Abstract The only conclusion that we can draw from these observations is that one must individualize the treatment for each hemorrhaging ulcer that is encountered. The following points should be stressed: (1) that patients may die from hemorrhage under conservative management regardless of whether they are transfused or not; (2) if the patient is operated we feel that it is essential to remove the ulcer or ulcers and this can best be accomplished by a subtotal resection; (3) hemorrhage does occur frequently following gastroenterostomy and the hemorrhage apparently comes from the original ulcer and not from a marginal ulcer, although this occassionally does happen. A negative roentgen study does not exclude a marginal ulcer, but, if the patient is free from pain plus a negative roentgenogram then one is fairly certain that a marginal ulcer does not exist. There were 12 cases that bled after gastroenterostomy for chronic ulcers and 4 that had recurring hemorrhage after gastroenterostomy for bleeding ulcers. In other words, of the 123 bleeding ulcers, 16 or 13 per cent, had had gastroenterostomies performed but have had severe hemorrhage following the gastroenterostomy.