Understanding causes of mortality in a population provides information about the selection pressures acting on individuals in that population. To obtain unbiased estimates of mortality, researchers must track individual animals until they die, retrieve dead animals, and make a forensic estimation to determine the cause. Given the limitations of traditional radio-tracking technology and the biases inherent in other methods of estimating mortality causes, little is known about causes of mortality in small mammals (<1 kg). Here we used an automated radio telemetry system to record activity patterns and detect mortality events in a wild population of Peromyscus leucopus (White-footed Mouse). Mice (n = 32) were fitted with small (1 g) radio-collars. Daily survival rate was 0.98 ± 0.01, which translates to a 50% chance of survival over a 29-day period. Although the species is well known as a vector for various diseases, we found only one death attributable to disease. Most mouse mortality (93%) was caused by predators, primarily Mustela spp. (weasels ). We found a significant relationship between Borrelia infection (the bacterium that causes Lyme disease in humans), and increased mortality (χ² = 6.92, df = 1, P = 0.008); there was no detectable effect of sex on survival. Our results suggest that, to the extent that predation risk is dependent on heritable phenotypes, predation risk is the most important evolutionary force acting on mortality in the population we studied.