The role of Pleistocene forest refugia in the evolution and biogeography of tropical biotas.

Refuge theory postulates that repeated oscillation of dry and moist climatic periods during the Pleistocene caused an alternating fragmentation and coalescence of areas of lowland tropical rainforest vegetation, leading to genetic differentiation and speciation in isolated populations of rainforest organisms and hence accounting for much of the high diversity now apparent in these habitats. The theory, which became widely accepted during the 1970s, is still based largely on inferences from modern species distribution patterns and on selected palaeoclimatic and geomorphic data. However, much of this evidence has multiple interpretations; indeed, modern population genetic theory argues against many of the tenets of refuge theory. Moreover, there is no palynological evidence against which refuge theory might be tested. Although large-scale dynamic processes have undoubtedly played many parts in promoting diversity in tropical rainforest, refuge theory must be seriously questioned on many counts.