A retrospective review of patients with a preoperative diagnosis of appendicitis or thought to have appendicitis at laparotomy showed a discrepancy between the surgeons' opinion of the macroscopic appearance of the appendix and the pathologist's opinion (which was assumed to be the most accurate) in 14.5% of cases. All cases were false positive diagnoses. Using the surgeons' descriptions of the appendix at appendicectomy, a sub-group can be identified with an 80% error rate. In this group, the terms used to describe the appendix were 'mild, early, slight or moderate inflammation'. This error rate is often ignored in studies on appendicitis. We recommend that only the terms 'normal' 'inflamed' or 'gangrenous or perforated' should be used to describe the appendix. If there is any doubt, it should be classed as normal. These recommendations will improve diagnostic accuracy at operation.