Vertebrate Mating Systems

The following section are included:

• CONTENTS

This book is a synthesis of recent advances in understanding of the ecological causes and consequences of variation in vertebrate mating systems. Each chapter was first presented as an invited lecture during a workshop of the International School of Ethology, held at the Ettore Majorana Centre for Scientific Culture in Erice, Sicily, in November 1998…

This review addresses current issues in the study of mating systems and mate choice and shows how sexual selection results in trade-offs for both sexes. Arbitrary female preferences serve as a null hypothesis for comparison with the alternative hypotheses of direct and indirect net benefits for female preferences. Indirect benefits require heritable male traits and thus raise the issue of decreasing additive genetic variance which could eliminate selection for female choice based on indirect benefits. Whenever heritable traits and preferences affect mating, the consequent assortative mating will produce genetic correlation and the possibility for accelerating evolution. Accelerating evolution can also result from frequency-dependent selection on any signals and responses. An expanded view of mate choice includes indirect as well as direct choice. With this view, the coevolution of male and female traits is seen to have much greater scope than currently appreciated. Either form of mate choice can lead to the evolution of exaggerated traits and reliable signaling systems as a result of males optimizing fitness (survival X reproduction) and females optimizing decision-making in noisy conditions. Discordances between preferences and traits in a population can result either from phylogenetic constraints or from current selection on preferences and traits. This review thus emphasizes that the evolution of exaggerated traits and preferences depends on inevitable trade-offs for the individuals involved.

A central question in the study of animal mating systems has been the possibility of female choice for males to obtain indirect benefits which will be expressed only in the next generation. Two main theories have been provoked as explanations for indirect mate selection, and there has been surprisingly strong feelings in favour of either of them over the recent years. The research history conforms very well with the paradigm theory of Thomas Kuhn. In early 1980's the only accepted theory was based on the 'Fisherian runaway process', where male ornaments are viewed as merely arbitrary traits that have evolved through non-adaptive female preferences. The 'good genes' alternative that ornaments are viability indicators allowing females to enhance offspring fitness through inheritance of viability enhancing genes was theoretically accepted only in late 1980's. This paradigm shift had obvious effects on the published empirical papers, suggesting that publication biases in favour of preferred hypotheses may be common. Indeed, this will make it difficult to apply meta-analyses on the published studies, in particular in the case of good gene effects that are likely to be slight and, thus, requiring particularly large sample sizes.

Darwin's theory of sexual selection predicts that morphological differences between the sexes (sexual dimorphism and dichromatism) should to be correlated with the intensity of sexual selection among species. The intensity of sexual selection co-varies with the mating system. Highly polygynous or polyandrous mating systems are characterized by intense sexual selection while in monogamous systems the intensity is relaxed. For these reasons we should expect more dramatic differences between the sexes in species with polygynous or polyandrous mating systems. The prediction is fulfilled in some but not all cross-specific analyses performed so far. Here I discuss why some highly polygynous species do not show dimorphism as predicted by theory. I have previously reported that in lekking birds there are quite a few examples of monomorphic species despite the high intensity of sexual selection imposed by the lek mating system and that when taking phylogeny into account there is no evidence that lekking promotes dimorphism. The reason is mainly because evolution is not an entirely deterministic process and that in a comparison some taxa share a common evolutionary background that others lack. One consequence of this is an evolutionary lag in sensory modality. As an example, nocturnal species display during night and may from signalling theory be expected to use vocal signals. In contrast diurnal species display in day light but often during windy conditions and hence the most optimal signals may be visual ones. Within taxonomic groups both sensory modality and mating system tend to be retained. In cases when the sexes have a similar morphology the physical conditions when the males are most active impose limits on what kinds of signals are the most optimal ones. I propose that those species that, despite intense sexual selection, have morphologically similar sexes are constrained by the physical environment in which the lek is performed.

Variation in mating systems can have demographic consequences such as differences in reproductive success due to dominance status or, in avian communal breeding systems, delayed reproduction by nonbreeding "helpers". Such sources of variation, that depend on social status or behavioral differences rather than age, can make traditional age-classified demographic analyses difficult or of limited validity. I present several case histories to show how stage-classified matrix population models have the flexibility to deal with such variation. The results provide powerful insights into evolutionary selective pressures, management options for managed or endangered species, or pointers to the sorts of demographic field data that are most essential in accurately assessing dynamics. By examining the effects of nonbreeding helpers in communally breeding Florida scrub-jays, the dynamics of lek-mating sage grouse based on the brood status of hens, and the demographic differences between males and females in lek-mating long-tailed manakins, I illustrate the wide range of problems and model formulations available. The techniques should be valuable to researchers who wish to combine the study of mating systems with analyses of evolutionary demography or population dynamics.

