56 THE PRINCIPLES OF HEREDITY 



tion of the parent's development, but it is not a complete 

 recapitulation of the life- history of the race. No child could 

 develop unless there were recapitulation; but no race could 

 undergo extensive evolution unless the recapitulation were in- 

 complete and inexact. Shortened and emended, a life-history 

 is like some toilsome work begun by an inexperienced author 

 on insufficient knowledge and with half- formed plans. As his 

 knowledge grows and his plans take shape, every page is 

 altered. Almost every line is written over or erased. Passages 

 or chapters which were formerly important become irrelevant 

 and are omitted. New passages carry on the continuity of 

 the narrative. The finished work leaves few traces of its 

 life-history ; yet not so few but that a skilful student may 

 detect here and there, by passages not quite in keeping with 

 the rest, traces of the method by which the author worked. 



93. The mode of development of every individual makes 

 recapitulation certain. They are unskilful students who 

 deny its existence. Its very incompleteness is due to the 

 recapitulation by descendants of the regressive variations 1 

 of the ancestors. Naturally the most ancient parts of the 

 history are most incomplete and most inexact, for on them 

 variation has longest been at work. Even if existing plant 

 and animal forms had arisen by special creation, there would 

 still have been recapitulation. But in that case it would 

 have been recapitulation without variations. There would 

 have been no history ; or it would have been history in a 

 sentence. As it is, the dim and fleeting resemblances to 

 lower animals, displayed by the embryos of all higher types, 

 present the strongest evidence of the truth of the doctrine of 

 evolution that exists in the whole range of science. 2 We have 

 here a real history retold in every generation with the additions 

 and omissions made by the preceding generation a history 

 which ever grows longer with the lapse of years, and ever 

 more and more inaccurate and incomplete in its earlier parts. 



94. We know that children follow in the developmental 

 footsteps of their parents. We know that they do so with 

 variations. We know there has been evolution. We know 

 that monsters are rarely produced, and that they perish when- 

 ever produced, and so leave no descendants. These facts 

 being known to be true, the truth of the doctrine of recapitu- 

 lation follows as certainly as if it had been mathematically 

 demonstrated. It " may be maintained with the same degree 

 of certainty as that with which astronomy asserts that the 



1 See 97. 



2 See Damoin and After Darwin^ Romanes, vol. i., pp. 98-155. 



