This study was designed to test the hypothesis that continuation of a regular running or aerobics program, or both, during the latter half of pregnancy would have a negative effect on the course and outcome of labor. The onset, course, and outcome of labor were independently monitored in 131 well-conditioned recreational athletes who had an uneventful first half of pregnancy. Daily exercise performance was quantitated before conception and throughout pregnancy. Comparisons were made between the 87 women who continued to exercise regularly at or above 50% of their preconceptional level throughout pregnancy and the 44 who discontinued their regular exercise regimen before the end of the first trimester. The incidence of preterm labor was similar in the two groups (9%). Labor began significantly earlier in the exercise group (277 +/- 6 vs 282 +/- 6 days). The women who continued to exercise had a lower incidence of abdominal (6% vs 30%) and vaginal (6% vs 20%) operative delivery, and active labor was shorter (264 +/- 149 vs 382 +/- 275 min) in those who were delivered vaginally. Finally, clinical evidence of acute fetal stress (meconium, fetal heart pattern, and Apgar score) was less frequent in the exercise group (50% vs 26%), although birth weight was reduced (3369 +/- 318 vs 3776 +/- 401 gm). These data negate the initial hypothesis and indicate that, in well-conditioned women who regularly perform aerobics or run, continuation of these exercise regimens has a beneficial effect on the course and outcome of labor.