Methods for measuring serum apolipoproteins, apo A-I and apo B, have been tested and used for measuring the apolipoproteins in 230 healthy individuals, volunteers from various working places and one old peoples home, 118 men and 112 women, aged 18 to 85, from the Reykjavik area. Cholesterol, triglycerides, HDL-cholesterol and LDL-cholesterol were also measured and relation of levels studied. The methods turned out to be precise, 4.98% for apo A-I and 1.10 for apo B, and results compared favourably with assigned values of a control serum subjected to replicate measurements by the methods (Table I). Serum apo A-I and HDL-cholesterol were significantly higher in women than in men. Apo A-I correlated highly significantly (p<0.0001) with HDL-C, a little less with cholesterol in women and significantly with age in men but not in women. Apo B had a highly significant correlation with age, cholesterol, triglycerides and LDL-C, but none with apo A-I and HDL-C. The present study is the second one published on the serum levels of apo A-I and apo B in Icelanders. Apart from few inconsistencies the two studies compare well in general. The other, very recent, study was done through the Heart Association (Hjartavernd) and they used random sampling for their group selection whereas we used a non biased sampling method based on practicability. Their mean apo A-I levels were lower than ours in men and women (10%) and their mean apo B levels higher in men (25%) and women (20%). HDL-cholesterol, but not cholesterol and LDL cholesterol, levels were also found to be higher in our group. These differences in serum concentrations are probably caused by different methods and/or standards used by the two groups of investigators. When compared to mean serum values for apo A-I and apo B published during the last 16 years by various investigators from different countries, the present serum values as well as the apo A-I/ apo B ratio, seem to be among the higher ones (Table IV).