W. A. Davis 347 



time A changes from + 41 per cent, to a negative value — 23 per cent. 

 After midnight the apparent dextrose suddenly begins to rise again 

 and the. laevulose to fall ; as would be expected the value for A becomes 

 less negative (changes from — 23 to — 7-4). 



As showing the gradual transference of the optically active impurities 

 from the tops to the bottoms of the stalks, it is interesting to compare the 

 values in Table VIII for say 6 a.m. with those for 12 noon. The impurity 

 in the 6 a.m. tops is such as to cause A to have a positive value + 10-3 

 per cent, (invertase) ; at the same time, however, the bottoms have a 

 negative value — 7-4, but at 12 noon the value of A for the bottoms has 

 become positive, viz. + 1-8 (the sum of + 10-3 and — 7-4 is + 2-9). 

 Similarly at 6 p.m. the value of A is negative in the tops (— 84-5) but 

 positive in the bottoms, but at midnight the bottoms show a negative 

 value, — 23-3 ; had the whole of the material causing the negative 

 value at 6 p.m. been transferred to the bottoms, the change expected 

 would be - 84-.5 + 41-2 or - 43-3. 



In the case of the stalk bottoms (Fig. 5), where the fluctuation of 

 the apparent laevulose is relatively small, it is the dextrose curve 

 which follows most closely the curve of total hexoses, but as pointed 

 out in the case of the laevulose in Fig. 4 this has no real significance 

 and is a result merely of the method of calculation. 



If one compares merely the relative position of the apparent dextrose 

 and laevulose curves in Figs. 4 and 5, dextrose seems to accumulate at 

 the bottoms of the stalks far more than the laevulose, the values for 

 dextrose (7-4 to 11-00 per cent.) being higher in Fig. 5 than in Fig. 4 

 (4-87 to 5-71), whilst the fluctuations of laevulose are smaller (0-23 to 

 1-8 per cent, in Fig. .5 as compared with to 4-5 per cent, in Fig. 4). 

 But from the considerations already brought forward it is clearly 

 quite unsafe to conclude that it is actually dextrose which accumulates 

 at the bottom of the stalks, as large quantities of other optically active 

 substances are undoubtedly present, which cause the wide divergences 

 between the results for cane sugar by the polarisation and reduction 

 methods. If dextrose were the principal sugar present (in some cases it 

 appears to be, as at 6 a.m., the sole hexose in the stalks) it would point, 

 as assumed by Lindet, to the laevulose being largely consumed on the 

 way from leaf to root for constructive purposes ; but it would necessitate 

 also that the saccharose in the root is built up from the dextrose being 

 conveyed to it. This would involve a transformation in the root of 

 dextrose into laevulose, followed by a synthesis of cane sugar from 

 dextrose and laevulose. Whilst this operation is a possible one, it is more 



