1. Since the inhibitory effect of direct or indirect cortical stimulation on cortical units can be overcome by excitation with even more L-glutamate, it is not likely to be due to an excessive depolarization.2. Further evidence that surface stimulation has a hyperpolarizing action on cortical cells was obtained by intracellular recording from over 120 pericruciate cells. Inhibitory post-synaptic potentials (IPSPs) are seen in most cells, which are comparable in threshold and duration with the inhibitory effect observed extracellularly. The IPSPs are usually not preceded by a discharge of the same cells.3. The extracellular slow wave corresponding to the inhibitory effect varies considerably with different preparations and different depths within the cortex. A predominantly positive wave is only seen occasionally. In general, the relevant wave recorded deep in the cortex tends to be mainly negative.4. This negative slow wave can be much potentiated by tetanic stimulation, or, especially, by a large local release of L-glutamate; the last procedure is most effective either very near the surface, or below a depth of 1.0 mm. These observations suggest that inhibitory synapses occur more profusely in the superficial half of the grey matter.5. Unlike L-glutamate, GABA tends to depress the ;inhibitory' slow wave.6. The inhibitory effect must be produced by intracortical neurones, since it is fully preserved in isolated cortical slabs. In both acute and chronic slabs, the inhibition is particularly well marked and long lasting, partly because spontaneous activity and the usual post-inhibitory rebound of excitability are absent.7. The intracortical pathways responsible for the spread of inhibition cannot be situated mainly in the superficial layers, as they are not readily blocked by surface cooling or the application of local anaesthetics.8. One can record unit discharges immediately after a surface shock. Some of these discharges could be from inhibitory interneurones, but they do not last more than 10-20 msec.9. We conclude from the observations described in this and a previous paper (Krnjević, Randić & Straughan, 1966a) that a widespread system of intracortical interneurones can be activated by direct or indirect stimulation of the cortex; these interneurones have a powerful and prolonged inhibitory action on most cortical cells.10. The identity and distribution of the postulated inhibitory interneurones is discussed in the light of some relevant morphological evidence.