Julian S. Huxley 267 



This could have been accomplished in one of two ways. Either one 

 of the X-chromosomes was transformed into a Y — a hypothesis so 

 improbable as not to be worth further consideration; or else the 

 normal effect of two X-chromosomes was overridden by the effect of 

 the delayed fertilization upon the eggs' metabolism. 



If this latter hypothesis is adopted, then we must assume that 

 50 °/^ of one of these " all-male " cultures are normal, of constitution 

 XY, but that the other 50 7o are what may be called "masculinized 

 females," or, better, to adopt Shull's useful term ('14), somatic males, of 

 constitution XX. 



All of this latter class of males, if bred to normal females under 

 normal conditions, should give only female offspring, since all sperma- 

 tozoa as well as all eggs in such a cross will contain an X-chromosome. 

 The remaining males mated in a similar way will give the normal 1 : 1 

 sex-ratio, so that the effect of mating normal females to a number of 

 males selected at random from an all-male culture will be to give a pre- 

 ponderance of females in the next generation in the proportion 3 $ : 1 </ 

 (a less preponderance than the previous preponderance of males). Both 

 males and females, however, of this generation will be of normal con- 

 stitution, so that their offspring would show a normal 1 : 1 sex-ratio. 



Similar effects of late fertilization on sex-ratio are given by Pearl 

 and Parshley ('13) for cattle. 



Other important evidence in favour of the possibility of external 

 factors overriding sex- chromosome constitution is botanical, and is 

 given by ShuU ('14) in his paper on Lychnis dioica (see especially 

 p. 285 et seqq.). 



In the first place, his experiments have led him to the conclusion 

 that hermaphroditism in Lychnis dioica may be either (a) genetic, 

 apparently due to the presence of a definite gene, and therefore herit- 

 able, or (6) somatic, due to some cause outside the germ-plasm causing 

 a shift of a genetically unisexual individual to a hermaphrodite condition. 

 These somatic hermaphrodites in the great majority of cases he found 

 to behave as males, and he concludes that they are genetically males in 

 which chromosome constitution has been overridden by external causes. 



One particular hermaphrodite plant, however, when used as pollen 

 parent in three separate crosses, gave in each case a large progeny 

 consisting entirely of female plants. In no other experiment has Shull 

 ever obtained even a moderate family composed only of females. His 

 explanation is that the plant in question, though also a somatic 

 hermaphrodite, was not a modified male but a modified female, of 



19—8 



