The diagnosis of iron deficiency rests upon the correct evaluation of body iron stores. Morphological interpretation of blood film and the red cell indices are not reliable and often absent in mild iron deficiency. Serum iron levels and iron-binding capacity are more sensitive indices of iron deficiency, but they are often normal in iron depletion and mild iron deficiency anemia. They are also subject ro many variables which may introduce substantial errors and influenced by many pathologic and physiologic states. Examination of the bone marrow aspirate for stainable iron has been regarded as one of the most sensitive and reliable diagnostic method for detecting iron deficiency, but this also has limitations. Thus, there is still need for a more practical, but sensitive and reliable substitute as a screening test of iron deficiency. Pollack et al. (1965) observed that the intestinal absorption of cobalt was raised in iron, deficient rats and Valberg et al. (1969) found that cobalt absorption was elevated in patients with iron deficiency. A direct correlation was demonstrated between the amounts of radioiron and radiocobalt absorbed. Unlike iron, excess cobalt was excreted by the kidney, the percentage of radioactivity in the urine being directly related to the percentage absorbed from the gastro-intestinal tract. Recently a test based on the urinary excretion of an oral dose of 57Co has been proposed as a method for detecting iron deficiency. To assess the diagnostic value of urinary cobalt excretion test cobaltous chloride labelled with 1 μCi of 58Co was given by mouth and the percentage of the test dose excreted in the urine was measured by a gamma counter. The mean 24 hour urinary cobalt excretion in control subjects with normal iron stores was 6.1%(1.9∼15.2%). Cobalt excretion was markedly increased in patients with iron deficiency and excreted more than 29% of the dose. In contrast, patients with anemia due to causes other than iron deficiency excreted less than 27%. Hence, 24 hour urinary cobalt excretion of 27% or less in a patient with anemia suggets that the primary cause of the anemia is not iron deficiency. A value greater than 27% in an anemic subject suggests that the anemia is caused by iron deficiency. The cobalt excretion test is a simple, sensitive and accurate method for the assessment of body iron stores. It may be particularly valuable in the epidemiological studies of iron deficiency and repeated evaluations of the body iron stores.