The effect of injecting ethanol directly into the myocardium to control aconitine-induced ventricular tachycardia (VT) was evaluated in anesthetized dogs. In 17 dogs, VT was induced by injecting aconitine (1.0 microgram, median dose) directly into the epicardium. After inducing persistent VT for up to 3 min, 0.6 ml (median volume) of 96% ethanol was injected into the same epicardial region. Regular sinus rhythm reappeared in 15 dogs with no change in systolic blood pressure; the other 2 dogs died of ventricular fibrillation (VF). In another 13 dogs, VT was induced by injecting aconitine directly into the endocardium using a Variocath needle catheter. After persistent VT for up to 3 min, a regular sinus rhythm was restored in 7 dogs by injecting 2.0 ml (median volume) of 96% ethanol; the remaining 6 dogs died of VF. Histology showed no transmural necrosis and the subendocardial necrotic areas were essentially the same in the dogs that recovered from VT as in those that died. There was no statistically significant relationship between doses of ethanol and VT duration. These preliminary results suggest that the injection of ethanol into the myocardium may efficiently terminate VT when other techniques fail.