There is no mention whatsoever of scurvy in animals previous to 1895, when Theobald Smith wrote: “When guinea-pigs are fed with cereal (it has been observed for some years in this laboratory), with bran and oats mixed, without any grass, clover, or succulent vegetable, such as cabbage, a peculiar disease, chiefly recognizable by subcutaneous extravasations of blood, carries them off in from four to eight weeks.” Smith did not pursue the subject further.

Coincident with the appreciation of the value of experimental methods, one would have expected attempts to produce scurvy in animals. The disorder had been recognized for a great many years, the fact that it was occasioned by a dietary deficient in fresh food was widely known, so that it would seem natural for clinicians or laboratory workers to have tried to induce the disease in animals by a similar restricted diet. There is, however, no suggestion of such an experiment in the voluminous literature on scurvy. The explanation of the neglect of an experimental study of scurvy appears to be that this was considered a disorder which required little investigation—its etiology, its prevention and its cure seemed thoroughly understood. It is strange that the mere fact that such a severe disturbance responded, as if by magic, to the administration of simple foodstuffs should not have awakened curiosity as well as wonder, and stimulated inquiry into the nature and significance of the curative agent.

In 1903 Bolle published the results of experiments on guinea-pigs which had been fed for various periods on raw or sterilized milk. He reported that the animals developed a marked fragility of the bones, varying in intensity in proportion to the degree of heat to which the milk had been subjected. These changes Bolle interpreted as typical scurvy, but it is difficult to decide whether the disorder was actually true scurvy, in view of the indefinite description of the bone lesions.

Bartenstein repeated Bolle’s work and described in detail the changes in the bones. From his work we learned that it was of little moment whether the animals were fed raw or sterilized milk—their nutrition failed within about the same length of time, and they died within a month. The chief pathologic change was a marked fragility of the bones, leading to spontaneous fractures, or to breaking of the bones in the course of ordinary manipulation. Bartenstein described the disorder as “beginning with the degeneration of the marrow, and secondarily leading to an atrophy of the bone due to increased absorption and deficient new formation of bone, especially at the endochondral lines of ossification. As the result of the osteoporosis, spontaneous fractures occur without noticeable hemorrhagic diathesis.” He found a deficiency of calcium in the bones of animals suffering from even a mild degree of this disorder. Hart and Lessing are of the opinion that we cannot accept Bartenstein’s diagnosis of scurvy. They compare this disorder to that described by Ziegler as “osteotabes infantum,” in which there is a more or less pronounced “jelly marrow” which has replaced the lymphoid cells in scattered areas. They sum up their analysis of the question with the statement that “the primary marrow changes in the diaphysis, the marked lacunar absorption by osteoclasts, the absence of subperiosteal hemorrhages, do not indicate typical scurvy.” Schmorl produced comparable lesions in dogs and came to the conclusion that the disorder was one “very similar” to scurvy but not identical with it, the essential difference being that in all animals there was a most marked absorption of bone by osteoclasts.

Similar results were obtained by others. In 1904 Peiper and Eichloff attempted to produce scurvy in dogs by means of a diet composed exclusively of raw or of sterilized milk. The following year we find a report by Esser of feeding goats on sterilized milk. None of these studies hits the mark; a condition of fragilitas ossium was produced rather than typical scurvy.

In 1907 the first systematic and convincing experimental study of scurvy appeared. In this year Holst and Froelich published a preliminary account of their classic work, undertaken in an endeavor to fathom the nature of “ship beriberi,” a disorder which disabled so many of the sailors in the Norwegian navy. We must remember that until recently attention has been focussed on the infectious diseases, and little thought given to the investigation of nutritional disorders. Although experimental investigation in this field may be stated to have been initiated by this work, it in turn had been stimulated by a conception of dietary diseases which had just begun to be realized. This new viewpoint was the result of the work of Eijkman. As has been mentioned in considering the pathogenesis of human scurvy, Eijkman demonstrated that hens developed polyneuritis, a disease resembling beriberi, when fed on polished rice, and that the simple change to a diet of unpolished rice, or the addition of rice polishings to the dietary, sufficed to protect or to cure. This work had been repeated and substantiated. It was evident to those who considered the question that the old lines of thought in regard to food and dietetics inadequately explained remarkable phenomena of this kind. It was also clear that nutritional disorders were subject to experimental investigation in a way similar to the infectious diseases, which were absorbing the interest of the various laboratories. It is unnecessary to review the rapid rise of investigative work in this field. It may be stated, in general, that during the past few years the investigations of scurvy have become so numerous that an established research technic may be said to have developed in connection with its study.

