This review provides an update on nutritional aspects of carotenoids (as distinct from retinoids), with specific relevance to anticarcinogenesis. Critical gaps remain in our knowledge of the nutritional functions of carotenoids despite an overwhelming accumulation of research data in areas tangential to human nutrition. In addition to their roles as precursors of retinol and retinoids, carotenoids have distinct functions of their own in animals and humans. In vitro they are antioxidants with a broad range of potencies. In vivo, they protect porphyrics against sunlight. The evidence for anticarcinogenic actions of beta-carotene in certain specified test situations is persuasive. Nevertheless, despite a large number of studies demonstrating protection by carotenoids, the characteristics that render a given carotenoid effective and the relative efficacy of the individual carotenoids are not known. Moreover, dose-response and pharmacokinetic relationships remain virtually unexplored. Research to uncover mechanisms of protection by carotenoids is, for technical reasons, painfully slow. Epidemiological studies reveal associations but not cause and effect. To explore cause and effect, intervention trials are underway, hampered by the paucity of data regarding optimal choice of carotenoid, dosage, and regimen. The in vitro test systems that would provide this information are not available because the molecular sites relevant to the chemopreventive action of carotenoids are obscure. Each of these problems has a solution, but not a simple one. Until these are resolved, blanket recommendations regarding supplementation will remain problematic. To this point, health authorities have not recommended dietary supplementation with carotenoids. Instead, they recommend increased consumption of yellow and dark green carotenoid-rich vegetables. In the future, an individual at risk for a particular carcinogenic process may be recommended a supplement of the most appropriate anticarcinogen, specific to their individual endowment of genetic and environmental risk factors. This review emphasizes not only what is known but also what is not known. Consequently we identify priorities for research that, if undertaken, will allow such recommendations to be made or discounted with more confidence.