362 PLANT SOCIOLOGY 



attempt to arrange the fundamental units of vegetation, the asso- 

 ciations, according to affinities based on their intrinsic characteristics 

 or to place them in a logically incontestable series of group concepts. 

 They aim at a system "which can be used without extensive study" 

 (Brockmann-Jerosch and Riibel, 1912, p. 13). Obviously, such a 

 system is possible only by making many concessions, and everyone 

 must admit that his own arrangement is just as far from being "natural" 

 as any other, in that it cannot give a clear picture of the situation to 

 the student (Warming-Graebner, 1918, p. 341). 



Some of the most ardent supporters of the physiognomic-ecologic 

 method are beginning to doubt its adequacy. "It seems practically 

 hopeless," writes Du Rietz (1924, p. 130), "to get plant sociologists 

 to agree on a unified system of formations. And one must admit that 

 in this grouping into formations, associations that from every other 

 consideration are very closely related must often be classed in different 

 formations. " 



To make things worse, the Brussels Congress (1910) gave this 

 archaic classification its official benediction. 



Meanwhile, influential voices have spoken against the Brussels 

 pronouncement. "Would we," remarked Pavillard (1912, p. 13), 

 "place white sheep and white rabbits in one genus just because they are 

 white? The preliminary arrangement of associations according to 

 their appearance is an expression of laziness, without philosophic 

 grounds. Nothing could be more illogical than to create in this way 

 ecologic genera whose species are the floristically defined associations. " 



Floristic Systems. — The most recent grouping of plant communities 

 according to similarity in floristic composition proceeds from con- 

 firmed observation that every species — indeed, every race — has a 

 definite, greater or lesser, indicator value. The species are used as 

 signs of certain synecologic, syngenetic, and synchorologic relations. 

 Similarly, therefore, the combination of species of two communities 

 shows similarity in their life conditions in the widest sense. Since 

 we combine floristically related communities into higher units, we are 

 uniting, by floristic characteristics, units which should be united also 

 on ecologic and historic grounds. 



However, floristic criteria cannot be applied mechanically. Floris- 

 tic coefficients of similarity, however carefully obtained, are not 

 adequate if used alone. 



Coefficients of Community. — Jaccard (1912) used the term coeffi- 

 cient of community for the similarity of the species lists of two regions, 

 expressed in percentage. The concept is equally applicable for the 

 comparison of two communities. 



