Advances in the Study of Behavior, Volume 30 (Advances in the Study of Behavior)

Advances in

THE STUDY OF BEHAVIOR VOLUME 30

Advances in THE STUDY OF BEHAVIOR Edited by

PETER J. B. SLATER JAY S. ROSENBLATT CHARLES T. SNOWDON TIMOTHY J. ROPER

Advances in THE STUDY OF BEHAVIOR Edited by PETER J. B. SLATER School of Biology University of St. Andrews Fife, United Kingdom

JAY S. ROSENBLATT Institute of Animal Behavior Rutgers University Newark, New Jersey

CHARLES T. SNOWDON Department of Psychology University of Wisconsin Madison, Wisconsin

TIMOTHY J. ROPER School of Biological Sciences University of Sussex Sussex, United Kingdom

VOLUME 30

San Diego San Francisco New York Boston London Sydney Tokyo

∞ This book is printed on acid-free paper. 

Copyright C 2001 by ACADEMIC PRESS All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2001 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0065-3454/01 $35.00 Explicit permission from Academic Press is not required to reproduce a maximum of two figures or tables from an Academic Press chapter in another scientific or research publication provided that the material has not been credited to another source and that full credit to the Academic Press chapter is given.

Academic Press A Harcourt Science and Technology Company 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com

Academic Press Harcourt Place, 32 Jamestown Road, London NW1 7BY, UK http://www.academicpress.com International Standard Book Number: 0-12-004530-3 PRINTED IN THE UNITED STATES OF AMERICA 01 02 03 04 05 06 QW 9 8 7 6 5 4 3 2 1

Contents

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

ix xi

The Evolution of Alternative Strategies and Tactics H. JANE BROCKMANN I. Introduction and Approach . . . . . . . . . . . . . . . . . . . . . . . . . II. How Are Alternative Allocation Patterns Maintained in Populations? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Case Studies of Alternative Allocation Patterns . . . . . . . . . IV. An Integrative Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Modeling Alternative Allocation Patterns . . . . . . . . . . . . . . VI. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 10 17 33 34 37 39

Information Gathering and Communication during Agonistic Encounters: A Case Study of Hermit Crabs ROBERT W. ELWOOD AND MARK BRIFFA I. II. III. IV. V. VI. VII. VIII. IX. X.

Game Theory and Animal Contests . . . . . . . . . . . . . . . . . . . Hermit Crab Shell Fights . . . . . . . . . . . . . . . . . . . . . . . . . . . Resource Value and Decisions during Contests . . . . . . . . . . Relative Size of the Opponents and Decisions in Fights . . . . A Model of Information Gathering and Motivational Change . . . . . . . . . . . . . . . . . . . . . . . . . . Motivational Change during Agonistic Encounters . . . . . . . Probing Motivational Change during Animal Contests . . . . The Vigor of Shell Rapping and Communication . . . . . . . . . Negotiation or Aggression in Hermit Crab Shell Fights? . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

53 56 58 61 63 73 75 77 91 93 95

vi

CONTENTS

Acoustic Communication in Two Groups of Closely Related Treefrogs H. CARL GERHARDT I. II. III. IV. V. VI.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hyla cinerea Species Group . . . . . . . . . . . . . . . . . . . . . . . . . Hyla versicolor Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparisons with Other Taxa . . . . . . . . . . . . . . . . . . . . . . . Geographic Variation in Acoustic Communication . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

99 102 126 147 154 157 159

Scent-Marking by Male Mammals: Cheat-Proof Signals to Competitors and Mates L. M. GOSLING AND S. C. ROBERTS I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Scent-Marking and Competition between Males . . . . . . . . . III. How Do Signalers Ensure That Their Scent Marks Are Detected? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. How Do Receivers Use Scent Marks to Assess Signalers? . . V. Costs and Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Eavesdropping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Scent-Marking and Mate Choice . . . . . . . . . . . . . . . . . . . . . VIII. Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . IX. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

169 172 174 181 190 195 197 201 204 206

Male Facial Attractiveness: Perceived Personality and Shifting Female Preferences for Male Traits across the Menstrual Cycle IAN S. PENTON-VOAK AND DAVID I. PERRETT I. II. III. IV. V.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Theories of Human Facial Attractiveness . . . . . . . . . . . . . . . Personality Attributions and Facial Characteristics . . . . . . . Menstrual Cycle Shifts in Face Preferences . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

219 220 232 238 252 253

vii

CONTENTS

The Control and Function of Agonism in Avian Broodmates HUGH DRUMMOND I. II. III. IV. V.

Introduction . . . . . Proximate Control Functions . . . . . . . Conclusion . . . . . . Summary . . . . . . . References . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

261 263 282 291 294 295

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

303

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

317

This Page Intentionally Left Blank

Contributors

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

MARK BRIFFA (53), School of Biology and Biochemistry, The Queen’s University of Belfast, Belfast, BT9 7BL, United Kingdom H. JANE BROCKMANN (1), Department of Zoology, University of Florida, Gainesville, Florida 32611 HUGH DRUMMOND (261), Laboratorio de Conducta Animal, Instituto de ´ Universidad Nacional Autonoma ´ Ecologia, de M´exico, A.P. 70-275, 04510 D.F., Mexico ROBERT W. ELWOOD (53), School of Biology and Biochemistry, The Queen’s University of Belfast, Belfast, BT9 7BL, United Kingdom H. CARL GERHARDT (99), Division of Biological Sciences, University of Missouri, Columbia, Missouri 65211 L. M. GOSLING (169), Evolution and Behaviour Research Group, Department of Psychology, University of Newcastle, Newcastle-upon-Tyne NE1 7RU, United Kingdom IAN S. PENTON-VOAK (219), School of Psychology, University of St. Andrews, St. Andrews, Fife, KY18 9JU, United Kingdom DAVID I. PERRETT (219), School of Psychology, University of St. Andrews, St. Andrews, Fife, KY18 9JU, United Kingdom S. C. ROBERTS (169), Evolution and Behaviour Research Group, Department of Psychology, University of Newcastle, Newcastle-upon-Tyne NE1 7RU, United Kingdom

ix

This Page Intentionally Left Blank

Preface

The aim of Advances remains as it has been since the series began: to serve the increasing number of scientists who are engaged in the study of animal behavior by presenting their theoretical ideas and research to their colleagues and to those in neighboring fields. We hope that the series will continue its “contribution to the development of cooperation and communication among scientists in our field,” as its intended role was phrased in the preface to the first volume in 1965. Since that time traditional areas of animal behavior research have achieved new vigor by the links they have formed with related fields and by the closer relationship that now exists between those studying animal and human subjects. Scientists studying behavior today range more widely than ever before: from ecologists and evolutionary biologists, to geneticists, endocrinologists, pharmacologists, neurobiologists, and developmental psychobiologists, not forgetting the ethologists and comparative psychologists whose prime domain is the study of behavior. It is our intention not to focus narrowly on one or a few of these fields, but to publish articles covering the best behavioral work from a broad spectrum. The skills and concepts of scientists in such diverse fields necessarily differ, making the task of developing cooperation and communication among them a difficult one. But it is one that is of great importance, and one to which the editors and publisher of Advances in the Study of Behavior are committed. We will continue to provide the means to this end by publishing critical reviews, by inviting extended presentations of significant research programs, by encouraging the writing of theoretical syntheses and reformulations of persistent problems, and by highlighting especially penetrating research that introduces important new concepts. The present volume illustrates these aims well. It is taxonomically varied, with two chapters primarily on invertebrates, one on amphibia, one on birds, and two on mammals. All of the chapters have a strong theoretical base and place their subjects in an evolutionary framework, including that on human facial expressions. Although the editors did not set out to achieve a common theme, much of the work described concerns the relationships between individuals, and especially communication, a growth area within the study of behavior at present. All the chapters tackle important topics and come up with insights of wide significance to those interested in the study of behavior.

xi

This Page Intentionally Left Blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 30

The Evolution of Alternative Strategies and Tactics H. JANE BROCKMANN DEPARTMENT OF ZOOLOGY UNIVERSITY OF FLORIDA GAINESVILLE, FLORIDA

32611

I. INTRODUCTION AND APPROACH Understanding variation and how variation is maintained in populations has been an important problem in animal behavior and in evolutionary biology since the time of Darwin (1859). In a scientific world that emphasized typological thinking, Darwin saw variation as an intrinsic property of life. He emphasized differences between species and sexes, but he was also interested in variation within one sex. For example, he described two types of males in the Brazilian amphipod Orchestia darwinii that differed in the structure of their chelae: one claw was enlarged and shaped differently from the other (Darwin, 1871). Darwin speculated on how the two forms could be maintained: “Both forms have derived certain special, but nearly equal advantages from their differently shaped organs.” He did not know the function of the chelae for certain, whether they were for securing the female or fighting with other males, but he understood that having two forms of one sex at the same time in one population required a special explanation. For many years such variation was largely ignored as having no adaptive function (Alcock, 1979; Majerus, 1998), but over the past 25 years much has been learned about the evolution of these alternative strategies and tactics (Maynard Smith, 1982; Andersson, 1994; Dugatkin and Reeve, 1998). A. PATTERNS OF VARIATION Variation in a continuous character such as size can be described with a frequency distribution (Fig. 1). For example, in the mud-daubing wasp, Trypoxylon politum, the female mass provisions a brood cell with a number of paralyzed spiders, lays a single egg, and seals off the brood cell where the offspring develops with whatever it was given by its mother (Brockmann and Grafen, 1989). The variation that we see in male and female 1

C 2001 by Academic Press Copyright  All rights of reproduction in any form reserved. 0065-3454/01 $35.00

2

H. JANE BROCKMANN

FIG. 1. Variation in the size of male and female Trypoxylon politum, the solitary pipe-organ mud-daubing wasp from Florida (overwintering generation). Two frequency distributions, one for males and one for females, are shown. The variation in these distributions is the result of allocation decisions by females toward son and daughter production. (Figure based on data presented in Table 5 of Brockmann and Grafen, 1992.)

offspring size could be derived from a number of sources: (1) genotypic differences between females in how much food they allocate to offspring of each sex; (2) genetic differences between young in metabolism or utilization of prey; (3) environmental differences between the provisioning females, such as differences in hunting conditions; (4) environmental differences between offspring, such as differences in temperature during development; and (5) differences in gene–environment interactions for mothers (e.g., females with different genotypes may allocate prey differently when hunting conditions are good) or for offspring. Natural selection may shift the mean or the variance of a population if fitness differences exist between different parts of the distribution, but we normally expect a single peak for each sex, that is, one optimal offspring size for males and one for females. What is much harder to explain is a bimodal distribution among adults of one sex (Fig. 2). For example, discontinuous variation in size and behavior may be found in colonies of ants (majors and minors) (Fig. 2a; Wilson, 1953) or in some species of horned beetles (Eberhard and Gutierrez, 1991). Discontinuous variation in body proportions and behavior is found in many species, such as short- and long-winged forms of crickets (Walker, 1986; Crnokrak and Roff, 2000) and water striders (Kaitala, 1988; Kaitala and Dingle, 1992, 1993) or fighting and nonfighting morphs of male mites

FIG. 2. (a) Frequency distribution of the sizes of workers from one nest of weaver ants, Oecophylla smaragdina, showing a bimodal distribution (modified from Wilson, 1953). (b) Frequency distribution of the behavior of single and multiple foundress females of the social wasp Mischocyttarus mexicanus. Single foundresses initiate nests on their own, whereas multiple foundresses initiate nests in collaboration with other females. Although the two types of females are indistinguishable morphologically, they show quite different behavior when forced to be alone on the nest, as the figure shows. Wing raises are an indication of defensive behavior in the presence of the observer (data from Clouse, 1997, and unpublished data).

4

H. JANE BROCKMANN

(Radwan, 1993), thrips (Crespi, 1986), and marine isopods (Shuster, 1990, 1992). Bimodal distributions in behavior may occur with no associated morphological differences, such as differences in the behavior of females on a paper wasp nest (Fig. 2b; Clouse, 1994, 1997), or in calling and noncalling behavior in crickets (Cade, 1979), or in the behavior of honeybees where some remain inside while others forage outside the nest (Seeley, 1985). In each case the maintenance of this kind of variation requires an explanation: surely, one of the phenotypes would, on average, have higher success than the other, so why is it that one trait does not drive the other from the population? Many cases of alternative phenotypes are not peaks in a continuous distribution, like size, but are discontinuous characters, like color patterns or distinct morphs. For example, female fig wasps lay eggs in figs and when the offspring emerge, some sons are winged and leave the natal fig to search for mating opportunities elsewhere, whereas others are wingless and remain within the fig and fight other wingless males for access to females that have not yet emerged (Hamilton, 1979). So two distinct, male morphs exist in one population at the same time (Cook et al., 1997). How is such variation maintained? Each of the different types of discrete variation has a different literature, a different way of describing the variation, and a somewhat different explanation for its maintenance. When discrete peaks in a frequency distribution of behavior or morphology occur and the differences are due to the animals being different sexes, we refer to this as sexual dimorphism. When differences occur among adults within a sex, for example in size, structure (including protein structure), or color, we call the variation polymorphism. When the peaks are differences in behavior within a sex, we refer to them as alternative behavior patterns (or alternative strategies and tactics) or, for social insects, polyethism. When the differences are associated with seasonal changes, they are called polyphenism, and when they are associated with other environmental factors, they are referred to as conditional polyphenism (Walker, 1986), phenotypic plasticity, or developmental reaction norms (Greene, 1999). Taken together, discrete phenotypic variation among adults may be referred to as variants (Dunbar, 1982) or alternative phenotypes (Gross, 1996) or, more loosely, alternative strategies and tactics (Davies, 1982). In this chapter I will argue that this diversity of discrete, alternative phenotypes found among adults of one sex is fundamentally similar in important ways (Dunbar, 1983; Waltz and Wolf, 1984; Moran, 1992; Charnov, 1993; Roff, 1996; Greene, 1999; Nijhout, 1999) despite being described and explained quite differently in the behavioral, sex-allocation, lifehistory, mimicry, and caste evolution literatures (Lloyd, 1987). By identifying similar patterns among a seemingly disparate set of behavioral, morphological, physiological, and life-history phenomena, we can examine whether similar underlying processes are shaping their evolution (Charnov, 1993).

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

5

I draw on these diverse literatures to illuminate how alternative phenotypes might evolve and be maintained in populations. Excellent reviews exist for vertebrates (e.g., Taborsky, 1994, 1999; Skulason ´ and Smith, 1995; Smith and Skulason, ´ 1996), so I will use examples from the invertebrate literature whenever possible. One of the reasons that different literatures deal with the problem of alternative phenotypes somewhat differently is that they ask somewhat different questions. For example, the alternative strategies and tactics literature has focused on predicting the frequencies of the two patterns in the population and on identifying the selective pressures that maintain those frequencies (e.g., Brockmann et al., 1979). The sex-allocation literature is also focused on frequencies of the two forms, that is, sex ratios, but emphasizes how females allocate their limited resources to son and daughter production (Charnov, 1982). The insect caste literature seeks to understand the developmental mechanisms that control morph phenotypes (Nijhout and Wheeler, 1982; Wheeler, 1991; Fahrbach, 1997) as well as the optimal morph ratios within a colony and the proximate cues that influence morph frequencies and their evolution (Oster and Wilson, 1978; Naug and Gadagkar, 1998). The mimicry literature studies the proximate factors that influence morph frequencies (Huheey, 1988; Turner, 1989; Bond and Kamil, 1998) and the evolutionary and developmental processes by which adaptations are built up within a morph (Turner, 1977). The polyphenism literature has tended to focus on the developmental control of differences (Pener, 1991; Nijhout, 1999) and on the proximate factors that cause individuals to switch from one form to another, whereas the reaction norm literature emphasizes the interacting effects of genotype and environment on the development of diverse phenotypes (Greene, 1999). For each question posed, both proximate and ultimate approaches are needed. (1) One major set of questions concerns the control and evolution of different frequencies of alternative phenotypes within a population (e.g., sex ratios). For this question we need to understand the selective (or other) factors and evolutionary processes that maintain those frequencies as well as the factors that control a change from one phenotype to another or a decision to invest in one pattern over another, which has the effect of altering frequencies. (2) The other major question concerns the control and evolution of phenotypic differences. What are the selective pressures that favor the appearance of a particular suite of characters that make up a phenotypic variant (e.g., why are males and females different in size)? The equivalent questions for alternative strategies are why do alternative phenotypes exist and what are the selective pressures and processes (e.g., disruptive selection) that favor different phenotypes. We also need to understand the proximate developmental and physiological factors that control and constrain the

6

H. JANE BROCKMANN

appearance of different phenotypes. In this chapter I will begin with some terminological issues and then focus on the first of these problems, how particular frequencies of alternative phenotypes are maintained in populations. B. DESCRIBING PATTERNS: TERMINOLOGY AND DEFINITIONS Alternative strategies and tactics refer to the presence of two or more discrete behavioral variants among adults of one sex and one population when those variants serve the same functional end, such as more than one way of foraging, or attracting mates, or nesting (Brockmann et al., 1979; Rubenstein, 1980; Dominey, 1984). Often discrete behavioral variants are associated with specific morphological, physiological, and life-history characters. As with polymorphism (Ford, 1940), these discrete variants occur and remain in populations at frequencies that cannot be explained by mutation or simple mistakes. From its inception, a clear distinction has been made between alternative phenotypes that are due to genetic differences, termed strategies, and those that are due to environmental or phenotypic conditions, termed tactics or outcomes (Alcock, 1979; Brockmann et al., 1979; Dawkins, 1980; Maynard Smith, 1982; Austad, 1984b; Austad and Howard, 1984; Dominey, 1984; Gross, 1984). Recently, Gross (1996) has identified two principal classes of alternative strategies: genetic polymorphism and genetic monomorphism (Table I). Gross reserves the term “alternative strategy” for cases of a known genetic polymorphism. If differences are not due to genetic differences, he argues, then the differences are due to decisionmaking processes within the individual. There are two types of decisionmaking processes. When the frequencies of the two tactics are set by a random decision rule, such as the toss of a coin, this is a “mixed strategy” (Maynard Smith, 1982) or what Dominey (1984) refers to as a “stochastic mixed” strategy (Brockmann et al., 1979; Dawkins, 1980). If the decision rule about which pattern to follow is facultative and made on the basis of environmental or individual conditional (or status-dependent) cues, then this is referred to as a “conditional strategy” with alternative tactics (Gross, 1996). Furthermore, in the original discussions of alternative strategies and tactics, a clear distinction is made between strategies that are affected by frequency-dependent selection (alternative strategies and stochastic mixed strategies) and those that are conditional on environment or phenotype, called conditional or making-the-best-of-a-bad-job strategies (Dawkins, 1980; Davies, 1982). When two alternative strategies are affected by frequency-dependent selection, there exists an equilibrium frequency (called the ESS) of the two patterns (Rubenstein, 1980). As long as the population remains at this equilibrium, the two patterns will be, on average, equally successful. Above or below the equilibrium, one pattern does better, thus

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

7

TABLE I CLASSIFICATION SYSTEMS FOR PATTERNS OF DISCRETE VARIATION WITHIN A POPULATION

Alternative strategiesa and tactics

Sex allocationb

Alternative allocation patterns; strategy allocationc alternative phenotypesd

One pattern

Asexual

One allocation pattern

Two or more patterns Genetic polymorphism or alternative strategy Genetic monomorphism or alternative tactics Mixed strategy Conditional strategy

Two sexes Dioecy (separate sexes) Hermaphroditism Simultaneous Sequential

Two or more allocation patterns Irreversible (fixed over individual’s lifetime) Reversible (changes over individual’s lifetime) Simultaneous (changes back and forth over individual’s lifetime) Sequential (switch from one phenotype to another at a particular point during individual’s life)

Combinations (e.g., gynodioecy) a Maynard Smith, 1982; Gross, 1996 (the patterns listed in the first column are not completely comparable to those listed in the second and third columns). b Charnov, 1982. c Lloyd, 1987. d Moran, 1992.

returning the population to the equilibrium frequency. However, with condition-dependent patterns, the tactics are not expected to do equally well (Dawkins, 1980; Dominey, 1984). Austad (1984b) classifies alternative behavior patterns as those that are, on average, equally successful (i.e., isogignous or alternative and mixed strategies) and those that are not (i.e., allogignous or conditional strategies). Three problems exist with these categories for alternative phenotypes. First, simply identifying the patterns (e.g., genetic vs nongenetic, equal vs nonequal fitness) requires the observer to know a great deal about the underlying mechanisms before the alternative phenotypes can be classified (Davies, 1982). To describe a pattern as an alternative strategy, for example, the observer must at least understand the underlying genetics. To be identified as a mixed ESS, the observer must establish that the decision rule is random and that the success of the two patterns is equal, a particularly difficult set of attributes to determine with any certainty (Brockmann et al., 1979; Parker, 1984; Caro and Bateson, 1986). So the classification that exists in this literature is not a classification of behavior, but a classification of its underlying mechanisms (Alcock, 1979; e.g., behavior due to individual flexibility or different genotypes) and their outcomes. In other words, we have not kept

8

H. JANE BROCKMANN

pattern separate from process in much of this literature. Second, classifying alternative phenotypes into genetic and environmental forces our thinking into nature versus nurture categories, a dichotomy that animal behaviorists have come to consider problematic (Caro and Bateson, 1986). Differences in behavior associated with alternative, mixed, or conditional strategies are all certainly affected by genes, environment, and gene–environment interactions. Although it is useful to understand how genes, environment, and their interactions work together to produce differences in phenotype, this is not the central focus of this field. Rather, we wish to understand the selective agents that act to maintain the frequencies of alternative phenotypes in the population and the factors that shape the evolution of alternative phenotypes (Taborsky, 1999). Third, the classification scheme has not allowed us to consider frequency-dependent effects on conditional strategies. What other classification system might we use to identify patterns among alternative phenotypes? Striking similarities exist among the three patterns of alternative phenotypes, alternative, mixed, and conditional strategy, and the well-known patterns of sex allocation (Table I). Charnov (1982) divides the world of sex-allocation phenotypes into (1) dioecious, when separate sexes are maintained throughout the lives of individuals, and (2) hermaphrodite, when sex changes occur over the adult life of the individual. Hermaphroditism is further subdivided into (a) sequential hermaphrodites when individuals change sex at a particular age, size, or condition during their lifetime, and (b) simultaneous hermaphrodites, when individuals switch back and forth between male and female depending on circumstances. Although a wide variety of sex-determining mechanisms are known, including heterogamety, various polygenic mechanisms, environmental sex determination, and haplodiploidy (Bull, 1983), these are not the basis for classification in the sex-allocation literature (Charnov, 1982). Rather, descriptive patterns of individual behavior are used. Using the sex-allocation literature as a model, we can rename alternative strategies and tactics as either irreversible (fixed over the adult lifetime of the individual) or reversible (facultative). Reversible patterns may be simultaneous when individuals can readily change back and forth between patterns or sequential when individuals switch from one pattern to another at a particular point during their lifetimes (Table I). Henson and Warner (1997) and Taborsky (1998, 1999) also categorize alternative phenotypes in this way for similar reasons. Caro and Bateson (1986) criticize the reversible/irreversible dichotomies that are often used in behavioral studies by pointing out that an observer cannot be certain that an animal will never switch, or once switched cannot change back, given the right circumstances. Although this is true for any one individual, by studying a population of individuals it is possible to assign most patterns to one or another of these categories (although some doubt may always exist).

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

9

When alternative phenotypes are grouped into descriptive categories as shown in Table I, three additional points become clear—points that are well known in the sex-allocation literature. First, the term “conditional strategy” refers to all categories: conditions (environmental, social, individual) can affect whether individuals take up different strategies throughout their adult lifetime (as with environmental sex determination) or whether they switch from one pattern to another (as with protandrous sex change). For this reason, “conditional strategy” or “condition-dependent strategy” is a particularly confusing term (and variants on the use of this term confuse its meaning further, see Gardner et al., 1987). Gross (1996) argues that, over the past 20 years, research has demonstrated that alternative and mixed strategies have generally turned out to be a complex set of condition-dependent strategies. In fact, he doubts whether mixed or alternative strategies exist at all. This view has arisen because we have had the expectation that alternative strategies and mixed strategies should not be affected by environmental conditions. Reasoning from sex-allocation theory, however, where no such expectation exists, we realize that environmental conditions, individual age or condition, population density, etc., are very likely to influence any pattern (if fitness increases from using such information) regardless of the sex-determining mechanism. In fact, the sex-allocation literature has been revolutionized by the realization that most sex ratios, once thought to be fixed (Williams, 1979), are, in fact, variable and that individuals adjust their sex-allocation decisions in response to environmental or individual conditions (Clutton-Brock and Iason, 1986; Kruuk et al., 1999). Second, frequency dependence is an intrinsic part of all sex-allocation theory including condition-dependent patterns. This suggests a possible role for frequency dependence in all alternative strategies and tactics, including those that are condition dependent or based on environmental cues. Third, combinations of sex-allocation patterns are well known in some groups. For example, gynodioecy occurs when some individuals are female and some are hermaphroditic (Charnov, 1982) and, in some species of fish, both simultaneous and sequential hermaphrodites occur within one population (St. Mary, 1994). These examples suggest that combinations of patterns may also be found in studies of alternative phenotypes. Descriptive, rather than mechanism-based categories, then, allow us to see new patterns and relationships. C. DECISION MAKING, TRADE-OFFS, AND ALTERNATIVE ALLOCATION PATTERNS Both sex-allocation and alternative strategies and tactics belong to a much broader class of problems known as “allocation strategies” (Lloyd, 1987). Their common features are (1) that time or resources are invested in two

10

H. JANE BROCKMANN

or more different activities that lead to the same functional end; (2) that individuals must have rules (called “decisions” in this literature, not implying any particular mechanism such as fixed rule, learning, or even rational evaluation) about how to allocate their time and resources among the alternatives (if they are in any sense mutually exclusive); and (3) that time or resources invested in one activity are time or resources that could be put into another (e.g., there is a trade-off between son and daughter production in sex-allocation theory). These are common themes of behavioral and evolutionary ecology (Charnov, 1997). For example, life-history studies examine how animals allocate their time and resources between growth and reproduction or present and future reproduction; the optimal foraging literature asks how individuals allocate time among patches, and within a patch; studies of reproductive behavior focus on how individuals allocate resources among mates and between parental care and additional matings, or how they divide their parental care among offspring; studies of social behavior develop models about when an animal should join a group or remain alone or how an individual should allocate resources between relatives and self (Rubenstein, 1980; Lloyd, 1987). In general, these studies have assumed that selection favors those allocation rules that maximize long-term reproductive success. The expectation is that there will be one best way of achieving that success. But in many species two or more allocation patterns exist (e.g., Alcock et al., 1977; Cade, 1981; Greenfield and Shelly, 1985; Arak, 1988; Kukuk and Schwarz, 1988; Waltz and Wolf, 1988; Arnqvist, 1992; Radwan, 1993; Alcock, 1997). For example, in a number of species, some males invest heavily in weapons and male–male combat, whereas others do not and instead acquire matings by sneaking around the fighting males (Gadgil, 1972; Rubenstein, 1980). Such variation is clearly maintained over generations and often involves the evolution of elaborate adaptations. For example, in the midshipman, Bass (1992, 1996, 1998; Brantley and Bass, 1994; Goodson and Bass, 2000) has shown that nearly every aspect of the fish’s morphology, physiology, and behavior differs depending on whether it is a singing/nest-guarding male or a nonsinging/sneaker male. What are the selective pressures and evolutionary processes that favor such alternative phenotypes?

II. HOW ARE ALTERNATIVE ALLOCATION PATTERNS MAINTAINED IN POPULATIONS? In the most general sense, only one phenotype will be maintained in a population when the fitness of that phenotype is higher than that of any alternative that has arisen in the population (Fig. 3a). Two phenotypes will

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

11

FIG. 3. General model for the maintenance of variation in populations. (a) Shows one phenotype, and (b) shows multiple phenotypes. A and B refer to the fitness curves (or gain curves) for two morphs or allocation patterns within one species. The fitness curves may be any shape, but as long as A remains above B for all values, B will not invade the population as shown in (a). However, if the fitness curves cross, then two patterns will be maintained in the population at one time as shown in (b). A number of different factors can cause fitness curves to cross, including frequency dependence, external factors such as season and density, and internal factors such as age and condition.

be maintained when fitness curves (gain curves) cross (Fig. 3b). What would make fitness curves cross? A wide array of possibilities exist. These include conditions that favor different patterns under different individual states, such as age or body condition, or under different environmental situations, such as density, temperature, or food plants. Selective pressures can change temporally, causing fitness curves to cross along a time scale (e.g., over a season), or selective pressures can differ between habitats, causing fitness curves to cross along a spatial scale. Frequency dependence is a particularly important process because it can act to stabilize frequencies of alternative phenotypes in populations.

12

H. JANE BROCKMANN

A. FREQUENCY-DEPENDENT SELECTION 1. Frequency Dependence and Sex Allocation Variation may be maintained in populations by negative frequencydependent selection (Gadgil, 1972; Hamilton, 1979; Dawkins, 1980; Rubenstein, 1980; Heino et al., 1998). This occurs when the success of a phenotype depends on its frequency or proportion in the population (Fig. 3b), which results in a stable mixture of phenotypes at the crossover frequency. When the population is at this frequency, the two patterns are equally successful. If phenotype A drops for some reason, then its fitness increases and the frequency of A rises again in the population. If A becomes too common, its fitness drops and the frequency of A returns to the equilibrium. Frequency dependence is the same evolutionary process that maintains equal investment in sons and daughters, that is, a sex ratio of 50:50 (males: females) in most species. Because every individual in the next generation has exactly one father and one mother, we know that success through son production is, on average, equal to success through daughter production. If any rule for producing offspring results in too many sons, then selection favors a different rule, one in which investment in sons more nearly equals investment in daughters (Trivers and Hare, 1976; Boomsma, 1991). For example, in a number of species of arthropods, a portion of the population is infected with a virus that makes the infected individuals produce only daughters, so selection favors uninfected individuals producing a male-biased sex ratio (Taylor, 1990). Frequency-dependent selection, then, acts to modify individual sex-allocation decision rules as long as the success through son and daughter production is equal. When it is not equal, as is true under local mate competition or local resource competition, frequency dependence continues to operate but the result is stable female-biased or male-biased sexallocation patterns that match the ratios of success per investment through son and daughter production (Charnov, 1982; Bourke, 1997; Beekman and van Stratum, 1998). The similarities between alternative strategies and sex-allocation patterns are particularly clear when we consider the case of a female faced with a decision about how much to invest in sons and daughters (Brockmann and Grafen, 1992) and in large and small males (Alcock, 1996a,b,c; Simmons et al., 2000). Her investment in those males will depend on her fitness through large and small male production. If the success of large and small males is frequency dependent, as is likely (small males do well when they are rare relative to larger males), then selection favors females that allocate resources in proportion to the ultimate success of the two types of males. She may tinker with her allocation pattern depending on conditions, but if individual decision rules result in a population-wide shift away from the equilibrium

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

13

frequency of large males, then selection will act on the decision rule, which returns the population to the ESS. Like much of sex allocation, the parent controls the allocation pattern for the next generation, so relative success must be measured from the parent’s point of view (and her trade-offs, i.e., reproductive value through son and daughter production) and not just as offspring fitness (Dawkins, 1980). 2. Frequency Dependence and Polymorphism Frequency-dependent or balancing selection is thought to explain most cases of polymorphism (Majerus, 1998). The underlying processes that account for the frequency-dependent effects are of several different types. (1) Recent studies have demonstrated that predation is often frequency dependent, that is, that predators prefer the more common prey phenotypes (apostatic selection) and that such behavior may have the effect of maintaining a polymorphism in the population (Endler, 1988; Roff, 1996). For example, Cain and Sheppard (1950, 1954) showed that thrushes are more likely to prey on the most common color forms of the highly polymorphic land snail Cepaea nemoralis. This behavior means that a rare color morph, even though it is not overtly more cryptic than the common morph, will have an advantage. The reason that common morphs are more likely to be preyed on is less clear and several explanations have been proposed, including the formation of a search image, effects on the rate of searching for prey, changes in the rate of handling prey, and neophobia (Guilford and Dawkins, 1991; Shettleworth, 1998). (2) Arms races between species can also result in frequency-dependent selection and the maintenance of polymorphism. For example, in the coevolutionary arms race between parasites and hosts, selection favors parasites that are adapted to the more common host genotypes, which will mean that rarer host forms will have an advantage (Sheldon and Verhulst, 1996; Wakelin and Apanius, 1997; Potts et al., 1997). The same principle applies in Batesian and Mullerian ¨ mimicry (Lindstrom ¨ et al., 1997). Rare polymorphic forms may gain an initial advantage from being similar to a distasteful form but as their frequency rises the advantage deteriorates (Mallett and Joron, 1999). (3) Competition within a species may also explain polymorphic phenotypes (Skulason ´ and Smith, 1995; Smith and Skulason, ´ 1996). For example, a number of species of insects show partial bivoltinism, a pattern in which some individuals produce two generations and some produce only one generation per year (Seger and Brockmann, 1987; Brockmann and Grafen, 1992). One explanation for this variable life-history pattern is competition among females for oviposition sites (Rauscher, 1986). The success of the rare individual who enters diapause and emerges in the following year is higher than the success of individuals who produce offspring that develop, emerge, and compete in the

14

H. JANE BROCKMANN

same season. (4) Sexual selection and nonrandom mating can also account for polymorphism (Majerus, 1986; Sinervo and Lively, 1996; Gillespie and Oxford, 1998). Although rare male effects have been described in a number of species (Partridge and Hill, 1984; Partridge, 1988), the underlying mechanisms for such preferences are not known. Possibilities include female preference for or habituation to cues produced by the more common types, female preference for dissimilar genotypes, and male–male competition. In addition, the advantage to the female for showing rare male preference (e.g., outbreeding) has not been established. The point here is that these explanations for the evolution of polymorphism based on nonrandom mating may apply to other cases of alternative phenotypes. For example, female mimicry has been described as behavior by a male that allows him to gain closer access to females or resources than he would have if he were to act like a male; rival males are duped into treating the female-mimicking male like a female (Thornhill, 1979; Dominey, 1980, 1981; Forsyth and Alcock, 1990). Frequency dependence means that such behavior will be most effective when rare. Heterozygous advantage has also been used to explain balanced polymorphism. If differences in behavior are due to the presence of two alleles at one locus and heterozygous individuals have the highest fitness (heterosis), then both alleles and their associated polymorphic phenotypes will be maintained in the population (Gadgil and Taylor, 1975). Furthermore, if females prefer heterozygous males (Rubenstein, 1980; Weatherhead et al., 1999) or males that create heterozygous genotypes in their offspring (Potts and Wakeland, 1993; Wedekind, 1994), then variation can be maintained (Brown, 1997). Despite its promise as an explanation, relatively few examples exist of heterozygous advantage maintaining polymorphism (Cook and Gao, 1996; Majerus, 1998). 3. Testing Frequency Dependence Frequency dependence is widely regarded as an important explanation for the maintenance of variation in the alternative strategy, sex-allocation, color polymorphism, and mimicry literatures (Rubenstein, 1980; Hedrick, 1986; Brockmann, 1986; Majerus, 1998). Despite its popularity as an explanation, few studies have evaluated frequency dependence (Hori, 1993; Roff, 1996; Giraldeau and Livoreil, 1998) even within the sex-ratio literature (Conover and van Voorhees, 1990; Basolo, 1994). The best test is one in which frequencies are altered and a population-level response is observed, but such large-scale experiments are difficult to conduct (Hamilton, 1979). Nonetheless, frequency dependence holds a special place in explaining the evolution of variation because of its ability to maintain patterns at stable frequencies

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

15

over generations (as long as the fitness curves do not change). But other processes may also cause fitness curves to cross.

B. BEHAVIORAL SWITCHES DUE TO ENVIRONMENTAL, SOCIAL, AND INDIVIDUAL EFFECTS Perhaps the most common means by which fitness curves cross are behavioral decision rules based on environmental or condition-dependent switches (Fig. 3b). Differences in day length, temperature, crowding, and the presence of predators are all known to affect the development of different colors, life histories, sizes, or forms of arthropods (Tauber et al., 1986; Greene, 1999; Taborsky, 1999). For example, caterpillars may develop into different colors or entirely different forms when feeding on different host plants (Fink, 1995; Greene, 1996). Migratory locusts change from a solitary, cryptic form (pale yellow, green, or brown) to a gregarious, conspicuous, aposematic form when they develop under crowded conditions (Pener, 1991; Sword et al., 2000; Sword and Simpson, 2000). Sex-allocation decisions may similarly be influenced by individual condition and environmental factors (Trivers and Willard, 1973; Bull, 1981; Charnov, 1982; Brockmann, 1986). For example, females of the parasitic wasp Lariophagus parasitize beetle pupae of different sizes, laying sons in relatively smaller pupae and daughters in relatively larger pupae (Charnov et al., 1981). Females that use this facultative rule (switching from son to daughter production when parasitizing prey of larger sizes) have higher success than those who are equally likely, for example, to lay sons in hosts of all sizes. Parker (1982) was one of the first to argue that if phenotypic differences correlate with differences in fitness, then selection will favor a strategy switch at a threshold T and the ESS will be to play Y below T and play X above T. A number of others have developed similar models for a variety of systems (Oster and Wilson, 1978; Barnard and Sibly, 1981; Waltz, 1982; Parker, 1984; Gross, 1984, 1991a,b; Walker, 1986; Vickery et al., 1991; Moran, 1992; Koprowski, 1993; Radwan, 1993; Clark, 1994; Hutchings and Myers, 1994; Roff, 1996; Henson and Warner, 1997; Reeve, 1998; Greene, 1999; Taborsky, 1999). The earlier literature on alternative strategies and tactics saw conditional strategies as those in which frequency-dependent selection was not operating (Davies, 1982). Conditional strategies were seen as less successful alternatives and as just something that happened to the animal (Eberhard, 1982; Austad, 1984). This view has now been replaced, however, because advantages to alternative phenotypes have been discovered, stable frequencies of alternatives have been observed, and frequency dependence has been found to act on decision rules that involve behavioral switches (Parker, 1982;

16

H. JANE BROCKMANN

Waltz, 1982; Swingland, 1983; Arak, 1984; Lundberg, 1988; Roff, 1994, 1996; Henson and Warner, 1997; Johnson et al., 2000). For example, sex-allocation decisions are certainly influenced by frequency-dependent effects. When the rule that causes female Lariophagus to lay daughters in large hosts and sons in small hosts results in excessive numbers of sons, frequency-dependent selection favors a different rule that results in more equal numbers of sons and daughters (Charnov et al., 1981; Brockmann, 1986). Recently, Repka and Gross (1995; Gross and Repka, 1998) have shown that the fitness associated with alternative tactics within a population depends on both their frequency and the phenotypic condition of the individuals of that population. There will be a unique ESS switch point that determines both the condition at which an individual switches between tactics and the resulting stable frequency of tactics in the population. At this ESS the fitness of the alternative tactics will be equal. When fitness is averaged over the population, however, the average success of the alternative tactics will not be equal. Density dependence is a common environmental effect that can cause fitness curves to cross: if one morph does better at low densities whereas the other has higher fitness at high densities (Fig. 3), then the two patterns can be maintained in the population, as long as densities vary regularly across the range in which both patterns are favored. A string of years at low densities, however, and the high-density adapted pattern would be driven to extinction. Dunbar (1982) argues that there is a close interaction between frequency and density dependence in many systems because a rare-type advantage can change to a rare-type disadvantage (or vice versa) as densities rise (Radwan, 1993; Rowell and Cade, 1993). Also, the costs of competition invariably rise with increasing density, which means that the success associated with frequency-dependent effects will change. Density may act to set the equilibrium point for a given set of strategies and frequencydependent processes may then act to stabilize the population at that level (Arak, 1983). C. SPATIAL AND TEMPORAL VARIATION Spatial and temporal variation may also maintain genetic polymorphism in populations (Seger and Brockmann, 1987). Stability can be maintained in a spatially heterogeneous environment with migration between locally adapted subpopulations but only under rather specific conditions (Maynard Smith and Hoekstra, 1980; Via and Lande, 1985; Hedrick, 1986; Moran, 1992; Sandoval, 1994; Roff, 1996). Riechert (1986) provides a nice example of how variation in aggressive behavior in a species of spider is maintained by different selective pressures in two nearby localities with frequent migration

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

17

between sites. Another well-known example is when melanic forms of a species are favored in industrial areas and nonmelanic forms in rural areas; the frequency of the two morphs in each area is the product of selection and the degree of migration between areas (Majerus, 1998). Periodic environmental fluctuations can also maintain diverse phenotypes in a population as long as each pattern has higher success than the others under at least one set of environmental conditions and the frequency of environmental fluctuations (relative to generation length) is such that extinction of one type does not occur (Rubenstein, 1980; Roff, 1996). This is particularly likely when generations overlap (Ellner and Hairston, 1994) and there is delayed diapause (Hedrick, 1995). However, if the animal can detect the sort of environment in which it is living and respond to that variation by changing phenotype, then switches will be favored (Dingle, 1984; Walker, 1986; Moran, 1992; Bronmark ¨ and Miner, 1992; Ellner and Hairston, 1994). In addition, in some systems, such as with alternative diapause or dispersal patterns, frequency dependence may also act to maintain the variable phenotypes in the population (Parker and Courtney, 1983; Swingland, 1983; Brockmann, 1986; Walker, 1986; Seger and Brockmann, 1987; Kaitala et al., 1989; Hutchings and Myers, 1994; Roff, 1996).

III. CASE STUDIES OF ALTERNATIVE ALLOCATION PATTERNS Now that the basic processes have been described, let us turn to some detailed examples of different types of discrete, alternative phenotypes (Table I). From these studies I will draw generalities about the factors that favor the evolution of alternative allocation patterns. A. IRREVERSIBLE OR ADULT LIFETIME ALLOCATION PATTERNS 1. Color Polymorphism in Damselflies Adult female Ischnura ramburi damselflies (Odonata: Zygoptera, Coenagrionidae) come in two color morphs, one colored and patterned like males and referred to as andromorphic (or androchromic) and the other different from the males and referred to as gynomorphic (gynochromic or heteromorphic) (Robertson, 1985). The inheritance of these color morphs is unknown in I. ramburi, but in several other species of Ischnura, the color morphs are due to two or three alleles at a single locus (Johnson, 1964, 1966; Cordero, 1990). In I. ramburi, mating is time-consuming and probably quite costly, taking on average 6 hours (Sirot and Brockmann, 2000). Damselfly mating begins when the male grasps the female’s neck with a pair of claspers located

18

H. JANE BROCKMANN

at the tip of his abdomen. The female then usually bends her abdomen up to the male’s penis, which is located on his second abdominal segment. In most ischnurans, the male first removes any previous male’s sperm from the female’s sperm storage organ (Waage, 1979; Cordero and Miller, 1992) and then replaces and packs in his own sperm (Robinson and Novak, 1997), so strong last male sperm precedence occurs (Waage, 1986; Cordero and Miller, 1992; Cooper et al., 1996). After copulation, the two separate and the female later lays eggs unattended by the male, inserting her eggs into the stems of submerged vegetation. Behavioral differences accompany the differences in color: andromorphic females are more malelike in color, pattern, and behavior and males respond to andromorphs and gynomorphs differently (Robertson, 1985; Sirot and Brockmann, 2000). Males pounce on gynomorphs, wrestle, and attach (Robertson, 1985) and the female is likely to respond by fleeing or hiding (Fig. 4). Andromorphs, on the other hand, are more likely to confront the approaching male, facing off with him, and the male usually responds by flying away as he would when confronted by another male (Robertson, 1985). So the two morphs show distinct behavioral differences although they do not differ in size or longevity (Sirot, 1999). The selective pressures favoring the maintenance of two color patterns in this species are something of a mystery but the current explanation combines male mimicry and sexual conflict (Robertson, 1985; Hinnekint, 1987; Forbes, 1994). When the frequency of andromorphs is low, that is, when they are rare, they do particularly well because they are not easily detected by males and are thus able to avoid costly, long matings that interfere with feeding and egg laying (Fig. 4). As their frequency rises, males are increasingly likely to detect the male mimics and so their fitness drops. The model suggests then that the two morphs are maintained at an equilibrium frequency by crossing fitness curves and frequency dependence. But there is an added effect. According to this hypothesis, andromorphs are particularly favored over gynomorphs at high densities as long as they remain rare because at high operational sex ratios male harassment is particularly costly. At low densities, however, andromorphs do worse when they are rare because males do not respond to them as females and they are more likely to go unmated (Fig. 4). So the hypothesis combines frequency-dependent and density-dependent effects. Testing this hypothesis has proven rather difficult (Fincke, 1994a,b), although there is some support from various polymorphic, related species (Robertson, 1985; Cordero, 1992; Forbes et al., 1997; Cordero et al., 1998). For example, as predicted there is a correlation between andromorph frequency and male density (Forbes et al., 1995; Cordero and Egido, 1998) and van Gossum et al., (1999) have shown that males choose to mate with the more common female morph. Sirot and Brockmann (2000) have experimental laboratory data for

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

19

FIG. 4. Gynomorphic and andromorphic females of the damselfly Ischnura ramburi differ in both color and behavior. The gynomorphic females are tan and often hide along grass stems, whereas andromorphic females have the same color patterns as males and perch in the open more often. The model explains how the two color morphs might be maintained in the population. When andromorphs are rare, they have an advantage over gynomorphs (top graph), particularly in high-density populations (bottom graph) as long as they remain rare (based on data presented in Sirot and Brockmann, 2001.)

I. ramburi to support the general shapes of the fitness curves, including the assumption that multiple matings are costly to andromorphs. What is now needed is an experimental manipulation of the frequencies, densities, and operational sex ratios of natural populations to see if the predicted changes

20

H. JANE BROCKMANN

occur in relative fitness (and subsequent changes in morph frequencies) of the two morphs. 2. Size Polymorphism in Male Isopods and Dung Beetles The marine isopod Paracerceis sculpta has three male morphs known to be controlled by three alleles at one locus (Shuster and Sassaman, 1997). The large, slow-developing, and ornamented ␣-males establish themselves at the entrance to a sponge, which they guard from intrusions by other males (Shuster, 1989a). Females that are about to molt are attracted to these sponges, sometimes in large numbers (Shuster and Wade, 1991a), where they apparently mate with any available male (Shuster, 1989a). The much smaller ␤-males are also attracted to the sponges, which they can enter without a fight, presumably because they resemble females in form and behavior (Shuster, 1992). The tiny, rapidly developing, and nonornamented ␥ -males are also attracted to the sponges but they enter and mate only by sneaking around the ␣-males. When few females are present ␣-males defend their sponges effectively, but when females aggregate, ␤- and ␥ -male success increases sharply (Shuster, 1989b). Female density, the availability of sponges, operational sex ratios, and the frequency of the three morphs interact to influence the success of males. Shuster and Wade (1991b) show that the three morphs are approximately equally successful and are found at Hardy– Weinberg equilibrium frequencies. Many lifetime alternative phenotypes are not associated with a genetic polymorphism. For example, males of some beetle species have large horns on the head or thorax that are used in combat. In some species males occur in long- and short-horned forms (Eberhard, 1979; Eberhard and Gutierrez, 1991). In Onthophagus beetles the female digs a burrow in the ground below a pile of dung, where she remains for a number of days provisioning brood cells for her offspring (Emlen, 1994). A large, horned male mates with her and guards her tunnel from intrusions by other males, whereas small, hornless males try to sneak past or dig around the guarding males, thus bypassing the guard and mating with the female (Emlen, 1997a, 2000a). Developmental studies show that differences in horn length depend on the size of the larva just prior to pupation (Emlen and Nijhout, 1999) and that the control of horn length involves a sharp threshold switch such that individuals above a certain size develop very large horns and those below a certain size have no horns, with very few individuals bearing horns of intermediate size (due to disruptive selection against the intermediates) (Fig. 5; Emlen, 1994). Emlen (1996) has shown that there is genetic variance on this threshold rule such that it responds to selection. He has also shown that the rule is flexible depending on local conditions that affect the likelihood that

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

21

FIG. 5. The dung beetle Onthophagus acuminatus comes in two morphs, small hornless males and large horned males. The top figure shows a frequency distribution for horn length in one population (inset) and the rapid switch from small to large horns when individuals reach a particular size. The bottom figure is a model for the maintenance of the two morphs in the population. Small individuals have higher fitness when hornless and large individuals are more successful when horned. (Top figure modified from Emlen, 2000b.)

males will have to fight other individuals (Emlen, 1997b). Emlen’s studies demonstrate that fitness curves cross: large males do better with large horns because of their improved ability to fight; small males do better with no horns because of their improved ability to dig and move around in burrows. Furthermore, large males with no horns and small males with large horns are not as successful at either activity (Fig. 5). Because the success of small males depends on both their frequency and the density of females, frequency- and density-dependent effects are operating in this system as in most others that we have discussed.

22

H. JANE BROCKMANN

3. Generalities about the Evolution of Irreversible Patterns The three examples of irreversible allocation patterns are from diverse systems, yet all illustrate similar processes in action. Why do animals evolve these irreversible patterns rather than what would seem to be a more adaptable, facultative behavior? This same question has been posed in the sexallocation literature: why are all animals not hermaphrodites? As Charnov (1982) argues, the most general answer to this question is that dioecy evolves when individuals who switch sexes have lower fitness than those who remain as one sex. Lifetime patterns allow complex suites of adaptations in association with male or female function, such as size, color, morphology, physiology, and behavioral and life-history differences, which are all coordinated through physiological and developmental processes. The distinguishing feature is that these complex suites of characters are closer to the optimum than are the traits that are produced by individuals that switch (Dewitt et al., 1998). This can occur when the costs of plasticity are high, such as the cost (in time or resources) of acquiring the necessary information to make an adaptive switch, or when there are limits to plasticity (Dewitt et al., 1998), such as in the accuracy of matching phenotype to environment (Moran, 1992). Lifetime, alternative allocation patterns evolve when the gains from switching are less than the benefits that would be achieved by remaining as one phenotype throughout the adult lifetime (Henson and Warner, 1997). B. REVERSIBLE PATTERNS: SIMULTANEOUS STRATEGY ALLOCATION In simultaneous hermaphrodites, individuals change flexibly from one pattern to another and back again. Similar allocation patterns can be found in other forms of behavior, such as in alternative patterns of nesting or mating. 1. Nesting Strategies in the Great Golden Digger Wasp Female Sphex ichneumoneus are large, solitary wasps (Sphecidae) that dig burrows in the ground that they mass provision with paralyzed katydids hunted down in the surrounding vegetation (Brockmann et al., 1979; Brockmann and Dawkins, 1979). It takes most of a day for a female to construct a nest and a couple of days to find, capture, and transport back to the nest the several large katydids that are necessary for rearing the single egg that is laid in each brood cell. Because productivity is low, with the most successful females laying only about six eggs during her 4- to 5- week lifetime, Brockmann and Dawkins (1979) argue that this is a system in which the time a female can save in digging is time she can put into hunting. We found that females sometimes construct burrows of their own and sometimes utilize the burrows that other females construct (Brockmann et al.,

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

23

1979). When they dig burrows on their own, they sometimes abandon nests that are later used by other females. When females utilize other females’ burrows, they sometimes end up in a good abandoned burrow but on occasion they end up in the nest of another female. In this species such nest sharing is costly because the two females fight and duplicate each others’ efforts (Brockmann and Dawkins, 1979). They seem unable to detect each other’s presence and simply continue to provision as though they were alone until they happen to return at the same time when they engage in a long, and sometimes damaging, fight and one ousts the other from their joint nest (Dawkins and Brockmann, 1980). We tried hard to determine the cues the females might use to choose nests. For example, we thought it likely that a female first looked for an abandoned nest and, finding none, dug a nest of her own. If this were the case, then the interval between completing the last nest and occupying an abandoned burrow would be less, on average, than that between the last nest and starting to dig a new burrow. Our data did not support this hypothesis or, in fact, any other that we tried. Brockmann, Grafen, and Dawkins (1979) capture the essential elements of the wasps’ behavior in a model that portrays the system as a market economy of burrows, with females sometimes constructing nests and sometimes entering nests (Fig. 6a). We hypothesize that the wasps are making the decision about whether to dig or enter at random because we did not detect any cues that influence their choices (Brockmann and Dawkins, 1979). We postulate that this random decision rule in the individual about digging and entering is set by frequency-dependent selection acting on the population. The frequency-dependent effect is based on the market economy of burrows: When diggers are rare, enterers do poorly because there are few abandoned burrows and they end up in a costly joint nest with another female, but when enterers are rare they do well because there are lots of unoccupied burrows around and they save the costs of digging. Of course, individuals might alter these frequencies based on immediate circumstances (although we had no evidence that they did), but if any individual decision rule results in too many enterers relative to diggers, then enterers would do worse and selection would favor a different decision rule that would lower the frequency of entering in the population. In this way, we argue, the two patterns expressed within each individual are maintained at a stable equilibrium in the population. Several researchers have criticized this study because it seems to them unlikely that an animal would simply choose to dig or enter at random and not use the available information (Caro and Bateson, 1986; Field, 1989; Wolf and Waltz, 1993; Gross, 1996). I agree. However, we tried in every way we could to show that females base their decision on some environmental condition or individual attribute: digging and entering are not characteristic of

24

H. JANE BROCKMANN

FIG. 6. Model of golden digger wasp nesting strategies. In the top figure, the decision structure and outcome of the alternative decisions are shown, and in the bottom figure the hypothesized frequency-dependent effect is shown. Females may either dig a nest or enter an already existing nest. Entering does well when it is rare and poorly when it is common. (Top figure modified from Brockmann et al., 1979.)

particular individuals, there are no seasonal effects, there is no correlation with an individual’s size, they are not choosing to dig or enter on the basis of past success or on the basis of how long they have been searching (Brockmann and Dawkins, 1979). Furthermore, when two females meet, they fight in relation to how many prey they have brought to the nest and not based on the true value of the nest, that is, the total number of prey in their joint nest (Dawkins and Brockmann, 1980), suggesting that one female does not know that the other is present in her nest. This result is made all the more surprising given subsequent studies that show that some sphecid wasps clearly brood parasitize the nests of conspecifics (Brockmann, 1980; Field, 1989, 1992). A number of alternative suggestions have been made, such as that females

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

25

base their dig or enter decision on some unmeasured variable or that our measurements are not accurate enough to detect a difference (Field, 1989; Wolf and Waltz, 1993) or that some unexamined constraint exists such as a limitation on egg-laying rate (Rosenheim et al., 1996) or energy to expend on hunting (Fagerstrom, ¨ 1982), any one of which would cause us to reexamine the data should support be found. Caro and Bateson (1986) suggest that the dig-or-enter decision is based on frequency-dependent choice, which seems unlikely given that the females seem unable to detect other females in the nest. Experimental manipulations that are designed to evaluate the underlying assumptions and decision mechanisms are surely needed to resolve these questions. Our model (Fig. 6) should be tested experimentally by manipulating the frequencies of the two patterns (dig and enter) to see if their relative success changes or by manipulating the success to see if the frequencies of dig and enter change over generations. 2. Simultaneous Hermaphrodites and Generalities about Simultaneous Allocation Patterns Reversible strategies such as those shown by the digger wasps have a sexallocation counterpart in simultaneous hermaphroditism (i.e., species that switch genders during a bout of mating). For example, seabass such as hamlets (Hypoplectrus) mate in pairs, change male and female roles after each spawning, and, in effect, take turns releasing eggs for the partner to fertilize (egg trading) (Fischer, 1988). In the tobaccofish (Serranus tabacarius, another seabass) large individuals specialize to some extent in male function but can also reproduce by egg trading (Petersen, 1995). Mating in freshwater pulmonate snails (Physa) also involves size asymmetries with smaller sperm donors and larger sperm recipients, although both are capable of reproducing as either gender (DeWitt, 1996). Clearly, simultaneous hermaphroditism is not common in the animal world and it can occur only when male and female function does not require morphological, physiological, and behavioral adaptations that preclude the animal’s performing well in either role, that is, only when it does not pay to specialize (Fischer, 1980; Charnov, 1982; Herre et al., 1987). When success through male function is constrained in some way (or when selfing is advantageous) or when sex change is so costly that it pays individuals to maintain both types of gonadal tissue (St. Mary, 1997), simultaneous hermaphroditism is favored. This occurs when the value of being one sex or the other changes frequently and unpredictably (such as with the size of the partner) (Charnov, 1979; DeWitt, 1996). Frequency dependence remains a part of the equation for outbred species, as it does with all sex-allocation problems (Charnov, 1996). Building from the hypotheses developed for simultaneous hermaphrodites, reversible allocation patterns should evolve either when no specialized

26

H. JANE BROCKMANN

adaptations (such as specialized morphology) are needed or when selection favors the animal investing in what is necessary to perform in either role. This is most likely when the environmental conditions that favor the two patterns are unpredictable and change frequently. A mixed strategy of the digger wasp sort can occur only under the more stringent conditions that the individual does not have information about what affects its fitness, yet must choose between exclusive alternatives and when frequency-dependent selection is operating (West-Eberhard, 1979; Maynard Smith, 1982; Dominey, 1984). If individuals have information, that is, cues relevant to fitness, then selection will strongly favor their using that information and taking up the appropriate pattern (Parker and Rubenstein, 1981; Dunbar, 1982), and this is surely what the vast majority of animals do (Gross, 1996; Taborsky, 1994, 1998, 1999). For example, while in mass mating aggregations, small males of the giant cuttlefish (Sepia apama) repeatedly assume the body shape and color patterns of females, which prevents attack by the larger males and often allows them to mate successfully with females (Norman et al., 1999). However, there are situations in which information about fitness is unreliable (or too expensive to acquire) and yet the animal must allocate time or resources to mutually exclusive alternatives. When this occurs or when frequency-dependent selection is operating, as is hypothesized in the digger wasp case and in sex-allocation systems, a coin-tossing (mixed) strategy may be favored.

C. REVERSIBLE PATTERNS: SEQUENTIAL ALLOCATION PATTERNS Let us now turn to the third set of possibilities, the situation in which multiple phenotypes arise because two or more patterns occur within the same individual at a particular point or state in that individual’s life, that is, alternative phenotypes that occur sequentially over the adult life of the individual. 1. Alternative Strategies in Horseshoe Crabs Horseshoe crabs, Limulus polyphemus, nest along the east coast of the United States on sandy beaches where pairs spawn either in the company of numerous satellite males or as isolated pairs (Brockmann, 1990). Some males swim offshore, searching for females, and attach as a female swims within range (Barlow et al., 1988). Males use their first pair of legs, which are modified into claws, to grasp the female’s terminal spines in a kind of amplexus and the pair travels together to the nesting beach (Barlow et al., 1986). Other males come ashore without females and position themselves around the nesting couples (Brockmann, 1996; Hassler, 1999). The female

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

27

buries herself at the top of the high-tide line (Brockmann and Penn, 1992) and lays her eggs directly into the sand, but only if a male is attached to her terminal spines (i.e., only if an attached male is present). Males fertilize eggs externally as they are being laid with aquatic, free-swimming sperm. When the pair nests as a lone couple, the attached male fertilizes all of the female’s eggs, but when satellite males are present, sperm competition occurs (Brockmann, 1990). We found that when one satellite male is present, he fertilizes on average 40% of the eggs laid (when he is in the best position over the female’s incurrent canal; Brockmann et al., 1994); when two males are present, each fertilizes 40% of the eggs (when in the best location), leaving the attached male with only 20% of the fertilizations; and when more satellites are present, they take away from each others’ paternity (Brockmann et al., 2000). By laying eggs at the top of the tide line on the new and full moon high tides, females are placing their eggs in the optimal environment for egg development (Penn and Brockmann, 1994). The eggs develop into trilobite larvae and then metamorphose into tiny horseshoe crabs 2 to 4 weeks later; after 9 to 10 years they undergo a terminal molt and begin to reproduce (Shuster, 1982) and may then live 8 years or more (Botton and Ropes, 1988). Females probably lay eggs only for a few days each year in a series of spawnings associated with one tidal cycle, but males return over several weeks, which is the likely explanation for the strongly male-biased operational sex ratios in many populations (Brockmann and Penn, 1992). Attached and satellite males differ in a number of important respects. Satellites are darker, are more worn, have more benthic organisms attached to their carapace, and are probably older, on average, than attached males (Brockmann and Penn, 1992). When they come ashore, the older satellite males are more likely to be overturned by waves and less likely to right themselves than the younger attached males that are in better condition (Penn and Brockmann, 1995). Performance tests also indicate that darker males pair more slowly, are less likely to find a female when released unattached, and are less likely to remain attached when paired. These results indicate that the alternative mating behavior patterns of horseshoe crabs are condition dependent. Two types of condition-dependent rules exist. It could be that all males follow one rule (attach to females) and that the observed differences between males result from differences in their ability to find and hold onto females (young males hold on, whereas old males cannot and so come ashore as satellites). This would mean that satellite behavior is a making-the-best-of-a-badsituation tactic. Alternatively, it could be that differences in male behavior are due to a decision rule, a condition-dependent switch in behavior (young males attach to females and older males come ashore as satellites). It is not easy to tell the difference between these two possibilities, but the horseshoe

28

H. JANE BROCKMANN

crabs afforded me such an opportunity. By preventing males from attaching (by covering their claws with small plastic bags), I could see whether young and old males behaved in the same way. If the animals are using a makingthe-best-of-a-bad-situation tactic, then all males should come ashore equally because none could attach. If the animals are using a decision rule, however, then young and old males should not return equally; young males should be more likely to remain at sea. The data clearly show that young males are significantly less likely to return as satellites (Brockmann, in preparation). This means that the observed differences in behavior between young and old males are not just a consequence of their condition, but rather result from a decision rule based on condition. I hypothesize, then, that males switch from behaving as attached males when they are young to satellites when they are older (Fig. 7). The model shows that young males have higher fitness by being attached than by being satellites, whereas older males do better when they are satellites. Selection sets the optimum switch point decision rule that tells individuals when to change from one pattern to the other. Three different kinds of factors may affect this switch point, in addition to the animal’s condition and the correlation between condition and fitness for the two patterns of behavior (Fig. 7). (1) When density is high, the success of attached males will drop relative to that of satellites and selection will favor males shifting to satellite behavior at a younger age (Fig. 7). This means that when most females in the population nest synchronously, satellite behavior will be favored, whereas in asynchronous populations satellite behavior will be less successful. (2) Environmental conditions such as waves, currents, and tidal flow probably alter the fitness curves for the two strategies. The best switch point may even be changing from one tide to the next depending on conditions and the expected outcomes. If males have the information and if they are physically able, selection should favor their changing back and forth between being attached and satellites depending on these conditions (or at least some males should do so). (3) Frequency dependence is also involved. Females will not lay if an attached male is not present, so the success of satellites depends on the presence of attached males. If all males are attached, then a rare satellite should do relatively well, but when satellites are abundant relative to attached males, the success of each satellite will decline as the sizes of the spawning groups increase (Brockmann et al., 2000). So superimposed on this switch point model are environmental, density-dependent, and frequency-dependent effects, all of which alter the payoffs, the equilibrium frequencies, and thus the switch point between staying at sea and becoming an attached male and coming ashore as a satellite. Testing this model (Fig. 7) is straightforward in principle, although difficult in practice. First, we need to verify the nature of the switch point decision

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

29

FIG. 7. Model of a condition-based decision rule in horseshoe crabs. Males were observed either to come to the nesting beach attached to the terminal spines of a female or to approach pairs as an unattached male and engage in sperm competition with the attached male (in the drawing two dark satellites are shown with a lighter male attached to the larger light female). The model hypothesizes that young horseshoe crabs or those in good condition have higher fitness when they attach to females at sea and accompany them to the nesting beach, whereas older males or those in poorer condition are more successful when they forego searching for females and come to the beach as satellites.

rule and the shapes of the associated fitness curves, that is, that animals in good condition have higher fitness by being attached and older animals have higher fitness by being satellite. Because animals will not normally do the wrong thing, the fitness curves must be evaluated experimentally. We can, for example, turn young males into satellites by covering their claws

30

H. JANE BROCKMANN

so they cannot attach to females and then evaluate their relative success as satellites. Second, the effects of density and various environmental factors on reproductive success of satellite and attached males will need to be measured under different conditions. Third, the effect of frequency dependence will need to be evaluated. Although this model still holds without it, frequency dependence will have the effect of maintaining attached and satellite patterns in the population over long periods. Such a study would require altering the frequencies of attached and satellite males and evaluating the effect of these changes on success. 2. Sequential Hermaphrodites, Social Insects, and Some Generalities Like the horseshoe crabs, sequential hermaphrodites show two phenotypes within an individual and they switch from one to the other at specific points in their lives. A sequential hermaphrodite reproduces early in life as one sex and then changes to another as it grows older. For example, in the hermaphroditic polychaete Ophryotrocha puerilis, small individuals are male and large ones are female. The point at which the animal switches depends on the relationship between fitness and size for individuals of each gender: large individuals have higher success as females because they have more eggs, and small individuals are more successful as males because females prefer them (Berglund, 1990). Small females and large males are less successful. Frequency-dependent selection is also operating: if individuals in the population behaved in ways that resulted in too much allocation to male function, then the success of males would drop and selection would favor a different rule that brought the population back to a more equal allocation to male and female function. An individual that switches at the ESS does better than any mutant individual that switches earlier or later (Charnov, 1993). The age or condition at which a sequential hermaphrodite switches is a rule set by a combination of individual cues, the shapes of the fitness curves, and frequency-dependent selection. Reasoning from the arguments developed for sequential hermaphroditism, if the reproductive success associated with alternative phenotypes is correlated with age, size, or environmental conditions (such as season or density), and the individual can detect and respond to them, then selection should favor individuals switching from one pattern to the other at the age or size that maximizes fitness (Waltz and Wolf, 1984; Taborsky, 1999). The cue that the animal uses to switch its behavior does not have to be the same as the factor that improves fitness; it only needs to be correlated with it (Moran, 1992; Nijhout, 1999). For example, diapause is usually cued by photoperiod even when death from freezing is the selective agent. Such anticipatory switches are expected if preparation time or a lag is required, such as the changes associated with entering diapause (Taylor, 1986).

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

31

Abrupt, age-related changes in behavior are used to organize work in social insect colonies (Wilson, 1975; Robinson, 1992). Remarkable similarities exist among independently evolved social insect species in both the pattern and the control (Fahrbach, 1997) of age-related changes. In a social insect such as the honey bee, changes in behavior occur continually throughout the life of the worker (they change from brood care to guarding to foraging as they age), and the switches from one activity to another can be abrupt, such as between nest maintenance early in life and foraging late in life (Seeley, 1985; Calderone and Page, 1996). These changes are affected by interactions with other workers and the colony age structure, mediated through the action of changing levels of juvenile hormone (Huang and Robinson, 1992, 1996). Age-related changes in behavior are associated with structural changes that appear at different life stages, such as wax gland secretions or alarm pheromone production, that allow the animal to perform specific tasks effectively. Abrupt changes in behavior have been favored by selection because intermediate conditions are less efficient. The likely explanation for age-related switches in behavior in social insect nests is that the more risky tasks are postponed until late in life (Seeley, 1985). As the individual ages, changes occur in reproductive value and in the expected gains from different activities, such as remaining reproductively active or foraging outside the nest (Jeanne, 1991). Early in life an individual can expect some direct reproduction if she remains in the nest but as she ages, her expectation of success through direct fitness drops and her expected success through indirect fitness increases so she then engages in activities that benefit the colony as a whole (Jeanne, 1991). Interestingly, individuals will change tasks if no other individuals are available to do the work (Wilson, 1984; Naug and Gadagkar, 1998), will speed up development if the colony is short of food, or will slow down development if there is no one to care for brood (Page et al., 1992; Schulz et al., 1998), but their efficiency at performing these tasks may be reduced as a consequence. In general, then, switches occur when fitness curves cross (e.g., young animals have higher fitness by remaining in the nest, whereas older individuals have higher fitness by leaving), when activities are mutually exclusive, and when there is disruptive selection (selection against intermediates) on the suites of characters associated with each pattern (Danforth and Desjardins, 1999; Taborsky, 1999). D. RECURRING THEMES These case studies reveal a number of recurring themes that point us toward a key set of factors that we must consider in studies on the evolution of alternative phenotypes. (1) You cannot separate alternative strategies into those that are genetic and those that are environmental in any of the

32

H. JANE BROCKMANN

different types of allocation patterns, a lesson learned from the study of sex allocation. Genes and environment are involved in all decisions that animals make whether reversible or not, whether a genetic polymorphism or not. (2) Patterns that are taken up for an entire adult life usually involve a complex suite of coadapted characters that require considerable developmental time or physiological adjustment. When animals switch back and forth easily between different patterns, as they do with hermaphroditism and behavioral switches, there are usually fewer alterations associated with the alternative allocation patterns, little cost, or few trade-offs and constraints. This is an argument that has been used to explain the circumstances that favor a lifetime as separate sexes over hermaphroditism. (3) In the broadest sense, two (or more) patterns of behavior are maintained in a population when fitness curves cross. The factor that is most likely to cause fitness curves to cross is a switch in behavior based on information available about the relationship between fitness and the condition, state, or status (Gross, 1996) of the animal. Condition (or state or status) can refer either to an internal condition such as age or energy reserves or to some external environmental cue such as photoperiod or weather conditions. (4) Condition-dependent alternative phenotypes (conditional strategies) are of two types: those that are a consequence of conditions (making-the-best-of-a-bad-situation) and those that result from evolved decision-making rules. Experimental manipulations can allow one to distinguish between these two possibilities. (5) Many decisionmaking processes can be described with threshold models, such as those used repeatedly here. These models make specific predictions about the relationship between behavior and fitness and they predict abrupt changes from one pattern to another when fitness benefits change. In many cases such abrupt switches occur and their underlying mechanisms have been described (e.g., with horn dimorphism), but in other cases incomplete switches have been described, such as with partial bivoltinism and partial migration. In these cases bet-hedging and frequency-dependent effects are a likely explanation (Swingland, 1983; Lundberg, 1988; Seger and Brockmann, 1987). (6) Because the density of conspecifics affects the resources or time available to allocate to different actions and because conspecifics often have a strong effect on fitness, density-dependent effects are among the most likely factors affecting fitness curves. (7) When no information is available and the animal still needs to make a decision between mutually exclusive options, selection may favor a coin-tossing decision rule. Frequency-dependent selection can act on this decision rule to adjust the frequencies of the two patterns to the ESS. (8) Understanding the maintenance of two patterns in a population is clearly not a “simple” matter of adding up offspring or matings to see if the two patterns are equally successful. These systems are dynamic and constantly evolving, so the chance of hitting a system at the equilibrium is small.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

33

When you view alternative strategies in this way, looking for equal success is pretty futile. (9) Finally, most systems have clear frequency-dependent effects operating, such as sex-allocation systems. These are often of the sort that are likely operating in horseshoe crabs and damselflies, where one pattern does particularly well when it is rare and the other pattern is little affected by the frequency of the other in the population. What lessons do these themes hold for future studies of alternative strategies and tactics?

IV. AN INTEGRATIVE APPROACH The study of alternative phenotypes is a study of decision-making processes and the factors that maintain particular decision-making rules in the population. For most systems this is a study of allocation strategies, how animals choose to allocate time or resources to different activities to maximize fitness. The theories associated with sex allocation, evolutionarily stable strategies, division of labor and social insect castes, mimicry polymorphism, polyphenism, and phenotypic plasticity have provided important insights. But studies of alternative phenotypes also require an understanding of how selection is operating on the underlying physiological (Moore et al., 1998; Lank et al., 1999; Nijhout, 1999) and developmental decision-making processes (Moran, 1992; Harvell, 1994). Let me give two examples that illustrate the importance of an integrative approach to the study of alternative phenotypes. Male speckled wood butterflies (Panarge aegeria), a northern, woodland species, switch between patrolling for females and waiting for them perched on territories. These territories are located where shafts of sunlight penetrate the thick forest canopy. Males without territories continually intrude on these sunlit patches but are nearly always driven out by territory holders (or they leave on their own) (Davies, 1978). When the “owner always wins” behavior was first described, it was called an uncorrelated asymmetry because no known cue or condition distinguished patrollers from perching territorial males (Austad et al., 1979; Davies, 1979). Subsequently, however, the importance of temperature and basking has become apparent (Wickman and Wiklund, 1983; Shreeve, 1987) and the differences in male behavior are now ascribed to the effect of temperature on flight performance (van Dyck and Matthysen, 1998; van Dyck et al., 1997). The lesson is an important one: an understanding of the controlling mechanisms and cues has completely altered our understanding of the behavior and the selective factors that maintain the alternative phenotypes in the population. Emlen (2000b) has shown that differences in the size of larval male dung beetles are associated with differences in the levels of juvenile hormone

34

H. JANE BROCKMANN

(JH) that are present during a 30-h window just prior to pupation. During this time large males release JH, thus stimulating rapid proliferation of cells in certain tissues in the head region that have JH receptors. No such release occurs in small males, so these tissues do not develop (but if JH is applied experimentally, small males with large horns can be produced) (Emlen and Nijhout, 1999). Emlen argues that differences in the placement of those JH sensitive tissues in different species result in different horn morphologies. In part, this placement is driven by the functional morphology of the horn structures, but in part it is also driven by certain developmental constraints. Emlen has found, for example, that nearby tissues (such as eyes) are reduced when large horns develop (on the head). A complete understanding of the evolution of alternative phenotypes in this species has required an integrative approach whereby the developmental, physiological, morphological, and behavioral bases for decisions are studied in concert. Not only are these processes interesting in their own right, but they also allow us to examine exactly how selection is acting on decision-making processes.

V. MODELING ALTERNATIVE ALLOCATION PATTERNS The study of alternative allocation patterns has been hindered by the sheer complexity of the problems being examined, the lack of clear predictions, and the failure to find ways to evaluate the systems experimentally. How can these difficulties be remedied? Any model of strategy allocation, be it a problem in alternative behavioral strategies or sex allocation or social insect division of labor or seasonal polyphenism, must include the following properties. 1. Trade-offs. When animals make decisions about how to behave, they are investing time or resources into one activity that could be invested in another, such as between present and future reproductive opportunities (Alonzo and Warner, 1999) or present mating and life span. This means that any model must have a way of estimating the effects of tradeoffs. 2. Fitness. The goal of any model is to understand how different decisions affect fitness. This will require a common currency that allows us to predict the effects of different kinds of decisions on fitness. 3. Information. Selection favors animals that use all the information they have available to them about the relationship between their actions and fitness, but because information varies, a model must take differences in information into account.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

35

4. State dependence. Selection favors animals that pay attention to their own condition, to the amount of energy reserves they have, and to the risks they are taking with the different options they might follow. Those gains, costs, and risks often change with the age and the condition of the animal, and so it is not surprising that age or size or condition-dependent switches in behavior are common in alternative strategy systems. If physiological or developmental constraints exist, then these, too, must be included. 5. Sequential control. Previous decisions usually alter or constrain what the animal can do or its future payoffs, so the whole process is hierarchical or sequential (Dunbar, 1983). This means that the best approach to modeling alternative phenotypes is one that allows us to track through the life of an animal, adding up its payoff for having taken different routes with different consequences. 6. Costs of changing. Any model also has to take into account the cost of changing from one pattern to another. If changing from A to B takes 2 weeks, as it does in some sex-change systems, then this would have to be factored into the equations. 7. Frequency and density dependence. Any model has to take into account the complex and interacting effects of frequency and density dependence because they change the shapes and values of the fitness curves and alter the risks and costs. For example, in the damselfly system, we argue that at high densities andromorphs do better than gynomorphs because they are able to avoid the high costs of male harassment, but this is true only if andromorphs are rare. If they are too common, then their male mimicry is broken, males recognize them as females, and they suffer the same costs, perhaps even higher costs, than gynomorphs. 8. Dynamics. All of these factors are constantly changing, so a model must be able to include the effects of fluctuating conditions. If we are to analyze alternative strategy allocation systems, we need complex models that simultaneously evaluate all of these properties and make predictions about how animals should behave under different conditions. Some species are better than others for evaluating such models. Useful attributes include the following: (a) knowledge of the decision rules that the animals are using and the information that the animal has available to it to make choices; (b) knowledge of the underlying physiological and developmental bases for the allocation rules, along with how those rules fit into the lifetime, developmental trajectory of the animal; (c) the ability to measure lifetime reproductive success when individuals use different allocation patterns; (d) knowledge of the basis for any frequency dependence that is

36

H. JANE BROCKMANN

operating in the system; (e) the ability to manipulate the system to alter frequency- and density-dependent effects as well as reproductive success; (f) knowledge of the underlying genetics, because all theories of allocation assume heritability of the decision-making processes. It will not be easy to find such a system (for discussions of some problems, see Brockmann and Grafen, 1992; Alcock, 1996b). One approach that has been very useful for sorting out a variety of interacting effects is dynamic programming. Houston and McNamara (1987) and Mangel and Clark (1988) have introduced a method that will allow us to develop and evaluate the type of complex model that is outlined above (Houston et al., 1988), that is, to evaluate the simultaneous effects of many variables on decision-making processes that occur over the lifetime of an animal. The models are particularly good at incorporating the interacting effects of internal state, external conditions such as changes over the season, and the effects of density dependence and changes in risk (Houston et al., 1988; McNamara and Houston, 1996). The models can also incorporate the effects of differences in availability of information, learning, the effect of frequency dependence (dynamic games), and the effects of stochastic processes. They also make specific predictions that can be evaluated in the field or laboratory (Hutchinson and McNamara, 2000). Quite a different picture of a system may be revealed when we consider all these factors together rather than considering each variable separately (Fogel, 1995; St. Mary, 1997; Mangel and Heimpel, 1998; Weber et al., 1998; Kokko and Johnstone, 1999; Yerkes and Koops, 1999; Luttberg and Warner, 1999). In a study that makes use of this approach, Lucas et al. (1996) show some quite unexpected results when multiple factors are considered together. They are interested in predicting the numbers of calling and satellite frogs that will be present in a breeding pond on a particular night. They take into account such factors as the cost of calling, the length of the breeding season, the numbers of females and males arriving at the pond and the overall frequencydependent effect resulting from the fact that females respond only when males are singing (and satellites do not sing). When all of these factors are considered together, there are some unexpected results. For example, the model predicts waves of males joining a chorus, rather than a steady stream of males from one night to the next, even when environmental conditions do not change. The waves are caused by several interacting effects. First, it is costly for males to call, and if a male calls for a few nights, he has to leave the pond to feed. Second, there are density-dependent effects on predation risk because males are safer when there are many frogs present, and density-dependent effects on female arrival rates, and obviously males do better when many females are arriving. Third, frequency dependence also affects the outcome because the success of a satellite depends on the

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

37

frequency of satellites in the population (they do better when they are rare). Together these factors interact to cause the unexpected waves of chorus singing on some nights, waves that have often been observed but were not understood because they did not correlate with any environmental factors. The model also predicts age- and condition-dependent effects on joining a chorus that were not previously suspected. For example, the satellites should be made up of young males and some old males that are in poor condition, something that was not previously recognized. Young males should take up the satellite, noncalling pattern not because they are less attractive to females but because this behavior is less risky and increases their survival chances to the next breeding season. The model identifies female arrival rate as a crucial element and one that drastically alters predictions about males, but little is known about female behavior. The dynamic game modeling approach has allowed Lucas et al. (1996) insights into unexplained behavior, has encouraged field workers to observe and measure phenomena not thought to be of importance in this system, and has enhanced their ability to evaluate the effects of multiple, interacting variables in a far more realistic fashion. Alternative strategy and tactic systems are diverse and complex. Clearly, their analysis should not be handled one factor at a time. Ignoring any one of the elements may lead to an incomplete—or even a completely wrong—explanation for the evolution of alternative behavioral traits. With the dynamic modeling approach (Mangel and Clark, 1988; Houston and McNamara, 1999; Clark and Mangel, 2000), we have the tools to put together realistic models that allow us to make specific predictions. With clear predictions we can subject the models to evaluation in real systems (Hutchinson and McNamara, 2000).

VI. SUMMARY Discrete variation where one sex has more than one means of achieving the same functional end, such as mating or nesting (alternative strategies and tactics), requires an evolutionary explanation. Discrete differences in behavior, morphology, color, and life-history patterns are found in many contexts, such as alternative reproductive patterns, social insect castes, sex allocation, and mimicry polymorphisms. In this chapter I draw on the literature of these fields to identify similar patterns and common models that can be applied to our understanding of alternative phenotypes in general. I emphasize the invertebrate literature throughout because excellent reviews are available for vertebrates.

38

H. JANE BROCKMANN

Traditionally, the behavioral literature has separated alternative phenotypes into those due to genetic differences (e.g., polymorphisms) and those due to environmental or individual cues (e.g., conditional tactics). Such categories are impractical and they force us to dichotomize behavior as due to nature or nurture. Drawing on the well-developed sex-allocation literature, I suggest a different system for categorizing alternative phenotypes: irreversible or adult lifetime patterns (e.g., dioecy) and reversible or facultative patterns (e.g., hermaphroditism), which can be further subdivided into sequential and simultaneous. This new system for organizing alternative phenotypes allows us to see that condition dependence and frequency dependence may occur under any category. In general, alternative phenotypes can be maintained in a population when their fitness curves cross, that is, when each does better than the other under some conditions. A variety of factors can cause fitness curves to cross. Negative frequency dependence is one, which has the added advantage of explaining how patterns can be maintained at a stable frequency. Probably the most common explanation for the maintenance of discrete variants is switches in phenotype based on environmental or individual, conditionbased cues. What are the circumstances that favor the evolution of irreversible phenotypes over switching? Lifetime patterns are favored when individuals that switch have lower fitness than those that do not. Lifetime patterns allow complex suites of adaptations to develop for each phenotype that are coordinated through physiological and developmental processes. When environmental conditions are predictable and correlate with fitness, selection favors the animal switching from one pattern to the other at the time that maximizes fitness. However, when the environmental conditions that favor two phenotypes are unpredictable and change frequently, selection favors animals that invest in what is necessary to perform in either role. Under most circumstances the animal will use decision rules based on information about what maximizes fitness, but if the individual does not have such information, yet must choose between exclusive alternatives, then selection will favor an arbitrary decision rule. A promising approach to studying alternative phenotypes is dynamic, state-variable modeling (Houston and McNamara, 1999; Clark and Mangel, 2000), which is particularly good for evaluating the interacting effects of the organism’s state, of environmental and social conditions, and of density and frequency dependence. When such multiple factors are evaluated together, surprising predictions about the conditions under which individuals should follow particular patterns sometimes result. An understanding of the evolution of alternative phenotypes requires a strong, integrative approach that seeks to understand the cues and mechanisms controlling

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

39

behavioral decisions. When this approach is combined with dynamic, statebased models, it is possible to make predictions about the circumstances that favor the evolution of particular decision-making patterns over others.

Acknowledgments This chapter began with invitations from the Animal Behaviour Society to present a Fellows’ Lecture at their 1998 meeting at Southern Illinois University and from the Association for the Study of Animal Behavior to present a plenary address at their 1998 meeting in Urbino, Italy. I thank both associations for these invitations and for the many valuable discussions following each lecture. During fall 1997, I conducted a graduate seminar course on alternative strategies and many of the ideas and examples presented here were developed with this seminar group. I thank Michael Taborsky, John Alcock, Douglas Emlen, Laura Sirot, Peter Slater, and Charles Snowdon for providing many helpful suggestions on the manuscript.

References Alcock, J. (1979). The evolution of intraspecific diversity in male reproductive patterns in some bees and wasps. In “Sexual Selection and Reproductive Competition in Insects” (M. S. Blum and N. A. Blum, eds.), pp. 381–402. Academic Press, New York. Alcock, J. (1996a). Male size and survival: The effects of male combat and bird predation in Dawson’s burrowing bees, Amegilla dawsoni. Ecol. Entomol. 21, 309–316. Alcock, J. (1996b). Provisional rejection of three alternative hypotheses on the maintenance of a size dichotomy in males of Dawson’s burrowing bee, Amegilla dawsoni (Apidae, Apinae, Anthophorini). Behav. Ecol. Sociobiol. 39, 181–188. Alcock, J. (1996c). The relation between male body size, fighting, and mating success in Dawson’s burrowing bee, Amegilla dawsoni (Apidae, Apinae, Anthophorini). J. Zool. (London) 239, 663–674. Alcock, J. (1997). Competition from large males and the alternative mating tactics of small males of Dawson’s burrowing bees (Amegilla dawsoni) (Apidae, Apinae, Anthophorini). J. Insect Behav. 10(1), 99–114. Alcock, J., Jones, C. E., and Buchmann, S. L. (1977). Male mating strategies in the bee Centris pallida, Fox (Anthophoridae: Hymenoptera). Am. Nat. 111, 145–155. Alonzo, S. H., and Warner, R. R. (1999). A trade-off generated by sexual conflict: Mediterranean wrasse males refuse present mates to increase future success. Behav. Ecol. 10, 105–111. Andersson, M. (1994). “Sexual Selection.” Princeton Univ. Press, New Jersey. Arak, A. (1983). Male–male competition and mate choice in anuran amphibians. In “Mate Choice” (P. P. G. Bateson, ed.), pp. 181–210. Cambridge Univ. Press, UK. Arak, A. (1984). Sneaky breeders. In “Producers and Scroungers” (C. J. Barnard, ed.), pp. 154– 194. Croom Helm, London. Arak, A. (1988). Callers and satellites in the natterjack toad: Evolutionarily stable decision rules. Anim. Behav. 36, 416–432. Arnqvist, G. (1992). The effects of operational sex ratio on the relative mating success of extreme male phenotypes in the water strider Gerris odontogaster (Zett.) (Heteroptera: Gerridae). Anim. Behav. 43, 681–683. Austad, S. (1984a). Alternative male reproductive behaviors. Am. Zool. 24, 309–319.

40

H. JANE BROCKMANN

Austad, S. N. (1984b). A classification of alternative reproductive behaviors and methods for field-testing ESS models. Am. Zool. 24, 309–319. Austad, S., and Howard, R. (1984). Introduction to the symposium: Alternative reproductive tactics. Am. Zool. 24, 307–308. Austad, S. N., Jones, W. T., and Waser, P. M. (1979). Territorial defence in speckled wood butterflies: Why does the resident always win? Anim. Behav. 27, 960–961. Barlow, R. B., Jr., Powers, M. K., Howard, H., and Kass, L. (1986). Migration of Limulus for mating: Relation to lunar phase, tide height, and sunlight. Biol. Bull. 171, 310– 329. Barlow, R. B., Powers, M. K., and Kass, L. (1988). Vision and mating behavior in Limulus. In “Sensory Biology of Aquatic Animals” (J. Atema, R. R. Fay, A. N. Popper, and W. N. Tavolga, eds.), pp. 419–434. Springer-Verlag, New York. Barnard, C. J., and Sibly, R. M. (1981). Producers and scroungers: A general model and its applications to captive flocks of house sparrows. Anim. Behav. 29, 543–550. Basolo, A. L. (1994). The dynamics of siherian sex-ratio evolution: Theoretical and experimental investigations. Am. Nat. 144, 473–490. Bass, A. (1992). Dimorphic male brains and alternative reproductive tactics in a vocalizing fish. Trends Neurosc. 15, 139–145. Bass, A. H. (1996). Shaping brain sexuality: Varying reproductive tactics of plainfin midshipman fish have neural correlates. Am. Scientist 84, 352–363. Bass, A. H. (1998). Behavioral and evolutionary neurobiology: A pluralistic approach. Am. Zool. 38, 97–107. Beekman, M., and van Stratum, P. (1998). Bumblebee sex ratios: Why do bumblebees produce so many males? Proc. R. Soc. London, B 265, 1535–1543. Berglund, A. (1990). Sequential hermaphroditism and the size-advantage hypothesis: An experimental test. Anim. Behav. 39, 426–433. Bond, A. B., and Kamil, A. C. (1998). Apostatic selection by blue jays produces balanced polymorphism in virtual prey. Nature (London) 395, 594–596. Boomsma, J. J. (1991). Adaptive colony sex ratios in primitively eusocial bees. Trends Ecol. Evol. 6, 92–95. Botton, M. L., and Ropes, J. W. (1988). An indirect method for estimating longevity of the horseshoe crab (Limulus polyphemus) based on epifaunal slipper shells (Crepidula fornicata). J. Shellfish Res. 7, 407–412. Bourke, A. F. G. (1997). Sex ratios in bumble bees. Philos. Trans. R. Soc. London, B 352, 1921–1933. Brantley, R. K., and Bass, A. H. (1994). Alternative male spawning tactics and acoustic signals in the plainfin midshipman fish Porichthys notatus Girard (Teleostei, Batrachoididae). Ethology 96, 213–232. Brockmann, H. J. (1980). Diversity in the nesting behavior of mud-dauber (Trypoxylon politum Say; Sphecidae). Fl. Entomol. 63, 53–64. Brockmann, H. J. (1986). Decision making in a variable environment: Lessons from insects. In “Behavior as a Factor in the Dynamics of Populations” (L. Drickamer, ed.), pp. 95–111. Privat, Toulouse, France. Brockmann, H. J. (1990). Mating behavior of horseshoe crabs, Limulus polyphemus. Behaviour 114, 206–220. Brockmann, H. J., and Dawkins, R. (1979). Joint nesting in a digger wasp as an evolutionarily stable preadaptation to social life. Behaviour 71, 203–245. Brockmann, H. J., and Grafen, A. (1989). Mate conflict and male behaviour in a solitary wasp, Trypoxylon (Trypargilum) politum (Hymenoptera: Sphecidae). Anim. Behav. 37, 232–255.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

41

Brockmann, H. J., and Grafen, A. (1992). Sex ratios and life-history patterns of a solitary wasp, Trypoxylon (Trypargilum) politum (Hymenoptera: Sphecidae). Behav. Ecol. Sociobiol. 30, 7–27. Brockmann, H. J., and Penn, D. (1992). Male mating tactics in the horseshoe crab, Limulus polyphemus Anim. Behav. 44, 653–665. Brockmann, H. J., Grafen, A., and Dawkins, R. (1979). Evolutionarily stable nesting strategy in a digger wasp. J. Theor. Biol. 7, 473–496. Brockmann, H. J., Colson, T., and Potts, W. (1994). Sperm competition in horseshoe crabs (Limulus polyphemus). Behav. Ecol. Sociobiol. 35, 153–160. Brockmann, H. J., Nguyen, C., and Potts, W. (2000). Paternity in horseshoe crabs when spawning in multiple male groups. Anim. Behav. (In press.) Brockmann, J. (1996). Satellite male groups in horseshoe crabs, Limulus polyphemus. Ethology 102, 1–21. Bronmark, ¨ C., and Miner, J. G. (1992). Predator-induced phenotypical change in body morphology in crucian carp. Science 258, 1348–1350. Brown, J. L. (1997). A theory of mate choice based on heterozygosity. Behav. Ecol. 8, 60–65. Bull, J. J. (1981). Sex ratio evolution when fitness varies. Heredity 46, 9–26. Bull, J. J. (1983). “Evolution of Sex Determining Mechanisms.” Benjamin/Cummings Publ. Co., Menlo Park, California. Cade, W. (1979). The evolution of alternative male reproductive strategies in field crickets. In “Sexual Selection and Reproductive Competition in Insects” (M. Blum and N. A. Blum, eds.), pp. 343–379. Academic Press, New York. Cade, W. H. (1981). Alternative male strategies: Genetic differences in crickets. Science 212, 563–564. Cain, A. J., and Sheppard, P. M. (1950). Selection in the polymorphic land snail Cepaea nemoralis (L.). Heredity 4, 275–294. Cain, A. J., and Sheppard, P. M. (1954). Natural selection in Cepaea Genetics 39, 89–116. Calderone, N. W., and Page, R. E. (1996). Temporal polyethism and behavioural canalization in the honey bee, Apis mellifera. Anim. Behav. 51, 631–643. Caro, T. M., and Bateson, P. (1986). Organization and ontogeny of alternative tactics. Anim. Behav. 34, 1483–1499. Charnov, E. L. (1979). Simultaneous hermaphroditism and sexual selection. Proc. Natl. Acad. Sci. USA 76, 2480–2484. Charnov, E. L. (1982). “The Theory of Sex Allocation.” Princeton Univ. Press, New Jersey. Charnov, E. L. (1993). “Life History Invariants. Some Explorations of Symmetry in Evolutionary Ecology.” Oxford Univ. Press, UK. Charnov, E. L. (1996). Sperm competition and sex allocation in simultaneous hermaphrodites. Evol. Ecol. 10, 457–462. Charnov, E. L. (1997). Trade-off-invariant rules for evolutionarily stable life histories. Nature (London) 387, 393–394. Charnov, E. L., Hartogh, R. L. L.-d., Jones, W. T., and van den Assem, J. (1981). Sex ratio evolution in a variable environment. Nature (London) 289, 27–33. Clark, C. (1994). Antipredator behavior and the asset-protection principle. Behav. Ecol. 5, 159–170. Clark, C., and Mangel, M. (2000). “Dynamic State Variable Models in Ecology.” Princeton Univ. Press, New Jersey. Clouse, R. (1994). Nesting decisions of the social paper was Mischocyttarus mexicanus. M.S. Thesis, Department of Zoology, University of Florida, Gainesville, FL. Clouse, R. M. (1997). Are lone paper wasp foundresses mainly the result of sister mortality. Fl. Sci. 60, 265–274.

