ARTICLE The phylogeography of the rodent genus Malacomys suggests multiple Afrotropical Pleistocene lowland forest

ABSTRACTAim This study aims to reconstruct the evolutionary history of the Africanrodent genus Malacomys and to identify factors driving diversification withinthis genus.Location African tropical lowland forest.Methods Analyses were based on sampling representatives from most of theknown geographical range of the genus. We assessed genetic structure and his-torical biogeography using a combination of mitochondrial and nuclear mark-ers. Morphological differences between lineages were analysed using ageometric morphometric approach.Results Three species of Malacomys are recognized within the genus. Two areendemic to West Africa, and one is endemic to Central Africa. Our analysesreveal a strong phylogeographical structure with 13 lineages, most of them allo-patric or parapatric. A complex biogeographical history, including dispersal–vicariance events, explains the current genetic structure of Malacomys. Discretedivergence events within the genus are dated to the mid-Pliocene (3.7 Ma,95% range: 2.4–5.2 Ma) and the Pleistocene (less than 1.9 Ma, with mostevents less than 1 Ma). Morphological variation is partly congruent withgenetic structure and may indicate local adaptations.Main conclusions Climatic oscillations, which led to periodic fragmentationof the forest habitat, seem to be the major driver of diversification within thisgenus. Our results support the existence of multiple small, rather than a fewlarge, forest refugia during glacial maxima. Rivers have played a significant rolein shaping boundaries of several regional haplogroups, either by promotingdiversification or by preventing secondary contact between previously isolatedlineages.KeywordsClimate changes, dispersal, forest refugia, Malacomys, phylogeography,Plio-Pleistocene, riverine barrier hypothesis, tropical Africa, vicariance.INTRODUCTIONThe tropical forest of Africa is one of the most biologicallydiverse regions in the world for mammals (Ceballos & Ehrlich,2006). The diversification of tropical African vertebrates has awide range of possible evolutionary causes, include the follow-ing three. Tectonic activity during the Mio-Pliocene may havestrongly influenced diversification events, particularly in theRift region and along the Cameroon Volcanic line (Moodley B Taylor et al., 2009). Climate fluctuations asso-ciated with Pleistocene glacial cycles led to periodic contrac-tions and expansions of vegetation and faunal assemblages(deMenocal, 2004). Such modifications of habitat may havecaused faunal diversification, either through allopatric diver-gence into distinct forest refugia (Pleistocene refuge theory;Haffer, 1982) or, for taxa with wide ecological preferences,