The responses of blood glucose and plasma free fatty acid (FFA), triglycéride (TG) and immunoreactive insulin concentrations to the oral and intravenous administration of long-chain TG have been measured in normal subjects. A TG emulsion and an aqueous solution of glycerol were each infused intravenously over five hours on different days in six subjects. Glycerol infusion produced no measurable effect. The TG infusion did not induce changes in plasma insulin and blood glucose concentrations despite increases in plasma FFA and TG levels. Eight subjects ingested safflower oil on one occasion and water on another. Oil ingestion caused an early and significant increase in the plasma insulin concentration, which preceded any increases in plasma FFA and TG concentrations. Blood glucose levels were unchanged. In six additional subjects, the ingestion of a safflower oil emulsion, in contrast to the unemulsified oil, led to earlier and larger increases in plasma FFA and TG concentrations. Despite differences in rates of TG absorption, there was no difference in the pattern of plasma insulin and blood glucose responses to emulsified or unemulsified oil. Therefore only the oral, but not the intravenous, administration of TG stimulated insulin release, which was not related to changes in plasma FFA and TG levels. The insulin response to ingested TG may therefore reflect the release of gastrointestinal hormones, such as pancreozymin, which in turn stimulates the pancreatic islets.