444 
Chas. W. Metz and Jose F. Nonidez 
Tliere seems to be only one plausible interpretation of events up to 
tliis point — namely, that syiiapsis lias occurretl cluring telophase, wlien 
the chromosomes were soinewhat drawn out, but were not in what could be 
ealled a trueleptotene stage. The latter seems to have been oniitted entirely. 
A comparison of these features with those found in other groups of 
organisms has been given in our earlier paper (’21, pp. 172—174) and need 
not be repeated here. 
Later growth stages. 
ln Asilus senceus we were able to follow the individual bivalent 
chromosomes of stage h directly through the growth period and into the 
prophase of the first spermatocyte division, without the intervention of 
any confused or diffuse stage. In the present species the growth period 
is complicated by the presence of a stage in which the chromosomes become 
spun out into long (bivalent) threads (figs. 19—22) and become tangled up 
to such an extent that they cannot be traced individually (figs. 23—25). 
As the bivalent bodies lengthen out after stage h each forms two, 
more or less granulär, paired threads, usually lying in a closely approxi- 
mated position. The detailed strncture of these threads is represented 
as accurately as possible in figs. 19 and 22 where only portions of the 
nuclei are shown. In figs. 20 and 21 entire, or nearly entire, nuclei are 
represented. The double threads of this stage correspond in nuniber and 
size relations to the paired spermatogonial chromosomes and to the ag- 
gregates of stage h. As they become more and more spun out and inter- 
mingled they gradually lose their identity (figs. 23—25) through over- 
lapping. At this stage, which may be designated stage c (figs. 23, 24) the 
threads are more or less polarized and lie near the nuclear wall, usually 
on one side, forming an umbrella-like canopy with its free edges toward 
the nucleolus on the opposite side. Figs. 19 to 24 indicate the derivation 
of this strncture. Some at least of the attenuating or um'avelling threads 
assunie a definitely oriented position with respect to the nucleolus as shown 
in figs. 20 to 22. 
In fig. 24 an attempt is made to show all of the chromatin, with the 
result that only the threads lying near the upper side of the nucleus can 
be figured distinctly. Those underneath, near the lower side, are made 
pale in color for the sake of contrast, giving an erroneous impression of 
the general appearence. Under the microscope the threads on the lower 
side are just as distinct as those above. 
At first there is no indication of polarized loops; instead one end of 
a thread may point away from the nucleolus, the other toward it; but with 
