218 BOTANICAL GAZETTE [SEPTEMBER 
resembles much more closely the process in vascular plants than in the alge 
to which Chara has been assumed to be more closely allied. The stages agree 
in general with those described by Osterhout. No functional centrosomes are 
to be found. “Extra nuclear nucleoles” are abundant, and seem to furnish 
material for the spindle fibers and cell plate. The structures assumed by 
Kaiser to be centrosomes are doubtless such nucleolar masses. 
In contrast to the method of spindle development described for the vascu- 
lar plants and Chara, Harper finds in the ascus of Erysiphe a type much 
more nearly resembling that described for animal cells by Hermann and 
Flemming. A disk shaped central body is present with each nucleus through- 
out nuclear division and ‘spore formation in the ascus. At the beginning of 
spindle building this body is surrounded by a system of radiating kinoplasmic 
fibers. Then two such centers appear beside the nucleus and separate grad- 
ually to form the poles of the spindle. From these centers fibers extend and 
are attached to the chromosomes. In the bounding off of the ascospores by 
free cell formation the polar radiations of the last preceding mitosis perform 
an entirely new function. They grow in length and swing back around the 
nucleus, which has been drawn out into a beak beneath the central body, and 
fuse laterally to form a new plasma membrane around the young spore. The 
bounding layer of the young ascospore is thus composed of the same kinoplas- 
mic substance as the polar radiations and spindle fibers. The spore wall is 
formed much later. 
» In Basidiobolus Fairchild finds very characteristic barrel shaped multi- 
polar spindles. The fibers converge in groups, and each group ends ina 
strongly staining body. These bodies, taken together, make up a sort of polar 
plate. Here also a typical cell plate is probably formed as in Chara. Fair- 
child further notes the very interesting fact that in Basidiobolus we have one 
of the shortest possible life histories, involving at the same time an alterna- 
tion of sexual and asexual fruit forms. Two successive nuclear divisions 
may complete the entire round of conidium and zygospore formation. 
Swingle has given for Stypocaulon the most complete description of a 
sharply differentiated centrosome, and its division and migration during 
spindle development, which has yet been worked out in plant cells. The 
process here, also, is much the same as in animal cells. The apical cell of 
this plant is almost constantly in division, and the polar radiations and cen- 
trosomes persist in it through the resting stages of the nucleus. The spindle 
has both “mantle” and “central” fibers, and at the time of its greatest 
development the polar radiations are much reduced. The amount of kino- 
plasm at the two poles is regularly unequal. The contrast in structure and 
reaction of the fibrous kinoplasm and alveolar trophoplasm is nowhere more 
sharply shown than in this apical cell of Stypocaulon. Cell division takes 
place without the aid of connecting “fibers” or constriction of the plasma 
| 
