208 



often tenacious. Not uncommonly the appearance and con- 

 sistence of the cultures however could best be described as 

 typical. On blood agar tlie growth was always considerably 

 heavier than in the case of Pfeiffer's bacillus and of a more 

 whitish appearance by reflected light. 



The appearance of isolated colonies differed a good deal. 

 In some cases they could not be distinguished from Pfeiffer's 

 bacillus, but they were more often darker, flatter and drier. 

 Granular formation occurred chiefly in those colonies which 

 most resembled Pfeiffer's bacillus colonies. 



The microscopic morphology of all the strains, a number 

 of which were examined several times, must be described as 

 distinctly „atypical", that is to say that coarse forms and 

 threads were predominant. In most cases the elements were 

 coarser than we occasionally observe in Pfeiffer's bacillus (see 

 plate 3). In other cases the microscopic picture however was 

 indistinguishable from that of microscopically atypical Pfeif- 

 fer's bacilli. The spherical bodies were often well developed 

 but scarcely differed from the corresponding bodies in Pfeif- 

 fer's bacillus. 



A few strains were examined for anaerobic growth, gelatine 

 and serum liquefaction, glucose fermentation, minimum tem- 

 perature for growth, thermal death-point, and resistance against 

 drying, but they did not differ from Pfeiffer's bacillus in any 

 of these respects. 



18 strains were examined for indol formation, with 4 nega- 

 tive results. Agglutination tests (1:50 and 1:100) with 3 of 

 the haemolylic strains and 14 Pfeiffer's bacillus sera were all 

 completely negative. Some other strains agglutinated spon- 

 taneously. 



We must therefore regard the haemolytic haemoglobinophilic 

 bacteria as a group which is closely allied to Pfeiffer's ba- 

 cilli, but can be easily distinguished from them, first of all 

 by the haemolysis itself, which is a very appropriate character 

 common to the group, because (1) it is constant; (2) it is either 

 absent, or present to a marked degree; (3) practically all the 

 strains examined haemolyse in exactly the same manner. Fur- 

 ther the group as a whole, though not always the individual 

 strain, is characterised by a series of other common qualities 

 which distinguish it from Pfeiffer's bacillus (occurrence, rela- 



