The nuclear coaiponents of tlie sex cells of four species of cockroaches. 507 
difficulty of finding cells where the chromosomes are sufficieutly 
separated from one another to warrant couclusions as to their number 
has been spoken of earlier in the paper, but the Statement just ma de 
concerning the invariability of the chromosome number in the sperma- 
togonia refers to cases where the counts are possible within a 
reasonable degree of certainty. 
Turning now to another point concerning the question of the 
odd chromosome of the spermatogonia, we find various accounts given 
by workers on different species of Orthoptera. Jordan (’08) States 
that "the accessory chromosome first appears as a definite charac- 
teristic nuclear structure in the resting stage of the first Order of 
secondary spermatogonia.” In the earlier series, he finds no such 
body. His description is similar to that of Stevens (’05) who 
identifies the chromatin nucleolus of the spermatogonia as the "ele- 
ment x” or odd chromosome, as Montgomery (’Ol) did in his earlier 
papers. Sutton (’02), Davis (’08) and Robertson (’08) find the odd 
chromosome occupying a distinct cyst in the nucleus of the resting 
stage between the earlier and later series of spermatogonia. There 
is no such condition in the species studied by the present writer and 
it is doubtful that the chromatin nucleolus of these stages in Peri- 
planeta is the odd chromosome, for exactly similar conditions exist 
in the oögonial prophases and the corresponding stages of the 
ovarian follicle cells. In this species, the odd chromosome appears 
for the first time in the very early growth period, immediately after 
the breaking down of the chromosomes of the telophase of the last 
series of spermatogonia. 
b) The synizesis stage. There has been much discussion as to 
whether synizesis is an artifact or not and a decided stand has been 
taken by Mc Clung (’02), Schaffner (’07), Guignard ( : 99), Janssens 
(’05), Miyake (’05), Davis (’08) and others against the view that it 
is a constanfS or normal phenomenon in the spermatogenesis. Miyake 
described a similar appearance for the chromatin of certain stages 
of the development of the central cell of Picea , but as B. M. Davis 
(’05) remarks, this probably has no fundamental relation to nuclear 
activity. Davis (’99) himself made a series of experiments on Antho- 
ceros with a number of fixing fluids but invariably synizesis occurred 
at the same time in the development of the spore and at no other 
time. H. S. Davis (’08) concludes that synizesis is an artifact, for 
he finds no such condition of the chromatin in those cells lying near 
the wall of the follicle where the fixing fluid operates first. The 
