112 JOHN H. SCHAFFNER Vol. XXII, No. 4 



favorable and unfavorable nutritive conditions and maturity 

 or senility seem to be among the most important factors. Dis- 

 turbances brought about by unequal distribution of autosomes 

 or characteristic distributions of allosomes appear also to be 

 conditions bringing about or influencing sexual states, probably 

 by an influence on the metabolism of the cell. Besides these 

 cases in which sexual states are associated with slight differences 

 in chromosome numbers are those in which a haploid or diploid 

 set may be associated with one sex or the other. It is well- 

 known, however, that doubling the number of chromosomes in 

 a gametophyte does not necessarily disturb the sexual condition. 

 It would appear that in cells with allosomes sex reversal should 

 be more difficult to bring about than in cells which have equal 

 distributions of chromosomes, but this is for future 

 experiments to determine. It is well known that in numerous 

 cases secondary sexual characters are reversed in spite of the 

 presence of allosomes. The probable reason that secondary 

 sexual characters become more prominent in the higher forms is 

 because the protoplast contains a greater number of hereditary 

 factors which can be influenced by the sexual state at the time 

 of their expression of characters. 



It becomes evident that sex cannot be associated primarily 

 with special chromosomes. The hypothesis of homozygous or 

 heterozygous combinations as determiners and controllers of 

 any given sexual state falls entirely outside of the complex 

 phenomena to be explained. Neither hereditary factors for sex 

 nor factors which control sex are in evidence; for in the vast 

 majority of plants the sexual state is determined in the somatic 

 cells in which neither segregation nor association of chrom- 

 osomes is taking place. So far as the gametophyte is concerned, 

 after passing through the evolutionary phase where her- 

 maphrodites are the rule, it is illuminating in the highest degree 

 to find that when the complete segregation of sex was attained 

 in the gametophyte it was not at all by the separation of units 

 with sexual control in the reduction division, but at a different 

 stage of the life cycle, namely, in the vegetative cells of the 

 sporophyte. Thus the megasporocytes do not segregate male- 

 ness from femaleness nor any properties or hereditary factors 

 which exercise a control over sexual expression, but all the cells 

 of a megasporocyte are differentiated with femaleness, and 

 give rise to female gametophytes, and in the same way the 



