toms of disease except a tendency to lie quietly in its pen. On examination the subcutaneous fat was found diffusely reddened. There was a slight peritonitis, indicated by a considerable quantity of straw-colored effusion and some fibrinous stringy deposits. There were also a few local excrescences on the small intestine, due to the irritation of echinorhynchi. Spleen somewhat enlarged; on its surface a few bright red punctiform elevations. Right heart distended with a clot. Local hepatizations in lungs, probably caused by lung worms, which were very numerous. Stomach but slightly reddened. A number of ulcers in the duodenum, the mucosa of which was reddened. The mucosa of the ileum for 14 feet from valve was completely necrosed, the walls thickened, and the serosa of this portion dotted with ecchymoses. Beyond this portion, near the jejunum, there were scattered ulcerations on a deeply congested membrane for 6 or 7 feet. The entire length of the large intestine was covered with dirty yellowish ulcerations varying in diameter from a pin's head to nearly an inch. The mucosa itself was very deeply congested in the cæcum and colon only and the walls much thickened. Ascarides and echinorhynchi numerous in small intestines. The liver attached to diaphragm in several places by whitish exudate.

A tube of meat infusion with peptone inoculated from the spleen of this animal was found to be a pure culture of the motile bacterium of hog-cholera. Line cultures on gelatine plates confirmed the microscopic examination. A tube of nutritive gelatine inoculated from the spleen at the same time contained in each needle track, several days later, from 10 to 15 colonies of the same bacterium. Two coverglass preparations revealed no bacteria. This fact, combined with the small number of colonies in the tube culture, gave evidence of the small number of germs in the spleen tissue. Inoculations on mice and guinea-pigs gave substantially the same results as those obtained last year.

No. 159 was fed with the viscera of No. 165 on January 28. February 5, its eyes were sore and nearly closed; it was quite weak. It died on the following day, only eight days after infection. The skin on abdomen was reddened in patches; the subcutaneous tissue diffusely. The superficial inguinals, as well as the glands in the abdomen, were deeply congested, the cortex more especially. Those of the thorax were nearly pale. The spleen was dotted with a few punctiform elevations. Beneath the epicardium and endocardium of both auricles and the endocardium of the left ventricle were extensive patches of extravasated blood. Kidneys enlarged and congested throughout. The lesions of the ileum, cæcum, and colon in this animal were quite as extensive as those of the case just described; there were no ulcers in the rectum, however. Those of the colon had black centers, pointing to a recent origin from blood extravasations on the surface of the mucous membrane. In the spleen of this case the characteristic bacteria of hog-cholera were exceedingly numerous, as determined by cover-glass preparations. Two liquid cultures proved pure when tested on gelatine plates. In the needle tracks of a tube culture in gelatine innumerable colonies appeared in a few days. Inoculations from subsequent cultures proved equally positive.

Pig No. 156 was fed with the viscera of No. 159 February 18, and, after manifesting the usual symptoms of hog-cholera, died February 25, seven days after feeding. Among the marked lesions produced by the disease was a complete necrosis of the upper two-thirds of the colon, with scattered ulcers along the lower third. About an eighth of a foot of the lower portion of the small intestine, beginning at the valve, was necrosed, without manifesting distinct ulceration, for which the period of disease was evidently too brief. In the spleen there were numerous small grayish spots, probably centers of necrosis, as they showed no longer cell structure when crushed on a slide and stained. The fundus of the stomach was also deeply congested.

The spleen, to which organ the microscopic examination was limited, contained the characteristic oval bacteria, as shown by cover-glass preparations. Three liquid cultures made from the same organ were found to be pure cultures of the same microbe when tested by line cultures. A tube culture in gelatine developed in each needle track numerous non-liquefying colonies.

In these animals the mode of introduction of the virus determined the seat of the severest lesions. It is probable that the food passes quite rapidly through the small intestine; that in the stomach the action of the bacteria is more or less limited, because they have not sufficient time to multiply, and probably because hindered by the acid condition of the organ, though they will multiply with considerable vigor in slightly acid infusions. The prolonged stay of the food in the large intestine permits multiplication, and thereby causes the

first and severest lesions to appear here. When these have become very extensive, so as to paralyze the action of the large intestine, the ileum becomes involved in a similar manner, possibly by a partial stoppage of the infectious matter in this portion of the intestine. This view is supported by the evidence of the above and other post mortem examinations in which the disease was produced by feeding. This mode of infection by feeding viscera was used to keep up the disease at the experimental station, as simple infection in pens could not always be relied upon in furnishing cases for investigation. These few cases might therefore be supplemented by many others to show the ease with which infection may take place in this way.

