M. Koidzumi 
297 
centre in mitosis. I also failed to find any structures suck as centrosomes at 
the ends of the strand, and I agree with the above view. As for the origin of 
the strand, I think it is probably derived from the wall of the axial core. 
In Teratonympha and Pseudotrichonympha the condition is somewhat differ¬ 
ent. Here the axial core does not function as a division-centre, but a special 
structure is formed for the purpose. In the early stages of division of Terato- 
nympha a strand is seen, quite independent from the nucleus, provided with 
a “central body” at both its ends, each of these being connected with the 
base of the daughter nipples by means of a thread. As regards the origin of 
these structures, I think they are derived from the axial core. Thus I believe 
that the axial core of both Teratonympha and Trichonympha is the source of 
the structures of extranuclear origin in mitosis; but these organisms differ 
in one point, namely, in Teratonympha the axial core gives rise to the body 
functioning as a centrosome, whilst in Trichonympha it acts itself as a division- 
centre. In Pseudotrichonympha the condition is quite similar to that of 
Teratonympha : namely, the spherical body and the thread connecting it 
with the nipple are derived from the tubular column, and probably, I think, . 
from its wall, whilst the strand arises from the spherical body. 
The homology of the two layers at the anterior region of Trichonympha, 
Teratonympha, and Pseudotrichonympha is readily recognizable; the differences 
being simply in degree of development. The axial core in Trichonympha is 
surrounded concentrically by the inner and outer layers, the basal granules 
of the flagella being localized at the basal or proximal surface of the outer 
layer. In Pseudotrichonympha almost the entire surface of the axial core is 
directly surrounded by the outer layer, and only a small portion of its posterior 
part is in contact with the inner layer; consequently the basal granules, in 
this region, are localized immediately on the surface of the axial core. In 
the bell, the distribution of the layers is identical in both genera. In Terato¬ 
nympha the basal granules of the flagella are found at the base of the outer 
layer, just as in the other genera, and the conical axial core is in contact with 
the inner layer (tubular rind) at its anterior end. As they are partly separated 
by the layer of peculiarly differentiated protoplasm, the axial core and the inner 
layer of Teratonympha seem not so intimately connected as in the other genera. 
But it is not only at the anterior end that these two parts are connected: 
there is also the vertical partition crossing the space between them, and 
keeping them in close connexion. Thus the inner layer seems to serve as a 
means of connexion between the locomotory organs and the part of the body 
which plays the role of the division-centre in mitosis. 
In Microspironympha the anterior portion of the body is not so highly 
differentiated as in the above three genera; but there is a peculiar tubular 
structure situated in the axis of the body, connecting the tip of the body 
and the anterior surface of the nucleus. It is easily imaginable that this is 
a structure corresponding to the axial core in Trichonympha. Moreover, it 
seems not unreasonable to interpret the tubular wall and its contents as 
