In these days of "mechanical brains" it is not surprising that our yearning continues for diagnostic machines that will replace the labor ordinarily associated with the solution of a difficult neurological problem. However, the most competent clinicians have always advised against too great a reliance on laboratory and mechanical diagnostic aids—and such advice is still sound. In 1929, Berger demonstrated that small changes in electrical potential can be picked up from various parts of the scalp. When suitably amplified these can be recorded, producing a wave-like pattern on a continuous record. This is known as an electroencephalogram. At first this attracted little attention, but in the past 20 years electroencephalography has come more and more to the fore. Societies and journals devoted to this subject have been founded. However, this technical method, which is of value both in clinical neurology and in research, now finds its reputation threatened as a