5 o4 HEREDITY AND SEX 



of the two sexes. It has been pointed out by Walker that pro- 

 duction of a preponderance of females when food is abundant and 

 a preponderance of males when food is scarce is an advantageous 

 automatic regulation which natural selection would tend to 

 perpetuate. 



Many experiments have been made with the Rotifer Hydatina 

 senta, but the results are conflicting. There is a striking sex 

 dimorphism, the males being small and gutless. The females 

 are from birth either male-producers or female-producers ; and, 

 according to Maupas and Nussbaum, this is determined before 

 birth, while the female embryo is still within its mother's uterus, 

 by conditions of temperature and nutrition. Well-fed mothers 

 produce females which produce females only ; starved mothers 

 produce females which produce males only. According to 

 Punnett's researches, however, changes of temperature and 

 nutrition have no effect ; but some stocks give rise to many 

 male-producing females, others to few or none. 



Against the theory of environmental influence are Stras- 

 burger's numerous experiments on dioecious Phanerogams, such 

 as Mercurialis perennis, spinach and hemp. He found that 

 changes in illumination, soil, crowding, and so on, had no effect 

 in altering the proportions of male and female offspring. He 

 is of opinion that in such cases the sex is fixed by the time the 

 seed is formed. 



As regards the fifth theory, then, we find (a) that in certain 

 cases there is some evidence that the nurture of the parents may 

 influence the proportions of the male-producing and female- 

 producing germ-cells, affecting either the number formed or the 

 number that survive, and (b) that in other cases there is no 

 hint of any such influence, the facts pointing rather to the view 

 that the sex of the future offspring is not only predestined but 

 predetermined at a very early stage in the germ-cells. 



With the facts as they are at present before us, it seems 

 impossible to give any one answer to the question under dis- 



