The evolution of female mating preferences: differentiation from species with promiscuous males can promote speciation

Females of many species are frequently courted by promiscuous males of their own and other closely related species. Such mating interactions may impose strong selection on female mating preferences to favor trait values in conspecific males that allow females to discriminate them from their heterospecific rivals. We explore the consequences of such selection in models of the evolution of female mating preferences when females must interact with heterospecific males from which they are completely postreproductively isolated. Specifically, we allow the values of both the most preferred male trait and the tolerance of females for males that deviate from this most preferred trait to evolve. Also, we consider situations in which females base their mating decisions on multiple male traits and must interact with males of multiple species. Females will rapidly differentiate in preference when they sometimes mistake heterospecific males for suitable mates, and the differentiation of female preference will select for conspecific male traits to differentiate as well. In most circumstances, this differentiation continues indefinitely, but slows substantially once females are differentiated enough to make mistakes rare. Populations of females with broader preference functions (i.e., broader tolerance for males with trait values that deviate from females' most preferred values) will evolve further to differentiate if the shape of the function cannot evolve. Also, the magnitude of separation that evolves is larger and achieved faster when conspecific males have lower relative abundance. The direction of differentiation is also very sensitive to initial conditions if females base their mate choices on multiple male traits. We discuss how these selection pressures on female mate choice may lead to speciation by generating differentiation among populations of a progenitor species that experiences different assemblages of heterospecifics. Opportunities for differentiation increase as the number of traits involved in mate choice increase and as the number of species involved increases. We suggest that this mode of speciation may have been particularly prevalent in response to the cycles of climatic change throughout the Quaternary that forced the assembly and disassembly of entire communities on a continentwide basis.     

Testing the correlated response hypothesis for the evolution and maintenance of male mating preferences in Drosophila serrata

Mate preferences are abundant throughout the animal kingdom with female preferences receiving the most empirical and theoretical attention. Although recent work has acknowledged the existence of male mate preferences, whether they have evolved and are maintained as a direct result of selection on males or indirectly as a genetically correlated response to selection for female choice remains an open question. Using the native Australian species Drosophila serrata in which mutual mate choice occurs for a suite of contact pheromones (cuticular hydrocarbons or CHCs), we empirically test key predictions of the correlated response hypothesis. First, within the context of a quantitative genetic breeding design, we estimated the degree to which the trait values favoured by male and female choice are similar both phenotypically and genetically. The direction of sexual selection on male and female CHCs differed statistically, and the trait combinations that maximized male and female mating success were not genetically correlated, suggesting that male and female preferences target genetically different signals. Second, despite detecting significant genetic variance in female preferences, we found no evidence for genetic variance in male preferences and, as a consequence, no detectable correlation between male and female mating preferences. Combined, these findings are inconsistent with the idea that male mate choice in D. serrata is simply a correlated response to female choice. Our results suggest that male and female preferences are genetically distinct traits in this species and may therefore have arisen via different evolutionary processes.     

Female discrimination thresholds frequently exceed local male display variation: implications for mate choice dynamics and sexual selection

Among the factors that can influence female mate choice decisions is the degree to which females differentiate among similar displays: as differences decrease, females are expected to eventually stop discriminating. This discrimination threshold, in conjunction with the magnitude of male trait variation females regularly encounter while making mate choice decisions, may have important consequences for sexual selection. If local display variation is above the discrimination threshold, female preferences should translate into higher mating success for the more attractive male. But if display variation is frequently below the threshold, the resulting increased pattern of random mating may obscure the existence of female mate choice. I investigated the interplay between female discrimination and male display variation in green treefrogs (Hyla cinerea) and found that call trait differences between nearest neighbour males were frequently smaller than what females are expected to discriminate. This finding has two important consequences for our understanding of sexual selection in the wild: first, low display variation should weaken the strength of selection on male display traits, but the direction of selection should mirror the one predicted from females choice trials. Second, caution is needed when interpreting data on realized mating success in the wild: a pattern of random mating with respect to male display traits does not always mean that female preferences are weak or that conditions are too challenging for females to express their preferences. Rather, insufficient display variation can generate the same pattern.     

