;15;15;55;13; 55;55;3;13;13; 13;13;13 3;3;3;3;3;3;56;57; 56;56;3;8;_;_;56;56 5;5;5;5;5;5;9;9; 9;9;9;9;9;9;9;9 6;6;6;6;_;_;6;_; 8;8;6;6:8; _;_;_ 6;6;6;6;_;_;6:_: _:_:6:_:6: _:6:6

317

Gallinago gallinago

111

128

134

134

66.6

68.5

1

AA

7

7

16.5

3.9

1

2

0

0

Gallinago hardwickii

145.5

160.4

157.4

161.5

67.7

70.7

–

–

–

–

–

–

–

–

–

–

Gallinago media

155

175

144

146

61.2

64.7

2

GD

0

7

23.2

3.5

0

1

4

0

Gallinago megala

–

–

139.2

141.9

61.8

66.3

–

AA

–

–

–

–

–

–

–

–

Gallinago nigripennis

–

–

130

132

70.7

77.5

1

AA

–

–

–

–

–

–

0

0

Gallinago stenura Glareola cinerea Glareola maldivarum

101 – 76.8

132 – 75.2

132 141 185.2

132 146 182.9

60.1 10 15.1

63.7 10.5 15.5

1 1 1

– – ANA

– – 7

– – 7

– – –

– 1.85 –

– 2 –

– – –

– – –

– – –

Glareola nordmanni

99

96

204

193

–

–

1

–

–

–

11

3.6

–

–

0

0

Glareola nuchalis

–

–

143

150

–

–

1

–

–

–

7.4

1.6

–

–

0

0

Glareola ocularis

103

82

196

198

11.2

11.2

–

–

–

–

–

–

–

–

–

–

Glareola pratincola

79

75

198

192

21.5

21.9

1

AA

7

7

10.1

3

2

0

0

0

Gygis alba

142

135

236.6

235.8

40.8

40

1

AA

7

7

21.6

1

2

–

0

0

Haematopus bachmani

–

–

245

247.5

68.3

73.6

1

–

7

7

45.2

2

2

–

0

0

Haematopus finschi

517

554

253

260

80.3

92.3

1

AA

7

7

44.2

2.33

2

–

0

0

13;13;13;13;15;15;13;13; 58;58;18;13;13;13; 13;13 12 59;59;13;13;15;15;13;15; 13;13;3;43;13; 13;13;13 _;_;12;12;12;12;_;3;_;_; _;_;_;_;_;_ 1 13 1 12;12;12;12;12;12;15;15; 12;12;_;_;_;_;_;_ 13;13;13;13;_;_;13;_;_; _;18;13;_;_;13;13 _;_;1;1;_;_;1;_;_;_;18;1; _;_;1;1 16;16;1;1;1;1;_;_;_;_;_;_; _;_;_;_ 13;13;13;13;13;13;13;13; 13;13;18;1;18;13; 13;13 12;12;12;12;12;12;12;6; 12;12;6;12;12; _;12;12 _;_;15;15;15;15;3;_; 60;60;18;61;60;_;6;6 10;10;10;10;10;10;10;7; 10;10;10;10;10; _;10;10

(Continued)

APPENDIX. TABLE 1. (Continued)

318

Male Female wing wing length length

Male Female bill bill length length

Social mating

Male display

Male

Female

Egg mass

Clutch

Chick

Breeding

Male

Female

(mm)

(mm)

system

type

care

care

(g)

size

feeding

habitat

polygamy

polygamy

–

0

0

2

–

0

0

1.7

2

–

0

0

46.7

2.8

2

2

2

0

7

44.9

2.6

2

–

1

0

7

7

48.2

2.4

2

–

0

0

ANA

7

7

21.8

3.5

0

2

0

0

1 1

GD –

7 7

7 7

21 –

3.9 –

0 0

– –

0 0

0 0

–

3

–

7

0

14.1

4

0

–

0

4

67.5

74

1

–

7

7

37

4

2

–

–

–

141.1

25.4

29.2

3

GD

7

0

7.1

4

0

2

0

4

137.9

31.4

32.6

3

GD

7

0

9.7

3.5

0

–

–

–

Male mass

Female mass

(g)

(g)

Haematopus fuliginosus

740.3

778.5

290.4

289.9

71.3

81.4

1

AA

7

7

69.5

2

2

Haematopus longirostris

602.3

626.3

274.7

280.9

73.8

84.8

1

AA

7

7

49

2.5

Haematopus moquini

668

730

275

279

63.2

71.6

1

AA

7

7

55.8

Haematopus ostralegus

500

536

254

255

69.6

78.4

1

AA

7

7

Haematopus palliatus

–

–

254.4

259.5

83

92.3

1

–

7

Haematopus unicolor

717

734

273

279

82.8

91.8

1

AA

Himantopus himantopus

164

157

247

232

63.7

61.1

1

Himantopus mexicanus Himantopus novaezelandiae

170.4 219

168.7 227

227 –

217.5 –

65.3 –

63.9 –

Hydrophasianus chirurgus Ibidorhyncha struthersii

126

231

–

–

–

–

–

233.5

255

Irediparra gallinacea

84

143

121.8

Jacana jacana

108.3

142.8

124.9

(mm)

(mm)

References 15;15;15;15;15;15;62;7; 62;62;3;62;62; _;10;10 10;10;10;10;10;10;62;7; 62;62;18;53;18; _;10;10 1;1;1;1;1;1;63;1; 63;63;1;1;1;_;1;1 8;8;13;13;13;13;64;15; 64;64;43;13;13; 13;13;13 _;_;15;15;15;15;3;_; 65;65;18;66;3;_;6;6 10;10;10;10;10;10;10;_; 3;3;10;10;10;_;10;10 10;10;13;13;13;13;13;13; 13;13;43;13;13;13; 13;13 6 10;10;_;_;_;_;67;_;67; 67;_;_;10;_;10;10 68;68;_;_;_;_;53;_;69; 69;3;69;70;_;53;53 _;_;3;3;3;3;71;_; 71;71;3;3;3;_;_;_ 10;10;10;10;10;10;7;7; 7;7;_;_;10;7;10;10 3;3;3;3;3;3;72;73; 72;72;72;73;3;_;_;_

319

Jacana spinosa

86.9

145.4

120.3

135.3

17.8

19.3

3

GD

7

0

8.3

4

0

2

0

4

Larosterna inca

–

–

281.3

270.6

45.3

44

–

GD

–

–

–

–

–

2

–

–

Larus argentatus

977

813

425

404

53.2

48.9

1

GD

7

7

87.5

2.7

2

–

1

0

Larus atricilla

294

306

321

312

40.2

38.9

1

GD

7

7

–

–

2

2

0

0

Larus audoinii

–

–

408

393

49.6

44.5

1

ANA

7

7

67

2.6

2

2

0

0

Larus bulleri

243

218

296.8

286.6

38.3

35.9

1

GD

7

7

–

1.85

2

–

–

Larus cachinnans Larus californicus Larus canus Larus cirrocephalus

1275 841 413 352

1033 710 360 297

– 412.2 360 324

– 391.3 341 306

– 53 34.5 37.7

– 47.8 31.1 35.2

– 1 1 1

– GD GD GD

– 7 7 7

– 7 7 7

– – 51 –

– – 2.46 2.4

– 2 2 2

– – 2 2

– – 1 –

– – 0 –

Larus crassirostris

591

522

379

368

46.7

46

–

–

–

–

–

–

–

2

–

–

Larus delawarensis

566

471

382

368

41

37.2

1

GD

7

7

58.6

2

2

0

0

Larus dominicanus

1080.7

935.3

429

414

57.4

52.5

1

GD

7

7

84.2

1.8

2

2

0

0

Larus fuscus Larus genei

880 275

755 236

427 310

406 295

52.1 43.2

47.7 39.9

1 1

GD ANA

7 7

7 7

– –

2.71 2.15

2 2

2 2

0 0

0 0

Larus glaucescens

1180

946.1

405.5

392

63.3

57

1

GD

7

7

95.7

–

2

2

0

0

Larus glaucoides Larus glaucoides thayeri

– 1093

– 899

417 –

403 –

43.6 –

41.9 –

– –

– –

– –

– –

– –

– –

– –

1.5 –

– –

– –

15;15;15;15;15;15;74;13; 74;74;3;3;75;7;6;6 _;_;20;20;20;20;_;76;_;_; _;_;_;7;_;_ 13;13;13;13;13;13;13;13; 13;13;1;1;6;_;13;13 13;13;1;1;1;1;6;6; 6;6;_; _;6;13;6;6 _;_;13;13;13;13;13;13; 13;13;13;1;13; 13;13;13 12;12;12;12;12;12;12;12; 12;12;_;12;_;7;_;_ 16 6 13 1;1;13;13;13;13;13;13; 13;13;_;1;13;13;_;_ 12;12;12;12;12;12;_;_; _; _;_;_;_;7;_;_ 6;6;1;1;1;1;6;6; 6;6;6; _;6;13;6;6 1;1;1;1;1;1;12;1; 12;12;12;1;1;7;12;12 13 13;13;13;13;13;13;13;13; 13;13;_;1;13;13;13; 13 6;6;6;6;6;6;6;6; 6;6;6; _;6;7;6;6 13 16;16

(Continued)

APPENDIX. TABLE 1. (Continued) Male mass

Female mass

(g)

(g)

Male Female wing wing length length

Male bill length

Female bill length

Social mating

Male display

Male

Female

Egg mass

Clutch

Chick

Breeding

Male

Female

(mm)

(mm)

(mm)

system

type

care

care

(g)

size

feeding

habitat

polygamy

polygamy

(mm)

References

320

Larus hartlaubii

–

–

280

266

33.8

32.5

–

–

–

–

–

1.8

–

2

–

–

_;_;1;1;1;1;_;_;_;_;_;1; _;7;_;_

Larus heermanni Larus hemprichii

– 558

– 461

347.3 343

337.7 330

– 50.1

– 46.9

– –

– GD

– 7

– 7

58.6

– 2.42

2 2

– 2

– –

– –

Larus hyperboreus

1576

1249

473

456

62.7

58.2

1

GD

7

7

112.4

3

2

2

0

0

6 1;1;1;1;1;1;_;1;1;1; _;1;1;13;_;_ 13;13;13;13;13;13; 13;6;13;13;6;13;6; 13;13;13

Larus ichthyaetus Larus leucopthalmus Larus livens Larus marinus Larus melanocephalus Larus minutus

1245 357 – 1806 – 99

1143 303 – 1407 – 98

505 325 427.6 499 309 225

472 309 414.4 474 303 218

64.3 48.8 21.77 65.5 33.8 22

55.8 44.5 19.46 59.8 34 21.8

– – 1 1 1 1

GD – – ANA GD ANA

7 – 7 7 7 7

7 – 7 7 7 7

112 – 99.66 – – –

2.22 2 – 2.9 3 2.17

2 – 2 2 2 2

2 2 – 2 2 2

– – 0 0 0 0

– – 0 0 0 0

13 1 6 13 13 13;13;13;13;13;13;13;13; 13;13;_;13;_;13;6;6

Larus modestus

–

–

332.2

318.7

41.6

39

–

GD

–

–

–

–

–

0

–

–

Larus novaehollandiae

313

264

298

288

36.5

34.4

1

GD

7

7

40.2

2

2

2

0

0

Larus occidentalis

972.99 799.78

423.1

398

56.5

51.8

1

GD

7

7

–

–

2

–

0

0

_;_;20;20;20;20;_;76;_;_; _;_;_;7;_;_ 12;12;12;12;12;12;12;12; 12;12;12;12; 12;7;12;12 6

Larus pacificus

1550

1077

461.7

436.7

61.6

53.8

1

–

–

–

–

2.2

–

2

–

–

Larus philadelphia

189

170

267

262

28.9

27.7

1

–

7

7

–

–

2

2

–

–

Larus pipixcan

281

279

288

284

29.6

29.5

–

GD

7

7

35.8

–

2

2

0

0

Larus relictus

519

463

–

–

–

–

–

–

–

–

–

–

–

–

–

–

12;12;12;12;12;12;12;_; _;_;_;12;_;7;12;12 13;13;13;13;13;13;6;_; 6;6;_;_;6;13;_;_ 12;12;12;12;12;12;_;6; 6;6;6;_;6;13;6;6 16

Larus ridibundus

294

267

306

295

33.4

31.6

1

ANA

7

7

–

2.54

2

2

0

0

13

321

Limicola falcinellus Limnodromus griseus

– 111

– 116

105 145

110 144

30.4 58.8

33.5 56.5

1 1

AA ANA

7 7

4 3

9.1 17.5

3.9 4.1

0 0

2 –

0 0

0 0

13 13;13;13;13;13;13;3;6;

Limnodromus scolopaceus Limnodromus semipalmatus Limosa fedoa

100

109

141

146

62.1

72.4

1

–

7

2

–

3.9

0

–

–

–

–

–

–

–

–

–

–

–

7

4

–

–

–

–

–

–

77;77;3;3;6;_;6;6 13;13;13;13;13;13;7;_; 7;7;_;6;6;_;_;_ 115

320

421

228.9

239.9

96.6

114.6

1

–

6

7

44.5

4.1

0

–

0

0

3;3;3;3;15;15;78;_;

Limosa haemastica

222

289

209

217.4

74.5

89.5

1

AA

7

7

37.5

4

0

–

–

–

Limosa lapponica

313

354

210

223

80.4

97.7

1

AA

7

7

37

3.72

0

–

0

0

Limosa limosa

264

315

207

218

92.1

107

1

AA

7

7

39

3.9

0

2

0

0

Lymnocryptes minimus Metopidius indicus

53.7 176.2

46.7 282.4

114 162

110 189

40.3 22.5

40.1 25.1

2 3

AA GD

0 7

7 0

– 11.9

– 4

– –

– –

– –

– –

Micropalama

55.8

60.4

132

134

39.5

41.3

1

AA

7

4

11.2

3.9

0

1

0

0

himantopus Microparra capensis

–

–

86.8

90.6

15.6

16.2

1

GD

7

7

4.5

2.8

0

2

–

–

Numenius americanus

640.1

758.6

279.3

291.3

145.3

184

1

AA

7

4

73

4

0

–

0

0

Numenius arquata

662

788

292

310

118

137

1

AA

7

6

76

3.9

0

2

0

0

78;78;3;3;_;_;6;6 3;3;3;3;15;15;3;20; 79;79;3;79;3;_;_;_ 13;13;13;13;13;13;13;13; 13;13;13;6;13; _;13;13 13;13;13;13;15;15;13;13; 80;80;18;13; 13;13;13;13 13 81;81;81;81;81;81;3;82; 82;82;3;3;_;_;_;_ 13;13;13;13;15;15;83;83; 83;83;3;3;3;13;6;6 _;_;1;1;1;1;84;84; 84;84;3;84;19;7;_;_ 13;13;13;13;15;15;6;6; 48;48;3;3;19;_;6;6 13;13;13;13;15;15;13;13; 85;85;3;13;13;13; 13;13

Numenius borealis Numenius madagascariensis Numenius minutus

– 697.2

– 807.9

206.3 308.5

203.8 310.1

51.3 158

54.9 179.5

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

6 12

162

182

183

188

41.6

45.3

–

–

–

–

–

–

–

–

–

–

13

(Continued)

APPENDIX. TABLE 1. (Continued) Male mass (g)

Female mass (g)

Male wing length (mm)

Female wing length (mm)

Male bill length (mm)

Female bill length (mm)

Social mating system

Numenius phaeopus

368

398

242

251

78.6

86.9

1

Numenius tahitiensis

378

489

230

245

82.8

77.2

Numenius tenuirostris

–

–

251

262

72.9

Pagophila eburnea

617

507

345

336

35.3

Pedionomus torquatus

54

72.4

88.9

96.5

Peltohyas australis Phalaropus fulicaria

80 50.8

88.2 61

141.8 129

Phalaropus lobatus

32.4

37.4

Philomachus pugnax

199

Pluvialis apricaria

Male display type

Female care

Egg mass (g)

Clutch size

Chick feeding

Breeding Male Female habitat polygamy polygamy

AA

7

6

50

3.9

0

–

0

0

1

AA

7

5

54.8

4

0

–

0

0

89.9

–

–

–

–

–

–

–

–

–

–

32.7

1

GD

7

7

1.69

2

1

0

0

12.3

12.8

3

GD

7

0

10

3.6

0

0.5

0

3

140 137

17.2 21.6

17 22.8

1 3

– ANA

7 7

7 0

– 7.5

– 3.8

– 0

– –

– 0

– 2

108

114

21.1

21.4

3

ANA

7

0

6.3

4

0

2

0

3

118

191

158

35

30.9

2

GD

0

7

21

3.7

1

2

4

0

175

176

190

190

21.8

21.6

1

ANA

7

6

32.8

3.9

0

1

0

0

Pluvialis dominica

145

146

182

182

23.2

23.3

1

–

7

7

26

4

0

–

0

0

Pluvialis fulva

130

140

165.9

168.7

24.1

23.5

1

–

7

6

–

–

0

–

0

0

322

Male care

References 13;13;13;13;15;15;13;15; 85;85;18;13;13; _;13;13 3;3;3;3;15;15;86;6; 86;86;3;3;6;_;6;6 13 13;13;13;13;13;13;13;13; 13;13;_;13;6;13;13; 13 16;16;10;10;10;10;88;10; 88;88;18;88;10;7; 10;10 10 13;13;13;13;15;15;13;13; 89;89;3;13;13;_;6;6 13;13;13;13;15;15;13;13; 90;90;3;13;13;13;6;6 13;13;13;13;13;13;13;13; 13;13;3;13;13;13; 13;13 13;13;13;13;15;15;13;13; 43;43;3;13;13;13; 13;13 3;3;3;3;15;15;91;_; 91;91;18;3;91;_;6;6 6;6;10;10;10;10;6;_; 6;6; _;_;6;_;6;6

Pluvialis squatarola

240.4

197

199

28.9

29

1

ANA

7

4

34.2

4

0

–

0

0

Pluvianellus socialis

89.1

79.5

137

137.4

22.2

20.6

1

–

7

7

11.3

2

2

–

–

–

Pluvianus aegyptius Ptychoramphus aleuticus Recurvirostra americana

– – 323.4

– – 309.8

139 – 222.4

139 – 217.7

– – 89.6

– – 88.8

1 1 1

– GD –

– – 7

– – 7

9.2 – 28.7

2.4 – 3.7

– – 0

– – –

0 0 0

0 0 0

Recurvirostra avosetta

–

–

226

225

86.1

78.3

1

GD

7

7

31.7

3.9

0

2

0

0

323

Recurvirostra novaehollandiae Rhinoptilus africanus

325.9

322.8

233.3

232

92.6

87.8

–

GD

–

–

–

–

–

–

–

–

86

88

–

–

–

–

1

GD

7

7

10.4

1

2

–

0

0

Rhinoptilus chalcopterus

151

155.2

180

180

–

–

–

–

–

–

–

2.5

–

–

–

–

Rhinoptilus cinctus Rhodostethia rosea Rissa brevirostris

125 – 400

– – 382

163.5 271 –

163.5 260 –

– 19.3 –

– 17.8 –

– – 1

– – GD

7 – 7

7 – 7

– – 49.4

2 – –

– – 2

– 2 –

– – 0

– – 0

Rissa tridactyla

443

393

311

307

34.2

33.7

1

GD

7

7

–

2.01

2

1.5

0

0

Rostratula benghalensis

146

159

129.4

139.7

48.9

50.6

3

GD

7

0

12.4

4

1

2

0

4

Rostratula semicollaris

–

–

103.9

107

42.3

42.7

1

–

7

7

11.3

2

–

2

–

–

Rynchops flavirostris Rynchops niger

164 349

164 254

336 380

334 342

– 69.5

– 56.4

1 1

– GD

7 7

7 7

– 26.9

1.8 1.8

2 2

– 2

0 0

0 0

Scolopax minor

168.4

210.5

128

142

63.9

71.5

2

AA

2

7

17

3.9

1

1

4

0

15;15;15;15;15;15;13;13; 13;13;3;13;13; _;13;13 92;92;92;92;92;92;3;_; 92;92;3;92;3;_;_;_ 13 6 3;3;3;3;3;3;93;_; 93;93;3;3;93;_;6;6 _;_;13;13;13;13;94;13; 94;94;3;13;13;13; 13;13 10;10;10;10;10;10;_;7; _; _;_;_;_;_;_;_ 1;1;_;_;_;_;1;1; 95;95;18;95;18;_;1;1 1 1 13 16;16;_;_;_;_;6;6; 6;6;6; _;6;_;6;6 13;13;13;13;13;13;13;13; 13;13;13;13;13;13;6; 6 13;13;13;13;15;15;13;10; 96;96;3;13;3;13;13 _;_;21;21;15;15;97;_; 97;97:18;3;_;7;_;_ 7 16;16;20;20;20;20;98;98; 98;98;6;1;98;7;6;6 99;99;3;3;6;6;99;15; 99;99;3;99;99;7;6;6

(Continued)

APPENDIX. TABLE 1. (Continued) Male mass

Female mass

(g)

(g)

Male Female wing wing length length

Male Female bill bill length length

Social mating

Male display

(mm)

(mm)

system

type

(mm)

(mm)

Male Female care

care

Egg mass

Clutch

Chick

Breeding

(g)

size

feeding

habitat

Male

Female

polygamy polygamy

Scolopax rusticola

306

313

202

200

70.9

73.5

2

ANA

0

7

24.5

3.8

1

1

4

0

Steganopus tricolor

50.2

68.1

125

136

30.7

33.6

3

GD

7

0

9.4

4

0

–

0

3

Stercorarius longicaudus

269.9

307.2

30.6

28.4

–

–

1

–

–

–

39.2

–

2

–

0

0

Stercorarius parasiticus

361

408

315

321

31.1

31.8

1

AA

7

7

–

2

2

1.5

0

0

324

Stercorarius pomarinus

648

740

363

373

39.8

40.9

1

AA

7

7

66

1.95

2

–

0

0

Sterna albifrons

60

55

181

175

30.2

28.7

1

–

7

7

10

2.05

2

–

0

0

Sterna aleutica Sterna anaethetus

– –

– –

271.8 261.8

271.5 258.6

– –

32.9 –

1 1

GD/ ANA –

– 7

– 7

19.6 20

– 1

2 2

– –

– 0

– 0

Sterna antillarum

46.9

49.3

166.9

166.4

28.2

26.7

1

ANA

7

7

8.1

2

1.5

0

0

Sterna balaenarum

–

–

171

171

29.3

29.3

1

–

7

7

–

1

2

–

0

0

Sterna bengalensis

–

–

–

–

–

–

1

–

7

7

35

1

2

–

–

–

Sterna bergii

–

–

351.1

344.2

61.2

57.7

1

ANA

7

7

–

1

2

–

0

0

Sterna caspia

680

588

421

412

72.4

67.8

1

ANA

7

7

65

1.54

2

2

0

0

Sterna dougallii

–

–

236

233

–

–

1

ANA

7

7

20.4

1.43

2

–

0

0

References 13;13;13;13;15;15;100;13; 100;100;13;13; 13;13;_;_ 13;13;13;13;15;15;101;6; 102;102;3;3;6;_;6;6 6;6;6;6;_;_;6;_; _;_;6; _;13;_;13;13 13;13;13;13;13;13;13;13; 13;13;13;13;13;7; 13;13 13;13;13;13;13;13;13;15; 13;13;13;13;13; _;13;13 9;9;9;9;9;9;9;_; 13;13;9;9;13;_;9;9 6 _;_;12;12;_;_;6;_; 12;12;9;1;12;_;9;9 6;6;6;6;6;6;6;6; 6;6;6; _;6;7;6;6 1 _;_;_;_;_;_;12;_; 1;1;9;1;12;_;_;_ _;_;12;12;12;12;9;12; 9;9;_;1;12;_;9;9 13;13;13;13;13;13;13;13; 13;13;9;1;13;9;9;9 _;_;9;9;_;_;12;12; 9;9;9;1;9;_;12;12

325

Sterna elegans

–

–

316.9

312.7

64.3

60.8

–

–

–

–

36.48

1.02

2

2

–

–

Sterna fuscata

–

–

294

287

–

–

1

–

7

7

40.2

1

2

–

0

0

Sterna hirundo

124

126

272

270

37.1

35.2

1

ANA

7

7

21

2.65

2

2

0

0

Sterna maxima

–

–

–

–

–

–

1

–

7

7

–

1

–

0

0

Sterna nereis

72.2

72.4

186.8

182.4

–

–

–

–

7

7

10.2

1.77

2

–

–

–

Sterna nilotica

–

–

–

–

–

–

–

–

–

–

–

2

–

–

0

0

Sterna paradisea Sterna repressa

112 –

117 –

279 247

274 250

37.4

35.7

1 –

ANA –

7 –

7 –

19 19

2.4

2 2

– –

0 0

0 0

Sterna sandvicensis

–

–

309

304

–

–

1

7

7

35

2

2

–

0

0

Sterna saundersi Sterna striata Sterna sumatrana Sterna vittata Stiltia isabella Synthliboramphus

– 125.8 106 – 66 206.3

– 131.9 106 – 65 205.7

168 282.1 228.1 – 203.3 141.4

165 277.5 228.6 – 194.7 142

29.6 – – – 17.2 –

28.3 – – – 16.7 –

1 1 1 1 1 1

– ANA ANA – GD –

– 7 7 7 7 7

– 7 7 7 7 7

– – 16.1 20 7 –

– 1.07 1.5 1.29 1.86 –

– 2 2 2 2 2

– – – – – –

0 0 0 0 – –

0 0 0 0 – –

1 12 12 12 12 6 6;6;5;5;5;5;6;_; 5;5;5;_;5; _;_;_ 103;103;103;103;15;15; 103;103;103;103;18; 103;103;_;_;_ 103;103;103;103;15;15;

antiquus Synthliboramphus hypoleucos Thinocorus orbignyianus

166.1

173.2

120.3

122.5

17

18

1

–

7

7

37.2

–

2

–

–

–

134

125

139.9

135.8

16.1

15.9

1

ANA

0

7

15.5

4

0

–

–

–

Thinocorus rumicivorus

49

60

114

115.6

14

14.3

1

ANA

0

7

8

4

0

0

–

–

Thinornis novaeseelandiae

–

–

121.2

121.1

23.9

22.7

1

AA

7

7

–

–

–

–

0

0

_;_;20;20;6;6;_;_;_; _;6;6;6;7;_;_ _;_;9;9;_;_;6;_; 6;6;12;6;6;_;12;12 13;13;13;13;13;13;9;9; 9;9;13;13;6;9;9;9 _;_;_;_;_;_;6;_; 6;6;_;1;_; _;9;9 12 _;_;_;_;_;_;_;_; _;_;_;1;_; _;9;9 9 _;_;9;9;9;9;_;_; _;_;9;1;1; _;1;1 _;_;9;9;_;_;9;_; 9;9;9;1;9; _;9;9

103;103;103;103;18; 103;103;7;_;_ _;_;10;10;10;10;10;7; 10;10;_;_;_;_;10;10

(Continued)

APPENDIX. TABLE 1. (Continued)

326

Male mass (g)

Female mass (g)

Male wing length (mm)

Female wing length (mm)

Male Female bill bill length length (mm) (mm)

Social mating system

Tringa cinerea Tringa erythropus

69.4 142

74.8 161

132 167

137 170

45.2 57.4

49.2 59.8

1 1

Tringa flavipes

80

83.7

159

163

36

36.5

Tringa glareola

62

73

126

129

28.3

Tringa hypoleucos

45.5

50

112

112

Tringa incana

101

116

163

Tringa macularia

36.9

48

Tringa melanoleuca

164

Tringa nebularia

Male display type

Male Female care care

Egg mass (g)

Clutch size

Chick Breeding Male Female feeding habitat polygamy polygamy

– AA

– 7

– 3

– 24.5

– 4

– 0

– –

0 0

0 1

1

AA

7

7

17.4

4

0

–

0

0

29.3

1

AA

7

4

13.5

4

0

–

0

0

24.6

24.5

1

ANA

7

5

12.5

3.9

0

2

0

1

165

38.5

39.9

–

–

7

7

–

–

0

–

–

–

105

109

23.2

24.1

3

GD

7

0

9

4

0

2

0

4

176

196

199

55.8

56.7

1

AA

7

7

27.9

3.7

0

–

–

–

172

175

191

193

55.1

57

1

AA

7

6

30.5

3.9

0

–

1

0

Tringa ochropus

75

85

144

146

34.5

34.6

1

AA

7

4

15.5

3.9

0

–

0

0

Tringa solitaria

45.1

52.1

131

136

29.2

29.6

1

AA

–

–

–

–

–

–

0

0

Tringa stagnatilis

67.1

76

139

142

39.5

40.4

1

AA

7

7

14

4

0

–

0

0

References 13 13;13;13;13;13;13;13;13; 104;104;3;[51];13; _;13;13 13;13;13;13;13;13;6;57; 6;6;6;_;6;_;6;6 13;13;13;13;13;13;13;13; 13;13;3;8;13;_;13;13 13;13;13;13;15;15;13;13; 105;105;3;43;13;13; 13;13 12;12;12;12;12;12;_;_; 6;6;_;_;6;_;_;_ 13;13;13;13;15;15;13;15; 106;106;3;13;13;13; 13;13 6;6;13;13;13;13;6;57; 6;6;6;6;6;_;_;_ 13;13;13;13;13;13;3;13; 43;43;3;13;13; _;13;13 3;3;13;13;13;13;13;13; 107;107;3;43;13; _;13;13 13;13;13;13;13;13;6;6; _; _;_;_;_;_;6;6 13;13;13;13;13;13;13;13; 13;13;3;107;13; _;13;13

327

Tringa totanus

123

135

159

161

41.7

42.7

1

AA

7

6

22.3

4

0

2

0

0

Tryngites subruficollis

70.5

53

136

129

20.1

18.5

2

GD

0

7

13

4

0

1

4

0

Uria aalge

981

985

207

209

44.6

43.2

1

GD

7

7

108

–

2

2

0

0

Uria lomvia

961

928

224

223

14

13.6

1

–

7

7

98.5

1

2

–

0

0

Vanellus albiceps Vanellus armatus

– 162

– 167

218 207

218 207

33.4 27.7

33 27.6

1 1

– ANA

7 7

7 7

– 16.5

– 3.1

0 0

– –

0 0

0 0

Vanellus chilensis

–

–

232

230.1

–

–

1

–

7

7

27.8

3

0

–

–

–

Vanellus coronatus

171

156

196

195

31.1

30.4

1

–

7

7

17

2.6

0

–

1

1

Vanellus crassirostris

170

170

211

207

–

–

1

–

7

7

19.6

3

0

–

0

0

Vanellus gregarius

252

200

206

204

29.4

28.7

1

–

7

7

26.5

4.2

0

–

0

0

Vanellus indicus Vanellus leucurus

– –

– –

240 179

230 177

33.7 29.4

34.3 29.1

– –

AA –

– –

– –

– –

– –

– –

– –

0 –

0 –

Vanellus lugubris

109.5

113

178

171

22.2

20.7

1

–

7

7

13.7

3

0

–

0

0

Vanellus malabaricus

–

–

195.5

193

–

–

1

–

7

7

13.5

4

0

–

–

–

Vanellus melanocephalus

–

–

233

234

25

25.3

–

–

–

–

–

–

–

–

–

–

Vanellus melanopterus

–

–

210

205

26.1

26.4

1

–

7

7

18.2

3

–

–

0

0

13;13;13;13;13;13;13;13; 13;13;18;13;13;13; 13;13 13;13;13;13;15;15;108; 15;108;108;3;3;6; 13;6;6 5;5;5;5;5;5;9;9; 9;9;13; _;9;9;9;9 5;5;5;5;5;5;9;_; 9;9;9;13;9;_;9;9 1 1;1;1;1;1;1;1;1; 109;109;1;1;1;_;1;1 _;_;3;3;_;_;110;_; 110;110;3;110; 110;_;_;_ 1;1;1;1;1;1;1;_; 111;111;1;1;1;_;1;1 1;1;1;1;_;_;1;_; 110;110;1;1;1;_;1;1 13;13;13;13;13;13;13;_; 13;13;3;13;13; _;13;13 13 13 1;1;1;1;1;1;1;_; 111;111;1;1;1;_;1;1 _;_;3;3;_;_;3;_; 112;112;3;3;112;_;_; _ 1 _;_;1;1;1;1;1;_; 111;111;1;1;_;_;1;1

(Continued)

APPENDIX. TABLE 1. (Continued) Male Female wing wing length length

Male Female bill bill length length

Social mating

Male display

(mm)

(mm)

system

type

328

Male mass

Female mass

(g)

(g)

Vanellus miles

264.2

252.8

227.2

223.2

36.1

35.2

1

ANA

7

7

32

3.7

0

–

0

0

Vanellus senegallus

–

–

236

237

34

33.9

1

–

7

7

23.3

3

0

–

0

0

Vanellus spinosus

191.5

183.8

203

200.7

27.2

26.7

1

ANA

7

7

16.4

4

0

2

0

0

Vanellus superciliosus Vanellus tectus Vanellus tricolor

– – –

– – –

– 191 196.3

– 190 193.4

– 24.3 22.5

– 23.3 23

1 1 1

GD – ANA

7 – 7

7 – 7

– – 21.6

– – 3.75

– – 0

– – –

0 0 0

0 0 0

Vanellus vanellus

211

226

229

224

24.1

23.9

2

AA

7

6

25.5

3.9

0

2

1

1

(mm)

(mm)

Male Female care

care

Egg mass

Clutch

Chick

Breeding

(g)

size

feeding

habitat

Male

Female

polygamy polygamy

References 10;10;10;10;10;10;113;7; 113;113;3;113;113; _;10;10 1;1;1;1;1;1;1;_; 114;114;1;1;1;_;1;1 13;13;13;13;13;13;13;13; 13;13;3;13;13;13;13; 13 1 1 _;_;10;10;10;10;10;7; 10;10;10;10;10; _;10;10 13;13;13;13;13;13;13;13; 43;43;13;13;13; 13;13;13

Xema sabini a

194

177

276

267

24.9

24.1

1

ANA

7

7

–

2

1

2

0

0

13

Scored variables are defined as follows: social mating system (1, monogamous; 2, polygynous; 3, polyandrous); male display type (GD, ground display; ANA, aerial nonacrobatic; AA, aerial acrobatic); male care and female care (scored independently for males and females: 0, no care; 1–3, provides care during incubation; 4–6, provides post‐hatching care; 7, full care until hatching; see Sze´kely and Reynolds, 1995 and Reynolds and Sze´kely, 1997 for full details); chick feeding (0, chicks self‐feed from hatching; 1, chicks self‐feed after approximately 7–10 days; 2, chicks fed by parents until fledging); breeding habitat (0, desert and

329

semidesert; 1, dry grassland and tundra, and dry forest; 2, wetland, marsh, seashore, lake, and river); male and female polygamy (scored independently for males and females: 0, monogamous; 1, rare polygamy (<1% or only anecdotal reports of polygamy); 2, occasional polygamy (1–5%, polygamy is known to occur but it is infrequent); 3, moderate polygamy (6–20%, polygamy is well known but is not regarded as typical of the species); 4, frequent polygamy (>20%, polygamy is considered the main mating system for the species); Underscores in reference column indicate no data available; see Thomas et al. (2006) for full details). (1) Urban et al. (1986); (2) Tarboton (1976); (3) Johnsgard (1981); (4) Pitman (1960); (5) Gaston and Jones (1998); (6) Poole and Gill (1992–2003); (7) del Hoyo et al. (1996); (8) Dementiev and Gladkov (1969); (9) Cramp (1985); (10) Marchant and Higgins (1993); (11) Hay (1979); (12) Higgins and Davies; (1996); (13) Cramp and Simmons (1983); (14) Nettleship (1973); (15) Jehl and Murray (1986); (16) Dunning (1993); (17) Handel and Gill (2000); (18) Scho¨nwetter (1967); (19) Walters (1984); (20) Blake (1977); (21) Hayman et al. (1986); (22) Anderson (1991); (23) Westwood (1983); (24) Brown (1948); (25) Ali and Ripley (1996); (26) Myers et al. (1982); (27) Parmelee and Payne (1973); (28) Soikkeli (1967); (29) Nettleship and Maher (1973); (30) Holmes and Pitelka (1964); (31) Parmelee et al. (1968); (32) Holmes (1973); (33) Holmes (1972); (34) Pitelka (1959); (35) Miller (1985); (36) Miller (1983); (37) Prater et al. (1977); (38) Gratto‐Trevor (1991); (39) Tomkovich (1989); (40) Hilde´n (1975); (41) Howe (1982); (42) Szekely and Lessells (1993); (43) Nethersole‐Thompson and Nethersole‐Thompson (1986); (44) Reiser and Hein (1974); (45) Graul (1973b); (46) Graul (1975); (47) McCulloch (1992); (48) Lenington (1984); (49) Bergstrom (1981); (50) Jones (1963); (51) von Blotzheim et al. (1975); (52) Burger (1981); (53) Pringle (1987); (54) Miskelly (1990); (55) Pullainen (1970); (56) Tomkovich (1990); (57) Figuerola (1999); (58) Tuck (1972); (59) Saether et al. (1986); (60) Webster (1941); (61) Groves (1984); (62) Wakefield (1988); (63) Summers and Cooper (1977); (64) Harris (1967); (65) Nol (1985); (66) Nol et al. (1984); (67) Pierce (1986); (68) Thong‐aree et al. (1995); (69) Hoffman (1949); (70) Roberts (1991); (71) Kelso (1972); (72) Osborne (1982); (73) Osborne and Bourne (1977); (74) Jenni and Collier (1972); (75) Jenni and Betts (1978); (76) Moynihan (1962); (77) Kitchinski and Flint (1973); (78) Nowicki (1973); (79) Hagar (1968); (80) Lind (1961); (81) Butchart (2000); (82) Mathew (1964); (83) Jehl (1973); (84) Tarboton and Fry (1986); (85) Grant, M.C. (in literature); (86) Gill et al. (1991); (87) Baker‐Gabb et al. (1990); (88) Bennett (1983); (89) Kistchinski (1975); (90) Hilde´n and Vuolanto (1972); (91) Parmelee et al. (1967); (92) Jehl (1975); (93) Gibson (1971); (94) Brown (1950); (95) Maclean (1967); (96) Kobayashi (1955); (97) Ho¨hn (1975); (98) Burger and Gochfeld (1990); (99) Mendall and Aldous (1943); (100) Hirons (1983); (101) Colwell (1986); (102) Ho¨hn (1967); (103) Maclean (1969); (104) Hilde´n (1979); (105) Yalden and Holland (1992); (106) Oring (1986); (107) Glutz von Blotzheim et al. (1977); (108) Pitelka et al. (1974); (109) Hall (1964); (110) Walters (1982); (111) Ward (1989); (112) Tuljapurkar (1986); (113) Barlow et al. (1972); (114) Little (1967); (115) Melnikov, J. I. (personal communication).

330

GAVIN H. THOMAS ET AL.

Acknowledgments We thank Clemens Ku¨pper, David Lank, Michael Weston, and an anonymous reviewer for insightful comments on the paper. G.H.T. is a NERC funded research associate at the Centre for Population Biology, Imperial College London. T.S. was supported by NERC (GR3/10957, NE/C004167/1), BBSRC (BBS/B/05788), a Research Fellowship by The Leverhulme Trust (RF/2/RFG/2005/0279), and a Hrdy Visiting Fellowship of Harvard University.

References Alcock, J. (2005). ‘‘Animal Behavior: An Evolutionary Approach, Eighth Edition.’’ Sinauer Associates, Sunderland. Alerstam, T. (1990). ‘‘Bird Migration.’’ Cambridge University Press, Cambridge. Ali, S., and Ripley, S. D. (1996). ‘‘Handbook of the Birds of India and Pakistan.’’ Oxford University Press, Oxford. Alonzo, S. H., and Warner, R. R. (2000). Dynamic games and field experiments examining intra‐ and intersexual conflict: Explaining counterintuitive mating behavior in a Mediterranean wrasse, Symphodus ocellatus. Behav. Ecol. 11, 56–70. Amat, J. A., and Masero, J. A. (2004). How Kentish plovers, Charadrius alexandrinus, cope with heat stress during incubation. Behav. Ecol. Sociobiol. 56, 26–33. Amat, J. A., Fraga, R. M., and Arroyo, G. M. (1999). Brood desertion and polygamous breeding in the Kentish Plover Charadrius alexandrinus. Ibis 141, 596–607. Anderson, G. J. (1991). The breeding biology of the bush thick‐knee Burhinus magnirostris and notes on its distribution in the Brisbane area. Sunbird 21, 33–61. Andersson, M. (2005). Evolution of classical polyandry: Three steps to female emancipation. Ethology 111, 1–23. Andersson, M. B. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton. Arnqvist, G., and Rowe, L. (2002). Antagonistic coevolution between the sexes in a group of insects. Nature 415, 787–789. Arnqvist, G., and Rowe, L. (2005). ‘‘Sexual Conflict.’’ Princeton University Press, Princeton. Arnqvist, G., Edvardsson, M., Friberg, U., and Nilsson, T. (2000). Sexual conflict promotes speciation in insects. Proc. Natl. Acad. Sci. USA 97, 10460–10464. Baker, A. J., Pereira, S. L., and Paton, T. A. (2007). Phylogenetic relationships and divergence times of Charadriiformes genera: Multigene evidence for the Cretaceous origin of at least 14 clades of shorebirds. Biol. Lett. 3, 205–209. Baker‐Gabb, D. J., Benshemesh, J. S., and Maher, P. N. (1990). A revision of the distribution, status and management of the plains‐wanderer Pedionomus torquatus. Emu 90, 161–168. Balshine, S., Kempenaers, B., and Sze´kely, T. (2002). Introduction. Philos. Trans. R. Soc. Lond. B 357, 237–240. Barlow, M. L., Muller, P. M., and Sutton, R. R. (1972). Breeding data on the spur‐winged plover in southland, New Zealand. Notornis 19, 212–249. Barraclough, T. G., Vogler, A. P., and Harvey, P. H. (1998). Revealing the factors that promote speciation. Philos. Trans. R. Soc. Lond. B 353, 241–249. Basolo, A. L. (1990). Male swords and female preferences. Science 250, 808–810. Basolo, A. L. (1995). Phylogenetic evidence for the role of a pre‐existing bias in sexual selection. Proc. R. Soc. Lond. B 259, 307–311. Bateman, A. J. (1948). Intra‐sexual selection in Drosophila. Heredity 2, 349–368. Bennett, A. T. D., and Cuthill, I. C. (1994). Ultraviolet vision in birds: What is the function? Vision Res. 34, 1471–1478.

PARENTAL CONFLICT IN SHOREBIRDS

331

Bennett, P. M., and Owens, I. P. F. (2002). ‘‘Evolutionary Ecology of Birds: Life Histories, Mating Systems and Ecology.’’ Oxford University Press, Oxford. Bennett, S. (1983). A review of the distribution, status and biology of the plains‐wanderer Pedionomus torquatus, Gould. Emu 83, 1–11. Bergsten, J., Toyra, A., and Nilsson, A. N. (2001). Intraspecific variation and intersexual correlation in secondary sexual characters of three diving beetles (Coleoptera: Dytiscidae). Biol. J. Linn. Soc. 73, 221–232. Bergstrom, P. W. (1981). Male incubation in Wilson’s plover (Charadrius wilsonia). Auk 98, 835–838. Birkhead, T. R., and Møller, A. P. (1992). ‘‘Sperm Competition in Birds.’’ Academic Press, London. Birkhead, T. R., Hatchwell, B. J., Linder, R., Blomqvist, D., Pellatt, E. J., Griffiths, R., and Lifjeld, J. T. (2001). Extra‐pair paternity in the common murre. Condor 103, 158–162. Blake, E. R. (1977). ‘‘Manual of Neotropical Birds, Volume I.’’ University of Chicago Press, Chicago. Blomqvist, D., Johansson, O. C., Unger, U., Larsson, M., and Flodin, L.‐A. (1997). Male aerial display and reversed sexual size dimorphism in the dunlin. Anim. Behav. 54, 1291–1299. Blomqvist, D., Andersson, M., Ku¨pper, C., Cuthill, I., Kis, J., Lanctot, R. B., Sandercock, B. K., Sze´kely, J., Wallander, J., and Kempenaers, B. (2002a). Genetic similarity between mates and extra‐pair parentage in three species of waders. Nature 419, 613–615. Blomqvist, D., Kempenaers, B., Lanctot, R. B., and Sandercock, B. K. (2002b). Genetic parentage and mate guarding in the Arctic‐breeding western sandpiper. Auk 119, 228–233. Bo´kony, V., Liker, A., Sze´kely, T., and Kis, J. (2003). Melanin‐based plumage coloration and flight displays in plovers and allies. Proc. R. Soc. Lond. B 270, 2491–2497. Brown, L. H. (1948). Notes on birds of the Katba, Ilorin and N. Benin provinces of Nigeria. Ibis 90, 525–538. Brown, P. E. (1950). ‘‘Avocets in England.’’ RSPB Occasional Publication No. 14, R Soc. Protect. Birds, London. Bukacinska, M., Bukacinska, D., Epplen, J. T., Sauer, K. P., and Lubjuhn, T. (1998). Low frequency of extra‐pair paternity in Common Gulls (Larus canus) as revealed by DNA fingerprinting. J. Ornithol. 139, 413–420. Burger, A. E. (1981). Time budgets, energy needs and kleptoparasitism in breeding lesser sheathbills (Chionis minor). Condor 83, 106–112. Burger, J., and Gochfeld, M. (1990). ‘‘The Black Skimmer. Social Dynamics of a Colonial Species.’’ Columbia University Press, New York. Burke, T., Davies, N. B., Bruford, M. W., and Hatchwell, B. J. (1989). Parental care and mating behaviour of polyandrous dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature 338, 249–251. Burt, A., and Trivers, R. (2006). ‘‘Genes in Conflict: The Biology of Selfish Genetic Elements.’’ Harvard University Press, Cambridge, MA. Butchart, S. H. M. (1999). Sexual conflicts and copulation patterns in polyandrous bronxe‐ winged jacanas (Metopidius indicus). Behaviour 136, 443–468. Butchart, S. H. M. (2000). Population structure and breeding system of the sex‐role reversed, polyanduous Bronze‐winged Jacana Metopidius indicus. Ibis 142, 93–102. Butchart, S. M., Seddon, N., and Ekstrom, J. M. M. (1999). Yelling for sex: Harem males compete for female access in bronze‐winged jacanas. Anim. Behav. 57, 637–646. Cartar, R. V. (1985). Testis size in sandpipers: The fertilization hypothesis. Naturwissenschaften 72, 157–158. Chapman, T., Arnqvist, G., Bangham, J., and Rowe, L. (2003). Sexual conflict. Trends Ecol. Evol. 18, 41–47.

332

GAVIN H. THOMAS ET AL.

Chapman, T., Liddle, L. F., Kalb, J. M., Wolfner, M. F., and Partridge, L. (1995). Costs of mating in Drosophila melanogaster females is mediated by male accessory gland products. Nature 373, 241–244. Clutton‐Brock, T. H. (1991). ‘‘The Evolution of Parental Care.’’ Princeton University Press, Princeton. Colwell, M. A. (1986). The first documented case of polyandry for Wilson’s Phalarope (Phalaropus tricolor). Auk 103, 611–612. Colwell, M. A., and Oring, L. W. (1989). Extra‐pair paternity in the spotted sandpiper: A female mate acquisition tactic. Anim. Behav. 38, 675–684. Cramp, S. (1985). ‘‘The Birds of the Western Palearctic, Volume IV.’’ Oxford University Press, Oxford. Cramp, S., and Simmons, K. E. L. (1980). ‘‘The Birds of the Western Palearctic, Volume II.’’ Oxford University Press, Oxford. Cramp, S., and Simmons, K. E. L. (1983). ‘‘The Birds of the Western Palearctic, Volume III.’’ Oxford University Press, Oxford. Dale, J. (2006). Intraspecific variation in bird colors. In ‘‘Bird Coloration, Volume 2, Function and Evolution’’ (G. E. Hill and K. J. McGraw, Eds.). Harvard University Press, Cambridge, MA. Dale, J., Montgomerie, R., Michaud, D., and Boag, P. (1999). Frequency and timing of extrapair fertilization in the polyandrous red phalarope (Phalaropus fulicarius). Behav. Ecol. Sociobiol. 46, 50–56. Davies, N. B. (1991). Mating systems. In ‘‘Behavioural Evology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), 3rd ed., pp. 471–487. Blackwell Sci. Publ., Oxford. Davies, N. B. (1992). ‘‘Dunnock Behaviour and Social Evolution.’’ Oxford University Press, Oxford. Darwin, C. (1871). ‘‘The Descent of Man and Selection in Relation to Sex.’’ J. Murray, London. Delehanty, D. J., Fleischer, R. C., Colwell, M. A., and Oring, L. W. (1998). Sex‐role reversal and the absence of extra‐pair paternity in Wilson’s phalaropes. Anim. Behav. 55, 995–1002. Dementiev, G. P., and Gladkov, N. A. (1969). ‘‘Birds of the Soviet Union, Volume 3.’’ Israel Programme for Scientific Translations Ltd., Jerusalem. Dunning, J. B. J. (1993). ‘‘CRC Handbook of Avian Body Masses.’’ CRC Press, Boca Raton. Edwards, P. J. (1982). Plumage variation, territoriality and breeding displays of the Golden Plover Pluvialis aprocaria in Southwest Scotland. Ibis 124, 88–95. Emlen, S. T., and Oring, L. W. (1977). Ecology, sexual selection and evolution of mating systems. Science 197, 215–223. Emlen, S. T., and Wrege, P. H. (2004). Size dimorphism, intrasexual competition, and sexual selection in Wattled Jacana (Jacana jacana), a sex‐role‐reversed shorebird in Panama. Auk 121, 391–403. Emlen, S. T., Demong, N. J., and Emlen, D. J. (1989). Experimental induction of infanticide in female wattled jacanas. Auk 106, 1–7. Emlen, S. T., Wrege, P. H., and Webster, M. S. (1998). Cuckoldry as a cost of polyandry in the sex‐role‐reversed wattled jacana, Jacana jacana. Proc. R. Soc. Lond. B 265, 2359–2364. Ens, B. J., Kersten, M., brenninkmeijer, A., and Hulscher, J. B. (1992). Territory quality, parental effort and reproductive success of oystercatchers (Haematopus Ostralegus). J. Anim. Ecol. 61, 703–715. Ericson, P. G. P., Envall, I., Irestedt, M., and Norman, J. A. (2003). Inter‐familial relationships of the shorebirds (Aves: Charadriiformes) based on nuclear DNA sequence data. BMC Evolutionary Biology 3, 16. Erckmann, W. J. (1981). ‘‘The Evolution of Sex‐role Reversal and Monogamy in Shorebirds.’’ PhD thesis, University of Washington.

PARENTAL CONFLICT IN SHOREBIRDS

333

Erckmann, W. J. (1983). The evolution of polyandry in shorebirds: An evaluation of hypotheses. In ‘‘Social Behaviour of Female Vertebrates’’ (S. K. Wasser, Ed.), pp. 113–168. Academic Press, New York. Fain, M. G., and Houde, P. (2004). Parallel radiations in the primary clades of birds. Evolution 58, 2558–2573. Fairbairn, D. J. (1997). Allometry for sexual size dimorphism: Pattern and process in the coevolution of body size in males and females. Annu. Rev. Ecol. Syst. 28, 659–687. Figuerola, J. (1999). A comparative study on the evolution of reversed sexual size dimorphism in monogamous waders. Biol. J. Linn. Soc. 67, 1–18. Fisher, R. A. (1930). ‘‘The Genetical Theory of Natural Selection.’’ Clarendon Press, Oxford. Franz, D. (1991). Mating system and strategy of reproduction in penduline tit Remiz pendulinus. J. Ornithol. 132, 241–266. Gage, M. J. G., Parker, G. A., Nylin, S., and Wiklund, C. (2002). Sexual selection and speciation in mammals, butterflies and spiders. Proc. R. Soc. Lond. B 269, 2309–2316. Gaston, A. J., and Jones, I. L. (1998). ‘‘The Auks.’’ Oxford University Press, Oxford. Gavrilets, S. (2000). Rapid evolution of reproductive barriers driven by sexual conflict. Nature 403, 886–889. Gavrilets, S., and Waxman, D. (2002). Sympatric speciation by sexual conflict. Proc. Natl. Acad. Sci. USA 99, 10533–10538. Gibson, F. (1971). The breeding biology of the American avocet (Recurvirostra americana) in Central Oregon. Condor 73, 444–454. Gilbert, L., Burke, T., and Krupa, A. (1998). No evidence for extra‐pair paternity in the western gull. Mol. Ecol. 7, 1549–1552. Gill, R. E., Lanctot, R. B., Mason, J. D., and Handel, C. M. (1991). Observations on habitat use, breeding chronology and parental care in Bristle‐thighed Curlews on the Seward Peninsula, Alaska. Wader Study Group Bull. 61, 28–36. Glutz von Blotzheim, U.N., Bauer, K.M., and Bezzel E. (1977). ‘‘Handbuch der vo¨gel Mltteleuropas, Vol. 7.’’ Wiesbaden, Germany: Akademische Verlagsgesellschaft. Gowaty, P. A. (1996). Battles of the sexes and origins of monogamy. In ‘‘Partnerships in Birds: The Study of Monogamy’’ (J. M. Black, Ed.), pp. 21–52. Oxford University Press, Oxford. Gratto‐Trevor, C. (1991). Parental care in semipalmated sandpipers Calidris pusilla: Brood desertion by females. Ibis 133, 394–399. Graul, W. D. (1973a). Possible functions of head and breast markings in Charadriinae. Wilson Bull. 85, 60–70. Graul, W. D. (1973b). Adaptive aspects of the mountain plover social system. Living Bird 12, 69–94. Graul, W. D. (1975). Breeding biology of the mountain plover. Wilson Bull. 87, 6–31. Griffith, S. C., Owens, I. P. F., and Thuman, K. A. (2002). Extra pair paternity in birds: A review of interspecific variation and adaptive function. Mol. Ecol. 11, 2195–2212. Griggio, M., Matessi, G., and Marin, G. (2004). No evidence of extra‐pair paternity in a colonial seabird, the common tern (Sterna hirundo). Ital. J. Zool. 71, 219–222. Grønstøl, G. B. (1996). Aerobatic components in the song‐flight display of male Lapwings Vanellus vanellus as cues in female choice. Ardea 84, 45–55. Groves, S. (1984). Chick growth, sibling rivalry, and chick production in American black oystercatchers. Auk 101, 525–531. Hagar, J. A. (1968). Nesting of the Hudsonian godwit at Churchill, Manitoba. Living Bird 5, 5–43. Haig, D. (2004). Genomic imprinting and kinship: How good is the evidence? Ann. Rev. Genet. 38, 553–585.

