THE PRINCIPLES OF CLASSIFICATION 69 



should necessarily spell exclusion from any other group, and 

 this is not the case in the American classifications. 



In the lengthy and elaborate schemes that have appeared 

 from America there has been no attempt to question the 

 validity of the characteristics used for the groupings. All the 

 attributes used in earlier groupings have been taken at their 

 face value. Those parts of Buchanan's scheme that deal 

 with the sulphur bacteria will be here noted. 



Review of the Aitributes used in the Subdivision of 

 THE Sulphur Bacteria. 



The chief difficulty in grouping these forms arises from the 

 fact that hitherto so little has been done to distinguish transient 

 from permanent characters. If the former arc used for the 

 diagnosis of bacteria, confusion inevitably follows the effort 

 to establish the genus and species of any particular unknown 

 organism on the basis of its morphological distinctions. The 

 range of constant attributes among bacteria is very limited. 

 The author has recorded (Ellis (i — 3)) the results of the ex- 

 amination of the more important morphological characters 

 generally used in the classification of bacteria. These are 

 given below : — 



(i) Division in one, two, or three planes. — It was found 

 experimentally that a Sarcina which normally divides in three 

 dimensions of space, could be made to break up into smaller 

 units of one, two, three, or four cells. An observer remarking 

 the cells in this last condition, would not be able to allocate 

 them to the genus Sarcina, for they would not show the 

 characteristic cell-arrangement in three dimensions of space. 



In the same way species of the genus Micrococcus, which 

 is distinguished by the divisions of its cells in two dimensions, 

 were induced to break up into smaller groups varying from 

 one to four or a small number of cells. 



On the other hand, the Streptococci, which divide in only 

 one dimension, do not, in the author's experience, depart from 

 that method of division. Hence it follows that an organism 

 showing globular cells in a culture may be a Sarcina, or Micro- 

 coccus, or a Streptococcus. It is possible only by an extended 



