MUNSON: SPERMATOGENESIS OF PAPILIO. 
95 
tail pole of it (pi. 17, figs. 134, 135), or close to the nucleus (pi. 17, 
fig. 143). At the point where this axial filament is attached to the 
nucleus, there is frequently a stainable body, which is easily mistaken 
for a centrosome. But as it stains like chromatin and is without 
doubt inside the nuclear membrane, I look upon it as an aggrega¬ 
tion of chromatin, caused by the convergence at that point of the 
fibrils which together constitute the axial filament. 
That the axial filament is composed of several threads seems sup¬ 
ported also by the fact that the constituent fibrils sometimes separate 
at various points along their course through the tail of the spermato¬ 
zoon (pi. 17, figs. 140, 141). 
It is not at all improbable that the cytoreticulum is continuous with 
the nuclear reticulum in other cells, and it ought not to be considered 
improbable, a 'priori, that the axial filament may have a similar con¬ 
nection with the nucleus. 
In my work on the ovarian egg of Limulus (’ 98 ), I removed the 
germinal vesicle from the living egg and found that thread-like proc¬ 
esses adhered to the nuclear membrane as if the broken cytoplasmic 
fibers of the reticulum were really continuous with the linin of the 
nuclear network. Other appearances in the sectioned material of 
those oocytes, led me to the same conclusion. The phenomena of 
mitosis, too, support that view. 
In my work on Clemmys (’ 04 ) I have described a division of the 
cytoplasm into a central portion and a peripheral portion. These 
two zones, which I also found in the oocyte of Limulus, I believe are 
present in most cells. I have called the line separating them, the cyto- 
coel. 
The cytocoel in the spermatid is the line between the cytoplasm 
and the nebenkern. At first the material of the nebenkern fills the 
inner space completely. But I feel convinced that if the material of 
the nebenkern could be removed entirely, there would remain some 
slight trace of a reticulum in the vacuole thus formed. This is sup¬ 
ported by the fact that the material of the nebenkern, in numerous 
instances, does not fill the space entirely (pi. 17, fig. 125). In those 
cases traces of a reticulum within the nebenkern vacuole can be seen. 
The fibrils resemble in appearance those which constitute the axial 
filament. 
My conclusion is that the axial filament of the spermatozoon repre¬ 
sents the greatly compressed cytoreticulum of that portion of the cell 
