Evolution of coloration, urotomy and coral snake mimicry in the snake genus Scaphiodontophis (Serpentes: Colubridae)

Abstract

The Neotropical hinged-tooth, coral snake mimics of the genusScaphiodontophisare characterized by extremely long and disproportionately thick tails that are extremely fragile. Both the coloration and tail structure are putative antipredator devices. While all examples have components of the coloration that match those of the venomous coral snakes (family Elapidae), the range of variation is extreme, leading to controversy on the status of various populations, including nine named taxa. Individual, ontogenetic and geographic variation in scutellation and head, body and tail coloration were analysed to evaluate population status and possible evolutionary trends based on a sample of 183 examples from Mexico, Central America and Colombia. Variation in subcaudal counts show population differences (higher in Mexico and upper Central America) but are not congruent with geographic variation in coloration. Generally snakes from north of Nicaragua and from central and eastern Panama have a pattern of dyads (black-light-black bands separating red bands), those from Atlantic slope Nicaragua to western Panama a pattern of monads (light-black-light bands separating the red ones) and those from Colombia have both pattern types on the same snake. The dyads and/or monads may be present the length of the body and tail, restricted to the anterior part of the body or on the entire body or on the anterior part of the body and on the tail. Two or more of these variants may occur at a single geographic locality or only a single one may be present. Head and nuchal colour patterns (Z, A, V, and Du) are relatively consistent geographically. The Atlantic slope Guatemala, Belize and Honduras population have the A pattern, those of Nicaragua, Costa Rica and western Panama the V pattern, and those in Colombia a Du pattern. Other populations have the Z coloration. Intermediate conditions in coloration of the body and tail and head and neck are found at localities intermediate between the main pattern types, indicating intergradation among adjacent populations. Consequently, we regard these snakes as representative of a single species,Scaphiodontophis annulatusDumeril and Bibron and the eight other names applied to various populations and individuals as synonyms. Analysis of colour pattern leads us to the conclusion that the tricolour pattern evolved from a uniform one through a lineate-spotted condition (usually present on the non-tricolour portions of the snake) through a bicolour red and black pattern to the dyadal condition. The monadal pattern in turn was derived from the dyadal one. The data further indicates that tricolour components first appeared anteriorly and progressively expanded posteriorly. The evolutionary sequence for the head and nuchal pattern appears to be A → Z → V → DuS. annulatushas a series of jaw and tooth specializations designed for rapid processing of hard-bodied prey found during diurnal foraging in the leaf-litter. Urotomy in this species involves intervertebral tail-breakage (pseudoautotomy) without regeneration. Evidence is presented supporting the long-tail multiple break hypothesis as applicable toScaphiodontophisand other snakes with similar tail morphology (specialized pseudoautotomy). This is in contrast to snakes with similar tail morphology (specialized pseudoautotomy). This is in contrast toConiophanesand other snakes with a high incidence of urotomy having long but unspecialized tails (unspecialized pseudoautotomy) without multiple breaks over time. AllScaphiodontophiscolour patterns have a general resemblance to that of venomous coral snakes and offer protection from generalizing predators having innate or other triggered responses to coral snake colours. The aposematic effect is enhanced by tail thrashing and head twitching behaviours. The characteristic foraging pose ofS. annulatus, which tends to expose the head and anterior body, makes even the incomplete tricolour pattern effective as an antipredator defense. No evidence supports the idea that tail thrashing or the incomplete tricolour pattern directs the predator attacks to the tail to expedite pseudoautotomy. Coral snake mimicry and specialized pseudoautotomy are shown not to be co-evolved and pseudautotomy seems to have evolved long before mimetic coloration in this genus.