We prospectively examined the incidence of epithelial ovarian cancer and its subtypes in relation to baseline fruit and vegetable consumption in the Swedish Mammography Cohort, a population-based cohort study of 61 084 women aged 38-76 years in 1987-1990. During an average follow-up of 13.5 years, 266 incident cases of invasive epithelial ovarian cancer were diagnosed. After adjustment for potential confounders, we observed a statistically significant inverse association between consumption of vegetables and ovarian cancer risk (P-value for trend=0.01); the multivariate rate ratio (RR) for the comparison of three or more servings of vegetables per day with one or fewer servings per day was 0.61 (95% confidence interval (CI), 0.38-0.97). For fruit consumption a modest, not statistically significant, positive association was found (P-value for trend=0.07); the multivariate RR for the highest compared with the lowest category of consumption being 1.37 (95% CI, 0.90-2.06). The associations with fruit and vegetable consumption did not vary by subtype of ovarian cancer. These findings suggest that high consumption of vegetables, but not of fruits, may reduce the risk of ovarian cancer.