Fetal growth is normally constrained by maternal factors. This constraint is demonstrated by the usual inverse linear relationship between litter size and mean fetal weight. Cross-breeding experiments between mice of lines selected for high or low plasma insulin-like growth factor (IGF-I) levels suggested that elevations in maternal IGF-I abolish (P <0.01) this constraining effect and reverse the usual positive relationship between fetal and placental size in late gestation. This was confirmed by treating mice and rats throughout pregnancy with IGF-I. In normal mice and in low IGF-I line mice treatment with IGF-I 10μg 8-hourly s.c. from day 1 to 19 of pregnancy) abolished maternal constraint whereas 0.9% (w/v) NaCl treatment did not. In Wistar rats osmotic pumps were implanted to deliver IGF-I (1μg/g body weight per day), bovine GH (bGH; 0.6μg/g body weight per day) or saline from day 1 to 19 of pregnancy. IGF-I therapy but not bGH or saline abolished (P < 0.01) maternal constraint and altered (P <0.01) the relationship between placental and fetal weight. When high or low IGF-I line mice embryos were transplanted into a normal line of mice, the expected negative relationship (P <0.05) between mean fetal weight and litter size was maintained. However the embryos of the high line were heavier (P <0.05) than those from the low line irrespective of fetal number, suggesting a direct role for IGF-I in the regulation of fetal growth. Thus both endogenous and exogenous elevations in maternal IGF-I indirectly promote fetal growth either by altering nutrient delivery to the placenta or by affecting placental function.