Extending Janzen's hypothesis to temperate regions: A test using subterranean ecosystems

Janzen's hypothesis (1967; American Naturalist) predicts that tropical habitats with reduced thermal seasonality would select for species with narrow thermal tolerance, thereby limiting dispersal among sites of different elevations showing little overlap in temperature. These predictions have so far been tested by confronting tropical and temperate mountain communities, leaving unresolved the question of their generalization to habitats with low thermal seasonality outside the tropics. We provide the first extension of Janzen's hypothesis to temperate habitats, by testing for differences in thermal tolerance and elevational range among congeneric alpine spiders (Araneae: Linyphiidae: Troglohyphantes) occurring along a steep gradient of decreasing thermal seasonality with increasing cave depth. Using species from the same temperate region rather than from distinct biogeographic regions avoids confounding the effects of short- and long-term climatic variability on thermal tolerance and elevational range extent. Following Jansen's assumptions, we predicted that cave habitats with low thermal seasonality would select for narrow thermal tolerance. Also, specialized subterranean species would exhibit both narrower elevational range extents and smaller realized thermal niche breadths. Initially, we showed that thermal seasonality and the overlap in temperature across caves were considerably lower in deep than in shallow cave habitats. Then, we measured thermal tolerance and used morphological traits to quantify the degree of specialization to subterranean life of 11 spider species. We found that thermal tolerance decreased with increasing subterranean specialization. Deep subterranean species reached their critical temperature at 1-4 degrees C above their habitat temperature, whereas shallow subterranean species withstood a twofold larger temperature increase. At last, we demonstrated that a species' elevational range extent and the variation of temperature encountered across its range decreased with increasing specialization to deep subterranean life. Our integrative work, being grounded in organismal and habitat measures, represents the first generalization of Janzen's framework to caves and provides a conceptual framework to disentangle the effect of long-term climate variability on subterranean biodiversity patterns. Extending Janzen's thoughts to a broader range of ecosystems is key to understanding how the ecological specialization-dispersal trade-off may constrain the response of species to climate change. A free Plain Language Summary can be found within the Supporting Information of this article.