The efficiency of blood transfusion in the treatment of hæmophilia has been very often demonstrated, and seems at present to afford the most certain means that we possess of arresting the symptoms. Presumably the transfused blood supplies directly the deficient factor in the coagulation complex, and it has been shown by Bernheim (1917) that the transfusion even of quite a small amount of blood will almost immediately stop the bleeding. In addition to comparative certainty and rapidity in action, transfusion has the advantage that it will replace the blood which has been lost, for often the patient has reached a stage at which he is in danger of his life from actual anæmia. This treatment, therefore, will always be useful in an emergency, whether the patient be bleeding to death from a slight wound, or whether he be suffering from acute appendicitis and so is in need of an immediate operation. If transfusion does not at once stop the bleeding, the treatment can be repeated, so that the patient should not be allowed to die from loss of blood. In most cases the bleeding will eventually stop if the patient’s life can be prolonged. Even if the treatment be immediately successful, the transfused blood necessarily contains only a limited quantity of the substance necessary for the coagulation complex, and this gradually disappears. Again, therefore, the effect is transitory, so that transfusion is in no sense curative. It has been noticed that the tendency of a hæmophilic to bleed decreases as age advances, and it has been suggested by Ottenberg and Libmann that small quantities of blood should be injected into his veins at regular intervals of one to three months. It is possible that in this way he might be brought safely through the more perilous years of his life. The proof of the effect of transfusion upon the coagulation time of the blood rests upon the evidence of a number of independent observers. Pemberton has recorded a case of a hæmophilic whose coagulation time before transfusion was estimated to be 23 minutes. Blood was given to the amount of 500 cc., and 5 minutes later the coagulation time was 3 minutes. Twelve hours later it was 8 minutes, and on the fourth day after transfusion it had risen again to 20 minutes. Other observations have been made as follows: