One hundred seven patients presenting with malignant melanoma of the extremities were included in a prospective randomized study, which was conducted to evaluate the effectiveness of adjuvant hyperthermic regional cytostatic perfusion. In a control group (A, N = 54) the tumors were excised widely and the regional lymph nodes were dissected. The patients in the second group (B, N = 53) received additional hyperthermic (42 C) perfusion with melphalan. The mean follow-up observation period was 554 days. We chose the disease-free survival time as the criterion for success. The study could be discontinued prematurely, since the intermediate evaluation revealed a highly significant difference between the groups (p = 0.0001). We observed 21 local recurrences in the control group and four recurrences in the perfusion group. The retrospective breakdown by clinical stages also showed significant differences. The recurrence rate in the control group was 27.8% in Stage I, 31.6% in Stage II, and 58.8% in Stage III. In the perfusion group we observed recurrences equaling 5.6% in Stage I, 5.5% in Stage II, and 12.5% in Stage III. The differences between the groups based on the target-criterion of disease-free survival represent significance levels of p = 0.09 in Stage I, p = 0.03 in Stage II, and p = 0.003 in Stage III. We feel that on the evidence provided by our study, shown in the above results, the adjuvant application of regional hyperthermic cytostatic perfusion has proven itself to be superior to conventional procedures alone.