U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter XV





The discovery of dichlorethylsulphide and the recognition of its toxic action have been usually ascribed to Victor Meyer in 1886.1 A more intensive searching of chemical literature than was possible during war time carries knowledge of this subject back to 1860 and to the researches of an English chemist, F. Guthrie,2 who contributed a study on some of the derivatives of the olefines, including an investigation of the action of sulphur chloride on ethylene, and studied the product of this action, which he designated Aethylendichlorosulfid, and to which he gave the formula C4H4S2Cl2, described this product as having a pungent, not unpleasant odor similar to mustard oil, with an astringent taste like that of horse-radish. As to its pathologic action, he noted that the small amount of vapor arising from it attacked the more tender parts of the skin, as, for example, between the fingers and around the eyes, and destroyed the epidermis. When applied in liquid form to the skin there developed a blister.

In the same year, A. Niemann 3 studied the action of the brown sulphur chloride upon ethylene, and obtained an impure product to which he gave the formula C4H4ClS2, as most closely corresponding to the results of his analysis. He described this as possessing an odor similar to horseradish oil, although less penetrating. The most characteristic property of the substance, however, was that regarded by him as most dangerous, in that the slightest trace of the liquid coming into contact with the skin caused at first no pain; but after several hours there developed at the point of contact a reddening, followed on the next day by the formation of a blister which very slowly suppurated, and which was extremely slow in healing, with the production of a marked scar. He noted the occurrence of these toxic effects in the same manner in different individuals, and warned against the danger of working with this product.

No further work upon the action of sulphur chloride upon ethylene appears to have been carried out until Victor Meyer;1 succeeded, in 1886, in producing a pure dichlorethylsulphide, to which he gave the name thiodiglykolchloride, with the formula:


In addition to a description of the chemical properties of this substance, Meyer noted also observations upon its toxic action. He considered it as most striking that this substance, apparently harmless, so slightly volatile, and almost in- soluble in water, of very slight odor and neutral in reaction, and from its chemical

a The material forming the main body of this chapter is taken from the investigations of Warthin and Weller, published as "The Medical Aspects of Mustard Gas Poisoning," C. V. Mosby Co., St. Louis, 1919. These experimental studies first appeared in the Journal of Laboratory and Clinical Medicine during 1918-1919; the first study, that on mustard- gas lesions of the skin, was published independently. After its publication the pathological laboratory of the University of Michigan became allied with the medical advisory board of the Chemical Warfare Service, and the first reports of the remaining investigations were given that board and published in abstract in its bulletins, later appearing in the journal mentioned. The literature of the subject was consulted, and differences of opinion and new observations were given full consideration.


constitution giving no expectation of aggressive properties, should exert such a specific toxic effect as it produced upon one of the workers in his laboratory in the form of severe skin lesions and a transitory inflammation of the eyes. As Meyer himself was not affected by his work with it, he was at first inclined to believe that its toxic effects were due to some individual idiosyncrasy. At his request animal experiments were carried out with the following results: Two medium-sized rabbits were confined three to four hours in a closed chamber, ventilated by a moderately strong current of air entering through a glass tube containing strips of filter paper saturated with thiodiglykolchloride (dichlorethylsulphide). The animals became restless, and rubbed their noses and mouths frequently with their paws. The nose, mouth, and conjunctiva became very much reddened and the eyes very moist. Perspiration appeared to be increased. On the following day the eyes were severely inflamed, the lids glued together by purulent secretion. Marked snuffles developed, the lobes of the ears were much swollen and the external auditory passage showed a purulent inflammation. On the evening of the third day the animals died. and the lungs of both showed a severe diffuse pneumonia. A larger-sized rabbit that had inhaled the vapor of the substance through a tracheal fistula for a few hours, with the surface of its body protected from contact with this vapor, died on the evening of the same day from a diffuse pneumonia, before other symptoms had had time to appear. In rabbits whose ear-tips had been painted upon the unbroken skin with a small amount of the chloride by meints of a fine pencil, no direct local effects were seen, but the entire ear becanme greatly swollen, and in one case a profuse purulent inflammation developed. Because of the small amount used and its external application, any entrance of the material into the external canal was excluded. In one case in whiich the hairs had been removed from the tip of the ear the local touching with the chloride produced a suppuration at the point of application, with an associated diffuse swelling of the whole ear and inflammation of the eyes. After subcutaneous injection of about two drops of the chloride into the back of a rabbit there developed all inflanmnation of both eves and severe snuffles, with death on the third day from pneumonia. No effects were noted at the point of application. Because of the fact that the vapor of the chloride affected those, engaged in this experimental work, it was discontinued. Meyer concluded that the toxic action reached its height only after its entrance into the blood stream.

During the next year Meyer4 made a second contribution on the physiologic action of dichlorethvlsulphide. Through further investigations he had found that the slightest trace of this substance produced on human skin severe and prolonged inflaminations. Rabbits, after a short exposure to its vapor, died regularly with pneumonia, and the local application of a trace of the substance to the ear produced severe inflammation of the ear and eyes, with marked swelling of the former. He noted also that animals surviving such a local application showed a persistent profuse suppuration, leading, after several weeks, to a complete necrosis and loss of the ear. Physiologic tests were made also with the monochlorethylsulphide, which was found to possess similar, although less intense, toxic properties. Chlorine-free ethylsulphide was also tested as to its physiologic action and found to be nontoxic. Meyer decided, therefore, that the toxicity of the mono- and dichlorethylsulphides depends entirely upon their chlorine content.


The remarkable toxic properties of this apparently nonaggressive substance seem strangely to have escaped the notice or interest of pharmacologists. It is apparently not mentioned in the literature again until 1891, when Th. Leber,5 on the ground of the two experimental studies of its action on the eye which he had made for Meyer in 1886 and 1887, concluded that dichlorethylsulphide belongs to the pus-producing class of inflammatory irritants.

It again falls back into obscurity for a period of over 20 years. No mention of the physiologic action of this substance is found again until 1912, when H. T. Clarke,6 corroborating Meyer's findings concerning the poisonous action of dichlorethylsulphide, made the following statement:

That its action is toxic, and not merely irritative, as is the case with acids, is shown by the absence of pungency in the odor and by the fact that it takes effect only after some hours have elapsed. It can, nevertheless, be handled with perfect safety, provided that care is taken not to inhale its vapor, or to allow it to come into contact with the skin.

After another period of neglect, as far as the literature of the subject is concerned, dichlorethylsulphide was forced upon the attention of a war-distraught world in a most sensational manner through its successful employment on a large scale by the German Army as a war gas, at Ypres, July 12 and 13, 1917. 7 From that time to the end of the war it proved to be their most effective war weapon, although the protective measures devised against it by the Allies led to a great reduction in the number of casualties caused by it. Within eight days after the first bombardment with this substance the peculiar physiologic effects produced by it led Lieut. Col. Harrison and Professor Baker to conclude that the active constituent of the gas shell must be dichlorethylsulphide, as described by Victor Meyer, and this supposition was confirmed a few days later by the chemical analysis of the shell liquid.

The recognition of the new poison gas as dichlorethylsulphide naturally developed an intense interest among the scientific men connected with the Allied armies, as well as in various scientific institutions and laboratories not directly connected with the armies, but engaged on problems of research relating to the war. In the pathological laboratory of the University of Michigan, at Ann Arbor, there had been begun an intensive study of poison gases that might be employed in warfare, as the especial war-problem of this laboratory.b Studies on chlorine were being prosecuted when news of the new gas was received. The laboratory was soon placed in possession of a quantity of pure dichlorethylsulphide through the kindness and skill of Dr. Moses Gomberg, of the chemical laboratory of the university.

Pure beta-beta-dichlorethylsulphide as prepared by Gomberg was a clear, colorless, heavy, oily fluid, having a very faint cress or mustardlike odor, more like that of a freshly bruised plant than that of prepared mustard. The boiling point was 217° C. Samples of this preparation were unchanged when nearly five years old. Later, other samples of mustard gas were received from various sources, and these were all more or less impure, having a slight or marked yellowish or brownish color and a more decided odor. With age these impure samples became more brownish and more odorous. The pure substance caused no staining of tissues with which it came in contact; the impure substance sometimes stained tissues a bright yellow. These impure samples varied between 60 to 85 percent of dichlorethylsulphide content and were, therefore, comparable to the percentages actually employed by the Germans in their yellow cross and green cross shell.

b These investigations were made by Doctor Warthin and members of his staff, Dr. C. V. Weller find Dr. George R. Hermann.


The studies on which this chapter is based took up, in turn, the pathologic action of mustard gas on the skin, the eye, the respiratory and gastrointestinal tracts, and finally its general pathologic action. They were concerned with the gross and microscopic lesions produced both in man and animals by the direct application of the liquid dichlorethylsulphide or by exposure to its vapor. The human material was obtained through auto-applications, amputation material with consent of the patient, and accidental lesions in laboratory workers or in men engaged in the manufacture of dichlorethylsulphide. By means of auto-applications complete series of lesions in all stages from inception to healing were studied. In factory workers all types of mustard-gas burns were observed, from the mildest forms to those of the greatest severity ending fatally. Lesions were produced in all possible ways by the direct application of pure mustard gas and its solutions of all possible strengths, and by exposure to varying concentrations of its vapors. To standardize the results a method of applying a known quantity of liquid mustard gas was devised by the use of a standard capillary pipette which would deliver uniform droplets estimated at about 0.0004 c.c. For the vapor exposures an especial gassing chamber was constructed for animal work, whereby known concentrations of the gas could be employed. (See fig. 190, p. 596.)

Mustard gas is essentially a local poison, exerting its action directly upon those tissues with which it comes into contact. The very mildest concentrations of its vapor usually affect the eyes first, then the skin and the respiratory tract. Stronger concentrations of its vapor usually affect the eyes first, then the skin and the respiratory tract. Stronger concentrations may act first upon the skin, particularly upon delicate or sweaty cutaneous areas as the genitalia and flexor surfaces; or skin, eyes, and respiratory tract may be affected at the same time. The cutaneous lesions, on the whole, are the most common and striking; the respiratory lesions, however, are the ones most likely to produce serious results, when they are severe. The toxic action of mustard gas on the skin, eye, respiratory and gastrointestinal tracts, and its general pathology are given in order. When a series of animals was used for individual experiments one typical protocol alone is given to illustrate it, but the final conclusions are based upon the total number of controlling investigations carried out.



Dichlorethylsulphide was applied by means of a capillary pipette in uniform droplets estimated to be about 0.0004 c.c. When applied to the skin this drop at once spread out over an area 3 to 4 mm. in diameter and was completely volatilized, or at least disappeared, in one to two minutes, according to atmospheric conditions.

Following is a protocol of the most important stages in the development of the skin lesion produced in this manner. It was found that the rate of production of the lesion and the intensity of the reaction varied over a considerable range in different individuals; notable in a Charcot-joint leg and in a case of malignant disease the development of the vesicle was delayed.




March 12, 1918, 2.30 p. m. Standard droplet applied to flexor surface of left forearm. In one minute the liquid completely disappeared, giving off a strong odor. There were no subjective symptoms. In about 10 minutes there appeared a delicate silvery gray sheen over the surface of the area of application. This was soon followed by a faint flush, which gradually deepened and spread until it was about 7 mm. across. Photograph (fig. 92) was taken one hour after application. At this stage the erythema was influenced by changes of temperature, etc., alternately paling and reddening. Whenever the area became somewhat paler the superficial silvery luster was visible. During the second and third hour a well-marked edema appeared and the erythematous zone became wider and more deeply colored.

FIG. 92.- Mustard-gas lesion at one hour

FIG. 93.- Mustard gas lesion at three hours

Same day, 5.30 p. m. At this time the lesion measured 1.1 mm. in diameter. It was slightly elevated and had a marginal zone somewhat deeper red than the central portion. Outside of the red zone there was a very faint, barely perceptible zone showing less color than the remainder of the skin. (Fig. 93, photograph three hours after application.) In the next several hours there was no change.

March 13, 1918, 6.30 a. m. Sixteen hours after the application a vesicle began to form. At 8.30 a. m. 18 hours after the application, this was at its height. (Fig. 94.) At this time the lesion measured 25 ml. in diameter. It presented an erythematous base slightly elevated, ad fading gradually into the surrounding skin. Upon the summit of this erythematous base there rose a tense vesicle 6 by 9 mm. in area and 3 mm. high. This was


filled with a clear pale-yellow fluid. Ip to this time there had been no subjective symptoms, but with the formation of the vesicle there was slight smarting, increased by pressure.

Same day, 12.30 p. m.- 22 hours after application. (Fig. 95.) At this time the vesicle covered no greater area but was 4 mm. high and still very tense. Its fluid content had become slightly cloudy or opalescent. With a hand lens the base of vesicle could be seen through the fluid content and appeared yellowish-white, opaque, and necrotic. Around the border of the inflamed base of the vesicle there was a definite secondary areola.

March 14, 1918, 12.30 p. m.- 46 hours after application. (Fig. 96.) By this time there was nearly complete absorption of the fluid from the vesicle. The epidermal covering of the vesicle was thrown into fine wrinkles and folds and had taken on a yellowish-brown

FIG. 94.- Mustard-gas lesion at 18 hours

FIG. 95.- Mustard-gas lesion at 22 hours

tint. The zone immediately around the base of the vesicle was pale pinkish-white and about 1 mm. in width. Outside of this the flushed areola persisted. The total width of the lesion was but 17 mm.

March 15, 1918, 2.30 p. m.- 72 hours after application. (Fig. 97.) The central portion of the lesion measured 8 by 4 mm. and was of a bright yellowish-brown color. Around this was a white zone 1 to 2 mm. in width. Outside of this was a zone of erythema most marked about the base of the collapsed vesicle and fading peripherally. The total width of the lesion was 18 mm.

During the night of March 15 to March 16 the delicate wrinkled epidermis was rubbed off, leaving an excavated area with a grayish yellow-white moist base. The excavated area measured 6 by 4 mm. The border was somewhat irregular and slightly overhanging. The white zone, about 1 mm. in width, still persisted at the border, and the erythematous zone outside of this was about the same width as before. The base became slightly glossy upon drying. Figure 98 was taken at 5.30 p. m., March 16, 1918.


March 17,1918, 9 p m.-Total excavated area somewhat diminished. Border smoother. Floor not quite so deep. Total width of lesion was 15 mm.

March 21, 1918, 4 p. m. A brownish crust representing the necrotic base was beginning to loosen at the edges, where it was white and slightly desquamating. About this was a marked erythematous areola, 4 mm. wide. (Fig. 99.)

March 22, 1918, 9 a. m. During the night there was marked itching, the only pronounced subjective symptom so far noted. The crust loosened entirely and came off. Beneath the crust there was a small amount of thin purulent fluid. The erythematous zone was less marked. The base of the excavated area was again dry and covered with a yellow- ish-brown crust. The margin of the excavation was whitish, with edges slightly puckered. The central portion of the lesion measured 5 by 9 mm.

FIG. 96.- Mustard-gas lesion at 46 hours

FIG. 97.- Mustard-gas lesion at 72 hours

March 30, 1918. Since March 22, the area of excavation had gradually become covered with a yellowish-brown crust which had become elevated nearly 2 mm. above the surrounding surface. The zone of erythema was fading, leaving a yellowish-brown pigmentation. The inner portion of this zone was shiny and somewhat puckered, and there were a few minute desquamating scales at the border of the dense crust. The entire width of the lesion, including the zone of pigmentation, was 3 cm., the reddened, somewhat shiny zone, was 2 to 3 mm. in width, the scaly desquamation about 1 him., and the elevated crust 4 by 7 mm. (Fig. 100.)

April 1, 1918. The dark crust or scab became loose and came away leaving a white, dry, slightly granular area nearly flush with the surface of the skin. This measured 4 by 8 mm. (Fig. 101.)


May 1, 1918. Healing was now nearly complete, the lesion consisting of a thin scar, pinkish-white in the central portion and whiter, more opaque, at the margin, with very slight puckering. Around this was a brown pigmented areola. The whole area, however, was redder than normal skin. (Fig. 102.)


The changes in human skin were studied microscopically from one-half hour up to four weeks, including the development of the vesicle and beginning eschar formation. As the lesions at one-half hour, 18 hours, and 36 hours represented three distinct stages, these will be described in detail.

FIG. 98.- Mustard-gas lesion at four days

FIG. 99.- Mustard-gas lesion at nine days


Epidermis.-The horny layer was relatively thicker than normal and split up into flat scales and layers, loosening readily from the stratum lucidum. The stratum lucidum had a slight brownish color. The granular layer was flattened; the cells drawn out parallel to the surface: the nuclei were pycnotic. The stratum germinativum was markedly shrunken, in many places only one- third to one-half as wide as normal, its nuclei pycnotic and the cytoplasm shrunken about the nuclei. Occasional vacuoles were found in thie lowest layers, but the most marked change was the shrunken appearance of the whole epidermis, both cytoplasm and nuclei. At the border of the lesion the epidermis passed gradually into the normal condition.

Papillary layer of corium.- In the central part of the lesion the capillaries were contracted and contained but little blood. In scattered capillaries the


red blood cells were agglutinated and stained with eosin as bright red hyaline masses. Such agglutination thrombi. however, were not a common feature of the picture, and in the larger vessels thrombosis did not occur. The endothelium of the capillaries of the papillary layer showed marked pycnosis, caryorrhexis, and disintegration of the nuclear chromatin. Chromatin dust was found around many of these capillaries and also many of the connective-tissue cells of the upper portion of the papillary layer showed marked caryorrhexis. The cytoplasm of many of the endothelial cells of the capillaries was vacuolated, showing hydropic degeneration or edema. About these capillaries there

FIG. 100.- Mustard-gas lesion at 18 days

FIG. 101.- Mustard-gas lesion at 20 days

was a clear space due to a perivascular edema. Around some capillaries this was very marked. Many of the capillaries showed diapedesis of leucocytes along their course (fig. 104), but in the central part of the lesion there was practically no hemorrhage and the vessels were conspicuous for their contraction and anemia.

Corium proper.-The vessels running through the corium showed similar changes in their endothelium but there was little leucocytosis or white cell migration. The larger vessels contained more blood. No thromboses or hemorrhages were present. The lymphatics were dilated and the nerve trunks showed caryorrhectic nuclei and edema


Hair follicles.- Along the hair follicles the squamous epithelium showed changes similar to those of the surface and the capillaries about the hair follicles also showed changes similar to those described above. In the neighborhood of the hair follicles the corium was affected more deeply than elsewhere, showing a distinct penetration through the hair follicles.

Sweat glands.-The epithelium of the sweat glands showed no apparent changes, although the vessels about them showed changes similar to those described above.

Sebaceous glands.-Shrinking and pycnosis of cells similar to that seen on the surface.

In the transition border there were small hemorrhages by diapedesis.

FlG. 102.- Mustard-gas lesion at 49 days

FIG. 103.-Typical mustard-gas vesicle about 20 hours after application


Epidermis.-In the central part of the lesion there was a marked liquefaction and hydropic change in the cytoplasm of the epithelium. This varied greatly in degree. The horny layer was in part desquamated and loosened. the stratum lucidum was widened and more dense than normal and stained brownish-red. Over mans of the papillae small vesicles had already formed, the majority of the epithelial cells having undergone liquefaction. In some places the epidermis was lifted from the papillae by the collection of fluid beneath it.


FIG. 104.- Lesion of human skin one-half hour after application of mustard gas

FIG. 105.- Human skin 18 hours after application of mustard gas. Transition between slightly damaged epithelium and epithelium showing hydropic degeneration. Early blister formation.


FIG. 106.- Human skin 18 hours after application of mustard gas. Early vesicle formation

FIG. 107.- Human skin 18 hours after application of mustard gas. High-power view of hydropic change with early vesicle formation


The liquefaction of the cell cytoplasm extended deep down into the hair follicles and into the sebaceous glands. The stratum germinativum had lost its continuity in many places and the cells were completely necrotic.

Papillary layer of the corium.- The connective tissue was edematous, stained bluish, and contained many degenerating nuclei. There was an increase in the number of wandering cells and many of these showed carvorrhexis. Around all the capillaries there was a zone of edema and small-celled infiltration. Small hemorrhages by diapedesis were scattered through the papillary laver and upper portion of corium, particularly around the hair follicles.

Corium proper.- The blood vessels contained more blood than in the earlier lesion, particularly the deeper ones. The larger ones showed a marked congestion and the lymphatics were dilated with a lymph rich in albumin.

FIG. 108.- Human skin 18 hours after application of mustard gas. High-power view of small vesicles. Separation of epidermis from papillary layer

Around the hair follicles the edema, liquefactive changes, and hemorrhages were more marked than elsewhere. The sweat glands showed a marked edema of the interstitial connective tissue, congestion of the capillaries, leucocyte infiltration, and small hemorrhages by diapedesis. Some of the glands showed a marked necrosis of the epithelium but these changes varied greatly in degree.

Subcutaneous tissues.-The vessels were congested, lymphatics dilated, and there was edema of the adipose tissue. The vascular changes extended along the smaller capillaries even into the subcutaneous tissue.


Epidermis.-The horny laver was more compact but ragged, and in many places infiltrated with leucocytes. The epidermis was nearly completely necrosed, in some places being lifted from the papillary layer. The remaining


nuclei were markedly pycnotic or fragmented. In many areas only the lowest layer of nuclei persisted. There were also collections of fluid between the horny layer and the portions of rete remaining.

Papillary layer.-The papillary layer showed marked edema, the capillaries were congested and there were many hemorrhages by diapedesis. The entire papillary layer was infiltrated with leucocytes, many of which showed caryorrhexis.

Corium.-There was a leucocyte infiltration throughout the entire corium, but less marked than in the papillary layer. It was most marked around the hair follicles and around the sebaceous glands and sweat glands. The congestion and edema were also most marked around these structures. Some of

FIG. 109.- Human skin 36 hours after application of mustard-gas vesicle formation in epidermis and leucocyte infiltration of papillae

the smaller vessels showed marked necrosis of the wall with leucocyte infiltration and diapedesis. Scattered areas of edema and small-celled infiltration extended even into the subcutaneous tissue where the vessels were markedly congested.


Inasmuch as the later stages in human skin paralleled those in the lower animals and as the specific differences between the action of mustard gas on human skin and on the skin of the rabbit, guinea pig, and cat existed only in the early stages, it seems advisable here to omit a more detailed description of these changes and to summarize them as follows:

1. About 40 to 50 hours after application, collapse of vesicles and progressive necrosis. 2. About 72 hours after application, progressive necrosis and beginning eschar formation. 3. Four to six days after application, necrosis Completed, beginning separation of slough. Edema and hyperemia persistent


The microscopic appearances of a droplet lesion on human skin at one week after application are shown in Figure 111. In the center the epidermis was completely necrosed and desquamated; the necrosis extended some distance into the corium; the surrounding area was intensely hyperemic; there was moderate edema, very little leucocyte infiltration, and no evidences of repair.4 By the 19th day, complete separation of slough. Slow healing and scar formation.5 For an indefinite period, congestion and pigmentation.

All of these descriptions apply to the effect of a standard drop of pure mustard gas in the absence of infection. The course naturally varied with the concentration, amount, time, and other factors. In very mild mustard-gas lesions, which appeared only as hyperemias, with intact epidermis, the microscopic examination might show a complete necrosis of the papillary layer of the corium with the exception of the chromatophores, which became larger,

FIG. 110.- Human skin 36 hours after application of mustard gas. High-power view at border of lesion, showing changes in epidermis and leucocyte infiltration and edema of papillary layer

increased in number, and heavily pigmented. The pigmentation was due to melanin and not to blood pigment. This necrosis of the upper part of the corium explained the secondary production of vesicles through slight trauma.


The rabbit, guinea pig, and cat were employed for these experiments. The skin of the abdomen was shaved and standard drops were applied after the irritation from shaving had subsided. The character of the tissue lesion and the reaction in these animals proved, in the early stages, to be essentially different from the lesions in human skin. For the purpose of brevity and conciseness the protocols are condensed as below.

Rabbit.- Within two hours after application of the standard drop there developed a very marked edema, much larger than the area touched by the


drop of mustard gas. This edema was subcutaneous, appearing as a definite tumor mass, rather sharply circumscribed, over which the cuticle might be moved. The surface of the area appeared gray and cloudy, the skin losing its normal translucency and appearing as if it had been cooked. In some cases the blanching appeared to extend into the deeper portion of the skin. About this gray area there was but slight hyperemia. By the third day after the application the epidermis over the area had undergone complete necrosis and there was formed a slough without any vesicle formation. Vesiclts were never observed in the rabbit. This slough was held on apparently by the hairs. It gradually became elevated, then separated and contracted, and might not be cast off for three or four weeks. When this slough disappeared, the lesion below was practically healed. The most striking features were the marked edema at the beginning, the persistence of this edema without vesicle formation, and the slow healing in the absence of infection.

FIG. 111.- Droplet lesion of mustard gas on human skin seven days after application. Low-power view showing area of necrosis of epidermis and upper portion of corium, with intense hyperemia of the surrounding vessels. Moderate edema and very little small-celled infiltration

Various protective experiments were tried out, a number of substances being used to prevent the lesion or lessen its severity--washing with water, soap and water, lead acetate, lead acetate and silver nitrate, zinc oxide ointment, zinc oxide paste, bleaching powder, sodium sulphide, tincture of green soap, potassium permanganate. The application of most of these substances five minutes after the application of the mustard gas lessened the edema and rendered the lesion more diffuse but did not prevent necrosis. The use of potassium permanganate resulted in even greater edema than the untreated control while strong sodium sulphide solution was found to be disadvantageous because of the necrosis produced. (Figs. 112, 113, and 114.)

Guinea pig.- Practically identical results were obtained in the guinea pig with the standard drop of pure mustard gas. namely, within a few hours marked subcutaneous edema, followed in a few hours by necrosis of epidermis and papillary layer without vesicle formation and exhlibiting very slow healing.


Cat.- The same results were obtained as for rabbits and guinea pigs. Subcutaneous edema without vesicle formation, followed by necrosis and slow healing.


The changes observed in the development of the lesions in the rabbit, guinea pig, and cat were essentially the same.


The most striking feature of this stage was the intense edema which, although sharply localized to the subcutaneous tissue and fascia, extended also

FIG. 112.- Rabbit. Application of mustard gas at 11:30 a.m. Droplet used was slightly larger than the standard . Marked subcutaneous edema as seen at 4 p.m. on the same day

Into the muscle of the abdominal wall. The connective-tissue fibrillae were widely separated and the tissue spaces filled with a heavy albuminous precipitate staining deep pink with eosin. The muscle fibers of the abdominal wall were separated, and even in two hours there was a leucocytic infiltration into the muscle. The edema extended 0.5 to 1 cm. below the epidermis. The epidermis was shrunken, ceils pycnotic, andi in the central portion of the lesion completely necrotic. In the upper layer of the corium numerous degenerating, nuclei were seen. The b1ood vessels showed degeneration of the endothelium


with small hemorrhages by diapedesis and leucocvte migration. Around each vessel there was an area of edema. There was, however, no vesicle forma- tion in or beneath the epidermis as in the human cases. The changes were more uniformly diffuse in the animal than in the human skin and the depth

FIG. 113.- Rabbit. Skin of belly shows results of four applications of standard drops of mustard gas. Above two areas of typical edema, the one on the rabbit's right untreated, the one to the left washed off in five minutes by water. The latter is more diffuse, larger in area, but less intense. Below, on the rabbit's right, an area washed after five minutes with soap containing an excess of free alkali. This area shows the least reaction. On the lower left is an area treated. after five minutes, with potassium permanganate. The reaction here is most marked.

of penetration greater. The localizedl pe netration along the hair follicles, so prominent in the human skin, did not show in the animal skin, the greater number of hair follicles permitting it more uniform access of the liquid. There were no thromboses in the damaged area. The hemorrhages were relatively small


and, as in the human skin, the vessels in the immediate lesion were contracted and anemic. Around the borders of the lesion they showed marked congestion.


The necrosis of the epidermis and of the underlying tissues steadily became more prominent because of the loss of nuclei in the epidermis and upper part of the corium of the central part of the lesion. By the fifth and sixth day after the application of the mustard gas the edema had subsided to a marked degree, and the central part of the lesion might be entirely without nuclei as far as the lower portion of the corium. It was bloodless, rather dry, and there

FIG. 114.- Rabbit. Two areas of mustard-gas application. Advanced eschar formation

was but little leucocyte infiltration. Surrounding this was a narrow zone of less marked necrosis and degeneration. This gradually became hyperemic and small hemorrhages by diapedesis took place from the damaged vessels. In this area there was a more marked infiltration of leucocytes but this rarely became diffuse, the leucocytes remaining collected in the neighborhood of the vessels. Outside of this zone the tissues were hyperemic, somewhat edematous, and showed an increased number of wandering cells for some distance.


There gradually began a separation of the completely necrotic tissue from the living. This eschar consisted of the dead epidermis and upper part of the


corium, sometimes as far down as the lower borders of the hair follicles. This dried, contracted, and became leathery, but was held in position by the hairs. There now developed in the neighboring living tissue a productive inflammation. The shrinking of the necrosed tissue and the demarcation, with the

FIG. 115.- Rabbit. Low-power view of mustard-gas lesion in rabbit two hours after application. Extreme subcutaneous edema. Epidermis but slightly changed

FIG. 116.- Guinea pig. Low-power views of mustard-gas lesion five and one-half hours after application. Extreme subcutaneous edema Epidermis necrosed in center of lesion

surrounding reparative inflammation, progressed very slowly until there was a regeneration of the epithelium beneath the eschar. The latter remained adherent, usually until complete repair had taken place. The repair of the epidermis occurred chiefly from the cells remaining in the hair follicles.

(For the histologic changes in animal tissues see figs. 11.5 to 126, inclusive.)


FIG. 117.- Border of lesion two hours after application of mustard gas. To the right of the middle, the epidermis is still living; to the left nearly completely necrosed, necrosis extending into the upper portion of the corium. Early edema

FIG. 118.- Rabbit. Two hours after application of mustard gas, changes in epidermis and corium. Marked vascular changes with beginning migration of leucocytes. Small hemhorrhages by diapedesis Early edema


FIG. 119.- Guinea pig five and one-half hours after application of mustard gas to skin of abdomen. Deep subcutaneous edema

FIG. 120.- Rabbit. Six days after application of mustard gas. Treatment with zinc oxide paste live minutes after use of mustard gas. Center of lesion, complete necrosis of epidermis, hair follicles, and upper portion of corium extending even to the sweat glands. No reaction


FIG. 121.- Rabbit. Six days after application of mustard gas. Treatment with zinc oxide paste five minutes after application. There was no edema stage. Epidermis is dead and there is a moderate inflammatory reaction in the corium. Reaction much less intense than in control

FIG. 122.- Rabbit. Six days after application of mustard gas. Treatment after five minutes with 2 per cent solution of silver nitrite and 5 percent lead acetate. Primary edema was nearly completely controlled but necrosis, six days later, is marked extending deeet into the coriuni, with more rapid separation of the slough


FIG. 123. - Rabbit. Periphery of same lesion as Figure 122. Area of less damage

FIG. 124.- Rabbit. Six days after application of mustard gas, untreated. Border of lesion. Necrosis less marked. Beginning repair


FIG. 125.- Rabbit. Six days after application untreated. Intermediate zone. Separation of necrotic epidermis and papillary layer, with infiltration of leucocytes into the necrotic tissue. Fibroblastic proliferation in lower part of dermis, with regeneration of hair follicles. Intense congestion of subcutaneous vessels.

FIG. 126.- Six days after application, untreated. Abherent slough, representing the necrotic epidermis and upper portion of the corium, involving the hair follicles.




Human skin exposed for varying periods of time to mild, medium, and strong concentrations of mustard gas developed lesions in proportion to the degree of concentration and time of exposure. These lesions were essentially the same as those produced by the direct application of the liquid gas. A third factor, that of susceptibility, appeared, however, to play a greater part in the cutaneous reaction to vapor, particularly to the mildest concentrations. This susceptibility was both a local and a general individual one. Local cutaneous susceptibility existed, particularly in the softer, more delicate areas of the skin, especially in the regions well supplied with sweat glands and exposed to friction. The skin of the genitalia and inner surfaces of thighs, although covered by clothing, was often the only part affected by the vapor. In other cases the skin between the fingers, flexor surfaces of elbows, axillae, and eyelids was affected, while other portions of the epidermis might show very slight effects or none at all. Individual susceptibility appeared also to exert an influence in determining the severity of the reaction. Individuals with soft delicate skin, particularly those of the lymphatic constitution, showed an especial susceptibility, while those with thick, coarse, or pigmented skin showed less susceptibility. Other physical conditions affecting the skin during or immediately after the exposure undoubtedly played a large part in determining the individual effects of the exposure. Warm sweaty skin was more susceptible pressure and friction upon the exposed areas were of the greatest importance in increasing the severity of the lesion. Slight rubbing would produce vesicles and eschar formation in parts exposed to vapor, when similarly exposed portions of the same area not rubbed would not pass beyond the hyperemic stage. Exposure of cutaneous areas to even the mildest vapors producing a barely recognizable hyperemia led to a remarkable tendency to pigmentation, much greater than that produced by exposure to sunlight or radioactivity. Frequent slight exposures might increase this pigmentation until the exposed individual became as dark as a negro. In many cases the pigmentation was very spotty. Exposures to very high dilutions (1:5,000,000) might produce itching of the scrotum and between the fingers without visible cutaneous lesions. Later pigmentation occurred in some instances.

The microscopic examination of cutaneous lesions resulting from exposure to mustard-gas vapor showed a pathology essentially the same as that resulting from the direct application of the liquid (figs. 127 to 138), in the stages of hyperemia, vesicle formation, eschar formation, and repair. The stage of pigmentation showed the production in the upper layers of the corium of large branching chromatophores that developed the pigment intracellularly. These cells were of endothelial or connective-tissue cell origin; never of epithelial. Later the cells of the rete came to contain the pigment also.


Exposures of shaved areas of animal skin to vapor of mustard gas produced lesions essentially the same as those caused by direct application of the liquid mustard gas, varying only in degree of severity in proportion to the duration


of exposure and concentration of the vapor. As with the direct application the lesions produced by the vapor in animal skin showed less tendency to vesicle formation than in the case of human skin, and the pigment formation was not a striking feature. Concentrations of 1:1,000,000 for a few minutes

FIG. 127.- Section of skin of penis, eight days after exposure to strong concentration of mustard-gas vapor. Intense hyperemia

FIG. 128.- Section of skin of scrotum, from same individual as in Figure 127. Necrosis of epidermis; intense hyperemia

might produce in rabbits a hyperemia and edema lasting for several days. The microscopical picture of the vapor lesions in animals was essentially the same, varying only in degree, as that of the lesions produced by the liquid applications.