42

H. JANE BROCKMANN

Clutton-Brock, T. H., and Iason, G. R. (1986). Sex ratio variation in mammals. Q. Rev. Biol. 61(3), 339–374. Conover, D., and van Voorhees, D. (1990). Evolution of a balanced sex ratio by frequencydependent selection in a fish. Science 250, 1556–1558. Cook, J. M., Compton, S. G., Herre, E. A., and West, S. A. (1997). Alternative mating tactics and extreme male dimorphism in fig wasps. Proc. R. Soc. London, B 264, 747–754. Cook, L. M., and Gao, G. (1996). Test of association of morphological variation with heterozygosity in the snail Cepaea nemoralis Heredity 76, 118–123. Cooper, G., Miller, P. L., and Holland, P. W. H. (1996). Molecular genetic analysis of sperm competition in the damselfly Ischnura elegans (Vander Linden). Proc. R. Soc. London, B 263, 1343–1349. Cordero, A. (1990). The inheritance of female polymorphism in the damselfly Ischnura graellsii (Rambur) (Odonata: Coenagrionidae). Heredity 64, 341–346. Cordero, A. (1992). Density-dependent mating success and colour polymorphism in females of the damselfly Ischnura graellsii (Odonata: Coenagrionidae). J. Anim. Ecol. 61, 769– 780. Cordero, A., and Egido, F. J. (1998). Mating frequency, population density, and female polychromatism in the damselfly Ischnura graellsii: An analysis of four natural populations. Etologia 6, 61–67. Cordero, A., and Miller, P. L. (1992). Sperm transfer, displacement, and precedence in Ischnura graellsii (Odonata: Coenagrionidae). Behav. Ecol. Sociobiol. 30, 261–267. Cordero, A., Carbone, S. S., and Utzeri, C. (1998). Mating opportunities and mating costs are reduced in androchrome female damselflies, Ischnura elegans (Odonata). Anim. Behav. 55, 185–197. Crespi, B. J. (1986). Size assessment and alternative fighting tactics in Elaphrothrips tuberculatus (Insecta: Thysanoptera). Anim. Behav. 34, 1324–1335. Crnokrak, P., and Roff, D. A. (2000). The trade-off to macroptery in the cricket Gryllus firmus: A path analysis in males. J. Evol. Biol. 13, 396–408. Danforth, B. N., and Desjardins, C. A. (1999). Male dimorphism in Perdita portalis (Hymenoptera, Andrenidae) has arisen from preexisting allometric patterns. Insects Soc. 46, 18–28. Darwin, C. (1859). “The Origin of Species by Means of Natural Selection or The Preservation of Favoured Races in the Struggle for Life.” John Murray, London. Darwin, C. (1871). “The Descent of Man, and Selection in Relation to Sex.” J. Murray, London. Davies, N. B. (1978). Territorial defence in the speckled wood butterfly (Parage aegeria): The resident always wins. Anim. Behav. 27, 1253–1267. Davies, N. B. (1979). Game theory and territorial behaviour in speckled wood butterflies. Anim. Behav. 27, 961–962. Davies, N. B. (1982). Behaviour and competition for scarce resources. In “Current Problems in Sociobiology” (King’s College Sociobiology Group, eds.), pp. 363–380. Cambridge Univ. Press, UK. Dawkins, R. (1980). Good strategy or evolutionarily stable strategy. In “Sociobiology: Beyond Nature/Nurture” (G. W. Barlow and S. Silverberg, eds.), pp. 331–367. Westview Press, Boulder, CO. Dawkins, R., and Brockmann, H. J. (1980). Do digger wasps commit the concorde fallacy? Anim. Behav. 28, 892–896. DeWitt, T. J. (1996). Gender contests in a simultaneous hermaphrodite snail: A size-advantage model for behaviour. Anim. Behav. 51, 345–351. DeWitt, T. J., Sih, A., and Wilson, D. S. (1998). Costs and limits of phenotypic plasticity. Trends Ecol. Evol. 13, 77–81.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

43

Dingle, H. (1984). Behavior, genetics, and life histories: Complex adaptations in uncertain environments. In “New Ecology: Novel Approaches to Interactive Systems.” (P. W. Price, C. N. Slobodchikoff, and W. S. Gaud, eds.), pp. 169–194. John Wiley & Sons, New York. Dominey, W. J. (1980). Female mimicry in male bluegill sunfish — a genetic polymorphism? Nature (London) 284, 546–548. Dominey, W. J. (1981). Maintenance of female mimicry as a reproductive strategy in bluegill sunfish (Lepomis macrochirus). Env. Biol. Fishes 6, 59–64. Dominey, W. J. (1984). Alternative mating tactics and evolutionarily stable strategies. Am. Zool. 24, 385–396. Dugatkin, L. A., and Reeve, H. K., (1998). “Game Theory and Animal Behavior.” Oxford Univ. Press, New York. Dunbar, R. (1982). Intraspecific variations in mating strategy. In “Perspectives in Ethology” (P. P. G. Bateson and P. Klopfer, eds.), Vol. 5, pp. 385–431. Plenum Press, New York. Dunbar, R. I. M. (1983). Life history tactics and alternative strategies of reproduction. In “Mate Choice” (P. P. G. Bateson, ed.), pp. 423–433. Cambridge Univ. Press, UK. Eberhard, W. G. (1979). The function of horns in Podischnus agenor (Dynastinae). In “Sexual Selection and Reproductive Competition in Insects” (M. S. Blum and N. A. Blum, eds.), pp. 231–258. Academic Press, New York. Eberhard, W. G. (1982). Beetle horn dimorphism: Making the best of a bad lot. Am. Nat. 119, 420–426. Eberhard, W. G., and Gutierrez, E. E. (1991). Male dimorphisms in beetles and earwigs and the question of developmental constraints. Evolution 45, 18–28. Ellner, S., and Hairston, N. G. (1994). Role of overlapping generations in maintaining genetic variation in a fluctuating environment. Am. Nat. 143, 403–417. Emlen, D. J. (1994). Environmental control of horn length dimorphism in the beetle Onthophagus acuminatus (Coleoptera: Scarabaeidae). Proc. R. Soc. London, B. 256, 131–136. Emlen, D. J. (1996). Artificial selection on horn length–body size allometry in the horned beetle Onthophagus acuminatus Evolution 50, 1219–1230. Emlen, D. J. (1997a). Alternative reproductive tactics and male-dimorphism in the horned beetle Onthophagus acuminatus (Coleoptera: Scarabaeidae). Behav. Ecol. Sociobiol. 41, 335–341. Emlen, D. J. (1997b). Diet alters male horn allometry in the beetle Onthophagus acuminatus (Coleoptera: Scarabaeidae). Proc. R. Soc. London, B 264, 567–574. Emlen, D. J. (2000a). Dig it, and they will come. Natural History 109, 64–69. Emlen, D. J. (2000b). Integrating development with evolution: A case study with beetle horns. Bioscience 50, 403–418. Emlen, D. J., and Nijhout, H. F. (1999). Hormonal control of male horn length dimorphism in the horned beetle Onthophagus taurus J. Insect Physiol. 45, 45–53. Endler, J. A. (1988). Frequency-dependent predation, crypsis, and aposematic coloration. Phil. Trans. R. Soc. London, B 319, 505–523. Fagerstrom, ¨ T. (1982). Maternal investment, female rivalry, and a fallacy. Oikos 39, 116– 118. Fahrbach, S. E. (1997). Regulation of age polyethism in bees and wasps by juvenile hormone. Adv. Study Behav. 26, 285–310. Field, J. (1989). Alternative nesting tactics in a solitary wasp. Behaviour 110, 219–243. Field, J. (1992). Intraspecific parasitism as an alternative reproductive tactic in nest-building wasps and bees. Biol. Rev. 67, 79–126. Fincke, O. M. (1994a). Female colour polymorphism in damselflies: Failure to reject the null hypothesis. Anim. Behav. 47, 1249–1266. Fincke, O. M. (1994b). On the difficulty of detecting density-dependent selection on

44

H. JANE BROCKMANN

polymorphic females of the damselfly Ischnura graellsii: Failure to reject the null. Evol. Ecol. 8, 328–329. Fink, L. S. (1995). Foodplant effects on colour morphs of Eumorpha fasciata caterpillars (Lepidoptera: Sphingidae). Biol. J. Linn. Soc. 56, 423–437. Fischer, E. A. (1980). The relationship between mating system and simultaneous hermaphroditism in the coral reef fish Hypoplectrus nigricans (Serranidae). Anim. Behav. 28, 620–634. Fischer, E. A. (1988). Simultaneous hermaphroditism, tit-for-tat, and the evolutionary stability of social systems. Ethol. Sociobiol. 9, 119–136. Fogel, A. (1995). Development and relationships: A dynamic model of communication. Adv. Study Behav. 24, 259–290. Forbes, M. (1994). Tests of hypotheses for female-limited polymorphism in the damselfly, Enallagma boreale Selys. Anim. Behav. 47, 724–726. Forbes, M. R. L., Richardson, J. M. L., and Baker, R. L. (1995). Frequency or female morphs is related to an index of male density in the damselfly Nehalennia irene. Ecoscience 2, 28–33. Forbes, M. R., Schalk, G., Miller, J. G., and Richardson, J. M. L. (1997). Male–female morph interactions in the damselfly Nehalennia irene (Hagen). Can. J. Zool. 75, 253–260. Ford, E. B. (1940). Polymorphism and taxonomy. In “The New Systematics” (J. S. Huxley, ed.), pp. 493–513. Clarendon Press, Oxford. Forsyth, A., and Alcock, J. (1990). Female mimicry and resource defense polygyny by males of a tropical rove beetle, Leistotrophus versicolor (Coleoptera: Staphylinidae). Behav. Ecol. Sociobiol. 26, 325–330. Gadgil, M. (1972). Male dimorphism as a consequence of sexual selection. Am. Nat. 106, 574– 579. Gadgil, M., and Taylor, C. E. (1975). Plausible models of sexual selection and polymorphism. Am. Nat. 109, 470–472. Gardner, R., Morris, M. R., and Nelson, C. E. (1987). Conditional evolutionarily stable strategies. Anim. Behav. 35, 507–517. Gillespie, R. G., and Oxford, G. S. (1998). Selection on the color polymorphism in Hawaiian happy-face spiders: Evidence from genetics structure and temporal fluctuations. Evolution 52, 775–783. Giraldeau, L.-A., and Livoreil, B. (1998). Game theory and social foraging. In “Game Theory and Animal Behavior” (L. Dugatkin and H. K. Reeve, eds.), pp. 16–37. Oxford Univ. Press, UK. Goodson, J. L., and Bass, A. H. (2000). Forebrain peptides modulate sexually polymorphic vocal circuitry. Nature (London) 403, 769–772. Greene, E. (1996). Effect of light quality and larval diet on morph induction in the polymorphic caterpillar Nemoria arizonaria (Lepidoptera: Geometridae). Biol. J. Linn. Soc. 58, 277–285. Greene, E. (1999). Phenotypic variation in larval development and evolution: Polymorphism, polyphenism, and developmental reaction norms. In “The Origin and Evolution of Larval Forms” (B. Hall and M. Wake, eds.), pp. 379–410. Academic Press, New York. Greenfield, M. D., and Shelly, T. E. (1985). Alternative mating strategies in a desert grasshopper: Evidence of density dependence. Anim. Behav. 33, 1192–1210. Gross, M. R. (1984). Sunfish, salmon, and the evolution of alternative reproductive strategies and tactics in fishes. In “Fish Reproduction: Strategies and Tactics” (R. Wooton and G. Potts, eds.), pp. 55–75. Academic Press, London. Gross, M. R. (1991a). Evolution of alternative reproductive strategies: Frequency-dependent sexual selection in male bluegill sunfish. Philos. Trans. R. Soc. London, B 332, 59–66. Gross, M. R. (1991b). Salmon breeding behavior and life history evolution in changing environments. Ecology 72, 1180–1186.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

45

Gross, M. R. (1996). Alternative reproductive strategies and tactics: Diversity within sexes. Trends Ecol. Evol. 11, 92–97. Gross, M. R., and Repka, J. (1998). Game theory and inheritance in the conditional strategy. In “Game Theory and Animal Behavior” (L. Dugatkin and H. K. Reeve, eds.), pp. 168–187. Oxford Univ. Press, UK. Guilford, T. C., and Dawkins, M. S. (1991). Receiver psychology and the evolution of animal signals. Anim. Behav. 42, 1–14. Hamilton, W. D. (1979). Wingless and fighting males in fig wasps and other insects. In “Sexual Selection and Reproductive Competition in Insects” (M. S. Blum and N. A. Blum, eds.), pp. 167–220. Academic Press, New York. Harvell, C. D. (1994). The evolution of polymorphism in colonial invertebrates and social insects. Q. Rev. Biol. 69, 155–185. Hassler, C. L. (1999). Satellite male groups in horseshoe crabs (Limulus polyphemus): How and why are males choosing females. M.S. Thesis, Department of Zoology, University of Florida, Gainesville, FL. Hedrick, P. (1995). Genetic polymorphism in a temporally varying environment: Effects of delayed germination or diapause. Heredity 75, 164–170. Hedrick, P. W. (1986). Genetic polymorphism in heterogeneous environments: A decade later. Ann. Rev. Ecol. Syst. 17, 535–566. Heino, M., Metz, J. A. J., and Kaitala, V. (1998). The enigma of frequency-dependent selection. Trends Ecol. Evol. 13, 367–370. Henson, S. A., and Warner, R. R. (1997). Male and female alternative reproductive behaviors in fishes: A new approach using intersexual dynamics. Ann. Rev. Ecol. Syst. 28, 571–592. Herre, E. A., Leigh, E. G., and Fischer, E. A. (1987). Sex allocation in animals. In “The Evolution of Sex and Its Consequences” (S. C. Stearns, ed.), pp. 219–261. Birkhauser Verlag, Basel. Hinnekint, B. O. N. (1987). Population dynamics of Ischnura elegans (vanderLinden) (Insecta: Odonata) with special reference to morphological colour changes, female polymorphism, multiannual cycles and their influence on behaviour. Hydrobiologia 146, 3–31. Hori, M. (1993). Frequency-dependent natural selection in the handedness of scale-eating cichlid fish. Science 260, 216–219. Houston, A. I., and McNamara, J. M. (1987). Singing to attract a mate: A stochastic dynamic game. J. Theor. Biol. 129, 57–68. Houston, A. I., and McNamara, J. M. (1999). “Models of Adaptive Behaviour.” Cambridge Univ. Press, UK. Houston, A., Clark, C., McNamara, J., and Mangel, M. (1988). Dynamic models in behavioural and evolutionary ecology. Nature (London) 332, 29–34. Huang, Z.-Y., and Robinson, G. E. (1992). Honeybee colony integration: Worker–worker interactions mediate hormonally regulated plasticity in division of labor. Proc. Natl. Acad. Sci. USA 89, 11726–11729. Huang, Z.-Y., and Robinson, G. E. (1996). Regulation of honey bee division of labor by colony age demography. Behav. Ecol. Sociobiol. 39, 147–158. Huheey, J. E. (1988). Mathematical models of mimicry. Am. Nat. 131, S22–S41. Hutchings, J. A., and Myers, R. A. (1994). The evolution of alternative mating strategies in variable environments. Evol. Ecol. 8(3), 256–268. Hutchinson, J. M. C., and McNamara, J. M. (2000). Ways to test stochastic dynamic programming models empirically. Anim. Behav. 59, 665–676. Jeanne, R. L. (1991). Polyethism. In “The Social Biology of Wasps” (K. G. Ross and R. W. Matthews, eds.), pp. 389–425. Cornell Univ. Press, Ithaca, NY. Johnson, C. (1964). The inheritance of female dimorphism in the damselfly, Ischnura damula Genetics 49, 513–519.

46

H. JANE BROCKMANN

Johnson, C. (1966). Genetics of female dimorphism in Ischnura demorsa. Heredity 21, 453–459. Johnson, K., DuVal, E., Kielt, M., and Hughes, C. (2000). Male mating strategies and the mating system of great-tailed grackles. Behav. Ecol. 11, 132–141. Kaitala, A. (1988). Wing muscle dimorphism: Two reproductive pathways of the waterstrider Gerris thoracicus in relation to habitat instability. Oikos 53, 222–228. Kaitala, A., and Dingle, H. (1992). Spatial and temporal variation in wing dimorphism of California populations of the waterstrider Aquarius remigis (Heteroptera: Gerridae). Ann. Entomol. Soc. 85, 590–595. Kaitala, A., and Dingle, H. (1993). Wing dimorphism, territoriality, and mating frequency of the waterstrider, Aquarius remegis Ann. Zool. Fenn. 30, 163–168. Kaitala, V., Kaitala, A., and Getz, W. M. (1989). Evolutionarily stable dispersal of a waterstrider in a temporally and spatially heterogeneous environment. Evol. Ecol. 3, 283–298. Kokko, H., and Johnstone, R. A. (1999). Social queuing in animal societies: A dynamic model of reproductive skew. Proc. R. Soc. Lond., B 266, 571–578. Koprowski, J. L. (1993). Alternative reproductive tactics in male eastern gray squirrels: “Making the best of a bad job.” Behav. Ecol. 4, 165–171. Kruuk, L. E. B., Clutton-Brock, T. H., Albon, S. D., Pemberton, J. M., and Guinness, F. E. (1999). Population density affects sex ratio variation in red deer. Nature (London) 399, 459–462. Kukuk, P. F., and Schwarz, M. (1988). Macrocephalic male bees as functional reproductives and probable guards. Pan-Pac. Entomol. 64, 131–137. Lank, D. B., Coupe, M., and Wynne-Edwards, K. E. (1999). Testosterone-induced male traits in female ruffs (Philomachus pugnax): Autosomal inheritance and gender differentiation. Proc. R. Soc. London, B 266, 2323–2330. Lindstrom, ¨ L., Alatalo, R. V., and Mappes, J. (1997). Imperfect Batesian mimicry—The effects of the frequency and the distastefulness of the model. Proc. R. Soc. London, B 264, 149– 153. Lloyd, D. G. (1987). Parallels between sexual strategies and other allocation strategies. In “The Evolution of Sex and Its Consequences” (S. C. Stearns, ed.), pp. 263–281. Birkhauser Verlag, Basel. Lucas, J. R., Howard, R. D., and Palmer, J. G. (1996). Callers and satellites: Chorus behavior in anurans as a stochastic dynamic game. Anim. Behav. 51, 501–518. Lundberg, P. (1988). The evolution of partial migration in birds. Trends Ecol. Evol. 3, 173–174. Luttberg, B., and Warner, R. R. (1999). Reproductive decision-making by female peacock wrasses: Flexible versus fixed behavioral rules in variable environments. Behav. Ecol. 10, 666–674. Majerus, M. E. N. (1986). The genetics and evolution of female choice. Trends Ecol. Evol. 1, 1–9. Majerus, M. E. N. (1998). “Melanism. Evolution in Action.” Oxford Univ. Press, UK. Mallett, J., and Joron, M. (1999). Evolution of diversity in warning color and mimicry: Polymorphisms, shifting balance, and speciation. Ann. Rev. Ecol. Syst. 30, 201–233. Mangel, M., and Clark, C. W. (1988). “Dynamic Modeling in Behavioral Ecology.” Princeton Univ. Press, New Jersey. Mangel, M., and Heimpel, G. E. (1998). Reproductive senescence and dynamic oviposition behaviour in insects. Evol. Ecol. 12, 871–879. Maynard Smith, J. (1982). “Evolution and the Theory of Games.” Cambridge Univ. Press, UK. Maynard Smith, J., and Hoekstra, R. (1980). Polymorphism in a varied environment: How robust are the models? Genet. Res. Camb. 35, 45–57. McNamara, J. M., and Houston, A. I. (1996). State-dependent life histories. Nature (London) 380, 215–221.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

47

Moore, M. C., Hews, D. K., and Knapp, R. (1998). Hormonal control and evolution of alternative male phenotypes: Generalizations of models for sexual differentiation. Am. Zool. 38, 133– 151. Moran, N. (1992). The evolutionary maintenance of alternative phenotypes. Am. Nat. 139, 971–989. Naug, D., and Gadagkar, R. (1998). The role of age in temporal polyethism in a primitively eusocial wasp. Behav. Ecol. Sociobiol. 42, 37–47. Nijhout, H. F. (1999). Hormonal control in larval development and evolution—Insects. In “The Origin and Evolution of Larval Forms” (B. K. Hall and M. H. Wake, eds.), pp. 217–254. Academic Press, New York. Nijhout, H. F., and Wheeler, D. E. (1982). Juvenile hormone and the physiological basis of insect polymorphisms. Q. Rev. Biol. 57, 109–133. Norman, M. D., Finn, J., and Tregenza, T. (1999). Female impersonation as an alternative reproductive strategy in giant cuttlefish. Proc. R. Soc. London, B 266, 1347–1349. Oster, G. F., and Wilson, E. O. (1978). “Caste and Ecology in the Social Insects.” Princeton University Press, New Jersey. Page, R. E., Robinson, G. E., Britton, D. S., and Fondrk, M. K. (1992). Genotypic variability for rates of behavioral development in worker honeybees (Apis mellifera L.). Behav. Ecol. 3, 173–180. Parker, G. A. (1982). Phenotype-limited evolutionarily stable strategies. In “Current Problems in Sociobiology” (King’s College Sociobiology Group, eds.), pp. 173–201. Cambridge Univ. Press, UK. Parker, G. A. (1984a). Evolutionarily stable strategies. In “Behavioural Ecology: An Evolutionary Approach” (J. R. Krebs and N. B. Davies, eds.), pp. 30–61. Blackwell Scientific Publications, Oxford, UK. Parker, G. A. (1984b). The producer-scrounger model and its relevance to sexuality. In “Producers and Scroungers: Strategies for Exploitation and Parasitism” (C. J. Barnard, ed.), pp. 127–153. Croom Helm, London. Parker, G. A., and Courtney, S. P. (1983). Seasonal incidence: Adaptive variation in the timing of life history stages. J. Theor. Biol. 105, 147–155. Parker, G. A., and Rubenstein, D. I. (1981). Role assessment, reserve strategy, and the acquisition of information in asymmetric animal conflicts. Anim. Behav. 29, 221–240. Partridge, L. (1988). The rare-male effect: What is its evolutionary significance? Philos. Trans. R. Soc. London, B 319, 525–539. Partridge, L., and Hill, W. G. (1984). Mechanisms for frequency-dependent mating success. Biol. J. Linn. Soc. 23, 113–132. Pener, M. P. (1991). Locust phase polymorphism and its endocrine relations. Adv. Insect Physiol. 23, 1–79. Penn, D., and Brockmann, H. J. (1994). Nest-site selection in the horseshoe crab, Limulus polyphemus. Biol. Bull. 187, 373–384. Penn, D., and Brockmann, H. J. (1995). Age-biased stranding and righting in male horseshoe crabs, Limulus polyphemus. Anim. Behav. 49, 1531–1539. Petersen, C. W. (1995). Reproductive behavior, egg trading, and correlates of male mating success in the simultaneous hermaphrodite, Serranus tabacarius. Env. Biol. Fishes 43, 351–361. Potts, W. K., and Wakeland, E. K. (1993). Evolution of MHC genetic diversity: A tale of incest, pestilence, and sexual preference. Trends Genet. 9, 408–412. Potts, W. K., Manning, C. J., and Wakeland, E. K. (1997). The role of infectious disease, inbreeding, and mating preferences in maintaining MHC genetic diversity: An experimental test. In “Infection, Polymorphism, and Evolution” (W. D. Hamilton and J. C. Howard, eds.), pp. 99–108. Chapman and Hall, London.

48

H. JANE BROCKMANN

Radwan, J. (1993). The adaptive significance of male polymorphism in the acarid mite Caloglyphus berlesei. Behav. Ecol. Sociobiol. 33, 201–208. Rauscher, M. D. (1986). Competition, frequency-dependent selection, and diapause in Battus philenor butterflies. Fl. Entomologist 69, 63–78. Reeve, H. K. (1998). Game theory, reproductive skew, and nepotism. In “Game Theory and Animal Behavior” (L. Dugatkin and H. K. Reeve, eds.), pp. 118–145. Oxford Univ. Press, UK. Repka, J., and Gross, M. R. (1995). The evolutionarily stable strategy under individual condition and tactic frequency. J. Theor. Biol. 176, 27–31. Riechert, S. E. (1986). Spider fights as a test of evolutionary game theory. Am. Sci. 74, 604–610. Robertson, H. (1985). Female dimorphism and mating behaviour in a damselfly, Ischnura ramburi: Females mimicking males. Anim. Behav. 33, 805–809. Robinson, G. E. (1992). Regulation of division of labor in insect societies. Annu. Rev. Entomol. 37, 637–665. Robinson, J. V., and Novak, K. L. (1997). The relationship between mating system and penis morphology in ischnuran damselflies (Odonata: Coenagrionidae). Biol. J. Linn. Soc. 60, 187–200. Roff, D. A. (1994). Evidence that the magnitude of the trade-off in a dichotomous trait is frequency dependent. Evolution 48, 1650–1656. Roff, D. A. (1996). The evolution of threshold traits in animals. Q. Rev. Biol. 71, 3–35. Rosenheim, J. A., Nonacs, P., and Mangel, M. (1996). Sex ratios and multifaceted parental investment. Am. Nat. 148, 501–535. Rowell, G. A., and Cade, W. H. (1993). Simulation of alternative male reproductive behavior: Calling and satellite behavior in field crickets. Ecol. Model. 65, 265–280. Rubenstein, D. I. (1980). On the evolution of alternative mating strategies. In “Limits to Action: The Allocation of Individual Behavior” (J. E. R. Staddon, ed.), pp. 65–100. Academic Press, New York. Sandoval, C. P. (1994). The effects of the relative geographic scales of gene flow and selection on morph frequencies in the walking-stick Timema cristinae. Evolution 48, 1866–1879. Schulz, D. J., Huang, Z.-Y., and Robinson, G. E. (1998). Effects of colony food shortage on behavioral development in honey bees. Behav. Ecol. Sociobiol. 42, 295–303. Seeley, T. (1985). “Honeybee Ecology.” Princeton Univ. Press, New Jersey. Seger, J., and Brockmann, H. J. (1987). What is bet-hedging? In “Oxford Surveys of Evolutionary Biology” (P. H. Harvey and L. Partridge, eds.), Vol. 4, pp. 182–211. Oxford Univ. Press, UK. Sheldon, B. C., and Verhulst, S. (1996). Ecological immunology: Costly parasite defences and trade-offs in evolutionary ecology. Trends Ecol. Evol. 11, 317–321. Shettleworth, S. J. (1998). “Cognition, Evolution, and Behavior.” Oxford Univ. Press, UK. Shreeve, T. G. (1987). The mate location behaviour of the male speckled wood butterfly, Pararge aegeria, and the efffect of phenotypic differences in hind-wing spotting. Anim. Behav. 35, 682–690. Shuster, C. N., Jr. (1982). A pictorial review of the natural history and ecology of the horseshoe crab Limulus polyphemus, with reference to other Limulidae. In “Physiology and Biology of Horseshoe Crabs: Studies on Normal and Environmentally Stressed Animals” (J. Bonaventura, C. Bonaventura, and S. Tesh, eds.), pp. 1–52. Alan R. Liss, New York. Shuster, S. M. (1989a). Female sexual receptivity associated with molting and differences in copulatory behavior among the three male morphs in Paracerceis sculpta (Crustacea: Isopoda). Biol. Bull. 177, 331–337. Shuster, S. M. (1989b). Male alternative reproductive strategies in a marine isopod crustacean (Paracerceis sculpta): The use of genetic markers to measure differences in fertilization success among ␣-, ␤- and ␥ -males. Evolution 43, 1683–1698.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

49

Shuster, S. M. (1990). Courtship and female mate selection in a marine isopod crustacean, Paracerceis sculpta. Anim. Behav. 40, 390–399. Shuster, S. M. (1992). The reproductive behaviour of ␣-, ␤- and ␥ -male morphs in Paracerceis sculpta, a marine isopod crustacean. Behaviour 121, 231–258. Shuster, S. M., and Sassaman, C. (1997). Genetic interaction between male mating strategy and sex ratio in a marine isopod. Nature 388, 373–377. Shuster, S. M., and Wade, M. J. (1991a). Equal mating success among male reproductive strategies in a marine isopod. Nature 350, 608–610. Shuster, S. M., and Wade, M. J. (1991b). Female copying and sexual selection in a marine isopod crustacean. Anim. Behav. 41, 1071–1078. Simmons, L. W., Tomkins, J. L., and Alcock, J. (2000). Can minor males of Dawson’s burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini), compensate for reduced access to virgin females through sperm competition? Behav. Ecol. 11, 319–325. Sinervo, B., and Lively, C. M. (1996). The rock-paper-scissors game and the evolution of alternative male strategies. Nature (London) 380, 240–243. Sirot, L. K. (1999). Intersexual conflict and mating avoidance in the damselfly, Ischnura ramburi. M. S. Thesis, Department of Zoology, University of Florida, Gainesville, FL. Sirot, L. K., and Brockmann, H. J. (2001). Costs of sexual interactions to females in Rambur’s forktail damselfly, Ischnura ramburi (Zygoptera: Coenagrionidae). Anim. Behav. (In press.) Skulason, ´ S., and Smith, T. B. (1995). Resource polymorphisms in vertebrates. Trends Ecol. Evol. 10, 366–370. Smith, T. B., and Skulason, ´ S. (1996). Evolutionary significance of resource polymorphisms in fish, amphibians, and birds. Ann. Rev. Ecol. Syst. 27, 111–134. St. Mary, C. (1994). Sex allocation in a simultaneous hermaphrodite, the blue-banded goby (Lythrypnus dalli): The effects of body size and behavioral gender and the consequences for reproduction. Behav. Ecol. 5, 304–313. St. Mary, C. (1997). Sequential patterns of sex allocation in simultaneous hermaphrodites: Do we need models that specifically incorporate this complexity? Am. Nat. 150, 73–97. Swingland, I. R. (1983). Intraspecific differences in movement. In “The Ecology of Animal Movement” (I. R. Swingland and P. J. Greenwood, eds.), pp. 102–115. Clarendon Press, Oxford. Sword, G. A., and Simpson, S. J. (2000). Is there an intraspecific role for density-dependent colour change in the desert locust? Anim. Behav. 56, 861–870. Sword, G. A., Simpson, S. J., El Hadi, O. T. M., and Wilps, H. (2000). Density-dependent aposematism in the desert locust. Proc. R. Soc. London, B 267, 63–68. Taborsky, M. (1994). Sneakers, satellites, and helpers: Parasitic and cooperative behavior in fish reproduction. Adv. Study Behav. 23, 1–100. Taborsky, M. (1998). Sperm competition in fish: “Bourgeois” males and parasitic spawning. Trends Ecol. Evol. 13, 222–227. Taborsky, M. (1999). Conflict or cooperation: What determines optimal solutions to competition in fish reproduction? In “Behaviour and Conservation of Littoral Fishes” (V. C. Almada, R. Oliveira, and E. J. Goncalves, eds.), pp. 1–47. Instituto Superior de Psicologia Aplicada, Lisboa, Portugal. Tauber, M. J., Tauber, C. A., and Masaki, S. (1986). “Seasonal Adaptations of Insects.” Oxford Univ. Press, UK. Taylor, D. R. (1990). Evolutionary consequences of cytoplasmic sex ratio distorters. Evol. Ecol. 4, 235–248. Taylor, F. (1986). Toward a theory for the timing of hibernal diapause. In “The Evolution of Insect Life Cycles” (F. Taylor and R. Karban, eds.), pp. 236–258. Springer Verlag, Berlin.

50

H. JANE BROCKMANN

Thornhill, R. (1979). Adaptive female-mimicking behavior in a scorpionfly. Science 205, 412– 414. Trivers, R. L., and Hare, H. (1976). Haplodiploidy and the evolution of the social insects. Science 191, 249–263. Trivers, R. L., and Willard, D. E. (1973). Natural selection of parental ability to vary the sex ratio of offspring. Science 179, 90–92. Turner, J. R. G. (1977). Butterfly mimicry: The genetical evolution of an adaptation. Evol. Biol. 10, 163–204. Turner, J. R. G. (1989). Mimicry: The palatability spectrum and its consequences. In “The Biology of Butterflies” (R. I. Vane-Wright and P. R. Ackery, eds.), pp. 141–161. Princeton Univ. Press, New Jersey. van Dyck, H., and Matthysen, E. (1998). Thermoregulatory differences between phenotypes in the speckled wood butterfly: Hot perchers and cold patrollers. Oecologia 114, 326– 334. van Dyck, H., Matthysen, E., and Dhondt, A. A. (1997). The effect of wing colour on male behavioural strategies in the speckled wood butterfly. Anim. Behav. 53, 39–51. van Gossum, H., Stoks, R., Matthysen, E., Valck, F., and De Bruyn, L. (1999). Male choice for female colour morphs in Ischnura elegans (Odonata, Coenagrionidae): Testing the hypotheses. Anim. Behav. 57, 1229–1232. Via, S., and Lande, R. (1985). Genotype–environment interaction and the evolution of phenotypic plasticity. Evolution 39, 505–522. Vickery, W. L., Giraldeau, L. A., Templeton, J. J., Kramer, D. L., and Chapman, C. A. (1991). Producers, scroungers, and group foraging. Am. Nat. 137, 847–863. Waage, J. K. (1979). Dual function of the damselfly penis: Sperm removal and transfer. Science 203, 916–918. Waage, J. K. (1986). Evidence of widespread sperm displacement ability among Zygoptera (Odonata) and the means for predicting its presence. Biol. J. Linn. Soc. 28, 285–300. Wakelin, D., and Apanius, V. (1997). Immune defence: Genetic control. In “Host–Parasite Evolution” (D. H. Clayton and J. Moore, eds.), pp. 30–58. Oxford Univ. Press, UK. Walker, T. J. (1986). Stochastic polyphenism: Coping with uncertainty. Fl. Entomol. 69, 46–62. Waltz, E. C. (1982). Alternative mating tactics and the law of diminishing returns: The satellite threshold model. Behav. Ecol. Sociobiol. 10, 75–83. Waltz, E. C., and Wolf, L. L. (1984). By Jove! Why do alternative mating tactics assume so many different forms? Am. Zool. 24, 333–343. Waltz, E. C., and Wolf, L. L. (1988). Alternative mating tactics in male white-faced dragonflies (Leucorhinia intacta): Plasticity of tactical options and consequences for reproductive success. Evol. Ecol. 2, 205–231. Weatherhead, P. J., Dufour, K. W., Lougheedd, S. C., and Eckert, C. G. (1999). A test of the good-genes-as-heterozygosity hypothesis using red-winged blackbirds. Behav. Ecol. 10, 619–625. Weber, T. P., Ens, B. J., and Houston, A. I. (1998). Optimal avian migration: A dynamic model of fuel stores and site use. Evol. Ecol. 12, 377–402. Wedekind, C. (1994). Handicaps not obligatory in sexual selection for resistance genes. J. Theor. Biol. 170, 57–62. West-Eberhard, M. J. (1979). Sexual selection, social competition, and evolution. Proc. Am. Philos. Soc. 123, 222–234. Wheeler, D. E. (1991). The developmental basis of worker caste polymorphism in ants. Am. Nat. 138, 1218–1238. Wickman, P.-O., and Wiklund, C. (1983). Territorial defence and its seasonal decline in the speckled wood butterfly (Pararge aegeria). Anim. Behav. 31, 1206–1216.

THE EVOLUTION OF ALTERNATIVE STRATEGIES AND TACTICS

51

Williams, G. C. (1979). The question of adaptive sex ratio in outcrossed vertebrates. Proc. R. Soc. London, B. 205, 567–580. Wilson, E. O. (1953). The origin and evolution of polymorphism in ants. Q. Rev. Biol. 28, 136–156. Wilson, E. O. (1975). “Sociobiology. The New Synthesis.” Harvard Univ. Press, Cambridge, MA. Wilson, E. O. (1984). The relation between caste ratios and division of labor in the ant genus Pheidole (Hymenoptera: Formicidae). Behav. Ecol. Sociobiol. 16, 89–98. Wolf, L. L., and Waltz, E. C. (1993). Alternative mating tactics in male white-faced dragonflies: Experimental evidence for a behavioural assessment ESS. Anim. Behav. 46, 325–334. Yerkes, T., and Koops, M. A. (1999). Redhead reproductive strategy choices: A dynamic state variable model. Behav. Ecol. 10, 30–40.

This Page Intentionally Left Blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 30

Information Gathering and Communication during Agonistic Encounters: A Case Study of Hermit Crabs ROBERT W. ELWOOD AND MARK BRIFFA SCHOOL OF BIOLOGY AND BIOCHEMISTRY THE QUEEN’S UNIVERSITY OF BELFAST BELFAST, BT9 7BL UNITED KINGDOM

I. GAME THEORY AND ANIMAL CONTESTS A major advance in our understanding of animal contests was made with the application of game theory (Maynard Smith and Price, 1973). Game theory models generally assume that individuals attempt to maximize individual fitness when contesting resources with nonrelatives. It is supposed that natural selection will act in such a way that a population of animals comes to adopt an evolutionarily stable strategy (ESS). This strategy, if employed by most members of the population, confers a higher fitness to any individual than does any other strategy that is within the capabilities of the animal or set of strategies within the game (Maynard Smith and Price, 1973). Two early models, the hawk/dove game and the symmetrical war of attrition game, were particularly influential and will help clarify basic concepts. In the mixed hawk/dove game (Maynard Smith, 1976), each contestant, at each encounter, may elect to play hawk or dove. The hawk option involves attacking the opponent and continuing to do so until injured or until the opponent retreats. The dove option involves displaying for a predetermined period or, if attacked, retreating before getting injured. Should both contestants elect to play hawk one will be injured and incur the cost of a wound (W) and the other will gain the resource of a particular value (V) but the fights will be short. If both elect to play dove, both incur the cost of time wasting (T ) and one, arbitrarily, gains the resource. Should one elect to play hawk and the other dove, the former will win the resource. It might appear that playing hawk is the better option but the solution depends on the values of the costs (W and T) and the benefit (V ). A key feature of this game is that 53

C 2001 by Academic Press Copyright  All rights of reproduction in any form reserved. 0065-3454/01 $35.00

54

ROBERT W. ELWOOD AND MARK BRIFFA

it is assumed that all those playing hawk are equal and all those playing dove are equal, in other words that the contests are essentially symmetrical and no assessment of the opponent or of the resource takes place. The game assumes that all wounds are equal within a population and the resource value is the same for each fight. The ESS may be calculated as follows: play hawk with probability (2T + V)/(2T + W) and play dove with probability 1 − [(2T + V)/(2T + W)]. If we provide arbitrary values of 20 for the cost of the wound, 10 for the value of the resource, and 3 for the cost of time wasting, we find the ESS is to play hawk on 61.5% of the occasions and dove on 38.5% of the occasions. In itself, this is not particularly illuminating but we gain more insights into how animals might behave by manipulating these arbitrary values. This shows that if the resources are typically of a high value, then that population should be more prone to adopt the hawk option. Indeed, if the value is the same or higher than the cost of the wound, then the ESS is to select hawk all the time. If, however, the cost of the wound is much higher than the potential value of the resource the ESS will shift so that dove is played more often. These shifts in the ESS will be set by natural selection. The second influential game, the symmetrical war of attrition, examines situations where contests are settled on the basis of contest duration, the winner being the animal that persists the longer (Maynard Smith and Price, 1973; Maynard Smith, 1974, 1976). The contest is settled purely on the basis of which animal is prepared to accept the greater cost of time. Thus, if one contestant competes for time m1 and the other for time m2, where the former is the greater, the payoff to the victor is V − m2 and for the loser is –m2. Note that the victor only has to fight up to the point when the loser quits. The problem is to find an ESS for the persistence time. There is no pure ESS in this model. Suppose a population adopts a strategy of playing m; then any mutant that played slightly more would win and there would be selection for increasing persistence times. The expected benefit of a contest, however, is V/2 as only one gains the resource and, if m were greater than V/2, it would pay not to fight as there would be a net loss from each encounter. It might seem that playing m = V/2 is the ESS, but a variable strategy with a distribution of potential “bids” that follows a negative exponential distribution turns out to be the only stable state (Maynard Smith, 1974). The distribution of bids is such that it produces a distribution of fight times the mean of which is V/2. Both of the above models assume that the animals are essentially equal and that they value the resource in the same way, but this is unrealistic. Animals within a species vary in size, power, possession of weapons, skill in fighting, position, etc., and this should influence the ability to impose costs on the opponent during an encounter. Any animal that has the ability to

INFORMATION GATHERING AND COMMUNICATION

55

gather information about the resource holding power (RHP) (Parker, 1974) of the opponent relative to its own RHP has a distinct advantage. With respect to the hawk/dove game, the animal that perceives itself as having the higher RHP should be more likely to assume the hawk mode and the animal perceiving that it has the lower RHP should be more likely to assume the dove mode. Resource value is also likely to vary from contest to contest and any animal that can gather information about the resource value may adjust its tactics. With a high value resource it pays to increase the probability of playing hawk. Furthermore, within a fight two opponents may value the resource differently. In this case it would pay the animals to adjust their tactics accordingly. Thus, although the assumptions of the hawk/dove game were unrealistic the game provided a conceptual basis for the development of more realistic models that account for information gathering prior to and during each contest. With respect to the war of attrition, the ability to gather information about costs and benefits changes the nature of the game and the ESS is different. The solution is not now to shift from one distribution of bids to another, rather, it is that each opponent should be prepared to incur costs up to the value of the resource (Hammerstein, 1981; Parker and Rubenstein, 1981; Hammerstein and Parker, 1982). A war of attrition may result in which case it will terminate only when the contestant prepared to pay the lower costs gives up. However, if animals were to have perfect information prior to the contest, then no contest would occur as it would pay the animal that can afford only the lower costs to quit before it incurs unnecessary costs. It is clear, however, that animals do fight or display and it is expected that much of the activity within contests functions to gather information. If information is gathered it should have an effect on the motivational state of the animal (sensu Elwood and Neil, 1992). For example, if, after interacting for a while, one animal begins to perceive that it has the greater RHP, then its motivation to continue the contest should increase. Likewise, if one animal perceives a great value in the resource, the motivation for that animal to compete should increase. The sequential assessment model examines this information gathering and consequent motivational change during a contest (Enquist and Leimar, 1983, 1987). The model assumes that at the start of the contest, information about RHP is low but that the information, and hence the reliability of the assessment, increases during the contest. With a low value resource, an animal may accept limited information about relative RHP to determine when to quit, but with high value resources the decision to quit should be based on more accurate information. Thus it is anticipated that fights will be of a longer duration with high value resources. Furthermore, when the difference between the contestants in RHP is great, it is expected that the information about the asymmetry will be more easily perceived and, therefore, long fights are

56

ROBERT W. ELWOOD AND MARK BRIFFA

predicted when the animals are similar in ability. In recently developed game theoretical models, such as the energetic war of attrition (Payne and Pagel, 1997), the influence of the rate of performance on the level of signaling is considered. This further refinement provides a set of predictions about how the temporal pattern of signaling is expected to vary, and these are discussed in the final section.

II. HERMIT CRAB SHELL FIGHTS Hermit crabs are decapod crustaceans that inhabit empty gastropod shells. The shell provides protection from predators, desiccation, and salinity changes (Elwood and Neil, 1992), but the effectiveness of the shell depends on its size relative to crab size. A shell that is too small does not have room for the hermit crab to withdraw and one that is too large and heavy imposes additional costs in carrying it about (Herreid and Full, 1986). Furthermore, shells of different species appear to have different qualities for the crabs and influence growth and reproductive success (Hazlett and Baron, 1989; Elwood et al., 1995). For example, the common European hermit crab Pagurus bernhardus shows a strong preference for Littorina obtusata shells over Gibbula cineraria, and individuals found in the preferred species have a higher reproductive success in terms of number of eggs per brood and number of broods per season (Elwood et al., 1995). Empty shells are typically rare and, therefore, hermit crabs have difficulty in locating optimal shells (Hazlett, 1970a; Childress, 1972; Scully, 1979); even if one is found it will cease to be optimal as the crab grows. Various field studies have confirmed that crabs tend not to be housed in optimal shells (Childress, 1972; Abrams, 1978; Imafuku, 1984). Crabs readily approach and investigate any empty shell encountered and will move into those shells if they offer an improvement (Elwood and Neil, 1992). Crabs may also attack wounded gastropods to obtain shells (Imafuku, 1983). Probably the most common way of obtaining a new shell, however, is by engaging in a “shell fight,” after which a shell exchange may occur. Shell fights have common features among the various genera of hermit crabs (Hazlett, 1966), but the following description is of P. bernhardus (Dowds and Elwood, 1983). The fight starts when two crabs encounter each other and a display of the chelipeds may occur. The larger of the two crabs then moves forward rapidly and grasps the shell of the smaller crab causing the latter to withdraw into its shell. The larger crab, which we will call the attacker, then runs its chelipeds over the exterior of the defender’s shell (that of the smaller crab) and turns the shell to access the aperture. The attacker then feels into the aperture and the defender often moves its major cheliped in an apparently defensive act called “cheliped flicking” (Hazlett, 1970b).

INFORMATION GATHERING AND COMMUNICATION

57

FIG. 1. A diagrammatic representation to clarify the terminology that we use about shell fights. “Shell raps” are shown individually as vertical bars on a horizontal time base and are divided into “bouts.” Within a bout the raps are separated by short “gaps” and between bouts there are longer “pauses.”

The attacker may then escalate the contest by grasping the defender’s shell and moving it around its columellar axis a few times (shell rocking). This is followed by a more vigorous activity known as shell rapping. The two shells are forcefully hit together in a bout of approximately 4–16 raps after which there is a pause (Fig. 1). During the pause the attacker may engage in shell rocking and also may pull at the chelipeds of the defender before resuming rapping. There may be numerous bouts of rapping and subsequent pauses until either the attacker gives up the contest or the defender relaxes its grip on the interior of the shell and is pulled free of the shell by the attacker. At this point the attacker lets go of the defender, which then remains naked in the vicinity of the encounter, and the attacker feels into the aperture of the now empty defender’s shell. The attacker may move into the defender’s shell but typically keeps hold of its old shell, which it may also investigate. After examining both shells the attacker may move back to the original shell or may move away in the new shell, leaving the defender to occupy whichever shell is left by the attacker, usually the discarded shell originally occupied by the attacker. The crabs are thus said to have “exchanged” shells. Hermit crab fights are unusual in that there are two resources. Each participant starts and finishes the fight with a shell and it is possible that both

58

ROBERT W. ELWOOD AND MARK BRIFFA

crabs could gain in shell quality. For example, a large crab in a small shell could fight a small crab in a large shell. This gave rise to the idea that these encounters are not fights but a process of “negotiation” (Hazlett, 1978, 1982, 1983, 1987, 1989, 1996). The negotiation model suggests that shell exchange should occur when both the defender and the attacker can make a gain in shell quality. It is a requirement of this model that both crabs can gather information about both shells during the encounter so that they can both determine which is the better shell for themselves. We will return to this idea, but it should be noted at the outset that in the species of our studies, P. bernhardus, there is no evidence of negotiation. Exchanges are likely when the attacker would improve the quality of its shell irrespective of the gain or loss to the defender (Elwood and Glass, 1981; Dowds and Elwood, 1983).

III. RESOURCE VALUE AND DECISIONS DURING CONTESTS It is clear that hermit crabs contest ownership of shells and shells may be quantified in terms of crab preferences. These preferences have been determined from choice experiments so that for any size of crab we know the optimal shell size (Jackson, 1988). We have used two main ways to vary the value of the shells. First, either the preferred Littorina, or the much less preferred Gibbula shells may be used. Second, we can vary the size (weight) of the shells relative to the optimum shell for a particular size of crab (Fig. 2). In subsequent experiments, crabs are typically cracked out of their original shells (this does not harm the crab) and offered a shell determined by the experimental design. Thus we can vary the quality of the shells occupied by crabs prior to placing those crabs together to engage in a shell fight. We can thus accurately judge the gain that may be made by the attacker and the defender in shell exchange. Hermit crabs are readily available, easily maintained and manipulated, and make ideal subjects to examine information gathering in contests. One experiment, involving four groups, examined how the value of the two shells influenced decisions of both participants during the course of the encounter (Dowds and Elwood, 1983). In the GL group the larger crab was given a Gibbula shell (the less preferred species) and the smaller crab a Littorina shell (the preferred species), both shells being the size suitable for the larger crab. In the GG group both crabs were in Gibbula shells, in the LL group both were in Littorina shells, and in the LG group the larger crab was in the Littorina shell and the smaller crab in the Gibbula shell. The two crabs were then placed together in a container and monitored. Large crabs in the GL group could make a gain if they obtained the smaller crab’s shell, those in the GG and LL groups would

INFORMATION GATHERING AND COMMUNICATION

59

FIG. 2. Crab weight is plotted against the preferred shell weight derived from shell selection experiments when (a) Littorina obtusata and (b) Gibbula cineraria shells were offered (data from Jackson, 1988).

neither gain nor lose, whereas those in the LG group would lose in shell quality if they exchanged shells. Of the 125 fights observed in the experiment 93.3% involved the larger crabs assuming the role of attacker and the few fights initiated by the smaller crabs are disregarded in the following analyses.

60

ROBERT W. ELWOOD AND MARK BRIFFA

The larger crabs in the poor quality Gibbula shells were more likely to initiate a fight than were those in Littorina shells, but the quality of the smaller crabs’ shells did not influence the decision of whether the larger crabs would attack or not. Thus, there was no evidence that attackers could assess the shell quality of their opponents prior to initiating fights. Once each fight started, however, the attacker could feel over the exterior of the opponent’s shell and there were strong effects of shell quality on the decision to escalate or not. Attackers in poor quality shells were more likely to escalate and, if the defenders were in the good quality shells for the attackers, escalated fights were more probable. If the attacker is basing its decision on the potential gain it must have information on both shells and we could predict the probability of escalation in the following order of the groups: GL > GG = LL > LG This prediction was supported by the data. In the group where attackers could make the highest potential gain in shell quality, GL, 75% of fights were escalated; in groups GG and LL, 31 and 27% of fights were escalated, respectively; in the group with the lowest potential gain in shell quality, LG, only 7% of fights were escalated. At the beginning of the encounter the attacker appears only to have information about its own shell but it gathers information about its opponent’s shell prior to deciding whether or not to escalate. The potential gain to the attacker also had a marked effect on whether or not the smaller crab was evicted, with attackers that had the most to gain also being the most likely to effect an eviction. Information is still gathered, however, after the defender is evicted and after the attacker moves into the defender’s shell. As stated above, the victorious attacker feels within the empty defender’s shell and, if it moves into it, may also feel its old shell. Of the 24 attackers in the GL group that changed shells, all remained in the new shell; of the 14 attackers in the GG and LL groups that changed shells, 50% changed back; and the single crab in the LG group that changed moved back to its original shell. It is clear that a series of decisions made by attackers depends on the information available at that time. It might be predicted, under the negotiation model, that smaller crabs would be more willing to enter into an encounter if they were in poor quality shells, but there was no evidence that this occurred. Defenders in poor quality shells do not appear to be inciting encounters with others. Once the encounter starts, the smaller crab assumes a different role from that of the attacker and withdraws into its shell. It thus seems unlikely that it could gather much information about the attacker’s shell by tactile means. It has been suggested that defenders could assess the acoustic fundamental

INFORMATION GATHERING AND COMMUNICATION

61

frequency of the rapping sound, which should be determined by the internal volume of the two shells and be independent of weight and force of the raps (Hazlett, 1987). Thus, it might be possible for the defender to gain information about the volume of the attacker’s shell. Under the negotiation model the main function of rapping is envisaged as the attacker providing information for the defender. Again, there is no evidence in our species that crabs can obtain information in this manner and there has been no supporting evidence provided for the idea of the fundamental frequency being used in any species. When withdrawn, however, the defender shows the one defensive activity of cheliped flicking. The function of this activity has not been investigated in detail. Perhaps it is an attempt to avoid being grasped by the attacker or perhaps it disrupts the ability of the attacker to gather information about the quality of the defender’s shell. Whatever the function, when escalated fights were examined, defenders in the preferred Littorina shells were more likely to flick their chelipeds than were those in Gibbula shells. However, as predicted by the aggression model, the quality of the attacker’s shell had no effect on the defender’s behavior. Thus, it seems from studies on P. bernhardus that the defender only has information about its own shell. Furthermore, there is no evidence of negotiation because defenders that were most likely to be evicted were in the GL group, that is, the group in which defenders had the most to lose by exchange.

IV. RELATIVE SIZE OF THE OPPONENTS AND DECISIONS IN FIGHTS It was noted above that the opponents display their chelipeds at the start of an encounter and, in the vast majority of cases, it is the larger crab that assumes the role of attacker. Thus, some assessment of size occurs at this stage and this appears to be based on displays of the chelipeds. These displays tend to be shorter in duration if the size difference of the animals is small (Dowds and Elwood, 1985), indicating that the decision about roles is easier if the size disparity is large. Other experiments involving crabs with autotomized chelipeds or small, regenerating chelipeds confirm the role of these appendages in establishing which animal will be the attacker and which will be the defender (Neil, 1985). Earlier work with models of crabs with large or small chelipeds demonstrated that the former tended to cause retreat, whereas the latter tended to be ignored (Hazlett, 1969). Information on size is also gathered at later stages. When smaller crabs attack they tend to do so without the larger having displayed. In these cases the smaller crab typically ceases the attack when it accesses the aperture of the larger crab’s shell and detects the chelipeds of the larger crab (Dowds

62

ROBERT W. ELWOOD AND MARK BRIFFA

and Elwood, 1985). Thus, relative RHP seems to influence decisions in these fights. However, there is a complication that makes it difficult to design unambiguous experiments. If we vary the relative size of crabs, then we cannot keep the shell quality constant for both animals. This is because each crab has an optimal shell that depends on the size of the crab. For example, if the size difference between the crabs is increased, and both shells are kept at a suitable size for the larger crab (so that motivation about the resource is the same for the attacker), then the shell occupied by the smaller crab will be much too large. This may change the accessibility of the smaller crab to the larger crab because the former may withdraw deep into the shell and thus avoid being grasped and pulled by the attacker. The large size of the defender’s shell may also alter the value with which the defender regards that shell. One way in which this problem can be reduced is to use a naked crab to fight for the shell of another crab. Naked crabs are extremely vulnerable in natural situations and are highly motivated to obtain almost any shell. In one such experiment four groups of crabs were used that varied in the size disparity of the opponents (Dowds and Elwood, 1985). In all cases one of the crabs weighed 0.25 ± 0.02 g and was in a Littorina shell of the optimal size for itself. The other crab in each pair was naked and in groups A, B, C, and D weighed 0.35, 0.30, 0.25, and 0.20 g, respectively. Thus, the motivation for the housed crab should be the same in terms of its retention of the shell. The value of the shell to the attacker is extremely high because any shell offers a very substantial improvement to a naked crab. Fights were more likely to be initiated by the naked crab when the housed crab was relatively small and the housed opponent was more likely to initiate a fight if it was relatively large. The latter cases resulted in short fights in which the housed crab grasped the opponent and violently shook it. Subsequently, the naked crab avoided the housed crab. Fights initiated by the naked crab were also unusual in that they had no shell with which to engage in shell rapping. Nevertheless, attackers still moved their abdomens as if they were rapping. The action, however, appeared to have little effect because the fights were of very long duration but with little chance of evicting the defender. Relatively large crabs persisted for a longer duration and made more rapping movements, indicating that relative RHP was assessed and influenced decisions about persistence. This is opposite to the prediction based solely on resource value. The shell offered a greater gain to the attacker in group C (where the crabs were the same size and hence the shell would be the optimum size for the naked attacker) than in groups A and B (where the shell would be smaller than the optimum for the attacker). Thus, the data show that relative size has a major influence on the motivation of attacker’s persistence.

INFORMATION GATHERING AND COMMUNICATION

63

There is also evidence that the relative size of opponents can continue to have effects after defenders are evicted in fights in which each crab starts with a shell (Dowds and Elwood, 1985). At this point the victorious attacker typically feels into the now empty shell of the defender and, if it moves into that shell, the attacker may withdraw deeply into it, apparently testing it for size and fit. The relative size of crabs influences the duration of these assessments. The key analysis compared fights in which the attackers were in Gibbula shells at the start and fought crabs in Littorina shells. In all cases the shells were of a size suitable for the crab in the Gibbula shell, but two groups differed in the size difference of the opponents. Either the opponents were the same size or the crab in the Gibbula shell was larger. When the victorious attacker was the larger crab, it took significantly more time feeling into the empty shell of the defender than if it was the same size. Furthermore, once the attacker changed into the new shell it was more likely to withdraw into the shell, apparently to continue the assessment, if the crabs differed in size. The victorious crabs that were larger than their opponents thus appeared to be more confident at this stage and less subject to counterattack by the naked animal.

V. A MODEL OF INFORMATION GATHERING AND MOTIVATIONAL CHANGE The process of shell assessment during a contest is likely to be similar to that undertaken by crabs when they encounter empty shells. The activities are similar; the empty shell is examined on the exterior and then within the aperture (Elwood and Stewart, 1985). Shells may be offered with items within the aperture so the crab has to work to clear the shell (Elwood and Adams, 1990) or with the aperture sealed with cement so that attempts to clear the shell are not effective (Neil and Elwood, 1986). Studies of shell assessment by hermit crabs have been fruitful toward understanding motivational change and decision making during information gathering. Here we review such studies and subsequently show how this understanding can be applied to studies of contest behavior. Crabs investigating shells that offer a high gain accept those shells more quickly than if they only offered a moderate gain (Elwood and Stewart, 1985). This finding stimulated the development of a simple threshold model of motivational change and decision making (Elwood and Neil, 1986, 1992; Jackson and Elwood, 1989; Elwood, 1995). A crab assessing a shell has three options: it may continue with the activity, it may move into the shell (or to the next stage of assessment), or it may reject the shell. Which one of these occurs will depend on the perceived quality of the shell being assessed.

64

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 3. The level of causal factors for shell acquisition is the result of two inputs: (a) the quality of the shell in possession (increasing input with decreasing quality) and (b) the quality of the shell being investigated (increasing input with increasing quality) (from Elwood and Neil, 1992).

Numerous factors may be assessed for each shell (volume, weight, shape, epibionts, etc.), but it is assumed that these various factors can be reduced to two inputs (Fig. 3). One is the level of causal factors associated with the shell in possession: the poorer the shell, the higher the causal factors for new shell acquisition. The other is the level of causal factors for the shell being investigated: the better the new shell, the higher the causal factors. The former will be known at the start of shell assessment, whereas the latter will change as information is gathered. These two factors will summate to create a single output that changes over time. If this output increases above a threshold the crab should enter the shell, if it decreases below a lower threshold the crab should reject the shell. First let us consider the initial level of causal factors at the start of the investigation (Fig. 4). This will be determined by the information the crab has at that time, and, thus, will be set primarily by the quality of the shell in possession. If that shell is relatively poor there will be a higher level of causal factors than if it is relatively good. Thus, if the crab is in a poor shell, the initial level of causal factors for shell acquisition may be near the threshold for shell entry. Conversely, if the crab is in a relatively good shell, the level will be near the threshold for shell rejection (respectively, points A and B in Fig. 4). As the assessment proceeds the level of causal factors is predicted to change in a direction dependent on the information that is gathered. Thus, if the new shell is of good quality, this will be perceived during the assessment and the level of causal factors will move upward, toward the shell entry threshold, whereas if the shell is of poor quality, the trajectory will be downward, toward the rejection threshold (Fig. 5). If the new shell is of good quality, then a crab in a poor shell (point A in Fig. 4) will accept it more quickly than will a crab in a shell that is not so poor (point B in Fig. 4).

INFORMATION GATHERING AND COMMUNICATION

65

FIG. 4. Point A represents the level of causal factors for a crab in a poor shell and point B for a crab in a moderate shell at the start of an investigation of a good shell. The causal factors for shell acquisition will rise in both cases, as an improvement is perceived during the assessment, but crab A accepts the shell earlier than does crab B (from Elwood and Neil, 1986).

FIG. 5. Points A and B are the same initial starting points as in Fig. 4, but in this case each crab encounters a particularly poor shell. The level of causal factors will thus decline as the assessment continues but crab B rejects the shell earlier than does crab A (from Elwood and Neil, 1986).

66

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 6. Lines A and B represent changes in the level of causal factors for crabs encountering poor and very poor shells, respectively. The worse the encountered shell is the earlier the rejection (from Elwood and Neil, 1986).

The reverse, however, should occur for shell rejection (Fig. 5). Especially good or poor shells are expected to be easily perceived as such and result in faster decisions (Fig. 6). For example, a shell that is too small and also has an obstruction in the aperture is likely to be perceived as being of poor quality sooner than one that is of the optimal size but with the aperture blocked. Thus, for any single starting point there may be different speeds with which a shell is accepted or rejected. If the new shell causes neither an upward nor a downward trajectory, the crab might investigate for a long time. However, investigation will not continue forever because causal factors for other activities are likely to increase and also because crabs seem to become increasingly distractible by other stimuli during a prolonged shell assessment (Neil and Elwood, 1986). One final refinement of the model is required. The assessment process consists of two major activities prior to moving to the new shell: external investigation and aperture investigation. It is expected that information gathered during external investigation will influence the level of causal factors at the start of aperture investigation. Thus, if the exterior of the shell indicates good quality, the crab will have a higher motivation to obtain the shell when it initiates aperture investigation than if the exterior indicates only moderate quality (Jackson and Elwood, 1989). We may thus envisage a series of thresholds, one for each stage of the information gathering process.

INFORMATION GATHERING AND COMMUNICATION

67

A. TESTING THE MODEL This is a simple model generated from observations on the timing of decisions to accept shells. It predicts changes in motivational state that are dependent on information gathered during the assessment process. A model is only useful, however, if it can be tested to provide new insights, and for this we need measures of the motivational state. Two main methods have been used. First, the completion of the sequence may be prevented in some manner so that we can measure how long the crab persists with attempts to complete the sequence. The higher the motivation the longer should be the persistence. Second, the crab may be interrupted at some time during the shell investigation with a novel, potentially startling stimulus. The higher the motivation, the shorter is the predicted startle (Culshaw and Broom, 1980; Moorehouse et al., 1987). 1. Persistence When the Sequence Cannot Continue The model predicts that a crab occupying a poor shell or investigating a good shell should have a high motivation to obtain the new shell (Fig. 4). Both situations are predicted to result in a longer persistence time when the sequence is stopped and these predictions have been tested in a number of experiments. In the first experiment, crabs were housed in shells that were of 25, 50, 75, or 100% of the optimum weight for those crabs (25% is much too small for the crab) (Neil and Elwood, 1986). These crabs were offered shells, either 100 or 25% of the optimum weight, but, in each case, the aperture was blocked with dental cement. All the shells used were of the preferred species for these crabs (Littorina) and the measure of interest was the time spent investigating the shell prior to rejection. A two-way analysis of variance showed that crabs persisted with the investigation longer if their current shells were of poor quality and also if the new shell was of good quality (Fig. 7). Thus, there is a clear fit with the predictions of the model in that information from both major sources, own shell and assessed shell, influence the motivational state as determined by persistence times. To test that these findings are not specific to the shell aperture being blocked, a different method of stopping the sequence was employed. In this test, crabs were housed in Littorina shells, either 50 or 75% of the optimum weight (i.e., too small), and then offered Littorina or Gibbula shells of the optimum size (Fig. 8). Each offered shell, however, was stuck, aperture down, to a piece of slate that was buried in the sand on the bottom of the experimental chamber, so that the shell was clearly visible above the sand but the slate hidden. The shell could not be moved and the measure was how long each crab persisted in attempting to turn it. The data showed a clear effect of the species of shell that was stuck to the slate, with crabs persisting longer

68

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 7. The median investigation (persistence) times are plotted for crabs in various sizes of shells (relative to preferred size) that investigate blocked shells, either of the preferred size (100%) or that are much too small (25%) (from Elwood and Neil, 1992).

FIG. 8. The mean time spent by crabs attempting to turn shells that are stuck aperture down. Crabs spent more time attempting to turn the preferred Littorina obtusata shells than the less preferred Gibbula cineraria shells.

INFORMATION GATHERING AND COMMUNICATION

69

with the preferred species. The effect of the size of shell inhabited by the crab, however, was not significant, although there was a trend in the predicted direction. These data demonstrate that information is gathered from the exterior of the shell and that this information alters the motivational state of the crab, as measured by the time spent attempting to access the aperture. Thus, these and other experiments not described here (see Jackson and Elwood, 1989) are congruent with the type of motivational change predicted by the threshold model. 2. Interrupting the Sequence with a Novel Stimulus By using the interruption technique, data may be gathered that are predicted to show an inverse relationship with the motivational state, in that a high motivation to acquire a shell should result in a short startle response. This prediction was tested using crabs housed in Littorina shells 75% of the optimum weight (i.e., slightly too small). Each crab was subsequently offered a single shell of the same species that was either 100 or 50% of their optimum weight, that is, a potential gain or loss could be made, respectively (Jackson and Elwood, 1990). During the shell investigation a novel stimulus was presented. This was a black card moved over the crab at a height of 10 cm causing a shadow to fall on the crab for about 1 s. The stimulus was applied at one of three points in the investigation: (1) when the crab first took hold of the shell, (2) when the crab was turning the shell, and (3) after 5 s of aperture investigation. Crabs stimulated when holding or investigating the aperture showed significantly shorter startle responses if they were investigating optimum shells, rather than shells that were too small (Fig. 9). That is, the duration of the startle response clearly reflected the information gathered during the assessment. The above experiment demonstrated that crabs had some information about the offered shell when they first took hold of it so it was possible that information had been gathered prior to contact, probably by visual means. To test this, a second experiment was conducted that applied the stimulus prior to the crabs contacting the shells (Jackson and Elwood, 1990). The quality of the occupied shell was also varied; crabs were in Littorina shells either 50 or 75% of the optimum weight. These crabs were then offered shells of the preferred size but some were of the preferred species (Littorina), whereas others were of the unpreferred species (Gibbula). The stimulus of the black moving card was applied as the crab approached within 1 cm of the shell but before it made contact with the shell. Crabs approaching the less preferred Gibbula shells had longer startle responses than did those approaching the preferred Littorina shells (Fig. 10). Thus, even prior to contact, some information had been gained that

70

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 9. The number of crabs showing different durations of interruption following presentation of a novel stimulus at different stages of shell assessment. Crabs in 100% shells showed shorter startle durations than did those in 50% shells at the holding and investigating (aperture) stages but not at the turning stage (Mann–Whitney U tests).

influenced the motivational state. When approaching Littorina shells, the size of the occupied shell had no significant effect. When approaching Gibbula shells, however, crabs in 50% shells had shorter startle responses than did those in 75% shells. Thus, the interruption technique helped elucidate differences in motivation that were due to information about the occupied shell as well as that gathered visually about the other shell during approach. Combining the appropriate data from these two interruption experiments allowed the effects of the time that the novel stimulus was applied to be explored (Fig. 11). In each case crabs in 50% Littorina shells were offered optimal Littorina shells to investigate. The data indicated that startle responses occurring prior to contact with the shell were considerably longer than were those that occurred after contact. This effect may be interpreted as being due to the more accurate information being gathered after contact. The new shells offered an improvement in quality and, as this was perceived

INFORMATION GATHERING AND COMMUNICATION

71

FIG. 10. The median duration of the interruption in the assessment of shells following presentation of a novel stimulus for crabs occupying L. obtusata shells, either 50 or 75% of the preferred weight, as they approached L. obtusata or G. cineraria shells of the optimum weight (from Jackson and Elwood, 1990).

FIG. 11. The median durations of interruption in assessment following presentation of a novel stimulus is shown for crabs in two experiments that occupied 75% L. obtusata shells and were offered 100% shells of the same preferred species. The data indicate longer interruptions when the crab had yet to contact the shell compared to those that were manipulating the shell. (From Jackson and Elwood, 1990).

72

ROBERT W. ELWOOD AND MARK BRIFFA

by the crabs, the causal factors for shell acquisition should have risen, and thus, the startle responses were shorter. Assessment activities, however, may also produce a decline in motivation. For example, if a poor quality shell was offered for assessment, then that should be perceived as being poor during the assessment and longer startle responses would be predicted toward the end of the assessment process. That is, as the level of causal factors declines toward the shell rejection threshold, the startle response should be longer. This possibility has also been tested, using both male and female crabs (in the experiments described above only males were used) housed in Littorina shells 50% of the optimum weight. The crabs were then offered an optimal weight shell of the same species but which had the aperture blocked with dental cement (Fig. 12). Crabs were exposed to the novel stimulus at one of three points: (1) 3 s after initial contact with the shell, at which point they should perceive a good quality shell; (2) 2 s after contacting the blocked aperture; or (3) 10 s after contacting the blocked aperture. Analysis of variance indicated no difference in the responses of males and females but there was a clear effect of the time of the stimulus with those stimulated late in the sequence having significantly longer startle responses. Thus, the prediction is upheld. When the blocked aperture is detected and investigated for 10 s the

FIG. 12. The mean interruptions are shown for male and female crabs investigating blocked shells at different stages in the investigation: A denotes 3 s after initial contact, B denotes 2 s after contacting the blocked aperture, and C denotes 10 s after contacting the blocked aperture. There was no gender difference but the startle responses were longer toward the end of the assessment (error bars show standard error).

INFORMATION GATHERING AND COMMUNICATION

73

animal seems to gather information as to the unsuitability of the shell. Thus, the level of causal factors declines toward the rejection threshold, resulting in the increased startle duration.

VI. MOTIVATIONAL CHANGE DURING AGONISTIC ENCOUNTERS The above experiments demonstrate that motivational state changes as information is gathered about shells, at least in nonagonistic situations. When assessing the exterior of a shell of optimum size and species the motivation of the crab to obtain that shell has been shown to rise (Fig. 11), but when an immovable blockage is encountered the motivation declines (Fig. 12). This latter case is analogous to the situation for an attacker in a shell fight, the shell being occupied by another crab rather than blocked by cement. Crabs persist with the shells for similar times in these two situations (Fig. 13, compare a and c), suggesting similar changes in motivation, although it should be noted that the activities differ in that shell rapping only occurs in a fight. We have seen that crabs persist longer, with a shell with a cement blockage, if they are in particularly poor shells and if the exterior of the assessed shell is good (Fig. 7). We predict that the same should occur in a shell fight. Certainly, attackers with much to gain are more likely to escalate the fight and are more likely to win the fight, presumably because they persist longer in the fight. This is because, if there is little to gain, attackers usually give up prior to escalating the fight and the encounters are of short duration (Fig. 13c). Thus, attackers only persist and escalate if there is a clear potential gain. However, if we examine only escalated fights there is little evidence to suggest that attackers with a high potential gain persist longer than those for which the potential gain is moderate. This analysis is, however, based on few data. When the potential gain is high, the attackers are remarkably successful in evicting their opponents and thus there are very few attackers that give up. It is only these unsuccessful attackers that show their full persistence times and thus divulge their motivation. An exception concerns experiments with naked attackers that have a considerable gain if they are successful but tend not to be able to evict the opponent as often as normally housed crabs. These naked crabs persist in fights considerably longer than housed crabs that give up (Dowds and Elwood, 1985). It is also predicted that the relative sizes of the crabs should influence the persistence time of attackers. Again, however, there are too few data to test this because too few attackers give up. The exception, again, is with naked attackers and we have noted previously that the size difference influences the persistence times in a manner consistent with theory (Dowds and Elwood,

74

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 13. The distribution of durations for which crabs persist in assessment of shells blocked with dental cement (a) is similar to that seen for fights when the attacker gives up (c). Note that the majority of incomplete fights are shorter than those that result in eviction (b). (From Elwood and Neil, 1992).

INFORMATION GATHERING AND COMMUNICATION

75

1985). To examine information gathering and motivational change during normal fights, however, a technique is required that is not dependent on the attacker being the one to give up the fight.

VII. PROBING MOTIVATIONAL CHANGE DURING ANIMAL CONTESTS Elwood et al. (1998) used the interruption technique to probe the motivational state of attackers at an early stage of the fight. In the experiment the key variables were (1) the potential gain to the attacker, and (2) the size difference between the crabs. It is expected that, if the attacker can detect a high potential gain and/or the defender is relatively small, the attacker should show a short interruption. Crabs were allocated in pairs of differentsized animals to one of two experimental groups. In all cases the smaller crab of the pair was given a Littorina shell of the correct size for the larger crab. In group A (n = 92), the larger crab of each pair was given a Gibbula shell that was too small for it, the shell being the correct size for the smaller crab. In group B (n = 117), the larger crab was given a Gibbula shell that was the correct size for it and, hence, too large for the smaller crab. Thus, in each case, the larger crab could obtain the preferred size and species by taking the opponent’s shell, but the gain would be greater in group A than in B. The variation in size difference of the crabs was similar in the two groups. If a fight commenced and the attacker had felt into the aperture of the defender’s shell for 10 s, a novel stimulus was applied. This was a black card dropped from a height onto the side of the glass dish, which reliably caused the attacker to temporarily stop fighting and withdraw into its shell prior to resumption of the fight. Interestingly, a more substantial stimulus was required to startle fighting crabs compared to previous studies in which crabs were assessing empty shells (Jackson and Elwood, 1990). The two groups did not differ in the probability of a fight, but there were significantly more evictions in the high gain group [A = 53/59 (90%), B = 47/62 (76%)]. The effects of potential gain and of relative size of the opponents on the startle duration for all fights and on various parameters of shell rapping for those that resumed the fight and effected an eviction were analyzed by analysis of covariance. Crabs in the high gain group (A) had significantly shorter startle responses than those in the low gain group (B) (means 10.2 and 14.3 s, respectively), but, unexpectedly, the relative size of crabs had no effect on the startle duration. Crabs in the high gain group fought significantly harder than did those in the low group. They performed more raps in total to evict their opponents (means 93.0 and 59.0, respectively) and more bouts of shell rapping (means 11.2 and 7.1, respectively)

76

ROBERT W. ELWOOD AND MARK BRIFFA

but not in the number of raps in each bout. In contrast, relative crab size had no effect on the total raps or on the number of bouts of rapping required to evict the opponent, however, fewer raps per bout were given by more evenly matched crabs. It is clear that attacking crabs assess the potential gain prior to escalation as shown by the shorter startle response in the high gain group. Furthermore, the high level of motivation of group A appears to persist throughout the contest as noted by the greater effort. However, although group A crabs gave more raps in total and more bouts of rapping, there was no difference between the groups in the duration of the contest. Rather, crabs in the high gain group fought more vigorously than did those with a lower gain, with shorter pauses between bouts, resulting in more raps in the same time. A subsequent study that also employed the interruption technique showed that attackers that went on to evict the opponent had shorter startle responses than did those that failed to evict (Briffa and Elwood, in preparation). Thus the high motivation of the attackers is reflected in their probability of victory. Victorious crabs also are those that fight with great vigor, which suggests that motivation may also be measured by the changing vigor during the fight. However, caution is required when interpreting ongoing behavior as fully reflecting motivational state during agonistic encounters, because there are sound theoretical reasons to suggest that animals will not signal their future intentions (Maynard Smith, 1982). Studies discussed later suggest that the vigor of the fight is markedly influenced by the stamina of the crab and may not always mirror the information available to the crab. A final complication is that the vigor of the fight is not always a simple function of number of raps per unit time. It may also be a function of the power of individual raps, with high power leading to a slower repetition (Briffa and Elwood, 2000b). Thus, the relationship between motivation and vigor is not clear. In an apparent contradiction to theory (Hammerstein and Parker, 1982; Enquist and Leimar, 1987), the relative size of the combatants did not appear to influence the motivation of the attacker as measured by the startle response (Elwood et al., 1998). However, the temporal pattern of the fight did vary with relative size, with fewer raps per bout being given to relatively large opponents. Relative size also failed to influence total effort to gain an eviction and it was not a predictor of victory. This is different from other studies of animal contests but is easily explained. The smaller crab was always housed in a shell of the right size for the larger crab. Thus, when there was a large size difference between the crabs, the small crab was in a shell much too large for it. This enabled the smaller crab to withdraw deeply into the shell, making it difficult for the attacker to pull at it during the contest. A second order polynomial regression between relative size and the

INFORMATION GATHERING AND COMMUNICATION

77

number of bouts required for eviction makes this point. This shows a U shaped function with particularly large or small size differences resulting in more bouts of shell rapping. A key feature to emerge from these studies is that, despite the superficial resemblance to an asymmetrical war of attrition, these fights appear not to be settled purely on the basis of persistence. There are additional features concerning the vigor with which the attacker fights that influences when the defender gives up. It is the vigor of fighting and how that fits with recent models of contest behavior that is the subject of the next section.

VIII. THE VIGOR OF SHELL RAPPING AND COMMUNICATION Recent work has concentrated on analyses of the pattern of shell rapping (Briffa et al., 1998; Briffa and Elwood, 2000a,b, 2001). The models of negotiation and aggression make different predictions about the possible functions of this activity. As discussed above, the negotiation model predicts that rapping must advertise the quality of the attacker’s shell to enable the defender to make a decision about whether to allow an exchange on the basis of the change in shell quality that this would entail. It has been suggested that this could happen by defenders determining the volume of the attacker’s shell by assessing the fundamental frequency of the raps (Hazlett, 1987). According to the aggression model, however, any information that shell rapping provided for defenders would concern either the motivation or the fighting ability (RHP) of the attacker. A second possibility under the aggression model is that shell rapping could have a detrimental effect on the defender that would reduce its ability to maintain an adequate grip on its shell, such that after a number of bouts of rapping it would no longer be capable of resisting eviction. Suggested detrimental effects include disruption of the respiratory water current, disorientation by disrupting the sensory organs (Elwood and Neil, 1992), and direct action on the abdominal muscles (Briffa and Elwood, 2000b). Note that the two possibilities, of communication and causing detrimental effects, are not mutually exclusive and the aggression model supports combined strategic and tactical functions for shell rapping. Under the aggression model, defenders would give up either because they would be rendered incapable of resisting or because they would anticipate high levels of detrimental effects, time costs, or possible injury if they continued to resist the attacker’s attempts at eviction or because of a combination of these factors. These two models of shell exchange thus predict different functions for shell rapping and make different predictions about how the outcome of an

78

ROBERT W. ELWOOD AND MARK BRIFFA

encounter should be affected by (1) the temporal pattern of the raps and (2) the power with which rapping is performed. If shell rapping was an aggressive activity, one would intuitively expect that the probability of the defender being evicted would increase with both the temporal vigor of rapping and the power with which the raps are delivered. Experimental studies of fighting in other animals, such as those conducted on roaring in red deer, where the maximum number of roars per bout that an individual is capable of performing is related to its fighting ability (Clutton-Brock and Albon, 1979), appear to support this. In addition, there are various theoretical models that specifically deal with repetition of displays used during agonistic interactions. Two of these models that appear to be relevant to shell fighting in hermit crabs are the sequential assessment game (Enquist and Leimar, 1983, 1987) and the energetic war of attrition (Payne and Pagel, 1996, 1997). According to the sequential assessment game, repeated performances allow receivers to accumulate increasingly accurate information such that contest durations are expected to be determined by the cost of obtaining information, the accuracy of assessment, and other factors that may affect these such as the presence of a predator (Brick, 1999). Under the energetic war of attrition model, however, the number of repetitions is integral to the level of advertisement. Both models make predictions about how the rate of signaling should vary during contests and the latter model in particular, which relates to situations in which displays of ability would be important, implies that the vigor of performance should influence the level of advertisement. The aggression model of shell exchange suggests that the temporal pattern of rapping and, possibly, the force with which the raps are delivered, should influence whether or not the defender is evicted. The negotiation model, however, makes no predictions about variation in either the vigor or power of rapping with fight outcome because neither of these factors would affect the fundamental frequency of the raps. Briffa et al. (1998) tested the predictions of the aggression model by analyzing the temporal structure of rapping during staged encounters. Crabs were allocated to pairs and the larger crab of each pair was supplied with either a shell of the unpreferred Gibbula species or the preferred Littorina species. This shell was either 50 or 80% of the crab’s optimal shell weight. The smaller crab of each pair was supplied with a Littorina shell of the optimal weight for the large crab with which it was paired. Thus there were four groups of staged fights defined by the size and species of shell supplied to the large crabs. The large crab would always gain an optimal shell by effecting an exchange, but the level of gain differed between the groups. The most gain occurred in the group where the large crabs were supplied

INFORMATION GATHERING AND COMMUNICATION

79

with 50% adequate Gibbula shells and the least gain occurred in the group where large crabs were supplied with 80% adequate Littorina shells. This enabled the effects of (1) the outcome of the encounter, (2) the species of shell supplied to the large crab, (3) the size of shell supplied to the large crab and, (4) the relative weight difference (RWD) between the two crabs on the pattern of rapping to be assessed. The second and third factors determine the potential gain available to the large crab by effecting an exchange and RWD determines the relative fighting abilities of the two participants, such that each of these three factors could potentially contribute to the motivation of the large crab to initiate an attack and perform shell rapping. The parameters of the pattern of rapping that were measured were the total number of bouts of rapping and the total number of raps, the mean number of raps per bout, and the mean duration of the pauses in rapping between bouts. As expected, large crabs supplied with very poor quality shells, ones that were only 50% adequate in size and of the unpreferred species (Gibbula 50%), were the most likely to initiate a fight, whereas those that were supplied with relatively good shells, ones that were 80% adequate and of the preferred species (Littorina 80%), were least likely to initiate a fight. There was no significant difference, however, in the likelihood of an exchange between the four groups. Of the large crabs that initiated a fight, those that effected an exchange performed more bouts of rapping and more raps in total than those that gave up (Fig. 14). Furthermore, the attackers supplied with 50% adequate shells performed more bouts of rapping in total and more raps than those supplied with 80% adequate shells. Analysis of the mean number of raps performed in each bout showed that successful attackers performed more raps per bout than those that did not evict the opponent over the last four bouts of the fight (Fig. 15) but not in the first four bouts, and that the number of raps per bout increased with RWD (Fig. 16). Attackers that evicted the opponent also left pauses of shorter duration than those that did not (Fig. 17) and there was a strong negative relationship between pause duration and RWD with attackers leaving shorter pauses when the weight difference between the two crabs was high (Fig. 18). Thus, when the whole fight is examined, it appears that successful attackers (1) perform more raps and (2) rap more vigorously, performing more raps per bout and leaving shorter pauses between bouts, than those that are unsuccessful. Whereas the simple asymmetric war of attrition model is based on the assumption that contests should be settled by duration, these data demonstrate that the rate of performance is also an important factor. One might expect a trade-off between these two factors, but from the evidence of our studies, it appears that individuals that both perform more repetitions and display more vigorously are likely to be victorious. The vigor with which

80

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 14. Attackers in poor quality shells (50%) produced more bouts of rapping than did those in medium quality shells (80%) and those that evicted their opponent produced more bouts of rapping than those that did not evict (From Briffa et al., 1998).

FIG. 15. The mean number of raps per bout increased over the last four bouts of rapping for crabs that evicted their opponent, whereas for those that failed to evict the number of raps per bout declined (From Briffa et al., 1998).

INFORMATION GATHERING AND COMMUNICATION

81

FIG. 16. The positive relationship between the RWD of the opponents and the number of raps performed by attackers in the fourth bout (From Briffa et al., 1998).

FIG. 17. The mean duration of each of the last three pauses for attackers that effected an exchange decreased, whereas the mean duration tended to increase for attackers that did not effect an exchange (From Briffa et al., 1998).

82

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 18. The negative relationship between the RWD of the opponents and the duration of the third pause left by attackers (From Briffa et al., 1998).

crabs rapped increased with the weight difference between the two crabs. Clearly, these findings support the aggression model of shell exchange but not the negotiation model and the vigor of rapping, the key determinant of success, appears to be related to the ability of attackers. Variation in the vigor of rapping over the course of shell fights was investigated by examining the number of raps performed in each of the first and last four bouts of rapping and the durations of the pauses that separated these bouts (Briffa et al., 1998). The number of raps performed during the first four bouts did not change from bout to bout but the duration of the pauses between these bouts became progressively longer (Fig. 19). During the final four bouts of rapping, attackers that went on to evict the opponent increased the number of raps from bout to bout, with the difference between the second last bout and the last bout being particularly marked (Fig. 15), and left pauses of progressively shorter duration between each of these bouts (Fig. 17). Attackers that gave up without evicting the opponent, however, performed fewer raps in each successive bout (Fig. 15) and left pauses of increasing duration (Fig. 17). Crabs thus decrease the vigor of rapping during the initial four bouts but may subsequently either decrease or increase the vigor of rapping over the last four bouts. Which pattern of change is chosen influences the outcome of the encounter. It remains unclear why attackers that are about to give up should apparently signal this by reducing the vigor of rapping when theory predicts that

INFORMATION GATHERING AND COMMUNICATION

83

FIG. 19. The mean duration of each of the first three pauses shows an increase in pause duration over the early stage of the fight irrespective of the eventual outcome of the encounter or of the size of shell supplied to attackers (From Briffa et al., 1998).

animals should attempt to conceal their future intentions during contests such as these (Maynard Smith, 1982). Reduction in the vigor of aggressive activity prior to giving up has also been shown in fighting in cichlid fish (Turner and Huntingford, 1986) and it was suggested that this information is revealed unintentionally, possibly because unsuccessful contestants become exhausted toward the end of the fight. In this case it might still pay them to continue signaling at a reduced rate as long as there is some chance of victory. The only theoretical model that allows both escalation and de-escalation in the rate of signaling is the energetic war of attrition (Payne and Pagel, 1996, 1997). This model differs from the simple war of attrition in that contest outcome and duration are expected to be determined by a combination of the time costs and the energetic costs incurred during the fight, whereas the simple war of attrition model only takes time into account. Thus the simple asymmetric war of attrition is in fact a special, limited case of an energetic war of attrition where the energy costs are very low. For this reason both models may make similar predictions under some circumstances. A condition of the energetic war of attrition is that any signal of ability must be a cumulative measure based on both contest duration and magnitude of performance. During the course of the encounter the signaler is expected to continually reassess its level of performance, in relation to the ability of its opponent, such that the level of signaling can be adjusted to allow the amount of energy expended to be minimized. De-escalation in the rate of signaling can only

84

ROBERT W. ELWOOD AND MARK BRIFFA

take place when the time-associated costs increase sublinearly such that a doubling of contest duration does not result in a two-fold increase in time costs. It has thus been suggested that animals that engage in this type of contest signal their ability by performing an exhausting activity that allows the receiver to assess their stamina. Stamina can be defined as an animal’s capacity to perform an energetically demanding activity against the effects of fatigue. In this sense, the term fatigue means that the rates of energy consumption during the activity should be high in relation to (1) the total available energy pool and (2) the speed of recovery of the muscles involved. Thus, fatigue would be expected to regulate the speed of performance of energetically demanding signaling activities such as shell rapping. To test this, Briffa and Elwood (2000a) examined changes in the very small periods of time between the raps within bouts over the course of fights. They suggested that, if the vigor of rapping is affected by fatigue, an increase in the duration of these “gaps” (Fig. 1) would be expected as the fight progressed. Furthermore, any effects of fatigue should become apparent relatively quickly such that an increase in gap durations within bouts would also be expected; attackers would be able to rap vigorously at the start of a bout, immediately following a pause, but toward the end of a bout the effects of fatigue would be expected to slow the rate of rapping until the bout is terminated to allow another pause. In addition to these effects on the duration of the gaps within bouts, it would also be expected that the number of raps per bout should decrease and that the duration of pauses between bouts should increase with the number of bouts performed and increasing levels of fatigue. Thus, there are four features of the temporal pattern of rapping that could indicate whether the activity is affected by fatigue: (1) variation in the duration of gaps from bout to bout, (2) variation in the duration of gaps within bouts, (3) variation in the number of raps from bout to bout, and (4) variation in the duration of pauses as the fight progresses. In order to determine how these features varied during shell fighting Briffa and Elwood (2000a) staged fights between pairs of crabs where the smaller crab of each pair was provided with a shell that was 100% adequate for the larger crab, but the larger crab was provided with a shell that was either 25 or 50% adequate for itself. To measure the duration of the periods of time between raps, the sound of the raps was recorded using a microphone that was attached to a computer via sound filtering equipment. This enabled the duration of the sound produced by each rap to be recorded by the computer. The duration of each gap could then be calculated simply by subtracting the time at which the record of the rap preceding it started from the start of the rap by which the gap was terminated. A similar approach was taken to

INFORMATION GATHERING AND COMMUNICATION

85

FIG. 20. The mean duration of gaps, between raps within a bout, shows a marked increase over the first six bouts (From Briffa and Elwood, 2000a).

the analysis of gap durations to that used by Briffa et al. (1998) in assessing variation in the number of raps per bout and in the duration of pauses as the fight progressed. Repeated measures ANOVAs were used to analyze gap durations during the first six bouts of rapping but within each of these bouts two separate analyses, each containing a second repeated measure, were included to take into account the position of the gap within the bout. The durations of the first and last four gaps of each bout were analyzed. By analyzing six bouts it was possible to obtain a measure of how the pattern varied during the course of the fight while still including a sufficient number of fights of adequate duration in the analysis. This analysis showed that there was a highly significant increase in the duration of gaps from bout to bout over the first six bouts (Fig. 20). In addition to this, there was a similarly strong pattern for the duration of gaps to increase within bouts during the first four gaps and continuing over the last four gaps until the end of the bout (Fig. 21). An unexpected feature of this within-bout pattern, however, was that the duration of the first gap was greater than that of the second and this appeared to be a characteristic feature of the pattern of gap durations. Thus, although there was a strong overall pattern of increasing gap durations within bouts this was only true from the second gap until the end of the bout. Further analysis of these fights showed that during the first six bouts the number of raps tended to decrease and the duration of pauses increased as the fight progressed.

