106 ^ CENTURY OF PROGRESS IN THE NATURAL SCIENCES 



characteristics in eacii group, naturally suggested the existence of a high degree 

 of specificity which was reflected in both physiological and morphological prop- 

 erties. Since each group contained representatives exhibiting minor differences, 

 one from the other, in shape, size, color, or physiology, it must have seemed 

 eminently rational to consider these as species and the entire group as a genus. 

 A logical consequence of this approach was Orla-Jensen's classification (1909) 

 in which the bacteria were assigned to genera that were defined by a combi- 

 nation of morphological and physiological characters. By considerably extend- 

 ing the number of differential morphological traits and incorporating the newer 

 concepts of the mechanisms of biochemical processes, derived from studies on 

 the comparative biochemistry of microorganisms, Kluyver and van Niel (1936) 

 sought to provide a more up-to-date system along the same general lines. 



Some systematists have, however, consistently condemned the use of physio- 

 logical criteria for the definition of even such small taxonomic units as genera. 

 They seem to agree with Lehmann and Neumann (1927, 2:190) who wrote: 



Dass die Systematik der Spaltpilze und der ihnen nahestehenden Mikroorganismen 

 genau so wie die aller anderen Lebewesen zunachst nacti morpliologischen Grundsatzen 

 (Form, Begeisselung, Sporenbildung) versucht werden muss, ist klar, trotz aller oben 

 angegebenen Schwierigkeiten. 



Statements to this effect can be found, for example, in Prevot's extensive paper 

 on the classification of the anaerobic cocci (1939, p. 50) : 



. . . nous pensions qu'il est possible aujourd'hui de chercher a adapter au monde bac- 

 terien les doctrines classiques qui ont reuissi pour le regne vegetal et le regne animal 

 entre les mains des freres de Jussieu, de Cuvier, de Geoffrey Saint-Hilaire, etc., et des 

 modernes: il existe une relation enti'e la valeur des characteres et le determinisme du 

 groupement des Bacteries, et cette relation est commune au trois mondes, vegetal, ani- 

 mal et bacterien: les characteres morphologiques ont la priorite sur les characteres 

 physiologiques. 



On the basis of such considerations Prevot has even developed a set of rules for 

 the delineation of taxa of higher order, as follows (ibid., p. 61) : 



Les characteres de morphologie generale sont des characteres de classe. 



Les characteres de reproduction (simple, par spore, par conidie) sont des characteres 

 d'ordre. 



Les characteres de structure cytochimique (coloration de Gram) sont des characteres 

 de famille. 



Les characteres de morphologie speciale (ectoplasme, biometvie, directions de division, 

 arrangement cellulaire) sont des characteres de genre. 



Les characteres physiologiques (culturaux, pathogenes, biochemiques) sont des charac- 

 teres d'espece. 



Les characteres physiologiques secondaires et serologiques (agglutination) sont des 

 characteres de variete ou race. 



From a scientific viewpoint it is, however, astonishing that the validity 

 of such verdicts generally seems to have been taken for granted; rarely, if ever, 

 has an attempt been made to .justify the belief that for the purpose of classifi- 

 cation of the bacteria morphological characters are more significant than physi- 

 ological or biochemical properties. Occasionally it is possible to infer from the 

 context the reasons for this notion. The reference to Jussieu, Cuvier, and Saint- 

 Hilaire in the above quotation from Prevost, for example, indicates the trend of 

 thought. And Kluyver and van Niel ( 1936, p. 370) expressed this still more directly: 



