During the three years 1967-69, 781 cases of squamous cell carcinoma of the skin were reported to the Manchester Regional Cancer Registry. The proportions of males to females were significantly different (p < 0 001) among the skin cancer sites. The age-specific incidence rates were significantly different (p < 0-001) between the sexes for the five-year age groups of 55 years and above. Full occupational histories were obtained on 598 (77 %) patients; a further 148 (19 %) patients gave one main occupation only, while the remaining 35 (4 %) patients were untraced. The numbers of patients observed in broad occupational groups (occupational orders) were compared with the numbers expected using the 1931 and 1951 censuses. For all skin cancer sites combined the occupations of farming and textiles were found to have highly significant excesses of 1500% and 13500 respectively for males. The corresponding excesses for females were 300% for textile workers and varied from 1140 % to 590 % for farmers, but only for the farmers were the excesses highly significant. For males the occupation of metal worker also showed excesses of 38 % and 23 % which were of borderline significance. The association between occupation and individual skin cancer sites was then considered. For males there were excesses in the arm for the occupational orders of chemical workers, paper/printing workers and fishermen, and in the ears for builders, but these excesses were of borderline significance. There was a significant difference (p < 0-05) in the proportion of male patients with atopic skin conditions in each cancer site. However, this was not found for the female patients. For both male and female patients no significant associations were found between the skin site and either eye colour, residence in the tropics or smoking habit.A review of scrotal cancer in Scotland in 1967(Lloyd Davies, 1970 showed that, of the 18 confirmed cases from the Hospital Inpatient Index, only 12 were probably of occupational origin. A morbidity study by Lee et al. (1972) found that, of 89 cases of scrotal cancer, 84 were probably of occupational origin. The occupations at risk have been shown to be textile workers, in particular mule spinners, and metal workers, in particular automatic lathe operators. Occupations possibly at risk are road workers, dye workers, chain makers and 'Stanford jointers' (Henry, 1946;Cruickshank and Squire, 1950;Spink et al., 1964;Lee et al., 1972). Because scrotal cancer may be regarded as one specific site of skin cancer the questions arise,