334

GAVIN H. THOMAS ET AL.

Hall, K. R. L. (1964). A study of the blacksmith plover (Hoplopterus armatus (Burchell)) in the Cape Town area. Ostrich 35, 3–16. Hamilton, W. D. (1990). Mate choice near and far. Am. Zool. 30, 341–352. Handel, C. M., and Gill, R. E. (2000). Mate fidelity and breeding site tenacity in a monogamous sandpiper, the black turnstone. Anim. Behav. 60, 471–481. Harris, M. P. (1967). The biology of oystercatchers Haematopus ostralegus on Skokholm Island, S. Wales. Ibis 109, 180–193. Hay, R. (1979). The wrybill—one of the world’s ornithological oddities. Forest and Bird 213, 13–17. Hayman, P., Marchant, J., and Prater, T. (1986). ‘‘Shorebirds. An Identification Guide to the Waders of the World.’’ Christopher Helm, London. Heg, D., Ens, B. J., Burke, T., Jenkins, L., and Krujt, J. P. (1993). Why does the typically monogamous oystercatcher (Haematopus ostralegus) engage in extra‐pair copulations? Behaviour 126, 247–289. Hegyi, Z., and Sasva´ri, L. (1998). Parental condition and breeding effort in waders. J. Anim. Ecol. 67, 41–53. Higgins, P. J., and Davies, S. J. J. F. (1996). ‘‘Handbook of Australian, New Zealand and Antarctic Birds, Volume III.’’ Oxford University Press, Melbourne. Hilde´n, O. (1975). Breeding system of Temminck’s Stint Calidris temminckii. Ornis Fenn. 52, 117–146. Hilde´n, O. (1979). The timing of arrival and departure of the Spotted Redshank Tringa erythropus in Finland. Ornis Fenn. 56, 18–23. Hilde´n, O., and Vuolanto, S. (1972). Breeding biology of the Red‐necked Phalarope Phalaropus lobatus in Finland. Ornis Fenn. 49, 57–85. Hirons, G. (1983). A five‐year study of the breeding behaviour and biology of the woodcock in England—A first report. In ‘‘Proceedings of the 2nd European Woodcock and Snipe Workshop’’ (H. Kalchreuter, pp. 51–67). IWRB, Slimbridge. Hockey, P., Douie, C., and Barlow, A. (1995). ‘‘Waders of Southern Africa.’’ Struik Winchester, Cape Town. ¨ ber die Brutflege des polyandrischen Wasserfasans, Hydrophasianus Hoffman, A. (1949). U chirurgus (Scop.). Zool. Jahrb. 78, 323–470. Ho¨hn, E. O. (1967). Observations on the breeding biology of Wilson’s Phalarope (Steganopus tricolor) in Central Alberta. Auk 84, 220–244. Ho¨hn, E. O. (1975). Notes on black‐headed ducks, painted snipe, and spotted tinamous. Auk 92, 566–575. Holland, B., and Rice, W. R. (1998). Perspective: Chase‐away sexual selection: Antagonistic seduction versus resistance. Evolution 52, 1–7. Holland, B., and Rice, W. R. (1999). Experimental removal of sexual selection reverses intersexual antagonistic coevolution and removes a reproductive load. Proc. Natl. Acad. Sci. USA 96, 5083–5088. Holmes, R. T. (1972). Ecological factors influencing the breeding season schedule of Western Sandpipers (Calidris mauri) in subartic Alaska. Amer. Midland Natur. 87, 472–491. Holmes, R. T. (1973). Social behaviour of breeding western sandpiper Calidris mauri. Ibis 115, 107–123. Holmes, R. T., and Pitelka, F. A. (1964). Breeding behavior and taxonomic relationships of the Curlew Sandpiper. Auk 81, 362–379. Houston, A. I., and Davies, N. B. (1985). The evolution of cooperation and life history in the dunnock Prunell modularis. In ‘‘Behavioral Ecology: Ecological Consequences of Adaptive Behavior’’ (R. M. Sibly and R. H. Smith, Eds.), pp. 471–487. Blackwell Sci. Publ., Oxford.

PARENTAL CONFLICT IN SHOREBIRDS

335

Houston, A. I., Sze´kely, T., and McNamara, J. M. (2005). Conflict between parents over care. Trends Ecol. Evol. 20, 33–38. Howe, M. A. (1982). Social organization in a nesting population of eastern willets (Catoptrophus semipalmatus). Auk 99, 88–102. Hoyo, J. D., Elliott, A., and Sargatal, J. (1996). ‘‘Handbook of the Birds of the World, Volume 4.’’ Lynx Edicions, Barcelona. Hrdy, S. B. (1979). Infanticide among animals: A review, classification, and examination of the implications for the reproductive strategies of females. Ethol. Sociobiol. 1, 13–40. Jehl, J. R. Jr. (1973). Breeding biology and systematic relationships of the stilt sandpiper. Wilson Bull. 85, 115–147. Jehl, J. R. Jr. (1975). Pluvianellus socialis: Biology, ecology, and relationships of an enigmatic Patagonian shorebird. Trans. San Diego Soc. Natur. Hist. 18, 25–73. Jehl, J. R. Jr., and Murray, B. G. Jr. (1986). The evolution of normal and reverse sexual dimorphism in shorebirds and other birds. In ‘‘Current Ornithology’’ (R. F. Johnston, Ed.), Vol. 3, pp. 1–86. Plenum Press, New York. Jenni, D. A., and Betts, B. J. (1978). Sex differences in nest construction, incubation, and parental behaviour in the polyandrous American jacana (Jacana spinosa). Anim. Behav. 26, 207–218. Jenni, D. A., and Collier, G. (1972). Polyandry in the American Jacana (Jacana spinosa). Auk 89, 743–765. Jennions, M. D. (1997). Female promiscuity and genetic incompatibility. Trends Ecol. Evol. 12, 251–253. Johnsgard, P. A. (1981). ‘‘The Plovers, Sandpipers, and Snipes of the World.’’ University of Nebraska Press, Lincoln and London. Johnson, D. D. P., and Briskie, J. V. (1999). Sperm competition and sperm length in shorebirds. Condor 101, 848–854. Jones, N. V. (1963). The sheathbill Chionis alba (Gmelin) at Signy Island, South Orkney Islands. Br. Antarct. Surv. Bull. 2, 53–71. Jones, A. G., Rosenqvist, G., Berglund, A., Arnold, S. J., and Avise, J. C. (2000). The Bateman gradient and the cause of sexual selection in a sex‐role‐reversed pipefish. Proc. R. Soc. Lond. B 267, 677–680. Kam, J. van de, Ens, B., Piersma, T., and Zwarts, L. (2004). ‘‘Shorebirds. An Illustrated Behavioural Ecology.’’ KNNV Publishers, Utrecht. Kelso, L. (1972). The Ibisbill—Ibidorhyncha struthersii Vigors. Fauna Tadzhik Sov. Soc. Rep. XIX, 239–245. Kempenaers, B., Congdon, B., Boag, P., and Robertson, R. J. (1999). Extrapair paternity and egg hatchability in tree swallows: Evidence for the genetic compatibility hypothesis? Behav. Ecol. 10, 304–311. Kistchinski, A. A. (1975). Breeding biology and behaviour of the Grey Phalarope Phalaropus fulicarius in east Siberia. Ibis 117, 285–301. Kitchinski, A. A., and Flint, V. E. (1973). On the biology of the dowitcher (Limnodromus griseus) in the tundras of East Siberia. In ‘‘Fauna and Ecology of Waders’’ (V. E. Flint, Ed.), pp. 52–55. Nauka, Moscow. Kobayashi, H. (1955). Observations of the Painted Snipe, Rostratula bengalensis. Tori 14, 1–13. Kokko, H., and Jennions, M. (2003). It takes two to tango. Trends Ecol. Evol. 18, 103–104. Kosztola´nyi, A., Sze´kely, T., and Cuthill, I. C. (2003). Why do both parents incubate in the Kentish plover? Ethology 109, 645–658. Ku¨pper, C., Kis, J., Kosztola´nyi, A., Sze´kely, T., Cuthill, I. C., and Blomqvist, D. (2004). Genetic mating system and timing of extra‐pair fertilizations in the Kentish plover. Behav. Ecol. Sociobiol. 57, 32–39.

336

GAVIN H. THOMAS ET AL.

Lack, D. L. (1968). ‘‘Ecological Adaptations for Breeding in Birds.’’ Methuen, London. Lanctot, R. B., and Weatherhead, P. J. (1997). Ephemeral lekking behavior in the buff‐ breasted sandpiper, Tryngites subruficollis. Behav. Ecol. 8, 268–278. Lanctot, R. B., Scribner, K. T., Kempenaers, B., and Weatherhead, P. J. (1997). Lekking without a paradox in the buff‐breasted sandpiper. Am. Natur. 149, 1051–1070. Lanctot, R. B., Weatherhead, P. J., Kempenaers, B., and Scribner, K. T. (1998). Male traits, mating tactics and reproductive success in the buff‐breasted sandpiper, Tryngites subruficollis. Anim. Behav. 56, 419–432. Lanctot, R. B., Sandercock, B. K., and Kempenaers, B. (2000). Do male breeding displays function to attract mates or defend territories? The explanatory role of mate and site fidelity. Waterbirds 23, 155–164. Lande, R. (1980). Sexual dimorphism, sexual selection, and adaptation in polygenic characters. Evolution 34, 292–305. Lande, R. (1981). Models of speciation by sexual selection on polygenic traits. Proc. Natl. Acad. Sci. USA 78, 3721–3725. Lank, D. B., and Smith, C. M. (1987). Conditional lekking in ruff (Philomachus pugnax). Behav. Ecol. Sociobiol. 20, 137–145. Lank, D. B., Oring, L. W., and Maxson, S. J. (1985). Mate and nutrient limitation of egg‐laying in a polyandrous shorebird. Ecology 66, 1523–1524. Lank, D. B., Smith, C. M., Hanotte, O., Ohtonen, A., Bailey, S., and Burke, T. (2002). High frequency of polyandry in a lek mating system. Behav. Ecol. 13, 209–215. Lenington, S. (1984). The evolution of polyandry in shorebirds. In Shorebirds. Breeding behavior and populations. In ‘‘Behavior of Marine Animals’’ (J. Burger and B. L. Olla, Eds.), Vol. 5, pp. 149–167. Plenum Press, New York. Lessells, C. M. (1983). The mating system of Kentish Plovers Charadrius alexandrinus: Some observations and experiments. Wader Study Group Bull. 39, 43. Lessells, C. M. (1999). Sexual conflict in animals. In ‘‘Levels of Selection in Evolution’’ (L. Keller, Ed.), pp. 75–99. Princeton University Press, Princeton. Lifjeld, J. T., Harding, A. M. A., Mehlum, F., and Øigarden, T. (2005). No evidence of extra‐ pair paternity in the little auk Alle alle. J. Avian Biol. 36, 484–487. Ligon, J. D. (1999). ‘‘The Evolution of Avian Breeding Systems.’’ Oxford University Press, Oxford, New York. Liker, A., and Sze´kely, T. (1997). Aggression among female lapwings, Vanellus vanellus. Anim. Behav. 54, 797–802. Liker, A., and Sze´kely, T. (1999). Parental behavior in the Lapwing Vanellus vanellus. Ibis 141, 608–614. Liker, A., and Sze´kely, T. (2005). Mortality costs of sexual selection and parental care in natural populations of birds. Evolution 59, 890–897. Lind, H. (1961). ‘‘Studies on the Behaviour of the Black‐Tailed Godwit (Limosa limosa (L.)).’’ Ph. D thesis, University of Copenhagen. Lindenfors, P., Sze´kely, T., and Reynolds, J. D. (2003). Directional changes in sexual size dimorphism in shorebirds, gulls and alcids. J. Evol. Biol. 16, 930–938. Lislevand, T., and Thomas, G. H. (2006). Limited male incubation ability and the evolution of egg size in shorebirds. Biol. Lett. 2, 206–208. Little, J. D. V. (1967). Some aspects of the behaviour of the Wattled Plover Afribyx senegallus (Linnaeus). Ostrich 38, 259–280. Maclean, G. L. (1967). The breeding biology and behaviour of the Double‐banded Courser Rhinoptilus africanus (Temm.). Ibis 109, 556–569. Maclean, G. L. (1969). A study of seedsnipe in southern South America. Living Bird 8, 33–80. Maclean, G. L. (1972). Cluth size and evolution in the Charadrii. Auk 89, 299–324.

PARENTAL CONFLICT IN SHOREBIRDS

337

Marchant, S., and Higgins, P. J. (1993). ‘‘Handbook of Australasian, New Zealand and Antarctic Birds, Volume II.’’ Oxford University Press, Melbourne. Martin, O. Y., and Hosken, D. J. (2003). The evolution of reproductive isolation through sexual conflict. Nature 423, 979–982. Mathew, D. N. (1964). Observations on the breeding habits of the Bronze‐winged Jacana [Metopidius indicus (Latham)]. J. Bombay Nat. Hist. Soc. 68, 295–301. McCulloch, N. (1992). The status and ecology of the St. Helena Wirebird. BTO Res. Rep. 97, 1–89. McNamara, J. M., Sze´kely, T., Webb, J. N., and Houston, A. I. (2000). A dynamic game‐ theoretic model of parental care. J. Theor. Biol. 205, 605–623. McNamara, J. M., Houston, A. I., Sze´kely, T., and Webb, J. N. (2002). Do parents make independent decisions about desertion? Anim. Behav. 64, 147–149. Mee, A., Whitfield, D. P., Thompson, D. B. A., and Burke, T. (2004). Extrapair paternity in the common sandpiper Actitis hypoleucos, revealed by DNA fingerprinting. Anim. Behav. 67, 333–342. Mendall, H. L., and Aldous, C. M. (1943). ‘‘The Ecology and Management of the American Woodcock.’’ Maine Cooperatuve Wildlife Research Unit, Orono, Maine. Millar, C. D., Anthony, I., Lambert, D. M., Stapleton, P. M., Bergmann, C. C., Bellamy, A. R., and Young, E. C. (1994). Patterns of reproductive success determined by DNA fingerprinting in a communally breeding oceanic bird. Biol. J. Linn. Soc. 52, 31–48. Millar, C. D., Lambert, D. M., and Young, E. C. (1997). Minisatellite DNA detects sex, parentage, and adoption in the south polar skua. J. Hered. 88, 235–238. Miller, E. H. (1983). Habitat and breeding cycle of the Least Sandpiper (Calidris pusilla) on Sable Island, Nova Scotia. Can. J. Zool. 61, 2880–2898. Miller, E. H. (1985). Parental behaviour in the Least Sandpiper (Calidris minutilla). Can. J. Zool. 63, 1593–1601. Mills, S. C., Grapputo, A., Koskela, E., and Mappes, T. (2007). Quantitative measure of sexual selection with respect to the operational sex ratio: A comparison of selection indices. Proc. R. Soc. Lond. B 274, 143–150. Miskelly, C. M. (1990). Breeding systems of New Zealand Snipe Coenocorypha aucklandica and Chatham Island Snipe C. pusilla; are they food limited? Ibis 132, 366–379. Monroe, B. L. Jr., and Sibley, C. G. (1993). ‘‘A World Checklist of Birds.’’ Yale University Press, New Haven. Moynihan, M. (1962). Hostile and sexual behavior patterns of South American and Pacific Laridae. Behaviour (Supplement VIII), 1–365. Myers, J. P., Hilde´n, O., and Tomkovich, P. (1982). Exotic Calidris species of the Siberian tundra. Ornis Fenn. 59, 175–182. Nebel, S. (2007). Differential migration of shorebirds in the East Asian‐Australasian Flyway. Emu 107, 14–18. Nethersole‐Thompson, D., and Nethersole‐Thompson, M. (1986). ‘‘Waders: Their Breeding, Haunts and Watchers.’’ Poyser, Calton. Nettleship, D. N. (1973). Breeding ecology of turnstone Arenaria interpres at Hazen Camp, Ellesmere Island, N.W.T. Ibis 115, 202–217. Nettleship, D. N., and Maher, W. J. (1973). The avifauna of Hazen Camp, Ellesmere Island, N. W. T. Polarforschung 43, 66–74. Nol, E. (1985). Sex roles in the American oystercatcher. Behaviour 95, 232–260. Nol, E., Baker, A. J., and Cadman, M. D. (1984). Clutch intitiation dates, clutch size, and egg size of the American oystercatcher in Virginia. Auk 101, 855–867. Nowicki, T. (1973). ‘‘A Behavioral Study of the Marbled Godwit in North Dakota.’’ M.Sc. thesis, Central Michigan University.

338

GAVIN H. THOMAS ET AL.

Olson, V. A., and Owens, I. P. F. (2005). Interspecific variation in the use of carotenoid‐based coloration in birds: Diet, life history and phylogeny. J. Evol. Biol. 18, 1534–1546. Orians, G. H. (1969). On evolution of mating systems in birds and mammals. Am. Natur. 103, 589–603. Oring, L. W. (1986). Avian polyandry: A review. In ‘‘Current Ornithology’’ (R. F. Johnston, Ed.), Vol. 3, pp. 309–351. Plenum Press, New York. Oring, L. W., and Lank, D. B. (1986). Polyandry in spotted sandpipers: The impact of environment and experience. In ‘‘Ecological Aspects of Social Evolution: Birds and Mammals’’ (D. I. Rubenstein and R. W. Wrangham, Eds.), pp. 21–34. Princeton University Press, Princeton. Oring, L. W., Fleischer, R. C., Reed, J. M., and Marsden, K. E. (1992). Cuckoldry through stored sperm in the sequentially polyandrous spotted sandpiper. Nature 359, 631–633. Oring, L. W., Reed, J. M., and Maxson, S. J. (1994). Copulation patterns and mate guarding in the sex‐role reversed, polyandrous spotted sandpiper. Anim. Behav. 47, 1065–1072. Osborne, D. R. (1982). Replacement nesting and polyandry in the Wattled jacana. Wilson Bull. 94, 206–208. Osborne, D. R., and Bourne, G. R. (1977). Breeding behavior and food habits of the Wattled jacana. Condor 79, 98–105. Osorno, J. L., and Sze´kely, T. (2004). Sexual conflict and parental care in magnificent frigatebirds: Full compensation by deserted females. Anim. Behav. 68, 337–342. Owens, I. P. F. (2002). Male only parental care and classical polyandry in birds: Phylogeny, ecology and sex differences in remating opportunities. Phil. Trans. R. Soc. Lond. B 357, 283–293. Owens, I. P. F., Burke, T., and Thompson, D. B. A. (1994). Extraordinary sex‐roles in the Eurasian dotterel—female mating arenas, female‐female competition, and female mate choice. Am. Natur. 144, 76–100. Owens, I. P. F., Dixon, A., Burke, T., and Thompson, D. B. A. (1995). Strategic paternity in the sex‐role reversed Eurasian dotterel (Charadrius morinellus): Behavioral and genetic evidence. Behav. Ecol. 6, 14–21. Packer, C., and Pusey, A. E. (1984). Infanticide in carnivores. In ‘‘Infanticide: Comparative and Evolutionary Perspectives’’ (G. Hausfater and S. Blaffer Hrdy, Eds.), pp. 31–42. Princeton University Press, Princeton. Pagel, M. D. (1994). Detecting correlated evolution on phylogenies—a general method for the comparative‐analysis of discrete characters. Proc. R. Soc. Lond. B 255, 37–45. Panhuis, T. M., Butlin, R., Zuk, M., and Tregenza, T. (2001). Sexual selection and speciation. Trends Ecol. Evol. 16, 364–371. Parker, G. A. (1979). Sexual selection and sexual conflict. In ‘‘Sexual Selection and Reproductive Competition in Insects’’ (M. S. Blum and N. A. Blum, Eds.), pp. 123–166. Academic Press, New York. Parker, G. A. (2006). Sexual conflict over mating and fertilization: An overview. Phil. Trans. R. Soc. Lond. B 361, 235–260. Parker, G. A., and Partridge, L. (1998). Sexual conflict and speciation. Phil. Trans. R. Soc. Lond. B 353, 261–274. Parker, G. A., Royle, N. J., and Hartley, I. R. (2002). Intrafamilial conflict and parental investment: A synthesis. Phil. Trans. R. Soc. Lond. B 357, 295–307. Parmelee, D. F., and Payne, R. B. (1973). On multiple broods and the breeding strategy of Arctic sanderlings. Ibis 115, 218–226. Parmelee, D. F., Stephens, H. A., and Schmidt, R. H. (1967). The birds of southern eastern Victoria Island and adjacent small islands. Nat. Mus. Can. Bull. 222, 1–229.

PARENTAL CONFLICT IN SHOREBIRDS

339

Parmelee, D. F., Greiner, D. W., and Graul, W. D. (1968). Summer schedule and breeding biology of the White‐rumped Sandpiper in the Canadian arctic. Wilson Bull. 80, 5–29. Paton, T. A., and Baker, A. J. (2006). Sequences from 14 mitochondrial genes provide a well‐ supported phylogeny of the Charadriiform birds congruent with the nuclear RAG‐1 tree. Mol. Phyl. Evol. 39, 657–667. Paton, T. A., Baker, A. J., Groth, J. G., and Barrowclough, G. F. (2003). RAG‐1 sequences resolve phylogenetic relationships within Charadriiform birds. Mol. Phyl. Evol. 29, 268–278. Payne, R. B. (1984). Sexual selection, lek and arena behavior, and sexual size dimorphism in birds. Ornith. Monogr. 33, 1–53. Pierce, R. (1986). ‘‘Black Stilt.’’ New Zealand Wildlife Service, Dunedin. Pierce, E. P., and Lifjeld, J. T. (1998). High paternity without paternity‐assurance behavior in the purple sandpiper, a species with high paternal investment. Auk 115, 602–612. Piersma, T., and Morrison, R. I. G. (1994). Energy‐expenditure and water turnover of incubating ruddy turnstones—high costs under high arctic climatic conditions. Auk 111, 366–376. Piersma, T., Lindstrom, A., Drent, R. H., Tulp, I., Jukema, J., Morrison, R. I. G., Reneerkens, J., Schekkerman, J., and Visser, G. H. (2003). High daily energy expenditure of incubating shorebirds on High Arctic tundra: A circumpolar study. Funct. Ecol. 17, 356–362. Pitelka, F. A. (1959). Numbers, breeding schedule, and territoriality in pectoral sandpipers in northern Alaska. Condor 61, 233–264. Pitelka, F. A., Holmes, R. T., and MacLean, S. F. Jr. (1974). Ecology and evolution of social organization in Arctic sandpipers. Am. Zool. 14, 185–204. Pitman, C. R. S. (1960). Lily‐Trotters. Uganda Wildl. Sport 1, 1–9. Poole, A., and Gill, F. (1992–2003). ‘‘The Birds of North America.’’ The Birds of North America Inc., Philadelphia. Prater, A. J., Marchant, J. H., and Vuorinen, J. (1977). ‘‘Guide to the Identification and Aging of Holarctic Waders.’’ BTO Guide 17, British Trust for Ornithology, Tring. Pringle, J. D. (1987). ‘‘The Shorebirds of Australia.’’ Angus and Robertson Publishers, North Ryde, Australia. Pullainen, E. (1970). On the breeding biology of the dotterel Charadrius morinellus. Ornis Fenn. 47, 69–73. Queller, D. C. (1997). Why do females care more than males? Proc. R. Soc. Lond. B 264, 1555–1557. Reiser, K. H., and Hein, K. (1974). Zur Vorkommen und zur Brutbiologie des Flussregenpfeifers (Charadrius dubius) in Schleswig‐Holstein. Corax 5, 9–30. Reynolds, J. D. (1987). Mating system and nesting biology of the red‐necked phalarope Phalaropus lobatus: What constrains polyandry? Ibis 129, 225–242. Reynolds, J. D. (1996). Animal breeding systems. Trends Ecol. Evol. 11, 68–72. Reynolds, J. D., and Sze´kely, T. (1997). The evolution of parental care in shorebirds: Life histories, ecology, and sexual selection. Behav. Ecol. 8, 126–134. Rice, W. R. (1996). Sexually antagonistic male adaptation triggered by experimental arrest of female evolution. Nature 381, 232–234. Rice, W. R. (2000). Dangerous liaisons. Proc. Natl. Acad. Sci. USA 97, 12953–12955. Roberts, T. J. (1991). ‘‘The Birds of Pakistan, Volume 1.’’ Oxford University Press, Karachi. Royle, N. J., Hartley, I. R., and Parker, G. A. (2002). Sexual conflict reduces offspring fitness in zebra finches. Nature 416, 733–736. Ryan, M. J., and Rand, S. (1998). Evoked vocal response in male tu´ngara frogs: Pre‐existing biases in male response? Anim. Behav. 56, 1509–1516.

340

GAVIN H. THOMAS ET AL.

Saether, B. E., Fiske, P., and Ka˚la˚s, J. A. (1999). Pushy males and choosy females: Courtship disruption and mate choice in the lekking great snipe. Proc. R. Soc. Lond. B 266, 1227–1234. Saether, B. E., Ka˚la˚s, J. A., Lofaldli, L., and Andersen, R. (1986). Sexual Size Dimorphism and Reproductive Ecology in Relation to Mating System in Waders. Biol. J. Linn. Soc. 28, 273–284. Schamel, D., and Tracy, D. M. (1977). Polyandry, replacement clutches, and site tenacity in the Red Phalarope (Phalaropus fulicarius) at Barrow, Alaska. Bird‐Banding 48, 314–324. Schamel, D., Tracy, D. M., Lank, D. B., and Westneat, D. F. (2004). Mate guarding, copulation strategies and paternity in the sex‐role reversed, socially polyandrous red‐necked phalarope Phalaropus lobatus. Behav. Ecol. Sociobiol. 57, 110–118. Scho¨nwetter, M. (1967). ‘‘Handbuch der Oologie, Volume I.’’ Akademia Verlag, Berlin. Schuh, R. T., and Stys, P. (1991). Phylogenetic analysis of cimicomorphan family relationships (Heteroptera). J. New York Entomol. Soc. 99, 298–350. Soikkeli, M. (1967). Breeding cycle and population dynamics in the dunlin (Calidris alpina). Ann. Zool. Fenn. 4, 158–198. Stephens, M. L. (1982). Mate takeover and possible infanticide by a female Northern Jacana (Jacana spinosa). Anim. Behav. 30, 1253–1254. Summers, R. W., and Cooper, J. (1977). The population, ecology and conservation of the black oystercatcher Haematopus moquini. Ostrich 48, 28–40. Sutherland, W. J. (1985). Chance can produce a sex difference in mating success and explain Bateman’s data. Anim. Behav. 33, 1349–1352. Sze´kely, T., and Cuthill, I. C. (1999). Brood desertion in Kentish plover: The value of parental care. Behav. Ecol. 10, 191–197. Sze´kely, T., and Reynolds, J. D. (1995). Evolutionary transitions in parental care in shorebirds. Proc. R. Soc. Lond. B 262, 57–64. Sze´kely, T., and Lessells, C. M. (1993). Mate change by Kentish plovers Charadrius alexandrinus. Ornis Scand. 24, 317–322. Sze´kely, T., and Williams, T. D. (1995). Costs and benefits of brood desertion in female Kentish plovers, Charadrius alexandrinus. Behav. Ecol. Sociobiol. 37, 155–161. Sze´kely, T., Cuthill, I. C., and Kis, J. (1999). Brood desertion in Kentish plover: Sex differences in remating opportunities. Behav. Ecol. 10, 185–190. Sze´kely, T., Webb, J. N., and Cuthill, I. C. (2000a). Mating patterns, sexual selection and parental care: An integrative approach. In ‘‘Vertebrate Mating Systems’’ (M. Apollonio, M. Festa‐Bianchet, and D. Mainardi, Eds.), pp. 159–185. World Scientific, Singapore. Sze´kely, T., Reynolds, J. D., and Figuerola, J. (2000b). Sexual size dimorphism in shorebirds, gulls, and alcids: The influence of sexual and natural selection. Evolution 54, 1404–1413. Sze´kely, T., Freckleton, R. P., and Reynolds, J. D. (2004). Sexual selection explains Rensch’s rule of size dimorphism in shorebirds. Proc. Natl. Acad. Sci. USA 101, 12224–12227. Sze´kely, T., Thomas, G. H., and Cuthill, I. C. (2006). Sexual conflict, ecology and breeding systems in shorebirds. Bioscience 56, 801–808. Szentirmai, I., Komdeur, J., and Sze´kely, T. (2005). What makes a nest‐building male successful? Male behavior and female care in penduline tits. Behav. Ecol. 16, 994–1000. Tanaka, Y. (1996). Sexual selection enhances population extinction in a changing environment. J. Theor. Biol. 180, 197–206. Tarboton, W. R. (1976). Notes on South African jacanas. Fauna and Flora 27, 5–7. Tarboton, W. R., and Fry, C. H. (1986). Breeding and other behaviour of the lesser jacana. Ostrich 57, 233–243. Thomas, G. H. (2004). ‘‘Sexual Conflict, Ecology and Breeding Systems in Shorebirds: Phylogenetic Analyses.’’ Ph.D thesis, University of Bath.

PARENTAL CONFLICT IN SHOREBIRDS

341

Thomas, G. H., and Sze´kely, T. (2005). Evolutionary pathways in shorebird breeding systems: Sexual conflict, parental care, and chick development. Evolution 59, 2222–2230. Thomas, G. H., Wills, M. A., and Sze´kely, T. (2004a). Phylogeny of shorebirds, gulls, and alcids (Aves: Charadrii) from the cytochrome‐b gene: Parsimony, Bayesian inference, minimum evolution, and quartet puzzling. Mol. Phyl. Evol. 30, 516–526. Thomas, G. H., Wills, M. A., and Sze´kely, T. (2004b). A supertree approach to shorebird phylogeny. BMC Evol. Biol. 4, 28. Thomas, G. H., Freckleton, R. P., and Sze´kely, T. (2006). Comparative analyses of the influence of developmental mode on phenotypic diversification rates in shorebirds. Proc. R. Soc. Lond. B 273, 1619–1624. Thong‐aree, S., Khobkhet, O., Lauhachinda, V., and Pong‐umpai, S. (1995). Breeding biology of pheasant‐tailed jacana Hydrophasianus chirurgus in Central Thailand. Nat. Hist. Bull. Siam. Soc. 43, 289–302. Thuman, K. (2003). ‘‘Female Reproductive Strategies in the Ruff (Philomachus pugnax).’’ Ph.D thesis. Uppsala University. Tomkovich, P. S. (1989). Parental roles and the mating system of the long‐toed stint Calidris subminuta. Wader Study Group Bull. 57, 42–43. Tomkovich, P. S. (1990). Social organisation in a breeding population of spoon‐billed sandpiper Eurynorhychus pygmaeus. Wader Study Group Bull. 59, 7. Tregenza, T., and Hosken, D. J. (2005). Mate choice: Been there, done that. Curr. Biol. 15, R959–R961. Tregenza, T., and Wedell, N. (2000). Genetic compatibility, mate choice and patterns of parentage: An invited review. Mol. Ecol. 9, 1013–1027. Tregenza, T., Wedell, N., and Chapman, T. (2006). Introduction. Sexual conflict: A new paradigm? Phil. Trans. R. Soc. Lond. B 361, 229–234. Trivers, R. L. (1972). Parental investment and sexual selection. In ‘‘Sexual Selection and the Descent of Man’’ (B. Campbell, Ed.), pp. 136–179. Aladine, Chicago. Tuck, L. M. (1972). ‘‘The Snipes. Monograph Series 5.’’ Canadian Wildlife Service, Ottawa. Tuljapurkar, V. B. (1986). The yellow‐wattled lapwing. Sanctuary (Asia) 6, 236–243. Urban, E. K., Fry, C. H., and Keith, S. (1986). ‘‘The Birds of Africa., Volume 2.’’ Academic Press, London. von Blotzheim, U. N. G., Bauer, K. M., and Bezzel, E. (1975). ‘‘Handbuch der Vogel Mitteleuropas, Volume 6.’’ Akademische Verlagsgesellscahaft, Wiesbaden. Wakefield, W. C. (1988). Breeding resource partitioning of a mixed population of pied and sooty oystercatchers. Stilt 13, 39–40. Wallander, J., Blomqvist, D., and Lifjeld, J. (2001). Genetic and social monogamy—does it occur without mate guarding in the ringed plover. Ethology 107, 561–572. Walters, J. R. (1982). Parental behavior in lapwings (Charadriidae) and its relationships with clutch sizes and mating systems. Evolution 36, 1030–1040. Walters, J. R. (1984). The evolution of parental behavior and clutch size in shorebirds. In ‘‘Shorebirds. Breeding Behaviour and Populations. Behavior of Marine Animals’’ (J. Burger and B. L. Olla, Eds.), Vol. 5, pp. 243–287. Plenum Publishing Corporation, New York. Ward, D. (1989). Behaviour associated with breeding of crowned, blackwinged and lesser blackwinged plovers. Ostrich 60, 141–150. Warriner, J. S., Warriner, J. C., Page, G. W., and Stenzel, L. E. (1986). Mating system and reproductive success of a small population of polygamous snowy plovers. Wilson Bull. 98, 15–37. Webb, J. N., Houston, A. I., McNamara, J. M., and Sze´kely, T. (1999). A game‐theoretic framework for variable patterns of parental care. Anim. Behav. 58, 983–993.

342

GAVIN H. THOMAS ET AL.

Webster, J. D. (1941). The breeding of the black oystercatcher. Wilson Bull. 53, 141–156. Wedell, N., Kvarnemo, C., Lessells, C. K. M., and Tregenza, T. (2006). Sexual conflict and life histories. Anim. Behav. 71, 999–1011. West‐Eberhard, M. J. (1983). Sexual selection, social competition, and speciation. Q. Rev. Biol. 58, 155–183. Westneat, D. F., and Stewart, I. R. K. (2003). Extra‐pair paternity in birds: Causes, correlates, and conflict. Annu. Rev. Ecol. Evol. Syst. 34, 365–396. Westneat, D. F., Sherman, P. W., and Morton, M. L. (1990). The ecology and evolution of extra‐pair copulations in birds. In ‘‘Current Ornithology’’ (D. M. Power, Ed.), pp. 331–369. Plenum Press, New York. Weston, M. A., and Elgar, M. A. (2005). Parental care in hooded plovers Thinornis rubricollis. Emu 105, 283–292. Westwood, N. J. (1983). Breeding stone‐curlews at Weeting Heath, Norfolk. Br. Birds 76, 291–304. Wetton, J. H., and Parkin, D. T. (1991). An association between fertility and cuckoldry in the house sparrow Passer domesticus. Proc. R. Soc. Lond. B 245, 227–233. Whitfield, D. P. (1990). Male choice and sperm competition as constraints on polyandry in the red‐necked phalarope Phalaropus lobatus. Behav. Ecol. Sociobiol. 27, 247–254. Whitfield, D. P. (1995). Behavior and ecology of a polyandrous population of Grey Phalaropes Phalaropus fulicarius in Iceland. J. Avian Biol. 26, 349–352. Whitfield, D. P., and Tomkovich, P. S. (1996). Mating system and timing of breeding in Holarctic waders. Biol. J. Linn. Soc. 57, 277–289. Widemo, F. (1997). The social implications of traditional use of lek sites in the ruff Philomachus pugnax. Behav. Ecol. 8, 211–217. Williams, G. C. (1975). ‘‘Sex and Evolution.’’ Princeton University Press, Princeton. Yalden, D. W., and Holland, P. K. (1992). Relative contributions of Common Sandpiper Actitis hypoleucos parents to guarding their chicks. Ringing and Migration 13, 95–97. Zharikov, Y., and Nol, E. (2000). Copulation behavior, mate guarding, and paternity in the semipalmated plover. Condor 102, 231–235.

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

Postcopulatory Selection in the Yellow Dung Fly Scathophaga stercoraria (L.) and the Mate‐Now‐Choose‐Later Mechanism of Cryptic Female Choice Paul I. Ward zoological museum of the university of zurich, 8057 zurich switzerland

I. INTRODUCTION Postcopulatory sexual selection is the continuation after copula has begun of precopulatory sexual selection and is dominated by the same two processes, male‐male competition and female choice. At the moment, there is controversy within studies of postcopulatory sexual selection over the importance of female choice, which is subtle and much more difficult to study than male‐male competition. It is, however, important that evidence in support of either process should be treated with equal rigor. The debate is similar to the previous one about the relative importance of the two processes in precopulatory sexual selection (Andersson, 1994). Male–male competition, especially precopulatory competition, has led to the evolution of many spectacular and easy‐to‐observe courtship morphologies and behaviors. In many cases, these are involved in the mating process before any aspect of female choice occurs, for example, males compete for the best territories before a female selects a mate either on the basis of the male’s own quality or on that of his territory, or even a combination of the two. This means that studies of female choice must consider the influence of male–male competition, while studies of male–male competition do not necessarily have to consider the influence of female choice if the processes occur before female choice occurs. This makes the study of postcopulatory female choice difficult because a male influence is probably ubiquitous, whereas a female‐influenced process may not occur or may only occur under particular circumstance, for example, if a female can avoid or counterbalance any influences of her male partners, such as attempts to prevent the female mating with other males or disturbing attempts to do so. 0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37007-1

343

Copyright 2007, Elsevier Inc. All rights reserved.

344

PAUL I. WARD

The traditional view of postcopulatory competition is of ‘‘sperm competition’’ (Parker, 1970), that is, the sperm from different males compete within a female for fertilization of her eggs. This view is being increasingly extended by findings that females play an active role after mating in determining the outcome of competition among males to obtain fertilizations, that is, cryptic female choice (Eberhard, 1996). Since the evolutionary or behavioral interests of males and females are seldom identical, essentially because the females of most species mate with more than one male (see Arnqvist and Rowe, 2005), much attention is being paid to the resolution of the different interests, usually referred to as sexual conflict (Parker, 1979; Rice, 1996). As there are a number of extensive reviews available (Andersson and Simmons, 2006; Arnqvist and Rowe, 2005; Bernasconi et al., 2004; Hosken and Snook, 2005), I do not cover general issues here. Instead, I will discuss the particularities of one well‐examined species, the yellow dung fly S. stercoraria (L.), with comments only on the general issues which are directly relevant to findings on S. stercoraria.

II. SEXUAL CONFLICT Parker’s pioneering theoretical work on sexual conflict (Parker, 1979), the importance of which in S. stercoraria will become clear below, received relatively little attention until Rice’s empirical work (Rice, 1996) on the topic restimulated the field (Arnqvist and Rowe, 2005). Arnqvist and Rowe discuss potential reasons for this lack of attention but do not mention that during that period biologists were moving away from using the ‘‘phenotypic gambit’’ (Grafen, 1984) of the early stages in the development of behavioral ecology to guide the precise details of their research. The phenotypic gambit is that we sacrifice the detailed examination of the underlying genetics of behavioral traits and interactions and instead concentrate on an examination of phenotypes and how they might evolve if the genetics were very simple. The disadvantage of the phenotypic gambit when examining reproductive traits of both males and females is that there are too many traits involved in the interactions and they are among the fastest evolving of all traits (Arnqvist and Nilsson, 2000; Eberhard, 1996). Additionally, an evolutionary response in one sex may provoke a response in the other sex in a quite different type of trait, for example, a morphological change in male genitalia may provoke a change in female physiology. This sort of evolutionary interaction may make it extraordinarily difficult (even impossible) to make the precise simplifications necessary for realistic theoretical explorations of the maintenance

POSTCOPULATORY SELECTION

345

of phenotypic variability in reproductive traits. The complexity of the male–female interactions in S. stercoraria described below makes it likely that such processes have occurred in this species. An additional assumption in the theoretical modeling is also often that environmental variation, and the accompanying variation in selection patterns for particular traits, is rather coarsely scaled, that is, a female only encounters one type of environment in her life. Parker (1979) made this assumption and then argued that female choice among male variants is unlikely to be of evolutionary importance. This was because his model predicted a stable cycle of coarse‐scaled habitats, in which males very quickly adapted to changes in the choice variation within the female population. However, if environments are always variable on a small spatial scale, that is, each female is regularly confronted with a variety of environmental conditions in which to place or raise her offspring, then the argument would lose its force. We may then expect that females may consider male variants not as good or bad in themselves but rather good or bad in particular circumstances, that is, male quality from a female perspective may depend not just on the intrinsic quality of the male but rather on an interaction between male genotype and the particular selection pattern offspring can be expected to experience. Thus, although it may be possible for a female, by allowing the males to ‘‘sort themselves out’’ and then mating with the winner, to obtain the sperm for the best father for her offspring in a particular situation, this may not carry over into other situations. This will be particularly relevant if there is a time gap between mating and the use of the sperm. I will argue below that this is precisely the situation in which yellow dung fly females find themselves, although the time gap is often very small. An ability to store the sperm from different males separately, at least to some extent, would allow females to be able to choose the father of their offspring at the time of fertilization not at the time of mating, that is, to select for a ‘‘mate‐now‐choose‐later’’ mechanism for the evolution of cryptic female choice. This type of selection may have led to the evolution of the complexity of female reproductive tracts; they certainly do seem more complicated than is necessary for the straightforward receiving, maintenance, and use of sperm, for example, the multiple spermathecae described below, the many sperm storage tubules in birds (Briskie et al., 1997), or the internally structured spermathecae of some snails (Beese and Baur, 2006). The complexity may also be a response to adaptations in males which have the effect of preventing females from choosing against their sperm. However, it seems likely that since the final stages of fertilization in sperm‐storing species occur within a female’s body, she would have the final, possibly decisive influence. The insemination

346

PAUL I. WARD

milieu, including its effects on the mortality of sperm as they are transferred, and the activities of female muscular during transfer are also under female control. Hellriegel and Ward (1998) modeled how the physiology and morphological complexity of a female’s reproductive tract could in principle be used to influence the fitness of a female’s offspring. They modeled how a female could alter the transfer and storage of sperm from different males. The hostility of the female tract for sperm, a very common observation in animals, and a control over the storage rates of males’ sperm to the site or sites of storage were considered. Interestingly, an ability to vary the degree of sperm mortality in relation to different males could give females with even a single storage site a considerable degree of influence over fertilization successes under sperm competition. Even small degrees of differential storage of different male’s sperm to multiple storage sites were shown to give females considerable potential control over the later use of stored sperm, that is, in a complex storage system a large element of cryptic female choice could be given by small changes in physiological reactions to males assessed by females to be better fathers for their offspring, for whatever reason.

III. WHAT IS IN A NAME? The current name for the yellow dung fly, S. stercoraria (L.) has a somewhat complicated history (Sabrosky, 1999). The (L.) indicates the species was known to Linnaeus, but under another name, Musca stercoraria Linnaeus 1758. This was in the 10th edition of Linnaeus’s great rearrangement of systematics, Systema Naturae, which is considered to be the basis of animal systematics. In 1803 Meigen made the systematic revision which recognized Scathophaga as sufficiently different to Musca to warrant a separate genus and gave the species the name stercoraria. In 1805 Fabricius incorrectly amended the genus name to Scatophaga, which was used, along with a large number of other junior and invalid synonyms, well into the 1960s. At some point it was noticed that there was a family in Pisces (fish) that also had the name Scatophagidae, based on the genus Scatophagus. These are coral reef fish related to the butterfly fish. Since two groups with the same name are unacceptable in systematics, the reversion to the priority spelling in the Diptera, that is, Scathophaga, would avoid ‘‘a disturbing change in Pisces’’ (Sabrosky, 1999). Systematists therefore now use the Scathophaga spelling for the fly species.

POSTCOPULATORY SELECTION

347

IV. PARKER’S PIONEERING WORK An excellent summary of Parker’s pioneering work on the reproductive biology, and in particular of sperm competition, of the yellow dung fly can be found in Parker (1978). When a fresh dung pat is produced, it and the surrounding grass are very quickly populated by male dung flies searching for arriving females, which come over about the next hour and a half. When a female arrives, she is very quickly grabbed by a male and copula ensues, either on the pat or in the surrounding grass. The female exhibits no apparent choice at this point but the arrival timing itself may be an element of choice (see Section V, below) and the male also appears to determine copula duration. Males distribute themselves around a pat so that all have an approximately equal chance of obtaining a female. After copula, the male takes the female back to the pat, where she lays her eggs. During this phase, the male guards the female from other males by riding on her back. Fights between males are common and in some cases an attacker can take the female from the male defending her. Successful attackers are usually larger than the defending males. The new male will then copulate with the female, again with no obvious resistance from her, and then take her back to the pat to continue oviposition. In a series of laboratory experiments, Parker was able to show that about 80% of the female’s eggs were fertilized by the sperm of the last male to copulate with her. This gave a clear justification of the male’s guarding behavior: he is securing his paternity of a majority of the female’s eggs. Parker was also able to produce an elegant model of the behavior which gave predictions for the optimal copula time very close to that he had observed. This fit of model and observed behavior was a breakthrough piece of work in behavioral ecology.

V. FEMALE ARRIVAL AT THE DUNG PAT AND THE EVOLUTION OF TESTES SIZE Later work on the arrival of the females showed the situation was even more complicated than Parker had found. Borgia (1980, 1982) found that male size and density affected male success rate in capturing females, with larger males being in general more successful and smaller males being relatively more successful at lower densities. Borgia (1981) speculated that females do not arrive completely randomly, and this could be an element of female choice. Reuter et al. (1998) followed up this idea theoretically and strongly suggested that the female arrival pattern does indeed have an important consequence for the pairing pattern, with females being paired with larger males than they would be if they arrived at the pat in a

348

PAUL I. WARD

completely random pattern. Females therefore do seem to show a subtle precopulatory choice for larger males. Jann et al. (2000) also demonstrated that the selection pattern on male size showed a strongly seasonal pattern, consistent with Borgia’s ideas. This variation also makes it unlikely that selection on other aspects of the reproductive biology of the flies will be similar throughout the seasons. The above work lead to an interest in the behavior of males of different sizes and the sperm reserves maintained by males. Testes size is often considered as a good measure of the importance of sperm competition, both within and between species. Across species, males with larger testes are assumed to need to use more sperm, either because they will mate more often or because the risk or intensity of sperm competition is higher. Within species, males more successful at obtaining matings may have larger testes but less successful males may also have the larger testes, as they may attempt to pass more sperm per mating to outweigh the disadvantages of fewer matings. Clearly, the balance of these two effects, mating frequency and the number of sperm transferred and stored per mating, should always be considered in within‐species studies. In yellow dung flies, Ward and Simmons (1991) found smaller males copulate for longer than do larger ones. They also showed that diet, through sperm reserve production, affects copula duration. Parker (1992) modeled these results and produced a qualitative fit. He predicted that males should show cycles of using and replenishing their sperm reserves by mating and foraging away from the dung. There is little or even no food which can be obtained at the mating sites. Parker et al. (1990) modeled postcopulatory sexual selection by looking for overall patterns of sperm storage within females, treating each female as having a single storage site, that is, by simplifying the storage system of multiple spermathecae (see below). Their goal was to provide suggestions for examining the mechanisms of sperm transfer, storage, and use in a variety of species. Parker and Simmons (1991) improved the models of Parker et al. (1990). Direct displacement of sperm from a female’s storage system with no consideration of female activity was also an important feature of all the models. Simmons and Parker (1992) followed up these modeling papers and examined intraspecific variation in P2, the proportion of the female’s clutch fertilized by the second male, in the yellow dung fly. It should be noted here that although P2 has in the past been used as a measure of ‘‘sperm competition success,’’ it would be better described as the outcome of all the processes, male and female, which could affect sperm use patterns. Simmons and Parker found body size to be important; their results were consistent with larger males having higher sperm displacement rates of

POSTCOPULATORY SELECTION

349

previously stored sperm than smaller ones. Copula duration also had a separate influence on P2. In these experiments about 18% of the variance in P2 could be accounted for. This strongly suggested to me that female influences may play a role in the outcome of sperm competition sensu stricto and I began to pursue this idea. VI. CRYPTIC FEMALE CHOICE, SPERM COMPETITION, AND MALE–FEMALE CONFLICT I have chosen to adopt an essentially historical presentation of the development of ideas and results on cryptic female choice in the yellow dung fly. I believe this gives a flavor of research in action as we have striven to accommodate alternative views and to show how the interests in behavior and ecology have been expanded into genetics, physiology, and morphology, in an attempt to gain as complete a picture as possible of the reproductive biology of a single species. Ward (1993) was the first attempt to examine the relationship between the complexity of the female reproductive tract and the storage and use of sperm. A female usually has three quite distinct spermathecae, each with its own duct to the bursa copulatrix (Fig. 1). These are typically arranged as two on one side of the body, referred to below as the doublet spermathecae, and one on the other side, referred to as the singlet. About 10% of females at the field site near Zurich, Switzerland have four spermathecae, two

Fig. 1. A schematic diagram of the female reproductive tract. The three spermathecae and two accessory glands are at the upper left, attached to the bursa copulatrix. The last segments of the abdomen are also shown.

350

PAUL I. WARD

on each side of the body. The general idea was to test if this morphology, and its variability, could be an adaptation to influence how the sperm from males of different genotypes could be stored and used to a female’s advantage. The initial male character of interest was male size, as this was already known to influence the outcome of pairing patterns and sperm use. The pattern of sperm storage over a female’s individual spermathecae was examined for the first time. The main findings were, first, that sperm were not evenly distributed over the spermathecae. Sperm from larger males tended to be stored more together than were those of smaller males, and this in a female’s doublet spermathecae. Second, P2 was not constant over three clutches of a female not subsequently allowed to copulate. In females which mated last to a large male, P2 remained very high over the three clutches whereas if a female had mated last to a small male then P2 was low at first but increased over the three clutches. Females release sperm from the spermathecae just before each egg is individually fertilized in the bursa copulatrix. The interpretation of the P2 variation then was that all females first used sperm from a spermatheca containing the mixture of sperm with the highest representation from the larger of her two mates and only later used sperm from a spermatheca where the smaller male was, at least relatively, better represented. Later work has confirmed in other species that sperm are not evenly distributed and used among a female’s sperm storage sites. For example, Fritz (2004) found in the Caribbean fruit fly Anastrepha suspensa that after a single copulation a female would always have sperm in her ventral receptacle but not always in her spermathecae. Twig and Yuval (2005) found in the Mediterranean fruit fly Ceratitis capitata that both a ventral fertilization chamber and the spermathecae were used to store sperm. They found that sperm are used from the fertilization chamber at fertilization and the chamber is periodically refilled from the spermathecae. This would give the female the opportunity to choose the order from which spermatheca sperm are moved to the chamber. Around the early 1990s attention started to be paid to sperm length as a possible character of males which could influence the outcome of postcopulatory sexual selection. Ward and Hauschteck‐Jungen (1993) examined the relationship between sperm length characters and male body size in the yellow dung fly. Sperm are exceptionally slender cells in flies and so attention focused on the lengths of the head, which contains the condensed DNA of the cell, and of the tail, which provides the propulsive force of the cell. Total length was also examined. There were no significant correlations between male body size and any of the sperm length characters. Strikingly however, and for the first time in any species, it was found that individual

POSTCOPULATORY SELECTION

351

males differed significantly in the sperm lengths they produced. There was thus clearly repeatability, suggestive of genetic variation, in sperm length characters in a randomly chosen animal species. This variation in sperm characters is not related to body size. Ward (2000b) examined the genetic determination of sperm length in yellow dung flies. There was indeed considerable genetic variation and significant heritability (h2 ¼ 0.673). There was also evidence for sex‐linkage and epistatic interactions between determinants on the X chromosome and determinants on the autosomes. This suggests that the outcome of selection on sperm length is far from straightforward and I suggested that there may even be selection on females which favors genes producing shorter sperm in their offspring. This could be due to shorter sperm being cheaper to maintain in storage or, more intriguingly, easier to otherwise manipulate, that is, there may be selection against longer sperm within females. The repeatability and evidence for genetic variation have been found in a number of other species [e.g., Ward (1998a); Morrow and Gage (2001) in the cricket Gryllus bimaculatus; Malo et al. (2006) in red deer Cervus elaphus]. Nonetheless, there is some environmental effect on sperm size. Hellriegel and Blanckenhorn (2002) found that temperature variation affected testes and sperm development. Sperm length variation within males increased significantly with increasing developmental temperature and decreasing adult food, showing that stress can affect the size of sperm produced by a particular male. This effect was, however, relatively small compared to the genetic differences. Parker and Simmons (1994) extended Simmons and Parker (1992) to examine in detail the relationship between copula duration and male body size, yielding a detailed explanation for the increased copula duration of smaller males, because of the predicted lower sperm displacement rates and the lower chances of obtaining a female from another male in a takeover. Again, the experiment was conducted under one set of laboratory conditions and the females treated as having a sperm storage system with one storage site. The percentage of the variance in P2 was higher than before but not above 20%. Simmons et al. (1996) conducted a new set of experiments, the results of which they interpreted as a lack of evidence for female influence over paternity. They measured P2 in the first clutches of females mated to both large and small males and found that, when differences in the males’ behaviors were accounted for, large males did not gain a larger share of paternity than did small ones. However, because they only examined first clutches, they could not have detected the main effect found by Ward (1993). In that study, three clutches were examined and smaller males which mated second were relatively more successful in later clutches than

352

PAUL I. WARD

in earlier ones. No change in P2 was found for larger males mating second. This difference suggests that female do indeed exert an influence over paternity, and they do so by somehow separating the sperm from different males within their reproductive systems [shown to some extent in Ward (1993)] and using those from large males first. Otronen et al. (1997) went on to use the differences in sperm length between individual males described above to follow the fate of sperm across females’ individual spermathecae. [Bernasconi et al. (2007) have shown that a male’s mean sperm length has the necessary consistency for these experiments to be reliable.] Females mated with a large and a small male, which differed in sperm length, with either male producing the longer sperm. Again, very significant differences in the sperm distributions of the males over the spermathecae were found. Sperm length was found to play an important role in the pattern of storage but only in complex interaction with other male and female characters. When the larger male was the female’s first mate and his sperm were the longer ones, then his sperm tended to be predominantly in her singlet spermathecae. However, they were equally likely to be in the doublet or singlet when he was her second mate. When the larger male produced the shorter sperm, the pattern was even more complicated. This was because fewer of his sperm were found in the doublet spermathecae when the larger male mated second. More sperm would have been expected in the doublets from the initial results above (Ward, 1993) but sperm length was not examined there. Nevertheless, the Otronen et al. (1997) results clearly show the situation is extremely complex and that sperm length is related to the process of sperm storage.