Contemporary sexual selection does not explain variation in male display traits among populations

Sexual selection is widely hypothesized to facilitate the evolution of reproductive isolation through divergence in sexual traits and sexual trait preferences among populations. However, direct evidence of divergent sexual selection causing intraspecific trait divergence remains limited. Using the wolf spider Schizocosa crassipes, we characterized patterns of female mate choice within and among geographic locations and related those patterns to geographic variation in male display traits to test whether divergent sexual selection caused by mate choice explains intraspecific trait variation. We found evidence of phenotypic selection on male behavior arising from female mate choice, but no evidence that selection varied among locations. Only those suites of morphological and behavioral traits that did not influence mate choice varied geographically. These results are inconsistent with ongoing divergent sexual selection underlying the observed intraspecific divergence in male display traits. These findings align with theory on the potentially restrictive conditions under which divergent sexual selection may persist, and suggest that long‐term studies capable of detecting periodic or transient divergent sexual selection will be critical to rigorously assess the relative importance of divergent sexual selection in intraspecific trait divergence.     

Independent Mating Preferences for Male Body Size and Coloration in Female Trinidadian Guppies

Although females in numerous species generally prefer males with larger, brighter and more elaborate sexual traits, there is nonetheless considerable intra‐ and interpopulation variation in mating preferences amongst females that requires explanation. Such variation exists in the Trinidadian guppy, Poecilia reticulata, an important model organism for the study of sexual selection and mate choice. While female guppies tend to prefer more ornamented males as mates, particularly those with greater amounts of orange coloration, there remains variation both in male traits and female mating preferences within and between populations. Male body size is another trait that is sexually selected through female mate choice in some species, but has not been examined as extensively as body coloration in the guppy despite known intra‐ and interpopulation variation in this trait among adult males and its importance for survivorship in this species. In this study, we used a dichotomous‐choice test to quantify the mating preferences of female guppies, originating from a low‐predation population in Trinidad, for two male traits, body length and area of the body covered with orange and black pigmentation, independently of each other. We expected strong female mating preferences for both male body length and coloration in this population, given relaxation from predation and presumably relatively low cost of choice. Females indeed exhibited a strong preference for larger males as expected, but surprisingly a weaker (but nonetheless significant) preference for orange and black coloration. Interestingly, larger females demonstrated stronger preferences for larger males than did smaller females, which could potentially lead to size‐assortative mating in nature.     

Conservation of multivariate female preference functions and preference mechanisms in three species of trilling field crickets

Divergence in mate recognition systems among closely related species is an important contributor to assortative mating and reproductive isolation. Here, we examine divergence in male song traits and female preference functions in three cricket species with songs consisting of long trills. The shape of female preference functions appears to be mostly conserved across species and follows the predictions from a recent model for song recognition. Multivariate preference profiles, combining the pulse and trill parameters, demonstrate selectivity for conspecific pulse rates and high trill duty cycles. The rules for integration across pulse and trill timescales were identical for all three species. Generally, we find greater divergence in male song traits than in associated female preferences. For pulse rate, we find a strong match between divergent male traits and female peak preferences. Preference functions for trill parameters and carrier frequency are similar between species and show less congruence between signal and preference. Differences among traits in the degree of trait–preference (mis)match may reflect the strength of preferences and the potential for linkage disequilibrium, selective constraints and alternative selective pressures, but appear unrelated to selection for mate recognition per se.     

Genetic Architecture of Sexual Selection: QTL Mapping of Male Song and Female Receiver Traits in an Acoustic Moth

Models of indirect (genetic) benefits sexual selection predict linkage disequilibria between genes that influence male traits and female preferences, owing to non-random mate choice or physical linkage. Such linkage disequilibria can accelerate the evolution of traits and preferences to exaggerated levels. Both theory and recent empirical findings on species recognition suggest that such linkage disequilibria may result from physical linkage or pleiotropy, but very little work has addressed this possibility within the context of sexual selection. We studied the genetic architecture of sexually selected traits by analyzing signals and preferences in an acoustic moth, Achroia grisella, in which males attract females with a train of ultrasound pulses and females prefer loud songs and a fast pulse rhythm. Both male signal characters and female preferences are repeatable and heritable traits. Moreover, female choice is based largely on male song, while males do not appear to provide direct benefits at mating. Thus, some genetic correlation between song and preference traits is expected. We employed a standard crossing design between inbred lines and used AFLP markers to build a linkage map for this species and locate quantitative trait loci (QTL) that influence male song and female preference. Our analyses mostly revealed QTLs of moderate strength that influence various male signal and female receiver traits, but one QTL was found that exerts a major influence on the pulse-pair rate of male song, a critical trait in female attraction. However, we found no evidence of specific co-localization of QTLs influencing male signal and female receiver traits on the same linkage groups. This finding suggests that the sexual selection process would proceed at a modest rate in A. grisella and that evolution toward exaggerated character states may be tempered. We suggest that this equilibrium state may be more the norm than the exception among animal species.     

