RECENT CRITICISMS AND HYPOTHESES. 569 



that isolation is a condition precedent to such changes. Ap- 

 parently it did not occur to me as needful to specify this iso- 

 lation as making possible the differentiation of species ; since 

 it goes without saying that members of a species spreading 

 east, west, north, south, and forming groups hundreds of 

 miles apart, must, while breeding with those of the same 

 group be prevented from breeding with those of other groups 

 — prevented from having their locally-caused modifications 

 mutually cancelled. 



The importance of isolation has of late been emphasized 

 by Dr. Romanes and others, who, to that isolation conse- 

 quent on geographical diffusion, have added that isolation 

 which results from difference of station in the same habitat, 

 and also that due to differences in the breeding periods 

 arising in members of the same species. Doubtless in what- 

 ever way effected, the isolation of a group subject to new 

 conditions and in course of being changed, is requisite as a 

 means to permanent differentiation Doubtless also, as con- 

 tended by Mr. Gulick and Dr. Romanes, there is a difference 

 between the case in which an entire species being subject to 

 the same conditions is throughout modified in character, thus 

 illustrating what Mr. Gulick calls " monotypic evolution,^' 

 and the case in which different parts of the species, leading 

 different lives, will, if they are by any means prevented from 

 inter-breeding with other parts, form divergent varieties: 

 thus illustrating " polytypic evolution." 



§ 174/. Beyond geographical and topographical isolation, 

 there is an isolation of another kind regarded by some as 

 having had an important share in organic evolution. Fore- 

 shadewed by Mr. Belt, subsequently enunciated by Mr. Catch- 

 pool, fully thought out by Mr. Gulick, and more recently 

 elaborated by Dr. Romanes, " Physiological Selection " is held 

 to account for the genesis of marked varieties side by side 

 with their parents. It is contended that without the kind of 

 isolation implied by it, variations will be swamped by inter- 



