This study was carried out to test the hypothesis that adrenergic coronary vasoconstriction limits blood flow to hypoperfused regions of myocardium during exercise. The vasoconstrictor influence of α-adrenergic receptor subtypes was assessed by use of selective adrenergic blocking agents. Dogs chronically instrumented with a circumflex coronary artery hydraulic occluder and an intra-arterial catheter underwent treadmill exercise in the presence of a coronary stenosis that decreased distal perfusion pressure to 40 mm Hg. Myocardial blood flow was measured with radioactive microspheres (15 μm) before and during selective α1- or α2afcadrenergic receptor blockade produced by intracoronary infusion of prazosin (1 /tg/kg/min×lO min) or idazoxan (1 μg/kg/min × lO min), respectively. Coronary perfusion pressure was held equal before and during receptor blockade with the hydraulic occluder. Compared with control exercise, subendocardial blood flow increased during α2-receptor blockade with prazosin from 0.60 ± 0.14 to 1.12 ± 0.17 ml/min/g (p<0.05), and mean transmural flow increased from 1.07 ± 0.19 to 1.60 ± 0.22 ml/min/g (p<0.05). In contrast, subendocardial and mean transmural blood flow were not different from control during selective α2-adrenergic receptor blockade with idazoxan (0.48 ± 0.10 vs. 0.67 ± 0.14 ml/min/g, p=0J3, and 0.82 ± 0.15 vs. 1.02 ± 0.2© ml/min/g, p-Q.45, respectively). These data indicate that even in the presence of a coronary stenosis that causes substantial myocardial underperfusion during exercise, residual coronary vasoconstrictor tone is present in iscbemic myocardium, and this vasoconstriction is mediated predominantly by the α1-adrenergic receptor.