To investigate the importance and the causes of myocardial infarction complicating the coronary bypass operation, we have reviewed clinical and preoperative angiographic data on 717 patients operated on over an 81/2 year period and postoperative angiography on all but one survivor. The hospital mortality was 0.6%. Of the 56 (7.8%) patients who sustained perioperative myocardial infarction, two died. Left ventricular angiography supported the diagnosis in 40 patients and aided in quantification of myocardial infarction. Age, preoperative symptomatic status, incidence of prior myocardial infarction, and perfusion and axoxia timed did not correlate with myocardial infarction, but extent of coronary disease, number of grafts, and associated endarterectomy did. There have been no late deaths in the 54 survivors of perioperative infarction, which appears usually to be a benign event. Graft occlusion was nearly three times commoner in patients with myocardial infarction, and in 42 of 54 survivors the myocardial infarction could be explained by new impairment of perfusion, usually owing to compromise of the grafted artery. The relationship of perioperative infarction to iatrogenic diminution of perfusion, of the infarcted segment suggests that technical factors, especially meticulous attention to the quality of the graft-coronary artery anastomosis might further reduce the perioperative infarction rate.