In regard to Prognosis wounds of the chest furnished the most hopeful class of the whole series of trunk or visceral injuries. Cases of wound of the heart and great vessels afforded the only exceptions to an almost universally favourable course, both as regards life and the non-occurrence of serious after-effects.

This was mainly explicable on two grounds: first, the sharply localised character of the lesion produced by the bullet of small calibre; and, secondly, the fact that the lung, the most frequently injured organ, is not materially affected by the grade of velocity with which the bullet strikes. In point of fact, wounds of this organ probably afford an instance in which high grades of velocity are distinctly favourable to the nature of the injury, and this is possibly true in the case of wounds of the chest-wall also.

The significance of the calibre of the bullet in wounds of the chest is evident. The late Mr. Archibald Forbes, in one of his letters from the seat of the Franco-German war, remarked that in crossing a battlefield it was easy to recognise the patients who had suffered a wound of the lung from the fact that the whistle of the air entering and leaving the chest was plainly audible. This was, indeed, not uncommonly the case in wounds produced by the older bullets of large calibre, but with the employment of the smaller projectile it has become an experience of the past. Some evidence as to the comparative severity of wounds produced by the larger forms of bullet was, moreover, afforded by the present campaign, since Martini-Henry wounds were occasionally met with. Of some instances observed by myself, in one, external hæmorrhage was a prominent symptom; in another, a piece of lung was prolapsed from a wound in the back, and twice I observed pneumothorax, an uncommon sequela to wounds from bullets of small calibre.

It may be remarked, however, that all these more serious injuries were recovered from, also that when we consider that the patients were comparatively young and healthy subjects, the favourable prognosis was what might have reasonably been expected. When, as occasionally happened, a patient of more mature years, with enlarged facial capillaries, received a wound of the lung, the course was in no way so favourable as that witnessed in the case of the younger men.

In support of this opinion I may add that wounds from shrapnel and fragments of shell also did remarkably well, although they sometimes gave rise to more troublesome symptoms than did wounds produced by bullets of the Mauser type. Again, these injuries as a whole were of nothing like so serious a nature as the lacerations of the lung produced by fractured ribs, which we commonly have to treat in civil practice, and are not accustomed to regard as especially dangerous.

It is also a striking fact that the most common and troublesome complication of wounds of the chest, hæmothorax, was usually the result of the wound of the chest-wall and not of the lung. I preface these remarks to the detailed account of the thoracic injuries, because I think the favourable course usually taken by patients with wounds of the lung has been accorded somewhat greater prominence than the circumstances warranted.

Non-penetrating wounds of the chest-wall.--Surface wounds were not very common, and were chiefly of interest in so far as they illustrated the very superficial course that may be occasionally taken by a bullet without breach of the integument, and as sometimes affording opportunity for the exercise of diagnostic skill when the track traversed the axilla.

The most common situation for tracks taking a long course on the surface of the thoracic skeleton was the back. Such wounds were usually received while the patients were prone on the ground; thus I might instance a case in which the bullet entered the posterior aspect of the shoulder 3 inches above the spine of the scapula, passed downwards, pierced that process, and emerged 2 inches below the inferior angle of the bone. Wounds of a similar nature coursing in transverse and oblique directions, and not implicating bone, were also seen. Those implicating the vertebræ have been already dealt with. The scapular region was also a favourite one for the lodgment of retained bullets, some resting in the supra- and infra-spinatus muscles, others lying beneath the bone itself.

On the anterior aspect of the chest, bullets coming from the front sometimes traversed and fractured the clavicle, and then took a short course downwards, emerging over the ribs or sternum. Figure 81 represents a particularly long track in this region. In other cases the precordial region was crossed, but I never witnessed any serious effect on the heart's action in any such injury at the time the patients came under my notice.

Wounds received with the arm outstretched and traversing the axilla sometimes gave considerable trouble in excluding with certainty a perforation of the thoracic cavity. Thus a bullet entered below the centre of the right clavicle and emerged 2-1/2 inches below, above the angle of the scapula, at its axillary margin. The arm was outstretched at the moment of the reception of the injury; but when the wound was viewed with the limb placed alongside the trunk, it seemed almost impossible that the chest cavity could have escaped. In some cases of this kind the difficulty was at once cleared up by noting evidence of injury to the axillary nerves.

A word will suffice as to the treatment of these wounds. The only special indication was to keep the scapula at rest for a sufficient period. I have dealt with the anatomy of them at such length only because in their extreme form they are so highly characteristic of the nature of the injuries which may be produced by bullets of small calibre.

Penetrating wounds of the chest.--Tracks crossing the thoracic cavity in every direction were common. When the erect attitude was maintained, frontal and sagittal wounds, pure or oblique, were received; when the prone position was assumed, longitudinal tracks, either purely or obliquely vertical, were the rule. Experience of wounds of the latter class was extensive in the present campaign, from the fact that so many of the advances were made in prone or crawling attitudes. The vertical and transverse tracks each possessed the special characteristic of frequently implicating both the thoracic and abdominal cavities, but the vertical were often prolonged into the neck, or even downwards through the pelvis. The vertical wounds in addition sometimes exhibited one very important feature, the fracture of several ribs from within, often at a very considerable distance from the aperture of either entry or exit.

[Illustration: FIG. 81.--Superficial Track in anterior Wall of Trunk]

Characters of the apertures of entry and exit.--As has already been mentioned, the chest-wall was one of the situations in which the aperture of entry was often large, and the oval form due to obliquity of impact on the part of the bullet was particularly well marked. The exit wounds were often smaller than those of entry, especially if the bullet emerged by an intercostal space; even when the ribs were comminuted, the fragments were, as a rule, too small to occasion more than a slightly enlarged and irregular aperture. Taken as a class, however, and putting aside explosive exit wounds, wounds of the chest afforded more numerous examples of irregular outline and variation in size than were met with in any other region of the body.

When the tracks penetrated the broad upper intercostal spaces, an interesting feature, due to the tense and rigid nature of the muscles closing the intervals, and their large admixture of fibrous tissue, was sometimes noticed. The bullet, especially if passing obliquely, was apt to cut a slit in the muscles far exceeding in size the opening in the overlying integument, with the result of leaving a palpable subcutaneous defect. Under these circumstances the yielding spot was often noticed to rise and fall with the movements of respiration, external palpation met with an absence of normal resistance, and there was impulse on coughing.

Fractures of the ribs.--These injuries were produced in either transverse or longitudinal coursing tracks, their special feature being a sharp localisation of the lesion of the bone.

