Serum ferritin concentration correlates with tissue iron stores in humans, horses, calves, dogs, and pigs but not in rats. Because serum iron and total iron-binding capacity can be affected by disorders unrelated to iron adequacy (such as hypoproteinemia, chronic infection, hemolytic anemia, hypothyroidism, and renal disease), serum ferritin is probably the most reliable indicator of total body iron stores in larger species. To test the hypothesis that serum ferritin might be correlated with tissue iron levels in cats, we developed a quantitative enzyme-linked immunosorbent assay that uses two monoclonal antibodies in a sandwich arrangement to measure feline serum ferritin. The recovery of purified ferritin added to feline sera ranged from 94% to 104%; the within-assay coefficient of variability was 8.4%, and the assay-to-assay variability was 13.2%. Mean serum ferritin from 40 apparently healthy cats was 76 ng/ml (SD = 24 ng/ml). Serum ferritin concentration was significantly correlated (P < 0.001, n = 101, r = 0.365) with the nonheme iron in the liver and spleen (expressed as milligrams of iron per kilogram of body weight), as determined by Pearson product-moment correlation analysis. Because serum iron can decrease in diseases other than iron deficiency, the combination of serum iron and serum ferritin should provide sufficient evidence to differentiate anemia of chronic inflammation from anemia of iron deficiency in the cat.