Genetic admixture in the Late Pleistocene

Contemporary human mtDNA is strikingly uniform. The replacement model of modern human origins holds that this uniformity is best explained by a model in which all modern Homo sapiens are descended from a small population within the last 100,000 years. On the other hand, if the population was large over the past 1 million years, as implied by the multiregional model, more diversity is expected. The multiregional model of modern human origins proposes regional evolution of modern Homo sapiens from their archaic ancestors and attributes overall similarity between modern Homo sapiens to gene flow between regional populations. In addition to these two extreme hypotheses, a continuum of intermediate hypotheses is conceivable. Suppose that an original small population of modern Homo sapiens expanded throughout the world mixing with surrounding peoples as they went. Denote by q the fraction of DNA from these surrounding peoples in the post-expansion population. The extreme hypotheses would -be reflected by q = 0 in the case of the replacement model and q = 1 in the case of the multiregional model. Intermediate values, such as q = 0.05 are hypotheses which have never been discussed. This thesis investigates the role of mtDNA in the debate on modern human origins and proposes to test hypotheses about intermediate values of divergent mtDNA. The results show that small values of q cannot be ruled out by current uniformity. They also show that the maximum level of q compatible with this evidence is low if effective female population size was large over the past 1 million years.