Occlusion of the same coronary artery at the same anatomic site in a group of dogs results in significant variation in the size of the resultant infarcts. The present study shows that much of this variation can be explained by differences in the pattern of distribution of the occluded coronary artery. The relationship between the anatomic size of an occluded coronary bed and the the amount of necrosis resulting from 40 minutes of proximal circumflex coronary (LCC) occlusion and 3 days of reperfusion was determined in randomly selected dogs using an in vitro latex coronary injection technique. The amount of necrosis, calculated from serial histologic sections taken through the posterior papillary muscle (PP), correlated closely with the size of the occluded coronary bed: 5 dogs with small LCC beds (26 +/- 2%) had 10 +/- 3% PP necrosis; 5 dogs with medium-sized LCC beds (37 +/- 2%) had 33 +/- 1% necrosis; and 7 dogs with large LCC beds (42 +/- 1%) had 51 +/- 2% necrosis. The results show that much of the variability in infarct size among dogs subjected to coronary artery occlusion at the same anatomic site is a function of differences in ischemic bed size. Grouping dogs by occluded coronary bed size may improve the resolution of experimental studies testing the effect of various therapies on infarct size.