VII. VARIATION AT THE PHOSPHOGLUCOMUTASE LOCUS Ward (1998b) examined cryptic female choice for single locus variation at the phosphoglucomutase (PGM) locus and quantitative variation in body size and development time in males. PGM is an important enzyme in the transfer of energy stored or obtained as glycogen into the glycolytic cycle. As glycogen is an important component of bacteria in cow dung, that is, the larval food, and is the adult flight fuel, PGM genotype may be crucial in a dung fly’s life history. Flies with different PGM genotype certainly show variation in their incidence in the field in relation to temperature over days, consistent with behavioral differences, and over seasons, consistent with changing selection patterns (Ward et al., 2004). In addition, fine‐scaled population structuring was only found for PGM in a study of a number of variable proteins, again suggesting variable selection at this locus (Kraushaar et al., 2002).

POSTCOPULATORY SELECTION

353

In a first experiment, females were mated to two males. Females were divided into three treatments after the matings: (1) they were dissected immediately; (2) they were dissected 30 min later, having been awake; or (3) they were dissected 30 min later, having been under CO2 anesthetic. The distributions of sperm in their spermathecae and the total numbers of sperm stored were analyzed. More sperm were stored from larger males, possibly a male effect, based on Parker and Simmons’s results described above, and the sperm were more likely to be in the singlet spermatheca the larger the male. It is hard to understand why the second effect could be favored by selection on males. This is because although his sperm could be better favored if they are stored together, the effect could also be reversed and they could be more disfavored. The treatment of the females also showed an interaction between their body size and the treatment. Immediately after copula, larger females had more sperm in the spermathecae. If the female had been anesthetized, then this relationship was even stronger. This shows that sperm are able to actively move on their own within the female tract. Interestingly however, if the female had been awake for 30 min, then there was actually a negative relationship between the total number of sperm stored and female body size. This was the first evidence that perhaps larger females are better able to resist manipulation by males or the movements of their sperm and that, although they need more sperm to fertilize their larger numbers of eggs than do smaller females, they are prepared to be selective about the particular sperm they store [see Fedina and Lewis (2004); Hellriegel and Bernasconi (2000) for later successful uses of CO2 anesthetic in yellow dung flies and in red flour beetles Tribolium castaneum respectively]. It is also possible that smaller females may be less selective because the males with which they mate vary less, for example in body size, than do the males with which larger females mate. The next experiment in Ward (1998b) examined whether females laid offspring of different genotype under different environmental conditions. Field‐collected females were allowed to oviposit in the laboratory either in bright or darkened conditions. This was done to simulate the differences in insolation in pats in different places within a meadow. Pats in the center of the meadow are continuously insolated during the day, whereas pats near or under trees, buildings, or hedges are at least partially shaded during the day. It was expected that these different habitats could provide different conditions for the growth of a female’s offspring and that this could lead to variation in a female’s assessment of a male’s quality, if there were genotype by environment interactions for offspring growth (see below).

354

PAUL I. WARD

Five PGM alleles were found in the population and eggs were not laid randomly with respect to PGM genotype in the ‘‘sunshine’’ and ‘‘shade’’ conditions. However, this effect was only found in one of four experimental blocks and more work was clearly needed (see below). A much stronger effect was found for the quantitative traits. In the three blocks in the same year there was a significant difference in offspring developmental times, with offspring laid in the sunshine taking longer to develop than those laid in the shade, especially toward the end of the season. Offspring laid in the sunshine were also larger than those laid in the shade, though significantly so only at the end of the season. These effects are most likely to be maternal effects rather than genetic ones, as females have been shown to be able to influence development times in their offspring (Blanckenhorn, 1998). An alternative, more speculative, explanation could be a complex interaction between genotype and different foraging conditions in sunshine or shaded conditions. This would clearly repay further examination. A possible genotype by environment interaction for larval growth at different temperature regimes was examined. Sibling offspring of different PGM genotypes were raised under either constant or variable temperatures. The variable temperatures were set so that the overall mean was the same as that of the constant temperature. This was done to simulate minimally the variation in insolation in the field described above. The idea was that if females responded to this small variation in temperature conditions, they would respond even more strongly to the much larger variation in the field. There were many higher order interactions for growth success in the different environments. Sex and family differences were expected, based on previous results showing similar effects in other experiments (Blanckenhorn, 1998). Most importantly, there was a significant PGM genotype by environment interaction. This showed that different PGM genotypes grow differently in different environments; homozygotes grew better in constant conditions and heterozygotes in variable ones. Thus, if a female can estimate the likely growing conditions for her offspring, perhaps by estimating the likely insolation and its variation at a particular site, she could choose the best father for that offspring and favor his sperm at fertilization as the egg is laid. This was tested in Ward (2000a), as described below.

VIII. EJACULATE LABELING AND DETAILED MORPHOLOGY OF THE FEMALE REPRODUCTIVE TRACT Simmons et al. (1999) used radiolabeling to follow the fates of ejaculates of different males within females. As radioactive amino acids were injected only a few days prior to copulation, it is possible that the accessory fluids

POSTCOPULATORY SELECTION

355

rather than sperm themselves were labeled. The arguments based on sperm movements may therefore need to be treated with some caution. All three spermathecae of a female were treated together in the data gathering. An indirect model of sperm displacement was constructed, that is, males were no longer considered to have direct access to the spermathecae as ‘‘the data, and morphology of the female, clearly preclude such a mechanism.’’ They found that there was evidence of an active female role but not for cryptic female choice since males of all sizes transferred equivalent numbers of sperm. Again the argument is relevant that the method of putting all spermathecae together and only examining one clutch under one environmental situation may remove any opportunity to detect variation in female mate choice. Birkhead (1998) supported Parker and Simmons’s position that there was no evidence for female choice. This difference in the interpretation of findings contributed to some later useful discussion on the precise conditions necessary to demonstrate female choice (Birkhead, 2000; Eberhard, 2000; Pitnick and Brown, 2000). Parker et al. (1999) extended the effects found above and showed that there was a relationship between copula duration and P2 and female size only in virgin females, that is, when an experiment simulated copulas with incoming virgin females. [A note of caution on the general importance of this observation is that this is an uncommon natural situation. Most females, around 98%, arriving at the dung to lay eggs already have sperm stored (Parker et al., 1993).] The results were interpreted as sperm displacement decreasing with increasing female size, another indication that female morphology influences the outcome of postcopulatory sexual selection. They also found that the new indirect displacement model was a very good fit to the data. Parker and Simmons (2000) improved the fit of the indirect displacement model to their data, concentrating on the effect of male size. More precise data on exactly how sperm are transferred, ideally in a variety of environmental situations, were clearly necessary. Hosken et al. (1999) provided a detailed histological examination of male and female anatomy during copula. Males were found to place the gonopore close to the exits of the spermathecal ducts in the bursa copulatrix and to transfer sperm there. Sperm in the spermathecae were extremely well ordered and appeared to be regularly packed. The male squeezes the female during copula, with the maximal squeezing very close to, or even at, the point where the gonopore is placed internally. The tip of the aedeagus (penis) has a stout chitinous hook which is pressed into the female epithelium. This looks like an anchor, which may also function to dissuade

356

PAUL I. WARD

the female from pushing against the male as the hook appears sharp enough to cause injury if wrongly moved. Males were also found to remove some sperm from the bursa at the end of copula. Hosken and Ward (2000) used the same sections to show that each spermatheca has a chitinous invagination projecting into the lumen opposite the entrance of the spermatheca, with a muscle attached to the end of the invagination. This muscle and invagination appear to move during copula, potentially moving sperm into or out of the spermatheca. This is another clear indication of a possible active female role in the process of sperm movements since the muscles of each spermatheca will be controlled, and thus be able to move, independently, potentially influencing the distribution of sperm stored across the spermathecae. Hosken and Ward (1999) examined the role of the large female accessory glands during copula and oviposition. The glands were clearly active during both behaviors, and apparently more so when a female was copulating with a larger male than with a smaller male. The function of the secretion could be lubrication during copula, antibacterial defense by the female or even acting directly to make the insemination site hostile to sperm. These possibilities could not be discriminated. One rather odd finding was of sperm in the accessory glands of some females. We have no clear explanation for this finding. The possibilities include: the male attempts to force fluid into all parts of the female tract, perhaps even to prevent the female fluid from the glands entering the bursa, sperm actively swimming but being disoriented and ending in the wrong place, or female muscular activity resulting in their accidental misdirection. There were now several indications of a female influence on sperm competition. It was time to directly examine cryptic female choice.

IX. CRYPTIC FEMALE CHOICE AT THE PGM LOCUS Ward (2000a) bred flies to be homozygotes of one of the two common PGM variants at a field site near Zurich, Switzerland. Female homozygotes of the most common field allele were maintained from eclosion either at a constant temperature or a variable one with a similar mean. This was to relate the results to the larval growth experiment described above (Ward, 1998b). When mature, they were mated in random order to both a similar male homozygote and a male homozygotic for the other allele in the experiment. The prediction, based on the growth results above, was that females from the constant environment would favor the males similar to themselves, since in a constant environment the offspring of these homozygotes would grow better than the heterozygotes, whereas the females

POSTCOPULATORY SELECTION

357

from the variable environment would favor the dissimilar males, producing more heterozygote offspring, which would grow best in the variable conditions. The predictions were partially confirmed. The females from the constant environment did indeed strongly favor the similar males (Fig. 2). Interestingly, they favored the second male even more in the second than in the first clutch, suggesting the female had been able to manipulate the sperm mixtures while maturing the second clutch of eggs (no new

A 0.6

P2 Residuals

0.4

First clutch Last male same genotype

0.2 −0.0 −0.2 −0.4

Last male different genotype

−0.6 Constant

Variable

Environment B 0.6

P2 Residuals

0.4

Second clutch Last male same genotype

0.2 −0.0 −0.2 −0.4

Last male different genotype

−0.6 Constant Variable Environment Fig. 2. The sperm use patterns of females raised under different environmental conditions. The successes of the males, P2 Residuals, are measured after the males’ influences have been accounted for. Reprinted from Ward (2000a) with permission from the Society for the Study of Evolution.

358

PAUL I. WARD

copulations took place between the clutches). However, the females from the variable environment showed no consistent preference, suggesting a cue other than PGM variation could be more important in these conditions. This will be the focus of future work. Nevertheless, this was clear evidence that cryptic female choice occurs at least sometimes in the yellow dung fly. It also shows that it is crucial to vary the conditions experienced by females. It does seem quite likely that conducting experiments under a single set of conditions could lead to all females favoring the same type of male and thus leading to the erroneous conclusion that cryptic female choice does not take place in the species. Ward (2000a) also successfully bred flies to have either three, the usual number, or four spermathecae, demonstrating that there is genetic variation in the study population for female internal morphology. In a subsequent sperm use experiment using females with either three or four spermathecae, high‐quality (large) females, that is, those laying an above‐ average clutch size, could better influence P2 in their preferred direction than could low‐quality (small) females when they had four spermathecae (Fig. 3). P2 was similar in all females with three spermathecae (Ward, 2000a). Again this shows that some aspect of the female has to be experimentally manipulated, here female internal morphology, if cryptic female choice is to be demonstrated.

0.8

P2 Residuals

0.4

Low-quality females

0.0

−0.4

−0.8

High-quality females

Three Four Number of spermatheca

Fig. 3. The sperm use patterns of females of different qualities, that is, different fecundities, with different numbers of spermathecae. The successes of the males, P2 Residuals, are measured after the males’ influences have been accounted for. Reprinted from Ward (2000a) with permission from the Society for the Study of Evolution.

POSTCOPULATORY SELECTION

359

X. A SELECTION EXPERIMENT ON SEXUAL CONFLICT Hosken and Ward (2001) bred replicate lines of flies under either monogamous or polyandrous regimes, in which each female mated to three males. A significant evolutionary response in testes size was detected, with males from the polyandrous lines investing more in testes size than males from the monogamous lines. This was the result predicted and confirms thinking on the evolution of testes size under differing risks of sperm competition. Hosken et al. (2001) further showed that females in the polyandrous treatment evolved larger accessory glands than those in the monogamous ones. We thought this may have been due to the glands producing a spermicidal substance, but this conclusion was premature (see Bernasconi et al., 2002 below). Hosken et al. (2001) conducted a sperm use experiment with the flies from the different selection treatments, controlling for inbreeding effects in the production of the experimental flies. Females from both selection regimes were mated in random order to a male from each selection regime. Both male and female effects on the fertilization successes of males from the different treatments were found. Mating with females from both treatments, males from polyandrous lines had higher fertilization successes than did those from monogamous lines, showing that breeding under sperm competition was directly related to success in experimental matings. Irrespective of the male types, females from the monogamous lines showed higher P2 values than those from the polyandrous ones. This was interpreted as monogamous breeding females having reduced effectiveness in some adaptation which allows females to influence the outcome of postcopulatory sexual selection. These matings were all conducted using the standard experimental protocol of all matings of a female being on the same day. It is therefore interesting that only the group of polyandrous females mated with males from the monogamous lines had a P2 much less than the normal laboratory value of around 80%. This is another indication that it is an interaction of male and female effects which most strongly affects the outcome of sperm competition. Hosken et al. (2002a) extended the variation among competing males, using males from distinct populations. They found that heteropopulation males were more successful in sperm competition, unrelated to copula duration, than males from the same population. They argued that this is best explained by females being less able to resist manipulation by males from populations to which they are not adapted.

360

PAUL I. WARD

XI. FIELD EXPERIMENTS ON CRYPTIC FEMALE CHOICE Ward et al. (2002a) conducted two field experiments to examine cryptic female choice in nature. These were done to test the expectations from the laboratory experiments (Ward, 2000a). Dung pats were experimentally manipulated either to have a conspicuous difference in insolation on their north and south aspects or were experimentally shaded over half their area to have a strongly insolated half of the pat and a strongly shaded half. The pats were then left undisturbed for at least an hour and a half for flies to oviposit on them. At the end of this period, all the dung fly eggs were collected, raised in the laboratory under standard conditions, and genotyped at the PGM locus. Both experimental setups yielded similar results: the offspring with more mobile PGM variants were significantly more likely to be found in the northern or shaded parts of the experimental pats. This could have been caused by females of different genotypes laying their eggs in different places. There were three lines of argument against this possibility. First, females almost always distribute their eggs over different parts of the habitat in the laboratory (Ward et al., 1999). Second, eggs are not laid in single clumps in the field, that is, each female appears to distribute her eggs over a relatively large area. Third, females spend most of the time when they are laying eggs walking around looking for oviposition sites, rather than actually laying the eggs, which is a very quick process. Again, this strongly indicates females distribute their eggs over the habitat. The most likely explanation for the differences in distribution of eggs of different PGM genotype is therefore that a female determines the paternity of each egg, to the extent possible given that each spermatheca contains a mixture of sperm from different males, by releasing sperm from the spermatheca most likely to result in the best PGM genotype for that particular offspring in that particular place, that is, depending on the genotype by environment interaction for the expected growth of that larva in that part of the dung pat.

XII. FEMALE ACCESSORY GLANDS AND MATERNAL EFFECTS Hosken et al. (2002b) followed up the function of the accessory glands. No antibacterial activity could be detected. Furthermore, there was no relationship with egg hatchability. These effects indicate that the secretion does not protect the female during copula and is also not important in producing substances which protect the eggs in the bacteria‐rich dung. Lubrication and altering the insemination milieu therefore remain the most likely function of the gland secretion. Bernasconi et al. (2002) did

POSTCOPULATORY SELECTION

361

find that sperm suffer serious mortality on transfer to females, up to 30%. However, they were unable to identify a part of the female tract which experimentally affected sperm survival in vitro. This left lubrication as the most likely gland secretion function. Nevertheless, more work on this area could be productive. Tregenza et al. (2003) attempted to estimate the interactions between maternal and genetic effects on offspring performance in a number of life‐ history traits in different thermal habitats. This was an initial trial at clarifying the relative importance of a female’s genotype and of her phenotypic experience for her offspring. Unfortunately, what we found was that such interactions are extremely complex even in a well‐controlled laboratory setting with a well‐understood species. More work is clearly needed in this area, in this and other species. It is very difficult to identify the characters a female might use when choosing a father for her offspring after mating with more than one male. In yellow dung flies, male body size, which is only loosely related to a male’s development time as a larva, has been the focus of attention on quantitative characters (Blanckenhorn, 1998; Mu¨hlha¨user et al., 1996; Simmons and Ward, 1991). However, it seems that male development time may well be important for the outcome of postcopulatory sexual selection. Hosken et al. (2003) found that males with shorter development times also had offspring which developed faster, as expected. They were also more successful in a sperm use experiment. Martin et al. (2004) also found that female adaptations were important in a similar way. It would thus be valuable to know in other species how many of a male’s or a female’s characters have a measurable influence of the outcome of competitive trials in postcopulatory sexual selection, that is, how complex is the pattern of selection. The heritabilities of the traits, and in particular their genetic correlations, would also be of interest. The evolution of the genetic architecture of male traits could be strongly influenced by the strength of female choice, for example balancing selection on two male traits caused by female choice could constrain the independent evolution of those traits.

XIII. CHECKING LABORATORY RESULTS WITH FIELD FLIES It is important to ensure that laboratory results are checked against field results so that we can be sure we are examining natural behavior and not laboratory artifacts. Ward et al. (2002b) measured the fertilization successes of field‐collected males in uninterrupted mating trials, using PGM variation to score the successes, in virgin laboratory‐reared females. Essentially, we

362

PAUL I. WARD

could confirm that the effects described in the earlier work on second male and female size effects on P2 variation were valid effects of natural male behavior. Blanckenhorn et al. (2004) also checked for relationships in the field. They found that the shapes of the relationships between male body size and (1) mating success in the field and (2) testes length were both quadratic increasing functions and were statistically similar. This suggests that male testes size is selected to be most closely related to the expected mean number of copulations a male of a particular size can expect to achieve. There was more variation in the testes length data than in the mating success data. I speculate that this is related to variation in female choice, either by different females, perhaps those of different sizes, or by choice variation at different parts of the season.

XIV. COMPARATIVE ANALYSES The evolutionary relevance of the experimental results has also been examined by comparing male and female reproductive characters across the family Scathophagidae. The importance of this kind of comparative analysis lies mainly in investigating if selection processes observable at the level of a single species could also be responsible for the evolution of the diversity of the same characters at a higher taxonomic level. Minder et al. (2005) used the independently assessed molecular phylogeny of Bernasconi et al. (2000) in their analysis. This corrects for any relationship between characters actually being due to a character or suite of characters not related to the characters of interests, because of a close genetic relationship between species. There were three important relationships found. First, sperm lengths were longer where female spermathecal duct lengths were longer. Second, testes sizes were larger where the sizes of the spermathecae themselves were larger. Third, testes sizes were also larger where spermathecal duct lengths were longer. All these associations are strong evidence that the experimental results described in detail above are also involved in longer‐term selection processes leading to overall coevolution of male and female reproductive characters. Comparative work has also been conducted on other species. Pitnick et al. (1999) had previously examined evolution of multiple sperm storage organs in Drosophila species. They found that loss of spermathecal function had occurred repeatedly, with only the ventral receptacle being used. The ventral receptacle appears to be the primary storage site in these species. Miller and Pitnick (2003) followed up this research; their main finding was that a male’s fertilization success depended on an interaction between his

POSTCOPULATORY SELECTION

363

own sperm and ejaculate traits and the female’s morphological reproductive tract traits. Furthermore, Bangham et al. (2003) used female Drosophila with two or three spermathecae to show that the temporal pattern of sperm use differed between the female morphologies. In a different kind of comparative study, Joly et al. (2004) examined clinal patterns in reproductive patterns in D. melanogaster. They found a clinal pattern in sperm length associated with drought in D. melanogaster but not in D. simulans. This set of papers shows again that investigation of related species and checking for results in the field are important sources of stimulation for research questions, and even for opening up new areas.

XV. CONCLUDING REMARKS The reproductive systems of males and females have clearly not evolved in complete independence of other aspects of their biology and such relationships will certainly be of interest in understanding the details of the coevolution of reproductive traits. For example, a topic of current interest in sexual selection in general is the relationship between reproduction and the immune system. We have found variation in the immune system of yellow dung flies (Schwarzenbach et al., 2005) and there is indeed a relationship between the immune system and reproductive biology. Flies bred to have an increased value of an important immune parameter, the phenoloxidase level (PO), have reduced longevity under starvation in the laboratory compared to flies bred to have a low PO level (Schwarzenbach and Ward, 2006). In general, much attention has been paid to sperm size, as it is an easy‐to‐ measure character and varies greatly between species. For example, Radwan (1996) found that copula duration, body size, and sperm number were not related to sperm competition success in the bulb mite Rhizoglyphus robini, but males producing larger sperm were more successful. This is however, an exceptional result, possibly because bulb mites have amoeboid sperm. In species with flagella sperm, the importance of sperm length is much less clear. For example, Baer et al. (2003) studied three species of Bombus bumble bees. They tenuously suggested that multiple mating may select for longer sperm, but acknowledged that the comparative work is based on too few species for the conclusion to be certain. They also found that longer or shorter sperm may be preferentially stored, that is, cryptic female choice may occur but not have a consistent direction. Simmons et al. (2003) found extensive intramale sperm length variability in the cricket Teleogryllus oceanicus. In a careful experiment they found no consistent relationship between sperm length and fertilization success in competitive

364

PAUL I. WARD

matings. Locatello et al. (2006) also found no relationship between male phenotype and sperm length in guppies Poecilia reticulata. It is very clear that much more work is necessary if we are to understand the evolution of one of the most obvious, and easily measured, sperm characters, simple sperm length. One possibility is that the interaction of sperm length and cryptic female choice generally leads to stabilizing rather than directional selection on the relevant characters. There seem to me to be two main reasons why cryptic female choice has been hard to determine and establish in the yellow dung fly and which have made its recognition in other species extremely difficult. Most investigations are conducted under a single set of laboratory conditions or under the natural conditions in the field. The laboratory situation produces almost complete lack of variance in the conditions under which the females are able or likely to choose. As under such a restricted set of conditions all females may indeed show the same choice, the conclusion may be reached that there is no choice in the field. However, this may well be a laboratory artifact and females would choose differently under different conditions. I therefore strongly recommend that experiments in cryptic female choice are conducted under a variety of conditions which mimic the expected natural variation females would encounter in the field. Second, studies in the field typically have lower sample sizes than those in the laboratory. Thus, even if the relevant environmental variation can be recognized (see the PGM results for yellow dung flies above), the number of females in a particular set of conditions is likely to be too small to establish a statistically significant difference between groups. This is perhaps the main reason behavioral ecologists should strive to establish model systems in which adequate laboratory and field work can be conducted to similarly high standards.

XVI. SUMMARY The yellow dung fly S. stercoraria (L.) has been a classic study organism for the study of sperm competition. Sperm competition occurs when the sperm from more than one male is present in a female’s reproductive tract during a fertilization bout. More recently, attention has been focused on female influences on the outcome of sperm competition, often called cryptic female choice. The interactions between male and female characters, and male–female conflict in particular, are currently a very active research area. A female yellow dung fly has three or four spermathecae in which she stores the sperm from a number of males. Cryptic female choice has been demonstrated in the yellow dung fly in the laboratory, and is partly related to her

POSTCOPULATORY SELECTION

365

number of spermathecae. Females also have accessory reproductive glands, which are active during copula. There is considerable sperm mortality during transfer. Variable sperm mortality at the site of transfer can be a powerful mechanism of cryptic female choice, even in the absence of multiple storage organs. Cryptic female choice in dung flies depends on a male’s genotype at the PGM locus. As there is a genotype by environment interaction for offspring growth between PGM genotype and growth temperature, the genetic quality of a male as a father depends on the environment in which his offspring must grow, which the female selects at oviposition, that is, the females have a mate‐now‐choose‐later choice mechanism. Field experiments have shown at the population level that the distribution of PGM genotypes is consistent with the laboratory choice results. Furthermore, artificial selection on mating system resulted in correlated changes in male testes size and in the size of female accessory reproductive glands, consistent with a sexual conflict over fertilization patterns in this species. Sperm length, which is variable among males but consistent within a male, may also be involved in how successfully a male can transfer his sperm. Postcopulatory selection in this species is clearly complex and not controlled completely by either sex. Rather, there is a complicated, many‐layered interaction between the sexes. This will also be true in most other animal species. I therefore strongly recommend that investigations of the details of postcopulatory sexual selection are conducted under a variety of environmental conditions, which should reflect the natural variation of the species under examination as closely as possible. This would enormously increase the chance of teasing apart male and female influences on the outcome of postcopulatory selection.

Acknowledgments Thanks to Gioia Schwarzenbach, Andrew Pemberton, Marc Naguib, Tracie Ivy, Luc Bussiere, and an anonymous reviewer for comments on the manuscript. Thanks also to Barbara Oberholzer and Rosemarie Keller for their help in its preparation and to Thomas Degen for Fig. 1.

References Andersson, M. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton. Andersson, M., and Simmons, L. W. (2006). Sexual selection and mate choice. TREE 21, 296–302. Arnqvist, G., and Nilsson, T. (2000). The evolution of polyandry: Multiple mating and female fitness in insects. Anim. Behav. 60, 145–164. Arnqvist, G., and Rowe, L. (2005). ‘‘Sexual Conflict.’’ Princeton University Press, Princeton.

366

PAUL I. WARD

Baer, B., Schmid‐Hempel, P., Hoeg, J. T., and Boomsma, J. J. (2003). Sperm length, sperm storage and mating system characteristics in bumblebees. Insect Soc. 50, 101–108. Bangham, J., Chapman, T., Smith, H. K., and Partridge, L. (2003). Influence of female reproductive anatomy on the outcome of sperm competition in Drosophila melanogaster. Proc. R. Soc. Lond. B. 270, 523–530. Beese, K., and Baur, B. (2006). Expandable spermatheca influences sperm storage in the simultaneously hermaphroditic snail Arianta arbustorum. Inver. Rep. Dev. 49, 93–101. Bernasconi, M. V., Pawlowski, J., Valsangiacomo, C., Piffaretti, J. C., and Ward, P. I. (2000). Phylogeny of the Scathophagidae (Diptera, Calyptratae) based on mitochondrial DNA sequences. Mol. Phylogenet. Evol. 16, 308–315. Bernasconi, G., Hellriegel, B., Heyland, A., and Ward, P. I. (2002). Sperm survival in the female reproductive tract in the fly Scathophaga stercoraria (L.). J. Insect Physiol. 48, 197–203. Bernasconi, G., Ashman, T. L., Birkhead, T. R., Bishop, J. D. D., Grossniklaus, U., Kubli, E., Marshall, D. L., Schmid, D. L., Skogsmyr, I., Snook, R. R., Taylor, D., Till‐Bottraud, I., et al. (2004). Evolutionary ecology of the prezygotic stage. Science 303, 971–975. Bernasconi, G., Ward, P. I., and Hellriegel, B. (2007). Repeatability of sperm size in outbred and inbred Scathophaga stercoraria (L.) flies. Can. Entomol. 139, 228–234. Birkhead, T. R. (1998). Cryptic female choice: Criteria for establishing female sperm choice. Evolution 52, 1212–1218. Birkhead, T. R. (2000). Defining and demonstrating postcopulatory female choice—Again. Evolution 54, 1057–1060. Blanckenhorn, W. U. (1998). Adaptive phenotypic plasticity in growth, development time, and body size in the yellow dung fly. Evolution 52, 1394–1407. Blanckenhorn, W. U., Hellriegel, B., Hosken, D. J., Jann, P., Altwegg, R., and Ward, P. I. (2004). Does testis size track expected mating success in yellow dung flies? Funct. Ecol. 18, 414–418. Borgia, G. (1980). Sexual competition in Scatophaga stercoraria: Size‐related and density‐ related changes in male ability to capture females. Behaviour 75, 185–206. Borgia, G. (1981). Mate selection in the fly Scatophaga stercoraria: Female choice in a male‐ controlled system. Anim. Behav. 29, 71–80. Borgia, G. (1982). Experimental changes in resource structure and male density: Size‐related differences in mating success among male Scatophaga stercoraria. Evolution 36, 307–315. Briskie, J. V., Montgomerie, R., and Birkhead, T. R. (1997). The evolution of sperm size in birds. Evolution 51, 937–945. Eberhard, W. G. (1996). ‘‘Female Control: Sexual Selection by Cryptic Female Choice.’’ Princeton University Press, Princeton. Eberhard, W. G. (2000). Criteria for demonstrating postcopulatory female choice. Evolution 54, 1047–1050. Fedina, T. Y., and Lewis, S. M. (2004). Female influence over offspring paternity in the red flour beetle Tribolium castaneum. Proc. R. Soc. B 271, 1393–1399. Fritz, F. H. (2004). Sperm storage patterns in singly mated females of the Caribbean fruit fly, Anastrepha suspensa (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 97, 1328–1335. Grafen, A. (1984). Natural selection, kin selection and group selection. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), 2nd ed., pp. 62–89. Blackwell, Oxford. Hellriegel, B., and Bernasconi, G. (2000). Female‐mediated differential sperm storage in a fly with complex spermathecae, Scatophaga stercoraria. Anim. Behav. 59, 311–317. Hellriegel, B., and Blanckenhorn, W. U. (2002). Environmental influences on the gametic investment of yellow dung fly males. Evol. Ecol. 16, 505–522.

POSTCOPULATORY SELECTION

367

Hellriegel, B., and Ward, P. I. (1998). Complex female reproductive tract morphology: Its possible use in postcopulatory female choice. J. Theor. Biol. 190, 179–186. Hosken, D., and Snook, R. (2005). How important is sexual conflict? Am. Nat. 165, S1–S4. Hosken, D. J., and Ward, P. I. (1999). Female accessory reproductive gland activity in the yellow dung fly Scathophaga stercoraria (L.). J. Insect Physiol. 45, 809–814. Hosken, D. J., and Ward, P. I. (2000). Copula in yellow dung flies (Scathophaga stercoraria): Investigating sperm competition models by histological observation. J. Insect Physiol. 46, 1355–1363. Hosken, D. J., and Ward, P. I. (2001). Experimental evidence for testis size evolution via sperm competition. Ecol. Lett. 4, 10–13. Hosken, D. J., Blanckenhorn, W. U., and Garner, T. W. J. (2002a). Heteropopulation males have a fertilization advantage during sperm competition in the yellow dung fly (Scathophaga stercoraria). Proc. R. Soc. Lond. B 269, 1701–1707. Hosken, D. J., Garner, T. W. J., and Ward, P. I. (2001). Sexual conflict selects for male and female reproductive characters. Curr. Biol 11, 489–493. Hosken, D. J., Garner, T. W. J., Tregenza, T., Wedell, N., and Ward, P. I. (2003). Superior sperm competitors sire higher‐quality young. Proc. R. Soc. Lond. B 270, 1933–1938. Hosken, D. J., Meyer, E. P., and Ward, P. I. (1999). Internal female reproductive anatomy and genital interactions during copula in the yellow dung fly, Scathophaga stercoraria (Diptera: Scathophagidae). Can. J. Zool. 77, 1975–1983. Hosken, D. J., Uhı´a, E., and Ward, P. I. (2002b). The function of female accessory reproductive gland secretion and a cost to polyandry in the yellow dung fly. Physiol. Entomol. 27, 87–91. Jann, P., Blanckenhorn, W. U., and Ward, P. I. (2000). Temporal and microspatial variation in the intensities of natural and sexual selection in the yellow dung fly Scathophaga stercoraria. J. Evol. Biol. 13, 927–938. Joly, D., Korol, A., and Nevo, E. (2004). Sperm size evolution in Drosophila: Inter‐ and intraspecific analysis. Genetica 120, 233–244. Kraushaar, U., Goudet, J., and Blanckenhorn, W. U. (2002). Geographical and altitudinal population genetic structure of two dung fly species with contrasting mobility and temperature preference. Heredity 89, 99–106. Locatello, L., Rasotto, M. B., Evans, J. P., and Pilastro, A. (2006). Colourful male guppies produce faster and more viable sperm. J. Evol. Biol. 19, 1595–1602. Malo, A. F., Gomendio, M., Garde, J., Lang‐Lenton, B., Soler, A. J., and Roldan, E. R. S. (2006). Sperm design and sperm function. Biol. Lett. 2, 246–249 Doi:10.1098/ rsbl.2006.0449. Martin, O. Y., Hosken, D. J., and Ward, P. I. (2004). Post‐copulatory sexual selection and female fitness in Scathophaga stercoraria. Proc. R. Soc. Lond. B 271, 353–359. Miller, G. T., and Pitnick, S. (2003). Functional significance of seminal receptacle length in Drosophila melanogaster. J. Evol. Biol. 16, 114–126. Minder, A. M., Hosken, D. J., and Ward, P. I. (2005). Co‐evolution of male and female reproductive characters across the Scathophagidae (Diptera). J. Evol. Biol. 18, 60–69. Morrow, E. H., and Gage, M. J. G. (2001). Artificial selection and heritability of sperm length in Gryllus bimaculatus. Heredity 87, 356–362. Mu¨hlha¨user, C., Blankenhorn, W. U., and Ward, P. I. (1996). The genetic component of copula duration in the yellow dung fly. Anim. Behav. 51, 1401–1407. Otronen, M., Reguera, P., and Ward, P. I. (1997). Sperm storage in the yellow dung fly Scathophaga stercoraria: Identifying the sperm of competing males in separate female spermathecae. Ethology 103, 844–854. Parker, G. A. (1970). Sperm competition and its evolutionary consequences in insects. Biol. Rev. 45, 525–567.

368

PAUL I. WARD

Parker, G. A. (1978). Searching for mates. In ‘‘Behavioural Ecology: An Evolutionary Approach’’ (J. R. Krebs and N. B. Davies, Eds.), pp. 214–244. Blackwell, Oxford. Parker, G. A. (1979). Sexual selection and sexual conflict. In ‘‘Sexual Selection and Reproductive Competition in Insects’’ (M. S. Blum and N. A. Blum, Eds.), 1st ed., pp. 123–166. Academic, New York. Parker, G. A. (1992). Marginal value theorem with exploitation time costs – diet, sperm reserves, and optimal copula duration in dung flies. Am. Nat. 139, 1237–1256. Parker, G. A., and Simmons, L. W. (1991). A model of constant random sperm displacement during mating: Evidence from Scatophaga. Proc. R. Soc. Lond. B 246, 107–115. Parker, G. A., and Simmons, L. W. (1994). Evolution of phenotypic optima and copula duration in dungflies. Nature 370, 53–56. Parker, G. A., and Simmons, L. W. (2000). Optimal copula duration in yellow dung flies: Ejaculatory duct dimensions and size‐dependent sperm displacement. Evolution 54, 924–935. Parker, G. A., Simmons, L. W., and Kirk, H. (1990). Analyzing sperm competition data: Simple models for predicting mechanisms. Behav. Ecol. Sociobiol. 27, 55–65. Parker, G. A., Simmons, L. W., and Ward, P. I. (1993). Optimal copula duration in dungflies: Effects of frequency dependence and female mating status. Behav. Ecol. Sociobiol. 32, 157–166. Parker, G. A., Simmons, L. W., Stockley, P., McChristie, D. M., and Charnov, E. L. (1999). Optimal copula duration in yellow dung flies: Effects of female size and egg content. Anim. Behav. 57, 795–805. Pitnick, S., and Brown, W. D. (2000). Criteria for demonstrating female sperm choice. Evolution 54, 1052–1056. Pitnick, S., Markow, T., and Spicer, G. S. (1999). Evolution of multiple kinds of female sperm‐ storage organs in Drosophila. Evolution 53, 1804–1822. Radwan, J. (1996). Intraspecific variation in sperm competition success in the bulb mite: A role for sperm size. Proc. R. Soc. Lond. B 263, 855–859. Reuter, M., Ward, P. I., and Blanckenhorn, W. U. (1998). An ESS treatment of the pattern of female arrival at the mating site in the yellow dung fly Scathophaga stercoraria (L.). J. Theor Biol. 195, 363–370. Rice, W. R. (1996). Sexually antagonistic male adaptation triggered by experimental arrest of female evolution. Nature 381, 232–234. Sabrosky, C. W. (1999). ‘‘Family‐Group Names in Diptera. North American Dipterist’s Society.’’ Backhuys Publishers, Leiden. Schwarzenbach, G. A., and Ward, P. I. (2006). Responses to selection on phenoloxidase activity in yellow dung flies. Evolution 60, 1612–1621. Schwarzenbach, G. A., Hosken, D. J., and Ward, P. I. (2005). Sex and immunity in the yellow dung fly Scathophaga stercoraria. J. Evol. Biol. 18, 455–463. Simmons, L. W., and Parker, G. A. (1992). Individual variation in sperm competition success of yellow dung flies, Scatophaga stercoraria. Evolution 46, 366–375. Simmons, L. W., and Ward, P. I. (1991). The heritability of sexually dimorphic traits in the yellow dung fly Scathophaga stercoraria (L.). J. Evol. Biol. 4, 593–601. Simmons, L. W., Parker, G. A., and Stockley, P. (1999). Sperm displacement in the yellow dung fly, Scatophaga stercoraria: An investigation of male and female processes. Am. Nat. 153, 302–314. Simmons, L. W., Stockley, P., Jackson, R. L., and Parker, G. A. (1996). Sperm competition or sperm selection: No evidence for female influence over paternity in yellow dung flies Scatophaga stercoraria. Behav. Ecol. Sociobiol. 38, 199–206.

POSTCOPULATORY SELECTION

369

Simmons, L. W., Wernham, J., Garcia‐Gonzalez, F., and Kamien, D. (2003). Variation in paternity in the field cricket Teleogryllus oceanicus: No detectable influence of sperm numbers or sperm length. Behav. Ecol. 14, 539–545. Tregenza, T., Wedell, N., Hosken, D. J., and Ward, P. I. (2003). Maternal effects on offspring depend on female mating pattern and offspring environment in yellow dung flies. Evolution 57, 297–304. Twig, E., and Yuval, B. (2005). Function of multiple sperm storage organs in female Mediterranean fruit flies (Ceratitis capitata, Diptera: Tephritidae). J. Insect Physiol. 51, 67–74. Ward, P. I. (1993). Females influence sperm storage and use in the yellow dung fly Scathophaga stercoraria (L.). Behav. Ecol. Sociobiol. 32, 313–319. Ward, P. I. (1998a). Intraspecific variation in sperm size characters. Heredity 80, 655–659. Ward, P. I. (1998b). A possible explanation for cryptic female choice in the yellow dung fly, Scathophaga stercoraria (L.). Ethology 104, 97–110. Ward, P. I. (2000a). Cryptic female choice in the yellow dung fly Scathophaga stercoraria (L.). Evolution 54, 1680–1686. Ward, P. I. (2000b). Sperm length is heritable and sex‐linked in the yellow dung fly (Scathophaga stercoraria). J. Zool. 251, 349–353. Ward, P. I., and Hauschteck‐Jungen, E. (1993). Variation in sperm length in the yellow dung fly Scathophaga stercoraria (L.). J. Insect Physiol. 39, 545–547. Ward, P. I., and Simmons, L. W. (1991). Copula duration and testes size in the yellow dung fly, Scathophaga stercoraria (L.) – the effects of diet, body size, and mating history. Behav. Ecol. Sociobiol. 29, 77–85. Ward, P. I., Foglia, M., and Blanckenhorn, W. U. (1999). Oviposition site choice in the yellow dung fly Scathophaga stercoraria. Ethology 105, 423–430. Ward, P. I., Vonwil, J., Scholte, E. J., and Knop, E. (2002a). Field experiments on the distributions of eggs of different phosphoglucomutase (PGM) genotypes in the yellow dung fly Scathophaga stercoraria (L.). Mol. Ecol. 11, 1781–1785. Ward, P. I., Wedell, N., Hosken, D. J., and Tregenza, T. (2002b). Measuring the sperm competition successes of field males of the yellow dung fly. Ecol. Entomol. 27, 763–765. Ward, P. I., Jann, P., and Blanckenhorn, W. U. (2004). Temperature‐mediated seasonal variation in phosphoglucomutase allozyme frequency in the yellow dung fly, Scathophaga stercoraria. Mol. Ecol. 13, 3213–3218.

This page intentionally left blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

The Evolution, Function, and Meaning of Marmot Alarm Communication Daniel T. Blumstein department of ecology and evolutionary biology, university of california, los angeles, california 90095‐1606, usa

I. INTRODUCTION When strolling through your favorite habitat, it is not uncommon to hear birds or mammals emit alarm calls (Klump and Shalter, 1984), particularly if you’re strolling with a domestic predator! These striking and often easily identified vocalizations are often loud and localizable (Hurd, 1996; Wood et al., 2000) and may be directed to both predators and conspecifics. Because predators may be better able to locate a caller, emitting these calls creates an evolutionary paradox: Why call if it increases the likelihood of a caller being detected by a predator and killed (Maynard Smith, 1965)? One solution to this paradox is that if, by calling, individuals save their relatives, kin selection can explain its adaptive utility (Keller and Reeve, 2002). Alarm calls are thus a system in which we can study the dynamics of altruism. Moreover, because alarm calls may be directed to conspecifics, we can study their meaning. By meaning, I refer here specifically to their information content (Halliday, 1983; Macedonia and Evans, 1993). Calls could contain potentially referential information about the specific type of predator, and/or calls could contain information about the degree of risk that the caller faces when calling (Evans, 1997; Macedonia and Evans, 1993). Alarm calling is thus a system in which we can study the evolution and adaptive utility of complex communication and referentiality—a necessary component of human language. Calls can also contain other information, such as the identity, sex, and age of the caller. Thus, alarm‐calling systems may offer us some unique insights into the adaptive significance of individuality. 0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37008-3

371

Copyright 2007, Elsevier Inc. All rights reserved.

372

DANIEL T. BLUMSTEIN

In this chapter, I summarize two decades of work studying marmot alarm communication. Marmots are large, ground‐dwelling sciurid rodents, and I have studied 8 of the 14 species. My work has been conducted in Canada (Vancouver Island marmots—Marmota vancouverensis), Berchtesgaden National Park, Germany (Alpine marmots––M. marmota); Khunjerab National Park, Pakistan (golden marmots––M. caudata aurea), The Chuvash Republic, Russia (steppe marmots—M. bobak), and the United States (Mt. Rainer National Park, hoary marmots—M. caligata; Olympic National Park, Olympic marmots—M. olympus; Kansas and Ohio, woodchucks—M. monax; Capital Reef National Park, Utah, The Rocky Mountain Biological Laboratory, Colorado, and around Boulder, Colorado, yellow‐bellied marmots—M. flaviventris). Marmots are an outstanding model system to study alarm call function and meaning because they are diurnal, live in discrete locations, alarm call when they encounter a variety of predators, and, unlike some of their more distant scuirid relatives (Leger et al., 1980), emitting alarm calls in nonpredator contexts is relatively rare. My work has focused on three of the four Tinbergian questions (Tinbergen, 1963): the evolution, adaptive utility, and meaning of alarm calls. I also have thought about the applied value of studying alarm communication; a type of question that I have suggested could be considered as a ‘‘fifth question’’ (Blumstein, in press‐b). II. EVOLUTION Marmots produce a variety of whistles, chirps, and chucks. Figure 1 illustrates spectrograms from all 14 species. It is immediately obvious that some species produce multiple alarm call types, while others produce only a single type of call. Interestingly, and unlike sexually selected vocalizations or those involved in species identification, these alarm calls are used in a single context: signaling alarm. What explains this variation in alarm call structure? Some variations in call structure might be explained by the intended recipient. Alarm calls can be directed both to predators and to conspecifics. Generally, calls may be directed to conspecifics to warn them about the presence of a predator (Blumstein and Armitage, 1997a; Sherman, 1977) or to create pandemonium (Neill and Cullen, 1974; Sherman, 1985) during which time the caller may escape. Calls that function in these contexts should occur in social species. Calls may also be directed to the predator and may function to discourage pursuit (Hasson, 1991) and may thus be a general case of detection signaling. They may also attract other predators––which would create competition or predation on one predator by another, thus allowing the prey to escape (Ho¨gstedt, 1983).

MARMOT ALARM COMMUNICATION

373

Himalayan himalayana Tarbagan sibirica Black-capped camtschatica Gray baibacina Steppe bobak Golden caudata Menzbier's menzbieri Brooks Range broweri Alpine marmota Woodchuck monax Hoary caligata Vancouver island vancouverensis Olympic olympus Yellow-bellied flaviventris 4 kHz 2 1 sec Number of alarm calls Equivocal

1

2

4

5

Fig. 1. Marmot phylogeny and spectrograms (256 point FFT, 50% overlap Canary spectrograms generating a frequency resolution of 5.75 ms
86.93 Hz) of the alarm call repertoire of all 14 species along with hypothesized reconstruction of alarm call repertoire size. Reproduced with permission from Blumstein (2003). The partially resolved phylogeny is based on Kruckenhauser et al. (1999) and Steppan et al. (2000). For species with multiple call types, boxes separate adjacent call types.

374

DANIEL T. BLUMSTEIN

A comparative study sheds light on the ancestral function of alarm calling in rodents (Shelley and Blumstein, 2005). Erin Shelley and I focused on 209 species of rodents and noted whether or not they produced alarm calls. Social species should benefit from producing alarm calls, either by nepotistic or potentially reciprocal benefits. Thus, if calling evolved to have a conspecific alarming function, then we would expect that the evolution of sociality would precede the evolution of calling. We scored species as social if they were likely to live near kin; either because they lived in family groups or because they lived in colonies. However, producing alarm calls is a potentially risky behavior that may attract the attention of predators. If calling evolved to be directed towards predators, we assumed that individuals would only do so when they were relatively safe. Thus, producing them in the dark, where it is difficult to assess and manage predation risk, may be particularly risky. If calling evolved to be directed toward predators, we might expect that the evolution of diurnality would precede the evolution of calling. We noted whether they were predominantly active during the day or night, and whether they were never active at night (this reduced our sample size to 156 species for which we had sufficient data). We used nonphylogenetic and phylogenetic techniques to study the evolution of calling in rodents using our 209 and 156 species data sets (Shelley and Blumstein, 2005). In nonphylogenetic logistic regressions, we found that more variation in the likelihood of calling was explained by diurnality (25–44%) than by sociality (7–8%). There was a weak relationship between sociality and diurnality (6–7% of the variation was explained). Phylogenetic analyses supported the hypothesis that calling was likely to evolve following the evolution of diurnality, but not following the evolution of sociality. These results are consistent with the hypothesis that the evolution of diurnality preceded the evolution of alarm calling. We inferred from this that calling may have initially evolved as a means to communicate with the predator, and we suggested that its initial function was detection signaling that was subsequently exapted (Gould and Vrba, 1982) to serve its conspecific warning function. Call structure may also shed some light on the target. For instance, we expect signals that are directed to a predator to be ‘‘obvious.’’ Marler (1955) suggested that songbird mobbing calls illustrate this in that they are broadband, rapidly repeated sounds that are easy to localize. In contrast, songbird alarm calls elicited by aerial predators are difficult to localize because they have a relatively narrow bandwidth and fade in and out (Marler, 1955). Unlike cockerels, which, when alarmed by terrestrial predators produce uniquely wide bandwidth, rapidly paced calls, and, when alarmed by aerial predators, produce high‐frequency hard to hear faint whistles (Evans, 1997; Evans et al., 1993); all marmot species I have studied do not have such

MARMOT ALARM COMMUNICATION

375

production specificity. Rather, calls seemingly communicate the degree of risk a caller faces when calling (see the description later). Some marmot species emit calls that are less obvious as risk increases. Using a human as a threatening stimulus, I found that golden marmots, which have multinote calls, emit calls with fewer notes as risk increases (i.e., as a person gets closer to the caller) (Fig. 2). Alpine marmots repeat a simple note different number of times and emit calls with fewer repetitions as risk increases (Fig. 2). By contrast, other species make themselves more obvious as risk increases. Yellow‐bellied marmots called more rapidly (Fig. 3) and emit more calls as risk increases, variables that influenced their responsiveness to playback (Fig. 3). Thus, even among congeners, mechanisms to potentially communicate risk vary, a theme that will be developed later.

Notes/Call

16

Golden marmot (M. caudata)

12 8 4 0 0

12

50

100 150 Distance (m)

200

250

Alpine marmot (M. marmota)

Notes/Call

9

6

3

0 0

50 100 Distance (m)

150

Fig. 2. Golden marmots produce first calls with fewer notes (adjusted R2 ¼ 0.254, p < 0.001), and alpine marmots produce first calls with fewer repeated notes (adjusted R2 ¼ 0.10, one‐ tailed p ¼ 0.031), as the distance to an approaching human decreases. [Redrawn from Blumstein (1995a) and Blumstein and Arnold (1995), and used with permission from Blackwell Publishing; inset: alarm call spectrograms.] A one‐tailed p‐value is reported for alpine marmots because of an a priori directional hypothesis.

376

DANIEL T. BLUMSTEIN

A Response to a human with a dog p < 0.001

Time (s)

6 4 2 0

Response score (±SE)

B

4

0

1

2

3 4 5 6 7 Inter-call interval

8

9

Response to playback p = 0.002

3

2

1

4 Slow 1 Call 4 Fast Playback stimulus

Fig. 3. (A) The response of yellow‐bellied marmots to a human approaching with a dog. The intercall interval 1 ¼ the time between the first and second call, intercall interval 2 ¼ the time between the second and third call, and so forth. Intercall interval decreased significantly (p < 0.01) demonstrating that marmots emitted calls at a faster pace as risk increased. (B) Yellow‐bellied marmots responded to both the number and rate of calls played back to them. Responses were scored: 0 ¼ no response; 1 ¼ stand and look, 2 ¼ rear and look, 3 ¼ rear‐ up on hind legs in upright position and look, and 4 ¼ retreat to burrow. All calls were played back at 93 dB SPL.

Another factor that could explain variation in call structure is the acoustic habitat. The environment modifies the structure of all signals, often in predictable ways (Bradbury and Vehrencamp, 1998; Slabbekoorn, 2004). Attenuation is inevitable and many environments create ‘‘excess attenuation’’ (attenuation greater than 6 dB per doubling distance––Bradbury and Vehrencamp, 1998). There is also degradation––the inevitable loss of fidelity. Sounds that are transmitted through predictable habitats should lead to predictable attenuation and degradation (Naguib and Wiley, 2001). If animals can adapt their vocalizations to this challenge, the acoustic adaptation hypothesis predicts that a species’ call should be best transmitted in its own habitat (Morton, 1975). This hypothesis was developed by ornithologists to explain the striking difference in birdsongs and birdcalls in dense forests (where they are frequently relatively tonal) as opposed to

MARMOT ALARM COMMUNICATION

377

Fig. 4. Average spectrogram correlation values for 3‐kHz pure tones broadcast through eight marmot species’ habitats illustrating significant differences in the transmission fidelity of the different habitats.

more open grasslands where they are more ‘‘buzzy.’’ Support for the hypothesis comes from intraspecific studies of tits (Hunter and Krebs, 1979) and sparrows (Handford, 1988; Handford and Lougheed, 1991). Interspecific studies supporting the hypothesis (Wiley, 1991) have relied on large data sets of many species and have found that the strongest effects were from the most different habitat types (dense forests and grasslands) suggesting that smaller differences in environments may not have the same effect (Blumstein and Turner, 2005). Janice Daniel and I tested the hypothesis that a marmot species’ alarm call is best transmitted in its own habitat (Daniel and Blumstein, 1998). First, we broadcast and rerecorded 3 kHz pure tones (chosen because this is the dominant frequency of marmot alarm calls) as well as a species’ own calls through a number of social groups within a species’ habitat. Calls were broadcast and rerecorded at 1, 10, 20, 30, and 40 m from the speaker. We used spectrogram correlation (Clark et al., 1987), a technique that compares the structure of two spectrograms, to quantify change in the original signal (i.e., that rerecorded at 1 m) when transmitted 10, 20, 30, or 40 m through the environment. We found significant intraspecific variation in the acoustic transmission properties of the habitats of the three species studied (golden marmots, Alpine marmots, and yellow‐bellied marmots—Blumstein and Daniel, 1997). Subsequent analyses have found significant variation in eight species’ habitats (Fig. 4). For pure tones, interspecific variation was greater than intraspecific variation in habitat transmission fidelity (Blumstein and Daniel, 1997).

