This study considers potential interaction effects of three measures of religiosity, organized (OR), non-organized (NOR), and intrinsic religiosity (IR), on depression and general mental health, controlling for socio-demographic characteristics and mobility. In-home interviews were conducted among a stratified random sample of Medicare beneficiaries from five central Alabama counties (the University of Alabama at Birmingham Study of Aging). Those who were high on all three dimensions of religiosity reported having fewer symptoms of depression and better mental health than did those who were low on all three dimensions of religiosity. Subjects who scored high on OR reported lower levels of depression (F (1,981) = 3.97, p<0.05). Neither IR nor NOR had salutary effects on the measure of depression nor on the general measure of mental health.The interpretation of the relationships of religiosity with the Geriatric Depression Scale (GDS) and the general mental health (Mental Component Score of the SF-12; MCS) measures was complicated by the presence of three way interactions (F (1,981) = 9.02, p<0.01 and F (1, 981) = 5.46, p<0.05, for GDS and MCS respectively). The presence of interaction effects between the different dimensions of religiosity and mental health affirms the importance of remaining sensitive to the multidimensional nature of religiousness and its relationships with measures of mental health.