Cell Structuie, Growth and Division in the Antheridia of Polytrichum etc. 135 
an antheridium of eight rows; and Figure 2 is from an antheridium with 
nine rows of androgones. On the other hand, Figures 11, Plate VI, 
and 54, Plate VII, showing cells with numerous kinetosomes, are from 
antheridia each of which contains only six rows of androgones at its 
widest part; Figure 16, Plate VI, is from an antheridium of four rows; and 
Figure 32, Plate VII, in which the kinoplasm at one pole at least is in the 
form of separate bodies, is from a young antheridium containing a single 
row of four androgones. In respect to the periocl at which the separate 
kinetosomes first appear, I have observed an almost unvaried uniformity 
among the cells of a single antheridium and a general uniformity among 
the antheridia of a single head, but considerable diversity between 
antheridia of about the same age from different heads. 
When the separate kinetosomes are distinguishable, those of a group 
often lie nearly or quite in a common plane, so as to give much the appear- 
ance, especially under moderate magnification, of a single plate (see the 
upper polar group in Fig. 31, PI. VII; the lower group in Fig. 32; also 
Figs. 20, PI. VI; 37, PI. VII; 63 and 65, PI. VIII). This arrangement is 
most common in the earlier cell generations in which kinetosomes appear. 
In many cases, however, and particularly in the later androgonial 
generations, the grouping is looser and more irregulär (e. g., Figs. 17, 
PI. VI; 38—40, 54, PI. VII). 
Individual kinetosomes differ greatly in size and shape, although 
most frequently their form approaches that of a straight or bent rod. Often 
their shape is not dissimilar to that of the chromosomes. A resemblance 
to the latter is most noticeable in haematoxylin preparations at the equa- 
torial plate stage (PI. VII, Figs. 37 — 40). It is less striking in triple-stained 
material at the same stage, because of the difference at this time between 
the staining properties of the kinoplasm and the chromatin. 
The behavior of the group of kinetosomes, as a group, is similar to 
that already described for the polar plates. Düring the telophases (PL VIII, 
Figs. 60, 63, 65), a single group is present at each spindle pole, passing 
consequently into the corresponding daughter cell (PI. VI, Fig. 11). Many 
of the kinetosomes, because of their position at the pole of the former 
spindle, lie close to the nucleus of the daughter cell; some of them com- 
monly seem to be in contact with the nuclear membrane. Soon they are 
separated into two groups, which move apart (Fig. 11, lower cell; Figs. 12 
— 15), until, as in the case of the polar plates, they lie close to opposite 
sides or ends of the cell (Figs. 16 — 18), with the nucleus about midway 
between them. Düring the migration of the daughter groups, some of 
the kinetosomes remain for a time seemingly in contact with the nuclear 
