Allegheny Mountain dusky salamander COSEWIC assessment and status report: chapter 5

Update
COSEWIC Status Report
on the
Allegheny Mountain Dusky Salamander
Desmognathus ochrophaeus

Great Lakes/St. Lawrence Population
Carolinian Population

in Canada

2007

Species Information

Name and Classification

The Allegheny Mountain Dusky Salamander, Desmognathus ochrophaeus, is a member of the Family Plethodontidae (lungless salamanders) (Order Urodela) (Fig. 1). This is the only Canadian species within the D. ochrophaeus complex, which consists of six morphologically and ecologically similar species: D. apalachicolae, D. carolinensis, D. imitator, D. ochrophaeus, D. ocoee, and D. orestes (Tilley and Mahoney 1996; Petranka 1998; Anderson and Tilley 2003). These species are all very closely related and are best identified by geographic range and genetic comparison. Desmognathus ochrophaeus is the least variable member of the complex with respect to genetic and morphological features, and is believed to have been derived from a southern form that underwent a rapid northern range expansion following the Pleistocene glacial period (Petranka 1998).


Figure 1: Desmognathus ochrophaeus, (A) Individual from the Niagara Population (B) Individual from the Covey Hill Population

Figure 1: Desmognathus ochrophaeus, A. Individual from the Niagara population. Photo by Rob Tervo. B. Individual from the Covey Hill population. Photo by David M. Green.

(A) Photo by Rob Tervo. (B) Photo by David M. Green.


Morphological Description

Desmognathusochrophaeusis a small, slender salamander with 14 costal grooves (Fig. 1). Mature D. ochrophaeus average 37 mm snout-vent length (svl) (range: 29-45 mm), or 70 to 100 mm in total length (Orr 1989; Conant and Collins 1998; Petranka 1998). Males are 6-20% longer in snout-vent length than females, and have well-developed premaxillary teeth, larger jaw muscles, smaller mental glands, and a mouth line that appears sinuate when viewed from the side (Orr 1989; Bruce 1993).

Over its global range (Fig. 2), this species exhibits a high degree of morphological variability (Fitzpatrick 1973; Petranka 1998). Mature individuals usually have a light mid-dorsal stripe that extends from the head to the tail. This stripe is straight-edged, and its colour varies from gray, brown, tan, yellow, orange, to red, depending on age and sex. Juveniles and young adults typically have yellowish or reddish dorsal stripes. Older individuals tend to become melanistic with age and the dorsal stripe and chevron pattern become less apparent. Commonly the stripe has a mid-dorsal row of chevron-shaped dark spots. Dark pigment borders the stripe and extends laterally and ventrally to the mottled lower sides. The belly becomes light grey to grayish black with age (Petranka 1998). Larvae have prominent yolk reserves in the belly region, well-developed gills, and a well-developed fin on the dorsal and ventral surfaces of the tail (Organ 1961).


Figure 2: Global Distribution of Desmognathus ochrophaeus

Figure 2: Global Distribution of Desmognathus ochrophaeus.

Arrows indicate Canadian populations at Covey Hill in Quebec and Niagara in Ontario.

Individuals with a very light dorsal band are sometimes confused with the Northern Two-lined Salamander (Eurycea bislineata), but that species' compressed tail and yellow underparts should distinguish it. Occasionally a specimen is found that has a striking superficial resemblance to some colour phase of the Red-backed Salamander (Plethodon cinereus), but the comparatively larger hind legs and the light bar from the eyes to the angle of the jaw in D. ochrophaeus distinguish it, even in juveniles (Bishop 1941; Orr 1989).

