The effects of seven Class-I antiarrhythmic drugs on the maximum upstroke velocity (Vmax) of action potential were examined in isolated guinea pig ventricular muscles in order to characterize their use- and state-dependent sodium channel blocking action. From the onset and offset kinetics of the use-dependent Vmax inhibition during stimulation trains, the seven drugs were subdivided into two groups; fast drugs (lidocaine, mexiletine, and tocainide), and slow drugs (quinidine, aprindine, disopyramide and flecainide). In experiments to assess the state-dependent sodium channel block, a conditioning clamp pulse to 0 mV was applied by using the single sucrose-gap voltage-clamp technique, and the Vmax of test action potential 100 msec after the clamp pulse was measured. The decrease in Vmax by 10 msec clamp pulse was defined as the activated channel block (ACB), and the decrease in Vmax as the clamp pulse duration was prolonged from 10 to 500 msec was defined as the inactivated channel block (ICB). The ratio of ICB to ACB was less than 1.0 for quinidine, disopyramide and flecainide, and much greater than 1.0 for aprindine, lidocaine, mexiletine, and tocainide. These characteristics may contribute to the differences in efficacy of each drug in treating various types of arrhythmias.