364 BOTANICAL GAZETTE [DECEMBER 
the process of fixation. Of the nature and function of these 
granules, which in staining properties resemble nucleoli, little 
was learned except that they are not starch. They are not 
peculiar to any phase of cell division, though rather more 
numerous in the resting mother cell, or the prophase. They 
cannot be extra-nuclear nucleoli, for there is no increase or dimi- 
nution in their size and number contemporaneous with the dis- 
solution or renewal of the nucleoli. Moreover, the nuclear 
membrane is yet unbroken, when, in the later prophase of divi- 
sion, the nucleoli disappear from view, and therefore, if the 
nucleolar substance passes out into the cytoplasm at this time, 
it must do so by osmosis, and any relation between it and the 
extra-nuclear granules must be conjectural. To the inquiry if 
they might not be the source, at least in part, of the material of 
the achromatic spindle, no satisfactory answer can be given; it 
may be so, but positive evidence is lacking. 
The mother cells enlarge for about two weeks after they 
have reached their full number, gradually separating from one 
another and becoming more and more rounded in outline. The 
nuclei also grow larger and pass first into a spirem condition 10 
which the chromatic ribbon is not a single thread but much 
branched and anastomosed, and later into synapsis. The details 
of these changes need not be considered in this paper. 
Meanwhile, the cytoplasm has retained its reticulate arn 
But towards the end of synapsis a change is discernible, an 
this change is the first indication of the future spindle. It 
begins as an aggregation of material immediately about boas 
nucleus, causing this region to stain more deeply with gentian- 
violet. The structure of this accumulation of cytoplasm, a 
it is first recognizable, is so delicate as absolutely to baffle rs 
powers of the microscope; nothing definite can be said << 
regard to it, except that it appears to be very finely granu es 
especially in preparations fixed in Flemming’s solution. eet 
similar layer about the nucleus of the pollen mother pes 
Cobaea scandens, Lawson (8) proposed to give the name a 
karyoplasm, in virtue of its position. Since Strasburger § 
