This report describes studies of anticonvulsants for the organophosphorus (OP) nerve agent soman: a basic research effort to understand how different pharmacological classes of compounds influence the expression of seizure produced by soman in rats, and a drug screening effort to determine whether clinically useful antiepileptics can modulate soman-induced seizures in rats. Electroencephalographic (EEG) recordings were used in these studies. Basic studies were conducted in rats pretreated with HI-6 and challenged with 1.6 x LD50 soman. Antimuscarinic compounds were extremely effective in blocking (pretreatment) or terminating soman seizures when given 5 min after seizure onset. However, significantly higher doses were required when treatment was delayed for more than 10 min, and some antimuscarinic compounds lost anticonvulsant efficacy when treatment was delayed for more than 40 min. Diazepam blocked seizure onset, yet seizures could recur after an initial period of anticonvulsant effect at doses =2.5 mg/kg. Diazepam could terminate ongoing seizures when given 5 min after seizure onset, but doses up to 20 mg/kg were ineffective when treatment was delayed for 40 min. The GABA uptake inhibitor, tiagabine, was ineffective in blocking or terminating soman motor convulsions or seizures. The glutamate receptor antagonists, NBQX, GYKI 52466, and memantine, had weak or minimal antiseizure activity, even at doses that virtually eliminated signs of motor convulsions. The antinicotinic, mecamylamine, was ineffective in blocking or stopping seizure activity. Pretreatment with a narrow range of doses of alpha2-adrenergic agonist, clonidine, produced variable protection (40-60%) against seizure onset; treatment after seizure onset with clonidine was not effective. Screening studies in rats, using HI-6 pretreatment, showed that benzodiazepines (diazepam, midazolam and lorazepam) were quite effective when given 5 min after seizure onset, but lost their efficacy when given 40 min after onset. The barbiturate, pentobarbital, was modestly effective in terminating seizures when given 5 or 40 min after seizure onset, while other clinically effective antiepileptic drugs, trimethadione and valproic acid, were only slightly effective when given 5 min after onset. In contrast, phenytoin, carbamazepine, ethosuximide, magnesium sulfate, lamotrigine, primidone, felbamate, acetazolamide, and ketamine were ineffective.