Male patients with proven urethral strictures (total 210) were prospectively randomised to undergo either dilatation (106 patients) or internal urethrotomy (104 patients) as an outpatient procedure under local anaesthesia. The incidence of complications or failure during the performance of the procedure did not differ significantly between the two treatment groups. Complications or failure in performing urethral dilatation were significantly more common in patients who presented with retention or complications compared with symptoms only, and in those with a positive compared with negative urine cultures. Complications or failure in performing internal urethrotomy were significantly more common in patients with a positive than with a negative urine culture, and long (> 2 cm) rather than short (< 2 cm) strictures, whereas the difference approached significance for patients with multiple rather than single strictures (P = 0.06). Failure alone in the performance of internal urethrotomy was significantly more common in patients with trauma compared with urethritis as aetiology, and in those without previous stricture treatment. With a mean follow-up of 15 months the cumulative percentage of recurrent urethral strictures did not differ significantly between the two treatment groups. We conclude that urethral dilatation and optical internal urethrotomy under local anaesthesia are equally successful as initial outpatient treatment. With regard to successful performance of the procedure itself, multiple, longer (> 2 cm), post-traumatic, and previously untreated strictures are better managed with dilatation, whereas patients with complications or retention are better managed with internal urethrotomy. A positive urine culture is associated with a higher complication and failure rate in the performance of both procedures.