Attempts have been made to produce scurvy in various animals, but it has been brought about regularly only in the guinea-pig and in the monkey. Experiments on the dog are as yet too few to justify our regarding this animal as definitely available for this study. The nutrition of rats, mice, rabbits, hogs, pigeons, fowl, receiving a diet identical with that which regularly occasions scurvy in guinea-pigs, either progresses normally or a state of malnutrition develops which cannot be identified with scurvy. Recently Hart, Steenbock and Smith have reported that cattle can be reared to maturity on a diet which will produce scurvy in the guinea-pig in four to five weeks. We are therefore confronted with the remarkable and inexplicable phenomenon, a fact concurred in by almost all investigators, that a diet induces either normal nutrition or malnutrition, according to the experimental animal employed. A diet of polished rice, or other decorticated grain, will lead to the development of scurvy in the guinea-pig, to polyneuritis in the pigeon or fowl, or, according to Holst and Froelich, to a combination of these disorders in the hog.35 The basis of these divergent results cannot be surmised, and is worthy of the most thorough investigation. It may well be that the elucidation of this problem—for example, why we are able to bring about scurvy in the guinea-pig but not in the rabbit, will shed light on the pathogenesis of this disorder. Are we to conclude that some of these animals have the power to synthesize the vitamine whereas others must depend upon the food for it, or are we to presuppose an ability to maintain normal life and function without any or with a minimal amount of this vitamine? The difference between these two groups, the susceptible and non-susceptible animals, probably is not as absolute as we have been wont to regard it. Recently Harden and Zilva have shown that although rats are able to thrive on a diet free from antiscorbutic vitamine, they show an appreciable gain in weight when this factor is added to the dietary. If such be proved to be the case, we must regard the non-susceptibility of the rat, the rabbit, etc., as relative rather than absolute.

From time to time a doubt has been raised as to whether we should accept guinea-pig scurvy as the counterpart of human scurvy. This question can be answered only by comparing the disorder in the one species with that in the other—as to mode of production, pathology, symptomatology, means of cure and all other phenomena. Viewed from these standpoints it is found that in almost every respect the disorder is identical in man and in the guinea-pig. The outstanding distinction is the difference in the length of time elapsing before the development of symptoms. In the child or in the adult it takes about six months of the deficient diet before clinical symptoms are manifest and a diagnosis can be established; in the guinea-pig the disorder can be recognized two weeks after restricting the diet. In the one instance we seem to be dealing with a nutritional disorder which is chronic or at least subacute, and in the other with a markedly acute condition. This distinction is open, however, to certain qualifications. In the first place, we must consider the duration of life of the two species, the comparatively short span of the guinea-pig compared with that of man. It must be borne in mind, furthermore, that the guinea-pig is placed on a diet absolutely devoid of all antiscorbutic vitamine, whereas this rarely obtains in human beings. For example, the diet which is most markedly scorbutic for infants is the “malt soup” previously mentioned, but even this food contains an amount of the antiscorbutic factor which is not negligible. But after taking these differences into consideration, it is nevertheless evident that the guinea-pig is far more sensitive to scurvy than man. This does not indicate that the guinea-pig is an unsuitable experimental animal, any more than the fact that the pigeon is more susceptible to polyneuritis than man indicates that it is unsuited to investigations of beriberi. It merely prevents our carrying out delicate quantitative experiments, and cautions against drawing too finely-spun deductions. In all nutritional investigations it should never be forgotten that conclusions drawn from experiments on animals are merely provisional, and must await substantiation on man, and, furthermore, that where differences in reaction are noted, the clinical data should be accorded full consideration.

Pathogenesis of Guinea-pig Scurvy.—From a pathogenetic point of view guinea-pig scurvy and human scurvy show remarkable points in common. Any diet that leads to the development of scurvy in man likewise brings it about in the guinea-pig, and contrariwise, any food which cures the disorder has the same beneficent effect on both species. This similarity extends so far that, as will be shown in the chapter on antiscorbutics, the relative potency of the various foods is approximately the same for man and for the guinea-pig. The parallelism generally is striking. The dietary which has been commonly employed in experimental scurvy has been that first suggested by Holst and Froelich, namely, oats, hay and water. Recently, however, this dietary has been enlarged in order to make it more complete, so as to include adequate protein, water-soluble and fat-soluble vitamine, and inorganic salts. To this end the group of workers at the Lister Institute (Chick and co-workers, Harden and Zilva) place their animals on a basal diet of one part of crushed oats and two of wheaten bran, and a daily ration of 50 to 60 c.c. of milk autoclaved for one hour at 120° C. This milk still retains a small amount of antiscorbutic vitamine. Cohen and Mendel have employed, apparently with good result, a “soy bean cracker,” containing soy bean flour which has been heated for 30 minutes at 15 pounds’ pressure (120° C.), 3 per cent. of sodium chloride, the same percentage of calcium lactate and of dried brewers’ yeast, and raw milk sufficient to supply 5 per cent. of butter-fat.

The use of raw milk was introduced by Jackson and Moore, and adopted by McCollum and Pitz and again by Pitz in a series of interesting experiments. The milk was given ad libitum. The results of these investigations were puzzling at first, until it was shown by Chick, Hume and Skelton that the dietary on which they were based contained a fundamental error which accounted for their lack of consistence. Although milk is not rich in the antiscorbutic factor, it possesses it in moderate degree, so that the outcome will be quite different according to whether a guinea-pig takes, for example, 50 c.c. or 100 c.c. daily. It is quite evident, therefore, that a food of this kind cannot be offered ad libitum, and that if this rule is not observed, most disconcerting results will follow.36 This basic error in framing the dietary has made it impossible to accept the deductions of these authors. The conclusions of McCollum and Pitz are so striking and have led to such wide discussion, that they require consideration, in spite of the fact that the error in the dietary is now recognized. These authors found that the cÆcum of their animals was greatly distended with putrefying fÆces. As the cÆcum is extraordinarily large and delicate in this species, they drew the deduction that the development of scurvy in the guinea-pig was due principally to the retention of fÆces. “An impacted cÆcum, the seat of putrefaction, may cause injury to the cÆcal wall, sufficient to permit the invasion of the tissues by bacteria, or the animals may perhaps be injured primarily by the absorption of toxic products of bacterial origin.” Accessory dietary factors or vitamines, according to this theory, are supposed to play no part in the disorder, and antiscorbutics, such as orange juice, are considered to owe their efficacy mainly to their laxative properties, and to be replaceable by other laxatives such as phenolphthalein or oleum petrolatum. The efficacy of orange juice is supposed to be due to its content of citrates, and to be replaceable by what was termed “artificial orange juice,” a mixture composed of the various salts, citric acid and sucrose, in the proportions in which they are found in the natural juice.

