C. A. BARBEH 139 



knowledge of the canes in their natural conditions and the opinions of local 

 cultivators regarding them, prevents us from analysing these characters at 

 present. Enough has been said to show that, to the already well recognized 

 groups of Iii.dian canes typified by Mungo, Pansahi, Nargori, two more classes, 

 less obviously marked to the casual observer but none the less genuine, must be 

 added. In these groups the usual separation of the canes into Ukk and Ganmi 

 breaks down, and, indeed, there are some indications of a further gradation 

 between the latter and the Painida class. If this turn out to be the case, we 

 shall perhaps be able ultimately to throw light o]\ the hitherto insuperable gulf 

 between the indigenous Indian and exotic tropical canes, a gulf which has led 

 to the idea that we have in them two groups of cultivated plants closely related 

 and yielding the same commercial substance, but arising from different wdld 

 parents. I know, at present, of no sin.gle, fundamental dif?eren.ce between 

 these two groups, and regard it as quite possible that, in India itself, we may 

 find the transition from one to the other. Historically, at any rate, the 

 cultivated canes of the tropics have been traced to Northern India, and we may 

 at present rest at that. 



But there is another aspect of this study to which attention may be drawn. 

 Our classification, is not merely an empirical statement of unconnected differ- 

 ences, a sort of analytical key for the separation of varieties, but presents the 

 data for a reasoned statement regarding the lines of evolution among a section 

 of cultivated canes. We also claim to have made distinct advances along the 

 difficult path of tracing the origin of cultivated canes from their wild ancestor. 



There are, as already stated, in. each class, a series of can.e varieties, passing 

 more or less imperceptibly from what are confessedly the thinnest and most 

 primitive canes in the world to thick, well-deyelopcd forms showing distinct 

 resemblances to tropical sugarcanes. By the careful study of the variation 

 between, these extremes, we are placed in a better position for tracing the 

 evolution of the latter from primitive ancestors. And, what is more significant, 

 we have established a series of connecting links between cultivated canes and 

 wild Saccharums now growing in India. 



Our study of Sacchanim spontaneum is still incomplete, as the full method 

 proposed is to examine and compare a large series of seedlings of this species, 

 in order to establish the limits of variation in the vegetative organs. But a 

 constant collection of specimens, while on tour in. different parts of India, hass 

 demonstrated the fact that there are some very distinct varieties of this species, 

 more or less confined to definite geographical regions. There is the common, 

 wire-leafed weed, usually known as hans or kahi grass, met with all over India, 