378

DANIEL T. BLUMSTEIN

To directly test the acoustic adaptation hypothesis, we systematically broadcast four different species’ calls (hoary, Olympic, woodchucks, and yellow‐bellied) in the four different species’ habitats (Daniel and Blumstein, 1998). If the acoustic adaptation hypothesis explained the variation in the structure of these species’ calls, we expected that each species call would be best transmitted in its own habitat. A significant interaction between habitat and call type in spectrogram correlation values would provide support for the acoustic adaptation hypothesis. Using MANOVA, we found significant habitat and species effects, but there was no significant interaction: a species’ call is not best transmitted in its own habitat. Importantly, the effect size of the call (partial
2 ¼ 0.80) was larger than the effect size of the habitat (partial
2 ¼ 0.31) and this was much larger than the effect size of the interaction (partial
2 ¼ 0.08). Together, the results suggest that there are some calls that transmit well and some that transmit poorly, and that there is substantial variation in the transmissibility of habitats, but that the habitat has little effect on the evolution of differences in call microstructure. By examining acoustic variation in isolated populations, among phylogenetically close relatives, and by examining the results of studies where sciurids have been isolated on islands, it is possible to hypothesize that drift, not selection, may be responsible for microstructural changes in call structure (Blumstein, 1999a; Daniel and Blumstein, 1998). For instance, long‐tailed marmot calls (the golden marmots that I studied are a named subspecies of the long‐tailed marmot) are geographically variable and this variation seems to be associated with a pattern of isolation by glaciation (Nikol’skii et al., 1999). Sibling species, such as the hoary and Olympic marmot, have acoustically similar calls as do the closely related steppe marmot and gray marmot (M. baibacina). Finally, the structure of squirrel alarm calls, isolated by sea level changes on islands, has begun to diverge in as few as 7500 years (Nikolsky, 1981). Given the potential importance of drift, it is surprising that we found no geographic variation in yellow‐bellied marmot alarm calls studied at three locations in Utah and Colorado (Blumstein and Armitage, 1997a). However, recent molecular evidence found substantial gene flow even between isolated populations, perhaps providing the solution to this puzzle (Floyd et al., 2005). If drift explains variation in the call microstructure, what explains variation in repertoire size? I studied three factors that could influence repertoire size: the acoustic environment, home range size, and sociality. I ranked the acoustic transmission characteristics for eight species by summing the spectrogram correlation values at 10, 20, 30, and 40 m. I hypothesized that species living in acoustic environments that better

MARMOT ALARM COMMUNICATION

379

allowed the transmission of 3‐kHz pure tones could potentially communicate more complex information and thus should have larger call repertoire sizes. In these and the following analyses, I fitted regressions on raw data and on phylogenetically independent contrasts; results were similar. Either way analyzed, there was no relationship between the transmission fidelity of a habitat and the number of alarm call types (Fig. 5). I collected home range size estimates from the literature and hypothesized that species living in larger home ranges would need to communicate more precise information and thus might have larger call repertoires. This might be expected because when distances between signalers and receivers are large, there is likely to be greater uncertainty about the true risk. If we assume that callers directing calls to conspecifics would benefit by the reduced ambiguity of using acoustically different calls, then we would expect that species with larger home ranges would have a larger call repertoire. In the raw data, there was a weak positive relationship between home range size and call repertoire size (Fig. 5). However, when results were examined in a contrast‐based analysis, this relationship disappeared. The conservative interpretation is thus that there was no relationship between home range size and call repertoire size. I used the Blumstein and Armitage (1997b, 1998) social complexity metric to quantify sociality. This metric focuses on the demographic roles present in a social group (adult males/females, 2‐year‐old males/females, yearling males/females, and pups) and uses information theory to quantify the variation in social structure. This acknowledges that social complexity requires some description of the number of roles and the number of individuals, and it acknowledges that social complexity emerges from variable social situations. Because species also vary in the time to natal dispersal, and thus social groups vary in the degree of relatedness within them, the number emerging from the information theory analysis is multiplied by the time to natal dispersal. By doing so, the metric assumes that kin groups are more socially complex than nonkin groups. In both analyses of species values and in contrast‐based analyses, I found that more socially complex species produce more alarm call types (Blumstein, 2003). In an analysis of independent contrasts that controlled for variation in alarm call repertoire size explained by transmission fidelity (one‐tailed p ¼ 0.49), 57% of the variation in repertoire size was explained by social complexity (one‐tailed p ¼ 0.04). Similarly, after accounting for nonsignificant variation explained by home range size (one‐tailed p ¼ 0.37), 59% of the variation in repertoire size was explained by social complexity (one‐ tailed p ¼ 0.05). Thus, it seems that social complexity is relatively more important than either the acoustic habitat or the home range size in explaining variation in alarm call repertoire size in the marmots studied

380

DANIEL T. BLUMSTEIN

A

vancouverensis 5 olympus

caligata

flaviventris

marmota

4 3 2 monax

caudata 1 0.5

0.6

0.7

0.8

0.9

1.0

Transmission fidelity B vancouverensis

Number of alarm call types

5

caligata

olympus

4 3

camtschatica

marmota 2

flaviventris

1 monax/ bobak/ sibirica/ baibacina/ caudata 0

4

8 12 Home range (ha)

14

C vancouverensis 5 olympus 4

caligata

3 flaviventris

2 monax

caudata

1 0.0

0.3

marmota

0.6

0.9

1.2

1.5

Social complexity Fig. 5. Bivariate relationships between (A) habitat transmission fidelity, (B) home range size, or (C) social complexity and alarm call repertoire size (number of alarm calls) in marmots. [Modified in part with permission from Blumstein (2003).] Species values are plotted. In an analysis of independent contrasts that controlled for variation in alarm call repertoire size

MARMOT ALARM COMMUNICATION

381

to date. These results suggest that social complexity (see also Blumstein and Armitage, 1997b), not simply the need to communicate over long distances, selects for complex communication. In summary, alarm calling seems to have initially evolved as a means of detection signaling to predators. Conspecific warning functions are thus an exaptation, the adaptive utility of which will be discussed in the next section. Because some species make themselves more obvious as predation risk increases, there is likely a dual function of alarm calling. The structure of alarm calls is variable and this variation is likely to reflect drift processes, rather than selection from the acoustic environment. Marmots produce different numbers of alarm call types and a significant amount of variation in repertoire size is explained by sociality. III. FUNCTION The apparent paradox I discussed in the introduction, as to why animals should produce potentially costly vocalizations that warn others, will be discussed here. I confess at the outset that I have not demonstrated that emitting calls increase predation risk (in thousands of hours watching marmots, I have only seen one complete predation event, and in many more thousands of hours, my assistants have only seen a few successful predatory episodes on adults), nor did I specifically document other costs of calling. In theory, alarm calls could have an energetic cost, but even studies of birdsong do not always quantify substantial energetic costs (Ward et al., 2003), and alarm calls are relatively rare events and thus should be even less costly. For instance, in yellow‐bellied marmots, we document a bout of calling every 2.1 h of observations (between 2002 and 2006, we noted 1677 bouts of calling in 3553 h of direct observation). However, individuals that call remain vigilant and reduce time allocated to other activities when they call. Individuals that respond to calls trade‐off foraging and other activities with antipredator vigilance. Thus, if we assume that calling (and responding to calls) has some opportunity cost (Blumstein, 2007), let us now focus on potential benefits from calling. I previously discussed potential benefits from calls directed to predators. Most species of marmots do not engage in contagious calling where the calls of one individual elicit calls from other individuals (e.g., as

explained by transmission fidelity (one‐tailed p ¼ 0.49), 57% of the variation in repertoire size was explained by social complexity (one‐tailed p ¼ 0.04). Similarly, after accounting for nonsignificant variation explained by home range size (one‐tailed p ¼ 0.37), 59% of the variation in repertoire size was explained by social complexity (one‐tailed p ¼ 0.05).

382

DANIEL T. BLUMSTEIN

found in some primates––Zuberbu¨hler, 2001). Contagious calling is consistent with the hypothesis that callers were signaling detection and were thus directing their calls at the predator where each individual benefits (Zuberbu¨hler et al., 1999). The one exception is the Olympic marmot, which in response to playback sometimes called (Blumstein, 1999a). In all species I have studied, it is not uncommon to only hear one or a few individuals’ alarm calls as a predator passes through an area. But marmots do respond to the calls of other marmots by increasing their vigilance and retreating to their burrows. Here, I focus on the adaptive utility of calls directed to conspecifics while acknowledging that calls could simultaneously function to deter pursuit (Caro, 2005). Reciprocal altruism (Trivers, 1971) appears to be unimportant in alarm‐ calling systems. There is no evidence from any sciurid rodent alarm‐calling system that individuals engage in reciprocal bouts of calling (Blumstein, 2007). In some respects, this is puzzling because calls are often sufficiently individually identifiable (see the description later), animals are able to remember important attributes associated with callers (see the description later), and because such a system would limit the cost to any given caller. In other respects, this is not puzzling because calling is a relatively rare and important thing to do, and because there is no guarantee that a recipient today will be around to warn an actor tomorrow. Moreover, reciprocity works best when there is a direct transfer of benefits between two individuals; eavesdropping by multiple recipients destabilizes the process (Blumstein, 2007). Finally, the legacy of calls being directed to predators may constrain calling when there is a direct benefit from doing so. That said, the sight of a predator does not inevitably elicit calling. In all the species of marmots I have studied, not all individuals call when a predator is within sight. It seems that individuals do not emit alarm calls unless they themselves are safe. This is in contrast to what has been reported in Belding’s ground squirrels (Sherman, 1985), other mammals (Caro, 2005), and some birds (Cresswell, 1993) which may call while being pursued by particularly threatening predators. By contrast, marmots do not call first and then run to safety. Rather, they retreat to safety, increase their vigilance, and then decide whether or not to emit an alarm call. So what influences the probability of calling? One factor that may influence the probability of calling is the caller’s endocrine state. We used a ‘‘trap‐calling assay’’ where we noted whether individuals called when we approached them in traps. Importantly, subjects who are more likely to call when in a trap are more likely to call when approached or when they encounter a predator (Blumstein et al., 1997). We amassed a data set of 29 breeding‐age females who called on one occasion, but not on another, and we compared fecal glucocorticoid levels on these

MARMOT ALARM COMMUNICATION

383

100

Glucocorticoid metabolite concentration (ng/g)

p = 0.007 80

60

40

20

0 Did not call

Called

Fig. 6. Fecal corticosteroid metabolites in adult female yellow‐bellied marmots on an occasion when they emitted a call and on another occasion when they did not emit a call. [Modified from Blumstein et al. (2006a) and used with permission from the Royal Society.]

two occasions (Fig. 6). Our results were not confounded by some factors known to influence glucocorticoid levels, including time of day, age, breeding status, and time of season. We found that fecal glucocorticoid metabolites in individuals that emitted calls on one day but not on another had systematically higher levels of this ‘‘stress’’ hormone on days when they called (Blumstein et al., 2006a). State, specifically stress level (as estimated by glucocorticoid metabolites), thus provides a mechanism that could explain why some individuals are more likely to call when particular relatives are within earshot (see the description later). State also provides a plausible mechanism that could influence caller reliability (discussed later). Building on excellent studies from Sherman (1977) and Hoogland (1995) that showed that Belding’s ground squirrels and black‐tailed prairie dogs are exquisitely sensitive to their audience and modify call production based on the presence or absence of both direct and indirect kin, we (Blumstein et al., 1997) asked whether and how yellow‐bellied marmot call production is sensitive to the presence of indirect kin within earshot. In a series of analyses that looked at how the propensity of calling was influenced by the presence, absence, or number of conspecifics, we found that female yellow‐ bellied marmots substantially, significantly, and somewhat uniquely increased the rate of calling after they had emergent young (Table I). Our observations do not support the hypothesis that the presence of indirect kin influenced either the rate of calling or the likelihood of calling [e.g., only mothers increased calling in response to the presence of vulnerable pups in

384

DANIEL T. BLUMSTEIN

TABLE I* CALLING RATES BEFORE AND AFTER PUP EMERGENCEa Adult female (mother)

Adult females

Adult males

Yearling females

Yearling males

1995 Rate < Rate > Rate (total)

N¼4 0.05 (0.04) 0.31 (0.12) 0.18 (0.07)

N ¼ 10 0.08 (0.05) 0.04 (0.08) 0.06 (0.06)

N¼3 0.06 (0.05) 0.01 (0.01) 0.03 (0.02)

N¼6 0.03 (0.02) 0.00 (0.00) 0.01 (0.01)

N¼5 0.05 (0.02) 0.07 (0.06) 0.06 (0.03)

1980, 1985, 1991 Rate < Rate > Rate (total)

N ¼ 20

N¼9

N¼6

N ¼ 14

N ¼ 12

0.05 (0.07) 0.02 (0.03) 0.03 (0.04)

0.02 (0.04) 0.04 (0.11) 0.02 (0.03)

0.01 (0.02) 0.02 (0.04) 0.01 (0.02)

0.02 (0.05) 0.08 (0.10) 0.05 (0.08)

0.04 (0.03) 0.003 (0.01) 0.01 (0.02)

Alarm‐calling rates (bouts of alarm calls/h  SD) before pups emerged above ground (rate <), after pups emerged above ground (rate >), and the overall seasonal average (total rate during the two periods over which marmot alarm calling was quantified: 1995 summer only, and the composite of 1980, 1985, and 1991). Note the substantial (and significant) increase in calling by adult females (bold type) after pups emerged. *Table from Blumstein et al. (1997) and used with permission from Elsevier. a

the social group; the number of other group members (likely to be relatives) had no effect on calling rate, etc.]. Taken together, these results suggest that calling is nepotistic (Sherman, 1980a,b). However, unlike other systems where calling is nepotistically directed to collateral kin, in yellow‐bellied marmots, nepotistic behavior is directed to vulnerable young (Blumstein et al., 1997). It is possible that the general patchiness of yellow‐bellied marmot colonies and their relatively limited size provides fewer opportunities for nepotism. By contrast, prairie dogs and Belding’s ground squirrels often live in much denser populations. Ultimately, it seems that these demographic differences may explain interspecific variation in the adaptive utility of alarm communication (Blumstein, 2007). Of course individuals other than mothers produce alarm calls, albeit less frequently. What is their function? Some of it could be detection signaling to predators. These calls also function to warn conspecifics because conspecifics clearly respond to callers (Blumstein and Daniel, 2004). Ultimately, teasing apart the variation explained by these audiences is a fundamentally important question that remains unanswered in any calling species. In summary, alarm calling appears to have limited costs but some benefits. A primary function of yellow‐bellied marmot alarm calling is for mothers to warn vulnerable young; calling by them is a form of parental care. Elevated basal glucocorticoid levels in mothers with young may

MARMOT ALARM COMMUNICATION

385

provide a proximate mechanism underlying this pattern. This hypothesis remains to be formally tested, and experiments must be conducted to properly identify causality. The adaptive significance of calling by non‐ mothers may be different, and identifying the relative importance of conspecific and heterospecific audiences remains to be clarified.

IV. MEANING Alarm calls could potentially contain information about: the species producing the call, the predator type eliciting the call, the risk a caller experiences when it calls, and caller’s age, sex, condition, and identity. My interests in alarm calling originally stemmed from a desire to study the evolution of complex communication and language (Cheney and Seyfarth, 1990; Evans and Marler, 1995; Snowdon, 1993). I adopted a trait‐based view of language evolution and focused on referential ability (Blumstein, 1999b). Human language is notable in the degree to which we can assign arbitrary acoustic labels to types of stimuli (Hockett, 1960) and the ability to communicate referential information is amenable to comparative study if one studies more than a single species. Referential signals communicate information about environmental events, objects, or perhaps actions (Evans, 1997). Marler et al. (1992) coined the term ‘‘functional reference’’ to focus specifically on animals’ abilities and to avoid any connotation of higher‐ level representational cognitive abilities. I believe this distinction is important because it is theoretically possible to have referential abilities without having the ability to form representations, and the methods to study representations are more rigorous than those that can be used to study functional reference (Evans, 1997). Like human speech, it is important to realize that referentiality is one type of information that could be encoded in an alarm signal; other types include the degree of risk or an individuals’ arousal (Manser et al., 2002). To study functional reference, we must document two things (Marler et al., 1992): the degree of a call’s production specificity and a call’s contextual independence (or ‘‘perception specificity’’). A call with a high degree of production specificity is one that is elicited by a narrow range of stimuli. Because some marmot species were said to have predator‐specific calls (Heard, 1977; Lenti Boero, 1992), while others did not, marmots seemed to be an ideal system to study the evolution of referential abilities. However, the second criterion is essential to demonstrate functional reference: calls, divorced from other contextual information, should reliably elicit the appropriate response to a particular referent.

386

DANIEL T. BLUMSTEIN

As I systematically elicited calls from a variety of species and conducted playback experiments to determine the degree to which uniquely different calls could elicit different responses, I discovered that none of the marmots I studied had functionally referential calls (Blumstein, 1995a,b, 1999a,b; Blumstein and Armitage, 1997a; Blumstein and Arnold, 1995). None of the species had sufficiently high production specificity to suggest that calls were functionally referential. However, I also learned that marmots used a variety of mechanisms to communicate risk (Table II). Such response urgency or risk‐based communication is well known from ground‐dwelling sciurid rodents (Owings and Hennessy, 1984; Robinson, 1981; Sloan and Hare, 2004; Warkentin et al., 2001). Golden and alpine marmots packaged calls into multinote signals that covaried with the degree of risk a caller experienced. Olympic, hoary, and Vancouver Island marmots had different call types, some of which covaried with the stimulus class (aerial or terrestrial) that elicited the calls. Most species responded to call type variation, while some did not (Blumstein, 1995b). None of the species responded in a way that suggested that unique calls elicited unique responses. And, I discovered things that I did not expect. Vancouver Island marmots appeared to have a simple form of syntax: the order in which different call types were produced influenced response (Blumstein, 1999a). Discovering this remarkable variation in mechanism suggested two important things. First, as studied so well in some insect and anuran systems (Gerhardt and Huber, 2002), communication mechanisms can evolve quickly: closely related species may have different mechanisms as illustrated by contrasting yellow‐bellied marmots (two call types, rated‐based

TABLE II DIVERSITY OF MECHANISMS USED BY MARMOTS TO COMMUNICATE RISK Species

No. calls

M. monax M. bobak

1 1

M. caudata aurea M. flaviventris M. marmota M. olympus M. caligata M. vancouverensis

1 2 2 4 4 5

Mechanism a

Rate Package Number/rate Package (number?) Number/rate & call type Rate? Call duration and bout composition

References Unpublished observation Unpublished observation, Nikol’skii et al., 1994 Blumstein, 1995a,b Blumstein and Armitage, 1997a Blumstein and Arnold, 1995 Blumstein, 1999a Blumstein, 1999a Blumstein, 1999a

a M. monax rarely emit alarm calls and I was unable to conduct playbacks to study mechanism.

MARMOT ALARM COMMUNICATION

387

mechanism) and Olympic marmots (four call types, rate‐based mechanism). Second, it is perhaps more profitable to study what and how animals communicate, rather than trying to force a human construct into a nonhuman system. This is an important lesson that still is not fully appreciated by those searching for ‘‘word’’ analogs in nonhuman signals. The signal may be an utterance, or the signal may have a longer time course. We must let the species under study tell us where to look for meaning. While I value teasing apart the information contained in signals, to do so we may have to look beyond word‐like analogies. In general, we should expect different attributes of signals to contain information about different things (Lambrechts and Dhondt, 1995; Marler, 1960), and the marmot alarm calls illustrate this nicely. By making a number of acoustic measurements, we found that in addition to predation risk, yellow‐bellied marmot alarm calls also contain potential information about caller’s age, sex, and identity (Blumstein and Armitage, 1997a; Blumstein and Munos, 2005). Interestingly, these attributes are encoded differently. Risk is communicated by varying the rate and number of calls produced (Blumstein and Armitage, 1997a). Microstructural differences encode sex, age, and identity (Blumstein and Armitage, 1997a; Blumstein and Munos, 2005). In a large data set containing multiple calls recorded from individuals on different occasions, stepwise discriminant function analysis classified over 62% of calls correctly to individual, compared with a less than 1% randomly expected classification (Blumstein and Munos, 2005). By broadcasting and rerecording calls through marmot habitat, Olivier Munos and I also discovered that the acoustic characteristics that encode identity degrade less than a randomly selected set of acoustic characteristics (Blumstein and Munos, 2005). This result provides some suggestion that there has been selection to encode individuality rather than it simply being an unselected by‐product of laryngeal morphology (Fitch and Hauser, 1995, 2002; see the description later). Surprisingly, there have been no studies that have focused strictly on sexual selection in marmots. It is conceivable that alarm calls can be used in mate choice decisions. Cockerels that emit risky aerial alarm calls have higher reproductive success than cockerels that emit fewer risky aerial alarm calls (C. S. Evans, unpublished data). As with other species (Clutton‐Brock and Albon, 1979; Davies and Halliday, 1978), it is conceivable that the acoustic characteristics of marmot alarm calls may reflect body condition or size, or simply emitting calls indicate good condition. These are areas open for exploration. Moving away from marmots, we can ask the broader question of what selects for different alarm call types. One hypothesis is that the need to communicate about mutually incompatible escape options may be

388

DANIEL T. BLUMSTEIN

important in selecting for functional reference (Macedonia and Evans, 1993). Specifically, if the best escape strategy for evading one predator is to freeze and lay low, while the best escape strategy for another predator is to retreat into a burrow, and if signalers can benefit from communicating this to conspecifics, then we should expect different alarm call types. We see this in fowl: cockerels produce aerial alarm calls, which are relatively cryptic, are associated with the signaler freezing, and also cause recipients to freeze upon hearing them, and terrestrial alarm calls, which are associated with the signaler making himself conspicuous once in a safe location and also cause recipients to flee to safe locations (Evans et al., 1993). The mutually incompatible response hypothesis also explains differences in lemur species that may or may not produce functionally referential calls (Macedonia and Evans, 1993) and seemingly explains why vervet monkeys, some other primates (Cheney and Seyfarth, 1990; Zuberbu¨hler, 2000), a prairie dog (Kiriazis and Slobodchikoff, 2006), a social mongoose (Manser, 2001; Manser et al., 2001, 2002), and a tree squirrel (Greene and Meagher, 1998) have functionally referential calls. I suggest that a comprehensive hypothesis about the evolution of functional reference will combine sociality (which demonstrably is associated with alarm call repertoire size in sciurid rodents) and the mutually incompatible response hypothesis. Such a hypothesis will also have to explicitly acknowledge the two‐step process of evolving production specificity and contextual independence (Blumstein, 1999b). In summary, marmots encode the relative predation risk a caller faces when they call; there is limited referential information contained in their calls. Marmots communicate risk using a variety of mechanisms. Some of these mechanisms potentially communicate risk immediately (producing different call types) while others require a longer time frame for a receiver to extract information about risk (varying the number or rate of calls). In addition to risk, other potentially useful information is also contained in marmot alarm calls. Identifying the potential information in calls and the mechanisms by which it is encoded should lead to a rich understanding of communication.

V. INDIVIDUALITY AND RELIABILITY Why should alarm calls be individually specific? From a proximate standpoint, laryngeal variation will inevitably lead to some acoustic variation (Fitch and Hauser, 1995, 2002). Is this acoustic variation useful? Individual‐specific contact calls (Rendall et al., 1996; Wanker and Fischer, 2001) maintain group stability, while the individual‐specific calls that creching birds and mammals

MARMOT ALARM COMMUNICATION

389

produce (Insley, 2000; Jouventin et al., 1999; Leonard et al., 1997) function to allow parents to reunite with their offspring. In these signaling systems, we should expect there to be selection on both the signaler and the receiver (Searcy and Nowicki, 2005). Selection on the signaler will lead to more distinctive vocalizations, while selection on the receiver will lead to an ability to discriminate among them. Alarm calls are a bit different because while it is easy to hypothesize the benefits of calling, and of responding to a call, it is more difficult to think about why selection should select for signalers to make distinctive calls when the desired response is to simply warn a vulnerable conspecific or tell a predator that it is been detected (Blumstein, 2007; Blumstein et al., 2004). To address this problem, let us consider both the signaler and the receiver in more detail, and let us assume that individuals differ in the reliability with which they emit calls. Variation in caller reliability is easy to envision if we acknowledge that experience or endocrine state may influence the likelihood of calling. Consider a naı¨ve individual that calls to many stimuli; with experience, fewer stimuli potentiate calls (Cheney and Seyfarth, 1990). If stress levels potentiate calling, and if individuals vary in their stress levels, some are more or less likely to call in any given situation. If stress is unrelated to the true risk of predation, then we can easily envision that stressed subjects may call when there is no risk of predation. In this circumstance, from the receivers’ perspective, discriminating reliable callers from unreliable callers is essential so as not to waste time responding to false alarms. From the signalers’ perspective, we still need to know what benefits they might obtain by emitting individually distinctive calls. Several lines of evidence suggest that there has been selection on signalers to make distinctive calls. First, acoustic characteristics that allow us to statistically discriminate between individuals degrade less than a randomly selected set of characteristics (Blumstein and Munos, 2005). Second, as I will discuss later, receivers can discriminate among individuals (Blumstein and Daniel, 2004). Third, more social species seem to have more information content about individuality contained in their calls than less social species (K. Pollard and D. Blumstein, unpublished data). Using information theory, Beecher derived a method to calculate the number of bits of information about individuality contained in a signal (Beecher, 1989a,b). Using these ‘‘Beecher statistics,’’ Oliver Munos and I found that there were about 3.4 bits of information about the individual contained in yellow‐bellied marmot alarm calls (Blumstein and Munos, 2005). Importantly, this information content would allow marmots to distinguish a maximum of 10 individuals (i.e., 23.4 ¼ 10.5 individuals)—a number that is consistent with the number of permanent residents in yellow‐bellied marmot groups. Working with the more

390

DANIEL T. BLUMSTEIN

social Olympic marmot, we discovered that these marmots have about 5.6 bits of information about individuality contained in their calls (K. Pollard, S. Cox Griffin, and D. T. Blumstein, unpublished data). Future comparative work will formally test the hypothesis that the evolution of sociality is correlated with the evolution of information contained about individuality; preliminary results are encouraging. Finally, Kim Pollard has developed a stochastic dynamic model suggesting that if callers are surrounded by relatives, and if individuals vary in their reliability, kin selection could select for greater individual distinctiveness (K. Pollard, unpublished data). Yellow‐bellied marmots are able to discriminate individuals solely on the basis of their reliability (Blumstein et al., 2004). To demonstrate this, we conducted a habituation‐recovery experiment (Evans, 1997), where we associated the calls of one individual with a threatening stimulus (and thus created a reliable caller) and the calls of another subject with no threatening stimulus (and thus created an unreliable caller). The threatening stimulus was a taxidermic mount of a badger (a marmot predator), and the unthreatening stimulus was the badger covered with a tarp. The habituation series consisted of 10 min of rapidly paced calling with or without the badger present. For such learning experiments, it is essential for pretest and posttest subjects to different exemplars of the habituation series. The key comparison is how a subject’s response to calls from the reliable and unreliable callers changes as a function of the habituation series. Thus, it is a before‐ after treatment‐control within‐subjects design. Our assay for all these experiments was the amount of time subjects continued to forage on a handful of bait. When not foraging, marmots looked around; in some cases they disappeared into their burrow. We found that marmots are exquisitely good at detecting variation in caller reliability. A single 10‐min exposure to an unreliable caller (a subject whose calls were broadcast while the badger was covered with a tarp) was sufficient for marmots to respond differently (Fig. 7). What was particularly extraordinary was how marmots responded. We expected that yellow‐ bellied marmots, like some primates (Cheney and Seyfarth, 1988, 1990; Gouzoules et al., 1996; Ramakrishnan and Coss, 2000) and other sciurid rodents (Hanson and Coss, 2001; Hare and Atkins, 2001; Nesterova, 1996) would follow the ‘‘crying wolf’’ phenomenon: unreliable individuals would elicit reduced vigilance because marmots had learned that they are unreliable. What we found was the exact opposite. Unreliable callers elicited more vigilance. Results from a second experiment were consistent with this finding: marmots hearing degraded calls (which presumably provided less reliable information about the true risk of predation), suppressed foraging and were more vigilant (Blumstein et al., 2004).

391

Difference (post-test—pre-test) in proportion of time allocated to foraging

MARMOT ALARM COMMUNICATION

0.8

p = 0.066

0.6 0.4 0.2 0 −0.2 −0.4 −0.6 Reliable caller

Unreliable caller

Fig. 7. Response of yellow‐bellied marmots (posttraining––pretraining) to a reliable caller and an unreliable caller. [Modified from Blumstein et al. (2004) and used with permission from the Royal Society.] Note: Because of the potentially disturbing nature of this experiment, for ethical reasons the sample size was limited and we (a priori) set our alpha to 0.1.

I interpret these results as being an alternative way of dealing with uncertainty––when unsure, allocate more time to independent investigation. Unreliable individuals are unreliable specifically because they do not communicate the true risk of predation. Why then would a reliable individual elicit a shorter bout of investigation? Perhaps because looking around a bit was sufficient to know that they must have made a mistake. The overall conclusion parallels lessons from insects and anurans (Gerhardt and Huber, 2002): mechanisms of perception may evolve rapidly and are likely to be as variable as mechanisms of production. In summary, yellow‐bellied marmots and other sciurids produce individually distinctive alarm calls. Individuality seems to correlate with social complexity: the calls of species that are more social contain more bits of information about individuality within them. One function of individually distinctive alarm calls is to allow recipients to differentiate subjects based on their reliability. Surprisingly, yellow‐bellied marmots illustrate a unique way by which animals may respond to unreliable subjects. Rather than reducing responsiveness to unreliable callers, they increase responsiveness. Thus, in addition to discovering a novel mechanism to respond to reliability, these results illustrate that, as in insects and anurans, the mechanisms by which individuals respond to signals are somewhat plastic.

392

DANIEL T. BLUMSTEIN

VI. APPLIED RELEVANCE OF ALARM‐CALLING BEHAVIOR ‘‘Conservation behavior’’ is the application of general principles and insights of behavioral biology to conservation biology (Blumstein and Ferna´ndez‐Juricic, 2004). Previous applications of communication to conservation have focused on using individually discriminable vocalizations to help census populations (Baptista and Gaunt, 1997; McGregor et al., 2000), estimate condition (McGregor et al., 2000), identify population differences (McGregor et al., 2000), and study the effects of anthropogenic noise (Brumm and Slabbekoorn, 2005; McGregor et al., 2000). Clearly, the substantial individual‐specific variability in the alarm calls of yellow‐bellied marmots (Blumstein and Munos, 2005) and many other species (Davidson and Wilkinson, 2002; Hare, 1998; Nikol’skii and Suchanova, 1994; Rendall et al., 1996; Semple, 2001) suggests that this indeed could be a useful tool to noninvasively census populations. While noninvasive censusing is important, there are other previously unappreciated conservation benefits from understanding alarm‐calling behavior. Fifty percent of marmot species are either IUCN red‐listed or are otherwise of conservation interest. The Vancouver Island marmot is one of the most endangered species in the world; at the time of writing, fewer than 50 individuals remained in the wild and managers are using captive breeding and reintroduction to try to recover the population (Bryant, 2005; Bryant and Page, 2005). One early lesson from my Vancouver Island marmot work was that this species had a full repertoire of antipredator behavior, including the ability to produce more alarm call types than any other species (Blumstein et al., 2001). More recent work found that calling and other antipredator behavior apparently persists in captivity (Blumstein et al., 2006b). Thus, alarm calling, and the propensity to alarm call, is a behavior that can be used to understand whether behavior has changed in the relaxed selection induced by some captive breeding programs. The observation that calls initially evolved as a form of detection signaling has an important conservation message: range expansions by novel predators could have deleterious consequences because novel predators may not have evolved to play the same games (Hugie, 2003) as alarm‐calling species. We know that species’ ranges change naturally (Kirkpatrick and Barton, 1997), and unnaturally (through anthropogenic habitat changes and deliberate and accidental translocation—Long, 2003; Low, 1999). We also know that there are likely to be deleterious consequences for prey species when predators change their ranges (Berger et al., 2001; Low, 1999). We should be wary of ritualized interactions that prey may have with their predators, particularly when predators did not evolve with them. Thus, alarm‐calling

MARMOT ALARM COMMUNICATION

393

species may be especially vulnerable to the introduction of novel predators. This counterintuitive suggestion requires proper study. The fecal glucocortioid metabolite results have at least two implications for conservation. First, if calling is a risky behavior in that it exposes animals to predators, we must be very careful about stressing free‐living animals—it could increase their exposure. Second, calling may be an indicator of stress (see also Bercovitch et al., 1995; Boinski et al., 1999; Norcross and Newman, 1999) that can be used in both captivity and in the wild. And, if stress is associated with reproductive failure (Abbott et al., 1997; Wasser, 1999), calling in captivity may be a noninvasive indicator that individuals are stressed. In the wild, sudden increases in calling behavior may help identify anthropogenic stressors. Of course, such relationships will be more easily detected in longer term studies of individually identified subjects. Finally, by studying marmot alarm‐calling behavior, we can gain insights into national defense (Blumstein, in press‐a). This may seem far‐fetched, but in 2005, I participated in an interdisciplinary workshop on Darwinian Security hosted by the National Center for Ecological Analysis and Synthesis. An edited volume (Sagarin and Taylor, in press) emerged from our working group. The book develops connections between behavior, ecology, and evolutionary biology and national defense. In the context of defense, we need to know how to respond to unreliable sources of information (which may come from human intelligence or from signals from adversaries or interactants). A lesson from marmots is that unreliable sources should elicit independent investigation. Another lesson is that detection signaling is an effective mean of reducing risk. In the context of defense, we should announce the discovery of terrorist plots with the aim that this would force terrorists to change operational methods or targets. In summary, a fundamental understanding of the proximate basis, evolutionary history, and function of calling has at least five important applied implications. First, it provides the tools for noninvasive population censuses. Second, it helps us evaluate a species’ antipredator abilities. Third, it provides a warning about a possible negative outcome from predator range shifts. Fourth, it gives us methods to noninvasively measure stress in captivity and in the wild, and may thus provide information that can help identify a cause of reproductive failure. Finally, it has implications for national defense.

VII. SUMMARY AND FUTURE WORK I have adopted a Tinbergian route to study alarm communication in marmots. Alarm signals probably evolved as a means to signal detection to predators and have become exapted into a conspecific warning system.

394

DANIEL T. BLUMSTEIN

Marmots have acoustically divergent alarm calls and species produce between one and five alarm call types. The acoustic environment probably explains limited variation in call structure. Alarm call repertoire size is explained by social complexity; more social species produce more types of alarm calls. Much work has focused on yellow‐bellied marmots. The current adaptive utility of yellow‐bellied marmot calling seems to be to warn vulnerable offspring, although some additional variation in calling may be explained by signaling to the predator. Calls are individually distinctive and this distinctiveness allows marmots hearing calls to modify their response based on caller reliability. Glucocorticoids appear to potentiate calling: individuals that called on one occasion but not another had systematically higher fecal glucocorticoid levels when they called. My alarm calling results have implications for studying and managing threatened or endangered marmots: populations can be acoustically censused, antipredator abilities can be documented, and stress can be noninvasively measured in captivity and in the field. While I have focused on alarm communication, marmots obtain other information about risk acoustically. Many species respond to the sounds which their predators make. A phylogenetic reconstruction of the ability to respond to the sound of predators predicts that yellow‐bellied marmots should respond to predators. Recent playback experiments suggest that they do, and experiments further suggest that marmots have an innate ability to respond to the sounds of locally extinct predators (Blumstein et al., 2007). Future work will focus on the specific acoustic cues that marmots use and which enable them to respond to novel predators. Sometimes, when we handle a yellow‐bellied marmot pup within about 10 days of emergence, it gives a long, disturbing scream. These screams are unique and structurally different from their alarm calls. Interestingly, these screams have noticeable nonlinearities. It has been hypothesized that vocalizations with nonlinearities may be more difficult to habituate to (Fitch et al., 2002). Future work will directly test this hypothesis by adding nonlinearities to calls and studying habituation. My past work has completely ignored an important Tinbergian dimension: the ontogeny of calling. All species I studied seem to be able to emit fully formed (although relatively higher frequency) calls about the time they first emerge from their natal burrows (i.e., by about the time they are a month old). The frequency, but not obviously the structure, changes as their body size increases. Thus, calls appear not to be learned. The context of calling, however, appears to be amenable to some degree of experience (Shriner, 1999). Future work may explore this.

MARMOT ALARM COMMUNICATION

395

Acknowledgments Major support for this research came from the University of California Davis (through a variety of graduate fellowships); the US National Institute of Health (NRSA MH10793); an Australian Research Council Postdoctoral Fellowship; the University of California Los Angeles (Assistant Professors’ Initiative, COR grants, and Life Science Deans’ Recruitment and Retention Funds), and the National Geographic Society. I thank the many students, assistants, and collaborators who have helped develop this story. I am particularly grateful to Chris Evans, James Hare, Joe Macedonia, Marta Manser, Peter Marler, Alexander Nikol’skii, Don Owings, and Judy Stamps for extensive and ongoing discussions about alarm communication, and to the editors (Mitani, Roper, and Wynne‐Edwards) for clarifying comments on a draft of this review. I thank Kim Pollard for allowing me to report preliminary results from her dissertation, as well as for comments on previous versions. I especially thank Janice Daniel, my wife and collaborator on many of these studies.

References Abbott, D. H., Saltzman, W., Schultz‐Darken, N. J., and Smith, T. E. (1997). Specific neuroendocrine mechanisms not involving generalized stress mediate social regulation of female reproduction in cooperatively breeding marmoset monkeys. Ann. N.Y. Acad. Sci. 807, 219–238. Baptista, L. F., and Gaunt, S. L. L. (1997). Bioacoustics as a tool in conservation studies. In ‘‘Behavioral Approaches to Conservation in the Wild’’ (R. Buchholz and J. R. Clemmons, Eds.), pp. 212–242. Cambridge University Press, New York. Beecher, M. D. (1989a). Signalling systems for individual recognition: An information theory approach. Anim. Behav. 38, 248–261. Beecher, M. D. (1989b). The evolution of parent‐offspring recognition in swallows. In ‘‘Contemporary Issues in Comparative Psychology’’ (D. A. Dewsbury, Ed.), pp. 360–380. Sinauer, Sunderland, MA. Bercovitch, F. B., Hauser, M. D., and Jones, J. H. (1995). The endocrine stress response and alarm vocalizations in rhesus macaques. Anim. Behav. 49, 1703–1706. Berger, J., Swenson, J. E., and Persson, I.‐L. (2001). Recolonizing carnivores and naive prey: Conservation lessons from Pleistocene extinctions. Science 291, 1036–1039. Blumstein, D. T. (1995a). Golden‐marmot alarm calls: I. The production of situationally specific vocalizations. Ethology 100, 113–125. Blumstein, D. T. (1995b). Golden‐marmot alarm calls: II. Asymmetrical production and perception of situationally specific vocalizations? Ethology 101, 25–32. Blumstein, D. T. (1999a). Alarm calling in three species of marmots. Behaviour 136, 731–757. Blumstein, D. T. (1999b). The evolution of functionally referential alarm communication: Multiple adaptations; multiple constraints. Evol. Comm. 3, 135–147. Blumstein, D. T. (2003). Social complexity but not the acoustic environment is responsible for the evolution of complex alarm communication/La complexite´ sociale est responsable de l’e´volution de la communication d’alarme complexe, non le milieu acoustique. In ‘‘Adaptative Strategies and Diversity in Marmots/Strate´gies Adaptatives et diversite´ chez les marmottes’’ (R. Ramousse, D. Allaine´, and M. Le Berre, Eds.), pp. 41–49. International Marmot Network, Lyon. Blumstein, D. T. (2007). The evolution of alarm communication in rodents: Structure, function, and the puzzle of apparently altruistic calling. In ‘‘Rodent Societies’’ (J. O. Wolff and P. W. Sherman, Eds.), pp. 317–327. University of Chicago Press, Chicago.

396

DANIEL T. BLUMSTEIN

Blumstein, D. T. (in press‐b). Tinbergen’s four questions In ‘‘The Encyclopedia of Applied Animal Behavior and Welfare’’ (D. mills, Ed.). CAB International, Wallingford, UK. Blumstein, D. T. (2007b). Tinbergen’s four questions. In ‘‘The Encyclopedia of Applied Animal Behaviour and Welfare’’ (D. Mills, Ed.). CAB International, Wallingford, UK. Blumstein, D. T., and Armitage, K. B. (1997a). Alarm calling in yellow‐bellied marmots: I. The meaning of situationally‐specific calls. Anim. Behav. 53, 143–171. Blumstein, D. T., and Armitage, K. B. (1997b). Does sociality drive the evolution of communicative complexity? A comparative test with ground‐dwelling sciurid alarm calls. Am. Nat. 150, 179–200. Blumstein, D. T., and Armitage, K. B. (1998). Life history consequences of social complexity: A comparative study of ground‐dwelling sciurids. Behav. Ecol. 9, 8–19. Blumstein, D. T., and Arnold, W. (1995). Situational‐specificity in alpine‐marmot alarm communication. Ethology 100, 1–13. Blumstein, D. T., and Daniel, J. C. (1997). Inter‐ and intraspecific variation in the acoustic habitats of three marmot species. Ethology 103, 325–338. Blumstein, D. T., and Daniel, J. C. (2004). Yellow‐bellied marmots discriminate between the alarm calls of individuals and are more responsive to the calls from juveniles. Anim. Behav. 68, 1257–1265. Blumstein, D. T., and Ferna´ndez‐Juricic, E. (2004). The emergence of conservation behavior. Cons. Biol. 18, 1175–1177. Blumstein, D. T., and Munos, O. (2005). Individual and age/sex class variation in marmot alarm calls. Anim. Behav. 69, 353–361. Blumstein, D. T., and Turner, A. C. (2005). Can the acoustic adaptation hypothesis predict the structure of Australian birdsong? Acta Ethol. 15, 35–44. Blumstein, D. T., Steinmetz, J., Armitage, K. B., and Daniel, J. C. (1997). Alarm calling in yellow‐bellied marmots: II. Kin selection or parental care? Anim. Behav. 53, 173–184. Blumstein, D. T., Daniel, J. C., and Bryant, A. A. (2001). Anti‐predator behavior of Vancouver Island marmots: Using congeners to evaluate abilities of a critically endangered mammal. Ethology 107, 1–14. Blumstein, D. T., Verenyre, L., and Daniel, J. C. (2004). Reliability and the adaptive utility of discrimination among alarm callers. Proc. R. Soc. Lond. B 271, 1851–1857. Blumstein, D. T., Patton, M. L., and Saltzman, W. (2006a). Faecal glucocorticoid metabolites and alarm calling in free‐living yellow‐bellied marmots. Biol. Lett. 2, 29–32. Blumstein, D. T., Holland, B.‐D., and Daniel, J. C. (2006b). Predator discrimination and ‘personality’ in captive Vancouver Island marmots (Marmota vancouverensis). Anim. Cons. 9, 274–282. Blumstein, D. T., Cooley, L., Winternitz, J., and Daniel, J. C. (2007). Do yellow‐bellied marmots respond to predator vocalizations? Behav. Ecol. Sociobiol. (in press). Blumstein, D. T. (in press‐a). Fourteen lessons from anti‐predator behavior. In ‘‘Natural Security: A Darwinian Approach to a Dangerous World’’ (R. Sagarin and T. Taylor, Eds.). University of California Press, Berkeley. Boinski, S., Gross, T. S., and Davis, J. K. (1999). Terrestrial predator alarm vocalizations are a valid monitor of stress in captive Brown Capuchins (Cebus apella). Zoo Biol. 18, 295–312. Bradbury, J. W., and Vehrencamp, S. L. (1998). ‘‘Principles of Animal Communication.’’ Sinnauer, Sunderland, MA. Brumm, H., and Slabbekoorn, H. (2005). Acoustic communication in noise. Adv. Study Behav. 35, 151–209. Bryant, A. A. (2005). Reproductive rates of wild and captive Vancouver Island marmots (Marmota vancouverensis). Can. J. Zool. 83, 664–673.

MARMOT ALARM COMMUNICATION

397

Bryant, A. A., and Page, R. E. (2005). Timing and causes of mortality in the endangered Vancouver Island marmot (Marmota vancouverensis). Can. J. Zool. 83, 674–682. Caro, T. (2005). ‘‘Antipredator Defenses in Birds and Mammals.’’ University of Chicago Press, Chicago. Cheney, D. L., and Seyfarth, R. M. (1988). Assessment of meaning and the detection of unreliable signals by vervet monkeys. Anim. Behav. 36, 477–486. Cheney, D. L., and Seyfarth, R. M. (1990). ‘‘How Monkeys See the World.’’ University of Chicago Press, Chicago. Clark, C. W., Marler, P., and Beeman, K. (1987). Quantitative analysis of animal vocal phonology: An application to swamp sparrow song. Ethology 76, 101–115. Clutton‐Brock, T. H., and Albon, S. D. (1979). The roaring of red deer and the evolution of honest advertisement. Behaviour 69, 145–169. Cresswell, W. (1993). Escape responses by redshanks, Tringa totanus, on attack by avian predators. Anim. Behav. 46, 609–611. Daniel, J. C., and Blumstein, D. T. (1998). A test of the acoustic adaptation hypothesis in four species of marmots. Anim. Behav. 56, 1517–1528. Davies, N. B., and Halliday, T. R. (1978). Deep croaks and fighting assessment in toads Bufo bufo. Nature 274, 683–685. Davidson, S. M., and Wilkinson, G. S. (2002). Geographic and individual variation in vocalizations by male Saccopteryx bilineata (Chiroptera: Emballonuridae). J. Mamm. 83, 526–535. Evans, C. S. (1997). Referential communication. Perspect. Ethol. 12, 99–143. Evans, C. S., and Marler, P. (1995). Language and animal communication: Parallels and contrasts. In ‘‘Comparative Approaches to Cognitive Science’’ (H. Roitblatt, Ed.), pp. 341–382. MIT Press, Cambridge, MA. Evans, C. S., Evans, L., and Marler, P. (1993). On the meaning of alarm calls: Functional reference in an avian vocal system. Anim. Behav. 46, 23–38. Fitch, W. T., and Hauser, M. D. (1995). Vocal production in nonhuman primates: Acoustics, physiology and functional constraints on ‘‘honest’’ advertisement. Am. J. Primatol. 37, 191–219. Fitch, W. T., and Hauser, M. D. (2002). Unpacking ‘‘honesty’’: Vertebrate vocal production and the evolution of acoustic signals. In ‘‘Acoustic Communication: Springer Handbook of Auditory Research’’ (A. Simmons, R. R. Fay, and A. N. Popper, Eds.), pp. 65–137. Springer, New York. Fitch, W. T., Neubauer, J., and Herzel, H. (2002). Calls out of chaos: The adaptive significance of nonlinear phenomena in mammalian vocal production. Anim. Behav. 63, 407–418. Floyd, C. H., Van Vuren, D. H., and May, B. (2005). Marmots on Great Basin mountaintops: Using genetics to test a biogeographic paradigm. Ecology 86, 2145–2153. Gerhardt, H. C., and Huber, F. (2002). ‘‘Acoustic Communication in Insects and Anurans.’’ University of Chicago Press, Chicago. Gould, S. J., and Vrba, E. S. (1982). Exaptation—a missing term in the science of form. Paleobiology 8, 4–15. Gouzoules, H., Gouzoules, S., and Miller, K. (1996). Skeptical responding in rhesus monkeys (Macaca mulatta). Int. J. Primatol. 17, 549–568. Greene, E., and Meagher, T. (1998). Red squirrels, Tamiasciurus hudsonicus, produce predator‐class specific alarm calls. Anim. Behav. 55, 511–518. Halliday, T. (1983). Information and communication. In ‘‘Animal Behaviour Volume 2: Communication’’ (T. R. Halliday and P. J. B. Slater, Eds.), pp. 43–81. Blackwell Scientific Publications Ltd., New York. Handford, P. (1988). Trill rate in the Rufous‐collared sparrow, Zonotrichia capensis, in northwestern Argentina. Can. J. Zool. 66, 2658–2670.

398

DANIEL T. BLUMSTEIN

Handford, P., and Lougheed, S. C. (1991). Variation in duration and frequency characteristics in the song of the rufous‐collard sparrow, Zonotrichia capensis, with respect to habitat, trill dialects and body size. Condor 93, 644–658. Hanson, M. T., and Coss, R. G. (2001). Age differences in the response of California ground squirrels (Spermophilus beecheyi) to conspecific alarm calls. Ethology 107, 259–275. Hare, J. F. (1998). Juvenile Richardson’s ground squirrels, Spermophilus richardsonii, discriminate among individual alarm callers. Anim. Behav. 55, 451–460. Hare, J. F., and Atkins, B. A. (2001). The squirrel that cried wolf: Reliability detection by juvenile Richardson’s ground squirrels (Spermophilus richardsonii). Behav. Ecol. Sociobiol. 51, 108–112. Hasson, O. (1991). Pursuit‐deterrent signals: Communication between prey and predators. Trends Ecol. Evol. 6, 325–329. Heard, D. C. (1977). The behaviour of Vancouver Island marmots, Marmota vancouverensis. M.Sc. thesis, University of British, Columbia. Hockett, C. F. (1960). The origin of speech. Sci. Am. 203(3), 89–96. Ho¨gstedt, G. (1983). Adaptation unto death: Function of fear screams. Am. Nat. 121, 562–570. Hoogland, J. L. (1995). ‘‘The Black‐tailed Prairie Dog: Social Life of a Burrowing Mammal.’’ University of Chicago Press, Chicago. Hugie, D. M. (2003). The waiting game: A ‘‘battle of waits’’ between predator and prey. Behav. Ecol. 14, 807–817. Hunter, M. L. Jr., and Krebs, J. R. (1979). Geographical variation in the song of the Great Tit (Parus major) in relation to ecological factors. J. Anim. Ecol. 48, 759–785. Hurd, C. R. (1996). Interspecific attraction to the mobbing calls of black‐capped chickadees (Parus atricapillus). Behav. Ecol. Sociobiol. 38, 287–292. Insley, S. J. (2000). Long‐term vocal recognition the northern fur seal. Nature 406, 404–405. Jouventin, P., Aubin, T., and Lengagne, T. (1999). Finding a parent in a king penguin colony: The acoustic system of individual recognition. Anim. Behav. 57, 1175–1183. Keller, L., and Reeve, H. K. (2002). Kin selection. In ‘‘Encyclopedia of Evolution’’ (M. Pagel, Ed.), pp. 595–600. Oxford University Press, Oxford. Kiriazis, J., and Slobodchikoff, C. N. (2006). Perceptual specificity in the alarm calls of Gunnison’s prairie dogs. Behav. Proc. 73, 29–35. Kirkpatrick, M., and Barton, N. H. (1997). Evolution of a species’ range. Am. Nat. 150, 1–23. Klump, G. M., and Shalter, M. D. (1984). Acoustic behaviour of birds and mammals in the predator context. I. Factors affecting the structure of alarm signals. II. The functional significance and evolution of alarm signals. Z. Tierpsychol. 66, 189–226. Kruckenhauser, L., Haring, E., Arnold, W., and Pinsker, W. (1999). Die Phylogenie der Gattung Marmota (Rodentia, Sciuridae): Ein Stammbaum auf der Basis von DNA‐ Sequenzdaten. Stapfia 63, 169–176. Lambrechts, M. M., and Dhondt, A. A. (1995). Individual voice discrimination in birds. Current Ornithol. 12, 115–139. Leger, D. W., Owings, D. H., and Gelfand, D. L. (1980). Single‐note vocalizations of California ground squirrels: Graded signals and situation‐specificity of predator and socially evoked calls. Z. Tierpsychol. 52, 227–246. Lenti Boero, D. (1992). Alarm calling in Alpine marmot (Marmota marmota L.): Evidence for semantic communication. Ethol. Ecol. and Evol. 4, 125–138. Leonard, M. L., Horn, A. G., Brown, C. R., and Fernandez, N. J. (1997). Parent‐offspring recognition in tree swallows, Tachycineta bicolor. Anim. Behav. 54, 1107–1116. Long, J. L. (2003). ‘‘Introduced Mammals of the World.’’ CSIRO Publishing, Collingwood, Victoria.

