AIMS--To determine why mucosal biopsy specimens of the gastrointestinal tract were taken and whether they were justified on clinical or pathological grounds. METHODS--A prospective audit of 190 consecutive biopsy specimens received in a university hospital histology department over six weeks. RESULTS--The 31 separate presenting symptoms included diarrhoea (34%), abdominal pain (16%) and rectal bleeding (15%). In 41% (78/190) the histology was normal, 28% (53/190) showed inflammatory changes and 11% 21/190) carcinoma. A clear justification for the procedure was identified in over 90% (171/190) of patients. In 36% (68/190) there was a change in patient management on receipt of biopsy reports and further investigations were ordered in 29% (55/190). The mean time taken to report biopsy specimens was 4.7 working days and there was no difference between the reporting time of a pathologist compared with a consultant or a trainee. CONCLUSIONS--There is no evidence that mucosal biopsy specimens are taken unnecessarily.