W. A. Davis 343 



J become nearly equal to unity. In all other cases the ratio varies 



from 2-5 to 5, or even higher in the case of the bottom halves of the 

 stalks. In passing from the leaves to mid-ribs, from mid-ribs to the 

 tops of stalks and from the tops of stalks to the bottom the proportion 

 of apparent dextrose steadily and rapidly increases. 



Series I. Tops and bottoms of stalks. The analyses given in Table IV 



show that the proportion of apparent dextrose to laevulose [ -- j is far 



higher in the bottom half of the stalks than in the top half at the same 

 time. At 6 a.m. of August 26th at first sight it would appear that during 

 the night the laevulose practically disappears from both top and bottom 

 halves of the stalk just as it appeared to do in the leaf (see Table I) 

 during the night in the same series. Lindet observed a similar pheno- 

 menon in the case of the sugar beet and attributed the predominance of 

 dextrose in the stalks to the laevulose being used more rapidly than the 

 dextrose for purposes of tissue building. But that it is quite unsafe 

 to rely upon the polarimetric data as affording any real index of the 

 proportions of dextrose and laevulose actually present in the stalks is 

 shown by the following considerations. The stalks stand out in striking 

 contrast to the leaves as regards the extraordinary divergences between 

 the results obtained for saccharose by the reduction method and by the 

 method of double polarisation. In some cases, for example at 6 p.m., 

 Table IV, the polarisation results for cane sugar in the bottom 

 halves are 40 per cent, higher than the values obtained by reduction ; 

 and yet at the same time the tops give polarisation results which are 

 85 per cent. lo^v. The following table (Table VIII) gives a conrfjarison 

 of the data obtained for cane sugar by the two methods (reduction 

 and polarisation), showing that the divergence is very much greater in 

 the stalks than in the leaves and much more variable in its nature. 

 The fact that the tops may give by polarisation a large apparent 

 deficiency of saccharose and the bottoms at the same time a large 

 excess as compared with the reduction values (see the data at 12 noon 

 and 6 p.m.. Table VIII), or vice versa as at midnight when the relations 

 are reversed, points to the presence in the top and bottom, halves of the 

 stalk at different periods of the day of quite different impurities, with 

 different and opposite rotatory powers (substances as different as d- 

 and ^glutamine or d- and ^asparagine). 



A careful comparison of Table VIII with the curves showing the 

 variation of apparent dextrose and laevulose in the stalks (Figs. 4 



