The geography of species diversification

Abstract

Speciation is largely a geographical process, but the role of geography in explaining patterns in species diversification remains poorly understood. Here I examine the effects of geographic ranges on the evolution of species diversity using a combination of phylogenetic and geographic approaches based on both simulated and real data. In the first part of the thesis, I develop a neutral model combining the evolution of species’ distributions and the geography of speciation to explore the effects of this relationship on the dynamics of species radiations. I show that neutral interactions between geographic ranges and speciation can lead to dramatic differences in diversity amongst clades and through time, and can mirror the phylogenetic patterns observed in real data derived from bird genera. The evolution of ranges following speciation is likely to be a key factor in how patterns of species diversification unfold through time. In the second part of the thesis I examine the dynamics of ranges over both ecological and evolutionary timescales. First, using neutral and deterministic models of range expansion, I show that the structure of species’ distributions in birds are highly deterministic and cannot be predicted by random ecological processes. I then use phylogenies of extant birds and mammals, to examine whether there is any evidence for systematic changes in range size through time. In both groups, I find that a model of random range evolution cannot be rejected. The results show that inferences regarding the dynamics of range evolution from extant phylogenies are likely to be confounded by the effects of speciation and extinction. In the final part of the thesis, I test whether the geographic ranges of species determine the potential for speciation, focussing on how range shape constrains gene flow between populations. Using estimates of population neutral genetic differentiation for birds, mammals and amphibians, I find that differences in the degree of genetic cohesion of a species cannot be predicted by the shape of its geographic range. Hence, if range shape is to influence speciation it must do so through alternative mechanisms. Taken together, these findings suggest that geography can have a profound effect on the patterns of diversification. Ignoring the role of geography may therefore result in misleading conclusions regarding the processes underlying variation in species diversity. While my findings show that geographic ranges are deterministic, they also imply that neutral processes may play a much larger role in the history of diversification than is generally appreciated. This geographical perspective therefore goes some way to reconciling the roles of neutrality and ecology in the evolution of species diversity.