Reproductive Skew in Vertebrates Reproductive skew is the study of how reproduction is partitioned in animal societies...
26 downloads
1088 Views
10MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Reproductive Skew in Vertebrates Reproductive skew is the study of how reproduction is partitioned in animal societies. In many social animals reproduction is shared unequally, leading to a reproductive skew among group members. Skew theory investigates the genetic and ecological factors causal to the partitioning of reproduction in animal groups and may yield fundamental insights into the evolution of animal sociality. This book brings together new theory and empirical work, mostly in vertebrates, to test assumptions and predictions of skew models. It also gives an updated critical review of skew theory. The team of leading contributors cover a wide range of species, from insects to humans, and discuss both ultimate (evolutionary) and proximate (immediate) factors influencing reproductive skew. Academic researchers and graduate students alike with an interest in evolution and sociality will find this material stimulating and exciting. r e i n m a r h a g e r is an NERC Research Fellow in evolutionary biology at the University of Manchester. He has been awarded University Fellowship by the German Science Foundation and was elected Senior Rouse Ball Scholar at Trinity College, Cambridge. d r . c l a r a b . j o n e s , p h . d . , p h . d . is a scientist and consultant trained by several universities, research agencies, and corporations. She has published more than 100 texts (both technical, theoretical, empirical, and popular), including several books. Currently, Dr. Clara resides in The Americas, continuing both academic, theoretical, scientific, applied, and corporate projects.
Reproductive Skew in Vertebrates Proximate and Ultimate Causes Edited by Reinmar Hager University of Manchester
Clara B. Jones Fayetteville State University, North Carolina
cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sa˜o Paulo, Delhi Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521864091 ª Cambridge University Press 2009 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2009 Printed in the United Kingdom at the University Press, Cambridge A catalog record for this publication is available from the British Library Library of Congress Cataloging in Publication data Hager, Reinmar, 1974– Reproductive skew in vertebrates : proximate and ultimate causes / Reinmar Hager, Clara B. Jones. p. cm. Includes bibliographical references and index. ISBN 978-0-521-86409-1 (hardback) 1. Vertebrates–Reproduction. 2. Sexual selection in animals. I. Jones, Clara B. II. Title. QP251.H13 2009 591.560 2–dc22 2009007300 ISBN 978-0-521-86409-1 hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
Contents
List of contributors page vii Foreword Sandra L. Vehrencamp xi Preface xix
Part I 1
Reproductive skew theory
1
Models of reproductive skew: outside options and the resolution of reproductive conflict 3 Rufus A. Johnstone and Michael A. Cant
2
Reproductive conflict and the evolution of menopause
24
Michael A. Cant, Rufus A. Johnstone, and Andrew F. Russell
Part II 3
Testing assumptions and predictions of skew models
51
Reproductive skew in female-dominated mammalian societies 53 Kay E. Holekamp and Anne L. Engh
4
The effects of heterogeneous regimes on reproductive skew in eutherian mammals 84 Clara B. Jones
5
Social skew as a measure of the costs and benefits of group living in marmots 114 Thea B. Wang, Peter Nonacs, and Daniel T. Blumstein
6
Explaining variation in reproductive skew among male langurs: effects of future mating prospects and ecological factors 134 Reinmar Hager
v
vi Contents
7
The causes and consequences of reproductive skew in male primates 165 Nobuyuki Kutsukake and Charles L. Nunn
8
Sociality and reproductive skew in horses and zebras
196
Daniel I. Rubenstein and Cassandra M. Nun˜ez
9
Reproductive skew in avian societies
227
Walter D. Koenig, Sheng-Feng Shen, Alan H. Krakauer, and Joseph Haydock
10
Reproductive skew in cooperative fish groups: virtue and limitations of alternative modeling approaches 265 Michael Taborsky
11. Reproductive skew in primitively eusocial wasps: how useful are current models? 305 Jeremy Field and Michael A. Cant
Part III 12
Resolving reproductive conflicts: behavioral and physiological mechanisms 335
Reproductive skew in female common marmosets: contributions of infanticide and subordinate self-restraint 337 David H. Abbott, Leslie Digby, and Wendy Saltzman
13
Reproductive skew in African mole-rats: behavioral and physiological mechanisms to maintain high skew
369
Chris G. Faulkes and Nigel C. Bennett
14
The causes of physiological suppression in vertebrate societies: a synthesis 397 Andrew J. Young
Part IV 15
Future directions
437
Understanding variation in reproductive skew: directions for future empirical research 439 Sarah J. Hodge
16
On the evolution of reproductive skew: a genetical view 467 W. Edwin Harris and Reinmar Hager
17
Social conflict resolution, life history, and the reconstruction of skew 480 Bernard J. Crespi
Taxonomic index Subject index
508 511
Contributors
David H. Abbott Wisconsin National Primate Research Center, University of Wisconsin, Madison, WI, USA
Nigel C. Bennett Mammal Research Institute, University of Pretoria, South Africa
Daniel T. Blumstein Department of Ecology and Evolutionary Biology, University of California, Los Angeles, CA, USA
Michael A. Cant Centre for Ecology and Conservation, University of Exeter, UK
Bernard J. Crespi Evolutionary Biology, Simon Fraser University, Burnaby, BC, Canada
Chris G. Faulkes School of Biological and Chemical Sciences, Queen Mary, University of London, UK
Jeremy Field Department of Biology and Environmental Science, University of Sussex, UK
Leslie Digby Department of Biological Anthropology and Anatomy, Duke University, Durham, NC, USA
Anne L. Engh Department of Zoology, University of Pennsylvania, Philadelphia, PA, USA
Reinmar Hager Faculty of Life Sciences, University of Manchester, UK
vii
viii List of Contributors
W. Edwin Harris Faculty of Life Sciences, University of Manchester, UK
Joseph Haydock Department of Biology, Gonzaga University, Spokane, WA, USA
Sarah J. Hodge Department of Zoology, University of Cambridge, UK
Kay E. Holekamp Department of Zoology, Michigan State University, East Lansing, MI, USA
Rufus A. Johnstone Department of Zoology, University of Cambridge, UK
Clara B. Jones Department of Psychology, Fayetteville State University, Fayetteville, NC, USA; and National Evolutionary Synthesis Center, Duke University, Durham, NC, USA
Walter D. Koenig Hastings Reservation and Museum of Vertebrate Zoology, University of California, Berkeley, CA, USA
Alan H. Krakauer Section of Evolution and Ecology, University of California, Davis, CA, USA
Nobuyuki Kutsukake Department of Evolutionary Studies of Biosystems, Graduate University for Advanced Studies Hayama, Miura-gun, Zushi, Kanagawa, Japan
Peter Nonacs Department of Ecology and Evolutionary Biology, University of California, Los Angeles, CA, USA
Cassandra M. Nun˜ez Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA
Charles L. Nunn Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany; and Department of Integrative Biology, University of California, Berkeley, CA, USA
Daniel I. Rubenstein Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA
Andrew F. Russell Department of Animal and Plant Sciences, University of Sheffield, UK
List of Contributors
Wendy Saltzmann Department of Biology, University of California, Riverside, CA, USA
Sheng-Feng Shen Department of Neurobiology and Behavior, Cornell University, Ithaca, NY, USA
Michael Taborsky Department of Behavioural Ecology, University of Berne, Switzerland
Sandra L. Vehrencamp Department of Neurobiology and Behavior, Cornell University, Ithaca, NY, USA
Thea B. Wang Department of Ecology and Evolutionary Biology, University of California, Los Angeles, CA, USA
Andrew J. Young Centre for Ecology and Conservation, University of Exeter, UK
ix
Foreword
A brief history of skew theory New ideas in science don’t spring out of nowhere; they combine and build from earlier ones. The concept of reproductive skew is no different, and it is nice to have this opportunity to look back, over 30 years ago now, and identify the various sources and give credit where it’s due. The story begins in 1974 at the 16th International Ornithological Congress in Canberra, Australia. During that long flight to Australia from Johannesburg, South Africa, the pilot came on the loudspeaker and announced that the United States President Richard Nixon had just resigned in disgrace over the Watergate scandal. Having been in the field in Africa all summer, this was a shocking return to civilization. Ian Rowley had convened the first symposium on cooperative breeding in birds, with himself, Lew Grimes, Glen Woolfenden, and Amotz Zahavi presenting surveys of the cooperatively breeding species in their respective continents (Australia, Africa, the Americas, and Europe) (Grimes 1976, Rowley 1976, Woolfenden 1976, Zahavi 1976). All four speakers noted that the helper-at-the-nest form of cooperative breeding, where offspring remain on the parental territory, delay breeding, and assist with the care of subsequent broods of their parents, was by far the most common one. Cooperative breeding was associated with a variety of habitats and climates, but most were characterized by sedentary residence on territories or fixed home ranges. Habitat saturation was identified by Woolfenden and Zahavi as the primary force favoring prolonged retention of offspring on the parental territory. During the contributed talk sessions, two of us, John Craig and I, spoke about our work on two species that differed significantly from the usual pattern in having multiple males and females breeding in a communal nest: groove-billed anis and pukeko. A lively workshop discussion followed, in which we argued about whether kin altruism or individually selected selfish
xi
xii Foreword behavior was the more important driving force for the evolution of cooperation. I tried to focus the discussion on the question of what might be driving the two different forms of cooperative breeding, the helper system versus the communal system, and suggested that lack of reproduction by helpers was the more pressing question to answer. Woolfenden glared at me and said no, the burden was on me, Craig, and others to explain when communal nesting would be favored, since helping was the norm and communal nesting the rarer form. The gauntlet had been thrown down, and I left the meeting determined to answer this question. In 1976, with my thesis finished and my first academic job beginning, a second motivator arose: teaching a new course (jointly with spouse Jack Bradbury). E. O. Wilson’s Sociobiology (Wilson 1975) had just appeared, and became the textbook for our course of the same name. Wilson’s review of the two alternative routes to sociality, the familial and the communal routes, was inspiring, and it quickly became clear that many animal groups, including social insects, spiders, birds, and mammalian carnivores, all showed these two forms of cooperative breeding (Eisenberg 1966, Wilson 1971, Lin & Michener 1972, Brown 1974). I felt compelled to come up with an explanation for the students. Hamilton’s rule (Hamilton 1964) could explain the altruism of helpers towards close kin, but it couldn’t explain the aggregation and mutual cooperation of unrelated individuals in a communal system. Richard Alexander’s insightful review of animal social behavior (Alexander 1974) held the next major key: he clearly articulated that an individual’s gain from remaining in a social group had to be compared to the alternatives of living alone or taking the risk of finding and joining another group. He also noted that different competitive dynamics were likely to occur in different types of groups, i.e. family groups versus groups of unrelated individuals. Finally, he asserted that group-living animals invariably form dominance hierarchies in which dominants often gain at the expense of subordinates, and coined the phrase “parental manipulation of progeny” to describe selection on parents to reduce the reproduction of certain offspring in order to increase their inclusive fitness via other offspring. A new perspective started to emerge: a focus on the power of the dominant and its greater ability to shape group composition and bias reproductive shares to its advantage. But the leverage of the dominant had limits if subordinates could opt to leave. Under conditions of habitat saturation, offspring would have poor outside options and therefore would be forced to remain with their parents, where they could be prevented from breeding by parental domination but able to increase their inclusive fitness by helping. By contrast, in unsaturated conditions groups would only form if there were some
Foreword type of mutualistic advantage, and dominants would be limited in their leverage to demand the cooperation of subordinates unless subordinates obtained enough direct fitness benefits to make staying in the group worthwhile. I first worked up the graphical analysis of the limits of dominant manipulation in unrelated groups, which then allowed me to incorporate Hamilton’s rule to compute the inclusive fitness break-even point within and outside the group for the case of related group members. Only three variables, combined in a very simple model, were required to explain the full range of high- and low-skew breeding systems: the benefit of group versus solitary breeding, the availability of options for breeding outside the group, and the coefficient of relatedness. I presented the bare bones of the model at the next Ornithological Congress in Berlin, 1978 (Vehrencamp 1980). Woolfenden was there, but he did not offer any comments. Stephen Emlen, my former thesis advisor, who also attended this congress, clearly did grasp the significance of the idea and subsequently came out with his own version of it, without formulating a fully quantitative model (Emlen 1982). It took a few more years to completely vet the mathematics of the model and explore the parameter space with different values of relatedness, group benefits, and ecological constraints. I appreciate the feedback from my mathematically competent colleagues at UC San Diego, Mike Gilpin, Ted Case, and Kurt Fristrup. I was pleased when John Maynard Smith took an interest in the model during our 1980 sabbatical leave in Sussex, UK; he not only incorporated it as an example of a two-stage game into the game theory book he was writing (Maynard Smith 1982), but he “did the sums” in a slightly different way by assuming that in larger groups subordinates made individual decisions about whether to stay or leave. To explain the occurrence of equitably breeding groups of relatives within the model’s framework, some factor that prevented enforcement by dominants such as constraints on control, reproductive inefficiency caused by within-group conflict, or coalitions of subordinates had to be invoked. If these factors were widespread, such that observed skew was often less than the maximum predicted by the model, I surmised that the original model would not be very useful. Occurrence of skews greater than predicted by the model would certainly falsify it. In the final revision of the 1983 paper (Vehrencamp 1983), a reviewer and editor were critical of the use of the word “skew” as a verb and insisted I use the term “bias.” Once the paper was published, I figured the idea would have to cook. Since it seemed very difficult to conduct experimental tests and manipulations on birds, and the molecular means for determining paternity and maternity were not fully developed, I switched to research on animal communication.
xiii
xiv Foreword Kern Reeve, along with colleagues Francis Ratnieks and Laurent Keller, gave the theoretical model a shot in the arm in the early 1990s by simplifying the math, limiting group size to two, greatly clarifying the specific predictions and tests of the model, and extending the model to make predictions about when groups should be stable versus unstable (Reeve 1991, Reeve & Ranieks 1993, Keller & Reeve 1994). In addition, they added the possibility that the degree of skew might be limited by subordinate fighting for complete control of the group’s resources, derived skew in parent–offspring associations, and generalized the skew model to N-person groups with diminishing group benefits as group size increases (Reeve & Keller 1995, 1997, Reeve & Emlen 2000). With their background in social insects, they shifted the terminology and mechanisms of group formation to a perspective different from my vertebrate approach. Whereas I conceived of groups of potential breeders initially forming for reasons of birth location or ecological conditions, followed by the emergence of a dominance hierarchy and the subsequent suppression of reproduction by subordinates just short of the point where they should leave, Reeve envisioned a dominant overwintering queen returning first to the birth nest site and then enticing later arrivals to join in a group. Words like “incentive,” “payment,” “concession,” “negotiation,” and “social contract” were incorporated to describe the processes of group formation and reproductive partitioning. Many lab groups were engaged in empirical testing of the model’s predictions. To my surprise, the reproductive biasing models suddenly became “skew theory.” A symposium devoted to skew theory was held in Sheffield in 1997 to bring together vertebrate and invertebrate perspectives. Along with the notoriety and testing came a wave of skepticism articulated by Tim CluttonBrock, who expected dominants to have perfect control and both dominants and subordinates to have perfect knowledge of reproductive shares and the availability of outside options (Clutton-Brock 1998). Beginning in 1998, an explosion of new skew models by Reeve, Michael Cant, Rufus Johnstone, Hanna Kokko, and others appeared (Cant 1998, Reeve et al. 1998, Cant & Johnstone 1999, Johnstone & Cant 1999, Johnstone et al. 1999, Kokko and Johnstone 1999, Crespi & Ragsdale 2000, Reeve 2000, Cant & Field 2001, Kokko et al. 2001, to mention only a few). In response to the criticism, these models variously modified the assumptions and mechanisms of dominant control, incorporated additional fitness components such as the potential for subordinates to inherit the dominant position, and considered the costs of enforcing skew and producing more young. A major alternative model was also developed, the tug-of-war model (Reeve et al. 1998), which assumed that
Foreword reproductive shares were the outcome of a costly conflict between two individuals of different competitive abilities. Ecological constraints do not influence reproductive partitioning in this model, but both participants reduce their combative effort when more closely related so skew tends to be unaffected or even lower than when they are less closely related. Renewed efforts to test the alternative predictions of the tug-of-war and optimal skew models sometimes found better support for the tug-of-war predictions. A new synthetic model by Reeve and Shen (2006) that combines the tug-of-war process within the constraints of options outside the group, called the bordered tug-of-war model, holds great promise as a single flexible model. This model assumes that dominant and subordinate continually adjust their payments and selfish tug-of-war efforts in response to each other (K. Reeve, personal communication). In a real way, it better incorporates the biasing mechanisms and constraints that I envisioned 30 years ago. By varying parameters that set whether one, both, or no parties concede some direct fitness to the partner to keep her/him from leaving, one can cover the whole range of traditional concessions, bordered tug-of-war, and pure tug-of-war models. Each of these three processes appears to operate under different conditions of ecological constraint, benefit to grouping, and relatedness. The current state of skew theory is well represented in the chapters of this book. Chapters by Taborsky on cooperatively breeding fish and by Kutsukake and Nunn on primates highlight the extreme range of ease and difficulty, respectively, of manipulating and measuring the key skew-determining parameters. It is heartening to read in the review of avian cooperative breeders by Koenig et al. that a meta-analysis largely supports the predictions of the traditional optimal skew model for birds. Chapters by Abbott et al., Faulkes and Bennett, and Young argue that in mammals, despite the high cost of offspring production, dominant females can attain a high skew through hormonal suppression and infanticide, for example. Jones, on the other hand, argues that high variability and unpredictability of ecological conditions limits the ability of dominants to control subordinates in socially flexible mammals. A useful review of reproductive skew studies in primitively eusocial wasps by Field and Cant points out that skew is often greater than predicted by the concessions model. They discuss whether this outcome is a clear falsification of the model, or whether either ecological constraints or inheritance of the dominant position has not been properly measured and considered. Hager also reports that future reproductive potential may be driving the observation of high reproductive skew in male langurs. Holekamp and Engh demonstrate the impact of the control one sex may have over skew in the opposite sex, a problem that continually arises in cases of vertebrate cooperative breeders containing
xv
xvi Foreword multiple members of both sexes. Cant and Johnstone explore a model similar to the bordered tug-of-war, but where mutual adjustments between dominant and subordinate are not allowed. Despite the skepticism and sometimes-heated disagreements (for example, see Magrath and Heinsohn 2000, Nonacs 2006, Nonacs 2007, and chapters by Hodge and Crespi), the collective body of theoretical and empirical work on reproductive skew has been and continues to be an engaging and fruitful field of study with broad ramification to many aspects of social behavior in animals and humans. Sandra L. Vehrencamp Cornell University
References Alexander, R. D. (1974). The evolution of social behavior. Annual Review of Ecology and Systematics, 5, 325–384. Brown, J. L. (1974). Alternative routes to sociality in jays. American Zoologist, 14, 63–80. Cant, M. A. (1998). A model for the evolution of reproductive skew without reproductive suppression. Animal Behaviour, 55, 163–169. Cant, M. A. & Field, J. (2001). Helping effort and future fitness in cooperative animal societies. Proceedings of the Royal Society of London B, 268, 1959–1964. Cant, M. A. & Johnstone, R. A. (1999). Costly young and reproductive skew in animal societies. Behavioral Ecology, 10, 178–184. Clutton-Brock, T. H. (1998). Reproductive skew, concessions and limited control. Trends in Ecology & Evolution, 13, 288–292. Crespi, B. J. & Ragsdale, J. E. (2000). A skew model for the evolution of sociality via manipulation: why it is better to be feared than loved. Proceedings of the Royal Society of London B, 267, 821–828. Eisenberg, J. F. (1966). The social organization of mammals. Handbuch der Zoologie, 10, 1–92. Emlen, S. T. (1982). The evolution of helping. I. An ecological constraints model. American Naturalist, 119, 29–39. Grimes, L. G. (1976). Cooperative breeding in African birds. Proceedings of the International Ornithological Congress, 16, 666–673. Hamilton, W. D. (1964). The genetical evolution of social behavior, I & II. Journal of Theoretical Biology, 7, 1–52. Johnstone, R. A. & Cant, M. A. (1999). Reproductive skew and the threat of eviction: a new perspective. Proceedings of the Royal Society of London B, 266, 275–279. Johnstone, R. A., Woodroffe, R., Cant, M. A., & Wright, J. (1999). Reproductive skew in multimember groups. American Naturalist, 153, 315–331.
Foreword Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology & Evolution, 9, 98–102. Kokko, H. & Johnstone, R. A. (1999). Social queuing in animal societies: a dynamic model of reproductive skew. Proceedings of the Royal Society of London B, 266, 571–578. Kokko, H., Johnstone, R. A., & Clutton-Brock, T. H. (2001). The evolution of cooperative breeding through group augmentation. Proceedings of the Royal Society of London B, 268, 187–196. Lin, M. & Michener, C. (1972). Evolution of sociality in insects. Quarterly Review of Biology, 47, 131–159. Magrath, R. D. & Heinsohn, R. G. (2000). Reproductive skew in birds: models, problems and prospects. Journal of Avian Biology, 31, 247–258. Maynard, Smith J. (1982). Evolution and the Theory of Games. Cambridge: Cambridge University Press. Nonacs, P. (2006). The rise and fall of transactional skew theory in the model genus Polistes. Annales Zoologici Fennici, 43, 443–455. Nonacs, P. (2007). Tug-of-war has no borders: it is the missing model in reproductive skew theory. Evolution, 61, 1244–1250. Reeve, H. K. (1991). The social biology of Polistes. In K. Ross & R. Matdiews, eds., The Social Biology of Wasps. Ithaca, NY: Cornell University Press, pp. 99–148. Reeve, H. K. (2000). A transactional theory of within-group conflict. American Naturalist, 155, 365–382. Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model. Behavioral Ecology, 11, 640–647. Reeve, H. K. & Keller, L. (1995). Partitioning of reproduction in mother–daughter versus sibling associations: a test of optimal skew theory. American Naturalist, 145, 119–132. Reeve, H. K. & Keller, L. (1997). Reproductive bribing and policing evolutionary mechanisms for the suppression of within-group selfishness. American Naturalist, 150, S42–S58. Reeve, H. K. & Ranieks, F. L. W. (1993). Queen–queen conflict in polygynous societies: mutual tolerance and repoductive skew. In L. Keller, ed., Queen Number and Sociality in Insects. Oxford: Oxford University Press, pp. 45–85. Reeve, H. K. & Shen, S. F. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Rowley, I. (1976). Cooperative breeding in Australian birds. Proceedings of the International Ornithological Congress, 16, 657–666. Vehrencamp, S. L. (1980). To skew or not to skew? In R. No¨hring, ed., Acta XVII Congressus Internationalis Ornithologici. Vol. I. Berlin: Verlag der Deutschen Ornithologen-Gesellschaft. pp. 869–874.
xvii
xviii Foreword Vehrencamp, S. L. (1983). A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667–682. Wilson, E. O. (1971). The Insect Societies. Cambridge, MA: Belknap Press. Wilson, E. O. (1975). Sociobiology: the New Synthesis. Cambridge, MA: Belknap Press. Woolfenden, G. E. (1976). Cooperative breeding in American birds. Proceedings of the International Ornithological Congress, 16, 674–684. Zahavi, A. (1976). Cooperative breeding in Eurasian birds. Proceedings of the International Ornithological Congress, 16, 685–694.
Preface
We have collaborated on projects since 2001, and it became evident to us by 2004 that a book about reproductive skew incorporating theoretical, empirical, and review chapters might be timely. Because of our own research specializations, and because a large body of literature exists on skew in social insects, we generated a plan to prepare a volume on reproductive skew in vertebrates, approaching Cambridge University Press with a proposal. The present text represents our attempt to provide a “state of the art” overview of reproductive skew in vertebrate societies by some of the most active and highly regarded researchers in this field. It is our intention to highlight the most fundamental questions for students of reproductive skew, to assess the strengths and weaknesses of skew models, to critically evaluate skew in insect societies and skew in social vertebrates, and to identify important directions for future theoretical and empirical work. In her foreword to our volume, Sandra Vehrencamp has provided a brief overview of the history, theory, and empirical highlights of models of reproductive skew. It is clear from her personalized account that advances in the study of reproductive skew (the within-sex partitioning of reproduction within social groups) were presaged by early work in behavioral ecology demonstrating a relationship between dispersion and quality of limiting resources, in particular, food and nesting sites, as well as variations in social behavior, social organization, and mating systems within and between populations. The relationship between these factors is analyzed in models of reproductive skew that attempt to explain the partitioning of reproduction among individuals of the same sex in animals and offer a theoretical framework for understanding the formation of social groups. In addition to the central role of kin selection, skew theory identifies other fundamental processes that are key to the evolution of complex sociality, such as suppression of reproduction or the control of group membership, and it may thus enable us to
xix
xx Preface study one of the major transitions in evolution, from primarily solitary to social lifestyles, across a range of different taxa. Our book provides the reader with theoretical, empirical, and review chapters on a variety of model vertebrate systems exemplifying high, intermediate, and low skew and the causes and consequences of biased reproduction within groups. Furthermore, since social insects have been the classic exemplars of high-skew societies, we include a chapter on these organisms and their similarities and differences to social vertebrates. We think that a particularly helpful feature of our text is its future value as a reference tool. A related utility is the book’s presentation of many ideas for future research that have not been thoroughly investigated to date: for example, the evolution of low-skew societies and underlying mechanisms of suppression. An additional feature of Reproductive Skew in Vertebrates: Proximate and Ultimate Causes is its presentation of caveats about skew models for those who may be uncritically enthusiastic about these paradigms. The investigation of all aspects of reproductive skew is in its early stages, and we hope that our volume will help this field develop into a more mature, and critical, enterprise. As editors, we would be remiss if we failed to acknowledge and to express gratitude to the many individuals who have facilitated the process of our book’s actualization – from initial proposal to published text. We thank our initial contacts at Cambridge University Press whose interest in our ideas never faltered. Our editor at Cambridge, Martin Griffiths, has provided significant input, guidance, support and encouragement for our project. Without Martin’s expert ability to diagnose and to resolve challenges, our book may not have gone to press. Subsequent to the decision by Cambridge University Press to invite our initial book proposal and to distribute it for review, we received helpful and constructive criticism and advice from several researchers investigating the topic of reproductive skew. These critiques influenced our final decisions regarding our volume’s conceptual framework and organization, and also prompted us to include additional chapters on topics we had not previously considered. Indeed, the initial feedback we received from reviewers of our proposal was critical in stimulating our thinking about target questions and heightened our enthusiasm for our project, and we gratefully acknowledge the comments of these researchers. Our most profound thanks are extended to our contributors, who have demonstrated professionalism, expertise, patience, and, most important, good humor at every stage of our project. Without the willingness of these individuals to engage in frequent communication about their submissions and to receive our suggestions with grace, even when they may have disagreed with them, our book would not have been realized. We hope that our contributors
Preface will be pleased with the final text, as we are, and that they, their colleagues, and their students will value their contributions to Reproductive Skew in Vertebrates: Proximate and Ultimate Causes for many years to come. Among the contributors to our book that we wish to acknowledge by name are Sandra Vehrencamp and Bernard Crespi, who assumed responsibility for writing the foreword and the concluding chapter, respectively. These experts met unique challenges requiring the ability to summarize the field of reproductive skew, both retroactively and for the future. We hope that both of these contributions will help to place the book’s chapters in a broad context by linking each topic with past work, with the present state of the field, and with the literature on reproductive skew and related topics yet to be published. Finally, we wish to express personal thanks to those who have been particularly influential in our careers. Reinmar Hager is especially grateful to Rufus Johnstone, his thesis advisor at Cambridge University, for introducing him to skew theory, and to Yfke Hager, for help with editing. Clara B. Jones, likewise, thanks her dissertation advisor at Cornell University, William C. Dilger, for encouraging her interests in plants and animals and for facilitating her research interest in the evolution of social behavior. In the final analysis, it may be both necessary and sufficient to stress that we hope you find our volume a “good read” that will provide information, stimulate thinking, and generate original research, both theoretical and empirical, for its readers and their extended academic families. We look forward to these outputs as well as your measured feedback.
xxi
I
Reproductive skew theory
1
Models of reproductive skew: outside options and the resolution of reproductive conflict rufus a. j ohnstone and m ichael a. ca nt
Summary The two main types of skew models, transactional and compromise models, make different assumptions about the division of reproduction. Transactional models assume that one individual has full control over reproduction within the group, but may have to refrain from claiming all reproduction in order to prevent others leaving or evicting it from the group. Compromise models, by contrast, ignore outside options such as departing to breed elsewhere, but allow for incomplete control over reproduction within the group. Attempts to synthesize these two approaches have proved controversial. Here, we show that this controversy can be resolved using a simple principle from the economic literature on bargaining – the “outside option principle.” Even if outside options are available, they will influence the outcome of reproductive conflict within a group only if they yield greater payoffs than are available within the group. We present a novel synthetic model based on this principle, in which individuals engage in a tug-of-war over reproduction within a group, but may “ease off ” in their competitive effort in response to the threat of departure or eviction. We show that over a large range of parameter space, particularly when group productivity and relatedness among group members are high, these threats are not credible, so that opportunities outside the group do not influence the stable level of skew. However, when group productivity and relatedness are low, one or other of the players will typically ease off in competition in order to maintain group stability. Under these circumstances, outside Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
3
4
R. A. Johnstone, M. A. Cant options do influence skew. Tests which examine the relationship between skew and factors such as group productivity or ecological constraints are thus expected to yield variable results. The essential question is whether or not any of the members of a group stand to gain from its dissociation. The answer will determine whether or not outside options come into play.
Introduction The term reproductive skew refers to inequality in the distribution of breeding success among members of a group (Vehrencamp 1983, Keller & Reeve 1994). In high-skew societies, such as those of honeybees (Apis mellifera), Mexican jays (Aphelocoma ultramarina), or meerkats (Suricata suricatta), the distribution of reproductive opportunities is markedly unequal (see, e.g., Chapter 13 in this volume). One or a few breeders monopolize reproduction, while others are denied the opportunity to mate or to raise offspring (and may even, in the case of eusocial insects or naked mole-rats (Heterocephalus glaber), develop as sterile workers). In low-skew societies, by contrast, all individuals have similar opportunities to breed (see, e.g., Chapter 15). Inequality in breeding success is not a precisely defined concept. There are many different plausible ways in which to conceive and measure inequality in a multi-member group, leading to many different indices of skew (Kokko & Lindstro¨m 1997, Tsuji & Kasuya 2001). Moreover, inequality in actual reproductive success may be expected to arise simply by chance, even if all individuals have similar opportunities to breed (e.g. Haydock & Koenig 2002). Nevertheless, however one chooses to measure inequality, it is clear that there are dramatic and consistent differences in skew within and among species (Keller & Reeve 1994, Reeve & Keller 2001). Even closely related species otherwise similar in their ecology and behavior may differ markedly in this respect – compare, for instance, dwarf mongooses (Helogale parvula), in which typically only one female in a group breeds, with banded mongooses (Mungos mungo), in which most females breed in each attempt (Cant 2000). These differences in reproductive skew cry out for explanation. Models of reproductive skew attempt to provide an adaptive account of variation in skew both between and within species. They assume that there exists a conflict of interest among members of a group, such that each would benefit by obtaining a greater share of reproduction than is in the best interest of the others. Each model then predicts how (at evolutionary equilibrium) this conflict of interest will be resolved, depending upon various factors such as the benefits of group membership, the opportunity for independent breeding, and the ability of each group member to compete for resources or breeding
Models of reproductive skew opportunities, as well as to evict or exclude others from the group (Reeve & Ratnieks 1993, Keller & Reeve 1994, Johnstone 2000). Models of skew may differ in their assumptions about the extent of conflict between group members (e.g. Cant & Johnstone [1999] suggested that when production of offspring entails accelerating costs, individuals might all benefit from sharing reproductive opportunities). Most, however, are distinguished by their assumptions about the relative power of dominant and subordinate individuals. The most striking contrast in this respect is between transactional and compromise models of skew.
Transactional models of skew Transactional models of skew were the first to be developed. Vehrencamp’s (1979, 1983) seminal papers, in which she introduced the concept of reproductive skew, were built around a transactional model, and it is this approach that has been followed in most later analyses (e.g. Reeve & Ratnieks 1993, Johnstone & Cant 1999, Buston et al. 2007). In fact, when biologists talk of skew theory it is usually the transactional approach that they have in mind. The basic assumption of transactional models is that animals may concede reproductive opportunities to others, despite being potentially able to claim these opportunities for themselves, in order to maintain the stability of the group. This is advantageous because cooperation is presumed to yield productivity benefits. It may therefore pay to yield some reproduction to others in order to gain (or continue to enjoy) the benefits of associating with them. There are in fact two types of transactional model. Early analyses focused on reproductive concessions offered by dominants to retain helpful subordinates in the group – in these models, dominance takes the form of complete control over reproduction, with dominant individuals yielding breeding opportunities to subordinates so as to make it worth their while remaining in the group rather than departing (Vehrencamp 1979, 1983, Reeve & Ratnieks 1993, Reeve & Emlen 2000). By contrast with this notion of “concessions,” the “restraint” model focuses on reproductive concessions offered by subordinates to prevent the dominant evicting them – in these models, dominance takes the form of control over group membership, with subordinates refraining from claiming as large a share of reproduction as they might, lest the dominant eject them from the group (Clutton-Brock 1998, Johnstone & Cant 1999). Both types of model can be formalised in a similar way (Reeve & Ratnieks 1993, Johnstone & Cant 1999, Johnstone 2000, Buston et al. 2007). Consider a pair of individuals, related by a coefficient r. In association, the combined reproductive success of the pair, relative to that of an established lone breeder,
5
6
R. A. Johnstone, M. A. Cant is given by the parameter k (typically > 1). The expected reproductive success of an individual that disperses to breed elsewhere, again relative to that of an established lone breeder, is given by the parameter d (typically < 1, since we assume that dispersal entails some risk or cost). In a stable association, each individual must obtain sufficient reproductive success within the group that it would not gain either by leaving or by evicting the other (where eviction is possible). In the concession model, one “dominant” individual is assumed to have complete control over the distribution of reproduction within the group, subject only to the threat of departure by the “subordinate.” In this case, the subordinate in a stable group is expected to receive the minimum share of reproduction, pmin, that is compatible with group stability, i.e. a share that is just sufficient to ensure that leaving is not profitable. This share is given by ðkpmin dÞ þ rðkð1 pmin Þ 1Þ ¼ 0
ð1:1Þ
where the first term on the left-hand side represents the direct fitness impact of staying (rather than leaving) on the subordinate, and the second term the indirect fitness impact on the reproductive success of the dominant. Rearranging, we obtain pmin ¼
d r ðk 1 Þ kð1 rÞ
ð1:2Þ
(if d < r (k 1), then the subordinate does best to remain even if the dominant completely monopolizes reproduction). In the restraint model, by contrast, the “subordinate” is free to claim unsanctioned reproduction, subject only to the threat of eviction by the “dominant”. In this case, the dominant is expected to receive the minimum share of reproduction, qmin, that is compatible with group stability, i.e. a share that is just sufficient to ensure that evicting the subordinate is not profitable. This share is given by ðkqmin 1Þ þ r ðkð1 qmin Þ dÞ ¼ 0
ð1:3Þ
where the first term on the left-hand side represents the direct fitness impact of tolerating the subordinate’s presence (rather than evicting it from the group) on the dominant, and the second term the indirect fitness impact of toleration on the reproductive success of the subordinate. Rearranging, we obtain qmin ¼
1 r ð k dÞ k ð1 r Þ
ð1:4Þ
Models of reproductive skew In both models, the association will prove stable provided that pmin þ qmin < 1
ð1:5Þ
i.e. provided that the pair are together sufficiently productive that both may simultaneously receive at least their minimum required share. Substituting Equations 1.2 and 1.4 into 1.5, this yields the condition 1þd
ð1:6Þ
implying that the association will prove stable provided that the total productivity of both individuals would be reduced if it were to break up. Concessions and restraint
From Equations 1.2 and 1.6, we see that in the concession model the subordinate’s share of reproduction in a stable group increases with the opportunity for independent breeding (d), but decreases with the productivity of the group (k) and the degree of relatedness between the players (r). This makes intuitive sense – when there are greater benefits to be obtained elsewhere, the subordinate must receive a greater share of group productivity to make staying worthwhile; at the same time, the greater the productivity of the group, and the greater the relatedness between the players, the smaller the share required to satisfy this requirement. From Equations 1.4 and 1.6, however, we see that in the restraint model, the dominant’s share exhibits precisely the same trends (with the exception that it decreases with the opportunity for independent breeding only for r > 0). Consequently, the two models generally yield opposite predictions regarding skew. For instance, when group productivity increases, the concession model predicts a decrease in the subordinate’s share, leading to greater skew (provided that the dominant enjoys greater reproductive success than the subordinate). By contrast, the restraint model predicts a decrease in the dominant’s share, leading to reduced skew (again provided that the dominant enjoys greater reproductive success than the subordinate). The difference between the two models lies in the roles assigned to the two individuals. In the concession model, it is the subordinate that threatens to break up the group, and the dominant that must concede reproduction it could otherwise claim in order to maintain the association; in the terminology of Buston et al. (2007) it is the dominant that allocates reproduction to its partner. By contrast, in the restraint model it is the dominant that threatens to break up the group, and the subordinate that must concede reproduction; in this case it is the subordinate that is the allocator. The terms
7
8
R. A. Johnstone, M. A. Cant “dominant” and “subordinate” thus carry different meanings in the two models. In the concession model, dominance denotes control over the division of reproduction (subject to the threat of departure by the subordinate), while in the restraint model, dominance denotes control over group membership (i.e. the power to evict).
Compromise models The concession and restraint models yield opposing predictions about skew, but both are instances of the transactional approach. In both, one individual is assumed to exercise full control over the division of reproduction, subject only to the threat of group breakup (which may be initiated by the other individual). Compromise models instead assume that each member of the group can act selfishly to claim a greater share of breeding opportunities, at a cost to the productivity of the group as a whole (Clutton-Brock 1998, Reeve et al. 1998, Johnstone 2000). The outcome of the conflict over reproduction depends upon the level of selfish effort invested by each individual, and on their relative “strength.” In this kind of model, “dominance” typically takes the form of greater competitive ability – the dominant individual may be able to invest more effort in competition than can subordinates, or it may obtain a greater share for the same level of investment, due to superior resource-holding potential (Reeve et al. 1998). The other factor that can affect the outcome of the conflict in these models is relatedness among group members, which potentially influences the level of competitive effort that each individual will invest at equilibrium. The most influential compromise model, the tug-of-war game of Reeve et al. (1998), focuses (like the simple transactional models described above) on the interaction between two individuals, one dominant and one subordinate, who are related by a coefficient r. Both players simultaneously choose how much effort to invest in selfish competition over reproductive opportunities within the group. The levels of effort will be denoted x for the dominant and y for the subordinate. Total group productivity is equal to k(1xy), where the parameter k specifies group productivity relative to that of a lone breeder in the absence of competition. Productivity thus declines linearly with total expenditure by both players on selfish competition. The fraction of reproduction claimed by the dominant is equal to x/(x þ by), where the parameter b (<1) specifies the competitive ability of the subordinate relative to that of the dominant. Dominance in this context is thus defined by superior competitive ability.
Models of reproductive skew Given the above assumptions, the direct fitness payoffs to the two players from the tug-of-war are given by x x þ by by Ws ðx; yÞ ¼ kð1 x yÞ x þ by
Wd ðx; yÞ ¼ kð1 x yÞ
ð1:7Þ
Reeve et al. (1998) derived the (unique) stable pair of effort levels x* and y* in the basic tug-of-war game, each of which simultaneously maximizes (for the player in question, given the other’s behavior) the sum of its own direct fitness payoff plus r times that of the other player. These effort levels satisfy the first-order condition @Wd ðx; yÞ @Ws ðx; yÞ @Ws ðx; yÞ @Wd ðx; yÞ þr ¼ þr ¼0 @x @x @y @y
ð1:8Þ
and are given by 2 x ¼
3
b 7 6 2 rð1 bÞ 4qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 15 2ð1 bÞ 2 r2 ð1 bÞ þ4b 3 2
ð1:9Þ
1 2b þ r ð1 bÞ 7 6 y ¼ 41 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi5 2ð1 bÞ r2 ð1 bÞ2 þ4b
The share of reproduction obtained by the subordinate at this equilibrium increases with its relative competitive ability, b, but surprisingly it is largely insensitive to the relatedness between the competitors, r. As relatedness increases, both players reduce their competitive effort, with the result that skew changes little.
Outside options Much recent discussion of reproductive skew has focused on the issue of control. Can the dominant really prevent subordinates from claiming any share of reproduction, as concession models assume? Or are subordinates able to gain some reproductive success even against the best interests of the dominant, as in compromise models? Can subordinates even claim a large enough share that the threat of eviction becomes relevant? Framed in these terms, compromise and transactional approaches start to seem less clearly distinct. Perhaps concession and restraint models simply represent extreme cases on a continuum of
9
10
R. A. Johnstone, M. A. Cant dominant control – the former deal with cases in which the dominant has full control over reproduction, the latter with cases in which the dominant has none (except that granted by the threat of eviction). There remains, however, another fundamental difference between transactional and compromise models. The transactional models incorporate outside options – the possibility of leaving to breed elsewhere, or of evicting a competitor from the group. Indeed, it is these outside options that, in transactional analyses, set limits on the level of skew possible in a stable group. The fitness prospects of individuals that leave or are forced out of a group, usually referred to in terms of ecological constraints, thus have a major influence on skew in transactional analyses. By contrast, compromise models have generally ignored outside options, and focus only on the resolution of conflict within the group. Ecological constraints thus have no influence on the outcome of these analyses. In reality, it seems likely that animals will often have the opportunity to leave a group and breed elsewhere, or to evict others from the association. But at the same time, it also seems likely that no one individual will enjoy complete control over reproduction within the group. A realistic model of skew, therefore, needs to incorporate both competition over reproduction within the group, as in compromise models, and outside options, as in transactional models. How are these two approaches to be combined? While there have been several attempts to “synthesize” skew models in order to answer this question, there is still little agreement over which approach is appropriate ( Johnstone 2000, Reeve & Shen 2006, Nonacs 2007).
Synthetic models In a previous review paper (Johnstone 2000), one of us outlined a synthetic model of skew that attempted to incorporate outside options into the tug-of-war model of Reeve et al. (1998). In this combined model, animals were assumed first to yield concessions to one another, in the form of uncontested shares of group productivity, sufficient to ensure that neither would benefit by leaving the group. Subsequently, both were assumed to engage in a tug-of-war competition over the remaining, contested fraction of reproduction. The model suffers from a problem, however, in that the concessions required to prevent departure in the first step are calculated on the basis of total group productivity, overlooking the fact that some of this productivity will be squandered in competition during the second step. Because of the costs of competition, the original concessions lose some of their value, and can no longer be relied upon to ensure group stability when the outcome of the tug-ofwar is taken into account.
Models of reproductive skew Reeve and Shen (2006) attempted to deal with this problem in a modified synthetic model. They too allow individuals to concede uncontested shares of reproduction to one another, and assume that a tug-of-war takes place over the remaining contested fraction. However, they assume that the magnitude of the concessions and the intensity of effort in the tug-of-war are decided simultaneously. Consequently, they suggest that an “evolutionarily stable” outcome must satisfy two simultaneous conditions: (1) each player must be just indifferent between remaining in the group and choosing its outside option, and (2) each player’s competitive effort in the tug-of-war must be a best response to that of the other, given the magnitude of the concessions made by each (i.e. selfish efforts are at a Nash equilibrium). Once again, however, this approach seems to us problematic. First, it rests on the questionable assumption that the players use up all the potential benefits of group membership in the process of finding a resolution (see Nonacs 2007). Second, the putative solution is also based on the assumption that each player gives away a “free” share of reproduction to its opponent while simultaneously fighting hard for the remainder. Intuitively, it would make more sense for each player to reduce the size of the free share offered to its partner, and invest less effort in wasteful competition. This intuition is borne out by formal analysis – one can show that any solution of the form suggested by Reeve and Shen (2006) can be invaded by a mutant that offers a reduced “handout” and at the same time invests less in wasteful competition over the remainder. Nonacs (2007) forcefully critiques both of the above models, highlighting some of the same problems we have pointed out here. He then presents results of a simulation analysis in which individuals may exchange shares of uncontested reproduction while simultaneously engaging in a tug-of-war over the remainder. The results are that such exchange always proves unstable, and that levels of competitive effort evolve to the solution of the basic tug-of-war game. However, since the analysis omits any possibility of departure or eviction, this is unsurprising. Removing all outside options renders concessions pointless, and the game merely reduces to the basic tug-of-war. The question thus remains, when outside options are available, when and how will this affect the outcome of competition over reproduction within the group? If the exchange of uncontested shares of reproduction (as proposed by Johnstone 2000 and Reeve & Shen 2006) is unfeasible or unstable, how should we expect competitors to respond to the threat of the group breaking up?
Bargaining theory and the outside option principle A possible solution to the problem of synthesis can be gleaned from work in economic bargaining theory, which is concerned with problems that
11
12
R. A. Johnstone, M. A. Cant are structurally very similar to those addressed by models of skew (Nash 1950, Osborne & Rubinstein 1990, Muthoo 1999). In economics, a “bargaining situation” is one in which two or more players have a common interest to cooperate, but disagree over how the profits of cooperation should be distributed. An obvious example is trade, although formally similar problems arise in a range of other contexts, from salary negotiations and litigation to military disputes between nation states (Fearon 1995, Muthoo 1999, 2000, Powell 2002). There is a large literature on bargaining theory in economics which we will not attempt to review here; readers seeking an accessible introduction are directed to Muthoo (1999, 2000). Rather, we need only refer to a simple principle of bargaining theory, the “outside option principle” (Binmore 1985, Sutton 1986), to help solve the problem of how to synthesize models of reproductive skew. The outside option principle suggests that a focal individual’s outside option will influence the resolution of within-group conflict only if this option yields a greater payoff to the focal individual than the payoff it expects to receive through negotiation (Nash 1953). In these circumstances the individual can use a credible threat to exercise its outside option to obtain a more favorable resolution, one that is sufficient to render the threat incredible. The principle chimes with our everyday intuition about bargaining situations. Imagine, for example, a bartender earning $10 per hour who demands a raise from his employer on the grounds that the bar across the street is offering an hourly rate of $9. Clearly, a rational employer will be left unmoved by this threat of departure. Only if the other bar is offering more than $10 per hour will the employer need to take the threat seriously, and decide whether it is worth raising her own offer to retain the worker. Threats to take action that both parties know would be self-defeating are not credible, and so should exert no influence on the resolution of conflict (Nash 1953, Binmore 1985, McNamara & Houston 2002). The outside option principle suggests that the models of Johnstone (2000) and Reeve & Shen (2006) encounter problems in part because they assume that players always concede shares of reproduction to one another, even under circumstances in which the threat of eviction and the threat of departure prove incredible. This conclusion was also reached independently by P. Buston and A. Zink, in an unpublished manuscript sent to us while this chapter was being written. In the next section, we apply the outside option principle to resolve these problems, and determine when and how the possibilities of departure and of eviction influence behavior in a tug-of-war over reproduction. The key difference between our approach and that of previous synthetic models is that, rather than start with the assumption that an equilibrium
Models of reproductive skew features concession of uncontested shares of reproduction, we first focus on the resolution of the tug-of-war, and then use this to determine whether concessions are necessary. The model We focus on the interaction between two individuals, one dominant and one subordinate, who are related by a coefficient r. Dominance in this context is defined by the ability to evict the subordinate from the local breeding territory, although we also allow for the possibility that the dominant may enjoy an advantage over the subordinate in competing for reproductive opportunities within the local territory. In the extreme, the dominant may enjoy complete control over reproduction. The interaction involves two stages. In the first stage, both players simultaneously choose how much effort to invest in selfish competition over reproductive opportunities within the group. This stage is identical to the “tug-of-war” game of Reeve et al. (1998). The levels of effort will be denoted x for the dominant and y for the subordinate. Following the first stage, both players simultaneously choose whether to remain in association and accept the payoffs from the tug-ofwar, or whether to terminate the interaction. If either player chooses to terminate the interaction, then we assume that the dominant retains control of the breeding territory and thus receives a direct fitness payoff of 1 (baseline productivity for a lone breeder), while the subordinate receives a direct fitness payoff of d (1), reflecting the cost or risk associated with locating alternative breeding opportunities. The “outside options” available to the players are thus departure for the subordinate, and eviction for the dominant (though when d ¼ 1 both players are indifferent as to who leaves and who stays). Evaluation of outside options
The direct fitness payoffs to the two players from the tug-of-war, if both remain in association, depend upon the effort levels of both, and are given by x x þ by by Ws ðx; yÞ ¼ kð1 x yÞ x þ by
Wd ðx; yÞ ¼ kð1 x yÞ
ð1:10Þ
In the second stage of the game, therefore, the subordinate stands to gain by departing if and only if ðd Ws ðx; yÞÞ þ rð1 Wd ðx; yÞÞ > 0
ð1:11Þ
13
14
R. A. Johnstone, M. A. Cant while the dominant stands to gain by evicting the subordinate if and only if ð1 Wd ðx; yÞÞ þ r ðd Ws ðx; yÞÞ > 0
ð1:12Þ
If neither condition is met, then both players do best to remain in association. Under these circumstances, neither the threat of departure nor the threat of eviction is credible. Stable levels of competitive effort
How do the threats of departure and eviction in the second step of the game affect the players’ choices of competitive effort during the first step? The (unique) stable pair of effort levels x* and y* in the basic tug-of-war game (derived by Reeve et al. 1998) were given above. If the resulting direct fitness payoffs to both players are great enough that neither condition (1.11) nor condition (1.12) is satisfied for effort levels x* and y*, then the stable outcome of the game is unaffected by the possibility of departure or of eviction, because neither threat is credible given the outcome of the tug-of-war. Only if one or both of the players stands to gain by departing or evicting the other will the availability of outside options influence the resolution of the conflict. The “concession” zone
If ð1 Wd ðx ; y ÞÞ þ rðd Ws ðx ; y ÞÞ < 0 < ðd Ws ðx ; y ÞÞ þ rð1 Wd ðx ; y ÞÞ
ð1:13Þ
then the dominant stands to gain from the association, while the subordinate does best (given effort levels x* and y* ) to depart. Under these circumstances, the dominant must “ease off ” in the tug-of-war if it is to retain the subordinate in the group. This potentially leads to what we shall call a “concession” equilibrium, at which the stable effort levels of the two players, denoted xc and yc, satisfy @Ws ðx; yÞ @Wd ðx; yÞ þr ¼ 0 for x ¼ xc ; y ¼ yc @y @y
ð1:14Þ
ðd Ws ðxc ; yc ÞÞ þ r ð1 Wd ðxc ; yc ÞÞ ¼ 0
ð1:15Þ
and
Equation 1.14 implies that the subordinate does not stand to gain from a change in effort level, because this would entail a net decrease in the inclusive fitness payoff from the tug-of-war. Equation 1.15 implies that the dominant
Models of reproductive skew does not stand to gain from an increase in its effort, because this would trigger departure by the subordinate, which obtains an inclusive fitness from the tugof-war that is just sufficient to make departure unprofitable. Such an outcome will, however, only prove stable provided that ð1 Wd ðxc ; yc ÞÞ þ rðd Ws ðxc ; yc ÞÞ < 0
ð1:16Þ
If this condition is not met, then the dominant will not be selected to reduce its competitive effort in the tug-of-war sufficiently to retain the subordinate, since it would do better simply to evict its competitor. The “restraint” zone
If ð1 Wd ðx ; y ÞÞ þ r ðd Ws ðx ; y ÞÞ > 0 > ðd Ws ðx ; y ÞÞ þ rð1 Wd ðx ; y ÞÞ
ð1:17Þ
then the subordinate stands to gain from the association, while the dominant does best (given effort levels x* and y* ) to evict its competitor. Under these circumstances, the subordinate must “ease off ” in the tug-of-war if it is to be allowed to remain in the group. This potentially leads to what we shall call a “restraint” equilibrium, at which the stable effort levels of the two players, denoted xr and yr, satisfy @Wd ðx; yÞ @Ws ðx; yÞ þr ¼ 0 for x ¼ xr ; y ¼ yr @x @x
ð1:18Þ
ð1 Wd ðxr ; yr ÞÞ þ r ðd Ws ðxr ; yr ÞÞ ¼ 0
ð1:19Þ
and
Equation 1.18 implies that the dominant does not stand to gain from a change in effort level, because this would entail a net decrease in the inclusive fitness payoff from the tug-of-war. Equation 1.19 implies that the subordinate does not stand to gain from an increase in its effort, because this would trigger eviction by the dominant, which obtains an inclusive fitness from the tug-of-war that is just sufficient to make eviction unprofitable. Such an outcome will, however, only prove stable provided that ðd Ws ðxr ; yr ÞÞ þ rð1 Wd ðxr ; yr ÞÞ < 0
ð1:20Þ
If this condition is not met, then the subordinate will not be selected to reduce its competitive effort in the tug-of-war sufficiently to be tolerated by the dominant, since it would do better simply to leave.
15
16
R. A. Johnstone, M. A. Cant Group breakup
If the solution of the original tug-of-war model is unstable because one or other of the players stands to gain by exercising their outside option, but at the same time neither a concession equilibrium nor a restraint equilibrium is feasible, then we expect that the association will break up. An equilibrium at which x < x* and y < y*, i.e. at which both players simultaneously “ease off ” in the tug-of-war in order to maintain group stability, is not possible. If both players stand to gain from the outcome of the basic tug-of-war, then neither the threat of departure nor the threat of eviction is credible, so that neither player needs to adjust its effort level in response. Conversely, if neither player stands to gain from the outcome of the basic tug-of-war, then neither has any incentive to “ease off ” in order to maintain an association that is unprofitable to both. Results
We have derived analytical solutions for the boundaries of the regions of parameter space in which one obtains the different outcomes described above. Since these expressions are in some cases complex, we will not give them here; instead, Figures 1.1 and 1.2 show graphically these regions of parameter space, for the case of unrelated (r ¼ 0) and related (r ¼ 0.5) competitors, respectively. The general qualitative pattern is simple: groups are more likely to break up when productivity (k) is low, the opportunity for independent breeding (d) is great, and competitors are less closely related. Breakup is also more likely for an intermediate level of asymmetry in competitive ability between dominant and subordinate. There is a substantial region in which both the threat of departure and the threat of eviction prove incredible, so that the solution of the basic tug-of-war game is stable and unaffected by outside options. This outcome is most likely when group productivity is high, there is little opportunity for independent breeding, and competitors are more closely related. Once again, it is also more likely for an intermediate level of asymmetry in competitive ability between dominant and subordinate. When the subordinate is much weaker than the dominant, and particularly when there are substantial opportunities for independent breeding and the competitors are unrelated, the threat of subordinate departure becomes credible. Under these circumstances we obtain a “concession” equilibrium at which the dominant “eases off ” in competition to retain the subordinate in the group. Conversely, when the subordinate is not too much weaker than the dominant, and again particularly when there are substantial opportunities for independent breeding and the competitors are unrelated, the subordinate may have to “ease off ” in competition for its presence to be tolerated.
Models of reproductive skew d = 0.8
d = 0.6
4
4 No threats Dom concession
No threats
Sub restraint
3.5 Group productivity (k)
Group productivity (k)
3.5
3
Group breaks up
2.5
2
1.5
0.6 0.8 0.4 Subordinate strength (b)
2
Group breaks up
0.2
0.6 0.8 0.4 Subordinate strength (b)
No threats
3.5
3 Sub restraint 2.5
2
Group breaks up
1
d = 0.2
4
Group productivity (k)
Group productivity (k)
2.5
1
d = 0.4
4
1.5
Sub restraint
1.5
0.2
3.5
Dom concession 3
No threats
3
2.5
Sub restraint
2
1.5 Group breaks up
0.2
0.6 0.8 0.4 Subordinate strength (b)
1
0.2
0.6 0.8 0.4 Subordinate strength (b)
1
Figure 1.1 The graphs show, for unrelated players (r ¼ 0), the regions of parameter space in which: (i) the solution of the basic tug-of-war game proves stable because both the threat of departure and the threat of eviction are incredible (labeled “no threats”); (ii) the dominant must reduce its competitive effort to prevent departure of the subordinate (labeled “Dom concession”); (iii) the subordinate must reduce its competitive effort to prevent eviction by the dominant (labeled “Sub restraint”); (iv) the association proves unstable. In each graph, the relative competitive ability of the subordinate (b) increases from left to right along the horizontal axis, and group productivity (k) increases from bottom to top along the vertical axis. Different graphs give results for different levels of opportunity for independent breeding (d): high opportunity in the top-left graph, and low opportunity in the bottom-right graph.
Implications for skew
What is the significance of these different solution regions? It matters in which region a population or species falls, because depending on the nature of the outcome, patterns of skew are expected to be very different. When
17
R. A. Johnstone, M. A. Cant d = 0.8
4
3.5 Group productivity (k)
Group productivity (k)
No threats 3
2.5 Dom concession 2 Group breaks up
1.5
0.2
0.4 0.6 0.8 Subordinate strength (b)
No threats
3
2.5
2
Group breaks up
0.2
1
Sub restraint
Dom concession
1.5
d = 0.4
4
0.6 0.8 0.4 Subordinate strength (b)
1
d = 0.2
4
3.5 Group productivity (k)
3.5
3
d = 0.6
4
3.5
Group productivity (k)
18
No threats
2.5
2
3 No threats 2.5
2
Sub restraint Sub restraint
1.5
1.5 Group breaks up
Group breaks up 0.2
0.6 0.8 0.4 Subordinate strength (b)
1
0.2
0.6 0.8 0.4 Subordinate strength (b)
1
Figure 1.2 Regions of parameter space as in Figure 1.1, for related players (r ¼ 0.5).
neither the threat of departure nor the threat of eviction is credible, outside options, as we have seen, do not come into play. Under these circumstances, neither ecological constraints nor the opportunity for independent breeding nor group productivity exert any influence whatsoever on the level of skew within a group. As mentioned previously, relatedness also has little influence on stable skew in the tug-of-war game, so that in this region the relative competitive ability of the subordinate will be the only significant influence on skew (with stronger subordinates gaining a greater share of reproduction). Outside options typically come into play, as described above, when the subordinate is either very weak or very strong, group productivity is low, relatedness is low, and there are opportunities for independent breeding. In
Models of reproductive skew these “concession” and “restraint” regions, group productivity, relatedness, and the opportunity for independent breeding will influence the level of skew in a group, just as in simple transactional models of skew. In the concession region, where the dominant is forced to “ease off ” in competition in order to retain the subordinate, the latter’s share of reproduction will increase with the opportunity for independent breeding and decrease with group productivity and relatedness. By contrast, in the restraint region, where it is the threat of eviction that comes into play, the subordinate’s share of reproduction will decrease with the opportunity for independent breeding and increase with group productivity. Lastly, across both concession and restraint regions, the subordinate’s share increases with its competitive ability (just as in the basic tug-of-war).
Discussion Taking account of the “outside option principle” leads to a synthetic model of skew that does not suffer from the stability problems of previous analyses. The threat of departure and the threat of eviction come into play only when they are credible; moreover, they lead to players “easing off ” in competition, rather than exchanging uncontested shares of reproduction. The question of how individuals could possibly “agree” to refrain from competing over a proportion of the available reproductive opportunities does not, therefore, arise. The results suggest a rather different view of reproductive skew from previous attempts at synthesis. While some have argued for the universal scope and power of transactional (particularly concession) models, and their potential for unifying the study of social evolution (e.g. Reeve & Keller 2001, Buston et al. 2007), the present analysis suggests that they are relevant only under certain restricted circumstances. Where group members are closely related and/or the productivity benefits of association are great, the threat of departure or of eviction is likely to prove incredible, because the direct and indirect benefits of group membership are too large to forgo. Under these circumstances, the whole basis of transactional skew models is eliminated. No concessions need be offered, because there is no opportunity for either player to gain by dissolving the association. Consequently, manipulating outside options is not expected to influence skew (see, e.g., Langer et al. 2004, Heg et al. 2006). Rather strikingly, these circumstances of high relatedness and high group productivity are precisely those in which selection is most likely to favor association in the first place. Consequently, the most obviously beneficial interactions are those least likely to be amenable to analysis in terms of skew theory.
19
20
R. A. Johnstone, M. A. Cant Should we then abandon skew models completely? We would argue not. The burgeoning literature on biological markets and partner choice (Noe¨ et al. 1991, Noe¨ & Hammerstein 1994, Bshary & Noe¨ 2003, Sachs et al. 2004, Foster & Wenseleers 2006) attests to a growing awareness among biologists that outside options are often important. Although most of the classical applications of game theory in biology focus on “forced play” among a small number of players, there are many situations in which one individual is not forced to interact with another, but can readily switch partners or perhaps forgo any interaction at all. Under these circumstances, it is not surprising that the threat of terminating an interaction and losing a partner should exert a significant influence on behavior within the association. Such “sanctions” have become the focus of much discussion in the study of mutualism and cooperation (see, e.g., Herre et al. 1999, Ferrie`re et al. 2002, Johnstone & Bshary 2002, West et al. 2002, Bshary & Noe¨ 2003, Kiers et al. 2003, Sachs et al. 2004, Foster & Wenseleers 2006), and an elegant experimental demonstration is provided by studies of interaction between cleaner-fish and clients, in which the threat of departure by the client induces cleaners to refrain from biting live tissue (Bshary & Grutter 2002, 2005). Although this example may seem rather remote from reproductive skew, it illustrates the same principle that lies at the heart of skew theory – that individuals’ prospects outside a given association can influence the resolution of conflicts inside it. The key feature of biological “markets” that renders the threat of departure (or rejection) credible is the ready availability of alternative partners. We suggest, accordingly, that transactional skew theory is likely to prove most relevant not to highly profitable associations involving small numbers of closely related individuals, who may have little opportunity to join a similar family group elsewhere, but to looser and less cooperative associations between more distantly related individuals. Individuals in such cases may have opportunities to move between alternative, unrelated groups, so that the threat of departure may become credible, just as the threat of eviction may when group members can be easily replaced or contribute little to productivity in any case. Indeed, it is surprising that, apart from some work by Reeve (1998), there has so far been little attempt to incorporate market effects explicitly into skew theory. The response of other group members to the prospect of departure or an attempt at eviction need not, as our model shows, involve the “exchange” of uncontested shares of reproduction, with all of the questions this raises about how animals might achieve such a feat, and what prevents cheating. Individuals may simply reduce their competitive efforts in response to the risk of triggering the breakup of a group. To detect such influences is likely to be
Models of reproductive skew difficult, because an effective threat is precisely one that elicits responses that make it unprofitable to carry out. Consequently, social behavior might potentially be influenced by many “invisible” threats that remain hidden until the “rules” they enforce are broken. However, threats of this kind can be exposed by experimentally staging such violations. Wong et al. (2007), for instance, have shown that the typical size hierarchy seen in groups of the coral-dwelling goby Paragobidon xanthosomus is maintained by the latent threat of eviction: when dominant fish were experimentally paired with subordinates larger than observed under natural circumstances, eviction was the result (while dominants tolerated individuals who were smaller than themselves). Conclusion To sum up, we suggest that outside options can influence the resolution of reproductive conflict within a group, but that they will do so only when the threat of departure or of eviction is credible. Consequently, tests which examine the relationship between skew and factors such as group productivity or ecological constraints are expected to yield variable results. The essential question is whether or not any of the members of a group stand to gain from its dissociation. The answer will determine whether or not outside options come into play.
References Binmore, K. (1985). Bargaining and coalitions. In A. Roth, ed., Game Theoretic Models of Bargaining. Cambridge: Cambridge University Press, pp. 269–302. Bshary, R. & Grutter, A. S. (2002). Experimental evidence that partner choice is the driving force in the payoff distribution among cooperators or mutualists: the cleaner fish case. Ecology Letters, 5, 130–136. Bshary, R. & Grutter, A. S. (2005). Punishment and partner choice cause cooperation in a cleaning mutualism. Biology Letters, 1, 396–399. Bshary, R. & Noe¨, R. (2003). Biological markets: the ubiquitous influence of partner choice on the dynamics of cleaner fish–client reef fish interactions. In P. Hammerstein, ed., Genetic and Cultural Evolution of Cooperation. Cambridge, MA: MIT Press, pp. 167–184. Buston, P. M., Reeve, H. K., Cant, M. A., Vehrencamp, S. L., & Emlen, S. T. (2007). Reproductive skew and the evolution of group dissolution tactics: a synthesis of concession and restraint models. Animal Behaviour, 74, 1643–1654. Cant, M. A. (2000). Social control of reproduction in banded mongooses. Animal Behaviour, 59, 147–158. Cant, M. A. & Johnstone, R. A. (1999). Costly young and the partitioning of reproduction in animal societies. Behavioral Ecology, 10, 178–184.
21
22
R. A. Johnstone, M. A. Cant Clutton-Brock, T. H. (1998). Reproductive skew, concessions and limited control. Trends in Ecology Evolution, 7, 288–292. Fearon, J. D. (1995). Rationalist explanations for war. International Organization, 49, 379–414. Ferrie`re, R., Bronstein, J. L., Rinaldi, S., Law, R., & Gauduchon, M. (2002). Cheating and the evolutionary stability of mutualisms. Proceedings of the Royal Society of London B, 269, 773–780. Foster, K. R. & Wenseleers, T. (2006). A general model for the evolution of mutualisms. Journal of Evolutionary Biology, 19, 1283–1293. Haydock, J. & Koenig, W. D. (2002). Reproductive skew in the polygynandrous acorn woodpecker. Proceedings of the National Academy of Sciences of the USA, 99, 7178–7183. Heg, D., Bergmuller, R., Bonfils, D., et al. (2006). Cichlids do not adjust reproductive skew to the availability of independent breeding options. Behavioral Ecology, 17, 419–429. Herre, E. A., Knowlton, N., Mueller, U. G., & Rehner, S. A. (1999). The evolution of mutualisms: exploring the paths between conflict and cooperation. Trends in Ecology and Evolution, 14, 49–53. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Johnstone, R. A. & Bshary, R. (2002). From parasitism to mutualism: partner control in asymmetric interactions. Ecology Letters, 5, 634–639. Johnstone, R. A. & Cant, M. A. (1999). Reproductive skew and the threat of eviction: a new perspective. Proceedings of the Royal Society of London B, 266, 275–279. Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 12, 99–103. Kiers, E. T., Rousseau, R. A., West, S. A., & Denison, R. F. (2003). Host sanctions and the legume–rhizobium mutualism. Nature 425, 78–81. Kokko, H. & Lindstro¨m, J. (1997). Measuring mating skew. American Naturalist, 149, 794–799. Langer, P., Hogendoorn, K., & Keller, L. (2004). Tug-of-war over reproduction in a social bee. Nature 428, 844–847. McNamara, J. M. & Houston, A. I. (2002). Credible threats and promises. Philosophical Transactions of the Royal Society of London B, 357, 1607–1616. Muthoo, A. (1999). Bargaining Theory with Applications. Cambridge: Cambridge University Press. Muthoo, A. (2000). A non-technical introduction to bargaining theory. World Economics, 1, 145–166. Nash, J. (1950). The bargaining problem. Econometrica, 18, 155–162. Nash, J. (1953). Two-person cooperative games. Econometrica, 21, 128–140. Noe¨, R. & Hammerstein, P. (1994). Biological markets: supply and demand determine the effect of partner choice in cooperation, mutualism and mating. Behavioral Ecology and Sociobiology, 35, 1–11.
Models of reproductive skew Noe¨, R., van Schaik, C. P., & van Hooff, J. A. R. A. M. (1991). The market effect: an explanation for payoff asymmetries among collaborating animals. Ethology, 87, 97–118. Nonacs, P. (2007). Tug-of-war has no borders: it is the missing model in reproductive skew theory. Evolution, 61, 1244–1250. Osborne, M. & Rubinstein, A. (1990). Bargaining and Markets. San Diego, CA: Academic Press. Powell, R. (2002). Bargaining theory and international conflict. Annual Review of Political Science, 5, 1–30. Reeve, H. K. (1998). Game theory, reproductive skew and nepotism. In L. A. Dugatkin & H. K. Reeve, eds., Game Theory and Animal Behaviour. Oxford: Oxford University Press, pp. 118–145. Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model. Behavioral Ecology, 11, 640–647. Reeve, H. K. & Keller, L. (2001). Tests of reproductive-skew models in social insects. Annual Review of Entomology, 46, 347–385. Reeve, H. K. & Ratnieks, F. L. W. (1993). Queen–queen conflict in polygynous societies: mutual tolerance and reproductive skew. In L. Keller, ed., Queen Number and Sociality in Insects. Oxford: Oxford University Press, pp. 45–85. Reeve, H. K. & Shen, S. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Sachs, J. L., Mueller, U. G., Wilcox, T. P., & Bull, J. J. (2004). The evolution of cooperations. Quarterly Review of Biology, 79, 135–160. Sutton, J. (1986). Non-cooperative bargaining theory: an introduction. Review of Economic Studies, 53, 709–724. Tsuji, K. & Kasuya, E. (2001). What do the indices of reproductive skew measure? American Naturalist, 158, 155–165. Vehrencamp, S. L. (1979). The roles of individual, kin and group selection in the evolution of sociality. In P. Marler & J. G.Vandenbergh, eds., Handbook of Behavioural Neurobiology 3. Social Behaviour and Communication. New York, NY: Plenum Press, pp. 351–394. Vehrencamp, S. L. (1983). Optimal degree of skew in reproductive societies. American Zoologist, 23, 327–335. West, S. A., Kiers, E. T., Pen, I., & Denison, R. F. (2002). Sanctions and mutualism stability: when should less beneficial mutualists be tolerated? Journal of Evolutionary Biology, 15, 830–837. Wong, M. Y. L., Buston, P. M., Munday, P. L., & Jones, G. P. (2007). The threat of punishment enforces peaceful cooperation and stable queues in a coral-reef fish. Proceedings of the Royal Society of London B, 274, 1093–1099.
23
2
Reproductive conflict and the evolution of menopause m i c h a e l a. c a n t , r u f u s a. j o h n s t o n e , and andrew f . r usse ll
Summary Human females (Homo sapiens) exhibit a dramatic form of reproductive skew in which half the age classes of adults contain only nonbreeders. Among other mammals, only pilot (Globicephala spp.) and killer whales (Orcinus orca) exhibit a similar pattern. The “grandmother” hypothesis suggests that selection can favor post-reproductive survival because older females help their offspring to reproduce. But the indirect fitness gains of helping appear insufficient to outweigh the potential benefits of continued direct reproduction, so this hypothesis cannot explain why women cease reproducing in the first place. Here we present some background on menopause and describe new research which helps to understand both the strange taxonomic distribution of menopause and the timing of reproductive cessation in humans. Specifically, recent models have explored the potential reproductive conflicts that may have arisen in ancestral human families, and the influence of demography on the resolution of such conflicts. These studies suggest that an integrated model which takes into account the potential costs of reproductive competition, as well as the benefits of helping, offers a fuller understanding of the evolution of menopause. Tabar ne maiet hate kana jane bakariyon, lardiyon jyoon! (How unbecoming of parents to procreate alongside their children like goats and sheep!) Saying of the Mogra, Rajasthan, India (quoted in Patel 1994) Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
24
Reproductive conflict and the evolution of menopause Reproductive skew in human societies Human societies are characterized by a dramatic and puzzling pattern of reproductive skew. In populations exposed to natural schedules of mortality and fertility (i.e. without access to modern medicine and technology), almost half the age classes of adult human females contain only nonbreeders (Figure 2.1). The mean ages at which women give birth to their last child in natural-fertility populations cluster around 39 years (Wood 1994, p. 442), but even in hunter–gatherer societies that lack modern medicine women who reach this age can expect to live well into their sixties (Pennington 2001, Blurton Jones et al. 2002). The restriction of reproduction to certain age classes is not in itself unusual for a cooperative vertebrate, but in other species it is almost always older females who breed and younger females who do not (Emlen 1991, 1995). The reverse pattern exhibited by humans is extremely rare – among vertebrates only killer whales and pilot whales are reported to exhibit a similar reversal of breeding roles with respect to age class (Marsh & Kasuya 1986, Olesiuk et al. 1990, Whitehead & Mann 2000). Early reproductive cessation represents an evolutionary puzzle because standard life-history theory suggests that there should be no selection for somatic
Figure 2.1 Survival and fecundity in a natural-fertility human population. Data from a Taiwanese population in 1906 (redrawn from Hamilton 1966).
