Apkil 21, 1916] 



SCIENCE 



553 



modifying influences of the surroundings 

 which according to them cause the varia- 

 tions. Selection came in only secondarily, 

 by sorting out these variations and espe- 

 cially by eliminating some of them. Such 

 was the particular doctrine developed by 

 my master, A. Giard, at the Sorbonne. 

 Others have more or less absolutely refused 

 to grant any value to selection. Such was 

 the case of the philosopher Herbert 

 Spencer. We must also recognize that, 

 since the time of Darwin, natural selection 

 has remained a purely speculative idea and 

 that no one has been able to show its efS- 

 caey in concrete indisputable examples. 



The neo-darwinists, on their side, have, 

 in a general way, gone further than Darwin 

 because they see in selection the exclusive 

 factor of evolution and deny all value to 

 Lamarckian factors. This was the doc- 

 trine of Wallace, and has been especially 

 that of Weismann. I will digress a mo- 

 ment to speak of the ideas of these last- 

 mentioned authors, because of the influ- 

 ence which they have exerted and still ex- 

 ert, correctly in some respects, incorrectly 

 in others, at least as I think. 



Weismann attacked the doctrine of the 

 inheritance of acquired characteristics and 

 has incontestably shown the weakness of 

 the facts which had been cited before his 

 time in support of this kind of heredity. 

 But he went too far when he tried to show 

 the impossibility of this form of heredity. 

 In so doing, he starts from a conception 

 which meets with great favor; the radical 

 distinction between the cells of the body 

 proper, or soma, and of the reproductive 

 elements or germ cells. He saw, in these 

 two categories, distinct and independent 

 entities, the one opposed to the other. 

 Soma which constitutes the individual, 

 properly speaking, is only the temporary 

 and perishable envelope of the germ which 

 is itself a cellular autonomous immortal 



line, which is continuous through succes- 

 sive generations, and forms the substratum 

 of hereditary properties. The germ alone 

 has some kind of absolute value. The soma 

 is only an epiphenomenon, to use the lan- 

 guage of philosophers. The soma is of 

 course modified by external conditions, but 

 for one to speak of the inheritance of ac- 

 quired characteristics, the local modifica- 

 tions of the soma would have to be regis- 

 tered in the germ and reproduced in the 

 same form in the soma of following genera- 

 tions, in the absence of the external cause 

 which produced them in the first place. 

 Now, says Weismann, the possibility of 

 such an inscription, as it were, upon the 

 germ of a modification undergone by the 

 soma is not evident a "priori, and when we 

 go over the facts we find none supporting 

 this conclusion. There are indeed modifi- 

 cations which appear in one generation and 

 which are reproduced in the following gen- 

 erations ; but Weismann goes on to attempt 

 to prove that at their first appearance they 

 were not the effect of external factors on 

 the soma, but that they proceeded from the 

 very constitution of the germ, that they 

 were not really acquired and somatic, but 

 were truly innate or germinal. 



Such reduced to its essential points is 

 the negative contention of the doctrine of 

 Weismann. It rests upon the absolute and 

 abstract distinction between the soma and 

 the germ. In spite of the support which 

 this conception has had and still has, I con- 

 sider it, for my part, as unjustifiable in the 

 degree of strictness which Weismann has 

 attributed to it. It is true that the advance 

 in embryology and cytology often allows 

 us to identify the reproductive tissue and 

 to follow it almost continuously through 

 successive generations, but the conception 

 of its autonomy is at least a physiological 

 paradox. Though the continuity of the 

 germ cells is sufficiently evident in many 



