74 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
leading up to the division represented in figure 28. It is of course 
next to impossible to determine just what happens. But it seems to 
me that there is first a conjugation of chromosomes (pi. 13, fig. 25), 
then a re-formation of mixed chromosomes (pi. 13, fig. 26), then a 
pulling apart of those new chromosomes into double the original 
number (pi. 13, fig. 27). A reunion of different halves is then brought 
about in the formation of the spindle, thus restoring the original 
number of twenty-eight chromosomes. I have seen nothing in these 
stages suggesting a pseudoreduction. It may have something to do 
with the comparatively long period of rest and growth which follows. 
The efficiency of the chromatin in metabolism of the cell may be 
enhanced by this new equilibrium of its chromosomes brought about 
by rearrangement of their constituent units. 
P seudomitosome and abnormal spermatogones .— At the close of 
this synaptic mitosis, the spindle, or intermediate fibers, disappears 
more slowly than usual, and consequently gives rise to a body corre¬ 
sponding to the “nebenkern” of the spermatid. This pseudoneben- 
kern gradually melts away at the periphery causing a large spherical 
vacuole extending partly if not completely around the nucleus (pi. 13, 
fig. 32). It is about the size of the original nucleus just before the 
formation of the spindle (pi. 13, fig. 27). At one pole of this vacuole 
lies the compact spherical nucleus (pi. 13, fig. 32, c) apparently without 
nuclear membrane, and evidently with little or no karyolymph. The 
original spindle substance, mitosome, gradually diminishes till nothing 
remains but a small spherical body in appearance not unlike an attrac¬ 
tion sphere (pi. 13, fig. 32, b). This occasionally has one or two 
central bodies suggesting centrosomes (pi. 13, fig. 32, a). 
As this occurs in cysts which I have to interpret as gonocysts and 
in cells of the size of spermatogones in synapsis, I have to select one 
out of two possible conclusions. Either there is here an abortive 
attempt of the spermatogones to form spermatids, thus omitting the 
spermatocyte stage, and hence the usual maturation divisions, or else 
it is a pathological condition of the cells, a fact which is suggested by 
the peculiar position and behavior of the nucleus. 
It is of course possible that the nucleus (pi. 13, fig. 32, c ) may ex¬ 
pand till it ultimately fills the vacuole, thus restoring the cell to its 
original resting condition. In that case the centrosome and sphere 
remaining after the pseudomitosome is absorbed, would lie close to 
the nuclear membrane between the nucleus and cytoplasm,— a posi- 
