116 CLASSIFICATION OF FISSION-FUNGI. 



seven species, which are sharply differentiated from each 

 other. Such groups afford pleasure to the classifier. 



It is entirely different with other groups. The family 

 rosacese possesses very sharply differentiated genera, but 

 in three of these (rubus, potentilla, rosa) the multiplicity 

 of species is so great that scarcely two classifiers, in the 

 endeavor to bring order out of chaos, arrive at the same 

 classification. Essentially two exactly opposite methods 

 exist for the solution of this problem. According to the 

 first, one distinguishes every form which differs in any 

 way whatever by a name (consequently, for example, 

 every individual rose-bush !) and then arranges the count- 

 less forms thus obtained in the most natural manner pos- 

 sible. Or — and this is to-day generally preferred — a 

 number of the most striking and widely distributed forms 

 are selected as species, and the others are grouped as sub- 

 species, forms, varieties, and transitional forms of these 

 main species. 



A strict classification of bacteria appears more difficult 

 than that of any other group in the vegetable kingdom 

 for the following reasons: 



1. Bacteria, because of their minuteness and simple 

 structure, possess very few morphologic characteristics 

 suitable for classification. 



2. The description of the individual varieties of bac- 

 teria represented in the literature has been absolutely 

 insufficient ; even recently there has been much sinning 

 in this direction. 



3. There are a great many rarely described varieties 

 of bacteria, which can no longer be obtained in cul- 

 ture, with which, therefore, there is no possibility of 

 comparison with an apparently new variety. 



4. Quite a number of those describing "new" varie- 

 ties have taken no trouble to look over the contribu- 

 tions of their predecessors, but this, to be sure, is often 

 excusable because of the conditions represented under 2 

 and 3. 



Still greater difficulties in the proper definition of 

 species among bacteria lie in the extremely great vari- 

 ability of bacteria, so often referred to in the general part. 

 Cohn and Koch could easily show that Nageli, who had 