In tracks crossing the chest transversely the injury to the ribs might consist in notching, perforation, or complete solution of continuity, sometimes with fine comminution. In the incomplete injuries some importance attached to the localisation of the lesion to the upper or lower border of the rib, in so far as the intercostal artery was concerned. Comminution at the wound of entry was, as a rule, not so extensive as at the aperture of exit, and in any case was less apparent, since the fragments were driven inward. The wider comminution at the exit aperture depends on the lesser degree of support afforded by the thoracic coverings to the convex outer surface of the rib, and on the fact that the velocity of the bullet has been lowered by its passage through the opposite rib and the chest cavity.

The splinters of comminuted ribs are small, and wide-reaching fissures rare. These characters depend on the elastic nature of the resistance offered by the curved rib to the passage of the bullet, which is calculated to preserve the bone from the full force of impact, except at the point actually impinged upon.

Fractures of the ribs, produced from within by bullets taking a longitudinal course through the thorax, were still more special in character. They were also more important, as giving rise to troublesome symptoms.

In these, again, the degree of injury to the bones varied considerably. In some cases the bones were merely grooved internally, without any external deformity; in other cases a sort of green-stick fracture was produced, accompanied by the projection of a tender salient angle externally; in others complete solution of continuity was effected.

Another feature of importance was the occasional implication of several ribs. In this case the symptoms accompanying the injury were very much more like those observed in the corresponding injuries resulting from indirect violence seen in civil practice.

Injuries to the costal cartilages closely resembled those to the ribs. Perforation, bending from injury to the inner aspect, and comminution were observed. The latter condition differed from the similar one seen in the case of the ribs only in so far as the tougher consistence of the cartilage did not lend itself to such free comminution, and the splinters remained in great part attached. The nature of the fractures, in fact, somewhat resembled that seen on breaking a piece of cane.

I saw no fracture of the sternum except of the nature of a pure perforation; these were not uncommon in the hospitals, either in the upper or the extreme lower portions of the bone. Fractures in other portions were no doubt usually associated with fatal injuries to the heart. The openings were usually so small as to be difficult of palpation, and I never had the opportunity of examining one post mortem.

Perforations of the body of the scapula were common, but they were of little importance in symptoms or prognosis.

Symptoms of fracture of the ribs.--Fractures accompanying transverse wounds of the chest were characterised by the insignificance of the symptoms produced. Every common sign of fracture of the rib was in fact absent. Neither pain, stitch on inspiration, nor crepitus, either audible or palpable, was, as a rule, present. This absence of signs was accounted for by the nature of the lesion: thus in perforations or notchings there was no loss of continuity, while in the freely comminuted fractures the loss of continuity was so absolute as to allow no possibility of the main fragments rubbing together. Again, part of the symptoms attending these injuries, as seen in civil practice, depends upon contusion and laceration of the surrounding structures--a condition precluded by the localised nature of the application of the violence by a bullet of small calibre. In order to establish a diagnosis, therefore, we were in many cases reduced to palpation, and occasionally to direct examination of the wound.

Fractures accompanying longitudinal tracks formed a class rather apart in the matter of symptoms. In these mere groovings might also be accompanied by no signs, or at the most by slight local pain and tenderness. When, however, the grooving was sufficiently deep to be accompanied by deformity, or a complete solution of continuity was effected, the signs were often severe. The tender salient angle, or, in the absence of this, a highly tender localised spot, often pointed to the less severe injuries, and when the fractures were complete or multiple, pain was a very prominent symptom, both constant and in the form of inspiratory stitch. The severity of the pain was probably to be in part ascribed to implication of the intercostal nerves, which in these injuries was direct and often multiple. Again, severe contusion or actual laceration of the nerves, with resulting anæsthesia, was less common than when the bullet directly implicated the nerves in transverse wounds. Free comminution and absolute solution of continuity were also less common than in the fractures accompanying transverse wounds; hence pain from rubbing of the fragments on inspiratory movement or palpation was more common, and crepitus, either on auscultation or palpation, was more often met with. Patients with this class of fracture often suffered greatly from painful dyspnoea, and were unable to assume the supine position.

External hæmorrhage of severity was rare from these thoracic wounds; in many cases it did not amount to more than local staining of the shirt; altogether I saw only one or two cases where any serious bleeding occurred. Internal hæmorrhage into the pleura, in consequence of the position of the intercostal arteries, was common, and often abundant; this will be treated of under the heading of hæmothorax.

Treatment of fractured ribs.--Transverse wounds of the thorax, with no symptoms of fractured ribs, needed to be dealt with as wounds of the soft parts alone.

In multiple fractures accompanying longitudinal tracks, bandaging or strapping for the purpose of fixation was necessary to relieve pain. A few fragments of bone sometimes needed primary removal, and occasionally small sequestra were removed at a later date; but necrosis was rare, unless some complication led to the development of a fistula.

Retained bullets were occasionally met with in the chest wall. In such cases the last remaining energy of the bullet often seemed to have been spent in diving under the margin of a rib and turning longitudinally up or down. Removal was sometimes necessary, either from the prominence produced, the presence of pain, or the continuance of suppuration. Some of the specimens removed offered interesting evidence of the capacity of the ribs to withstand considerable violence from a bullet. These were slightly bent, and marked by a half-spiral groove. I saw such bullets removed from the thoracic and the abdominal wall, and the evidence seemed rather against the groove having been produced prior to their entrance into the body.

[Illustration: FIG. 82.--Spirally grooved Mauser Bullet]

Wounds of the diaphragm.--Perforations of the diaphragm were very frequent, and as a rule of small significance. When, however, the course taken by the bullet was parallel with that of the slope of the diaphragm, a more or less extensive slit was the result. I saw such a wound still gaping, and 2 inches in length, in the body of a patient who died three weeks after the infliction of a fatal abdominal injury.

In several other obliquely transverse thoracic wounds there was reason to assume the existence of similar slits. Certain signs were more or less constant under these circumstances. These consisted in shallow respiration, often accompanied by a groan or the slightest degree of hiccough on inspiration, and considerable increase in respiratory frequency. In one patient the respirations were at first 48, only dropping to 36 some seventy hours after the reception of the injury. In some of the cases in which the abdominal cavity was implicated, wound to the diaphragm seemed a more likely explanation of early, frequent, and painful vomiting than did visceral injury. The possibility of the later development of diaphragmatic herniæ in some of these patients will have to be borne in mind in the future.

