The mechanism by which halothane interferes with catecholamine-induced vasoconstriction was examined, utilizing specific agonists at postjunctional alpha 1- and alpha 2-adrenoceptors on vascular smooth muscle. Stimulation of either adrenoceptor subtype normally produces vasoconstriction. Two experimental models of drug-induced vasoconstriction were used: in vivo blood pressure response in pithed rats, and in vitro isometric tension development in canine saphenous vein rings. These models were then utilized to examine the anti-vasoconstriction properties of halothane. In vivo, halothane (1 MAC) produced a significant depression in the vascular response to azepexole (an alpha 2-adrenoceptor agonist), but halothane did not alter vasoconstriction by phenylephrine (an alpha 1-adrenoceptor agonist). Halothane caused a 24% reduction of maximal response (P less than 0.0001) to azepexole in pithed rats, and a 3.2-fold rightward shift of the log dose-response curve (P less than 0.0001). Similarly, in vitro, halothane significantly attenuated alpha 2- but not alpha 1-adrenoceptor responsiveness. Halothane (4%) depressed maximal vein contraction to azepexole by 26% (P less than 0.0001), and shifted the log concentration-response curve 2.4-fold to the right (P less than 0.0001). The observed selective interference with alpha 2-mediated vasoconstriction by halothane is unlikely to represent drug antagonism at the receptor level. Our observations may suggest, indirectly, that halothane interferes with Ca+2 entry into vascular smooth muscle. The phenomenon of selective anti-vasoconstriction at alpha 2-adrenoceptors by halothane may explain why alpha 1-adrenergic agonists often appear to retain their vasopressor activity during halothane anesthesia. The mechanism of halothane-induced vasodilation thus includes attenuation of alpha 2- but not alpha 1-adrenergic vasoconstriction; this further demonstrates the multifactorial nature of halothane-induced vasodilation.