C. A. BARBER 165 



Many sununation series have been allowed to stand unaltered, in others 

 the last few numbers have been altered to allow a reasonable drop to the last 

 eight, in others it has been found necessary to emphasize a maximum or to 

 indicate a late maximum by the increase of one or two fiorures in the non- 

 representative columns. In doing this, it is borne in mind that the averages 

 of a series such as this are much more uniform than the individuals 

 comprising it, whether early or late, long or short, as explained above. 



In obtaining the averages for the typical cane representing the whole 

 Saretha or Sunnabile group, exactly the same procedure has been adopted, 

 but, in place of the figures for individual cane varieties, the summation 

 series of all the varieties are placed in the table, these varieties being of the 

 same age and growing at the same place, under similar conditions. Similar 

 precautions are taken as to the average number of organs, the position of the 

 maxima and the smoothness of the series at the end. As will be seen in the 

 sequel, the method is of service, in that, not only do varieties differ from one 

 another in these series, but the whole groups of Saretha and Sunnabile forms 

 show differences which are of value in distinguishing them. 



But long series of figures are difficult to follow, and they have, accordingly, 

 been plotted out in curves in which the characteristics of growth can be detected 

 at a glance (Chart I). As examples of such curves, are appended those, 

 representing the lengths of leaf sheaths in Khar, in the shortest and longest 

 canes (Nos. 4 & 19) of the twenty measured and in one which has the average 

 number of joints (No. 8). Added to these is the ideal summation curve of 

 the whole twenty canes examined. This latter is of mean length, is much 

 more uniform than the curves of the individual canes, in that excessive 

 variations from joint to joint have been ruled out. and the early and late 

 maxima are indicated, although greatly reduced. The last few, immature 

 organs are not all shown, and the crosses indicate where they commence. For 

 other length curves, reference may be made to the paragraphs in Section V. 

 dealing with the length of the several organs in different parts of the cane, 



