To determine the degree to which endogenous and exogenous hormonal factors influence the risk for subarachnoid hemorrhage in women. A population-based case-control study. King County, Washington. 103 women with an incident, spontaneous subarachnoid hemorrhage and 2 age- and sex-matched controls per case-patient who were identified through random-digit dialing. Information on exposures was collected during in-person interviews of case-patients, controls, and their surrogates. Premenopausal women, especially those without a history of smoking or hypertension, were at a reduced risk for subarachnoid hemorrhage compared with age-matched postmenopausal women (odds ratio, 0.24; 95% CI, 0.09 to 0.68). The use of hormone replacement therapy was associated with a reduced risk (odds ratio, 0.47; CI, 0.26 to 0.86); the reduction was significantly greater in women who had smoked than in those that had never smoked. Of the 23 premenopausal case-patients, 74% were either menstruating when hemorrhaging occurred or had had their last menstrual period 21 or more days before hemorrhaging compared with the expected 43% (difference, 31%; CI, 4% to 58%). Premenopausal women are at reduced risk for subarachnoid hemorrhage, especially those without a history of smoking or hypertension. Hormone replacement therapy reduced the risk only in postmenopausal women who had ever smoked. Among women still menstruating, the risk for hemorrhage was greatest in the perimenstrual period.