A preparation is described in which feline hearts were perfused with arterial blood drawn from a blood-donor cat. Ventricular function was measured with a fluid-filled latex balloon within the left ventricle. Left ventricular developed pressure, maximum left ventricular dP/dt, ventricular compliance, and coronary blood flow changed only slightly during two hours of perfusion at a constant pressure of 75 mmHg. Water, sodium, potassium, and calcium contents, and creatine kinase activity of isolated hearts did not differ from the values obtained for intact hearts from the blood-donor cats. Isolated hearts subjected to 40 minutes of normothermic global ischemia with one hour of reperfusion exhibited significantly decreased contractility, but no change in ventricular compliance. Myocardial water and sodium contents were increased after 40 minutes of ischemia and reperfusion. Hearts subjected to 60 minutes of ischemia with reperfusion exhibited decreases in both contractility and compliance. A prolonged reactive hyperemic response was maintained throughout the hour of reperfusion. Myocardial sodium content increased more than could be accounted for by edema formation; decreases were observed in potassium content and creatine kinase activity while calcium content was increased. The applicability of this model for evaluating the effects of pharmacological interventions on myocardial ischemic injury are discussed.