Janzen (1967) proposed that broader temperature tolerances could facilitate dispersal across mountain passes and lead to latitudinal differences in the potential for speciation. This elegant hypothesis has since been extended to account for the effects of other environmental barriers to dispersal and suggested as a possible mechanistic explanation for the origin of latitudinal gradients of diversity. Nevertheless, its key prediction –that gene flow across environmental gradients is more likely in species with broader physiological tolerances– has never been tested in a large-scale comparative analysis. Here we test these ideas through a phylogenetically-informed analyses of the potential for speciation in terrestrial mammals (N=3136) and birds (N=6694). Our analyses account for well-known correlates of subspecies richness, including species’ age, area of breeding distributions, adult body size, island dwelling, habitat heterogeneity, historical exposure to glaciation, and migration (birds) / hibernation (mammals).
Results/Conclusions
We show that although environmental tolerances are significant predictors of the potential for phenotypic divergence among populations within a species, the resulting patterns do not always reflect Janzen’s proposed mechanism. Specifically, our results indicate that although a broader tolerance to rainfall unpredictability leads to significantly lower numbers of subspecies (as predicted by Janzen), a greater tolerance to temperature variability has the opposite effect even among species whose ranges are dissected by mountains. We explore how temperate and tropical species vary along these and other ecological axes and show that although individual effects vary in direction, their net combined effect is a higher (in mammals) or similar (in birds) potential for speciation in temperate than tropical taxa. We interpret these findings in light of the current controversy on the effect of latitude on diversification rates, and discuss their implications for our understanding of latitudinal diversity gradients.