Sympatric speciation by sexual selection alone is unlikely

According to Darwin, sympatric speciation is driven by disruptive, frequency-dependent natural selection caused by competition for diverse resources. Recently, several authors have argued that disruptive sexual selection can also cause sympatric speciation. Here, we use hypergeometric phenotypic and individual-based genotypic models to explore sympatric speciation by sexual selection under a broad range of conditions. If variabilities of preference and display traits are each caused by more than one or two polymorphic loci, sympatric speciation requires rather strong sexual selection when females exert preferences for extreme male phenotypes. Under this kind of mate choice, speciation can occur only if initial distributions of preference and display are close to symmetric. Otherwise, the population rapidly loses variability. Thus, unless allele replacements at very few loci are enough for reproductive isolation, female preferences for extreme male displays are unlikely to drive sympatric speciation. By contrast, similarity-based female preferences that do not cause sexual selection are less destabilizing to the maintenance of genetic variability and may result in sympatric speciation across a broader range of initial conditions. Certain groups of African cichlids have served as the exclusive motivation for the hypothesis of sympatric speciation by sexual selection. Mate choice in these fishes appears to be driven by female preferences for extreme male phenotypes rather than similarity-based preferences, and the evolution of premating reproductive isolation commonly involves at least several genes. Therefore, differences in female preferences and male display in cichlids and other species of sympatric origin are more likely to have evolved as isolating mechanisms under disruptive natural selection.     

Can non‐directional male mating preferences facilitate honest female ornamentation?

Recent studies have demonstrated male mate choice for female ornaments in species without sex-role reversal. Despite these empirical findings, little is known about the adaptive dynamics of female signalling, in particular the evolution of male mating preferences. The evolution of traits that signal mate quality is more complex in females than in males because females usually provide the bulk of resources for the developing offspring. Here, we investigate the evolution of male mating preferences using a mathematical model which: (i) specifically accounts for the fact that females must trade-off resources invested in ornaments with reproduction; and (ii) allows male mating preferences to evolve a non-directional shape. The optimal adaptive strategy for males is to develop stabilizing mating preferences for female display traits to avoid females that either invests too many or too few resources in ornamentation. However, the evolutionary stability of this prediction is dependent upon the level of error made by females when allocating resources to either signal or fecundity.     

Mate‐sampling costs and sexy sons

Costly female mating preferences for purely Fisherian male traits (i.e. sexual ornaments that are genetically uncorrelated with inherent viability) are not expected to persist at equilibrium. The indirect benefit of producing ‘sexy sons’ (Fisher process) disappears: in some models, the male trait becomes fixed; in others, a range of male trait values persist, but a larger trait confers no net fitness advantage because it lowers survival. Insufficient indirect selection to counter the direct cost of producing fewer offspring means that preferences are lost. The only well‐cited exception assumes biased mutation on male traits. The above findings generally assume constant direct selection against female preferences (i.e. fixed costs). We show that if mate‐sampling costs are instead derived based on an explicit account of how females acquire mates, an initially costly mating preference can coevolve with a male trait so that both persist in the presence or absence of biased mutation. Our models predict that empirically detecting selection at equilibrium will be difficult, even if selection was responsible for the location of the current equilibrium. In general, it appears useful to integrate mate sampling theory with models of genetic consequences of mating preferences: being explicit about the process by which individuals select mates can alter equilibria.     

THE EVOLUTION OF MATE PREFERENCES FOR MULTIPLE SEXUAL ORNAMENTS

Males of many species use multiple sexual ornaments in their courtship display. We investigate the evolution of female sexual preferences for more than a single male trait by the handicap process. The handicap process assumes that ornaments are indicators of male quality, and a female benefits from mate choice by her offspring inheriting “good genes” that increase survival chances. A new handicap model is developed that allows equilibria to be given in terms of selection pressures, independent of genetic parameters. Multiple sexual preferences evolve if the overall cost of choice is not greatly increased by a female using additional male traits in her assessment of potential mates. However, only a single preference is evolutionarily stable if assessment of additional male traits greatly increases the overall cost of choice (more than expected by combining the cost of each preference independently). Any single preference can evolve, the outcome being determined by initial conditions. The evolution of one preference effectively blocks the evolution of others, even for traits that are better indicators of male quality. Comparison is made with sexual selection caused by Fisher's runaway process in which male traits are purely attractive characters. This shows that sexual preferences for multiple Fisher traits are likely to evolve alongside preference for a single handicap trait that indicates male quality. This is a general difference in the evolutionary outcome of these two causes of sexual selection.     

