C. A. BAtlBER 143 



it has been necessary to take a very large series of measurements, and this 

 part of the work has been very laborious. For instance, on going through the 

 varieties, it soon became evident that the organs in the Sunnabile forms were 

 on the whole shorter an.d broader than in the Saretha varieties. As each 

 series contained thick and thin, large and small varieties, it became necessary 

 to include proportion.al measurements, and modules have been introduced, 

 founded on the length of the organ divided by its breadth, and this has given 

 satisfactory results. But one was also soon convinced that the shape of the 

 leaf was different in the two groups, and the module fails to indicate this fully. 

 For the demonstration of this difference in shape, it became necessary to take 

 a series of width measurements at different places up the leaf till the widest 

 l)oint was reached. The points fixed were the base, 1" from it, and also 6" 

 and 12" and the widest place, wherever that was. By studyiji.g these in each 

 variety an.d averaging them in the whole group, we have been able to express 

 the shape of the leaf by mean.s of the ymdieiit of increase in width, and this 

 gives an expression of the general shape of the leaf. The width of the midrib 

 was measured at the same places, for it was noted that this part of the leaf 

 was nmch more prominent in. the Saretha series (as in Saccharum n'pon- 

 taneum) than in the Suiui.abile grouj3, and it was desired to state this accurately. 



Many of these measurements were undertaken, not so nmch to give exact 

 data, as to convince one's self that an observed difference was actually present 

 as a group character — a matter of pecuUar difficulty in varieties with such a 

 wide range in. general size — and. if present, what exceirtions were observable. 

 The results are in.terestin.g, as indicating the relative closeness of the different 

 varieties in the groups, for, generally, where exceptions occur, such are also 

 traceable in other characters, and we see that in each group there are 

 fluctuations in various direction.s, suggestive of new points of departure iii the 

 evolution of the t>q)e. 



The variation in the length of organs at different periods of growth has 

 received special attention, and it is hoped will form the subject of another 

 memoir. The length of joint, for instance, is very different at the base, in the 

 middle and at the top, ajid the two groups under consideration show marked 

 differences in this variation ; there is greater \-ariation in the Saretha series, 

 an.d the longest joints are always nearer the base than in the Suimabile series, 

 which is more uniform all through. But to pub this down in definite terms was 

 very difficult. In each variety 20 canes were taken at haphazard, and each 

 joint, leaf sheath and lamina was measured in succession from base to apex. 

 As the mimber of joints varies from cane to cane, it becomes an extremely 



