9i5 
Gates.—*Pollen Formation in Oenothera gig as. 
A study of many nuclei in the stage represented by Figs. 6 and 7 
makes it evident that a certain amount of collapse together of the thread- 
work does take place. This is shown by the fact that the synaptic spireme 
not only occupies less space than the presynaptic reticulum, but has a 
greater density, showing that the material is more closely aggregated. 
That the spireme occupies less space may perhaps be partly a result of 
osmotic effects during the preparation of the slides. It is probable, how¬ 
ever, that it is not to be accounted for as the result of contraction on the 
part of the threadwork. The threads of the presynaptic nuclear reticulum 
are anchored to the nuclear membrane, and this perhaps serves to keep 
the threadwork taut and occupying the whole of the nucleus. This is 
clearly shown by the fact that, when the nuclear expansion without chro¬ 
matin growth begins, some of the threads frequently remain attached for 
some time to the membrane, and are pulled outwards with it as the pressure 
from within distends the membrane. Finally, when the nucleus has reached 
its definitive size, only a few such threads remain and the threadwork floats 
freely in the nuclear cavity. The expansion of the nucleus is completed 
before the rearrangement of its contents from the reticulum to the spireme 
condition takes place. It is probable that during this rearrangement a 
certain amount of collapse together of the threads, very few of which have 
any attachment to the nuclear membrane, takes place, so that the spireme, 
which is always placed unilaterally in the nucleus, occupies somewhat less 
space than did the reticulum. But while the spireme occupies usually 
somewhat less space than the original reticulum, yet this by no means 
accounts for all the empty space in the nucleus. This empty space is 
produced by a rather sudden expansion of the nucleus unaccompanied by 
any change in its reticulum. The transformation of the nuclear chromatin 
content from the reticulum to the spireme condition occurs apparently 
only after this nuclear growth is completed. 
Nuclear Extrusions. 
During the whole growth period of the pollen mother-cells, and the 
early synaptic stages (represented by Figs. 1-7), the nucleus occupies 
a central position in the pollen mother-cell. Then, while in the typical con¬ 
dition of synapsis or synizesis, the nucleus moves to one side of the cell until 
the nuclear membrane is in direct contact with the cell-wall. This condition 
has frequently been figured by previous investigators, e. g. in the papers 
of Overton (’ 09 , PI. Ill, Pigs. 4-6) and Schaffner (’ 09 , PI. XII, Fig. 10), 
but no special significance has been attached to it. It is probable that this 
lateral movement of the nucleus always takes place, after its growth is 
completed and the transformation of its contents from a reticulum to the 
spireme of synizesis has occurred. This movement seems to be concerned 
