Naturwissenschaften (2005) 92:121–127 DOI 10.1007/s00114-004-0602-0 SHORT COMMUNICATION Ulrich Kuch · J. Scott Keogh · John Weigel · Laurie A. Smith · Dietrich Mebs Phylogeography of Australias king brown snake (Pseudechis australis) reveals Pliocene divergence and Pleistocene dispersal of a top predator Received: 1 October 2004 / Accepted: 6 December 2004 / Published online: 2 February 2005  Springer-Verlag 2005 Abstract King brown snakes or mulga snakes (Pseude- chis australis) are the largest and among the most dan- gerous and wide-ranging venomous snakes in Australia and New Guinea. They occur in diverse habitats, are important predators, and exhibit considerable morpho- logical variation. We infer the relationships and historical biogeography of P. australis based on phylogenetic analysis of 1,249 base pairs from the mitochondrial cy- tochrome b, NADH dehydrogenase subunit 4 and three adjacent tRNA genes using Bayesian, maximum-likeli- hood, and maximum-parsimony methods. All methods reveal deep phylogenetic structure with four strongly supported clades comprising snakes from New Guinea (I), localities all over Australia (II), the Kimberleys of Western Australia (III), and north-central Australia (IV), suggesting a much more ancient radiation than previously believed. This conclusion is robust to different molecular clock estimations indicating divergence in Pliocene or Late Miocene, after landbridge dispersal to New Guinea had occurred. While members of clades I, III and IV are medium-sized, slender snakes, those of clade II attain large sizes and a robust build, rendering them top predators in their ecosystems. Genetic differentiation within clade II is low and haplotype distribution largely incongruent with geography or colour morphs, suggesting Pleistocene dispersal and recent ecomorph evolution. Significant haplotype diversity exists in clades III and IV, implying that clade IV comprises two species. Members of clade II are broadly sympatric with members of both northern Australian clades. Thus, our data support the recognition of at least five species from within P. aus- tralis (auct.) under various criteria. We discuss biogeo- graphical, ecological and medical implications of our findings. Electronic Supplementary Material Supplementary material is available in the online version of this article at http://dx.doi.org/10.1007/s00114-004-0602-0 Introduction Molecular techniques have proven to be crucial for the characterization of morphologically conservative species. Many recent studies revealed a common pattern of phy- logenetic segregation according to geographic breaks and not morphology or life history (Bonett and Chippindale 2004), parallel evolution of ecomorphs (Keogh et al. in press), and discordance with taxonomic concepts based on phenetic similarity, especially subspecies classifications (Burbrink et al. 2000). Various studies demonstrated deep genetic structure indicating a long history of isolation, but the possibility that cryptic species occur in sympatry throughout much of their range has rarely been addressed (Olson et al. 2004). In the case of venomous snakes, this may be of great medical importance because the com- position, biological activities and antigenic properties of snake venoms can vary dramatically between species (Chippaux et al. 1991). King brown snakes (Pseudechis australis) are Aus- tralia’s largest venomous snakes, attaining maximum lengths in excess of 3 m and considerable bulk (3–6 kg body mass at 2–2.5 m total length; Maryan 1997). Their U. Kuch ( ) ) · D. Mebs Zentrum der Rechtsmedizin, Klinikum der Johann Wolfgang Goethe-Universität, Kennedyallee 104, 60596 Frankfurt am Main, Germany e-mail: U.Kuch@em.uni-frankfurt.de Tel.: +49-69-63017418 Fax: +49-69-63015882 J. S. Keogh School of Botany and Zoology, Australian National University, Canberra, ACT 0200, Australia J. Weigel Australian Reptile Park, P.O. Box 737, Gosford, NSW 2250, Australia L. A. Smith Department of Terrestrial Vertebrates, Museum of Natural Science, Francis Street, Perth, WA 6000, Australia