Eastern foxsnake (Elaphe gloydi) (Carolinian)COSEWIC assessment and status report: chapter 6

Distribution

Global Range

Foxsnakes are thought to have extended their range eastward along a post-Wisconsin glaciation prairie (steppe) corridor during the warm, arid Xerothermic period approximately 4000–6000 years ago (Schmidt 1938, Conant 1940, Smith 1957). According to Schmidt (1938) the foxsnake is a mid-western endemic, either restricted to, or centred in, the steppe peninsula and not widely distributed toward the southwest. This distribution has been interpreted as a postglacial spread, favoured by the impoverishment of the fauna of the coniferous forest during the glacial retreat (Schmidt 1938). Subsequent invasion of the prairie peninsula by forest resulted in the species being separated into two allopatric groups of populations (Schmidt 1938, Conant 1940, Smith 1957). The Eastern Foxsnake survived as a Xerothermic relict along portions of the Lake Huron-Lake Erie waterway (Schmidt 1938, Conant 1940, Smith 1957).

Elaphe gloydi is found in Ontario, southeastern Michigan, and northern Ohio, USA (Figure 1). In Michigan, Eastern Foxsnakes are reported in Iosco, Macomb, Monroe, Saginaw, St. Clair, and Wayne Counties, and in Ohio they are reported to occur in Erie, Lucas, Ottawa, and Sandusky Counties (NatureServe 2006). Records of Eastern Foxsnakes in Wood County, Ohio were reported by Conant (1938); however, recent data suggest that the species no longer persists there (Harding, 1997).


Canadian Range

The Eastern Foxsnake’s Canadian range lies entirely within Ontario. Much of the insight into the distribution of E. gloydi in Ontario is made possible by the observation records compiled and organized by the Ontario Herpetofaunal Summary (OHS; Oldham and Weller 2000) maintained by Ontario’s Natural Heritage Information Centre (NHIC). Refinements to the species’ distribution were informed by focal studies recently conducted at several locations in Ontario, through discussions amongst members of the Eastern Foxsnake/Eastern Hognosed Snake recovery team, and increased scrutiny of observation records by (Doucette 2005) and (Willson and Rouse 2006).

Within Ontario, the species’ distribution is highly disjunct, occupying three discrete regions (hereafter regional populations) along the Lake Erie-Lake Huron waterway shoreline, including tributaries, and several islands in Lake Erie, Detroit River-Lake St. Clair, and the 30 000 islands of Georgian Bay. The three regional populations from south to north are (1) Essex-Kent, (2) Haldimand-Norfolk, and (3) Georgian Bay Coast (Figures 2, 3). The Extent of Occurrence, calculated as a single Minimum Convex Polygon encompassing all NHIC records is 68 505 km2.


Figure 1: Eastern Foxsnake (Elaphe gloydi) Distribution in North America

Figure 1. Eastern Foxsnake (Elaphe gloydi) distribution in North America

 


Figure 2: Eastern Foxsnake (Elaphe gloydi) Distribution in Ontario, Canada

Figure 2. Eastern Foxsnake (Elaphe gloydi) distribution in Ontario, Canada

 


Figure 3: Eastern Foxsnake (Elaphe gloydi) Observations in Haldimand-Norfolk Region Showing the Northern Limit of Observations (green line), 5 Snakes Observed Per Year (yellow line), Zones of Most Frequent Reports Usually Ending with Death of Observed Foxsnake (purple line)

Figure 3. Eastern Foxsnake (Elaphe gloydi) observations in Haldimand-Norfolk region showing the northern limit of observations, 5 snakes observed per year, zones of most frequent reports usually ending with death of observed foxsnake.

Map courtesy of M. Gartshore.

The northernmost reliable record of E. gloydi comes from a small unnamed island in Georgian Bay, approximately 50 km northwest of Pointe au Baril, Ontario (recorded 17 June 1982; Mills et al. 1983). The farthest west that Eastern Foxsnakes are found in Ontario is Fighting Island in the Detroit River, whereas the tip of the Long Point sandspit, probably represents the species’ easternmost extent. Elaphe gloydi has been reported from Canada’s southernmost point, Middle Island, within the Lake Erie archipelago; however, recent snake surveys (2001–2006) of that island have failed to find foxsnakes (D. Jacobs and Pt Pelee NP unpublished data). The small size of the island (23-ha) may preclude establishment of a resident, self-sustaining population; instead, occasional migrants from Pelee Island (the “Fish Point” sandspit is only 5 km to the north) may result in the periodic observation of foxsnakes on Middle Island.