FIG. 129.- Human skin one week after exposure to strong concentration of mustard-gas vapor. Microscopically, the changes consist of increased cornitication, pycnosis of the cells of the epidermis, and necrosis of the papillary layer of the corium. The only living cells in the upper portion of the corium are pigmented chromatophores

FIG. 130.- Human skin one week after exposure to strong concentration of mustard-gas vapor. Edge of large vesicle showing the necrosis of the upper portion of the corium, congestion of vessels, and separation of the epidermis


FIG. 131.- Section of skin from same case as Figure 130. Area of collapsed vesicles; necrosis of epidermis and corium; congestion of vessels

FIG. 132.- Skin from axilla of same patient as Figure l30.Necrosis of skin to the depth of the large sweatglands. These show also partial necrosis with some early regeneration


FIG. 133.- Infected gangrenous area from skin of back

FIG. 134.- Section of corium from skin of same patient as Figure 130, showing dilated lymphatics filled with fibrin thrombi, in the lower portion of the corium. Sone of these lymphatics contain partially hemolyzed red blood cells in small numbers


FIG. 135. - Microscopic section of eschar four weeks after exposure. Areas of regenerating epithelium from the sweat glands

FIG. 136.- Microscopic appearance of mustard-gas decubitus four weeks after exposure. Destruction or tissue too great for regeneration. Necrosis extends below the level of the sweat glands


FIG. 137.- Photomicrograph of regenerating epidermis under the wet Dakin and saline method of treatment, four weeks after injury. Note the regeneration of the epithelium from the remains of the hair follicles and sweat glands

FIG. 138.- Completely healed mustard-gas lesion four weeks after injury treated 12 days with grease method, with increasing infection and gangrene. Under the wet Dakin and saline method infection was checked promptly and healing begun. Regeneration of eipdermis from hair follicles and sweat glands



A large number of mustard gas lesions of the skin were observed in men engaged in the manufacture and handling of dichlorethylsulphide. Some of these cases were extremely severe, two of them terminating fatally. All possible decrees of cutaneous lesions were seen. From these the following seven cases are selected for illustration: c

CASE 1.- Pvt. Mc. (Case 3, pp. 590, 610). Seen one week after an exposure for 40 minutes, in four shifts. to a strong concentration of mustard gas. Patient wore gas mask, rubber gloves, and rubber boots. One hour after exposure, after eating supper, he felt slightly nauseated, went out doors and began to feel very warm. Rolling up his sleeves he found his skin to be very red, and on opening his shirt his chest was also seen to be very red. He immediately started for the emergency room, but on the way became sicker and more nauseated. He then took a kerosene rub and a hot-water shower with soap, but as soon as the water struck him he vomited his supper. Then he took another shower, after which he vomited almost continuously for about five hours, altogether "about 100 times." At the same time his eyes began to smart intensely, and for three days he was in intense agony from the eye pain. He developed a very dry throat, a hacking cough, and difficulty in speaking. His severe symptoms lasted for three days, but his voice remained husky and he had a bronchial cough for some time. The erythema of the skin persisted but vesication and necrosis of the epidermis did not show for two days. When seen a week later, he presented the appearances seen in Figure 139. Over the entire area of erythema there was a branny desquamation of the epidermis and the skin appeared markedly pigmented, the pigmentation being deepest over the neck and forehead, region of the nipples, wrists, lower portion of abdomen, pubic region, and legs. In the hairiest part of the axillae the lesions showed eschar formation. Eschars were present also at the bends of the elbows, around the genitals, and in the popliteal spaces. The whole back desquamated. Over the back there were numerous small pustules. In the small of the back, just below the belt line, there were areas of necrosis covered with a greenish-gray crust. There were also minute pustules over the forearms and legs, varying in size from a pinhead to a pea. There was an odor of putrefaction about the genitals. The surface of the entire glans penis was necrotic; the skin of the penis was edematous, desquamating, and showed numerous necrotic hair follicles. The skin of the scrotum was edematous, desquamating. with a necrotic gangrenous surface, and exuding pus from all of the follicles. The inguinal lymph nodes on both sides were enlarged. The surfaces of contact between the genitals and the inner surface of the left thigh showed a deeper necrosis than elsewhere. Over the legs there was a large flaky exfoliation of the epidermis. The umbilicus was necrotic and infected. The ocular lesions are described on page 590.

CASE 2.- Pvt. M. (Case 5, p. 610). Exposed to the same strong concentration of mustard gas one-half hour in one shift. Thirty minutes after the end of the exposure he became nauseated and his eyes were irritated and burned. He took the routine kerosene rub and showers, but soon developed severe photophobia with symptoms of shock. Within a few hours entire body was erythematokis. Vesication and desqualnation began on the second day. When seen one week after the exposure his skin showed a general erythema and pigmentation except where protected by his rubber gloves, rubber boots, and belt. The skin of the neck was almost black; the horny layer had completely desquamated. Pigmentation was very marked over the forearms, hands, and abdomen, increasing toward the pubic region. The part of the face protected by the gas mask contrasted markedly with the deep black pigmentation outside of the borders of the mask. The back was erythematous and desquamating. The skin of the small of the back was almost black, except for the white belt line running across it. The buttocks showed intense erythems and desquamation. In the right gluteal fold there was an area of deeper necrosis. In the axillae, at the bends of the elbows, and around the genitals there were deep necrotic areas. exuding pus and giving off an odor of putrefaction. There was deep necrosis of the glans pellis, of the scrotal skin, and of the contact surfaces of the genitals and skin of left thigh. There was a bilateral inguinal adenitis. There were numerous small pustules over the forearms and lower portion of the back.

c These seven cases were studied with respect, also, to lesions of the respiratory and gastrointestinal tracts.


FIG. 139.- Photograph taken one week after 40 minutes exposure to strong concentration of mustard-gas vapor Treated with grease method during this time with increasing intection and gangrene of epidermis. Change of treatment to the wet Dakins saline method effected prompt healing


CASE 3.-Pvt. H. (Case 7, p. 610). (Figs. 140 and 141.) Exposed 10 to 12 minutes in one shift to the same concentration of mustard gas as the above cases. First symptom was burning of the eyes; later mild shock, with vomiting; soreness of the throat and coryza. By the next morning had slight erythema over the body, most marked about the genitals. Vesication began on second day. When seen one week after exposure the face showed some pigmentation which was more marked in the neck, and the skin of the face and neck showed areas of desquamation with erythematous bases. The back was mottled with pigmented areas. Low on back the areas became confluent. Across the small of the back there was a sharply demarcated area which was protected by the belt. Over the pigmented areas there were flakes of desquamated epithelium. When these were removed an erythematous base appeared. The skin of the axilae was very erythematous with areas of deep necrosis in the creases where the hair was thickest. There were also areas of deep necrosis in the bends of the elbows. Pigmented bands were present in the groin parallel with Poupart's ligament. The skin of the penis was erythematous without necrosis. The skin of the scrotum was necrotic throughout, with an odor of putrefaction. Areas of necrosis were present in the crotch. The buttocks were erythematous, pigmented, and desquamating. There were areas of necrosis about the anus. The skin of the popliteal spaces showed erythema, desquamation and slight necrosis.

CASE 4.- Pvt. E. (Case 6, p. 610). (Figs. 142 and 143.) Exposed 10 minutes in one shift at the same concentration as the others. He claimed to be especially susceptible to mustard gas because he had had frequent burns previously from slight exposures. He first noted irritation of the eyes and found that his left arm was red to the elbow. He took an American oil rub, after which he developed a general erythema. Vesication began on the second day. When seen a week later, the skin showed a general pigmentation except where protected by the gloves, boots, and gas mask. The pigmentation was the most marked over the face, thorax, and abdomen. The back was erythematous, with large areas of collapsed vesicles over the scapulae. The axillae were erythemiatous, the skin desquamated, with deeper necrosis where the hairs were thickest in the folds. The deepest necroses on the upper part of the body were in the bends of the elbows. Wherever the skin had been rubbed over the bony prominences there were thick crusts of dead skin. Over the buttocks and the inner surface of the left thigh where the scrotum was in contact the skin was markedly erythematous, with branny desquamation over the entire body. The penis was markedly edematous and phimosed, with pus exuding from the meatus. The skin of the scrotum was completely necrosed and exfoliated. In the hair follicles there were numerous pustules. There were also small pustules over the forearms, shoulders, and back. Axillary lymph glands were enlarged. This patient had old mustard burn scars on his face, arms, and feet.

CASE 5.-Pvt. W. (Case 4, p. 610). (Figs. 144 and 145.) Exposed one-half hour in two or three shifts. Same concentration as in other cases. On coming from work he felt slightly ill and after eating supper was nauseated and vomited. He became very pale and showed severe shock. His throat was dry and sore and voice impaired. He had worn gas mask, rubber gloves, and rubber boots. He had also had mustard-gas burns of the hands previously. Showed general erythema a few hours after exposure. Vesication began on second day. When seen a week after exposure he was still in a state of severe shock and presented the signs of a diffuse bronchopneumonia. At this time his skin showed a generalized necrosis over the back, axille, chest, arms, abdomen, buttocks, groins, thighs, and popliteal spaces. The skin of the back exuded pus and there was a marked odor of gangrene over the entire surface, especially over the back. The chest and neck were deeply pigmented and the epidermis had exfoliated. Large patches of exfoliation were present over the chest and shoulders. The abdomen showed a marked line of demarcation corresponding to his belt. Everywhere a foul pus exuded from the necrotic surfaces. The genitals showed complete necrosis of the skin of the penis and scrotum, the surface being covered with a greenish gray slough, with numerous miliary abscesses corresponding to the hair follicles. The axillse, buttocks, and popliteal spaces showed the deepest necrosis. The portion of the face covered by the gas mask showed a milder lesion. For several weeks this patient was at the point of death, but after changing the method of treatment from the grease to the Dakin's solution the infection was finally conquered and regeneration slowly took place. When seen several months afterwards, the patient was engaged in office work and his skin showed extensive areas of cicatrization resembling those of severe thermal burns.


FIG.140.- Diffuse erythema the skin due to exposure for l to l2 minutes to stronge concentration of mustard gas. Treated one week by the grease method, with increasing infection of the dead skin, particularly around the genitals end arms Change of treatment to the wet Dakin and saline methods resulted in prompt healing


FIG. 141.- Rear view of same patient shown in Figure 140


CASE 6.- Pvt. Ha. (Case 1, p. 609.) Exposed one hour in two or three shifts at the same concentration. Almost immediately he became nauseated and pale, and vomited. He took a hot bath and kerosene rub, but quickly developed symptoms of severe shock. He vomited violently, had a severe diarrhea and extreme thirst. His face was pale but his body was very red. Two days later large vesicles developed over the back, chest, legs, and genitals. A condition of severe shock persisted. On the fourth day his temperature began to rise and an odor of putrefaction was noticed for the first time. The epidermis over his whole back was said to have been rubbed off at this time. By the fifth day the skin was discharging pus over the back, the patient became delirious, the pulse was rapid and thready, and the gangrenous condition of the skin increased. Death took place on the eleventh day. (For lesions of respiratory and gastrointestinal tracts see pp. 611, 620.) A description of the skin changes is given in full in the autopsy protocol on page 626.

FIG. 142.- Acute urethritis and phimosis due to mustard gas. One week after exposure

CASE 7.- Pvt. S. (Case 2, p. 610.) (Figs. 146, 147, and 148.) Exposed three-quarters of an hour in two or three shifts to the same concentration. On changing his clothes, hie noticed that his head, face, and neck were very red except where the mask had covered the skin. He took the kerosene rub and bath, but soon became nauseated and developed symptoms of severe shock. At the same time an intensely painful conjunctivitis and a severe irritation of the entire skin developed. During the next three days vesicles developed over the entire body. These were opened and drained. Temperature began to rise on the fourth day. On the sixth day there was marked exfoliation of the epidermis. When seen a week after the exposure his skin was gangrenous and infected throughout and his condition was very had. With the change of treatment to the Dakin's sponge bath and redluction of the infection his condlition improved somewhat. At one time it was thought that he was out of danger, but his skin was almost completely.


exfoliated and large areas of decubitus developed over the back. Regeneration of the surface epithelium did not take place, and with the increasing decubitus the patient died four weeks after the exposure. An autopsy was refused, but a careful study was made of his skin and the localization of the lesions carefully determined, as in Figure 146. The nature of these lesions was as follows: The entire back showed a deep necrosis extending through the corium into the subcutaneous tissue in many places. In part, the necrotic surface was covered by a purulent exudate. Over the anterior surface of the body there were large irregular areas of deep necrosis, exuding pus. Between the areas of deep necrosis the epidermis was completely lost, the denuded surface was congested and the capillary tufts in the papillae of the epidermis could be easily seen. The portion of the face covered by the mask showed a deep erythema, pigmentation, and desquamation, with areas of necrosis on the lips and at the angles of the mouth. The skin of the genitals was gangrenous and there was a foul odor of putrefaction.

FIG. 143.- Buttocks of same patient as Figure 142

The surface was covered with a thick tenacious eschar. greenish-yellow or grayish in color. When this was removed, a purulent hemorrhagic base was exposed. Over the sacrum there was a large area of decubitus extending to the bone. (Figs. 147 and 148.)

Microscopic examination of the necrotic skin showed a complete necrosis of the epidermis and corium, with secondary infection, the necrosis in many places extending through the corium to the sulbcutaneous tissues. No evidences of regeneration were seen in these areas of deepest necrosis, but in those portions of the skin where the necrosis involved only the upper part of the corium areas of regenerating epithelium from the sweat glands were present. In the skin of the scrotum the regeneration of the hair follicles formed small solid masses of epithelium as large as a millet seed, which could be distinctly seen and felt through the skin. These had been regarded its follicular abscesses, hut microscopic examination showed them to be regenerative areas of squamous epithelium.



In addition to these cases, about 70 cases of local mustard-gas lesions of the skin of varying degrees, both in acute and chronic stages and presenting a great variety of clinical pictures, were seen.

As the result of observations on these cases the following conclusions were drawn:

1. Even after slight mustard-gas erythemas, particularly when these have been several times repeated, there develops a dry, desquamative eczema or dermatitis, particularly between the fingers and on the genitals, which may be mistaken for itch. This chronic lesion, otherwise trifling, is especially annoying for its constant itching. The genital lesions may be mistaken for venereal affections.

FIG. 144.- Photograph 1 week after one-half hour exposure to strong concentration of mustard gas. During this time treated by the grease method. Photograph shows very well the protection afforded by the tight belt. The more marked lesions in the axillae, bends of the elbows and genitals, and the large, flaky character of the primary disquamation and the pigmentation are well shown

2. Slight mustard-gas lesions, not proceeding beyond the stages of erythema and pigmentation, may often be made to vesicate by slight trauma, such as rubbing, pressure, or by being struck. This phenomenon, which is analogous to Nikolsky's sign, has been misinterpreted as indicating a persistence of the mustardl gas in the injured area with secondary action, or as a delayed reaction. It is due entirely to the lowered vitality or partial necrosis of the papillary layer of the dermis with the epidermis adherent, so that relatively slight injuries cause the epidermis to separate and form a vesicle.
3. It has been stated that the fluid of a mustard-gas vesicle when applied to uninjured areas of the epidermis will produce fresh lesions. This is not true. The fluid of these vesicles is absolutely without action upon either the skin or conjunctiva.
4. No specific susceptibility to mustard-gas lesions of the skin is acquired by repeated exposures. Healed mustard-gas lesions, like all other healed lesions, have a lower degree of vitality and are more susceptible to all forms of injury.


FIG. 145.- Back of patient shown in Figure 144


FIG. 146.- Diagram illustrating the distributon of mustard-gas eschars, four weeks after exposure to strong concentration of mustard gas. The hatched area represents the lesions


FIG. 147.- Mustard-gas lesions of back at one week


FIG. 148.- Skin lesions of mustard gassing one week after exposure.



1. Dichlorethylsulphide (mustard-gas) is an escharotic, specific in its action upon the epidermis and tissues of corium, particularly upon the endothelium of the vessels. The lesion may vary according to the concentration and period of exposure, presenting either a hyperemia, vesicle, or eschar. There is no essential difference between the lesions produced by direct application of liquid mustard gas and those caused by exposures to its vapors
2. The lesion is a chemical burn unlike that produced by heat, electricity, or the ordinary corrosives such as sulphuric, nitric, and hydrochloric acids, or strong alkalies. Of all these agents, the effects are most closely allied to those of hydrochloric acid, but are much greater in intensity. It differs from a heat burn in the absence of thrombosis, in the greater degree of fluid exudation, in the greater moistness of the affected area and in the fact that the necrosis as shown by the loss of nuclei requires hours, or even days, for its complete development. The coagulated, shrunken and cooked appearance of the tissues in heat burns is not apparent in the tissues of mustard-gas burns.
3. The vessels in the affected area are severely damaged and collapsed and there is a local anemia in the earlier stages, with a marked fluid exudation and leucocyte migration. The process is nonhemorrhagic and nonthrombosing.
4. In clinical cases of general gassing with mustard gas, the skin is at first pale, then becomes erythematous within a few hours. With the development of the erythema there is usually intense shock with extreme nausea and vomiting. Vesication begins usually on the second day and progresses for several days, eschar formation appearing on the sixth or seventh day. The temperature usually does not rise until the development of escharization and secondary infection.
5. In man the necrosis of the epidermis is usually evident, microscopically, in two hours through the hydropic change in the epithelium and early vesicle formation. There is no deep edema. It is confined to the epidermis and to the papillary layer in the early stages.
6. In animals the intense and deep edema is most striking and altogether different from that seen in man. Vesicle formation was not noted in animals. The fluid from vesicles has no irritating property and can produce no secondary lesions.
7. The deep penetration of the smallest quantities applied to the surface is a most striking feature. There is an undoubted entrance through the hair follicles, sebaceous and sweat glands.
8. The slowly progressive development of the necrosis is a specific characteristic, the height of the necrosis being reached five to ten days after application. In this respect, also, there is a similarity to the X-ray burn. This may he explained, in part, by contraction and death of the vessels with resulting anemia in the affected area.
9. The painlessness of the lesion is also a marked characteristic. This may be explained by the edema and degeneration of the nerve endings in the affected portion.
10. Areas of the skin damaged by mustard gas may not show vesication or exfoliation unless they are subjected to pressure or rubbing. Slight trauma upon such damaged areas may produce vesicles or blebs some time after exposure. This phenomenon is analogous to Nikolsky's sign in pemphigus.


11. In none of the animals and in none of the clinical cases was there any conjunctivitis or irritation of the respiratory tract produced by local cutaneous applications. There is no evidence of metastasis from the local lesion, as claimed by both Meyer1 and Haldane.8 The conjunctival and respiratory lesions are due alone to the direct action of mustard gas, and when animals are protected from the vapor no lesions in these organs will result, no matter how severe the skin burn.
12. Contrary to the statements of certain English and French observers, the admixture of water does not increase the escharotic action, but if the oil is immediately washed away the lesion is greatly reduced in intensity. Washing within two minutes with tincture of green soap may entirely prevent the lesion or result in only a slight hyperemia.
13. The lesions observed in the axillae, between the fingers and toes, around the genitals, and between the thighs of men gassed in action are probably due to the greater moisture of these parts from perspiration and the resulting resolution of the gas, as well as to the more abundant gland supply of these parts.
14. The slow healing is probably chiefly due to the vessel injury and the relatively slight leucocytic demarcating infiltration. In this respect the lesion is strikinglv like an X-ray burn of the skin. Regeneration of the epidermis after complete necrosis takes place from the epithelium of the sebaceous and sweat glands. Marked pigmentation may persist for long periods. This pigmentation is due to the formation of many pigmented chromatophores in the upper portion of the corium. These findings present positive evidence for the production of melanin by connective tissue and endothelial cells. Chronic eczema, with desquanmation and intense itching, is a frequent sequel to repeated slight mustard-gas burns.


In spite of the fact that Victor Meyer, the discoverer of mustard gas, noted the conjunctival lesions in man and in experimental animals, the literature contains no reports of further investigations in this direction. The material presented below must therefore be regarded as opening up a new field in ophthalmic pathology.



Application of dichlorethylsulphide to the conjunctiva was made in two ways: (1) By direct application of liquid dichlorethylsulphide; (2) bv exposure to dichlorethylsulphide vapor. In order to secure results comparable to those obtained by direct application to the skin, as previously reported. an extensive series of animals was made use of, in which the liquid dichlorethylsulplide was applied directly to the center of the cornea bv means of a fine pipette in uniform minute droplets determined to be about 0.0004 c.c. in size. The animal's lids were at once released and by its blinking the mustard-gas liquid was spread in the conjunctival sac. By using a uniform height of column of dichlorethylsulphide in the pipette the effort to maintain a standard size of droplet was quite successful. However, it is realized that slight variations in the amount applied were unavoidable. In another series, the animals were exposed to


dichlorethylsulphide vapor in varying concentrations and for varying times in a respiration chamber especially devised for the purpose. (See fig. 190.) Both pure and crude forms of mustard gas were employed.


Although the objective symptomatology and gross pathology of the lesions produced by direct application of the liquid mustard gas and by exposure to the vapor were essentially similar in the two cases, there were slight differences depending entirely upon method of administration which makes it desirable to consider them separately.


Rabbit.- At once, upon application of a standard droplet (0.0004 c.c.) of dichlorethylsulphide to the center of the cornea, the rabbit blinked repeatedly but showed no other evidence of irritation for one or two minutes, when there was usually a period of rapid blinking and rubbing of the eye with the fore paws. At the same time there was a definite increase in lacrymation, but not to the point of epiphora. When at rest the eye might be held partly closed. Thereafter the rabbit showed but slight signs of irritation throughout the earlier stages, except that at long intervals the eye and nose were rubbed with the paws; the affected eye was opened as widely as the other, and lacrymation was moderately increased. Fifteen minutes after application there was beginning congestion of the vessels of the superior palpebral conjunctiva. At the same time there might be a slight decrease in the normal luster of the eye. The fluorescein test (see p. 567) was positive as early as 10 minutes. Thirty minutes after application there was increased lacrymation, and the congestion of the palpebral conjunctival vessels could be so marked that there was a pinkish reflex through the upper lid. These changes gradually increased dur- ing the next half hour.

At one hour the first evidences of edema appeared in the form of thicken- ings of the conjunctiva about 1 mm. in diameter. These occurred most frequently in the superior palpebral conjunctiva near the fornix and at the upper border of the nictitating membrane. These areas were so small and so translucent that they could scarcely be seen except with the aid of the loupe and electric illumination. From the second to the sixth hour there was a progressive increase in the conjunctival edema, the localized edematous thickenings extending and coalescing until the ring of swollen edematous conjunctiva so encroached upon the palpebral fissure that the sclera could not be made visible even by forced separation of the lids. The conjunctiva of the nictitating membrane shared in a marked degree in the formation of this edematous ring. The bulbar conjunctiva also showed a well-marked edema, a definite chemosis being noted about the fifth hour. Between the fifth and sixth hours after application the cornea began to show a faint clouding or haziness, especially in that portion of its lower half exposed in the palpebral fissure. The cornea was less lustrous than normal and when the surface was viewed under oblique illumination it was found to be somewhat roughened and irregular, indicating areas of destruction of the corneal epithelium. There was an increasing formation of seropurulent exudate. Photophobia gradually became more marked, the animal resisting all attempts to separate the lids. Lacrymation became excessive to the point of epiphora. The pupil reacted promptly to light.


From the sixth to the twelfth hour after application the edema of both palpebral and bulbar conjunctiva continued to increase. There was now a very lefinite chemosis. At eight hours the clouding of the cornea had reached such a degree that it might be properly described as porcelain-like because of its bluish-white opalescence. At this stage, likewise, there was noted an irregularly

FIG. 149.-Twenty-four hours after direct application of standard droplet of dichlorethylsulphide to cornea of right eye. Marked edema of lids and surrounding subcutaneous tissue rounded area devoid of epithelium located in the mesial superior portion of the cornea. At ten hours the lids were found sealed by the accumulated seropurulent exudate along their margins. This adhesion of the lids, first noted at this stage, was a constant feature until the gradual subsidence of the acute inflammatory process about the third week. Very minute subconjunctival hemorrhages were likewise first noted about the tenth hour.


From the twelfth to the twenty-fourth hour the picture remained fairly constant. There was a gradual increase in the edema which spread markedly into the periorbital tissues. (Figs. 149, 150, 151.) The stiffened lids stood out widely from the eyeball, and along the lid margins there was desquamation of the epithelium and other evidences of necrosis. The corneal opacity was still more marked, and in some instances there was a purulent fluid int the anterior chamber. This hypopyon was never in great degree in this stage.

During the second day the edema remained so marked that the eye could not be thoroughly examined except by excision. Eyes thus examined at 2-hour intervals showed a econtinuous progression in the lesion. The edema no longer continued to increase and in the 36-hour specimens a definite decrease was noted. In spite of this decrease the lids became even more indurated, indicating an increased cellular infiltration, and stood out far from the eyeball. The edges of the lids remained sealed unless forcibly separated, and in the pocket thus

FIG. 150.- Twenty-four hours after direct application of standard droplet of dichlorethylsuphide to cornea. Marked edema of lids, flecks of purulent exudate. The marked congestion of the conjunctival vessels is best seen in the building edematous mass of the superior palpebral conjunctiva.

FIG. 151.- Twenty-four hours after direct application of standard droplet of crude mustard-gas liquid to the cornea. Extreme edema of conjunctiva, especially marked in upper lid and nictitating membrane. Seropulent exudate.

formed thick purulent exudate accumulated. From the 44-hour stage on, this accumulation of thick purulent exudate became a very important feature. The denuded area upon the cornea increased in size and in practically all cases assumed a somewhat circular or elliptical form, occupying the greater part of the corneal surface but always approaching more nearly the inner quadrant of the limbus than the outer. The corneal opacity increased and frequently showed a curious distribution, similar to that which we have since learned has been described in human cases by Pissarello. 9 This consisted of a greater degree of opacity in the lower half of the cornea terminating in a more opaque band or zone runningr horizontally through the cornea just below its transverse diameter. This was best seen in a 30-hour stagre. The marginal lid changes became much more marked during that period.

The decrease in the edema continued through the third day, but was noted especially, in the palpebral conjunctiva, while the bulbar conjunctiva, in contrast, exhibited a chemosis which appeared even more marked than hitherto.


The necrosis of the palpebral conjunctiva at the lid margins, and even of a zone on the dermal side of the lids, became more evident, for at that time there appeared numerous shallow ulcerations resulting from separation of the necrotic material.

On the fourth day the diminishing edema was followed by an increased congestion. More numerous and larger subconjunctival hemorrhages were frequently noted. These did not occur, however, in either such size or numbers as to permit the lesion to be characterized as hemorrhagic. The lids became increasingly sensitive to pressure so that the animals gave evidence of pain when an attempt was made to separate them more widely. In the earlier stages the photophobia seemed to be the feature that gave the animal distress. The upper lid, near the inner canthus, began to show a "kinking" or "ruffling" of the margin. This was a constant feature in the later stages. The marked seropurulent exudate continued to persist.

There were no marked changes in the lesions during the fifth day. The seropurulent exudate was unchanged. The "ruffling" of the upper lid was more marked and the lids were still indurated, standing far out from the eyeball.

FIG. 152.- One week after direct application of standard droplet of dichlorethylsulphide to the cornea. Lids still somewhat edematous. They were sealed by a purulent exudate which adheres along the lid margins and to the adjacent hair. A marked purulent rhinitis, referable to involvement of the mucosa through the nasolacrymal duct, is evident

In rabbits that had been repeatedly examined, thereby separating the lids and removing the accumulations of purulent and necrotic material, the exudate was less in amount and more serous on the sixth day. In animals allowed to remain with the eyelids sealed for days at a time, the seropurulent exudate persisted much longer and was of course much more destructive. At this time the necrotic lid hairs and the hairs of the face near the inner canthus began to drop out, resulting in some instances in an extensive depilation.

On the seventh day there were no marked changes from those previously described. The depilation about the inner canthus and over the lids was more complete, and there was a beginning entropion of the upper lid which had the effect of drawing the stiff outstanding lid somewhat toward the eyeball. (Fig. 152.)

During the second week, the lids of eyes which had been frequently examined no longer became sealed between examinations, and the exudate decreased to a small amount of serous fluid carrying a few flakes of pus and necrotic tissue. The upper lid border showed a marked degree of "ruffling" or folding at the lid margin. This was accompanied by an irregularly distributed entropion. The lid margin itself became smooth, rounded, and devoid of lashes. The lower lid showed toward the end of the second week a marked ectropion, with the same smooth, rounded margin found on the upper lid. The periorbital depilation at times was very extensive. The corneal opacity was unchanged. The normal corneal curvature might be distorted by small, irregular, staphylomalike projections occurring most constantly inferiorly toward


the inner canthus. However, these on section were found to be thickenings of the damaged cornea and not areas of bulging. Hypopyon might be present. (Figs. 153, 154, 155.)

During the third and subsequent weeks the changes in the lesions were slowly progressive, demonstrating the sluggish character of the reparative process following injury with mustard gas. The "ruffling" of the upper lid border persisted and gradually became more exaggerated up to the eighth week, which terminated our period of observation. The ectropion of the lower lid also increased in degree, reaching in some cases an almost complete eversion. The margins of the lids were rounded, thickened, smooth, and glossy, having the appearance of scar tissue. (Figs. 156, 157, and 158.) From the fourth week on there was a progressive vascularization of the injured cornea. As the newly formed vessels traversed the limbus the usual differentiation between sclera and cornea became effaced. By the sixth week this process of organization became far advanced and the new-formed vessels might be traced to the central region

FIG. 153.-Two weeks after direct application of standard droplet of dichlorethylsulphide to the cornea. Marked reduction of edema. Much less purulent exudate. Indurated lids exhibit the characteristic "ruffling" and partial entropion of the upper lid in the later stage and the smooth ectropion of the lower lid. The lower half of the cornea shows a marked clouding

FIG. 154.- Two weeks after direct application of standard droplet of dichlorethylsulphide to the cornea. Marked depilation about the eye. Characteristic "ruffling" and entropion of lower lid. The corneal cloudiness and lack of luster are very apparent, likewise the staphyloma in the lower half of the anterior quadrant

of the cornea as is clearly shown by Figure 159. The organization was grossly most evident in that portion of the cornea which showed the primary, porcelain-like cloudiness. When this involved the entire cornea, as in the case figured, there resulted, a complete opacity of the cornea, so that the iris and pupil could no longer be seen and blindness would be nearly, if not quite, complete. (Figs. 159, 160, 161.)

Dog.- Following the direct application of a standard droplet of liquid dichlorethylsulphide to the cornea of the dog, the changes ran closely parallel to those described above for the rabbit, and it seems unnecessary to describe them in detail. The earliest changes were identical. At the period, however, at which the rabbit's eye became sealed by the adhesion of the lid margins, the powerful orbicularis muscle of the dog prevented this occurrence. As a result there was never the empyema of the conjunctival sac that was found in the rabbit, and the exudate seemed to be less in amount. The congestion in the earlier stages was more marked, or at least more evident, than in the rabbit. The edema and the corneal necrosis were similar with the exception that the localization of the corneal opacity in the lower segment was never noted.



Rabbit.-The symptoms and gross pathology produced by exposure to mustard-gas vapor varied with the concentration of the vapor and the length of exposure. In any case, they differed in degree, but not in kind, from those produced by the direct application of the liquid dichlorethylsulphide to the cornea and its immediate spread throughout the conjunctival sac by blinking. An exposure for 15 minutes to a concentration of approximately 1:20,000 was found to give results of about the same degree of severity as those produced by the standard droplet of liquid directly applied. The following notes are abstracted from a protocol of a rabbit subjected to mustard-gas vapor for that time and at that strength.

FIG. 155.- Two weeks after direct application of standard droplet of dichlorethylsulphide to cornea. Specimen obtained by excision of lids and orbital evisceration. Marked depilation especially at the inner canthus. Characteristic "ruffling" and entropion of upper lid. Corneal cloudiness

About five minutes after being placed in the gassing chamber the animal commenced to show signs of irritation, increased blinking, rubbing of the eyes and nose, and a change of position to bring the head as far as possible from the affluent opening by which the impregnated air was being introduced. Immediately after removal from the chamber the animal was quiet, but after a short interval continued to show increasing signs of irritation of the conjunctival and respiratory mucosa. These continued for the next few hours. At 5 hours after gassing a well-marked conjunctivitis was present. There was increased lacrymation, redness of the lid borders, beginning edema, and a marked photophobia. After 16 hours the signs and symptoms were all much more marked. The exudate had become somewhat mucopurulent; congestion of the lid borders and conjunctiva was increased and likewise the edema. The animal kept both


eyes closed because of the photophobia, which seemed much more marked. The edema reached its height about 24 hours after exposure, and a well-marked chemosis was then evident. All skin surfaces where the hair was thin showed a definite erythema. This was especially marked on the lids and convexity of the ears. A bilateral rhinitis, with abundant serous exudate also appeared at this time. The inflammatory changes of the upper respiratory tract were much worse at 36 hours, the wheezing respiration being distinctly audible at a distance. The eyes were sealed, and upon separation of the lids a large amount of thick purulent exudate appeared. The conjunctivie were edematous and congested, and there were a few minute subconjunctival hemorrhages.

The lesions progressed slowly, as described for those produced by direct application. At 72 hours the clouding of the lower half of the cornea had appeared to a marked extent. The lids, having lost their edema to a consider- able degree, remained stiff and stood out far from the eyeball, showing the first signs of the "ruffling" previously described. At 96 hours the lid borders began

FIG. 156.- Three weeks after direct application of standard droplet of dichlorethylsulphide to center of cornea. Same rabbit as Figure 154. Marked ectropion of lower lid. Porcelainlike cloudiness of the cornea most marked in the lower half

FIG. 157.- Three weeks after direct application of standard droplet of dichlorethylsulphide to cornea. Specimen obtained by excision of lids and evisceration of orbit. Anterior segment of globe in profile to show apparent staphyloma of cornea toward inner canthus. The corneal cloudiness is well shown

to show loss of the necrotic epithelium, with the formation of shallow ulcers. The corneal surface became roughened through loss of the epithelium, but no deeper ulceration could be demonstrated.

The lesions produced by this method have not been observed up to the stage of complete healing, as was done with those produced by direct application, but there are no apparent differences in the nature of the sluggish reparative processes in the two cases.


An attempt was made to verify the statement of Victor Meyer that subcutaneous injection of dichlorethylsulphide determines the occurrence of a conjunctivitis, indicating not only a metastasis of the toxic substance through the blood stream, but also a selective affinity or special susceptibility of the conjunctiva. For this purpose we used a series of 4 albino rabbits, 2 hares,


and 2 dogs. The albino rabbits offered the greatest prospect of success, since in them the slightest conjunctival congestion could be readily seen. In the case of the rabbits and hares 1 minim of dichlorethylsulphide was given as a subcutaneous injection. In no instance was any trace of conjunctivitis or any other ocular change produced. The site of injection developed a waferlike induration which dried down to a deep-seated eschar. In every case there developed a foul diarrhea, to which most of the animals succumbed on about the sixth day. One large dog received 4 minims subcutaneously. There was no conjunctivitis, but a severe diarrhea developed, and death occurred on the fourth day. Another dog was fed 4 minims in meat. Severe burns of the

FIG. 158.- Rabbits' eye at 4 weeks after direct application of standard droplet of dichlorethylsulphide to cornea

mouth and upper alimentary tract were produced, but there was no conjunctivitis. These lesions in the gastrointestinal tract will be considered more fully later in this chapter. Attempts to produce conjunctivitis or other eye lesions byv intravenous injection also failed; although this fact has little weight since death occurred too soon to permit the development of marked changes.


