To investigate the effects of participation in aerobic exercise on pregnancy outcome, 388 women (mean age = 31.7, range = 18-42) were followed from a mean 16.5-wk gestation through delivery. Frequency, duration, and mode of aerobic exercise prior to conception and during the first trimester were determined by in-person interviews. Activity patterns during the second and third trimesters were assessed by telephone interviews. For each time period, women were categorized into one of the following exercise groups: Level I = aerobic exercise, excluding vigorous walking, at least three times a week for at least 20 min a time; Level II = aerobic exercise at least three times a week and 20 min at a time, if and only if vigorous walking is included; Level III = aerobic exercise less than three times a week, 20 min a time; and Level IV = aerobic exercise less than once a week. Mean birthweight was statistically unrelated to level of exercise preconceptionally or in any trimester. Gestational age, weight gain, and other pregnancy outcomes were also unassociated with exercise level. However, pregnancy symptoms were inversely associated with level of exercise; women who exercised more earlier in pregnancy reported fewer discomforts later in pregnancy (P = 0.01). These data suggest that participation in aerobic exercise during pregnancy at a level great enough to produce or maintain a training effect does not adversely affect birthweight or other maternal and infant outcomes but may be associated temporally with fewer perceived pregnancy-associated discomforts.