86

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 21. The gap duration first shows a decrease and then an increase over the first four gaps of each bout (a) and then continues to increase over the last four gaps until the end of the bout (b) with the bout terminating after a particularly long gap. The initial decrease in gap duration is curious but appears to be a characteristic feature that is present in most bouts of rapping (From Briffa and Elwood, 2000a).

INFORMATION GATHERING AND COMMUNICATION

87

The pattern of variation in the vigor of rapping thus exhibited all of the features that would be expected if vigor was affected by the level of fatigue. Another possibility, however, is that gap durations are actively varied by attackers during the course of the fight and that this variation is a design feature of the signal. This could explain the unexpected increase in duration between the first and second gap of each bout. In order to establish more firmly whether or not fatigue does affect the pattern of rapping, physiological evidence is needed. One possible result of sustained exhausting activity would be anaerobic respiration, and the associated accumulation of the metabolic by-products. In particular, the accumulation of lactic acid in the blood is known to restrict activity rates in crustaceans and might represent a significant cost of rapping. Thus blood lactate concentration might play an important role in mediating the decision to give up for attackers. This might be because either they were unable to continue with sufficient vigor or the anticipated costs of eventual victory would be greater than the benefit arising from improved shell fit. One reason that defenders might give up is because the accumulation of lactate would prevent them from maintaining an adequate grip on their shell or because the anticipated costs would be greater than the perceived benefit of retaining their present shell (Elwood and Neil, 1992). During rapping, defenders remain tightly withdrawn into their shells and it was suggested that this might prevent them from respiring effectively (Elwood and Neil, 1992). To test if lack of oxygen influences the course and outcome of shell fights the physiological condition of crabs was altered prior to fights such that lactate levels were expected to be raised (Briffa and Elwood, in press). This involved four groups of staged fights in which the oxygen concentration of the water in which the crabs were maintained prior to fighting was varied. In the first group, OH, potential attackers were held in oxic water (100% saturated at 10◦ C) prior to fighting, whereas defenders were held in hypoxic water (30% saturated at 10◦ C). In the second group, HO, this was reversed, and in the third and fourth groups, HH and OO, both crabs were given hypoxic and oxic treatment, respectively. This allowed the two hypotheses, that the vigor of rapping was mediated by fatigue, and hence could act as a signal of stamina, and that defenders give up because rapping disrupts aerobic respiration, to be tested. It was found that attackers treated with hypoxic water rapped with decreased vigor and the increase in gap durations, both within bouts and from bout to bout, was more pronounced than those maintained in oxic water. Seemingly as a consequence of this decreased vigor, attackers that were treated with hypoxic water were less likely to evict the defender than those that were treated with oxic water, which shows that the condition of attackers does appear to affect the chance of success. In contrast to the prediction, however, hypoxic treatment applied

88

ROBERT W. ELWOOD AND MARK BRIFFA

to defenders had no effect on the likelihood of their being evicted. This appears to rule out the possibility suggested by Elwood and Neil (1992) that shell rapping reduces their ability to maintain an adequate grip on their shell by disrupting respiration but supports the view that rapping could advertise stamina. Recent work has involved analysis of lactic acid levels in the blood of crabs after fights (Briffa and Elwood, in preparation). In attackers it was found that lactate concentration increases with the total number of bouts of rapping that they have performed. Furthermore, lactate concentration in attackers that gave up was greater than in attackers that evicted the opponent, but there was no difference in lactate levels between defenders that retained their shell and those that were evicted. Again, this suggests that fatigue is important in mediating the behavior of attackers, but does not affect the decision of defenders. Thus, crabs in the two roles might use different decision rules. These findings support the idea that rapping advertises stamina but they do not rule out the possibility that rapping may advertise ability in some other way or that the primary function of rapping is to impose detrimental effects on defenders. The variation in the pattern of rapping shows marked similarities to changes that take place in the pattern of signaling during aggressive interactions in other animals. The vocalizations utilized by birds such as the great tit Parus major, for example, have been studied in detail (Lambrechts and Dhondt, 1988; Weary et al., 1991), these signals being used by males during territorial disputes over perching sites. Ryan (1988) suggested that the energetic costs of vocalization might limit the ability of some males to produce the full repertoire of calling patterns. Furthermore, Lambrechts and Dhondt (1988) found that the proportion of time spent singing decreased between bouts and suggested that “song switching” was a strategy to overcome this effect of fatigue. It is likely, however, that some of this variation is mediated by changes in motivation and a further study of this system (Weary et al., 1991) was not able to distinguish between the two possible causes. Recent analysis of the bout structure of the “perch coo” in male collared doves (Ballintijn and ten Cate, 1999) shows that the duration of gaps between coos increases with bout duration, whereas the sound percentage, an index of the energy used in each call, decreases. It appears then that in these other signaling systems there is similar variation in the pattern of signaling, during the course of the encounter, to that seen during shell rapping. In addition to variation in the timing of rapping, variation in the force with which the raps are performed might also be expected, both within fights and between attackers. In particular, the impact of rapping should vary with relative RHP, and the likelihood of effecting an exchange should increase with the impact of rapping. Furthermore, under the sequential assessment model,

INFORMATION GATHERING AND COMMUNICATION

89

one would expect that crabs that rapped with a high level of force should be able to effect an exchange more quickly than those that rap with less force, because the stronger signal would indicate higher fighting ability. Under the energetic war of attrition, however, the force with which rapping is performed should be less important because the level of advertisement is given by the number of performances. Measurement of the force with which individual raps are performed would be difficult because it would involve attaching a pressure sensor to one of the shells. This could obstruct the movement of either of the crabs or alter the perceived value of the shell to which it was attached and prevent the fight from proceeding as normal. An alternative approach, adopted by Briffa and Elwood (2000b), is to dampen the force of the raps by applying an elastic coating to the exterior of the shells supplied to attackers. Four groups of fights between pairs of crabs were staged in which the large crab of each pair was supplied with a shell that was either 25 or 50% adequate and had an elastic coating applied, either to the part of the surface that makes contact with the defender’s shell during rapping (damped) or to the back of the shell (undamped). This did not allow measurements of variation in the power of rapping during fights to be made, but did enable the effects of the force of rapping on (1) the likelihood of exchange and (2) the pattern of rapping to be investigated. The key finding of this study was that attackers that were supplied with undamped control shells were more likely to evict the defender than those that were supplied with the damped shells. The rubberized shells clearly prevented the attacker from achieving the normal impact when shell rapping and, thus, it was concluded that the impact is an important variable in shell fights. This is congruent with the finding that naked crabs, which do initiate fights and rap their soft abdomens against the shells of defenders but with little impact, are unlikely to effect an exchange (Elwood and Glass, 1981). In normal fights, the impact of rapping would vary with the effort with which it is performed and thus might contain information about the attacker. This information may be about RHP and/or motivation and could signal likely future costs to the defender should it persist in the fight. It was also found that the duration of fights was increased when the shells were damped and this is broadly congruent with the sequential assessment model because the reduced force of raps would indicate poor fighting ability and, thus, defenders would be less likely to give up early in the fight. It is also conceivable, however, that this effect would be expected under the energetic model if information about RHP is contained in the force of rapping as well as in the temporal pattern. Another possibility is that these differences between crabs supplied with damped and undamped shells are due to differing levels of detrimental effects that the raps cause because the magnitude of these

90

ROBERT W. ELWOOD AND MARK BRIFFA

FIG. 22. The mean number of raps per bout is lower for crabs with rubberized shells than with control shells, particularly during the first few bouts (From Briffa and Elwood, 2000b).

effects would normally be expected to vary with the effort with which the raps are performed. There is, however, no conclusive evidence to support any of the various suggested detrimental effects that might be caused by receiving raps. Because, the energy used, and hence the impact of rapping, appears to be an important feature of this activity, one would expect that attackers that are prevented from rapping with their usual level of impact would attempt to compensate in some way. They could either produce more raps per bout or increase their effort with each rap in an attempt to hit harder to overcome the effects of the rubber. The data, however, did not support the former possibility. Attackers supplied with rubberized shells performed fewer raps per bout than those supplied with control shells (Fig. 22). This suggests that these crabs might use more effort in each rap in an attempt to compensate for the damping. If this is the case, then the crabs would be subjected to higher levels of fatigue and be unable to perform as many raps per bout. The idea that fewer raps per bout indicate higher levels of fatigue was supported by a further finding that crabs supplied with 25% adequate shells performed more raps per bout that those occupying the heavier 50% adequate shells; this is because the lighter shells may be rapped using less energy. Thus, the differences in the temporal pattern of rapping may be caused by heightened levels of fatigue because of an increase in the effort that is put into the raps. This would be congruent with the analyses of the temporal pattern of rapping and with the idea that rapping is used to advertise stamina.

INFORMATION GATHERING AND COMMUNICATION

91

IX. NEGOTIATION OR AGGRESSION IN HERMIT CRAB SHELL FIGHTS? Hermit crab studies have proved useful in investigating the effects of resource value and fighting ability on the motivation of animals to perform assessment activities and to engage in apparently agonistic encounters. As mentioned previously, analysis of shell rapping provides good support for the aggression model as opposed to the negotiation model but the data from the other aspects of these encounters are less clearly interpretable. As described above, a problem inherent in the shell exchange system is that the change in resource value available to the defender, by occupying the shell discarded by the attacker, is likely to covary with the size difference between the two crabs, particularly during staged fights of the kind that have been described. In the field, crabs generally initiate shell fights so as to improve their shell fit by obtaining a shell larger than the one that they currently occupy. In this situation, if there is a large size difference between the two crabs, the defender is also likely to be in a shell that is too large and thus might improve its shell fit by occupying the smaller shell discarded by the attacker. If the size difference between the crabs is small, however, then the defender will be less likely to reduce the deviation from its preferred shell size by occupying the shell that the attacker discards. Because size difference contributes to relative RHP, under the aggression model attackers would be expected to initiate fights with crabs that are considerably smaller in size than themselves because the chance of victory would be high. Thus, crabs that attackers should choose to rap are also those which would be the most likely to gain by allowing an exchange to take place. This makes it difficult to distinguish between the two models of shell exchange either on the basis of which crabs attackers choose to fight or by the outcomes of these interactions. Hazlett’s (1996) study of the North American intertidal hermit crab Clibanarius vittatus, for example, found that occupied shells selected by crabs for the characteristic deep aperture investigation, which precedes rapping, were close to the preferred size of the attacker and also did not fit the defender, although the potential change in shell fit for the defender did not have any effect. This can be explained by both models: under the aggression model in terms of the likelihood of victory and under the negotiation model in terms of the suitability of the shell in question for the defender. If the aggressive interpretation was correct, however, one would expect that the relative size difference between the pairs of crabs would also explain why initiators selected certain shells, whereas in this analysis no such relationship was found, suggesting that the fighting abilities of the crabs were unimportant during this decision-making process. These findings then provide good support for the negotiation model in this species. Interestingly, Hazlett (1996) also found

92

ROBERT W. ELWOOD AND MARK BRIFFA

that the size difference between the crabs had a significant effect on which shells were rapped following the initial investigation and that the magnitude of gain available to noninitiators also appeared to affect this decision. Thus, although the initial decision to investigate shells follows the negotiation model, Hazlett’s data show that the decision to escalate the fight and commence rapping is congruent with both models. The question of aggression or negotiation might be resolved by analyzing the temporal pattern of rapping in the various species for which negotiation has been suggested. As described above, analysis of this type supports the aggression model, but not the negotiation model, in P. bernhardus. The advantage of analyzing the pattern of rapping is that clear predictions can be tested while avoiding the difficulties of interpreting results based on the gain in shell quality that may be made by each crab. According to the negotiation model, then, one would not expect to see any difference in the vigor of rapping between attackers that effect an exchange and those that do not in hermit crabs such as Clibanarius antillensis, Calcinus tibicen and Clibanarius vittatus, for which negotiation has been suggested (Hazlett, 1987, 1989, 1996). Furthermore, rapping should not be exhausting, such that initiators should rap at a constant rate throughout the encounter. Analyses of shell investigation and exchange in P. bernhardus demonstrate that hermit crabs make good subjects for the investigation of information gathering and decision making, and how these processes affect motivational state. The value of the resource may be readily manipulated and, during shell fights, the roles of signaler and receiver are fixed throughout the encounter. In addition, shell fights have been used to test the predictions of theoretical models of communication during fighting, such as the asymmetric war of attrition (Parker and Rubenstein, 1981), the energetic war of attrition (Payne and Pagel, 1996, 1997) and the sequential assessment model (Enquist and Leimar, 1983, 1987). Further resolution of the question of aggression or negotiation during shell fighting would consolidate the view that the shell exchange system provides a good empirical model of communication in an aggressive context. Signaling to advertise fighting ability appears to be a key function of shell rapping. It is likely, however, that there are additional components to the overall function of this activity, which allows attackers to evict defenders, that are not yet fully understood; why, for example, should defenders release their shells in response to rapping in the absence of any obvious detrimental effects? By conducting further investigations into the nature of shell exchange it may be possible to gain further insight into the function and “design” of repeated aggressive signals in general. Of particular interest might be the question of a dual “tactical” and “strategic” function that has been suggested for some forms of signal (Enquist and Leimar, 1987; Grafen, 1990; Bradbury and Vehrencamp, 1998). It is envisaged that such

INFORMATION GATHERING AND COMMUNICATION

93

activities might have direct effects on receivers in addition to advertising factors related to the signaler and this seems relevant to the situation during shell fights.

X. SUMMARY 1. Game theoretical models of animal contests assume that animals attempt to maximize individual fitness during contests with nonrelatives. The asymmetric war of attrition takes into account the probability that the ratio of cost to benefit for each opponent will be different, but the information that each contestant has about its relative cost to benefit ratio, in comparison to that of its opponent, is unlikely to be perfect. The sequential assessment model suggests that the accuracy of this information increases with each performance of the agonistic activity in question. Recent models, however, such as the energetic war of attrition, assume that aggressive signals are performed repeatedly because they accumulate to give a signal of stamina. 2. Hermit crabs interact in pairs in apparently agonistic encounters over the ownership of gastropod shells. The key activity during these interactions is “shell rapping,” where the attacker brings its shell rapidly and repeatedly into contact with that of the defender in a series of bouts. At the end of an encounter the attacker may evict the defender from its shell and permanently occupy, the new, empty shell. The evicted defender is then free to occupy the shell discarded by the attacker and the crabs are said to have “exchanged shells.” Under certain circumstances both crabs may benefit from these interactions. 3. This possibility of mutual gain from shell exchanges has lead to the hypothesis that crabs negotiate over the ownership of shells during these interactions. The negotiation model suggests that the function of shell rapping is to provide the nonrapping crab with information about the shell of the attacker. The defender would then be able to make a decision about whether to release its own shell on the basis of the change in shell quality that this would entail. Under the aggression model, however, it is expected that any information transferred by shell rapping concerns either the fighting ability of attackers, their motivation to continue, or a combination of these two factors. In addition, shell rapping might incur some direct detrimental effect on defenders. 4. The shell selection process in nonagonistic situations has been used to provide a model of information gathering and motivational change. A

94

ROBERT W. ELWOOD AND MARK BRIFFA

threshold model of motivational change during shell investigation relates two causal factors, the quality of the shell currently occupied and the quality of the potential new shell, to the decisions available to the crab: continue investigating the new shell, enter the new shell, or reject the new shell. This model allows various trajectories of motivational change to be calculated, which relate investigation time to the likely decision of the crab, for different scenarios defined by the quality of the crab’s original shell and that of the empty shell under investigation. This model was tested by probing the motivational state of crabs during shell investigation by (1) preventing adequate investigation and measuring persistence times and (2) using a novel stimulus to startle the crabs during shell investigation and measuring the time taken to return to investigatory behavior. Experiments conducted using these techniques confirmed that crabs use the two sources of information predicted by the model. 5. Motivational state is expected to vary in attackers during shell fights. An indication of this is provided by the duration of contests where attackers give up, but the majority of fights end in an eviction of the defender, hence there are too few data to test attacker motivation in this way. The novel stimulus technique was thus used to probe the motivational state of attackers during fights. It was found that the duration of the startle response was related to the potential gain, with those with the greater potential improvement in shell quality showing the shortest startle durations. Furthermore, attackers that were eventually victorious were shown to have a higher motivational state during the early escalated phase of the fight than those that did not evict the opponent. Surprisingly, the motivation of attackers did not appear to be related to the size difference between the two crabs. 6. The key determinants of the outcome of encounters are the power of the impact of the raps, the duration of the pauses between bouts of rapping, and the mean number of raps performed in each bout, with powerful and vigorous attackers being more likely to evict the opponent. Differences in vigor are also associated with the potential gain available to the attacker and with the relative size difference between the two crabs. Vigor also varies during the course of the fight: at the start of the fight the duration of pauses increases such that the vigor is de-escalated, however, at the end of the fight successful attackers increase the rate of rapping prior to evicting the opponent, whereas those that give up decrease the rate of rapping before doing so. The only game theoretical model of fighting that allows both increase and decrease in the rate of signaling during the contest is the energetic war of attrition. According to this model the function of repeated signals is to provide information about the stamina of receivers. That the rate of rapping is mediated by fatigue was confirmed by staging fights with crabs that had

INFORMATION GATHERING AND COMMUNICATION

95

been preexposed to hypoxic water and by analysis of lactic acid concentration in the hemolymph, following fights. We conclude that hermit crab shell fights represent a useful model of animal contests that involve communication between the two participants. First, the value of the resource in question and the relative fighting ability of the crabs can be easily manipulated. Second, unlike other signaling systems, the roles of attacker and defender are fixed throughout the fight such that the decision rules used by signalers and receivers can be examined separately.

Acknowledgments We thank the Department of Education for Northern Ireland for funding the early stages of the work discussed here and the BBSRC for funding recent studies.

References Abrams, P. (1978). Shell selection and utilisation in a terrestrial hermit crab, Coenobita compressus (H. Milne Edwards). Oecologia 34, 239–253. Ballintijn, M. R., and ten Cate, C. (1999). Variable element number of the perch-coo of collared doves (Streptopelia decaocto): Exhaustion and stuttering? Behaviour 136, 847–864. Bradbury, J. W., and Vehrencamp, S. L. (1998). “Principles of Animal Communication.” Sinauer Associates, Sunderland, MA. Brick, O. (1999). A test of the sequential assessment game: The effect of increased cost of sampling. Behav. Ecol. 10, 726–732. Briffa, M., and Elwood, R. W. (2000a). Analysis of the fine-scale timing of repeated signals: Does shell rapping in hermit crabs signal stamina? Anim. Behav. 59, 159–165. Briffa, M., and Elwood, R. W. (2000b). The power of shell rapping influences rates of eviction in hermit crabs. Behav. Ecol. 11, 288–293. Briffa, M., and Elwood, R. W. Cumulative or sequential assessment during hermit crab shell fights: effects of oxygen on decision rules. Proc. R. Soc. B. In press. Briffa, M. Elwood, R. W., and Dick, J. T. A. (1998). Analysis of repeated signals during shell fights in the hermit crab Pagurus bernhardus Proc. R. Soc. B 265, 1467–1474. Childress, J. P. (1972). Behavioural ecology and fitness theory in a tropical hermit crab. Ecology 53, 960–964. Clutton-Brock, T. H., and Albon, S. D. (1979). The roaring of red deer and the evolution of honest advertisement. Behaviour 69, 145–170. Culshaw, A. D., and Broom, D. M. (1980). The imminence of behavioural change and the startle response of chicks. Behaviour 73, 64–76. Dowds, B. M., and Elwood, R. W. (1983). Shell wars: Assessment strategies and the timing of decisions in hermit crab shell fights. Behaviour 85, 1–24. Dowds, B. M., and Elwood, R. W. (1985). Shell wars II: The influence of relative size on decisions made during shell fights. Anim. Behav. 29, 1239–1244. Elwood, R. W. (1995). Motivational change during resource assessment by hermit crabs. J. Exp. Mar. Biol. Ecol. 193, 41–45. Elwood, R. W., and Adams, P. (1990). How hermit crabs (Pagurus bernhardus) deal with obstructions in the apertures of shells. Irish Nat. J. 23, 180–185.

96

ROBERT W. ELWOOD AND MARK BRIFFA

Elwood, R. W., and Glass, C. W. (1981). Negotiation or aggression during shell fights of the hermit crab Pagurus bernhardus? Anim. Behav. 29, 1239–1244. Elwood, R. W., and Neil., S. J. (1986). Information and motivational change. In “Quantitative Models in Ethology” (P. W. Colgan and R. Zayan, eds.), pp. 97–107. Privat I. E. C., Toulouse. Elwood, R. W., and Neil, S. J. (1992). “Assessments and Decisions: A Study of Information Gathering by Hermit Crabs.” Chapman and Hall, London. Elwood, R. W., and Stewart, A. (1985). The timing of decisions during shell investigation by the hermit crab Pagurus bernhardus. Anim. Behav. 33, 620–627. Elwood, R. W., Marks, N., and Dick, J. T. A. (1995). Consequences of shell species preferences for female reproductive success in the hermit crab Pagurus bernhardus. Mar. Biol. 123, 431–434. Elwood, R. W., Wood, K. E., Gallagher, M. B., and Dick, J. T. A. (1998). Probing motivational state during agonistic encounters in animals. Nature 393, 66–68. Enquist, M., and Leimar, O. (1983). Evolution of fighting behavior: Decision rules and assessment of relative strength. J. Theor. Biol. 102, 387–410. Enquist, M., and Leimar, O. (1987). Evolution of fighting behavior: The effect of variation in resource value. J. Theor. Biol. 127, 187–205. Grafen, A. (1990). Biological signals as handicaps. J. Theor. Biol. 144, 517–546. Hammerstein, P. (1981). The role of asymmetries in animal contests. Anim. Behav. 29, 193–205. Hammerstein, P., and Parker, G. A. (1982). The asymmetric war of attrition. J. Theor. Biol. 96, 647–682. Hazlett, B. A. (1966). Social behavior of the Paguridae and Diogenidae of Curacao. Studies on the fauna of Curacao and other Caribbean Islands 88, 1–143. Hazlett, B. A. (1969). Further investigations of the cheliped presentation display in Pagurus bernhardus (Decapoda, Anomura). Crustaceana 17, 31–34. Hazlett, B. A. (1970a). The effect of shell size and weight on the agonistic behavior of a hermit crab. Z. Tierpsychol 27, 369–374. Hazlett, B. A. (1970b). Tactile stimuli in the social behavior of Pagurus bernhardus Behaviour 36, 20–48. Hazlett, B. A. (1978). Shell exchange in hermit crabs: Aggression, negotiation, or both? Anim. Behav. 26, 1278–1279. Hazlett, B. A. (1982). Resource value and communication strategy in the hermit crab Pagurus bernhardus (L.). Anim. Behav. 30, 135–139. Hazlett, B. A. (1983). Interspecific negotiations: Mutual gain in exchanges of a limiting resource. Anim. Behav. 31, 160–163. Hazlett, B. A. (1987). Information transfer during shell exchange in the hermit crab Clibanarius antillensis. Anim. Behav. 35, 218–226. Hazlett, B. A. (1989). Shell exchanges in the hermit crab Calcinus tibicen. Anim. Behav. 37, 104–111. Hazlett, B. A. (1996). Assessments during shell exchange in the hermit crabs Clibanarius vittatus: The complete negotiator. Anim. Behav. 51, 567–573. Hazlett, B. A., and Baron, L. C. (1989). Influence of shells on mating behavior in the hermit crab Calcinus tibicen. Behav. Ecol. Sociobiol. 24, 369–376. Herreid, C. F., and Full, R. J. (1986). Energetics of hermit crabs during locomotion: The cost of carrying a shell. J. Exp. Biol. 120, 297–308. Imafuku, M. (1983). New shell acquisition in the hermit crab Pagurus geminus. J. Ethol. 91– 100. Imafuku, M. (1984). Quality of shells occupied by Pagurus geminus: How many hermit crabs are satisfied with their shells? J. Ethol. 2, 31–36. Jackson, N. W. (1988). “Information gathering and decision making during shell selection by the

INFORMATION GATHERING AND COMMUNICATION

97

hermit crab Pagurus bernhardus.” Ph. D. Dissertation. The Queen’s University of Belfast, UK. Jackson, N. W., and Elwood, R. W. (1989). How animals make assessments: Information gathering by the hermit crab Pagurus bernhardus. Anim. Behav. 38, 951–957. Jackson, N. W., and Elwood, R. W. (1990). Interrupting an assessment process to probe changes in the motivational state. Anim. Behav. 39, 1068–1077. Lambrechts, M. M., and Dhondt, A. A. (1988). The anti-exhaustion hypothesis: a new hypothesis to explain song performance and song switching in the great tit. Anim. Behav. 36, 327–334. Maynard Smith, J. (1974). The theory of games and the evolution of animal conflict. J. Theor. Biol. 111, 475–491. Maynard Smith, J. (1976). Evolution and the theory of games. Am. Scientist 64, 41–45. Maynard Smith, J. (1982). Do animals convey information about their intentions? J. Theor. Biol. 97, 1–5. Maynard Smith, J., and Price, G. R. (1973). The logic of animal conflict. Nature 246, 15–18. Moorehouse, J. E., Fosbrooke, I. H. M., and Ludlow, A. R. (1987). Stopping a walking locust with sound: An analysis of variation in behavioral threshold. Exp. Biol. 46, 193–201. Neil, S. J. (1985). Size assessment and cues: Studies of hermit crab contests. Behaviour 92, 22–38. Neil, S. J., and Elwood, R. W. (1986). Factors influencing shell investigation in the hermit crab Pagurus bernhardus. Ethology 73, 225–234. Parker, G. A. (1974). Assessment strategy and the evolution of fighting behavior. J. Theor. Biol. 47, 223–243. Parker, G. A., and Rubenstein, D. I. (1981). Role assessment, reserve strategy, and acquisition of information in asymmetric animal contests. Anim. Behav. 29, 221–240. Payne, R. J. H., and Pagel, M. (1996). Escalation and time costs in displays of endurance. J. Theor. Biol. 183, 185–193. Payne, R. J. H., and Pagel, M. (1997). Why do animals repeat displays? Anim. Behav. 54, 109– 119. Ryan, M. J. (1988). Energy, calling, and selection. Am. Zool. 28, 885–898. Scully, E. P. (1979). The effects of gastropod shell availability and habitat characteristics on shell utilization by the intertidal hermit crab Pagurus longicarpus Say. J. Exp. Mar. Biol. Behav. Ecol. 37, 139–152. Turner, F. G., and Huntingford, F. A. (1986). A problem for game theory analysis: Assessment and intention in male mouthbrooder contests. Anim. Behav. 4, 961–970. Weary, D. M., Lambrechts, M. M., and Krebs, J. R. (1991). Does singing exhaust male great tits? Anim. Behav. 41, 540–542.

This Page Intentionally Left Blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 30

Acoustic Communication in Two Groups of Closely Related Treefrogs H. CARL GERHARDT DIVISION OF BIOLOGICAL SCIENCES UNIVERSITY OF MISSOURI COLUMBIA, MISSOURI

65211

I. INTRODUCTION Vocal communication in anurans (frogs and toads) has served as a premier system for studying basic issues in ethology, neurobiology, and evolution. A major part of this work has been conducted with species in, or closely related to, two groups of North American treefrogs (genus Hyla) that were originally recognized by Blair (1959). My goal is to provide a summary and progress report on the research conducted with these treefrogs over the past 40 years. I also compare some of the major results and conclusions of this research to those derived from studies of other taxa, primarily other anuran taxa and acoustic insects. The major advantage of studying these animals is that the acoustic modality dominates communication, which can be assessed in a robust fashion through playback experiments. After briefly describing systematic relationships, breeding ecology, vocal repertoires, and signaling interactions in the two groups of treefrogs, I focus on the main results of my research: (1) the identification of the acoustic properties of natural communication signals of males that are used for mating decisions by females; and (2) the estimation of female preference functions based on these acoustic properties. Comparisons of the acoustic criteria used in mate choice among different species can provide insights concerning broadscale patterns of evolutionary change in communication systems, especially if phylogenetic relationships can be established by other criteria. One of the major themes of this review, however, is that the preferences of even closely related species are often based on completely different acoustic criteria. Thus, sensory mechanisms underlying receiver biases and preferences might sometimes be even more evolutionarily labile than the acoustic structure of the signals. 99

C 2001 by Academic Press Copyright  All rights of reproduction in any form reserved. 0065-3454/01 $35.00

100

H. CARL GERHARDT

Information from playback experiments also bears on issues concerning sexual selection by female choice in contemporary populations. First, these data (preference functions), in conjunction with estimates of signal variation among males, can be used to predict patterns of selection on signals by female mate choice (Gerhardt, 1991). Second, the attributes of the signalers that produce signals with especially attractive properties can be analyzed in order to predict how mating with these males might affect the female’s fitness either directly or indirectly. Indeed, the anuran system is especially suitable for robust experimental tests of these predictions (e.g., Welch et al., 1998). Although the present-day consequences of mating preferences may not, of course, be the same as those occurring in the past, the same kinds of selective forces are likely, in general, to play a significant role in the evolution of preferences. Another theme of this review is that realistic assessments of acoustic selectivity of anurans and other animals in nature require that investigators also generalize the results of their playback experiments by varying stimulus intensity over the range experienced in nature by the species in question. As I will show, selectivity can be affected significantly by both the relative and absolute playback levels of alternative stimuli. Other factors that affect selectivity include the number of signals presented in a playback test, their timing relationships, the background noise level, and temperature (Gerhardt 1978a, 1987; Gerhardt and Mudry, 1980; Gerhardt and Klump, 1988a; Dyson and Passmore, 1988a,b; Marquez ´ and Bosch, 1997a; Wollerman, 1999). Thus, the degree of receiver selectivity estimated in most simplified laboratory situations must seldom, if ever, be fully realized in the field. Furthermore, the translation of acoustic selectivity into actual mating choices will also be affected to some extent by the operational sex ratio, the effects of predators, and the assessment patterns used by females (e.g., Telford et al., 1989; Morris, 1989; Morris and Yoon, 1989; Jennions and Petrie, 1997; Wagner, 1998). Anurans have been frequent subjects of studies of the evolution and mechanisms of acoustic communication for a number of reasons. First, males produce a small set of stereotyped, unlearned signals (Gerhardt, 1988, 1994a). The analysis of calls recorded at close range can quantify the potential for particular acoustic properties to convey biologically important information (Gerhardt, 1991; Runkle et al., 1994; Bee and Gerhardt, submitted). Analysis of more distantly recorded calls can show how signal intensity and acoustic properties change as these sounds propagate through the environment (e.g., Gerhardt and Klump, 1988b; Ryan and Sullivan, 1989). Second, many pairs of species are highly genetically compatible (e.g., Mecham, 1965; Blair, 1972). Both the calls and phonotactic selectivity of interspecific hybrids have been studied (e.g., Gerhardt, 1974a; Doherty and Gerhardt, 1984), and these systems offer opportunities to gain knowledge

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

101

about the genetic bases of differences in signals and female selectivity comparable to that available for some insect systems (Shaw, 1996; for review, see Ritchie and Phillips, 1998). Third, male and female frogs reliably respond to playbacks of prerecorded natural calls and synthetic calls. Thus, the presumed functions of different signals in the repertoire can be confirmed, and playbacks of synthetic calls can determine if the potential information encoded in particular acoustic properties is actually used by receivers. These kinds of studies are particularly powerful when applied to treefrogs because there is no evidence that cues from any other sensory modality are used in mate choice. For example, females of at least one species (Hyla cinerea) sometimes attempt to induce mating with males of another species (H. gratiosa) tethered near a source of conspecific calls, even though these males differ significantly in size, odor, and appearance from conspecific males (Gerhardt, 1968). The strong reliance on acoustic signals for mate choice probably explains why these frogs show preferences for signals that differ in very subtle ways. Fourth, anurans fertilize eggs externally, and hence, in most species, paternity is certain because the male of an amplexed (mated) pair fertilizes all of the female’s eggs. External fertilization also makes it possible to assess experimentally the potential direct (e.g., fertilization success) and indirect (fitness differences in offspring) consequences of mating with males with particular attributes (including advertisement calls) without the complication of possible maternal effects (e.g., Bourne, 1993; Welch et al., 1998). Fifth, numerous species of frogs and toads have wide geographical distributions, which often overlap with those of other species in the same group. These distribution patterns provide opportunities to explore broadscale patterns of variation in signals and female selectivity and to test hypotheses about the possible roles of interspecific interactions and environmental factors in shaping or constraining the evolution of acoustic communication (e.g., Nevo and Capranica, 1985; Ryan et al., 1996; for review, see Gerhardt and Schwartz, 1995). Still another advantage of studying vocal communication in anurans is that an extensive body of knowledge exists about their auditory system. Indeed, another theme of this review is that the acoustic specifications for behavioral selectivity have often been used to frame questions about acoustic pattern recognition that have been approached with biophysical and neurophysiological techniques (e.g., Capranica and Moffat, 1983; Feng and Schellart, 1999). In turn, mechanistic studies can identify constraints on receiver selectivity that are imposed by the sensory system. Research with anurans, and especially treefrogs, has also contributed to our understanding of the mechanisms of sound localization (for reviews, see Feng and Schellart, 1999; Lewis and Narins, 1999), but this topic is beyond the scope of this review.

102

H. CARL GERHARDT

II. Hyla cinerea SPECIES GROUP Blair (1959) used call structure and, to a lesser extent, information about genetic compatibility to define three groups of North American treefrogs. Members of the Hyla cinerea group are H. cinerea (green treefrog), H. gratiosa (barking treefrog), and H. andersonii (pine barrens treefrog). Little controversy exists about the close relationship between H. cinerea and H. gratiosa. Not only do their calls have the same basic structure, but also they are highly genetically compatible, if not interfertile (Mecham, 1965; but see Lamb and Avise, 1986). Phylogenetic analyses, including those based on “total evidence,” invariably place these two species as sister taxa (Cocroft, 1994; da Silva, 1997). The phylogenetic position of H. andersonii relative to H. cinerea and H. gratiosa is more problematic, and cladistic analyses of several combined data sets group H. andersonii with core members of the H. versicolor group (Section III). However, in these analyses, the only support for the placement of H. andersonii is a set of polymorphic allozyme loci, for which methods of analysis are problematic (Cocroft, 1994). Artificial crosses indicated a high level of genetic compatibility between H. andersonii and H. cinerea (Gerhardt, 1974b), and a natural hybrid between these two was reported by Anderson and Moler (1986). These same authors, however, also found a natural hybrid between H. andersonii and H. femoralis. My criteria for including H. andersonii in the H. cinerea group are bioacoustic: (1) the overall call structure (as opposed to the values of quantitatively varying, constituent properties) of H. andersonii is more similar to that of H. cinerea than to the calls of any other North American treefrog (Section II.A), and (2) females of H. andersonii and H. cinerea show phonotactic responses to each other’s calls under some conditions (Section II.D). The breeding seasons (spring and summer) and distributions of the three species overlap extensively. Much of the range of H. gratiosa and H. andersonii falls within that of the more widely distributed H. cinerea (Conant et al., 1998). Scattered populations of H. gratiosa occur in more inland locations than do those of H. cinerea, and H. andersonii is the only member of the group found in southern New Jersey, United States. Whereas H. cinerea usually breeds in permanent ponds and lakes, H. gratiosa usually forms choruses in fish-free, temporary, or semipermanent ponds. Habitat disturbances frequently bring the two species into contact, however, and numerous hybrids and backcross products have been found in nature (Gerhardt, 1974a; Gerhardt et al., 1980). Indeed, introgressive hybridization has been documented over an extended period of years in a series of artificial ponds near Auburn, Alabama, United States (Mecham, 1960; Lamb and Avise, 1986; Schlefer et al., 1986). Mismatings between females of H. gratiosa and males of H. cinerea were most common at the Auburn ponds, probably

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

103

because males of H. gratiosa typically call from floating positions, whereas males of H. cinerea usually call from the ground or emergent vegetation along the shoreline (Lamb and Avise, 1986). Females of H. gratiosa thus had to pass near calling males of H. cinerea, which often have silent, satellite males associated with them, in order to reach conspecific males. Calling males, and especially satellite males, frequently attempt to mate with other frogs of comparable size that move toward them (Perrill et al., 1978). Although extensively sympatric with H. cinerea and H. gratiosa in the southeastern United States, H. andersonii has a specialized habitat (small pools and streams in thick vegetation in otherwise xeric habitats; Bullard, 1965) that provides considerable ecological isolation from these other species (Gerhardt, 1974b). Males call from the ground or from emergent vegetation near these pools and streams, or from man-made ditches. A. ADVERTISEMENT CALLS All of the advertisement calls of the members of this group are relatively short, broadband (noisy) signals, whose physical properties are described in the next two sections. Mean sound pressure levels (SPL in decibels [dB] re 20 ␮Pa, fast root-mean-square) at 50 cm from males of H. cinerea and H. gratiosa were 89 and 90 dB SPL, respectively; maximum peak levels were 102.5 and 107.5 dB SPL, respectively (Gerhardt, 1975). The mean peak level of the calls of three males of H. andersonii at the same distance was 98 dB SPL (Gerhardt, 1974b). 1. Spectral Properties Typical advertisement calls produced by males of H. gratiosa from their usual floating position consist of a series of harmonically related components. The fundamental (lowest) frequency of the call is also the dominant frequency (contains the most acoustic energy) and ranges from about 400 to 550 Hz (Figs. 1A, 2A, 8A). The third or fourth harmonic is also emphasized, giving the call its typically bimodal spectrum (Oldham and Gerhardt, 1975). Calls produced from trees or bushes or from the ground along shoreline are much more variable than typical calls and tend to be much noisier (Fig. 2B). Playback tests of two females indicated that such calls were unattractive relative to typical calls (Oldham and Gerhardt, 1975). In H. cinerea, the frequency components displayed in a sonogram of a typical call appear to be harmonics of a missing fundamental frequency of about 200–450 Hz. In fact, these components are either harmonics or sidebands resulting from the modulation of a harmonic series having a fundamental frequency three times higher (Oldham and Gerhardt, 1975; Gerhardt, 1998). I have also recorded two exceptional individuals of H. gratiosa that

104

H. CARL GERHARDT

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

105

produced calls with this kind of structure (harmonics and sidebands) (Gerhardt, 1981b). The low-frequency peak in H. cinerea nearly always consists of a single component, the fundamental frequency of the harmonic series; its frequency ranges from about 640–1340 Hz (Oldham and Gerhardt, 1975). The high-frequency peak, whose frequency ranges from about 2200– 3600 Hz, is made up of two or three components with about the same relative amplitude. The low- and high-frequency peaks also typically have about the same relative amplitude. Interspecific hybrids between the two species also produce calls with bimodal spectra: the two peaks are generally intermediate with respect to those in the parental species (Fig. 3; Gerhardt, 1974a; Gerhardt et al., 1980). Some male hybrids, both confirmed and putative ones recorded in the field, produce calls consisting of a harmonic series as in typical H. gratiosa (Fig. 3, parts 1–5B), whereas others produce harmonics and sidebands as in H. cinerea (Fig. 3, parts 7–11B). Males of H. andersonii also produce calls that consist of harmonics and sidebands (Figs. 1 and 2). In this species, the fundamental frequency (mean of about 1100 Hz) is usually dominant, and the second harmonic or a sideband of the second harmonic (range of about 2000–3000 Hz) is secondarily emphasized (Gerhardt, 1974b). Sidebands between these two peaks are not always strongly attenuated, and hence the bimodality of the spectrum is not always as distinctive as in H. cinerea and H. gratiosa. The sideband nature of several components is easily seen in variant calls of the same male in Fig. 2D,E. 2. Temporal Properties The mean duration of the advertisement calls of the three species range from 120 ms to 200 ms; the shortest individual calls were about 100 ms, and the longest, about 300 ms (Fig. 1; Gerhardt, 1970; 1974b; Oldham and Gerhardt, 1975). The calls of H. cinerea and H. gratiosa have relatively abrupt, pulsatile beginnings (one to three cycles), which are especially salient

FIG. 1. Typical advertisement calls of (A) Hyla gratiosa, (B) Hyla cinerea, and (C) and (D) Hyla andersonii. The first column shows oscillograms of one complete call. The second column is a series of oscillograms with an expanded time base to show two cycles of the repeating waveform. The period of the repeating waveform (indicated by the horizontal brackets) is reflected in the power spectra of the third column. That is, the frequency interval between frequency components is equal to the reciprocal of the period of the repeating waveform. In H. andersonii, the two calls (C and D) were produced one after the other by the same male; notice that the period of the repeating waveform and frequency interval between components differs, but that the frequencies of the fundamental (at about 1100 Hz) and second harmonic (at about 2200 Hz) remain the same. From Figure 11 of Gerhardt (1998), with permission.

106

H. CARL GERHARDT

FIG. 2. Sonograms of typical advertisement calls of (A) Hyla gratiosa, (B) Hyla gratiosa (“tree” call), (C) Hyla cinerea, and (D,E) Hyla andersonii. Notice that the different calls of the same male of the last species differ not only in frequency intervals, but that some components in the call labeled E change abruptly in frequency with time. These components are sidebands, and the components that do change gradually in a smooth fashion (indicated by arrows) are harmonics of a series with a fundamental frequency of about 1100 Hz.

in H. gratiosa (Fig. 1A,B). The calls of H. andersonii have a smooth gradual onset and slightly more abrupt offset (Fig. 1C,D). The fine time-structure of the waveform in most of the call (after the pulsatile beginning in H. cinerea and H. gratiosa) is quasiperiodic. By this I mean that a complex waveform repeats at a regular rate (the reciprocal of the period), although its exact shape changes slightly from cycle to cycle. The repetition rate of the repeating waveform is equal to the fundamental frequency in typical calls of H. gratiosa that consist of a harmonic series; the repetition rate equals the frequency interval between harmonics and sidebands in the calls of the other two species (Fig. 1; Oldham and Gerhardt, 1975; Gerhardt, 1978a; 1998). Males of all three species repeat these calls in a more or less regular fashion: about 55–60 calls/min in H. gratiosa, slightly faster (about 85 calls/min) in H. cinerea, and faster still (146 calls/min) in H. andersonii (Gerhardt, 1974b; Oldham and Gerhardt, 1975). B. OTHER CALLS IN THE REPERTOIRE Neighboring males of H. cinerea often produce aggressive or encounter calls (Wells, 1977), which can also be elicited by playbacks of advertisement calls (Gerhardt, unpubl. data). Sometimes, and especially early in the evening, isolated males add a few aggressive calls to the end of a long series of advertisement calls. Aggressive calls are produced by pulsing the advertisement call at rates of about 50 Hz. As shown by the oscillogram and sonograms of Fig. 4B, the depth of modulation in aggressive calls is about 85–95%, and the spectral properties of the two kinds of calls are similar. Males occasionally produce aggressive calls that are pulsed for only part of

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

107

FIG. 3. Sonograms of advertisement calls of Hyla gratiosa, Hyla cinerea, and natural hybrids. Sonograms labeled 1–5B show hybrid calls having the same kind of structure (harmonic series) as that of typical calls of H. gratiosa. Sonograms labeled 7–11B show hybrid calls having the same kind of structure (harmonics and sidebands) as that of typical calls of H. cinerea. Sonograms with different letters show call variants produced by the same individual. Modified from Figure 1 of Gerhardt et al., 1980.

their total duration (Gerhardt, 1978b). Aggressive calls are usually slightly shorter and repeated at higher rates (157 calls/min) than are advertisement calls (Oldham and Gerhardt, 1975). “Release” calls, which are produced by fighting males or by males and unreceptive females that are clasped by a male, have a structure that is similar to that of aggressive calls. The pulsing of release calls is, however, generally more irregular in both rate and depth than that in aggressive calls (Gerhardt, unpubl. data). I have rarely heard and never recorded the aggressive calls of H. gratiosa. These signals were pulsed, and sounded similar to the (pulsed) release calls of this species (Gerhardt, unpubl. data). Aggressive calls of H. andersonii have not been described in the literature

108

H. CARL GERHARDT

FIG. 4. Advertisement and aggressive calls of the green treefrog, Hyla cinerea. (A) Advertisement call: Left, oscillogram; right, sonogram; (B) Aggressive call of the same male: Left, oscillogram; right, sonogram; (C) Synthetic calls used to test females for preferences based on pulsing (amplitude modulation) at the rate (50 Hz) typical of aggressive calls. Left, unmodulated call (standard advertisement call); right, calls with modulation depths of about 10% (top) and 50% (bottom). Females did not show a preference for the unmodulated call unless the modulation depth was at least 50%.

and, in my limited field experience with this species, I have not heard such signals. Males of H. cinerea and H. gratiosa also produce distress calls, which are loud, broadband screams made with the mouth open (Gerhardt, unpubl. data). These signals are produced when an animal is seized by a predator, especially a snake, and can occasionally be elicited by human handling. No evidence exists that distress calls in any anuran species elicit any kind of

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

109

reaction in other conspecific individuals (Bogert, 1960). Recent speculation is that distress calls could startle the predator or attract another predator, whose attack on the first predator may give the frog a chance to escape (LeVering, 1997). C. SIGNALING INTERACTIONS As in many other species of frogs and some acoustic insects, neighboring males of species in the H. cinerea group tend to produce advertisement calls in an alternating fashion (for reviews, see Narins, 1992; Greenfield, 1994). Alternation is not perfect in the sense that equal periods of silence occur between the signals of the two males, and one male’s calls (lagging calls) are usually produced soon after those (leading calls) of the other male. Stereo recordings of pairs of neighboring H. cinerea were analyzed by Klump and Gerhardt (1992), who found that pairs of interacting males usually change the leadership role within and between bouts of calling. Some interesting exceptions included pairs of males that did not alternate very well, and other pairs, in which there was a consistent leader–follower relationship. The leading male ignored the calls of the other male, which seemingly waited for the other male to trigger its calls. Males also varied considerably in the degree to which they timed their calls relative to playbacks of conspecific advertisement calls (Klump and Gerhardt, unpubl. data). I return to this topic in the general discussion of the effects of timing interactions on female preferences (Section IV.B). D. PHONOTACTIC RESPONSES TO PRERECORDED CALLS 1. Responses to Heterospecific Advertisement Calls Females of all three members of the H. cinerea group sometimes show phonotactic approaches to playbacks of the calls of other members of the same group (Gerhardt, 1974b; Oldham and Gerhardt, 1975; Gerhardt et al., 1994b). These results suggest that females potentially treat the signals of other members of the group as signals rather than biologically meaningless noise, at least in such no-choice situations (Gerhardt, 1982). Another possibility, which I do not think is likely, is that responses to heterospecific signals are an artifact of testing highly motivated females. Most females tested in playback experiments are captured in amplexus and separated from the male prior to testing. Finding females before they mate is difficult, and these animals are somewhat less likely to respond phonotactically than females from mated pairs. This state of affairs raises an important question. Because females are seemingly committed to egg laying by the

110

H. CARL GERHARDT

time they are mated (they will oviposit within a matter of hours even if separated from the male), might mated females be less selective than unmated females, which are presumably in the early stages of phonotactic readiness? Murphy and Gerhardt (1996) tested females of H. gratiosa that were captured soon after they entered a breeding pond, and then retested the same females later in the evening after they had mated. No difference in the selectivity was detected: females that failed to discriminate between synthetic calls late in the evening after mating also had failed to do so when tested just after entering amplexus. Although not reported in that paper, Murphy and Gerhardt also found that females of H. gratiosa tested before mating were as likely to respond phonotactically to playbacks of the calls of H. cinerea as they were after mating. Thus, in this species at least, responses to heterospecific calls do not appear to be an artifact of the female’s reproductive condition. More experiments of this sort with other species are needed to generalize this result. Phonotactic readiness declines as oviposition begins, so a failure to respond could either reflect this reduced motivational state or the ineffectiveness of the stimulus. It is thus important to conduct a control test, which consists of playing back a signal known to be highly attractive, after a female fails to respond in a single-stimulus test (e.g., Gerhardt and Klump, 1988b; Ryan and Rand, 1993a). 2. Responses to Chorus Sounds Single-speaker tests of females of H. gratiosa indicate that females orient and show phonotaxis toward a loudspeaker playing back conspecific chorus sounds at 38–40 dB SPL (Gerhardt and Klump, 1988b). The maximum distance at which the SPL of chorus sounds would drop to 38–40 dB is estimated to be somewhat less than 300 m from the pond, suggesting that gravid females could potentially use chorus sounds to guide them to breeding ponds where conspecific males are present. A field study in which males were captured as they arrived at a breeding pond, however, found that the numbers and arrival times of females were independent of whether or not the captured males were allowed to form a chorus (Murphy, submitted for publication). Whether or not they use sound to locate choruses, females of H. gratiosa can distinguish a mixed-species chorus containing conspecific calls and those of H. cinerea from a pure chorus of H. cinerea (Gerhardt and Klump, 1988b). 3. Masking of Individual Calls by the Chorus and Release from Masking Detection of the signals of an individual against a background chorus of conspecifics is a particularly difficult task because the spectral properties

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

111

FIG. 5. Masking of individual calls by a conspecific chorus in Hyla cinerea. (A) Sonogram showing the chorus background and two advertisement calls (marked by arrows) of an individual male when the SPL of the chorus sound was 87 dB and that of the individual calls was 81 dB, measured at the release point of the female., (B) Sonogram showing the chorus background and two advertisement calls (marked by arrows) when the SPL of the chorus sound and that of the individual calls was 87 dB., (C) Polar diagram of the playback arrangement. Chorus sound was played from two speakers positioned at 0 and 180◦ (white rectangles). The calls of an individual male were played back from a speaker (gray rectangle) immediately adjacent to the speaker playing back chorus sound at 0◦ . The dots indicate the locations at which females left the circular arena, and the arrow shows the mean directional vector. Its length (r-value) quantifies the directionality of the sample of females. In these tests, the SPL of the individual calls was 6 dB less than that of the chorus sound at the female’s release point, and the distribution of directions taken by the 15 females was not significantly different from that of a uniform distribution. (D) Polar diagram and symbols as in (C). In these tests, the SPL of individual calls was the same as that of the chorus sound, and the distribution of directions taken by the 15 females was significantly different from a uniform distribution, and most females left the arena in the sector where the speaker playing back the individual calls was located or in an adjacent sector. Modified from Figure 1 of Gerhardt and Klump (1988a).

of the signals and noise are so similar. In tests of H. cinerea, Gerhardt and Klump (1988a) played back chorus sounds from two speakers on opposite sides of the release point in a circular arena (Fig. 5). A third speaker, located immediately adjacent to one or the other of these speakers, played back the advertisement calls of a single male. Eighty percent of the females showed accurate phonotaxis to the single male’s calls when their SPL was about the same as that of the chorus noise; only 47% of the females did so when the

112

H. CARL GERHARDT

SPL of the single male’s calls was 6 dB lower than that of the chorus noise. Considering the normal intermale spacing in this species, only three to five males are likely to produce calls that equal or exceed the level of the chorus background from any one point in a dense chorus (Gerhardt and Klump, 1988a). This means that a female must move around in the chorus if she is to assess more than just a few males before choosing one. Another study, using broadband masking stimuli, showed that females of H. cinerea can detect the calls of individuals better if the noise source is spatially separated from the signal sources by 90◦ (Schwartz and Gerhardt, 1989). This improvement in detection, which is termed “release from masking,” is not very strong (equivalent to about a 3-dB increase in the signal-tonoise ratio). Although they detected signals at lower signal-to-noise ratios when signals and noise were spatially separated, females did not discriminate between advertisement and aggressive calls, which are easily distinguished in quiet conditions (Section II.D.4). 4. Preferences for Advertisement Calls to Aggressive Calls Based on field observations of interacting males and phonotactically active females, I assumed that the unpulsed call of H. cinerea was the advertisement call and that the pulsed call was an aggressive signal (Fig. 4). This hypothesis was supported by tests of 16 females, which chose the advertisement calls of a conspecific male rather than his aggressive calls in 63 of 65 tests (Oldham and Gerhardt, 1975). As mentioned above, aggressive calls are so rare that they have not even been recorded in H. andersonii and H. gratiosa. The relative unattractiveness of aggressive calls is widespread among anurans, and avoiding males producing such calls might reduce the chances that the female is attacked by aggressively interacting males (Section IV.A). 5. Preferences for Conspecific Advertisement Calls to Heterospecific Calls and Hybrid Calls The great majority of females of all three species chose prerecorded conspecific advertisement calls to prerecorded calls of other species in the group. In these early tests, speakers were separated by 4 m, and after a female responded to one speaker, I frequently switched the stimuli between speakers without replacing the female at the central release point. A response to the closer, louder call was termed a “secondary response.” In tests of 31 females of H. cinerea from Georgia, United States, for example, I recorded 277 responses to speakers playing back conspecific calls and only two responses to calls of H. gratiosa (Gerhardt, 1968). Of these responses to conspecific calls, 194 responses began from immediately in front of the speaker emitting the calls of H. gratiosa after the sounds had been switched between the speakers. Secondary responses were rare. Subsequent experiments showed that females reliably chose synthetic calls with spectral properties modeled after

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

113

conspecific signals that were 12 dB less intense than synthetic calls with spectral properties typical of H. gratiosa (Gerhardt, 1982); some females chose the “heterospecific” signal when the SPL of the conspecific alternative was reduced by another 6 dB. Comparable results are available for females of H. gratiosa: all 48 responses by 19 females were to conspecific calls rather than to the calls of H. cinerea (Oldham and Gerhardt, 1975). Subsequent tests using synthetic calls showed that the spectral differences between the calls of the two species afford the same 12-dB margin for error as in H. cinerea (Gerhardt, 1982). Females of H. andersonii were somewhat less selective than females of H. cinerea and H. gratiosa in situations where SPL differences favored the heterospecific signal. Although 12 females of H. andersonii chose conspecific calls in 43 of 47 tests (and all initial responses were to conspecific calls), nine of the females made “secondary” contacts with the speaker playing back H. cinerea calls (Gerhardt, 1974b). By comparison, six females of H. cinerea responded exclusively to conspecific calls in 18 tests (10 of these responses began in front of the speaker playing back H. andersonii calls after a speaker switch) (Gerhardt, 1974b). Secondary contacts were rare. Tests of more subtle differences in natural calls indicate a species difference in selectivity between H. cinerea and H. gratiosa. Whereas 38 females of H. gratiosa responded exclusively (112 tests) to conspecific calls rather than to hybrid calls, females of H. cinerea frequently chose hybrid calls over conspecific calls (Gerhardt, 1974a). Females responded to hybrid calls in 4 of 14 initial trials and in 31 of all 78 tests. Indeed, two females responded exclusively to hybrid calls. The relative attractiveness of hybrid calls to females of H. cinerea was also corroborated by the fact that in 27 of 31 trials, eight females chose hybrid calls over conspecific aggressive calls (Gerhardt, 1974a). Females of both species also discriminated between the calls of different hybrids, preferring the hybrid calls with a frequency spectrum that more closely resembled that typical of their respective conspecific calls. Taken as a whole, these results show that although females within the H. cinerea species group are, in general, highly selective in choosing conspecific calls, there are nevertheless species differences in phonotactic selectivity. Furthermore, as documented in the next sections, extensive tests using synthetic calls reveal that females of H. cinerea and H. gratiosa also differ in the acoustic criteria underlying their preferences for different signals. E. PHONOTACTIC RESPONSES TO SYNTHETIC CALLS 1. Generating and Testing Standard Synthetic Calls The main purpose of playback experiments employing synthetic calls is to identify the physical properties (acoustic criteria) used by females in mate

114

H. CARL GERHARDT

choice. The first step is to synthesize sounds, which, in a direct competition, are as attractive as prerecorded natural calls. Thus, we can be sure that such “standard” synthetic calls contain all of the features that account for the attractiveness of natural signals. Failure to do so can lead to erroneous conclusions. If, for example, synthetic signals are of marginal effectiveness, many females will not respond at all, and those that do might be less likely to discriminate on the basis of small differences than if both alternative signals were highly attractive (Schmitt et al., 1993). Problems of interpretation could also arise if synthetic standard calls were to derive their attractiveness from inadvertently using supernormal values of one or more properties. In my studies, the properties of standard signals had values close to the means of the same properties in the calls of conspecific males from the same populations from which females were collected. No standard call was ever more attractive than a prerecorded, natural exemplar. A significant, general result of these experiments is that not all of the stereotyped properties of natural advertisement calls are required for a standard synthetic call to be equally potent. This discovery is consistent with the results of many classical ethological studies showing that very crude models are often as effective as very realistic ones in eliciting appropriate behavioral responses (Tinbergen, 1951). Another important point, which I emphasize in this review, is that the effectiveness of an acoustic signal also depends on whether females are presented with that signal alone or given a simultaneous choice of two or more signals. Once a standard call has been generated, the next task is to synthesize alternative synthetic calls in which one or more acoustic properties are varied in a systematic fashion. These alternative signals are then tested against the standard synthetic call. I usually first test relatively large differences that result in a preference for or against the standard call. Reducing the magnitude of the differences until the preference is abolished then provides estimates of just-meaningful differences (Nelson and Marler, 1990). Such estimates are useful for studies of both proximate mechanisms and sexual selection. In general, however, the acoustic differences in an array of synthetic signals will depend on the questions being addressed by the investigator. 2. Behavioral Thresholds and Preferences Based on Spectral Structure A diagnostic property of the advertisement calls of H. cinerea and H. gratiosa is their bimodal spectrum (Figs. 1 and 2). Are both of these spectral peaks necessary to elicit phonotaxis? Do females prefer calls having just the low-frequency peak to calls with just the high-frequency peak? Are calls with both spectral peaks more attractive than calls with a single peak? Are females sensitive to differences in the relative amplitudes of the two peaks? Experiments with H. gratiosa showed that females can be attracted

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

115

by calls having only the low-frequency peak, that calls with a bimodal spectrum are preferred to calls with a single peak, and that calls with just the low-frequency peak are preferred to calls with just the high-frequency peak (Gerhardt, 1981b). More quantitative answers to the questions just posed have been provided by extensive studies of H. cinerea. Single-speaker playbacks estimated behavioral thresholds (minimum SPL at which females of H. cinerea reliably show phonotaxis) for synthetic calls with different components of natural advertisement calls (Fig. 6A) (Gerhardt, 1974c, 1976, 1981a). First, notice that the threshold for the combination of two high-frequency peaks (2700 + 3000 Hz) is significantly lower than that for the single high-frequency peak (3000 Hz). This suggests that female preferences are likely to be influenced by the 300 Hz waveform periodicity of the natural advertisement call, which is represented in the two-component synthetic call by the 300 Hz beats arising from the frequency difference in the 2700 + 3000 Hz stimulus. I return to this issue later in Section II.E.6. Second, notice that the threshold for a stimulus of 900 Hz alone was about the same as that for the combination of 900 and 3000 Hz. Furthermore, the thresholds for both of these stimuli were more than 40 dB less than that for the component of 3000 Hz alone. These estimates predict that energy (contributed by a single component) in the high-frequency part of the spectrum of the call will not affect preferences unless signals are played back at very high levels (90 dB SPL), corresponding to locations very close to a calling male. This last prediction was not confirmed by subsequent experiments. Indeed, two-speaker (choice) experiments showed that the combination of 900 + 3000 Hz components was more attractive than a 900 Hz component alone at playback levels of about 55 dB and higher (Gerhardt, 1981a; Fig. 6B). Females also based preferences on a deficiency in the relative amplitude of the high-frequency peak at playback levels as low as 72 dB SPL (Fig. 6C, left panel). Deficiencies in the low-frequency peak affected female preferences over the entire range of playback levels, including a level corresponding to the behavioral threshold for the low-frequency peak alone (Fig. 6C, right panel). To summarize, females are attracted to low-frequency signals corresponding to the low-frequency spectral peak in the conspecific advertisement call at levels as low as 48 dB SPL. This component alone can thus elicit phonotaxis at a distance (maximum of about 60 m, assuming spherical spreading alone). As females approach the chorus or individual males (and SPL at the female’s position increases), their preferences would then be influenced by the high-frequency peak in the conspecific call. Thus, although ineffective by itself in eliciting phonotaxis, high-frequency energy in combination with low-frequency energy affects selective phonotaxis. At even closer distances,

116

H. CARL GERHARDT

FIG. 6. Behavioral relevance of the two spectral peaks in the advertisement call in Hyla cinerea. (A) Estimated behavioral thresholds for synthetic calls with the spectra diagrammed. (B) Results of two-speaker playback experiments in which females were offered a choice of synthetic calls: 0.9 kHz alone versus 0.9 + 3.0 kHz. Notice that although females had behavioral thresholds of 78 to 90 dB SPL to high-frequency components alone, the combination of low and high components elicited preferences at 54 dB SPL. (C) Summaries of experiments in which females were given choices between synthetic calls differing in the relative amplitudes of the two peaks; the overall playback level of the two alternatives (bottom), was equalized at different sound pressure levels at the female’s release point.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

117

females also become sensitive to differences in the relative amplitudes of the two spectral peaks. Differences in the relative amplitudes of the two peaks would become progressively less as the female approached the male because excess attenuation (drop in amplitude over and above that expected from spherical spreading) is directly related to frequency. This change in relative amplitude potentially could, along with an increase in absolute SPL, be used by females to estimate the distance to the male (Gerhardt, 1976). These behavioral results correlate well with our present knowledge about the spectral sensitivity of the peripheral auditory system of H. cinerea (for review, see Capranica and Moffat, 1983). First, many of the primary auditory neurons innervating the amphibian papilla, one of the two main inner ear organs of frogs, are most sensitive to frequencies corresponding to the lowfrequency peak in the conspecific advertisement call; a few of these neurons have thresholds as low as 20 dB SPL and many more would respond well to such sounds at 48 dB SPL. Second, the primary auditory neurons innervating the other main organ, the basilar papilla, are most sensitive to frequencies corresponding to the broader, high-frequency peak of the call; the absolute sensitivity (threshold) of the most sensitive of the neurons innervating the basilar papilla is about 55 to 60 dB SPL, which is just about the same level at which females begin to prefer the combination of low- and high-frequency peaks to the low-frequency peak alone (Fig. 6B). Third, studies of evoked potentials and single auditory neurons at the level of the auditory thalamus revealed that neural responses to appropriate combinations of low- and highfrequency energy are much greater than to either component presented in isolation (Mudry and Capranica, 1987; Fuzessery, 1988). This response property obviously correlates with the general preference for signals with a bimodal spectrum, although there is some question about the role of the thalamus in mediating rapid responses to acoustic signals (Schmidt, 1992; Walkowiak, Endepols, Feng, Schul, and Gerhardt, unpubl. data). 3. Midfrequency Inhibition The previous section deals with the relevance of the two spectral peaks in the advertisement call, but another feature of a bimodal spectrum is the relative deficiency of acoustic energy between the two spectral peaks. In tests of H. cinerea, females preferred a standard synthetic call of 900 + 2700 + 3000 Hz to an otherwise similar synthetic call to which two additional components of 1800 and 2100 Hz were added (Gerhardt, 1974c). In natural advertisement calls, components with frequencies in this intermediate range normally have relative amplitudes at least 10 to 20 dB less than those of components making up the two peaks. Further tests showed that the addition of a single component of 1800 Hz also reduced the relative attractiveness of a signal, even when the overall SPL of this stimulus was increased slightly

118

H. CARL GERHARDT

so that its attractive components had the same amplitude as those in the standard call (Gerhardt, unpubl. data). Additions of single components of 1500 or 2100 Hz had no influence on female preferences. The sensory basis of this midfrequency inhibition is unknown, but one of its expected biological consequences is a reduction in the relative attractiveness of the calls of H. gratiosa, in which the frequency of the high-frequency peak is usually close to 1800 Hz. Midfrequency inhibition appears to be absent in H. gratiosa, even though the second harmonic of the advertisement call is suppressed relative to the fundamental and third or fourth harmonics (Fig. 2A). Females did not prefer a standard synthetic call with a bimodal spectrum to an alternative in which the second harmonic had the same amplitude as the low- and high-frequency peaks (Gerhardt, 1981a). Indeed, females did not even prefer a signal of 500 + 2000 Hz to an alternative of 500 + 1000 Hz, which does not have energy in the high-frequency part of the spectrum emphasized in natural calls. 4. Preferences Based on Variation in Frequencies of Spectral Peaks Considerable between-male variation exists in the frequencies of the lowand high-frequency peaks in the advertisement calls of H. cinerea. In populations sympatric with H. gratiosa, for example, the ranges of variation were 730–1340 and 2190–3620 Hz, respectively (Oldham and Gerhardt, 1975; Gerhardt, 1974c). In two-speaker tests at 75 dB, females from eastern Georgia (United States) prefer synthetic calls with frequencies that are close to means (about 900 and 3000 Hz, respectively) in conspecific calls in the same local population. Under these conditions, females would be expected to discriminate against the nearly 20% of males in the population whose call frequencies fall well above or well below the mean (Gerhardt, 1974c, 1987). The strength of frequency preferences varies, however, with the absolute SPL to which alternatives are equalized (Fig. 7) and with reductions in the SPL of the standard synthetic call (Gerhardt, 1987). At all playback levels (65, 75, and 85 dB), females preferred the mean low-frequency peak of 900 Hz to an alternative of 1100 Hz, but they preferred the 900-Hz stimulus to an alternative of 700 Hz only at 65 and 75 dB (Fig. 7A). With respect to variation in the high-frequency peak, females discriminated smaller departures from the mean frequency at high SPLs than they did at low SPLs (Fig. 7B). These last results might reflect the lower absolute sensitivity of primary neurons tuned to the high-frequency region of the spectrum; perhaps some minimum proportion of these neurons has to be excited in order to mediate discrimination of relatively small differences in frequency. Frequency preferences based on variation in the low-frequency peak are less precise when females have to choose among four alternatives, each

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

119

FIG. 7. Preference functions in green treefrogs, Hyla cinerea, at different playback levels. Points connected by lines show the proportions of females responding to each of two alternatives. (A) Responses to alternatives differing only in the frequency of the low-frequency spectral peak. (B) Responses to alternatives differing only in the frequency of the high-frequency peak. *, Statistically significant preferences ( p < 0.05, two-tailed binomial test).

played back from a separate speaker (Gerhardt, 1987). Whereas females always chose the mean frequency of 900 Hz to an alternative of 600 Hz in two-speaker tests, a substantial proportion of females chose the 600-Hz stimulus in four-speaker trials. In these experiments, the timing relationships of the four stimuli were fixed to avoid any temporal overlap, and the SPL of the four sounds was equalized at the female’s release point. We might thus expect even less selectivity in nature, where both the timing relationships (Section IIC) and the SPLs of neighboring males vary. Females would experience even greater differences in SPL because they are unlikely to assess calls and begin phonotaxis from a point equidistant from the different males. Frequency preferences in H. gratiosa are similar to those in H. cinerea in that females, on average, prefer calls with low-frequency peaks of about average frequency to alternatives with lower or higher values (Gerhardt, 1981b). More extensive testing of individual females, however, shows considerable

120

H. CARL GERHARDT

FIG. 8. Examples of preference functions of individual females of Hyla gratiosa based on variations in (A) frequency of the low-frequency peak, and (B) call rate. Solid lines connect the frequencies or call rates of the pairs of stimuli presented to females and indicate the number of choices (0–4) the female made to each stimulus of the pair. Shaded histograms show the distribution of the two call properties in the advertisement calls of 31 males (frequency) or 29 males (call rate).

individual variation. When females did not prefer frequencies near the mean in the population, they usually preferred higher-than-average signals (e.g., the bottom-most histogram of Fig. 8A; Murphy and Gerhardt, 2000). Females were not very selective with respect to variation in the high-frequency band (see Section II.D.2; Gerhardt, 1981a). 5. Temperature Effects on Frequency Preferences Unlike the temporal properties of pulsed signals (Section III.A.1), the spectral properties of anuran calls change relatively little with temperature. Thus, it was surprising to find that in H. cinerea, significant shifts in frequency

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

121

preferences occurred when females were tested at different temperatures. More specifically, females reversibly changed their preference for the frequency of the low-frequency peak: at normal breeding temperatures of about 24–27◦ C, a 900-Hz call was preferred to alternatives of 500 and 600 Hz; at temperatures of 18–20◦ C, females preferred the alternatives of lower frequency (Gerhardt and Mudry, 1980). These results almost certainly reflect the temperature-dependence of the tuning of auditory receptors (hair-cells) in anurans (Lewis and Narins, 1999). The consequence of the temperature-dependence of frequency preference is that the match between spectral properties and auditory tuning at normal temperatures becomes much poorer at low temperatures. In H. cinerea, for example, the frequency of the low-frequency peak in the calls of individual males that were recorded at both 18 and 27◦ C differed by less than 100 Hz. Furthermore, because males of H. cinerea and H. gratiosa form choruses at temperatures as low as 18◦ C, and because males of the latter species produce calls with frequencies as high as 550 Hz, the temperature effect on preferences could increase the chances that a female of H. cinerea would approach a male of H. gratiosa. Limited fieldwork suggests, however, that few females arrive for mating at these temperatures, which mostly occur in the early part of the breeding season of H. cinerea (Daniel, pers. comm.). Another, more likely, consequence of the temperature effect on female preferences is that large males, which have relatively lower frequencies than average-sized and small males (Gerhardt et al., 1987), might have a mating advantage during the early part of the season when nights are cooler than later in the season. 6. Preferences Based on Fine-Temporal Properties Recall that waveforms of the signals of species in this group are, for the most part, periodic (Section II.A.2). Whereas females of H. cinerea are somewhat selective for synthetic calls having a waveform periodicity that is typical of that in conspecific advertisement calls (Gerhardt, 1978c), females of H. gratiosa are unselective (Gerhardt, 1981a). Preferences in H. cinerea do not depend on differences in the fine details of the waveform but rather on whether or not the typical periodicity of about 300 Hz is reflected in the overall amplitude-time waveform of the stimulus, as in natural advertisement calls (Fig. 1B). This conclusion follows from studies showing that a signal of 900 + 3000 Hz, which has a 300-Hz periodicity (Fig. 9B), was much less attractive than a stimulus of 900 + 2700 + 3000 Hz, which also has a 300-Hz periodicity (Fig. 9C). The difference-frequency between the two high-frequency components of the latter stimulus gives rise to beats, which make the 300-Hz periodicity of the waveform obvious in the stimulus envelope. If this hypothesis is correct, then even large differences in waveform periodicity should fail to elicit preferences for conspecific values if the

122

H. CARL GERHARDT

FIG. 9. Synthetic signals used to test for fine-temporal preferences in Hyla cinerea. (A) Oscillogram (left) shows six cycles of the repeating waveform of a synthetic call with a periodicity of 900 Hz (one period is indicated by the horizontal line); power spectrum (right) shows the two components of 900 and 2700 Hz. (B) Oscillogram (left) shows two cycles of the repeating waveform of a synthetic call with a periodicity of 300 Hz (one period is indicated by the horizontal line); power spectrum (right) shows the two components of 900 and 3000 Hz. (C) Oscillogram (left) shows two cycles of the repeating waveform of another synthetic call with a periodicity of 300 Hz; power spectrum (right) shows the three components of 900, 2700, and 3000 Hz. Beats arising from the summation of the two high-frequency components make the 300-Hz periodicity evident in the overall amplitude-time envelope of the waveform, whereas periodicity information in the signals shown in (A) and (B) has to be extracted from the fine details of the waveform. Females did not show a preference between the calls shown in (A) and (B) and strongly preferred the call in (C) to that in (B). See the text for further explanation.

periodicity is not evident in the envelope. In fact, females do not show a preference between such alternatives (Figs. 9A and 9B) with periodicities of 300 Hz and 900 Hz, respectively (Gerhardt, 1978c; see also Gerhardt et al., 1990). Females of H. cinerea discriminated strongly against signals with low rates of amplitude modulation (pulsing) that is similar to the pulsing (at about 50 Hz) of aggressive calls (Gerhardt, 1978c). Again, the preference depended on the expression of the time pattern in the amplitude-time envelope, and occurred when the depth of modulation was about 50% or greater (Fig. 4C). I consider this result to exemplify “correct rejection” in terms of signal detection theory (e.g., Wiley, 1994): the detection of modulation reduces the attractiveness of the signal relative to that of the unmodulated signal.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

123

Because, in periodic signals, waveform periodicity is represented in the frequency domain as the frequency interval between components, the frogs could be basing their preferences on a spectral rather than a temporal analysis. For example, if tones are modulated at 50 Hz, then additional frequency components (sidebands) arise with frequencies 50 Hz above and below that of the tone. Because the long-term spectra of amplitude modulated noise is the same as that of unmodulated noise, however, any preference between such signals must be based on the detection of the modulation (pulsing) in the time domain. The preference of females of H. cinerea for a synthetic call consisting of unmodulated noise to the same noise modulated at 50 Hz, supports the hypothesis that the frogs use time-domain processing to resolve differences in the temporal properties of these two kinds of signals (Gerhardt, 1978c). Females of H. cinerea also showed preferences based on the number of cycles of 50-Hz pulsing. They preferred signals with one pulse to signals with two pulses, signals with two pulses to signals with three pulses, and so forth (Gerhardt, 1978b). These results indicate that this species has the potential to use a graded signaling system, even though males rarely produce calls that are acoustically intermediate. Another striking difference between H. cinerea and H. gratiosa concerns the relevance of the amplitude envelope of the initial part of the advertisement call. Recall that, in both species, the call begins with one to several cycles of amplitude modulation at a rate of about 100–150 Hz (Fig. 1A,B). Females of H. cinera did not prefer a synthetic call having such a pulsatile beginning to a completely unmodulated call nor did they prefer a typical natural call played normally (forward) to the same signal played backward so that the pulsatile part occurred at the end rather than at the beginning of the call. By contrast, females of H. gratiosa preferred a synthetic call with a pulsatile beginning to an unmodulated call and discriminated against natural calls played backward (Gerhardt, 1981a). 7. Preferences Based on Gross-Temporal Properties The two gross-temporal properties of the calls of H. cinerea and H. gratiosa—call duration and call rate—differ considerably in within-male variability. For example, in a sample of calls from 20 males of H. cinerea, the mean coefficient of variation for call duration was 3.7% and for call rate 12.3% (Gerhardt, 1991). These values are representative of two ends of a continuum of variability that I have termed “static” and “dynamic” (Gerhardt, 1991). Dominant frequency is another example of a static property. In these and other treefrogs, static properties tend to mediate stabilizing or only weakly directional selection (see Section II.D.5), whereas dynamic properties tend to mediate highly directional preferences (Gerhardt, 1991).

124

H. CARL GERHARDT

Preferences based on these two gross-temporal properties also exemplify this trend. Females of H. cinerea, for example, preferred a standard call with a duration of 160 ms to an alternative of 120 ms (Gerhardt, 1987). Females did not pick the standard call over a longer alternative of 480 ms, a duration that is about 180 ms beyond the upper end of the range of variation, but neither did they prefer this long call. Thus, at most, the preferences based on this static property are weakly stabilizing. By contrast, females of H. cinerea preferred calls played back at a rate of 86 calls/min (standard) to calls played back at 75 calls/min (Gerhardt, 1987). Twenty of 21 females also chose calls with a rate of 150 calls/min (approaching the maximum observed in nature) to an alternative with a rate of 75 calls/min. Calls with the latter rate were, however, preferred to alternatives with a rate of 300 calls/min, which did not even elicit phonotactic responses in nochoice tests. Thus, although female preferences are highly directional, they are not open-ended. More extensive experiments, in which call-rate differences were varied in steps of 5 calls/min, were conducted with females of H. gratiosa (Murphy and Gerhardt, 2000). In most individual females, preferences were highly directional (Fig. 8B), and the mean difference in call rate that resulted in a preference for the higher-rate alternative was about 30%. F. PREFERENCES BASED ON RELATIVE TIMING OF CALLS In all of the experiments discussed so far, the timing relationships of alternative signals (even in 4-speaker tests) were fixed so that equal intervals of silence occurred between presentations of each stimulus. Recall, however, that timing between neighboring males is variable and there are often leader–follower relationships (see Section II.C). In experiments with H. cinerea, females preferred a synthetic call that led a nearly identical call (digital copy) in time by 40, 160, or 300 ms (Klump and Gerhardt, 1992). The preference was especially strong with the 40-ms delay, which resulted in a 78% overlap of the lagging call by the leading call. This preference was abolished, however, by reducing the SPL of the leading call by 6 dB. These results are consistent with those for some other anurans and acoustic insects (Greenfield, 1994) and might represent an example of the precedence effect (see full discussion in Section IV.B). G. PREFERENCES BASED ON DIFFERENCES IN SOUND PRESSURE LEVEL Weber’s law predicts that the minimum detectable difference in intensity between two sounds should be a constant proportion of their absolute level. In fact, preferences based on differences in SPL depend on both absolute

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

125

and relative values in H. cinerea. Females preferred the more intense of two synthetic calls that differed by 3 dB when the absolute SPLs of the stimuli were low to moderate (60 vs 63 dB and 70 vs 73 dB), but a difference of 6 dB was required to elicit a preference at higher absolute levels (80 vs 86 dB) (Gerhardt, 1987). Similar results have been reported in studies of acoustic insects (Forrest, 1994; Romer ¨ et al., 1998). This result again emphasizes the fact that preferences are likely to depend on both absolute and relative differences in amplitude. Predicting female preferences in nature thus requires observations of the distances (and hence SPLs) at which females assess potential mates. H. SUMMARY Members of the Hyla cinerea species group have repertoires of two to four different signals. Male advertisement calls are relatively short and typically produced rapidly and in an alternating fashion between neighboring males. Males of one species, H. cinerea, frequently produce distinctly pulsed aggressive signals, which are acoustically similar to release calls produced by both sexes in all three species. Advertisement calls typically have two species-specific spectral peaks. Some females of all three species respond to the calls of the other species in single-speaker tests, but prefer conspecific calls in choice tests. Females of H. andersonii were the least selective of the three species. Females of H. gratiosa also chose conspecific calls to those of interspecific hybrids (H. cinerea × H. gratiosa) as did most females of H. cinerea. Experiments using synthetic calls indicate that spectral differences alone can promote selective phonotaxis, and, for H. cinerea and H. gratiosa, females prefer signals with bimodal spectra to signals with a single spectral peak. In H. cinerea, selectivity based on spectral structure (relative amplitude and frequencies of the two spectral peaks) is affected by the absolute intensity of signals, and the behavioral results correlate well with different frequency sensitivity and thresholds of the two auditory organs of anurans: the amphibian and basilar papillae. Frequency preferences but not call frequencies are reversibly dependent on temperature. At normal breeding temperatures, preferences of H. cinerea and H. gratiosa based on differences in the frequencies of the spectral peak represent stabilizing or weakly directional selection. Preferences based on finescale temporal properties depend on the expression of temporal differences in the overall amplitude-time envelope of the signals. The advertisement call of H. cinerea, for example, becomes less attractive when it is pulsed (as in aggressive calls) to a sufficient degree (50% modulation depth). This preference might represent “correct rejection” of signals associated with the

126

H. CARL GERHARDT

possibility of an attack. In H. cinerea and H. gratiosa, females prefer high to low values of call rate, a dynamic property that often varies significantly within a bout of calling. Females of H. cinerea species show preferences for leading over lagging signals, especially when the beginning of the lagging signal is overlapped. Although closely related, significant differences exist in the acoustic criteia used by females. Unlike females of H. gratiosa, females of H. cinerea discriminate against calls with added midfrequency components, and they are selective for variation in the frequency of both peaks typical of advertisement calls. Females of H. gratiosa are selective mainly for variation in the low-frequency peak, and, unlike females of H. cinerea, they discriminate against calls that lack the pulsatile beginning typically found in conspecific calls of both kinds.

III. Hyla versicolor GROUP As defined by Blair (1959), the members of the H. versicolor group are H. arenicolor (canyon treefrog), H. avivoca (bird-voiced treefrog), H. femoralis (pinewoods treefrog), H. chrysoscelis (diploid gray treefrog), and H. versicolor (tetraploid gray treefrog). Although the interpretation of hybridization data relative to phylogeny reconstruction is problematic, this grouping of the diploid species is completely supported by a phenogram based on genetic compatibility assayed by laboratory crosses (Ralin, 1970). Phylogenetic analyses consistently group H. avivoca with H. chrysoscelis and H. versicolor (hereafter these three species will be labeled as the “core” group), but H. arenicolor and H. femoralis fall out separately (Cocroft, 1994; da Silva, 1997). This topology is again supported solely by allozyme data. Recent studies using mtDNA sequences suggest, as did the more limited studies of Pierce (1975), that H. arenicolor might consist of more than one taxon (Barber 1999a,b). The clades identified by these two studies are not, however, geographically congruent. One of the three major clades identified by Barber (1999a) is much more closely related to H. chrysoscelis and H. versicolor than are the other two clades. A tentative phylogenetic tree that represents a summary of the available evidence is shown in Fig. 10. Although closely related to the core group which is supported by all available evidence, H. femoralis is probably a good candidate for outgroup comparisons. This conclusion is also supported to some extent by bioacoustic considerations discussed below. The calls of Hyla “eximia,” which may consist of two or more taxa (Fouquette, pers. comm.), are not illustrated. Their structure is completely unlike any of the other species, and no data are available from playback experiments.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

127

FIG. 10. Phylogenetic relationships and advertisement calls of some species in or closely related to the Hyla versicolor species group. Left: Tentative phylogenetic tree based on the strict consensus tree of Da Silva (1997), who used a “total evidence” approach, plus new relationships suggested by analyses of mtDNA (cytochrome b) sequences in Hyla arenicolor, H. “eximia,” H. chrysoscelis, and H. versicolor (Ptacek et al., 1994; Barber, 1999a). “H. eximia” may consist of two or more different taxa, whose calls are completely different in structure than those of the other species. All lineages of H. versicolor are biparental tetraploids that arose from one of the two weakly defined lineages of H. chrysoscelis or an extinct (or as yet undiscovered) third lineage. Center: Oscillograms showing about 4 s of calling. Time markers: 1 s. Notice that Hyla femoralis pulses are produced nearly continuously at a relatively high and irregular rate. The other species organize pulses into discrete calls that are about the same duration in H. arenicolor, H. chrysoscelis, and H. versicolor; calls of H. avivoca are longer and produced at slower rates. Right: Oscillograms with an expanded time base to show the amplitude-time structure of single pulses from calls of H. femoralis and H. avivoca, two pulses from calls of H. arenicolor and H. versicolor, and three pulses from a call of H. chrysoscelis. Males were recorded at temperatures between about 20–23◦ C. Time markers: 25 ms; y-axis tick marks represent arbitrary amplitude units.

H. arenicolor is found in the southwestern United States and northwestern Mexico, where it breeds in small streams in rocky canyons; other species are found in the eastern one-third of the United States (Conant et al., 1998). In the southeastern United States, the distributions of H. avivoca, H. chrysoscelis, and H. femoralis overlap extensively. H. avivoca prefers

128

H. CARL GERHARDT

extensive river swamps and lakes, whereas H. femoralis breeds in temporary pools and ditches in otherwise xeric habitats. Mixed species choruses are rare (Gerhardt, 1970). By contrast, H. chrysoscelis breeds in a wide range of habitats, and frequently forms mixed choruses with H. avivoca and H. femoralis (Gerhardt, 1970). Although H. avivoca tends to call more often from arboreal positions, the calling sites used by all of these species overlap considerably (Gerhardt, unpubl. obs.). The range of H. versicolor overlaps with that of H. chrysoscelis in several parts of the mid-Atlantic region in the eastern part of the range and in scattered areas throughout the midwestern United States. In Missouri, there was some tendency for calling males of H. chrysoscelis to occupy lower calling positions than calling males of H. versicolor in ponds where both species occurred (Ptacek, 1992). The two gray treefrogs, nominally H. chrysoscelis and H. versicolor, pose special challenges and provide exciting opportunities for comparative studies. Although indistinguishable by external morphology, differences in call types were noted more than 60 years ago (Noble and Hassler, 1936). Subsequently, frogs of a slow-trilling type were found to be biparental tetraploids and designated as H. versicolor; “fast trillers” were found to be diploids, as are all other North American treefrogs, and were designated as H. chrysoscelis (for review, see Bogart and Wasserman, 1972). Crosses between female H. versicolor and male H. chrysoscelis can produce viable, though infertile, triploid offspring; reciprocal crosses are unsuccessful (Mable and Bogart, 1991). Studies of variation in allozymes, chromosome polymorphisms, and mitochondrial DNA sequences indicate that there are at least two, weakly differentiated lineages of H. chrysoscelis, from which three or more lineages of tetraploids (H. versicolor) have arisen independently by autopolyploidy (Ptacek et al., 1994, and references therein). The two lineages of H. chrysoscelis, as defined by allozyme markers and mtDNA sequences, come into contact over a fairly broad area in the central United States and appear to interbreed freely. The distributions of the three tetraploid lineages are largely disjunct, except for a small area of contact (southwestern and northwestern lineages) in southwestern Missouri and perhaps in an area in the north-central part of the range (northwestern and eastern lineages) (Gerhardt and Ptacek, unpubl. data). The existence of multiple tetraploid lineages provides opportunities to conduct replicated studies of evolutionary changes in communication systems after speciation from the parental form (H. chrysoscelis). I concentrate on research with the gray treefrogs because relatively little information is available about acoustic communication in the other members of the group. Data from the other species in or closely related to the group are needed for comparative studies that will almost certainly provide insights concerning broadscale patterns of evolution.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

129

FIG. 11. Sonograms (left) and diagrams of power spectra (right) of the advertisement calls of (A) Hyla avivoca (part of one call), (B) H. chrysoscelis, (C) H. versicolor, (D) H. arenicolor, and (E) H. femoralis.

