We examined the effects of exogenous 5-hydroxytryptamine (5-HT) and selective 5-HT receptor agonists and antagonists on proximal, middle and distal colonic motility in conscious fasted dogs with extraluminal force transducers implanted chronically. 5-HT (0.003-0.1 mg/kg i.v.) dose-dependently enhanced motility along the entire length of the colon. The 5-HT (0.03 mg/kg i.v.)-induced response was inhibited by 0.1-1.0 mg/kg i.v. methysergide, a 5-HT1/2 antagonist, at all recording sites and by 0.1-1.0 mg/kg i.v. ketanserin, a 5-HT2A antagonist, at the middle and distal sites only. At 1 mg/kg i.v., YM060, a 5-HT3 antagonist, reduced the amplitude of the initial transient high-amplitude contractions induced by 5-HT, but did not affect the tonic contraction induced by 5-HT. At doses up to 3 mg/kg i.v., 2-methoxy-4-amino-5-chlorobenzoic acid 2-(diethylamino) ethyl ester (SDZ205-557), a 5-HT4 antagonist, and hexamethonium (up to 10 mg/kg i.v.) did not affect 5-HT-induced responses at any recording site. Renzapride, a 5-HT4 agonist, also stimulated motility along the entire length of the colon at 0.3 mg/kg i.v.. The renzapride-induced response was inhibited by 1 mg/kg i.v. SDZ205-557 or 3 mg/kg i.v. hexamethonium. m-Chlorophenylbiguanide (m-CPBG), a 5-HT3 agonist, (1 mg/kg i.v.) produced a transient high-amplitude contraction at all recording sites and this contraction was eliminated by pretreatment with 0.03 mg/kg i.v. YM060. The contraction produced by m-CPBG declined rapidly, so the increase in the motility index by m-CPBG was not significant at any recording site. Of the antagonists tested, 0.1-1 mg/kg i.v. methysergide produced a delayed and prolonged contractile response at the middle and distal sites. The onset of the response was delayed about 20 min after application and the response was maintained over the subsequent 60-min observation period. The methysergide (1 mg/kg i.v.)-induced response was inhibited by 3 mg/kg i.v. hexamethonium. The other antagonists, ketanserin, YM060 and SDZ205-557, had no contractile effect at any recording site. These results indicate that exogenous 5-HT stimulates motility along the entire length of the fasted canine colon and that 5-HT-induced responses in the proximal colon are mediated mainly by 5-HT1, whereas those in the middle and distal colon are mediated by both 5-HT1 and 5-HT2 receptors. Renzapride and methysergide also stimulate colonic motility via additional mechanisms. The activation of 5-HT4 receptors and the blockade of endogenous 5-HT inhibitory regulation via 5-HT1 receptors may be involved in the action of renzapride and methysergide respectively.