One approach to the development of an effective red cell substitute has been chemical modification of human hemoglobin to optimize oxygen transport and plasma half-life. Human hemoglobin A₀ and two of these modified hemoglobins, one prepared from the cross-linking of the α-chains at lysine residue 99 by bis(3,5-dibromosalicyl)fumarate (Hb-DBBF) and the other by acylation of lysine residue 82 of the β-chain by mono-(3,5-dibromosalicyl)fumarate (Hb-FMDA), were tested by HPLC for their susceptibility to oxidative damage caused by H₂O₂. Such oxidative insult may occur during ischemia and reperfusion of tissues after transfusion of red cell substitutes to patients with hypovolemic shock and trauma. Hb-DBBF was extremely susceptible to damage of its heme and protein moieties with stoichiometric amounts of H₂O₂, whereas Hb-FMDA was highly resistant, even at 10-fold molar excess and at an acidic pH of 4.7. Hemoglobin A₀ was of intermediate susceptibility, exhibiting alteration of heme and protein moieties at acidic but not neutral pH. Since the degradation of heme can release the potentially toxic agent iron, Hb-FMDA may be a more promising candidate than Hb-DBBF for development as a red cell substitute. A similar approach may be used to assess the susceptibility of other hemoglobin-based red cell substitutes to oxidative damage in order to determine the molecular basis of heme and protein alteration.