1896.] On Heredity and Rejuvenation. 99 



sumes that the majority of variations fall below the value 

 maintained by natural selection, and consequently that when 

 the influence of natural selection is eliminated (as in disuse), 

 the mere variation will bring the traits concerned to disap- 

 pearance. It marks Weismann's style of thought to find that 

 he has entirely omitted to determine whether his assumption 

 was correct, and nevertheless in his book, " The Germ Plasm," 

 presents panmixia as an established law. As a matter of fact, 

 the statistics of variations which we already have, show that 

 his assumption is erroneous, and that it is equally probable 

 that mere variation will magnify a characteristic as it is that 

 it will diminish it. 



Let us return to the embryo. The following hypothesis 

 may be advanced : 



The loss of ancestral clwreiet< ristics in the embryo is due to post- 

 selection, the cells being kept in a rejuvenated stage, in order that 

 they may after n-<wh xeromplish new differentiations. 



This conclusion follows directly from the preceding consid- 

 erations, and, therefore, needs no further defense. 



IV. Concluding Remarks. 



The views presented in the preceeding sections are inti- 

 mately connected one with another and collectively determine 

 our conception of the process of heredity. The conception 

 concerns only the process and not the essential character or 

 cause of heredity. According to my view, heredity exists in 

 all cells, but a of the liv- 



ing substance, and can be complete only in embryonic cells ; 

 embryonic cells arise under very various conditions. That 

 which is novel in this theory is the significance attributed to 

 embryonic cells. Embryonic cells I prefer to designate as re- 

 juvenated cells. 



The theory above presented is an unavoidable consequence 

 of the facts known, and stands in absolute contradiction with 

 Weismann's theory of the germ plasm. 



I have read with the greatest conscientiousness every article 

 hitherto published by Weismann, which deals with his theo- 

 ries of heredity. My final impression from this study is that 



