longer has been the exposure, since the absorption of the excess will be a gradual process, taking place through the lungs in respiration, but it will be more rapid as the pressure is greater. These conditions agree with observed facts. After returning to the normal atmospheric pressure, the excess of gas probably passes off gradually by the lungs. in most instances; but if extreme, gas may escape from the blood within the body. The occurrence of this has been proved experiment. ally (Hoppe-Seyler and P. Bert). Gas, so escaping, is no doubt quickly reabsorbed, but must exert a pressure capable of arresting the function of the structures of the nervous system. If abundant, it may conceivably rupture these. Further, the special effect on the nerve-centres may be connected with the position of these within cavities that are practically closed. These conditions, coupled with the extremely circuitous course of the blood from the cord, may explain the incidence of the effects on the nervous system. The few ascertained facts harmonise with this pathology. Such escape of gas has been proved to result from diminution of the atmospheric pressure. In many cases no visible lesions have been found, and it is obvious that gas, widely effused, may exert dangerous pressure, and yet may be speedily removed, so that no indication may remain of the cause that has completely arrested function. It was once thought that the symptoms were due to local hæmorrhages, but extravasations are seldom met with, even in the results of experiment, and it is certain that they take but a trifling part in the production of symptoms. The only positive pathological observation agrees with the opinions above stated. Leydent found (in a case of characteristic paraplegia) small irregular fissures in the mid-dorsal region, chiefly within the posterior and hinder parts of the lateral column. The fissures were filled with round-cells, but contained no red blood-corpuscles, and from their well-defined edges they were certainly not produced by the infiltration of the cells found within them. The only explanation that is satisfactory, or in any harmony with their features, is that they were produced by the sudden escape of gas, and were afterwards occupied by the round-cells. The physical firmness of the cord is far less in the dorsal region than in the enlargements; and probably, as Leyden suggests, this is the reason why the dorsal region suffers structural damage from the escape of gas more than other parts. It is evident, however, that such escape may occur, in the brain, for instance, even to a fatal degree, and leave no traces unless it occurs so rapidly as to cause laceration of the tissue. The absence of hæmorrhage of the ordinary character may be further explained by the fact that the effused air will resist the effusion of blood. Some other anomalous features, such as the transient character of the symptoms in some cases, are also explained by the peculiar nature of the lesion, which may exert a wide-spread and considerable * Hoppe-Seyler, ‘Müller's Archiv,' 1887; P. Bert, Comptes Rend.,' 1871–2. Arch. f. Psych.,' ix, Heft 2. + influence, and in a short time may have vanished by the reabsorption of the compressing air. Hæmorrhage must be regarded as an accidental consequence, and even the production of such fissures as were found by Leyden may not be a necessary part of the lesion, even in grave cases. The mechanical effects of the gas may be considerable, and arrest the function of the nerve-elements, without the passage of the gas outside the capillary vessels. The fact that the spinal cord suffers more than other organs may be partly due to the plexuses through which the blood can return only slowly to the lungs, where its relief from the surcharge of gas is effected. The same considerations apply to the brain. In both organs the escape of air is more instantly disastrous than elsewhere, and, indeed, may not only occur, but act in a peculiar way, on account of the position of the organs within chambers that are, to a large degree, closed. The conditions are too complex to permit us to follow them in detail, or analyse fully the mechanisms that determine the effect on the nerve-centres. The escape of the excess of gas in the lungs can only take place gradually, as successive quantities of the surcharged blood pass through the capillaries of the air-cells. If, in consequence of the slowness of the circulation in the cord, any escape of air occurs in its capillaries, the effect of this will be still further to hinder the local circulation, and to favour the further escape of gas. In this way we are able to discern something of the mechanism which determines the special affection of the spinal cord. The less the exterual support the less will be the resistance to the escape of gas, and the incidence of the lesion on the posterior part of the cord may be in part at least determined by the adoption of the recumbent posture when the first symptoms