EXTENSION OF EVOLUTIONARY THEORIES 345 



be due to special variations, which he calls mutations. 

 These mutations have long been recognised ; they 

 rarely occur, and the feature of them is that the 

 organism is modified by them at a bound, in several 

 directions at once. Thus, amongst a number of plants 

 that de Vries was cultivating, most of which varied in 

 the usual manner, there were some that departed from 

 the species in several directions simultaneously, and so 

 were very conspicuously altered. When de Vries 

 crossed these " mutations " with each other they 

 produced plants of just the same kind. In other words, 

 contrary to the variations, the mutations retained their 

 new characters from the beginning when purely cultivated. 



Species have been brought about, he thinks, by these 

 mutations. Owing to certain internal causes the species 

 have, he thinks, suddenly divided into several with a 

 number of new characters which they all retained. 

 These mutations take no particular direction ; they are 

 partly useful, partly indifferent, partly injurious to the 

 organism. Natural selection weeds out the unfit, but 

 has no influence on the survivors. It cannot improve 

 the latter ; it must take the individuals as they are, 

 modified simultaneously in several directions. 



Thus, whereas we have so far considered the trans- 

 formation of species to be due to the fact that each 

 organ was slowly modified in connection with its 

 environment, in the sense of adaptation, this new theory 

 suddenly shatters the whole conception. When we 

 reflect on this we are disposed to join those who reject 

 the mutations as the formers of species. We know 

 that the species are made up of adaptations, and these 



