In a study of 95 presumably healthy, 40-42-year old males from Northen Sweden, the Lp(a) phenotype distribution differed between those who had, and those who did not have one or more close relatives (parent or sib) with coronary heart disease. In the former group, 60% of the males were Lp(a+), as opposed to 28% in the latter group. Thus, in the homogeneous population sample studied, analysis of the normal inherited Lp(a) variation permitted the identification of distinct subpopulations, with respect to familial occurrence of coronary heart disease. None of a series of other parameters distinguished such sub-populations. The results reported are in agreement with our previous finding of a close association between phenotype Lp(a+) and risk of contracting coronary heart disease.