vagina, or as the result of breaking up adhesions about the cervix is readily controlled by the packing, but the operator must positively assure himself that there is no hemorrhage from the higher adhesions or ovarian branches. If these cannot be reached per vaginam, the abdomen should be opened and the vessels secured. I have a specimen in my office, brought to me

few days ago by Dr. Pond, City and County Hospital, where there is a large vein running direct from the left ovarian vein to the left renal, entering the left renal vein about a quarter of an inch inside the spermatic. A vaginal hysterectomy in this case, with removal of the ovary, is often done (and which was done in some of the cases reported), might give fatal hemorrhage unless the vein has valves. These cases show that vaginal hysterectomy is a safe operation; that the younger the patient the greater tendency the disease has to return, and that when the uterus is fixed by adhesions the prognosis is bad. When the uterus is free and movable the operation is satisfactory and will probably yield as good results as the removal of cancer in any other portion of the body; but this requires an early diagnosis, and how can an early diagnosis be made? There is no pathognomonic symptom. Sometimes by removing a portion of the cervix or by curetting, the microscope gives quite positive assistance, not always. The last case reported, operated on ten days ago, I felt morally sure of the diagnosis; Dr. Rivas, who saw the case in consultation, considered the diagnosis absolute. I submitted a small portion that had been removed to Dr. Montgomery, with negative results, notwithstanding, I operated. The adhesions to the bladder and rectum were so absolute and the adnexa so fixed that the uterus could not have been removed without endangering the patient's life. Dr. Montgomery examined the removed cervix with the following results:

"I have examined the cervix uteri you removed from Mrs. E., and although I do not get indubitable evidence of cancerous infiltration, yet many things point that way, and so strongly, that for the patient's sake it ought to be treated as cancerous. In the cervix there are many cysts lined with actively proliferating epithelium, and in some places there are alveoli filled with actively proliferating and a typical epithelium. In no section examined, however, had the epithelial infiltration extended at all deeply. There was also found a considerable amount of hyaline degeneration-a sign also looking to the presence of malignant disease."

The early diagnosis of cancer must be made from the family history, age, emaciation, cachexia, pain, hemorrhage, character of the discharges, and a good deal must finally rest with the personal experience of the examiner. It is not more difficult to diagnose cancer of the uterus, than to differentiate cancer of the stomach, or bowels, or kidney, or bladder, or pancreas, etc.

GASTROSTOMY BY WITZEL'S METHOD FOR PRIMARY CANCER OF THE ESOPHAGUS.

By DUDLEY TAIT, M. D.

(Read before the California Academy of Medicine, May, 1894.) The patient, aged 72 years, came under my observation early in April, 1894. His hereditary antecedents are of considerable interest. His father died of cancer of the breast, and one brother of cancer of the oesophagus at the age of 84 years. No history of having swallowed any corrosive liquids, no syphilis nor alcoholism. With the exception of slight attacks of pharyngitis, his health has always been excellent. Seven months ago he began to complain of dysphagia and of a dull pain in the inter-scapular and precordial regions. At first only solid food caused discomfort, but after the third month a similar condition was induced by the ingestion of liquids. When taking nourishment, patient was aware of the existence of what he called "two obstacles," the first, probably spasmodic, corresponding to the origin of the œsophagus, the second, at about two inches above the lower end of the ensiform appendage. No hæmatemesis nor melæna; frequent regurgitation, emaciation rapid, patient having lost sixty pounds during the last five months; general condition indicates inanition rather than cachexia. The oesophageal bougie (circumference 2.8 centimeters) shows a resistance at a point 12 inches from the teeth and penetrates with difficulty the obstruction which apparently measures 1 inches in length. On being withdrawn, the bougie brings away a brownish mucus mixed with shreds of tissue, analogous to that scraped from the section of a carcinomatous growth. The urine contains no sugar nor albumen; urea 16 grammes per diem. For three weeks, at the patient's request, bougies were passed daily, a painful, dangerous and absolutely useless procedure, inasmuch as each attempt invariably induced severe pain and distress. Fortunately, the irritation and rapid ulceration resulting therefrom did not pro

duce perforation into surrounding cavities nor rupture of the œsophagus. The constriction grew more and more pronounced until the patient could not even pass liquids. During the week prior to the operation a fetid odor was discernible in the regurgitated fluids.

