When drugs are administered intravenously, the risk of causing injury to nerves in front of the elbow can be considerably minimized if the anatomical structure of this part is kept in mind. The operation appears to be so simple that the novice is frequently permitted to dig his way to proficiency, without first having been the recipient of instruction or warning regarding the procedure.

This case has shown so great an improvement by symptomatic treatment with parasympathetic stimulants and a sympathetic depressant that it seems worthy of record for its own value, apart from the interesting investigations to which it leads.

As there is still considerable hesitancy in the adoption of this valuable form of therapy, I am glad to have an opportunity of stating the case for a more widespread application of prolonged narcosis to the manic-depressive disorders. A legitimate diversity of opinion exists as to the propriety of using any drug with a toxic action over a long period, especially so when it introduces an extraneous, and maybe serious, risk into a disease of itself non-fatal, and with a large recovery-rate. With attention to detail in the treatment, and with the addition of the Cardiff modification of prolonged narcosis, we are satisfied that toxic symptoms arising during the administration of somnifaine—the drug used—can be controlled, and the risk heretofore associated with this form of therapy eliminated.

Narco-analysis is an eclectic technique based on the observation (6) that a combination of narcosis with psycho-therapy is quicker and sometimes more effective than the formal methods of analytical psychology. The earlier attempts of this kind were limited to the non-analytical methods of overt suggestion augmented by drugs.

In the following paper a case of état dysmyélinisé of the globus pallidus, or as it is often synonymously called, “Hallervorden-Spatz disease”, will be presented. Records of thoroughly studied cases of this condition are still rather scanty and there is much need for a systematic extension of experience, the more so since, as Helfand (16, 17) has recently pointed out, progress in the elucidation of the condition has not always been enhanced by generalizations from findings based on too small a case-material. We are happy in having at our disposal the vast resources of the London County Mental Hospitals. It is hoped to follow up this paper by publishing further cases of a rather different nature, though with some characteristics in common with the case described here.

During 1931, 1932 and 1933 at least 20 female cases of mania were admitted to the Warwick County Mental Hospital, and I propose to review the treatment of these cases, omitting melancholia because of the difficulty of excluding involutional illnesses, because mania is more difficult to treat (1), and finally because the results of treatment are easier to follow.

Since we do not yet know the real nature of Schilder's disease nor see a direct way of investigating it, research on this problem can usefully be carried out in two ways: (1) careful description of cases in all possible clinical and pathological detail, and (2) broad comparative studies. That collection of further case-material is still useful was shown, we hope, in the description of a family by one of us (A. M.) in collaboration with Tennent (24). In this family there was a pronounced tendency to gastro-intestinal troubles preceding the nervous symptoms, and the same has been reported by Globus and Strauss (15), Globus (14), and F. Bielschowsky (2). Too much importance must not be attached to such findings, as the onset of the disease just as often follows measles, influenza and other acute infectious conditions; and there is the striking familial taint of tuberculosis in the families described by Symonds (45) and van Bogaert and Scholz (47), and atrophy of the suprarenals was associated with the disease in the latest case reported by R. Pfister (29). The great variety of such preceding illnesses makes us rather chary of regarding them as of direct ætiological or pathogenic significance. It is, on the other hand, of the greatest importance to keep such occurrences carefully in mind, because we do not know yet what common pathogenic factor may lie hidden behind these various initial symptoms. As regards gastro-intestinal troubles, their significance in various avitaminous and other deficiency conditions is generally known, and they have been a very obvious initial symptom in the demyelinizing disease of monkeys, the histopathological kinship of which to Schilder's disease has already been shown in A. Meyer and Tennent's paper, and which will be discussed later in this contribution. This kinship appears to us to be closer than was recognized by Schob (35), Spielmeyer (42), Scherer (33), and Davison (9).

Since October, 1934, we have treated 40 cases of psychoses by somnifaine narcosis in the Glasgow Royal Mental Hospital. The technique employed by us was that recommended by Ström-Olsen. The patients were isolated in a quiet room, in partial darkness, and a nurse was constantly in attendance by night and day. As a preliminary to the treatment, insulin and glucose were given for a few days beforehand, while the patient received daily bowel wash-outs. The amount of glucose given was 1 oz. to 1½ oz. three times a day, along with 10 units of insulin. This insulin and glucose routine was continued during the treatment, and for some days after the administration of the somnifaine was discontinued. The somnifaine was injected intra-muscularly in 2-c.c. doses. The degree of narcosis one wished to produce was that approximating as nearly as possible to natural sleep. Throughout, one tried to keep the patient in such a state that they were able to be roused from time to time for feeding. The amount of somnifaine necessary to produce this state varied with each patient, but as a rule we found that 6 c.c. of somnifaine given in 2-c.c. doses in the 24 hours was the average amount required. The treatment was continued for 10 to 14 days.

Dr. Tómasson's interesting remarks on the therapeutic effect of adrenergic and cholinergic drugs lead me to mention some experiments with benzedrine recently carried out at the Maudsley Hospital. These experiments primarily originated from different considerations. They were part of more extensive investigations into the connection between mental phenomena and blood-pressure and its fluctuations. Benzedrine seemed to be suitable for experimental alterations of the blood-pressure because its effect on arterial tension lies between that of adrenaline and ephedrine with regard to rapidity and intensity of action. Experiments on normal persons demonstrated a peculiar stimulating and euphorizing effect of the drug. The experimental subjects were selected as normal in regard to blood-pressure. They included several cases of mild depression. These also showed either a slight elation, or at least a decrease of their depression after the administration of the drug. This result encouraged further experiments on depressives of various kinds. These were always carried out in the following way: Patients whose mental state was well known and who had been constant for a period with regard to their blood-pressure were given at first small, and later slowly increasing doses of the drug. There were intervals without medication in which the subjects received inert tablets of the same appearance and taste as the benzedrine. The blood-pressure was taken several times daily.

It was with some diffidence that I agreed to contribute towards this discussion because I felt that, having had no personal experience of the use of hypnotic drugs in the modern method of prolonged narcosis, anything I might say might be held by some to be spoken without authority.