Introduction
When electroencephalography is used as a diagnostic test in persons with seizures, it is open to two criticisms. First, markedly abnormal interseizure records are often found in patients who have mild seizures, while, on the other hand, slightly abnormal and normal records are often found in patients who have severe seizures. Consesequently, electroencephalograms afford no evidence of the severity of seizures and vice versa. Second, in patients with seizures an abnormal record can rarely be used with confidence as evidence of organic cerebral disease. This second criticism can still be maintained when the EEG abnormalities are focal, for the clinician reasonably refuses to be satisfied with an electrical focus from which an abnormal histological lesion cannot be inferred with confidence.The first criticism rests on a distinction which is usually drawn between severe and mild seizures in terms of the violence of convulsions and the depth and duration of