Twelve newborn infants were given morphine intravenously for postoperative analgesia. They received a continuous infusion of 6.2 to 40 micrograms/kg/hr for 9 to 105 hours (mean +/- SEM 59.5 +/- 10.2 hours); in four the infusion was preceded by a loading dose of 50 to 100 micrograms/kg. Morphine plasma concentrations correlated with the rate of infusion, but with large variability. There was a tendency for plasma morphine concentrations to decrease in some patients receiving a constant infusion rate, suggesting improvement in morphine clearance rate. Elimination half-life of morphine (13.9 +/- 6.4 hours) was significantly longer than in older children and adults (about 2 hours). Similarly, morphine concentrations in neonates receiving 20 micrograms/kg/hr for 24 hours were three times higher (52 +/- 31 ng/ml) than in older children receiving the same schedule. Two infants who received 32 and 40 micrograms/kg/hr, respectively, developed generalized seizures. Because of the apparently greater sensitivity to morphine and the lower elimination rate in newborn infants, the infused dose should not exceed 15 micrograms/kg/hr.