Following local anesthetic use, maternal and umbilical serum levels of lidocaine were determined at delivery by means of a gas-chromatography-mass-spectrometry technique in 13 cases. In six cases, where delivery was performed by cesarean section, lidocaine was used for epidural analgesia. The dose given averaged 4.0 +/- 1.7 mg/kg, and the time between analgesia and delivery was 22.0 +/- 4.5 minutes. The mean umbilical serum level of lidocaine was 1.19 +/- 0.79 micrograms/ml and that of the maternal serum was 2.18 +/- 1.25 micrograms/ml. The fetal to maternal ratio was 0.52 +/- 0.18. Lidocaine levels of neonatal plasma were followed at 3, 6, 12, and 24 hours after delivery, and the mean half-life was found to be 6.7 +/- 1.3 hours. In the other seven cases, lidocaine was given in normal vaginal delivery for pudendal nerve block, and the dose was as small as 0.79 +/- 0.06 mg/kg. The mean umbilical and maternal serum concentrations of lidocaine were 0.064 +/- 0.039 micrograms/ml and 0.143 +/- 0.071 micrograms/ml, respectively, and the ratio was 0.45 +/- 0.16. Lidocaine given to the mothers crossed to the fetuses readily and resulted in neonatal plasma levels that were half those of the mothers'. The elimination of lidocaine from the newborn after birth was prolonged so that it might prevent the adaptation of the infant to postnatal circumstances. Viewed from the standpoint of infant care, anesthetics at delivery should be given to the mother only when the benefit obtained by their use outweighs any possible disadvantages.