THE EVOLUTION OF TYPES 281 



change the form of the whole race but the same or closely simi- 

 lar mutations appearing at the same time in a host of individ- 

 uals, it may be of various species within a genus, can and cer- 

 tainly do result in the origination of a new type less or equally 

 or more in harmony with the ever changing environment. 



If mutations occur at random we are faced with the 

 problem of how the type in which they occur can persist. We 

 must admit that in order to insure persistence all the mutations 

 occurring in a single type must be linked or complementary. 

 A modification occurring in the dentition unaccompanied by 

 suitable changes in skull, in musculature, in the Avhole organ- 

 ism, cannot become dominant. Thus we find in time complete 

 specific and later generic differences profound in nature and 

 far-reaching in influence, the beginnings of the formation of 

 new families, new orders, new classes. As Bergson puts it 

 there must be postulated an original impetus which is inherited. 

 "This impetus, sustained right along the lines of evolution 

 among which it gets divided, is the fundamental cause of 

 variations, at least of those that are regularly passed on [muta- 

 tions], that accumulate and create new species. In general, 

 when species have begun to diverge from a common stock, 

 they accentuate their divergence as they progress in their evolu- 

 tion. Yet, in certain definite points, they may evolve iden- 

 tically ; in fact they must do so if the hypothesis of a common 

 impetus be accepted." 