Correlated changes in male plumage coloration and female mate choice in cardueline finches

The coevolution of extravagant male traits and female mate preferences is a central tenet of sexual selection theory. In lineages in which males have developed more elaborate sexual characters, females favour the most extreme expression of the trait. In some taxa, however, ornamental displays have evolved from more to less exaggerated states. Under these circumstances, it is unclear whether females show preferences for an ancestral male condition or for the current, less elaborate display. Here, we tested female mate preferences relative to male ornamental coloration in two species of cardueline finch (the American goldfinch, Carduelis tristis, and pine siskin, Carduelis pinus) that have evolved less elaborate carotenoid-based colour displays from more elaborately coloured ancestral states. We presented females of each species with a choice of males having either large patches of red colour (the elaborate, ancestral condition) or with species-typical patches of yellow colour (the less elaborate, derived state). Female goldfinches and siskins showed consistent preferences for the natural colour displays of males, and not for the more elaborate, ancestral colour pattern. Previous research on another cardueline finch taxon (a subspecies of the house finch, Carpodacus mexicanus griscomi), however, showed that females prefer more elaborate, ancestral coloration to the current form of reduced colour expression. The lack of congruence between male trait expression and female trait preference in the lineage with the most recently derived reduction in trait expression suggests that there may be evolutionary lags in the correspondence between male traits and female preferences. A shift in the expression of male coloration appears to be the first step towards the evolution of reduced colour displays in these finches.     

Different mate preferences by parasitized and unparasitized females potentially reduces sexual selection

Parasite‐mediated models of sexual selection predict that females should avoid parasitized mates, thereby generating selection on male traits revealing health. The strength of this selection, however, may depend on the prevalence of parasitism among females if their infection status alters their mate preferences. We evaluated the effects of a socially transmitted parasite on male traits and female behaviour in spadefoot toads. Parasitized males were larger and in better condition than unparasitized males. Moreover, better condition males produced longer calls. Unparasitized females preferred longer calls indicative of good‐condition males that are more likely parasitized. By contrast, parasitized females as a group possessed no preference for call duration. Presumably because of reduced selection by these parasitized females, male mating success was not associated with any measured traits. Thus, when females are parasitized, sexual selection on condition‐dependent traits is potentially reduced.     

Testing the fisherian mechanism: examining the genetic correlation between male song and female response in waxmoths

Models of indirect (genetic) benefits sexual selection predict linkage disequilibria between genes that influence male traits and female preferences, owing to either non-random mate choice or physical linkage. Such linkage disequilibria, a genetic correlation, can accelerate the evolution of male traits and female preferences to exaggerated levels. But relatively few empirical studies have measured the genetic correlation between male traits and female responses in natural populations, and the findings of those few are mixed: often, genetic correlations are not found. We tested the above prediction in an acoustic pyralid moth, Achroia grisella, in which males attract females with a rhythmic train of sound pulses, and females respond only to song that exceeds a minimum pulse rhythm. Both male song rhythm and female threshold response are repeatable and heritable characters. Because female choice in A. grisella is based largely on male song, and males do not appear to provide direct benefits at mating, genetic correlation between male song rhythm and female response is expected. We studied 2 A. grisella populations, bred them according to a full-sib/half-sib design, split the progeny among 4 different environmental conditions, and measured the male song/female response genetic correlation in each of the 8 resulting groups. While song rhythm and response threshold were generally heritable, we found no evidence of significant genetic correlation between these traits. We suggest that the complexity of the various male song characters, of female response to male song, and of correlations between male song characters and between aspects of female response have mitigated the evolution of strong genetic correlation between song and response. Thus, exaggerated levels of trait development may be tempered.     

