of balls, at which when heated she would habitually sit at open windows, and often sleep in clothes saturated with perspiration. In a few cases, severe exertion, such as excessive athletics, has been the apparent cause. A severe concussion of the spine has sometimes preceded the first symptoms by a few months. The disease may also follow great sexual excess. In many cases no cause can be traced, immediate or remote. Its general etiology thus resembles that of simple spastic paraplegia.

SYMPTOMS.-The onset of the disease is usually slow and gradual, two or three years passing before walking power is much impaired; it is rarely subacute, so as to reach a considerable degree in two or three months. The early symptoms resemble those of spastic paraplegia, with the addition of ataxic unsteadiness, which may be at first the most prominent symptom. The legs suffer first, often alone, or the arms may also be involved. The patient finds that he tires more readily than before, and also becomes unsteady on turning or walking in the dark, and these symptoms gradually increase. The weakness may be slight, but if all the movements are tested, a defect in power will be found in some, often most in the flexors of the knee and hip, and frequently more in one leg than the other. The patient is unsteady when be stands with feet together, and he tends to fall if the eyes are then closed. If the feet are bare, the irregular action of the muscles is shown by the movement of the tendons on the dorsum of the foot, as in tabes. In early cases, the inco-ordination is revealed by the patient's gait, which is distinctly unsteady. There is rarely the high movement and sudden descent of the feet often seen in tabes, but (as in many cases of tabes) the patient is unsteady, reels on turning, and has often to bring his foot suddenly to the ground to maintain his equilibrium. He may even have to steady himself with a stick, or to catch hold of some adjacent object, to save himself from falling. The ataxy is equally evident when he lies, and attempts, his eyes being closed, to touch some object with his foot.

The sensory and reflex symptoms present a marked contrast to those of tabes. Lightning pains are almost always absent; I have only met with them in one case, and in this they were a transient symptom. Sometimes there is a slight dull pain in the legs, felt especially on fatigue. Dull pain in the sacral region or in the spine is not uncommon, and is often an early symptom. The sacral pain, indeed, is sufficiently frequent to deserve special note. A girdle-pain is met with only in rare cases. As a rule there is no loss of sensation either on the legs or trunk. Hyperesthesia is equally rare. Reflex action from the sole may be normal or increased; less commonly it is diminished. The cremasteric and abdominal reflexes are sometimes lost. The most striking difference from tabes, however, is in the condition of myotatic irritability, which, in the vast majority of cases, is greatly increased. The knee-jerk is quick and extensive; it can be obtained from above (see p. 16), and there is generally a distinct

rectus-clonus. The foot-clonus is also commonly to be obtained. This myotatic excess persists and increases, and when the arms are involved they present symptoms similar to those in the legs-incoordination, weakness, and marked excess of the myotatic irritability. There may be conspicuous ataxy of the hands, and a tendency to cramp-like spasm on an attempt to use them. The muscles in both arms and legs are usually well nourished.

Sexual power is often lost early in the disease. The sphincters may be impaired, sometimes early, but they also often escape. As in other diseases, inability to empty the bladder is apt to develop insidiously, without the patient's knowledge, and the organ may thus become habitually distended.

The iris usually acts to light, but I have seen loss of the light-reflex in two or three cases, and have once known accommodation to be lost, the action to light being normal. These symptoms occur chiefly in cases following syphilis. Optic nerve atrophy occurs only in rare cases, far less frequently than in tabes; I have seen one case in which there were indications of a retro-ocular axial neuritis. The external ocular muscles are unaffected as a rule, except for nystagmus, which often exists on movement of the eyes, but seldom when they are at rest.

Slight impairment of articulation is not uncommon; sometimes there are irregular tremulous movements of the face resembling those of general paralysis, and this in cases in which there is no mental change. Westphal has observed marked ataxy of the facial muscles. As a rule the mental state is either normal, or there is merely slight failure of memory.

