THE IMPROVEMENT OF THE SUGAR-CANE BY HYBRIDISATION, 313 
probably practical value. As to the cause of the real nature of these bud 
varieties very little is known at present. It has been suggested that the. 
striped or ribbon canes are the results of cross-fertilisation, and that, 
therefore, the sports are to be considered as cases of dissociation and then 
segregation of hybrid characters, or of atavism. 
It is supposed that unfavourable influences, either external or internal, 
temporarily encumber the growth of the young buds and predispose them 
to reversions. But if sports are of an atavistic nature and are favoured 
by influences that impede normal growth, how is it that they, almost 
without exception, give such excellent results when cultivated, being 
hardier and richer than the original stock from which they arose ? 
Variations in Sugar-contents of Individual Canes (e).—Bearing in 
mind the classical experimental work, with which the name of de Vilmorin 
is associated, in selection of the sugar-beet, by which, through the aid 
of science, the sugar-contents have been raised from 10 to 18 per cent., 
workers with sugar-cane were led to commence investigations with 
regard to the chances of obtaining canes of higher saccharine content. 
The occurrence of a wide range of variation in the percentage of 
saccharose in the juices of canes of the same age and variety was soon 
noticed, and the fact that the sugar-cane was propagated by cuttings 
naturally suggested that any improvement inherent in the plant could be 
developed more rapidly than if it had to be grown from seeds. Investi- 
gations in chemical selection have been carried out in the West Indies, 
but the results, so far obtained, are not at all conclusive. 
d’Albuquerque, Barbados, at the last West Indian Agricultural 
Conference, 1905, stated that it would appear “ that, with a given variety, 
the richness or poorness of the seed-cane (i.e. cane used for planting) 
does not affect the quality of the juice of the resulting crop.” Harrison, 
British Guiana, concludes that the “relative richness of seedlings is 
qualitatively, if not quantitatively, constant.’ Watts, Antigua, states that 
“some difference is induced by the process of selection and, while this 
method of work is not likely to be followed by practical planters as a 
means of improving their canes, yet the fact is interesting from its 
scientific aspect as indicating that plants propagated by cuttings are 
subject to slight alterations.” 3 
In Queensland it has been stated that improvement from single-stalk 
selection is not as great as would be expected, while in Java the evidence 
seems to point to the fact that selection among “ cane-clumps ”’ is likely 
to give better results than selection among individual canes. 
The following is the summary of the results of chemical selection 
investigations by Kobus in Java up to 1902, and has been abstracted 
from the JourNAL of the Royal Horticultural Society, 1903 (vol. xxviii. 
p. 298) : 
1. “ The amount of sugar in the individual haulms of one sugar plant 
was apt to vary greatly.” 
2. “The variability of the amount of sugar in the different varieties 
was greatest in thick-stemmed varieties that had long been in cultivation, 
and least in young ones more recently selected from seed.” 
8. “The amount of sugar in the cane varied directly with the weight 
of the same.”’ 
