For many years blood transfusion has been a favorite method of treatment in severe complications following otitic and mastoid infections and has likewise been used by physicians generally in a variety of other diseases. In a small percentage of cases the benefit derived from the procedure has seemed to be out of all proportion to the replacement value of the blood elements introduced and has, I believe, been generally attributed to the presence of specific immune bodies in the donor's blood.
With this idea in view I have endeavored in the past to secure blood for transfusion from donors in whom immune bodies specific for the patient's disease might be assumed to be present. This assumption could reasonably be made when it was known that the donor had recently suffered from the same or from a similar disease caused by an identical organism, so far as could be determined, and