84 



SCIENCE 



[N. S. Vol. XXXIX. No. 994' 



type, and it is absurd to give them generic 

 rank, and group together widely different 

 types which are alike in no single respect 

 except that they have flagella. These gen- 

 era based on motility are on a par with 

 a division of animals into those with wings 

 and those without, which would place bats 

 and birds and flying fishes and bees in one 

 group and cats and ordinary fishes and 

 worker ants in another. 



The unsatisfaetorj- nature of the Migula 

 classification, which, even if the motility 

 genera were accepted, left over one third 

 of all known bacteria in the genus Bacillus, 

 led many bacteriologists to abandon any 

 attempt at a natural classification and to 

 seek refuge in frankly arbitrary schematic 

 groupings. The logical outcome of this 

 tendency is the decimal group number 

 which our society has adopted, the history 

 of which has been so ably presented by Pro- 

 fessor Harding (1910). 



The group number, according to which 

 the characters of bacteria are indicated by 

 a conventional series of decimals, has an 

 undoubted vak\e and has proved a godsend 

 to workers who study a large series of new 

 cultures and desire a concise record of their 

 behavior. It is a sort of index to the chief 

 characters of the organisms in question, a 

 method of cataloging reactions observed. 

 It is obvious however that it is artificial, 

 and that it does not furnish a " classifica- 

 tion," an arrangement of bacteria accord- 

 ing to their natural relationships. 



There is some danger, I think, that this 

 important distinction between the group 

 number on our standard card and a real 

 biological classification may be forgotten. 

 When the student notes that 100 means 

 that endospores are produced and 200 that 

 they are not produced, he is likely to draw 

 the conscious or subconscious conclusion 

 that all bacteria producing endospores are 

 more closely related to each other, are more 



of a kind, than are the members of the two 

 separate groups. I think that this is very 

 probably a fact. Then, of the non-spore 

 formers, he notes that strict aerobes fall 

 under 210, strict anaerobes under 230 and 

 facultative foiins under 220. Again he is 

 likely to draw a similar conclusion as to rel- 

 ative relationships and again perhaps the 

 conclusion is reasonably correct. Inthethird 

 place of the whole numbers, however, any 

 such deduction as to natural relationships 

 from the group number would certainly be 

 erroneous. Of the facultative non-spore 

 formers, all which liquefy gelatine fall 

 under 221 ; and all which fail to do so un- 

 der 222. That is, the group number throws 

 together on the one hand B. cloaca; and the 

 liquefying strains of fluorescent water bac- 

 teria and the liquefying proteus forms, and 

 on the other hand B. coli and the non- 

 liquefying fluorescent and proteus types. It 

 is reasonably certain however that liquefy- 

 ing and non-liquefying fluorescent bacteria 

 are more closely related to each other, that 

 B. coli and B. cloacce are more closely re- 

 lated to each other, that liquefying and non- 

 liquefying proteus types are more closely 

 related to each other than are the liquefy- 

 ing or the non-liquefying members of the 

 three respective pairs. Precisely as in the 

 ease of Migula 's motility genera the use of 

 a single arbitrarily chosen character in 

 classification leads to misleading results. 



It is sometimes held that the difficulties 

 we experience in bacterial classification are 

 due to the fact that we must necessarily 

 rely in the main on physiological rather 

 than on morphological characteristics. I 

 do not believe this to be the case. There is 

 no fundamental distinction between mor- 

 phological and physiological properties, 

 since all are at bottom due to chemical 

 differences in germ plasm, whether they 

 happen to manifest themselves in the size 

 and arrangement of parts or in the ability 



