The effect of nifedipine, 0.010 mg/kg intravenously, on myocardial blood flow was studied in 15 dogs 4 weeks after placement of an Ameroid constrictor on either the left circumflex or left anterior descending coronary artery to produce total coronary occlusion. Myocardial blood flow was measured with radionuclide-labeled microspheres at rest and during two levels of treadmill exercise to acheive a heart rate of 190 (light exercise) and 230 (heavy exercise) beats/min. During control conditions, increasing exercise resulted in a progressive increase in myocardial blood flow in normally perfused areas, but was associated with worsening subendocardial hypoperfusion in collateral-dependent areas. Nifedipine administration resulted in a transient reduction of arterial pressure and an increase in heart rate. To determine whether nifedipine exerted significant persistent effects on the coronary collateral circulation, measurements of myocardial blood flow were repeated beginning 30 minutes after nifedipine administration, at a time when heart rate and arterial pressure had returned to control levels. In normally perfused areas, nifedipine did not significantly alter myocardial blood flow at rest, but increased mean myocardial blood flow from 2.06 ± 0.15 to 2.40 ± 0.20 ml/min per g during light exercise (p < 0.01), while blood flow during heavy exercise was not significantly altered. In collateral-dependent myocardial areas, the volume and transmural distribution of myocardial blood flow were not significantly altered after nifedipine administration either at rest or during exercise. These results fail to demonstrate persistent vasodilation of the coronary collateral vessels after the systemic hemodynamic effects of nifedipine have subsided.