The fatty acid intake is in part reflected by the fatty acid composition of adipose tissue and serum lipids. We evaluated whether the proportions of myristic (14:0), pentadecanoic (15:0) and heptadecanoic (17:0) fatty acids in the adipose tissue triacylglycerols and serum cholesterol esters and phospholipids reflect long-term dairy fat consumption in free-living men. In 114 healthy men aged 40-76 y, we compared the relative content of 14:0, 15:0 and 17:0 in subcutaneous adipose tissue and in serum lipids with relative intake (g/100 g of total fat) assessed by two 1-wk weighed food records made 6 mo apart and assessed by fourteen 24-h dietary recall interviews equally distributed during 1 y. According to food records, the mean +/- SD dairy fat intake was 24.9 +/- 13.1 g/d (29.6 +/- 10.5 g/100 g total fat); intake of 14:0, 15:0 and 17:0 was 4.6, 0.23 and 0.16 g/100 g total fat, and the content in adipose tissue was 3.6, 0.36 and 0.25 g/100 g fatty acids, respectively. Pearson correlation coefficients between intake of dairy fat (based on 24-h recalls) and fatty acid composition of adipose tissue were 0.64 (P: < 0.001) for 14:0, 0.74 (P: < 0.001) for 15:0 and 0.60 (P: < 0.001) for 15:0 + 17:0. Corresponding correlations with serum cholesterol esters were 0.34 (P: < 0.001) (14:0), 0.45 (P: < 0.001) (15:0) and 0.56 (P: < 0.001) (15:0 + 17:0), and with serum phospholipids the values were 0.30 (P: < 0.01) (14:0), 0.50 (P: < 0.001) (15:0) and 0.50 (P: < 0.001) (15:0 plus 17:0). In our study population, the relative content of 15:0 or 14:0 in adipose tissue is a valid biomarker for long-term dairy fat intake in free-living individuals. When adipose tissue is not available, 15:0 content in serum cholesterol esters or phospholipids might be used. Intake data based on repeated 24-h recalls are equally informative and may be an equivalent choice in nutritional studies.