Five-lined skink (Eumeces fasciatus) COSEWIC assessment and status report: chapter 3

Species Information

Name and classification

Eastern Canada’s only lizard, the five-lined skink (Eumeces fasciatus), belongs to the family Scincidae and is known as le scinque pentaligne in French. The Family Scincidae is the largest family of squamate reptiles and is distributed worldwide (Pough et al., 2004). The genus Eumeces is also broadly distributed across North and Central America, North and Southeast Asia, and North Africa (Fitch, 1954). The first comprehensive examination of the genus revealed no less than fifty recognized species that share a relatively conserved morphology (Taylor, 1936). Recent genetic work suggested that Eumeces is not a monophyletic group and that the genus should be split into multiple genera. Taxonmists have recommended that all Eumeces species in North America adopt the genus name Plestiodon (Schmitz et al., 2004; Brandley et al., 2005). Following Crother et al. (2000), the genus name Eumeces will be retained for this report.

Within the genus Eumeces, several species groups have been identified including the fasciatus group that has three North American representatives (E. fasciatus, E. laticeps and E. inexpectatus). Allozyme data suggested that these three species are each other’s closest relatives (Murphy et al., 1983), but more recent genetic evidence based on multiple mitochondrial genes suggests the fasciatus species group is not monophyletic (Schmitz et al., 2004; Richmond and Reeder, 2002). Based on this work, the sister species of E. fasciatus is E. septentrionalis. Eumeces septentrionalis occurs in Canada but is limited to two discrete areas in southwestern Manitoba,and the species is listed federally as Endangered (Committee on the Status of Endangered Wildlife in Canada (COSEWIC), 2004a). A third species of Eumeces(E. skiltonianus) has its Canadian range limited to southcentral British Columbia. Eumeces skiltonianus is listed federally as Special Concern (COSEWIC, 2002).

Despite the expansive geographic range of E. fasciatusand the variety of habitat and environmental conditions it occupies, there are no recognized subspecies. Wherever possible, this report supplements information gathered from throughout the species’ range with information specific to populations within its Canadian range. The term “population” is used throughout this report and is generally defined as a group of individuals within 1-2 km (e.g. Fitch, 1954; Seburn, 1993; Hecnar 1994; Hecnar and M’Closkey, 1998; Howes et al., 2006). This designation is consistent with the definition of an “Element Occurrence” employed by the Natural Heritage Information Centre (NHIC, 2006).

Morphological description

Newly hatched E. fasciatus are approximately 25 mm in snout-vent length (SVL) and have five cream stripes on their green-black bodies. The species’ characteristic bright blue tail is most obvious in hatchlings and juveniles. Body and tail colouration fades with age to become a solid bronze in both sexes, although females generally retain more of the juvenile colouration pattern than males. During breeding season, adult males develop bright orange colouration around the jaws and chin, and very large females may show some pink colouration around the chin. The scales are unkeeled, giving individuals a smooth, shiny appearance and perhaps explaining why the species is often misidentified as a salamander by the general public (Fitch, 1954; B. Howes, pers. obs.).

Individuals reach a maximum of approximately 86 mm in SVL and have a wedge-shaped head, and a slender, elongated body ending with a tail that can be autotomized and regenerated. Their laterally flattened bodies and moderately developed limbs make them adept burrowers, and enable them to find refuge under a variety of cover objects while their well developed toes and strong claws provide them with agility over a variety of substrates (Fitch, 1954).

Previous research suggested that males and females have similar body sizes (Fitch, 1954; Vitt and Cooper, 1986a; Seburn, 1990), and that sexual dimorphism is apparent in head size only, with males having larger heads than females (Vitt and Cooper, 1986a; Seburn, 1990). Males probably have larger relative head sizes than females because of sexual selection rather than resource partitioning (Vitt and Cooper, 1986a). Both males and females are capable of eating larger prey than they typically ingest based on their gape limit (Vitt and Cooper, 1986a), and males exhibit intrasexual aggression (e.g. Fitch, 1954; Cooper and Vitt, 1987).

Recent work indicates that some populations differ in the degree of sexual size dimorphism in head size as well as body size. Morphological data collected from populations throughout the species’ range showed that male-biased sexual size dimorphism in SVL significantly increased with latitude (Howes and Lougheed, unpublished data). Mean male SVL for individuals from nine Ontario populations (75.2 ± 5.3 mm, n=50) was significantly greater (P<0.0001) than mean female SVL for individuals from nine Ontario populations (70.5 ±5.8 mm, n=97).

Genetic description

Range-wide

A recent phylogeographic study spanning the entire range of E. fasciatus revealed six major mitochondrial lineages within the species. Similar to other eastern North American herpetofauna, E. fasciatus is structured in a manner that reflects divergence from east to west (longitudinal phylogeographic structure). Phylogeographic patterns are consistent with fragmentation due to refugial and post-glacial dynamics, but deep divergences among some lineages imply historical fragmentation that predates the Pleistocene (Howes et al., 2006, see Figure 1).

