A study of the Chromosomes in three species of Pseudococcus. 
55 
It is only fair to say that the characteristics of this process are less 
marked than in spermatocyte cells, but they are nevertheless immista- 
kable (fig. 52). In tbe female all of the ten chromosomes evolve simul- 
taneously. 
As already implied, whether this sex differentiation of somatic cells 
is present from the very beginning of development er not, must be left 
undecided. Twentyfour larvae or embryos in the gastrula stage showed 
no clumping of chromatic material, but such evidence may indicate only 
that no males were encountered in this number^). Since the sex can not 
be determined by the gonads at this early stage, this possibility must 
be left open. 
Spermatogonia and Oogonia. 
Arrangement of chromosomes into two groups is not as marked in 
the spermatogonia as in the spermatocytes. In P. citri there is at most 
only a trace of clumping, and shortly before the chromosomes become 
arranged in a plate, even this has disappeared (fig. 41). In P. nipae 
however, the tendency of part of the. chromatin to clump is more marked 
than in P. citri, a feature to be expected from the comparison of develop- 
ments in other cells of the two species. But even in this case, the clumping 
is a much more definite character in spermatocyte cells (fig. 42 and 43). 
The number of chromosomes is always ten. 
This same number can be counted without difficulty in follicle and 
oogonial cells (fig. 40). In these cells of the female as well as in those 
already discussed, no clumping of the chromatin is ever in evidence. It 
can be regarded as established that no cell of the female shows that 
feature. 
Conclusion. 
I have had occasion to point out in a previous paper (’21) that 
there is a marked similarity between the five chromosomes of Pseudo- 
coccus which show a tendency to remain in dose association and are 
the first to condense, and the ordinary heterochromosome. It has be- 
come more and more evident that such a chromosome is not struc- 
turally different from the ordinary autosome and in the great majority 
of cases any exceptional behavior on its part can be attributed to the 
fact that it has no mate. A case in point is the monoseme of Locusta 
^) Later investigations have indicated that clumping does occur at the gas» 
trula stage. 
