Serum iron and ferritin concentrations were measured in 1,532 regular blood donors from South Wales who were undergoing HLA typing prior to registration on the British Bone Marrow and Platelet Donor Panel. Serum transferrin concentrations were determined for donors with serum iron concentrations > 24 mumol/l. There were 25 donors with transferrin saturations > 50% and 11 with transferrin saturations > 60%. There were five donors with serum ferritin concentrations > 200 micrograms/l (women) or > 300 micrograms/l (men). Two of the male donors had transferrin saturations > 50% and serum ferritin > 300 micrograms/l on repeat blood samples and are being treated by venesection. Donors with HLA-A3 did not differ from those without A3 in serum iron or ferritin concentrations. Even in the group of donors who were apparently homozygous for A3 there were neither abnormal serum iron nor ferritin concentrations. Although it is well established that measurements of transferrin saturation are required to detect homozygous haemochromatosis (HFE) in its earlier stages, the number of 'false-positive' results is likely to be unacceptably high for screening blood donors. Serum ferritin assays should identify donors with HFE and iron overload before the onset of liver damage. With two million regular donors and 300,000 new donors each year, a significant proportion of the U.K. population will be screened within 10 years. The assay of serum ferritin identifies donors with low levels of storage iron who are at risk of developing iron-deficiency anaemia. Furthermore, donation frequency may be increased for those donors with higher ferritin concentrations when blood supplies are low.