A. ADVERTISEMENT CALLS The temporal properties of advertisement calls both help to define the group and to distinguish among the species (Fig. 10 and 11) (Blair, 1959; Gerhardt, 1974a, 1982; Pierce and Ralin, 1972). The advertisement calls of all species consist of a series of pulses (Fig. 10). In all except H. femoralis,

130

H. CARL GERHARDT

the pulses are organized into discrete trains, which are usually designated as calls (middle column of Fig. 10). The pulses of H. femoralis are produced in a highly irregular fashion for minutes at a time, hence, defining the beginning and the end of a call that is analogous to those of the other species is impossible. The pulses of H. femoralis and H. arenicolor differ from those of the other species in the group by having distinctive subpulses, which have a stereotyped, temperature-dependent interval (Gerhardt, 1970, 1974a, 1998; Gerhardt and Marshall, unpubl. data). Mean SPLs of the calls of H. avivoca, H. chrysoscelis, H. femoralis, and H. versicolor at 50 cm from calling males were 85.5 dB (98.5 dB peak), 93.5 dB (105 dB peak), 88.5 dB (105.3 dB peak), and 96 dB (108.5 dB peak), respectively (Gerhardt, 1975). 1. Spectral Properties I consider three spectral properties of the advertisement calls: (1) the number of spectral peaks; (2) the frequency or frequencies of the spectral peak(s); and (3) the extent, if any, of frequency modulation (change) within pulses (Figs. 10 and 11). The calls of H. avivoca have a single spectral peak, which ranged in frequency from about 2000–2600 Hz (mean = 2340 Hz) in a sample of males recorded in eastern Georgia and South Carolina (Gerhardt, 1970). The frequency of each pulse increases slightly from beginning to end. Males of H. femoralis produce noisy calls with a bimodal spectrum, but the low-frequency peak contains most of the acoustic energy. The centerfrequency in the low-frequency band of 49 males from eastern Georgia and South Carolina ranged from 1950–2350 Hz (mean = 2150 Hz); the (usually) much weaker high-frequency band ranged from 3800–4200 Hz (Gerhardt, 1970). The intrapulse pulsing is represented in the frequency domain by a series of components (separated by about 120 Hz at about 23◦ C) within each band; these components change little, if at all, in frequency from the beginning to the end of a pulse. Males of H. arenicolor produce calls with a distinctly bimodal spectrum: a low-frequency peak occurs at about 500 Hz, and a high-frequency peak is centered at about 2200–2500 Hz (Blair, 1959; Pierce and Ralin, 1972). As shown in the sonogram of Fig. 11, sidebands occur at about 230 Hz on either side of the low-frequency peak; as in pulses of H. femoralis, these components reflect the pulse rate of the subpulses within each pulse and are fairly constant in frequency. In the gray treefrogs (H. chrysoscelis and H. versicolor), the high-frequency peak of the calls (centered between about 2000–2800 Hz) nearly always has more acoustic energy (mean of 8–10 dB) than the low-frequency peak (centered between about 1000–1500 Hz) (Gerhardt, 1982, and unpubl. data). The mean values of these emphasized frequencies vary to a limited extent geographically; in

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

131

Missouri, the means for the low-frequency peak are about 1100 and 1200 Hz, for H. versicolor and H. chrysoscelis, respectively; the corresponding means for the high-frequency peak are about 2200 and 2400 Hz, respectively (Gerhardt, unpubl. data). The calls of several putative natural hybrids between H. avivoca and H. chrysoscelis and between H. femoralis and H. chrysoscelis have been analyzed (Gerhardt, 1974a), and descriptions of the calls are also available for several laboratory reared individuals that were the products of crosses between the latter two species (Doherty and Gerhardt, 1984). All of the hybrid individuals produced advertisement calls with two distinct spectral peaks that were roughly intermediate in frequency in comparison with the analogous peaks in the calls of the parental species (Fig. 12). 2. Fine-Temporal Properties The advertisement calls of species in or closely related to this group differ in pulse duration, pulse shape, pulse rate, or some combination of these properties. The pulses in the calls of 32 males of H. avivoca from eastern Georgia and South Carolina ranged from 50–82 ms in duration (mean 62 ms); the pulses rise gradually in amplitude from beginning to end (Fig. 10, right column) (Gerhardt, 1970). Pulse duration and the interval between pulses both vary somewhat within a call. The mean pulse rate (reciprocal of the interpulse interval) was about 6.6 pulses/s, and the mean maximum rate was 7.7 pulses/s. The pulses in the calls of 49 males of H. femoralis from eastern Georgia and South Carolina ranged from 38–62 ms in duration and were made up of three to six subpulses (Fig. 10, right column). The interval between the longer pulses was extremely variable: mean pulse rate was 7.9 pulses/s and mean maximum pulse rate was 10.9 pulses/s (Gerhardt, 1970). Pulse duration and pulse rates are correlated with temperature (for H. avivoca and H. femoralis, pulse duration: r = −0.87 and −0.88; and average pulse rate: r = 0.88 and 0.71, respectively). No published data concerning pulse duration are available for H. arenicolor; the range of values measured from recordings of five males from Arizona was 26–52 ms; pulses had 3 or 4 subpulses with a period of about 4.5–6.3 ms (temperature range: 17– 22◦ C; Gerhardt and Marshall, unpubl data). The mean pulse rate for six populations from the southwestern United States and Mexico ranged from 14.1 to 23.6 pulses/s at 20◦ C (Pierce, 1975). At a common temperature, the pulses of H. chrysoscelis and H. versicolor differ in duration by about a factor of two (Fig. 10, right column). The amplitude-time envelopes of the pulses are species-typical at all temperatures: the amplitude of pulses of H. versicolor rises slowly in a nearly linear fashion, reaching full amplitude at about 70–80% of the pulse duration; pulses of H. chrysoscelis have an approximately logarithmic rise and

132

H. CARL GERHARDT

FIG. 12. Sonograms of the advertisement calls of (A) Hyla avivoca (part of one call), (B) a putative hybrid (H. avivoca × H. chrysoscelis), (C) H. chrysoscelis, (D) a putative hybrid (H. chrysoscelis × H. femoralis), and (E) H. femoralis (about 1 s from a calling bout).

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

133

usually reach full amplitude at about 45–50% of the pulse duration (Gerhardt and Doherty, 1988). In both species, the pulse duty cycle (ratio of pulse duration to pulse period) tends to be constant (0.45–0.59) over the entire range of breeding temperatures (Gerhardt and Doherty, 1988; Gerhardt, 1991). Pulse duration and period (pulse rate) are typically uniform within each call. Considerable variation in mean pulse rate exists within the range of distribution of H. chrysoscelis: mean values at 20◦ C are as low as 35 pulses/s in the southeastern United States, and as high as 50 pulses/s in the central Midwest (Gerhardt, 1999) (see Section V.A). Less variation in pulse rate occurs in H. versicolor: mean values (at 20◦ C) for eastern populations (Maryland, Virginia, West Virginia, and New Jersey) are about 19 pulses/s; values for populations in the southwestern United States (Texas, Oklahoma) are about 24 pulses/s (Gerhardt, Ptacek, and Keller, unpubl. data). Mean values for populations in the central Midwest are about 20 pulses/s. These three regions represent the known centers of distributions of the three lineages of H. versicolor as defined by analyses of sequences of the cytochrome b mitochondrial gene (eastern, southwestern, and northwestern, respectively) (Ptacek et al., 1994). Pulse rate in field recorded and laboratory reared hybrids between H. chrysoscelis and H. versicolor is roughly intermediate in comparison with that in the calls of the parental species (Gerhardt et al., 1994a). Pulse shape and duration in the calls of putative hybrids between H. avivoca and H. chrysoscelis are also intermediate in comparison to their values in the calls of the parental species (Fig. 12; Gerhardt, 1974a). In the calls of putative hybrids between H. femoralis and H. chrysoscelis, however, the pulses were about the same duration as those in H. chrysoscelis (Fig. 12). Some laboratory raised individuals of this combination produced pulses with crude, intrapulse modulation reminiscent of the pulses produced by H. femoralis (Doherty and Gerhardt, 1984). No difference in the call structure of reciprocal crosses was detected. 3. Gross-Temporal Properties Although the duration of the advertisement calls of H. arenicolor, H. chrysoscelis, and H. versicolor is similar (range: 0.38–1.5 s), the calls of H. avivoca are significantly longer (range: 1.5–5.7 s; mean = 2.85 s) (Gerhardt, 1970, 1994a; Pierce, 1975; Ralin, 1977). Call duration in H. chrysoscelis also varies geographically (Gerhardt, 1974d; Ralin, 1977); more recent and extensive analyses show weak clinal variation in mean duration, which increases from east to west and from south to north (Gerhardt, Keller, Ptacek, and Forester, unpubl. data). Because males of H. femoralis do

134

H. CARL GERHARDT

not organize pulses into discrete trains, these signals are better analyzed with interval histograms (Gerhardt, 1998). The histograms show that while the intervals between the “long” pulses are highly irregular, the intervals (about 8 ms at 23◦ C) between the “subpulses” are highly stereotyped. Call rate is highly variable within and between males in this species group, but formal analyses are available only for H. versicolor (Gerhardt, 1991; Runkle et al., 1994; Gerhardt et al., 1996). Call rate is slowest in H. avivoca, intermediate in the gray treefrogs, and fastest in H. arenicolor (Pierce, 1975; Gerhardt and Marshall, unpubl. data). Call duration and call rate are usually inversely correlated in H. versicolor and H. chrysoscelis: males tend to produce long calls with slow rates in dense choruses and short calls at high rates in sparse choruses or when calling in relative isolation (Wells and Taigen, 1986; Gerhardt and Marshall, unpubl. data). In a sample of H. versicolor, call duration was positively correlated with the SPL of the calls of the nearest neighbor at the position of the male being recorded (Gerhardt, 1991). This relationship could be used to correct statistically for the effects of chorus density in comparisons of night-to-night or geographical variation in call duration. B. OTHER CALLS IN THE REPERTOIRE The aggressive calls of H. avivoca, H. arenicolor, and the gray treefrogs are distinctly different from their advertisement calls (Fig. 13). In H. avivoca, the aggressive call is distinctly pulsed at 73–80 pulses/s (Gerhardt and Marshall, unpubl. data). Aggressive-call structure is more variable in the other species, and pulsing is so obscure in some of the signals that pulse rate would be difficult or impossible to estimate. In calls from which estimates could be made, Pierce and Ralin (1972) reported values of 60–240 pulses/s in H. arenicolor and 50–85 pulses/s in the gray treefrogs. The dominant frequencies of these signals are similar to those of advertisement calls in H. avivoca and the gray treefrogs (Figs. 11 and 13). Unlike those in advertisement calls, however, these components rise in frequency from the beginning to the end of the calls, and the higher harmonics tend to have more energy than those in advertisement calls. Analysis of a limited sample of H. arenicolor aggressive calls indicates that the low-frequency peak is similar to that in the advertisement call, but the high-frequency peak is somewhat lower in frequency and not as strongly emphasized (Fig. 13; Gerhardt and Marshall, unpubl. data). Although sometimes produced singly, aggressive calls are usually produced in a series of 2–8 calls, and occasionally males of H. avivoca and gray treefrogs add one or more aggressive calls after an advertisement call even if they are not interacting vocally with a neighbor (Gerhardt, pers. obs.). To

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

135

FIG. 13. Aggressive calls of members of the Hyla versicolor group. Sonograms (top panels) and oscillograms (lower panels) show representative but not necessarily typical calls because of the large variation in these signals. (A) H. arenicolor, (B) H. avivoca, (C) H. chrysoscelis, and (D) H. versicolor. The calls shown were produced as pairs, except for the H. avivoca, which produced a long series of these signals.

my knowledge, aggressive calls have not been reported in H. femoralis, and this is another bioacoustic reason for suggesting that this species is more distantly related to the core group of species than is H. arenicolor. Aggressive calls in each species of gray treefrog can be elicited by playbacks of the advertisement calls of the other species, and advertisement calls of gray treefrogs also elicit aggressive calls from males of H. arenicolor and H. avivoca (Gerhardt and Marshall, unpubl. data). Thus, these heterospecific calls are not only audible but are treated as signals rather than ignored as background noise. The effects of playbacks of aggressive calls on male behavior and their possible attractiveness to females have not yet been studied. The aggressive calls of H. chrysoscelis and H. versicolor sound very similar, and the acoustic analyses of small samples of these signals have not yet revealed any consistent differences (Pierce and Ralin, 1972). The most parsimonious explanation for this similarity is not convergent evolution, but rather the conservation of call structure in H. versicolor after its polyploid speciation from H. chrysoscelis (Ptacek et al., 1994). The possibility exists, however, that more smapling will show that aggressive-call structure

136

H. CARL GERHARDT

differs among the different lineages (species) of H. versicolor. As described by Pierce and Ralin (1972), the release calls of H. arenicolor and the gray treefrogs are similar to their respective aggressive calls, but are even less stereotyped in their temporal properties and also differ in minor ways in their frequency profiles. When approached or touched by a female, especially if the male does not succeed in clasping her, males of H. chrysoscelis and H. versicolor sometimes produce “long” advertisement calls, which have been termed courtship calls (Fellers, 1979; Gerhardt, unpubl. data). These signals are two to three times longer than the normal advertisement calls, and the male normally produces only one or two such signals. The duration and rate of advertisement call production remains somewhat above average for about 2–3 min after production of long calls if the male does not achieve amplexus with the female (Gerhardt, unpubl. data). C. SIGNALING INTERACTIONS Males of H. versicolor, H. chrysoscelis, and H. avivoca usually alternate calls with neighbors. Computer-based monitoring of the calls of up to 8 males of H. versicolor at a time revealed that males also selectively interacted with their neighbor (Schwartz, 2001; Schwartz, Buchanan, and Gerhardt, in preparation). Whereas the other anurans and insects that have been studied selectively avoided overlap with their nearest or loudest neighbors (Brush and Narins, 1989; Schwartz, 1993; Minckley et al., 1995), the opposite pattern was found in H. versicolor. Nevertheless, the overlap time was only about 11% of the total calling time and, as chorus size was reduced, males were more likely to avoid overlap with their neighbor. This last result suggests that perhaps males were not deliberately overlapping the neighbor’s calls but that they have difficulty in detecting a neighbor’s calls against the background noise of a large chorus. Recall that females of H. cinerea also have difficulty detecting the calls of individual males against a loud chorus of conspecific calls (Gerhardt and Klump, 1988a; Section II.D.3). D. PHONOTACTIC RESPONSES TO PRERECORDED CALLS Most of our knowledge about phonotactic selectivity in the gray treefrogs comes from studies of females. Brown and Pierce (1965) described phonotaxis by a female of H. arenicolor in the field, and Vincent Marshall and I recently observed phonotaxis to playbacks of a prerecorded conspecific call by a female of the clade found in the Grand Canyon (Arizona). I have tested a few females of H. femoralis (Section III.D.2), but I am unaware of any studies of phonotaxis by females of H. avivoca.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

137

1. Phonotaxis to Heterospecific Advertisement Calls Females of H. versicolor sometimes respond to playbacks of the advertisement calls of H. chrysoscelis if they have no other choice: 3 of 8 females moved toward the speaker and one touched it at a playback level of 75 dB SPL; 4 of 7 females moved toward the speaker and one touched it at a playback level of 85 dB SPL (Gerhardt and Doherty, 1988). The phonotactic movements in the vicinity of the speaker were much less persistent than responses to conspecific calls, and even the 2 females that touched the speaker moved away without returning. Some females of H. chrysoscelis have also been observed to show similar, weak phonotactic responses to playbacks of the calls of H. versicolor (Littlejohn et al., 1960; Gerhardt, unpubl. data). The one female of H. arenicolor tested by Marshall and me showed vigorous and repeated phonotactic approaches to playbacks of synthetic calls of H. versicolor. 2. Preferences for Conspecific Advertisement Calls to Heterospecific and Hybrid Calls Littlejohn et al. (1960) first demonstrated selective phonotaxis of gray treefrogs in two-speaker choice tests that were conducted when the existence of two species was just being confirmed. These authors showed that females found in amplexus with a “fast-trilling” male (= H. chrysoscelis) chose speakers emitting prerecorded calls of a male producing fast-trilled calls; females found in amplexus with a “slow-trilling” male (= H. versicolor) preferred slow-trilled calls. As in the H. cinerea group, females preferred conspecific calls to those of putative hybrids. Females of H. chrysoscelis preferred conspecific advertisement calls to those of putative hybrids of two combinations: H. avivoca × H. chrysoscelis and H. femoralis × H. chrysoscelis (Gerhardt, 1970, 1974a). In the only published tests of H. femoralis that I am aware of, 10 females chose the conspecific call in all 21 tests against the calls of a putative hybrid (H. femoralis × H. chrysoscelis) (Gerhardt, 1974a). Laboratory reared females of H. femoralis × H. chrysoscelis (both reciprocal crosses) preferred the calls of hybrids to those of H. chrysoscelis but not to the calls of H. femoralis. The last result might reflect the fact that the calls of H. femoralis are not organized into discrete calls and provide nearly constant, though irregular, acoustic stimulation (Fig. 10). Tests using a pair of synthetic calls, which provided more nearly equal acoustic stimulation, showed that hybrid females preferred signals with pulse rates typical of hybrid males to signals with a much higher pulse rate typical of H. chrysoscelis (Doherty and Gerhardt, 1984). The generally intermediate structure of hybrid calls and the overall preference of hybrid females for hybrid calls fits with

138

H. CARL GERHARDT

the results of comparable studies of acoustic insects (for review see, Ritchie and Phillips, 1998). E. PHONOTACTIC RESPONSES TO SYNTHETIC CALLS 1. Standard Synthetic Calls: Irrelevance of Frequency-Modulated Pulses The first standard synthetic signals I used to test female gray treefrogs were just as attractive as natural exemplars. These artificial signals incorporated the within-pulse frequency modulation of the natural calls. Given choices between these synthetic calls and synthetic calls with constant-frequency pulses, however, females did not show a preference (Gerhardt, 1978a; Gerhardt and Schul, in preparation). Thus, this highly stereotyped feature of the calls appears to be irrelevant for signal selection. All other standard calls used in tests of gray treefrogs lacked frequency-modulated pulses because its incorporation would have added another variable difference (either the rate of change in frequency or the extent of frequency change) to pairs of alternatives in which pulse duration was unequal. 2. Preferences Based on Spectral Patterns As in the H. cinerea species group, the bimodal spectrum is a relevant acoustic property for female gray treefrogs. Both species show phonotactic responses to playbacks of synthetic calls having just the low-frequency peak or just the high-frequency peak at playback levels of 75–85 dB SPL. Behavioral thresholds have not yet been estimated. Females also prefer synthetic calls with both spectral peaks to alternatives with only one of the two peaks (Gerhardt and Doherty, 1988; Gerhardt and Tanner, in preparation). Females preferred a stimulus in which the amplitude of the low-frequency component was as much as 24 dB less than that of the high-frequency component in tests against a stimulus consisting of the high-frequency peak alone. Thus, a spectral component does not necessarily have to be emphasized in the call to have a significant effect on female preferences. Recall that the relative amplitude of the low-frequency peak is typically 8–10 dB less than that of the high-frequency peak (Section III.A.1). One surprising difference in the two species of gray treefrogs concerns the relative attractiveness of the two spectral peaks. Whereas females of H. versicolor prefer the calls with just the high-frequency peak to alternatives with just the low-frequency peak, females of H. chrysoscelis do not show such a preference (Gerhardt and Doherty, 1988; Gerhardt and Schul, in preparation). This is a particularly intriguing result because the preference of the newly derived species has changed while the spectral structure of its advertisement call has been conserved. These behavioral results are consistent with differences in audiograms based on recordings of evoked

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

139

potentials from the midbrain of the two species (Lombard and Straughan, 1974; Hillery, 1984). 3. Preferences Based on Variation in Frequencies of Spectral Peaks Female gray treefrogs of both species show relatively weak preferences based on differences in the frequencies of the two spectral peaks (Gerhardt and Tanner, in preparation). That is, preferences based on differences of less than 400 Hz are usually abolished by reducing the SPL of the preferred signal by just 6 dB. The pattern of preference is weakly stabilizing because frequencies near the mean are preferred to alternatives of lower and higher frequency. Frequency preferences in H. versicolor depend not only on playback level but also on whether pairs of alternatives have one or both spectral peaks. In initial studies, using synthetic calls having only the high-frequency peak, the pattern of preference was found to be weakly directional in that females preferred a call with a frequency of 1900 Hz to the standard call of 2200 Hz (near the population mean; Gerhardt and Doherty, 1988). By contrast, in the experiments using signals with both spectral peaks, females showed no preference for the stimulus of 950 + 1900 Hz to that of 1100 + 2200 Hz (Gerhardt and Tanner, in preparation). Repeating the tests of single-component calls gave the same result as the original experiment, but females failed to prefer the 1900-Hz stimulus when the playback level was reduced from 85 to 75 dB SPL (Gerhardt and Tanner, in preparation). These results suggest that the neural mechanisms underlying frequency preferences in H. versicolor almost certainly involve intensity-dependent interactions between neurons in the ascending auditory pathways that process the outputs of the two hearing organs. Neurophysiological studies indicate that at playback levels of 85 dB SPL, acoustic energy in the high-frequency peak, to which the basilar papilla is tuned, would activate a substantial number of the most sensitive neurons in the low-frequency channel derived from the amphibian papilla (Lombard and Straughan, 1974; Diekamp and Gerhardt, 1995; Schul and Gerhardt, in preparation). Similarly, lowfrequency energy presented at this level would effectively stimulate some of the most sensitive neurons in the high-frequency channel. Preferences that are seemingly based on frequency differences between signals with bimodal spectra are thus likely to be influenced at least in part by differences in the ratio of excitation of the two channels. 4. Preferences Based on Fine-Temporal Properties: Pulse Rise-Time The two species of gray treefrogs differ significantly in their phonotactic selectivity with respect to the shape and rise-time of the pulses that make up the calls of males of the two species (Fig. 10). Females of H. versicolor

140

H. CARL GERHARDT

show strong preferences for synthetic calls in which pulse shape simulates that in conspecific calls to alternatives in which pulse shape is typical of the H. chrysoscelis calls (Gerhardt and Doherty, 1988; Diekamp and Gerhardt, 1995; Gerhardt and Schul, 1999). In tests at 20◦ C, for example, preferences survived a 6-dB reduction in the relative SPL of the preferred (conspecific) shape, and most females still chose the standard stimulus after a further 6-dB reduction (Diekamp and Gerhardt, 1995). Females discriminate differences in pulse shape on the basis of the form of the rise (linear vs logarithmic), on its absolute value, or both. Surprisingly, females of H. chrysoscelis show no preference for pulses having the conspecific shape to alternatives simulating the shape of H. versicolor pulses (Gerhardt and Schul, in preparation). Preferences based on pulse rise-time in H. versicolor are both frequencyand intensity-dependent. Females discriminate between signals that differ by only 5 ms in pulse rise-time, preferring alternatives with slow rise-times to alternatives with fast rise-times (Gerhardt and Schul, 1999). When the frequency of the alternative calls was 1100 Hz, this preference occurred at playback levels of 75 and 85 dB SPL, and a 10-ms difference in rise-time elicited a preference at 65 dB SPL. When the frequency of the alternative calls was 2200 Hz, however, pulse rise-time preferences occurred only at a playback level of 85 dB SPL. This last result suggests that pulse rise-time selectivity depends on stimulating sensitive auditory neurons that are tuned to the low-frequncy peak. As discussed above (Section III.E.2), sensitive neurons showing maximal sensitivity to frequencies representative of the low-frequency peak would also be stimulated by a sound of 2200 Hz at 85 dB SPL. Tuning curves of auditory neurons at the level of the dorsolateral nucleus (first station in the pathway) indicate that few, if any, of these neurons would be excited by such a stimulus at 75 dB SPL (Schul and Gerhardt, in preparation). Comparable results are available concerning the frequency dependence of preferences based on pulse duration and pulse rate (Gerhardt and Schul, in preparation). 5. Preferences Based on Variation in Fine-Temporal Properties: Pulse Rate and Duration As discussed above, the difference in temperature-corrected pulse rate provided the first clue that there might be more than one species of gray treefrog. Experiments with females of both species confirm that relatively small differences in pulse rate are adequate to promote selective phonotaxis to conspecific calls (Fig. 14; Gerhardt and Doherty, 1988; Gerhardt and Schul, in preparation). This statement, however, has to be qualified in several ways. First, females must be tested with alternative stimuli in which the pulse rate of the standard call matches the pulse rate produced by conspecific males at the test temperature. This requirement follows from the fact that

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

141

FIG. 14. Summary of tests of female preferences based on pulse rate in Hyla chrysoscelis (top) and (B) Hyla versicolor (bottom). Lines connect points showing the proportions of females choosing each of two alternatives. w, Indicates that the two alternatives were adjusted to the same SPL. Other symbols indicate that the SPL of the standard call (pulse rate near the mean) was reduced relative to that of the alternative by: 6 dB (), 12 dB (), or 18 dB (). In tests of H. chrysoscelis, all pulses simulated the shape typical of that in conspecific calls. In one test of H. versicolor, the pulses of the alternative call with a higher pulse rate (indicated by “Hch”) had pulses simulating the shape typical of H. chrysoscelis. The thin horizontal lines between the two panels show the ranges of variation in pulse rate (no correction for temperature); the thick horizontal lines show the ranges of variation in temperature-corrected pulse rate (20◦ C—the same as the temperature at which females were tested); the vertical lines indicate the mean temperature corrected pulse rates.

142

H. CARL GERHARDT

temperature affects pulse rates and female selectivity for pulse rate in a parallel fashion. This phenomenon has been termed “temperature coupling” (Gerhardt, 1978a), and a neurophysiological correlate is the temperature dependence of the “tuning” of band-pass neurons in midbrain (torus semicircularis) to a particular rate of amplitude modulation (equivalent to pulse rate) (Rose et al., 1985). The behavioral coupling is poorer at the high end of the normal range of breeding temperatures in H. versicolor. That is, females tested at 16◦ C preferred pulse rates of males calling at that temperature to pulse rates of males calling at 20◦ C; the preference reversed when the females were tested at 20◦ C. Females tested at 24◦ C, however, did not prefer synthetic calls with pulse rates typical of males at this temperature to pulse rates that males produce at 20◦ C (Gerhardt and Doherty, 1988). By comparison, pulse rate varies in a nearly linear fashion with temperature up to at least 30◦ C (Gayou, 1984). Second, for H. versicolor, the pulse duty cycle of both alternatives has to be held constant at about 50%, as it tends to be in the calls of both species over the normal range of breeding temperatures (Section III.A.2). When the pulse duty cycle of a signal with a heterospecific pulse rate is greater than that of the standard (conspecific pulse rate) stimulus, females may no longer prefer the standard call (Gerhardt and Schul, in preparation). Third, for H. versicolor, the pulses in the alternative to the standard call have to have pulse shapes modeled after those making up the calls of H. chrysoscelis. If so, then preferences for the standard call are much stronger than when high-pulse-rate alternatives (even as high as 35–40 Hz) have pulse shapes typical of H. versicolor (Fig. 14B). Thus, there is a synergistic effect of preferences based on these two fine-temporal properties. A neurophysiological correlate is that auditory neurons in the midbrain were more likely to be selective for conspecific pulse rates if pulse shape simulated that in conspecific calls than if pulse shape was similar to that in the calls of H. chrysoscelis (Diekamp and Gerhardt, 1995). The new behavioral results with H. versicolor concerning the relevance of pulse duty cycle might force a reinterpretation of the results of neurophysiological studies of temporal selectivity of auditory neurons in the midbrain (Rose et al., 1985; Diekamp and Gerhardt, 1995). In these studies, the artificial stimuli (e.g., sinusoidally modulated noise or tones) and synthetic calls used to study pulse rate selectivity effectively kept the duty cycle constant while varying pulse rate. Thus, although band-pass neurons were common, their selectivity in the face of systematic variation in pulse duty cycle is unknown. The behavioral results predict that if band-pass neurons play a significant role in pulse-rate selectivity, then their response properties (and selectivity) will be affected strongly by variation in pulse duty cycle in H. versicolor but not in H. chrysoscelis.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

143

6. Preferences Based on Gross-Temporal Properties: Call Rate and Call Duration The gross-temporal properties of the advertisement calls of the gray treefrogs are highly dynamic, sometimes varying by as much as 100% or more from call to call (Wells and Taigen, 1986; Morris and Yoon, 1989; Gerhardt, 1991). Females of both species prefer synthetic or digitized natural calls with high rather than low values of call rate and call duration (Klump and Gerhardt, 1987; Sullivan and Hinshaw, 1992; Gerhardt et al., 1996). When offered a choice of calls that differed in duration or call rate by a factor of two, preferences for the longer and higher rate alternatives were maintained in the face of 6–10 dB reductions in their amplitude (Gerhardt et al., 1996). In H. versicolor, females preferred calls with one or two more pulses than alternatives, corresponding to differences in call duration of about 50–100 ms (Fig. 15A; Gerhardt et al., 2000). Subtle differences in the number of pulses per unit time, a function of variation in both call rate and call duration, can also influence preferences in choices between synthetic alternatives (Gerhardt and Watson, 1995). Variability per se in call duration, call rate, or both does not influence female preferences (Gerhardt and Watson, 1995).

FIG. 15. Preferences of females of Hyla versicolor based on differences in the number of pulses per call (= call duration) in synthetic calls. Lines connect points showing the proportions of females choosing each of two alternatives. The horizontal line at the top of the figure shows the range of variation in male calls, the vertical line shows the mean, and the boxes show ±1 standard deviation.

144

H. CARL GERHARDT

Preferences based on duration are not, however, completely open-ended. Although all 8 females of H. versicolor that responded chose a call of supernormal duration (about 4 s or 80 pulses/call) to a call of about average duration (975 ms or 20 pulses/call), 17 other females did not respond to either call, suggesting in this particular choice-situation neither call was very attractive (Klump and Gerhardt, 1987). Recent experiments show that both the absolute and relative durations of alternative signals affect the strength of preferences, as estimated by their maintenance in the face of SPL reductions in the longer alternative (Gerhardt et al., 2000). In these experiments, females strongly preferred the longer of two synthetic calls that differed in duration by about 50% when both had absolute durations below the mean in the population but not when both alternatives had durations above the mean. The result just mentioned supports the idea that females might more strongly prefer calls of about average duration to very short calls than they prefer very long calls to calls of average duration, a hypothesis that was suggested by an earlier study (Gerhardt et al., 1996). In that study, playback of long calls was stopped when a female of H. versicolor had moved about one-half of the distance from the release point toward the speaker playing long calls. Because playback of a short call alternative was not interrupted, females then moved back toward the source of short calls. When females reached a point about half-way between the original release point and the source of short calls, the playback of long calls was resumed. The question was then whether or not the female would return to the source of long calls. The answer depended on the absolute durations of the alternative stimuli. When the short call had a duration representative of the low-end of the distribution, most females returned to the source of longer calls, even if they were just average in duration. But when the short call had a duration near the mean in the population, only a small percentage of females returned to the source of calls with a duration that was exceptionally long. These experiments also show that females do not necessarily make long-term assessments of total acoustic stimulation if the two alternatives differ markedly in duration. Over the course of each test, the total time of acoustic stimulation by the uninterrupted short calls was obviously much greater than that of the interrupted long calls. Preferences based on differences in call duration also occur when stimulation time is equalized on a cycle-by-cycle basis. That is, females of both species preferred long calls even if their call rates were lowered so that their acoustic stimulation time was the same as that of short calls (Klump and Gerhardt, 1987; Gerhardt et al., 1996). These, and the other results discussed above, indicate a distinct bias for long calls, which might be reliable indicators of male quality (see Section IV.C.3). Future experiments that incorporate systematic variation of call duration and call rate will be required to understand “assessment rules” and the time required for females to make a

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

145

FIG. 16. Timing relationships of synthetic calls used to tests for preferences for leading calls in Hyla versicolor. (A) In one test, the leading call overlapped the lagging call in time, (B) in the other test, the leading call did not overlap the lagging call.

decision about the relative attractiveness of stimuli that vary in these two dynamic temporal properties. F. TESTS FOR PREFERENCES BASED ON RELATIVE TIMING AND AN EXAMPLE OF THE “COCKTAIL PARTY EFFECT” Klump and Gerhardt (1992) offered females choices between synthetic calls with different timing relationships (Fig. 16) and found, unlike the situation in H. cinerea, no preference for either leading or lagging calls. However, because females have strong preferences based on pulse rate, pulse duration and shape, substantial overlap in the calls of two nearby neighbors are likely to obscure these fine-temporal properties. Indeed, the effects of masking interference were evident in the early experiments of Littlejohn et al. (1960), who noted that females did not respond to playbacks when the two alternative prerecorded calls overlapped but did so when they drifted out of phase. This phenomenon was formally confirmed by Schwartz (1987), who showed that females of H. versicolor preferred loudspeakers broadcasting nonoverlapping calls rather than overlapping ones. One common solution to reducing masking interference is to use directional hearing to facilitate the discrimination of overlapping sounds that arrive from different directions. This mechanism has been called the “cocktail party effect” because of the well known ability of humans to focus selectively on a particular conversation in a crowded, noisy environment such as a party (Cherry, 1953). Recent experiments with females of H. versicolor showed that separation of speakers playing back two overlapping sounds that

146

H. CARL GERHARDT

disrupt the temporal pattern of the advertisement call can bias phonotactic choices. Specifically, females chose calls played from one of two such speakers when they were separated by 120◦ (but not 90◦ ) rather than responding to one of a second pair of adjacent speakers playing back similarly disrupted calls (Schwartz and Gerhardt, 1995). The same degree of spatial separation of speakers did not, however, result in a release from masking in the smaller treefrog H. microcephala (Schwartz, 1993). G. PREFERENCES BASED ON SOUND PRESSURE LEVEL Another difference between H. cinerea and H. versicolor concerns the dependence of preferences based on differences in intensity on the absolute playback level. Fellers (1979) reported that 3 of 5 females of H. versicolor reliably chose playbacks of a prerecorded call played back at 92 dB SPL over the same call played back at 90 dB SPL. This general result was corroborated by Gerhardt et al. (2000) who found that all 12 females they tested consistently chose a synthetic call played back at 85 dB SPL rather than the same call played back at 82 dB SPL. Thus, the intensity-discrimination ability of this species is better at higher absolute levels than is that of H. cinerea (see Section II.G). Whether gray treefrogs show preferences based on small differences in SPL at lower absolute levels is unknown. H. SUMMARY Members of the H. versicolor species group (H. arenicolor, H. avivoca, H. chrysoscelis, and H. versicolor) produce relatively long advertisement calls made up of distinctive pulses that have species-typical shapes and repetition rates. Each species also produces distinctive aggressive calls, which can be elicited by playbacks of both conspecific and heterospecific advertisement calls. The phylogenetic relationship of the pine woods treefrog H. femoralis to the core group is ambiguous despite its genetic compatibility with H. chrysoscelis. Pulses are not organized into discrete pulse trains (calls), and aggressive calls are rare or lacking in its repertoire. Extensive experimental studies are available for the two species of gray treefrogs. With regard to spectral properties, females of both species: (1) do not require frequency modulated pulses such as those produced by conspecific males, (2) prefer signals with two spectral peaks such as those produced by conspecific males rather than signals with a single peak, and (3) show weak stabilizing preferences with regard to variation in dominant frequency. Some frequency preferences in H. versicolor depend on whether the alternative stimuli have just the high-frequency peak or both peaks. These results suggest complex interactions (cross talk) between the two sensory channels that are tuned to, and process, acoustic energy in these two bands.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

147

Female gray treefrogs tested at about 20◦ C prefer synthetic signals having the same pulse rate as those in the calls of a conspecific male recorded at 20◦ C to alternatives in which the pulse rate is 20–25% lower or 50% higher. Resolution of some fine-temporal differences in H. versicolor appears to require processing by auditory neurons that are most sensitive to the lowfrequency peak in the call. Both pulse rate and pulse-rate preferences are positively correlated with temperature and shift in a roughly parallel fashion over the normal range of breeding temperatures. Females of both species of gray treefrogs show strongly directional and modestly intensity-independent preferences based on call rate and call duration. Preferences based on call duration are stronger when both alternatives are short than when both alternatives are long, even if the percentage difference in duration is the same. Some preferences for the longer of two alternatives are not explained by differences in total acoustic stimulation. Unlike females of H. cinerea, females of H. versicolor show no preference for leading signals over lagging signals and show preferences based on small differences in amplitude at moderateto-high absolute playback levels. Gray treefrogs appear to use their directional hearing to resolve the temporal properties of attractive signals that overlap in time. As in the comparison of H. cinerea and H. gratiosa, differences in acoustic criteria between the two species of gray treefrogs are numerous. Despite the conservation of the spectral structure of the call, females of H. versicolor prefer the high-frequency spectral peak to the low-frequency peak, whereas females of H. chrysoscelis do not show a preference. In H. chrysoscelis, pulserate preferences are not affected by varying pulse duration (and hence changes in pulse duty cycle), and preferences are much more intensityindependent when alternatives have lower-than-average pulse rates than when alternatives have higher-than-average rates. In H. versicolor, preferences are based on joint variation in pulse duration and intervals between pulses, which means that there has been a change in how these temporal properties are evaluated rather than a simple shift in the temporal tuning of a common mechanism. Females of H. versicolor but not H. chrysoscelis are also highly selective with respect to pulse shape and rise-time. These species differences are somewhat surprising because H. versicolor arose by polyploidy from H. chrysoscelis.

IV. COMPARISONS WITH OTHER TAXA A. SIGNAL REPERTOIRES AND AGGRESSIVE SIGNALS The typical repertoire of species in the green and gray treefrog groups is intermediate in size compared with other anuran taxa. Whereas spadefoot

148

H. CARL GERHARDT

toads (family Pelobatidae) and toads (family Bufonidae) apparently produce only advertisement calls and release calls, territorial members of the family Ranidae (true frogs) may produce a variety of aggressive calls in addition to advertisement, release, and distress calls (Capranica, 1965; Duellman and Trueb, 1986). In some tropical anurans (Rana nicrobariensis, Philatus leucorhynus, Boophis madagascariensis), repertoires have been conservatively estimated as consisting of as many as eight different signals (Arak, 1983; Jehle and Arak, 1998; Narins et al., 2000). Moreover, the variety of acoustic signals in these and other species can be magnified because aggressive signals are often graded in their acoustic structure (for review, see Wells, 1988; Grafe, 1995). The functions of the signals in extended repertories still have to be established. Indeed only a few studies have investigated conspecific reactions to graded signals (e.g., Grafe, 1995) or how other males react to aggressive signals that do not show graded variation (e.g., Brenowitz and Rose, 1997). Males that produce distinctive aggressive calls face a dilemma because females usually discriminate against such signals in favor of advertisement calls (e.g., Afrixalus brachycnemis, Hyla cinerea, H. ebraccata, H. microcephala, Hyperolius marmoratus, Pseudacris regilla) (Oldham and Gerhardt, 1975; Schwartz and Wells, 1985; Wells and Bard, 1987; Backwell, 1988; Grafe, 1995; Brenowitz and Rose, 1999). One solution adopted by some species is the production of diphasic advertisement calls. Some evidence exists for two species that one part of the signal functions to attract females, and the other part to repel rival males (e.g., Eleutherodactylus coqui; Narins and Capranica, 1978, but see Stewart and Bishop, 1994; Geocrinia victoriana; Littlejohn and Harrison, 1985). In still other species, males that are interacting vocally with other males add components to advertisement calls that enhance their attractiveness to females (e.g., Physalaemus pustulosus, Hyla ebraccata, H. microcephala; Ryan, 1985; Wells, 1988). These additional components are presumably not as effective in repelling males as aggressive calls. Recall that males of H. cinerea, H. avivoca, and the two gray treefrogs often add a few aggressive calls to the end of a series of advertisement calls even when calling in isolation. Could these combinations be more attractive than advertisement calls alone? Or are the males merely providing a warning to other males of possible escalation to more aggressive behavior while still producing signals that effectively attract females? As in the two species of gray treefrogs, the aggressive calls of three closely related species of neotropical treefrogs (Hyla ebraccata, H. microcephala, and H. phlebodes) are similar in acoustic structure (Wells, 1988). Playbacks and field observations indicate that males of all three species react in a similar fashion to the aggressive calls of the other species (Wells and Schwartz, 1984a). By contrast, the advertisement calls of these three species are well differentiated in pulse rate, and females of the first two species prefer

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

149

conspecific calls (Wells and Schwartz, 1984a; Schwartz, 1986; Backwell and Jennions, 1993). Thus, in these neotropical species, aggressive-call structure might have been conserved, whereas the advertisement calls have diverged. Better evidence for the conservation of aggressive-call structure (as opposed to convergence) exists for the two species of gray treefrogs because we know that the tetraploid species is derived from the diploid one (Section III.B). B. SIGNALING INTERACTIONS AND THE PRECEDENCE EFFECT For the most part, the treefrogs discussed in this review are typical of most anurans and some acoustic insects in that males generally avoid acoustic overlap with their nearest or loudest neighbors (except perhaps in very dense choruses: Section III. C), resulting in alternation of calls (Klump and Gerhardt, 1992; Greenfield, 1994). Greenfield (1994) emphasizes that in most species, whether or not males synchronize their calls or tend to alternate them, males compete to produce calls that lead those of their neighbor in time. This pattern of behavior may be favored by preferences for leading signals, which are widespread, and especially strong if the two signals overlap (Greenfield, 1994; Grafe, 1996). Greenfield (1994) considers such preferences to be examples of the precedence effect, a term has been applied to a variety of masking and binaural (stereophonic) effects (Wallach et al., 1949; for review, see Zurek, 1987). These rules of signal timing and preference are, however, not universal, and variation exists both within (Sections II.C and III.F) and between species. Recall that females of H. cinerea prefer leading calls, whereas females of H. versicolor show no preference. In one species of African anuran, Kassina fusca, males deliberately overlap the calls of neighbors, and females prefer lagging calls, provided that the degree of overlap is limited to about 10–25% (Grafe, 1999). Females of Hyla ebraccata, a neotropical hylid, also prefer lagging calls (Wells and Schwartz, 1984b). These are clear counterexamples to the so-called precedence effect. Discovering the mechanisms and functional significance in these exceptional systems should aid our understanding of the causes and consequences of signaling interactions in general. C. PHONOTACTIC SELECTIVITY IN FEMALES 1. Phonotaxis to Heterospecific Signals, Species Recognition, and Mate Choice Phonotaxis to heterospecific calls in single-speaker tests is not restricted to members of the H. cinerea and H. versicolor groups. Females of other species of anurans and some acoustic insects also respond to some kinds

150

H. CARL GERHARDT

of heterospecific signals in this situation, and, as in the treefrogs discussed here, the great majority of these same females prefer conspecific signals to the same heterospecific signals in choice situations (Backwell and Jennions, 1993; Ryan and Rand, 1993a; Schul et al., 1998). Phonotaxis to the calls of heterospecific males indicates that these sounds contain some minimum set of attractive properties. Discrimination against these same sounds in favor of conspecific signals indicates that the values of those properties are suboptimal, that heterospecific signals lack other attractive components found in conspecific calls, or both. Recall that in the H. cinerea group females chose hybrid calls that had spectral properties most similar to conspecific calls; these preferred hybrid calls were, however, seldom as attractive as conspecific calls (Section II.D.1). Studies using synthetic calls also show that quantitative variation in pertinent properties generally results in graded differences in preference strength. Moreover, there is no sudden discontinuity in preference strength once the value of an acoustic property exceeds the range of variation in conspecific calls or falls into the range of heterospecific calls. Thus, selective phonotaxis that results in preferences for conspecific calls to heterospecific calls is qualitatively equivalent to preferences that lead to differential mating within a species. Thus, as emphasized by Littlejohn (1981), discrimination against heterospecific calls (often considered as a kind of “isolating mechanism”) must often be a by-product (consequence instead of cause) of intraspecific choice (see also Paterson, 1985). This fact does not rule out the possibility that negative consequences of mismating can sometimes play a role in the evolution of call structure and preferences, but this hypothesis has to be tested on a case by case basis (Section V; Gerhardt, 1982; Pfennig, 1998). Females of some species do not respond to heterospecific calls in no-choice tests (Morris and Fullard, 1983; Ryan and Rand, 1993a; Schul, 1998). Because females of the katydid Concephalus nigropleurum responded to playbacks of songs of an allopatric congener but not a sympatric one (Morris and Fullard, 1983), Gwynne and Morris (1986) suggested that this species might have a special mechanism for recognizing and rejecting heterospecific signals. However, both this species and the katydid Tettigonia viridissima (Schul, 1998) ignored rather than avoided speakers playing back heterospecific signals. In experiments with treefrogs (the two gray treefrogs and H. gratiosa), females were tested in situations where they had the opportunity to avoid sources of heterospecific signals during phonotactic responses to conspecific signals (Gerhardt et al., 1994b). The females did not avoid the heterospecific signals, which, in my view, would provide good evidence that females “classify” heterospecific signals differently than conspecific signals. This concept is exemplified by many species of acoustic insects, which show positive phonotaxis to conspecific songs and negative phonotaxis or evasive

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

151

behaviors in response to signals simulating the echolocation calls of bat predators. These two classes of signals are largely processed by different ascending auditory neurons and obviously elicit incompatible motor responses (for review, see Pollack, 1998). 2. Species Differences in Acoustic Criteria One recurring theme in this review is that different species often use entirely different acoustic criteria for selecting among signals. The best examples are very closely related: H. cinerea and H. gratiosa in the green treefrog group, and H. chrysoscelis and H. versicolor in the gray treefrog group. Even when females selectively respond to synthetic calls with the same acoustic difference, they may evaluate that difference in an entirely novel way. Recall that both species of gray treefrog respond selectively to the joint differences in pulse rate and pulse duration that simulate the species differences between their advertisement calls. Yet, deeper analysis shows that females of H. chrysoscelis use the difference in pulse rate alone, whereas the selectivity of females of H. versicolor is based on combinations of pulse duration and pulse intervals (Section III.E.4). A similar conclusion emerges from studies of European katydids. As in H. chrysoscelis, females of Tettigonia cantans appear to rely purely on pulse-rate differences, whereas females of T. viridissima have a recognition space based on joint variation in pulse duration and intervals (Schul, 1998). Closely related species of grasshoppers (genus Chlorthippus) also have strikingly different preference criteria for their complex, two-part signals (Stumpner and Helversen, 1994) and, within one species of the same genus (C. biguttulus), males and females (which also signal) differ in the criteria used to discriminate between the signals of the two sexes (Helversen and Helversen, 1997). These results suggest that the evolution of different receiver mechanisms, which often involves more than merely “retuning,” appears to be at least as labile as the mechanisms leading to changes in signals. Moreover, these results serve to emphasize again that it is dangerous to assume that an acoustic property is relevant for a given species just because that property is stereotyped and can be used by humans for distinguishing between species. 3. Phonotactic Preferences and Potential Consequences for Intraspecific Mate Choice a. Spectral Properties. Females of other taxa (anurans and acoustic insects) show either little selectivity for signals that differ in frequency within the conspecific range or patterns of preference that can be characterized as stabilizing or weakly directional (20 of 27 species or populations; Gerhardt and Huber, in preparation). That is, frequencies near the mean in the

152

H. CARL GERHARDT

population are usually as attractive or more attractive than frequencies representative of the two ends of the range of distribution. Even so, detailed studies usually document some directionality, which arises because the most preferred frequency does not always correspond exactly to the mean or because females discriminate against signals from one end of the distribution more strongly than they discriminate against signals from the other end of the distribution. These general results probably reflect the fact that although the auditory sensitivity of a given species usually matches reasonably well with the frequencies emphasized in conspecific advertisement calls, the maximum sensitivity seldom corresponds exactly with the mean frequencies in the population. Moreover, a few large mismatches have also been documented (for review, see Gerhardt and Schwartz, 2001; Gerhardt and Huber, in preparation). Large-male mating advantages have been found in some of the species that prefer calls of lower-than-average frequency (Gerhardt, 1994a). This mating pattern is expected because dominant frequency is negatively correlated with body size in these and in other frogs and toads (Ryan, 1988; Gerhardt, 1991). However, call intensity, which is positively correlated with male size (Gerhardt, 1975), can also mediate female preferences for large males (e.g., Arak, 1988). When tested, even directional preferences for lowerthan-average frequencies are usually abolished or reversed by decreasing the relative amplitude of the preferred signal by 6–9 dB (exception, one population of Acris crepitans studied by Ryan et al., 1992). Such preferences can also be abolished or reversed by having the less-preferred signal lead the preferred signal in time (e.g., Dyson and Passmore, 1988a,b; Howard and Palmer,1995). In species in which frequency preferences probably contribute to sizedependent mating success (including size-assortative mating), females obtain a direct benefit in the form of increased fertilization success (e.g., Ryan, 1985; Robertson, 1990; Bourne, 1993). For Physalaemus pustulosus, this direct benefit is interpreted as a selective force that maintains but does not explain the origin of the preference (Ryan, 1990; Ryan and Rand, 1993b). This view stems from a phylogenetic analysis indicating that the frequency preference, mediated by “chucks” that are added to the advertisement call, arose before the evolutionary appearance of chucks. Because call frequency and body size are well correlated in interspecific comparisons (Ryan, 1988; Gerhardt, 1991), species differences in dominant frequency might often be an indirect consequence of direct selection on body size. Better evidence for long-term effects of direct selection on dominant frequency is probably provided by comparative studies that identify species such as P. pustulosus, in which frequency is much lower than expected from male body size (Ryan, 1988).

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

153

b. Fine-Temporal Properties. Preferences based on fine-temporal properties such as pulse rate are almost universally stabilizing (Gerhardt, 1988, 1991). This result, and the fact that pulse rate is not only highly static within males but also varies little between males in a population, supports the bias of many biologists that the main result of preferences based on pulse rate is the choice of conspecific males. Indeed, the range of between-male variation in (temperature-corrected) pulse rate in a typical population is usually smaller than the range of equally attractive values. This observation may explain why, to my knowledge, no study of anurans has ever set out to correlate intrapopulation variation in pulse rate with mating success nor has such a relationship ever been reported. There are two problems with this generalization. First, recall that preference functions based on pulse rate are sometimes asymmetrical, suggesting the possibility of weakly directional selection. As shown in Fig. 14, for example, females of H. chrysoscelis more strongly discriminate against alternatives with lowerthan-average pulse rates than they discriminate against alternatives with higher-than-average pulse rates. Second, pulse rate preferences are not always strongly stabilizing. Recent studies report, for example, that even when playback levels of alternatives were equalized, substantial proportions of females of two anuran species (neotropical treefrogs, Hyla ebraccata, and green toads, Bufo viridis) responded to synthetic stimuli with pulse rates well outside the range of variation (Castellano and Giacoma, 1998; Wollerman, 1998). c. Dynamic, Gross-Temporal Properties. Surveys of other anurans and acoustic insects indicate a general trend for highly directional preferences to be based on dynamic call properties. In 26 of 30 species (or populations) of insects and anurans, females preferred high values of dynamic properties to values near the mean (Gerhardt and Huber, in preparation; see also Ryan and Keddy-Hector, 1992). In three of the exceptions, researchers found that females choosing calls with high values of duration or call rate might risk mating with a heterospecific male (see Section V.B). Male mating success has been shown to correlate directly with high values of call rate in eight studies of anurans (e.g., Passmore et al., 1992; Grafe, 1997; for review, see Gerhardt, 1994a; Schwartz, Buchanan and Gerhardt, in press). Female preferences for such high values are likely to have contributed to these differences in male mating success. However, males producing calls at high rates might also have increased success simply because their signals should be more easily detectable in dense choruses than those of males producing calls at low rates (Gerhardt, 1991). Comparisons of mating success in large and small choruses might be able to estimate the relative contributions of these two factors to increased male mating success.

154

H. CARL GERHARDT

Preferences for signals that provide extra acoustic stimulation, as do calls with high values of dynamic properties, have previously been noted in a wide variety of taxa by Ryan and Keddy-Hector (1992). These authors suggest that this pattern reflects a general preexisting bias for high levels of stimulation. This kind of explanation cannot, however, identify the selective forces that might be responsible for the evolutionary origins and maintenance of the preferences. As I argued earlier (Section I), the same kinds of selective pressures that currently maintain a trait must often have favored its origin and further evolution. A preference mechanism based solely on differences in the amount of acoustic stimulation is also inadequate to explain some patterns of preference, such as those found in gray treefrogs (Klump and Gerhardt, 1987; Section: III.E.6). Moreover, in the painted reed frog Hyperolius marmoratus broadleyi (Grafe, 1997) and the katydid Requena verticalis (Schatral and Bailey, 1991), females favored some signals that provided less acoustic stimulation than alternatives. Dynamic properties seem especially suitable as honest indicators of male attributes that could benefit females either directly or indirectly because of the strong correlation between the values of dynamic properties and energetic costs. Evidence for direct benefits comes from studies of spadefoot toads Scaphiopus multiplicata, in which females paired with males producing calls at a high rate had greater fertilization success than did females paired with males producing calls at a slow rate (Pfennig, 2000). Evidence for indirect benefits is provided by a study of H. versicolor, in which the eggs of a series of females were fertilized both with the sperm of males that produced the long calls (preferred by females) and males that produced very short calls (Welch et al., 1998). All statistically significant differences in larval performance (growth rates, size at metamorphosis, and time to metamorphosis), which correlate with adult survival, favored the offspring of long callers.

V. GEOGRAPHIC VARIATION IN ACOUSTIC COMMUNICATION So far I have reviewed results of studies that primarily examined acoustic signals and female preferences in one or a few nearby populations of a given species. Here I consider geographic variation in both signals and preferences. Such differences are expected to arise from differences in selection occurring in different parts of the range and from stochastic processes such as mutation and drift. Identifying sources of selection that explain patterns of geographical variation in communication systems is difficult because the interplay of these evolutionary forces is likely to be complex and also to vary in time and space.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

155

A. ADVERTISEMENT CALLS Geographic variation in advertisement calls has been extensively documented in nearly a dozen species of anurans. A major goal of many of these studies was to seek evidence for reproductive character displacement or reinforcement (sensu Loftus-Hills and Littlejohn, 1992). This process predicts a geographic pattern in which the divergence of acoustic properties of known or probable significance for female choice of conspecific signals is greater in sympatric areas than in allopatric areas. Although studies of other groups of animals find scant evidence for reproductive character displacement (orthopteran insects: Butlin, 1987; Alexander et al., 1997; birds: Miller, 1982; but see Otte, 1989, and Howard, 1993, for a different view), three of five such studies of anurans, in which sampling was adequate, yielded positive results. The species pairs whose advertisement calls show the expected pattern of divergence in sympatry include the chorus frogs Pseudacris feriarum and P. nigrita in the southeastern United States (Fouquette, 1975), the treefrogs Litoria ewingii and L. verreauxii in southeastern Australia (Littlejohn, 1965), and the narrowmouth toads Gastrophryne carolinesis and G. olivacea in the south-central United States (Loftus-Hills and Littlejohn, 1992). The last two of these studies could be improved by additional sampling in allopatric areas to make sure that divergence in sympatry is not a fortuitous consequence of a trend established in allopatric areas for other reasons (Grant, 1972). It will also be important to estimate the minimum differences in geographically variable (displaced) properties that are sufficient to elicit selective phonotaxis (Gerhardt and Schwartz, 1995; Gerhardt, 1999). A stronger case for reproductive character displacement could be made if the minimum difference were always found in sympatric and nearby populations and smaller, nondiscriminable differences occurred in allopatric populations. Some evidence along these lines is available for Litoria ewingii and L. verreauxii (Littlejohn and Loftus-Hills, 1968). No evidence for reproductive character displacement of call properties was found in the cricket frog (Acris crepitans) in comparisons between the calls of males from areas of allopatry and areas of sympatry with a congener (Acris gryllus) (Nevo and Capranica, 1985) or between the calls of two subspecies of A. crepitans in Texas, United States (Ryan and Wilczynski, 1991). Instead, the geographical variation in advertisement calls within Texas is mainly clinal, explained in part by a parallel cline in male body size. Significant geographic variation remains after statistically removing the effect of body size, and some nonclinal variation in temporal properties and dominant frequency is correlated with differences in habitat (for review, see Wilczynski and Ryan, 1999).

156

H. CARL GERHARDT

Reproductive character displacement of pulse rate might well be expected between the genetically incompatible gray treefrogs. Ralin (1977) reported that males in sympatric populations of the gray treefrog H. chrysoscelis in east-central Texas had higher pulses rates than allopatric populations in west-central Texas. However, a more extensive study involving three detailed transects of more than 1000 km each found that sympatry between diploid and tetraploid gray treefrogs explains very little of the geographical variation in pulse rate (Gerhardt, 1999; Gerhardt, Keller, Ptacek, and Forester, in preparation). Variation is primarily clinal, with pulse rate increasing strongly from east to west and less so from south to north. In the eastern part of the range, a weak correlation exists between geographic variation in pulse rate and a karyotypic polymorphism involving the location of the nucleolar organizing region (Wiley, 1983; Gerhardt, Keller, Ptacek, and Forester, in preparation). An extensive study of call and allozyme variation in the wide-ranging neotropical species, Physalaemus pustulosus, found that some acoustic properties show clinal variation, but most other properties are correlated with temporally separated invasions of Panama. These invasions are suggested by the existence of two major allozyme groups (Ryan et al., 1996). Effects of isolation by distance on call similarity were also demonstrated. This is a rich system for future studies because so much is known about properties of importance in eliciting selective phonotaxis, at least in populations from Panama (e.g., Wilczynski et al., 1995). It is not a good system for studying reproductive character displacement because of the absence of areas of extensive sympatry with other species in the group. B. FEMALE PREFERENCES Relatively few studies of geographic variation in female preferences are available. Waage (1979) suggested that evidence for reproductive character displacement might be uncommon because researchers usually focus on geographical variation in signals rather than preferences. If differences in signals are adequate for discrimination between two taxa that reestablish contact, then selection might simply act to favor receivers (usually females) that prefer conspecific signals rather than promoting further divergence in the signals. Moreover, selection on female insects and anurans is likely to be stronger than on males because females usually make mating decisions and have more to lose from a mistake than do males. These ideas are supported by studies of the gray treefrog (H. chrysoscelis). Females prefer the pulse rates close to the average in the calls of local males to lower pulse rates (in the direction of H. versicolor), regardless of whether H. versicolor also occurs in the same area (Gerhardt, 1999). However, the strength of the preference varies geographically in a way that is predicted

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

157

by reproductive character displacement (Gerhardt, 1994b, 1999). That is, the preferences of many females of H. chrysoscelis from allopatric areas are readily abolished or reversed by increasing the call duration or amplitude of the synthetic call with the lower pulse rate (more similar to that in H. versicolor), whereas most females from sympatric areas continue to choose signals with conspecific values under these conditions. These results have now been generalized over four remote allopatric and four sympatric populations (Gerhardt, 1994b, 1999). Recent comparisons of the preferences of midwife toads (Alytes cisternasii and A. obstetricans) and spadefoot toads (Scaphiopus multiplicatus) from allopatric and sympatric areas (unfortunately, only one population of each type) also support the hypothesis that interspecific interactions can affect the evolution of preferences (Marquez ´ and Bosch, 1997b; Pfennig, 2000). Females from allopatric areas showed the usual directional preference functions based on dynamic call properties, whereas females from sympatric populations did not. As mentioned in Section IV.C.3, females of S. multiplicatus that were mated with males producing high call rates had more eggs fertilized than females that were mated with males producing low call rates (Pfennig, 2000). The lack of a directional preference in sympatry suggests that the costs of mating with a genetically incompatible (heterospecific) mate must outweigh the benefits of mating with a conspecific male who can increase fertilization success. Evidence for increased selectivity in receivers in sympatric versus allopatric areas has been documented in insects and birds as well (references in Gerhardt, 1994b).

VI. SUMMARY Studies of vocal behavior in anurans provide significant insights about the mechanisms and evolution of animal communication. Females extract the information required for a mating decision from relatively stereotyped advertisement calls that are produced by males. Aggressive signals, used in exchanges between males, are more variable in form, as might be expected of communication between rivals, but overall, their structure is more similar in closely related species than is that of advertisement calls. One open question is why, within the groups of treefrogs considered in this review, aggressive signaling is so prevalent in some species and rare or absent in close relatives. Comparative studies of the acoustic criteria used in signal selection reveal significant differences among species. What are important properties for one species are seemingly irrelevant for its closest relative. Moreover, as exemplified by gray treefrogs and European katydids, even when females apparently base preferences on the same acoustic property, deeper analysis reveals that

158

H. CARL GERHARDT

different mechanisms are used by each species to extract a different aspect of that property. In several of the treefrog species considered in this review, some preferences and preference-mechanisms are more divergent than call structure (Sections II.E and III.E). More research with additional species is needed before we can even attempt to make broad generalizations about the coevolution of signalers and receivers in animal communication systems. Communication usually takes place in acoustically difficult situations: dense choruses, where background noise presents special challenges for signal detection and pattern recognition. Although they can quantify acoustic preferences, playback experiments in simplified laboratory situations are unlikely to accurately predict patterns of mate choice in nature. Studies with species in the green and gray treefrog groups show that the optimum values of particular acoustic properties can vary depending on the absolute playback level to which alternatives are equalized, the relative playback levels of alternatives, the number of sounds presented and their timing relationships, and the ambient temperature. Comparable results are available for other anurans and acoustic insects. Technology that allows researchers to monitor groups of calling males, whose signals and mating success can be quantified, promises to refine our estimates of how female selectivity contributes to mate choice in nature (e.g., Passmore et al., 1992; Schwartz, 1994, 2001). Studies of geographical variation in communication systems are limited because we know so little about patterns of variation in receiver selectivity. Even where acoustic properties have diverged in sympatric areas relative to allopatric areas, and hence show the pattern expected of reproductive character displacement, we often do not always know whether these properties are even relevant to females. A few studies have shown that female selectivity diverges in sympatric areas, whereas signal structure does not, a result that may reflect the fact that the choosing sex usually has more to lose from mating mistakes than the signaling sex. I predict, however, that studies of geographical variation in the acoustic criteria and selectivity of receivers will ultimately provide significant insights concerning the evolution of communication systems that will go well beyond the narrow issue of reproductive character displacement. Acknowledgments I thank the following individuals who contributed to the work reviewed here, provided useful comments on the manuscript, or both: Mark Bee, Bryant Buchanan, Sarah Bush, Reginald Cocroft, Richard Daniel, Bettina Diekamp, John Doherty, Sarah Humfeld-Conditt, Michael Keller, Georg Klump, Vincent Marshall, Christopher Murphy, Stephen Perrill, Margaret Ptacek, Michael Ritchie, Johannes Schul, Joshua Schwartz, Peter Slater, Charles Snowdon, Steven Tanner, and Allison Welch. The National Science Foundation and National Institutes of Health have generously funded my research.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

159

References Alexander, R. D., Marshall, D. C., and Cooley, J. R. (1997). Evolutionary perspectives on insect mating. In “The Evolution of Mating Systems in Insects and Arachnids” (J. C. Choe and B. J. Crespi, eds.), pp. 4–31. Cambridge Univ. Press. UK. Anderson, K., and Moler, P. E. (1986). Natural hybrids of the pine-barrens treefrog Hyla andersonii with Hyla cinerea and Hyla femoralis (Anura: Hylidae): Morphological and chromosomal evidence. Copeia 1986, 70–76. Arak, A. (1983). Vocal interactions, call matching, and territoriality in a Sri Lankan treefrog, Philautus leucorhinus (Rhacophoridae). Anim. Behav. 31, 292–302. Arak, A. (1988). Female mate selection in the natterjack toad: Active choice or passive attraction? Behav. Ecol. Sociobiol. 22, 317–327. Backwell, P. R. Y. (1988). Functional partitioning in the two-part call of the leaf folding frog Afrixalus brachycnemis. Herpetologica 44, 1–7. Backwell, P. R. Y., and Jennions, M. D. (1993). Mate choice in the neotropical frog Hyla ebraccata: Sexual selection, mate recognition, and signal selection. Anim. Behav. 45, 1248–1250. Barber, P. H. (1999a). Phylogeography of the canyon treefrog, Hyla arenicolor (Cope) based on mitochondrial DNA sequence data. Mol. Ecol. 8, 547–562. Barber, P. H. (1999b). Patterns of gene flow and population genetic structure in the canyon treefrog, Hyla arenicolor (Cope). Mol. Ecol. 8, 563–576. Blair, W. F. (1959). Call structure and species groups in U.S. treefrogs (Hyla). Southwest. Nat. 3, 77–89. Blair, W. F. (ed.). (1972). “Evolution in the Genus Bufo.” Univ. of Texas Press, Austin. Bogart, J. P., and Wasserman, A. O. (1972). Diploid–polyploid cryptic species pairs: A possible clue to evolution by polyploidization in anuran amphibians. Cytogenetics 11, 7–24. Bogert, C. M. (1960). The influence of sound on the behavior of amphibians and reptiles, In “Animal Sounds and Communication” (W. E. Lanyon and W. N. Tavloga, eds.), pp. 137–320. American Institute Biological Sciences, Washington, DC. Bourne, G. R. (1993). Proximate costs and benefits of mate acquisition at leks of the frog Ololygon rubra. Anim. Behav. 45, 1051–1059. Bullard, A. J. (1965). Additional records of the treefrog Hyla andersonii from the coastal plain of North Carolina. Herpetologica 21, 154–155. Brenowitz, E. A., and Rose, G. J. (1997). Plasticity of aggressive thresholds in Hyla regilla: Discrete accommodation to encounter calls. Anim. Behav. 53, 353–361. Brenowitz, E. A., and Rose, G. J. (1999). Female choice and plasticity of male calling behaviour in the Pacific treefrog. Anim. Behav. 57, 1337–1342. Brown, L. E., and Pierce, J. R. (1965). Observations on the breeding behavior of certain anuran amphibians. Texas J. Sci. 26, 313–317. Brush, J. S., and Narins, P. M. (1989). Chorus dynamics of a Neotropical amphibian assemblage: Comparison of computer simulation and natural behaviour. Anim. Behav. 37, 33–44. Butlin, R. K. (1987). Speciation by reinforcement. Trends Ecol. Evol. 2, 8–13. Capranica, R. R. (1965). “The Evoked Vocal Response of the Bullfrog:. A Study of Communication by Sound.” M.I.T. Press, Cambridge, MA. Capranica, R. R., and Moffat, A. J. M. (1983). Neurobehavioral correlates of sound communication in anurans. In:“Advances in Vertebrate Neuroethology” (J.-P. Ewert, R. R. Capranica, and D. J. Ingle, eds.), pp. 701–730. Plenum, New York. Castellano, S., and Giacoma, C. (1998). Stabilizing and directional female choice for male calls in the European green toad. Anim. Behav. 56, 275–287. Cherry, E. C. (1953). Some experiments on the recognition of speech, with one and with two ears. J. Acoust. Soc. Am. 25, 975–979.

160

H. CARL GERHARDT

Cocroft, R. B. (1994). A cladistic analysis of chorus frog phylogeny (Hylidae: Pseudacris). Herpetologica 50, 420–437. Conant, R., Collins, J. T., and Hunt, I. C. (1998). “A Field Guide to Reptiles and Amphibians: Eastern and Central North America.” Houghton Mifflin, Boston. da Silva, H. R. (1997). Two character states new to hylines and the taxonomy of the genus Pseudacris. J. Herpetology 31, 609–613. Diekamp, B. M., and Gerhardt, H. C. (1995). Selective phonotaxis to advertisement calls in the gray treefrog Hyla versicolor: Behavioral experiments and neurophysiological correlates. J. Comp. Physiol. A 177, 173–190. Doherty, J. A., and Gerhardt, H. C. (1984). Acoustic communication in hybrid treefrogs: Sound production by males and selective phonotaxis of females. J. Comp. Physiol. 154, 319–330. Duellman, W. A., and Trueb, L. (1986). “Biology of Amphibians.” McGraw-Hill New York. Dyson, M. L., and Passmore, N. I. (1988a). Two-choice phonotaxis in Hyperolius marmoratus (Anura: Hyperoliidae): The effect of temporal variation in presented stimuli. Anim. Behav. 36, 648–652. Dyson, M. L., and Passmore, N. I. (1988b). The combined effect of intensity and the temporal relationships of stimuli on phonotaxis in female painted reed frogs Hyperolius marmoratus. Anim. Behav. 36, 1555–1556. Fellers, G. M. (1979). Aggression, territoriality, and mating behavior in North American treefrogs. Anim. Behav. 27, 107–119. Feng, A. S., and Schellart, N. A. M. (1999). Central auditory processing in fish and amphibians. In: “Comparative Hearing: Fish and Amphibians” (R. R. Fay and A. N. Popper, eds.), pp. 218–268. Springer, New York, Berlin, Heidelberg. Fouquette, M. J. (1975). Speciation in chorus frogs. I. Reproductive character displacement in the Pseudacris nigrita complex. Syst. Zool. 24, 16–23. Forrest, T. G. (1994). From sender to receiver: Propagation and environmental effects on acoustic signals. Am. Zool. 34, 644–654. Fuzessery, Z. M. (1988). Frequency tuning in the anuran central auditory system. In “The Evolution of the Amphibian Auditory System” (B. Fritzsch, T. Hetherington, M. J. Ryan, W. Wilczynski, and W. Walkowiak, eds.), pp. 253–273. Wiley, New York. Gayou, D. C. (1984). Effects of temperature on the mating call of Hyla versicolor. Copeia 1984, 733–737. Gerhardt, H. C. (1968). “Evolutionary Aspects of Vocal Communication and Reproductive Behavior in the Hyla cinerea Species Group.” M. A. Thesis, University of Texas, Austin. Gerhardt, H. C. (1970). “Selected Aspects of the Reproductive Biology of Some Southeastern United States Hylid Frogs.” Ph.D. Thesis, University of Texas, Austin. Gerhardt, H. C. (1974a). Vocalizations of some hybrid treefrogs: Acoustic and behavioral analyses. Behaviour 49, 130–151. Gerhardt, H. C. (1974b). Behavioral isolation of the treefrogs Hyla cinerea and Hyla andersonii. Am. Midl. Natur. 91, 424–433. Gerhardt, H. C. (1974c). The significance of some spectral features in mating call recognition in the green treefrog (Hyla cinerea). J. Exp. Biol. 61, 229–241. Gerhardt, H. C. (1974d). Mating call differences between eastern and western populations of the treefrog Hyla chrysoscelis. Copeia 1974, 534–536. Gerhardt, H. C. (1975). Sound pressure levels and radiation patterns of the vocalizations of some North American frogs and toads. J. Comp. Physiol. A. 102, 1–12. Gerhardt, H. C. (1976). Significance of two frequency bands in long distance vocal communication in the green treefrog. Nature (London) 261, 692–694. Gerhardt, H. C. (1978a). Temperature coupling in the vocal communication system of the gray treefrog Hyla versicolor. Science 199, 992–994.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

161

Gerhardt, H. C. (1978b). Discrimination of intermediate sounds in a synthetic call continuum by female green tree frogs. Science 199, 1089–1091. Gerhardt, H. C. (1978c). Mating call recognition in the green treefrog (Hyla cinerea): The significance of some fine-temporal properties. J. Exp. Biol. 74, 59–73. Gerhardt, H. C. (1981a). Mating call recognition in the green treefrog (Hyla cinerea): Importance of two frequency bands as a function of sound pressure level. J. Comp. Physiol. 144, 9–16. Gerhardt, H. C. (1981b). Mating call recognition in the barking treefrog (Hyla gratiosa): Responses to synthetic calls and comparisons with the green treefrog (Hyla cinerea). J. Comp. Physiol. 144, 17–25. Gerhardt, H. C. (1982). Sound pattern recognition in some North American treefrogs (Anura: Hylidae): Implications for mate choice. Am. Zool. 22, 581–595. Gerhardt, H. C. (1987). Evolutionary and neurobiological implications of selective phonotaxis in the green treefrog (Hyla cinerea). Anim. Behav. 35, 1479–1489. Gerhardt, H. C. (1988). Acoustic properties used in call recognition by frogs and toads. In “The Evolution of the Amphibian Auditory System.” (B. Fritzsch, T. Hetherington, M. J. Ryan, W. Wilczynski, and W. Walkowiak, eds.), pp. 455–483. Wiley, New York. Gerhardt, H. C. (1991). Female mate choice in treefrogs: Static and dynamic acoustic criteria. Anim. Behav. 42, 615–635. Gerhardt, H. C. (1994a). The evolution of vocalization in frogs and toads. Annu. Rev. Ecol. Syst. 25, 293–324. Gerhardt, H. C. (1994b). Reproductive character displacement of female mate choice in the grey treefrog H. chrysoscelis. Anim. Behav. 47, 959–969. Gerhardt, H. C. (1998). Acoustic signals of animals: Field measurements, recording, analysis, and description. In “Techniques in the Study of Animal Communication by Sound” (S. Hopp and M. Owren, eds.), pp. 1–25. Springer, Berlin, Heidelberg, New York. Gerhardt, H. C. (1999). Reproductive character displacement and other sources of environmental selection on acoustic communication systems. In “The Design of Animal Communication” (M. Hauser and M. Konishi, eds.), pp. 515–534. M.I.T. Press, Cambridge, MA. Gerhardt, H. C., and Doherty, J. A. (1988). Acoustic communication in the gray treefrog, Hyla versicolor: Evolutionary and neurobiological implications. J. Comp. Physiol. A. 162, 261– 278. Gerhardt, H. C., and Klump, G. M. (1988a). Masking of acoustic signals by the chorus background noise in the green treefrog: A limitation on mate choice. Anim. Behav. 36, 1247– 1249. Gerhardt, H. C., and Klump, G. M. (1988b). Phonotactic responses and selectivity of barking treefrogs (Hyla gratiosa) to chorus sounds. J. Comp. Physiol. A. 163, 795–802. Gerhardt, H. C., and Mudry, K. M. (1980). Temperature effects on frequency preferences and mating call frequencies in the green treefrog, Hyla cinerea (Anura: Hylidae). J. Comp. Physiol. 137, 1–6. Gerhardt, H. C., and Schul, J. (1999). A quantitative analysis of behavioral selectivity for pulserise time in the gray treefrog, Hyla versicolor. J. Comp. Physiol. A. 185, 33–40. Gerhardt, H. C., and Schwartz, J. J. (1995). Interspecific interactions in anuran courtship. In “Amphibian Biology, Vol. 2: Social Communication” (H. Heatwole and B. K. Sullivan, eds.), pp. 603–632. Surrey, Beatty and Sons, Chipping Norton. Gerhardt, H. C., and Schwartz, J. J. (2001). Auditory tuning and frequency preferences in anurans. In “Anuran Communication” (M. J. Ryan, ed.). Smithsonian, Washington. (In press.) Gerhardt, H. C., and Watson, G. F. (1995). Within-male variability in call properties and female choice in the grey treefrog. Anim. Behav. 50, 1187–1191.

162

H. CARL GERHARDT

Gerhardt, H. C., Guttman, S. I., and Karlin, A. A. (1980). Natural hybrids between Hyla cinerea and Hyla gratiosa: Morphology, vocalization, and electrophoretic analysis. Copeia 1980, 577–584. Gerhardt, H. C., Daniel, R. E., Perrill, S. A., and Schramm, S. (1987). Mating behaviour and male mating success in the green treefrog. Anim. Behav. 35, 1490–1503. Gerhardt, H. C., Allan, S., and Schwartz, J. J. (1990). Female green treefrogs (Hyla cinerea) do not selectively respond to signals with a harmonic structure in noise. J. Comp. Physiol. A 166, 791–794. Gerhardt, H. C., Ptacek, M. B., Barnett, L., and Torke, K. (1994a). Hybridization in the diploidtetraploid treefrogs Hyla chrysoscelis and Hyla versicolor. Copeia 1994: 51–59. Gerhardt, H. C., Dyson, M. L., Tanner, S. D., and Murphy, C. G. (1994b). Female treefrogs do not avoid heterospecific calls as they approach conspecific calls. Anim. Behav. 47, 1323– 1332. Gerhardt, H. C., Dyson, M. L., and Tanner, S. D. (1996). Dynamic acoustic properties of the advertisement calls of gray treefrogs: Patterns of variability and female choice. Behav. Ecol. 7, 7–18. Gerhardt, H. C., Tanner, S. D., Corrigan, C. M., and Walton, H. C. (2000). Female preferences based on call duration in the gray treefrog (Hyla versicolor). Behav. Ecol. 11, 663–669. Grafe, T. U. (1995). Graded aggressive calls in the African painted reed frog Hyperolius marmoratus (Hyperoliidae). Ethology 101, 67–81. Grafe, T. U. (1996). The function of call alternation in the African reed frog (Hyperolius marmoratus): Precise call timing prevents auditory masking. Behav. Ecol. Sociobiol. 38, 149– 158. Grafe, T. U. (1997). Costs and benefits of mate choice in the lek-breeding reed frog, Hyperolius marmoratus. Anim. Behav. 53, 1103–1117. Grafe, T. U. (1999). A function of synchronous chorusing and a novel female preference shift in an anuran. Proc. R. Soc. London B 266, 2331–2336. Grant, P. R. (1972). Convergent and divergent character displacement. Biol. J. Linnean Soc. 4, 39–68. Greenfield, M. D. (1994). Cooperation and conflict in the evolution of signal interactions. Annu. Rev. Ecol. Syst. 24, 97–126. Gwynne, D. T., and Morris, G. K. (1986). Heterospecific recognition and behavioral isolation in acoustic orthoptera. Evol. Theory 8, 33–38. Helversen, O. v., and Helversen, D. v. (1997). Recognition of sex in the acoustic communication system of the grasshopper Chorthippus biguttulus (Orthoptera: Acrididae). J. Comp. Physiol. A. 180, 373–386. Hillery, C. M. (1984). Seasonality of two midbrain auditory responses in the treefrog, Hyla chrysoscelis. Copeia 1984, 844–852. Howard, D. S. (1993). Reinforcement: Origin, dynamics, and fate of an evolutionary hypothesis. In “Hybrid Zones and the Evolutionary Process” (R. G. Harrison, ed.), pp. 46–69. Oxford Univ. Press, UK. Howard, R. D., and Palmer, J. G. (1995). Female choice in Bufo americanus: Effects of dominant frequency and call order. Copeia 1995, 212–217. Jehle, R., and Arak, A. (1998). Graded call variation in the Asian cricket frog Rana nicrobariensis. Bioacoustics 9, 35–48. Jennions, M. D., and Petrie, M. (1997). Variation in mate choice and mating preferences: A review of causes and consequences. Biol. Rev. 72, 283–327. Klump, G. M., and Gerhardt, H. C. (1987). Use of non-arbitrary acoustic criteria in mate choice by female gray treefrogs. Nature (London) 326, 286–288. Klump, G. M., and Gerhardt, H. C. (1992). Mechanisms and function of call-timing in

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

163

male–male interactions in frogs. In “Playback and Studies of Animal Communication” (P. K. McGregor, ed.), pp. 153–174. Plenum, New York, London. Lamb, T., and Avise, J. C. (1986). Directional introgression of mitochondrial DNA in a hybrid population of tree frogs: The influence of mating behavior. Proc. Natl. Acad. Sci. USA 83, 2526–2530. LeVering, K. (1997). Distress calling in frogs. Abstract, 77th Annual Meeting of the American Society of Ichthyologists and Herpetologists. p. 194. Seattle, WA. Lewis, E. R., and Narins, P. R. (1999). The acoustic periphery of amphibians: Anatomy and physiology. In “Comparative Hearing: Fishes and Amphibians” (R. R. Fay and A. N. Popper, eds.), pp. 101–154. Springer, Berlin Heidelberg, New York. Littlejohn, M. J. (1965). Premating isolation in the Hyla ewingi complex (Anura: Hylidae). Evolution 19, 234–243. Littlejohn, M. J. (1981). Reproductive isolation: A critical review. In “Evolution and Speciation” (W. R. Atchley and D. S. Woodruff eds.), pp. 298–334. Cambridge Univ. Press, UK. Littlejohn, M. J., and Harrison, P. A. (1985). The functional significance of the diphasic advertisement call of Geocrinia victoriana (Anura: Leptodactylidae). Behav. Ecol. Sociobiol. 16, 363–373. Littlejohn, M. J., and Loftus-Hills, J. J. (1968). An experimental evaluation of premating isolation in the Hyla ewingi complex (Anura: Hylidae). Evolution 22, 659–663. Littlejohn, M. J., Fouquette, M. J., and Johnson, C. (1960). Call discrimination by female frogs of the Hyla versicolor complex. Copeia 1960, 47–49. Loftus-Hills, J. J., and Littlejohn, M. J. (1992). Reinforcement and reproductive character displacement in Gastrophryne carolinensis and G. olivacea (Anura: Microhylidae): A reexamination. Evolution 46, 896–906. Lombard, R. E., and Straughan, I. R. (1974). Functional aspects of anuran middle ear structures. J. Exp. Biol. 61, 57–71. Mable, B. K., and Bogart, J. P. (1991). Call analysis of triploid hybrids resulting from diploidtetraploid species crosses of hylid treefrogs. Bioacoustics 3, 111–119. Marquez, ´ R., and Bosch, J. (1997a). Female preference in complex acoustical environments in the midwife toads Alytes obstetricans and Alytes cisternasii. Behav. Ecol. 8, 588–594. Marquez, ´ R., and Bosch, J. (1997b). Male advertisement call and female preference in sympatric and allopatric midwife toads. Anim. Behav. 54, 1333–1345. Mecham, J. S. (1960). Introgressive hybridization between two southeastern treefrogs. Evolution 14, 445–457. Mecham, J. S. (1965). Genetic relationships and reproductive isolation southeastern frogs of the genera Pseudacris and Hyla. Am. Midl. Nat. 74, 269–308. Miller, E. H. (1983). Character and variance shift in acoustic signals of birds. In “Acoustic Communication in Birds” (D. E. Kroodsma and E. H. Miller, eds.), pp. 253–295. Academic Press, New York. Minckley, R. L., Greenfield, M. D., and Tourtellot, M. K. (1995). Chorus structure in tarbush grasshoppers: Inhibition, selective phonoresponse, and signal competition. Anim. Behav. 50, 579–594. Morris, G. K., and Fullard, J. H. (1983). Random noise and congeneric discrimination in Conocephalus (Orthoptera: Tettigoniidae). In “Orthopteran Mating Systems: Sexual Competition in a Diverse Group of Insects” (D. T. Gwynne and G. K. Morris, eds.), pp. 73–96. Westview Press, Boulder, CO. Morris, M. R. (1989). Female choice of large males in the treefrog Hyla chrysoscelis: The importance of identifying the scale of choice. Behav. Ecol. Sociobiol. 25, 275–281. Morris, M. R., and Yoon, S. L. (1989). A mechanism for female choice of large males in the treefrog Hyla chrysoscelis. Behav. Ecol. Sociobiol. 25, 65–71.

164

H. CARL GERHARDT

Mudry, K. M., and Capranica, R. R. (1987). Correlation between auditory thalamic area evoked potentials and species-specific call characteristics. II. Hyla cinerea (Anura: Hylidae). J. Comp. Physiol. A. 161, 407–416. Murphy, C. G., and Gerhardt, H. C. (1996). Evaluating experimental designs for determining mate choice: The effect of amplexus on mate choice by barking treefrogs. Anim. Behav. 51, 881–890. Murphy, C. G., and Gerhardt, H. C. (2000). Preference functions of individual female barking treefrogs, Hyla gratiosa. Evolution 54, 660–669. Narins, P. M. (1992). Biological constraints on anuran acoustic communication: Auditory capabilities of naturally behaving animals. In “The Evolutionary Biology of Hearing” (D. B. Webster, R. R. Fay, and A. N. Popper, eds.), pp. 439–454. Springer, Berlin, New York, Heidelberg. Narins, P. M., and Capranica, R. R. (1978). Communicative significance of the two-note call of the treefrog Eleutherodactylus coqui. J. Comp. Physiol. A 127, 1–9. Narins, P. M., Lewis, E. R., and McClelland, B. E. (2000). Hyperextended call note repertoire of the endemic Madagascar treefrog Boophis madagascariensis (Rhacophoridae). J. Zool. (London) 250, 283–298. Nelson, D. A., and Marler, P. (1990). The perception of birdsong and an ecological concept of signal space. In “Comparative Perception, Vol. II: Complex Signals” (W. C. Stebbins and M. A. Berkley, eds.), pp. 443–478. Wiley, New York. Nevo, E., and Capranica, R. R. (1985). Evolutionary origin of ethological isolation in cricket frogs, Acris. Evol. Biol. 19, 147–214. Noble, G. K., and Hassler, W. G. (1936). Three salientia of geographic interest from southern Maryland. Copeia 1936, 63–64. Oldham, R. S., and Gerhardt, H. C. (1975). Behavioral isolation of the treefrogs Hyla cinerea and Hyla gratiosa. Copeia 1975, 223–231. Otte, D. (1989). Speciation in Hawaiian crickets. In “Speciation and Its Consequences” (D. Otte and J. A. Endler, eds.), pp. 482–526. Sinauer, Sunderland. Passmore, N. I., Bishop, P. J., and Caithness, N. (1992). Calling behaviour influences mating success in male painted reed frogs, Hyperolius marmoratus. Ethology 92, 227–241. Paterson, H. E. H. (1985). The recognition concept of species. In “Species and Speciation” (E. S. Vrba, ed.), pp. 21–29. Transvaal Museum Monograph No. 4, Pretoria. Perrill, S. A., Gerhardt, H. C., and Daniel, R. (1978). Sexual parasitism in the green treefrog (Hyla cinerea). Science 200, 1179–1180. Pfenning, K. S. (1998). The evolution of mate choice and the potential for conflict between species and mate-quality recognition. Proc. R. Soc. London B. 265, 1–6. Pfenning, K. S. (2000). Female spadefoot toads compromise on mate quality to ensure conspecific matings. Behav. Ecol. 11, 220–227. Pierce, J. R. (1975). “Isolation and Differentiation in the Canyon Treefrog Hyla arenicolor” Ph.D. Thesis,, University of Texas, Austin. Pierce, J. R., and Ralin, D. B. (1972). Vocalizations and behavior of males of three species in the Hyla versicolor complex. Herpetologica 28, 329–337. Pollack, G. S. (1998). Neural processing of acoustic signals. In “Comparative Hearing: Insects” (R. R. Hoy, A. N. Popper, and R. R. Fay, eds.), pp. 139–196. Springer, New York. Ptacek, M. B. (1992). Calling sites used by male gray treefrogs, Hyla versicolor and H. chrysoscelis, in sympatry and allopatry. Herpetologica 28, 329–337. Ptacek, M. B., Gerhardt, H. C., and Sage, R. D. (1994). Speciation by polyploidy in treefrogs: Multiple origins of the tetraploid, Hyla versicolor. Evolution 48, 898–908. Ralin, D. B. (1970). “Genetic Compatibility and a Phylogeny of the Temperate North American Hylid Fauna.” Ph.D. Dissertation, University of Texas, Austin.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

165

Ralin, D. B. (1977). Evolutionary aspects of mating call variation in a diploid-tetraploid species complex of treefrogs (Anura). Evolution 31, 721–736. Ritchie, M. G., and Phillips, S. D. F. (1998). The genetics of sexual isolation. In “Endless Forms: Species and Speciation” (D. J. Howard and S. H. Berlocher, eds.), pp. 291–308. Oxford Univ. Press, UK. Robertson, J. G. M. (1990). Female choice increases fertilisation success in the Australian frog, Uperoleia laevigata. Anim. Behav. 39, 639–645. Romer, ¨ H., Spickermann, M., and Bailey, W. J. (1998). Sensory bias for sound intensity discrimination in the bushcricket Requena verticalis (Tettigoniidae, Orthoptera). J. Comp. Physiol. A. 182, 595–607. Rose, G. J., Brenowitz, E. A., and Capranica, R. R. (1985). Species specificity and temperature dependency of temporal processing by the auditory midbrain of two species of treefrogs. J. Comp. Physiol. 157, 763–769. Runkle, L. S., Wells, K. D., Robb, C. C., and Lance, S. L. (1994). Individual, nightly, and seasonal variation in calling behavior of the gray treefrog, Hyla versicolor: Implications for energy expenditure. Behav. Ecol. 5, 318–325. Ryan, M. J. (1985). “The Tungara Frog: A Study in Sexual Selection and Communication.” Univ. of Chicago Press Illinois. Ryan, M. J. (1988). Constraints and patterns in the evolution of anuran acoustic communication. In “The Evolution of the Amphibian Auditory System” (B. Fritszch, W. Wilczynski, M. J. Ryan, T. Hetherington, and W. Walkowiak, eds.), pp. 637–677. Wiley, New York. Ryan, M. J. (1990). Sensory systems, sexual selection, and sensory exploitation. Oxford Surveys in Evol. Biol. 7, 157–195. Ryan, M. J., and Keddy-Hector, A. (1992). Directional patterns of female mate choice and the role of sensory biases. Am. Natur. 139, S4–S35. Ryan, M. J., and Rand, A. S. (1993a). Species recognition and sexual selection as a unitary problem in animal communication. Evolution 47, 647–657. Ryan, M. J., and Rand, A. S. (1993b). Sexual selection and signal evolution: The ghost of biases past. Philos. Trans. R. Soc. London B. 340, 187–195. Ryan, M. J., and Sullivan, B. K. (1989). Transmission effects on temporal structure in the advertisement calls of two toads, Bufo woodhousei and Bufo valliceps. Ethology 80, 182– 189. Ryan, M. J., and Wilczynski, W. (1991). Evolution of intraspecific variation in the advertisement call of a cricket frog (Acris crepitans, Hylidae). Biol. J. Linnean Soc. 44, 249–271. Ryan, M. J., Perrill, S. A., and Wilczynski, W. (1992). Auditory tuning and call frequency predict population-based mating preferences in the cricket frog, Acris crepitans. Am. Natur. 139, 1370–1383. Ryan, M. J., Rand, A. S., and Weight, L. A. (1996). Allozyme and advertisement call variation in the tungara frog, Physalaemus pustulosus. Evolution 50, 2435–2453. Schatral, A., and Bailey, W. J. (1991). Decisions during phonotaxis in the bushcricket Requena verticalis (Orthoptera: Tettigoniidae): Do females change direction to alternative male calls? Ethology 88, 320–330. Schlefer, E. K., Romano, M. A., Guttman, S. I., and Ruth, S. B. (1986). Effects of twenty years of hybridization in a disturbed habitat in Hyla cinerea and Hyla gratiosa. J. Herpetol. 20, 210–221. Schmidt, R. S. (1992). Neural correlates of frog calling: Production by two semi-independent generators. Behav. Brain Res. 50, 17–30. Schmitt, A., Friedel, T., and Barth, F. G. (1993). Importance of pause between spider courtship vibrations and general problems using synthetic stimuli in behavioural studies. J. Comp. Physiol. A 172, 707–714.

