Development of a viable blood substitute began by focusing on recreating the oxygen-carrying capacity of blood, leading to the recognition that haemoglobin (Hb) is presently unequalled for this function. However, as human Hb is the only realistic source of this protein, the production of a blood substitute that solves transfusional blood availability problems and shortages must introduce a multiplying factor between supply of natural blood and blood substitute, while maintaining equivalency of function/efficacy. In other words, a unit of blood should produce several units of equivalent blood substitute. This expansion is now possible because of new understanding of how blood delivers oxygen in the microcirculation and the consequences of reducing oxygen-carrying capacity in haemorrhage. This information is used to provide improved resuscitation capacity and maintenance of tissue metabolism by tailoring the properties of a blood substitute to the task of maintaining microvascular function, rather than oxygen delivery capacity. Resuscitation in an organism that is haemorrhaging requires maintenance perfusion, a process directly linked to the maintenance of adequate levels of shear stress on the endothelium, induced by either increased blood/plasma viscosity or increased blood flow velocity in the microcirculation. This process must also be intimately coupled with the requirement that no portion of the tissue is anoxic. This disparate set of requirements can be satisfied with high viscosity Hb solutions that have high affinity for oxygen, a combination of properties that causes the microcirculation to remain functional, and a requirement that supersedes restoration of oxygen-carrying capacity in the treatment of haemorrhage.