As the disease increases, the muscular weakness and reflex spasm become more and more considerable; while the ataxy, after reaching a certain degree (not sufficient to prevent locomotion), necessarily sinks into the background as the paralysis increases. With the increased weakness, the aspect of the patient comes to be that of spastic paraplegia, described in the last section. Indications of unsteadiness may still be observed in isolated movements. The arms, if previously free, may begin to suffer, but sometimes they escape altogether. The motor weakness may go on to complete paralysis. This is quickly reached in some cases, but so slow is the usual progress of the disease that one patient, in whom the disease was at no time absolutely stationary, was still able to stand, after eight years. In spite of the progress of the weakness, sensation remains unimpaired, and the cranial nerves do not suffer.

The symptoms present, in some cases, variations from this type. Sensation on the legs may be impaired, and the knee-jerk may be lost. These cases are, however, very rare, and are probably cases of true tabes with lateral sclerosis added. Other cases occur in which a girdlepain, impaired sensation, &c., suggest extensive changes in the cord, although the state of the legs is that common in the disease. Such

cases, as we shall see, are probably examples of chronic myelitis, or are intermediate between that and the disease now under consideration.

Ataxic paraplegia has little tendency to cause death. Indeed, the fatal cases have, for the most part, been untypical, and do not convey an accurate idea of the characters of the disease. The chief danger to life is from the accidents common to all chronic spinal affections, -bedsores, and especially kidney disease from undiscovered or untreated imperfect action of the bladder.

Among complications the most important are mental changes resembling those of general paralysis of the insane, of which indeed this combined sclerosis may form part. Slight muscular atrophy sometimes occurs. Arthritis of doubtful significance has been once observed. A patient of Westphal's presented derangement of the sympathetic, and died from peculiar spasm of the muscles of respiration. Visceral crises, however, are practically unknown.

PATHOLOGICAL ANATOMY.-In all cases the spinal cord has presented sclerosis of both posterior and lateral columns; but the precise extent and degree of the degeneration are subject to considerable variations. As a general rule, the sclerosis of the posterior columns differs from that of tabes in two particulars. First, it is not more intense, and often it is less intense, in the lumbar than in the dorsal region of the cord. Sometimes, indeed, as in the case shown in Fig. 129, in the middle and lower parts of the lumbar region, the posterior columns may be free from sclerosis, although it is considerable in the dorsal region and at the junction of this with the lumbar enlargement (B). The second difference is that the sclerosis has not that special intensity in the root-zone of the postero-external column which characterises the lesion of tabes. In rare cases the whole posterior column in the lumbar region is diseased. The part of the external column near the commissure and near the neck of the posterior horn usually remains free. Sometimes the degeneration does not extend up to the posterior surface of the cord; for instance, as in the case figured, it affects chiefly the middle three fifths of the posterior columns. When the degeneration is considerable, the posterior median columns, in the upper part of the cord, may present the usual ascending degeneration of secondary origin. When the lesion is slight in degree in the lower half of the cord, there may be only a diffuse degeneration of the columns in the cervical region, similar to that below, and not the more intense limited affection of the median part which occurs when there is a typical ascending degeneration from a considerable lesion of the column lower down.

The degeneration in the lateral columns is also variable in extent and position, and is often not strictly "systemic" in character, i. e. is not strictly limited to a single system of fibres, although the pyramidal tracts are chiefly affected. In one or two cases, indeed, the whole pyramidal tract, and this alone, has been degenerated; but more often the sclerosis, while intense in this tract,

also extends in front of it into the mixed zone of the lateral columns (Fig. 129, A, B). A similar extension, however, is usually met with in the sclerosis that attends degeneration of the anterior cornua; compare Fig. 127. The lateral limiting layer, between the pyramidal tract and the grey matter, may be also invaded. The direct cerebellar tract often escapes, as in Fig. 129, A (B is below its level of origin), but it is affected in some cases. Occasionally a zone of sclerosis has existed in the whole periphery of the cord, extending deeply into the lateral column in the position of the pyramidal tract. This tract has been found diseased, in some degree, in all cases, and in several the extent of its degeneration has been greater in the lower part of the cord than in the upper. The direct (anterior) tract may be affected as well as the lateral, and is commonly involved when the lesion is considerable in the upper part of the cord. As in most other degenerative diseases of the cord, there may be some increase of connective tissue in the unaffected columns. As a rule no morbid change has

