1. We have recently reported that patients taking beta 1-adrenoceptor-selective antagonists exhibit marked sensitization of beta 2-adrenoceptor responses but unaltered beta 1-adrenoceptor responses in the heart, both in vitro and in vivo. We therefore investigated beta 1- and beta 2-adrenoceptor-mediated relaxant responses in rings of human internal mammary artery and saphenous vein without endothelium, taken from beta 1-blocked and non-beta-blocked patients undergoing coronary artery bypass graft surgery, for comparison. We also examined alpha 1-adrenoceptor-mediated contraction in these vessels, to determine whether beta 1-blockade had any cross-regulatory effect. 2. Following alpha-blockade with 10 microM phenoxybenzamine, both noradrenaline adrenaline produced concentration-dependent relaxations in both blood vessels, their effects being mediated predominantly through beta 2-adrenoceptors; a lesser beta 1-adrenoceptor component to relaxation was also found in internal mammary artery and a minor beta 1-adrenoceptor component was present in saphenous vein. No differences were found in beta 1- or in beta 2-adrenoceptor-mediated vasorelaxation between beta 1-blocked and non-beta-blocked patients. 3. Methoxamine produced concentration-dependent contractions in both blood vessels, and the potency and efficacy were not significantly different between vessels from beta 1-blocked and from non-beta-blocked patients. 4. These findings indicate that, in these tissues, which possess a relatively minor beta 1-adrenoceptor component in contrast to myocardial tissue, chronic beta 1-blocker treatment does not alter either beta 1- or beta 2-adrenoceptor responses. Likewise, in such tissues, alpha 1-adrenoceptor responses are unaffected by prior beta 1-blockade.