ABSTRACT.-Several species of migratory songbirds exhibit a distinct form of habitat segregation while on their Neotropical wintering grounds in which males and females occupy different habitat types. In the American Redstart (Setophaga ruticilla), that sexual habitat segregation is a result of behavioral dominance of older males. In that study, we examined whether such dominance behavior and the resulting differential habitat segregation has consequences for the condition or survival of excluded individuals. We quantified the physical condition and survival of redstarts (both males and females) occupying two habitat types that differed in the proportion of males and females present in Jamaica. Both sexes of redstarts occupying female-biased habitat lost significantly more mass over-winter and had lower annual survival and longevity compared to individuals in male-biased habitat. These results suggest that nonbreeding habitats differed in suitability, with the least suitable habitat being occupied predominately by females. Because most female redstarts are forced to over-winter in these kinds of habitats, they may often be in poor physiological condition prior to departing on spring migration for the breeding grounds. This in turn may influence dynamics of the breeding period by determining their condition and perhaps reproductive success. Furthermore, because winter habitat segregation appears to lower female survivorship, it may also limit the number and availability of breeding females. These results implicate events that occur during the nonbreeding period as playing a critical role in the annual dynamics of this migratory species. Received 30 September 1999, accepted 8 August 2000 IDENTIFYING THE FACTORS that limit abundance of organisms is essential to understanding their population dynamics. Unfortunately, this is difficult because of the complex life cycles of most organisms and problems in assessing relative importance of factors such as weather, food, habitat, and intra-and interspecific interactions. The complexities associated with determining limiting processes becomes especially apparent for long-distance migratory birds that range over large geographical areas.Progress toward identifying factors that control population dynamics of migratory birds has been further hindered because most studies have focused on describing patterns (e.g. distribution and habitat associations: see papers in Keast and Morton 1980, Hagan and Johnston 1992, Martin and Finch 1995 rather than understanding population processes. This emphasis is especially true for research on migrant songbirds in their Neotropical wintering