166

H. CARL GERHARDT

Schul, J. (1998). Song recognition by temporal cues in a group of closely related bushcricket species (genus Tettigonia). J. Comp. Physiol. A 183, 401–410. Schul, J., Helversen, D. v., and Weber, T. (1998). Selective phonotaxis in Tettigonia cantans and T. viridissima in song recognition and discrimination. J. Comp. Physiol. A 182, 687–694. Schwartz, J. J. (1986). Male call behavior and female choice in a neotropical frog. Ethology 73, 116–127. Schwartz, J. J. (1987). The function of call alternation in anuran amphibians: A test of three hypotheses. Evolution 41, 461–471. Schwartz, J. J. (1993). Male calling behavior, female discrimination, and acoustic interference in the neotropical treefrog Hyla microcephala under realistic acoustic conditions. Behav. Ecol. Sociobiol. 32, 401–414. Schwartz, J. J. (1994). Male advertisement and female choice in frogs: Recent findings and new approaches to the study of communication in a dynamic acoustic environment. Am. Zool. 34, 616–624. Schwartz, J. J. (2001). Call monitoring and interactive playback systems in the study of acoustic interactions among male anurans. In “Frogspeak” (M. J. Ryan, ed.), Smithsonian, Washington, DC. Schwartz, J. J., and Gerhardt, H. C. (1989). Spatially mediated release from auditory masking in an anuran amphibian. J. Comp. Physiol. A 166, 37–41. Schwartz, J. J., and Gerhardt, H. C. (1995). Directionality of the auditory system and call pattern recognition during acoustic interference in the gray treefrog Hyla versicolor. Auditory Neurosci. 1, 195–206. Schwartz, J. J., and Wells, K. D. (1985). Intra- and interspecific vocal behavior of the neotropical treefrog Hyla microcephala. Copeia 1985, 27–38. Schwartz, J. J., Buchanan, B., and Gerhardt, H. C. (2001). Female mate choice in the gray treefrog (Hyla versicolor) in three experimental environments. Behav. Ecol. Sociobiol. In press. Shaw, K. L. (1996). Polygenic inheritance of a behavioral phenotype: Interspecific genetics of song in the Hawaiian cricket genus Laupala. Evolution 50, 256–266. Stewart, M. M., and Bishop, P. J. (1994). Effects of increased sound level of advertisement calls on calling male frogs, Eleutherodactylus coqui. J. Herpetol. 28, 46–53. Stumpner, A., and Helversen, O. v. (1994). Song production and song recognition in a group of sibling grasshopper species (Chorthippus dorsatus, Ch. dichrous, and Ch. loratus: Orthoptera: Acrididae). Bioacoustics 6, 1–27. Sullivan, B. K., and Hinshaw, S. H. (1992). Female choice and selection on male calling behaviour in the grey treefrog Hyla versicolor. Anim. Behav. 44, 733–744. Telford, S. R., Dyson, M. L., and Passmore, N. I. (1989). Mate choice only occurs in small choruses of painted reed frogs (Hyperolius marmoratus). Bioacoustics 2, 47–53. Tinbergen, N. (1951). “The Study of Instinct.” Oxford Univ. Press, New York. Waage, J. K. (1979). Reproductive character displacement in Calopteryx (Odonata: Calopterygidae). Evolution 33, 104–116. Wagner, W. E., Jr. (1998). Measuring female mating preferences. Anim. Behav. 55, 1029–1042. Wallach, H., Newman, E. B., and Rosenzweig, M. R. (1949). The precedence effect in sound localization. Am. J. Psych. 62, 315–336. Welch, A. M., Semlitsch, R. D., and Gerhardt, H. C. (1998). Call duration as an indicator of genetic quality in male gray tree frogs. Science 280, 1928–1930. Wells, K. D. (1977). The social behaviour of anuran amphibians. Anim. Behav. 25, 666–693. Wells, K. D. (1988). The effects of social interactions on anuran vocal behavior. In “The Evolution of the Amphibian Auditory System” (B. Fritszch, W. Wilczynski, M. J. Ryan, T. Hetherington, and W. Walkowiak, eds.), pp. 433–454. Wiley, New York.

ACOUSTIC COMMUNICATION IN CLOSELY RELATED TREEFROGS

167

Wells, K. D., and Bard, K. M. (1987). Vocal communication in a neotropical treefrog, Hyla ebraccata: Responses of females to advertisement and aggressive calls. Behaviour 101, 200–210. Wells, K. D., and Schwartz, J. J. (1984a). Vocal communication in a neotropical treefrog, Hyla ebraccata: Aggressive calls. Behaviour 91, 128–145. Wells, K. D., and Schwartz, J. J. (1984b). Vocal communication in a neotropical treefrog, Hyla ebraccata: Advertisement calls. Anim. Behav. 32, 405–420. Wells, K. D., and Taigen, T. L. (1986). The effect of social interactions on calling energetics in the gray treefrog (Hyla versicolor). Behav. Ecol. Sociobiol. 19, 9–18. Wilczynski, W., and Ryan, M. J. (1999). Geographic variation in communication systems. In “Geographic Variation in Behavior: Perspectives of Evolutionary Mechanism” (S. A. Foster and J. A. Endler, eds.), pp. 234–261. Oxford Univ. Press, New York. Wilczynski, W., Rand, A S., and Ryan, M. J. (1995). The processing of spectral cues by the call analysis system of the tungara ´ frog, Physalaemus pustulosus. Anim. Behav. 49, 911–929. Wiley, J. E. (1983). Chromosome polymorphism in Hyla chrysoscelis. Copeia 1983, 273–275. Wiley, R. H. (1994). Errors, exaggeration, and deception in animal communication. In “Behavioral Mechanisms in Evolutionary Ecology” (L. A. Real, ed.), pp. 157–189. Univ. of Chicago Press, Illinois. Wollerman, L. (1998). Stabilizing and directional preferences of female Hyla ebraccata for calls differing in static properties. Anim. Behav. 55, 1619–1630. Wollerman, L. (1999). Acoustic interference limits call detection in a neotropical frog Hyla ebraccata. Anim. Behav. 57, 529–536. Zurek, P. M. (1987). The precedence effect. In “Directional Hearing” (W. A. Yost and G. Gourevitch, eds.), pp. 85–105. Springer, Berlin.

This Page Intentionally Left Blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 30

Scent-Marking by Male Mammals: Cheat-Proof Signals to Competitors and Mates L. M. GOSLING AND S. C. ROBERTS EVOLUTION AND BEHAVIOUR RESEARCH GROUP DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF NEWCASTLE NEWCASTLE UPON TYNE NE1 7RU UNITED KINGDOM

I. INTRODUCTION Uniquely among social signals, scent marks are placed on objects in the environment, often in the absence of any receiver, and may only be detected much later, often in the absence of the signaler. This is curious because social odors in mammals can be transmitted directly from signaler to receiver in a way that is closely analogous to the mode of action in other sensory modalities. Examples include the rapid airborne transmission of volatiles from the dorsal gland of springbok, Antidorcas marsupialis (Bigalke, 1972), or the rump gland of pronghorn antelopes, Antilocapra americana (Moy, 1970), in antipredator contexts. Although animals clearly do not have time to make scent marks when trying to evade predators, examples of this kind show that direct transmission is possible. If so, why is scent-marking, the most ubiquitous form of chemical signaling in the mammals, so curiously indirect? Why is information transmitted in a fashion which, in many ways, appears to be highly inefficient? Signalers are often not present to reinforce their scent signals in the way that is possible for visual or auditory signals and often they cannot know whether a mark will be detected or who the receiver will be. Scent marks may often be degraded before they can be detected, for example by rain (Alberts, 1992). Despite these apparently severe limitations, scent marks are a very common form of signaling by male mammals. They occur in complex patterns, often involving hundreds of marks that are regularly replenished in active marking and remarking sessions. Most male mammals scent-mark and most, perhaps all, resource defense territories are scent-marked. What information is transmitted by patterns of scent marks 169

C 2001 by Academic Press Copyright ° All rights of reproduction in any form reserved. 0065-3454/01 $35.00

170

L. M. GOSLING AND S. C. ROBERTS

and what are the benefits of sending signals in this way that outweigh its inefficiencies and account for its widespread role in mammalian intrasexual competition? Here, we review the development of one idea that addresses this problem, namely that scent-marking by resource holders provides a means of competitor assessment. This idea takes advantage of the theoretical advances made in understanding competitive interactions between individuals, particularly in terms of game theory and competitor assessment (Parker, 1974; Parker and Rubenstein, 1981; Maynard Smith, 1982, 1996). It has been developed explicitly over the past 20 years (Gosling, 1982, 1990; Richardson, 1993) and was implicit in earlier treatments of scent-marking (Uexkull ¨ and Kriszat, 1934; Hediger, 1949; Geist, 1965; Mykytowycz, 1965, 1970, 1973, 1975; Ralls, 1971; Richter, 1972). The empirical basis for this development is a large body of observational and experimental work that links scent-marking and male intrasexual competition, and, in the following section, we review this evidence. The notion that scent marks allow assessment by potential opponents is an interpretation rooted in evolutionary theory rather than a new idea. It also has the advantage of providing a unified theoretical basis for what appears to be a unitary phenomenon. Alternative explanations have been reviewed elsewhere (Gosling, 1982; Brown and Macdonald, 1985) and will not be considered here. The link between scent-marking and territoriality has been helpful in drawing attention to the association between scent-marking and male intrasexual competition, but it is not an obligate one (although the converse, an obligate link between territoriality and scent-marking, may be true). Ralls (1971) noted that scent-marking occurs in a similar form in both territorial and dominance mating systems and argued against a simple link with area defense. Her more general alternative, that in both cases, scent marks function as threats, is reminiscent of the earlier idea that scent marks in territories function as extensions of the owner’s body, even in its absence (Hediger, 1949; Geist, 1965). The problem with such ideas is that many intruders appear to be undeterred by scent marks. They smell the marks but then move on through the territory or stay to use its resources. This difficulty can be removed by using a more fundamental view of a threat, namely that it signals fitness costs to a receiver (Gosling, 1990). Scent marks can thus be regarded as signaling fitness costs to an animal that detects them (in the absence of the signaler), where these costs are a product of the probability that the signaler will return and its relative competitive ability. The receiver’s decision will be influenced by these costs but it should also take into account other factors such as the relative value of the marked resources to both individuals. Given the indirect nature of signaling by scent marks, how do signalers ensure that marks are detected by their competitors? We review the evidence

SCENT-MARKING BY MALE MAMMALS

171

that males place marks to intercept their intended receivers and we also pose the question of whether receivers seek out marks. This is a possibility because receivers may benefit if they can more accurately assess their chances of success in future contests and thus adjust the risks they take. Such decisions could in turn affect the economics of territory defense in a way that depends on the proportion of competitors that opt to avoid contests with the signaler. Next we consider how receivers assess the competitive ability of signalers from scent marks. Reflecting the current literature, we subdivide this treatment into mechanisms appropriate for decisions made without ever meeting the signaler and decisions deferred until after meeting the signaler. In the latter case we again ask what information receivers can get from scent marks that they cannot get directly from the signaler. These questions have implications for general ideas about signal honesty and we review the evidence for the views that scent marks are signals of status or condition-dependent signals. Whereas this review leads us to think that patterns of marks could provide a cheat-proof record of social status, recent information suggests that scent-marking may also have large intrinsic costs and that signal costs may trade-off against life-history traits including growth and reproductive tenure. Paradoxically, scent-marking could also reduce the costs of area defense if many receivers decide to avoid escalated contests (Gosling, 1986a). We discuss the implications of these reduced costs of area defense for the evolution of resource-defense mating systems. Recent developments in social signaling have focused attention on the possibility that individuals may intercept signals that are being sent between two other animals, to their own advantage (McGregor, 1993; Johnstone, 1998). This behavior is called eavesdropping. For example, predators such as frog-eating bats may use the mating calls of frogs to locate their prey (Tuttle and Ryan, 1981). Scent marks are very persistent signals and the interval between placing a mark and its reception is often long. Because of this, scent-marking provides exceptional opportunities for eavesdropping by animals who are not the principal target of the signal. Eavesdropping by predators and parasites may place important constraints on status advertisement by scent-marking. Eavesdropping by male competitors could be a significant selective force in the evolution of scent-marking but it has not been investigated. It could also be argued that eavesdropping by female mammals on scent marks is an important mechanism for assessing the quality of potential mates. Although the strong association between scent-marking and male intrasexual competition suggests that the primary targets for the information carried in scent marks are male competitors, female mammals are the higher investing sex and should be more choosy than males about mate quality (Trivers, 1972). Thus, if scent marks are signals of quality, it might be expected that

172

L. M. GOSLING AND S. C. ROBERTS

females would use this information to assess mate quality, even if they also use odors directly from the body of the signaler at a later stage. The alternative is that scent marks signal directly to females. This is not a trivial distinction because it deals with the extent to which female receiver psychology has shaped the design of scent marks. But, regardless of which of these alternatives is correct, there is a large body of information showing that females respond to the information in scent marks. Females show physiological (priming) responses and behavioral preferences in relation to the familiarity and status of the male odor donor. Most recently, females have been shown to use odor mediated by the major histocompatibility complex of genes (MHC) to choose mates in relation to their genetic relatedness (Yamazaki et al., 1976; Jordan and Bruford, 1998; Penn and Potts, 1999) and to use odor to distinguish healthy and diseased mates (Penn and Potts, 1998). Most of these studies have been on mice and most use male urine or soiled bedding, but how does this relate to scent-marking? The only studies that explicitly use scent marks are those showing that females compare the odor of a potential mate with marks previously found in the environment to select mates (Reece-Engel, 1990; Johnston et al., 1997a,b; Rich and Hurst, 1998, 1999; Ferkin et al., 1999). Do females obtain information about mate quality using such mechanisms and, if so, how do they trade-off quality against information about genetic relatedness or disease status?

II. SCENT-MARKING AND COMPETITION BETWEEN MALES The link between male scent-marking behavior and intrasexual competition is well established and is seen most directly in the common occurrence of marking during contests. The aptly named “stink fights” between groups of sifakas, Propithecus verreauxi, a territorial lemur, are a well-known example (Jolly, 1966). Many of the boundary encounters between male antelopes in neighboring territories consist mainly of ritualized exchanges in which males alternately scent-mark and attempt to remove the marks of their opponent (e.g., wildebeest, Connochaetes taurinus; Estes, 1969). In view of this link it is not surprising that scent-marking and the glands that contribute specialized products to marks are androgen-dependent. This dependence, which has been demonstrated by castration and androgen restoration (Bronson and Whitten, 1968; Mugford and Nowell, 1970; Jones and Nowell, 1973, 1974), extends to the chemical constituents of marks that are known to function in agonistic contexts (Novotny et al., 1984; Harvey et al., 1989). While there is a clear and well-documented association between scentmarking and the defense of territories, Ralls (1971) noted that marking is

SCENT-MARKING BY MALE MAMMALS

173

also common by males in dominance mating systems. She suggested that, in both cases, scent marks may act as a kind of threat. If this is the case, then marks perhaps signal potential fitness costs to opponents in general rather than just to intruders into territories (Gosling, 1990). If the signaler is absent when the receiver detects a mark, these costs will depend on the probability that the signaler will return, its relative competitive ability, and the relative value of any marked resources to the two individuals (see Section IV.A). The involvement of these various factors may explain why the responses of intruders to scent marks is so variable. Some intruders into territories appear to be undeterred by scent marks. They smell the marks but then move on through the territory or stay to use its resources. In other cases, best known from studies of mice, males avoid scent-marked substrates, especially when they are of low competitive ability (Gosling et al., 1996a,b) or when the scent is from dominant males (Jones and Nowell, 1989; Hurst, 1993), and are more reluctant to risk or prolong fights with males whose scent suggests that they are territory owners (Gosling and McKay, 1990; Hurst et al., 1994). Further, there is generally a correlation between marking frequency and social status. Resource holders, both territorial males and dominant males in dominance mating systems, in general mark more than nonresource holders (Miller et al., 1987; Rozenfeld et al., 1987; Hurst, 1990; Allen et al., 1999). This distinction is not always simple, perhaps because there is expected to be greater variance in scent-marking rates (and any form of status signaling) within nonresource holders than within resource holders: the latter have passed through a competitive filter and should be relatively homogeneous. In addition, the scent-marking rates of young males have been found to be the best predictor of intrasexual dominance in later life (Collins et al., 1997). Further evidence for a link between scent-marking and competition is that investment in marking appears to be regulated by the level of threat from potential opponents. Thus, marks in territories are more dense where the threat of intrusion is greatest (e.g., Thomson’s gazelle, Gazella thomsoni: Walther, 1978; klipspringer, Oreotragus oreotragus: Roberts and Lowen, 1997). Intrusion from competitors may sometimes be confounded by visits by potential mates but a recent study of oribi, Ourebia ourebia, avoids this complication. Brashares and Arcese (1999a) found that territorial males marked at common boundaries in relation to the number of male helpers in neighboring territories but not in relation to numbers of females. Investment in marking may also reflect the overall level of competition in a population: in coypus, Myocastor coypus, the mean size of the anal gland used for territorial marking is predicted by estimates of the number of male competitors entering the population but not by that of potential mates (Gosling and Wright, 1984). Similarly, male mice housed adjacent to other males were found to develop

174

L. M. GOSLING AND S. C. ROBERTS

larger preputial glands (which are known to be important in scent-marking and maintaining dominance relationships: Bronson and Marsden, 1973; Yamashita et al., 1989; Novotny et al., 1990; Collins et al., 1997), whereas the glands of those housed next to females became smaller (Hayashi, 1986). Investment in marking also appears to be adjusted in relation to the resourceholding potential (RHP) of individual opponents. Dominant male mice that are smaller than their subordinate partners have higher marking rates and larger preputial glands than dominant males that are larger than their partners, suggesting that a higher rate of scent-marking can compensate for relatively low RHP (Gosling et al., 2000). Lastly, males appear to use scent-marking when trying to take over territories. Male hartebeest that find vacant territories intensively mark the dung piles in the territory, often giving this high priority even when involved in contests with invasion by neighboring males (Gosling, 1974). Male hartebeest also try to take over territories by daily intrusion in a target territory from small temporary territories nearby. During such intrusions they systematically mark at dung piles rather than confronting the resident male. The resident male follows, paws away the new marks, and replaces them with its own before chasing after the intruder. Again, marking appears to receive higher priority than aggression. Over a period of days or weeks the invading male sometimes takes over ownership of the territory having apparently succeeded in a process of attrition. In similar fashion, scent-marking on territories by intruding aardwolves, Proteles cristatus, prior to the mating season, may be a prelude to challenge for ownership or mating opportunities (Richardson, 1987, 1991).

III. HOW DO SIGNALERS ENSURE THAT THEIR SCENT MARKS ARE DETECTED? Most of this review deals with the kinds of information signalers transmit to intended receivers through their scent marks, and on how receivers respond to this information. In this section, however, we outline some of the specific problems associated with the process of signal transmission and detection through the use of scent marks. How do signalers ensure that their scent marks are detected by their intended receivers and how does this elicit an appropriate response? In an influential paper, Guilford and Dawkins (1991) coined the phrase “receiver psychology” to describe ways in which the environment and sensory capabilities of receivers act as important agents of selection on signal design (Endler, 1993; Guilford and Dawkins, 1993; Endler and Basolo, 1998; Rowe, 1999). As this section illustrates, scent marks as signals impose several particular constraints that are quite distinct from those

SCENT-MARKING BY MALE MAMMALS

175

relevant to acoustic or visual signals: signalers that use scent marks must be exceptionally good receiver psychologists. A. TEMPORAL VARIATION IN EFFICACY The crucial parameter of temporal variation for scent-marking signalers is the mark’s persistence time, the interval between deposition and the time when the mark can no longer be detected (Bossert and Wilson, 1963; Alberts, 1992). The most widespread solution to the problem of scent-mark decay is the inclusion of relatively larger molecules than in other types of chemical signal and Alberts (1992) found that territorial scent marks were of larger molecular mass than, for example, alarm signals. In most socially functioning mammalian integumental glands there are both apocrine and sebaceous components (see reviews in Grau, 1976; Adams, 1980; Albone, 1984), the latter mainly contributing large lipids (Dryden and Conaway, 1967; Gorman et al., 1974). Other secretions contain lipocalin protein molecules (Singer and Macrides, 1992). Whereas large molecular mass will in itself result in lower volatility and increased persistence (Bossert and Wilson, 1963), large molecules are thought to act as vehicles for volatile constituents in scent secretion (Bacchini and Gaetani, 1992; Bacchini et al., 1992; Ryg et al., 1992; Robertson et al., 1993) and may act as a controlled release system to regulate their emission (Regnier and Goodwin, 1977; Hurst et al., 1998). Recent evidence has, however, suggested that the major urinary proteins (MUPs) in mouse scent marks are an integral part of the signal itself, stimulating countermarking by receivers, whereas the volatiles they release do not, perhaps serving instead to attract receivers to the marks (Humphries et al., 1999). A small number of studies have estimated persistence times by observing responses of receivers to scent marks of known age. As it is difficult to prove that marks are no longer detectable, these studies use the interval between deposition and the time when the mark fails to elicit a response as a reasonable estimate of effective persistence. Urine of dominant male mice is avoided by subordinates for up to 48 h but its aversive properties had disappeared after 72 h (Jones and Nowell, 1977). However, scent marks appeared to remain active for as long as 7 days in klipspringer antelope (Roberts, 1998), 10 days in dwarf mongooses, Helogale parvula (Rasa, 1973), and 100 days in hamsters, Mesocricetus auratus (Johnston and Schmidt, 1979), whereas odor from anal gland marks of hyenids can be detected even by humans after 1 to 6 months (Gorman and Mills, 1984; Apps et al., 1989). In female house mice, Mus domesticus, two urinary pheromones elicit ultrasonic vocalizations from males, one of which loses efficacy within 18 h, whereas the other lasts up to 30 days (Sipos et al., 1993, 1995). Persistence times of meadow vole, Microtus pennsylvanicus, posterolateral and anogenital scents similarly vary, and

176

L. M. GOSLING AND S. C. ROBERTS

there also appear to be sex differences, with anogenital scent from females and males producing responses up to 10 and 25 days, respectively (Ferkin et al., 1995). The extent to which receivers can accurately discriminate information conveyed by aged, yet detectable, scent marks, remains unclear. Signal degradation appears to at least permit receivers to estimate the age of the mark. Indeed brown hyena, Hyaena brunnea, scent marks may explicitly offer this information to receivers. Hyenas deposit two discrete scent secretions each time they mark: a sebaceous portion and a volatile apocrine one. Mills et al. (1980) interpreted this as a possible bifunctional mechanism, where the more persistent sebaceous portion provides information about identity and status, while the apocrine secretion mainly allows receivers to estimate the age of the mark and, thus, how recently the owner had left. Differential responses to the age of scent marks by receivers have been shown in recent studies to influence the perceived threat of encounter with competitors (Roberts, 1998) and the attractiveness of potential mates to females, where individuals depositing more recent scent were preferred (Ferkin et al., 1995; Rich and Hurst, 1999), perhaps because this provides the most up to date, or least corrupted, information. When receivers approach scent marks, they often sniff and lick the scent deposit (Brown and Macdonald, 1985; Idris, 1994; Roberts, 1998). Alberts (1992) has suggested that by introducing moisture in this way, receivers elicit a release of chemicals from the mark and that a series of investigations may thus cause repeated rise and fade-out cycles from a single mark. B. SPATIAL RANGE The trade-off between odor persistence (which as we have seen results in scent marks of low or controlled volatility) and detectable range (which is directly related to volatility) means that in most scent marks, detection is frequently probabilistic and depends largely on the interaction between movements of intended receivers and the spatial deployment of scent marks. The economic theory of scent-marking (Gosling, 1981, 1986a) was developed in recognition of this fact and the time and energy constraints under which signalers operate. Measuring the distance over which scent marks are detected is problematic, especially outside of the laboratory, as a number of factors will influence detectability (see following section). However Muller¨ Schwarze (1974) estimated a detection distance of about 2–5 m in blacktailed deer, Odocoileus hemionus columbianus, and it seems likely that, at least in larger mammals, distances will be measured in meters, rather than tens of meters. This is in stark contrast to other forms of olfactory signal, notably the insect mate attractant pheromones, which often function over

SCENT-MARKING BY MALE MAMMALS

177

ranges of hundreds of meters (Caro, 1982; Boake et al., 1996; Zhang and Schlyter, 1996). C. SPATIAL DEPLOYMENT AND RECEIVER INTERCEPTION As spatial range of scent marks is typically low, Gosling (1981) stressed the importance of optimal spatial deployment of a limited number of marks across territories, so that they were positioned where they were most likely to intercept intruders (see Fig. 1). Studies that have examined the distribution of scent marks within a territory have usually found that they are either clustered at the territory boundary or toward its center. Perimeter marking has been described in several taxa, including hyenids (Kruuk, 1972; Richardson, 1991), viverrids (Bearder and Randall, 1978), canids (Macdonald, 1979; White et al., 1989; Allen et al., 1999), mustelids (Roper et al., 1986, 1993; Pigozzi, 1990), felids (Smith et al., 1989), rodents (Bel et al., 1995; Boero, 1995; Rosell et al., 1998), and artiodactyls (Gilbert, 1973; Franklin, 1974; Gosling, 1981, 1987; Sun et al., 1994). Dung middens in particular are often placed at the perimeter (Hendrichs and Hendrichs, 1971; Dunbar and Dunbar, 1974; Walther, 1978; Ono et al., 1988; Brashares and Arcese, 1999b). Marking toward the core, or hinterland (Mills et al., 1980), is perhaps less striking and is reported less frequently than perimeter marking. It occurs in preorbital gland marking of small antelopes (Hendrichs and Hendrichs, 1971; Norton, 1980; Ono et al., 1988) and under some circumstances in badgers (Meles meles: Roper et al., 1993), but is best documented in studies of

FIG. 1. Map of the scent marks in the territory of a gerenuk, Litocranius walleri. The oval of marks and radiating arms may be designed to intercept the movements of competitors moving into the territory. The pattern is shown using a nearest-neighbor mapping technique. Redrawn from Gosling (1981).

178

L. M. GOSLING AND S. C. ROBERTS

hyenids (Mills et al., 1980; Mills, 1983; Gorman and Mills, 1984; Mills and Gorman, 1987; Gorman, 1990; Richardson, 1991). The shift from predominantly peripheral to predominantly core marking appears to be dependent on the increasing size of the scent-marked area. For example, hyenas living in large clans with small territories mark along the perimeter, but ecologically stressed populations, living in small clans and very large territories, mark toward the territory center. Scent-mark density is also correlated with smaller territory sizes (Richardson, 1991). These trends occur both within and across species and indicate that the marking strategy is not species-specific but dependent on ecological factors that dictate the optimal distribution of scent (Gorman and Mills, 1984; Gorman, 1990). The trends are also consistent with predictions from scent-marking economics; owners of smaller territories can scent-mark at the boundary because the distances between marks are small and the chance of intruders (or potential mates) missing them as they encroach are small. As territory size expands, intermark distances and thus the number of missed detections also increase and, in consequence, a greater proportion of marks are placed toward the core (Gorman, 1990). An attempt to formalize this trend was made by Roberts and Lowen (1997) using an analytical model that examined a territory owner’s trade-off between minimizing the cost of an intrusion (estimated by calculating the average area available to intruders before detecting a scent mark) and maximizing the probability of mark detection by the intruder. They found a surprisingly robust relationship between the two variables such that intrusion costs were minimized when a ring of scent marks was positioned at approximately 0.8 of the territory radius. This appears to be because, although positioning the ring closer to the territory center would reduce the likelihood of intruders missing a mark as it passed through the ring, the chances of an intruder missing the ring altogether were increased if it traveled at an oblique angle to the center (cf. Gosling, 1981). Roberts and Lowen (1997) suggest that marks surplus to those required for the optimally positioned ring are therefore added toward the periphery in small territories and toward the core with increasing territory size. This kind of approach assumes for simplicity that movements of receivers are probabilistic and random across spatially homogeneous territories. In reality, marks are positioned in ways that are more sensitive to intruder movements. The most notable example is the obvious tendency for marks to be placed along trails and pathways (e.g., Peters and Mech, 1975; Gosling, 1981, 1985; Smith et al., 1989; Sun et al., 1994). The strength of this tendency is likely to be correlated with habitat heterogeneity, particularly in relation to topographical features, which effectively channel animal movements. Nonetheless, the probabilistic approach is useful in understanding the kinds of constraints operating on signalers, particularly if a proportion

SCENT-MARKING BY MALE MAMMALS

179

of intrusions occurs away from trails (e.g., while feeding), where habitats are relatively homogeneous or where a large number of trails intersect the territory; indeed, when trail density is sufficiently high, intruder movements will approximate randomness. In addition to marking along trails, marks are more likely to intercept intruders if they are clustered along particular boundary areas frequented or contested by receivers (e.g., Walther, 1978; Roberts and Lowen, 1997) or if they adjoin territories whose residents pose a greater threat than elsewhere (Brashares and Arcese, 1999a). Such deployment can be flexible to adapt to changing social circumstances, as illustrated by switching of latrine location in response to experimental manipulation of intruder pressure in the solitary and territorial blind mole-rat, Spalax ehrenbergi (Zuri et al., 1997). Lastly, scent marks may be concentrated around especially clumped and contested resources, such as near the burrows of alpine marmots, Marmota marmota (Bel et al., 1995; Boero, 1995) and badgers (Roper et al., 1986). Although it could be argued that these patterns simply reflect where animals spend most of their time, several observations suggest that animals frequently travel to certain areas specifically to scent-mark them before returning or moving elsewhere. Boundary scent-marking patrols are one example (e.g., Gilbert, 1973; Tilson and Tilson, 1986), especially if there is evidence of a recent intrusion from a particular boundary (Sliwa and Richardson, 1998). D. DO RECEIVERS SEEK OUT SCENT MARKS? The economics of scent-marking would be transformed if instead of having to place scent marks where receivers might pass, the receivers could be relied on to seek out the marks. Theoretically, such behavior should only evolve if there is a fitness benefit for receivers that outweighs any costs of searching. A benefit of this kind seems likely because, if marks signal RHP, then receivers should benefit from being able to assess the competitive ability of the signaler and use this information to reduce the costs of contests. Thus, it would be expected that receivers should actively seek out marks. Empirical evidence shows that this is often the case. For example, Muller¨ Schwarze (1974) observed that both captive and wild black-tailed deer search for scent marks after entering a new area. In ring-tailed lemurs, Lemur catta, which were free-ranging within a large enclosure, 62% of scent marks were investigated within 10 min, with a median latency of only 30 s (Kappeler, 1998). If signalers can rely on receivers to seek out marks, then it might be expected that signalers would advertise their presence (Roberts and Gosling, in press). Two sorts of behavior are consistent with this expectation. First, marks

180

L. M. GOSLING AND S. C. ROBERTS

are placed in locally conspicuous places and may be multimodal (Rowe, 1999), with a strong visual component contributing to the detectability of the olfactory signal (Alberts, 1992; Roberts, 1997). For example, klipspringer antelopes prefer to scent-mark on dead trees or branches (mainly of preferred food species) where there are few leaves, most commonly in an area slightly elevated above its surroundings and immediately above a significant break in slope: receivers may thus form a visual search image of likely scent-marked sites (Roberts, 1997). Signalers may also manufacture suitable marking sites where there are none locally available (Gosling, 1972), go to strenuous lengths to place marks as far from the ground as possible to maximize active range (Rasa, 1973; Peters and Mech, 1975; Roberts, 1997), or contribute to signaling sites used by other species (Gosling, 1980; Paquet, 1991), thus gaining in detectability without compromising associated benefits. Second, many signalers actively create visual anomalies, for example by disturbing vegetation by antler thrashing near scent marks, or marking the ground by pawing and scraping with claws or hooves (Gilbert, 1973; Johansson and Liberg, 1996). In some felids and ungulates (Graf, 1956; Barette, 1977; Bowyer et al., 1994; Feldman, 1994), signalers damage or tear off strips of bark before marking, creating visible wounds to trees at scent-marking sites. Although such wounds might, in some cases, serve to prolong mark persistence, they are generally separate from the secretion and thus appear to be unrelated to the olfactory function of the marks. There is no evidence that receivers are attracted to such visual features, but it would be difficult to explain their widespread existence if receivers did not respond to them. These behaviors could increase the detectability of marks, in the first case by providing a conventional site at which a receiver is more likely to find a mark than elsewhere, and in the second by reinforcing the visual element of scent-mark location, further drawing the attention of receivers to the marks. They might also make the mark more memorable by providing a component of the signal in an additional sensory modality. If these visual features function wholly or partly as advertisements, then signalers may be subverting the advantage of assessment to opponents. Because competitors are prepared to incur the costs of seeking out marks, signalers should be able to mark at lower density (Roberts and Gosling, in press). An alternative to the idea that such behavior advertises scent marks is that costly, perhaps condition-dependent signals reinforce the status signal in the scent mark. Behaviors such as pawing or antler thrashing would be more likely to function in this way than placing marks at conspicuous and/or conventional sites. However, there could also be indirect costs of putting marks at conventional sites such as energetic costs of reaching them.

SCENT-MARKING BY MALE MAMMALS

181

IV. HOW DO RECEIVERS USE SCENT MARKS TO ASSESS SIGNALERS? A. BACKGROUND TO THE RECEIVER’S RESPONSE TO SCENT MARKS A unique characteristic of scent marks as a social signal is that receivers often encounter the signal in the absence of the signaler. Because responses to social signals may be influenced by the chance of reinforcement (for example, the chance that a threat will be followed up by attack), the question arises, why do receivers respond to scent marks at all? The answer must be that there is a probability that the signaler will return or that the receiver will encounter it if it proceeds. The difference from a nonolfactory signal is thus one of degree not of kind, although the lack of any immediate reinforcement gives the receiver an option for delaying its response. The second factor that should influence the receiver’s decision is the cost of a contest with the signaler. This will depend principally on relative competitive ability (RHP) and resource value. In general, the RHP of resource holders is greater than that of nonresource holders but the degree of the difference will affect the costs of the contest. For example, some animals of very low quality may opt to avoid all signalers—they would be unlikely to be able to take over ownership of a resource and so any costs of contests would outweigh the benefits. Alternatively, some high quality individuals might be prepared to incur high costs in assessing an opponent because the benefits are potentially high. In general, receivers should withdraw from a scent-marked area when: p(C ∗ RHPS · VS /RHPR · VR ) > (1 − p)VR ,

(1)

where p = probability of meeting the signaler; C = costs of an average contest; RHPS = RHP of signaler; VS = value of the resource to the signaler; RHPR = RHP of receiver; and VR = value of the resource to the receiver. In reality, the value of the resource to the receiver is complex. For example, it is important to distinguish situations where the receiver only intends to use the defended resources in the absence of the signaler (for example, to feed) and those where it intends to try to take over ownership of the resources. The second alternative will usually involve costly escalated contests because resource value is high. Thus, while the receiver’s decision depends on a complex of factors, crucially it needs to assess the competitive ability of the signaler. However, receivers will vary in the accuracy of the information that they need, depending on the likely net benefits or costs. Some decisions can be made in the absence of the signaler and some can be deferred until after meeting it. But how could scent marks provide information that allows receivers to assess signalers in these two contexts? We will review the evidence for the

182

L. M. GOSLING AND S. C. ROBERTS

mechanisms that have been identified and then evaluate the benefits and costs of these mechanisms. B. DECISIONS IN THE ABSENCE OF THE SIGNALER 1. Intrinsic Meaning Scent marks can potentially give information to receivers in the absence of any knowledge about the signalers themselves and without any further encounter with the signaler. The simplest information is that a signaler has been present in the area. But although there is evidence that animals avoid scent-marked areas even when they have had no contact with the signalers, it is difficult to exclude the possibility that they have used additional information. However, it is possible to imagine that where the receiver’s RHP is low and the value of the defended resource is low, then any potential costs will outweigh the benefits and the receiver should move to a different area. Additional characteristics on a scale of increasing complexity of information are those of mark abundance, freshness, and the regularity with which marking sites are remarked. Marks also contain chemical information and here the best data are from mice. The scent marks of dominant male mice have stronger aversive properties than those of subordinate males and castrates (Jones and Nowell, 1973, 1989). A number of androgen-dependent volatile compounds, particularly terpenes, thiazole, and brevicomin, have been identified that appear to be responsible for this response (Novotny et al., 1990b; Bacchini et al., 1992; Robertson et al., 1993; Hurst et al., 1998). There is, thus, some evidence that receivers use intrinsic properties of marks to make decisions about use of space. However, real-world observations show that many, perhaps most, animals that detect marks do not leave the marked area (Gosling, 1982; Simons et al., 1997). Indeed this may often be impossible because marks are so widespread. Intrinsic information in scent marks is not sufficient to make a decision about whether or not to leave a scent-marked area. Receivers also need to compare their own competitive ability with that of the individual who made the scent mark, a process analogous to estimating relatedness through selfreferent phenotype matching (Heth et al., 1998; Mateo and Johnston, 2000a). Experiments on mice provide the best empirical evidence, albeit indirect, for the existence of this mechanism. The finding that subordinate mice are more likely to avoid scent-marked areas than dominant mice (Jones and Nowell, 1989) at first appears to support such a mechanism, but the subjects could have been negatively conditioned to the odors (general, not individual) that were present when they became subordinate. However, on the assumption that body size reflected competitive ability, Gosling et al. (1996a,b) showed that small adult male mice were more likely to avoid scent-marked areas

SCENT-MARKING BY MALE MAMMALS

183

than large ones. The males in these experiments were raised in isolation, suggesting that they may recognize their own competitive ability innately. Extensive studies, conducted mainly on mice, have shown recently that receivers can infer the genetic relatedness of the signaler from odors mediated by the MHC part of the genome (reviewed by Jordan and Bruford, 1998; Penn and Potts, 1999) and also its disease status (reviewed by Penn and Potts, 1998). Such investigations usually focus on a mate choice paradigm but intrinsic information about disease status and relatedness could also be selected for as signals of quality to same-sex competitors. For example, social odors could honestly signal quality if it was too costly for a sick animal to fake a healthy odor. Males should also modify their status signals and thus their competitive behavior in relation to their genetic relatedness (Johnstone, 1998). Information about relatedness and disease status is conveyed by volatile chemicals carried in urine and is thus potentially placed in scent marks. However, no studies have tried to integrate the information available to receivers from patterns of marks and their androgen-dependent volatiles with the chemical information available about disease status (potentially redundant) and that about genetic relatedness (potentially additive). 2. Learned Association Receivers could also assess the potential costs of meeting the signaler from a learned association between the smell of the mark and the memory of previous contests with the individuals that made the marks. The outcome of these contests could allow predictions about the costs of future meetings with the same individual and receivers could opt to leave the marked area when these costs outweigh the benefits of using the marked resource. There are a number of prerequisites for the existence of this mechanism. First, subjects must be able to distinguish between the odors of individual conspecifics. This ability has been demonstrated in numerous species in several taxa (reviewed by Halpin, 1980, 1986, and by Voznessenskaya et al., 1992; see also Johnston et al., 1993; Wilcox and Johnston, 1995). Individual odors may also be recognized across species (e.g., Beauchamp et al., 1985; Johnston and Robinson, 1993; Settle et al., 1994). Individual variation in mouse odors has been linked to genetic variation and in particular to variation in the MHC region of the genome. Mice can be trained to discriminate nearly identical mice that differ only at the MHC, whereas they cannot distinguish between genetically identical mice (Brown et al., 1987, 1990; see also Nevison et al., 2000). The chemicals responsible for these distinctive odors are carried in urine and appear to be a mixture of volatile carboxylic acids, which vary in their relative proportions in each individual (Singer et al., 1997). However, it should be emphasized that demonstrations of sensory capacity cannot be used directly to infer how information about individuality is used. For

184

L. M. GOSLING AND S. C. ROBERTS

example, the capacity to discriminate at the level of the individual is also needed for scent-matching (see Section IV. C. 2) a mechanism that does not require past knowledge of individual behavior. The second prerequisite for individual learning is that receivers can remember the smell of individuals they have encountered in the past. This has been shown in experiments by Roeder (1983), in which two female genets, Genetta genetta, were found to be able to remember scent marks of males for between 9 and 12 weeks. With larger sample sizes, Johnston (1993) has shown that the length of time over which individual flank gland odors can be remembered (assessed using habituation methods) is at least 10 days in hamsters, whereas there are indications of memory for up to 4 weeks in guinea pigs (Beauchamp and Wellington, 1984). It has been shown that Belding’s ground squirrels, Spermophilus beldingi, can even remember familiar versus unfamiliar social odors (from the oral gland) after overwinter hibernation (Mateo and Johnston, 2000b). Subordinate male mice are more likely to avoid the urine odor of a dominant male that defeated them than an unfamiliar odor (Carr et al., 1970). Memorizing individual odors seems most likely, at least in theory, in the case of animals that meet frequently (as with those that live in closed social groups or in neighboring territories). However, there must be limits imposed by the number of potential opponents and the dynamics of changing status within groups. No studies appear to have been carried out on these constraints on adaptive patterns of individual recognition. C. DECISIONS DEFERRED UNTIL AFTER MEETING THE SIGNALER 1. Intimidation A number of authors have suggested that scent marks may alter the psychological state of the signaler and/or receiver such that the confidence of the signaler is enhanced and the receiver is intimidated (Geist, 1965; Mykytowycz, 1965; Richardson, 1993; Sliwa and Richardson, 1998). Although the identification of subjective states cannot be tested directly, they can theoretically be linked to behavioral consequences that can then be tested. However, unfortunately there are no cases where predictions of this hypothesis can be separated from those made from the other candidate mechanisms. Rather it seems to be adopted as an explanation when alternatives have been either excluded or, more commonly, not considered. It may be best to regard ideas about intimidation as suggestions about the mental state of animals that make adaptive decisions based on intrinsic information, learned association, and/or scent-matching, rather than as a separate mechanism.

SCENT-MARKING BY MALE MAMMALS

185

2. Scent-Matching The receiver could memorize the smell of the scent marks as it enters a new area and compare this odor with that of any potential opponent that it meets. When the scents matched, the male would know that the opponent was of high status and, depending on its own relative competitive ability and the value of the marked resource, it could withdraw from the encounter at an appropriate stage (Gosling, 1982). Resource holders would benefit from the reduced costs due to these decisions (see the following). There are a number of prerequisites for the existence of scent-matching and these have interesting parallels to, and crucial distinctions from, some of those outlined above. Scent-matching requires animals to be able to discriminate odors at the level of the individual. It does not require long-term memory of the odor of individuals or a learned association of their smell with past behavior. Scent-matching requires the ability to remember the odor of scent marks that the animal encountered in the recent past and the ability to compare these odors with the smell of potential opponents. The most remarkable demonstration of such sensory capacity comes from tests of the ability of domesticated dogs, Canis familiaris, to remember the human odor associated with forensic objects and to later match these either with additional objects that have been impregnated with these smells (Settle et al., 1994; Schoon, 1997) or with their ultimate source, a crime suspect (Schoon and DeBruin, 1994; Schoon, 1996). Less obviously, scent-matching also requires a sequential and spatial memory that allows the receiver to interpret the pattern of scent marks it encounters as a marked area. For example, the majority of the marks that an animal has encountered in the recent past may be from male A, but the most important ones may be the small number it encountered as it walked from the territory of male A into that of male B. The idea of scent-matching was initially advanced to explain a wide range of observations of scent-marking and linked behavior in wild animals that had previously been difficult to explain (Gosling, 1982). In particular, scentmatching is consistent with the widespread observations that territory owners and other high-status males anoint their own bodies with the substances used for scent-marking and make themselves available for olfactory inspection at the start of many encounters (Fig. 2); many males evert their scent glands as they approach opponents (e.g., hyaenas: Kruuk, 1972). Males also remove marks made by other males and overmark the site with their own odor. It was realized that many early observations that had been interpreted using other mechanisms might be explained more simply by scent-matching. An example is the classic observation that when two male rabbits, Oryctolagus cuniculus, are placed in an arena with the scent marks of one of them, that individual is more likely to win any contest (Mykytowycz, 1973, 1975).

186

L. M. GOSLING AND S. C. ROBERTS

FIG. 2. Territorial male hartebeest, Alcelaphus buselaphus, (a) rubbing substances used in scent-marking onto its side and (b) allowing a nonterritorial intruder to sniff the same area. The intruder may thus identify the owner by matching the smell of the owner with the smell of scent marks previously encountered in the territory. The marking substances are secretions from an antorbital gland and feces. Males often lie in their own feces (“dung piles”) and rub their head and neck on the ground to further transfer the smell to their body (“self-anointing”). Redrawn from Gosling (1982).

The use of scent-matching by intruders into mouse territories has been confirmed experimentally (Gosling and McKay, 1990). Males that approached a resident male fought at lower intensity when they approached on a substrate that matched the resident than when they approached on substrate marked by a third, unknown male. The use of scent-matching has also been demonstrated in the context of mate choice (discussed later) in hamsters and rabbits (Steel, 1984; Reece-Engel, 1988, 1990). One of the predictions of the scent-matching hypothesis is that resource holders should remove (e.g., Sun and Muller-Schwarze, ¨ 1998) or replace scent marks that do not match their own odor (Gosling, 1982). Replacing scent marks is especially common and is known as overmarking or

SCENT-MARKING BY MALE MAMMALS

187

countermarking (Johnston et al., 1994; Wilcox and Johnston, 1995; Roberts, 1998; Sliwa and Richardson, 1998; Ferkin, 1999; Roberts and Dunbar, 2000). Debate has included the idea that marks are placed next to the competitor’s scent mark rather than on top of it in order to signal an asymmetry between the two males (Ferkin, 1999; Rich and Hurst, 1999). The circumstances in which scent-matching is a possible mechanism depends, by definition, on the detection of scent marks prior to an encounter between signaler and receiver (Gosling, 1982). We have recently shown (Roberts and Gosling, unpublished), using an analytical model, that the likelihood of this requirement being fulfilled is crucially dependent on the interaction between the efficacy of the network of scent marks within the defended area and its size. The opportunity for scent-matching generally declines with increasing territory size if mark detection is purely probabilistic. However, in reality, increasing the number, effective range, and detectability of marks (Section III; particularly marking on trails along which intruders enter the territory) will substantially enhance the range of territory sizes over which matching is possible. D. INTRASPECIFIC VARIATION IN MECHANISMS OF ASSESSMENT There has been a tendency in the literature to associate particular mechanisms of assessment using scent marks with particular species. However, it would be surprising if this were the case because selection should favor flexibility in mechanisms where selection pressures, including ecological and social factors, show equal or greater variation within than between species. The available evidence supports this expectation, particularly in the best studied species, the house mouse and its laboratory descendants. Thus, mice are known to respond to intrinsic properties of scent marks and in particular to androgen-dependent chemicals that reflect social dominance. Some mice, but not all, avoid marked areas (Jones and Nowell, 1989; Hurst, 1993). Mice also learn to associate a record of past encounters with their odor and use this information in avoidance decisions (Gosling et al., 1996a,b). However, they also use scent-matching to decide whether or not to escalate contests with novel opponents (Gosling and McKay, 1990; Hurst et al., 1994) and take into account the relative proportions of an opponent’s marks and that of its competitors (competitive countermarking: Rich and Hurst, 1998, 1999). Individual receivers thus appear to use a number of different forms of assessment, the prevalent mechanism probably being determined by the balance of costs and benefits in particular ecological and social contexts. This may mean that particular species tend to use the same mechanism but does not necessarily imply that mechanisms are species-specific.

188

L. M. GOSLING AND S. C. ROBERTS

We have recently explored the circumstances that dictate the switch between mechanisms involved in competitor assessment via scent marks (Roberts and Gosling, unpubl. data), using a game theoretical approach. Game theory seeks to understand how individual behavior varies in relation to the behavior of others (Parker, 1974; Parker and Rubenstein, 1981; Maynard Smith, 1982, 1996); in this case, we were interested in how the strategies of the scent-marking signaler and receiver vary within an owner– intruder paradigm, and how these are dependent on the degree of knowledge about owners that intruders are likely to have gleaned from scent marks before encounters occur. We specifically compared the likely expression of the two most commonly cited mechanisms from the literature, using intrinsic properties of scent marks (cf. Hediger, 1949; Richardson, 1993) and scentmatching (cf. Gosling, 1982). The model examined the circumstances under which it is beneficial for intruders to attempt to assess their opponent further by scent-matching, and those in which owners should either present themselves for scent-matching by intruders or escalate immediately. The stable strategy for intruders (providing benefits of escalation outweigh potential costs of injury) is to attempt to scent-match over most parameter values, especially if there is a moderate degree of owner advantage, but in some circumstances the best option may be immediate escalation or withdrawal (Fig. 3). On the other hand, it pays owners to allow assessment where mark detection probability is high (usually in small territories), but where it is low they do best by escalating immediately. Thus, although intruders do best when they can maximize their information-gathering about owners, the opportunity to match is effectively denied if owners escalate immediately. The model’s predictions are consistent with observed variation in behavior and demonstrate that different mechanisms may result from the balance of costs and benefits along a continuum of information acquisition by intruders. E. THE ACCURACY OF RHP ASSESSMENT USING SCENT-MARKING We have seen that when receivers detect a pattern of marks, they can infer that they are in an area occupied by a male of high competitive ability and, moreover, one that has “owner advantage” (Davies, 1978; Alcock and Bailey, 1997). But how should they use this information? Different receivers must vary in the accuracy of the information that they need to make an adaptive decision. For example, individuals of very low competitive ability simply need to know that an area is defended before deciding to withdraw. They do not need to have detailed information about the competitive ability of the signaler because such information is redundant when all resource holders must be of relatively high RHP. This decision is based on the cost:benefit analysis in Eq. 1 (Section IV.A), and, in particular, the assessment of relative

SCENT-MARKING BY MALE MAMMALS

189

FIG. 3. Variation in evolutionarily stable strategies of intruders. Where both the degree of owner advantage (x; here, x = 0.5 indicates no advantage) and probability of scent mark detection prior to an encounter (m) are low, it pays intruders to escalate contests without assessment (that is, play Hawk, H). In this example, H (shaded black) is the ESS for all m < 3. Successive curves represent the critical value of x, for a given m (here 0.9, 0.7, 0.5, 0.3), above which H is supplanted by the strategy A . Where x is high (above the dashed line), intruders should adopt the strategy A. The strategies A and A are scent-matching assessment strategies defined by playing Dove (D) if the matching process identifies an opponent as the territory owner, and H if not; they differ in their response to immediate escalation by owners (A plays D, A plays H). The results illustrate the potential for flexibility in assessment mechanisms in different socioecological circumstances (Roberts, S. C. and Gosling, L. M., unpublished).

competitive ability. Bearing this procedure in mind, how accurate is the information available in each of the mechanisms of assessment outlined above and under what circumstances would it be useful or not? We know that animals can obtain information about the status of a signaler directly from intrinsic properties of its scent marks. However, intrinsic information may be too general to give precise information to a male about the likely costs of remaining in a scent-marked area. For example, the concentration of androgen-dependent volatiles could indicate that an opponent was in a generally dangerous category, but it could not indicate which one of a range of potential opponents had made the marks. There could thus be advantages in using intrinsic information, but probably only where the differences in RHP between the signaler and receiver are large and therefore easily perceived and assessed. This difference may often be compounded by resource value, particularly where reproductive tenure is dependent on the resources in a territory and where the benefits to an intruder are limited to a short bout of feeding. Where a receiver meets its competitors frequently and remembers their odors, their identities, and the outcome of these encounters (wins:losses),

190

L. M. GOSLING AND S. C. ROBERTS

the detection of scent marks made by these individuals could provide a precise indication of the costs and outcome of future encounters. However, this information should become progressively less accurate as the frequency of such meetings declines and as the rate of status change in the group increases. It will also generally be less effective in large groups than in small groups because of the effect of group size on the first two variables. Where resource holders meet competitors rarely and there is a good chance that their status will have changed in the meantime, memorized information could be actively misleading. This transition may reflect the fundamental transition from an individual reference for dominance in the multimale groups of dominance mating systems to a spatial reference for dominance in resource-defense polygyny and monogamy (Gosling, 1986b). As noted by Ralls (1971), it is particularly interesting that scent-marking is maintained in dominance mating systems when intrinsic information from odor could be directly transmitted and received. Where resource holders contend with large numbers of potential competitors, and where they meet individual competitors infrequently, spatial references for dominance must be unambiguous and verifiable. In these circumstances, verification may only be achieved reliably by comparing the odor of the territory’s scent marks with that of the owner. In many systems, the number of competitors, and hence the need for such a mechanism, are frequently underestimated. Where a network of resource holders defends contiguous territories, as in the aardwolf (Richardson, 1991) or the monogamous antelopes (Brotherton and Manser, 1997; Roberts and Lowen, 1997), the majority of the population appears to be resource holders. However, offspring are continually produced and become transient “floaters”; these are cryptic in their behavior to avoid detection by resource holders and are consequently also rarely detected by observers. Nonetheless, they are a continual threat to resource holders and indeed eventually replace them. V. COSTS AND BENEFITS A. HONESTY AND SIGNAL COSTS Some olfactory information can be simply passed between individuals and does not involve the use of scent marks. For example, the subauricular glands of oribi and rump glands of pronghorn antelopes, Antilocapra americana, are designed for direct, airborne transmission of volatile chemicals to conspecifics. Why are scent marks so commonly used as an intermediate vehicle for chemical signals? The answer may be that patterns of scent marks leave a cheat-proof record of individual behavior. The historical element of this record is impossible to fake because the signaler has to be present over the time that it takes to make the pattern of marks. Because it has to be

SCENT-MARKING BY MALE MAMMALS

191

dominant in space (territoriality) or over competing individuals (in dominance systems) while this is done, the scent marks are cheat-proof. Scent marks are thus a record that is not only of status but that has been probed by competitors and shown to reflect honestly the RHP of the signaler. If intruders or subordinates deposit scent marks the resource holder quickly overmarks them; indeed, overmarking may have higher priority than aggression in such circumstances (e.g., hartebeest: Gosling, 1974; aardwolf: Sliwa and Richardson, 1998). Although it might be expected that scent-marking is a costly trait, there has been little investigation of the costs of this or any other olfactory signal. However, data suggest that scent-marking in mice may have important life-history costs and that these may ultimately affect reproductive success. This has additional significance because scent-marking rates in mice are highly variable between litters (Collins et al., 1997), although it is not yet known whether this variation has a genetic or developmental origin. Data have shown that scent-marking rates are inversely correlated with growth rate and asymptotic body size in male mice housed singly (Fig. 4; Gosling and Roberts, unpubl. data). Although at first sight surprising, major urinary proteins that play a key role in mouse scent-mark function (Section III.A) are produced at urine concentrations of between 10 and 20 mg ml−1 (Nevison et al., 2000) and this rate of protein synthesis could account for the observed growth reductions. In male mice housed in pairs, the dominant individual generally marks at a

FIG. 4. Relationship between scent-marking rates and growth in singly housed male mice. The inverse relationship suggests that scent-marking has an energetic cost and that, as marking levels increase, resources are progressively diverted from growth. There was also an inverse relationship between marking rate and asymptotic weight (mean of weeks 20–25) in this sample of mice (Gosling, L. M. and Roberts, S. C., unpublished data).

192

L. M. GOSLING AND S. C. ROBERTS

higher rate. But, if the dominant male is smaller than its subordinate partner, it scent-marks at a higher rate than if it is larger than its subordinate partner. Smaller dominants also have larger preputial glands, the site of most mouse pheromone production, than larger dominants, even though their body sizes are smaller. As a result of greater costs of scent-marking and associated gland development, poor competitors grow even more slowly and, as a result, experience a higher rate of dominance reversals (Gosling et al., 2000). These findings suggest a life-history trade-off between the costs of scentmarking, to help establish and maintain social status, and the time over which dominance can be maintained. Relatively poor competitors may have to invest more heavily in status signaling to maintain dominance, and consequently they incur relatively high costs. But although small dominant males only maintain their dominance for short periods, this period could provide critical fitness benefits particularly when the probability of survival is low, as in r-selected species including the wild progenitors of the laboratory mouse, or if small males are relatively vulnerable to some other source of mortality such as predation (Dickman et al., 1991; Koivunen et al., 1998). Males that invest heavily in signaling to maintain dominance may thus be opting for a strategy of breeding early. But why do some larger mice appear to accept a subordinate role early in life? The strategic options appear to be principally some form of sneak-breeding or waiting strategies (Kozlowski, 1992; Maynard Smith, 1996; Pilastro et al., 1997; Kokko and Sutherland, 1998) with low associated costs of scent-marking. There is no direct evidence for the former, although subordinate male mice do occasionally obtain some matings (Wolff, 1985; Franks and Lenington, 1986). Life-history benefits from delaying reproduction are consistent with the dominance reversals observed in our experiments, but the consequences for lifetime reproductive success have not yet been tested. These data about the costs of scent-marking in mice are consistent with the theoretical notion that selection should favor the evolution of costly status signals because they provide reliable, cheat-proof information about the quality of the signaler (Zahavi, 1975; Grafen, 1990). B. THE ECONOMICS OF SCENT-MARKING IN TERRITORY DEFENSE Scent marks provide an opportunity for intruders to identify resource holders and then to modify their behavior in a way that takes account of the benefits of utilizing the defended resource and the costs of meeting the resource holder. Because, on average, animals excluded from resource-holding status will have lower RHP than resource holders, many receivers will opt to avoid contests. Others will delay and meet the resource holder to maximize use of the defended resource but, perhaps subject to confirmation of the

SCENT-MARKING BY MALE MAMMALS

193

resource holder’s identity, will withdraw at an early stage of an escalated contest. Resource holders thus avoid or reduce the costs of contests with a large proportion of potential opponents. However, they cannot avoid contests with high-status animals that are seeking to achieve resource-holding status. Scent marks could actually incur additional costs from these contests because they may inadvertently advertise a resource that is worth trying to take over. However, this effect will be offset by owner advantage and, in any case, such individuals are relatively rare. The net costs of these encounters must be low in relation to the costs saved from large numbers of contests with lower status individuals. The extent of the savings can be envisaged as the sum of the difference in the costs of contests resolved by prior assessment and the comparable costs of escalated contests. These two sets of costs can be expressed as two cost functions that increase linearly at different rates with the value of the defended resource (Fig. 5). Comparison with an asymptotic benefit function show that

FIG. 5. The economic consequences of scent-marking in territory defense. The steep cost function (Cesc) shows the cost of defending a territory of increasing size (or value) when competitors are expelled in contests without prior assessment by scent-marking. The shallow cost function (Cassess) shows the cost of defending a territory of increasing size (or value) when competitors assess the owner using its scent marks. Costs are lower because most intruders are of lower RHP than resource holders and so most withdraw without escalation. The broken line shows the increasing fitness benefits of monopolizing an area of increasing size. Intercepts between the benefit and cost functions show the increase in territory area that can be defended with scent-marking. The fitness benefit of scent-marking is the difference between the intercepts of the cost and benefit functions on the ordinate. From Gosling (1986).

194

L. M. GOSLING AND S. C. ROBERTS

the value of the defended resource is higher with scent-marking than with escalated contests. The fact that most, perhaps all, resource-defense territorial systems in mammals are characterized by scent-marking suggests an obligate relationship. Scent-marking occurs in some but not all dominance systems. For example, it is absent in a number of tragelaphine antelopes and in the bovines (Gosling, 1985). Perhaps resource-defense mating systems would not be economically viable without the reduction in costs from prior assessment using scent marks. Evidence to support this comes from a spatially explicit mathematical model of territoriality in wolves, Canis lupus, which suggests that territory formation cannot occur in the absence of scent-marking (Lewis and Murray, 1993; Lewis et al., 1997). In an important empirical demonstration, Stenstrom ¨ (1998) showed that in fallow deer, Dama dama, resource-holding stags scent-marked more when their defended resources were challenged, but that those that scent-marked at high frequencies were subjected to fewer agonistic encounters than those marking at lower rates (Fig. 6). These findings suggest that the evolution of scent-marking may have been necessary before resource-defense systems could replace more primitive systems based on an individual rather than a spatial reference for dominance.

FIG. 6. Relationship between the frequency of scent-marking by fallow deer stags, Dama dama, and the number of agonistic challenges in which they are involved. Stags that scentmark at high rates are involved in fewer encounters. The results lend support to the idea that investment in scent-marking reduces the costs of directly defending territories. Thick lines indicate median values. From Stenstrom ¨ (1998).

SCENT-MARKING BY MALE MAMMALS

195

VI. EAVESDROPPING A. COMPETITORS AND MATES Whenever signals are sent, there is potential for individuals other than the intended receiver to detect and use the information transmitted. This sort of interception is termed eavesdropping (McGregor, 1993). Eavesdropping may have substantial benefits for receivers, and may be cost-neutral, or even disadvantageous, for signalers. For example, it could allow males that observe competitive interactions between other males to assess them as potential opponents at lower cost than by direct assessment. McGregor (1993) was among the first to recognize the potential for eavesdropping, particularly of wide-ranging, conspicuous signals emitted within networks of territorial animals. There is substantial evidence for eavesdropping by receivers of acoustic signals, particularly of territorial birdsong (e.g., Dabelsteen et al., 1997; Naguib et al., 1999; Otter et al., 1999). In contrast, the incidence of eavesdropping in olfactory signals, and scent marks in particular, has rarely been explicitly described. This is surprising, because scent marks are of central importance in mammalian territorial signaling networks (see Section III.C) and their persistence over time offers ample opportunity for many competitors to gather information from them before they become redundant. It may be that difficulties in distinguishing intended receivers and eavesdroppers in scent-marking studies (in view of the often large intervals between signal emission and reception) are responsible for the neglect of this intriguing aspect of scent-marking. If, as seems likely, scent marks are honest signals of competitive ability to male competitors, it would be surprising if females did not eavesdrop on these signals for mate choice. However, males could also signal their quality and genetic relatedness directly to females. This is not a purely semantic debate because the role of females as intended receivers or as eavesdroppers could affect signal design. Because of the possibility that males signal directly to females, we will treat scent-marking and mate choice separately in the following section. Because of the long-lasting nature of scent marks, eavesdroppers also have a unique opportunity to remove, conceal, or amend marks to their own advantage. Scent marks may attract a number of receivers in succession and early receivers may remove the marks, for example by pawing, or add their own marks to amend the information in the mark for subsequent receivers. Thus, in some rodents, males may scent-mark next to, or partially overlapping, the earlier scent mark; this has been termed countermarking. Receivers use the region of overlap and the age difference between the marks to discriminate between the scent marks (Johnston and Bhorade, 1998;

196

L. M. GOSLING AND S. C. ROBERTS

Ferkin et al., 1999; Rich and Hurst, 1999). Alternatively, overmarking of one scent mark by another is often interpreted as an attempt to mask the underlying scent, thus denying later arrivals an opportunity to detect the signal (Woodward et al., 1999). In this way, males might overmark female scent to withhold information about mate receptivity from competitors (Tyler, 1972; Moodie and Byers, 1989; Komers, 1996; Brotherton and Manser, 1997). A number of experiments examining the ability of receivers to discriminate the underlying scent mark indicate that overmarking may be successful in masking scent (Johnston et al., 1994; Woodward et al., 1999). However, this is not always the case, for example in the overmarks of male ring-tailed lemurs, where at least some information in the original scent is conserved after overmarking (Kappeler, 1998). The extent to which overmarking actually masks underlying scent in other taxa, as opposed to being in essence a subcategory of countermarking, remains to be shown (Roberts and Dunbar, 2000). B. PREDATORS, PREY, AND PARASITES Just as vocalizations (Cade, 1975; Tuttle and Ryan, 1981; Belwood and Morris, 1997) and insect airborne pheromones (Stowe et al., 1995; Haynes and Yeargan, 1999) may attract and direct the attention of parasitoids and predators, scent marks also enable such unintended receivers to reliably predict the location and movements of signalers. The best documented example concerns the ability of avian predators to eavesdrop on the scent marks of small rodents, notably voles. Captive kestrels, Falco tinnunculus, are able to detect vole (Microtus, Clethrionomys) scent marks in ultraviolet light, due to the reflectant properties of protein constituents of the marks (Viitala et al., 1995). Field experiments have also demonstrated the ability of the diurnal kestrel (Viitala et al., 1995) and roughlegged buzzard, Buteo lagopus (Koivula and Viitala, 1999; but not nocturnal Tengmalm’s owls, Aegolius funereus: Koivula et al., 1997), to discriminate and preferentially hunt above areas that contained artificially enhanced levels of scent marks. The use of information about scent mark density by raptors is thus an important potential cost of scent marking. Koivula et al. (1999b) have additionally shown that kestrels are even able to discriminate between sex and age of vole prey, apparently on the basis of variable levels of ultraviolet reflectance that exist between classes (Koivula et al., 1999a). This highlights the particular risks to territorial males that may arise from their scent marks. A recent experiment (our unpublished data) has examined the degree to which male mice varying in their signaling investment are prepared to investigate scent marks of an unfamiliar individual in the presence of the odor of a predator (ferret urine). Males that scent-mark at high frequencies are quicker to approach but spend less time investigating the marks, than

SCENT-MARKING BY MALE MAMMALS

197

males signaling at low rates. These results indicate that there are individual differences in sensitivity to potential eavesdropping by predators, which are related to the individual’s own signaling strategy. Whereas mammalian predators, like their avian counterparts, may eavesdrop on scent marks of their prey in order to hunt more effectively, prey animals might use the same strategy to reduce the risk of predation. Whitetailed deer, Odocoileus virginianus, densities are highest in buffer zones between wolf territories (Mech, 1977), the location of which are demarcated by wolf scent-mark patterns (Peters and Mech, 1975), although whether this results from eavesdropping is unknown. The aversive effects of predator scent on their prey are especially well known in rodents and include reduced foraging and travel times (Kotler et al., 1992; Epple et al., 1993; Jedrzejewski et al., 1993; Perrot Sinal et al., 1999) and the induction of avoidance behavior (e.g., Stoddart, 1976; Gorman, 1984; Robinson, 1990; Baretto and Macdonald, 1999). How much increased hunting costs resulting from these responses impinge on the intraspecific benefits of scent-marking by predators is unknown, but they could be considerable where prey density is low. Lastly, a striking example of eavesdropping by parasites is that of a southern African tick and the klipspringer, its small antelope host. The tick, Ixodes matopi, aggregates on the preorbital gland scent marks of klipspringers, which are deposited on the ends of low branches (Roberts, 1997; Roberts and Lowen, 1997), in order to gain access to the host on subsequent visits to the marks (Rechav et al., 1978; Spickett et al., 1980). Adult ticks are predominantly active in the rainy season (Colborne et al., 1981) and locate scent marks by following the trail of an aqueous active component of the secretion as it is washed by rainwater down the branches of the shrub (Rechav et al., 1978). Klipspringers tolerate preorbital gland gleaning by passerine birds in an attempt to reduce their parasite load, but this may also lead to direct ingestion of glandular secretion by the birds, thus introducing an (albeit probably small) additional cost (Roberts, 1995).

VII. SCENT-MARKING AND MATE CHOICE As discussed above, it is not yet known whether males signal directly to females or whether females simply eavesdrop on signals aimed at other males. But, regardless of which of these, or both, occur, there is good evidence that females respond to a wide variety of male odors. In research on mice, many of these odors originate in urine or soiled bedding because these are convenient sources of odor in laboratory experiments. Urine is used for scent-marking and so scent marks could be the normal mode of information transmission in nature. Some evidence does exist for responses to

198

L. M. GOSLING AND S. C. ROBERTS

scent marks and to patterns of marks. Female white-tailed deer investigate and scent-mark at male marking sites (Moore and Marchinton, 1974), and make unusual visits outside their normal range (Sawyer et al., 1989) when they are in estrus, leading Sawyer et al. to suggest that they use marking sites to assess potential mates. In addition, female rabbits, hamsters, and mice prefer males who match the scent marks, or the predominant scent marks, that they had previously been exposed to (Steel, 1984; Reece-Engel, 1988; Rich and Hurst, 1998, 1999). Females thus appear to discriminate between males on the basis of their ability to maintain territory integrity by scentmarking, preferring males whose territories contain only the owner’s marks over those that had been partially marked by intruders (Rich and Hurst, 1998) or the male that most effectively countermarked the scent of intruders onto its territory (Johnston et al., 1997b; Rich and Hurst, 1999; Fig. 7). Females also prefer males with larger scent-marking glands and higher marking rates (Clark et al., 1992), and the scent of their mate to that of unfamiliar individuals (Newman and Halpin, 1988; Tang-Martinez et al., 1993). Female mice also show a wide range of behavioral and physiological responses to intrinsic properties of male urine. Thus female mice prefer the urine of dominant over subordinate males (Parmigiani et al., 1982; Drickamer, 1989; Hayashi, 1990; other rodents: Carr et al., 1982; Evsikov et al., 1995) and intact over castrate urine (Scott and Pfaff, 1970; Hayashi and Kimura, 1978). The chemical basis of this preference is known to include at least four volatile chemicals, a thiazole, a brevicomin, and ␣ and ␤ farnesenes, all of which are attractive to females (Jemiolo et al., 1985, 1991). The thiazole and brevicomin bind to MUPs (Bacchini et al., 1992) and because of this are lasting components of scent marks (Hurst et al., 1998), but this is less certain in the case of the farnesenes (Novotny et al., 1999). Further indications of mate quality are that female mice can distinguish the odor of parasitized and unparasitized males (Kavaliers and Colwell, 1992) and that the urine odor of parasitized males loses its attractiveness (Kavaliers and Colwell, 1995a,b; Penn et al., 1998). Significantly, the odor of infected males does not become aversive to females suggesting that this response is adapted to avoiding mating with infected males rather than avoiding infection. Such data are consistent with the more general suggestion that females prefer to mate with males that have extravagant sexual displays because they are the healthiest and the most resistant to parasites (Hamilton and Zuk, 1982). Perhaps linked to these ideas about the role of immunocompetence in sexual selection is the finding that rats, mice, and humans use odors mediated by variation in the MHC for mate choice. The most extensive data are from inbred mouse strains and, in general, these show that mice choose mates disassortatively with respect to MHC variation (e.g., Yamazaki et al., 1976; Egid and Brown, 1989; Jordan and Bruford, 1998). The best data on female

FIG. 7. Female preferences for male mice, Mus domesticus, which scent-mark near competitor’s marks in response to simulated territorial intrusions. Males that deposit scent marks in response to introduced intruder scent (countermarking males) are preferred over those that fail to do so (countermarked males). (a) Time spent by females in male territories within a 2 h observation period, (b) time spent exploring tubes carrying odors of the territorial males, and (c) number of scent marks deposited by females while exploring the tubes. From Rich and Hurst (1999).

200

L. M. GOSLING AND S. C. ROBERTS

choice are experiments under seminatural conditions that showed that females often mated with males outside their own territory. In these matings, the other males were more MHC-dissimilar than their own territorial males (Potts et al., 1991, 1992). Choice appears to be based on odor and tests involving the responses of trained mice in Y-mazes suggest that variable proportions of volatile carboxylic acids are used to separate nearly identical mice that differ only at the MHC (Singer et al., 1997). MHC-disassortative mating may be a mechanism to increase heterozygosity specifically at the MHC, for example to increase the resistance of their offspring to parasites and diseases, or to prevent inbreeding in general (Jordan and Bruford, 1998; Penn and Potts, 1998). All of the data reviewed in this section could possibly be linked to signals of quality that reflect immune response genotype and immunocompetence. Differences in approach could account for the differences in emphasis on androgen-dependent features such as marking frequencies and MUP-borne volatiles and odors that signal disease resistance. However, it is too early to arrive at any such conclusions. Further work is necessary to integrate the various approaches and to test if females take into account both mate quality and genetic relatedness when choosing mates. All of the male odors dealt with so far lead to behavioral responses by females and in particular preferences for males. However, females in a wide range of rodents also show a number of priming or physiological responses. Best known are the effects of the urine odor of males on advancing female puberty (Vandenburgh, 1971; Bronson and Desjardins, 1974; Colby and Vandenbergh, 1974), induction of estrus (Whitten, 1956; Marsden and Bronson, 1964; Bronson and Whitten, 1968; Chipman and Albrecht, 1974), and, when an unfamiliar male is involved, in inducing abortions (Bruce, 1959; Chipman and Fox, 1966; Dewsbury, 1982; Hafer, 1990). Much is also known about the specific chemical components responsible for these effects (Vandenbergh et al., 1975, 1976; Jemiolo et al., 1986, 1989; Novotny et al., 1990a,b). Nearly all of the research on these phenomena has been carried out in captivity and male urine has been presented in a number of ways (soiled bedding, sprayed onto the female, applied directly to the nares). Because females are the higher investing sex (Trivers, 1972), females are expected to respond selectively to males in relation to their mate quality and genetic relatedness. To our knowledge such intrasexual variation in response has not been measured. Similarly, no studies have measured physiological responses to simulated patterns of scent marks and so the potential influence of any signal of RHP through this medium is unknown. We cannot exclude the possibility that the intrinsic information in rodent urine discussed above is designed to be transmitted directly between individuals or that the chemicals emitted are highly volatile and thus transient

SCENT-MARKING BY MALE MAMMALS

201

components of urine rather than longer-lasting components of scent marks (MUPs and/or the volatiles that bind to them). Self-anointing with scentmarking substances is a common behavior and some glands are designed for direct transmission of odor. Some experiments have shown that the response of females to male odors is greater when the signaler is present than when the odor is presented alone (Milligan, 1975; review in Brown, 1985). However, this could indicate a mechanism that involves information in scent marks with conditional or obligate reinforcement from the signaler (as in scent-matching, Section IV.C). VIII. CONCLUSIONS AND FUTURE DIRECTIONS A. SIGNALING TO COMPETITORS A large body of historical and contemporary data supports the consensus that emerged in the 1980s that scent-marking by male mammals provides a means of competitor assessment. However, relatively little of this information has been collected with this hypothesis in mind (exceptions are reviewed in Gosling and McKay, 1990, and Stenstrom, ¨ 1998) and there is a need for further experimental tests, both in the laboratory and in the field. Patterns of scent marks provide a uniquely spatial element that may simply indicate a spatial reference for dominance or, in game theory terms, an owner’s signal in a bourgeois strategy. However, asymmetries in RHP or resource value are ignored in this conditional strategy and this seems unlikely in the real world. Patterns of scent marks also indicate an animal’s ability to defend an area over the time that it takes to mark. This important information seems unlikely to be ignored, although no experiments have addressed this hypothesis explicitly. The idea that selection should favor signals that honestly reflect the signaler’s quality has received surprisingly little attention in studies of scentmarking (Penn and Potts, 1998). However, recent experimental results suggest that scent-marking is costly both in energetic terms and in risks of attracting predators. Thus scent marks could be condition-dependent signals of quality as well as signals of ownership in a bourgeois strategy. The circumstances under which these signaling strategies operate remain to be clarified. Ideas about signals of quality and the economics of scent-marking also need to be integrated with current findings that information about the signaler’s genotype and disease status are conveyed in social odors. How does intrinsic information about relatedness and disease status modify information from patterns of marks? Most, perhaps all, territories of male mammals are scent-marked (although not all males that scent-mark are territorial). This suggests that

202

L. M. GOSLING AND S. C. ROBERTS

resource-defense polygyny might not be economically feasible without the reduction in costs of contest behavior that occurs when territory owners are assessed using scent marks. This proposition has been tested directly only once (Stenstrom, ¨ 1998) but further experiments in a wider range of species are needed. The general conclusion about reduction in costs would also benefit from comparative analysis. Costs of area defense would be further reduced if intruders seek out marks. The benefits of assessment, the widespread existence of visual advertisements, and the small range of marks (most may need water or water vapor from a receiver before volatiles are released) suggest that they do. Perhaps the practice of placing a signal in the environment in a fashion that does not allow rapid reinforcement might not exist if the benefits to receivers from the information in marks did not exceed the costs of seeking them out. We have reviewed the evidence and theoretical support for the idea that scent marks may be condition-dependent signals of quality. In this they appear to be consistent with the handicap principle, namely that only animals of high quality can afford to use costly signals. We have also outlined the argument that resource holders use scent-marking to manipulate the behavior of nonresource holders to reduce the costs of resource defense. We have even suggested that resource-defense polygyny in mammals might generally not be economically viable without the cost savings that follow from competitor assessment by scent-marking. How can this apparent paradox be resolved? Perhaps only animals that can pay the high costs of scent-marking can also afford the high risk of escalated contests that is involved in acquiring a territory. But having done so, the additional costs of area defense are lower than those of contest behavior in the competing strategy that would involve escalated contests with every competitor. The universal association of resource-defense territoriality and scent-marking suggests that the sum of the costs of scent-marking and a reduced level of escalated contest should be less that those of escalated contests in the absence of assessment by scentmarking. However, this proposition has not been tested empirically, and this remains a priority for understanding the economics of scent-marking in territories. B. MECHANISM Much of the literature on scent-marking has been divided according to the taxon studied and according to favored mechanisms for information transmission. We find no evidence that this reflects biological reality. In the best known species, the mouse, it is known that receivers make decisions on the basis of all three principal mechanisms: the use of intrinsic information (for example mark density and the concentration of androgen-dependent

SCENT-MARKING BY MALE MAMMALS

203

volatiles), the use of memorized information about past opponents and their odors, and the scent-matching of new opponents with the smell of scent marks recently encountered. These mechanisms are also known across a range of species and we believe that when more information is collected, it will prove unusual for animals to be restricted to only one mechanism. Game theoretical analysis has defined the circumstances under which it pays intruders to switch between an assessment of a signaler using intrinsic information and scent-matching. Further investigation is required to explore the importance of the accuracy of assessment that is possible using different mechanisms. For example, it might be expected that learning from past encounters will have limits due to changing social relationships and the number of individuals that can be remembered. Scent-matching is theoretically most accurate and should thus be used where the fitness benefits of a decision outweigh the risks of close approach to a potentially dangerous opponent. Future understanding of alternative mechanisms may depend mainly on advances in receiver psychology, including the limits to spatial and sequential memory and interpretation of patterns of marks. C. SIGNALS TO MATES Although the patterns of scent marks in wild mammals show that they are designed principally as signals to male competitors, selection should also favor choice within the highest investing sex (Trivers, 1972) and so females are expected to use information about male quality in scent marks for mate choice. Empirical studies confirm that females do use information in scent marks although we do not know whether they obtain this information by eavesdropping on signals intended for other males or whether the signals are designed for reception by females. This is a research priority that could be addressed using existing techniques. There is a striking lack of data about the responses of females to male scent marks in natural systems. If the spatial element of patterns of scent marks provides a cheat-proof signal of competitive ability, we would expect females to use these patterns for mate choice. A small body of experimental data suggests that they do and that they use scent-matching between the predominant odor of scent marks and that of potential mates. However, most information about female preferences comes from responses to intrinsic properties of male social odors, a process that need not involve reference to patterns of marks. Some of the information conveyed by qualitative characteristics of marks is potentially redundant with respect to patterns of marks. For example, dominance status and freedom from parasites, both of which can be detected by females using volatiles in mouse urine, may be correlated with an animal’s ability to

204

L. M. GOSLING AND S. C. ROBERTS

establish and maintain patterns of scent marks. We know nothing about the interaction between intrinsic information about quality and that available from patterns of scent marks for mate choice. Information about genetic relatedness from volatiles mediated by MHC variation may provide additional information to that about quality. If females benefit from disassortative mating either to avoid inbreeding depression or because their progeny benefit by increased disease resistance, then they should take account of this information during mate choice. Once again, we do not know how the signals interact or whether their design is influenced by female psychology. Nearly all accounts of physiological responses to male odors, such as puberty advancement or abortion, are treated in the literature as successful manipulations by males of female reproductive physiology. But because females are the higher investing sex, selection should favor their ability to resist manipulation by males and to control their own reproduction. The outcome of such sexual conflict may vary in relation to ecological and frequencydependent factors but little research has been done using this paradigm. For example, is there variation between females in their estrus advancement in relation to male quality or genetic relatedness? If a female aborts and reconceives with a new male, does the female gain because the new male then invests more heavily in the litter? Is a female more likely to abort when there is a chance of mating with an MHC dissimilar male? The questions are legion.

IX. SUMMARY Scent-marking is a ubiquitous form of olfactory signaling in male mammals and both territorial males in resource-defense mating systems and dominant males in dominance mating systems scent-mark. A large body of evidence suggests a link between scent-marking by male mammals and intrasexual competition. Resource holders appear to mark to help establish and maintain their status. They may do this because scent marks allow potential opponents to assess the status or RHP of the signaler. Nonresource holding competitors benefit because they can adjust the level of escalation in relation to potential costs and benefits and avoid risky contests. Resource holders benefit through reduced costs because many nonresource holders withdraw to avoid escalated contests. Three basic mechanisms allow receivers to make decisions after detecting scent marks. Receivers may (1) detect intrinsic properties of scent marks (e.g., concentrations of androgen-dependent volatiles), (2) remember past contests and the odor of each individual involved and associate these with

SCENT-MARKING BY MALE MAMMALS

205

the odor of scent marks, and (3) remember the smell of marks recently encountered and match this smell with potential opponents that they meet subsequently. It is now known that all of these mechanisms are used, sometimes within one species (e.g., mice) and we argue that the mechanisms are used conditionally, depending on information available and potential costs and benefits to receivers. Game theoretical analysis has recently shown how territorial intruders may switch from using intrinsic properties of marks to scent-matching when making decision about whether to remain in a territory. Scent-marking may be a uniquely cheat-proof signal of status because males must be able to defend their territory or dominance status over the time taken to mark it. A pattern of marks is thus a signal of status that has been tested in intrasexual competition. It also seems likely that marks are intrinsically costly both in energetic terms and by increasing predation risk. Mice can detect whether urine is from a parasitized or nonparasitized individual and these odors could potentially signal immunocompetence if mediated by variation at the MHC region of the genome. This remains to be tested. It is known that mice can detect relatedness via urine volatiles mediated by the MHC and it has been predicted that males should modify their competitive behavior in the light of this information. Again this remains to be tested. Information about disease status and genetic relatedness does not explain why males maintain patterns of scent marks. Most, perhaps all, territories are scent-marked. This may be because most intruders are of lower RHP than resource holders and these males should usually withdraw after assessing the resource holder by its scent marks. The costs of defending a territory may thus be substantially reduced. The obligate link between scent-marking and territoriality suggests that resource-defense polygyny in mammals may not be economically viable without this reduction in the costs of area defense. A little information is available to show that females use information from patterns of scent marks and a great deal of information shows that they use intrinsic information. It is not known whether males signal to females to enable mate choice or if females eavesdrop on signals sent between male competitors. Most known responses are to male urine by female rodents. For example, females show physiological (priming) responses to male odors (e.g., advancing and synchronizing estrus, inducing abortion). Other research has identified factors responsible for female mate preferences in choice tests. For example, the dominance status of the signaling male is a predictor of female interest and such studies have identified androgen-dependent volatiles responsible for the response. More recently, females have been shown to use odor mediated by the MHC locus to choose mates in relation to their genetic relatedness and to use odor to distinguish healthy and diseased mates. Most of these studies have been on mice and most use male urine, but the effect

206

L. M. GOSLING AND S. C. ROBERTS

of patterns of urine scent marks has not been investigated. The only studies that explicitly use scent marks are those showing that females match the odor of potential mates with marks previously found in the environment to select mates. Future research should aim to clarify how information about the quality of potential mates is transmitted and how females trade-off such information against genetic relatedness. Acknowledgments We thank Peter Slater and Tim Roper for helpful comments, David Stenstrom ¨ and Tracy Rich for permission to reproduce figures, and Gilbert Roberts for advice about Eq. 1. SCR was supported by a grant from the Leverhulme Trust while this chapter was written.

References Adams, M. G. (1980). Odour-producing organs of mammals. Symp. Zool. Soc. London 45, 57–86. Alberts, A. C. (1992). Constraints on the design of chemical communication systems in terrestrial vertebrates. Am. Nat. 139, (Suppl.), S62–S89. Albone, E. S. (1984). “Mammalian Semiochemistry: The Investigation of Chemical Signals between Mammals.” Wiley, New York. Alcock, J., and Bailey, W. J. (1997). Success in territorial defense by male tarantula hawk wasps Hemipepsis ustulata: The role of residency. Ecol. Entomol. 22, 377–383. Allen, J. J., Bekoff, M., and Crabtree, R. L. (1999). An observational study of coyote (Canis latrans) scent-marking and territoriality in Yellowstone National Park. Ethology 105, 289– 302. Apps, P. J., Viljoen, H. W., Richardson, P. R. K., and Pretorius, V. (1989). Volatile components of anal gland secretion of aardwolf (Proteles cristatus). J. Chem. Ecol. 15, 1681–1688. Bacchini, A., and Gaetani, E. (1992). Mouse urinary proteins bind pheromones. Chem. Senses 17, 818. Bacchini, A., Gaetani, E., and Cavaggioni, A. (1992). Pheromone binding proteins of the mouse, Mus musculus. Experientia 48, 419–421. Barette, C. (1977). Scent-marking in captive muntjacs, Muntiacus reevesi. Anim. Behav. 25, 536–541. Baretto, G. R., and Macdonald, D. W. (1999). The response of water voles, Arvicola terrestris, to the odours of predators. Anim. Behav. 57, 1107–1112. Bearder, S. K., and Randall, R. M. (1978). The use of faecal marking sites by spotted hyaenas and civets. Carnivore 1, 32–38. Beauchamp, G. K., and Wellington, J. L. (1984). Habituation to individual odors occurs in brief, widely-spaced presentations. Physiol. Behav. 32, 511–514. Beauchamp, G. K., Yamazaki, K., Wysocki, C. J., Slotnick, B. M., Thomas, L., and Boyse, E. A. (1985). Chemosensory recognition of mouse major histocompatibility types by another species. Proc. Nat. Acad. Sci. 82, 4186–4188. Bel, M. C., Porteret, C., and Coulon, J. (1995). Scent deposition by cheek rubbing in the alpine marmot (Marmota marmota) in the French Alps. Can. J. Zool. 73, 2065–2071. Belwood, J., and Morris, G. K. (1997). Bat predation and its influence on calling behaviour in Neotropical katydids. Science 238, 64–67.

SCENT-MARKING BY MALE MAMMALS

207

Bigalke, R. C. (1972). Observations on the behaviour and feeding habits of the springbok, Antidorcas marsupialis. Zool. Afr. 7, 333–359. Boake, C. R. B., Shelly, T. E., and Kaneshiro, K. Y. (1996). Sexual selection in relation to pest-management strategies. Ann. Rev. Entomol. 41, 211–229. Boero, D. L. (1995). Scent-deposition behaviour in alpine marmots (Marmota marmota L.): Its role in territorial defence and social communication. Ethology 100, 26–38. Bossert, W. H., and Wilson, E. O. (1963). The analysis of olfactory communication among animals. J. Theor. Biol. 5, 443–469. Bowyer, R. T., Vanbellenberghe, V., and Rock, K. R. (1994). Scent marking by Alaskan moose— characteristics and spatial distribution of rubbed trees. Can. J. Zool. 72, 2186–2192. Brashares, J. S., and Arcese, P. (1999a). Scent marking in a territorial African antelope: I. The maintenance of borders between male oribi. Anim. Behav. 57, 1–10. Brashares, J. S., and Arcese, P. (1999b). Scent marking in a territorial African antelope: II. The economics of marking with faeces. Anim. Behav. 57, 11–17. Bronson, F. H., and Desjardins, C. (1974). Circulating concentrations of FSH, LH, estradiol, and progesterone associated with acute, male-induced puberty in female mice. Endocrinology 94, 1658–1668. Bronson, F. H., and Marsden, H. M. (1973). The preputial gland as an indicator of social dominance in male mice. Behav. Biol. 9, 625–628. Bronson, F. H., and Whitten, W. K. (1968). Estrus-accelerating pheromone of mice: Assay, androgen-dependency and presence in bladder urine. J. Reprod. Fert. 15, 131–134. Brotherton, P. N. M., and Manser, M. B. (1997). Female dispersion and the evolution of monogamy in the dikdik. Anim. Behav. 54, 1413–1424. Brown, R. E. (1985). The rodents: I Effects of odours on reproductive physiology (primer effects). In “Social Odours in Mammals” (R. E. Brown and D. W. Macdonald, eds.), Vol. 1, pp. 245–344. Clarendon Press, Oxford. Brown, R. E., and Macdonald, D. W., eds. (1985). “Social Odours in Mammals.” Clarendon Press, Oxford. Brown, R. E., Singh, P. B., and Roser, B. (1987). The major histocompatibility complex and the chemosensory recognition of individuality in rats. Physiol. Behav. 40, 65–73. Brown, R. E., Roser, B., and Singh, P. B. (1990). The MHC and individual odours in rats. In “Chemical Signals in Vertebrates V” (D. W. Macdonald, D. Muller-Scharze, ¨ S. E. Natynczuk, eds.), pp. 230–243. Plenum Press, New York. Bruce, H. M. (1959). An exteroceptive block to pregnancy in the mouse. Nature 184, 105. Cade, W. H. (1975). Acoustically orienting parasitoids: Fly phonotaxis to cricket song. Science 190, 1312–1313. Caro, J. H. (1982). The sensing, dispersion and measurement of pheromone vapors in air. In “Insect Suppression with Controlled Release Pheromone Systems” (A. F. Kydonieus and M. Beroza, eds.), Vol. I, pp. 145–158. CRC Press, Boca Raton Fl. Carr, W. J., Martorano, R. D., and Krames, L. (1970). Responses of mice to odors associated with stress. J. Comp. Physiol. Psychol. 2, 223–228. Carr, W. J., Kimmel, K. R., Anthony, S. L., and Schlocker, D. E. (1982). Female rats prefer to mate with dominant rather than subordinant males. Bull. Psychonomic Soc. 20, 89–91. Chipman, R. K., and Albrecht, E. D. (1974). The relationship of the male preputial gland to the acceleration of estrus in the laboratory mouse. J. Reprod. Fert. 38, 91–96. Chipman, R. K., and Fox, K. A. (1966). Oestrous synchronization and pregnancy blocking in wild house mice (Mus musculus). J. Reprod. Fert. 12, 233–236. Clark, M. M., Tucker, L., and Galef, B. G., Jr. (1992). Stud males and dud males: Intrauterine position effects on the reproductive success of male gerbils. Anim. Behav. 43, 215– 221.

