Sixteen cohorts of men aged 40-59 years at entry were examined with the measurement of some risk factors and then followed-up for mortality and causes of death for 25 years. These cohorts were located in the USA (1 cohort), Finland (2), the Netherlands (1), Italy (3), the former Yugoslavia (5), Greece (2), and Japan (2), and included a total of 12,763 subjects. Large differences in age-adjusted coronary heart disease (CHD) death rates were found, with extremes of 45 per 1000 in 25 years in Tanushimaru, Japan, to 288 per 1000 in 25 years in East Finland. In general, higher rates were found in the US and Northern European cohorts as compared to the Southern European and Japanese cohorts. However, during the last 10 years of follow-up large increases of CHD death rates were found in some Yugoslavian areas. Out of 5 measured entry characteristics treated as age-adjusted levels (serum cholesterol, systolic blood pressure, cigarette smoking, body mass index and physical activity at work), only serum cholesterol was significant in explaining cohort differences in CHD death rates. Over 50% of the variance in CHD death rates in 25 years was accounted for by the difference in mean serum cholesterol. This association tended to decline with increasing length of follow-up, but this was due to the great changes in mean serum cholesterol in the two Yugoslavian cohorts of Velika Krsna and Zrenjanin. When these two cohorts were excluded the association increased with time. Changes in mean serum cholesterol between year 0 and 10 helped in explaining differences in CHD death rates from year 10 onward. It can be concluded that this study suggests that mean serum cholesterol is the major risk factor in explaining cross-cultural differences in CHD.