CRUSTACEA AND MOLLUSCS 26 1 



is brought about. Still, natural selection may, of course, 

 act with such intensity as to give predominance to a 

 small number of animals that possess a variation, and 

 leave all their fellows to be destroyed. Thus in very 

 severe cold a few deer with particularly thick coats may 

 be preserved. 



It is due to amphimixis alone that the species remain 

 within their limits for thousands of years. If the 

 variations were not constantly neutralised, they would 

 long ago have modified animals so much that they 

 would no longer have the least resemblance to their 

 ancestors. If it were not for amphimixis there would 

 not be to-day any fishes resembling the fishes of earlier 

 days. 



On the other hand, amphimixis allows no division 

 of a species. How can a new species be developed 

 from an older one, and this be preserved, if the 

 promiscuous crossing is continually at work destroying 

 or generalising all new characters .'* In that case, we 

 can safely say, there would not be a number of different 

 species if there were not a force that prevented the 

 crossing of the members of the new species with those 

 of the parental stock. We have alluded several times 

 already to this force. It is isolation. We shall deal 

 with it in the tenth chapter, and see why there are 

 many different species, when we have learned why the 

 species may be arranged in a system according to their 

 greater or less resemblance, and why there are unifying 

 types of species. But we have other questions to 

 answer first. 



