Epidemiological evidence consistently shows that diets high in fresh fruit and vegetables significantly lower cancer risk. Given the postulated role of oxidative DNA damage in carcinogenesis, the assumption has been made that it is the antioxidant properties of food constituents, such as vitamin C, E and carotenoids, which confer protection. However, epidemiological studies with specific antioxidants, either singly or in combination, have not, on the whole, supported this hypothesis. In contrast, studies examining the in vitro effect of antioxidants upon oxidative DNA damage have generally been supportive, in terms of preventing damage induction. The same, however, cannot be said for the in vivo intervention studies where overall the results have been equivocal. Nevertheless, recent work has suggested that some dietary antioxidants may confer protective properties through a novel mechanism, unrelated to their conventional free-radical scavenging abilities. Upregulation of antioxidant defence, xenobiotic metabolism, or DNA-repair genes may all limit cellular damage and hence promote maintenance of cell integrity. However, until further work has clarified whether dietary supplementation with antioxidants confers a reduced risk of cancer and the mechanism by which this effect is exerted, the recommendation for a diet rich in fruit and vegetables remains valid empirically.