Arif et al. 2007

Arif S, Adams DC, Wicknick JA. 2007. Bioclimatic modelling, morphology, and behaviour reveal alternative mechanisms regulating the distributions of two parapatric salamander species. Evolutionary Ecology Research 9: 843-854.

These authors examined the factors responsible for shaping the geographic ranges of two species of salamanders in the genus Plethodon. One species is broadly distributed in eastern North America (including Guelph), the other occurs only on a group of mountain peaks in western Virginia. The primary working hypothesis was that the more restricted species is so restricted partly by competitive interactions with its congener, that is it cannot expand to lower elevations because it cannot outcompete the other, and is persists in its home range mainly via the congener’s intolerance of local abiotic conditions.

This hypothesis was tested by collecting data on diet, local distribution, head morphology, agonistic behaviour, and climate factors for both species in the region where the very localized species lives. Diet was examined by stomach contents of previously-collected specimens; no difference was found, suggesting these species are not applying pressure to each other by exploitative competition. Local distribution included measurements of parameters in 18 sites, two where both species occur (sympatry), and eight sites for each species where it occurs alone (allopatry). I suppose that taken together, at a larger spatial scale, this site distribution qualifies as parapatric. Head morphology was measured for each site, and analysed by PCA. The two species differ consistently and significantly in head shape, but the sympatric populations are not more or less divergent from each other than expected by chance, implying no character displacement for these traits. Behaviour was measured by a series of encounter trials, in which individual males from allopatric populations were allowed to interact as either “residents” or “intruders”. The highly-restricted species was found, perhaps surprisingly, to behaviourally dominate the more cosmopolitan species, suggesting any competitive exclusion that may be occurring is driven in the opposite direction to the initial prediction.

Climate data came from the WORLDCLIM data set (Hijmans et al. 2005). There was high congruence between the predicted ranges of these species through modelling and the actual observed ranges, with some exceptions – the models predicted some species occurrences where they do not occur, on the other side of obvious geographic barriers such as rivers.

The main conclusion of this paper is that the very restricted range size of one species is apparently constrained by abiotic climate factors, not by competitive exclusion or similar processes. In contrast, the absence of the more cosmopolitan species from these mountain tops is apparently driven more by interspecific interactions than by abiotic factors, a situation precisely opposite that predicted by the initial hypothesis. Additionally, these authors suggest that the differences in morphology may be indicative of local adaptation, but further refinement of those data will be required to test such hypotheses.