Predation and evolutionary diversification in Gambusia fishes

Understanding the mechanisms responsible for phenotypic evolution and speciation is a major goal of evolutionary biology. This thesis examines the role of predation in driving morphological and locomotor evolution, and speciation, in a group of livebearing fishes (Family Poeciliidae) in the genus Gambusia (mosquito fishes). In the first chapter, I provide a review and synthesis of the general importance of predation in evolutionary diversification. Predation is of critical and widespread evolutionary importance despite its recent neglect compared with alternative ecological factors (e.g., resource competition). Second, I investigate the ecology and evolution of fish locomotion, highlighting a predictive approach to understanding the role of natural selection in phenotypic evolution. Thirdly, I evaluate the importance of natural selection (via predation) and premating sexual selection in the evolution of male genital size in two Gambusia species ( Gambusia affinis, G. hubbsi ). Then in the fourth chapter, I examine population differentiation in morphology and locomotion among populations of G. affinis inhabiting low- and high-predation regimes, illustrating that divergent natural selection has apparently led to the rapid evolution of differences in body stiffness, size of the anterior body/head region, size of the caudal region, three-dimensional streamlining, and steady and unsteady locomotor abilities. The fifth chapter demonstrates that adaptation to divergent predator regimes is driving ecological speciation as a by-product in a post-Pleistocene radiation of Bahamas mosquitofish ( Gambusia hubbsi ) inhabiting blue holes. The study also reveals that not only is predation driving reproductive isolation among recently isolated populations, but appears to have been historically important in past speciation events of Gambusia fishes. Overall, this research employs a pluralistic, interdisciplinary approach--synthesizing prior work to formulate novel frameworks and predictions, and conducting comparative and experimental analyses integrating morphological, locomotor, behavioral, and molecular data across multiple temporal scales--and should advance our understanding of the generation of phenotypic diversity and the origin of species.