[Illustration: FIG. 4. A TUBERCULOUS FALLOPIAN TUBE AND OVARY: ENTIRE AND IN SECTION. Caseous matter has exuded through the cœlomic ostium of the tube and become encapsuled. Natural size.] As soon as the diseased parts are extracted, a dab is pressed into the hollow to check the oozing: the pedicle is clamped with forceps and the tube and ovary detached. It is the common practice in dealing with inflamed and septic ovaries and tubes to transfix and ligature the pedicles as in a simple clean ovariotomy. The consequences of this practice are not satisfactory, for the pedicles being infected often give rise to trouble, because the silk acts as a seton, an abscess forms which may open up through the abdominal wound, the rectum, or perforate into the bladder, and leads to the establishment of a sinus which persists for many months until the ligature is extruded. There are several methods of avoiding this: for example, the arteries in these broad pedicles may be ligatured separately with thin silk, and the edges of the peritoneum drawn together by two or three mattress sutures (Fig. 11, p. 40). In cases where the Fallopian tube is thickened quite up to the uterine angle, it may be exsected from the uterus: in such cases the uterine artery will be tied and the flaps at the uterine angle can be brought into apposition by a mattress suture. In acute cases of salpingitis the cœlomic ostium is open and the infective material can be seen leaking from it (Fig. 3). In chronic cases this ostium is firmly occluded (Fig. 4). Acute cases are dangerous as they are apt to cause post-operative peritonitis. Chronic cases are difficult on account of visceral adhesions. The most serious complication likely to arise in the enucleation of a pyosalpinx, especially on the left side, is a firm adhesion to the rectum; this may be occasionally anticipated when the patient gives a clear history of one or more sudden discharges of pus from the anus. An accidental tear of the rectum through comparatively healthy tissues may be repaired by interrupted sutures, but when the injury is in tissues altered by chronic suppuration, the only course open to the surgeon is to drain with a wide rubber tube, and it is surprising as well as gratifying to know that a fistula of this kind low in the rectum will often close in a week or ten days. It is important to bear in mind that an undetected tear into the rectum, if the abdomen be closed without drainage, will, in all probability, lead to fatal peritonitis. It has happened that a surgeon in removing a pyosalpinx tore a hole in the rectum; he was unaware of the accident, and a few hours after the operation ordered 10 ounces of saline solution to be injected into the bowel. This fluid passed through the rent in the gut direct into the pelvis with fatal consequences.