We studied survival of radio-marked western gray squirrels (Sciurus griseus) and quantified causes of mortality including incidence and severity of notoedric mange in south-central Washington, 1998–2005. We used known-fate models in Program MARK to explore alternative hypotheses on factors related to survival and correlation analysis to investigate parameters that might be related to incidence of mange. The best-supported models indicated that survival varied by year and by sex and that survival of males was lower during the breeding season compared to the non-breeding season. We found little support for differences in survival between juvenile (5–12 months old) and adult squirrels, or for winter severity or size of the acorn crop as significant influences on survival. We determined the likely proximate cause of death for 81 animals; 63% were killed by predators and 37% succumbed to disease, with most disease deaths attributed to mange. Mange was documented in the population during all years and occurred more frequently in animals captured in spring than in animals captured in fall. Counter to our predictions, occurrence of mange was not correlated with 2 measures of winter severity but was strongly correlated with mildness of the preceding winter (number of days with mean air temperature ≥0° C). Sequential use of nests by individual squirrels during mild winters with temperatures conducive to survival of ephemeral, free-living mites may partially explain the periodic epizootics of notoedric mange in this western gray squirrel population. Continued deterioration of squirrel habitat through fragmentation will place additional stressors on the population and may compound the effects of mange on this threatened species. © 2013 The Wildlife Society.