The colubrid snakes of the genus Dendrelaphis Boulenger, 1890 are widely distributed, ranging from Pakistan in the West to the northern and eastern coast of Australia in the East and South and to southern China in the North (Ziegler and Vogel, 1999). Members of the genus Dendrelaphis are slender, diurnal species that are predominantly arboreal and feed mainly on lizards and amphibians.
Boulenger (1894), Wall (1921a), Meise and Henning (1932), Mertens (1934) and Smith (1943) have in turn revised the systematics of this genus and have in turn disagreed with one another. As such, it is not surprising that the systematics of this genus have remained ambiguous as well as incomplete, a fact that was underlined by the recent descriptions of two wide-spread as well as rather common Southeast Asian species, Dendrelaphis kopsteini Vogel and Van Rooijen, 2007 and Dendrelaphis haasi Van Rooijen and Vogel, 2008.
Dendrelaphis tristis (Daudin, 1803), as hitherto defined, inhabits Sri Lanka, India, Pakistan, Bangladesh and Myanmar (Ziegler and Vogel, 1999). Daudin described D. tristis on the basis of a plate and some additional information provided by Russell (1796). A type was not deposited in a collection as was ususal at that time. D. tristis superficially resembles D. pictus (Gmelin, 1789). Consequently, Meise and Henning (1932) doubted the validity of its specific status. This illustrates the rather crude approach to the systematics of this genus at that time. D. tristis actually differs in many aspects from D. pictus. For instance, D. tristis has a substantially more stocky build, smaller vertebral scales and a conspicuous vertebral stripe which is absent in D. pictus. The names D. tristis and D. pictus in fact refer to separate clades each composed of two or more species (this report; Van Rooijen and Vogel, 2008; Vogel and Van Rooijen, in prep).
We initially examined a dozen specimens referred to D. tristis in the context of a taxonomic revision of the genus and noticed that the distribution of the number of ventral scales was bimodal. Examination of additional specimens and further research demonstrated the two superimposed distributions correspond with other differences in morphology as well as differences in coloration. This dichotomy of forms could not be attributed to sexual dimorphism and thus corresponds with two distinct taxa. In this paper, univariate and multivariate statistical techniques are used to illustrate and confirm these findings.