In the ocular lesions produced by the standard droplet and by exposure to a vapor concentration of 1: 20,000 for 15 minutes, with ordinary care given to the eyes so treated, by separation of the lids to allow the purulent exudaite to drain out, no evidence of panophthalmitis was observed, and by the end of


the fourth week healing was usually complete. But in animals exposed as above, with eyes untreated, lids allowed to remain sealed and exudate to accumulate, within three weeks there frequently developed a purulent panophthalmitis which by the sixth week usually destroyed the entire eyeball with complete suppuration of all orbital tissues. When a larger dosage was given, and the eyes untreated, a panophthalmitis might develop more rapidly. (Fig. 162.)

One series of four rabbits was treated with a heavy application of liquid dichlorethylsulphide to the center of the cornea. The amount applied was

FIG. 159.- Six weeks after direct application of standard droplet of dichlorethylsulphide to cornea. Combined ectropion and entropion of upper lid with resulting "ruffling" of lid margin. Ectropion of lower lid. Organization of the necrotic cornea with extensive arborizations of newly formed blood vessels, best seen in the upper half of the cornea. Even in the photographs these can be traced from the sclera across the superior arc of the limbus to the central portion of the cornea. The same eye is shown in Figures 160 and 161

about twice that of the standard application. In three rabbits the eyes were left untouched for five weeks, in the fourth for six weeks. After the first sealing of the lids there was no separation of them to allow the purulent exudate to escape. As a result an empyema of the conjunctival sac developed, with subsequent destruction of the eye from a suppurative panophthahnitis. These eyes all showed the same gross picture: Perforation of the cornea, necrosis and abscess formation, loss of the contents of the globe, and suppuration of all orbital contents.


In our cases the occurrence of panophthalmitis was due entirely to secondary infection, which occurred ultimately after several weeks in the untreated cases after slight gassing, but much more rapidly after heavier dosage. The pathologic changes produced by dichlorethylsulphide were covered up by those due to secondary infection.


The standard method of applying an alkaline aqueous solution of fluorescein for the determination of corneal ulceration was carried out on a series of eyes at different time intervals following exposures to dichlorethylsulphide. A 2 percent alkaline watery solution was used. Ten minutes after the direct application of the standard droplet to the center of the cornea several minute

FIG. 160.- Enlargement of Figure 159, to show details of changes

pin-point areas at the vertex of the cornea retained the fluorescent green coloration. These minute areas increased in number so that within 15 minutes after the exposure the vertex of the cornea appeared peppered with these spots in a small circular area. These pin-point areas increased in size and became confluent in about 30 minutes after exposure. By 1 hour they were fairly uniformly confluent. From this time on the fluorescein staining showed a gradually increasing depth of penetration and a spreading of the circumference of the staining area over the cornea, especially toward the inner canthus. The depth of the intravital staining reaction was greatest at the vertex of the cornea and decreased laterally. At the end of 48 hours a narrow cres- centic area toward the outer canthus remained unstained, showing that the necrosis in this area occurred more slowly than elsewhere. After exposure to


the vapor of dichlorethylsulphide the fluorescein staining reactions were the same as after direct application. This intravital fluorescein staining was parallel in intensity to the necrosis of the corneal epithelium as shown by the microscopic study of the same stages. The earliest intravital staining occurred at the same time that the pycnosis of the corneal epithelium became evident, before any ulceration or abrasion took place. The use of this test was therefore of the greatest importance clinically in determining the severity and progress of the lesions and the effects of therapy.


1. The standard drop of 0.0004 c.c. when applied directly to the cornea of animals was found to produce results practically identical with those produced by an exposure of 15 minutes to a vapor concentration of 1: 20,000. The

FIG. 161.- Seven weeks after direct application of standard droplet ofdichlorethylsulphide to cornea. Same eye as in Figure 159. Here given in profile to show marked irregularity of covered surface. Note especially the prominent apparent staphyloma in the sclera toward the inner canthus

criticism, therefore, that the use of the standard drop produces changes not comparable to those observed in exposure to dichlorethylsulphide vapor does not hold. Further, the use of the standard drop is a much more convenient and accurate way, as well as a safer method, of handling this material for experimental purposes. Moreover, the use of the direct-application method avoids the complication of respiratory or general cutaneous involvement following the exposure of the animal in a gas chamber.

2. Dichlorethylsulphide produces after one or two minutes exposure to drop or highly concentrated vapor a definite irritation of the conjunctiva with increase of lacrymation. Usually within 30 minutes there is a well-marked hyperemia, followed in an hour by the development of edema, which progresses rapidlly up to the twelfth hour when there is usually a well-marked chemosis. Minute subconjunctival hemorrhages may develop as early as the tenth hour


3. In animals the edema develops first and most markedly in the palpebral conjunctiva, following the direct application, while in the exposure to mustard gas vapor it frequently develops first in the bulbar conjunctiva, this being practically the only difference observed in the effects of the two methods. By the end of the third day the edema begins to subside slightly, but persists to some degree for several weeks. In man the edema is less marked, more irregular, while the hyperemia is more marked, and minute vesicles may be found on the conjunctival surface.
4. The necrosis of the cornea is shown by a definite cloudiness developing in 5 to 6 hours, which usually at 8 hours has reached such a degree that the cornea takes on a porcelainlike appearance in the form of a very characteristic

FIG. 162.- Five weeks after direct application of dichlorethylsuiphide to cornea. Dosage about twice the size of the standard droplet. Lids not separated. Eye untreated. Resulting panophthalmitis with collapse of eyeball

bluish-white opalescence. In the mildest cases the lesion does not progress beyond a slight cloudiness. Intravital staining with an alkaline aqueous solution of fluorescein shows very early the development of the corneal necrosis, even before ulceration has occurred. A striking phenomenon is the frequent occurrence of a more opaque band or line running horizontally across the cornea just below its transverse diameter.
5. A seropurulent exudate is well developed by the fifth to the sixth hour and increases until the eyelids are usually sealed by the accumulated exudate by the tenth hour. This adhesion of the lids remains a constant feature until the gradual subsidence of the inflammatory process about the third week. The edges of the lids remain sealed unless forcibly separated. If the eyes are frequently


examined with consequent separation of the lids and removal of the accumulated exudate, the stage of purulent exudation is perceptibly shortened.
6. With the subsidence of the edema a characteristic kinking or "ruffling" of the upper lid, a combined entropion and ectropion, appears, usually by the fifth or sixth day. At the same time the lower lid begins to exhibit a smooth ectropion.
7. Depilation of the lid hairs and of the face hairs, eventually about the entire orbit, takes place.
8. During the second week changes in the corneal curvature are constantly noted, some of these so marked as to appear staphylomalike.
9. Hypopyon sometimes occurs. Clouding of the contents of the anterior chamber occurs quite regularly in the later stages.
10. From the third week on the lesions slowly progress in a manner characteristic of the mustard gas lesion of the skin toward resolution and repair. The "ruffling" of the upper lid increases up to the eighth week, while the ectropion of the lower lid frequently reaches the point of complete eversion. Progressive vascularization of the cornea takes place, the vessels usually reaching the center of the vertex by the end of the sixth week. Corneal cicatrization, marked impairment of vision, and thickening of the eyelids and nictitating membrane are the ultimate sequelae. Even in the lighter cases in man, the edema and hyperemia of the conjunctivae tend to run a chronic course with resulting disturbances and reduction of vision. An increased susceptibility to the vapor may develop. This susceptibility, however, is not a specific one.
11. Secondary infection. In animals exposed to the standard drop or vapor concentrations of 1: 20,000, followed by treatment of the eyes, purulent panophthalmitis has not been observed to develop. In the case, however, of untreated eyes, and following heavier dosage, suppurative panophthalmitis does develop with complete destruction of the eyeball.



Every care was taken to prevent the formation of artefacts, either post-mortem or technical. The eye was removed at once from the freshly killed animal by a wide incision encircling the lids and eviscerating the orbit. It was at once placed in the fixing fluids, a neutral formol being used for the greater part. The specimen was left in the fixing fluid about 48 hours before being sectioned, at which time uniform blocks of tissue were selected. These were so oriented as to give vertical sections through the entire eyeball, a section through the inner canthus and one through the lacrymal gland from each eye. The lens was removed from the eyeball after thorough hardening, in order to facilitate the cutting of thin sections. These tissues were embedded in paraffin. the sections were transferred to the celloidin sheet and stained with hemalum and eosin and bv other ordinarv staining methods.


RABBIT.- One-quarter hour: The changes observed were of slight degree, consisting of contraction and pycnosis of the corneal epithelium, which was more marked at the vertex and less marked over the linibus. Likewise the cells of the substantia propria showed a slightly greater degree of pycnosis and contraction than in control preparation. The bulbar and palpebral conjunctivae showed a similar pycrosis and contraction. In the fornix there was


a more decided vacuolation of the conjunctival epithelium. The epithelium of the cutaneous surface of the eyelid showed also slight pycnosis and contraction. In the region of the inferior fornix alone did the subconjunctival connective tissue show a well-marked edema. In the upper fornix the edema was less marked. The lacrymal gland showed a marked distention of its gland spaces which were filled with secretion. Harder's gland and the tarsal glands showed no changes, likewise the sinus hairs. The blood vessels of all parts were anemic rather than congested. There was no hemorrhage and no thrombosis.

One-half hour: Changes identical with those above, except a little more marked in degree. Pycnosis and vacuolation were a little more pronounced. The edema was much greater, particularly around the lacrymal gland. The distention of the lacrymal gland was greater, and its cells showed a marked vacuolation. Greater congestion in all parts.

One hour
: The only changes were an increase in the edema and congestion.

One and one-half hours: The pycnosis of the corneal epithelium was as above, but the substantia propria showed a distinctly more marked pycnosis of the corneal cells extending entirely through the cornea at the vertex. A large portion of the endothelium of the anterior chamber showed marked pycnosis and contraction. The conjunctival epithelium, both bulbar and palpebral, showed an increased degree of pycnosis, while the subconjunctival

FIG. 163.- Cornea two hours after application of standard droplet of dichlorethylsulphide. First stage of necrosis of corneal epithelium and of the cells of the interstitial substance. Marked pycnosis of the corneal epithelium, the cells of the lowest layer alone being barely distinguishable. The nuclei of the interstitial substance and of the endothelial lining of the anterior chamber are also pycnotic. Section taken at corneal vertex

connective tissue showed a marked edema, extending entirely through the eyelids. It was especially noteworthy that many of the blood vessels appeared partly collapsed, with a relative anemia, while others were moderately congested. The lacrymal gland showed the same active secretion, many of the acini being greatly dilated, and filled with secretion. The vessels of the gland were congested and there was an increased number of wandering cells throughout the gland. The stroma of the gland and the surrounding tissues showed extreme edema. The tarsal glands showed groups of pycnotic alveoli. The epithelium of the membrana nictitans showed less change than that of the bulbar and palpebral conjunctivae.

Two hours: (Fig. 163.) Pycnosis of the corneal epithelium was complete. It was impossible to recognize the individual cells. The substantia. propria stained more diffusely blue and the endothelium of the anterior chamber showed a greater degree of pycnosis and vacuolation. Similar changes occurred in the palpebral and bulbar conjunctivae and in the epidermis of the eyelids. The edema of the eyelids was now extreme, being most marked on the conjunctival surface. (See fig. 160.) There was an increase of wandering cells throughout both upper and lower eyelids, but no well-marked cellular infiltrations. Many of the blood vessels showed a well-marked congestion. The membrana nictitans showed a marked pycnosis of its epitbelium and a marked edema of its stroma, its thickness being increased about tenfold. The lacrymal gland showed a marked distention of all its gland spaces and a marked vacuolation


of its epithelium which appeared distinctly more columnar in shape. The cells of Harder's gland stained very deeply and were contracted and pyenotic and had lost their granular appearance to a marked degree. The Meibomian glands showed a similar marked pycnosis involving all of the alveoli. The adipose tissue around the eye and about the tarsal glands showed a marked edema and gave a marked mucin staining reaction.

Three hours: The cornea was as before. The epithelium of the conjunctiva, both bulbar and palpebral, showed a greater degree of necrosis. The edema was much more marked beneath both the bulbar and palpebral conjunctiva, amounting almost to a liquefaction of the tissue. There was a well-marked polynuclear infiltration beneath the conjunctival epithelium extending throughout the entire lid. The epidermal surface of the eyelids showed now a well-marked pycnosis and caryorrhexis of the nuclei of the dermis, particularly of the papillary layer, with a well-marked polynuclear infiltration, the changes being identical with those described in the chapter on the skin lesions, with the exception that the cellular infiltration was somewhat more prominent than was found in the skin. The vessels showed a more marked congestion than in any of the previous stages. The lacrymal gland showed the same

FIG. 164.- Section of cornea at vertex five hours after direct application of standard droplet of dichlorethylsulphide. Desquamation of necrotic epithelium in center of vertex. Pyenosis of remaining epithelium and of cells of the interstitial substance. Complete necrosis of endothelial cells of anterior chamber

distention of the alveoli, but the cells were less colttnnar and the nuclei showed some pyenosis and caryorrhexis. The gland of Harder and the tarsal glands showed the same pycnosis as in the preceding stage. The changes in the epidermal surface of the eyelid had advanced in necrosis and cellular infiltration.

Four hours: Changes in the cornea appeared as in the preceding. Bulbar and palpebral conjunctivtae, membrana nictitans, and epidermal surface of eyelids showed increasing edema and cellular infiltration. The lacrymal gland showed extreme distention and the cells of many acini were flattened and vacttolated as in a state of exhaustion atrophy. Harder's gland presented a well-marked vacuolation of its epithelium, many of the cells showing large clear droplets, edema. Likewise the tarsal glands showed edema. In the palpebral and tulbar conjunctivae the cellular infiltration was increased, and the lymph follicles in the palpebral conjunctiva contained great ntmbers of wandering cells. Vessels were greatly congested.

Five hours: (Fig. 164.) The corneal epithelium was in part desquamated over the region of the vertex, and about this the epitheliumn showed marked dissociation and vacuolation. The substantia propria stained lighter in color, many of its nuclei being only shadows.


At the sclerocorneal junction there was a well-marked infiltration of polynuclears, most marked just beneath the pycnotic epithelium. The upper and lower lids showed extreme edema (see fig. 160); and there was a marked infiltration of polynuclears throughout, which was more marked beneath the palpebral conjunctiva near the palpebral border. In this region the conjunctival epithelium was desquamating and an ulcerating surface was developing. The adherent dead epithelium was infiltrated with polynuclears. The blood vessels showed congestion. There was no thrombosis. On the epidermal border of the lids there was beginning desquamation of the dead epidermis. The sebaceous glands near the margin of the lid showed sebum retention. Likewise on the lower lid near the palpebral margin, there was noted the first development of ulcer or slough, and the membrana nictitans showed loss of its epithelium in areas. Lacrymal, Harder's, and tarsal glands, as in the preceding.

Six hours: Changes identical with above.

Eight hours: The only change noted from above was an increase in the extent of the ulcer on the conjunctival surface near the palpebral margin in both upper and lower lids, more marked in the upper. (Fig. 165.)

FIG. 165.- Section of palpebral conjunctiva eight hours after application of standard droplet of dichlorethylsulphide. Complete necrosis of epithelium. Marked congestion. Minute hemorrhages. Polynuclear infiltration

Ten hours: Cornea showed gradual loss of chromatin. Stained more palely. On the eyelids the necrosis of the epidermis and eschar fortmation bad progressed while almost the entire conjunctival epithelium was completely disintegrated, forming a granular layer infiltrated with polynuclears. The extidate was most marked near the palpebral border where there was a distinct ulcer covered with exudate. Intense edema and congestion as in the preceding. Small pin-point hemorrhages ocurred in the subconjunctival connective tissues.

Twelve hours: At the corneal vertex the cornea was dead throughout, having completely lost its nuclei. The corneal epithelium was desquamated over the vertex, and only here and there were there faint otltlines of the nuclei of the substantia propria. The pycnotic and vacuolated endothelium of the anterior chamber was still intact. The escharization of the epidermal surface of the lids was complete, and desquamiation of both bulbar and palpebral conjunctival epithelium was nearly complete. Congestion was very marked and there were numerous minute hemorrhages by diapedesis both beneath the


epidermal surface and in the subconjunctival connective tissue. The edema was even more marked than in the preceding stages, involving the whole eyelid, and the lymphatics were enormously distended. The lacrymal gland was as before, but Harder's and the tarsal glands showed marked edema and liquefaction necrosis. There were masses of exudate, rich in polynuclears, adherent to the conjunctiva at the lid margins, the result of drying and concentration of the secretion, but near this, minute collections of polynuclears beneath the conjunctiva suggested beginning infection. The fluid of the edema at this stage assumed the same hyaline, finely granular, deeply pink staining appearance as was found in the subcutaneous tissues.

Fourteen hours: Changes as above, but slightly advanced. Intense congestion. Polynuclear infiltration of the muscle. Numerous hemorrhages about the greatly distended vessels. Secondary thrombosis beginning. More marked polynuclear infiltration into the subconjunctival connective tissue. Heavy granular precipitate in the distended alveoli of the lacrymal gland. These contained, likewise, great numbers of albuminous spherules, staining deep violet or pink, derived from the disintegration of the cells.

Fifteen hours: Changes as above with more marked polynuclear infiltration at the sclerocorneal junction, in the ocular muscles and in the orbital tissues. Beginning infection of the conjunctival ulcers near the palpebral border. Diffuse inflammation of the entire peribulbar tissues. It was a notable fact that the sebaceous glands on the epidermal surface of the lids showed penetration with the dichlorethylsulphide and necrosis, while the large ones at the palpebral margin seemed to escape.

Sixteen hours: Complete loss of the necrotic epithelium over the corneal vertex. Increasing polynuclear infiltration at the sclerocorneal junction. Edema of the ciliary body and ciliary ring. Increasing polynuclear infiltration Eand greater collection of pus at the palpebral borders. Eighteen hours: Nearly complete necrosis of the cornea, except toward the periphery of the limbus. On the lids escharization was extending, the necrosis reaching into the subconjunctival tissue and into the subepidermal tissues on the cutaneous side. It was notable, however, that the subcutaneous necrosis was nearly through the depth of the lower border of the sebaceous glands. The epithelium of the entire margin of the lids was now completely necrosed and desquamated. Over the entire surface of the conjunctiva, including the membrana nictitans, the epithelium was necrotic and desquamating, but the necrosis extended but very little into the subepithelial connective tissues. For the first time changes were observed in the erectile sinus hairs of the upper lid, the epithelium of the hair follicle showing some pycnosis and contraction.

Twenty hours: Complete necrosis of the cornea, in the central portion extending to the endothelial lining of the anterior chamber which was partly necrotic and desquamating. Conjunctival, lid, and gland changes as above. The large sebaceous glands at the lid margin showed a more marked polynuclear infiltration than in any of the preceding. Congestion, edema, and minute hemorrhages as in preceding stages.

Twenty-two hours: Complete death of cornea in central portion, with loss of epithelium and endothelium of anterior chamber. Only scattered nuclei through the dead substantia propria. Eyelids showed the same extreme edema, congestion and numerous hemorrhages by diapedesis around the congested vessels as in the preceding, but with an increasing small-celled infiltration. Fibroblastic proliferation first observed in the subconjunctival connective tissue.

Twenty-four hours: Cornea completely necrotic except near the sclerocorneal junction. Its laminae were separated and there was a collection of fluid between the separated laminae. Edema at its height; other changes as above.

Twenty-six hours: Necrosis of the cornea complete with marked separation of the laminae into irregular spaces. Edema as in preceding, but polynuclear infiltration much more marked, extending throughout the entire thickness of the lids. Marked distention of the acini and atrophy of the cells of the lacrymal gland. Numerous capillary hemorrhages by diapedesis throughout the subconjunctival tissues, and the dermis of the lids. (Fig. 166.)

Twenty-eight hours: As in the preceding, no difference.

Thirty hours: (Fig. 167.) Identical with the preceding stages.


Thirty-two hours: Most marked changes in the cornea, particularly at the vertex. Surface roughened, knobby, laminae separated. Changes in lids and glands as in preceding.

Thirty-four hours: The central portion of the cornea showed complete necrosis but at the periphery there was a marked cellular infiltration and proliferation of corneal cells. At the sclerocorneal junction there was an evident fibroblastic proliferation. The edema of the upper palpebral conjunctiva was less marked; cellular infiltration about as before, but more cells of a fibroblastic type. The edema of the lower lid was the most marked yet seen, a large portion of the subconjunctival connective tissue showing complete liquefaction necrosis with marked polynuclear infiltration. (Figs. 168, 169.) The cutaneous border of the lower lid showed marked depilation. The lacrymal gland showed in various areas a very marked hypertrophy of the cells. These were more columnar, with a very deeply staining protoplasm. Other portions showed the light vacuolated cells as described above.

FIG. 166.- Palpebral conjunctiva 26 hours after application of standard droplet of dichlorethylsulphide, showing advancing necrosis, more marked infiltration, congestion, minute hemorrhages and edema

Thirty-six hours: Changes as in the preceding, except for the membrana nictitans which showed marked congestion of its vessels and multiple capillary hemorrhages. The ocular muscles showed marked edema.

Thirty-eight hours: The most marked necrosis of the cornea, the complete necrosis extending almost to the sclerocorneal junction. Complete necrosis of the endothelial lining of the anterior chamber with increased number of leucocytes in the fluid of the anterior chamber. On the lids there was beginning regeneration of the epidermis and hair follicles, but no evidences of regeneration of the conjunctiva. Almost complete depilation. Edema persistent in the subconjunctival. tissues, and at the palpebral margins there was a marked collection of pus.

Forty hours: Necrosis of cornea less marked than in preceding individual; endothelial lining preserved. Appearances indicated a somewhat earlier stage, probably due to less intense action.

Forty-two hours
: (Fig. 170.) Marked necrosis of the central portion of the cornea. The membrane of Descemnet and the endothelium of the anterior chamber appeared as a hyaline red line bordering the anterior chamber. Proliferation of corneal cells at border of cornea;


fibroblastic proliferation at sclerocorneal junction. Beginning separation of eschar on epidermal side of lids and also on the conjunctiva. Cutaneous lesion much deeper and more severe than the conjunctival, although the edema was much more marked on the conjunctival side, as was also the leucocytic infiltration.

Forty-four hours: Changes the same as at 42 hours, but of less intensity. Probably due to dilution of dose on the cornea. The skin surface of the lids showed marked escharization.

Forty-six hours: No essential differences.

Forty-eight hours: Corneal changes same as in the preceding. More marked escharization of the lids and conjunctival surfaces. Beginning separation of the eschar on epidermal portion of lids. Almost complete depilation.

Fifty hours: Very marked separation of corneal lamellae and formation of irregular spaces. Partial separation of the slough on the custaneous surface of the eyelids. Well-marked regeneration of the hair follicles. Old hairs nearly entirely shed. The membrana nictitans was markedly hemorrhagic. The large sinus hairs showed marked necrosis of the epithelium of the hair follicle. Very marked fibroblastic proliferation in upper lid.

Fifty-four hours: Changes as in preceding, except for more marked escharization of the epidermal surface of lids, and more marked fibroblastic proliferation of the subcutaneous tissue. Marked congestion and numerous hemorrhages, particularly on the epidermal side of the lids.

FIG. 167.- Cornea 30 hours after exposure to vapor of dIichlorethylsulphide. Complete necrosis of cornea

Sixty hours: Shrinking and drying of the cornea at its vertex. Fibroblastic proliferation on the dermal side of the lids very marked. Epithelial regeneration well advanced. Beginning regeneration of conjunctival epithelium, but the greater part of the conjunctiva remained dented. The lower lid showed many hemorrhages throughout its substance; the hemorrhages were more marked than in any other case. The eschar on the dermal side completely separated with the regeneration of the epidermis and hair follicles. Very few of the old hairs were left. Edema was still persistent on the conjunctival side, but less marked.

Sixty-five hours: Cornea showed regeneration of the corneal cells throughout its entire extent, but most marked at the periphery. There was marked regeneration of the endothelium of the anterior chamber with plasmodial masses of epithelium on the surface. Anterior chamber contained large numbers of leucocytes. Lids showed marked edema, extreme congestion, and multiple hemorrhages. Eschar on epidermal side was very deep and not completely separated. Changes much more severe than in the preceding instance with less regeneration and repair.

Seventy-two hours: Edema of lids still very marked. Deep escharization on epidermal side with eschar separating. Very little regeneration and repair. In the lower lid the edema was very marked, with liquefaction necrosis of the subconjunctival connective tissues.

Eighty-four hours: (Figs. 171, 172.) Complete necrosis of cornea, with no evidence of repair except at the limbus. Very marked congestion and numerous hemorrhages in the lids. Nearly complete depilation with regeneration of epidermis and hair follicles. Membrana nictitans markedly hemorrhagic and showed some fibroblastic proliferation.


Ninety-six hours: Cornea completely necrosed in the vertex. Proliferation of corneal cells toward the limbus. Fibroblastic proliferation near sclerocorneal junction. Very deep escharization of epidermal side of lids with separation of eschar. (Figs. 173, 174.) Regeneration of conjunctival epithelium in various portions of both upper and lower lids. Edema of lids much less marked, but lids were thickened by the marked fibroblastic proliferation.

One hundred ten hours: Cornea completely necrosed in central portion. At limbus it showed polynuclear infiltration and fibroblastic proliferation. Anterior chamber contained numerous polynuclears, pus. Iris and ciliary body showed cellular infiltration. Conjunctiva showed areas of regeneration of epithelium. Edema was less marked. The gland of Harder showed liquefaction necrosis, as did also the tarsal glands. Necrosis of the epidermal side was deep, but the eschar had nearly completely separated and there was regeneration of the epithelium in some areas and of the hair follicles. Edema of lids was much less marked but the lids were thickened by the connective-tissue proliferation. The membrana nictitans showed almost complete regeneration of its epithelium.

FIG. 168.- Palpebral conjunctiva 34 hours after application of standard droplet of dichlorethylsulphide. Section at fornix of upper lid showing the complete loss of the necrotic surface, extreme edema, and polynuclear infiltration.

Five days: There was some regeneration of the corneal epithelium and proliferation of corneal cells and infiltration of leucocytes throughout the entire cornea. The edema of the bulbar conjunctiva was still very marked while that of the palpebral was very much less. The subconjunctival lymph follicles were very hyperplastic. Large dilated lymphatics showed, however, in the subconjunctival connective tissue of the lids. The escharization of the epidermal side was very deep, extending below the hair follicles, and there was no repair. In the lower lid the edema fluid showed marked fibrin formation.

Six days: Regeneration of the corneal epithelium over the limbus and extending in a delicate layer over the vertex. Endothelium of anterior chamber was regenerating. Marked regeneration of the epithelium of the palpebral conjunctiva with overformation of epithelium. Regeneration less marked on the epidermal surface. Lymph follicles hyperplastic. The bulbar conjunctiva still showed marked edema without lunch cellular infiltration.


Seven days: (Fig. 175.) Marked regeneration of the cornea with great thickening, the cornea being more than twice as thick as normal, but the thickness varied in different portions. Marked infiltration of leucocytes and marked proliferation of the cells of the substantia propria. Around the periphery there was a marked formation of new blood vessels extending into the limbus and reaching nearly to the vertex. Marked regeneration of the endothelium of the anterior chamber. Sections cut in some planes of the cornea showed a failure of regeneration of the corneal epithelium and a very marked infiltration of the anterior lamella. The conjunctival epithelium showed marked regeneration while the eschar of the epidermal portion of the lids had not yet separated. The bulbar conjunctiva still showed marked edema of the subconjunctival tissue. The lids were thickened by the marked fibroblastic proliferation. There was intense congestion and many capillary hemorrhages.

Eleven and one-half days: Cornea thickened irregularly from leucocyte infiltration and proliferation of cells of the substantia propria. At the periphery the escharization was quite marked extending onto the bulbar conjunctiva, beneath which there was a marked leucocytic

FIG. 169.- Section from the same region is in Figure 168, but taken deeper down, showing the extreme edema and liquefaction necrosis, below the narrow band of the sphincter orbicularis

infiltration. The conjunctiva showed marked regeneration of its epithelium, infiltration of the subepithelial tissue and hyperplasia of the lymph nodes. On the epidermal surface there was a very marked eschar formation which was partly adherent and partly separated. Beneath it there was advanced regeneration of the surface epithelium and of the hair follicles. The membrana nictitans was much thickened, its epithelium nearly regenerated, and it showed a marked proliferation of new blood vessels with very hypertrophic endothelium.

Twelve and one-half days: Changes identical with the preceding. The hyperplasia of the subconjunctival lymph nodes was very marked.

Two weeks: Cornea very irregular in thickness. Showed less evidence of regeneration and repair than the preceding. New blood vessels extending into it from the periphery. Eyelids showed marked regeneration of epithelium both onl the conjunctival and on the skin side. Large nests of cells proceeding from the hair follicles.

Three weeks: Regeneration of cortical epithelium and endothelium of anterior chamber. Marked proliferation of blood vessels around the periphery of cornea and formation of scar tissue replacing the corneal substantia propria. Great, irregularity in thickness of cornea. Eyelids showed advanced repair and regeneration of the epidermal surfaces and of the conjunctival


surface. Edema still present on the conjunctival side. The regenerating hair follicles and sebaceous glands formed large atypical cell nests. The sinus hairs showed fibroblastic obliteration of the cavernous spaces and atypical proliferations of squamous epithelium suggesting newly formed sebaceous glands.

Four weeks
: (Figs. 176, 177, 178.) The cornea showed throughout regeneration of its epithelium and of the endotheliumi of the anterior chamber. It was very irregularly thickened

FIG. 170.- Cornea 42 hours after application of standard droplet of dichlorethylsulphide. Membrane of Descemet appears as a bright hyaline line staining red with eosin

and had an uneven surface. From the periphery great numbers of newly formed vessels extended up toward the corneal vertex, on one side reaching halfway across the cornea. These vessels were much more numerous on the anterior side of the cornea, and many of them were very large and dilated. On one side near the scleral junction there was a large area of subepithelial edema infiltrated with leucocytes, producing an elevation on the cornea, as described in the gross notes. Other section of the cornea showed it to be heavily

FIG. 171.- Cornea 3 ½ days after application of standard droplet of dichlorethylsulphide. Complete necrosis of corneal tissue; ulceration of surface; beginning infiltration with polynuclear leucocytes and collection of polynuclear leutcocyes along the line of the necrotic endothelium

infiltrated with pus cells between the newly formed blood vessels. Regeneration of conjunctival epithelium was complete, except near the palpebral margin, where there was an ulcer. Edema of the subconjunctival tissues was much less marked than in the preceding. Harder's and the tarsal glands showed marked lymphoid infiltration. Lacrymal gland more than in appearance. Hair follicles showed new formation of hairs. The large sebaceous glands at the palpebral margin were hyperplastic, and the lymph modes of the conjunctiva were also hyperplastic.


FIG. 172.- Section of sclerocorneal junction 3½ days after application of standard droplet of dichlorethylsulphide. Infiltration of leucocytes beginning flbroblastic and angioblastic proliferation

FIG. 173.- Section of upper lid at palpebra margin four days after direct applieation of standard droplet of dichlorethylsulphide Advanced ulceration beginning repair


FIG. 174.- Section of same lid near fornix. Regeneration of the conjunctival epithelium. Disappearance of the edema and advancing cicatrization of the subconjunctival tissues

FIG. 175.- Section of corneal vertex seven days after application of standard droplet of dichlorethylsulphide. Ulcerated surface. Infiltration of necrotic cornea with polynuclears and scattered flbroblasts. Beginning regeneration of endothelium


FIG. 176.- Section of corneal vertex four weeks after application of standard droplet of dichlorethylsulphide, showing the marked irregularities in the cortieal surface; regeneration of corneal epithelium and endothelium of anterior chamber; edema of the interstitial substance with some fibroblastic repair

FIG. 177.- Sclerocorneal junction of same eye as in Figure 176, four weeks after application of standard droplet, showing vascularization and repair proceeding from the sclera


Seven weeks: (Figs. 179, 180, 181, 182.) The cornea was markedly but very irregularly thickened throughout, heavily infiltrated with leucocytes, and showed throughout its extent a marked new formation of blood vessels. Substantia propia was largely replaced by fibrous connective tissue. The membrane of Descemet showed as a clear hyaline line and the endothelium of the anterior chamber was regenerated in a flat layer over this. The regeneration of the epidermal surface of the lids was complete. The sebaceous glands were approaching normal size. The edema has almost entirely disappeared except over the bulbar conjunctiva. Lacrymal gland still hypertrophic. Lymphoid tissue in Harder's gland was hyperplastic, as were also the suhconjunctival lymph nodes. The lids and nictitating membrane were thickened from the new formation of connective tissue.

DOG.- A small series was also carried out upon dogs. The changes were essentially the same in kind, although apparently of greater intensity. A marked hemorrhagic condition of the conjunctiva was noted in the earlier stages.

FIG. 178.- Section of corneal limbus 4 weeks after application of standard droplet of dichlorethylsulphide. Advancing repair into the cornea from the sclerocorneal junction


Rabbit.- Series of animals exposed for varying periods to varying concentrations of mustard gas vapor (1:20,000 to 1:50,000) and killed at intervals of 12 hours to 4 days after exposure in the gassing chamber showed changes of precisely the same kind as the animals treated with the standard droplet of the liquid. It Was found that a 15-minute exposure to a 1:20,000 concentration produced changes identical in degree with those produced by the standardized drop of liquid. It is noteworthy in these experiments with the weaker concentrations of the gas that the cornea first showed evidences of necrosis and then the skin surface of the lids, while the conjunctival epithelium did not undergo necrosis, and the conjunctiva itself presented only the picture of a mild conjunctivitis with an unusual degree of edema, particularly of the bulbar conjuinctiva in the earlier stages, and of the palpebral in the later.


The epidermal changes were in all cases much more severe than the conjunctival, marked necrosis of the former occurring when no necrosis of the conjunctival epithelium was observed. The corneal necrosis was always most severe at the corneal vertex. In all cases exposed to the gas the lesions were most intense in the portion of the conjunctiva and cornea exposed in the pal- pebral fissure, and the lesions therefore were much less diffuse than when direct application was used, thus resembling more nearly the human cases.


Microscopic examination of eyes to which dichlorethylsulphide had been applied post-mortem showed no changes attributable to the action of the mustard gas liquid.

FIG. 179.- Section from the inferior portion of the cornea seven weeks after application of the standard droplet of dichlorethylsulphide. Persistent ulcer; marked polynmclear infiltration of the cornea, and repair. Blood vessels have reached the center of the cornea


Animals so treated died within five hours after the intravenous injection of 1 minim and uniformly in four to five days after the subcutaneous injection of 1 minim. Microscopic examination of the eyes of these animals showed no evidence of conjunctivitis or other lesions.


