74 INVASION 



importance of the two methods can hardly be conjectured as 

 yet, but origin by monogenesis would seem to be the rule. 

 The arguments adduced by Engler in support of poly- 

 genesis are in themselves conclusive, but the investigations 

 of the past decade have brought to light additional proofs, 

 especially from the experimental side. In determining the 

 physical factors of prairie and mountain formations, and 

 especially by methods of experimental ecology, the author 

 has found that habitats are much less complex than they are 

 ordinarily thought to be, since water-content and humidity, 

 and to a less degree light, constitute the only factors which 

 produce direct modification. In addition, it has been ascer- 

 tained that the minimum difference of water-content, hu- 

 midity or light, necessary to produce a distinguishable 

 morphological adjustment, is much greater than the unit 

 differences recorded by the instruments. In short, the 

 differences of habitats, as ascertained by thermograph, psy- 

 chrometer and photometer, are much greater than their 

 efficient differences, and, with respect to their ability to pro- 

 duce modification, habitats fall into relatively few categories. 

 A striking illustration of this is seen in the superfi- 

 cially very different habitats, desert, strand, alkali plain, 

 alpine moor, and arctic tundra, all of which are capable of 

 producing the same type of xerophyte. It follows from this 

 that many more or less plastic species of extensive 

 geographical area will find themselves in similar or identical 

 situations, measured in terms of efScient differences, and 

 will be modified in the same way in two or more of these. 

 In mountain regions, where interruption of the surface and 

 consequent alternation are great, the mutual invasion of con- 

 tiguous formations is of frequent occurrence, often resulting 

 in habitat forms.* The spots in which these nascent species, 

 such as Galium boreale hylocolum. Aster levis lochmocolus, etc., 

 are found, are often so related to the area of the parent 

 species as to demonstrate conclusively that these forms are 

 the result of polygenesis and not of migration. Naturally, 



'Clements, F. E. A System of Nomenclature for Phytogeography. 

 Engler 31 :b 17 1903. 



