go BOTANICAL GAZETTE [aucusT 
periphery and the spindle there is a thin layer of kinoplasmic ~ 
material, looking as if it were made up of small portions of fibers 
in a tangled mass (fig. 23). A similar appearance was described | 
by Mottier (of. cit., p. 180) in an earlier stage of spindle forma 
tion in the pollen mother cell of the lily. He observed, how 
ever, that this layer forms a part of the spindle, z.¢., it is the same 
as the felted layer described by Strasburger and Belajeff. That 
it forms no part of the existing spindle in the larch, howevel, — 
would be indicated by its late appearance after the spindle 8 ~ 
formed. It may be that there is in the larch more kinoplasm — 
formed than is needed for this stage in the development of the | 
spindle, and that it becomes absorbed into the cell protoplasm 
or is used directly for the later stages in the growth of the fibers 
(p. 47.) Between this kinoplasmic layer and the spindle there 
often appears a thick layer of finely granular trophoplasm which 
stains readily with the orange. A similar layer in the pollet | 
mother cells of Hermerocallis was described by Juel, but he | 
he did not mention any strong affinity for the orange stain, such 
as I have observed in the larch. In some of my own prepare | 
tions of Hermerocallis, I have observed the orange-stained layet F 
very frequently. ie 
Throughout the cytoplasm, and in some cases in the spindles 
also, there often appeared, in this as well as in later stages 
large blue-stained granules. Their distribution was irregulat: | 
They were more apparent usually in the larch than in the oni” 
ort 
mentioned and not as secondary structures formed simp 
between the groups of chromosomes. They show in the re 
other hand, the portions between the receding groups of daug 
