The transferrin receptor in serum provides a useful measure of tissue iron deficiency and the rate of erythropoiesis, whereas serum ferritin reflects the amount of storage iron in normal subjects. We studied the serum transferrin receptor and the transferrin receptor-ferritin ratio in 57 healthy prepubertal or early pubertal boys and followed them at 3-mo intervals for 24 mo to evaluate their iron status during puberty. The mean laboratory parameters changed as follows: Hb from 13.0 to 13.3 g/dL (p = 0.01), mean corpuscular volume from 85 to 84 fL (p = 0.0001), transferrin receptor from 6900 to 7200 micrograms/L (p = 0.03) ferritin from 36 to 23 micrograms/L (p = 0.0001), and transferrin receptor-ferritin ratio from 230 to 400 (p = 0.0001). At the start of the investigation, the serum transferrin receptor was elevated (> 9000 micrograms/l) or ferritin low (< = or 12 micrograms/L) in fewer than 2% of the boys. During the subsequent 2 y the proportion of boys with an elevated transferrin receptor or low ferritin value increased. The two parameters were simultaneously abnormal in none of the boys initially, but in about 3% of the boys 2 y later. The change in transferrin receptor-ferritin ratio was closely correlated with genital development. The proportion of elevated transferrin receptor-ferritin ratios increased 4.5-fold during the 2 y, indicating the high responsiveness of the ratio. At the end of the study, iron therapy was started to eliminate any iron deficiency. In response to the therapy, the mean transferrin receptor-ferritin ratio fell to 210 +/- 19, i.e. close to the level at the beginning of the study. The marked responses of the transferrin receptor and the receptor-ferritin ratio to iron therapy reflect the dependence of these parameters on iron status rather than on physiologic differences in the rate of erythropoiesis.