Thirty patients with stones in the common bile duct were allocated alternately to have choledocholithotomy carried out with either T-tube drainage or with primary closure. Choledochoscopy was done during every operation, and the patency of the common bile duct tested by perfusion. There were no operative deaths. The length of operation was shorter with primary closure (p less than 0.01) but there were no differences between the groups in operative blood loss, days in hospital after operation, postoperative morbidity and mortality, and final outcome at follow up. There was no change in the incidence of postoperative bacteraemia, the number of adverse reactions, and the incidence of bile peritonitis after removal of the T-tube. Patients who had T-tubes reported greater discomfort and inconvenience than those without, and their treatment cost more. We conclude that primary closure of the common bile duct is a reasonable alternative to T-tube drainage in selected cases.