Abstract.-The turnover-pulse hypothesis (TPH) makes explicit predictions concerning the potential responses of species to climate change, which is considered to be a major cause of faunal turnover (extinction, speciation, and migration). Previous studies have tested the TPH primarily by examining temporal correlations between turnover pulses and climatic events. It is rarely possible to dissect such correlations and observe turnover as it is occurring or to predict how different lineages will respond to climate change. Thus, whether climate change drives faunal turnover in the manner predicted by the TPH remains unclear. In this study, we test the underlying mechanics of the TPH using well-dated Quaternary ungulate records from southern Africa's Cape Floristic Region (CFR). Changes in sea level, vegetation, and topographic barriers across glacial-interglacial transitions in southern Africa caused shifts in habitat size and configuration, allowing us to generate specific predictions concerning the responses of ungulates characterized by different feeding habits and habitat preferences. Examples from the CFR show how climatically forced vegetation change and allopatry can drive turnover resulting from extinction and migration. Evidence for speciation is lacking, suggesting either that climate change does not cause speciation in these circumstances or that the evolutionary outcome of turnover is contingent on the nature and rate of climate change. Migrations and extinctions are observed in the CFR fossil record over geologically short time intervals, on the order of Milankovitchscale climate oscillations. We propose that such climate oscillations could drive a steady and moderate level of faunal turnover over 10 4 -year time scales, which would not be resolved in paleontological records spanning 10 5 years and longer. A turnover pulse, which is a marked increase in turnover relative to previous and subsequent time periods, requires additional, temporally constrained climatic forcing or other processes that could accelerate evolutionary change, perhaps mediated through biotic interactions.