If the transfusion is being done at leisure, the donor’s blood must be tested for a positive Wassermann reaction. Even this test, however, has been known to fail, and since, in an emergency, the most careful inquiry, aided by a desire on the part of the donor to arrive at the truth, may reach an erroneous conclusion, the risk of infection with syphilis can never be completely eliminated. Since reasonable care can make the danger a remote one, it need not hinder the performance of a transfusion any more than an occasional death under anæsthesia prevents the frequent use of general anæsthetics. The mere existence of such a danger is, however, an argument in favour of the general use of the “professional blood donor,” whose Wassermann reaction, personal history, and mode of life are well known to the practitioner; the previous use of his blood on perhaps more than one occasion, if unattended by any ill results, will give an added confidence. The tragedy of such a misfortune is so great that no precaution which can possibly be taken should be regarded as absurd. The second danger present in the inherent qualities of the donor’s blood has been already alluded to in the historical sketch of the subject. Before the existence of the “blood groups” was realized, a number of fatalities due to an unexplained cause had occurred. Even after the existence of the groups had been demonstrated, the warning that resulted was apt to be disregarded, and it was not until still further fatalities due to this incompatibility of bloods had taken place that the very important nature of the discovery came to be understood. The chances are, on the whole, that the blood of any donor chosen at random will not prove fatal to a given recipient; nevertheless, it must frequently happen that the transfusion without being fatal will be wasted, or to some degree detrimental. It is therefore evident that the existence of blood groups must be seriously regarded, and it is necessary to enter into a detailed consideration of their relations to one another and the symptoms which they may produce. In the next chapters will be found a further description of their physiology and pathology and of the methods of testing for them.