These reports stimulated renewed interest in experimental scurvy, suggesting new aspects, and therefore directing attention to points requiring investigation. There were no data at hand on the consistency of the contents of the bowel in guinea-pigs, so that it was necessary to make appropriate observations in normal and in scorbutic animals. It was soon reported by various workers, Rappleye, Cohen and Mendel, Hess and Unger, and others, that there was no definite relationship between the occurrence of scurvy and impaction of the cÆcum. In Fig. 8 we see portrayed the stool output of a guinea-pig during the period in which it was developing scurvy, and during a subsequent period when it was being cured by means of orange juice. It is clear that there was no significant variation in the output during these divergent periods. A similar conclusion was arrived at by those who examined the cÆcums of animals postmortem. Cohen and Mendel write: “Summarizing our experience with nearly one hundred scorbutic animals, we conclude that actual impaction of the fÆces in the cÆcum occurred in about one-quarter of the cases, and visible damage to wall, i.e., congestion or hemorrhage, or impaction, or both, was found in perhaps half of the cases. It should be noted that this statement covers all the diets we have tried.” We concur in this conclusion. Not infrequently we found the cÆcums of markedly scorbutic pigs to contain semi-fluid fÆces, the consistency of its contents depending on the character of the diet, quite apart from its adequacy and lack of the antiscorbutic factor. For example, a diet rich in milk, containing 100 c.c. or more, led to the formation of rather solid fÆces; if oats were added to the milk diet the fÆces in the cÆcum were found to be still more solid, and this portion of the gut more often impacted. This condition could be detected by palpation even during life. Just as we encountered scorbutic animals on an oat, hay and water diet, who had semi-fluid fÆces in the cÆcums, so we met with others which were on a milk diet, and showed no signs of scurvy, although their cÆcums were impacted with fÆces of a putty-like consistence. Guinea-pigs do not seem to be able to tolerate a diet containing a large quantity of the fat of cow’s milk. Such a diet leads to impaction of the large intestine resulting in death, but does not induce scurvy. Jackson and Moore produced a condition of this kind by feeding pigs with cream containing 26 to 28 per cent. fat. “In every case,” they write, “the large intestine was distended with light mustard-colored semi-solid fÆces.” This pathological condition is not understood, but is quite distinct from scurvy, and remarkable in view of the fact that the milk of the guinea-pig contains as high as 25 per cent. of fat. It is an interesting illustration of the marked biological difference in the butter-fat of various species.

Nor was it found, as McCollum and Pitz claimed, that antiscorbutics were replaceable by laxatives in the diet. Without entering into the details of this aspect of the subject, which will be considered in the chapter on symptomatology, it may be stated that attempts to prevent the occurrence of scurvy or to cure it by means of laxatives invariably failed. Chick, Hume and Skelton, as well as Hess and Unger, gave oleum petrolatum to a series of pigs without the slightest favorable effect. Hart, Steenbock and Smith recently reported that they had administered 1 c.c. of this oil on alternate days to one series of pigs, and 2 mg. of phenolphthalein on alternate days to another series, without relieving the scurvy. Cohen and Mendel, in order to test the adequacy of their diet as to roughage, supplemented it with additions of considerable filter paper and sawdust, “without averting the appearance of scurvy.” It is evident, therefore, that constipation does not play an essential rÔle in the pathogenesis of scurvy in the guinea-pig, and that antiscorbutics are by no means synonymous or interchangeable with laxatives. These results accord with observations on infantile scurvy.

The Pathology.—The pathology of guinea-pig scurvy is essentially that of human scurvy. Hemorrhages and separations of the epiphyses or fractures of the long bones dominate the macroscopic picture. The hemorrhages are found rarely in the gums, but are common about the joints, in the muscles of the jaw or in those of the hind legs. They may be subcutaneous and appear as bluish discolorations at various parts of the body, especially if the disorder has resulted in death or if infection has been superadded. On stripping the skin we often note hemorrhages in the intercostal muscles, and beading of the ribs at the site of the costochondral junctions, least marked in the upper and in the floating ribs. This has been frequently described in connection with guinea-pig scurvy; its similarity to the rosary of human rickets has been drawn attention to recently by Jackson and Moore. It should not be regarded as “pseudo-rhachitic,” but as typically scorbutic, from a microscopic as well as a macroscopic viewpoint. On closer examination a yellowish-white transverse line may be seen at the epiphyseal junction of the ribs, and frequently some subperiosteal hemorrhage. Beading of this character has been reported by Ingier also in the “snuffles” of hogs and as the result of trauma. A similar enlargement of the chondral junctions of the sternum may be found on examining its posterior surface.

