To better understand why plasma immunoreactive insulin (IRI) concentration is high in the rat fetus during the last 3 d of gestation and why fetal hyperinsulinemia abruptly subsides after birth, insulin secretion and metabolic clearance rates were estimated in fetuses and nursed pups. Intravenously injected [(125)I]monoiodoinsulin was cleared from the plasma of prematurely delivered pups at least as rapidly as from that of 7- to 10-d-old pups, suggesting that fetal hyperinsulinemia is not a result of slow clearance of the hormone. The fetal liver bound 35% of the injected label within 3 min, and binding was saturable. The uptake of radioactivity by the fetal kidney was nonsaturable and much lower than that of adult rat kidney. Isolated fetal islets were already reactive to glucose on the 19th d of gestation. Pancreatic insulin secretory capacity was estimated by measuring (a) the insulin release of isolated islets incubated in the presence of 2.8 mM glucose, (b) the insulin content of the same islets, and (c) the total insulin extracted from the pancreas, using the formula (a x c)/b. 2 d before birth, the pancreatic insulin secretory capacity was high, accounting for fetal hyperinsulinemia. It was even higher after birth, not accounting for the postnatal decrease in plasma IRI concentration. Pups delivered by caesarian section 1 d before term exhibited a brisk decrease in plasma IRI concentration when the cord was cut. By contrast, if the feto-placental unit was removed from the dam, maintaining fetal blood circulation through the placenta, fetal plasma IRI concentration remained as high as in utero. These experiments suggest that a placental factor stimulates fetal insulin secretion. After birth, when the cord is cut, insulin secretion is rapidly turned off, and the pups switch from a state of unlimited fuel supply by the mother to a state of fuel saving.