Relative incidence rates for colon and rectal cancer vary greatly between populations in the world. While Japanese have historically had low prevalence, immigration to the United States has now resulted in equal if not higher rates than in Caucasian- or African-Americans. Furthermore, recent data from some population-based registries in Japan itself are also pointing to particularly high susceptibility. Of particular interest is the fact that Japanese in both the home country and the US in fact have far higher rates for rectal cancer than the other two ethnic groups. An intriguing question is whether they might also demonstrate variation from Caucasian- and African-Americans in the raltive incidence rates for proximal and distal colon cancers, given the clear differences in risk factors like diabetes, physical exercise, smoking, alcohol consumption, meat and fish intake and calcium exposure which have been shown to operate in these two sites. A comprehensive epidemiological research exercise is here proposed to elucidate ethnic variation in colorectal cancer development, based on cross-cancer registry descriptive and case control approaches. It is envisaged that additional emphasis on screened populations should further provide important insights into causal factors and how primary and secondary prevention efforts can be optimized.