A total of 112 white females residents of King County, Washington, aged 35-54 years, who had received a first diagnosis of invasive breast cancer between July 1977 add August 1978, were investigated concerning prior use of oral contraceptives. Their responses were compared with those of a random sample of 469 demographically comparable women from the same population. Overall, oral contraceptive use in cases and controls was similar. However, use of oral contraceptives in preparous women was more common among cases than controls, with the estimated risk of breast cancer associated with such use being 2.2 times that of nonusers (90% confidence interval = 1.1-4.6). This relationship could be explained only in part by the effect of oral contraceptives in postponing or preventing childbirth. The association of breast cancer with use of oral contraceptives prior to ever giving birth has been observed in three studies, including this one, suggesting that the susceptibility of breast tissue to hormonal factors that influence the development of malignancy may be altered by having been exposed to the events of pregnancy.