Epidemiologic studies of diet and cancer have been facilitated in Hawaii by the multiethnic composition of its population and the consequent heterogeneity in dietary intakes. Studies of migrant populations, particularly the Japanese, have firmly supported the conclusions that environmental factors are of predominant etiologic significance for most major sites of cancer, and that these factors may exert their influences at particular periods of life. Recent observations on Filipino migrants reproduce most of the findings in the Japanese, although they do not show the same abrupt increase in colon cancer rates to the high levels found in Caucasians. Data on dietary intakes in these populations support several of the prevailing hypotheses regarding the etiology of certain gastrointestinal and hormone-dependent cancers. Several case-control studies of diet and cancer have been completed or are ongoing in Hawaii. Some of these have included comparable studies in Japan, but the findings in Hawaii have generally not been reproduced in Japan. Weak associations with dietary fat have been found in Hawaii for breast cancer (particularly in Japanese women) and for prostate cancer (particularly in men greater than or equal to 70 years of age). Vitamin A (especially carotene) has been shown to be inversely associated with lung cancer risk in men, but positively associated with prostate cancer risk in older men. Vitamin C may be inversely related to bladder cancer risk, but has shown no relationship to lung or prostate cancer risk. These and other findings are discussed in terms of future needs for epidemiologic research in this field.