25
26
M. A. Cant et al. maintenance after the end of reproduction. Why then do women cease reproducing so long before they die? Despite almost 50 years of research on the evolution of menopause, this important question remains open. Current models invoke the kin-selected benefits of helping as a grandmother to explain post-reproductive survival in women, but, as we describe below, quantitative analyses suggest that these models cannot explain why women stop breeding at the time they do. Part of the problem is that current models focus solely on the direct fitness consequences of reproduction, and compare this with the indirect fitness consequences of helping. Helping is assumed to affect the fitness of other group members, while breeding is not. This approach is one-sided because it ignores the potential impact of reproduction on the fitness of other group members. Where there are limited resources within a group for reproduction, the decision to reproduce will depend on whether other females in the group will also reproduce, how many young they will produce, and how one’s own young will fare in competition if they do. Reproduction in a social context, therefore, is a game-theoretic rather than an optimization problem. Reproductive skew theory was developed to study exactly this type of problem, and so is an apt framework within which to study the evolution of patterns of reproduction in humans. In this chapter we take a fresh look at the puzzle of menopause by examining the potential reproductive conflicts in ancestral human societies, and the way in which these conflicts are likely to be resolved. We first review the main adaptive explanations for menopause, the “mother” and “grandmother” hypotheses, and highlight the empirical and theoretical difficulties that these hypotheses have encountered. We then describe results from our own recent research, which focuses on the impact of demography on kin selection across individual lifespans, and how this will affect the resolution of conflicts over reproduction within human social groups. Our aim is to show that these new models offer a fuller explanation for the pattern and timing of reproductive cessation in humans, and help to explain why, of all long-lived, social mammals, it is specifically among the lineages of great apes and toothed whales that menopause has evolved.
How old is menopause? A possible non-adaptive explanation for menopause is that it is a simple artifact of the reduction in mortality that followed agriculture and improved sanitation (reviewed by Peccei, 2001a, 2001b). The idea is that the reproductive lifespan of women reflects the expected female lifespan prior to these technological developments, but there has been insufficient time for
Reproductive conflict and the evolution of menopause selection to extend the reproductive period to match the newly elongated lifespan. If this hypothesis were correct, menopause should be absent in hunter–gatherer populations without agriculture or modern medicine because women would rarely survive beyond the age of 50. On the contrary, in the three best-studied hunter–gatherers (the !Kung of the Kalahari, Ache of Paraguay, and Hadza of Tanzania) a large fraction of women survive to postmenopausal age. For example, 64% of non-nomadic !Kung, 46% of Hadza, and 42% of Ache women live until age 50 or more (Pennington 2001). Moreover, women who survive to 45 can expect an average of 20 or more years of life thereafter (Pennington 2001, Blurton Jones et al. 2002). A pattern of menopause coupled with prolonged post-reproductive life can be inferred from ancient texts: the Bible (Psalms 90: 10) refers to an expected lifespan of 70 (the familiar “threescore years and ten”), rising to 80 years “by reason of strength”, while Aristotle (c. 360 BC) and Pliny (c. AD 77) cite an age at menopause of around 50 years (Amundsen & Diers 1970). Finally, a recent analysis of fossil molar wear (Caspari & Lee 2004) suggests that the fraction of humans surviving to become grandparents increased five-fold between the Middle and Upper Paleolithic (c. 300 000–10 000 years ago), i.e. before the emergence of agriculture. The evidence suggests, therefore, that menopause has been a feature of the life history of human females for at least the last 10 000 or 20 000 years, and possibly much longer.
Reproductive senescence Menopause is best viewed as the endpoint of unusually rapid senescence of the reproductive system relative to somatic systems. Senescence, the general decline in efficiency of bodily functions with age, is an inescapable property of both somatic and reproductive systems in iteroparous organisms. This is because random mortality ensures that older individuals always make up a smaller fraction of the breeding population than younger individuals, so genes which have negative effects on reproduction or survival early in life are more strongly opposed by selection than are genes with negative effects later in life (Medawar 1952, Williams 1957, Hamilton 1966). In theory, senescent decline should strike all body functions at a similar rate, since if one system or organ (for example, the cardiovascular system, or the renal system) declined much more rapidly than others, selection to maintain other systems would weaken, accelerating their rate of decline to match that of the most rapidly senescent (Williams 1957). Consequently, the capacity for reproduction is predicted to decline in tandem with, and at a similar rate to, other somatic systems. In most organisms this expectation is borne out: fertility in old age, like other
27
28
M. A. Cant et al. functions, is much reduced but reproduction is nevertheless still possible (Rose 1991). What in humans (and some whales) could cause the rate of reproductive senescence to become decoupled from the rate of somatic senescence? To answer this question it is not sufficient to invoke the personal costs of breeding at late ages, since these costs are themselves an evolved property of the system. Increases in, say, the rate of birth defects, or stillbirth, that occur after the age of 40 in humans are a result of, rather than a cause of, the rapid senescence of the reproductive system. Menopause marks the end of a process of rapid reproductive senescence that begins a decade earlier, and one must be careful not to invoke the effects of this rapid senescence in order to explain it. Rather, we need to consider the potential benefits of early reproductive cessation in an ancestral, non-menopausal hominid species in which fertility declined at the same rate as other bodily functions.
Adaptive explanations Two closely related adaptive explanations for the evolution of menopause are known as the “mother” and “grandmother” hypotheses. These hypotheses differ primarily in whether menopause is assumed to boost the survival or the fertility of existing young. Williams (1957) suggested that early reproductive cessation is a consequence of the long period of offspring dependency in humans. According to this hypothesis, older women may at some point gain from ceasing reproduction to invest in raising their existing children to adulthood, rather than engaging in increasingly risky breeding attempts which could leave their dependent offspring motherless (Peccei 2001a, 2001b). By contrast, other authors have emphasized the benefits of menopause for enhancing the fertility of a woman’s existing children (Hamilton 1966, Alexander 1974, Hawkes et al. 1998). An older woman whose fertility is declining due to senescence may at some point do best to switch resources from her own breeding attempts to helping to rear grandchildren. In both cases, beyond the age at which helping is more profitable than breeding there is no selection to maintain fertility, and the rate of senescence of the reproductive system is expected to increase sharply. The result is a decoupling of the rates of somatic and reproductive senescence, and a pattern of reproductive cessation long before death. Empirical data offer little support for the mother hypothesis. The chance of dying in childbirth must be very large for females to prefer reproductive cessation over continued reproduction. We can illustrate this with a simple numerical example. Consider a female at age 40 with four dependent offspring, faced with a decision of whether to cease reproduction or produce one
Reproductive conflict and the evolution of menopause more child. Let d be the chance that she dies in childbirth, which we will assume leads to the certain death of all her existing young. In this example, reproductive cessation will be favored over continued breeding if 4 > 5(1 d), or d > 20%. In reality, the chance of dying in childbirth is minuscule even among hunter–gatherers (e.g. around 1/150 in the Ache: Hill & Hurtado 1996). Data from the United States in the late nineteenth and early twentieth centuries suggest that even for mothers who gave birth at age 45–50, the risk of dying in childbirth was around 5% (Loudon 1993). Quantitative analyses using data from natural-fertility populations have concluded that the risk of death in childbirth is far too small to account for the evolution of menopause (Rogers 1993, Hill & Hurtado 1996). It is tempting to invoke other factors which might devalue late-life reproduction, such as elevated rates of fetal wastage and birth defects, but again it is important to remember that these factors are themselves the outcome of selection for rapid reproductive senescence. Investigations of the grandmother hypothesis are more numerous (see Voland et al. 2005). Data from modern hunter–gatherers and historical populations provide evidence that grandmothers can indeed boost the reproductive success of their children. Significant positive effects of grandmothers on grandchild survival have been reported in six out of seven studies of grandmothering in pre-medical societies (Lahdenpera¨ et al. 2004a, 2004b, Mace & Sear 2005). However, while these studies demonstrate a potential kin-selected benefit of grandmotherhood, quantitative analyses suggest that these benefits are not sufficiently large to explain the evolutionary maintenance of menopause at the age at which it occurs (Rogers 1993, Hill & Hurtado 1996). Hill & Hurtado (1991, 1996), for example, use data from one well-studied hunter– gatherer society (the Ache of Paraguay) to calculate the inclusive fitness payoffs of grandmothering versus continued reproduction for older women. To estimate the latter payoff they assume that in the absence of menopause fertility would decline at the same rate as somatic function. Their calculations suggest that menopause cannot be favored by kin selection in this population because there are few close kin alive for an older woman to help, and because her help has too little impact on the survival or reproduction of these kin. Rogers (1993) uses a different model and dataset (the Taiwan 1906 data used in Hamilton’s [1966] classic paper on senescence) but shows a similar result, namely that the impact of an older woman’s help on her close kin must be more substantial than has been so far documented for menopause to be favored by kin selection. Recently, however, Shanley & Kirkwood (2001) presented a model which they claim can account for the evolution of menopause around the age of 50 or 55, again using the Taiwan 1906 dataset. Unfortunately, their model uses the intrinsic rate of increase of the population as the measure of fitness to be
29
30
M. A. Cant et al. maximized by natural selection. This means that offspring and grandoffspring are counted as equally “valuable” to an older female faced with the decision of whether to continue reproduction versus help as a grandmother. As Hamilton (1966) demonstrated, selection for helping versus breeding will depend not only on the number of young produced as a result of helping or breeding, but on the relatedness to these offspring. A mother’s relatedness to her own offspring is twice that to her grandoffspring, so the fitness payoff of helping to raise an extra grandchild is, other things being equal, half that of producing another child herself. Shanley & Kirkwood’s (2001) model does not take into account relatedness, and so overestimates the fitness benefits of grandmothering by a factor of around two compared to the analyses of Hill & Hurtado (1996) and Rogers (1993). This may explain why Shanley & Kirkwood’s analysis can account for the evolution of menopause at around 50 whereas the other analyses cannot.
Physiological constraints and phylogenetic inertia While women can clearly gain fitness benefits by grandmothering, these fitness benefits are insufficient to account for the timing of reproductive cessation in human women or the evolution of menopause. The problems raised by these quantitative studies can be circumvented, however, if we assume that timing of menopause is a phylogenetic artifact or reflects some form of physiological constraint. Hawkes and co-workers (Hawkes et al. 1998, Hawkes 2003) note that the endpoint of reproductive senescence in human females occurs at the same age as in chimpanzees (Pan troglodytes), i.e. in the fifth decade of life. Consequently, they argue, it is the extended post-reproductive life of human females, not the timing of menopause, that is the derived trait to be explained (Hawkes et al. 1998). Grandmothering effects are invoked to explain the extension of the female lifespan long past the end of the phylogenetically conserved age at reproductive cessation (why males have similarly extended lifespans is not explained by this hypothesis, but see Marlowe [2000] for one perspective). This argument is unsatisfactory on its own because it assumes that stasis in the face of evolutionary change requires no special explanation. The conservation of patterns of reproductive senescence in the human lineage, despite lengthening lifespan, implies: (1) some physical or physiological constraint prohibiting evolutionary change; (2) an absence of genetic variation upon which selection can act; or (3) some form of stabilizing selection. Comparative evidence lends no support to the first possibility, i.e. that the reproductive lifespan of human females cannot be extended much past the age of 50 due to physiological constraints. Other long-lived mammals continue to breed until
Reproductive conflict and the evolution of menopause the end of life: African elephants (Loxodonta africana) reproduce in their sixties (Moss 2001) and blue whales (Balaenoptera musculus) into their nineties (Mizroch 1981). Across species, oocyte stocks are evolutionarily labile and are adjusted to lifespan and body weight (Gosden & Telfer 1987). In addition, the initial oocyte stock and rate of follicular attrition in human females is commensurate with a longer reproductive lifespan, but at around the age of 40 there is a marked increase in the follicular hazard rate so that by age 50 follicle stocks have dropped below a minimum required to sustain menstrual activity (Faddy et al. 1992, Faddy & Gosden 1996, Figure 2.2). By contrast, in laboratory rodents and rhesus macaques (Macaca mulatta) (the only other species for which similar data are available), there is no indication that the rate of follicular attrition increases later in life (Jones & Krohn 1961, Nichols et al. 2005).
Figure 2.2 The bi-phasic model of declining follicle numbers in pairs of human ovaries from neonatal age to 51 years old (modified from Faddy et al. 1992). Data are from four different autopsy studies. Note the logarithmic scale on the y-axis. A bi-exponential regression model offers a significantly better fit to the data than a single exponential regression. Follicle numbers decline at a constant exponential rate from birth until reaching a critical figure of around 25 000 at age 37.5 years, after which the exponential rate parameter increases in magnitude. Menopause occurs on average when a threshold of around 1000 follicles remain (Faddy et al. 1992).
31
32
M. A. Cant et al. Turning to the second possibility, age at menopause varies widely among individuals (with ages between 40 and 59 considered normal in both modern and natural-fertility populations), and estimates of the heritability of age at menopause range from 40% to 63% (Snieder et al. 1998, Peccei 1999). There would thus seem to be sufficient genetic variation on which natural selection could act, if prolonged fertility were advantageous. Moreover, recent evidence from a preindustrial Finnish population (Figures 2.3, 2.4) suggests that prolonged reproduction can have a substantial positive influence on a woman’s fitness (Helle et al. 2005), and that age at last reproduction is also highly heritable (Pettay et al. 2005). Taken together, this evidence suggests that any selection to extend the female’s reproductive span has been held in check by some form of opposing selection. Indeed, Hawkes (2003, p. 389) reaches a similar conclusion: Overall then, the available data on ages at last birth and menopause in chimpanzees show age-specific fertility declines in that species not substantially different from our own. Mammalian fertility, however, can extend to much older ages than it does in humans. This evidence is consistent with the argument that ancestral age-specific fertility declines have been maintained in our lineage, perhaps conserved by stabilizing selection.
Figure 2.3 Extended Lummaa family from nineteenth-century Finland showing several generations. Courtesy of Virpi Lummaa.
Reproductive conflict and the evolution of menopause
Figure 2.4 Photo of nineteenth-century nuclear Lummaa family from Finland with mother, father and all their children. Courtesy of Virpi Lummaa.
This stabilizing selection means that mutations for a later age at menopause must have been selected against, despite selection for a longer lifespan. We are thus back to our initial question: what is the nature of this selection? Why has the reproductive lifespan of human females not increased in line with their longevity, as it has in other long-lived mammals?
Reproductive competition: a new perspective on menopause We believe that previous models offer an incomplete account of the evolution of menopause because they focus solely on the kin-selected benefits of parenting and helping, and ignore the potential kin-selected costs of cobreeding. Conflict over reproduction in animal societies is expected wherever communal resources are used for the production of offspring, particularly where helpers increase the success if group members and breeders compete
33
34
M. A. Cant et al. for monopolization of those helpers. If two females produce young in the same group at the same time, each offspring will necessarily receive less food (unless twice as much food, or help, is available). Per capita success of young in a communal brood will therefore decrease with the number of young produced, as assumed in standard clutch-size theory (Lack 1947, Cant 1998, Cant & Johnstone 1999). Reproductive conflict is ubiquitous in other cooperative breeders in which there is more than one potential breeder per social group (Keller & Reeve 1994, Clutton-Brock 1998, Beekman et al. 2003, Ratnieks et al. 2006). Unlike other cooperative breeders, however, the possibility of reproductive competition in ancestral human families has been ignored. Recently, we have argued that the pattern and timing of reproductive cessation in humans is best understood as an adaptation to minimize the degree of reproductive competition between generations (Cant & Johnstone 2008). Certainly one of the consequences of the mean age at reproductive cessation in humans is that it leads to very low reproductive overlap between generations. Many primate species exhibit a post-reproductive lifespan, but there is nevertheless considerable overlap in the period for which females of older and younger generations are reproductively active. Figure 2.5A shows, for those primate species classified as exhibiting a post-reproductive lifespan (Cohen 2004), the relationship between generational overlap (calculated as the proportional overlap between the maximum lifespans of mother and daughter) and reproductive overlap (calculated as the proportional overlap between the mean reproductive spans of mother and daughter). Humans are not unusual in respect of their degree of generational overlap (77%, compared to 71% for chimpanzees, Pan troglodytes, and 73% for gorillas, Gorilla gorilla). However, they show an extraordinarily low degree of maximum reproductive overlap compared to other primates (30%, compared to 50% or more for all other species in the sample), far lower than would be expected on the basis of their generational overlap. On average, females from one generation stop breeding at just the time that the females in the next generation start to reproduce (Cant & Johnstone 2008; Figure 2.5B). If we used the regression line shown in Figure 2.5A to “reverse-engineer” the reproductive lifespan of human females, we would predict a mean age at last birth of 62, followed by menopause around 10 years later – an intriguing match to the predicted age at menopause obtained by extrapolating the constant rate of follicular attrition before the age of 40 (Figure 2.2). If the pattern of rapid reproductive senescence leading to low reproductive overlap in humans is a response to reproductive competition, why do none of the other primate species in Figure 2.5 show similar adaptations to the costs of co-breeding? Reproductive competition is expected where breeding resources are communal or the need for helpers is important and helpers are limiting.
A
1 Mouse lemur
Maximum reproductive overlap
0.9
Hanuman langur Rhesus macaque
0.8
Tamarin
Pig-tailed macaque Barbary macaque Japanese macaque
0.7
Olive baboon
0.6
Orangutan
0.5
Gorilla Chimpanzee
0.4 Human
0.3 0.5
0.6
0.7
0.8
0.9
1
Maximum generational overlap B
Orangutan
Japanese macaque 5
22
23
Chimpanzee 15
39 47
12
38 44
Human 19 38
82
Figure 2.5 Patterns of reproductive overlap in 12 primate species recently classed as exhibiting a post-reproductive lifespan (Cant & Johnstone 2008). Panel A shows maximum reproductive overlap versus maximum generational overlap. Maximum generational overlap is defined as (MLS AFB)/MLS, where AFB is average age at first birth and MLS is maximum recorded lifespan. Maximum reproductive overlap is defined as (MRS AFB)/MRS, where MRS is the maximum reproductive span, calculated as maximum age at last birth (MALB) minus AFB. For four species (chimpanzees, orangutans, Japanese macaques, and humans), published data are sufficiently detailed to calculate average reproductive overlap, defined as (ARS AFB)/ARS, where ARS is the average reproductive span (i.e. mean ALB minus AFB). Panel B summarizes the pattern of overlap for these four species. For each, horizontal bars represent the maximum lifespans of three successive generations, scaled to a standard length and offset in accordance with the value of AFB relative to MLS, with average reproductive spans shaded. The average reproductive overlap values for Japanese macaques, orangutans and chimpanzees were 0.71, 0.52, and 0.39 respectively, compared to an average reproductive overlap for humans of 0.00. For reference sources and values used to plot the figure see Cant & Johnstone (2008).
36
M. A. Cant et al. Compared to most other primates, such competition will be particularly intense in humans because they exhibit a unique degree of food sharing, and are cooperative breeders, in the sense that adults other than parents make important contributions to raising offspring (Emlen 1991, Mace & Sear 2005). In most other primates mothers look after their own young in the early years of life, provisioning them with milk during infancy, and helping them to gain experience in finding their own food – usually fruits and other plant matter. Mothers and their offspring are not reliant on shared food resources obtained by adult helpers, so co-breeders are likely to have little direct impact on each other’s reproductive success. Humans, by contrast, provision their children into adulthood and parents rely on the other group members to gather food. Food acquisition is divided between family members on the basis of sex and age class, and food collected or hunted by different group members is combined into a shared resource (Kaplan & Hill 1985, Gurven et al. 2004, Gurven 2005). Additional offspring within the same family will therefore draw on the same communal resource pool. Unless the presence of an extra female breeder somehow generates twice as much food for the group, co-breeding females will have fewer resources with which to raise their children than a female who is able to monopolize reproduction within the group. Unfortunately, the reproductive separation of generations in humans is so pronounced that it is difficult to obtain direct data on the costs of co-breeding. For example, in the natural-fertility Gambian dataset (for which data were collected between 1950 and 1974), only 5.6% of children (89/1588) had a reproductively active maternal grandmother when they were born, and there were no children born who had a reproductively active paternal grandmother (Sear et al. 2000); similar patterns also occur in pre-modern Western populations (V. Lummaa, personal communication). In many societies the reproductive separation of generations is further reinforced by cultural taboos. Among the Nyakyusa of Tanzania and some West African populations, women are required to stop breeding as soon as their first grandchild is born (Wilson 1957 p. 137, Cavalli-Sforza 1983), and similar proscriptions are found in some Asian societies (Patel 1994 p. 162 [quote at head of this chapter], Islam & Yadava 1997, Skinner, 2004). Though there are almost no data on the costs of co-breeding between generations, we can get some insight into the potential costs of co-breeding within a single family by examining offspring success under polygyny. Women in polygynous marriages (i.e. marriages in which two or more women are married to a single man) are on average less fertile than their monogamous counterparts (Isaac & Feinberg 1982, Pebley & Mbuga 1989). A large dataset from six West African countries suggests that, after controlling for socioeconomic and demographic factors, polygyny is associated with a 50%
Reproductive conflict and the evolution of menopause increase in neonatal mortality, and a doubling of post-neonatal child mortality (Amey 2002). These costs of polygyny to children have been attributed to crowding and disease transmission (Isaac & Feinberg 1982, Roth & Kurup 1988) and to reduced access to resources (Strassmann 1997) – two factors that would also apply to co-breeding between generations. While co-breeding is likely to involve costs, it remains puzzling why older women should choose to cease reproduction in the face of competition from younger women. In other animals, when two or more generations of females are present in a social group it is almost always older females that “win” the conflict over reproduction and retain breeding status, while younger females remain in the group as reproductively suppressed helpers. Why should humans be different? A critical factor is the unusual demography of humans (Cant & Johnstone 2008, Johnstone & Cant 2008). A “kinship dynamics” model suggests that the unusual dispersal and mating patterns of great apes, and some cetaceans, predispose these species to the evolution of early reproductive cessation and late-life helping behavior, in contrast to the majority of other mammalian species (Johnstone & Cant 2008). Moreover, relatedness asymmetries that arise as a result of demography are predicted to give younger females a decisive advantage in reproductive conflict with older females (Cant & Johnstone 2008). Analyses that incorporate demography can therefore help to explain both the unusual taxonomic distribution of menopause and the timing of reproductive cessation in humans, as we describe below.
Demography and kin selection across the lifespan Demography is important to consider in a model of menopause because the relatedness between a female and other breeders in her group can change as she ages, thereby affecting the strength of kin selection for acts such as helping or early reproductive cessation. Building on the infinite-island modeling framework (Wright 1931, Taylor 1992), Johnstone & Cant (2008) construct a general model of “kinship dynamics” to explore the impact of sex-biased dispersal and intra- versus extra-group mating on the strength of kin selection across the lifespan. They derive general formulae which track the changes in genetic similarity between males, females, and offspring in a local group (or “island”) as some individuals disperse, and others die and are replaced. These patterns of relatedness are then used to determine how the strength of kin selection across the lifespan varies for eight representative patterns of demography, which differ as to whether dispersal is male- or female-biased, and whether mating occurs locally or outside the group.
37
38
M. A. Cant et al. The results indicate that where dispersal is male-biased and mating occurs locally, the average relatedness of a female to the offspring produced in her local group decreases as she ages (Figure 2.6A). In these circumstances acts which benefit relatives, such as helping or reproductive restraint, are favored more strongly earlier rather than later in life. By contrast, female-biased dispersal and local mating leads to an increase with age in the average relatedness of a female to the offspring produced in her local group (Figure 2.6B). This effect arises under female-biased dispersal because a female’s relatedness to local male breeders, initially low, increases as her sons mature and remain in the group (relatedness among local female breeders, by contrast, starts low and remains low because daughters disperse). In these circumstances acts which benefit other group members are favored later, rather than earlier in life. Female-biased dispersal, therefore, predisposes females to early reproductive cessation and late-life helping. Most social mammals exhibit male-biased dispersal and female philopatry (Greenwood 1980, Pusey & Packer 1987, Clutton-Brock 1998, Lawson Handley & Perrin 2007). By contrast, three lines of evidence suggest that the evolutionary history of Homo has been characterized by female-biased dispersal and male philopatry. First, our closest primate relatives, chimpanzees, bonobos (Pan paniscus), and gorillas, are unusual among primates because they exhibit female-biased dispersal, and male dispersal is rare (Pusey et al. 1997, Boesch & Boesch-Ackermann 2000, Nishida et al. 2003, Stokes et al. 2003, Yamigawa & Kahekwa 2004, Eriksson et al. 2006). Second, patterns of variation in mitochondrial DNA and the Y-chromosome are consistent with greater rates of female than male dispersal (Seilstad et al. 1998, Oota et al. 2001), at least on the relevant, local scale (Wilder et al. 2004). Finally, female-biased transfer is common in modern human hunter–gathers (Ember 1978). For example, an influential analysis by Ember (1978) concluded that only 16.2% of 179 hunter– gatherer societies show a matrilocal pattern of residence. More recent analyses (Alvarez 2004, Marlowe 2004) have contested Ember’s classification, mainly because in the majority of human societies dispersal is merely biased towards, rather than restricted to, one sex or the other (and so, it is argued, should be classed as “bilocal”). Leaving this controversy over “strict” patrilocality aside, it remains the case that female-biased dispersal is considerably more common than the reverse pattern (Marlow 2004). Taken together, these three independent lines of evidence suggest that mutations affecting female reproductive lifespan are likely to have arisen in an ancestral social environment in which dispersal was female-biased. Johnstone and Cant’s (2008) model suggests, therefore, that humans were predisposed to evolve early reproductive cessation and late-life helping because the
Figure 2.6 Effect of demography on patterns of age-specific relatedness and selection for social acts across the lifespan (modified from Johnstone & Cant 2008). Graphs on the left show age-specific relatedness to a breeding female of other females (solid lines) and of males (dotted lines) in her group, as a function of her age. The dashed curves show mean relatedness to a female of other breeders, averaging across both sexes. Age is scaled relative to mean generation time. Results are plotted for three different demographic systems. (A) A high rate of male dispersal and a low rate of female dispersal, with mating occurring within the local group. (B) High female dispersal and low male dispersal, with mating occurring within the local group. (C) Low male and low female dispersal, with mating occurring outside the local group. Graphs on the right show the patterns of kin selection across the lifespan associated with these three systems. A female can perform social acts which result in an immediate gain of b offspring for other breeders in the group, at an immediate cost of c offspring to herself. The graphs plot the absolute magnitude of the c/b ratio below which a social action may be favored in females of different ages. Positive values indicate that selection will favor helping behavior (i.e. acts for which b > 0) when c/b falls above zero but below the value shown (in the lightly shaded area), while negative values indicate that selection will favor harming behavior (i.e. b < 0) when c/b falls below zero but above the value shown (in the heavily shaded area). For more details see Johnstone & Cant (2008).
40
M. A. Cant et al. relatedness of a female to the offspring produced in her local group increased as she aged. By contrast, most other social mammals exhibit male-biased dispersal, and so are predisposed to the evolution of early-life, rather than late-life, helping. The same model predicts that another unusual pattern of dispersal and mating can give rise to an age-specific increase in local relatedness: one in which both sexes are philopatric but mating occurs outside the local group (Figure 2.6C). Interestingly, this is precisely the demographic system exhibited by menopausal cetaceans. Both male and female resident killer whales are philopatric, but mate outside of the local group (Baird 2000, Whitehead & Mann 2000); it is not known whether transient killer whales, which do disperse, also exhibit menopause). Short-finned pilot whales (Globicephala macrorhynchus) are thought to exhibit a similar pattern – there is clear evidence for their sister species the long-finned pilot whale (G. melas) (Amos 1993), and the available genetic data suggest that the short-finned and long-finned species are comparable (Amos 1998). Again, the predicted increase in relatedness with age in this case is driven by relatedness through males. Consequently in these species mothers should direct their help towards sons, with the aim of improving their extra-group mating success. Observations of resident killer whales fit well with this prediction: mothers maintain closer associations with their adult sons than with their adult daughters, and may aid their son’s foraging efforts, or form effective alliance partners for them in agonistic encounters with other males (Baird 2000).
Relatedness asymmetries and conflict resolution What would be the consequences of female-biased dispersal for the resolution of reproductive conflict in ancestral hominids? This question can be explored using a simple model of the human social unit (Figure 2.7A; Cant & Johnstone 2008). We assume that males and females are socially monogamous, but allow for a proportion p of offspring to be fathered by unrelated males. For simplicity we assume in the basic model that only females disperse, but the qualitative results of our analysis hold where dispersal is merely biased toward, rather than restricted to, females. Females leave their natal groups at maturity, pair with a male of similar age, and join his natal social group. Consequently, when a young female first arrives in the group, she has no other genetic relatives present. This female can choose to breed herself and produce offspring to whom she is related by 1/2, or to refrain from breeding and assist the breeding attempts of the older female (i.e. the mother of her mate). This older female produces offspring to whom the younger female is
Reproductive conflict and the evolution of menopause
Figure 2.7 Schematic representation of relatedness asymmetries between generations under sex-biased dispersal (from Cant & Johnstone 2008). Male and female symbols represent parents. (A) Where females disperse and immigrate into a patrilocal group, a mother is related to the offspring of a daughter-in-law by (1 p)/4, where p is the probability of extra-pair paternity. The daughter-in-law, by contrast, is completely unrelated to the mother’s offspring. Thus the difference in relatedness to own versus other offspring is greater for younger than for older females. (B) Where males disperse and immigrate into matrilocal groups, by contrast, the difference in relatedness of a female to her own versus the other breeder’s offspring is greater for the mother than for the daughter.
unrelated. The difference in relatedness to offspring produced directly versus offspring produced by helping is therefore 1/2. The older female, by contrast, can choose to breed and produce offspring of relatedness 1/2, or refrain from breeding and help to rear grandoffspring, to whom she is related by (1 p)/4. The relatedness differential between breeding and helping for the older female is therefore 1/2 (1 p)/4, or (1 þ p)/4. This means that so long as there is any chance that her son fathered her putative grandchildren (i.e. p < 1), the difference in relatedness to offspring produced by breeding rather than helping is lower for the older female than for the younger female. As a result, a younger female will have an advantage in reproductive competition with older females because she is insensitive to the costs she inflicts on an older female by breeding. This contrasts with the situation where dispersal is male-biased (as in most social mammals). Here relatedness asymmetry favors older females over younger females, so older females are expected to have an advantage in reproductive conflict with the younger generation (Figure 2.7B).
41
42
M. A. Cant et al. A formal game-theoretic treatment of this model is provided by Cant and Johnstone (2008). Their analysis is based on the “tug-of-war” model of Reeve et al. (1998), which is the standard model for the analysis of reproductive conflict when both parties exhibit partial control over the outcome (Johnstone 2000, Cant & Shen 2006, Reeve & Ho¨lldobler 2007). Both females engage in competition over reproductive shares at the expense of total group resources. Increased selfish investment, therefore, results in a larger slice of a smaller reproductive “pie.” Regardless of the relative strength of the two females, the evolutionarily stable solution is for the older female to commit to zero reproduction and allow the younger female to claim all of the available reproduction. This is an example of an “endogenous” or “natural” Stackelberg solution in which both players prefer to act in sequence (rather than submit simultaneous “sealed bids”), and both agree on who should move first (Albaek 1990, van Damme & Hurkens 1999). Endogenous Stackelberg equilibria are interesting from a biological perspective because they can explain the evolution of commitment strategies which are profitable precisely because they cannot credibly be changed (Nesse 2001, Cant & Shen 2006). Thus, in the conflict between older and younger females the older female’s first move of zero investment is advantageous only if it is perceived to be irreversible by the younger female. Permanent sterility as a consequence of rapid reproductive senescence would be one highly effective way to commit credibly to a first move of zero investment in reproduction.
Conclusion To summarize the above arguments, a social environment in which dispersal is female-biased generates relatedness asymmetries between older and younger females and the offspring they produce. Conflict over reproduction in these circumstances is predicted to favor older females who commit to zero reproduction when females of a younger generation start to breed. Given a pattern of female-biased dispersal during the period of lengthening human lifespan, there would be little selection to extend or maintain female reproductive capacity beyond the age at which a woman might expect to become a paternal grandmother. This can account for the high and accelerating rate of oocyte loss in humans (Figure 2.2) leading to sterility in mid-life, and the exceptionally low reproductive overlap between generations in humans (Figure 2.5). The intensity of reproductive competition and the magnitude of the benefits that can be conferred by helping must also be important, however, because chimpanzees and bonobos exhibit strongly female-biased dispersal but are not unusual in their degree of reproductive overlap (Figure 2.5A).
Reproductive conflict and the evolution of menopause A number of studies have reported that the presence of a paternal grandmother has relatively little effect on offspring survival compared to that of a maternal grandmother (Bereczkei & Dunbar 1997, Sear et al. 2000, 2002), or even a negative effect on offspring survival (Beise & Voland 2002). For example, data from the Krummho¨rn region of Germany in the eighteenth and nineteenth centuries show that the chance of stillbirth for a daughter-in-law was increased by 35% if the paternal grandmother was present in the household, and these mortality costs are particularly high at the start of the daughter-inlaw’s marriage (Beise & Voland 2002, Voland & Beise 2005). While these data offer evidence of reproductive conflicts between generations in human families, the positive impact of maternal grandmothers seems at odds with our assumption of a female-biased dispersal system. It is important to distinguish, however, between the evolutionary origins of a lifehistory trait and the behavioral strategies that are employed once that trait has evolved. The universality of menopause in modern humans, despite vast differences in social systems and access to resources, illustrates the flexibility of behavior compared to the physiological processes underlying rapid reproductive senescence. The reproductive conflict model does not imply that older females should not help daughters if the social system subsequently changes to become less female-biased, or mothers are able to maintain kin ties with their daughters. Indeed, given a choice between helping daughters versus sons, mothers should direct their help preferentially towards daughters, since grandchildren through sons may have been fathered by extra-pair males. From a woman’s perspective, therefore, a flexible or “bilocal” system that allowed her to direct care toward daughters late in life would be preferable to strict patrilocality throughout her life. Most modern forager societies exhibit a degree of flexibility of this kind (Alvarez 2004, Marlowe 2004). The model is of course simplistic in many respects, but it remains a useful tool with which to make testable predictions. For example, the assumptions of the model could be tested by examining whether relatedness asymmetries exist within family units of natural-fertility populations using genealogical and/or genetic data. Given a larger sample size, from either current or historical populations, one should be able to detect a cost to females of breeding alongside reproductive grandmothers, similar to the demonstrated costs of co-breeding within generations in polygynous marriages. The fact that the ancestral social system may have changed in more recent times offers an opportunity to test our model. Attempts to test the predictions of the model could utilize the wide variety of patterns of dispersal and marital residence exhibited by modern humans (Ember 1978, Marlowe 2004). If a system of male-biased dispersal and matrilocality were to persist for many generations, our model predicts that
43
44
M. A. Cant et al. selection to minimize reproductive overlap would weaken. Consequently, we would predict mean age at last birth to be higher in historically matrilocal compared to patrilocal societies, and cultural restrictions on reproduction by grandmothers to be less prevalent in the former than the latter. The stability of the dispersal system over time could be inferred from genetic data, since genetic studies show that matrilocal societies show less variation in mitochondrial DNA, and greater variation in Y-chromosome DNA, than patrilocal societies (Oota et al. 2001). More speculatively, if the rate of reproductive senescence is adjusted on an individual level we might predict that women who have only daughters (or have daughters first) should cease reproduction (and perhaps undergo menopause) later than women who have only sons (or have sons first). The models described in this chapter raise their own questions about the evolutionary origins and taxonomic distribution of menopause. Why is this particular life-history pattern so unusual among mammals, and absent from other vertebrates? Male philopatry and female-biased dispersal is characteristic of other primates which do not exhibit menopause, such as the hamadryas baboon (Papio hamadryas: Pusey & Packer 1987, Hammond et al. 2006) and red colobus (Piliocolobus rufomitratus: Marsh 1979). It is also the pattern exhibited by large cooperatively breeding canids (Moehlman & Hofer 1997), and the majority of cooperatively breeding birds (Greenwood 1980, Brown 1987). Clearly, neither male-biased philopatry nor a benefit of cooperative breeding are sufficient to account for the evolution of menopause. The magnitude and nature of the potential fitness benefits that can be conferred by helping will be of key importance. Cetacean biologists have suggested that the main benefit that can be conferred by older female killer whales and pilot whales is information and experience (McAuliffe & Whitehead 2005). The potential benefits of information transfer are probably even greater in humans, and other aspects of human social biology (e.g. communal food-gathering, short inter-birth intervals, tool use) may also contribute to making helping a cost-effective strategy later in life. It is important to remember, however, that the indirect fitness benefits of helping represent just one side of the equation. The reproductive life history of highly social animals such as humans and toothed whales will also be shaped, in substantial part, by conflict over reproduction. An integrated approach which considers all the potential inclusive fitness consequences of social acts within ancestral families promises to yield a much-improved understanding of menopause. Acknowledgments We thank Bill Amos, Michael Briga, Rebecca Chicot, Tim CluttonBrock, Nick Davies, Jeremy Field, Øistein Holen, Reinmar Hager, Sarah Hrdy,
Reproductive conflict and the evolution of menopause Clara B. Jones, Rebecca Kilner, Nobu Kutsukake, Laurent Lehmann, Virpi Lummaa, Ruth Mace, Katie McAuliffe, Francis Ratnieks, and Joan Silk, and two anonymous reviewers, for discussion and helpful comments on earlier manuscripts. MAC and AFR are funded by Royal Society University Research Fellowships. References Albaek, S. (1990). Stackelberg leadership as a natural solution under cost uncertainty. Journal of Industrial Economics, 38, 335–347. Alexander, R. D. (1974). The evolution of social behaviour. Annual Review of Ecology and Systematics, 5, 325–383. Alvarez, H. P. (2004). Residence groups among hunter–gatherers: a view of the claims and evidence for patrilocal bands. In B. Chapais & C. M. Berman, eds., Kinship and Behavior in Primates. Oxford: Oxford University Press, pp. 420–442. Amey, F. K. (2002). Polygyny and child survival in sub-Saharan Africa. Social Biology, 49, 74–89. Amos, W. (1993). Use of molecular probes to analyse pilot whale pod structure: two novel analytical approaches. Symposia of the Zoological Society of London, 66, 33–48. Amos, W. (1998). Culture and genetic evolution in whales. Science, 284, 2055a. Amundsen, D. W. & Diers, C. J. (1970). Age of menopause in classical Greece and Rome. Human Biology, 42, 79–86. Baird, R. W. (2000). The killer whale: foraging specializations and group hunting. In J. Mann, R. Connor, P. L. Tyack, & H. Whitehead, eds., Cetacean Societies. Chicago, IL: University of Chicago Press, pp. 127–153. Beekman, M., Komdeur, J., & Ratnieks, F. L. W. (2003). Reproductive conflicts in animal societies: who has power? Trends in Ecology and Evolution, 18, 277–282. Beise, J. & Voland, E. (2002). A multilevel event history analysis of the effects of grandmothers on child mortality in a historical German population (Krummho¨rn, Ostfriesland, 1720–1874). Demographic Research, 7, 469–497. Bereczkei, T. & Dunbar, R. I. M. (1997). Female-biased reproductive strategies in a Hungarian Gypsy population. Proceedings of the Royal Society of London B, 264, 17–22. Blurton Jones, N. G., Hawkes, K., & O’Connell, J. F. (2002). The antiquity of postreproductive life: are there modern impacts on hunter–gatherer postreproductive lifespans? American Journal of Human Biology, 14, 184–205. Boesch, C. & Boesch-Achermann, H. (2000). The Chimpanzees of the Tai Forest. Oxford: Oxford University Press. Brown, J. L. (1987). Helping and Communal Breeding in Birds. Princeton, NJ: Princeton University Press. Cant, M. A. (1998). A model for the evolution of reproductive skew without reproductive suppression. Animal Behaviour, 55, 163–169. Cant, M. A. & Johnstone, R. A. (1999). Costly young and reproductive skew in animal societies. Behavioral Ecology, 10, 178–184.
45
46
M. A. Cant et al. Cant, M. A. & Johnstone, R. A. (2008). Reproductive conflict and the separation of reproductive generations in humans. Proceedings of the National Academy of Sciences of the USA, 105, 5332–5336. Cant, M. A. & Shen, S. -F. (2006). Endogenous timing in competitive interactions among relatives. Proceedings of the Royal Society of London B, 273, 171–178. Caspari, R. & Lee, S. H. (2004). Older age becomes common late in human evolution. Proceedings of the National Academy of Sciences of the USA, 101, 10895–10900. Cavalli-Sforza, L. L. (1983). The transition to agriculture and some of its consequences. In D. J. Ortner, ed., How Humans Adapt: a Biocultural Odyssey. Washington, DC: Smithsonian Institute Press, pp. 103–126. Clutton-Brock, T. H. (1998). Female transfer and inbreeding avoidance in social mammals. Nature, 337, 70–72. Cohen, A. A. (2004). Female post-reproductive lifespan: a general mammalian trait. Biological Reviews, 79, 733–750. Ember, C. (1978). Myths about hunter gatherers. Ethnology, 17, 439–448. Emlen, S. T. (1991). Evolution of cooperative breeding in birds and mammals. In J. R. Krebs and N. B. Davies, eds., Behavioural Ecology: an Evolutionary Approach, 3rd edn. Oxford: Blackwell, pp. 301–337. Emlen, S. T. (1995). An evolutionary theory of the family. Proceedings of the National Academy of Sciences of the USA, 92, 8092–8099. Eriksson, J., Siedel, H., Lukas, D., et al. (2006). Y-chromosome analysis confirms highly sex-biased dispersal and suggests a low male effective population size in bonobos (Pan paniscus). Molecular Ecology, 15, 939–949. Faddy, M. J. & Gosden, R. G. (1996). Ovary and ovulation: a model conforming the decline in follicle numbers to the age of menopause in women. Human Reproduction, 11, 1484–1486. Faddy, M. J., Gosden, R. G., Gougeon, A., Richardson, S. J., & Nelson, J. F. (1992). Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Human Reproduction, 7, 1342–1346. Gosden, R. G. & Telfer, E. (1987). Numbers of follicles and oocytes in mammalian ovaries and their allometric relationships. Journal of Zoology, 211, 169–175. Greenwood, P. J. (1980). Mating systems, philopatry and dispersal in birds and mammals. Animal Behaviour, 28, 1140–1162. Gurven, M. (2005). To give or not to give: the behavioural ecology of human food transfers. Behavioral and Brain Sciences, 27, 543–559. Gurven, M., Hill, K., & Jakugi, F. (2004). Why do foragers share and sharers forage? Explorations of social dimensions of foraging. Research in Economic Anthropology, 23, 19–43. Hamilton, W. D. (1966). The moulding of senescence by natural selection. Journal of Theoretical Biology, 12, 12–45. Hammond, R. L., Lawson Handley, L. J., Winney, B. J., Bruford, M. W., & Perrin, N. (2006). Genetic evidence for female-biased dispersal and gene flow in a polygynous primate. Proceedings of the Royal Society of London B, 273, 479–484.
Reproductive conflict and the evolution of menopause Hawkes, K. (2003). Grandmothers and the evolution of human longevity. American Journal of Human Biology, 15, 380–400. Hawkes, K., O’Connell, J. F., Blurton Jones, N. G., Alvarez, H., & Charnov, E. L. (1998). Grandmothering, menopause, and the evolution of human life histories. Proceedings of the National Academy of Sciences of the USA, 95, 1336–1339. Helle, S., Lummaa V., & Jokela, J. (2005). Are reproductive and somatic senescence coupled in humans? Late, but not early reproduction correlated with longevity in historical Sami women. Proceedings of the Royal Society of London B, 272, 29–37. Hill, K. & Hurtado, A. M. (1991). The evolution of reproductive senescence and menopause in human females. Human Nature, 2, 315–350. Hill, K. & Hurtado, A. M. (1996). Ache Life History: the Ecology and Demography of a Foraging People. New York, NY: De Gruyter. Isaac, B. L. & Feinberg, W. E. (1982). Marital form and infant survival among the Mende of rural upper Bambara Chiefdom, Sierra-Leone. Human Biology, 54, 627–634. Islam, M. M. & Yadava, R. C. (1997). Estimation of fecundability: levels and trends in Bangladesh. Journal of Family Welfare, 43, 13–20. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Johnstone, R. A. & Cant, M. A. (2008). Sex differences in dispersal and the evolution of helping and harming. American Naturalist, 172, 318–330. Jones, E. C. & Krohn, P. L. (1961). The effects of hypophysectomy on age changes in the ovaries of mice. Journal of Endocrinology, 21, 469–496. Kaplan, H. & Hill, K. (1985). Food sharing among Ache foragers: tests of explanatory hypotheses. Current Anthropology, 26, 223–245. Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 9, 98–102. Lack, D. (1947). The significance of clutch size. Ibis, 89, 309–352. Lahdenpera¨, M., Lummaa, V., & Russell, A. F. (2004a). Menopause: why does fertility end before life? Climacteric, 7, 327–331. Lahdenpera¨, M., Lummaa, V., Helle, S., Tremblay, M., & Russell, A. F. (2004b). Fitness benefits of prolonged post-reproductive lifespan in women. Nature, 428, 178–181. Lawson Handley, L. J. & Perrin, N. (2007). Advances in our understanding of mammalian sex-biased dispersal. Molecular Ecology, 16, 1559–1578. Loudon, I. (1993). Death in Childbirth: an International Study of Maternal Care and Maternal Mortality 1800–1950. Oxford: Oxford University Press. Mace, R. & Sear, R. (2005). Are humans cooperative breeders? In E. Voland, A. Chasiotis, & W. Shiefenhovel, eds., Grandmotherhood: the Evolutionary Significance of the Second Half of Female Life. New Brunswick, NJ: Rutgers University Press, pp. 143–159. Marlowe, F. W. (2000). The patriarch hypothesis: an alternative explanation of menopause. Human Nature, 11, 27–42. Marlowe, F. W. (2004). Marital residence among foragers. Current Anthropology, 45, 277–284.
47
48
M. A. Cant et al. Marsh, C. W. (1979). Female transference and mate choice among Tana River red colobus. Nature, 281, 568–569. Marsh, H. & Kasuya, T. (1986). Evidence for reproductive senescence in female cetaceans. Reports of the International Whaling Commission (Special Issue), 8, 57–74. McAuliffe, K. & Whitehead, H. (2005). Eusociality, menopause and information in matrilineal whales. Trends in Ecology and Evolution, 20, 650. Medawar, P. B. (1952). An Unsolved Problem in Biology. London: HK Lewis. Mizroch, S. A. (1981). Analyses of some biological parameters in the Antarctic fin whale. Reports of the International Whaling Commission, 31, 425–434. Moehlman, P. D. & Hofer, H. (1997). Cooperative breeding, reproductive suppression, and body mass in canids. In N. G. Solomon & J. A. French, eds., Cooperative Breeding in Mammals. Cambridge: Cambridge University Press, pp. 76–129. Moss, C. J. (2001). The demography of an African elephant (Loxodonta africana) population in Amboseli, Kenya. Journal of Zoology, 255, 145–156. Nesse, R. M., ed. (2001). Evolution and the Capacity for Commitment. New York, NY: Sage Foundation Press. Nichols, S. M., Bavister, B. D., Brenner, C. A., et al. (2005). Ovarian senescence in the rhesus monkey (Macaca mulatta). Human Reproduction, 2, 79–83. Nishida, T., Corp, N., Hamai, M., et al. (2003). Demography, female life history and reproductive profiles among the chimpanzees of Mahale. American Journal of Primatology, 59, 99–121. Olesiuk, P., Bigg, M. A., & Ellis, G. M. (1990). Life history and population dynamics of resident killer whales (Orcinus orca) in the coastal waters of British Columbia and Washington State. Reports of the International Whaling Commission (Special Issue), 12, 209–243. Oota, H., Settheetham-Ishida, W., Tiwawech, D., Ishida, T., & Stoneking, M. (2001). Human mtDNA and Y-chromosome variation is correlated with matrilocal versus patrilocal residence. Nature Genetics, 29, 20–21. Patel, T. (1994). Fertility Behaviour: Population and Society in a Rajasthan Village. Oxford; Delhi: Oxford University Press. Pebley, A. R. & Mbugua, W. (1989). Polygyny and fertility in sub-Saharan Africa. In R. J. Lesthaeghe, ed., Reproduction and Social Organization in Sub-Saharan Africa. Berkeley, CA: University of California Press, pp. 339–365. Peccei, J. S. (1999). First estimates of heritability in the age of menopause. Current Anthropology, 40, 553–558. Peccei, J. S. (2001a). Menopause: adaptation or epiphenomenon? Evolutionary Anthropology, 10, 43–57. Peccei, J. S. (2001b). A critique of the grandmother hypothesis: old and new. American Journal of Human Biology, 13, 434–452. Pennington, R. (2001). Hunter–gatherer demography. In C. Panter-Brick, R. H. Layton, & P. Rowley-Conwy, eds., Hunter–Gatherers: an Interdisciplinary Perspective. Cambridge: Cambridge University Press, pp. 170–204.
Reproductive conflict and the evolution of menopause Pettay, J. E., Kruuk, L. E. B., Jokela, J., & Lummaa, V. (2005). Heritability and genetic constraints of life-history trait evolution in preindustrial humans. Proceedings of National Academy of Sciences of the USA, 102, 2838–2843. Pusey, A. E. & Packer, C. (1987). Dispersal and philopatry. In B. B. Smuts, D. L. Cheney, R. M. Seyfarth, & R. W. Wrangham, eds., Primate Societies. Chicago, IL: University of Chicago Press, pp. 250–266. Pusey, A. E., Williams, J., & Goodall, J. (1997). The influence of dominance rank on the reproductive success of female chimpanzees. Science, 277, 828–831. Ratnieks, F. L. W., Foster, K. R., & Wenseleers, T. (2006). Conflict resolution in insect societies. Annual Review of Entomology, 51, 581–608. Reeve, H. K. & Ho¨lldobler, B. (2007). The emergence of a superorganism through intergroup competition. Proceedings of the National Academy of Sciences of the USA, 104, 9736–9740. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Rogers, A. R. (1993). Why menopause? Evolutionary Ecology, 7, 406–420. Rose, M. R. (1991). Evolutionary Biology of Aging. New York, NY: Oxford University Press. Roth, E. A. & Kurup, K. B. (1988). Demography and polygyny in a southern Sudanese agro-pastoralist society. Culture, 8, 67–73. Sear, R., Mace, R., & McGregor, I. A. (2000). Maternal grandmothers improve nutritional status and survival of children in rural Gambia. Proceedings of the Royal Society of London B, 267, 1641–1647. Sear, R., Steele, F., McGregor, I. A., & Mace, R. (2002). The effects of kin on child mortality in rural Gambia. Demography, 39, 43–63. Seilstad, M. T., Minch, E., & Cavalli-Sforza, L. L. (1998). Genetic evidence for a higher female migration rate in humans. Nature Genetics, 20, 278–280. Shanley, D. P. & Kirkwood, T. B. L. (2001). Evolution of the human menopause. Bioessays, 23, 282–287. Skinner, G. W. (2004). Grandparental effects on reproductive strategizing Nobi villagers in Early Modern Japan. Demographic Research, 11, 111–148. Snieder, H., MacGregor, A. J., & Spector, T. D. (1998). Genes control the cessation of a woman’s reproductive life: a twin study of hysterectomy and age at menopause. Journal of Clinical Endocrinology and Metabolism, 83, 1875–1880. Stokes, E. J., Parnell, R. J., & Olejniczak, C. (2003). Female dispersal and reproductive success in wild western lowland gorillas (Gorilla gorilla gorilla). Behavioral Ecology and Sociobiology, 54, 329–339. Strassmann, B. I. (1997). Polygyny as a risk factor for child mortality among the Dogon. Current Anthropology, 38, 688–695. Taylor, P. D. (1992). Altruism in viscous populations: an inclusive fitness approach. Evolutionary Ecology, 6, 352–356. Van Damme, E. & Hurkens, S. (1999). Endogenous Stackelberg leadership. Games and Economic Behavior, 28, 105–129.
49
50
M. A. Cant et al. Voland, E. & Beise, J. (2005). “The husband’s mother is the devil in the house”: data on the impact of the mother-in-law on stillbirth mortality in historical Krummho¨rn (18th–19th century Germany) and some thoughts on the evolution of postgenerative female life. In E. Voland, A. Chasiotis, & W. Shiefenhovel, eds., Grandmotherhood: the Evolutionary Significance of the Second Half of Female life. New Brunswick, NJ: Rutgers University Press, pp. 239–255. Voland, E., Chasiotis, A., & Shiefenhovel, W., eds. (2005). Grandmotherhood: the Evolutionary Significance of the Second Half of Female Life. New Brunswick, NJ: Rutgers University Press. Whitehead, H. & Mann, J. (2000). Female reproductive strategies of cetaceans. In J. Mann, R. Connor, P. L. Tyack, & H. Whitehead, eds., Cetacean Societies. Chicago, IL: University of Chicago Press, pp. 219–246. Wilder, J. A., Kingan, S. B., Mobasher, Z., Pilkington, M. M., & Hammer, M. F. (2004). Global patterns of human mitochondrial DNA and Y-chromosome structure are not influenced by higher migration rates of females versus males. Nature Genetics, 36, 1122–1125. Williams, G. C. (1957). Pleiotropy, natural selection, and the evolution of senescence. Evolution, 11, 398–411. Wilson, M. (1957). Rituals of Kinship among the Nyakyusa. London: Oxford University Press. Wood, J. W. (1994). Dynamics of Human Reproduction. New York, NY: De Gruyter. Wright, S. (1931). Evolution in Mendelian populations. Genetics, 16, 97–159. Yamagiwa, J. & Kahekwa, J. (2004). Dispersal patterns, group structure, and reproductive parameters of eastern lowland gorillas at Kahuzi in the absence of infanticide. In M. M. Robbins, P. Sicotte, & K. J. Stewart, eds., Mountain Gorillas: Three Decades of Research at Karisoke. Cambridge: Cambridge University Press, pp. 89–122.
II
Testing assumptions and predictions of skew models
3
Reproductive skew in femaledominated mammalian societies k a y e . ho l e k a m p a n d a n n e l . e ng h
Summary We review available data documenting reproductive skew in the small group of mammals characterized by female dominance over males, focusing mainly on lemurs and spotted hyenas (Crocuta crocuta). Although most females in all lemur species examined here appear to bear young at each opportunity, we know very little about variation in longer-term reproductive success or rates of reproduction among female lemurs. Therefore we cannot draw firm conclusions in regard to reproductive skew among female lemurs except that at present this appears to be slight. However, current data show that female lemurs typically mate with multiple males, and that a substantial fraction of litters containing multiple offspring is sired by more than one male. The extent of reproductive skew in male lemurs varies among species, but there is a slight trend, among the lemur species for which genetic data exist, for male skew to decrease as the intensity of female dominance increases. Variance in reproductive success among female spotted hyenas appears to be substantially greater than it is in male-dominated species in which plural breeding occurs. In this species, female dominance, combined with virilization of the external genitalia, may increase female control over mating to its extreme limit, such that we find very little reproductive skew among males relative to that found in other polygynous mammals. The most dominant male hyenas often achieve very little reproductive success. Overall, reproductive skew among females in female-dominated mammals appears to be the same as or slightly greater than that in male-dominated species, whereas skew among males in femaledominated species generally tends to be relatively low. Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
53
54
K. E. Holekamp, A. L. Engh Introduction Early models of reproductive skew (e.g. Vehrencamp 1983a, 1983b) assumed that the distribution of reproduction within vertebrate social groups is determined by interactions occurring within each sex. However, as the study of sexual selection revealed new mechanisms of multisexual mate choice and intrasexual competition for mates operating both before and after copulation in many species (e.g. Birkhead & Moller 1993; Eberhard 1996), it became apparent that there is in fact much potential for females to affect reproductive success in males and vice versa. Although reproductive success is a property of individuals, usually measured as the number of surviving offspring an individual produces in its lifetime, reproductive skew describes the extent to which reproductive success varies among members of a specific population of animals. Recent studies of reproductive skew have emphasized that many gregarious vertebrates occur in stable mixed-sex groups, and that under these circumstances in particular, control over reproduction exerted by members of one sex might affect reproductive skew in the other (e.g. Whittingham & Dunn 1998, Cant 2000, Van Hooff 2000). Cant and Reeve (2002) addressed the question of how female control influences the distribution of paternity among males in cooperatively breeding vertebrates. However, a related question that, to our knowledge, has not previously been addressed asks more broadly whether, and how, reproductive skew within animal societies is affected by which sex is socially dominant. In most mammals, sexual dimorphisms in body size, weaponry, and aggressive behavior occur in association with contest competition among males for access to females or for control of the resources needed by females for successful reproduction (Darwin 1871, Short & Balaban 1994). As males are usually larger, better armed, and more aggressive than females, they seldom have difficulty achieving social dominance over females, and males easily win contests for resources needed by both sexes. Examples abound among ungulates (e.g. red deer, Cervus elaphus), cercopithecine primates (e.g. baboons, Papio spp.), and both pinniped and fissiped carnivores (e.g. elephant seals, Mirounga angustirostris, and lions, Panthera leo, respectively). Indeed, males are larger than females, and socially dominant to them, in all but a handful of mammalian species. Our focus in this chapter is on the small group of “role-reversed” species in which females are socially dominant to males. We define female dominance as the ability of adult females to win against adult males in contests over resources, and to evoke submissive behavior from males in dyadic contexts. Although females in some species dominate males seasonally (e.g. reindeer, Rangifer tarandus; Espmark 1964), here we consider
Reproductive skew in female-dominated mammalian societies only species in which female dominance occurs year-round. We focus our attention on those species for which data exist documenting individual variation in reproductive success. The species that satisfy these criteria include two species of mole-rats (naked mole-rats, Heterocephalus glaber, and common molerats, Cryptomys hottentotus), spotted hyenas (Crocuta crocuta), and several species of lemurs. Although females dominate males in some populations of bonobos (Pan paniscus: Stanford 1998, Vervaecke et al. 2000; but see Paoli et al. 2006) and Garnett’s greater bushbabies (Otolemur garnettii: Hager & Welker 2001), data documenting reproductive skew are rare in bonobos (Gerloff et al. 1999) and unavailable in O. garnettii, so we discuss these species only briefly here. The species satisfying our inclusion criteria range in size from roughly 30 g (naked mole-rats) to 60 kg (spotted hyenas), and they occupy a variety of ecological niches. Some of these female-dominated mammals are seasonal breeders (common mole-rats and lemurs) whereas others breed year-round (naked mole-rats and spotted hyenas). Mean group size in these species ranges from one (e.g. gray mouse lemurs, Microcebus murinus: Eberle & Kappeler 2004b) to approximately 80 (naked mole-rats: Bennett & Faulkes 2000). Thus female dominance represents one of a very small number of traits all these species have in common. Here we will examine effects of female dominance in these species on the partitioning of reproduction within each sex (Table 3.1). Because the mole-rats are the subject of their own chapter in this volume (Chapter 13), we focus here on lemurs and spotted hyenas.
Reproductive skew among lemurs Lemurs are arboreal forest-dwelling primates that consume a wide array of plant material and insects. All but two of the roughly 50 extant lemur species are endemic to Madagascar (Richard & Dewar 1991, Goodman & Benstead 2005), where, due to long dry seasons and highly seasonal rainfall, reproduction among lemurs is also highly seasonal. Most lemurs are characterized by low basal metabolic rate, lack of sexual dimorphism in body size, roughly even adult sex ratios, relatively small group sizes compared to haplorhine primates, targeted female–female aggression, high infant mortality, and female dominance (Richard & Dewar 1991, Wright 1999). To date, no lemur species are known to be dominated by males (Wright 1999). However, the degree of female dominance among lemur species is variable. In one species, Eulemur fulvus, there is no obvious sex-related dominance structure, whereas in others females have feeding priority (Propithecus verreauxi, Hapalemur griseus, Eulemur mongoz, Phaner furcifer), show partial dominance (Eulemur coronatus, Daubentonia madagascarensis, Mirza coquereli), or are clearly dominant to
55
56
K. E. Holekamp, A. L. Engh Table 3.1 Social system and partitioning of reproduction in female-dominated mammals Degree
Degree of skew among
of female Species
dominance
Social system
Males
Females
Indri (Indri indri)
High
One-male, one-female
Unknown
Unknown
Multi-male, multi-female Moderate
Moderate
groups with offspring Ring-tailed lemur
High
(Lemur catta) Ruffed lemur (Varecia
groups High
Loose multi-male,
High
Multi-male, multi-
variegata) Milne-Edwards sifaka
Unknown
Unknown
Moderate?
Unknown
Unknown
Unknown
Unknown
Unknown
Moderate
Moderate
Moderate
Moderate
Moderate
Very high
Moderate
Very high
Multi-male, multi-female Unknown
Unknown
multi-female groups
(Propithecus diadema
female groups
edwardsi) Golden-crowned sifaka
High
(Propithecus tattersalli) Black lemur (Eulemur
female groups High
macaco) Grey mouse lemur
Multi-male, multiMulti-male, multifemale groups
High
(Microcebus murinus)
Solitary with small female sleeping groups
Spotted hyena (Crocuta
High
crocuta) Naked mole-rat
multi-female groups High
Large, multi-male,
High
Large, multi-male,
(Heterocephalus glaber) Common mole-rat
multi-female groups Moderate
coronatus) Alaotran gentle lemur
to high
multi-female groups
(Cryptomys hottentotus) Crowned lemur (Eulemur
Large, multi-male,
groups Moderate
(Hapalemur griseus
One-male, multi-female groups
Moderate to
Very low
high
alaotrensis) Aye-aye (Daubentonia
Moderate
Solitary
Moderate?
Unknown
Moderate
Small multi-male,
Moderate to
Low
madagascariensis) Verreaux’s sifaka (Propithecus verreauxi)
multi-female groups
Coquerel’s mouse lemur Moderate (Mirza coquereli) Fat-tailed dwarf lemur
Low
Low?
Moderate?
Low?
Unknown
Unknown
Unknown
Unknown
kin clusters Moderate
(Cheirogaleus medius) Garnett’s greater
Solitary in matrilineal
high
One-male, one-female groups with offspring
Moderate
Solitary in matrilineal kin clusters
bushbaby (Otolemur garnettii) Mongoose lemur (Eulemur mungoz)
Moderate
One-male, one-female groups with offspring
Reproductive skew in female-dominated mammalian societies Table 3.1 (cont.) Degree
Degree of skew among
of female Species
dominance
Social system
Males
Females
Fork-marked lemur
Moderate
One-male, one-female
Unknown
Unknown
Moderate
Multi-male, multi-
Low
Moderate?
High
Low?
High
Low?
Unknown
Unknown
(Phaner furcifer) Bonobo (Pan paniscus)
groups with offspring female groups
Red-fronted brown lemur None (Eulemur fulvus rufus) Mayotte brown lemur
Small multi-male, multi-female groups
None
(Eulemur fulvus
Small multi-male, multi-female groups
mayottensis) Pygmy mouse lemur (Microcebus berthae)
Unknown
Solitary, possibly with female kin clusters
males in all contexts (Microcebus murinus, Lemur catta, Indri indri, Varecia variegata, Eulemur macaco: reviewed in Radespiel & Zimmerman 2001). Here we will examine skew in lemur species with well-characterized mating systems, presented in broad order of increasing clarity with which these species exhibit female dominance over males. Verreaux’s sifaka (Propithecus verreauxi)
P. verreauxi (Figure 3.1) live in multi-male multi-female groups that range in size from 2 to 14 members (Lawler et al. 2003). Most groups contain one or two adult females, several adult males, and immatures (Jolly 1998). Both males and females have a clear dominance hierarchy (Brockman 1999, Kraus et al. 1999). Females are usually philopatric, but if a female matures in a group that already contains several breeding females, these adults may force her to disperse (Lawler et al. 2003). All males disperse at puberty (Richard 1992), but this appears to be voluntary. Females often attack older males and force them to leave the group, possibly as a means of avoiding mating with their fathers (Richard 1992, Lawler et al. 2003). Females give birth to only one infant each year (Richard et al. 2002), and only females care for young. Prime-aged females are more fertile and have higher infant survival than do very young and very old females (Richard et al. 2002). Heavier females are more likely to give birth than lighter females, and their young are more likely to survive to weaning (Richard et al. 2000, Lewis & Kappeler 2005). Richard et al. (2002) allege that reproductive success varies
57
58
K. E. Holekamp, A. L. Engh
Figure 3.1 Vereaux’s sifaka (Propithecus verreauxi) is a lemur species characterized by female dominance. The photo shows a young individual. Courtesy of Claudia Fichtel.
substantially among females; however, besides body weight and longevity, they do not identify characteristics associated with increased success. Given that only 48% of adult females give birth each year, infant survival is poor (52%), and female reproduction is not contingent on the previous year’s reproduction, there could potentially be moderate reproductive skew among female P. verreauxi. However, using data from 14 female P. verreauxi that were observed for at least 10 years (Richard et al. 2002), we calculated that there is little variance in female reproductive success (mean ¼ 0.309 offspring surviving to age 1 per year, variance ¼ 0.018, n ¼ 14). Although P. verreauxi are highly seasonal breeders (Lawler et al. 2003), the estrous periods of individual females are asynchronous (Brockman & Whitten 1996). Females often mate with more than one male, though never with all group males (Brockman 1999). Multiple mating may be a strategy to confuse paternity and thereby prevent infanticide, which has been observed in this species (Richard et al. 2002). During the breeding season, males often make forays into neighboring groups (Brockman 1999, Lawler et al. 2003). Females, however, seem to prefer residents (i.e. established immigrants). They mate more often with older, dominant males than with young subordinates and non-residents (Brockman et al. 1998, Brockman 1999), and resident males sire
Reproductive skew in female-dominated mammalian societies 68% of the offspring each year (Lawler et al. 2003). Dominant males try to guard receptive females and harass them if they mate with other males (Richard 1992, Lawler et al. 2003). However, females may refuse to mate with dominant males, and may mate surreptitiously at the group’s periphery or during intergroup conflicts (Brockman 1999). Males engage in numerous, drawn-out fights during the breeding season, leaving ample opportunity for females to mate with whomever they choose (Lawler et al. 2005). In groups without a stable male hierarchy, females may actually provoke fights between males, and usually mate with the winner (Richard 1992). Male reproductive success is highly skewed in this species. Of 70 males sampled in one population, 36 did not father any young, while the remaining 34 fathered 90 infants; on average, each male fathered 1.3 offspring, but the high variance (3.3) indicated strong skew (Lawler et al. 2005). Alaotran gentle lemur (Hapalemur griseus alaotrensis)
This highly endangered subspecies of the gray gentle lemur is found only in the marshes of Madagascar’s Lac Alaotra. They are small, cathemeral folivores (Mutschler 2002) that live in groups of 2–9 individuals, including one or two breeding females, their offspring, and one breeding male (Mutschler et al. 2000, Nievergelt et al. 2002). Females appear to be largely philopatric, whereas males disperse after puberty (Nievergelt et al. 2002). Because both male and female offspring may remain in the group even after they have reached adult size, and male offspring never reproduce until after dispersal, groups contain more breeding females than males despite the fact that adult sex ratios are even (Nievergelt et al. 2002). Reproductive females may mitigate competition from offspring by evicting maturing daughters (Mutschler 1999). Emigrant females then attempt to establish new groups with males, but their mortality rates are assumed to be high (Nievergelt et al. 2002). Both males and females participate in territory defense (Nievergelt et al. 1998), but only females care for young. On average, females produce 1.03 offspring each year (Nievergelt et al. 2002), with no difference in yearly reproductive rate between females in one-female groups and females in two-female groups (Nievergelt et al. 2002; but see Mutschler et al. 2000). Most females give birth over several consecutive years, suggesting that reproductive skew among females is extremely low (Nievergelt et al. 2002). Unlike many other lemur species, H. griseus have a long breeding season, lasting six months (Mutschler et al. 2000). In light of the relatively low levels of reproductive synchrony among females, it should be possible for males to monopolize within-group reproduction. Genetic data suggest that this is exactly what happens. Ninety-two percent of 59 infants were or could have been
59
60
K. E. Holekamp, A. L. Engh fathered by the group male (Nievergelt et al. 2002). A minimum of 8% of all offspring were sired by extra-group males; however, in 60% of these cases, the group male was closely related to the mother of the extra-group infants, and in the other 40%, the female mated with a solitary male after two or more years of unsuccessful reproduction with the group male (Nievergelt et al. 2002). Male reproductive success is twice as high in two-female groups as in one-female groups (Nievergelt et al. 2002), so there may be substantial reproductive skew among males in this species. However, turnover of group males is quite high, presumably because new immigrants expel established males (Nievergelt et al. 2002). If males compete more aggressively for two-female groups, the extent of skew might be tempered by shorter male tenure in these groups. Coquerel’s mouse lemur (Mirza coquereli)
M. coquereli are solitary lemurs that forage alone and rarely share sleeping nests with other adults (Kappeler 1997a). Females’ ranges overlap with those of several other females. Males’ home ranges, in contrast, do not overlap with those of other males, except during the breeding season, when their ranges quadruple in size. Although genetic analyses suggest that dispersal is facultative in both sexes, females tend to settle close to their natal range, creating clusters of matrilineal kin (Kappeler et al. 2002). Reproduction in this species is highly seasonal (Kappeler 1997a). Females give birth to 1–2 infants each year, and most appear to bear young (Stanger 1995, Kappeler et al. 2002), but the degree of reproductive skew among females is unknown. Paternity among M. coquereli males is relatively evenly divided. Because females emit estrus advertisement calls, as do female Daubentonia and Microcebus (Stanger 1995), it has been inferred that they seek to attract multiple mates. Males appear to compete for access to receptive females through scramble competition. Kappeler et al. (2002) found that half of twin litters were fathered by multiple males, and that there was no spatial clustering of infants sired by particular males. In addition, almost one-third of infants could not have been fathered by resident males, suggesting that their sires made brief visits from afar during the mating season. Thus, skew among males in this species appears to be quite low. Fat-tailed dwarf lemur (Cheirogaleus medius)
Fat-tailed dwarf lemurs enter a 6- to 8-month-long torpor during the dry season, so they must mate, wean their young, and fatten for the next dry season during a brief active period each year (Mu¨ller & Thalmann 2002). C. medius live in small family groups that usually consist of a mating pair, infants, and sometimes their young from the previous year (Fietz 1999, Mu¨ller & Thalmann 2002). This
Reproductive skew in female-dominated mammalian societies species is socially monogamous. Adult pairs stay together on the same home range for several years and cooperate in rearing their offspring (Fietz 1999; Mu¨ller & Thalmann 2002). Paternal care is obligate; females who lose their partners are unable to raise young (Fietz 1999). Both males and females disperse, though males usually do so earlier than females (Mu¨ller & Thalmann 2002). Mating is seasonal, starting soon after females emerge from torpor, but not highly synchronous (Fietz 1999). Females usually twin, but can give birth to as many as five infants in a single litter (Fietz 1999, Mu¨ller & Thalmann 2002). Although there are no published data on female skew in this species, it is likely to be low, given the necessity of parental care by males. Given their relatively small testes, obligate paternal care, and social monogamy, C. medius might be expected to be genetically monogamous and show low skew as well, particularly since paired males should be able to monitor their partners’ reproductive status better than should neighboring or floating males. Male C. medius produce sperm plugs (Fietz et al. 2000), suggesting that, if paired males tend to mate with their partners before other males do, they might also have a competitive advantage over extra-pair males. However, of 16 juveniles whose paternity could be determined by Fietz et al. (2000), only 56% were sired by their social fathers. The remaining 44% were fathered either by neighboring (13%) or unsampled (31%) males. No non-territorial male floaters fathered any young. Given these data, there is potential for moderate skew among male C. medius. Some males may father several litters, whereas others clearly do not father any young. However, the exact extent of skew among male C. medius is unknown. Grey mouse lemur (Microcebus murinus)
M. murinus forage alone at night, but they spend the day sleeping in tree holes that they often share with conspecifics (Radespiel 2000). Sleeping groups are stable associations, typically consisting of 2–4 closely related females (Radespiel 2000, Radespiel et al. 2001, Eberle & Kappeler 2003), though larger groups of up to 15 have been observed (Martin 1972). Co-sleeping females regularly groom and sometimes nurse each other’s offspring (Eberle & Kappeler 2003) and have extensive home range overlap (Radespiel et al. 2001). Males usually sleep alone (Radespiel 2000). Molecular evidence suggests that most females remain in their natal areas whereas most males disperse (Radespiel et al. 2001, Wimmer et al. 2002, Fredsted et al. 2004). As the mating season approaches, males expand their home ranges so that, on average, they overlap the ranges of 11 females (Radespiel 2000, Eberle & Kappeler 2004a). Although reproduction is highly seasonal in this species, most female M. murinus are receptive on different nights than nearby females (Eberle &
61
62
K. E. Holekamp, A. L. Engh Kappeler 2002, 2004b). Females entering estrus advertise their status by increasing their frequency of scent marking (Eberle & Kappeler 2004b) and emitting ultrasonic vocalizations (Buesching et al. 1998). Females give birth to 1–3 young, and at least 70% of females conceive litters each year (Radespiel 2000, Eberle & Kappeler 2004a). Heavier females produce heavier offspring, and heavier offspring are more likely to survive to one year (Eberle & Kappeler 2004b). It is unclear whether maternal weight is correlated with other characteristics, such as age or sleeping-group size, or whether there is much variance in female reproductive success. The overall sex ratio in M. murinus populations is male-biased (Radespiel 2000), and this pattern is even more pronounced in the operational sex ratio. Because females enter estrus slightly asynchronously, and because each female is receptive for only a few hours, up to 18 or more males might compete for access to a single receptive female each night (Eberle & Kappeler 2002, 2004b). Behavioral observations confirm that as many as six males approach a receptive female at once (Eberle & Kappeler 2004b). Males gather outside the nest holes of receptive females at dusk, waiting for them to emerge (Radespiel 2000, Eberle & Kappeler 2004a). Once females emerge, males attempt to mate and to guard females from the advances of other males. Heavier males are most successful at repelling competitors (Eberle & Kappeler 2004a), but even heavy males must contend with uncooperative females. Though females show little evidence of direct mate choice, they regularly counteract males’ attempts to monopolize mating, escaping from guarding males in 70% of observed matings (Eberle & Kappeler 2004b). Females are typically promiscuous, mating with as many as seven males in one evening (Eberle & Kappeler 2002, 2004b). Data from captive animals suggest that dominant males father the vast majority of all young (Andres et al. 2001; but see Radespiel et al. 2002) and inhibit testicular development and reproductive behavior in subordinate males (Perret 1992), but there is no evidence of reproductive suppression in the wild (Schmid & Kappeler 1998). Seventeen of 26 wild litters had multiple sires (Eberle & Kappeler 2004b), but reproductive success was biased towards older and heavier males (Eberle & Kappeler 2004a). The large testes of male M. murinus suggest that sperm competition may be very important in this species (Kappeler 1997b, Schmid & Kappeler 1998), and there does appear to be a reproductive advantage to males who mate with females early in their receptive periods (Eberle & Kappeler 2004a). Schmelting (2001) suggests that male reproductive success is constrained by poor male survival. He proposes that newly immigrated males maintain small territories in order to mitigate the risks of predation. In their second breeding season, when they are more familiar with their territories, they expand their ranges and thus the number of females they encounter.