The joints of the body always show some changes. The epiphyses are enlarged to a variable degree, resembling the epiphyses typical of human rickets. As in infants, this change is met with most commonly at the wrist joint, involving the ulna and the radius. The knee-joint is likewise often involved, especially the ends of the tibia; the elbow, ankle, and the shoulder may also show an articular swelling of the bones. About these joints hemorrhages in the subcutaneous tissues may be seen, or edema extending along the extremities. Not infrequently a fracture of one of the long bones is found, which may or may not have been diagnosed during life. The common site of fracture is the lower part of the tibia or fibula. Much more frequent than fractures are separations of the epiphyses, which long since have been recognized as typical of infantile scurvy. Even gentle handling in the course of performing the necropsy may occasion a lesion of this kind—of the tibia at the knee, of the radius or ulna at the wrist, or indeed at any of the epiphyseal junctions. The infracted ends occasionally may be seen held insecurely by a delicate band of periosteum. The shafts of the bones are brittle, rarefied, and easily broken.

On opening the chest, slight hemorrhages may be noted in the pericardium and in the visceral and costal pleurÆ. The heart is frequently enlarged, and the pericardial sac contains an excess of serum; the right ventricle, however, is not found disproportionately hypertrophied. Pneumonia is met with very frequently and constitutes a common terminal infection.

On opening the abdomen we may note subperitoneal hemorrhages of the muscular wall or of the coils of intestine. The liver and spleen are generally normal, as is the pancreas. The kidney frequently shows minute hemorrhages beneath the capsule and on section.

The adrenals not infrequently are large. This fact was first brought out by Rondoni and Montagnani, and is of added interest in view of its confirmation by McCarrison as well as by LaMer and Campbell. Its significance is discussed at length in the consideration of the adrenal gland in human scurvy. In view of the fact that fasting leads to a similar hypertrophy, and that guinea-pig scurvy frequently is complicated by fasting, these observations should be extended. In all investigations of this kind, bacterial cultures should be taken of the adrenals, or other organs, to be certain that there is no complicating infection.

There have been no reports as to the effect of scurvy on the size of the testicles, ovary, thymus, thyroid or parathyroid glands in the guinea-pig, such as have been made by McCarrison in relation to avian polyneuritis.

The alimentary tract occasionally shows macroscopic changes similar to the lesions found in man. The entire canal is frequently very empty, especially the stomach and the small intestine, due to the lack of appetite for some days previous to death. In the stomach we may find areas of congestion or numerous small superficial ulcers surrounded by congestion and covered with mucus; occasionally these ulcers are somewhat larger and deeper. The larger ulcers are more frequently situated in the first part of the duodenum, often proximal to the papilla of Vater. Holst and Froelich described this lesion in one of their early communications. More common than ulceration of the duodenum is a diffuse congestion of its upper part. This lesion is of note because it has been described frequently in the protocols of human necropsies, and is found in beriberi, in avian polyneuritis, and in pellagra in man. Its significance is unknown, and has indeed never been discussed.

Lower down in the bowel there are occasionally areas of congestion and ulceration, but no section seems particularly predisposed to these lesions. The contents of the bowel, especially of the cÆcum, as mentioned above, depend more on the character of the food than on the existence of scurvy. If the diet has consisted of milk and oats, the cÆcum will be found full and perhaps impacted, whereas if hay and oats have been fed, the cÆcum will be less full and its contents semisolid.

In view of the fact that many of the animals have taken very little food for some days previous to their death, it will be well to describe briefly the macroscopic picture of simple starvation in guinea-pigs. When guinea-pigs are given only water they live about one week; if orange juice is added to this water diet they succumb a little later to starvation. Under all these conditions the striking pathologic change—absent in scurvy—is edema. It is true that the limbs may show slight edema in scurvy, and that the pericardial and the pleural sacs, and even the peritoneal cavity, occasionally contain a small quantity of serum, but it is comparatively an insignificant amount. Moreover it is difficult to decide to what extent this edema is due to scurvy, and to what extent to starvation. In typical starvation, on the other hand, such as occurs on the limited diets enumerated above, we find marked subcutaneous edema, sometimes a true anasarca, and frequently also ascites. We are reminded of the “war edema” and its frequent association with starvation. Another distinction between the two conditions is the fact that the marrow in starvation is yellow and not red as in scurvy. In passing, it may be mentioned that the ascites was greater when orange juice had been given than where the animal received only water.

In perusing the literature but one study has been noted on the effect of a scorbutic diet on the foetus. This investigation was carried out on a large series of guinea-pigs by Ingier (1915). The following comprise her conclusions:

“1. Pronounced cases of Barlow’s disease may be produced in the foetus as early as ten to fifteen days after the commencement of dieting pregnant guinea-pigs with oats and water. There are wide individual variations. The scorbutic changes in the skeleton are greatest in the earlier embryonic stages. The foetuses of that period, with practically no exceptions, die and show marked traces of impeded growth.

“2. Foetuses from the later period of pregnancy are born alive, and apparently fully developed, with comparatively slight changes in the osseous system.

“3. Even a short extension of the period of extra-uterine dieting on milk from scorbutic mothers, and later on oats and water, is sufficient to change the latent scurvy into a highly-pronounced case.

