The mechanism by which acetylcholine (ACh) decreases systemic arterial pressure and hindlimb vascular resistance was investigated in the anesthetized rat. ACh injections caused dose-dependent decreases in systemic arterial pressure and hindlimb vascular resistance. N(omega)-nitro-L-arginine methyl ester (L-NAME) had little effect on the magnitude of depressor and vasodilator responses but decreased response duration when baseline parameters were corrected by a nitric oxide (NO) donor infusion. The decrease in the duration of the ACh depressor response was prevented by the administration of excess L-arginine. The L-NAME-resistant component of the depressor response to ACh was attenuated by ebselen, a glutathione peroxidase mimic. The calcium-activated potassium (K(Ca)) antagonists charybdotoxin (ChTX) and apamin decreased the magnitude but not the duration of the hindlimb vasodilator response to ACh. The combination of L-NAME, ChTX, and apamin reduced the magnitude and duration of the vasodilator response to ACh but not to sodium nitroprusside. Vasodepressor and hindlimb vasodilator responses to ACh were not modified by cytochrome P-450 and cyclooxygenase pathway inhibitors. These results suggest that the hindlimb vasodilator response to ACh has an initial L-NAME-resistant component mediated by the activation of K(Ca) channels and a sustained L-NAME-dependent component. The results with ebselen suggest that the L-NAME-resistant component of the depressor response involves a peroxide-sensitive mechanism. The present study suggests that vasodilator responses to ACh are not mediated by cytochrome P-450 products, since miconazole and 1-aminobentriazole alone or in combination did not affect either component of the response. The present data suggest that the hindlimb vasodilator response to ACh in the rat is mediated by two mechanisms with an initial ChTX- and apamin-sensitive, L-NAME-resistant phase not mediated by cytochrome P-450 products and a secondary sustained phase mediated by NO.