man, aged 39, strictly temperate and with no history of syphilis, had for six months suffered from severe frontal headache and occasional vomiting, loss of memory, alteration in speech, which became slow and deliberate. There was weakness without any definite paralysis in the lower limbs. He gradually developed intense double optic neuritis, became subject to delusions, and passed into a condition of mental hebetude, terminated by sudden death a year after the beginning of his illness. The autopsy showed enormous dilatation of the ventricles and evidence of inflammatory thickening in the ependyma of the fourth ventricle. Probably a diagnosis is impossible in some cases of this class. The following points may suggest a diagnosis:

1. The skull may be enlarged, either a pre-existing enlargement dating from early childhood or birth, and the morbid conditions underlying which, having in later life again become active, or because even in adults the skull when there is internal hydrocephalus may occasionally enlarge.

2. Intermissions and remissions of symptoms lasting even for years. 3. Absence of localising symptoms; if the cranial nerves are affected the paralysis (except in the case of the optic nerve) is often of a fluctuating character.

4. Slight exophthalmos, fine tremors of the tongue, hands and lips, and weakness of the lower limbs.

5. Examination of the cerebro-spinal fluid obtained from lumbar puncture may throw some light on the pathological condition.

Lastly, in considering this question, Oppenheim mentions that an acute encephalitis may become engrafted on a chronic hydrocephalus and definite localising symptoms ensue.

Multiple Sclerosis. The symptoms of this disease may be simulated by tumour about the cerebellum and its peduncles, and the peculiar intention tremor so often seen in disseminated sclerosis may also be found in tumour. Combinations of ataxia with tremor also may be present in tumour of the corpora quadrigemina, the pons and cerebellum. As points of distinction, optic neuritis does not occur in disseminated sclerosis, at least never in the condition of choked disc; there may, however, be a slight transient neuritis. States of stupor form no part of the symptomatology of sclerosis; cortical symptoms are also against this diagnosis, as also is tenderness on pressure over the skull. In children disseminated sclerosis is rare, tumour more common, yet in some cases the diagnosis may be very difficult. The following case, which was under the care of Dr. Percy Marsh at the Liverpool Children's Hospital illustrates this:

J. M., a boy, æt. 10, said to have been quite well until six months ago, when he became unsteady in his movements. There was nothing in his family history of importance. The boy could not walk without support and then was able to do so only in an ataxic spastic manner. There was great increase of the knee-jerks, double ankle clonus, and extensor plantar reflex. In the upper extremities there was the characteristic intentional tremor, and speech was distinctly slow and staccato. There was no optic neuritis, no vomiting, no mental

disturbance and no affection of the bladder, no nystagmus, no scoliosis or club-foot. It was thought on the whole to be a case of disseminated sclerosis.

DIAGNOSIS OF THE SITE OF THE TUMOUR.

Oppenheim does not deal very fully with this section of the subject in a special chapter, as he does so in the body of the book. The chapter in Brun's is more complete. From these two works I have prepared the following table of differential features between frontal and cerebellar lesions-a distinction which it is often difficult to make:

1. Headache. In frontal lesions, mostly in the front of the head, and is less severe than in cerebellar growths where it is referred chiefly to the back of the head and region of the neck.

2. Percussion. Often circumscribed tenderness in frontal lesions sometimes with a tympanitic note. In cerebellar growths the tenderness is not circumscribed and is not seldom universal.

3. Optic neuritis. In frontal lesions rather late in development, and may be unilateral and attended with symptoms of hemianopsia owing to involvement of a tract at the base of the brain. In cerebellar lesions it is early, intense and bilateral.

4. Cortical symptoms. In frontal lesions Jacksonian convulsions and paralytic symptoms on the opposite side of the body from cortical involvement may occur; such signs are absent in cerebellar tumours. Motor aphasia may also be present in frontal lesions, not in cerebellar.

