THEIR SIGNIFICANCE IN HEEEDITY. 373 



' reducing- division ' the greatest possible variability in the offspring- 

 is secured. 



If my interpretation of the second polar body be accepted, it 

 is obvious that the late occurrence of the ' reducing division ' is 

 proved. At the same time we receive an explanation of the ad- 

 vantage gained by the postponement of the reduction of the germ- 

 plasm until the end of the ovarian development of the egg ; 

 because the greatest possible number of individual variations in 

 the offspring are produced in this way. 



If I am not mistaken, this argument lends additional support to 

 the idea which I have previously propounded, that the most 

 important duty of sexual reproduction is to preserve and con- 

 tinually call forth individual variability, the foundation upon which 

 the transformation of species is built ] . 



But if it be asked whether the postponement of the ' reducing 

 division ' to the end of the ovarian development of the egg is incon- 

 sistent with the preservation of the other half of the dividing nucleus, 

 I should be inclined to reply that a ' reducing division ' of the 

 mature egg, resulting in the production of two eggs, was probably 

 the phyletic precursor of the present condition. I imagine that the 

 division of the mature egg-cell although it is now so extremely 

 unequal was equal in very remote times ; but that for reasons of 

 utility, connected with the specialization of the eggs of animals, it 

 gradually became more and more unequal. It is now hardly pos- 

 sible to give in detail the various reasons of utility which have 

 brought about this condition, but it may be assumed that the 

 enormous size attained by many animal egg-cells has been espe- 

 cially potent in producing the change. 



A careful consideration of this last point seems to me to be 

 demanded by a comparison of the egg-cells with the male germ- 

 cells. Just as the female germ-cells of animals are distingiiished 

 by the attainment of a large size, the male germ-cells are generally 

 remarkable for their minute proportions. In most cases it would 

 be physiologically impossible for a large egg-cell, rich in yolk, to 

 attain double its specific size in order to undergo division into two 

 equal halves and yet to remain of the characteristic size. Even 

 without the additional difficulties imposed by the necessity for such 



1 See the preceding Essay on 'The Significance of Sexual Reproduction in the 

 theory of Natural Selection.' 