Visceral injuries.--The frequent escape of the thoracic viscera from injury, putting aside the lungs which fill so great a part of the cavity, was very remarkable. I never saw a case in which I could assume injury to any of the posterior mediastinal viscera, although such may have occurred on the field of battle. An injury to the oesophagus, for instance, would almost of necessity be accompanied by wound of either one of the large vessels, even the thoracic aorta, or the spinal column. I was somewhat surprised, however, to learn on enquiry from surgeons who had seen a large number of the dead and dying on the field, that thoracic wounds, putting aside those that directly implicated the heart, were responsible for but a small proportion of the fatalities.

The escape of the posterior mediastinal viscera, the great vessels, and the heart, is, I believe, to be explained by the fact that all are supported and held in position by the loose meshed mediastinal tissue, which allows for their displacement after the manner observed in the case of the vessels and nerves lying in the loose tissue of the great vascular clefts.

Wounds of the heart.--Perforating wounds of the heart were probably fatal in all instances, in spite of the fact that, in some patients who survived, the position of wound apertures on the surface of the body made it difficult to believe that the heart had not been penetrated. (See cases below.)

In the case of this organ, we must bear in mind its constant variations in bulk, its elastic compressibility, and its variations in position in systole and diastole. The variations in bulk and position would be capable of explaining the escape of the organ from injury at some particular moment, when a second shot apparently through the same wound track might implicate it. Beyond this, reasoning from the case of analogous hollow viscera, as the arteries or the intestine, a bullet might readily score the surface of the heart without perforating its cavity.

Such accidents were observed. Thus, in a case examined by Mr. Cheatle, the patient died of suppurative pericarditis, secondary to a wound of which the external apertures had closed. In this patient both auricle and ventricle were scored externally by the passage of the bullet.

I am, however, disinclined to allow that many patients survived direct blows on the heart, since I believe that in the majority if not in all cardiac wounds the actual cause of death was not hæmorrhage, but sudden stoppage of the heart's action. This is to be inferred from the fact that severe external hæmorrhage did not occur; in some cases the shirt was hardly stained, and in all death occurred in the course of a very few minutes. Again, in none of the patients whom I saw who had received possible wounds of the heart-wall were there evident signs of hæmo-pericardium. In view of the difficulty of detecting this condition from physical signs, this argument is naturally not of great weight, but must be allowed.

One or two death scenes from cardiac wound were described to me. In one the patient muttered 'They have got me this time,' and died quietly; in a second the patient's face became ghastly pale, he lay on his back with the knees flexed, clutching the ground, gasping for breath, and died only after some minutes of evident great agony. The absence of any post-mortem details as to the condition of the heart in these injuries is much to be regretted.

    (145) Entry, in the seventh left intercostal space, in the posterior axillary line; exit, immediately below the ninth costal cartilage, close to the position of the gall bladder.

    This track in all probability involved the diaphragm twice, both lungs and pleuræ, and passed immediately beneath the heart. The liver was also perforated, but the spleen and stomach probably escaped as far as could be judged from the symptoms. The patient afterwards developed a pneumo-hæmo-thorax on the right side. The immediate symptoms were great distress in breathing and rapid irregular pulse. The difficulty in respiration was probably in part accounted for by the injuries to the lung and diaphragm. The pulse remained from 112 to 120 for three days, at first soft and hardly perceptible, later very irregular, and dropping one every fifth or sixth beat; and it seemed fair to attribute this to the shock to the nervous mechanism of the heart. The patient recovered from the chest injury.

    In some other patients in whom the track passed close below the heart a disturbance of the pulse rate was noted, but this was in some cases a slowing, not below 48, in others quickening to 100, with irregularity both in force and beat.

    (146) Entry, in the fourth right interspace, 3 inches from the middle line; exit, in the seventh left interspace, in the mid-axillary line. This wound was received at a distance of 500-600 yards, but the bullet penetrated both sides of a stout silver cigarette case and some cigarettes before entering the body. There were minor signs of pulmonary injury, 'coughing day and night,' and slight discoloration of the sputum on three or four occasions. The respirations were quickened to 32, and as much as ten days after the injury the pulse only beat 48 to the minute; it then rose to 56, but beat in a very deliberate manner.

In other cases the signs were almost nil.

    (147) Entry, in the fourth right intercostal space 3/4 of an inch from the sternum; exit, in the sixth left interspace in the posterior axillary line. This patient had no symptoms, beyond quickening of the pulse to 100, and a 'feeling of tightness at the heart.' He shortly returned to active duty.

    (148) Entry, situated in the third right interspace 3 inches from the sternal margin; exit, in the fourth left space 2-3/4 inches from the sternal margin. In this case the bullet without doubt passed through the anterior mediastinum, and slight injury to the lung was evidenced by transient hæmoptysis.

Some remarks regarding wounds of the thoracic vessels have already been made in Chapter IV., where instances of injury to the innominate and left subclavian arteries are recounted. The escape of the large trunks was generally quite as astonishing as in other parts of the body, especially in the superior mediastinum.

    (149) Entry, over the first right intercostal space beneath the centre of the clavicle; exit, at left anterior axillary fold. The great vessels must have been crossed here in immediate contact, and considerable hæmorrhage from the wound of entry caused great anxiety; this ceased spontaneously, however, and, beyond transient hæmoptysis and a right pneumo-thorax, no further trouble occurred.

    (150) Entry, in the ninth interspace, just anterior to the anterior axillary line; exit, through the right half of the sternum, 1/2 an inch below the upper border. No primary hæmorrhage of importance followed, but I believe this patient subsequently died. The wound was received at a range of within fifty yards.

Wounds of the lungs.--Numerically, pulmonary wounds formed the most important series of visceral injuries met with in the thorax, the frequency of incidence corresponding with the proportionate sectional area occupied by the organs. Although these injuries did well, and needed little interference on the part of the surgeon, many points of interest were raised by them.

Thus the comparative importance of the wound in the chest-wall to that in the lung itself, was scarcely what, without actual experience, would have been expected, the former proving so very much the more important element of the two.

The question of velocity on the part of the bullet took a very secondary position in these injuries. I saw a number of cases in which the patients estimated the range at which they received their wounds as from 30 to 50 yards, and although some of the wounds were of a severe type, the increased gravity depended rather on the injury to the chest-wall than to that of the lung. If the bullet passed by the intercostal space, avoiding the rib, I very much doubt if the relative velocity was of any importance, further than from the fact that a sufficiently low degree to allow of lodgment of the bullet was distinctly unfavourable.

In view of the general lack of significance in these injuries it was interesting to note how very definite was the ill effect of early transport on the after course. This depended on the frequent development of parietal hæmothorax in patients who were not kept absolutely at rest.

