The effect of 100 mg of phenobarbital (PB) at bedtime for the last few wk of pregnancy on the incidence and severity of neonatal hyperbilirubinemia was studied. No effect was observed in the newborns of mothers who took less than ten tablets. In the 1310 newborns of adequately treated mothers (PB greater than or equal to 1.0 g), the incidence of marked jaundice (bilirubin > 16.0 mg/dl) and the need to perform an exchange transfusion were reduced by a factor of six in relation to the incidence in 1553 control infants. A randomly selected group of 415 children (182 control, 233 PB) were reexamined at 61 to 82 months of age. There was no difference in the overall morbidity and mortality between the control and treatment group. A detailed neurologic assessment failed to reveal any differences between the two groups. In the VisuoMotor Integration test, the PB group scored significantly better than the control group. In the Draw-A-Woman and the Verbal Intelligence Test, the difference was in the same direction but was not statistically significant. The degree of jaundice was not found to significantly influence the performance in the neurological examination and the intelligence tests. Sensorineural hearing defect was significantly more common in the children with moderate or marked jaundice (bilirubin > 12 mg/dl) than in those with lesser degrees of jaundice. Prenatal PB is a practical, effective, and safe method for decreasing the incidence of neonatal hyperbilirubinemia.