We conducted a meta-analysis of the literature concerning breast cancer and estrogen replacement therapy. The overall relative risk of breast cancer associated with this therapy was 1.07. However, the variation of the estimated risks among the studies was far greater than could plausibly be explained by chance alone. To explain this variation, we looked at the effects of type, duration, and dosage of treatment. Overall, women who took 0.625 mg/d or less of conjugated estrogens had a risk of breast cancer that was 1.08 times that of women who did not receive this therapy (95% confidence interval [CI], 0.96 to 1.2). The relative risks from these individual studies of low-dosage therapy did not differ significantly from each other. Women who took 1.25 mg/d or more of conjugated estrogens had a breast cancer relative risk of 2.0 or less in all studies. However, the variation in observed risks at this higher dosage was significant. This implies that other risk factors varied among these studies, making it difficult to estimate the overall risk associated with this dosage. The relative risk of breast cancer associated with estrogen replacement therapy among women with a history of benign breast disease was 1.16 (95% CI, 0.89 to 1.5). The combined results from multiple studies provide strong evidence that menopausal therapy consisting of 0.625 mg/d or less of conjugated estrogens does not increase breast cancer risk.