A method combining data on fatty acid composition into subsets is used to illustrate general relative competitive selectivities in the metabolic and transport events that maintain fatty acid compositions in tissue lipids and to minimize differences among tissues or species in the amount of individual fatty acids. Fatty acid compositions of triglycerides and phospholipids in several tissues of the rat were maintained with simple relationships between the exogenous n-3 and n-6 dietary polyunsaturated fatty acids and the endogenous n-7 and n-9 types of fatty acid. The general pattern of fatty acids in triglycerides was similar for liver, plasma and adipose tissue, averaging about 30% as saturated acids, 67% as 16- and 18-carbon unsaturated acids and only about 2% as 20- and 22-carbon highly unsaturated acids. The tissues maintained a linear relationship between the amount of 18-carbon polyunsaturated fatty acids in the diet and in the tissue triglycerides, with the proportionality constant for 18:3n-3 being 60% of that for 18:2n-6. The total phospholipids of liver, plasma and red blood cells maintained about 45% of the fatty acids in the form of saturated fatty acids and 20-30% as 20- and 22-carbon highly unsaturated fatty acids irrespective of very different proportions of n-3, n-6 and n-9 types of fatty acids. In all three tissues, the 20-carbon highly unsaturated fatty acids of the n-3, n-6 and n-9 type were maintained in a competitive hyperbolic relationship with apparent EC50 values for dietary 18:2n-6 and 18:3n-3 near 0.1% of dietary calories. The consistent quantitative relationships described in this study illustrate an underlying principle of competition among fatty acids for a limited number of esterification sites. This approach may be useful in predicting the influence of diet upon tissue levels of the substrates and antagonists of eicosanoid biosynthesis.