Female mate choice and the potential for ornament evolution in túngara frogs Physalaemus pustulosus

The potential for ornament evolution in response to sexual selection rests on the interaction between the permissive-ness or selectivity of female preferences and the constraints on male development of signaling related traits. We investigate the former by determining how latent female preferences either exaggerate the magnitude of current traits (I.e. Elaborations) or favor novel traits (I.e. Innovations). In tungara frogs, females prefer complex mating calls (whine-chucks) to simple calls (whine only). The whine is critical for mate recognition while the chuck further enhances the attractiveness of the call. Here we use a combina-tion of synthetic and natural stimuli to examine latent female preferences. Our results show that a diversity of stimuli, including conspecific and heterospecific calls as well as predator-produced and human-made sounds, increase the attractiveness of a call when added to a whine. These stimuli do not make simple calls more attractive than a whine-chuck, however. In rare cases we found stimuli that added to the whine decrease the attractiveness of the call. Overall, females show strong preferences for both elaborations and innovations of the chuck. We argue that the emancipation of these acoustic adornments from mate recognition allows such female permissiveness, and that male constraints on signal evolution are probably more important in explaining why males evolved their specific adornment. Experimentally probing latent female preferences for stimuli out of the species' range is a useful means to gain insights about the potential of female choice to influence signal evolution and thus the astounding diversity in male sexually-selected traits.     

The evolution of male mate choice in insects: a synthesis of ideas and evidence

Mate choice by males has been recognized at least since Darwin's time, but its phylogenetic distribution and effect on the evolution of female phenotypes remain poorly known. Moreover, the relative importance of factors thought to underlie the evolution of male mate choice (especially parental investment and mate quality variance) is still unresolved. Here I synthesize the empirical evidence and theory pertaining to the evolution of male mate choice and sex role reversal in insects, and examine the potential for male mating preferences to generate sexual selection on female phenotypes. Although male mate choice has received relatively little empirical study, the available evidence suggests that it is widespread among insects (and other animals). In addition to 'precopulatory' male mate choice, some insects exhibit 'cryptic' male mate choice, varying the amount of resources allocated to mating on the basis of female mate quality. As predicted by theory, the most commonly observed male mating preferences are those that tend to maximize a male's expected fertilization success from each mating. Such preferences tend to favour female phenotypes associated with high fecundity or reduced sperm competition intensity. Among insect species there is wide variation in mechanisms used by males to assess female mate quality, some of which (e.g. probing, antennating or repeatedly mounting the female) may be difficult to distinguish from copulatory courtship. According to theory, selection for male choosiness is an increasing function of mate quality variance and those reproductive costs that reduce, with each mating, the number of subsequent matings that a male can perform ('mating investment') Conversely, choosiness is constrained by the costs of mate search and assessment, in combination with the accuracy of assessment of potential mates and of the distribution of mate qualities. Stronger selection for male choosiness may also be expected in systems where female fitness increases with each copulation than in systems where female fitness peaks at a small number of matings. This theoretical framework is consistent with most of the empirical evidence. Furthermore, a variety of observed male mating preferences have the potential to exert sexual selection on female phenotypes. However, because male insects typically choose females based on phenotypic indicators of fecundity such as body size, and these are usually amenable to direct visual or tactile assessment, male mate choice often tends to reinforce stronger vectors of fecundity or viability selection, and seldom results in the evolution of female display traits. Research on orthopterans has shown that complete sex role reversal (i.e. males choosy, females competitive) can occur when male parental investment limits female fecundity and reduces the potential rate of reproduction of males sufficiently to produce a female-biased operational sex ratio. By contrast, many systems exhibiting partial sex role reversal (i.e. males choosy and competitive) are not associated with elevated levels of male parental investment, reduced male reproductive rates, or reduced male bias in the operational sex ratio. Instead, large female mate quality variance resulting from factors such as strong last-male sperm precedence or large variance in female fecundity may select for both male choosiness and competitiveness in such systems. Thus, partial and complete sex role reversal do not merely represent different points along a continuum of increasing male parental investment, but may evolve via different evolutionary pathways.     

Evolution of Novel Signal Traits in the Absence of Female Preferences in Neoconocephalus Katydids (Orthoptera,Tettigoniidae)

Background Significance

Communication signals that function to bring together the sexes are important for maintaining reproductive isolation in many taxa. Changes in male calls are often attributed to sexual selection, in which female preferences initiate signal divergence. Natural selection can also influence signal traits if calls attract predators or parasitoids, or if calling is energetically costly. Neutral evolution is often neglected in the context of acoustic communication.