The tracks produced in the lungs by the bullets were very minute, and in the few cases in which opportunity arose for their examination post mortem some little time after the infliction of the wound, there was great difficulty in localising them. The slight damage incurred by the pulmonary tissue is due to its elasticity and non-resistent character.

Pulmonary hæmothorax was distinctly rare. Reasoning from the analogous wounds of the liver, tracks scoring the surface of these organs might be much more to be feared than clean perforations. The elasticity of the lung tissue, however, must make such lesions rare. In point of fact, there is no reason why a perforation by a bullet of small calibre should be much more feared than a puncture from an exploring trocar, and the danger of the two wounds is probably very nearly the same.

The only points of importance as to the particular region of the lung traversed were the distance from the periphery as affecting the probable size of the vessels injured, and perhaps the implication of the base or apex of the organ respectively. I am under the impression that wounds in the apical region were somewhat more liable to be followed by the development of pneumothorax, and possibly hæmothorax, while wounds at the base gained their chief importance from the frequency of concurrent injury to the abdominal viscera. I had no experience of the immediate results of wound of the great vessels at the root of the lung, but assume that they led to speedy death.

Symptoms of wound of the lung.--I shall describe the whole complex usually observed, although it is obvious that the wound of the chest-wall is responsible for a large proportion of the signs.

The majority of these injuries were accompanied by a certain degree of systemic shock, and this was more marked in wounds received at a short range. The shock was, however, rather to be attributed to the injury to the chest-wall and thoracic concussion than to that to the lung itself. I think it may also be stated that few patients were inclined to walk or remain in the erect position after receiving these wounds; this feature was also noted in horses in whom a bullet passed through the lungs.

The remarks made as to the pain accompanying fractures of the ribs apply equally here. Pain was not a prominent symptom, except in so far as the actual impact caused temporary suffering. It was striking how often patients who received wounds through the arm prior to the same bullet traversing the chest appreciated the chest wound only, yet the chest might pass unnoticed when a still more sensitive part was struck later, as has been already mentioned in the section on wounds in general.

Dyspnoea was not a prominent primary symptom. The patients sometimes had 'all the wind knocked out of them' at the moment of impact, but when seen at the Field hospitals a short time later, the respirations were shallow, but easy and regular, and only moderately quickened; thus 24 was a not uncommon rate. Naturally if accumulation of blood in the pleura began early and continued, these remarks do not hold good; and again in some older men of full-blooded type and the subjects of recurrent attacks of bronchitis, a considerable degree of pain, dyspnoea, and even cyanosis was sometimes present soon after the injury. The complication of wound of the diaphragm has already been referred to in this relation.

Local respiratory immobility of the thoracic parietes and consequent asymmetry of movement were constant. This was especially a marked feature when the upper part of the chest was implicated on one side only. It rather corresponded, however, to the local shock observed in wounds of the limbs than to the instinctive immobility accompanying fractures of the ribs; since, as already explained, small-calibre bullet wounds of the ribs are not necessarily painful on movement, and the sign existed even when the bullet had passed by an intercostal space. This sign was naturally a transitory one.

Hæmoptysis was a fairly constant sign, but sometimes quite absent when no doubt could exist as to the perforation of the lung. As a rule, a considerable quantity of blood might be coughed up shortly after the injury; but I never knew this to be sufficient in amount to give rise to any misgivings as to danger from the hæmorrhage. After the first evacuation of blood from the wounded lung, the sign varied much; in the majority of instances the patients continued to expectorate small quantities of blood mixed with mucus, for some three or four days, the blood gradually assuming a coagulated condition. Sometimes only the primary hæmoptysis was noted, and still more rarely the expectoration of clots was continued for a week, or even longer. This probably depended partly on personal idiosyncrasy, partly on the size of the vessels which had been implicated in the track.

Cough was not commonly the troublesome symptom noted in the contused wounds of the lung seen in civil practice accompanying fracture of the ribs. Moist sounds were usually audible on auscultation, but in many cases over a very limited area and only on the first few days.

Cellular emphysema was distinctly rare, and usually limited in extent: thus I saw it in the posterior triangle of the neck alone in an apical wound; over about a third of the upper part of the thorax in another wound through the second intercostal space, and in this case oddly enough the emphysema was the only sign of injury to the lung; and very occasionally widely distributed--in the latter case there were also usually multiple fractures of the ribs. Neither issue of air from the external wound nor frothy blood was ever seen with small-calibre wounds, but I saw one instance in a case of Martini-Henry wound.

Pneumothorax was also rare. I saw pneumothorax three times out of about half a dozen Martini-Henry wounds, but I do not think it occurred as often in 100 small-calibre wounds. The Martini-Henry wounds all recovered; but convalescence was very prolonged, and the same remark to a less degree holds good in the small-calibre cases.

That the slow recovery in cases of pneumothorax in the Martini-Henry wounds was due mainly to the size of the opening in the thoracic parietes was, I think, proved by the fact that in the small-calibre bullet wounds, followed by the development of pneumothorax, the external wounds were usually large and irregular in type; also, that in the only pneumothorax which I saw produced during an extraction operation, the air was very rapidly absorbed. In the latter case, however, there was little reason to conclude that wound of the lung had occurred primarily, and certainly no opening existed at the time the thorax was incised.

Hæmothorax.--This was the most frequent and also the most interesting of the complications of wound of the chest. In 90 per cent. or more of the cases, the hæmorrhage was of parietal source, and due either to direct injury to the intercostal vessels by the bullet or to laceration by spicules of comminuted ribs. For this reason, the passage of the bullet whether by an intercostal space, or through a rib, provided the wound was not at the posterior part of the space where the artery crosses, was a point of considerable prognostic importance. Exclusion of the lung as the source of hæmorrhage was, I think, amply justified by the absence of continuous recurrent or progressive hæmoptysis in the majority of the cases, and by the very small trace of injury found in the lungs of patients who died some weeks after the injury. In such it was difficult to discriminate the tracks at all. I only happened to see one case where free hæmoptysis, during the course of development of a hæmothorax, pointed to the lung as the source of the blood.

Hæmorrhage into the pleural cavity occurred in some degree in a very large proportion of the chest wounds, but it was especially interesting to note how greatly its extent was influenced by the amount of transport to which the patients were subjected in the early stages after the injury. During the early part of the campaign, on the western side, I saw a large number of chest wounds, and had I been asked my opinion as to the relative frequency of occurrence of hæmothorax I should have placed it at about 30 per cent. The patients in these early battles needed little wagon transport, and when sent down to the Base travelled in comfortable ambulance trains. After the commencement of the march from Modder River to Bloemfontein, however, these conditions were changed, and all the chest as other cases were exposed to the necessity of three days and nights' journey to the Stationary hospitals and afterwards to the long journey to Cape Town. Of these patients, at least 90 per cent. suffered with hæmothorax of varying degrees of severity.

