XII.] CONJUGATION AND SEXUAL REPRODUCTION. I31 



the individuality of the being which is produced by this 

 ontogeny. 



There is therefore, in my opinion, nothing inadmissible in the 

 idea of the breaking up of the chromatin rods or idants, during 

 each nuclear resting stage, if only the single ids remain un- 

 changed ; but certain facts in heredity, to be mentioned imme- 

 diately, support the view that the specific hereditary substance 

 from one or both parents can be contained in the germ-cells 

 of the child, and this presupposes that it is at least possible and 

 perhaps the rule, for the order and arrangement of the ids in 

 the idants to remain unchanged from the germ-cells of the 

 parent to those of the offspring. I would, then, assume that, at 

 least on the way from germ-cell to germ-cell, the views of van 

 Beneden and Boveri are upon the whole correct, viz. that the 

 chromatosomes (idants) only apparently break up during the 

 nuclear resting stage, but in reality persist. I imagine that, after 

 the period of the resting stage, they are generally composed of 

 the same ids, for the most part arranged in series similar to those 

 which existed before the preceding nuclear division. We are 

 already acquainted with such astonishingly delicate mechanical 

 arrangements in cells, that the existence of special provision 

 for maintaining the original arrangement of the rod elements 

 (ids) might be looked upon as very far from an impossibility. 

 Even if direct observation should fail to answer this question in 

 the future, some certainty might be reached by those indirect 

 means which often lead us to a final decision in such excessively 

 minute biological questions — viz. the means provided by an 

 examination of the facts of heredity. Even now there is, I think, 

 one such fact, supporting the idea of a continuity of the idants ; 

 I mean the frequently observed fact that a child may pre- 

 dominantly or even exclusively resemble one of its parents alone. 

 If the elements of the chromatin rods, i. e. the ancestral plasms, 

 were irregularly mingled together in each nuclear resting stage, 

 to be rearranged at random in the idants, it would scarcely 

 ever happen that the scattered ids would come together in a 

 series like that which existed in the original paternal or 

 maternal idants. The individual stamp of a nuclear rod (idant) 

 must entirely depend upon its construction out of particular ids. 

 Nevertheless, we must not regard this constitution as for ever 

 unchangeable. The universally observed change of indivi- 



K 2 