Four cases showing panophthalmitis, three at five weeks and one at six weeks, showed microscopically a diffuse suppurative process involving all the structures of the eye and of the orbit. The suppurative process began first in the cornea, and in the earlier stages might show as small pin-point abscesses


in the substantia propria, each surrounding a colony of cocci. The process extended through the anterior chamber toward the posterior portion of the eyeball until ultimately all was involved. Similar changes were observed in a dog three weeks after exposure, the earlier infection in this case resulting from the lack of care and the lessened resistance of this especial animal.


In the uninfected cases the changes consisted of congestion and edema, and a mild diffuse inflammation involving particularly the ocular muscles. In the infected cases there was a diffuse suppurative process.

FIG. 180.- Section from inferior portion of corneal limbus, seven weeks after application of standard droplet of dichlorethylsulphide. Partial regeneration of corneal epithelium. Marked polynuclear infiltration and advanced vascularization and repair of the substantia propria. Regeneration of the endothelium of the anterior chamber


1. The microscopic changes produced by the direct application of the standard droplet of the liquid and by exposure for 15 minutes to a vapor con- centration of 1: 20,000 were identical.
2. Changes in the cornea.- The earliest changes noted were in the cornea, consisting of pycnosis and contraction of the epithelium and of the substantia propria, most marked at the corneal vertex and extending to Descemet's membrane. This was followed by a loss of nuclei until by the twelfth hour the corneal vertex showed complete necrosis, the necrosis often extending through the limbus nearly to the scleral junction. Desquamation of the dead corneal epithelium began in about 5 hours. Polynuclear infiltration of the sclerocorneal junction began in 5 to 6 hours. Fibroblastic proliferation was first noted at 34


hours at the selerocorneal junction. Earliest signs of regeneration of the corneal substantia propria were noted at the periphery at 65 hours. New formation of blood vessels into the limbus was well marked as early as 7 days. Slow repair of cornea continued for several weeks, with development of a highly vascularized corneal cicatrix. Marked changes in the corneal thickness occurreti as the result of separation of the lamellve, edema, cellular infiltration, and fibroblastic proliferation. The severity of the corneal lesion was in direct proportion to the concentration of the gas and the period of exposure.
3. Changes in the conjunctiva.- Necrosis and desquamation of a large part of the conjunctival epithelium resulted, but it is noteworthy that this necrosis was much less in degree than that of the cornea or of the skin surfaces

FIG. 181.- Section front the superior half of the corneal vertex seven weeks after application of standard droplet of dichlorethylsulphide, showing the greater degree of cicatrization usually found in this portion

of the eyelids. The primary necrosis rarely extended beneath the basement membrane of the palpebral and bulbar conjunctiva, except at the palpebral margin, where collections of a serofibrinopurulent exudate occurred. Here shallow ulcers were produced. The most striking feature of the conjunctival involvement was the extreme edema of the subconjunctival connective tissues, which was usually most marked in the bulbar conjunctiva near the scleral sulcus and in the palpebral conjunctiva of the upper lid. This was so extreme that liquefaction necrosis in this tissue, with marked leucocyte infiltration usually followed. Petechial hemorrhages were of frequent occurrence in the subconjunctival connective tissue. Regeneration of the conjunctival epithelium occurred readily and healing took place with a permanent thickening of the conjunctiva due to the formation of fibroblastic tissue.


4. Structures of eyeball.- (Figs. 183, 184.) Iritis and iridocyclitis, with exudation into the anterior chamber, occurred without infection in the severest forms of gassing, and were common occurrences at about the third to sixth week in the uncared-for cases as the result of secondary infection even when the gassing was light. In the cared-for cases of average exposure, no changes were observed in iris, ciliary body, choriod, retina and optic nerve, except congestion and edema.
5. Nictitating membrane.- Changes observed in rabbits were necrosis of the epithelium, extreme edema, multiple hemorrhages, congestion, and eventually more or less marked thickening from connective-tissue proliferation.

FIG. 182.- Section of corneal limbus from same eye as Figure 181, showing advanced cicatrization

6. Lacrymal gland.- Increased functional activitv was noted at all times; ultimately, overuse atrophy and subsequent hypertrophy. Increased albumin content was observed in the secretion.
7. Harder's gland.- Parenchymatous degeneration and inflammatory infiltration occurred.
8. Tarsal glands.- Evidences of penetration into the tarsal glands were shown by degeneration and necrosis of certain of the acini. Leucocyte infiltration occurred about and into these glands.
9. Skin surface of eyelids.- The same changes were observed in the cutaneous surface of the eyelids as were described in the chapter on cutaneous lesions. The important influence of the sebaceous glands and hair follicles in permitting the entrance of the gas into the deeper tissues of the dermis was strikingily shown here, the penetration into and resulting escharization of the skin being greater than in the case of the cornea or the conjunctiva.


10. Subconjunctival tissues.-The relatively slight penetration into the subconjunctival tissues might be explained as the result of the protection afforded by the moistness of the surface and lacrymation.
11. Orbital tissues.- In the cared-for cases congestion, edema, and a mild diffuse cellular infiltration, particularly in the orbital muscles, were noted; in infected cases a diffuse suppurative cellulitis occurred.


CASE 1 (fig. 185).- Exposed to strong concentration of mustard-gas vapor for 10 to 12 minutes. Patient wore a gas mask. The eyes began to burn and be painful, lacrymated freely, and were red and inflamed two hours after exposure. Three to four hours after exposure he began to feel as though there were granules beneath the eyelid. He could not

FIG. 183.- Ciliary body from eye of rabbit exposed 12 hours in gassing chamber to a concentration of I:.50,000. Animal died 92 hours later. Marked collection of polynuclear leucocytes in anterior chamber, in the ciliary body, and in the iris of the left eye, which had received no treatment with the dichloramine-T solution

keep his eyes closed because there was much more pain when the lid borders were approximated. He was very sleepy, because of much loss of rest due to night work, and his eyelids felt heavy. He went to bed. Boric acid compresses (cold) were applied. American oil was instilled once; it did no good. The compresses were to be continuous; they relieved the pain, but he could bear them only about one-half hotlr at a time through the night. For 3 to 4 days argyrol was instilled once in 24 hours, and did very little good. The palpebral conjunctiva appeared "blistered" the morning after the exposure. The irritation continued for 3 to 4 days, but gradually decreased in severity. The congestion decreased but was still marked. In fact it was present and quite distinct 31 dlays after exposure. The patient complained of "misty" blurring of vision, which was marked at first and then decreased, but was still present three weeks after exposure. The bright sunlight hurt his eyes, the lids felt heavy. There was no interference with accommodation so far as the patient knew. He was able to read about the tenth day, but could read only a page at a time.

d For the opportunity of examining the clinical cases, and for their histories, we are indebted to Capt. L. L. Roos, M. C.


Examination.- Congested dilated vessels on the bulbar conjunctiva from the border of cornea, over the limbus, to the inner and outer canthus in each eye. The congestion and thickening of the bulbar conjunctiva showed in the palpebral aperture. The palpebral conjunctiva, especially superiorly, showed slight thickening.

CASE 2 (fig. 186).- Exposed 10 to 12 minutes to a strong concentration of mustard gas vapor. He wore a gas mask at the time. Sulphur dichloride and sulphur monochloride had always irritated his eyes. He first noted irritation under the arms about four hours after exposure, and just one hour later-that is, five hours after exposure--there was a stinging irritation "just like salt" in the eyes. One drop of "Silvol" solution was put in each eye. This increased the pain. The eyes lacrymated profusely and the eyeballs were red and the vessels congested. Boric-acid compresses (cold) were applied at intervals all night. At 3 a. m. he found that he could open his eyes only to a slight extent, the aperture was only about one-half inch at the greatest. The lids were edematous, and this prevented the opening of the eyes. The morning after the exposure the eyes were completely closed

FIG. 184.- Iris of same eye as in Figure 183, showing congestion, edema, and polynuclear infiltration. Marked polynuclear exudate in the anterior chamber

and the lashes were "glued together" by the thick purulent exudate. He could not open the eyes wide enough to see anything. At the first-aid room argyrol and sterile American oil were put into the eyes, and he was put to bed and cold or iced boric compresses were put on continuously for 5 days. Argyrol was instilled 3½ days after exposure for about 30 minutes and everything both near and far was found to be blurred. The conjunctiva was congested. For more than 10 days the eves lacrymated profusely, especially the left eye. On the fifth day his vision was clearer but very much impaired. He could not distinguish letters at all on a printed page. He could not read ordinary print for 10 days, and then he could read the print for only 10 minutes at a time until the eyes ached and things became blurred. It was not until the eighteenth day that he could read for an hour at a time. His vision for distance gradually improved, but was still somewhat blurred. After 18 days, at the end of the third week, near vision was such that he could not read much more than 1½ hours at a time and then he must rest 2 to 3 hours.

Examination.-Thickening of the conjunctiva in the palpebral fissure was especially marked in the triangle with apex at the inner canthus and base at the inner limbus. The thickened conjunctiva was yellowish. The vessels were slightly congested. The palpebral conjunctiva was not very much changed, possibly slightly thickened.


CASE 3.- Pvt. Mc. Exposure of 40 minutes in four shifts to strong concentration of mustard gas vapor while wearing a gas mask. The eyes had been congested for two days because of exposure to hydrochloric-acid gas. On entering the infirmary at 6 p. m. he had no eye symptoms, but because of the congestion the eyes were irrigated with sodium bicarbonate solution or boric acid. He was very nauseated and sick, and vomited and retched very much. At the same time he complained of severe sharp pains in the eyes. Boric acid compresses were apllied continuously for three days, and the eyes were washed with the boric acid about every 15 minutes for the entire first night. The eyes were very painful, burned, and felt as though an electric current was passing across and through the anterior

FIG. 185- Congestion of the conjunctival vessels persisting to a marked degree, four weeks after exposure to dichlorethylsulphide vapor

part of the eye. The eyes were swollen and a purulent exudate glued the eyelashes down. On the third day, while his eyes were being opened and argyrol instilled, he had a flash of vision for a second, just long enough to see his nurse. On the fourth day he recognized individuals. He had been able to open his eyes about the fourth night, but he could see only a dull light. On the fifth day he opened his eves slightly and everything was blurred; near and distant vision were equally impaired. He was not able to read for about 14 days, and then only for about 5 minutes at a time, gradually becoming able to read more and more, but at 6 weeks he could read only about 1 to 2 hours at a time. When in the sunlight, however, there was a feeling of "wideness" (of the palpebral opening), some blur, and some discomfort, and slight photophobia.

FIG. 186.- Persistent congestion four weeks after exposure to dichlorethylsulphide vapor. Acute symptoms were very severe and the patient still complained of dimness of vision when he left the hospital after five weeks

Examination.- In the morning there was slight purulent exudation, which glued the eyelashes together; there was a small drop of purulenit exudate in the inner canthus. There was a slight blepharitis marginalis. The ocular conjunctivae showed thickened, slightly yellowish, elevated areas, pingueculae, on either side of the corneae. The vessels were large, slightly dilated and congested. The palpebral conjunctiva showed nothing of any great interest.

CASE 4 (fig. 187).- Exposure of 30 minutes to strong concentration of the vapor of dichlorethylsuliphide while wearing the gas mask. He was first exposed at about 3.30 p. m. and noticed his first symptoms at about 5 p. m. These were irritation, burning, and lacrymation of the eyes and a "faint feeling' in the pit of the stomach. Then on leaving the


infirmary and getting into "'the air" the eyes felt better and he was not troubled until he returned to the infirmary. Here, after taking the routine "shower, kerosene rub, and shower" treatment and retiring, his eves felt very painful, with the sensation of fine sand under the lids,and there was profuse lacrymation. A thick serous exudate made the eyelashes mat together and seal the eye. His vision was blurred, probably due to the increased lacrymatory secretion. The eyes were red and congested, according to his roommate. His eyes were irrigated with boric acid; American oil was instilled, and cold boric-acid com- presses were applied continuously for five days. Argyrol was used from the second day until about the fifth day. The eyes could not be opened, because the lids were swollen, edematous, and glued together at the margins by the matted lashes. After the third day he was able to hold the eyes open for a few minutes at long intervals, but it was not until the fifth day that he could dispense with the compresses. His vision had been very blurred, but cleared fairly well on the fifth day. There was some yellowish exudation with considerable serous exudation. The eyes were washed frequently during the day with boric-acid solution. A hordeolum formed on the upper eyelid, near the inner canthus, on about the fourteenth day. The patient did not attempt to read for two weeks, and when he did begin he noticed no trouble, except that the eyes ached after he had read a quarter to half an hour. He could not read magazines until about ten days later.

Examination.- The bulbar conjunctiva showed the same features as in the other cases, but to a more marked degree; there was a thickening and a yellowish color, with marked congestion of the vessels of the part of the conjunctiva that was exposed in the palpebral

FIG. 187.- Marked conjunctival congestion and hordeolum of left upper lid in a case of severe mnustard gas conjunctivitis, four weeks after exposure. The hordeolum is a part of the general staphylococcus furunculosis which may characterize the later stages of the severe skill burns

fissures on either side of the cornea. At the limbic border a number of straight, small, deep vessels were seen to be injected and suggested a deeper penetration and a sclerociliary injection. The conjunctiva from the inner limbos to the inner canthus was more markedly involved.

CASE 5 (fig.188).- Exposed 30 to 45 minutes in several shifts to strong concentration of dichlorethylsulplhide vapor while wearing gas mask. Began on the evening shift at 4 p. m. and first exposed himself at this time. He noticed no eye symptoms until about 10 p. m., when there was burning, a feeling as though granules were under the lids, and a fairly profuse lacrymation. Cold boric-acid compresses and boric-acid eve washes were employed during the night and for several days. The routine treatment, as given, together with argyrol instillations, was employed. According to the physician in charge, the patient was in severe shock. There was some pain, blurred vision for near and distant objects, and a profuse lacrymation. At first the lacrymation was serous in type, but soon it became yellowish, thicker, and almost purulent. The lids were tightly sealed and somewhat edematous. The symptoms continued to be fairly severe and the treatment was kept up for about a week. It was not until this time that the vision was at all clear. The lacrymation continued for some time.

Examination.- Slightly congested areas in the palpebral aperture on each side of the cornea. The conjunctiva showed slight thickening in the same exposed areas. The discharge resulting from the collection of the pus was found both in the inner and in the outer canthi. The palpebral conjunctiva showed nothing of significance.


CASE 6 (fig. 189).- Exposed 45 minutes in several shifts to strong concentration of mustard gas vapor, while wearing gas mask. The irritation of the eyes began at about 9 p. m., some five hours after exposure. There was severe burning and a feeling of sand and glass scratching in the eyes. There was profuse lacrymation and congestion. The next morning after the exposure the exudate from the eyes was seen to be more or less purulent and glued the eyelashes together and held the eyes closed. After the parts had been treated with cold boric acid compresses during the whole first night, as had been done in all the cases, it was necessary to "flush" the eyes with the solution in order to dissolve and loosen the plastic purulent exudate. When opened, the eyes were found to be much congested and small blebs were seen, especially on the lower palpebral conjunctiva. Argyrol was instilled as in the other cases and continuous compresses were applied for five days. The eyes were opened for short spaces of time on the fourth and fifth days and a blurring and dimness of vision was noted.

Examination.- Showed the same condition that had been found in the other cases. The acute congestion had decreased and the signs remaining were those of a somewhat chronic irritative process. There was the slight thickening in the exposed part of the bulbar conjunctiva in the palpebral fissure.

CASE 7.- Exposed directly to strong vapor. The vapor irritated the eyes severely and caused pain and lacrymation for 10 to 15 minutes. There were no blebs, but there was some congestion. After this there was no more pain and the patient continued to work where there was practically always some vapor. The patient's eyes were not painful but became more and more congested.

FIG. 188.- Dichlorethylsulphide conjunctivitis four weeks after exposure to vapor. In the acute stage there was extreme photophohia, lacrymation, pain, edema, and purulent exudation. The residual congestion and seropurulent exudation are still evident

Examination.-The whole conjunctiva was inflamed, congested, and slightly edematous, the lacrymation was increased in amount. The vision was apparently not involved. The bulbar conjunctiva was more markedly congested and affected than was the palpebral. Under the boric acid, continuous-compress treatment, there was a decided improvement, and the congestion was fairly well relieved in about four days. On the fifth day the area which was most markedly injected was in the bulbar conjunctiva of the palpebral fissure. Some of the scleral vessels in this region were injected. A small subconjunctival fatty area (pinguecula) was present.

CASE 8.- Exposed to fumes of ethyl alcohol, ethylene, and some dichlorethylsulphide. The eyes showed chronic irritation, thickening of the bulbar conjunctiva in the palpebral aperture, and some yellowish discoloration and areas that appeared to be fatty deposits. The patient complained of impaired vision and lacrymation. He previously had one severe arm burn, but there were no eye symptoms at that time.

CASE 9.-The patient had a slight irritation and itching and some congestion of the conjunctiva for about two weeks. At first the symptoms were very mild and noticeable only after exposure to dichlorethylsulphide. The symptoms were exaggerated and prolonged if the patient sat through a moving-picture show. Bright sunlight also increased the symptoms. The conditions were more or less cumulative and the symptoms gradually increased. During the second week he had lacrymation, congestion of the bulbar conjunctiva in the palpebral fissure and difficulty in accommodation in the right eye with more or less blurring of vision. There was also some sticky seropurulent exudate, and in the morning the eyelids and lashes were glued together, and a drop of pus was seen in the inner canthus. He used only the boric acid wash.


Examination.- He presented a picture similar to the previous cases, with a thickening of the conjunctiva of the eyeball that was exposed in the palpebral fissure. The right eye showed slightly more congestion and thickening than the left, and there was some ciliary injection about the sclerocorneal junction. The palpebral conjunctiva, especially over the lower right eyelid, showed some congestion and slight thickening.

CASE 10.- Exposure without mask to strong concentration of mustard-gas vapor. Symptoms developed three hours later. He complained of severe irritation and pain in the eyes and the lids were swollen, edematous and could hardly be separated. The routine cold boric acid compresses were used, and it was not until three days later that the edema disappeared.

CASE 11.- Presented a somewhat different aspect. He was exposed for a few minutes to quite a heavy dose of dichlorethylsulphide vapor. He had some slight irritation and lacrymation, but no congestion to any marked degree. There was, however, a diminution of vision in the right eye which was said to be very marked. Both near and far vision were greatly reduced. There was blurring of all images.

Report of eye specialist.-"In regard to Case 11, whom I this day examined, I find his vision in right eye 4/200. Under atropine I find he has at least three diopters of hypermetropia, with some astigmatism, the correcting of which does not improve his vision. If this condition has been brought about in the last three weeks, as he claims, though the fundus does not show this, it must be toxic. Otherwise, it is amblyopic and his previous vision was not normal."

FIG. 189.- Dichlorethylsulphide conjunctivitis four weeks after exposure to vapor. The severity of the original process is indicated by the severe skin changes. The persistent congestion is the sole evidence of the severe conjunctivitis that was present

CASE 12.- Had been exposed to dichlorethylsulphide for about one month. He had some irritation and congestion about one week ago. The condition improved, but the patient complained of failing vision, dimness, and blurring. Both eyes were affected and the condition progressed rapidly; both near vision and distant vision were involved. The patient had to bring a printed page to within 6 inches of the eye in order to be able to read. A close examination of the corneae failed to reveal any opacities to account for the trouble. The corneal epithelium showed no apparent thickening or dulling. The conjunctiva showed some thickening and signs of chronic irritation in the part exposed in the palpebral fissure. It was considered that this case, like Case 11, was one of some internal fundus pathology or an accommodation disturbance.

CASE 13.- Exposure of one hour to very dilute concentration of mustard gas vapor. Six hours later, marked burning and irritation of conjunctiva, lacrymation, blurring of vision, and reddening of conjunctival surfaces. These symptoms lasted three days and gradually decreased, but a feeling of roughness of lids and visual disturbance persisted for nearly two months


From these cases it will be seen that exposure to varying concentrations of vapor of dichlorethylsulphide for varying periods produced a conjunctivitis showing all stages and degrees of intensity, from a simple acute type to a severe chronic proliferative conjunctivitis. The symptomatology and clinical picture varied greatly. The lesions were most marked in that portion of the bulbar


conjunctiva exposed in the palpebral fissure. The milder cases recovered after several days or weeks, but the more severe cases developed chronic hyperemia of the conjunctiva, new formation of vessels and scar tissue in the most severe, with more or less permanent disturbances of vision. One of the cases observed developed an almost complete amblyopia in one eye, so that only the perception of light and shadows was possible. In another case marked bilateral gradual reduction of vision was noted. A xanthomalike pigmentation was also noted in the chronic cases, the pigmentation developing near the outer or inner sclerocorneal junction, or over the corneal limbus.

There was no evidence of any metastatic involvement of the eye. In cases showing severe burns of other parts of the body, arm, leg, etc., no conjunctivitis, even of the simplest type, developed. Such exceptions are explained entirely by the fact that the vapor did not reach the eyes externally. In other cases, with severe burns over the entire body with the exception of the face, which was protected by the gas mask, no eye or conjunctival symptoms were noted.


1. The action of mustard gas upon the cornea and conjunctiva is essentially the same as that upon the skin. The conjunctiva is, however, less susceptible to the action, or better protected, as the degree of necrosis produced in it is always less than that in the cornea or the epidermis.
2. Exposures to dilute concentrations of the vapor produce slight degenerations of the corneal and conjunctival epithelium, followed by a simple conjunctivitis. The use of a 2 percent alkaline aqueous fluorescein solution in demonstrating the necrosis of the corneal epithelium within 10 to 15 minutes after exposure to gassing has great clinical value.
3. Exposures to stronger concentrations produce a more or less complete necrosis of the corneal vertex, extending throughout the entire depth of the cornea. Purulent exudation into the anterior chamber may occur; but no changes except congestion and edema were observed in the posterior chamber or optic nerve in noninfected cases. In severe cases iridocyclitis and iritis may occur without secondarv infection. The conjunctival epithelium also suffers necrosis, and there results an intense edema of the subconjunctival tissues with marked congestion, multiple hemorrhages, leucocyte infiltration, and frequently secondary liquefaction necrosis. The depth of the necrosis in the conjunctiva is much less than that in the palpebral epidermis. This difference in degree of escharization can be explained in part by the penetration of the hair follicles on the skin surface, and in part by the moistness of the conjunctival surfaces and the lacrymation. A diffuse mild inflammation of the peribulbar tissues occurs, often with marked infiltration of the ocular muscles. Purulent panophthalmitis may result from secondary infection, but is rare.
4. No metastatic lesions of the eye could be produced experimentally by applications of mustard gas to other regions of the body, or by subcutaneous or intraperitoneal injections.
5. For the milder forms of mustard-gas conjunctivitis immediate irrigation with the 0.5 to 1 percent chlorcosane solution of dichloramine-T, followed by frequent irrigation with saturated boracic acid solution, is recommended: for the severe forms the same initial treatment, followed by frequent irrigations with the dichlorarmine-T alternating with boracic acid. We advise against the


use of bandages or compresses bringing pressure upon the eye, against the use of colloidal silver preparations, and against the use of cocaine. During exposure to mustard-gas vapor the dichloramine-T solution may be used as a prophylactic agent.
6. Healing in the more severe forms results in vascularization and cicatrization of the cornea with marked disturbances in vision. Even in the milder forms of conjunctivitis, localized roughness or irregularity in the conjunctival surface may persist for weeks as the result of localized edema, hyperemia, cellular infiltration, etc. Serious refractive errors and reduction of vision result, even in mild cases. For the correction of the disturbances of vision the patient should be referred to a competent specialist.



In the gassing investigations use was made of a simple apparatus which served our purposes most effectively and is to be recommended for its simplicity as well as for the relative accuracy with which the gas concentration could be estimated. As shown in Figure 190, it consisted of:
A. One or more bottles containing sulphuric acid for drying the air admitted.
B. A containerfor the liquid mustard gas so arranged as to be easilv detached from the tubing, so that when stoppered with a ground-glass stopper it could be weighed with its contents. Above this was at shunt tube for varying the amount of pure air admitted to the chamber, thus varying the concentration of gas obtained. Within the mustard gas bottle strips of absorbent paper were so arranged as to increase the evaporation surface. When used for experimental work with toxic substances, which are gaseous at room temperature, the gas intake replaced this weighing bottle.
C. A gassing chamber which was large enough to hold several small animals or a large dog. It had a removable plate-glass top, so that the animals could be observed during gassing; this was sealed during the experiment with the gutta-percha-tallow mixture, commonly used for making air-tight seals for museum jars, and could be held in position by a heavy weight if necessary.
D. and E. Bottles for removing the gas from the air flowing from the chamber, so that the amount passing over could again be estimated here if desired. When used for mustard gas these bottles might contain a chlorinated solution for the destruction of the gas.
F. A standard gas meter for measuring the amount of air passing through the chamber.
G. A suction pump attached to the city water supply.

After detaching the weighing bottle, air was drawn through the gassing chamber for some time in order to wash out the mustard or other gas remaining. The weight of mustard gas used was obtained by difference, and the amount of air drawn through during the period in which the bottle was attached was recorded from the gas meter. The concentration represented by these figures was obtained by reference to at transformation table. The results were as approximately correct as could be obtained byt any form of gassing chamber, and the apparatus band the advantage of simplicity. As far as mustard gas is concerned,


such factors as variable absorption by the hair of the animal, etc., introduce unavoidable errors, so that absolute accuracy of method could not be obtained. regardless of the limitations of the apparatus.

The animals were killed by a direct blow upon the neck, to avoid the changes in the respiratory system produced by an anesthetic. Autopsies were made as quickly as possible after the death of the animal. The tissues were fixed in formol, embedded in paraffin, and the usual stains, such as hematoxylin and eosin, were employed.


Various series of animals were exposed to varying concentrations of mustard gas in the gassing chamber for varying periods of time. It was found that rabbits would survive a 40-minute exposure to a dilution of 1: 1 10,000, recovery taking place after a period of respiratory involvement. For purposes of brevity, typical protocols have been selected and are given here.


RABBIT 32.- Exposed in the gassing chamber for 40 minutes to a 1: 110,000 concentration. During the gassing the rabbit frequently changed position, rubbed its nose, and showed

FIG. 190.- Experimental gassing apparatus. Pathological laboratory, University of Michigan. A, washer containing sulphuric acid; B, gas container; C, gassing box; D and E, degassing bottles; F, standard gas meter; G, suction pump

signs of irritation. When removed from the gassing chamber the rabbit appeared unaffected in any way. In two and one-half hours conjunctival erythema and increased lacrymation were evident, these symptoms increasing until the animal was killed. Seven and one-half hours after removal from the box, symptoms of coryza manifested themselves. Twelve hours after removal from gassing chamber, rabbit was killed by blow upon tlle neck. The conjunctivae presented marked congestion and edema, and were covered with puruletit flakes. The bulbar conjunctivae showed a marked collar of edema at the limbos. The cornea showed a slight haziness, and at its vertex there was a definite necrosis of the superficial epitheliums. This area was irregularly oval in shape. The lid margins showed congestion and the skin about mouth and nostrils was erythematous.

Autopsy showed right-sided dilatation of the heart. No fluid was found in the pleural cavity, and there were no pleural changes. On section the lungs showed marked congestion and edema without hemorrhages or atelectasis visible to the naked eye. No differences between upper and lower lobes were noted. The trachea was filled with frothy mucus and the mucosa was congested, particularly in the upper portion, the congestion diminishing below toward the bronchi. The mucosa of the entire upper respiratory tract, nose, mouth. pharynx and larynx showed congestion without hemorrhages.

Microscopic findings.-Nose: Sections from the skin about the nostrils showed slight pycnosis of the epidermis and congestion of the vessels, without other changes. The mucous membrane of the nose showed a marked congestion, more marked pycnosis of the epithelium and slight edema. Sections from the pharynx and larynx showed a marked congestion with extreme mucous degeneration of the mucous glands. There was pycnosis of the upper layers


of the epithelium of the mucous membrane but no definite necrosis. The acini of the mucous glands were greatly enlarged and filled with deep blue-staining mucus. The ducts were dilated and filled with mucus. There was but slight edema of the mucous membrane and no hemorrhages. These signs diminished in the trachea except for the edema, which was somewhat greater in the lower part of the trachea than above, and there was an increased number of wandering cells in the mucosa. The columnar cells of the tracheal mucosa showed more marked changes than the squamous epithelium of the upper respiratory tract. The great majority showed a hydropic or a mucoid degeneration. The epithelium was intact for the greater part but small patches of desquamation occurred. The larger bronchi presented the same appearance as the trachea, though to a somewhat lesser degree. The smaller bronchi were filled with mucus. The epithelium showed mucoid degeneration and occasional areas of desquamation. The walls of the larger bronchi were edematous, the vessels markedly congested, but there was no increase in leucocytes. The exudate in the bronchi was entirely mucoid or albuminous in character, not fibrinous. There was no hemorrhage into the bronchi.

FIG. 191.- Rabbit. Exposed 40 minutes to a 1:110,000 concentration. Killed 12 hours after removal from gassing chamber. Section of ting. Marked congestion, edema, and area of partial atelectasis alternating with those of emphysema

The lung tissue showed extreme congestion and marked edema with numerous minute hemorrhages, too small to be seen with the naked eye. (Fig. 191.) A majority of the alveoli were filled with a heavy albuminous precipitate, but scattered throughout the lung were numerous emphysematous alveoli and dilated bronchioles. No areas of complete atelectasis were seen, although edematous areas showed partial collapse.

The changes in the respiratory tract of this animal were those of an acute catarrhal rhinitis, pharngitis, laryngitis, tracheitis, and bronchitis, decreasing somewhat in intensity from above downward, with pulmonary congestion and edema.

RABBIT 33.- Exposed 20 minutes to a concentration of 1:15,000. During the exposure the rabbit changed its position from that of facing the inflowing mustard gas to the opposite direction. Three hours after removal from the gassing chamber the animal showed increased lacrymation in both eyes. The borders of the eyelids, skin areas about the month and nostrils, the ears and all parts of the body where the hair was short and thin exhibited


a marked erythema. The animal showed marked photophobia and irritation of the eyes. Eight hours after removal from the chamber flakes of purtulent material were seen over the conjunctivae and a definite coryza had developed. The conjunctive were edematous and congested. These symptoms increased for 36 hours, when the animal was killed.

Autopsy showed a severe conjunctivitis and coryza. Right-sided dilatation of the heart. The upper air passages were filled with foamy exudate and the mucosa was congested. Pleural cavities and pleurae negative. The lungs were markedly congested, the right lung more so than the left. The lungs appeared air-containing throughout, except for the middle lobe on the right, which was solid in areas, dark red in color. On section it bled but slightly. Beyond congestion no other changes were found in any organs or tissues.

Microscopic findings.-The cutaneous borders of the nostrils showed necrosis, edema, congestion and marked leucocyte infiltration of the corium. The mucous membrane of the nose presented patches of necrosis of the epithelium, congestion and edema, areas of small-celled

FIG. 192.- Exposed 20 minutes to a concentration of 1:15,000. Killed 36 hours after gassing. Section of trachea showing acute catarrhal desquamative tracheitis; marked mucoid generation of the epithelium; congestion and edema of the submucosa. Lumen filled with mucous containing many desquamated cells

 infiltration of the submucosa, and marked mucous degeneration of the mucous glands. Mouth and pharynx: Sections of tongue and pharyngeal wall showed a contraction of the upper half of the squamous epithelium with pycnosis of the nuclei, congestion of the vessels, slight edema of the submucosa and slight small-celled infiltration. The larynx presented patches of necrosis in the hvperemic mucous membrane, with marked edema extending to the cartilages. Small-celled infiltration was well marked. Mucous glands showed mucoid degeneration. The surface of the mucosa of the trachea was covered with patches of mucus containing desquamated cells. The epithelium showed marked mucoid and hvdropic degeneration; there were large areas of complete necrosis with desquamation. In the submucosa there was a marked edema extending to the cartilage rings. The vessels were markedly congested (Fig. 192). The larger bronchi showed marked degeneration and necrosis and desquamation of the epithelium, the epithelium being represented for the greater part by a single line of nuclei at the base. Many of the bronchi were filled with an exudate of mucous containing many desquamated and degenerating cells but few leucocytes. There was some edema of the walls of the bronchi and the number of leucocytes was increased around the bronchi. The smaller bronchioles showed a better preserved mucous membrane


but many of the cells were vacuolated, presenting mucoid or hydropic degeneration, and desquamation was frequent. The lung showed practically the same picture as in the preceding; marked congestion and edema, small hemorrhages into the alveoli and emphysematous alveoli and dilated bronchioles. Many of the edematous areas showed partial atelectasis. The apparently solid area from the right lung presented a more marked atelectasis and a greater degree of edema but no pneumonia. Other organs showed marked congestion without other changes.

The microscopical picture in this case was similar to that in the preceding, but the changes were somewhat greater in intensity with a greater degree of necrosis and a well-defined leucocyte reaction.

RABBIT 30.- Exposed 30 minutes to a concentration of 1:15,000. Killed 44 days after removal from the gassing chamber. Within 5 minutes after removal from the box the animal showed the first signs of irritation, rubbing its eves and nose frequently. Six hours afterwards there was a well-developed conjunctivitis. By 24 hours the conjunctivitis had

FIG. 193.- Rabbit. Exposed 30 minutes to a concentration of 1:l5,000. Killed 4¼ days after gassing. Section of larger bronchus showinig lumen flled with edema fluids. Bronchial epithelium shows marked mucoid and hydropic degeneration

greatly increased, with marked conjunctival edema, and the animal showed a marked bilateral coryza. By the second day the snuffles and coryza were much worse, the animal showing a marked respiratory wheezing, audible several feet away. The conjunctivitis had become distinctly purulent in character, with multiple subeonjunctival pin-point hemorrhages, extreme edema and beginning corneal ulceration. Throughout the day the snuffles and wheezing greatly increased, the animal appeared sick, restless, with respirations greatly increased and shallow, these symptoms reaching their height in the evening of the second day. From the morning of the third day the respiratory symptoms gradually improved until the animal was killed, 4¼ days after gassing.

Autopsy.-The eves showed a very severe purulent conjunctivitis, with characteristic porcelain appearance of the cornea. The mucosa of upper respiratory tract showed marked congestion and edema and mucous exudate, diminishing in intensity frona the nostrils to the nasopharynx. In the anterior 2 cm. of the nasal tract the exudate was purulent in character. The nasopharnx, larynx, and trachea presented marked congestion of the mucosa, and the lumen of the trachea was filled with a frothy mucous extending into the bronchi. No hemorrhages


or ulcerations were seen in the mucosa of the upper respiratory tract. Pleural cavities and pleurae were negative. The lungs were uniformly markedly congested and apparently air-containing throughout. No pneumonic areas or hemorrhages could be felt or seen. The heart presented a marked right-sided dilatation. Beyond a marked congestion, other organs and tissues showed nothing.

