Rediscovery of Liophis taeniurus Tschudi (Reptilia, Serpentes Colubridae) and Its Relationship to Other Andean Colubrid Snakes

Abstract

The holotype of Tschudi's (1845) Liophis taeniurus was found and data obtained on the holotype and additional specimens were compared with similar data on other colubrid species and genera that have been considered related to the former species. Liophis festae is here considered a distinct species. Tschudi (1845) described Liophis taeniurus on the basis of one specimen although he indicated that they had found a few specimens of this species in the hot forested region (Peru). Over the years this species has been recorded from Guayaquil, Ecuador (Boulenger 1894), Costa Rica and Panama (Dunn 1937), and Colombia (Dunn 1944). All of these localities are on the Pacific side of the Andes or in interAndean valleys while the proposed type locality is Cisandean. Until now, L. taeniurus has remained an obscure species, usually found on museum shelves under the species names of albiventris, bimaculatus, reginae, cobella, and epinephalus. Tschudi's (1845) original description was based primarily on color and the only squamative data presented was the number of ventrals and caudals. The number of caudals was thought to be incomplete by Tschudi because the tail appeared to be broken off. His color description was accurate, but since the tail appeared broken off, the ventral tail color may have been in error. Therefore any snake resembling the color pattern described by Tschudi, with approximately the same range of ventrals and possibly caudals, may have been erroneously labeled taeniurus, especially since the type was considered lost and verification was impossible at the time. Tschudi's scanty description led Boulenger (1894) to include all specimens at the British Museum that fit the description of taeniurus into his account of that species in the Catalogue of Snakes.... This basic error led to a definition of taeniurus that was incorrect because Boulenger's series contained at least two species, neither of which were taeniurus. Boulenger described a midbody scale row variation of 17 to 19 rows, ventrals 148 to 185, subcaudals 53 to 57, 1 preocular, 2 postoculars, temporals 1 + 2, a tail length of 20 to 23 percent of the total length, and a color pattern that superficially resembled that of Tschudi's (1845) original description. Amaral (1929) accepted Boulenger's account and was convinced that taeniurus was polytypic, and relegated albiventris and bipraeocularis as subspecies of taeniurus. This led Dunn (1937) to describe a new subspecies, taeniurus juvenalis, from the mountain slopes of Costa Rica and northern Panama, add epinephalus as a subspecies of taeniurus, and place Liophis reginae albiventris Jan, as a synonym of L. taeniurus ephinephalus. By 1944, Dunn realized the complexity of the problem, returned epinephalus to specific rank and placed juvenalis as a subspecies of epinephalus. Dunn (1944) also indicated that a specimen from Pampolona, Colombia (2340 m) may have served as the basis of a Colombian record for taeniurus. The latter record probably represents bimaculatus because Dunn indicated that taeniurus did not occur in Colombia. Dunn (1937) apparently surmised that Liophis and Leimadophis were closely related and transferred taeniurus and epinephalus to Leimadophis without explanation. Most species with dorsolateral light or dark tail lines have been placed in Leimadophis for more than 40 years. The generic status of Leimadophis has been questioned by Maglio (1970), and he suggested that Dromicus and Leimadophis are congeneric and Dromicus is the older name. The generic status of 1 Present Address: Dept. of Biological Sciences, Illinois State University, Normal, IL 61761