Endogenous hormones may play a role in ovarian carcinogenesis. Postmenopausal obesity, although associated with higher circulating levels of estrogen and androgens, has not been linked consistently to ovarian cancer. The present study examined the relationship between body mass index (BMI), height, and ovarian cancer mortality among postmenopausal women in a large prospective mortality study of 300,537 women who were cancer free at enrollment in 1982 and had no history of hysterectomy or ovarian surgery. During 16 years of follow-up, 1,511 deaths occurred from ovarian cancer. Cox proportional hazard modeling was used to compute rate ratios (RRs) and to adjust for confounders. Ovarian cancer mortality rates were higher among overweight [BMI >/=25;RR, 1.16; 95% confidence interval (CI), 1.04-1.30] and obese women (BMI >/=30; RR, 1.26; 95% CI, 1.07-1.48) compared with women with BMI <25. Use of postmenopausal estrogens modified the association between BMI and ovarian cancer mortality (P = 0.05). The increased risk associated with obesity (BMI >/=30) was limited to women who never used postmenopausal estrogens (RR, 1.36; 95% CI, 1.12-1.66) and was not seen among ever users (RR, 0.93; 95% CI, 0.62-1.41). Height was positively associated with ovarian cancer mortality. Compared with women 152-156 cm tall, ovarian cancer mortality rates were lowest for the shortest women (RR, 0.72; 95% CI, 0.47-1.10 for women <152 cm) and highest for the tallest (RR, 1.41; 95% CI, 0.95-2.09 for women >/=177 cm). In this study, obesity and height appear to be independently associated with ovarian cancer mortality. The 36% increase in risk associated with obesity among women who had never used postmenopausal estrogens may have important public health implications because obesity is a growing problem in the United States.