For investigation of the role of linoleic acid (LA) and its biologically significant metabolites in mammary tumor promotion by dietary fat, a detailed study of the fatty acid group composition of serum lipids, tumor neutral lipids, tumor phospholipids, and tumor prostaglandins (PG's) was conducted in female inbred F344 rats initiated with N-nitrosomethylurea (CAS: 684-93-5) and fed diets containing various types and amounts of fat. The oils, safflower [23%, high fat (HF); 5%, low fat], corn (23%, 5%), olive (23%, 5%), and coconut (23%) varied widely with respect to their LA content and their polyunsaturate:monounsaturate:saturate ratios (9:1:1, 4.6:2.6:1, 0.6:5.9:1, and 0.008:0.05:1, respectively, for safflower, corn, olive, and coconut oils). A modified hexane-based technique was used for extraction of serum and tumor lipids. Total tumor lipids ranged from a low of 5 to a high of 228 mg/g (wet wt) with no differences found among the 7 dietary groups. The phospholipid content of the tumors ranged from 27 to 47% of total tumor lipid, again with no differences seen among dietary groups. Total serum lipids varied from a low of 3.77 mg/ml (safflower oil, 23%) to a high of 6.11 mg/ml (coconut oil, 23%), and an overall inverse trend was observed between total serum lipids and tumor incidence for the 4 HF diet groups. Serum cholesterol levels were significantly depressed in the HF safflower oil and corn oil groups compared to those in all other dietary groups and, in general, varied inversely with respect to mammary tumor incidence. Serum and tumor neutral fatty acid profiles closely reflected those of the diet, while tumor phospholipids appeared more resistant to diet-induced changes. Olive oil-fed animals exhibited high levels of oleic acid in both serum and tumor lipids. The levels of the major metabolite of LA, arachidonic acid (AA), in tumor phospholipids were highly variable and did not equate with dietary or serum LA levels. A positive association was found among dietary LA, tumor PGE2, and mammary tumor incidence among the 4 HF groups; however, no association was found between tumor AA levels and either tumor PGE2 levels or mammary tumor incidence. The results of this study suggest that dietary LA may exert its effects on mammary tumor promotion by virtue of its role as a PG precursor; but the precise steps in this sequence and possible competitive interactions between essential fatty acids, monoenes, and saturates and the PG-synthesizing system remain to be determined.