We have carried out a study of large malignant skin tumours (squamous-cell carcinomas) and other lesions in "hairless" mice (in groups of 45 or 60 mice) intermittently exposed to ultraviolet light over a period of 15 weeks, beginning when the mice were about 8 weeks old. Various groups were given a standard diet (Wayne Lab-Blox) or the same food with added vitamin C or vitamin E throughout the study. Lesions, classified by histopathologic study as atypical squamous-cell proliferations varying from early actinic keratoses to invasive poorly differentiated squamous-cell carcinomas, had begun to develop by the end of the period of irradiation. They were counted twice a month for five months. The observed fraction of mice that developed lesions during successive time periods was analyzed by the statistical method recommended by a committee of the International Agency for Research on Cancer. A pronounced effect of vitamin C in decreasing the incidence of the malignant lesions was observed with very high statistical significance. No significant effect of vitamin E was observed. We conclude that vitamin C should be given special attention with respect to the relation between diet and cancer.