Hemostasis is the oldest problem of operative surgery, and is still one of the most interesting and important. Our conception of the amount of blood lost in the course of a given operation is, in general, vague. It is founded on such indications as the color of the patient, the pulse rate and the number of sponges used. Likewise, our idea concerning the amount of blood that has been lost when alarming symptoms occur is based on speculation. Our opinions in regard to this phase of hemorrhage are influenced largely by the behavior of the donor after transfusion and the results of animal experiments. At the Robert W. Long Hospital, from 700 to 750 c.c. of blood has been taken from donors many times without the least apparent harm resulting. In fact, donors have usually walked away from the hospital soon after the transfusion. However, when as much as 900