One hundred and nineteen consecutive patients were studied prospectively after uncomplicated myocardial infarction by maximal exercise electrocardiography at two weeks and coronary angiography at six weeks. At angiography 87 patients had a stenosis greater than 70% in one major coronary artery supplying residual viable myocardium outside the infarction zone. In 82 (94%) of these the 12 lead maximal exercise electrocardiogram correctly identified these areas of ischaemic but viable myocardium. Based on ST criteria alone five patients had a false negative exercise electrocardiogram for additional disease. Nevertheless, three developed angina or a significant fall in systolic blood pressure or both at a low workload. On the basis of the anatomical lesions, symptoms, and the results of the European Coronary Surgery Study Group 55 patients were allocated for surgery. Of these, 54 underwent coronary artery bypass grafting within three months of myocardial infarction. One patient died perioperatively and another died after a reinfarction at four months while awaiting surgery. The remaining 53 were symptom free during a mean follow up period of 37 months. Sixty four patients received medical treatment. At angiography 32, 24, and eight patients had one, two, and three vessel coronary disease respectively. The exercise electrocardiogram correctly predicted the anatomy in 60 (94%), with two false positive and two false negative results for additional disease. The eight patients with three vessel disease treated medically had generalised inoperable disease, and at follow up three had died after a further infarction and five remained symptomatic with full medical treatment. Thus of those designated as at high risk and considered suitable for surgery the 37 month survival was 53 or 54 patients treated surgicall7y.