372 ON THE NUMBER OP POLAR BODIES AND 



resulting from this division would already contain only half the 

 number of ancestral germ-plasms from the father and the mother, 

 contained in the fertilized egg-cell. In this case the main object, 

 the reduction of the ancestral germ-plasms, would be gained by a 

 single division, and all the succeeding nuclear divisions, causing the 

 multiplication of these two first germ-cells, might take place by 

 the ordinary form of nuclear division, viz. 'equal division.' But 

 perhaps nature not only cares for this one main object alone, but 

 also secures certain secondary advantages at the same time. In the 

 case which we have supposed the egg-cells of the mature ovary 

 would only contain two different combinations of germ-plasm, 

 which we may call combinations A and B. Even if millions of 

 egg-cells were formed, every one of them would contain either A 

 or , and hence (at least as far as the female pronucleus is con- 

 cerned) only two kinds of individuals could arise from such eggs 

 viz. offspring A' and _Z?'. All the offspring A' would be as similar 

 to one another as identical twins, and the same would be true 

 of offspring S. 



But if the icoth instead of the 1st embryonic germ-cell entered 

 upon the ' reducing division,' a hundred cells would undergo this 

 division at the same time, and thus two hundred different com- 

 binations of ancestral germ-plasm would arise, and two hundred 

 different kinds of germ-cells would be found in the mature ovary. 

 A still greater number of different combinations of hereditary ten- 

 dencies would arise if the ' reducing division ' occurred still later ; 

 but undoubtedly the diversity in the composition of the germ- 

 plasm must be greatest of all when the ' reducing division ' does 

 not take place during the period in which the germ-cells undergo 

 multiplication, but at the end of the entire course of ovarian 

 development, and separately in each full-grown mature egg ready 

 for embryonic development. In such a case there will be as many 

 different combinations of ancestral germ-plasms as there are eggs, 

 for, as I have shown above, it is hardly conceivable that such a 

 complex body as the nuclear substance of the egg-cell composed 

 of innumerable different units would ever divide twice in pre- 

 cisely the same manner. Every egg will therefore contain a 

 somewhat different combination of hereditary tendencies, and thus 

 the offspring which arise from the different germ-cells of the same 

 mother can never be identical. Hence by the late occurrence of the 



