Rats were surgically prepared to allow perfusions of anatomically limited portions of the gastrointestinal (GI) surface during test meals. The results demonstrated that at least one potent satiety signal was generated when ingested food accumulated in the stomach and did not enter the small intestine. This gastric satiety signal did not require the vagus nerve for its operation. In addition, at least one other potent satiety signal was generated when food perfused the small intestine. This intestinal satiety signal did not require gastric distension for its operation. We tested a variety of GI peptides to determine whether any met the criteria imposed by this evidence for regionally specific satiety signals. Bombesin (BBS), a peptide present in high concentration in the stomach, was a potent and behaviorally specific inhibitor of food intake. Its satiating effect was not altered by subdiaphragmatic vagotomy. Cholecystokinin (CCK), a peptide hormone that is released from the small intestine by food, was also a potent and behaviorally specific inhibitor of food intake; its satiating effect did not require gastric distension for its expression, but its satiating effect was markedly reduced or abolished by subdiaphragmatic vagotomy. Thus, BBS and CCK may mediate at least part of the satiating effect of food acting in the stomach and in the small intestine, respectively.