VARIATIONS AND HEREDITY 321 



heredity, a failure of heredity to work perfectly? Neither of these 

 views is entirely wrong, nor yet entirely right. Logically, there is no 

 more reason for considering one of two necessary factors first and the 

 other second, but pedagogically, there seem to be some reasons why 

 at least a general consideration of the variation factor is necessary as 

 a preliminary to an understanding of the principles of heredity. On 

 the other hand, it has come to be felt that a rather thorough knowledge 

 of the mechanism of heredity is prerequisite to any adequate under- 

 standing of the origin of or the causes of variations. In other words, 

 it seems necessary to consider variation both before and after heredity 

 We shall, therefore, in the next chapter present a general survey of 

 variation, viewed broadly with reference to heredity and environment. 

 The authors of that chapter assume as proven some of the principles 

 that can be adequately discussed only after we know a great deal about 

 heredity. For example, modifications (acquired characters) are classi- 

 fied as non-inheritable; mutations are classed as heritable before we 

 have any knowledge of this complex category of variations; combina- 

 tions are dealt with before Mendelian heredity is discussed; and the 

 possible induction of new hereditary conditions by changes of the en- 

 vironment is dealt with briefly before we have had a chance to discuss 

 the normal origin of new hereditary characters. If, therefore, the read- 

 er will consider this chapter as a preliminary survey and no more, and 

 recognize that it is used as a pedagogical device to facilitate an under- 

 standing of the intricate features of heredity, he will find it excellent 

 for this purpose. 



The more detailed study of variation (including the statistical 

 study of variation, variation and selection in pure lines, the inheritance 

 of somatic modifications, the mutation phenomenon and its causes) is 

 much more favorably pursued after we have learned as much as we 

 can about heredity. 



Before proceeding to a detailed consideration of Mendehan 

 heredity, it is essential that the reader appreciate the fact that, though 

 Mendel's laws appear to be merely the laws of hybridization, they have 

 a much wider application: they are really the laws of sexual reproduction. 

 The method of hybridization is simply a convenient way of bringing 

 out the laws sharply, for in different varieties of animals or plants the 

 contrasting characters stand out more definitely than in members of 

 the same race or variety. As a matter of fact, however, every indi- 

 vidual, except those belonging to closely inbred strains or those pro- 

 duced asexually, is a hybrid with regard to many of its characters. 



