Fig 1.
Operational Taxonomic Units in the Western Rattlesnake Complex.
The Western Rattlesnake, historically composed of 9 subspecies (sensu Klauber), displays a level of phenotypic variability that is concomitant with its vast geographic range in North America. This variability is captured in the following images, identified to subspecies and location, and published with permission of the photographer (as indicated). Subspecies are as follows (from left to right, top to bottom): Crotalus cerberus (Pinal County AZ: Martin Feldner); C. oreganus (San Luis Obispo County CA: Martin Feldner); C. o. helleri (Los Angeles County CA: Martin Feldner); C. o. caliginis (San Coronado Island, Baja California: Rob Olivier); C. o. concolor (Coconino County AZ: Martin Feldner); C. o. lutosus (Mohave County AZ: Martin Feldner); C. o. abyssus (Coconino County AZ: Martin Feldner); C. vviridis viridis (Harding County SD: Mark Davis); C. v. nuntius (Coconino County AZ: William Wells).
Fig 2.
The topographic distribution of nine Western Rattlesnake OTUs (Operational Taxonomic Units) in North America.
The nine Western Rattlesnake OTUs vary greatly in size and extent of distributional area. The Colorado Plateau represents a suture between eastern and western lineages, as six of nine occur there as either parapatric or sympatric.
Fig 3.
Lateral and dorsal landmarks used for subsequent shape analysis in the Western Rattlesnake complex.
A total of 33 dorsal and 50 lateral landmarks were selected to assess shape in the Western Rattlesnake complex. Landmarks were composed of both Type I (white) and sliding semi (black) landmarks.
Fig 4.
Phylogeny, transformation grids, and head shape variation within and among subspecies of the Western Rattlesnake (Crotalus viridis) complex.
Left: A well-resolved Bayesian phylogenetic hypothesis was derived from six concatenated mtDNA sequences (to include an outgroup, Crotalus scutulatus, subsequently pruned from the tree). Nodes are numbered according to ancestral character states estimated (Table 1). Posterior node probabilities are 1.00, except for node 14, which is 0.99. Transformation grids that denote the deviation from mean form were derived for all nine subspecies using both dorsal and lateral landmark configurations. Transformation grids illustrate the shift from a more ovoid head shape in the eastern (viridis + nuntius) lineage to a more stereotypic, spearheaded morphology in the western lineage (cerberus + oreganus + concolor + helleri + caliginis + lutosus + abyssus). In addition, sister subspecies reflect a shift from an elongate snout and compressed head in the larger, more widespread form (i.e. viridis, helleri, and lutosus), to a shorter, more compact, and less compressed head shape in the diminutive forms (i.e. nuntius, caliginis, and abyssus, respectively). Right: The subspecies shapes correspond to mean positions in the among-subspecies PC plots. Ellipsoids represent scaling of one standard error of the mean (top) and 95% confidence limits (bottom) for each PC. The three PCs account for 58.0% of the variation among subspecies. Colors of ellipsoids match subspecies, as depicted in terminal branches of the phylogeny. The ancestral states and phylogeny edges are projected into the PC plots to facilitate interpretations. The bolder red edge corresponds to node 10, which separates clades.
Table 1.
Results of a non-parametric multivariate analysis of variance (np-MANOVA) for head shape conducted on the Western Rattlesnake (Crotalus viridis) complex.
Type I (sequential) sums of squares (= SS) were used to calculate sums of squares and cross-products matrices (SSCP) (see Supplemental Information for details). Head size = log(CS) whereas subspecific identification = Subspecies. P-values (= P) were computed from 10,000 random permutations of the randomization procedure. Z-values (= Z) are standard deviates of observed SS from sampling distributions. MS = Mean-Squares; R2 = coefficient of determination; F = F-statistic.
Table 2.
Pairwise head shape comparisons among subspecies within the Western Rattlesnake (Crotalus viridis) complex.
Pairwise subspecific shape differences (i.e., Procrustes distances) are arrayed below the diagonal, whereas P-values that stem from 10,000 random permutations of data are above the diagonal. The single non-significant Procrustes distance (i.e., cerberus versus abyssus) is depicted in bold, as is its corresponding P-value. Values in bold italics represent Procrustes distance comparisons among the three sister-taxa (i.e., lutosus/ abyssus; helleri/ calignis; viridis/ nuntius), with corresponding P-values in bold blue as well.
Table 3.
Pairwise head shape comparisons among ancestral character states within the Western Rattlesnake (Crotalus viridis) complex.
Distances compiled among ancestral character states are provided for comparisons. Numbers refer to ancestral nodes in the phylogeny (Fig 4)
Fig 5.
Results of a disparity through time (DTT) analysis plotted for the Western Rattlesnake (Crotalus viridis) complex.
The solid line indicates the DTT of mean within-subclade head shapes. The dashed line indicated the median with the gray area representing 95% confidence limits, as derived from 10,000 simulations of Brownian evolution. The morphological disparity index (MDI), computed as average squared Euclidean distances between the DTT and median values, was 0.4271.
Fig 6.
Plots depicting Bayesian posterior probabilities of assigning specimens to subspecies with phylogenetically informed prior probabilities in the Western Rattlesnake (Crotalus viridis) complex.
Interquartile ranges are shown as error bars, with median values denoted by bolder notches. Each plot indicates the actual subspecific designation in the title.
Table 4.
A revised taxonomy for six species of the Crotalus viridis complex based upon molecular and morphological data and an integrated taxonomy approach.