Geographic variation in the long-nosed snake, Rhinocheilus lecontei (Colubridae): beyond the subspecies debate

Scalation, colour pattern, linear and geometric morphometrics were used to quantify geographical differentiation in the long-nosed snake, Rhinocheilus lecontei, and to test the hypothesis that all four subspecies are morphologically distinct. Also investigated were potential associations between morphological (scalation, colour pattern, linear measurements) and environmental variables (climate, vegetation, soil). Sexual dimorphism was weakest for geometric and strongest for linear morphometric variables. Morphological variables differed widely in their ability to differentiate subspecies. Linear morphometric variables achieved the most statistically significant pairwise Mahalanobis distances between subspecies, while geometric morphometrics largely failed to differentiate them. Colour pattern showed the strongest and linear morphometrics the weakest correlation with environment. Several characters varied continuously along latitudinal or longitudinal gradients, such that, in some cases, the clines for closely related traits were discordant. No one subspecies was consistently divergent in all analyses, leading to the conclusion that the three mainland subspecies are not sufficiently distinct to warrant separate subspecies status. The island subspecies, though not always statistically distinct, is geographically separate from other populations and differs in characters related to size. Given the small number of specimens available, a decision regarding its taxonomic status (i.e. elevation to species level) is best deferred until additional specimens can be examined and data on molecular variation can be analysed. © 2004 The Linnean Society of London, Biological Journal of the Linnean Society, 2004, 83, 65–85.