Lack of social support is a known risk factor for postpartum depression (PPD). However, the association between lack of social support from a partner or others and PPD remains unknown. We examined this association among Japanese mothers. We distributed an original questionnaire to mothers participating in a three- or four-month health check-up program over October to November 2012 in Aichi Prefecture, Japan. Of the 9707 eligible mothers, 6590 responded to the questionnaire (response rate: 68%). Social support from a partner or others was assessed based on whether the mother can consult with her partner or others (i.e., parents, relatives, and friends who are close by or far) on childcare. PPD was assessed with the Edinburgh Postnatal Depression Scale. The data were analyzed using multiple logistic regression analysis for four categories: no social support from either a partner/others, social support from a partner only, social support from others only, and social support from both, adjusted for possible covariates. Mothers who have no social support from either a partner/others, have social support from a partner only, and have social support from others only were 7.22 (95% confidence interval [CI], 1.76–29.6), 2.34 (95% CI, 1.37–3.98), and 3.13 (95% CI, 2.11–4.63) times more likely to show PPD, respectively, in comparison with mothers who have social support from both, after adjustment of possible covariates. Mothers with no social support from a partner, but have social support from others, showed significant risk for PPD, which may be invisible. Further prevention effort is needed to detect PPD cases, with a focus on mothers without support from their partner.