THE CYTOLOGY OF THE SUGARCANE 519 
seedlings after selfing, while its use as the male parent in hybridisation 
is frequently quite succesfull. These facts show that in certain condi- - 
tions EK 28 can be satisfactorily fertile, so that it was worth while to 
examine, wheter EK 28 shows in certain inflorescences a more re- 
gular reduction division, with which a greater fertility is likely to be 
coupled. 
In the two cases examined, as well in that of Ardjoeno as in that of 
EK 28, irregular reduction-division was coupled with sterility. In the 
case of Ardjoeno we saw that regular reduction division resulted in 
fertility. On theoretical grounds it is therefore probable that a connec- 
tion between the course of the reduction division and fertility does 
exist; observations in the field point the same way. 
All individuals examined, arisen from crosses of different kinds of 
sugarcane with one another: 247 B, SW 3, SW rrr and EK 2, have 
exactly 80 chromosomes in the diploid phase. None of these shows even 
the slightest deviation of this number. All original sugarcane kinds of 
which it is probable that they arose, in part at least, by spontaneous 
crossing or by self-fertilisation from other kinds of Saccharum offici- 
narum, also possess 80 chromosomes. Probably the gametes, which con- 
tributed to the formation of these kinds, have therefore had 40 chro- 
mosomes each. The diploid chromosome number 136 of the crosses S. 
officinarum x spontaneum can also be best explained by assuming 
that the eggcells of S. officinarum which contributed to the formation 
of these crosses, always contained 40 chromosomes 1). All this points 
towards fertility of those gametes of S. officinarum with 40 chromoso- 
mes and to infertility — as a rule — of those having another chromo- 
some-number. 
This also, it seems to me, makes a relation between the course of the 
reduction-division and fertility probable, but more observations are 
required to establish such a relation with certainty. 
As to Lahaina, it should still be pointed out, that this cane which is 
absoluty sterile in the male line, showed a reduction division, which, it 
is true, was slightly irregular in the majority of cases, but that pollen- 
. mothercells, showing a complete reduction- division, were by no means 
rare. The fact that this cane is sterile in the male line, while nor- 
1) In the megaspore-mothercells of sugarcane I could also occasionally see an 
r regular reduction-division. 
