Ca(2+) clearance in frog motor nerve terminals was studied by fluorometry of Ca(2+) indicators. Rises in intracellular Ca(2+) ([Ca(2+)](i)) in nerve terminals induced by tetanic nerve stimulation (100 Hz, 100 or 200 stimuli: Ca(2+) transient) reached a peak or plateau within 6-20 stimuli and decayed at least in three phases with the time constants of 82-87 ms (81-85%), a few seconds (11-12%), and several tens of seconds (less than a few percentage). Blocking both Na/Ca exchangers and Ca(2+) pumps at the cell membrane by external Li(+) and high external pH (9.0), respectively, increased the time constants of the initial and second decay components with no change in their magnitudes. By contrast, similar effects by Li(+) alone, but not by high alkaline alone, were seen only on 200 stimuli-induced Ca(2+) transients. Blocking Ca(2+) pumps at Ca(2+) stores by thapsigargin did not affect 100 stimuli-induced Ca(2+) transients but increased the initial decay time constant of 200 stimuli-induced Ca(2+) transients with no change in other parameters. Inhibiting mitochondrial Ca(2+) uptake by carbonyl cyanide m-chlorophenylhydrazone markedly increased the initial and second decay time constants of 100 stimuli-induced Ca(2+) transients and the amplitudes of the second and the slowest components. Plotting the slopes of the decay of 100 stimuli-induced Ca(2+) transients against [Ca(2+)](i) yielded the supralinear [Ca(2+)](i) dependence of Ca(2+) efflux out of the cytosol. Blocking Ca(2+) extrusion or mitochondrial Ca(2+) uptake significantly reduced this [Ca(2+)](i)-dependent Ca(2+) efflux. Thus Ca(2+)-dependent mitochondrial Ca(2+) uptake and plasmalemmal Ca(2+) extrusion clear out a small Ca(2+) load in frog motor nerve terminals, while thapsigargin-sensitive Ca(2+) pump boosts the clearance of a heavy Ca(2+) load. Furthermore, the activity of plasmalemmal Ca(2+) pump and Na/Ca exchanger is complementary to each other with the slight predominance of the latter.