Death feigning is a potentially important behaviour used by a wide variety of animals to increase the probability of escape from a would-be predator. Few data are available on the influence of various factors on death feigning in vertebrates, especially ectotherms, because of difficulties in consistently stimulating the behaviour under controlled conditions. I examined the effects of temperature, body size and locomotor performance on death feigning in neonate brown snakes, Storeria dekayi, in the laboratory. Brown snakes consistently feigned death in water, and contrary to predictions, were more likely to feign death and to feign death longer as temperature increased. Q 10 values for death-feigning durations (mean=2.79) were greater than those for maximal swimming velocities (mean=1.77) between 10°C and 20°C. However, no statistical difference was detected between Q 10 values for feigning durations (mean=1.11) and swimming velocities (mean=1.28) between 20°C and 30°C. At 30°C, swimming velocity was negatively correlated with death-feigning duration. Moreover, body size was negatively related to death-feigning duration at 30°C. These results suggest that temperature probably plays a large role in the decision by ectotherms to death feign, and that an animal's locomotor abilities and body size potentially influence the likelihood and duration of death feigning at optimal temperatures. However, physiological constraints greatly reduce the use of death-feigning behaviours at suboptimal temperatures, regardless of locomotor abilities and body size. Therefore, other stationary defensive behaviours are probably more important at suboptimal temperatures.