It has been suggested that high fat, high cholesterol, and low fibre intakes play a role in the causation of colon cancer, but since they are highly intercorrelated, it is difficult to determine which (if any) variable is truly related to colon cancer. Food disappearance data for 1954--65 and mortality data for 1967--73 from 20 industrialised countries were used to assess which variables are independently related to colon cancer. Simple correlation analysis indicated that intake of total fat, saturated fat, monounsaturated fat, cholesterol, and fibres are each highly correlated with mortality-rate for colon cancer. The partial correlation of dietary cholesterol with colon cancer remains highly significant when fat or fibre is controlled. However, the partial correlations of fat or of fibre iwth colon cancer are no longer significant when cholesterol is controlled. Cross-classification showed a highly signficant main effect for cholesterol, but nor for fat or fibre. The findings support the possibility of a causal relationship between cholesterol intake and colon cancer.