The physical and mechanical effects of injecting crystalloid cardioplegic solution under various pressures and flows was studied (in canine hearts) to establish a safe method for administering it in the presence of normal coronary arteries. A constant pressure system (300 mm Hg = 15 psi) was maintained in the solution reservoir, and flows and pressures were varied with the use of cannulas of different inner diameters: 0.8, 1.35, 1.6, 2.3, 2.58, and 2.80 mm. Cardioplegia distribution was measured by 15 μm radioactive microspheres. Peak flow rate, total flow, and mean flow rate per infusion were measured by an inline electromagnetic flowmeter probe. Direct aortic root pressure, time to standstill, and myocardial temperatures were recorded by continuous monitoring. Cardiac isoenzymes were measured in the coronary sinus, peripheral blood, and directly in the myocardial tissue. Histologic changes in the left ventricle were examined by light microscopy. The results showed that the higher the flow and pressure, the shorter the prearrest period, the better flow distribution, and the faster the myocardial temperature drop. Mean aortic root pressures higher than 110 mm Hg and peak flow rates greater than 1500 ml/min caused a higher incidence of mechanical-physical trauma to the vascular endothelium and the endocardium, but cellular protection was good. Low pressure (less than 30 mm Hg) and peak flows (less than 125 ml/min) showed a higher incidence of cellular (myocardial) ischemia, focal necrosis, and uneven flow distribution. An aortic root pressure of 61 ± 5 mm Hg, a mean peak flow rate of 622 ± 52 ml/min, and a total of 600 ml for the first injection seem to offer the best cellular protection with minimal physical injury to the endothelium and endocardium for a mean canine heart weight of 236 gm.