Microscopic findings.-Nose: Sections from the anterior nostrils showed a marked necrosis and ulceration of the mucosa with a marked edema and congestion of the submucosa and a polynuclear infiltration. Great numbers of staphylococci were seen on the necrotic mucosa. Throughout the submucosa were numerous minute hemorrhages by diapedesis, and in one medium-sized blood vessel a definite thrombus. The mucous glands showed the squamous epithelium to be intact for the greater part, but the outer half was necrotic, dry, looking like stratum corneum. Larynx: The surface epithelium was completely necrotic in many areas, particularly where a mucoid exudate lay upon the surface. In other areas it was preserved, but had lost its columnar appearance, and was reduced to a layer of pycnotic nuclei staining almost black with hematoxylin. In the mucoid exudate in the lumen there

FIG. 194.- Lung of same rabbit as Figure 193. Acute congestion and edema

were great numnbers of swollen desquamated mucoid epithelial cells and very few leucocytes. The submucosa was edematous, the edema, however, being very irregularly distributed. The blood vessels showed marked congestion and the number of leucocytes in the submucosa was not increased. The laryngeal mucous glands showed marked mucoid degeneration. Many of the larger bronchi were filled with mucus. The columnar cells showed marked mucoid degeneration and pycnosis. Practically every cell was converted into a goblet cell. There was no increase of leucocytes in or about the bronchi. The pulmonary vessels showed extreme congestion and in many of these were large masses of fibrin, irregularly scattered through the red blood cells, or at times somewhat laminated, presenting the appearance of recent thrombosis. Throughout the lung atelectatic areas alternated with emphysematous. The alveolar spaces of the atelectatic areas showed marked edema, being filled with a pink-staining finely granular precipitate. The bronchioles in these atelectatic areas were distended and were filled with either a similar edematous fluid or a more mucous fluid. No pneumonic areas were found and no large hemorrhages. Minute hemorrhages by diapedesis were found along the walls of the greatly distended capillaries. The other organs showed nothing but intense congestion. (Figs. 193, 194.)


The lesions were practically the same as in rabbit 33, except that they were more severe in the nose and mouth. The purulent character of the nasal lesions was probably to be explained by the development of a secondary infection due to a staphylococcus.

RABBIT 31.- Exposed 35 minutes to a concentration of 1: 30,000. On removal from the gassing box the animal showed some irritation of the eyes and nose. Four hours after removal the animal had developed a well-marked conjunctivitis, with erythema of the exposed skin surfaces and beginning snuffles with scant nasal secretion. Within 24 hours there was marked increase of conjunctivitis and coryza. The animal was killed 30 hours after gassing.

Autopsy.- Eyes showed a marked conjunctivitis, with purulent exudate. Exposed skin surfaces showed erythema and edema. The mucosa of entire respiratory tract from nose to bronchi showed congestion, slight edema, and abundant mucous exudate. Pleural cavities negative. Pleurm negative. Lungs presented uniform congestion without consolidation, the lung tissue being apparently air-containing throughout. No hemorrhage or atelectatic or pneumonic areas visible to the naked eye. The heart showed dilatation of the right side. In other organs nothing notable but congestion.

Microscopic findings.- Nose: Squamous epithelium of the anterior nostrils showed patches of complete necrosis, these minute erosions or ulcerations being covered with a fibrinopurulent exudate. The remaining epithelium showed more or less pycnosis. Thern was very little edema of the submucosa and no small-celled infiltration or hemorrhages. On the skin side of the nostrils the edema was more marked than on the mucosal side, and there was marked necrosis, pycnosis, and desquamation of the epidermis. Pharynx: Squamous epithelium of mouth and pharynx showed patches of pyenosis, contraction of the upper third with pycnosis of the nuclei resembling cornification, congestion of the blood vessels and slight edema of the submucosa. Larynx: The columnar epithelium showed marked mucoid or hydropic degeneration. The majority of the cells of the upper layer were either vacuolated or were goblet cells. The nuclei stained very heavily, but the epithelium was for the greater part intact, without desquamation. In the lumen there was a thin albuminous fluid containing very few desquamated cells, leucocytes, no fibrin, and but a small amount of mucin. The submucosa showed a rather marked edema with the leucocytes slightly increased. The vessels were markedly congested. The mucous glands showed increased mucus formation. Trachea: The epithelium was intact throughout the greater portion, but showed mucoid and hydropic degeneration. The lumen was filled with a thin mucoserous fluid. Well-marked edema of the submucosa with some increase in leucocytes. Marked congestion of the vessels. Bronchi: The lungs presented practically the same appearance as in the previously described cases but with less edema and on the whole a less marked congestion. The bronchi contained a smaller amount of fluid. The epithelium was better preserved, although showing mucoid change and vacuolar change in the larger ones. There was very little atelectasis, the lungs being rather emphysematous throughout. There was no pneumonia and no hemorrhage. The other organs showed chronic congestion. the liver a chronic coccidiosis with cirrhosis. (Figs. 195, 196.)

In this rabbit the lesions in the anterior nares were severe, the degree of necrosis there being comparable to that on the conjunctiva and cornea, but the laryngeal, tracheal, and bronchial lesions were of the mildest degree.

RABBIT 46.- Gassed six hours at a concentration of 1:50,000. Animal died 60 hours after removal from the gassing chamber. During the gassing the animal showed irritation within half an hour. At the end of an hour it became drowsy and was quiet, with its eyes half closed. Within three hours there was a definite lacrymation and conjunctival edema, these conditions increasing during the remainder of the gassing. When taken from the box the eyes were very much congested, edematous, and showed profuse lacrymation. By the next day the rabbit showed marked respiratory involvement, coughing all the time, and bubbling rȃles could be heard throughout its chest. These conditions increased during the next 2 days, the animal dying 60 hours after removal from the gassing chamber.

Autopsy.- Both eyes presented marked corneal opacity with mucopurulent conjunctivitis. Pharyngeal mucosa very markedly congested and edematous and covered with mucopurulent exudate. Buccal mucosa negative in appearance. Marked edema of all


laryngeal structures and tracheal mucosa, the tracheal edema being most marked in the upper part, just beneath the larynx, where there were also small hemorrhages in the mucosa. Pleural cavities and pleurte negative. Pericardial fluid increased, clear. Heart dilated; both sides filled with partly clotted blood. Both lungs were voluminous and air-containing without any airless areas except a small atelectatic area in the border of the lowest left lobe. On section the lungs showed moderate congestion and edema without any hemorrhages or pneumonic or atelectatic areas visible to the naked eye. Other organs showed no changes except congestion.

Microscopic findings.- Nose: Anterior nares showed a shrinking and pycnosis of the epithelium, the outer layers being desquamated and light staining, resembling cornified epithelium with pycnotic nuclei. In many areas this pycnosis and shrinking extended to the lowest layer of the epithelium. Through the lowest layer there were many hydropic cells and occasionally small vesicles were formed. The ciliated columnar epithelium showed a more marked necrosis, vacuolization and desquamation. There was marked edema of the

FIG. 195.- Section of laryngeal mucosa of rabbit. Exposed 35 minutes to a concentration of 1:30,000. Killed 36 hours after gassing. Pycnosis and mucoid degeneration of the epithelium Marked congestion and edema of the submucosa

submucosa extending to the cartilages and through the muscles. The mucous glands showed extreme mucoid degeneration; the blood vessels, extreme congestion. There were minute hemorrhages about the mucous glands and the number of wandering cells was increased. The buccal mucosa and the tongue showed necrosis of the upper half to two-thirds of the squaniouis epithelium, with partial desquamation of these dead cells in lamellae. There was slight edema and moderate congestion of the subepithelial tissues. The mucosa of the pharynx showed the same changes with a more marked edema of the submucosa. Larynx: In the lumen, lying upon the surface, there was a fibrinopurulent membrane which was firmly attached in areas where the epithelium was completely lost. In these areas the picture was that of a diphtheritis. Colonies of staphylococci were present in the diphtheritic membranes and on the necrosecd epithelium. Where the epithelium was preserved it showed marked vaciolation with hydropic and mucoid degeneration. In some areas the epithelium was still attached but was necrotic and infiltrated with polynuclears and eosinophile cells. There was a very marked subepithelial edema, extreme congestion of the blood vessels and


areas of leucocyte infiltration, many of these being eosinophiles. This infiltration was most marked in the neighborhood of the diphtheritic areas. The cervical lymph nodes in this case showed marked eosinophilia and great numbers of hemophages in the sinuses. Trachea: Mucosa of the trachea showed marked mucoid and hydropic degeneration. There was a diphtheritic necrosis, but the process was less marked than in the larynx. Edema and congestion of the wall were about the same. Bronchi: The epithelium of the larger bronchi showed marked degeneration and necrosis. There was an increase in leucocytes in the walls of the larger bronchi, the submucosa was edematous and the vessels were markedly congested. The majority of the smaller bronchioles contained a mucopurulent exudate and the bronchial wall was infiltrated with leucocytes. Around many of the bronchioles there were definite areas of a hemorrhagic purulent bronchopneumonia. Colonies of staphylococci were found in each of these areas. Between these areas the alveoli were overdistended and emphysematous. Other alveoli contained a heavy albuminous precipitate of edema. The liver presented a marked nutmeg liver, central necrosis and congestion of lobules. The kidneys showed marked cloudy swelling and congestion. (Figs. 197, 198.)

FIG. 196.- Section of tracheal wall of same rabbit as Figure 195. Similar changes in epithelium and submucosa

The respiratory lesions in this animal were more pronounced than in any of the preceding, as was shown by the diphtheritic necroses in the larynx and trachea, the purulent bronchitis, and bronchopneumonia. The cause of death was considered to be an infective (staphylococcus) bronchopneumonia, secondary to the lesions produced by the gassing.

RABBIT 45.- This rabbit was gassed at the same time as rabbit 46, in the same box, for six hours at a concentration of 1: 50,000. Its reaction during the gassing and afterwards appeared to be identical with that of No. 46, but the animal lived for seven days.

Autopsy.- The autopsy findings were the same as for No. 46, but the changes were more severe in character. The anterior nares showed extensive ulcers covered with diphtheritic membrane, and the nasal cavity was filled with a mucoid fibriniopurulent membrane, this membrane extending through the nasopharynx and into the larynx and trachea. This membrane was in part firmly adherent to the wall, particularly along tile posterior wall,


while anteriorly it was loose and separated as a cast of the tracheal tube. Heart showed marked dilatation of both cavities. Pleurae and pleural cavities negative. Large bronchi filled with purulent exudate. Lungs voluminous, markedly congested and edematous, with the picture of a diffuse bronchopneumonia.

Microscopic findings.- The microscopic examination of the nose, nasopharynx, larynx, and trachea revealed the presence of a diphtheritic inflammation, with a membrane containing numerous colonies of cocci. This membrane was firmly attached wherever the epithelium was completely lost, but was loose and separated from the surface wherever the epithelium was still intact. In the larynx and trachea a large part of the columnar epithelium was replaced by a regenerating layer of large, deeply staining pavement or cuboidal cells. The submucosa showed marked congestion, edema, scattered hemorrhages, and leucocyte infiltration. The larger bronchi were filled with a fibrinopurulent membrane which was adherent to the wall and contained many large colonies of cocci. The epithelium was entirely absent except in a few places where there was a regenerating layer. Lungs: The lungs presented

FIG. 197.- Exposed six hours to a concentration of 1:50,000. Died 60 hours after gassing, section of large bronchus, showing purulent necrotic bronchitis

the picture of an advanced fibrinopurulent bronchopneumonia with large areas of consolidation. All of the bronchioles appeared as abscesses. In some of these there were many large epithelial giant cells, syncytial in type, showing an attempt at regeneration, but the great majority showed complete destruction of the epithelium. Syncytial epithelial giant cells were also found in some of the alveoli. Colonies of cocci were found everywhere. Many of the pneumonic areas showed older and fresh hemorrhages and marked inflammatory edema. (Figs. 199, 200.)

This case presented a more marked diphtheritic inflammation of the upper respiratory tract and a diffuse purulent bronchopneumonia due to secondary infection following gassing.

RABBIT 43.- Exposed for 12 hours to a concentration of 1: 50,000. Died 54 hours after removal from the gassing chamber. In the first stages of the gassing the animnal showed nasal irritation, but later became drowsy and depressed, and sat with eyelids nearly closed. At the end of two hours increased lacrymation and conjunctival edema were first noted.


When removed from the gassing chamber there was a well-marked conjunctivitis, with coryza and salivation. The animal would not eat or drink, but appeared stupefied. The coryza gradually increased in intensity and within 6 hours after removal from the chamber bubbling respiratory sounds were heard over its thorax. Coryza and conjunctivitis increased during the first 24 hours, at which time there was a profuse seropurulent nasal discharge. The wheezing and bubbling respiratory sounds could be heard some distance from the cage. Respirations were rapid and apparently difficult. Thirty-six hours after gassing the rabbit was still much depressed and the respiratory sounds were louder. The animal frequently sneezed and coughed. There was a marked purulent discharge from the eyes and nose. Forty-eight hours after removal from the gassing chamber the animal was very weak, the respiratory movements more rapid and much forced, the thorax heaving. Both eyes were sealed with a thick purulent exudate. After increasing respiratory difficulty the animal died, 54 hours from the time of exposure.

FIG. 198.- Section of lung from same case as Figure 197. Area of purolent bronchopneumonia with colony of cocci in the center of the field

Autopsy- Marked purulent conjunctivitis with corneal ulceration. Erythema of the exposed portions of the skin. Marked edema of the subcutaneous tissue, particularly in the region of head, neck, and thorax. Nose filled with purulent exudate, the mucosa showing deep necroses. Nasopharynx filled with mucopurulent fluid. Larynx and trachea filled with frothy exudate beneath which there was a membranous cast, partly adherent but decreasing in intensity downward. The mucosa of the entire upper respiratory tract was markedly congested and showed marked necrosis of the surface epithelium. The lungs showed marked congestion and edema. At the base of each lung there was an olive green area, the central portion of which was lighter in color. Bronchi filled with a thick mucopurulent exudate. Pleurae were negative. Heart showed marked dilatation of the right ventricle. All other organs showed marked congestion. Some cloudy swelling in the kidneys. The external genitals were very erythematous and edematous.

Microscopic findings.- The nose was filled with a diphtheritic fibrinopurulent exudate and the epithelium was completely necrosed, except for very small areas in the folds. The membrane wvas for the greater part adherent, but loosened in some areas. The submucosa showed marked congestion, leucocyte infiltration, and edema, extending into the muscles


and cartilages. The mucosa of the pharynx showed a more marked necrosis of the squamous epithelium than in any of the above cases. The cells of the lowest layers were markedly pycnotic. The larynx and the trachea showed a diphtheritic process, the surface being covered with a fibrinopurulent membrane which was firmly attached where the epithelium was entirely absent, but lying loosely on the surface where the epithelium was still intact or regenerating. In the membrane there were colonies of cocci. The epithelium of the larynx and trachea was reduced to a single layer of pycnotic nuclei over a large extent of surface. The submucosa showed marked fibroblastic proliferation, leucocyte infiltration and edema, with minute hemorrhages. The mucous glands of the larynx and trachea showed some acini and ducts distended with mucus, but the greater number of the acini appeared small, collapsed, and pycnotic. The inflammatory infiltration and edema extended into the muscle. The mucosa of the upper part of the esophagus showed marked pycnosis and necrosis of the upper layers of the epithelium. The large bronchi showed a nearly complete necrosis of the

FIG. 199.- Rabbit. Exposed for six hours to a concentration of 1:50,000. Died seven days after gassing. Eschar from upper portion of larynx

epithelium and a fibrinous membrane adherent to the surface forming casts of the bronchial tubes. There was marked congestion of the wall and leucocyte infiltration of the surrounding tissues. The smaller bronchioles were filled with mucous and albuminous exudate with increased leucocytes. The epithelium was intact, but showed mucoid degeneration and pycnosis of the nuclei. A very few bronchi contained a mucopurulent exudate. The lungs showed extreme congestion, numerous hemorrhages by diapedesis, a few small areas of early bronchopneumonia and atelectatic edematous areas beneath the pleuira. The only bacteria found were those in the bronchioles containing the mucopurulent exudate. Marked congestion of all other organs. Cloudy swelling of the liver and kidney cells.

This case differed from the preceding in the severe diptheritis of the upper respiratory tract with more marked evidences of healing in the larynx and trachea. It showed a severe secondary infective process of the larynx and trachea with less involvement of the bronchi and lungs.


RABBIT 44.- Gassed for 12 hours at a concentration of 1: 50,000, at the same time with rabbit 43. Reaction during the gassing identical. Developed a corresponding conjunctivitis and coryza with similar pulmonary symptoms. This animal, however, survived until 92 hours after the end of the exposure.

Autopsy.- Generalized mustard-gas burn of the skin, especially where the hair was thin, in the form of a marked erythema, marked subcutaneous edema and dilatation of the superficial vessels. The untreated left eye showed a severe purulent conjunctivitis with corneal opacity and superficial ulceration. The right eye had been treated with dichloramine-T and showed a much less severe conjuncietivitis and in jury to the cornea. The upper respiratory tract was filled with frothy exudate. The mucosa, particularly of the trachea, showed a diphtheritic necrosis with membrane, most marked at the larynx and just below it, but extending into the bifurcation. Bronchi filled with frothy exudate. The lungs were voluminous, with marked congestion and very edematous on section. Atelectatic areas seemed to alternate with emphysematous. At the anterior border of the lowest lobe of the left lung, there was

FIG. 200.- Same rabbit as Figure 199. Section of larynx showing diphtheritic ulcer an area greenish gray in color, softened and apparently necrotic. Right ventricle and comus markedly dilated. All other organs congested. In the intestine, small necrotic areas, with inflammatory reaction about them, were found, the entire intestine being markedly congested.

Microscopic findings.-Nose: Slough of the skin of the nose adherent. On the mucous membrane side was an ulcer with slough adherent in some areas. Fibroblastic proliferation well developed. Mouth: Tongue showed numerous small areas of ulceration and sloughs extending down to the muscle. The remaining portion of the epithelium was intact but showed pycnosis and contraction of the outer half. Pharyngeal mucosa showed the same lesions. Mucous glands of nose and nasopharynx showed very marked mucoid degeneration. Intense congestion of the neighboring vessels. Patches of leucocyte infiltration. The larynx showed a purulent infiltration of the wall extending to the cartilages. A fibrinopurulent membrane was adherent to the surface over the greater portion and beneath the membrane there was a phlegmonous infiltration of the wall of the larynx. Small patches of epithelium upon the surface had the appearance of regenerating epithelium, regeneration taking place apparently from the ducts of the mucous glands. Numerous colonies of cocci


were found in the pus. Below the larynx the epithelium was present over the greater part of the tracheal mucosa, which showed mucoid and hydropic degeneration. There were also islands of regenerated epithelium with some fibroblastic proliferation in the submucosa. The tracheal wall showed congestion, edema, small-celled infiltration, and fibroblastic proliferation. The larger bronchi were filled with pus. Small diphtheritic areas were found on the mucosa, where the epithelium was entirely gone and the fibrinopurulent exudate was adherent. Walls of the bronchi were edematous, infiltrated with leucocytes, and the blood vessels were congested. Any epithelium remaining showed mucoid and hydropic degeneration. The lungs showed extreme congestion and numerous small hemorrhages. The alveoli for the greater part were overdistended, emphysematous, although there were many small atelectatic areas in the neighborhood of plugged bronchi. The atelectatic areas showed a marked edema. Practically all the bronchi and bronchioles were filled with a purulent exudate and around many there were areas of purulent bronchopneumonia. Colonies of cocci were found in the pus in the bronchi. Other organs showed intense congestion. No other organs showed specific changes with the exception of stomach and intestine, in which small sloughs were found, probably the result of local action of mustard gas swallowed in food or saliva.

This rabbit showed a severe purulent and diphtheritic inflammation of the upper respiratory tract and a purulent bronchopneumonia, due to a staphylococcus, secondary to the more severe lesions caused by the gassing.


From repeated and carefully controlled animal experiments the following conclusions were arrived at in regard to the production and nature of the respiratory lesions due to mustard gassing:

1. The respiratory lesions are proportionate to the concentration of mustard gas employed and to the length of exposure.
2. The mildest lesions are those produced by short exposures, such as ten to fifteen minutes, to dilutions of 1:110,000 or over; or very short exposures, one to several minutes, to higher concentrations.
3. Exposure to mustard gas causes almost immediate signs of nasal irritation on the part of the animal. It will almost immediately rub its nose and turn its back to the inflowing gas. Conjunctival symptoms, in the form of photophobia and increased lacrymation, usually appear within two to three hours, the respiratory symptoms developing two to three hours later, as a rule, in the form of snuffles, increased nasal secretion, or a more or less severe coryza. In the case of short exposure, or low concentration of the gas, the respiratory symptoms do not progress beyond a catarrhal stage.
4. In prolonged exposures, or exposures to higher concentrations, the respiratory symptoms appear at relatively the same time, but usually later than the conjunctivitis. They then increase in intensity for several days until a picture of diffuse diphtheritic inflammation of the respiratory tract is produced, as manifested by increasing difficulty in respiration, exudation, Wiles, and coughing. The animal may die within 48 hours, or later.
5. The pathologic lesions in the mildest cases consist of a superficial degeneration or necrosis of the epithelium of the mucous membrane, with congestion and edema and increased mucus secretion. These milder lesions are found chiefly in the anterior nares, the pharynx. larynx, and upper portion of trachea; but animals killed after such exposures invariably show a more or less marked pulmonary congestion and edema, increased bronchial secretion, and small. scattered atelectatic areas. From lesions of this degree recovery. without secondary infection, usually takes place.


6. The severer lesions consist of deeper necrosis of the mucosa of the respiratory tract and the formation of a diphtheritic membrane in the anterior nares, nasopharynx, larynx, trachea, and larger bronchi. The most severe lesions are always in the nares, pharynx, larynx, and upper part of the trachea, the intensity decreasing toward the bronchi. The respiratory columnar epithelium suffers more than the epithelium of the mouth cavity, although irregular areas of complete necrosis of the latter may be produced. The entire epithelium of the larynx, trachea, and larger bronchi may be killed outright and the surface covered with a diphtheritic membrane, which, for the greater part, is easily detached, forming a cast of the larynx, trachea, and larger bronchi. The laryngeal and tracheal walls show edema, congestion and small-celled infiltration, but the edema is always less marked than in the case of the subcutaneous tissues or the conjunctivae.
7. In the severe cases, secondary infection with staphylococci almost invariably occurs, and the lesions in the larynx, trachea, and bronchi usually take on a purulent character within four to six days, if the animal lives that long.
8. In the mildest cases the lungs show congestion and edema with hydropic and mucoid degeneration of the epithelium of the bronchioles. In the more severe cases the necrosis of the epithelium extends into the smaller bronchial divisions and bronchioles and a secondary purulent bronchopneumonia, apparently usually due to staphylococci, follows. In these severer pulmonary lesions, atelectatic areas, due to the plugging of the bronchioles with exudate, usually alternate with emphysematous areas. The atelectatic areas are usually edematous or hemorrhagic. Secondary hemorrhage into the purulent bronchopneumonic areas is also frequent, in some cases the condition assuming the picture of a hemorrhagic bronchopneumonia. In the pneumonic areas following gassing, colonies of staphylococci are always present.
9. In the severe cases death may occur more quickly as the result of a diphtheritic or purulent laryngitis without the development of pneumonia, but in the majority of cases the cause of death, so far as the respiratory tract is concerned, is the development of an infective purulent bronchopneumonia secondary to the injury produced by the gassing.
10. Cases with localized diphtheritic necroses in nose, mouth, larynx, and trachea, without extensive bronchopneumonia, may recover with healing and cicatrization of these areas. Fibroblastic proliferation in such areas was noted as early as the fourth day after gassing. Cicatricial contractions of trachea or larynx may result from such healed lesions.


The clinical features of mustard gas lesions of the respiratory tract in man were studied in the cases of seven soldiers who were severely gassed in the manufacture of mustard gas, although wearing gas masks. The concentration of the gas to which these men were exposed was so strong that all received severe burns, two of the cases terminating fatally. In addition to these severe cases, data were secured from 63 other workers with mustard gas who had had at one time or another milder respiratory symptoms following accidental exposure to mustard gas vapor.

CASE l.- Pvt. Ha. (Case 6, p. 549). Exposed one hour in two or three shifts. Within a few hours after the exposure he developed erythema of the skin, nausea, and vomiting and


complained of intense thirst. This patient developed no conjunctivitis, his eyes being sufficiently protected by the gas mask. Later, burns about the lips appeared, and within the next few days diphtheritic necroses developed over the dorsum of tongue, uvula, hard and soft palate, pillars of fauces, tonsils, and pharynx. On the seventh day after the accident, large moist rȃles; were present over both lungs, and he was thought to have a diffuse bronchopneumonia. The patient died on the eleventh day after gassing. The autopsy report of his respiratory tract changes will be found below.

CASE 2.- Pvt. S. (Case 7, p. 549). Exposed three-quarters of an hour, in two or three shifts, to the same concentration as the preceding. During the next few hours, in addition to the development of conjunctivitis, nausea, and vomiting and erythema of the skin, he complained of intense thirst, and later showed patches of diphtheritic necrosis on the lips, angles of mouth, hard and soft palate, uvula, pillars, tonsils, buccal surfaces, dorsum of tongue, and gums. All of the mucous membrane of the mouth showed erythema and desquamation of the epithelium. When seen on the eighth day he could barely whisper and had physical signs of a diffuse bronchopneumonia. He died four weeks after the accident. Autopsy not permitted.

CASE 3.- Pvt. Mc. (Case 1, p. 544). Exposed 40 minutes in four shifts. Within a few hours after the exposure, coincidently with the nausea and vomiting and erythema of the skin, he complained of a very dry throat. For three days he had a hacking cough and complete aphonia. The roof of his mouth was very painful, and on the third day one of the sore places in his throat "seemed to burst" and he coughed up membranelike material, after which his symptoms were improved, but his voice remained husky and a bronchial cough persisted for some time. When examined eight days after the accident he had a catarrhal rhinitis and patches of diphtheritic necrosis over the mucous membrane of the mouth, particularly on the right pillar and in the posterior pharyngeal wall. Occasional loud, moist rȃles were heard over his thorax. Respirations increased.

CASE 4.- Pvt. W. (Case 5, p. 546). Exposed one-half to three-quarters of an hour in two or three shifts. In the first few hours after the exposure he developed mouth and throat symptoms in the form of a dry sore throat and could barely whisper. When examined eight days after the accident he showed diphtheritic necrosis of the mucous membrane of the lips, angles of the mouth, hard and soft palate, uvula, anterior and posterior pillars, tonsils, and posterior pharyngeal wall. The roof of the mouth was covered with a grayish-green diphtheritic membrane through which numerous discrete yellowish-white elevations appeared, suggesting small pustules on erythematous bases. On the eighth day he developed symptoms of a diffuse bronchopneumonia from which he later recovered, after, in the meantime, having expectorated a fibrinous bronchial cast, containing streptococci.

CASE 5.- Pvt. M. (Case 2, p. 544). Exposed one-half hour in one shift. Had severe thirst but no soreness of mouth or throat. He developed a temporary erythema of the mucosa of the hard palate, with a small patch of diphtheritic necrosis. Voice husky, frequent cough, and respiratory rate increased.

CASE 6.- Pvt. E. (Case 4, p. 546). Exposed 10 minutes in one exposure. On the next day complained of a very dry throat. He showed only a temporary erythema of the mucous membrane over the hard palate. When seen eight days later his voice was husky, and he had an occasional dry cough.

CASE 7.- Pvt. Hu. (Case 3, p. 546). Exposed 10 to 12 minutes in one shift. He developed a severe rhinitis and a sore throat, but showed only an erythema of the mucous membrane of the mouth. When seen eight days after the accident his voice was still husky, and he had a laryngeal cough.

A large number of other cases of mild respiratory lesions were also seen. All of these were in cases with conjunctivitis. These patients complained usually of a dry or sore throat, huskiness of voice, with or without an accompanying rhinitis. A dry laryngeal cough usually persisted for several days, a definite huskiness for a longer period. In two cases, short exposures to weak concentrations produced in susceptible individuals a fibrinopurulent rhinitis lasting several days. In the severe cases of mouth burns the most marked lesions were always on the hard and soft palate and the dorsum of the tongue. The mildest cases showed erythema of the same areas.



CASE 1.- Pvt. Ha.


Nose: Examination of anterior nares negative. No skin burns on nose.
Lips: No burns of mucosa of lips.
Mouth and neck organs: Tip of the tongue was negative. On the dorsurn there were patches of grayish and grayish-white coating beneath which the mucous membrane was desquamated. When the coating was removed there was a denuded, reddened surface. These lesions increased in severity toward the root of the organ where they became diphtheritic in character and were covered with a greenish-gray membrane. (Fig. 201.) The tonsils and palatal arch showed areas of diphtheritic necrosis covered with a grayish membrane. Mucosa of the palate was markedly edematous and presented elevated, reddened and grayish areas. The tonsils were about normal in size, with deep crypts. The right was negative on section. The left tonsil showed membranous patches and, on section, crypts containing fibrinous exudate. Mucosa of pharynx was covered with a thick, grayish, tenacious membrane, which, when pulled off, left a denuded, reddened surface. The epiglottis was covered with a thick, grayish mucus and the glottis was filled with a similar mucus. Beneath this the mucosa was congested and showed small patches of superficial necrosis, but these lesions were less marked than in the pharynx. Below the glottis, the mucosa of the trachea was covered with a thick grayish mucus, beneath which the mucosa showed a congestion diminishing in intensity from above downward. There was no membrane in the trachea but in its lower portion and at the bifurcation there was a thick, grayish mucus in the lumen. The posterior wall of the trachea showed marked hypostasis but there was no visible necrosis of its mucosa. The esophagus showed congestion of its mucosa with superficial desquaamation at the mouth but no changes below this .
Lungs: There was no fluid in either pleural cavity. The lungs met in the midline above the heart. Both lungs were free throughout. Pleural surfaces negative.
Left lung: There were delicate old adhesions between the upper and lower lobes. On the anterior surface of the lower lobe there was a small pigmented nodule beneath the pleura, a healed tubercle. There was no thickening or puckering of tie pleura over the apex. Beneath the pleural surface there were numerous minute petechial hemorrhages which were largest and most abundant posteriorly. The lung was air-containing throughout. The upper lobe was voluminous and pink in color; the lower lobe showed marked hypostasis posteriorly. On section, the cut surface showed marked congestion and edema. Foamy fluid, thicker than the ordinary stasis edema, exuded abundantly from the cut surface. The lower lobe presented a marked hypostatic congestion. The small hemorrhages appeared soft and fresh and the blood showed no discoloration. Small fresh agonal lardaceous clots were found in the larger pulmonary vessels, but no thrombi or emboli were visible to the naked eye. No pneumonic areas, no abscesses, no hemorrhagic infarctions and no airless areas were found. Anthracosis was moderate. The main bronchi were filled with a thick, frothy fluid which was rather viscous and was grayish-white in color, but in the smaller branches became distinctly yellowish and thicker. Beneath this the mucosa of the bronchi was markedly congested .
Right lung: The right lung was very much heavier than the left and much deeper in color. There were delicate old adhesions between tipper and middle and middle and lower lobes. The lower lobe posteriorly was almost black in color, the middle lobe deep red, while the upper lobe was more pinkish. The subpleural hemorrhages were more marked in this lung than on the left side. The anterior margins of the lobes showed emphysema. The greater part of the lower lobe showed a partial atelectasis and posteriorly a firm airless area almost black in color, about the size of a hen's egg. On section, there was extreme congestion and edema, with areas of atelectasis throughout the lower lobe. Throughout the entire lung there were numerous firm, elevated, airless areas of small size, that were not wedge-shaped. The large airless area in the lower lobe was wedge-shaped, with its base toward the surface, deep red, almost black in color, and showed in its center a softer, lighter area, apparently around the blood vessel. No thrombi or emboli could be seen with the naked eye. The bronchi showed the same fluid content and changes inl the mucosa as on the left.
Bronchial nodes: The bronchial nodes on both sides were rather small, soft, edematous and but slightly pigmented.


FIG.201.- Mouth and neck organs of fatal human case. Mustard-gas lesions of tongue, pharynx larynx, and trachea. Dorsum of tongue shows diphtheritic eschar. Diphtheritic necrosis of pharynx mucosa of larynx and trachea. Marked edema with diphtheritic necrosis of the arytenoepiglottideau fold.



Mouth: Sections from the dorsum of the tongue showed patches of diphtheritic necrosis and small ulcers from which the membrane had been loosened. The submucosa showed intense congestion, small-celled infiltration, some fibroblastic proliferation and numerous capillary thromboses. Some of the dilated vessels, either capillaries or lymphatics, were filled with coarse threads of fibrin without red blood-cells and with only an occasional leucocyte. The mucous glands showed intense mucoid degeneration with small-celled infiltration and many of the ducts were dilated and filled with mucus. Regeneration of the epithelium from the deepest portions of the folds and mouths of the mucous glands had begun. (Fig. 202.)
Pharynx: Sections of the pharyngeal wall showed intact epithelium covered with a thick mucus. The submucosa showed marked congestion, edema, and polynuclear infiltration, with small areas of hemorrhage. There were numerous dilated capillaries or lvmphatics filled

FIG. 202.- Case 1, Private Ha. Mustard-gas lesion of dorsum of tongue. Base of ulcer from which the diphtheritic membrane has become detached

with coarse threads of fibrin without red blood cells or leucocytes. The mucous glands showed marked mucoid degeneration. The tonsils showed marked congestion and edema. The greater part of the epithelium was intact, but there were localized patches of diphtheritic necrosis. There was an increase of the stroma and many sclerotic vessels.
Larynx: Sections from larynx showed a desquamative catarrhal laryngitis with small areas of diphtheritis. (Fig. 203.) The epithelium was for the greater part necrotic, either entirely gone or represented by a single layer of nuclei. On the surface there was a thick mucofibrinous exudate forming a cast of the entire lumen and for the greater part separated from the necrotic surface. The submucosa showed congestion, edema and slight leucocyte infiltration.
Trachea: The trachea presented the same condition as the larynx, a desquamative necrotic tracheitis with the formation of a thin mucofibrinous membrane over the surface.
Bronchi: The larger bronchi were filled with a thick mucous exudate containing many desquamated cells, some polvnuclear leucocytes and a few red blood cells, with little or no


fibrin. The epithelium of the mucosa was in part intact, showing hydropic or mucoid degeneration, and in part desquamated. The bronchial walls showed intense engorgement of the vessels, well-marked edema and a moderate small-celled infiltration. The mucous glands showed marked mucoid degeneration. The picture was that of a catarrhal desquamative bronchitis. (Fig. 204.)
Lungs: The lungs showed extreme congestion and edema, with numerous minute hemorrhages by diapedesis. Areas of emphysema and atelectasis alternated throughout the lung. The atelectatic areas showed the most marked edema and were usually the areas in which hemorrhages were found. No pneumonic areas were found. The airless area in the lower right lobe was a recent hemorrhagic infarct, the central lighter portion being an artery obturated by a recent mixed thrombus. Other smaller recent hemorrhagic infarets were found beneath the pleura of the right lung, with recent thrombi in the vessels leading to these areas. All the smaller bronchi and bronchioles were dilated and filled with a seromucoid

FIG. 203.- Case I, Private Ha. Section of diphtheritic lesion on vocalcord. Epithelium of mucosa completely destroyed and a mucofibrinous membrane partly detached from the surface. Extreme hyperemia of the vessels. Some small-celled infiltration

fluid containing desquamated cells with some fibrin and occasional red cells. The epithelium of the smaller bronchi was intact for the greater part, but showed marked mucoid or hydropic degeneration. In many bronchioles desquamation of the epithelium had taken place. In no bronchi or bronchioles was there any purulent exudate. There were no old thrombi or emboli, and fat stains of frozen sections showed no fat emboli. Smears of the bronchial exudate and from the cut surface of the lung showed only occasional diplococci.

