Transfusion of blood has become such a frequent therapeutic measure in the armamentarium of the pediatrician that it is now used almost as a routine measure by some clinicians. Transfusion is used in treating the moderately sick infant rather than as a final resort in an attempt to save one who is desperately ill. Certain definite precautions are, or always should be, observed to protect the recipient against accident. These precautions especially include the grouping of the bloods, and, in some instances, direct typing to prove positive compatibility of the serums and cells. Another protective measure is the use of the Wassermann or a similar laboratory test to rule out syphilis. There is never freedom from the possibility of the occurrence of an untoward accident in any technical procedure that deals with an organism as sensitive as the human being. It would be utterly impossible to rule out every known