The spermatogenesis of Hydra. 405 
the middle piece. If this droplet of nutritive material derived large- 
ly from the liquid and absorbed cell body is to be regarded as 
a greatly enlarged centrosome, as seems likely from the facts given, 
then we may justly assign to the centrosome the function of lique- 
faction and assimilation of the cell body. The centrosome then 
becomes a digestive ferment. This can only be a suggestion render- 
ed possible in Hydra by what we know of the centrosome in 
animals where it can be plainly followed. 
The acrosome seems to be derived from no special body as in 
many higher forms for no Nebenkern or idiosome has been found in 
the spermatid. It arises apparently from that portion of the cell 
protoplasm which adheres to the exterior tip of the elongating 
nucleus. This spermatozoon is small. From tip of head to base of 
tail it measures 42 u. Of this the acrosome is 0.2 «, the head proper 
3.2 u, and the middle piece 0.8 u. The diameter of the head at its 
widest point is 1.2 u, of the middle piece 1.3 w. 
Summary of the phenomena of mitosis. 
It will not be amiss to recapitulate the phenomena of cell 
division, classified under several headings. 
The centrosome. This structure is followed with difficulty. 
It is apparently absent in the spermatogonia until just prior to the 
equatorial plate stage and disappears usually shortly after division. 
This is true also for the spermatocytes. Occasionally howewer it 
persists until the early prophase of the succeeding division (Fig. 49). 
That its disappearance is apparent rather than real seems likely, 
for it is occasionally found dividing before its disappearance (Fig. 20). 
And then too a central spindle is not infrequently seen (Fig. 57 
and 58). It is probable then that during or after cell division of 
the centrosome occurs, the new daughter centrosomes separating 
and moving each 90 degrees to the new poles of the daughter cells. 
In the spermatid then it remains (Fig. 61) at the pole, as no new 
daughter cells are to be formed, and is there incorporated in the 
middle piece. 
Occasionally some detail could be made out for the centro- 
some, when it appears as a small sphere with a more deeply stain- 
ing central granule (Fig. 40). 
The mitotic figure. This is extremely simple as asters 
are wanting. It was noticed by Braver in the formation of 
the polar bodies that the polar spindle is barrel-shaped and 
