The relationship between plasma iron, transferrin saturation, and plasma iron turnover was studied in 53 normal subjects whose transferrin saturation varied between 17% and 57%, in 25 normal subjects whose transferrin saturation was increased by iron infusion to between 67% and 100%, and in five subjects with early untreated idiopathic hemochromatosis whose transferrin saturation was continually elevated to between 61% and 86%. The plasma iron turnover of all of these subjects ranged from 0.45 to 1.22 mg/dL whole blood/d. The mean values for the above-mentioned three groups were 0.71 +/- 0.17, 1.01 +/- 0.11, and 1.01 +/- 0.13 mg/dL whole blood/d, respectively. Most of this variation, estimated at 72% by regression analysis, was due to a direct relationship between transferrin saturation and plasma iron turnover. This effect was attributed to a competitive advantage of diferric over monoferric transferrin in delivering iron to tissues. This was confirmed by the demonstration of a more rapid clearance of diferric as compared to monoferric transferrin in an additional group of eight normal subjects. Calculations were made of the amount of transferrin reacting with membrane receptors per unit time. Allowance was made for the noncellular (extravascular) exchange and for the 4.2:1 preference of diferric over monoferric transferrin demonstrated in vitro. The amount of iron-bearing transferrin leaving the plasma to bind to tissue receptors for 53 subjects with a transferrin saturation between 17% and 57% was 71 +/- 13; for 25 subjects with a saturation from 67% to 100%, 72 +/- 12; and for five subjects with early idiopathic hemochromatosis, 82 +/- 11 mumol/L whole blood/d. There were no significant differences among these groups. These studies indicate that while the number of iron atoms delivered to the tissues increases with increasing plasma iron and transferrin saturation, the number of iron-bearing transferrin molecules that leave the plasma per unit time to bind to tissue receptors is relatively constant and within the limits studied, independent of transferrin saturation.