The species has three broadly distributed (East, Central, and West), and three geographically restricted lineages (Carolinas, Oklahoma, and Wisconsin). The most broadly distributed is the East lineage. It spans from the Mississippi River east to the Atlantic Ocean, and includes all Ontario populations. The West lineage extends from the Mississippi River west to Texas and Minnesota, while the narrowly distributed Central lineage includes populations in northeast Texas, southeast Louisiana, northwest Mississippi, and a population in central Wisconsin (Figure 1).

The Carolinas lineage consists of two populations in the Atlantic coastal plain, and the Oklahoma lineage consists of one population located at the extreme western periphery of the species range. The Wisconsin lineage consists of one disjunct population (Figure 1).

Figure 1. Distribution and mitochondrial lineage groupings of Eumeces fasciatus (range distribution based on Conant and Collins, 1998). States and provinces are indicated by abbreviations and sampling sites are marked with circles. Species’ range borders are marked with thick lines and include three disjunct series of populations (MN, WI, and IA). Based on analysis of 769 bp (base pairs) of the mitochondrial DNA and include three main lineages (East, Central, West) and three geographically isolated lineages (Carolinas, Oklahoma, Wisconsin). A simplified phylogeny from Howes et al. (2006) in the lower right of this figure shows the relationships among these different lineages. Adapted from Howes et al. (2006).

Figure 1.Distribution and mitochondrial lineage groupings of Eumeces fasciatus (range distribution based on Conant and Collins, 1998).

Genetic characteristics of Ontariopopulations

Ontario populations of E. fasciatus may have inherent genetic characteristics that make them more at risk of local extinction because they are peripheral populations located at the northern margin of the species’ range. For instance, northern peripheral populations of a variety of species may carry relatively low levels of genetic diversity possibly due to rapid post-glacial expansion and founder effect (see Hewitt, 1996). Peripheral populations may also be smaller and more genetically isolated relative to more central populations (abundant centre hypothesis; e.g. Brown, 1984).

Howes and Lougheed (in review) found that northern (including Ontario populations) and western peripheral populations had significantly lower intra-population genetic diversity relative to central populations and peripheral populations to the east and south (bordered by the Atlantic Ocean) (Figure 2). This lower genetic diversity can increase the level of homozygosity of a population and may result in reduced individual fitness (e.g. Shaffer, 1981; Milligan et al. 1994; Lande and Shannon, 1996). Reduced intra-population genetic diversity may also limit the potential for populations to adapt to future changes (e.g. climate change, novel parasites or disease), as a population’s evolutionary potential is proportional to its additive genetic variance (Fisher, 1958). Genetic variation can be a major determinant in the long-term persistence of a population, and Ontario’s populations of E. fasciatus may ultimately face a greater risk of local extinction relative to more southern populations. It should be noted that this finding is based on neutral genetic markers (microsatellite loci), and that it is assumed these neutral genetic markers reflect total genomic variability in E. fasciatus.

Figure 2. The relation between a population’s location within the species’ range and its intra-population genetic diversity (Average allelic richness) based on six microsatellite loci for Eumeces fasciatus. Eastern, western, northern, and southern populations are located within 200 km of the species’ range border and are defined according to their most proximate border. Central populations are defined as any population occurring more than 200 km within the species’ range border. Mean diamonds are drawn for each group, where the vertical span of the diamond represents the 95% confidence interval, and the middle line represents group mean. The horizontal line indicates the grand mean for all groups. From Howes and Lougheed (in review).

Figure 2. The relation between a population’s location within the species’ range and its intra-population genetic diversity (Average allelic richness) based on six microsatellite loci for Eumeces fasciatus.

Although northern populations (e.g. Ontario’s populations) have lower intra-population genetic diversity, estimates of their effective population size (Ne) were not significantly reduced compared to other populations across the range (Howes and Lougheed, unpublished data). Reduced population size could negatively impact intra-population genetic diversity by promoting inbreeding, and increase the impact of genetic drift. It is often assumed that peripheral populations are smaller (and therefore have smaller effective population sizes) than more geographically central populations. The mean Nein nine surveyed Ontario populations (273) was lower than the mean for 21 other populations across the range (339), but this difference was not significant (Howes and Lougheed, unpublished data).

Northern and western populations were more genetically differentiated from each other across all distances than were central and eastern populations (Figure 3). The degree of genetic differentiation between population pairs can be used as a surrogate measure for the degree of genetic isolation of a population. Mean values of FST (a standard measure of pairwise population differentiation) for northern and western populations (0.18 and 0.21 respectively) were significantly less than those of central and eastern populations (0.069 and 0.037 respectively). These low values imply that genetic connectivity among Ontario’s populations is reduced relative to levels elsewhere in the range, which in turn indicates that the likelihood of natural recolonization following a local extinction event would be low in Ontario populations (Howes and Lougheed, unpublished data).

