Current Investigations in Economic Botany. 111 
and 0-199. Further evidence is needed, and is being sought, before 
any definite conclusions can be drawn as to whether this method is 
a reliable and practicable one for improving the sugar contents of a 
cane to the extent of raising races of cane superior to the original 
stock. 
In the experiments conducted by Thompson and Edson, Bovell, 
Borame and Watts the unit of selection has throughout been the 
individual cane or stalk of which several are borne by each cane 
plant. It is possible that the individual canes of one plant vary 
considerably in sucrose contents, and that whilst one is rich the 
others are poor in sugar. Richness in sucrose of an individual 
# < 
cane might on this view be only an accidental circumstance and 
not necessarily a characteristic of the plant as a whole. Accor¬ 
dingly it is possible that it might be advisable to select plants rich 
in sucrose, rather than individual canes. This, in short, is the view 
of Kobus of Java and the following summary will indicate the 
evidence on which it is based and the results obtained by him, 
working along these lines. 
Kobus first endeavoured to ascertain whether the percentage 
of sugar was constant or not in the different stems, or canes, of 
one plant. To this end he analysed about 500 canes belonging to 
134 plants of “ Fiji Cane.” He found that the percentage of sugar 
varied considerably, that in fact in only a few plants were the 
differences less than two per cent., variations of three to four per 
cent, being more usual, and as much as seven to eight per cent, 
not uncommon. The following instances will suffice to indicate the 
variation found, the figures on each line referring to the individual 
canes of one plant:— 
Percentage of S 
ucrose in the Canes of Separate Plants. 
Variation. 
15-42 
15-38 
14-12 
9-94 
8-74 
— 
6-68 % 
13-52 
12-14 
6 92 
— 
— 
— 
660 „ 
5-74 
9-24 
11-22 
8-42 
13-76 
6-76 
8 02 „ 
12-52 
13-40 
13-02 
— 
— 
— 
0-88 „ 
12-26 
12-02 
12-68 
— 
— 
— 
0-64 „ 
14-74 
14-64 
14-10 
— 
— 
— 
0-64 „ 
Other varieties were also analysed and differences corres- 
