Lower limb blood flow, oxygen uptake, and femoral vein 02 content were measured at rest and during maximal bicycle exercise, performed with two legs and one leg, in four normal subjects and in five patients with severe congestive heart failure. While in normal subjects femoral vein blood flow and lower limb vascular conductance were significantly greater during one-leg exercise than during two-leg exercise (6084 + 745 vs 5370 + 803 ml/min, p < .05, and 52.3 8.0 vs 45.1 + 8.2 U x 1o3, p < .05, respectively), in patients with severe congestive heart failure these values were similar during the two forms of exercise (1082 ± 459 vs 1053 ± 479 m/min and 9.6 ± 3.7 vs 9.4 + 3.5 U x 103, respectively). In five additional patients, one-leg maximal bicycle exercise was performed before and after administration of phentolamine into the femoral artery of the active leg. Regional a-adrenergic blockade with phentolamine did not alter maximal oxygen uptake attained during one-leg bicycle exercise (9.8 ± 1.5 vs 10.3 ± 1.9 ml/kg). Lower limb blood flow and femoral vein 02 content attained during maximal one-leg exercise were also similar before and after phentolamine. Thus, in contrast with normal subjects, patients with severe congestive heart failure were unable to further increase limb blood flow during one-leg bicycle exercise. Moreover, local a-adrenergic blockade does not augment blood flow to the active limb during maximal one-leg bicycle exercise. This suggests that the ability of the muscular vasculature to vasodilate during exercise is impaired and may be a limiting factor to maximal exercise capacity in such patients.