88 BOTANICAL GAZETTE [aveust 
and in this he is confirmed by GrécorrE and WyGAERTS (9). — 
This vesicle, which he interprets as a Hautschicht, resolves itself into 
fibers and furnishes at least a part of the material for the spindle. I 
do not find such a vesicle separate and distinct from the nuclear 
‘membrane, but I find strong evidence that the nuclear membrane 
itself becomes resolved into fibers. This view is quite compatible 
with the generally accepted theory of the nature of a plasma mem- 
brane, and the evidence is presented by such appearances as are 
shown in figs. 12-14. In fig. 12 we have a nucleus which in one 
plane shows a number of prominent lobes. A few fibers are visible — 
over one lobe, and at several other places the nuclear cavity i 
apparently bounded by a weft of fibers. These are either derived — 
from a layer of kinoplasm which closely invests the nucleus or 
from the nuclear membrane itself. The fact that the nuclear 
membrane disappears as these fibers come into view would lend force — 
to the latter supposition. In fig. 13 fibers are shown over one lobe 
of a nucleus which is very much elongated, and in fig. 14 they may 
be seen at both ends of a similarly elongated nucleus. In the latter 
case the nuclear membrane persists in several places, seeming t0 — 
merge gradually into the fibrillar condition. The fibers appear 10 — 
conform to the irregularities of the surface, giving strong indications — 
that they are derived from the nuclear membrane. ae 
HarPER (10) has shown a close relation between membranes and 
fibers in Erysiphe, where, in free spore formation in the ascus, the 
fibers which mark out the boundary of the future spore fuse side 
side to form a plasma membrane. The nuclear membrane is gener 
ally believed to be of kinoplasmic origin, and so are the fibers of the : 
achromatic spindle. Evidently then, the transition from the oné 
to the other may be easily accomplished. 2 
Soon after the appearance of the first fibers, the number is greatly 
increased, but I have not been able to determine the origin of - 
remainder. The completed spindle is bipolar, and may be pointed 
(fig. 15) or blunt (fig. 16). The ends may terminate near OF abe 
distance from the cell wall. It happens frequently that one end 
extends into a lobe of the spore mother cell, and the other abuts 0? 
the infolded wall between the two adjacent lobes which stand oppo 
a it, a producing a very much flattened pole or even a forked ont 
(fig. 16). 
see 
