Maternal depression is the most common complication of the postpartum, having devastating and long lasting effects on mother and infant. Lactation is associated with attenuated stress responses, especially that of cortisol, and the lactogenic hormones, oxytocin and prolactin, are associated with anti-depressant and anxiolytic effects. These associations suggest that breast-feeding may decrease maternal depressive symptoms, yet empirical results have been conflicting. Recent findings have indicated that parity may mediate the association between breast-feeding and stress response. Because a decreased stress response is associated with a decreased risk for depression, parity may also mediate the association between infant feeding method and depressive symptoms. Specifically, the benefits of breast-feeding may appear in multiparous but not primiparous mothers. In the present study, data drawn from a national sample of primiparous and multiparous mothers were examined for possible associations between infant feeding method and depressive symptoms, as assessed by the Center for Epidemiological Survey-Depression scale (CES-D). After controlling for several possible confounding variables, breast-feeding by multiparas was associated with significantly decreased odds of having depression compared with bottle-feeders (OR = 0.41, CI 0.19-0.87, p = 0.02); however, no risk reduction from breast-feeding was evident among primiparas. The results support a parity-mediated association between lactation and maternal depressive symptoms. The results provide a reason for earlier conflicting findings, present new research avenues, and suggest possible clinical approaches.