70 REPRODUCTION AND LIFE HISTORY. 



short, the individual's recapitulation of racial history is 

 general, but not precise. 



But we do not understand how the recapitulation is sustained. Has 

 the protoplasm of the embryo some unconscious memory of the past ? 

 Have the protoplasmic molecules, as Haeckel puts it, learned long since 

 some rhythmic dance which they cannot forget ? And, to what extent 

 must there be similarity of external conditions if the recapitulation, " the 

 perigenesis of the plastidules," is to be sustained? 



(4) Organic continuity between generations. — Heredity. 

 — Everyone knows that like tends to beget like, that off- 

 spring resemble their parents, and sometimes their ancestors 

 (atavism). Not only are the general characteristics trans- 

 mitted, but minute features, idiosyncrasies, pathological 

 conditions, innate or congenital in the parents, may be 

 transmitted to the offspring. 



Many attempts have been made to explain this, but the 

 first suggestion with any scientific pretensions was that the 

 reproductive cells, which may become offspring, consist of 

 samples accumulated from the different parts of the body. 



This was a very old idea, but Herbert Spencer and 

 Charles Darwin gave it new life. According to Darwin's 

 " provisional hypothesis of pangenesis," the reproductive 

 cells accumulate gemmules liberated from all parts of the 

 body. In development these gemmules help to give rise to 

 parts like those from which they originated. This hypo- 

 thesis has been repeatedly modified, but except in the 

 general sense that the body may influence its reproductive 

 cells, " pangenesis " is discredited by most biologists. 



The idea which is now accepted with general favour is, 

 that the reproductive cells which give rise to the offspring 

 are more or less directly continuous with those which gave 

 rise to the parent. This idea, suggested by Owen, Haeckel, 

 Rauber, Galton, Jager, Brooks, Nussbaum, and especially 

 emphasised by Weismann, is fundamentally important. 



At an early stage in the development of the embryo the 

 future reproductive cells of the organism are distinguishable 

 from those which are forming the body. These, the somatic 

 cells, develop in manifold variety, and, as division of labour 

 is established, they lose their likeness to the fertilised ovum 

 of which they are the descendants. The future reproductive 

 cells, on the other hand, are not implicated in the formation 



