208 HEREDITARY CHARACTERS 



which has been kept in the category of individual characters 

 and has been prevented from becoming a racial character by 

 the action of natural selection, in spite of the fact that it is 

 far more ancient than a great many of the racial characters. 



The hypothesis that sex has continued as an individual 

 character in some important respects and is transmitted in 

 a similar manner to individual characters, would rest upon 

 a very insecure basis, were it not that there is a mechanism 

 in cell division which provides for the alternative distribu- 

 tion of certain bodies before fertilisation. All that has been 

 said with regard to the alternative transmission of individual 

 characters applies equally to the alternative transmission of 

 sex. Sex, however, is the only character upon which the 

 action of natural selection has been such as to keep it 

 alternative, and therefore we should expect to find some 

 differences between the way in which it is transmitted and 

 the results of the Mendelian experiments. In the case of 

 every other individual character, the final result must be 

 either elimination or the production of a racial character. 



It is of course necessary to assume that there are several 

 entities in every cell representing both sexes. It is quite 

 possible that in the case of the insects observed by Wilson 

 these entities may be collected in one or two chromosomes. 

 It is not necessary, though, to suppose that this has 

 happened in the majority of cases. We constantly find 

 instances of peculiar differentiation with regard to par- 

 ticular characters in living organisms, and it may well be 

 that the cases of the animals where a difference exists 

 between the cells of the male and female, are due to some 

 special differentiation, the exact purport of which we do not 

 as yet understand. Where we have no evidence as to the 

 existence of any particular sex-producing chromosome, it 

 is not at all necessary to assume one. The entities that 

 control the appearance of one sex or the other might be 

 present in one or in several chromosomes. It certainly 

 seems that both sexes are represented in all cells. The 

 alternative distribution of the chromosomes and their re- 



