32 KANSAS UNIVERSITY QUARTERLY. 
stance it is wholly of nuclear origin, and that the nucleolus has 
contributed the substance for its formation. The abundance of the 
nucleolar material and its early solution suggest this, and no evi- 
dence of a contribution from the cytoplasm is to be seen. But 
judging from the amount of the nucleolar substance originally 
present it is probable that only a fraction of it is used up when the 
spindle is first formed. After the daughter nuclei have gone into 
the resting state the spindle still persists and increases in size, as 
shown by Fig. 6, plate V. The cell plate is formed, as usual, at 
the equator of the spindle, and with the spindle it continues to 
grow in diameter until it has traversed the cell (Fig. 7, plate V. 
While the spindle has been broadening it has been decreasing 
in length, as if it were being drawn in to form the cell plate. After 
the cell plate is complete the remnants of the spindle lose their 
distinct thread-like character and appear to merge gradually into 
the honey-comb structure of the cytoplasm (Fig. 8, plate V). When 
the spindle is first formed it takes on a violet color with the safranin- 
gentiana violett-orange stain, but after the daughter nuclei have 
gone into the resting stage the spindle fibres assume the pale brown 
color of the cytoplasm; so that both as to construction and reaction 
to stains the kinoplasm in this instance appears to merge gradu- 
ally into the cytoplasm, or, in other words, the differences between 
kinoplasm and trophoplasm slowly fade away, as if these structures 
were morphologically the same thing. It will be observed that the 
daughter nuclei enter into the resting state and lose all connection 
with the spindle at an early stage in the formation of the cell plate, 
so that the growth of the plate proceeds without any kinoplasmic 
connection of the nucleus with the region of growth. This is the 
very opposite of the process of the formation of the plasma mem- 
brane about the spores of Erysiphe and Peziza as described by 
Harper, and stands out as evidence that the kinoplasmic connec- 
tion of the nucleus with the growing membrane is not necessary in 
all cases. Reference has already been made to the opinion of 
Strasburger that the nucleolus contributes its substance to the for- 
mation of the plasma membrane of ascospores. We see in the 
pollen mother cells of Asclepias that the cell plate and the kino- 
plasmic spindle continue to grow after they have severed their 
connection with the daughter nuclei and the latter has entered into 
the resting state with their nucleoli already present. Accordingly 
if the nucleolar substance of the mother cell takes part in the con- 
tinued growth of the cell plate and spindle a portion must have 
remained dissolved in the cytoplasm after the completion of the 