208

L. M. GOSLING AND S. C. ROBERTS

Colborne, J., Norval, R. A. I., and Spickett, A. M. (1981). Ecological studies on Ixodes (Afrixoides) matopi Spickett, Keirans, Norval and Clifford, 1980 (Acarina: Ixodidae). Onderstepoort J. Vet. Res. 48, 31–35. Colby, D. R., and Vandenbergh, J. G. (1974). Regulatory effects of urinary pheromones on puberty in the mouse. Biol. Reprod. 11, 268–279. Collins, S. A., Gosling, L. M., Hudson, J., and Cowan, D. (1997). Does behaviour after weaning affect the dominance status of adult male mice (Mus musculus)? Behaviour 134, 989–1002. Dabelsteen, T., McGregor, P. K., Holland, J., Tobias, J. A., and Pedersen, S. B. (1997). The signal function of overlapping singing in male robins. Anim. Behav. 53, 249–256. Davies, N. B. (1978). Territorial defence in the speckled wood butterfly (Pararge aegeria): The resident always wins. Anim. Behav. 26, 138–147. Dewsbury, D. A. (1982). Pregnancy blockage following multiple male copulation or exposure at the time of mating in deer mice, Peromyscus maniculatus. Behav. Ecol. Sociobiol. 11, 37–42. Dickman, C. R., Predavec, M., and Lynam, A. J. (1991). Differential predation of size and sex classes of mice by the Barn Owl, Tyto alba. Oikos 62, 67–76. Drickamer, L. C. (1989). Odor preferences of wild stock female house mice (Mus domesticus) tested at three ages using urine and other cues from conspecific males and females. J. Chem. Ecol. 15, 1971–1987. Dryden, G. L., and Conaway, C. H. (1967). The origin and hormonal control of scent production in Suncus murinus. J. Mammal. 48, 420–428. Dunbar, R. I. M., and Dunbar, E. P. (1974). Social organization and ecology of the klipspringer (Oreotragus oreotragus) in Ethiopia. Z. Tierpsychol. 35, 481–493. Egid, K., and Brown, J. L. (1989). The major histocompatibility complex and female mating preferences in mice. Anim. Behav. 38, 548–549. Endler, J. A. (1993). Some general comments on the evolution and design of animal communication systems. Philos. Trans. R. Soc. London B 340, 215–225. Endler, J. A., and Basolo, A. L. (1998). Sensory ecology, receiver biases and sexual selection. Trends. Ecol. Evol. 13, 415–420. Epple, G., Mason, J. R., Nolte, D. L., and Campbell, D. L. (1993). Effects of predator odors on feeding in the mountain beaver (Aplodontia rufa). J. Mammal. 74, 715–722. Estes, R. D. (1969). Territorial behavior of the wildebeest (Connochaetes taurinus Burchell, 1823). Z. Tierpsychol. 26, 284–370. Evsikov, V. I., Nazarova, G. G., and Potapov, M. A. (1995). Female odor choice, male social rank, and sex ratio in the water vole. Adv. Biosciences 93, 303–307. Feldman, H. N. (1994). Methods of scent marking in the domestic cat. Can. J. Zool. 72, 1093– 1099. Ferkin, M. H. (1999). Meadow voles (Microtus pennsylvanicus, Arvicolidae) over-mark and adjacent-mark the scent marks of same-sex conspecifics. Ethology 105, 825–837. Ferkin, M. H., Burda, J., O’Connor, M. P., and Lee, C. J. (1995). Persistence of the attractiveness of two sex-specific scents in meadow voles, Microtus pennsylvanicus. Ethology 101, 228– 238. Ferkin, M. H., Dunsavage, J., and Johnston, R. E. (1999). What kind of information do meadow voles (Microtus pennsylvanicus) use to distinguish between the top and bottom scent of an over-mark? J. Comp. Psychol. 113, 43–51. Franklin, W. L. (1974). The social behaviour of the vicuna. In “The Behaviour of Ungulates and Its Relation to Management” (V. Geist and F. R. Walther, eds.), Vol. 2, pp. 477–487. IUCN, Morges. Franks, P., and Lenington, S. (1986). Dominance and reproductive behavior of wild house mice in a seminatural environment correlated with T-locus genotype. Behav. Ecol. Sociobiol. 18, 395–404.

SCENT-MARKING BY MALE MAMMALS

209

Geist, V. (1965). On the rutting behaviour of the mountain goat. J. Mammal. 45, 551–568. Gilbert, B. K. (1973). Scent marking and territoriality in pronghorn (Antilope americana) in Yellowstone National Park. Mammalia 37, 25–33. Gorman, M. L. (1984). The response of prey to stoat (Mustela erminea) scent. J. Zool. (London) 202, 419–423. Gorman, M. L. (1990). Scent marking strategies in mammals. Rev. Suisse Zool. 97, 3–29. Gorman, M. L., and Mills, M. G. L. (1984). Scent marking strategies in hyaenas (Mammalia). J. Zool. (London) 202, 535–547. Gorman, M. L., Nedwell, D. B., and Smith, R. M. (1974). An analysis of the contents of the anal scent pockets of Herpestes auropunctatus (Carnivora: Viverridae). J. Zool. (London) 172, 389–399. Gosling, L. M. (1972). The construction of antorbital gland marking sites by male oribi (Ourebia ourebia, Zimmerman 1783). Z. Tierpsychol. 30, 271–276. Gosling, L. M. (1974). The social behaviour of Coke’s hartebeest (Alcelaphus buselaphus cokei). In “The Behaviour of Ungulates and Its Relation to Management.” (V. Geist and F. R. Walther, eds.), Vol. 1, pp. 485–511. IUCN, Morges. Gosling, L. M. (1980). Defence guilds of savannah ungulates as a context for scent communication. Symp. Zool. Soc. London 45, 195–212. Gosling, L. M. (1981). Demarkation in a gerenuk territory: An economic approach. Z. Tierpsychol. 56, 305–322. Gosling, L. M. (1982). A reassessment of the function of scent marking in territories. Z. Tierpsychol. 60, 89–118. Gosling, L. M. (1985). The even-toed ungulates: Order Artiodactyla. In “Social Odours in Mammals” (R. E. Brown and D. W. Macdonald, eds.), Vol. 2, pp. 550–618. Oxford University Press, UK. Gosling, L. M. (1986a). Economic consequences of scent marking in mammalian territoriality. In “Chemical Signals in Vertebrates IV” (D. Duvall, D. Muller-Schwarze, ¨ and R. M. Silverstein, eds.), pp. 385–395. Plenum Press, New York. Gosling, L. M. (1986b). The evolution of mating strategies in male antelopes. In “Ecological Aspects of Social Evolution” (D. I. Rubenstein and R. W. Wrangham, eds.), Princeton Univ. Press, New Jersey. Gosling, L. M. (1987). Scent marking in an antelope lek territory. Anim. Behav. 35, 620– 622. Gosling, L. M. (1990). Scent marking by resource holders: Alternative mechanisms for advertising the costs of competition. In “Chemical Signals in Vertebrates V” (D. W., Macdonald, S. Natynczuk, and D. Muller-Schwarze, ¨ eds.), pp. 315–328. Oxford University Press, UK. Gosling, L. M., and McKay, H. V. (1990). Competitor assessment by scent matching: An experimental test. Behav. Ecol. Sociobiol. 26, 415–420. Gosling, L. M., and Wright, K. H. M. (1984). Scent marking and resource defence by male coypus (Myocastor coypus). J. Zool. (London) 234, 423–436. Gosling, L. M., Atkinson, N. W., Collins, S. A., Roberts, R. J., and Walters, R. L. (1996a). Avoidance of scent-marked areas depends on the intruder’s body size. Behaviour 133, 491–502. Gosling, L. M., Atkinson, N. W., Dunn, S., and Collins, S. A. (1996b). The response of subordinate male mice to scent marks varies in relation to their own competitive ability. Anim. Behav. 52, 1185–1191. Gosling, L. M., Roberts, S. C., Thornton, E. A., and Andrew, M. A. (2000). Life history costs of olfactory status signalling in mice. Behav. Ecol. Sociobiol. 48, 328–332. Graf, W. (1956). Territorialism in deer. J. Mammal. 37, 165–170. Grafen, A. (1990). Biological signals as handicaps. J. Theor. Biol. 144, 517–546. Grau, G. A. (1976). Olfaction and reproduction in ungulates. In “Mammalian Olfaction,

210

L. M. GOSLING AND S. C. ROBERTS

Reproductive Processes and Behaviour” (R. L. Doty, ed.), pp. 219–242. Academic Press, New York. Guilford, T., and Dawkins, M. S. (1991). Receiver psychology and the evolution of animal signals. Anim. Behav. 42, 1–14. Guilford, T., and Dawkins, M. S. (1993). Receiver psychology and the design of animal signals. Trends in Neurosci. 16, 430–436. Hafer, A. A. (1990). Dominance status and its influence on the Bruce effect in house mice. In “Chemical Signals in Vertebrates V” (D. W. MacDonald, D. Muller-Scharze, ¨ and S. E. Natynczuk, eds.), pp. 269–275. Plenum Press, New York. Halpin, Z. T. (1980). Individual odors and individual recognition: Review and commentary. Biol. Behav. 5, 233–248. Halpin, Z. T. (1986). Individual odors among mammals: Origins and functions. Adv. Stud. Behav. 16, 39–70. Hamilton, W. D., and Zuk, M. (1982). Heritable true fitness and bright birds: A role for parasites? Science 218, 384–387. Harvey, S., Jemiolo, B., and Novotny, M. (1989). Pattern of volatile compounds in dominant and subordinate male mouse urine. J. Chem. Ecol. 15, 2061–2072. Hayashi, S. (1986). Effects of a cohabitant on preputial gland weight of male mice. Physiol. Behav. 38, 299–300. Hayashi, S. (1990). Social condition influences sexual attractiveness of dominant male mice. Zool. Sci. 7, 889–894. Hayashi, S., and Kimura, T. (1978). Effects of exposure to males on sexual preference in female mice. Anim. Behav. 26, 290–295. Haynes, K. F., and Yeargan, K. V. (1999). Exploitation of intraspecific communication systems: Illicit signalers and receivers. Annals Entomol. Soc. Am. 92, 960–970. Hediger, H. (1949). Saugetier-Territorien ¨ und ihre Markierung. Bijdr. Dierkd. 28, 172–184. ¨ Hendrichs, H., and Hendrichs, U. (1971). “Dikdik und Elefanten: Okologie und Soziologie zweier Afrikanischer Huftiere.” R. Piper, Munich. Heth, G., Todrank, J., and Johnston, R. E. (1998). Kin recognition in golden hamsters: Evidence for phenotype matching. Anim. Behav. 56, 409–417. Humphries, R. E., Robertson, D. H. L., Beynon, R. J., and Hurst, J. L. (1999). Unravelling the chemical basis of competitive scent marking in house mice. Anim. Behav. 58, 1177–1190. Hurst, J. L. (1990). Urine marking in populations of wild house mice Mus domesticus Rutty. I. Communication between males. Anim. Behav. 40, 209–222. Hurst, J. L. (1993). The priming effects of urine substrate marks on interactions between male house mice, Mus musculus domesticus, Schwartz and Schwartz. Anim. Behav. 45, 55–81. Hurst, J. L., Hayden, L., Kingston, M., Luck, R., and Sorensen, K. (1994). Response of the aboriginal house mouse Mus spretus Lataste to tunnels bearing the odours of conspecifics. Anim. Behav. 48, 1219–1229. Hurst, J. L., Robertson, D. H. L., Tolladay, U., and Beynon, R. J. (1998). Proteins in urine scent marks of male house mice extend the longevity of olfactory signals. Anim. Behav. 55, 1289–1297. Idris, M. (1994). Behavioural responses of the Indian desert gerbil, Meriones hurriane, towards conspecific and interspecific sebum odor. Annals of Arid Zone 33, 137–141. Jedrzejewski, W., Rychlik, L., and Jedrzejewska, B. (1993). Responses of bank voles to seven species of predators: Experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251–257. Jemiolo, B., Alberts, J., Sochinski-Wiggins, S., Harvey, S., and Novotny, M. (1985). Behavioural and endocrine responses of female mice to synthetic analogues of volatile compounds in male urine. Anim. Behav. 33, 1114–1118.

SCENT-MARKING BY MALE MAMMALS

211

Jemiolo, B., Harvey, S., and Novotny, M. (1986). Promotion of the Whitten Effect in female mice by synthetic analogs of male urinary constituents. Proc. Nat. Acad. Sci. 83, 4576–4579. Jemiolo, B., Andreolini, F., Xie, T., Wiesler, D., and Novotny, M. (1989). Puberty-affecting synthetic analogs of urinary chemosignals in the house mouse, Mus domesticus. Physiol. Behav. 46, 293–298. Jemiolo, B., Xie, T. M., and Novotny, M. (1991). Socio-sexual olfactory preference in female mice: Attractiveness of synthetic chemosignals. Physiol. Behav. 50, 1119–1122. Johansson, A., and Liberg, O. (1996). Functional aspects of marking behaviour by male roe deer (Capreolus capreolus). J. Mammal. 77, 558–567. Johnston, R. E. (1993). Memory for individual scent in hamsters (Mesocricetus auratus) as assessed by habituation methods. J. Comp. Psychol. 107, 201–207. Johnston, R. E., and Bhorade, A. (1998). Perception of scent over-marks by golden hamsters (Mesocricetus auratus): Novel mechanisms for determining which individual’s mark is on top. J. Comp. Psychol. 112, 230–243. Johnston, R. E., and Robinson, T. A. (1993). Cross-species discrimination of individual odors by hamsters (Muridae, Mesocricetus auratus, Phodopus campbelli). Ethology 94, 317– 325. Johnston, R. E., and Schmidt, T. (1979). Responses of hamsters to scent marks of different ages. Behav. Neural Biol. 26, 64–75. Johnston, R. E., Derzie, A., Chiang, G., Jernigan, P., and Lee, H. C. (1993). Individual scent signatures in golden hamsters: Evidence for specialisation of function. Anim. Behav. 45, 1061–1070. Johnston, R. E., Chiang, G., and Tung, C. (1994). The information in scent over-marks of golden hamsters. Anim. Behav. 48, 323–330. Johnston, R. E., Sorokin, E. S., and Ferkin, M. H. (1997a). Female voles discriminate males’ over-marks and prefer top-scent males. Anim. Behav. 54, 679–690. Johnston, R. E., Sorokin, E. S., and Ferkin, M. H. (1997b). Scent counter-marking by male meadow voles: Females prefer the top-scent male. Ethology 103, 443–453. Johnstone, R. A. (1998). Conspiratorial whispers and conspicuous displays: Games of signal detection. Evolution 52, 1554–1563. Jolly, A. (1966). “Lemur Behaviour: A Madagascar Field Study.” Univ. of Chicago Press, Illinois. Jones, R. B., and Nowell, N. W. (1973). The effect of urine on the investigatory behaviour of male albino mice. Physiol. Behav. 11, 35–38. Jones, R. B., and Nowell, N. W. (1974). Effects of androgen on the aversive properties of male mouse urine. J. Endocrinol. 60, 19–25. Jones, R. B., and Nowell, N. W. (1977). Aversive potency of male mouse urine: A temporal study. Behav. Biol. 19, 523–526. Jones, R. B., and Nowell, N. W. (1989). Aversive potency of urine from dominant and subordinate male laboratory mice (Mus musculus): Resolution of a conflict. Aggressive Behav. 15, 291– 296. Jordan, W. C., and Bruford, M. W. (1998). New perspectives on mate choice and the MHC. Heredity 81, 127–133. Kappeler, P. M. (1998). To whom it may concern: The transmission and function of chemical signals in Lemur catta. Behav. Ecol. Sociobiol. 42, 411–421. Kavaliers, M., and Colwell, D. D. (1992). Aversive responses of female mice to the odors of parasitized males: Neuromodulatory mechanisms and implications for mate choice. Ethology 95, 202–212. Kavaliers, M., and Colwell, D. D. (1995a). Discrimination by female mice between the odors of parasitized males and non-parasitized males. Proc. R. Soc. London B 261, 31–35.

212

L. M. GOSLING AND S. C. ROBERTS

Kavaliers, M., and Colwell, D. D. (1995b). Odors of parasitized males induce aversive responses in female mice. Anim. Behav. 50, 1161–1169. Koivula, M., and Viitala, J. (1999). Rough-legged buzzards use vole scent marks to assess hunting areas. J. Avian Biol. 30, 329–332. Koivula, M., Korpimaki, ¨ E., and Viitala, J. (1997). Do Tengmalm’s owls see vole scent marks visible in ultraviolet light. Anim. Behav. 54, 873–877. Koivula, M., Koskela, E., and Viitala, J. (1999a). Sex and age-specific differences in ultraviolet reflectance of scent marks of bank voles (Clethrionomys glareolus). J. Comp. Physiol. A 185, 561–564. Koivula, M., Viitala, J., and Korpimaki, ¨ E. (1999b). Kestrels prefer scent marks according to species and reproductive status of voles. Ecoscience 6, 415–420. Koivunen, V., Korpimaki, ¨ E., and Hakkarainen, H. (1998). Refuge sites of voles under owl predation risk: Priority of dominant individuals? Behav. Ecol. 9, 261–266. Kokko, H., and Sutherland, W. J. (1998). Optimal floating and queuing strategies: Consequences for density dependence and habitat loss. Am. Nat. 152, 354–366. Komers, P. E. (1996). Obligate monogamy without paternal care in Kirk’s dikdik. Anim. Behav. 51, 131–140. Kotler, B. P., Blaustein, L., and Brown, J. S. (1992). Predator facilitation—The combined effects of snakes and owls on the foraging behaviour of gerbils. Annal. Zool. Fennici 29, 199– 206. Kozlowski, J. (1992). Optimal allocation of resources to growth and reproduction: Implications for age and size at maturity. Trends Ecol. Evol. 7, 15–19. Kruuk, H. (1972). “The Spotted Hyaena, A Study of Predation and Social Behaviour.” Univ. of Chicago Press, Illinois. Lewis, M. A., and Murray, J. D. (1993). Modelling territoriality and wolf-deer interactions. Nature 366, 738–740. Lewis, M. A., White, K. A. J., and Murray, J. D. (1997). Analysis of a model for wolf territories. J. Math. Biol. 35, 749–774. Macdonald, D. W. (1979). Some observations and field experiments on the urine marking behaviour of the red fox, Vulpes vulpes L. Z. Tierpsychol. 51, 1–22. Marsden, H. M., and Bronson, F. H. (1964). Estrous synchrony in mice: Alteration by exposure to male urine. Science 144, 1469. Mateo, J. M., and Johnston, R. E. (2000a). Kin recognition and the “armpit effect”: Evidence of self-referent phenotype matching. Proc. R. Soc. London B 267, 695–700. Mateo, J. M., and Johnston, R. E. (2000b). Retention of social recognition after hibernation in Belding’s ground squirrels. Anim. Behav. 59, 491–499. Maynard Smith, J. (1982). “Evolution and the Theory of Games.” Cambridge Univ. Press, UK. Maynard Smith, J. (1996). The games lizards play. Nature 380, 198–199. McGregor, P. K. (1993). Signalling in territorial systems: A context for individual identification, ranging and eavesdropping. Philos. Trans. R. Soc. London B 340, 237–244. Mech, L. D. (1977). Wolf-pack buffer zones as prey reservoirs. Science 198, 320–321. Miller, K. V., Marchinton, R. L., Forand, K. J., and Johansen, K. L. (1987). Dominance, testosterone levels, and scraping activity in a captive herd of white-tailed deer. J. Mammal. 68, 812–817. Milligan, S. R. (1975). Further observations on the influence of the social environment on ovulation in the vole, Microtus agrestis. J. Reprod. Fert. 44, 543–544. Mills, M. G. L. (1983). Behavioural mechanisms in territory and group maintenance of the brown hyaena Hyaena brunnea in the southern Kalahari. Anim. Behav. 31, 503– 510. Mills, M. G. L., and Gorman, M. L. (1987). The scent marking behaviour of the spotted hyaena (Crocuta crocuta) in the southern Kalahari. J. Zool. (London) 212, 483–497.

SCENT-MARKING BY MALE MAMMALS

213

Mills, M. G. L., Gorman, M. L., and Mills, M. E. J. (1980). The scent marking behaviour of the brown hyaena Hyaena brunnea. S. Afr. J. Zool. 15, 240–248. Moodie, J. D., and Byers, J. A. (1989). The function of scent marking by males on female urine in pronghorns. J. Mammal. 70, 812–814. Moore, W. G., and Marchinton, R. L. (1974). Marking behaviour and its social function in white-tailed deer. In “The Behaviour of Ungulates and Its Relation to Management” (V. Geist and F. Walther, eds.), pp. 447–456. IUCN, Morges. Moy, R. F. (1970). Histology of the subauricular and rump glands of the pronghorn (Antilocapra americana Ord.). Am. J. Anat. 129, 65–88. Mugford, R. A., and Nowell, N. W. (1970). Pheromones and their effect on aggression in mice. Nature 226, 967–968. Muller-Schwarze, ¨ D. M. (1974). Social functions of various scent glands in certain ungulates and the problems encountered in experimental studies of scent communication. In “The Behaviour of Ungulates and Its Relation to Management” (V. Geist F. R. Walther, eds.), pp. 107–113. IUCN, Morges. Mykytowycz, R. (1965). Further observations on the territorial function and histology of the submandibular cutaneous (chin) glands in the rabbit, Oryctolagus cuniculus (L.). Anim. Behav. 13, 400–412. Mykytowycz, R. (1970). The role of skin glands in mammalian communication. In “Advances in Chemoreception. Communication by Chemical Senses” (J. W. Johnston, D. G. Moulton, and A. Turk, eds.), Vol. 1, pp. 327–360. Appleton-Century-Crofts, New York. Mykytowycz, R. (1973). Reproduction of mammals in relation to environmental odours. J. Reprod. Fert. Suppl. 19, 433–446. Mykytowycz, R. (1975). Activation of territorial behaviour in the rabbit, Oryctolagus cuniculus, by stimulation with its own chin gland secretion. In “Olfaction and Taste” (D. A. Denton J. P. Coghlan, eds.), Vol. 5, pp. 425–432. Academic Press, New York. Naguib, M., Fichtel, C., and Todt, D. (1999). Nightingales respond more strongly to vocal leaders of simulated dyadic interactions. Proc. R. Soc. London B 266, 537–542. Nevison, C. M., Barnard, C. J., Beynon, R. J., and Hurst, J. L. (2000). The consequences of inbreeding for recognising competitors. Proc. R. Soc. London B 267, 687–694. Newman, K. S., and Halpin, Z. T. (1988). Individual odours and mate discrimination in the prairie vole (Microtus ochrogaster). Anim. Behav. 36, 1779–1787. Norton, P. M. (1980). The habitat and feeding ecology of the klipspringer Oreotragus oreotragus (Zimmerman 1783) in two areas of the Cape Province. Unpubl. M.S. Thesis. University of Pretoria, South Africa. Novotny, M., Schwende, F. J., Wiesler, D., Jorgeson, J. W., and Carmack, M. (1984). Identification of a testosterone-dependent unique volatile constituent of male mouse urine: 7-exo-ethyl-5-methyl-6, 8-dioxabicyclo[3.2.1]-3-octene. Experientia 40, 217–219. Novotny, M., Harvey, S., and Jemiolo, B. (1990a). Chemistry of male dominance in the house mouse, Mus domesticus. Experientia 46, 109–113. Novotny, M., Jemiolo, B., and Harvey, S. (1990b). Chemistry of rodent pheromones: Molecular insights into chemical signalling in mammals. In “Chemical Signals in Vertebrates V” (D. W. Macdonald, D. Muller-Schwarze, ¨ and S. E. Natynczuk, eds.), pp. 1–22. Oxford University Press, UK. Novotny, M. V., Ma, W., Zidek, L., and Daev, E. (1999). Recent biochemical insights into puberty acceleration, estrus induction and puberty delay in the house mouse. In “Advances in Chemical Communication in Vertebrates” (R. E. Johnston, D. Muller-Schwarze, ¨ and P. Sorenson, eds.), pp. 99–116. Plenum, New York. Ono, Y., Doi, T., Ikeda, H., Baba, M., Takeishi, M., Izawa, M., and Iwamoto, T. (1988). Territoriality of Guenther’s dikdik in the Omo National Park, Ethiopia. Afr. J. Ecol. 26, 33–49.

214

L. M. GOSLING AND S. C. ROBERTS

Otter, K., McGregor, P. K., Terry, A. M. R., Burford, F. R. L., and Peake, T. M. (1999). Do female great tits (Parus major) assess males by eavesdropping? A field study using interactive song playback. Proc. R. Soc. London B 266, 1305–1309. Paquet, P. C. (1991). Scent marking behaviour of sympatric wolves (Canis lupus) and coyotes (C. latrans) in Riding Mountain National Park. Can. J. Zool. 69, 1721–1727. Parker, G. A. (1974). Assessment strategy and the evolution of fighting behaviour. J. Theor. Biol. 47, 223–243. Parker, G. A., and Rubenstein, D. I. (1981). Role assessment, reserve strategy, and acquisition of information in asymmetric animal conflicts. Anim. Behav. 29, 221–240. Parmigiani, S., Brunoni, V., and Pasquali, A. (1982). Behavioural influences of dominant, isolated and subordinated male mice on female socio-sexual preferences. Bollettino di Zoologia 49, 31–35. Penn, D., and Potts, W. K. (1998). Chemical signals and parasite-mediated sexual selection. Trends Ecol. Evol. 13, 391–396. Penn, D. J., and Potts, W. K. (1999). The evolution of mating preferences and major histocompatibility complex genes. Am. Nat. 153, 145–164. Penn, D., Schneider, G., White, K., Slev, P., and Potts, W. (1998). Influenza infection neutralizes the attractiveness of male odour to female mice (Mus musculus). Ethology 104, 685– 694. Perrot Sinal, T. S., Ossenkopp, K. P., and Kavaliers, M. (1999). Effects of repeated exposure to fox odor on locomotor activity levels and spatial movement patterns in breeding male and female meadow voles (Microtus pennsylvanicus). J. Chem. Ecol. 25, 1567–1584. Peters, R. P., and Mech, L. D. (1975). Scent-marking in wolves. Am. Sci. 63, 628–637. Pigozzi, G. (1990). Latrine use and the function of territoriality in the European badger, Meles meles, in a Mediterranean coastal habitat. Anim. Behav. 39, 1000–1002. Pilastro, A., Giacomello, E., and Bisazza, A. (1997). Sexual selection for small size in male mosquitofish (Gambusia holbrooki). Proc. R. Soc. London B 264, 1125–1129. Potts, W. K., Manning, C. J., and Wakeland, E. K. (1991). Mating patterns in seminatural populations of mice influenced by MHC genotype. Nature 352, 619–621. Potts, W. K., Manning, C. J., and Wakeland, E. K. (1992). MHC-based mating preferences in Mus operate through both settlement patterns and female controlled extra-territorial matings. In “Chemical Signals in Vertebrates VI” (R. L. Doty and D. Muller-Schwarze, ¨ eds.), pp. 229–235. Plenum, New York. Ralls, K. (1971). Mammalian scent marking. Science 171, 443–449. Rasa, O. A. E. (1973). Marking behaviour and its social significance in the African dwarf mongoose, Helogale undulata rufula. Z. Tierpsychol. 32, 293–318. Rechav, Y., Norval, R. A. I., Tannock, J., and Colborne, J. (1978). Attraction of the tick Ixodes neitzi to twigs marked by the klipspringer antelope. Nature 275, 310–311. Reece-Engel, C. (1988). Female choice of resident male rabbits Oryctolagus cuniculus. Anim. Behav. 36, 1241–1242. Reece-Engel, C. (1990). Scent marking, residency, and female choice in the European rabbit (Oryctolagus cuniculus). In “Chemical Signals in Vertebrates V” (D. W. MacDonald, D. Muller-Scharze, ¨ and S. E. Natynczuk, eds.), pp. 329–335. Plenum Press, New York. Regnier, F. E., and Goodwin, M. (1977). On the chemical and environmental modulation of pheromone release from vertebrate scent marks. In “Chemical Signals in Vertebrates I” (D. Muller-Schwarze ¨ and M. M. Mozell, eds.), pp. 115–133. Plenum Press, New York. Rich, T. J., and Hurst, J. L. (1998). Scent marks as reliable signals of the competitive ability of mates. Anim. Behav. 56, 727–735. Rich, T. J., and Hurst, J. L. (1999). The competing countermarks hypothesis: Reliable assessment of competitive ability by potential mates. Anim. Behav. 58, 1027–1037.

SCENT-MARKING BY MALE MAMMALS

215

Richardson, P. R. K. (1987). Aardwolf mating system: Overt cuckoldry in an apparently monogamous mammal. S. Afr. J. Sci. 83, 405–410. Richardson, P. R. K. (1991). Territorial significance of scent marking during the nonmating season in the aardwolf Proteles cristatus (Carnivora: Protelidae). Ethology 87, 9–27. Richardson, P. R. K. (1993). The function of scent marking in territories: A resurrection of the intimidation hypothesis. Trans. R. Soc. S. Afr. 48, 195–206. Richter, W. von (1972). Territorial behaviour of the black wildebeest. Zool. Africana 7, 207– 231. Roberts, S. C. (1995). Gleaning in klipspringer preorbital glands by Redwinged Starlings and Yellowbellied Bulbuls. Ostrich 66, 147–148. Roberts, S. C. (1997). Selection of scent-marking sites by klipspringers (Oreotragus oreotragus). J. Zool. (London) 243, 555–564. Roberts, S. C. (1998). Behavioural responses to scent marks of increasing age in klipspringer Oreotagus oreotragus. Ethology 104, 585–592. Roberts, S. C., and Dunbar, R. I. M. (2000). Female territoriality and the function of scentmarking in a monogamous antelope (Oreotragus oreotragus). Behav. Ecol. Sociobiol. 47, 417–423. Roberts, S. C., and Gosling, L. M. The economic consequences of advertising scent mark location in territories. In “Chemical Signals in Vertebrates IX” (A. Marchlewska-Koy, ed.). Plenum Press. In press. Roberts, S. C., and Lowen, C. (1997). Optimal patterns of scent marks in klipspringer (Oreotragus oreotragus) territories. J. Zool. (London) 243, 565–578. Robertson, D. H. L., Beynon, R. J., and Evershed, R. P. (1993). Extraction, characterisation, and binding analysis of two pheromonally active ligands associated with major urinary protein of house mouse (Mus musculus). J. Chem. Ecol. 19, 1405–1416. Robinson, I. (1990). The effect of mink odour on rabbits and small mammals. In “Chemical Signals in Vertebrates V” (D. W. Macdonald, D. Muller-Schwarze, ¨ and S. E. Natynczuk, eds.), pp. 566–572. Oxford Univ. Press, New York. Roeder, J. J. (1983). Memorisation des marques olfactives chez la genette (Genetta genetta L.): Duree de reconaissance par les femelles de marques olfactives de males. Z. Tierpsychol. 61, 311–314. Roper, T. J., Shepherdson, D. J., and Davies, J. M. (1986). Scent-marking with faeces and anal secretion in the European badger (Meles meles): Seasonal and spatial characteristics of latrine use in relation to territoriality. Behaviour 97, 94–117. Roper, T. J., Contradt, L., Butler, J., Christian, S. E., Ostler, J., and Schmid, T. K. (1993). Territorial marking with faeces in badgers (Meles meles): A comparison of boundary and hinterland latrine use. Behaviour 127, 289–307. Rosell, F., Bergan, P., and Parker, H. (1998). Scent-marking in the Eurasian beaver (Castor fiber) as a means of territory defense. J. Chem. Ecol. 24, 207–219. Rowe, C. (1999). Receiver psychology and the evolution of multicomponent signals. Anim. Behav. 58, 921–931. Rozenfeld, F. M., Le Boulange, ´ E., and Rasmont, R. (1987). Urine marking in male bank voles (Clethrionomys glareolus Schreber, 1780; Microtidae, Rodentia) in relation to their social rank. Can. J. Zool. 65, 2594–2601. Ryg, M., Solberg, Y., Lydersen, C., and Smith, T. G. (1992). The scent of rutting male ringed seals (Phoca hispida). J. Zool. (London) 226, 681–689. Sawyer, T. G., Marchinton, R. L., and Miller, K. V. (1989). Response of female white-tailed deer to scrapes and antler rubs. J. Mammal. 70, 431–433. Schoon, G. A. A. (1996). Scent identification lineups by dogs (Canis familiaris): Experimental design and forensic application. Appl. Anim. Behav. Sci. 49, 257–267.

216

L. M. GOSLING AND S. C. ROBERTS

Schoon, G. A. A. (1997). Scent identifications by dogs (Canis familiaris): A new experimental design. Behaviour 134, 531–550. Schoon, G. A. A., and DeBruin, J. C. (1994). The ability of dogs to recognise and cross-match human odors. Forensic Sci. Intl. 69, 111–118. Scott, J. W., and Pfaff, D. W. (1970). Behavioral and electrophysiological responses of female mice to male urine odors. Physiol. Behav. 5, 407–411. Settle, R. H., Sommerville, B. A., McCormick, J., and Broom, D. M. (1994). Human scent matching using specially trained dogs. Anim. Behav. 48, 1443–1448. Simons, R. R., Jaeger, R. G., and Felgenhauer, B. E. (1997). Competitor assessment and area defense by territorial salamanders. Copeia 1, 70–76. Singer, A. G., and Macrides, F. (1992). Lipocalins associated with mammalian pheromones. In “Chemical Signals in Vertebrates VI” (R. L. Doty and D. Muller-Schwarze, ¨ eds.), pp. 119–124. Plenum Press, New York. Singer, A. G., Beauchamp, G. K., and Yamazaki, K. (1997). Volatile signals of the major histocompatibility complex in male mouse urine. Proc. Nat. Acad. Sci. 94, 2210–2214. Sipos, M. L., Nyby, J. G., and Serran, M. F. (1993). An ephemeral sex pheromone of female house mice (Mus domesticus): Pheromone fade-out time. Physiol. Behav. 54, 171–174. Sipos, M. L., Alterman, L., Perry, B., Nyby, J. G., and Vandenburgh, J. G. (1995). An ephemeral pheromone of female house mice—degradation by oxidation. Anim. Behav. 50, 113–120. Sliwa, A., and Richardson, P. R. K. (1998). Responses of aardwolves, Proteles cristatus, Sparrman 1783, to translocated scent marks. Anim. Behav. 56, 137–146. Smith, J. L. D., McDougal, C., and Miquelle, D. (1989). Scent marking in free-ranging tigers, Panthera tigris. Anim. Behav. 37, 1–10. Spickett, A. M., Keirans, J. E., Norval, R. A. I., and Clifford, C. M. (1980). Ixodes (Afrixoides) matopi n.sp. (Acarina:Ixodidae): A tick found aggregating on preorbital gland scent marks of the klipspringer in Zimbabwe. Onderstepoort J. Vet. Res. 48. Steel, E. (1984). Effect of the odour of vaginal secretion on non-copulatory behaviour of male hamsters (Mesocricetus auratus). Anim. Behav. 32, 597–608. Stenstrom, ¨ D. (1998). Mating behaviour and sexual selection in non-lekking fallow deer (Dama dama). Ph.D. Thesis. University of Uppsala, Sweden. Stoddart, D. M. (1976). Effect of the odour of weasels (Mustela nivalis L.) on trapped samples of their prey. Oecologia 22, 439–441. Stowe, M. K., Turlings, T. C. J., Loughrin, J. H., Lewis, W. J., and Tumlinson, J. H. (1995). The chemistry of eavesdropping, alarm and deceit. Proc. Nat. Acad. Sci. 92, 23–28. Sun, L., and Muller-Schwarze, ¨ D. (1998). Beaver response to recurrent alien scents: Scent fence or scent match? Anim. Behav. 55, 1529–1536. Sun, L., Xiao, B., and Dai, N. (1994). Scent marking behaviour in the male Chinese water deer. Acta Theriol. 39, 177–184. Tang-Martinez, Z., Mueller, L. L., and Taylor, G. T. (1993). Individual odours and mating success in the golden hamster, Mesocricetus auratus. Anim. Behav. 45, 1141–1151. Tilson, R. L., and Tilson, J. W. (1986). Population turnover in a monogamous antelope (Madoqua kirki) in Namibia. J. Mammal. 67, 610–613. Trivers, R. L. (1972). Parental investment and sexual selection. In “Sexual Selection and the Descent of Man, 1871–1971” (B. Campbell, ed.), pp. 136–179. Heinemann, London. Tuttle, M. D., and Ryan, M. J. (1981). Bat predation and the evolution of frog vocalisations in the neotropics. Science 214, 677–678. Tyler, S. (1972). The behaviour and social organisation of the New Forest Ponies. Anim. Behav. Monogr 5, 85–196. Uexkull, ¨ J. V., and Kriszat, G. (1934). Striefzuge ¨ durch die Umwelten von Tieren und Menschen. Cited in Hediger (1944).

SCENT-MARKING BY MALE MAMMALS

217

Vandenburgh, J. G. (1971). The influence of the social environment on sexual maturation in male mice. J. Reprod. Fert. 24, 383–390. Vandenbergh, J. G., Whitset, J. M., and Lombardi, J. R. (1975). Partial isolation of a pheromone accelerating puberty in female mice. J. Reprod. Fert. 43, 515–523. Vandenburgh, J. G., Finlayson, J. S., Dobrogosz, W. J., Dills, S. S., and Kost, T. A. (1976). Chromatographic separation of puberty accelerating pheromone from male mouse urine. Biol. Reprod. 15, 260–265. Viitala, J., Korpimaki, ¨ E., Palokangas, P., and Koivula, M. (1995). Attraction of kestrels to vole scent marks visible in ultraviolet light. Nature 373, 425–427. Voznessenskaya, V. V., Parfyonova, V. M., and Zinkevich, E. (1992). Individual odortypes. In “Chemical Signals in Vertebrates VI” (R. L. Doty and D. Muller-Schwarze, ¨ ed.), pp. 503–508. Plenum Press, New York. Walther, F. R. (1978). Mapping the structure and the marking system of a territory of the Thomson’s gazelle. E. Afr. Wildl. J. 16, 167–176. White, P. J., Kreeger, T. J., Tester, J. R., and Seal, U. S. (1989). Anal-sac secretions deposited with faeces by captive red foxes (Vulpes vulpes). J. Mammal. 70, Whitten, W. K. (1956). Modification of the oestrous cycle of the mouse by external stimuli associated with the male. J. Endocrinol. 13, 399–404. Wilcox, R. M., and Johnston, R. E. (1995). Scent counter marks: Specialised mechanisms of perception and response to individual odours in golden hamsters, Mesocricetus auratus. J. Comp. Psychol. 109, 349–356. Wolff, R. J. (1985). Mating behaviour and female choice: Their relation to social structure in wild caught House mice (Mus musculus) housed in a semi-natural environment. J. Zool. (London) 207, 43–51. Woodward, R. L., Schick, M. K., and Ferkin, M. H. (1999). Response of prairie voles, Microtus ochrogaster (Rodentia, Arvicolidae), to scent over-marks of two same-sex conspecifics: A test of the scent-masking hypothesis. Ethology 105, 1009–1017. Yamashita, J., Hayashi, S.-I., and Hirata, Y. (1989). Reduced size of preputial glands and absence of aggressive behaviour in the genetically obese (ob/ob) mouse. Zool. Sci. 6, 1033–1036. Yamazaki, K., Boyse, E. A., Mike, V., Thaler, H. T., Mathieson, B. J., Abbott, J., Boyse, J., Zayas, Z. A., and Thomas, L. (1976). Control of mating preferences in mice by genes in the major histocompatibility complex. J. Exp. Med. 144, 1324–1335. Zahavi, A. (1975). Mate selection: A selection for a handicap. J. Theor. Biol. 53, 205–214. Zhang, Q. H., and Schlyter, F. (1996). High recaptures and long sampling range of pheromone traps for fall web worm moth Hyphantria cunea (Lepidoptera: Arctiidae) males. J. Chem. Ecol. 22, 1783–1796. Zuri, I., Gazit, I., and Terkel, J. (1997). Effect of scent-marking in delaying territorial invasion in the blind mole-rat Spalax ehrenbergi. Behaviour 134, 867–880.

This Page Intentionally Left Blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 30

Male Facial Attractiveness: Perceived Personality and Shifting Female Preferences for Male Traits across the Menstrual Cycle IAN S. PENTON-VOAK AND DAVID I. PERRETT SCHOOL OF PSYCHOLOGY UNIVERSITY OF ST. ANDREWS ST. ANDREWS, FIFE, KY16 9JU UNITED KINGDOM

I. INTRODUCTION A review of the facial attractiveness literature describes the field as “among the success stories in cognitive science” (Thornhill and Gangestad, 1999). Certainly, studies of female attractiveness have achieved remarkable consensus as to the facial features that are preferred by males, and the biological signals that such characteristics may convey (e.g., Johnston and Franklin, 1993; Perrett et al., 1994, 1998; Jones, 1995). Yet studies of male attractiveness, often using techniques identical to those used in studies of female faces, have failed to isolate features that are consistently found attractive. Cross-cultural studies of mate preferences indicate that men are more influenced by physical appearance than women in mate choice (Buss, 1989). Observation suggests that women, rather than men, are the sex that advertise through extensive use of makeup and clothing: “Faced with these facts, a biologist would be forced to suspect that he was looking at a society in which females compete for males, rather than vice versa”(Dawkins, 1976). To those familiar with sexual selection in nonhuman species this seems to present a paradox, as Homo sapiens is moderately sexually dimorphic in body size, males mature later than females, and testis size tentatively suggests a unimale mating system (Harcourt, 1996). These sex differences indicate male competition for females, not vice versa. Indeed, the majority of documented human societies are best described as facultatively polygynous (Murdock, 1967; Daly and Wilson, 1983). In an attempt to resolve this paradox, many evolutionary theorists have been quick to point out that in a social species with biparental care such 219

C 2001 by Academic Press Copyright ° All rights of reproduction in any form reserved. 0065-3454/01 $35.00

220

IAN S. PENTON-VOAK AND DAVID I. PERRETT

as ours, both males and females should be choosy. Female reproductive potential is more easily assessed from physical appearance than male reproductive potential due to declining female fertility with age. Male ability to invest in children may even increase with age as status gains allow greater access to resources (e.g., Symons, 1979). Women, however, as well as men, are concerned with good looks in a mate. Although both males and females claim in self-report that physical attractiveness is not of primary importance when choosing a partner (Buss, 1989), the single best predictor of young adults’ satisfaction with a “blind date” is facial attractiveness for both men and women (Walster et al., 1966). Given the apparent importance of the face in mate choice decisions by both sexes, and the centrality of mate choice theories to evolutionary approaches to behavior, men’s facial characteristics may reflect the action of sexual selection. The purpose of this review is to examine both theories and studies of male facial attractiveness (Section II) and studies of personality attributions made to face shapes that may potentially influence female judgments of male attractiveness (Section III). Finally, we outline some of our recent work investigating how shifting female preferences for male faces across the menstrual cycle may balance the costs and benefits associated with different male traits (Section IV). We hope that this summary may at least partially unite some of the conflicting findings in the current literature.

II. THEORIES OF HUMAN FACIAL ATTRACTIVENESS Despite increasing professional and public interest in the field, systematic studies of facial attractiveness have not yet reached consensus on the characteristics that make human male faces attractive to females. The three main hypotheses are that (1) symmetrical faces, (2) average faces, or (3) faces with exaggerated secondary sexual characteristics are found attractive. Each of these hypotheses (which are not mutually exclusive) can be justified by “good genes” theories, and each has received some empirical support, briefly reviewed here. All three of these theories rely on facial characteristics honestly advertising biological quality: symmetry is related to developmental stability, which may be linked to genotypic quality; averageness may indicate heterozygosity; and exaggerated sex-typical traits may represent an “honest handicap.” All, then, are thought to be phenotypic signals linked to heritable genotypic quality and reproductive potential, and, hence, preferences for such characteristics should be favored by selection. Empirical proof that facial attractiveness is associated with characteristics that advertise health is, however, conflicting in contemporary populations (Kalick et al., 1998; Shackleford and Larsen, 1999; see also Daly and Wilson, 1999).

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

221

A. FACIAL SYMMETRY Symmetry is a marker of developmental stability that may reflect genotypic as well as phenotypic quality (Møller, 1997; Møller and Thornhill, 1997, for reviews). Bodily symmetry appears to be related to reproductive success in many species, including humans (Gangestad and Thornhill, 1997; Manning et al., 1997; Møller and Thornhill, 1998). Meta-analysis suggests that symmetry is more important to male attractiveness than to female attractiveness across species (Møller and Thornhill, 1998). A simple prediction is that facial symmetry will be important in judgments of attractiveness. Studies addressing this question are reviewed in the following two sections. 1. Studies of Naturally Occurring Facial Asymmetries There is no standardized technique for measuring the absolute symmetry of a human face, which may be a factor contributing to the relatively small amount of research that has been conducted into the symmetry of natural faces. Grammer and Thornhill (1994) estimated asymmetry in male and female faces by bisecting lines that delineated left and right positions of bilateral features, marked on each face in a small sample of 16 men and 16 women. Horizontal symmetry measured in this way correlated with attractiveness judgments of both male and female faces. Using a similar measurement technique, but with the inclusion of a measure of vertical as well as horizontal symmetry, Scheib et al. (1999) also found that symmetry and rated attractiveness correlated for a sample of 40 male faces. The relationship between symmetry and facial attractiveness, however, was still observed when only the left or right half of each face was presented (i.e., without cues to symmetry), suggesting that some covariate of symmetry that can be ascertained from half-faces may influence attractiveness judgments. Scheib et al. found that size of the lower face and cheekbone prominence were related to symmetry in both half- and full-face attractiveness judgments. It should be noted, however, that some cues to symmetry may still be present in half-faces. The midline used to generate the half-face stimuli bisected the middle of the nose tip and the “v” of the upper lip. In faces with low symmetry, other points on this midline may reveal more (or less) than half of centrally placed features (e.g., bottom of chin, bridge of nose). Hence, some cues to symmetry may still remain in half-faces. Mealey et al. (1999) studied the covariation of symmetry and attractiveness in monozygotic twin pairs. Such twins are genetically but not developmentally identical, and, hence, manifest differing levels of facial symmetry when adult. The symmetry of each twin pair was assessed by presenting left–left (L–L) and right–right (R–R) chimeric image pairs of each of the two twins to adult participants, and asking them to rate the similarity of each pair of images (Twin 1 L1–L1 and R1–R1; Twin 2 L2–L2 and R2–R2). The most similar

222

IAN S. PENTON-VOAK AND DAVID I. PERRETT

pair of images (i.e., L1–L1 and R1–R2 or L2–L2 and R2–R2) was assumed to belong to the more symmetric of the two twins. Secondly, the natural images of the twins were presented to separate raters, who were asked to rate the more attractive of each twin pair using a rating task. The results indicated a significant correlation between symmetry and attractiveness for both male and female twins—the more symmetrical of each twin pair was generally preferred, and bigger differences in estimated symmetry led to larger attractiveness differentials within twin pairs. 2. Studies Using Computer Graphic Symmetry Manipulations Given the theoretical benefits ascribed to symmetry (e.g., Møller and Thornhill, 1998), and the results of studies of naturally occurring asymmetry in faces, it is surprising that several studies directly manipulating human facial images have found that asymmetry is generally preferred to symmetry (Kowner, 1996; Langlois et al., 1994; Samuels et al., 1994; Swaddle and Cuthill, 1995). Most of these studies have created symmetric face images by aligning one half-face with its mirror reflection (Kowner 1996; Langlois et al., 1994; Samuels et al.,1994). These techniques may induce additional stimulus differences unrelated to symmetry. The mirror-reflecting technique can introduce abnormal feature shapes. For example, a mouth of normal width displaced to the right of the midline will assume atypical widths in left-mirrored and right-mirrored chimeric face images (Fig. 1, C,D). Asymmetry may have been preferred in a further study (Swaddle and Cuthill, 1995) because asymmetric original faces were compared to symmetric images with different skin textures (induced by combining original and mirror images); see Perrett et al. (1999) for a discussion of these problems, and Fig. 1. Despite the fact that experiments using chimeric stimuli have failed to detect a preference for symmetry, several studies have demonstrated that symmetry predisposes favorable facial attractiveness judgments, reconciling the results of computer graphic studies with studies of naturally occurring asymmetry. Perrett et al. (1999) present three experiments using novel techniques for manipulating symmetry: these experiments are described in some detail next.

FIG. 1. Symmetry manipulation for facial images with natural skin textures. Real face images with normal and symmetric shapes. (A, B) Note that asymmetries in pigmentation and shadows present in the original faces (A) remain in the more symmetrically shaped versions (B). (C, D) Images made with techniques employed in previous studies of facial symmetry. Chimeric faces made by combining the left sides of the original faces with their mirror reflections (C) and similarly for the right sides of the faces (D) illustrate the shape abnormalities that this technique induces. From Perrett et al. (1999).

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

223

224

IAN S. PENTON-VOAK AND DAVID I. PERRETT

In Experiment 1, symmetry in face shape was improved without changing the symmetry of face textures; natural asymmetries in skin pigmentation were present in both the original and more symmetric versions of the same face (Fig. 1). Adults’ responses to pair-wise presentation of these two versions of each face indicated a clear preference for the symmetrically remapped stimuli. Experiment 2 used stimuli with average texture information generated from a set of faces. This average texture was rendered into both the original shapes and the symmetrically remapped shapes of a set of individual faces, giving perfect symmetry in the remapped version. Pair-wise presentation showed a preference for perfectly symmetrical face stimuli. Experiment 3 used a rating task rather than a forced choice paradigm (stimuli were presented one at a time rather than in pairs), replicating the preference for symmetry shown in Experiments 1 and 2. In Experiment 1, symmetry manipulations were rather subtle; perhaps because of this subtlety of the manipulation, 75% of subjects were unaware that symmetry had been manipulated. Despite this unawareness, subjects preferred symmetrical faces, demonstrating that symmetry influences attractiveness judgments even without conscious awareness. Experiment 1 goes some way toward establishing the ecological validity of computer graphic research into facial symmetry because it demonstrates a preference for more symmetrical male and female facial images with entirely natural skin pigmentation and texture. Rhodes et al. (1998) used a technique similar to that of Swaddle and Cuthill (1995), but avoided stimulus artifacts such as double blemishes by retouching the manipulated images. Using this improved technique, they too found a preference for symmetry over asymmetry in faces of both sexes. Further support for a preference for symmetry in human faces comes from Hume and Montgomerie (1999) who replicated Rhodes et al.’s technique with similar results. Thus, the three most recent, and perhaps methodologically superior, computer graphic studies (Rhodes et al., 1998; Perrett et al., 1998; Hume and Montgomerie, 1999) parallel the findings of investigations into naturally occurring facial asymmetries (Grammer and Thornhill, 1994; Mealey et al., 1999; Scheib et al., 1999). There are progressive changes in the shape of faces, particularly in soft tissues, throughout life (e.g., Burt and Perrett, 1995). Asymmetries in face shape due to differential growth and aging will, therefore, be more prominent in older faces. Kowner (1996) demonstrated that asymmetry itself may be a perceptual cue to age because asymmetric faces were perceived as older than symmetric faces. To summarize, it is apparent that symmetry is one factor contributing to judgments of facial attractiveness. It is also apparent, however, that symmetrical faces may possess other properties that are associated with attractiveness.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

225

For example, sex hormones may influence the symmetry of growth (Thornhill and Gangestad, 1993) and chin shape and size, which independently affect attractiveness (Perrett et al., 1994; Scheib et al., 1999). A covariate of symmetry, rather than symmetry itself, may be a primary cue to attractiveness judgments (Scheib et al., 1999). B. AVERAGENESS The most influential and most investigated theory of facial attractiveness is that average features, not exaggerated characteristics, are optimal, as such faces are thought to indicate high levels of heterozygosity in their owners (Symons, 1979; Thornhill and Gangestad, 1993). As extreme genotypes that are more likely to be homozygous for deleterious alleles (Symons, 1979) and less likely to have alleles to which pathogens are poorly adapted (Thornhill and Gangestad, 1993) are selected against, a preference for average faces may be adaptive. A preference for averageness is compatible with both cognitive theories of prototyping (see Langlois and Roggman, 1990) and work in theoretical biology suggesting that a preference for “average” phenotypes would rapidly replace random mating (Koeslag, 1994; Koeslag, 1994). However, the averageness hypothesis has received only mixed empirical support. Modern computer graphic techniques develop a method first suggested by Francis Galton (1878) to allow composite “average” faces to be constructed from much larger sets of individual photographs. Langlois and Roggman (1990) found that composite (average) faces were rated as more attractive than the individual faces that formed them. A preference for average faces over extremely distinctive faces has been demonstrated using line drawings, although the comparison is confounded by symmetry (Rhodes and Tremewan, 1996). A preference for averageness may explain the popular belief that people are attracted to opposite sex individuals that resemble themselves. Penton-Voak et al. (1999a) presented female subjects with a series of synthetic male faces, one of which had been generated from their own face (a gender transform—see Rowland and Perrett, 1995). It was found that subjects rated faces more similar to their own as more attractive than other faces. A preference for averageness, however, rather than similarity, could generate this finding, because faces very far from average receive low attractiveness ratings and such atypical faces differ from the faces of most individuals more than average faces. Alley and Cunningham (1991) proposed that average-sized features would be attractive but not optimal, and briefly reviewed studies using methodologies not reliant on digital averaging that failed to support the averageness hypothesis. Perrett et al. (1994) demonstrated that, although average composite male and female faces are indeed attractive, more attractive composites can

226

IAN S. PENTON-VOAK AND DAVID I. PERRETT

FIG. 2. An original male face (top left) can be warped into an average face shape derived from a sample of 26 young adult males (bottom left). Color information from an average face can be warped into an individual face shape (top right). The average face is shown bottom right.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

227

be generated by using a subset of the 15 most attractive faces from a set of 60 individuals. A counterargument to this finding suggests that the 15 most attractive individuals in the group may be, in fact, the 15 most “average” faces in the group, and, hence, a composite of them represents the population average better than the full 60-face group. This argument does not explain why a caricature of the 15-face female average (exaggerating the differences between the 60- and 15-face average by 50%) creates a yet more attractive face (Perrett et al., 1994). Pollard et al. (1999) used facial-metric measurements of feature size rather than composites to calculate averageness, and found that faces with features close to the average size of the population studied were rated as exactly that—average—in attractiveness. Average composite faces tend to have smooth skin and be symmetric; these factors, rather than averageness per se, may lead to the high attractiveness attributed to average faces (Alley and Cunningham, 1991; Benson and Perrett, 1992). Manipulations of averageness using combinations of average shape and average color information from composites reveal that both skin texture and face shape influence attractiveness. Individual faces warped into average face shapes (Fig. 2) are rated as more attractive than the original (Figs. 2 and 3). Similarly, rendering the color information from a blended average into an original shape improves attractiveness ratings. Average shape

FIG. 3. Mean and SE of attractiveness ratings given by 28 subjects to 16 original male images, 16 original face shapes with average color, 16 faces with original color and average shape, and the average image (N = 28).

228

IAN S. PENTON-VOAK AND DAVID I. PERRETT

and color together typically result in the highest attractiveness ratings (Figs. 2 and 3; Benson, 1993; Little, 1999). Dissociating symmetry and averageness is problematic. As no reliable directional asymmetries have been found in human faces at rest (e.g., Perrett et al., 2001), average faces are by definition symmetric. Average size and symmetry of real faces, however, may not be simply related. Studies of bird species indicate that larger than average secondary sexual characteristics (such as swallows’ tails) are more symmetric than smaller traits (Møller and Hoglund, 1991). In women, breast asymmetry is negatively related to breast size (large breasts are relatively more symmetric than small breasts; Manning et al., 1997) and, in men, facial symmetry appears to be related to jaw and cheekbone size (Scheib et al., 1999). So although in composite images symmetry increases as facial averageness increases, this may not reflect patterns of symmetry in real faces. Symmetry, rather than averageness, may drive the attractiveness judgments of composite faces. However, a study by Rhodes et al. (1999) independently manipulated averageness and symmetry and concluded that both positively influence attractiveness judgments.

C. SECONDARY SEXUAL CHARACTERISTICS AND SEXUAL DIMORPHISM Symons (1994) proposes that averageness may be a “default” position that defines facial attractiveness unless another psychological mechanism is operating. A preference for secondary sexual characteristics may be one such alternative mechanism operating in women’s judgments of male facial attractiveness, as it appears to be in other species (Andersson, 1994). The growth of male secondary sexual characteristics in mammals reflects androgenic activity; such hormones are thought to be immunosuppressive (e.g., Hillgarth and Wingfield, 1997). Consequently, male traits are often hypothesized to represent an “honest handicap” (Zahavi, 1975; Folstad and Karter, 1992). 1. Sex Differences in Facial Growth Sexual selection is often implicated in sexual dimorphism; male–male competition leads to sexual selection for successful (i.e., large, strong) males. Male–female dimorphism in human faces radically increases at puberty under the influence of sex steroids, particularly testosterone and estrogen. Levels of both hormones increase in both pubescent boys and girls, although the ratio of androgens to estrogens is dependent on sex. High testosterone levels cause forward growth of the brow ridges, and an increase in the size of the bones of the jaw, lower face and cheekbones (Thornhill and Gangestad, 1996; Gage et al., 1999). As masculinity in mammals is

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

229

testosterone dependent, whereas femininity represents the more neutral developmental state (Haaq and Donahoe, 1998; Owens and Short, 1995), male faces grow more than female faces, which remain relatively childlike. High estrogen apparently prevents facial bone growth in pubertal females, and leads to fat deposition in the lips and cheek area (Thornhill and Gangestad, 1999). Considerable evidence suggests that extremely feminine female faces are considered attractive. A wide variety of techniques ranging from (1) measurement of facial photographs of women (Cunningham, 1986; Grammer and Thornhill, 1994; Jones and Hill, 1993) through (2) studies of facial composites (Perrett et al., 1994, 1998; Rhodes et al., 2001) to (3) the generation of attractive female face shapes using genetic algorithms (Johnston and Franklin, 1993), show that feminine features indicating estrogenized female faces increase their attractiveness cross-culturally. Sex-typical female features (small lower face, a relatively flat mid-face, full lips and high eyebrows associated with a lack of brow ridge prominence) may indicate youth (estrogen levels decrease in adult females with age) and reproductive health (Symons, 1979, 1994; Thornill and Gangestad, 1993, 1996, 1999; Singh, 1993). A wealth of studies indicate that femininity (an exaggeration of female sextypical features) rather than averageness is most attractive in female faces. Given the probable “signaling” properties of estrogenized female faces, a male preference for such features is potentially adaptive. There is some evidence for female preferences for exaggerated male facial characteristics. Scheib et al. (1999) found a positive relationship between attractiveness and two markers of facial masculinity (cheek bone prominence and jaw size). Cunningham et al. (1990) and Grammer and Thornhill (1994) used facial measurements, and found a female preference for large jaws in males. “Masculine” features, such as a large jaw and a prominent brow ridge, are also reliably associated with ratings of dominance in photographic, indenti-kit and composite stimuli by male and female raters (McArthur and Apatow, 1983–1984; Berry and Brownlow, 1989; Berry and Wero, 1993; McArthur and Berry, 1987; Perrett et al., 1998). Facial dominance appears to correlate with status in some human hierarchies (Mueller and Mazur, 1997) and facial dominance in adolescent males is associated with earlier age at first copulation (Mazur et al., 1994). Nonetheless, the relationship between facial dominance and attractiveness is unclear—some studies find a positive relationship (Keating, 1985) while others find the opposite (McArthur and Apatow, 1983–1984; Berry and McArthur, 1985; Perrett et al., 1998). Berry (1991) had a large set of male photographs rated for attractiveness and, independently, babyishness. She found, essentially, evidence for two types of attractiveness—sincerity (associated with attractive babyfaces) and

230

IAN S. PENTON-VOAK AND DAVID I. PERRETT

power (associated with attractive “mature” faces). As well as preferences for some typically “masculine” features, Cunningham et al. (1990) also found preferences for more feminine features (such as large eyes and an expressive mouth). Cunningham et al. suggest that women have “multiple motives” when choosing mates: optimally attractive faces may simultaneously appear dominant and yet elicit nurturing responses. 2. Manipulating Sexual Dimorphism in Faces Using Computer Caricaturing Techniques Although comparative approaches to human facial preferences suggest that exaggerated sex-typical traits will be attractive in both male and female faces, earlier studies (reviewed above) suggest that although this may hold true for female faces, preferences for male faces are more complex. Our recent work has used computer graphic techniques to manipulate facial sexual dimorphisms in stimuli used in preference tests; some of this work is reviewed below. Computer graphic techniques can be used to construct average male and female faces by digitally blending photographs of individuals of one sex. Sexual dimorphism in face shape can then be enhanced or diminished by looking at the geometrical differences between male and female face shapes and either exaggerating or minimizing them. For example, “masculinizing” a male face shape by increasing the differences between a male and female average increases the size of the jaw and reduces lip thickness (among all other dimorphic characteristics), as male jaws are larger than female jaws and the lips of men are thinner than those of women. These “masculinized” or “feminized” versions of male and female composite faces are useful stimuli for testing the influence of sexually dimorphic characteristics on judgments of facial attractiveness, although they are not without critics. Meyer and Quong (1999) suggest that as our stimuli are based on male–female differences rather than within-sex differences they fail to accurately simulate differing androgen levels in male faces. Although we concede that stimuli based on within-sex differences in sex steroids would be superior to our stimuli, we assume that male–female differences parallel the differences between individuals with high and low androgens (Perrett and Penton-Voak, 1999). Femininity in mammals represents the more neutral developmental state: masculinization is dependent on androgens (Owens and Short, 1995). Typically male XY individuals with complete androgen insensitivity syndrome have a female appearance (although they do have “male” dentition; Haaq and Donahoe, 1998). It seems unlikely that feminizing a face by applying male–female differences produces facial shapes associated with higher than average levels of androgens.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

231

FIG. 4. Male images: (left) 50% feminized and (right) 50% masculinized.

Using this technique, the amount of “masculinity” or “femininity” transform applied to a face is specified as a percentage. The shape differences between male and female faces are described by a set of vectors between marked delineation points on the features of the male and female averages (172 delineation points define the outline of the face, the eyes, the mouth, the nose, etc.). Transforms are expressed as a percentage of the distance traveled along these vectors: in a 25% feminized male face shape, each delineation point is moved 25% of the way along the vector to the female average face. The color information from the original male average is then warped into this new shape. To masculinize male face shapes, the direction of the male– female vector is reversed before the points are moved along it (see Fig. 4 for examples of feminized and masculinized male face stimuli). Such manipulations of both Japanese and Caucasian face stimuli were presented to Japanese and Caucasian adult males and females in their country of origin (Perrett et al., 1998). Participants could alter the appearance of a face on a computer monitor by using a computer mouse. Moving the mouse left or right would change the amount of masculinity or femininity in the face in real time, under the subject’s control. For the male face stimuli, the shape selected by Caucasians as most attractive (from the shape range available) was significantly feminized for both the

232

IAN S. PENTON-VOAK AND DAVID I. PERRETT

Caucasian male face and the Japanese male face continua. Japanese participants also selected significantly feminized versions of the male stimuli for both the Japanese and Caucasian male face continua. There were no effects of subject sex, subject population, type of stimulus face or any interactions between main effects (Perrett et al., 1998). Previous studies show crosspopulation consistency in attractiveness judgments. Our study showed crosscultural (between population) agreement in the preference for feminized over average face shapes, which further refutes the averageness hypothesis. This preference for feminized male faces seems contrary to predictions from indicator mechanisms of sexual selection and to some other published studies of male facial attractiveness reviewed briefly above. Yet, other studies report a somewhat similar trend [e.g., Berry and Zebrowitz McArthur’s (1986) reported preference for “babyfaced” men; the “multiple motives” theory of Cunningham et al. (1990)]. Rather than preferring typically masculine faces (with prominent brow ridges and large jaws), both male and female adults appear to favor a small amount of femininity in men’s faces, using stimuli prepared in this way. One possible explanation for this somewhat unusual finding may rest on stereotypical personality attributions that individuals make to faces. Evidence in favor of this hypothesis is reviewed next.

III. PERSONALITY ATTRIBUTIONS AND FACIAL CHARACTERISTICS A. IS PHYSIOGNOMY ACCURATE? The explanation for the unexpected preferences for feminized male faces reported above may lie in the personality attributions these composite faces elicit (Perrett et al., 1998). The attribution of personality to faces, physiognomy, has had a long and checkered history across many cultures. Yet, on the whole, scientists of the twentieth century have dismissed physiognomists as being, at best, misguided: Respectable science holds . . . that there is not any connection between the features of the face and the character of the person . . . Any connection between these two wholly different kinds of personal qualities would bespeak of some mystical system of correspondences between the mind and body; scientists could only regard such a system as absurd (Brandt, 1980; in Berry and Brownlow, 1989).

Despite such condemnation, Liggett (1974) demonstrated that the advice of science has clearly been ignored: 90% of university students see the face as a valid guide to a person’s character. Contrary to the orthodox scientific view that physiognomy is a worthless area for study that should be classified (and condemned) along with phrenology, recent literature contains a number

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

233

of studies concerning personality judgments from physical characteristics. These will be briefly reviewed next. B. STUDIES OF ZERO ACQUAINTANCE AND THE “BIG FIVE” PERSONALITY FACTORS Although physiognomists have often been dismissed as charlatans or worse (e.g., Cleeton and Knight, 1924), their work may have been, in at least part, based on observable relationships between physical appearance and personality. Studies using “zero acquaintance” paradigms (in which participants rate the personality of strangers) have found a surprising degree of correlation between self-ratings and stranger ratings on personality factors, often using five-factor models of personality (e.g., Norman, 1963). Many trait theorists agree that five trait dimensions seem to represent a reasonable compromise that is at least a partially accurate measure of people’s personalities (e.g., McCrae and Costa, 1987; Watson, 1989; Barrett and Pietromonaco, 1997). These five dimensions have been derived from factor analyses of ratings of individuals using trait adjectives (e.g., talkative-quiet, cold-warm, etc., See Norman, 1963; McCrae and Costa, 1987; Botwin et al., 1997). Although different researchers disagree on the best names for these five factors, they are commonly referred to as Extraversion, Agreeableness, Conscientiousness, Emotional Stability (Neuroticism) and Openness-Intellect (or Culture). Without verbal interaction, individuals in small groups are to some extent able to make accurate perceptions of others’ self-reported personalities. Using small group paradigms, in which four or five unacquainted individuals assess each others’ personalities, factors of extraversion, agreeableness, and conscientiousness are moderately accurately inferred; culture and emotional stability less so (e.g., Passini and Norman, 1966; Albright et al., 1988; Watson, 1989; see Kenny et al., 1994 for a review). These findings seem to generalize cross-culturally: Albright et al. (1997) used a similar paradigm with Chinese students in Beijing, and reported a similar pattern of results to those found in Western studies. Small group paradigms, of course, allow nonverbal communication and interaction to convey personality. Nonetheless, Kenny et al. (1992) replicated consensus and accuracy in personality perception using videotapes of individuals, and Borkenau and Liebler (1992) showed that somewhat legitimate personality judgments can be made from still photographs alone. There are both individual and sex differences in accuracy of personality attributions. Ambady et al. (1995) report that women are more accurate judges of strangers’ personalities than men, consistent with other literature on nonverbal behavior.

234

IAN S. PENTON-VOAK AND DAVID I. PERRETT

C. THE EFFECTS OF “MATURE” AND “BABYFACE” CHARACTERISTICS ON ATTRIBUTIONS OF FACIAL DOMINANCE Several researchers have devoted much effort to examining the characteristics and consequences of faces that are described as “dominant” or “babyfaced.” Dominance can be considered as a trait encompassed by the extraversion factor in the five-factor model (although it appears to load more heavily on the agreeableness factor in some studies), but it has received considerable attention in studies outside of this theoretical orientation. Dominance is a characteristic that is closely linked to male reproductive opportunities and success in many species including humans. Dominant looking male teenagers, for example, copulate earlier than less dominant looking peers (Mazur et al., 1994). Thus, the characteristics that lead to attributions of dominance may influence attractiveness judgments. Most studies agree on the characteristics that make male faces appear dominant: mature features, such as a large jaw and a prominent brow ridge, are reliably associated with ratings of dominance in photographic, indentikit, and schematic stimuli (Berry and Wero, 1993; Berry and Brownlow, 1989; McArthur and Berry, 1987; McArthur and Apatow, 1983–1994). Similarly, the literature agrees that babyfaces are characterized by smaller chins, high eyebrows, and larger eyes. Such faces are generally rated as being warmer, more honest, and more sincere, but also more naive and less physically strong (McArthur and Apatow, 1983–1984; Berry and McArthur, 1985; Berry, 1991). Babylike and dominant faces reliably elicit personality attributions cross-culturally (e.g., Keating et al., 1981; McArthur and Berry, 1987) but their effect on attractiveness judgments is equivocal as discussed above. Such dominance attributions, however, influence potentially important social interactions. Facial maturity has been shown to have more of an effect on male daily social experience than on female routine interactions (Berry and Landry, 1997). By simulating a trial situation, Berry and Zebrowitz McArthur (1986) demonstrated that manipulating the maturity of the “defendant” influences both the chance of “conviction” and the “sentence” given. Babyfaced individuals were less likely to be found guilty of charges involving intentional criminal behavior, although they were more likely to be found guilty of a negligent crime. Babyfaced defendants were also given lighter sentences for negligent crimes, reflecting the effects of attributed naivety and honesty. In a study with perhaps more ecological validity, Mueller and Mazur (1997) showed that facial dominance (as rated by undergraduates) of the graduates from the West Point Military Academy in 1950 predicted their final rank at the end of their careers. Such an environment may reward success in physical male–male competition more than many other walks of life.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

235

Consensus in judgments of the personality of mature or babylike faces does not necessarily indicate that such attributions are valid, even though they influence potentially important social situations. Many of the personality attributions made to babylike faces are also, unsurprisingly, made to infants themselves. Lorenz (1943) suggested that positive behavioral responses to infants are adaptive in that they increase offspring viability. Similar responses to adult faces with some childlike qualities may be inappropriate generalizations of an otherwise adaptive behavior. This argument alone, however, fails to explain the evidence that babyfaced or mature individuals do to some extent self-report that they have the personality characteristics attributed to them. Berry and Brownlow (1989) found that ratings of male babyishness were positively correlated with the face owner’s self-reported approachability and warmth, but negatively related to self-reported aggression. For female faces, babyishness was associated with low self-reported levels of physical power and assertiveness. Bond et al. (1994) demonstrated that individuals whose faces were rated as being “less honest” were more likely to volunteer for experiments that involved them in deceiving others than people who were judged to look more honest. Such correspondence between self-reported and observed personality variables is often explained in terms of a behavioral confirmation or “selffulfilling prophecy” effect (Watson, 1989; Berry and Wero, 1993). It seems that certain facial configurations reliably elicit certain personality judgments. If someone is regularly perceived and treated as if they are submissive (or dominant), they may modify their behavior so that they act in a more submissive (or dominant) manner. This, in turn, may lead to an internalization of such behavior, and an individual may come to see themself as possessing the characteristics with which they are stereotypically attributed. The behavioral confirmation theory may operate to strengthen the relationship between expected and observed behaviors of male strangers. It does not, however, provide a convincing explanation of the origin of stereotypes necessary for consistent ratings of dominance and other characteristics (Watson, 1989). Cultural factors are often invoked, and yet, cross-cultural studies show considerable agreement in attributions of dominance. This has led some researchers to posit a biological basis for the personality ratings given to faces. As the differences between “dominant” and “baby” faces clearly parallel the differences between “masculine” and “feminine” composite stimuli, they probably reflect underlying androgen levels. Testosterone is also linked to male dominance behaviors (Mazur and Booth, 1998), tentatively providing a biological link between facial appearance, behavior, and valid (though stereotypical) personality judgments. Given the reported importance of personality to human mate choice (e.g., Buss, 1989), it seems likely that stereotypical personality judgments may

236

IAN S. PENTON-VOAK AND DAVID I. PERRETT

influence attractiveness judgments of unfamiliar faces. First impressions last, after all. 1. Personality Judgments Made to Composite Faces To see whether personality attributions influenced the attractiveness judgments of our composite stimuli, we presented sets of three Caucasian and Japanese male composite stimuli (in masculinized, average, or feminized versions) to university undergraduates and research staff in Scotland (Perrett et al., 1998). Increasing the masculinity of face shape across the three set members increased ranking of perceived dominance, masculinity, and age, but decreased ranking of perceived warmth, emotionality, honesty, cooperativeness, and quality as a parent (Friedman’s ␹ 2 ; N = 20, p < 0.0005 for each dimension). Even with the small number of raters in our sample, and the relatively small differences between the stimuli, consensus in attributions was nearly total. These results demonstrate that exaggerating or reducing sexual dimorphism in composite faces alters their perceived masculinity or femininity in the predicted direction. Thus, manipulations of this sort clearly influence perceptions that convey the “psychological meaning” of masculinity and femininity (Meyer and Quong, 1999; Perrett and Penton-Voak, 1999). In addition, the considerable consensus in the ranking of stimuli indicates that, although masculinized and feminized versions of stimuli are fairly similar, they must contain stereotypical cues to personality. Socially valued traits such as honesty, warmth, cooperation, and skill as a parent are associated with feminized versions of male faces, whereas traits such as dominance are, as predicted, associated with masculinized face shapes. A female preference for slightly feminized male faces can perhaps be at least partially explained by these personality attributions. Although biological predictions such as handicap theory indicate that females should prefer masculinized faces, such faces elicit negative personality attributions. Crossculturally, personality factors are reported to be the most important factor in mate choice by both sexes (Buss and Barnes, 1986; Buss, 1989). So, it seems inconceivable that personality attributions could have no effect on attractiveness judgments. Feminization of male face shape may increase attractiveness because it softens particular features that are perceived to be associated with negative personality traits. 2. Testosterone, Behavior, and Face Shapes The behavioral confirmation theory of accurate personality ratings and a putative biological theory of facial dominance may in fact interact and reinforce one another. Male testosterone levels are known to increase in the winners and decrease in the losers of competitions (e.g., Mazur et al., 1997). Success in dominance interactions may increase testosterone levels in

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

237

a way that influences the growth of male facial characteristics, especially in adolescents. A positive reciprocal relationship between dominance behavior influenced by stereotypical personality attributions and levels of testosterone is therefore plausible (Gage et al., in preparation; Manning, pers. comm.). Increasing the testosterone level in adult males is also associated with more troubled sexual relationships. A sample of 4462 former members of the U.S. armed forces showed that men with testosterone levels one standard deviation above the mean (estimated from one measurement) were 50% more likely never to marry than men with testosterone one standard deviation below the mean (Booth and Dabbs, 1993). The same study demonstrated that married men with high testosterone were also more likely to suffer troubled relationships, increased incidence of domestic violence, and extra-pair sex. The implications of these findings for theories of sexual selection in humans are clear: whatever indirect “good genes” benefits high testosterone may offer in terms of offspring viability, there is an apparent cost in terms of likely paternal investment. It is worth reiterating that women make more accurate personality judgments in zero-acquaintance studies than men (Ambady et al., 1995). As lower initial investment in offspring makes males the sex most likely to desert (Symons, 1979), selection pressures to make accurate personality judgments might be expected to be stronger in females than males. In the holistic transforms employed to masculinize and feminize the stimuli used in our experiments, all male–female facial differences visible in frontal facial photographs were manipulated. Our methodology has generated a robust and reliable preference for slightly feminized male stimuli using participants (university students) matched in age with those stimuli, both in the United Kingdom and Japan and also in South Africa (Perrett et al., 1998). This preference is unexpected given that dominant-looking males copulate earlier than their less-dominant-looking peers (Mazur et al., 1994) and succeed in competitive hierarchies (Mueller and Mazur, 1997), and that “cultural success” as reflected by status in social hierarchies is positively correlated with copulation frequency in contemporary Canada (Perusse, 1993). Cues to cultural success (such as high-status clothing) undoubtedly positively influence attractiveness judgments of males (e.g., Townsend and Roberts, 1993; Townsend and Wasserman, 1998). Testosterone appears to be related to dominance behavior and so, by extension, position in hierarchies, but simultaneously testosterone is associated with less likelihood of extended paternal investment (Mazur and Booth, 1998). One aspect of female mate choice may be a cost–benefit analysis of such conflicting factors: a high status/high testosterone partner may have access to more resources, and offer heritable benefits to offspring; but he may also be more likely to desert or even injure the female who chooses him (Mazur and Booth, 1998). A female preference for a lower status partner

238

IAN S. PENTON-VOAK AND DAVID I. PERRETT

who is more likely to provide long-term paternal investment may, overall, pay a greater reproductive dividend. Given that stereotypical personality judgments have some accuracy, such attributions may form an important component in women’s mate-choice decisions. With our stimuli, personality attributions apparently outweigh biological fitness cues in attractiveness judgments of male faces. In real faces, perceived personality may not be as closely linked to masculine characteristics: combinations of masculine and feminine features, indicating both status and kindness may be optimally attractive. There is some support for such multiple motives influencing judgments of facial attractiveness (Cunningham et al., 1990; see also Scheib et al., 2001). The preferences we have found indicate a selection pressure that has acted against exaggerated dimorphism in male and female faces. Because more feminine face shapes are perceived as younger, the preferences would encourage a “youthful” facial structure in the human species generally. There is some paleo-anthropological evidence consistent with such a hypothesis, as both cranial and postcranial robusticity have been declining throughout the recent evolution of Homo (e.g., Brace et al., 1987; Ruff et al., 1993). A move toward more gracile morphology is often attributed to advances in food preparation and other technologies, and it is, of course, unclear whether sexual selection has played a part.

IV. MENSTRUAL CYCLE SHIFTS IN FACE PREFERENCES A. THE HUMAN OVARIAN CYCLE—CONCEALED OVULATION AND CONSTANT RECEPTIVITY? In contrast to most other mammals and many other primates, human female reproductive cycles do not have a clearly defined estrus period of sexual activity around ovulation. Instead ovulation is hidden both from males and to (most) females themselves. Women have intercourse, and similarly can refrain from intercourse, at any point in their menstrual cycle, leading some theorists to view human sexual behavior as “emancipated” from hormonal influences (Ford and Beach, 1951). To consider women continuously receptive implies a continuous state of estrus in which females are active solicitors of male attention (Hrdy and Whitten, 1987). This is an overstatement, as Frank Beach memorably noted in 1974: “Any male who entertains this illusion must be a very old man with a short memory or a very young man due for a bitter disappointment,” a sentiment echoed by Symons (1979): “The sexually insatiable woman is to be found primarily, if not exclusively, in the ideology of feminism, the hopes of boys, and the fears of men.”

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

239

The evolution of concealed ovulation and continuous receptivity has nonetheless attracted much attention from evolutionarily minded scholars of human behavior. Various hypotheses have suggested that, for example: (1) Concealed ovulation and continual sexual receptivity would strenthen the human pair-bond, and hence, cooperation within the group necessary for successful hunting/gathering lifestyles (e.g., Morris, 1968; Lancaster, 1975); (2) Ovulation was concealed as an adaptation to force desirable males into consort relationships, and thereby increase paternal investment (e.g., Alexander and Noonan, 1979; Symons, 1979); (3) Hominid females, with increasing intelligence, wished to avoid the danger and pain of pregnancy by consciously avoiding copulation around ovulation, but those females that could detect ovulation and avoided conception were selected against (Burley, 1979); (4) Cryptic ovulation confuses paternity, preventing infanticide (Hrdy, 1981); or (5) It allows hominid females to gain good genes benefits in extra-pair partners (Benshoof and Thornhill, 1979; Schroder, ¨ 1993). There are several in-depth reviews of these theories in the literature (Strassman, 1981; Gray & Wolfe, 1983; Steklis and Whiteman, 1989). Choosing between such hypotheses is difficult, as many are based on the assumptions that human concealed ovulation is a derived trait (from an ancestor with cyclic estrus swellings) and that something is known about the social structure and mating patterns of ancestral hominids. In fact, it is not at all clear that concealed ovulation is a derived trait from either comparative research with extant primates or studies of primate phylogeny (e.g., Dixson, 1983; Hrdy and Whitten, 1987; Burt, 1992; Sillen-Tulberg ´ and Møller, 1993). Furthermore, little can be concluded about mating systems from sexual dimorphism in fossil hominids. First, small sample sizes for separate hominid species make sexing fossil specimens problematic (e.g., Hausler and Schmid, 1995; Tague and Lovejoy, 1998) and estimating dimorphism unreliable (McHenry, 1994). Second, those estimates of sexual dimorphism that do exist reveal a curious pattern in the australopithecines. Whereas body size dimorphism is large, indicating strong male–male competition in early hominids, canine dimorphism is unexpectedly low (Plavcan and van Schaik, 1997). Although it is probably safe to say that neither polyandry nor monogamy was the mating system of ancestral hominid species, it is hard to say whether a unimale or multimale system was prevalent in our ancestors (Sillen-Tulberg ´ and Møller, 1993; Plavcan and van Schaik, 1997). B. CHANGES IN SEXUAL BEHAVIOR ACROSS THE MENSTRUAL CYCLE Despite the obvious lack of a true estrus phase in women, many researchers have proposed that female sexual desire, behavior, and preferences may vary across the menstrual cycle. The sources of this variation may

240

IAN S. PENTON-VOAK AND DAVID I. PERRETT

be social (e.g., sex during menstruation and postpartum is taboo in many cultures; Hrdy, 1981) or biological (resulting from the influence of fluctuating levels of sex hormones). Women’s menstrual cycles last between 21 and 35 days, with a mean duration of around 28 days. Many commentators have pointed out the great irregularity of cycle lengths both within and between women, which is especially interesting in comparison to the regularity of most mammals (e.g., Burley, 1979). Irregularity, however, also seems common in some nonhuman primates (e.g., Papio ursinus and Pan troglodytes, Burt, 1992). Nevertheless, most standard models of the menstrual cycle are based on a 28-day duration. In such models, ovulation occurs on approximately day 14 at the end of the follicular phase (Regan, 1996). The first 5 days (early follicular phase) is the period of menstruation, in which levels of estrogen, progesterone, and androgen are all low. In the midfollicular phase (days 6–10), estrogen levels rise steadily, androgens increase somewhat and progesterone levels remain low. Days 11–14 (the late-follicular phase) are characterized by a rise in androgen and progesterone and a rapid increase of estrogen levels, which peak approximately 2 days before ovulation. This estrogen peak causes a surge of luteinizing hormone, which, in turn, leads the follicle to rupture and the ovum to be released on day 14 in standard models. The 14 days following ovulation, the luteal phase, are characterized by secretion of progesterone from the corpus luteum. In this phase, progesterone and estrogen levels rise until a drop in the level of both hormones occurs before the onset of menstruation (the beginning of the next cycle). As human sperm may survive in the female reproductive tract for around 5 days (Baker and Bellis, 1995), copulations in the 5 days preceding ovulation (mid- to late-follicular phase) are most likely to result in conception. Despite the model given above, ovulation is actually fairly unpredictable, especially in the first years following menarche. Social factors, such as a lot of time spent in the presence of males, may shorten the cycle; external factors such as stress may result in anovulatory cycles (Metcalf et al., 1983). There is even the possibility that women, like some other mammals, may ovulate in response to copulation during the follicular phase (Jochle, ¨ 1973; Baker and Bellis, 1995). Peaks in sexual desire and activity have been reported around ovulation (e.g., Urdy and Morris, 1968; Adams et al., 1978; Stanislaw and Rice, 1988); in the midfollicular phase (Urdy and Morris, 1977; Bancroft et al., 1983), and also in the late-luteal/premenstrual period (e.g., Stewart, 1989). Further studies report multiple peaks of sexual desire and/or activity, often at ovulation and in the late-luteal phase (e.g., Silber, 1994). Dissociating various factors (e.g., male or female initiation of sexual activity) in this body of

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

241

research is problematic. Cultural proscriptions against sex during menstruation may influence reported patterns of desire. Some evidence that culturally enforced abstinence may account for perimenstrual peaks in sexual desire comes from Matteo and Rissman (1984), who investigated peaks of sexual activity in lesbian couples who did not reduce their sexual activity during menstruation, but did report an increased activity midcycle. Regan (1996) presents a comprehensive review of the literature and concludes that there is substantial variation between women who may have no, one, or two peaks of sexual desire. Women who do have peaks in sexual desire tend to have them either in the midfollicular, late-follicular (ovulation), or late-luteal phases of the menstrual cycle. Follicular phase or midcycle peaks in sexual interest would parallel many other primate species, as such peaks would be associated with likely conception following sex (Hrdy, 1981). Sexual desire, then, varies across individual women’s cycles and between different women (Regan, 1996). There appears to be no simple species-typical pattern of human female proceptivity or receptivity that links sexual desire to hormonal activity. Humans are not unique in this regard—sexual receptivity in other primates varies across species, and between and within groups (Rowell, 1972), and is clearly influenced by social, as well as hormonal factors (Keverne, 1976). Despite confusion as to exactly when peaks in sexual interest occur, two studies report that women with partners may be more inclined to seek sex outside the pair bond when conception following sex is most likely. Baker and Bellis (1990, 1995) report data indicating that the rate of female extrapair copulations (EPCs) is around 2.5 times higher during the follicular phase than in the luteal phase, which possibly promotes sperm competition when conception is likely. A decrease in EPC frequency during nonfertile phases of the menstrual cycle may reduce the risk of being discovered, and hence, the risk of desertion by the cuckolded male. Baker and Bellis’ study has generated much controversy as the data were collected from a questionnaire in Company magazine, introducing the possibility of some self-selection bias among participants. A further study (Worthman, 1978; reported in Hrdy, 1981) supports the findings in a hunter–gatherer society, using hormonal measurements to determine menstrual status. Worthman reports that there is a midcycle increase in sexual activity with both husbands and lovers among women of the !Kung San. A series of studies of women’s clothing in Viennese discotheques (e.g., Grammer et al., 1997) indicate that women dressed in tighter, more revealing clothes in the ovulatory phase of the menstrual cycle (as revealed by estradiol assays). As such clothing increases women’s self-rated “sexiness” in the context of a nightclub, this may be an example of proceptive female sexual behavior related to cyclic hormonal fluctuations.

242

IAN S. PENTON-VOAK AND DAVID I. PERRETT

C. CHANGES IN FEMALE PREFERENCES FOR MALE SCENT ACROSS THE MENSTRUAL CYCLE Some evidence also suggests that female preferences for male characteristics may change across the menstrual cycle. Fitness benefits to offspring can only be realized if conception follows copulation, so females may be more attentive to phenotypic markers indicating fitness during the follicular phase of the menstrual cycle when conception is most likely (days 6–14; Baker and Bellis, 1995; Regan, 1996; Gangestad and Thornhill, 1998). Female preferences for odor seem to have a cyclic component, with females being more receptive to odors at ovulation, coincident with increases in estrogen and luteinizing hormone (Doty et al., 1981). Odor cues appear to play an important role in human mate choice, and women report odor to be a more important cue in mate choice than men (Herz and Cahill, 1997). In addition, there is some evidence that women are able to make biologically significant judgments of male odor that may be fitness relevant. Wedekind et al. (1995) performed a “T-shirt experiment,” in which males wore clean T-shirts for 2 nights. The worn shirts were then presented to female subjects, who rated their odor. Both female (the smellers) and male (the smelled) participants were typed for various genes in the major histocompatibility complex (MHC). Women rated the odors of men who had dissimilar MHC genes to themselves as preferable to MHC-similar men, which might maximize immune system function in offspring through heterozygosity (Wedekind and Furi, 1997). Although comparisons across various phases of the menstrual cycle were not made in this experiment, all women were tested around the 12th day of their menstrual cycle, coinciding with the occurrence of peak olfactory sensitivity and peak fertility. These preferences for men differing in immune system genotypes were reversed in women using oral contraception. Grammer (1993) studied female perception of androstenone, a chemical structurally related to testosterone that is found mainly in male rather than female sweat and urine. In contrast to the closely related androstenol, which has a pleasant, sandalwoodlike smell that is attractive to females, androstenone has an unplesant, urinelike odor. The exact mechanism of androstenone and androstenol production is unclear: they may both be byproducts of enzyme action on bacteria, or androstenol may oxidize to androstenone (see Grammer, 1993, for a brief review). Grammer 1993 showed that the generally unpleasant odor of androstenone becomes more acceptable in midcycle to women who do not use hormonal contraception. Furthermore, androstenol influences female mood at midcycle: females who placed a drop of androstenol on their top lip each morning for a month reported themselves to be more submissive than a placebo group at midcycle (Benton, 1982; see also Kirk-Smith et al., 1978).

