Previous work has shown that if pulsatile left atrial-femoral artery bypass is instituted after occlusion of the left anterior descending coronary artery for from 15 minutes to 2 hours, it can significantly limit the size of the infarct resulting 4 hours later. This study investigated whether pulsatile left atrial-femoral artery bypass begun after more clinically pertinent periods of initial ischemia can still significantly limit infarct expansion. After baseline measurements of hemodynamics, tension-time index, and regional myocardial blood flow in 73 open-chest, adult dogs, the left anterior descending coronary artery was ligated for 15 minutes or 1, 2, 4, or 6 hours of unprotected ischemia. In the five control groups, the initial ischemic period was merely extended for another 4 hours. In the five experimental groups, the animals were placed on pulsatile left atrial-femoral artery bypass for another 4 hours after the initial ischemic period. At the end of each procedure, gentian violet was used to identify the area at risk of infarction, and triphenyltetrazolium chloride was used to delineate the area of infarct. The results showed a significant reduction in the area of infarct as a percentage of the area at risk in each bypass group compared with its control group for all ischemic periods of less than 6 hours. These findings suggest that the maximum permissible ischemic time delay for myocardial salvage by pulsatile left atrial-femoral artery bypass is one which is pertinent in a clinical setting. The results justify continued attempts to develop appropriate techniques for percutaneous application of this modality to patients with an evolving myocardial infarction.