315 patients with untreated penile warts were recruited. Patients were excluded from the study if they had non-penile warts requiring separate treatment, risk for HIV infection, another painful penile condition, intrameatal warts, had already received some treatment in the previous 12 months or were aged under 16. Patients were randomly allocated to 3 groups of treatment. 106 were allocated to 3 ml of PDX 0.5%, 103 to 5 ml of PODO 0.5% and 106 to 5 ml of PODO 2.0%. The groups were comparable in terms of entry characteristics such as the length of time warts were present, the number of warts and their type.277 patients were protocol eligible (19 did not conform to the entry criteria and 52 "violated" the protocol). Results were obtained from 158 patients (65% of protocol eligible, 50% of all entrants) who attended the clinical examination in week 5 or who returned a questionnaire. The trial size gave the expected 90% power to detect medium differences at the 5% significance level which was estimated by using Cohen's methods.