Summary In the BUPA Study, a prospective study of 22,000 men attending a screening centre in London, serum samples were collected and stored. The concentration of beta-carotene was measured in the stored serum samples from 271 men who were subsequently notified as having cancer and from 533 unaffected controls, matched for age, smoking history and duration of storage of the serum samples. The mean betacarotene level of the cancer subjects was significantly lower than that of their matched controls (198 and 221 Mg I1 respectively, P=0.007). The difference was apparent in subjects from whom blood was collected several years before the diagnosis of the cancer, indicating that the low beta-carotene levels in the cancer subjects were It has been suggested that beta-carotene (and other carotenoids) may play a role in reducing the incidence of cancer (Peto et al., 1981). Beta-carotene has anti-oxidant activity (Burton & Ingold, 1984); it is an efficient quencher of singlet oxygen (Foote, 1979). Singlet oxygen is a toxic, and possibly cancer inducing, form of oxygen that occurs as a result of many metabolic reactions.There is limited evidence to suggest that beta-carotene supplementation reduces the risks of chemically induced tumours in animals. Since an important function of carotenoid pigments is to protect organisms from photosensitisation, and hence probably skin cancer which can occur as a result, the administration of beta-carotene to animals is likely to reduce the incidence of skin cancers occurring in those animals. What is, perhaps, of greater interest is whether the administration of beta-carotene can reduce the incidence of cancers that are not induced by ultraviolet light, and in particular of cancers that affect tissues other than skin. Dorogokupla (cited by MathewsRoth, 1982) induced subcutaneous tumours in rats with injections of 9,10-dimethyl-1-2-benzanthracene (DMBA) and skin tumours in mice by the topical application of DMBA; animals fed a diet supplemented with unlimited amounts of red carrots developed tumours at a lower rate than did the animals receiving the unsupplemented diet. Mathews-Roth (1982) administered about 6.7 grams of beta-carotene per kilogram of diet per day to mice and showed that this led to a considerable increase in pigment accumulation in the skin (593 mg 100 g-1 skin) and that the induction of skin tumours by 7,12-dimethylbenzanthracene promoted by croton oil, was inhibited by the stored beta-carotene, there being 3.4 tumours per mouse in the beta-carotene group compared to 11.6 tumours per mouse in the control group (P<0.01). Mathews-Roth used canthaxanthin, a carotenoid without vitamin A activity, as a control substance. It showed no significant anti-cancer activity. Rettura and his colleagues (1982) found that beta-carotene (and retinyl (Stocks, 1958;Bjelke, 1975;Phillips, 1975;MacLennan et al, 1977;Bjelke, 1978;Tuyns et al., 1978;Cook-Mozaffari et al., 1979;Hirayama, 1979;Mettlin et al., 1979;Gregor et al., 1980;Mettlin et al., 1981;Modan et al., 1981;Shekelle et al....