In general two types of disease appear. In one the lesions are limited quite exclusively to the alimentary tract, involving the stomach, the large intestine, and often the ileum, less frequently the jejunum. There may be complete necrosis of the mucosa in the colon and ileum, with intense reddening of the fundus of the stomach. The internal organs are but slightly affected. There are few or no hemorrhages, and the bacterium is very scarce in the spleen and other organs, so that its presence is only determinable by culture.

In the other type the extensive local lesions are replaced by hemorrhagic lesions of the internal organs, involving the spleen, kidneys, lymphatic glands, lungs, and serous membranes generally. Besides these, the mucous membrane of the stomach and intestines may be congested, and extensive hemorrhages into the submucous tissue, often into the lumen of the digestive tube, take place. These lesions have been described somewhat in detail in the last report. The spleen and blood are found to contain a large number of the hog-cholera bacteria.

Both types of the disease produced by feeding lead to a speedy termination by death in from six days to two weeks. The difference above given may perhaps be referred to a difference in the virulence of the bacteria. In the type first described the bacteria may be less adapted to a parasitic life. Their poisonous effects are exerted locally in destroying the mucous membrane. In the second type the bacteria are capable of entering the blood, to be distributed to all the organs where the hemorrhagit lesions are caused by their growth.

(b) Feeding pure cultures of the Bacterium of hog-cholera.-In the preceding report (p. 207) two very severe cases of hog-cholera are reported as having been produced by the feeding of pure liquid cultures. These positive results are not always obtained, as some of the following experiments indicate:

Pigs Nos. 155 and 156 were fed February 8, 1886, with 200cc of a beef-infusion peptone culture derived from a mouse which had succumbed to inoculation. The animals remained well. Pig No. 155 was fed February 1 with 100% of liquid cultures of the bacterium of hog-cholera without manifesting any symptoms of disease.

The rapidly fatal effect from the ingestion of the viscera of swine containing the specific bacterium may be harmonized with the negative results above recorded when we consider the different condition of the bacterium in the liquid cultures and in the infectious viscera. In the latter case the bacteria are enveloped by connective and cellular tissue, which protect them from the destructive effect of gastric digestion, so that they are carried into the intestine where the pathological lesions are first manifested. In culture fluids the bacteria are in the most vulnerable state, and are easily accessible to the action of the gastric juice, which very probably does not permit any to pass alive into the duodenum.

That the condition of the stomach is a very important factor in the production of the disease when the virus has entered it seems a very reasonable assumption. If the virus reaches the empty stomach coated with an alkaline mucus it is more likely to multiply and reach the duodenum than when the stomach is filled with food which is being actively digested. In order to test this assumption the following experiment was tried:

December 13.-Three pigs were fed each with 300cc of beef infusion in which the bacterium of hog-cholera had been multiplying for three days at a temperature of 90° to 95° F. The beef infusion had been neutralized, and sterilized in two flasks, and the cultures, when examined before the experiments, were found to contain the motile bacterium only.

The pigs were prepared for the feeding as follows: No. 348 received no food for over twenty-four hours. A 2 per cent. solution of sodium carbonate in beef infusion was then given to increase the alkalinity of the stomach. Of this about 1 liter was consumed. It was then fed with 300cc of culture liquid mixed with beef-broth to make 1 liter. No. 350 was starved in the same way, but received no alkali before consuming the culture. No. 342 was not deprived of food before eating the culture. The result confirmed our anticipations. No. 348 showed signs of disease in two days. On the third it was unable to rise, and died on the same day. The post mortem examination showed a considerable congestion of the mucous membrane of the duodenum and jejunum, as well as of the large intestine. The fundus of stomach affected in the same way. The liver was gorged with blood, as well as the portal system. There were no marked lesions of the other viscera. That the hog-cholera bacterium had also entered the blood was shown by two pure cultures in beef infusion obtained from the spleen. A gelatine culture from the liver contained about 6 or 7 colonies.

No. 350 was a more typical case, and demonstrated the severe local effects of the bacterium much better, since the animal lived longer. It ate fairly well until the fourth day, when its appetite gave way and diarrhea set in. From this time it grew weak and thin, being scarcely able to walk. It died on the tenth day after feeding. The lesions of the alimentary tract were exceedingly grave. Beginning with the stomach, the mucous membrane was dotted with closely set elevated masses as large as split peas, and larger patches of a whitish viscid substance, made up entirely of cellular elements (diphtheritic?). When removed, a raw depressed surface was exposed. The membrane itself was pale. Besides a general injection of the ileum, Peyer's patches were more deeply congested, and the uppermost covered with a thin yellowish film, not removable, and most likely dead epithelium. In the cæcum and colon the mucosa was superficially necrosed, and converted into a continuous layer of a dirty whitish mass about 1mm thick. The walls of the intestine were greatly thickened and very friable.

