We report the results of a prospective randomised controlled trial of the management of 101 Colles' fractures in patients over the age of 55 years. Within two weeks of initial reduction 43 fractures had displaced with either more than 10 degrees dorsal angulation or more than 5 mm radial shortening. These patients were randomly divided into two groups: 21 were remanipulated and held by an external fixator; in the control group of 22 patients, the redisplacement was accepted and conservative treatment was continued. Patients treated with external fixation had a good anatomical result, but their function was no better than that of the control group. We found no correlation between final anatomical and functional outcome, and concluded that the severity of the original soft-tissue injury and its complications are the major determinants of functional end result.