122 THEORIES OF HEREDITY 



a comparatively limited area, the reproductive organs 

 (ovaries or testes). These primary sexual organs can be 

 removed in the operation of castration without essentially 

 affecting any somatic cells except the components of 

 characters indicative of sex. When the influence of the 

 germ-cells is withdrawn such secondary sexual characters, 

 as they are called, tend to be transformed into those 

 indicative of the opposite sex. Although the succession 

 of individuals is of course prevented by the removal of 

 the primary organs, the life of the individual may continue 

 to its normal length. 



The problem of heredity may be stated as follows : — 

 How is it that a single germ-cell can produce, by repeated 

 division, an organism in which the peculiarities of the 

 somatic units of the parent are reproduced ? A single 

 cell separates from a small area in the body of the parent, 

 but it controls the development of the offspring, so that 

 the characters of every part of the parent are repeated 

 with more or less accuracy. 



It seems that there are only two possible ways in which 

 this marvellous fact can be explained. First, the whole 

 of the somatic cells may be so intimately connected with 

 the germ-cells that each of the latter bears within itself 

 the influence of the whole of the former — an influence, 

 too, of such a nature as to lead to the reappearance of 

 the corresponding somatic cell in the course of develop- 

 ment ; clearly, therefore, an influence of a material nature. 

 Secondly, we may look upon the germ-cells as directly 

 developed from the germ-cell from which the parent 

 arose. Parent and offspring would then resemble each 

 other, because they are developed from the same thing, 

 although at different times. 



There is an essential difference between these two 

 theories of heredity. In the first, the germ-cells may 

 bear the impress of every event which happens to the 

 somatic cells during the life of the parent, and such 

 characters may therefore be looked for in the offspring ; 

 in the second, offspring and parent can only resemble 

 each other in characters which were predetermined in 

 the germ-cell from which the parent arose. These latter 



