UC Merced

Aim: To test the congruence of phylogeographic patterns and processes between a woodland agamid lizard (Diporiphora australis) and well-studied Australian wet tropics fauna. Specifically, to determine whether the biogeographic history of D. australis is more consistent with a history of vicariance, which is common in wet tropics fauna, or with a history of dispersal with expansion, which would be expected for species occupying woodland habitats that expanded with the increasingly drier conditions in eastern Australia during the Miocene-Pleistocene. Location: North-eastern Australia. Methods: Field-collected and museum tissue samples from across the entire distribution of D. australis were used to compile a comprehensive phylo-geographic dataset based on c. 1400 bp of mitochondrial DNA (mtDNA), incorporating the ND2 protein-coding gene. We used phylogenetic methods to assess biogeographic patterns within D. australis and relaxed molecular clock analyses were conducted to estimate divergence times. Hierarchical Shimodaira-Hasegawa tests were used to test alternative topologies representing vicariant, dispersal and mixed dispersal/vicariant biogeographic hypotheses. Phylogenetic analyses were combined with phylogeographic analyses to gain an insight into the evolutionary processes operating within D. australis. Results: Phylogenetic analyses identified six major mtDNA clades within D. australis, with phylogeographic patterns closely matching those seen in many wet tropics taxa. Congruent phylogeographic breaks were observed across the Black Mountain Corridor, Burdekin and St Lawrence Gaps. Divergence amongst clades was found to decrease in a north-south direction, with a trend of increasing population expansion in the south. Main conclusions: While phylogeographic patterns in D australis reflect those seen in many rain forest fauna of the wet tropics, the evolutionary processes underlying these patterns appear to be very different. Our results support a history of sequential colonization of D. australis from north to south across major biogeographic barriers from the late Miocene-Pleistocene. These patterns are most likely in response to expanding woodland habitats. Our results strengthen the data available for this iconic region in Australia by exploring the understudied woodland habitats. In addition, our study shows the importance of thorough investigations of not only the biogeographic patterns displayed by species but also the evolutionary processes underlying such patterns. © 2010 Blackwell Publishing Ltd.