to this there is described a suppurative, a sequestrating and a gangrenous form of pancreatitis. As many of you know, pancre atitis, especially of the hemorrhagic form, is a peculiarly fatal disease in man. Its onset is often sudden, not preceded apparently by any definite train of symptoms, and its termination is extremely rapid and usually fatal.

The experimental investigations of pancreatitis come into very close relationship chronologically with those of experimental fat-necrosis, for the reason that some of the procedures used to excite changes in the pancreas that would lead to fat-necrosis, were found to give rise at times to lesions in the pancreas that bear a distant relation to pancreatitis in man; but especial attention was paid to the experimental production of pancreatitis beginning four or five years ago with the work of a Frenchman, Carnot, and a Bohemian, Hlava, both of whom undertook to carry out a wide series of experiments upon the production of this disease. Those which Hlava undertook were perhaps the most interesting and I think the most important. He conceived the idea that the onset of pancreatitis being as sudden as it seems to be, the cause probably is not to be sought in the presence of bacteria. He further conceived the idea that perhaps under some conditions not understood at present, the gastric juice which gains access, of course, into the intestinal canal, may through an anti-peristaltic action of the intestine be driven into one of the pancreatic ducts, either into the main duct or the accessory duct, and having gained access to the pancreas, set up there changes which might lead to pancreatitis.

In the course of the experiments which he carried out to prove or to disprove this view, he injected into the parenchyma of the pancreas of a dog, besides gastric juice cultures of bacteria in suspension, and he found as had others, Carnot especially, before him, that whereas bacteria could be made to set up inflammatory changes in the pancreas, that they

did so with great difficulty, and that large quantities were needed; that forms of bacteria which have as a rule nothing to do with lesions in that location in the body, such as the diphtheria bacillus, were most effective perhaps on account of the potency of the toxins they produced, and that it seems improbable, so far as the hemorrhagic form of the disease is concerned, that bacteria have anything to do with its production. On the other hand it was soon discovered that pyogenic organisms introduced into the substance of the pancreas were capable of setting up suppurative lesions bearing a close resemblance to suppurativ pancreatitis which occurs in human beings. Hlava's experiments with gastric juice were more convincing, but they were still unsatisfactory; they did not act regularly, nor did it reproduce accurately the disease in human beings, but gave rise to definite inflammation, associated with hemorrhage, much more closely resembling pancreatitis in man than any previously obtained by other experimenters.

My own interest in this subject dated a few years before the experiments of Hlava and came from the fact of a favorable case of fat-necrosis having come to my attention, in which I undertook to demonstrate chemically what Langerhans had failed to prove, viz. the existence of the fat-splitting ferment in the necrotic areas. I chanced to be successful and my demonstration has since been confirmed, so that the evidence is now cmplete that the fat-necroses are due to the presence of this ferment.

My interest was soon transferred from the study of fat-necrosis to that of pancreatitis, but I shall not weary you with details as to the variety of experiments which were carried out on the dog to determine whether or not it was possible to produce a pathological condition that might be regarded as identical or comparable with that which occurs naturally in man. For this purpose a variety of irritants were injected, usually through the duct of Wirsung, into the pancreas. You may be interested

to know that in carrying out this experiment, which of course is done under the influence of an anesthetic, the duodenum is split open and a canula introduced into the duct, the injections being carried directly into the substance of the gland. The injections if fluid, pass quickly through the main duct into the smaller ducts, thence into the acini of the gland, but they do not remain there but quickly filter out and get into the lymphatic spaces, and eventually into the peri-pancreatic tissues. It was quickly found that bacteria could not produce the typical disease found in man, unless extraordinary quantities or unusual forms were used, thus confirming the previous experiments of Carnot. It was also found that rapid disorganization of the gland of the animal could be produced by chemicals, and if the animal lived 48 hours the necrotic tissue, when invaded by putrefactive bacteria from the intestine, produced a condition resembling gangrenous pancreatitis in man. Most interesting were the experiments in which gastric juice was introduced in sufficient quantity to filter out into the lymphatic spaces of the acini. The anımals died at the end of 24 to 48 hours, much depressed, never coming really out from the influence of the anesthetic. At least they remained in a depressed or comatose condition until death; sugar appeared in the urine; the pancreas when examined showed a picture that in all respects, I think, can be compared with that which occurs in man. It was the seat of a widespread hemorrhagic infiltration. Hemorrhage was sometimes so abundand as to pass into the omentum, and the peritoneal cavity was covered with fat-necroses, as in human beings; and altogether the pathological picture is that which we are acquainted with in the acute disease in men.

