616 Guo. W. TANNREUTHER, 
males or females, i. e, males and females are not produced by the 
same presexual parent. We can not consider the individuals of 
the fifth parthenogenetic generation as sex hybrids with either male 
or female characters dominant. If either were true, it would be 
difficult to account for the production of both sexes under the same 
conditions. Nor can we consider the parthenogenetic forms (CASTLE, 
1903) as sex hybrids in which there is a uniform dominance of the 
female character. Were this the case, it would be equally difficult 
to account fo the appearance of males, as the parthenogenetic eggs 
which produce the males have but one polar body formed and the 
female character which dominates would not be eliminated. 
The view is presented by (Stevens, 1905) that in partheno- 
genetic eggs which undergo no reduction, dominance of sex character 
may be reversed by external conditions. This view can not be 
applied in the case of the aphids I have studied. If it were true 
we would not have the appearance of both male and female under 
like conditions at the same season or time of the year, as occurs in 
the species studied. 
There are no differences in the size and structure of the ova 
which produce males and females. The number and form of chromo- 
somes are alike in both instances. We could hardly conceive that 
the sex determinant is found in any particular chromosome. If it is, 
its behavior is the same in the parthenogenetic eggs that produce 
either male or female. It is possible that the distribution or division 
of the chromosomes in the formation of the polar body, which may 
be qualitatively different, plays the role of sex determinant. Just 
what the determining factor is, as to whether a parthenogenetic egg 
shall produce a male or female under the same external conditions 
can only be conjectured, but the weight of evidence seems so indicate 
that the determination of sex is due to a structural difference in 
the chromosomes. 
As stated above, the chromosomes are constant in size and 
number, but vary in the number and size of chromomeres for each 
chromosome. Again it seems possible that the chromosomes may 
perform a definite and special function in sex production without 
being in themselves qualitatively or quantitatively different, except 
in the degree of their special activity. This later view is suggested 
very forcibly from the fact that the completely formed chromosomes 
are alike in number and size in the male and female producing ova. 
