HERITABLE DIFFERENCES 39 



ity, transitory so far as their inheritance is concerned. 

 They are the result of soil rather than seed. 



"Combinations" and "mutations" are more deep- 

 seated. They are conditioned by the germinal nature 

 of the organism and may, therefore, be passed on as 

 hereditary. 



Combinations are the result of a new deal after a 

 reshuffling of the cards. Nothing essentially new, which 

 was not already present in one or the other of the 

 parental lines, is introduced but a different arrange- 

 ment or bringing together of old qualities is effected. 

 This process of variation through hybridization is the 

 concern of Mendelism and will be considered further on. 



Mutations, like Minerva springing full-fledged from) 

 the head of Jove, are something qualitatively new which' 

 appear abruptly without transitional steps and that! 

 breed true from the very first. 



A distinctive qualitative character marks mutations, 

 like the discontinuous differences between such chemical 

 compounds as carbon monoxide (CO) and carbon 

 dioxide (CO 2 ), as Bateson has pointed out, but the 

 leap from one to the other may be so small that it is 

 difficult to ascertain by inspection whether the differ- 

 ence is due to mutation or to modification. The test 

 comes in breeding, since the progeny of a modification, 

 or "fluctuation" as deVries terms it, will revert to the 

 old average of the parental generation, while the 

 progeny of a mutation will vary around a new average 

 set by the mutation itself. 



The series of positions taken by the lower end of a 

 swinging pendulum illustrate what is meant by these 



