Abstract To characterize the prandial plasma corticosterone peak, three experiments were done on adult male Wistar rats: (1) the prandial hormone peak was followed at three different times of day, 0800 h, 1900 h, and 0100 h, in starved rats; (2) a 20% glucose solution was infused into the jugular vein or was given as a drinking solution after a 24-h starvation; (3) in order to determine the role of vagal proprioceptive inputs, the hormone peak was examined at two times of day, 0800 h and 0100 h, in starved vagotomized rats. The prandial hormone peak was more conspicuous at midnight than in the daytime. Exposure to food pellets without feeding failed to induce a hormone rise. The peak appeared after oral ingestion, but not after intravenous infusion, of glucose solution. Ingestion of water alone was without effect. The peak was not affected by bilateral vagotomy. Thus, as far as glucose as a metabolite is concerned, the possibility that glucose per se works directly on the brain or indirectly through glucoreceptors in the gut was excluded. Possible modes of action of glucose ingestion are discussed.