Several lines of evidence suggest that ulcerative colitis could be caused by excessive bacterial production of H2S in the colon. A rodent model of colitis involves the feeding of nonabsorbable, carbohydrate-bound sulfate in the form of dextran sulfate or carrageenan. The observation that metronidazole blocks the development of this colitis suggested that the injurious agent could be a sulfur-containing compound (such as H2S) that is released during the bacterial metabolism of the nonabsorbed sulfate. We tested this possibility by feeding rats dextran sulfate, with or without bismuth subsalicylate, a compound that avidly binds H2S. Bismuth subsalicylate reduced the fecal release of H2S in dextran sulfate-treated rats to values well below that of controls. Nevertheless, all the animals developed colitis. We conclude that excessive H2S production does not play a role in the dextran sulfate model of colitis.