are perceived. At the onset, when the first symptoms are experienced, it is probable that a return to a greater degree of pressure might arrest the development of the mischief. It is certain that, in all cases of exposure to a high degree of pressure, the transition to the normal pressure should be made gradually. When developed symptoms indicate that a definite lesion of the cord has occurred, the further treatment must be conducted on the same principles as in acute myelitis-the morbid process that has, in fact, been set up. HÆMORRHAGE INTO THE SPINAL CORD; Primary hæmorrhage into the spinal cord, sufficient to cause symptoms, is a very rare disease, and it is even more rare than is suggested by the cases now and then recorded as such. We have already seen that hemorrhage may accompany myelitis; a considerable extravasa tion may occur when the inflammation is only commencing, during the state of congestion, and while the symptoms are slight. Such cases are easily mistaken for primary hæmorrhage. The risk of error is not always removed by pathological examination. Inflammation results from hæmorrhage, and when indications of myelitis are found about a clot, it may be impossible to say whether these are primary or secondary. It is probable that many cases of secondary myelitic hæmorrhage have been regarded and described as primary, and it is possible that a few cases of primary hemorrhage have been regarded as secondary. One writer, Hayem, goes so far as to deny the occurrence of primary non-traumatic hæmorrhage; but such an exclusive view is unwarranted. It is certain, however, that the history of primary hæmorrhage has been largely written from uncertain data, and will need extensive revision when a sufficient number of exact observations have accumulated. * ETIOLOGY.-The rarity of hæmorrhage into the cord is especially great in comparison with the frequency of hæmorrhage into the brain. The difference probably depends on the tortuous and long course of the arterial path to the cord, whereby the vessels are preserved from the high pressure which is the chief cause of the degeneration, dilatation, and rupture of the cerebral arteries. Miliary aneurisms are not found within the spinal cord. Hæmorrhage is far more common in males than in females. It may occur at any age, and has been met with in young children, even so early as seven months,† while some of the subjects have been in advanced life; but it is most common between twenty and forty, i. e. during the first half of adult life. Many cases occur in young adults apart from injury or obvious exciting cause, and apart also from initial myelitis, which accounts, however, for some of the cases. It has been met with as a consequence of a hæmorrhagic tendency, as in a case in which it succeeded severe epistaxis, to which the patient, a young man of twenty-four, was liable.‡ Of immediate causes, injury is the most frequent, especially falls which involve a severe concussion of the spine; the spinal column may or may not be injured at the same time. Over-exertion and exposure to cold have in rare cases preceded the onset. Chronic alcoholism and sexual excess have been thought to predispose. In one case within my knowledge, an extensive hæmorrhage into the grey substance at the top of the lumbar enlargement resulted from coitus four times repeated, the symptoms commencing suddenly during the fourth act. Minute extravasations are often found after death from diseases which * Des Hémorrhagies intra-rachidiennes,' Paris, 1872. Clifford Allbutt, Lancet,' 1870, vol. ii, p. 84. Numerous hæmorrhages were observed in the grey matter of the lumbar and cervical enlargements (probably polio-myelitic, but occurring the day after a fall) in a child of four by Chaffey (Path. Trans.,' 1885). ‡ Sinclair, ‘Lancet,' 1885, ii, p. 1043. interfere with respiration and cause venous congestion, and they are especially frequent in maladies which, at the same time, cause functional excitement of the cord, as tetanus and all severe convulsions. They cause no symptoms, and are probably produced during the last moments of life. They have been terined “accessory." The diseases of the cord that lead to secondary hæmorrhage are chiefly inflammation, tumours, and cavities in the cord. The last-named condition is especially important. It is probable that hæmorrhage occurs more readily in spinal cords that contain a congenital cavity or fissure, due to an arrest of development, and surrounded by embryonal neuroglial tissue. Hence this condition will be found in disproportionate frequency in cases of hæmorrhage. PATHOLOGICAL ANATOMY-The minute extravasations just mentioned, as met with after asphyxial and convulsive diseases, are found in both the grey and white substance, but especially in the former. They are usually microscopic, or visible to the naked eye as minute red points, distinguishable from distended vessels only by their slighter resistance to a stream of water. The extravasation