With the advent of direct bypass surgery, the dramatic clinical responses in some patients has been followed by enormous enthusiasm in many centers. Surgical technics have advanced to such a point that nearly all patients with obstructive coronary disease could have the obstructions bypassed to one or many areas. Criteria for evaluating surgery include mortality, operative infarction rate, patency of grafts, clinical response, and stress testing. Many reports fail to correlate results with angiographic studies of vein function and with completeness of revascularization. When cine studies are performed, a good correlation exists between patent veins and relief of angina and improved stress response. Different patterns of response to stress, sometimes independent of relief of angina, are obtained depending on the completeness of revascularization. Predictable relief of angina is found if revascularization is complete. At times dramatic, but much less predictable, relief of failure (improved ventricular function) follows revascularization. Angina is a valuable aid in selecting some patients for surgery. Stress testing (bicycle ergometry) can now define general groups of patients who are likely, and who are not likely, to show improved myocardial response to stress after surgery. These studies also demonstrate the need for the surgeon to provide complete revascularization whenever possible. The criteria for selection for surgery of patients with symptoms of gross heart failure remain unclear. While revascularization technics could be technically applied to nearly all coronary patients, present methods are unable accurately to define who really needs the surgery and, equally important, which hearts will respond once revascularization is completed.