The relationship between ecological factors and mating systems has been the focus of interest for an impressive number of studies, especially during the last two decades. Here I review some main environmental effects on mating systems of birds and mammals. Interspecific variability as a result of adaptation to the environment has been shown in a number of taxa so that mating systems are commonly better explained by environmental rather than phylogenetic effects, producing adaptive convergent outcomes of the conflict of interests between the sexes. If mating systems are strongly affected by local environment, they should differ between populations of the same species (spatial variation) and also between different years in the same population (temporal variation). Intraspecific variation illustrates the response of a particular genetic lineage to different environments, and in this sense, constitutes a useful tool for understanding environmental effects on the evolution of mating systems. Many environmental factors such as resource dispersion, predation, sociality, etc. may affect female and in turn male dispersion, although male tactics may also affect female dispersion and tactics. Changes in the mating system may be experimentally induced by changing particular variables of the environment, providing a powerful tool to test hypotheses. At the same time variability in mating systems also raises important implications for management: mating systems may also change as a consequence of human alterations of the environment. The degree of behavioural plasticity of mating patterns may vary among species. Different factors may be involved in the degree of observed variability. On the one hand, are those related to research bias, either in space or time scales. On the other hand, are those related to evolutionary history (e.g. environmental stability or perception abilities), to the associated benefits (e.g. short- versus long term benefits) and to the costs and limits of changing behaviour (e.g. assessment and uncertainty). Finally, I believe that the way individuals respond to environmental variation by changing their mating patterns, may be crucial to our understanding of sexual selection processes and their consequences.

This review addresses how the environment can cause a change in the mating pattern of a population through its direct effect on the potential reproductive rates of males and females and through environmentally mediated changes in the costs and benefits to individuals in choosing among mating partners. In addition, the way in which the environment can constrain the ability of individuals to mate with preferred partners by influencing the perceptual capabilities is also discussed. In ectotherms in particular, temperature and food availability determine to a large extent the PRR of individuals by influencing both the time needed to brood eggs and the time required to produce a clutch. This in turn effects the OSR of the population and ultimately, which sex competes. The sex with the higher PRR is the sex towards which the OSR is biased and is also the sex that has more to gain from competing for multiple mates. The relationship between the OSR and mate choice is less clear because environmental factors can set the costs and benefits to males and females in being choosy independently of the OSR.

Mating patterns, sexual selection and parental care are central topics in behavioural ecology, but they are often analysed in isolation from each other. We propose a new conceptual framework to investigate these topics in relation to each other. We argue that it is beneficial to study both mating behaviour and parental care of all types of individual in a population, because the behaviours of different individuals are interrelated in many ways. In particular, we propose a framework in which the parental care adopted is the best response to the mating behaviour and the mating behaviour adopted is the best response to the parental behaviour. The backbone of the proposed framework is the feedback relationship between mating strategies (e.g. accepting or rejecting a mate), mating opportunities (related to the number and quality of animals searching for a mate) and parental care strategies (e.g. caring for the offspring or deserting them). For instance, mating opportunities should influence both the mating and parental strategies. The mating and parental strategies, in turn, have an effect on mating opportunities. We emphasise the conceptual significance of these feedback loops as well as referring to empirical studies which have demonstrated some of these feedbacks. The strength of these feedbacks probably vary between species and may be negligible in some systems. Unlike most previous approaches to mating behaviour and parental care, we do not assume that mating systems, parental investments by males and females, operational sex ratios, reproductive rates, or the intensity of sexual selection are fixed in a population. Rather, these characteristics emerge when one specifies the behavioural options of males and females, and their consequences. Mating and parental decisions can have consequences beyond the immediate breeding attempt and the proposed framework allows us to investigate such decisions from a life-history perspective. Mating and caring decisions involve various interactions among members of a population (e.g. conflicts between prospective mates, and between male and female parents), thus studying mating and caring behaviour benefits from the use of game theory. Since the state of animals (e.g. whether they are mated or not, their age, energy reserves or the number of their offspring) and the time in the breeding season commonly influence the payoffs from different behavioural options, we advocate the use of state-dependent dynamic game theory as a suitable approach for the analysis of such decisions. Finally, we call for a new generation of theoretical models and empirical studies to understand the diverse mating and parental behaviour of animals which have fascinated evolutionary biologists from Darwin onwards.

