Loneliness is a mechanism through which marital quality relates to older adults’ mental health. Links between marital quality, loneliness, and depressive symptoms, however, are often examined independent of older adults’ functional health. The current study therefore examines whether associations between marital quality, loneliness, and depressive symptoms are contextually dependent on individuals’ own (or their spouse’s) functional limitations, as well as on gender. Data came from couples (N = 1084) who participated in the Health and Retirement Study (HRS), a nationally representative dataset of older adults (age 50+). We utilized data from the 2014 leave-behind psychosocial questionnaire to measure spousal support/strain and loneliness, and interview data from 2014 to measure baseline depressive symptoms and demographic covariates (e.g., race and education). Depressive symptoms in 2016 served as the focal outcome variable. Findings from a series of path models estimated in MPLUS indicated that loneliness is a mechanism through which spousal support predicts older adults’ depressive symptoms. Such linkages, however, were dependent on individuals’ own functional limitations and gender. For functionally limited males in particular, spousal support was shown to reduce depressive symptoms insofar as it was associated with lower levels of loneliness; otherwise, it was associated with higher levels of depressive symptoms. Such findings reinforce the importance of taking a contextualized approach when examining associations between support and emotional well-being later in life.