THE SIGNIFICANCE OF AN ABNORMAL ELECTROENCEPHALOGRAM.

ATTENTION has recently been focused upon the occurrence of abnormalities in the electro-encephalogram (e.e.g.) of apparently normal subjects, and upon the high incidence of these abnormalities in subjects with behaviour disturbances and in other members of their families. Davis (1941) analysed the variations in the e.e.g. of groups of normal subjects and the abnormalities in wave form and frequency which she found in epileptic and psychotic patients. Similarly Lindsley and Cutts (1940) measured the number of waves in different frequency bands in small groups of normal adults and children, and in groups of children considered to be " constitutionally inferior " or presenting definite behaviour problems. Previously Jasper, Solomon, and Bradley (1938) had studied the abnormalities seen in the e.e.g. of behaviour problem children, while Hoagland, Cameron, and Rubin (1937), and Davis and Davis (1939) showed that a relatively high proportion of psychotic, especially schizophrenic, patients had abnormal e.e.g.'s. In a statistical evaluation of the e.e.g. in epilepsy made by Golla, Graham, and Walter (1937), abnormalities were found in the e.e.g.'s of 60 per cent. of a mixed group of epileptics. It has since then been generally accepted as the common experience of most workers that a proportion of undoubted epileptics have a normal e.e.g. between fits. Jasper and Kershman (1941) have recently reported abnormalities in the e.e.g.'s of over 95 per cent. of a large group of epileptic subjects, and in surveying the electro-encephalographic phenomena found in epilepsy they have excluded epileptic subjects with " normal" records as being atypical. Lennox, Gibbs and Gibbs (1940) found that with their methods of interpreting records about 3 per cent. of patients with overt seizures had normal e.e.g.'s. This discrepancy in the percentage abnormality rate of epileptics may possibly be due to a difference in the norm adopted, but as there are not comparable figures for control groups of normals in all the communications this point is obscure. Lowenbach (1939), Strauss, Rahm, and Barrera (1939), and Lennox, Gibbs, and Gibbs (1939 and 1940) have 257 coright.