To determine if, during elective cardiac arrest, the myocardial protection afforded by a cold (4° C) crystalloid potassium cardioplegic solution could be improved by oxygenation of the solution, we placed 16 dogs on cardiopulmonary bypass and subjected their hearts to 4 hours of cold cardioplegic arrest. Group I hearts (n = 8) received aerated crystalloid solution perfused through the aortic root every 20 minutes. Group 2 hearts (n = 8) were treated identically except that the crystalloid cardioplegic solution was fully oxygenated. Left ventricular function curves (ejecting heart) were generated before arrest (control) and after 45 minutes of reperfusion. A cardiac output of 1,0000 ml/min could be attained in only two hearts of Group I after reperfusion, whereas all but one heart of Group 2 had excellent functional preservation. Mean postreperfusion adenosine triphosphate (ATP) levels in Group 1 and Group 2 hearts were 62% and 89% of control, respectively (p <0.01). Myocardial water content had increased significantly (p <0.002) after reperfusion in Group 1, but not in Group 2. During cardioplegic solution infusion, myocardial oxygen consumption (MV̇O2) was 1.42 ± 0.15 ml O2/min/100 gm LV for Group 1 and 6.91 ± 1.27 ml O2/min/100 gm LV for Group 2 (p <0.001). Oxygen consumed per minute of arrest was 0.027 ± 0.003 ml O2/min/100 gm LV for Group 1 and 0.128 ± 0.015 ml O2/min/100 gm LV for Group 2 (p <0.001). Postreperfusion ultrastructural evaluation of two of the Group 1 hearts revealed severe ischemic damage in contrast to the normal ultrastructural appearance of two of the Group 2 hearts. With careful attention given to maintenance of myocardial hypothermia and cardioplegic delivery methods, the myocardial protection afforded by an oxygenated crystalloid cardioplegic solution exceeds that provided by the aerated control and compared favorably with other methods of myocardial protection during ischemic arrest.