Reproductive skew in female-dominated mammalian societies Accordingly, he found that 21% of recent immigrants sired offspring in their first season, whereas 44–50% of males reproduced in their second to fourth seasons. Scramble competition, sperm competition, contest competition, and female preferences for multiple partners all appear to play roles in determining how reproduction is partitioned among grey mouse lemurs. Although reproduction is skewed towards older, heavier males who hold larger territories, females limit males’ efforts to monopolize mating, so the resulting skew among males is probably moderate. Ring-tailed lemur (Lemur catta)
On average, each troop of ring-tailed lemurs (Figure 3.2) contains 12–15 individuals comprising one or more matrilines of females, their offspring, and one or more immigrant males (Sussman 1991, Gould et al. 2003). As in many Old World monkeys, closely related female L. catta spend more time together and groom each other more often than do unrelated females (Nakamichi & Koyama 1997). Females never outrank their mothers, and kin tend to have similar ranks, but female L. catta do not inherit their mothers’ ranks, as do cercopithecine monkeys and spotted hyenas (Holekamp & Smale 1993,
Figure 3.2 Female dominance in ring-tailed lemurs (Lemur catta), with the female on the left and the male on the right. Courtesy of Peter Kappeler.
63
64
K. E. Holekamp, A. L. Engh Nakamichi & Koyama 1997, Nunn & Pereira 2000). Ring-tailed lemurs also differ from monkeys and hyenas in that they rarely form third-party alliances, and their ranks are sometimes circular and may change rapidly (Nakamichi & Koyama 1997). Female L. catta cooperate to defend their core home range against incursions by other troops (Gould et al. 2003). Large troops may be more successful at defending their home ranges, but as troop size increases, annual birth rate decreases (Jolly et al. 2002, Takahata et al. 2005). Troops that contain numerous adult females often fission when some of the females force others out ( Jolly et al. 2002, Gould et al. 2003). Breeding among ring-tails is highly seasonal, but each female is usually receptive on a different day (Periera 1991, Sauther 1991). Females give birth to 1–3 young at the start of the dry season (Parga & Lessnau 2005). Seventy-five to ninety-five percent of females reproduce each year, suggesting that there is little skew in reproductive output in this species (Sussman 1991, Sauther et al. 1999, Jolly et al. 2002, Gould et al. 2003, Parga & Lessnau 2005). Infant mortality, however, is very high (52%), so there may be great variation in the number of surviving young produced by females in a troop (Sussman 1991). High-ranking females are reported to have higher reproductive success than lower-ranking females in some populations (Taylor 1986, Takahata et al. 2005). However, female hierarchies are often unstable (Takahata et al. 2005), making it difficult to draw firm conclusions about variation in lifetime reproductive success among female L. catta, and therefore also about reproductive skew. Male L. catta disperse shortly before the breeding season (Sussman 1992). Most males emigrate for the first time at around 3–4 years of age, disperse again a few years later, then disperse every three to four years thereafter (Sussman 1992). They generally attempt to join a new troop along with one or two partners, but usually only one male successfully transfers at a time (Sussman 1992). Though males rarely join females in resource defense, they frequently cooperate to keep dispersing males from joining the troop (Nakamichi & Koyama 1997). If males are able to withstand prolonged challenges from troop members, they typically enter the troop with a low rank and remain peripheral for several months (Gould 1997). Both male and female L. catta have been observed attacking infants, but infanticide appears to be rare in this species (reviewed in Sauther et al. 1999, Ichino 2005). Males do not participate in care of young. In the week before estrus, cycling females attract a great deal of male attention. Females advertise their reproductive status by displaying their swollen vulvas and scent-marking (reviewed in Jolly 1998). Males begin to follow females several days before they are receptive, and if unreceptive females do not aggressively resist amorous males, the males constantly harass
Reproductive skew in female-dominated mammalian societies them (Sauther 1991). L. catta’s large testes and sperm plugs suggest strong sperm competition in this species (Sauther et al. 1999), and evidence from captive ring-tails suggests an advantage to being the first to mate (Pereira & Weiss 1991). In the wild, males fight viciously for access to estrous females, and some data suggest high-ranking males are the first to mate with receptive females (Jolly 1966, Sauther et al. 1999, Cavigelli & Pereira 2000). After mating, males often try to guard females against mating attempts by lower-ranking males, but most females nonetheless mate with multiple males (Sauther et al. 1999). Despite constant harassment from troop males, females frequently present to and mate with low-ranking males, transferring males, and males from other troops (Sauther 1991, Sauther et al. 1999, Parga 2002). The only males with which females consistently refuse to mate are natal males (Sauther 1991). Though they are promiscuous, the order in which females accept males suggests that they do have preferences (Sauther 1991). Genetic analysis of paternity among semi-free-ranging L. catta indicates that reproductive skew among males may be considerable (Pereira & Weiss 1991). Surprisingly, however, there is not yet a molecular analysis of paternity in wild L. catta. Behavioral data from the wild suggest that males engage largely in contest and sperm competition, but that female choice may curtail male–male competition. If females do in fact tend to mate with the highest-ranking males first, then there may be strong skew among males. In contrast, if female preferences are less predictable, or if females prefer to mate with different males in consecutive estrous periods, there may be very little male skew.
Reproductive skew among bonobos (Pan paniscus) Bonobos inhabit the forests of central Africa, living in mixed-sex groups that contain 7–8 males and 8–15 females (Kano 1996, Gerloff et al. 1999). Intensive aggression is relatively rare in this species, and social rank plays a less important role in determining priority of resource access in bonobo society than it does in societies of most other gregarious primates (de Waal 1995). Controversy exists in the literature regarding the extent of female dominance in this species, but it seems to be generally agreed that female bonobos at least have feeding priority over males (Stanford 1998, Vervaecke et al. 2000, Paoli et al. 2006). In a recent review, Kutsukake & Nunn (2006) used two different indices to estimate reproductive skew among male primates, the binomial skew index of Nonacs (2000, 2003) and the lambda index of Kokko and Lindstro¨m (1997). Although the binomial index suggested that the distribution of copulations among male bonobos differs significantly from expectations based on random
65
66
K. E. Holekamp, A. L. Engh mating, the lambda index was far lower than that found in most populations of chimpanzees (Pan troglodytes), a species in which males inevitably dominate females (Kutsukake & Nunn 2006). However, the lambda values calculated for male bonobos did not differ from those calculated for many other primate species in which males dominate females (Kutsukake & Nunn 2006). It is important to note that the analysis of Kutsukake & Nunn (2006) was based on mating frequency rather than on molecular analyses of paternity in each species. Similarly, although observations of captive bonobos suggest that highranking females may interfere with other females’ copulations (Vervaecke & van Elsacker 2000), the lack of molecular paternity analyses leaves it unclear whether this behavior results in any measurable skew among females in wild populations.
Reproductive skew among spotted hyenas (Crocuta crocuta) Spotted hyenas are large terrestrial predators occurring throughout sub-Saharan Africa. They live in complex multi-generational social groups, called clans, ranging in size from 6 to 90 individuals. Each Crocuta clan contains one to several matrilines of natal females and their offspring, as well as one to several adult immigrant males. Most habitats in which spotted hyenas occur appear to be saturated such that clan territories form a mosaic covering the entire landscape (Kruuk 1972, Boydston et al. 2001). Relatedness is high within matrilines but, on average, clan members are only very distantly related due to high levels of male-mediated gene flow among clans, and mean relatedness declines only slightly across clan borders (Van Horn et al. 2004). Clans are fission–fusion societies in which all members recognize each other, defend a common territory against neighboring clans, and rear their cubs at a single communal den (Kruuk 1972, Henschel & Skinner 1991). Individual hyenas spend much of their time alone or in small groups, particularly when foraging, but they also join together during territorial defense, interactions with competitors, and at ungulate kills. Members of multiple hyena matrilines frequently join forces to defend their kills against lions or hyenas from other clans, and by doing so risk serious injury or death (Kruuk 1972, Mills 1990, Henschel & Skinner 1991, Hofer & East 1993, Boydston et al. 2001). Thus unrelated clan-mates serve as essential allies in competition for resources. The carcasses of large ungulates represent extremely rich, rare, and ephemeral food patches that occur unpredictably in space and time, so clan members compete intensively over them. Every Crocuta clan is structured by a rigid linear dominance hierarchy, and an individual’s position in this hierarchy determines its priority of access to food (Kruuk 1972, Tilson & Hamilton 1984, Frank 1986,
Reproductive skew in female-dominated mammalian societies Mills 1990). Within a clan, all adult females are socially dominant to adult males not born in the clan (Smale et al. 1993, 1997). Before cubs reach puberty, they attain ranks in the clan’s dominance hierarchy immediately below those of their mothers (Holekamp & Smale 1993, Smale et al. 1993). Virtually all males disperse between 24 and 62 months of age, and they do this voluntarily, but females typically spend their entire lives in the natal clan (Henschel & Skinner 1987, Smale et al. 1997, Boydston et al. 2005). Excluding the occasional adult male who never disperses, adult natal males always dominate adult females ranked lower than their own mothers in the clan’s hierarchy for as long as they remain in the natal clan. However, when males disperse, they behave submissively to all new hyenas encountered outside the natal area, and this is the point at which females come to dominate males (Smale et al. 1993, 1997). By joining a new clan, each immigrant assumes the lowest rank in that clan’s dominance hierarchy (Smale et al. 1997). Males disperse alone, and relatedness among immigrant males is extremely low (Van Horn et al. 2004). Dispersing females never join existing clans elsewhere, but instead leave the natal clan alone or with some clan-mates to form an entirely new clan if an opening occurs in the mosaic of neighboring territories (Mills 1990, Holekamp et al. 1993). Because females who disperse alone appear to experience very low reproductive success, female dispersal effectively occurs only by clan fission. Although subordinate females monitor reproductive opportunities outside their natal territory (Holekamp et al. 1993), vacancies rarely occur in the local mosaic of territories, so clan fission events are similarly rare. Overall it appears that opportunities for male dispersal arise frequently, but opportunities for female dispersal seldom occur. Whereas male spotted hyenas are quite mobile, their immigration to new clans appears constrained by the severe aggression directed at dispersers by resident immigrant males (Boydston et al. 2001, Szykman et al. 2003). Males appear to compete intensively for membership in new clans, so reproductive skew among male hyenas might be expected to vary with the intensity of this competition (Reeve & Emlen 2000). Intrasexual social rank among immigrant males is highly correlated with immigrants’ tenure in the new clan, such that those arriving first dominate those arriving later (Smale et al. 1997). Social status among males is not linked with variation in body size or weaponry. Low-ranking males who have recently joined a clan rise in rank as high-ranking, longer-tenured males die or leave the clan (East & Hofer 2002, Engh et al. 2002). Thus, rather than fighting with other males for social status, males acquire status by queuing (Smale et al. 1997, East & Hofer 2001, Alberts et al. 2003). The queuing convention observed among immigrant males is relatively strict (Kokko & Johnstone 1999); males who
67
68
K. E. Holekamp, A. L. Engh manage to join a new clan rarely use aggression to enhance their status, but instead simply wait their turn to rise in rank (East & Hofer 2001). Female Crocuta are slightly larger than males, and they are heavily masculinized with respect to many aspects of their morphology and behavior (Lindeque & Skinner 1982, Glickman et al. 1992). In addition to being socially dominant to males, females are also more aggressive (Smale et al. 1993, Szykman et al. 2003), and their external genitalia are heavily virilized, such that the female urinates, copulates, and gives birth through a fully erectile pseudopenis (Matthews 1939). The female hyena’s masculinized genitalia, combined with her social dominance over breeding males, allow her an extraordinarily high degree of control over mating relative to that experienced by female mammals living in male-dominated societies (Engh et al. 2002, East et al. 2003). In most Crocuta populations studied to date, births occur throughout the year, so reproductive synchrony among females is generally very low (Kruuk 1972, Lindeque & Skinner 1982, Mills 1990, Holekamp et al. 1999). Litters born to wild hyenas usually contain one or two cubs, although triplets occur occasionally. Males do not participate in care of young, nor do females participate in the care of offspring other than their own. Approximately 8% of all hyena deaths with known causes in East African Crocuta clans are due to infanticide (Hofer & East 1995). Most successful attempts at infant-killing are perpetrated by adult females (Frank 1996), although adult males have also been observed to make failed attempts (Kruuk 1972, East & Hofer 1995). Dominant females sometimes kill cubs of subordinates, but subordinate females also kill dominant cubs (Frank 1996, East et al. 2003). Infanticide by subordinates can potentially decrease reproductive skew (Young & CluttonBrock 2006). Interestingly, reproductive skew among female spotted hyenas appears to be greater than that documented among females of male-dominated species characterized by plural breeding (Holekamp & Smale 2000). Although all female Crocuta produce young, they do so at rates that vary strongly with social rank. For example, the highest-ranking female hyenas produce an average of approximately 2.5 offspring per year, while the lowest-ranking females produce only 0.5 offspring per year, a five-fold difference (Holekamp et al. 1996). This stands in marked contrast to the far more egalitarian distribution of reproduction among females in other plural breeders. For example, in female lions the most extreme difference in annual cub production between any two pride-mates is only 0.35 cubs (Packer et al. 2001). Clans of spotted hyenas are similar in size, structure, and composition to the societies of many cercopithecine primates. Like spotted hyenas, these primates are plural breeders. Although high-ranking female primates garner
Reproductive skew in female-dominated mammalian societies greater reproduction than do subordinates, the rank-related variation in female reproductive success in all these primate species is small compared to that in spotted hyenas (Holekamp & Smale 2000). In addition to producing offspring at higher rates, high-ranking female hyenas bear daughters that generally grow faster, are more likely to survive to adulthood, start breeding earlier, and enjoy longer reproductive lifespans than do daughters of mid- or low-ranking females (Holekamp et al. 1996, Hofer & East 1996). Direct aggressive harassment does not appear to function importantly in the reproductive suppression of subordinate females in this species (Holekamp et al. 1996). Furthermore, the occurrence of infanticide is also often inconsistent with suppression of reproduction in subordinate hyenas. Instead, the key determinant of reproductive success among female Crocuta is rank-related priority of access to food resources. Those female hyenas enjoying the greatest reproductive success are those most effectively able to use aggression to displace conspecifics from carcasses during competitive feeding. Reproductive skew among male Crocuta
Crocuta’s mating system is polygynous, but matings are not monopolized by high-ranking males (Engh et al. 2002, East et al. 2003). Dispersal status and length of residence as immigrants in new clans after dispersal are the strongest determinants of reproductive success among male hyenas. Adult natal male hyenas are socially dominant to immigrant males, and often show strong sexual interest in clan females (Holekamp & Smale 1998), yet they sire only 3% of cubs in their natal clans, whereas immigrants sire 97% (Engh et al. 2002), indicating that females prefer to mate with immigrants. Among resident immigrant males, social rank is correlated with male reproductive success, but tenure predicts this far better than does male rank. Immigrants do not typically begin to sire offspring until they have resided in their new clan for one or two years, during which time they occupy low rank positions in the male queue (Engh et al. 2002, East et al. 2003). At least 40% of female Crocuta mate with multiple males during any given estrous period, and 25–40% of twin litters are multiply sired (Engh et al. 2002, East et al. 2003). Work by Cant & Reeve (2002) underscores the importance of evaluating whether males or females control paternity in multi-male vertebrate societies before attempting to examine predictions of skew models. The authors suggest experimental manipulations in which alpha males are removed briefly when females are fertile to assess the extent to which females control the distribution of paternity. If females control paternity, removal of alpha males should have little consistent effects on their share of paternity. By contrast, if alpha males normally control paternity, their removal should lead to a reduction in
69
70
K. E. Holekamp, A. L. Engh their share of paternity. Crocuta offer a natural experiment of this sort. Patterns of paternity in spotted hyenas do not conform, as they do in many other gregarious mammals, to a dominance-based “priority of access” model (Altmann et al. 1996) predicting that the number of offspring sired by males should directly reflect both male social rank and the number of females simultaneously in estrus. Hyena immigrants in the top half of the male hierarchy father more cubs than those in the bottom half, but males of all ranks sire offspring, and the alpha male generally sires fewer cubs than do males in lower rank positions (Engh et al. 2002); again, this suggests an important role for female choice. Overall reproductive skew among male Crocuta is considerably less than that documented in many other group-living mammals in which males dominate females (ungulates: Clutton-Brock et al. 1988, Pemberton et al. 1992, Hirotani, 1994; rodents: Sherman et al. 1991; primates: Smith 1993, de Ruiter et al. 1994, Altmann et al. 1996, Bercovitch & Nurnberg 1997, Setchell et al. 2005; carnivores: Keane et al. 1994, Creel et al. 1997, Girman et al. 1997). In contrast to what occurs in most polygynous mammals, male–male contest competition appears to have little influence over male reproductive success in the spotted hyena. Instead, female choice of mates appears to be the key determinant of patterns of paternity. Females clearly prefer immigrants over adult natal males, and they frequently choose immigrant males other than the alpha male as mates (Engh et al. 2002). High-ranking male hyenas cannot monopolize reproduction if females prefer not to mate with them. Female control over mating has thus reduced selection for male fighting ability, and has led to low levels of male–male combat and the evolution of a male social queue (East et al. 2003).
How well do female-dominated mammals satisfy assumptions of skew models? In accordance with transactional models, dominant lemurs benefit from the presence of subordinates in the group; subordinate lemurs are often needed by dominants for cooperative defense of territories. Ecological constraints clearly limit dispersal in lemurs, group stability appears to be important to all group members, and subordinates in some lemur species are constrained by the threat of expulsion from the group by dominants. Similarly, mole-rats (see Chapter 13) also appear to satisfy the assumptions of transactional skew models (e.g. Vehrencamp 1983a, 1983b, Reeve & Emlen 2000) quite well. Spotted hyenas also appear to satisfy some of the assumptions underlying transactional models. Specifically, dominant spotted hyenas need subordinate allies, without which high-ranking animals risk loss of individual ungulate
Reproductive skew in female-dominated mammalian societies carcasses and the group territory. Furthermore, group stability is important to both high- and low-ranking hyenas; rates of female–female aggression and wounding are greatly elevated during periods of social upheaval (Van Meter & Holekamp, unpublished data). Finally, ecological constraints limit dispersal in hyenas, particularly among females. However, lemurs and spotted hyenas fail to satisfy various other assumptions underlying transactional models. Most importantly, transactional models assume that dominant individuals have complete control over allocation of reproduction within the group, a condition that is clearly not met by either lemurs or hyenas. Furthermore, transactional models assume that dominants can evict subordinates when this is in the best interests of the dominants, but this is not true in hyenas. The ability of dominant hyenas to control subordinate reproduction is severely constrained by the fission–fusion sociality characteristic of this species, and eviction potential is low because subordinates are very well armed and often larger than dominants. In accordance with compromise skew models (e.g. Clutton-Brock 1998, Reeve et al. 1998), both dominants and subordinates reproduce in the societies of lemurs and hyenas because neither has complete control over reproduction. Moreover, individuals clearly struggle over the distribution of reproduction rather than conceding each other a share of it to their mutual benefit. However, contrary to assumptions underlying compromise models, ecological constraints are extremely important in limiting dispersal in both lemurs and hyenas, and group stability does enhance reproductive success of both dominant and subordinate group-mates. Furthermore, in violation of another assumption of compromise models, dominant hyenas may be constrained by the threat of subordinate departure, as clans regularly fission whenever vacant habitat becomes available (e.g. Holekamp et al. 1993). Overall it appears that the female-dominated mammals considered in this chapter satisfy the assumptions of transactional skew models slightly better than those of compromise models. That is, ecological constraints and benefits of group membership appear to be important determinants of skew in all these species, as they are in many male-dominated mammals. However, lemurs and hyenas fail to meet some of the critical assumptions of both transactional and compromise models, so we would expect predictions derived from those models to be of similarly limited value in regard to these species. We suggest that, although skew models are useful in predicting partitioning of reproduction in mole-rats (see Chapter 13), perhaps current models invoke too many simplistic assumptions to account for patterns of reproduction and aggression in societies like those of lemurs and spotted hyenas. For example, most skew models ignore individual variation within dominant or subordinate
71
72
K. E. Holekamp, A. L. Engh classes, and incorrectly assume that all members of each class are identical (Smith et al. 2007). One possible exception to this generalization is the recruiter–joiner model, in which Hamilton (2000) uses optimal skew theory to predict group size and how resources will be divided within groups of social foragers. Assuming that reproductive output by female spotted hyenas is directly related to their food intake, an assumption strongly supported by empirical field data (Holekamp et al. 1996, Hofer & East 2003), the “limited control” transactional model presented by Hamilton (2000) appears to predict partitioning of reproduction among female spotted hyenas better than do other skew models.
Conclusion Although reproductive skew in both sexes has been well documented in mole-rats and spotted hyenas, the paucity of data from other species with female dominance (e.g. lemurs and bonobos) does not presently allow us to arrive at any firm general conclusions about skew in female-dominated mammals. Variation in reproductive skew among these species is enormous: it is extremely high among female mole-rats, moderate among female spotted hyenas, and apparently quite low in female lemurs. Thus, overall variation in skew among these species appears to be no different than that documented in male-dominated mammals. Dominant female mole-rats can control reproduction in subordinates (Faulkes et al. 1991, Lacey & Sherman 1997) far more effectively than can dominant lemurs or hyenas, suggesting that female molerats satisfy the assumptions of original transactional skew models (e.g. Vehrencamp 1983a, 1983b) much better than do these other animals. With the exception of naked mole-rats, reproductive skew among males in female-dominated mammals is greater than that among females, even in socially monogamous species, and is similar to that found in male-dominated species. Females in female-dominated societies must be somewhat freer from constraints upon reproduction than their male-dominated counterparts, because they have higher priority of access to food and other resources critical for reproduction than males. Therefore, in female-dominated species, we might expect the average productivity of females to be higher than in male-dominated species. However, since reproductive skew largely reflects intrasexual competition, there is no reason to expect a priori that female skew should necessarily differ between female-dominated and male-dominated societies. Indeed, in some cases the question of which sex is socially dominant appears to be moot in regard to skew. For example, female skew among male-dominated Damaraland molerats (Cryptomys damarensis) does not differ from that in female-dominated naked
Reproductive skew in female-dominated mammalian societies and common mole-rats (Bennett & Faulkes 2000). Similarly, in the one lemur species (Eulemur fulvus) that is not characterized by female dominance, older, heavier females have higher reproductive success than younger, lighter females (Overdorff et al. 1999), a pattern of skew virtually identical to that in Propithecus verreauxi, in which females dominate males. Although this implies that skew is independent of female dominance patterns, many gaps remain in our knowledge about the species discussed in this review. Furthermore, the analysis presented here may be confounded by our attempts to compare reproductive skew among species with different types of social organization. For example, we have attempted here to evaluate patterns of reproductive skew in gregarious, monogamous, and solitary lemurs. However, although we see no reason why skew cannot be calculated within a population of solitary or monogamous animals as well as within a particular social group of gregarious animals, “reproductive skew” traditionally refers only to the latter. In any case, until more studies examine variance in lifetime reproductive success, it will be difficult to come to any firm conclusions about the degree or determinants of skew among female lemurs in particular. More generally, conclusions about the extent of reproductive skew among female mammals, regardless of which sex is socially dominant, should ideally be based on data documenting offspring survivorship rather than simply the number of females breeding within a particular social group, because early mortality among offspring can profoundly affect long-term reproductive success. There is a suggestion in the work reviewed in this chapter that skew among females in some female-dominated mammals may be slightly greater than that in male-dominated species with comparable group compositions. The degree of skew among females is likely to be affected by the extent to which priority of access to resources influences reproductive output. In hyenas and mole-rats this appears to be considerably greater than in lemurs, and the reproductive skew among female lemurs appears to be correspondingly low. If in fact reproductive skew among females tends to be greater in female-dominated than in male-dominated species, then males in these species should be selective with respect to their own choice of mates. Indeed, we already know that male spotted hyenas strongly prefer to mate with high-ranking females (Szykman et al. 2001). Female dominance, and the associated high level of female control over reproduction, tends to decrease skew among males. Although few studies have used molecular tools to assess partitioning of reproduction among male lemurs, virtually all female lemurs mate with multiple males at each estrus, and they often choose to mate with low-ranking resident males or males from other groups. Most male lemurs invest little in reproduction and therefore
73
74
K. E. Holekamp, A. L. Engh might be expected to exhibit a higher degree of skew than females. However, compared to male-dominated species, reproductive success among male lemurs is probably influenced relatively weakly by male–male competition and relatively strongly by female preferences. As a result, we might expect that skew among male lemurs, on average, should be less than that documented in male-dominated mammals living in groups of comparable size. On the other hand, if females prefer males who excel in intraspecific competition, this prediction would be nullified. We find that skew among male lemurs spans a wide spectrum. Male P. verreauxi show strong skew, whereas skew among male M. coquereli is essentially absent. Female dominance is expressed at a low level in P. verreauxi, but is much more strongly pronounced in M. coquereli. In several lemur species, females counteract males’ attempts to monopolize reproduction, so it is likely, on the whole, that there is less skew among male lemurs than among maledominated mammals. However, the significance of this difference is unclear, since even in male-dominated species, females often successfully resist males’ attempts to control mating (e.g. Eberhard 1996). Among spotted hyenas, the female’s masculinized genitalia may increase female control over mating to its extreme limit, and in this species we see a much lower degree of reproductive skew among males relative to that found in other polygynous mammals (e.g., compare Engh et al. 2002 with Altmann et al. 1996). The most dominant males in this species often achieve very little reproductive success. In neither lemurs nor hyenas do dominant males determine whether females are permitted some promiscuity, as females in both groups clearly make their own decisions regarding whether or not to mate with more than one male. Competition among male mammals for access to females is often intense, and can result in high variance in reproductive success among males, especially in polygynous species. In most mammals, sexual selection favors traits in males that are attractive to females and that enhance male competitive ability. The range of reproductive skew among males in female-dominated species is generally less than it is in male-dominated mammals. Among female dominated species, we seldom find cases in which one male monopolizes all reproduction. Even in naked mole-rats, multiple males usually get to mate, and females bear litters sired by multiple males (Bennett & Faulkes 2000). Given the apparent strength of the effects of female choice in female-dominated mammalian species, it appears that males have been obliged to develop strategies to maximize their reproductive success that supplement or replace male–male combat. We see much more reliance in female-dominated species than in the majority of mammals on alternative modes of sexually selected interactions including endurance rivalry (e.g. queuing), sperm competition, use of copulatory plugs,
Reproductive skew in female-dominated mammalian societies and mate-guarding. Coercive sex is not possible in female-dominated species, and female dominance makes it difficult or impossible for males to monopolize females. Each female determines whether or not she will be monopolized by a single male and, if so, which male this will be. Dominant females can tolerate or refuse male mating attempts according to their own reproductive interests, and this unusual degree of female control appears to reduce the strength of the relationship between social status and reproductive success in males. Acknowledgments This chapter was produced with partial support of funds from the US National Science Foundation in grants IBN0343381 and IOB0618022. References Alberts, S. C., Watts, H. E., & Altmann, J. (2003). Queuing and queue-jumping: longterm patterns of reproductive skew among male savannah baboons, Papio cynocephalus. Animal Behaviour, 65, 821–840. Altmann, J., Alberts, S. C., Haines, S. A., et al. (1996). Behavior predicts genetic structure in a wild primate group. Proceedings of the National Academy of Sciences of the USA, 93, 5797–5801. Andres, M., Gachot-Neveu, H., & Perret, M. (2001). Genetic determination of paternity in captive grey mouse lemurs: pre-copulatory sexual competition rather than sperm competition in a nocturnal prosimian? Behaviour, 138, 1047–1063. Bennett, N. C. & Faulkes, C. G. (2000). African Mole-Rats: Ecology and Eusociality. Cambridge: Cambridge University Press. Bercovitch, F. B. & Nurnberg, P. (1997). Genetic determination of paternity and variation in male reproductive success in two populations of rhesus macaques. Electrophoresis, 18, 1701–1705. Birkhead, T. & Moller, A. (1993). Female control of paternity. Trends in Ecology and Evolution, 8, 100–104. Boydston, E. E., Morelli, T. L., & Holekamp, K. E. (2001). Sex differences in territorial behavior exhibited by the spotted hyena (Crocuta crocuta). Ethology, 107, 369–385. Boydston, E. E., Kapheim, K. M., Van Horn, R. C., Smale, L., & Holekamp, K. E. (2005). Sexually dimorphic patterns of space use throughout ontogeny in the spotted hyaena (Crocuta crocuta). Journal of Zoology, London, 267, 271–281. Brockman, D. K. (1999). Reproductive behavior of female Propithecus verreauxi at Beza Mahafaly, Madagascar. International Journal of Primatology, 20, 375–398. Brockman, D. K. & Whitten, P. L. (1996). Reproduction in free-ranging Propithecus verreauxi: estrus and the relationship between multiple partner matings and fertilization. American Journal of Physical Anthropology, 100, 57–69.
75
76
K. E. Holekamp, A. L. Engh Brockman, D. K., Whitten, P. L., Richard, A. F., & Schneider, A. (1998). Reproduction in free-ranging male Propithecus verreauxi: the hormonal correlates of mating and aggression. American Journal of Physical Anthropology, 105, 137–151. Buesching, C. D., Heistermann, M., Hodges, J. K., & Zimmermann, E. (1998). Multimodal oestrus advertisement in a small nocturnal prosimian, Microcebus murinus. Folia Primatologica, 69, 295–308. Cant, M. A. (2000). Social control of reproduction in banded mongooses. Animal Behaviour, 59, 147–158. Cant, M. A. & Reeve, H. K. (2002). Female control of the distribution of paternity in cooperative breeders. American Naturalist, 160, 602–611. Cavigelli, S. A. & Pereira, M. E. (2000). Mating season aggression and fecal testosterone levels in male ring-tailed lemurs (Lemur catta). Hormones and Behavior, 37, 246–255. Clutton-Brock, T. H. (1998). Reproductive skew, concessions and limited control. Trends in Ecology and Evolution, 13, 288–292. Clutton-Brock, T. H., Albon, S. D. & Guinness, F. E. (1988). Reproductive success in male and female red deer. In T. H. Clutton-Brock, ed., Reproductive Success. Chicago, IL: University of Chicago Press, pp. 325–343. Creel, S. (1997). Rank and reproduction in cooperatively breeding African wild dogs: behavioral and endocrine correlates. Behavioral Ecology, 8, 298–306. Darwin, C. (1871). The Descent of Man and Selection in Relation to Sex. London: John Murray. de Ruiter, J. R., van Hooff, J. A. R. A. M., & Scheffrahn, W. (1994). Social and genetic aspects of paternity in wild long-tailed macaques (Macaca fascicularis). Behaviour, 129, 203–224. de Waal, F. B. M. (1995). Bonobo sex and society. Scientific American, 272, 82–88. East, M. L. & Hofer, H. (2001). Male spotted hyenas (Crocuta crocuta) queue for status in social groups dominated by females. Behavioral Ecology, 12, 558–568. East, M. L. & Hofer, H. (2002). Conflict and cooperation in a female-dominated society: a reassessment of the “hyperaggressive” image of spotted hyenas. Advances in the Study of Behavior, 31, 1–30. East, M. L., Burke, T., Wilhelm, K., Greig, C., & Hofer, H. (2003). Sexual conflicts in spotted hyenas: male and female mating tactics and their reproductive outcome with respect to age, social status and tenure. Proceedings of the Royal Society of London B, 270, 1247–1254. Eberhard, W. G. (1996). Female Control: Sexual Selection by Cryptic Female Choice. Princeton, NJ: Princeton University Press. Eberle, M. & Kappeler, P. M. (2002). Mouse lemurs in space and time: a test of the sociobiological model. Behavioral Ecology and Sociobiology, 51, 131–139. Eberle, M. & Kappeler, P. M. (2003). Cooperative breeding in grey mouse lemurs (Microcebus murinus). Folia Primatologica, 74, 367–368. Eberle, M. & Kappeler, P. M. (2004a). Sex in the dark: determinants and consequences of mixed male mating tactics in Microcebus murinus, a small solitary nocturnal primate. Behavioral Ecology and Sociobiology, 57, 77–90.
Reproductive skew in female-dominated mammalian societies Eberle, M. & Kappeler, P. M. (2004b). Selected polyandry: female choice and intersexual conflict in a small nocturnal solitary primate (Microcebus murinus). Behavioral Ecology and Sociobiology, 57, 91–100. Engh, A. L., Funk, S. M., Van Horn, R. C., et al. (2002). Reproductive skew among males in a female-dominated society. Behavioral Ecology, 13, 193–200. Espmark, Y. (1964). Studies in dominance–subordination relationship in a group of semi-domestic reindeer (Rangifer tarandus). Animal Behaviour, 12, 420–426. Faulkes, C. G., Abbott, D. H., Liddell, C. E., George, L. M., & Jarvis, J. U. M. (1991). Hormonal and behavioral aspects of reproductive suppression in female naked mole rats. In P. W. Sherman, J. U. M. Jarvis, & R. D. Alexander, eds., The Biology of the Naked Mole Rat. Princeton, NJ: Princeton University Press, pp. 426–445. Fietz, J. (1999). Monogamy as a rule rather than exception in nocturnal lemurs: the case of the fat-tailed dwarf lemur, Cheirogaleus medius. Ethology, 105, 259–272. Fietz, J., Zischler, H., Schwiegk, C., et al. (2000). High rates of extra-pair young in the pair-living fat-tailed dwarf lemur, Cheirogaleus medius. Behavioral Ecology and Sociobiology, 49, 8–17. Frank, L. G. (1986). Social organization of the spotted hyaena (Crocuta crocuta). II. Dominance and reproduction. Animal Behaviour, 35, 1510–1527. Frank, L. G. (1996). Female masculinization in the spotted hyena: endocrinology, behavioral ecology and evolution. In J. L. Gittleman, ed., Carnivore Behavior, Ecology, and Evolution. Ithaca, NY: Cornell University Press, vol. 2, pp. 78–131. Fredsted, T., Pertoldi, C., Olesen, J. M., Eberle, M. & Kappeler, P. M. (2004). Microgeographic heterogeneity in spatial distribution and mtDNA variability of gray mouse lemurs (Microcebus murinus, Primates: Cheirogaleidae). Behavioral Ecology and Sociobiology, 56, 393–403. Gerloff, U., Hartung, B., Fruth, B., Hohmann, G., & Tautz, D. (1999). Intracommunity relationships, dispersal patterns and paternity success in a wild-living community of bonobos (Pan paniscus) determined from DNA analysis of faecal samples. Proceedings of the Royal Society of London B, 266, 1189–1195. Girman, D. J., Mills, M. G. L., Geffen, E., & Wayne, R. K. (1997). A molecular genetic analysis of social structure, dispersal, and interpack relationships of the African wild dog (Lycaon pictus). Behavioral Ecology and Sociobiology, 40, 187–198. Glickman, S. E., Frank, L. G., Pavgi, S., & Licht, P. (1992). Hormonal correlates of “masculinization” in female spotted hyenas (Crocuta crocuta). I. Infancy to sexual maturity. Journal of Reproduction and Fertility, 95, 451–462. Goodman, S. M. & Benstead, J. P. (2005). Updated estimates of biotic diversity and endemism for Madagascar. Oryx, 39, 73–77. Gould, L. (1997). Intermale affiliative behavior in ringtailed lemurs (Lemur catta) at the Beza-Mahafaly Reserve, Madagascar. Primates, 38, 15–30. Gould, L., Sussman, R. W., & Sauther, M. L. (2003). Demographic and life-history patterns in a population of ring-tailed lemurs (Lemur catta) at Beza Mahafaly Reserve, Madagascar: a 15-year perspective. American Journal of Physical Anthropology, 120, 182–194.
77
78
K. E. Holekamp, A. L. Engh Hager, R. & Welker, C. (2001). Female dominance in African lorises (Otolemur garnettii). Folia Primatologica, 72, 48–50. Hamilton, I. M. (2000). Recruiters and joiners: using optimal skew theory to predict group size and the division of resources within groups of social foragers. American Naturalist, 155, 684–695. Henschel, J. R. & Skinner, J. D. (1987). Social relationships and dispersal patterns in a clan of spotted hyaenas, Crocuta crocuta, in the Kruger National Park. South African Journal of Zoology, 22, 18–23. Henschel, J. R. & Skinner, J. D. (1991). Territorial behavior by a clan of spotted hyaenas, Crocuta crocuta. Ethology, 88, 223–235. Hirotani, A. (1994). Dominance rank, copulatory behavior and estimated reproductive success in male reindeer. Animal Behaviour, 48, 929–936. Hofer, H. & East, M. L. (1993). The commuting system of Serengeti spotted hyaenas: how a predator copes with migratory prey. II. Intrusion pressure and commuter’s space use. Animal Behaviour, 46, 559–574. Hofer, H. & East, M. L. (1995). Population dynamics, population size, and the commuting system of Serengeti spotted hyaenas. In A. R. E. Sinclair and P. Arcese, eds., Serengeti II. Chicago, IL: University of Chicago Press, pp. 332–363. Hofer, H. & East, M. L. (1996). The components of parental care and their fitness consequences: a life-history perspective. Verhandlungen der Deutschen Zoologischen Gesellschaft, 89, 149–164. Hofer, H. & East, M. L. (2003). Behavioral processes and costs of co-existence in female spotted hyenas: a life-history perspective. Evolutionary Ecology, 17, 315–331. Holekamp, K. E. & Smale, L. (1993). Ontogeny of dominance in free-living spotted hyaenas: juvenile rank relations with other immature individuals. Animal Behaviour, 46, 451–466. Holekamp, K. E. & Smale, L. (1998). Dispersal status influences hormones and behavior in the male spotted hyena. Hormones and Behavior, 33, 205–216. Holekamp, K. E. & Smale, L. (2000). Feisty females and meek males: reproductive strategies in the spotted hyena. In K. Wallen & J. Schneider, eds., Reproduction in Context. Cambridge. MA: MIT Press, pp. 257–285. Holekamp, K. E., Ogutu, J. O., Dublin, H. T., Frank, L. G., & Smale, L. (1993). Fission of a spotted hyaena clan: consequences of prolonged female absenteeism and causes of female emigration. Ethology, 93, 285–299. Holekamp, K. E., Smale, L., & Szykman, M. (1996). Rank and reproduction in female spotted hyaena. Journal of Reproduction and Fertility, 108, 229–237. Holekamp, K. E., Smale, L., & Cooper, S. (1997). Hunting rates and hunting success in the spotted hyaena. Journal of Zoology, London, 242, 1–15. Holekamp, K. E., Szykman, M., Boydston, E. E., & Smale, L. (1999). Association of seasonal reproductive patterns with changing food availability in an equatorial carnivore. Journal of Reproduction and Fertility, 116, 87–93.
Reproductive skew in female-dominated mammalian societies Ichino, S. (2005). Attacks on a wild infant ring-tailed lemur (Lemur catta) by immigrant males at Berenty, Madagascar: interpreting infanticide by males. American Journal of Primatology, 67, 267–272. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Jolly, A. (1966). Lemur Behavior. Chicago, IL: University of Chicago Press. Jolly, A. (1998). Pair-bonding, female aggression and the evolution of lemur societies. Folia Primatologica, S1, 1–13. Jolly, A., Dobson, A., Rasamimanana, H. M., et al. (2002). Demography of Lemur catta at Berenty Reserve, Madagascar: effects of troop size, habitat and rainfall. International Journal of Primatology, 23, 327–353. Kano, T. (1996) Male rank order and copulation rate in a unit-group of bonobos at Wamba, Zaire. In W. C. McGrew, L. F. Marchant, & T. Nishida, eds., Great Ape Societies. Cambridge: Cambridge University Press, pp. 135–145. Kappeler, P. M. (1997a). Intrasexual selection in Mirza coquereli: evidence for scramble competition polygyny in a solitary primate. Behavioral Ecology and Sociobiology, 45, 115–127. Kappeler, P. M. (1997b). Intrasexual selection and testis size in strepsirhine primates. Behavioral Ecology, 8, 10–19. Kappeler, P. M., Wimmer, B., Zinner, D., & Tautz, D. (2002). The hidden matrilineal structure of a solitary lemur: implications for primate social evolution. Proceedings of the Royal Society of London B, 269, 1755–1763. Keane, B., Waser, P. M., Creel S. R., et al. (1994). Subordinate reproduction in dwarf mongooses. Animal Behaviour, 47, 65–75. Kokko, H. & Johnstone, R. A. (1999). Social queuing in animal societies: a dynamic model of reproductive skew. Proceedings of the Royal Society of London B, 266, 571–578. Kokko, H. & Lindstro¨m, J. (1997). Measuring the mating skew. American Naturalist 149, 794–799. Kraus, C., Heistermann, M., & Kappeler, P. M. (1999). Physiological suppression of sexual function of subordinate males: a subtle form of intrasexual competition among male sifakas (Propithecus verreauxi)? Physiology and Behavior, 65, 855–861. Kruuk, H. (1972). The Spotted Hyena: a Study of Predation and Social Behavior. Chicago, IL: University of Chicago Press. Kutsukake, N. & Nunn, C. L. (2006). Comparative tests of reproductive skew in male primates: the role of demographic factors and incomplete control. Behavioral Ecology and Sociobiology, 60, 695–706. Lacey, E. A. & Sherman, P. W. (1997). Cooperative breeding in naked mole rats: implications for vertebrate and invertebrate sociality. In N. G. Solomon & J. A. French, eds., Cooperative Breeding in Mammals. Cambridge: Cambridge University Press, pp. 267–301. Lawler, R. R., Richard, A. F., & Riley, M. A. (2003). Genetic population structure of the white sifaka (Propithecus verreauxi verreauxi) at Beza Mahafaly Special Reserve, southwest Madagascar (1992–2001). Molecular Ecology, 12, 2307–2317.
79
80
K. E. Holekamp, A. L. Engh Lawler, R. R., Richard, A. F., & Riley, M. A. (2005). Intrasexual selection in Verreaux’s sifaka (Propithecus verreauxi verreauxi). Journal of Human Evolution, 48, 259–277. Lewis, R. J. & Kappeler, P. M. (2005). Seasonality, body condition, and timing of reproduction in Propithecus verreauxi verreauxi in the Kirindy Forest. American Journal of Primatology, 67, 347–364. Lindeque, M. & Skinner, J. D. (1982). Fetal androgens and sexual mimicry in spotted hyaenas (Crocuta crocuta). Journal of Reproduction and Fertility, 65, 405–410. Martin, R. D. (1972). A preliminary field study of the lesser mouse lemur (Microcebus murinus J. F. Miller 1777), Tierpsychologie Zeitschrift fu¨r, 9, 43–89. Matthews, L. H. (1939). Reproduction in the spotted hyaena, Crocuta crocuta (Erxleben). Philosophical Transactions of the Royal Society B, 230, 1–78. Mills, M. G. L. (1990). Kalahari Hyaenas: The Behavioural Ecology of Two Species. London: Unwin Hyman. Mu¨ller, A. E. & Thalmann, U. (2002). Biology of the fat-tailed dwarf lemur (Cheirogaleus medius E. Geoffroy 1812): new results from the field. Evolutionary Anthropology, S1, 79–82. Mutschler, T. (1999). The Alaotran gentle lemur (Hapalemur griseus alaotrensis): a study in behavioural ecology. Unpublished Ph.D. thesis, University of Zurich, Switzerland. Mutschler, T. (2002). Alaotran gentle lemur: some aspects of its behavioral ecology. Evolutionary Anthropology, S1, 101–104. Mutschler, T., Nievergelt, C. M. & Feistner, A. T. C. (2000). Social organization of the Alaotran gentle lemur (Hapalemur griseus alaotrensis). American Journal of Primatology, 50, 9–24. Nakamichi, M. & Koyama, N. (1997). Social relationships among ring-tailed lemurs (Lemur catta) in two free-ranging troops at Berenty Reserve, Madagascar. International Journal of Primatology, 18, 73–93. Nievergelt, C., Mutschler, T., & Feistner, A. T. C. (1998). Group encounters and territoriality in wild Alaotran gentle lemurs (Hapalemur griseus alaotrensis). American Journal of Primatology, 46, 251–258. Nievergelt, C. M., Mutschler, T., Feistner, A. T. C., & Woodruff, D. S. (2002). Social system of the Alaotran gentle lemur (Hapalemur griseus alaotrensis): genetic characterization of group composition and mating system. American Journal of Primatology, 57, 157–176. Nonacs, P. (2000). Measuring and using skew in the study of social behavior and evolution. American Naturalist, 156, 577–589. Nonacs, P. (2003). Measuring the reliability of skew indices: is there one best index? Animal Behaviour, 65, 615–627. Nunn, C. L. & Pereira, M. E. (2000). Group histories and offspring sex ratios in ringtailed lemurs (Lemur catta). Behavioral Ecology and Sociobiology, 48, 18–28. Overdorff, D. J., Merenlender, A. M., Talata, P., Telo, A., & Forward, Z. A. (1999). Life history of Eulemur fulvus rufus from 1988–1998 in southeastern Madagascar. American Journal of Physical Anthropology, 108, 295–310.
Reproductive skew in female-dominated mammalian societies Packer, C., Pusey, A. E., & Eberly, L. E. (2001). Egalitarianism in female African lions. Science, 293, 690–693. Paoli, T., Palagi, E., & Borgognini Tarli, S. M. (2006) Reevaluation of dominance hierarchies in bonobos (Pan paniscus). American Journal of Physical Anthropology, 130, 116–122. Parga, J. A. (2002). Male dominance rank reversal during the breeding season in ringtailed lemurs (Lemur catta): changes resulting from female mate choice. American Journal of Physical Anthropology, S34, 123. Parga, J. A. & Lessnau, R. G. (2005). Female age-specific reproductive rates, birth seasonality, and infant mortality of ring-tailed lemurs on St. Catherine Island: 17-year reproductive history of a free-ranging colony. Zoo Biology, 24, 295–309. Pemberton, J. M., Albon, S. D., Guinness, F. E., Clutton-Brock, T. H., & Dover, G. A. (1992). Behavioral estimates of male mating success tested by DNA fingerprinting in a polygynous mammal. Behavioral Ecology, 3, 66–75. Pereira, M. E. (1991). Asynchrony within estrous synchrony among ringtailed lemurs (Primates: Lemuridae). Physiology and Behavior, 49, 47–52. Pereira, M. E. & Weiss, M. L. (1991). Female mate choice, male migration, and the threat of infanticide in ringtailed lemurs. Behavioral Ecology and Sociobiology, 28, 141–152. Perret, M. (1992). Environmental and social determinants of sexual function in the male lesser mouse lemur (Microcebus murinus). Folia Primatologica, 59, 1–25. Radespiel, U. (2000). Sociality in the gray mouse lemur (Microcebus murinus) in northwestern Madagascar. American Journal of Primatology, 51, 21–40. Radespiel, U. & Zimmermann, E. (2001). Female dominance in captive gray mouse lemurs (Microcebus murinus). American Journal of Primatology, 54, 181–192. Radespiel, U., Sarikaya, Z., Zimmermann, E., & Bruford, M. W. (2001). Sociogenetic structure in a free-living nocturnal primate population: sex-specific differences in the grey mouse lemur (Microcebus murinus). Behavioral Ecology and Sociobiology, 50, 493–502. Radespiel, U., Dal Secco, V., Drogemuller, C., et al. (2002). Sexual selection, multiple mating and paternity in grey mouse lemurs, Microcebus murinus. Animal Behaviour, 63, 259–268. Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model . Behavioral Ecology, 11, 640–647. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Richard, A. F. (1992). Aggressive competition between males, female-controlled polygyny and sexual monomorphism in a Malagasy primate, Propithecus verreauxi. Journal of Human Evolution, 22, 393–406. Richard, A. F. & Dewar, R. E. (1991). Lemur ecology. Annual Review of Ecology and Systematics, 22, 145–175.
81
82
K. E. Holekamp, A. L. Engh Richard, A. F., Dewar, R. E., Schwartz, M., & Ratsirarson, J. (2000). Mass change, environmental variability and female fertility in wild Propithecus verreauxi. Journal of Human Evolution, 39, 381–391. Richard, A. F., Dewar, R. E., & Ratsirarson, J. (2002). Life in the slow lane? Demography and life histories of male and female sifaka (Propithecus verreauxi verreauxi). Journal of Zoology, London, 256, 421–436. Sauther, M. L. (1991). Reproductive behavior of free-ranging Lemur catta at Beza Mahafaly Special Reserve, Madagascar. American Journal of Physical Anthropology, 84, 463–477. Sauther, M. L., Sussman, R. W., & Gould, L. (1999). The socioecology of the ringtailed lemur: thirty-five years of research. Evolutionary Anthropology, 8, 120–132. Schmelting, B. (2001). Reproductive tactics in male grey mouse lemurs (Microcebus murinus, J. F. Miller 1777) in Northwestern Madagascar. Unpublished Ph.D. thesis, School of Veterinary Medicine, Hannover, Germany. Schmid, J. & Kappeler, P. M. (1998). Fluctuating sexual dimorphism and differential hibernation by sex in a primate, the gray mouse lemur (Microcebus murinus). Behavioral Ecology and Sociobiology, 43, 125–132. Setchell, J. M., Charpentier, M., & Wickings, E. J. (2005). Mate guarding and paternity in mandrills: factors influencing alpha male monopoly. Animal Behaviour, 70, 1105–1120. Sherman, P. W., Jarvis, J. U. M., & Alexander, R. D. (1991). The Biology of the Naked MoleRat. Princeton, NJ: Princeton University Press. Short, R. V. & Balaban, E., eds. (1994). The Differences Between the Sexes. Cambridge: Cambridge University Press. Smale, L., Frank, L. G., & Holekamp, K. E. (1993). Ontogeny of dominance in free-living spotted hyaenas: juvenile rank relations with adults. Animal Behaviour, 46, 467–477. Smale, L., Nunes, S., & Holekamp, K. E. (1997). Sexually dimorphic dispersal in mammals: patterns, causes, and consequences. Advances in the Study of Behavior, 26, 181–250. Smith, D. G. (1993). A 15-year study of the association between dominance rank and reproductive success of male rhesus macaques. Primates, 34, 471–480. Smith, J. E., Memenis, S. K., & Holekamp, K. E. (2007). Rank-related partner choice in the fission–fusion society of the spotted hyena (Crocuta crocuta). Behavioral Ecology and Sociobiology, 61, 753–765. Stanford, C. B. (1998). The social behavior of chimpanzees and bonobos. Current Anthropology, 39, 399–420. Stanger, K. (1995). Vocalizations of some cheirogaleid prosimians evaluated in a phylogenetic context. In L. Alterman, G. Doyle, & M. Izard, eds., Creatures of the Dark. New York, NY: Plenum, pp. 353–376. Sussman, R. W. (1991). Demography and social organization of free-ranging Lemur catta in the Beza Mahafaly Reserve, Madagascar. American Journal of Physical Anthropology, 84, 43–58. Sussman, R. W. (1992). Male life history and intergroup mobility among ringtailed lemurs (Lemur catta). International Journal of Primatology, 13, 395–413.
Reproductive skew in female-dominated mammalian societies Szykman, M., Engh, A. L., Van Horn, R. C., et al. (2001). Association patterns between male and female spotted hyenas reflect male mate choice. Behavioral Ecology and Sociobiology, 50, 231–238. Szykman, M., Engh, A. L., Van Horn, R. C., et al. (2003). Rare male aggression directed toward females in a female-dominated society: baiting behavior in the spotted hyena. Aggressive Behavior, 29, 457–474. Takahata, Y., Koyama, N., Ichino, S., & Miyamoto, N. (2005). Inter- and within-troop competition of female ring-tailed lemurs: a preliminary field report. African Study Monographs, 26, 1–14. Taylor, L. (1986). Kinship, dominance, and social organization in a semi-free-ranging group of ring-tailed lemurs (Lemur catta). Unpublished Ph.D. thesis, Washington University, St. Louis, MO, USA. Taylor, L. & Sussman, R. W. (1985). A preliminary study of kinship and social organization in a free-ranging group of Lemur catta. International Journal of Primatology, 6, 601–614. Tilson, R. T. & Hamilton, W. J. (1984). Social dominance and feeding patterns of spotted hyaenas. Animal Behaviour, 32, 715–724. Van Hooff, J. A. R. A. M. (2000). Relationships among non-human primate males: a deductive framework. In P. M. Kappeler, ed., Primate Males: Causes and Consequences of Variation in Group Composition. Cambridge: Cambridge University Press, pp. 183–191. Van Horn, R. C., Engh, A. L., Scribner, K. T., Funk, S. M., & Holekamp, K. E. (2004). Behavioral structuring of relatedness in the spotted hyena (Crocuta crocuta) suggests direct fitness benefits of clan-level cooperation. Molecular Ecology, 13, 449–458. Vehrencamp, S. L. (1983a). A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667–682. Vehrencamp, S. L. (1983b). Optimal skew in cooperative societies. American Zoologist, 23, 327–355. Vervaecke, H. & van Elsacker, L. (2000). Sexual competition in a group of captive bonobos (Pan paniscus). Primates, 41, 109–115. Vervaecke, H., de Vries, H., & van Elsacker, L. (2000). Dominance and its behavioral measures in a captive group of bonobos (Pan paniscus). International Journal of Primatology, 21, 47–68. Whittingham, L. A., & Dunn, P. O. (1998). Male parental effort and paternity in a variable mating system. Animal Behaviour, 55, 629–640. Wimmer, B., Tautz, D., & Kappeler, P. M. (2002). The genetic population structure of the gray mouse lemur (Microcebus murinus), a basal primate from Madagascar. Behavioral Ecology and Sociobiology, 52, 166–175. Wright, P. C. (1999). Lemur traits and Madagascar ecology: coping with an island environment. Yearbook of Physical Anthropology, 42, 31–72. Young, A. J. & Clutton-Brock, T. (2006). Infanticide by subordinates influences reproductive sharing in cooperatively breeding meerkats. Biology Letters, 2, 385–387.
83
4
The effects of heterogeneous regimes on reproductive skew in eutherian mammals c l a r a b. j on e s
Summary Although several models of reproductive skew have been proposed, “transactional” models, on the one hand, and “tug-of-war” or “incomplete control” models, on the other hand, are most commonly debated. The former hold that dominants control group size by yielding “incentives” (a share of total group productivity) to subordinates, while the latter advance the idea that the dominant’s control over one or more subordinates is incomplete. While high skew has been identified in several mammalian societies, most research on this topic shows that social mammals are likely to display intermediate, low, or variable reproductive skew. In an attempt to explain this pattern of results, the present chapter shows that mammals have evolved to cope with heterogeneous environmental regimes (abiotic and biotic), yielding a eutherian adaptive complex comprised of endothermy, relative brain enlargement, and behavioral flexibility. It is argued that these adaptations to environmental unpredictability favor the evolution of alternative phenotypes as well as situation- and condition-dependent responses decreasing the likelihood that dominants will be able to monopolize subordinates, including females, and that high skew will be observed. Additional research is required to highlight similarities and, most important, differences in the routes to sociality between insects, birds, and mammals (see Vehrencamp 1979). A simple mathematical model is presented linking reproductive suppression by a dominant to his/her influence on a subordinate and consequent Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
84
The effects of heterogeneous regimes on reproductive skew ability to monopolize the subordinate. This treatment shows that, depending upon conditions, both types of skew models are realistic. Reproductive skew is discussed in relation to allocation decisions and the different reproductive tactics of female and male mammals. All other things being equal, reproductive skew in heterogeneous regimes may be limited for taxa with relatively generalized phenotypes, because exposure to unpredictable conditions increases the likelihood of error, uncertainty, and risk of response outcomes, decreasing the ability to control a conspecific.
Introduction Individuals living in reproductive units (e.g. colonies, groups, troops) may share reproduction more or less equally. Alternatively, one or a few individuals may monopolize a group’s reproductive output. The apportionment of reproduction within groups by sex (“reproductive skew”: Vehrencamp 1983) may predict other features of animal sociality, such as agonistic (Hager 2003a) and foraging (Jones 2004) behaviors or responses to predation (Bian et al. 2005). Several recent treatments have argued that theories of reproductive skew may yield general formulations for the evolution of social behavior (Reeve & Keller 1996, Heinze & Keller 2000, Reeve & Emlen 2000, Reeve 2001). There is some disagreement, however, about the relative utility of “transactional” models, on the one hand, and “tug-of-war” or “incomplete control” models, on the other (Johnstone 2000, Hager 2003b), and alternate models of skew besides the most influential ones have also been proposed (e.g. Crespi & Ragsdale 2000). Each of these models analyzes social behavior as rule-governed, evolutionarily stable outcomes of variables (e.g. “ecological constraints”, coefficients of relatedness) predicting the apportionment of reproduction within groups. In concession models of skew, a subtype of transactional skew models, the presence of one or more same-sex subordinates is beneficial to the dominant that may extend “incentives” (i.e. one or more shares of total group reproductive productivity) to the subordinate in order to decrease the likelihood that (s)he will emigrate or escalate an interaction. Incentives are theorized to induce a subordinate to remain in the group and, since it is the dominant who “decides” to yield or not to yield incentives to a subordinate, the dominant ultimately determines group size in these models. In “tug-of-war” models (Clutton-Brock 1998), a class of compromise models of reproductive skew, dominants have “incomplete control” over the reproduction of same-sex group members. In this condition, subordinates may have options other than leaving the reproductive unit (dispersal) if the costs of remaining outweigh the
85
86
C. B. Jones associated benefits. Reproductive skew in groups with “incomplete control” by dominants may often be lower, on average, than reproductive skew in groups described by the classic transactional models if subordinates in the former condition have more behavioral options to escape a dominant’s control (e.g. alternative reproductive behaviors). Although competing skew models can be found in the literature, there is broad agreement that the degree of reproductive skew within groups is a function of three primary factors: (1) dispersal costs, (2) costs of reproduction, and (3) the coefficient of within-group relatedness (r) (Vehrencamp 2000). A primary focus of the present chapter concerns the consequences of environmental heterogeneity for variability in these three factors among social mammals. While no general model of reproductive skew has received broad consensus, researchers have shown that Hamilton’s (1964) seminal treatments require expansion beyond their exclusive emphasis upon coefficients of relatedness (r) and inclusive fitness maximizing (Johnstone 2000, West et al. 2001, 2002, Reeve 2001, Reeve & Shen 2006). In addition, several investigators have pointed out that factors determining reproductive skew may vary within and between taxa (e.g. Emlen 1984, Solomon & French 1997, Lacey & Sherman 1997, Vehrencamp 2000, Crespi 2005). The present chapter treats reproductive repression in heterogeneous regimes as one measure of “repression of competition” (Frank 1995, 2003, 2006) influencing patterns of reproductive skew in social mammals. For this report, environmental heterogeneity is defined as spatial and temporal variation in a stimulus or stimulus array endogenous or exogenous to the organism and representing selection pressures inducing alternative behavioral phenotypes (Piersma & Drent 2003, Jones 2005a). I propose in this chapter that, in general, social mammals, including humans, are adapted to cope with, to manage, and to respond to changing conditions with responses likely to yield intermediate, low, or variable – rather than high – reproductive skew, all other things being equal. A-priori evaluation of social evolution in heterogeneous regimes suggests that environmental unpredictability is likely to decrease the value, accuracy, and thus utility of a dominant’s “staying incentives” or “peace incentives” (see Vehrencamp 1983), leading to conditions favoring “incomplete control.” This hypothetical scenario supports the conclusions of Johnstone (2000) that “concession” models and “tug-of-war” models are not mutually exclusive.
Eutherian adaptations to environmental heterogeneity Extensive reviews of the evolutionary history of mammals may be found elsewhere (Eisenberg 1981, Vaughan et al. 2000, Feldhamer et al. 2004,
The effects of heterogeneous regimes on reproductive skew Rose & Archibald 2005). In Eisenberg’s (1981, p. 7) words, “There have been many natural experiments resulting from the early separation of the various phyletic lines that we recognize today as the class Mammalia”. While this statement highlights diversity among mammals, the present treatment focuses upon particular selection pressures and conserved features of eutherians which may help to explain variable configurations of reproductive skew among social taxa. Following Eisenberg’s (1981; see also Rose & Archibald 2005) report, the morphological grade characterizing mammals was attained by descendants of therapsids during the Triassic, more than 250 Ma. The emphasis in this chapter concerns changes in the mammalian line beginning in the early Cretaceous period (125 Ma), when early placentals first appeared. These groups and related mammalian assemblages (e.g. marsupials) demonstrated dentition characteristic of an insectivore–frugivore trophic level, limiting food resources with an unpredictable dispersion in time and space. The radiations of mammals during the Cretaceous occurred in consort with the radiation of angiosperms, events inducing plant–mammal coevolution. Marsupial and eutherian radiations were distinct and well advanced by the late Cretaceous, and differentiation of modern orders was recognizable by the Eocene (Tertiary period), 54.8–33.7 Ma. Mammalian adaptations to environmental heterogeneity are assumed to have been driven by plant–mammal interactions, factors thought to explain phenotypic plasticity and relative brain enlargement in these animals (Eisenberg 1981, Lillegraven et al. 1987, Jones 1995, Feldhamer et al. 2004, Gingerich 2006, Wilbur & Rudolf 2006).
The eutherian adaptive complex: endothermy, relative brain enlargement, and behavioral flexibility The eutherian adaptive complex was driven by environmental heterogeneity, in particular, variability in food abundance and dispersion (Eisenberg 1981, Jerison 1983, Vaughan et al. 2000, Rose & Archibald 2005, Gingerich 2006, McNab 2006). In brief, first principles of ecology indicate that the size and composition of groups change in response to temporal environmental heterogeneity and may have important consequences for the survival and fecundity of organisms (Roughgarden 1979, Pulliam & Caraco 1984, Wang et al. 2006). Population abundance and structure (including group size) through time is an attribute of resource predictability (Roughgarden 1979). High resource predictability and high resource quality, relatively homogeneous spatial dispersion of resources and resource tracking by the animal population is expected to favor resource defense (e.g. contest competition or territoriality)
87
88
C. B. Jones by individuals or small groups, on average, whereas low resource predictability and large distance or high variation in distance between patches may make resources indefensible (unmonopolizable), yielding large average group size (Schoener 1971, Emlen & Oring 1977, Roughgarden 1979, Pulliam & Caraco 1984). Since temporal unpredictability of resources may be positively correlated with spatial uncertainty (“patchiness”), foraging in groups may reduce average searching time per individual group member. Thus, environmental predictability will be inversely correlated with group size (Wittenberger 1980, Pulliam & Caraco 1984). The resultant population structure has significant consequences for genes and the individuals that carry them (Hewitt & Butlin 1997). Population structure may be evident as subdivision into demographic subunits or groups which represent an evolutionary compromise among those parameters yielding optimal inclusive fitness to individuals (Wilson 1975, Wittenberger 1980, Pulliam & Caraco 1984, Dunbar 1996). As Wilson (1975) pointed out, the frequency distribution of group sizes will be a function of those phenomena leading individuals to join and to leave groups combined with the selection pressures on individual responses to these forces. The parameters determining modal group size in a population, thus, are ultimately expressed as adaptations of individuals to local environments (Wilson 1975, Wittenberger 1980, Dunbar 1996). In the same local conditions, males and females may adopt different adaptive tactics and strategies for optimizing lifetime reproductive success due to the constraints of anisogamy (differential investment in gametes by each sex: Trivers 1972), intragenomic conflict (Burt & Trivers 2006), and/or intersexual conflict (costs imposed by one sex upon the other: Rice 2000). In mammals, these factors have consequences for the reproductive strategies of females, who are expected to adopt those behavioral programs conferring the greatest benefits from the conversion of resources, especially food, into offspring. In the same conditions, the fitness of mammalian males will depend upon their ability to monopolize females (or resources required by females: Bradbury & Vehrencamp 1977, Emlen & Oring 1977, Wittenberger 1980, Wrangham 1980, Nunn 1999) (see Figure 4.1). Eisenberg’s review (1966; see also Eisenberg 1981) demonstrated that most mammals exhibit a spatially dispersed (“solitary”) dispersion pattern in time and space excluding periods of mating and maternal care of young. Apparently, then, the benefits of group life have not outweighed sociality’s costs for most mammalian taxa. Nonetheless, complex social behavior in which adults of one or both sexes exhibit tolerance and consequent association in time and space may be found in several mammalian groups, in particular in
The effects of heterogeneous regimes on reproductive skew
Figure 4.1 An adult female mantled howler monkey (Alouatta palliata palliata) displaying genital hypertrophy which may have evolved with a suite of traits (e.g. female choice) characterizing female emancipation (Emlen & Oring 1977, Jones 2005a: 61–78; 91–92) in this species. Female emancipation reflects the unreliability of female monopolization by males and probably arises as a result of the unpredictability of resources, in particular, food, in time and space. Female emancipation and related characteristics of this species (e.g. multiple mating) are likely to constrain the potential for high skew among males (see Jones 1985, Jones & Corte´s-Ortiz 1998).
carnivores, cetaceans, and primates (Feldhamer et al. 2004; see also Eisenberg 1981, pp. 424–425, Gingerich 2006). Some members of these taxa, in addition to some rodents (e.g. the naked mole-rats, Heterocephalus glaber: Lacey & Sherman 1997; see Solomon & French 1997), exhibit noteworthy mechanisms
89
90
C. B. Jones of reproductive control, cooperation, or division of labor. While several authors have advanced classification systems for social taxa (see review in Jones 2005a, pp. 115–121; also Jones 2005b, p. 20), Crespi’s (2005) recent schema is employed in this chapter. In brief, assessing social states in terms of trade-offs (differential optima) of costs and benefits to reproductive success in different environmental regimes, Crespi (2005, p. 570) envisions three classes of social system: (1) “eusocial,” in which “the trade-off involves two permanently distinct types of individual, and this permanence results in two independently evolving phenotypic systems;” (2) “cooperatively breeding,” in which “the difference between ‘breeders’ and ‘helpers’ is not permanent – individuals are ‘totipotent’ and can switch roles;” and (3) “communal,” in which individuals are also totipotent but “there is only one type of individual, who both breeds and engages in helping.” The adaptive significance of endothermy
Feldhamer et al. (2004, pp. 113–115) provide a detailed summary of the significance of endothermy (maintaining a relatively constant body temperature by means of heat produced inside the body), pointing out that this adaptation explains the ability of mammals to survive “inhospitable environments.” Endothermy becomes, then, the primary component of the eutherian adaptive axis, permitting mammals to remain active over a broad range of environmental conditions (Vaughan 1978). Endothermy is most likely a necessary condition for relative brain enlargement, possibly favored initially in changing environments by the role of the hypothalamus and thyroid gland in regulating body temperature and by benefits associated with increasing endothermy’s basic functions of centralized coordination and control (see Jerison 1983). The adaptive significance of relative brain enlargement
The brain, in particular the neocortex, affords a central location for organismal coordination and control (Eisenberg 1981, Jerison 1983) in response to environmental stimuli. For mammals and many other vertebrates, the neocortex is associated with behavioral flexibility (Matzel et al. 2003, Hsu et al. 2005) and mechanisms of learning that may vary in complexity (e.g. habituation, imprinting, associative processes, social learning, cognition). The mammalian neocortex is also capable of assessing likelihoods of reward and selfish gain in conditions of uncertainty (Rilling et al. 2002, de Quervain et al. 2004, King-Casas et al. 2005, Schultz 2006), features that may enhance fitness in heterogeneous regimes (e.g. Bian et al. 2005, Jansen & Stumpf 2005). Ecological and social factors are thought to be the primary selective factors promoting
The effects of heterogeneous regimes on reproductive skew relative brain enlargement in mammals (Shultz & Dunbar 2005). These factors are not mutually exclusive, since the social decisions that individuals make will reflect resource abundance and dispersion in local conditions, a perspective fundamental to behavioral ecology (Emlen & Oring 1977). Since females are expected to be “energy maximizers” (Schoener 1971) and more sensitive than males (“time minimizers”: Schoener 1971) to ecological regimes, all other things being equal, future investigations of the relative import of ecological and social variables for relative brain enlargement require that males and females be analyzed separately. These fundamental differences between the sexes are expected to have broad implications for the behavioral decisions made by eutherians, including tactics and strategies of reproductive allocation (Feldhamer et al. 2004, Jones 2005a, Kussell & Leibler 2005). Related to the evolution of endothermy and large brains is the observation that both adaptations are energetically costly (Eisenberg 1981, Jones 2005a, Isler & van Schaik 2006), with consequent impacts upon metabolic processes and behavioral, including sociosexual, decisions made by females and males. Relative brain enlargement will be associated with a suite of traits including longevity, late sexual maturity, and iteroparity (Eisenberg 1981, Kappeler & Pereira 2003), characteristics associated with adult over juvenile survival (see, for example, Jones 1997). Temporal and spatial allocation patterns dependent upon “sensing,” learning, and memory rather than stochastic (“bethedging”) tactics and strategies (Jansen & Stumpf 2005, Kussell & Leibler 2005) are diagnostic of many eutherians, and, combined with other mammalian traits (e.g. modes of locomotion and reproduction), probably distinguish these taxa from other families (insects, birds) noteworthy for high grades of sociality. Many eutherians, then, exploited an open niche favoring the capacity for “responsive switching” (Jansen & Stumpf 2005, Kussell & Leibler 2005) in heterogeneous regimes. Responsive switching entails facultative physiological and behavioral changes in the face of environmental fluctuations, with consequent benefits (e.g. adjustment to changing rates of abiotic and biotic environmental change, generation of novel responses) and costs (e.g. the maintenance of requisite molecular and physiological machinery, energetic “neglect” if the environment rarely changes, production of maladapted individuals). Investigating the trade-offs of responsive switching in social mammals may increase our understanding of patterns of reproductive skew, including sociosexual patterns of energy and time allocation. It will also be important to study when and under what conditions stochastic phenotype switching and responsive switching are employed within and between eutherians
91
92
C. B. Jones and other taxa, possibly yielding differential patterns of reproductive skew. Taxa exhibiting responsive switching may be more likely to display low and intermediate reproductive skew, since individuals of these groups are expected to have more control over their responses to environmental change, possibly yielding a broader range of reproductive tactics and strategies to achieve reproduction directly or to avoid control or influence by conspecifics. The adaptive significance of behavioral flexibility
The third component of the mammalian axis, behavioral flexibility, reflects selection in heterogeneous regimes, optimizing genotypic and phenotypic success in conditions of uncertainty or risk (Meyers & Bull 2002, Piersma & Drent 2003, Reader & MacDonald 2003, Jones 2005a; see also Lewontin 1957). Behavioral flexibility may be promoted by the coordination and control functions accompanying relative brain enlargement, facilitating an organism’s adjustment to changing conditions, abiotic (e.g. temperature, humidity) or biotic (e.g. feeding or social regimes), and endogenous (e.g. immunological, hormonal) or exogenous (e.g. mate quality, interaction rates) stimuli (see Mayr 1963, Jerison 1983, Rutherford 2000). Recent evidence demonstrates that mechanisms switching responses from one behavioral pattern to another are genetically induced in mammals (Choi et al. 2005; see also WestEberhard 1979, Jones & Agoramoorthy 2003). Behavioral flexibility, a reversible component of the phenotype (Piersma & Drent 2003), yields alternative phenotypes across time and space for the same individual, whose patterns of response are more or less “totipotent” (the ability of an individual to perform most or all of a society’s roles). Because most social mammals, including humans (Homo sapiens), are communal rather than eusocial or cooperatively breeding, the study of alternative behavioral phenotypes in mammals is primarily a study of polyphenisms (environmentally switched alternative phenotypes) rather than polymorphisms (genotypically regulated alternative phenotypes), characteristic of social insects (Wilson 1971) and naked mole-rats (O’Riain et al. 2000). Totipotency renders relatively generalized phenotypes or the ability to change one’s patterns of response rapidly, all other things being equal, in response to abiotic or biotic (including social) regimes (see West-Eberhard 1979, 2003). This trait also allows an individual to exhibit relatively specialized responses for varying periods of time (e.g. temporarily displaying a helper role: Nicolson 1987, adjusting lactation to food shortages: Dall & Boyd 2004; modifying intensities of offspring competition in response to growth rates: Stockley & Parker 2002, exploiting seasonal opportunities: Gockel & Ruf 2001).
