AND SPECIES HYBRIDS WITHIN THE GENUS SACCHARUM 139 
most parallel to one another. The same is seen in the left side of the 
figure, a little below the aequator. Of two pairs, at the lower side of the 
figure, lying close to the nuclear wall, it is also certain, that the chroso- 
somes are separated. Finally one sees, when focussing at a low level, 
two very small, doubtless univalent chromosomes, at a great distance 
from one another. A small, perfectly transparent nucleolus lies between 
them. These two chromosomes also. almost certainly belong together. 
If these 10 chromosomes had been united to pairs the number of ge- 
mini in this nucleus would consequently have been 40. Frequently ho- 
wever the number of non-pairing chromosomes is still larger, so that 
the microscopical images are still more convincing, but then the picture 
becomes so complicated, that the exact number of univalent and biva- 
lent chromosomes is usually undeterminable. Fig. 36 shows a similar 
case. At the upper side of the nucleus we see a mass of bivalent and 
univalent chromosomes, which cannot with certaintly be dissolved. 
In the center of the nucleus 4 sets of chromosomes are seen when focus- 
sing at a high level, which have not paired. Focussing at a low level, 
we see three other-sets, one of which is covered by the nucleolus. It is 
not improbable that 27 gemini and 13 sets of unpaired chromosomes 
are present. 
One could, of course, suppose that the univalent chromosomes still 
had to pair; this however is certainly not the case, because one meets 
again with the unpaired chromosomes in metaphase and anaphase. It 
seems that the mutual affinity of these chromosomes has become wea- 
kened. It is, however, not impossible, that during prophase these chro- 
mosomes have been united. 
In G. G. N. G. one meets in nearly all cases with unpaired chromoso- 
mesin the diakinesis-nuclei ; the nucleus pictured in fig. 34 is an excep- 
tion and even in that nucleus probably not all chromosomes were com- 
pletely paired. It may even happen in the case of G. G. N. G. that pret- 
ty nearly all chromosomes remain unpaired. That however is the ex- 
ception and leads to an entirely abnormal division of the PMC, so that 
no normal tetrads and pollengrains are formed. Preparations of this 
kind very frequently show abnormal division figures which will be 
_ treated at the end of this paragraph. We shall now speak of the divi- 
sion in which, in prophase, univalent chromosomes may occur besides 
bivalent ones. This one leads not to the formation of abnormal tetrads. 
If one views the nuclear plates of G. G. N. G. from the side, we see 
