452 DIRECTIVE FACTORS IN EVOLUTION: 



they could afford a starting-point for the formation of a new 

 species. And when closely related species in nature are ex- 

 amined, such as the different kinds of warblers, or of sedges, 

 it seems impossible that the slight differences separating 

 them should represent gaps produced by natural selection 

 through an elimination of intermediate forms. Thus an in- 

 spection of nature reveals a state of affairs which many inves- 

 tigators have come to believe to be much too refined to be a 

 product of natural selection." Some who admit that natural 

 selection is " capable of rough-hewing a species " doubt 

 its ability to put on " the polishing touches ". The answers 

 to this objection are three. (1) The idea that established 

 differences between species are too refined to be the work 

 of natural selection, shows a lack of appreciation of the fact 

 that the selection is often in relation to a very intricate and 

 subtle web of life, where the shibboleth that decides survival 

 or failure may be a very refined criterion indeed. (2) Vari- 

 ations are sometimes correlated, and a minor variation which 

 ia not itself of sufficient magnitude to have survival value 

 may be carried in the wake of one that has. (3) Some 

 variations are not minute fluctuations, but are brusque muta- 

 tions, springing fully formed into existence and therefore 

 at once of a magnitude to be sifted in the sieves of natural 

 selection. 



A second objection, also familiar to Darwin, is that indi- 

 viduals possessing an advantageous variation would have to 

 pair with others not possessing it, and that the new departure 

 would be swamped by the inter-crossing. To this there are 

 three answers: that similar variations often occur about the 

 same time in several individuals ; that many factors of iso- 

 lation operate towards reducing the range of inter-crossing 

 and bringing similar forms together ; and, thirdly, that many 