This case presented a marked degree of rmustard-gas degeneration and necrosis of the epithelium of the mouth cavity, larynx and trachea, decreasing in intensity to the larger bronchi and bronchial subdivisions, without infection or pneumonia. Death, in this case, probably resulted from shock and toxemia from the extensive areas of necrosed skin and secondary infection.



From the clinical cases coming under observation, mustard-gas lesions of the respiratory tract in man were of the same nature as those produced experimentally in animals. In the milder cases there was an injury to the epithelium of the mucosa of the upper respiratory tract, particularly of the anterior nares, hard and soft palate, dorsum of tongue, pharynx, larynx, and upper part of trachea. Rhinitis and laryngeal huskiness and cough, with sore throat and thirst, were the most common symptoms in individuals exposed to light and moderate concentrations. From these conditions recovery was usually prompt, but laryngeal huskiness and irritation might continue for some days or even weeks. No instance of secondary infection was observed in such cases.

FIG. 204.- Case 1, Private Ha. Section of main division of bronchus. Picture of catarrhal bronchitis. Marked mucoid degeneration and vacuolation of the bronchial epithelium. Congestion, small-celled infiltration and edema of the bronchial wall

In the more severe human cases, there was a greater degree of degeneration and necrosis of the epithelium of the respiratory tract, extending into the bronchi, but diminishing from the upper part of the trachea down, the most severe lesions being eschars on the palate, dorsum of tongue, pharynx and larvnx.

Associated with these was a widespread catarrhal tracheitis and bronchitis, with congestion and edema of the lungs. The one case autopsied showed no pneumonia. Under the conditions of warfare more severe exposures to mustard gas may produce diphtheritic lesions of larynx, trachea, and bronchi, terminating in bronchopneumonia.



1. The action of dichlorethylsulphide (mustard gas) upon the epithelium of the respiratory tract is escharotic in character, similar to its action upon the epidermis, cornea, and conjunctival epithelium. Degeneration or complete necrosis of the epithelium, or deeper necrosis extending into the submucosa constitute the primary lesions produced in the respiratory tract by mustard gas in weak or strong concentrations. These lesions are most marked in the anterior nares, mouth, pharynx, larynx, and upper part of trachea, diminishing in intensity toward the lungs. In the more severe cases, however, the bronchial epithelium may also undergo degeneration or necrosis. The subepithelial edema in the upper respiratory tract is not at all comparable in degree to that associated with mustard-gas lesions of the skin and conjunctiva.

FIG. 205.- Case I Private Ha. Section of upper lobe of lung. Congestion and edema. Acute emphysema

2. Following the primary lesion there develops a catarrhal or a diplitheritic rhinitis, localized stomatitis, pharyngitis, laryngitis, tracheitis, or bronchitis. Cicatricial contractions or sclerosis may result from the healing of such lesions.
3. In even the mildest cases there is marked congestion and edema of the lungs, catarrhal bronchitis, and localized atelectasis due to bronchial plugging. In the more severe cases the bronchitis may be diphtheritic in character. Secondary infection may occur, and, following the development of a purulent bronchitis, a purulent bronchopneumonia may result. We have observed no pneumonia resulting directly from the gassing. All of the pneumonias observed, following the gassing, have been the result of secondary infection.




Capsules containing 0.06 to 0.24 c.c. of dichlorethylsulphide were given to rabbits and guinea pigs in olive oil, butter, and lard; similar capsules were given in meat to dogs. Animals were kept fasting, but were given water to drink. They were killed in series to get the complete pathologic picture from beginning to end of the process. For the purpose of brevity protocols are omitted, and the results of the experimental work are condensed as follows:


The symptoms, which appeared, usually in from 1 to 2 hours, were vomiting; irritation of mucous membranes of mouth; profuse salivation; ultimately foul

FIG. 206.- Case 1, Private Ha. Section of lower lobe of lung. More intense congestion. Minute hemorrhages by diapedesis. Area of partial atelectasis, alternating with emphysematous areas. Edema

discharge from nose and mouth; foul diarrhea, at times tarry stools; depression; refusal of food and water. When the animal was allowed to live there was progressive anorexia, and weakness; by the sixth day it was so weak that it lav prostrate. Animals became greatly emaciated, with odor of foul putrefaction. Foul discharge from mouth. Death occurred usually by the twelfth nlay, from 0.03 to 0.06 c.c. doses, within three to five days from the 0.24 c.c. lose, for dogs; sooner for guinea pigs and rabbits.


Delayed rigor mortis in animals not immediately autopsied. Emaciation. No free gas or fluid in peritoneal cavity. No general peritonitis. Liver congested;

e For the gastrointestinal lesions produced by subcutaneous and intravenous injections of mustard gas the reader is referred to general pathology, p.632.


gall-bladder dilated; spleen anemic, small and dry. Heart showed dilatation of right side. Peculiar chicken-fat clots in right ventricle of dogs. Lungs showed varying degrees of congestion; no pneumonia. Kidneys congested. Stomach and duodenum distended with gas and containing greenish or black fluid. Mucous membrane showed varying degrees of necrosis, localized in patches or extending over the entire surface. The necrosed areas appeared greenish-brown or gray, to black, without much inflammatory reaction. In the milder cases the mucosa at times showed marked congestion. Very little hemorrhage in stomach. Necrotic areas were usually thinned, but the animal died before perforation occurred. Early peritonitis, over the thinned areas, occurred. Self-digestion by the stomach juices of the necrotic surface was delayed, from the inhibition of stomach secretions. Clean ulcers were not seen in the early cases, but were found in from 6 to 12 days, the ulcer extending into the muscle coat. The stomach lesions did not show the characteristic edema produced by dichlorethylsulphide in the skin and conjunctiva. A moderate edema was observed onlv in the cardiac end. The nonhemorrhagic character of the gastric lesions was also striking. Throughout the intestines there were patches of necrosis with congestion and slight hemorrhage, and more or less marked catarrh. Mesenteric glands swollen, edematous and congested. In the mouth, nose, pharynx, and esophagus of a certain proportion of the animals marked necrotic lesions, with secondary infection, occurred as the result of local contact with the gas in swallowing or vomiting.


Gastrointestinal mucosa showed varying stages and degrees of necrosis, involving the upper layers of the mucosa or extending through the submucosa into the muscle coats. In no case did the ulcer extend through the muscle. Preserved portions of the mucosa showed dilated glands filled with granular or thready precipitate, or colloid masses, resembling hyaline casts. The epithelium of the deeper portions of the glands often showed a marked hydropic degeneration or liquefaction necrosis. There was usually little or no edema of the submucosa or tissues of the neighboring stomach wall. The eschar might be adherent to the surface or the floor of the ulcer might consist of the muscle coats. These showed polynuclear infiltration, congestion, some fibrinous exudate into the tissues, and occasionally hemorrhage. In the older cases there was often a marked small-celled infiltration of the muscle coats, extending to the subserosa, with fibroblastic proliferation of the latter and beginning peritonitis. The necrotic areas in the intestine were similar but showed a more marked leucocyte infiltration, often extending to the serosa, but without any overlying peritonitis. The necrosis of the mucosa was most marked over the lymphoid tissue, Peyer's patches often being nearly completely necrosed. Numerous putrefactive bacteria were found in the necrotic areas of the mucosa. The entire mucosa was congested and inflamed, showing occasional hemorrhages. The inflamed mucosa was but moderately edematous. There was marked mucoid degeneration of the epithelium. The mesentery was edematous and showed at the insertion areas of inflammation. The mesenteric glands showed sinus catarrh and were congested. (Figs. 207, 208.)



In animals gassed in the gassing chamber and not dying within several days there were found practically always a more or less severe congestion and catarrhal inflammation, which were more severe in the small intestine than in the stomach. In some cases more severe lesions were found in the upper part of the digestive tract (diphtheritic necrosis, catarrhal inflammation, ulceration, etc.), but these could be explained as the result of gas swallowed with the air, saliva, or with food, and must be interpreted as direct local lesions. (Fig. 209.) The generalized catarrhal condition of the small intestine may be explained by the assumption that dichlorethylsulphide, or some substance resulting from it, is excreted by the intestine or through the bile, or by the direct local action of minute quantities of mustard gas taken in with food or saliva and diffused throughout the gastrointestinal contents.

FIG. 207.- Dog. Received four minimis of dichlorethylsulphide on meat. Died five days afterwards. Small mustard-gas eschar in stomach mucosa. Early fibroblastic proliferation


In the severe cases of mustard gassing observed, marked nausea and vomiting, gastrointestinal pain, and diarrhea either preceded or were synchronous with the development of the cutaneous burns. These symptoms were reflex in nature and may be regarded as symptoms of the severe shock from which the patients suffered at that time. The gastrointestinal symptoms seen in several patients after the symptoms of severe shock had disappeared were reflex and dependent upon the laryngotracheal-bronchial irritation (coughing).

In the fatal case the following gross appearances in the gastrointestinal tract were noted.


The stomach contained a small amount of thin, grayish fluid, slightly blood stained, and some milk curds. Throughout the entire mucosa there were numerous small petechial hemorrhages, these being most numerous toward the cardia, where the mucosa was sprinkled with them, many of them being confluent. They all appeared very fresh, the blood not being discolored and having no erosions or ulcerations over them. The gastric mucosa was thin, smooth, and soft, with early post-mortem change. The duodenum contained a thin, grayish bile-stained fluid, the mucosa was congested and covered with a tenacious, thick, grayish mucus. Just below the pylorus there were numerous hemorrhages in the mucosa. The small intestine was distended with gas, and at intervals there were small collections of thin fecal matter slightly bile stained. The mucosa was congested, and there was marked

FIG. 208.- Dog. Received 0.06 c.c. of dichlorethylsulphide in capsule. Died 12 days later. Portion of base of very large eschar of stomach wall, extending nearly to the serosa. Marked leucocyte infiltration

hypostasis in portions of the loops without other changes. The colon contained gas, and some formed fecal masses. The mucosa was negative.

Microscopic findings.-The stomach showed a marked congestion of the vessels of the mucosa, with numerous fresh hemorrhages. Post-mortem digestion of the upper portion of the mucosa. The duodenum showed a post-mortem desquamation of the surface epithelium, slight catarrh of the glands, and congestion and edema of the mucosa. Colon showed at similar condition. Appendix showed evidences of old inflammation. No active process. The mesenteric glands showed a marked sinus catarrh, with many hemophages present in the simuses.

It was hardly probable that the gastrointestinal changes observed in this case were the direct result of the mustard gassing. It is much more likely that they wvere entirely secondary phenomena.



1. Direct application of mustard gas to the mucosa of stomach or intestine by means of contaminated food produced localized degeneration, necrosis, and ulceration similar in type to the lesions of the respiratory tract.
2. The mild catarrhal conditions and small localized eschars of the stomach and upper portion of the small intestine seen in animals gassed in the gassing chamber may be explained as the result of minute quantities of mustard gas swallowed with air or dissolved in the saliva.
3. The gastrointestinal symptoms seen in gassed human beings are probably chiefly reflex, associated either with shock or with the respiratory irritation. As in other forms of gassing, it is probable that erosions or ulcers of

FIG. 209.- Rabbit. Exposed 12 hours to a concentration of 1 to 50,000. Died 92 hours after gassing. Mustard-gas eschar on tongue the stomach and intestine may be embolic in character, the emboli arising in the primary or infected mustard-gas lesions of the skin or elsewhere. It is also possible that in man, as in the case of animals, localized eschars of the gastrointestinal mucosa may be produced by the direct action of mustard gas swallowed in contaminated food or saliva.


From observations made upon a large number of animals exposed to mustard gas, either by direct applications to the skin or eyes, or in the gassing chamber, the systemic symptoms shown consisted of gastrointestinal disturbances


such as vomiting and diarrhea; disturbance of heart rate, usually an increased rate, but in severe cases the rate may be slowed; lowering of the temperature except with infection; decreased urinary elimination; in the blood a secondary anemia, with or without leucocytosis; and nervous symptoms, as increased reflex excitability, tremors, and convulsions, or marked depression, stupor, and coma. The great majority of slightly or moderately gassed animals showed no systemic phenomena. The most constant of the general symptoms were the gastrointestinal. While a certain number of cases showing these symptoms could be explained as resulting from slight quantities of mustard gas swallowed in the air or saliva, or on food particles, it is probable that the gastrointestinal symptoms were chiefly reflex to the respiratory irritation, or were a part of the general phenomena of shock. In all severe dichlorethylsulphide gassing an initial shock was very common, and the changes in temperature, circulation, etc., formed part of the clinical picture of this condition. With the development of pneumonia or localized infection the usual symptoms attending such processes were observable.

Nothing was seen that corroborated Victor Meyer's idea that the toxic action of mustard gas is exerted chiefly through the blood; indeed, all observations made one certain that local applications of mustard gas to the skin of distant parts can not produce conjunctivitis or local lesions at other sites.

In rabbits gassed in the gassing chamber at varying concentrations there occurred within 24 hours following the gassing a distinct concentration of the blood, as shown by a polycythemia varying from one to two millions increase, and an increase in the leucocytes. This concentration might last for several days, after which the number of red blood cells slowly came down to normal. With the advent of secondary infection, which might take place at any time from the second to seventh day after gassing, there was a leucocytosis which might reach 50,000. If the animal lived and showed increasing cachexia as the result of infection, the blood picture of a secondary anemia developed. This, however, had no direct relationship to the toxic action of dichlorethylsulphide. These investigations, therefore, confirmed the previous observations made in this laboratory that cutaneous, ocular or respiratory gassing with dichlorethylsulphide had no direct action upon the blood or bone marrow.

Four selected protocols illustrating these points are given:

RABBIT 90.-Gassed six hours at a concentration of 1 to 20,000



RABBIT 91.- Gassed Six hours at a concentration of 1 to 20,000:

RABBIT 92.- Gassed one hour at a concentration of 1 to 30,000:



RABBIT 93.-Gassed one hour at 1 to 30,000:


Likewise, the pathologic findings in gassed animals gave no convincing evi'dence of the absorption of mustard gas into the blood from the skin, eyes, or respiratory tract. The general pathology was as follows:

Central nervous system.-Congestion; very rarely minute hemorrhages; ill infected cases emboli were seen rarely.
Thyroid.-Congestion; no other changes, except as involved directly fronm the laryngeal lesions. This event was common enough in the severe cases of laryngeal and tracheal diphtheritis, when the thyroid was congested, edematous and infiltrated with leucocytes.
Heart.-Ventricles usually dilated. No changes observable in heart muscle.
Aorta.-Intima of great vessels occasionally showed fatty streaks.
Lung.-When not involved directly through the respiratory tract, as in local application to the skin, the lung showed congestion, with or without edema. In the case of infected skin or eye lesions pulmonary emboli and metastatic pneumonia may occur.
Spleen.-Congested. In some cases an increase in the number of pigmented phagocytes. Adrenals.-Congestion. Occasionally minute hemorrhages in medulla. (These were probably post-mortem.)
Kidneys.-Congestion. Frequently slight cloudy swelling. In infected cases more marked cloudy swelling, casts.
Liver.-Marked congestion. In the majority of cases liver cells were normal; slight cloudy swelling and fatty changes were not rare, but their occurrence did not suggest any direct relationship with the gassing.
Gall-bladder.-A thinner, watery bile was usually present.
Gastrointestinal tract.-Congestion and mucous catarrh were always present. Minute hemorrhages and erosions were frequently seen in stomach


(vomiting?).Very rarely embolic hemorrhages and erosions occurred, especially in animals with large esehars, or infected lesions. At times local eschars of the gastric mucosa occurred; these were explainable as due to ingestion of contaminated food or saliva. The marked salivation seen in many animals and the swallowing of quantities of such saliva containing small amounts of mustard gas, offered a plausible explanation for the catarrhal conditions found in the gastrointestinal tract, without assuming, as some writers have done, that they were due to an excretion of mustard gas through the gastrointestinal mucosa. The intense splanchnic congestion usually present might also serve to explain the gastrointestinal conditions.
Mesenteric lymph nodes.-Congestion and edema.
Genital organs.-No changes noted.
Urinary bladder.-Negative.


The clinical pathology of a series of thirty cases gassed by exposure to mustard-gas vapors was studied. f Nine of these cases were very severe, two resulting fatally. The others presented lesser grades of severity from a mild conjunctivitis only to severe skin burns. Unfortunately the study of the severe cases could not be begun until the tenth day after gassing, and observations as to the clinical pathology of these cases must be correspondingly modified. The results of this study are summed up in the following:

1. Mild cases of mustard-gas burns of the skin showed no changes in the blood or urine.
2. Moderately severe and severe cases of mustard-gas burns of the skin with some involvement of the upper respiratory tract showed after the first week definite changes in urine, blood urea, and blood.
3. The urinary changes consisted in a diminution of the urinary output, increased concentration and acidity, albuminuria, and diminished urea and chloride output. In the sediment there might be found casts, renal epithelium, red blood cells, and an increased number of leucocytes. Under forced fluids prompt improvement occurred.
4. Coincident with these urinary changes the blood urea was found to be high, but approached normal with the improvement in the urinary condition when fluids were forced.
5. The blood showed a slight secondary anemia, with a well-marked polymorphonuclear leucocytosis, a definite eosinophilia, and the appearance of mvelocytes and young forms of leucocytes.g The blood platelets were usually

f By Dr. George R. Hermann.
g Krumbhaar (Role of the Blood and the Bone Marrow in Certain Forms of Gas Poisoning. Journal of the American Medical Association, 1919, lxxii, No.1,139) discussing the hematologic examinations made in a series of patients gassed with mustard gas, concluded that there is an initial leucocytosis followed by a more or less extreme degree of leucopenia, which persists even in the presence of bronchopneumonia, and is very probably an important contributing factor to the high mortalityofsevere cases. Should the leucocyte count fall below 1,000 per c. mm. a "myelocyte crisis" may bring about a partial amelioration; but in the two cases of this kind observed, this did not serve to protect them from a fatal outcome. Lessened blood formation is also indicated by the production of anemia without blast-cell formation and diminution in the number of blood platelets. In the earlier stages the coagulation time is decreased, and in the later stages of severe cases it is increased.
In the observations recorded in this chapter experience with human cases of mustard-gas poisoning, even the fatal cases, and with experimental animal gassing gave results in contradiction to those obtained by Krumbhaar. Leucopenia was never observed; on the contrary, there was a definite leucocytosis in all severe and moderately severe cases, with a mild econdary anemia, and an increase in the blood platelets instead of a decrease. There was also very frequently an increase in the eosinophiles.


increased. No evidence of hemolysis was found. These changes indicated a disturbance in the white cell formation rather than in the red blood cell group. No leucopenia was noted at any time. The leucocytosis reached its height coincidentally with the height of the secondary infection and fell with the improvement of the infection.
6. The temperature, pulse, and respiration charts showed in the severe cases an initial period of shock. With the development of the necrosis and the secondary infection there was a corresponding febrile reaction.
7. The bacteriologic examination of the infected skin lesions and furuncles showed constantly the presence of staphylococcus pyogenes aureus. In the one bronchial cast obtained streptococci were present.
8. We believe that the changes in the blood and urine may be interpreted as dependent upon the secondary infection and, in part, possibly, to the absorption of toxic products from the necrotic skin, rather than to any direct toxic action of mustard gas.


No evidences of a toxic action upon the internal organs were seen in any of the lesions of mustard gas produced by direct application or by exposure to its vapor. Clinical studies, as far as carried out, showed increased blood urea, decreased urea and chloride output, decreased quantitative urinary output, and increased specific gravity, with increased acidity. The toxic symptoms observed could be explained as the result of the skin destruction, damage to the respiratory tract, secondary infection, decubitus, with resulting absorption, etc. In the earlier stages symptoms of severe shock might be present.


After healing of the skin lesions a severe itching might persist for many weeks or months. This was true, even of the mildest lesions; accompanying the itching the affected part might show a branny desquaination, or secondary vesicles might develop after rubbing (Nikolsky's sign).


The protocol of the gross and microscopic general pathology of the one human autopsy case of fatal poisoning from mustard gas is given in full, with the exception of the respiratory and gastrointestinal tracts, which have already been quoted (pp. 611, 620)


(Autopsy by Dr. A. S. Warthin, 2 p. in., July 8, 1918.)

Pvt. Ha. Died, 12.30 a. m., July 8, 1918.

External examination
.-Young adult male body; length 170 cm.; strong build, frame large, thorax deep, epigastric angle a right angle, abdomen on level with ribs. Musculature good, well developed. Panniculus abundant, particularly over abdomen and thighs. Hips rounded, slightly of feminine type. Neck thick, thyroid small, facies slightly suggestive of mouth breathing; general appearances of body suggest the lymphatic constitution. All regional lymph nodes prominent. Hair of head abundant, dull brownish in color, dry; scalp shows much dandruff. Face shaven, beard not heavy, body hair fairly abundant; pubic hair of feminine type. External genitals small, scrotum small, tight; testicles small; penis medium size, moderate phimosis. No anomalies. No deformities, defects, or mutilations. No surgical wounds or scars.
Hypostasis.- Moderate post-mortem hypostasis, pale in color, except over hyperemic areas. Superficial veins on upper portion of body contain but little blood.


Rigor mortis- Marked rigor mortis throughout body, except in right arm, where it has been broken by manipulation of the arm.
Body heat.-Body is cold. (Cooled by undertaker.)
Odor.- Marked odor of gangrene over the skin, particularly from that of back.
Skin.-Marked pigmentation of skin of face, hands, and forearms. Right hand shows irregular patches of pigmentation alternating with paler recent scars of older (mustard gas) burns. Hands show irregular patches, particularly between the fingers, of desquamation of the horny layer. No recent burns on hands .

Over the greater part of the body, from the collar line down to the boot line, the skin presents the appearance of a chemical burn, varying in degree from hyperemia, slight desquamation, dried blebs and bullae, denuded areas, to areas of well-marked necrosis and eschar formation. These lesions are most severe in the axillary regions, inner aspect of arms, bends of elbows, flanks, genitals, thighs, and back. Severe lesions completely encircle both thighs, and are particularly marked posteriorly where there is extreme necrosis, secondary infection, and pus formation. The necrosis is very marked over the entire back, neck, shoulders, and buttocks, where the skini is discolored, mottled red, grayish, vellowish-grav, brown, black, or greenish, with marked gangrenous odor. Deeper points of hyperemia corresponding to the hair follicles show in these discolored areas. Over a large part of these lesions the epidermis is desquamated in large bullae, and the desquamated surface is brown, yellow, or red. There are no large hemorrhages into these lesions, and very few petechial hemorrhages, the reddened areas being hyperemic only, the redness disappearing on pressure. From the discolored moist areas a cloudy foul-smelling serum exudes. This is not blood stained in any region. Through the hyperemic areas in many regions the hair follicles appear as opaque, yellowish miliary nodules. These are particularly marked over the scrotum.

The skin of the abdomen is hyperemic, with a pale belted line corresponding to the protection afforded by belt worn by patient. This belted area of less injury disappears in the flanks and shows but slightly on the back. The anterior surfaces of the thighs are bright scarlet; marked hyperemia. The hyperemia stops rather abruptly over the legs about 15 cm. below the lower border of the patella. The lower portions of the legs show practically no involvement anteriorly; posteriorly there is slight discoloration and hyperemnia (hypostasis?). The feet show no lesions.

The face shows no lesions except on the right side, around and below the right ear, where there is a slight desquamation of the epidermis, with some reddening of the edges and elevated portions of the lobe of the ear. Below the left ear there is slight desquamation without reddening. Over the neck, below the chin, are patches of dried vesicles, brownish red where the epidermis is desquamated, and red where it is still intact. The hair shows no changes. On the scalp the only lesion found is an area of hyperemiia above the right mastoid prominence. Here the scalp is scaly with a large amount of dandruff.

It is notable that the skin over the bony prominences, particularly over clavicles, scapulae, etc., shows circumscribed patches of deeper eschar formation.

The areolae of the nipples show strikingly the greater intensity of the injury around the hair follicles and sebaceous glands. Each nipple is surrounded by a deep zone of hyperemia in which the epidermis has beeln lost; through the hyperemic skin the yellowish, enlarged hair follicles can be seen.

The skin of penis and scrotum is discolored, necrotic, and gives a foul gangrenous odor. There is a well-marked phimosis, but the small portion of the glans exposed, around the meatus, is necrotic, yellowish, and inifiltrated with pus. A drop of pus exudes from the meatus. The remaining portion of the mucosa of the glans shows a marked inflammatory reaction (hyperemia, infiltration, and edema). Through the discolored skin of the scrotum the yellowish necrotic enlarged hair follicles and sebaceous glands stand out prominently. Around the anus there is a zone of marked gangrene and suppuration of the skin.

The areas of skin (face, hands, legs, and feet) not affected by the lesions show marked anemia.

Mucous membranes
.- These are pale, particularly the lips. They are dry but show no burns. Tip of the tongue shows no lesions. (See infra.)
Eyes.- The cornea and conjunctivae of both eyes are clear; there is no hyperemia, no discharge, no opacity. No signs at all of conjunctival involvement.
Nose.- Nasal openings negative.
Ears.- Auditory passages negative.


Edema.- Trace only about ankles; none in eyelids or below eyes. Over the burned areas the skin is swollen, tumefied, infiltrated, consistency increased, but pits only slightly on pressure.
Main section.- On section the panniculus is abundant, pale yellow, moist shining, and only slightly edematous. The superficial veins are markedly anemic. No free gas in peritoneal cavity. Abdominal recti deep red in color, with lighter patches of hyaline or Zenker's necrosis (as in typhoid). Omentum lies a hand breadth below the umbilicus, is very rich in fat; its vessels are empty.

There is no free fluid in the cavity. The peritoneal surface is clear, shining, rather dry. but shows no signs of inflammation. Upper coils of small intestine are distended with gas. Cecum is greatly distended, also transverse colon, and descending colon as far as the sigmoid.
Hepatic and splenic flexures collapsed.
Stomach in normal position, contains a small amount of fluid and gas. Vessels of stomach enormously congested. Pylorus is in normal position. Old adhesions around the appendix, but no active process. The appendix is 10 cm. long, patent throughout, and empty. Lower border of liver in median line is three finger breadths below the ensiform, and about one finger breadth below costal margin in the right nipple line. The spleen is in normal position, the lower pole about two fingers above the costal margin.
The diaphragm on the left is in the fifth intercostal space, and at the fifth rib on the right.
The thoracic muscles are deep red, very dry, and anemic.
The costal cartilages are white, and cut easily.
No free gas or fluid in the pleural cavities.

The sternum shows normal consistence, and its red marrow is in normal amount and appearance. The anterior mediastinal fat is very abundant and shows numerous petechial hemorrhages. The fat is very abundant over the pericardial sac, and shows numerous petechial hemorrhages.
The thymic fat is abundant and contains numerous petechial hemorrhages. In it, the thymus is still present in the form of two distinct lobes, each of which is 5 cm. long and 3 to 5 mm. thick. Its color is pink, consistence firm; no hemorrhages in its substance.

The lungs meet in the middle line above the heart. The apex of the heart is in the fifth intercostal space inside the left nipple line.
  Both lungs are free throughout. There is no fluid in either pleural cavity.

.- No gas in pericardial sac; fluid normal in amount, clear amber. Pericardial lining smooth, clear, moist, shining.
Heart.- Heart is small, smaller than cadaver's right fist; in marked rigor mortis, the left ventricle completely contracted, and the right one nearly so. The auricles and vena cavae are collapsed, nearly empty; on section, contain but a small amount of dark fluid blood. The subepicardial fat is in excess; over the pulmonary artery, just above the conus, is a sclerotic patch ("soldier's spot") about the size of a dime; along the coronary branches. posteriorly over the left ventricle, are narrow stripes of sclerosis. There are numerous subepicardial hemorrhages, most marked in the neighborhood of the auriculoventricular groove. On section the heart contains a very little fluid blood and small thin lardaceous agonal clots, particularly in the right ventricle, extending into the pulmonary artery.
The mitral opening admits two fingers; the flaps are negative. In the left auricle a small agonal lardaceous clot. The left ventricle wall measures 20 mm. in thickness, the muscle is deep brownish-red, firm, and shows no cloudiness, fibroid patches, or fatty change. No fatty change seen beneath the endocardium.
The tricuspid opening admits three fingers barely. Its flaps are normal. Small agonal clot in right auricle. The foramen ovale is patent, an oval slit admitting probe 3 mm. in diameter, but well guarded by membranous curtain. The pulmonary artery admits two fingers, the pulmonary semilunar valves are negative. The right ventricle wall measures 3 to 8 mm. in thickness, about one-half of this being fat tissue. The aortic opening admits the thumb easily. Its semilunar flaps negative. The beginning of the aorta shows very fine stripes and patches of fatty degeneration of the intima. Arch of aorta negative; also its main branches. Pulmonary artery and main branches empty and negative.

(For protocol of lungs, see p. 611.)
Examination of thoracic duct negative.


Thoracic aorta shows linear stripes of fatty degeneration.

(For protocol of mouth and neck organs, see p. 611.)
The cervical esophagus shows some injection of its mucosa at its mouth, with superficial desquamation; below this no changes except post-mortem hypostasis. A small amount of stomach contents in the lower portion.
The thyroid is small, but on section shows normal amount of colloid.
Parathyroids, four in number, brownish in color, about normal size.
Cervical nodes are about normal in size, some larger, translucent, slightly edematous.
Parotid and submaxillary glands are normal in size and appearances.
Great vessels and nerves of neck region are negative.

Abdominal organs
.-The spleen is somewhat enlarged, soft, flattens on the board, its capsule slightly wrinkled. On its surface are the remains of a few old adhesions to the diaphragm. On section the pulp is soft, deep brownish-red, rises above the trabeculae, the cut surface slightly shagreened. Bleeds rather poorly. Follicles barely visible. There are no infarets, abscesses, or other lesions apparent.
The adrenals are surrounded by abundant fat tissue. Both are normal in size, the cortex pale yellow and somewhat fatty, the medullary portions showing no post-mortem changes. No pathologic changes apparent.
The kidneys have large fatty capsules. Fibrous capsules strip very easily. Both kidneys somewhat enlarged, plump, and markedly congested. Surfaces smooth, veins congested. On section both kidneys bleed freely. The cut surface shows extreme congestion; the outlines between labyrinths and medullary rays are not distinct. The surface is slightly cloudy (slight or moderate cloudy swelling). The pelvis of the left kidney is negative; that of right is dilated, filled with cloudy urine, and its mucosa slightly injected.

The left ureter is normal in size and appearance; the right is dilated with cloudy urine.
The urinary bladder is moderately distended, containing about 60 to 70 c.c. of turbid urine. Mucosa is pale, negative.
The prostate is of normal size and shape; on section normal in appearance.
The seminal vesicles contain a small amount of thin, brownish semen on the right side; on the left they are nearly empty, and the walls show some thickening.
The testes show congested tunics. No fluid in sacs. On section both organs are rather small, slightly congested, and slightly edematous. The left testis shows slight fibrosis extending from the rete.

(For protocol on gastrointestinal lesions see p. 620.)
The liver is much enlarged, particularly the right lobe; but the lower border is rather sharp. Capsule negative. Surface smooth. Color deep brownish red, with paler anemic areas. On section bleeds very freely. Consistency fairly firm.  Cut surface uniform deep reddish-brown, the lobules enlarged, with congested central areas and slightly cloudy parenchyma (slight cloudy swelling). Anemic areas show slight fatty shine.
The bile passages are patent. Gall-bladder small; contains a moderate amount of thick brownish bile; no concretions.
The portal vein, common duct, and inferior vena cava negative on section.
The entire splanchnic area is congested.
The retroperitoneal and mesenteric lymph nodes are enlarged, markedly congested, and somewhat edematous. The prevertebral hemolymph nodes are hyperplastic, deep red, and very numerous.
The thoracic and abdominal aorta small in size (hypoplastic), empty; the intima shows linear stripes of fatty degeneration, most marked in the neighborhood of the celiac axis.
Semilunar ganglia and solar plexus negative.

Nervous system
.-The scalp is negative except for changes mentioned above. Periosteum of cranium negative. In the right parietal bone a circular area of slight hyperostosis resulting from a cephaihematoma at birth. Corresponding to this area of elevation there is on the inner surface of the skull cap a marked pacchionian depression almost perforating the greatly thinned outer table. The skull cap is thin; dura adherent all over; the meningeal grooves and pacchionian depressions unusually marked for age of cadaver; the lamina vitrea rough and dull; the dura is thickened throughout. The arachnoid shows numerous focal thickenings; the subarachnoid fluid is markedly increased (edema); no evidences of inflammation. Central longitudinal sinus is negative. The basal meninges are slightly thickened and unusually tough.
The basal vessels are negative.
The entire leptomeninges are somewhat thickened, but strip easily.


The cerebral convolutions are rather sharp, but show no focal lesions. The ventricles are somewhat dilated; the cerebrospinal fluid is increased greatly, but is clear. The ependyma is normal.
The choroid plexus is markedly congested on both sides. The pineal gland is normal in size and appearance. On section the cerebrum shows congestion and edema. No hemorrhages found. The cerebellum shows marked congestion and edema. Dentate nuclei are normal. No lesions in cerebellar lobes. The basal ganglia, internal and external capsules, are negative.
The pons and medulla are congested and edematous. No lesions apparent.
The cervical cord shows slight congestion and edema. No lesions apparent.
The hypophysis is of normal size, congested.
Basal sinuses congested. Base of skull negative.


Mustard-gas burns of skin and upper respiratory tract; gangrene and secondary infection of skin; shock; toxemia; edema, congestion, and early hemorrhagic bronchopneumonia. Passive congestion of all organs. Splanchnic congestion marked. Moderate parenchymatous degeneration of kidneys and liver (toxic); multiple petechial hemorrhages; congestion and edema of brain and cord; thymicolymphatic constitution.


