142 EVOLUTION AND GENETICS 



of a given advantage each change must add to give 

 it a chance to become estabhshed in a population of 

 a given number. Since only relatively few of the in- 

 dividuals produced in each generation become the 

 parents of future generations, numbers count heav- 

 ily against any one individual establishing itself. 

 This is a most difficult problem for which we have 

 practically no data, and as yet only the beginning 

 of a theoretical analysis has been made of this side of 

 the selection problem. Haldane has developed a par- 

 tial analysis of the problem for a few Mendelian sit- 

 uations. He points out that the problem is extremely 

 complex and that there is at present not much quan- 

 titative information to furnish material for such a 

 study of natural selection by means of gene mutations. 



The Diagnostic Characteristics of Species and the 

 Origin of Species by Natural Selection 



It has often been pointed out that the characters 

 used by systematists to separate species have as a 

 ride nothing whatsoever to do with the adaptive fea- 

 tures of species. The latter are largely physiological. 

 Yet if species have originated through adaptive 

 modifications, it might be expected tliat physiolog- 

 ical characters would be the most distinctive ones 

 that distinguish species from one another. 



The solution of this paradox is, I think, to be 

 found in the many-sided effects produced by each 

 gene. It has been shown, particidarly clearly in the 



