422 



THE GRAMMAR OF SCIENCE 



times argue concerning reproduction as if like produced 

 like, so that if the individual corresponding to an organ 

 a were self-fertilised, we should have a number of progeny 

 also with organs a, or all individuals being equally fertile 

 the frequency polygon of the next generation would 



Fig. 30. 



simply be a magnified image of XCY, each vertical ordi- 

 nate being multiplied by a constant number 11 represent- 

 ing the degree of individual fertility. Now, such a con- 

 ception is very far from the truth ; just as all boys of 

 eleven years of age of the same height do not grow into 

 a group of boys of the same height at thirteen, but into 

 an array of definite variability (see p. 404), so all the 

 progeny of an individual of organ or character a, form 

 an array with definite variability, nor is the type of this 

 array, that is, its mean, identical with a, but with an 

 organ whose quantitative value is nearer to the modal 

 value c than a. These, the main features of inheritance, 

 are well established as we shall see later. They probably 

 hold as closely for asexual as for bisexual reproduction, 

 and with this mode of inheritance, the frequency distribu- 

 tion, even without selection, reroains the same from 

 generation to generation. In the case of bisexual re- 

 production we can conveniently replace the individual of 

 character a, by an individual of an artificial nature con- 



