253 



19. The medium must contain a small amount of hae- 

 moglobin"; details are given of the form in which it must be 

 present. 



20. The cultures of the growth-promoting organism and 

 Pfeiffer's bacillus must not be brought into immediate contact. 



21. It can be assumed that the different species of bac- 

 teria form the same growth-promoting substance, but in dif- 

 ferent amounts. 



22. There. are strains which must be classed as Pfeiffer's 

 bacilli although the symbiosis phenomenon is poorly developed 

 or even completely absent. 



23. From investigations into the agglutination reac- 

 tions it may be asserted that any two strains of Pfeiffer's ba- 

 cillus chosen at random will prove in the majority of cases 

 to be fundamentally distinct when compared by simple agglu- 

 tination and absorption of agglutinins. 



Out of 6 strains of Pfeiffer's bacillus cultivated from au- 

 topsies on influenzal pneumonia cases in the same epidemic, 

 only two were identical as regards agglutination. The others 

 were distinctly different from them and from one another. 



24. No connection could be shown to exist between the 

 agglutination reaction and the classification into indol-positive 

 and indol-negative strains. 



25. The larger number of typical strains however are indol- 

 producers, and the majority of atypical strains, non-indol- 

 producers. 



26. As regards microscopic morphology there are 

 in all probability true individual differences between strains 

 of Pfeiffer's bacillus, but on account of the great variability 

 of the individual strains and the gradual transitions between 

 the different appearances observed it has been impossible to 

 determine their exact nature. 



The occurrence of round bodies has been shown to be a 

 feature common to all Pfeiffer's bacilli, which exhibit wide 

 variations with regard to size and staining reactions. 



Some of these bodies are stained black by Gram's method 

 and also with fuchsin or any simple staining method. 



27. A comprehensive survey is given of the way in which 

 the haemoglobinophilic bacteria probably ought to be classified. 

 Only three species are considered as belonging to this group, 



