GERMINAL SELECTION 161 



sents one ancestor and the total number of ids repre- 

 sents the total number of the individual's ancestors. 

 All those ancestors would then contribute their part 

 to the constitution of the cell's characters. Moreover, 

 there are in the cytoplasm of every cell biophors of 

 every ancestral cell. If all those elements became 

 superadded, the cells would, at the end of the onto- 

 genesis, possess all those accumulated characters. 

 This being the case, they could not be determined in 

 any direction. 



If there was to be any determination at all, if one 

 character was to succeed another character instead of 

 being simply superadded to it, the biophors of every 

 mother-cell would have to die off; but Weismann 

 never mentions that biophors can be wiped out by 

 death. 



There is another difficulty. Why do biophors mi- 

 grate from the nucleus? This question is as unan- 

 swerable as the question why Darwin's gemmules are 

 attracted by certain cells. Weismann mentions a cer- 

 tain stage of maturity, but we fail to see how biophors 

 could mature and why they should mature in one cell 

 and not in another cell, in a soma cell and not in a 

 germ cell. 



This could only be accounted for by a difference 

 in the conditions which influence them, but such an 

 explanation would not be consistent with the postu- 