The effects of heterogeneous regimes on reproductive skew The eutherian adaptive complex and reproductive skew Exposure to environmental heterogeneity favored adaptations predisposing eutherians to intermediate, low, or variable levels of reproductive skew, since responses are likely to be condition-dependent and/or facultative (Vehrencamp 2000; see also Waynforth et al. 1998). Condition-dependent responses will be enhanced by large brains and behavioral flexibility favoring alternative behaviors, and some of these responses will be a function of learning and higher-order processes (e.g. imitation, problem solving). Outcomes of sociosexual interactions are likely to be highly variable in many, if not most, eutherian societies, since exposure to environmental heterogeneity will decrease the accuracy of behavioral decisions (e.g. attempts by dominants to repress subordinates’ competition, including reproduction) by increasing response variability and, consequently, error. Further, alternative response patterns may arise where phenotypes are exposed to unpredictable regimes, often creating sociosexual opportunities (e.g. parasitic or cooperative phenotypes: Taborsky 1994). These alternative phenotypes may obtain shares of a group’s reproductive output over the short or long terms, decreasing within-group skew. Large brains and behavioral flexibility may also facilitate decisions to play a “waiting game” (e.g. queueing: Voigt & Streich 2003, Alberts et al. 2003; see Kokko & Johnstone 1999) or to adopt a temporarily unproductive or maladaptive phenotype, possibly imposed by an influential group member (see Jones 2005a). Phenotypic hitchhiking, whereby one individual, usually a subordinate, “hitchhikes” on the phenotype of another, usually a dominant, may permit individuals to “buy time” by adopting an alternative strategy until the abiotic or biotic (including social) environment becomes more auspicious for the expression of optimal responses. Subordinate males, for example, may form a temporary coalition or alliance with one or more dominant males while ascending a group’s dominance hierarchy and/or while waiting for opportunities to displace dominants. For example, Jones (1980) reported a coalition between a young male pretender and a dominant male. Over time, the young male rose to dominant rank in his group, and the former dominant lost this position. A coalition between these two males expelled another high-ranking male from the group. In another case, a high-ranking male was observed to assume temporary subordinate rank within his group while a wound healed (C. B. Jones, personal observation). These and other examples (e.g. where individuals become “helpers” to other group members) of phenotypic hitchhiking have the potential to influence reproductive skew within a group.
93
94
C. B. Jones Research on sex and reproduction in mammals has been dominated by studies of male tactics and strategies until relatively recently (Andersson 1994, Ellis 1995, Jennions & Petrie 1997, Dixson 1998, Shahnoor & Jones 2003). Research is needed to investigate causes and consequences of within-group relative contributions of direct (selfish) reproduction (“reproductive skew”) by mammals of both sexes in an attempt to describe and test patterns and processes of despotic (high skew) or egalitarian (intermediate or low skew) relationships or variations in reproductive skew across conditions (e.g. season, habitat) or groups of the same species. Low, intermediate, or variable reproductive skew among males in polygynous and/or multi-male groups may also be promoted by the virtually ubiquitous tendency of eutherian females to mate multiply (Wolff & Macdonald 2004, Jones 2005a, pp. 100–102). This female adaptation may impose costs of time and of ejaculate production upon males (Preston et al. 2001, Eady & Hardy 2001), a condition likely to constrain high skew in this sex. Investigations are also required to identify those features of mammals, mammalian phylogeny, and mammalian regimes differentiating their patterns of reproductive skew from those found in birds and insects, taxa more predisposed, on average, to high-skew societies than documented for mammals. One possibility is that individualistic strategies are more highly elaborated in social mammals, a condition that would dampen the evolution of high grades of sociality (Jones 2005a) and thus high-skew societies. Most, if not all, groups of social mammals are distinguished by the presence of helpers, usually subordinate females assisting the reproductive output of dominant females, to whom they are often related genetically. High skew has been reported for some social mammals (naked mole-rats: Sherman et al. 1991; male lions, Panthera leo: Packer et al. 1991; some marmoset, Callithrix spp., and tamarin, Saguinus spp., monkeys: Abbott 1993; black rhinoceros, Diceros bicornis: Garnier et al. 2001; male Antarctic fur seal, Arctocephalus gazella: Hoffman et al. 2003; female house mice, Mus musculus: Rusu & Krackow 2004; male whitefaced capuchin monkeys, Cebus capucinus: Jack & Fedigan 2006). Nonetheless, as summarized in Table 4.1, most research on social mammals supports the view that reproductive skew within groups is low, intermediate, or highly variable and, as this chapter suggests, differentially responsive to local conditions. This observation supports the argument that, all other things being equal, mammals are selected to endure and to manage direct exposure to environmental uncertainty, an evolutionary scenario leading not only to the suite of characteristics presented as the eutherian adaptive complex but also to a predisposition to make the “best of a bad job” (see Austad 1984, Brockmann 2001, Ebensperger & Blumstein 2006).
The effects of heterogeneous regimes on reproductive skew Table 4.1 Selected overview of eutherian societies with low, intermediate, or variable reproductive skew. In these studies, skew was generally not calculated quantitatively. Thus, classification is typically based on inferences from researchers’ descriptions. Reproductive Skew Order (species)
(sex)
Notes/references
Low (females)
Coefficient of relatedness poor predictor of
Chiroptera Bechstein’s bat (Myotis bechsteinii)
group structure and reproductive skew (Kerth et al. 2002)
Brown long-eared bat
Low (males)
(Plecotus auritus)
Kin selection did not account for patterns of natal philopatry (Burland et al. 2001)
Primates Brown lemur (Eulemur
Variable (males)
Reproductive skew strongly influenced by
Low (females)
Suppression of competition frequently
fulvus mayottensis) Ring-tailed lemur
female choice (Gachot-Neveu et al. 1999)
(Lemur catta)
initiated by dominants to subordinates (Cavigelli et al. 2003)
Moustached tamarin (Saguinus mystax)
Variable (moderate to high (males)
Woolly spider monkey Low (females and or muriqui
males)
(Brachyteles
Evidence of multiple mating by females (Goldizen 1988, Huck et al. 2005) No evidence of dominance hierarchies or reproductive suppression (Strier 1986, 1992)
arachnoides) Vervet monkey
Variable (males)
(Cercopithecus aethiops)
Results supported “limited control” model of reproductive skew (Whitten & Turner 2004)
Rhesus macaque
Variable (males)
(Macaca mulatta)
Coefficients of relatedness did not predict patterns of reproductive skew, and results supported “limited control” model (Widdig et al. 2004)
Barbary macaque (Macaca sylvanus)
Low (females and males)
Correlation between social rank and reproductive success appears to be absent; low skew a function of weak environmental constraints (Ku¨mmerli & Martin 2005)
Mandrill (Mandrillus sphinx)
Variable (females and males)
Success of dominant males generally high but decreased as number of reproductive males in group increased; skew influenced by heterozygocity (Setchell et al. 2005, Charpentier et al. 2005)
Yellow baboon (Papio cynocephalus)
Variable (males)
Reproductive skew a function of male tenure and group size (Alberts et al. 2003)
95
96
C. B. Jones Table 4.1 (cont.) Reproductive Skew Order (species)
(sex)
Notes/references
Mountain gorilla
Low to intermediate
Results support “tug of war” model of
(Gorilla beringei
(males)
beringei)
reproductive skew since subordinate male can escape monopolization by silverback (Bradley et al. 2005)
Chimpanzee (Pan
Variable (females)
troglodytes)
Linear dominance hierarchies not evident, and females have dispersed configuration of foraging (Pusey et al. 1997)
Human (Homo sapiens)
Variable (females)
Menopause increases reproductive skew among females (Foster & Ratnieks 2005; see also Austad 1997, Packer et al. 1998, Fedigan & Pavelka 2001)
Human (Homo sapiens)
Variable (males)
Promiscuity and slavery increase reproductive skew among males (Dickemann 1997, Betzig, 1997a, 1997b)
Wealthy modern human (Homo sapiens)
Variable (females and males)
“Snowballing” resources increases reproductive skew (Hill & Reeve 2005)
Carnivora Meerkat (Suricata suricatta)
Variable but usually high (females)
Results support “limited control” models of reproductive skew (Clutton-Brock et al. 2001, Young & Clutton-Brock 2006)
Banded mongoose
Low (females)
(Mungos mungo)
No evidence of reproductive suppression of subordinates by dominants or of inbreeding avoidance; egalitarianism among females may be induced by benefits of cooperative breeding (de Luca & Ginsburg 2001, Gilchrist et al. 2004)
Spotted hyena (Crocuta Intermediate to low crocuta)
(males)
Results support “limited control” models of reproductive skew, of which female choice and male tenure are significant determinants (Engh et al. 2002)
African lion (Panthera
Low (females)
leo)
Egalitarianism among females may be induced by benefits of cooperative breeding and high costs of female–female aggression (Packer et al. 2001)
Gray seal (Halichoerus grypus)
Variable (males)
Reproductive skew influenced by tendency of females to mate multiply, in particular, with males at sea (Worthington Wilmer et al. 1999, Ambs et al. 1999)
The effects of heterogeneous regimes on reproductive skew Table 4.1 (cont.) Reproductive Skew Order (species)
(sex)
Notes/references
Intermediate to low
Levels of reproductive skew influenced by
Cetacea Humpback whale (Megaptera
(males)
novaeangliae)
operational sex ratio and consequently low competition among males for mates (Cerchio et al. 2005)
Artiodactyla White-tailed deer
Variable (males)
Some evidence of assortative mating by age
Variable (males)
Evidence of multiple mating by females
(Odocoileus virginianus) Pronghorn antelope
(Sorin 2004)
(Antilocapra americana)
including multiple paternity of litters (Carling et al. 2003)
Soay sheep (Ovis aries)
Intermediate to low
Camargue stallion
Low (females);
(males) (Equus caballus)
variable (males)
Depletion of ejaculate with subsequent matings limits skew (Preston et al. 2001) Male reproductive skew a function of maternal rank, female choice, and success of coalitions (Feh 1990)
Rodentia Wood mouse
Variable (females)
(Apodemus sylvaticus)
High reproductive skew associated with mother–daughter breeding groups and with large differences in maternal investment in litters (Gerlach & Bartmann 2002)
African striped mouse (Rhabdomys pumilio)
Variable (females and males)
Patterns of philopatry and consequent reproductive skew a function of habitat (“social flexibility”) (Schradin & Pillay 2005)
Alpine marmot
Variable (females)
(Marmota marmota)
Reproductive suppression costly for dominant females, and skew decreased with increasing group size (King & Allaine´ 2002, Hackla¨nder et al. 2003)
Yellow-bellied marmot Variable (females) (Marmota flaviventris)
Reproductive skew a function of female group size, and lower skew in larger groups was associated with increased female–female aggression (Armitage & Schwartz 2000)
97
98
C. B. Jones Reproductive skew and reproductive allocation patterns in social mammals All other things being equal, mammalian females are more likely than males to exhibit offspring care (Eisenberg 1966, 1981, Trivers 1972) and cooperation (Jones 2005b, Box 4.1), although both responses may be induced in
Box 4.1. Why are females more likely than males to be social? The simple answer to this question is, Because the benefits of sociality are greater to females than to males. Why? It has been suggested that sociality is most likely to evolve as an energy-saving strategy (Heinze & Keller 2000; see also Jones & Agoramoorthy 2003, Jones 2005b). Females (“energymaximizers”: Schoener 1971) have more to gain, all other things being equal, from energy-saving opportunities, and are expected to be more
The effects of heterogeneous regimes on reproductive skew Box 4.1. (cont.) sensitive than males to energetic costs (Jones 2005b). These conditions will predispose females to sociality (e.g. altruism, philopatry) where these responses deliver energetic gains benefiting inclusive fitness (Queller 1997). The graphical model describes the costs (C ) or benefits (B) to female inclusive fitness (expected lifetime reproductive success) of the relative degree of sociality as a function of differential energy savings, from low (–) to high (þþ). Benefits will increase and then level off as the costs increase linearly (because resources, in particular food, are limiting), and the maximum net benefit (benefit minus cost) to females should occur at threshold x. The location of x will depend upon the positions and shapes of the benefit and cost curves, a function of environmental unpredictability over the short and long terms. Energy limitation may explain why the highest grades of sociality (eusociality and cooperative breeding) are relatively uncommon among mammals (Jones & Agoramoorthy 2003), for which endothermy and large brain size are energetically costly (Eisenberg 1981, Feldhamer et al. 2004; see also Jansen & Stumpf 2005, Kussell & Leibler 2005), and broad patterns of reproductive skew in mammals may correspond to regional differences in resource dispersion in the tropics (Fleming et al. 1987).
conspecifics by force (taking away behavioral control), coercion (imposing costs on non-cooperators), manipulation (exerting influence for selfish advantage), or exploitation (use of another’s resources for selfish ends: Crespi & Ragsdale 2000; see also Frank 1995, 2003, 2006, Helms Cahan 2001, Jones 2005a, 2005b; Figure 4.2). Since most eutherian taxa are characterized by sexual dimorphism, with females the smaller sex (Crook 1972, Eisenberg 1981), and since energy limitation will usually be more important to a female than to a male mammal (Schoener 1971, Jones 2005a, pp. 61–78, 2005b; Box 4.1), female mammals are expected to be more vulnerable to environmental perturbations, particularly in food resources, than a male in similar conditions, all other things being equal. As a result, allocation decisions determining reproductive skew among group-living mammalian females, such as the differential benefits and costs of cooperating with or suppressing another female’s reproduction, or of dispersing away from a natal group rather than remaining, will generally have greater consequences for a female’s lifetime reproductive success. This condition obtains because of a female’s greater expected parental investment compared with males (Trivers 1972), because
99
100
C. B. Jones
Figure 4.2
such decisions are likely to increase female inter-birth intervals, because males are likely to parasitize the reproductive allocation patterns of females (Alexander et al. 1997, Jones 2007), and because sexual conflict will usually benefit males (Rice 2000, Jones 2005a, pp. 61–78). Male mammals, on the other hand, with an inherently higher reproductive capacity than females in the same conditions (Trivers 1972), should benefit more from selfish tactics and strategies (e.g. avoidance of other males) likely to maximize their reproductive potential, and should be capable and willing to bear greater costs (e.g. exposure to heterogeneous conditions or higher tolerance for behavioral inaccuracy and risk) in their pursuit of fertilizable females. Simply, males can afford to make more reproductive errors than females, all other things being equal. This asymmetry between the sexes will dampen the potential benefits that mammalian males, including those with lower resource-holding potential, might gain from associating with their mate(s) or other males, since these gains are likely to be outweighed by the benefits of individualized tactics and strategies, including risk taking, to maximize future reproduction.
The effects of heterogeneous regimes on reproductive skew The literature on mammals supports this preliminary scenario (see, for example, Wilson 1975, Wittenberger 1980, Feldhamer et al. 2004). Mammalian females living in groups, for example, are more likely to be philopatric than are males of the same species in the same conditions (Greenwood 1980), and male mammals are unlikely to be social (Eisenberg 1966, 1981, Wilson 1975). This pattern suggests that dispersal costs, costs of reproduction, and/or the benefits of remaining within matrilines are high for female mammals, and that the benefits of sociality are low for mammalian males. The inherent differences in adaptations to group living between the sexes of this class are likely to explain the outcomes summarized in Table 4.1. As Le Galliard et al. (2005, p. 206) point out, “Sociality typically requires, first, some form of altruistic behavior through which individuals sacrifice their own fitness for the benefit of others . . . and, second, some reduction in individual mobility, allowing for sustained interaction, which exacerbates competition for local resources.” All other things being equal, female and male mammals will differentially weight the costs and benefits of the two requirements relative to their abiotic and biotic environments, including the temporal and spatial distribution of conspecifics (kin and non-kin). The resulting allocation decisions will determine patterns of within-sex reproductive skew within groups. Reproductive skew and reproductive suppression in heterogeneous regimes Crespi & Ragsdale (2000) argue that the apportionment of reproduction within groups results from cooperation (mutualistic benefits), force, persuasion (providing benefits to cooperators), or coercion. These four factors describe competitive relations between group members of the same sex yielding asymmetries in their contribution to total group productivity, and may be determined by repression of reproductive behaviors, usually repression of subordinate reproduction by dominants. More effective reproductive repression (e.g. punishment, policing) will generally lead to higher reproductive skew (but see Young & Clutton-Brock 2006). Factors such as environmental heterogeneity, however, may dampen the potential for high-skew societies by increasing the uncertainty, and thereby error, of any behavioral decision by a potential repressor. Heterogeneous regimes will also favor alternative phenotypes and behavioral flexibility, increasing the likelihood that potential targets of reproductive repression, usually subordinates, will partially or completely escape repression attempts (e.g. by policing, such as infanticide: see Hager & Johnstone 2004, Jones 2005c, 2007, by coalitions and alliances with other group members against the
101
102
C. B. Jones repressor: see Flack et al. 2006; by social parasitism, including phenotypic manipulation: see Jones 2005a, 2005b). Such post-repression events have the potential to significantly modify power relations within mammalian groups (Feh 1990, Jones 2000, 2005d, 2006, Gros-Louis et al. 2003, Kutsukake & Hasegawa 2005, Saj & Sicotte 2005). A simple model by May & Anderson (1990) can be modified to describe the potential benefits and costs of reproductive repression. This model measures the fitness of a parasite as a measure of the parasite’s influence on its host. Assume that the fitness of a potential repressor (a social parasite) can be measured as reproductive rate (R0), a density-dependent value. The effectiveness of reproductive repression may be linked to a repressor’s influence on the recipient (host) of its act (e.g. increased inter-birth intervals, increased mortality rates, or decreased litter size experienced by the recipient will decrease his/her fitness). May & Anderson’s formulation can be written for reproductive repression such that R0 ¼ yðNÞ=ða þ b þ vÞ
where y is effectiveness of reproductive repression, N is population density of potential recipients of the act, a is rate of recipient cost (e.g. mortality rate: see Le Galliard et al. 2005) from intensity of reproductive repression, b is rate of recipient’s cost from all but intensity of reproductive repression, and v is recovery rate (the recipient’s ability to completely or partially escape the deleterious effects of reproductive repression). The report by Bradley et al. (2005) on male mountain gorillas (Gorilla beringei) with low to intermediate skew provides a preliminary example of May & Anderson’s (1990) model where y is incomplete, a is not high because of the subordinate’s ability to adopt alternative behavioral tactics and strategies, b is lower than the costs of dispersal, and v is intermediate to high because of the subordinate’s ability to avoid or escape monopolization by the dominant (“negative reinforcement”: Jones 2002). Following May & Anderson (1990), R0 increases as a decreases when y, b, and/ or v are independent. Under these conditions, reproductive repression would not be favored by selection, since the costs of attempted repression would outweigh its benefits. This situation is likely to arise in heterogeneous regimes where the correlation of abiotic and biotic events is decreased. In such “noisy,” stochastic circumstances, the potential for high reproductive skew should be minimized, and “limited control” models of reproductive skew are more likely to apply. Where y, b, and/or v are correlated, however, reproductive repression should be favored, and the intensity of reproductive repression should be determined by the relative degree of benefit to the actor (the repressor), all
The effects of heterogeneous regimes on reproductive skew other things being equal, consistent with “concession” models of reproductive skew. Correlation of y, b, and/or v determines environmental predictability and may be employed as cues for sociosexual decision making.
Conclusions and prospects The arguments advanced in the present chapter hold that social mammals exposed to unpredictable environments are pre-adapted to heterogeneous regimes, increasing the likelihood that these taxa will exhibit low, intermediate, or variable patterns of reproductive skew compared to social insects and cooperative birds. In future, it will be important to measure environmental heterogeneity (see, for example, Jones 1997) in relation to dispersal costs, costs of reproduction, and within-group relatedness for a broad range of social animals, including humans, in order to explain variations in reproductive skew within and between populations and species. Similar treatments have argued that the benefits of assisting the reproduction of kin and their competitors (West et al. 2002) and the benefits of dispersal (Jones 2005a, pp. 18–25) are a function of local competition, and variations in the intensity of local competition are likely, as well, to strongly influence patterns of altruism and philopatry among all group members whose interests conflict. Since the intensity of local competition will vary with environmental predictability (e.g. Emlen & Oring 1977), reproductive skew should reflect these conditions. Frank (2003, p. 693) asserted, “Repression of competition within groups joins kin selection as the second major force in the history of life shaping the evolution of cooperation.” Future studies need to evaluate the differential significance of dispersal costs, costs of reproduction, and the coefficient of within-group relatedness (r) within and between the orders (insects, birds, and mammals) characterized by expansion of sociality (Table 4.1; see also Vehrencamp 1979, 2000, Kokko & Johnstone 1999, Reeve & Shen 2006). It is expected that, for social mammals, decisions to respond or not to respond are more dependent upon an act’s consequences for the competitive regimes of the actor’s direct (progeny) or indirect (other relatives) kin (Jones 2005c; see also West et al. 2002), rather than r, per se, possibly due to the fluctuating effects of heterogeneous regimes. It is possible that, for eutherians, the benefits of altruism and restricted mobility are more often outweighed by the benefits of dispersal, compared to social insects and birds. Partially supporting these ideas, unpredictable conditions induce increased rates of bisexual dispersal in some social mammals (Jones 1999). Repression of reproduction may be the primary determinant of altruism and limited mobility, and thus reproductive skew, in mammals. Studies are
103
104
C. B. Jones required to identify the range of response thresholds to heterogeneous conditions characteristic of the sexes in each mammal species. Finally, because anthropogenic perturbations continue to increase in frequency, duration, and intensity in many parts of the world, increasing the likelihood of mammal extinctions, patterns of reproductive skew may be linked to the persistence of populations in some areas, elevating the importance of studies of the relationship between reproductive skew and reproductive rate within groups. A recent search for the phrase “reproductive skew” in the Journal of Mammalogy yielded only 13 citations. Hopefully, students of mammals will embrace this topic for a more complete understanding of conserved and derived social traits within the class. Acknowledgments Jack Bradbury, Bob Johnston, Fred Stollnitz, “Griff” Ewer, Don Wilson, Norm Scott, Richard W. “Thor” Thorington, and John Eisenberg were my first models in mammalogy. I am fortunate that these specialists shared their knowledge with me during my earliest years as a student and researcher. I am indebted to my postdoctoral advisor, Dick Lewontin, who, at every turn, addressed stochasticity’s impact on animal, including human, populations. I appreciate the constructive criticism provided by Lee Drickamer and Reinmar Hager on an early version of this chapter. This chapter was discussed by the NESCent Mammal Reading Group (Kathleen Smith, Samantha Hopkins, Samantha Price, Louise Roth, and Clara Jones), whose comments proved useful in revising the manuscript. Chris Maher shared an anecdote about woodchucks highlighting the behavioral flexibility of “solitary” mammals. This chapter is dedicated with gratitude to the memory of R. F. “Griff ” Ewer, from my 1973 field season at Barro Colorado Island, Panama. Supported by the National Evolutionary Synthesis Center (NESCent), NSF #EF-0423641. References Abbott, D. H. (1993). Social conflict and reproductive suppression in marmoset and tamarin monkeys. In W. A. Mason & S. P. Mendoza, eds., Primate Social Conflict. Albany, NY: State University of New York Press, pp. 331–372. Alberts, S. C., Watts, H. E., & Altmann, J. (2003). Queuing and queue-jumping: longterm patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Animal Behaviour, 65, 821–840. Alexander, R. D., Marshall, D. C., & Cooley, J. R. (1997). Evolutionary perspectives on insect mating. In J. C. Choe & B. J. Crespi, eds., The Evolution of Mating Systems in Insects and Arachnids. Cambridge: Cambridge University Press, pp. 4–31.
The effects of heterogeneous regimes on reproductive skew Ambs, S. M., Boness, D. J., Bowen, W. D., Perry, E. A., & Fleischer, R. C. (1999). Proximate factors associated with high levels of extraconsort fertilization in polygynous grey seals. Animal Behaviour, 58, 527–535. Andersson, M. (1994). Sexual Selection. Princeton, NJ: Princeton University Press. Armitage, K. B. & Schwartz, O. A. (2000). Social enhancement of fitness in yellowbellied marmots. Proceedings of the National Academy of Sciences of the USA, 97, 12149–12152. Austad, S. N. (1984). A classification of alternative reproductive behaviors and methods for field-testing ESS models. American Zoologist, 24, 309–319. Austad, S. N. (1997). Postreproductive survival. In S. A. Austad, ed., Between Zeus and the Salmon: the Biodemography of Longevity. Washington, DC: National Academy of Sciences, pp. 161–174. Betzig, L. (1997a). Roman polygyny. In L. Betzig ed., Human Nature: a Critical Reader. Oxford: Oxford University Press, pp. 375–398. Betzig, L. (1997b). Why a despot? In L. Betzig, ed., Human Nature: a Critical Reader. Oxford: Oxford University Press, pp. 399–401. Bian, J., Wu, Y., & Liu, J. (2005). Breeding behavior under temporal risk of predation in male root voles (Microtus oeconomus). Journal of Mammalogy, 86, 953–960. Bradbury, J. W. & Vehrencamp, S. L. (1977). Social organization and foraging in emballonurid bats III: mating systems. Behavioral Ecology and Sociobiology, 2, 1–17. Bradley, B. J., Robbins, M. M., Williamson, E. A., et al. (2005). Mountain gorilla tug-of-war: silverbacks have limited control over reproduction in multimale groups. Proceedings of the National Academy of Sciences of the USA, 102, 9418–9423. Brockmann, H. J. (2001). The evolution of alternative strategies and tactics. Advances in the Study of Behavior, 30, 1–51. Burland, T. M., Barratt, E. M., Nichols, R. A., & Racey, P. A. (2001). Mating patterns, relatedness and the basis of natal philopatry in the brown long-eared bat, Plecotus auritus. Molecular Ecology, 10, 1309–1321. Burt, A. & Trivers, R. (2006). Genes in Conflict. Cambridge, MA: Harvard University Press. Carling, M. D., Wiseman, P. A., & Byers, J. A. (2003). Microsatellite analysis reveals multiple paternity in a population of wild pronghorn antelopes (Antilocapra americana). Journal of Mammalogy, 84, 1237–1243. Cavigelli, S. A., Dubovick, T., Levash, W., Jolly, A., & Pitts, A. (2003). Female dominance status and fecal corticoids in a cooperative breeder with low reproductive skew: ring-tailed lemurs (Lemur catta). Hormones and Behavior, 43, 166–179. Cerchio, S., Jacobsen, J. K., Cholewiak, D. M., Falcone, E. A., & Merriwether, D. A. (2005). Paternity in humpback whales, Megaptera novaeangliae: assessing polygyny and skew in male reproductive success. Animal Behaviour, 70, 267–277. Charpentier, M., Setchell, J. M., Prugnolle, F., et al. (2005). Genetic diversity and reproductive success in mandrills (Mandrillus sphinx). Proceedings of the National Academy of Sciences of the USA, 102, 16723–16728.
105
106
C. B. Jones Choi, G. B., Dong, H., Murphy, A. J., et al. (2005). Lhx6 delineates a pathway mediating innate reproductive behaviors from the amygdale to the hypothalamus. Neuron, 46, 647–660. Clutton-Brock, T. H. (1998). Reproductive concessions and skew in vertebrates. Trends in Ecology and Evolution, 13, 288–292. Clutton-Brock, T. H., Brotherton, P. N., Russell, A. F., et al. (2001). Cooperation, control, and concession in meerkat groups. Science, 291, 478–481. Crespi, B. J. (2005). Social sophistry: logos and mythos in the forms of cooperation. Annales Zoologici Fennici, 42, 569–571. Crespi, B. J. & Ragsdale, J. E. (2000). A skew model for the evolution of sociality via manipulation: why it is better to be feared than loved. Proceedings of the Royal Society of London B, 267, 821–828. Crook, J. H. (1972). Sexual selection, dimorphism, and social organization in the primates. In B. Campbell, ed., Sexual Selection and the Descent of Man, 1871–1971. Chicago: Aldine, pp. 231–281. Dall, S. R. X. & Boyd, I. L. (2004). Evolution of mammals: lactation helps mothers to cope with unreliable food supplies. Proceedings of the Royal Society of London B, 271, 2049–2057. de Luca, D. W. & Ginsberg, J. R. (2000). Dominance, reproduction, and survival in banded mongooses: towards an egalitarian social system? Animal Behaviour, 61, 17–30. de Quervain, D. J., Fischbacher, U., Treyer, V., et al. (2004). The neural basis of altruistic punishment. Science, 305, 1246–1247. Dickemann, M. (1997). Paternal confidence and dowry competition: a biocultural analysis of purdah. In L. Betzig, ed., Human Nature: a Critical Reader. Oxford: Oxford University Press, pp. 311–328. Dixson, A. F. (1998). Primate Sexuality: Comparative Studies of the Prosimians, Monkeys, Apes, and Human Beings. Oxford: Oxford University Press. Dunbar, R. I. M. (1996). Determinants of group size in primates: a general model. In W. G. Runciman, J. Maynard Smith, & R. I. M. Dunbar, eds., Evolution of Social Behaviour Patterns in Primates and Man. Oxford: Oxford University Press, pp. 33–58. Eady, P. E. & Hardy, I. C. W. (2001). Overt versus covert competition in Soay sheep. Trends in Ecology and Evolution, 16, 279–280. Ebensperger, L. A. & Blumstein, D. T. (2006). Sociality in New World hystricognath rodents is linked to predators and burrow digging. Behavioral Ecology, 17, 410–418. Eisenberg, J. F. (1966). The social organizations of mammals. Handbuch der Zoologie, 10, 1–92. Eisenberg, J. F. (1981). The Mammalian Radiations: an Analysis of Trends in Evolution, Adaptation, and Behavior. Chicago, IL: University of Chicago Press. Ellis, L. (1995). Dominance and reproductive success among nonhuman animals: a cross-species comparison. Ethology and Sociobiology, 16, 257–333. Emlen, S. T. (1984). Cooperative breeding in birds and mammals. In J. R. Krebs & N. B. Davies, eds., Behavioural Ecology: An Evolutionary Approach, 2nd edn. Oxford: Blackwell, pp. 305–339.
The effects of heterogeneous regimes on reproductive skew Emlen, S. T. & Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science, 197, 215–223. Engh, A. L., Funk, S. M., Van Horn, R. C., et al. (2002). Reproductive skew among males in a female-dominated mammalian society. Behavioral Ecology, 13, 193–200. Fedigan, L. M. & Pavelka, M. S. M. (2001). Is there adaptive value to reproductive termination in Japanese macaques? A test of the maternal investment hypothesis. International Journal of Primatology, 22, 109–125. Feh, C. (1990). Long-term paternity data in relation to different aspects of rank for Camargue stallions, Equus caballus. Animal Behaviour, 40, 995–996. Feldhamer, G. A., Drickamer, L. C., Vessey, S. H., & Merritt, J. F. (2004). Mammalogy: Adaptation, Diversity, Ecology, 2nd edn. Boston, MA: McGraw-Hill. Flack, J. C., Girvan, M., de Waal, F. B., & Krakauer, D. C. (2006). Policing stabilizes construction of social niches in primates. Nature, 439, 426–429. Fleming, T. H., Breitwisch, R., & Whitesides, G. H. (1987). Patterns of tropical vertebrate frugivore diversity. Annual Review of Ecology and Systematics, 18, 91–109. Foster, K. R. & Ratnieks, F. L. W. (2005). A new eusocial vertebrate? Trends in Ecology and Evolution, 20, 363–364. Frank, S. A. (1995). Mutual policing and repression of competition in the evolution of cooperative groups. Nature, 377, 520–522. Frank, S. A. (2003). Perspective: repression of competition and the evolution of competition. Evolution, 57, 693–705. Frank, S. A. (2006). Social selection. In C. W. Fox & J. B. Wolf, eds., Evolutionary Genetics: Concepts and Case Studies. Oxford: Oxford University Press, pp. 350–363. Gachot-Neveu, H., Petit, M., & Roeder, J. J. (1999). Paternity determination in two groups of Eulemur fulvus mayottensis: implications for understanding mating strategies. International Journal of Primatology, 20, 107–119. Garnier, J. N., Bruford, M. W., & Goossens, B. (2001). Mating system and reproductive skew in the black rhinoceros. Molecular Ecology, 10, 2031–2041. Gerlach, G. & Bartmann, S. (2002). Reproductive skew, costs, and benefits of cooperative breeding in female wood mice (Apodemus sylvaticus). Behavioral Ecology, 13, 408–418. Gilchrist, J. S., Otali, E., & Mwanguhya, F. (2004). Why breed communally? Factors affecting fecundity in a communal breeding mammal: the banded mongoose (Mungos mungo). Behavioral Ecology and Sociobiology, 57, 119–131. Gingerich, P. D. (2006). Environment and evolution through the Paleocene–Eocene thermal maximum. Trends in Ecology and Evolution, 21, 246–253. Gockel, J. & Ruf, T. (2001). Alternative seasonal reproductive strategies in wild rodent populations. Journal of Mammalogy, 82, 1034–1046. Goldizen, A. W. (1988). Tamarin and marmoset mating systems: unusual flexibility. Trends in Ecology and Evolution, 3, 36–40. Greenwood, P. J. (1980). Mating systems, philopatry, and dispersal in birds and mammals. Animal Behaviour, 28, 1140–1162. Gros-Louis, J., Perry, S., & Manson, J. H. (2003). Violent coalitionary attacks and intraspecific killing in wild white-faced capuchin monkeys (Cebus capucinus). Primates, 44, 341–346.
107
108
C. B. Jones Hackla¨nder, K., Mo¨stl, E., & Arnold, W. (2003). Reproductive suppression in female Alpine marmots, Marmota marmota. Animal Behaviour, 65, 1133–1140. Hager, R. (2003a). Models of reproductive skew applied to primates. In C. B. Jones, ed., Sexual Selection and Reproductive Competition in Primates: New Perspectives and Directions. Norman, OK: American Society of Primatologists, pp. 65–101. Hager, R. (2003b). The effects of dispersal costs on reproductive skew and withingroup aggression in primate groups. Primate Report, 67, 85–98. Hager, R. & Johnstone, R. A. (2004). Infanticide and control of reproduction in cooperative and communal breeders. Animal Behaviour, 67, 941–949. Hamilton, W. D. (1964). The evolution of social behavior. Journal of Theoretical Biology, 7, 1–52. Heinze, J. & Keller, L. (2000). Alternative reproductive strategies: a queen perspective in ants. Trends in Ecology and Evolution, 15, 508–512. Helms Cahan, S. (2001). Co-operation and conflict in ant foundress associations: insights from geographical variation. Animal Behaviour, 61, 819–825. Hewitt, G. M. & Butlin, R. K. (1997). Causes and consequences of population structure. In J. R. Krebs & N. B. Davies, eds., Behavioural Ecology: an Evolutionary Approach. Oxford: Blackwell, pp. 350–372. Hill, S. E. & Reeve, H. K. (2005). Low fertility in humans as the evolutionary outcome of snowballing resource games. Behavioral Ecology, 16, 398–402. Hoffman, J. I., Boyd, I. L., & Amos, W. (2003). Male reproductive strategy and the importance of maternal status in the Antarctic fur seal Arctocephalus gazelle. Evolution, 57, 1917–1930. Hsu, M., Bhatt, M., Adolphs, R., Tranel, D., & Camerer, C. F. (2005). Neural systems responding to degrees of uncertainty in human decision-making. Science, 310, 1680–1683. Huck, M., Lottker, P., Bohle, U. R., & Heymann, E. W. (2005). Paternity and kinship patterns in polyandrous moustached tamarins (Saguinus mystax). American Journal of Physical Anthropology, 127, 449–464. Isler, K. & van Schaik, C. P. (2006). Metabolic costs of brain size evolution. Biology Letters, 2, 557–60. Jack, K. M. & Fedigan, L. M. (2006). Why be alpha male? Dominance and reproductive success in wild white-faced capuchins (Cebus capucinus). In A. Estrada, P. A. Garber, M. Pavelka, & L. Luecke, eds., New Perspectives in the Study of Mesoamerican Primates: Distribution, Ecology, Behavior, and Conservation. New York, NY: Springer, pp. 367–386. Jansen, V. A. A. & Stumpf, P. H. (2005). Making sense of evolution in an uncertain world. Science, 309, 2005–2007. Jennions, M. D. & Petrie, M. (1997). Variation in mate choice and mating preferences: a review of causes and consequences. Biological Reviews, 72, 283–327. Jerison, H. J. (1983). The evolution of the mammalian brain as an informationprocessing system. In J. F. Eisenberg & D. G. Kleiman, eds., Advances in the Study of Mammalian Behavior. Shippensburg, PA: American Society of Mammalogists, pp. 113–146.
The effects of heterogeneous regimes on reproductive skew Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Jones, C. B. (1980). The functions of status in the mantled howler monkey (Alouatta palliata Gray): intraspecific competition for group membership in a folivorous Neotropical primate. Primates, 21, 389–405. Jones, C. B. (1985). Reproductive patterns in mantled howler monkeys: estrus, mate choice, and copulation. Primates, 26, 130–142. Jones, C. B. (1995). Howler monkeys appear to be preadapted to cope with habitat fragmentation. Endangered Species UPDATE, 12, 9–10. Jones, C. B. (1997). Life history patterns of howler monkeys in a time-varying environment. Boletin Primatologico Latinoamericano, 6, 1–8. Jones, C. B. (1999). Why both sexes leave: effects of habitat fragmentation on dispersal behavior. Endangered Species UPDATE, 16, 70–73. Jones, C. B. (2000). Alouatta palliata politics: empirical and theoretical aspects of power. Primate Report, 56, 3–21. Jones, C. B. (2002). Negative reinforcement in primate societies related to aggressive restraint. Folia Primatologica, 73, 140–143. Jones, C. B. (2004). The number of adult females in groups of polygynous howling monkeys (Alouatta spp.): theoretical inferences. Primate Report, 68, 7–25. Jones, C. B. (2005a). Behavioral Flexibility in Primates: Causes and Consequences. New York, NY: Springer. Jones, C. B. (2005b). Social parasitism in mammals with particular reference to Neotropical primates. Mastozoologı´a Neotropical, 12, 19–35. Jones, C. B. (2005c). A preliminary test of the Van Schaik model of male coalitions for Costa Rican mantled howler monkeys (Alouatta palliata). Laboratory Primate Newsletter, 44, July, 3–6. Jones, C. B. (2005d). A preliminary test of the Van Schaik model of male coalitions for Costa Rican mantled howler monkeys (Alouatta palliata). Laboratory Primate Newsletter, 44, www.brown.edu/primate/current.html. Jones, C. B. (2006). Exploring the utility of Van Schaik’s model of male coalitions for female mantled howler monkeys (Alouatta palliata). Laboratory Primate Newsletter, 45, www.brown.edu/primate/current.html. Jones, C. B. (2007). The evolution of exploitation in humans: “Surrounded by strangers I thought were my friends”. Ethology, 113, 499–510. Jones, C. B., & Agoramoorthy, G. (2003). Alternative reproductive behaviors in primates: towards general principles. In C. B. Jones, ed., Sexual Selection and Reproductive Competition in Primates: New Perspectives and Directions. Norman, OK: American Society of Primatologists, pp. 103–139. Jones, C. B. & Corte´s-Ortiz, L. (1998). Facultative polyandry in the howling monkey (Alouatta palliate): Carpenter was correct. Boletin Primatologico Latinoamericano, 7, 1–7. Kappeler, P. M. & Pereira, M. E., eds. (2003). Primate Life Histories and Socioecology. Chicago, IL: University of Chicago Press. Kerth, G., Safi, K., & Ko¨nig, B. (2002). Mean colony relatedness is a poor predictor of colony structure and female philopatry in the communally breeding Bechstein’s bat (Myotis bechsteinii). Behavioral Ecology and Sociobiology, 52, 203–210.
109
110
C. B. Jones King, W. J. & Allaine´, D. (2002). Social, maternal, and environmental influences on reproductive success in female Alpine marmots (Marmota marmota). Canadian Journal of Zoology, 80, 2137–2143. King-Casas, B., Tomlin, D., Anen, C., et al. (2005). Getting to know you: reputation and trust in a two-person economic exchange. Science, 308, 78–83. Kokko, H. & Johnstone, R. A. (1999). Social queuing in animal societies: a dynamic model of reproductive skew. Proceedings of the Royal Society of London B, 266, 571–578. Ku¨mmerli, R. & Martin, R. D. (2005). Male and female reproductive success in Macaca sylvanus in Gibraltar: no evidence for rank dependence. International Journal of Primatology, 26, 1229–1249. Kussell, E. & Leibler, S. (2005). Phenotypic diversity, population growth, and information in fluctuating environments. Science, 309, 2075–2078. Kutsukake, N. & Hasegawa, T. (2005). Dominance turnover between an alpha and a beta male and dynamics of social relationships in Japanese macaques. International Journal of Primatology, 26, 775–800. Lacey, E. A. & Sherman, P. W. (1997). Cooperative breeding in naked mole-rats: implications for vertebrate and invertebrate sociality. In N. G. Solomon & J. A. French, eds., Cooperative Breeding in Mammals. Cambridge: Cambridge University Press, pp. 267–301. Le Galliard, J. F., Ferrie`re, R., & Dieckmann, U. (2005). Adaptive evolution of social traits: origin, trajectories, and correlations of altruism and mobility. American Naturalist, 165, 206–224. Lewontin, R. C. (1957). The adaptations of populations to varying environments. Cold Spring Harbor Symposium of Quantitative Biology, 22, 395–408. Lillegraven, J. A., Thompson, S. D., McNab, B. K., & Patton, J. L. (1987). The origin of eutherian mammals. Biological Journal of the Linnean Society, 32, 281–336. Matzel, L. D., Han, Y. R., Grossman, H., et al. (2003). Individual differences in the expression of a “general” learning ability in mice. Journal of Neuroscience, 23, 6423–6433. May, R. M. & Anderson, R. M. (1990). Parasite–host coevolution. Parasitology, 100, S89–101. Mayr, E. (1963). Animal Species and Evolution. Cambridge, MA: Belknap Press. McNab, B. K. (2006). The energetics of reproduction in endotherms and its implication for their conservation. Integrative and Comparative Biology, 46, 1159–1168. Meyers, L. A. & Bull, J. J. (2002). Fighting change with change: adaptive variation in an uncertain world. Trends in Ecology and Evolution, 17, 551–557. Nicolson, N. A. (1987). Infants, mothers, and other females. In B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham, & T. T. Struhsaker, ed., Primate Societies. Chicago, IL: University of Chicago Press, pp. 330–342. Nunn, C. L. (1999). The number of males in primate social groups: a comparative test of the socioecological model. Behavioral Ecology and Sociobiology, 46, 1–13.
The effects of heterogeneous regimes on reproductive skew O’Riain, M. J., Jarvis, J. U. M., Alexander, R., Buffenstein, R., & Peeters, C. (2000). Morphological castes in a vertebrate. Proceedings of the National Academy of Sciences of the USA, 97, 13194–13197. Packer, C., Gilbert, D., Pusey, A. E., & O’Brien, S. J. (1991). A molecular genetic analysis of kinship and cooperation in African lions. Nature, 351, 562–565. Packer, C., Tatar, M., & Collins, A. (1998). Reproductive cessation in female mammals. Nature, 392, 759–761. Packer, C., Pusey, A. E., & Eberly, L. E. (2001). Egalitarianism in female lions. Science, 293, 690–693. Piersma, T. & Drent, J. (2003). Phenotypic flexibility and the evolution of organismal design. Trends in Ecology and Evolution, 18, 228–233. Preston, B. T., Stevenson, I. R., Pemberton, J. M., & Wilson, K. (2001). Dominant rams lose out by sperm depletion. Nature, 409, 681–682. Pulliam, R. & Caraco, T. (1984). Living in groups: Is there an optimal group size? In J. R. Krebs & N. B. Davies, eds., Behavioral Ecology: an Evolutionary Approach. Oxford: Blackwell, pp. 122–147. Pusey, A., Williams, J., & Goodall, J. (1997). The influence of dominance rank on the reproductive success of female chimpanzees. Science, 277, 828–831. Queller, D. C. (1997). Why do females care more than males? Proceedings of the Royal Society of London B, 264, 1555–1557. Reader, S. M. & MacDonald, K. (2003). Environmental variability and primate behavioural flexibility. In S. M. Reader & K. M. Laland, eds., Animal Innovation. Oxford: Oxford University Press, pp. 83–116. Reeve, H. K. (2001). In search of unified theories in sociobiology: help from social wasps. In L. A. Dugatkin, ed., Model Systems in Behavioral Ecology: Integrating Conceptual, Theoretical, and Empirical Approaches. Princeton, NJ: Princeton University Press, pp. 57–71. Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model. Behavioral Ecology, 11, 640–647. Reeve, H. K. & Keller, L. (1996). Relatedness asymmetry and reproductive sharing in animal societies. American Naturalist, 148, 764–769. Reeve, H. K., & Shen, S. -F. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Rice, W. R. (2000). Dangerous liaisons. Proceedings of the National Academy of Sciences of the USA, 97, 12953–12955. Rilling, J., Gutman, D., Zeh, T., et al. (2002). A neural basis for social cooperation. Neuron, 35, 395–405. Rose, K. D. & Archibald, J. D., eds. (2005). The Rise of Placental Mammals. Baltimore, MD: Johns Hopkins University Press. Roughgarden, J. (1979). Theory of Population Genetics and Evolutionary Ecology: An Introduction. New York, NY: Macmillan.
111
112
C. B. Jones Rusu, A. S. & Krackow, S. (2004). Kin-preferential cooperation, dominance-dependent reproductive skew, and competition for mates in communally nesting female house mice. Behavioral Ecology and Sociobiology, 56, 298–305. Rutherford, S. L. (2000). From genotype to phenotype: buffering mechanisms and the storage of genetic information. Bioessays, 22, 1095–1105. Saj, T. L. & Sicotte, P. (2005). Male takeover in Colobus vellerosus at Boabeng-Fiema Monkey Sanctuary, central Ghana. Primates, 46, 211–214. Schoener, T. W. (1971). Theory of feeding strategies. Annual Review of Ecology and Systematics, 2, 369–404. Schradin, C. & Pillay, N. (2005). Intraspecific variation in the spatial and social organization of the African striped mouse. Journal of Mammalogy, 86, 99–107. Schultz, W. (2006). Behavioral theories and the neurophysiology of reward. Annual Review of Psychology, 57, 87–115. Setchell, J. M., Charpentier, M., & Wickings, E. J. (2005). Mate guarding and paternity in mandrills: factors influencing alpha male monopoly. Animal Behaviour, 70, 1105–1120. Shahnoor, N. & Jones, C. B. (2003). A brief history of the study of sexual selection and reproductive competition in primatology. In C. B. Jones, ed., Sexual Selection and Reproductive Competition in Primates: New Perspectives and Directions. Norman, OK: American Society of Primatologists, pp. 1–43. Sherman, P. W., Jarvis, J. U., & Alexander, R. D., eds. (1991). The Biology of the Naked Mole-Rat. Princeton, NJ: Princeton University Press. Shultz, S. & Dunbar, R. I. M. (2005). Both social and ecological factors predict ungulate brain size. Proceedings of the Royal Society of London B, 273, 207–215. Solomon, M. G. & French, J. A. (1997). Cooperative Breeding in Mammals. Cambridge: Cambridge University Press. Sorin, A. B. (2004). Paternity assignment for white-tailed deer (Odocoileus virginianus): mating across age classes and multiple paternity. Journal of Mammalogy, 85, 356–362. Stockley, P. & Parker, G. A. (2002). Life history consequences of mammal sibling rivalry. Proceedings of the National Academy of Sciences of the USA, 99, 12932–12937. Strier, K. B. (1986). The behavior and ecology of the woolly spider monkey, or muriqui (Brachyteles arachnoids E. Geoffroy 1806). Unpublished Ph. D. dissertation, Harvard University. Strier, K. B. (1992). Atelinae adaptations: behavioral strategies and ecological constraints. American Journal of Physical Anthropology, 88, 515–524. Taborsky, M. (1994). Sneakers, satellites, and helpers: parasitic and cooperative behavior in fish reproduction. Advances in the Study of Behavior, 23, 1–100. Trivers, R. L. (1972). Parental investment and sexual selection. In B. Campbell, ed., Sexual Selection and the Descent of Man, 1871–1971. Chicago, IL: Aldine, pp. 136–179. Vaughan, T. A. (1978). Mammalogy. Philadelphia, PA: Saunders. Vaughan, T. A., Ryan, J. M., & Czaplewski, N. J. (2000). Mammalogy, 2nd edn. Fort Worth, TX: Harcourt College Publishers. Vehrencamp, S. L. (1979). The roles of individual, kin, and group selection in the evolution of sociality. In P. Marler & J. Vandenbergh, eds., Handbook of Behavioral
The effects of heterogeneous regimes on reproductive skew Neurobiology: Social Behavior and Communication (Volume 3). New York, NY: Plenum, pp. 351–394. Vehrencamp, S. L. (1983). A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667–682. Vehrencamp, S. L. (2000). Evolutionary routes to joint-female nesting in birds. Behavioral Ecology, 11, 334–344. Voigt, C. C. & Streich, W. J. (2003). Queuing for harem access in colonies of the greater sac-winged bat. Animal Behaviour, 65, 149–156. Wang, G., Hobbs, N. T., Boone, R. B., et al. (2006). Spatial and temporal variability modify density dependence in populations of large herbivores. Ecology, 87, 95–102. Waynforth, D., Hurtado, A. M., & Hill, K. (1998). Environmentally contingent reproductive strategies in Mayan and Ache males. Evolution and Human Behavior, 19, 369–385. West, S. A., Murray, M. G., Machado, C. A., Griffin, A. S., & Herre, E. A. (2001). Testing Hamilton’s rule with competition between relatives. Nature, 409, 510–513. West, S. A., Pen, I., & Griffin, A. S. (2002). Cooperation and competition between relatives. Science, 296, 72–75. West-Eberhard, M. J. (1979). Sexual selection, social competition, and evolution. Proceedings of the American Philosophical Society, 123, 222–234. West-Eberhard, M. J. (2003). Developmental Plasticity and Evolution. Oxford, UK: Oxford University Press. Whitten, P. L., & Turner, T. R. (2004). Male residence and the patterning of serum testosterone in vervet monkeys (Cercopithecus aethiops). Behavioral Ecology and Sociobiology, 56, 565–578. Widdig, A., Bercovitch, F. B., Streich, W. J., et al. (2004). A longitudinal analysis of reproductive skew in male rhesus macaques. Proceedings of the Royal Society of London B, 271, 819–826. Wilbur, H. M. & Rudolf, V. H. W. (2006). Life-history evolution in uncertain environments: bet hedging in time. American Naturalist, 168, 398–411. Wilson, E. O. (1971). The Insect Societies. Cambridge, MA: Belknap Press. Wilson, E. O. (1975). Sociobiology: the New Synthesis. Cambridge, MA: Belknap Press. Wittenberger, J. F. (1980). Group size and polygamy in social mammals. American Naturalist, 115, 197–222. Wolff, J. O. & Macdonald, D. W. (2004). Promiscuous females protect their offspring. Trends in Ecology and Evolution, 19, 127–134. Worthington Wilmer, J., Allen, P. J., Pomeroy, P. P., Twiss, S. D., & Amos, W. (1999). Where have all the fathers gone? An extensive microsatellite analysis of paternity in the grey seal (Halichoerus grypus). Molecular Ecology, 8, 1417–1429. Wrangham, R. W. (1980). An ecological model of female-bonded primate groups. Behaviour, 75, 262–299. Young, A. J. & Clutton-Brock, T. (2006). Infanticide by subordinates influences reproductive sharing in cooperatively breeding meerkats. Biology Letters, 2, 385–387.
113
5
Social skew as a measure of the costs and benefits of group living in marmots t h e a b . w a n g , pe t e r n o n a c s , a n d d a n i e l t . b l u m s t e i n
Summary In group-living animals reproduction is often skewed such that some group members reproduce more than others. In addition to reproductive skew, group members may also exhibit social skew, where some individuals show particular behaviors more often than others. Significant social skew in behaviors such as anti-predator defense or social interactions may influence survival and reproduction. Therefore, social skew has the potential to translate into reproductive skew and affect group productivity. We measured social skew across groups in a population of yellow-bellied marmots (Marmota flaviventris). Several behaviors such as agonistic interactions, affiliative interactions, and first emergence were significantly skewed in most groups. Alarm calling, however, was infrequently skewed more than would be expected by random chance. Thus, marmot groups do not appear to have behavioral roles in terms of individuals acting like sentinels. Although significant social skew was present, it did not obviously affect fitness as measured by female reproductive success for each group. However, skew in individual-directed behavior (e.g. agonistic and affiliative interactions) did significantly correlate with the level of reproductive skew. Finally, the results were independent of the scale at which groups were defined. Behavioral variability appears to occur similarly across the entire marmot population. The results of this study illustrate that the quantification of social skew has potential to be a powerful tool for understanding the evolution of sociality. Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
114
Social skew and group living in marmots Introduction Animal groups are characterized by variation in the degree to which individuals exhibit specific behaviors and variance across group members in the distribution of such variables has come to be called “skew” (Reeve & Ratnieks 1993, Keller & Reeve 1994, Ruzzante et al. 1995, Kokko & Lindstro¨m 1997, Kokko et al. 1999, Nonacs 2000). Within cooperative groups, research has largely focused on reproductive skew (Johnstone 2000, Magrath & Heinsohn 2000, Nonacs 2001, Reeve & Keller 2001). Indeed, the degree of reproductive skew has been proposed as a valuable evolutionary metric for comparing species (Sherman et al. 1995, Lacey & Sherman 2005). Such a eusociality index would differentiate between species by how strictly breeder and non-breeder roles are defined within groups. Similar levels of reproductive skew could therefore imply similar evolutionary pressures across taxonomically very different groups (e.g. ants similar to naked mole-rats and paper wasps similar to cooperatively breeding birds). This eusociality index for comparing species has been criticized by Costa & Fitzgerald (1996, 2005) on the basis that many species show complex cooperative behavior, but do not skew reproduction. However, these authors offer no mathematical solution to this problem, and it may be impossible as a practical matter to have a single skew index that measures variance simultaneously across various social behaviors and reproduction. Instead, Nonacs (2000, 2001) suggested that skew indices could be applied separately across reproductive and non-reproductive behaviors. The degree of skew in nonreproductive behaviors could correlate with outcomes for the group in survival, foraging, or reproduction. Thus, the degree to which individual group members have defined roles may have positive or negative consequences for group success in terms of survival and productivity. Here, we will extend the use of skew to study roles in yellow-bellied marmots (Marmota flaviventris; Figure 5.1), a moderately social ground-dwelling rodent (Frase & Hoffmann 1980). A “role” within a group is defined in this chapter by the presence of significant skew. For example, if the frequency of alarm calling in a group of marmots shows significant skew, this could imply that the more alert individuals are acting as sentinels. Conversely, a lack of skew would imply that there is no sentinel role within groups and all individuals are equally likely to watch for predators. Roles could exist for any activity with variance in events or productivity between individuals. The combination of all the roles will define the structure and patterns of relationships within the group.
115
116
T. B. Wang et al.
Figure 5.1 Yellow-bellied marmot pup, sex unknown. Courtesy of Daniel T. Blumstein.
Skew indices
Skew can be described in over 20 different mathematical ways (Kokko et al. 1999, Nonacs 2000), and, for purposes of comparing social groups, not all methods are equally valid (Nonacs 2003). Suppose we want to measure whether alarm-calling behavior in marmots is skewed. We record K alarm calls over some time period, and if there are N marmots in our defined group, the mean would be K/N alarm calls per marmot. The first problem in quantifying skew is that our null hypothesis of no skew is not an expectation that each marmot gives exactly K/N alarms. This would imply that marmots are somehow dividing up sentinel duty exactly equally, and this would require a behavioral explanation as much as would a finding of significant skew. Instead, our null expectation would be a random distribution around the mean, with some animals calling more and others calling less due to random chance. Therefore, a skew index must have a set value for a random distribution. A second problem arises if all marmots are not present for equal periods of time. If for instance, a marmot does not emerge from its burrow, it would not be expected to give an alarm call. Therefore, apparent skew in alarm calling could be an artifact of different times spent above ground. Any skew index that
Social skew and group living in marmots cannot adjust for differing times spent in the group is not reliably measuring behavioral differences and roles (Crespi & Yanega 1995). Nonacs (2000, 2003) compared a variety of skew indices and recommended the B index as the most useful and reliable under a wide range of assumptions. The B index was found to be sensitive to robust differences in skew and can compare groups with different productivities, sizes, and differential residence times. This allows us to account for random processes and differential survival or duration within a group. Therefore, the B index is resistant to bias due to group numerical characteristics. It was also determined to be more powerful and to have better-defined statistical properties than other skew indices (Nonacs 2000). The B index as we use it here is not a direct replacement for the eusociality index, but rather a way of measuring skew in the performance of any social behavior. The B index calculates variance across i individuals in a group, R(pi – ni/N)2, where pi is the proportion of total events performed or benefits received by the i-th individual and ni is the time the i-th individual spent in the group. Differential time spent in the group is handled by changing N from the number of individuals to the total time spent in the group across all individuals. Thus, variance is the observed proportion of all acts across all individuals, minus their proportional contribution to the total group time. A random distribution has a positive variance, the magnitude of which varies with K and N. Fortunately, we can estimate a random distribution of K events across N individuals as following a binomial distribution (Sokal & Rohlf 1995). The B index, therefore, is the observed variance minus the expected binomial variance, and would equal zero if the behavioral activity was distributed randomly. Social skew
To study behavioral roles in social groups, the group itself must first be defined. Groups can be determined according to geographic boundaries, but these boundaries and therefore group composition may depend on the scale at which interactions occur (e.g. alarm calls can potentially connect individuals that never physically interact). Thus, one individual can be part of many different spatially nested groups. For example, marmots that live in a meadow can be considered as a group. However, within the meadow, members may share different burrow systems and at a finer scale some individuals may be associated with specific burrows. Which geographic level to use for study may depend on the question being asked and its scale of biological relevance. Returning to our example of alarm calling, we could use a definition of a group that contains the active space of a call (i.e. all individuals within earshot), or we could focus on the set of subjects that could both hear and see a potential
117
118
T. B. Wang et al. caller. This distinction is important for the specific benefits that subjects can obtain. Potentially more information might be transmitted by a multimodal signal (i.e. seeing and hearing the caller: Partan & Marler 2005). Biologically meaningful group boundaries, however, may become apparent only after experimental observation when statistical patterns emerge. These patterns could identify group membership from similar individual responses to detected signals or resulting fitness consequences. For any given defined group, it is possible to calculate skew in behavior. There are no a-priori predictions that marmot groups with well-defined behavioral roles would be more or less efficient, in terms of productivity and survival, than groups without defined behavioral roles. Certainly there will be trade-offs in time allocation. If individuals spend less time in social interactions they could allocate more time to foraging, grooming, or other forms of self-maintenance (e.g. K. A. Pollard & D. T. Blumstein, unpublished data). Energy gain and injury avoidance could come from reduced territory defense, when only a few individuals fight off transient individuals. In addition, animals that engage in the majority of the activity could draw predators away from the rest of the group. The rest of the group may be able to forage and rest unnoticed by predators. On the other hand, the existence of behavioral roles could cause group productivity to decline. One individual performing more aggressive interactions could disrupt the other members. An individual dominating all behavioral activity could suppress other group members and prevent them from engaging in personally beneficial activities. In this way, behavioral roles in non-reproductive behaviors could have a strong influence on mating and reproductive skew. A few aggressive individuals could prevent others from gaining enough nutrition to be able to reproduce, or even prevent contact with members of the opposite sex. Do behavioral roles have fitness consequences, and are they evolutionarily important? This might depend on whom they affect and the level of the skew. For example, if juveniles, but not adult females, have structured behavioral roles, these roles might have less of an impact on offspring production. On the other hand, if group success is limited by juvenile survival, then behavioral roles may have more fitness consequences. Roles could also vary in the degree of the consequences of skew. Behavioral roles within groups where dominants completely suppress subordinates from grooming and foraging activities would have a large fitness effect. Groups could also compensate for the effects of behavioral roles. If dominant individuals are able to prevent subordinates from reproducing but they end up having more offspring themselves, total group productivity would stay the same.
Social skew and group living in marmots In this chapter we will examine yellow-bellied marmot reproductive and non-reproductive behavior for evidence of skewed social roles. We will use the B index as a metric for suggesting roles within groups and to identify potential fitness consequences of behavioral skew.
Methods Study animals and study site
Yellow-bellied marmots, a semi-fossorial ground-dwelling sciurid rodent, were studied in the upper East River valley near the Rocky Mountain Biological Laboratory, Colorado, USA (Figure 5.2). This population has been observed continuously since 1962 (Armitage 1991, Ozgul et al. 2006) and social groups subdivide most colony sites (Blumstein et al. 2006). Each year of the study, all subjects were live-trapped and marked (with fur dye to facilitate observations from afar, and with ear tags for permanent identification). Subjects were observed from a distance known not to influence their behavior (this varied by site and ranges from 50 to 200 m), through 8 · 40 binoculars and 15–45 · spotting scopes, during morning (06:30–10:00) and afternoon (16:00– 19:00) activity bouts. All-event recording (Martin & Bateson 1993) quantified social interactions (occurring about once every 20 minutes), and alarm calling
Figure 5.2 Example of yellow-bellied marmot habitat at the Rocky Mountain Biological Laboratory, Colorado. The shown site is called “Picnic” where groups P18, P20, and P21 can be found. Courtesy of Lucretia Olson.
119
120
T. B. Wang et al. (which is much less frequent). For the following analysis of social skew, we focus on data collected in the 2004 active season (April through August). Group composition
Social groups were defined by calculating a simple ratio association index (Cairns & Schwager 1987). The ratio calculates the proportion of observations in which individuals are seen together (at the same burrow entrance) as a fraction of all times they were seen both together and alone. We used the entire set of observations and trap locations from 2004 to calculate ratios. We measured skew across a range of association levels (0.1, 0.3, 0.5, and 0.7). The low 0.1 association level compared groups that included individuals observed in the same general geographic colony (which consists of several burrows) and the higher levels are increasingly smaller subdivisions within the colony. Different association levels allow us to examine behavioral roles at different scales in the population. We used a 0.5 association level to compare skew across different age–sex classes. This level is appropriate for individuals known to share burrows, and has been used in previous studies (Baird & Whitehead 2000, Nanayakkara & Blumstein 2003). The age–sex classes that we considered included: all marmots, all marmots excluding juveniles, adult females and yearling females, adult males and yearling males, only adult females, and only adult males. We used SOCPROG 2.2 (Whitehead 2004), a program for analyzing social structure, to calculate association indices and plot dendrograms that helped us identify social groups. Group names consist of a one- or two-letter prefix that indicates a geographic colony and a specific number. Non-reproductive behaviors
We analyzed skew in individual-directed and group-directed behaviors. Individual-directed behaviors are defined as those that have a clear recipient, such as allogrooming or biting. We grouped individual-directed behaviors into two categories: (1) affiliative or positive, or (2) agonistic or negative (Table 5.1). In contrast, group-directed behaviors may have no targeted individual and potentially a large number of recipients. Examined group-directed behaviors were alarm calling and first emergence. Alarm calls have a demonstrable conspecific warning function (Blumstein et al. 1997). Each time a marmot initiated a bout of alarm calling, we recorded the number of individual calls within a bout, the duration of the bout, and the likely cause for the alarm. For analysis, we used bouts of alarm calls as a measure rather than the number of separate calls within a bout, because these are likely to be directed at the same stimulus. Bouts were considered separate if at least 1 minute elapsed between calls.
Social skew and group living in marmots Table 5.1 Individual-directed behaviors Affiliative (positive)
Agonistic (negative)
Follow another marmot
Posture aggressively
Forage with another marmot
Vocalize aggressively
Lie down with another marmot
Snarl
Greet
Snap
Sniff
Hiss
Play
Displace
Groom
Fight
Nurse Mate
The first marmot to emerge from the burrow each day can be thought of as a “scout.” This marmot incurs an extra degree of predation risk because it may be the first to be detected by a lurking predator. All subsequent marmots potentially gain information from what happens to the first marmot through group eavesdropping. Because the visibility and total activity levels during the end of the summer are reduced (thus making it difficult in our subalpine system for eavesdroppers to benefit), we only looked at emergences before July 1, 2004. First emergence was recorded during observations begun before 07:30 hours, and only when the observer could clearly identify which marmot emerged first. Skew calculation
Skew was calculated using the B index as described in Nonacs (2000). It determines whether K events are spread randomly over i number of individuals in a group. For affiliative and agonistic social behaviors and alarm calls, we transformed for each marmot its observed number of social behaviors (ki) into its proportional contribution to the total number of social interactions (ki/K ¼ pi). Because individuals were observed for different lengths of time, group size was defined relative to the total number of minutes that all individuals were observed (Nt), such that weighted mean group size equals Nt/nmax, where nmax is the maximum time any individual could be present (often equal to the length of time the group was observed). Therefore, the expected proportion of K events performed by the i-th individual is the number of minutes it was observed, divided by the total time (¼ ni/Nt). Observed variance is the sum across all marmots of (pi ni/Nt)2.
121
122
T. B. Wang et al. To calculate the B index value for a group we subtracted the expected variance from the observed variance. The expected variance follows the binomial distribution such that it equals (1 – 1/N)/K. For first emergence, Nt was total number of animals known to be present in a group (independent of whether they did or did not emerge on a particular day), and all ni equalled one. The significance of skew for each group was determined by simulation. We assigned a probability level that the observed B value was due to random chance for each group. All B index values and their significance levels were calculated using Skew Calculator (available at: www.eeb.ucla.edu/Faculty/ Nonacs). Statistically significant skew as measured by the B index values implies that marmots within a group have different behavioral propensities that might be considered behavioral roles. Fitness consequences
The next question is whether strongly defined roles within groups have fitness consequences. Therefore, we correlated observed social skew with reproductive skew (measured again with the B index) and overall reproductive success of groups. Individual reproductive success was measured by the number of pups born to each female. For mixed litters, we calculated the average number of pups per associated adult female (Armitage 2004). Group reproductive success was measured by the total number of pups divided by the total number of adult females in the group. The range of the B index increases with larger group size (N) and both N and K affect the absolute values that represent complete skew and completely equal sharing (see Nonacs 2000 for details on calculating the potential minimum and maximum B index values). Therefore, we cannot use them for across-group comparisons. Instead, we converted the B index values to standardized values by dividing positive and negative B values by the absolute values of the maximum and minimum possible B value (Nonacs 2000). This creates a parameter range of 1 to 1. The adjusted B index should be used with caution because it tends to overweight distributions with less than random variance (Nonacs 2003). For our purposes, however, this does not represent a problem as we do not use the adjusted B index to determine deviations from random expectations. Results
When groups were defined using association indices ranging from 0.1 to 0.7, we found no substantial differences in the proportion of groups that indicated significant skew. This was found for both group-directed and individual-directed behaviors (Figure 5.3). Higher association levels subdivided the
Social skew and group living in marmots
Proportion of significant results
(a)
1
5
Alarm Calling
5
First Emergence
0.8
0.6 7 6 0.4
0.2
9
10
6
11
0 0.1
(b)
0.3 0.5 Association level value
1
Affiliative 7
Proportion of significant results
6 0.8
0.7
Agonistic 8
6 8
9
9 11
0.6
0.4
0.2
0 0.1
0.3 0.5 Association level value
0.7
Figure 5.3 Proportion of significantly skewed groups at each of the four levels of association. The total number of groups for each behavior at that association level is indicated at the top of each bar. (a) Group-directed behavior; (b) Individual-directed behavior.
population into more groups but similar proportions of them were significantly skewed. Not all groups for the four behaviors were used because we could not calculate skew for groups with only 1 member or only 1 event. We
123
124
T. B. Wang et al. found consistently significant skew in first emergence, affiliative, and agonistic social behaviors. At least 60% of the groups at each association level were significantly skewed in these three behaviors. Alarm calling was much less often significantly skewed. A detailed analysis at the 0.5 level of association shows that skew for all behaviors was consistent across different demographic and age–sex classes (Table 5.2). In general, groups that were significantly skewed across all marmots were also skewed when juveniles were excluded from the analysis, and when only one sex was included. In addition, neither age nor sex was a significant predictor for which marmots were the most likely to emerge first in the morning (Table 5.3). Skew in individual or group-directed behaviors did not significantly correlate with a group’s reproductive success. Reproduction itself, however, was significantly skewed across females in four of the seven groups, and marginally skewed in another (Table 5.4). Two groups, however, shared reproduction more equally than would have been predicted by chance, and across all seven groups reproduction was more skewed than expected by chance. There was no significant relationship between a group’s reproductive skew and the mean number of offspring produced per adult female in the group (Figure 5.4), nor did reproductive skew predict the success of individual females. Skew in both individual-directed social interactions was significantly associated with reproductive skew (Figure 5.5). Group P20, however, is an outlier in these comparisons, and if excluded, neither relationship remains significant. Although both individual-directed behaviors exhibited significant levels of skew, there was no significant correlation between the skew of affiliative and the skew of agonistic social behaviors (Figure 5.6). Skew in neither of the group-directed behaviors significantly correlated with reproductive skew.
