SELECTED PAPERS 



furthermore to the great importance attached to the character of 

 pigment production. Whilst in genera like Spirillum, Propionibacterium, 

 Mycobacterium the production of yellow, orange, brown and red pig- 

 ments is not considered to be of sufficient importance for making 

 generic distinctions, here a special tribe is created for uniting the 

 pigmented forms which are then divided into four genera on the basis 

 of the colour. The occurrence of yellow pigments which in the genus 

 Micrococcus is deemed inessential is here raised to a character of generic 

 rank (Flavobacterium) . Not all bacteriologists will realize, moreover, 

 that for the correct determination of the common fluorescents he must 

 reject the designation of the pigment as yellow but characterize it as 

 green or blue-green. If he succeeds in avoiding this pitfall he will still 

 be lost in case he is dealing with one of the numerous representatives 

 of the genus Phytomonas which also produce a fluorescent pigment. 

 For it is impossible to arrive at the last mentioned genus without this 

 time having designated the same colour as pale yellow! 



The arbitrariness may have been sufficiently demonstrated by the 

 foregoing examples. Yet it is worth-while to note two more instances 

 which at the same time clearly show the lack of homology in the vari- 

 ous systematic units. In the tribe Erwineae the two genera are distin- 

 guished by the mode of insertion of the flagella; on the contrary the 

 genera Flavobacterium, Chromobacterium and Achromobacter contain or- 

 ganisms with polar as well as such with peritrichous flagella. The 

 second instance is the arbitrary use of the character of the Gram-stain. 

 It has been mentioned above that the main subdivision of the sapro- 

 phytes is based on this property; on the other hand in the group of 

 animal parasites we encounter Gram-positive and Gram-negative 

 species within the same genus (cf. e.g. Alcaligenes, Bacteroides). 



We shall not continue the detailed criticism of Bergey's classification 

 of the Eubacteriales but only make a few remarks regarding the re- 

 maining orders, documenting our opinion that in these orders the same 

 inconsistency, lack of homology etc. are to be found. The second order, 

 Actinomycetales, owes its origin to the just evaluation by Lehmann and 

 Neumann of the natural relationship between the Actinomycetes and the 

 bacteria of the genera Mycobacterium and Cory neb acterium. The com- 

 mon characteristics of these groups are morphological and comprise 

 positive Gram-stain, the lack of ability to form true endospores and 

 permanent immobility. These properties are indeed included in the 



296 



