Previous studies have demonstrated that post-junctional alpha(1)- and alpha(2)-adrenoceptors mediate vasoconstriction in the human forearm. However, the relative contributions of the alpha-adrenoceptor subtypes to basal limb vascular tone are unknown. In healthy young men, forearm blood flow (FBF; venous occlusion plethysmography) responses to brachial artery administration of prazosin (an alpha(1)-adrenoceptor antagonist), yohimbine (an alpha(2)-adrenoceptor antagonist) and phentolamine (a non-selective alpha-adrenoceptor antagonist) were determined after local beta-adrenoceptor blockade with propranolol. In 10 subjects, prazosin increased FBF from 2.4 +/- 0.3 to 5.8 +/- 1.0 ml (100 ml)(-1) min(-1) (approximately 140 %; P < 0.001 vs. baseline). Subsequently, phentolamine further increased FBF to 11.7 +/- 1.6 ml (100 ml)(-1) min(-1) (approximately 385 %; P < 0.001 vs. baseline). Thus, the average calculated increase in FBF due to removal of alpha(2)-vasoconstrictor tone was greater than that due to removal of alpha(1)-tone (5.9 +/- 0.8 vs. 3.4 +/- 0.8 ml (100 ml)(-1) min(-1); P < 0.01) and represented approximately 63 % of basal sympathetic tone. Complete alpha(1)-adrenoceptor blockade was confirmed by a minimal reduction in FBF in response to phenylephrine after prazosin (46 +/- 3 vs. 6 +/- 4 %; before vs. after blockade) and in a separate group of four subjects, increasing the dose of prazosin threefold did not evoke further forearm vasodilatation. Additionally, the reduction in FBF in response to tryamine (evokes endogenous noradrenaline release) was abolished after phentolamine (40 +/- 3 vs. 2 +/- 1 %; before vs. after blockade), documenting complete pharmacological sympathectomy. In another group of seven subjects, administering yohimbine prior to phentolamine resulted in similar findings. These observations indicate that vasoconstricting post-junctional alpha(2)-adrenoceptors contribute more to basal vascular tone than alpha(1)-adrenoceptors in the forearms of young healthy men. The potential physiological and pathophysiological implications of these findings are discussed.