THE CELLULAR BASIS OF HEREDITY 121 



but always the division of the chromosomes is equal and non-differential. 

 When once the various tissues have been differentiated the further divi- 

 sions in these tissue cells are usually non-differential even in the case of 

 the cell bodies. 



There can be no doubt that this remarkably complicated process of 

 cell division has some deep significance ; why should a nucleus divide in 

 this peculiarly indirect manner instead of merely pinching in two as was 

 once supposed to be the case? "What is the relation of cell division to 

 embryonic differentiation? In this process of mitosis, or indirect cell 

 division, two important things take place: (1) Each chromosome, 

 chromomere and centrosome is divided exactly into two equal parts so 

 that each daughter structure is at the time of its formation quantita- 

 tively one half the size and qualitatively precisely like its mother struc- 

 ture. (2) Accompanying the formation of radiations, which go out 

 from the eentrosomes into the cell body, diffusion currents are set up 

 in the cytoplasm which lead to the localization of different parts of the 

 cytoplasm in definite regions of the cell, and this cytoplasmic localiza- 

 tion is sometimes of such a sort that one of the daughter cells may con- 

 tain one kind of cell substance and the other another kind. Thus while 

 mitosis brings about a scrupulously equal division of the elements of the 

 nucleus, it may lead to a very unequal and dissimilar division of the 

 cytoplasm. In this is found the significance of mitosis and it suggests 

 at once that the nucleus contains undifferentiating material, viz., the 

 idioplasm or germplasm, which is characteristic of the race and is 

 carried on from cell to cell and from generation to generation ; whereas 

 the cell body contains the differentiating substance, the personal plasm 

 or somatoplasm which gives rise to all the differentiations of cells, 

 tissues and organs in the course of ontogeny. 



Weismann supposed that the mitotic division of the chromosomes 

 during development was of a differential character, the daughter chro- 

 mosomes differing from each other at every differential division in some 

 constant and characteristic way, and that these differentiations of the 

 chromosomes produced the characteristic differentiations of the cyto- 

 plasm which occur during development. But there is not a particle of 

 evidence that the division of chromosomes is ever differential; on the 

 contrary, there is the most complete evidence that their division is 

 always remarkably equal both quantitatively and qualitatively. If 

 daughter chromosomes and nuclei ever become unlike, as they sometimes 

 do, this unlikeness occurs long after division and is probably the result 

 of the action of different kinds of cytoplasm upon the nuclei, as is true, 

 for example, in the differentiation of the chromosomes in the somatic 

 cells as contrasted with the germ cells of Ascaris (Fig. 32). But while 

 the chromosomes invariably divide equally, other portions of the nucleus 

 may not do so. Achromatin and oxychromatin, like the cytoplasm, may 



VOL. LXXXV. — 9. 