5. Rotation of the head and eyes. In frontal lesions this is towards the opposite side, and is an irritation symptom. In cerebellar growths there may be paralysis of the eye movements owing to interference with the 6th nucleus, so that there is deficiency in the conjugate movement towards the side of the lesion. The two phenomena are essentially different. "A definite attitude of the head is not infrequently seen in cases of cerebellar disease in man, the position being with the ear approximated to the shoulders on the side opposite to the lesion, and with the face turned up to the side of the lesion." (Batten. Brain, 1903.) 6. Tonic contractions of the back muscles may occur from lesions in either place.

7. Ataxia. The ataxia may be of exactly similar features, but in cerebellar lesions it is more intimately connected with vomiting and vertigo.

8. Affection of cranial nerves. In frontal lesions the 3rd and 6th may be paralysed as distant symptoms; there may also be cortical paralysis of the 7th and 12th on the opposite side. Anosmia, if present, is usually unilateral. The 8th nerve is never affected. If the 5th is affected it is the ophthalmic branch only. In cerebellar lesions peripheral palsy of the nerves in the posterior fossa at the base of the skull is very characteristic. Anosmia when bilateral is more likely to be from a cerebellar growth. The 8th nerve is not infrequently affected and all the sensory branches of the 5th.

9. Psychical symptoms develop earlier in frontal growth, and a peculiar form of punning (witzelsucht) has been mentioned as of some significance in tumours in these regions.

THE COURSE AND PROGNOSIS.

The usual course is chronic, progressive and towards death. Acute symptoms may be developed from hæmorrhage into the tumour, especially in the glioma, softening of the surrounding tissue, sudden swelling or increase in size of a cyst or the rupture of an aneurysm; of more importance are those cases in which remission or intermission of the symptoms occurs and occasionally even complete cessation—a so-called cure. Quite a number of cases are recorded in which the duration has been five years or even a much longer period of time. Spontaneous cure has been known to take place in aneurysm, cysticercus and echinococcus. Solitary tubercle also occasionally heals, and localised tuberculous meningo-encephalitis may be curable.

Clinically, an important group of cases is that in which all symptoms except blindness from optic atrophy disappear, sometimes for a very long period. This fact is to be remembered in considering the advisability of a palliative operation, since simple opening of the dura mater in many cases causes the disappearance of optic neuritis.

Favourable spontaneous terminations are more likely to occur in children than adults. Osteoma, fibroma, angioma, cholesteatoma and psammoma are of better prognosis than growths of other kinds. Cases of long duration are usually tumours in the great brain or those growing from the base. Remarkable cases of cure sometimes follow "auto-trepanation and consequent relief of pressure by escape of cerebro-spinal fluid. Dr. T. R. Glynn very kindly allows me to mention an unpublished case of his, as an example. A young man in 1900 began to suffer from headache, vomiting, weakness in arms, and difficulty in walking. Vision gradually became much affected and double optic neuritis was present. In 1903 a discharge of clear fluid occurred from the nose and was shown to be cerebro-spinal fluid. The patient then gradually recovered, the weakness and optic neuritis disappeared, vision became normal, and the patient was apparently completely restored to health. (The pathological basis of this case seems uncertain. W. B. W.)

TREATMENT.

Medical. The administration of large doses of potassium iodide was introduced by Wernicke in 1881. It is to be used in rapidly increasing doses, from 100-300 grains per diem in an adult, half that quantity in a child. Apart from syphilitic lesions, there is a great deal of evidence showing that under this treatment amelioration of symptoms and remissions for considerable periods of time have taken place. Aneurysm, tubercle, sarcoma and glioma have been thus favourably influenced, or tumours with cyst formation, according to Oppenheim. Mercury, except in syphiloma, seems to do harm.

Surgical Treatment. Oppenheim gives a most thorough analysis of the indications for operation, derived from a study of the records of 371 cases in which the skull was opened for brain tumour. We have:140 (37.7%) died at operation, or within a week.

88 (23-24%) marked improvement.