Microscopic sections showed an extensive cellular infiltration of the submucous connective tissue which had separated the masses of fat cells, concealed the connective tissue fibers, and caused a great thickening of the entire layer. The mucosa itself was greatly altered. The surface was necrossed and converted into an amorphous mass. In some places the necrosis involved the entire depth of the crypts of Lieberkühn, a series of striæ indicating their former existence. Those whose epithelium still remained were plugged with a cylindrical mass, filled with broken-down nuclei. The bacteria had exerted their poisonous effects from the surface of the mucosa towards the depths, destroying the surface epithelium and glandular structures and involving secondarily the submucous layer. Near the rectum this continuous mass of dead tissue was replaced by isolated ulcers embedded in an intensely reddened mucosa. Plate II, taken from another case, illustrates well the superficial death of the mucosa. The ileo-cæcal valve was much swollen, but the necrosis did not extend into the ileum, although there were a few ulcers near the valve, and the epithelium had a pale, lusterless aspect, as if dead. The liver was filled with blood, which readily clotted as it flowed from the cut surface. Spleen congested and but slightly enlarged. Lungs hypostatic. The lymphatic glands in general not much affected. Two liquid cultures from the blood were turbid next day, and contained the hogcholera bacterium only. In a gelatine tube culture from the liver about a dozen colonies developed in each needle track.

No. 342, which was fed with the same quantity of culture liquid but was not deprived of food previously, was somewhat ill on the following day. It recovered, however, and continued apparently well for several weeks. It began thereupon to grow thin and weak. On January 26 it was no longer able to rise, and was therefore killed for examination, in order to conclude the experiment. On opening the

abdominal cavity it was at once perceived that the animal had been suffering from a very intense disease of the large intestine, a portion of which was firmly attached to the bladder. When dissected out and slit open, the mucous membrane of the cæcum and colon was found replaced by a brownish friable layer of necrosed tissue. The wall of the intestine was infiltrated to such an extent that it was nearly inch thick, and so degenerated that the forceps easily tore through it. The thickness of the walls prevented the intestine from collapsing after it was opened. Its only contents was a brownish liquid mass. The glands of the meso-colon were very large, some like horse-chestnuts. On section the entire tissue was very pale, almost white. The spleen was somewhat enlarged; the malpighian corpuscles unusually large and prominent on section. Lungs and heart normal; kidneys deeply reddened throughout on section.

This case is very interesting in completing the information gained by this feeding experiment. No. 348, which had been fed with sodium carbonate besides being deprived of food, died three days after the ingestion of the culture. No. 350, which was simply starved, died ten days thereafter, while No. 342, which ate the culture without being previously starved, was dying on the thirty-fourth day.

These results show how easily infection may occur by way of the digestive system, provided the destructive action of gastric digestion be prevented, as was done by starving and by the use of an alkaline carbonate.

They also indicate how purely local this destructive action may be. Gelatine cultures from these animals showed that the internal organs contained but very few bacteria. So few were they in fact that the microscope alone could not have demonstrated their presence, as they could not be found on cover-glass preparations.

Other successful experiments by feeding pure cultures will be given in connection with a description of the bacterium from different parts of the country.

(c) Subcutaneous inoculation with pure cultures.-The least successful method of producing the disease is the subcutaneous inoculation of pure cultures. In the report for 1885 at least three out of four inoculations produced a rapidly fatal form of the disease. In the numerous experiments to be described later, in which pigs were inoculated with cultures to determine whether any future protection was thereby granted, only five died from the inoculation. In these experiments two subcutaneous injections were practiced, a small quantity being followed by a larger quantity of culture liquid. The deaths occurred from the first injection when this was made comparatively large; the second dose, which was usually quite large, was borne without any ill effects.

These successful inoculations, reported in extenso in another section, are briefly as follows: No. 239, inoculated April 27 with ec culture liquid, died May 2, only six days after inoculation. Hemorrhagic condition of vital organs. Though seven others were treated in the same way none took sick. It may be that in this individual the needle entered a superficial vein, and in this way introduced the virus directly into the blood.

Nos. 204 and 212, inoculated April 12 with 14 culture liquid, died eleven and seven days after inoculation, respectively. In the former the mucosa of large intestine was entirely necrosed. In the latter ulceration was just beginning.

