Gating current (Ig) has been studied in relation to inactivation of Na channels. No component of Ig has the time course of inactivation; apparently little or no charge movement is associated with this step. Inactivation nonetheless affects Ig by immobilizing about two-thirds of gating charge. Immobilization can be followed by measuring ON charge movement during a pulse and comparing it to OFF charge after the pulse. The OFF:ON ratio is near 1 for a pulse so short that no inactivation occurs, and the ratio drops to about one-third with a time course that parallels inactivation. Other correlations between inactivation and immobilization are that: (a) they have the same voltage dependence; (b) charge movement recovers with the time coures of recovery from inactivation. We interpret this to mean that the immobilized charge returns slowly to "off" position with the time course of recovery from inactivation, and that the small current generated is lost in base-line noise. At -150 mV recover is very rapid, and the immobilized charge forms a distinct slow component of current as it returns to off position. After destruction of inactivation by pronase, there is no immobilization of charge. A model is presented in which inactivation gains its voltage dependence by coupling to the activation gate.