Morphological and molecular evidence indicates that the Gulf Coast box turtle (Terrapene carolina major) is not a distinct evolutionary lineage in the Florida Panhandle

Abstract

Four extant subspecies of Terrapene carolina in eastern North America, Terrapene carolina bauri, Terrapene carolina carolina, Terrapene carolina triunguis, and Terrapene carolina major, are recognized based on morphological studies. A fifth subspecies, Terrapene carolina putnami, has been described from Pleistocene deposits but is very similar morphologically to T. c. major. Questions concerning the relationship of the Gulf Coast box turtle (T. c. major) to other box turtles have been pervasive ever since it was described. We used a combined morphological and genetic analysis to address the status of T. c. major and other T. carolina lineages. Terrapene c. bauri, T. c. carolina, and T. c. triunguis are distinct based on a discriminate function analysis of 25 morphological characters, including characters traditionally used to assign subspecies. The results of the present study confirm that box turtles phenotypically diagnosed as T. c. bauri, T. c. carolina, and T. c. triunguis all occur within the hypothesized range of T. c. major, and that the latter does not possess a diagnosable morphology. The three morphological lineages also possess divergent mitochondrial haplotypes that are present within the hypothesized range of T. c. major. In addition, a fourth distinct mtDNA lineage co-occurs within the putative range of T. c. major. This unique lineage may include mitochondrial DNA variation from the Pleistocene T. c. putnami. Analysis of nine nuclear DNA microsatellites revealed no population structure in box turtles currently assigned to T. c. major from the Florida Panhandle, suggesting a complete admixture of lineages in this region. The results of the present study indicate that box turtles traditionally assigned to T. c. major based on phenotype are the result of introgression between eastern extant (predominantly T. c. carolina) and an extinct subspecies, T.

Phylogeographic analyses of the southern leopard frog: the impact of geography and climate on the distribution of genetic lineages vs. subspecies

“…2006; Lemmon et al. 2007; Beamer & Lamb 2008; Jackson & Austin 2010; Butler et al. 2011), other studies of southeastern taxa show phylogenetic breaks that are not specific to the suture zone (e.g.…”

“…Although this pattern is common among some North American taxa (e.g. Austin et al 2002;Masta et al 2002;Church et al 2003;Lemmon et al 2007;Beamer & Lamb 2008;Jackson & Austin 2010;Butler et al 2011), other studies of southeastern taxa show phylogenetic breaks that are not specific to the suture zone (e.g. Avise et al 1979Avise et al , 1987Donovan et al 2000;Zamudio & Savage 2003;Gamble et al 2008;Degner et al 2010;Makowsky et al 2010).…”

<strong>On the reclassification of Box Turtles (<em>Terrapene</em>): A response to Martin <em>et al.</em> (2014)</strong>

“…However, for the rapidly evolving D-loop Butler et al (2011) found T. carolina not to be paraphyletic with respect to T. coahuila. Yet, they reported nearly identical haplotypes of a T. carolina from Escambia County (Florida) and T. coahuila, but these haplotypes were highly distinct from all other haplotypes of T. carolina.…”

“…However, if 14% are hybrids, 86% of the concerned populations must be pure, and this is evidence for the maintenance of largely distinct gene pools in sympatry. By contrast, T. carolina forms panmictic populations whenever different subspecies meet (Butler et al 2011), and this is fundamentally different from the situation in T. carolina and T. ornata. Consequently, the recognition of T. mexicana as a species distinct from T. carolina, as proposed by Martin et al (2013Martin et al ( , 2014, is unwarranted.…”

“…The region of sympatry for T. carolina and T. ornata corresponds to a vast area including the eastern parts of Texas, Oklahoma and Kansas, most of Missouri, western Arkansas, western Louisiana, and parts of Illinois and Indiana (Ernst & Lovich 2009). In contrast, the genetic and morphological distinctness of the subspecies of T. carolina, including T. c. triunguis, dissolves in their contact zone (Butler et al 2011), which constitutes a true genetic melting pot. As a legacy of the distinct subspecies, only the deeply divergent mtDNA lineages remain, which are inherited through the maternal line (without recombination).…”