Gates.—Pollen Formation in Oenothera gigas. 931 
fourteen chromosomes is a distinct advantage in the study of mitosis in 
nuclei, which are so small in comparison with other seed plants. Two points 
have been definitely determined from my study of these somatic divisions 
in O. gigas. (1) In the prophase the spireme segments into a single chain 
of twenty-eight chromosomes, all of which were arranged end to end in 
a single series. The spireme is usually wound regularly just within the 
periphery of the nucleus. (2) Soon after this segmentation, while still in 
the prophase condition, the chromosomes show, in many cases at least, 
a clear longitudinal split. Thus the split in the chromosomes occurs long 
before the equatorial plate stage. 
Discussion. 
Several special features in connexion with the data in this paper call 
for further discussion. Phenomena of chromatin extrusion, evidently very 
similar to those described here for Oenothera , have been described by Miss 
Digby (’ 09 ) in the pollen mother-cells of Galtonia . The chief difference 
exhibited by these processes in the two cases is that in Galtonia the nucleolus 
as well as the spireme extrudes bodies, while in Oenothera the nucleolus 
apparently takes no part in the process. In Oenothera the extrusion takes 
place, usually at least, through cytoplasmic connexions already present 
between mother-cells, the viscous chromatic material extruding through these 
pores and forming a solid mass in the other cell. 
This may have an important bearing on our present theories of nuclear 
structure, for the exuded or extruded material quickly forms a pseudo¬ 
nucleus by the accumulation of karyolymph and the precipitation of a mem¬ 
brane where this comes in contact with the cytoplasm. Furthermore, the 
extruded chromatin, though it at first forms a body as solid as a nucleolus, 
soon spreads out and forms a structure closely resembling a reticulum or 
spireme. It seems probable from these observations that far too much 
objective morphological significance has been attached to various reticulum 
and spireme stages of the nucleus. It is evident that the greatest care must 
be exercised in interpreting spireme and reticulum stages as structures 
having a definite morphological meaning. Rather would it seem that these 
appearances depend upon the condition of physical aggregation of the 
chromatin material. For example, the spireme condition may be assumed 
as the result of the electrical charges held by the chromatin molecules. 
While cytological observation is of course very important as indicating 
different physiological conditions and stages of development in the nucleus, 
yet the chromosomes are not necessarily to be regarded as a congeries of 
definite bodies or particles which persist and multiply individually from stage 
to stage. This is an important modification of the ordinary method of cyto¬ 
logical interpretation, for it assumes that we already see with our higher 
