Unsaturated cobalamin-binding capacity in plasma (P-UBBC) is determined by use of a silica gel (QUSO G32) to separate haptocorrin (P-ApoHC) and transcobalamin (P-ApoTC). The method is sensitive and precise: detection limit 13 pmol/L, interassay coefficients of variation 3% for P-UBBC (mean = 1080 pmol/L), 4% for P-ApoTC (mean = 700 pmol/L) (n = 30). Values for P-UBBC, P-ApoHC, P-ApoTC, and P-TBBC (P-UBBC plus P-cobalamin) determined in a population study of 228 individuals, ages 21-87 years, did not differ by sex. These values increased with age, whereas the cobalamin saturation (P-cobalamin as percentage of P-TBBC) decreased with age. However, these changes were statistically significant but marginal and thus not clinically important. We therefore suggest using combined reference intervals (central 95 percentiles) for all age groups: 500-1200 pmol/L for P-UBBC, 90-275 pmol/L for P-ApoHC, 400-930 pmol/L for P-ApoTC, 850-1600 pmol/L for P-TBBC, and 20-50% for cobalamin saturation. Results for 277 inpatients show high P-ApoHC in myeloproliferative disorders or acute nonlymphatic leukemia, whereas P-ApoTC concentrations are high in some patients with lymphoproliferative disorders or autoimmune diseases.