Abstract It remains unclear whether physical activity is associated with epithelial ovarian cancer risk. We therefore examined the association between recreational physical activity and risk of ovarian cancer in a national population-based case-control study in Australia. We also systematically reviewed all the available evidence linking physical activity with ovarian cancer to provide the best summary estimate of the association. The case-control study included women ages 18 to 79 years with a new diagnosis of invasive (n = 1,269) or borderline (n = 311) epithelial ovarian cancer identified through a network of clinics, physicians, and state cancer registries throughout Australia. Controls (n = 1,509) were randomly selected from the national electoral roll and were frequency matched to cases by age and state. For the systematic review, we identified eligible studies using Medline, the ISI Science Citation Index, and manual review of retrieved references, and included all case-control or cohort studies that permitted assessment of an association between physical activity (recreational/occupational/sedentary behavior) and histologically confirmed ovarian cancer. Meta-analysis was restricted to the subset of these studies that reported on recreational physical activity. In our case-control study, we observed weakly inverse or null associations between recreational physical activity and risk of epithelial ovarian cancer overall. There was no evidence that the effects varied by tumor behavior or histologic subtype. Twelve studies were included in the meta-analysis, which gave summary estimates of 0.79 (95% confidence interval, 0.70-0.85) for case-control studies and 0.81 (95% confidence interval, 0.57-1.17) for cohort studies for the risk of ovarian cancer associated with highest versus lowest levels of recreational physical activity. Thus, pooled results from observational studies suggest that a modest inverse association exists between level of recreational physical activity and the risk of ovarian cancer. (Cancer Epidemiol Biomarkers Prev 2007;16(11):2321–30)