436 BOTANICAL GAZETTE [JUNE 
doubt in my mind that this is the case, although it is difficult to 
determine, but there is a bare possibility that the chromosomes 
may be turned the other way. However, their position in the 
nuclear area, and their appearance and behavior on the spindle, 
indicate that the loops are turned outward. At this stage also 
the chromosome is generally stained so dark that no trace of 
the chromatin band is discernible, but in sections stained with 
anilin safranin, and gentian violet the central part stains darker 
and clearly indicates the position of the chromatin band ( fg. 
30). The splitting of the chromosome is gradual ( jigs. 29- 
32), and consists in the separation or untwisting of the chro- 
matin coil, which is gradually pulled out until it lies like a 
straight band or rope on the spindle threads (figs. 33-36). 
The untwisting of the chromosome can be seen easily in all 
stages, and by proper focusing the entire coil can be traced. 
Figs. 33-35, which represent the later phases of this process, do 
not represent the coiled appearance as well as the sections. 
After the chromatin coils have straightened, the splitting 
takes place in the middle of each one at the equator. Thus 
there is an actual transverse division of the chromosomes, the 
half of each original chromatin loop passing to opposite poles 
of the spindle. Each daughter nucleus, therefore, receives 
about as many chromatin granules as there were in the mother 
nucleus, and although there is no diminution in the number of 
chromatin granules, only half of the granules originally present 
in the mother nucleus are represented in each daughter nucleus. 
‘It will be seen that although the chromosomes are not all of the 
: same size and length, yet if the chromatin band breaks at prac- 
: tically the middle point each daughter nucleus receives about 
sa same number of chromatin granules ; and since the chroma- 
tin granules are the same in number as in the mother nucleus it 
cannc jot be proper to speak of a reduction in the amount of chro- 
c matin, although only half ~ the * original chromatin granules are 
-Tepres ented. There is no number but a reduction of 
hi cm in n kind, Whether there is a . reduction in the number 
matin granules befor id ~ ee nucleus: is ecas aes oe 