“4. The foetus cannot be kept alive longer than the adult animal, about twenty-eight days, either by intra-uterine dieting alone or by combined intra- and extra-uterine dieting.

“5. The mothers show signs of the disease at an early period and are more severely attacked than non-pregnant animals. Death also occurs comparatively often in the first period of gestation.”

In these experiments intra-uterine fractures, premature births and still-born litters are frequently mentioned. This suggests naturally an inquiry as to whether similar occurrences have been observed in human scurvy. In considering the pathogenesis of human scurvy, we have remarked on the meagreness of the data on this most interesting aspect of this disorder. In view of the similarity between human and guinea-pig scurvy, we should expect not only miscarriages and still-births to result, but cases of congenital scurvy, especially of the latent or rudimentary type.

Scurvy Has Been Induced Also in the Monkey.—The most extensive investigation of this kind is that of Hart and Lessing, who brought about scurvy in several young Macaci Rhesi which had been fed for some months on boiled condensed milk with the addition of cooked rice and pig-nuts. Typical scurvy resulted in all but one instance, and in this case an old animal was used and a different pathological picture developed. These experiments require brief reviewing, as the clinical signs and gross pathology in monkeys differ somewhat from those encountered in guinea-pigs, and particularly as these distinctions render scurvy in monkeys and in infants practically identical.

Apart from general listlessness and lack of activity, the first sign is hemorrhage of the gums, the characteristic sign of infantile scurvy. This is stated to be constant; in the guinea-pig it is most exceptional. Another sign characteristic of infantile scurvy is the subperiosteal hemorrhage, which in the guinea-pig may be found as a small effusion near one of the larger joints or the costochondral junctions, but in the monkey consists of large effusions, similar to those described by Barlow. Subperiosteal hemorrhages of the cranial bones were constantly seen, and not infrequently involvement of the scapula and of the maxilla. Hemorrhage into the orbit leading to exophthalmos and to hemorrhagic discoloration of the upper lid—a lesion not infrequent in human scurvy—is also described; in guinea-pigs we have met with exophthalmos only in two instances. In all other respects scurvy in the monkey resembles that in the guinea-pig, even to the extent of the scorbutic rosary of the lower true ribs. Hart and Lessing describe the presence of the “white line” of Fraenkel, which is referred to at length in treating of the symptomatology of human scurvy. This is a shadow, seen by means of the X-ray, traversing the long bones near their epiphyses—a definite “shadow band” associated with a narrow light zone lying just beneath it. This “white line” has been noted likewise by Talbot, Dodd and Peterson in the scurvy of monkeys.

Microscopic Pathology.—Turning to the microscopic pathology, we find that the changes are similar to those described elsewhere in connection with human scurvy. It will be unnecessary, therefore, to give more than a general survey of the typical alterations. Little has been added to the description of lesions so carefully depicted in the first report of Holst and Froelich (1907). The bone marrow at the ends of the diaphyses in proximity to the epiphyseal junction loses its normal lymphoid character and is replaced by a reticular or fibrillated substance, the so-called “framework” marrow (Geruestmark of Schoedel and Nauwerk) containing a homogeneous mucoid tissue and only a few osteoblasts and marrow cells. The number of blood-vessels is considerably reduced and fresh hemorrhage or blood pigment is frequently seen. The osseous tissue itself shows marked changes, corresponding to the rarification and brittleness noted on gross examination. The osseous trabeculÆ are fewer in number and those which remain are slender and irregular, and frequently appear as isolated islets. The cortical substance also becomes very thin. There are marked alterations in the intermediate cartilages, especially of the ribs.37 Instead of the cells being arranged in orderly rows, they are irregularly placed, and frequently greatly reduced in size and number. The bone trabeculÆ on which they abut are not well formed or of equal length, and do not present an even and transverse plane, but are misshapen, small, so that the line of junction with the cartilage is zigzag. In cases of marked scurvy the junction may be entirely disorganized and deformed, showing fractures of the rarefied bone and hemorrhages in the neighborhood. This leads frequently to a macroscopic deformity of bone, a bulging of the surface at the costochondral junction—the scorbutic “beading” or “rosary” of the ribs. Recently Delf and Tozer have described these changes, classifying them as those occurring in “incipient,” “definite,” “acute,” “chronic definite,” and “chronic acute” scurvy. Fig. 9 shows these types in diagrammatic form. In this figure we see how varied may be the manifestations of this nutritional disorder, according to whether it has existed for a shorter or a longer time; in some instances the picture is very puzzling. The “incipient scurvy” corresponds to what we have termed clinically “latent scurvy” in infants, a condition which cannot be diagnosed and is manifested merely by a faulty nutrition which responds promptly to the addition of an antiscorbutic to the diet. There have been no histologic examinations in man at this incipient stage, but we may infer that they are similar to the bony changes found in the guinea-pig. When the scurvy has become chronic in an animal which has lived for months on a quota of antiscorbutic food sufficient to preserve life but insufficient to prevent the development of scurvy, we find a microscopic picture at the costochondral junction differing widely from that seen in the acute stage. Not only are the columns of cartilage cells represented merely by misshapen vestiges, but an ossified band (Fig. 9⁵) is seen at the junction. It is probable that this is frequently the cause of the marked cessation of growth which has been described in connection with this type of scurvy in infants. Delf and Tozer interpret this ossified band at the junction as “an attempt to strengthen the junction in an abnormal manner, the normal process having broken down.” If the animal is again deprived of antiscorbutic food the ossified band breaks down, the junction becomes deformed and disorganized, and a condition of “chronic scurvy (acute)” is stated to have developed.

