In the seven years following initial examination (1964--65) 758 deaths were observed in 11,121 Yugoslav men aged 35--62. The mortality rate was similar to that found for men in the Framingham (Massachusetts) study. However, coronary heart disease (CHD) mortality was much lower in Yugoslavia but mortality from cancer, accidents and violence and respiratory disease was much higher. The excess cancer mortality was chiefly from stomach cancer. The excess respiratory mortality was from both tuberculosis and chronic obstructive respiratory disease. Within Yugoslavia there were area differences in mortality by cause but little difference in total mortality: in Tuzla (Bosnia) there was a higher respiratory disease mortality but a lower mortality from accidents and violence than in Remetinec (Croatia). In the rural areas of the study there was a higher mortality from respiratory diseases and accidents and violence than in urban areas but a counterbalancing lower mortality from cardiovascular diseases. The low coronary death rates and high cancer of the stomach death rates in this population present an important epidemiological challenge.