EVOLUTION AND CLASSIFICATION OF BACTERIA 451 



There is no reason, indeed, for making a separate family of the 

 phototrophic bacteria. As Jensen points out, most of them are 

 rods with polar flagella, a fact which suggests that they are 

 closely related to the heterotrophic Pseudomonas forms. They 

 may easily be placed in the family Pseudomonadaceae recog- 

 nized by the committee. It is true, as remarked by Buchanan 

 (1917) that the spherical N itrosococcus should not be sepa- 

 rated from the rod-shaped Nitrosomonas; but this can easily be 

 avoided without making a separate family for these organisms. 

 One of Jensen's most valuable contributions to systematic bac- 

 teriology is in pointing out that the shape of cell or form of 

 body is not a fundamental character and that there is not only a 

 possibility but even a probability that transformations of cocci 

 to rods and of rods to cocci may have taken place more than once 

 in the course of development. This conception makes it entirely 

 possible to place the autotrophic and possibly some simple hetero- 

 trophic cocci in the Pseudomonadaceae. It would merely require 

 a definition of the family so worded as not to exclude all coc- 

 cus forms. ^ ^. J. 

 The acceptance of Jensen's ideas as to the transformations of 

 cocci into rods and rods into cocci does not force us to accept 

 his theory that the most primitive forms were rods with polar 

 flagella. In the natural course of development, a non-motile 

 spherical organism would probably have preceded the motile, 

 rod-shaped Pseudomonas type. As soon as a spherical organism 

 becomes motile it tends to become elongated, a fact which 

 undoubtedly explains the rarity of motile cocci. This thought 

 gives another reason for hesitation before accepting the family 

 Nitrobacteriaceae as the most prunitive group of bacteria. 



Family 2. Mycohacieriaceae. This family is placed by the 

 committee immediately after the Nitrobacteriaceae. In this 

 they follow Jensen. Jensen, however, uses this arrangement 

 on the fallacious assumption that RhizoUum (the organism oi 

 legume nodules) has a polar flagellum, as formerly claimed by 

 Harrison and Barlow (1907). On this assumption Azotobacter 

 and RhizoUum, both with polar flagella and both able to utilize 

 atmospheric nitrogen, formed natural stepping-stones between 



