The EEGs of 459 patients with and 442 patients without epileptic seizures showed generalized spike wave (s/w) paroxysms in 126 cases and generalized slow wave paroxysms in 63 cases before and/or after sleep deprivation. In patients with epileptic seizures s/w paroxysms appeared in 23%, whereas in patients without epileptic seizures. S/w paroxysms were present in only 5%. The difference is statistically highly significant. The frequency of generalized slow wave paroxysms showed with 6% and 8% respectively no statistical significance. In 91% of patients with s/w paroxysms the latter appeared already in the waking phases of the EEG following sleep deprivation. All s/w paroxysms which appeared in the sleep EEG were demonstrable already in stage I according to Rechtschaffen and Kales. The rate of activation of s/w paroxysms following sleep deprivation was 42% in patients with epileptic seizures, that of slow wave paroxysms was 38%. Although the number of patients with s/w paroxysms increases by approximately 100% following sleep deprivation they comprise still only 9.6% of all 459 patients with epileptic seizures. Combined with EEGs which show slow wave paroxysms or focal disturbances the incidence increases to 17%. Sleep deprivation as an activating procedure therefore provides a significant contribution to the confirmation of the diagnosis of epilepsy as well as control of therapy of epileptic patients whose routine EEG does not show typical pathological signs.