Thirty-two subjects affected by various types of epilepsy have been studied as follows: 1) from the clinical point of view with careful case history or by direct observation of the seizures; 2) from the EEG point of view by means of prolonged recordings during wakefulness and nocturnal sleep or with pharmacological activations; 3) from a radiological point of view with standard skull radiography, air-encephalography (PNX), brain scanning and, when necessary, with cerebral angiography, mono or bilateral. Furthermore, all the patients underwent a CAT at least once; in the majority of cases this examination was repeated after administration of contrast medium. The following results were obtained by comparing the various examinations: 1) anatomo-elector-clinical correlation was present only in some cases; 2) only in patients with cerebral neoplasms was there proof of an agreement in site between EEG, CAT and PNX; 3) it was not possible to detect in a definite manner epileptic glial lesions with CAT; 4) the EEG analysis, when repeated several times with different methods of investigations, whoed epileptic foci in a higher number compared to the anatomical focal lesions proved by CAT; 5) compared with air-encephalography and morphological brain scanning, CAT usually pointed out a higher number of focal and/or diffused cerebral lesions with a higher degree of precision; viceversa cerebral angiography proved to be irreplaceable in cases where it was necessary to study possible circulatory alterations.