The role of endogenous opiates and cholecystokinin (CCK) in the control of postprandial pyloric myoelectric activity was investigated in conscious dogs with chronically implanted intraparietal electrodes at the gastroduodenal junction. Meals consisted of either 20 g/kg of canned food (standard meal) or the same food supplemented with 0.5 mL/kg of arachis oil (fat meal). During the 6 hours after standard and fat meals, the number of pyloric spike bursts, 2-4 seconds in duration, was 61.8 +/- 15.8 and 49.9 +/- 12.7/15 minutes, respectively. Administered 15 minutes before a fat meal, naloxone (50 micrograms/kg IV) decreased the number of spike bursts by 31.4%, whereas methyl-levallorphan, a peripheral opiate antagonist, increased postprandial spike activity by 22.2% when administered IV (0.5 mg/kg) and decreased it when administered intracerobroventricularly at a dose of 10 micrograms/kg. These two antagonists administered in the same conditions before a standard meal had no effect on the postprandial spike activity. A 1-hour infusion of cholecystokinin octapeptide (CCK-8), 500 ng.kg-1.h-1 IV and 50 ng.kg-1.h-1 intracerebroventricularly, performed 1 hour after a standard meal induced a 19.6% and 15.8% decrease in the number of pyloric spike bursts, respectively. Both naloxone IV (50 micrograms/kg) and methyl-levallorphan intracerebroventricularly (10 micrograms/kg) administered before the infusion of CCK-8 reinforced this pyloric inhibition, which was antagonized by methyl-levallorphan IV (0.5 mg/kg). The CCK antagonist asperlicin, 200 micrograms/kg IV and 20 micrograms/kg intracerebroventricularly, administered before a fat meal increased pyloric spike bursts by 22.0% and 31.5%, respectively. These results indicate that after a fat meal, endogenous opiates exert a peripheral inhibitory and central stimulatory control of pyloric motility; they suggest the involvement of both peripheral and central release of CCK.