The chief dangers of blood transfusion are two-fold--that of introducing into the recipient a disease carried by the donor, and that due to the inherent properties of the donor’s blood which may interact in a serious manner with the blood of the recipient. The first of these dangers is obvious, and common sense will suggest what steps should be taken to avoid it. Danger of communicating disease is almost restricted to conditions in which an infective agent is actually circulating in some form in the blood. Inquiry will usually be enough to establish the possible presence in the prospective donor’s blood of an organism such as the malaria parasite. Nevertheless, a case has been recorded by van Dijk, in which malaria was transmitted by injecting into a patient suffering from influenza some serum obtained from another patient who was supposed to be convalescent from influenza, but had been treated for malaria a few months earlier. Another case is reported by Bernheim, who transmitted a double infection of malaria--tertian and æstivo-autumnal--by means of a blood transfusion. Blood infections, such as those due to the exanthemata, may be avoided by the precaution of never employing a blood donor who shows any signs of present illness, even though a raised temperature be the only symptom. In certain cases, when, for instance, the prospective donor may be suffering from tuberculosis in some form or from gonorrhœa, the organism is extremely unlikely to be present in the blood in numbers sufficient to communicate disease. Nevertheless, on general principles, such donors should be eliminated if circumstances permit. The most subtle form of infection, the most dangerous, and the most difficult to eliminate, is syphilis. Definite cases have been recorded in which syphilis has been communicated by blood transfusion. In one instance recorded by Sydenstricker and by Bernheim a father was infected by blood taken from his son, who had refused beforehand to allow himself to be tested. Fortunately such occurrences are rare. Still rarer and still more curious is the transmission of horse asthma recorded by Ramirez. In this instance, in which the disease is to be regarded as a form of anaphylaxis, the patient had received an amount of serum sensitive to horse protein great enough to provide him with the corresponding symptoms for some time afterwards.