“…Independently of whether the inactivation blocking site is a component of the normal gating machinery or not, it is clear that if some fraction of resting channels have this site occupied, by e.g., K, then these channels inherently have inactivation coupled to activation in that a conducting channel is necessary to clear the sites and so permit the inactivation gates to close. The nature of the coupling between activation and inactivation certainly involves more just this effect, however, as immobilization of the gating charge (Armstrong and Bezanilla, 1977;Meves and Vogel, 1977;Neumcke et al, 1976;Nonner et al, 1978;Nonner, 1980;Bullock and Schauf, 1979;Swenson, 1980;Starkus et al, 1981) establishes that the operation of the activation gates is itself dependent on the inactivation state of the channel, and because channels can also inactivate without conducting (Gillespie and Meves, 1980;Bean, 1981;Horn et al, 1981;Aldrich et al, 1983;Horn and Vandenberg, 1984). I thank Drs.…”