Cerebrum.- Sections taken from all parts of the cerebral cortex, basal ganglia, floor of ventricles, internal capsule, show intense congestion, edema and a few minute perivascular hemorrhages.
Pons, medulla, and cerebellum.- Show a similar congestion and edema.
Meninges.- Old thickenings. No active process. Marked congestion of the vessels and edema. In the larger pial veins there are agonal clots rich in leucocytes.
Hypophysis.- Intense congestion; otherwise negative.
Pineal gland.- Large calcareous concretion. Gland substance smaller in amount than normal. Marked edema.
Spinal cord. -Edema. Congestion. Post-mortem myelinosis.
Heart.- Subepicardial fat abundant. Vessels congested. Numerous petechial hemorrhages throughout the fat. Heart muscle fibers somewhat smaller than normal-acute simple atrophy. Stroma somewhat increased around the coronaries in the left ventricle wall. Endocardium is negative. Heart muscle shows very slight parenchymatous degeneration. No fat. Mixed agonal clots taken from the heart show white clot with very few leucocytes, while in the red portions of the clot there are large collections of leucocytes.
Aorta.- Stripes of marked fatty degeneration in the intima. These are longitudinal and stain rather brownish-red with sudan III and also a brownish-red with scarlet-red. Osmic acid not successful. (For respiratory tract and neck organs, see p. 611.)
Thyroid.- Vessels congested. Otherwise negative.
Cervical lymph nodes.- Show marked sinus catarrh with numerous hemophages in the sinuses. The cervical hemolymph nodes show marked congestion and hemolysis, many hemophages in the sinuses.
Thymus.- Remains of thymus scattered throughout the thymic fat with large corpuscles of Hassall, many of which are calcified.
Spleen.- Marked acute passive congestion. Small hemorrhages scattered throughout the pulp. Acute lymphoid exhaustion. Follicles very small with central exhaustion. No degeneration of the central portion of the follicles seen as in burns of the skin. No thrombi, no emboli or infarcts. Adrenals.-Slight cloudy swelling of the cortex. No increase in fat content. Congestion. Edema. Chromaffinic substance small in amount.
Kidneys.- Intense congestion. Cloudy swelling, particularly of the convoluted tubules. Numerous small precipitates of phosphates in the straight tubules of the cortex (concentrated urine). Marked edema of the connective tissue around the larger vessels and of the medullary portion. Very few casts. Frozen sections show no fat.
Bladder.- Congestion and edema. Otherwise negative.


    Prostate.- Congestion. Gland spaces filled with secretion and desquamated epithelium. No pathologic changes. The prostatic plexus contains an organizing thrombus.
Seminal vesicles.- Negative. The one apparently hyaline shows only collapse of the wall; empty, but no fibrosis.
Posterior urethra.- Negative.
Testes.- Very little normal spermatogenesis. Many atypical division figures and desquamation of spermatids. Stroma is increased throughout, basement membrane thickened. Appearances suggest an old mumps orchitis. Intense congestion of the vessels.
Epididymis.- Negative.
Penis.- Foreskin edematous. Vessels congested, with areas of complete necrosis of the epithelium of the subepithelial tissues. Where the epidermis is intact, the horny layer is thickened and the epithelium beneath shows evidences of regeneration. The glans shows likewise areas of necrotic epithelium with areas of regeneration and other patches in which the horny layer is greatly thickened and adherent. The meatus shows a slight inflammation. The body of the penis shows areas of regenerating epithelium covered with a dense desquamated horny layer. Large areas of the surface are denuded, the epithelium necrotic, the necrosis extending into the subepithelial tissues. Marked congestion and edema throughout the organ and a mild diffuse inflammation.
.-  Acute passive congestion. Slight cloudy swelling. Fat stains show no fatty degeneration but a few scattered cells containing large droplets of fat.
Pancreas.- Congestion. Early post-mortem softening. No other pathologic changes. Islands numerous, many of them very large.
Stomach.- Post-mortem digestion of upper part of mucosa. Congestion. Hemorrhages.
Duodenum.- Post-mortem necrosis of mucosa. Marked mucoid change in glands of Brunner.
Small intestine
.- Post-mortem desquamation of epithelium. Congestion. Edema. Slight catarrh. Colon.-Similar changes.
Appendix.- Evidences of old inflammation. No active process.
Mesenteric glands.- Lymphoid hyperplasia, with exhaustion of the germ centers. Marked sinus catarrh. Many hemophages.
Retroperitoneal hemolymph nodes.- Marked congestion of sinuses. Great numbers of hemophages.
Skin.- The skin taken from various portions of the body shows complete necrosis of the epidermis and greater part of the corium reaching as far as the sweat glands over the greater part of the surface, but in many instances reaching the subcutaneous tissues. Many colonies of bacteria are found upon the necrotic corium but there is very little leucocytic infiltration in the necrotic corium or at the border between the living and dead tissues. Only in scattered areas where there is evidently a localized secondary infection are there any notable collections of polynuclears. The vessels in the lower portion of the corium show intense congestion, stasis. In the lymphatics there is a heavy albuminous precipitate with coarse fibrin threads. The lower border of the corium and upper portion of the subcutaneous fat shows usually a well-marked edema. No old thrombi are found and very few hemorrhages, only here and there have a few red blood cells escaped from the vessels. More recent clots with heavy threads of fibrin are found in many of the distended vessels. In these clots many hemolyzed red blood cells are seen. Some hyaline clots are found in the areas of marked necrosis and secondary infection, but these may be secondary to the latter process. Fibroblastic proliferation has begun at the lower border of the necrotic corium and is particularly marked around many of the blood vessels and sweat glands. Regeneration of epithelium from the hair follicles and sweat glands has begun in many areas and in some instances extends to the surface. Separation of the eschar has begun in some areas. Every degree of change is shown from areas where the necrotic epithelium is still adherent, to those areas where the slough is beginning to separate, and to areas of beginning regeneration. A striking feature of the


process is that the sebaceous glands are everywhere completely destroyed, but islands of regenerating squamous epithelium mark the site of these glands. The sweat glands also show marked degeneration, the majority of the acini being necrotic or hydropic. Regeneration of epithelium occurs almost entirely from the hair follicles and ducts of the sweat glands. Around the nipple, in the axillary region, and over the scrotum the necrosis is deeper and more marked and the large sweat glands of the axilla show marked degeneration. In the scrotum the necrosis extends deep into the dartos, involving the superficial bundles of involuntary muscle. In the scrotum, also, regeneration of epithelium proceeds from the hair follicles.


Mustard-gas burns of skin and upper respiratory tract. Necrosis, secondary infection' and gangrene of skin. Acute necrotic pharyngitis. Acute catarrhal laryngitis, tracheitis, and bronchitis. Congestion and edema of the lungs. Acute passive congestion of all organs. Multiple petechial hemorrhages. Marked fatty degeneration of intima of aorta. Parenchymatous degeneration of kidneys and liver. Splanchnic congestion. Shock. Toxemia. Secondary anemia. Hypoplasia of heart and aorta. Thymicolymphatic constitution. Hyperplasia of hemolymph nodes with excessive hemolysis. Organizing thrombus in prostatic plexus. Old orchitis (mumps?).


The pathologic findings in this autopsy case give no evidence of a systemic action of dichlorethylsulphide. All of the changes seen can be explained as due to the direct local action of the mustard-gas vapor or as secondary to the shock and secondary infection of the lesions.


Series of animals, rabbits and dogs, were given varying amounts of dichlorethylsulphide by subcutaneous and intravenous injection. Two samples of very pure dichlorethylsulphide were used, one furnished by the chemical laboratory of the University of Michigan, and the other by the Chemical Warfare Service. For the subcutaneous injections in rabbits, a site on the right side of the back was constantly employed. The hair was clipped or shaved over an area 6 cm. in diameter and the liquid dichlorethylsulphide was injected into the subcutaneous fascia above the spinal muscles. In dogs, also, the injection was made beneath the skin of the back. For the intravenous injections in rabbits, the superficial femoral and jugular veins were used. The pure substance was used, instead of oily solutions, because of the great danger of fatty embolism with the latter. In most cases, the animals were allowed to die, but some were killed in order to avoid post-mortem change in the tissues. In all cases autopsies were done, the tissues were fixed in formol, and sections of all organs were stained in hematoxylin and eosin and other routine stains.


The following condensed protocols are selected as representative of the reactions following injection of varying amounts of dichlorethylsulphide. Rectal temperatures, daily weights and character of feces are given, and these. taken together, serve to indicate the severity of the reaction.



RABBIT 55.- Subcutaneous injection of 0.015 c.c. of dichlorethylsulphide. Very slight general reaction, with apparently complete recovery, except for the local lesion, 26 days after injection

RABBIT 48.- Subcutaneous injection of 0.03 c.c. of dichlorethylsulphide. Slight general reaction. Living, 37 days after injection, with apparent complete recovery, except for the local lesion.



RABBIT 47.- Subcutaneous injection of 0.03 c. c. of dichlorethylsulphide. Moderately severe reaction. Death on the twelfth day


Autopsy.-Autopsy at 1.45 p. m. Body warm. No rigor. Pleural cavities and pleurw negative. Heart contracted. The right auricle still pulsates when stimulated. The lungs show a moderate congestion, with hypostatic areas posteriorly in the lower lobes. The liver is congested; otherwise negative. The gall-bladder is not distended. The spleen shows a moderate congestion, likewise the kidneys, in which there are also some edema and slight cloudy swelling. The stomach is nearly empty and the mucosa shows no changes. The upper small intestines are slightly edematous. Formed stools are found in the colon. No evidence of diarrhea. At the root of the mesentery there is a lobulated mass of edematotIs hyperplastic lymph nodes.

Microscopic findings
.-Lungs: Slight congestion. No edema. Heart: Negative. Spleen: Congestion. In the sinuses are great numbers of pigmented hemophages, the pigment varying in color from brown to almost black. Hemosiderosis. Kidneys: Moderate congestion. Very slight cloudy swelling. Practically negative. Adrenals: Negative. Stomach: Negative. Small intestine: Congestion and edema of mucous membrane with necrosis of the superficial epithelium of the tops of the folds and villi. Glands of Lieberkuhn show excessive mucus formation and individual glands show necrosis. Minute erosions occur along the mucosa. Liver shows marked passive congestion; a nutmeg liver with acute central necrosis in many lobules. The bile-ducts show an unusual mucous degeneration of the epithelium, with edema of the surrounding connective tissue. There is a complete necrosis of the epithelium of the gall-bladder and of its basement membrane. The neighboring liver tissue also shows a zone of necrosis as from the diffusion of some necrosing substance from the gall-bladder. At the site of injection there is a large area of necrosis extending through the striped muscle into the fascia. About this there is a zone of fibroblastic proliferation and leucocyte infiltration. The striped muscle in the neighborhood shows a marked Zenker's necrosis.

RABBIT 50.- Subcutaneous injection of 0.06 c.c. of dichlorethylsulphide. Moderately severe reaction. Apparent recovery, except for local lesion. Animal killed on the thirty-second day after injection.



Autopsy.-Pleurae, pleural cavities, heart, and lungs are negative. The stomach is distended with food; its mucosa appears negative. The gall-bladder is small; the bile, pale. Liver: Congestion. Spleen: Congestion. Kidneys: Slight cloudy swelling and congestion. The upper small intestines contain a small quantity of thin fluid material. There are no formed stools in the colon, but no evidence of diarrhea. The intestinal mucosa shows no lesions to the naked eye.

Microscopic findings
.-Heart: Negative. Lungs: Congestion without edema. Small patches of bronchopneumonia. Spleen: Marked congestion with a fairly large number of pigmented phagocytes. Kidneys and adrenals: Negative. Stomach: Negative. Mucosa well preserved. Liver: Congestion. Slight cloudy swelling. Bile-ducts negative. Gallbladder: Epithelium well preserved. No necrosis. Intestine: Marked catarrhal enteritis. Epithelium well preserved. No desquamation. Lumen nearly empty. Pancreas: Congestion. Mesenteric lymph nodes. Extreme edema .

RABBIT 63.- Subcutaneous injection of 0.08 c.c. of dichlorethylsulphide. Severe reaction. Animal killed on the fifteenth day.



Autopsy.-Pleural cavities, pleurtc, heart and lungs negative, except for numerous firm, slightly caseous areas at root of right lung, probably an old tuberculosis. Peritoneal cavity contains about 50 c.c. of clear fluid. Liver: Congestion. Gall-bladder: Moderately distended with thin bile. Stomach: Moderately filled. No visible lesions of the mucosa. Intestines: Peristalsis very active. Thin mucoid fluid in upper small intestine. In the cecal pouch there is much thick gruel-like fecal material with denser oval masses adherent to the mucosa. Formed stools in color. Kidneys: Congestion and slight cloudy swelling. Adrenals: Congestion. Mesenteric nodes: Somewhat hyperplastic.
Microscopic findings.-Heart: Negative. Lungs: Pulmonary abscesses. Congestion. Spleen: Marked congestion. No pigmented phagocytes. Kidneys: Congestion, otherwise negative. Adrenals: Excess of fat. Marked lipoidosis of fascicular and reticular zones. Stomach: Negative. Intestines: Epithelium well preserved for the greater part. No mucoid degeneration except in one portion, where the mucous glands are distended and filled with hyaline cast-like masses. Many of the glands are shallow mucous cysts. Liver: Cells small, simple atrophy. Gallbladder: Epithelium perfectly preserved. No necrosis. Papillae less marked than normal. Submucosa edematous. Site of injection: Large area of necrotic muscle. Deposit of lime salts. Bacterial infection.

RABBIT 65.- Subcutaneous injection of 0.12 c.c. of dichlorethylsulphide. Very severe reaction. Death on the fifth day.


Autopsy.-Body cold. Marked rigor. Entire posterior portion of body is soaked with fluid feces. At the site of injection there is a small brownish area. On incising this area the subcutaneous fascia is found to be a bright sulphur yellow in color. There is very little local edema. Pleural cavities and pleurae are negative. The heart shows a slight dilatation. There is a moderate congestion of the lungs most marked posteriorly. The peritoneum is negative and there is no fluid in the abdominal cavity. The liver is firm, deeply fissured, and shows a marked congestion. The gallbladder is nearly empty, collapsed, its wall wrinkled. It contains but a few drops of a thin yellowish mucoid fluid. The stomach is moderately distended. The stomach mucosa appears negative. The spleen and kidneys show congestion; otherwise they are negative. Adrenals: Negative. The intestines contain only fluid fecal material. No lesions of the intestinal tract can be seen. All splanchnic veins show a marked congestion.

Microscopic findings.-Heart: Negative. Large white clot in right ventricle. Lungs: Intense congestion and edema. Patches of atelectasis. One of the pulmonary arteries contains a large laminated thrombus or embolus, rich in leucocytes. Spleen: Intense congestion. In the sinuses there are great numbers of phagocytes containing hemosiderin. Kidneys: Congestion, otherwise negative. Adrenals: In one adrenal there is an area of fibrosis. Liver: Intense congestion. Cloudy swelling. Cirrhosis. Coccidiosis. Gallbladder is completely necrotic. Necrotic diffusion zone in surrounding liver tissue. The larger bile ducts show hyaline swelling of their columnar epithelium. Excessive mucus formation. Stomach: Post-mortem change, otherwise negative, except at the pyloric end, where there is a marked mucoid degeneration. Intestine: Extreme catarrh. All glands show extreme mucous degeneration. Desquamation of the epithelium. Skin lesion: Large area of necrosis. Hemorrhage. Edema. No reaction.

RABBIT 64.- Subcutaneous injection of 0.12 c.c. of dichlorethylsulphide. Severe reaction. Death on sixth day. The close parallelism between this case and the preceding is noteworthy.



Autopsy.-Body cold. No rigor. Edema of abdominal wall has diminished. There is a small red-brown eschar at the site of injection. Heart: Moderate dilatation. Fibrin and leucocyte clot in right auricle. Lungs are somewhat mottled, particularly in upper lobes, probably an early lobular pneumonia. Congestion. Peritoneum: Moist, shining. Liver: Congested. Gall-bladder: Well filled. Bile, thin and greenish-yellow in color. Small intestines and greater part of colon distended with very abundant yellowish-brown fluid fecal material. Formed stools in descending colon. Kidneys: Moderate congestion. Splanchnic vessels all show marked congestion.

Microscopic findings.-Heart: Negative. Large white clot with very few leucocytes. Lungs: Purulent bronchopneumonia. Marked congestion and edema. There is a large thrombus in a pulmonary vessel, the wall of which shows necrosis. Around the vessel there is a zone of necrosis which shades off to partial necrosis, as though due to the diffusion of a necrosing substance from the vessel. There are enormous numbers of staphylococci through- out the lung. Kidneys: Congestion. Numerous casts in one. Gall-bladder: Wall completely necrosed, with area of diffusion necrosis in surrounding liver tissue. Liver: Congestion. Cloudy swelling. Small areas of lime salt deposit in necrotic liver cells. Some of the larger bile-ducts show a polynuclear infiltration around the columnar epithelium. The epithelium itself shows a slight cloudy swelling. Adrenals: Increased lipoid content in cortex. Conges- tion of medulla. Stomach: Negative. Intestines: Marked desquamative catarrh. Necrosis of upper portion of mucosa. Extreme mucoid degeneration of the glands. Site of injection: Enormous area of necrosis extending clear through the skin and subcutaneous tissues. Hemorrhage. Thrombosis. Areas of polynuclear infiltration. Secondary infection.

RABBIT 67.- Subcutaneous injection of 0.18 c.c. of dichlorethylsulphide. Very severe reaction. Severe diarrhea within 24 hours and death during the third day.


Autopsy.-Body cold. Rigor mortis present. Head retracted. Fur about the anus is wet with liquid feces. The abdominal wall shows a marked edema which extends up the right flank. At the site of injection the subcutaneous tissues show but very little edema and no surface lesion is visible. Upon incising this area the fascia is found to be yellow in color. Pleural cavities and pleura are negative. Heart: Moderate dilatation. Blood coagulated. Lungs: Marked congestion. Peritoneum: Moist, shining. No free fluid in peritoneal cavity. Liver: Congestion. Gallbladder: Moderately filled with thin bile. The stomach is partly distended. Its wall is somewhat edematous and in the mucosa several minute erosions are visible. Spleen and kidneys: Congestion. Adrenals: Negative. The intestines are distended with gas and fluid feces. Marked diarrhea. Splanchnic congestion.

Microscopic findings
.-Heart: Negative. Lungs: Congestion. Some edema. Areas of atelectasis. Spleen: Congestion without pigmentation. Kidneys, adrenals, and pancreas: Negative. Liver: Congestion. Slight cloudy swelling. The larger bile-ducts show desquamation of the epithelium and some of them are completely necrosed. Gall-bladder: The wall is completely necrosed and there is some necrosis from diffusion into the surrounding liver tissue. Stomach: Negative. Intestines: Catarrhal enteritis. Marked mucoid degeneration throughout.


RABBIT 66.- Subcutaneous injection of 0.18 c.c. of dichlorethvlstilphide. Severe diarrhea. Marked fall in temperature. Death on the eleventh day.


Autopsy.-Body still warm. No rigor. Both eyes sealed with exudate. When lids are separated, a large quantity of mucopurulent exudate escapes. No superficial lesion at site of injection and only a slight subcutaneous induration can be felt. There is a firm mass, about 2.5 cm. in diameter, in the abdominal wall at the site of the marked edema noted while the animal was living. On incising this mass it is found to be somewhat edematous still, and the tissues are stained a bright sulphur-yellow color. Heart: Right-sided dilatation. Lungs: Pale pink in color. Slight congestion. Liver: Marked congestion. Gallbladder: Distended. Bile highly pigmented. Stomach: Nearly empty; its contents bile stained and very mucoid. In the mucosa there are numerous small erosions, particularly in the fundus, along the greater curvature. These are covered with an abundant mucus and have shreds of brownish-red material, probably blood clot, streaking the mucus over them. Spleen: Negative. Kidneys: Congestion and slight cloudy swelling. The intestines contain gas and a small quantity of fluid material. There are no formed stools. Adrenals: Congestion.

Microscopic findings
.-Heart: Negative. Lungs: Moderate congestion. Emphysema. No edema. Kidneys: One kidney shows many casts in the medullary tubules. Otherwise negative. Spleen: Markedly congested, with a fairly large number of pigmented phagocytes. Liver: Congestion and cloudy swelling. Stomach: Extreme mucoid degeneration of the epithelium of the outer portion of the mucosa, with desquamation. Marked cloudy swelling of the parietal cells. Small postmortem erosions. Gallbladder: Mucosa fairly well preserved. No necrosis of wall or neighboring liver tissue. Intestines: Marked mucoid degeneration. Extreme catarrhal enteritis.

RABBIT 69.- Subcutaneous injection of 0.24 c.c. of dichlorethylsulphide. Diarrhea. Death within 12 hours.


Autopsy.-Body cold. Rigor present. Soft formed stools matted in hair about anus. No superficial lesion at site of injection, only a slight subcutaneous edema. Upon incising this area there is a very strong odor of dichlorethylsulphide. Heart: Marked dilatation. Lungs: Congestion, with small hemorrhages beneath the pleura of the diaphragmatic surfaces of the lower lobes. Liver: Congestion. Stomach: Moderately distended. No lesions of mucosa visible. Spleen, kidneys, and adrenals: Congestion, otherwise negative. Intestines: Distended with gas and fluid fecal material. No formed stools. Marked splanchnic congestion.

Microscopic findings
.-Heart: Negative. Lungs: Congestion. Edema. Small capillary hemorrhages. Adrenals: Negative. Spleen: Congestion. No pigmented phagocytes. Pancreas: Negative. Kidneys: Rather marked cloudy swelling, even to simple necrosis. Gallbladder: Complete necrosis of wall, neerosis extending into liver tissue for some distance. Slight cloudy swelling of the liver cells. Stomach: Negative. Intestines: Congestion; otherwise negative. Site of injection: Large area of necrosis. Edema. No reaction.


RABBIT 70.- Subcutaneous injection of 0.30 c.c. of dichlorethylsulphide. Diarrhea. Death within 12 hours.


Autopsy.- Body cold. Rigor present. Much soft fecal material about the anus. Strong odor of dichlorethylsulphide when area of injection is excised. Heart: Marked dilatation. Lungs: Congestion, otherwise negative. Liver, spleen, kidneys, and adrenals: Congestion. Gall-bladder: Well filled. Intestines: The tipper portion of the small intestine is filled with yellowish fluid material. No formed stools in lower colon. Marked splanchnic congestion.

Microscopic findings.- Heart: Hypertrophy and dilatation. Lungs: Intense congestion and edema. Spleen: Congestion. Very few pigmented phagocytes. Pancreas: Negative. Adrenals: Congestion. Kidneys: Congestion and some cloudy swelling. Liver: Congestion and fatty degeneration. Gallbladder: Necrosed. Diffusion zone of necrosis in the surrounding liver tissue. Intestines: Epithelium well preserved. Early stage of mucous degeneration; practically every cell filled with mucus and the majority of them intact. Site of injection: Large eschar involving all tissues.

RABBIT 72.- Subcutaneous injection of 0.60 c.c. of dichlorethylsulphide. Death on the second day.


.-Body still somewhat warm. Perianal hair soiled. Very strong odor of dichlorethylsulphide at site of injection. Heart: Right-sided dilatation. Lungs: Congestion. Liver, spleen, kidneys, and adrenals: Congestion. Otherwise negative. Stomach: Mod- erately distended. In the mucosa there are several small erosions with brownish-red bases. Intestines: Much fluid material. No formed stools in lower colon. Diarrhea. Edema of thoracic and abdominal walls.

Microscopic findings.-Brain: Congestion. Heart: Negative. Lungs: Congestion without edema. Spleen: Intense congestion without pigmented phagocytes. Kidneys: Congestion. Otherwise negative. Stomach: Negative. Some post-mortem change. Small intestine: Marked post-mortem change, with the picture of a mucous catarrh involving the greater part of the tract. Liver: Cirrhosis. Coccidiosis. Large bile ducts dilated; their epithelium well preserved. The small bile-ducts are unchanged. Slight fatty degeneration. Gall-bladder: Nearly complete necrosis of wall. Narrow necrotic zone of diffusion into surrounding liver tissue.

RABBIT 74.- Subcutaneous injection of 0.60 c.c. of dichlorethylsulphide, 0.15 c.c. being injected in each of four widely separated areas on the back. Rabbit killed when dying, five hours after injection. An apparent acceleration of reaction with increased opportunity for absorption.



Autopsy.-Head somewhat retracted. Perianal region covered with soft fecal material. Heart: Moderately dilated. Lungs: Negative. Liver: Congestion. Gallbladder: Nearly empty. Negative. The stomach is moderately distended. The mucosa of the fundus, particularly along the greater curvature, shows a large number of small, shallow erosions with dark red-brown bases. Intestines: Contain much fluid material. No formed stools in colon.

Microscopic findings.-Liver: Marked congestion. Slight cloudy swelling. The small bile-ducts arc apparently normal. Gall-bladder: Epithelium perfectly preserved. No changes observable. Stomach: Well preserved. Portions examined show no changes. Intestines: Catarrhal enteritis.

DOG 3.- White bull terrier; weight 8 kilos. Subcutaneous injection of 0.24 c.c. dichlorethylsulphide. Severe diarrhea. Death in four days.
August 27.- 0.24 c.c. of dichlorethylsulphide injected subcutaneously in back.
August 28.- Appears sick. Is much less active and coat is roughened.
August 29.- Appears very sick. No diarrhea.
August 30.- Very severe diarrhea. Stools fluid and brownish black in color, resembling altered blood. The animal remains quiet.
August 31, 12 noon.- Lying on side. Respiration slow and shallow. Diarrhea persists. 6 p. m.- Dog died during the afternoon. Autopsy at 7 p. m.

Autopsy.- The body is in complete rigor. The posterior portion is soiled with fluid fecal material. At the site of injection there is an indurated lamellar area, about 10 cm. in diameter, which involves both skin and subcutaneous tissue. Pleural cavities and pleurae negative. Heart: Marked right-sided dilatation, the left ventricle being in rigor. Small amount of very dark fluid blood in the chambers. Lungs: Free throughout; air-containing. Middle lobe of right lung congested. No pneumonia. No free gas or fluid in peritoneal cavity. Omentum free. The subperitoneal vessels over the intestines are congested and there are a few minute subperitoneal hemorrhages. The intestines are empty and firmly contracted. The duodenum contains a bile-stained fluid and the mucosa is bile stained. Throughout the small intestine the mucosa shows a marked congestion and there are small hemorrhages in the summits of some of the folds. The Peyer's patches are hyperplastic. In the colon the mucosa is much congested and there are a few very minute hemorrhages. The appendix is negative. The stomach wall is contracted, the rugae prominent. The mucosa is much congested and there are a few pin-point hemorrhages. The kidneys show slight cloudy swelling. The adrenals were well preserved and negative. The bladder is contracted; its mucosa, negative.

Microscopic findings.- The heart shows dilatation and its vessels are congested. Lungs: Marked edema. Intense congestion. Small hemorrhages. Several areas of hemorrhagic infarction, with embolic blocking of the vessels. Spleen: Congestion. Atrophy. Kidneys: Marked congestion. Slight cloudy swelling. Pancreas: Marked cloudy swelling. Liver: Cloudy swelling. Marked congestion. Marked mucoid degeneration and cloudy swelling of the epithelium of the smaller bile-ducts. All of the bile-ducts contain a violet-staining, hyaline substance and the epithelium is higher than normal, the cells appearing larger and swollen. The nuclei appear increased in number. The cytoplasm stains violet, but many of the cells are vacuolated. In a few of the larger bile-ducts there are small areas of necrosis. The stomach mucosa is intact. There is marked congestion and some increase in mucus in the upper portion of the mucosa. The mucosa of the small intestine is congested, edematous, and infiltrated with leucocytes. The epithelium of the glands shows syncytial formations of regenerating epithelium. The columnar form of the cells is lost and there is no mucus formation except in the lower small intestine, where there are many mucous cysts, especially in the deeper portions of the mucosa. The fundi of the glands of Lieberkuhn are dilated, filled with stringy mucin or a colloid material. The epithelium is cuboidal, or flattened, or syncytial, and stains a deep violet with hematoxylin and eosin. In some of the dilated glands of Lieberkuhn the epithelium is entirely gone and the lumen is filled with a hyaline cast. The epithelium of the upper part of the glands and of the surface is entirely gone. The picture is that of a severe degenerative and desquamative catarrhal enteritis with beginning regeneration. The lymph follicles are hyperplastic and the germ centers show lymphoid exhaustion. Mesenteric nodes: Marked congestion. Great numbers of hemophages filled with blood cells. Adrenals: Marked congestion of medulla. Site of injection: Large eschar with marked edema. In the vessels there are large thrombi. The borders of the lesion show an abundant polynuclear infiltration and large areas of hemorrhagic extravasation.



Since it has been assumed by some that the local and general effects of dichlorethylsulphide are due to the action of the hydrochloric acid produced by its hydrolytic cleavage, a series of animals was given varying amounts of hydrochloric acid by subcutaneous injection. The following protocol is selected as illustrative of this group.


RABBIT 76.- Subcutaneous injection of 0.60 c.c. of hydrochloric acid. Both local and general reactions entirely unlike those produced by dichlorethysulphide.


.-Large excavated lesion, measuring 5 by 7 cm. at the site of injection. The borders are smooth. The base consists of charred, dry, fragmented muscle in coarse bundles and muasses. There is no pturulent exudate. The whole lesion is nearly black in color, as though charred, and is verv foul snselling, with the odor of dry gangrene. Heart: Still heating. Negative. Lungs: Bright pink its color. Moderate congestion. Liver: Congested. Extensive coccidiosis. Spleen: Negative. Stomach: Contains a small amount of food and water, and much swallowed mucus. There are no visible lesions of the mucosa. The intestines are nearly empty except the cecum, which is moderately filled. The scant amount of fecal matter in the small intestine is very thin, mucoid, and bile stained. In the lower colon the fecal material is soft and but slight formed. There is, however, no evidence of diarrhea about the anus.

Microscopic findings.- Heart: Negative. Lungs: Congestion. No edema and no hemorrhages. Liver: Coccidiosis. No diffusion necrosis into liver tissue around the gallbladder. Gallbladder: Epithelium well preserved. No necrosis. No desquamation. Many mucin threads in lumen. Stomach, intestines, kidneys, and adrenals: All negative. Tissue from site of injection: Very extensive eschar formation.


A series of animals was given varying doses of dihydroxyethylsulphide (hydrolyzed mustard gas) by subcutaneous injection, in order to ascertain Whether the severe general effects could be due to the absorption of this substance from the site of injection. The following condensed protocols are selected from this series.


RABBIT 80.- Subcutaneous injection of 0.30 c.c. of dihydroxyethylsulphide. No local lesion, no diarrhea, and no loss of weight.



RABBIT 81.- Subcutaneous injection of 0.60 c.c. of dihydroxyethylsulphide. No local lesion. No diarrhea.


Autopsy.-There is a slight congestion of all organs. Stomach is well filled, its mucosa grossly negative. Intestines: Apparently negative. Formed feces in lower colon. No evidence of diarrhea. The bladder contains a very turbid yellowish-white urine. No other changes.

Microscopic findings.- Lung: Slight congestion. No edema. Kidneys: Negative. Slight congestion. Adrenals: Negative. Spleen: Negative. Liver: Marked fatty change. Gall-bladder: Epithelium perfectly preserved. No changes. Small intestine: Slight, but well-defined catarrh. Congestion of mucosa. Excessive mucus formation. Stomach: Negative.


The following condensed protocols are selected from the various series of intravenous injections. In every case in which dichlorethylsulphide was given, it was injectedl without dilution. Although bland oils will serve as suitable diluents so far as mutual solubility is concerned, the certainty of producing one degree of fatty embolism, with resulting confusion in the pathologic picture, renders this method inadvisable. In addition, the much greater solubility of dichlorethylsulphide in the oil than in the body fluids may be expected to retard its action and delay the production of its characteristic symptomatology.


RABBIT 53.- Intravenous injection of 0.06 c.c. of dichlorethylsulphide. General Convulsions. Marked reduction of temperature. Prostration. Death in less than three hours.


.-Body nearly cold. Rigor mortis marked. Abdomen moderately distended. No diarrhea. Pleural cavities and pletlrae negative. Heart: Moderately dilated, particularly on the right side. Dark fluid blood, clotting tardily. Lungs: Air-containing throughout. Congestion moderate, except posteriorly. Spleen: Negative. Kidneys: Slight congestion. Adrenals: Negative. Liver: Marked congestion. Lobules large. Central zones dark red, outer two-thirds of lobule, grayish-red. Gall-bladder: Well filled with a pale mucoid bile. Stomach: distended with food. Mucosa shows scattered areas of post-mortem change along greater curvature. Upper small intestines contain a thin yellowish mucoid fluid. Lower small intestine is well filled and there are formed stools in the lower colon.


Microscopic findings
.- Central nervous system: Congestion. Heart: Negative. Lungs: Congestion. Atelectasis. No edema. No hemorrhage. Spleen: Marked congestion but no pigment. Adrenals and pancreas: Acute congestion. Stomach: Slight cloudy swelling of parietal cells. Congestion. Intestines: In portion examined the mucosa is well preserved. Congestion of villi. No necrosis. Liver: Moderate congestion. Slight fatty degeneration.

RABBIT 51.- Accidental intravenous injection of dichlorethylsulphide in the course of a deep subcutaneous injection. Total injection 0.12 c.c., of which probably less than 0.06 c.c. entered a small subcutaneous vein. Convulsions. Diarrhea. Death in four hours .

Autopsy.- Body still warm but rigor mortis is present. Soft fecal material from anus. Posterior portion of body is much soiled with feces. Pleural cavities and pleurae: Negative. Heart: Right-sided dilatation. Blood, fluid, clotting slowly. Lungs: Free. Slight con- gestion. Air-containing throughout. Spleen: Negative. Kidneys: Moderate congestion. Adrenals: Negative. Liver: Moderate congestion. Stomach: Distended with food. The first portion of the small intestine contains a very abundant slightly yellow fluid, which is somewhat mucoid. In the descending colon and rectum there are no formed stools.

Microscopic findings.- Heart: Negative. Lungs: Moderate congestion. No edema. Small patches of atelectasis. Bronchioles dilated. No hemorrhages. Spleen: Moderate congestion with few pigmented cells. Kidneys: Congestion. Intestine: Mucosa well preserved. Stomach: Negative. Liver: Congestion. Slight cloudy swelling.



RABBIT 89.- Intravenous injection of 0.06 c.c. of hydrochloric acid. Animal living on fourth day. No diarrhea. No evidence of general reaction.


.- Gross findings all negative, except thrombosis of jugular at site of injection.

Microscopic findings.- All negative, except for thrombophlebitis at site of injection.


RABBIT 87.- Intravenous injection of 0.30 c.c. of dihydroxyethylsulphide. Animal living on the fifth day. No diarrhea. No marked loss of weight.


.-All gross findings negative.

Microscopic findings.- Negative, except for marked fatty liver.