MARMOT ALARM COMMUNICATION

399

Low, T. (1999). ‘‘Feral Future.’’ Viking, Ringwood, Victoria. Macedonia, J. M., and Evans, C. S. (1993). Variation among mammalian alarm call systems and the problem of meaning in animal signals. Ethology 93, 177–197. Manser, M. (2001). The acoustic structure of suricates’ alarm calls varies with predator type and the level of response urgency. Proc. R. Soc. Lond. B 268, 2315–2324. Manser, M. B., Bell, M. B., and Fletcher, L. B. (2001). The information receivers extract from alarm calls in suricates. Proc. R. Soc. Lond. B 268, 2485–2491. Manser, M. B., Seyfarth, R. M., and Cheney, D. L. (2002). Suricate alarm calls signal predator class and urgency. Trends Cogn. Sci. 6, 55–57. Marler, P. (1955). Characteristics of some animal calls. Nature 176, 6–8. Marler, P. (1960). Bird song and mate selection. In ‘‘Animal Sounds and Communication’’ (W. F. Lanyon and W. Tavolga, Eds.), pp. 348–367. Am. Inst. Biol. Sci., Washington, DC. Marler, P., Evans, C. S., and Hauser, M. D. (1992). Animal signals: Motivational, referential, or both? In ‘‘Nonverbal Vocal Communication: Comparative and Developmental Approaches’’ (H. Papousek, U. Jurgens, and M. Papousek, Eds.), pp. 66–86. Cambridge University Press, Cambridge. Maynard Smith, J. (1965). The evolution of alarm calls. Am. Nat. 99, 59–63. McGregor, P. K., Peake, T. M., and Gilbert, G. (2000). Communication behavior and conservation. In ‘‘Behaviour and Conservation’’ (L. M. Gosling and W. J. Sutherland, Eds.), pp. 261–280. Cambridge University Press, Cambridge. Morton, E. S. (1975). Ecological sources of selection on avian sounds. Am. Nat. 109, 17–34. Naguib, M., and Wiley, R. H. (2001). Estimating the distance to a source of sound: Mechanisms and adaptations for long‐range communication. Anim. Behav. 62, 825–837. Neill, S. R. S. J., and Cullen, J. M. (1974). Experiments on whether schooling by their prey affects the hunting behaviour of cephalopods and fish predators. J. Zool. Lond. 172, 549–569. Nesterova, N. L. (1996). Age‐dependent alarm behaviour and response to alarm call in bobac marmots (Marmota bobac Mu¨ll.). In ‘‘Biodiversity in Marmots’’ (M. Le Berre, R. Ramousse, and L. Le Guelte, Eds.), pp. 181–186. International Network on Marmots, Moscow‐Lyon. Nikol’skii, A. A., and Suchanova, M. V. (1994). Individual variability of alarm call in steppe marmot (Marmota bobac Mu¨ll., 1776). In ‘‘Actual Problems of Marmots Investigation’’ (V. Y. Rumiantsev, Ed.), pp. 169–181. ABF Publ. House, Moscow. Nikol’skii, A. A., Nesterova, N. L., and Suchanova, M. V. (1994). Situational variations of spectral structure in Marmota bobac Mu¨ll. alarm signal. In ‘‘Actual Problems of Marmots Investigation’’ (V. Y. Rumiantsev, Ed.), pp. 127–148. ABF Publ. House, Moscow. Nikol’skii, A. A., Kotlyakov, V. M., and Blumstein, D. T. (1999). Glaciation as a factor of geographic variation in the long‐tailed marmot (bioacoustical analysis). Dokl. Biol. Sci. 368, 509–513. Nikolsky, A. A. (1981). Initial stage of ethologic divergence in isolated populations of Arctic ground squirrel (Citellus parryi Rich.). J. Gen. Biol. 42, 193–198. Norcross, J. L., and Newman, J. D. (1999). Effects of separation and novelty on distress vocalizations and cortisol in the common marmosette (Callithrix jacchus). Am. J. Primatol. 47, 209–222. Owings, D. H., and Hennessy, D. F. (1984). The importance of variation in sciurid visual and vocal communication. In ‘‘The Biology of Ground‐Dwelling Squirrels’’ (J. O. Murie and G. R. Michener, Eds.), pp. 169–200. University of Nebraska Press, Lincoln. Ramakrishnan, U., and Coss, R. G. (2000). Age differences in the responses to adult and juvenile alarm calls by bonnet macaques (Macaca radiata). Ethology 106, 131–144.

400

DANIEL T. BLUMSTEIN

Rendall, D., Rodman, P. S., and Emond, R. E. (1996). Vocal recognition of individuals and kin in free‐ranging rhesus monkeys. Anim. Behav. 51, 1007–1015. Robinson, S. R. (1981). Alarm communication in Belding’s ground squirrels. Z. Tierpsychol. 56, 150–168. Sagarin, R., and Taylor, T. (Eds.). (In press). ‘‘Natural Security: A Darwinian Approach to a Dangerous World.’’ University of California Press, Berkeley. Searcy, W. A., and Nowicki, S. (2005). ‘‘The Evolution of Animal Communication: Reliability and Deception in Signaling Systems.’’ Princeton University Press, Princeton. Semple, S. (2001). Individuality and male discrimination of female copulation calls in the yellow baboon. Anim. Behav. 61, 1023–1028. Shelley, E. L., and Blumstein, D. T. (2005). The evolution of vocal alarm communication in rodents. Behav. Ecol. 16, 169–177. Sherman, P. W. (1977). Nepotism and the evolution of alarm calls. Science 197, 1246–1253. Sherman, P. W. (1980a). The meaning of nepotism. Am. Nat. 116, 604–606. Sherman, P. W. (1980b). The limits of ground squirrel nepotism. In ‘‘Sociobiology: Beyond Nature/Nurture?’’ (G. W. Barlow and J. Silverberg, Eds.), pp. 505–544. Westview Press, Boulder. Sherman, P. W. (1985). Alarm calls of Belding’s ground squirrels to aerial predators: Nepotism or self‐preservation? Behav. Ecol. Sociobiol. 17, 313–323. Shriner, W. M. (1999). Antipredator responses to a previously neutral sound by free‐living adult golden‐mantled ground squirrels, Spermophilus lateralis (Sciuridae). Ethology 105, 747–757. Slabbekoorn, H. (2004). Habitat‐dependent ambient noise: Consistent spectral profiles in two African forest types. J. Acoust. Soc. Am. 116, 3727–3733. Sloan, J. L., and Hare, J. F. (2004). Monotony and the information content of Richardson’s ground squirrel (Spermophilus richardsonii) repeated calls: Tonic communication or signal certainty? Ethology 110, 147–156. Snowdon, C. T. (1993). Linguistic phenomena in the natural communication of animals. In ‘‘Language and Communication: Comparative Perspectives’’ (H. L. Roitblat, L. M. Hearman, and P. E. Nachtigall, Eds.), pp. 175–194. Erlbaum, Hillsdale, New Jersey. Steppan, S. J., Akhverdyan, M. R., Lyapunova, E. A., Fraser, D. G., Vorontsov, N. N., Hoffmann, R. S., and Braun, M. J. (2000). Molecular phylogeny of the marmots (Rodentia: Sciuridae): Tests of evolutionary and biogeographic hypotheses. Syst. Biol. 48, 715–734. Tinbergen, N. (1963). On aims and methods of ethology. Z. Tierpsychol. 20, 410–433. Trivers, R. L. (1971). The evolution of reciprocal altruism. Q. Rev. Biol. 46, 35–57. Wanker, R., and Fischer, J. (2001). Intra‐ and interindividual variation in the contact calls of spectacled parrotlets (Forpus conspicillatus). Behaviour 138, 709–726. Ward, S., Speakman, J. R., and Slater, P. J. B. (2003). The energy cost of song in the canary, Serinus canaria. Anim. Behav. 66, 893–902. Warkentin, K. J., Keeley, A. T. H., and Hare, J. F. (2001). Repetitive calls of juvenile Richardson’s ground squirrels (Spermophilus richardsonii) communicate response urgency. Can. J. Zool. 79, 569–573. Wasser, S. K. (1999). Stress and reproductive failure: An evolutionary approach with applications to premature labor. Am. J. Obstet. Gynecol. 180, S272–S274. Wiley, R. H. (1991). Associations of song properties with habitats for territorial oscine birds of eastern North America. Am. Nat. 138, 973–993. Wood, S. R., Sanderson, K. J., and Evans, C. S. (2000). Perception of terrestrial and aerial alarm calls by honeyeaters and falcons. Aust. J. Zool. 48, 127–134.

MARMOT ALARM COMMUNICATION

401

Zuberbu¨hler, K. (2000). Referential labelling in Diana monkeys. Anim. Behav. 59, 917–927. Zuberbu¨hler, K. (2001). Predator‐specific alarm calls in Cambell’s monkeys, Cercopithecus campbelli. Behav. Ecol. Sociobiol. 50, 414–422. Zuberbu¨hler, K., Jenny, D., and Bshary, R. (1999). The predator deterrence function of primate alarm calls. Ethology 105, 477–490.

This page intentionally left blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 37

The Evolution of Geographic Variation in Birdsong Jeffrey Podos* and Paige S. Warren{ *department of biology, graduate program in organismic and evolutionary biology, university of massachusetts amherst, massachusetts 01003 { department of natural resources conservation, graduate program in organismic and evolutionary biology, university of massachusetts amherst, massachusetts 01003

I. INTRODUCTION Evolutionary biologists have a long‐standing interest in how organisms vary geographically. This interest is motivated in part by recognition of a relationship between geographic variation and the process of speciation. Traits that vary over a given species’ range may serve as neutral, noncontributing indicators of the early stages of divergence, for example as different populations adapt to distinct environments and undergo corresponding genotypic and phenotypic divergence (Schluter, 2000). In other cases, traits that vary geographically might contribute to the speciation process. Primary examples of such traits are mating ornaments and displays which, in many animals, are centrally involved in mate recognition and mate selection (Andersson, 1994; Boughman, 2001; Foster, 1999; Panhuis et al., 2001; Wells and Henry, 1998; West‐Eberhard, 1983). Geographic divergence of mating signals can, under particular circumstances, facilitate assortative mating, reproductive isolation, and thus continued divergence among populations (Irwin et al., 2001; Lachlan and Servedio, 2004; Liou and Price, 1994; Payne et al., 2000; Ptacek, 2000; Slabbekoorn and Smith, 2002a). Divergence of mating signals and mate recognition systems is increasingly recognized as an important factor in speciation (Ryan, 1986). Studies of vocal signals in birds offer potentially useful opportunities for empirical tests of the relationships among geographic signal divergence, reproductive isolation, and speciation (reviewed by Edwards et al., 2005). Many bird vocalizations express significant geographic variation, a phenomenon that can be attributed largely to the tendency for many birds to 0065-3454/07 $35.00 DOI: 10.1016/S0065-3454(07)37009-5

403

Copyright 2007, Elsevier Inc. All rights reserved.

404

JEFFREY PODOS AND PAIGE S. WARREN

learn to vocalize through imitation. Imitative vocal learning enables the ready generation and rapid transmission of novel patterns of vocal structure (Slabbekoorn and Smith, 2002a; Slater, 1989). Indeed, it has been argued that consequent plasticity of the vocal phenotype has been instrumental in generating the high species diversity that characterizes some avian taxa (Fitzpatrick, 1988; Vermeij, 1988; cf. Baptista and Trail, 1992; see also Irwin and Price, 1999; ten Cate, 2000). A Science Citation Index query (Fig. 1A) attests to rapidly expanding activity, over the past 15 years, in the study of the causes and consequences of geographic variation in birdsong. Prospects for this field of inquiry, however, did not look very promising just a few decades ago. Mounting unease centered especially on questions about the evolution of ‘‘dialects,’’ a particular form of vocal geographic variation. This unease was well illustrated in Baker and Cunningham’s influential review of the phenomenon of dialects, a main goal of which was to articulate a ‘‘synthetic theory’’ on dialect origins and maintenance (Baker and Cunningham, 1985). Responses to this review, provided by a panel of

Number of literature citations

90

60

30

0 1965– 1970– 1975– 1980– 1985– 1990– 1995– 2000– 1969 1974 1979 1984 1989 1994 1999 2004 5-year span FIG. 1. Number of primary literature citations, binned over 5‐year spans, for two searches on Thompson’s ISI Web of Science. (A) Filled bars, search query as follows: topic ! [bird* AND song AND (geographic* OR dialect*)]. At the time of the search, Web of Science queries did not include abstract text for literature pre‐1990; thus only data post‐1990 are presented. A marked increase in citations is evident, pointing to increased activity in the field. (B) Open bars, number of publications that Marler and Tamura (1964), a classic paper on birdsong dialects. The post‐1990 trend mirrors that of the broader query (closed bars). Additionally, this search reveals a temporary reduction in activity in the field in the early 1990s, corresponding to the timing of the Baker and Cunningham (1985) exchange.

GEOGRAPHIC VARIATION IN BIRDSONG

405

peer experts, were forceful and included laments about difficulties in quantifying patterns of vocal geographic variation, in comparing data across taxa, in extrapolating laboratory data to field situations, and, most critically for present purposes, in identifying evolutionary factors that facilitate dialect formation and maintenance (Baptista, 1985; Brenowitz, 1985; Kroodsma, 1985a; Lemon, 1985; McGregor, 1985). With regard to this latter issue, Baker and Cunningham argued that dialects are necessarily maintained through adaptive processes, and thus, in order to understand their evolution we must examine their present function. A broad alternative hypothesis, that dialects emerge as epiphenomena of other evolutionary processes (Andrew, 1962), was summarily dismissed as being ‘‘pointless’’ (Baker and Cunningham, 1985, p. 86). The peer expert panel generally found Baker and Cunningham to be overly supportive of local adaptation hypotheses of dialect evolution and overly dismissive of alternative hypotheses (Baptista, 1985; McGregor, 1985; Waser, 1985). Given the wide range of stated unresolved issues, along with limited evidence at the time that could support any particular hypothesis of dialect evolution, it is perhaps not surprising that publication of the Baker and Cunningham exchange was followed by a period of dampened enthusiasm for the field (Fig. 1B). Our goal in this chapter is to assess the present state of the field, from both empirical and conceptual perspectives. Prior reviews on the topic of geographic variation in bird vocalizations have been numerous (Baker and Cunningham, 1985; Krebs and Kroodsma, 1980; Mundinger, 1982; Slabbekoorn and Smith, 2002a), and readers may question the value of yet another contribution. Yet new information continues to accrue. Moreover, recent general advances in the study of birdsong—a field that has remained active in realms ranging from mechanisms to ecology and evolution—have renewed the way that we can study geographic variation, in at least two ways. First, we have gained numerous insights into the range of possible functions of song learning, particularly in the arenas of social and sexual selection (Beecher and Brenowitz, 2005; Kroodsma and Byers, 1991; Nowicki et al., 2002). Our increasingly detailed understanding of the myriad functions of song learning suggests that patterns of vocal geographic variation may emerge as secondary by‐products of selection on other functions, rather than through direct selection for geographic patterns themselves (Slater, 1989). Second, advances in our understanding of mechanisms of vocal production suggest additional ‘‘non‐functional’’ scenarios by which geographic vocal variation may emerge. Whereas earlier models of song evolution focused on vocal imitation and cultural evolution, geographic variation in song may also emerge through evolution of the morphological and physiological underpinnings of song production.

406

JEFFREY PODOS AND PAIGE S. WARREN

We begin with a brief overview of the study of geographic variation in birdsong, focusing in particular on song dialects and hypothesis that have traditionally been put forward to explain their evolution (Section II). To evaluate the status of these hypotheses, we next conduct a comparative survey based on data gleaned from published literature (Section III). Our focus on dialects in these two sections is not motivated by an opinion that dialects are inherently more interesting than other patterns of geographic variation, or by an opinion that dialects are a unique phenomenon requiring special explanation. Rather it is simply because dialects have been the focus of the majority of published studies in this area. We then discuss how recent advances in the study of birdsong have enriched our ability to address questions about vocal geographic variation, arguing in particular that modern studies on mechanisms of song learning and production support ‘‘by‐ product’’ models of vocal geographic evolution (Section IV). As we argue below, the by‐product hypothesis of vocal geographic evolution can be regarded as a broader version of the ‘‘epiphenomenon’’ hypothesis of vocal dialect evolution. In Section V, we summarize the factors that may contribute together to the evolution of geographic variation in bird vocalizations.

II. EVOLUTION OF GEOGRAPHIC VARIATION IN SONG: LITERATURE OVERVIEW A. THE IMPORTANCE OF LEARNING MECHANISMS AND DISPERSAL PATTERNS Perhaps the most appropriate place to begin an overview of the literature on geographic variation in birdsong is with Marler and Tamura’s classic work on white‐crowned sparrows, Zonotrichia leucophrys nuttalli (Marler and Tamura, 1962, 1964). The phenomenon of geographic variation in birdsong had been observed and reported on previously, but only a handful of published studies (Borror, 1961) had made use of sound spectrograms, which provide an invaluable visual aid for assessing patterns of vocal geographic variation. Marler and Tamura described, for Z. leucophrys nuttalli of Northern California, the now‐classic ‘‘dialect’’ pattern of geographic variation, in which songs within particular populations ‘‘all share certain salient characteristics . . . which differ in certain consistent respects from the patterns found in neighboring populations’’ (Marler and Tamura, 1964, pp. 1483–1484). A Berkeley population of birds, for instance, was found to sing trills with a specific structure (fewer notes, of descending frequencies) distinct from trills sung by other nearby populations (Marler and Tamura, 1962). As suggested by Marler and Tamura and as generally corroborated by later studies (reviewed by Kroodsma et al., 1985), two main

GEOGRAPHIC VARIATION IN BIRDSONG

407

factors appear to facilitate dialect evolution in Z. leucophrys nuttalli. The first is that songs are culturally transmitted across generations, via vocal learning. A central role for vocal learning in song development in these birds was first demonstrated in a series of laboratory studies in which young males were exposed to various training regimes (Marler and Tamura, 1964). Training models presented over loudspeakers, during the sensitive phase of song acquisition, were shown to be copied precisely by these birds, even when training models had been recorded from nonnatal localities (Marler and Tamura, 1964). By contrast, in the absence of training models, birds were found to develop songs with atypical, degraded acoustic structure. It was thus argued that song learning enables the ready transmission of song patterns across generations, not just from fathers to sons but also potentially to other young males in a population. Song learning facilitates dialect formation by providing a mechanism for generating vocal novelties, which may emerge through copying ‘‘errors’’ (Marler and Tamura, 1964; see also Baptista, 1977; Lemon, 1975; Marler and Peters, 1987, 1988; Slater, 1986, 1989). The second main factor that appears to contribute to song dialect evolution in Z. leucophrys nuttalli is that of limited or biased dispersal. Toward this end, Marler and Tamura (1962) suggested two possibilities: that dialects may emerge if male birds remain on the grounds where they learned to sing, or that dialects may emerge if males do disperse but then settle preferentially in locations where they hear songs similar to their own. Either explanation would be consistent with the observation of vocal ‘‘neighborhoods.’’ Banding/recapture studies (Baker and Mewaldt, 1978; Petrinovich et al., 1981) confirmed that dispersal distances in this subspecies are indeed moderate, most commonly within a few hundred meters and rarely exceeding 1 km. By contrast, data in support of biased dispersal in this subspecies have been equivocal, and their interpretation controversial (Baker and Mewaldt, 1981; Baker et al., 1985; Hafner and Peterson, 1985; Petrinovich et al., 1981). The Z. leucophrys nuttalli system turns out to be comparatively, although by no means absolutely, straightforward; other systems examined to date feature their own complications and deviations from the nuttalli scenario. This point is well illustrated by variation found even within this one species. Two additional white‐crowned sparrow subspecies, Z. leucophrys pugetensis and Z. leucophrys oriantha, also exhibit song dialects. In the former subspecies, dialects occur over larger neighborhood areas (Baptista, 1977; Chilton and Lein, 1996a; DeWolfe and Baptista, 1995; Heinemann, 1981; Nelson and Soha, 2004), and in the latter subspecies, songs vary systematically among subalpine meadow populations (Chilton et al., 1995; Harbison et al., 1999; Orejuela and Morton, 1975). Birds of both subspecies turn out

408

JEFFREY PODOS AND PAIGE S. WARREN

to be less philopatric than their nuttalli counterparts, especially Z. leucophrys oriantha, which is fully migratory. Recent field and laboratory studies by Nelson and colleagues indicate that birds of both subspecies memorize multiple models at their natal grounds, and then, postdispersal, ‘‘select’’ among memorized models to best match song types present at their breeding grounds (‘‘overproduction’’ and ‘‘selective attrition’’: Nelson, 2000; Nelson et al., 1995, 1996a). An alternative model of song learning, postdispersal acquisition and memorization of song models, as posited by Heinemann (1981), appears not to apply in these birds. A fourth subspecies, Z. leucophrys gambelli, a long‐distance migrant that breeds in the sub‐Arctic, shows no evidence of dialects (Austen and Handford, 1991; DeWolfe et al., 1974; Nelson, 1998). Nelson (1999) argues that the postdispersal, prebreeding time frame in this subspecies is too brief to allow the kinds of extended interactions that are required for song matching via selective attrition. Studies of white‐crowned sparrows thus demonstrate that learning strategies, dispersal patterns, and resulting geographic song patterns can be highly variable even within a single species. This variation of course represents just a tiny sample of the diversity present in the songbirds as a group (Krebs and Kroodsma, 1980; Kroodsma, 1996; Slater, 1989). Indeed, a primary message of prior reviews of geographic song variation has been of caution in extrapolating results across species (Kroodsma, 1996). One widespread phenomenon that does not apply in white‐crowned sparrows, for instance, is postdispersal acquisition and learning of song models, as has been shown, for example, in the brood‐parasitic brown‐headed and bronzed cowbirds, Molothrus ater and M. aeneus (Rothstein and Fleischer, 1987; Warren, 2002). The white‐crowned sparrow system also does not address the potential influence of improvisation on geographic song patterns (Kroodsma and Verner, 1978; Kroodsma et al., 1997; Marler et al., 1972). Moreover, unlike many other species, adult white‐crowned sparrows tend to produce only a single song type. Patterns of geographic variation, and the ecological and learning‐based causes of these patterns, tend to be more challenging although sometimes still possible to document in species with song repertoires (Searcy et al., 2002; Slater et al., 1984; Williams and Slater, 1990). Explaining the formation of dialects might be relatively straightforward if patterns of geographic variation in song evolved solely as an incidental by‐product of particular learning mechanisms and patterns of dispersal (Fig. 2A). However, since the earliest studies of avian vocal geographic variation, authors have also spent considerable energy contemplating the potential fitness benefits of particular geographic song patterns. Implicit in this exercise is the hypothesis that selection for particular geographic patterns

GEOGRAPHIC VARIATION IN BIRDSONG

409

FIG. 2. Traditional framework to explain the evolution of vocal geographic variation in songbirds. (A) Solid lines: Geographic variation in song emerges, in a proximate sense, as a by‐ product of specific learning mechanisms and dispersal patterns. The evolution of learning mechanisms that foster early, accurate imitation, combined with the evolution of limited dispersal distances, for example, will result in the evolution of sharp dialects. (B) Dashed lines: Selection for particular patterns of geographic variation may alter the evolution of learning mechanisms and patterns of dispersal, in a feedback loop. For instance, selection favoring strong assortative mating (local adaptation hypothesis) may conceivably favor the evolution of early song imitation and limited dispersal, and thus the evolution of sharp dialects.

alters, in a feedback loop, the evolution of the learning mechanisms and dispersal patterns that shape vocal geographic divergence (Fig. 2B; Jenkins, 1985). To explore this scenario further we turn again to the phenomenon of dialects and hypotheses that have been put forward to explain their evolution. We provide brief commentary on definitions of song dialects, and then examine three broad categories of hypotheses that have traditionally been forwarded to explain their evolution.

B. DEFINITION OF SONG DIALECTS As noted by Slater (1989, p. 33), ‘‘geographic variation [in song] is seldom the simple matter that the word ‘dialect’ might imply.’’ Indeed, ‘‘dialects’’ have been described across a range of geographic scales (e.g., compare Leader et al., 2000 with Warren, 2002; see also Mundinger, 1982) and for a variety of vocal parameters. Schematically, we can represent geographic structure in vocal parameters as taking a range of spatial patterns, including random variation (Fig. 3A), gradual and shallow clines (Fig. 3B), or steep clines with stepped variation (Fig. 3C). This latter form is consistent with classic definitions of dialects (Marler and Tamura, 1962; Mundinger, 1982). Numerous researchers have noted that strict dialects (Fig. 3C) may be a comparatively rare phenomenon (Slater, 1986, 1989). The functional hypotheses that we review in Section II.C have drawn on the presence of sharp boundaries between dialect neighborhoods as justification for regarding dialects as being actively maintained by selection.

410

JEFFREY PODOS AND PAIGE S. WARREN

Vocal parameter (s)

Random

Clinal

Dialect

Geographic locality FIG. 3. Schematic representation of three patterns of geographic variation in song. Geographic variation can be manifested at varying scales. Dialect variation features sharp transitions in vocal parameters between localities, and consistency in vocal parameters within localities.

C. HYPOTHESES TO EXPLAIN THE EVOLUTION OF SONG DIALECTS 1. Local Adaptation Hypothesis The local adaptation hypothesis, alluded to by Marler and Tamura (1962) and then formalized by Nottebohm (1969), posits that females gain fitness advantages when they are able to mate with males from their natal regions, in preference to males from other regions. According to this hypothesis, birds that select mates from their natal regions will gain fitness advantages because their offspring will more likely express adaptations to local ecological conditions. Because song is a key mating signal in many species of birds, song structure is thus posited to diverge by locality, under selection for accurately marking birds’ natal localities. Baker and colleagues argued more specifically that dialects serve as markers for ‘‘coadapted gene complexes,’’ and that dialect boundaries represent secondary contact zones between partially

GEOGRAPHIC VARIATION IN BIRDSONG

411

isolated populations (Baker, 1982; Baker and Thompson, 1985; Baker et al., 1982). The local adaptation hypothesis makes four predictions, each of which has been subject to considerable scrutiny. The first prediction is that birds should learn their vocalizations early, before dispersing from their natal regions (MacDougall‐Shackleton and MacDougall‐Shackleton, 2001; Payne, 1981; Rothstein and Fleischer, 1987). Decades of study have now shown wide diversity in the timing of song acquisition, and so this prediction may hold for some species but clearly does not hold for others. Brown‐headed cowbirds, for instance, learn songs postdispersal, and thus dialects in this species can best be attributed to other hypotheses (Rothstein and Fleischer, 1987). The timing of learning has traditionally been studied in laboratory conditions, and a general point of discussion concerns the applicability of laboratory results to field contexts. In particular, it has been noted that species that only learn early in the laboratory, from taped tutor songs, may still retain the ability to learn songs later in life, if trained by live tutors (Baptista and Petrinovich, 1984). The relative impact of social influences on song learning continues as an area of active study (Beecher and Brenowitz, 2005; Johannessen et al., 2006). Second, the local adaptation hypothesis predicts that dispersing birds will tend to settle, more than would be expected by chance, in localities where birds sing their natal dialects, as opposed to localities in which birds sing ‘‘foreign’’ dialects. Tests of this prediction require mark/recapture studies, and results so far have been inconclusive (Baker and Mewaldt, 1978; Baker and Mewaldt, 1981; Baker et al., 1985; Hafner and Peterson, 1985; Petrinovich et al., 1981). A call by Kroodsma (1985a) for additional empirical work in this area still rings true. Third, the local adaptation hypothesis predicts that dialect groups should come to be genetically differentiated, as a result of recent histories of assortative mating. The main challenge in tests of this prediction has been to identify genetic parameters in which dialect neighborhoods differ. Again the evidence has been inconclusive and open to interpretation (Baker et al., 1982; Lougheed and Handford, 1992; Payne and Westneat, 1988; Zink and Barrowclough, 1984). In the most comprehensive study on this topic to date, MacDougall‐Shackleton and MacDougall‐Shackleton (2001) analyzed variation in microsatellite allele frequencies among eight dialect regions of Z. leucophyrs oriantha, and concluded that ‘‘dialect borders are associated with some reduction in gene flow, but they appear to be very low walls rather than barriers in any strict sense’’ (MacDougall‐Shackleton and MacDougall‐Shackleton, 2001, p. 2574). The fourth prediction of the local adaptation hypothesis is that females should evolve mating preferences for males from their natal dialects. A number of assays have been developed to measure the impact of vocal

412

JEFFREY PODOS AND PAIGE S. WARREN

variation on female preferences (Searcy, 1992). In tests of the influence of dialect variation on female preferences, researchers have generally turned to the copulation solicitation display assay. In this assay, females are captured, acclimated to laboratory conditions, and then treated with exogenous estradiol, normally via silastic implants, to increase sexual receptivity. Songs of different dialects are then presented over loudspeakers, and the strength and vigor of resulting solicitation displays documented. Females of a number of species have been found to respond more often and more vigorously to local songs than to foreign songs (Baker et al., 1981, 1987a; Searcy et al., 1997, 2002). Differential preferences for local songs presumably emerge because of greater familiarity with local dialects (Baker et al., 1981), or through learned associations between local songs and social feedback experienced early in life (Riebel, 2003). Learned preferences for males that sing local songs could foster, through coevolution of dialect and preference, the divergence of dialect forms (Riebel, 2003). It is useful to note that of these four predictions, only the second is unique to the local adaptation hypothesis. The others predictions are also consistent with other hypotheses of dialect evolution (see below). Early studies of vocal geographic variation in birds focused on the local adaptation hypothesis (Marler and Tamura, 1962; Nottebohm, 1969), which is perhaps not surprising given that birdsong studies of that era were generally geared toward questions about species recognition. Many birds produce species‐specific vocalizations, and ornithologists have long hypothesized that vocalizations thus aid conspecific recognition (Marler, 1957, 1960). With the advent of portable tape recorders and loudspeakers, this hypothesis was tested and supported in a host of playback studies, which showed time and again that birds respond more strongly to playback of conspecific than heterospecific song (Falls, 1963; Gill and Murray, 1972; Martin and Martin, 2001; Milligan, 1966; reviewed by Becker, 1982). Similar patterns of elevated responses to playback of conspecific versus heterospecific song have also been demonstrated in females from numerous species (Searcy, 1992). From an evolutionary perspective, interspecific divergence in vocal structure is thought to be driven by two related factors, namely selection for avoiding interspecific acoustic competition (Nelson and Marler, 1990) and selection against cross‐species mating and hybrid production (Butlin, 1995; de Kort and ten Cate, 2001; Haavie et al., 2004; Seddon, 2005; see Sætre et al., 1997, for a parallel argument for the evolution of plumage divergence). According to these hypotheses, individuals within a population that best transmit the correct species identity, through production of the most species‐typical songs, experience lower probabilities of fitness‐reducing interspecific hybridization (Noor, 1999). It was a small logical step from the species recognition hypothesis to the local adaptation

GEOGRAPHIC VARIATION IN BIRDSONG

413

hypothesis. Both hypotheses focus on the same evolutionary factors, notably selection against mating ‘‘errors.’’ The only substantive difference between the hypotheses is that mating errors occur at inter‐ versus intraspecific scales. As with tests of conspecific recognition, tests of local recognition abilities have relied heavily on playback designs (Baker et al., 1987a; Harris and Lemon, 1974; Lemon, 1967; McGregor, 1983; Milligan and Verner, 1971; Petrinovich and Patterson, 1981; Ratcliffe and Grant, 1985; Searcy et al., 1997, 2002). 2. Social Adaptation Hypotheses A second class of hypotheses for dialect evolution also focuses on the role of song in recognition, but with regard to social groups rather than locality. These ‘‘social adaptation hypotheses’’ suggest that males gain fitness advantages by singing songs similar to those of other males in their region, whereas males that sing nonlocal songs are subject to social penalties. One version of this hypothesis, the deceptive mimicry hypothesis (Payne, 1981), posits that subordinate males that successfully mimic the vocalizations of dominant males are able to improve their access to mates, and also to reduce probabilities of aggressive interactions with dominant males. Under this scenario, dialects should be temporally unstable and depend on which individuals are dominant at a given time (Payne, 1981; Rothstein and Fleischer, 1987). A related hypothesis is that of ‘‘honest convergence,’’ which posits that dialects serve as honest signals of long‐ term residence (Rothstein and Fleischer, 1987). Similarly, the colony password hypothesis (Feekes, 1977) proposes that dialects can serve as markers of group membership in colonial species, and thus facilitate the identification of intruders into a colony. Social adaptation hypotheses predict that individuals will learn new vocalizations on dispersal to a new dialect area, in order to match (mimic) vocalizations at the new locality (Payne, 1981). If individuals are constrained to acquire and crystallize new vocalizations early, prior to juvenile dispersal, then the social dynamics of adults cannot play a role in the maintenance of dialects. Postdispersal learning has been implicated in a number of species, including the cowbird systems mentioned earlier (Rothstein and Fleischer, 1987; Warren, 2002). Not surprisingly, male–male interactions have been a primary focus in tests of social adaptation hypotheses. The deceptive mimicry hypothesis predicts that accurate mimics should incite less aggressive responses from dominant males, and moreover that males should be more aggressive toward other males from neighboring dialects as opposed to from their home dialects (Rothstein and Fleischer, 1987). The colony password hypothesis similarly predicts that members of a colony should be more

414

JEFFREY PODOS AND PAIGE S. WARREN

aggressive toward males from foreign rather than local dialects, because foreign dialects would indicate potential intruders from other colonies (Feekes, 1977). Dialect size is also an important component of social adaptation hypotheses. The colony password and deceptive mimicry hypotheses predict that dialect areas should be small, corresponding to ‘‘socially cohesive units’’ (Payne, 1981). The honest convergence hypothesis does not require that dialects be small, but does, however, require them to be not substantially larger than average adult dispersal distance (Warren, 2002). If a dialect area is too large, dialect identity can no longer serve as an honest signal of local residence because newly arrived males would produce vocal signals indistinguishable from those of local residents (Warren, 2002). In general, it is difficult to conceive of a scenario in which social adaptation maintains boundaries between larger regional dialects; if dialects were too large, most individuals within a dialect would be unlikely to ever encounter other dialects. 3. Epiphenomenon Hypothesis The epiphenomenon hypothesis, first articulated (although only briefly) by Andrew (1962), posits that dialects emerge as incidental by‐products of particular patterns of learning and dispersal, both of which evolve under selection pressures unrelated to dialect formation. In other words, selection on dialect patterns per se need not produce dialects or maintain the discrete boundaries that characterize dialects. This has been considered by some to be the null hypothesis for dialect evolution, and it indeed makes fewer assumptions than do the other classes of hypotheses (Fig. 2A vs B; Lemon, 1975; Wiens, 1982). As we understand it, the epiphenomenon hypothesis differs from functional hypotheses of dialect evolution in that it does not require a history of continuous contact between birds from different dialect groups or a history of negative selection against birds that sing foreign songs. Continuous contact is necessary in the functional hypotheses, for example as intruders that sing foreign songs are rejected, in order to generate the negative selection pressures implied therein. Most simply, nonfunctional divergence among dialect groups can result from differential trajectories of selection among isolated populations. To illustrate, numerous lines of evidence now suggest that songs undergo selection for optimal transmission through the acoustic environment (Nottebohm, 1969, 1985; Wiley and Richards, 1978; reviewed by Slabbekoorn, 2004). Such selection pressures may drive intraspecific vocal divergence, and thus the emergence of dialect patterns, when different populations come to occupy distinct habitats (Doutrelant et al., 1999; Handford and Lougheed, 1991; Hunter and Krebs, 1979; Patten et al., 2004; Slabbekoorn and

GEOGRAPHIC VARIATION IN BIRDSONG

415

Smith, 2002b). Limited dispersal, and thus limited contact among localities, may facilitate the nonfunctional evolution of dialects, as vocal novelties persevere only in limited regions (Lemon, 1975; Slater, 1989). (A note on terminology: vocal divergence through acoustic adaptation to different habitats itself clearly has functional bases. However, any resulting geographic patterns would be considered nonfunctional, given that the locus of acoustic adaptation was site specific and thus did not favor geographic diversification per se). Slater (1985, 1986, 1989) offered a compelling defense of the epiphenomenon hypothesis for dialect evolution. As noted by Slater (1986, p. 96), ‘‘It is possible that dialects have no functional significance, but that the differences they represent are simply spurious byproducts of vocal learning. If song is learnt and dispersal after learning is restricted, some sort of variation in both time and space seems inevitable. . .’’ Slater defened the epiphenomenon hypothesis by raising three points. First, song learning may evolve for a range of functions besides dialect recognition. Such functions may include helping birds to better match the local acoustic environment or to enable accurate transmission of complex songs (Slater, 1986). Second, selection on bird vocalizations generally acts at the level of individuals, whereas dialects are population‐level phenomena. Third, variations in dispersal patterns and the accuracy of song learning, as studied in computer simulations, illustrate that the epiphenomenon mechanism can indeed generate discrete dialect patterns, particularly in species with small repertoire sizes (Goodfellow and Slater, 1986; Lachlan et al., 2004; Slater, 1989). We regard the epiphenomenon hypothesis as a type of ‘‘by‐product’’ hypothesis of vocal evolution. By‐product mechanisms of interpopulation divergence refer generally to situations in which selection on one trait or suite of traits drives incidental changes in, or limits on the expression of, other traits or suites of traits. Correlated evolution of multiple traits may arise through shared genetic or phenotypic mechanisms, such as when multiple functions make use of the same anatomical or physiological components (Patek et al., 2006; Podos and Hendry, 2006). Correlated evolution of multiple traits may also arise as result of life history constraints, as metabolic or energetic resources allocated to certain traits limit the development or expression of other traits (Roff, 1992). By‐product mechanisms have been of particular interest in studies of ‘‘ecological speciation,’’ which posits that incidentally modified trait(s) can impact patterns of mating and reproductive isolation (Boughman, 2001; Dobzhansky, 1951; Mayr, 1942; Orr and Smith, 1998; Podos, 2001; Podos and Hendry, 2006; Rice and Hostert, 1993; Ruegg et al., 2006; Schluter, 2000, 2001). Rundle et al. (2000), for instance, provided evidence that adaptive divergence of body size in benthic versus limnetic stickleback fishes has facilitated assortative

416

JEFFREY PODOS AND PAIGE S. WARREN

mating by morph, presumably because body size is used as a cue in mate assessment and selection. According to by‐product models of trait divergence, the emergence of reproductive isolation among populations is not necessarily a product of selection for that isolation, but potentially a secondary consequence of ecological adaptation in other traits (Podos and Hendry, 2006).

III. ASSESSING HYPOTHESES OF DIALECT EVOLUTION The potential relevance of the three hypothesis classes outlined above has now been addressed in a moderate number of species. Our goal in this section is to survey the primary literature, in order to identify potentially common conditions that underlie dialect formation and maintenance. Of particular interest is the timing of song acquisition, given that a key prediction distinguishing the two functional hypotheses is whether acquisition occurs pre‐ or postdispersal. Other potential correlates of dialect patterns examined here include social systems, degree of seasonal mobility, and vocal repertoire size. Examination of these parameters cannot provide definitive support for or against the three classes of dialect hypotheses, especially given difficulties noted by many other researchers in comparing dialect data across species. Rather, our intent is to gain a sense of trends or conditions that may favor dialect evolution as a general phenomenon.

A. OUR APPROACH We surveyed the literature for species reported to exhibit dialects (Table I). Our survey included not just birds but also primates, cetaceans, anurans, and insects. To meet our definition of dialects, we required evidence of sharp geographic boundaries between signal forms, of discrete differences among signal forms, and of uniformity of signal forms within given localities (Fig. 3C). Thus, a species exhibiting geographic variation in a continuous character such as pulse rate or dominant frequency was not included unless this variation was shown to be partitioned discretely among sites (e.g., as shown in Leader et al., 2000). We excluded studies for which data were ambiguous about one or more of the above criteria (Naguib et al., 2001; Podos, 2007). Moreover, a species exhibiting many discretely different signal forms was not included unless these signal types were geographically partitioned such that narrow boundaries between types could be clearly mapped. We thus excluded from our analysis ‘‘island’’ dialects (Ratcliffe, 1981), in which dialect localities are geographically isolated

TABLE I SPECIES REPORTED TO EXHIBIT VOCAL DIALECTS, INCLUDED IN OUR ANALYSES #

Species

417

Common name

Family

Signal

1 Amazona auropalliata

Yellow‐naped Amazon

Psittacidae

group calls

2 Phaethornis longuemareus 3 Miliaria calandra

Little Hermit

Trochilidae

song

Corn bunting

Fringillidae

song

4 Emberiza citrinella

Yellowhammer

Fringillidae

song

5 Emberiza hortulana

Ortolan

Fringillidae

song

6 Zonotrichia leucrophrys Nuttall’s white‐crowned nuttalli Sparrow

Fringillidae

song

7 Zonotrichia leucophrys Montane White‐crowned oriantha Sparrow

Fringillidae

song

References Wright, 1996; Wright and Wilkinson, 2001; Wright et al., 2005 Wiley, 1971 Holland et al., 1996; McGregor, 1980, 1983; McGregor and Krebs, 1984; McGregor and Thompson, 1988; McGregor et al., 1988 Baker et al., 1987a; Møller, 1982; Hansen, 1985, 1999; Glaubrecht, 1989, 1991; Rutkowska‐Guz and Osiejuk, 2004 Conrads, 1976; Conrads and Conrads, 1971; Thielcke, 1969 Baker, 1974, 1975, 1983; Baker and Mewaldt, 1978; Baker and Thompson, 1985; Baker et al., 1981, 1982, 1984a,b, 1987b; Baptista, 1975; Baptista et al., 1997; Cunningham et al., 1987; Marler and Tamura, 1962, 1964; Milligan and Verner, 1971; Petrinovich and Patterson, 1981; Trainer, 1983; Zink and Barrowclough, 1984 Baptista, 1977; Chilton and Lein, 1996a,b; Nelson, 2000; Nelson et al., 1996b, 2004; Soha et al., 2004 (Continued)

TABLE I (Continued) #

Species

Common name

Family

Signal

418

8 Zonotrichia leucophrys Puget Sound White‐crowned pugetensis Sparrow

Fringillidae

song

9 Zonotrichia capensis

Rufous‐crowned Sparrow

Fringillidae

song

Vesper Sparrow Sage Sparrow Spotted Towhees Indigo Bunting

Fringillidae Fringillidae Fringillidae Fringillidae

song song song song

Fringillidae

song

Fringillidae

song

16 Molothrus ater

Yellow‐rumped Cacique (Surinam) Yellow‐rumped Cacique (Panama) Brown‐headed Cowbird

Fringillidae

song

17 Molothrus aeneus

Bronzed Cowbird (Winter)

Fringillidae

song

10 11 12 13

Pooecetes gramineus Amphispiza belli Pipilo maculatus Passerina cyanea

14 Cacicus cela 15 Cacicus cela

References Baptista and King, 1980; Baptista and Morton, 1982, 1988; Harbison et al., 1999; MacDougall‐Shackleton and MacDougall‐Shackleton, 2001; MacDougall‐Shackleton et al., 2002; Nelson et al., 1996a King, 1972; Lougheed and Handford, 1992; Nottebohm, 1969, 1975; Tubaro and Segura, 1994; Zink et al., 1991 Kroodsma, 1972 Rich, 1981 Borror, 1975 Payne, 1981, 1982, 1983; Payne and Westneat, 1988 Feekes, 1977 Trainer, 1988, 1989; Trainer and Parsons, 2002 Alderson et al., 1999; Dolbeer, 1982; Fleischer and Rothstein, 1988; O’Loghlen, 1995; O’Loghlen and Rothstein, 1993, 1995; Rothstein and Fleischer, 1987; Rothstein et al., 1999; Teather and Robertson, 1986; Yokel, 1986 Dolbeer, 1982; Rothstein, 1980; Warren, 2000, 2003

419

18 Molothrus aeneus

Bronzed Cowbird (Breeding)

Fringillidae

song

19 Dolichonyx oryzivorus

Bobolink

Fringillidae

song

20 Fringilla coelebs

Chaffinch

Fringillidae

21 22 23 24 25 26 27 28

Greenfinch House Finch Smith’s longspur Apapane Bewicks Wren European Wren (UK) European Wren (France) Short‐toed Treecreeper

Fringillidae Fringillidae Fringillidae Fringillidae Certhiidae Certhiidae Certhiidae Certhiidae

call (‘‘rain call’’) song song song song song song song song

29 Turdus iliacus

Redwing

Passeridae

song

30 Vidua chalybeata

Village Indigobird

Passeridae

song

31 Vidua purpurascens 32 Poecile atricapillus

Dusky Indigobird Black‐capped Chickadee

Passeridae Paridae

33 Poecile atricapillus

Black‐capped Chickadee

Paridae

34 Poecile carolinensis

Carolina Chickadee

Paridae

song call (‘‘gargle’’) song (‘‘fee bee’’) song (‘‘fee bee’’)

Chloris chloris Carpodacus mexicanus Calcarius pictus Himatione sanguinea Thryomanes bewickii Troglodytes troglodytes Troglodytes troglodytes Certhia brachydactyla

Carter, 1984; Clotfelter, 1995; Dolbeer, 1982; Rothstein, 1980; Warren, 2000, 2002, 2003 Avery and Oring, 1977; Trainer and Peltz, 1995 Baptista, 1990; Sick, 1939; Sorjonen, 2001; Thielcke, 1969, 1988a,b, 1989 Gu¨ttinger, 1974, 1976 Mundinger, 1975, 1982 Briskie, 1999 Ward, 1964 Kroodsma, 1974, 1985b Catchpole and Rowell, 1993 Kreutzer, 1974 Seitz et al., 1994; Thielcke, 1969, 1984, 1986, 1987; Thielcke and Wuestenberg, 1985 Bjerke, 1974, 1980, 1982, 1984; Bjerke and Bjerke, 1981; Espmark, 1981, 1982; Fonstad et al., 1984; Mork, 1974 Payne, 1973, 1981, 1987; Payne and Payne, 1977 Payne, 1973, 1981, 1987 Ficken et al., 1978, 1985 Kroodsma et al., 1999 Ward, 1966 (Continued)

TABLE I (Continued) #

Species

Common name

Family

420

35 Pachycephala olivacea 36 Cyanocitta cristata

Olive Whistler Blue Jay

Corvidae Corvidae

37 Creadion carunculatus 38 Menura novaehollandiae 39 Nectarinia osea 40 Orcinus orca 41 Pyseter macrocephalus 42 Saguinus labiatus labiatus

Saddleback Superb Lyrebird Orange‐tufted Sunbirds Orca, Killer Whale Sperm Whale Red‐chested Moustached Tamarin

Signal

References White, 1985, 1986, 1987 Kramer and Thompson, 1979

Callaeatidae Menuridae

song group calls (‘‘bell’’ call) song song

Nectariniidae Cetacea Certacea Callithricidae

song group calls group calls long calls

Leader et al., 2000, 2002, 2005 Ford, 1991 Weilgart and Whitehead, 1997 Masataka, 1988

Jenkins, 1978 Powys, 1995; Robinson and Curtis, 1996

GEOGRAPHIC VARIATION IN BIRDSONG

421

from each other. This is not to say that vocal variation in species distributed across islands cannot be considered dialectal; our criteria here were set for purposes of analysis only.

B. LITERATURE SURVEY Our search for published studies of dialects was conducted using both digital databases (ISI Web of Science) and searches of citation sections of published works on dialects. We identified about 200 studies of 52 species in which the authors suggested that the vocal system could be described as dialectal. Of these, 42 cases met our present definition of dialects (Fig. 3C). We include data from 141 studies of these 42 cases in our analysis (Table I). We compiled data for them (Table II), as follows. 1. Dialect Characteristics We coded three descriptive characteristics of dialects: spatial extent (dialect scale), temporal stability of dialect boundaries, and presence or absence of bilingualism at boundaries between dialects. Dialect scale was coded according to four categories: microgeographic, small, medium, and large, coded, respectively, as 0, 1, 2, or 3. Microgeographic dialects feature 10 individuals or less in each dialect area, and span less than 2 km in any given direction. Small dialects consisted of dialects areas that span 2–10 km. Medium dialects were considered to span 10–100 km, and large dialects greater than 100 km. In some cases, dialects were only described in terms of numbers of individuals. Small, medium, and large dialects were then classified as containing less than 100 individuals, less than 1000 individuals, and more than 1000 individuals, respectively. Most dialect areas do not greatly exceed 1000 km. The temporal stability of dialect boundaries was described explicitly by many authors. Some taxa retain similar boundaries and acoustic characteristics of song types over long periods of time (Hansen, 1999; Thielcke, 1987) while others change rapidly, even within a season (Trainer, 1989). We classified dialect stability as short‐term, moderate‐ term, and long‐term, coded as 1, 2, or 3, respectively. Short‐term dialects maintained 2–6 years of stability. In the case of most passerine species, this probably corresponds to the life span of individuals of the species (Weatherhead and Forbes, 1994). Moderate‐term dialects were defined as being stable for 6–20 years, and long‐term dialects as being stable for greater than 20 years. It is probable that some taxa coded with moderate‐term dialects actually are stable over the long‐term, and that long‐term stability has not yet been demonstrated.

TABLE II VOCAL AND ECOLOGICAL PARAMETERS OF DIALECT SPECIESa

422

#

Species

Spatial scale

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Amazona auropalliata Phaethornis longuemareus Miliaria calandra Emberiza citrinella Emberiza hortulana Zonotrichia leucrophrys nuttalli Zonotrichia leucophrys oriantha Zonotrichia leucophrys pugetensis Zonotrichia capensis Pooecetes gramineus Amphispiza belli Pipilo maculatus Passerina cyanea Cacicus cela Cacicus cela Molothrus ater Molothrus aeneus Molothrus aeneus Dolichonyx oryzivorus Fringilla coelebs Chloris chloris

2 0 1 3 2 1 3 3 2 1 1 1 0 2 2 2 2 3 1 2 2

Temporal stability 2 2 3 1 3 3 2 3

Bilingual

Territoriality

Mobility

Repertoire size max

1

0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 0 1 0 1 1

0 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0

1 1 3 5 2 1 2 2 1 43 1 2 1 4 8 3 1 1 43 1 35

1 1 1 0 1 1 1 0 1

0 0 2 1 1 1 3 0

0 1 1 1 1 0

Timing of learning post post

pre pre pre pre

post post post post

post post

423

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 a

Carpodacus mexicanus Calcarius pictus Himatione sanguinea Thryomanes bewickii Troglodytes troglodytes Troglodytes troglodytes Certhia brachydactyla Turdus iliacus Vidua chalybeata Vidua purpurascens Poecile atricapillus Poecile atricapillus Poecile carolinensis Pachycephala olivacea Cyanocitta cristata Creadion carunculatus Menura novaehollandiae Nectarinia osea Orcinus orca Pyseter macrocephalus Saguinus labiatus labiatus

1 0 1 1 0 1 3 1 1 1 1 1 2 2 2 0 1 0 2 1 2

1 2 1

1 3 2 0 0

1

1 0 1 1 1 1 0 0

1 2 1 3 1

1 1 0 0

0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1

See text for an explanation of coding; blank cells indicate a lack of sufficient available data.

1 1 1 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 1 0

4 1

pre post

16 6 1 5 2 12 12 1 9

post

10 2 1 3 1 17 30 1

pre pre post post post post

post

pre pre post

424

JEFFREY PODOS AND PAIGE S. WARREN

Bilingualism refers to the propensity of individuals to acquire more than one dialect, particularly in areas near a dialect boundary. Following our definition of dialects, the proportion of a population that is bilingual should be low in most cases. But bilingualism was reported regularly among the species we reviewed. We classified bilingualism as either absent or present (coded as 0 or 1, Table II) based on the authors’ reporting. We classified bilingualism as absent only when an author either stated it to be so or gave sufficiently exhaustive detail on vocal behavior of the focal populations to allow us to make this inference. 2. Ecological Correlates We compiled data for two ecological and life history characteristics commonly cited in the literature: social system and degree of seasonal mobility. First, we assessed, based on the literature, whether taxa with dialects were predominantly territorial or, alternatively, social (e.g., colonial) on their breeding grounds. With regard to seasonal mobility, taxa were classified as either sedentary or migratory, again on the basis of published descriptions. 3. Vocal Correlates Data on one vocal parameter, repertoire size, was compiled for each species or subspecies in our analysis. Rather than analyzing the ranges of repertoire sizes within each species, we included in our analyses only maximum repertoire size. We focus on maximum repertoire size because this provides a good indication of the potential for individuals to acquire songs of more than one dialect. Coding each cell with a single number aided our multivariate analyses below. 4. Timing of Song Acquisition While the song acquisition phase of song learning has not been characterized in detail for many species, it can sometimes be inferred from observational studies of song acquisition in the field. This is particularly so in demonstrations of postdispersal learning, in which individuals are shown to match their songs to the local dialect after dispersing from their natal grounds. Thus, we classified any taxa in which individuals regularly alter their songs after dispersal as a ‘‘postdispersal learner’’ for the purposes of this analysis. This includes species that delete song types learned earlier that do not match neighbors. We classified taxa as predispersal learners when the predominant pattern appears to be early song acquisition, with no known examples of modification of song after natal dispersal. These categories parallel the open‐ended and close‐ended song learning categories as

GEOGRAPHIC VARIATION IN BIRDSONG

425

defined by Beecher and Brenowitz (2005) and others. Notably, song learning in some species likely bridges pre‐ and postdispersal periods. Z. leucophrys oriantha and Z. leucophrys pugetensis, for instance, are believed to memorize song models predispersal but then crystallize only a subset of memorized models postdispersal, based on vocal interactions with neighbors (Nelson, 2000). For such species, we coded song learning as primarily predispersal, based on the supposed period of song model acquisition. One notable shortcoming in our analysis is that we did not include dispersal distances, in spite of the fact that these are considered a key factor in shaping vocal geographic variation. Unfortunately, dispersal is difficult to measure and therefore not available in the literature for many species. Moreover, our system of classifying taxa as pre‐ versus postdispersal learners incorporates dispersal distance, at least to some extent. Taxa identified as having postdispersal learning have most likely been so identified because individuals have been observed dispersing across dialect boundaries. Thus, these taxa will have longer dispersal distances than will predispersal learners, almost by definition. 5. Analytical Approaches While our survey included a wide range of taxa, we decided to limit our quantitative analysis to birds, which constituted the large majority of vocal dialect cases identified (Table I, taxa #1–39). We first tested whether dialect characteristics in birds differed systematically between pre‐ and postdispersal learners. The lack of information on vocal ontogeny for many species greatly limited our sample size, and therefore the number of parameters we could include in a multivariate analysis. Because dialect characteristics such as spatial scale are potentially linked to ecological variables such as territoriality, we conducted two separate discriminant function analyses with song ontogeny as the grouping variable in both cases. In the first, we asked whether our three ecological and vocal parameters classified pre‐ and postdispersal learners as distinct groups. In the second, we asked whether the three dialect parameters classified pre‐ and postdispersal learners as distinct groups. We explored potential correlates of dialect variation using nonparametric methods, comparing, using univariate analyses, each of the dialect parameters to the ecological and vocal parameters shown in Table II. It would have been preferable to apply the comparative method, that is, taking phylogenetic relationships into account in our analysis. This was not practical, however, given the limited information currently available about the relationships of the taxa examined. It seems unlikely that phylogenetic factors would have strongly influenced our findings, given that dialect characteristics such as spatial scale can vary widely even among closely related taxa (see, e.g., Zonotrichia species and subspecies, #6–9, Table II).

