C. A. BAKHEli 



157 



of dots indicating the falls of laiii. There is a remarkable descent in the curve 

 until the middle of June, after which recovery is rapid. Similar variations 

 in sucrose content were met with in many of the seedlings, as shown by 

 their petty and bulk analyses. 



It is. thus, of obxious importance to get the seedlings as ^,()()n as possible 

 into line with the local time of sowing and reaping, and to avoid these irregu- 

 larities as far as may be. At jiresent, after reaping rheni in -June to August, 

 they are planted in beds and will ripen about a year later. But. in future, 

 cuttings will if possible be taken from the young canes in February and these 

 will be planted in the ordinary varietal plots, which ha^■e been liitherto. ior one 

 reason or another. })lanted late each year. According to rhis arrangement, the 

 seedlings collected, say. in December 1915, will be first analysed in June 1917, 

 the second analysis being in February 1919. These dates throw a side light 

 on the length of time required for a seedling to be tested, the 1919 analysis 

 being by no means the latest one necessary before forming an opinion as to 

 its value. An additional advantage of this February planting will be that 

 sets will be available for distribution to North India at the usual planting 

 time there. 



As in the case with cane varieties extraordinary \-ariations have been met 

 with in a seedling, in analysing the different canes in the clump. This has 

 led to increased care in sampling. For pettv analyses, the canes chosen are 



