News of Science.

Two corrections should be made in the item "Blood groups and disease" [Science 124, 674 (12 Oct. 1956)] which reports a paper on associations between blood groups and disease read to the British Association for the Advancement of Science. The three diseases for which the evidence is overwhelming are (i) duodenal ulcer, (ii) gastric ulcer, (iii) cancer of the stomach. As is stated, the ulcers are commoner in persons of group O, and cancer of the stomach in persons of group A. There is fairly strong evidence in connection with diabetes mellitus and pernicious anemia; but the incidence is raised in group A, not groups O and B. It should perhaps be added that practically all the data so far come from Western Europe and North America, and the associations might be different or absent in other peoples.