1. Gastric motility and neurotransmitter release in response to vagal stimulation were studied using a vascularly isolated perfused rat stomach. Gastric motor responses were recorded by a strain gauge force transducer implanted on the proximal stomach. 2. Electrical stimulation of vagal trunk (0.5-20 Hz) produced a triphasic response which was composed of a rapid transient relaxation (first phase) followed by a phasic contraction (second phase) and a delayed prolonged relaxation (third phase). Maximum responses of the first, second and third phase were observed at 2.5, 5 and 10 Hz, respectively. Intra-arterial infusion of tetrodotoxin (0.1 microM) or hexamethonium (100 microM) completely abolished the triphasic response. 3. The nitric oxide (NO) biosynthesis inhibitor NG-nitro-L-arginine (L-NNA; 100 microM) significantly antagonized the rapid relaxation but had no effect on the delayed relaxation, while vasoactive intestinal polypeptide (VIP) antagonist (1 microM) significantly reduced the delayed relaxation without affecting the rapid relaxation. 4. In response to vagal stimulation, NO production ([3H]citrulline formation in gastric tissue preloaded with [3H]arginine) was maximum at 2.5 Hz, whereas VIP release into the venous effluent was largest at 10 Hz. Hexamethonium abolished vagal-stimulated NO production and VIP release. L-NNA had no effect on VIP release in response to vagal stimulation. 5. The nicotinic receptor agonist 1,1-dimethyl-4-phenylpiperizinium (DMPP; 100 microM) also caused a triphasic response similar to that observed with vagal stimulation and produced a significant increase in VIP and NO formation. DMPP-evoked VIP release was not affected by L-NNA. Similarly, DMPP-evoked NO production was not antagonized by VIP antagonist. 6. These results suggest that vagus nerve stimulation evokes NO and VIP release via nicotinic synapses which cause different modes of relaxation of the stomach. There is no interaction between NO and VIP release in response to vagal stimulation.