426

JEFFREY PODOS AND PAIGE S. WARREN

C. RESULTS The majority of research on vocal geographic variation has been conducted on songbirds from temperate regions. Considerably, less is known about vocal geographic variation in other bird groups in other regions, notably tropical suboscine passerines. Thus, the sample analyzed here does not represent an unbiased snapshot of natural vocal variation. Our first impression from our survey is that the absolute number of species exhibiting sharp dialects is lower than the abundant literature on dialects might lead one to suspect. Most entries are for dialects in song (n ¼ 36), but 5 are for group calls. In five cases for which intraspecific variation could not easily be summarized, we use multiple entries for a single species. One of these is the black‐capped chickadee, for which dialects are reported in two different vocal signals, the ‘‘gargle’’ and the ‘‘fee bee’’ (Table II). In the other cases, populations or subspecies are so distinct in their reported dialect or other characteristics as to warrant separate entries. For example, three white‐crowned sparrow subspecies (Z. leucophrys nuttalli, Z. leucophrys oriantha, and Z. leucophrys pugetensis, #6–7) differ both in the characteristics of the dialects they exhibit as well as in reported patterns of migratory behavior (Table II). In the bronzed cowbird (M. aeneus), dialects occur in distinct geographic patterns in breeding versus wintering populations (Warren, 2002). The majority of the cases of vocal dialects we identified occur in passerine birds (n ¼ 37). These were distributed among just eight families (sensu Sibley and Monroe, 1990), with most examples from four subfamilies of the Fringillidae (n ¼ 18 species, Fig. 4). The Fringillids represent roughly 40% of the species that were found to exhibit some degree of song sharing, according to a survey of the Fringillids by Handley and Nelson (2005). Slater (1989) and others have argued that dialects are a rare phenomenon. According to our chapter, they certainly seem to be uncommon in passerines, occurring in less than a quarter of passerine families, at least as based on available data. Yet, dialects may actually be a common form of geographic variation in the Fringillids, a group with a high propensity for song sharing (45 of the 65 taxa reviewed by Handley and Nelson, 2005).

1. Variation in Dialect Characteristics The three dialect characteristics, spatial scale, temporal stability, and bilingualism, were not correlated with one another (all Spearman’s r < 0.25, p > 0.2, N ¼ 27, except N ¼ 20 for temporal stability‐bilingualism comparison). Spatial scale and temporal stability were each normally distributed across our sample. It was more difficult to assess bilingualism than the other

427

GEOGRAPHIC VARIATION IN BIRDSONG

20 18

18

Number of species

16 14 12 10 8 6 4

3

3 2

2

2 1

1

1

1

1

ae lid

Tr

oc

hi

id ac

iid

Ps

itt

in ar ct

ae

ae

ae

ae

id

ur

tid ea

en

M

lla Ca

Ne

ae

ae

rid

id rv

Co

Pa

ae id

ae

er

iid

ss Pa

llid

Ce

gi in Fr

rth

ae

0

FIG. 4. Number of species in which vocal dialects were identified, according to family, in birds. All but two families, the Psittacidae and Trochilidae, are passerines.

two variables, with many cases lacking sufficient detail. Of the 27 cases in which this could be assessed with confidence, the majority, 74%, exhibited some level of bilingualism.

2. Ecological Correlates of Dialect Characteristics The 42 dialect systems we found exhibit a wide variety of ecological characteristics, with representatives of both territorial and social species and of both sedentary and migratory species (Fig. 5, Table II). Given the dominance of passerine birds in the sample, we were not surprised to find that a majority of the birds with dialects are territorial (79%, Fig. 5). It was more surprising to find that sedentary species were the minority in the sample (Fig. 5), since it is thought that dialects should evolve more readily in sedentary species. Across our dialect sample, territorial species tended to be sedentary more frequently than did social species (w2 ¼ 3.83, p ¼ 0.05). This relative dearth of sedentary species with dialects might be due to the dominance in our sample of temperate and arctic species, many of which are seasonally migratory.

428

JEFFREY PODOS AND PAIGE S. WARREN

100% 80% Mig 60% 40% Sed 20% 0%

Territoriality

Seasonal mobility

FIG. 5. Distribution of ecological correlates among species with vocal dialects. Bars indicate the percentage of species that are territorial (black) versus nonterritorial (white) and sedentary (‘‘sed’’) versus migratory (‘‘mig’’).

We found few significant relationships between ecological variables and dialect variables, but the trends uncovered suggest potential mechanisms underlying variation among dialect systems. Migratory species tend to have larger dialect regions (Kruskal‐Wallis, Z ¼
1.85, p ¼ 0.06), and were somewhat more likely than sedentary species to show bilingualism at dialect boundaries (w2 ¼ 2.82, p ¼ 0.09). But we found no relationship between seasonal mobility and the stability of dialect regions over time. By contrast, territorial species tend to have more temporally stable dialects than do social species (Kruskal‐Wallis, Z ¼
1.9, p ¼ 0.05). But we find no effect of territoriality on spatial scale of dialects or the propensity for bilingualism. 3. Vocal Correlates of Dialect Characteristics We find that 32% (12 of 37) of songbird species with dialects in our review have repertoires of only a single song type, and that 68% (25 species) have repertoires of fewer than five song types. According to Beecher and Brenowitz (2005), the corresponding percentages for all passerines are 30% of single song type and 50% repertoires of less than five song types. The two data sets thus correspond closely in this regard. As with the ecological variables, there are no statistically significant correlations between repertoire size and dialect characteristics. However, species with smaller repertoires tend to have more stable dialects over time (Spearman’s r ¼
0.38, p ¼ 0.05) and a lower propensity for bilingualism (Spearman’s r ¼
0.36, p ¼ 0.12). Repertoire size is not associated with dialect scale.

429

GEOGRAPHIC VARIATION IN BIRDSONG

4. Song Ontogeny

Number of taxa

Dialect species were somewhat more likely to show postdispersal rather than predispersal learning (Fig. 6). While this lends greater support for the social adaptation than local adaptation hypothesis, neither is strongly supported. The social adaptation hypothesis can be rejected in the nine dialect systems in which individuals rarely disperse across dialect boundaries and apparently do not acquire the local song type when they do cross boundaries. While some of the cases of predispersal learning have been disputed (Baptista and Petrinovich, 1984) or are based on single studies (Kreutzer, 1974), many others are well established (Lachlan and Slater, 2003). Thus, each of the functional hypotheses can be rejected in at least some cases. Furthermore, there are many missing cells in the table. The timing of song acquisition and memorization relative to natal dispersal is not known for more than a third of the cases in our table. Multivariate approaches identified few good predictors of the timing of song acquisition. First, we conducted a discriminant function analysis using the ecological and vocal parameters, territoriality, seasonal mobility and repertoire size, and song ontogeny as the grouping variable. This discriminant function was not significant (Exact F ¼ 1.20, df ¼ 3, N ¼ 18, p ¼ 0.33), with a canonical correlation of 0.45 for the first canonical axis. We conducted a second discriminant function analysis using three factors specifically describing the dialect systems, the spatial scale of dialects, their

20 18 16 14 12 10 8 6 4 2 0

16

16

Postdispersal

Unknown

10

Predispersal

FIG. 6. Numbers of all taxa with dialects that exhibit either predispersal learning or postdispersal learning, or for whom timing of learning has not been described. The two functional hypotheses, local adaptation and social adaptation, make mutually exclusive predictions regarding timing of song memorization and acquisition.

430

JEFFREY PODOS AND PAIGE S. WARREN

temporal stability, and presence of bilingualism. This was also not significant (Exact F ¼ 1.89, df ¼ 3, N ¼ 10, p ¼ 0.20) with a canonical correlation of 0.60 for the first canonical axis. Eliminating bilingualism from the analysis, a variable with many missing data points, improves the classification considerably, as shown in Fig. 7. This classification is statistically significant (Exact F ¼ 6.01, df ¼ 3, N ¼ 15, p ¼ 0.01), with the first canonical axis (Canon1) accounting for 67% of the variation among the cases of dialects. Temporal stability shows the highest correlation with Canon1 (r ¼ 0.88, p < 0.0001), but dialect size is also strongly correlated with Canon1 (r ¼ 0.59, p ¼ 0.002). Thus, the temporal and spatial features of dialects appear to be moderately successful predictors of learning ‘‘strategy’’ (sensu Beecher and Brenowitz, 2005). Temporal stability of dialects remained a significant predictor of learning strategy in univariate tests. This and repertoire size were the only parameters to show a tendency to differ between pre‐ and postdispersal learners. Postdispersal learners were slightly more likely to have larger maximum repertoire sizes than predispersal learners (Kruskal‐Wallis, 3.5 18

14, 15, 21

3.0

8

2.5

Spatial scale

5

16

+

1.5

Pre

Post

Canonical 2

4, 7, 28 30, 31

2.0

+

19, 22, 24, 27

9 Temporal stability

1.0 3, 29, 38 37, 39

0.5

6 0.0

2, 23

−0.5 1.0

1.5

2.0

2.5

3.5 3.0 Canonical 1

4.0

4.5

5.0

5.5

FIG. 7. Results of discriminant function analysis, using the spatial scale and temporal stability of dialect regions as independent variables and the timing of song acquisition as a grouping variable. Circles indicate the 50% confidence interval for predispersal (dark circle) and postdispersal learners (light circle). All bird species are plotted and numbered according to their entries in Table II.

GEOGRAPHIC VARIATION IN BIRDSONG

431

Z ¼
1.54, p ¼ 0.12). This finding is supported by the observation that all seven predispersal cases have repertoires with five or fewer song types. Dialects in predispersal learners are also significantly more stable over time (Kruskal‐Wallis, Z ¼ 2.43, p ¼ 0.015), though there is clearly considerable variation in both groups. Some dialect systems in postdispersal learners are quite stable over time, for example in brown‐headed cowbirds (M. ater) (Fleischer and Rothstein, 1988). Dialects in predispersal learners thus occur in species with smaller repertoires, and when they occur, they tend to be relatively stable in time. By contrast, dialects in postdispersal learners tend to cover smaller regions and be less stable over time, though there is considerable variation among dialects exhibited by postdispersal learners.

D. DISCUSSION Our survey illustrates both the rarity and diversity of dialect systems in nature. No single functional hypothesis can account for all or even a majority of published examples of dialects, and a substantial portion of the cases reject at least one of these hypotheses. The local adaptation hypothesis is rejected for 16 cases of dialects in which individuals are capable of modifying songs after dispersal to new dialect regions (Fig. 6). Likewise, the social adaptation hypothesis is rejected for 10 cases in which individuals appear not to disperse across dialect boundaries or to modify their signals after natal dispersal (Fig. 6). Although timing of song acquisition remains unknown for more than a third of all dialect systems (Table II), it seems unlikely that further work will support one of these functional hypotheses absolutely over the other. We suggest that the diversity of dialect systems, occurring as it does across a range of social systems, scales, and ecological conditions, argues against any given functional hypothesis of dialect evolution. The assembled data provide additional insights into the selective forces underlying the evolution of these diverse patterns of dialect variation. First, we note some common features across the species in our chapter. Although we searched intensively for cases of dialects in such well‐studied acoustic performers as insects and anurans, we only found dialects in species with evidence of vocal learning. In birds, these were the passerines, psitticines, and trochilids. Among mammals, the cetaceans and primates were the only groups represented. The majority of the cases (37 of 42) were in passerine birds. The predominance of passerines in the sample no doubt reflects, to some extent, historical and taxonomic biases in research on geographic variation in acoustic signaling. However, it remains noteworthy that no examples of dialects to date have been

432

JEFFREY PODOS AND PAIGE S. WARREN

found in species lacking imitative learning. We conclude that imitative learning, as predicted by Kroodsma (1996) among others, is a necessary condition for dialect formation. Our survey also reveals a strong incidence of dialects within the Fringillids, including species that share songs with neighbors (compare our data with Handley and Nelson, 2005). At the surface, this relationship suggests potential support for social adaptation hypotheses such as the deceptive mimicry hypothesis. However, as we argue in Section IV, and as was argued by Slater (1989), evidence of relationships between song sharing and vocal dialect evolution provides more direct support for by‐product hypotheses of dialect evolution. 1. Ecological and Vocal Correlates of Dialect Parameters Our comparisons of vocal and ecological parameters in dialect species suggest several issues worth pursuing in future work. First, the spatial characteristics of dialects appeared to be associated with seasonal mobility, with migratory species tending to evolve larger dialect regions. These interspecific comparisons appear to corroborate patterns described for the three subspecies of the white‐crowned sparrow with dialects, which range from the sedentary Z. leucophrys nuttalli with its small dialect regions to the migratory Z. leucophrys oriantha and Z. leucophrys pugetensis with their larger dialect region (Nelson, 1999). Further refinement of comparisons to include distances traveled by migratory populations may reveal an even stronger relationship with dialect size. Second, territorial species tend to maintain the same dialects for longer periods of time than do social species such as the colonial yellow‐rumped cacique (Trainer, 1989). Perhaps more revealing is the lack of a relationship between migratory behavior and stability of dialects. This suggests that seasonal mobility alone does not disrupt dialect patterns. Other life history characteristics that could be addressed include the degree of site fidelity and the rate of population turnover (Handley and Nelson, 2005; Kroodsma, 1996; Wiens, 1982). Third, larger song repertoires are associated with more stable dialect regions, but not with other dialect characteristics. This may reflect underlying selection on either repertoire size or accuracy of vocal imitation (Beecher and Brenowitz, 2005). Species in which accurate imitation is advantageous typically have lower repertoire sizes and are expected to have more stable song neighborhoods (Beecher and Brenowitz, 2005). We note, however, that predispersal learners also tend to have smaller repertoires and more stable dialects. The significant correlation between repertoire size and dialect stability disappears when we treat pre‐ and postdispersal learners separately. Thus, an unresolved question is whether

GEOGRAPHIC VARIATION IN BIRDSONG

433

the stability of geographic variation in song is a consequence of timing of song acquisition, or of selection on traits such as repertoire size or imitative learning. The timing of song acquisition itself may be a consequence of selection on other traits such as dispersal distances or length of breeding seasons (Nelson, 1999). 2. Role of Song Ontogeny The evolutionary consequences of geographic variation in song are thought to be determined in large part by song ontogeny, in particular by the timing of song acquisition and memorization (Slabbekoorn and Smith, 2002a). Our survey of birds with dialects found examples of species with both pre‐ and postdispersal learning. These groups differ somewhat in the characteristics of their dialects, particularly in the stability of dialects over time (Fig. 7). But these differences are not overwhelming, and lack of information on song ontogeny in our sample significantly hampers our ability to draw conclusions about its role in dialect evolution. Nevertheless, it seems clear that song ontogeny is only one of many characteristics of species that influence how geographic variation in song evolves.

IV. RECENT STUDIES OF AVIAN VOCAL EVOLUTION, AND HOW THEY SUPPORT BY‐PRODUCT MODELS OF VOCAL GEOGRAPHIC DIVERGENCE As of two decades ago, researchers had mustered little direct empirical support for functional hypotheses of dialect evolution (Kroodsma, 1985a). Our survey in the preceding section suggests that little has changed on this front. There is and probably will never be a simple, universal explanation for the phenomenon of dialects. Perhaps more importantly, we suggest that the traditional focus on dialects has eclipsed investigation into the broader phenomenon of vocal geographic evolution, of which dialects are but one form expressed. In this section, we argue that by‐product hypotheses of vocal geographic evolution—the only one of the three sets of dialect hypotheses that seems to have broader applicability—have been receiving new sources of support, through recent general advances in the study of birdsong. A. PHYLOGENETIC SIGNAL IN VOCAL EVOLUTION Over decades, the study of behavioral evolution has been transformed by increased attention to phylogenetic factors (Brooks and McLennan, 1991; Martins, 1996). This has been the case for the study of birdsongs, in spite of the prior presumption that these signals are too plastic to permit

434

JEFFREY PODOS AND PAIGE S. WARREN

historical analyses (Irwin, 1996; Payne, 1986; Price and Lanyon, 2002; ten Cate, 2004). The principal approach used so far to account for historical factors in bird vocal evolution has involved comparative analyses (Catchpole, 1980; Irwin, 1990; Kroodsma, 1977; Podos, 1997; Read and Weary, 1992; Slabbekoorn et al., 1999; Wiley, 1991). Toward this end, reference to hypotheses of phylogenetic relationships has proven particularly helpful. Phylogenetic hypotheses enable researchers to estimate vocal ancestral character states, as has now been done in a number of bird groups (de Kort and ten Cate, 2004; Irwin, 1988; Payne, 1986; Price and Lanyon, 2002, 2004). In their studies of oropendolas and caciques, to illustrate, Price and Lanyon (2002, 2004) used ancestral state reconstruction to infer that some vocal parameters (e.g., note structure, peak frequencies) are highly variable, whereas other vocal parameters (e.g., the presence or absence of a trill or a click within songs, and note and song duration) have remained stable over evolutionary time. Phylogenetic hypotheses have also allowed for independent contrast and similar statistical analyses, to assess correlated evolution in other taxonomic groups between song traits and neural, morphological, or ecological parameters (Podos, 2001; Seddon, 2005; Sze´kely et al., 1996; Van Buskirk, 1997). Returning to the example of oropendolas and caciques, phylogenetic reconstruction enabled analysis of correlations between rates of vocal evolution and the intensity of sexual selection in different lineages (Price and Lanyon, 2002, 2004). B. MULTIPLE FUNCTIONS AND TRADE‐OFFS IN VOCAL EVOLUTION Evidence of historical signal in vocal evolution suggests alternatives to functional hypotheses of evolution, which assume that vocal traits are sufficiently plastic to be easily molded to whatever selective pressures are presently in play. Instead, vocal features may be evolutionarily conserved when they are subject to multiple evolutionary pressures (Beecher and Brenowitz, 2005; Gil and Gahr, 2002; Nowicki and Podos, 1993). A rigorous empirical example of how song may evolve under multiple selection pressures was provided by Seddon (2005), who documented concurrent impacts of morphological adaptation, interspecific competition, and acoustic adaptation on the divergence of vocal structure in Neotropical antbirds (Passeriformes: Thamnophilidae). When traits are subject to multiple selection pressures, they may face trade‐offs in their evolution, that is, by responding only partially to some pressures (Roff, 1992). Evolutionary responses to specific functions, through their impacts on morphology, physiology, and behavior, may also alter how other traits are expressed, or even provide opportunities for the evolution of new functions (Gould and Vrba, 1982; Patek et al., 2006). The relevance of these issues for questions about

GEOGRAPHIC VARIATION IN BIRDSONG

435

bird vocal evolution has been highlighted by recent advances in two areas: on mechanisms of vocal learning and on mechanisms of vocal production. Advances in both areas, we argue below, support by‐product models of vocal geographic divergence. 1. Mechanisms of Vocal Learning Researchers have long wondered why some animals have evolved imitative learning as a mechanism in vocal signal ontogeny (Nottebohm, 1972). After all, many other animals develop effective vocal signals without recourse to imitation. Two traditional explanations for the evolution of vocal imitation are that it enables transmission of particularly complex patterns of vocal structure across generations, and that it helps animals to adapt their vocal signals to local acoustic environments (Slater, 1986). Recent work in songbirds has expanded this list in two significant directions. a. Vocal imitation and the development of song sharing among territorial neighbors One of the main documented functions of birdsong is to mediate interactions among neighboring territorial males (Catchpole and Slater, 1995; Hyman, 2002; Searcy and Andersson, 1986; Todt and Naguib, 2000). Research has focused on the use of ‘‘shared’’ songs by interacting neighbors and the potential fitness benefits of song sharing (Brown and Farabaugh, 1997; Handley and Nelson, 2005; Lachlan et al., 2004; Molles and Vehrencamp, 2001; Payne and Payne, 1997; Todt and Naguib, 2000). Two hypotheses to explain song sharing are that it provides a reliable means by which males can discriminate neighbors from strangers, and that it enables increased precision in communicating aggressive intent among territorial neighbors. Birds do well to distinguish neighbors from strangers because, unlike neighbors, strangers are ‘‘inherently expansionist’’ (Beecher and Brenowitz, 2005, p. 147), and thus require closer monitoring. Research by Beecher and colleagues on song sparrows (Melospiza melodia) illustrates how neighboring males may employ shared songs to escalate or de‐escalate aggressive interactions. To escalate an interaction, males may first respond to a singing neighbor with a nonshared song, then by singing a ‘‘repertoire match’’ (a shared song, although not the one being sung by the neighbor at the moment), and then by singing a precise type match (Beecher and Brenowitz, 2005). Selection for song sharing may influence the evolution of vocal geographic patterns—beyond obvious effects on the finest‐scale geographic patterns, that is, among neighbors—because of how it shapes the evolution of song learning strategies in populations (Beecher and Brenowitz, 2005; Handley and Nelson, 2005; Slater, 1989). First of all, selection for song sharing may presumably influence the number of song types that birds will evolve to learn, that is song repertoire size. As stated by Beecher and Brenowitz (2005,

436

JEFFREY PODOS AND PAIGE S. WARREN

p. 147), ‘‘selection for song sharing and selection for large song repertoires are at least partially contrary . . . as a logical consequence of the fact that a song learning strategy cannot optimize both goals.’’ This is because selection for song sharing may favor, Beecher and Brenowitz (2005) argue, the evolution of smaller repertoires with correspondingly greater percentages of songs shared among neighbors. Expanding to a broader geographic scale, species that evolve smaller repertoires may be more likely to express the classic dialect pattern (Williams and Slater, 1990). A second potential impact of selection for song sharing is on the timing of song learning. Song sharing may be promoted in species that retain sufficient flexibility to match songs produced by neighbors on their breeding grounds, after dispersal from natal grounds (Beecher and Brenowitz, 2005; Martens and Kessler, 2000; Trainer, 1989). For species or subspecies with significant postnatal dispersal, ‘‘open‐ended’’ song learning programs should be favored because such learning programs increase the likelihood that birds would be able to match the songs of neighboring males (Rothstein and Fleischer, 1987). Evolution of the classic dialect pattern may thus be facilitated indirectly, via selection for song sharing. The observation that many Fringillids express both song sharing and dialects (Handley and Nelson, 2005; Section III) supports this hypothesis. b. Vocal imitation and male quality Another proposed function of vocal imitation is that it enables song structure to be used as an accurate indicator of male quality (Buchanan et al., 1999, 2003; Nowicki et al., 1998, 2002). This hypothesis suggests that learned songs serve as honest indicators of male quality because their accurate reproduction requires successful brain development in the face of potentially severe nutritional and developmental stress. Males who accurately reproduce vocal features of song tutors, or who are able to develop complex vocal features, in effect advertise high‐ quality genes, developmental histories, and learning abilities. Such males would presumably offer higher quality genetic input and rearing environments for females that choose them as mates. Studies suggest that nutritional or developmental stress may indeed impair the normal development of vocal brain nuclei (Buchanan et al., 2004; MacDonald et al., 2006; Nowicki et al., 2002; but see Gil et al., 2006). Empirical research on the potential consequences of developmental stress for song development has focused especially on three vocal structural parameters: syllable or song repertoire size, the accuracy of vocal tutor matching, and rates or durations of vocal output (Buchanan et al., 2003; Nowicki et al., 2002; Spencer et al., 2003, 2004, 2005). Presumably, males of higher quality will be able to develop larger vocal repertoires, given the correlations between early stress, song nuclei volume, and vocal repertoire

GEOGRAPHIC VARIATION IN BIRDSONG

437

size (Nowicki et al., 2002). An honest indicator mechanism such as this might help to explain female preferences for large vocal repertoires, as shown in some species (Searcy and Yasukawa, 1996). Accuracy in imitation may similarly provide an honest indicator of a male’s brain developmental status, given the complexity of the auditory and neural systems that are devoted to the process, and given the numerous environmental challenges that may impede song nucleus development (Nowicki et al., 2002). Rates or durations of vocal output may provide an indicator of the quality of a male’s overall health and developmental history (Spencer et al., 2005). Nowicki et al. (1998) noted a possible trade‐off in the evolution of copying accuracy and repertoire size development, in which a premium on imitation accuracy (quality of copying) might impede the development of large repertoires (quantity of copying) and vice versa. In terms of neural mechanisms, selection for high imitation accuracy may perhaps constrain the development of larger repertoires if the brain space and developmental resources required for accurate imitation secondarily limit the quantity of vocal material that can be imitated in the first place. Our point with this example is that selective pressures on repertoire size and the accuracy of vocal tutor matching may impose, either independently or jointly, secondary effects on the evolution of vocal geographic patterns. Species with larger repertoires are generally thought not to evolve stringent dialects (reviewed by Searcy et al., 2002), a supposition that is partly supported by our analysis in Section III. Beyond the question of whether dialects occur or not, species that have evolved a premium on copying accuracy may evolve greater stability and within‐neighborhood similarity in the structure of their songs (Slater, 1986; see also our analyses in Section III). Vocal geographic structure may thus arise indirectly through selection favoring males with learning programs that render them more adept at ‘‘managing’’ interactions for selfish gain (Kroodsma, 1996), or through selection on song structure as an honest indicator of male genetic and developmental quality (Nowicki et al., 2002). These arguments echo Slater’s suggestion that selection at the level of individuals drives geographic patterns that appear at the level of populations (Slater, 1989). 2. Mechanisms of Vocal Production The ontogeny and evolution of vocalizations are impacted not just by learning but also by the mechanisms that underlie vocal production (Elemans et al., 2004; Fee et al., 1998; Nowicki et al., 1992; Podos and Nowicki, 2004a; Suthers and Goller, 1997). Some recent studies have emphasized the fact that vocal production in birds requires the input and activity of not just the sound source (the syrinx) but also other motor components, including the respiratory system and vocal tract (Beckers et al., 2003; Hoese

438

JEFFREY PODOS AND PAIGE S. WARREN

et al., 2000; Nowicki, 1987; Nowicki and Marler, 1988; Riede et al., 2006; Suthers, 2004; Suthers et al., 1999). These additional motor components of vocal production can impose performance limits on the expression and evolution of song features (reviewed by Podos and Nowicki, 2004a). Movements of respiratory muscles, to illustrate, are coordinated precisely with syrinx activity, and appear to be essential in controlling the timing of vocal output (Suthers et al., 1999). Maximal rates of breathing cycles may thus limit the evolution of temporal modulations in song. Components of the vocal tract, including the trachea, larynx, and beak, modify the spectral structure of song, and in particular serve to dampen harmonic overtones and thus enable the production of pure‐tonal songs (Beckers et al., 2003; Hoese et al., 2000; Nowicki and Marler, 1988; Riede et al., 2006; Westneat et al., 1993). Maximal rates of vocal tract reconfiguration, such as those achieved through changes in beak gape, can limit trill rates and frequency bandwidth within trilled vocalizations (Nowicki et al., 1992; Podos, 1997). Recent empirical studies suggest two related effects of production constraints on vocal evolution. First, production constraints can bias the evolution of individual vocal features. The evolution of trill rate in swamp sparrows (Melospiza georgiana), to illustrate, appears to be limited by individual birds’ vocal performance abilities. This was revealed in a study in which young male swamp sparrows were trained with tutor songs in which trill rates had been artificially elevated (Podos, 1996; see also Podos et al., 1999). Birds proved able to memorize the rapid tutor songs, but unable to produce accurate copies of these songs, in manners consistent with a hypothesis of motor constraints on song production (Podos, 1996). Second, production constraints on vocal evolution may be manifest not only in individual features but also as trade‐offs among multiple vocal features. Songs of birds of the sparrow family Emberizidae, to illustrate, exhibit a trade‐off between trill rates and frequency bandwidth (Podos, 1997). Similar patterns have now been described in additional taxa (Ballentine et al., 2004; Draganoiu et al., 2002; Illes et al., 2006), and evidence also suggests that songs produced at higher performance levels—that is, with greater trill rates and/or frequency bandwidths—are more effective with regard to both inter‐ and intraspecific function (Ballentine et al., 2004; Illes et al., 2006). An acoustic trade‐off between trill rate and frequency bandwidth is consistent with a hypothesis of physical constraint: in order to achieve particularly rapid trill rates, the requirement for pure‐tonal quality (and thus rapid vocal tract reconfigurations) sets performance limits on the ranges of frequencies that can be produced over a given time interval (Podos and Nowicki, 2004a). Physical limits or trade‐offs in vocal evolution are worth attention because they may counter selection for particular functions. Thus, to

GEOGRAPHIC VARIATION IN BIRDSONG

439

illustrate, selection for songs with particularly rapid trills, for example under a local adaptation scenario, would presumably be impeded by physical constraints on trill production. Selection on components of the vocal apparatus for nonvocal functions may also invoke secondary effects on vocal evolution (Podos and Hendry, 2006; Podos and Nowicki, 2004b). Body size, to illustrate, evolves in many animals in response to a range of selective factors such as thermoregulation, fecundity, reproductive rate, and dispersal (Blanckenhorn, 2000; Roff, 1992). Resulting changes in body size may impose secondary impacts on the fundamental frequencies of bird vocalizations, given tight correlations between body size and syrinx size, and the functional relationship of syrinx size and vocal frequency production (Bertelli and Tubaro, 2002; Cutler, 1970; Ryan and Brenowitz, 1985). A second scenario, involving beak and song evolution, has been illustrated recently for Darwin’s finches of the Gala´pagos Islands, Ecuador. In these birds, beak form and function has been shown to evolve in precise correspondence with varying ecological parameters, namely food availability and interspecific competition (Grant and Grant, 1995, 2002, 2006). Analyses of songs of birds with known morphologies have now revealed that the same two vocal parameters mentioned above, trill rate and frequency bandwidth, correlate with variation in beak morphology (Huber and Podos, 2006; Podos, 2001). This correlation seems likely to be the result of proximate constraints on beak gape changes and thus vocal tract configurations during vocal production. Consistent with this hypothesis, birds with larger beaks, predicted to suffer greater constraints on vocal performance (Nowicki et al., 1992), have evolved songs with slower trill rates and narrower frequency bandwidths (Huber and Podos, 2006; Podos, 2001; Podos and Nowicki, 2004b; Podos et al., 2004a). Thus, within given lineages of Darwin’s finches, morphological adaptation under selection for feeding opportunities is predicted to impact vocal performance, and thus the evolution of some vocal features, that is those vocal features that require precise tract reconfigurations for their production. These scenarios of vocal evolution suggest that adaptation to divergent environments may produce, on its own account, structured patterns of vocal geographic variation. To continue with the example of Darwin’s finches, populations of some species have diverged in genetics and morphology on different islands, presumably as a result of the different selective environments on those islands, and a result of limited gene flow between islands (Lack, 1947; Petren et al., 2005). Given the role of the beak in vocal production, adaptive divergence in beak morphology may thus have driven intraspecific, between‐island vocal divergence, with, for example, the largest‐beaked

440

JEFFREY PODOS AND PAIGE S. WARREN

populations of a given species experiencing the most severe constraints on trill evolution (Podos and Nowicki, 2004b). The main point of this example, for present purposes, is that geographic patterns of song variation may emerge as incidental by‐products of selection for nonvocal functions, without need for selection for the patterns themselves, as is implied by functional adaptation hypotheses of vocal geographic evolution.

V. EVOLUTION OF GEOGRAPHIC VARIATION IN AVIAN VOCAL SIGNALS: PROSPECTUS As we argued above, advances on both empirical and conceptual fronts provide increasing support for a role of by‐product models of vocal geographic variation. We do not, however, intend to suggest that all facets of song evolution are explained through by‐product mechanisms. Rather, there are myriad factors that may impact geographic divergence of the vocal phenotype. This final section, which follows closely from Podos et al. (2004b), is devoted to surveying the range of scenarios by which song features may diverge among different populations of a species. A. INTERPLAY OF MEMES AND MECHANISMS IN VOCAL EVOLUTION To better address the range of factors involved in vocal evolution we find it useful to distinguish two distinct ‘‘substrates’’ of vocal evolution: memes and mechanisms. Memes refer to song parameters that are transmitted across generations via learning, whereas mechanisms refer to phenotypic bases of vocal expression (development and production) that are transmitted across generations via genetic inheritance (Podos et al., 2004b). Traditional explanations for patterns of vocal geographic evolution have, in our viewpoint, been hampered by a nearly exclusive focus on meme evolution. Part of the reason for the relative neglect of mechanisms, we believe, is that their effect is normally manifest over comparatively broad timescales (Podos, 1997; Ryan and Brenowitz, 1985) and are thus more difficult to identify and study. In the evolution of learned vocalizations, vocal memes and vocal mechanisms may evolve on nonintersecting trajectories. Thus, for instance, selection for increased trill rates may augment trill rates in a population, in the event that the mechanisms responsible for trill production in that lineage are able to accommodate such increases. Similarly, evolutionary changes in body size may have no impact on the vocal frequencies expressed in a population, in the event that vocal frequencies were initially not produced near their limits of possibility. But memes and mechanisms may also interact in vocal evolution. Evolutionary changes in mechanisms

GEOGRAPHIC VARIATION IN BIRDSONG

441

of vocal production and learning may adjust potential routes of meme evolution, and evolutionary stability in vocal mechanisms may limit the response of memes to directional selection (Podos et al., 2004b). Recognition of the interplay of memes and mechanisms in vocal evolution allows us to identify five categories of potential causes for the evolution of vocal geographic variation (Podos et al., 2004b). B. POTENTIAL CAUSES OF VOCAL GEOGRAPHIC EVOLUTION Songs, like any other phenotype, evolve through the combined effects of drift and selection. We identify two scenarios that involve drift and three that involve selection. We do not claim the scenarios to be mutually exclusive or collectively exhaustive. Rather, song divergence likely involves all of these processes, emphasized to varying degrees and at different times in any lineage’s evolutionary history. We do not attempt to integrate details about dispersal patterns or the timing of learning, which must play a central role in vocal geographic evolution (Ellers and Slabbekoorn, 2003; Krebs and Kroodsma, 1980). Nor do we attempt to evaluate how long‐term ecological processes, such as changes in land use or impacts of fire on habitat, may influence bird distributions and thus dialect formation (Laiolo and Tella, 2005). 1. Cultural Drift Song features may evolve as a result of inaccurate transmission of song memes across generations because of ‘‘errors’’ in learning (Grant and Grant, 1996; Payne, 1996). Distinct trajectories of cultural evolution via copy errors may explain vocal differences among diverging populations, especially during the initial stages of divergence (Lemon, 1975; Slabbekoorn and Smith, 2002a). To illustrate, evolutionary divergence in the phonology (fine structure) of notes, resulting from inaccurate imitation, may explain interisland differences in note structure in some species of Darwin’s finches (Grant and Grant, 1996). Lineages that readily express cultural errors, along with some isolation of descendent populations, seem likely to generate vocal geographic variation. 2. Genetic Drift Song features may also evolve via random changes in the anatomical, physiological, and neural mechanisms that underlie vocal ontogeny and production—and, more specifically, in the genetic loci that underpin these mechanisms. Genetic drift may presumably impact vocal geographic evolution when song memes in a lineage are produced at or near some anatomical, developmental, or performance limit. Consider, for instance, drift in the genetic loci that underlie syrinx mass. Syrinx mass appears to set lower limits

442

JEFFREY PODOS AND PAIGE S. WARREN

on vocal frequencies, such that only larger syringes can produce lower frequency sounds. If genetic drift leads to reduced syrinx mass within a given populations, we would expect the potential for frequency production to follow suit. Thus, vocal frequencies, if initially produced near maximal performance capacities (comparatively low frequencies), may accordingly be ‘‘bumped’’ to higher levels in the offshoot population (Podos et al., 2004b). Genetic drift may also alter the structure and function of the brain nuclei involved in song learning, for instance, through random alterations in the timing of interactions between song nuclei (Livingston et al., 2000). Such random changes may have consequences for the timing and content of song acquisition. 3. Cultural Selection Cultural selection occurs when certain vocal memes are favored over others, as a result of the differential effectiveness of those memes in the process of communication. A primary example concerns selection for optimal sound transmission. Songs are known to vary in how well they transmit in different environments, and cultural selection is thought to thus shape certain vocal parameters (Slabbekoorn, 2004; Wiley and Richards, 1978). Songs with slow repetition rates and low frequencies, to illustrate, have been shown to evolve more often in forested habitats than in other habitats, presumably because slow, low‐frequency songs suffer relatively less degradation in forested habitats than elsewhere. Songs with more effective transmission properties may be favored by selection not only in the context of interactions among adults but also in song model imitation by juveniles (Hansen, 1979). Cultural selection for optimal sound transmission has been implicated in the divergence of song among populations of a number of species (Doutrelant et al., 1999; Handford and Lougheed, 1991; Hunter and Krebs, 1979; Ruegg et al., 2006; Slabbekoorn and Smith, 2002b; Wiley, 1991). 4. Natural Selection Natural selection may drive vocal geographic divergence through its influence on either memes or mechanisms. With respect to memes, natural selection may facilitate vocal divergence via ‘‘reinforcement,’’ in which selection against hybrid production favors those birds that produce the most species, population, or locality distinctive songs (Butlin and Ritchie, 1994; Marler, 1957, 1960; Nelson and Marler, 1990; Ptacek, 2000). This is the broader context in which we would place the local adaptation hypothesis of dialect evolution. The example of the multifunctional role of the beak in singing and feeding, discussed in the previous section, illustrates how natural selection on mechanisms can cause incidental vocal evolution (Nowicki et al., 1992; Podos and Nowicki, 2004a,b). To reiterate, natural selection in the context of selection for food availability, food type, and interspecific competition is known to

GEOGRAPHIC VARIATION IN BIRDSONG

443

drive precise changes in beak form and function (Grant and Grant, 1995, 2002, 2006). Given the role of the beak in vocal production, such evolution can drive vocal changes as a secondary consequence (Podos and Nowicki, 2004b). 5. Sexual Selection Sexual selection is traditionally regarded as favoring elaborate or complex forms of vocal signals, especially as a result of female choice (Andersson, 1994; Catchpole and McGregor, 1985; Searcy and Andersson, 1986; Searcy and Yasukawa, 1996). Sexual selection may also favor vocal features that challenge males’ developmental and performance capacities (Nowicki et al., 2002) or that enable increased precision in communication in male–male interactions (Beecher and Brenowitz, 2005; Todt and Naguib, 2000). As a general observation, the course of sexual selection is often haphazard, with different signal parameters favored or exaggerated in different lineages (Boughman, 2001; Panhuis et al., 2001). Divergent pathways of sexual selection on song may similarly result in signal divergence in offshoot populations, at least to the extent that populations remain in genetic and cultural isolation. To illustrate we turn again to potential trade‐offs involving repertoire size. In some lineages, female preferences for complex signals may favor the evolution of large repertoires, whereas in other lineages female preferences for accurate imitation may favor small repertoires. Moreover, selection for song sharing among males in other lineages may favor moderate‐sized repertoires. Divergence of sexual selection pressures among populations may thus presumably lead to geographic divergence in repertoire size.

VI. SUMMARY Our goal in this chapter has been to evaluate, from both empirical and conceptual perspectives, the factors that facilitate the evolution of geographic variation in bird vocalizations. Studies on this topic have traditionally focused on the evolution of song ‘‘dialects,’’ and have emphasized functional hypotheses to explain their evolution. Two such hypotheses, ‘‘local adaptation’’ and ‘‘social adaptation’’ hypotheses, focus on the potential role of song in aiding recognition of males, either by locality or by social group. A quantitative survey of results from papers published on dialects, between 1962 and 2006, however, suggests limited direct support for functional hypotheses. An alternative set of hypotheses suggests that song features may diverge through ‘‘by‐product’’ scenarios, in which selection for nonrecognition functions drives incidental changes in song structure, and geographic variation therein. Examples of such functions involve the evolution of song learning in neighbor–neighbor song sharing and the

444

JEFFREY PODOS AND PAIGE S. WARREN

evolution of song learning in the context of sexual selection for male quality. We also describe scenarios by which songs may diverge indirectly through selection on components of the vocal apparatus such as body size and beak form and function. To conclude, we outline scenarios by which songs may diverge geographically; via cultural drift, genetic drift, cultural selection, natural selection, and sexual selection. Empirical study of these scenarios, together with countinued descriptions of vocal learning strategies and patterns of dispersal, may provide insights into vocal geographic evolution and thus propensities for speciation by reproductive isolation.

Acknowledgments J.P. gratefully acknowledges financial support from the National Science Foundation (NSF IOB‐0347291). P.W. gratefully acknowledges financial support from the National Science Foundation (NSF IBN‐98‐01490) and the Zoology Scholarship Fund for Excellence (Dorothea Stengl) at University of Texas. Helpful comments on previous versions of this chapter were provided by M. Naguib, P. Slater, L. Higgins, D. Hillis, M. Kirkpatrick, C. Sexton, M. Ryan, and W. Wilczynski.

References Alderson, G. W., Gibbs, H. L., and Sealy, S. G. (1999). Determining the reproductive behaviour of individual brown‐headed cowbirds using microsatellite DNA markers. Anim. Behav. 58, 895–905. Andersson, M. (1994). ‘‘Sexual Selection.’’ Princeton University Press, Princeton, NJ. Andrew, R. J. (1962). Evolution of intelligence and vocal mimicking. Science 137, 585–589. Austen, M. J. W., and Handford, P. (1991). Variation in the songs of breeding Gambel’s white‐ crowned sparrows near Churchill, Manitoba. Condor 93, 147–152. Avery, M., and Oring, L. W. (1977). Song dialects in the bobolink (Dolichonyx oryzivorus). Condor 79, 113–118. Baker, M. C. (1974). Genetic structure of two populations of white‐crowned sparrows with different song dialects. Condor 76, 351–356. Baker, M. C. (1975). Song dialects and genetic differences in white‐crowned sparrow (Zonotrichia leucophrys). Evolution 29, 226–241. Baker, M. C. (1982). Vocal dialect recognition and population genetic consequences. Am. Zool. 22, 561–569. Baker, M. C. (1983). The behavioral response of female Nuttall’s white‐crowned sparrows to male song of natal and alien dialects. Behav. Ecol. Sociobiol. 12, 309–315. Baker, M. C., and Cunningham, M. A. (1985). The biology of bird‐song dialects. Behav. Brain Sci. 8, 85–133. Baker, M. C., and Mewaldt, L. R. (1978). Song dialects as barriers to dispersal in white‐ crowned sparrows, Zonotrichia leucophrys nuttalli. Evolution 32, 712–722. Baker, M. C., and Mewaldt, L. R. (1981). Response to ‘‘Song dialects as barriers to dispersal: A re‐evaluation.’’ Evolution 35, 189–190. Baker, M. C., and Thompson, D. B. (1985). Song dialects of sparrows: Historical processes inferred from patterns of geographic variation. Condor 87, 127–141.

GEOGRAPHIC VARIATION IN BIRDSONG

445

Baker, M. C., Spitler‐Nabors, K. J., and Bradley, D. C. (1981). Early experience determines song dialect responsiveness of female sparrows. Science 214, 819–821. Baker, M. C., Thompson, D. B., Sherman, G. L., Cunningham, M. A., and Tomback, D. F. (1982). Allozyme frequencies in a linear series of song dialect populations. Evolution 36, 1020–1029. Baker, M. C., Baker, A. E. M., Cunningham, M. A., Thompson, D. B., and Tomback, D. F. (1984a). Reply to ‘‘Allozymes and song dialects: A reassessment.’’ Evolution 38, 449–451. Baker, M. C., Tomback, D. F., Thompson, D. B., Theimer, T. C., and Bradley, D. C. (1984b). Behavioral consequences of song learning: Discrimination of song types by male white‐ crowned sparrows. Learn. Motiv. 15, 428–440. Baker, M. C., Tomback, D. F., Thompson, D. B., and Cunningham, M. A. (1985). Reply to Hafner and Petersen. Evolution 39, 1177–1179. Baker, M. C., Bjerke, T. K., Lampe, H. U., and Espmark, Y. O. (1987a). Sexual response of female yellowhammers to differences in regional song dialects and repertoire sizes. Anim. Behav. 35, 395–401. Baker, M. C., Spitler‐Nabors, K. J., Thompson, A. D. Jr., and Cunningham, M. A. (1987b). Reproductive behaviour of female white‐crowned sparrows: Effect of dialects and synthetic hybrid songs. Anim. Behav. 35, 1766–1774. Ballentine, B., Hyman, J., and Nowicki, S. (2004). Vocal performance influences female response to male bird song: An experimental test. Behav. Ecol. 15, 163–168. Baptista, L. F. (1975). Song dialects and demes in sedentary populations of the white‐crowned sparrow (Zonotrichia leucophrys nuttalli). Univ. Calif. Pub. Zool. 105, 1–52. Baptista, L. F. (1977). Geographic variation in songs and dialects of the Puget Sound white‐ crowned sparrow. Condor 79, 356–370. Baptista, L. F. (1985). Bird‐song dialects: Social adaptation or assortative mating? Behav. Brain Sci. 8, 100–101. Baptista, L. F. (1990). Dialectal variation in the raincall of the chaffinch (Fringilla coelebs). Die Vogelwarte 35, 249–256. Baptista, L. F., and King, J. R. (1980). Geographical variation in song and song dialects of montane white‐crowned sparrows. Condor 82, 267–284. Baptista, L. F., and Morton, M. L. (1982). Song dialects and mate selection in montane white‐ crowned sparrows. Auk 99, 537–547. Baptista, L. F., and Morton, M. L. (1988). Song learning in montane white‐crowned sparrows: From whom and when. Anim. Behav. 36, 1753–1764. Baptista, L. F., and Petrinovich, L. (1984). Social interaction, sensitive phases and the song template hypothesis in the white‐crowned sparrow. Anim. Behav. 32, 172–181. Baptista, L. F., and Trail, P. W. (1992). The role of song in the evolution of passerine diversity. Syst. Biol. 41, 242–247. Baptista, L. F., Jesse, A., Bell, D. A., and Cebrian, C. (1997). Acquisition and recall of Gambel’s sparrow dialects by Nuttall’s white‐crowned sparrows in the wild. Wilson Bull. 109, 516–521. Becker, P. H. (1982). The coding of species‐specific characteristics in bird sounds. In ‘‘Acoustic Communication in Birds’’ (D. E. Kroodsma and E. H. Miller, Eds.), Vol. 1, pp. 213–252. Academic Press, New York. Beckers, G. J. L., Suthers, R. A., and ten Cate, C. (2003). Pure‐tone birdsong by resonance filtering of harmonic overtones. Proc. Natl. Acad. Sci. USA 100, 7372–7376. Beecher, M. D., and Brenowitz, E. A. (2005). Functional aspects of song learning in songbirds. Trends Ecol. Evol. 20, 143–149. Bertelli, S., and Tubaro, P. L. (2002). Body mass and habitat correlates of song structure in a primitive group of birds. Biol. J. Linn. Soc. 77, 423–430.

446

JEFFREY PODOS AND PAIGE S. WARREN

Bjerke, T. (1974). Geographic variation in the song of the redwing, Turdus iliacus. Sterna 13, 65–76. Bjerke, T. (1982). Origin of song traditions in the redwing Turdus iliacus: Evidence from field studies. Fauna Norvegica, Series C., Cinclus 5, 84–92. Bjerke, T. K. (1980). Continuity and change in dialect structure of the redwing Turdus iliacus. Fauna Norvegica Series C, Cinclus 3, 73–79. Bjerke, T. K. (1984). The response of male redwings Turdus iliacus to playback of different dialects. Fauna Norvegica, Series C, Cinclus 7, 24–27. Bjerke, T. K., and Bjerke, T. H. (1981). Song dialects in the redwing Turdus iliacus. Ornis Scandin. 12, 40–50. Blanckenhorn, W. U. (2000). The evolution of body size: What keeps organisms small? Q. Rev. Biol. 75, 385–407. Borror, D. J. (1961). Intraspecific variation in passerine bird songs. Wilson Bull. 73, 57–78. Borror, D. J. (1975). Songs of the rufous‐sided towhee. Condor 77, 183–1985. Boughman, J. W. (2001). Divergent sexual selection enhances reproductive isolation in sticklebacks. Nature 411, 944–948. Brenowitz, E. A. (1985). Bird‐song dialects: Filling in the gaps. Behav. Brain Sci. 8, 101–102. Briskie, J. V. (1999). Song variation and the structure of local song dialects in the polygynandrous Smith’s longspur. Can. J. Zool. 77, 1587–1594. Brooks, D. R., and McLennan, D. H. (1991). ‘‘Phylogeny, Ecology, and Behavior.’’ University of Chicago Press, Chicago. Brown, E. D., and Farabaugh, S. M. (1997). What birds with complex social relationships can tell us about vocal learning: Vocal sharing in avian groups. In ‘‘Social Influences on Vocal Development’’ (C. T. Snowden and M. Hausberger, Eds.), pp. 98–127. Cambridge University Press, Cambridge, UK. Buchanan, K. L., Catchpole, C. K., Lewis, J. W., and Lodge, A. (1999). Song as an indicator of parasitism in the sedge warbler. Anim. Behav. 57, 307–314. Buchanan, K. L., Spencer, K. A., Goldsmith, A. R., and Catchpole, C. K. (2003). Song as an honest signal of past developmental stress in the European starling (Sturnus vulgaris). Proc. Roy. Soc. B 270, 1149–1156. Buchanan, K. L., Leitner, S., Spencer, K. A., Goldsmith, A. R., and Catchpole, C. K. (2004). Developmental stress selectively affects the song control nucleus HVC in the zebra finch brain. Proc. Roy. Soc. Lond B. 271, 2381–2386. Butlin, R. K. (1995). Reinforcement: An idea evolving. Trends Ecol. Evol. 10, 432–434. Butlin, R. K., and Ritchie, M. G. (1994). ‘‘Behavior and speciation.’’ In Behavior and Evolution, (P. J. B. Slater and T. R. Halliday, Eds.), pp. 43–79. Cambridge University Press, Cambridge, UK. Carter, M. D. (1984). The Social organization and parasitic behavior of the bronzed cowbird in South Texas. PhD thesis. University of Minnesota, Minneapolis, MN. Catchpole, C. K. (1980). Sexual selection and the evolution of complex songs among European warblers of the genus Acrocephalus. Behaviour 74, 149–166. Catchpole, C. K., and McGregor, P. K. (1985). Sexual selection, song complexity and plumage dimorphism in European buntings of the genus Emberiza. Anim. Behav. 33, 1378–1380. Catchpole, C. K., and Rowell, A. (1993). Song sharing and local dialects in a population of the European Wren Troglodytes troglodytes. Behaviour 125, 67–78. Catchpole, C. K., and Slater, P. J. B. (1995). ‘‘Bird Song: Biological Themes and Variations.’’ Cambridge University Press, Cambridge, UK. Chilton, G., and Lein, M. R. (1996a). Long‐term changes in songs and song dialect boundaries of Puget Sound white‐crowned sparrows. Condor 98, 567–580.