Discussion Within a social group or neighborhood, functional roles can result from individuals that are recognizably different in their personalities and propensities towards certain actions (Bouchard & Loehlin 2001, Sih et al. 2004). While the majority of the observed yellow-bellied marmot groups exhibited significant skew across several behaviors, the functional implications of this skew were not dramatic. Sentinels and scouts
We found that in almost all groups no sentinel role could be detected through a differential likelihood to issue alarm calls. This result parallels the
125
126
Table 5.2 Skew in behaviors at the 0.5 level of association. For some groups, there were not enough observed behaviors to calculate B values Alarm calling Group All marmots
N
BR10
11
B 0.0077
Scouting P-level
B
0.735
0.0396
P-level 0.034
Social
Social
affiliative
agonistic
B
P-level
B
P-level
0.0106
<0.001
0.0752
<0.001
BR12
4
0.2536
0.283
BR14
48
0.0005
0.486
0.0717
<0.001
0.0274
<0.001
0.0903
<0.001
M5
30
0.0296
0.076
0.0844
0.001
0.0218
<0.001
0.0582
<0.001
P18
22
0.0155
0.333
0.1236
0.057
0.0964
<0.001
0.0723
0.012
P20
15
0.2484
0.030
0.0257
0.841
0.3849
<0.001
P21
22
0.0352
0.769
0.0772
<0.001
0.0169
0.107
T2
5
0.0517
0.483
0.5178
0.033
0.4724
0.005
0.0228
0.014
0.0277
0.094
T8
10
0.4080
0.003
T9
13
0.0107
0.699
0.0373
0.054
BR10
11
0.0077
0.714
0.0396
0.028
BR12
3
0.1903
0.385
0.0897
0.104
0.0303
<0.001
All adults and yearlings 0.0106
<0.001
0.0752
<0.001
0.0342
<0.001
0.0844
<0.001
BR14
20
0.0041
0.545
0.0436
0.004
M5
20
0.0318
0.113
0.0689
0.003
0.0265
<0.001
0.0516
<0.001
P18
13
0.0234
0.202
0.0985
0.109
0.1089
<0.001
0.0679
0.008
P20
6
0.2518
0.033
0.0289
0.851
0.3837
0.003
P21
12
0.0309
0.658
0.0792
<0.001
0.0161
0.117
0.9479
0.024
T2
2
0.0160
0.607
T8
2
0.0409
0.382
T9
13
0.0107
All adult females and
N
0.0043
0.298
0.061
0.0228
P-level
B
0.0373
P-level
B 0.1353
0.015
0.0162
0.010
0.0385
0.036
0.0771
0.003
0.0570
<0.001
0.0525
<0.001
P-level
0.0303
<0.001
0.697
B
B
0.006 P-level
yearling females BR10
7
BR14
15
0.0054
0.419
M5
7
0.0082
0.421
P18
9
P20
3
0.0714
0.101
0.0227
0.004
0.2159
<0.001
0.0711
0.25
0.0570
0.006
0.0566
0.046
0.0025
0.559
P21
8
0.0501
0.772
0.0186
0.203
0.0302
0.007
0.0129
0.227
T9
5
0.0328
0.801
0.0556
0.269
0.0617
0.003
0.0288
0.237
BR10
4
0.0200
0.662
0.0247
0.583
0.0034
0.793
0.0640
0.001
BR14
5
0.1530
0.760
0.0327
0.291
0.0175
0.004
0.3705
<0.001
0.0294
0.565
0.1785
0.048
0.0321
<0.001
0.0291
0.011
0.0738
<0.001
All adult males and yearling males
M5
13
P18
4
P20
3
P21
4
T9
8
0.3259 0.0134
0.0414
0.713
0.5179
0.006
0.1667
1.000
0.1364
<0.001
0.1637
0.073
0.0875
0.045
0.0318
<0.001
0.0077
0.224
0.130 0.362
Social skew and group living in marmots Table 5.3 The distribution of the number of observed first-emergence events across marmots within groups, by age–sex class (no. first / no. in class) Group
Adult females
Adult males
Yearling females
Yearling males
BR10
1/3
2/1
11/4
7/2
BR14
14/8
2/3
7/6
5/2
M5
4/3
3/3
6/4
2/10
P18
5/8
0/1
0/1
0/3
P21
10/7
2/2
1/1
1/2
T9
2/3
0/2
4/2
10/6
Total
36/32
9/12
29/18
25/25
Table 5.4 Reproductive skew among adult females Group
Females
Pups
B value
P
BR10
3
12
0.1111
0.064
BR14
8
31
0.0018
0.447
M5
3
9
0.1481
0.040
P18
8
13
0.0514
0.015
P20
3
12
0.0417
0.927
P21
7
12
0.2163
<0.001
T9
3
14
0.1293
0.022
Means
5.0
14.7
0.0875
<0.001
5.0 Mean pups per female
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 –1
–0.8
–0.6
–0.4
–0.2
0
0.2
0.4
Reproduction B value (adjusted) Figure 5.4 Mean number of pups per female for each group relative to the adjusted B index values (P ¼ 0.526).
127
T. B. Wang et al.
Reproduction B value (adjusted)
0.2 0.1 0 –0.1 –0.2 –0.3 –0.4 –0.5
P20
–0.6 –1
–0.8
–0.6
–0.4
–0.2
0
0.2
0.4
Affiliative B value (adjusted) 0.4 Reproduction B value (adjusted)
128
0.2 0.0 –0.2 –0.4 –0.6 –0.8 –1.0 –0.1
P20 0.0
0.1 0.2 0.3 0.4 Agonistic B value (adjusted)
0.5
0.6
Figure 5.5 Skew in behavior versus skew in reproduction: (a) P ¼ 0.0023 and (b) P ¼ 0.0003 with all the data points. If the outlier point (group P20 in both panels) is removed, there is no significant relationship.
finding of Blumstein et al. (1997) that most age–sex classes call at similarly low rates. In contrast, however, the majority of groups had individuals that were significantly likely to act as scouts (defined as the first marmot to emerge from its burrow in the morning). Emerging first is a risky behavior as this individual would be the first to encounter a waiting predator (predators were observed outside marmot burrows early in the morning), and other marmots would gain information from the experience of the first one out. Overall, which marmots had a propensity to emerge first was not significantly predicted by age or sex. Thus, a tendency to act as a scout may reflect consistent behavioral differences in the personalities of marmots relative to their risk
Social skew and group living in marmots 0.50 0.45 Agonistic B value
0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 –0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
Affiliative B value Figure 5.6 The relationship between index values for agonistic and affiliative behaviors.
taking (Blumstein et al. 2004). In contrast, Armitage et al. (1996) found that adult males spent more time above ground than other age–sex classes. This would suggest that adult males act more often as scouts and have increased opportunities to use other marmots as scouts. Whether scouting truly has a social role requires further study. Individual-directed versus group-directed behaviors
Behavioral skew was more pronounced across groups in individualdirected behaviors than in group-directed behaviors (Figure 5.3). This also suggests stable personality differences across individuals in willingness to positively or negatively interact. These differences are not created by different patterns of interaction across the sexes, as both females and males show the same skew patterns (Table 5.2). We cannot rule out, however, that relatedness differences rather than personality differences play a role in creating skew in individual-directed behavior. We currently do not have precise coefficients of relatedness calculated for the observed animals. Therefore, particularly affiliative pairs may be close relatives, and agonistic pairs may be genetically distant or unrelated. A mixture of such pairs within a group could create an overall skewed pattern of behavior. A pattern of interaction that is based primarily on relatedness would, however, also predict similar and correlated levels of skew across groups in affiliative and agonistic acts. Such a correlation does not exist between these two measures (Figure 5.6), which suggests that aggressive and cooperative behavior may vary separately across individual
129
130
T. B. Wang et al. yellow-bellied marmots. Therefore, at least one of the traits must be independent of genetic relatedness. The results described in this study are not highly sensitive to the size of the defined group as categorized by association index. This implies that the underlying factors affecting social skew are scale-independent. In other words, similar skew at low and high association levels implies that behavioral variability in personality appears to be the same within groups as it is across groups. This was unexpected, because some behaviors are limited to individuals that physically interact (e.g. affiliative behaviors), while others (e.g. alarm calling) can affect individuals with whom the caller has no physical relationship. Lower association indices imply a reduced opportunity for physical interaction. Given variation in the nature of who could be influenced by a behavior, we expected that social skew would be influenced by our definition of group. We also found that within colonies there was no evidence of individuals associating with others into groups that have similar behavioral patterns. It is possible that behavioral roles are constrained in some way, and that groups cannot be that different from each other. Further studies are required to elucidate why social behaviors appear to be scale-independent. Functional implications
We found no relationship between either behavioral or reproductive skew and per capita fitness of females in terms of reproductive success. The degree to which a group had defined behavioral roles did not predict either more or less reproduction. Therefore, within the population we studied there is no evidence for strong selection operating to favor more advanced levels of sociality with pronounced social and reproductive skew. Yellow-bellied marmots appear stably ensconced at an intermediate level of social complexity (Nonacs 2001, Helms Cahan et al. 2002), a finding that parallels other classifications of this species (Armitage 1981, Michener 1983, Blumstein & Armitage 1998). We are unaware of other studies with vertebrates or social insects that have correlated social skew to group success, but the results from other species and populations would be interesting for comparisons. We found two interesting relationships between behavioral skew and reproductive skew. Reproductive skew was positively correlated with affiliative skew and negatively with agonistic skew (Figure 5.5). This suggests that when reproductive skew occurs, it is not created through agonistic actions of a very dominant individual. It seems possible that because the existence of skew in different behaviors is not due to one particular individual, patterns of social skew in groups would be more resistant to changes in group composition. However, it should be noted that the results for reproductive, affiliative, and
Social skew and group living in marmots agonistic skew are strongly influenced by the presence of one group (P20). Our dataset is currently not large enough to determine whether this group is a true outlier, or indicative of a more robust relationship. However, the analysis has identified this one group as behaving in a very different manner from others. Therefore, it will be interesting to pursue why this group appears to have different dynamics within its interactions. Following the entire RMBL population for several more years will allow us to answer this question. Overall, we did not find a large effect of behavioral roles on group productivity. It is possible that our productivity measurement was insensitive to effects of behavioral roles. This could be taken into account by measuring weight gain or some other group factor. In this population of yellow-bellied marmots, productivity could be influenced more by past behavioral interactions or environmental factors, such as snowfall the year before, than by current behavioral interactions. We were also limited by a small number of groups and low event numbers for some behaviors. Increased size in both of these factors could possibly elucidate the effects of behavioral roles.
Conclusion In summary, the study of social skew and the identification of behavioral roles can lead to insight and clarification regarding social evolution and social complexity. Using social skew for comparisons avoids taxa-based terminology, which is an important step in identifying patterns of sociality in social insects and vertebrates (Brockmann 1997). We can look at groups with similar social skew and behavioral roles to identify common constraints between social species as well as look at the effect of common constraints. In this chapter, we have outlined the first steps of an approach to studying group behavior that can potentially be applied across numerous other taxa, and that may become a framework for comparison across groups that do not have clear reproductive skew (see Costa & Fitzgerald 2005). In regard to yellowbellied marmots, there are several future areas of research. For example, additional group- and individual-directed behaviors can be examined to see if the latter have consistently higher levels of skew. Also, the B index identifies the presence of behavioral roles within a group, but not the specific individuals responsible. More detailed analysis may identify predictive characters for particular behavioral types. Finally, it is of interest to be able to follow the same groups across years and investigate whether social skew changes over time and whether there are any fitness consequences. The application of skew metrics to behavioral phenomena may have a very informative future.
131
132
T. B. Wang et al. References Armitage, K. B. (1981). Sociality as a life-history tactic of ground squirrels. Oecologia, 48, 36–49. Armitage, K. B. (1991). Social and population dynamics of yellow-bellied marmots: results from long-term research. Annual Review of Ecology and Systematics, 22, 379– 407. Armitage, K. B. (2004). Lifetime reproductive success of territorial male yellowbellied marmots. Oecologia Montana, 13, 28–34. Armitage, K. B., Salsbury, C. M., Barthelmess, E. L., Gray, R. C., & Kovach, A. (1996). Population time budget for the yellow-bellied marmot. Ethology, Ecology and Evolution, 8, 67–95. Baird, R. W. & Whitehead, H. (2000). Social organization of mammal-eating killer whales: group stability and dispersal patterns. Canadian Journal of Zoology, 78, 2096–2105. Blumstein, D. T. & Armitage, K. B. (1998). Life history consequences of social complexity: a comparative study of ground-dwelling sciurids. Behavioral Ecology, 9, 8–19. Blumstein, D. T., Steinmetz, J., Armitage, K. B., & Daniel, J. C. (1997). Alarm calling in yellow-bellied marmots: II. Kin selection or parental care? Animal Behaviour, 53, 173–184. Blumstein, D. T., Runyan, A., Seymour, M., et al. (2004). Locomotor ability and wariness in yellow-bellied marmots. Ethology, 110, 615–634. Blumstein, D. T., Ozgul, A., Yovovich, V., Van Vuren, D. H., & Armitage, K. B. (2006). Predation risk predicts presence and persistence of yellow-bellied marmot colonies. Journal of Zoology London, 270, 132–138. Bouchard, T. J. Jr. & Loehlin, J. C. (2001). Genes, evolution, and personality. Behavior Genetics, 31, 243–273. Brockmann, H. J. (1997). Cooperative breeding in wasps and vertebrates: the role of ecological constraints. In J. C. Choe & B. J. Crespi, eds., Evolution of Social Behavior in Insects. Ithaca, NY: Cornell University Press: pp. 348–371. Cairns, S. J. & Schwager, S. J. (1987). A comparison of association indices. Animal Behaviour, 35, 1454–1469. Costa, J. T. & Fitzgerald, T. D. (1996). Developments in social terminology: Semantic battles in a conceptual war. Trends in Ecology and Evolution, 11, 285–289. Costa, J. T. & Fitzgerald, T. D. (2005). Social terminology revisited: Where are we ten years later? Annales Zoologici Fennici, 42, 559–564. Crespi, B. J. & Yanega, D. (1995). The definition of eusociality. Behavioral Ecology, 6, 109–115. Helms Cahan, S., Blumstein, D. T., Sundstrom, L., Liebig, J., & Griffin, A. (2002). Social trajectories and the evolution of social behavior. Oikos, 96, 206–216. Frase, B. A. & Hoffmann, R. S. (1980). Marmota flaviventris. Mammalian Species, 135, 1–8. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26.
Social skew and group living in marmots Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 9, 98–102. Kokko, H. & Lindstrom, J. (1997). Measuring the mating skew. American Naturalist, 149, 794–799. Kokko, H., Mackenzie, A., Reynolds, J. D., Lindstrom, J. & Sutherland, W. J. (1999). Measures of inequality are not equal. American Naturalist, 154, 358–382. Lacey, E. A. & Sherman, P. W. (2005). Redefining eusociality: concepts, goals and levels of analysis. Annales Zoologici Fennici, 42, 573–577. Magrath, R. D. & Heinsohn, R. G. (2000). Reproductive skew in birds: models, problems and prospects. Journal of Avian Biology, 31, 247–258. Martin, P. & Bateson, P. (1993). Measuring Behaviour, 2nd edn. Cambridge: Cambridge University Press. Michener, G. R. (1983). Kin identification, matriarchies, and the evolution of sociality in ground-dwelling sciurids. In J. F. Eisenberg & D. G. Kleiman, eds., Advances in the Study of Mammalian Behavior: Stillwater, OK: American Society of Mammalogists, pp. 528–572. Nanayakkara, D. D. & Blumstein, D. T. (2003). Defining yellow-bellied marmot social groups using association indices. Oecologia Montana, 12, 7–11. Nonacs, P. (2000). Measuring and using skew in the study of social behavior and evolution. American Naturalist, 156, 577–589. Nonacs, P. (2001). A life history approach to group living and social contracts between individuals. Annales Zoologi Fennici, 38, 239–254. Nonacs, P. (2003). Measuring the reliability of skew indices: is there one best index? Animal Behaviour, 65, 615–627. Ozgul, A., Armitage, K. B., Blumstein, D. T., Van Vuren, D. H. & Oli, M. K. (2006). Effects of patch quality and network structure on patch occupancy dynamics of a yellow-bellied marmot metapopulation. Journal of Animal Ecology, 75, 191–202. Partan, S. R. & Marler, P. (2005). Issues in the classification of multimodal communication signals. American Naturalist, 166, 231–245. Reeve, H. K. & Keller, L. (2001). Tests of reproductive-skew models in social insects. Annual Review of Entomology, 46, 347–385. Reeve, H. K. & Ratnieks, F. L. W. (1993). Queen–queen conflict in polygynous societies: mutual tolerance and reproductive skew. In L. Keller, ed., Queen Number and Sociality in Insects. Oxford: Oxford University Press, pp. 45–85. Ruzzante, D. E., Hamilton, D. C., Kramer, D. L., and Grant, J. W. A. (1995). Scaling of the variance and the quantification of resource monopolization. Behavioral Ecology, 7, 199–207. Sherman, P. W., Lacey, E. A., Reevem, H. K., and Keller, L. (1995). The eusociality continuum. Behavioral Ecology, 6, 102–108. Sih, A., Bell, A. M., Johnson, J. C., & Ziemba, R. E. (2004). Behavioral syndromes: An integrative overview. Quarterly Review of Biology, 79, 241–277. Sokal, R. R. & Rohlf, F. J. (1995). Biometry, 3rd edn. New York, NY: W. H. Freeman. Whitehead, H. (2004). SOCPROG: programs for analyzing social structure. Halifax, Nova Scotia: Department of Biology, Dalhousie University.
133
6
Explaining variation in reproductive skew among male langurs: effects of future mating prospects and ecological factors re inmar hager
Summary The partitioning of reproduction in a group, or reproductive skew, is affected by multiple ecological, genetic, and population parameters. The ways in which these factors interact to determine reproductive skew have been formally modeled, but few attempts have been made to assess the utility of skew theory in advancing our understanding of primate social systems and observed levels of skew. In the study presented here, I demonstrate how reproductive skew among male Hanuman langurs (Semnopithecus entellus) can be calculated using paternity data. In contrast to expectations, skew was found to increase with the number of males in a group. In a further analysis, I estimated skew for 21 populations of Asian langurs and investigated the ecological and population factors influencing variation in this parameter. The results lend support to model predictions that future mating prospects, as measured in chances of individual survival and number of potential mates, can be a significant predictor of skew. Furthermore, increased population density positively affected skew, which may reflect ecological constraints on dispersal for subordinates. These results suggest that skew theory may offer a useful framework for understanding the effects of multiple factors determining variation in reproductive skew.
Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
134
Male langurs: future mating prospects and ecological factors Introduction One of the key aims of studies in evolutionary biology, behavioral ecology, and primatology is to analyze the factors influencing variation in individual fitness and associated behavioral traits (e.g. Clutton-Brock & Harvey 1979, Krebs & Davies 1993). The interactions between key genetic and ecological parameters and features of animal societies were formally modeled in the late 1970s in the pioneering work of Sandra Vehrencamp, who developed a model of the degree to which fitness can be biased among group members (Vehrencamp 1979, 1983a, 1983b). This skew in individual fitness was assumed to reflect an equilibrium between competition and cooperation in groups, and might be applied to categorize animal societies. In the following 20 years, the initial models were further developed, focusing on the bias in the distribution of reproduction among group members (Reeve & Ratnieks 1993, Emlen 1994, 1995, Keller & Reeve 1994), or reproductive skew. Skew theory investigates associated genetic and ecological correlates of reproductive skew and could thus be used to quantify features of sociality and compare levels of sociality between and within species (Shermann et al. 1995). This has prompted some initial enthusiasm that skew theory might be a good foundation upon which to build a framework to understand sociality, in particular where skew models focused on conditions for stable groups (Reeve & Ratnieks 1993, Keller & Reeve 1994, Reeve & Keller 2004, Reeve & Shen 2006). For example, a feature of group-living animals that can be directly predicted is the degree to which dominants monopolize reproduction (high skew) and under what conditions a more egalitarian distribution of reproduction is to be expected (low skew). In many skew models (to be precise, in transactional models: see below), one can make explicit predictions about the expected stable group size. Its implications for group living address the fundamental question of why individuals should remain in a group rather than disperse. Furthermore, group foraging has been analyzed using the skew framework (Hamilton 2000) as well as other characteristics of sociality such as within-group aggression (e.g. Reeve & Nonacs 1997, Reeve & Keller 1997, Cant & Johnstone 2000) and levels of infanticide (Johnstone & Cant 1999a, Hager & Johnstone 2004). Assuming that levels of reproductive skew reflect a way of quantifying sociality, we can then proceed to analyze the multiple factors that account for variation in this parameter. While key parameters have been identified in the early skew models (i.e. relatedness, group productivity, costs of solitary breeding), the original set has been extended to incorporate individual fighting abilities, costs of offspring production for individual group members, and the
135
136
R. Hager ability to commit discriminate or indiscriminate infanticide (Reeve & Nonacs 1997, Cant & Johnstone 1999, Johnstone & Cant 1999a, Hager & Johnstone 2004), allowing, on the one hand, a much more detailed investigation of skew tailored to different social systems. On the other hand, an ever-increasing number of more detailed models with sometimes disparate predictions (see e.g. Johnstone 2000, Reeve & Keller 2001) may render it more difficult to assess the utility of skew theory. Testing predictions of skew models in primates
Despite the relative popularity of skew theory among theoretical scientists, the majority of empirical studies with tests of model predictions were limited to social insects or birds (Vehrencamp 2000, Reeve & Keller 2001, Magrath et al. 2004), with only a few studies in mammalian systems (e.g. meerkats, Suricata suricatta: Clutton-Brock et al. 2001; naked mole-rats, Heterocephalus glaber: Faulkes & Bennett 2001; marmots, Marmota marmota: Allaine´ 2000). While the initial treatments on skew theory may be of a more technical nature, the paucity of data on paternity (which would allow determining levels of skew), and the focus on social insects have to some degree impeded potential applications of skew theory in the field of primatology. The question at issue is to determine whether our understanding of the ultimate and proximate causes of primate sociality can be advanced by skew theory. More specifically, we can ask if specific features of primate social behavior are better explained using skew theory than by other existing hypotheses (e.g. the priority of access model: Altmann 1962) or if any additional insight might be gained. One of the first attempts to make primatologists aware of the potential scope of skew theory for the study of primate social systems (Hager 2003a) was followed by a small number of studies investigating paternity data in primates and compared model predictions with empirical results to infer evolutionary causes of the observed distribution of reproduction in these species (e.g. common marmoset, Callithrix jacchus: Dietz 2004; mountain gorilla, Gorilla beringei: Bradley et al. 2005; rhesus macaques, Macaca mulatta: Widdig et al. 2004). In the following section, I will briefly review the main concepts of skew theory, summarize the key model predictions, and examine their application to primate systems. Models of reproductive skew
Reproductive-skew models may be classified into two basic categories, transactional and compromise models, while so-called synthetic models attempt to accommodate features of both transactional and compromise models (Johnstone 2000, Reeve 2000, Reeve & Keller 2001, Hager 2003a).
Male langurs: future mating prospects and ecological factors In transactional skew models it is assumed that a dominant, who is in full control of group reproduction, concedes a share of reproduction to subordinates because the presence of the latter increases the group’s (and thus the dominant’s) fitness (see e.g. Buston 2004) – in essence reflecting a “transaction” between dominant and subordinates. Subordinates are limited in dispersing by ecological constraints on breeding outside the group, and may accept a share of the group’s reproduction conceded by the dominant, who will offer just enough fitness benefits to make staying worthwhile for the subordinate. In theory, the share of reproduction offered by the dominant is the minimum share needed to retain the subordinate. Implicit in this argument is that subordinate reproduction occurs at the expense of the dominant’s share, weighted by the effects of relatedness between the two. This basic framework is referred to as a concession model. A variation on this treatment has been developed (the restraint model), in which it is assumed that dominants only control group membership but that reproduction is controlled by subordinates (Johnstone & Cant 1999b). By contrast, compromise models ignore constraints imposed by group stability and assume that control over group reproduction is incomplete. Rather than assuming a transaction between group members, each individual struggles in a kind of tug-of-war for its share, which is determined mainly by individual fighting abilities and relatedness between group members (Reeve et al. 1998, Johnstone 2000). In compromise models, ecological factors do not play a direct role in affecting skew. Finally, the subordinate is assumed to be less efficient in converting group resources into individual reproduction than the dominant. Skew models thus use a small set of key genetic and ecological variables (e.g. relatedness, cost of dispersal and offspring production, fitness of subordinates and dominants: see Reeve & Keller 2001, Hager 2003a) to explain, in a simplified manner, variation in levels of reproductive skew. For instance, concession models predict that skew increases with dispersal costs because the dominant concedes a lower share of reproduction due to the limited breeding options for subordinates outside the group. A comprehensive summary of the main model assumptions and predictions can be found in Hager (2003a), Johnstone (2000), and Reeve & Keller (2001). To date, two of three studies in primates that have used skew theory to analyze the bias in reproduction (rather than just investigating differences in individual reproductive success outside this framework) found some support for compromise models (Widdig et al. 2004, Bradley et al. 2005). By contrast, one study reported that neither transactional nor compromise models can adequately elucidate the diversity of social and mating systems in
137
138
R. Hager cooperatively breeding callitrichids (Dietz 2004). Work in a semi-free-ranging population of rhesus macaques on Cayo Santiago found no evidence that the pattern of reproductive skew followed the predictions of concession models (Widdig et al. 2004). For instance, distantly related males had a lower reproductive success than close relatives of the dominant male, which is in contrast to concession-model predictions. Further lack of support for concession models comes from a study in wild mountain gorillas, which showed that neither the dominant nor the subordinate male had full control over reproduction, suggesting a tug-of-war situation described by compromise models (Bradley et al. 2005). It should be noted, however, that none of the above studies set out to explicitly test predictions of compromise versus transactional models. Conclusions were based mainly on lack of support for some predictions of one class of models rather than a test between the two (see also Chapter 10 on testing predictions of skew models). Future prospects and skew
Skew theory has advanced over the years, and a number of new models have been developed, introducing additional parameters that are thought to affect reproductive skew, including the benefits of inheriting breeding status (Kokko & Johnstone 1999, Cant & Johnstone 1999, 2000, Hager & Johnstone 2004). In primatology, it has long been acknowledged that rank as well as chances of inheriting a dominant position affect individual reproductive success in both sexes (Cowlishaw & Dunbar 1991, Ellis 1995, Packer et al. 1995, Alberts et al. 2003). Kokko & Johnstone (1999) formally modeled the effects of this on the distribution of reproduction in a group. Skew is said to be influenced by the subordinate’s chances of inheriting breeding status at a later stage because in weighing the costs and benefits of leaving (and attempting to breed elsewhere) subordinates may remain in the group and “queue” for their turn. Under certain conditions, even if subordinates never obtain any concessions from the dominant, staying is still better because of the chances to breed later; skew will thus be lower than if the future prospects of mating were low. In other words, a subordinate’s decision to leave or remain in the group is likely to be influenced by its future prospects to reproduce (Stacey & Ligon 1987). Moreover, results of the model have shown that the delayed benefits of queuing may have a far greater effect on skew and group composition than other key parameters such as relatedness. In primates, this line of reasoning is supported by the observation of social queuing in wild savannah baboons (Papio cynocephalus). Investigating 32 group years, Alberts et al. (2003) showed that individuals may attain dominant status or higher ranks, often by waiting, and that this significantly affects the level of (mating) skew.
Male langurs: future mating prospects and ecological factors Given these results of empirical and theoretical studies, it seems crucial to incorporate measures of future mating prospects in analyses that seek to explain variation in reproductive skew in primates. From this inference follows the question: which factors are most likely to influence an individual’s future prospects? An individual’s prospects of gaining matings in the future will be determined first by its chances of survival, and second by the number of potential mates and competitors. Both factors, in turn, mutually depend on multiple ecological and genetic factors such as resource availability, and it is one key (and difficult) task in studies of skew to quantify these measures in a meaningful way. The study system: Hanuman langurs
Langurs appear to be a good system in which to explore whether skew theory can further our understanding of primate social systems and the underlying causes for variation in the distribution of reproduction. First, social organization is extremely diverse within and between species, male dispersal is high, and individuals are likely to face the trade-offs as described in skew models. Second, it is possible to obtain data on paternity (for one population at present: Ramnagar) as well as on other ecological and population variables, thus allowing an analysis of the factors influencing skew. Finally, many studies suggest that key assumptions of skew models are met, such as control over reproduction and the potential for group benefits due to subordinates’ presence, the potential for queuing, or the propensity to disperse. Hanuman langurs (Presbytis entellus) live in one-male or multi-male groups consisting of up to 120 individuals throughout the Indian subcontinent in habitats ranging from desert to mountains (Hrdy 1977, Winkler et al. 1984, Koenig & Borries 2001). Male dispersal patterns are characterized by dispersal from the natal group or secondary dispersal. Further, males may form all-male bands that attempt to take over other groups (Rajpurohit & Sommer 1991, Newton & Dunbar 1994, Borries 2000). While it is unverified whether the dominant male fully controls reproduction in multi-male groups, he certainly dominates the reproductive output of the group and is also most likely to interrupt subordinate matings (Sommer 1989, Newton & Dunbar 1994). The presence of subordinates in the group may confer several advantages to the dominant, for example in the context of predator detection and communal foraging. In addition, it has been observed that subordinate resident males may act as infant protectors (Borries et al. 1999) and that subordinate males were tolerated by the dominant to prevent a rival from entering the group (Hrdy 1977). Further details on langur life history may be found for example in Koenig & Borries (2001) and Hrdy (1977).
139
140
R. Hager
Objective of the study
The aim of this study is first to show how reproductive skew among primate males can be calculated and how ecological and population factors influencing variation in levels of skew can be investigated, with a specific focus on the effects of future mating prospects. Second, I ask the question whether the explanations given by skew theory present a realistic way of interpreting the effects of multiple variables on skew in this system and whether we gain additional insights from applying skew theory to understand sociality in langurs, in particular in the light of existing theories such as the priority-of-access model (Altmann 1962). Third, I highlight a number of assumptions in skew models that need to be revisited in order to give a more realistic description of the factors affecting male skew. The study presented here suggests directions for future analyses of skew in primates that include both ecological and population variables as predictors of skew. I do not attempt, however, to explicitly test whether skew in langurs is best described by a tug-of-war or a transactional mechanism. Methods Measuring reproductive skew
While the concept of reproductive skew seems an intuitively straightforward way of measuring how reproduction is shared among group members, the quantification of skew has been more contentious (Tsuji & Kasuya 2001, Nonacs 2003). At one end of a continuum, high skew occurs if one or a few individuals monopolize reproduction in a group, whereas there is no skew when reproduction is egalitarian. Thus, reproductive skew refers to the bias in reproduction among individuals of the same sex (Reeve & Keller 1995, Clutton-Brock 1998, Johnstone 2000). Essentially, a group of animals is divided into non-breeders and breeders of one sex, and skew is calculated among the breeders, although the definition of who should be included in this category may prove more difficult in some species (see e.g. Chapter 9). In most skew indices, the level of skew ranges from 0 (even distribution of reproduction) to 1 (monopolized reproduction), although other measures have been devised (Kokko et al. 1999, Tsuji & Kasuya 2001). On statistical and mathematical grounds, a number of potential problems when measuring skew have been pointed out over the years (Nonacs 2003). First, some level of skew will arise purely due to random processes. Thus, it is problematic that a level of skew is returned in the analysis which in fact does not exist. Second, when mean reproductive output is very low among group members, the level of
Male langurs: future mating prospects and ecological factors skew may appear high simply due to sampling error (Tsuji & Kasuya 2001). To address these problems, initial skew indices have been refined, leading to a large number of indices at hand (Kokko et al. 1999, Tsuji & Kasuya 2001). Which index should be chosen when studying reproductive skew and its underlying causes in primates? To date, only two indices have been commonly applied in empirical studies: Nonacs’ B index (Nonacs 2000) and Pamilo & Crozier’s S3 index (Pamilo & Crozier 1996), mainly because of their reliable statistical properties. In addition to the need to accommodate statistical accuracy, practical considerations are likely to play a role in the choice of index, such as the availability of specific data. In studies on skew in primates, Nonacs’ B index has been applied to measure reproductive skew in groups of mountain gorillas (Bradley et al. 2005) and in male rhesus macaques (Widdig et al. 2004) because it allows comparing groups of different sizes and productivity. The calculation of this index requires data on the tenure of individuals, although this will be simplified if all individuals were present throughout the period over which skew is measured. However, if no information on individual tenure is available, or if it is incomplete, a different index may be chosen. In the dataset used in this study (Launhardt et al. 2001), a given subordinate’s contribution to the group’s reproductive output was unknown and could thus not be related to measures of tenure. Therefore, and due to uncertainties in the tenure of subordinates, I shall use the effective-number-ofbreeding-individuals index (Pamilo & Crozier 1996), which has been used in other empirical work on skew (Vehrencamp 2000). Calculating levels of skew in Hanuman langurs
I calculated reproductive skew in Hanuman langurs from populations in Ramnagar, India, using the effective-number-of-breeding-individuals skew index S3 (Pamilo & Crozier 1996) and published paternity data (Launhardt et al. 2001). This index measures skew as a function of the total number of reproductive individuals and their individual reproductive contribution, and is expressed as X Skew ¼ N T 1= p2i =ðNT 1Þ
ð6:1Þ
where NT is the total number of actually reproductive individuals and pi is the reproductive contribution of the i-th individual. Skew was calculated for each group and year for which paternity data were available, rather than per group across years, because mean tenure of alpha males in multi-male groups was only 1.47 mating seasons and subordinates as well as females were joining and leaving the groups across years in both one-male and multi-male groups
141
142
R. Hager (Launhardt et al. 2001). This pattern is caused by multiple social and ecological factors, which in turn may affect an individual’s trade-off of staying versus leaving (subordinates) or conceding a share of reproduction to subordinates or not (dominant). In another study in macaques, reproductive skew has been calculated for each year (Widdig et al. 2004), while Bradley et al. (2005) used longterm data and calculated skew per group (see also below in the Discussion). The original study on paternity assigned the same share of the overall proportion of reproduction to non-alpha resident males while only the alpha male’s contribution was determined (Launhardt et al. 2001). Thus, I assumed an equal share of reproduction among the adult subordinates in my analysis. However, a rank order may exist among subordinate males (see e.g. Borries 2000), which, in turn, may result in an unequal share of reproduction among them. In order to estimate how this would affect calculated levels of skew, I have conducted an additional analysis assuming that only the highest-ranking subordinate obtained any successful matings. Furthermore, it should be noted that skew in one-male groups will not invariably be complete because unidentified non-resident males are able to sire young (Launhardt et al. 2001), possibly leading to lower skew in one-male groups with multiple offspring (e.g. Ohsawa et al. 1993). With a few exceptions (Jodhpur), all Asian langur populations are characterized by the existence of one-male groups (Srivastava & Dunbar 1996), which have thus been included in the calculation of skew. Groups in which no young were born were excluded both from the calculation of skew and from the analysis of factors that may affect skew. Note that there is one dominant male per group. Estimating and analyzing reproductive skew in langur populations
In order to explore which ecological and population factors may affect variation in levels of skew in populations for which no paternity data are available, I first analyzed data on the Ramnagar population to find variables that significantly predict skew and are available for most Asian langur populations. Using the Ramnagar dataset, I conducted a quadratic regression analysis that predicts the level of skew using the number of males as the independent variable. Then I collated data from the literature on 11 original or derived variables from 21 Asian langur populations (Srivastava & Dunbar 1996, Treves & Chapman 1996). Table 6.1 lists the parameters that are used in the skew analysis. Ecological parameters
As a measure of ecological constraint for a given population, I followed Emlen’s model (1982) and used the amount of annual rainfall in addition to the
Male langurs: future mating prospects and ecological factors Table 6.1 Ecological and life-history parameters used in the GLM analysis of potential predictors of estimated levels of skew in 21 Asian langur populations. Details of individual parameters are given in the text. Ecological Annual precipitation (mm) Number of months with less than 50 mm of rain Predation risk Dispersal costs Life-history/population Number of potential mates per male Population density (n individuals per km2) Proportion of one-male groups in the population Number of non-group males per bisexual group Birth rate Home-range size (km2) Birth index
number of months per year with less than 50 mm of rain. High levels of rainfall may reflect overall better conditions (Emlen 1982, Clutton-Brock et al. 2001). Predation risk was assigned following Treves & Chapman (1996) as an index of disturbance of the predator community whereby 1 refers to an intact and 0 to a disturbed community. In the latter case, the risk of predation is assumed to be lower than in the former (Treves & Chapman 1996). Furthermore, in transactional skew theory, dispersal costs play a crucial role in determining whether an individual will leave and attempt breeding outside the focal group or stay and accept a share of reproduction (e.g. Clutton-Brock 1998, Reeve & Emlen 2000, Hager 2003b). While it is widely accepted that dispersal costs are affected by multiple variables such as predation risk or uncertainties of resource availability (e.g. Moore & Ali 1984), quantifying costs of dispersal has been consistently difficult (e.g. Greenwood 1980, Pusey & Packer 1987). Given the data available, here I have approximated dispersal costs as a composite measure of predation risk and population density, for which published data from langurs are available. First, predation risk is generally accepted as a key cost of dispersal such that costs increase with higher predation risk due to unfamiliar habitat, loss of group protection, etc. (e.g. Isbell 1994, Alberts & Altmann 1995). Assuming that population density at moderate to high densities reflects more saturated habitats (Sterck 1998), the resulting increase in competition for limited resources such as food and mates is thought to affect the costs of dispersal (Tanner 1966, Emlen 1982, Morris 1992, 2002). In the
143
144
R. Hager analyses presented here, I have assumed that costs of dispersal increase with both predation risk and population density. In order to quantify this measure, population density was divided into three categories: high, intermediate, and low, the boundaries defined by the upper and lower limits of the first, second, and third quartiles. Predation risk was used as a categorical variable as described above. Both factors were then used in an additive model such that, for example, dispersal costs were highest when both population density and predation risk were high. Life-history/population parameters
In order to approximate future male mating prospects, I calculated the number of females per male in a given group or population, assuming that, everything else being equal, more females per male reflect better chances of mating per male. Furthermore, I added the number of extra-group males per bisexual group, which may serve as an indicator of potential intrasexual competition to monopolize groups. The birth rate in a population refers to the number of births per female per year, and the birth index, ranging from 1 to 12, denotes seasonality. Small numbers in this index indicate that births occur during a few months per year (1 means that all births occurred during one month) while the maximum value, 12, refers to an even distribution of births over the year (Srivastava & Dunbar 1996). Finally, home-range size has been included. Reproductive skew in all populations was estimated using a quadratic regression. Data on within-group relatedness were not available. As a final step in the analysis, I employed general linear models to assess the effects of the above parameters on levels of skew using a stepwise approach (Sokal & Rohlf 1995, Grafen & Hails 2002) with predation risk, birth index, and dispersal costs as categorical variables and the remainder as covariates. Error structures were checked for normality. Results Reproductive skew in the Ramnagar population
Using the effective-number-of-breeding-individuals skew index (Equation 6.1; Pamilo & Crozier 1996) and published paternity data (Launhardt et al. 2001), I first calculated levels of reproductive skew in the Ramnagar populations following the assumption that all subordinates obtained an equally large share of reproduction (Launhardt et al. 2001). However, if a hierarchy exists among subordinates it is reasonable to assume that reproduction may be unequally distributed, with higher-ranking subordinates gaining a greater share of the offspring not sired by the dominant in the group. This, in turn,
Male langurs: future mating prospects and ecological factors Table 6.2 Skew in the Ramnagar population. Adapted and modified from Launhardt et al. (2001). It should be noted that the number of infants sired by the dominant male represents a statistical measure of the probability of paternity (Launhardt et al. 2001). Skew 2 assumes that all offspring not sired by the alpha male were fathered by the highest-ranking subordinate. Group O-troop
P-troop
A-troop
X-troop B-troop
Year
Adult males
Infants
Alpha sired
Skew
Skew 2
1992
5
2
1.57
0.852
0.873
1993
5
9
6.21
0.750
0.813
1994
4
3
2.57
0.883
0.891
1995
7
5
4.57
0.967
0.967
1996
3
7
2.14
0.005
0.631
1992
3
7
1.14
0.173
0.813
1994
3
3
2.57
0.828
0.837
1995
3
3
1.57
0.210
0.502
1992
2
2
1.57
0.490
0.490
1993
1
1
1
1
1
1994
1
2
2
1
1
1995
1
2
2
1
1
1993
1
5
5
1
1
1994
1
1
1
1
1
1995
1
5
5
1
1
may affect the level of overall skew. Therefore I performed an additional analysis assuming that all young not sired by the dominant male were fathered by the highest-ranking subordinate. Table 6.2 shows that skew is complete in all one-male groups and ranges from very high levels to an almost egalitarian distribution in multi-male groups. Reproductive skew can vary strongly over the years, even within groups. For example, in the O-troop skew reaches a peak in 1995 with a level of 0.967 while in the following year reproduction was shared almost equally among males. The variation of skew across years and groups is less if we assume that the highest-ranking subordinate sired all young that were not fathered by the dominant. Comparing levels of skew assuming an equal subordinate share to those when assuming a monopolized subordinate share for the same year and group we find only small differences, with the exception of the P-troop in 1992 and the O-troop in 1996. Here, skew is much higher when one subordinate sires all non-dominant offspring. Because the dominant male in P-troop sired only 1.14 young out of 7, the subordinate male is assigned all remaining young such that he dominated reproduction in this year, which in turn results in high skew. It should be
145
R. Hager
1.0
Skew
146
0.5
0.0 1
2
3
4 5 Number of males
6
7
Figure 6.1 Reproductive skew in the Hanuman langur population from Ramnagar, India. The graph follows Equation 6.2.
noted that the number of infants sired by the dominant male represents a statistical measure of the probability of paternity (Launhard et al. 2001). In order to estimate levels of skew in populations for which no paternity data are available, we need to find a significant correlation between predictor variables and skew in the Ramnagar population. Assuming an equal share for all subordinates, reproductive skew is given by a quadratic regression, with the number of males as the independent variable: Skew ¼ 1:36292 0:481220 Nm þ 0:0648801ðNm Þ2
ð6:2Þ
where Nm is the number of males in the group in a given year. Figure 6.1 shows that skew in Ramnagar across all years and groups follows a U-shaped curve of group size accounting for 46.8% of the variation (F2,14 ¼ 5.27, p ¼ 0.023, n ¼ 16). Skew is complete in all one-male groups and falls to a low when groups contain 3–5 males (including the dominant), increasing again thereafter. If we assume that reproduction in the group was divided only between the dominant and the highest-ranking subordinate, we find a similarly significant relationship (F2,14 ¼ 4.97, p ¼ 0.027, n ¼ 16). In other words, whether or not subordinate reproduction follows a rank order or is monopolized does not affect the overall relationship between the number of males in the group and skew. Estimating skew in Asian langur populations
While several life-history and ecological parameters are known for many populations of langurs, genetic data on paternity and relatedness of
Male langurs: future mating prospects and ecological factors group members remain scarce. Therefore I estimated levels of skew in these populations using Equation 6.2 (Table 6.3). This shows that across all populations the distribution of reproduction is predicted to be skewed, although within each population the level of skew may vary between groups and years, as suggested by the data from Ramnagar. The highest levels of skew are predicted for populations in Deotala, Jaipur, and Mahdav, while skew takes intermediate values in Orcha Forest and Rajaji. Figure 6.2 illustrates the differences in estimated skew among Asian langur populations. If we were able to investigate whether or not the same set of factors influences skew in populations with similar skew, we could then attempt to assess whether skew could be used to classify and categorize populations. Predictor variables of estimated skew levels
Having estimated levels of skew in langur populations, we can now explore which of the life-history variables listed in Table 6.3 may explain variation in skew. All main effects and meaningful two-way interactions were analyzed using general linear models and the results of this analysis are summarized in Table 6.4. The minimal model shows that three predictor variables have significant main effects: the number of potential mates per male, the level of average annual rainfall in the focal population, and population density. First, the measure of future male mating prospects has a significant positive effect on skew (GLM; F1,12 ¼ 48.26, p < 0.001). In addition, the amount of average annual rainfall also positively impacts on skew (F1,12 ¼ 27.62, p ¼ 0.001), as does population density (GLM; F1,12 ¼ 8.84, p ¼ 0.021). In other words, skew is predicted to increase with increasing future mating prospects, annual rainfall, and population density. These effects are additive in that they explain part of the variation in levels of skew, over and above the effects of the other variables. Skew is affected to roughly the same extent by annual rainfall as it is by population density and is given by the following model equation: Estimated skew ¼ 0:145 þ 0:04½future prospects þ 2 · 103 ½population density þ 1:75 · 104 ½annual rain
ð6:3Þ
The positive effect of rainfall on skew is displayed in Figure 6.3, which shows the residual variation explained by rainfall after controlling for number of mates and population density. No significant second-order interactions between the predictor variables can be reported. Further, I detected no significant effects of birth rate, home range size, the average group size in the population, or the birth index characteristic of a population detected.
147
148
149
Table 6.3 Predicted levels of skew in 21 Asian langur populations in order of latitude. Levels of skew were estimated using Equation 6.2. Adapted from Treves & Chapman (1996) and Srivastava & Dunbar (1996). Empty cells refer to unavailable data, and populations with no multi-male groups have been excluded. Days/ Population
Birth rate
density
(births per
Home
month with
Estimated
Potential
Precipitation
% one-male
(individuals/
female per
Range
Predation
50 mm
Birth
Location
mates/male
per year (mm)
groups
km2)
year)
size (km2)
index
rain
index
skew
Kanha
7.67
1600
87
46.0
0.43
0.75
0
8
2.5
0.879
Dharwar
4.71
1492
77
91.5
0.20
0.20
1
5
8
0.732
Mundanthurai
2.63
-
-
-
-
-
-
-
0.811
Karnataka 1
7.17
-
-
-
-
-
1
-
0.707
level of
Karnataka 2
5.07
-
-
-
-
-
1
-
0.816
Andhra
5.25
-
-
-
-
-
1
-
0.759
Pradesh Junbesi
3.27
2500
33
1.5
0.11
6.60
1
5
2.5
0.787
Orcha forest
1.97
2030
0
4.4
0.72
5.30
0
5
7.0
0.471
Jaipur
17.80
648
84
12.5
0.51
-
1
8
0.947
Ambagarh
33.50
-
-
66.0
-
-
0
-
0.879
Rajaji
3.97
1200
25
90.0
-
1.50
0
6
2.5
Abu forest
6.74
1800
88
31.6.0
0.65
0.40
1
8
9
Abu town
5.56
1800
88
71.8
0.65
0.40
1
8
0.477 0.683 0.707
Ranthambhore
8.48
900
67
14.6
0.32
-
0
8
4.0
0.537
Kanha meadow Deotalao
4.79 10.50
-
-
46.1 -
-
-
0 -
-
2.5
0.582 0.947
1.0
0.477
Gir forest
5.53
674
40
121.5
0.31
0.40
0
8
Mahdav
13.30
-
-
-
-
-
-
-
0.947
Bangladesh
3.80
-
-
-
-
-
1
-
0.660
Raipur
5.61
388
57
-
0.35
-
1
7
1.0
0.524
Simla
3.33
1800
42
24.6
0.51
2.80
-
5
1.5
0.476
R. Hager
1 0.9
Estimated level of reproductive skew
150
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21
Figure 6.2 Estimated levels of reproductive skew for 21 Asian langur populations. The populations are numbered 1–21 in the same order as in Table 6.3.
Moreover, neither dispersal costs, predation risk, the mean number of infants, nor the proportion of one-male groups and their number per bisexual group were found to exert a significant effect on levels of skew. It should be noted that the number of females as a separate predictor does not significantly influence skew either (GLM; F1,12 ¼ 0.01, p ¼ 0.952).
Discussion The objective of this study was to illustrate how variation in reproductive skew among male primates may be explored, and to assess the potential of skew theory in helping to understand why specific factors influence skew and ultimately primate social systems. Using the S3 index, I calculated reproductive skew for five groups of Hanuman langurs in Ramnagar across several years. The analysis revealed a U-shaped curve of skew when plotted against the number of male group members. Skew was complete in all one-male groups, fell to a minimum with 2–3 subordinate males in the group, and increased again thereafter. Finally, several predictor variables of estimated skew levels in Asian langur population were identified using multiple regression. Chances of gaining matings in the future, as reflected in number of mates per male and survival chances (approximated by annual rainfall), were found
Male langurs: future mating prospects and ecological factors Table 6.4 GLM model of factors influencing predicted levels of reproductive skew in 21 populations of Asian langurs, listed in order of p value. Full model
DF
F
p
Number of potential mates per male
1
48.26
0.001
Annual precipitation (mm)
1
27.62
0.001
Population density (n individuals per km2)
1
8.84
0.021
Birth rate
1
5.32
0.069
Dispersal costs
2
14.98
0.180
Home-range size (km2)
1
1.70
0.262
Mean group size
1
0.88
0.386
Birth index
1
0.28
0.622
Predation risk
1
0.25
0.666
Mean number of infants
1
0.08
0.793
Proportion of one-male groups
1
0.05
0.833
Number of non-group males per bisexual group
1
0.04
0.860
Minimal model
Coefficient
SE coefficient
Constant
0.145
0.122
Number of potential mates per male
0.040
0.009
Population density (n individuals per km2)
0.002
7.9 · 104
Annual precipitation (mm)
0.002
5.7 · 105
Percentage variance accounted for: 88.10
to exert a significant positive effect on skew. In addition, environmental constraints, as measured by population density, positively affected skew. Reproductive skew is predicted to increase additively with all three parameters. In contrast to the other two studies on skew in primates (Widdig et al. 2004, Bradley et al. 2005), which found no support for predictions of concession models, here I demonstrate that variation in male reproductive skew may be explained by considering ecological and population variables, and that their effects can be explained in the framework of transactional skew models. However, it should be noted that no genetic data were available on withingroup relatedness, so the effect of r on skew could not be ascertained. Reproductive skew in Ramnagar populations: a potential role for infanticide?
Reproductive skew among male Langurs in the Ramnagar population plotted as a function of the number of males in the group followed a U-shaped curve with a minimum at just over three males in the group (Figure 6.1). It is surprising to see that skew increases when more than 3–4 males are present in
151
R. Hager Model residuals (number mates and population density)
152
0.2
0.1
0.0
–0.1
–0.2 500
1500 Annual rain [mm]
2500
Figure 6.3 GLM residuals of the two predictors potential mates and population density as a function of annual rainfall.
the group, since one would have expected it to become more difficult for the dominant to control subordinate reproduction when more males are present. Several factors need to be considered when attempting to explain this pattern. First, it should be noted that the observed pattern is mainly due to skew being complete in one-male groups, although skew declines also from two to three males in the group and increases thereafter. Second, even if one-male groups were excluded, skew would be low with around 3–4 males and higher with more males in the group. This suggests that the analysis reflects a realistic pattern of the distribution of skew, so we can now explore adaptive explanations. The role of infanticide in the control of reproduction has been modeled by Hager & Johnstone (2004), who found that the ability to discriminate own young from those of other group members, coupled with the option of infanticide, can have strong effects on reproductive skew because the threat of infanticide is effective in preventing subordinate reproduction. Hanuman males are certainly capable of committing infanticide, with most attacks being reported in the context of group takeovers both in one-male groups (e.g. Vogel & Loch 1984) and in multi-male groups (Boggess 1984, Borries 1997). The Hager & Johnstone (2004) model shows in particular that in order for the threat of infanticide to affect subordinate reproduction, little or no actual infanticide needs to occur, because subordinates refrain from siring young that otherwise would be killed by the dominant. Applied to the pattern of skew found in the Ramnagar population, one could speculate that with only a few extra males (one or two), the dominant may be able to directly control subordinate reproduction, for example by interrupting mating attempts or controlling access to females in the group. Infanticide is not invoked in this situation
Male langurs: future mating prospects and ecological factors because of the potential costs of accidentally killing one’s own young relative to possible fitness gains. However, the more males there are in the group, the more difficult it will become to physically control subordinate reproduction, and the dominant may threaten to kill any young that are being sired by other males, even at the risk of incurring fitness costs by killing his own young. In other words, the dominant may employ alternative strategies of controlling reproduction depending on the number of males in the group which affect his trade-off of the costs and benefits of physical control versus the threat of infanticide. The transition between these two strategies may occur when there are 2–3 extra males in the group, at which point it becomes difficult to directly control subordinate reproduction, and too costly to attempt to do so with more males. Testing this hypothesis will remain difficult, however, because of the constraints on experimental manipulation of male numbers in primate groups that could potentially lead to elevated levels of infanticide. As an alternative to the infanticide hypothesis, skew may be low in groups with 2–3 subordinate males because all males (dominant and subordinate) may be able to monopolize a given female to a greater extent than with more males in the group. Within-group aggression may increase with the number of males in the group, and the costs of attempting to monopolize a female may become too high such that queuing, yielding less costly but also less immediate chances of breeding, may become the better alternative for subordinates. In turn, this would result in higher skew, as suggested by the Ramnagar data. Factors influencing reproductive skew in langur populations: future mating prospects
Estimated levels of skew in Asian langur populations were found to be affected by chances of obtaining future matings and population density. Concession theory predicts that skew should be higher when future prospects are good, because benefits from queuing increase. Under such conditions, subordinates are expected to accept a smaller share of current reproduction when remaining in the group (Kokko & Johnstone 1999). Results of the analysis presented here lend support to this hypothesis: skew among males was positively influenced by their future mating prospects, which has been approximated by chances of survival and potential mating partners per male, whereby a higher ratio reflects better chances. One caveat here is that as the number of females per male increases the variance in reproductive success in the limited sex is predicted to increase (Emlen & Oring 1977). This means that mating chances may increase asymmetrically between males, with higher-ranking males gaining most of the benefits. However, it is reasonable to assume that monopolization becomes generally more difficult as the number of females increases.
153
154
R. Hager In langurs, staying in the group and queuing for a breeding position may become an option for subordinate males due to costs associated with dispersal (e.g. Rajpurohit & Sommer 1991). While to date no study has explicitly investigated the possibility of queuing in langurs, several observations provide evidence for this hypothesis. In both multi-male groups and all-male bands, it has been reported that individuals may rise in the hierarchy to attain the breeding position, with the result that most rank changes occur during male immigration, but also among resident males (Rajpurohit et al. 1995, Borries 1997). Other studies in primates have shown that an individual’s rank position is a good predictor of mating and reproductive success (Cowlishaw & Dunbar 1991, Ellis 1995). Altmann (1962) suggested that a ranking system may indeed function as a queue to attain dominance rank, for example in savannah baboons (Alberts et al. 2003) or Japanese macaques (Macaca fuscata: Sprague et al. 1998). Furthermore, in langurs it has been observed that during a group takeover subordinate males provided coalitional support, and Sommer (1988) suggested that such males gain delayed benefits, for instance through increased mating chances – a situation equivalent to that described in transactional skew models. Thus, queuing as a result of weighing costs and benefits of dispersal may be a viable strategy for subordinate males to increase their reproductive success. It should be noted that queuing in general may not apply to female langurs, due to their inverse dominance system whereby younger females are at the top of the hierarchy and then decline thereafter (Borries et al. 1991). Implicit in the argument that queuing for future matings can be decisive in affecting skew is the assumption that both individual survival and number of potential mates at a given point in time indeed reflect the conditions projected into the future. If the number of females per male is highly variable over the years (e.g. in populations with a high emigration/immigration rate), or if rainfall, and thus the chance of survival, is unpredictable, such measures of future prospects are more likely to estimate short-term mating prospects. Therefore, it seems important to ask how reliable the measures of future mating prospects for a given population are. In addition, it may be helpful to investigate in a separate study whether any of the population or ecological parameters influence the dominant’s ability to control subordinate reproduction. For instance, the structure of the habitat in which the group lives could make it more or less easy to interrupt subordinate mating attempts. Ecological constraints
Ecological constraints determine whether the trade-off of costs and benefits associated with dispersal and attempting to breed independently is
Male langurs: future mating prospects and ecological factors positive or negative. In concession models, for example, this will affect the share of reproduction the dominant needs to offer in order to retain the subordinate. Analyzing the factors influencing variation in skew provides support for the assumptions made in skew models about the significance of ecological constraints. First, I found that the amount of annual rainfall positively affects skew. Higher levels of rainfall have previously been assumed to reflect good environmental conditions (e.g. Emlen 1982), which, everything else being equal, may mean better chances of survival. Assuming that queuing for mating chances is an option for subordinate males, this would result in higher skew. Second, population density was found to affect skew such that skew is predicted to increase with population density. Following Sterck (1998), higher population densities occur in saturated habitats, in which we may expect higher competition for limiting resources (Tanner 1966, Morris 1992). Such conditions place constraints on dispersal and successful joining of other groups (Jones et al. 1988), which, in turn, make staying in a given group more attractive, so that skew should be higher. In addition, when resources are very limited in a habitat, retaining subordinates in the group may effectively reduce group productivity due to increased within-group competition over resources. In the concession framework, the staying incentive may then be reduced by the dominant, resulting in higher skew. It is important to note again that the analysis was based on estimated levels of skew derived from the significant relationship between number of males and skew in the Ramnagar population, which may deviate from that in other populations. Costs of dispersal
Many studies in primates have demonstrated that dispersing incurs fitness costs in terms of higher mortality, predation risk, and fighting with other males (Rajpurohit & Sommer 1991, van Schaik & Ho¨rstermann 1994). Concession models of skew predict, everything else being equal, high costs of dispersal to result in high skew (Johnstone 2000). Results of my analysis, however, did not detect any effect of dispersal costs on estimated skew in Asian langurs. Several factors could account for this finding. The most parsimonious of these would be that my proxy measure of dispersal costs (taking into account the additive effects of predation risk and population density) does not adequately reflect costs of dispersal in the langur system, implying the possibility that skew may still be affected by this variable had it been measured in a different way. Given the paucity of existing data and the non-existence of a generally agreed parameter to quantify dispersal costs, we are left in such an analysis with making assumptions about which recorded parameters are most likely to represent costs of dispersal (see Methods). For example, it could be
155
156
R. Hager that dispersal costs for a given individual are not at their highest when both predation risk and population density are high, but rather that the chance of individual predation is a negative function of population density because the attack rate per individual may be lower. At present, it is difficult to ascertain whether dispersal costs really do not impact on skew, or whether any effect could not be detected with the measures used. Assuming that dispersal costs can, indeed, be approximated as conducted in this analysis, we can ask why costs of dispersal did not affect levels of skew as posited by skew models. It may be that in langurs predation risk plays only a minor role in determining costs of dispersal (as opposed to fighting with other males, unknown habitat, etc.). Indeed, only one predator attack has been reported in the literature (Borries 1997). It should be noted that animals are usually not observed for the whole duration of their lifetime and may simply disappear for unknown reasons. This illustrates, again, the immense difficulty in quantifying the true effects of predation mortality and hence costs of dispersal. Future directions
Future studies of reproductive skew and the factors that influence variation in this measure will benefit from a number of improvements both in data collection and in conceptual aspects, which I will only briefly discuss here. Foremost is the collection of data that allow calculation of skew across years and groups, i.e. paternity data with complete records of individual tenure and, ideally, values for within-group relatedness. Second, primatologists should endeavor to collect and quantify the key variables identified by skew theory (see above), the more problematic being dispersal costs, future mating prospects, and group productivity. Third, as yet, there is no consensus regarding the period over which one should measure skew (skew per group year or over several years; see Bradley et al. 2005, Widdig et al. 2004), nor does it appear to be conceptually clear why one should prefer one option over the other. In primates, a starting point would be to consider over what time period individuals in a given species are likely to trade off, for example, staying versus leaving (from the subordinate’s perspective), or what the potential gains of subordinate presence are compared to the loss of direct reproduction if a share is conceded to subordinates (from the dominant’s perspective). The length of the breeding season and average tenure are variables that could be considered here (Figure 6.4). Of course, it would be highly interesting to compare how skew varies between years against the level of skew over longer periods of time (see Altmann et al. 1996). Can we identify groups by a reasonably constant level of skew, and can we correlate this to specific population and ecological parameters that are characteristic of the focal population?
Male langurs: future mating prospects and ecological factors
Figure 6.4 Among the Hanuman langurs at Mount Abu, Rajasthan, India. The average tenure length for a resident male was 27 months. This picture shows a former alpha male (grinding his teeth in the foreground) together with young males, all ousted when a rival male took over their troop. Two mothers with infants, attempting to avoid attacks on their infants by the new male, are also temporarily traveling with them (details can be found in Hrdy’s 1977 book The Langurs of Abu). Courtesy of Sarah B. Hrdy/Anthro-Photo.
Within-group aggression
One key feature of group-living animals is the level of within-group aggression, which is thought to be linked to reproductive skew in a group (Reeve & Nonacs 1997, Reeve 2000, Reeve & Keller 2001, Hager 2003b). A particular role in determining levels of aggression between members of one sex is assigned to dispersal costs. For example, in a concession model by Cant & Johnstone (2000), high costs of dispersing result in an increase of aggression because the subordinate (who is constrained in dispersing) stands to gain a potentially large share of reproduction by being highly aggressive, the extent of which depends on how aggression affects group productivity. On the other hand, in a tug-of-war situation, skew will be determined largely by the subordinate’s fighting ability, and will be high when the latter is poor because the
157
158
R. Hager subordinate can claim only a small fraction of reproduction (Johnstone 2000). In the context of queuing, it is reasonable to assume that a consequence of better future mating prospects could be to reduce costly activities such as fighting to gain matings within a group. Therefore, one testable prediction is that within-group aggression should be lower in groups in which subordinates have good future prospects than in groups with poor prospects, all other things being equal. The effects of female choice on male reproductive skew
All models of skew analyze how reproduction is shared among samesex individuals in a group and how population, ecological, and individual factors influence the dominant’s and subordinate’s trade-offs. However, in many systems, including primates, females are likely to exert an influence on male reproductive skew, for example by sneaky matings with subordinates, by extra-group copulations, or by totally disregarding any male except the alpha. Among primates are numerous examples of female choice in some form (e.g. Stumpf & Boesch 2005). Without attempting any formal description, it seems reasonable that skew among males could be negatively affected by females choosing to occasionally mate with a male different from the dominant, perhaps to increase the genetic diversity of their young or as a security against infanticide should the subordinate become the dominant at a later stage (Hausfater & Hrdy 1984). While it seems clear from many empirical studies that female choice impacts male reproductive skew, to date this has not been given due attention in skew theory (but see Cant & Reeve 2002). For example, if subordinates can expect to obtain a certain level of matings through female choice, they may accept a lower “direct” share conceded by the dominant, and skew might be higher than without female choice.
Conclusion The study presented here shows that skew theory can help explain causes of variation in reproductive skew in langurs. The factors found to influence levels of estimated skew among males can be explained by a concession mechanism if queuing for mating opportunities is a realistic option in langurs. The question at issue, then, is: do we advance our understanding of primate sociality by using skew theory to explain the distribution of reproduction, compared to alternative explanations? A central tenet in primatology is the priority-of-access model (Altmann 1962), which asserts that access to estrous females is largely determined by a male’s position in the dominance hierarchy. A general prediction of this model is that male
Male langurs: future mating prospects and ecological factors reproductive success is a function of his tenure and the number of estrous females at a given point in time, which has been confirmed by several studies (Altmann et al. 1996, Pope 1990). We therefore need to ask whether we gain any additional insight into primate social systems by applying skew theory? The answer, for the present author, is “Yes.” While the priority-of-access model makes predictions about individual reproductive success, the model does not offer any further explanations as to why individuals show variation in reproductive success. Which factors influence a dominant’s standing in the hierarchy? What is the role of ecological factors in determining skew? Skew theory holds the possibility of linking ecological, genetic, and population characteristics in a conceptual approach (described as transactional, compromise, or synthetic approaches: Johnstone 2000, Reeve & Shen 2006). Clearly, the factors underlying variation in skew can be analyzed regardless of skew theory, but these models have formally demonstrated the way in which such factors can be expected to interact under conditions of natural selection, and they offer testable hypotheses. Moreover, a more comprehensive analysis of reproductive skew allows us then to compare and categorize groups of animals, and perhaps species, according to their level of skew. Future studies should compare populations that feature similar skew. It will also be important to explore whether or not these comparisons are affected by similar sets of parameters and have similar consequences, for instance, for group sizes.
Acknowledgments I would like to thank Carola Borries, Volker Sommer, and two anonymous reviewers for help with the literature, and Clara B. Jones for constructive comments on the manuscript. References Allaine´, D. (2000). Sociality, mating system and reproductive skew in marmots: evidence and hypotheses. Behavioural Processes, 51, 21–34. Alberts, S. C. & Altmann, J. (1995). Balancing costs and opportunities: dispersal in male baboons. American Naturalist, 145, 279–306. Alberts, S. C., Watts, H. E. & Altmann, J. (2003). Queuing and queue-jumping: longterm patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Animal Behaviour, 65, 821–840. Altmann, J., Alberts, S. C., Haines, S. A., et al. (1996). Behavior predicts genes structure in a wild primate group. Proceedings of the National Academy of Sciences of the USA, 93, 5797–5801.
159
160
R. Hager Altmann, S. A. (1962). A field study of the sociobiology of rhesus monkeys, Macaca mulatta. Annals of the New York Academy of Sciences, 102, 338–435. Boggess, J. (1984). Infant killing and male reproductive strategies in langurs (Presbytis entellus). In G. Hausfater & S. B. Hrdy, eds., Infanticide: Comparative and Evolutionary Perspectives. New York, NY: Aldine, pp. 283–310. Borries, C. (1997). Infanticide in seasonally breeding multimale groups of Hanuman langurs (Presbytis entellus) in Ramnagar (South Nepal). Behavioral Ecology and Sociobiology, 41, 139–150. Borries, C. (2000). Male dispersal and mating season influxes in Hanuman langurs living in multi-male groups. In P. M. Kappeler, ed., Primate Males. Cambridge: Cambridge University Press, pp. 146–158. Borries, C., Sommer, V. & Srivastava, A. (1991). Dominance, age, and reproductive success in free-ranging female hanuman langurs (Presbytis entellus). American Journal of Primatology, 12, 231–257. Borries, C., Launhardt, K., Epplen, C., Epplen, J. T. & Winkler, P. (1999). Males as infant protectors in Hanuman langurs (Presbytis entellus) living in multimale groups: defence pattern, paternity and sexual behaviour. Behavioral Ecology and Sociobiology, 46, 350–356. Bradley, B. J., Robbins, M. M., Williamson, E. A., et al. (2005). Mountain gorilla tug-of-war: silverbacks have limited control over reproduction in multi-male groups. Proceedings of the National Academy of Sciences of the USA, 102, 9418–9423. Buston, P. M. (2004). Does the presence of non-breeders enhance the fitness of breeders? An experimental analysis in the clown anemonefish Amphirion percula. Behavioral Ecology and Sociobiology, 57, 23–31. Cant, M. A. & Johnstone, R. A. (1999). Costly young and reproductive skew in animal societies. Behavioral Ecology, 10, 178–184. Cant, M. A. & Johnstone, R. A. (2000). Power struggles, dominance testing, and reproductive skew. American Naturalist, 155, 406–417. Cant, M. A. & Reeve, H. K. (2002). Female control of the distribution of paternity in cooperative breeders. American Naturalist, 160, 602–611. Clutton-Brock, T. H. (1998). Reproductive skew, concessions and limited control. Trends in Ecology and Evolution, 13, 288–292. Clutton-Brock, T. H. & Harvey, P. H. (1979). Comparison and adaptation. Proceedings of the Royal Society of London B, 205, 547–565. Clutton-Brock, T. H., Brotherton, P. N., Russell, A. F., et al. (2001). Cooperation, control, and concession in meerkat groups. Science, 291, 478–481. Cowlishaw, G. & Dunbar, R. I. M. (1991). Dominance rank and mating success in male primates. Animal Behaviour, 41, 1045–1056. Dietz, J. M. (2004). Kinship structure and reproductive skew in cooperatively breeding primates. In B. Chapais & C. M. Berman, eds., Kinship and Behavior in Primates. Oxford: Oxford University Press, pp. 223–241. Ellis, L. (1995). Dominance and reproductive success among nonhuman animals: a cross-species comparison. Ethology and Sociobiology, 16, 257–333.
Male langurs: future mating prospects and ecological factors Emlen, S. (1982). The evolution of helping. I. An ecological constraints model. American Naturalist, 119, 29–39. Emlen, S. (1994). Benefits, constraints and the evolution of family. Trends in Ecology and Evolution, 9, 282–285. Emlen, S. (1995). An evolutionary theory of the family. Proceedings of the National Academy of Sciences of the USA, 92, 8092–8099. Emlen, S. & Oring, L. (1977). Ecology, sexual selection, and the evolution of mating systems. Science, 197, 215–223. Faulkes, C. G. & Bennet, N. (2001). Family values: group dynamics and social control of reproduction in African mole-rats. Trends in Ecology and Evolution, 16, 184–190. Grafen, A. & Hails, R. (2002). Modern Statistics for the Life Sciences. Oxford: Oxford University Press. Greenwood, P. J. (1980). Mating systems, philopatry, and dispersal in birds and mammals. Animal Behaviour, 28, 1140–1162. Hager, R. (2003a). Models of reproductive skew applied to primates. In Jones, C. B. ed., Sexual Selection and Reproductive Competition in Primates: New Perspectives and Direction. Norman, OK: American Society of Primatologists, pp. 65–101. Hager, R. (2003b). The effects of dispersal costs on reproductive skew and withingroup aggression in primate groups. Primate Report, 67, 941–949. Hager, R. (2004). Behavioural ecology. Encyclopedia of Life Sciences. Wiley: Chichester. Hager, R. & Johnstone, R. A. (2004). Infanticide and control of reproduction in cooperative and communal breeders. Animal Behaviour, 67, 941–949. Hamilton, I. M. (2000). Recruiters and joiners: using optimal skew theory to predict group size and the division of resources within groups of social foragers. American Naturalist, 155, 684–695. Hausfater, G. & Hrdy, S. B. (1984). Comparative and evolutionary perspectives on infanticide: introduction and overview. In G. Hausfater & S. B. Hrdy, eds., Infanticide: Comparative and Evolutionary Perspectives. New York, NY: Aldine, pp. xiii-xxxv. Hrdy, S. B. (1977). The Langurs of Abu. Cambridge, MA: Harvard University Press. Isbell, L. A. (1994). Predation on primates: ecological patterns and evolutionary consequences. Evolutionary Anthropology, 3, 61–71. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 105, 5–26. Johnstone, R. A. & Cant, M. A. (1999a). Reproductive skew and indiscriminate infanticide. Animal Behaviour, 57, 243–249. Johnstone, R. A. & Cant, M. A. (1999b). Reproductive skew and the threat of eviction: a new perspective. Proceedings of the Royal Society of London B, 266, 275–279. Jones, W. T., Waser, P. M., Elliott, L. F., Link, N. E. & Bush, B. B. (1988). Philopatry, dispersal, and habitat saturation in the banner-tailed kangaroo rat, Dipodomys spectabilis. Ecology, 69, 1466–1473. Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 9, 98–102. Koenig, A. & Borries, C. (2001). Socioecology of Hanuman langurs: the story of their success. Evolutionary Anthropolgy, 10, 122–137.
161
162
R. Hager Kokko, H. & Johnstone, R. A. (1999). Social queuing in animal societies: a dynamic model of reproductive skew. Proceedings of the Royal Society of London B, 266, 571–578. Kokko, H., Mackenzie, A., Reynolds, J. D., Lindstro¨m, J. & Sutherland, W. J. (1999). Measures of inequality are not equal. American Naturalist, 72, 358–382. Krebs, J. R. & Davies, N. B. (1993). An Introduction to Behavioural Ecology. 3rd edn. Oxford: Blackwell Science. Launhardt, K., Borries, C., Hardt, C., Epplen, J. T. & Winkler, P. (2001). Paternity analysis of alternative male reproductive routes among the langurs (Semnopithecus entellus) of Ramnagar. Animal Behaviour, 61, 53–64. Magrath, R. D., Heinsohn, R. G., & Johnstone, R. A. (2004). Reproductive skew. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge: Cambridge University Press, pp. 157–176. Moore, J. & Ali, R. (1984). Are dispersal and inbreeding avoidance related? Animal Behaviour, 32, 94–112. Morris, D. W. (1992). Scales and costs of habitat selection in heterogeneous landscapes. Evolutionary Ecology, 6, 412–432. Morris, D. W. (2002). Measuring the allele effect: positive density dependence in small mammals. Ecology, 83, 14–20. Newton, P. N. & Dunbar, R. I. M. (1994). Colobine monkey society. In A. G. Davies & J. F. Oates, eds., Colobine Monkeys: Their Ecology, Behaviour and Evolution. Cambridge: Cambridge University Press, pp. 311–346. Nonacs, P. (2000). Measuring and using skew in the study of social behavior and evolution. American Naturalist, 156, 577–589. Nonacs, P. (2003). Measuring the reliability of skew indices: is there one best index? Animal Behaviour, 65, 615–627. Ohsawa, H., Inoue, M., & Takenaka, O. (1993). Mating strategy and reproductive success of male patas monkeys. Primates, 34, 533–544. Packer, C., Collins, D. A., Sindimwo, A., & Goodall, J. (1995). Reproductive constraints on aggressive competition in female baboons. Nature, 373, 60–63. Pamilo, P. & Crozier, R. H. (1996). Reproductive skew simplified. Oikos, 75, 533–535. Pope, T. R. (1990). The reproductive consequences of male cooperation in the red howler monkey: paternity exclusion in multi-male and single-male troops using genetic markers. Behavioral Ecology and Sociobiology, 27, 439–446. Pusey, A. E. & Packer, C. (1987). Dispersal and philopatry. In B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham, & T. T. Struhsacker, eds., Primate Societies. Chicago, IL: Chicago University Press, pp. 250–281. Rajpurohit, L. S. & Sommer, V. (1991). Differences in mortality among langurs (Presbytis entellus) of Jodhpur, Rajasthan. Folia Primatologica, 56, 17–27. Rajpurohit, L. S., Sommer, V., & Mohnot, S. M. (1995). Wanderers between harems and bachelor bands: male Hanuman langurs (Presbytis entellus) at Jodhpur in Rajasthan. Behaviour, 132, 255–299. Reeve, H. K. (2000). A transactional theory of within-group conflict. American Naturalist, 155, 365–382.
