"A central goal of evolutionary biology is to understand a population's evolutionary trajectory from fundamental population-level characteristics. The mathematical framework of population genetics provides the tools to make these predictions. And while population genetics provides a well-studied framework to understand how adaptation and neutral evolution quantitatively alter population fitness, less attention has been paid to using population genetics to predict qualitative evolutionary outcomes. For instance, do different populations evolve alternative genetic mechanisms to encode similar phenotypic traits, and if so, which processes lead to these differences? This dissertation investigates the role of population size in altering the qualitative outcome of evolution. It is difficult to experimentally investigate qualitative evolutionary outcomes, especially in small populations, due to the time required for novel evolutionary features to appear. To get around this constraint, I use digital experimental evolution. While digital evolution experiments lack aspects of biological realism, in some regards they are the only methodology that can approach the complexity of biological systems while maintaining the ease of analysis present in mathematical models. Digital evolution experiments can never prove that certain evolutionary trajectories occur in biological populations, but they can suggest hypotheses to test in more realistic model systems. First, I explore the role of population size in determining the evolution of both genomic and phenotypic complexity. Previous hypotheses have argued that small population size may lead to increases in complexity and I test aspects of those hypotheses here. Second, I introduce the novel concept of 'drift robustness' and argue that drift robustness is a strong factor in the evolution of small populations. Finally, I end with a project on the role of genome size in enhancing the extinction risk of small populations. I conclude with a broader discussion of the consequences of this research, some limitations of the results, and some ideas for future research."--Page ii.
Read
