The fasting concentration of cholesterol and triglycerides in serum and in very low (VLDL), low (LDL) and high (HDL) density lipoproteins (LP) was determined 3 months after a myocardial infarction (MI) in 54 men, and the values obtained were compared to those in 61 healthy male control subjects. The mean triglyceride concentration in MI patients was significantly increased in serum, VLDL, LDL and HDL by 74 %, 110 %, 30 % and 12 % respectively, compared to controls. The mean cholesterol concentration was significantly raised by 16 %, 120 % and 14 % in serum, VLDL and LDL but decreased by 22 % in HDL. Hypertriglyceridaemia occurred in 58 % of MI patients. Of these patients, two-fifths had hypertriglyceridaemia only and three-fifths had combined hyperlipidaemia. The hypertriglyceridaemia was caused by elevation of only VLDL triglycerides in 26 %, only LDL triglycerides in 19 %, VLDL and LDL triglycerides in 23 % and by various other combinations of raised LP triglyceride levels in 25 % of cases. Hypercholesterolaemia was found in 41 % of MI subjects. Of these, one-sixth had elevation of cholesterol levels, while five-sixths had combined hyperlipidaemia. The LP abnormalities underlying hypercholesterolaemia were increased of only VLDL cholesterol levels in 36 %, only LDL cholesterol in 14 % and both VLDL and LDL cholesterol in 50 % of cases. The low HDL cholesterol values in comparison to controls were related to higher VLDL triglyceride values in MI patients, since HDL cholesterol fell significantly with increasing VLDL triglyceride levels. When HDL cholesterol was related to similar VLDL triglyceride levels, there were no major differences between controls and MI. It was impossible to determine the type of hyperlipoproteinaemia solely from determination of patients' fasting cholesterol and triglycerides. All 4 lipid groups (normal lipids, elevation of cholesterol levels, elevation of triglyceride levels and elevation of both lipid levels) each comprised at least 2 different types of hyperlipoproteinaemia. Upon LP analysis, the mixed hyperlipidaemic group (elevation of both cholesterol and triglyceride levels) was, as expected, found to be most complex and was comprised of subjects with normal LP levels (11 %), and those with Type IIA (26 %), Type IIB (31 %), Type IV (20 %) and Type III hyperlipoproteinaemia (11 %). In addition, LP abnormalities (Type IIA and others) were found in MI patients with normal total serum lipids. Thus, 38 % of MI had normal lipid levels but only 17 % were normal when all LP classes were considered. The modified Fredrickson system for categorising types of hyperlipoproteinaemia did not cover all LP abnormalities encountered in the MI population. One such abnormality not defined under the present system was the elevation of LDL triglyceride levels, an important cause of hypertriglyceridaemia alone or in combination with raised triglyceride levels in other LP classes. LDL hypertriglyceridaemia occurred with or without simultaneous elevation of LDL cholesterol levels. The importance of quantitative LP analysis in the correct diagnosis of hyperlipidaemia and in the evaluation of the presence of hyperlipoproteinaemia is stressed.