A slice preparation was used to study the spread of epileptiform activity in the rat entorhinal cortex. Interictal-like discharges were induced in the medial entorhinal cortex by blocking synaptic inhibition mediated via GABAA-receptors. Recorded intracellularly, these discharges consisted of an initial paroxysmal depolarizing shift followed by a variable number of afterdischarges. There was no apparent difference between these events whether they were recorded in isolated cortical slices or in slices where the hippocampus and subicular complex remained attached. The events were also unaffected by droplets of a xylocaine solution applied to sites in the hippocampus, subicular complex or superficial layers of the entorhinal cortex but applications to layer IV/V, lateral or medial to the recording site could reduce the number of afterdischarges without affecting the initial paroxysmal shift. Simultaneous intracellular recordings from neurons in layer IV/V and layer II of the medial entorhinal cortex showed that the paroxysmal depolarizing shift and all afterdischarges in the deeper layer always preceded those recorded in the superficial layer, and these events invariably occurred on a one-to-one basis. This was true whether the events were evoked or occurred spontaneously. The delay varied between 2 and 11 ms but was consistent for a given cell pair. A similar relationship existed between discharges recorded simultaneously in layer IV/V neurons and layer VI neurons, events in the layer IV/V cells preceding those in the deeper layer. Discharges recorded simultaneously in pairs of layer IV/V neurons showed more complex relationships. Paroxysmal depolarizing shifts were always recorded in both cells and the discharge could occur at the more medial site before the more lateral, or vice versa. For a given pair the temporal relationship was invariable. It was often the case, however, that the temporal relationship between afterdischarges was reversed with respect to the initial paroxysmal shift. This relationship was also invariable in a given pair of cells. Interictal-like discharges in layers II or IV/V neurons could be abolished by perfusion with 6-cyano-7-nitro-quinoxaline-2,3-dione which is an antagonist for the non-N-methyl-D-aspartate (i.e. quisqualate/kainate) subtype of excitatory amino acid receptor. The afterdischarges associated with the events were abolished in an all-or-none fashion whereas the blockade of the paroxysmal depolarizing shift was progressive. Antagonists of N-methyl-D-aspartate receptors also abolished afterdischarges but only reduced the initial paroxysmal shift. It is concluded that the interictal-like discharges arise intrinsically within the cortex and are not influenced by input from hippocampal or subicular structures.(ABSTRACT TRUNCATED AT 400 WORDS)