Operation April 21st, 1894. An incision two and a half inches in length was made, beginning one-quarter of an inch to the left of the median line, parallel to and one inch below the border of the ribs. The few fibers of the rectus encountered were easily separated. As soon as the peritoneal cavity was opened the liver and stomach came into view, the latter resembling very much, in dimensions, the transverse colon. The anterior wall of the stomach was seized by two fingers, brought outside, surrounded by sterilized gauze and held by the assistant. After puncturing the stomach with a bisNo. 1-Witzel's method, show-toury, a No. 26 (Charrière) rubber catheter ing application of sutures in wall of stomach. was introduced towards the cardiac orifice, one inch and a half lying within the stomach. (Cut No. 1.) The catheter was then infolded, for a distance

of two inches, by two rows of sixteen Lembert sutures, in accordance with Witzel's1 method. The stomach was returned after inserting four silk stitches into its walls. These stitches were thrust through the abdominal wall and made secure after inserting the silk-worm-gut sutures for the abdominal incision. (Cut No. 2.) The latter were then tied, thus terminating the operation. The tube emerged at the upper third of the cutaneous incision. Aristol and sterilized gauze dressing. A clamp was placed on the end of the tube to prevent the escape of the contents of the stomach.

Rectal alimentation during the first twelve hours. Sixteen hours after the operation a nutritive injection of four ounces was made into the stomach through the tube. Patient stated that it felt good." On the second day these injections were 'Witzel, Centralbl. f. Chir., 1891, 601.

given every four hours, and after the third day the amount was doubled. Each injection comprised three eggs, 100 gms. milk, 60 gms. liquid peptonoids and 15 gms. of whiskey.

The average daily consumption, with slight variance, is 18 eggs, 600 gms. milk, soup or malted milk; 350 gms. liquid peptonoids, and 60 gms. whiskey. The tube escaped through inadvertence on the second day, but no difficulty was experienced in

No. 3-Gastrostomy by Witzel's method, 21 days after operation.

replacing it and no harm resulted therefrom. Twelve hours after the operation the patient began to be troubled with an incessant desire to "clear his throat," probably due to the irritating contents of the oesophageal pouch above the stricture. This symptom was at first sufficiently pronounced to prevent sleep for more than fifteen minutes at a time. The frequent coughing and regurgitation did not, however, prevent good adhesions between the stomach and abdominal walls. The fetid

ity, noted prior to the operation, suddenly increased, coinciding with the expectoration of the bloody mucus. Frequent irrigation of the upper half of the oesophagus with a 4 per cent solution of boracic acid and the injection, followed by the immediate regurgitation, of an infusion of coffee checked in great part these distressing symptoms. The wound gave no trouble; the silkworm-gut sutures were removed on the 13th day. Patient sat up on the 6th day and walked about on the 14th day. There has been absolutely no leakage, consequently the surrounding integument is healthy in appearance. (Cut No. 3.) After giving an injection of 350 gms., if the tube be withdrawn and the patient made to cough while standing, nothing escapes. I ex. pect him later to discard the tube, i. e., to use it only when taking the nutritive injections. He evidently receives sufficient nourishment, as his weight is increasing and the color of the skin and his general condition have greatly improved. Urea, 37 gms. per diem.

Remarks: Too much care cannot be taken in securing the stomach to the abdominal wall. The majority of reported failures may be ascribed to the absence of sufficiently firm and extensive adhesions. For this reason, in the present case, four sutures were employed, two on each side of the tube. I have used as many as seven on the dog and have obtained strong adhesions. Drainage of the peritoneal cavity, or of the oblique artificial canal, is entirely unnecessary. This was, however, resorted to, for reasons unknown to me, by W. Meyer,1 of New York, who has had more experience in gastrostomy for malignant oesophageal strictures than any other American surgeon. Von Hacker's method of utilizing the fibers of the rectus for a sphincter is a purely theoretical view; it looks well on a cadaver, but in the living, careful suturing will prove much more valuable. I draw this conclusion from; first, the presence of the dense fibrous ring at the external opening of the gastric fistula; second, the inactivity of the muscular fibers in Littre's operation for artificial anus; third, the negative results obtained on the dog. The hemorrhage resulting from the incision of the stomach may be immediately arrested by introducing the catheter. An apparently more elegant method would be to use the Paquelin cautery in place of the knife, but the extensive cedema and

1 W. Meyer, Annals of Surg., November, 1893, p. 555.

"Von Hacker, Wien. Med. Woch., 1886; XXXI, 1073-1110 and Wien. Klin. Woch., 1890, 693.