Male langurs: future mating prospects and ecological factors Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model. Behavioral Ecology, 11, 640–647. Reeve, H. K. & Keller, L. (1995). Partitioning of reproduction in mother–daughter versus sibling associations: a test of optimal skew theory. American Naturalist, 145, 119–132. Reeve, H. K. & Keller, L. (1997). Reproductive bribing and policing as evolutionary mechanisms for the suppression of within-group selfishness. American Naturalist, 150, 42–58. Reeve, H. K. & Keller, L. (2001). Tests of reproductive-skew models in social insects. Annual Review of Entomology, 46, 347–385. Reeve, H. K. & Nonacs, P. (1997). Within-group aggression and the value of group members: theory and a field test with social wasps. Behavioral Ecology, 8, 75–82. Reeve, H. K. & Ratnieks, F. L. W. (1993). Queen–queen conflict in polygynous societies: mutual tolerance and reproductive skew. In L. Keller, ed., Queen Number and Sociality in Insects. Oxford: Oxford University Press, pp. 45–85. Reeve, H. K. & Shen, S. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Reeve, H. K., Emlen, S. & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Sherman, P. W., Lacey, E. A., Reeve, H. K., & Keller, L. (1995). The eusociality continuum. Behavioral Ecology, 6, 102–108. Sokal, R. R. & Rohlf, F. J. (1995). Biometry. 3rd edn. New York, NY: Freeman. Sommer, V. (1988). Male competition and coalitions in langurs (Presbytis entellus) at Jodhpur, Rajasthan, India. Human Evolution, 3, 261–278. Sommer, V. (1989). Sexual harassment in langur monkeys (Presbytis entellus): competition for ova, sperm, and nurture? Ethology, 80, 205–217. Sprague, D. S., Suzuki, S. S., Takahashi, H., & Sato, S. (1998). Male life history in natural populations of Japanese macaques: migration, dominance rank, and troop participation of males in two habitats. Primates, 39, 351–363. Srivastava, A. & Dunbar, R. I. M. (1996). The mating system of Hanuman langurs: a problem in optimal foraging. Behavioral Ecology and Sociobiology, 39, 219–226. Stacey, P. B. & Ligon, J. D. (1987). Territory quality and dispersal options in the acorn woodpecker, and a challenge to the habitat saturation model of cooperative breeding. American Naturalist, 130, 654–676. Sterck, E. H. M. (1998). Female dispersal, social organization, and infanticide in langurs: are they linked to human disturbance? American Journal of Primatology, 44, 235–254. Stumpf, R. M. & Boesch, C. (2005). Does promiscuous mating preclude female choice? Female sexual strategies in chimpanzees (Pan troglodytes verus) of the Tai National Park, Cote d’Ivoire. Behavioral Ecology and Sociobiology, 57, 511–524. Tanner, J. T. (1966). Effects of population density on growth rates of animal populations. Ecology, 47, 733–745.
163
164
R. Hager Treves, A. & Chapman, C. A. (1996). Conspecific threat, predation avoidance, and resource defense: implications for grouping in langurs. Behavioral Ecology and Sociobiology, 39, 43–53. Tsuji, K. & Kasuya, E. (2001). What do the indices of reproductive skew measure? American Naturalist, 158, 155–165. van Schaik, C. P. & Ho¨rstermann, M. (1994). Predation risk and the number of adult males in a primate group: a comparative test. Behavioral Ecology and Sociobiology, 35, 261–272. Vehrencamp, S. (1979). The roles of individual, kin, and group selection in the evolution of sociality. In P. Marler & J. Vandenbergh, eds., Handbook of Behavioral Neurobiology: Social Behavior and Communication. New York, NY: Plenum Press, pp. 351–394. Vehrencamp, S. (1983a). A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667–682. Vehrencamp, S. (1983b). Optimal degree of skew in cooperative societies. American Zoologist, 23, 327–335. Vehrencamp, S. (2000). Evolutionary routes to joint-female nesting in birds. Behavioral Ecology, 11, 334–344. Vogel, C. & Loch, H. (1984). Reproductive parameters, adult-male replacements, and infanticide among free-ranging langurs (Presbytis entellus) at Jodphur (Rajasthan), India. In G. Hausfater & S. B. Hrdy, eds., Infanticide: Comparative and Evolutionary Perspectives. New York, NY: Aldine, pp. 237–255. Widdig, A., Bercovitch, F., Streich, W. J., et al. (2004). A longitudinal analysis of reproductive skew in male rhesus macaques. Proceedings of the Royal Society of London, B, 271, 819–826. Winkler, P., Loch, H. & Vogel, C. (1984). Life history of Hanuman langurs (Presbytis entellus). Folia Primatologica, 43, 1–23.
7
The causes and consequences of reproductive skew in male primates n o b u y u k i k u t s u k a k e an d c h a r l e s l . n u n n
Summary This chapter discusses the underlying causes and consequences of reproductive skew in male primates. Although our understanding of the causes of skew is still in its infancy, empirical studies thus far support the compromise framework (e.g. tug-of-war model) rather than the concession model. Our assessment of the different models also suggests that the priority-of-access (POA) model makes predictions that are very similar to the compromise framework, but that skew models expand significantly on the POA model by including additional factors that relate to patterns of reproduction within groups. Our phylogenetic comparative analyses on mating skew in male primates also provide supporting evidence for the tug-of-war model, as mating skew decreased as the number of males increased, suggesting that monopolization of females becomes more difficult when there are more rivals (Emlen & Oring 1977). However, there have been no studies that represent strong tests of skew models, possibly because of difficulties in estimating parameters that are necessary for quantitative analyses. Future research could help to clarify the causes of skew, including development of mathematical models that are more suitable to primate societies, empirical studies based on paternity tests, and comparative approaches to investigate interspecific patterns of skew in other biological systems. Previous studies commonly investigated the causes of skew, but fewer have considered the consequences of skew on other physiological and social parameters such as within-group relatedness and sexually transmitted diseases. Of
Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
165
166
N. Kutsukake, C. L. Nunn these, it appears that effects on within-group relatedness could have the largest effects on patterns of primate sociality. The introduction of reproductive-skew models into primate research is likely to provide new insights into primate social and reproductive behavior in the future, while a primate perspective is likely to stimulate new theoretical work on reproductive skew.
Introduction Reproductive-skew theory attempts to explain the uneven distribution of reproductive success among same-sexed group members by multiple social, ecological, and genetic factors (reviewed in Johnstone 2000). Reproductiveskew theory has often been divided into two broad categories known as transactional and compromise frameworks. These frameworks differ according to the assumptions that each of them makes. In a version of the transactional framework known as the concession model, the dominant individual controls the reproduction of subordinates and allows them to reproduce in return for the subordinate staying in the group (i.e. the dominant offers a “staying incentive”: Vehrencamp 1983a, 1983b, Keller & Reeve 1994, Clutton-Brock 1998, Johnstone 2000). Retaining the subordinate is assumed to increase group productivity (i.e. total reproductive output of a group) and fitness benefits of a dominant, relative to the alternative of the subordinate leaving the group. In contrast, the tug-of-war model, which is part of the compromise framework, suggests that the dominant individual is unable to control the reproduction of subordinates completely (Reeve et al. 1998, Cant 1998, Clutton-Brock 1998); the division of reproduction is therefore determined by competition between a dominant and subordinate (Reeve et al. 1998, Cant 1998, Clutton-Brock 1998), which is assumed to decrease group productivity. These models can be expanded into systems with more than two individuals competing for reproduction (Johnstone et al. 1999, Reeve & Emlen 2000), including queuing systems (i.e. a subordinate acquiring a higher dominance position in the future: Kokko & Johnstone 1999, Ragsdale 1999, Mesterton-Gibbons et al. 2006). In this chapter, we consider the causes and the consequences of skew in male primates. Although recent research has synthesized the transactional and compromise frameworks into single conceptual models (Johnstone 2000, Reeve & Shen 2006), the classic dichotomy of the transactional (e.g. concession model) and compromise frameworks (e.g. tug-of-war model) provides a useful starting point for investigating reproductive skew in primates and will therefore be used here. Social primates live in relatively stable social groups. In these groups, males can exhibit considerable variation in the degree to which reproduction or matings are skewed (Cowlishaw & Dunbar 1991, Bulger 1993,
Causes and consequences of skew in male primates Kutsukake & Nunn 2006). Although inter-individual variation in male reproductive success has been a central topic in primate research (e.g. Cowlishaw & Dunbar 1991, Bulger 1993, Alberts et al. 2003, van Noordwijk & van Schaik 2004), only recently have researchers begun to investigate patterns of mating and reproduction in male primates using the theoretical framework of reproductive skew (Hager 2003, Widdig et al. 2004, Bradley et al. 2005, Kutsukake & Nunn 2006). Figure 7.1 provides an overview of the topics covered in this chapter. First we focus on the causes of skew, starting with an explanation of the priority-ofaccess model (POA) model and how this model corresponds to newer theoretical frameworks for understanding reproductive skew. In this first section, we also discuss key assumptions and predictions of the tug-of-war and concession models. Further, we review four case studies that have explicitly introduced and used paternity data to investigate predictions of skew models in primates, and we discuss a new research direction to examine predictions from skew theory using phylogenetic comparative methods (Kutsukake & Nunn 2006). In the second part of this chapter, we investigate the consequences of reproductive skew on other biological traits (Figure 7.1), focusing on two examples. The first involves the effects of skew on relatedness within groups, and the second considers how patterns of skew might influence the spread of sexually transmitted diseases.
The causes of reproductive skew The priority-of-access (POA) model
The POA model (Altmann 1962) has been the most influential framework used to explain variation in reproduction among male primates (Altmann 1962, Altmann et al. 1996, Boesch et al. 2006). The model predicts that the dominant male monopolizes reproduction within a group. However, the degree to which the dominant male succeeds in this goal is affected by the number of estrous females in the group. When two or more females are in estrus at the same time, the dominant male is unable to mate-guard all of them effectively, thus providing an opportunity for subordinate males to mate (Figure 7.2). The model therefore makes predictions for the distribution of matings within groups, with the dominant male obtaining the largest share and subordinates obtaining lesser amounts in proportion to their ranks. Empirical studies provide evidence for the dominant male’s advantages in both mating (Cowlishaw & Dunbar 1991, Bulger 1993, Ellis 1995, Alberts et al.
167
168
N. Kutsukake, C. L. Nunn
Causes Demographic Variables Male number Female number
Female influences
Ecological constraints
Estrous overlap Female choice Incest avoidance
Probability of solitary reproduction
Compromise framework
Genetic factors Relatedness among males Transactional framework
Skew among males
Within-group kin structure
Sexually transmitted disease
Consequences Figure 7.1 Scheme of the causes and consequences of reproductive skew discussed in this chapter. Black arrows indicate the effects predicted from the compromise framework (i.e. tug-of-war model), and the dashed arrows indicate the effects predicted from the transactional framework (i.e. concession model).
2003, Kutsukake & Nunn 2006) and paternity success (van Noordwijk & van Schaik 2004). In addition, some studies have investigated the effect of estrous synchrony on the distribution of matings, reproductive success, and the number of males in a group (e.g. Bulger 1993, Paul 1997, Nunn 1999a, Soltis et al. 2001, Takahashi 2004, van Noordwijk & van Schaik 2004, Alberts et al. 2006, Boesch et al. 2006). In general, these studies have shown that the more females are in estrus, the more difficult it becomes for the dominant male to control access to females. This effect of estrous synchrony has also been demonstrated in other species (e.g. domestic cats, Felis catus: Say et al. 2001). The POA model has contributed greatly to primate research, but studies in primates have produced variable results (Dunbar 1988, Cowlishaw & Dunbar 1991, Bulger 1993, van Noordwijk & van Schaik 2004, Kutsukake & Nunn 2006). In some cases, researchers have uncovered the biological reasons for departures from the POA model. For example, mate choice by females may also affect the distribution of reproduction in ways that differ from predictions of the POA model (Dunbar 1988, Soltis 2004). Females may confuse paternity by mating promiscuously and concealing ovulation – both of which should
Causes and consequences of skew in male primates
Figure 7.2 Mating in wild Japanese macaques (Macaca fuscata) at Shiga Heights, Nagano, Japan. A male is mounting an estrous female. Photo by N. Kutsukake.
decrease skew – or females may increase skew by copulating with the dominant male during periods in which the probability of fertilization is high (van Schaik et al. 1999, 2000, Nunn 1999b, van Noordwijk & van Schaik 2004). Some researchers have incorporated the effect of the number of males in evaluating the POA model (e.g. Alberts et al. 2003, 2006, Boesch et al. 2006), based on the reasoning that it should be more difficult for a dominant male to monopolize females when there are more males in the group that are competing for females (Cowlishaw & Dunbar 1991). Our comparative work – discussed below – provides evidence for this effect in analyses that control for phylogeny, suggesting that male number is the primary factor affecting skew in social primates. We refer to this framework as the extended POA model, in order to separate it from the original POA model, in which the number of males was not explicitly included (Altmann 1962). The POA and skew models are not fundamentally different in their goals of explaining the distribution of reproduction within groups. Compared to the POA model, however, the reproductive-skew framework uses a greater number of genetic and environmental variables, allowing a more detailed investigation
169
170
N. Kutsukake, C. L. Nunn of variation in male mating success. For example, skew models take into account the possibilities for males to leave the group and either attempt to breed on their own or join another group where their fitness would be greater, and thus also the need for dominant males to provide staying incentives. The concession model from the transactional framework also makes explicit assumptions about the degree of control that dominant males have over reproduction, with the tug-of-war models explicitly challenging the assumption of the dominant’s complete control of subordinate reproduction. Skew models make use of data on relationships among males, with greater skew predicted under the concession model when males are more closely related. In addition to male reproductive success, these models can be applied to investigate female reproductive success. Testing the reproductive-skew frameworks
Evaluating whether a particular skew model applies to a species requires information on multiple genetic and ecological parameters. Quantification of these parameters is difficult in any species, including primates. Moreover, experimental studies on skew conducted in other systems (e.g. Hymenoptera) are difficult or unethical to attempt in primates, in part because most primates have long lifespans and many are highly threatened. In this chapter, we first investigate the assumptions of different skew models (Johnstone 2000, Magrath et al. 2004), and second test specific predictions in observational and comparative studies. Testing assumptions of different skew models
The first assumption of the transactional framework is that the presence of subordinates increases group productivity and the dominant’s fitness. Positive relationships between male number and group productivity (e.g. efficiency of defense against extra-group males) have been reported in male primates (Wrangham 1999, Treves 2001). In wild chimpanzees (Pan troglodytes), for example, inter-group aggression is mainly conducted by males (Wrangham 1999), and the number of offspring and probability of infant survival increases with the number of males (Boesch et al. 2006). This pattern could occur through the combined effects of attracting females to the group and better defense of the territory or offspring. In another population, at Mahale, Tanzania (Figure 7.3), researchers documented that a decrease in the number of males in a small group, possibly caused by inter-group killing by the larger neighboring group, resulted in the transfer of females to the larger group (Nishida et al. 1985). Thus, this assumption that subordinates provide fitness benefits and higher group productivity could be met in species where males defend a territory or a
Causes and consequences of skew in male primates
Figure 7.3 Four male chimpanzees at Mahale Mountains National Park, Tanzania, patrolling the border to a neighboring group. Photo by N. Kutsukake.
group of females, and in other situations in which dominant males benefit from membership in multi-male groups. A key assumption of the concession model within the transactional framework is that the dominant individual has complete control over reproduction by subordinates. Field studies provide weak support for this assumption. In most species of social primates, for example, the presence of a dominant individual does not suppress the reproductive states of subordinates, and complete control must be difficult to achieve if there are too many rivals in a group. A dominant male can often interrupt subordinate matings, but in many cases he is ineffective in completely preventing copulations by subordinate males (Soltis 2004). Various studies have further shown that the degree to which the alpha male succeeds in reproduction decreases as the number of rivals increases (van Noordwijk & van Schaik 2004). Finally, complete control should be especially difficult in species living in fission–fusion societies (Dunbar 1988), where subdivision of the group into foraging parties should make it more difficult for males to monitor mating attempts by other males.
171
172
N. Kutsukake, C. L. Nunn These species include chimpanzees, bonobos (Pan paniscus), and spider monkeys (Ateles spp.). An important assumption of the compromise framework is that group productivity decreases as a result of competition between the dominant and subordinate. Infanticide by males is widely observed in primates (van Schaik & Janson 2001) and reduces group productivity. Correlational studies showed that groups with multiple males are less productive than single-male groups (black-and-white colobus, Colobus guereza: Dunbar 1987; Hanuman langurs, Semnopithecus entellus: Srivastava & Dunbar 1996), although the behavioral mechanism for how the presence of multiple males affects male–male competition – and ultimately group productivity – is largely unknown. Moreover, some studies reported positive effects of subordinate males on group productivity (red howler monkeys, Alouatta seniculus: Crockett & Janson 2000; mountain gorillas, Gorilla beringei: Watts 2000). These results suggest that the links between the number of subordinate males in a group and competition among males or group productivity is not universal. Future studies should test this assumption more broadly across species, including species living in multimale groups. This brief summary suggests that males are unlikely to have complete control over reproduction (as assumed in the concession model), and that group productivity can either increase or decrease with the number of males as well as the intensity of competition among dominant and subordinate males (as predicted by transactional and compromise frameworks, respectively). Thus, a particular skew model could be appropriate for some species but not for others, and quantitative testing of the assumptions could help to disentangle which models should be investigated in different species. Other models (and extensions of these models, such as social queuing or models that incorporate multiple individuals) make additional assumptions (Kokko & Johnstone 1999, Johnstone et al. 1999, Reeve & Emlen 2000) that would be worth investigating as skew frameworks are applied to primate mating systems. Testing specific predictions of skew models
Reproductive skew models also make different predictions regarding the effects of demographic variables (number of males and females), female reproductive traits (estrous synchrony), and relatedness among males on patterns of reproductive skew (Table 7.1). The tug-of-war model and the extended POA model predict that skew decreases as the number of males in a group increases, based on the reasoning that it will be difficult for a dominant male to control or monitor reproductive attempts by other males when more rivals are
Causes and consequences of skew in male primates Table 7.1 Predicted relationships between reproductive skew and the number of males and females, estrous overlap, and relatedness among males from three models Models “Extended” Effects on
priority-of-
reproductive skew
Priority-of-access access
Tug-of-war
Concession
Number of males
No prediction
No prediction
No prediction
No prediction
No prediction No prediction
in group Number of females in group Estrous overlap
Relatedness among
No prediction
No prediction
No prediction
males
þ
or
þ, positive relationship between variable and degree of skew; , negative relationship with skew
present (Cowlishaw & Dunbar 1991, van Noordwijk & van Schaik 2004). Similarly, increases in the number of females in a group should decrease skew if this provides more mating opportunities for subordinate males (Altmann 1962, Cowlishaw & Dunbar 1991, Bulger 1993, van Noordwijk & van Schaik 2004). Another prediction from the tug-of-war model and the POA model involves female estrous overlap (Table 7.1). Increased estrous overlap, which results from a long mating season, a long estrus duration, or socially mediated synchrony, should make it more difficult for a dominant male to monopolize a receptive female, thus decreasing skew among males (Ridley 1986, Cowlishaw & Dunbar 1991, Paul 1997, Shuster & Wade 2003). Similar effects can arise if the costs of guarding are high, causing dominant males to guard females for only part of their cycles (Packer 1979, Bercovitch 1983, Alberts et al. 1996). Few mathematical or empirical models of reproductive skew among males have considered the influence of female reproduction and behavior; exceptions include the female control model developed by Cant and Reeve (2002), and studies of acorn woodpeckers (Melanerpes formicivorus: Haydock & Koenig 2002), brown jays (Cyanocorax morio: Williams 2004), and some studies of primates (e.g. Soltis et al. 2001, Charpentier et al. 2005a, Boesch et al. 2006). Many of these exceptions involve female effects on male monopolization, and therefore address assumptions of the tug-of-war model. Furthermore, the socioecological model focuses explicitly on how female reproductive strategies influence male
173
174
N. Kutsukake, C. L. Nunn behavior (Wrangham 1980, Sterck et al. 1997, Dunbar 1988, Nunn 1999a, van Schaik et al. 1999), an area where primatology has the potential to contribute to further development of skew models. Finally, the tug-of-war model predicts no relationship between male relatedness and skew (Table 7.1). This relationship could even be negative in circumstances in which males exert weaker control over close relatives (Reeve et al. 1998). In some circumstances, for example, dominants could increase their inclusive fitness by exerting fewer restrictions on mating by related subordinates, thus generating a negative association between relatedness and skew. The concession model predicts no association between demographic factors or estrous synchrony and skew (Table 7.1). Instead, this model predicts that relatedness will affect patterns of skew, with high relatedness associated with high skew, assuming related subordinate males can receive their “staying incentive” in the form of inclusive fitness benefits (Keller & Reeve 1994, Johnstone 2000). Case studies of the causes of reproductive skew
Data on patterns of reproductive success in primates are becoming increasingly available (van Noordwijk & van Schaik 2004), offering potential for investigating whether compromise or transactional frameworks are more appropriate for studying skew in male primates (Widdig et al. 2004, Setchell et al. 2005, Charpentier et al. 2005a, Bradley et al. 2005, Boesch et al. 2006). As reviewed below, the compromise framework appears to offer a better fit for primate males, and the extended POA model may be equally powerful in explaining patterns of reproductive skew among male primates. Even so, transactional frameworks may account for additional variation in male skew, particularly when males defend territories, as this is one way by which group productivity can increase with the number of males (see above). In what follows, we review four case studies of male skew in multi-male multi-female primate groups (Table 7.2). These examples are not meant to be exhaustive; rather we use selected examples to reveal how skew theory provides new insights to variation in male reproductive success in primates. Rhesus macaque (Macaca mulatta)
In rhesus macaques, males disperse from their natal groups, while females remain in the group in which they were born. Widdig et al. (2004) investigated reproductive skew in a population on Cayo Santiago, Puerto Rico, and found that the top sire fathered between 19% and 30% of the offspring per year over a 6-year period, while 69–79% of males sired no infants at all. In terms
Table 7.2 Summary of empirical studies investigating reproductive skew in male primates Model “Extended” Study site, Species
group
Rhesus
Cayo Santiago,
macaque
a
# male
# female
Other
Relatedness
Estrous
Priority of
priority of
Tug-of-
among males
overlap
access
access
war
Concession
factors
ns (but two
ns
Yes
No
Hetero-
Puerto Rico
most
(Macaca
(free-ranging
successful
mulatta)
provisioned)
males were
important
zygosity
related in two of five years) Mandrill
b, c
CIRMF mandrill
(Mandrillus
colony, Gabon
sphinx)
(semi-freeranging provisioned) Setchell
b
Yes
Yes
-—
—
Yes
Partially yes
+ in one
Not
—
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
et al. 2005
Mountain gorilla
c
Charpentier
+
et al. 2005a Karisoke,
ns
f
avoidance ns
Virunga,
(of four)
investigated,
(Gorilla
Volcanoes
group
but overlap is
beringei)
National Park,
unlikely
Rwanda (wild) Chimpanzee
d
Taı¨ National
(Pan
Park, Cote
troglodytes)
d’Ivoire
—
—
Yes
Yes
(wild) Comparative analyses
Incest
—
ns
ns
ns
Partially yes
f
e
(31 species) þ, positive relationship; , negative relationship; ns, uncorrelated a Widdig et al. 2004 b controlled for the effect of estrous overlap by calculating the deviance of observed data from expected value from the priority-of-access model c Bradley et al. 2005 d Boesch et al. 2006 e Kutsukake and Nunn 2006 f “partial” because the effect of estrous overlap was not confirmed
Causes and consequences of skew in male primates of specific tests, the authors showed that: (1) males exhibited significant variation in skew, with a measure of skew (the B index: Nonacs 2000) significantly different from zero in most tests; (2) the B index was not significantly associated with either average pairwise relatedness among males or female synchrony (estimated indirectly from births); (3) heterozygosity of MHC genes predicted male reproductive success, highlighting the potential role of female choice. The authors concluded that their results support the compromise framework, as the concession model would predict a lower level of relatedness among breeders and stronger control of group reproduction by resident (dominant) males. Mandrill (Mandrillus sphinx)
In the wild, mandrills live in groups of up to several hundred individuals (Abernethy et al. 2002). Behavior in these groups has not been investigated, largely due to the difficulties of habituating and observing behavior of mandrills in their natural habitat. Important information on this species has been provided by research from a semi-free-ranging captive colony (CIRMF Mandrill Colony, Gabon). In this population, only the alpha male exhibits the distinctive secondary sexual traits (e.g. bright facial color) characteristic of this extremely sexually dimorphic species. Although multiple males are present in the colony, paternity analyses have shown that the alpha male fathers 69% of offspring, indicating extreme reproductive skew among males (Setchell et al. 2005). Two studies have investigated different aspects of reproductive skew in this colony. Although the authors studied the same groups, some results differed between the studies, in part due to differences in the specific aims of each study, in samples collected and variables analyzed, and in statistical approaches. In one of these studies, Setchell et al. (2005) investigated deviations in the alpha male’s reproductive success from the expected value based on the POA model. The authors showed that departures from the POA model increased as the number of males in a group increased (Table 7.2), which fits predictions from the extended POA and the tug-of-war models. By comparison, Charpentier et al. (2005a) studied factors affecting the failure of alpha males to sire offspring. These authors reported that relatedness among males, female estrous synchrony, relatedness between the dominant male and females, and the number of males affected paternity of the dominant male. Specifically, (1) the dominant male’s reproduction decreased as relatedness among males increased; (2) estrous overlap decreased reproduction by the dominant male; (3) relatedness between the dominant male and females negatively affected reproduction by the alpha male. Although the behavioral mechanism for this
177
178
N. Kutsukake, C. L. Nunn result is unknown, incest avoidance may have played a role, because the degree of heterozygosity correlated positively with individual reproductive success (Charpentier et al. 2005b). (4) In contrast to predictions of the tug-of-war model, the number of males correlated positively with the proportion of offspring sired by the alpha male (Charpentier et al. 2005a). To explain this result, Charpentier et al. (2005a) suggested that competition among subordinates increased with the number of males, deflecting competition away from the dominant male. Although the effect of male number differed between the studies, both Setchell et al. (2005) and Charpentier et al. (2005a) concluded that the limited control model best characterized this species; predictions of the concession model were never supported. Setchell et al. (2005) also noted that conditions in wild mandrills might produce weaker patterns of control than those found in the colony studied by these authors. Mountain gorilla (Gorilla beringei)
There is variation in the number of males in groups of mountain gorillas in the Virunga Mountains, Rwanda, with multi-male groups representing 40% of the groups in the population (Robbins et al. 2001). Female reproductive cycles are short, and it is rare that the receptive periods of two or more females overlap. The lack of overlap should tend to enable the dominant male to monopolize reproduction within a group. However, paternity analyses have shown that subordinates also reproduce to some extent (about 15%), suggesting that reproductive skew (estimated by the B index) is high but not complete (Bradley et al. 2005). In another study, Robbins & Robbins (2005) used an individual-based simulation model with demographic parameters from the same population studied by Bradley et al. (2005) to investigate the expected reproductive success of subordinates that remain in their group. The model revealed that remaining in a group benefits a subordinate more than dispersing. However, the model revealed that the dominant does not benefit from retention of subordinates, suggesting that dominant males do not concede reproduction. Thus, both Bradley et al. (2005) and Robbins & Robbins (2005) concluded that reproductive skew in this population corresponds better to predictions of the tug-of-war model than those of concession models. Chimpanzee (Pan troglodytes)
In chimpanzees, males remain in their natal group and exhibit a high degree of fission–fusion sociality. Females develop sexual swellings when they are in estrus, with synchronous estrous relatively common. The dominant
Causes and consequences of skew in male primates male has higher reproductive success, but subordinate males also reproduce (Constable et al. 2001). Boesch et al. (2006) investigated paternity in chimpanzees from Taı¨ National Park, Cote d’Ivoire, using long-term records. They found that the alpha male’s proportion of reproduction decreased as the number of males increased and when female estrous overlap increased. These results therefore agree with predictions from both the extended POA model and the tug-of-war model. Summary of case studies
Overall, these studies suggest that limited control is a characteristic of male behavior in primates (Table 7.2), and that the tug-of-war model or the extended POA model can explain variation in skew among male primates. However, these studies do not completely reject the concession model, for the following reasons. First, empirical studies mainly tested predictions from mathematical models that were designed for systems other than primates, often assuming that the group contains only two individuals – a dominant and a subordinate. In contrast, mathematical models that incorporate more realistic parameters, such as three or more group members or the possibility of social queuing by subordinates, predict a reduced necessity of offering incentives by a dominant individual to a subordinate (in particular to unrelated subordinates) relative to the two-player models (Kokko & Johnstone 1999, Ragsdale 1999, Johnstone et al. 1999, Reeve & Emlen 2000, Reeve & Shen 2006). This makes it difficult to draw firm predictions for how relatedness should correlate with patterns of skew. Second, no studies in primates have succeeded in accurately quantifying parameters that are necessary to test the skew model, in large part because it is difficult to conduct experimental manipulations in primates and to quantify model parameters. Finally, the concession and tug-of-war models are not mutually exclusive, and can in fact coexist within a single framework (Johnstone 2000, Reeve & Shen 2006). Phylogenetic comparative analyses
Skew models have been regarded as a unifying framework for understanding the diversity of social systems seen in animals (Keller & Reeve 1994, Sherman et al. 1995), but surprisingly few studies have examined broad evolutionary patterns of skew within one clade of either vertebrates or invertebrates (Faulkes et al. 1997, Boomsma & Sundstro¨m 1998, Duffy et al. 2000). Such comparative perspectives are important in reproductive skew research for at least four reasons. First, comparative studies provide a means to understand the factors generating broad evolutionary patterns of skew (Nonacs 2000) and therefore can assess the generality of a pattern across species, leading to
179
180
N. Kutsukake, C. L. Nunn
Figure 7.4 Phylogenetic comparative analyses on the relationship between male number in a group and the maximum mating proportion (the proportion of mating by the most successful male). For further details on these analyses, see Kutsukake & Nunn (2006).
greater unification of models of social evolution. Second, comparative approaches offer an opportunity to test assumptions and predictions of skew models from an evolutionary perspective. Third, by identifying differences among species, comparative results can point to new variables to investigate in future field or laboratory research. Finally, comparative research can be used to generate new hypotheses, which can then be tested in the field or laboratory, or refined through theoretical models. Prior cross-species comparisons in primates examined the effects of seasonality on variance in mating or reproductive success (Cowlishaw & Dunbar 1991, Paul 1997). We conducted a phylogenetic comparative analysis on the determinants of “mating” skew in male primates, based on a database of species in multi-male primate groups (in total from 84 studies, representing 31 species in 17 genera: Kutsukake & Nunn 2006). Since few studies have investigated the distribution of paternity for a sufficient number of primate species, we investigated mating distribution. While many studies have shown that mating frequency predicts reproductive success (e.g. Smith 1981, Pope 1990, Ohsawa et al. 1993, de Ruiter et al. 1994, Paul & Kuester 1996, Soltis et al. 1997, Alberts et al. 2006), other studies failed to find such links (e.g. Curie-Cohen et al.
Causes and consequences of skew in male primates 1983, Shively & Smith 1985, Inoue et al. 1991, 1993), possibly because many matings in primates are likely to be non-reproductive (Soltis 2004). To deal with this problem, we used data that are most tightly linked to male reproductive success whenever possible; specifically, we preferred data on ejaculation frequency more than copulation frequencies, and copulation data at times when conception was most likely to take place (Kutsukake & Nunn 2006). Genetic information on actual reproduction in groups would clarify these issues and allow skew to be examined more directly, but such data are not yet sufficiently available to test the predictions in a comparative context. In quantifying the magnitude of mating skew, we focus here on results using the “maximum mating proportion” (Bulger 1993), which is the proportion of matings by the most successful male. We also examined other skew indices, including the B index (Nonacs 2000) and lambda (Kokko & Lindstro¨m 1997). We investigated the effects of three variables: demographic factors (the number of males or females in a group), female reproductive factors associated with the difficulties of monopolizing estrous females (i.e. duration of the breeding season, duration of estrus, and measures of estrous overlap), and male dispersal pattern (categorized as male philopatry or male dispersal). Regarding male dispersal pattern, the concession model predicts high skew in male-philopatric species relative to species in which males disperse because there is (1) a high probability that a dominant male has a brother within a group and (2) a lower probability that subordinates will disperse. Taken together, these factors reduce the need for the dominant male to provide a staying incentive. The main results of our study (Kutsukake & Nunn 2006) can be summarized as follows. First, based on Nonac’s B index, mating was significantly skewed among males in 75.4% of cases (43/57 cases), and the alpha male or resident male tended to mate more frequently. Second, using the independent contrasts method (Felsenstein 1985) and stepwise multiple regression, we found that only male number correlated with mating skew (P < 0.001), with the proportion of mating by the most successful male falling as the number of males in a group increases (Figure 7.4). Finally, neither female reproductive proxies nor male dispersal pattern affected mating skew. Overall, these results are most consistent with the tug-of-war model and partially consistent with the extended POA model (in the sense that the number of males negatively affected skew). This result raises the possibility that the effects of estrous synchrony are not universal to all primate species, that its effects are weak, or that synchrony is difficult to quantify, all of which would limit our ability to detect a significant association in comparative analyses given existing data. Although the intensity
181
182
N. Kutsukake, C. L. Nunn of the correlation between dominance rank and reproductive success was affected by seasonality (Paul 1997), so far, few studies have investigated paternity among males in relation to estrous synchrony (Setchell et al. 2005, Charpentier et al. 2005a, Boesch et al. 2006). One could argue that the concession model also predicts that mating skew should decrease as the number of males increases, specifically if the dominant male needs to pay staying incentives to each subordinate male. However, we also found a similar negative relationship in an intraspecific analysis of wild chimpanzees (Kutsukake & Nunn 2006). The negative relationship is not expected in a male-philopatric species, such as the chimpanzee, because subordinate males have few opportunities for reproduction outside of their natal communities, and therefore do not need an incentive to stay. Even with this intraspecific analysis, however, we cannot firmly reject the concession model. For example, a negative relationship between the number of males and mating skew can also be explained by the concession model because reproductive skew may decrease when the power difference between a dominant and subordinate is small (e.g. in a group with many males: Cowlishaw & Dunbar 1991); therefore, the dominant may concede the reproduction as a “peaceful” incentive to avoid a risky fight with powerful rival males (Reeve & Ratnieks 1993). Indeed, the power differences may be smaller in a group with a large number of males because one would expect that males are, on average, more similar in age (and therefore competitive ability). This idea needs further testing, but highlights the difficulty of testing between different skew models, even in well-studied mammalian species. In addition, our comparative study does not reject the possibility that the concession model applies to a particular primate species, even if it is not a general explanation for patterns of skew across primates. As is shown by a recent synthetic model, the transactional framework and compromise framework are not mutually exclusive (Johnstone 2000, Reeve & Shen 2006). So one model may fit one species but not others, or in certain demographic or ecological situations but not in others within a species. For example, even within a species, the dominant male may be able to exert complete control in a small group in which there is only one subordinate, but not in a large group with multiple subordinates. This possibility can be tested by investigating how the effects of relatedness on reproductive skew vary according to the number of subordinate males in groups. Although we focused on short time intervals, such as a single breeding season, our approach can be used to examine complex life-history trajectories (patterns of lifetime reproductive success). In addition, it could be applied to both sexes. For example, reproductive success among female primates can be
Causes and consequences of skew in male primates estimated using long-term data. Finally, it would be interesting to apply this approach to other clades in which data on reproduction and phylogeny are widely available, such as birds and social insects, and in other well-studied mammalian groups, such as rodents, ungulates, and carnivores. Applying comparative approaches to other biological systems
Comparative tests can focus on either the predictions or the assumptions of skew models, and testing is possible if researchers have quantitative data for the distribution of reproduction or mating among group males. Here, in an attempt to stimulate further comparative research in other clades of animals, we list several methodological practices for conducting comparative tests of predictions related to reproductive skew models. (1)
(2)
(3)
(4)
Carefully choose the hypotheses, predictions, or assumptions to be tested. Within the framework of the models and the biology of the organisms, the researcher needs to consider alternative explanations and how different parameters might influence the predictions of a skew model. It is also important to incorporate the characteristics that are specific to the study animals, because some parameters are difficult to quantify in some clades. Collect data on mating or reproductive skew and other important variables such as group composition (e.g. number of males and females), relatedness, female behaviors, and reproductive biology. Data on reproduction are available in many non-primates (e.g. Ellis 1995), which could be used for comparative analyses. It is also important to obtain a phylogeny for the group of species being studied. “Supertrees” and other large-scale, dated phylogenies are now available for many species (Bininda-Emonds 2004), making this process easier than in the past. Quantify the distribution of reproduction using several skew proxies (Nonacs 2003). Many studies will not provide these measures directly, and may not even provide information for the comparative biologist to calculate the measures. Thus it might be necessary to use a simple index that maximizes sample size (in terms of the number of species). Test the hypotheses using phylogenetic comparative methods, such as independent contrasts (Felsenstein 1985, Harvey & Pagel 1991, Nunn & Barton 2001). It is important to check whether the data show phylogenetic signal (Blomberg & Garland 2002), to test the statistical and evolutionary assumptions, and to determine whether the results are robust to alternative assumptions.
183
184
N. Kutsukake, C. L. Nunn Consequences of reproductive skew Previous studies have mainly investigated the causes of skew, and have tested specific models. An important new direction in skew research is to consider the consequences of reproductive skew on other biological traits, including social structure and individual social strategies (Figure 7.1; Heinze 1995, Widdig et al. 2001, Cant & English 2006). For example, in some systems, the number of breeders and characteristics of the breeding queue could influence optimal group size (Cant & English 2006). With the goal of developing new questions for future studies, we briefly discuss two consequences of reproductive skew in male primates: (1) effects on within-group relatedness and (2) the spread of disease. Reproductive skew and within-group relatedness
In species characterized by high skew, infants born within a short period are more likely to be paternally related. For example, Widdig et al. (2004) found that in a high-skew rhesus macaque troop at Cayo Santiago, 74% of the infants had at least one paternal sibling in the group, and individuals had almost four times as many paternal as maternal siblings. In contrast, infants in low-skew societies are more likely to be fathered by different males, thus tending to reduce the level of relatedness at the group level. The paternal relatedness among group members should have a major impact on a wide range of social behaviors, including affiliation, cooperation, competition, and mate choice (Hamilton 1964, Chapais & Berman 2004). Several studies have suggested that individuals recognize paternal relationships and adjust their behavior accordingly. For example, skew is high in male western gorillas, and the silverbacks of different groups are closely related (Bradley et al. 2004). This result may explain the occurrence of nonagonistic encounters between groups observed in this species – unexpected in such a sexually dimorphic species, in which male–male competition is likely to be intense. Paternal half-siblings are more affiliative with one another than unrelated individuals in rhesus macaques (Widdig et al. 2001) and in savannah baboons (Papio cynocephalus: Smith et al. 2003, see also Silk et al. 2006). Furthermore, in baboons, paternal half-siblings showed less affiliative and sexual behavior during consortships than did unrelated pairs (Alberts 1999). As a final example, infants were supported by a biological father (Buchan et al. 2003) or were not the target of infanticide by the biological father in species living in multi-male groups (Borris et al. 1999a, 1999b, Soltis et al. 2000).
Causes and consequences of skew in male primates When reproductive skew is high and the dominant male’s tenure is long enough for his female offspring to become sexually mature, it may be adaptive for the dominant male to discriminate the paternity of the offspring and avoid mating with his daughters. In wild white-faced capuchin monkeys (Cebus capucinus), for example, the probability of reproduction by the alpha male varied with whether or not a female was a daughter of the alpha male, with a lower probability of reproduction between the alpha male and his daughter (Muniz et al. 2006). It would be interesting to investigate whether such incestavoidance mechanisms are more common in primates, especially since previous studies yielded inconclusive results (Table 7.2; Constable et al. 2001). If incest avoidance is an important selective force, strong skew combined with long male tenure could reduce future opportunities for the alpha male to reproduce within a group, thus creating an incentive for secondary dispersal. In addition, it will be important to uncover the proximate mechanisms responsible for identifying paternal kin (Rendall 2004). It is possible, for example, that dominant males make use of their information on the monopolization of receptive females as a proximate cue to assess the probability of their paternity. Similarly, for human observers, it may be possible to estimate the magnitude of reproductive skew a posteriori from the genetic relatedness among infants and juveniles in a group. Reproductive skew and the spread of infectious disease
Reproductive skew can also have consequences for patterns of social contact within social units, thus affecting the spread of disease within primate groups (Nunn & Altizer 2006). In a high-skew primate group under the tug-ofwar model, for example, one or a few males will gain access to receptive females. Thus there is likely to be intense competition among males as they fight to improve or maintain their dominance ranks. This fighting causes wounds through biting and scratching and may result in the spread of disease, as demonstrated in the case of retroviruses (SIV and STLV) in a semi-freeranging colony of mandrills (Nerrienet et al. 1998). In addition to being involved in male intrasexual competition, a high-ranking male in a high-skew society also has better access to mates, resulting in higher rates of sexual contact. Thus, such a male can act as a contact point for sexually transmitted diseases (STDs: Graves & Duvall 1995), potentially even selecting for reduced skew (Thrall et al. 2000, Kokko et al. 2002). If a female is already infected with an STD at the time a new male rises in rank, this male is likely to become infected shortly after he attains high rank; he can thus serve as the source of infection for the many females that he mates with during his tenure as the alpha male.
185
186
N. Kutsukake, C. L. Nunn A number of models have investigated the epidemiology of STDs in both human (Anderson & May 1991) and non-human systems (Thrall et al. 1997, Boots & Knell 2002, Kokko et al. 2002). In the context of variance in male mating skew, for example, Thrall et al. (2000) developed an individual-based model to explore how variance in mating success, patterns of female dispersal, and mortality rates of both sexes influence the spread of STDs. Given that the simulated population had an equal number of males and females, every male would have one female if there was no skew (equivalent to monogamy); each additional female assigned to a male means one less female for another male, resulting in increased reproductive skew. The simulations revealed that the prevalence of STDs is higher as the degree of polygyny (reproductive skew) increases. A challenge in applying these concepts to generate testable predictions is that low skew in multi-male multi-female primate groups can also favor the spread of an STD. Thus, if males have relatively equal access to females, this could result in a higher rate of mating with more males throughout the female’s cycle, possibly as a strategy to reduce the risk of infanticide (Hrdy & Whitten 1987, van Schaik et al. 1999). And of course, increased promiscuity should increase the spread of an STD (Anderson & May 1991). This promiscuity is likely to increase the prevalence to even higher levels than revealed by models of STD spread under skew (e.g. Thrall et al. 2000), especially if most subordinates have some mating success. The STD model of Thrall et al. (2000) provides a way out of this conundrum, however, because it also predicts a higher prevalence of infection in females than in males as reproductive skew increases, i.e. a sex difference is predicted. Kokko et al. (2002), in a different modeling approach, confirmed that female choice for a particular (presumably high-ranking) male can also lead to higher prevalence of infection in females. Thus, a critical prediction is that higher skew will produce not only high prevalence (relative to, say, monogamy); increasing skew should also produce a sex difference in the prevalence of an STD, with higher prevalence in females than in males. This prediction has been tested and supported using data on STDs in primates (Nunn & Altizer 2004). A next step is to examine whether sex differences also correlate with skew and other variables influencing the establishment of an STD, including mortality rates, dispersal, and differences in transmission probabilities between the sexes (e.g. with females potentially being more susceptible to an STD). In addition, it will be important to bring queuing or age-dependency into the STD models, because if most individual males have some mating opportunities over their lifetimes, the difference in STD exposure between the sexes may become more narrow.
Causes and consequences of skew in male primates Conclusion This chapter has discussed the causes and consequences of reproductive skew in male primates. Empirical studies have shown that the tug-ofwar model may better explain the pattern of skew among males than the concession model. Our comparative study revealed a negative association between the number of males in a group and skew, which agrees with previous findings in primates (Setchell et al. 2005, Boesch et al. 2006, reviewed in van Noordwijk & van Schaik 2004) and also agrees with predictions from the tug-ofwar model. Therefore, we tentatively conclude that incomplete control is a general characteristic of male primates, but more studies are needed to test the assumptions or predictions of the concession model. The priority-of-access (POA) model (Altmann 1962) highlighted the effect of estrous overlap on the distribution of reproduction among males in multi-male multi-female groups, including non-primates. A major conclusion of our chapter is that the POA model – especially an extended version that incorporates the number of males – is almost indistinguishable from the compromise framework. This is particularly true with regard to the model predictions (Table 7.1). It might therefore seem that the skew framework represents “new wine in an old bottle.” This would be misleading, however, as the skew framework is actually much broader than the previous POA model. For example, similar to models of reproductive skew, it builds significantly on POA by encapsulating factors involving relatedness, breeding opportunities, and costs of dispersal. Several challenges remain for the future. First, the present mathematical models of reproductive skew were not designed to investigate primate social systems. In particular, it would be worthwhile to develop skew models that incorporate three or more players (Johnstone et al. 1999, Reeve & Emlen 2000), social queuing (Kokko & Johnstone 1999, Ragsdale 1999, Mesterton-Gibbons et al. 2006), female influences such as incest avoidance (Cant & Reeve 2002, Johnstone 2000), and female choice for males with particular biological traits (“good genes” or high dominance rank). Recent mathematical models in which one individual adjusts behavior in response to the behavior of the other individual (negotiation game: McNamara et al. 1999, Cant & Shen 2006) may be more appropriate in primates, because social interactions in primates change temporally according to the strategy of opponents. Also, an individual-based model based on empirical demographic parameters would be a useful tool for generating more refined predictions for patterns of skew in primates (e.g. Robbins & Robbins 2005).
187
188
N. Kutsukake, C. L. Nunn Second, no empirical studies of primates have successfully estimated the parameters that are needed to distinguish among the different skew models. These parameters include the links between competition within groups and group productivity and ecological constraint that determine the probability of solitary reproduction. This may represent a limitation of skew theory, with very few predictions distinguishing the different models. Nonetheless, experimental studies, including manipulating group composition, would help to more formally test skew theory in primates. Such tests could be conducted in semi-free-ranging groups. Third, most of the studies in primates estimate skew in a relatively short time period. Thus, it is unknown how short-term skew is associated with longterm (i.e. lifespan) reproductive success (Altmann et al. 1996). The consequences of reproductive skew have been largely unexplored, yet these topics offer great opportunities for future research in primates. Independent of the causes of skew, how a given level of skew affects social structure, individual decision making, and other biological traits that relate to reproduction is a promising area for both empirical and theoretical research. For example, investigating the relationship between skew and the prevalence of STDs could have important implications for conservation biology, given that STDs often cause sterility (Canfield et al. 1991, Lockhart et al. 1996). In conclusion, bringing the skew paradigm to primatology may yield new perspectives for understanding primate behavior, specifically by integrating more diverse factors that are relevant to male and female decisions on group formation, interactions within groups, and reproductive strategies. Thus, skew models could play a major role in developing an integrative model of primate socioecology. Key future directions will involve developing skew models that are more appropriate for primates, collecting data to test the assumptions and predictions of these models, and investigating the consequences of reproductive skew for primate behavior. Moreover, a primate perspective on reproductive skew should help to ground models of skew more firmly, specifically in the context of multiple competitors and queuing within groups.
Acknowledgments We thank Reinmar Hager and Clara B. Jones for their invitation to contribute this chapter, and Mike Cant, Sarah Hodge, Kavita Isvaran, Jo Setchell, and two anonymous reviewers for helpful comments and discussion. This study was supported by JSPS Research Fellowships, RIKEN Special PostDoctoral Researchers Program, financed by JSPS core-to-core program HOPE (to NK) and the Max Planck Society (to CN).
Causes and consequences of skew in male primates References Abernethy, K. A., White, L. J. T., & Wickings, E. J. (2002). Hordes of mandrills (Mandrillus sphinx): extreme group size and seasonal male presence. Journal of Zoology, London, 258, 131–137. Alberts, S. C. (1999). Paternal kin discrimination in wild baboons. Proceedings of the Royal Society of London B, 266, 1501–1506. Alberts, S. C., Altmann, J., & Wilson, M. L. (1996). Mate guarding constrains foraging activity of male baboons. Animal Behaviour, 51, 1269–1277. Alberts, S. C., Watts, H. E., & Altmann, J. (2003). Queuing and queue-jumping: longterm patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Animal Behaviour, 65, 821–840. Alberts, S. C., Buchan, J. C., & Altmann, J. (2006). Sexual selection in wild baboons: from mating opportunities to paternity success. Animal Behaviour, 72, 1177– 1196. Altmann, S. A. (1962). A field study of the sociobiology of the rhesus monkey, Macaca mulatta. Annals of the New York Academy of Sciences, 102, 338–435. Altmann, J., Alberts, S. C., Haines, S. A., et al. (1996). Behavior predicts genetic structure in a wild primate group. Proceedings of the National Academy of Sciences of the USA, 93, 5797–5801. Anderson, R. M. & May, R. M. (1991). Infectious Diseases of Humans: Dynamics and Control. Oxford: Oxford University Press. Bercovitch, F. B. (1983). Time budgets and consortships in olive baboons (Papio anubis). Folia Primatologica, 41, 180–190. Bininda-Emonds, O. R. P., ed. (2004). Phylogenetic Supertrees: Combining Information to Reveal the Tree of Life. Dordrecht: Kluwer. Blomberg, S. P. & Garland, T., Jr. (2002). Tempo and mode in evolution: phylogenetic inertia, adaptation and comparative methods. Journal of Evolutionary Biology, 15, 899–910. Boesch, C., Kohou, G., Nene, H., & Vigilant, L. (2006). Male competition and paternity in wild chimpanzees of the Tai forest. American Journal of Physical Anthropology, 130, 103–115. Boomsma, J. J. & Sundstro¨m, L. (1998). Patterns of paternity skew in Formica ants. Behavioral Ecology and Sociobiology, 42, 85–92. Boots, M. & Knell, R. J. (2002). The evolution of risky behaviour in the presence of a sexually transmitted disease. Proceedings of the Royal Society of London B, 269, 585–589. Borries, C. K. L., Epplen, C., Epplen J. T., & Winkler, P. (1999a). Males as infant protectors in Hanuman langurs (Presbytis entellus) living in multimale groups: defense pattern, paternity and sexual behavior. Behavioral Ecology and Sociobiology, 46, 350–356. Borries, C., Launhardt, K., Epplen, C., Epplen, J. T., & Winkler, P. (1999b). DNA analyses support the hypothesis that infanticide is adaptive in langur monkeys. Proceedings of the Royal Society of London B, 266, 901–904.
189
190
N. Kutsukake, C. L. Nunn Bradley, B. J., Doran-Sheehy, D. M., Lukas, D., Boesch, C., & Vigilant, L. (2004). Dispersed male networks in western gorillas. Current Biology, 14, 510–513. Bradley, B. J., Robbins, M. M., Williamson, E. A., et al. (2005). Mountain gorilla tug-ofwar: silverbacks have limited control over reproduction in multimale groups. Proceedings of the National Academy of Sciences of the USA, 102, 9418–9423. Buchan, J. C., Alberts, S. C., Silk, J. B., & Altmann, J. (2003). True paternal care in a multi-male primate society. Nature, 425, 179–181. Bulger, J. B. (1993). Dominance rank and access to estrous females in male savanna baboons. Behaviour, 127, 67–103. Canfield, P. J., Love, D. N., Mearns, G., & Farram, E. (1991). Chlamydial infection in a colony of captive koalas. Australian Veterinary Journal, 68, 167–169. Cant, M. A. (1998). A model for the evolution of reproductive skew without reproductive suppression. Animal Behaviour, 55, 163–169. Cant, M. A. & English, S. (2006). Stable group size in cooperative breeders: the role of inheritance and reproductive skew. Behavioral Ecology, 17, 560–568. Cant, M. A. & Reeve, H. K. (2002). Female control of the distribution of paternity in cooperative breeders. American Naturalist, 160, 602–611. Cant, M. A. & Shen, S.-F. (2006). Endogenous timing in competitive interactions among relatives. Proceedings of the Royal Society of London B, 273, 171–178. Chapais, B. & Berman, C. M. (2004). Kinship and Behavior in Primates. Oxford: Oxford University Press. Charpentier, M., Peignot, P., Hossaert-McKey, M., et al. (2005a). Constraints on control: factors influencing reproductive success in male mandrills (Mandrillus sphinx). Behavioral Ecology, 16, 614–623. Charpentier, M., Setchell, J. M., Prugnolle, F., et al. (2005b). Genetic diversity and reproductive success in mandrills (Mandrillus sphinx). Proceedings of the National Academy of Sciences of the USA, 102, 16723–16728. Clutton-Brock, T. H. (1998). Reproductive skew, concessions and limited control. Trends in Ecology and Evolution, 13, 288–292. Constable, J., Ashley, M., Goodall, J., & Pusey, A. (2001). Noninvasive paternity assignment in Gombe chimpanzees. Molecular Ecology 10, 1279–1300. Cowlishaw, G. & Dunbar, R. I. M. (1991). Dominance rank and mating success in male primates. Animal Behaviour, 41, 1045–1056. Crockett, C. M. & Janson, C. H. (2000). Infanticide in red howlers: female group size, male membership, and a possible link to folivory. In C. P. van Schaik & C. H. Janson, eds., Infanticide by Males and its Implications. Cambridge: Cambridge University Press, pp. 75–98. Curie-Cohen, M., Yoshihara, D., Luttrell, L., et al. (1983). The effects of dominance on mating behavior and paternity in a captive troop of rhesus monkeys (Macaca mulatta). American Journal of Primatology, 5, 127–138. de Ruiter, J. R., van Hooff, J. A. R. A. M., & Scheffrahn, W. (1994). Social and genetic aspects of paternity in wild long-tailed macaques (Macaca fascicularis). Behaviour, 129, 203–224. Duffy, J. E., Morrison, C. L., & Rios, R. (2000). Multiple origins of eusociality among sponge-dwelling shrimps (Synalpheus). Evolution, 54, 503–516.
Causes and consequences of skew in male primates Dunbar, R. I. M. (1987). Habitat quality, population dynamics, and group composition in colobus monkeys (Colobus guereza). International Journal of Primatology, 8, 299–329. Dunbar, R. I. M. (1988). Primate Social Systems. Ithaca, NY: Cornell University Press. Ellis, L. (1995). Dominance and reproductive success among nonhuman animals: a cross-species comparison. Ethology and Sociobiology, 16, 257–333. Emlen, S. T. & Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science, 197, 215–223. Faulkes, C. G., Bennett, N. C., Bruford, M. W., et al. (1997). Ecological constraints drive social evolution in the African mole-rats. Proceedings of the Royal Society of London B, 264, 1619–1627. Felsenstein, J. (1985). Phylogenies and the comparative method. American Naturalist, 125, 1–15. Graves, B. M. & Duvall, D. (1995). Effects of sexually transmitted diseases on heritable variation in sexually selected systems. Animal Behaviour, 50, 1129–1131. Hager, R. (2003). Models of reproductive skew applied to primates. In C. B. Jones, ed., Sexual Selection and Reproductive Competition in Primates: New Perspectives and Directions. Norman, OK: American Society of Primatologists, pp. 65–101. Hamilton, W. D. (1964). The genetical evolution of social behaviour. Journal of Theoretical Biology, 7, 1–16, 17–52. Harvey, P. H. & Pagel, M. D. (1991). The Comparative Method in Evolutionary Biology. Oxford: Oxford University Press. Haydock, J. & Koenig, W. D. (2002). Reproductive skew in the polygynandrous acorn woodpecker. Proceedings of the National Academy of Sciences of the USA, 99, 7178–7183. Heinze, J. (1995). Reproductive skew and genetic relatedness in Leptothorax ants. Proceedings of the Royal Society of London B, 261, 375–379. Hrdy, S. B. & Whitten, P. L. (1987). Patterning of sexual activity. In B. B. Smuts, D. L. Cheney, R. W. Seyfarth, R. W. Wrangham, & T. T. Struhsaker, eds., Primate Societies. Chicago, IL: University of Chicago Press, pp. 370–384. Inoue, M., Mitsunaga, F., Ohsawa, H., et al. (1991). Male mating behaviour and paternity discrimination by DNA fingerprinting in a Japanese macaque group. Folia Primatologica, 56, 202–210. Inoue, M., Mitsunaga, F., Nozaki, M., et al. (1993). Male dominance rank and reproductive success in an enclosed group of Japanese macaques: with special reference to post-conception mating. Primates, 34, 503–511. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Johnstone, R. A., Woodroffe, R., Cant, M. A., & Wright, J. (1999). Reproductive skew in multimember groups. American Naturalist, 153, 315–331. Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 9, 98–102. Kokko, H. & Johnstone, R. A. (1999). Social queuing in animal societies: a dynamic model of reproductive skew. Proceedings of the Royal Society of London B, 265, 571–578.
191
192
N. Kutsukake, C. L. Nunn Kokko, H. & Lindstro¨m, J. (1997). Measuring mating skew. American Naturalist, 149, 794–799. Kokko, H., Ranta, E., Ruxton, G., & Lundberg, P. (2002). Sexually transmitted disease and the evolution of mating systems. Evolution, 56, 1091–1100. Kutsukake, N. & Nunn, C. L. (2006). Comparative tests of reproductive skew in male primates: the roles of demographic factors and incomplete control. Behavioral Ecology and Sociobiology, 60, 695–706. Lockhart, A. B., Thrall, P. H., & Antonovics, J. (1996). Sexually transmitted diseases in animals: ecological and evolutionary implications. Biological Reviews of the Cambridge Philosophical Society, 71, 415–471. Magrath, R. D., Heinsohn, R. G., & Johnstone, R. A. (2004). Reproductive skew. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge: Cambridge University Press, pp. 157–176. McNamara, J. M., Gasson, C. E., & Houston, A. I. (1999). Incorporating rules for responding into evolutionary games. Nature, 401, 368–371. Mesterton-Gibbons, M., Hardy, I. C. W., & Field, J. (2006). The effect of differential survivorship on the stability of reproductive queueing. Journal of Theoretical Biology, 242, 699–712. Muniz, L., Perry, S., Manson, J. M., et al. (2006). Father–daughter inbreeding avoidance in a wild primate population. Current Biology, 16, R156–157. Nerrienet, E., Amouretti, X., Muller-Trutwin, M. C., et al. (1998). Phylogenetic analysis of SIV and STLV type I in mandrills (Mandrillus sphinx): indications that intracolony transmissions are predominantly the result of male-to-male aggressive contacts. Aids Research and Human Retroviruses, 14, 785–796. Nishida, T., Hiraiwa-Hasegawa, M., Hasegawa, T., & Takahata, Y. (1985). Group extinction and female transfer in wild chimpanzees of the Mahale Mountains. Zeitschrift fu¨r Tierpsychologie, 67, 284–301. Nonacs, P. (2000). Measuring and using skew in the study of social behavior and evolution. American Naturalist, 156, 577–589. Nonacs, P. (2003). Measuring the reliability of skew indices: is there one best index? Animal Behaviour, 65, 615–627. Nunn, C. L. (1999a). The number of males in primate social groups: a comparative test of the socioecological model. Behavioral Ecology and Sociobiology, 46, 1–13. Nunn, C. L. (1999b). The evolution of exaggerated sexual swellings in primates and the graded signal hypothesis. Animal Behaviour, 58, 229–246. Nunn, C. L. & Altizer, S. M. (2004). Sexual selection, behavior and sexually transmitted diseases. In P. M. Kappeler & C. P. van Schaik, eds., Sexual Selection in Primates: a Comparative Perspective. Cambridge: Cambridge University Press, pp. 117–130. Nunn, C. L. & Altizer, S. M. (2006). Infectious Diseases in Primates: Behavior, Ecology and Evolution. Oxford: Oxford University Press. Nunn, C. L. & Barton, R. A. (2001). Comparative methods for studying primate adaptation and allometry. Evolutionary Anthropology, 10, 81–98.
Causes and consequences of skew in male primates Ohsawa, H., Inoue, M., & Takenaka, O. (1993). Mating strategy and reproductive success of male patas monkeys (Erythrocebus patas). Primates, 34, 533–544. Packer, C. (1979). Male dominance and reproductive activity in Papio anubis. Animal Behaviour, 37, 37–45. Paul, A. (1997). Breeding seasonality affects the association between dominance and reproductive success in non-human male primates. Folia Primatologica, 68, 344–349. Paul, A. & Kuester, J. (1996). Differential reproduction in male and female Barbary macaques. In J. E. Fa & D. G. Lindburg, eds., Evolution and Ecology of Macaque Societies. Cambridge: Cambridge University Press, pp. 293–317. Pope, T. R. (1990). The reproductive consequences of male cooperation in the red howler monkey: paternity exclusion in multi-male and single-male troops using genetic markers. Behavioral Ecology and Sociobiology, 27, 439–446. Ragsdale, J. E. (1999). Reproductive skew theory extended: the effect of resource inheritance on social organization. Evolutionary Ecological Research, 1, 859–874. Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model. Behavioral Ecology, 11, 640–647. Reeve, H. K. & Ratnieks, F. L.W. (1993). Queen–queen conflicts in polygynous societies: mutual tolerance and reproductive skew. In L. Keller, ed., Queen Number and Sociality in Insects. New York, NY: Oxford University Press, pp. 45–85. Reeve, H. K. & Shen, S. -F. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Rendall, D. (2004). “Recognizing” kin: mechanisms, media, minds, modules and muddles. In B. Chapais & C. Berman, eds., Kinship and Behavior in Primates. Oxford: Oxford University Press, pp. 295–316. Ridley, M. (1986). The number of males in a primate troop. Animal Behaviour, 34, 1848–1858. Robbins, A. M. & Robbins, M. M. (2005). Fitness consequences of dispersal decisions for male mountain gorillas (Gorilla beringei beringei). Behavioral Ecology and Sociobiology, 58, 295–309. Robbins, M. M., Sicotte, P., & Stewart, K. J. (2001). Mountain Gorillas: Three Decades of Research at Karisoke. Cambridge: Cambridge University Press. Say, L., Pontier, D., & Natoli, E. (2001). Influence of oestrus synchronization on male reproductive success in the domestic cat (Felis catus L.). Proceedings of the Royal Society of London B, 268, 1049–1053. Setchell, J. M., Charpentier, M., & Wickings, E. J. (2005). Mate-guarding and paternity in mandrills (Mandrillus sphinx): factors influencing monopolisation of females by the alpha male. Animal Behaviour, 70, 1105–1120. Sherman, P. W., Lacey, E. A., Reeve, H. K., & Keller, L. (1995). The eusociality continuum. Behavioral Ecology, 6, 102–108.
193
194
N. Kutsukake, C. L. Nunn Shively, C. & Smith, D. G. (1985). Social status and reproductive success of male Macaca fascicularis. American Journal of Primatology, 9, 129–135. Shuster, S. M. & Wade, M. J. (2003). Mating Systems and Mating Strategies. Princeton, NJ: Princeton University Press. Silk, J. B., Altmann, J., & Alberts, S. C. (2006). Social relationships among adult female baboons (Papio cynocephalus) I. Variation in the strength of social bonds. Behavioral Ecology and Sociobiology, 61, 183–195. Smith, D. G. (1981). The association between rank and reproductive success of male rhesus monkeys. American Journal of Primatology, 1, 83–90. Smith, K., Alberts, S. C., & Altmann, J. (2003). Wild female baboons bias their social behaviour towards paternal half-sisters. Proceedings of the Royal Society of London B, 270, 503–510. Soltis, J., Thomsen, R., & Takenaka, O. (2001). The interaction of male and female reproductive strategies and paternity in wild Japanese macaques, Macaca fuscata. Animal Behaviour, 62, 485–494. Soltis, J. (2004). Mating systems. In B. Thierry, M. Singh, & W. Kaumanns, eds., Macaque Societies: a Model for the Study of Social Organization. Cambridge: Cambridge University Press, pp. 135–151. Soltis, J., Mitsunaga, F., Shimizu, K., et al. (1997). Sexual selection in Japanese macaques, II: female mate choice and male–male competition. Animal Behaviour, 54, 737–746. Soltis, J., Thomsen, R., Matsubayashi, K., & Takenaka, O. (2000). Infanticide by resident males and female counter-strategies in wild Japanese macaques (Macaca fuscata). Behavioral Ecology and Sociobiology, 48, 195–202. Srivastava, A. & Dunbar, R. I. M. (1996). The mating system of Hanuman langurs: a problem in optimal foraging. Behavioral Ecology and Sociobiology, 39, 219–226. Sterck, E. H.M., Watts, D. P., & van Schaik, C. P. (1997). The evolution of female social relationships in nonhuman primates. Behavioral Ecology and Sociobiology, 4, 291–309. Takahashi, H. (2004). Do males have a better chance of mating when the number of estrous females is equal to or greater than the males’ ordinal rank? Testing the hypothesis in Japanese macaques. American Journal of Primatology, 63, 95–102. Thrall, P. H., Antonovics, J., & Bever, J. D. (1997). Sexual transmission of disease and host mating systems: within-season reproductive success. American Naturalist, 149, 485–506. Thrall, P. H., Antonovics, J., & Dobson, A. P. (2000). Sexually transmitted diseases in polygynous mating systems: prevalence and impact on reproductive success. Proceedings of the Royal Society of London B, 267, 1555–1563. Treves, A. (2001). Reproductive consequences of variation in the composition of howler monkey (Alouatta spp.) groups. Behavioral Ecology and Sociobiology, 50, 61–71. van Noordwijk, M. A. & van Schaik, C. P. (2004). Sexual selection and the careers of primate males: paternity concentration, dominance acquisition tactics and
Causes and consequences of skew in male primates transfer decisions. In P. M. Kappeler & C. P. van Schaik, eds., Sexual Selection in Primates: a Comparative Perspective. Cambridge: Cambridge University Press, pp. 208–229. van Schaik, C. P. & Janson, C. H. (2001). Infanticide by Males and its Implications. Cambridge: Cambridge University Press. van Schaik, C. P., van Noordwijk, M. A., & Nunn, C. L. (1999). Sex and social evolution in primates. In P. Lee, ed., Comparative Socioecology of Primates. Cambridge: Cambridge University Press, pp. 204–240. van Schaik, C. P., Hodges, K., & Nunn, C. L. (2000). Paternity confusion and the ovarian cycles of female primates. In C. P. van Schaik & C. H. Janson, eds., Infanticide by Males and Its Implications. Cambridge: Cambridge University Press, pp. 361–387. Vehrencamp, S. L. (1983a). A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667–682. Vehrencamp, S. L. (1983b). Optimal skew in cooperative societies. American Zoologist, 23, 327–355. Watts, D. P. (2000). Causes and consequences of variation in male mountain gorilla life: histories and group membership. In P. M. Kappeler, ed., Primate Males: Causes and Consequences of Variation in Group Composition. Cambridge: Cambridge University Press, pp. 169–180. Widdig, A., Nu¨rnberg, P., Krawczak, M., Streich, W. J., & Bercovitch, F. B. (2001). Paternal relatedness and age proximity regulate social relationships among adult female rhesus macaques. Proceedings of the National Academy of Sciences of the USA, 98, 13769–13773. Widdig, A., Bercovitch, F. B., Streich, W. J., Sauermann, U., Nuernberg, P., & Krawczak, M. (2004). A longitudinal analysis of reproductive skew in male macaques. Proceedings of the Royal Society of London B, 271, 819–826. Williams, D. A. (2004). Female control of reproductive skew in cooperatively breeding brown jays (Cyanocorax morio). Behavioral Ecology and Sociobiology, 55, 370–380. Wrangham, R. W. (1980). An ecological model of female-bonded primates. Behaviour, 75, 262–299. Wrangham, R. W. (1999). The evolution of coalitionary killing. Yearbook of Physical Anthropology, 42, 1–30.
195
8
Sociality and reproductive skew in horses and zebras d a n i e l i . r u b e n s t e i n a n d ca s sa nd r a m . n u n˜ e z
Summary The outcome of competition for resources or mates often leads to individual differences in reproductive success. In populations of equids, such as those of horses and zebras, skewed distributions of reproduction emerge because a limited number of individuals achieve disproportionate gains. For both sexes, skew results from differences in rank, age, and degree of social stability, although skew is generally greater for males than for females. Adult male horses and zebras typically establish “harem” groups by bonding with a number of mature females. Although the number of females that dominants bond with can be quite variable, potentially high levels of skew are rarely reached because subordinate males adopt alternative mating tactics that exact concessions from partners, whether they are dominant stallions or other subordinates. Successful breeding females also rely on support from subordinates to minimize feeding competition by keeping group size small, and this, too, reduces skew among females. The conflict of interest between the sexes arising over differences in optimal group size, along with the tendency for females to leave groups when sexually harassed, induces, but limits, the aggression that males direct towards females. Thus female concessions can shape both female and male levels of skew, but they also can be modulated by male behavior. Thus it appears that for equids the level of skew that emerges depends on key phenotypic traits and how their distribution among individuals constrains reproduction, as well as on how social relationships within and between the sexes affect the ability of a small number of individuals to monopolize resources involved in reproduction.
Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
196
Sociality and reproductive skew in horses and zebras Inequality is a pervasive feature of virtually all societies, as is the attempt by the disadvantaged to attenuate it. Humans (Homo sapiens) have gone so far as to construct political systems to reduce, if not eliminate, differences in payoffs or outcomes. The Marxist ideal “From each according to his abilities, to each according to his needs” (Marx 1875) represents an egalitarian extreme. While animals may not codify behavior into philosophical norms, actions within many animal societies often redistribute payoffs away from the successful few, thus moderating heavily skewed resource allocations and reproductive opportunities. In mammals alone, skewed patterns of reproductive success are seen in male macaques (Macaca mulatta: Widdig et al. 2004), rhinoceroses (Diceros bicornis: Garnier et al. 2001), and even sex-role-reversed species such as hyenas (Crocuta crocuta: Engh et al. 2002). And in gorillas (Gorilla beringei: Robbins et al. 2006), meerkats (Suricata suricatta: Clutton-Brock et al. 2001), marmots (Marmota spp.: Allaine´ 2000), and mountain zebras (Equus zebra: Lloyd & Rasa 1989), skewed patterns of reproduction occur among females as well. Reductions in inequality often emerge when the less favored change the rules of the game by adopting strategies that deviate from the typical strategy (Rubenstein 1980). Alternatively, when the haves and the have-nots need each other to prosper, the status quo is maintained by the actions of those receiving the least. Either those having gained the most provide incentives to the less welloff, or the well-off are unable to completely impose their will when confronted by the actions of others (Rubenstein 1978, Vehrencamp 1983, Reeve et al. 1998, Johnstone 2000, Reeve & Shen 2006). In order to understand why particular strategies and relationships evolve and produce skewed payoffs and outcomes, knowing the underlying causes of inequality is critical. In many animal societies inequality results from differences in rank or fighting ability. Dominant females often monopolize resources needed to produce and raise offspring, while dominant males often monopolize reproductive access to females (Rubenstein 1994). For males, status and the bodily condition necessary for achieving and maintaining dominance is driven by sexual selection. But less well understood are the factors affecting status in non-sex-role-reversed females, especially in species such as equids, where competition for essential ecological resources is minimal (Rubenstein 1994). Dominance itself is influenced by many other factors such as age, bodily condition, reproductive state, and environmental harshness, thus pushing an understanding of the causes of reproductive skew one step back. Factors other than dominance and its determinants are also likely to play a role in shaping patterns of reproductive skew. These might include kinship as well as the nature and strength of relationships with individuals of different ages and sex. In this chapter we will explore these issues in order to understand the causes
197
198
D. I. Rubenstein, C. M. Nun˜ez and consequences of unequal and skewed reproduction in populations of feral horses (Equus caballus) and plains zebras (E. burchelli). We will illustrate the degree to which reproduction is skewed in each sex and we will identify causes. We will then examine strategies that develop to attenuate these differences among both females and males. Lastly, we will explore management and conservation consequences of these attenuation strategies.