It has been noted by many investigators that changes in the teeth take place in the course of scurvy. They become somewhat yellow and lose their glistening appearance, and occasionally break off. The molars commonly become loosened, so that they can readily be removed from their alveolar sockets; less frequently this is true of the incisors. Until recently, however, this subject has not been studied in detail, and no significant histologic changes in the teeth have been described. Jackson and Moore showed that with marked changes in the teeth there was often “great dilatation of the veins in the pulp attended by more or less hemorrhage into the pulp,” and that “in guinea-pigs fed on oats and hay there was almost complete necrosis of the pulp of the incisor teeth, also more or less necrosis in the pulp of the molars.”

Recently an intensive study of this subject has been made by Zilva and Wells, which is of special interest because it describes the first beginnings of these lesions, and particularly because we have no knowledge whatsoever of the dental changes which occur in human scurvy. These investigators found a fibroid degeneration of the pulp of the teeth, a pulpar fibrosis. “It is clear at once,” they write, “and it is an important fact that no trace of cellular organization, no trace of cell nuclei, no trace of interstitial cement substances can be found anywhere. Nerves, cells, blood-vessels, and odontoblasts have all shared the process of fibrication and are no longer recognizable.” These radical changes in the teeth, brought about by a deficiency of antiscorbutic vitamine, were demonstrated not only in guinea-pigs but also in monkeys. In some instances they were found where a histologic examination of the costochondral junctions showed nothing abnormal. “Profound changes were recorded where the scorbutic changes during life were so slight as to be almost unrecognizable,” and, they continue, “the mildest degree of scurvy which could just be discovered at the postmortem examination produced well-defined changes in the structure of the teeth.” If this work is confirmed, we must consider the teeth as one of the first tissues of the body to be affected by scurvy. The authors quite rightly raise the question whether the teeth of young children may not likewise be injured by a deficiency of antiscorbutic vitamine, whether this may not play a rÔle in the dental caries so prevalent among civilized communities. It is evident, they state, that such transient conditions of infantile scurvy as have been described by Hess as “subacute” or “latent” scurvy, may occur more often than is usually suspected, and may reasonably be expected to influence dentition. It seems quite possible that the caries of the permanent teeth is due not only to infantile rickets but also to infantile scurvy.

Besides the typical histologic changes in the bones there are alterations in other organs which require mention. All investigators have found a degeneration of the muscles, showing a loss of their striations, swelling of the fibres, and the presence of irregularly-distributed vacuoles and granules. The interstitial tissue frequently is permeated with edema, as we should expect from gross appearances. Holst and Froelich have reported a fatty degeneration of the heart muscle, as well as of the epithelium of the mucous membrane of the glands of the stomach and of the intestine. Hart and Lessing, in their protocols of necropsies on monkeys, describe an interesting lesion associated with the degenerated muscle fibres—a collection of granules staining deep blue with hÆmatoxylin and dissolving on the addition of acid. These granules, interpreted as being composed of calcium, were found in the muscles of the limbs, of the tongue, and in the heart. It is reasonable to attribute their formation to an absorption of bone throughout the body. Similar calcium deposits were seen frequently in the adrenal glands, in their cortex, or at the border of the cortex and medulla. This lesion gains special interest in view of the calcium deposits described so frequently in connection with mercurial poisoning, more particularly as the symptoms of scurvy and of this toxic condition have marked clinical resemblances.

There has been but little histologic investigation of the nerves in experimental scurvy. In fact, the only systematic study of the kind is that of Holst and Froelich, whose attention was drawn to this field in an attempt to solve the relationship between scurvy and ship beriberi. These writers found a true Wallerian polyneuritis in only two pigs, one of which had been fed on wheat bread made with yeast, and the other on decorticated barley. In many instances, however, there was extensive degeneration of the axis cylinders without degeneration of the sheaths. They do not, however, attribute great importance to these changes, as the same lesions were found in the nerves of animals fed on cabbage and fresh potatoes. In view of the confusing reports on the nerves of birds in experiments on polyneuritis, one cannot be too careful in drawing conclusions from histologic studies of this kind.

In the study by Jackson and Moore on experimental scurvy in guinea-pigs, the histology of the blood-vessels is carefully considered. “Marked thinning of the wall” was found and depicted; “the wall as a whole had partially melted away, leaving few traces.” These parts of the wall contained many small round bodies resembling cocci, which were stained a deep blue by the Wright and the Giemsa methods. These bodies were present also in the lumen of the vessel and in the inner layers of the more normal portions of the wall. In addition to such changes in the veins, “lesions having the shape, location, and characteristics of infarcts, were found in the ends of the diaphyses of the long bones.” As a result of this pathologic picture the authors are of the opinion that they may have been dealing with a mild infection. This is quite possible, as scurvy tends to render the tissues less resistant to the entrance of bacteria. We believe, however, that even if such were the case, the phenomenon must be regarded merely as secondary in its relation to the pathogenesis of scurvy.

