114 Carbon Assimilation 



(ii.) The completeness of the inversion of cane sugar. 



(iii.) The completeness of the fermentation of sugars other than 

 pentoses by bakers' yeast. 



(iv.) The completeness of the fermentation of sugai-s other than 

 maltose and pentoses by maltase-free yeasts. 



Finally the accuracy of the glucose and fructose determinations 

 depends upon the accuracy of the following determinations: — 



(i.) The reducing power of the original plant extract. 



(ii.) The optical rotation of the original plant extract. 



(iii.) The reducing power of the extract after inversion. 



(iv.) The reducing power of the extract after fermentation 

 with maltase-free yeasts. 



(v.) The reducing power of the extract after fermentation 

 with bakers' yeast. 



(vi.) The estimation of pentoses by the Krober-Tollens method. 



As the accuracy of the glucose and fructose determinations thus 

 depends on the accuracy of 13 separate assumptions, operations and 

 determinations, it is not to be expected that the results given for 

 glucose and fructose are likely to have a high order of accuracy. 

 Indeed, Davis (1916) points out that the extracts probably contain 

 optically active substances other than sugars, e.g., amino-acids and 

 amides. As Davis iiimself says, "the values given as dextrose 

 and Ijevulose probably do not, in most cases, represent real values " ; 

 he therefore prefers to designate them as " apparent dextrose " and 

 "apparent Isevulose." 



A method by which it may be possible to make more 

 satisfactory estimations of fructose is suggested in a recent 

 publication of Miss Wilson and Atkins (1916). The sucrose is first 

 estimated by measuring the reducing power and optical rotation of 

 the solution of mixed sugars before and after treatment with 

 invertase. After inversion, fructose may then be estimated by 

 oxidising other sugars (glucose and maltose) by means of bromine. 

 Under certain definite conditions, the glucose and maltose are 

 destroyed by this means, and the fructose remains almost entirely 

 unchanged. The method is not very exact, but it is possible that 

 further research on it may i-ender it more accurate, and in any case, 

 the results obtained by its means are not open to all the objections 

 of the indirect method previously described. It is, moreover, 

 considerably more rapid. 



Although Miss Wilson and Atkins worked out the method in 

 order to apply it to the analysis of leaf extracts, no account of work 

 involving its use is as yet available. 