Patterns of sociality and methods Equid societies come in two forms (Klingel 1977, Moehlman 1979, Rubenstein 1986). In one, typified by feral and wild horses (Equus caballus) as well as plains (E. burchelli) and mountain zebras (E. zebra), females and their immature offspring associate for long periods with one male. This is the socalled “harem,” and as offspring of both sexes mature they disperse. Thus the breeding members of these family groups are typically non-relatives, and kinship is unlikely to play a strong role in shaping social or reproductive decisions. In the other, typified by Grevy’s zebras (E. greryi), Asiatic (E. hemionus) and African wild asses (E. africanus), and the kiang (E. kiang), females form loose and transitory associations with many other females and males. This is a fission–fusion society and, typically, females of different reproductive states segregate and use different habitats. Since both classes of females contain reproductively active individuals, males forgo bonding with any class, instead establishing territories around water. In this way these territories contain the home ranges of lactating females with young foals as well as the routes that non-lactating females use to come to water (Rubenstein 1986, 1994, Sundaresan et al. 2007). Variants on these themes do emerge, and they provide insights into the ecological forces that shape social relationships as well as reproductive strategies and payoffs. For example, some horse populations have harems that defend territories (Rubenstein 1980, 1981) and some have “harems” with two stallions rather than one (Rubenstein 1982, 1986, Feh 1999, Linklater & Cameron 2000a, 2000b). Moreover, in plains zebra societies harems often coalesce into herds. Thus “harem species” can form multi-male as well as unimale groups, providing polygynous opportunities for males and polyandrous opportunities for females. Harem males can also defend territorial space and, at least in plains zebras, males can bring their harems together to form herds. A second tier of sociality, with fission–fusion elements, emerges as herds form and dissolve. The social and reproductive patterns described in this chapter come from long-term studies (1983 to the present) of free-ranging horses on Shackleford
Sociality and reproductive skew in horses and zebras
Figure 8.1 A highly successful Shackleford female and her young on the Shackleford Banks, North Carolina, USA. Photo courtesy of Cheryl Burke, Carteret County News Times.
Banks, North Carolina, USA (Figure 8.1), and plains zebras on the Ol Pejeta Conservancy and unfenced commercial ranches around the Mpala Research Center, Kenya, from 2000 to 2004 (Figure 8.2). Standard behavioral observations on time budgets (Rubenstein 1986, Rubenstein & Hack 2004), associations (Rubenstein et al. 2007, Sundaresan et al. 2007), and movements (Rubenstein 1986, 1994) were made on individuals identifiable by unique color patterns and freeze brands (horses) or stripes (plains zebras). Patterns of reproductive success (means, variances, and skew) are represented by standard statistical moments. Paternity analyses were performed by the US National Park Service using allozymes and DNA (E. G. Cothran, personal communication).
Background features and framework Equids illustrate some special social, ecological, and demographic features. First, equids, as large-bodied, hindgut fermenters with wide tooth rows, can subsist on low-quality forage. As a result, competition for such abundant food should be low, especially in mesic habitats. Instead of contesting for individual food items, females move to nearby items of equivalent
199
200
D. I. Rubenstein, C. M. Nun˜ez
Figure 8.2 Two male plains zebras fighting. This species is characterized by harem structures where one, but occasionally a second, male defends a group of females. Fights occur both between intruder and resident males and between two resident males. Photo by Dan Rubenstein.
value. Rubenstein (1994) shows that contests among female horses are infrequent, occurring at a rate of about one every 10 hours. Similarly, foraging success of plains zebras is not impeded as harem size increases (Rubenstein 1994). This suggests that vegetation eaten by horses and plains zebras facilitates, rather than hinders, group living. As a result, contests over foraging location or individual food items should not be a major factor exacerbating reproductive differences among females. Second, as Figure 8.3 illustrates, sex-specific survivorship schedules for Shackleford horses are somewhat atypical for polygynous, sexually selected mammals. In this population male survivorship is not always lower than that of females. Instead the curves cross shortly after maturity. As Berger (1986) notes, competition is intense among juvenile equid males as they attempt to establish dominance relationships early in life. Not surprisingly, female survival is much higher during this period, when females are building up bodily condition. Once a male’s place in society is established, however, the emergence of stable linear dominance hierarchies among harem males (Rubenstein 1994) lowers the costs of maintaining dominance, thus increasing male
Sociality and reproductive skew in horses and zebras Sex Differences in Survival
1000
Number of Horses
800
Male Female
600
400
200
0
0
2
4
6 Age
8
10
12
Figure 8.3 Age-specific survivorship of male and female horses on Shackleford Banks.
survival. For females the opposite is true: once foaling commences, female mortality increases and remains high and constant from year to year. When taken together, both features suggest that augmenting feeding success to maintain good body condition will matter to females, but strategies for increasing intake should not involve contest competition. As hindgut fermenters and throughput feeders, foraging success is likely to be mediated more by time available for feeding, than by excluding neighbors from particular resource patches. Group-living females could increase foraging time by coordinating foraging and vigilance activities among themselves, or they could rely on males to assume most of the vigilance costs while they devote more time to foraging. Equid females choose the latter by selecting males that limit driving and herding behavior and whose heightened vigilance reduces sexual approaches and harassment by other males (Rubenstein 1986). Since male reproductive success is determined by the number of females in the group, reproductive patterns of both males and females will be affected more by social than by ecological factors. Features such as age, rank, and social stability are likely to have important consequences on differences in reproduction and what individuals can do to ameliorate them. Since reproductive
201
202
D. I. Rubenstein, C. M. Nun˜ez success depends both on how many young are born over a lifetime and how many survive to the age of independence – the age when direct parental investment ceases – we will use total number born over an extended period of time and total number that survive until 1 year of age as measures of success. Our analyses will necessarily examine windows of time during which individuals of different ages will span different portions of these times. Consequently, reproductive rate as measured by number of foals born per unit time and interbirth interval will also be measured. Since these measures of reproductive success are likely to be influenced by age, residuals about best-fit regression lines of each of the key reproductive variables on age will be used to control for age effects. Patterns of reproductive inequality and skew Reproduction among females and males in both horses and zebras is highly variable, but not always skewed (Table 8.1). For female horses having left their natal group for at least 3 years, only a few have failed to give birth during their lifetimes. The distribution of total births is bell-shaped, with 50% of the population giving birth to between two and six offspring during this period. Yearly foaling rate is similarly bell-shaped, with the average female giving birth approximately once every other year (Table 8.1). For plains zebras, moments of variance and skew are greater than those of horses for both the total number of young born and yearly foaling rate (Table 8.1). Reproduction in males for both species is more variable and more skewed than for females (Table 8.1). Figure 8.4 shows the distribution of the total number of foals sired by male horses; similar distributions and moments are seen for male plains zebras (Table 8.1). In general, measures of male fertility are more variable and show more skew than measures of female fertility in both harem-dwelling equids (Table 8.1).
Factors affecting female reproductive success Age
Many features of reproduction in horses and plains zebras are affected by age. On Shackleford Banks the fraction of females giving birth in any given year increases with age (Figure 8.5). Whereas only 27% of females less than 6 years of age have given birth, 55% of those older than 10 have done so. Moreover, the total number of foals produced during the 17-year period increases the longer a female lives (Figure 8.6) as does the number of offspring surviving to independence (Figure 8.7). In addition, as females age, their yearly
Sociality and reproductive skew in horses and zebras Table 8.1 Summary statistics of distributions of various measures of reproductive successes for male and female horses and plains zebras. Means, variances, and skews are calculated from the first three moments of the frequency distributions. For horses, statistics before and after management are presented. Males Mean
Females
Variance
Skew
Mean
Variance
Skew
Horses (natural conditions) Total young born
4.92
24.83
1.96
3.76
6.14
0.59
No. surviving to
–
–
–
3.04
3.08
0.69
1.11
0.44
0.30
0.42
0.04
0.48
independence Yearly foaling rate
Horses (after management) Total young born
2.08
5.51
1.16
2.05
0.79
0.74
No. surviving to
–
–
–
1.46
0.67
2.03
0.52
0.26
0.70
0.20
0.04
0.54
independence Yearly foaling rate Plains zebras Total young born
2.34
2.07
1.00
0.84
0.66
0.54
No. surviving to
–
–
–
0.16
0.15
2.32
0.66
0.21
1.78
0.29
0.07
0.81
independence Yearly foaling rate
25 20 15 10 5
io
ny si s Ad am Pa c St hin eg o as W us in st on Sp oc Li k za rd M er li D n ig ge Ph r oe ni x Sl as h Za ne C ae sa r Ev an To ob a Ax l C as e Ed g H e om e Lu r c Si ifer sy ph Sp us hy nx
0
D
Number of offspring sired
30
Stallion
Figure 8.4 Distribution of the number of offspring sired by individual male horses over their lifetimes. Moments of the distribution found in Table 8.1.
203
D. I. Rubenstein, C. M. Nun˜ez 0.6
Fraction giving birth/yr
0.5 0.4 0.3 0.2 0.1 0 Young
Middle Age Female Age
Old (> 10 yrs)
Figure 8.5 Fraction of female horses giving birth per year as a function of age. v2 ¼ 24.7, p < 0.0001; df ¼ 2; n ¼ 610.
12
B
10
No. of foals born
204
B BB
8 6 4 2
BBBBBB BBB BBB B BB BB B BB B BBBBBB BBBBBBB BBBBBBB
0
B
B
BBBBBBB 0 2 4 6 8 10 12 14 16 18 Female Age
Figure 8.6 Total number of foals produced by female horses since leaving their natal harems. F [1, 155] ¼ 679.48; p < 0.0001; r ¼ 0.91.
birth rate also increases (Figure 8.8). For plains zebra females, age-dependent birth and survival patterns are similar (Table 8.2). Rank
Although a greater proportion of high-ranking females give birth each year than middle- or low-ranking females (Table 8.3), rank is age-dependent (v2 ¼ 18.9; df ¼ 2; p < 0.001). When residuals of total number of foals born, or
No. foals surviving to independence
Sociality and reproductive skew in horses and zebras 7
B
6
B
5
BBB
4
B
3
BB
2
BBBB BBB BB
BBB BB
B
B
BBBBBBBBBB
1
BBBBBBBBB
B
0 0
2
4
6
8 10 12 14 16 18 Female Age
Figure 8.7 Number of foals surviving to independence at 1 year of age as a function of female age. F [1,97] ¼ 175.96; p < 0.0001; r ¼ 0.83.
Yearly foaling rate (no. born/yr.)
1
B
0.9 0.8
B
0.7
B
B
0.6
B
B B
B
B BB B B B B B B B B B B B B B B B B B B B B B B B B B B B B BB B B
0.5 0.4 0.3 0.2 0.1 0 0
B
B
B
BBBBBBB 2 4 6 8 10 12 14 16 18 Female Age
Figure 8.8 Yearly foaling rate by female horses of different ages. F [1, 155] ¼ 44.12; p < 0.0001; r ¼ 0.47.
total number surviving to age of independence, are compared for high-, middle-, and low-ranking females, high-ranking females outperform females of lower rank (Table 8.3). Male rank also has a strong effect on both the total number of young produced and the total reared to the age of independence by females after accounting for age (Table 8.3). Females bonded to males of higher rank bear
205
206
D. I. Rubenstein, C. M. Nun˜ez Table 8.2 Age-dependent birth and survival patterns in plains zebras x-variable
y-variable
Female age
Birth rate (mean)
(SE)
Old
0.30
0.02
Middle
0.26
0.04
Young
0.10
0.12
Female age
Total foals born (mean)
(SE)
Old
0.92
0.05
Middle
0.68
0.08
Young
0.21
0.13
F
df
p value
r
3.32
2,298
< 0.05
0.21
F
df
p value
r
6.58
2,352
< 0.002
0.25
more young, and more of them survive to independence, than females associating with males of lower rank (Table 8.3). Social stability
Since males vary in the degree to which they provide females with material rewards (Rubenstein 1986), females faring less well should switch groups in search of better conditions. Since older females have had more opportunities to change groups than younger females, residuals from the best-fit line of number of moves regressed against age were compared with residuals of the best-fit lines regressing total number of young produced, or surviving to 1 year of age, against age. In both cases, after controlling for age, females’ reproductive success is inversely related to the number of different harems a female joins. Females moving frequently among males bear fewer young (F [1, 152] ¼ 11.99; p < 0.001) and have fewer of them survive to independence (F [1, 107] ¼ 13.72; p < 0.0005) than females bonded more tightly to males. How long-term residency leads to enhanced reproductive success is related to reduced harassment by males. In general, females associating with higherranking males, or males rising in rank quickly, are herded and harassed less than females associating with lower-ranking males (Figure 8.9; Rubenstein 1994). And females that are harassed more tend to be females that are most likely to switch harems (Figure 8.10). How females determine with which male to associate after leaving a harassing male could be based either on cues learned from witnessing male–male contests (Rubenstein & Hack 1992) or from direct feedback of male herding tendencies after joining a new harem. If indirect cues were the deciding factor, then females should generally move from high- to low-harassment males on their initial foray. If such information
Sociality and reproductive skew in horses and zebras Table 8.3 Factors affecting female reproductive success in horses x-variable
y-variable
x-variable
y-variable
Female rank
Proportion giving birth
High
0.61
Middle
0.38
Low
0.31
x2 10.68
df
p value < 0.005
2
Residuals of total foals Female rank
versus age
F
(mean)
(SE)
High
0.28
.25
Middle
0.20
.31
Low
0.61
.27
3.33
df 2.54
p value
r
< 0.05
0.32
Residuals of number Female rank
surviving to independence (mean)
(SE)
High
0.34
0.18
Middle
0.21
0.28
Low
0.67
0.27
Female rank
Residuals of foaling rate (mean)
(SE)
High
0.08
0.04
Middle
0.05
0.05
Low
0.02
0.04
Male rank
Residuals of total foals (mean)
(SE)
High
0.016
0.15
Low
0.61
0.20
F 4.38
F 1.03
F 5.87
df 2.37
df 2.54
df 1.45
p value < 0.02
p value < 0.5
r 0.44
r 0.19
p value
r
< 0.02
0.33
Residuals of number Male rank
surviving to independence (mean)
(SE)
High
0.01
0.24
Low
0.63
0.27
F 3.27
df 1.26
p value
r
< 0.08
0.32
is not perceivable or is unreliable, then the initial male chosen should not on average lead to reduced harassment. As Figure 8.11 shows, females often find themselves in a worse situation after their initial choice, and only after many moves do they find males that harass them less than their original partner.
207
D. I. Rubenstein, C. M. Nun˜ez
Herds/Hr/Female
0.3
0.2
0.1
0.0 High
Low Male Rank
Figure 8.9 Rates of herding received by female horses in harems led by high- and low-ranking males (t ¼ 2.16; p < 0.05; df ¼ 14).
3 Male Harassment Rate
208
2
1
0
Mover
Stayer Type of Female
Figure 8.10 Rates of herding received by harem female horses that remain in the harem or ultimately leave (t ¼ 2.23; p < 0.05; df ¼ 22).
Factors affecting male reproductive success Age
Both the number of total young sired by a male and his yearly siring rate increase with age. As males mature they accumulate more offspring over time (F [1,60] ¼ 173.9; p < 0.0001; r ¼ 0.86), yet they also tend to sire more offspring per year (F [1,60] ¼ 5.84; p < 0.02; r ¼ 0.28). Both measures of
12 11 10 9 8 7 6 5 4 3 2 1 -0.5
Individual Females
Individual Females
Sociality and reproductive skew in horses and zebras
0.0 0.5 1.0 1.5 Initial Change in Harassment Rate
2.0
12 11 10 9 8 7 6 5 4 3 2 1 –2
–1
0
1
Final Change in Harassment Rate
Figure 8.11 Paired comparisons of changes in harassment rates received by 12 female horses leaving their harems. Initial changes refer to differences in the rates they received from their original male and the male of the first group they joined (paired t ¼ 6.2; p < 0.005; df ¼ 11). Final changes refer to differences in the rates received from their original male and the male whose group they finally move to (paired t ¼ 4.24; p < 0.005; df ¼ 11).
reproductive success result from the fact that males accumulate females as they age. On becoming a harem leader, the number of females in the harem is smaller than in the year before they lose their harem (Table 8.4). Rank
Dyadic interactions among Shackleford males yield clear-cut dominance rankings (Rubenstein 1994). These manifest themselves in two ways. First, for harems headed by single males, rankings are highly linear and second, harems headed by single males are always dominant to those led by two males. When the population of males whose rank remained the same throughout the study period is divided into thirds, males in the top third of the hierarchy sire more total offspring and have higher yearly reproductive rates than males in the middle or lower third of the hierarchy (Table 8.4). As before, these differences are the result of differences in the number of mares in a male’s harem (Table 8.4). When rank is assessed by type of harem, males that are sole harem leaders sire more total offspring and have higher yearly siring rates than males in dualmale harems (Table 8.4). Most multi-male harems (80%), however, are led by younger males that become single harem leaders later in life. When the effect of age is controlled for by plotting the residuals of total number of offspring sired or yearly siring rate against age, single-male harem leaders fare significantly better with respect to siring rate and slightly better in terms of total foals sired than multi-male leaders (Table 8.4). Even more striking is the fact
209
210
D. I. Rubenstein, C. M. Nun˜ez Table 8.4 Factors affecting male reproductive success in horses x-variable
y-variable
Harem longevity
Number of females (mean)
First year
1.73
0.29
Penultimate year
3.42
0.28
Harem longevity
Number of foals sired (mean)
df
p value
r
4.17
78
< 0.0001
0.82
t
df
p value
r
4.99
78
< 0.0001
0.20
F
df
p value
r
3.60
2,28
< .05
.61
F
df
p value
r
8.02
2,107
< 0.0006
0.73
F
df
p value
r
6.98
2,107
< 0.002
0.58
F
df
p value
r
2.22
1,45
0.14
0.20
F
df
p value
r
4.00
1,47
< 0.05
0.51
t
df
p value
r
2.15
11
< 0.05
0.54
t
df
p value
r
1.51
4
< 0.20
060
(SE)
First year
1.9
0.16
Penultimate year
0.75
0.14
Harem rank
Total foals sired (mean)
(SE)
High
8.3
1.62
Middle
3.8
1.61
Low
2.5
1.54
Harem rank
Yearly siring rate (mean)
(SE)
High
1.58
0.15
Middle
1.29
0.14
Low
0.60
0.19
Harem rank
Number of females (mean)
t
(SE)
(SE)
High
3.88
0.33
Middle
2.78
0.31
Low
1.96
0.23
Harem type
Total young sired (mean)
(SE)
Single male
4.62
0.75
Multi-male
1.20
2.16
Harem type
Yearly siring rate (mean)
(SE)
Single male
1.29
0.09
Multi-male
0.86
0.22
Harem history:
Residuals of number of
males leading both
foals sired
harem types Single male
0.21
0.24
Multi-male
0.45
0.19
Herding intensity
Residuals of number of
(number / hour)
foals sired
Low (< 1/hour)
1.78
0.41
High (> 1/hour)
0.91
0.25
Sociality and reproductive skew in horses and zebras that when the siring rates of males experiencing both roles over a lifetime are compared after correcting for age, siring rate is significantly higher for males when they become sole harem leader than when they share leadership with a partner (Table 8.4). Social stability
Just as harassment rate affects female movements and fertility, it affects the stability of harems and male reproductive performance as well. Males that direct the movements of their females by overzealous herding experience higher female movement rates (F [1,3] ¼ 44.22; p < 0.007; r ¼ 0.96). And, although only a strong trend, when the population of single harem leaders is divided into high-intensity and low-intensity herders, the yearly siring rate of low-intensity males is almost twice as high as that of highintensity males (Table 8.4). Adopting alternative reproductive strategies or balancing selfish effort and the granting of concessions for access to critical resources are often thought of as different mechanisms that individuals use to reduce inequality and skew. For male equids this need not be the case. Male horses display a variety of routes to adulthood (Figure 8.12). The typical route involves leaving the natal group between 2 and 3 years of age and joining a bachelor group consisting of other non-breeding males. Bachelorhood is a period where males wander and socialize with other bachelors and harem males. Consequently dominance relationships begin to be developed with a significant portion of the male population. When a bachelor rises to the top of a group’s hierarchy he begins to challenge harem leaders. In the typical route to adulthood about half of these high-ranking bachelors take over a harem intact (harem route); the other half form a bond with a young female dispersing from her natal group (monogamous route). Two other less common alternative routes to adulthood are possible (Figure 8.12). In one, a young dispersing male forgoes joining a bachelor group, instead joining a harem group first as a peripheral male and finally as a secondary male (satellite route). This type of coalition illustrates how dominants sometimes offer concessions to retain the services of subordinates, as assumed by concession models of skew (Johnstone 2000). In the other, the dispersing male joins a bachelor group, but rather than spending time socializing and ascending the dominance queue, he forms a partnership with another recently arrived young male and together they leave the bachelor group and form a bond with a young dispersing female (polyandrous route). The probabilities associated with each of these routes to adulthood are shown in Figure 8.12, and Table 8.5 combines these likelihoods with the
211
212
D. I. Rubenstein, C. M. Nun˜ez Male Mating Tactics Harem Group
j
j
subadult .30
Bachelors Satellite
subadult .70 Harem
.10 .90 .50
Multi Multi
.10 .40
Polyandrous
Monogamous
Figure 8.12 Pathways to adulthood of Shackleford horse males. All juveniles leave their natal groups. Females either join an existing harem group or bond to a highranking bachelor male. Males have four options available. They can follow the typical harem route by joining a bachelor group, rising in rank and taking over an existing harem. Or they can leave the bachelor group as an experienced young adult and bond with an emigrating female (monogamous route). Alternatively, they can join a bachelor group for a short period, leaving with a similar-aged low-ranking subadult and bond with an emigrating female (polyandrous route). Some males, however, forgo joining a bachelor group altogether and take up residence as a secondary male in an existing harem (satellite route). Probabilities of the various pathways are shown adjacent to pathway arrows.
reproductive values presented above to generate estimated lifetime reproductive success of males pursuing these different ontogenetic pathways. Not too surprisingly the typical harem route is expected to lead to the highest lifetime output. But what is striking is that two of the other trajectories are not far behind. And given that the patterns presented above show that there are large differences in both male and female reproductive success associated with
Sociality and reproductive skew in horses and zebras Table 8.5. Male reproductive payoffs Male mating strategy Reproductive component
Satellite
Polyandrous
Monogamous
Harem 6
Potential breeding years
10
8
8
Number of females
3 – >5
1 – >5
1 – >5
3 – >5
Inter-birth interval (years)
3 – >2
3 – >2
3 – >2
2
Number of receptive females
1 – >2.5
0.33 – >2.5
0.33 – >2.5
1.5 – >2.5
Percentage paternity
10 – >100
50
100
100
Yearly reproductive success
0.1 – >2.5
0.2 – >2.5
0.7 – >2.5
1.2 – >2.5
Lifetime reproductive success
10.3
7.3
10.4
12.8
rank of harem males, males adopting the satellite and monogamous routes are likely to perform as well as subordinates adopting the harem route. The typical harem route derives its reproductive advantage because males take over an entire harem, usually containing a full complement of breeding females (4–6). As described above, females in single-male harems, especially if they are stable and the male is of high rank, have high yearly foaling rates and low inter-birth intervals. If takeover males are able to prevent other males from stealing these females and their harassment rate does not drive females away, then the expected reproductive success of takeover males will be great. But this route comes with an opportunity cost associated with time lost while rising in rank within bachelor society. It is precisely this cost that allows males adopting alternative routes to adulthood to flourish. Males adopting the monogamous route trade off time spent in acquiring dominance within bachelor groups with time spent increasing the size of their harems from one female to many. And while males adopting the satellite tactic reduce this opportunity cost to a minimum by associating with many reproductive females with almost no delay, their fertility is likely to be low initially since single harem stallions typically sire about 90% of a harem’s offspring. While pairs of males adopting the polyandrous route also reduce the time to adulthood, they must share paternity, which will range from 10% to 50%. Behavioral alternatives, inequality, and skew Males
While this series of trade-offs makes it possible for alternatives to coexist, their existence reduces inequality in male reproductive success. If males did not avail themselves of these alternatives the reproductive disparity
213
214
D. I. Rubenstein, C. M. Nun˜ez among dominant and subordinate single harem stallions is likely be even greater. Moreover, the behaviors that contribute to reducing reproductive inequality and stabilize these alternatives are the same behaviors that reduce reproductive skew. When males are challenged by other males for access to females, males must balance driving the challenger away and insuring that their females are not poached or do not stray. If only the typical route to adulthood existed, and males were only challenged by one stallion at a time, then balancing these competing demands would be straightforward: after a short period of assessment the subordinates would submit and herd their females away; if not, then dominants would escalate (Rubenstein & Hack 1992). Although herding females away from trouble solves a subordinate’s immediate problem, the likelihood of female departures in the long run is high and thus provides the impetus for males to adopt alternative ontogenetic pathways. Since two-thirds of the alternatives entail young males attempting to get a head start on breeding by forming alliances either with similar age-mates (polyandrous route) or mature stallions on the decline (satellite route), the evolutionary stability of the alternatives depends on the behavioral dynamics that sustain these coalitions. In both types of dual-stallion harems, the subordinate stallion is almost always first in approaching and engaging an intruder male. If his approach causes the intruder to retreat then little is risked and little energy is expended. Given his subordinate status, however, these initial contests usually escalate. Sniffing, squealing, and then biting, chasing, or rearing are commonplace (Rubenstein & Hack 1992). Since the intruder is far from his home area, the resident often has an ownership advantage and many (44%) of the contests end with the resident subordinate winning. But in slightly more than half of such contests (56%) the subordinate retreats, inducing the dominant male to join the fray. But this is where the two types of coalitions differ. In coalitions emerging via the satellite route, in virtually all contests (98%) the resident dominant wins. What is important to note, however, is that before the partner joined the harem, the solo stallion was losing encounters and was dropping in rank. By associating with a younger partner the situation becomes reversed. In the three instances of satellite coalitions forming, each original harem holder had been losing females. With the creation of these partnerships the harems began to grow again. In one instance the group grew from two to ten females. In coalitions begun via the polyandrous route, both males are usually in the bottom half of the breeding hierarchy. As a result, even with the dominant engaging the intruder first, resident dominants win less than half of encounters (38%) they assume. In no case, however, was a female ever abducted or seen to stray during a contest.
Sociality and reproductive skew in horses and zebras Clearly, coalitions of males protect against losing females while increasing the likelihood of winning, and perhaps even create long-term benefits by deterring future intrusions from neighboring males. For their services, subordinate males receive mating opportunities. In polyandrous-originating coalitions, subordinate males are rarely prevented (7%) from mating with estrous females. They are allowed to flehmen, inspect and lick female genital areas and, if the female doesn’t kick while a subordinate is mounting or walk out from under him once he is mounted, then thrusting and ejaculation occurs without interference by the dominant stallion. In 96% of these cases the dominant male subsequently mates with the female. If the dominant mates first, however, the subordinate male usually attempts to mate but succeeds only 88% of the time. Thus sperm mixing is common and paternity sharing is likely (Ginsberg & Rubenstein 1990). Although the situation is similar for subordinate males opting for satellite-route partnerships, the percentage of mating attempts by subordinates is lower. Given that group size can be up to five times larger for satellite partnerships than polyandrous ones, absolute paternity benefits for subordinates in these two coalition types are not likely to be that different. In over 17 years of studying the Shackleford horse population before population control began, only 12 coalitions formed and only three lasted until one partner died (Rubenstein 1982). As the above analysis shows, the expected lifetime reproductive success of these alternative strategies is predicted to be lower than that of the typical route to adulthood. Therefore it is not surprising that most coalitions disband after a few years. But while they last, what mechanism maintains them? Females generally do not stray far from a group even when in estrus. And given that the initiation and culmination of a successful mating takes between 2 and 5 minutes, it is unlikely that matings by a coalition partner will go unnoticed. That subordinates mate so often and without interference suggests that dominants are allowing them to do so. While these behavioral events tend to support the idea that concession giving by dominants is a mechanism for maintaining the stability of groups, dominants continuously reinforce their dominance by initiating fights before, during, and after the breeding season. Maintaining rank differential within coalitions appears to be important, but whether it plays a role in conditioning inferiority in subordinates so that they initiate fewer matings, thus increasing inequality within the coalition, or whether it is part of the overall maturation process that leads to breaking up the partnership, is hard to determine. In one of the three cases of satellite-originating partnerships, the subordinate spent most of his time associating with all new females joining the harem as it grew. On a regular basis the dominant would disrupt these associations by chasing
215
216
D. I. Rubenstein, C. M. Nun˜ez the subordinate away. Eventually the subordinate left the coalition to take over a neighboring harem, and within a few days those females from the old harem with whom the subordinate had formed strong associations moved to his new harem. Thus it appears that dominants continuously reinforce their dominance over subordinates for good reason. Frequent low levels of aggression by dominants appear to limit freedom of movement by subordinates. But making subordinates “think twice” before attempting to mate also reveals that dominants are unable to enforce exclusive control of mating. Consequently, granting concessions and enforcing only limited control over subordinates – a form of selfish effort – appear to be operating jointly as Johnstone (2000) and Reeve et al. (1998) suggest and as Reeve & Shen (2006) explicitly capture in the “bordered tug-of-war” model. Females
As we have seen above, reproductive inequality and skew also exists among females, even if the magnitude of these differences is smaller than among males. Since female rank affects reproductive success, dominant females have some control over the reproduction of subordinates. Yet subordinates stay. Why do they do so? In part they stay because they are tolerated by the dominants. Subordinates help limit the number of immigrating females joining harem groups (personal observation). Even in species where scramble competition is the norm, as it is in equids, competition will increase if groups get too large. Therefore, allowing subordinates to stay as an aid in preventing even more females from joining represents a cost to dominants. As predicted by concession models of skew (Johnstone 2000), granting residency appears to be a price that dominants pay to derive even greater gains. But do dominants actually have the ability to determine the optimal size of their harems? The answer is mostly likely no – and understanding why provides a more compelling reason why subordinates stay. Stallions typically stand near immigrating females, protecting them from approaches by other females and thus limiting the ability of resident females to force immigrants from harems. Because male rank and the social stability it provides affects female reproduction more than does female rank (Rubenstein 1986), when male and female reproductive interests come into conflict, females are caught in a bind. Consequently, they rarely confront males, and male interests tend to prevail. The limited ability of females to control their own reproductive success is underscored by two unfortunate “experiments” that occurred on Shackleford Banks when two harem stallions inadvertently died during immobilization by another team of researchers. Since this immobilization was being carefully monitored, we were able to get detailed, fine-grained behavioral data on the
Sociality and reproductive skew in horses and zebras social behavior of the unguarded females. Almost immediately these groups were besieged by harem leaders and high-ranking bachelors. For up to four days these females fended off advances, but the intrusion rate was over 10 times higher than when their harem males were present. The percentage of time females foraged decreased from 72% of every hour to less than 40%, not only because of direct approaches by males, but because all female groups retreated to less frequented areas where forage was less abundant and of lower quality. After four days each group disintegrated. This inadvertent perturbation to the normal social structure highlights the central role that males play in the lives of females. Their increased vigilance reduces their grazing time to 80% of that of their females, but it also buffers their females from extensive intrusions. As a result, females are provided with increased time for grazing and a means of enhancing their reproductive success (Rubenstein 1986). Our observations also suggest that frequent reciprocal and mutualistic actions that would be required of females to hold groups together in the absence of males are costly. Thus female–female transactions in which dominants would limit their aggression towards subordinates could in theory encourage group cohesion as predicted by concession models of skew (Johnstone 2000). But even if females attempted to grant such concessions to foster residency, their actions would be superseded by female–male transactions. But what form are these transactions likely to take? Males have much to gain by keeping as many females as possible in harems. Subordinate males do this by adopting a “best of a bad job” strategy of herding females away from competitors rather than driving competitor males away from females (Rubenstein 1994). In the long run such behavior reduces subordinate harem size and ultimately reproductive success. Dominants, however, provide more stable and less stressful environments for females, where their reproductive success is enhanced. Dominants also attract more females than subordinates. For males, both factors increase reproductive success. Skew dynamics
Understanding how males and females balance the tension between cooperation and competition provides insights into the processes determining levels of reproductive inequality and skew. For group-living species, reproductive skew emerges when three conditions pertain: dominants must be able to limit the reproduction of subordinates; subordinates must do better by staying than by leaving; and dominants must do worse if subordinates leave rather than stay. For horses all three conditions are met, but because of novel social dynamics. First, although competition among females is limited and involves “scramble” rather than the “contest” behavior, dominant females
217
218
D. I. Rubenstein, C. M. Nun˜ez control the reproduction of subordinates to a limited degree (Table 8.3). Second, subordinate females do help dominants by preventing additional females from entering the group and intensifying competition. And third, subordinates faring poorly sometimes do leave harems in search of better opportunities, but the process of integrating into new ones is difficult and not without costs (Figures 8.9–8.11). Thus for some subordinate females that are balancing the marginal gains of staying with leaving, concessions could tip the balance enticing them to stay and join in mutualistically beneficial behavior. But dominants do not appear to offer such concessions because males intervene to keep these subordinates – as well as many future subordinates – in their groups. Since the size of harem groups that benefit polygynous males will usually be larger than those that benefit females (Armitage 1981, 1986), reproductive tension is created among males and females. While agonism by dominant females does induce some subordinates to leave when groups get large, aggressive herding by males usually overrides the actions of their dominant females. But if males are too aggressive, subordinate females receiving herding actions by stallions will leave anyway. Thus, to maintain large harems, dominant males must curtail excessive aggressive behavior. Ultimately, the ability of males to balance agonistic and affiliative tendencies towards females will determine how many, and which, females stay. Just as dominant males in multi-male harems maintain their coalitions by conceding reproduction while simultaneously engaging in occasional bouts of aggression to condition subordinates to limit reproductive competition, harem males will strive to maintain harems larger than preferred by their females by using aggression against females in limited and discretionary ways. Rarely in animal societies are reproductive interests among group-living individuals of one sex mediated by interactions and transactions with members of the other sex. But it may be widespread in the equids. In plains zebras, herds form as harems coalesce (Rubenstein & Hack 2004). Harems come together because cuckolding pressure imposed by bachelors is high, higher than levels reached in horses. Presumably high levels of predation favor bachelor males forming long-term associations with a large number of males facing similar risks. When confronted with coordinated actions of many bachelors, harem stallions reduce their individual cuckolding risk by banding together and keeping bachelors on the edge of the herd (Rubenstein & Hack 2004). Clearly males benefit by forming herds, but they can only do so because female foraging success is similar whether harems are in or out of herds. To insure that females remain indifferent, males in herds incur reductions in time spent foraging above those incurred when their harems are alone on a
Sociality and reproductive skew in horses and zebras landscape (Rubenstein & Hack 2004). This represents a significant concession to maintaining herd stability.
Impact of management on reproductive inequality and skew In 2000 the US National Park Service began managing the Shackleford horse population with PZP immunocontraception of females. Along with yearly removal of females, this kept horse numbers well below carrying capacity. One immediate consequence of these interventions was an increased birth rate and a release from density-dependence. What little competition for resources existed in the more natural state, the current state offers even less. Other consequences were seen with respect to changes in behavior. Young females, especially those coming into estrus for the first or second time, showed reductions in time spent feeding and increases in time spent standing (Rogers 2001, Constantino 2002). Males also showed greater tendencies to herd females, especially those having received contraception (Stroeh 2001). After six years of intervention, patterns of reproductive inequality and skew of males and females have changed. As shown in Table 8.1, the variance in the distribution of total births, yearly foaling and siring rate, and the number of young surviving to the age of independence have all decreased. In many respects, the underlying factors affecting these measures of reproductive success remain unchanged after management. Age still has a major impact on the number of foals born during the 6 years (F [1,42] ¼ 8.15; p < 0.007; r ¼ 0.40), on the foaling rate (F [1,77] ¼ 56.79; p < 0.0001; r ¼ 0.64), and on the number of harem changes made by females (F [1,70] ¼ 8.95; p < 0.005; r ¼ 0.33). After computing appropriate residuals to account for the influence of age, female rank still affects the total number of foals born throughout the period as well as the annual foaling rate (Table 8.6). And while females associating with males of higher rank perform better reproductively than females associating with lower-ranking males, the effect is of reduced significance (Table 8.6). Since females receiving contraception continue to cycle throughout the entire reproductive season, genital inspection and harassment remains high for all males (Stroeh 2001). Consequently, differences in the way males of different ranks interact with females have been diminished. Reducing both the variance and skew in the number of foals born or sired during this six-year period should increase effective population size, since more females will contribute similar numbers of genes to future generations and a greater proportion of males will sire offspring (Falconer 1960). Given that the US Park Service is holding population numbers between 120 and 130, supporting
219
220
D. I. Rubenstein, C. M. Nun˜ez Table 8.6 Factors affecting reproductive success in horses after management commenced x -variable
y-variable
Female rank
Total foals sired (mean)
F
df
p value
r
37.17
2,31
< 0.003
.57
3.72
2,48
< 0.03
0.36
1.59
1,70
< 0.25
0.24
2.92
2,19
<0.07
0.28
(SE)
High
0.54
0.19
Middle
0.24
0.36
Low
0.43
0.18
Female rank
Yearly foaling rate (mean)
(SE)
High
0.075
0.04
Middle
0.003
0.07
Low
0.60
0.03
Female rank
Residual numbers of mares (mean)
(SE)
High
0.25
0.26
Middle
-0.27
0.47
Low
0.32
0.23
Male rank
Foaling rate (mean)
(SE)
High
0.27
0.05
Middle
0.20
0.08
Low
0.01
0.07
inherent features of a species’ social biology that reduce the loss of genetic diversity is important.
Discussion In our populations of horses and plains zebras reproductive inequalities and skew occur for males and females. At least for male horses, variation and skew would have been greater had it not been for the opportunity for lowperforming males to join groups of different types. Such variants are maintained because of the many ways in which adjustments in concessional and selfish effort can be made. In all multi-male harems, complete control by dominants is absent either because aging males are dropping in rank (satellite route) or because young partners begin their reproductive lives as inexperienced equals (polyandrous route). Both situations meet the assumptions of the
Sociality and reproductive skew in horses and zebras bordered tug-of-war model, and the relatively peaceful coexistence that emerges fits one of the model’s predictions (Reeve & Shen 2006). If there are no strong ecological or social reasons for maintaining groups, the model predicts that the degree of strife within groups will be low. Given that joining bachelor groups is a ready alternative that only incurs a moderate cost of delaying reproduction for a few years, it is not surprising that fights among coalition partners are infrequent and of low intensity. Coalitions among males are not limited to our horse and plains zebra populations. While coalitions are relatively rare in our study (< 10% of harem groups in any given year), they are common in other horse populations. On a nearby North Carolina island 50% of the harem groups were led by two males (Stevens 1990); multi-male groups comprise 12% of harem groups in the Great Basin (Berger 1986); 33% of groups in the Kaimanawa Range are led by multiple males; and 38% of the reproductive units in the Camargue horses contain two or more males (Feh 1999). Recently there has been much debate as to whether these multi-male units are coalitions formed by mutualistic or reciprocal relationships (Feh 1999, Linklater & Cameron 2000a). Linklater (2000) and Linklater & Cameron (2000b) claim that multi-male associations are not true coalitions in part because the authors assume that the reproductive success of males and females inhabiting multi-male groups must equal that of males and females in uni-male harems. Otherwise, they could not be maintained by selection. Game theoretic models that account for differences in phenotype (Parker 1984) offer many alternative predictions concerning reproductive payoffs that would permit multi-male social units to be maintained. And if males and females residing in them do so only temporarily as part of a complex life cycle, then the existence of such social variants is easier to sustain; as payoffs produced during one period are incorporated into measures of overall lifetime reproductive success their contribution is diluted. Feh’s field study (1999) and critique (2001) provide ample evidence that multi-male groups are functional coalitions. As a result, understanding the behavioral dynamics that hold them together can provide insights into how evolutionary forces generate skewed reproductive patterns. In accord with concession models of skew (Johnstone 2000), we have shown the degree to which both dominants and subordinates act selfishly and provide concessions that maintain these groups until balance among the payoffs changes. And change they will as phenotypes and environmental conditions change. Feh (1999) describes trade-offs associated with the different roles that are similar to those we find: subordinate coalition members take more risks and dominants rarely interfere with matings of subordinates, although they are aware that they are occurring. Feh also goes one step further and calculates the degree to
221
222
D. I. Rubenstein, C. M. Nun˜ez which dominants allow subordinates to mate. In the Camargue population, subordinates sire about 25% of multi-male groups’ offspring, a value that is consistent with our expectations. What makes our study unique is that we also explore the more complicated dynamics underlying unequal and skewed reproduction of females. While not as great as in males, female reproductive skew is also significant. Despite abundant vegetation and physiologies that make subsistence on low-quality food economical, agonism persists among females and dominance hierarchies emerge. That dominance affects reproduction demonstrates harem females invest in selfish behavior. Mutual support among females in keeping additional females from joining harems benefits females of all ranks, and shows that dominants are prepared to provide concessions to retain the services of subordinates. But group cohesion can only be maintained if subordinates also offer concessions to dominants at critical times. Such concessions are seen occasionally at watering points where access is extremely limited (Rubenstein 1994). Yet even though these conditions meet the assumptions of the bordered tug-of-war model, they are not the conditions that matter most in shaping skewed reproduction of females. Male preferences for even larger harems generate pressures to override the preferences of females. But in meeting their selfish needs, males could often drive away precisely those females they want to retain. Thus males are left balancing agonistic and affiliative tendencies with those of their females. As the model suggests, the optimal mix is likely to change with changes in group size as well as demography. Thus we expect that when there are many harems available rates of herding and harassment should decrease. The bordered tug-of-war model predicts that when ecological conditions are harsh, group breakup will be constrained and destructive conflict within groups should increase. What makes equids special is that they are among the few grazers that experience little competition. Marmots (Marmota spp.) are also grazers, but they are less free to wander widely in search of forage since they are tied to burrows for safety from predators (Armitage 1981, 1986). Thus they should experience greater competition than horses and plains zebras and they should exhibit more strife among dominants and subordinates. This appears to be true for most of the socially monogamous species. The need to maintain warmth during hibernation apparently selects for small numbers of young and thus complete reproductive suppression of subordinates by dominants (Allaine´ 2000). Reproductive skew is less pronounced in the yellow-bellied marmots (Marmota flaviventris), even though all females compete for limited resources. Apparently females cooperate in keeping other groups of females from usurping parts of the home range (Allaine´ 2000), much as female equids cooperate to keep additional immigrant females from joining harems. In both species emigration
Sociality and reproductive skew in horses and zebras by females is possible, thus reducing skew, but in marmots, males appear not to be involved in adjusting concessional and selfish investments. In our study we showed that the dynamics shaping sociality in horses operate in plains zebras as well. Dominance differences set the stage for inequitable and skewed reproduction in males and females of both haremliving species. The importance of male rank in balancing concessional and selfish effort is a thread in both species’ social cloth. Yet in plains zebras this thread appears stronger where ecological conditions make it more difficult for males to maintain control of females and their reproduction. Incomplete control of reproduction necessitates transactions between dominants and subordinates in order to stabilize social orders. When these transactions involve individuals within and between the sexes it is not surprising that a variety of mating and social structures can coexist. This is especially true for horses, and should be searched for in other species where conflicts of interest among many phenotypes are common.
Conclusions Inequality in reproductive success in horses and zebras exists in both sexes, although both variance and skew is greater for males than for females. Differences in age and status shape the distributions of reproduction for both males and females. Social stability also determines degree of skew. After controlling for age, females that change groups often breed less often than those that remain attached to particular males for long periods of time. And for males, greater reproductive success accrues to those that curtail harassment of females. Perhaps what is most striking about equids that form long-term breeding associations among males and females, as is the case in horses and zebras, is that the magnitude of skew is lower than could be expected. Such attenuation emerges because of two factors. One is that males can adopt a number of alternative reproductive tactics. Each alternative relies on concessions to subordinates that boost reproduction above what would be achieved at that age by adopting the typical tactic. Moreover, each alternative tactic enhances the reproduction of the concession giver over what he would have gained without joining the partnership. Similarly for females, concessions enable non-harem females to gain access to a dominant male while limiting the loss of reproduction that females already bonded to such an “attractive” male would incur if group size were to grow even larger. The other factor is that intersexual conflict reduces reproductive differences within the sexes because neither males nor females can completely enforce their self-interest. While males would prefer larger harems, which would enhance male skew,
223
224
D. I. Rubenstein, C. M. Nun˜ez the aggression necessary to effectively maintain large groups would most likely drive females away. And for females the ability to keep group size small by peripheralizing additional females, thus reducing female skew by allowing wandering females to be spread more equitably among males, is modulated by male herding. Thus, while phenotypic differences such as age and size and their distribution within populations play important roles in shaping patterns of skew, a variety of intra- and intersexual relationships that characterize harem-living equids also influence sex-specific patterns of skew. The extent to which the interplay between these forces shapes skew in other harem-living species, or even species exhibiting other mating systems, needs to be explored. Acknowledgments We thank the US National Park Service, people and the Government of Kenya (Research permit MOST 13/001/29C 80 vol. II) for allowing us to carry out this study, and Cape Lookout National Seashore (North Carolina), the Mpala Research Centre (Kenya) and the Ol Pejeta Conservancy (Kenya) for allowing us to work on their lands. Support for these studies came from the National Science Foundation awards CNS-0205214, IOB-9874523, and BSR-8352137. Thanks also go to Mace Hack, Justine Cordingley, Jessica Rogers, Oliver Stroeh, and Virginia Costantino for their assistance in gathering field data.
References Allaine´, D. (2000). Sociality, mating system and reproductive skew in marmots: evidence and hypotheses. Behavioural Processes, 51, 21–34. Armitage, K. B. (1981). Sociality as a life-history tactic of ground squirrels. Oecologia, 48, 36–49. Armitage, K. B. (1986). Marmot polygyny revisited: determinants of male and female reproductive strategies. In D. I. Rubenstein & R. W. Wrangham, eds., Ecological Aspects of Social Evolution: Birds and Mammals. Princeton, NJ: Princeton University Press, pp. 303–331. Berger, J. (1986). Wild Horses of the Great Basin. Chicago, IL: Chicago University Press. Clutton-Brock, T. H., Brotherton, P. N. M., Russell, A. F., et al. (2001). Cooperation, control, and concession in meerkat groups. Science, 291, 478–481. Costantino, V. S. (2002). Immunocontraception on feral mares: intended and unintended effects on hormone levels and behavior. Unpublished B. A. senior thesis, Princeton University. Engh, A. L., Funk, S. M., Horn, R. C. V., et al. (2002). Reproductive skew among males in a female-dominated mammalian society. Behavioral Ecology, 13, 193–200.
Sociality and reproductive skew in horses and zebras Falconer, D. S. (1960). Introduction to Quantitative Genetics. New York, NY: Ronald Press. Feh, C. (1999). Alliances and reproductive success in Camargue stallions. Animal Behavior, 57, 705–713. Feh, C. (2001). Alliances between stallions are more than just multimale groups: reply to Linklater and Cameron (2000). Animal Behaviour, 61, F27–F30. Garnier, J. N., Bruford, M. W., & Goossens, B. (2001). Mating systems and reproductive skew in the black rhinoceros. Molecular Ecology, 10, 2031–2041. Ginsberg, J. R. & Rubenstein, D. I. (1990). Sperm competition and zebra mating behaviour. Behavioral Ecology and Sociobiology, 26, 427–434. Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Klingel, H. (1977). Observations on social organization and behaviour of African and Asiatic wild asses (Equus africanus and Equus hemionus). Zeitschrift fur Tierpsychologie (Ethology), 44, 323–331. Linklater, W. L. (2000). Adaptive explanation in socio-ecology: lessons from the Equidae. Biological Reviews of the Cambridge Philosophical Society, 75, 1–20. Linklater, W. L. & Cameron, E. Z. (2000a). Distinguishing cooperation from cohabitation: the feral horse case study. Animal Behaviour, 59, F17–F21. Linklater, W. L. & Cameron, E. Z. (2000b). Tests for cooperative behaviour between stallions. Animal Behaviour, 60, 731–743. Lloyd, P. H. & Rasa, O. A. E. (1989). Status, reproductive success and fitness in Cape mountain zebra (Equus zebra zebra). Behavioral Ecology and Sociobiology, 25, 411–420. Marx, K. (1875). Marx/Engels Selected Works, Vol. 3. Moscow: Progress Publishers, 1970. Moehlman, P. D. (1979). Behavior and ecology of feral asses (Equus asinus). National Geographic Society Research Reports, 405, 405–411. Parker, G. A. (1984). Evolutionary stable strategies. In J. R. Krebs & N. B. Davies, eds., Behavioural Ecology: an Evolutionary Approach, 2nd edn. Oxford: Blackwell, pp. 30–61. Reeve, H. K. & Shen, S. -F. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders. Behavioral Ecology, 9, 267–278. Robbins, A. M., Robbins, M. M., Gerald-Steklis, N. & Steklis, H. D. (2006). Age-related patterns of reproductive success among female mountain gorillas. American Journal of Physical Anthropology, 131, 511–521. Rogers, J. E. (2001). The effects of an immunocontraceptive on the social behavior of female feral horses (Equus caballus). Unpublished B.A. senior thesis, Princeton University. Rubenstein, D. I. (1978). On predation, competition, and the advantages of group living. Perspectives in Ethology, 3, 205–232. Rubenstein, D. I. (1980). On the evolution of alternative mating strategies. In J. E. R. Stadden, ed., Limits to Action: the Allocation of Individual Behaviour. New York, NY: Academic Press, pp. 65–100.
225
226
D. I. Rubenstein, C. M. Nun˜ez Rubenstein, D. I. (1981). Behavioural ecology of island feral horses. Equine Veterinarian Journal, 13, 27–34. Rubenstein, D. I. (1982). Reproductive value and behavioural strategies: coming of age in monkeys and horses. Perspectives in Ethology, 5, 469–487. Rubenstein, D. I. (1986). Ecology and sociality in horses and zebras. In D. I. Rubenstein & R. W. Wrangham, eds., Ecological Aspects of Social Evolution: Birds and Mammals. Princeton, NJ: Princeton University Press, pp. 282–302. Rubenstein, D. I. (1994). The ecology of female social behavior in horses, zebras and asses. In P. Jarman and A. Rossiter, eds., Animal Societies: Individuals, Interactions and Organisations. Kyoto: Kyoto University Press, pp. 13–28. Rubenstein, D. I. & Hack, M. (1992). Horse signals: the sounds and scents of fury. Evolutionary Ecology, 6, 254–260. Rubenstein, D. I. & Hack, M. (2004). Natural and sexual selection and the evolution of multi-level societies: insights from zebras with comparisons to primates. In P. Kappeler and C. P. van Schaik, eds., Sexual Selection in Primates: New and Comparative Perspectives. Cambridge: Cambridge University Press. Rubenstein, D. I., Sundaresan, S., Fischhoff, I., & Saltz, D. (2007). Social networks in wild asses: comparing patterns and processes among populations. In A. Stubbe, P. Kaczensky, K. Wesche, R. Samjaa, & M. Stubbe, eds., Exploration into the Biological Resources of Mongolia, Vol. 10. Halle: Martin-Luther-University HalleWittenberg, pp. 159–176. Stevens, E. (1990). Instability of harems of feral horses in relation to season and presence of subordinate stallions. Behaviour, 112, 149–161. Stroeh, O. M. (2001). The effects of management strategies on the behavioral ecology of the Shackleford Banks male horses. Unpublished B.A. thesis, Princeton University. Sundaresan, S. R., Fischhoff, I. R., & Rubenstein, D. I. (2007). Male harassment influences female movements and associations in Grevy’s zebra (Equus grevyi). Behavioral Ecology, 18, 860–865. Vehrencamp, S. L. (1983). Optimal degree of skew in cooperative societies. American Zoologist, 223, 327–335. Widdig, A., Bercovitch, F. B., Streich, W. J., et al. (2004). A longitudinal analysis of reproductive skew in male rhesus macaques. Proceedings of the Royal Society of London, B, 271, 819–826.
9
Reproductive skew in avian societies w a l t e r d . k o e n i g , sh e n g - f e n g s h e n , al a n h . kr a k a u e r , and joseph haydock
Summary Reproductive skew theory provides a predictive theory of the extent of reproductive sharing that is expected to occur in societies consisting, at least potentially, of multiple co-breeders. Here, we discuss some of the challenges that skew theory faces as it attempts to form the basis of a unified theory of social evolution in birds. These include the problem of distinguishing potential versus actual reproductive roles, encompassing extra-group parentage and sexual conflict, predicting the distribution of group size, and determining the appropriate null model against which to test empirical results. Despite these and other problems with skew theory as currently developed, a compilation of prior studies indicates some degree of consistency with the predictions of concession or optimal skew theory. More surprisingly, a meta-analysis indicates that interspecific patterns of sociality offer reasonably good matches to the predictions of the concession model of reproductive skew. Strong support for skew theory remains lacking, and experimental tests sufficient to reject alternative skew models have yet to be performed in birds. Nonetheless, these results offer encouragement that additional theoretical work in this field may eventually yield a useful framework for understanding the remarkable diversity of avian sociality.
Introduction An irony of kin-selection and inclusive-fitness theory (Hamilton 1964) is that these concepts, conceived primarily as a solution to the paradox of Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
227
228
W. D. Koenig et al. altruism, should turn out to be so useful in providing a framework for interpreting conflict among close relatives. Indeed, as more social species have been studied in detail, it has become clear that cooperation and conflict go hand in hand in most, if not all, societies. Nowhere is the uneasy coexistence of cooperation and conflict more evident than in cooperative breeders, in which more than a pair of individuals assist in the production of offspring. Cooperatively breeding birds have proved a particularly fertile taxon for studying social behavior, both because they show a considerable range in mating systems and because they are generally diurnal, observable, and easily manipulated. Equally important, the close associations of individuals from multiple generations found in such societies result in opportunities for conflict and competition rarely found in non-group-living taxa. Simultaneously, the fact that associated individuals are often close genetic relatives opens the door for the evolution of altruistic and cooperative behaviors that are less likely to evolve among non-relatives that may only interact one or a few times during their lives. The standoff that appears to exist between the highly developed cooperation and equally sophisticated competition observed in cooperative breeders is to a considerable extent what makes such species so exciting and interesting to study. Most commonly, cooperatively breeding groups consist of nuclear families in which only a pair of individuals produce offspring that then delay dispersal, remaining on their natal territories as “helpers-at-the-nest” for a variable length of time, often several years, rather than disperse and attempt to breed on their own. Such helpers are typically non-breeders, presumably because of incest avoidance, despite being reproductively mature and fully capable of breeding. A classic example is the Florida scrub-jay (Aphelocoma coerulescens), which is both socially and genetically monogamous but lives in territorial groups containing up to several non-breeding helpers (Woolfenden & Fitzpatrick 1984, Quinn et al. 1999). Helpers in this situation can be considered subordinates in a high-skew society, although, as discussed below, whether this approach yields new insight into the situation or not is debatable. Beyond such cases, there is now an increasing number (although still a minority) of cases in which three or more individuals share reproduction, at least potentially, to a greater or lesser extent. These systems are quite diverse. A few species, most notably several species of manakins in the family Pipridae and the wild turkey (Meleagris gallapavo), exhibit what is known as “cooperative courtship” in which males form coalitions for the purpose of attracting and mating with females, but do not otherwise contribute parental care (Foster 1977, MacDonald 1989, Krakauer 2005a). More commonly birds form bisexual groups within which the mating system is quite variable, including
Reproductive skew in avian societies “cooperative polygyny” (if groups consist of one male and multiple females), “cooperative polyandry” (one female and multiple males), or “cooperative polygynandry” (multiple breeders of both sexes). Two examples of the latter are the acorn woodpecker (Melanerpes formicivorus: Koenig & Mumme 1987) and the dunnock (Prunella modularis: Davies 1992), both of which exhibit cooperative polygynandry in which more than one male may compete for matings with multiple females. The focus of this book, reproductive skew, provides a predictive and potentially quantitative theory as to the degree of reproductive sharing that should take place in such societies. Skew theory currently involves at least two distinct approaches. The first approach encompasses what are known as “transactional” skew models, in which participants are hypothesized to engage in a kind of social contract by which they agree to cooperate in the group in exchange for the assurance of obtaining sufficient fitness benefits to outweigh the alternative of leaving the group and breeding solitarily. Transactional models generally assume complete control of reproduction by one individual and a high degree of mutualism among the participants, focus on different assumptions (such as the possibility of reproduction being increasingly costly as more young are produced: Cant & Johnstone 1999), and explicitly consider group stability as well as the degree of reproductive sharing. Thus far, most empirical work has focused on testing transactional models, and thus we will primarily focus on these models here as well. A second framework, known as “compromise” models, ignores group stability, instead postulating that skew is determined solely by the relative competitiveness of the participating individuals (Reeve et al. 1998, Johnstone 2000). Thus, the most common form of compromise models, “tug-of-war models,” do not assume any social contract among the individuals involved. On the other hand, by eliminating the often unrealistic assumption of complete control, tug-of-war models accommodate the likelihood of overt competition among co-breeders, a frequently observed phenomenon that is not expected under transactional models (K. Reeve, personal communication). Attempts to combine the cogent aspects of these two approaches have been made by both Reeve (2000) and Johnstone (2000), and more recently by Reeve & Shen (2006) by means of a “bordered” tug-of-war model.
Goals As discussed in detail elsewhere in this volume, skew theory provides a framework integrating ecologic, genetic, and social factors to predict the patterns of reproductive partitioning in animal societies. The goal of skew theory
229
230
W. D. Koenig et al. is ambitious: to provide a unified theory of social evolution ultimately leading to “a comprehensive theory that relates all of the major features of animal societies to their ecological contexts, genetic structures, and internal power asymmetries” (Reeve & Keller 2001, p. 348). Although no one is likely to be surprised if this goal is not easily attained, it is clearly desirable to know whether skew theory as we currently understand it is heading in the right direction and appears capable of providing insight into the social behavior of different kinds of societies. One of our primary aims here is to see whether these goals are being met with the data available from avian societies. First we discuss several key issues related to avian social behavior that skew theory has largely ignored, or that are difficulties in applying skew theory as currently developed. Although there is overlap, we do not attempt to duplicate all the issues raised by the several excellent prior reviews of reproductive skew in birds (Johnstone 2000, Magrath & Heinsohn 2000, Magrath et al. 2004). As these reviews make clear, the available empirical data in birds fall far short of what is needed to unambiguously test the assumptions of any particular skew model, much less test quantitative predictions in ways that allow discriminating among the different models. Although we fully agree with this rather damning criticism, our goal here, as in any good review, is to stimulate rather than stifle further research. Toward this end, we devote the latter half of the chapter to examining the empirical evidence, both intra- and interspecific, relevant to reproductive skew and avian social behavior, concluding with a prospectus of how we envision skew theory evolving in the future and a discussion of the extent to which it is ever likely to meet Reeve & Keller’s (2001) ambitious goal.
Skew theory and birds: issues Before summarizing the available reproductive skew data for birds, we discuss a series of problems having to do generally with applying skew theory to avian social organization. The issues we focus on are not necessarily limited to birds and are by no means comprehensive. For additional critiques of skew theory as it applies to birds, see Johnstone (2000), Magrath & Heinsohn (2000), and Magrath et al. (2004). Incest avoidance and potential versus actual reproductive roles
Skew quantifies the extent to which reproduction is shared among potentially reproductive individuals. Thus, in order to measure skew, individuals that are not potentially reproductive should be excluded. Often this is neither controversial nor ambiguous: for example, it is generally clear that
Reproductive skew in avian societies fledglings or other immature, non-reproductive individuals should be excluded, since such individuals are incapable of breeding. Unfortunately, this distinction quickly becomes problematical when considering adult helpers that are physiologically capable of breeding but constrained socially from doing so. If the constraint involves reproductive suppression on the part of a dominant, such individuals should clearly be considered as potential breeders, as this is exactly the kind of situation for which skew theory was developed. But what if the constraint is incest avoidance? Incest avoidance is a genetic mechanism that functions as a constraint on breeding by reducing the potential fitness benefits of reproduction between closely related individuals. Helpers in many cooperative breeders live in groups where their only breeding options are incestuous, and consequently are non-breeders not because they are physiologically incapable of breeding but because the fitness benefits of doing so are presumably too low to make the effort worthwhile. Although this contrasts sharply with co-breeders that do not receive a share of reproduction, distinguishing the two can be problematical. Consider the acorn woodpecker, whose social system involves both matesharing by relatives and offspring that delay dispersal to become non-breeding helpers (Figure 9.1). We know that incest is not only avoided in this population, but that incest avoidance, rather than reproductive competition, is the primary mechanism restricting reproduction by offspring (Koenig et al. 1998, Koenig & Haydock 2004). Consequently, we are able to define non-breeding helpers operationally as birds fledged within the group whose parent of the opposite sex (or an opposite-sex co-breeder of their parent) is still present within the group. Meanwhile, co-breeders are defined based on their potential, rather than their actual, success at parenting offspring. For joint-nesting females, reproductive skew is absent and in most cases all females classified as breeders lay eggs (Haydock & Koenig 2002). For co-breeding males, however, things are not so straightforward. Bias in reproductive success between co-breeder males appears to be quite strong, on the order of 6 : 1 or more for individual nests, such that a majority of broods are sired by a single male and many co-breeding males do not sire offspring in any individual nesting attempt. Even if they successfully copulated with the female and their sperm had good opportunity to fertilize eggs, many males will by chance alone fail to sire offspring in a particular nest, since the number of young fledged is usually small, on the order of 2–5. Such “failed breeders” cannot be distinguished from nonbreeding helpers based on patterns of parentage, since neither sire any offspring. Instead, the distinction between the two classes of birds rests solely on
231
232
W. D. Koenig et al.
Figure 9.1 Two adult male acorn woodpeckers, a group-breeding species common in oak woodlands of western North America. Acorn woodpeckers exhibit both helping at the nest by non-breeders and cooperative polygynandry in which multiple co-breeder males compete for matings with 1–3 joint-nesting females, all of which lay their eggs in the same nest cavity. Co-breeders of the same sex (males and females) are almost always close relatives, usually siblings but sometimes parents and offspring or a combination of the two. Skew is low among joint-nesting females, who regularly destroy each other’s eggs but ultimately share maternity of eggs equally. In contrast, one male often sires all young in a nesting attempt, and skew among co-breeding males is thus apparently quite high. This results in two distinct kinds of non-reproductive males: helpers that do not attempt to mate due to incest avoidance, and potential cobreeders that fail to sire young in a particular breeding attempt. Photo ª 2000 by Ron Mumme.
the difference between their perceived reproductive roles. Is it necessary or desirable to distinguish these two sets of individuals? Dismissing the potential reproductive status of failed breeders and classifying them as non-breeding helpers along with the offspring that are constrained
Reproductive skew in avian societies from breeding due to incest avoidance clearly obscures an important functional difference between them and can be misleading. Reeve & Keller (1995), for example, proposed that the lack of reproduction by helpers observed in many cooperative breeders in which helpers are offspring, compared to systems in which helpers are siblings, supports a prediction of optimal skew theory stemming from the difference in symmetry of genetic relatedness between the potential breeders in the two situations and their offspring. As pointed out by Emlen (1996), however, the observation that helpers that are offspring do not usually breed is more parsimoniously explained by incest avoidance rather than optimal skew theory. That is, because of incest avoidance, such helpers are not part of the breeding activities of the group and therefore their lack of reproduction can neither be explained by skew theory nor provide support for it. On the other hand, there are at least two arguments for including nonbreeding helpers that are offspring of the “true” breeders in a group when considering or trying to measure reproductive skew. First, although such individuals have been empirically found not to successfully parent offspring in many species, excluding them rests on the assumption that reproductive suppression is primarily, if not exclusively, due to incest avoidance rather than reproductive competition or some other mechanism that is accommodated by skew theory. This is a reasonable assumption only if one agrees that incest avoidance is a strong and universal factor affecting mating patterns in vertebrate societies, a hypothesis that has certainly not been generally accepted historically (Bengtsson 1978, Shields 1982, Pusey & Wolf 1996), and remains controversial today for both theoretical and empirical reasons (Pusey & Wolf 1996, Koenig & Haydock 2004). The second argument for including non-breeding helpers when measuring skew is that even if incest avoidance is important in a society, “non-breeding helpers” are not necessarily excluded from reproduction. In rare cases they may breed incestuously; although incest avoidance appears to be strong in most vertebrate societies, it is not absolute (e.g. Haydock et al. 2001, Koenig & Haydock 2004). More problematically, as discussed in the next section, such helpers in at least some cases attempt to breed outside of their home group, contributing to high levels of extra-group matings. To the extent that either of these occur, excluding such helpers from consideration of reproductive skew becomes all that much more subjective. Equally disturbingly, by excluding the occurrence of incest and extra-group matings, the range of behaviors that reproductive-skew theory is left to address narrows, leaving only a small fraction of the diversity of avian mating systems within its purvey. How much of a difference does this make? Jamieson (1997) calculated skew using the Pamilo & Crozier (1996) index for multi-male and multi-female
233
234
W. D. Koenig et al. Table 9.1 Reproductive skew (mean ± SE [n broods]) of 128 broods (n offspring > 1) within multi-male and multi-female groups of acorn woodpeckers at Hastings Reservation for which parentage data were determined (Dickinson et al. 1995, Haydock et al. 2001) using the B index of Nonacs (2000, 2003). Category of individuals
Males
Females
Excluding non-breeding helpers
0.215 ± 0.025 [85]
0.034 ± 0.036 [32]
Including non-breeding helpers
0.298 ± 0.022 [106]
0.221 ± 0.030 [69]
groups of pukeko (Porphyrio porphyrio) at one of his study sites two ways and found a difference of a factor of 2.5 between the value calculated for females including helpers (0.42) and that excluding helpers (0.17). As a second example, we have calculated skew for acorn woodpeckers using the B index (Nonacs 2000, 2003) both including and excluding non-breeding helpers (Table 9.1). For males, skew increases a modest 39% if non-breeding helpers are included in the calculations. However, for females, skew increases over six-fold from near 0 to 0.221, concurrently increasing from 16% to 74% of the comparable value for males. Thus, both the magnitude and to some extent the pattern of reproductive skew are dependent on the assumptions being made about who is and who is not a potential breeder in the population. Extra-group parentage
There is considerable variation within cooperatively breeding birds in the frequency of extra-group reproduction, including both extra-group matings and conspecific brood parasitism (see Tables 9.3 and 9.4, below). Among 15 studies of female co-breeding species, for example, conspecific brood parasitism is known to occur in at least six, while at least four (brown jay, Cyanocorax morio; white-browed scrubwren, Sericornis frontalis; Taiwan yuhina, Yuhina brunneiceps [Figure 9.2], and Seychelles warbler, Acrocephalus sechellensis) are known to exhibit a significant (10%) incidence of extra-group matings. In the extreme, females may mate preferentially or exclusively with extra-group males, in which case skew theory is useless in predicting partitioning of reproduction among males within groups. This approaches what has in fact been found in the cooperatively breeding superb fairy-wren (Malurus cyaneus: Mulder et al. 1994, Dunn et al. 1995). Extra-group parentage clearly affects reproductive partitioning within groups, and thus reproductive skew. However, skew theory, designed thus far to address reproductive sharing among co-breeders within groups, has yet to be extended to integrate the potential for reproductive “sharing” by extra-pair
Reproductive skew in avian societies
Figure 9.2 A group of four yuhinas, a small species of joint-nesting babbler endemic to Taiwan, where it is found in a wide range of forest habitats. Yuhina social groups consist of 1–3 monogamous pairs that are generally non-relatives and share the same nest along with an occasional unpaired individual. Skew is low among both males and females, and egg destruction is rarely observed. How yuhinas resolve breeding conflict and achieve such seemingly peaceful cooperation with non-relatives remains to be answered. Photo ª 2007 by Jinn-Juang Hsu.
or extra-group males. Empirically, no current index of skew takes into account the loss of parentage breeders suffer within a group due to extra-group reproduction, and there is no obvious way to apply skew indices to situations in which a resident male, of which there may be only one, shares paternity with one or more extra-group males with which he may or may not interact. Theoretically, this situation clearly illustrates that monopolization of reproduction by a resident male may require control over more individuals than one’s co-breeders. Rather, such males must exert control over their female mate, or alternatively over all potential sires in the population attempting to obtain – by force in some cases – extra-pair matings with that female. Within socially monogamous species, variation in extra-pair matings appears to be related to a variety of factors, including breeding density, genetic variation, and sexual conflict (Petrie & Kempenaers 1998), and it is likely that similar factors are also important in influencing the frequency of extra-group matings in cooperative breeders. Skew models have yet to be extended to
235
236
W. D. Koenig et al. encompass this intriguing phenomenon, although work by Shellman-Reeve & Reeve (2000) applying a transactional model to the problem of extra-pair matings in socially monogamous species suggests that skew models could readily be developed to address this issue in the more complex cooperative breeding arena as well. Sexual conflict
Extra-group matings are a specific example of the more general omission of sexual conflict, defined as individuals of one sex imposing reproductive costs on the other for the benefit of the imposing sex, in most skew models. Specifically, it is evident that behavior on the part of one sex can often subvert the ability of the dominant member of the opposite sex to control reproduction of subordinates, thus violating this key assumption of skew theory. For example, females may actively choose to mate with more than one male to ensure that their eggs are fertilized, to increase genetic diversity in their offspring, or to obscure paternity and thereby increase the likelihood of obtaining parental care from more than one male (e.g. Stacey 1982). If females successfully seek matings from more than one male, or different males, within the social group, then male reproduction may end up being more egalitarian than predicted by skew theory alone. Alternatively, even when reproductive skew among males within the group is high due to female choice, if females seek matings from males outside the social group and prefer different extragroup mates than those preferred as within-group mates, variance in reproductive success among males in the population may be low (Webster et al. 1995). Among cooperatively breeding birds, the classic example of sexual conflict is the dunnock, in which Davies (1990, 1992) has shown that the fitness consequences of mate sharing are positive for females that are able to monopolize the parental effort of multiple males, but not for the males themselves. Conversely, having access to plural-nesting females allows male dunnocks to achieve higher fitness but reduces the reproductive success of the females themselves. Thus, the mating system, including how reproduction is shared within groups, appears to be mainly driven by sexual conflict rather than factors considered by skew models. Even more generally, skew models, to the extent that they focus exclusively on within-sex competition, ignore any means by which the opposite sex may affect the pattern of reproductive sharing among potential co-breeders within a group. Given the importance of such behaviors in so many systems (Andersson 1994), this is clearly a serious omission. Thus far, the only attempt at extending a transactional skew framework to include sexual conflict and the possibility of female control over the
Reproductive skew in avian societies distribution of paternity is that of Cant & Reeve (2002). Their results demonstrate that incorporating this basic aspect of many vertebrate societies yields contrasting predictions about patterns of reproductive skew depending on the assumptions made about the relationship between paternity, paternal care, and the options available to subordinate males. It is likely that some level of female control is at work in many, if not all, avian cooperative breeders, and thus that skew models dealing with such societies will need to consider the behavior of both sexes in order to successfully predict patterns of reproductive partitioning. Group size and sociality
Several papers have extended transactional models to groups of more than two co-breeding individuals (Johnstone et al. 1999, Reeve & Emlen 2000, Reeve & Jeanne 2003). What has yet to be addressed, however, is variation in group size and in the proportion of the population living in groups. This omission is paradoxical, given the explicit focus of skew theory on whether subordinates should remain in or leave groups. In acorn woodpeckers at Hastings Reservation, California, USA, for example, only half of groups consist of two or more co-breeder males and only 23% consist of joint-nesting females (usually two, but rarely three). Indeed, the most common breeding group composition is a breeding pair of individuals (plus a variable number of nonbreeding helpers), which make up 42% of all breeding groups. Among social birds in general, variation in the frequency of breeding coalitions varies considerably, from <10% among female oystercatchers (Haematopus ostralegus) and Arabian babblers (Turdoides squamiceps) to 90% in Taiwan yuhina (see Table 9.3, below). Can skew theory provide guidelines to assist in explaining the frequency of co-breeding groups in the population? Or, even more ambitiously, can it help explain differences in group size across populations within and among species? Thus far, the answer is no. Transactional skew theories are predicated on the assumption that individuals are able to choose to leave and breed independently if they do not receive sufficient fitness benefits from remaining in a group. In acorn woodpeckers, adult breeders do frequently leave reproductive positions for other opportunities (Haydock & Koenig, unpublished data), and it is likely that such fluidity exists in other systems also. However, current skew models do not predict the frequency with which individuals should share mates or variation in group size within or among species. This is clearly another area in which future development of skew theory might be particularly useful. As an example involving the concordance between group formation and relatedness, consider male coalitions of wild turkeys. Considered a textbook example of kin selection (Watts & Stokes 1971), recent molecular evidence has
237
238
W. D. Koenig et al. confirmed that males are closely related, and only the dominant male reproduces (Krakauer 2005a). Coalitions fathered offspring at approximately seven times the level of solitary males, and the indirect fitness benefits for subordinate members more than made up for the loss of independent breeding opportunities. Such high skew among co-breeding relatives is generally consistent with predictions of optimal skew theory. Conversely, skew theory predicts greater reproductive sharing among male coalitions of non-relatives, for which there should be strong selection given the immense reproductive advantage of coalitions compared with solitary males. Such coalitions of unrelated males have not been observed. Optimal skew theory, which assumes that conflict is avoided because of complete dominance, offers little to resolve this discrepancy. Instead, it may be explainable if coalitions of non-relatives resolve conflict in a completely different way than do related males, perhaps by means of a costly “tug-of-war” that ultimately renders cooperation unprofitable. Unfortunately, in lieu of unrelated coalitions to observe or the possibility of appropriate field experiments, this hypothesis, entailing the intriguing possibility that different skew models apply in different situations within the same population, remains speculative. Null models and relatedness asymmetries
Finally, there are several technical issues related to measuring reproductive skew in birds that deserve comment. It is important to keep in mind that although the basis of most avian societies is a set of closely related individuals, such societies still can differ in ways that are critical in terms of how they affect the predictions of the various skew models. For example, when cobreeders are siblings, relatedness is symmetrical: that is, each individual is equally closely related to the other’s offspring. In contrast, when co-breeders are parents and offspring, relatedness is potentially asymmetrical, with the offspring being more closely related to the parent’s subsequent young than the parent is to his or her grandchildren. Such differences have important consequences to predictions of reproductive skew in all models that are sensitive to relatedness (Reeve & Keller 1996). Specifically, when relatedness of cobreeders is asymmetrical, offspring produced by the dominant are typically equivalent, from the subordinate’s perspective, to the subordinate’s own offspring, and thus skew is generally expected to be greater than if relatedness is symmetrical. Unfortunately, asymmetrical relatedness only occurs if the parent mates with a relative (most often, presumably, the other parent) of the co-breeding offspring while the co-breeding offspring mates with a different, unrelated individual. It does not arise in the situation, such as occurs in acorn woodpeckers, in which
Reproductive skew in avian societies offspring only co-breed after all breeders of the opposite sex have disappeared and been replaced by non-relatives. Thus, asymmetrical relatedness is only likely to be an issue in very complex societies in which related and unrelated individuals of the same sex potentially co-breed, such as white-winged choughs (Corcorax melanorhamphos: Heinsohn et al. 2000). A second important issue, given the small sample sizes generally available from vertebrate societies, is the need to quantify skew using an index that is not sensitive to factors such as group size or differences in productivity that may differ between the groups being compared in a way that will obscure the pattern of interest (Nonacs 2000, 2003). Related to this problem is the need to test values statistically against an appropriate random model (Nonacs 2003). Such models can be modified to deal with quite complicated situations, such as when both productivity varies among groups and there is variability among group size. This latter difficulty is particularly problematical, since parentage cannot necessarily be divided exactly the same way among different numbers of co-breeders (Haydock & Koenig 2003). Unfortunately, using an appropriate skew index and testing it against an apparently valid null model does not guarantee that the results will be meaningful. There are almost always going to be multiple ways to calculate skew, even using the same index. In many cases it will be possible to calculate skew using different categories of group members (Jamieson 1997). Often it will be possible to calculate skew on a variety of time scales, including a broodby-brood basis, a seasonal basis, or (at least theoretically) a lifetime basis, differences that can lead to very different conclusions. Moreover, skew can be calculated for differing subsets of the population or across the population as a whole, although an index encompassing all groups, whether they involve cobreeders or not, clearly measures something other than reproductive sharing as usually envisioned in skew theory. A related issue that is becoming more important as molecular techniques continue to advance is to keep in mind that knowledge of outcomes – specifically of parentage – does not necessarily provide insight as to mechanisms and thus may complement but usually not replace behavioral observations. Indeed, by failing to provide any hints concerning potential mechanisms, a purely molecular approach can be distinctly misleading. As an example, consider again co-breeding male acorn woodpeckers. When measured within broods, such coalitions exhibit significant reproductive skew, with a single male frequently gaining paternity of all young. However, across broods there is often switching of paternity, leading to considerably lower skew when multiple nesting attempts are combined (Haydock & Koenig 2002, 2003). Such switching, which has also been observed in brown jays (Williams
239
240
W. D. Koenig et al. 2004) and to some extent in Gala´pagos hawks (Buteo galapagoensis: Faaborg et al. 1995), raises the possibility that, contrary to what is usually assumed to be the case, paternity may not be independently determined for each egg within broods. Indeed, for acorn woodpeckers, an excellent fit to the observed level of reproductive skew can be obtained when nesting attempts for the same set of males are combined and it is assumed that paternity is a winner-take-all affair 72% of the time, the observed probability of single-sired broods (Haydock & Koenig 2002). Thus the observed skew level can simply be attributed to nonindependence of paternity. Further behavioral observations and physiological studies on sperm competition may help to resolve how molecular results should be interpreted. To the extent that such non-independence of paternity is the case, the observed outcome of high skew within broods is a red herring, suggesting that there is a strongly unequal probability of paternity among co-breeder males (that is, high skew) when in fact the chances of co-breeders fathering any particular brood (and thus any particular offspring) may actually be equal. The take-home message is that in order to determine the appropriate null model and thus to interpret a particular pattern of skew, it may often be necessary to understand the mechanisms by which the pattern is obtained.
