Charge movement was measured in frog cut twitch fibers with the double Vaseline gap technique. Five manipulations listed below were applied to investigate their effects on the hump component (Iv) in the ON segments of TEST minus CONTROL current traces. When external CI-was replaced by MeSO~ to eliminate CI current, I v peaked earlier due to a few millivolts shift of the voltage dependence of I, kinetics in the negative direction. The Q-V plots in the TEA.CI and TEA.MeSO~ solutions were well fitted by a sum of two Boltzmann distribution functions. The more steeply voltage-dependent component (Q,) had a F ~ 6 mV more negative in the TEA.MeSO s solution than in the TEA.CI solution. These voltage shifts were partially reversible. When creatine phosphate in the end pool solution was removed, the I v hump disappeared slowly over the course of 20-30 min, partly due to a suppression of Qv' The hump reappeared when creatine phosphate was restored. When 0.2-1.0 mM Cd 2+ was added to the center pool solution to block inward Ca current, the I v hump became less prominent due to a prolongation in the time course of I v but not to a suppression of Qv" When the holding potential was changed from -90 to -120 mV, the amplitude of Ia was increased, thereby obscuring the I v hump. Finally, when a cut fiber was stimulated repetitively, I, lost its hump appearance because its time course was prolonged. In an extreme case, a 5-rain resting interval was insufficient for a complete recovery of the waveform. In general, a stimulation rate of once per minute had a negligible effect on the shape of I v. Of the five manipulations, MeSO~ has the least perturbation on the appearance of I v and is potentially a better substitute for C1-than SO~-in eliminating CI current if the appearance of the I v hump is to be preserved.