Vomiting at the time of perforation, or within a few minutes. after, occurred in 8 of my 9 cases. I believe that vomiting, coming on soon after pain in the bowels or without any other apparent cause, is a valuable corroborative symptom.

Chill.-Chilly sensations, shivering, or a marked chill occur in a number of cases. It was present in at least 33 per cent. of my patients, and Goodall found it present in 26 per cent. of his 96 patients.

Dulness on percussion and tympany are unreliable symptoms except where there is absence of liver dulness, and even this may be caused by the presence of gas within the intestines. Dulness in the dependent portion of the flanks, especially the right, sometimes occurs after perforation and the accumulation of fluid in this region.

Stools. It is said that a change from constipation to diarrhoea, or vice versa, often signalizes the occurrence of perforation. The cases in which I have noted this symptom, three in number, were characterized by the discharge in rapid succession of from three to five stools, preceded by pain in the abdomen.

Diagnosis. It is sometimes impossible to make the differential diagnosis from appendicitis, perforation of gastric ulcer, peritonitis without perforation, and gallbladder infection, but as all of these conditions require operation, it is not a matter of such great importance. Gallbladder infection can usually be distinguished by the location of the pain in the epigastric or hypochondriac region. In a patient operated on by me for perforation of the gallbladder during the course of typhoid fever there was no difficulty in making the diagnosis of gallbladder infection and finally rupture, chiefly from the location of the pain and

tenderness.

Prognosis.—That recovery from perforation can take place without operation has been proved, but such a result is in the nature of a surgical curiosity on account of its extreme rarity. Hare reports a patient, quoted by Goodall, on whom a post mortem, made six weeks later, death occurring from another cause, showed that a perforation had healed. Conservative, or over sanguine, estimates, place the mortality without operation at 90 per cent. at least. The table referred to of 362 patients operated on, gives a mortality of only 74.03 per cent. This percentage may perhaps be a little too low on account of the fact that many

unsuccessful operations have not been published.

Some opera

tors have in a small number of cases had a high percentage of success. Thus of Shepherd's three patients, all recovered. On the other hand, other operators have had a high percentage of deaths -of Munro's 12 patients all died. Those who have had the largest experience have had average results. Harte in this country operated in 20 cases with 4 recoveries. Lavroff of Russia operated on 33 patients with only 4 recoveries.

Harte and Ashhurst sum up the conditions which give the most favorable prognosis as follows: "Girls from ten to fifteen years. of age, the perforation occurring in the first week of the disease when the constitution is still strong, or in convalescence when the frame is already re-established; who have been operated on in the third hour after perforation, the single perforation being pinhead in size, within a few inches of the caecum or in the appendix; and when neither fecal extravasation had occurred nor adhesions were present."

Haggard collected, published and unpublished, the history of 137 patients operated on with 43 recoveries, a mortality of only 69 per cent.

Treatment.-After all that has been said the argument could scarcely be much stronger in favor of operation in every patient with typhoid perforations unless already dying, as the patient has everything to gain and nothing to lose. Laparotomy should be performed just as soon as possible after the perforation occurs, within thirty minutes if possible, before peritonitis begins. Those who have had the largest experience prefer a general anaesthetic and with them I fully agree. Local anaesthesia may have a limited field in patients in a semi-comatose condition. The incision should be made through the right rectus muscle, the caecum located and the examination of the small intestines should begin at that point. Openings should be closed when small—less than 4-inch-with one row of silk sutures which inverts the peritoneal coats, or by a purse-string suture. Larger openings should have two rows of sutures-one through all the coats near the edges of the opening, the other through all the coats except the mucous, inverting the first row. The continuous glover's suture is as good as any, and better than most of the complicated sutures which have been invented. The line of sutures should be at right angles to the long axis of the bowel in order to avoid constric

tion. Resection may be necessary on account of bad condition of the bowel. Irrigation with salt solution should be done if the contents of the intestines have escaped in quantity or when peritonitis is extensive, and drainage should be provided for. In limited peritonitis with little or no escape of intestinal contents, irrigation may be omitted.

It may not be considered out of place at this point to suggest some method of treatment of intestinal hemorrhage in typhoid fever, from which, according to some authors, as many patients die as from intestinal perforation. It has been estimated that from 30 to 50 per cent. of those who have intestinal hemorrhage die from the hemorrhage. I believe that the mortality would be lowered to something like 5 per cent. by timely operation. I would by no means advise operation in every case of hemorrhage, but when the patient's condition becomes serious, either from a sudden copious hemorrhage or from repeated small hemorrhages, no time should be wasted in the use of cold applications and the administration of morphine, ergot or adrenalin, but the abdomen should be opened at once and the bleeding point secured by sutures. The bleeding comes from an ulcerated Peyer's patch, therefore those patches should be carefully examined and those most affected should be excised and the intestines sutured, or the hemorrhage could be arrested by telescoping a section of the bowel so as to compress the ulcerated surface, using sutures to keep the invagination in position.

Indeed the time may come when typhoid fever will be treated as a surgical disease from the beginning. As the primary infection is in the glands which make up Peyer's patches, and the septicemia which constitutes the general symptoms is the result of absorption from the point primarily infected, why not, as soon as the diagnosis is established, open the abdomen, excise the affected Peyer's patches, or if many are involved excise the lower part of the ileum-thus removing the depot of infection in the same way that we remove the appendix, the gallbladder, or glands from the neck, axilla, or groin for various forms of infection.*

*Since writing the above I have observed the following extract from the Berliner Klinischer Wochenschrift in the Jour. Amer. Med. Ass., January 20, 1906.

