Quantification of intramyocardial hemorrhage was performed in 69 pigs submitted to various protocols of coronary artery occlusion and reperfusion. The study groups include 1) permanent occlusion; 2) reperfusion after periods of coronary occlusion of 30, 45, 60, 90, and 120 minutes; 3) reperfusion with diltiazem and with 4) methoxamine after a 60-minute occlusion period; and 5) permanent reocclusion after a 30-minute period of reperfusion. Red blood cell counts were directly assessed by visual examination of histologic slices of myocardium and in a subgroup of animals by counts of red blood cells labeled with 99m-technetium pertechnetate. Hemorrhage occurs in infarcts reperfused after a duration of 45 minutes or more of coronary occlusion and after a period of reperfusion maintained for at least 30 minutes. Red blood cell counts were maximal in the mid portions of transmural sections of the infarcts, with decreasing values toward epicardium and endocardium. Diltiazem decreased total red blood cell counts, whereas methoxamine increased it and also caused subendocardial hemorrhage. The most powerful predictors of the severity of hemorrhage after sustained reperfusion were infarct size and higher blood pressure.