ABSTRACT. Eleven children with typical absence seizures were studied clinically and by repeated 24-h EEGs with portable cassette tape recorder before and during anticonvulsant treatment. The history, the observation of seizures and the amount of spike-and-wave activity (episodes >3 sec) were studied in relation to the plasma levels. The EEG revealed spike-and-wave episodes, considered as seizures, in a higher frequency than the number of absences reported by the parents. Twenty-four hour EEG recording was however not superior to long term observation by relatives in establishing freedom from absences. The drug primary used was ethosuximide. Eight of eleven children responded completely on this drug, i.e. no absences observed and EEG was normalized. Of two children in whom the therapy was changed to sodium valproate, one responded completely and one partly. The relation between the ethosuximide dosage given and the plasma concentration level was good. Two cases responded at low plasma levels (<200 μmol/1). Tonic-clonic seizures occurred in 3/11 children.