Cell Structure, Growth and Division in the Antheridia of Polytrichum etc. 141 
nudeus does not appear so much elongated as do the nuclei in Figures 
20 — 22. Apparently the time of disappearance of the nucleole is not quite 
the same in all cases; at least, in Figures 22 and 24 there is nothing eer- 
tainly identifiable as a nucleole, although in Figures 25 and 26, represent- 
ing, to judge from the appearance of the spirem, a later stage, such a body 
is present, large, rounded, and separate from the spirem. In the cell 
shown in Figure 26, the spirem was blue, the nucleole red; the other body, 
of irregulär shape, which appears in the nucleus was stained pale blue; 
as to its nature I am uncertain. The presence of a distinct nucleole as 
late as the stages here represented corroborates the view already expressed 
as to the distinction between nucleolar substance and chromatin; and it 
is plain that in these nuclei the disappearance of the nucleole is as wholly 
unrelated to chromosome development as the most careful investigations 
liave shown is the case in the nuclei of seed plants. 
Somewhat later (Figs. 27, 28), the spirem has segmented, and the 
segments lie close against the nuclear membrane, the interior of the nuclear 
cavity being quite free from chromatic material. The chromosomes 
are elongated in Figure 27; in Figure 28 most of them at least are much 
shorter and thicker, but their shapes are very diverse and thev do not 
yet approach very closely to their appearance at the time of the equatorial 
plate. In the nuclei here shown (Figs. 27, 28), rounded bodies are present 
which may be nucleoles; but, because of the variable shape of the chromo- 
somes and their affinity for safranin at this stage, it is also quite possible 
that the rounded bodies in question are chromosomes. 
The stages just described are quickly followed by a shrinkage or 
collapse of the nucleus (Figs. 29 a — g). I have not seen, previous to this 
collapse, a penetration of the nuclear cavity by the spindle fibers, nor a 
breaking down of the nuclear membrane on one side — phenomena com- 
monly observed at corresponding stages of mitosis in the higher plants. 
The appearance is rather as though, because of some change in the character 
of the membrane, the nuclear sap gradually diffuses outward, so that the 
membrane contracts and the nuclear cavity becomes smaller. The mem- 
brane is visible for some time during the shrinking (Figs. 29 a — c); but 
it disappears as the chromosomes become closely massed together (Figs. 
29 d — g ). As a final result of this nuclear collapse (PI. VII, Figs. 30, 31), 
the chromosomes form a dense knot of irregulär outline, whose individual 
constituents are not plainly distinguishable. Soon, however, the chromo- 
somes become disentangled from one another (Figs. 32 — 34) and spread 
out in the equatorial plane of the completed spindle (Figs. 35 — 40). It 
is not perfectly clear whether the chromosome group of Figure 32 is in 
