Few studies have examined longitudinal associations between fish consumption and depression; none have defined depression using a diagnostic tool. We investigated whether fish consumption was associated with fewer new depression episodes in a national study of Australian adults. In 2004–2006, 1,386 adults aged 26–36 years (38% males) completed a 127-item (9 fish items) food frequency questionnaire. Fish intake was examined continuously (times/week) and dichotomously (reference group: <2 times/week). During 2009–2011, the lifetime version of the Composite International Diagnostic Interview was administered by telephone. New episodes of major depression/dysthymic disorder (since baseline) were defined using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. During follow-up, 160 (18.8%) women and 70 (13.1%) men experienced depression. For women, each additional weekly serving of fish consumed at baseline decreased the risk of having a new depressive episode by 6% (adjusted relative risk = 0.94, 95% confidence interval: 0.87, 1.01). Women who ate fish ≥2 times/week at baseline had a 25% lower risk of depression during follow-up than those who ate fish <2 times/week (adjusted relative risk = 0.75, 95% confidence interval: 0.57, 0.99). Reverse causation was also suggested but appeared to be restricted to persons with recent depression. Fish consumption was not associated with depression in men. These findings provide further evidence that fish consumption may be beneficial for women's mental health.