111 (29-30%) slight improvement.

32

371

insufficient detail given by authors.

Again, of the 371 cases, in 205 a palliative operation only was attempted, in 166 an attempt to enucleate was made, and in these 82, or about 50% showed marked improvement.

Allan Starr, in an analysis of 365 cases, collected up to Jan., 1903, gives:

51

8

16

Tumour removed, patient died = 59

Tumour removed, patient recovered = 168... 152

The results appear fairly good, but statistics are misleading, since failures are not published, and since there is considerable divergence of opinion as to what constitutes a successful result.

Very instructive is the record of 22 cases in Oppenheim's own practice:

False general diagnosis

Tumour not correctly located

=0

= 1

Tumour not discovered at operation, but found

at the site of operation post-mortem = 1

That is, a

cure in 4-5%, marked improvement in 20-21%, rapid death in 35%. Bruns, writing in 1897, came to the same conclusions, and the figures may probably be considered as giving the present-day results of surgical treatment.

The prognosis varies greatly according to the region in which the tumour is. Thus, of frontal tumour 46 cases are reported, with favourable results in 11 23.9%. The motor area is the most favourable site. In Bergmann's statistics of 116 cases, in which a radical operation was performed; in 75% the tumour was in this region.

Cerebellar tumours, on the other hand, are the most unfavourable

for surgical interference. Of 45 cases, 32, or 71%, died from the operation, whilst in 15% the operation was followed by considerable relief. Bruns, Bergmann and Allen Starr consider cerebellar tumours inoperable; but Oppenheim will not go so far as this; he considers an operation may be performed if the diagnosis is quite certain and general symptoms marked; in these cases also a palliative operation may do good. The cranial neuro-fibroma may also be successfully removed from the posterior fossa.

In summing up, Oppenheim makes some very valuable remarks. An operation is to be considered and the possible issues submitted to the patient or the patient's friends in cases of brain tumour in which the general and local diagnosis is thought to be certain and the growth is in a region reachable by the surgeon. It may be undertaken when the local diagnosis is certain and at least one of the general symptoms present, occasionally without general symptoms if the local signs indicate a slow gradual invasion by a cerebral lesion; this is especially important in tumours involving the Rolandic area, and it is only in cases of the latter class that the medical adviser is justified in actively advising an operation. Bruns gives very similar advice.

Palliative operations are a very important branch of treatment in tumours. Both Oppenheim and Bruns emphasise the necessity of opening the dura mater in order to allow a free flow of cerebro-spinal fluid. Even this operation is not without danger, and success may be fleeting or none at all. The indications are (1) certain general diagnosis with severe symptoms not otherwise capable of relief; (2) threatened blindness. When there is advanced optic neuritis, with diminishing acuity of vision, the propriety of opening the dura mater is to be considered. It is of great importance that the opening should not be made over a part of the cortex of high functional value, lest a resulting hernia cerebri cause damage. Hence in left-sided tumours the right frontal, parietal or temporo-sphenoidal lobes are sites of election. If there is any chance that the local diagnosis may be correct the opening can be made over the suspected place in the hope that a further radical operation may be performed later.

Is lumbar puncture or the puncture of the lateral ventricle any good? It is doubtful if either of these methods can claim to be of much use; neither is without danger. Gumprecht has recorded a number of cases of sudden death after lumbar puncture in cases of brain tumour, the explanation given being that the cerebellum and medulla are pushed downwards almost on to the bony ring of the foramen magnum, and the withdrawal of a little fluid from the spinal canal may cause the medulla to be suddenly pressed on to the bone. Since it is in cases of hydrocephalus that lumbar puncture is especially indicated and that this is usually the result of cerebellar tumour, in which lesion the danger above mentioned is most likely to arise, we see that the indications for the operation can rarely arise. Ventrical puncture also is probably not so valuable as the opening of the dura mater, though in the simple hydrocephalus, due to ependymitis, it may be useful.