10 059 civil servants and municipal employees, aged 40--65, were examined in the Israeli Ischaemic Heart Disease Study. 475 (4.72%) died during a 7-year follow-up. High density lipoprotein (HDL) cholesterol levels were determined in the last 6562 men entering the study. 305 (4.65%) of these men died. Coronary heart disease (CHD) accounted for 37% of the total deaths. A "J" shaped relation which persisted after removing data on early mortality (first 2 years) was observed between total cholesterol and total mortality. An inverse relation was observed between HDL cholesterol and total mortality. Multivariate analysis of the data to adjust for possible confounding effects of additional mortality risk factors demonstrated that total cholesterol made no independent contribution to total mortality, but that the contribution of low HDL to mortality persisted after adjustment. CHD mortality consistently increased with rising concentrations of total cholesterol. CHD mortality rates decreased markedly with increasing HDL cholesterol concentrations. After adjustment for age and other risk factors the relation of coronary mortality to HDL cholesterol emerged as the dominant one. There was no clear-cut association between total of HDL cholesterol and cancer mortality. These results indicate that, particularly in older age-groups, measures designed to increase HDL cholesterol may prove as valuable in preventing CHD as those designed to reduce low density lipoprotein cholesterol.