Cell Structure, Growth and Division in the Antheridia of Polytrichum etc. 137 
generally less, even relatively to the size of the cell, than in those in 
which plates are present (compare Figs. 16 — 18 with Figs. 8 — 10, PI. VI; 
Figs. 38 — 40 with Figs. 34 — 36, PI. VII). This difference is less apparent 
during the telophases, at which time large numbers of fibers always 
appear in Connection with cell plate formation. 
Apart from the comparative paucity of fibers, the developing spindle 
figure in the cells now in question is similar to that observed in connection 
with the polar plates. The spindle is broad-poled (PI. VI, Figs. 16 — 18), 
the fibers being attached usually to the various kinetosomes. Karely 
(PI. VI, Fig. 17 ; PI. VII, Fig. 37), some of the fibers can be traced past 
the polar structures to the plasma membrane. 
As we have now seen, the spindle itself is outlined, and a considerable 
part of its fibrous material is present, before any visible changes preparatory 
to division occur within the nucleus. The further history of the spindle 
is substantially the same in all cases, whatever the form of the structures 
at its poles. While the chromatin is becoming arranged into a spirem 
(PI. VI, Figs. 19 — 21), the nucleus increases in size, chiefly though not ex- 
clusively in the direction of the long axis of the spindle, until the nuclear 
membrane comes in contact with the polar structures (Figs. 22 — 24). 
The nucleus now takes up practically the space that was formerly occupied 
by the spindle ; the fibers of the latter, most of them now curved outward 
and closely appressed against the nuclear membrane, are no longer con- 
spicuous, though a few of them can usually be distinguished (Figs. 22, 
23). There is no evidence at any stage that any of the spindle fibers are 
formed in whole or in part within the nucleus. 
After the nuclear membrane disappears, the chromosomes become 
compacted into a tight knot (PI. VII, Fig. 30); the spindle now appears 
much as it did before the swelling of the nucleus (compare Fig. 30 with 
Figs. 10 and 19). Immediately surrounding the knot of chromosomes 
there is still a clear region, comparatively free from fibers. Soon there 
occurs a rapid increase in the amount of fibrous material (Figs. 31, 32), 
involving an increase in number of fibers as well as, in all probability, 
a growth in length of some of those already present. The clear region 
about the chromosomes disappears, and the whole spindle area is now 
quite uniformly filled with approximately parallel fibers. As the chromo- 
somes become disentangled from the knot (Fig. 34) and arranged in the 
equatorial plate (Fig. 35), a further rearrangement of the fibers results 
in the appearance shown in Figures 35 — 37. The mature spindle, like 
the spindles found in dividing cells of seed plants (Timberlake, 1900), 