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

243

1. The Scent of Symmetry? Three other studies of women’s olfactory preferences across the cycle show that women may also make discriminations of male quality in the follicular phase of the menstrual cycle that they do not make at other times (Gangestad and Thornhill, 1998; Rikowski and Grammer, 1999; Thornhill and Gangestad, 1999). These studies employed similar methodologies to Wedekind et al. (1995) in which females rated the odor of T-shirts that had been worn by males for 2 nights. Although these studies are concerned with signaling qualities of male odor, unlike the studies mentioned above no specific chemical is proposed as a signal. Instead these studies concentrate on bodily and facial symmetry of the male T-shirt weavers. All three studies demonstrate that women are sensitive to and prefer the scent of males who are symmetrical, through some as yet undetermined chemical signal. This discrimination, however, appears to be influenced by hormonal status across the menstrual cycle. In the first of these scent of symmetry experiments (Gangestad and Thornhill, 1998), the level of fluctuating asymmetry of 41 men had been estimated from measurements of digit length, elbow width, wrist width, ankle width, foot breadth, and ear length and width. These men wore T-shirts for 2 nights before they were returned for rating by 45 female participants, 17 of whom used oral hormonal contraception. The nonpill using women were split into two groups according to menstrual cycle phase (with the follicular phase labeled “high conception risk,” and menses and the luteal phase characterized as “low conception risk”). Men’s fluctuating asymmetry significantly predicted the attractiveness ratings given to their scent by “high” but not by “low” conception risk women, and had no relationship with the ratings of women on the pill. Thornhill and Gangestad (1999) replicate their original finding with a larger sample (74 males, 78 females) and better controls of personal hygiene. Additionally, this more recent study showed that women prefer the scent of men whose facial photographs are rated as attractive at times of high fertility. Men, on the other hand, showed no preference for the odor of symmetrical females. Rikowski and Grammer (1999) replicated the findings of Thornhill and Gangestad in a similar study in Austria. Sixteen males, whose body asymmetry had been estimated from 7 bilateral body traits and whose facial asymmetry had been estimated from the position of bilateral facial structures (see Grammer and Thornhill, 1994), wore T-shirts for 3 nights. Forty female raters (14 of whom were estimated to be in the follicular phase when tested) assessed the odor of the worn shirts. Overall, women in the follicular phase rated male odor as sexier overall than women in other menstrual cycle phases. In addition, it was found that the ratings of male odor “sexiness”

244

IAN S. PENTON-VOAK AND DAVID I. PERRETT

given by women in the follicular phase of their menstrual cycle correlated significantly with bodily symmetry and nonsignificantly with facial symmetry and facial attractiveness (r = 0.48 and 0.47, respectively, p < 0.1). These relationships were not present in women in other, low conception risk phases of the menstrual cycle. 2. Interpretation of Cyclic Preferences These cyclic changes have been interpreted from a “good genes” perspective. Symmetry is a marker of (possibly) heritable developmental stability and is a correlate of secondary sexual trait size in some species (Møller and Hoglund, 1991; Andersson, 1994), and hence of high level of androgens in mammals (Thornhill and Gangestad, 1996; Scheib et al., 1999). A preference for symmetric men when conception is most likely may be adaptive, increasing offspring viability. In combination with other evidence, such a mechanism may be especially important in extra-pair sexual activity. Females prefer symmetrical men as extra-pair copulation partners, and evidence reviewed above indicates that EPC activity peaks in the follicular phase (Thornhill and Gangestad, 1994; Baker and Bellis, 1995; Gangestad and Thornhill, 1997). As symmetry appears to be inversely related to the investment that men make in relationships, such choices probably do not reflect selection for paternal effort (Thornhill and Gangestad, 1997, 1998, 1999). D. CYCLIC SHIFTS IN PREFERENCES FOR FACES Cyclic changes analogous to those found for odors have been shown to operate in the visual modality, specifically in preferences for faces. Frost (1994) developed photographs of 3 pairs of male faces to manipulate the skin darkness. Skin color is sexually dimorphic within all races, with males having darker skin than females (Van Den Berghe and Frost, 1986). This may be due to a link between the production of melanin and that of gonadotropins by the anterior pituitary. Frost’s study showed that French Canadian women (N = 98) preferred the lighter face, but for women not using oral contraception (N = 56) this preference was attenuated when the estrogen/progesterone ratio was high—the follicular phase of the menstrual cycle. Breakdown of women by percentage of cycle completed shows a peak around ovulation in the number of dark faces preferred. Given the sexual dichromatism of skin color, a preference for darker skin at ovulation could be considered a preference for exaggeration of a male trait at ovulation. 1. Cyclic Preferences for Masculinity in Male Face Shapes The findings that traits relating to possible male “quality” are apparently more attractive to women at high-fertility phases of the menstrual cycle,

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

245

coupled with Frost’s evidence for female cyclic visual preferences for male skin color, suggested that a study of cyclic female preferences for male facial shape might be fruitful. We performed a series of experiments in which composite male face stimuli with manipulated levels of masculinity and femininity were presented to women whose menstrual phase was estimated. We hypothesized that females may be attracted to more masculine faces with exaggerated secondary sexual facial characteristics when conception is most likely, that is, in the follicular phase of the menstrual cycle. This prediction was supported by the three experiments briefly described below. 2. A United Kingdom–Based Questionnaire Study We presented 5 male stimuli with differing levels of masculinity/femininity (50 and 30% feminized, the average, and 30 and 50% masculinized) in the BBC Tomorrow’s World magazine, along with a short questionnaire with details of the female respondents’ age, use of oral contraception, and number of days since the onset of their last period. Of 178 women who completed all of the relevant sections of the questionnaires, 39 used oral contraception. Following Gangestad and Thornhill (1998), a standard 28-day model of the female menstrual cycle was used to assign the 139 remaining female respondents (mean age of 30.7 years with a range of 14–50) to one of two groups based on their chance of conception. This was estimated from the number of days since the onset of the participants last menses: high conception risk, (N = 55; the follicular phase, days 6–14) or low conception risk (N = 84; menses, days 0–5, and the luteal phase, days 15–28). Figure 5 shows the percentage of women in each of the two groups who selected each of the 5 available faces as most attractive. Those in the lowconception risk group effectively picked faces at chance levels (though showing a tendency to prefer feminized faces), whereas the high risk group showed a significant preference for masculinized faces (Penton-Voak and Perrett, 2000). These preliminary data indicated that women are attracted to exaggerated male traits when conception following coitus is most likely (the follicular phase) and not at other times of the menstrual cycle. 3. A Longitudinal Study of Japanese Females A second experiment aimed to address some of the methodological weaknesses of the questionnaire study. Instead of a single-shot methodology, a repeated-measures design was employed, allowing the same women to be tested in different phases of their menstrual cycle. This study was conducted in Japan with Japanese students, and used male facial stimuli sets of both Japanese and white British origin. Oral contraception alters the hormonal

246

IAN S. PENTON-VOAK AND DAVID I. PERRETT

FIG. 5. Percentage of subjects in high- and low-conception risk groups selecting each of the 5 stimuli (with varying levels of masculinity and femininity) as most attractive in the U.K. questionnaire study. From Penton-Voak and Perrett (2001).

processes of the menstrual cycle and appears to disrupt cyclic and other preferences (e.g., Wedekind et al., 1995; Thornhill and Gangestad, 1999). At the time of the study, oral contraception was unavailable in Japan. The young Japanese women tested here are very unlikely to have used hormonal contraception and hence provide a good population with which to test hypotheses based on normal cyclic changes in hormonal activity. If cyclic preferences are an adaptation to maximize fitness in offspring through extra-pair copulations and such shifts represent the interaction of hormonal and social factors, it is possible that women within relationships may manifest different patterns of response than those without a partner. Previous research suggests that women in all cultures value positive personality characteristics in a partner above all else (Buss, 1989). Research has indicated that stereotypical personality characteristics do indeed influence attractiveness judgments (Perrett et al., 1998; see review earlier in chapter). Once women have secured a partner, however, those seeking extra-pair partners may be influenced by considerations other than personality and future paternal investment. For example, cues to possibly heritable quality (as signaled by androgenic effects on face shape) may become more important. This experiment investigated such a possibility by presenting two sets (a Caucasian male average, and a Japanese male average) of 5 stimuli (40 and 20% masculinized, the average, and 20 and 40% feminized) to 39 Japanese students (Penton-Voak et al., 1999b). The participants took part in two experimental sessions, one in the follicular phase of their menstrual cycle, and one

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

247

FIG. 6. Effects of conception risk on femininity preferred by Japanese subjects in Japanese and Caucasian faces (top, from Penton-Voak et al., 1999b) and for subjects with (n = 20) and without (n = 19) a partner (below). Mean and SE illustrated.

in the luteal phase, as estimated from self-reports of average cycle length. In addition, participants reported whether they were currently in a “steady” heterosexual relationship. Overall, subjects preferred feminized faces (a replication of Perrett et al., 1998), and the origin of stimuli (Japanese or Caucasian) had no effect. As in the questionnaire experiment, women in the follicular phase showed a preference for significantly less-feminized faces (Fig. 6). Although cross-cultural consistency does not prove that a behavior is an adaptation (ecological pressures may cause local adaptations), crosscultural replication adds some weight to an evolutionary interpretation of the results. Although no significant effects or interactions were found when the relationships of participants were analyzed, there was a tendency for women in relationships to prefer more masculine faces in general ( p = 0.066; Fig. 6) and to show a greater follicular phase shift toward masculinity than women without partners ( p = 0.08; Fig. 6). Both of these trends are consistent with an extra-pair copulation interpretation of cyclic preferences: Once a main

248

IAN S. PENTON-VOAK AND DAVID I. PERRETT

partner has been acquired, women may be less concerned with perceived personality indicating beneficial paternal behavior, and more concerned with cues to genetic quality in possible extra-pair partners (i.e., those features associated with testosterone). Needless to say, conscious awareness of such concerns is not necessarily implicated here. These cues may be especially important when extra-pair copulations are most likely to occur and when conception is most likely during the follicular phase of the menstrual cycle. 4. A United Kingdom–based Longitudinal Study Another laboratory-based within-subjects experiment tested the preferences of U.K. university students across their menstrual cycles (Penton-Voak et al., 1999b). An interactive methodology was used allowing female undergraduates to alter the apparent masculinity and femininity of the five male facial continua (four Caucasian and one Japanese). To further investigate the possible role of cyclic preference changes in short-term or possibly extrapair sexual relationships, participants who were not using oral contraception were asked to make one of two attractiveness judgments: to pick a face that would be attractive in a “long-term partner” (N = 26), or a face that would be attractive in a “short-term sexual partner” (N = 23). In this sample, cyclic shifts favoring relative masculinity in the follicular phase only occurred in females judging attractiveness for a short-term sexual partner; preferences for a long-term partner did not differ significantly across the cycle (Fig. 7, Penton-Voak et al., 1999b).

FIG. 7. U.K. longitudinal study data. Effects of relationship context and conception risk on mean femininity preferred in five male facial continua. From Penton-voak et al., 1999b. Mean and SE illustrated.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

249

Females asked to pick a long-term partner in this experiment always preferred the same level of feminization in male face shapes across the menstrual cycle; such shapes elicit favorable personality attributions that may be related to behavior as discussed above. When likely paternal investment is low, however, as in short-term sexual encounters or extra-pair copulations, more weight is given to characteristics that may indicate “good genes” in partners, through facial traits thought to be linked to androgens. To some extent, this finding reconciles the overall preference for femininity in male faces with other literature on male sexual activity. As reported earlier, dominant (presumably high testosterone) men copulate earlier (Mazur et al., 1994) and achieve more sexual access as adults (Perusse, 1993). Townsend et al. (1995) report that high-status male college students, such as athletes, often reported more than 100 sex partners. Such a number of partners obviously reflects low male investment in each female. It has been suggested that this low investment in a given woman reflects the quality of the male involved: they are more attractive to women and can therefore obtain more partners, despite the women’s alleged desire for a monogamous relationship (e.g., Symons, 1979, and pers. comm.; Townsend et al., 1995; Gangestad and Thornhill, 1997; Wiederman and Dubois, 1998). Nonetheless, it is widely accepted that biparental care is an adaptation: men and women desire a long-term pair bond as it has led in the past to greater reproductive success. Men do not have to be tricked into socially monogamous relationships: strategies of zero male parental investment will be selected against if they result in lower reproductive success (Burley, 1979). High-testosterone men apparently fail either to attempt or to succeed in marital relationships (Mazur and Booth, 1998). The data of our U.K.-based longitudinal cycle study imply that higher testosterone men may be favored for short-term sexual relationships when conception is likely, whereas men with faces that indicate relatively lower testosterone, with their concomitant likelihood of extended paternal investment, are preferred for long-term relationships. At present, however, this hypothesis clearly relies on several assumptions that remain untested. In two of three of these cycle/face preference experiments, and all of our previous work using composite stimuli, we have found an overall preference for more feminine than average faces. Facial masculinity, however, is associated with dominance and some aspects of reproductive success such as sexual access (as reviewed above). Access to reproductive opportunities will reflect diverse influences. Male–male competition may favor different characteristics to those favored by female choice. Nonetheless, success in male–male competition and the accompanying status and dominance advantages that it bestows will be another factor influencing judgments of attractiveness in real-life situations.

250

IAN S. PENTON-VOAK AND DAVID I. PERRETT

E. CYCLIC PREFERENCES, MIXED FEMALE MATING STRATEGIES, AND “GOOD GENES” The experiments summarized above found evidence of a cyclic preference for male face shape: women exhibit preferences for biologically relevant aspects of facial structure that may signal heritable genetic characteristics when conception is most likely. The results are also consistent with other cyclic preferences relevant to mate choice. Menstrual phase may be one variable that contributes to the difficulty of defining what females find attractive in male faces. Female sexual interest has a peak at ovulation, and preferences for men honestly advertising immunocompetence at this time may be adaptive. In spite of this, women are potentially sexually receptive across the menstrual cycle. Sex when conception is unlikely probably serves purposes not directly linked to fertilization (Hrdy, 1981). Our work, and the work of other researchers into the influence of face shape on personality, indicates that dominance and quality as a parent are attributions made to opposite ends of the continuum that relates to facial masculinity and femininity (or mature and babyfaces). Both masculinity and femininity may be associated with behaviors that have costs and benefits to reproductive success (Perrett et al., 1998; Berry and Wero, 1993). A preference for males with a more masculine (i.e., high testosterone) appearance when conception is most likely may confer benefits for offspring in terms of heritable immunocompetence but also costs due to potentially decreased paternal investment. Femininity in male faces may be associated with the opposite collection of characteristics. In humans, paternity uncertainty results from internal fertilization, concealed ovulation, limited visual similarity between offspring and their fathers, and the apparently cross-cultural finding that couples prefer to copulate clandestinely (Schroder, ¨ 1993; Pagel, 1997; Christenfeld and Hill, 1995). Rates of extra-pair paternity are certainly nonzero, although well-controlled studies are the exception rather than the rule (Macintyre and Sooman, 1991). As with some other species (Graves et al., 1993; Andersson, 1994), human females may have been selected to pursue a mixed mating strategy in some circumstances. Adaptations allow female sexual behavior to serve multiple functions in addition to fertilization, and different male characteristics may indicate different potential benefits. Some females may choose a primary partner whose relatively low masculine appearance suggests cooperation in parental care, while occasionally pursuing extra-pair copulations with males with a relatively masculine appearance when conception is most likely. Potentially, sexual behavior arising from cyclic preferences may provide the benefits of polyandry (genetic diversity in offspring, good genes benefits in offspring, encouraging competition at the level of sperm) without losing

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

251

the advantages of ostensive monandry (such as extended paternal investment), thus realizing a reproductive advantage that may have been favored by selection. The behavioral adaptation that allows such mixed mating strategies is the female ability to copulate across the menstrual cycle, allowing most sexual encounters to serve purposes not directly involved in fertilization. For the majority of time, women appear to prefer smaller than average male facial secondary sexual characteristics. This is in sharp contrast to the many species in which female choice of male characteristics reveals preferences for exaggerated male traits (Andersson, 1994). Most species, however, copulate only when conception is likely during brief periods of female receptivity. In social species (such as many primates), with long-term, reciprocal relationships between individuals, females may have adapted to benefit from nonreproductive sex. Perhaps, then, we should not be surprised if the criteria of male attractiveness differ in these species. The results of the studies reported here have several caveats that can be addressed by future research. First, confirmation of an overall preference for femininity and cyclic shifts in such preferences using real rather than composite faces would add strength to both hypotheses. Facial-metric measurements of such stimuli may also isolate the features driving attractiveness judgments and allow multiple-motives models (e.g., Cunningham et al., 1990) to be tested thoroughly. Cyclic shifts, coupled with context-dependent mate choice criteria, may explain the difficulty of defining male facial attractiveness and the inconsistencies in the current literature. Although the preferences we have described in this chapter and in earlier papers generalize to participants in the United Kingdom and Japan, further cross-cultural work is needed to investigate the possible consequences of local environmental conditions. Yu and Shepard (1998) indicate that male preferences for low female waist-to-hip ratios (e.g., Singh, 1993) are not universal, as previously thought. Similarly, overall preferences for feminized male faces may not be universal: in societies living under the pressure of a higher parasite load than modern United Kingdom and Japan, possible indicators of good genes (i.e., testosterone-dependent features) may be given more weight in attractiveness judgments (Gangestad and Buss, 1993). Pathogen prevalence appears to be one factor influencing human mating systems (Low, 1990). Fashion and media coverage also seem likely to influence judgments of male attractiveness in Japanese and British society: contemporary film stars, musicians and other male icons may be relatively feminine in comparison with past stars of stage and screen. However, in studies examining the preferences of a more diverse sample of women than undergraduate students, subject age has not influenced attractiveness judgments of male faces (e.g.,

252

IAN S. PENTON-VOAK AND DAVID I. PERRETT

Penton-Voak and Perrett, 2001). This indicates that even if older women once preferred the more rugged stars of their youth, the recent crop of boyish idols has influenced their current preferences. Additionally, it seems unlikely that cyclic preferences result from cultural factors. The possibility, however, that cultural fads have a radical impact on mate choice cannot be ignored. Further cross-cultural research may shed some light on such issues. Finally, and perhaps most importantly, further studies of actual partner choice must be conducted. Although the timing of extra-pair sex, and the characteristics of men chosen for such extra-pair sex, appear consistent with the hypotheses proposed here (Baker and Bellis, 1995; Gangestad and Thornhill, 1997), it is unclear whether the small shifts in preferences that we report here influence actual behavior, and hence have biological significance. Ethological studies are necessary to expand our knowledge of human sexual behavior.

V. SUMMARY On the evidence of experimental studies, female preferences for even static images of male faces represent a complex set of decision-making processes, and the differing techniques employed by different researchers often generate conflicting results. The findings reported by Perrett et al. (1998) indicate that women do not have clear preferences for masculinized (high testosterone) face shapes as predicted by indicator models of sexual selection, and some other studies of male faces (e.g., Grammer and Thornhill, 1994). Stereotypical personality judgments attributed to static faces appear to influence attractiveness judgments. Masculinized faces (indicating high levels of androgens) are considered to possess fewer desirable personality traits than feminized faces. These attributions may have some validity, and a reasonable (although, as yet unsupported) biological model linking androgen levels, behavior, and facial shape fits in with the observed preference pattern and, apparently, the fossil record: an overall preference for relatively low-testosterone men may be a somewhat unexpected adaptation. Biological facial characteristics that are considered putative indicators of good genes are not, however, ignored. They appear to be appraised in the light of stereotypical personality judgments and in the context of personal life-history factors (such as the type of relationship sought and possibly the relationship status of the woman). Furthermore, preferences are mediated and interact with cyclic hormonal changes linked to the likelihood of conception following sex. Relative masculinity in faces seems to be preferred at

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

253

times when conception is most likely, paralleling other researchers’ work in the olfactory modality. These preferences have been found in two cultures (United Kingdom and Japan), with student-aged and older women, and with multiple stimulus sets. Cyclic preferences for male face shapes are consistent with other reported cyclic preferences for male odor. If preference data are reliably linked to actual sexual behavior (a question that should be addressed by future research), a model linking likelihood of parental investment and facial masculinity preferred by females is suggested: when parental investment is sought (i.e., for a long-term relationship) facial shapes associated with relatively lower testosterone levels are preferred. Such faces may reliably indicate prosocial personality characteristics. When, however, likelihood of parental investment is low (in short-term relationships or possibly extra-pair copulations when the likelihood of conception is high) relatively more masculine faces are preferred, in a fashion more consistent with “good genes” hypotheses.

Acknowledgments IPV was supported by an ESRC Ph.D. studentship. This work was supported by project grants to DP from Unilever Research and the ESRC-ROPA. We are indebted to the help from D. Carrington at the BBC Tomorrow’s World magazine, D. Castles at the Hasegawa Laboratory, University of Tokyo, and S. Yoshikawa at the Department of Cognitive Psychology in Education, University of Kyoto. We thank A. Whiten, R. Byrne, M. Goodale, J. Graves, R. Thornhill, and T. Roper for comments and suggestions on much of the work presented here.

References Adams, D., Burt, A., and Gold, A. R. (1978). Rise in female-initiated sexual activity at ovulation and its suppression by oral contraception. N. Engl. J. Med. 299, 1145–1150. Albright, L., Kenny, D. A., and Malloy, T. E. (1988). Consensus in personality judgments at zero acquaintance. J. Pers. Soc. Psychol. 55(3), 387–395. Albright, L., Malloy, T. E., Dong, Q., Kenny, D. A., Fang, X., Winquist, L., and Yu, D. (1997). Cross-cultural consensus in personality judgments. J. Pers. Soc. Psychol. 72(3), 558–569. Alexander, R. D., and Noonan, K. M. (1979). Concealment of ovulation, parental care and human social evolution. In “Evolutionary Biology and Human Social Behavior” (W. N. Chagnon & Irons, ed.), Duxbury Press, Massachusetts. Alley, T. R., and Cunningham, M. R. (1991). Average faces are attractive, but very attractive faces are not average. Psychol. Sci. 2(2), 123–125. Ambady, N., Hallahan, M., and Rosenthal, R. (1995). On judging and being judged accurately in zero-acquaintance situations. J. Pers. Soc. Psychol. 69(3), 518–529. Andersson, M. (1994). “Sexual Selection.” Princeton Univ. Press, New Jersey. Baker, R. R., and Bellis, M. A. (1995). “Human Sperm Competition: Copulation, Masturbation and Infidelity,” 1st ed. Chapman & Hall, London.

254

IAN S. PENTON-VOAK AND DAVID I. PERRETT

Bancroft, J., Sanders, D., Davidson, D., and Warner, P. (1983). Mood, sexuality, hormones, and the menstrual cycle. 3. Sexuality and the role of androgens. Psychosom. Med. 45, 509– 516. Barrett, L. F., and Pietromonaco, P. R. (1997). Accuracy of the five-factor model in predicting perceptions of daily social interactions. Pers. Soc. Psychol. Bull. 23, 1173–1187. Beach, F. (1974). Human sexuality and evolution. In “Reproductive Behavior” (W. Montagna and W. Sadler, eds.) Plenum, New York. Bellis, M. A., and Baker, R. R. (1990). Do females promote sperm competition? Data for humans. Anim. Behav. 40, 997–999. Benshoof, L., and Thornhill, R. (1979). The evolution of monogamy and concealed ovulation in humans. J. Soc. Biol. Struct. 2, 95–106. Benson, P. (1993). Perception and recognition of computer-enhanced facial attributes and abstracted prototypes. Ph.D. Thesis, University of St. Andrews. Benson, P., and Perrett, D. (1992). Face to face with the perfect image. N. Sci. 133, 32–35. Benton, D. (1982). The influence of androstenol—a putative human pheromone—on mood throughtout the menstrual cycle. Biol. Psychol. 15, 249–256. Berry, D. (1991). Attractive faces are not all created equal: Joint effects of facial babyishness and attractiveness on social perception. Pers. Soc. Psychol. Bull. 17(5), 523–533. Berry, D. S., and Brownlow, S. (1989). Were the physiognomists right? Personality correlates of facial babyishness. Pers. Soc. Psychol. Bull. 15, 266–279. Berry, D. S., and Landry, J. C. (1997). Facial maturity and daily social interaction. J. Pers. Soc. Psychol. 72, 570–580. Berry, D. S., and McArthur, L. Z. (1985). Some components and consequences of a babyface. J. Pers. Soc. Psychol. 48(2), 312–323. Berry, D. S., and McArthur, L. Z. (1988). What’s in a face? Facial maturity and the attribution of legal responsibility. Pers. Soc. Psychol. Bull. 14(1), 23–33. Berry, D. S., and Wero, J. L. F. (1993). Accuracy in face perception: A view from ecological psychology. Ecol. Psychol. 497–519. Berry, D. S., and Zebrowitz McArthur, L. (1986). Perceiving character in faces: The impact of age-related craniofacial changes on social perception. Psychol. Bull. 100, 3–18. Blaffer Hrdy, S., and Whitten, P. (1987). Patterning of sexual activity. In “Primate Societies” (B. B. Smutts, D. L. Cheney, R. M. Seyfarth, R. Wrangham, and T. T. Strusaker, eds.), pp. 370–384. Univ. Chicago Press, Illinois. Bond, C. F., Berry, D. S., and Omar, A. (1994). The kernal of truth in judgements of deceptiveness. Basic Appl. Soc. Psychol. 15, 523–534. Booth, A., and Dabbs, J. (1993). Testosterone and men’s marriages. Soc. Forces 72, 463–477. Borkenau, P., and Liebler, A. (1992). Trait inferences: Sources of validity at zero acquaintance. J. Pers. Soc. Psychol. 62(4), 645–657. Botwin, M. D., Buss, D. M., and Shackleford, T. K. (1997). Personality and mate preferences: Five factors in Mate selection and marital satisfaction. J. Pers. 65(1), 107–136. Brace, C. L., Rosenberg, K. R., and Hunt, K. D. (1987). Gradual change in human tooth size in the late pleistocene and post-pleistocene. Evolution 41, 705–720. Brandt, A. (1980). Face reading: The persistence of physiognomy. Psychol. Today (Dec.) 90–96. Burley, N. (1979). The evolution of concealed ovulation. Am. Nat. 114, 835–858. Burt, A. (1992). “Concealed ovulation” and sexual signals in primates. Folia Primatol. 58, 1–6. Burt, D. M., and Perrett, D. I. (1995). Perception of age in adult Caucasian male faces: Computer graphic manipulation of shape and colour information. Proc. R. Soc. London B 259, 137–143. Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behav. Brain Sci. 12, 1–49.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

255

Buss, D. M., and Barnes, M. (1986). Preferences in human mate selection. J. Pers. Soc. Psychol. 50, 559–570. Christenfeld, N., and Hill, E. A. (1995). Whose baby are you? Nature 378, 669. Cleeton, G. U., and Knight, F. B. (1924). Validity of character judgement based on external criteria. J. Appl. Psychol. 8, 215–231. Cunningham, M. R. (1986). Measuring the physical in physical attractiveness: Quasiexperiments on the sociobiology of female facial beauty. J. Pers. Soc. Psychol. 50(5), 925– 935. Cunningham, M. R., Barbee, A. P., and Pike, C. L. (1990). What do women want? Facialmetric assessment of multiple motives in the perception of male facial physical attractiveness. J. Pers. Soc. Psychol. 59(1), 61–72. Daly, M., and Wilson, M. (1983). “Sex, Evolution and Behavior” 2nd ed. PWS, Boston. Daly, M., and Wilson, M. I. (1999). Human evolutionary psychology and animal behaviour. Anim. Behav. 57, 509–519. Dawkins, R. (1976). “The Selfish Gene.” Oxford Univ. Press, UK. Dixson, A. F. (1983). Observations on the evolution and behavioral significance of “sexual skin” in female primates. Adv. Study Behav. 13, 63–106. Doty, R. L., Snyder, P. J., Huggins, G. R., and Lowry, L. D. (1981). Endocrine, cardiovascular, and psychological correlates of olfactory sensitivity changes during the human menstrual cycle. J. Comp. Physiol. Psychol. 95, 45–60. Folstad, I., and Karter, A. J. (1992). Parasites, bright males and the immunocompetence handicap. Am. Nat. 139, 603–622. Ford, C. S., and Beach, F. A. (1951). “Patterns of Sexual Behavior.” Eyre and Spottiswood, London. Frost, P. (1994). Preference for darker faces in photographs at different phases of the menstrual cycle: Preliminary assessment of evidence for a hormonal relationship. Percept. Motor Skills 79, 507–514. Gage, A. R., Manning, J. T., Scutt, D., Diver, M. J., and Fraser, W. D. (1999). Testosterone, cortisol and jaw size in men. Galton, F. J. (1878). Composite Portraits. Anthropological Institute of Great Britain and Ireland. 8, 132–142. Gangestad, S. W., and Buss, D. M. (1993). Pathogen prevalence and human mate preferences. Ethol. Sociobiol. 14, 89–96. Gangestad, S. W., and Thornhill, R. (1997). The evolutionary psychology of extrapair sex: The role of fluctuating asymmetry. Evol. Hum. Behav. 18(2), 69–88. Gangestad, S. W., and Thornhill, R. (1998). Menstrual Cycle Variation in Women’s Preferences for the scent of symmetrical men. Proc. R. Soc. London B 265, 927–933. Grammer, K. (1993). 5-alpha-androst-16en-3alpha-on: A male pheromone? A brief report. Ethol. Sociobiol. 14, 201–208. Grammer, K., and Thornhill, R. (1994). Human (Homo sapiens) facial attractiveness and sexual selection: The role of symmetry and averageness. J. Comp. Psychol. 108, 233–242. Grammer, K., Jutte, A., and Fischmann, B. (1997). Der Kampf der Geschlechter und der Krieg der Signale. In “Liebe, Lust und Leidenschaft. Sexualitat im Spiegal der Wissenschaft” (B. Kanitscheider, ed.), Hirzel. Stuttgart. Graves, J., Ortega-Ruano, J., and Slater, P. J. B. (1993). Extra-pair copulations and paternity in shags: Do females choose better males? Proc. R. Soc. London B 253, 3–7. Gray, J. P., and Wolfe, L. D. (1983). Human female sexual cycles and the concealment of ovulation problem. J. Soc. Biol. Struct. 6, 345–352. Haaq, C. M., and Donahoe, P. K. (1998). Regulation of Sexual Dimorphism in Mammals. Physiol. Rev. 78(1), 1–33.

256

IAN S. PENTON-VOAK AND DAVID I. PERRETT

Harcourt, A. H. (1996). Sexual selection and sperm competition in primates: What are the male genitalia good for? Evol. Anthropol. 4, 121–129. Hausler, M., and Schmid, P. (1995). Comparison of the pelves of Sts 14 and AL 288-1: Implications for birth and sexual dimorphism in australopithecines. J. Hum. Evol. 29, 363–383. Herz, R. S., and Cahill, E. D. (1997). Differential use of sensory information in sexual behavior as a function of gender. Hum. Nat. 8, 275–286. Hillgarth, N., and Wingfield, J. C. (1997). Testosterone and immunosuppression in vertebrates: Implications for parasite mediated sexual selection. In “Parasites and Pathogens” (N. E. Beckage, ed.). Chapman and Hall, New York. Hrdy, S. B. (1981). “The Woman That Never Evolved.” Harvard Univ. Press, Cambridge, MA. Hume, D. K., and Montgomerie, R. (1999). Facial attractiveness and symmetry in humans. Paper presented at the Human Behavior and Evolution Society, Utah. Jochle, ¨ W. (1973). Coitus-induced ovulation. Contraception 7(6), 523–564. Johnston, V. S., and Franklin, M. (1993). Is beauty in the eye of the beholder? Ethol. Sociobiol. 14, 183–199. Jones, D. (1995). Sexual selection, Physical attractiveness, and facial neoteny. Curr. Anthropol. 36(5), 723–748. Jones, D., and Hill, K. (1993). Criteria of facial attractiveness in five populations. Hum. Nat. 4(3), 271–296. Kalick, S. M., Zebrowitz, L. A., Langlois, J. H., and Johnson, R. M. (1998). Does human facial attractiveness honestly advertise health? Longitudinal data on an evolutionary question. Psychol. Sci. 9(1), 8–13. Keating, C. F. (1985). Gender and the physiognomy of dominance and attractiveness. Soc. Psychol. Q. 48, 61–70. Kenny, D. A., Horner, C., Kashy, D. A., and Chu, L. (1992). Consensus at zero acquaintance: Replication, behavioral cues and stability. J. Pers. Soc. Psychol. 62, 88–97. Kenny, D. A., Albright, L., Malloy, T. E., and Kashy, D. A. (1994). Consensus in interpersonal perception: Acquaintance and the big five. Psychol. Bull. 116, 245–258. Keverne, E. B. (1976). Sexual receptivity and attractiveness in the female rhesus monkey. Adv. Study Behav. 7, 155–200. Kirk-Smith, M., Booth, D., Carroll, D., and Davies, P. (1978). Human social attitudes affected by androstenol. Res. Commun. Psychol. Psychiatr. Behav. 3, 379–384. Koeslag, J., and Koeslag, P. (1994). Koinophilia. J. Theor. Biol. 167, 55–65. Koeslag, J. H. (1994). Koinophilia replaces random mating in populations subject to mutations with randomly varying fitnesses. J. Theor. Biol. 171, 341–345. Kowner, R. (1996). Facial asymmetry and attractiveness judgment in developmental perspective. J. Exp. Psychol.: Hum. Percept. Perform. 22(3), 662–675. Lancaster, J. (1975). “ Primate Behavior and the Emergence of Human Culture.” Holt, Reinhart, and Winston, New York. Langlois, J. H., and Roggman, L. A. (1990). Attractive faces are only average. Psychol. Sci. 1(1), 115–121. Langlois, J. H., Roggman, L. A., and Musselman, L. (1994). What is average and what is not average about attractive faces? Psychol. Sci. 5, 214–219. Liggett, J. (1974). “The Human Face.” Constable, London. Little, A. C. (1999). The influence of facial masculinity and distinctiveness on judgements of attractiveness in human male faces. M. S. Thesis, Department of Psychology. Stirling University. Lorenz, K. (1943). The innate forms of potential experience. Z. Tierpsychol. 5, 234–409. Low, B. S. (1990). Marriage systems and pathogen stress in human societies. Am. Zool. 30, 325–339.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

257

Macintyre, S., and Sooman, A. (1991). Non-paternity and prenatal genetic screening. Lancet 338, 869. Manning, J. T., Scutt, D., Whitehouse, G. H., and Leinster, S. J. (1997). Breast asymmetry and phenotypic quality in women. Evol. Hum. Behav. 18, 223–236. Matteo, S., and Rissman, E. F. (1984). Increased sexual activity during the midcycle portion of the human menstrual cycle. Horm. Behav. 18, 249–255. Mazur, A., and Booth, A. (1998). Testosterone and dominance in men. Behav. Brain Sci. 21, 353–371. Mazur, A., Halpern, C., and Udry, J. R. (1994). Dominant looking males copulate earlier. Ethol. Sociobiol. 15, 87–94. Mazur, A., Susman, E. J., and Edelbrock, S. (1997). Sex difference in testosterone response to a video game contest. Evol. Hum. Behav. 18, 317–326. McArthur, L., and Apatow, K. (1983–1984). Impressions of baby-faced adults. Soc. Cog. 2(4), 315–342. McArthur, L. Z., and Berry, D. S. (1987). Cross-cultural agreement in perceptions of babyfaced adults. J. Cross-cultural Psychol. 18(2), 165–192. McCrae, R. R., and Costa, P. T. (1987). Validation of the Five Factor model of personality across instruments and observers. J. Pers. Soc. Psychol. 52(1), 81–90. McHenry, H. M. (1994). Behavioral ecological implications of early hominid body size. J. Hum. Evol. 27, 77–87. Mealey, L., Bridgestock, R., and Townsend, G. (1999). Symmetry and perceived facial attractiveness. J. Pers. Soc. Psychol. 76, 151–158. Metcalf, M. G., Skidmore, G. F., Lowry, G. F., and Mackenzie, J. A. (1983). Incidence of ovulation in the years after the menarche. J. Endocrinol. 97, 213–219. Meyer, D. A., and Quong, M. W. (1999). The bio-logic of facial geometry. Nature 397, 661–662. Møller, A. P. (1997). Developmental stability and fitness: A review. Am. Nat. 149, 916– 942. Møller, A. P., and Hoglund, J. (1991). Patterns of fluctuating asymmetry in avian feather ornaments: Implications for models of sexual selection. Proc. R. Soc. London B 245, 1–5. Møller, A. P., and Thornhill, R. (1997). A meta-analysis of the heritability of developmental stability. J. Evol. Biol. 10, 1–16. Møller, A. P., and Thornhill, R. (1998). Bilateral Symmetry and Sexual Selection: A metaanalysis. Am. Nat. 151, 174–192. Morris, D. (1994). “The Naked Ape.” Vintage. Mueller, U., and Mazur, A. (1997). Facial dominance in Homo sapiens as honest signaling of male quality. Behav. Ecol. 8(5), 569–579. Murdock, G. P. (1967). “Ethnographic Atlas.” Univ. of Pittsburgh Press, Pennsylvania. Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes: Replicated factor structure in peer nominated personality ratings. J. Abnorm. Soc. Psychol. 66, 574– 583. Owens, I. P. F., and Short, R. V. (1995). Hormonal basis of sexual dimorphism in birds: Implications for new theories of sexual selection. Trends Ecol. Evol. 10(1), 44–47. Pagel, M. (1997). Desperately concealing father: A theory of parent-infant resemblance. Anim. Behav. 53, 973–981. Passini, F. T., and Norman, W. T. (1966). A universal conception of personality structure? J. Pers. Soc. Psychol. 4, 44–49. Penton-Voak, I. S., and Perrett, D. I. (2000). Female preference for male faces changes cyclically— Further evidence. Evol. Hum. Behav. 21, 39–48. Penton-Voak, I., Perrett, D., and Pierce, J. (1999a). Computer graphic studies of the role of facial similarity in attractiveness judgements. Curr. Psychol. 18, 104–117.

258

IAN S. PENTON-VOAK AND DAVID I. PERRETT

Penton-Voak, I. S., Perrett, D., Castles, D., Burt, M., Koyabashi, T., and Murray, L. K. (1999b). Female preferences for male faces change cyclically. Nature 399, 741–742. Perrett, D., and Penton-Voak, I. (1999). The Bio-logic of facial geometry. Nature 397, 662. Perrett, D. I., May, K. A., and Yoshikawa, S. (1994). Facial shape and judgements of female attractiveness. Nature 368, 239–242. Perrett, D. I., Lee, K. J., Penton-Voak, I. S., Rowland, D. R., Yoshikawa, S., Burt, D. M., Henzi, S. P., Castles, D. L., and Akamatsu, S. (1998). Effects of sexual dimorphism on facial attractiveness. Nature 394, 884–887. Perrett, D. I., Burt, D. M., Penton-Voak, I. S., Lee, K. J., Rowland, D. A., and Edwards, R. (1999). Symmetry and Human Facial Attractiveness. Evol. Hum. Behav. 20, 295–308. Perusse, D. (1993). Cultural and reproductive success in industrial societies: Testing the relationship at proximate and ultimate levels. Behav. Brain Sci. 16, 267–322. Plavcan, J. M., and van Schaik, C. P. (1997). Interpreting hominid behavior on the basis of sexual dimorphism. J. Hum. Evol. 32, 345–374. Regan, P. C. (1996). Rhythms of desire: The association between menstrual cycle phases and female sexual desire. Can. J. Hum. Sex. 5(3), 145–156. Rhodes, G., and Tremewan, T. (1996). Averageness, exaggeration and facial attractiveness. Psychol. Sci. 7(2), 105–110. Rhodes, G., Proffitt, F., Grady, J., and Sumich, A. (1998). Facial Symmetry and the perception of beauty. Psychonom. Bull. Rev. 5, 659–669. Rhodes, G., Sumich, A., and Byatt, G. (1999). Are average facial configurations attractive only because of their symmetry? Psychol. Sci. 10, 52–58. Rhodes, G., Byatt, G., Hickford, C., and Jeffery, L. (2001). Sex-typicality and facial attractiveness: Are supermale and superfemale faces super-attractive? Br. J. Psychol. Rikowski, A., and Grammer, K. (1999). Human body odour, symmetry and attractiveness. proc. R. Soc. London B 266, 869–874. Rowell, T. E. (1972). Female reproduction cycles and social behavior in primates. Adv. Study Behav. 4, 69–105. Ruff, C. B., Trinkaus, E., Walker, A., and Larsen, C. S. (1993). Postcranial robusticity in Homo. 1. Temporal trends and mechanical interpretation. Am. J. Phys. Anthropol. 91, 21–53. Samuels, C. A., Butterworth, G., Roberts, T., Graupner, L., and Hole, G. (1994). Facial aesthetics: Babies prefer attractiveness to symmetry. Perception 23, 823–831. Scheib, J. E., Gangestad, S. W., and Thornhill, R. (1999). Facial attractiveness, symmetry, and cues to good genes. Proc. R. Soc. London B. 266, 1913–1917. Schroder, ¨ I. (1993). Concealed ovulation and clandestine copulation: A female contribution to human evoluton. Ethol. Sociobiol. 14, 381–389. Silber, M. (1994). Menstrual cycle and work schedule: Effects of women’s sexuality. Arch. Sex. Behav. 23, 397–404. Sillen-Tulberg, ´ B., and Møller, A. P. (1993). The relationship between concealed ovulation and mating systems in anthropoid primates: A phylogenetic analysis. Am. Nat. 141(1), 1–25. Singh, D. (1993). Adaptive significance of female physical attractiveness: Role of waist-to-hip ratio. J. pers. Soc. Psychol. 65(2), 293–307. Stanislaw, H., and Rice, F. J. (1988). Correlation between sexual desire and menstrual cycle characteristics. Arch. Sex. Behav. 17, 499–508. Steklis, H. D., and Whiteman, C. H. (1989). Loss of estrus in human evolution: Too many answers, too few questions. Ethol. Sociobiol. 10, 417–434. Stewart, D. E. (1989). Positive changes in the pre-menstrual period. Acta Psychiatr. Scand. 79, 400–405. Strassman, B. I. (1981). Sexual selection, paternal care, and concealed ovulation in humans. Ethol. Sociobiol. 2, 31–40.

SHIFTING FEMALE PREFERENCES ACROSS THE MENSTRUAL CYCLE

259

Swaddle, J. P., and Cuthill, I. C. (1995). Asymmetry and human facial attractiveness: Symmetry may not always be beautiful. Proc. R. Soc. London 261, 111–116. Symons, D. (1979). “The evolution of human sexuality.” Oxford Univ. Press, UK. Symons, D. (1994). Beauty is in the adaptions of the beholder: The evolutionary psychology of human female sexual attractiveness. In “Sexual Nature/Sexual Culture” (P. R. P. Abramson, S. D., ed.). Univ. of Chicago Press, Illinois. Tague, R. G., and Lovejoy, C. O. (1998). AL 288-1—Lucy or Lucifer: Gender confusion in the Pliocene. J. Hum. Evol. 35, 75–94. Thornhill, R., and Gangestad, S. W. (1993). Human facial beauty: Averageness, symmetry, and parasite resistance. Hum. Nat. 4(3), 237–269. Thornhill, R., and Gangestad, S. W. (1994). Human fluctuating asymmetry and sexual behavior. Psychol. Sci. 5, 297–302. Thornhill, R., and Gangestad, S. W. (1996). The evolution of human sexuality. Trends Ecol. Evol. 11(2), 98–102. Thornhill, R., and Gangestad, S. W. (1999). The scent of symmetry: A human sex pheromone that signals fitness? Evol. Hum. Behav. 20, 175–202. Thornhill, R., and Gangestad, S. W. (1999). Facial Attractiveness. Trends Cog. Sci. 3, 452–460. Townsend, J. M., and Roberts, L. W. (1993). Gender differences in mate preference among law students: Divergence and convergence of criteria. J. Psychol. 127(5), 507–528. Townsend, J. M., and Wasserman, T. (1998). Sexual attractiveness: Sex differences in assessment and criteria. Evol. Human. Behav. 19, 171–191. Townsend, J. M., Kline, J., and Wasserman, T. H. (1995). Low investment copulation: Sex differences in motivations and emotional reactions. Ethol. Sociobiol. 16, 25–51. Udry, J. R., and Morris, N. (1968). Distribution of coitus in the menstrual cycle. Nature 200, 593–596. Urdy, R. J., and Morris, N. M. (1977). The distribution of events in the human menstrual cycle. J. Reprod. Fert. 51, 419–425. Van Den Berghe, P. L., and Frost, P. (1986). Skin color preference, sexual dimorphism and sexual selection: A case of gene culture co-evolution? Ethn. Rac. Studies 9(1), 87–113. Walster, E., Aronson, V., Abrahams, D., and Rottman, L. (1966). Importance of physical attractiveness in dating behavior. J. Pers. Soc. Psychol. 4(5), 508–516. Watson, D. (1989). Strangers’ rating of the five robust personality factors: Evidence of a surprising convergence with self-report. J. Pers. Soc. Psychol. 57, 120–128. Wedekind, C., and Furi, S. (1997). Body odour preferences in men and women: Do they aim for specific MHC combinations or simply heterozygosity? Proc. R. Soc. London B 264, 1471–1479. Wedekind, C., Seebeck, T., Bettens, F., and Paepke, A. J. (1995). MHC-dependent mate preferences in humans. Proc. R. Soc. London B 260, 245–249. Wiederman, M. W., and Dubois, S. L. (1998). Evolution and sex differences in preferences for short-term mates: Results from a plociy capturing study. Evol. Hum. Behav. 19, 153–170. Worthman, C. (1978). Psychoendocrine study of human behavior: some interactions of steroid hormones with affect and behavior in !Kung San. Ph.D. Thesis, Harvard University. Yu, D. W., and Shepard, G. H. (1998). Is beauty in the eye of the beholder? Nature 396, 321–322. Zahavi, A. (1975). Mate selection: A selection for a handicap. J. Theor. Biol. 53, 205–214.

This Page Intentionally Left Blank

ADVANCES IN THE STUDY OF BEHAVIOR, VOL. 30

The Control and Function of Agonism in Avian Broodmates HUGH DRUMMOND LABORATORIO DE CONDUCTA ANIMAL INSTITUTO DE ECOLOG´ıA

´ ´ UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO A.P. 70-275, 04510 D.F., MEXICO

I. INTRODUCTION Although much of the research into sibling aggression in the past 20 years has been framed as inquiry into the adaptiveness of within-family conflict, the links between the (largely untested) models of behavioral ecology and our factual knowledge are still tenuous (review in Mock and Parker, 1997). Meanwhile, almost unremarked, our understanding of the proximate control of agonism (aggression and responses thereto) in nestling birds has advanced steadily, through descriptive and experimental analyses, to the point where we now know more about nestling agonism than about broodmate or littermate agonism in any other class of vertebrates. There are detailed and illuminating studies of violent aggression and dominance among infant hyenas, Crocuta crocuta (e.g., Frank et al., 1991; Smale et al., 1995) and pigs, Sus scrofa (e.g., McBride, 1963; Hartsock and Graves, 1976) but, even there, experimental analysis has been minimal (an exception is Fraser and Thompson, 1991). Beyond those species, agonism among mammalian littermates has received little systematic attention. Most of the taxonomically diverse minority of avian species known to perform nestling aggression show facultative brood reduction, in which the youngest chicks in broods of two or more sometimes die, to the supposed benefit of the remaining chicks, or obligate brood reduction, in which the younger chick nearly always dies, to the supposed benefit of the single survivor (Mock, 1984a). There is also a category of (understudied) species with aggressive chicks that are precocial, nidifugous, and self-feeding (e.g., Canada goose, Branta canadensis, Radesater, ¨ 1974, 1976; Japanese quail, Coturnix coturnix, Boag and Alway, 1980), including some that are parentally 261

C 2001 by Academic Press Copyright ° All rights of reproduction in any form reserved. 0065-3454/01 $35.00

262

HUGH DRUMMOND

fed early in development (e.g., sandhill crane, Grus canadensis, Groves, 1984, and Layne, 1982). The adaptive brood reduction concept is not applied to this category because death of a chick is unlikely to free up food for its broodmates (except when it occurs during the parental feeding phase, as probably happens in sandhill cranes, Miller, 1973), but aggression of self-feeding chicks could prejudice growth, survival, or competitiveness of brood members. It should be noted that facultative brood reduction is also widespread in avian species that show no aggression (Lack, 1954; Clark and Wilson, 1985; Magrath, 1989), but occurs through begging competition and differential starvation. Aggression includes physical assaults directed at broodmates, such as pecking, biting, and pushing (but not jostling for access to food, where behavior is directed at the food or parent), and also threatening by vocal and visual displays. Here, it is considered siblicidal when it reduces the victim’s probability of surviving to independence, whether directly, through lesions, or indirectly, through exclusion from resources. Death due to nonaggressive competition for resources by begging and jostling would not be considered siblicide (cf. similar definition in Mock, 1984a). This is wider than Mock and Parker’s (1997, p. 11) definition of siblicide, which requires that aggression be “substantial” and “fatal” and has been interpreted to exclude aggressively enforced starvation when there are no visible lesions (e.g., deaths of great blue heron chicks, Ardea herodias, in Mock, 1985). I review the evidence for the different factors that influence nestling aggression and then make inferences about the two related functions that it serves: increasing personal access to parental resources and obtaining dominant status. I argue that in some species the struggle for dominance explains much of the temporal patterning and control of broodmate agonism and emphasize the different types of dominance relationship that may exist in avian species. Even though most fieldworkers have recognized that dominance exists, attempts to define or analyze it are rare. Where information is available, I also highlight the behavior of the underling, particularly submissive behavior, a neglected component of nestling agonism that often defines the relationship between two individuals (cf. Rowell, 1974). Alternative perspectives on nestling aggression are given by Mock and Forbes (1992), Forbes and Mock (1994), and Mock and Parker (1997). When mentioning two-chick broods, I refer to the first-hatched and secondhatched members as senior and junior chicks, respectively. For threechick and larger broods, seniors and juniors refer to the two eldest and two youngest, respectively. I also refer to chicks according to their order of hatching as a-chick, b-chick, and so on, or simply a, b, and so on. Reference to the dominant or the subordinate is to the chicks that habitually occupy those roles in the natal brood.

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

263

II. PROXIMATE CONTROL Numerous factors affect the occurrence and intensity of nestling agonism and they are not independent. Here, I critically examine all factors hitherto implicated as proximate influences on aggression or submission. (For further details of the critique of effects of ingestion and parental feeding method on the aggression of ardeids and other birds, see Drummond, 2001). A. FOOD AMOUNT In brood-reducing species, the resource ultimately driving nestling conflict is assumed to be food, so the amount of food supplied by parents is expected to influence their offspring’s aggression. This “food amount hypothesis” (Mock et al., 1987a) is supported by field studies of diverse facultative brood reducers, including the black-legged kittiwake, Rissa tridactyla, the osprey, Pandion haliaetus, the western grebe, Aechmophorus occidentalis, and the cattle egret, Bubulcus ibis. These studies showed more fighting in days, seasons, or localities where food availability or parental provisioning was poorer, increased attacking during meals and decreased attacking after them, as well as other associations (e.g., Poole, 1979, 1982; Braun and Hunt, 1983; Irons, 1992; review in Mock et al., 1987a). Field experiments on four taxonomically and ecologically diverse species of facultative brood reducers have demonstrated a causal relationship between food deprivation and aggression by manipulating the feeding of chicks in three distinct ways: obstructing ingestion, removing chicks from the nest, and preventing parental access. When pairs of blue-footed booby, Sula nebouxii, nestmates were prevented from ingesting parentally provided fish by taping their necks for a few days, rates of aggressive pecking and threatening by senior chicks increased several-fold (Drummond and Garc´ıa Chavelas, 1989; Nunez ˜ de la Mora et al., 1996), in contrast with the stable rates in control broods where tapes were periodically removed to allow feeding (Fig. 1). Moreover, when two unfamiliar junior chicks were briefly paired after manipulating their level of deprivation using neck tapes and supplementary feeds, the more deprived individual usually became aggressively dominant unless it was much smaller than the other (Rodr´ıguez-Girones ´ et al., 1996). In black-legged kittiwake broods temporarily prevented from ingesting parental food by wires around the bill, the longer the deprivation the more nestlings became aggressive, and after 20 h the rate of aggression was proportional to the weight deficit (Irons, 1992). When black guillemot, Cepphus grylle, parents were prevented from approaching and feeding their two-chick broods during a 6-hour period, a-chick attack rates roughly doubled (Cook et al., 2000). Furthermore, when normal feeding resumed,

264

HUGH DRUMMOND

FIG. 1. Relationship between weight loss and aggression in two-chick broods of the bluefooted booby. J = pecks by junior sibling, S = pecks by senior sibling. Tapes inhibiting food ingestion were fastened to experimentals and controls at end of day 1, but periodically removed from controls to allow feeding. Aggression by seniors increased as their weight declined. (Modified from Drummond and Garc´ıa Chavelas, 1989).

aggressive pecking subsided in all three species, demonstrating reversibility. Osprey broods removed for 3 h from the home nest showed less aggression if hand-fed before placement in an observation nest than if they were merely sham-fed, even though food was freely available in the observation nest (Machmer and Ydenberg, 1998). This food-sensitive aggressiveness in nestling birds may be an exception to the general vertebrate rule. Experiments show that food restriction per se does not provoke increased aggressiveness in birds and other vertebrates (Archer, 1976). The increase in aggression of experimentally food-restricted boobies, kittiwakes, and ospreys could therefore have been due to thwarting of feeding responses and consequent frustration, exacerbated by hunger (cf. Duncan and Wood-Gush, 1971). By contrast, the black guillemot chicks showed elevated aggression even in the total absence of food or parents, strongly implying that food deprivation per se elicited increased aggressiveness. Great egrets, Casmerodius albus, and ardeids generally, are allegedly insensitive to food ingestion: amount of food ingested by wild broods did not relate to their number of fights, food supplements did not lead to a reduction in fighting, and broods that were hand-fed different amounts in the

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

265

FIG. 2. Differences between siblicide and nonsiblicide cattle egret broods (siblicide— nonsiblicide) in percent of time performing aggression and volume of food ingested. Compared with counterparts in nonsiblicide broods, junior chicks in siblicide broods showed declining food consumption and increasing aggression as the day of death approached. (Modified from Creighton and Schnell, 1996).

laboratory did not fight at different rates (Mock et al., 1987a). However, as argued elsewhere (Drummond, 2001), these findings are inconclusive because, respectively, major variables such as age and brood size were not adequately controlled, food supplements scarcely reached the particular brood members responsible for most of the aggression (seniors), and hand-fed broods in both “low” and “high” food amount treatments were severely food deprived and exceptionally aggressive. Furthermore, some evidence in those descriptive and experimental studies (Mock et al., 1987a) and in the descriptive studies of Fujioka (1985a) and Creighton and Schnell (1996) reveals associations between food deprivation and increased aggression in great egrets and cattle egrets (Fig. 2). There has been no attempt to reduce the food consumption of ardeid chicks in the field experimentally. Food deprivation may lead to increased aggressiveness in facultatively siblicidal birds generally. There are no experimental data showing that it does so in obligately siblicidal species, and observations of persistent aggression in the face of apparently adequate food provision or ingestion (e.g., Meyburg, 1977; Gerhardt et al., 1997) suggest that obligately siblicidal species

266

HUGH DRUMMOND

are insensitive to it. No observations relating food amount to aggression have been reported for self-feeding chicks although Kruijt (1964) suggested that in broods of Burmese red junglefowl, Gallus g. spacideceus, hunger may make chicks peck more frequently during fights. B. PARENTAL FEEDING METHOD According to the proximate level “prey-size hypothesis” (Mock, 1984b, 1985), reframed as the “feeding-method hypothesis” (Drummond, 2001), nestlings are more likely to fight when they receive monopolizably small food parcels directly (mouth to mouth) from parents than when large parcels are dumped on the substrate for indirect consumption. Large prey are frequently converted by some species into small parcels that can be delivered directly and monopolized; for example, when boobies and pelicans predigest fish into mush for hatchlings and raptors tear prey into fragments. This attractive hypothesis potentially accounts for variation in fighting among populations of the same species and during ontogeny, but its validity is in doubt because supporting studies failed to control adequately for food deprivation and the hypothesis is contradicted by a descriptive field study and developmental data for two species (Drummond, 2001), as described below. When great blue heron broods were fostered into great egret nests, and consequently fed smaller prey than in the home nest, chicks ate more directly and fought more frequently than in normal nests of the same heron population (Mock, 1984b). However, assuming heron aggression is food sensitive, the observed increase in fighting could have been due to underfeeding by foster parents belonging to a much smaller species. No measures of food consumption or nestling growth were taken, but increased mortality in the fostered heron broods is evidence of underfeeding. Furthermore, the counterpart fostering experiment plainly contradicted the hypothesis: when great egret broods were raised by great blue herons, despite feeding mostly indirectly on the large prey provided, they fought intensely, at the normal egret level. A comparison of two great blue heron populations, each observed in a single season, showed that nestlings in Texas consumed mostly large fish indirectly and seldom fought, while nestlings in Quebec consumed mostly small fish directly and fought at considerably higher rates (Mock et al., 1987b). But greater food deprivation in Quebec cannot be ruled out because there was no comparison of food amount consumed, growth, or mortality of chicks in the two populations. Furthermore, in the only other population of this species where behavior has been quantified, the main prediction of the feeding method hypothesis was contradicted. Ontario heron parents usually regurgitated a single fish of the (large) size range regurgitated in Texas, and

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

267

FIG. 3. Transition from indirect to direct feeding methods in broods of great egrets (open circles) but not in broods of great blue herons (solid circles). No developmental association between feeding method and aggression has yet been shown. (Modified from Mock, 1985).

chicks were generally unable to monopolize it (David and Berrill, 1987). Even so, chicks frequently seized the fish from the parental bill before disputing it among themselves, and violent aggression (noted in 65% of broods) led to wounding, exclusion from feedings, or expulsion from the nest in at least 39% of broods. Rates of pecking were low, as in Texas herons, but aggressive pecking frequently served to exclude rivals completely from unmonopolizably large prey. Finally, the only study that has attempted to document an increase in aggression during the natural (in many species) developmental switch from indirect to direct feeding (Fig. 3), failed to find any evidence for an increase in aggression in brown pelicans, Pelecanus occidentalis (Pinson and Drummond, 1993), although the sample of broods was small. Moreover, the reports of Mock et al. (1987b) and Mock (1984b) actually include observations contradicting the developmental prediction of the feeding-method hypothesis: heron chicks appear to have been no less aggressive during early indirect feeding, whether in natural conditions (in Quebec) or with egret foster parents (in Texas), than they were when they switched later on to direct feeding. We need tests of the main prediction of the feeding-method hypothesis that control for effects of food amount, and tests of its main assumption that aggression is more effective for monopolizing direct feeds than indirect feeds. Drummond (2001) suggests an alternative hypothesis to explain variation among and within species: aggression is more likely whenever food

268

HUGH DRUMMOND

is concentrated in a restricted area (as often occurs when parents provide it), so that approaching it brings broodmates close enough to attack each other. If attacking is feasible and can improve the attacker’s feeding priority, then attack is expected, whether feeding is direct or indirect. Notice that the hypothesis predicts aggression between self-feeding chicks at a spatially concentrated food source, at least when food is scarce elsewhere. The efficacy of aggression for improving feeding priority probably varies with, for example, beak morphology and the number and mobility of competitors, factors whose effects are largely unknown. C. RELATIVE SIZE AND AGE OF CHICKS Large inequalities in age or size are often associated with increased or earlier mortality of the smaller broodmate, in both facultatively and obligately siblicidal species. Size differences are initially due mostly to staggered hatching and subsequently become modified by differential growth of seniors and juniors or of males and females (Fig. 4). Doubling the hatch intervals of cattle egrets, Bubulcus ibis, increased the probability that last-hatched chicks

FIG. 4. Development of sibling weight difference (senior—junior) according to sexual composition of 27 2-chick blue-footed booby broods, with an average 4-day hatching interval. F = female, M = male, and order of these letters specifies hatching order. Females with younger brothers enjoyed a persistent size advantage that declined over the first 40 days. Males with younger sisters were outgrown by them. (Modified from Drummond et al., 1991).

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

269

would die (Mock and Ploger, 1987). Naturally long hatch intervals in twochick broods seemed to promote the death of South Polar skua, Catharacta maccormicki, juniors (Young, 1963; Spellerberg, 1971a), but had no detectable effect on probability or age of death, or even on the growth of blue-footed booby junior chicks (Drummond et al., 1991). In the masked booby (Sula dactylatra), the longer the natural hatch interval the sooner the b-chick died, and experimental shortening of the interval prolonged the b-chick’s brief life (Anderson, 1989). Junior raptor chicks seem to survive longer if the hatch interval is short, provided parental brooding quells aggression (Meyburg, 1974, 1977; Newton, 1977). However, large hatch intervals could lead to increased mortality simply because the aggressor’s superior size and maturity make it a more formidable opponent. The questions are Does the direction of size difference determine which agonistic role a chick adopts? and, Does the magnitude of size disparity affect the strength of the agonistic tendencies of dominant and subordinate chicks? As a rule, within-brood dominance hierarchies follow the hatching order of chicks, even when hatching intervals are just several hours (e.g., Safriel, 1981). Relative size and maturity undoubtedly affect relative fighting ability and the outcome of fights, but chicks also seem able to detect and respond to relative size without need for a battle. When blue-footed booby singletons (chicks that grew up without a nestmate) were experimentally paired in artificial nests, in 12 of 13 pairs the larger chick performed more aggressive acts and fewer submissive acts than the smaller chick. Each pair of chicks slipped into this dominance relationship in a matter of minutes, requiring little behavioral interaction. Size differences were of a similar magnitude to natural differences between broodmates, and the larger singleton dominated even in the three most evenly matched pairs, where chicks differed in linear size and weight by just a few percent (Drummond and Osorno, 1992). Hence, the decision to attack or submit probably depends partly on assessment of relative size. This was illustrated by a field experiment (Drummond and Osorno, 1992) in which most junior (hitherto subordinate) chicks suddenly became aggressive when confronted with new (hitherto dominant) nestmates smaller than themselves, but none did so when the newcomers were larger (see Section II.F). That is, even habitually submissive chicks could be induced to attack by a newly perceived size advantage. The experimental evidence shows that aggression in facultatively siblicidal species tends to intensify as broodmates become more equally matched. Two experiments on food-sensitive aggression in natural osprey and black guillemot broods revealed incidentally that small size differences were associated with more fighting (Machmer and Ydenberg, 1998; Cook et al., 2000). The most informative studies have been those that minimized hatch intervals by swapping hatchlings between nests. In even-aged broods of cattle egrets,

270

HUGH DRUMMOND

FIG. 5. Aggression in even-aged and control 2-chick broods of the blue-footed booby, by dominants (solid circles) and subordinates (open circles). Based on each chick’s mean daily score over its own 11–20 day age interval. Dominants paired with similar-sized nestmates nearly tripled their attacking. (Modified from Osorno and Drummond, 1995).

dominance relationships always emerged, but in both reported experiments dominance was less clear-cut than in control broods, and serious fighting was more frequent (Fujioka, 1985b; Mock and Ploger, 1987). Likewise, between ages 11 days and 20 days, dominant chicks in experimentally evenaged broods of the blue-footed booby pecked their nestmates roughly three times as frequently as dominant chicks of the same ages in control broods (Fig. 5). Intriguingly, over the same age range (other) dominant booby chicks also intensified their pecking against much younger nestmates, when the experimental hatch interval was doubled. Aggression in black kite (Milvus migrans) chicks also seems to intensify with both greater synchrony and exaggerated asynchrony (Vinuela, ˜ 1999), but no increase with exaggerated asynchrony was detected in cattle egrets (Mock and Ploger, 1987) or ospreys (Forbes, 1991). Precisely which stimuli elicit increased aggression in even-aged broods? We can discount the possibility that elevated aggression in even-aged broods of cattle egrets and blue-footed boobies was due to reduced feeding because those broods actually consumed more food than control broods (Fujioka, 1985b; Mock and Ploger, 1987; Osorno and Drummond, 1995). In the evenaged boobies it was clearly dominant chicks that made the adjustment in fighting rate, whereas their subordinate nestmates showed the same minimal aggression that characterized control subordinates (Fig. 5). Hence, (1) subordinate chicks did not noticeably modify their agonistic behavior

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

271

in response to size similarity, and (2) dominants probably intensified their attacking in response to size similarity itself, rather than to meet unusual aggressive challenges. In contrast, elevated aggression in even-aged cattle egret broods may be partly a response to greater challenges, because dominance was unstable (Fujioka, 1985b; Mock and Ploger, 1987). To my knowledge, no effect of size difference on direction or intensity of aggressiveness has been shown for any obligately brood-reducing species. Observations of self-feeding chicks suggest that direction of aggressiveness is not related to relative weight in canvasback, Aythya valisineria, or redhead ducklings, A. americana (Collias and Collias, 1956), or in greylag goslings, Anser anser (Kalas, 1977; Lorenz, 1991). D. BROOD SIZE In facultatively siblicidal species, larger broods seem to exhibit more intense aggression (e.g., Jamieson et al., 1983). Great egret broods of three or four showed a higher rate of fights per dyad than broods of two (Mock and Parker, 1986) and cattle egret broods appeared to fight more before natural brood reduction (Fujioka, 1985a). Fighting rates of great blue herons increased with brood size in one population (David and Berrill, 1987) but not in another (Mock and Parker, 1986). If we can show that fighting increases with brood size, the critical question is whether brood size per se affects aggression or if the effect is mediated by diminished food ingestion. Thus far, there is inconclusive evidence that brood size affects aggression in cattle egrets and more convincing evidence that it has little or no effect on aggression in brown pelicans, two species that could well differ in this respect. The study of cattle egrets by Mock and Lamey (1991) is difficult to interpret because analyses emphasized within-brood mean scores rather than individual scores and control samples exhibited substantial unexplained variation. My interpretation of the analyses presented in the original article differs from that of the authors. Food consumption and aggression were recorded in three-chick broods before and after removing either c-chicks (Experiment 1) or a-chicks (Experiment 2) for 3 days. In control broods, chicks were briefly handled then returned to the nest. Aggression was analyzed mainly by computing the mean score of all dyads within each brood (three dyads before removal, one dyad after removal), a dubious procedure because distribution of aggression in three-chick broods is systematically uneven: roughly 90% of aggression in control broods was in the b–c dyad, and in this dyad the initiator is invariably the b-chick (Fujioka, 1985a). After removal of c-chicks, average aggression per dyad declined precipitously in experimental broods but not in control broods. However, for some reason the aggression between the two

272

HUGH DRUMMOND

seniors declined similarly in experimental and control broods. Thus, Experiment 1 shows only that when the most junior chick is removed the other chicks (especially the b-chick) stop attacking it intensely; there was no sound evidence that aggression between the remaining two seniors was affected. In Experiment 2, in addition to a reduction in mean aggression per dyad, an effect on the aggression of the two remaining chicks was clearly shown. After removal of a-chicks, aggression between the two junior chicks declined steeply, whereas the aggression of the two control juniors remained stable throughout, showing that disappearance of the most senior chick resulted in diminished aggression by the remaining pair. (The design also included replacement of removed chicks after 3 days absence, and replaced a-chicks greatly increased their aggression to both sibs, which was properly interpreted as reflecting reassertion of dominance rather than a simple response to brood size.). But could food amount have been responsible? The authors concluded that decline in aggression in Experiment 2 was not caused by the 24% increase in per capita food consumption (Xb,c after − Xa,b,c before ; p< 0.06) because control broods showed a similar (unexplained) increase in per capita consumption (27%, Xa,b,c after − Xa,b,c before ) and stable aggression. However, the calculated food increase in experimental broods was certainly underestimated by including a-chicks (normally the greatest food consumers) in the baseline figure. Food presumably affects the aggressiveness of the individual that consumes it, so to reject the possibility that experimental b-chicks were pacified by extra food, we would need to know that they themselves did not enjoy a greater feeding increase than control b-chicks. That was not shown, so the positive effect of brood size on aggression in cattle egrets cannot be confidently attributed to brood size per se rather than to food amount. There was no effect of brood size in the brown pelican. Three-chick broods of this species show different dynamics from equivalent great egret broods in that the two seniors fight as frequently as any other dyad, (Ploger, 1997). After removal of the c-chick, the rate of fighting per dyad, and also apparently within the a–b dyad, remained unchanged over the succeeding block of 6–9 days (Ploger, 1997). (The a-chick’s rate of ingestion was unaffected by removal of c and the b-chick’s rate declined, because parents reduced food provision to the brood. So it is conceivable that reduction of brood size elicited a tendency toward decreased fighting that was masked by increased fighting by the b-chick in response to reduced food intake, but this is highly speculative.) The contrast between the results for cattle egrets and brown pelicans points to the need, when dealing with multiple-chick hierarchies, to analyze particular behavioral roles. The literature also shows a contradictory trend for aggression to be weaker or less likely in species with parental feeding and larger broods. Among

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

273

sulids, intense aggression almost invariably leading to prompt death occurs in the two-chick broods of masked and brown boobies, whereas aggression in the two- and three-chick broods of the blue-footed booby is fooddependent, usually more moderate and frequently not fatal. Chicks in the characteristically larger broods of the Peruvian booby (Sula variegata) are reputedly nonviolent (Drummond, 1987, 1989). The Accipitridae, to which most raptors belong, is characterized by (1) large-bodied species with broods of two and obligate aggression, (2) medium-sized species with larger broods and variable, apparently food-dependent aggression, and (3) small species with even larger broods and no agonism at all (Newton, 1979). In the behaviorally best known ardeids, fighting appears greater in the cattle egret and the great egret, with typical clutches of 2–5 and 3–5, respectively, than in the little egret (Egretta garzetta, Inoue, 1985), the little blue heron (E. caerulea, Werschkul, 1979) and the great blue heron (Mock, 1985), with clutches of 2–6, 3–6 and 2–7, respectively (clutch sizes from Mart´ınez-Vilalta and Motis, 1992). Most avian species with precocial, parentally fed young known to be aggressive have small broods; the oystercatcher, Haematopus ostralegus, the American black oystercatcher, Haematopus bachmani, the sandhill crane, the western grebe, and the South Polar skua typically have broods of two chicks (Young, 1963; Nuechterlein, 1981; Safriel, 1981; Layne, 1982; Groves, 1984). The apparent trend for diminished aggression in species with parental feeding and larger broods requires comparative study. It is possible to reconcile the contradiction between these trends among species and within species. First, I propose that aggression occurs only in those brood-reducing species where aggression is generally effective, and that effectiveness of any individual’s aggression declines with increasing brood size because (1) doses of punishment received by individuals are diluted, and (2) the aggressor’s capacity to concentrate punishment on particular individuals is diminished, thereby limiting the scope for excluding broodmates from feeding. Targeting particular individuals should be increasingly hard the more broodmates there are to obstruct access and distract, and also if the ability to recognize individuals is numerically limited. In species with precocial young, the herring gull (Larus argentatus), the ring-billed gull (Larus delawarensis), the common tern (Sterna hirundo), the Canada goose, and the domestic chicken, discrimination of broodmates versus nonbroodmates emerges as early as age 2 to 4 days (Evans, 1970; Zajonc et al., 1975; Noseworthy and Lien, 1976; Radesater, ¨ 1976; Burger et al., 1988; Palestis and Burger, 1999). Furthermore, newly fledged budgerigars (Melopsitticus undulatus) are able to distinguish individuals among their erstwhile nestmates (Stamps et al., 1990). However, we do not know how many broodmates can be individually discriminated by size or other features, by either precocial or altricial nestlings. Second, I suggest that within

274

HUGH DRUMMOND

brood-reducing species that show nestling aggression, intensity could initially increase with brood size (as an adaptive response to food shortage or brood size itself) then decline with additional increases in brood size as larger numbers make aggression less effective. In species where some broods are large (say > 4 chicks), the combination of these two effects may yield a relationship between brood size and aggressiveness shaped like an inverted U. E. MATURATION/AGE Pioneers described the emergence (e.g., Nice, 1962) and ontogeny (e.g., Kruijt, 1964) of broodmate agonism in captivity, and more recently fieldworkers have provided piecemeal information on how the frequency of broodmate aggression varies with age. For every species there may be a typical curve of aggression against time for chicks of each dominance rank in each brood size, with variation around that curve due to differences among broods in such factors as food amount and relative size of broodmates. Patterns of age-related change can potentially reveal the influence of intrinsic, maturational control of aggressiveness as well as effects of environmental variables such as food amount, food-parcel size, relative size of broodmates, and broodmate behavior. Maturational change in aggressiveness could result from neuromuscular development or be due to changing patterns of sleeping or responsiveness to stimuli. I will consider when aggression begins, when it ends, and how it varies over time. Aggression starts shortly after hatching in obligate and facultative brood reducing species, and in self-feeding chicks. Onset within hours or days of the victim’s hatching has been reported for the masked booby (Anderson, 1989), the white pelican, P. erythrorhynchos (Cash and Evans, 1986), the lesser spotted eagle (A. pomarina, Meyburg, 1974), the black eagle (Gargett, 1978), egrets (Mock and Parker, 1997), the great blue heron (David and Berrill, 1987), the European crane, Grus grus (Heinroth and Heinroth, 1924–33, in Nice, 1962), the sandhill crane (Miller, 1973), the whooping crane, Grus americana (Wellington et al., 1996), the western grebe (Nuechterlein, 1981), the black-legged kittiwake (Cullen, 1957; Braun and Hunt, 1983), the great northern diver, Gavia immer (Beebe, 1907), the blue-footed booby (Osorno and Drummond, 1995), the South Polar skua (Young, 1963; Spellerberg, 1971b), the black kite (Vinuela, ˜ 1999), the domestic chicken (Nice, 1962; Zajonc et al., 1975), the western capercaillie, Tetrao urogallus, the black grouse, T. tetrix (Rajala, 1962), the Burmese red junglefowl (Kruijt, 1964), the Canada goose (Radesater, ¨ 1974, 1976), and the greylag goose (Kalas, 1977). For example, a-chicks of the great white pelican (P. onocrotalus) attacked when they had scarcely developed the motor coordination to lift their own heads, and even when the b-chick was still in its eggshell (Cooper,

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

275

1980). Late onset of aggression has never been documented. Although it seems that in obligate brood reducers early aggression by senior chicks is invariably present, implying that its occurrence is obligate (independent of environmental variables such as food amount or brood size), in some species of facultative brood reducers its occurrence could vary among broods, implying conditionality. The exhaustive longitudinal observations required to show total absence of aggression in particular broods have not been made, but there are indications of this in great blue herons (Mock, 1984b; David and Berrill, 1987) and ospreys (review in Poole, 1979). In contrast, occurrence of aggression in blue-footed booby broods seems to be universal (hence, probably obligate), although its intensity is clearly calibrated by food amount and other factors. Aggressiveness appears generally to be sustained over most of the nestling period or until the victim dies. In obligate brood reducers, intense aggression persists until the junior chick is dead, which usually takes just a few days (e.g., Gargett, 1978; Anderson, 1989). But would aggression occur at greater ages if a potential victim were available? When Anderson (1990) prevented the obligate siblicide of masked booby chicks by creating even-aged broods and removing the larger nestmate during feeding sessions, aggression apparently continued well after age 10 days. In brown booby broods, where siblicide was frustrated by similar methods or by tethering pairs of nestmates, fierce aggression persisted until at least age 23 days (Drummond, unpubl. data). An 18-day-old tawny eagle a-chick, suddenly confronted by the sibling that had been taken into protective custody shortly after hatching, pecked it almost to death in less than 2 hours (Steyn, 1973). On the other hand, junior hatchlings of obligately siblicidal raptors that are removed into foster nests can cohabit with the a-chick in the natal nest by the time they are both feathered (Meyburg, 1977; and see Ingram, 1959, and Newton, 1979). Hence the aggressiveness of these obligate brood reducers can last well beyond the period when siblicide is normally completed, but may expire before independence. Among self-feeding chicks and in facultatively siblicidal species, the rather sparse data show that aggression persists until broodmates die or fledge, for example in the sandhill crane (Miller, 1973), the American black oystercatcher (Groves, 1984), the black-legged kittiwake (Braun and Hunt, 1983), the great blue heron (David and Berrill, 1987) and the blue-footed booby (Drummond et al., 1986). In the booby it persisted until at least age 118 days, by which time fledglings were partially hunting for themselves (Drummond et al., 1991). Intensity of aggression varies with age in the blue-footed booby. Representative curves show that pecking by senior chicks was modest at age 5–10 days, quadrupled at ages 10–20 days, then declined more or less progressively over the next 40 days (Drummond et al., 1986). Pecking by juniors

276

HUGH DRUMMOND

peaked modestly when it was first registered, at age 10–15 days, and stayed close to zero thereafter. The pattern of the senior chick is consistent with an initial maturational increase associated with its developing motor skills and stamina (and increasing exposure of the nestmate as it emerges from brooding), and eventual decline in response to the nestmate’s acquired submissiveness (and decreasing exposure as its locomotor ability develops). The pattern of the junior chick is consistent with initial maturational increase followed by abeyance as it learns to adopt a subordinate role. There is also evidence that in this species responsiveness to stimuli varies with age: the positive effect of artificial food deprivation on frequency of aggressive pecking by senior chicks increased with age up to 6 weeks and declined afterward (Drummond and Garc´ıa Chavelas, 1989). Alternatively, the first effect could be due not to changing responsiveness but to improvement in personal motor skills and stamina, and the second to improvement in the broodmate’s capacity to evade. For most other species, the pattern of variation in intensity with age is confusing and permits no inferences about maturation. In black guillemots, which fledge at age 30–40 days, aggression appears to peak at age 4–12 days and then declines with age (Cook et al., 2000), similar to the pattern in blue-footed boobies. Kittiwake nestmates seemed to fight most fiercely in the first days after hatching (Cullen, 1957). In cattle egrets observed up to brood age 25 days (one week short of fledging), aggression associated with feeding bouts apparently became more prolonged as chicks got older, whether siblicide occurred or not, with a peak in nonsiblicide broods at age 21 days (Creighton and Schnell, 1996), and in a Japanese population, serious fighting did not start until age 19 days (Fujioka, 1985a). However, elsewhere fighting by cattle egrets appeared not to increase with age (Ploger and Mock, 1986), and fighting by great egrets appeared to decrease during the first month (Mock and Parker, 1986). Mock and Lamey (1991) claimed that ardeid broods generally diminish their fighting during the first 3 weeks, as dominance relationships become established. F. AGONISTIC EXPERIENCE Agonistic interactions with broodmates can affect the ontogeny of aggressive and submissive behavior. Drummond and Osorno (1992) tested this idea on 12- to 55-day old blue-footed boobies by pairing unfamiliar chicks that had grown up in their natal two-chick brood. They had thus experienced the natural schedule of habitually attacking and threatening their nestmate (dominants) or habitually absorbing pecks and emitting submissive displays (subordinates). As predicted, in these trials each category tended to assume the agonistic role it had customarily adopted. When dominants were paired

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

277

for 4 h with similar-sized subordinates in artificial nests, the former behaved aggressively and nonsubmissively and the latter behaved submissively and nonaggressively. When two subordinates were paired in the nest of one of them for 30 min, both stood about passively, and when two dominants were similarly paired, there were escalated fights. The robustness of this effect was put to the test by fostering among a fresh set of 15–31 day old dominants and subordinates. This pitted dominants against subordinates that were on average 32% heavier and 3.9 days older, effectively inverting the natural age/size difference between nestmates (Drummond and Osorno, 1992). Controls were created by similar fostering, but consisted of a dominant with a smaller subordinate. So strong was the effect of prior experience, that 11 of 12 experimental dominants dominated their imposing new nestmates during the 3 days after pairing, performing an average 5.6 aggressive acts per hour compared to 1.8 acts by subordinates. Most interestingly, experimental subordinates were more aggressive than control subordinates, evidently detecting their newfound size advantage. Seven of them showed elevated aggression, as if attempting to invert the dominance relationship, but only one of them succeeded. As in natural dominance inversions (e.g., Drummond et al., 1991), these rebellions were hard-fought and sustained over many days. The issue of which category of chick changes as a result of prior experience was resolved by pairing a further sample of agonistically experienced chicks against singletons (Drummond and Canales, 1998). Subjects were 2–3 weeks old and pairs were created by fostering one chick into the nest of the other (controlling, in this and all other experiments, for residence effects). During the 4 hours after pairing, dominants dominated singletons, whereas subordinates were submissive to singletons, showing that early experience trains the senior chick in each brood as a winner and the junior chick as a loser. Moreover, the effects of training were sufficient to overwhelm the effects of size difference, because in these pairings dominants were slightly smaller than singletons and subordinates were slightly larger than singletons. However, training as a winner had less persistent effects than training as a loser, so that by the sixth day after pairing, whereas most subordinates remained submissive only about half of the dominants were winning (Fig. 6). Importantly, the dominance relationship in dominant-singleton pairs remained unstable for weeks after chicks were paired, as if each chick’s lack of consistent early training in subordination made it difficult for the opponent to impose that role on it. Trained winning and losing were postulated by comparative psychologists as general learning phenomena on the basis of laboratory experiments (e.g., Scott and Marston, 1953), but these processes have seldom been confirmed in natural contexts. No other experimental analyses of avian nestlings

278

HUGH DRUMMOND

FIG. 6. Persistence of agonistic roles of blue-footed boobies after pairing with an unfamiliar singleton. Proportions of subordinates that lost against singletons and dominants that won against singletons, on the 10 days after pairing. Prior training as a loser had more persistent effects than prior training as a winner. (Modified from Drummond and Canales, 1998).

have been made, but intimidation, hesitation and deference of junior chicks have been mentioned in reports on other facultative brood reducers (e.g., the great egret, Mock, 1985), obligate brood reducers (e.g., the lesser spotted eagle, Aquila pomarina, Meyburg, 1974), and self-feeding chicks (e.g., Canada goslings, Radesater, ¨ 1974, 1976). G. BEHAVIOR OF THE BROODMATE Additional comparisons of behavior in the above pairings of unfamiliar chicks suggested that current behavior of the broodmate is an important proximate influence on the agonism of blue-footed boobies. First, dominant chicks escalated their aggression when challenged. Dominants performed roughly 2 aggressive acts per hour to similar-sized chicks that were submissive (subordinates), compared with roughly 20–60 acts to similar-sized chicks that were highly aggressive (dominants); and dominants performed twice as many aggressive acts to subordinates that were occasionally aggressive (being the larger pairmate) as to subordinates that remained submissive (being the smaller pairmate) (Drummond and Osorno, 1992). Second, chicks that would otherwise be submissive adopted an aggressive role when they were confronted by a habitually submissive pairmate: although blue-footed booby singletons spontaneously submit to larger singletons (Drummond and Osorno, 1992), when confronted by larger subordinates they became aggressive and managed to dominate, apparently responding to the larger

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

279

chick’s submissive demeanor (Drummond and Canales, 1998). However, the first and last of these three results require confirmation since in each comparison the data are from two experiments with different observation protocols. H. SEX No study of a brood-reducing species documents intrinsic differences in agonistic tendencies of male versus female chicks, or even reports any data that could have revealed such differences, so this is an open question. However, sex composition of a brood seems likely to influence agonism in species where males and females differ in size. Some authors have speculated that enhanced or prolonged size differential in raptor chicks due to sexual dimorphism could affect agonism or its outcome, leading to sex-biased mortality of nestlings (Craighead and Craighead, 1956; Edwards and Collopy, 1983; Bortolotti, 1986; Edwards et al., 1988), but firm evidence is wanting (critique in Drummond et al., 1991). In captive American kestrels, Falco sparverius, the larger body size of female chicks enabled them to outscramble their brothers for some prey (Anderson et al., 1993), but no aggression was involved. In 54 two-chick broods of blue-footed boobies, followed from shortly after hatching until age 100 days, age and sexual composition interacted to generate substantial variation in size difference between nestmates, including systematic size inversion (Fig. 4). Dominance relationships were generally stable, whatever the sexual composition and size difference, but because intensity of aggression was not measured we do not know whether, for instance, pecking intensified when females outgrew their elder brothers. In (self-feeding) chicks, males fought more frequently and vigorously than females, although this difference did not emerge until the third week of life in junglefowl (Kruijt, 1964) and the fourth week in grouse (Rajala, 1962). I. HORMONES It has long been known that sex hormones can affect aggressiveness and status in hierarchies of avian adults (Allee et al., 1939) and that androgen injections can precipitate the onset of fighting in precocial chicks (Kruijt, 1964). Two routes have now been identified by which natural hormones potentially influence nestling agonism in facultative brood reducers, but tests have yet to be made. First, hormones can be supplied to embryos in different amounts by the mother. Testosterone incorporated in canary (Serinus canaria) eggs in amounts that increased with laying order resulted in superior postnatal

280

HUGH DRUMMOND

FIG. 7. Yolk contents (mean, 95% confidence limits) of first-, second- , and third-laid eggs in cattle egret clutches. Effects of these differences on agonistic behavior have not been tested. (Modified from Schwabl et al., 1997).

growth and begging of later-laid offspring, and even differentially boosted social rank in juveniles (at 7–19 weeks) as determined by supplanting at a food dish (Schwabl, 1993, 1996). It is therefore quite possible that mothers of some species modulate their nestlings’ aggressiveness by lacing eggs with hormones. Schwabl et al. (1997) have indeed found more androgens in the first two eggs of three-egg clutches of cattle egrets (Fig. 7), but the consequences for broodmate agonism are as yet unknown. Second, hormones made by the chicks themselves, in response to agonistic or other experience, could be important. At age 15–20 days, junior chicks of the blue-footed booby show double the level of circulating corticosterone shown by senior chicks or singletons (Fig. 8; see also Ramos-Fernandez et al., 2000). This is a hormone that can facilitate submissiveness in adult vertebrates (Leshner, 1981, 1983), but whose effects on the behavior of nestlings have not been examined. Secretion of extra corticosterone could be a response to the subordination or other social experience of juniors, or to their restricted food intake.

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

281

FIG. 8. Corticosterone levels in blood samples of 15- to 20-day- old blue-footed booby chicks. After two weeks in a subordinate role, subordinates show elevated corticosterone. Effects of this hormone on agonistic behavior of nestlings have not been tested. (Modified from Nunez ˜ de la Mora et al., 1996).

J. PARENTAL BEHAVIOR Parents clearly create the initial competitive asymmetries between broodmates, and may even modify the outcome of their aggression by controlling nest location and architecture (Anderson, 1995), but do they exert any ongoing influence over agonism? The theory of parent–offspring conflict predicts parental suppression of hostilities in some circumstances (Trivers, 1974; O’Connor, 1978; Godfray and Harper, 1990), but overt behavioral conflict is likely to disappear as fundamental genetical conflict is resolved over evolutionary time (e.g., Parker, 1985). Parents generally give every appearance of being indifferent to even conspicuous violence among their nestlings (reviews in Mock and Forbes, 1992, and Drummond, 1993), but in many cases they are probably influencing it. For example, if food amount affects aggressiveness, then parental food allocations necessarily influence nestling agonism. Moreover, sometimes parents are expected to modify total food provision to the brood to head off aggressive brood reduction (Forbes, 1993; Rodr´ıguez-Girones ´ et al., 1996). Blue-footed boobies and black guillemots increased the differential feeding of dominant chicks when artificial food shortage provoked severe agonism (Drummond and Garc´ıa Chavelas, 1989; Cooke et al., 2000), and because their aggression is food sensitive, parents may have fed dominants extra to appease them. However, we are often unsure to what extent food allocation is controlled by parents rather than chicks.

282

HUGH DRUMMOND

One experiment showed higher mortality for blue-footed booby chicks when fostered into nests of the obligately siblicidal masked booby. This result was attributed to experimental suspension of normal parental control of nestling aggression (Lougheed and Anderson, 1999), but behavior of adults was not observed and the hypothetical means of control was not indentified. There are also reports of parents suppressing aggression by brooding chicks (e.g., Meyburg, 1974; Cash and Evans, 1986), giving (false) alarm calls (Young, 1963), or attending two mobile broodmates separately (Procter, 1975; Layne, 1982); and an unspecified aspect of parental presence somehow depressed the incidence of severe aggression in broods of great blue herons (Mock, 1987). When Cohen Fernandez ´ (1988) and her team watched 13 two-chick brown booby broods day and night during a total of 452 brood/hours, parents at six nests were seen to accommodate nest material around recently expelled b-chicks with their bills as if to protect them from siblicide, and parents at all 13 nests on at least one occasion interrupted violent interchanges by bill-nudging the chicks themselves. Drummond (1993) and Rodr´ıguez-Girones ´ et al. (1996) both reported rare observations of blue-footed booby adults pecking highly aggressive fostered chicks on the cranium, which briefly dampened their attacks. Certainly, parents of most species sometimes influence agonism by feeding and brooding chicks, but there is little indication that such effects are adaptive functions. We need to keep an open mind and avoid overinterpreting adult behavior that may be vestigial, that may produce incidental results, or that may be selected for its effects in other contexts, such as dealing with parasitic adoptees.

III. FUNCTIONS The functions of nestling agonism can only be inferred from the contexts in which it occurs, its proximate control, and its short term consequences, because there are virtually no unambiguous data showing how nestling agonism affects the inclusive fitness of the actors. Experimental studies are few, and in descriptive studies, agonism and seniority are usually confounded; we usually cannot be sure whether an elder broodmate’s feeding priority is due to its violence or merely to its superior size and begging. There is a virtually unbridged gap between our empirical knowledge and the sophisticated models that apply brood reduction theory and parent– offspring conflict theory to nestling agonism (review in Mock and Parker, 1997) and which tend to govern our thinking. We need to understand the basics.

