14 BOTANICAL GAZETTE [JULY 
heterotypic division would be quadrivalent, instead of bivalent, 
as in the case of the normal tetrad. 
As the chromosomes are drawn toward the nuclear plate, they 
become bent, so that each one represents a short arc of a circle 
t 
or a broad open U form, as if drawn more forcibly by threads 
attached to the middle portion. As they are oriented on the 
nuclear plate transverse to the axis of the spindle they at first’ = 
stand in various positions. Some lie tangentially, with the convex 
side toward the axis of the spindle and with both ends at the 
periphery, while others lie so that one end is directed toward © 
the axis, with the other end at the periphery. In this way 
they are sometimes ‘‘convolute” or more or less “ imbricate”’ 
in the nuclear plate, and may remain so for some time during 
the metaphase, or while separation at the nuclear plate is begin- 
ning to take place. Before separation has proceeded far, how- 
ever, the chromosomes are usually more bent upon themselves, — 
with the free ends of the U nearer together and directed out- 
ward in the usual way. 
While the monaster is forming, and many of the chromo- 
ER ERT es 
sainanaoraerpene sete Sp mere 
somes stand with one end directed toward the axis of the i 
spindle and the other radiating therefrom, certain ones which are . 
strongly lobed at the ends present a figure which leads one to 
think that the ends might be separated first, and that the 
chromosomes might then, in the anaphase, move to the poles in 
the form of aU, but with the concave or open side directed © 
toward the poles, or that the inner ends might be separated first, — 
somes has begun in such figures, and it is suggested that they 
are not yet drawn into proper position in the nuclear plate. 
As separation of the paired chromosomes begins, the larget - 
number of spindle threads appear to be attached in one tuft at 
the middle. The flattened U-shaped chromosome now broadens 
at the middle, in response to the tension of the tuft of threads" 
attached at this point on either half, and the chromatin substance 
is soon drawn out into a short process on either side which 
x 
