ADAPTIVE VARIATIONS. 391 



the reader who desires more detailed discussion is re- 

 ferred. 



Admitting that somatic variations are, on the whole, 

 adaptive, and admitting also to a very limited extent 

 the cumulative influence of changed conditions of life, 

 are we to agree with Henslow * that the close adapta- 

 tion of plants to their environment is due entirely to 

 the responsive power of protoplasm to the external en- 

 vironmental forces, and that it is absolutely unneces- 

 sary to call in the aid of Natural Selection? By no 

 means. Adaptive variation may be responsible for a 

 good deal of the adaptation observed in plants, and for 

 a very small part of that observed in animals, but prob- 

 ably in each case by far the larger portion must be as- 

 cribed to the ever present and ever acting agency of 

 Natural Selection. For instance, Henslow argues very 

 plausibly that inasmuch as certain plants when kept in 

 a dry atmosphere develop spines and other characters 

 similar to those possessed by desert plants, it is valid to 

 conclude that these desert plants owe their peculiar 

 characters to the direct action of the dry hot climate, 

 and to that alone. Supposing this explanation to be 

 correct, however, we ought, as Wallace points out,f 

 to find plants with spines and the other characteristics 

 of desert plants abounding in all dry countries, but 

 very rare or wanting in moist and fertile districts. But 

 this is by no means the case. Wallace states that 

 many of the peculiarities of desert plants are present 

 in the flora of the Brazilian Campos, and in that of the 

 Galapagos and the Sandwich Islands, but very few of 



*Ibid., pp. 14 and 32. 



f Nat. Science, vol. v. p. 177, 1894. 



