Skeptics of the currently en vogue mitochondrial DNA sequencing techniques for generating phylogenetic trees have occasionally joked that even such an "obviously" homogeneous taxon as the Western Diamond-backed Rattlesnake (Crotalus atrox) would doubtless be split into several "cryptic" species if ever subjected to mtDNA-based study. Imagine my surprise to discover that not only have a couple of such a studies been done, but that the results support the traditional notion of atrox as a relatively homogeneous species despite its broad range.
Castoe, Spencer, and Parkinson (2007) sampled tissues of C. atrox from across its range (including Mexico but excluding peripheral populations in Nevada, Oklahoma, and Arkansas) for mtDNA sequencing. Although numerous haplotypes were discerned during the study, most samples resolved into two clades ("Eastern" and "Western"), which are hypothesized to have diverged slightly from each other as a result of Pleistocene climatic cooling events pushing populations further to the south into refugia about 1.36 million years ago. Prior to this, the Eastern and Western clades were apparently in contact along the Cochise filter barrier (roughly between the Animas and Chiricahua mountains in NM and AZ, respectively) and this contact has resumed in post-Pleistocene times. This study also indicates that the Eastern clade experienced a much greater contraction during the Pleistocene than was experienced by the Western clade, but the Eastern clade currently occupies a substantially larger range than the Western one, implying a dynamic post-Pleistocene expansion in the east.
Carol L. Spencer (2008), who was one of the investigators in the above mtDNA study, also conducted a complementary morphological study of C. atrox as part of her doctoral requirements at the University of Texas - Arlington. Originally, Spencer collected measurements and counts of 68 morphological characters from 922 museum specimens of all age groups from throughout the range of the species. The number of characters was then reduced to 37 by eliminating those that were repetitive or that showed no variation across specimens. Counts and measurements of the juveniles were also eliminated because of purported high rates of variation among subadults and the fact that the proportions of various measurements may change from the juvenile to the adult stage. This data winnowing left Spencer with information from 673 adults, which were then subjected to multivariate and univariate statistical programs for analysis.
The results of this analysis failed to show any appreciable patterns for most of the morphological characters investigated, although a number of east-west clines were detected, notably in a few scale counts, and a north-south cline in body size (which seems to differ from the conventional wisdom and which the author attributes to an adherence to Bergman's Rule). Regardless of these clinal variations, the author found more variation within the three groups she investigated (Eastern, Central, and Western) than between them, effectively discounting the possibility that differentiation occurs along the most probable zone (the aforementioned Cochise Filter Barrier, which extends along the Continental Divide in SE AZ and SW NM). This conclusion agreed with the previous mtDNA study (Castoe et al., above and op cit), which showed an intergradation of mitochondrial genes in that region, indicating that although different haplotypes had developed during their more southerly Pleistocene sequestration, both clades had been freely interbreeding since their post-Pleistocene reintroduction to each other.
These two studies offer up an example of what we should see more of in herpetological taxonomy: the application of various, complementary methodologies to a problem instead of making a quixotic decision to disturb taxonomic stability based upon a single study involving a technique (mtDNA sequencing) that is becoming increasingly suspect when compared to results from other methods. The chauvinism exhibited by molecular taxonomists towards other (especially morphological) methods tends toward an unscientific incuriosity in many cases. All morphological characters are, after all, determined by nuclear DNA, not by mitochondrial DNA, and it is becoming increasingly evident that mitochondrial DNA sequencing is inadequate (by itself) to detect many intraspecific populations and subspecies that are "obvious" in a cursory morphological inspection. Additionally, I am becoming increasingly suspicious that the "phylogenetic history" recorded within the two respective genomes (mitochondrial and nuclear) will eventually prove to be significantly different for most species.
Castoe, T.A., Spencer, C.L., and C.L. Parkinson. 2007. Phylogeographic structure and historical demography of the western diamondback rattlesnake (Crotalus atrox): A perspective on North American desert biogeography. Molecular Phylogenetics and Evolution 42(2007): 193-212. [PDF]
Spencer, Carol L. 2008. Geographic variation in Western Diamond-backed Rattlesnake (Crotalus atrox) morphology. Pp. 55-78. In Hayes, W.K., Cardwell, M.D., Beaman, K.R., and S.P. Bush (eds.). 2008. The Biology of Rattlesnakes. Loma Linda, CA: Loma Linda Univ. Press, xvi + 606 pp., 20 color plates. [PDF]