GROWTH AND REPRODUCTION 59 



important, for on it must depend the occurrences of 

 evolution. For, obviously enough, if the rhythm 

 of ontogeny were firmly fixed, a species could never 

 originate new variations, but would remain rigid 

 and invariable. 



There are two different ways in which we may 

 think of these variations as taking place. Accord- 

 ing to Weismann, the rhythm of ontogeny can be 

 altered only by a basic change in the very first stage 

 of growth in the germ cell itself. This is, indeed, 

 the view generally held by students of biology. On 

 the other hand, it is maintained by the upholders of 

 the Hering-Semon hypothesis that the rhythm of 

 ontogeny can be changed by habit, that is, by linking 

 new engrams to old ones. Assuming that this can 

 happen in the neural structures of an individual, and 

 in lesser degree in other kinds of cells, how can the 

 engrams in question influence the germ cells ? This 

 is, indeed, the crux of the whole matter. The 

 mnemic theory holds that the germ cells share with 

 the nerve cells the possession of engrams engrams 

 linked by association so that they come into play in 

 definite succession in an automatic way, without the 

 repetition of the original stimuli. It assumes also 

 that these organized series of engrams in the germ 

 cells are due in part at least to an engram action of 

 the growing individual upon its own germ cells. 

 Clearly this is only possible if there is some kind of 

 connection between the nervous system and other 

 cells and the germ plasma, a neural telegraphy, 