In some cases, the least common, signs of considerable intra-pleural hæmorrhage immediately followed the wound; in others, the accumulation of blood was gradual, and only manifest in any degree at the end of three or four days, when it became stationary if the patient was kept at rest. In a second series the hæmorrhage was of the recurrent variety; these cases differing little in character from those of slight continuous hæmorrhage. In a third, the bleeding was definitely of a secondary character, corresponding with one of the classes of secondary hæmorrhage described in Chapter IV., and occurring on the eighth or tenth day from giving way of an imperfectly closed wounded vessel. In either of the two latter classes the development of the hæmothorax often corresponded with a journey, or with allowing the patient to get up.

The general course of these effusions was towards spontaneous absorption and recovery. Coagulation of the blood took place early, the fluid serum separated, and tended to undergo absorption with some rapidity, leaving a small amount of coagulum at the base, which evidenced its presence for many weeks by a persistence of a certain degree of dulness on percussion. Early coagulation, I think, accounted for the usual absence of gravitation ecchymosis as a sign.

The course to recovery was sometimes broken by signs of slight pleuritic inflammation, which, as affecting the amount of effusion, will be spoken of under the heading of symptoms. In some cases the amount of blood was so great as to necessitate means being taken for its removal; in these a reaccumulation often took place. Occasionally an empyema followed in cases thus treated.

The nature of the blood evacuated on tapping varied much. In very early aspirations unchanged blood was often met with, but clot sometimes made evacuation difficult and necessitated a second puncture. In the tappings done at the end of a week or more a dark porter-like fluid was common, while when suppuration was imminent a brick-red-coloured grumous fluid replaced normal blood. In the cases where early incision was resorted to, blood both fluid and in clots was often mixed with a certain proportion of lymph flakes, perhaps indicating the part taken by inflammatory reaction to the irritation of the clot in producing the rise of temperature.

Symptoms of hæmothorax.--In the more severe cases of primary bleeding the symptoms did not, as a rule, reach their full height until the third or fourth day after the injury. The patients then often suffered severely. The pulse and temperature rose, and to general symptoms of loss of blood were added: occasional lividity of countenance; severe dyspnoea, accompanied by inability to lie on the sound side or to assume the supine position; absence of respiratory movement on the injured side; pain, restlessness, cough, and sometimes continuance of hæmoptysis, small clots usually being expectorated.

Accompanying these symptoms were the usual physical signs of fluid in the pleura in differing degrees and combination. Dulness of varying extent up to complete absence of resonance on one side, often accompanied in the incomplete cases by well-marked skodaic resonance anteriorly. Loss of vocal resonance, and fremitus; oegophony, tubular respiration over the root of the lung or at the upper limit of the dulness, and more or less extensive displacement of the heart. Obvious increase in girth, fulness of the intercostal spaces, or gravitation ecchymosis was rare. The latter was most common in instances in which multiple fracture of the ribs existed (see fig. 83). I think the rarity of the last sign must have been due to the early coagulation of the blood, and its retention by the pleura, as I saw well-marked gravitation ecchymosis in one or two cases of mediastinal hæmorrhage.

The above complex of symptoms was common to all the cases, but in the slighter ones they gave rise to little trouble, and cleared up with great rapidity.

[Illustration: FIG. 83.--Gravitation Ecchymosis in a case of Hæmothorax, accompanying fracture of three ribs from within. The influence of the fractures on the development of the ecchymosis is shown by the linear arrangement of the discoloration]

The most interesting feature was offered by the temperature, as this was very liable to lead one astray. A primary rise always occurred with the collection of blood in the pleura, this reaching its height on the third or fourth day, usually about 102° F. in well-marked cases; it then fell, and in favourable instances remained normal. In a large number of cases, however, where the amount of blood was considerable, this was not the case, the primary fall not reaching the normal, and a second rise occurred which reached the same height as before or higher. The second rise was accompanied by sweating, quickened pulse, and the probability of the development of an empyema had always to be considered. I believe in most cases this secondary rise was an indication of a further increase in the hæmorrhage, for the dulness usually increased in extent, and such rises were often seen when the patient had been moved or taken a journey. Again, the temperature often fell to normal after paracentesis and removal of the blood, to rise again with a fresh accumulation, which was not uncommon. I have already mentioned the large proportional incidence of hæmothorax observed in the patients who had to travel down from Paardeberg, and I might instance another case related to me by Dr. Flockemann of the German ambulance, which was very striking. A Boer, wounded at Colesberg, developed a hæmothorax which quieted down, and he was removed to Bloemfontein; on arrival at the latter place the temperature rose, and other signs of fever suggested the development of an empyema; an exploring needle, however, only brought blood to light. After a short stay at Bloemfontein the symptoms entirely subsided, and the man was sent to Kroonstadt, when an exactly similar attack resulted, again quieting down with rest.

Similar recurrent attacks of hæmorrhage and fever occurred, however, in patients confined to their beds without moving after the first journey. Some temperature charts, in illustration of this point, are added to the cases quoted later. The explanation of the recurrent hæmorrhages is, I think, to be found in the reduction of the intra-thoracic pressure with coagulation and shrinkage of the clot in the pleura in the patients kept quiet in bed, while in the patients who had to travel it was probably the result of direct mechanical disturbance.

In many of these cases a pleural rub was audible at the upper margin of the dulness with the development of the fresh symptoms. Whether this was due to actual pleurisy or to the rubbing of surfaces rough from the breaking down of slight recent adhesions which had formed a barrier to the effusion, I am unable to say, but the signs were fairly constant. In some instances the increase in the amount of fluid was, no doubt, due to pleural effusion resulting from irritation from the presence of blood-clot, or perhaps the shifting of the latter; in these the secondary rise of temperature may well be ascribed to the development of pleurisy.

I am inclined to believe, however, that the primary rise of temperature was similar to that seen when blood accumulates in the peritoneal cavity as the result of trauma, and the secondary rises in most cases to those which we saw so frequently accompanying the interstitial secondary hæmorrhages spoken of in Chapter IV., and are to be explained on the theory of absorption of a blood ferment. The secondary rises always occurred with a fresh effusion, often of blood, occasioning an extension, which broke down probable light adhesions and exposed a fresh area of normal pleural membrane to act as a surface for absorption.

