The effects of adriamycin were compared to a combination program of methyl-CCNU and imidozole carboxamide (DTIC) in 44 patients with disseminated malignant melanoma. There were objective clinical responses in 6 of 21 patients with the combination of DTIC and methyl-CCNU who received this program as primary treatment and none in 23 patients receiving adriamycin as primary treatment. Secondary responses were not observed with either treatment regimen. Toxicity with the combination program included leukocyte depression (less than 3,000/cu mm) in 25% and platelet depression (less than 100,000/cu mm) in 40% compared to 52% leukocyte depression and 16% platelet depression after adriamycin. There were no responses after the combination treatment program in the absence of myelosuppression. There was nausea and vomiting in virtually all patients, which was moderately severe in one third of the patients receiving the combination and in only 10% of those receiving adriamycin. Alopecia developed in all who received adriamycin but in only 15% of the combination treatment group. The combination treatment response of 28% was of the same order as most response rates previously reported in this disease. This randomized controlled clinical trail found adriamycin without clinical benefit and not worthy of further trial in patients with disseminated malignant melanoma.