61 



number of healthy individuals as well. Perhaps the microbe 

 eventually acquires so great an increase in its vitality that it 

 thrives not only as a saprophyte in healthy persons but also 

 may be able by itself to produce infections resembling in- 

 fluenza. When the influenza dies out Pfeiffer's bacillus may exist 

 in normal people for some months still, but as it is incapable 

 of living for a prolonged period, as a. pure saprophyte it will, 

 as before the influenza appeared, gradually be limited to oc- 

 curring in the endemic diseases. 



The reason that in some cases Pfeiffer's bacillus is found 

 to have a great distribution among influenza patients imme- 

 diately after the beginning of the epidemic in that particular 

 locality, may partly be due to wide distribution among the 

 patients from whom the infection came, and partly to its 

 being widespread in the same place before influenza appeared 

 there. The converse will be observed when a locality in which 

 Pfeiffer's bacillus is rare becomes infected with influenza not 

 complicated with this microbe. 



If one of the smaller „influenza'" epidemics develops in a 

 place where Pfeiffer's bacillus is not found, or only rarely, 

 it may happen that the epidemic runs its whole course without 

 Pfeiffer's bacillus being demonstrated in the patients at all. 



We can easily imagine that with regard to the distribu- 

 tion of Pfeiffer's bacillus in different ^nfluenza^-epidemics there 

 may exist all possible transitions from complete absence to con- 

 stant occurrence, occasioned partly by the original distribution 

 of the microbe in the place in question and partly by the 

 interplay between this and other competing or amicable or- 

 ganisms. \ 



As far as one can see, these views are in good agreement 

 with the actual facts, that is to say that on the basis of what 

 is already known about the occurrence of Pfeiffer's ba- 

 cillus it is reasonable to conclude that it is not the primary 

 influenza microbe. But this opinion forces itself upon us to 

 a still greater extent from our knowledge of the manifold dif- 

 ferences between individual strains of Pfeiffer's bacillus. In this 

 connection it will suffice to reckon with serological differences. 

 It has already been intimated that the serological reactions of a 

 bacterium- in as far as these can most simply and clearly 

 be demonstrated by agglutination and agglutinin absorption, 