Following the study on the pathology of experimental scurvy, Jackson and Moore undertook to determine primarily whether the small stained bodies seen in the sections of the scurvy lesions were bacteria. This investigation has been cited frequently as presenting cogent evidence in favor of the infectious nature of scurvy, so that it will be necessary to consider it fully; the general question of whether scurvy is a bacterial infection is discussed under the consideration of etiology.

As is well known, Morpurgo, a generation ago, claimed to have produced rhachitic lesions in young rats by means of artificial infection with a gram-positive diplococcus. Pappenheimer brought about similar lesions in rats by the injection of a suspension of bone marrow from a rhachitic animal. Koch injected a streptococcus longus intravenously into young dogs, occasioning gross bony changes of the epiphyses and costochondral junctions, and microscopic changes resembling scurvy—an irregular line of ossification and “a framework” marrow, which, however, showed regions of osteoid. Jackson and Moody were able to isolate from the crushed tissue of their guinea-pigs “a diplococcus of low virulence with a tendency to form chains and produce green (color) on blood agar.” Pure strains of these organisms inoculated into the circulation of guinea-pigs and rabbits, living under ordinary conditions (a mixed diet consisting of green vegetables, hay and oats), gave rise in most instances to hemorrhagic and other lesions in the bones, joints, muscles, lymph-glands or gums. Hemorrhages were found beneath the periosteum in the region of the lower incisor teeth and the acetabulum and ribs. These results are far from constituting evidence in favor of the microbic origin of scurvy. They show merely that the tissues of scorbutic animals frequently harbor bacteria, and that injections of these bacteria will bring about hemorrhages which may be subperiosteal in character. They are open to the specific criticism that scurvy was produced readily in the rabbit, an animal which otherwise does not develop scurvy, and, furthermore, they differ from feeding experiments in inciting scurvy notwithstanding the fact that the animals were receiving an antiscorbutic diet (green vegetables). Cultures of the hearts’ blood of the affected animals were sterile in every instance; a result obtained likewise by Holst and Froelich.

Further studies of this kind should be carried out and should include cultures of the blood and tissues of guinea-pigs in the various stages of scurvy, especially the early stage. In addition, a histologic study should be made of the bones of animals injected with bacteria (preferably streptococci), in order to ascertain whether notwithstanding an unrestricted diet, typical lesions can be produced by this means.

For further details of the pathology of scurvy, the reader is referred to the chapter on human pathology.

Symptoms.—Let us consider the symptomatology of guinea-pig scurvy. In the course of an observation of many hundreds of animals we have been struck by the striking uniformity of the signs and symptoms. The animals made use of were almost invariably of moderate size, weighing from 200 to 300 grams. Where heavier pigs were employed the disease progressed less rapidly, but the signs were the same; they were, however, more difficult to elicit, owing to the subcutaneous fat. Most of the animals were on a diet of hay, oats and water ad libitum, but there was no variation in symptoms where fat and fat-soluble vitamine were supplied by an addition of egg yolk or of cod liver oil, or where egg albumen was fed to render the protein adequate, or where the inorganic salts were supplemented by additions of sodium or calcium chloride.

There is a variability in the sign which signalizes the onset of the disorder—sometimes it consists of a flattening of the weight curve, at others of an inordinate excitability of the animal, or frequently of a tender joint, generally a wrist. The joints almost invariably become tender early in the disease, causing the animal to wince and cry when it is examined. Accompanying this tenderness there is often slight swelling due to edema, or perhaps some hemorrhage, which alters the sharp, clean-cut contour of the joint. This edema may extend upward along the tendon sheaths. Soon the animal becomes lethargic rather than nervously active, and may look ill, as manifested by a roughness of its coat and its unnatural posture. Frequently it sits on three legs with the tender hind leg drawn upward and outward so as to escape pressure—a posture termed by Chick, Hume and Skelton (1918, 2) “the scurvy position,” and indicative of hemorrhage into the joints or muscles. At times it lies curled up, with the side of its face resting on the floor, as if to support its painful or sensitive jaw; this they have termed the “face-ache position.” The two diagnostic signs, however, are the hemorrhages about the joints and the loosening of the teeth. The diagnosis frequently can be established by the twelfth to the fourteenth day; the earliest diagnosis was made on the eighth day. Hemorrhages appear somewhat later than tenderness, and are situated at the joints, most frequently at the knee, which may be markedly swollen and show a bluish or reddish discoloration, extending upward or downward for some distance. Other joints are often involved, frequently the wrist, the ankle or the shoulder. In other cases hemorrhages into the muscles are noted, especially of the leg or of the thigh, and later, especially toward the end, hemorrhages from the bowel. Fractures or separations of the epiphyses may be found on examination or may be occasioned by the physical examination. A frequent site of this lesion is at the wrist or at the knee, involving the head of the tibia or the lower end of the femur. These fractures knit rapidly when an antiscorbutic is given, but result at times in deformity.

