The authors examined models of the relationships between religious activities, physical health, social support, and depressive symptoms in a sample of 4,000 persons age 65 and over. Religious activity was examined first as a single composite construct and then split into three component variables that were examined individually. Religious activity as a single construct was correlated with both social support and good physical health but was unrelated to depression. Split into the three components, model fit was significantly increased. Frequency of church attendance was positively related to physical health and negatively related to depression, but was surprisingly unrelated to social support. Frequent churchgoers were about half as likely to be depressed. Private prayer/Bible reading was negatively correlated with physical health and positively correlated with social support, but unrelated to depression. Religious TV/radio listening was unrelated to social support, negatively related to good physical health, and, unexpectedly, positively associated with depression.