It is, of course, manifest that the fever might also be ascribed to the infection of the clot or serum from without, and in the first cases I saw I was inclined to take this view, since we had in every case the primary wounds of chest-wall, and possibly of lung, and in some the addition of a puncture by an exploring needle between the first and second rise. After a wider experience, however, I abandoned the infection theory, as it seemed opposed by the very infrequent sequence of suppuration. The effect of simple removal of the blood or serum was also often so striking as to strongly suggest that it alone was responsible for the fever. Exactly the same result, moreover, followed evacuation of the interstitial blood effusions already mentioned elsewhere.

The common course of all the cases of hæmothorax was to spontaneous recovery, the rapidity of the subsidence of the signs depending mainly on the quantity of the primary hæmorrhage, and the occurrence of further increases. The blood serum tended to collect at the upper limit of the original blood effusion (as was often proved on tapping), and this was first absorbed; the clot deposited on the pleural surface and at the basal part of the cavity was, however, not absorbed with the same rapidity. In the majority of the patients when they left the hospitals, at the end of six weeks on an average, some dulness and deficiency of vesicular murmur always remained, and the clot and the surrounding surface, irritated by its presence, will, no doubt, be responsible for permanent adhesions in many cases. That such adhesions do form in the majority of cases I feel certain, as, although these patients when they left the hospital were to all intents and purposes apparently well, few of them could undertake sustained exertion without getting short of breath, and sometimes suffering from transitory pain, and for this reason it became customary to invalid them home.

In a small proportion of the cases empyema followed; but I never saw this in any case that had neither been tapped nor opened, and I saw only one patient die from a chest wound uncomplicated by other injuries. This case was an interesting one of recurrent hæmorrhage followed by inflammatory troubles:--

[Illustration: TEMPERATURE CHART 2.--Secondary Hæmorrhages in a case of Hæmothorax. Case No. 151]

    (151) The wound was received at short range, probably at from 100 to 200 yards. Entry, 1 inch from the left axillary margin in the first intercostal space; exit, at the back of the right arm 1-1/2 inch below the acromial angle; both pleuræ were therefore crossed. The patient expectorated at first fluid, then clotted, blood in considerable quantity. When brought into the advanced Base hospital on the third day, there were signs of blood in the left pleura, cellular emphysema over the right side of the chest, and signs of collapse of the right lung. The temperature chart gives shortly the course of the case: the right pneumo-thorax cleared up spontaneously, also the emphysema; but the left pleura needed tapping to relieve symptoms of pressure on four occasions, the 13th, 15th, 19th, and 25th days respectively. On the first two occasions blood was removed, on the third blood serum only, and on the last pus. The patient was relieved after each aspiration; after the third, the temperature fell to normal, the general condition also improved, and he promised to do well. None the less, reaccumulation took place, the evacuated fluid assumed an inflammatory character, and an incision to evacuate pus was eventually followed by death on the twenty-seventh day. The amount of hæmoptysis throughout was considerable, and the case was possibly one of pulmonary hæmothorax, as after death no source of hæmorrhage could be localised in the intercostal space. The track in the lung was almost healed, and although a part of it allowed the introduction of a probe for about an inch, it could be traced no further even on section of the organ, and no special vessel could be located as the original bleeding spot.

Empyema.--I may here add the little that I have to say on this subject. During the whole campaign the single case of primary empyema that I saw was the one recorded below, which deserves special mention as illustrating the disadvantage of extracting bullets on the field. Under the conditions which necessarily accompanied this operation the ensurance of asepsis was impossible, and the additional wound no doubt proved the source of infection.

    (152) Entry, at the posterior margin of the sterno-mastoid muscle, 2 inches above the clavicle; the bullet came to the surface beneath the skin over the fifth rib, in the nipple line of the right side. There was never any hæmoptysis, but the patient suffered with some dyspnoea throughout. After a three days' stay in the Field hospital, where the subcutaneous bullet was removed, the patient was transported by wagon and train to the Base, a journey of about 600 miles.

    On the fifth day pus escaped from the extraction wound, and when the case was examined at the Base, the temperature was 101°, the pulse over 100, the respirations 30, and the whole side of the chest was dull, with the exception of a patch of boxy resonance over the apex anteriorly. On the following day the chest was drained, and a considerable amount of pus evacuated, which was mixed with breaking-down blood-clot. A fortnight later a second operation had to be performed to improve the drainage, and the patient made a tedious recovery.

The following case well illustrates the symptoms in a severe case of hæmothorax, and empyema following aspiration:--

    (153) The patient was wounded at Paardeberg at a range of from 500 to 700 yards. Entry, just to the left of the episternal notch; exit, in the fifth left interspace posteriorly, midway between the spine and vertebral margin of the scapula. A quantity of bright blood was brought up at once, and later blood was coughed up in clots.

    There was no great pain at the moment of the injury; the man again got up to the firing line, and later walked two miles to the Field hospital without aid. He remained here a week, when he was sent down to the Base, and during the first three days' journey in the wagon he began to get worse. On the fourth day cough began to be very troublesome.

    When he arrived at the Base, fifteen days after the original injury, there was much dyspnoea; the temperature was 102°, and the pulse 110. The left side of the chest was dull throughout; an aspirating needle was introduced, and a pint of very dark liquid blood drawn off. The whole of the blood was not removed on account of the very severe cough and pain which the evacuation occasioned. The man appeared to steadily improve until three weeks later, when the temperature, which throughout had been uneven, became consistently high, and signs of fluid at the base increased. An aspirating needle was introduced, and 16 ounces of pus were drawn off. Two days later a piece of rib was resected (Mr. Pegg) and another pint of pus evacuated. After this, rapid improvement took place, and in ten days the man was able to be up and dressed, although a small amount of discharge still persisted. He eventually made an excellent recovery.

Secondary empyemata not uncommonly followed incision of the chest, or excision of a rib for draining a hæmothorax. These operations in the early part of the campaign were more freely undertaken on the supposition that rise of temperature and other symptoms of fever pointed to incipient breaking down of the clot. Subsequent experience showed this not to be the case, and early operations for drainage ceased to be undertaken. In these operations a primary difficulty was met with in effectively clearing out the clot, a drain had to be left, and suppuration occurred later in a considerable proportion. The suppurations were most troublesome; local adhesions formed, and the pus collected in small pockets, which were difficult to find and to drain, and even when the collections seemed to have been successfully dealt with at the time, residual abscesses often followed at a very late date. Thus, I saw a case with a contracted chest and a fresh abscess the day before I left Cape Town, in whom I had advised and witnessed an operation for the evacuation of clot in the presence of signs of fever a week after my arrival in the country, nine months previously. I saw another case where general infection followed incision of a hæmothorax, but the patient fortunately recovered.

