Cell Structure, Growth and Division in the Antheridia of Polytrichum etc. 151 
This body stains deep black in iron-alum-haematoxylin, retaining 
the stain after the rest of the cytoplasm, including the other dark-staining 
grantdes already mentioned, has been nearly or quite decolorized; the 
rays of the aster also appear somewhat darker than the rest of the cyto- 
plasm. In the most successful triple-stained preparations, the central 
body is dark red or purplish red, the rays are blue or violet. 
So far as my preparations show, the central body is a new structure, 
appearing for the first time in the cytoplasm of the androcyte mother 
cell. It may be imagined to have been really present among the similarly 
stained kinetosomes of earlier cell generations; or one may suppose that 
the progressive diminution of the visible kinoplasm by successive cell 
divisions has left in each cell only this single granule, which, as will appear, 
behaves with reference to spindle formation just as did the groups of 
kinetosomes, or the polar plates in earlier divisions. But I have found 
no direct evidence to Support either of these possibilities; and they are 
in a measure negatived by the occurrence of an astral figure at this time 
and its absence in Connection with any of the structures observed at 
earlier periods. I have sought carefully, also, for evidence of a nuclear 
origin of the central body, such as is maintained for granules of somewhat 
similar appearance and behavior in Marchantia by Ikeno (1903), and in 
mosses by the Leeuwen-Reijnvaans (1907 b, 1908); but on this point 
my results are entirely negative. 
Many antheridia are found, all of whose inferior cells have the charac- 
teristics that have now been described — including the possession of a 
central body. Cells so characterized have also been found in antheridia, 
other cells in which were in stages preparatory to the next ensuing division ; 
but they have never been seen in Company with cells which were under- 
going the preceding division or were in stages preliminary thereto. Since 
a single antheridium commonly contains representatives of a series of 
stages, it follows that there is in all probability a relatively considerable 
lapse of time between the end of the last androgonial division and the 
initiation of the division of the androcyte mother cells. Düring this time, 
as is evidenced by the appearances already described, the mother cells, 
and consequently the whole antheridium, grow considerably; the partition 
walls between the mother cells begin to soften and swell, preparatory to 
their ultimate dissolution, and the cells begin to draw away from the walls 
and round up; such kinetosomes as may have been inherited from the 
previous cell generations either disappear or become so modified as to be 
unrecognizable; and a central body with its accompanying aster appears 
in the cytoplasm of each cell. 
