24 BOTANICAL GAZETTE [JULY 
in the same plant qualitative reduction may take place in some 
cells, while quantitative or numerical reduction only takes place 
in others. This seems to me, as a working hypothesis, more 
reasonable than to insist, because one type has been found 
in one or several plants, that all plants must conform to it. 
CORNELL UNIVERSITY. 
EXPLANATION OF PLATES I-VI. 
(All the figures are drawn to the same scale, using a Zeiss microscope, compen- 
sation ocular 12, and the 2™" homogeneous immersion objective, the image being 
down even with the base of the microscope. The figures are reproduced from 
these drawings without any reduction.) 
PLATES I AND Il, Ariz: triphyll: 
_ 
Fic. 1. Spirem stage of nucleus showing the dividing thread with chromatin 
masses on it. The masses of chromatin are partly divided, and appear as paired 
masses along the thread. 
Fic. 2. Portions of partly formed spirem, showing indications of longitudinal 
division. 
I Early stage of chromosomes, just after segmentation of spirem band, 
een different stages of longitudinal division, forming oblong plates with ends 
maaan ae bik or those approaching X and U forms, where the division is more 
ents of linin are still attached. 
gre 4, © 6 7. Different stages, showing the gradual shortening of the chromo- 
somes in some cases, in others different forms of chromosomes, in same stage 45 
those in fg. 3. In figs. g and 5, some are opening out to form rings; in fg. 6 one of 
the oblong rings is twisted to form a figure 8, and in fig. 7 one shows two openings 
along the line of longitudinal division. All of the chromosomes in these figures are 
os more or less irregular, sph ah and show ee of the linin attached. 
s. 8, 9, 10. Chromosomes much shortened, an large number of them = 
nda 
in te ring form. In fg. “9 two have suo divided to form rods, two form 
oblong plates with forked ends, and one forms a ring, while in fg. zo nearly all are in Z 
the ring form. Some of them are still angular, and show the peculiar form of somany 
of the rings of Ariseema. The tetrads are being formed by the accumulation of the 
chromatin in denser portions near the ends of the rods, and near the ends of each — 
half of the rings, the paler zone across the middle showing the line of transvers¢ 
division. Linin threads are still attached to some of the angles. The nuclear 
membrane is still intact, and the ch 
from the 
Fie: II, 12, 13. The nuclear sie trae has disappeared, and threads of king - 
m are entering to form the spindle; chromosomes in the form o 
pl . 
rods, or angular plates, which are solid or with a central perforation indicating the — 
ring form 
Se ene 
romosomes are arranged around the periphery % 
the ea eat in the positions occupied by them when they separated as segments — 
eo ee ee 
i 
