A uniform etiology of this affection is not admitted by a number of writers, according to whom it is probably a sequel of a great variety of pathological conditions of the genital system, including gonorrheal infection and puerperal complications, as well as perityphlitic disease, congenital anomalies, and transmigration of the ovum. A certain connection between salpingitis and tubal pregnancy can hardly be denied, on the basis of the more or less incomplete clinical statements, as well as the anatomical findings. Suppurative changes were much in evidence in the specimen of my case, but there was no history of gonorrheal infection.

Concerning the finer mechanism of the faulty inbedding of the ovum, a good deal of uncertainty still prevails. Rabinowitz concludes that pathological and clinical studies furnish sufficient data to justify the deduction that gonorrheal salpingitis is the predominant cause of tubal pregnancy and that the lesion, most likely responsible, is the destruction of the ciliated epithelium. However, the presence of fairly complete and even actively motile cilia has been demonstrated by Zedel and other investigators. The absence of cilia may, moreover, be an artefact. Also, it must not be overlooked that the persistent hyperemia of pregnancy may give rise to processes entirely analogous to chronic inflammation. Again, the bacteriology of chronic perimetritis is not very well known, and Puppel calls attention to the fact that existing pelvio-peritonitic adhesions are not necessarily of gonorrheal character. The formation of solid adhesions is regarded by him as the cause of the repeated tubal

pregnancy in his two cases, and in his opinion, it makes no difference if these are derived from gonorrhea, puerpeal fever, a drained laparotomy, or a slight postoperative infection.

The best prospects for the elucidation of the etiology of tubal pregnancy in general, and its recurrence in particular, are held out by the utilization of all new specimens for careful microscopical study. In the case recently operated upon by me, the excised tubes have been photographed and examined under the microscope. The report of the pathologist follows: PATHOLOGY OF THE OVIDUCTS IN CASE OF EXTRA-UTERINE PREGNANCY OCCURING IN BOTH TUBES AT SHORT INTERVALS.

The fetal tumors were enclosed in unruptured tubes. Length of left fetal tumor 30 m. m.; breadth, 25 m. m. Length of right fetal tumor 25 m.m.; breadth, 15 m. m. In both tubes the fetal structures are outlined in the centre of blood clots and attached

to the posterior walls. Sections are taken from the tubes close to the fetal growths and from the tumor region.

The external connective tissue coats are not distinct from the muscular coats. The external longitudinal muscular coats consist of a loose connective tissue containing hypertrophied smooth muscle in scattered bundles. They are, in both cases, infliltrated with pus, which is more concentrated in some regions than in others; especially dense is it around the small blood

vessels. The internal circular muscular coats show much less pus infiltration, but small areas appear here and there, separating the bundles of muscle. The mucous coat shows the ciliated folds increased in size and deeply infiltrated with pus. The epithelium is normal in some places, while in most parts of the sections, the cilia are absent, and the cells are degenerated. It is quite possible that the absence of cilia from those cells which maintain their usual sizes and shapes, is due to manipulation. There is very little pus in the lumina of the tubes. As far as the tubes are concerned, they both show evidences of a suppurative inflammation with degeneration of the ciliated epithelium. The process is more intense in the left tube than in the right.

PATHOLOGY OF THE WALLS OF THE TUMOR.

The external longitudinal muscular coat shows an extensive suppuration. Dense areas of pus occur at frequent intervals and especially around the blood vessels. The internal circular muscular coat shows some pus collections, but much less than the external coat. The mucous coat shows an adenomatous arrangement of folds covered with columnar epithelium, beneath which are dense accumulations of embryonic and pus cells. Pus and blood with a central fetal structure form the contents of the tumor. Here and there scattered through the cell areas are amylaceous bodies. A suppurative inflammation is therefore shown in the walls of the tumor. In all sections the suppuration appears intense in the outside coats, light in the middle coat, and intense in the mucous coats. The order of appearance is not evident.

In the case of bilateral successive tubal pregnancy, described by Kermauner, the two ova were implanted in approximately the same region, at the transition into the ampullary segment of the tube. The time of occurrence of the first pregnancy was difficult to ascertain, on account of the imperfect anamnesis, but the histological picture indicated that the ovum in the left tube was the older of the two. The patient was a woman 37 years of age, who was operated upon on account of ruptured tubal pregnancy of the right side; the coagulum contained a fetus 4 to 5 cm. in length. The left tube, which was enlarged and club-shaped, was ablated at a distance of 2 cm from the uterine end. The description of the specimen illus. trates the insertion of both ova in the muscular layer of the tubal wall. The right tube formed a cylindrical body, 8.5 cm. long and 3 cm. thick, covered superficially with hemorrhagic adhesions. At a distance of 2 cm. from the beginning of the