Genotype-by-environment interactions and sexual selection

Genotype-by-environment interactions (G x Es) describe genetic variation for phenotypic plasticity, such that the relative performance of genotypes varies across environments. These interactions have been studied in the context of natural selection for decades, but research interest in the evolutionary consequences of G x Es in sexual traits is more recent. Theory suggests that G x Es in sexual traits could be of fundamental importance to the operation of sexual selection across heterogeneous environments, but empirical research lags behind the theory. In this thesis, I review the current literature on the role of G x Es in sexual selection and identify areas for further research. Using cuticular hydrocarbons (CHCs) in the fruit fly Drosophila simulans as a model system for sexual selection, I examine G x Es in trait expression and quantify the effect of these G x Es in terms of sexual signal reliability and the coevolution of male and female sexual traits. To do so, I use a combination of quantitative genetics and laboratory environmental manipulations. First, I demonstrate that male CHC profile is subject to sexual selection through female mate choice and find some variation in patterns of mate choice across diets and temperatures (Chapter 3). Next, I identify G x Es in male and female CHC expression across diets and temperatures, although G x Es in male CHC profile across temperatures are weak (Chapter 4). I find that G x Es in male CHC expression can cause sexual signal unreliability, as predicted by theory, since male CHCs do not reliably signal heritable aspects of male attractiveness across diets and temperatures (Chapter 5). I also find G x Es in some aspects of female mate choice across temperatures (Chapter 6). In spite of the evidence for signal unreliability and variation in female mate choice across environments, I show that the overall outcome of mate choice is unaffected by G x Es, such that the same male genotypes are attractive across diets and temperatures (Chapters 5 and 6). From my results, it seems likely that females assess male attractiveness based on multiple male sexual signals, so that whilst male CHCs influence mate choice, CHC profile does not necessarily correlate well with overall male attractiveness. I discuss the implications of these results for the evolution of sexual traits and the genetic covariance between male and female sexual traits across environments. The research in this thesis highlights the importance of multivariate studies of sexual selection across environments for a more complete understanding of the evolution of sexual traits.EThOS - Electronic Theses Online ServiceESFGBUnited Kingdo