I have here a reproduction of such an experimental form of hemorrhagic pancreatitis, that has lost some in the reproduction, but I think will show accurately enough what can be thus produced in the dog.

(Drawing exhibited.)

It seems to me therefore as proven that hemorrhagic pancreatitis in the dog can be made to resemble that condition in man, and that the latter is brought about, not by the action of bacteria, nor through the influences of traumatism, but probably through the perversion of some normal secretion which, gaining access to the pancreas, sets up a profound pathological disturbance, giving rise to inflammation and also to hemorrhage.

A most important contribution to this subject was made about two years ago by Dr. Opie of Baltimore, who had the good fortune to do an autopsy on a man who died of acute pancreatitis, in which case he found that a gallstone had lodged in the duodenum, compressing the common bile-duct and the pancreatic duct and permitting a back-flow of bile through the pancreatic duct into the pancreas. He was able to trace, by the bile-staining of the duct of Wirsung, the entrance of bile into the pancreas. The case was one of acute pancreatitis, hemorrhagic in character, from which he drew the conclusion that in all probability the perversion of the secretion of bile and its entrance into the pancreas are capable of setting up pancreatitis, and he transferred this idea from man to the dog, introducing experimentally into the duct of Wirsung bile taken from the dog, and he was able by this means readily to reproduce the picture of acute pancreatitis such as is found in man, and such as is producible by the injection of gastric juice.

Now it chances that a study of the cases of human pancreatitis shows that a large percentage of them are associated with cholelithiasis. This is a fact which had been previously pointed out by Dr. Fitz, as well as by many others to whose attention pancreatitis had come, and had been especially noted in respect to the occurence of fat-necrosis in the peritoneal cavity.

I think there is little doubt that Dr. Opie's demonstration of the effects of bile must be taken as conclusive, and as showing that that is at least one of the ways in which this acutely fatal disease comes to be produced.

I think it is probable and indeed so does he, that it fails to explain all cases of pancreatitis. Evidence from the study of human cases has accumulated which tends more and more to support the view that gastric juice may also enter the pancreas, although the mechanism of the entrance is not understood, and set up this condition. So, in short, I think it may be concluded that acute pancreatitis in man is in all probabilty a disease due to a perversion of normal secretions, these secretions being the bile on the one hand and the gastric juice on the other, and has nothing to do with the presence of microorganisms at all in its simple and uncomplicated form.

Another consideration has reference to a fact of especial interterest to surgeons, namely: that when the experimental injections are of a localized type, the animal may not succumb, and usually survives; and after weeks and months there develops in association with the pancreas, abscesses, cysts, and sequestra, such as now are diagnosed by the physician and surgeon and operated on by the surgeon with relief, in human subjects.

We are, therefore, I think, at present, informed rather fully as to the etiology of fat-necrosis, of the etiology of hemorrhagic pancreatitis, of cysts and sloughs of the pancreas, and incidentally of the origin of gangrenous pancreatitis, which seems to be nothing more than one of the other forms in which putrefactive bacteria from the intestine find in the injured and dead tissue an opportunity for their development and the production of putrefaction.

There has always been a question as to the relation of hemorrhage to inflammation in the pancreas. You will find in the literature abundant discussion as to whether hemorrhage is the primary thing, or the secondary thing; whether the hemorrhage sets up the inflammation, or inflammation predisposes to the hemorrhage. It does not seem possible to settle from human cases this point. The two conditions in the acute form