The question of pleurisy has already been mentioned in connection with hæmothorax; it no doubt accounted for secondary effusion in some cases, and beyond this I have nothing to add to what has been there said.

Pneumonia was rare; there were occasionally signs of consolidation, but, I think, quite as often in the opposite lung as in the one injured. I never saw a fatal case, and I am inclined to think that when it occurred it was as often the result of cold and exposure as of the injury to the lung. Abscess of the lung I only saw once, and that in a case in which the injury to the chest was complicated by paraplegia from spinal injury and septicæmia, and it was possibly pyæmic.

Diagnosis.--No difficulties special to small-calibre wounds were experienced, except such as have been already dealt with. The only class of case which frequently gave rise to difficulty was hæmothorax. Here two points especially needed consideration. (1) The source of the hæmorrhage as parietal or visceral. As has been already foreshadowed, this was mainly to be decided by the amount and persistence of the hæmoptysis, but naturally free hæmoptysis did not negative concurrent parietal bleeding. Then the actual source of the bleeding other than from the lung had to be considered; in the great majority of cases the intercostal vessels were responsible, and attention to the course of the tracks often allowed this to be definitely decided upon.

A case included in the chapter on Injuries to the Blood Vessels (No. 5, p. 127) is of great interest in this particular; in that instance feebleness of the radial pulse, together with the position of the wound, was a valuable indication of injury to the subclavian artery, but weakened somewhat by the fact of retention of the bullet, and hence uncertainty as to the exact course that it had taken, and as to whether the bullet itself was not responsible for pressure on the vessel. Such indications, however, should make one very chary of interference with a hæmothorax, even with extremely urgent symptoms, in the light of our present knowledge of the nature of the lesions to the great vessels produced by small-calibre bullets, and their tendency to be incomplete.

(2) The imminence of suppuration or its actual occurrence.--In most cases it sufficed to preserve an expectant attitude, and in the persistence or increase of symptoms, to have recourse to an exploratory puncture as the best means of solution of the difficulty.

Prognosis.--The prognosis both as to life and as to subsequent ill-effects was remarkably good; in many cases of uncomplicated injury to the lung the patients rejoined their regiments at the end of a month or six weeks. In the more serious cases complicated by the collection of blood in the pleura, convalescence was more prolonged, and an average time of six to eight weeks often elapsed before the patients could be safely discharged from hospital. In the more serious a certain amount of dulness always persisted at this time over the base of the lung, and the chest was usually somewhat contracted on the injured side, with evidence in the way of decreased vesicular murmur that the lung was still not free from compression. With regard to the persistence of dulness on percussion, it is well to bear in mind that a thin layer of blood apparently produces as serious impairment of resonance as a much larger quantity of serum. The signs appeared to favour the view that the space necessary for the location of the hæmorrhage had been obtained at the expense of the lung rather than by distension of the thoracic parietes, and also, I think, denoted the presence of adhesions. Possibly they will entirely disappear with the return of full excursion movements of respiration, the latter being often still somewhat restricted when the patients left hospital. All the patients with such signs were liable to attacks of pain and shortness of breath on actual bodily exertion. I happened to meet with an officer, the subject of a Lee-Metford wound of the thorax, sustained five years previously, and he told me that he was nine months before he could take active exercise without feeling short of breath.

As to the cases of hæmothorax and empyema which needed drainage, all did well; but expansion of the lung was much less satisfactory than would have been expected, probably on account of especially firm adhesions. The importance of concurrent injury I need hardly dwell on; but I might add that perforation of one or both arms, the most common one, did not materially affect the general statements above made.

Treatment.--In the early stages of the pulmonary wounds rest was the all-important indication, and when this was assured few serious cases of hæmothorax occurred. Beyond simple rest, the administration of opium with a view to checking internal hæmorrhage was used with good effect. The wounds needed simple dressing only.

The treatment of hæmothorax at a later date, however, was of much interest and difficulty. I think the following lines may be laid down for guidance in such cases:--

(i) Hæmothorax, even of considerable severity, will undergo spontaneous cure. An early rise of temperature may be disregarded.

(ii) Tapping the chest is indicated when pressure signs on the lung are sufficiently severe to cause serious symptoms, and the removal of the blood undoubtedly shortens the period of recovery, as well as relieves symptoms.

In such cases the collection of blood has usually been rapid and continuous; hence a fresh hæmorrhage is always probable when the local pressure has been removed. Tapping therefore should not necessarily mean complete evacuation, and should be followed by careful firm binding up of the chest, the administration of opium, and the most stringent precautions for rest.

(iii) Tapping may be needed as a diagnostic aid, and in such circumstances as much fluid as can be removed should be evacuated with the same precautions as mentioned in the last paragraph.

(iv) Tapping may be indicated for the evacuation of serum expressed from the blood-clot, or due to pleural effusion, on the same lines as in any other collection of fluid in the pleural cavity.

(v) Early free incision is, as a rule, to be steadfastly avoided. Some cases already quoted fully illustrate its disadvantages.

(vi) Cases in which an incision and the ligature of a parietal artery are indicated are very rare. I never saw such a one myself.

(vii) If a hæmothorax suppurates, it must be treated on the ordinary lines of an empyema. In view of the constant formation of adhesions and difficulty in drainage, a portion of a rib should always be resected in order to ensure sufficient space for after-treatment. The cavities, as a rule, are better irrigated, the usual precautions being taken where there is any reason to fear that the lung is still in communication with the cavity.

Care in carrying out asepsis in tapping, which should be performed with an aspirator, need hardly be more than mentioned. It will be noted that in some of the cases quoted suppuration followed tapping, but it must be remembered that in these the two primary wounds already existed as possible channels of infection.

Retained bullets of small calibre in the thoracic cavity were not common, unless the lodgment had occurred in the bodies of the vertebræ. I saw very few. Shrapnel bullets and fragments of shells, however, were, in proportion to the frequency of wounds from such projectiles, more commonly retained. The rules to be followed in such cases do not materially deviate from those to be observed in the body generally.

