The effects of physical activity during pregnancy and lactation on the fetal outcome and the growth of pups was studied in Wistar rats (n 144). Rats were trained to swim for 2 h every day, 6 d/week through pre-pregnancy, pregnancy and lactation. Maternal exercise during pregnancy, despite the dams having ad lib. access to food, resulted in low-birth-weight pups (5.6 (SD 0.7) g; n 178 in exercised dams v. 6.2 (SD 0.8) g; n 238 in sedentary dams). Maternal exercise continued through lactation exaggerated further the growth retardation of these pups (30.0 (SD 4.7) g; n 78 in exercised dams v. 36.0 (SD 6.9) g; n 126 in sedentary dams). The effects of maternal exercise during pregnancy and lactation studied over two successive generations revealed a reduction in the growth rates of the second generation progeny of both exercised (5.3 (SD 0.9) g; n 125 at birth and 25.1 (SD 6.8) g; n 54 at weaning) and sedentary rats (6.0 (SD 0.2) g; n 110 at birth and 31.3 (SD 4.3) g at weaning) born to first-generation exercised rats. While slower growth in the former indicates a cumulative effect of exercise stress over two generations, that of the latter indicates that the generational effects are manifest even though the dams of the F1 generation were not exposed to exercise stress during pregnancy and lactation. These findings suggest that the adverse effect of maternal exercise during pregnancy and lactation on fetal outcome in one generation is transferred to the subsequent generation.