This study was designed to test the hypothesis that pregnancy has an added training effect (increases "absolute" VO2max) in well-conditioned, recreational athletes. VO2max was measured serially in 20 nonpregnant recreational athletes who maintained their exercise within +/- 10% of initial levels over a 15-month period and 20 similar women who conceived and continued exercise at a reduced level during pregnancy with a return to within 20% of initial levels by 12 wk postpartum. Initially the two groups were similar in terms of age (30 +/- 1 vs 30 +/- 2 yr), weight 57.6 +/- 7.2 vs 59.7 +/- 7.5 kg), max pulse rate (189 +/- 8 vs 187 +/- 10 bpm), and absolute (3083 +/- 469 vs 3138 +/- 464 ml.min-1) VO2max. In the nonpregnant group the values obtained 15 months later were unchanged (weight = 57.8 +/- 6.6 kg, max pulse = 191 +/- 7 bpm, VO2max = 2977 +/- 397 ml.min-1) while those who conceived had a significant increase in absolute VO2max that was evident 12-20 wk postpartum and was maintained at the time of final testing 36-44 wk postpartum (3368 +/- 435 ml.min-1). Both weight (60.1 +/- 8.1 kg) and maximum pulse rate (185 +/- 12 bpm) were unchanged. These data indicate that pregnancy is followed by a small but significant increase in VO2max in recreational athletes who maintain a moderate to high level of exercise performance during and after pregnancy.