414 Experimental Zoology 



sperm, one of them functionless, etc. It is apparent, therefore, 

 that Beard's theory resolves itself into two assumptions : first, that 

 sex is determined by the egg alone ; and secondly, that two kinds 

 of eggs are produced. The solution of the problem is merely 

 shifted to a new field, since it is assumed that the male and female 

 eggs are produced, and no explanation of how or when this occurs 

 is forthcoming. 



Beard's view encounters, moreover, special difficulties that 

 can only be explained away by further assumptions. For ex- 

 ample, in the aphids there is a long succession of female partheno- 

 genetic eggs, ending finally in the production of male and female 

 parthenogenetic eggs. It would seem that external conditions 

 must determine whether the eggs are to be all female eggs or 

 some male and some female, but there is nothing in Beard's 

 theory that indicates how such a thing is possible. 



In the case of the bee, Beard is forced into the position of as- 

 suming that the sex of the individual is not dependent on whether 

 it is or is not fertilized, although the clearest evidence that we 

 have points unmistakably in that direction. He argues that 

 only female eggs can be fertilized, while male eggs cannot be 

 fertilized; but if this were the case, it is not evident how, the 

 queen could distinguish between the two kinds of eggs and lay 

 each in its appropriate cell. 



Castle has also proposed a hypothesis of "The Heredity of 

 Sex." His hypothesis is avowedly an. attempt to account for 

 the determination of sex by means of the Mendelian method. 

 It had already been suggested by Strasburger and by Bateson 

 that the sexual forms might bear the same relation to each 

 other as do the offspring of Mendelian hybrids. Castle has 

 elaborated this idea into an ingenious hypothesis. He assumes 

 that there are two kinds of spermatozoa, male and female; 

 and two kinds of eggs, male and female. If we assume, as 

 in the case of Mendelian hybrids, that all possible chance 

 combinations occur between the germ-cells, we should expect 

 three kinds of individuals (as in the Mendelian proportion), 

 males, females, and hermaphrodites. There would be as many 