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

283

Nestling aggression most obviously serves to secure immediately a greater share of parental investment by outcompeting or eliminating competitors, but a second function is often critical: obtaining or maintaining dominant status. The second function is largely nested in the first one, but this is how we habitually conceive of functions. For example, a courtship display can function to (1) obtain a better mate, (2) thereby produce better quality infant offspring, and (3) thereby increase lifetime reproductive success by generating more recruits. Recognizing status in the brood dominance hierarchy as a function enables us to explain many aspects of the occurrence, timing, and intensity of agonism that would otherwise be inexplicable. It also prepares us to look for additional consequences of dominance status beyond food consumption and the nestling period, particularly in self-feeding chicks, which may seldom obtain a feeding advantage through dominance. It needs to be mentioned that in 16 species, mostly raptors, nestling aggression can lead to death followed by broodmate cannibalism (review in Stanback and Koenig, 1992), although the proportion of corpses ingested is unknown for any species. Most authors have inferred that the function of such siblicide is competition among broodmates for parentally provided food, and that ingestion of the victim is opportunistic scavenging (e.g., Mock and Parker, 1986; Stanback and Koenig, 1992). A. BIASING PARENTAL INVESTMENT With varying degrees of rigor, aggressively dominant broodmates have been shown to enjoy feeding advantages in obligate and facultative brood reducers, including the white pelican (Cash and Evans, 1986; Evans and McMahon, 1987), the blue-footed booby (Drummond et al., 1986; Anderson and Ricklefs, 1995), the double-crested cormorant (Phalacrocorax auritus, Lewis, 1929; Mendall, 1936), the osprey (Poole, 1982; Jamieson et al., 1983), the cattle egret (Fujioka, 1985a; Ploger and Mock, 1986), the little blue heron (Florida caerulea, Werschkul, 1979), the South Polar skua (Procter, 1975), the black-legged kittiwake (Braun and Hunt, 1983), the western grebe (Nuechterlein, 1981), and the oystercatcher (Safriel, 1981). As far as I know, there are no comparable observations of self-feeding chicks of any species. Of course, even if parents schedule food deliveries such that each brood member gets the same amount at any particular age, dominant chicks will appear to eat more food than subordinates when we observe them in the early nestling period because they are generally older than their broodmates and food consumption increases with age at this time. This confounds the results of most studies, but when age of blue-footed booby chicks was analytically controlled, dominants still outperformed subordinates during the early part of the nestling period, both in frequency of feeds (Drummond et al., 1986) and

284

HUGH DRUMMOND

FIG. 9. Feeding frequency and food mass consumed by dominants (solid bars) and subordinates (hollow bars) in two-chick broods of the blue-footed booby. Nestmates are compared at the same age, although dominants were on average four days older than subordinate broodmates. (Modified from Guerra and Drummond, 1995).

grams of food ingested (Fig. 9). Of course, this does not entitle us to dismiss the possibility that superior size and motor coordination could simply lead to a similar feeding advantage in the absence of aggression. Why does the elder/larger/maturer/aggressive chick get more food? Field observers have inferred that subordinate chicks obtain less food not only because inferior age/size allows them to be nudged aside or outreached, but also because of their very subordination: they retreat or are driven from favorable feeding locations, they are reluctant to stand, approach, beg, or reach for food (e.g., Nuechterlein, 1981), or they are prevented from begging or reaching when food transfer is imminent. A causal link between aggression and feeding priority was indicated when pairs of blue-footed booby broodmates were experimentally prevented from swallowing parentally provided food: as the pecking of senior chicks increased by several hundred percent, their feeding advantage increased from 37 to 57% more attempted parental feeds than the broodmate received (Drummond and Garc´ıa Chavelas, 1989). In control broods, pecking and differential feeding remained stable. Both food-deprived broodmates approximately doubled their baseline rate of begging, and the more a chick begged relative to its sibling, the greater the proportion of food offers it received. Sometimes it seemed that seniors enhanced their feeding priority partly by forcibly suppressing their rivals’ begging. Similarly, when black guillemot a-chicks in two-chick broods responded to experimental food deprivation by increasing their rate of attacking, their share of feeds increased from roughly half to

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

285

three-quarters, mostly because sibs were intimidated into crouching down when parents were offering food (Cook et al., 2000). Cohen Fernandez’ ´ (1988) descriptive study of an obligately siblicidal species was particularly revealing. During the few days the two brown booby chicks shared the nest, despite begging less frequently than its sibling the a-chick was fed five times more often, largely because its huge advantage in size and maturity (due to a 5-day hatch interval) enabled it not only to beg more conspicuously but also to influence when the b-chick begged. Aggression can also affect access to other parentally provided resources. For example, aggressive expulsion of egret, blue-footed booby, brown booby, western grebe, and skua chicks can deprive them of parental protection and brooding, exposing them to predation, thermal stress, and infanticidal attacks of colony neighbors (Dorward, 1962; Young, 1963; Nuechterlein, 1981; Mock, 1985; Drummond et al., 1986). Self-feeding chicks of many species benefit from parental brooding, but I am not aware of cases where they aggressively exclude each other from it (but see Rajala, 1962). We know extremely little about the benefits of aggression-induced deaths of broodmates. It is widely assumed that perpetrators feed, grow, and survive better after eliminating competitors, but only one experiment demonstrates (short-term) benefits for them or for their parents. Although Mock and Parker’s (1986) descriptive study found that great egret c-chicks survived better in broods where a senior nestmate died (an unusual event), senior chicks did not survive any better in broods where c-chicks died (the normal siblicidal event). Anyway, because naturally reduced broods are probably at some baseline disadvantage compared to unreduced broods (for example, their parents may be less successful foragers), demonstrating the benefits of a death requires an experiment. Even in obligate brood reducers the situation is far from clear. When siblicide was suppressed in masked boobies by temporarily removing the dominant chick during meals and whenever its size advantage grew to be threatening, the apparent result was increased nest success (Anderson, 1990). Per capita mortality of chicks increased, but having an extra chick for longer made it more likely that at least one chick would approach fledging. However, at age 50 days, experimental chicks (“doubletons”) were 12–17% lighter and 10–23% smaller than controls from broods with normal siblicide, probably indicating poor survival prospects (Fig. 10; and see Drummond, 1993). This study shows that normal siblicidal aggression benefits perpetrators in terms of personal growth and survival during the period of parental feeding (cf. interpretation in Anderson, 1990). Much hinges on whether parents respond to siblicidal deaths by reducing the amount of food they bring to the brood, thereby nullifying the perpetrator’s potential payoff. Theoretical models of brood reduction usually assume

286

HUGH DRUMMOND

FIG. 10. Effect of siblicide on growth of the surviving masked booby chick. Mean (  se) growth of normal perpetrators/survivors was superior to that of doubletons (pairs of chicks in broods where siblicide was experimentally prevented). (Modified from Anderson, 1990).

that parents maintain the same level of provisioning after a death, generating a bonus for survivors (e.g., Lack, 1954; Smith and Fretwell, 1974; Bonabeau et al., 1998). So it is surprising that, in nonaggressive brood reducers, the evidence from all seven experimental studies where food amounts were estimated shows that parents cut back on food delivery after disappearance of a chick (review in Ploger, 1997, and see Lessells, 1993), and that after natural brood reductions in such species broodmates do not appear to survive or grow any better than nestlings in unreduced broods (reviews in Stouffer and Power, 1991; Hillstrom and Olson, 1994). The critical test was made (for an aggressive, facultative brood-reducing passerine) when naturally reduced broods of magpies, Pica pica, were compared with broods where victims were experimentally replaced (Husby, 1986). Results were inconclusive: offspring from reduced broods survived the period of postfledging parental care better than those from broods where victims were replaced, but there was no evidence that they consumed more food and the increased reproductive success of their broods was not statistically significant. What about parental response to a death in typical siblicide species? Experimental reduction of cattle egret broods by removing either the a-chick or the c-chick for three days (Mock and Lamey, 1991) was interpreted as showing that parents respond to chick loss by neatly tailoring food provision to the new brood size (Mock and Parker, 1997, p. 135), but the data actually hint at a short-term food bonus for perpetrators of siblicide. Per capita consumption of chicks that remained in the nest increased significantly during

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

287

removals (a-chick removals: 24% increase, p < 0.06; c-chick removals: 62% increase, p < 0.001; F-test for pre vs during vs post, percentages calculated from Fig. 3, Mock and Lamey, 1991), although control broods for some unknown reason showed similar increases. It is critical that we analyze the effects of siblicidal deaths on parental provisioning up to fledging, in both facultative and obligate brood reducers. We also know extremely little about the feeding and survival consequences of aggression that fall short of causing nestling death, beyond the fact that juniors initially feed and grow less than seniors. In some avian species low fledging mass is associated with poor postfledging survival (e.g., Jarvis, 1974; Tinbergen and Boerlijst, 1990), and in species with nonaggressive young, later hatched broodmates can show poorer postfledging survival (e.g. Spear and Nur, 1994). In species with aggressive young, we do not usually know whether underfeeding and poor growth of subordinate chicks is sustained throughout the whole period of parental feeding, because measures of ingestion and growth usually cover only about the first half of it. Diminished fledging mass, survival beyond independence, and fitness seem likely, but blue-footed booby juniors actually catch up with their sibs well before independence, despite their persisting agonistic subordination: in 27 broods where both chicks survived, seniors and juniors of each sex did not differ in maximum mass or asymptotic linear dimensions, or show more variance in these features (Drummond et al., 1991). This result is consistent with behavioral and other evidence showing that senior chicks of this species often tolerate food-sharing even during food shortage and when hungry, and despite being capable of aggressively suppressing feeding by junior chicks (Drummond et al., 1986; Anderson and Ricklefs, 1995). Such aggressive restraint is consistent with kin selection and may be present in other facultatively siblicidal species, but we should not expect juniors to attain fully equal size and mass unless the capacity of seniors to influence food allocation declines as the nestling period advances (cf. Trivers, 1974). B. OBTAINING AND MAINTAINING DOMINANCE Dominance (a concept discussed in Drummond, 1999) is used here in the broad sense of an asymmetrical agonistic relationship between two individuals. It may be a simple all-out struggle between unequal contestants or involve an element of yielding by one contestant at the start of interactions. When yielding is to a particular individual, as a result of earlier agonistic encounters between them, it is a relationship of true dominance (cf. Bernstein, 1981; Drews, 1993), but when an animal consistently yields to or attacks conspecifics generally, then it usually reflects training as a loser or winner, respectively (Ginsburg and Allee, 1942). Yielding can also occur (due to a

288

HUGH DRUMMOND

congenital tendency or learning) as a response to some asymmetrical characteristic, such as size difference. Five observations support the claim that dominance itself is an important function of aggression in facultative brood reducers. First, chicks commonly attack their sibs at times when no food or other present benefit is being contested (e.g., Mock, 1985; Drummond et al., 1986). For example, grey herons, Ardea cinerea, commonly attacked their broodmates after a feeding bout ended, sustaining their hostilities for a full minute after the parent’s departure from the nest (Milstein et al., 1970). Second, aggression generally occurs in every brood in every season, implying that aggression happens whatever the inflow of food (see Section II.E). (Ospreys and great blue herons may be exceptions but this needs confirmation because hierarchies can be established and maintained by interactions that are subtle, infrequent, or occur before the start of observations.) Third, when a chick is replaced in the nest after a period of removal, intense aggression results (Mock and Lamey, 1991; Drummond and Osorno, 1992), consistent with the dominant chick reasserting its dominance. Fourth, aggression starts shortly after hatching and continues through fledging, even in broods where all members fledge and do so at a similar size and weight (Drummond et al., 1991; see Section II. E). Fifth, rates of aggression increase when dominance relationships are rendered unstable or weak by experimentally equalizing the ages of broodmates, even though food consumption actually increases (see Section II. C). Only the first three observations have been reported for obligate brood reducers (e.g., Gargett, 1978; Cohen Fernandez, ´ 1988; Anderson, 1991; respectively), and these are consistent with the simple function of promptly killing the nestmate, so in these species it is less evident that subordination of broodmates is a function of aggression. In self-feeding chicks, there is evidence of dominance hierarchies in canvasback and redhead ducks (Collias and Collias, 1956), the magpie goose, Anseranas semipalmata (Davies, 1963), the Canada goose (Radesater, ¨ 1974), the greylag goose (Kalas, 1977; Lorenz, 1991), the red junglefowl (Kruijt, 1964), the domestic chicken (Baeumer, 1955, in Nice, 1962), the western capercaillie, the black grouse, and the willow grouse (Rajala, 1962). In brood-reducing species, dominance gives a chick priority in competition for current parental investment and can also assure that its dominance and priority will endure through the nestling period. The latter is clearly illustrated by the agonistic inertia in blue-footed booby broods composed of a male and younger female. After he, being the smaller sex, is outgrown by her (at an age of about five weeks), he remains dominant despite the fact that she eventually grows to be 21% heavier than him, simply because their early agonistic interactions have trained her as a loser and him as a winner (Drummond and Osorno, 1992; Drumond and Canales, 1998). Even in great

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

289

egrets, where nestling violence is unritualized and submissive postures are absent, it has been suggested that the function of much early aggression is to train the victim into a degree of subordinance that will prejudice its performance in future competition for parentally provided food (e.g., Mock, 1985; Mock and Parker, 1997). Where parental provisioning may eventually be insufficient, such that one or more brood members will be discarded, dominant status is evidently what prevents the individual chick from becoming one of the victims: in all brood reducing species with nestling agonism, mortality is concentrated on last-hatched, subordinate brood members. Potentially, dominance between broodmates also influences survival and competitive interactions with broodmates and others after fledging, independent of its effects on nestling resource competition. Certainly there is plenty of documented scope for benefits to accrue to dominant juveniles and adults. Dominance (between nonsibs) is associated with superior survival in juvenile birds (De Laet, 1985; Arcese and Smith, 1985) and also with greater survival and reproductive success in adults of numerous avian species (review in Arcese and Smith, 1985). Dominance interactions occur among acorn woodpecker, Melanerpes formicivorous, broodmates after independence and possibly well into adulthood (Stanback, 1994). Simmons (1988, 1989, 1991) speculated that the black eagle’s second chick functions as the sparring partner that enables the first one to develop its aggressive potential for later use against adult competitors. In species with self-feeding chicks, there is often rich opportunity for dominance in infancy to translate into social benefits after fledging. Tetraonids and Phasianids typically spend part of their life in flocks, and in Rio Grande turkeys, Meleagris gallopavo, vigorous fighting and stable dominance hierarchies occur among juvenile and sibling groups in flocks (Watts and Stokes, 1971). Flocks of geese are often made up of families comprising mated pairs and their adult offspring, and some species of goose show communal nesting (e.g., Davies, 1963). Among groupings of geese, dominance relations within and between families tend to pervade social relations (e.g., Lamprecht, 1985; Lorenz, 1991). However, we do not know for any species of bird whether agonistic dominance during infancy affects subsequent dominance, survival, or competitiveness. There are hints of possible effects. In one experiment, dominance status acquired in artificially composed broods of Japanese quail, Coturnix coturnix, during the first 6 weeks of life predicted the outcome of subsequent agonistic encounters with unfamiliar peers (Boag and Alway, 1980). But the experimental situation was highly unnatural and the effect could have been due to intrinsic differences among chicks affecting performance during both stages, rather than to early experience with broodmates. Intriguingly, little egrets that were low in the brood-size hierarchy at fledging subsequently

290

HUGH DRUMMOND

bred less successfully as adults than those that were high in the hierarchy (Thomas et al., 1999), implying some sort of long-term handicap due to low rank as a nestling. However, we do not know if the fledgling size hierarchy correlated with an agonistic hierarchy, and in this species broodmate aggression is very mild (Inoue, 1985). Whether there are long-term developmental consequences of being dominant or subordinate during infancy is one of the most fascinating unanswered questions about broodmate agonism. C. DEFENSE AND SUBMISSION The avoidance behaviors (e.g., crouching, head-hiding, withdrawing, fleeing) performed by younger broodmates of all species, surely function to reduce tissue damage, and might also minimize training as a loser. Interestingly, junior chicks in a minority of species also show submissive displays or aggression, including fighting back and initiating aggression. Interspecific variation in agonism of subordinate chicks has attracted little attention and its adaptiveness is poorly understood. Blue-footed booby juniors do not ordinarily initiate aggression or fight back, but will do so repeatedly in the course of a natural dominance inversion. These unusual events take place over a period of many days, during which dominant status seems to switch back and forth until a final winner emerges (e.g., Drummond et al., 1991). In more normal circumstances, blue-footed booby juniors occasionally initiate aggression by pecking the dominant chick briefly, as if to assess the rival (who usually pecks back vigorously), and this sort of probing can be induced by experimentally reducing the size of the dominant chick (Drummond and Osorno, 1992). Hence, in this booby both probing pecks and fighting back appear to function in dominance inversion. Either fighting back or initiating aggression occur at very low rates in junior chicks of the black-legged kittiwake (Braun and Hunt, 1983), the South Polar skua (Procter, 1975), the tawny eagle (A. rapax, Steyn, 1973), the osprey (Poole, 1979; Jamieson et al., 1983), the black guillemot (Cook et al., 2000), and the American black oystercatcher (Groves, 1984), and at a much higher rate in the brown pelican (Pinson and Drummond, 1993; Ploger, 1997). Fighting back is routine in great egrets (Mock, 1985) and cattle egrets, but in cattle egrets, at least, junior chicks do not initiate aggression (Fujioka, 1985a). Fighting back by ardeids seems to be related to immediately obtaining more food or dissuading attacks. Submissive displays are assumed to inhibit attacking by the other bird, although this has not been proven. The bill-down-and-face-away posture is the habitual response of blue-footed booby junior chicks to aggression, and it is also often emitted (usually briefly) by attacked senior chicks (e.g., Drummond and Canales, 1998).

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

291

IV. CONCLUSION Nestling aggressiveness can be influenced by amount of food consumed, relative size/age of nestlings, agonistic experience, behavior of the broodmate, and parental behavior; probably also by spatial concentration of food, brood size, and maturation; and possibly even by food-parcel size, hormones, and sex. This list will surely be extended. We have only started to explore the nature or shape of the relationships between the identified factors and aggression, and to discover how the factors interact with each other. Encouragingly, tests of interaction between agonistic experience and size difference (Drummond and Osorno, 1992), and between food amount consumed and size difference (Rodr´ıguez-Girones ´ et al., 1996), both yielded significant effects. However, nearly all experimental work has been on facultative brood reducers and we will not achieve a comprehensive understanding of the control and function of nestling aggression until basic questions about obligate brood reducers and self-feeding chicks are answered. The documented variation in types of agonistic relationship between broodmates is substantial. For example, the blue-footed booby shows discrete dominant and subordinate roles resulting from training. Brown pelican broods show a linear hierarchy in which seniors are aggressive and juniors perform submissive displays but are also frequently aggressive. Broods of the cattle egret also show a linear hierarchy, but pitched dyadic fights are commonplace and submissive displays are not used, suggesting a situation closer to an all-out contest, except that junior chicks do not initiate attacks on seniors and sometimes all brood members survive. In three-chick broods, whereas cattle egret a-chicks direct little aggression at either broodmate, brown pelican a-chicks are as aggressive as any chick in the brood. In the brown booby, the aggression of senior chicks is so intense and apparently unconditional that the junior one is usually killed before it even has an opportunity to go on the offensive. However, when we postponed siblicide by tethering nestmates (Drummond, unpubl. data), both of them showed fierce aggression except when temporarily intimidated by a barrage of attacks, and their behavior resembled an all-out contest. Unfortunately, available descriptions of hierarchies in self-feeding chicks do not give a clear idea of their nature. Obtaining such descriptions should surely be a priority. We cannot yet account for these interspecific differences in type of agonistic relationship at the proximate level. We know that individuals of a few brood-reducing species can adjust their aggression in response to personal food ingestion, relative broodmate size, agonistic experience, and so on, but experimental coverage of different species has been so sparse and uneven that few comparisons can be made. We should use similar experimental methods on different species and take roles in hierarchies into account. We could

292

HUGH DRUMMOND

start by looking for effects of food amount, agonistic experience and relative size in obligate brood reducers and self-feeding chicks. We should delve into behavioral differences within broods, asking for example, whether the mild aggressiveness of cattle egret a-chicks (in comparison with b-chicks) is due to privileged food intake or to social stimuli. We can ask whether blue-footed booby b-chicks in three-chick broods are trained simultaneously as winner and also as loser, whether they can adopt either role with equal effectiveness, and what are the stimuli that govern their choice of role. Tackling these sorts of questions, with descriptive and experimental methods, will require a focus on individual status, agonistic experience, food ingestion, and so on. We need to dissect hierarchies, and we should do this by analyzing individual roles. True dominance has not been shown for any species, although Kruijt (1964) suspected that it emerges in junglefowl broods after age 7 weeks, and trained winning and losing have only been looked for and demonstrated in a single species. I suspect some such training occurs in most avian species that show nestling aggression, although in obligately siblicidal species training effects are probably very weak or absent. Prima facie evidence for similar training is there in every descriptive account that mentions a brood dominance hierarchy, and even reportedly nonviolent hierarchies (e.g., in wild oystercatcher broods, Safriel, 1981) probably bear witness to earlier or undercover violent exchanges that escaped observation. Perhaps the clearest indication that chicks assume particular agonistic roles (probably as a result of training) rather than simply engaging in an all-out contest is the occurrence of dominance interactions mediated by displays. Displays have not been consistently reported and discriminating display postures and coordinations from unritualized behavior can be difficult, although calls are unambiguously communicative. Even so, threat displays have been noted in at least eleven species, seven of which use calls, and (silent) submissive displays in at least eight species, plus unspecified species of cranes whose chicks purr while cowering (Table I). Facultative brood reducers and species with self-feeding chicks are well represented in this list, but no species with obligate brood reduction shows evident aggressive or submissive displays. Scrutiny for displays will doubtless uncover examples in other species with nestling aggression, but experiments are urgently needed to test for trained winning and losing. Analysis of proximate control should go hand in hand with analysis of function. Competition for status in the brood dominance hierarchy explains why senior chicks of some brood-reducing species fight in the absence of food, attack from an early age, fight more frequently against similar-sized opponents, sustain aggression throughout the nestling period, and escalate when attacked. In species with self-feeding chicks, status in the brood hierarchy

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

293

TABLE I SPECIES REPORTED TO SHOW AGONISTIC DISPLAYS BETWEEN BROODMATES

Threat1

Submissive posture2

Brown pelican Osprey

Call Stretch up

BDFA3

Blue-footed booby

Upright posture Call

BDFA3

Western grebe South Polar skua

Call Upright posture Call

Turn away

Black-legged kittiwake Bald ibis Cranes Oystercatcher

Beak hide Beak hide Cower and purr

Reference Pinson and Drummond, 1993 Poole, 1979, 1982 Jamieson et al., 1983 Forbes, 1991 Nelson, 1978 Drummond et al., 1986 Pers. observ. Nuechterlein, 1981 Young, 1963 Spellerberg, 1971b Cullen, 1957 Braun and Hunt, 1983 Hirsch, 1979 Swengel et al., 1996

Dash forward Call Dash forward

Safriel, 1981

Davies, 1963

Canada goose

Concert Call Cackling

Greylag goose Junglefowl Domestic chicken

Frontal threat Challenge

American black oystercatcher Magpie goose

Groves, 1984

Appeasing movement Face away Call

Radesater, ¨ 1974 Kalas, 1977 Kruijt, 1964 Nice, 1962

1

Nonviolent stereotyped behavioral pattern that appears to elicit submission. Stereotyped behavioral pattern that appears to discourage aggression. Does not include patterns that appear simply to physically remove chick from risk of being attacked. 3 Bill-down-and-face-away posture. 2

is the most likely function of nestling agonism, but the practical benefits of high status are obscure and may well accrue after fledging. As expected, food deprivation influences aggressiveness in facultative brood reducers, but we need to test the converse prediction that neither obligate brood reducers nor self-feeding chicks respond in this way. I expect that comparative functional analysis will eventually enable us to understand the conspicuous interspecific differences in agonism of junior chicks. Junior chicks with very poor prospects of surviving and fledging alongside their broodmates (because parents will not bring enough food to

294

HUGH DRUMMOND

fledge them all) may fight fiercely to supplant those broodmates, whereas those with good prospects of surviving alongside broodmates may optimize those prospects by accepting a subordinate role (Drummond, 1993). The first category (exemplified by the brown booby) should show unconditional aggression except when physically overwhelmed, no submissiveness beyond temporary evasion and avoidance, and little or no susceptibility to training as a loser; the second category (exemplified by the blue-footed booby) should renounce aggression, should signal subordination (while monitoring the rival for any weakness that could open the door to a forced role inversion), and should be trainable as a loser. Critically, elder sibs of the first category should attack unconditionally to eliminate the growing threat before it becomes unmanageable, whereas elder sibs of the second category can safely show aggressive restraint as long as the junior chick is manifestly under control. In this scenario, which is susceptible to functional and causal analysis, it is the ecological prospects of junior chicks that drive the evolution of agonism and its proximate control in both juniors and seniors.

V. SUMMARY Agonism among avian broodmates is now better known and understood than agonism among infant broodmates or littermates in any other group of vertebrates. Birds whose nestlings show broodmate agonism include species that practice obligate brood reduction, species that practice facultative brood reduction and species in which chicks feed themselves. Research into the proximate control of agonism has concentrated almost exclusively on facultative brood reducers. Studies show that a chick’s aggression can be influenced by food deprivation, relative size/age of the broodmate, agonistic experience with broodmates, behavior of the broodmate, and parental behavior. Other likely factors include spatial concentration of food, brood size, maturation and, possibly, food parcel size, hormones, and sex. Food-sensitive aggression has now been demonstrated in four species of facultative brood reducers and seems to be an exception to the rule that food deprivation per se does not induce aggression in vertebrates. Acquisition of an aggressive or submissive personality through social interaction with broodmates represents an ecologically relevant example of trained winning and losing, a phenomenon postulated by comparative psychologists but seldom tied so clearly to a natural context. Analysis of the functions of broodmate agonism is at an early stage. Mathematical models related to brood reduction theory and parent–offspring conflict theory are relevant to broodmate agonism, but have scarcely been empirically tested. Inferring from the context, proximate control, and

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

295

short-term consequences of aggression, it serves two major functions: securing status in the brood dominance hierarchy and selfish biasing of parental investment (particularly food provisioning). Dominant status in broodreducing species seems to assure personal food consumption and exempt the individual from being the victim of siblicide. Experiments are needed to determine whether siblicide affects parental food provisioning and whether siblicide benefits perpetrators by increasing their food consumption, survival, and fitness. Longer term benefits of broodmate dominance in terms of greater dominance or competitiveness during juvenile or adult life have not been explored—and could be an important function of agonism, particularly in species with self-feeding chicks. Subordinate chicks of different species vary greatly in their agonistic behavior, and this variation may be related to their prospects of fledging alongside their broodmates. Understanding the diverse types of agonistic relationship found in different species requires focusing on individual roles in hierarchies. It also demands that we open the door to illuminating comparisons by extending descriptive and experimental analyses to obligate brood reducers and selffeeding chicks.

Acknowledgments I thank Rodolfo Dirzo, Robyn Hudson, Constantino Macias Garc´ıa, and Jose´ Luis Osorno for comments on an earlier draft, Cristina Rodr´ıguez Juarez ´ for help with literature searches and with figure and manuscript preparation, Judy Stamps for help getting literature, and Chris Perrins for hospitality at the Edward Grey Institute. The field research in which this review is grounded was financed by grants from the Universidad Nacional Autonoma ´ de Mexico ´ (DGAPA-IN211491), CONACYT (D112-903581, PCCNCNA-031528,), the National Geographic Society (3065-85, 4535-91), and the Conservation and Research Foundation.

References Allee, W. C., Collias, N. E., and Lutherman, C. Z. (1939). Modification of the social order in flocks of hens by the injection of testosterone propionate. Physiol. Zool. 12, 412–440. Anderson, D. J. (1989). The role of hatching asynchrony in siblicidal brood reduction of two booby species. Behav. Ecol. Sociobiol. 25, 363–368. Anderson, D. J. (1990). Evolution of obligate siblicide in boobies. II. Food limitation and parent– offspring conflict. Evolution 44, 2069–2082. Anderson, D. J. (1991). Parent blue-footed boobies are not infanticidal. Ornis. Scand. 22, 169– 170. Anderson, D. J. (1995). The role of parents in siblicidal brood reduction of two booby species. Auk 112, 860–869. Anderson, D. J., and Ricklefs, R. E. (1995). Evidence of kin-selected tolerance by nestlings in a siblicidal bird. Behav. Ecol. Sociobiol. 37, 163–168.

296

HUGH DRUMMOND

Anderson, D. J., Budde, C., Apanius, V., Martinez Gomez, J. E., Bird, D. M., and Weathers, W. W. (1993). Prey size influences female competitive dominance in nestling American kestrels. Ecology 74, 367–376. Arcese, P., and Smith, J. N. M. (1985). Phenotypic correlates and ecological consequences of dominance in song sparrows. J. Anim. Ecol. 54, 817–830. Archer, J. (1976). The organization of aggression and fear in vertebrates. In “Perspectives in Ethology” (P. P. G. Bateson and P. Klopfer, eds.). Vol. 2, pp. 231–298. Plenum New York. ¨ Tierpsychol. 12, 387–401. Baeumer, E. (1955). Lebensart des Haushuhns. Z fur Beebe, C. W. (1907). Notes on the early life of loon chicks. Auk 24, 34–41. Bernstein, I. S. (1981). Dominance: The baby and the bathwater. Behav. Brain Sci. 4, 419–457. Boag, D. A., and Alway, J. H. (1980). Effects of social environment within the brood and dominance rank in gallinaceous birds. Can. J. Zool. 58, 44–49. Bonabeau, E., Deneubourg, J. L., and Theraulaz, G. (1998). Within-brood competition and the optimal partitioning of parental investment. Am. Nat. 152, 419–427. Bortolotti, G. R. (1986). Influence of sibling competition on nestling sex ratios of sexually dimorphic birds. Am. Nat. 127, 495–507. Braun, B. M., and Hunt, G. L., Jr. (1983). Brood reduction in black-legged kittiwakes. Auk 100, 469–476. Burger, J., Gochfield, M., and Boarman, W. I. (1988). Experimental evidence for sibling recognition in common terns (Sterna hirundo). Auk 105, 142–148. Cash, K. J., and Evans, R. M. (1986). Brood reduction in the American white pelican (Pelecanus erythrorhynchos). Behav. Ecol. Sociobiol. 18, 413–418. Clark, A. B., and Wilson, D. S. (1985). The onset of incubation in birds. Am. Nat. 125, 603–611. Cohen Fernandez, ´ E. J. (1988). La reduccion ´ de la nidada en el bobo cafe´ (Sula leucogaster nesiotes Heller and Snodgrass 1901). Undergraduate thesis, UNAM, Fac. de Ciencias. Collias, N. E., and Collias, E. (1956). Some mechanisms of family integration in ducks. Auk 73, 378–400. Cook, M. I., Monaghan, P., and Burns, M. D. (2000). Effects of short-term hunger and competitive asymmetry on facultative aggression in nestling black guillemots Cepphus grylle. Behav. Ecol. 11, 282–287. Cooper, J. (1980). Fatal sibling aggression in pelicans: A review. Ostrich 51, 173–186. Craighead, J. J., and Craighead, F. C. (1956). “Hawks, Owls and Wildlife.” Stackpole, Harrisburg PA. Creighton, J. C., and Schnell, G. D. (1996). Proximate control of siblicide in cattle egrets: A test of the food amount hypothesis. Behav. Ecol. Sociobiol. 38, 371–377. Cullen, E. (1957). Adaptations in the Kittiwake to cliff-nesting. Ibis 99, 275–302. David, S., and Berrill, M. (1987). Siblicidal attacks by great blue heron, Ardea herodias, chicks in a southern Ontario heronry. Can. Field Natur. 101, 105–107. Davies, S. J. J. F. (1963). Aspects of the behaviour of the Magpie goose, Anseranas semipalmata. Ibis 105, 76–98. De Laet, J. V. (1985). Dominance and aggression in juvenile great tits, Parus major major L. in relation to dispersal. In “Behavioural Ecology: Ecological Consequences of Adaptive Behaviour” (R. M. Sibly and R. H. Smith, eds.), pp. 375–381. Blackwell Scientific Publications, Oxford, UK. Dorward, E. F. (1962). Comparative biology of the white booby and brown booby, Sula spp. at Ascension. Ibis. 103b, 174–220. Drews, C. (1993). The concept and definition of dominance in animal behaviour. Behaviour 125, 283–313. Drummond, H. (1987). A review of parent–offspring conflict and brood reduction in the Pelecaniformes. Col. Waterbirds 10, 1–15.

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

297

Drummond, H. (1989). Parent–offspring conflict and brood reduction in boobies. In “Proceedings of XIX International Ornithological Congress,” pp. 1244–1253. Univ. of Ottawa Press, Ottawa. Drummond, H. (1993). Have avian parents lost control of offspring aggression? Etolog´ıa 3, 187–198. Drummond, H. (1999). Agonism and dominance in nestling birds. In “Proceedings of 22nd International Ornithological Congress, Durban” (N. Adams and R. Slotow, eds.), pp. 1621– 1631. Birdlife South Africa, Johannesburg. Drummond, H. (2001). A reevaluation of the role of food in broodmate aggression. Anim. Behav., 61. Drummond, H., and Canales, C. (1998). Dominance between booby nestlings involves winner and loser effects. Anim. Behav. 55, 1669–1676. Drummond, H., and Garc´ıa Chavelas, C. (1989). Food shortage influences sibling aggression in the blue-footed booby. Anim. Behav. 37, 806–819. Drummond, H., and Osorno, J. L. (1992). Training siblings to be submissive losers: Dominance between booby nestlings. Anim. Behav. 44, 881–893. Drummond, H., Gonzalez, E., and Osorno, J. L. (1986). Parent–offspring cooperation in the blue-footed booby (Sula nebouxii): Social roles in infanticidal brood reduction. Behav. Ecol. Sociobiol. 19, 365–372. Drummond, H., Osorno J. L., Garc´ıa, C., Torres, R., and Merchant, H. (1991). Sexual dimorphism and sibling competition: Implications for avian sex ratios. Am. Nat. 138, 623–641. Duncan, I. J. H., and Wood-Gush, D. G. M. (1971). Frustration and aggression in the domestic fowl. Anim. Behav. 19, 500–504. Edwards, T. C., and Collopy, M. W. (1983). Obligate and facultative brood reduction in eagles: An examination of the factors that influence fratricide. Auk 100, 630–635. Edwards, T. C., Collopy, M. W., Steenhof, K., and Kochert, M. N. (1988). Sex ratios of fledgling golden eagles. Auk 105, 793–796. Evans, R. M. (1970). Imprinting and mobility in young ring-billed gulls, Larus delawarensis. Anim. Behav. Monogr 3, 193–248. Evans, R. M., and McMahon, B. (1987). Within-brood variation in growth and condition in relation to brood reduction in the American white pelican. Wilson Bull. 99, 190–201. Forbes, L. S. (1991). Hunger and food allocation among nestlings of facultatively siblicidal ospreys. Behav. Ecol. Sociobiol. 29, 189–195. Forbes, L. S. (1993). Avian brood reduction and parent–offspring “conflict.” Am. Nat. 142, 82–117. Forbes, L. S., and Mock, D. W. (1994). Proximate and ultimate determinants of avian brood reduction. In “Infanticide and Parental Care” (S. Parmigiani and F. S. vom Saal, eds.), pp. 237–256. Harwood, Chur, Switzerland. Frank, L. G., Glickman, S. E., and Licht, P. (1991). Fatal sibling aggression, precocial development, and androgens in neonatal spotted hyaenas. Science 252, 702–704. Fraser, D., and Thompson, B. K. (1991). Armed sibling rivalry among suckling piglets. Behav. Ecol. Sociobiol. 29, 9–16. Fujioka, M. (1985a). Sibling competition and siblicide in asynchronously hatching broods of the cattle egret Bubulcus ibis. Anim. Behav. 33, 1228–1242. Fujioka, M. (1985b). Food delivery and sibling competition in experimentally even-aged broods of the cattle egret. Behav. Ecol. Sociobiol. 17, 67–74. Gargett, V. (1978). Sibling aggression in the black eagle in the Matopos, Rhodesia. Ostrich 49, 57–63. Gerhardt, R. P., Gerhardt, D. M., and Vasquez, ´ M. A. (1997). Siblicide in Swallow-tailed Kites, Wilson Bull. 109, 112–120.

298

HUGH DRUMMOND

Ginsburg, B., and Allee, W. C. (1942). Some effects of conditioning on social dominance and subordination in inbred strains of mice. J. Physiol. Zool. 15, 485–506. Godfray, H. C. J., and Harper, A. B. (1990). The evolution of brood reduction by siblicide in birds. J. Theor. Biol. 145, 163–175. Groves, S. (1984). Chick growth, sibling rivalry, and chick production in American black oystercatchers. Auk 101, 525–531. Guerra, M. del C., and Drummond, H. (1995). Reversed sexual size dimorphism and parental care: Minimal division of labour in the blue-footed booby. Behaviour 132, 479–496. Hartsock, T. G., and Graves, H. B. (1976). Neonatal behaviour and nutrition-related mortality in domestic swine. J. Anim. Sci. 42, 235–241. Heinroth, O., and Heinroth, M. (1924–33). Die Vogel ¨ Mitteleuropas. 4 vols., Berlin. Hillstrom, L., and Olson, K. (1994). Advantages of hatching asynchrony in the pied flycatcher Ficedula hypoleuca. J. Avian Biol. 25, 205–214. Hirsch, U. (1979). Studies of West Palearctic birds: Bald ibis. British Birds 72, 313–324. Husby, M. (1986). On the adaptive value of brood reduction in birds: Experiments with the magpie Pica pica. J. Anim. Ecol. 55, 75–83. Ingram, C. (1959). The importance of juvenile cannibalism in the breeding biology of certain birds of prey. Auk 76, 218–226. Inoue, Y. (1985). The process of asynchronous hatching and sibling competition in the little egret Egretta garzetta. Col. Waterbirds 8, 1–12. Irons, D. B. (1992). Aspects of foraging behavior and reproductive biology of the black-legged kittiwake. Ph.D. dissertation, Univ. of California, Irvine. Jamieson, I. G., Seymour, N. R., Bancroft, R. P., and Sullivan, R. (1983). Sibling aggression in nestling ospreys in Nova Scotia. Can. J. Zool. 61, 466–469. Jarvis, M. J. F. (1974). The ecological significance of clutch size in the South African gannet (Sula capensis). J. Anim. Ecol. 43, 1–17. Kalas, V. S. (1977). Ontogenic und Funktion der Rangordnung innerhalb einer Geschwisterschar von Graugansen ¨ (Anser anser L.). Z. Tierpsychol. 45, 174–198. Kear, J. (1970). The adaptive radiation of parental care in waterfowl. In “Social Behaviour of Birds and Mammals” (J. H. Crook, ed.). pp. 357–392. Academic Press, London. Kruijt, J. P. (1964). Ontogeny of social behaviour in Burmese red junglefowl (Gallus gallus spadiceus) Bonnaterre. Behaviour, Supplement XII. E. J. Brill, Leiden. Lack, D. (1954). “The Natural Regulation of Animal Numbers.” Clarendon Press, Oxford. Lamprecht, J. (1985). Structure and causation of the dominance hierarchy in a flock of barheaded geese (Anser indicus). Behaviour 96, 28–48. Layne, J. N. (1982). Status of sibling aggression in Florida sandhill cranes. J. Field Ornithol. 53, 272–274. Leshner, A. I. (1981). The role of hormones in the control of submissiveness. In “Multidisciplinary Approaches to Aggression Research.” (P. F. Brain and D. Benton, eds.), pp. 309–321. Elsevier, Amsterdam. Leshner, A. I. (1983). The hormonal responses to competition and their behavioural significance. In “Hormones and Aggressive Behaviour” (B. B. Svare, ed.), pp. 393–404. Plenum Press, New York. Lessells, C. M. (1993). The cost of reproduction: Do experimental manipulations measure the edge of the options set? Etologia 3, 95–111. Lewis, H. F. (1929). “The Natural History of the Double-Crested Cormorant.” Ru-Mi-Lou Books, Ottawa, Canada. Lorenz, K. (1997). “Here am I–Where are you?” Harcourt Brace Jovanovich, New York. Lougheed, L. W., and Anderson, D. J. (1999). Parent blue-footed boobies suppress siblicidal behavior of offspring. Behav. Ecol. Sociobiol. 45, 11–18.

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

299

Machmer, M. M., and Ydenberg, R. C. (1998). The relative roles of hunger and size asymmetry in sibling aggression between nestling ospreys Pandion haliaetus. Can. J. Zool. 76, 181– 186. Magrath, R. (1989). Hatch asynchrony and reproductive success in the blackbird. Nature 339, 536–538. Mart´ınez-Vilalta, A., and Motis, A. (1992). Family Ardeidae (herons). In “Handbook of the Birds of the World,” vol. 1. (J. del Hoyo, A. Elliott, and J. Sargatal, eds.). Lynx Editions, Barcelona. McBride, G. (1963). The teat order and communication in young pigs. Anim. Behav. 11, 53–56. Mendall, H. L. (1936). The home life and economic status of the double-crested cormorant (Phalacrocorax auritus auritus) (Lesson). Maine Bull. 39, 1–159. Meyburg, B. U. (1974). Sibling aggression and mortality among nestling eagles. Ibis 116, 224–228. Meyburg, B. U. (1977). Sibling aggression and cross-fostering of eagles. In “Endangered Birds” (S. A. Temple, ed.). Univ. of Wisconsin Press, Madison. Miller, R. S. (1973). The brood size of cranes. Wilson Bull. 85, 436–441. Milstein, P., Le, S., Prestt, I., and Bell, A. A. (1970). The breeding cycle of the grey heron. Ardea 58, 171–257. Mock, D. W. (1984a). Infanticide, siblicide and avian nestling mortality. In “Infanticide: Comparative and Evolutionary Perspectives” (G. Hausfater and S. B. Hrdy, eds.), pp. 3–30. Aldine, New York. Mock, D. W. (1984b). Siblicidal aggression and resource monopolization in birds. Science 225, 731–733. Mock, D. W. (1985). Siblicidal brood reduction: The prey-size hypothesis. Am. Nat. 125, 327–343. Mock, D. W. (1987). Siblicide, parent-offspring conflict and unequal parental investment by egrets and herons. Behav. Ecol. Sociobiol. 20, 247–256. Mock, D. W., and Forbes, L. S. (1992). Parent–offspring conflict: A case of arrested development? Trends Ecol. Evol. 10, 130–134. Mock, D. W., and Lamey, T. C. (1991). The role of brood size in regulating egret sibling aggression. Am. Nat. 138, 1015–1026. Mock, D. W., and Parker, G. A. (1986). Advantages and disadvantages of ardeid brood reduction. Evolution 40, 459–470. Mock, D. W., and Parker, G. A. (1997). “The Evolution of Sibling Rivalry.” Oxford Univ. Press, UK. Mock, D. W., and Ploger, B. J. (1987). Parental manipulation of optimal hatch asynchrony in cattle egrets: An experimental study. Anim. Behav. 35, 150–160. Mock, D. W., Lamey, T. C., Williams, C. F., and Ploger, B. J. (1987a). Proximate and ultimate roles of food amount in regulating egret sibling aggression. Ecology 68, 1760–1772. Mock, D. W., Lamey, T. C., Williams, C. F., and Pelletier, A. (1987b). Flexibility in the development of heron sibling aggression: An intraspecific test of the prey-size hypothesis. Anim. Behav. 35, 1368–1393. Nelson, J. B. (1978). “The Sulidae: Gannets and Boobies.” Oxford Univ. Press, UK. Newton, I. (1977). Breeding strategies in birds of prey. Living Bird 16, 51–82. Newton, I. (1979). “Population Ecology of Raptors. ” Berkhamsted, Herts, Poyser, UK. Nice, M. M. (1962). Development of behavior in precocial birds. Trans. Linnean Soc. VIII, 1–211. Noseworthy, C. M., and Lien, J. (1976). Ontogeny of nesting habitat recognition and preference in neonatal herring gull chicks, Larus argentatus Pontoppidan. Anim. Behav. 24, 637–651. Nuechterlein, G. L. (1981). Asynchronous hatching and sibling competition in western grebe broods. Can. J. Zool. 59, 994–998. Nunez ˜ de la Mora, A., Drummond, H., and Wingfield, J. (1996). Hormonal correlates of

300

HUGH DRUMMOND

dominance and starvation-induced aggression in chicks of the blue-footed booby. Ethology 102, 748–761. O’Connor, R. J. (1978). Brood reduction in birds: Selection for fratricide, infanticide, and suicide. Anim. Behav. 26, 79–96. Osorno, J. L., and Drummond, H. (1995). The function of hatching asynohrony in the bluefooted booby. Behav. Ecol. Sociobiol. 37, 265–273. Palestis, B. G., and Burger, J. (1999). Individual sibling recognition in experimental broods of common tern chicks. Anim. Behav. 58, 375–381. Parker, G. A. (1985). Models of parent–offspring conflict. V. Effects of the behavior of the two parents. Anim. Behav. 33, 519–533. Pinson, D., and Drummond, H. (1993). Brown pelican siblicide and the prey-size hypothesis. Behav. Ecol. Sociobiol. 32, 111–118. Ploger, B. J. (1997). Does brood reduction provide nestling survivors with a food bonus? Anim. Behav. 54, 1063–1076. Ploger, B. J., and Mock, D. W. (1986). The role of sibling aggression in distribution of food to nestling cattle egrets (Bubulcus ibis). Auk 103, 768–776. Poole, A. L. (1979). Sibling aggression among nestling ospreys in Florida Bay. Auk 96, 415–417. Poole, A. L. (1982). Brood reduction in temperate and subtropical ospreys. Oecologia 53, 111– 119. Procter, D. L. C. (1975). The problem of chick loss in the South Polar skua Catharacta maccormicki. Ibis 117, 452–459. Radesater, ¨ T. (1974). On the ontogeny of orienting movements in the triumph ceremony of two species of geese (Anser anser and Branta canadensis). Behaviour 50, 1–50. Radesater, ¨ T. (1976). Individual sibling recognition in juvenile Canada geese (Branta canadensis). Can. J. Zool. 54, 1069–1072. Rajala, P. (1962). On the ecology of the broods of Capercaillie (Tetrao urogallus), Black Grouse (Lyrurus tetrix), and Willow Grouse (Lagopus lagopus). Suom Riista 15, 28–52. Ramos-Fernandez, G., Nunez ˜ de la Mora, A., Wingfield, J., and Drummond, H. (2000). Endocrine correlates of dominance in chicks of the blue-footed booby (Sula nebouxii): Testing the challenge hypothesis. Ethol. Ecol. Evol. 12, 27–34. Rodr´ıguez-Girones, ´ M. A., Drummond, H., and Kacelnik, A. (1996). Effect of food deprivation on dominance status in blue-footed booby (Sula nebouxii) broods. Behav. Ecol. 7, 82–88. Rowell, T. E. (1974). The concept of social dominance. Behav. Biol. 11, 131–154. Safriel, U. N. (1981). Social hierarchy among siblings in brood of the oystercatcher Haematopus ostralegus. Behav. Ecol. Sociobiol. 9, 59–63. Schwabl, H. (1993). Yolk is a source of maternal testosterone for developing birds. Neurobiology 90, 11446–11450. Schwabl, H. (1996). Maternal testosterone in the avian egg enhances postnatal growth. Comp. Biochem. Physiol. 114A, 271–276. Schwabl, H., Mock, D.W., and Gieg, J. A. (1997). A hormonal mechanism for parental favoritism. Nature 386, 231. Scott, J. P., and Marston, M. V. (1953). Nonadaptive behavior resulting from a series of defeats in fighting mice. J. Abnorm. Soc. Psychol. 48, 417–428. Simmons, R. (1988). Offspring quality and the evolution of cainism. Ibis 130, 339–357. Simmons, R. (1989). The Cain and Abel riddle in eagles and other birds. Afr. Wildl. 43, 35–43. Simmons, R. (1991). Offspring quality and sibling aggression in the black eagle. Ostrich 62, 89–92. Smale, L., Holekamp, K. E., Weldele, M., Frank, L. G., and Glickman, S. E. (1995). Competition and cooperation between litter-mates in the spotted hyaena, Crocuta crocuta. Anim. Behav. 50, 671–682.

CONTROL AND FUNCTION OF AGONISM IN AVIAN BROODMATES

301

Smith, C. C., and Fretwell, S. D. (1974). The optimal balance between size and number of offspring. Am. Nat. 108, 499–506. Spear, L., and Nur, N. (1994). Brood size, hatching order, and hatching date: Effects on four life history stages from hatching to recruitment in western gulls. J. Anim. Ecol. 63, 283–298. Spellerberg, I. F. (1971a). Aspects of McCormick skua breeding biology. Ibis 113, 357–363. Spellerberg, I. F. (1971b). Breeding behaviour of the McCormick skua Catharacta maccormicki in Antarctica. Ardea 59, 189–230. Stamps, J., Kus, B., Clark, A., and Arrowood, P. (1990). Social relationships of fledgling budgerigars, Melopsitticus undulatus. Anim. Behav. 40, 688–700. Stanback, M. T. (1994). Dominance within broods of the cooperatively breeding acorn woodpecker. Anim. Behav. 47, 1121–1126. Stanback, M. T., and Koenig, W. D. (1992). Cannibalism in birds In “ Cannibalism: Ecology and Evolution among Diverse Taxa” (M. A. Elgar and B. J. Crespi, eds.), pp. 277–298. Oxford Univ. Press, UK. Steyn, P. (1973). Observations on the Tawny Eagle. Ostrich 44, 1–23. Stouffer, P. C., and Power, H. W. (1991). An experimental test of the brood-reduction hypothesis in European starlings. Auk 108, 519–531. Swengel, S. R., Archibald, G. W., Ellis, D. H., and Smith, D. G. (1996). Behavior management. In “Cranes: Their Biology, Husbandry, and Conservation” (D. H. Ellis, G. F. Gee, and C. M. Mirande, eds.), Hancock House, Blaine, Washington, USA. Thomas, F., Kayser, Y., and Hafner, H. (1999). Nestling size rank in the little egret (Egretta garzetta) influences subsequent breeding success of offspring. Behav. Ecol. Sociobiol. 45, 466–470. Tinbergen, J. M., and Boerlijst, M. C. (1990). Nestling weight and survival in individual great tits Parus major. J. Anim. Ecol. 59, 1113–1128. Trivers, R. L. (1974). Parent–offspring conflict. Am. Zool. 14, 249–264. Vinuela, ˜ J. (1999). Sibling aggression, hatching asynchrony, and nestling mortality in the black kite (Milvus migrans). Behav. Ecol. Sociobiol. 45, 33–45. Watts, C. R., and Stokes, A. W. (1971). The social order of turkeys. Sci. Am. 224, 112–118. Wellington, M., Burke, A., Nicolich, J. M., and O’Malley, K. (1996). Chick rearing. In “Cranes: Their Biology, Husbandry, and Conservation” (D. H. Ellis, G. F. Gee & C. M. Mirande, eds.), Hancock House, Blaine, Washington, USA. Werschkul, D. F. (1979). Nestling mortality and the adaptive significance of early locomotion in the little blue heron. Auk 96, 116–130. Young, E. C. (1963). The breeding behaviour of the South Polar skua, Catharacta maccormicki. Ibis 105, 203–233. Zajonc, R. B., Wilson, W. R., and Rajecki, D. W. (1975). Affiliation and social discrimination produced by brief exposure in day-old chicks. Anim. Behav. 23, 131–138.

This Page Intentionally Left Blank

Index

A Aardwolves, 174, 190, 191 Acorn woodpecker, 288–289 Acoustic communication female preferences in, 156–157 geographic variation of, 154–157 of treefrogs. See Treefrog acoustic communication Acoustic criteria species differences in, 151 Acoustic pattern recognition, 101–102 Acris crepitans, 152 advertisement calls of geographic variation of, 155 Acris gryllus advertisement calls of geographic variation of, 155 Adult lifetime allocation patterns, 17–22 Advertisement calls, 148 bimodal spectrum of, 114–115, 116f conspecific, 115–116 vs. heterospecific calls, 112–113 geographic variation of, 155–156 heterospecific phonotaxis to, 137 responses to, 109–110 of Hyla cinerea, 103–106 amplitude envelope of, 123 spectral properties of, 103–105 temporal properties of, 105–106 of Hyla versicolor, 127f, 129–134 fine-temporal properties of, 131–133 gross-temporal properties of, 133–134 spectral properties of, 130–131 spectral peak frequency variation in, 118–120 vs. aggressive calls, 112

Aechmophorus occidentalis food deprivation and aggression, 263–264 Aegolius funereus, 196 Afrixalus brachycnemis advertisement calls of, 148 Age and aggressiveness, 274–276 Age-related behavioral changes in social insects, 31 Aggression in hermit crab shell fights, 91–93 in spiders, 16–17 Aggression model of shell rapping, 77–78 Aggressive calls of Hyla, 106–108 of Hyla versicolor, 134–136 vs. advertisement calls, 112 Aggressiveness food-sensitive of avian broodmates, 263–265 Agonism avian broodmate. See Avian broodmate agonism nestling, 282 Agonistic encounters hermit crab shell fights motivational changes during,73–75 Agreeableness, 233 Alcelaphus buselaphus, 174, 186 Allocation patterns alternative. See Alternative allocation patterns Allocation strategies, 9–10 Alpine marmots, 179 Alternative allocation patterns and behavioral switches, 15–16 case studies of, 17–26 303

304

INDEX

Alternative allocation patterns and behavioral switches (continued) and frequency-dependent selection, 12–15 maintenance of, 10–17, 11f modeling of, 34–37 properties of, 34–35 and spatial and temporal variation, 16–17 Alternative behavior patterns, 4 Alternative phenotypes, 4 control and evolution of, 5–6 genetic differences vs. environmental conditions, 6 Alternative strategies classes of, 6, 7t and sex-allocation patterns, 12–13 Alternative strategies and tactics, 4 alternative allocation patterns modeling of, 34–37 definition of, 6 frequency-dependent selection vs. phenotype, 6 integrative approach to, 33–34 literature of, 5 Alytes cisternasii acoustic communication female preferences in, 157 Alytes obstetricans acoustic communication female preferences in, 157 American black oystercatcher, 273, 290 American kestrels, 279 Amphibian papilla auditory neurons of, 117 Anaerobic respiration, 87 Andosterone female perception of, 242 Andromorphic females vs. gynomorphic Ishnura ramburi, 17–19, 19f Anseranas semipalmata, 273 Antelope klipspringer, 173, 175, 180 Antelopes, 177–178, 190 pronghorn, 169, 190 Antidorcas marsupialis, 169 Antilocapra americana, 169, 190 Ants discontinuous variation of, 2–4, 3f weaver bimodal distribution of, 2–3, 3f

Anurans vocal communication in, 99 Aperture investigation in hermit crab shell fights, 66 Apostatic selection, 13 Aquila pomarina, 274 Aquila rapax, 290 Ardea herodias, 262 Artiodactyls, 177 Attackers compensation of, 90–91 hermit crab shell fights motivational changes during, 73–75 startle responses of in hermit crab shell fights, 76 undamped shells vs. damped shells, 89–90 Auditory receptors tuning of temperature-dependence of, 121 Auditory thalamus auditory neurons of, 117 Averageness of male facial attractiveness, 225–228 Avian broodmate agonism, 261–294 functions, 282–290 biasing parental investment, 282–287 defense and submission, 289–291 dominance, 287–289 proximate control, 263–282 agonistic experience, 276–278 broodmate behavior, 278 brood size, 271–274 chick size and age, 268–271 food amount, 263–264 hormones, 279 maturation/age, 274–276 parental behavior, 279–282 parental feeding method, 266–268 sex, 279

B Badgers, 179 Balancing selection, 13 Barking treefrog. See Hyla gratiosa Basilar papilla auditory neurons of, 117 Batesian mimicry, 13

305

INDEX

Beetles dung size polymorphisms in, 19–20 horned discontinuous variation of, 2, 4 Behavior age-related in social insects, 31 bimodal distribution in, 4 and face shapes, 236–238 Behavioral advertising of scent marks, 180–181 Behavioral confirmation theory, 235–237 Behavioral ecology themes of, 10 Behavioral switches, 15–16 Behavior patterns alternative, 4 Belding’s ground squirrels, 184 Bimodal distribution in behavior, 4 of Oecophylla smaragdina, 2–3, 3f Birdsong, 195 Bird-voiced treefrog. See Hyla avivoca Black guillemot and aggression, 263–264, 290 Black kite chicks, 270 Black-legged kittiwake aggression, 263–264, 290 Black oystercatcher American, 273, 290 Black-tailed deer, 176–177 Blind-mole rat, 179 Blue-footed booby, 275, 278, 283 Blue-footed booby chicks, 281 Blue heron broods and parental feeding method, 266–268 Booby blue-footed, 275, 278, 283 Peruvian, 273 Boophis madagascariensis calls of, 148 Branta canadensis, 261, 273 Brazilian amphipod, 1 Breeding season of Hyla, 102–103 Brevicomin, 182, 198 Broodmate behavior of, 278 cannibalism of, 282

Brown hyena, 176 Brown pelicans and aggression, 267, 272, 290 Bubulcus ibis, 268 food deprivation and aggression, 263–264 Bufo viridis, 153 Buteo lagopus, 196 Butterflies wood, 33 Buzzard rough-legged, 196–197

C Calling frogs, 36–37 Calls advertisement. See Advertisement calls aggressive. See Aggressive calls distress of Hyla, 107–108 encounter of Hyla cinerea, 106–107 fine-temporal properties of, 121–122 gross-temporal properties of preferences on, 123–124 heterospecific vs. conspecific advertisement calls, 112–113 lagging, 109 leading, 109 of Hyla cinerea, 149 prerecorded phonotactic responses to, 109–113, 136–138 release, 107 sound pressure of, 124–125 spectral properties of temperature effects on, 120–121 synthetic. See Synthetic calls timing of, 124 within-male variation of, 123 Canada goose, 261, 273 Canadian goslings, 288 Canary, 279 Canids, 177 Canis familiaris, 185 Canis lupus, 194 Canyon treefrog. See Hyla arenicolor

306 Caricature computer of sexual dimorphism, 230–232 Casmerodius albus food deprivation and aggression, 264 Caterpillars behavioral switches of, 15 Catharacta maccormicki, 268 Cattle egrets, 263–264, 268, 271 Causal factors in hermit crab shell fights, 64, 64f–66f, 66 Cepaea nemoralis, 13 Cepphus grylle food deprivation and aggression, 263–264 Cheliped flicking, 56–57 Chorus sounds masking of, 110–112, 111f responses to, 110 Clethrionomys, 196 Clibanarius antillensis, 92 Clibanarius tibicen, 92 Clibanarius vittatus, 91–92 Clothing women’s and menstrual cycle, 241–242 Cocktail party effect and Hyla versicolor, 145–146 Color polymorphism and behavioral differences, 18 in damselflies, 17–19, 19f Communication acoustic female preferences in, 156–157 geographic variation of, 154–157 vocal in anurans, 99–101 Competition and male mammalian scent-marking, 172–174, 201–204 Competitors assessment of and scent marks, 170–171 and scent-marking, 195–196 Composite faces personality judgments of, 236 Computer caricature of sexual dimorphism, 230–232 Computer graphic manipulation of facial symmetry, 221–225

INDEX

Concephalus nigropleurum, 150 Conditional polyphenism, 4 Conditional strategy, 6, 9, 32–33 Condition-dependent strategy, 6, 9, 32–33 Connochaetes taurinus, 172 Conscientiousness, 233 Conspecific advertisement calls, 115–116 vs. heterospecific calls, 112–113 Cormorant double-crested, 283 Cost of changing in strategy allocation, 35 Coturnix coturnix, 261 Countermarking, 195–196 Coypus, 173 Crabs hermit, 53–95 European, 56 horseshoe. See Horseshoe crabs naked, 62 Crane sandhill, 273 Cricket frog advertisement calls of geographic variation of, 155 Crickets bimodal distribution in, 4 discontinuous variation of, 2, 4 Culture, 233 Cuttlefish giant and hermaphroditism, 26

D Dama dama, 194 Damselflies color polymorphism in, 17–19, 19f Darwin and variation, 1 Decisions, 10 Deer black-tailed, 176–177 fallow, 194 white-tailed, 197 wild black-tailed, 179 Defense avian broodmate agonism, 289–291 Density dependence, 16 in strategy allocation, 35

307

INDEX

Developmental reaction norms, 4 Dioecious phenotypes, 8 Diploid gray treefrog. See Hyla chrysoscelis Disassortative mating MHC, 200 Discontinuous variation of ants, 2–4, 3f Discrete variation classification systems of, 7t Distress calls of Hyla, 107–108 Dogs domesticated, 185 Domesticated dogs, 185 Dominance avian broodmate agonism, 287–289 Dominant frequency, 123 Dominant male faces, 234–235 Double-crested cormorant, 283 Dung beetles size polymorphisms in, 19–20 Dung middens, 177 Dwarf mongooses, 175 Dynamic programming, 36 Dynamics in strategy allocation, 35

E Eagle spotted, 274 tawny, 290 Eavesdropping, 171, 195–197 Ecology behavioral, 10 evolutionary, 10 Egrets, 273 cattle, 263–264, 268, 271 great and aggression, 264 Egretta garzetta, 273 Eleutherodactylus coqui advertisement calls of, 148 Emotional stability, 233 Encounter calls of Hyla cinerea, 106–107 Energetic war of attrition, 83 Environmental conditions vs. genetic differences in alternative phenotypes, 6

Equilibrium frequency (ESS), 6–7 mixed, 7–8 European hermit crabs, 56 Evolutionary ecology themes of, 10 External investigation in hermit crab shell fights, 66 Extraversion, 233

F Faces characteristics of mature vs. babyface, 234–238 composite personality judgments of, 236 shapes of and testosterone, 236–238 Facial attractiveness theories of, 220–232 Facial dominance and mature vs. babyface characteristics, 234–238 Facial growth sex differences in, 228–230 Facial symmetry, 221–225 computer graphic manipulation of, 221–225 naturally occurring, 221–222 Falco sparverius, 279 Falco tinnunculus, 196 Fallow deer, 194 Farnesenes, 198 Fast trillers, 128, 137 Fatigue, 84 Feeding-method hypothesis, 266–268 Female mating strategies mixed, 240–242 Female mice and male urine, 198 Female perception of andosterone, 242 Female preferences in acoustic communication, 156–157 for male scent during menstrual cycle, 242–244 Females andromorphic vs. gynomorphic Ishnura ramburi, 17–19, 19f Japanese

308

INDEX

Females (continued) and masculinity, 245–248 phonotactic selectivity in, 149–151 Femininity psychological meaning of, 236 Fitness in strategy allocation, 34 Floaters, 190 Florida caeruleauritus, 283 Food amount hypothesis, 263–264 Food-sensitive aggressiveness of avian broodmates, 263–265 Frequency dependence and horseshoe crabs, 28 in strategy allocation, 35 testing of, 14–15 Frequency-dependent selection, 12–15 and polymorphism, 13–14 and sex allocation, 12–13 vs. phenotype in alternative strategies and tactics, 6 Frogs calling, 36–37 cricket advertisement calls of geographic variation of, 155 geographical distribution of, 101 satellite, 36–37 true calls of, 148 vocal communication in, 99–101

G Game theory, 53 Gastrophyrne carolinesis advertisement calls of geographic variation of, 155 Gastrophyrne olivacea advertisement calls of geographic variation of, 155 Gazella thomsoni, 173 Gazelle Thomson’s, 173 Genes good, 240–242 Genetic differences vs. environmental conditions in alternative phenotypes, 6 Genetic monomorphism, 6, 7t

Genetic polymorphism, 6, 7t Genets, 184 Genetta genetta, 184 Geocrinia victoriana advertisement calls of, 148 Gerenuk scent mark map of, 177f Giant cuttlefish and hermaphroditism, 26 Gibbula cineraria shells, 56, 58–61, 69–70 Good genes, 240–242 Goose Canada, 261, 273 Goslings Canadian, 288 Gray treefrog diploid. See Hyla chrysoscelis tetraploid. See Hyla versicolor Great egrets food deprivation and aggression, 264 Great golden digger wasp nesting strategies of, 22–23, 24f Great tit vocalization of, 88 Green treefrog. See Hyla cinerea Ground squirrels Belding’s, 184 Grus canadensis, 262 Guillemot black and aggression, 263–264, 290 Gull ring-billed, 273 Gynomorphic females vs. andromorphic Ishnura ramburi, 17–19, 19f

H Haematopus ostralegus, 273 Hamlets and hermaphroditism, 25 Hartebeest, 174, 186, 191 Hawk/dove game, 53–54 Helogale parvula, 175 Hermaphrodite phenotypes, 8 Hermit crab, 53–95 European, 56 intertidal, 91

INDEX

Hermit crab shell fights, 56–58 aggression in, 91–93 decisions during, 58–61 information gathering model, 63–73 model testing in, 67–73 motivational changes during agonistic encounters, 73–75 during animal contests, 75–77 negotiation in, 91–93 resource value in, 58–61 shell rapping, 77–90 size of, 61–63 Heterospecific advertisement calls phonotaxis to, 137 responses to, 109–110 Heterospecific calls vs. conspecific advertisement calls, 112–113 Heterospecific signals phonotaxis to, 149–151 Honeybees age-related behavioral changes in, 31 bimodal distribution in, 4 Horizontal symmetry, 221 Hormones avian broodmate agonism, 279 Horned beetles discontinuous variation of, 2, 4 Horseshoe crabs alternative strategies in, 26–30 and frequency dependence, 28 males attached vs. satellite, 27–28, 29f House mice, 175 Human facial attractiveness theories of, 220–232 Human ovarian cycle, 238–239 Hyaena brunnea, 176 Hyenas, 178, 185 brown, 176 Hyenids, 177 Hyla andersonii advertisement calls of, 105–106 conspecific, 112–113 aggressive calls of, 107–108 breeding season of, 102–103 Hyla arenicolor, 126 advertisement calls of, 129–134 aggressive calls of, 134–136 distribution of, 127 prerecorded calls of phonotactic responses to, 136–138

309

Hyla avivoca, 126 advertisement calls of, 129–134, 129f, 132f, 148 aggressive calls of, 134–136 distribution of, 127–128 prerecorded calls of phonotactic responses to, 136–138 signaling interactions of, 136 Hyla chrysoscelis, 126 acoustic communication female preferences in, 156–157 advertisement calls of, 129–134, 129f, 132f geographic variation of, 156 distribution of, 127–128 signaling interactions of, 136 synthetic calls pulse rate and duration of, 142 pulse rise-time of, 140 spectral patterns of, 138–139 Hyla cinerea, 101–126 advertisement calls of, 103–106, 109–110, 118–120, 148 amplitude envelope of, 123 behavioral thresholds of, 115 breeding season of, 102–103 calls fine-temporal properties of, 121–122 sound pressure of, 124–125, 146 timing of, 124 chorus sounds of, 110–112, 111f distress calls of, 107–108 encounter calls of, 106–107 leading calls of, 149 prerecorded calls of phonotactic responses to, 109–113 signaling interactions of, 109 synthetic calls phonotactic responses to, 113–124 spectral patterns of, 138–139 Hyla ebraccata advertisement calls of, 148 Hyla examia calls of, 126 Hyla femoralis, 126 advertisement calls of, 129–134, 129f, 132f distribution of, 127–128 prerecorded calls of phonotactic responses to, 136–138

310

INDEX

Hyla gratiosa advertisement calls of, 103–106, 106f–107f, 109–110, 118–120 nonspecific, 112–113 aggressive calls of, 107 breeding season of, 102–103 calls fine-temporal properties of, 121–122 chorus sounds of, 110 distress calls of, 107–108 Hyla marmoratus advertisement calls of, 148 Hyla microcephala, 146 advertisement calls of, 148 Hyla phlebodes advertisement calls of, 148 Hyla versicolor, 126–147 acoustic communication female preferences in, 156–157 advertisement calls of, 127f, 129–134 aggressive calls of, 134–136 calls sound pressure level of, 146 and cocktail party effect, 145–146 conspecific advertisement calls of vs. heterospecific and hybrid calls, 137–138 prerecorded calls of phonotactic responses to, 136–138 signaling interactions of, 136 synthetic calls call rate and duration of, 143–145 pulse rate and duration of, 142 pulse rise-time of, 139–140 spectral patterns of, 138–139 spectral peak frequency variations of, 139 synthetic calls of phonotactic responses to, 138–145 Hyperolius marmoratus broadleyi, 154 Hypoplectrus and hermaphroditism, 25

I Information in strategy allocation, 34 Information gathering model of hermit crab shell fights, 63–73

Insect caste literature of, 5 Interruption technique in hermit crab shell fights, 69–73 in motivational state of attackers, 75 Intertidal hermit crab, 91 Intimidation in scent marks, 184 Intraspecific mate choice fine-temporal properties in, 153 gross-temporal properties in, 153–154 phonotactic preferences in, 151–154 spectral properties in, 151–152 Irreversible lifetime allocation patterns, 17–22 Irreversible patterns evolution of, 21–22 Ishnura ramburi andromorphic vs. gynomorphic females, 17–19, 19f color polymorphism in, 17–19, 19f Ixodes matopi, 197

J Japanese females and masculinity in male face shapes, 245–248 Japanese quail, 261

K Kestrels, 196–197 American, 279 Kittiwake black-legged aggression, 263–264, 290 Klipspringer antelope, 173, 175, 180

L Lagging calls, 109 Land snail, 13 Lariophagus behavioral switches of, 15–16

INDEX

Larus delawarensis, 273 Leading calls, 109 of Hyla cinerea, 149 Lemur catta, 179 Lemurs ring-tailed, 179 Life-history studies of, 10 Limulus polyphemus alternative strategies in, 26–30 Litoria ewingii advertisement calls of geographic variation of, 155 Litoria verreauxii advertisement calls of geographic variation of, 155 Littocranius walleri scent mark map of, 177f Littorina obtusata shells, 56, 58–61, 69–70 Locusts migratory behavioral switches of, 15 Luteal phase, 240

M Major histocompatibility complex (MHC) disassortative mating, 200 and odor, 172 Making-the-best-of-a-bad-job strategy, 6 Male faces dominant, 234–235 Male face shapes masculinity in cyclic preference for, 244–245 Male facial attractiveness, 219–253 averageness of, 225–228 facial symmetry, 221–225 and menstrual cycle shifts, 238–252 and personality attributions, 232–238 and secondary sexual characteristics, 228–232 theories of, 220–232 Male isopods size polymorphisms in, 19–20 Male mammalian scent-marking, 169–206 assessment of, 187–190

311

and competition, 172–174, 201–204 costs and benefits of, 190–194 detection of, 174–180 and eavesdropping, 195–197 frequency and social status, 173–174 and mate choice, 197–201, 203–204 receiver interception of, 177–180 signaler assessment in, 181–190 spatial range of, 176–177 temporal variation in, 175–176 Male scent physiological response to, 204 Male urine, 200 and female mice, 198 Marmota marmota, 179 Marmots alpine, 179 Masculinity in male face shapes cyclic preference for, 244–245 in Japanese females, 245–248 United Kingdom study of, 245, 248–249 psychological meaning of, 236 Masking of chorus sounds, 110–112, 111f Mates choice of female vs. male, 219 intraspecific, 151–154 and male mammalian scent-marking, 197–201, 203–204 phonotaxis to, 149–151 and scent-marking, 195–196 Mating disassortative MHC, 200 nonrandom, 14 Maturity and aggressiveness, 274–276 Meadow vole, 175–176 Melanerpes formicivorous, 288–289 Melanerpes gallopavo, 289 Menstrual cycle female preferences for male scent, 242–244 and sexual behavior, 239–242 and women’s clothing, 241–242 Menstrual cycle shifts and male facial attractiveness, 238–252 Mice house, 175

312

INDEX

Microtus, 196 Microtus pennsylvanicus, 175–176 Middens dung, 177 Midwife toads acoustic communication female preferences in, 157 Migratory locusts behavioral switches of, 15 Milvus migrans chicks, 270 Mimicry Batesian, 13 literature of, 5 Mullerian, 13 Mixed equilibrium frequency (ESS), 7–8 Mixed female mating strategies, 240–242 Mixed strategy, 6 Mongooses dwarf, 175 Monomorphism, 6, 7t Motivational changes in hermit crab shell fights, 63–73 during agonistic encounters, 73–75 during animal contests, 75–77 Mud-daubing wasp variation in, 1–2, 2f Mullerian mimicry, 13 Mus domesticus, 175 Mustelids, 177 Myocastor coypusurebia, 173

O Odocoileus hemionus columbianus, 176–177 Odocoileus virginianus, 197 Odor MHC mediated, 172 Oecophylla smaragdina bimodal distribution of, 2–3, 3f Onthopagus beetles, 20–21, 21f Openness-intellect, 233 Ophryotrocha puerilis, 30 Orchestia darwinii, 1 Oreotragus oreotragus, 173 Oribi, 173 Oryctolagus cuniculus, 185–186 Osprey and aggression, 263–264, 290 Ourebia ourebia, 173 Outcomes, 6 Ovarian cycle, 238–239 Overmarking, 196 Owls Tengmalm’s, 196–197 Oxygen influence on shell fights, 87 Oystercatcher, 273 American black, 273, 290

P N Naked crabs, 62 Natural selection, 2 Nature vs. nurture, 8 Negative frequency-dependent selection, 11f, 12 Negotiation in hermit crab shell fights, 91–93 Nesting strategies of great golden digger wasp, 22–23, 24f Nestling agonism, 282 Neuroticism, 233 Nonrandom mating, 14 North American treefrogs vocal communication in, 99 Nurture vs. nature, 8

Pagurus bernhardus, 56, 92 Panarge aegeria, 33 Pandion haliaetus food deprivation and aggression, 263–264 Paracerceis sculpta size polymorphisms in, 19–20 Parasites and scent-marking, 196–197 Parental behavior avian broodmate agonism, 279–282 Parental investment avian broodmate agonism, 282–287 Partial viboltinism, 13 Parus major vocalization of, 88 Pelecanus erythrorhynchos, 274 Pelecanus occidentalis food-induced aggression, 267

313

INDEX

Pelicans brown, 267, 272, 290 white, 274, 283 Perch coo, 88 Persistence time in hermit crab shell fights, 67, 68f, 69 Personality attributions and male facial attractiveness, 232–238 Personality factors and zero acquaintance, 233 Personality judgments of composite faces, 236 Peruvian booby, 273 Phalacrocorax auritus, 283 Phenotypes alternative, 4 control and evolution of, 5–6 genetic differences vs. environmental conditions, 6 vs. frequency-dependent selection in alternative strategies and tactics, 6 Phenotypic plasticity, 4 Philatus leucorhynus calls of, 148 Phonotactic preferences in intraspecific mate choice, 151–154 Phonotactic selectivity in females, 149–151 Phonotaxis selective, 115, 117 Physa and hermaphroditism, 25 Physalaemus pustulosus advertisement calls of, 148 geographic variation of, 156 call frequency of, 152 Physiognomy accuracy of, 232–233 Pine barrens treefrog. See Hyla andersonii Pinewoods treefrog. See Hyla femoralis Polyethism, 4 Polymorphism, 6, 7t and frequency-dependent selection, 13–14 Polyphenism literature of, 5 Precedence effect and signaling interactions, 149 Predation frequency dependency of, 13 Predators and scent-marking, 196–197

Preputial glands, 174 Prerecorded calls phonotactic responses to of Hyla cinerea, 109–113 of Hyla versicolor, 136–138 Prey and scent-marking, 196–197 Prey-size hypothesis, 266–268 Pronghorn antelopes, 169, 190 Propithecus verreauxi, 172 Proteles cristatus, 174 Pseudacris feriarum advertisement calls of geographic variation of, 155 Pseudacris nigrita advertisement calls of geographic variation of, 155 Pseudacris regilla advertisement calls of, 148

Q Quail Japanese, 261

R Rabbits, 185 Rana nicrobariensis calls of, 148 Rat blind-mole, 179 Reaction norm literature of, 5 Receiver psychology, 174–175 Release calls, 107 Release from masking, 112 Reproductive behavior studies of, 10 Resource-holding potential (RHP), 174, 181, 188–190 Reversible patterns and sequential allocation patterns, 26–31 and simultaneous strategy allocation, 22–25 Ring-billed gull, 273 Ring-tailed lemurs, 179 Rio Grande turkeys, 289

314

INDEX

Rissa tridactyla food deprivation and aggression, 263–264 Rodents, 177 Rough-legged buzzard, 196–197

S Sandhill crane, 273 Satellite frogs, 36–37 Scaphiopus multiplicata, 154 acoustic communication female preferences in, 157 Scent male female preferences for, 242–244 physiological response to, 204 Scent-marking economics of in territory defense, 192–194 economic theory of, 176–177 male mammalian. See Male mammalian scent-marking RHP assessment of, 188–190 Scent marks competitor assessment in, 170–171 degradation of, 169 intrinsic meaning of, 182–184 learned association in, 183–184 Scent-matching, 185–186 Seabass and hermaphroditism, 25 Secondary sexual characteristics and male facial attractiveness, 228–232 Selective phonotaxis, 115, 117 Sepia apama and hermaphroditism, 26 Sequential allocation patterns and reversible patterns, 26–31 Sequential assessment model, 55–56 Sequential control in strategy allocation, 35 Sequential hermaphrodite phenotypes, 8 Sequential hermaphrodites, 30–31 Sequential social insects, 30–31 Serinus canaria, 279 Serranus tabacarius and hermaphroditism, 25

Sex avian broodmate agonism, 279 Sex allocation and frequency-dependent selection, 12–13 literature of, 5, 9 patterns of and alternative strategies, 12–13 phenotypes of classification of, 8 Sex differences in facial growth, 228–230 Sexual behavior and menstrual cycle, 239–242 Sexual characteristics secondary and male facial attractiveness, 228–232 Sexual desire peaks in, 240–241 Sexual dimorphism, 4 computer caricature of, 230–232 and secondary sexual characteristics, 228–232 Sexual selection, 14 Shell fights of hermit crabs, 56–58, 77–90 oxygen influence on, 87 Shell rapping, 57, 57f aggression model of, 77–78 force variation within, 88–89 hermit crab shell fights, 77–90 vigor of, 82–83 Sibicidal aggression, 285–286 Sifakas, 172 Signaling interactions of Hyla cinerea, 109 of Hyla versicolor, 136 and precedence effect, 149 Signals heterospecific phonotaxis to, 149–151 Simple war of attrition, 83 Simultaneous allocation patterns and simultaneous hermaphrodites, 25–26 Simultaneous hermaphrodite phenotypes, 8 Simultaneous hermaphrodites and simultaneous allocation patterns, 25–26 Simultaneous strategy allocation and reversible patterns, 22–25 Size polymorphisms, 19–20

315

INDEX

Skua South Polar, 268, 273, 290 Snails and hermaphroditism, 25 land, 13 Social behavior studies of, 10 Social insects age-related behavioral changes in, 31 Social status and male mammalian scent-marking, 173–174 South Polar skua, 268, 273, 290 Spade-foot toads acoustic communication female preferences in, 157 calls of, 147–148 Spalax ehrenbergi, 179 Spatial variation, 16–17 Species recognition phonotaxis to, 149–151 Spermophilus beldingi, 184 Sphex icheneumoneus nesting strategies of, 22–23, 24f Spiders aggressive behavior in, 16–17 Spotted eagle, 274 Springbok, 169 Squirrels Belding’s ground, 184 Stamina, 84 Standard synthetic calls frequency-modulated pulses of, 138 State dependence in strategy allocation, 35 Stink fights, 172 Stochastic mixed strategy, 6 Strategies, 6 alternative classes of, 6, 7t and sex-allocation patterns, 12–13 Strategies and tactics alternative. See Alternative strategies and tactics Striders water discontinuous variation of, 2, 4 Submission avian broodmate agonism, 289–291 Sula variegata, 273

Switches behavioral, 15–16 Symmetrical war of attrition game, 53–54 Symmetry scent of, 243–244 Synthetic calls call rate and duration of, 143–145 midfrequency inhibition of, 117–118 phonotactic responses to of Hyla cinerea, 113–124 of Hyla versicolor, 138–145 pulse rate and duration of, 140–142 pulse rise-time of, 139–140 spectral patterns of, 138–139 spectral peak frequency variations of, 139 standard frequency-modulated pulses of, 138 testing of, 113–114

T Tactics, 6 Tawny eagle, 290 Temporal variation, 16–17 Tengmalm’s owls, 196–197 Terpenes, 182 Territoriality and scent marks, 170, 201–202 Territory defense scent-marking in economics of, 192–194 Testosterone and face shapes, 236–238 Tetraploid gray treefrog. See Hyla versicolor Thiazole, 182, 198 Thomson’s gazelle, 173 Thrushes predation of, 13 Ticks, 197 Tit great vocalization of, 88 Toads geographical distribution of, 101 midwife acoustic communication, 157 spade-foot acoustic communication, 157 calls of, 147–148

316

INDEX

Toads (continued) vocal communication in, 99 Tobaccofish and hermaphroditism, 25 Trade-offs in strategy allocation, 34 Treefrog acoustic communication, 99–158. See also Hyla cinerea; Hyla versicolor female preferences in, 157–158 phonotactic selectivity in females with, 149–154 signaling interactions with, 149 signal repertoires and aggressive signals in, 147–149 Treefrogs barking. See Hyla gratiosa bird-voiced. See Hyla avivoca canyon. See Hyla arenicolor diploid gray. See Hyla chrysoscelis green. See Hyla cinerea North American, 99 pine barrens. See Hyla andersonii pinewoods. See Hyla femoralis tetraploid gray. See Hyla versicolor True frogs calls of, 148 Trypoxylon politum variation in, 1–2, 2f T-shirt experiment, 242 Turkeys Rio Grande, 289

U United Kingdom study of masculinity in male face shapes, 245, 248–249 Urine male, 200 and female mice, 198

discontinuous of ants, 2–4, 3f discrete classification systems of, 7t patterns of, 1–9 terminology of, 6–9 Variation polymorphism, 4 Vertical symmetry, 221 Viboltinism partial, 13 Viverrids, 177 Vocal communication in anurans, 99–101 Vole, 196–197 meadow, 175–176

W Wasps behavioral switches of, 15–16 bimodal distribution in, 4 great golden digger, 22–23, 24f mud-daubing, 1–2, 2f Water striders discontinuous variation of, 2, 4 Waveform periodicity, 123 Weaver ants bimodal distribution of, 2–3, 3f Weber’s law, 124–125 Western grebe, 273 food deprivation and aggression, 263–264 White pelican, 274, 283 White-tailed deer, 197 Wild black-tailed deer, 179 Wildebeest, 172 Wolves, 194 Women’s clothing and menstrual cycle, 241–242 Wood butterflies, 33 Woodpecker acorn, 288–289

V Z Variants, 4 Variation Darwinian view of, 1

Zero acquaintance and personality factors, 233

Contents of Previous Volumes

Volume 18 Song Learning in Zebra Finches (Taeniopygia guttata): Progress and Prospects PETER J. B. SLATER, LUCY A. EALES, AND N. S. CLAYTON Behavioral Aspects of Sperm Competition in Birds T. R. BIRKHEAD Neural Mechanisms of Perception and Motor Control in a Weakly Electric Fish WALTER HEILIGENBERG Behavioral Adaptations of Aquatic Life in Insects: An Example ANN CLOAREC The Cicadian Organization of Behavior: Timekeeping in the Tsetse Fly, A Model System JOHN BRADY

The Evolution of Courtship Behavior in Newts and Salamanders T. R. HALLIDAY Ethopharmacology: A Biological Approach to the study of Drug-Induced Changes in Behavior A. K. DIXON, H. U. FISCH, AND K. H. MCALLISTER Additive and Interactive Effects of Genotype and Maternal Environment PIERRE L. ROUBERTOUX, MARIKA NOSTEN-BERTRAND, AND MICHELE CARLIER Mode Selection and Mode Switching in Foraging Animals GENE S. HELFMAN Cricket Neuroethology: Neuronal Basis of Intraspecific Acoustic Communication FRANZ HUBER Some Cognitive Capacities of an African Grey Parrot (Psittacus erithacus) IRENE MAXINE PEPPERBERG

Volume 19 Polyterritorial Polygyny in the Pied Flycatcher P. V. ALATALO AND A. LUNDBERG Kin Recognition: Problems, Prospects, and the Evolution of Discrimination Systems C. J. BARNARD Maternal Responsiveness in Humans: Emotional, Cognitive, and Biological Factors CARL M. CORTER AND ALISON S. FLEMING

Volume 20 Social Behavior and Organization in the Macropodoidea PETER J. JARMAN The t Complex: A Story of Genes, Behavior, and Population SARAH LENINGTON The Ergonomics of Worker Behavior in Social Hymenoptera PAUL SCHMID-HEMPEL

317

318

CONTENTS OF PREVIOUS VOLUMES

“Microsmatic Humans” Revisited: The Generation and Perception of Chemical Signals BENOIST SCHAAL AND RICHARD H. PORTER Lekking in Birds and Mammals: Behavioral and Evolutionary Issues R. HAVEN WILEY

Volume 21 Primate Social Relationships: Their Determinants and Consequences ERIC B. KEVERNE The Role of Parasites in Sexual Selection: Current Evidence and Future Directions MARLENE ZUK Conceptual Issues in Cognitive Ethology COLIN BEER

Parasites and the Evolution of Host Social Behavior ANDERS PAPE MØLLER, REIJA DUFVA, AND KLAS ALLANDER The Evolution of Behavioral Phenotypes: Lessons Learned from Divergent Spider Populations SUSAN E. RIECHERT Proximate and Developmental Aspects of Antipredator Behavior E. CURIO Newborn Lambs and Their Dams: The Interaction That Leads to Sucking MARGARET A. VINCE The Ontogeny of Social Displays: Form Development, Form Fixation, and Change in Context T. G. GROOTHUIS

Response in Warning Coloration in Avian Predators W. SCHULER AND T. J. ROPER

Volume 23

Analysis and Interpretation of Orb Spider Exploration and Web-Building Behavior FRITZ VOLLRATH

Sneakers, Satellites, and Helpers: Parasitic and Cooperative Behavior in Fish Reproduction MICHAEL TABORSKY

Motor Aspects of Masculine Sexual Behavior in Rats and Rabbits GABRIELA MORALI AND CARLOS BEYER On the Nature and Evolution of Imitation in the Animal Kingdom: Reappraisal of a Century of Research A. WHITEN AND R. HAM

Volume 22 Male Aggression and Sexual Coercion of Females in Nonhuman Primates and Other Mammals: Evidence and Theoretical Implications BARBARA B. SMUTS AND ROBERT W. SMUTS

Behavioral Ecology and Levels of Selection: Dissolving the Group Selection Controversy LEE ALAN DUGATKIN AND HUDSON KERN REEVE Genetic Correlations and the Control of Behavior, Exemplified by Aggressiveness in Sticklebacks THEO C. M. BAKKER Territorial Behavior: Testing the Assumptions JUDY STAMPS Communication Behavior and Sensory Mechanisms in Weakly Electric Fishes BERND KRAMER

CONTENTS OF PREVIOUS VOLUMES

Volume 24 Is the Information Center Hypothesis a Flop? HEINZ RICHNER AND PHILIPP HEEB Maternal Contributions to Mammalian Reproductive Development and the Divergence of Males and Females CELIA L. MOORE Cultural Transmission in the Black Rat: Pine Cone Feeding JOSEPH TERKEL The Behavioral Diversity and Evolution of Guppy, Poecilia reticulata, Populations in Trinidad A. E. MAGURRAN, B. H. SEGHERS, P. W. SHAW, AND G. R. CARVALHO Sociality, Group Size, and Reproductive Suppression among Carnivores SCOTT CREEL AND DAVID MACDONALD Development and Relationships: A Dynamic Model of Communication ALAN FOGEL Why Do Females Mate with Multiple Males? The Sexually Selected Sperm Hypothesis LAURENT KELLER AND HUDSON K. REEVE Cognition in Cephalopods JENNIFER A. MATHER

Volume 25 Parental Care in Invertebrates STEPHEN T. TRUMBO Cause and Effect of Parental Care in Fishes: An Epigenetic Perspective STEPHEN S. CRAWFORD AND EUGENE K. BALON Parental Care among the Amphibia MARTHA L. CRUMP

319

An Overview of Parental Care among the Reptilia CARL GANS Neural and Hormonal Control of Parental Behavior in Birds JOHN D. BUNTIN Biochemical Basis of Parental Behavior in the Rat ROBERT S. BRIDGES Somatosensation and Maternal Care in Norway Rats JUDITH M. STERN Experiential Factors in Postpartum Regulation of Maternal Care ALISON S. FLEMING, HYWEL D. MORGAN, AND CAROLYN WALSH Maternal Behavior in Rabbits: A Historical and Multidisciplinary Perspective ´ GABRIELA GONZALEZ-MARISCAL AND JAY S. ROSENBLATT Parental Behavior in Voles ZUOXIN WANG AND THOMAS R. INSEL Physiological, Sensory, and Experiential Factors of Parental Care in Sheep ´ F. LEVY, K. M. KENDRICK, E. B. KEVERNE, R. H. PORTER, AND A. ROMEYER Socialization, Hormones, and the Regulation of Maternal Behavior in Nonhuman Simian Primates CHRISTOPHER R. PRYCE Field Studies of Parental Care in Birds: New Data Focus Questions on Variation among Females PATRICIA ADAIR GOWATY Parental Investment in Pinnipeds FRITZ TRILLMICH Individual Differences in Maternal Style: Causes and Consequences of Mothers and Offspring LYNN A. FAIRBANKS

320

CONTENTS OF PREVIOUS VOLUMES

Mother-Infant Communication in Primates DARIO MAESTRIPIERI AND JOSEP CALL Infant Care in Cooperatively Breeding Species CHARLES T. SNOWDON

Volume 26 Sexual Selection in Seawood Flies THOMAS H. DAY AND ANDRE´ S. GILBURN Vocal Learning in Mammals VINCENT M. JANIK AND PETER J. B. SLATER

Volume 27 The Concept of Stress and Its Relevance for Animal Behavior DIETRICH VON HOLST Stress and Immune Response VICTOR APANIUS Behavioral Variability and Limits to Evolutionary Adaptation P. A. PARSONS Developmental Instability as a General Measure of Stress ANDERS PAPE MØLLER

Behavioral Ecology and Conservation Biology of Primates and Other Animals KAREN B. STRIER

Stress and Decision-Making under the Risk of Predation: Recent Developments from Behavioral, Reproductive, and Ecological Perspectives STEVEN L. LIMA

How to Avoid Seven Deadly Sins in the Study of Behavior MANFRED MILINSKI

Parasitic Stress and Self-Medication in Wild Animals G. A. LOZANO

Sexually Dimorphic Dispersal in Mammals: Patterns, Causes, and Consequences LAURA SMALE SCOTT NUNES, AND KAY E. HOLEKAMP

Stress and Human Behavior: Attractiveness, Women’s Sexual Development, Postpartum Depression, and Baby’s Cry RANDY THORNHILL AND F. BRYANT FURLOW

Infantile Amnesia: Using Animal Models to Understand Forgetting H. MOORE ARNOLD AND NORMAN E. SPEAR Regulation of Age Polyethism in Bees and Wasps by Juvenile Hormone SUSAN E. FAHRBACH Acoustic Signals and Speciation: The Roles of Natural and Sexual Selection in the Evolution of Cryptic Species GARETH JONES

Welfare, Stress, and the Evolution of Feelings DONALD M. BROOM Biological Conservation and Stress HERIBERT HOFER AND MARION L. EAST

Volume 28

Understanding the Complex Song of the European Starling: An Integrated Ethiological Approach MARCEL EENS

Sexual Imprinting and Evolutionary Processes in Birds: A Reassessment CAREL TEN CATE AND DAVE R. VOS

Representation of Quantities by Apes SARAH T. BOYSEN

Techniques for Analyzing Vertebrate Social Structure Using Identified

CONTENTS OF PREVIOUS VOLUMES

Individuals: Review and Recommendations HAL WHITEHEAD AND SUSAN DUFAULT Socially Induced Infertility, Incest Avoidance, and the Monopoly of Reproduction in Cooperatively Breeding African Mole-Rats, Family Bathyergidae NIGEL C. BENNETT, CHRIS G. FAULKES, AND JENNIFER U. M. JARVIS Memory in Avian Food Caching and Song Learning: A General Mechanism or Different Processes? NICOLA S. CLAYTON AND JILL A. SOHA Long-Term Memory in Human Infants: Lessons in Psychobiology CAROLYN ROVEE-COLLIER AND KRISTIN HARTSHORN Olfaction in Birds TIMOTHY J. ROPER Intraspecific Variation in Ungulate Mating Strategies: The Case of the Flexible Fallow Deer SIMON THIRGOOD, JOCHEN LANGBEIN, AND RORY J. PUTMAN

321

Volume 29 The Hungry Locust STEPHEN J. SIMPSON AND DAVID RAUBENHEIMER Sexual Selection and the Evolution of Song and Brain Structure in Acrocephalus Warblers CLIVE K. CATCHPOLE Primate Socialization Revisited: Theoretical and Practical Issues in Social Ontogeny BERTRAND L. DEPUTTE Ultraviolet Vision in Birds INNES C. CUTHILL, JULIAN C. PARTRIDGE, ANDREW T. D. BENNETT, STUART C. CHURCH, NATHAN S. HART, AND SARAH HUNT What Is the Significance of Imitation in Animals? CECILIA M. HEYES AND ELIZABETH D. RAY Vocal Interactions in Birds: The Use of Song as a Model in Communication DIETMAR TODT AND MARC NAGUIB

This Page Intentionally Left Blank