1. When molluscan neural somata are filled with the calcium-indicating photo-protein aequorin and subjected to a 1 Hz train of depolarizing pulses (0.3 sec duration to + 15 mV) under voltage clamp, the successive photo-emissions due to calcium influx facilitate. The origin of this phenomenon was investigated in identified neurones from the abdominal ganglion of Aplysia californica.2. Since outward currents inactivate cumulatively in successive pulses, the effective depolarization increases due to a series resistance error. Elimination of this error by electronic compensation or pharmacological block of outward current reduced aequorin response facilitation by only about 30%, on the average.3. When voltage-dependent sodium and potassium currents are blocked in tetraethylammonium (TEA)-substituted zero-sodium sea water, the remaining inward calcium currents display no facilitation. On the contrary, a slow decline during a pulse and a slight progressive depression in successive pulses are observed. Barium-substitution for calcium in the same medium eliminates a small residual potassium current insensitive to external TEA. The remaining inward barium currents also display depression instead of facilitation.4. A non-pharmacological separation of calcium current was accomplished by measuring tail currents at the potassium equilibrium potential following depolarizing pulses. Calcium tail currents activate rapidly and then decline gradually and incompletely as depolarizing pulse duration is lengthened. Tail currents also show no evidence of facilitation; there is instead a slight depression of currents after successive pulses.5. Increments of optical absorbance in neurones filled with the calcium-sensitive dye arsenazo III show a depression rather than facilitation to successive depolarizations in a train. The time course of these absorbance signals is consistent with the time-dependent depression of calcium current.6. Calibration of arsenazo III response amplitude indicates that the dye reports only about 1% of the calcium concentration increment expected from knowledge of cell volume and the charge carried by calcium current during a depolarizing pulse. This suggests that cytoplasmic buffering of free calcium must occur rapidly, on a time scale comparable to the response time of arsenazo III (about 1 msec) or more rapidly.7. The slow potassium tail current following a depolarizing pulse is calcium-dependent and probably provides an approximate index of the internal sub-membrane calcium concentration. Increments in this current after repetitive pulses display a slight progressive depression rather than facilitation.8. Since neither calcium currents nor the concentration transients show facilitation, we conclude that aequorin response facilitation is due to the non-linear dependence of aequorin photo-emissions on calcium concentration. This conclusion is supported by a finding that the very different kinetics of arsenazo III responses and aequorin responses can be reconciled by a simple model representing calcium accumulation and known response properties of the two indicator substances.9. In a train of impulses evoked by injecting depolarizing current into a neurone, the successive action potentials grow in duration. Nevertheless, a nearly constant calcium influx signalled by arsenazo III accompanies broadening action potentials.