may occupy the perivascular sheath, or the cavity in which the vessel lies, or extend between the nerve-elements. The larger non-traumatic hæmorrhages, which cause symptoms, always begin in the grey substance, and are often confined to it, extending into the white columns only when large in size. The vessels of the grey substance are more numerous than those of the white, have less external support, and probably undergo more considerable changes in state. effusion forms a cavity in the cord, sometimes rounded, sometimes irregular, in transverse section, and half an inch or more in vertical extent. The cord is enlarged at the seat of the hæmorrhage, and this may be visible externally as a dark swelling the size of a nut or a bean. Very rarely the hæmorrhage tears the layer of cord which limits it, and blood, usually only in small quantity, escapes into the membranes. The tissue adjacent to the clot is usually broken down, stained, and softened, and inflammatory changes may be recognised in it with the microscope, just as in the neighbourhood of bæmorrhages into the brain. As in the latter, the effused blood slowly undergoes changes in tint, becoming first rusty and then yellow, and ultimately a cyst may remain. Several extravasations may coexist, usually in the same part of the cord. In contrast to this focal hæmorrhage there may be an infiltration of the grey matter with punctiform extravasations, which may appear, until closely examined, to be a single hæmorrhage. The tissue between these small extravasations is broken down. It is probable that this form is always secondary to myelitis, as in the case mentioned in the foot-note on p. 391. In such secondary cases a careful microscopical examination usually reveals indications of inflammation much more extensive than the area affected by the hæmorrhage. A growth into which hæmorrhage occurs is usually a glioma, unaffected parts of which will be found in the neighbourhood of the extravasation. The cavities in the cord into which blood may escape are sometimes of considerable vertical extent. I have known a fissure in the posterior column to be filled with blood through almost the whole length of the cord. This condition thus favours both the occurrence and the extension of hæmorrhage. SYMPTOMS.-Slight symptoms, "prodromata," have been observed in some cases, chiefly in the form of trifling sensory disturbance, tingling, &c., in the limbs afterwards paralysed. They have existed for a few hours or days, or even for two or three weeks before the onset. It is probable, however, that these have been cases of secondary myelitic hemorrhage, and that there are no premonitory symptoms in primary extravasations. The actual onset is always sudden; the symptoms attain a considerable degree in the course of a few minutes, but they sometimes increase during one or two hours, when the hemorrhage is from a small vessel and slowly increases in size-probably augmented by the rupture of other vessels which are torn in the tissue lacerated by the blood. Occasionally there has been transient loss of consciousness without any cerebral lesion, probably from the upward influence of the shock. Rarely the onset has been by a series of sudden augmentations of the symptoms. Sometimes the symptoms come on during sleep. The suddenness of the onset is the characteristic of the disease. The symptoms which thus develop vary according to the seat and extent of the extravasation. In the majority of cases there is paraplegia, complete motor and sensory paralysis up to the level of the lesion, with loss of power over the sphincters. Pain commonly but not invariably accompanies the sudden palsy; it may be felt in the spine or in the sacrum, round the trunk, at the front of the thorax, or at the epigastrium, sometimes seeming to pass thence through the trunk to the spine, or it may be felt in the legs. If in the spine it is local, and does not extend through a considerable length of the spine, as in meningeal hæmorrhage, and there is not the initial spasm and rigidity which characterise the latter disease. This pain may precede the palsy, even for half an hour or an hour; it begins suddenly, and is no doubt due to the first slight extravasation, which afterwards increases (when some resistance yields), and effects the compression that produces the paralysis. Thus a girl aged fifteen was seized with sudden sharp pain, referred to the sternum at the mammary level, and to the corresponding region of the spine; this continued for half an hour, when the legs suddenly became powerless and insensitive. It is highly probable that, when there is such initial medial or bilateral pain, the hæmorrhage occurs in the central region of the grey matter, perhaps from one of the commissural or anastomotic arteries (see p. 188), and that the pain is due to the irritation of the fibres of the posterior commissure, in which the paths for pain cross the middle line. The spinal column may be tender opposite the |