There is increasing evidence that the technique of reinfusing an athlete's stored blood prior to competition to improve performance has been used on many occasions. Although early experimental results were controversial and the precise mechanism by which the technique improves performance is still debated, there is now strong evidence that if the blood doping produces a sufficient rise in total red cell mass there are significant improvements in physiological variables such as maximum oxygen uptake, lactate buffering and thermoregulation. These physiological changes are matched by improvements in endurance performance. These may persist in diminishing degree for several weeks, but have to be weighed against the detraining effect produced by the repeated venesection required to obtain an adequate amount of stored blood for autologous reinfusion. Experimental evidence suggests that the transient increase in blood volume and cardiac output following reinfusion is too short lived to be of any real importance and the major effect is related to the increase in total red blood cell mass and haemoglobin enabling an increased transport of oxygen and therefore a potentially greater reserve of blood which can be diverted to non-exercising tissues to improve thermoregulation. The increased red cell mass also improves lactate buffering. Although these benefits have been shown in several studies the increases in performance and measured physiological parameters do not bear a direct relationship to the changes in haematological variables. Blood doping is of considerable importance, not only as an abuse of fair competition, but also because of the light it throws on the physiological limits to endurance performance. It has reawakened controversy as to whether oxygen transport is the limiting factor in endurance.