1. Recording with glass micropipette electrodes inserted close to the synaptic region, in the presynaptic and in the postsynaptic fibers of the giant synapse in the stellate ganglion of the squid, has been accomplished.
2. The forms of the spike and of the synaptic potential are very much like those reported earlier (Bullock, 1948) from macroelectrodes. The crest time and the rate of fall are labile and depend on the state of fatigue, though the time of initiation of the postsynaptic potential does not.
3. It is concluded after examination of both intra- and extracellular recordings that there is a real synaptic delay of the order of 1 or 2 milliseconds at 15–20°C.
4. There is sometimes a very small and sometimes no visible deflection in the intracellular postsynaptic record attributable to the presynaptic spike. It is concluded that transmission cannot be electrical.
5. The amplitude of the postsynaptic potential can be controlled over some range by the amplitude of the presynaptic potential.
6. Hyperpolarization of the postsynaptic membrane results in increase in amplitude of spikes up to 200 millivolts, in increase in the membrane potential level at which the spike flares up, but in no considerable change in the amplitude in postsynaptic potential.
7. The postsynaptic potential can add to the late falling phase and the undershoot of an antidromic spike in the postfiber but cannot add to the crest or early part of the falling phase. The earliest part of the antidromic spike during which the postsynaptic potential can add is probably a period of refractoriness to electrical shock by analogy with the properties of the axon.