Current Investigations in Economic Botany. 107 
conditions. They bred true to colour, and shewed on analysis the 
following differences :— 
Canes per 
acre tons. 
Juice per 
acre gains. 
Sucrose 
per gallon. 
Quotient 
of Purity. 
Sucrose 
per acre. 
Striped Canes. 
21-80 
2-696 
2-310 
93-03 
6,228 
Yellow Canes 
(Sport). 
27-27 
3-555 
2-270 
91-64 
8,070 
The sport cane in this instance gave a yield of sugar at a rate 
nearly 25 per cent, above that of the parent cane, the result being 
due to the heavier yield of the sport cane rather than to its 
intrinsic superiority, its juice being, as a matter of fact, less rich in 
sucrose, and less pure than that of the parent. The result of one 
year’s experiment is not sufficient to demonstrate conclusively that 
the sport will permanently be superior to the parent cane, and 
experiments are being continued to allow of further evidence being 
obtained. The example will serve, incidentally, to indicate some 
of the different factors which have to be taken into account in 
determining the value of a sugar cane as a commercial source of 
sugar. 
Experience in Queensland, Mauritius and Barbados proves 
that in bud variation there is a possible means of obtaining new 
varieties of sugar cane which may be superior to the parent plants. 
The process cannot, however, be controlled exactly, and it is only 
by close observation and experimental rearing, and testing of sports 
as they arise that beneficial results are to be looked for. This 
method of improving the stock of canes lacks opportunities for 
working steadily and systematically to a definite end, such as is 
afforded by chemical selection and controlled hybridization. 
Chemical Selection of Sugar Canes. 
We will now turn to the consideration of the methods 
employed and the results obtained in the second mode of improving 
the sugar cane, namely that which is usually known as “ Chemical 
Selection.” As has already been pointed out canes vary greatly in 
richness, or to be more precise in the percentage of cane-sugar or 
sucrose contained in their juice. On a priori grounds it would appear 
reasonable that if a large number of canes of a given variety were 
analysed it would be found that the percentage of sucrose in their 
juice varied, that a comparatively few would have a very low per- 