Loosening of the teeth is another typical sign. It is, however, one which does not appear early and is somewhat difficult to elicit. The molar teeth are generally involved, especially those of the upper jaw, which may be so completely separated from their alveolar sockets that they can readily be removed by forceps. It is, however, impossible to examine the molar teeth satisfactorily during life. Far less frequently an incisor tooth becomes loosened; more often it loses its glistening appearance and looks dull and yellowish. Occasionally an incisor tooth fractures. The gums are rarely altered sufficiently to aid diagnosis; not infrequently they are congested or bluish, rarely hemorrhagic, and never ulcerous or spongy. This is the chief difference between the symptomatology of scurvy in the guinea-pig and in man, and probably is the result of lesser susceptibility of the former to infection by pyogenic bacteria. In general, however, the disorder in the guinea-pig bears a closer analogy to infantile than to adult scurvy; due to the fact, possibly, that young pigs are generally used for the experiments.

A sign of great interest, although not of diagnostic importance, is the “beading” of the ribs noted by many observers, and emphasized by Jackson and Moore. It has been described also in monkeys by Hart and Lessing. This is an enlargement or swelling of the costochondral junctions of the ribs, especially of the lower true ribs. It corresponds clinically to the “beading” and the “rosary” so characteristic of infantile rickets and mistakenly termed the “rhachitic rosary.” As pointed out, in discussing the symptomatology and pathology of human scurvy, this sign must be regarded as truly scorbutic in animals as well as in infants. The “rosary” is difficult to palpate in pigs which have considerable subcutaneous tissue; in thin animals, however, its course can be followed, the gradual development and subsequent disappearance. In this connection the enlargement of the epiphyses must be mentioned, another sign supposed to be characteristic of rickets in infants. Marked swelling of the wrists is frequently encountered in guinea-pigs suffering from scurvy—a bony enlargement involving the lower epiphyses of the ulna and of the radius. This is met with far more commonly in chronic scurvy than in the usual acute case. Where the disorder has existed for a long period, these bony knobs may persist indefinitely, constituting the sole residual sign of a former scorbutic condition.

The relation of loss of weight to the development of the scurvy requires consideration. In experiments carried out on rats to test the diets in respect to the water-soluble and the fat-soluble vitamines, the weight curve is used as the main criterion to judge whether the foodstuff is adequate. In guinea-pigs we cannot employ the weight curve as a criterion. Not infrequently an animal develops scurvy, and nevertheless does not lose in weight, but even gains slowly and steadily. This occurs when the appetite remains good, and the dietary is complete except for antiscorbutic vitamine. The same holds true for human scurvy, as we have noted in connection with the symptomatology. We have attempted to make up for this defect by adding to the charts a curve representing the clinical course (Fig. 10). Although this curve is computed on an empirical basis, it gives a comparatively true picture of the disease and is far more exact than attempting to portray the disease by means of a weight curve.38 Indeed, when we rely on the latter method it is impossible frequently to illustrate graphically the reaction of guinea-pigs to various influences. Another disadvantage of the weight curve is that it is influenced by factors having no direct relation to scurvy, especially infections of various types which retard the gain.

Guinea-pigs generally die of scurvy after having lost about one-third of their body weight; occasionally the loss is greater, reaching almost 50 per cent. This loss is due partly to the scorbutic condition, but to a greater extent to starvation occasioned by a marked lack of appetite. In this connection it may be noted that guinea-pigs frequently lose for a few days following the addition of an antiscorbutic to the dietary (Fig. 11). This reaction is evident from a perusal of the weight charts of other investigators, and occurs likewise in human scurvy. This loss is accompanied, as Gerstenberger has pointed out, by diuresis, which may be so marked that it is evident to the casual observation of those caring for the animals.

Exophthalmos may be mentioned again in this connection as a very rare sign of guinea-pig scurvy. HÆmaturia also occurs at times; it is not known how frequently it is present, whether it is an early manifestation in the guinea-pig as in the infant, or whether the source of the blood is the kidney or the bladder.

The superficial lymphatic glands are frequently palpable in scurvy, especially those in the inguinal region. This sign is emphasized by some pathologists. It has seemed to us attributable less to the nutritional condition than to the infections which so frequently complicate the disorder.

According to Jackson and Moore a rise of temperature does not accompany scurvy in the guinea-pig. As the result of an examination of nine guinea-pigs they conclude that “experimental scurvy is a non-febrile disease in the majority of affected animals.” Nor did they find a leucocytosis, the average leucocyte count of eight scorbutic pigs being about 8000.

As the data are meagre, a consideration of the chemical alterations associated with the scurvy of guinea-pigs and monkeys will be taken up in conjunction with the metabolism of human scurvy. The therapy of animal scurvy will also be deferred (chapter VI), as the reaction to dietetic measures is practically the same in man and in animals. In concluding this chapter we would call attention to the following interesting statement contained in the recent paper by Delf and Tozer: “In other experiments, however, where a liberal supply of an antiscorbutic was given and where the fat-soluble A growth factor was known to be deficient, the resulting histological changes in the rib-junctions of the animals examined were found to resemble closely those of ‘Definite’ or of ‘Definite Chronic’ scurvy. In these test cases the antiscorbutic chosen was, we believed, deficient in the fat-soluble A growth factor (for example, orange juice). This fact is mentioned because in the case of an animal not receiving an adequate supply of fat-soluble A the resulting changes in the junctions are not dissimilar from, and are likely to be confused with, those caused by scurvy alone.” If this observation is confirmed, it emphasizes the necessity in nutritional experiments, of constructing a dietary which is complete in every respect except the one under investigation; it also suggests the development of new pathologic entities in relation to other nutritional disorders resulting from a lack of the accessory food factors.