Nos. 208 and 209, inoculated at the same time with 1c of the same culture, died fifteen and six days after inoculation respectively. Numerous extensive ulcers in the large intestine of 208. In 209 general congestion and extravasation along digestive tract and in internal organs. With each pair two others had been inoculated without any untoward results.

The local swelling at the point of inoculation is usually proportionate to the quantity of culture fluid injected. The following cases show how large quantities may be borne without inducing the disease:

Nos. 116 and 154 were inoculated February 8 by the subcutaneous injection of 31ee of the second culture from the spleen of No. 165 in beef-infusion peptone, one-half

being injected into each thigh. A very large swelling appeared at the seat of inoculation in No. 154, causing considerable lameness. March 4 this animal was killed, although evidently not diseased. The inoculation tumor was over 1 inch long and i inch thick; firm, yellowish-white, developed in the loose connective, and only loosely attached to skin and subjacent mucular tissue. There was considerable serum in the abdominal cavity, and the spleen was somewhat enlarged. In the fatty tissue lining the dorsal wall of the abdominal cavity numerous worms were found (Sclerostoma pinguicola), occupying tunneled spaces in the fibro-adipose mass. No indications of hog-cholera. No. 116 was not affected.

No. 181 was inoculated February 13 with 7 of the second liquid culture from the spleen of pig No. 159. Within a few days the animal became lame, but this passed away. At the seat of inoculation large tumors had developed, no doubt causing the stiffness of the hind limbs. This animal was killed March 4. At each point of inoculation were found firm fibrous masses, from 2 to 3 square inches in extent and nearly an inch thick. No suppuration. There were no lesions pointing directly to hog-cholera. There was, however, a considerable quantity of pale serum in the abdominal cavity. Spleen enlarged; cortex of kidneys dull, thickened; lymphatic glands of large intestine somewhat prominent, but pale.

These animals may have suffered from the absorption of ptomaines from the place of inoculation, but dissemination of the bacteria through the internal organs evidently did not take place to any extent.

The failure to produce the disease in even a small proportion of animals by the injection of liquid cultures raised the question whether the cultivation in itself did not attenuate the bacteria. Consequently two experiments were made by inoculating with blood directly. Numerous gelatine cultures of heart's blood had demonstrated the very small number of bacteria compared with the number present in the spleen.

September 10.-A pig dying with the disease was killed, the heart carefully exposed, and the blood drawn with a disinfected hypodermic syringe. Nos. 329 and 333 received subcutaneously 5cc each, one half in each thigh. No. 329 in a few days lost its appetite, became weak and stupid. Found dead October 5. Slight local swelling at the points of inoculation; superficial inguinals greatly enlarged; hypostatic congestion of lungs; complete necrosis of mucous membrane in cæcum; large scattered ulcers in colon, showing as whitish patches on serous surface and encircled by a crown of enlarged blood vessels; bacteria in spleen.

No. 333, slightly ill for a time; fully recovered. Died December 2 with no other lesions than engorgement of liver. No signs of former ulceration.

A second experiment was made in the same way:

October 13.-Nos. 324 and 325 inoculated as in the preceding experiment, 10cc of blood being used for each animal. No.324 was found dead November 1, after being off feed for a time. Deeply reddened skin over caudal half of abdomen; extremely large and serously infiltrated superficial lymphatics; on section, hemorrhagic points. At point of inoculation the connective tissue is infiltrated; 50 to 75 clear amber serum in abdominal cavity; papillæ of kidneys deeply reddened; slight congestion, but no ulceration in large intestine; lymphatics in general moderately tumefied and congested.

No. 325 found dead October 29. Reddening of skin as in 324; extravasation in connective tissue; spleen greatly enlarged, purplish; lymphatics of thorax and abdomen purplish, enlarged; petechia on section of kidney and in pelvis, also over entire surface of epicardium; lung tissue mottled both on surface and on section with purple spots, due to blood extravasation into alveoli, so that it scarcely floats. Mucous and serous surface of small intestine dotted with petechiæ; small hemorrhages on the surface of the mucous membrane and into the submucous tissue of the cæcum and upper colon. Ulceration beginning.

These results are more positive than those obtained with cultures, and on first thought we may be inclined to attribute them to a greater virulence of the germs in the injected blood. This view needs further confirmation, however. The injected blood coagulating in the connective tissue contains in it the bacteria, which are not only protected from the aggression of cellular elements, but have actually a store of nourishment upon which they may live and multiply. No such H. Mis. 156-3