The relation between childbearing and breast cancer risk was investigated in a prospective study of 63,090 Norwegian women, among whom 1,565 breast cancer cases occurred in a follow-up from 1961 through 1980. The authors observed a strong and highly significant inverse association between the number of full-term pregnancies and the risk of breast cancer. The association was consistently found in all subgroups according to demographic variables and was obtained for all histologic subtypes except for cases classified as Paget's disease. The relation with parity could not be explained by confounding with age at first birth or other reproductive factors. The apparent protective effect of high parity was strongest in the groups with an early first birth, and no such effect could be established among women with their first birth at age 35 or later. The protective effect of multiparity was observed for cancers in all age groups. Uniparous women had, however, higher risk of developing breast cancer under age 60 than nulliparous women. Those who reported at least one abortion had somewhat lower risk than those who did not, but no trend according to the number of abortions was observed.