332 



KELLOGG AND BELL 



ought to be pretty closely similar on both sides of the insect. 

 That as much bilateral variety as actuall}^ exists, in many of the 

 species examined by us, should exist — a variety comparable in 

 certain cases even with the degree of variety revealed by the 

 comparison of considerable series of individuals — is a state of 

 affairs that only confirms us in the belief that these innumerable 

 small continuous variations, on which for so long the thorough- 

 going selectionists have put their faith as the sufficient bases for 

 natural selection's species-forming work, are clearly not com- 

 petent to serve as such bases. If these " continuous " variations 

 are the foundation stones of new species, some other agent than 

 selection must be found or invoked to build several courses on 

 them, to produce some cumulation of them, before natural selec- 

 tion finds them of that life and death worth which is the prere- 

 quisite for her potent interference. 



