170 PteRIODICtTY IN THR GROWTH OF SUGARCANE 



cases quoted above. But it is also obvious, if there is anything like periodicity 

 in growth, a rotation of zones of longish and shortish joints following one 

 another, that this periodicity, in any averaging, will also tend to be ruled out 

 and leave little trace behind. We must therefore in this study hark back 

 to the individual curves of separate canes. And, in noting the oscillations in 

 length in successive joints, we must see that the canes compared have as 

 nearly as possible the same number of joints. Three canes of Pansahi, grown 

 at Samalkota in 1916, with the same number of joints, have been selected 

 and their joint length curves plotted out on squared paper (Plate VI). The 

 oscillations in these curves, as was to be expected, are very great, but we see 

 at once that there is a regularity in these ups and downs in the three curves, 

 so that we may easily pick out the successive maxima. There being an equal 

 number of joints in the three canes, we can without difficulty average them, 

 and we see that, at fairly regular intervals, there is a zone of increased growth 

 in length, the maxima occurring at the 3rd, 10th, 15th, 20th, 25th and 28th 

 joints. The 28th joint is in the region of these which have not yet completed 

 their growth, while the 3rd is at the base of the plant where disturbances may 

 well occur owing to earthing up and the omission of the joints below the 

 surface. In the middle of the cane there is a regular periodicity in the growth 

 in length of joints. Our example has, frankly, been chosen where this periodi- 

 city is rather clearly shown, but it will be seen directly that it is no isolated 

 instance. A method has been devised, by which each set of twenty canes 

 may be judged, as to any periodicity in the length of joints in the individual 

 canes and in the whole series taken together (Plates VII and VIII). The twenty 

 canes are arranged across the page and the number of joints in each cane is 

 indicated by a row of equal squares, the cane with the greatest number of 

 joints being placed at the top, and the others in succession below it, until the 

 cane with the smallest number of joints is reached at the bottom. By this 

 arrangement, any set of canes of equal length are placed together. The first 

 diagram (Plate VII, fig. 1) represents twenty Pansahi canes grown at Samal- 

 kota in 1915-16, and, in this diagram, the canes numbered 14, 15 and 16 are 

 the ones whose curves are given in Plate VI. The maxima in the length of 

 joints, indicated in the diagram by circles, occur with a certain amount of 

 regularity, the increment being most frequently at every fifth or sixth joint. 

 Thus, in Cane No. 5, they occur at joints 2, 7, 13-14, 19 and 25 ; in No. 6, at 

 joints 2-3, 8-9, 14-15, 20-21, 26 and 30 ; in No. 7, at joints 2, 5, 10, 15, 20^ 

 32-33 and so on. Connecting lines have been drawn between the apparently 

 related maxima in adjoining canes, namely, such canes as have the same or 

 nearly the same numbers of joints ; and, where such related maxima are 