Figure 3. Comparison of the relationship between population pairwise genetic distance (based on FST) and geographic distance among north, west, east and central populations in Eumeces fasciatus based on six microsatellite loci. Northern populations (blue) and western populations (green) show significant isolation by distance (p=0.001, n=67 and p=0.021, n=7 respectively). Central populations (orange) and eastern populations (red) do not show significant isolation by distance.

Figure 3. Comparison of the relationship between population pairwise genetic distance (based on FST) and geographic distance among north, west, east and central populations in Eumeces fasciatus based on six microsatellite loci.

In summary, it is often assumed that peripheral populations possess a variety of genetic characteristics that may threaten their persistence. In this light, Ontario’s northern peripheral populations of E. fasciatus do not appear to be at risk because of low effective population size. However, results do suggest that they have reduced intra-population genetic diversity (Howes and Lougheed, unpublished data) and that they are more genetically isolated relative to more southern populations (Howes and Lougheed, unpublished data). This reduction in genetic diversity and increase in genetic isolation could elevate the likelihood of local extinction in Ontario populations.

Designatable units

Canadian populations of Eumeces fasciatus can be split into two designatable units based on genetic evidence, range disjunction, and biogeographic distinction. The first unit is distributed along the southern Canadian Shield in Ontario (hereafter called Great Lakes/St. Lawrence population) and the second is found within the Carolinian region of southwestern Ontario (hereafter called Carolinian population)

Genetic evidence

Although Ontario’s Carolinian and Great Lakes/St. Lawrence populations belong to the same mitochondrial lineage, they show considerable genetic divergence based on more rapidly evolving microsatellite markers.

Pairwise genetic differences in allele frequencies among thirty populations from across the species’ range were estimated using Nei’s standard genetic distance (Nei, 1978). An unrooted neighbour-joining tree was constructed based on these pairwise genetic distances among populations and support for each cluster of the tree was based on bootstrapping genotypes among populations and is indicated as a percent (Figure 4). Great Lakes/St. Lawrence populations (n=7) form an exclusive cluster, whereas Carolinian populations (n=2) form another cluster with a population from eastern Michigan (Howes et al., 2006).

Genetic differentiation (based on FST) between all pairs of Ontario populations was calculated to determine if average genetic differentiation between Great Lakes/ St. Lawrence and Carolinian populations exceeded that within each series of populations. FST (Wright, 1969) is a standard measure of genetic differentiation between two populations and values can range from 0 (no genetic differentiation) to 1 (complete genetic differentiation). Mean genetic differentiation within Great Lakes/St. Lawrence populations was 0.10 (n=21), and the genetic differentiation between the only pair of Carolinian populations was also 0.10 (Howes and Lougheed, unpublished data). In contrast, the mean genetic differentiation between Great Lakes/St. Lawrence and Carolinian populations was higher at 0.15 (n=14). All pairwise comparisons were highly significant (Howes and Lougheed, unpublished data), suggesting that Great Lakes/ St. Lawrence and Carolinian populations in Ontario show highly significant genetic isolation from each other.

Figure 4. Neighbour-joining dendrogram based on Nei’s (1978) genetic distance among populations of Eumeces fasciatus as determined by six microsatellite loci. Bootstrap values (>50%) from 1,000 replicates are shown. The state (U.S.A.) or county (Ontario) where each population was sampled is indicated.

Figure 4. Neighbour-joining dendrogram based on Nei’s (1978) genetic distance among populations of Eumeces fasciatus as determined by six microsatellite loci.

Range disjunction

The nearest Great Lakes/St. Lawrence and Carolinian populations are separated by approximately 250 km. Perhaps more importantly, the two series of Ontario populations are separated by Canada’s most dense urban area (Greater Toronto Area) and expansive agricultural land, rendering exchange of individuals and genetic material between the two regions virtually impossible (Figures 5, figure6).

Biogeographic distinction

Great Lakes/St. Lawrence and Carolinian populations are biogeographically distinct based on the map of faunal provinces of terrestrial amphibians, reptiles, and molluscs in Canada (Figure 5; COSEWIC, 2004b). Shield populations occur in the northern portion of the Great Lakes /St. Lawrence faunal province (7), whereas southwestern Ontario populations occur in the Carolinian faunal province (8) (Figure 5).

Figure 5. Map of the faunal provinces of terrestrial amphibians, reptiles, and molluscs in Canada (COSEWIC, 2004). The distribution of each province is indicated by a unique pattern that corresponds to the legend in the upper right. Populations of Eumeces fasciatus on the Canadian Shield occur in the northern portion of the Great Lakes/St. Lawrence faunal province (7), whereas southwestern Ontario populations occur in the Carolinian faunal province (8).

Figure 5. Map of the faunal provinces of terrestrial amphibians, reptiles, and molluscs in Canada(COSEWIC, 2004).

Page details

Date modified: