Epidemiologic evidence on the relation between nutrition and bladder cancer is reviewed. A role of diet and nutrition in bladder carcinogenisis is plausible since most substances or metabolites, including carcinogens, are excreted through the urinary tract. Ecologic studies on populations have found positive correlations between fats and oils and bladder cancer, but these are reflected only partly in the international differences in bladder cancer rates, which are systematically higher in Europe than in the United States. Ten case-control and three cohort studies of bladder cancer published in English between 1979 and 1994, and including some information on dietary factors, were reviewed. Of seven studies which considered various types and measures of fruit and vegetable consumption, six found a reduced risk with increasing consumption, which was more consistent for vegetables, with relative risk (RR) estimates between 0.5 and 0.7 for the highest cf the lowest consumption level. There is, therefore, suggestive evidence that a diet rich in fresh fruit and vegetables is a correlate-or an indicator-of reduced bladder cancer risk. No clear association emerged for other foods investigated, including meat and milk. With reference to nutrients, total fat intake was related to bladder cancer risk in three case-control studies, with relative risks between 1.4 and 1.7 for the highest cf the lowest consumption level. However, no relationship between fats and bladder cancer emerged in c cohort study on Japneese-Americans in Hawaii. No consistent association emerged between protein or carbohydrate consumption and bladder cancer risk. Among micronutrients, vitamin A, and particularly carotenoids, showed an inverse association with bladder cancer risk in four case-control studies, including one allowing for a measure of total caloric intake, but were not related consistently in two other studies. There oere only scattered and inconclusive data on vitamin C and E. Finally, two studies suggested that calcium and sodium and sodium may be related to bladder cancer risk. Thus, available data on diet and bladder cancer are still inconclusive. This is at least partly attributable to the limited number of cohort studies and the paucity of case-control studies, including satisfactorily detailed and validated dietary questionnaires. Despite these limitations, available data suggest that a diet rich in fresh fruit and vegetables, and, hence, possibly in carotenoids, is a correlate of reduced bladder cancer risk.