Deer mouse (Peromyscus maniculatus) populations increase dramatically after wildfire. These increases are puzzling because there are no obvious food sources or vegetation cover in severely burned areas. We conducted a capture-mark-recapture study of deer mice in a mosaic of burned and unburned montane forest in western Montana to determine if their postfire increase could be explained by source-sink dynamics, with burned areas acting as a sink. When overall mouse densities were very low, the vast majority of the population was found in burned areas. Mice appeared regularly in unburned forest only when the densities were high. This pattern is precisely opposite to the expected results if the sink hypothesis were correct. Moreover, mice in burned areas did not show decreased body weight, reproductive performance, or survival when compared to mice in unburned areas. Age structure and sex ratio did not differ between burned and unburned sites. We conclude that burned areas are unlikely to function as population sinks; rather, they appeared to represent high-quality habitat for deer mice.