An Application of Randomization for Detecting Evidence of Thermoregulation in Timber Rattlesnakes (Crotalus horridus) from Northwest Arkansas

Most reptiles maintain their body temperatures within normal functional ranges through behavioral thermoregulation. Under some circumstances, thermoregulation may be a time-consuming activity, and thermoregulatory needs may impose significant constraints on the activities of ectotherms. A necessary (but not sufficient) condition for demonstrating thermoregulation is a difference between observed body temperature distributions and available operative temperature distributions. We examined operative and body temperature distributions of the timber rattlesnake (Crotalus horridus) for evidence of thermoregulation. Specifically, we compared the distribution of available operative temperatures in the environment to snake body temperatures during August and September. Operative temperatures were measured using 48 physical models that were randomly deployed in the environment and connected to a Campbell CR-21X data logger. Body temperatures (n=1,803) were recorded from 12 radiotagged snakes using temperature-sensitive telemetry. Separate randomization tests were conducted for each hour of day within each month. Actual body temperature distributions differed significantly from operative temperature distributions at most time points considered. Thus, C. horridus exhibits a necessary (but not sufficient) condition for demonstrating thermoregulation. However, unlike some desert ectotherms, we found no compelling evidence for thermal constraints on surface activity. Randomization may prove to be a powerful technique for drawing inferences about thermoregulation without reliance on studies of laboratory thermal preference.