Recent interest has focused increasingly on the role of behavioural research in conservation. Within this field, the study of mating systems can make a highly important contribution. Mating systems both affect, and are affected by, interactions between conspecifics. Of particular importance are Allee effects, which arise as a result of the benefits of conspecific presence. Recognition of the consequences of Allee effects for mating systems, conservation, and behaviour, has also increased recently. The inter-relationship between these areas is highly complex, and both mating systems and Allee effects have consequences for each other, as well as for conservation. In this way, both have direct and indirect consequences for conservation. We discuss these in the light of current, and potential, contributions of the study of mating systems to conservation.

Conservation science is often interested in predicting the future state of populations living in variable environments. Behavioral ecology is often interested in understanding individual behavior in terms of adaptive strategies for dealing with variable ecological, social, or phenotypic states. To the extent that individual behavior is strategic, predictable in different environments, and has population-level consequences, the study of behavior can advance the predictive goals of conservation. Mating system studies, via data on variation in individual reproductive success, have most frequently been applied to conservation in the estimation of effective population size and prediction of genetic trajectories of single populations. However, mating strategy may also influence dispersal, gene flow, and genetic trajectories at spatial scales involving many populations. Mating competition within populations and dispersal will often have opposing effects on effective population size. Future studies of mating systems might aid conservation (i) in evaluating the genetic effects of mating strategy at different spatial scales, and (ii) by providing, in place of point estimates, descriptions of the way in which variation in lifetime reproductive success and population dispersal pattern change in reaction to environmental variation. Such information may allow conservation science to provide better guidance regarding the appropriate spatial scale of habitat protection and restoration and effective strategies for reserve design and re-design.

Genetic tools have recently altered our views of mating patterns in many taxa. The results have expanded the range of hypotheses being tested in several major subject areas; including the types of selective forces acting on female reproductive behavior, the relationship between paternity and paternal care, the operation of sexual selection, and the ecology of mating. Despite the progress genetics have made in these areas, many fundamental questions remain unanswered. Moreover, genetic tools have revealed considerable unexplained diversity in mating patterns within many taxa. Current conceptual approaches provide only partial explanations for such diversity, and tend to emphasize one sex over the other. I review some new advances in selection theory and quantitative genetic models which, when integrated with previous paradigms, offer some new insights on mating. Although the use of molecular genetic tools to study mating will continue, these other genetic approaches could also become important for progress toward understanding the evolution of male and female interactions over reproduction.

Traditional population genetics theory assumed that at some level in the hierarchy of genetic diversity there existed 'demes' in which males and females mate at random. Polygynous mating systems or individual variance in reproductive success were considered complications which altered estimates of effective population size and the calculation of F-statistics. Recent breeding group models bridge the gap between behavioural ecology and population genetics, and illustrate how genetic approaches can aid the study of mating systems, and vice versa.

At the practical level, molecular genetic tools such as DNA fingerprinting and microsatellites have revolutionized our understanding of the fitness consequences of different mating behaviours. We describe how the molecular methods we have used in studies of avian mating systems illustrate two important practical challenges. The first example uses the tree swallow to describe the difficulties of working with a large, open population of potential fathers. The second example summarizes work on two Australian miner species, which breed cooperatively. The practical challenges came in dealing with genetic relatedness of helpers at nests, both to explain helping, and to assign parentage in the few cases it was necessary. The fact that helpers were kin, coupled with some inbreeding, revealed the limits of even the most modern molecular approaches to mating systems. Traditional behavioural observations of known individuals will often still be needed to understand the complex pedigrees in such systems.

On the theoretical side, it is increasingly apparent that evolutionary explanations for the diversity of mating systems will have to be supported by rigorous genetic models. Most passerines were assumed to be monogamous, yet we now know there is little correspondence between their behavioural and genetic mating systems. It appears that EPP is often under female control, and much effort is being devoted to understanding how females distinguish between behavioural and genetic mates. Genetic explanations for EPP include the genetic diversity, genetic compatibility and good genes models. Our work with chickadees allows us to predict which males are likely to be chosen by females for EPP, but only a small number of other studies have found any correlates of male EPP success. Often genetic explanations of EPP are not well articulated, and there remains a real need to determine if behavioural ecologists are proposing genetical theories which are likely to apply in real populations in a reasonable amount of evolutionary time.