The impact of climate change during the Pleistocene on organisms living in temperate areas of the world has been documented for more than fifty years (Rand, 1948). A major impact of Pleistocene glacial cycles on vertebrates in temperate zones is hypothesized to be the isolation of species into allopatric populations (Avise and Walker, 1998). Refugial isolation and subsequent lineage formation during the Pleistocene has been invoked as a primary mechanism resulting in modern species diversity (Johnson and Cicero, 2004). Under the Pleistocene speciation model, these glacial cycles acted as a ‘species pump,’ producing the majority of extant organisms inhabiting temperate areas. These arguments, usually focused on dating the origin of sister species using a fixed rate of divergence in birds, have suggested that the majority of speciation events occurred in the Pleistocene (Johnson and Cicero, 2004; Weir and Schluter, 2007). In a more limited study, Avise et al. (1998) suggested that the Pleistocene was important for speciation in reptiles, amphibians, and fish as well.
In contrast to the Pleistocene diversification model, it has been shown that many speciation events in birds occurred much earlier than the Pleistocene (Zink and Slowinski, 1995; Klicka and Zink, 1997), potentially calling into question the effects of glacial cycles on the formation of species. Unfortunately, most of these studies have invoked a molecular clock using a fixed rate of mtDNA substitution over time, and thus did not consider the impact of molecular rate variation for assessing error in divergence dates. The effect of Pleistocene climatic cycles on speciation has not been investigated in other large groups of vertebrates, particularly those that may be more sensitive to the effects of glacial cycles, such as ectotherms like reptiles. We investigate these patterns in a major New World (NW) group of colubroid snakes, the tribe Lampropeltini, inferring a robust phylogeny for the group and estimating the timing of speciation and diversification. We use multiple nuclear and mitochondrial genes and ‘relaxed phylogenetics’ methods (Drummond et al., 2006) to estimate phylogeny and dates of origin while accounting for the possibility of rate heterogeneity across branches.
The colubroid snakes are a diverse (>2500 species), globally distributed group (Lawson et al., 2005) which date to the early Cenozoic (Burbrink and Pyron, 2008). Of the several NW representatives of the group (Natricinae, Crotalinae, Elapinae, Colubrinae, and Xenodontinae), the colubrine tribe Lampropeltini is one of the most conspicuous and well-studied (Williams, 1978; RodríguezRobles and de Jesús-Escobar, 1999). The lampropeltinines (rat, corn, and fox [Pantherophis, Bogertophis, and Pseudelaphe], king and milk [Lampropeltis], short-tailed [Stilosoma], bull, gopher, and pine [Pituophis], glossy [Arizona], scarlet [Cemophora] and longnosed [Rhinocheilus] snakes) are common constrictors, distributed from Canada to Ecuador (Williams, 1978; Conant and Collins, 1998; Stebbins, 2003). Several recent studies have found that the Lampropeltini form a monophyletic clade endemic to the NW, thus rendering the cosmopolitan genus Elaphe paraphyletic (RodríguezRobles and de Jesús-Escobar, 1999; Utiger et al., 2002; Burbrink and Lawson, 2007).
Based primarily on trees inferred using mitochondrial evidence, the taxonomy of the group is in a state of flux and the monophyly of several genera (i.e. Pantherophis, Pituophis, and Lampropeltis) has been disputed, including the erection of a new genus (Mintonius) for the fox snakes (Pantherophis vulpinus; Bryson et al., 2007; Burbrink and Lawson, 2007; Collins and Taggart, 2008). Additionally, while many phylogeographic studies have used mtDNA to investigate biogeographic structure (Burbrink et al., 2000; Burbrink, 2002; Mulcahy, 2008; Rodríguez-Robles and de Jesús-Escobar, 2000), higher-level phylogenies based solely or primarily on mitochondrial data have not been well-supported (Rodríguez-Robles and de Jesús-Escobar, 1999; Burbrink and Lawson, 2007). Thus, multiple independent loci are desirable to infer phylogenies and estimate tree-based quantities such as divergence times (i.e. Wiens et al., 2008). Here, we present a phylogeny based on three nuclear genes (3368 bp), two of which are newly presented in this study, and six mitochondrial genes (4926 bp). We included representatives from all 31 of the traditionally described species of lampropeltinine. We use this phylogeny to address hypotheses regarding the timing of origin and diversification of the lampropeltinines, as well as generate a revised taxonomy of the group.