The Georgian Bay coast regional population occurs within the Ontario Shield Ecozone (Great Lakes/St. Lawrence Forest Region). As illustrated in Figures 2 and 4, the species’ distribution in this region is tightly coupled to the shoreline of Georgian Bay. In fact, except for the southernmost extent of the foxsnake’s distribution in this region, the species occurs predominantly within a 1-km band of the shoreline (mainland and islands). The southern extent of the Georgian Bay Coast population is ≈225 km from the bounds of the central population, Haldimand-Norfolk.

The other two regional populations occur in the Mixedwood Plains Ecozone (Deciduous Forest Region) of SW Ontario. Despite this commonality, their closest extents are ≈88 km apart (Figures 2, 5). The gap between these two populations is seemingly real because it is unlikely to be the result of inadequate search coverage in that area--particularly because there is no shortage of roads close to the shoreline of Lake Erie (Figures 3, 4, 6). Furthermore, although small, isolated populations could exist within this distribution gap, it is highly unlikely that they would be numerous enough to function as a continuous band. Therefore, treating the Haldimand-Norfolk and Essex-Kent populations separately seems prudent. Within the Essex-Kent regional population, further subdivision and delineation of sites or populations may be beneficial because of the considerable barriers to movement (e.g., roads) that almost certainly prevent movement of individuals between populations in some cases (See also Genetics section). Regardless of whether all the sites are completely isolated from their nearest neighbour, their tentative delineation is simply helpful to visualize the current state of knowledge and possibly to define conservation or management units. Despite its grouping with the Essex-Kent regional population, the foxsnake populations of the Walpole Island wetland complex are technically within the bounds of Lambton County.


Figure 4: Locations (NHIC see text) of Eastern Foxsnakesin the Great Lakes/St. Lawrence populations showing those recorded from 1998-present (black dots), those recorded from 1995-present (black + gray dots) and those recorded from 1984-present (all dots)

Figure 4. Locations of Eastern Foxsnakesin the Great Lakes/St. Lawrence populations showing those recorded from 1998-present, those recorded from 1995-present and those recorded from 1984-present.

 


Figure 5: Locations (NHIC-see text) of Eastern Foxsnakes in the Carolinian Faunal Province Showing those Recorded from 1998-present (black dots), those from 1993-present (grey +black dots), and those from 1984-present (all dots)

Figure 5. Locations of Eastern Foxsnakes in the Carolinian Faunal Province showing those recorded from 1998-present, those from 1993-present, and those from 1984-present.

Further details are in Technical Summary.

Although historic records from the southwestern shoreline of Lake Ontario exist (i.e., Hamilton-Wentworth and Niagara Regional municipalities), Lamond (1994) concluded that the records were likely sightings and/or captures of released or escaped individuals. The lack of evidence suggesting viable populations in that area since 1994 supports that assertion. Several other Ontario records from northern Lambton, Middlesex, Huron, and Bruce Counties are suspect, and are possibly released/escaped captives, but more probably, are inaccurate identifications of Milksnakes (Lampropeltis triangulum) as Eastern Foxsnakes. Milksnakes are often mistaken as foxsnakes and vice versa, but the former is more widespread and common. For example, most reports of foxsnakes from Middlesex, including some with photographs, have been confirmed to be milksnakes (S Gillingwater, pers. comm.). Interestingly, specimens deemed to be of local origin were apparently collected from southern Bruce County in the early 1900s (Logier and Toner 1961).

Areas of Occupancy of the two DUs were calculated as follows. Observational records were available from the past 50+ years, but there were few before 1984, the year the Ontario Herpetofaunal Survey was begun in 1984. Therefore, only records from 1984 to the present were used. All NHIC records from 1984 onward were scrutinized and retained or rejected according to utility (some were rejected because location could not be verified or there were errors in identification, location etc.) (Wilson and Rouse 2006). On inspection, it was clear that many of the older records were no longer viable, particularly in the Carolinian populations, because habitat had been destroyed. Therefore, for the Carolinian populations, it was decided that the data would be considered on a temporal basis such that Area of Occupancy (AO) would be calculated every 5 years (roughly the generation length of this species). These dates were 1984, 1988, 1993, and 1998. For the Great Lakes/St. Lawrence populations, the AO was only calculated for all observations from 1984, and 1998, onward. The 1998 value was considered the current AO. In all cases, the AO was calculated as the sum of a 2x2 km2 grid overlain on all locations. As a result, the AO of the Carolinian region was 188km2 and, by remarkable coincidence, the AO of the Great Lakes/St. Lawrence was also 188 km2. (See Technical Summaries for all computed values.) It is of interest that the AO values declined from 1984 to 1998, especially in the Carolinian region, indicating that both DUs are declining (see Figure 4, 5).

Page details

Date modified: