The Recapitulation Theory. 89 



organism can, as it were assert a power over the cells in maintain- 

 ing a unity of structure during development to a degree which is 

 not possible with the thicker cellulose walls of plants, though in 

 woody stems growth-pressure produces a certain amount of such 

 readjustment of dead cells. In any case, whatever the reason, it 

 remains true that while development in plants is usually direct, 

 and recapitulation the exception, the animal embryologist is 

 confronted with recapitulation on all hands with such amazing 

 profusion that a comprehensive principle is more obviously required 

 for their explanation. 



Another explanation of the difference between plants and 

 animals in the occurrence of recapitulation may lie in a greater 

 frequency of mutations in the phylogeny of plants. Also, animal 

 groups in their evolution have probably passed more frequently by 

 adaptation from one habitat to another. Witness, for example, the 

 number of groups from Ephemeridae to whales or penguins which 

 have become secondarily modified for an aquatic life. Even when 

 a like occurrence happens in plants, the stages of it are often nearly 

 or quite obliterated by short-circuiting. It is therefore quite unsafe 

 to argue in the case of plants that because a given adaptational 

 character shows no recapitulation in development it must have 

 originated through a germinal change. Such an attitude stands 

 a better chance of being sound in the case of animals. 



Concerning the facts of recapitulation, MacBride's textbook of 

 embryology Invertebrata (1914) is a veritable mine of information. 

 His attitude represents a return to an interpretation of the 

 significance of larval stages in relation to phylogeny on a frankly 

 neo-Lamarckian basis. Embryologists of the last two decades have 

 largely endeavoured to avoid this attitude, but without conspicuous 

 success. MacBride assumes that larval stages represent actual 

 ancestral groups of organisms. It is also significant, as we pointed 

 out in the case of plants, that the recapitulation seems always to 

 have originated in connection with the adaptation of the animal to 

 a new set of conditions. 



Although we have never seen it explicitly stated that embryonic 

 recapitulation implies the inheritance of acquired characters, 1 yet 

 it is probably the tacit recognition of this fact which has led to the 

 denial of recapitulation by those who believe only in germinal 

 variations as material for evolution. 



1 Since this was written I have received from Professor MacBride (1917) 

 a paper which I had not previously seen, in which it is definitely stated that 

 recapitulation implies the inheritance of acquired characters, and citing a 

 number of cases in support of this view. 



