Gates,—Pollen Formation in Oenothera gig as. 917 
same wall, through which no extrusion takes place (Fig. 13). After extrusion 
has taken place, the nucleus moves back again towards the centre of the cell, 
and retains this position in the later stages of its development. In Fig. 12 
the extrusion has evidently taken place, as shown by the dark-staining areas 
in the contiguous region of the adjacent cell, and the nuclei have begun to 
move back towards the centres of their cells. In this particular case it will 
be observed that two nuclei have extruded into the same mother-cell. The 
nuclear membrane is usually definite, sharp, and conspicuous at this time, 
but sometimes, as in one of the nuclei in Fig. 1 a, portions of it may be quite 
indefinite. 
As already mentioned, in the cases observed (and the same process has 
been seen in certain anthers of several different flowers) the extrusion takes 
place simultaneously from all the mother-cells of a given anther, and so far 
as observed is always in the same direction in all the cells, indicating that 
the nuclear movement and extrusion are perhaps to bring about an equali¬ 
zation of pressure. It is very probable that the presence of these conspicu¬ 
ous openings or connexions between mother-cells allows of an equalization 
of pressure in the mother-cells from one end of an anther to the other. The 
movement of the nuclei and the nuclear extrusion may occur in connexion 
with this pressure-equalization, yet the latter seems scarcely adequate to 
account for it. Not infrequently the nucleus, instead of approaching the end 
wall of a mother-cell, moves to a side wall next the tapetum, which at this 
time is in direct contact with the row of mother-cells. In such cases no 
extrusion has ever been seen to take place, probably because there are no 
openings through which the extrusion could occur. 
We may now trace the history of the extruded material. As shown 
in Figs. 8-10, extrusion of a portion of the chromatin material forming 
the spireme of synizesis takes place through the cytoplasmic connexions, 
just as a viscous fluid might, under pressure, flow through an opening or 
a tube, replacing the cytoplasm present. This material, which is forced 
through to the adjoining cell, first accumulates as a dense structureless mass 
or body. It is well known that, in the telophase of a nuclear division, when 
karyolymph begins to accumulate about the daughter group of chromosomes, 
a nuclear membrane is precipitated where this karyolymph comes in contact 
with the cytoplasm. In the same manner, an equally definite membrane is 
formed where the liquid accumulated about this extruded chromatin comes 
in contact with the cytoplasm. The result is a cavity containing chromatin 
and surrounded by a definite membrane. We may very well call this a 
pseudo-nucleus. The liquid contained in this cavity is probably the same 
as the karyolymph of the nucleus, and this liquid is either secreted by the 
extruded chromatin or at any rate produced in connexion with its activity. 
The chromatin in this pseudo-nucleus is at first in a compact mass, but this 
soon becomes looser in structure (Fig. 11, to the right), and finally (Fig. 11, 
