Insulin acts directly as a mitogen on isolated embryonic tissues but has been shown to be inactive at physiological concentrations on several fetal cell types. Since insulin availability is obligatory for optimal fetal growth, we have investigated its mitogenic actions on chondrocytes isolated from the epiphyseal growth plates of fetal lambs. Chondrocytes were isolated from the proximal tibial growth plate of lamb fetuses between 40 and 130 days gestation using collagenase and were cultured in monolayer before use between passages 2 and 6. The synthesis of DNA was assessed from the incorporation of [3H]thymidine after incubation in medium supplemented with glucose (0.7 mM-25 mM) with or without insulin (0.08 nM-167 nM). Increase in cell number was assessed after incubation with test medium for up to 8 days. Insulin substantially increased both DNA synthesis, and cell number, compared to control incubations with a biphasic dose response; an initial 3- to 5-fold increase in DNA synthesis occurring at approximately 1 nM insulin with a second response seen at approximately 50 nM. Within the physiological range of concentrations insulin was only 50% as active as insulin-like growth factor I (IGF I), but was 15 times more active than IGF II. Similar effects of insulin were observed throughout the fetal age range, although the DNA synthetic rate in basal medium declined with both fetal age and cell passage number. The mitogenic actions of insulin were glucose-dependent and were maximal in the presence of 2.7 mM glucose. Insulin did not cause any change in chondrocyte cell cycle duration. Chondrocytes released immunoreactive IGF II but no detectable IGF I. While exposure to insulin concentrations of approximately 50 nM or greater resulted in a statistically significant increase in IGF II release from chondrocytes, no change in IGF II release was seen in response to physiological insulin concentrations. However, exposure of cells to a blocking monoclonal antibody against human IGF I, McAb SM 1.2, which also negates the mitogenic actions of IGF II, consistently reduced insulin-stimulated DNA synthesis suggesting that IGF II presence may be necessary for optimal insulin action. Combination experiments using maximal concentrations of IGF I (13.3 nM) and increasing amounts of insulin (0.16 nM-1.67 nM) showed additive effects on DNA synthesis, suggesting that each hormone was acting through distinct receptor populations. We conclude that insulin, at physiological concentrations, may exert direct growth-promoting actions at the epiphyseal growth plates of the fetal lamb throughout gestation.