ROENTGEN examination of the stomach with the aid of a barium meal is generally considered entirely innocuous, on which account barium studies of the stomach are employed more or less routinely and without discrimination or restriction. That this method of examination, however, is not altogether harmless is attested to by the number of cases in which radiographic investigation with a contrast medium is followed by acute perforation of a peptic ulcer. Publications dealing specifically with the subject of gastro-duodenal rupture as a sequel to the employment of the x-ray include 36 patients, 34 of whom had benign lesions and two malignant. Of the 34 perforated peptic ulcers, Panek reports eight, Eckman five, Hummelman four, Bittrolff three, Amberger, Colmers, and Nussbaum each two. Single instances are recorded by Ein Waldt, Frankenthal, Hijmans van den Bergh, Lang, Meijer, Rosenthal, Steiger, and Utili. Of the two carcinomatous ulcers which perforated following x-ray examination, one is reported by Lepennetier and Deruas and the other by Rovida. That the number of cases published under appropriate titles is by no means representative of the actual incidence is indicated by the frequency with which perforation following x-ray is mentioned incidentally in cases presented for other reasons. Casual mention of rupture occurring shortly after roentgenography is found, for instance, in the articles of Brütt, Dahm, Falta, Fermaud, Finsterer, Speck, and Wolfson and Gray, dealing primarily with other phases of perforated ulcer. Furthermore, there is evidence to show that experience with this accident is not confined to only a few but that it has been observed by many. Eckman states that after citing his five cases at a regional meeting of the American College of Surgeons, “Numerous admissions of such catastrophes were forthcoming.” I have met with the same response at meetings where I have had occasion to mention my own cases. Assuming that the cause and effect relationship between perforation and x-ray examination is a definite one, the question arises as to how serious a hazard is constituted by this diagnostic procedure. The only published statistical reports upon which to base an opinion are those of Panek, Hummelman, and Eckman. Panek writes that among 36 cases of perforated peptic ulcer observed in the Second Surgical Clinic of the University of Vienna between the years 1910 and 1926, eight followed roentgen examination with a contrast medium. These eight cases are divided by the author into two groups, the first including four cases in which rupture occurred within several hours of the x-ray observation, and the second group comprising the remaining four in which perforation followed after an interval of between 30 hours and six days. If the time element in Panek's series is disregarded, it is found that 22 per cent of his perforated ulcers followed the x-ray.