376 



THE AMERICAN NATURALIST [Vol. XL 



weaker determinants get less food, grow less well, and consequently 

 become in turn still less capable of nutrition. Once started on the 

 downward path the determinant descends more and more until it is 

 wholly eliminated. Nothing can save the determinant from this fate 

 except the elimination by natural selection of the adults of the strain 

 possessing the deteriorating determinant — and this will only happen 

 when the degenerating organ again becomes necessary to the welfare 

 of the postembryonic organism. 



Not only the degeneration but also the upbuilding of an organ can 

 be explained by the hypothesis of germinal selection. For a certain 

 strong determinant once having been selected, it will gather to itself 

 all available nutrition at the expense of the other determinants; it 

 will flourish in the race and will only stop its continual accretions 

 when it produces organs so developed as to be disadvantageous to the 

 active organism. 



In criticising this hypothesis one can only admit that it explains so 

 many facts that we hope it will some day be demonstrated. As it is, 

 it stands to-day a bald hypothesis based on numerous probable but 

 unproven assumptions. * 



Coming now to Weismann's position on the origin of the specific 

 type we find it of interest as being clearly opposed to that of de Vries, 

 Both theories accept the idea of unit characteristics which are repre- 

 sented in the germ by particles. The theories begin to diverge in 

 respect to these particles. De Vries concludes that these particles 

 change suddenly, probably by molecular changes within them, so that 

 a new characteristic arises suddenly and tends thereafter to persist. 

 The characteristic may be modified by selection, but its essential 

 nature cannot be changed thereby. Weismann on the contrary 

 regards these particles as being in a constant state of variation which, 

 when continued long in one direction, will result in the elimination 

 of a character or in its excessive development. Species are originally 

 connected by transition forms as are to-day the terrestrial snails of the 

 Celebes (page 299). Weismann repeatedly emphasizes the idea that 

 all variations are quantitative and that "what appears to us a quali- 

 tative variation is, in reality, nothing more than a greater or loss" 

 (vol. 2, p. 151). Here then we have clearly set forth the issii(> lift ween 



genetic significance are always qualitative; the othrr, that tiny are 

 quantitative only. This difference between the two schools would 

 seem to be a qualitative one. But alas for the peace of mind of him 

 who seeks clear distinctions, the quantitative may produce the qualita- 



