THE DISEASES OF THE SUGAR-CANE. 475 



perhaps be the case, but it does not explain the hardiness 

 of the resistant varieties which have been propagated in a 

 similar manner. The nearest allies of the sugar-cane among 

 wild forms appear also to be apogamous/ so that vegetative 

 reproduction is probably a natural mode of increase. There 

 seems, in fact, to be no more reason for deterioration in the 

 sugar-cane from this cause than there is in oranges, vines, 

 pine-apples, and other fruits propagated in this way." 



As far as can be gathered from the past history of these 

 "soft" canes, one of their characters has a/ways been that 

 they are especially liable to disease. The first references 

 to the Bourbon cane in Jamaica, India, Antigua, Queens- 

 land and Pernambuco seem, indeed, to unite its great value 

 as a sugar producer with a certain tenderness. There seems 

 then to be no valid reason to assume that the canes most 

 prized in cultivation are deteriorating. They are the best 

 results of the careful human selection of centuries. Suffi- 

 cient care has not been exercised to bring out disease- 

 resisting properties, as well as those of sugar production ; 

 but this is a point now held to be of first importance in all 

 suQ^ar-orowinQT countries. 



15. The search after new varieties of the sugar-cane is 

 by no means a new one. Selection has been at work for 

 thousands of years, and, thanks to it, we have at present 

 hundreds of varieties, many of them capable of producing 

 excellent results. 



Within recent years attempts have been made to deter- 

 mine the limits of asexual or vegetative selection. A series 

 of laborious experiments have been conducted by Edson in 

 Louisiana with the object of finding out whether, by plant- 

 ing the richer canes of a field, a richer variety may be built 

 up.^ 



The results thus far are in favour of such a method in 

 the United States. It has, however, been urged in Java, 

 and elsewhere in the Tropics, that the richer canes of these 

 localities are not capable of improvement to the same extent 

 as the poorer Louisiana canes — the sugar content of the 



^ Sagot ef Raoul. ^' Mobius, see also Went (3). ^ Edson. 



32 



