The cellular electrophysiologic consequences of both regional and global experimental ischemia and reperfusion were studied in the isolated cat myocardium, using conventional microelectrode techniques. Oxygenated Tyrode's solution was perfused through the left anterior descending and circumflex coronary arteries, while the preparation was superfused with Tyrode's solution gassed with 95% nitrogen and 5% carbon dioxide. Electrophysiologic characteristics of endocardial muscle cells were normal during coronary perfusion. When perfusion was discontinued for 30 minutes, resting membrane potential was decreased by 21.6 +/- 4.1%, action potential amplitude was decreased by 29.1 +/- 8.6% and action potential duration was decreased by 54.1 +/- 12.5% (p less than 0.001). Ectopic activity occurred after 5 to 10 minutes of ischemia and was more frequent in regional than in global ischemia (p less than 0.05). Rapid ventricular activity was observed in only 5 (17%) of 29 preparations during ischemia, whereas it occurred in 24 (83%) of 29 preparations during reperfusion. Rapid ventricular activity began 5 to 40 seconds (mean 19) after the start of reperfusion, stopped spontaneously after a mean of 113 +/- 211 seconds and occurred after both regional and global ischemia. The cellular electrophysiologic changes induced by ischemia returned to baseline values within the next 5 minutes. Repeated ischemia and reperfusion runs reproduced the same electrophysiologic changes and rapid ventricular activity. Coronary perfusion with procainamide (20 mg/liter) aggravated the ischemic depressions of action potential amplitude and action potential duration and increased conduction delay during ischemia, but it did not prevent rapid ventricular activity induced by reperfusion. In contrast, verapamil (1 mg/liter) perfusion did not affect the changes in action potential variables during ischemia but prevented reperfusion-induced rapid ventricular activity. Perfusion with calcium ion (Ca2+)-free Tyrode's solution just before ischemia and during reperfusion slowed or prevented reperfusion-induced rapid ventricular activity, without affecting the action potential changes during ischemia. It is concluded that, in these isolated perfused ventricular muscle preparations, different mechanisms may be operative in ischemic and reperfusion arrhythmias and Ca2+ may play an important role in the development of arrhythmias during the reperfusion phase of ischemia/reperfusion sequences.