Low-fat, low-cholesterol, high-P:S-ratio diets greatly affect plasma lipid levels. There is no information as to whether such diets affect only lipoprotein levels or also levels of apoproteins and lipoprotein compositions. It is important to have information on the latter to understand diet-induced changes in the metabolism of lipoproteins. Since vegetarians regularly eat an extremely low-cholesterol, low-fat, and-high-P:S ratio diet, they represent an ideal group to study. Fifty-eight vegetarians who eat no animal products and live on a farm commune were examined. Venous bloods were drawn after 12-14 hr fasts and analyzed for lipoprotein-lipids by Lipid Research Clinic procedures and for apoA-I and apoB by radioimmunoassay. Their normal dietary intake was evaluated with 24-hr food diaries. They averaged 2200 kcal/day with 17% protein, 32% fat, and 51% carbohydrate. Negligible amounts of cholesterol (<10 mg/day) was consumed and the P:S ratio was 1.9. Average lipoprotein-cholesterol levels were depressed to about 60% of age- and sex-specific normal levels compared to a group of normolipemic free-living non-vegetarians of a like age and sex distribution. ApoA-I and apoB values were also decreased, but while LDL-cholesterol:apoB ratios did not change, HDL-cholesterol: apoA-I ratios appeared higher in the vegetarians. HDL-cholesterol:LDL-cholesterol ratios of vegetarians were also elevated. The data suggest that the vegetarian diet depressed VLDL and LDL levels without producing major changes in their compositions, whereas both the levels and compositions of HDL were altered. The changes in LDL levels, in HDL-cholesterol:LDL-cholesterol ratios, and in HDL-cholesterol:apoA-I ratios would each place vegetarians in a lower risk category for the development of clinically manifest atherosclerosis.