Methodology/Principal Findings

We describe a signal trait that appears to have evolved in the absence of either sexual or natural selection. In the katydid genus Neoconocephalus, calls with a derived pattern in which pulses are grouped into pairs have evolved five times independently. We have previously shown that in three of these species, females require the double pulse pattern for call recognition, and hence the recognition system of the females is also in a derived state. Here we describe the remaining two species and find that although males produce the derived call pattern, females use the ancestral recognition mechanism in which no pulse pattern is required. Females respond equally well to the single and double pulse calls, indicating that the derived trait is selectively neutral in the context of mate recognition.

Conclusions/Significance

These results suggest that 1) neutral changes in signal traits could be important in the diversification of communication systems, and 2) males rather than females may be responsible for initiating signal divergence.     

Female Attraction to Higher Pitched Male Enticement Calls in Barn Swallows

Current sexual selection studies demonstrate that female mate preference almost always favors exaggerated male traits, such as colorful plumage and elongated tails. However, such studies focused mainly on secondary sexual characteristics, i.e., traits that are fully developed at maturation, while ignoring others, which precludes generalization of the findings. We studied the expression of the male enticement call, a nestling‐like courtship trait, in relation to female attraction and to the expression of secondary sexual ornaments in the Japanese barn swallow Hirundo rustica gutturalis. First, we showed that the peak amplitude frequency of male enticement calls was, on average, lower than that of nestling food‐begging calls, indicating that higher pitched calls are more similar to nestling food‐begging calls than are lower pitched calls. Second, we conducted an experimental playback of male enticement calls during the courtship period with the female's mate removed, and found that females were more attracted to heightened‐pitch calls than to control calls. Lastly, the peak amplitude frequency of male enticement calls was negatively related to the ornamentation, as measured by throat coloration. Taken together, these findings suggest the following: (1) females are attracted to a more nestling‐like trait, and (2) similarity to nestling food‐begging calls increases with decreasing secondary sexual ornaments. Nestling‐like male traits that are not secondary sexual characteristics should be considered when studying the general pattern of female attraction to male traits.     

Divergence of a speciation trait through artificial selection: Insights into constraints,by-product effects and sexual isolation

Speciation and sexual isolation often occur when divergent female mating preferences target male secondary sexual traits. Despite the importance of such male signals, little is known about their evolvability and genetic linkage to other traits during speciation. To answer these questions, we imposed divergent artificial selection for 10 non-overlapping generations on the Inter-Pulse-Interval (IPI) of male courtship songs; which has been previously shown to be a major species recognition trait for females in the Drosophila athabasca species complex. Focusing on one of the species, Drosophila mahican (previously known as EA race), we examined IPI's: (1) rate of divergence, (2) response to selection in different directions, (3) genetic architecture of divergence and (4) by-product effects on other traits that have diverged in the species complex. We found rapid and consistent response for higher IPI but less response to lower IPI; implying asymmetrical constraints. Genetic divergence in IPI differed from natural species in X versus autosome contribution and in dominance, suggesting that evolution may take different paths. Finally, selection on IPI did not alter other components of male songs, or other ecological traits, and did not cause divergence in female preferences, as evidenced by lack of sexual isolation. This suggests that divergence of male courtship song IPI is unconstrained by genetic linkage with other traits in this system. This lack of linkage between male signals and other traits implies that female preferences or ecological selection can co-opt and mould specific male signals for species recognition free of genetic constraints from other traits.     

Male choice generates stabilizing sexual selection on a female fecundity correlate

We know very little about male mating preferences and how they influence the evolution of female traits. Theory predicts that males may benefit from choosing females on the basis of traits that indicate their fecundity. Here, we explore sexual selection generated by male choice on two components of female body size (wing length and body mass) in Drosophila serrata. Using a dietary manipulation to alter female size and 828 male mate choice trials, we analysed linear and nonlinear sexual selection gradients on female mass and wing length. In contrast to theoretical expectations and prevailing empirical data, males exerted stabilizing rather than directional sexual selection on female body mass, a correlate of fecundity. Sexual selection was detected only among females with access to standard resource levels as an adult, with no evidence for sexual selection among resource-depleted females. Thus the mating success of females with the same body mass differed depending upon their access to resources as an adult. This suggests that males in this species may rely on signal traits to assess body mass rather than assessing it directly. Stabilizing rather than directional sexual selection on body mass together with recent evidence for stabilizing sexual selection on candidate signal traits in this species suggests that females may trade-off resources allocated to reproduction and sexual signalling.