We studied the local anesthetic effects of the quaternary lidocaine analogues QX-314, QX-572, and QX-222, the tertiary amine lidocaine, its analogues tocainide, 6603, 6211, and the neutral local anesthetic benzocaine to determine if molecular charge of antiarrhythmic agents influences their local anesthetic effects on heart fibers. We used standard microelectrode techniques and canine cardiac Purkinje fibers to compare the effects of stimulation rate, drug concentration, and K+-induced changes in resting membrane potential on the reduction of fast inward sodium current using the maximum rate of rise of the action potential upstroke, Vmax, as an index of changes in peak sodium current. Use-dependent block, defined as a modulation of the reduction in Vmax by local anesthetics due to changes in the stimulation rate, was observed with the permanently charged analogues and was most prominent for agents existing predominantly in the charged form, but was absent for the neutral local anesthetic benzocaine. The development of use-dependent block during rapid stimulation preceded by prolonged periods of quiescence was an exponential process which became more rapid with increasing drug concentration. Recovery from use-dependent block during quiescence was an exponential process that was not influenced by similar drug concentration changes. All local anesthetics caused tonic block, defined as a drug-induced reduction of Vmax from control that attained a constant value at slow stimulation rates (cycle length range 15 seconds to 2 minutes) and was not changed by prolonged (up to 8 minutes) periods of quiescence. These findings suggest that the charged form of lidocaine and its analogues is responsible for use-dependent block of cardiac sodium channels.