The 1978-1979 mortality rates for cancers of the breast, prostate, ovary, and colon in 26 to 30 countries were related to the average 1979-1981 food availability data published by the United Nations. The previously described relationship between breast cancer mortality rates and animal fat consumption continues to be evident, and applies also to the other three tumor types. The correlation with breast cancer was particularly strong in postmenopausal women. Since 1964, particularly notable increases in both breast cancer mortality rate and dietary fat intake have occurred in those countries with a relatively low breast cancer risk. The international comparisons support evidence from animal experiments that diets in which olive oil is a major source of fat are associated with reduced breast cancer risk. The excess in mortality rates for breast and ovarian cancer in Israel relative to the national animal fat consumption may be due to the mixed ethnic origin of the Israeli population. Positive correlations between foods and cancer mortality rates were particularly strong in the case of meats and milk for breast cancer, milk for prostate and ovarian cancer, and meats for colon cancer. All four tumor types showed a negative correlation with cereal intake, which was particularly strong in the case of prostate and ovarian cancer. Although, in general, there was a good positive correlation between prostate and breast cancer mortality rates and between prostate cancer and animal fat, discrepancies in national ranking indicate the operation of other etiologic factors that modify risk. The observed positive correlations between the four cancer mortality rates and caloric intake from animal sources, but negative correlations for vegetable-derived calories, suggest that, of the two, animal fat and not energy is the major dietary influence on cancer risk.