PIP: This review considers the possibility that female smokers have a modified risk of estrogen-related disease. The association of smoking with 2 estrogen deficiency states--early menopause and postmenopausal osteoporosis--is examined, and data on possible negative relationships between smoking and cancers of the breast and endometrium are reviewed. Data were obtained through literature searches and consultations with investigators having unpublished results. Epidemiologic evidence strongly supports an association of smoking with early menopause. Toxic effects on the ovary, enhancement of estrogen metabolism, and effects on central nervous system hormone release have been proposed as the mechanism of this relationship, but none has been fully investigated. Smoking also appears to be associated with osteoporotic fractures in postmenopausal women, particularly among thin women not on estrogen. Although data are sparse and confounding questions exist, female smokers appear to be at lower risk of endometrial cancer than nonsmokers. Ex-smokers may be at higher risk of breast cancer than current smokers, perhaps reflecting hormonal changes associated with stopping smoking. The precise role of estrogens in breast cancer etiology remains undefined, however. Data are insufficient to determine whether a smoking association with osteoporosis and breast and endometrial cancer arises through smoking's association with early menopause. Overall, the data suggest a smoking impact on estrogen-related phenomena. The mechanism of this association is probably extraovarian. It is noted that whatever protective effect smoking may have for estrogen excess disease, smokers have a substantially higher overall mortality than nonsmokers.