"The patient was a young woman who was shot in the abdomen, the injury requiring resection of the ileum for 98 centimeters or about one yard. The course of the case showed that the injury had been received during the onset of typhoid fever. An epidemic of typhoid fever prevailing at the time was of an unusually severe type, but the disease was very mild in this case, possibly owing to the fact that the part of the bowel removed was exactly the part generally affected by the typhoid process.'

CASE I.-H. M., male, colored, aged 35, native of Maryland; admitted to Emergency Hospital on June 24, 1900, with a history of having been taken suddenly about noon with a severe pain in the abdomen, at first generally distributed but finally becoming worse on the right side. He vomited several times and had chilly sensations. Examination six hours after the symptoms came on revealed the following symptoms: abdomen hard and rigid, the rigidity being distinctly more marked on the right side, temperature 101, pulse about 110. He stated that he had been as well as usual when the severe pain came on, but that he frequently suffered from pain in the bowels which he thought was colic. A diagnosis of appendicitis was made and immediate operation advised, but this was declined until next morning, when the temperature had fallen nearly to normal but there was still considerable pain in the bowels.

About 18 hours after the symptoms began ether was given and the abdomen was opened on the right side.

as the peritonenm was opened, a small quantity of thin yellow fluid, much like pea soup, escaped. No adhesions could be felt, and the first coil of intestine which presented was withdrawn. This was the ileum, considerably swollen and inflamed, with patches of lymph on it, and about three feet from the colon a perforation was found in the part opposite the mesentery. The opening was round and about two millimeters in diameter. It was excised and closed transversely to the long axis of the bowel with three rows of fine silk sutures, so many rows being used on account of the great friability of the tissues. The appendix was removed, although nothing was found wrong with it except that it shared in the general inflammation which existed over a limited area in the peritoneal cavity. No irrigation and no drainage. The patient had slight elevation of temperature for several days after the operation, but was discharged recovered at the end of 19 days.

CASE II.-Perforation and Cancer of the Colon.-Referred by Dr. S. S. Adams.-R. C., white, male, age 50, blacksmith; was admitted to Georgetown University Hospital April 15, 1902, having been ill three weeks with typhoid fever. On admission the pulse was 96, temperature 100.6, respiration 20, mind clear; tongue pointed, dry and tremulous. He was ordered calomel

grain every half hour till 2 grains were taken, to be followed by half an ounce of magnesium sulphate. This produced three actions of the bowels between the 15th and the 17th of April. Early in the morning of the 17th he was suddenly seized with severe pains and cramps in the bowels, followed by vomiting. These symptoms continued all day until 11 o'clock at night. His temperature about one hour after the perforation at 8 A. M., was 100.4; at 2.30 P.M., 102.4; pulse 124; at 6 P. M., temperature

100, pulse about the same; at 11 P.M., when I first saw him, the pulse was over 160 and could not be accurately counted; he was thirsty, vomiting about every five minutes, but his mind was perfectly clear. He stated that the pain was not so severe as it had been in the morning. The abdomen was soft throughout, not distended, but was tender to pressure in both iliac regions. There was dulness on percussion in the right iliac fossa extending up to the liver, and tympany on the left side of the median line.

A diagnosis of perforation was made and immediate operation advised and this was accepted. Under ether sixteen hours after perforation, the abdomen was opened through the right rectus muscle, evacuating a large quantity, at least 1,000 c.c., of peasoup-like fluid. There was general peritonitis, and an opening large enough to admit the tip of the little finger was found in the ileum, about twenty inches from the caecum. This was closed with one row of silk sutures. A tumor was then felt in the sigmoid flexure of the colon, and on bringing it forward for examination the finger was accidentally thrust through the diseased gut. As there was not time to resect and anastomose the tumor, an adeno carcinoma was removed, the lower portion of bowel closed and the upper segment was brought through the abdominal wound, making an artificial anus. The abdominal cavity was irrigated freely with salt solution and strips of gauze put in for drainage. The patient stood the operation well but death occurred 33 hours later. The necropsy showed general peritonitis, enlarged spleen, and 12 ulcers in the lower part of the ileum.

CASE III.—Two Perforations.-F. D. H., white, male, aged 30, laboratory attendant; was admitted to Providence Hospital with typhoid fever in June, 1902. June 14, in the third week of the disease, a dose of castor oil was given. June 15 he had a moderate hemorrhage from the bowels and the temperature fell from 102 to 101, pulse kept about 100. The patient did not suffer much, but stated that just before the hemorrhage occurred he felt something in his abdomen give way. A diagnosis of perforation was made but was not concurred in by my colleague, so that I did not insist on operating. June 16 he had another hemorrhage but his symptoms were not alarming, although there was now pain and tenderness in the right iliac region. June 17 his condition was about the same, with tenderness, pain and swelling. June 18, early in the morning, he began to vomit and the pulse was up to 140, mind perfectly clear. There was no longer any doubt as to the existence of perforation, and under ether, three days after perforation, the abdomen was opened through the right rectus muscle. There was general peritonitis, the exudate being thick and attached to intestine and omentum. Two perforations, one large enough to admit the tip of the little finger, the other slightly smaller, were found in the ileum about two feet