ORIGIN OF ADAPTATIONS AND OF SPECIES 249 



cross pollination was not necessary to the production of new 

 species by mutation, and when employed did not accelerate the 

 results materially. As a botanist, De Vries confined his inves- 

 tigations to plants, but his general conclusions are perhaps 

 equally applicable to animals, and his experiments are doubt- 

 less being repeated by zoologists. 



Through his exhaustive experiments, De Vries has partly 

 attained a long-desired object, in that he has removed the ques- 

 tion of the origin of some species " from the purely theoretical 

 to the concrete." 



The mutation theory is not primarily a theory of the origin 

 of adaptive characters. It endeavors to account for the origin 

 of certain characters, which may or may not prove useful to 

 their possessors. Indeed, one great merit of De Vries' theory 

 is that it affords an explanation for the existence of variations 

 which are not useful. Now Darwin does not pretend to 

 account for the origin of variations, but he shows how given 

 variations, if useful, may be preserved and accumulated. 

 Thus the theory of De Vries supplements that of Darwin and 

 does not antagonize it; even though De Vries himself takes 

 much pains to contrast the two theories, and even asserts that 

 new species arise exclusively as mutations. Both theories, 

 indeed, are theories of the origin of species; but according to 

 De Vries, specific characters spring into existence, irrespective of 

 their usefulness ; while according to Darwin, useful characters, 

 and these only, are premised, as the starting point of the evolu- 

 tion of certain kinds of species. Thus, as another has said, 

 natural selection begins where the mutation theory leaves off. 



Isolation. The theory of isolation as given by Gulick and 

 by Romanes is highly important as affording an explanation 

 of " the rise and continuance of specific characters which need 

 not necessarily be adaptive characters." By isolation is meant 

 " simply the prevention of intercrossing between a separated 

 section of a species or kind and the rest of that species or 

 kind. ... So long as there is free intercrossing, heredity 

 cancels variability, and makes in favor of fixity of type. Only 



