WEISMANN'S FIRST THEORY 45 



one hand, Weismann attributed to amphimixis, if not all, at 

 all events the vast majority, of those variations so necessary 

 for natural selection to work upon in the interest of the adapta- 

 tion of the species ; and, on the other hand, he denied the possi- 

 bility of an influence being exerted by the soma on the germ- 

 plasm. How, then, it was justly asked, could so many variations 

 be effected ? How, more especially, was the phenomenon of 

 correlated adaptation to be explained ? If every individual in 

 the animal world possesses an ancestral plasm, then it must be 

 Ms own plasm, and it must contain traces of the modification 

 of his body through external conditions. If the purely somatic 

 changes are not transmissible, how, then, can amphimixis bring 

 about such immense numbers of variations ? Amphimixis 

 reduces itself in this case to a mingling of identical germ-plasms. 

 Or even supposing certain variations to have been bequeathed 

 to us by the Protozoa, whose imicellular organisms are directly 

 exposed to the influence of the surrounding conditions, all the 

 possible combinations of the germ elements of the Protozoa 

 could hardly explain the correlated adaptations of the higher 

 Metazoa. On the one hand, somatic changes are not trans- 

 missible ; on the other hand, the germ-plasm is not exposed, 

 or is exposed in the very slightest degree, to modification by 

 external conditions ; and yet amphimixis, or the mingling of 

 germ-plasms, is the source of almost every variation in the 

 animal world ! But this affords no explanation as to the source 

 from which amphimixis derives the variations which it multiplies 

 and propagates. For amphimixis is not in itself a creative 

 source of variation — amphimixis can but multiply and disseminate 

 those variations and tendencies to variation which it finds at its 

 disposal. 



Germinal selection supplies the answer to the inquiry con- 

 cerning the original source of variation. It gives an explanation 

 of those variations and tendencies to variation of the germ- 

 plasm, which remained unexplained as long as the influence of 



