A large body of evidence has suggested a role for the endogenous opiates and their receptors in the regulation of appetite. In this study we have examined the relative effects of ketocyclazocine (KC), cyclazocine and ethylketocyclazocine, all putative kappa opiate receptor agonists, and morphine, a putative mu receptor agonist, on food consumption. All the kappa agonists induced feeding when administered at 8 AM as did morphine. KC failed to induce feeding during the nocturnal feeding period (2000 and 0200 hours) and morphine suppressed feeding at these times. KC and morphine suppressed starvation induced feeding when food was made available immediately after injection and had no effect when food was presented 2 and 4 hours after injection. High doses of naloxone (5 mg/kg) suppressed KC induced feeding while actually enhancing high dose morphine (25 mg/kg) induced feeding. Repeated injections of KC or morphine for 5 days resulted in enhancement of the feeding response with initiation of feeding occuring earlier. Taken together with the studies showing that the endogenous kappa ligand, dynorphin, enhabces feeding the most parsimonious interpretation of these studies is that kappa agonists are endogenous initiators of feeding and that kappa receptors are maximally saturated at times of food deprivation and during spontaneous feeding. The mu (or one of the other) opiate receptors inhibit feeding due to their sedative effect and antagonism of this effect leads to enhancement of the feeding response. It is postulated that kappa opiate receptors represent an important component of the natural feeding drive.