The absence of a simple, clinically relevant, animal model of inflammatory bowel disease (IBD) hampers research into this disease. In this study, colitis was induced in rabbits by intracolonic installation of 2,4,6-trinitrobenzene sulphonic acid (TNB) in 25% ethanol. Rabbits were killed from zero hours to 6 weeks and their colons examined. Rabbits were examined by endoscopy at weekly intervals. A single dose of TNB in ethanol produced dose dependent inflammation and ulceration, which at its optimum (40 mg) resulted in cobblestoning, strictures, and bowel wall thickening. The damage score at endoscopy was consistent with the score on macroscopic examination of the colon. Histopathological features of inflammation and ulceration observed in all animals that received 40 mg TNB included crypt abscesses, ulceration, crypt architectural distortion and, occasionally, granulomas and pseudopolyps. These changes, which are similar to those observed in IBD, persisted for 6 weeks. No lasting abnormalities were observed in control animals treated with TNB in saline, with ethanol alone, or with saline only. Histopathological similarity and the prolonged duration of inflammation, compared to other models, make this a suitable model for investigating inflammation in the colon. Furthermore, the model is accessible to endoscopy which adds to its value in experimental studies.