Analysis of the Bacterial Group 

 „Pfeiffer's Bacillus". 



Introductory Remarks. 



When we place a collection of organisms in a group 

 (species, race etc.) on the basis of a series of characters, the 

 classification will be rational provided the three following con- 

 ditions are fulfilled: 



1. Constancy. Each of the characteristics as far as it 

 can be examined, must be invariably present. (By a character 

 is partly understood that feature (as regards morphology, 

 growth etc.) which is brought out by given conditions, and 

 partly — in a wider sense — the way in which the feature 

 is dependent upon the conditions. If we consider a manifesta- 

 tion of life as a mathematical function of the conditions, y --= 

 f (x l5 x 2 ., . . .), then we may use as a distinguishing character 

 either the value of y when definite values are assigned to 

 x 1} x 2 . . . , or what is more rational but also much more dif- 

 ficult, — the nature of the function f ( ) itself). 



2. Discontinuity. Each of the qualities employed to 

 characterise a group, must appear in a discontinuous manner, 

 that is to say either in the quantity which is looked upon as 

 characteristic of the group, or in quite another quantity. 



3. Correlation. There must be a connection between 

 the characters employed, so that there is a group of characters 

 which go together, thus affording the same dividing line which- 

 ever of them is used as a basis of classification. 



These requirements which have been laid down, are the 

 ideal ones, but in bacteriology it will seldom, perhaps never, 



