1. The axo-axonic giant synapse in the stellate ganglion of the squid has been used to study synaptic transmission.2. When nerve impulses have been eliminated with tetrodotoxin, synaptic transfer of potential changes can still be obtained by applying brief depolarizing pulses to the presynaptic terminal.3. Suitably matched pulses are as effective as the normal presynaptic spike in evoking post-synaptic potentials. The synaptic delay and the time course of the post-synaptic potential are very similar to that in the normal preparation.4. The synaptic transfer (input/output) characteristic has been studied under different experimental conditions. With brief (1-2 msec) current pulses, post-synaptic response becomes detectable when the presynaptic depolarization exceeds about 30 mV. The post-synaptic potential increases about tenfold with 10 mV increments of presynaptic depolarization.5. Calcium increases, magnesium reduces the slope of the synaptic transfer curve. The influences on this curve of (i) duration of the pulse, (ii) preceding level of membrane potential, (iii) position of recording electrode, (iv) rate of repetitive stimulation are described.6. After loading the synaptic terminal with tetraethylammonium ions, large inside-positive potentials can be produced in the terminal and maintained for many milliseconds.7. By raising the internal potential to a sufficiently high level, synaptic transfer becomes suppressed during the pulse, and the post-synaptic response is delayed until the end of the pulse.8. This observation is in accord with a prediction of the ;calcium hypothesis', viz. that inward movement of a positively charged Ca compound, or of the calcium ion itself, constitutes one of the essential links in the ;electro-secretory' coupling process of the axon terminal.