Basic rhythms of EEG were quantitatively analyzed in 92 adult idiopathic epileptic patients and in 57 age-matched control subjects. Epileptic patients have been administered one or both of anticonvulsants, phenobarbital (PB) and diphenyl-hydantoin (DPH). Before the EEG study, blood samples were taken from patients and serum concentration of PB and DPH was examined with UV method. Dose and serum level correlation was statistically significant in both drugs but fairly inferior in DPH. Dominant posterior rhythms in awake subjects were restored in a medical computer. These data were transformed with a Fast Fourier method into amplitude spectra, which were expressed in percentage distribution of each frequency band, i.e. delta (1.0--3.5 Hz), theta (4.0--7.5), alpha (8.0--12.5) and beta activity (13.0--29.5 Hz). As the first result, it was found that slow wave activity significantly increased and alpha activity decreased in epileptics compared with control subjects. No difference was found in beta activity. So as the next step, the etiologies of abnormal slowness of epileptic EEG were investigated in relation to serum level of anticonvulsants, age of seizure onset, seizure frequency, duration of medication and seizure types. These analysis confirmed that slowing of epileptic EEG has statistically significant correlation with the following factors, 1) high serum level of anticonvulsants (PB greater than 20 microgram/ml), DPH greater than 15 microgram/ml) 2) high frequency of seizures (more than once per month) 3) seizure onset at young ages (0--5 yrs.) 4) psychomotor seizure type. When these slowing factors are plurally involved in one patient, the degree of EEG slowness increased in proportion to the numbers of these factors. On the other hand, if the comparison was made between epileptics without these factors and control subjects, strikingly no minimum differences was found in basic rhythms. This result induces the very important conclusion that epilepsy itself has no direct effect on slowing of basics rhythms and if there exists slowing, some secondary reasons must be searched. The author also speculated pathogenetic causes of EEG slowing by these factors and insisted that there exist very close interrelation between basic EEG rhythms and prognosis of epilepsy.