The role of depression in nonresident fathers' involvement with their infant children is poorly understood. A three-factor model of father involvement was evaluated, and its association with parental relationship quality and depressive symptoms in both parents were tested. Data on 569 families from the Early Childhood Longitudinal Study were used. Confirmatory factor analysis supported a three-factor model of nonresident father involvement, which was then examined in a model consistent with Belsky's determinants of parenting framework. Noncohabitating mothers and fathers evidenced a significant correlation between their quantitative levels of depression. Relationship quality predicted all factors of father involvement and was negatively associated with depression in either parent. Disruptions in relationship quality mediated the link between depression in both parents and reduced father involvement. Perhaps because of depression's association with relationship quality, depressive severity was significantly correlated between nonresident mother and father. (Contains 2 tables and 2 figures.)