The study of an anaerobic bacillus isolated from an infection in a French soldier has shown it to belong to the group of butyric acid bacilli classified as Bacillus Welchii. It has been found to produce no true exotoxin nor was the bacterial protein toxic. Nevertheless the bacillus has killed 5 pigs acutely, in spite of the fact that no general distribution occurred, and the bacilli were found in the blood stream at death either in small numbers, or not at all.

Since death could be produced in these animals by injection of the bacterial free filtrate, and the toxicity of this filtrate could be rendered neutral with sodium hydroxide we feel inclined to attribute much of the toxemia produced by the bacillus to its ability to form acid. Evidence has been advanced to show that large amounts of acid were formed in the lesions, probably by splitting of glycogen within muscle tissue.

We believe, therefore, that acid formation of Bacillus Welchii in the tissues, constitutes a powerful factor in the injury of the animal body.

Of the lesions distant from the focus of infection, the most striking one has been gastric ulceration. This has induced us to scrutinize closely the literature on experimental ulceration. As far as we can determine, only the acute, rapidly healing ulcers have been experimentally produced, and not the more chronic ones observed so frequently in human beings. These ulcers have been found to be produced through various agencies many of which are apparently unrelated to the question of acidity.1 However, there are some experiments which more or less parallel our own. Türck has found gastric ulcers in dogs following the feeding of large quantities of broth inoculated with colon bacilli, which doubtless was acid in reaction. Bolton has produced ulcerations of the gastric mucosa in cats, monkeys and guinea pigs by feeding them with 0.7 to 1.6 per cent HCl. The gastric mucosa of the guinea pigs in this case, was invariably ulcerated after doses of 0.7 to 0.9 per cent HCl.

We have found that ulcerations of the stomach in guinea pigs occurred regularly after intravenous injections of 4 to 4.5 cc. of 1.0 per cent acetic acid.

In view of these facts, we feel justified in concluding that the excess of acid produced by the tissue necrosis is connected with the formation of the gastric ulcers, probably modifying the reaction of the blood stream. Whether the modified blood reaction causes the death or lessened vitality of many of the body cells and the subsequent digestion of these cells of the gastric mucosa which have been injured, rendering them noticeable as a gross lesion, or whether it is possible for the stomach through its acid forming mechanism to lessen the blood acidosis by markedly increasing the acidity of the gastric juice and thus bring about the ulceration of the mucosa, can only be determined by a further investigation.

Of course the ulcer formation, although interesting, is but an incidental lesion. The point we wish to emphasize is that this organism forms no true toxin, nor does it regularly invade the blood stream. It is the acid produced by the bacillus in the breaking down of tissue that is a factor of great importance in causing the death of the infected animal.

We believe that the important principle demonstrated by this work is the fact that severe injury can be produced within the animal body by products of the bacterial activities upon the body tissues; products which are in no sense bacterial poison, but represent substances derived from the cleavage of the tissue constituents. Libmann, some years ago, suggested this possibility in his work on the streptococci, but subsequent workers have, we believe, failed to take cognizance of this possibly important means of injury to the animal body.


Rosenow's production of gastric ulcers by the injection of streptococci and his views on the specific localization of certain strains streptococci are too well known to be repeated here. Mann of Mayo's Clinic has noted the formation of gastric ulcerations in a large number of dogs and cats following double adrenalectomy. The ulcers developed in the absence of pancreatic juice or bile and could be prevented by continuous etherization or feeding with sodium bicarbonate. Mann quotes Durante as stating that section of the median or minor splanchnic nerve on either side is followed by ulcerations of the gastric mucosa, while Elliot has observed the formation of gastric ulcers in guinea pigs following the subcutaneous injection of tetra hydro-β-naphthylamine hydrochloride, a drug which causes rapid exhaustion of adrenalin from the glands by impulses passing over the splanchnic nerves. The ulcers could be prevented by feeding sodium bicarbonate. The injection of thyroid extract and adrenalectomy have produced gastric ulcers in the hands of Friedman and the same author has noted duodenal ulcers following the injection of adrenalin. This is especially interesting in the face of the fact that Fawcett, Rogers, Rahe and Beebe state that thyroid extract subcutaneously was the most potent stimulant of gastric secretion examined by them, while adrenal extract tended to lessen the gastric secretions. Interesting in connection with the adrenal work is the observation of Rosenau and Anderson that gastric ulcerations are not infrequent in guinea pigs killed with diphtheria toxin a substance causing marked adrenal lesions in these animals. Bolton has produced ulcerations of the gastric mucosa by injecting guinea pigs intraperitoneally with the sera of rabbits immunized with guinea pig gastric mucosa or liver. Finally Bardeen quotes Schjerning's statistics compiled from about two hundred autopsy reports, collected from the literature, of death following extensive burns. In those cases in which death occurred in less than forty-eight hours after the burn no duodenal ulcers were noted, but 19 per cent of those dying after forty-eight hours showed this lesion. Associated lesions were hyperemia of the brain, lungs and intestines, nepthrites, pneumonia, pleurisy and meningitis. Bardeen observed five cases of extensive burns but all died within ten hours after injury and no duodenal ulcers were found.

This content is only available via PDF.