Over 10,000 male civil servants and municipal employees in Israel, aged 40 years and above, underwent an extensive clinical, biochemical, anthropometric, sociodemographic and psychosocial evaluation in 1963, 1965 and 1968. Follow-up for mortality was continued through 1986. Over 23 years, a number of previously established risk factors for coronary heart disease (CHD) incidence were found to predict mortality. The long-term follow-up assisted in illustrating temporal patterns. A single causal assessment of blood pressure retained high prediction for long-term mortality. Blood lipids, while significantly associated with both coronary and all-cause mortality, exhibited a small contribution to the latter, when compared to hypertension, cigarette smoking habits and diabetes. Weak associations of long-term coronary mortality with the dietary intake patterns of fatty acids, as reported at baseline, were probably fully mediated by the effect of the diet on serum cholesterol. Religious orthodoxy appeared to provide a degree of immunity, part of which was independent of life-style correlates. A number of now well-established associations in cardiovascular epidemiology were first demonstrated, or amplified, in the study. Patterns of ethnic diversity in the risk factor and prevalence rates of CHD persisted, as viewed from the angle of mortality rates, over nearly a quarter of a decade, highlighting the enigma of a migrant country as a cardiovascular melting pot.