SELECTED PAPERS 



should be rated. On the one hand we may refer to Orla-Jensen's [1909] 

 classification which is an extreme example of the concept that phys- 

 iological characters should determine the main lines of the system. In 

 contrast herewith Prevot [1933] recently formulates one of his laws 

 of bacteriological systematics as follows: 'les caracteres physiologiques 

 sont des caracteres specifiques', thus forbidding the use of physiological 

 characters for the demarcation of units of higher systematic rank than 

 the species. However, in another passage of his treatise Prevot obvious- 

 ly takes exception to his law by allowing for the creation of genera on 

 a physiological basis in such cases where otherwise the morphological 

 genera would become overburdened. 



Personally we are of the opinion that Prevot rightly emphasizes the 

 priority of morphological over physiological characters. Yet his restric- 

 tion that the application of the latter should be confined to delimita- 

 tion of species is of a quite arbitrary nature. In accepting the taxon- 

 omic value of physiological characters it cannot be understood why 

 they could not also be applied for the demarcation of higher system- 

 atic units. 



In support of this view we wish to observe that such a procedure is 

 not at all limited to bacteriological classification. A typical example is 

 offered by algologists who do not hesitate to support the separation of 

 the class of ' Heterocontae from that of the Isocontae by such typical physiol- 

 ogical features as the nature of the storage products, the proportion 

 of the pigments present, and the chemical constituents of the cell wall 

 [West and Fritsch, 1927]. 



There is, moreover, a second reason why it is only rational to use 

 physiological characters not exclusively for differentiation of species 

 but also for uniting them into higher groups. This reason is to be 

 found in the fact, that the mere act of cultivating a bacterium, of thus 

 gradually familiarizing oneself with the various aspects of the organism, 

 leads to a more or less unconscious inclusion of the physiological char- 

 acteristics in fixing its systematic position. In other words, no bacteriol- 

 ogist can be satisfied with a classification which combines in one genus 

 organisms which behaved quite dissimilarly in his preliminary work. 

 To cite just one example: the microscopical appearance of some of the 

 aerobic sporeforming bacteria and that of certain types of butyric acid 

 bacteria may show a striking resemblance, yet the technique in hand- 

 ling and the nutritional requirements of both groups are so obviously 



284 



