EEVIEW OF BIOLOGY OF SEX-DETERMINATION 
9 
whereas parthenogenesis may result either in females or males, 
the latter coming from smaller eggs. This is all the more inter- 
esting because both Whitney and Shull, working on rotifers, have 
by external conditions been able to alter the normal cycle and to 
cause the discontinuance of parthenogenesis, thus bringing on 
the production of male individuals and eggs which require fertili- 
zation. They have evidently through these conditions influenced 
the type of egg produced. The type of egg would then control 
its own maturation if the case were parallel to that ot Aphis 
and Phylloxera. 
In the Cladocerans, where parthenogenesis alternates with a 
sexual cycle and at least three kinds of eggs are produced (thick 
shelled, fat laden, ephippial eggs which must be fertilized; thin 
shelled, glycogen laden, parthenogenetic eggs, developing into 
females; and thin shelled, smaller parthenogenetic eggs, develop- 
ing into males), the type of egg produced may be influenced by 
temperature and food, as was shown by Geoffrey Smith. It is 
not improbable that we may yet see in the maturation of these ova 
differences in chromosomal behavior correlated with each type 
of ovum and the sex of the resulting offspring. 
Closely allied to the foregoing problem is the question of sex 
determination in the Hymenoptera. Even before chromosomes 
were known, Dzierzon postulated that m.ales (drones) were formed 
from unfertilized eggs and females (workers and queens) from 
fertilized eggs. This has been substantiated nicely, both from 
cytological and genetical investigation. Newell showed in bee 
matings of Italian (grey) queens and German (dark) drones, and 
also in reciprocal crosses, that the male offspring of such matings 
were purely maternal while the females were hybrid in character. 
Cytological observations by Petrunkevitch and Nachtsheim have 
also established the validity of the Dzierzon theory. Coupled 
with this, observations on the spermatogenesis of Hymenoptera 
have revealed interesting results. The spermatogonia possess 
* merely the haploid number of chromosomes. In order that this 
number be not further reduced in the process of spermatogenesis 
only one division of the chromatin takes place. In the bee the 
first spermatocyte division results in all the chromatin passing 
