562 L. DONCASTER. 
intervals, so that their age is known with some exactness. 
They have, however, some disadvantages; the eggs are not 
readily penetrated by preserving agents, and the great 
amount of yolk, and the smallness of the mitotic figures, 
cause considerable technical difficulties. 
According to Castle’s hypothesis every germ-cell is either 
male-bearing or female-bearing, and when fertilisation takes 
place a male spermatozoon meets a female egg, or vice 
versa, and the sex is determined by the dominance of one 
or the other sex. In the bee he supposes that all the eggs 
are male-bearing, and hence, when unfertilised, give rise to 
drones, but the queen bee was derived from a fertilised egg, 
and hence contained both sexes. The separation of the sexes 
is supposed to take place at the second maturation division of 
the egg or spermatozoon, and, since all the eggs of the bee 
are male, the second polar body must always be female- 
bearing, since it is at this division that the separation of the 
male from the female character takes place.1 Petrunkewitsch 
finds that the second polar body, with one of the two nuclei 
formed by the first, gives origin to the genital cells of the 
drone, and Castle, therefore, supposes that the spermatozoa 
of the drone, descended from these polar bodies, are all 
female bearing, and that the female character is dominant, so 
that the fertilised egg always produces a female. 
If the hypothesis here outlined were correct it would give 
a satisfactory solution of that most difficult of problems, the 
determination of sex, and it therefore seems of great im- 
portance to test it as thoroughly as possible. In animals 
which resemble the bee in having two polar bodies and pro- 
ducing males from parthenogenetic eggs we should expect to 
find the same course of events as described by Petrunkewitsch, 
1 The view that the separation of the sexes may take place at a nuclear 
division is supported by the case of Sagitta where tlie ovaries and testes are 
derived from the halves of one original cell; and by ‘’rypanosoma, where 
tbe zygote nucleus divides into two, and when one of these develops the 
organism becomes female, when the other, male. See Schaudin’s paper 
(‘ Arbeiten a. d. kais. Gesundheitsamt,’ Berlin, vol. xx, 1904, p. 387). In 
neither of these cases, however, is the division a “ reducing ” one. 
