Possible associations between childbearing and the risk of brain cancer were explored in a case-control study "nested" within a large nationwide cohort defined by the Swedish Fertility Registry. Among women born between 1925-1975, 1,088 patients with meningiomas and 1,657 patients with gliomas were identified in the Swedish Cancer Registry. For every woman diagnosed with brain tumor, 5 age-matched controls were selected among those in the Fertility Registry. Relative risks were estimated by odds ratios from conditional logistic regression. Ever-parous women were at a reduced risk of glioma compared to nulliparous women, while parity was unrelated to meningioma risk. Age at first birth was unrelated to both meningioma and glioma risk. The gradient in risk between ever-parous and nulliparous women for gliomas, but not meningiomas, is difficult to explain biologically. A possible explanation is that pregnancy-induced alterations in androgen levels reduce the risk of glioma in parous women. Alternatively, childlessness may represent a marker of an occult glioma, negatively affecting fecundity. Overall, our present results do not support the notion that hormonal changes, or other physiological changes induced by childbearing, play an important role in the development of brain tumors.