DETERMINATION OF SUGARS IN GRAIN AND CATTLE FOODS. 7 



As has been suggested, errors are likely to be introduced in the 

 cuprous oxid weighing method by the presence of mineral salts and 

 organic matter. Fehling solution being strongly alkaline will cause 

 in some cases a precipitation of salts on the introduction of the sugar 

 solution. For this reason it is better to determine the copper in the 

 precipitated cuprous oxid by some such method as Low's. 1 Burning 

 the cuprous oxid to cupric oxid will only correct for the organic mat- 

 ter, so it could not replace an actual copper determination. 



A comparison of these methods on samples of commercial feeding 

 stuffs of which molasses was one of the ingredients resulted as follows: 



Results on molasses feeds by the two methods. 



1 No results by this method. 



These figures show some peculiar variations. In the sodium car- 

 bonate digestion method when lead was used as the clarifying agent 

 the results are lower and more closely approach the alcohol digestion 

 figures. In samples No. 8049 and 8050 the results by the two methods 

 are fairly comparable, but in the others, and especially in No. 7900, 

 the sodium carbonate digestion method gives much higher figures for 

 the percentage of total sugars, while in Nos. 7894, 7901, and 7934 the 

 increase in total sugars is not so large as the increase in reducing 

 sugars. The original samples left were not large enough to determine 

 the cause of this increase. For sample No. 7900 the increase in total 

 sugars amounts to 3.33 per cent. The results on this sample were 

 carefully checked and the determination was also made by the 

 section of plant physiological chemistry and the results as shown 

 in the table agreed with those obtained by the sugar laboratory. 

 A portion of the sample was placed in a Soxhlet extractor and ex- 

 tracted with alcohol until no sugar was found in the alcohol by the 

 ar-naphthol test. Another portion of the sample was boiled for two 

 and a half hours with a measured quantity of water and the sugar 

 determined in this extract. Little or no starch was present in this 



U. S. Dept. Agr., Bureau of Chemistry Bui. 107, Revised, p. 241. 