When the bullet is causing no trouble, and is lodged in either the bone of the spine or the lung substance, no interference is advisable. When, on the other hand, the bullet as viewed by the X-rays is seen to be in the pleural cavity, and any symptoms of its presence exist, it may be justifiable to remove it. I saw this done in one case for the removal of a shrapnel bullet from the lower reflexion of the pleura on account of fixed pain and tenderness complained of by the patient. The bullet, a shrapnel, had perforated the arm, which the patient was sure was by his side at the moment of injury, and the X-rays showed it to lie at the bottom of the pleural cavity, where we assumed it had fallen. When, however, the bullet was removed by Mr. Watson, he found that the fixed pain and tenderness had been the result of a fracture of a rib from the inner side, not involving loss of continuity; hence the actual indication for the operation had been a delusive one, since the bullet had not fallen, but expended its last force in injuring the rib. The patient made an excellent recovery, and rejoined his regiment at the end of six weeks. I saw several cases in which the bullet was lodged in either the lung or bones of the spine do well with no interference. The great disadvantage of primary removal in inducing an artificial pneumo-thorax and in laying open a hæmothorax is obvious.

In case of lodgment of the bullet in the lung, bearing in mind the infrequency of untoward symptoms, the latter should be watched for prior to interference.

The following cases illustrate some typical instances of wound of chest accompanied by the development of hæmothorax:--

[Illustration: TEMPERATURE CHART 3.--Primary Hæmothorax, with rise of temperature. Secondary rise, with fresh effusion and pneumonia. Spontaneous recovery. Case No. 154]

    (154) Severe hæmothorax. Spontaneous recovery.--Wounded at Modder River at a distance of 30 yards. Entry, at the junction of the left anterior axillary fold with the chest-wall; exit, immediately to the left of the seventh dorsal spinous process. The patient arrived at the Base with signs of an extensive hæmothorax, accompanied by a temperature which reached 102° on the fourth day, and on the evening of the tenth 103°. The man was very ill, and an exploring needle was inserted, by which about an ounce of blood was evacuated. The signs of fluid in the left pleura were accompanied by those of consolidation over the lower fourth of the right lung, and the sputa were rusty. Evidence of perforation of the left axillary artery existed in feebleness of the radial pulse; and there was musculo-spiral paralysis.

    After the preliminary puncture, the man refused any further operative treatment, although a second rise of temperature commenced on the fifteenth day, culminating in a temperature of 103.2° on the eighteenth. The further treatment of the patient consisted in the ensurance of rest and the alleviation of pain. A steady fall in the temperature extended over another three weeks, together with diminution in the signs of fluid in the pleura. At the end of seventy-four days the man was sent home, some slight dulness at the left base, and contraction of the chest sufficient to influence the spine in the way of lateral curvature, being the only remaining signs.

[Illustration: TEMPERATURE CHART 4.--Primary Hæmothorax. Secondary rise of temperature, with increase in the effusion. Spontaneous recovery. Case No. 155]

    (155) Severe hæmothorax. Secondary effusion. Spontaneous recovery.--Wounded at Koodoosberg Drift, at a distance of 200 yards. Entry, at angle of the right scapula; exit, at the junction of the left anterior axillary fold with the chest-wall. No signs of spinal cord injury. The patient was brought in from the field twelve miles by an ambulance wagon on the second day, and in crossing the Modder River he was accidentally upset into the stream. For the first four days there was no hæmoptysis, but for the succeeding nine days small brightish red clots were expectorated. There was some tenderness over the ribs from the fifth to the ninth in the axillary line, and on the ninth day some gravitation ecchymosis appeared over the same region. Cough was an early troublesome symptom in this case, and when admitted to the Base hospital, about the seventh day, there was evidence of fluid extending about a third of the way up the back.

    On the tenth day after admission a pleural rub was detected at the upper margin of the dulness, and the latter shortly extended upwards over a little more than half the back. Meanwhile, there was no further hæmoptysis, respiration was fairly easy, 24 per minute, but accompanied by slight dilatation of the alæ nasi, and the temperature, which had been ranging from 99° to 100°, began to rise steadily, on the fifteenth day reaching 102.5°. The patient refused even an exploratory puncture, and was treated on the expectant plan. The temperature slowly subsided, with a steady improvement in the physical signs, and at the end of about ten weeks he left for home with only slight dulness and incapacity for active exertion remaining. (Now again on active service.)

[Illustration: TEMPERATURE CHART 5.--Hæmothorax, primary and secondary rises of temperature, on each occasion falling on the evacuation of the blood. Case No. 156]

    (156) Severe hæmothorax. Recurrent secondary effusion. Tapping on two occasions. Cure.--The patient was wounded at Paardeberg, and arrived at the Base on the eighteenth day. Entry, below the first rib, just external to its junction with the costal cartilage; exit, through the ninth rib, just within the posterior axillary line. The whole right side of the chest was dull, with signs of the presence of fluid, the heart being displaced to the left. There was considerable distress; the respirations averaged 40, the pulse 100, and the temperature reached 101.5° the first evening after arrival.

    On the nineteenth day the thorax was aspirated (Mr. Hanwell) and 50 ounces of dirty red-coloured fluid, half clot, half serum, were evacuated. Considerable relief was afforded; the respirations became slightly less frequent; the heart returned to a normal position, and distant tubular respiration was audible. The temperature dropped to normal the third day after evacuation of the fluid, but on the sixth day it again commenced to rise, and meanwhile fluid again began to collect.

    On the twenty-sixth day a second aspiration resulted in the evacuation of 35 ounces of bloody fluid in which flakes of lymph were found. Three days later the temperature became normal. The respirations fell to 22, and the patient made an uninterrupted recovery.

[Illustration: TEMPERATURE CHART 6.--Wound of Lung. Secondary development of Hæmothorax, with rise of temperature. Spontaneous recovery. Case No 157]

    (157) Moderate hæmothorax. Secondary effusion at the end of twenty days. Spontaneous recovery.--Wounded at Paardeberg; range from 700 to 1,000 yards. Entry, in the centre of the second right intercostal space, anteriorly; exit, at the level of the sixth rib posteriorly, through the scapula, close to its vertebral margin.

    The patient arrived at the Base on the sixth day; he said he expectorated some blood at the end of about ten minutes after being shot, and experienced a 'half-choking sensation.' A small quantity of phlegm and occasional clots had been expectorated since. He had walked about a good deal; movement occasioned cough, and he became 'blown' very rapidly.

    On admission there were signs of fluid in the lower third of the pleural cavity, but no general symptoms beyond an evening rise of temperature to an average of 99°. About the twentieth day the temperature commenced to rise, and on the twenty-third and four following evenings reached 102°. The fever was accompanied by some distress, and a well-marked increase in the physical signs of the presence of fluid in the chest. The pulse rose to 96, and the respirations considerably above the average of 24, which was at first noted. A strictly expectant attitude was maintained, and the temperature steadily fell in a curve corresponding to the rise, gradually reaching the normal at the end of a week. The physical signs at the base steadily cleared up, and at the end of six weeks the patient returned to England convalescent.