We evaluated antiarrhythmic effects of N-696 using canine digitalis ventricular arrhythmia models and electrophysiological actions of N-696 using canine ventricular muscles and isolated blood perfused A-V node preparations and compared the effects with those of propranolol. 1) N-696 at 30 mg/kg, i.v. was not effective on digitalis arrhythmia, but reduced the total heart rate and atrial rate for 60 min and transiently decreased the blood pressure. Propranolol at 3 mg/kg, i.v. transiently suppressed digitalis ventricular arrhythmia and decreased the total heart rate, atrial rate and blood pressure. The minimum effective plasma concentration of propranolol was 1.7 +/- 0.4 microgram/ml. 2) Ten to 100 micrograms/ml N-696 and 3 to 10 micrograms/ml propranolol did not affect the resting potential and the action potential duration at 75% repolarization of the canine ventricular muscle, but decreased the maximum rate of rise of the action potential in a dose-dependent fashion, and 100 micrograms/ml N-696 decreased significantly the action potential amplitude. The minimum effective drug concentration of N-696 was 60 micrograms/ml, and that of propranolol was 6 micrograms/ml. 3) N-696 (30 micrograms-3mg) dose-dependently delayed the A-V conduction time of the isolated blood perfused A-V node preparation when it was administered into the posterior septal artery (PSA) and the anterior septal artery (ASA). Propranolol (10-600 micrograms) also had similar effects, but the doses of N-696 delaying the A-V conduction time by 15% was about 8 to 21 times higher than those of propranolol.