Variation and Hybridization in Elaphe bairdi (Serpentes: Colubridae)

Abstract

-The distributional status, and patterns of external and genetic variation in Elaphe bairdi are reviewed and compared with Elaphe obsoleta. Low-level hybridization between the two taxa, as inferred from morphological criteria, is confined to a distinct geographic and ecological region in central Texas. The two parapatric taxa are otherwise morphologically distinct. Genic data from an electrophoretic survey demonstrate no fixed electromorph differences between E. bairdi and E. obsoleta, and the geographic distribution of alleles at three putative loci suggest remnants of widespread allelic distributions in the common ancestral lineage. There is some ambiguous evidence of genetic introgression, but the overall genic/morphological information indicates that E. bairdi has undergone differentiation in isolation from an E. obsoleta precursor, and has since secondarily re-established geographic contact with descendant populations of E. obsoleta in the present zone of parapatry. Baird's Rat Snake, Elaphe bairdi, is a rather specialized member of the genus with a limited distribution in northwestern Mexico and westcentral Texas. Because of its restricted habitat and secretive habits, it is seldom encountered in the field and has been historically rare in museum collections. Originally described by Yarrow in 1880 from a single specimen, fewer than 20 individuals were available for study prior to 1950. About this time, however, enough information had become available to suggest that the species represented a western race of the widespread Elaphe obsoleta of the eastern United States (Milstead et al., 1950; see also dated field notes in Wright and Wright, 1957). The first publication of the combination Elaphe obsoleta bairdi appeared in Dowling's (1952) checklist of American Elaphe. The basis for his action is contained in his unpublished Ph.D. dissertation (1951), where he documented intergradation between Elaphe obsoleta lindheimeri and E. bairdi at the zone of contact on the margin of the Edwards Plateau in central Texas. By the late 1960s greatly augmented collecting activity in western Texas began to produce more specimens of the bairdi form. In a preliminary study of Texas Elaphe obsoleta, Lieb (1971) demonstrated pronounced differences in features between E. o. bairdi and E. o. lindheimeri. He suggested that the zone of intergradation between the two must be quite narrow, and that additional study of populations on the Edwards Plateau would have to be undertaken before a definitive statement on the status of E. bairdi could be made. A review of the status of Baird's Rat Snake was made by Olson (1977), who re-evaluated the differences between E. o. bairdi and E. o. lindheimeri from the standpoint of the amount of intergradation observed. In contrast to Dowling (1951), Olson interpreted the intermediate forms as rare hybrids of distinct species, and suggested that Elaphe bairdi be recognized at full specific rank. The crux of both Dowling's and Olson's arguments lies in the nature and amount of gene exchange between the two forms at the putative contact zone. Both accounts, however, suffer from a lack of documentation of the variation in the populations involved. Dowling (1951) had two specimens of E. bairdi, eight E. o. lindheimeri from Texas, and an unspecified number of intermediates. Olson (1977) reported data on 20 specimens of E. bairdi, including six specimens from the eastern edge of the range, and 20 E. o. lindheimeri from populations adjacent to the range of E. bairdi. Of the latter, three individuals from part of the contact zone were designated as hybrids. Olson further stated that there was no evidence of introgression in the two forms. This paper reviews the evidence for intergradation or hybridization between the nominal forms E. bairdi and E. o. lindheimeri through a review of the distributional status, external character variation with discriminant function analysis, and allozyme patterns of the relevant populations. Although our terminology follows Olson (1977) in the use of the combination Elaphe This content downloaded from 207.46.13.101 on Sat, 08 Oct 2016 05:23:13 UTC All use subject to http://about.jstor.org/terms STATUS OF ELAPHE BAIRDI bairdi, such usage is for convenience rather than an a priori taxonomic recommendation. MATERIALS AND METHODS Museum specimens were examined to document geographic distributions and collect data on the following characters: numbers of ventral and subcaudal scales, supralabials, and body blotches; the presence of dorsal body striping; and body blotch length. These characters were selected for their relevance (Wright and Wright, 1957; Dowling, 1951; Lieb, 1971; Olson, 1977). The sample included 85 specimens of Elaphe bairdi from all parts of the range, and 222 individuals of Elaphe obsoleta lindheimeri. A second aspect of this study was to see if gene products existed that would serve as markers to identify populations of Elaphe bairdi and E. obsoleta. For this purpose, 10 E. bairdi, four E. o. lindheimeri, four E. o. quadrivittata, two E. o. obsoleta, and one each of E. o. rossalleni and E. o. spiloides were obtained. Two methods were used to obtain tissues for protein assay. Snakes collected in the field were transported to the laboratory where they were killed and samples of blood plasma, packed blood cells, skeletal muscle, liver, kidney and heart were obtained (see Lawson and Dessauer, 1979). Blood was also collected from living snakes (principally E. bairdi) held in captivity by zoological parks, scientific institutions, or individuals. These blood samples were obtained by cardiac puncture from adult snakes, centrifuged into plasma and packed cell fractions, and frozen in liquid nitrogen for transportation to the laboratory. Individual tissues were homogenized in an equal volume of 0.25 M sucrose solution containing 10 mM dithiothreitol. Using a standard horizontal starch-gel system (Selander et al., 1971), plasma and homogenates were assayed for the non-enzymatic proteins hemoglobin, transferrin, and albumin, and the enzymes Octanol dehydrogenase (E.C. 1.1.1.73), Glycerol-3phosphate dehydrogenase (E.C. 1.1.1.8), Lactate dehydrogenase-1, -2 (E.C. 1.1.1.41), Superoxide dismutase-1, -2 (E.C. 1.15.1.1), Aspartate aminotransferase-1, -2 (E.C. 2.6.1.1), Creatine kinase-1 (E.C. 2.7.3.2), Adenylate kinase (E.C. 2.7.4.3), Arylesterase (E.C. 3.1.1.2), Tripeptide aminopeptidase (E.C. 3.4.11.4), Fumarate hydratase (E.C. 4.2.1.2), Mannose-6-phosphate isomerase (E.C. 5.3.1.8), Glucose-6-phosphate isomerase (E.C. 5.3.1.9), and Phosphoglucomutase (E.C. 5.4.2.2), Electromorphs were rendered visible using staining techniques described by Courtright et al. (1966), Brewer (1970), Matthews (1975), and Harris and Hopkinson (1976). Electrophoretic and buffer conditions followed Lawson and Dessauer (1979). 104 102 100 98 96 94 FIG. 1. Generalized distributional map for Elaphe bairdi and E. obsoleta in southwestern North America. Geographic sampling areas for Table 2 and Figs. 3-6 are indicated: ET, East Texas; XT, Cross-Timbers; NP, Northern Edwards Plateau; OP, Peripheral to Edwards Plateau; SP, Southeastern Edwards Plateau; EB, Eastern E. bairdi; WB, Western E. bairdi; SB, Southern E. bairdi. The rectangle indicates the area detailed in Fig. 2. Following the initial survey of allelic variation, additional samples of blood and other tissues were sought from specimens of E. bairdi and E. obsoleta, concentrating on the area of parapatry between the two forms in central Texas. A total of 82 individuals was eventually used in electrophoretic studies. All specimens utilized in morphological and electrophoretic analyses are listed in Appendix I.