Rooted maximum likelihood phylogeny of the mitochondrial deoxyribonucleic acid gene ND2 for the eastern garter snake (Thamnophis sirtalis sirtalis, STX), the red-sided garter snake (T. s. parietalis, PTX), and the Texas garter snake (T. s. annectens, ATX), with geographic representation of sample locations. APTX represents individuals collected where the ranges of T. s. annectens and T. s. parietalis met. Numbers to the left indicate nonparametric bootstrap values (>50%) for those recovered in maximum likelihood analysis. Colored squares (G1, G2, and G3) correspond with locality data of samples used in genetic analysis. Gray areas represent historic ranges also shown in Fig. 1. Bold black line represents known boundaries. Dashed black line represents uncertain boundaries.

Contexts in source publication

Context 1

... into two groups (Dunithan, 2012). We based models on original identification from specimen sources (i.e., morphological and geographic location). (Table 1a). Intrasubspecific variation was low for each subspecies for the COI gene (i.e., less than 1.0%; Table 1b). A rooted maximum likelihood groups individuals into three clades: G1, G2, and G3 (Fig. 2). G1 consists of only T. s. sirtalis, G2 consists of all T. s. parietalis, two T. s. annectens, and two T. s. sirtalis individuals, and G3 consists of only T. s. annectens individuals (Fig. 2). These groupings separate across the landscape with G1 mostly in Louisiana, G2 in central Texas and central Oklahoma, and G3 in far western ...

Context 2

... low for each subspecies for the COI gene (i.e., less than 1.0%; Table 1b). A rooted maximum likelihood groups individuals into three clades: G1, G2, and G3 (Fig. 2). G1 consists of only T. s. sirtalis, G2 consists of all T. s. parietalis, two T. s. annectens, and two T. s. sirtalis individuals, and G3 consists of only T. s. annectens individuals (Fig. 2). These groupings separate across the landscape with G1 mostly in Louisiana, G2 in central Texas and central Oklahoma, and G3 in far western Oklahoma (Fig. 2). In G2, two T. s. sirtalis (STX14 and STX16) are more closely related to T. s. parietalis and T. s. annectens individuals than to other T. s. sirtalis (Fig. 2), suggesting that ...

Context 3

... 2). G1 consists of only T. s. sirtalis, G2 consists of all T. s. parietalis, two T. s. annectens, and two T. s. sirtalis individuals, and G3 consists of only T. s. annectens individuals (Fig. 2). These groupings separate across the landscape with G1 mostly in Louisiana, G2 in central Texas and central Oklahoma, and G3 in far western Oklahoma (Fig. 2). In G2, two T. s. sirtalis (STX14 and STX16) are more closely related to T. s. parietalis and T. s. annectens individuals than to other T. s. sirtalis (Fig. 2), suggesting that these individuals were misclassified as T. s. parietalis/annectens (see ''Discussion'' for detailed explanation). Variation within T. s. sirtalis is reduced by ...

Context 4

... T. s. annectens individuals (Fig. 2). These groupings separate across the landscape with G1 mostly in Louisiana, G2 in central Texas and central Oklahoma, and G3 in far western Oklahoma (Fig. 2). In G2, two T. s. sirtalis (STX14 and STX16) are more closely related to T. s. parietalis and T. s. annectens individuals than to other T. s. sirtalis (Fig. 2), suggesting that these individuals were misclassified as T. s. parietalis/annectens (see ''Discussion'' for detailed explanation). Variation within T. s. sirtalis is reduced by 3.0% (from 4.5 to 1.5%; Table 1b) when these T. s. sirtalis individuals in G2 are removed from the analysis, and when we excluded G2 T. s. sirtalis samples in ...

Context 5

... 1b) when these T. s. sirtalis individuals in G2 are removed from the analysis, and when we excluded G2 T. s. sirtalis samples in the divergence analysis, rates were still above 4.65% (Table 1a) for each pairing of T. s. sirtalis with other subspecies. We found T. s. annectens in two groups (G2 and G3); G3 is composed solely of T. s. annectens (Fig. 2) and G2 contains all T. s. parietalis individuals with two T. s. sirtalis (mentioned above). G3 with only T. s. annectens was within the larger G2 (Fig. 2). When G3 T. s. annectens individuals (ATX12, ATX13, and ATX18) are treated as the only ''true'' T. s. annectens group, the intersubspecific sequence divergence rate for ND2 between ...

Context 6

... rates were still above 4.65% (Table 1a) for each pairing of T. s. sirtalis with other subspecies. We found T. s. annectens in two groups (G2 and G3); G3 is composed solely of T. s. annectens (Fig. 2) and G2 contains all T. s. parietalis individuals with two T. s. sirtalis (mentioned above). G3 with only T. s. annectens was within the larger G2 (Fig. 2). When G3 T. s. annectens individuals (ATX12, ATX13, and ATX18) are treated as the only ''true'' T. s. annectens group, the intersubspecific sequence divergence rate for ND2 between T. s. parietalis and T. s. annectens is still <1% (Table 1a). Both G2 and G3 that include T. s. annectens are placed more closely to T. s. parietalis than ...

Context 7

... G3 T. s. annectens individuals (ATX12, ATX13, and ATX18) are treated as the only ''true'' T. s. annectens group, the intersubspecific sequence divergence rate for ND2 between T. s. parietalis and T. s. annectens is still <1% (Table 1a). Both G2 and G3 that include T. s. annectens are placed more closely to T. s. parietalis than to T. s. sirtalis (Fig. ...

