SUGAR 129 



diminutive frost-resistant member of the Chin group, 

 and the resemblance of the Katha to the Kahi grass 

 (Saccharum spontaneum) is sufficiently close for the 

 cultivators to assert that Katha originally sprang from 

 that grass. Seedlings of S. spontaneum selected and 

 raised at Coimbatore have yielded 3 to 5 per cent, of 

 sucrO'Se in the juice. They are perfectly fertile when 

 crossed with North Indian canes, and a number of hybrids 

 have been obtained with 7 to 12 per cent, of sucrose. 

 These facts seem to lend support to the belief of the 

 Punjab ryots, but it is full early for us to frame theories 

 as to the origin of the Indian canes, as there are many 

 more to be collected and studied, and we have little 

 knowledge of the wild Saccharums of the Malay 

 Peninsula. 



Considering the variability of the North Indian canes 

 according to locality, it will be obvious that seedlings 

 suitable for o.ne part will not be likely to do equally well 

 in others. The ideal would be to cross local canes with 

 South Indian ones, so that in every case one of the 

 parents would have adapted itself by long acclimatization 

 to the peculiarities of the tract. And the immediate- 

 problem is to induce the desired parents to flower at the 

 same time. But even if they do it is by no means 

 certain that crosses can be effected. The flowers are so 

 numerous and small that it is practically impossible to 

 emasculate them, at any rate in sufficient numbers to be 

 economically useful. And we prefer for the present 

 simply to bring the arrows together and then to examine 

 the seedlings for parental characters. And for this to 

 be done effectively we shall have to make a very much 

 more complete morphological study of the sugar-cane 

 than has ever been attempted before. 



Many kinds of cane have steadily refused to flower at 

 all. Some produce infertile flowers, and even those 

 which flower do not so do every year. There will thus 

 be small chance of applying Mendelian methods on the 

 new farm. Where possible this line will not be lost sight 

 of, and the main line of work will be to collect parents 

 of known useful properties and attempt to accumulate 

 these properties in individuals by crossing whenever we 

 9 



