In rats with cobalt implanted in the right frontal cerebral cortex, acetylcholine (ACh) levels were depressed in the visually non-necotic, surrounding cortex at 7 and 14 days after surgery in comparison with values for controls treated with glass. At 21 days post-implantation, ACh levels were not different for glass and cobalt treatments. Effects of drugs affecting cholinergic function on electro-corticographic (ECoG) epileptiform activity were determined in rats implanted bilaterally with cobalt. The cholinesterase inhibitors, physostigmine and diisopropylfluorophosphate reduced both seizure activity and interictal spiking in these cobalt-treated rats. Hemicholinium-3 (HC-3), given subacutely initially inhibited seizures, but seizure frequency increased later during treatment. HC-3 did not appear to inhibit interictal spiking. These results suggest an involvement of brain cholinergic system in chronic cobalt experimental epilepsy. Seven days after cobalt implantation, HC-3 was less effective in depleting ACh in cerebral cortex adjacent to the cobalt-lesion than in similar tissue from rats with no cobalt implants. This suggests that the cholinergic neurons adjacent to the implant are not highly active at a time when seizure frequency is maximal.