410 



HISTORY OF THEORIES 



characters x, y, z in all their entirety, to start a new organism 

 again with the same capital. Balbiani, who was not influenced 

 by theoretical considerations, observed in Chironomus that the 

 future reproductive cells were isolated before even the blastoderm 

 was completed ; that is to say, before almost any differentiation 

 had occurred, a portion of the unspecialised ovum was insulated 

 to continue the constancy of the species. 



In this aspect the reproductive cells form a continuous chain, 

 and the reproduction of like is as natural and necessary as it 

 was in the Protozoa. No special theory is required. Similar 

 conditions produce similar results. Unfortunately, however, 

 a serious difficulty besets this easy theory. Such an early appear- 

 ance and insulation of the reproductive cells, continuous with 

 the very ovum itself, does indeed occur, and where it does the 

 problem of heredity is simple. Early origin of special germ-cells, 

 distinguished from those of the general " body," has been ob- 

 served in some " worm-types " (leeches, Sagitta, threadworms, 

 many Polyzoa) and in some Arthropods (Moina and Cyclops 

 among crustaceans, not a few insects, Phalangidae among 

 spiders), while indications of the same early separation are not 

 wanting in a number of other organisms. But it must be dis- 

 tinctly allowed that in most cases it is only after differentiation 

 is relatively advanced that the future reproductive cells make 

 their appearance. Thus we have to pass from the few cases as 

 yet known of the continuity of the germinal cells, to the more 

 general fact of the " continuity of the germ-plasma." 



Weismann's Theory. — Weismann, like the previous investi- 

 gators, had reached his conclusion independently. In the fact 

 of continuity between the reproductive elements of generations, 

 the solution of likeness must be found. But a direct chain 

 of cellular continuity can only be said to exist in a few cases. 

 The solution which is proposed for the majority of cases is 

 as follows : 



(i) "In each development a portion of the specific germinal 



