132 ADAPTATIONS. OECOLOGICAL CLASSIFICATION SECT. HI 



the soil are of supreme importance in determining the production of 

 formations, and they must therefore be the foundation of oecological 

 classification. Clements, 1 with reason, has objected to Schimper's 

 scheme of distinguishing between climatic and edaphic formations, if 

 indeed it was Schimper's meaning that a sharp distinction is throughout 

 possible, and that both groups of factors are of equal potency. 



The importance of soil in determining the development of definite 

 plant-communities is clearly revealed in the topographical distribution 

 of these ; there is not a single community of land-plants that extends 

 without interruption over great stretches of land ; all are discontinuous 

 and, according to the nature of the soil, interrupted by other communities, 

 however uniform the climate may be. On the other hand, it is often 

 the case that one and the same formation, either of water-plants or of 

 land-plants, is developed in very different climates. 



The differences existing in the climate of various parts of such a 

 country as Denmark are quite inadequate to account for the great differ- 

 ences in vegetation. A prominent part is played by chemical differences in 

 soils (amount of common salt), also by their fertility or amount of nutrient 

 salts ; for instance, soil poor in food-material favours the preponderance 

 of heath. Grabner 2 regards the percentage of nutritive salts dissolved 

 in the soil-water as the factor controlling the character of vegetation ; 

 he therefore divides formations into three great groups according as the 

 water is rich or poor in mineral matter, or contains common salt : 



1. Formation, where the water is rich in mineral salts ; 



2. Formation, where the water is poor in mineral salts ; 



3. Formation, where the water is saline. 



The same opinion is expressed by A. Nilsson 3 , but this view seems to 

 be based chiefly upon observations in the open air, and too little tested 

 by soil-analyses. Grabner apparently exaggerates the importance of the 

 factor in question, for his scheme would, on the one hand, unnaturally 

 separate formations, such as those of sour-meadow, heath-bog, and ling- 

 heath, which belong together, while, on the other hand, it would un- 

 naturally group together formations, such as those of sand-field and peat 

 vegetation, which are not allied. It would appear that the most potent 

 and decisive factor is the amount of water in soil ; and this, in turn, 

 depends upon the depth of the water-table and upon the physical 

 characters of soil (its water-capacity, the amount of air-content, the 

 plants and animals living in it, the production of humus, and the like). 



We must, however, admit that climate may favour a certain formation 

 by causing this to become less exacting as regards its edaphic require- 

 ments, and consequently enabling it to be distributed over a large area on 

 very diverse soils ; whereas in another climate this formation will be 

 vanquished by communities better adapted to localities where certain 

 special soils occur, and will be more broken up and restricted in its dis- 



1 Clements, 1899, 1904. Clements (p. 27) writes : ' From the above it follows 

 that Schimper's so-called climatic formations, forest, grassland, and desert, are 

 merely a somewhat incomplete expression of water-content association. As to 

 the validity of his division of all formations into climatic and edaphic, there is also 

 room for grave doubt ... all plants . . . are primarily influenced by soil, i.e. 

 they are edaphic.' 



1 Grabner, 1898, 1909. * A. Nilsson, 19026. 



