Abstract A case of perforated duodenal ulcer is reported in which the peritoneal cavity was not drained. Subsequently a pelvic abscess developed. The most likely origin of the abscess was an accumulation of fluid which had gravitated to the pelvis from the site of the perforation. The present trend in the treatment of perforation is away from drainage. It is thought to be futile if not harmful. A case is cited also in which much serous fluid was drained from the pelvis several days after the closure of a perforation. As a result of the leakage from a perforation peritoneal contamination is inevitable. The type of infection and its virulence must always remain an unknown factor. Prolonged drainage of the abdomen is admittedly impossible but clinical experiments with a dye indicate that the peritoneal cavity, contrary to general opinion, can be drained for thirty-six hours through a tube in the pelvis. Therefore, as a means of preventing the occurrence of peritonitis and the formation of pelvic abscess, drainage deserves reconsideration.