A retrospective study of 103 patients with histologically proved cutaneous malignant melanoma was made in order to assess the validity of a simple prognostic index, based upon a multivariate analysis of weighted pathological and clinical factors, in predicting the expected outcome in a given case. The results of the study were almost identical to those observed by Cochran (1968); 16 per cent of the patients with a low prognostic index (less than 20) died or developed recurrence of melanoma during the period of follow-up, in contrast to those with a high prognostic index (greater than 25) of whom 64 per cent were similarly affected. It is concluded that the predicted outcome based upon such an index is valid for the 5 postoperative years in some 90 per cent of patients with cutaneous malignant melanoma. It is suggested that the use of such an index may be of great value in selecting patients in whom the expected outcome to surgical therapy is poor and in whom adjuvant therapy may be of value.