GEOGRAPHIC VARIATION IN BIRDSONG

447

Chilton, G., and Lein, M. R. (1996b). Song repertoires of Puget Sound white‐crowned sparrows Zonotrichia leucophrys pugetensis. J. Avian Biol. 27, 31–40. Chilton, G., Baker, M. C., Barrentine, C. D., and Cunningham, M. C. (1995). White‐crowned sparrow (Zonotrichia leucophrys). In ‘‘The Birds of North America’’ (A. Poole and F. B. Gill, Eds.), pp. 1–28. Academy of Natural Sciences, Philadelphia, PA. Clotfelter, E. D. (1995). Courtship displaying and intrasexual competition in the bronzed cowbird. Condor 97, 816–818. Conrads, K. (1976). Studien an fremddialekt‐sa¨ngern und dialekt‐mischsa¨ngern des ortolans (Emberiza hortulana). J. Ornithol. 117, 438–450. Conrads, K., and Conrads, W. (1971). Regionaldialekte des ortolans (Emberiza hortulana) in Deutschland. Die Vogelwelt 92, 81–100. Cunningham, M. A., Baker, M. C., and Boardman, T. J. (1987). Microgeographic song variation in the Nuttall’s white‐crowned sparrow. Condor 89, 261–275. Cutler, B. (1970). ‘‘Anatomical studies of the syrinx of Darwin’s finches.’’ Master’s thesis. San Francisco State University, San Francisco. de Kort, S. R., and ten Cate, C. (2001). Response to interspecific vocalizations is affected by degree of phylogenetic relatedness in Streptopelia doves. Anim. Behav. 61, 239–247. de Kort, S. R., and ten Cate, C. (2004). Repeated decrease in vocal repertoire size in Streptopelia doves. Anim. Behav. 67, 549–557. DeWolfe, B. B., and Baptista, L. F. (1995). Singing behavior, song types on their wintering grounds and the question of leap‐frog migration in Puget Sound white‐crowned sparrows. Condor 97, 376–389. DeWolfe, B. B., Kaska, D. D., and Peyton, L. J. (1974). Prominent variations in the songs of Gambel’s white‐crowned sparrows. Bird Banding 45, 224–252. Dobzhansky, T. (1951). ‘‘Genetics and the Origin of Species.’’ Columbia University Press, New York. Dolbeer, R. A. (1982). Migration patterns for age and sex classes of blackbirds (Agelaius phoeniceus) and starlings (Sturnus vulgaris). J. Field Ornithol. 53, 28–46. Doutrelant, C., Leitao, A., Giorgo, M., and Lambrechts, M. (1999). Geographical variation in blue tit song, the result of an adjustment to vegetation type? Behaviour 136, 481–493. Draganoiu, T. I., Nagle, L., and Kreutzer, M. (2002). Directional female preference for an exaggerated male trait in canary (Serinus canaria) song. Proc. Roy. Soc. B 269, 2525–2531. Edwards, S. V., Kingan, S. B., Calkins, J. D., Balakrishnan, C. N., Jennings, W. B., Swanson, W. J., and Sorenson, M. D. (2005). Speciation in birds: Genes, geography, and sexual selection. Proc. Natl. Acad. Sci. USA 102, 6550–6557. Elemans, C. P. H., Spierts, I. L. Y., Mu¨ller, U. K., van Leeuwen, J. L., and Goller, F. (2004). Birdsong: Superfast muscles control dove’s trill. Nature 431, 146. Ellers, J., and Slabbekoorn, H. (2003). Song divergence and male dispersal among bird populations: A spatially explicit model testing the role of vocal learning. Anim. Behav. 65, 671–681. Espmark, Y. (1981). Song dialects in redwings Turdus iliacus. Var Fagelva¨rld 40, 81–90. Espmark, Y. (1982). Microgeographical and individual variation in the morphology of song of the redwing Turdus iliacus. Fauna Norvegica, Series C., Cinclus 5, 73–83. Falls, J. B. (1963). Properties of bird song eliciting responses from territorial males. Proc. Internat. Ornithol. Cong. 13, 259–271. Fee, M. S., Shraiman, B., Pesaran, B., and Mitra, P. P. (1998). The role of nonlinear dynamics of the syrinx in the vocalizations of a songbird. Nature 395, 67–71. Feekes, F. (1977). Colony‐specific song in Cacicus cela (Icteridae, Aves): The password hypothesis. Ardea 65, 197–202.

448

JEFFREY PODOS AND PAIGE S. WARREN

Ficken, M. S., Ficken, R. W., and Witkin, S. R. (1978). Vocal repertoire of the black‐capped chickadee. Auk 95, 34–48. Ficken, M. S., Ficken, R. W., and Apel, K. M. (1985). Dialects in a call associated with pair interactions in the black‐capped chickadee Parus atricapillus. Auk 102, 145–151. Fitzpatrick, J. W. (1988). Why so many passerine birds? A response to Raikow. Syst. Zool. 37, 71–76. Fleischer, R. C., and Rothstein, S. I. (1988). Known secondary contact and rapid gene flow among subspecies and dialects in the brown‐Headed cowbird. Evolution 42, 1146–1158. Fonstad, T., Espmark, Y., Lampe, H., and Bjerke, T. (1984). Breeding biology of the redwing Turdus iliacus in central Norway. Fauna Norvegica, Series C., Cinclus 7, 75–82. Ford, J. K. B. (1991). Vocal traditions among resident killer whales (Orcinus orca) in coastal waters of British Columbia [Canada]. Can. J. Zool. 69, 1454–1483. Foster, S. A. (1999). The geography of behaviour: An evolutionary perspective. Trends Ecol. Evol. 14, 190–195. Gil, D., and Gahr, M. (2002). The honesty of bird song: Multiple constraints for multiple traits. Trends Ecol. Evol. 17, 133–141. Gil, D., Naguib, M., Riebel, K., Rutstein, A., and Gahr, M. (2006). Early condition, song learning, and the volume of song brain nuclei in the zebra finch. J. Neurobiol. 66, 1602–1612. Gill, F. B., and Murray, B. G. (1972). Discrimination behavior and hybridization of the blue‐ winged and golden‐winged warblers. Evolution 26, 282–293. Glaubrecht, M. (1989). Zur geographischen variabilita¨t des gesangs der goldammer, Emberiza citrinella, im norddeutschen Dialekt‐Grenzgebiet. J. Ornithol. 130, 277–292. Glaubrecht, M. (1991). Gesangsvariation der goldammer (Emberiza citrinella) in Norddeutschland und auf den Da¨nischen Inseln. J. Ornithol. 132, 441–445. Goodfellow, D. J., and Slater, P. J. B. (1986). A model of bird song dialects. Anim. Behav. 34, 1579–1580. Gould, S. J., and Vrba, E. S. (1982). Exaptation: A missing term in the science of form. Paleobiology 8, 4–15. Grant, P. R., and Grant, B. R. (1995). Predicting microevolutionary responses to directional selection on heritable variation. Evolution 49, 241–251. Grant, P. R., and Grant, B. R. (1996). Cultural inheritance of song and its role in the evolution of Darwin’s finches. Evolution 50, 2471–2487. Grant, P. R., and Grant, B. R. (2002). Unpredictable evolution in a 30‐year study of Darwin’s finches. Science 296, 707–711. Grant, P. R., and Grant, B. R. (2006). Evolution of character displacement in Darwin’s finches. Science 313, 224–226. Gu¨ttinger, H. R. (1974). Gesang des Gru¨nlings (Chloris chloris). Lokale Unterscheide und Entwicklung bei Schallisolation. J. Ornithol. 115, 321–337. Gu¨ttinger, H. R. (1976). Variable and constant structures in greenfinch songs (Chloris chloris) in different locations. Behaviour 60, 304–318. Haavie, J., Borge, T., Bures, S., Garamszegi, L. Z., Lampe, H. M., Moreno, J., Qvarnstro¨m, A., To¨ro¨k, J., and Sætre, G. P. (2004). Flycatcher song in allopatry and sympatry—convergence, divergence, and reinforcement. J. Evol. Biol. 17, 227–237. Hafner, D. J., and Peterson, K. E. (1985). Song dialects and gene flow in the white‐crowned sparrow, Zonotrichia leucophrys nuttalli. Evolution 39, 687–694. Handford, P., and Lougheed, S. C. (1991). Variation in duration and frequency characteristics in the song of the rufous‐collared sparrow, Zonotrichia capensis, with respect to habitat, trill dialect and body size. Condor 93, 644–658.

GEOGRAPHIC VARIATION IN BIRDSONG

449

Handley, H. G., and Nelson, D. A. (2005). Ecological and phylogenetic effects of song sharing in songbirds. Ethology 111, 221–238. Hansen, P. (1979). Vocal learning: Its role in adapting sound structures to long‐distance propagation and a hypothesis on its evolution. Anim. Behav. 27, 1270–1271. Hansen, P. (1985). Geographic song variation in the yellowhammer (Emberiza citrinella). Natura Jutlandica 21, 209–219. Hansen, P. (1999). Long‐term stability of song elements in the yellowhammer Emberiza citrinella. Bioacoustics 9, 281–295. Harbison, H., Nelson, D. A., and Hahn, T. P. (1999). Long‐term persistence of song dialects in the mountain white‐crowned sparrow. Condor 101, 133–148. Harris, M. A., and Lemon, R. E. (1974). Songs of song sparrows: Reactions of males to songs of different localities. Condor 76, 33–44. Heinemann, D. (1981). Song dialects, migration, and population structure of Puget Sound white‐crowned sparrows. Auk 98, 512–521. Hoese, W. J., Podos, J., Boetticher, N. C., and Nowicki, S. (2000). Vocal tract function in birdsong production: Experimental manipulation of beak movements. J. Exp. Biol. 203, 1845–1855. Holland, J., McGregor Peter, K., and Rowe Candida, L. (1996). Changes in microgeographic song variation of the corn bunting Miliaria calandra. J. Avian Biol. 27, 47–55. Huber, S. K., and Podos, J. (2006). Beak morphology and song production covary in a population of Darwin’s finches. Biol. J. Linn. Soc. 88, 489–498. Hunter, L. M., and Krebs, J. R. (1979). Geographical variation in the song of the great tit (Parus major) in relation to ecological factors. J. Anim. Ecol. 48, 759–785. Hyman, J. (2002). Conditional strategies in territorial defense: Do carolina wrens play tit‐for‐ tat? Behav. Ecol. 13, 664–669. Illes, A. E., Hall, M. L., and Vehrencamp, S. L. (2006). Vocal performance influences male receiver response in the banded wren. Proc. Roy. Soc. B 273, 1907–1912. Irwin, D. E., and Price, T. (1999). Sexual imprinting, learning and speciation. Heredity 82, 347–354. Irwin, D. E., Bensch, S., and Price, T. D. (2001). Speciation in a ring. Nature 409, 333–337. Irwin, R. E. (1988). The evolutionary significance of behavioral development: The ontogeny and phylogeny of bird song. Anim. Behav. 36, 814–824. Irwin, R. E. (1990). Directional sexual selection cannot explain variation in song repertoire size in the New World blackbirds (Icterinae). Ethology 85, 212–224. Irwin, R. E. (1996). The phylogenetic content of avian courtship display and song evolution. In ‘‘Phylogenies and the Comparative Method in Animal Behavior’’ (E. P. Martins, Ed.), pp. 234–252. Oxford University Press, Oxford. Jenkins, P. F. (1978). Cultural transmission of song patterns and dialect development in a free‐ living bird population. Anim. Behav. 26, 50–78. Jenkins, P. F. (1985). Song learning, competition, and dialects. Behav. Brain Sci. 8, 108. Johannessen, L. E., Slagsvold, T., and Hansen, B. T. (2006). Effects of social rearing condition on song structure and repertoire size: Experimental evidence from the field. Anim. Behav. 72, 83–95. King, J. R. (1972). Variation in the song of the Rufous‐collared Sparrow, Zonotrichia capensis, in Northwestern Argentina. Z. Tierpsych. 30, 340–373. Kramer, H. G., and Thompson, N. S. (1979). Geographic variation in the bell calls of the blue jay (Cyanocitta cristata). Auk 96, 423–425. Krebs, J. R., and Kroodsma, D. E. (1980). Repertoires and geographical variation in birdsong. Adv. Study Behav. 11, 143–177.

450

JEFFREY PODOS AND PAIGE S. WARREN

Kreutzer, M. (1974). Ste´re´otypie et variations dans les chants de proclamation territoriale chez le troglodyte (Troglodytes troglodytes). Revue du Comportement Animal 8, 270–286. Kroodsma, D. E. (1972). Variations in songs of Vesper sparrows in Oregon. Wilson Bulletin 84, 173–178. Kroodsma, D. E. (1974). Song learning, dialects, and dispersal in the Bewick’s wren. Z. Tierpsychol. 35, 352–380. Kroodsma, D. E. (1977). Correlates of song organization among North American wrens. Am. Nat. 111, 995–1008. Kroodsma, D. E. (1985a). Limited dispersal between dialects? Hypotheses testable in the field. Behav. Brain Sci. 8, 108–109. Kroodsma, D. E. (1985b). Geographic variation in songs of the Bewick’s wren: A search for correlations with avifaunal complexity. Behav. Ecol. Sociobiol. 16, 143–150. Kroodsma, D. E. (1996). Ecology of passerine song development. In ‘‘Ecology and Evolution of Acoustic Communication in Birds’’ (D. E. Kroodsma and E. H. Miller, Eds.), pp. 3–19. Cornell University Press, Ithaca, NY. Kroodsma, D. E., and Byers, B. E. (1991). The function(s) of bird song. Am. Zool. 31, 318–328. Kroodsma, D. E., and Verner, J. (1978). Complex singing behaviours among Cistothorus wrens. Auk 95, 703–716. Kroodsma, D. E., Baker, M. C., Baptista, L. F., and Petrinovich, L. (1985). Vocal ‘‘dialects’’ in Nuttall’s white‐crowned sparrow. Curr. Ornithol. 2, 103–133. Kroodsma, D. E., Houlihan, P. W., Fallon, P. A., and Wells, J. A. (1997). Song development by grey catbirds. Anim. Behav. 54, 457–464. Kroodsma, D. E., Byers, B. E., Halkin, S. L., Hill, C., Minis, D., Bolsinger, J. R., Dawson, J.‐A., Donelan, E., Farrington, J., Gill, F. B., Houlihan, P., Innes, D., et al. (1999). Geographic variation in black‐capped chickadee songs and singing behavior. Auk 116, 387–402. Lachlan, R. F., and Servedio, M. R. (2004). Song learning accelerates allopatric speciation. Evolution 58, 2049–2063. Lachlan, R. F., and Slater, P. J. B. (2003). Song learning by chaffinches: How accurate, and from where? Anim. Behav. 65, 957–969. Lachlan, R. F., Janik, V., and Slater, P. J. B. (2004). The evolution of conformity‐enforcing behaviour in cultural communication systems. Anim. Behav. 68, 561–570. Lack, D. (1947). ‘‘Darwin’s Finches.’’ Cambridge University Press, Cambridge, UK. Laiolo, P., and Tella, J. L. (2005). Habitat fragmentation affects culture transmission: Patterns of song matching in Dupont’s lark. J. Appl. Ecol. 42, 1183–1193. Leader, N., Wright, J., and Yom‐Tov, Y. (2000). Microgeographic song dialects in the orange‐ tufted sunbird (Nectarinia osea). Behaviour 137, 1613–1627. Leader, N., Wright, J., and Yom‐Tov, Y. (2002). Dialect discrimination by male orange‐tufted sunbirds (Nectarinia osea): Reactions to own vs. neighbor dialects. Ethology 108, 367–376. Leader, N., Wright, J. W., and Yom‐Tov, Y. (2005). Acoustic properties of two urban song dialects in the orange‐tufted sunbird (Nectarinia osea). Auk 122, 231–245. Lemon, R. E. (1967). The response of cardinals to songs of different dialects. Anim. Behav. 15, 538–545. Lemon, R. E. (1975). How birds develop song dialects. Condor 77, 385–406. Lemon, R. E. (1985). Functional studies in bird song. Behav. Brain Sci. 8, 109–110. Liou, L. W., and Price, T. D. (1994). Speciation by reinforcement of premating isolation. Evolution 48, 1451–1459. Livingston, F. S., White, S. A., and Mooney, R. (2000). Slow NMDA‐EPSCs at synapses critical for song development are not required for song learning in zebra finches. Nat. Neurosciences 3, 482–488.

GEOGRAPHIC VARIATION IN BIRDSONG

451

Lougheed, S. C., and Handford, P. (1992). Vocal dialects and the structure of geographic variation in morphological and allozymic characters in the rufous‐collared sparrow Zonotrichia capensis. Evolution 46, 1443–1456. MacDonald, I. F., Kempster, B., Zanette, L., and MacDougall‐Shackleton, S. A. (2006). Early nutritional stress impairs development of a song‐control brain region in both male and female juvenile song sparrows (Melospiza melodia) at the onset of song learning. Proc. Roy. Soc. B 273, 2559–2564. MacDougall‐Shackleton, E. A., and MacDougall‐Shackleton, S. A. (2001). Cultural and genetic evolution in mountain white‐crowned sparrows: Song dialects are associated with population structure. Evolution 55, 2568–2575. MacDougall‐Shackleton, E. A., Derryberry, E. P., and Hahn, T. P. (2002). Nonlocal male mountain white‐crowned sparrows have lower paternity and higher parasite loads than males singing local dialect. Behav. Ecol. 13, 682–689. Marler, P. (1957). Specific distinctiveness in the communication signals of birds. Behaviour 11, 13–19. Marler, P. (1960). Bird song and mate selection. In ‘‘Animal Sounds and Communication’’ (W. E. Lanyon and W. N. Tavolga, Eds.), pp. 348–367. American Institute of Biological Science, Washington, DC. Marler, P., and Peters, S. (1987). A sensitive period for song acquisition in the song sparrow, Melospiza melodia: A case of age‐limited learning. Ethology 76, 89–100. Marler, P., and Peters, S. (1988). The role of song phonology and syntax in vocal learning preferences in the song sparrow, Melospiza melodia. Ethology 77, 125–149. Marler, P., and Tamura, M. (1962). Song ‘‘dialects’’ in three populations of white‐crowned sparrows. Condor 64, 368–377. Marler, P., and Tamura, M. (1964). Culturally transmitted patterns of vocal behavior in sparrows. Science 146, 1483–1486. Marler, P., Mundinger, P., Waser, M. S., and Lutjen, A. (1972). Effects of acoustical stimulation and deprivation on song development in red‐winged blackbirds (Agelaius phoeniceus). Anim. Behav. 20, 586–606. Martens, J., and Kessler, P. (2000). Territorial song and song neighbourhoods in scarlet rosefinch Carpodacus erythrinus. J. Avian Biol. 31, 399–411. Martin, P. R., and Martin, T. E. (2001). Behavioral interactions between coexisting species: Song playback experiments with wood warblers. Ecology 82, 207–218. Martins, E. P. (1996). ‘‘Phylogenies and the Comparative Method in Animal Behavior.’’ Oxford University Press, New York. Masataka, N. (1988). The response of red‐chested moustached tamarins to long calls from their natal and alien populations. Anim. Behav. 36, 55–61. Mayr, E. (1942). ‘‘Systematics and the Origin of Species.’’ Columbia University Press, New York. McGregor, P. K. (1980). Song dialects in the corn bunting (Emberiza calandra). Z. Tierpsychol. 54, 285–297. McGregor, P. K. (1983). The response of corn buntings to playback of dialects. Z. Tierpsychol. 62, 256–260. McGregor, P. K. (1985). Song dialects: What has to be explained, and with what? Behav. Brain Sci. 8, 110. McGregor, P. K., and Krebs, J. R. (1984). Song learning and deceptive mimicry. Anim. Behav. 32, 280–287. McGregor, P. K., and Thompson, D. B. A. (1988). Constancy and change in local dialects of the corn bunting. Ornis Scand. 19, 153–159.

452

JEFFREY PODOS AND PAIGE S. WARREN

McGregor, P. K., Walford, V. R., and Harper, D. G. C. (1988). Song inheritance and mating in a songbird with local dialects. Bioacoustics 1, 129–130. Milligan, M. M. (1966). Vocal responses of white‐crowned sparrows to recordings of their own and another species. Anim. Behav. 14, 356–361. Milligan, M. M., and Verner, J. (1971). Inter‐population song dialect discrimination in the white‐crowned sparrow. Condor 73, 208–213. Møller, A. P. (1982). Song dialects in a population of yellowhammers Emberiza citrinella in Denmark. Ornis Scand. 13, 239–246. Molles, L. E., and Vehrencamp, S. L. (2001). Songbird cheaters pay a retaliation cost: Evidence for auditory conventional signals. Proc. Roy. Soc. B 268, 2013–2019. Mork, K. (1974). Ringing results for redwing Turdus iliacus in Norway. Sterna 13, 77–107. Mundinger, P. C. (1975). Song dialects and colonization in the house finch, Carpodacus mexicanus, on the East Coast. Condor 77, 407–422. Mundinger, P. C. (1982). Microgeographic and macrogeographic variation in the acquired vocalizations of birds. In ‘‘Acoustic Communication in Birds’’ (D. E. Kroodsma and E. H. Miller, Eds.), Vol. 2, pp. 147–208. Academic Press, New York. Naguib, M., Hammerschmidt, K., and Wirth, J. (2001). Microgeographic variation, habitat effects and individual signature cues in calls of chiffchaffs Phylloscopus collybita canarensis. Ethology 107, 341–355. Nelson, D. A. (1998). Geographic variation in song of Gambel’s white‐crowned sprrow. Behaviour 135, 321–342. Nelson, D. A. (1999). Ecological influences on vocal development in the white‐crowned sparrow. Anim. Behav. 58, 21–36. Nelson, D. A. (2000). Song overproduction, selective attrition and song dialects in the white‐ crowned sparrow. Anim. Behav. 60, 887–898. Nelson, D. A., and Marler, P. (1990). The perception of birdsong and an ecological concept of signal space. In ‘‘Comparative Perception, Complex Signals’’ (W. C. Stebbins and M. A. Berkley, Eds.), Vol. 2, pp. 443–478. John Wiley and Sons, New York. Nelson, D. A., and Soha, J. A. (2004). Perception of geographical variation in song by male Puget Sound white‐crowned sparrows, Zonotrichia leucophrys pugetensis. Anim. Behav. 68, 395–405. Nelson, D. A., Marler, P., and Palleroni, A. (1995). A comparative approach to vocal learning: Intraspecific variation in the learning process. Anim. Behav. 50, 83–97. Nelson, D. A., Whaling, C., and Marler, P. (1996a). The capacity for song memorization varies in populations of the same species. Anim. Behav. 52, 379–387. Nelson, D. A., Marler, P., and Morton, M. L. (1996b). Overproduction in song development: An evolutionary correlate with migration. Anim. Behav. 51, 1127–1140. Nelson, D. A., Hallberg, K. I., and Soha, J. A. (2004). Cultural evolution of Puget sound white‐ crowned sparrow song dialects. Ethology 110, 879–908. Noor, M. A. F. (1999). Reinforcement and other consequences of sympatry. Heredity 83, 503–508. Nottebohm, F. (1969). The song of the chingolo, Zonotrichia capensis, in Argentina: Description and evaluation of a system of dialects. Condor 71, 299–315. Nottebohm, F. (1972). The origins of vocal learning. Am. Nat. 106, 116–140. Nottebohm, F. (1975). Continental patterns of song variability in Zonotrichia capensis: Some possible ecological correlates. Am. Nat. 109, 605–624. Nottebohm, F. (1985). Sound transmission, signal salience, and song dialects. Behav. Brain Sci. 8, 112–113. Nowicki, S. (1987). Vocal‐tract resonances in oscine bird sound production: Evidence from birdsongs in a helium atmosphere. Nature 325, 53–55.

GEOGRAPHIC VARIATION IN BIRDSONG

453

Nowicki, S., and Marler, P. (1988). How do birds sing? Music Percept. 5, 391–426. Nowicki, S., and Podos, J. (1993). Complexity, coupling, and contingency in the production of birdsong. In ‘‘Perspectives in Ethology, Vol. 10: Behavior and Evolution’’ (P. P. G. Bateson, P. Klopfer, and N. Thompson, Eds.), pp. 159–186. Plenum Press, New York. Nowicki, S., Westneat, M. W., and Hoese, W. J. (1992). Birdsong: Motor function and the evolution of communication. Sem. Neurosci. 4, 385–390. Nowicki, S., Peters, S., and Podos, J. (1998). Song learning, early nutrition, and sexual selection in songbirds. Am. Zool. 38, 179–190. Nowicki, S., Searcy, W. A., and Peters, S. (2002). Brain development, song learning and mate choice in birds: A review and experimental test of the ‘‘nutritional stress hypothesis.’’ J. Comp. Physiol. A. 188, 1003–1014. O’Loghlen, A. L. (1995). Delayed access to local songs prolongs vocal development in dialect populations of Brown‐headed Cowbirds. Condor 97, 402–414. O’Loghlen, A. L., and Rothstein, S. I. (1993). An extreme example of delayed vocal development: Song learning in a population of wild brown‐headed cowbirds. Anim. Behav. 46, 293–304. O’Loghlen, A. L., and Rothstein, S. I. (1995). Culturally correct song dialects are correlated with male age and female song preferences in wild populations of brown‐headed cowbirds. Behav. Ecol. Sociobiol. 36, 251–259. Orejuela, J. E., and Morton, M. L. (1975). Song dialects in several populations of mountain white‐crowned sparrows (Zonotrichia leucophrys oriantha) in the Sierra Nevada. Condor 77, 145–153. Orr, M. R., and Smith, T. B. (1998). Ecology and speciation. Trends Ecol. Evol. 13, 502–506. Panhuis, T. M., Butlin, R., Zuk, M., and Tregenza, T. (2001). Sexual selection and speciation. Trends Ecol. Evol. 16, 364–371. Patek, S. N., Baio, J. E., Fisher, B. L., and Suarez, A. V. (2006). Multifunctionality and mechanical origins: Ballistic jaw propulsion in trap‐jaw ants. Proc. Natl. Acad. Sci. USA 103, 12787–12792. Patten, M. A., Rotenberry, J. T., and Zuk, M. (2004). Habitat selection, acoustic adaptation, and the evolution of reproductive isolation. Evolution 58, 2114–2155. Payne, R. B. (1973). ‘‘Behavior, Mimetic Songs and Song Dialects and Relationships of the Parasitic Indigobirds (Vidua) of Africa.’’ American Ornithologists’ Union (Allen Press, Inc.), Lawrence Kansas. Payne, R. B. (1981). Population structure and social behavior: Models for testing the ecological significance of song dialects in birds. In ‘‘Natural Selection and Social Behavior: Recent Research and New Theory’’ (R. D. Alexander and D. W. Tinkle, Eds.), pp. 108–120. Chiron Press, New York. Payne, R. B. (1982). Ecological consequences of song matching: Breeding success and intraspecific song mimicry in indigo buntings. Ecology 63, 401–411. Payne, R. B. (1983). The social context of song mimicry: Song matching dialects in indigo buntings (Passerina cyanea). Anim. Behav. 31, 788–805. Payne, R. B. (1986). Bird songs and avian systematics. In ‘‘Current Ornithology’’ (R. J. Johnston, Ed.), pp. 87–126. Plenum Press, New York. Payne, R. B. (1987). Song dialects and neighborhood habitats in the indigobirds Vidua chalybeata and Vidua purpurascens at Lochinvar National Park, Zambia. J. Field Ornithol. 58, 152–170. Payne, R. B. (1996). Song traditions in indigo buntings: Origin, improvisatin, dispersal, and extinction in cultural evolution. In ‘‘Ecology and Evoluton of Acoustic Communication in Birds’’ (D. E. Kroodsma and E. H. Miller, Eds.), pp. 198–220. Cornell University Press, Ithaca, NY. Payne, R. B., and Payne, K. (1977). Social organization and mating success in local song populations of Village Indigobirds, Vidua chalybeata. Z. Tierpsychol. 45, 113–173.

454

JEFFREY PODOS AND PAIGE S. WARREN

Payne, R. B., and Payne, L. L. (1997). Field observations, experimental design, and the time and place of learning bird songs. In ‘‘Social Influences on Vocal Development’’ (C. T. Snowden and M. Hausberger, Eds.), pp. 57–84. Cambridge University Press, Cambridge, UK. Payne, R. B., and Westneat, D. F. (1988). A genetic and behavioral analysis of mate choice and song neighborhoods in indigo buntings. Evolution 42, 935–947. Payne, R. B., Payne, L. L., Woods, J. L., and Sorenson, M. D. (2000). Imprinting and the origin of parasite‐host species associations in brood‐parasitic indigobirds, Vidua chalybeata. Anim. Behav. 59, 69–81. Petren, K., Grant, P. R., Grant, B. R., and Keller, L. F. (2005). Comparative landscape genetics and the adaptive radiation of Darwin’s finches: The role of peripheral isolation. Mol. Ecol. 14, 2943–2957. Petrinovich, L., and Patterson, T. (1981). The responses of white‐crowned sparrows to songs of different dialects and subspecies. Z. Tierpsychol. 57, 1–14. Petrinovich, L., Patterson, T., and Baptista, L. F. (1981). Song dialects as barriers to dispersal: A re‐evaluation. Evolution 35, 180–188. Podos, J. (1996). Motor constraints on vocal development in a songbird. Anim. Behav. 51, 1061–1070. Podos, J. (1997). A performance constraint on the evolution of trilled vocalizations in a songbird family (Passeriformes: Emberizidae). Evolution 51, 537–551. Podos, J. (2001). Correlated evolution of morphology and vocal signal structure in Darwin’s finches. Nature 409, 185–188. Podos, J. (2007). Discrimination of geographical song variants by Darwin’s finches. Anim. Behav. 73, 833–844. Podos, J., and Hendry, A. P. (2006). The biomechanics of ecological speciation. In ‘‘Ecology and Biomechanics: A Mechanical Approach to the Ecology of Animals and Plants’’ (A. Herrel, T. Speck, and N. Rowe, Eds.), pp. 301–321. CRC Press, Boca Raton. Podos, J., and Nowicki, S. (2004a). Performance limits on birdsong. In ‘‘Nature’s Music: The Science of Birdsong’’ (P. Marler and H. Slabbekoorn, Eds.), pp. 318–342. Elsevier Academic Press, San Diego, CA. Podos, J., and Nowicki, S. (2004b). Beaks, adaptation, and vocal evolution in Darwin’s finches. Bioscience 54, 501–510. Podos, J., Nowicki, S., and Peters, S. (1999). Permissiveness in the learning and development of song syntax in swamp sparrows. Anim. Behav. 58, 93–103. Podos, J., Southall, J. A., and Rossi‐Santos, M. R. (2004a). Vocal mechanics in Darwin’s finches: Correlation of beak gape and song frequency. J. Exp. Biol. 207, 607–619. Podos, J., Huber, S. K., and Taft, B. (2004b). Bird song: The interface of evolution and mechanism. Annu. Rev. Ecol. Evol. Syst. 35, 55–87. Powys, V. (1995). Regional variation in the territorial songs of superb lyrebirds in the central tablelands of New South Wales. Emu 95, 280–289. Price, J. J., and Lanyon, S. M. (2002). Reconstructing the evolution of complex bird song in the oropendolas. Evolution 56, 1514–1529. Price, J. J., and Lanyon, S. M. (2004). Patterns of song evolution and sexual selection in the oropendolas and caciques. Behav. Ecol. 15, 485–497. Ptacek, M. B. (2000). The role of mating preferences in shaping interspecific divergence in mating signals in vertebrates. Behav. Processes 51, 111–134. Ratcliffe, L. M. (1981). Species recognition in Darwin’s ground finches (Geospiza Gould) PhD thesis. McGill University, Montreal Canada. Ratcliffe, L. M., and Grant, P. R. (1985). Species recognition in Darwin’s finches (Geospiza Gould). III. Male responses to playback of different song types, dialects, and heterospecific songs. Anim. Behav. 33, 290–307.

GEOGRAPHIC VARIATION IN BIRDSONG

455

Read, A. F., and Weary, D. M. (1992). The evolution of bird song: Comparative analyses. Phil. Trans. R. Soc. London B. 338, 165–187. Rice, W. R., and Hostert, E. E. (1993). Laboratory experiments on speciation: What have we learned in 40 years? Evolution 47, 1637–1653. Rich, T. (1981). Microgeographic variation in the song of the sage sparrow. Condor 83, 113–119. Riebel, K. (2003). The ‘‘mute’’ sex revisited: Vocal production and perception learning in female songbirds. Adv. Study Behav. 33, 49–86. Riede, T., Suthers, R. A., Fletcher, N. H., and Blevins, W. E. (2006). Songbirds tune their vocal tract to the fundamental frequency of their song. Proc. Natl. Acad. Sci. USA 103, 5543–5548. Robinson, F. N., and Curtis, H. S. (1996). The vocal displays of the lyrebirds (Menuridae). Emu 96, 258–275. Roff, D. A. (1992). ‘‘The Evolution of Life Histories: Theory and Analysis.’’ Chapman and Hall, New York. Rothstein, S. I. (1980). The preening invitation or head‐down display of parasitic cowbirds: 2. Experimental analysis and evidence for behavioral mimicry. Behaviour 75, 148–184. Rothstein, S. I., and Fleischer, R. C. (1987). Vocal dialects and their possible relation to honest status signaling in the brown‐headed cowbird. Condor 89, 1–23. Rothstein, S. I., Verner, J., and Farmer, C. (1999). Cowbird vocalizations: An overview and the use of playbacks to enhance cowbird detestability. In ‘‘Ecology and Management of Cowbirds and Their Hosts’’ (J. N. M. Smith, T. L. Cook, S. K. Robinson, S. I. Rothstein, and S. G. Sealy, Eds.), pp. 69–81. University of Texas Press, Austin Texas. Ruegg, K., Slabbekoorn, H., Clegg, S., and Smith, T. B. (2006). Divergence in mating signals correlates with ecological variation in the migratory songbird, Swainson’s thrush (Catharus ustulatus). Mol. Ecol. 15, 3147–3156. Rundle, H. D., Nagel, L., Boughman, J. W., and Schluter, D. (2000). Natural selection and parallel speciation in sympatric sticklebacks. Science 287, 306–308. Rutkowska‐Guz, J. M., and Osiejuk, T. S. (2004). Song structure and variation in yellowhammers Emberiza citrinella from western Poland. Polish J. Ecol. 52, 333–345. Ryan, M. J. (1986). Neuroanatomy influences speciation rates among anurans. Proc. Natl. Acad. Sci. USA 83, 1379–1382. Ryan, M. J., and Brenowitz, E. A. (1985). The role of body size, phylogeny, and ambient noise in the evolution of bird song. Am. Nat. 126, 87–100. Saetre, G. P., Moum, T., Bures, S., Kral, M., Adamjan, M., and Moreno, J. (1997). A sexually selected character displacement in flycatchers reinforces premating isolation. Nature 387, 589–592. Schluter, D. (2000). ‘‘ The Ecology of Adaptive Radiation.’’ Oxford University Press, Oxford. Schluter, D. (2001). Ecology and the origin of species. Trends Ecol. Evol. 16, 372–380. Searcy, W. A. (1992). Measuring responses of female birds to male song. In ‘‘Playback and Studies of Animal Communication’’ (P. K. McGregor, Ed.), pp. 175–189. Plenum, New York. Searcy, W. A., and Andersson, M. (1986). Sexual selection and the evolution of song. Annu. Rev. Ecol. Syst. 17, 507–533. Searcy, W. A., and Yasukawa, K. (1996). Song and female choice. In ‘‘Ecology and Evolution of Acoustic Communication in Birds’’ (D. E. Kroodsma and E. H. Miller, Eds.), pp. 454–473. Cornell University Press, Ithaca, NY. Searcy, W. A., Nowicki, S., and Hughes, M. (1997). The response of male and female song sparrows to geographic variation in song. Condor 99, 651–657. Searcy, W. A., Nowicki, S., Hughes, M., and Peters, S. (2002). Geographic song discrimination in relation to dispersal distances in song sparrows. Am. Nat. 159, 221–230.

456

JEFFREY PODOS AND PAIGE S. WARREN

Seddon, N. (2005). Ecological adaptation and species recognition drives vocal evolution in neotropical suboscine birds. Evolution 59, 200–215. Seitz, A., Reh, W., Veith, M., and Wolfes, R. (1994). Gene flow between and within natural vertebrate populations. Anim. Biotech. 5, 155–168. Sibley, C. G., and Monroe, B. L. J. (1990). ‘‘Distribution and Taxonomy of the Birds of the World.’’ Yale University Press, New Haven, CT. ¨ ber die Dialektbildung beim ‘‘Regenruf’’ des Buchfinken. J. Ornithol. 87, Sick, H. (1939). U 568–592. Slabbekoorn, H. (2004). Singing in the natural environment: The ecology of birdsong. In ‘‘Nature’s Music: The Science of Birdsong’’ (P. Marler and H. Slabbekoorn, Eds.). Elsevier Academic Press, San Diego, CA. Slabbekoorn, H., and Smith, T. B. (2002a). Bird song, ecology and speciation. Phil. Trans. R. Soc. London B 357, 493–503. Slabbekoorn, H., and Smith, T. B. (2002b). Habitat‐dependent song divergence in the little greenbul: An analysis of environmental selection pressures on acoustic signals. Evolution 56, 1849–1858. Slabbekoorn, H., de Kort, S., and ten Cate, C. (1999). Comparative analysis of perch coo vocalizations in Streptopelia doves. Auk 116, 737–748. Slater, P. J. B. (1985). White rats and general theories. Behav. Brain Sci. 8, 115–116. Slater, P. J. B. (1986). The cultural transmission of bird song. Trends Ecol. Evol. 1, 94–97. Slater, P. J. B. (1989). Bird song learning: Causes and consequences. Ethol. Ecol. Evol. 1, 19–46. Slater, P. J. B., Clements, F. A., and Goodfellow, D. J. (1984). Local and regional variations in chaffinch song and the question of dialects. Behaviour 88, 76–97. Soha, J. A., Nelson, D. A., and Parker, P. G. (2004). Genetic analysis of song dialect populations in Puget Sound white‐crowned sparrows. Behav. Ecol. 15, 636–646. Sorjonen, J. (2001). Long‐term constancy of two rain‐call dialects of the Chaffinch Fringilla coelebs in Finnish and Russian Karelia: A consequence of site‐fidelity? Ornis Fennica 78, 73–82. Spencer, K. A., Buchanan, K. L., Goldsmith, A. R., and Catchpole, C. K. (2003). Song as an honest signal of developmental stress in the zebra finch (Taeniopygia guttata). Horm. Behav. 44, 132–139. Spencer, K. A., Buchanan, K. L., Goldsmith, A. R., and Catchpole, C. K. (2004). Developmental stress, social rank and song complexity in the European starling (Sturnus vulgaris). Proc. Roy. Soc. B 271, S121–S123. Spencer, K. A., Buchanan, K. L., Leitner, S., Goldsmith, A. R., and Catchpole, C. K. (2005). Parasites affect song complexity and neural development in a songbird. Proc. Roy. Soc. B 272, 2037–2043. Suthers, R. A. (2004). How birds sing and why it matters. In ‘‘Nature’s Music: The Science of Birdsong’’ (P. Marler and H. Slabbekoorn, Eds.), pp. 272–295. Elsevier Academic Press, San Diego, CA. Suthers, R. A., and Goller, F. (1997). Motor correlates of vocal diversity in songbirds. In ‘‘Current Ornithology’’ (V. Nolan Jr., E. D. Ketterson, and C. F. Thompson, Eds.), pp. 235–288. Plenum Press, New York. Suthers, R. A., Goller, F., and Pytte, C. (1999). The neuromuscular control of birdsong. Phil. Trans. R. Soc. London B 354, 927–939. Sze´kely, T., Catchpole, C. K., DeVoogd, A., Marchl, Z., and DeVoogd, T. J. (1996). Evolutionary changes in a song control area of the brain (HVC) are associated with evolutionary changes in song repertoire among European warblers (Sylviidae). Proc. Roy. Soc. London B 263, 607–610.

GEOGRAPHIC VARIATION IN BIRDSONG

457

Teather, K. L., and Robertson, R. J. (1986). Pair bonds and factors influencing the diversity of mating systems in brown‐headed cowbirds [Molothrus ater]. Condor 88, 63–69. ten Cate, C. (2000). How learning mechanisms might affect evolutionary processes. Trends Ecol. Evol. 15, 179–181. ten Cate, C. (2004). Birdsong and Evolution. In ‘‘Nature’s Music: The Science of Birdsong’’ (P. Marler and H. Slabbekoorn, Eds.), pp. 296–317. Elsevier Academic Press, San Diego, CA. Thielcke, G. (1969). Geographic variation in bird vocalizations. In ‘‘Bird Vocalizations: Their Relations to Current Problems in Biology and Psychology’’ (R. A. Hinde, Ed.), pp. 311–339. Cambridge University Press, Cambridge, UK. Thielcke, G. (1986). Constant proportions of mixed singers in tree creeper populations (Certhia familiaris). Ethology 72, 154–164. Thielcke, G. (1987). Long‐term consistency of short‐toed tree creeper dialect Certhia brachydactyla. J. Ornithol. 128, 171–180. Thielcke, G. (1988a). Buchfinken (Fringilla coelebs) eliminieren erlernte gesa¨nge vom baumpieper (Anthus trivialis). Vogelwarte 34, 319–336. Thielcke, G. (1988b). Neue befunde besta¨en Baron PERNAUs (1660–1731) angaben u¨ber lauta¨u ßerungen des buchfinken (Figilla coelebs). J. Ornithol. 129, 55–70. Thielcke, G. (1989). Experiments on sensitive phases and song variation in the chaffinch Fringilla coelebs. J. Ornithol. 130, 435–454. Thielcke, G., and Wu¨sternberg, K. (1985). Experiments on the origin of dialects in the short‐ toed tree creeper Certhia brachydactyla. Behav. Ecol. Sociobiol. 16, 195–201. Todt, D., and Naguib, M. (2000). Vocal interactions in birds: The use of song as a model in communication. Adv. Study Behav. 29, 247–296. Trainer, J. M. (1983). Changes in song dialect distributions and microgeographic variation in song of white‐crowned sparrows Zonotrichia leucophrys nuttalli. Auk 100, 568–582. Trainer, J. M. (1988). Singing organization during aggressive interactions among male yellow‐ rumped caciques. Condor 90, 681–688. Trainer, J. M. (1989). Cultural evolution in song dialects of yellow‐rumped caciques in Panama. Ethology 80, 190–204. Trainer, J. M., and Parsons, R. J. (2002). Delayed vocal maturation in polygynous Yellow‐ rumped Caciques. Wilson Bull. 114, 249–254. Trainer, J. M., and Peltz, B. S. (1995). Song repertoire of the bobolink: A reassessment. Ethology 102, 50–62. Tubaro, P. L., and Segura, E. T. (1994). Dialect differences in the song of Zonotrichia capensis in the southern Pampas: A test of the acoustic adaptation hypothesis. Condor 96, 1084–1088. Van Buskirk, J. (1997). Independent evolution of song structure and note structure in American wood warblers. Proc. Roy. Soc. London B 264, 755–761. Vermeij, G. J. (1988). The evolutionary success of passerines: A question of semantics? Syst. Zool. 37, 69–71. Ward, R. (1966). Regional variation in the song of the carolina chickadee. Living Bird 5, 127–150. Ward, W. V. (1964). The songs of the Apapane. Living Bird 3, 97–117. Warren, P. S. (2000). Song dialects in the bronzed cowbird: A model system for testing evolutionary hypotheses. In ‘‘Integrative Biology,’’ p. 267. University of Texas, Austin, TX. Warren, P. S. (2002). Geographic variation and dialects in songs of the bronzed cowbird (Molothrus aeneus). Auk 119, 349–361.

458

JEFFREY PODOS AND PAIGE S. WARREN

Warren, P. S. (2003). Winter dialects in the bronzed cowbird and their relationship to breeding‐ season dialects. Anim. Behav. 65, 1169–1178. Waser, P. M. (1985). Are dialects epiphenomena? Behav. Brain Sci. 8, 117. Weatherhead, P. J., and Forbes, M. R. L. (1994). ‘‘Natal philopatry in passerine birds: Genetic or ecological influences?’’ Behavioral Ecology 5, 426–433. Weilgart, L., and Whitehead, H. (1997). Group‐specific dialects and geographical variation in coda repertoire in South Pacific sperm whales. Behav. Ecol. Sociobiol. 40, 277–285. Wells, M. M., and Henry, C. S. (1998). Songs, reproductive isolation, and speciation in cryptic species of insects: A case study using green lacewings. In ‘‘Endless Forms: Species and Speciation’’ (D. J. Howard and S. B. Berlocher, Eds.), pp. 217–233. Oxford University Press, New York, NY. West‐Eberhard, M. J. (1983). Sexual selection, social competition, and speciation. Q. Rev. Biology 58, 155–183. Westneat, M. W., Long, J. H. Jr., Hoese, W., and Nowicki, S. (1993). Kinematics of birdsong: Functional correlation of cranial movements and acoustic features in sparrows. J. Exp. Biol. 182, 147–171. White, F. W. G. (1985). Microgeographic variation in the songs of the olive whistler in Kosciusko national park. Emu 85, 181–187. White, F. W. G. (1986). Summer‐winter movements of olive whistlers Pachycephala olivacea in the Snowy mountains, New South Wales, Australia. Corella 10, 125–126. White, F. W. G. (1987). Macrogeographic variation in the olive whistler in Australia. Emu 87, 14–25. Wiens, J. A. (1982). Song pattern variation in the sage sparrow (Amphispiza belli): Dialects or epiphenomena? Auk 99, 208–229. Wiley, R. H. (1971). Song groups in a singing assembly of little hermits. Condor 73, 28–35. Wiley, R. H. (1991). Associations of song properties with habitats for territorial oscine birds of eastern North America. Am. Nat. 138, 973–993. Wiley, R. H., and Richards, D. G. (1978). Physical constraints on acoustic communication in the atmosphere: Implications for the evolution of animal vocalizations. Behav. Ecol. Sociobiol. 3, 69–94. Williams, J. M., and Slater, P. J. B. (1990). Modeling bird song dialects: The influence of repertoire size and numbers of neighbours. J. Theor. Biol. 145, 487–496. Wright, T. F. (1996). Regional dialects in the contact call of a parrot. Proc. Roy. Soc. London B 263, 867–872. Wright, T. F., and Wilkinson, G. S. (2001). Population genetic structure and vocal dialects in an amazon parrot. Proc. Roy. Soc. London B 268, 609–616. Wright, T. F., Rodriguez, A. M., and Fleischer, R. C. (2005). Vocal dialects, sex‐biased dispersal, and microsatellite population structure in the parrot Amazona auropalliata. Mol. Ecol. 14, 1197–1205. Yokel, D. A. (1986). Monogamy and brood parasitism: An unlikely pair. Anim. Behav. 34, 1348–1358. Zink, R. M., and Barrowclough, G. F. (1984). Allozymes and song dialects: A reassessment. Evolution 38, 444–448. Zink, R. M., Dittmann, D. L., and Rootes, W. L. (1991). Mitochondrial DNA variation and the phylogeny of Zonotrichia. Auk 108, 578–584.

Index

A Achaearanea disparata, 85–87, 118 Achaearanea tepidariorum, 118 Achaearanea tesselata, 118 Achaearanea vervoorti, 85–87, 113, 118 Achaearanea wau, 85–87, 99, 100, 104–108, 111–113, 118 division of labor in, 102 Acoustic adaptation hypothesis, 378 Acoustic habitat, 376, 387 Acrocephalus arundinaceus, 150, 207 Acrocephalus australis, 150 Acrocephalus schoenobaenus, 216 Acrocephalus sechellensis, 205, see also Seychelles warbler Actophilornis africanus, 284, 311 Adaptive sex ratio control concept, 148 Aebutina binotata, 85–87, 102, 105–106 division of labor in, 102 Aegithalos caudatus, 235 Aethia cristatella, 311 Aethia pusilla, 311 Aethia pygmaea, 311 Agelena aperta, 109 Agelena consociata, 85–87, 100, 106, 112–113, 128 Agelenidae, 85–87, 90–91, 99, 109 Aggression in spiders, 97 random acts, 59 Aggressive behavior and heterozygosity, 215 Agile antechinus, 246 Alarm calls, see also Marmot alarm communication before and after pup emergence, 384

directed to predators and conspecifics, 372, 374 in rodents, 374 indicator of stress, 392–393 individuality and reliability of, 388–391 repertoire size, factors influencing, 380 acoustic environment, 378 home range size, 379 sociality, 379 signaler and receiver, 389 Alca torda, 311 Alcidae, 284 Alle alle, 287, 311 Allogrooming, in female-bonded primate groups, 45, 54–55 Allozymes, 199–200, 202–203 Alpine marmots, 239, 241, 259, 375 Alpine newt, 245 Altruism, 1, 3, 14, 16–17 Altruistic inbreeding, 221–222 Amaurobiidae, 90–91 Amaurobioidea, 99 Amaurobius ferox, 90–91, 129 Ameridion cf. petrum, 118 Ameridion sp., 118 Anarhynchus frontalis, 311 Anas platyrhynchos, 248 Anastrepha suspensa, 350 Anelosimus analyticus, 118 Anelosimus arizona, 90–91, 115–118, 121, 128–129, 130 Anelosimus baeza, 90–91, 118 Anelosimus biglebowski, 118 Anelosimus cf. studiosus, 90, 92, 94, 98, 117, 129, 132, 134 459

460

INDEX

Anelosimus chickeringi, 118 Anelosimus domingo, 85–87, 107, 113, 118 Anelosimus dubiosus, 85–87, 102, 118 Anelosimus dude, 118 Anelosimus elegans, 118 Anelosimus ethicus, 118 Anelosimus eximius, 85–87, 95, 100–104, 106–107, 110–113, 115, 118, 124 Anelosimus fratemus, 118 Anelosimus guacamayos, 85–87, 118 Anelosimus inhandava, 118 Anelosimus jabaquara, 90, 92, 94, 98, 104, 118, 132, 134 Anelosimus jucundus, 90, 92, 118, 129 Anelosimus kohi, 118 Anelosimus lorenzo, 85–87, 118 Anelosimus may, 118 Anelosimus misiones, 118 Anelosimus nelsoni, 118 Anelosimus nigrescens, 118 Anelosimus octavius, 118 Anelosimus ontoyacu, 118 Anelosimus oritoyacu, 85–87 Anelosimus pacificus, 115–118 Anelosimus pantanal, 118 Anelosimus pulchellus, 118 Anelosimus puravida, 85–87, 118 Anelosimus rabus, 118 Anelosimus republicana, 85–87 Anelosimus rupununi, 85–87, 102, 118 Anelosimus sallee, 118 Anelosimus sp., 98–117, 118 Anelosimus studiosus, 96, 98, 106, 109, 117–118, 126, 129–130, 132–134 Anelosimus tosum, 118 Anelosimus tungurahua, 118 Anelosimus vittatus, 118 Anous minutus, 311 Anous stolidus, 311 Anous tenuirostris, 311 Antechinus agilis, 252 Aphriza virgata, 311 Arctocephalus gazella, 256 Arenaria interpres, 301, 311 Arenaria melanocephala, 311

Argyrodes projiciens, 106 Argyrodinae, 106 Ateles, 48–49 Atlantic salmon, 232, 236, 256–257 Attagis gayi, 312 Attenuation and degradation, 376 Australian sleepy lizard, 231 Avian vocal evolution geographic variation in, 440 historical signal in, 434 multiple functions and trade-offs in, 434–435, 439 phylogenetic signal in, 434 study of, 433 Aviles model, 124

B Banner-tailed kangaroo rat, 233, 236 Barn swallow, 230, 240 Bartramia longicauda, 312 Battle of the Sexes game, 5–6 Behavioral traits of spiders, 89–99 aggression, 97 colony composition, 96 colony foundation, 96 group activities, 96 mating behavior, 97 sex ratio, 94 Belding’s ground squirrels, 383 Benefits of philopatry hypothesis, 154–155 Bicyclus anynana, 215, 216, 219 Bilingualism, 424, 430 Biological markets (BM), theory of, 54 Bird vocal evolution, see Avian vocal evolution Birdsong and birdcalls, 376 Berkeley population of, 406 functions of, 435 geographic variation in, evolution of (see Vocal geographic evolution) Black-capped marmot, 378 Black-tailed prairie dogs, 383

INDEX

Black-throated blue warbler, 238, 240–241, 257, 259 Blue tit, 230, 238, 240–241, 245, 258–259 Bluegill sunfish, 242 Bonds serviced by grooming model, 73 Brachyramphus brevirostris, 312 Brachyramphus marmoratus, 312 Breeding systems and sexual conflict framework for, 307–308 in Shorebirds, 279–281, 283–288, 304–308, 310 Brown skua, 287 Buff-breasted sandpiper, 287, 289 Bufo calamita, 218 Burhinidae, 284 Burhinus spp., 312 Buteo galapagoensis, 213 By-product mutualism, 1, 34 in spiders, 127–128

C Calidris acuminata, 285, 312 Calidris alba, 301, 312 Calidris alpina, 285, 312 Calidris bairdii, 312 Calidris fuscicollis, 301, 313 Calidris maritima, 287, 301, 313 Calidris mauri, 287, 294, 297, 313 Calidris melanotos, 284, 313 Calidris minuta, 301, 313 Calidris spp., 313 Calidris temminckii, 305, 313 Callithrix jacchus, 27 Carpodacus mexicanus, 214 Catharacta antarctica, 287 Catharacta maccormickii, 287, 313 Catharacta skua, 313 Catoptrophorus semipalmatus, 314 Cebus apella, 10, 54 Cebus capucinus, 10, 49, 54 Cebus olivaceous, 54 Cebus sp., 8, 48 Cepphus carbo, 314 Cepphus columba, 314

461

Cepphus grylle, 314 Ceratitis capitata, 350 Cercopithecus, 48 Cercopithecus mitis, 49 Cerorhinca monocerata, 314 Cervus elaphus, 212 Charadriidae, 284 Charadriiformes, see Shorebirds Charadrius alexandrinus, 287, 297, 314 Charadrius hiaticula, 287, 301, 314 Charadrius montanus, 286, 314 Charadrius morinellus, 287 Charadrius rubricollis, 300, 315 Charadrius semipalmatus, 285, 287, 315 Charadrius spp., 314–315 Chase-away model, 282, 309 Chicken game, 6–7 Chionidae, 284 Chionis alba, 315 Chionis minor, 315 Chlidonias albostriatus, 315 Chlidonias hybridus, 315 Chlidonias leucopterus, 315 Chlidonias niger, 315 Chlorocebus, 48 Chlorocebus aethiops, 8 Chrysso cf. nigriceps, 118 Cladorhynchus leucocephalus, 315 Coal tit, 238, 240 Coalitions, in female-bonded primate groups, 46–51, 74–75 Coarse-scaled habitats, 345 Cockerels, aerial alarm calls of, 387–388 Coelotes terrestris, 90–91, 129, 133 Coenocorypha aucklandica, 315 Coenocorypha pusilla, 316 Coleosoma floridanum, 118 Colobus Cercocebus, 48 Colonial species of spiders, 83–84 Colony composition in spiders, 94 definition of, 84 foundation in spiders, 94, 101–105 Common gull, 287 Common mole-rat, 240–241, 259 Common murre, 287

