Gates.'—Pollen Formation in Oenothera gigas. 919 
observed cytoplasmic connexions. The nucleolus has never been observed 
to be in the ‘ sickle-stage ’ at this time, but is always spherical and floating 
freely in contact with the spireme. It takes no part in this extrusion. 
It is probable that this lateral movement of the nucleus in the mother¬ 
cell always occurs, whether it be followed by an actual extrusion of chroma¬ 
tin or not. The conditions of pressure within the cells, and the presence 
of cytoplasmic connexions at the point of contact, probably determine 
whether such extrusion will occur. 
It may be stated that the same process of chromatin extrusion occurs 
also in O. biennis , although Davis (TO), in his recent paper on this form, 
failed to observe it. In December, 1908, I showed (Gates, ’09 c) that the 
phenomena of reduction are just the same in O. biennis and O. laevifolia as 
in O. Lamarckicina and its mutants. A recent re-examination of some of 
the preparations made at that time furnishes certain and conclusive evidence 
that the same process of chromatin extrusion occurs in O. biennis. Dark 
areas of cytoplasm had been noticed, but their significance was not under¬ 
stood, and it was supposed that they were connected with degeneration. 
They can now be clearly interpreted in the light of my observations on 
O. gigas , in which particularly favourable sections showing the extrusion 
were obtained. A few cases showing the actual extrusion taking place were 
also observed in O. biennis. This process, therefore, doubtless occurs in all 
the forms, and this is a further point showing the identity of the reduction 
processes in O. biennis and O. Lamarckiana. 
Later Synaptic Stages. 
After the nucleus moves back to the centre of the mother-cell, it 
retains this central position until the nuclear membrane disappears after 
diakinesis. The synizesis condition soon changes to one such as is to be 
seen in Fig. 14, in which the thread work now occupies the whole of the 
nucleus once more, but the threads are beginning to show the progressive 
shortening and thickening which characterizes in part the succeeding stages 
of the nucleus. All the nuclei in one anther were in this particular con¬ 
dition, while other anthers of the same flower were in typical synizesis with 
the nucleus still central in position, and yet other anthers showed the begin¬ 
ning of the nuclear expansion which leads to the synizetic condition, as in 
Figs. 2-4. As shown by Fig. 14, the threadwork of the nucleus is fairly 
coarse, and the larger threads are characteristically straight and undeviating, 
quite unlike their usual meandering course. It is evident that a complete 
rearrangement of the contents of the nucleus must have taken place between 
this stage and the synapsis of Figs. 6 and 7. The interpretation of this stage 
will be referred to again later. 
