In vivo and in vitro electrophysiological experiments were performed on the rat dorsal vagal complex (DVC, i.e. nucleus of the tractus solitarius, NTS, and dorsal motor nucleus of the vagus, DMV) to examine the effects of corticotropin releasing hormone (CRF) on the central components of the vago-vagal reflex control of gastric function. When applied to gastrointestinal projecting DMV neurones, CRF (10-300 nM) induced a concentration-dependent membrane depolarization, an increase in action potential firing rate and decrease in amplitude of the action potential afterhyperpolarization (P < 0.05). Pretreatment with the non-selective CRF antagonist, astressin (0.5-1 microM) or the selective CRF(2) receptor antagonist, astressin 2B (500 nM) attenuated the CRF-induced increase in firing rate but did not alter basal discharge rate. CRF (30-300 nM) increased the amplitude of excitatory postsynaptic currents (EPSCs) evoked by stimulation of the NTS (P < 0.05). An alteration in the paired pulse ratio indicated the EPSC's increase occurred due to actions at presynaptic sites. In the in vivo anaesthetized rat preparation, bilateral microinjections (20 fmol in 20 nl for each site) of CRF in the DVC decreased gastric motility in rats pretreated with the muscarinic agonist, bethanecol (P < 0.05). The effects of CRF were abolished by systemic administration of the NOS inhibitor, L-NAME, or by bilateral vagotomy. We concluded that CRF had both a direct and an indirect excitatory effect on DMV neurones via activation of CRF(2) receptors and the decrease in gastric motility observed following microinjection of CRF in the DVC is due to the activation of an inhibitory non-adrenergic non-cholinergic input to the gastrointestinal tract.