68 



After investigation of the relations of the most important 

 representative of the haemoglobinophilic bacteria, three of tne 

 best known of the organisms which generally are classed in 

 the same category are examined in a similar manner. One of 

 the results attained by this procedure is the justification of 

 our continuing to class all the organisms going under the 

 name Pfeiffer's bacillus, as one species, as it is shown that 

 the differences between this and other species are distincL 

 and regular (pp. 205—224). 



Finally, the last section is devoted partly to giving an 

 account of the procedure adopted in this investigation and 

 partly to discussing the different media which are adapted to 

 the cultivation of Pfeiffer's bacillus, a term which is used as 

 equivalent to haemoglobinophilic bacteria in the strictest sense 

 (pp. 225-247). 



The characters of Pfeiffer's bacillus which are employed 

 in this investigation are the following: 



1. The morphology and behaviour to staining 

 by Gram's method. Pfeiffer's bacillus is a Gram-negative 

 rod. The first half of this term needs no further elucidation. 

 But we must without delay attempt to define what is meant by 

 the word „rod u . It happens in fact, that by no means every 

 single Pfeiffer's bacillus is distinctly rod-shaped. On the con- 

 trary it is very common to* find that the majority of indivi- 

 dual bacilli in a preparation show so little difference between 

 their length and breadth, that they might well be called cocci. 

 The reasons Pfeiffer's bacillus must however be classed with 

 the rod-shaped organisms are explained in the following: 



(a). In every culture of Pfeiffer's bacillus the majority of 

 the individual bacilli will show a perceptible difference between 

 length and breadth; and when two elements are arranged as a 

 pair or constitute the two halves of a bacillus which is divi- 

 ding, the longest dimension, at least for some of the organisms, 

 is at right angles to a line separating the two elements. 



The latter in particular always permits of a sure diffe- 

 rentiation from GramMnegative cocci, which are either round 

 or oval with the longest diameter parallel to the line of demar- 



