28 



It is well here to mention that sugar of milk is the only sugar 

 which can accurately be determined by the use of this method, 

 provided the original plan of manipulation is adhered to. 



The author proposes to prepare two separate solutions to avoid 

 the changes which Fehling's solution always undergoes. To this 

 end he prepares a standard copper solution, containing in 1 liter at 

 15° C. 34.64 cupric sulphate, purified by repeated crystallizations, 

 grinding, and drying between filter papers. The second solution 

 is prepared by dissolving 173 grs. of rochelle salts in 350 c.c. of 

 water, and boiling. After cooling, 600 c.c. of sodium hydrate solu- 

 tion (sodium hydrate purified by crystallization from alcohol) of 1.12 

 specific gravity, which had likewise been heated to boiling, are 

 added. The mixture is finally brought to 1 liter at 15° C. 



These two solutions are kept separate in a number of small 

 well-corked bottles, and are only mixed when used. 



For the titration of the sugar, the following plan has been 

 adopted : 5 c.c. of each of these solutions are brought into a por- 

 celain dish, with 20 c.c. of water, and heated to boiling. The sugar 

 solution is then gradually added from a burette, care being taken 

 that the solution is kept boiling constantly. 



Evaporated water should, from time to time, be replaced by 

 the addition of distilled water. The formation of dry crusts on 

 any part of the dish must be prevented. They are likely to get 

 overheated, and thus may give rise to the formation of products 

 which also reduce copper solution. The titration should be re- 

 peated at least once, and must be completed within five to ten 

 minutes. The process is at an end when all the copper is thrown 

 down as red cuprous oxide, which is best determined by the com- 

 plete disappearance of the blue color in the liquid. To observe 

 this better, the dish is slightly inclined, so as to enable one to ex- 

 amine the color against a white background. In doubtful cases, a 

 trace of the fiuid is filtered through a very small filter, acidified 

 and tested with potassium ferrocyanide. A red color indicates the 

 presence of copper, and a few more drops of the sugar solution 

 are added, the boiling being continued until the precipitation of 

 the copper has been completed. Care should also be exercised that 

 no sugar remains in the fluid. If, therefore, no copper reaction 

 ensues, we must always test for sugar in a separate portion by the 

 addition of copper solution, and boiling. 



