This research examined the extent to which religiosity was predictive of level of depression, even after controlling for race, gender, social support, income sufficiency, and physical health. Data were collected using in-home interviews conducted from 1999 to 2001 with 1,000 adults age 65 to 106. Subjects were recruited from a stratified, random sample of Medicare beneficiaries age 65 and older in five central Alabama counties (three rural and two urban). The sample was stratified by county, race, and sex and included balanced numbers of African American males and females and White males and females. Highly religious persons had lower levels of depression, even when controlling for other known covariates, β = -.16, t(972) p < .001. Females reported higher levels of depression, β = .07, t(972) p < .05. Although race was unrelated to depression in the model including gender and religiosity only, African Americans reported fewer symptoms of depression than did Whites when social support, income sufficiency, and physical health were added to the model, β= -.08, t(972) p < .01. Results suggest the importance of health and social service professionals' drawing upon older adults' positive spirituality in professional interventions to prevent and treat depression. © 2004 by The Haworth Press, Inc. All rights reserved.