Cell Structure, Growth and Division in the Antheridia of Polytrichum etc. 145 
ance most of the fibers of the spindle appear to end on the respective 
membranes. It wonld seem from these observations that upon the con- 
traction of the fibers the closely grouped chromosomes probably separate 
somewhat to occupy the clear space so left, and that a membrane is then 
formed in the region where the fibers end; but I have not been able to 
follow in such detail the formation of the nuclear membrane that I can 
state the precise sequence of events. Certain it is that the distance from 
the nuclear membrane, immediately after its formation, to the equator 
of the spindle is considerably less than was the distance from the ehromo- 
some groups to the equator before the appearance of the membrane (com- 
pare Figs. 60, 61, PI. VIII, with Figs. 57, 58, PL VII, and Fig. 59, PI. VIII). 
The distance between the nucleus and the equatorial plane of the spindle 
(or later between the nucleus and the cell plate or the division wall) de- 
creases still further with the continued growth of the nucleus (Figs. 63 — 65, 
PL VIII; Figs. 1, 11, PL VI); and from a study of the relative position of 
the daughter nucleus within the daughter cell during this period of growth, 
it appears that that portion of the nuclear periphery nearest the former 
spindle pole remains in approximately the same position, so that the 
growth of the nucleus is in effect mainly or entirelv toward the equatorial 
plane of the former spindle. Excepting as this method of growth would 
shift the centers of the daughter nuelei nearer to the equatorial plane, 
and consequently nearer to each other, I have seen no evidence of a move- 
ment toward each other of the daughter nuelei, such as several authors 
have found to occur at about this stage in the cells of other plants. 
At about the time that the ends of the spindle fibers begin to pull 
away from the chromosome groups (PL VIII, Fig. 59), the spindle shows 
a transverse zonation — a phenomenon first observed by Went (1887) 
and first rightly interpreted by Timberlake (1900) as due to a thickening 
of the ends of the spindle fibers while the equatorial portion of each fiber 
remains unthickened, the result being the appearance of a light equatorial 
zone separating two darker sub-nuc-lear zones. In the present case, the 
darker zones, at the earliest stage at which I have observed them, are 
separated from the respective chromosome groups by clear spaces that 
result, as already described, from the contraction of the fibers away 
from the chromosomes; consequently, the whole space between the two 
groups of chromosomes is divided into five zones, alternately light and 
dark. The appearance, as Timberlake has pointed out, is as though 
the fibers were composed of a plastic material which is beginning to flow 
from the end of each fiber toward its center in order to take part in the 
construction of the cell plate. 
Archiv f. Zellforschung. VIII. 
10 
