Species distributions are constrained by environmental factors such as climate and habitat. However, biogeography cannot be explained entirely in terms of now-existing environmental situations. All currently observed distribution patterns resultfrom an interplay of historic as well as ecological factors. A substantial amount of evidence suggests that climatic changes during the Pleistocene that resulted in severe glaciations in the Northern Hemisphere also caused drastic changes in the distribution of animals and vegetation in Africa. During that period, the climate in much of Africa became progressively arid and periods of reduced rainfall led to the temporary fragmentation of previously continuous rainforests into forest islands separated by savannas. These events also had severe consequences for aquatic habitats and even led to the complete drying out of some large lakes like Lake Victoria or the split of Lake Tanganyika into separate waterbodies. Because of these drastic environmental changes, faunas became isolated in refuges where species communities diverged and eventually speciated. Following this scenario, with the return of less arid climatic conditions, forest islands expanded, savannas receded and lake levels increased, thereby establishing contact zones between previously isolated taxa. Consequently, a substantial proportion of the modern tropical African faunas diversified and speciated on ecological islands that were subjected to cycles of fragmentation and subsequent expansion. The majority of hypotheses that concern possible outcomes of secondary contact among such isolates are based on zoogeographical studies of Neotropical and African faunas, often represented by birds and butterflies. However, recent studies on other faunal components challenge the so-called refuge hypothesis for the lowland forest faunas of the Amazonian basin and question the refuge hypothesis under its present form for the Central African forest faunas. For example, one study on speciation patterns in the Amazon basin suggests that centers of endemism - invoked to support the concept of unchanging refugia - may equally well support the hypothesis that the regions containing high numbers of endemics have been areas of maximal disturbance rather than of maximum stability. Secondly, a recent study on primates from the Central African Zaire Basin suggests that the two recognized refuges explained by the Pleistocene refuge theory - i.e. the Mount Cameroon and Central African Rift regions - may be the result of a incomplete and/or biased species sampling. The latter study suggests the existence of a Quaternary Major Fluvial Refuge and questions the palaeoenvironmental history that is accepted for that part of tropical Africa. Our objective is to infer the biogeographical history of tropical Africa from patterns of relatedness between taxa and populations of terrestrial (forest, savanna and montane) and aquatic (lacustrine) vertebrates. This approach assesses the extent to which populations of a series of independent lineages ocurring in the same region show concordant geographic patterns and times of separation. Such concordance would imply that they were subjected to the same climatic and geological changes that separated them by the same geographic barriers and over the same time spans. Our proposal aims to improve our understanding how some representative vertebrates and their distributions change over time and how these changes are influenced by climatologic and geographical events in tropical Africa. More particularly, we intend to test the assumption that climatic cyclic changes are an important driving mechanism for speciation in terrestrial and aquatic organisms. Finally, our results will also contribute to a better systematic knowledge of the studied taxa. The molecular characterization of historically isolated sets of populations can also be important for the management of faunas that are threatened by deforestation, fragmentation, habitat degradation and by ongoing and future climate changes.