Cell Structure, Growth and Division in the Antheridia of Polytrichum etc. 155 
pliosis of the androcyte into an antherozoid, is the blepharoplast. The 
wall between two daughter cells is at first very thin (Figs. 123, 124). By 
raeans of the contrast between this and the older walls, which by this time, 
as already noted, are considerably thickened, it is possible in many cases 
to determine whether two adjacent androcytes are or are not sister cells, 
even after all traces of the central spindle fibers have disappeared. When 
a pair of sister cells are thus identifiable (as in Figs. 123, 124), it is plain 
in inost instances that the position of the blepharoplast in each cell coin- 
cides approximately with that of a pole of the former spindle. This posi- 
tion of the blepharoplasts, together with other evidence already presented, 
seems to place beyond reasonable doubt the persistence during mitosis 
of the central bodies of the androcyte mother cell and their identity with 
the blepharoplasts. An occasional shifting of the position of one or both 
of the central bodies during the telophases, such as is suggested byFigure 
118 (upper cell), would explain the rather rare instances in which a blepharo- 
plast is located elsewhere within the androcyte than at what was obviously 
a spindle pole. 
As we have seen, most of the fibers of the spindle rudiment are in- 
distinguishable when the swelling nucleus has reached its greatest size 
(PI. VIII, Figs. 82 — 84), because they are closely appressed against the 
nuclear membrane. Later, as the nucleus contracts (Figs. 86 — 88) and 
the membrane disappears (PI. VIII, Fig. 90; PI. IX, Figs. 93 — 96) ,the 
spindle fibers become again visible. It now appears that not all of the fibers 
running toward either pole are attached to the corresponding central 
body, even when the latter is plainly visible — the result being a more or 
less broad-poled spindle (compare, e. g., Fig. 90, PI. VIII, with Fig. 80). 
The changed Orientation of the fibers suggests that in the course of the 
nuclear swelling the resultant lateral pressure upon the fibers caused a 
breaking of their attachment to the central body. Such loss of attach- 
ment is apparent, on the part of some of the few fibers then distin- 
guishable, while the nucleus is swelling and after it has attained its 
maximum size (Figs. 81, 82, 84); but the change is naturally more no- 
ticeable when all of the fibers again come into view during and after the 
nuclear collapse. 
Some of the fibers, however, remain attached to the central body, 
as appears whenever the latter is recognizable (PI. VIII, Figs. 87, 90; 
PI. IX, Figs. 94 — 96). Figure 94 shows an extreme case, in which nearly 
all the fibers retain their former orientation; the spindle is consequently 
sharp-poled. This cell, too, shows a marked development of polar radia- 
