CLASSIFICATION 31 



side with the medically trained ' bacteriologist,' to whom the principles 

 and meaning of classification were unfamiliar, there have been investigators 

 from every other branch of biology — brewers, agriculturists, and even 

 chemists, all vying with each other in the manufacture of species. It is 

 not argued for a moment that botanists alone have the right to set up 

 a system; but it cannot be too often repeated that other investigators, if 

 their taxonomic schemes are to be of any value, must proceed on the 

 recognized and established principles of classification. As things are, there 

 is no agreement even as to the value to be attached to the few clearly 

 marked morphological features which do exist. The only point of unanimity 

 is the adoption of Cohn's groups founded on the shape of the cell (coccus, 

 rod, spiral, or filament). The importance of the distinction between fila- 

 mentous bacteria and those forms which consist of a single cell deserves 

 more emphasis than it usually receives. The filamentous bacteria, tricho- 

 bacleria, constitute an order, separate from the haplobacteria, in which the 

 colonies of cells are mere ' growth-forms ' or social aggregates of individuals, 

 such as, for example, the packets of Sarcinae, or the cobweb-like zoogloea 

 of B. vulgaris. 



The power of movement or its absence has very properly been regarded 

 as a distinctive feature, but the constancy of the ciliation (mono-, lopho-, 

 peritrichous) has been undervalued. A cholera vibrio or a bacillus pyocyaneus 

 has only one flagellum (as a rare exception two) ; the typhoid bacillus, 

 B. subtilis, and many others are always peritrichous. Lophotrichous forms 

 like the spirilla and other aquatic bacteria, or the bacillus of blue milk 

 (B. syncyaneus), always have a polar tuft of cilia, the number of which is 

 approximately known and fairly constant. That their numbers in a stained 

 preparation are often irregular is due to the delicacy of the structures, which 

 are easily thrown off, and not to any original irregularity. In the motile 

 bacteria, as in the flagellate infusoria, the number and arrangement of the 

 cilia are morphological characters of fundamental systematic value. Another 

 is the form of the sporulating rod, which, although described by some as 

 variable, possesses in reality the constancy required of systematic characters. 

 The anthrax bacillus retains its cylindrical shape, the tetanus bacillus 

 becomes invariably swollen at one end, like a drum-stick (pleclron), and some 

 of the butyric bacteria assume the shape of a spindle at the time of spore- 

 formation. Exceptions to the rule are generally deformed individuals, and 

 are very rare, not one in a hundred. A careful discrimination of these two 

 characters, ciliation and sporulation, enables us to arrange the chaotic multi- 

 tude of rod-shaped bacteria into a few well-defined genera and sub-genera. 

 It may be argued that in many cases the spores are unknown. This is true, 

 but it is no reason why we should not classify the better known forms into 

 ' good ' genera, relegating the others provisionally, according to the shape of 

 their cilia, to genera where the sporulation is unaccompanied by change 