462

INDEX

Common sandpiper, 238, 287 Common tern, 287 Compatible allele, mate choice for, 194–195, 198–199 Competitive ability, of males and females, heterozygosity and fitness effects on, 215–216 Complementary genes, 194 Conflict over mating among shorebirds, 288–297 infanticide, 293–294, 310 multiple and extra-pair paternity, 286, 290–293 multiple mating, 288–290 optima, 281–282, 288–294 Conflict resolution, 56, 58 Conservation behavior, 392 Conspecific versus heterospecific song, 412 Conspecific warning function, 374, 381 Contact-sitting, 56 Contagious calling, 382 Contingent reciprocity, in primates evidence of, 17–18 experimental settings, 8 laboratory experiments, 10–17 limitations preclude deployment of, 18 naturalistic experiments, 8–10, 34 nonexperimental settings, 7–8 Cooperation among nonrelatives in primates, 2 and competition in spiders, 99–102 definition of, 2–3 evolution of, 1 game theory models of forms, 4–7 Battle of the Sexes, 5–6 Chicken game, 6–7, 31–32 Iterated Prisoner’s Dilemma, 4, 7–18 Stag Hunt, 5, 18–31 Cooperative breeding, 149 among Seychelles warbler, 151, 153–166, 176 direct benefits, 163–166

in oscine passerine species, 153 indirect benefits, 156–158 kin discrimination by subordinates, 157–159 kin selection mechanism, 159–163 system, 153–154 Cooperative coalition, in femalebonded primate groups, 45 Cooperative species, of spiders, 83 Copula duration, 348 and male body size, 351 female accessory gland during, role of, 356 male and female anatomy during, 355 Copulation partner choice, heterozygosity and mate choice evidence for, 191, 229–234 Cordylochernes scorpiodes, 225 Coturnix japonica, 251 Courtship behavior and heterozygosity, 215–216 Cricket, 245 Crocidura russula, 229 Crocuta crocuta, 33 Cross-fostering experiment, among Seychelles warbler, 161–162 Cryptic choice, 191 Cryptic female choice ‘‘mate-now-choose-later’’ mechanism for, 345 at PGM locus, 356–358 field experiments on, 360 in S. stercoraria (L.), 349–352 Cryptomys hottentotus hottentotus, 243 Ctenophorus ornatus, 212 Cunningham’s skink, 232, 236 Cursorius coromandelicus, 316 Cursorius cursor, 316 Cursorius rufus, 316 Cursorius temminckii, 316 Cyanistes caeruleus, 209 Cyclorhynchus psittacula, 316 Cyrtophora moluccensis, 112

463

INDEX

D Delena cancerides, 84, 90–91, 97–98, 100, 130, 132–133 Dendroica caerulescens, 260 Desertion, and parental care in shorebirds, 302–307 Desidae, 90–91, 99 Diaea ergandros, 90, 93, 99, 110, 121, 129 Diaea megagyna, 85–87, 97, 98 Diaea socialis, 85–87, 97, 100, 104, 107, 129 Dialect boundaries, temporal stability of, 421, 430–431 Dialect species, vocal and ecological parameters of, 422–424, 432 Dialect systems, diversity of, 431 Dialects characteristics, 421 and repertoire size, 429, 432 ecological correlates for, 424, 427– 428, 432 in passerine birds, 425–427 variation in, 427 vocal correlates for, 424, 428–429, 432 definition of, 409, 416 discriminant function analysis of, 430 evolution, 405 assessing hypotheses of, 416 by-product hypotheses for, 415 colony password hypothesis for, 413 deceptive mimicry hypothesis for, 413 epiphenomenon hypothesis for, 414–416 honest convergence hypothesis for, 413 in Z. leucophrys nuttalli, 407 local adaptation hypothesis for, 410–413, 431 social adaptation hypotheses for, 413–414 in Fringillids, 426

in predispersal and postdispersal learners, 431 literature survey, 421 scale of, 421, 430 sedentary species with, 428 size, 414, 432 species reported to exhibit, 416–420 variation on female preferences, 412 Dictynidae, 85–87 Differential mating opportunities, and parental care in shorebirds, 303– 304 Diplosoma listerianum, 245 Dipodomys spectabilis, 236 Diurnality, evolution of, 374 Division of labor, in spiders, 100–101 Divorce and second broods, heterozygosity and mate choice evidence for, 234–235 Dominance and heterozygosity, 215 Dromadidae, 284 Dromas ardeola, 316 Drosophila melanogaster, 281, 288, 363 Drosophila montana, 216 Drosophila sp., 207, 247 multiple sperm storage organs in, 362

E Ecological benefits, of sociality in spiders, 123–124 Ecological constraints hypothesis, 154 of sociality in spiders, 124 Ecological speciation, 415 Egernia cunninghami, 236 Eisenia fetida, 219 Elseyornis melanops, 316 Embracing, 56 Eresidae, 84–87, 90–91, 109, 116 Eresus cinnaberinus Rhine, 113 Eresus sp., 116 Eresus walckenaeri Crete, 113 Erythrocebus, 48 Erythrogonys cinctus, 316

464

INDEX

Eudromias morinellus, 285, 316 Eurasian dotterel, 287 Eurynorhynchus pygmeus, 316 Evolution and maintenance of sociality in spiders by-product mutualism, 127–128 ecological benefits, 124–125 ecological constraints, 126 fostering model, 125 game theory models, 127 group selection, 122–125 kin recognition, 121–122 kin selection, 119–122 multilevel selection, 122–125 Experimental mate choice studies, heterozygosity and mate choice evidence for, 251–253 Extra-group paternity, in Seychelles warbler, 157, 160, 164 Extra-pair copulations (EPCs), 223, 225 in Seychelles warbler, 171–175 Extra-pair fertilizations (EPFs), in Seychelles warbler, 171 Extra-pair mate choice, heterozygosity and evidence for, 192, 224, 228, 237–243, 255 Extra-pair offspring, 192 Extra-pair paternity (EPP), 192–193, 227 and conflict over mating among shorebirds, 286, 290–293 female choice for increase offspring heterozygosity and, 238–242 in Seychelles warbler, 168–170, 172–173, 175–176

F Faiditus spp., 106 Fecal corticosteroid metabolites, in yellow-bellied marmots, 383 Fecal glucocorticoid levels, 382 Female philopatry, 73, 75 Female reproductive tract

ejaculate labeling and morphology of, 354–356 of S. stercoraria (L.), 349 Female-biased colony sex ratio, in spiders, 107–108, 123–125, 131, 133 Female-bonded primate groups coexistence in, 43–73 conciliatory tendency, 59 female-female aggression, 47 kinship and competition, 44–46 allogrooming, 45, 54–55 cooperative coalition, 45 reconciliation, 45, 56–63 social dynamics, 45–46 long-term partnerships, 46 male-male coalitions, 47 organising principles, 46–63 coalitions, 46–51, 74–75 grooming, 51–57, 63–64, 74–75 reconciliation, 56–63, 74–75 relationship, 63–70 affliation levels, 65 bereavement and, 67–68 bond strength and duration, 66–67 evolutionary function of, 69–70 matrilines splitting, 67, 71 propositional knowledge in, 64–65 seasonal dissolution, 66 social interaction levels, 65–66 social relationship, 68–69 variation with seasons, 66 spatial approach to social interaction, 70–73 vocal alliances, 47 Females choice, between good and compatible alleles, 222–224 Ficedula albicollis, 209 Ficedula hypoleuca, 212 Field flies, laboratory results with, 361–362 Fisher–Zahavi process, 193 Fisherian run away process, 193 Fisherian sex ratio, 107 Foraging societies, 83 Fortress defenders, 125

465

INDEX

Fostering model, of sociality in spiders, 125 Fratercula arctica, 316 Fratercula cirrhata, 316 Fratercula corniculata, 316 Fregata magnificens, 298 Fringillids, dialects in, 426 Fruitfly, 245

G Gallinago gallinago, 317 Gallinago hardwickii, 317 Gallinago media, 287, 317 Gallinago megala, 317 Gallinago nigripennis, 317 Gallinago stenura, 317 Gallus gallus, 249–250 Game theory models Battle of the Sexes, 5–6 Chicken game, 6–7, 31–32 Iterated Prisoner’s Dilemma, 4, 7–18 of forms of cooperation, 4–7 of sociality in spiders, 125 Stag Hunt, 5, 18–31 Gandanameno spenceri, 116–117 Gasterosteus aculeatus, 208 Gazella cuvieri, 213 Gelochelidon nilotica, 301 Genetic benefits compatible genes, 194–195 good genes, 194–195 of mate choice, 189, 192–199 Genetic compatibility, 194 Genetic incompatibility and mate choice, 193 Genetic monogamy, 148 Genetic quality, 197 Genetic similarity, and hatching success, 210 Genome incompatibility mechanism, 199 Genomic imprinting, 280 Geospiza fortis, 196 Geospiza scandens, 218

Glareola cinerea, 317 Glareola maldivarum, 317 Glareola nordmanni, 317 Glareola nuchalis, 317 Glareola ocularis, 317 Glareola pratincola, 317 Glareolidae, 284 Glycogen, 352 Golden marmots, 375 Good allele, for mate choice, 194–195, 197–198, 208, 223 Good genes models, of sexual selection, 193 Good sperm hypothesis, 226 Gorilla berengei berengei, 31 Great reed warbler, 230, 240, 257 Great snipe, 230, 287 Great tit, 258 Grooming among primates, 8–10 and tolerance, 9–10 for access to food, 9 in female-bonded groups, 51–57, 63–64, 74–75 Group activities, in spiders, 96 Group augmentation, 1, 34 Group living form, in spiders, 83 Group selection, in spiders, 122–125 Gryllus bimaculatus, 211 Guppy, 245 Gygis alba, 317

H Haematopodidae, 284 Haematopus bachmani, 317 Haematopus finschi, 317 Haematopus fuliginosus, 318 Haematopus longirostris, 318 Haematopus moquini, 318 Haematopus ostralegus, 287, 318 Haematopus palliatus, 318 Haematopus unicolor, 318 Hamilton’s theory, of kin selection, 1 Hatching success, 210–212 Hawk-Dove game, see Chicken game

466

INDEX

Helvibis cf. longicaudatus, 118 Helvibis thorelli, 90, 93 Heterosis, 199 Heterozygosity and environmental factors effects on inbreeding, 217–219 fitness correlations and Mhc, 207–208 determinants, 217–221 direct effect hypothesis, 205 evidence for, 205–207 general effect hypothesis, 204 interpretation of, 202–208 linear/quadratic relationship, 208–209 local effect hypothesis, 205 magnitude of, 201–204 traits affected by, 201–203 fitness effects, 208–217 on competitive ability of males and females, 215–216 on parasite load and immune system, 213–214 on reproductive success of males and females, 216–217 on survival, 209–213 methods and estimators, 199–200 Heterozygosity and mate choice, 189, 219–251, 258–259, see also Mate choice evidence for divorce and second broods, 234–235 experimental mate choice studies, 251–253 extra-pair mate choice, 192, 237–243 increase offspring heterozygosity through, 229 relatedness-dependent paternity bias, 244–251 sire choice, 232–236 social mate or copulation partner choice, 191, 229–234 for heterozygous individuals evidence for, 256–260

extra-pair mate choice, 189, 257–260 females preference for heterozygous males, 253–255 heterozygous females choice, 256 inbred females choice, 260 social mate or sire choice, 256–260 to optimize offspring heterozygosity, 221–229, 260 altruistic inbreeding, 221–222 costs and mechanisms of choice, 226–229, 261 females choice between good and compatible alleles, 222–224 promiscuity and inbreeding avoidance, 224–226 Heterozygote advantage mechanism, 97 Heterozygous individuals mate choice for, 253–260 evidence for, 256–260 extra-pair mate choice, 189, 224, 257–260 females preference for heterozygous males, 253–255 heterozygous females choice, 256 inbred females choice, 260 social mate or sire choice, 256–260 Himantopus himantopus, 318 Himantopus mexicanus, 318 Himantopus novaezelandiae, 318 Hirundo rustica, 229 Hoary marmot, 378 House finch, 230, 240 House sparrow, 230, 238, 240–241, 257–258 House wren, 240, 259 Hybrid vigor, 199 Hydrophasianus chirurgus, 286, 318

I Ibidorhyncha struthersii, 318 Imitative vocal learning, 404, 432 Inbred sociality in spiders, 84–88, 99–113

467

INDEX

boom and bust colony dynamics, 111–113 colony foundation, 103–107 cooperation and competition, 99–101 division of labor, 102–103 female-biased colony sex ratio, 107–108, 123–125, 131, 133 mating system, 108–111 operational sex ratios, 107–108 population-genetic consequences, 108–111 primary sex ratios, 107–108 propagule dispersal and fission, 103–107 Inbreeding altruistic, 221–222 and inbreeding avoidance in Seychelles warbler, 166–170 and kin selection in spiders, 120 avoidance and promiscuity, 224–227 depression, 199, 201, 204, 206, 209, 213, 215–217, 221, 224, 227, 252–253 environmental factors and heterozygosity effects on, 217–219 methods and estimators, 199–200 Increase offspring heterozygosity, 229 Individual heterozygosity, see also Heterozygous individuals methods and estimators, 199–200 traits affected by, 201–203 Infanticide, and conflict over mating among shorebirds, 293–294, 310 Interlocus sexual conflict, 281, 310 Intralocus sexual conflict, 281, 310 Irediparra gallinacea, 318 Iterated Prisoner’s Dilemma in primate groups, 4, 7–18, 31, 34 contingent reciprocity evidence of, 17–18 experimental settings, 8 laboratory experiments, 10–17 limitations preclude deployment of, 18

naturalistic experiments, 8–10, 34 nonexperimental settings, 7–8

J Jacana jacana, 287, 318 Jacana spinosa, 289, 291, 319 Jacanidae, 284

K Kentish plover, 238, 287 Kin recognition, in spiders, 121–122 Kin selection, 153 Hamilton’s theory, 1 in spiders, 119–122 Kinship deceit hypothesis, 163 Kissing, 56 Kleptoparasites, 106 Kochiura aulica, 118 Kochiura rosea, 118

L Lacerta agilis, 244 Laridae, 284 Larosterna inca, 319 Larus canus, 287, 319 Larus occidentalis, 287, 320 Larus spp., 319–320 Lek paradox, 192 Lekking polygymy concept, 280, 284 Lepomis gibbosus, 217 Lepomis macrochirus, 216 Liasis fuscus, 214 Lifetime reproductive success (LRS), 197 Limicola falcinellus, 321 Limnodromus griseus, 321 Limnodromus scolopaceus, 321 Limnodromus semipalmatus, 321 Limosa fedoa, 321 Limosa haemastica, 321

468

INDEX

Limosa lapponica, 321 Limosa limosa, 301, 321 Lip smacking, 56 Little auk, 287 Local mate competition (LMC) model, 124 Long-tailed marmot calls, 378 Lymnocryptes minimus, 285, 321

M Macaca arctoides, 59 Macaca fasicularis, 9 Macaca fuscata, 21 Macaca mulatta, 59, 207 Macaca spp., 8, 48 Macaca tonkeana, 21 Machetes pugnax, 283 Malagasy giant jumping rat, 231 Male ornaments and heterozygosity, 215 Male-male coalitions, 47 Male–female conflict, in S. stercoraria (L.), 349–352 Males and females heterozygosity and fitness effects on competitive ability of, 215–216 on reproductive success of, 216–217 Mallard, 246 Mallos gregalis, 85–87, 128 Market forces, 1, 34 Marmot phylogeny and spectrograms, 373 risk-based communication of, 386 Marmot alarm communication acoustic characteristics of, 387 evolution of, 372 function of, 381 meaning of, 385 nonphylogenetic and phylogenetic techniques of, 374 referential information, 385 Marmot species’ habitats acoustic transmission properties of, intraspecific variation in, 377

broadcasting and rerecording calls, 387 Marmota marmota, 219, 236 Mate choice benefits to females, 189, 191–193, 262 compatible allele, 194–195, 198–199 cryptic choice, 191 direct (resources) and indirect (genetic) benefits, 191–192 evolution of, 192–193 for heterozygous individuals, 253–260 evidence for, 256–260 extra-pair mate choice, 189, 257–260 females preference for heterozygous males, 253–255 heterozygous females choice, 256 inbred females choice, 260 social mate or sire choice, 256–260 genetic benefits, 189, 192–199 compatible genes, 194–195 good genes, 194–195 genetic quality, 197 good allele, 194–195, 197–198 heterozygosity and (see Heterozygosity and mate choice) in Seychelles warbler, 170–175 indirect genetic benefits, 189, 192–193 mechanisms of, 191 nature of indirect genetic benefits, 193 postcopulatory process, 191 precopulatory process, 191 prefertilization process, 191 social or copulation partner choice, 191, 229–234 types of, 191 Material care to cooperative breeding, in spiders, 129–131 Mating opportunities, 302–307 and parental care in shorebirds, 303–304 Mating optima, conflict over in shorebirds, 281–282, 288–294 Mating signals, 410 geographic divergence of, 403

469

INDEX

Mating system, in spiders, 84, 97–99, 108–111 Melanerpes formicivorus, 160 Meleagris gallopavo, 194 Melospiza georgiana, evolution of trill rate in, 438 Melospiza melodia, 214, 218, 435 Metopidius indicus, 286, 321 Mexican jay, 238 Mhc alleles, 194–195, 208–209, 252, 261 and heterozygosity-fitness correlations, 202–203, 207–208 diversity, 194, 209 MHC-dependent patterns of mating, in Seychelles warbler, 171–178 Micropalama himantopus, 321 Microparra capensis, 321 Microsatellite fingerprinting, 148 Molothrus aeneus, 408 Molothrus ater, 408 Monias benschi, 215 Monogymy concept, 280, 284–285 Multilevel selection, in spiders, 120–123 Multilocus Mhc markers, 199–200 Multilocus minisatellite markers, 199–200, 202–203 Multiple mating, and conflict over mating among shorebirds, 288–290 Multiple paternity in shorebirds, 286–287, 290–293 Mus musculus, 223 Mutualism, 3 Mycoplasma gallisepticum, 214

N Natal dialects, 409, see also Dialects Neotropical antbirds, vocal structure in, 434 Nepotistic behavior, 384 Nesticodes rufipes, 118 Nonhuman primates, see Primates Nonresource-based mating systems, 192

Nonterritorial permanent-social species, in spiders, 83–84, 89 Nonvocal functions, 439 North American pika, 233 Northern water snake, 232 Numenius spp., 321–322

O Ochotona princeps, 236 Olympic marmot, 378 individuality in, 390 Oncorhynchus tshawytscha, 213 Operational sex ratios, in spiders, 107–108 Ornate dragon lizard, 231–232 Oropendolas and caciques, 434 Oryctolagus cuniculus, 206 Oryx leucoryx, 213 Oscine passerine species, cooperative breeding in, 153 Outbreeding avoidance, 237, 248 depression, 209, 212–213, 248 to inbreeding in spiders, 130–131 Ovis aries, 196, 213 Oxyopidae, 85–87 Oystercatcher, 287

P Pagophila eburnea, 322 Pan troglodytes, 7, 56 Papio, 48 Papio baboons, 66 Papio cynocephalus, 7 Papio hamadryas, 28, 60 Papio hamadryas cynocephalus, 47 Papio hamadryas ursinus, 47 Papio papio, 21 Parental care, in shorebirds 280, 282–283, 288, 297–307 constraints on duration of care, 280, 299–302

470

INDEX

Parental care, in shorebirds (continued ) desertion and mating opportunities, 302–307 differential mating opportunities, 303–304 social mating system and, 279, 283, 285, 305–307 Parental investment theory, 306 Parental relatedness, methods and estimators, 199–200 Parus ater, 235 Parus major, 209, 218 Passerculus sandwichensis, 207 Passerine birds, 425–427 Paternity bias, and genetic similarity, 244–251 Pavo cristatus, 255 Pedigrees, 199–200, 202–203 Pedionomidae, 284 Pedionomus torquatus, 285, 322 Peltohyas australis, 322 Phalaropus fulicaria, 286, 322 Phalaropus fulicarius, 287 Phalaropus lobatus, 286–287, 322 Phalaropus spp., 284 Phasianus colchicus, 216 Philomachus pugnax, 244, 283, 287, 322 Phosphoglucomutase (PGM) locus, 352–354 at cryptic female choice, 356–358 offspring of, 354 Phryganoporus candida, 97 Phryganoporus candidus, 90–91 Phylogenetic relationship, in spiders, 114–117 Physalaemus pustulosus, 251 Pied flycatcher, 238 Pluvialis apricaria, 297, 322 Pluvialis dominica, 322 Pluvialis fulva, 322 Pluvialis squatarola, 323 Pluvianellidae, 284 Pluvianellus socialis, 323 Pluvianus aegyptius, 323 Poecilia reticulata, 215 Polyandry concept, 280, 284–285

Polymerase chain reaction (PCR)based techniques, 148 Pongo pygmaeus, 22 Population-genetic consequences, in spiders, 106–109 Post-conflict behavior, 56, 58, 60–63 Postcopulatory process, in mate choice, 191 Postcopulatory sexual selection, processes in, 343 Postcopulatory sperm competition, 195 Postdispersal learning, 413, 424 Postzygotic sexual conflict, 282–283 Precopulatory process, in mate choice, 191 Prefertilization process, in mate choice, 191 Premating to postmating dispersal, in spiders, 128–130 Presbytis, 48–49 Prezygotic sexual conflict, 281–282 Primary sex ratios, in spiders, 107 Primate groups, see also Femalebonded primate groups Chicken games, 31–32 contingent reciprocity evidence of, 17–18 experimental settings, 8 laboratory experiments, 10–17 limitations preclude deployment of, 18 naturalistic experiments, 8–10 nonexperimental settings, 7–8 Iterated Prisoner’s Dilemma in, 4, 7–18 social tolerance, 29 Stag Hunt, 18–31 Primates cooperation among nonrelatives in, 2 games played by, 7–31 Chicken games, 31–32 Iterated Prisoner’s Dilemma, 7–18 Stag Hunt, 18–31 grooming females among, 8–9 nature of cooperation among, 1–2 reciprocal altruism in, 8 studies of food exchanges in, 10–14

471

INDEX

Primitively eusocial wasps, 83 Prisoner’s Dilemma, 3–4, 6–7 Procolobus, 48 Promiscuity and inbreeding avoidance, in heterozygosity and mate choice, 224–226 Propagule dispersal and fission, in spiders, 103–107 Pseudopompilus funereus, 112 Ptychoramphus aleuticus, 323 Purple sandpiper, 287

Q Quasi-social species, in spiders, 83

R Radioactive amino acids, 354 Rana sylvatica, 219 Rana temporaria, 206, 219–220 Rangifer tarandus, 234 Reciprocal altruism, 19, 71, 382 in primates, 8 primitive form of, 14 theory of, 1, 4 Reconciliation, in female-bonded primate groups, 56–63, 74–75 Recurvirostra americana, 323 Recurvirostra avosetta, 323 Recurvirostra novaehollandiae, 323 Recurvirostridae, 284 Red flour beetle, 245 Red phalarope, 287, 290 Red-necked phalarope, 287, 289–290, 292, 297 Reed bunting, 238, 240–241, 257–259 Reindeer, 232 Relatedness-dependent paternity bias, heterozygosity and mate choice evidence for, 244–251 Relationship repair hypothesis, 58, 60–61

Remiz pendulinus, 282 Reproductive success of males and females, heterozygosity and fitness effects on, 216–217 Resource-based mating systems, 192 Rhesus macaque, 257 Rhinolophus ferrumequinum, 212 Rhinoptilus africanus, 323 Rhinoptilus chalcopterus, 323 Rhinoptilus cinctus, 323 Rhodostethia rosea, 323 Ringed plover, 287 Rissa brevirostris, 323 Rissa tridactyla, 323 Robust reciprocity, 17 Rostratula benghalensis, 283, 286, 323 Rostratula semicollaris, 323 Rostratulidae, 284 Ruff, 246, 287 Rynchopidae, 284 Rynchops flavirostris, 323 Rynchops niger, 323

S Stegodyphus africanus, 116 Stegodyphus bicolor, 116 Stegodyphus dufuori, 116 Stegodyphus dumicola A5, 116 Stegodyphus dumicola B2, 116 Scytodes fusca, 90–91 Stegodyphus lineatus, 116 Scytodes pallida, 90–91 Saguinus oedipus, 14 Saimiri, 48 Salmo salar, 215 Salmo trutta, 215 Salvelinus alpinus, 228 Savannah sparrow, 230, 234, 239, 258–259 Scathophaga stercoraria (L.) copula duration, 348 cryptic female choice in, 345, 349–352 female accessory glands and maternal effects, 360–361

472

INDEX

Scathophaga stercoraria (L.) (continued ) female arrival at dung pat, 347–349 female reproductive tract of, 349 male body size, 361 male-female conflict in, 349–352 male-female interactions in, 345 Parker’s pioneering work in behavioral ecology of, 347 sexual conflict in, 344–346 selection experiment on, 359 sperm competition in, 344, 347, 349–352 sperm length in, 350–352 testes size, evolution of, 347–349 Scathophagidae, male and female reproductive characters of, 362 Scatophagus, 346 Schooling fish, predator inspection in, 1 Scolopacidae, 284 Scolopax minor, 323 Scolopax rusticola, 324 Scytodes socialis, 90–91 Self-directed behaviors, 58 Selfish genetic element, 280 Semipalmated plover, 287 Semnopithecus entellus, 49 Senseless acts of intimidation, 59 Sex allocation concept, 148 Sex ratio, in spiders, 96, 107–108, 123–125, 130, 133–134 Sexual conflict and Shorebirds breeding systems, 279–281, 283–288, 304–308, 310 interlocus conflict, 281, 310 intralocus conflict, 281, 310 postzygotic conflict, 282–283 prezygotic conflict, 281–282 theory, 281–283, 310 Sexual size dimorphism (SSD), and mate choice in shorebirds, 280, 293–297, 309–310 Sexually antagonistic coevolution (SAC), 282, 310 Sexually selected sperm hypothesis, 195

Seychelles warbler, 231, 239–241, 257–260 cooperative breeding, 151, 153–166, 176 direct benefits, 163–166 indirect benefits, 156–158 kin discrimination by subordinates, 157–159 kin selection mechanism, 159–163 cross-fostering experiment, 161–162 extra-group paternity in, 157, 160, 164 extra-pair copulations, 171–175 extra-pair fertilizations, 171 extra-pair paternity, 168–170, 172–173, 175–176 inbreeding and inbreeding avoidance, 166–170 mate choice, 170–175 MHC-dependent patterns of mating, 171–178 pedigree based analysis, 177 shared maternity in, 156–157, 160 species and population, 149–153 Seyfarth’s model, 53–54 Shared maternity, in Seychelles warbler, 156–157, 160 Shorebirds breeding systems, 283–288 and sexual conflict, 279–281, 304–308, 310 conflict over mating, 288–297 infanticide, 293–294, 310 multiple and extra-pair paternity, 286, 290–293 multiple mating, 288–290 optima, 281–282, 288–294 data on, 311–328 mate choice and sexual size dimorphism, 280, 293–297, 309 mating opportunities in, 302–307 multiple paternity in, 286–287, 290–293 need of study on, 283, 309–310 display traits, 309 diversification, 309–310

INDEX

extension risks and population decline, 310 parental care, 280, 282–283, 288, 297–307 constraints on duration of care, 280, 299–302 desertion and mating opportunities, 302–307 differential costs of, 302–303 differential mating opportunities of, 303–304 social mating system and, 279, 283, 285, 305–307 tug-of-war over, 297–299 Single-locus Mhc markers, 199–200 Single-locus microsatellite markers, 199–200 Sire choice, heterozygosity and mate choice evidence for, 232–236 Snowdrift game, see Chicken game Soay sheep, 233 Social and subsocial species, in spiders, 89–97 Social brain hypothesis, 45–46, 51, 64, 69 Social dynamics, in female-bonded primate groups, 45–46 Social interactions, among animals, 147–148 Social mate choice, heterozygosity and mate choice evidence for, 191, 229–234 Social mating system, and parental care in shorebirds, 279, 283, 285, 305–307 Social monogamous passerine species, 148 Social partner choice, in mate choice, 191, 229–234 Social spiders, see also Sociality in spiders; Spiders behavioral and life-history characteristics, 84–88 behavioral traits, 89–97 aggression, 97 colony composition, 94

473

colony foundation, 94 group activities, 96–97 mating behavior, 97–99 colonial species, 83–84 cooperative species, 83 evolutionary dead end, 117–119, 134 group living form in, 83 inbred sociality in, 84–88, 99–113 mating, 84, 97–99 nonterritorial permanent-social species, 83–84, 89 phylogenetic relationship, 114–117 quasi-social species, 83 sex ratio, 96, 107–108, 123–125, 131, 133–134 social and subsocial species, 89–97 social evolution, 114–117 subsocial behavior, 89 territorial permanent-social species, 83–84, 89 Social status and heterozygosity, 215 Sociality in spiders, 117–128, see also Social spiders evolution and maintenance, 117–128 by-product mutualism, 127–128 ecological benefits, 125–126 ecological constraints, 126 fostering model, 125 game theory models, 127 group selection, 122–125 inbreeding and kin selection, 122 kin recognition, 121–123 kin selection, 119–122 multilevel selection, 122–125 inbreed sociality, 99–113 models of, 118–128 transitions in evolution from, 128–135 material care to cooperative breeding, 131–133 outbreeding to inbreeding, 130–131 premating to postmating dispersal, 128–130 subsocial to inbred social, 89–99, 133 Song acquisition and memorization, 429, 433

474 Song acquisition (continued ) timing of, 416, 424–425, 429–430 Song learning, 425 dispersal patterns and accuracy of, 415 social influences on, 411 timing of, 411, 436 Song models, postdispersal acquisition and learning of, 408 Song ontogeny, 429–431 role of, 433 Song sharing and vocal imitation, 435 hypotheses of, 435 selection for, 436 Song sparrow, 231 Songbird alarm calls, see Marmot alarm communication South Polar skua, 287 Sparassidae, 90–91 Sperm competition fertilization successes under, 346 in S. stercoraria (L.), 344, 347, 349–352 Sperm displacement, 355 Sperm length differences in, 352 genetic determination of, 351 in S. stercoraria (L.), 350–352 Sperm quality and heterozygosity, 216 Sperm storage, 350 Spermathecae, 345, 350 sperm distributions in, 353 Spiders, see also Social spiders cooperation and permanent group living in, 120 inbred sociality, 84–88, 99–113 boom and bust colony dynamics, 111–113 colony foundation, 103–107 cooperation and competition, 99–102 division of labor, 102–103 female-biased colony sex ratio, 107–108, 123–125, 131, 133

INDEX

mating system, 108–111 operational sex ratios, 107–108 population-genetic consequences, 108–111 primary sex ratios, 107–108 propagule dispersal and fission, 103–107 sociality in, 117–128 (see also Sociality in spiders) Splendid fairy-wren, 239, 241, 259 Spotted sandpiper, 287, 289 Squirrel alarm calls, 378 Stag Hunt in primate groups, 5, 18–31 conditions for success, 28–32 in wild, 18–20 solving collaboration problems in laboratory, 20–28 Steganopus tricolor, 286–287, 324 Stegodyphus (Eresidae), molecular phylogeny of, 115–116 Stegodyphus dumicola, 84–87, 100–105, 107–109, 111, 113, 116, 121, 124, 128, 133 Stegodyphus lineatus, 90–91, 109, 116, 121–122, 129–131, 133 Stegodyphus mimosarum, 85–87, 100, 104–105, 108, 112, 115, 116 Stegodyphus sarasinorum, 85–87, 100, 104, 109, 115–116 Stegodyphus tentoriicola, 98, 115, 134 Steppe marmot, 378 Stercorariidae, 284 Stercorarius spp., 324 Sterna hirundo, 287, 325 Sterna spp., 324–325 Sternidae, 284 Stiltia isabella, 325 Stress-related behavior, 58, 60 Sturnus unicolor, 212, 218 Subsocial to inbred social, in spiders, 89–99, 134–135 Syngnathus typhle, 288 Synthliboramphus antiquus, 325 Synthliboramphus hypoleucos, 325 Syrinx mass, 442 Systema Naturae, 346

475

INDEX

T Tachycineta bicolor, 234 Tapinillus sp. 1, 85–87, 95–96, 100, 111, 114 Tarbagan, 378 Tegenaria atrica, 129 Teleogryllus commodus, 211 Teleogryllus oceanicus, 212 Territorial permanent-social species, in spiders, 83–84, 89 Territorial polygymy concept, 280, 284 Territorial size and heterozygosity, 215 Territorial status and heterozygosity, 215 Testis size and heterozygosity, 216 Tetrao tetrix, 215 Thendula emertoni, 118 Theridiidae, 84–87, 90–91, 99, 109 Theridion frondeum, 118 Theridion longipedatum, 118 Theridion nigroannulatum, 85–87, 103–104, 106, 118 Theridion pictum, 90, 93, 99, 118, 129 Theridion varians, 118 Thinocoridae, 284 Thinocorus orbignyianus, 325 Thinocorus rumicivorus, 325 Thinornis novaeseelandiae, 325 Thomisidae, 85–87, 90, 93, 99, 110 Thymoites unimaculatum, 118 Tidarren sisyphoides, 118 Tiliqua rugosa, 229 Tits and sparrows, intraspecific studies of, 377 Trap-calling assay, 382 Tree sparrow, 231 Tree swallow, 239 Tringa cinerea, 326 Tringa erythropus, 326 Tringa flavipes, 326 Tringa glareola, 326 Tringa hypoleucos, 287, 326 Tringa incana, 326 Tringa macularia, 286–287, 326 Tringa melanoleuca, 326 Tringa nebularia, 326

Tringa ochropus, 326 Tringa solitaria, 326 Tringa stagnatilis, 326 Tringa totanus, 327 Tryngites subruficollis, 284, 287, 327 Tungara frog, 231, 257 Turdus helleri, 218 Turnicidae, 284 Turnix silvatica, 286

U Uria aalge, 287, 327 Uria lomvia, 327 Uroctea durandi, 116–117 Urocteidae, 117

V Valuable relationship hypothesis, 56 Vancouver Island marmots antipredator behavior, 392 call types of, 386 Vanellus spp., 327–328 Vanellus vanellus, 284, 328 Vocal alliances, in female-bonded primate groups, 47 Vocal frequencies, 441 Vocal geographic evolution, 404, 406 by-product models of, 433 literature overview on, 406 memes and mechanisms in, 440 potential causes of cultural drift, 441 cultural selection, 442 genetic drift, 442 natural selection, 442 sexual selection, 443 traditional framework to, 409 types of, 410 Vocal imitation, see also Imitative vocal learning accuracy in, 437 and male quality, 436–437 and song sharing, 435

476

INDEX

Vocal learning mechanisms, 435 and dispersal patterns, 406–409 training models of, 407 Vocal mating signals, in birds, 403 Vocal parameters, 434, 437 geographic structure in, 409 Vocal production mechanisms of, 438–440 motor components of, 438 role of body size and beak in, 439–440 Vocal tract, components of, 438

W Wattled jacana, 287 Western gull, 287 Western sandpiper, 239, 287 White-crowned sparrows, see Zonotrichia leucophrys nuttalli; Zonotrichia leucophrys oriantha; Zonotrichia leucophrys pugetensis White-toothed shrew, 233, 236 Wilson’s phalarope, 287 Wood frog, 232 X Xema sabini, 328

Y Yellow dung fly, see Scathophaga stercoraria (L.) Yellow-bellied marmots alarm calling, 375–376, 381 function of, 384 individuality in, 389 predation risk, 387 reliability in, 390 fecal corticosteroid metabolites in, 383 habituation-recovery experiment, 390 to reliable and unreliable caller, response of, 391

Z Zahavian handicap process, 193 Zalophus californianus, 213 Zonotrichia leucophrys gambell, 408 Zonotrichia leucophrys nuttalli, 406, 432 dialect evolution in, 407 Zonotrichia leucophrys oriantha, 214, 407, 411, 425, 432 Zonotrichia leucophrys pugetensis, 407, 425, 432 Zootermopsis angusticollis, 219

Contents of Previous Volumes

Volume 18 Song Learning in Zebra Finches (Taeniopygia guttata): Progress and Prospects PETER J. B. SLATER, LUCY A. EALES, AND N. S. CLAYTON Behavioral Aspects of Sperm Competition in Birds T. R. BIRKHEAD Neural Mechanisms of Perception and Motor Control in a Weakly Electric Fish WALTER HEILIGENBERG Behavioral Adaptations of Aquatic Life in Insects: An Example ANN CLOAREC The Circadian Organization of Behavior: Timekeeping in the Tsetse Fly, A Model System JOHN BRADY

The Evolution of Courtship Behavior in Newts and Salamanders T. R. HALLIDAY Ethopharmacology: A Biological Approach to the Study of Drug-Induced Changes in Behavior A. K. DIXON, H. U. FISCH, AND K. H. MCALLISTER Additive and Interactive Effects of Genotype and Maternal Environment PIERRE L. ROUBERTOUX, MARIKA NOSTEN-BERTRAND, AND MICHELE CARLIER Mode Selection and Mode Switching in Foraging Animals GENE S. HELFMAN Cricket Neuroethology: Neuronal Basis of Intraspecific Acoustic Communication FRANZ HUBER Some Cognitive Capacities of an African Grey Parrot (Psittacus erithacus) IRENE MAXINE PEPPERBERG

Volume 19 Volume 20 Polyterritorial Polygyny in the Pied Flycatcher P. V. ALATALO AND A. LUNDBERG Kin Recognition: Problems, Prospects, and the Evolution of Discrimination Systems C. J. BARNARD Maternal Responsiveness in Humans: Emotional, Cognitive, and Biological Factors CARL M. CORTER AND ALISON S. FLEMING

Social Behavior and Organization in the Macropodoidea PETER J. JARMAN The t Complex: A Story of Genes, Behavior, and Population SARAH LENINGTON The Ergonomics of Worker Behavior in Social Hymenoptera PAUL SCHMID-HEMPEL 477

478

CONTENTS OF PREVIOUS VOLUMES

‘‘Microsmatic Humans’’ Revisited: The Generation and Perception of Chemical Signals BENOIST SCHAAL AND RICHARD H. PORTER

Parasites and the Evolution of Host Social Behavior ANDERS PAPE MOLLER, REIJA DUFVA, AND KLAS ALLANDER

Lekking in Birds and Mammals: Behavioral and Evolutionary Issues R. HAVEN WILEY

The Evolution of Behavioral Phenotypes: Lessons Learned from Divergent Spider Populations SUSAN E. RIECHERT

Volume 21

Proximate and Developmental Aspects of Antipredator Behavior E. CURIO

Primate Social Relationships: Their Determinants and Consequences ERIC B. KEVERNE The Role of Parasites in Sexual Selection: Current Evidence and Future Directions MARLENE ZUK Conceptual Issues in Cognitive Ethology COLIN BEER Response to Warning Coloration in Avian Predators W. SCHULER AND T. J. ROPER Analysis and Interpretation of Orb Spider Exploration and Web-Building Behavior FRITZ VOLLRATH Motor Aspects of Masculine Sexual Behavior in Rats and Rabbits GABRIELA MORALI AND CARLOS BEYER On the Nature and Evolution of Imitation in the Animal Kingdom: Reappraisal of a Century of Research A. WHITEN AND R. HAM

Volume 22 Male Aggression and Sexual Coercion of Females in Nonhuman Primates and Other Mammals: Evidence and Theoretical Implications BARBARA B. SMUTS AND ROBERT W. SMUTS

Newborn Lambs and Their Dams: The Interaction That Leads to Sucking MARGARET A. VINCE The Ontogeny of Social Displays: Form Development, Form Fixation, and Change in Context T. G. GROOTHUIS

Volume 23 Sneakers, Satellites, and Helpers: Parasitic and Cooperative Behavior in Fish Reproduction MICHAEL TABORSKY Behavioral Ecology and Levels of Selection: Dissolving the Group Selection Controversy LEE ALAN DUGATKIN AND HUDSON KERN REEVE Genetic Correlations and the Control of Behavior, Exemplified by Aggressiveness in Sticklebacks THEO C. M. BAKKER Territorial Behavior: Testing the Assumptions JUDY STAMPS Communication Behavior and Sensory Mechanisms in Weakly Electric Fishes BERND KRAMER

CONTENTS OF PREVIOUS VOLUMES

Volume 24 Is the Information Center Hypothesis a Flop? HEINZ RICHNER AND PHILIPP HEEB Maternal Contributions to Mammalian Reproductive Development and the Divergence of Males and Females CELIA L. MOORE Cultural Transmission in the Black Rat: Pine Cone Feeding JOSEPH TERKEL The Behavioral Diversity and Evolution of Guppy, Poecilia reticulata, Populations in Trinidad A. E. MAGURRAN, B. H. SEGHERS, P. W. SHAW, AND G. R. CARVALHO Sociality, Group Size, and Reproductive Suppression among Carnivores SCOTT CREEL AND DAVID MACDONALD Development and Relationships: A Dynamic Model of Communication ALAN FOGEL Why Do Females Mate with Multiple Males? The Sexually Selected Sperm Hypothesis LAURENT KELLER AND HUDSON K. REEVE

479

An Overview of Parental Care among the Reptilia CARL GANS Neural and Hormonal Control of Parental Behavior in Birds JOHN D. BUNTIN Biochemical Basis of Parental Behavior in the Rat ROBERT S. BRIDGES Somatosensation and Maternal Care in Norway Rats JUDITH M. STERN Experiential Factors in Postpartum Regulation of Maternal Care ALISON S. FLEMING, HYWEL D. MORGAN, AND CAROLYN WALSH Maternal Behavior in Rabbits: A Historical and Multidisciplinary Perspective GABRIELA GONZA¨LEZ-MARISCAL AND JAY S. ROSENBLATT Parental Behavior in Voles ZUOXIN WANG AND THOMAS R. INSEL Physiological, Sensory, and Experiential Factors of Parental Care in Sheep F. LE¨VY, K. M. KENDRICK, E. B. KEVERNE, R. H. PORTER, AND A. ROMEYER

Cognition in Cephalopods JENNIFER A. MATHER

Socialization, Hormones, and the Regulation of Maternal Behavior in Nonhuman Simian Primates CHRISTOPHER R. PRYCE

Volume 25

Field Studies of Parental Care in Birds: New Data Focus Questions on Variation among Females PATRICIA ADAIR GOWATY

Parental Care in Invertebrates STEPHEN T. TRUMBO Cause and Effect of Parental Care in Fishes: An Epigenetic Perspective STEPHEN S. CRAWFORD AND EUGENE K. BALON Parental Care among the Amphibia MARTHA L. CRUMP

Parental Investment in Pinnipeds FRITZ TRILLMICH Individual Differences in Maternal Style: Causes and Consequences of Mothers and Offspring LYNN A. FAIRBANKS

480

CONTENTS OF PREVIOUS VOLUMES

Mother–Infant Communication in Primates DARIO MAESTRIPIERI AND JOSEP CALL Infant Care in Cooperatively Breeding Species CHARLES T. SNOWDON Volume 26 Sexual Selection in Seaweed Flies THOMAS H. DAY AND ANDRE¨ S. GILBURN Vocal Learning in Mammals VINCENT M. JANIK AND PETER J. B. SLATER Behavioral Ecology and Conservation Biology of Primates and Other Animals KAREN B. STRIER How to Avoid Seven Deadly Sins in the Study of Behavior MANFRED MILINSKI Sexually Dimorphic Dispersal in Mammals: Patterns, Causes, and Consequences LAURA SMALE, SCOTT NUNES, AND KAY E. HOLEKAMP Infantile Amnesia: Using Animal Models to Understand Forgetting MOORE H. ARNOLD AND NORMAN E. SPEAR Regulation of Age Polyethism in Bees and Wasps by Juvenile Hormone SUSAN E. FAHRBACH Acoustic Signals and Speciation: The Roles of Natural and Sexual Selection in the Evolution of Cryptic Species GARETH JONES

Volume 27 The Concept of Stress and Its Relevance for Animal Behavior DIETRICH VON HOLST Stress and Immune Response VICTOR APANIUS Behavioral Variability and Limits to Evolutionary Adaptation P. A. PARSONS Developmental Instability as a General Measure of Stress ANDERS PAPE MOLLER Stress and Decision-Making under the Risk of Predation: Recent Developments from Behavioral, Reproductive, and Ecological Perspectives STEVEN L. LIMA Parasitic Stress and Self-Medication in Wild Animals G. A. LOZANO Stress and Human Behavior: Attractiveness, Women’s Sexual Development, Postpartum Depression, and Baby’s Cry RANDY THORNHILL AND F. BRYANT FURLOW Welfare, Stress, and the Evolution of Feelings DONALD M. BROOM Biological Conservation and Stress HERIBERT HOFER AND MARION L. EAST

Volume 28

Understanding the Complex Song of the European Starling: An Integrated Ethological Approach MARCEL EENS

Sexual Imprinting and Evolutionary Processes in Birds: A Reassessment CAREL TEN CATE AND DAVE R. VOS

Representation of Quantities by Apes SARAH T. BOYSEN

Techniques for Analyzing Vertebrate Social Structure Using Identified

CONTENTS OF PREVIOUS VOLUMES

Individuals: Review and Recommendations HAL WHITEHEAD AND SUSAN DUFAULT Socially Induced Infertility, Incest Avoidance, and the Monopoly of Reproduction in Cooperatively Breeding African Mole-Rats, Family Bathyergidae NIGEL C. BENNETT, CHRIS G. FAULKES, AND JENNIFER U. M. JARVIS Memory in Avian Food Caching and Song Learning: A General Mechanism or Different Processes? NICOLA S. CLAYTON AND JILL A. SOHA Long-Term Memory in Human Infants: Lessons in Psychobiology CAROLYN ROVEE-COLLIER AND KRISTIN HARTSHORN Olfaction in Birds TIMOTHY J. ROPER Intraspecific Variation in Ungulate Mating Strategies: The Case of the Flexible Fallow Deer SIMON THIRGOOD, JOCHEN LANGBEIN, AND RORY J. PUTMAN

Volume 29 The Hungry Locust STEPHEN J. SIMPSON AND DAVID RAUBENHEIMER Sexual Selection and the Evolution of Song and Brain Structure in Acrocephalus Warblers CLIVE K. CATCHPOLE Primate Socialization Revisited: Theoretical and Practical Issues in Social Ontogeny BERTRAND L. DEPUTTE

481

Ultraviolet Vision in Birds INNES C. CUTHILL, JULIAN C. PARTRIDGE, ANDREW T. D. BENNETT, STUART C. CHURCH, NATHAN S. HART, AND SARAH HUNT What Is the Significance of Imitation in Animals? CECILIA M. HEYES AND ELIZABETH D. RAY Vocal Interactions in Birds: The Use of Song as a Model in Communication DIETMAR TODT AND MARC NAGUIB

Volume 30 The Evolution of Alternative Strategies and Tactics H. JANE BROCKMANN Information Gathering and Communication during Agonistic Encounters: A Case Study of Hermit Crabs ROBERT W. ELWOOD AND MARK BRIFFA Acoustic Communication in Two Groups of Closely Related Treefrogs H. CARL GERHARDT Scent-Marking by Male Mammals: Cheat-Proof Signals to Competitors and Mates L. M. GOSLING AND S. C. ROBERTS Male Facial Attractiveness: Perceived Personality and Shifting Female Preferences for Male Traits across the Menstrual Cycle IAN S. PENTON-VOAK AND DAVID I. PERRETT The Control and Function of Agonism in Avian Broodmates HUGH DRUMMOND

482

CONTENTS OF PREVIOUS VOLUMES

Volume 31 Conflict and Cooperation in a Female-Dominated Society: A Reassessment of the ‘‘Hyperaggressive’’ Image of Spotted Hyenas MARION L. EAST AND HERIBERT HOFER Birdsong and Male–Male Competition: Causes and Consequences of Vocal Variability in the Collared Dove (Streptopelia decaocto) CAREL TEN CATE, HANS SLABBEKOORN, AND MECHTELD R. BALLINTIJN Imitation of Novel Complex Actions: What Does the Evidence from Animals Mean? RICHARD W. BYRNE Lateralization in Vertebrates: Its Early Evolution, General Pattern, and Development LESLEY J. ROGERS Auditory Scene Analysis in Animal Communication STEWART H. HULSE Electric Signals: Predation, Sex, and Environmental Constraints PHILIP K. STODDARD How to Vocally Identify Kin in a Crowd: The Penguin Model THIERRY AUBIN AND PIERRE JOUVENTIN

Volume 32 Self-Organization and Collective Behavior in Vertebrates IAIN D. COUZIN AND JENS KRAUSE Odor-Genes Covariance and Genetic Relatedness Assessments: Rethinking

Odor-Based Recognition Mechanisms in Rodents JOSEPHINE TODRANK AND GIORA HETH Sex Role Reversal in Pipefish ANDERS BERGLUND AND GUNILLA ROSENQVIST Fluctuating Asymmetry, Animal Behavior, and Evolution JOHN P. SWADDLE From Dwarf Hamster to Daddy: The Intersection of Ecology, Evolution, and Physiology That Produces Paternal Behavior KATHERINE E. WYNNE-EDWARDS Paternal Behavior and Aggression: Endocrine Mechanisms and Nongenomic Transmission of Behavior CATHERINE A. MARLER, JANET K. BESTER-MEREDITH, AND BRIAN C. TRAINOR Cognitive Ecology: Foraging in Hummingbirds as a Model System SUSAN D. HEALY AND T. ANDREW HURLY

Volume 33 Teamwork in Animals, Robots, and Humans CARL ANDERSON AND NIGEL R. FRANKS The ‘‘Mute’’ Sex Revisited: Vocal Production and Perception Learning in Female Songbirds KATHARINA RIEBEL Selection in Relation to Sex in Primates JOANNA M. SETCHELL AND PETER M. KAPPELER

CONTENTS OF PREVIOUS VOLUMES

Genetic Basis and Evolutionary Aspects of Bird Migration PETER BERTHOLD Vocal Communication and Reproduction in Deer DAVID REBY AND KAREN MCCOMB Referential Signaling in Non-Human Primates: Cognitive Precursors and Limitations for the Evolution of Language KLAUS ZUBERBU«HLER Vocal Self-stimulation: From the Ring Dove Story to Emotion-Based Vocal Communication MEI-FANG CHENG

483

HARALD LACHNIT, MARTIN GIURFA, AND RANDOLF MENZEL Begging, Stealing, and Offering: Food Transfer in Nonhuman Primates GILLIAN R. BROWN, ROSAMUNDE E. A. ALMOND, AND YFKE VAN BERGEN Song Syntax in Bengalese Finches: Proximate and Ultimate Analyses KAZUO OKANOYA Behavioral, Ecological, and Physiological Determinants of the Activity Patterns of Bees P. G. WILLMER AND G. N. STONE

Volume 35 Volume 34 Reproductive Conflict in Insect Societies ˆ RGEN HEINZE JU Game Structures in Mutualistic Interactions: What Can the Evidence Tell Us About the Kind of Models We Need? REDOUAN BSHARY AND JUDITH L. BRONSTEIN Neurobehavioral Development of Infant Learning and Memory: Implications for Infant Attachment TANIA L. ROTH, DONALD A. WILSON, AND REGINA M. SULLIVAN Evolutionary Significance of Sexual Cannibalism MARK A. ELGAR AND JUTTA M. SCHNEIDER Social Modulation of Androgens in Vertebrates: Mechanisms and Function RUI F. OLIVEIRA Odor Processing in Honeybees: Is the Whole Equal to, More Than, or Different from the Sum of Its Parts?

Mechanisms and Evolution of Communal Sexual Displays in Arthropods and Anurans MICHAEL D. GREENFIELD A Functional Analysis of Feeding GEORGE COLLIER The Sexual Behavior and Breeding System of Tufted Capuchin Monkeys (Cebus apella) MONICA CAROSI, GARY S. LINN, AND ELISABETTA VISALBERGHI Acoustic Communication in Noise HENRIK BRUMM AND HANS SLABBEKOORN Ethics and Behavioral Biology PATRICK BATESON Prenatal Sensory Ecology and Experience: Implications for Perceptual and Behavioral Development in Precocial Birds ROBERT LICKLITER Conflict and Cooperation in Wild Chimpanzees MARTIN N. MULLER AND JOHN C. MITANI

484

CONTENTS OF PREVIOUS VOLUMES

Trade-Offs in the Adaptive Use of Social and Asocial Learning RACHEL L. KENDAL, ISABELLE COOLEN, YFKE VAN BERGEN, AND KEVIN N. LALAND

Functional Genomics Requires Ecology LARA S. CARROLL AND WAYNE K. POTTS

Volume 36

Preexisting Male Traits Are Important in the Evolution of Elaborated Male Sexual Display GERALD BORGIA

Suckling, Milk, and the Development of Preferences Toward Maternal Cues by Neonates: From Early Learning to Filial Attachment? RAYMOND NOWAK A Neuroethological Approach to Song Behavior and Perception in European Starlings: Interrelationships Among Testosterone, Neuroanatomy, Immediate Early Gene Expression, and Immune Function GREGORY F. BALL, KEITH W. SOCKMAN, DEBORAH L. DUFFY, AND TIMOTHY Q. GENTNER Navigational Memories in Ants and Bees: Memory Retrieval When Selecting and Following Routes THOMAS S. COLLETT, PAUL GRAHAM, ROBERT A. HARRIS, AND NATALIE HEMPEL-DE-IBARRA

Signal Detection and Animal Communication R. HAVEN WILEY

Adaptation, Genetic Drift, Pleiotropy, and History in the Evolution of Bee Foraging Behavior NIGEL E. RAINE, THOMAS C. INGS, ANNA DORNHAUS, NEHAL SALEH, AND LARS CHITTKA Kin Selection, Constraints, and the Evolution of Cooperative Breeding in Long-Tailed Tits BEN J. HATCHWELL AND STUART P. SHARP How Do Little Blue Penguins ‘‘Validate’’ Information Contained in Their Agonistic Displays? JOSEPH R. WAAS