232 BOTANICAL GAZETTE [OCTOBER 
entirely. Whether they remain in the cytoplasm, or are with- 
drawn into the nucleus, or furnish part of the material for the 
nuclear membrane, are all matters of mere conjecture. The 
nucleoli begin to appear a little before the time when the cell 
wall has been completely formed. 
fig. 47 is an interesting case in that it shows the centrosome 
not yet divided in a very late stage. This body appears as a 
long, black, rod-like body forming a slender dumb-bell. The 
chromatin bands seem to be distributed again or spread out in 
a fine network, and the nucleus continues to swell out and 
become more rounded until the complete resting stage is again 
attained. The depression formed at the pole, however, remains, 
and in this there can often be seen exceedingly distinct centro- 
Spheres. Although the cases in the resting condition are not 
numerous where these bodies appear very distinct, yet in such 
cases as fig. 42 there can be no doubt of the continuance of the 
centrospheres into the resting stage of the nucleus. In the 
example given in Jig. 42 the whole cell is remarkably clear and 
free from granules, the two prominent bodies lying alone in the 
polar depression. To claim ‘that these bodies are not centro- 
spheres would be exceedingly dogmatic, and the only recourse 
left would be to name and describe two new organs of the cell 
which have the same appearance and occupy the same position 
as do real centrospheres. 
The general process of karyokinesis for the onion root may 
be summarized as follows: 
I. PROPHASE, 
I. The division begins with the separation of the centro- 
spheres, and when these have moved apart nearly 180° the 
incept of the achromatic spindle appears, forming two dome- 
shaped projections on opposite sides of the nucleus, at the sum- 
mits of which the centrospheres are situated, forming the poles 
around which are cytoplasmic radiations. At the same time 
the chromatin network is transformed into a continuous ribbon 
or spirem producing the figure known as the close mother skem 
(figs. 2-70). , 
