To evaluate the influence of pregnancy on resting and exercise pulmonary function, we studied 16 nonpregnant and 26 pregnant females in a two-phase study. During phase I, all pregnant subjects (P1) (mean gestational age = 29.5 +/- 7 weeks [X +/- SD]) were compared with 16 control (C) subjects. In phase II, 8 of the pregnant subjects (P2) (mean gestational age = 32.6 +/- 6 weeks) were evaluated again at 12.5 +/- 7 weeks' postpartum (PP). Exercise respiratory function was measured during work on a cycle ergometer during rest and at three submaximal work levels (25, 50, and 75 W), during the last minute of 5 to 7 minutes of steady-level power output. Ventilatory parameters [ventilation (VE), ventilatory frequency (VF), tidal volume (TV), ventilatory equivalent for oxygen (VE/VO2) and ventilatory equivalent for carbon dioxide (VE/VCO2)] and metabolic parameters [oxygen consumption (VO2), oxygen pulse, carbon dioxide production (VCO2), respiratory exchange ratio (RER), and plasma lactate (HLA)] were measured during the submaximal tests and throughout the progressive increase in work after the 75 W work load, until peak VO2 was achieved. The results from the phase I comparison indicated (as expected) higher resting ventilation, tidal volume, oxygen consumption, carbon dioxide production, and respiratory exchange ratio associated with pregnancy, and these findings were similar when the pregnant subjects were compared with their postpartum values. Plasma lactate levels, although similar in the phase I comparison, were significantly higher during pregnancy when compared with postpartum values. No resting differences were observed for VF, O2 pulse, VE/VO, or VE/VCO2 in either phase of the study. The results of the submaximal exercise tests indicated higher submaximal levels of VE, VF, VE/VO2, VE/VCO2 in P1 versus C; however, P2 had significantly higher VE, TV, VO2, VE/VCO2, when compared with PP. The influence of pregnancy on VCO2 and RER differed by workload in both sets of comparisons. Furthermore, no significant differences were observed between P1 and C and P2 and PP for plasma lactate levels. These findings suggest that pregnancy/control versus pregnancy/postpartum studies yield different results in both ventilatory and metabolic comparisons. Consideration must be given to the type of study when discussing and applying the findings.