The clinical course of the present epidemic of influenza has emphasized what appears to be two important points worthy of investigation:
The rapidity of onset, suggestive of an overwhelming intoxication, which may be due to a soluble toxin.
A marked leukopenia present in most of the cases.
The question of the leukopenia has not, in our opinion, received the attention it deserves, and a succeeding report will deal with that question. The present report deals with the filtrable poison. Although these experiments do not by any means prove that the influenza epidemic is caused by the Pfeiffer bacillus, yet they would seem to indicate this possibility.
Five strains of the influenza bacillus were used in this work. For two of these we are indebted to Dr. Anna Williams; the other three were isolated by us from influenzal pneumonia sputums.
The poison was formed by the growth of B. influenzae on