The etiology and pathophysiology of seasonal affective disorder (SAD) has been linked to the seasons and to light since its first conceptualization. Aspects of SAD that make it particularly amenable to biological investigation include the predictable recurrent episodes, the rapid response to a nonpharmacologic treatment, the specific neurovegetative features, and the availability of rich animal models of seasonality. This paper reviews new findings for the major biological hypotheses for SAD, focusing on circadian rhythms, neurotransmitters, and molecular genetics. Integrative issues and future directions for the study of SAD, including the heuristic value of a dual-vulnerability hypothesis that conceptualizes seasonality as a dimensional construct and the importance of studying endophenotypes, will be discussed.