Context 8

... genes for three subspecies of Thamnophis sirtalis. NA = not applicable. Three other notable results are implied by our phylogeny. Two involve unusual groupings of individuals from our data set. The first includes the two individuals (APTX57 and APTX58) in G2, mostly consisting of other T. s. parietalis and T. s. annectens individuals (Fig. 2). We collected these individuals from locations in Oklahoma where T. s. parietalis and T. s. annectens are thought to interbreed ( Fig. 2; Rossman et al., 1996); therefore, we believe that these two individuals represent T. s. annectens/T. s. parietalis intergrades. If this is the case, we are the first to find empirical support of ...

Context 9

... second notable result suggested by our phylogeny is the placement of two individuals in G2 (STX14 and STX16) that were originally identified as T. s. sirtalis (Fig. 2). We believe this to be a simple case of misidentification, because T. sirtalis is known to exhibit a very wide range of morphological features throughout its range (e.g., Rossman et al., 1996), and different subspecies are occasionally misidentified in areas where their ranges overlap. Note that the two T. s. sirtalis (STX14 and STX16) ...

Context 10

... are occasionally misidentified in areas where their ranges overlap. Note that the two T. s. sirtalis (STX14 and STX16) in G2 that are most closely related to T. s. parietalis and T. s. annectens individuals were both collected in Arkansas counties close to the Oklahoma border, near the range boundaries of both T. s. parietalis and T. s. sirtalis (Fig. 2). A reduction in divergence among the T. s. sirtalis group was seen when these two individuals (from G2) were removed (Table 1). Genetic changes within the T. s. parietalis grouping were not observed when the two individuals from G2 (STX 14 and 16) were included, supporting the idea that they are indeed T. s. parietalis (Table 1). ...

Context 11

... two individuals from G2 (STX 14 and 16) were included, supporting the idea that they are indeed T. s. parietalis (Table 1). Another interesting result revealed by our phylogeny involves the G3 clade, consisting of three individuals (ATX12, ATX13, and ATX18) identified as T. s. annectens on the basis of collection location (Rossman et al., 1996; Fig. 2). These three individuals are from counties in far western Oklahoma (Fig. 2) close to the Texas Panhandle. We believe that these individuals are indeed T. s. annectens, but represent a genetically distinct lineage, perhaps reflective of a collection area somewhat extreme relative to the rest of the subspecies. All other T. s. annectens ...

Context 12

... that they are indeed T. s. parietalis (Table 1). Another interesting result revealed by our phylogeny involves the G3 clade, consisting of three individuals (ATX12, ATX13, and ATX18) identified as T. s. annectens on the basis of collection location (Rossman et al., 1996; Fig. 2). These three individuals are from counties in far western Oklahoma (Fig. 2) close to the Texas Panhandle. We believe that these individuals are indeed T. s. annectens, but represent a genetically distinct lineage, perhaps reflective of a collection area somewhat extreme relative to the rest of the subspecies. All other T. s. annectens individuals from central Texas were undifferentiated from the larger G2 (T. ...

Context 13

... the Texas Panhandle. We believe that these individuals are indeed T. s. annectens, but represent a genetically distinct lineage, perhaps reflective of a collection area somewhat extreme relative to the rest of the subspecies. All other T. s. annectens individuals from central Texas were undifferentiated from the larger G2 (T. s. parietalis) clade (Fig. 2). The distribution of T. s. parietalis does not currently extend into central Texas (Fig. 1); however, the majority of records for T. s. annectens are from central Texas, suggesting that the Texas Panhandle populations represented in G3 might have originated from populations in central Texas. Future investigation in central Texas and ...

Context 14

... T. s. annectens has been most abundant in tall grass prairie habitat such as the Blackland Prairie (Tennant, 2003); however, in recent decades these prairies have been disturbed and often completely destroyed by conversion to agricultural areas. Our model suggests that the Great Plains floodplain system and eastcentral Texas plains oak savannah and woodlands may now be more suitable habitat (Table 2). ...

Context 15

... niche modeling and molecular ecology analyses complement each other and indicate that T. s. annectens is distinctive relative to the other two subspecies of T. sirtalis that occur in Texas. Despite the limitations of restricted sampling, we reveal that both T. s. annectens and T. s. parietalis show high genetic divergence from T. s. sirtalis ( Fig. 2; Table 1). This genetic differentiation may be because of a lack of sympatry throughout much of the ranges of the three subspecies in Texas as evidenced by our niche models (Fig. 3). Thamnophis s. annectens and T. s. parietalis are shown here to be ecologically different, although they are similar genetically except for the Texas ...

Context 16

... two subspecies is considered (Rossman et al., 1996). This ecological difference may have started as phenotypic plasticity within T. s. annectens, which, over time, resulted in niche evolution via genetic assimilation ( Pigliucci et al., 2006). These lineages may be in the process of speciation and mitochondrial DNA has not yet fully yet diverged (Fig. 2) as seen between Butler's garter snake (Thamnophis butleri) and the plains garter snake (Thamnophis radix;Fitzpatrick et al., 2008). Given its ecological differences from T. s. parietalis, combined with genetic distinctiveness of the Texas Panhandle group from the central Texas grouping, T. s. annectens is distinguishable from the two ...