This study investigated the effects of intravenous morphine on gastric antral and duodenal motility in healthy volunteers. Intravenous morphine (both infusion and bolus) increased duodenal motility, typically as bursts of contractions similar to phase III of the migrating motor complex. Intravenous infusion of morphine 40 microg/kg/hr rapidly increased duodenal motility in nine of 10 subjects; in eight it was phase III-like. Intravenous infusion of naloxone (40 microg/kg/hr) blocked this effect of morphine infusion in five of six subjects. Morphine bolus injection (5-20 microg/kg) in six subjects (30-42 min following a spontaneous phase III) induced further duodenal phase III-like activity; also, morphine bolus injection (5-20 microg/kg) in five subjects (30-42 min following a liquid meal) induced duodenal phase III-like activity. Atropine (10 microg/kg intravenously) was able to prevent the action of morphine (both intravenous infusion and intravenous bolus injection) in inducing this phase III-like activity. These observations show: (1) morphine in very low dose is able to stimulate maximal duodenal contractility; (2) the motility response is typically phase III-like; and (3) morphine acts on opioid receptors to initiate this phase III-like activity, with the effect blocked by antimuscarinic drugs.