146 THE THEORIES OF EVOLUTION 



The grandson would then resemble his grand- 

 father without resembling his father. Such illustra- 

 tions could be easily multiplied; but the point to be 

 remembered is that variation results from purely in- 

 ternal causes, from the very fact of amphimixia. 

 //Natural selection acts upon characters thus trans- 

 / mitted and we can now dispense with the hereditary 

 transmission of characters, with environmental influ- 

 ^-ence and other causes of variation suggested by 

 Lamarck. Xatural selection is the only origin of 

 those chance variations, innate and individual, which 

 are one of the postulates of Darwinism. 



Every detail of the Weismannian doctrine is thus 

 explained logically by another detail. Still Weis- 

 mann did not succeed in preserving the integrity of 

 his doctrine, and certain facts in contradiction to it 

 necessitated additions, amendments, compromises and 

 even the adjunction to his original system of a new 

 theory, the theory of germinal selection. 



Among those modifications and additions we will 

 only mention those which concern the relations be- 

 tween soma plasm and germ plasm, a question ger- 

 mane to that of the hereditary transmission of acquired 

 characters. According to the original theory, the cell 

 of the adult individual contains only such deter- 

 minants as correspond to its proper characters. This 

 is contradicted, however, by facts of a-sexual repro- 

 duction. The end cell of a bud, for instance, must 



