Repeated brief episodes of total coronary artery occlusion (i.e., severe ischemia), each separated by brief periods of reperfusion, reduce infarct size after a subsequent sustained ischemia. The importance of the intensity of ischemia during these coronary artery occlusions and the role of the following transient reflow are poorly understood. Therefore, our objective was to determine whether moderate preconditioning ischemia induced by partial coronary artery stenosis (reducing coronary flow to approximately 50% of its baseline values), with or without a brief period of total reperfusion, could precondition the canine myocardium. Dogs were randomized to receive one of three preconditioning “treatments”: the R(−) group underwent 15 minutes of partial coronary stenosis without subsequent brief reperfusion (n=8); the R(+) group underwent 15 minutes of partial coronary stenosis followed by 10 minutes of full reflow (n=8); and the control group underwent no intervention (n=8). All dogs then underwent 1 hour of total coronary artery occlusion and 4.5 hours of reperfusion. Both treated groups were equally and moderately ischemic during partial stenosis: myocardial blood flow in the inner two thirds of the left ventricular wall averaged 0.25±0.05 and 0.31±0.07 ml/min per gram in the R(−) and R(+) groups, respectively (p=NS). Furthermore, all three groups were equally and severely ischemic during sustained total occlusion: myocardial blood flow in the inner two thirds of the left ventricular wall averaged 0.06±0.05, 0.05±0.03, and 0.07±0.03 ml/min/g in control, R(−), and R(+) groups, respectively (p=NS). Infarct size (expressed as percentage of the area at risk) in the R(+) group was 9.0±2.5%, which was significantly smaller (p<0.01) than the value of 22.8±5.5% observed in control animals. Stenosis followed by full reflow in the R(+) group did not, however, attenuate contractile dysfunction during the sustained total coronary artery occlusion and did not preserve function in peri-infarct tissue after reperfusion. In contrast, stenosis without reperfusion in the R(−) group did not limit infarct size (28.4±5.4%,p=NS versus control) and did not preserve wall motion during total occlusion/reperfusion. Therefore, we conclude that 1) partial coronary artery stenosis can precondition the heart, 2) complete reflow after the preconditioning ischemia is mandatory to induce myocardial protection, and 3) preconditioning with moderate ischemia does not preserve contractile function.