Two random samples of women aged 55-64 living in the Vale of Glamorgan and the Rhondda Fach were followed up for 20 years. In each sample heights, weights, serum cholesterol concentrations, and specific causes of death were noted and the Quetelet index (weight/height2) calculated. The two samples were combined for analysis of total mortality. In the combined sample there was a strong positive association between Quetlet index and both total mortality and mortality from ischaemic heart disease. Multiple logistic regression analysis showed that both of these associations were significant, as was the strong positive association between Quetelet index and mortailty from ischaemic heart disease in the samples considered separately. There was no significant association between cholesterol concentration and mortality from ischaemic heart disease in each sample. There was a small negative association between cholesterol concentration and both total mortaility in the combined sample and ischaemic heart disease in the Rhondda Fach sample. These results suggest that the Quetelet index is important for prognosis, whereas the serum cholesterol concentration is not.