350 The Dcxtrose-lAUVidose Ratio hi the MidKjolil 



4. In the mid-ribs and stalks, especially at the bottoms of the 

 latter, the dextrose always appears to be in very large excess as compared 

 with the laevulose ; this is probably due to the proportion of the dextro- 

 rotatory impTirity being relatively greater in these parts than in the 

 leaf, as is shown by the divergences between the polarisation and 

 reduction values of saccharose being far greater. 



5. The apparent fluctuations in the ratio of dextrose to laevulose 

 are probably due to fluctuations in the optically active impurities 

 rather than to variations in the sugars themselves. This is shown by 

 the fact that these fluctuations can be correlated with the differences 

 between the cane sugar values as determined by reduction and polari- 

 sation. When the apparent laevulose increases faster than the dextrose 

 the results for cane sugar obtained by polarisation are lower than the 

 reduction values ; when the apparent dextrose increases faster than the 

 laevulose or the laevulose falls more rapidly than the dextrose, the 

 polarisation results are in excess of the true values. 



6. The fluctuations of the apparent dextrose and apparent laevulose 

 take place more or less regularly during the 24 hours ; this points to a 

 regular variation in the optically active impurities. 



7. In the leaves the values of saccharose obtained by the double 

 polarisation method are almost always higher than the reduction values ; 

 in the stalks, however, they are sometimes very high and sometimes 

 very low. This is probably due to the presence of at least two different 

 optically active substances at different times of the day. The increase 

 of the apparent laevulose corresponds with the increase of the substance 

 causing low results for cane sugar by the double polarisation method ; 

 the increase of apparent dextrose corresponds with a falling off of this 

 substance and the formation of the impurity which gives high results. 



8. Until more reliable results can be obtained for the true dextrose 

 and laevulose bj^ methods which are independent of the polarimetric 

 data, it seems justifiable, from the results brought forward, to assume 

 that the dextrose and laevulose exist in the leaves and stalks as invert 

 sugar and travel in nearly, if not exactly, equal proportions to the root, 

 where retransformation into saccharose occurs. This assumption best 

 agrees with the regular rise and fall of the total hexoses in the stalks 

 and mid-ribs along almost straight lines during the night, as contrasted 

 with the more irregular fluctuation of the apparent dextrose and laevulose 

 taken separately. 



9. It is impossible in the present state of our knowledge to draw 

 any conclusions from the proportion of apparent dextrose or laevulose 