Tests of skew theory in birds We now turn to empirical studies of skew in birds in order to examine the extent to which current theory successfully predicts patterns of skew within and among populations of birds. In general, we focus on the predictions of optimal skew or concessions theory, since this is the only framework that has thus far been tested in birds to any significant extent. However, it is important to keep in mind that the predictions of optimal skew theory are in some cases opposite to those of other skew models (e.g. the restraint or the tugof-war model), and thus that support or non-support for this particular model does not constitute a test of skew theory in general. With this in mind, we first consider intraspecific tests and then present a meta-analysis across species using published data on reproductive skew among birds in which co-breeding is known to occur. Intraspecific tests
We are aware of 12 intraspecific tests of concession skew theory involving seven different species (Table 9.2). Tests involved a wide range of comparisons, from different populations to differences among years or the sexes. All three major ecological and social factors encompassed by skew
Reproductive skew in avian societies theory have been tested, although there appears to be no study that claims to compare differences in the benefits of group living independent of ecological constraints. Two of the tests focused on the frequency and apparent strength of competitive interactions rather than differences in reproductive skew. All studies entail one or more caveats limiting the extent to which the comparison is a strong test of skew theory (Table 9.2). We discuss each briefly. Pukeko (Porphyrio porphyrio)
Jamieson (1997) contrasted skew in two populations of this polygynandrous species, one of which involved closely related co-breeders apparently living under conditions of high constraints and the other of which apparently involved unrelated co-breeders living under less ecologically restricted conditions. Comparisons were made in the degree of reproductive skew and the amount of dominance-related aggression, both of which were predicted to be higher in social groups composed of close kin in which ecological constraints were high. As predicted, skew was higher in the population living under higher ecological constraints in which co-breeders were genetic relatives. There was also some indication that dominance interactions were more prevalent and intense in the high-constraint, high-relatedness population. There are several caveats to these results. With respect to the dominance interactions, the differences were not entirely clear-cut and, as pointed out by Jamieson (1997), the assertion of dominance could by itself lead to high skew rather than being a consequence of high skew. As for the calculated skew values, differences between the populations were not significantly different for males, values for the high-skew population were estimated from a nearby study, and the index used (Pamilo & Crozier 1996) is influenced by group size and productivity (Nonacs 2000, 2003). Consequently, the observed differences in skew between the populations may to some extent be artifactual rather than due to differences in ecological constraints or relatedness of cobreeders. Common moorhen (Gallinula chloropus)
McRae (1996a) found that the proportion of eggs laid by dominant females in joint nests of this species was significantly higher when co-breeders were first-order relatives compared to when they were not known to be relatives, in accordance with optimal skew theory. However, as she points out, the sample size for apparently unrelated females was small (n ¼ 3, one of which was an adopted daughter) and the result is potentially confounded by age effects, which were highly significant in the population.
241
242
Table 9.2 Summary of intraspecific tests of concessions theory in social birds Results consistent with Species
Comparison
Difference
Prediction
prediction?
Comments
References
Pukeko (Porphyrio
Between
Constraints;
Greater skew in
Yes
Difference not
Jamieson 1997
porphyrio)
populations
relatedness
population with higher
significant for males;
constraints and related
skew not tested against
breeders
random models; data from one population inferred from prior nearby study.
Pukeko
Dominance
Constraints;
Greater dominance in
interactions
relatedness
high-skew societies
quantitative data;
with high constraints
possible circularity
Yes
No unambiguous
Jamieson 1997
and close relatedness Common moorhen
Female coalitions
Relatedness
(Gallinula chloropus)
Greater skew among
Yes
related cobreeders
Small sample size;
McRae 1996b
possibly confounded by age effects
Acorn woodpecker (Melanerpes formicivorus)
Between sexes
Constraints;
Greater skew in sex
Yes
No skew among
Haydock &
benefits of group
with higher constraints
females; male skew
Koenig 2002,
living
and greater group
depends on
2003
benefits
assumption of paternity independence
243
Acorn woodpecker
Male coalitions
Relatedness
Greater skew among
No
related cobreeders
Difference in
Haydock &
relatedness probably
Koenig 2003
small; small sample size Acorn woodpecker
Male coalitions
Constraints
Greater constraints and No
Male skew depends on Haydock &
(coalition size)
greater skew in smaller
assumption of
coalitions
paternity
Koenig 2003
independence Acorn woodpecker
Among years
Constraints
Greater skew in years
Partially
No significant
Haydock &
(reproductive
with fewer vacancies
relationship when
Koenig 2003
vacancies)
and higher constraints
relative, rather than absolute, number of vacancies used
Brown jay (Cyanocorax
Between sexes
morio)
Constraints;
Greater skew in sex
relatedness
with higher constraints
Relatedness
Greater competition
Yes
Female competitive interactions
Williams 2004
nests/years may change
and higher relatedness Brown jay
Switching between observed pattern
No
among closely related
Relatedness based on
Williams 2004
band sharing
cobreeders White-winged chough
Coalition structure
(Corcorax
Constraints;
Greater skew during
Complex social
Heinsohn et al.
relatedness
period of higher
structure consisting of
2000
constraints and higher
mix of related and
melanorhamphos)
Yes
relatedness Arabian babbler (Turdoides squamiceps)
Male coalitions
Relatedness
Greater skew among related cobreeders
unrelated coalitions Yes
Relatedness based on
Lundy et al.
band sharing; possibly
1998
confounded by group stability
244 Table 9.2 (cont.) Results consistent with Species
Comparison
Difference
Prediction
prediction?
Comments
White-browed
Male coalitions
Relatedness
Greater skew among
Yes
Possible confound with Whittingham
scrubwren (Sericornis frontalis)
related cobreeders
extra-group paternity
References
et al. 1997
Reproductive skew in avian societies Acorn woodpecker (Melanerpes formicivorus)
Polygynandry occurs regularly in this species in conjunction with helping at the nest by non-breeders. Based on comparative demography, Haydock & Koenig (2002, 2003) found that males appeared to be living under conditions of greater ecological constraints and were able to gain greater fitness benefits by living in groups than females. Optimal skew theory predicts that these differences should result in greater skew among co-breeder males, and this was indeed found to be the case. Although the difference was in accordance with skew theory, other findings were not. Females, for example, exhibited no skew whatsoever, sharing reproduction more equally than expected by chance. This pattern is enforced by egg destruction (Mumme et al. 1983, Koenig et al. 1995), a phenomenon eliminating any opportunity for females to control reproduction. Meanwhile, male skew is high within broods but is unrelated to male age or physical characteristics and often switches between broods. This suggests the possibility that paternity of eggs may not be independently determined, as is normally assumed, and that co-breeder males may actually have equal chances of siring eggs despite the high skew found in individual nests. Haydock & Koenig (2003) tested several other predictions of optimal skew theory as well. Among males there was no difference in skew in coalitions that were close relatives (r ¼ 0.5) versus coalitions where males were less closely related, contrary to expectations. There also tended to be an increase, rather than the expected decrease, in skew with increasing coalition size, which was assumed to correlate with lower ecological constraints since larger coalitions are known to be more competitive for reproductive vacancies and thus should provide more opportunities for leaving the group. The last test involved comparing reproductive skew in different years, indexing the degree of ecological constraints by counting the number of male reproductive vacancies occurring within the population during the three months prior to each nesting attempt. Skew was higher when there were fewer vacancies, in accord with skew theory. However, when dividing the mean number of vacancies by the number of individuals potentially competing for those vacancies, the results were no longer significant, thus providing at best only partial support for skew theory. Brown jay (Cyanocorax morio)
Females in this tropical jay can apparently be either joint or plural nesting, although the latter was not observed in the parentage study by Williams (2004). Williams compared skew in the two sexes, predicting that it should be greater among females, for which dispersal is apparently more
245
246
W. D. Koenig et al. constrained and relatedness was higher based on band-sharing. As expected from concession theory, skew was significantly higher among females. Williams (2004) also compared whether he observed aggressive interactions between joint-nesting females and found that females that have higher band-sharing scores, and thus were presumably more closely related, were less, rather than more, likely to engage in nesting aggression. White-winged chough (Corcorax melanorhamphos)
Heinsohn et al. (2000) documented the amalgamation of coalitions of relatives that occurred in this species following a severe drought and reported that among such groups males were more likely to share reproduction than females and that skew was higher prior to the drought, when groups were composed exclusively of relatives, than afterwards, when they were composed of related and unrelated birds. This latter finding supports the expected pattern of higher skew when co-breeders are close relatives. Arabian babbler (Turdoides squamiceps)
In a genetic analysis of this species, Lundy et al. (1998) found significantly higher band-sharing between males that did not share paternity compared to males that did, supporting the prediction of skew theory that there should be higher skew among more closely related individuals. A possible confound in this study was group stability, since shared paternity was more likely to occur following major changes in group structure, which were also more likely to involve non-relatives. White-browed scrubwren (Sericornis frontalis)
Similar to the Arabian babbler, Whittingham et al. (1997) found that males in this species were more likely to share paternity among non-relatives, a result they carefully disentangled from incest avoidance. However, this population also exhibited a fairly high incidence of extra-group paternity, which was more frequent in groups where the subordinate male was unrelated to the dominant. As a result, although subordinate males gained more parentage when unrelated to the dominant, the overall share of reproductive monopolization by the dominant did not differ appreciably between groups in which males were related (17 of 26 young; 66% paternity) compared to groups in which males were unrelated (12 of 24 young; 50% paternity; Fisher exact test p ¼ 0.21). To conclude, of the 12 intraspecific tests summarized in Table 9.2, the results of nine (75%) are fully or partially consistent with optimal skew theory. Of the seven involving differences in ecological constraints, six (86%) are consistent with skew theory at least in part, while only three of five (60%) tests
247
Table 9.3 Summary of the life-history and behavioral correlates of skew in female social birds. “Joint” species are those in which females nest communally; “plural” species are those in which females produce separate nests within a larger social group. CBP, conspecific brood parasitism. Mean
Extra-
Frequency
coalition
Relatedness
Ecological
Benefits of
Dominance
Competition group
Species
of coalitions
size
of coalitions
constraints
coalitions
obvious?
obvious?
matings?
Magpie goose
“Common”
2.0 (joint)
High
?
Low
Yes
Skew?
References
Yes (egg
Low (?) (but Low
Frith and
(Anseranas
tossing
CBP
Davies 1961,
semipalmata)
likely)
occurs?)
Horn et al.
Egg tossing
Low (but
possible
CBP
1986, McRae
common)
1996a, 1996b
1996 Common
21%
2.1 (joint)
High
High
Moderate
No
moorhen (Gallinula
Low
Gibbons
chloropus) Pukeko
26%
2.0 (joint)
Low
Low
Low
Yes
No
(Porphyrio
No (but
Low
CBP occurs)
Jamieson et al. 1994,
porphyrio)
Lambert
(Otokia)
et al. 1994, Jamieson 1997
Oystercatcher
1%
2.0 (joint)
Low
High
Low
No
No
?
Low
(Haematopus
Heg & van Treuren 1998
ostralegus) Acorn woodpecker (Melanerpes formicivorus)
23%
2.1 (joint)
High
High
Low
No
Yes (egg tossing)
No
Low/none
Haydock &
(switching) Koenig 2002, 2003
248 Table 9.3 (cont.) Mean
Extra-
Frequency
coalition
Relatedness
Species
of coalitions
size
of coalitions
Guira cuckoo
94%
3.1 (joint)
Low
Ecological
Benefits of
Dominance
Competition group
constraints
coalitions
obvious?
obvious?
matings?
Skew?
References
Low
Moderate (?) No
Yes (egg
No (but
Low/none
Macedo 1992,
tossing;
CBP occurs) (switching) Macedo et al.
(Guira guira)
infanticide) Groove-billed
30%
2.7 (joint)
Low
High
Moderate
Yes
ani (Crotophaga
Yes
2004, 2005 No (but
Low
(egg-tossing) CBP occurs)
Vehrencamp 1977, Koford
sulcirostris)
et al. 1986, 1990, Vehrencamp & Quinn 2004
Dunnock
38%
2.2 (plural) Low
Low
Low
No
No
Low
Low
(Prunella
Davies 1990, 1992
modularis) Brown jay
2.1 (plural; Moderate
Moderate to Low
(Cyanocorax
29%
sometimes (62% related)
low
morio)
joint)
Yes
Yes (some
17%
Low
egg tossing)
Lawton & Lawton 1985, Williams et al. 1994, Williams 2004
Taiwan yuhina 90% (Yuhina brunneiceps)
2.3 (joint)
Low
Low
High
Yes
Yes (egg tossing)
Low
Low
Yuan et al. 2004, S.-F. Shen unpubl. data
249
Pukeko
79%
2.0 (joint)
High
High
Low
Yes
No
(Shakespear)
No (but
Moderate
CBP occurs)
Jamieson et al. 1994, Lambert et al. 1994, Jamieson 1997
Tasmanian
42%
2.0 (joint)
High
High
native hen
Moderate to No
No
No
low
Moderate
Goldizen et al.
(> males)
1998, 2000
(Gallinula mortierii) Seychelles
30%
2.0 (joint)
High
High
Present on
warbler
high, but
(Acrocephalus
not low
sechellensis)
quality
Yes
No
40%
Moderate
Komdeur 2005
territories White-winged
32%
2.5 (joint)
chough
Moderate
High
(50% related)
(foraging
destruction;
Heinsohn et
constraints)
egg tossing)
al. 2000
(Corcorax
High
Yes
Yes (nest
Low (?)
High
Rowley 1978,
melanorhamphos) Arabian babbler (Jurdoides squamiceps)
9%
2.0 (joint)
High
High
Moderate to low
Yes
No (?)
No
High
Zahavi 1990, Lundy et al. 1998
250
Table 9.4 Summary of the life-history and behavioral correlates of skew in male social birds. The last two species exhibit cooperative courtship; all other species are either cooperatively polyandrous, in which males share one or more females (“unpaired”), or species in which male–female pairs form breeding coalitions (“paired”). Freq. of
Coalition
Relatedness
Ecological
Benefits of
Dominance
Species
coalitions
size
of coalitions
constraints
coalitions
obvious?
matings
Skew?
References
Gala´pagos hawk
68%
2.5
Low
High (?)
Low
No
Low
Low
Faaborg &
(switching)
Bednarz 1990,
(Buteo galapagoensis)
Extra-group
(unpaired)
Faaborg et al. 1995 Groove-billed ani
30%
2.7 (paired)
Low
High
Moderate
Yes
Low
Low
Vehrencamp
(Crotophaga
1977, Koford
sulcirostris)
et al. 1986, 1990, Vehrencamp & Quinn 2004
Taiwan yuhina
90%
2.5 (paired)
Low
Low
High
Yes
10%
Low
(Yuhina brunneiceps)
Yuan et al. 2004, S.-F. Shen unpubl. data
Dunnock (Prunella
64%
2.0
25%
2.4
modularis) Tasmanian native hen (Gallinula mortierii)
Low
Low
Low
No
Low
Low
High
High
Moderate
No
No
Moderate
Goldizen et al.
to low (<
1998,
females)
Goldizen et al.
(unpaired) (unpaired)
Davies 1990, 1992
to low
2000
251 White-winged
79%
chough (Corcorax
3.5 (partly
Moderate to
High
unpaired)
low (23% of
(foraging
pairs related)
constraints)
melanorhamphos) White-browed
31%
scrubwren
High
2.2
Moderate (50% Moderate (?) Low
(unpaired)
of coalitions
(Sericornis frontalis)
Yes
Yes
Low (?)
12%
Moderate to
Heinsohn
low
et al. 2000
Moderate to
Whittingham
low
et al. 1997,
related)
R. Magrath pers. comm.
Pukeko (Porphyrio
46%
porphyrio) (Otokia)
2.3
Low
Low
Low
(unpaired)
Yes (but
No
Moderate
Jamieson et al.
unrelated to
1994, Lambert
paternity)
et al. 1994, Jamieson 1997
Acorn woodpecker
49%
(Melanerpes
2.6
High
High
Moderate
No
No
(unpaired)
Moderate (?) Haydock & (switching)
Koenig 2002,
Moderate
Lawton &
(switching;
Lawton 1985,
> females)
Williams et al.
formicivorus) Brown jay
2003 High
? (unpaired)
Low
Low
Yes
No
17%
(Cyanocorax morio)
1994, Williams 2004 Brown skua
33%
(Catharacta
2.2
Low
Moderate (?) Low
No
No
(unpaired)
Moderate
Millar et al.
(switching)
1994
Moderate
Jamieson et al.
to high
1994, Lambert
antarctica) Pukeko (Shakespear)
100%
3.2 (unpaired)
High
High
Low
Yes
No
et al. 1994, Jamieson 1997
252 Table 9.4 (cont.) Freq. of
Coalition
Relatedness
Ecological
Benefits of
Dominance
Species
coalitions
size
of coalitions
constraints
coalitions
obvious?
matings
Skew?
References
Arabian babbler
34%
2.0
High
High
Moderate to
Yes
No
High (but
Zahavi 1990,
< females)
Lundy et al.
High
MacDonald
(Turdoides
(unpaired)
Extra-group
low
squamiceps) Long-tailed
1998 100%
manakin
2.0 (þ 7.1
Low
?
High
Yes
No
auxilliaries)
1989,
(Chiroxiphia linearis)
MacDonald & Potts 1994
Wild turkey (Meleagris gallapavo)
33%
2.4
High
?
High
Yes
No
High
Krakauer 2005a, 2005b
Reproductive skew in avian societies involving differences in relatedness were in the direction predicted. There are several reasons why these results provide at best only weak support of optimal skew theory, including that the predictions are qualitative, in some cases relatively superficial, may be biased toward positive findings, and generally fail to consider alternative models. Nonetheless, they offer at least some encouragement that optimal skew theory may be useful in understanding, if not predicting, patterns of reproductive sharing within some avian populations. Clearly additional work remains to be done. Recent advances in quantifying and statistically testing skew, and in molecular techniques allowing estimation of relatedness, should help future workers obtain more accurate and less biased estimates of these variables. Better knowledge of mechanisms will help resolve questions concerning the significance of paternity switching between nests and whether it is always appropriate to assume that paternity of eggs is determined independently of other eggs within the same brood. These advances, combined with increased sample sizes and additional studies, should ultimately provide the opportunity for considerably stronger intraspecific tests of optimal skew theory than have been conducted thus far. Interspecific tests: a meta-analysis
Based on the literature, we summarize the cogent information currently available on reproductive skew for different populations of birds, dividing the data by sex (Tables 9.3 and 9.4). Each species is listed once, except for the two populations of Pukeko studied by Jamieson (1997), which were considered separately due to the contrasting conditions under which they appeared to be living. In each case, we estimated, based on the data provided, the frequency of co-breeding coalitions and the mean size of coalitions (i.e. the mean number of individuals in coalitions excluding singletons). We then went on to categorize (usually as being “high,” “moderate,” or “low”) the overall level of reproductive skew exhibited by the population along with relatedness of coalitions, the benefits of reproductive coalitions, and the relative strength of ecological constraints to independent reproduction within the system. These latter three variables are widely acknowledged to be key ecological factors in the evolution of cooperative breeding, although their precise role and relative importance in particular systems is the subject of considerable debate (e.g. Dickinson & Hatchwell 2004, Ekman et al. 2004). We also noted whether dominance was obvious or not, whether (for females) competition among co-breeders was evident, whether extra-group matings in the form of conspecific brood parasitism and/or extra-group paternity occurs, whether males and females are generally paired or unpaired, and (for males) whether male parental care occurs or not. Male parental care is found in the majority of
253
254
W. D. Koenig et al. Table 9.5 Relationships between life-history parameters and coalition frequency, coalition size, and reproductive skew across 15 populations of co-breeding female birds (Table 9.3) and 15 populations of co-breeding male birds (Table 9.4). Not all parameters were available for all species. Tests are Spearman rank correlations (rs listed) or Mann–Whitney U tests (z value listed; ‘ þ ’ larger coalitions or greater skew with increases in or greater presence of the variable; ‘ ’ smaller coalitions or less skew with increases in or greater presence of the variable). þ ¼ p 0.10; * ¼ p 0.05; ** ¼ p 0.01). Coalition frequency Males Coalition frequency
–
Females –
Mean coalition size
Reproductive skew
Males
Females
Males
Females
0.09
0.44
0.01
0.06
0.06
0.31
(rs) Coalition size (rs) Genetic relatedness
0.09
0.44
0.20
0.21
0.32
–
0.53*
–
0.03
0.43
0.73**
0.36
0.16
0.34
0.35
0.67
0.26
0.61
0.51
–
0.44þ
0.51*
(rs) Ecological
0.11
0.53*
0.23
0.44þ
1.03
1.13
constraints (rs) Benefits of group
0.44þ
living (rs) Dominance
0.50
(z value) Competition
–
2.57**
–
1.52
(z value) Extra-group mating
0.00
0.32
0.74
0.12
0.89
1.04
0.39
1.09
1.58
2.29*
1.92þ
0.26
0.55
–
0.28
–
2.26*
–
(z value) Paired vs. unpaired (z value) Paternal care (z value)
species (which are either cooperatively polyandrous or in which pairs of birds form cobreeding coalitions), but not in the last two species (long-tailed manakin, Chiroxiphia linearis, and wild turkey), in which males exhibit cooperative courtship. The results confirm the existence of considerable variability across each of these parameters. They also confirm some interesting differences between male and female mate-sharing. In particular, both coalition frequency and mean coalition size are significantly larger among males than females (Wilcoxon signed-ranks tests; n ¼ 9; both p < 0.05), supporting the hypothesis that
Reproductive skew in avian societies females have a harder time sharing breeding opportunities (although not necessarily sharing reproduction per se) than males. This difference is most likely related to the fact that female competition, often taking the form of egg destruction and infanticide, is often much more destructive than male competition, and thus co-breeding is often likely to be more beneficial for males than for females. To what extent do other interspecific patterns correspond to those predicted from optimal skew theory? We used Spearman rank correlations and Mann–Whitney U tests to look for associations between coalition frequency, mean coalition size, and reproductive skew and the various life-history variables summarized in Tables 9.3 and 9.4. Results (Table 9.5) yielded no significant correlations with coalition frequency. However, several significant (p 0.05) and near-significant (p 0.10) correlations emerged with the other two variables. For males, mean coalition size was larger in populations living under greater ecological constraints, whereas for females, mean coalition size was larger in populations where female co-breeders were less closely related, the benefits of group living were greater, competition between females was more evident in the form of egg tossing and infanticide, and birds tended to be paired rather than unpaired. Further, the skew analysis also yielded several promising results. Skew in both sexes tended to be higher when co-breeders were more closely related. In addition, for females, skew was higher in populations living under conditions of greater ecological constraints and higher benefits of group living. Skew tended to be greater among males when they were unpaired rather than paired to females, and among the two species that do not exhibit male parental care. Thus, all three predicted relationships between ecological and genetic factors and reproductive skew were upheld to some degree in females. Only one of the three predicted relationships was upheld in males, but this was the one between skew and genetic relatedness, which is the most counterintuitive relationship predicted by optimal skew theory (Keller & Reeve 1994). As discussed earlier, skew theory does not in general predict any explicit relationships about coalition size. Encouragingly, however, the results involving this variable are intuitively reasonable. That is, we expect mean coalition size to increase with greater ecological constraints, as found in males, and with greater benefits of group living, as found in females. It is also satisfying that overt competition among co-breeding females appears to be greater in populations with larger coalitions, and interesting that coalitions tend to be larger among species in which male–female pairs form, since this may offer at least the start of a hypothesis explaining perplexing variation in the pair-bonding behavior (or lack thereof) observed in different cooperative breeders.
255
256
W. D. Koenig et al. The finding that coalition size is inversely correlated with genetic relatedness is also intriguing, and may have to do with factors other than individual strategic decisions. For example, if potential coalition members are limited to only close relatives, such as apparently is the case in wild turkeys (Krakauer 2005a, 2005b), then the frequency of those relatives will depend upon clutch size and patterns of mortality, and coalitions in such species are likely to be smaller for purely demographic reasons rather than as any reflection of differences in optimal group size. These results offer support for the intriguing possibility that not only particular ecological and genetic factors may predict reproductive skew, but that skew is associated with specific behavioral traits across species. Greater female skew is found in species in which ecological constraints and the benefits of group living are relatively large and genetic relatedness relatively high. Such species tend to exhibit more obvious dominance but less overt competition than low-skew species. Among males, greater skew is associated with closer genetic relatedness, while high-skew species tend to be species in which birds are unpaired and males do not participate in raising young. The correlation between pair-bonding and lower skew suggests that pairing behavior within complex social groups containing multiple breeders counters monopolization by a single individual that might otherwise result in high skew, while the relationship between high skew and lack of male parental care is interesting due to its possible relationship to multiple mating by females (Krakauer 2005b). In the two species in which males do not provide paternal care, there is no potential phenotypic benefit to females mating with more males than necessary to fertilize their eggs, and skew tends to be high. In contrast, in the species where males provide paternal care, females could potentially benefit by mating with multiple males and decreasing male skew because of the relationship between paternal care and opportunity of parentage (Stacey 1982, Hartley & Davies 1994, Cant & Reeve 2002). The lower skew in these species is consistent with this possibility. We conclude that the predictions of optimal skew theory are, at least to some extent, upheld across populations with the currently available data. This level of support is more than we had originally expected. Again, however, these analyses are at best a preliminary attempt at examining skew theory at an interspecific level. More data, phylogenetically controlled analyses, and tests in which alternative models are explicitly pitted against each other will be necessary before it will be possible to conclude that there is strong support for optimal skew theory among birds. Nonetheless, they yield hope that despite its shortcomings, skew theory may eventually provide a valuable tool for understanding many aspects of avian social behavior.
Reproductive skew in avian societies Conclusion Given the logistic difficulties of applying skew theory, combined with the failure of skew theory as usually envisioned to address several key variables in avian social behavior, including the frequency of extra-group matings, variation in group size and sociality, and sexual conflict, it would seem a priori unlikely that skew theory can fulfill the goal of providing a framework for unifying the major features of sociality. Given this starting point, the summary of empirical results reviewed here provides a relatively optimistic outlook for the application of skew theory to an understanding of avian social behavior. Within social birds, numerous key questions remain and experimental studies have yet to be performed, yet a number of studies provide at least some comparative evidence supporting the possibility that reproductive skew can be predicted based on knowledge of ecological constraints, the relative benefits of group living, and genetic relatedness (Table 9.2). More impressively, across species there is evidence not only that these factors can predict reproductive sharing but that such sharing correlates, at least to some extent, with several of the major features of avian societies including the degree of competition between co-breeders, whether birds are paired or unpaired, whether males provide paternal care or not, and to a lesser extent even the extent and obviousness of dominance and reproductive competition. The fact that many of these same parameters correlate with mean coalition size is particularly exciting, since this suggests the possibility that the factors needed to predict reproductive skew may also provide insight into the degree of sociality expressed by different species. Unfortunately, the weakness of many of these relationships, combined with our inability to reject alternative hypotheses, confirms that we have a long way to go before a unifying theory of social evolution, even just for birds, can be achieved. The fundamental spirit of all skew models is to capture the means by which animals resolve their breeding conflicts. Thus far there are mainly two nonmutually exclusive approaches that have been proposed: transactional and compromise models. Transactional models describe the situation in which a controlling dominant may yield certain benefits to a subordinate in order to change the latter’s behavior to the former’s benefit. In contrast, in tug-of-war models breeding conflicts are resolved by costly competition between group members (see also chapter 1). Adding to this central idea, we envision several directions for future advances, both theoretically and empirically. First, it would be useful to broaden the scope of skew theory and relax some of its assumptions in order to apply it to a greater number of systems. One logical frontier for skew theory to encompass, for example, would be to
257
258
W. D. Koenig et al. explain reproductive partitioning in aggregations such as leks (Kokko & Lindstro¨m 1996). Given the large variation in lek size within species (Ho¨glund & Alatalo 1995), as well as recent findings of divergent patterns of male relatedness (Ho¨glund et al. 1999, Petrie et al. 1999, Shorey et al. 2000, Gibson et al. 2005), leks may eventually provide excellent systems for testing models of reproductive skew. Theoretically, a promising step in this direction is potentially the “bordered” tug-of-war model (Reeve & Shen 2006), which combines the option of independent reproduction and reproductive incentive with the potential for within-group conflict, which, as the authors point out, is in theory rendered pointless in optimal skew models as a consequence of the assumption of complete dominant control. Their synthesis model demonstrates that, despite their proliferation, the diversity of skew models can be understood within a general theoretical framework by allowing researchers to ask not only how conflict is resolved in a particular society, but also why a particular mechanism is used in a specific case, thereby moving significantly closer to the goal of understanding reproductive partitioning in a given situation. Most theoretical developments thus far have concentrated on extending transactional models, albeit in various realistic and thus potentially useful ways. We suggest that theoreticians should focus more on modeling alternative conflict-resolution mechanisms and clarifying the relationships between different models. For empiricists, at least two approaches are available. First is attempting to distinguish between different models by testing suites of predictions (e.g. Langer et al. 2004). Second is direct experimental testing of conflict-resolution mechanisms, such as have been performed in social wasps (Reeve & Nonacs 1992) and cooperatively breeding fishes (Heg et al. 2006). Such experiments clearly provide methods of testing the predictions of alternative reproductive-skew models. Although progress is being made, experimental investigation of reproductive skew remains lacking in avian systems. In addition, other key aspects of cooperative-breeding societies, including extra-group matings, sexual conflict influencing reproductive partitioning, and variability in group size and composition, are still largely left behind by current skew models. A truly unified theory of avian social behavior, let alone social evolution in general, will clearly require much further investigation.
Acknowledgments We thank Kern Reeve, Michael Taborsky, an anonymous reviewer, and the editors for their comments, and the Department of Neurobiology and
Reproductive skew in avian societies Behavior, Cornell University, for accommodating the senior author during the preparation of the manuscript. Our work on avian social behavior has been supported by the National Science Foundation.
References Andersson, M. (1994). Sexual Selection. Princeton, NJ: Princeton University Press. Bengtsson, B. O. (1978). Avoiding inbreeding: at what cost? Journal Of Theoretical Biology, 73, 439–444. Cant, M. A. & Johnstone, R. A. (1999). Costly young and reproductive skew in animal societies. Behavioral Ecology, 10, 178–184. Cant, M. A. & Reeve, H. K. (2002). Female control of the distribution of paternity in cooperative breeders. American Naturalist, 160, 602–611. Davies, N. B. (1990). Dunnocks: cooperation and conflict among males and females in a variable mating system. In P. B. Stacey & W. D. Koenig, eds., Cooperative Breeding in Birds: Long-term Studies of Ecology and Behavior. Cambridge: Cambridge University Press, pp. 455–485. Davies, N. B. (1992). Dunnock Behaviour and Social Evolution. Oxford: Oxford University Press. Dickinson, J. L. & Hatchwell, B. J. (2004). Fitness consequences of helping. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge, Cambridge University Press, pp. 48–66. Dickinson, J. L., Haydock, J., Koenig, W. D., Stanback, M. T., & Pitelka, F. A. (1995). Genetic monogamy in single-male groups of acorn woodpeckers, Melanerpes formicivorus. Molecular Ecology, 4, 765–769. Dunn, P. O., Cockburn, A., & Mulder, R. A. (1995). Fairy-wren helpers often care for young to which they are unrelated. Proceedings of the Royal Society of London B, 259, 339–343. Ekman, J., Dickinson, J. L., Hatchwell, B. J., & Griesser, M. (2004). Delayed dispersal. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge: Cambridge University Press, pp. 35–47. Emlen, S. T. (1996). Reproductive sharing in different types of kin associations. American Naturalist, 148, 756–763. Faaborg, J. & Bednarz, J. C. (1990). Gala´pagos and Harris’ hawks: divergent causes of sociality in two raptors. In P. B. Stacey & W. D. Koenig, eds., Cooperative Breeding in Birds: Long-term Studies of Ecology and Behavior. Cambridge: Cambridge University Press, pp. 359–383. Faaborg, J., Parker, P. G., Delay, L. et al. (1995). Confirmation of cooperative polyandry in the Gala´pagos hawk (Buteo galapagoensis). Behavioral Ecology and Sociobiology, 36, 83–90. Foster, M. S. (1977). Odd couples in manakins: a study of social organization and cooperative breeding in Chiroxiphia linearis. American Naturalist, 111, 845–853. Frith, H. J. & Davies, S. J. J. F. (1961). Ecology of the magpie goose Anseranas semipalmata. CSIRO Wildlife Research, 113, 555–557.
259
260
W. D. Koenig et al. Gibbons, D. W. (1986). Brood parasitism and cooperative breeding in the moorhen, Gallinula chloropus. Behavioral Ecology and Sociobiology, 19, 221–232. Gibson, R. M., Pires, D., Delaney, K. S., & Wayne, R. K. (2005). Microsatellite DNA analysis shows that greater sage grouse leks are not kin groups. Molecular Ecology, 14, 4453–4459. Goldizen, A. W., Putland, D. A., & Goldizen, A. R. (1998). Variable mating patterns in Tasmanian hens (Gallinula mortierii): correlates of reproductive success. Journal of Animal Ecology, 67, 307–317. Goldizen, A. W., Buchan, J. C., Putland, D. A., Goldizen, A. R., & Krebs, E. A. (2000). Patterns of mate-sharing in a population of Tasmanian native hens Gallinula mortierii. Ibis, 142, 40–47. Hamilton, W. D. (1964). The genetical evolution of social behaviour. I, II. Journal of Theoretical Biology, 7, 1–52. Hartley, I. R. & Davies, N. B. (1994). Limits to cooperative polyandry in birds. Proceedings of the Royal Society of London B, 257, 67–73. Haydock, J. & Koenig, W. D. (2002). Reproductive skew in the polygynandrous acorn woodpecker. Proceedings of the National Academy of Sciences of the USA, 99, 7178– 7183. Haydock, J. and Koenig, W. D. (2003). Patterns of reproductive skew in the polygynandrous acorn woodpecker. American Naturalist 162, 277–289. Haydock, J., Koenig, W. D., & Stanback, M. T. (2001). Shared parentage and incest avoidance in the cooperatively breeding acorn woodpecker. Molecular Ecology, 10, 1515–1525. Heg, D. & van Treuren, R. (1998). Female–female cooperation in polygynous oystercatchers. Nature, 391, 687–691. Heg, D., Bergmu¨ller, R., Bonfils, D., et al. (2006). Cichlids do not adjust reproductive skew to the availability of independent breeding options. Behavioral Ecology, 17, 419–429. Heinsohn, R., Dunne, P., Legge, S., & Double, M. (2000). Coalitions of relatives and reproductive skew in cooperatively breeding white-winged choughs. Proceedings of the Royal Society of London B, 267, 243–249. Ho¨glund, J. & Alatalo, R. V. (1995). Leks. Princeton, NJ: Princeton University Press. Ho¨glund, J., Alatalo, R. V., Lundberg, A., Rintamaki, P. K., & Lindell, J. (1999). Microsatellite markers reveal the potential for kin selection on black grouse leks. Proceedings of the Royal Society of London B, 266, 813–816. Horn, P. L., Rafalski, J. A., & Whitehead, P. J. (1996). Molecular genetic (RAPD) analysis of breeding magpie geese. Auk, 113, 552–557. Jamieson, I. G. (1997). Testing reproductive skew models in a communally breeding bird, the pukeko, Porphyrio porphyrio. Proceedings of the Royal Society of London B, 264, 335–340. Jamieson, I. G., Quinn, J. S., Rose, P. A., & White, B. N. (1994). Shared paternity among non-relatives is a result of an egalitarian mating system in a communally breeding bird, the pukeko. Proceedings of the Royal Society of London B, 257, 271–277.
Reproductive skew in avian societies Johnstone, R. A. (2000). Models of reproductive skew: a review and synthesis. Ethology, 106, 5–26. Johnstone, R. A., Woodroffe, R., Cant, M. A., & Wright, J. (1999). Reproductive skew in multimember groups. American Naturalist, 153, 315–331. Keller, L. & Reeve, H. K. (1994). Partitioning of reproduction in animal societies. Trends in Ecology and Evolution, 9, 98–102. Koenig, W. D. & Haydock, J. (2004). Incest and incest avoidance. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge: Cambridge University Press, pp. 142–156. Koenig, W. D. & Mumme, R. L. (1987). Population Ecology of the Cooperatively Breeding Acorn Woodpecker. Princeton, NJ: Princeton University Press. Koenig, W. D., Mumme, R. L., Stanback, M. T., & Pitelka, F. A. (1995). Patterns and consequences of egg destruction among joint-nesting acorn woodpeckers. Animal Behaviour, 50, 607–621. Koenig, W. D., Haydock, J., & Stanback, M. T. (1998). Reproductive roles in the cooperatively breeding acorn woodpecker: incest avoidance versus reproductive competition. American Naturalist 151, 243–255. Koford, R. L., Bowen, B. S., & Vehrencamp, S. L. (1986). Habitat saturation in groove– billed anis (Crotophaga sulcirostris). American Naturalist, 127, 317–337. Koford, R. R., Bowen, B. S., & Vehrencamp, S. L. (1990). Groove-billed anis: joint nesting in a tropical cuckoo. In P. B. Stacey & W. D. Koenig, eds., Cooperative Breeding in Birds: Long-term Studies of Ecology and Behavior. Cambridge: Cambridge University Press, pp. 333–356. Kokko, H. & Lindstro¨m, J. (1996). Kin selection and the evolution of leks: whose success do young males maximize? Proceedings of the Royal Society of London B, 263, 919–923. Komdeur, J. (2005). No evidence for adaptive suppression of joint laying by dominant female Seychelles warblers: an experimental study. Behaviour, 142, 1669–1684. Krakauer, A. H. (2005a). Kin selection and cooperative courtship in wild turkeys. Nature, 434, 69–72. Krakauer, A. H. (2005b). The evolution of cooperative male courtship: kin selection and the mating system of the wild turkey Meleagris gallopavo. Unpublished Ph.D. thesis, University of California, Berkeley. Lambert, D. M., Millar, C. D., Jack, K., Anderson, S., & Craig, J. L. (1994). Single- and multilocus DNA fingerprinting of communally breeding pukeko: do copulations or dominance ensure reproductive success? Proceedings of the National Academy of Sciences of the USA, 91, 9641–9645. Langer, P., Hogendoorn, K., & Keller, L. (2004). Tug-of-war over reproduction in a social bee. Nature, 428, 844–847. Lawton, M. F., & Lawton, R. O. (1985). The breeding biology of the brown jay in Monteverde, Costa Rica. Condor, 87, 192–204. Lundy, K. J., Parker, P. G., & Zahavi, A. (1998). Reproduction by subordinates in cooperatively breeding Arabian babblers is uncommon but predictable. Behavioral Ecology and Sociobiology, 43, 173–180.
261
262
W. D. Koenig et al. MacDonald, D. W. (1989). Correlates of male mating success in a lekking bird with male–male cooperation. Animal Behaviour, 37, 1007–1022. MacDonald, D. W. & Potts, W. K. (1994). Cooperative display and relatedness among males in a lek-mating bird. Science, 266, 1030–1032. Macedo, R. H. (1992). Reproductive patterns and social organization of the communal guira cuckoo (Guira guira) in central Brazil. Auk, 109, 786–799. Macedo, R. H., Cariello, M. O., Graves, J., & Schwabl, H. (2004). Reproductive partitioning in communally breeding guira cuckoos, Guira guira. Behavioral Ecology and Sociobiology, 55, 213–222. Macedo, R. H. F., Quinn, J. S., & Lima, M. R. (2005). Reproductive skew and individual strategies: infanticide or cooperation? Acta Ethologica, 8, 92–102. Magrath, R. D. & Heinsohn, R. G. (2000). Reproductive skew in birds: models, problems and prospects. Journal of Avian Biology, 31, 247–258. Magrath, R. D., Johnstone, R. A., & Heinsohn, R. G. (2004). Reproductive skew. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge: Cambridge University Press, pp. 157–176. McRae, S. B. (1996a). Family values: costs and benefits of communal nesting in the moorhen. Animal Behaviour, 52, 225–245. McRae, S. B. (1996b). Brood parasitism in the moorhen: brief encounters between parasites and hosts and the significance of an evening laying hour. Journal of Avian Biology, 27, 311–320. Millar, C. D., Anthony, I., Lambert, D. M., et al. (1994). Patterns of reproductive success determined by DNA fingerprinting in a communally breeding oceanic bird. Biological Journal of the Linnean Society, 52, 31–48. Mulder, R. A., Dunn, P. O., Cockburn, A., Lazenby-Cohen, K. A., & Howell, M. J. (1994). Helpers liberate female fairy-wrens from constraints on extra-pair mate choice. Proceedings of the Royal Society of London, Series B, 255, 223–229. Mumme, R. L., Koenig, W. D., & Pitelka, F. A. (1983). Reproductive competition in the communal acorn woodpecker: sisters destroy each other’s eggs. Nature, 305, 583–584. Nonacs, P. (2000). Measuring and using skew in the study of social behavior and evolution. American Naturalist, 156, 577–589. Nonacs, P. (2003). Measuring the reliability of skew indices: is there one best index? Animal Behaviour, 65, 615–627. Pamilo, P. & Crozier, R. H. (1996). Reproductive skew simplified. Oikos, 75, 533–535. Petrie, M. & Kempenaers, B. (1998). Extra-pair paternity in birds: explaining variation between species and populations. Trends in Ecology and Evolution, 13, 52–58. Petrie, M., Krupa, A., & Burke, T. (1999). Peacocks lek with relatives even in the absence of social and environmental cues. Nature, 401, 155–157. Pusey, A. & Wolf, M. (1996). Inbreeding avoidance in animals. Trends in Ecology and Evolution, 11, 201–206. Quinn, J. S., Woolfenden, G. E., Fitzpatrick, J. W., & White, B. N. (1999). Multi-locus DNA fingerprinting supports genetic monogamy in Florida scrub-jays. Behavioral Ecology and Sociobiology, 45, 1–10.
Reproductive skew in avian societies Reeve, H. K. (2000). A transactional theory of within-group conflict. American Naturalist, 155, 365–382. Reeve, H. K. & Emlen, S. T. (2000). Reproductive skew and group size: an N-person staying incentive model. Behavioral Ecology, 11, 640–647. Reeve, H. K. & Jeanne, R. L. (2003). From individual control to majority rule: extending transactional models of reproductive skew in animal societies. Proceedings of the Royal Society of London B, 270, 1041–1045. Reeve, H. K. & Keller, L. (1995). Partitioning of reproduction in mother–daughter versus sibling associations: a test of optimal skew theory. American Naturalist, 145, 119–132. Reeve, H. K. & Keller, L. (1996). Relatedness asymmetry and reproductive sharing in animal societies. American Naturalist, 148, 764–769. Reeve, H. K. & Keller, L. (2001). Tests of reproductive-skew models in social insects. Annual Review of Entomology, 46, 347–385. Reeve, H. K. & Nonacs, P. (1992). Social contrasts in wasp societies. Nature, 359, 823–825. Reeve, H. K. & Shen, S.-F. (2006). A missing model in reproductive skew theory: the bordered tug-of-war. Proceedings of the National Academy of Sciences of the USA, 103, 8430–8434. Reeve, H. K., Emlen, S. T., & Keller, L. (1998). Reproductive sharing in animal societies: reproductive incentives or incomplete control by dominant breeders? Behavioral Ecology, 9, 267–278. Rowley, I. (1978). Communal activities among white-winged choughs Corcorax melanorhamphos. Ibis, 120, 178–197. Shellman-Reeve, J. S. & Reeve, H. K. (2000). Extra-pair paternity as the result of reproductive transactions between paired mates. Proceedings of the Royal Society of London B, 267, 2543–2546. Shields, W. M. (1982). Philopatry, Inbreeding Avoidance, and the Evolution of Sex. Albany, NY: State University of New York Press. Shorey, L., Piertney, S., Stone, J., & Ho¨glund, J. (2000). Fine-scale genetic structure on Manacus manacus leks. Nature, 408, 352–353. Stacey, P. B. (1982). Female promiscuity and male reproductive success in social birds and mammals. American Naturalist, 120, 51–64. Vehrencamp, S. L. (1977). Relative fecundity and parental effort in communally nesting anis, Crotophaga sulcirostris. Science, 197, 403–405. Vehrencamp, S. L. & Quinn, J. S. (2004). Joint laying systems. In W. D. Koenig & J. L. Dickinson, eds., Ecology and Evolution of Cooperative Breeding in Birds. Cambridge, UK: Cambridge University Press, pp. 177–196. Watts, C. R. & Stokes, A. W. (1971). The social order of turkeys. Scientific American, 224 (6), 112–118. Webster, M. S., Pruett-Jones, S., Westneat, D. F., & Arnold, S. J. (1995). Measuring the effects of pairing success, extra-pair copulations and mate quality on the opportunity for sexual selection. Evolution, 49, 1147–1157.
263
264
W. D. Koenig et al. Whittingham, L. A., Dunne, P. O., & Magrath, R. D. (1997). Relatedness, polyandry and extra-group paternity in the cooperatively-breeding white-browed scrubwren. Behavioral Ecology and Sociobiology, 40, 261–270. Williams, D. A. (2004). Female control of reproductive skew in cooperatively breeding brown jays (Cyanocorax morio). Behavioral Ecology and Sociobiology, 55, 370–380. Williams, D. A., Lawton, M. F., & Lawton, R. O. (1994). Population growth, range expansion, and competition in the cooperatively breeding brown jay, Cyanocorax morio. Animal Behaviour, 48, 309–322. Woolfenden, G. E. & Fitzpatrick, J. W. (1984). The Florida Scrub Jay: Demography of a Cooperative-Breeding Bird. Princeton, NJ: Princeton University Press. Yuan, H.-W., Liu, M., & Shen, S.-F. (2004). Joint nesting in Taiwan yuhinas: a rare passerine case. Candor, 106, 862–872. Zahavi, A. (1990). Arabian babblers: the quest for social status in a cooperative breeder. In P. B. Stacey & W. D. Koenig, eds., Cooperative Breeding in Birds: Long-Term Studies of Ecology and Behavior. Cambridge: Cambridge University Press, pp. 105–130.
10
Reproductive skew in cooperative fish groups: virtue and limitations of alternative modeling approaches michael taborsky
Summary Four types of cooperation between conspecific competitors can be differentiated in fish reproduction: joint defense of a spawning site or territory, joint preparations for spawning, cooperative spawning, and cooperative brood care. Long-lasting associations allowing for different reproductive shares of partners are mainly found in species that cooperate in territory defense and brood care. Here I use skew theory to scrutinize patterns of reproductive participation among related and unrelated group members in cooperatively reproducing cichlids and wrasses. A comparison of five species from which sufficient data are available does not reveal an obvious relationship between average relatedness, or group size, and reproductive skew levels, as predicted from respective skew models. It is remarkable that superficially similar cooperative systems in fish may be based on distinctly different parameter combinations, even in closely related species. I discuss five alternative schemes to understand the patterns of reproductive participation in cooperatively reproducing fish, including kin-selection theory, reciprocity models, manipulation or coercive strategies, models of alternative reproductive tactics, and a dynamic modeling approach. A comparison of approaches suggests that conventional skew models do not account for the complexity of evolutionary mechanisms involved in reproductive skew among members of fish groups. Alternative approaches, such as reciprocity theory or models to explain the coexistence of alternative reproductive tactics, may have greater explanatory Reproductive Skew in Vertebrates: Proximate and Ultimate Causes, ed. Reinmar Hager and Clara B. Jones. Published by Cambridge University Press. ª Cambridge University Press 2009.
265
266
M. Taborsky potential, at least in some cases. However, they have generally not been developed sufficiently to derive predictions allowing for a conclusive test. To understand how decisions of distinctly different types of group members evolve, an approach is needed that takes account of the state dynamics involved, as in fish, conditions constantly change due to their continued growth after maturation. I conclude by outlining widespread fallacies in the application of theory to interpret data, caused by the use of general models that are based on purposely simple assumptions to explain case studies while neglecting the mechanisms involved.
Introduction Reproductive-skew theory has been developed to explain the extent to which breeding is monopolized by dominant individuals in cooperatively breeding animals (Johnstone 2000). Transactional models focus on group stability and the constraints this places on the division of reproduction (Vehrencamp 1979, 1983, Stacey 1982, Keller & Reeve 1994, Clutton-Brock 1998, Reeve et al. 1998, Johnstone & Cant 1999). Dominant individuals with full control over the reproductive share of subordinates trade reproductive concessions against the help they receive (concession model), or, when their control is incomplete, subordinates may trade the threat of eviction against their reproductive participation (restraint model). In contrast, compromise models view the division of reproduction as the outcome of a conflict between group members, where each individual has some ability to enforce its own optimum, ignoring issues of group stability (Cant 1998, Reeve et al. 1998). In a “tug-of-war” dominants and subordinates face a trade-off between maximizing overall group productivity or their own share of reproduction (Reeve et al. 1998). Synthetic models of reproductive skew view these different approaches as special cases of a general underlying theory rather than as alternative paradigms and aim at combining them into a more general model that incorporates the possibility of both transactions and intra-group conflict (Johnstone 2000, Reeve & Shen 2006). Reproductive-skew models have primarily been used to understand the extremely uneven reproductive shares of females in social insects (Reeve & Ratnieks 1993, Langer et al. 2004; see Reeve & Keller 2001 for review), but they have also been applied to the study of reproductive share in male and female vertebrates showing cooperative breeding (Cant 2000, Faulkes & Bennett 2001, Haydock & Koenig 2002, 2003, Rusu & Krackow 2004, Heg et al. 2006; see also Chapter 9). In this chapter I aim to review the suitability of reproductive-skew theory to explain the sharing of reproduction between members of groups in
Cooperative fish groups: alternative modeling approaches fish. I shall first outline how reproductive competitors in fish cooperate in one way or another to illustrate the scope of social patterns that might be better understood by the use of reproductive-skew theory. I shall then relate the available information about these systems to the predictions of various skew models and discuss alternative concepts that might be applied in order to understand the evolutionary mechanisms underlying reproductive sharing in fish societies. Finally, I shall examine the explanatory power of reproductiveskew models for the behavior of fish species where group members cooperate during reproduction.
Cooperative reproductive behavior in fish With more than 25 000 described species, fish are by far the most speciose and diverse class of vertebrates (there are more fish species than all other vertebrate classes combined: Reid & Hall 2003). This is reflected by their unparalleled variation in reproductive patterns (Breder & Rosen 1966, Keenleyside 1979, Blumer 1982, Taborsky 1987, 1994, 1998, 1999, 2001, 2008a, Turner 1993, Berglund 1997, Petersen & Warner 1998, Wisenden 1999, Avise et al. 2002). Taborsky (1994) listed 95 fish species involved in cooperative reproductive relationships or alloparental care, and more examples have been described since (Taborsky 2001, Heg & Bachar 2006). Some of these cases involve interspecific alloparental care, which is most often parasitic rather than cooperative in nature and in any case is not relevant for the question of reproductive skew among group members. If we only include cases that involve cooperation between male reproductive competitors, we can differentiate between four different forms of cooperation: (1) joint defense of spawning site or breeding territory, (2) joint preparations for spawning (nest building, courtship), (3) cooperative spawning, and (4) cooperative brood care (see Table 10.1, listing 38 species of 10 families). Regarding the latter, there are various conditions leading to brood care of foreign offspring that are not of interest for the question of reproductive skew among group members, including egg stealing, adoption of young, nest takeovers, and joint offspring defense by neighboring pairs (see Taborsky 1994 for review). In category (4) I consider only cases of alloparental care between members of long-lasting social groups, a situation resembling “helpers-at-the-nest” in birds or alloparental care in mammals. For our purpose here I only include cases with reproductively mature helpers. Below I shall describe five case examples in detail belonging to the categories (1) joint defense and (4) alloparental care outlined above. The discussion will be confined to cases from which sufficient information is available on cooperative
267
268
M. Taborsky effort of potential competitors and their reproductive participation and success, to be able to investigate the explanatory power of skew theory. Cooperative reproduction of bourgeois and satellite males (“joint defense”)
At least eight species of fish have been studied in which bourgeois males1 defending a spawning site, courting females, and providing brood care are assisted by satellite males that are tolerated in their territories (see Taborsky 1994 for review). Usually these satellites join in defense against reproductive competitors, and in turn they participate in attempts to fertilize eggs deposited by females at the defended nests or spawning sites. In the Mediterranean ocellated wrasse (Symphodus ocellatus), between one and three satellites cooperate with nest owners in defense against small reproductive parasites or sneakers (Figure 10.1), and these male partners stay together for most of the spawning period at a particular nest (Taborsky et al. 1987). Males of this species belong to one of three different life histories, which relates to their growth pattern in the first year of life (Soljan 1930, Alonzo et al. 2000). Satellites do not join in brood care or in defense against egg predators, and they leave the nest after spawning has ceased. Nest males with satellites appear to be much more successful than those without, because females visit such nests more frequently and apparently prefer to spawn there (Taborsky 2001). Reproductive cooperation in polygynandrous breeding groups (“alloparental care”)
In the West African cichlid Pelvicachromis pulcher male helpers are permanently accepted in harem territories, where they help in defense against reproductive competitors and predators of eggs and fry. In turn they participate in fertilizing eggs produced in these groups, but only from a minimum harem size of three females (Martin & Taborsky 1997). This breeding system is characterized by a high degree of reciprocity among males. Harem owners tolerate up to three male helpers, at the expense of the latters’ participation in offspring production, but the dominant males gain by their enormous defense effort. Helpers pay a high price by risky defense that often results in injuries, but dominant helpers obtain a similar reproductive success to monogamous males (Martin & Taborsky 1997). Helpers in the second and third rank positions expend more effort per success than dominant ones, but their competitive ability is apparently lower.
1
The term “bourgeois males” refers to individuals investing in privileged access to mates, by behavioral (e.g. defense, courtship), physiological, or morphological means (e.g. pheromones, secondary sexual characters: Taborsky 1997).
Cooperative fish groups: alternative modeling approaches
Figure 10.1 Spawning in Symphodus ocellatus seen from above. Three fish are in the nest built mainly of brown and red algae by the nest male, which is on the bottom of the nest (below two other fish) just in the movement of spawning (curved body). The female is in the foreground adjacent to the nest male, showing her extended belly and white genital papilla, just about to release eggs. The satellite male is above the spawning nest male and female in an S-bent position with spread opercula, attempting to herd the female into the nest. In the upper left corner of the picture, a small sneaker male is just about to rush to the nest to release sperm (blurred due to its fast approach). (Photo by Michael Taborsky taken at Pointe de la Revellata, Corsica)
Brood-care helpers (“alloparental care”)
Cooperative brood care has been described in at least 17 species of fish, most of which belong to the substrate-breeding cichlids of Lake Tanganyika (Table 10.1; Taborsky 1994; Heg & Bachar 2006 for review). In all species studied in sufficient detail to date, reproductive competition is found among group members, especially among males (Taborsky 1985, Kohler 1998, Dierkes et al. 1999, 2008, Awata et al. 2005, 2006, Heg et al. 2006). Mature brood-care helpers either were born in the territory where they help and share in reproduction, or they have joined the group at a later stage (Taborsky & Limberger 1981, Stiver et al. 2004, Awata et al. 2005, Bergmu¨ller et al. 2005a, Dierkes et al. 2005, Heg & Bachar 2006). Relatedness is variable between dominant breeders and their mature helpers, ranging from zero or very low degrees of relatedness (Awata et al. 2005, Dierkes et al. 2005, Stiver et al. 2005) to relatedness levels considerably exceeding the population mean (Taborsky & Limberger 1981,
269
270
M. Taborsky Kohler 1998). In the species from which we know most, Neolamprologus pulcher (Figure 10.2), helpers of both sexes share in direct and indirect brood care by cleaning and fanning eggs, cleaning larvae, digging out shelters, removing snails, and defending the brood and territory against predators and competitors (Taborsky & Limberger 1981, Taborsky 1984, Taborsky et al. 1986, BalshineEarn et al. 1998, Stiver et al. 2005). After reaching maturity, male helpers share in reproduction by surreptitious release of sperm while the breeders spawn (Taborsky 1985, Dierkes et al. 1999, 2008, Heg et al. 2006). Female helpers may breed in the dominant’s territory as well, or they may split off part of the territory to initiate a harem with the male breeder or pair up with a foreign male (Taborsky 1985, Heg et al. 2006, Heg & Hamilton 2008). Superficially the social system of N. savoryi looks similar, but the groups are usually haremic in this species and subdivided into subgroups, with each female assisted by her own set of helpers, and the male breeder defending the entire area encompassing all subgroups (Heg et al. 2005b). In the congener N. multifasciatus, young do not disperse from their natal territories and they share in territory defense and maintenance, especially digging (Kohler 1998). Usually more adult females than males are present per group (Taborsky 2001). Both mature male and female helpers participate regularly in reproduction (Kohler 1998). In the closely related cichlid Julidochromis ornatus, helpers stay with a breeding pair or harem in roughly 40–95% of breeding groups depending on the population, with male helpers outnumbering female helpers about twofold (Awata et al. 2005, Heg & Bachar 2006). Usually there are only one or two helpers present, but helper numbers may go up to six. These helpers are usually unrelated to the dominant breeders, they share in the defense of broods and in digging, and they participate in offspring production (Awata et al. 2005, 2006, Heg & Bachar 2006). This system is somewhat reminiscent of the polygynandrous groups in Pelvicachromis pulcher (Martin & Taborsky 1997), but in J. ornatus the pattern is more flexible, with monogamous pairs and male and female harems all occurring with and without helpers of both sexes. The group structure and cooperative brood care seems similar in the congener J. marlieri (Taborsky 1994, unpublished data, Sunobe 2000), but there is no information available for this species on reproductive participation of group members.
Use of skew theory to understand patterns of reproductive participation in cooperatively reproducing fish In this section I shall examine how reproductive-skew models can help us understand the reproductive shares of group members in the examples
271
Cooperative fish groups: alternative modeling approaches
M. Taborsky
Table 10.1 Cooperative reproduction between adult male conspecifics in fish Catostomidae Catostomus commersonii
joint spawning
Reighard 1920
Hypentelium nigricans
joint spawning
Raney & Lachner 1946
Moxostoma aureolum
joint spawning
Reighard 1920
M. duquesnei
joint spawning
Bowman 1970, Kwak & Skelly 1992
M. carinatum
joint courtship & spawning
Hackney et al. 1968
M. macrolepidotum
joint spawning
Jenkins 1970, Burr & Morris 1977
M. erythrurum
joint spawning
Jenkins 1970, Kwak & Skelly 1992
M. valenciennesi
joint spawning
Jenkins 1970, Jenkins & Jenkins
Erimyzon oblongus
joint spawning
Page & Johnston 1990
1980 Cyprinidae Nocomis micropogon
joint nest building
Reighard 1943
Notropis leptocephalus
joint nest building
Wallin 1989
joint defense
Barlow 1961
Cyprinodontidae Cyprinodon macularis Cottidae Hemilepidotus hemilepidotus joint brood defense
DeMartini & Patten 1979
Osphronemidae a
Betta brownorum
joint defense
B. persephone
joint defensea
Witte & Schmidt 1992 Witte & Schmidt 1992
Blenniidae Parablennius sanguinolentus joint defense
Santos 1985, Santos & Almada 1988, Oliveira et al. 2002
Cichlidae Pelvicachromis pulcher
joint brood defense
Martin & Taborsky 1997
Neolamprologus pulcher b
cooperative breeding
Taborsky & Limberger 1981, Taborsky 1984, 1985, Taborsky & Grantner 1998, Balshine et al. 2001, Brouwer et al. 2005, Dierkes et al. 2005, Heg et al. 2005a, Stiver et al. 2005
N. gracilis
cooperative breeding
Woodland 2002, H. H. Bu¨scher
N. helianthus
cooperative breeding
H. H. Bu¨scher pers. comm. in Heg
N. marunguensis
cooperative breeding
H. H. Bu¨scher pers. comm. in Heg
N. olivaceous
cooperative breeding
H. H. Bu¨scher pers. comm. in Heg
N. falcicula
cooperative breeding
& Bachar 2006 own obs. at Kigoma, Tanzania
N. splendens
cooperative breeding
H. H. Bu¨scher pers. comm. in Heg
N. savoryi
cooperative breeding
Taborsky & Limberger 1981;
N. multifasciatus
cooperative breeding
Kohler 1998
N. similis
cooperative breeding
Bu¨scher 1992 and pers. comm. in
Julidochromis marlieri
cooperative breeding
Taborsky & Limberger 1981,
J. regani
cooperative breedinga
Taborsky & Limberger 1981
J. ornatus
cooperative breeding
Taborsky & Limberger 1981, Awata
pers. comm. in Heg & Bachar 2006 & Bachar 2006 & Bachar 2006
& Bachar 2006 Kondo 1986; Heg et al. 2005b
Heg & Bachar 2006 Taborsky 1994, Yamagishi & Kohda 1996, Sunobe 2000
et al. 2005, 2006, Heg & Bachar 2006 J. transcriptus Chalinochromis brichardi
cooperative breeding
Kuwamura 1997
cooperative breeding
Awata et al. 2005
joint defense
Fricke 1979
Symphodus ocellatus
joint defense
Fiedler 1964, Lejeune 1985,
S. roissali
joint defense
Lejeune 1985
Pomacentridae Amphiprion akallopisos Labridae Taborsky et al. 1987 S. tinca
joint defense
Lejeune 1985
Halichoeres maculipinna
joint defense
Thresher 1979
joint courtship
Fahy 1954
Percidae Etheostoma blennioides
Examples are included only if adult males cooperate in some reproductive activity. “Joint defense” refers to defense of a spawning site or breeding territory; if broods are jointly defended, this is called “joint brood defense.” Cases where neighboring breeding pairs defend their young jointly (see Taborsky 1994 for review and examples) are not included, because it is unlikely that these associations persist during the spawning phase. In cooperatively breeding species, both adult male and female brood-care helpers may be tolerated by dominant breeders. a
Only observations from captivity are available.
b
Neolamprologus pulcher and N. brichardi have been regarded as two species (Poll 1974), but
recent molecular, morphological, and ecological evidence suggests that this is not justified (Duftner et al. 2007, E. Skubic et al. unpublished).
272
Cooperative fish groups: alternative modeling approaches
Figure 10.2 In front of a breeding shelter, a dominant breeder of Neolamprologus pulcher (left) approaches a subordinate helper, which responds with a submissive display (tail vibration, with folded fins). Below these two adult fish there is a small young (only front half visible). (Photo taken by Michael Taborsky at Kasakalawe Point, Zambia)
described above. Table 10.2 provides basic information about the reproductive biology and behavior of these example species. Reproductive skew between bourgeois and satellite males in Symphodus ocellatus
Nest males and satellite males of this species defend the nest jointly against opportunistic reproductive competitors (sneakers). Nest males are always bigger than satellite males and always dominant. They seem to be in control of group membership of their satellites, which is mediated by an elaborate exchange of display behaviors between the two parties (Taborsky et al. 1987). The cooperators in S. ocellatus are not related to each other above the population mean (Basieux 2007), because they have been subject to a planktonic dispersal phase at their larval stage and, in addition, they often belong to different age classes (Alonzo et al. 2000). The reproductive skew is high between the cooperators: as revealed by DNA microsatellite analyses, nest males sired 47% of offspring produced in their nests (median; range 2–72%; n ¼ 15 nests) when satellites were present, and satellites sired 0.63% (median; range 0–22%; n ¼ 15 nests), which is only 1.34% of the nest male’s share (Figure 10.3); all other young were sired by non-cooperative, parasitic males (Basieux 2007), which visit these nests in great numbers (Taborsky et al. 1987,
273
274
M. Taborsky Table 10.2 Basic information on the reproductive biology and behavior of the five species presented as examples in the text Symphodus
Pelvicachro-
Neolamprologus N.
Trait
ocellatus
mis pulcher
pulcher
multifasciatus ornatus
Julidochromis
Brood care of
male
biparental
biparental
biparental
male
male, female & male, female male, female
biparental
breeders Sex of subordinates male
immature
&
&
immature
immature
Brood care of helpers: defence of brood no
yes
yes
yes
yes
territory
no
yes
yes
yes
no
maintenance direct brood carea no Subordinates raise
no
yes
no
yes
no
yesb
yes
?
yesb
low
high
medium
medium
high
breeders’ success Reprod. share of male subordinates Control of group
male breeder male breeder pair members pair
membership
(þ/-)
Classification of
fixed and
ARTsc
members sequential
sequential
sequential
plasticd
(reversible
(fixed
(fixed
(fixed
sequence)
sequence)
sequence)
pair members sequentiale
sequence) Information compiled from Taborsky & Limberger 1981, Taborsky 1984, 1985, 1994, Martin & Taborsky 1997, Kohler 1998, Dierkes et al. 1999, Alonzo et al. 2000, Schradin & Lamprecht 2000, Awata et al. 2005, Brouwer et al. 2005, Heg & Bachar 2006, Basieux 2007. a
In fish this includes fanning of eggs, removing infected eggs, and cleaning of eggs, larvae,
and free-swimming fry with the mouth. b
Only correlative evidence available.
c
Alternative reproductive tactics can be fixed for life or plastic; plastic tactics may be either
performed simultaneously depending on conditions, or sequentially with a fixed or reversible sequence (see Figure 1 in Taborsky et al. 2008). d
In S. ocellatus, males do not switch between bourgeois and subordinate tactics within a
season. Males that are already large in their first season may reproduce as satellites and turn into bourgeois nest males in their second season. Males that are small in their first season reproduce as sneakers and turn into satellites in their second season. They never reproduce as bourgeois nest males (Alonzo et al. 2000). e
A sequential tactic with fixed sequence seems most likely in J. ornatus, but some individuals
might not pass through a helper phase at all (as derived from size distributions: Awata et al. 2005).
Cooperative fish groups: alternative modeling approaches
x adult subordinates/group
2.5
2.29
females
2.16
2.0 1.60 1.5
1.31
1.27 1.09
1.0
0.86 0.63
0.5
Species: N helpers:
0.4 Degree of relatedness (r)
males
S.o. 104
P.p. 50
N.p. 31
N.m. 32
J.o. 35
0.375
to male breeder to female breeder
0.3 0.205 0.2
0.158
0.1
0.043 0.029
0.033 (0) (0)
(0) (0)
S.o. a
P.p. b
0 Species: N helpers:
N.p. 99
77.1
80
N.m. 12+22c
78.8
Paternity of young in groups [%]
67.5 60
J.o. 35
sired by dom. sired by subdom.
50.4
47 40
34.7
32.5 21.2
20 10.3 0.63
Species: N helpers:
S.o. 15
P.p. 10
N.p. 12+7d
N.m. 10
J.o. 22
Figure 10.3 Number of adult subordinates per group (upper panel), relatedness between subordinate and dominant cooperators (middle panel), and levels of reproductive skew among male group members (lower panel), as determined in five species of fish with cooperative reproduction (Symphodus ocellatus, Pelvicachromis pulcher, Neolamprologus pulcher, N. multifasciatus, Julidochromis ornatus). See text for details and references. Exact values are given above each column or point, and sample sizes are provided below the abscissa. Paternity estimates were obtained
275
276
M. Taborsky Alonzo & Warner 1999). The paternity of individual sneakers is unknown, but when comparing the reproductive behavior of satellites with sneakers, by being tolerated the former stay much closer to the nest than the latter (median distance 35 cm vs. 75 cm), which provides a better position for fertiliz