Total maternal plasma cortisol levels were measured by a radioassay method in 9 patients who were in spontaneous labor and 10 patients who were electively induced at term with oxytocin. Determinations were made at onset of labor and repeated at full cervical dilatation. Total maternal plasma cortisol levels were also measured in 7 patients undergoing elective cesarean section without labor, determinations being made just prior to the procedure and at the time of uterine incision. Computerized analysis showed the mean initial cortisol level (+/- SE) in the spontaneous labor group (15.4 +/- 1.6 mug/100 ml) to be significantly less than the mean initial level of the group electively delivered by oxytocin induction (37.2 +/- 6 mug/100 ml), with P less than 0.01. The former value was also found to be significantly less than that of the group electively delivered by cesarean section (32.1 +/- 9.3 mug/100 ml), with P less than 0.05. A significant rise was noted at full cervical dilatation in the spontaneous labor group (P less than 0.05), whereas no change occurred in the two elective groups. No significant correlation was found between the maternal cortisol levels on the one hand and the cord cortisol levels. These findings indicate that a) maternal participation is unlikely in bringing about a surge of fetal plasma cortisol which is thought to precede spontaneous labor, b) elective termination of term pregnancy by oxytocin induction or cesarean section may be initially more stress-provoking to the mother than spontaneous labor, and c) maternal stress as measured by plasma cortisol level is not reflected in the fetus.