Abstract
Tables are shown which record the course of bacteriæmia investigated by frequent blood-cultures in a series of cases of pneumonia. The proportion of cases having a bacteriæmia in the course of the disease agrees fairly closely with other findings both in this country and in America. Attention is drawn to the frequency with which the organisms disappear from the blood in the absence of any specific treatment, and to the frequency of recovery of bacteriæmic cases. It is therefore submitted that it is unsafe to regard the disappearance of a bacteriæmia as the direct result of specific serum treatment, and that this cannot be used as evidence of successful serum therapy.