66 DISCONTINUOUS VARIATION. 



" the more diversified the descendants from any one 

 species become in structure, constitution, and habits, by 

 so much the more will they be better enabled to seize 

 on many and widely diversified places in the polity of 

 nature, and so be enabled to increase in numbers." 

 He also attached considerable importance to geograph- 

 ical isolation of a portion of a species, as an element in 

 the modification of species through natural selection. 



Though Darwin's principle of diversification of 

 structure is doubtless a very true one, yet it does not in 

 itself contain sufficient clue as to why a species should 

 split up into two or more varieties. Thus, if by some 

 means these actually arose, but both continued to in- 

 habit the same area, it is difficult to understand why 

 intercrossing should not rapidly reduce them to the 

 single species from which they took their origin. It 

 was to overcome this difficulty that Romanes suggested 

 his theory of " Physiological Selection." * This theory 

 is founded on the fact that individuals of a species, 

 though fertile with some, may be perfectly sterile with 

 other individuals, and this apparently independent of 

 any differences of form, colour, or structure. Romanes 

 thought that this incompatibility might run through a 

 whole race or strain, and so a group of individuals of a 

 species be in a physiological sense isolated from the rest, 

 and therefore able to vary independently, without hav- 

 ing their newly acquired characteristics swamped by 

 intercrossing. As Wallace has very clearly shown, t 

 this theory, in the form originally proposed by its au- 



*Journ. Linn. Soc. (Zool.), vol. xix. p. 337, 1886. Also "Dar- 

 win and after Darwin," vol. iii. p. 41. 

 f " Darwinism," p. 181. 



