IV.] FOUNDATION OF A THEORY OF HEREDITY. 1 97 



of the idioplasm' for the conception of germs of structure 

 ('Anlagen'). The different varieties of nucleoplasm which 

 arise during ontogeny represent, as it were, the germs of 

 Nageli ('Anlagen'), because, by means of their molecular 

 structure, they create a specific constitution in the cell-bodies 

 over which they have control, and also because they determine 

 the succession of future nuclei and cells. 



It is in this sense, and no other, that I can speak of the 

 presence of preformed germs ('Anlagen') in the idioplasm. 

 We may suppose that the idioplasm of the first segmentation 

 nucleus is but slightly different from that of the second onto- 

 genetic stage, viz. that of the two following segmentation nuclei. 

 Perhaps only a few groups of micellae have been displaced or 

 somewhat differently arranged. But nevertheless such groups 

 of micellae were not the germs (' Anlagen ') of a second stage 

 which pre-existed in the first stage, for the two are dis- 

 tinguished by the possession of a difterent molecular structure. 

 This structure in the second stage, under normal conditions of 

 development, again brings about the change by which the 

 different molecular structure of the third stage is produced, 

 and so on. 



It may be argued that von Baer's well-known and funda- 

 mental law of development is opposed to the hypothesis that 

 the idioplasm of successive ontogenetic stages must gradually 

 assume a simpler molecular structure. The organization of the 

 species has, on the whole, increased immensely in complexity 

 during the course of phylogeny : and if the phyletic stages are 

 repeated in the ontogeny, it seems to follow that the structure 

 of the idioplasm must become more complex in the course of 

 ontogeny instead of becoming simpler. But the complexity of 

 the whole organism is not represented in the molecular struc- 

 ture of the idioplasm of any single nucleus, but by that of all 

 the nuclei present at any one time. It is true that the germ- 

 cell, or rather the idioplasm of the germ-nucleus, must gain 

 greater complexity as the organism which arises from it be- 

 comes more complex ; but the individual nucleoplasms'of each 

 ontogenetic stage may become simpler, while the whole mass 

 of idioplasms in the organism (which, taken together, repre- 

 sent the stage in question) does not by any means lose in 

 complexity. 



