APPENDIX. 



159 



be conveniently determined by Olerget's method, both readings (before and after 

 inversion), however, being made alter addition of ammonia. Should the sugar in 

 these materials have become partly inverted, the reducing sugar must be determined 

 by Fehling's process, and due allowance made for it. 



Class III. Direct readings of rotation in ammoniacal solution are inadmissible in 

 analyses of the substances of this class on account of the influence of iron and 

 aluminum on the rotation of tartaric acid, and on account of the small but unknown 

 rotation of the trace of inverted starch. 



Accurate determinations, however, may be made in the presence of excess of 

 ammonium molybdate in neutral solution. The latter substance has the property 

 of greatly increasing the rotation of tartaric acid, so that by its use the small rotation 

 of the inverted starch is made insignificant. It is to be noted, however, that this 

 increased rotation is very sensitive to the presence of alkali and acid, and is, more- 

 over, modified by phosphates. It is therefore necessary, in the first place, to remove 

 the phosphoric acid, and, secondly, to bring the solution to a definite state of 

 neutrality Hoth these results are attained by the following procedure, the details 

 of which must be carefully adhered to: 



Solutions required. The following solutions must be prepared, but need not be 

 made up very accurately: 



Molybdate solution: 44 grains ammonium heptamolybdate in 250 cc. 



Citric acid solution: 50 grams citric acid in 500 cc. 



Magnesium sulphate solution: 60 grams MgSO 4 . 7H 2 O in 500 cc. 



Ammonia solution: 80 cc concentrated ammonia (sp. gr. 0.880) in 500 cc. 



Hydrochloric acid: 60 cc concentrated hydrochloric acid in 500 cc. 



Methyl orange solution: 



Method of i <><<'< Inre. An amount of material containing not more than 0.2 gram 

 tartaric acid, not more than 0.3 gram alum, and not more than 0.3 gram calcium 

 superphosphate is accurately weighed and placed in a dry flask. To this, 5 cc. of 

 citric acid and 10 cc. of molybdate solution are added and allowed to react with 

 the substance for 10 or 15 minutes (with an occasional shake). Next, 5 cc. of mag- 

 nesium sulphate solution are added and 15 cc. of ammonia solution stirred in. After 

 a few minutes (not more than an hour) the solution is filtered through a dry filter, a 

 slight turbidity of the filtrate being disregarded. To 20 cc. of the filtrate are then 

 added a few drops of methyl orange and hydrochloric acid, from a burette, till the 

 pink color appears ( '2 or .'> drops too much or too little are of no consequence). 

 Finally, 10 cc. more of the molybdate solution are added to the pink solution, 

 which now becomes colorless or pale yellow; and w r ater is added to make up the 

 volume to 50 cc. This s< dution, after filtering if necessary, is polarized in a 20-cm. tube. 



The amount of tartaric acid in grams (y) in the weight of substance originally 

 taken is given by the following formula, in which x is the rotation in minutes: 



// = 0. 001086 x + 0.001601 */x. 

 But if the rotation is not less than 40', the simpler formula, 



y = 0.0075 + 0.001 168 x, 

 may be employed. 



The following table gives the tartaric acid in grams for every 10 minutes rotation: 



