Fifty healthy mothers, with normal placental function, were anaesthetised with ketamine (2 mg/kg body mass) for Caesarean section. Surgery was conducted with the patient in the lateral tilt position and anaesthesia was maintained with nitrous oxide, oxygen, muscle relaxants and controlled ventilation. Eight of the 50 infants delivered were clinically depressed, judged on the basis of the modified Apgar score at 2 minutes after delivery. The average time to sustained respiration was 58, 1 minutes. Studies of maternal blood gases before induction and at delivery revealed mild respiratory alkalosis, associated with an appropriate degree of compensatory metabolic acidosis. Umbilical cord blood gas analysis showed the presence of a fetal respiratory acidosis. The average derived fetal base excess levels were similar to those obtained in a previous study with thiopentone anaesthesia, but calculated mean maternal-to-fetal pH and base excess gradients were slightly greater in the present study. Prolonged induction-to-delivery intervals were associated with an increase in maternal metabolic and fetal respiratory acidosis. Slow delivery of the infant after invasion of the uterus magnified the degree of fetal metabolic acidosis, and widened the maternal-to-fetal acid-base gradients. Convincing evidence of maternal awareness during surgery was not obtained in this study. Five patients had hallucinations in the immediate postanaesthetic period. Unpleasant dreams were reported by 10% of patients. Delirium on emergence from anesthesia was not encountered. In this study, ketamine appeared to maintain fetoplacental exchange adequately, but may have been responsible for some degree of drug-induced neonatal depression. It is suggested that ketamine should be re-evaluated, using a lower dosage schedule, for Caesarean section.