This report describes maternal and perinatal risk factors for Wilms' tumor analyzed in a case-control study nested in a nationwide cohort in Sweden. The Swedish National Cancer Registry ascertained 110 cases from among successive birth cohorts from 1973 through 1984, identified by the Swedish Medical Birth Registry, the latter based on medical records. From the Birth Registry, we matched 5 controls without cancer to each case by sex and date of birth. Wilms'-tumor children were more likely to have mothers who had been exposed to penthrane (methoxyflurane) anesthesia during delivery than mothers of controls (odds ratio (OR) = 2.4; 95% confidence interval (CI) 1.1 to 5.1); this excess risk was higher in females than males and increased with age at diagnosis. Wilms'-tumor cases were also more likely to have had physiologic jaundice (OR = 2.3; 95% CI 1.1 to 5.0). Higher parity of the mother decreased the risk of Wilms' tumor among females (OR = 0.7; 95% C1 0.5 to 1.0). We were unable to confirm the reported increased risks of Wilms' tumor for those with high birth weights or with a maternal history of hypertension or fluid retention during pregnancy, nor did we find any association with mother's age at delivery, previous stillbirth, previous live birth, gestational length or height of the child.