We performed a multi-institutional, incident case-control study of 328 endometrioid and 26 serous carcinomas to assess whether risk factors and circulating hormone levels in women with serous carcinoma differ from the expected profile for endometrial carcinoma We also evaluated exposures potentially related to endometrial cancer risk, anthropometric measurements, and circulating levels of sex hormones and related carrier proteins. Histopathologic specimens were reviewed without knowledge of the other data. As expected, a statistically significant association was observed for high body mass index (BMI) (relative risk, 3.5) and use of menopausal estrogens (relative risk, 2.4) in the endometrioid carcinoma cases, whereas serous carcinomas were not strongly associated with these factors. Smoking and oral contraceptive use decreased risk for both tumor types. For five of six sex hormones tested, age-adjusted mean serum levels in patients with serous carcinoma were significantly lower than those in women with endometrioid carcinoma. After adjustment for BMI, these differences were narrowed, but levels of albumin-bound estradiol and estrone remained significantly lower in the serous cases. Age and BMI-adjusted levels of sex hormone-binding globulin were significantly higher in patients with serous carcinoma than in women with endometrioid carcinomas. In conclusion, risk factors and sex hormone levels in patients with uterine serous carcinoma seem to differ from those in women with endometrioid carcinoma, suggesting that there may be at least two different pathways of endometrial carcinogenesis.