Male Long-Evans rats were kindled via daily electrical stimulation of the left prepyriform cortex. The animals were then used in two experiments which examined the pharmacological basis of cocaine's effects on three mutually exclusive components of the kindled seizure, which were the following: (a) latency to clonus, (b) clonus duration, and (c) duration of AD outlasting clonus. The first experiment compared the effects produced by cocaine HCl (20 mg/kg, IP), lidocaine HCl (20 mg/kg, IP), and amphetamine sulfate (2.5 mg/kg, IP). The results indicated that both cocaine and lidocaine reduced the duration of kindled AD, latency to clonus, and duration of AD persisting beyond clonus, thus suggesting that these cocaine effects are mediated by local anesthetic meechanisms. Only cocaine reduced clonus duration, which suggest that this cocaine effect is not produced by a local anesthetic action. The second experiment examined the effects of cocaine following the adminiostration of three dose levels of the monoamine antagonists haloperidol, prazosin, yohimbine, propranolol, or metergoline (selected for their ability to block dopamine, alpha-1-norepinephrine, alpha-2-norepinephrine, beta-norepinephrine, and serotonin receptors, respectively). The results of this experiment found no support for a monoaminergic contribution to cocaine's effect on clonus latency or AD after clonus. However, results for prazosin, which reduced clonus duration and exhibited an additive effect with cocaine on this variable, suggest that cocaine's norepinephrine action (especially on the alpha-norepinephrine systems) may modulate clonus duration.