Revisiting Thermal Gradient Experiments: Effects of Thermal Heterogeneity on Salamander Behavior.

Thermal gradient experiments are commonly used in studies of ectothermic organisms for a variety of scientific inquiries. Such gradient experiments, performed in the laboratory, are often used to infer the climatic preferences of animals in the absence of other variables. However, the ability to extrapolate laboratory results to the field is only as good as the accumulation of ecological data for that organism. When the variable quantified is interpreted as thermal "preference," there are some assumptions that come with it, namely that the organism selects a particular preferred temperature by positive thermotaxis. Amphibians, as well as most ectotherms, tend to be thermoconformers, so conclusions from thermal gradient experiments carry different meanings than they do for organisms such as heliothermic ectotherms that maintain a narrow range of body temperatures in the lab and field. We tested whether and how the Eastern Red-backed Salamander (Plethodon cinereus) behaves when presented with a heterothermal gradient arena in comparison to a control (homothermal) arena. Salamanders in the control arena unambiguously moved toward either end of the arena, despite no variation in temperature being available. We found that salamanders did respond to a thermal gradient, but that their thermoregulatory behavior was limited to the avoidance of the hottest end of the gradient, and not a positive thermotaxis toward a specific temperature as assumed of a thermal "preference." Our results encourage a broader consideration of how laboratory-measured behaviors relate to the predicted behaviors of organisms in natural settings, and a re-evaluation of the terminology used to describe movement behaviors in thermal gradients.