No description available.

No description available.

Thesis (Ph.D.) - University of St Andrews, March 2009.

Swordtail fishes of the genus Xiphophorus have been studied as a model of sexual selection for many years. Many single-trait manipulation studies have been performed, determining female preferences for individual male traits. I characterized how five traits (standard body length, body depth, dorsal fin width, sword length and vertical bar number) correlate to one another within natural variation of populations of X. birchmanni, X. malinche and three hybrid populations and created synthetic 3- dimensional animations exhibiting these traits within ranges of natural variation. I then performed choice tests on females of the above populations using a computer system that automatically played these stimulus videos and simultaneously tracked a female’s position within a test tank to determine female preference for different male phenotypes. Only X. birchmanni females showed significant preferences. Their preferences were in line with past research of univariate trait manipulation experiments. They showed significant preference for larger bodies and dorsal fins and smaller or no swords. They also showed a non-significant preference for vertical bar numbers. My results also confirmed univariate studies in which X. malinche females showed reduced preference for conspecific males and being rather indifferent to the presence of swords. Hybrid females were also shown to have reduced preferences for any specific trait, suggesting that they express recombinant preferences, which can also be explained by reduced color vision at low levels of light.

There is much evidence that mate-choice decisions made by humans are affected by social/contextual information. Women seem to rate men portrayed in a relationship as more desirable than the same men when portrayed as single. Laboratory studies have found evidence suggesting that human mate choice, as in other species, is dependent on the mate choice decisions made by same-sex rivals. Even though non-independent mate choice is an established and well-studied area of mate choice, very few field studies have been performed. This project aims to test whether women’s evaluation of potential mates desirability is dependent/non-independent of same-sex rivals giving the potential mates sexual interest. This is the first field study performed in a modern human’s natural habitat aiming to test for non- independent mate choice in humans. No desirability enhancement effect was found. The possibilities that earlier studies have found an effect that is only present in laboratory environments or have measured effects other than non-independent mate choice are discussed. I find differences in experimental design to be the most likely reason why the present study failed to detect the effect found in previous studies. This field study, the first of its sort, has generated important knowledge for future experimenters, where the most important conclusion is that major limitations in humans ability to register and remember there surrounding should be taken in consideration when designing any field study investigating human mate choice.

Female choice is an important selective force shaping the evolution of communication and speciation in animals. However, predation risk can impose severe costs on longer searches and choosiness, thereby limiting the expression of female preferences for specific male traits. The work detailed in this dissertation explores how mate choice and sexual selection can be influenced by predation risk in túngara frogs. I begin by examining the effects of multiple simulated cues of predation risk on female search behavior and mate choice, taking a departure from the standard presence/absence paradigm used in similar studies to explore responses to quantitative variation in perceived predation risk. I demonstrate that light, longer travel times, and acoustic cues of predators are all sufficient to sway females away from otherwise more attractive conspecific males. Next, I explore the role of predation risk in altering female permissiveness, or the range of signals females will respond to. Using an artificial series of calls intermediate between heterospecific and conspecific, I demonstrate that predation risk dramatically increases the range of signals females will respond to, including a small number of females choosing pure heterospecific calls. Next I attempt to bridge a logical gap with our understanding of search costs, testing questions about how female search paths change with increasing distance. I demonstrate that females use more direct paths and move faster under higher light conditions, potentially reducing sampling but also reducing encounter rates with predators. Lastly, I examine factors that influence how individual females vary in their response to perceived risk, particularly hormonal state and experience. I demonstrate that naïve, captive-bred females respond to acoustic cues produced by natural predators in a manner similar to wild females and that, while hormonal state is obviously important in determining female receptivity, it has little effect directly on how females respond to predators. Together, these studies demonstrate that predation risk not only changes how females respond to conspecific males, but also increases female permissiveness and constrains search behavior. Predation risk can strongly influence and potentially even negate the expression of female preferences, having profound consequences for communication and the evolution of reproductive isolation between populations. / text

No description available.

The evolution of mate choice remains one of the most controversial topics within evolutionary biology. In particular, the coevolutionary dynamics between ornaments and mating preferences has been extensively studied, but few generalizations have emerged. From a theoretical standpoint, the nature of the genetic covariance built up by the process of mate choice has received considerable attention, though the models still make biologically unrealistic assumptions. Empirically, the difficulty of estimating parameters in the models has hindered our ability to understand what processes are occurring in nature. Thus, it is the goal of this dissertation to contribute to the field both theoretically and empirically. I begin with a review of the evolution of mate choice and demonstrate how the lack of cross-talk between theoretical and empirical pursuits into studying mate choice has constrained our ability to extract basic principles. The review is followed by a new model of intersexual selection that relaxes some of the critical assumptions inherent in sexual selection theory. There are two empirical studies whose goal is to measure mating preference functions and genetic correlations in a way that can be related back to theory. Finally, I conclude by setting the stage for future endeavors into exploring the evolution of mate choice. The results presented herein demonstrate four things: (i) a lack of communication between theoretical and empirical studies of mate choice; (ii) genetic drift plays a much larger role in preference evolution than previously demonstrated; (iii) genetic correlations other than those explicitly modeled are likely to be important in preference evolution; and (iv) variation in mating preferences can eliminate intersexual selection altogether. From these four findings it can be concluded that a tighter link between theory and empiricism is needed, with a particular emphasis on the importance of measuring individual-level preference functions. Models will benefit from integrating the specific phenotypes measured by empiricists. Experimentation will be more useful to theory if particular attention is paid to the exact phenotypes that are measured. Overall, this dissertation is a stepping stone for a more cohesive and accurate understanding of mate choice evolution.

The ‘good genes’ model of sexual selection predicts that sexual and natural selection should act concordantly. However sexual selection can favour alleles in males that are costly when expressed in daughters, placing the two in opposition. The relationship between natural and sexual selection depends on the nature of genetic variation for fitness. Laboratory adaptation may deplete sexually concordant fitness variation, overestimating sexually antagonistic variation and obscuring good genes. I investigated sire-offspring fitness correlations in Drosophila melanogaster populations expected to differ in their levels of sexually concordant fitness variation. In maladapted populations, successful sires produced fitter daughters than unsuccessful sires; this pattern was reversed in adapted populations. Several generations later, successful sires in both population types produced lower fitness daughters than unsuccessful sires, consistent with predictions. However, subsequent generations revealed no effect of sire status on daughter fitness, highlighting the difficulty in testing predictions on the evolutionary dynamics of fitness heritability.

The ‘good genes’ model of sexual selection predicts that sexual and natural selection should act concordantly. However sexual selection can favour alleles in males that are costly when expressed in daughters, placing the two in opposition. The relationship between natural and sexual selection depends on the nature of genetic variation for fitness. Laboratory adaptation may deplete sexually concordant fitness variation, overestimating sexually antagonistic variation and obscuring good genes. I investigated sire-offspring fitness correlations in Drosophila melanogaster populations expected to differ in their levels of sexually concordant fitness variation. In maladapted populations, successful sires produced fitter daughters than unsuccessful sires; this pattern was reversed in adapted populations. Several generations later, successful sires in both population types produced lower fitness daughters than unsuccessful sires, consistent with predictions. However, subsequent generations revealed no effect of sire status on daughter fitness, highlighting the difficulty in testing predictions on the evolutionary dynamics of fitness heritability.