Most of those involved in research into the origin of our species now agree that anatomically modern humans evolved only in Africa sometime within the last 150,000 years. Beyond this, there is less understanding about the evolutionary processes which have produced the genetic differences observed both within and between populations, or about those which have given rise to the cultural and linguistic diversity which characterizes our species. Therefore, this thesis models the role of geography in the evolutionary process in order to provide greater understanding about the mechanisms involved in the generation and maintenance of modern human diversity. Three main questions are addressed. First, what patterns can be identified in the Pleistocene evolutionary history of modern humans in Africa? Second, what is the geographical and environmental context for this evolution? Third, to what extent can the patterns observable in Pleistocene African evolution be said to be the products of geography? To answer these questions a review of the evidence for modern human evolution is conducted. This covers genetic, palaeontological, archaeological and palaeoenvironmental research. From these data a model is developed, suggesting that geography and environmental change have promoted cycles of expansion and contraction in the size and distribution of the human population, and that these mechanisms have led to differentiation and diversification. In order to test this prediction a spatially explicit population model is constructed. This couples a model of population growth and spread to a geographically realistic and novel method of environmental change. The motor for this change is the oscillation of global temperatures over the past. A variety of simulations are conducted using the model and the results are consistent with the hypothesis that environmentally stimulated fluctuations in population size and distribution are driving forces in the evolution of modern human diversity. Comparisons with linguistic, archaeological and ecological data on past populations, as well as with models of genetic population history, suggest that the patterns observed in the output of the model are very similar to those observed in other types of evidence. Thus, the model provides valid, quantifiable and new information about the evolution of the modern human population and its demographic history within Africa.