Desmognathus ochrophaeus is most easily confused with the Northern Dusky Salamander (Desmognathus fuscus), which is a more stocky, less terrestrial species often found in similar habitats (Orr 1989). Desmognathus ochrophaeus is a montane species whose global range is contained within the range of D. fuscus, which is primarily a lower elevation form (Houck and Francillon-Vieillot 1988). Both species are commonly found in the same type of habitat, and the overlapping ranges of these two species, their high degree of intraspecific variation, and their propensity for hybridization (Karlin and Guttman 1981; Sharbel et al. 1995), have presented difficulties for field identification in regions of sympatry. Molecular analysis has therefore become necessary for the identification of species, especially within hybrid zones (Sharbel and Bonin 1992; Tilley and Mahoney 1996; Tilley 1997). Nevertheless, morphological characters can often be used with some success. Both species have an eye stripe extending from the eye to the angle of the jaws, a characteristic of all the species in the genus (Orr 1989). Desmognathus ochrophaeus, however, usually has a round slender tail, whereas that of D. fuscus is larger in diameter at the base and is keeled with a sharp crest along the dorsal edge (Bider and Matte 1996). Both species have a mid-dorsal stripe, but in D. fuscus the stripe usually has irregular edges rather than regular ones and D. fuscus lacks the chevron-shaped markings on the dorsum. Larvae of D. ochrophaeus have a broad, light, dorsal band that is entirely lacking in D. fuscus (Bishop 1941; Logier 1952).


Genetic Description

Both Canadian populations of D. ochrophaeus are found adjacent to the Canadian/U.S. border, but are geographically isolated by distance or physical barriers from neighbouring populations in the United States. It is not known how long populations on either side of the border have been separated.

The Covey Hill population in Quebec is found approximately 90 km north of the nearest population in upper New York State. Sharbel and Bonin (1992) used allozyme electrophoresis to identify D. ochrophaeus and found low levels of hybridization between D. ochrophaeus and the closely related D. fuscus in southern Quebec. Hybrids can be distinguished from pure parent D. ochrophaeus or D. fuscus through molecular analysis and occasionally morphological features. Their mix of identifying characteristics, however, can make them difficult to identify in the field (Bonin 1993; Sharbel et al. 1995; Boutin 2006). A genetic study of the population by Boutin (2006) using a 500 base pair fragment of the 12S rRNA mitochondria gene, nuclear (RAG-1 region) DNA, and restriction enzymes also identified low levels of hybridization. Backcrossing of hybrids, where hybrids mate with a pure parental type, has further complicated morphological identifications, and considerable backcrossing with D. ochrophaeus is apparent in most suspected hybrids (Sharbel et. al. 1995; Boutin 2006). While hybridization between pure forms of the two species remains infrequent, it is unknown what percentage of seemingly “pure” D. ochrophaeus contain D. fuscus genes. The Quebec population is believed to exist as a relatively small, genetically isolated, inbred unit with a small percentage occasionally interbreeding with D. fuscus (Sharbel et al. 1995). No genetic comparisons have been made between the Quebec population and populations south of the border, so levels of genetic differentiation are unknown.

In Ontario, Canadian populations of D. ochrophaeus are separated from neighbouring American populations by the Niagara Gorge and a small geographic distance. Rivers are known to impede salamander dispersal (Howard et al. 1983; Bonnett and Chippindale 2004), and the fast-flowing Niagara River that runs through the gorge is likely a significant barrier to gene flow. No populations are known from within the Niagara gorge on the New York side. The closest known population is located 22 kilometres east of the Canadian Niagara population along the Niagara Escarpment in New York State.

In the fall of 2005, a molecular study was undertaken to compare samples in the Niagara Gorge of Ontario with six neighbouring populations from across western New York State (Markle and Green 2006). The objective of the study was to determine if the Niagara population was genetically distinct from neighbouring U.S. populations. The analysis compared variable regions of two mitochondrial genes (cyt b and 12S rRNA) for a total sequence length of 954 base pairs. Results of the analysis did not reveal any genetic variation between any of the populations. Other studies based on allozymes have also found low genetic variation between populations of D. ochrophaeus, even over distances of 1000 km (Tilley and Mahoney 1996; Tilley 1997).

A total of 22 individuals from the Niagara site have now been identified through molecular analysis as D. ochrophaeus, suggesting that the entire population is D. ochrophaeus. So far, no variation has been found within the Ontario population (using mitochondrial markers cyt b and 12S rRNA) (Markle 2006 unpubl. data).

Page details

Date modified: