AND SPECIES HYBRIDS WITHIN THE GENUS SACCHARUM 413 
than the sessile one. S. officinarum behaves reversely ; both in the case 
of Chunnee and in that of Ruckree II the peduncled spikelets are fur- 
ther advanced than the sessile ones, but the difference is much less pro- 
nounced than in the case of S. spontaneum. It may happen in the case 
of S. spontaneum that in the pairs of spikelets close to the top of an axis 
the anthers of the peduncled spikelets contain pollen already, when in 
those of the sessile spikelets PMC can be found in diakinesis or meta- 
phase; the anthers of the peduncled spikelets at the foot of the axis 
may then contain pollendiads or mothercells in metaphase, while in 
those of the sessile spikelets mothercells will generally still be in synap- 
sis. 
In the three anthers of one flower one may either find similar or dif- 
ferent stages of division ; it may also happen that different stages occur 
simultaneously in one anther; this may even happen in the different 
sporangia of one anther. 
If one examines the anthers of fresh material under the microscope, 
one can very well distinguish through the wall between PMC,diads, te- 
trads and pollengrains. PMC still at rest, touch one another entirely 
still, when they have already passed to the synapsis-stage (see fig. 10. p. 
121). During this stage however a central lacuna begins to form between 
them. As long as this lacuna is not present the stages of division requi- 
red for a study of the chromosomes are certainly not yet formed; the 
material is too young. By putting aside a little materail, during each 
fixation, for microscopial examination in a fresh condition, one is able 
to controle whether the material is of the desired developmental stage 
or not. 
As a general rule it is safe to assume, that certain stages occurring 
now in the peduncled spikes of the top of the panicle of Saccharum offi- 
cinarujm will reappear 6 days or so later at the base of that panicle. In 
the case of Saccharum spontaneum, a species with smaller panicles, this 
interval will be shortened to 4 days. 
It türned out to best suit my purpose to collect panicles of Saccha- 
rum spontaneum 1 or 2 days after the beginning of the , flag’’-stage. In 
the case of Saccharum officinarum it was better to wait until the plant 
flagged” 4 or 5 days. If one waited longer, no pollen-mothercells were 
dividing or only a few of them at the base of the panicle were still in 
division. 
Spikelets in which, with the naked eye, bright yellow anthers are vi- 
Genetica V 8 
