ABSTRACT
Snakes are among the most charismatic and highly-studied organisms (Greene 1997), yet our understanding of their early evolution and phylogeny remains in a state of flux. Extensive anatomical information (e.g., Underwood 1967; McDowell 1974, 1975, 1979), analyzed using quantitative phylogenetic methods (e.g., Kluge 1991; Cundall et al. 1993), had led to a broad consensus on relationships among living snakes (Lee and Scanlon 2002; Rieppel et al. 2003). The tiny, burrowing, worm-like blindsnakes (scolecophidians) were considered the most basal clade of living snakes, followed by other small burrowing taxa with restricted gapes (pipesnakes and shieldtail snakes). The partly surface-active, and moderategaped sunbeam snakes (Xenopeltis and Loxocemus) were transitional forms, while the typical, generally surface-active and large-gaped snakes (such as pythons, boas, and colubroids) were inferred to represent a single, derived radiation (“core macrostomatans”). This phylogeny implied that snake evolution involved consistent trends towards greater surface activity, increased body size, and enlarged gape (e.g., Underwood 1967; Rodríguez-Robles et al. 1999). However, some studies of primitive fossil snakes with large body size and extensive gapes did not support this scenario, although the exact phylogenetic position of these fossils remains debated (e.g., Caldwell 2007; Wilson et al. 2010). Most recently, increasingly large molecular sequence datasets have further challenged the traditional scenario (Slowinski and Lawson 2002; Wilcox et al. 2002; Lawson et al. 2005; Vidal et al. 2007a, 2007b; Wiens et al. 2008; Burbrink and Crother 2010).
