The times required for a steady rate of miniature end-plate potential discharge to be reached in response to changes in extracellular [K(+)], [Na(+)], and [Ca(++)] have been measured. In the presence of 15 mM KCl, Ca(++) raises and Na(+) lowers the steady-state mepp frequency; but the depressive effect on Na(+) is not specific: Li(+) can replace Na(+) to a large extent. Mepp frequency has been found to depend on the ratio of [Ca(o) (++)]/[Na(o) (+)]. It is assumed that in the steady state, intracellular sodium will change when extracellular sodium is changed. Because both intracellular and extracellular sodium at motor nerve endings affect acetylcholine release, it is proposed that mepp frequency depends on the ratio [Ca(o)] [Na(i)](2)./[Na(o)](2) Two models are proposed. Firstly, to account for the action of sodium and calcium a carrier is postulated for which Ca(++) and Na(+) compete. The carrier determines a maximum level of intracellular Ca(++) far lower than predicted by the Nernst equation for Ca. Secondly, to account for activation of acetylcholine release by a small influx of Ca(++), the ions are presumed to enter the nerve ending in a two stage process through a small intermediate compartment and to act on the acetylcholine release site in this region rather than after entering directly into the cell.