FIG. 129.-Ataxic paraplegia, combined lateral and posterior sclerosis. A, upper dorsal; B, last dorsal; C, mid-lumbar. The posterior columns are free from sclerosis in C, except in their anterior parts; in B the disease involves the middle and anterior parts of both post.-med. and post.-ext. columns except in the neighbourhood of the neck of the horn; in A the sclerosis is slighter and is confined to the middle three fifths of these columns. The sclerosis of the lateral columns in C is limited, on the right, to the pyramidal tract, on the left it extends in front of this; in B it is very dense in the whole lateral column, involving not only the pyramidal tract but the limiting layer, and part of the "mixed zone;" in A it is similar in extent but slighter in degree. There is some increase of tissue throughout the anterior columns, and a focus of sclerosis near the anterior fissure, on the right in A and B, on both sides in C.

⚫ I am indebted to Mr. J. Hopkins for the opportunity of drawing these sections. The patient was a man aged twenty-one, in whom weakness of the legs commenced at twenty, after a wetting, improved, and then slowly increased. He was admitted

been recognised in the grey matter or in the membranes. In one case only was there slight meningitis. Disease of the ascending root of the fifth has been found in some cases;* in one there was also atrophy of the cells of the oculo-motor nucleus, and the case thus differed considerably from the type. The muscles have been found normal.

PATHOLOGY.-The double lesion in the posterior and lateral columns supplies an explanation of the two sets of symptoms which characterise the disease. The inco-ordination must be referred to the disease of the posterior columns, which, greatest in the dorsal region, involves. there the fibres that conduct impulses from the muscles, probably to the cerebellum.+ The effect must be to lessen the cerebellar guidance, and this explains the resemblance of the inco-ordination to that in cerebellar disease. The fact that the disease does not extend into the lumbar root-zone accounts for the integrity of the muscle-reflex action, and enables us to understand this marked difference from tabes. Disease of the direct cerebellar tract probably has an influence similar to that of the posterior median columns (see p. 215). That interruption of the conducting path in the cord will produce inco-ordination is, as we have seen (p. 205), well established. The sclerosis in ataxic paraplegia occupies also the region of the cord in which the short vertical fibres run, connecting the posterior grey matter at different levels, and it is possible that the damage to these fibres may contribute to the ataxy.

The only recognisable lesion that can be regarded as the cause of the paralysis is the degeneration of the pyramidal tract. The statements made regarding spastic paraplegia apply also to the disease, and need not be repeated. The lesion has been found greater on the side on which one leg was weaker than the other. The variations in the amount of disease, which does not always correspond to the degree of palsy, are probably due to the fact that the latter may be caused by degeneration of the termination of the fibres, which may be much greater than that of the fibres higher up the cord. In most cases that have been examined, the visible disease was greater in the lower than a year after the onset, with considerable weakness of the legs, a reeling, unsteady gait, increased knee-jerk, foot-clonus, but with no anesthesia, wasting of muscles, pains, or affection of the arms. The symptoms slowly increased, paraplegia became absolute, and the spasm very intense, sometimes flexor and sometimes extensor. There was a doubtful impairment of sensibility to touch on the legs, but no loss to pain. The sphincters became affected, bedsores formed, and the patient died two years after the onset. Fuller details will be found in 'Brain,' October, 1883, p. 383. Oppenheim, Neur. Cent.,' 1888, p. 647.

+ Much evidence has been obtained of the connection of the posterior median columns with the cerebellum through the grey matter of its bulbar nucleus, and also of the effect of interruption of these columns in the dorsal cord, in causing inco-ordination like that of cerebellar disease. Besides the experiments of Bechterew on the latter point already mentioned (p. 215), a recent important investigation into the connections of the cerebellum fully confirms that with the post.pyramidal nucleus (Brosset, Contrib. à l'étude des Connexions du Cervelet,' Paris, 1891).