A case-control study including 78 patients with malignant melanoma of the skin and 131 controls with the diagnoses of malignant lymphoma, testicular cancer or bone and soft-tissue sarcoma, was performed at the Norwegian Radium Hospital 1974-1975. The questionnaire contained questions for evaluating the comparability between cases and controls, and questions bearing on the relation between sun exposure and malignant melanoma. No essential difference between cases and controls was found as regards hair and eye colour, time spent outdoors during leisure, and degree of solar exposure. The melanoma patients liked sunbathing less than controls, worked more outdoors, more often went to Southern Europe for sunbathing and more often used sun lotions. These differences, however, were not clearly statistically significant. Highly significant differences were demonstrated as regards the tolerance of sun exposure and propensity to freckling. The melanoma patients tolerated sun exposure less well and freckled more easily than the controls. Although attempts were made to eliminate bias, there are still limitations and loop-holes in the study, and the relative risks demonstrated are not large enough to be of great immediate practical or scientific value. It seems justified, however, to advise persons with a low sun exposure tolerance to be cautious in following the sun-tan-demanding fashions. The study may also provide certain clues for the planning of future epidemiological and clinical studies regarding the etiology of malignant melanoma.