Phylogenetic systematics and tempo of evolution of the Viverrinae (Mammalia, Carnivora, Viverridae) within feliformians: implications for faunal exchanges between Asia and Africa.

We reconstructed the phylogeny of the subfamily Viverrinae (Mammalia, Carnivora, Viverridae) using a approximately 3kb data set in order to reassess timing and patterns of faunal exchanges between Asia and Africa. Maximum parsimony, maximum likelihood, and Bayesian analyses of separated and combined matrices (cytochrome b, transthyretin intron I and IRBP exon 1 [IRBP1]) recovered all the well-supported relationships within feliformian lineages. In addition, IRBP1 supported paraphyly of genus Herpestes and contributed to the resolution of equivocal hypotheses within Viverridae, including (1) the monophyly of Viverrinae, and (2) Viverricula sister-group of the other terrestrial civets (Civettictis and Viverra). The combined analysis yielded a robust phylogeny, recovering monophyly of Prionodontidae and yielding high posterior probabilities for nodes (1) (Prionodontidae, Felidae) and (2) ((Felidae, Prionodontidae), ((Hyaenidae, (Herpestidae, Eupleridae)), Viverridae)). Using a fossil cross-validation method, we estimated the emergence of Viverridae at 34.29Myr, with a separation between the three traditional subfamilies Hemigalinae, Paradoxurinae, and Viverrinae during the Late Oligocene-Early Miocene. The terrestrial civets and the splits between (1) Civettictis and Viverra and (2) Poiana and Genetta were estimated to appear during the Middle Miocene. Parsimony- and maximum likelihood-based methods yielded unambiguous ancestral area reconstructions, including the Asian origin of the family Viverridae, the subfamily Viverrinae, the terrestrial civets and the clade (Civettictis, Viverra). On the grounds of genetic distances, morphological divergence, and divergence time estimates, we propose the erection of the subfamily Genettinae (including Genetta and Poiana). Our analyses suggested two independent migration events from Asia to Africa, during the Middle Miocene (Civettictis) and between the Late Oligocene and Middle Miocene (Genettinae). These results are in agreement with the hypothesis of Miocene routes from Asia to Africa-via the Arabian microplate-that would have involved several independent events of migrations. Couched in the context of the viverrid fossil record, our study calls for a revision of the paleontological data in order to fully appreciate the complexity of Afro-Asian faunal exchanges.