Subcutaneous injections
.- When pure dichlorethylsulphide was injected subcutaneously in doses of from 0.015 up to 0.60 c.c. the injections were apparently painless and the animals exhibited no signs of discomfort. Even with the


largest doses the animals resumed feeding as soon as released and gave no evidence of local irritation. At varying periods of time, from one hour to several days, depending in part upon the size of the dose and in part upon the rate of absorption, toxic symptoms appeared, usually in the form of salivation, diarrhea, and marked depression of temperature. At first the respirations were quickened; later they became slow. From the largest doses the animal might die within two hours, without diarrhea, but with a short period of nervous excitement followed by coma and gradual failure of respiration. Doses of from 0.015 to 0.06 c.c. were not necessarily fatal to rabbits, although at certain proportion of the animals receiving such doses did die. Death from these doses occurred from the fourth to the tenth day. Diarrheal symptoms might appear as early ats the second day, but in the great majority of cases diarrhea did not appear until the fourth to seventh day. Coincident with the diarrhea there was a marked fall of temperature, which could be as much as seven to eight degrees below normal. If the animal survived the diarrheal period, the temperature might come back nearly to, or quite to, normal, to fall again just before death, if the animal died. During the diarrheal stage the animal rapidly lost weight and this loss of weight continued after the cessation of diarrhea, so that when the animal died it might have lost as much as one-third of its body weight. The diarrheal stools were fluid, mucoid, brown to black, sometimes tarry, and very foul smelling. Accompanying the diarrhea, there was a marked anorexia and great thirst; the animal was quiet, depressed; and both circulation and respiration were slowed. In the animals that recovered, the diarrhea might last for several days, accompanied by a marked lowering of temperature and a general depression. As the stools become normal the temperature rose and the general condition improved until the animal was again apparently perfectly normal.

At the site of the subcutaneous injection there developed a local edema which was usually much less than the edema produced by cutaneous applications. In a certain number of cases there was a marked edema of the belly wall, although the injections were routinely made on the back of the animal, apparently due to the hypostasis of body fluid containing mustard gas, from the seat of the injection. The edematous area at the seat of injection gradually chalanged into an indurated eschar (figs. 210 and 211) which underwent a slow demarcation from the neighboring living tissue. In some animals injected subcutaneously with the dichlorethylsulphide furnished by the Chemical Warfare Service and living two days or longer, a sulphur-yellow coloration of the tissue at site of injection was noted. Such a coloration was not noticed in any animal injected with the Gomberg preparation.

The gross pathology shown by animals killed or dying at varying periods after subcutaneous injections consisted of a general passive congestion of all organs with occasional minute hemorrhages, emboli, and infarctions. The most specific changes were those found in the gastrointestinal tract in the form of an intense splanchnic congestion and a more or less severe catarrhal enteritis. In practically every instance it was noted that the contents of the gallbladder consisted of a thin, pale, yellowish, mucoid bile.

The microscopic findings confirmed the gross appearances. No specific changes were found in any organs except the intestines, spleen. and possibly the bile-ducts. (Figs. 212 to 221.) These specific changes consisted in a mucoid


degeneration, necrosis, and desquamation of the epithelium of the intestinal mucosa, marked congestion of the whole splanchnic area, with edema and occasional petechial hemorrhages and minute erosions of the mucosa. Degenerative changes were also noted in the epithelium of the bile-ducts, and the post-mortem changes in the mucosa of the gallbladder and larger bile-ducts appeared to be greatly hastened, and more severe than normal, as though from the presence of some necrosing substance in the bile. A very striking feature of the pathology of many cases of subcutaneous injections of dichlorethylsulphide was the presence in the blood spaces and sinusoids of the spleen of great numbers of large pigmented phagocytes and hemophages containing .altered red blood cells, indicating a greatly increased hemolysis. (Fig. 221.)

FIG. 210.- Rabbit. Eschar resulting from subcutaneous injection of 0.03 c.c. of dichlorethylsulphide, 18 days after injection

The microscopic appearances of the tissues at the site of injection were those of an extensive eschar extending entirely through the skin and fascia, and deep into the muscles with extensive extravasations and infiltrations of leucocytes. The appearances of secondary infection were frequently added to those of the primary lesion. In several instances deposits of lime salts were noted in the necrotic area.

Intravenous injections
.-When dichlorethylsulphide was injected directly into the external jugular or superficial femoral veins in doses varying from 0.0075 to 0.18 c.c., the animal showed no signs of pain and returned to eating, but within a short time began to show symptoms of hyperexcitability in the form of very rapid respiration and slight convulsive movements. Within an hour there might be general convulsions and opisthotonos with head turned to one side. The animal quickly became very weak, and tended to lie upon its side


with rapid jerking movements of the legs, as though running. The animal could not turn over. Soft feces were frequently involuntarily discharged. The animal acted as though nauseated and there were slight salivation and lacrymation. The pupils were dilated. The temperature rapidly fell 6° to 8° before death. The convulsions ceased, the pupils contracted, salivation increased, the animal passed into coma, and there was a gradual failure of circulation and respiration, although the respiration ceased some time before the heart stopped beating. Death usually occurred in one to four hours, depending upon the size of the dose. Following the intravenous injection of larger doses, the first convulsive movements might appear in 10 minutes and be followed rapidly by wild clonic convulsions, very much resembling those of rabies.

FIG. 211.- Rabbit. Eschar resulting from subcutaneous injection of 0.06 c.c. of dichlorethylsulphide, 11 days after injection

The gross pathologic changes following intravenous injections of dichlorethylsulphide were dilatation of the heart, particularly on the right side, and marked congestion of the lungs with numerous petechial hemorrhages. All other organs showed congestion with numerous minute hemorrhages. Mucosa of gastrointestinal tract was covered with thick mucus, was congested, and might show minute hemorrhages. The content of the gall-bladder occasionally was a thin and pale bile. No specific changes were found in the brain and cord and no thrombus in the vein at the site of injection.

Microscopic examination confirmed the gross appearances, in that it showed extreme congestion, stasis and hemorrhages in the lungs; congestion of the gastrointestinal tract, with an increased formation of mucus scattered irregularly throughout, not comparable, however, to the changes found in animals injected subcutaneously; and marked general congestion.


Epicrisis.-The introduction of dichlorethylsulphide into the blood stream caused death within a few hours with characteristic symptoms. When injected subcutaneously, with resulting slower absorption into the circulation, death occurred later, from 12 hours to several days, likewise with a characteristic symptomatology. The symptoms could be classed roughly as (1) nervous and (2) intestinal. In the case of large doses, with death occurring within an hour, the nervous symptoms alone might be exhibited, while in te animals living for some time the nervous symptoms were slight and those of the intestinal group predominated. In the correlation of the pathologic findings with the symptomatology, no change was found in the central nervous system offering an adequate explanation for a death dependent entirely upon a lesion of the central nervous system. In the delayed cases (subcutaneous) the intestinal tract

FIG. 212.- Rabbit. Received subcutaneous injection of .045 c.c. of dichlorethylsulphide. Died on third day during mild diarrhea. General mucoid degeneration

offered a pathologic picture commensurate with the symptoms. In addition to the specific pathology just mentioned, incidental pathologic findings in the form of thrombi, emboli, embolic infarctions and hemorrhages could at times be interpreted as explaining some of the observed clinical phenomena.

Of all the pathologic lesions produced by mustard gas, apparently the most specific and most interesting were the intestinal changes following intravenous and subcutaneous injections. The intense catarrhal enteritis observed after such injections suggested the excretion of dichlorethylsulphide or some product of its decomposition through the intestinal mucosa. No changes were observed in the kidneys indicating the excretion of any toxic product through the renal epithelium. In the case of the liver there were observed, at times, degenerative changes in the biliary epithelium, suggesting the presence


of some injurious substance in the bile. Further, the probability of this hypothesis seems the greater in view of certain changes, found in a certain proportion of cases, in the gallbladder, in the form of an apparently earlier post-mortem necrosis of the gallbladder wall, and a diffusion of the bile into the neighboring liver tissue. It has been a notable observation that, in many cases, when autopsy was performed immediately upon death of the animal, such early post-mortem change was found in the gall-bladder; while in animals that had not received dichlorethylsulphide injections, such an immediate post-mortem change was not observed.

A large number of experiments were made in the effort to demonstrate the presence of mustard gas or its products in the intestinal wall and contents, and

FIG. 213.- Rabbit. Received subcutaneous injection of O.18 c.c. of dichlorethylsulphide. Began to have diarrhea seven days after injection and died four days later. Section of upper portion of small intestine

showing acute catarrhal enteritis in the liver and bile. In the tissues at the site of the injection, the odor of mustard gas would persist for several days, and cutaneous applications of the ether extract of these tissues produced a well-marked mustardgas burn, even when the ether extract was much diluted. Further, in oneI case in which death occurred within one hour after jugular injection, with numerous small embolic hemorrhages in the lungs, the odor of mustard gas was clearly evident on section of the lung, and an ether extract of the lung gave a slight positive test. With this single exception we were unable to obtain any odor of mustard gas or any positive skin reaction from extracts of any organ. Ether and aqueous extracts of the liver and bile, the intestinal tract and its contents, the blood, heart and lungs, and of the urine, even when concentrated, vielded no odor of mustard gas and gave no positive skin test.


It has been generally assumed that the injurious action of mustard gas follows its hydrolysis in the living cells and tissues. The products resulting from the hydrolysis of dichlorethylsulphide are hydrochloric acid and dihydroxyethylsulphide.

From the chemistry department of the University of Michigan a quantity of pure dihydroxyethylsulphide was obtained. Varying series of experiments made with this substance showed it to be apparently inert, both as applied locally, by ingestion, and by subcutaneous and intravenous injections, in quantities up to many times the lethal doses for dichlorethylsulphide. No diarrhea, no fall in temperature, and no other symptoms were produced, and no pathologic lesions were found in such animals with the exception of marked fatty changes in the liver cells.

FIG. 214.- Same rabbit as in Figure 213. Extreme mucoid degeneration. Catarrhal enteritis. Mucous diarrhea

A similar set of experiments with hydrochloric acid injections was also carried out. The subcutaneous injection of hydrochloric acid, even in a dose of 0.12 c.c., produced intense local irritation and pain, in striking contrast to the anesthetic action of dichlorethylsulphide. The course of the local lesion was entirely unlike that produced by mustard gas. The liquefaction of the eschar was very rapid and there quickly resulted a deep excavated lesion, the bottom and sides of which appeared charred. (See fig. 222.) No diarrhea was produced; the temperature remained normal or rose; no symptoms were produced, and the animal recovered unless secondary infection set in.

The intravenous injection of 0.06 c.c. of hydrochloric acid (diluted to prevent local injury during injection) produced no symptoms comparable to those resulting from the intravenous injection of dichlorethvlsulphide.


The manner of causation of the intestinal lesions remains the special problem of the pathology of dichlorethylsulphide poisoning. Neither hydrochloric acid nor dihydroxyethylsulphide, when introduced into the circulation, will cause similar changes. Therefore these lesions can not be the result of the action of hydrochloric acid produced at the site of the lesion and circulating in the blood. The next most plausible explanation of the cause of the intestinal lesions is that mustard gas itself or some other unknown decomposition product circulates in the blood and is excreted through the mucosa of the intestine, or possibly also through the bile.

It has been stated that the most delicate chemical test for mustard gas is a color reaction with a solution of platinic chloride and sodium iodide. The

FIG. 215.- Section of cecal wall from same rabbit as in Figure 213. Marked catarrhal inflammation. Mucous diarrhea

liquid to be tested is applied to absorbent paper moistened with this solution. In the presence of dichlorethylsulphide the pink color is changed to a faint purple, which becomes blue or deep blue, depending upon the concentration of the mustard gas solution. When this test was applied to ether and aqueous extracts of the liver, bile and intestine from animals injected subcutaneously and intravenously with dichlorethylsulphide, a definite blue color was obtained more marked in the case of the aqueous extracts. This reaction at first was believed to be a positive test for mustard gas in these extracts. Continuing the control experiments, it was found that dihydroxyethylsuphide gives the same reaction with the platinic clhloride-sodium iodide as dichlorethylsulphide. The aqueous extracts of liver, bile, intestine and urine, which might be expected to have dihydroxyethylsulphide in them, if it were present at all, since it is soluble in water. were then taken an(l heated with concentrated hydrochloric acid, with


the expectation of reconverting the hydrolyzed product into mustard gas. The products thus obtained, however, gave no odor of mustard gas, and their ether extracts gave neither a. skin reaction nor a positive color test. On the other hand, these aqueous extracts of body fluids and organs did give bluish-green color tests with the test solution that might be interpreted as faint positive reactions for mustard gas. Unfortunately, however, the aqueous extracts of these same organs and fluids from animals untreated with mustard gas produce an identical color reaction. It was impossible, therefore, to demonstrate the presence of mustard gas or its hydrolyzed products in the liver, bile, intestines, feces, blood, or urine.

FIG. 216.- Rabbit. Received subcutaneous injection of 12 c.c. of dichlorethylsuliphide. Diarrhea began on second day, the animal dying three days later. Extreme mucoid degeneration of the entire intestinal epithelium. Mucous diarrhea


1. Dichlorethylsulphide when injected internally in doses of approximately 0.06 c.c. per kilo for the rabbit, and 0.03 c.c. per kilo for the dog, causes a fatal intoxication, characterized, when death occurs quickly, by symptoms referable to the central nervous system; but when death takes place more slowly, by intense diarrhea, anorexia and reduction of temperature.
2. The only specific pathology of such fatal poisonings is a marked degeneration of the epithelium of the gastrointestinal tract in the form of a severe catarrhal enteritis, and the occasional occurrence of similar changes in the epithelium of the bile-ducts and gallbladder.
3. In a certain number of animals injected subcutaneously a marked hemosiderosis of the spleen wtas observed. There seemed to be a relationship between the degree of the splenic pigmentation and that of the local extravasations at site of injection. As such splenic pigmentation was not constant, it seems


more likely that it was the result of the local hemolysis and not due to any specific action upon the blood or blood-forming organs. The splenic hemosiderosis also bore a definite relationship to the number of pigmented hemophages seen in the sinuses of the lymph and hemolymph nodes. The latter can be similarly explained. However, since hemosiderosis of the spleen is not an uncommon finding in laboratory rabbits, its occurrence in these animals may have been purely coincident.
4. These changes suggest the excretion of mustard gas or some poisonous product resulting from its decomposition into the gastrointestinal tract. No positive proof, however, of this mode of excretion could be obtained. The hydrolysis of mustard gas circulating in the blood may take place in the intestinal mucosa, thereby producing the degenerative changes seen in these cells.

FIG. 217.- Dog. Received subcutaneous injection of 0.24 c.c. of dichlorethylsulphide. Died in four days with a very severe diarrhea. Extreme catarrhal desquamative enteritis. Mucoid degeneration and necrosis of the glandular epithelium On the other hand, the intestinal conditions may not be due to the direct action of any substance upon the epithelium of the mucosa, but may be secondary to the splanchnic congestion and the hyperexcitability of the nervous centers.

5. It seems probable that the characteristic symptoms and death following subcutaneous and intravenous injections of dichlorethylsulphide are due to the direct action of this substance circulating in the blood, upon the central nervous system, either with or without hydrolysis in the cells of the nervous tissue affected. As the result of intracellular hydrolysis, hydrochloric acid may be liberated within the cell and give rise to the toxic effects observed. The specific character and constancy of the symptomatology offer sufficient argument against any assumption that the process is embolic in character.


In the series of acutely fatal intravenous injections of dichlorethylsulphide described above, no specific changes in blood or blood-forming organs were noted. Since the appearance of Pappenheimer's preliminary report 10 there was carried out in the pathological laboratory of the University of Michigan a series of investigations using much smaller doses of dichlorethylsulphide in alcoholic solutions. These experiments confirmed Pappenheimer's observation that in rabbits living to the third or fourth day after such injections, there occurs an initial leucocytosis, followed quickly by an extraordinary drop, so that before death the leucocytes may practically vanish from the circulation. Two protocols are given:

FIG. 218.- Same dog as in Figure 217. Middle portion of small intestine, showing desquamation of the superficial epithelium and necrosis of the epithelium of the gland of Lieberkuhn. Severe enteritis


RABBIT 103.- Injected intravenously with 0.006 c.c. of dichlorethylsulphide in alcoholic solution. A marked leucocytosis was noted in four hours. On the next day the white cells began to fall, and on the third day had reached 275. During this period the red cells remained slightly higher than normal. Blood smears showed only an occasional white cell. These were about equally divided between polynuclear leucocytes and degenerating mononuclears. Died on the fifth day.

RABBIT 104.- Given 0.010 c.c. of dichlorethylsulphide intravenously in alcoholic solution. Showed in 17 hours a leucocytosis of 23,600. On the next day the white cells began to fall rapidly, on the fourth day reaching 325. Smears showed only occasional white cells, about equal numbers of polynuclears and degenerating mononuclears. Died on the fifth da y.

Both of these animals showed extraordinary depletion of the bone marrow and, to a lesser degree, of the spleen and lymphoid tissues. The second rabbit showed marked general edema.


Intravenous injections of small amounts of pure dichlorethylsulphide do, therefore, produce a marked leucopenia before death.


In Meyer's first laboratory experiments with dichlorethylsulphide he noticed an apparent difference in individual susceptibility, in that he, himself, was not affected by exposure to it, while a laboratory worker engaged in making it developed conjunctivitis and a severe skin eruption.1 Meyer's conclusion that individual susceptibility must vary greatly seems hardly warranted by this observation. Undoubtedly Meyer was working with a very impure compound, the concentration of which must have varied from time to time, and

FIG. 219.- Rabbit. Received subcutaneous injection of 0.006 c.c. of dichlorethylsulphide. Very severe diarrhea from fourth to ninth day. Killed 5 days after injection. Mucosa of intestines intensely congested and edematous. Marked mucoid degeneration with cystic glands. Regeneration of superficial epithelium

the laboratory worker must have been much more exposed to its fumes than Meyer himself. Further, the dissemination of mustard gas is such an insidious matter of physical conditions and pure chance, that in a group of workers exposed to apparently identical conditions the greatest diversity of effects may be produced, leading to an incorrect supposition of especial resistance or susceptibility in the individual members of the same group. The question of the existence of individual susceptibility can be settled, therefore, only by the application of experimental methods in which the conditions of exposure are identical.

An acquired hypersensitivity to mustard-gas vapor as the result of previous or repeated burns is claimed by many workers and accepted by some observers.


That such a localized susceptibility may be shown in scars, healing burns, injured conjunctiva and mucous membranes, is very probable, and our own observations bear this out, but there is no evidence to the effect that exposures to mustard gas increase the individual's susceptibility on the part of the whole organism to this substance.

The individual human sensitivity is in part a racial one and, therefore, intrinsic and constitutional. Marshall 11 would explain it as due to differences in the skin, cutaneous lipoids, etc. Among the individuals of the same race showing differences in cutaneous sensitivity to mustard gas he noted no other constitutional differences or peculiarities. In the cases seen by us there were five individuals who seemed to be cutaneously hypersensitive, as shown by their

FIG. 220.- Rabbit. Received subcutaneous injection of 0.06 c.c. of dichlorethylsulphide. Very severe diarrhea on fifth to seventh day afterwards. Apparent recovery. Killed on the thirty-second day after injection. Mucosa shows excessive mucous formation

receiving frequent and severe burns from vapor exposures not affecting other workers so severely, although apparently similarly exposed. The two fatal cases belonged to this group. It is noteworthy that all five of these cases presented constitutional stigmata or clinical symptoms of a definite pathologic constitution, the thymicolymphatic.



Skin.-Dichlorethylsulphide (mustard gas), in liquid or in vapor form even in very low concentrations, is an escharotic poison for the animal tissues (skin, conjunctivae, cornea, mucous membranes of respiratory and gastrointestinal tracts) with which it comes in direct contact. The degree of the injury


is proportionate to the concentration of the gas, the time of exposure, individual susceptibility, and local physical conditions, such as moisture, sweating, warmth, pressure, and friction. The escharotic action, is, for the greater part, painless, the anesthetic effect being especially notable upon the skin; while upon the mucous membranes its action may be more irritant, probably chiefly reflex in character. The cutaneous surfaces most susceptible are those with thinner, more delicate skin, well supplied with sweat glands and hair follicles, where sweat may collect, and which are exposed to friction or pressure, such as the axillae, flexor surfaces, genitals, inner surface of arms and corresponding surface of trunk, inner surfaces of thighs, and between the fingers. There is a penetration of the gas into the sweat and sebaceous glands, and a resolution

FIG. 221.- Rabbit. Received subcutaneous injection of 0.03 c.c. of dichlorethylsulphide. Very severe diarrhea on fourth to eighth day. Died on twelfth day after injection. Intestines showed severe catarrhal enteritis. Section of spleen showing the great ntmber of pigmented phagocytes in the blood sinuses

of mustard-gas vapor in sweat and sebum occurs. The injuries are particularly striking in their insidious, slowly progressive development, becoming first apparent only some hours after exposure. Upon human skin the lesion appears as a hyperemia, followed by vesication, eschar formation, sloughing, and slow healing, with more or less pigmentation. Depilation may occur; in severe cases the eschar may extend entirely through the corium into the subcutaneous tissues. Secondary infection and gangrene of the eschars occurs invariably in cases not properly treated. Milder lesions may show only the earlier stages of hyperemia, vesication, or pigmentation. In general the injuries may be classed as burns of first, second, or third degree. Following extensive hyperemias in human skin a most marked pigmentation, exceeding in degree the most marked forms of solar tan, may be quickly developed and fade slowly. The pigmentation may


be diffuse or spotted. In human skin vesication is pronounced; in animals the cutaneous lesions are characterized by the development of marked subcutaneous edema in the injured area. The fluid of the vesicle or of the edema is nonirritating when applied to uninjured areas. In the case of human skin frequently exposed to very dilute concentrations (only perceptible by odor), an eczematous itching condition between the fingers, on the genitals, and other parts, may develop; rubbing or scratching of the itching part may lead to the quick development of a blister or superficial eschar (Nikolsky's sign). Such interdigital lesions in laboratory workers may resemble clinically those produced by the itch mite. The genital lesions may be mistaken for venereal sores. Cutaneous areas injured by mustard gas are rendered more susceptible to trauma or

FIG. 222.- Rabbit - Sloughing lesion produced by subcutaneous injection of 0.00 c.c. of hydrochloric acid, 24 hours after injection

other forms of injury, including new exposures to mustard gas. This local susceptibility is, however, a general one, and not a specific lowered resistance to the action of dichlorethylsulphide. Subcutaneous injections of pure dichlorethylsulphide produce painless eschars, followed by dry sloughing with edema less marked than in the case of external cutaneous application; a hypostatic edema may develop on the animal's belly when injected in the back. In the tissues at the site of the injection and in the hypostatic edema mustard gas may be present for some days after the injection as shown by odor and physiologic reaction. The resolution of mild skin injuries is often attended by troublesome itching. Healing of the deep cutaneous eschars is very slow; during the healing of extensive deep lesions the patients complain of a sensation


of "tightness" or contraction of the skin; large scars may be produced resembling those resulting from deep thermal burns. The hairs maybe lost; but when regenerated they may be white in color. Mild burns may be more painful than severe ones.

Eye.- Upon the cornea mustard gas exerts an especially injurious action, particularly at the vertex. Within 10 to 15 minutes after exposure to dilute concentrations, degeneration or necrosis of the corneal surface may be demonstrated by the application of a 2 percent alkaline aqueous solution of fluorescein, the injured cells retaining a greenish fluorescent coloration. In more severe injuries the cornea may be killed throughout its entire thickness at the vertex. The mildest cases show a slight cloudiness; the severe cases present a characteristic porcelain appearance of bluish-white opalescent cloudiness, often with a more opaque band or line running horizontally across the cornea just below its transverse diameter. The injury to the conjunctiva is shown by the development of a more or less severe catarrhal, seropurulent or purulent conjunctivitis with marked edema of the subconjunctival tissues leading often to "ruffling" of the lids, entropion, ectropion, or a combination of these. Even the lighter cases tend to run a chronic course with disturbances and reduction of vision. In the severe cases cicatrization and vascularization of the cornea take place slowly with resulting impairment or loss of vision. The injured eye is more susceptible to infection; and in infected cases suppurative panophthalmitis may develop with complete destruction of the eyeball. Recovered cases of mild mustard-gas conjunctivitis often show an increased sensitivity to the action of light, dust, and other irritants, including mustard-gas fumes.

Respiratory tract.- Upon the mucosa of the respiratory tract mustard-gas vapor produces a local injury to the epithelium as shown by the development of a catarrhal, desquamative, membranous, diphtheritic or purulent inflammation (rhinitis, stomatitis, pharyngitis, laryngitis, tracheitis, and bronchitis), these lesions being most severe in the nose, back of tongue, palate, pharynx, and larynx, decreasing in intensity downward. Coryza, salivation, dryness of mouth and throat, aphonia, and persistent cough are the chief symptoms, with physical signs of laryngeal, tracheal and bronchial involvement, and atelectasis, emphysema, and edema of the lungs. As a result of secondai v in- fection a purulent bronchopneumonia may develop.

Gastrointestinal tract.- Through the swallowing of air, saliva, or secretions from the upper respiratory tract containing mustard gas, or from the ingestion of contaminated food, local corrosive action upon the alimentary mucosa may be produced, varying from a catarrhal inflammation to large areas of eschar formation with resulting gastric ulcer, perforation, etc. The symptomatology of the mildest lesions is covered up by that resulting from the more severe burns elsewhere; the more severe ones will produce marked symptoms referable to the stomach and intestines.

Susceptibility.-There exists a racial (whites more susceptible than negroes) and an individual susceptibility to the action of dichlorethylsulphide, particularly in the case of the skin, and probably also of the respiratory tract. The individual susceptibility, in some cases at least, is associated with the constitutional stigmata anti symptomatology of the thymicolymphatic constitution. Acquired susceptibility is not specific. Animals show also generic and individual differences in sensitivity to mustard gas.


Systemic action.- There is no evidence of any systemic poisoning by the absorption of dichlorethylsulphide from the skin, eyes, or mucous membranes of the respiratory or gastrointestinal tracts. There is no metastatic action of the gas from the site of local external application.

Shock.-In all severe cases of mustard-gas burns of skin, eyes, or mucous membranes there is usually the clinical picture of severe shock, in the form of intense pallor, depression of pulse and temperature, general collapse, nausea, and vomiting. The mildest cases show no systemic reaction.

Blood and urine.-No changes are observable in the blood or urine of mild cases. In cases with large infected burns of skin or respiratory tract the blood presents a mild secondary anemia, with leucocytosis; the blood urea is increased; the urine is diminished, concentrated, and contains casts and albuminuria. Under forced fluids the urinary symptoms improve and the blood urea diminishes. In severe infected cases the general picture may be that of a severe toxemia. In experimental animals, after more severe gassing with involvement of the respiratory tract, there occurs a distinct concentration of the blood, with a polycythemia of two to three million above the normal, and a corresponding leucocytosis.

Intravenous and subcutaneous injection.- When injected intravenously or subcutaneously dichlorethylsulphide is an active poison, causing death in one to four hours intravenously and two hours to three weeks after subcutaneous injections (for rabbits intravenous injections of 0.0075 c.c. per kilo may be lethal within four hours), according to size of dose, individual animals, etc. When death takes place quickly the symptoms are (chiefly those of an action upon the central nervous system, such as hyperexcitability, rapid respirations, general convulsions, opisthotonos, gradual failure of respiration and circulation, coma, and death. When the animal lives longer after small intravenous injections, or after subcutaneous injection, there develops a characteristic symptomatology of salivation, marked diarrhea, leucopenia, and fall of temperature, with marked anorexia, emaciation, and depression. With subcutaneous injections of 0.015 to 0.06 c.c. death usually takes place from the fourth to the tenth day.


The specific microscopic pathology of the local lesions of dichlorethylsulphide poisoning consists in degeneration and necrosis of the cells with which it comes in contact. The earliest microscopic change is pvenosis of the nucleus and cell body, followed by hydropic degeneration, liquefaction or coagulation necrosis. In the skin, hyperemia, with regeneration of the damaged cells, pigmentation, vesicle formation, desquamation of the dead epidermis or eschar formation mark varying stages of severity of the lesion. The degenerative changes extend deepest in the hair follicles and sweat glands. In mild burns without vesication the papillary layer of the corium may show a greater degree of necrosis than the epidermis itself, thus explaining the frequent occurrence of Nikolsky's sign. Large, heavily pigmented chromatophores may be the only living cells left in the papillary layer. In severe burns the necrosis may extend entirely through the corium. In the cornea, pycnosis and simple or coagulation necrosis of the corneal epithelium and interstitial substance, even to the endothelial layer, in extent varying with the degree of exposure, constitute the microscopic features. On the conjunctive the epithelium shows pycnosis, hydropic degeneration, liquefaction necrosis, or there may be a deeper necrosis


extending into the subconjunctival tissues. The conjunctival surface suffers to a less degree proportionately than the epidermis. On the mucous membranes the epithelium shows pycnosis,. hydropic or mucoid degeneration, desquamation. liquefaction or coagulation necrosis. The necrosis may extend into the submucosa, but the depth of the lesions on the conjunctivan and the mucous membranes of the respiratory tract is never so great from identical exposures as it is in the skin. Following the necrosis there is marked hyperemia, and the development of an edema, more marked in the subcutaneous and subconjunctival tissues in animals, but less marked in man. Human skin, however, shows a much greater tendency to vesication. The blood vessels in the necrotic area are killed, the blood cells hemolyzed to some extent without thrombus formation or much extravasation, except minute hemorrhages by diapedesis. Following the lesion there is a demarcating inflammation, with slow regeneration, repair, or cicatrization. The regeneration of the epidermis proceeds from the epithelium of the sweat and sebaceous glands. On the mucous membranes there results in the severe cases a localized eschar or ulcer, or a more diffuse diphtheritis. With secondary infection the inflammatory process becomes purulent or suppurative. The influence of secondary trauma and infection is well shown in the early development of deep areas of decubitus in the injured regions of the skin. Multiple furuncles may develop, or large cutaneous areas become gangrenous. In the eye purulent involvement of the anterior chamber, iris. and cillarv body may occur, or even a suppurative panophthalmitis. In the respiratory tract secondary infection of the injured mucosa may lead to a purulent bronchopneumonia.

The internal organs in animals with mustard gas lesions of the skin, eves. respiratory or gastrointestinal tract offer nothing of a specific pathologic nature. There is general congestion, marked splanchnic congestion, acute catarrh of the intestines, and, in infected cases, some cloudy swelling of the kidnevs.

In fatal cases the cause of death is to be found in shock, secondary infection with toxemia, or local conditions as laryngitis, tracheitis, bronchitis, and bronchopneumonia. It is also possible that the entrance into the body of shell fragments carrying liquid dichlorethylsulphide might cause a relatively speedy deathl through absorption.

At the site of subcutaneous injections there is found a local eschlar with demarcating hemorrhage, edema and inflammatory infiltration; in the large veins into which injections have been made, no changes have been found except occasional thrombosis.

The general pathology of the injected cases presents a specific pathologic picture in the intestinal tract in the form of a severe mucoid, desquamative or necrotic enteritis, the intestinal epithelium showing the most marked hvdropic or mucoid degeneration, even to liquefaction necrosis. Similar changes may be found in the epithelium of the bile-ducts. In a certain number of cases the spleen, lymph nodes, and hemal nodes show a marked hemosiderosis, the hemosiderin being contained in large hemophages. It is most probable that these evidences of increased hemolysis are explainable by the extravasations and blood destruction occurring at the site of the injection, or are simply coincident pathology due to some other cause as such hemosiderosis is not a rare finding in laboratory animals. Marked depletion of the bone marrow is produced by small doses intravenously. In the other organs no pathologic changes but congestion and edema have been found, with the rare exception of emboli or thrombi.



The cause of death in intravenous and subcutaneous injections would appear to be the direct action of minute quantities of free dichlorethylsulphide, or some poisonous product resulting from its decomposition, upon the cells of the central nervous system. It has been assumed that the pathologic action of dlichlorethylsulphide is due to its hydrolysis within the tissue cells. The products of this hydrolysis, hydrochloric acid and dihydroxyethylsulphide, when injected into the blood, do not produce the same effects. Dihydroxyethylsulphide and hydrochloric acid, when injected into the circulation in much larger doses than would result from the hydrolysis of the fatal doses of mustard gas, are harmless. The effect upon the cells of the central nervous system, however, may depend upon hydrolysis, with the liberation of hydrochloric acid (Marshall) in these cells, of minute quantities of mustard gas from the circulation, or these cells may be injured without such hydrolysis occurring. It is probable that the gastrointestinal catarrh resulting from the injections of dichlorethysulpbide is secondary to the nervous injury, rather than to an excretion of the poison or poisonous products through this tract, although this point remains unsettled. No positive tests for dichlorethylsulphide or dihydroxyethylsulphide have been obtained in the bile, intestinal contents, or urine. Investigations, therefore, failed to confirm Marshall's statement that dihydroxyethylsulphide is excreted in the urine in mustard gas poisoning. Incidentally, it has been shown that the platinic chloride-sodium iodide color test for dichlorethylsulphide is not applicable to the body fluids or extracts of various organs and tissues, as similar color changes are produced by some of these.


(1) Meyer, Victor: Ueber Thiodiglycol Verbindungen. Berichte der deutschen chemischer Gesellschaft, Berlin, 1886, xix, No. 3, 3259.
(2) Guthrie, F.: On Some Derivatives from the Olefines. Journal of the Chemical Society of London, 1860, xii, 109. Also: Ueber einige Derivate der Kohlenwasserstoffe CnHn. Annalen der Chemie und Pharmacie, Leipsig und Heidelberg, il. s. xxxvii, Part 3, 1860, 266.
(3) Niemann, A.: Ueber die Einwirkung des Braunen Chlorschwefels auf Elaylgas. Annalen der Chemie und Pharmacie, Leipsig, n. s. xxxvii, Part 3,1860, 288.
(4) Meyer, Victor: Berichte der deutschen chemischer Gesellschaft, Berlin, 1887, xx, No. 2, 1729.
5) Leber, T.: Die Entstehung der Entzündung und die Wirkung der entzündungserregenden Schadlichkeiten, Leipsig, Wilhelm Engelmann, 1891, 338.
(6) Clarke, H. T.: Journal of the Chemical Society of London, 1912, ci, No. 2, 1583.
(7) Monthly Summary of Information Gas Warfare (British) No. 1, July, 1917, S. S. 184. On file, Medical Division, Chemical Warfare Service.
(8) Haldane, J. S.: Reports of the British Chemical Warfare Service Medical Committee, January, 1918. On file, Technical Division, Chemical Warfare Service.
(9) Pissarello, C.: Relazione sugli effetti morbosi delle granate cariche ad "Yprite." Giornale di Medicina Militare, Roma, 1918, 1xvi, 128.
(10) Pappemheimer, A. M.: The Effects of Intravenous Injections of Dichlorethylsulphide in Rabbits with Special Reference to the Blood and Haematopoietic Tissues. Proceedings of the Society for Experimental Biology and Medicine, New York, 1919, xvi, 92. (Meeting of March 19, 1919.)
(11) Monograph No. 1, Dichlorethyl and Homologues. Published in Bulletin No. v, prepared by Medical Division, Chemical Warfare Service, Issued by the Director, Chemical Warfare Service, Washington, D. C., August 30, 1918, 348. On file, Medical Division, Chemical Warfare Service.