1.1. A study of 136 consecutive tubal pregnancies which occurred in a 30 month period is presented. The racial incidence was 84.5 per cent Negro and 15.5 per cent Caucasian. The ratio of ectopic pregnancy to term pregnancy in our clinic was 1:56 in Negroes, and 1:88 in Caucasians.2.2. There were four fatalities, a gross rate of 2.9 per cent, which is correctable to a fatality rate of 0.8 per cent.3.3. The factors contributing to mortality in tubal pregnancy are: (1) failure of diagnosis; (2) delay in surgery; (3) inadequate blood replacement; (4) elective surgery beyond the immediate need; (5) the patient's delay in seeking care.4.4. Sterility was a common factor; 54 per cent of the patients in the study had been sterile for over four years.5.5. Pelvic inflammatory disease was proved by pathologic study in only 28 per cent of the cases. We believe that salpingitis, although an important proved cause of tubal pregnancy, today appears to play a less prominent role in the etiology, and this point needs to be somewhat clarified in our teachings.6.6. The history, gentle yet thorough pelvic examination, and an “ectopic pregnancy conscious” attitude, when combined with centesis of the cul-de-sac, will usually permit the ready diagnosis of extrauterine pregnancy. The most important diagnostic finding at time of pelvic examination, in our hands, is tenderness on manipulation of the cervix. The most valuable aid in diagnosis for us has been aspiration of the cul-de-sac with a 17 gauge needle. This was positive in 95 per cent of the cases of this study.7.7. Only essential surgery should be performed in patients with tubal pregnancy, and salpingectomy should be done so as to spare the adjacent ovary with its blood supply intact in every possible instance. Appendectomy at the time of laparotomy for ectopic pregnancy is an inadvisable surgical procedure.8.8. The postulation that defibrination of circulating blood might occur in ruptured tubal pregnancy is suggested. In the few cases in which we made tests, the circulating blood fibrinogen levels were within normal range, but no fibrinogen nor accelerator globulin was found in the fluid blood obtained from the abdomen at the time of laparotomy in the five cases so studied.